CereTom - Carpeta de Producto by deleccientifica

CONTENIDO C E R E TO M 5 DELEC CIENTÍFICA

A B O C A D O S A L A I N N OVA C I Ó N

C O N S U LTO R Í A E S P E C I A L I Z A D A

S E R V I C I O T É C N I C O E X P E R TO

C A R A C T E R Í S T I C A S P R I N C I PA L E S

ÁREAS DE APLICACIÓN

F O L L E TO S 2 9 ESPECIFICACIONES TÉCNICAS

C ATÁ L O G O 4 7 BIBLIOGRAFÍA 59

Esta carpeta fue generada por el equipo de consultores de DeLeC Científica. Buenos Aires, Argentina. 2019. DeLeC Científica Uruguaya - Representante Regional Exclusivo Fco. García Corina 2357 – Piso 1. Montevideo - Uruguay DeLeC Científica Argentina – Agente Comercial Local Aráoz 821 -C1414DPQ - Buenos Aires – Argentina. Tel: (+54-11) 4775 5844

as patologías neurológicas son foco de atención de la mayoría de las instituciones asistenciales de todo el mundo.

Estadísticamente en Argentina, el ACV es la cuarta causa de muerte en adultos y la principal causa de discapacidad. Si no se trata a tiempo, por cada minuto que pasa, los efectos empeoran. Pueden causar, pérdida de memoria, problemas de comportamiento y problemas en el habla. Por otro lado, según estadísticas de la Organización mundial de la Salud (OMS), cada año mueren más de 1.2 millones de personas por accidentes automovilísticos y entre 20 y 50 millones padecen traumatismos no mortales. Las personas que sobreviven, suelen en su mayoría sufrir consecuencias de tipo físico, cognitivo y conductual, por lo que los “Traumatismos Encéfalocraneano (TEC)” son considerados la primera causa de discapacidad neurológica en adultos jóvenes. Ante este panorama nos gustaría presentarles una solución que se involucra completamente en esta situación, una herramienta que permite disminuir los tiempos de atención, mejorando la calidad del servicio y brindando la posibilidad de una mejor y más rápida recuperación de estos pacientes. CereTom es un tomógrafo de 8 cortes que permite realizar angiografías y perfusión sin importar la localización del paciente. Se logró transformar las tecnologías de tomografía fija en plataformas móviles que nos permiten cambiar el paradigma a “La tecnología se acerca al paciente” (ver página 17). La combinación de su rápido escaneo, la flexibilidad de seteos y la posibilidad de tener acceso inmediatamente a las imágenes, lo hace una herramienta indispensable para cualquier profesional médico que necesite información en tiempo real de pacientes en estado crítico. Posee ventajas que lo transforman en un equipo versátil, deseado por muchas áreas de atención de la salud: es un equipo autoblindado, por lo cual CereTom no requiere un búnker y puede utilizarse a lo largo de todas las instalaciones. Es compatible con todos los neuronavegadores que trabajen con el formato DICOM, funciona con baterías externas que se cargan en la red normal de 220V, su peso y su sistema de traslado avanzado permiten que la tecnología alcance cualquier punto donde se la solicite (ver página 23).

Por esta sumatoria de beneficios creados por las características específicas de CereTom, en DeLeC estamos seguros que es el esquipo adecuado y necesario para toda la comunidad médica que desee y se ocupe de brindar una mejor calidad de atención, disminuir los riesgos de vida y mejorar los tiempos y resultados de la rehabilitación. Es una mirada hacia al futuro en la atención de emergencias neurológicas, a la cual, ya tenemos posibilidad de acceso. Invitamos a que lo conozcan y consulten las publicaciones científicas que respaldan sus resultados.

CERETOM

CERETOM, DE SAMSUNG

¿Qué es el sistema CereTom®? Cuando se le pregunta a Eric Bailey, presidente de NeuroLogica, acerca de su motivación para comenzar con la compañía, recuerda dos hechos que lo marcaron para siempre. El primero, cuando un ataque cerebrovascular dejó a su abuelo en un estado muy débil durante diez años. Y el segundo, cuando su hermano murió debido a una lesión en la cabeza que sufrió en un accidente automovilístico. De allí la idea de fundar NeuroLogica. Una tomografía computarizada podría haber salvado a su hermano y quizá mejorado la calidad de vida de su abuelo. CereTom® es el primer tomógrafo portátil y autoblindado del mundo, diseñado principalmente para tomar imágenes de cabeza y cuello. Fue desarrollado en el año 2004 por la empresa norteamericana NeuroLogica. Su estructura compacta y software de avanzada permiten trasladar el equipo por todas las áreas de un hospital. Al estar autoblindado, no existe la necesidad de un búnker o la asignación de protección radiológica en el sector donde se deba realizar la tomografía. Estas características hacen que CereTom® sea más que un equipo de imágenes: es una solución multi-departamental que ofrece la posibilidad de realizar tomografías en todos los servicios, evitando la complicación y el costo de trasladar los pacientes. En la terapia, permite evaluar pacientes complejos cuyo traslado implica riesgo, tiempo, y costo. En la guardia, con CereTom® se puede realizar una rápida evaluación de las personas que ingresan con síntomas de stroke o con politraumatismos, ganando tiempo en situaciones donde cada segundo cuenta. En neurocirugías, el equipo permite tomar imágenes dentro del mismo quirófano, antes, durante y después del procedimiento, brindando mayor seguridad y calidad de atención. En pacientes pediátricos y neonatos incluso se pueden realizar tomografías de cuerpo entero. CereTom® es una solución flexible para todo el hospital.

DELEC CIENTÍFICA

¿Quiénes somos?

ABOCADOS A LA

DeLeC Científica es una empresa exitosa, en constante crecimiento, líder en innovación tecnológica aplicada a la medicina que fue fundada en el año 2003 por un grupo empresario de capitales nacionales, con vocación de servicio y alto grado de Responsabilidad Social.

“Si en el futuro somos lo que proyectamos, en el presente somos lo que hacemos en virtud de aquella decisión y proyecto”. José Ortega y Gasset

¿Cuál es nuestra misión y concepto de negocio? Nuestra misión es impulsar la mejora continua de nuestro sistema de salud regional, ofreciendo para ello los mejores productos de la revolución tecnológica del siglo XXI.

A tal fin, nos hemos propuesto hacer foco en lo especial y proveer soluciones a problemas de los que nadie se ha ocupado, teniendo en cuenta que hay muchas empresas de electromedicina, de todo tipo y tamaño, que se dedican a atender el mercado de volumen con soluciones estandarizadas que dejan afuera a mucha gente.

¿A quiénes servimos y con qué tipo de productos? Servimos a las comunidades científicas y médicas proveyéndoles productos y servicios de última generación y alto valor agregado.

Nos concentramos exclusivamente en aquellos que son seguros y están debidamente certificados y aprobados por los organismos internacionales de control - FDA y CE - y también los nacionales - ANMAT y ARN –; todo en el marco de un soporte pre y post venta de excelencia.

¿En qué nos especializamos? Somos consultores altamente especializados en el desarrollo e implementación de programas médicos de excelencia soportados por equipos y sistemas de real innovación tecnológica.

I N N O VA C I Ó N

¿Cómo estamos organizados? La firma cuenta con seis áreas de trabajo con roles bien definidos. • División de Sistemas Médicos: Se distingue por proveer la mejor tecnología de punta disponible, a partir de la revolución tecnológica del siglo XXI, para cubrir necesidades de equipamiento de diagnóstico y también de tratamiento. Busca optimizar resultados clínicos y al mismo tiempo mejorar la calidad de la experiencia vivida por los pacientes antes, durante y después del proceso de tratamiento. • División de Cirugía Robótica y Simulación: pone a disposición de la comunidad médica la posibilidad inigualable de asistencia robótica a la cirugía mínimamente invasiva y también los únicos simuladores quirúrgicos verdaderamente realísticos; considerados los mejores del mundo en cirugía virtual. • División de Radioterapia y Radiocirugía: Ofrece la trilogía ideal de equipos para tratamiento de tumores malignos y benignos. • División de Ingeniería, Soporte Técnico y Aplicaciones Clínicas: Asegura el correcto funcionamiento de la base de equipos de innovación tecnológica instalada, incluyendo su actualización continua, y brinda servicio docente a los usuarios para asegurar su correcta utilización y las buenas prácticas. • División de Comunicación y Marketing: Área en constante crecimiento que busca llegar estratégicamente a los públicos de relevancia para la empresa y trasmitir un mensaje claro sobre su misión y objetivos. Para que “las cosas sean” es necesario comunicar que existen. “No se desea lo que no se conoce”. • División de Administración, Personal y Finanzas: Optimiza los resultados económicos de la empresa, cuidando que haya una distribución equitativa de los recursos entre los seis grupos de interés: proveedores, clientes, personal, accionistas, bancos/inversores y el fisco. Su objetivo principal es velar por una gestión eficiente y ecuánime al momento de crear valor económico produciendo, al mismo tiempo, valor social.

Precisión submilimétrica e innovación en tratamientos de radioterapia y radiocirugía.

Radioterapia intraoperatoria, Pequeño, ligero, móvil, autoblindado y efectivo.

Tomógrafo de mamas con imágenes 3D realmente isotrópicas.

Hospitales móviles diseñados a la necesidad del cliente.

Tomógrafos móviles autoblindados de uso en clínicas y ambulancias.

Adquisición continua de imagen en bipedestación 2D y 3D con baja dosis.

Simuladores de alta y mediana fidelidad y sistema audiovisual con herramientas de gestión.

Cabezas para intubación, torso para trauma y monitor multiparamétrico simulado.

Tecnología no invasiva que ayuda a visualizar venas no visibles a simple vista.

Sistema de adquisición de señales biológicas y software para educación e investigación.

¿Cuál es nuestra filosofía? Nuestro lema es ganar cuando el cliente también gana, cumplir con lo prometido y hacerlo a tiempo.

¿Qué nos diferencia del resto de las empresas del rubro? Nuestro diferencial es que no sólo proveemos equipos, sino que desarrollamos programas médicos de excelencia, acompañando al cliente desde la etapa embrionaria del proyecto hasta su optimización operativa. • Seleccionamos el equipamiento necesario y lo instalamos. • Brindamos soporte técnico con garantía oficial. • Nos encargamos del entrenamiento de los médicos y del personal de la institución local. • Una vez que llegamos a una institución, nunca nos vamos.

¿Cuál es nuestra modalidad de trabajo? Trabajamos en equipo con proveedores y clientes a través de una continua actividad de docencia.

No tenemos vendedores, pero nos atenemos al perfil y la descripción habitual de las tareas de un vendedor técnico. Nuestro equipo se compone de profesionales universitarios, frecuentemente con posgrado que, a poco de ingresar a la compañía, son enviados a las distintas fábricas representadas para capacitarse de la mejor forma para ofrecer cada producto con solvencia técnica y científica.

¿Contra quiénes competimos? No poseemos competidores, ya que vamos en una dirección distinta al resto, usualmente: cambiando paradigmas.

¿Dónde comercializamos y soportamos nuestros productos? La firma comercializa la mayoría de sus productos en Argentina, Uruguay, Paraguay y Bolivia. Sin embargo, muchas veces, a pedido de distintos fabricantes, extendemos nuestro radio de acción llegando a otros países de América del Sur. En el 2010 introdujimos la cirugía robótica en Colombia, preparando el terreno para que luego se nombrara un distribuidor local. Antes, en el 2008, habíamos instalado las dos primeras unidades da Vinci que hubo en Brasil. Lo hicimos en el Hospital Albert Einstein y en el Sirio Libanes, ambos de Sao Paulo. Cabe destacar que también en este país existe actualmente un distribuidor local para esta tecnología.

¿Cuál es nuestra visión de futuro? En el mediano plazo esperamos ser líderes regionales y referentes indiscutidos en innovación tecnológica aplicada a la medicina.

Esto lo lograremos gracias a nuestro comportamiento empresarial, calidad de relaciones que establezcamos, eficiencia y efectividad de nuestros productos y servicios. Apostamos a ser una organización de renombre y prestigio regional, comprometiéndonos con la comunidad y ayudando para la construcción de un mundo mejor a través de la mejora continua del servicio de salud.

Oficinas de DeLeC Científica

C O N S U LTO R Í A

Nuestra experiencia en el ámbito de la innovación tecnológica en salud nos dice que

ESPECIALIZADA

el mejor equipo de diagnóstico por imágenes no hace una intervención de calidad por sí mismo. Tan importante como la herramienta son la formación, la comprensión de la tecnología, el buen uso, el asesoramiento y los objetivos que orientan la práctica. Por eso en DeLeC Científica acompañamos a las instituciones desde el desarrollo de los proyectos, el diseño de nuevas áreas o servicios de salud, el asesoramiento en la adquisición de nuevas tecnologías, los requerimientos normativos y legales, la diagramación logística, el mantenimiento y el monitoreo del uso. Un asesoramiento adecuado es clave para: - obtener planificaciones que permitan optimizar el tiempo de los proyectos, - prever los riesgos potenciales para garantizar la viabilidad en el mediano y largo plazo, - identificar todos los stakeholders alrededor del proyecto y prever cuál será el impacto en ellos, - conseguir una mirada profunda sobre la inversión, el retorno y reconocer oportunidades que no están a la vista. Con el fin de asesorar tomando como referencia los máximos estándares de calidad, los consultores de DeLeC nos actualizamos de acuerdo a los programas de formación de las firmas que representamos y participamos de forma activa en la agenda más relevante de la innovación tecnológica médica en Occidente.

Ofrecer un servicio técnico de alta performance, alineado tanto a las exigencias y

SERVICIO TÉCNICO

estándares de las marcas con las que trabajamos, como a los requerimientos de

EXPERTO

nuestros clientes es un punto destacado en nuestra empresa. El equipo técnico asiste en la interpretación de los requerimientos previos (condiciones eléctricas, infraestructura, etc.), se ocupa de la instalación, cuando el equipo lo requiere, y luego monitorea el funcionamiento y el uso para garantizar el desempeño óptimo de la tecnología. El servicio de instalación de DeLeC Científica para el equipo CereTom cumple con todos los procesos recomendados por la fábrica. Nuestros técnicos deben cumplir con un cronograma de formación y capacitación anual, en las casas matrices de las firmas que representamos. Por lo tanto, desde DeLeC Científica ofrecemos una asistencia de instalación y posventa certificada por fábrica.

CARACTERÍSTICAS PRINCIPALES

C A R AC T E R Í S T I C A S P R I N C I PA L E S

CereTom® es un tomógrafo portátil de 8 cortes que entrega la mejor calidad de imagen para estudios sin contraste, angiotomografías y estudios de perfusión por contraste, sin importar dónde se encuentre el paciente. Su rapidez de escaneo, flexibilidad de configuración y la visualización instantánea de las imágenes hacen de CereTom® una herramienta indispensable para cualquier médico que necesite información en tiempo real del estado de sus pacientes más críticos.

Dimensiones: 134 x 153 x 73 cm (ancho x alto x profundidad) Peso: 438kg

Las capacidades de software de la Workstation permiten visualización 2D, 3D y reconstrucción multiplanar, y las imágenes producidas tienen compatibilidad DICOM 3.1. Estas características permiten total integración con cualquier sistema de PACS con el que se trabaje en la institución. CereTom® también incluye una tabla radiolúcida universal, adaptable a todo tipo de camas. Así, todas las camillas del hospital se pueden convertir rápidamente en plataformas de escaneo estables y seguras, evitando traslados riesgosos para los pacientes.

- Tomógrafo multislice de 8 cortes de 1,25mm. - Posibilidad de escanear en forma axial, helicoidal y dinámica. - Tomografías con y sin contraste, estudios de perfusión y angiotomografías. - Calidad de imagen equivalente a la de un tomógrafo fijo. - Apertura: 32cm. FOV: 25cm. - Tubo de rayos X de ánodo fijo, refrigerado por aire. - Voltaje del tubo: 100, 120, 140kV. Corriente del tubo: 1 - 7mA. - Capacidades de software: 2D, 3D, reconstrucción multiplanar. - Compatibilidad DICOM 3.1 - Imágenes compatibles con todos los sistemas de navegación quirúrgica, PACS, HIS, RIS. - Protocolos personalizados. - Comunicación wireless o por cable. - Alimentación de red y a baterías. Autonomía de las baterías: 2 horas.

La mejor calidad de imagen para estudios sin contraste, angiotomografías y estudios de perfusión por contraste, sin importar dónde se encuentre el paciente. 19

BENEFICIOS

Gracias a su portabilidad y autoblindaje, CereTom® permite que sea el tomógrafo el que vaya hacia el paciente, y no a la inversa. Esto implica grandes beneficios en distintas áreas: mayor calidad de atención para los pacientes, mayor eficiencia en el funcionamiento de los servicios, y beneficios económicos para la institución en general. Tomando imágenes dentro del quirófano, los neurocirujanos pueden planear los procedimientos con mayor precisión, y por lo tanto más seguridad. CereTom® es compatible con todos los neuronavegadores, por lo tanto las imágenes cargadas en estos sistemas pueden ser obtenidas inmediatamente antes de la cirugía, o incluso durante la misma. Finalmente, con imágenes postoperatorias dentro del mismo quirófano es posible corroborar la eficacia del procedimiento. Con CereTom® se evita el traslado de pacientes delicados desde la unidad de terapia intensiva al tomógrafo fijo para control, con lo cual: • Permite un mejor seguimiento de la evolución del paciente, sin riesgos. • Brinda mayor disponibilidad de tiempo a los intensivistas para cubrir otras necesidades. • Libera el uso del tomógrafo fijo para diagnóstico de otros pacientes. El uso de CereTom® en la guardia reduce el tiempo de diagnóstico para los pacientes con síntomas de ACV. En este tipo de pacientes el tiempo es un factor fundamental, y con cada segundo ganado se evita la pérdida de millones de neuronas: tiempo es cerebro. Todas estas ventajas son posibles gracias a la portabilidad y calidad de imagen del CereTom®. Además, a diferencia de los tomógrafos fijos, no requiere de la construcción de un búnker por estar autoblindado. Podrá obtener mayor información sobre los beneficios de la tomografía portátil de CereTom® dirigiéndose a la sección de Áreas de aplicación.

SEGURIDAD RADIOLÓGICA

Frente a la ventaja de poder usar el tomógrafo en las distintas áreas del hospital, surge la duda de cuánta radiación de dispersión es emitida y qué nivel de seguridad se le brinda al personal y a los pacientes. CereTom® entrega la misma dosis a los pacientes que los tomógrafos fijos de última generación. Para estos niveles de dosis, NeuroLogica realizó estudios para evaluar la radiación dispersa. Los resultados indican que el técnico radiólogo, posicionado a una distancia de 2 metros, perpendicular a la cama, puede realizar más de 26 tomografías por día sin superar los límites de radiación permitidos.

Estos estudios fueron evaluados según las normas del American College of Radiology, y la empresa también sigue el estándar ALARA, que busca utilizar la mínima cantidad de dosis posible. Por otro lado, cabe destacar que el tomógrafo cuenta con la aprobación de la FDA, el marcado CE, está habilitado por la ANMAT, y está siendo utilizado en más de 500 instituciones en todo el mundo con excelentes resultados. Finalmente, el CereTom® puede incluir de manera opcional el software de reducción de radiación RadRedux, que gracias a sus filtros de eliminación de ruido permiten trabajar con un nivel de dosis 50% menor al habitual, sin sacrificar calidad de imagen.

ÁREAS DE APLICACIÓN

T E R A P I A I N T E N S I VA

Frecuentemente en las terapias intensivas se enfrentan con el desafío de trasladar pacientes críticos al área de tomografía, que en muchos hospitales suele estar en otros pisos o a grandes distancias. El problema surge cuando estos pacientes se encuentran conectados a varios equipos de soporte de vida y monitoreo. En muchos casos se decide directamente no trasladarlos ya que hacerlo puede llegar a producir su descompensación y ponerlos en riesgo. CereTom® posee la ventaja de ser movilizado fácilmente por un operador y de llegar a todos los pacientes, incluso a aquellos que por su gravedad no pueden ser trasladados. Su portabilidad y la rapidez para obtener imágenes de alta calidad permiten evaluarlos de una manera más segura y eficiente, mejorando la calidad de tratamiento. CereTom® hace que la tomografía pueda ser un estudio rutinario para los pacientes complejos. El uso de la tomografía móvil en terapia conlleva varias ventajas. Por un lado, permite que el paciente pueda permanecer en su cama sin interrumpir la atención que recibe. Evitar traslados implica evitar complicaciones respiratorias o de oxigenación. Por otro lado, gana tiempo en el cuidado del paciente. Las imágenes diagnósticas son obtenidas con rapidez y evaluadas junto a la cama. En caso de existir alguna complicación, el tratamiento puede comenzar más rápidamente. Finalmente, evita que el personal de terapia tenga que abandonar la sala. Esto se traduce en mejores cuidados para el resto de los pacientes, mayor seguridad para el personal, que no necesita dejar su lugar de trabajo, y un manejo más eficiente de los recursos1.

1. Para mayor información ver en la sección Publicaciones el artículo: Kaitlin Peace et al. The Use of a Portable Head CT Scanner in the Intensive Care Unit.

Q U I RÓ FA N O S

Por su portabilidad, CereTom® puede ser utilizado dentro de un quirófano, sin necesidad de blindar la sala. Esto le permite al neurocirujano obtener imágenes tomográficas pre, intra y postoperatorias dentro de la misma sala de tratamiento, y así cerciorarse de la efectividad del procedimiento. El uso de CereTom® en quirófano puede ser aplicado por los servicios de neurocirugía, cirugía de cabeza y cuello, cirugía reconstructiva, otorrinolaringología, y pediatría, donde incluso se pueden obtener imágenes de cuerpo entero. Algunos ejemplos del uso del CereTom® en quirófano son: • Evaluación del alcance de resecciones tumorales, como glioblastomas, meningiomas y tumores de la hipófisis. Un estudio realizado para evaluar el uso de CereTom® indica que lesiones residuales fueron encontradas en 32% de los pacientes que sobrellevaron una cirugía de estas características. Estudiando estas lesiones y su proximidad a estructuras anatómicas delicadas, se decidió continuar con la resección o finalizar el procedimiento2. • Colocación de catéteres ventriculares, donde es recomendable realizar una tomografía en la sala de operaciones para confirmar que se encuentren en la posición adecuada, y de no ser así modificarla. De esta manera, se evitan complicaciones, se gana tiempo, y aumenta la confianza del neurocirujano3. • Estimulación cerebral profunda. Con CereTom® es posible verificar la correcta colocación de los electrodos de manera intraoperatoria y sin la necesidad de mantener consciente al paciente. Así, el procedimiento se puede realizar bajo anestesia general, con mayor rapidez y confort para los pacientes. • La tomografía intraoperatoria también permite actuar con rapidez frente a diversas complicaciones clínicas que pueden ocurrir durante las neurocirugías. Con CereTom® se pueden realizar chequeos inmediatos en caso de edemas, hemorragias, hidrocefalia aguda o isquemias2.

2. Para mayor información ver en la sección Publicaciones el artículo: Mobile Computed Tomography: Early Experience in Korea, Jin Wook Kim, MD et al. 3. Puede encontrar más ejemplos y casos ilustrativos en la sección Publicaciones en el artículo: Alterations in Surgical Plan Based on Intraoperative Portable Head Computed Tomography Imaging, Andrew P. Carlson, MD, et al.

GUARDIA

CereTom® permite mejorar los tiempos de diagnóstico en ACV. Un stroke o accidente cerebro vascular (ACV) es caracterizado por una brusca interrupción del flujo sanguíneo neto que irriga al cerebro y que origina una serie de síntomas variables en función del área cerebral afectada. Existen de dos tipos de ACV: • ACV isquémico (o infarto cerebral) Se debe a la oclusión de alguna de las arterias que irrigan la masa encefálica, generalmente por ateroesclerosis o bien por un émbolo (embolia cerebral) que viaja desde otra localización, fundamentalmente el corazón u otras arterias. • ACV hemorrágico (o hemorragia cerebral) Se deben a la ruptura de un vaso sanguíneo encefálico debido a un pico de hipertensión o a un aneurisma congénito. La ventaja de CereTom® en los ACV reside en lo que se llama efecto ventana. Cuando un paciente sufre un ACV hay un período de 3 horas a partir de ocurrido el accidente, dentro del cual, si el paciente recibe el tratamiento, tiene un 85% de posibilidades de recuperarse sin daño cerebral. Con CereTom® instalado en la guardia hospitalaria reducimos este tiempo de diagnóstico al mínimo. El paciente llega de la ambulancia, se hace el diagnóstico que permitirá saber en forma temprana qué tipo de ACV está padeciendo y así determinar rápidamente el tratamiento a seguir. Se estima que en la Argentina ocurren entre 130.000 y 190.000 ataques cerebrales por año, es decir, uno cada 4 minutos. El 30% de las personas que sufren un ataque cerebral fallece en el primer mes, o sea que en nuestro país fallecen entre 39.000 y 60.000 personas al cabo de los primeros 30 días luego del ataque cerebral.

OTORRINOLARINGOLOGÍA

Con CereTom®, los pacientes pueden tener la consulta, realizar la tomografía, recibir su diagnóstico y discutir sus opciones de tratamiento, todo en una sola visita. De manera opcional, el CereTom® puede incluir una silla coronal, cuya altura y el ángulo del asiento pueden ajustarse de forma simultánea para acomodar pacientes de todos los tamaños, desde el más bajo al más alto. Confort mejorado significa menos movimiento, dando como resultado mejores imágenes. Esta silla permite tomar imágenes directamente coronales, evitando transformarlas desde las imágenes axiales (procedimiento en el cuál se pierde calidad de imagen).

N E O N ATO LO G Í A Y P E D I AT R Í A

Posee todas las ventajas nombradas en las otras áreas, con el diferencial de que en neonatos y pacientes pediátricos hasta cierto tamaño pueden realizarse tomografías de cuerpo completo. En una unidad de cuidados intensivos pediátricos se realizó un estudio para evaluar la utilidad de la tomografía de cabeza para el servicio. Se trabajó sobre un total de 391 pacientes. En el 62% de los casos se debió utilizar el CereTom® para evitar el traslado, debido a la gravedad del estado del paciente y el grado de atención que necesitaba. Entre las conclusiones más importantes, se rescata que mediante el uso de tomografía portátil en más de un tercio de los pacientes se encontraron nuevas patologías, y en un cuarto se decidió modificar el tratamiento de acuerdo a lo observado. Por otro lado, se confirmó que la dosis de radiación entregada por el CereTom® cumple con los estándares de los Estados Unidos4. Cabe destacar que para los pacientes pediátricos y neonatos, en quienes es recomendable utilizar la menor cantidad de radiación posible, es ideal incluir el software de reducción de ruido RadRedux, que permite obtener calidad de imagen diagnóstica al 50% de la dosis.

4. Para mayor información ver en la sección Publicacioes el artículo: Head Computed Tomography Scanning During Pediatric Neurocritical Care: Diagnostic Yield and the Utility of Portable Studies, Kerri L. LaRovere et al.

FOLLETOS

ESPECIFICACIONES TÃ&#x2030;CNICAS

CATÁLOGO

BIBLIOGRAFÍA

KAITLIN PEACE ET AL. (2009). The Use of a Portable Head CT Scanner in the Intensive Care Unit. JIN WOOK KIM, MD ET AL. (2010). Mobile Computed Tomography: Early Experience in Korea. ANDREW P. CARLSON, MD, ET AL. (2011). Alterations in Surgical Plan Based on Intraoperative Portable Head Computed Tomography Imaging KERRI L. LAROVERE ET AL. (2011). Head Computed Tomography Scanning During Pediatric Neurocritical Care: Diagnostic Yield and the Utility of Portable Studies.

CONTENIDO

PUBLICACIONES CIENTĂ?FICAS

M o b i l e C o m p u t e d To m o g r a p h y : Ea r l y e x p e r i e n ce i n Ko re a

Reducing hospital-acquired infections among the neurologically critically III

The economic annddiniiacal bfits of portable hed/neck CTimaging inthe intensive care unit

The use of a portable head CT scanner in the intensive care unit

The use of mobile computed tomography in intensive care: regulatory compliance and radiation protection

Head computed tomography scanning during pediatric neurocritical care: Diagnostic yield and the utility of portable studies

Intraoperative CT and nexframe-guided placement of bilateral hippocampal based responsive neurostimulation: A technical note

Alterations in surgical plan based on intraoperative portable head computed tomography imaging

Retrograde partial migration of ventriculoperitoneal shunt with chambre: Review of causative factors and its prevention

Early low cerebral blood flow and high cerebral lactate: prediction of delayed cerebral ischemia in subarachnoid hemorrhage

Pre-hospital CT diagnosis of subarachnoid hemorrhage

Derivation of predictive score for hemorrhagic progres of ecrebral contusions in moderate and severe traumatic brain injury

103

Mobile stroke units for prehospital thrombolysis, triage, and beyond: benefits and challenges

110

PUBLICACIONES CIENTÍFICAS

Mobile Computed www.jkns.or.kr Tomography: Early J Korean Neurosurg experience in Korea

10.3340/jkns.2010.48.1.31

Print ISSN 2005-3711

On-line ISSN 1598-7876

Soc 48 : 31-36, 2010

Clinical Article

Mobile Computed Tomography : Early Experience in Korea Jin Wook Kim, M.D.,1 Sang Hyung Lee, M.D.,2 Young-Je Son, M.D.,2 Hee-Jin Yang, M.D.,2 Young Seob Chung, M.D.,2 Hee-Won Jung, M.D.1 Department of Neurosurgery,1 Seoul National University Hospital, Seoul, Korea Department of Neurosurgery,2 Seoul Metropolitan Government-Seoul National University Boramae Medical Center, Seoul, Korea

Objective : With improved technology, the values of intraoperative computed tomography (iCT) have been reevaluated. We describe our early clinical experience with a mobile CT (mCT) system for iCT and discuss its clinical applications, advantages and limitations. Methods : Compared with intraoperative magnetic resonance imaging, this mCT system has no need for major reconstruction of a preexisting operating room for shielding, or for specialized instruments or equipment. Patients are placed on a radiolucent head clamp that fits within the gantry. Because it consists simply of a scanner and a workstation, it can be moved between locations such as an operating room, an intensive care unit (ICU) or an emergency room without difficulty. Furthermore, it can achieve nearly all types of CT scanning procedures such as enhancement, temporal bone imaging, angiography and three-dimensional reconstruction. Results : For intracranial surgery, mCT can be used for intraoperative real-time neuronavigation by interacting with preoperative images. It can also be used for intraoperative confirmation of the extent of resection of intracranial lesions and for immediate checks for preventing intraoperative unexpected accidents. Therefore, the goals of maximal resection or optimal treatment can be achieved without any guesswork. Furthermore, mCT can achieve improved patient care with safety and faster diagnosis for patients in an ICU who might be subjected to a ventilator and/or various monitoring devices. Conclusion : Our initial experience demonstrates that mCT with high-quality imaging offers very useful information in various clinical situations. KEY WORDS : Intraoperative computed tomography ˙ Mobile computed tomography.

neuronavigation2,13). As a result, intraoperative assessment of the precise orientation or localization of structures and the extent of resection of intracranial lesions such as brain tumors can be improved. These approaches are also useful for intraoperative monitoring of various unexpected intraoperative brain events including acute hemorrhages, diffuse brain swelling and acute hydrocephalus. Intraoperative imaging systems can be divided into several types, such as magnetic resonance imaging (MRI), computed tomography (CT) and ultrasound13). Among them, intraoperative MRI (iMRI) has been used more extensively than other systems. However, with modern technology, systems based on intraoperative CT (iCT) are now more common. Here, we describe our early clinical experience of iCT with a mobile CT (mCT) system and discuss its clinical applications, advantages and limitations.

INTRODUCTION Over the past couple of decades, intraoperative neuronavigation has been used widely as a standard procedure for precise guidance, orientation and localization in most neurosurgical operations1,2,9). The importance of obtaining realtime intraoperative imaging is being emphasized increasingly instead of being limited to neuronavigation. There could be several reasons for this. Existing neuronavigation systems cannot reflect exactly the “brain shifts” resulting from the loss of cerebrospinal fluid (CSF) or the removal of an intracranial lesion during an operation, so the acquisition of intraoperative images is mandatory for more accurate intraoperative • Received : January 26, 2010 • Revised : May 26, 2010 • Accepted : June 21, 2010 • Address for reprints : Sang Hyung Lee, M.D.

Department of Neurosurgery, Seoul Metropolitan Government-Seoul National University Boramae Medical Center, Seoul National University College of Medicine, 39 Boramae-gil, Dongjak-gu, Seoul 156-707, Korea Tel : +82-2-870-2302, Fax : +82-61-870-3863 E-mail : nslee@snu.ac.kr

MATERIALS AND METHODS For intraoperative imaging scanning, we used a portable 31

J Korean Neurosurg Soc 48 | July 2010

Fig. 1. Photograph of mobile computed tomography (mCT) scanner, gantry and workstation in the operation room and the intensive care unit. A : Intraoperative photograph. For mCT scanning in the operating room, there is a need for additional procedures to confirm safety, such as wrapping the gantry in sterile draping, moving the gantry to the patientâ&#x20AC;&#x2122;s head and performing a simulation without radiation exposure. B : MCT scanning for a patient in an intensive care unit. This demonstrates a portable workstation that can receive images through a wireless system and display three-dimensional reconstructed images.

CT scanner, the CereTom (Neurologica, Danvers, MA, USA). This is one of a number of recently developed mobile units and is designed for multi-slice scanning of the head and neck (Fig. 1A). It comprises a mobile system with a portable workstation. It is smaller than other portable CT units, with a weight of 362 kg, a height of 153 cm, a length of 134 cm, a width of 72 cm and a gantry diameter of 32 cm10). Thus, it requires only one person for its transport and can be moved to any places including an operating room (OR), an intensive care unit (ICU) or an emergency room (ER) without difficulty (Fig. 1B). Typical scanning parameters are 120 kV, 7 mA, a scanning time of 2 s and a 10-mm-wide collimated beam with eight 1.25-mm-wide detectors. It can provide a reconstructed section thickness of 1.25 mm, as described10). The average patient effective dose for a routine head CT examination on the CereTom is -1.7 mSV, typical for head examinations performed on any CT scanner during which adults receive between 1 and 2 mSV10). For acquiring mCT images in the OR, only a few additional procedures are required as follows. The gantry needs to be wrapped in sterile drapes to maintain sterile conditions during the scanning procedure. The gantry of the mCT is then moved to the patientâ&#x20AC;&#x2122;s head to perform a simulation scan without using radiation and to confirm the safety of the position. Contrast medium can be infused intravenously as needed. A required preoperative procedure for intraoperative CT scanning is that the patient must be placed on a conventional operating table fitted with a radiolucent extension or a radiolucent head fixator.

RESULTS Between January and August 2009, 63 patients underwent scanning using the mCT system in the OR or the ICU. Among these, 20 patients with postoperative routine CT scanning were excluded. Among the remaining 43 patients, we measured the extent of resection of intracranial lesions in 22 (51%). Nine patients (21%) underwent scanning in the ICU, eight (19%) underwent ventricular catheter placement surgery and four (9%) were scanned for monitoring intraoperative brain events. Among the 22 procedures for assessing the extent of resections, there were 10 pituitary adenomas, four vestibular schwannomas, four meningiomas, two intracranial hematomas and one each of a brain metastasis and an abscess. Assessing the extent of resection

We used the mCT for assessing the extent of resection. For example, a 61-year-old male patient underwent an operation to remove a pituitary macroadenoma that had extended considerably to the suprasellar area (Fig. 2A). After an initial surgical resection using a transsphenoidal approach, an iCT scan was performed using the mCT for assessing the extent of resection of the tumor. After the injection of contrast medium, a residual enhanced lesion was detected over the suprasellar area and surgical resection was resumed for additional resection (Fig. 2B). After a second intraoperative CT scan, the surgeon decided to finish the operation without additional resection because of the close vicinity of the tumor to the 32

Mobile Computed Tomography | JW Kim, et al.

Fig. 2. Mobile CT for a pituitary macroadenoma. A : Preoperative axial and sagittal images with contrast enhancement. B : Initial intraoperative mobile computed tomography (mCT) scanning with contrast enhancement showing axial images (upper) and coronal reconstructed images (lower). A small residual enhancing mass was detected at the top of the lesion. C : After the second intraoperative CT scanning, the small residual enhancing mass was still located close to the anterior communicating artery.

Fig. 3. Mobile computed tomography (mCT) of a tuberculum sellae meningioma. A : A meningioma was located at the tuberculum sellae area and compressed the left optic nerve. B : Intraoperative mCT axial images with contrast enhancement. A small residual enhancing mass was detected beneath the left optic nerve.

scan was performed for immediate review (Fig. 4A). This revealed that the ventricular catheter was located in the temporal horn of the right lateral ventricle. Catheter repositioning of the ventricular catheter into the frontal horn was performed in the OR (Fig. 4B). In another patient undergoing a ventriculo-peritoneal shunt, like aforementioned case, optimal catheter repositioning was achieved through the intraoperative mCT scan. In the 6 remaining cases, intraoperative mCT scan was used due to the difficulty such as the very small size of ventricles to confirm the position of the ventricular catheter.

anterior cerebral artery, although some residual lesion remained (Fig. 2C). The second case demonstrated another application of the mCT for improving the extent of resection. Patient was a 51year-old female who had presented with decreased visual acuity of the left eye caused by a tuberculum sellae meningioma (Fig. 3A). We achieved a total removal of the tumor based on additional information from an intraoperative mCT scan, which demonstrated an enhanced residual mass beneath the left optic nerve (Fig. 3B). Among the 22 procedures in which introperative mCT scan was used to assess the extent of resection, residual lesions were actually detected in 7 (32%) patients. Surgery was continued for additional resection in 3 cases.

ICU and monitoring intraoperative events

Another application of mCT was for patients in our ICU who underwent insertion of an intracranial cerebral pressure monitor and pentothal coma therapy (Fig. 1B). Nine patients who had taken an ICU management for pentothal coma therapy or monitoring for increased intracranial cerebral pressure received an immediate CT scan without the need for transportation to a remote conventional CT facility. The preoperative CT images of a 7-year-old boy with sudden-onset of projectile vomiting and lethargy had a large hemorrhage in the cerebellum (Fig. 5A). During an emergency evacuation of the hematoma, inexplicable severe cerebellar swelling was observed although a major part of the hematoma had been removed. After intraoperative mCT scanning, a

Ventricular catheter placement

The mCT can be also useful for the intraoperative evaluation of ventricular catheter placement operations such as ventriculo-peritoneal shunting or extraventricular drainage. Fig. 4 shows a patient who underwent a ventriculo-peritoneal shunt operation for the relief of hydrocephalus. Once the ventricular catheter had been placed through the right parieto-occipital point without a problem, the flow of CSF drainage was slightly diminished. An intraoperative mCT 33

J Korean Neurosurg Soc 48 | July 2010

Fig. 4. Mobile CT for performing a ventriculo-peritoneal shunt. A : Intraoperative photograph of a ventriculo-peritoneal shunt operation in the operating room. B : The proximal ventricular catheter was located in the right temporal horn (left). Adjustments for correcting malposition of the ventricular catheter were made immediately into the frontal horn in the operation room.

Fig. 5. Mobile computed tomography (mCT) used for monitoring intraoperative brain events. A : The preoperative CT images showed a large hemorrhage in the cerebellum. B : A diffuse acute subdural hemorrhage was detected in the left fronto-temporal area after intraoperative mCT scanning.

supratentorial subdural hemorrhage was confirmed on the left side (Fig. 5B). After close radiological follow-up without an additional operation, the subdural hemorrhage had disappeared gradually. Another monitoring was performed to confirm the site of stereotactic brain biopsy in which the intraoperative frozen biopsy was repeatedly reported as normal brain tissue. Remaining 2 procedures were performed to evaluate a sudden and unexpected brain swelling during the evacuation of epidural hemorrhage and subacute subdural hemorrhage.

ma2,13). When this occurs, there is an inherent risk that any neuronavigation based on preoperative images becomes progressively inaccurate5). The degree of displacement caused by a brain shift is difficult to assess intraoperatively, so the usefulness of neuronavigation may diminish during the surgical procedure2). The best potential approach for solving this problem is the acquisition of intraoperative real-time images showing anatomical structures and intracranial lesions accurately13). These intraoperative real-time images, which facilitate more precise orientation and localization, are increasingly important for contemporary neurosurgical procedures. Since iMRI was applied to the neurosurgical field in the mid-1990s, it has been more extensively used than other systems, because it offers higher resolution of brain structures and intracranial lesions such as brain tumors than did earlier CT scanners5,12). However, it has some disadvantages such as the need for shielding, which requires major reconstruction of a preexisting OR. It also requires specialized operative instruments and equipment, especially for high-field systems12). Furthermore, it needs considerable expense for installation and maintenance and the installation of iMRI has been limited mainly to a few large hospitals12). On the other hand, although they were introduced in the early 1980s, iCT systems have not been used extensively because

DISCUSSION Since neuronavigation has been introduced to neurosurgery, better clinical outcomes have been achieved through the better guidance, orientation and localization it provides for neurosurgeons5). Such neuronavigation is now a routine method for almost all neurosurgical procedures. However, one of the most important drawbacks for the traditional neuronavigation is undoubtedly the problem of â&#x20AC;&#x153;brain shiftâ&#x20AC;?. Spatial displacement between the preoperative images and the surgical field can occur after opening of the dura and this will increase gradually because of CSF drainage, changes in arterial pCO2, removal of intracranial lesion and brain ede34

Mobile Computed Tomography | JW Kim, et al.

images in the OR. Although only a few studies for ventricular catheter placement surgery used by iCT have been reported, the usefulness of iCT for these operations is obvious. Ventricular catheter placement surgery is one of the most frequent neurosurgical procedures. In this condition, iCT can help in diagnosing any suboptimal placement of ventricular catheter and complications, allowing for correction before the patient leaves the OR. In particular, it can improve the confidence of a neurosurgeon when placing a ventricular catheter in complex cases or in various operations into nontraditional sites including cysts, small ventricles and Ommaya reservoir insertion. The frequency in use for ventricular catheter placement surgery in our study was around 20%. If there were no mCT available, we would have to use guesswork and decide on a treatment plan only after the confirmation of postoperative CT images. One of the advantages of our iCT system is its mobility. This mCT technology could provide a better way for evaluating patients in the ICU and emergency department as well as the OR, because it takes a relatively short time to obtain images and it improves the quality of treatment for the patients. Transporting ventilated, sedated patients who are critically ill and physiologically unstable from the ICU to a conventional CT room is often slow, and is a potential risk that can result in aggravated secondary injuries to the brain4,8). Further, mCT could also prove useful in the emergency department for evaluating patients suspected of stroke or head trauma as it decreases the evaluation time10). Rumboldt et al.10) reported that the inherent risks associated with such intrahospital transport could reach more than 50% and adverse events could occur in a further 15% of cases, even with a well-trained transport team. Gunnarsson et al.4) reported that the staff workload for one scanning procedure used by a portable CT system was reduced by 60% to 75%. The mCT can perform almost all CT scanning procedures including image enhancement, angiography, fluoroscopy, temporal bone scans and various three-dimensional reconstructions10). Through these varied protocols, it can be used extensively not just for patients undergoing neurosurgery, but also for patients in other specialties such as orthopedics, pediatrics and otorhinolaryngology. Hoelzle et al.6) reported that the iCT monitoring of orbital floor fractures is considered a very useful surgical aid for immediate monitoring of a surgical reduction. Stieve et al.11) demonstrated that iCT could assist in exposure of the skull base and lamina papyracea during endonasal surgery or cochlear implantation in otorhinolaryngology. Actually, economic efficacy as well as clinical benefits must be considered. Previous studies have concluded that the introduction of an iCT with or without a portable

they give limited image quality, are time-consuming and give relatively high radiation exposure to patients12). However, with modern technology, iCT systems have become smaller and more compact and can be moved freely in and out of the OR. They do not require specialized equipment as they give a low radiation dose, nor do they need new space for intraoperative imaging9). Thus, the advantage of iCT-based technology compared with iMR imaging is its simplicity combined with adequate image quality. For these reasons, the value of iCT has been reevaluated and the demand for iCT systems is increasing. Clinical applications

Mobile CT, as one of the iCT systems, was installed at our hospital in late 2008. The clinical applications of mCT can be categorized as follows. First, the acquisition of intraoperative images by mCT can provide more precise information for real-time orientation and localization than does neuronavigation based on preoperative images. Using this information, it is now possible to assess the extent of resection of intracranial lesions such as brain tumors or intracranial hematomas, intraoperatively. Consequently, maximal resection or optimal resection for intracranial lesions can be achieved. Gumprecht and Lumenta3) reported the clinical application of iCT scanning for removing brain tumors. According to their report, the tumors remaining in 32/76 patients (42%) could be detected by intraoperative scanning. Among these, surgery was continued for additional tumor resection in 16 cases. In our study, mCT scanning for the same purpose was performed on 22 patients (51%). Although we cannot confirm that this additional resection results in significantly improved clinical outcomes such as long-term tumor control rate or survival benefits, we believe that at least it provides a realistic possibility for surgeons to avoid unexpected or unwanted residual lesions. The combination of realtime neuronavigation and intraoperative assessing of the extent of resection could be more useful for various operations at the skull base. Another advantage of mCT is that it can reduce uncertainty in various clinical situations during neurosurgical procedures. Whenever intraoperative images are needed, an immediate imaging check can be performed to check on unexpected complications such as diffuse brain edema, localor more rarely, distant-brain hemorrhages, acute hydrocephalus or ischemia2). The discovery of such an unexpected brain event provides an opportunity for immediate action and helps to avoid serious consequences2). In our study, we could achieve a fast diagnosis and appropriate decisionmaking without any guesswork through the confirmation of distant intracranial hemorrhages based on intraoperative CT 35

J Korean Neurosurg Soc 48 | July 2010

system in the OR or ICU is feasible and cost-effective4,7,10,12).

322, 2003 4. Gunnarsson T, Theodorsson A, Karlsson P, Fridriksson S, Bostrรถm S, Persliden J, et al. : Mobile computerized tomography scanning in the neurosurgery intensive care unit : increase in patient safety and reduction of staff workload. J Neurosurg 93 : 432-436, 2000 5. Haberland N, Ebmeier K, Hliscs R, Grnewald JP, Silbermann J, Steenbeck J, et al. : Neuronavigation in surgery of intracranial and spinal tumors. J Cancer Res Clin Oncol 126 : 529-541, 2000 6. Hoelzle F, Klein M, Schwerdtner O, Lueth T, Albrecht J, Hosten N, et al. : Intraoperative computed tomography with the mobile CT Tomoscan M during surgical treatment of orbital fractures. Int J Oral Maxillofac Surg 30 : 26-31, 2001 7. Masaryk T, Kolonick R, Painter T, Weinreb DB : The economic and clinical benefits of portable head/neck CT imaging in the intensive care unit. Radiol Manage 30 : 50-54, 2008 8. Matson MB, Jarosz JM, Gallacher D, Malcolm PN, Holemans JA, Leong C, et al. : Evaluation of head examinations produced with a mobile CT unit. Br J Radiol 72 : 631-636, 1999 9. Nakao N, Nakai K, Itakura T : Updating of neuronavigation based on images intraoperatively acquired with a mobile computerized tomographic scanner : technical note. Minim Invasive Neurosurg 46 : 117120, 2003 10. Rumboldt Z, Huda W, All JW : Review of portable CT with assessment of a dedicated head CT scanner. AJNR Am J Neuroradiol 30 : 1630-1636, 2009 11. Stieve M, Schwab B, Haupt C, Bisdas S, Heermann R, Lenarz T : Intraoperative computed tomography in otorhinolaryngology. Acta Otolaryngol 126 : 82-87, 2006 12. Uhl E, Zausinger S, Morhard D, Heigl T, Scheder B, Rachinger W, et al. : Intraoperative computed tomography with integrated navigation system in a multidisciplinary operating suite. Neurosurgery 64 : 231239; discussion 239-240, 2009 13. Willems PW, van der Sprenkel JW, Tulleken CA, Viergever MA, Taphoorn MJ : Neuronavigation and surgery of intracerebral tumours. J Neurol 253 : 1123-1136, 2006

Limitations

Because this system is CT-based, it offers limited evaluation for several lesions such as low grade gliomas or poorly delineated lesions and it has an inherent risk of radiation exposure. In addition, because it has a relatively smaller gantry than other iCT equipment in order to permit effective transport, it cannot evaluate other body parts below the head and neck. Because of the initial experience with mCT, it still needs an expert technician in managing of the mCT system for various CT procedures. A long-term follow-up study is mandatory for determining the clinical efficacy, costs and benefits of mCT. CONCLUSION Our initial experience shows promising results that mCT with various high-quality images offers a very reasonable technology in various clinical situations such as intraoperative imaging and ICU care. References

1. Enchev Y : Neuronavigation : geneology, reality, and prospects. Neurosurg Focus 27 : E11, 2009 2. Foroglou N, Zamani A, Black P : Intra-operative MRI (iop-MR) for brain tumour surgery. Br J Neurosurg 23 : 14-22, 2009 3. Gumprecht H, Lumenta CB : Intraoperative imaging using a mobile computed tomography scanner. Minim Invasive Neurosurg 46 : 317-

The benefits o a portable head CT scannr

international

hospital Equipment & Technology

Volume 37 November 2011

Medical Management

Weekly news updates on www.ihe-online.com

The benefits of a portable head CT scanner The small bowel in children: diagnostic and therapeutic approaches Management of POCT in a teaching hospital also in this issue

Ultrasound: highlights from WFUMB 2011 Hospital spotlight: mobile DR in an ICU New initiatives at RSNA 2011

central blood pressure and arterial stiffness monitoring Page 31

new generation X-ray trolleys Page 33

Portable colour doppler electrosurgical pencil ultrasound system with smoke evacuation Page 35 Page 36

The Magazine for healThcare decision Makers 10

– November 2011

CoMPUtEd toMoGrAPHy

the benefits of a portable head Ct scanner the development of a portable head Ct scanner represents a significant technologic advance. Avoiding transportation of iCU patients can decrease transportation-related morbidity. Portable Ct speeds up the time for scanning, which is critical in conditions such as stroke, where “time is brain.” independent cost benefit analysis shows a 169% return on investment with portable Ct. by dr Andrew P. Carlson Computed tomography (CT) has become the most accessible standard technique in the diagnostic toolkit for head and brain imaging. The use of head CT in general has led to a major paradigm shift since its advent in the 1970s and 80s, making the days of exploratory burrholes unnecessary. The development of CT imaging with funding from the recording company EMI led to a Nobel prize in Physiology or Medicine

in 1979 [1]. Since the time of the installation of the first EMI scanner of the head at the Mayo clinic, the hardware and software improvements have been monumental. A more recent development in CT imaging has been the development of a portable head CT scanner [2,3]. Though there are several types, the NeuroLogica CereTom (Danvers, MA, USA) is the best evaluated

and will be discussed here. There are many theoretical and actual benefits to such a technology. Besides having utility in clinical decision-making, the scanner is a cost-effective investment as well.

Overview

The scanner itself weighs around 750lb, and can be wheeled either manually or by a joystick control by one person. The scanner and its separate control tower can be stored in a hallway or spare room. [Figure 1]. Once the scanner is positioned, the scans are carried out by precisely advancing the scanner itself on a “centipede” system of tracks. The scanner has eight detectors, and has been carefully independently assessed and found to produce high quality, diagnostic images with similar radiation dose to conventional scanning [4]. All standard CT settings are adjustable to the radiologist’s preferences and there are several dose settings and types of reconstruction algorithms for images.

Clinical utility of portable CT

There are at least three key advantages to the use of portable CT head imaging:

1. Portable CT allows for critically ill patients to remain in one place without risking transportationrelated morbidity.

Figure 1. The portable head CT scanner stored in a hallway alcove in the Intensive Care Unit.

Figure 2. A portable scan with blood flow mapping being done in the intensive care unit on a patient with multiple monitoring devices.

The risk of intra hospital transportation to patients is increasingly being recognised. A typical visit to radiology at our institution requires that all of the patient’s monitoring cables be shifted to a small transport monitor. Most invasive neuromonitors are disconnected and not monitored during transportation. The endotracheal tube is either connected to a portable ventilator or a bag-valve mask (though recently efforts are being made to transport without any circuit disconnection, so the entire ventilator is brought). The nurse, one of the two patient care technicians for the unit, and the only respiratory therapist then accompany the patient down an elevator and through a hall. The patient and all the lines are then slid onto the CT table, the scan is performed, and the patient and all the lines are slid back. During this period, there are many occasions when a central line, arterial line,

9 external ventricular drain or other monitoring device can be displaced. One report noted that such mishaps occur in 40-60% of patient transports [5]. Performing the CT at the patient’s bedside has been shown to decrease the risk of complications from 25% to 4.3% in high risk patients [6]. Besides these kinds of events, many critical monitoring parameters may not be recorded during the scan and transport. For neurosurgical patients, this may be particularly relevant if the CT scan is attempting to answer a physiologic question such as is the case with a CT perfusion or xenon/CT blood flow study. If the patient’s intracranial pressure or oxygenation probes are not recording, it may be difficult to correlate the tomographic data with the continuous data back in the ICU [Figure 2].

2. Portable CT provides fast access to information for making emergent patient care decisions.

Speed in decision-making is often critical in neurological situations and the now familiar “time is brain” mantra is true for many critical conditions. Pooled estimates of recovery with tissue plasminogen activator (t-PA) in cases of acute stroke show a continuous slope of decline in benefit over the first several hours of administration times. Earlier imaging of a patient with hemiplegia and aphasia may mean the difference between being able to administer thrombolytics or not (or may mean an earlier trip to the operating room if a large subdural haematoma is seen instead). With conventional scanning, obtaining the CT may take an hour or more, and with a 4.5 hour window for administration of t-PA from onset of symptoms, unfortunately many potential candidates simply time out. A situation where patients would be prioritised and the appropriate diagnostic techniques would be brought to the patient seems ideal in such critical situations. In the ICU, many patients are balancing on a fine line of stability and deterioration. In the case of a change in the neurologic exam, key developments may be missed during the “packaging” and transportation to the CT scanner. During the hour spent in transport and imaging, the patient may have deteriorated further. If the scanner can be brought to the patient, the physician can continue to track the changes in the ICU. Furthermore, respiratory or haemodynamic parameters may cause a patient to be too unstable to be transported safely. Bedside imaging would then be the only way to obtain information with

regard to intracranial processes. Patients with a post-operative haematoma can be diagnosed quickly and brought back to the operating room if needed. The scanner can be used to check the position of ventricular catheters in the brain at the bedside immediately after placement or even during the procedure [Figure 3].

– November 2011

We offer full range of eye protection during anesthesia

EyeGard™ The safe, effective way to protect the eyes during surgery.

3. Portable CT allows for anatomic imaging in locations within the hospital that typically do not have access to CT imaging.

Besides the intensive care unit, there are several areas that might not have access to scanning where it may be useful. We found that when portable CT scanning was used in the operating room for a specific procedure such as the placement of a shunt catheter or tumour resection, changes were made in 32% of cases [2], [Figure 4]. Intraoperative imaging is increasingly being used, but for many institutions it is cost prohibitive. A fixed CT or MRI scanner in the operating room will be unused for much of the time, so hybrid systems are in use where routine scans can be performed in an adjacent room, but this still limits scanning to one specially equipped operating room. CT certainly does not offer the resolution of MRI that is needed for some specific applications, but for most situations it is probably adequate. The versatility of a portable system means that a scan can be performed in any operating room. Some uses that have been explored are confirmation of catheter placement, confirmation of deep brain stimulation lead placement, ruling out haematoma or mistargeting after biopsy, facial fracture reduction and foreign body removal. Intra-operative scanning can also be used to update neuronavigation with changing intra-operative landmarks to improve accuracy. CT scanning in the angiography suite may occasionally be desired, especially in the case of balloon test occlusion for performing tomographic SPECT or other physiologic CT based CBF studies. These kinds of real-time tomographic assessment could not otherwise be made. A role has been suggested for a portable scanner in the outpatient clinic setting, particularly in otolaryngology, where sinus imaging could be performed immediately if it was considered necessary, potentially avoiding several clinic visits and a delay in diagnosis.

Cost benefits of portable CT

With increased reliance in medicine placed on assessment of cost benefit, it is critical to understand both the actual and potential

Bat Mask

Protect your patient’s eyes during general anesthesia against corneal abrasion, lacerations and scleral hemorrhage with the Bat Mask eye cover.

U.S.A. 813-889-9614 • Fax 813-886-2701

see us at MEDICA - Hall 16, E06 Distributors Wanted

www.ihe-online.com & search 45908 Oxygen-therapy and Suction Devices

French Manufacturer ISO 9001 & 13485 - CE marking

NEW RVTM3 VACUUM REGULATOR

We are pleased to present you our new RVTM3 vacuum regulator.

The RVTM3 is used to measure and to adjust the vacuum level within the context of medical suctions. It enables to drain substances out of the patient’s body during surgical procedures. The new RVTM3 is compact, strong, ergonomic and easy-to-use: manual adjustment of the vacuum gauge from -45° to +45°, easy-click rotation of the safety jar, integrated outlet tubing nipple... It is available in mbar/hPa (0-250, 0-600, 0-1000) as well as in mmHg (0-400, 0-760). Many different standards are available according to various markets. Visit us at Medica «Hall 11 Booth E40» TECHNOLOGIE MEDICALE S.A.S.

101 rue Vaillant Couturier - P.O. Box 46 - F-93136 NOISY LE SEC (Paris) - FRANCE Tel. + 33 (0)1 48 45 58 95 - Fax. + 33 (0)1 49 42 90 21 e-mail: info@technologiemedicale.com - www.technologiemedicale.com

www.ihe-online.com & search 45963

– November 2011

CoMPUtEd toMoGrAPHy

N costs of implementing a new technology such as portable CT. A group from the Cleveland Clinic has transitioned to exclusive use of the portable CT scanner in the neuro ICU. In the first six month period, 502 scans were performed, and an estimated cost savings of over €39, 000 based on personnel costs for scanning was reported [7]. An internal rate of return of 169% on the investment with a break-even point of 6.9 months was reported. In addition, it was noted that there was increased efficiency in use of the conventional scanner. This is intuitive as well, since in an institution where both emergent and scheduled scans are performed on the same scanner, there will often be times when the scanner is waiting for an emergent case.

The above-mentioned review, however, did not explore many of the additional potential cost related issues of implementing a scanner as they may be more difficult to estimate. Any true cost/benefit assessment must also consider the costs of transportation related morbidity mentioned above. This is especially true with regard to central line and ventilator tube disconnection, which may not only increase costs of replacement, but of hospital acquired infection as well. Other less defined costs, such as the burden placed on the remaining staff in the intensive care unit when a nurse, a patient care technician and respiratory therapist are off the floor for the scan are difficult to quantify, but are nevertheless important factors. Increasing speed of access to imaging has the potential to allow for intervention earlier in situations such as stroke, and it is clear that earlier intervention can improve clinical outcomes in stroke and related conditions. This offers a cost benefit that may be significant over a population if it results in a decreased rate in morbidity, dependency and loss of productivity.

Conclusion

Figure 3. The scanner is being used for immediate confirmation of correct placement of an external ventricular drain in the intensive care unit. The arrow points to the patient’s head positioned in the scanner. After the scanner and head is draped, the scan can be performed from within the sterile field.

Portable head CT imaging represents a significant advance in diagnosis and treatment of patients with cranial pathology. Bringing the technology to the patient can increase the speed and safety of imaging, which is critical, particularly for many neurologic emergencies. There is a further benefit in that it decreases personnel costs in ICU patients and increases efficiency of fixed scanning. It is likely, for the reasons discussed, that such technology will become standard of care over the next decade.

References

1. Alexander RE, Gunderman RB. EMI and the first CT scanner. J Am Coll Radiol 2010;7(10):778-781. 2. Carlson AP, Phelps J, Yonas H. Alterations in Surgical Plan Based on Intraoperative Portable Head Computed Tomography Imaging. Journal of neuroimaging : official journal of the American Society of Neuroimaging Apr 7 2011. 3. Carlson AP, Yonas H. Portable Head Computed Tomography ScannerTechnology and Applications: Experience with 3421 Scans. Journal of neuroimaging : official journal of the American Society of Neuroimaging. Jun 23 2011. 4. Rumboldt Z, Huda W, All JW. Review of Portable CT with Assessment of a Dedicated Head CT Scanner. AJNR Am J Neuroradiol 2009;30(9):1630-6 5. Smith I, Fleming S, Cernaianu A. Mishaps during transport from the intensive care unit. Critical care medicine 1990;18(3):278-281. 6. Gunnarsson T, Theodorsson A, Karlsson P et al. Mobile computerized tomography scanning in the neurosurgery intensive care unit: increase in patient safety and reduction of staff workload. J Neurosurg 2000;93(3):432-436. 7. Masaryk T, Kolonick R, Painter T, Weinreb DB. The economic and clinical benefits of portable head/neck CT imaging in the intensive care unit. Radiol Manage 2008;30(2):50-54.

The author Figure 4. Use of portable head CT in the operating room in 50 cases and effect on operative plan. (Adapted from Carlson AP, Phelps J, Yonas H. Alterations in Surgical Plan Based on Intraoperative Portable Head Computed Tomography Imaging. Journal of neuroimaging : official journal of the American Society of Neuroimaging. Apr 7 2011.)

Andrew P. Carlson, MD, MSCR Department of Neurological Surgery University of New Mexico School of Medicine Albuquerque New Mexico USA

Neurocrit Care DOI 10.1007/s12028-016-0286-2

ORIGINAL ARTICLE

Reducing Hospital-Acquired Infections Among the Neurologically Critically Ill 2 • 2,3 • 2,3 • John J. Halperin1,2 • Stephen Moran Doriann Prasekinfections Ann Richards Reducing hospital-acquired among the 1,2 • 2,3 Charlene Ruggiero Christina Maund

neurologically critically III

Springer Science+Business Media New York 2016

Abstract Background Hospital-acquired infections (HAIs) result in excess morbidity, mortality, and resource consumption. Immobilized, ventilator-dependent ICU patients are at the highest risk of HAI. Methods Despite broad implementation of relevant bundles, HAI incidence in our neuro ICU remained high, particularly catheter-associated urinary tract infections (CAUTIs) and ventilator-associated events (VAEs). We reviewed the administrative data and nosocomial infection markers (NIMs) for all neurology and cranial neurosurgery patients admitted to our neuro ICU between January 2011 and May 2014, identified and implemented interventions, and measured effects using National Healthcare Safety Network (NHSN)-defined CAUTIs and VAEs. Interventions included (1) reviewing Foley catheter use, including indications and alternatives, and instituting daily rounds, continuously questioning the ongoing need for a catheter; (2) re-educating neuro ICU personnel in insertion and maintenance technique, introducing a new kit that simplified and standardized sterile insertion; and (3) placing a mobile CT in the neuro ICU since our patients required repeated transports for brain imaging and since we found correlations between frequencies of these transports, and both respiratory and urinary NIMS. Results VAEs decreased 48 %, Foley use decreased 46 %, CAUTIs decreased from 11/1000 catheter days to 6.2. & John J. Halperin john.halperin@atlantichealth.org 1

Department of Neurosciences, Overlook Medical Center, 99 Beauvoir Avenue, Summit, NJ 07902, USA

Atlantic Health System, Morristown, NJ, USA

Overlook Medical Center, Summit, NJ 07902, USA

Overall complication rate decreased 55 %, ICU length of stay 1.5 days, and risk-adjusted mortality 11 %. Conclusions Combining a multidisciplinary approach with rigorous analysis of objective data, we decreased total HAIs by 53 % over 18 months. Key drivers were decreased urinary catheter use and decreased patient transport from the ICU for imaging. Keywords Hospital-acquired infections Neurocritical care CAUTI Ventilator-associated events Reduction

Background The management of critically ill patients with severe central nervous system insults has advanced dramatically in recent years. However, morbidity and mortality remain substantial. These patients are often bedridden and immobilized, frequently requiring ventilator support for extended periods of time. Such patients are at high risk of hospitalacquired infections (HAIs), particularly pulmonary [1, 2] and urinary [3, 4] infections. These in turn can prolong hospitalization [5] and increase the risk of additional complications and death. Despite widespread utilization of bundles of care for ventilator [6], central line, and other aspects of care [7], HAIs still occur with some frequency in this population. These issues have been of particular concern in our institution—a Joint Commission-designated Comprehensive Stroke Center and regional referral center for neurologic disorders—where we have a large population of very ill patients with intracranial hemorrhages, large ischemic strokes, and other severe nontraumatic brain injuries (Table 1), with over 2000 patients admitted

123 14

Neurocrit Care Table 1 Principal diagnoses in the prospective patient group (% of each patient group with a given diagnosis, January 2013– March 2016)

All neuro ICU (n = 1.913) Hemorrhagic stroke

17 % (n = 329)

14 % (n = 58)

Ischemic stroke

18 % (n = 338)

10 % (n = 41)

Aneurysm repair

14 % (n = 276)

24 % (n = 94)

Seizures

3 % (n = 64)

9 % (n = 34)

Other intracranial neurosurgery

51 % (n = 982)

60 % (n = 236)

Other

11 % (n = 20)

7 % (n = 26)

annually with primary diagnoses of neurological or cranial neurosurgical disorders; of these patients, more than one in four require a stay in our critical care unit—a dedicated 10-bed neurosciences intensive care unit (neuro ICU) staffed by two fellowship-trained neurointensivists. In this group of patients, 458 required mechanical ventilation during their ICU stay during our pilot study. The range of diagnoses and other demographic information are summarized in Tables 1 and 2. In trying to decrease the incidence of HAIs, the first challenge is accurate measurement [8]—to understand the magnitude of the problem, to benchmark performance against other institutions with comparable patient populations, to identify specific aspects of care that are the most problematic, and to measure changes in outcome in response to interventions. Although administrative data are the most readily available, these have well-appreciated limitations of sensitivity and specificity—limitations that become increasingly confounding the smaller the sub-population being studied. For a number of years, our institution used the reproducible and explicitly defined Nosocomial Infection Marker (NIM) methodology (CareFusion Inc., San Diego CA) as a surrogate measure of pulmonary and urinary infections. This method is probably more sensitive than specific, but as an objectively defined laboratory-test-based approach, it is internally consistent and sufficiently reproducible to permit temporal trending and interinstitutional comparisons. Table 2 Patient demographics: neuro ICU, on ventilator Prospective N

Ventilator (n = 393)

393

Male/female

198/195

Age (mean ± SD)

61.7 ± 15.7

Deaths (expected)

137 (132)

Mortality %; O:E

35 %; 1.12

All neuro ICU NHSN respiratory

58 (3 %)

NHSN CAUTI

107 (10 %)

NHSN National Healthcare Safety Network, CAUTI catheter-associated urinary tract infection

The major limitations we encountered were a lack of diagnostic specificity and of sufficiently detailed comparator data, making benchmarking challenging at best. It was only when we turned to the more specific Centers for Disease Control and Prevention (CDC)/National Healthcare Safety Network (NHSN) definitions and processes that we were able to measure our infection rates with sufficient specificity and compare rates in our institution to similar ones elsewhere. Although, as with any epidemiologic definition, the NHSN approach sacrifices some sensitivity, it is reasonably specific—although imperfect [9]—that notwithstanding the highly specific case definitions for catheter-associated urinary tract infections (CAUTIs) and ventilator-associated events (VAEs) and pneumonias, allowed us to identify areas in which we had the greatest opportunities for improvement, and then track the results of our interventions over time.

Methods To better understand factors contributing to HAIs in our institution, we first retrospectively reviewed outcomes in all patients (a) aged 18 or older; (b) admitted to our neuro ICU; (c) between January 1, 2011 and May 31, 2014; (d) with primary neurologic diagnoses; and (e) who required intubation and ventilator support. We reviewed the data regarding respiratory and urinary tract infections (UTIs) in this population, using NIM markers from 2011 to 2014 as a surrogate measure. This automated, algorithm-driven approach allows for identification of hospital-acquired infections without requiring systematic chart review [10, 11], and has been shown to be more accurate than administrative data. Case identification begins with microbiology laboratory findings; excludes (a) infections present on admission, (b) organisms usually present as a result of contamination, and (c) duplicate specimens; and has been validated particularly as a measure of urinary and pulmonary infections. Beginning in January 2013, NHSN-defined CAUTIs, VAEs, and probable/possible ventilator-associated

123

Neurocrit Care

pneumonias (PVAPs) were tracked prospectively by the hospitalâ&#x20AC;&#x2122;s Infection Prevention nurses, using strict NHSN definitions. The current NHSN criteria for CAUTIs require a culture identifying no more than two different bacterial pathogens with a colony count of at least 100,000/ml [12] for at least one of the pathogens, in a patient with other symptoms of a UTI or fever, and a urinary catheter that has been in place for more than two calendar days (or was removed within a day prior to recognition of the infection). Prior to a 2015 definition update, cultures with nonbacterial pathogens were included as were colony counts <100,000. To assure consistency, we applied the 2015 criteria retrospectively to our CAUTI data from January 2013 onward, although we found this affected only a very small number of cases. Under the current NHSN rubric [12, 13], VAEs are divided into ventilator-associated conditions (VACs), defined as a period of worsened oxygenation in a patient whose ventilatory status was previously stable; infectious VAC (IVAC), defined as a VAC with evidence of infection such as elevated white blood cell count or fever with implementation of antimicrobial therapy; and PVAP, defined as an IVAC with purulent sputum or compelling laboratory evidence of a respiratory pathogen. As we reviewed different potential contributors to HAIs in our institution, we recognized that our patients require frequent transports from the neuro ICU to Radiology for neuroimaging. In our hospital, this requires transfer to a stretcher, transport of some distance including an elevator ride, and then transfer to and from the scanner table. Each step provides an opportunity for unintentional manipulation of the endotracheal tube or urinary catheter. Despite best efforts during transport, patients do not invariably have the head of the stretcher elevated optimally; and the height of the urinary catheter relative to the bladder can fluctuate. Therefore, we specifically tested whether there was a relationship between HAIs, as reflected in urinary and respiratory NIMs, and the frequency of transport in and out of the neuro ICU for CT scans, independent of the duration of intubation or ICU stay. We reviewed the number of scans each patient had during the period prior to their positive NIM, then stratified patients by number of days intubated and, for each duration, calculated the average number of CT scans among those patients who developed a NIM and those who did not (Fig. 1). Since we had systematic NHSN data from 2013 onwards, and NIM data through December 2014, we were able to assess the relationship between these two quite different measures. From 2013 onward, we measured the impact of our interventions using NHSN-defined CAUTIs, VAEs, and PVAPs.

Interventions In July 2014, we began an extensive review of our utilization of urinary catheters, including the indications for insertion and discontinuation [6], and the techniques used both to insert and maintain them. Since we were unable to identify specific contributory factors to our high rate of CAUTIs, we re-educated all caregivers in, and universally adopted, the Ann Arbor criteria for appropriateness of catheter placement and removal [14] and assured that daily rounds in the ICU always include questioning the continued need for every urinary catheter. We obtained new Foley kits (BARD SURESTEPTM) that simplified and helped standardize the steps involved in site preparation and catheter insertion, and we changed to routinely use 14 French catheters instead of 16. We found that some catheters were being inserted in the operating room or intervention suite. We changed the procedures at those sites to match those in use in the neuro ICU. We also trained personnel at those sites to remove catheters before sending the patient to the neuro ICU, unless there were a compelling need to continue them. All neuro ICU nursing staff were required to retrain in Foley catheter insertion, including both watching a video and then performing an observed simulation. All RNs and patient care technicians were retrained in soap and water perineum care, which was then done every (12-h) shift on all patients with indwelling catheters. Finally, processes were implemented to assure real-time feedback to the neuro ICU staffâ&#x20AC;&#x201D;including immediate notification and dissemination of all positive culture findings and maintaining a calendar in the neuro ICU indicating the number of days since the last CAUTI or other HAI. Following all requisite training, this process was fully implemented in August 2014. In late December 2014, when our preliminary data suggested that patient transports out of the unit were contributing to the problem, and the impact of the enhanced Foley protocol was still suboptimal, we placed a mobile CT in the neuro ICU to perform routine head CTs at the bedside. By February 2015, this was being used routinely for approximately half of all neuro ICU head CTs, with Nursing and Radiology performing morning imaging rounds for 2 h early every day. Urgent or more complex scans, including CT angiography, continued to be performed in the imaging department; but these now represent a minority of all imaging in this population. Statistical Analysis For our initial exploratory analysis, we used hospital administrative data to collect demographic information as well as the numbers and timing of scans, combining this with the NIMs data. We used University HealthSystem

123

Neurocrit Care

R NIM+

18 16

Ave CTs/patient

Fig. 1 Horizontal axis days on ventilator before positive respiratory NIM. Vertical axis, average # of CT scans/patient among patients with a + respiratory NIM versus those with negative respiratory NIM. While average number of CTâ&#x20AC;&#x2122;s/patient was similar overall in the 2 groups (mean 3.3 vs. 3.5 scans/patient, positive NIM vs. negative), there is an apparent divergence after 2 weeks (mean number of CTs/patient 11.3 vs. 6.4, positive vs. negative)

R NIM-

14 12 10 8 6 4 2 0

Consortium (UHC now VizientTM) data for high-level external benchmarking. Descriptive statistics and comparisons were performed using StatPlus:mac Pro 6.0.3; (AnalystSoft Inc., Walnut, CA). Continuous variables were described as mean with 95 % confidence intervals. Statistical comparisons were performed using 2-tailed t tests (continuous variables) or chi-square tests [2] (dichotomous variables). Our prespecified primary outcome measures were numbers of CAUTIs, CAUTIs/1000 Foley days, and numbers of VAEs.

Results From the outset, we recognized that the incidence of HAIs in our population was high, but we initially attributed this to our neuro ICU population being atypical. Although our risk-adjusted length of hospital stay (LOS) for these patients was indistinguishable from that in comparable UHC hospitals, our mean neuro ICU LOS was 1.3 days (29 %) longer; and our expected mortality for this population was 11.2 % compared to 7.7 % in the comparator group. Among patients with neuro ICU LOS of 2 or more days, 18 % of our patients had parenchymal hemorrhagic strokes (vs. 12 % in the comparison group), 17 % aneurysm repairs (vs. 11 %), 13 % medically managed ischemic strokes (vs. 12 %), and 4 % mechanical interventions for ischemic strokes (vs. 2 %). Taken together, this suggested our population was considerably more ill and at greater risk of all complications. When we analyzed HAIs retrospectively, we found that, in the study period, 458 patients required ventilator

9 10 11 12 13 14 15 16 17 18 19 20 21 30

Ventilator days pre R NIM+

support; 144 of these required intubation for 7 or more days. Eighty-four developed positive respiratory NIMs (18 %); on average, these patients underwent 3.5 head CTs prior to the NIM; and 108 developed positive urinary NIMs (23 %), with 3.9 CTs each on average prior to the NIM. We reviewed the occurrence of positive respiratory and urinary NIMs in relation both to the duration of intubation and to the number of transports off the unit for CT scans. In exploratory univariate analyses, we found that the total number of scans performed during the hospitalization was significantly greater in patients who developed positive respiratory or urinary NIMs than in those who did not (Table 3; p < 0.001 for each). The average number of scans performed prior to the development of respiratory (3.5) and urinary (3.9) NIMs was greater than the total 3.4 (NS) and 3.3 (p = 0.04) number of scans, respectively, obtained throughout the entire hospitalization of patients who never developed respiratory and urinary NIMs. Since this is potentially confounded by the fact that patients with positive NIMs had longer LOS and more prolonged intubation, we focused the analysis on patients who developed NIMs while still on the ventilator, stratifying patients by the number of days intubatedâ&#x20AC;&#x201D;either in total without a NIM or preceding a respiratory NIM (Fig. 1). Overall, the difference in mean number of scans was not statistically significant. However, inspection of the distribution of numbers of scans suggests that while scan frequency was comparable for patients intubated up to 2 weeks, beyond this time, the number of scans in patients with positive NIMs was greater than that in those with negative ones, although the sample size was too small for this to be statistically significant (p = 0.13).

123 17

Neurocrit Care Table 3 Retrospective data analysis of neuro ICU patients requiring mechanical ventilation, admitted between January 2011 and May 2015 Urinary NIMs Positive

Respiratory NIMs Negative

Positive

Negative

108 (23 %)

352

84 (18 %)

376

Age (mean, CI)

58.7 (58.6–61.8)

61.1 (59.4–62.9)

NS (p = 0.19)

58.7 (55.5–62.0)

61. 0 (59.2–62.7)

NS (p = 0.26)

NIMs(M/F)

31/76 (29 %)

189/173

0.0002 v2

37/47

179/197

NS p = 0.56

NHSN + (M/F)

9/26 (2 %)

207/218

0.009 v

8/10

208/234

NS p = 0.83

LOS

28.1 (25.6–30.6)

12.0 (10.6–13.4)

<0.0001

32.3 (26.2–38.5)

13.1 (11.9–14.2)

<0.0001

Days pre NIM

9.7 (8.0–11.5)

9.4 (6.6–12.1)

Vent days

7.9 (6.3–9.6)

5.4 (4.8–5.9)

0.0002

15.1 (12.9–17.2)

4.6 (4.1–5.1)

<0.0001

Observed mortality

12.9 % (2.8–14.7)

45.7 % (41.1–51.5)

<0.0001

39 % (29–50)

38 % (33–43)

NS p = 0.80

Mortality O:E

0.71

1.36

1.27

# CTs

6.7 (5.8–7.6)

3.3 (3.0–3.5)

<0.0001

7.2 (6.0–8.3)

3.4 (3.1–3.7)

CTs pre NIM

3.9 (3.4–4.7)

0.04

3.5 (2.8–4.2)

<0.0001 NS

(Denominators slightly different between urinary and respiratory datasets due to incomplete data on some patients). Comparison by v2 for dichotomous variables, 2-tailed t test for continuous ones NIM nosocomial infection marker, NHSN National Healthcare Safety Network, LOS length of hospital stay

Among patients for whom we had both NIM and NHSN data (2013 and 2014 data; 463 patients), urinary NIMs were positive in 108, versus 36 NHSN CAUTIs; respiratory NIMs were positive in 84, versus 19. Compared to NHSNbased diagnoses, urinary NIMs were 94 % sensitive and 83 % specific; respiratory 58 % and 84 %, respectively. CAUTIs, as measured by either urinary NIMS or NHSN criteria, were about three times as frequent among women as among men, a finding that was highly statistically significant (Table 3). In contrast, men and women were equally likely to have a positive respiratory NIM. We then prospectively tracked NHSN data as we implemented our corrective actions (Fig. 2). In 2013–2014, we averaged 1.8 VAEs/month, a third of which were PVAPs, with 24 VAEs and 3 PVAPs in 2013 and 17 and 10, respectively, in 2014. In late December 2014, we placed the mobile CT in the neuro ICU and spent the next 8 weeks implementing protocols for its regular use. By mid-February, we were routinely performing half of all neuro ICU head CTs at the bedside, a proportion that has been maintained ever since. In January 2015, we had 4 VAEs including 1 PVAP; in February, we had 1 VAC. For the remainder of 2015, we had 10 additional VAEs, 3 of which were PVAPs (total for the year 15 VAEs, 4 PVAPs). We had no VAEs in 2 of the first 3 months in 2016, with 2 VACs and no PVAPs in the third. From January 2015 through March 2016, we had 6 months with zero VAEs, more than the total VAE-free months in the prior 2 years combined. Comparing the period from 2013 through January 2015 to that from February 2015 through March 2016, the number of VAEs/month decreased from 1.8 (95 % CI, 1.3 %–2.4 %) to 0.9 (95 % CI, 0.4 %–1.5 %) (p = 0.035, 2-tailed t test); the number of PVAPs/month declined by

more than 50 %, from 0.56 (95 % CI, 0.2 %–0.9 %) to 0.21 (95 % CI, -0.03 % to 0.45 %; NS). The August 2014 re-education program was followed by an immediate decline in Foley utilization but an increase in CAUTIs/1000 Foley days with no net change in CAUTIs/month. The monthly data (Fig. 3; Table 4) indicate an overall decline in CAUTIs starting in the first few months of 2015, following the introduction of the mobile CT. Overall (Table 4), the number of CAUTIs/month, and the rate per 1000 Foley days, decreased from a baseline of 2.4/month to 0.8/month (p < 0.001, 2-tailed t test) and 10.9 CAUTIs/1000 Foley days to 6.2 (p = 0.04, 2-tailed t test), the latter corresponding to a standardized infection ratio (SIR, NHSN; compared to all neurosurgical ICUs) of under 1.0. The number of Foley days/month decreased from 222 (95 % CI, 209.0 %–236.2 %) to 182 (95 % CI, 149.2 %–214.1 %) (p = 0.006, 2-tailed t test) for the period from Aug 2014 through January 2015, followed by a further reduction to 119 (95 % CI, 97.1 %–141.5 %) (p < 0.003, 2-tailed t test) compared to the prior 6 months. Together, these two effects resulted in an overall reduction in monthly CAUTIs by 2/3.

Discussion Critically ill patients with severe central nervous system compromise, typically intubated with indwelling intravenous and urinary catheters, are at considerable risk of nosocomial infections. In this study, we retrospectively reviewed our experience with these HAIs, identified and implemented interventions, and prospectively assessed their impact. In the process, we assessed the relative

123

Neurocrit Care Fig. 2 Monthly NHSN VAEs and PVAPs (2015 definitions). Notably in 2013, we had 24 VAEs (3 PVAPs), 17 in 2014 (10 PVAPs), and 15 in 2015 (4 PVAPs). In Q1 2016, we had 2 VAEs but no PVAPs. NHSN National Healthcare Safety Network, VAE ventilator-associated event, PVAP probable/possible ventilator-associated pneumonia

PVAP

VAE

4 3 2 1 0

25 20

15 16

10 5 0

8 10 10

12 8

9 6

Foley Focus ↓ 22

6 10

Mobile CT ↓

800 700 600 500

12 11

7 3

7 5

CAUTI/Q

300 200

2012 2012 2012 2012 2013 2013 2013 2013 2014 2014 2014 2014 2015 2015 2015 2015 2016 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1

Foley days

400

100 0

CAUTI/1,000 FD

Fig. 3 Quarterly NHSN catheter-associated urinary tract infections, Foley days/quarter, and CAUTIs/1000 Foley days. The increase in CAUTIs in Q3 2014 was driven by a spike of 5 in September, the first month we fully implemented the new catheter care procedures. Prior to implementing this protocol, we averaged 3.2 CAUTIs monthly in 2014. For the remainder of 2014, we had 2 in each of October and

November, 3 in December. In 2015, we had 2 in March and May, none in February, June, September, or December, 1 in each of the remaining months. There were 2 in January 2016; none in February or March. NHSN National Healthcare Safety Network, CAUTI catheterassociated urinary tract infection

sensitivity and specificity of two different objective measures of urinary and respiratory HAIs—NIMs and NHSN surveillance techniques. To our knowledge, the relationship between HAIs and patient transports has not previously been investigated. Since patients requiring multiple CT scans are typically the same ones requiring prolonged intubation, it would be reasonable to assume that the relationship between number of scans and frequency of infections simply reflects the fact that both occur in patients requiring prolonged immobilization in the ICU. Because of the limitations of the administrative dataset that we needed

to use in our preliminary analysis, our initial findings were ambiguous on this subject, but did suggest a potential association. On the one hand, patients with positive urinary or respiratory NIMs had, on average, twice as many scans as those who had negative NIMs (p < 0.001). On the other hand, they were on ventilators substantially longer. Focusing on the number of CTs obtained prior to the date of the positive NIM was confounded by an inability to determine the appropriate period of time to review in the comparison group of patients with negative NIMs, making a straightforward comparison challenging.

123

Neurocrit Care Table 4 NHSN infection rates PVAP/month

VAE/month 1.80 (1.25–2.35)

Jan 2013–Jan 2015

0.56 (0.22–0.89)

Feb 2015–Mar 2016

0.21 (-0.03–0.45)

0.93 (0.40–1.46)

t test

p = 0.15

p = 0.035

CAUTIs/month

Foley days/m

CAUTI/1000 FD

Jan 2013–Jul 2014

2.42 (1.79–3.05)

222.6 (209.0–236.2)

10.85 (87.99–13.71)

Aug 2014–Jan-2015

2.67 (1.23–4.10)

181.7 (149.2–214.1)

15.2 (5.93–24.42)

Jan 2013–Jan 2015 Feb 2015–Mar 2016

2.56 2.01–3.11) 0.79 (0.32–1.25)

212.8 (199.0–225.6) 119.3 (97.1–141.5)

12.2(9.4–15.1) 6.2 (2.5–9.9)

t test, period 1 vs. 2

0.69

0.006

0.18

t test period 2 vs. 3

0.001

0.003

0.02

t test period 1 vs. 3

0.003

<0.001

0.04

t test (1&2 vs. 3))

p < 0.0001

0.001

0.015

Baseline for VAE was January 2013 through January 2015, for CAUTIs from October 2013 to July 2014. Intermediate period is from August to January 2015. Current period for both is January 2015 through March 2016. Values are mean (95 % confidence interval). The numbers of VAEs and PVAPs decreased substantially following implementation of the mobile CT. The number of CAUTIs was impacted both by the 46 % decline in Foley days and the decrease in CAUTIs/Foley day. Much of the decline in the number of CAUTIs/1000 Foley days occurred following implementation of the mobile CT. Since there was no decrease in CAUTIs/1000 Foley days in the August 2014–January 2015 period, in the third line of CAUTI data, the entire period prior to January 2015 was pooled as the baseline

Given the ambiguous, if suggestive, nature of the preliminary data, and the fact that our HAI rates remained higher than we wished, we proceeded with two separate initiatives. First, we focused on decreasing CAUTIs, by implementing the Ann Arbor recommendations on urinary catheter utilization and decreasing Foley days by 46 % compared to baseline. At the same time, we retrained all involved in Foley catheter placement and maintenance. Several months later, we proceeded with implementation of bedside CT scans in the neuro ICU, primarily in the hope of decreasing VAEs. With this, the decreases in both VAEs—and CAUTIs have been both unambiguous and sustained. We conclude that a multidisciplinary approach to care of the neurocritically ill patient—combining standard ventilator bundles with enhanced attention to urinary catheter maintenance and strict adherence to the Ann Arbor criteria limiting Foley catheter utilization as well as limiting avoidable transports of the patient in and out of the ICU— can result in a significant decrease in the prevalence of HAIs. Compliance with Ethical Standards Conflict of Interest The authors declare that they have no conflicts of interest.

References 1. Klompas M, Branson R, Eichenwald EC, et al. Strategies to prevent ventilator-associated pneumonia in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35:915–36.

2. Johnson M, Rocque B, Kamps T, Medow J. Reduction of ventilator-associated pneumonia in the Neuroscience Intensive Care Unit: a multimodality prevention and testing protocol. J Neurosurg Sci. 2013;57:293–6. 3. Tominaga GT, Dhupa A, McAllister SM, Calara R, Peters SA, Stuck A. Eliminating catheter-associated urinary tract infections in the intensive care unit: is it an attainable goal? Am J Surg 2014;208:1065–1070; discussion 1069–1070. 4. Lo E, Nicolle LE, Coffin SE, et al. Strategies to prevent catheterassociated urinary tract infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35:464–79. 5. Hayashi Y, Morisawa K, Klompas M, et al. Toward improved surveillance: the impact of ventilator-associated complications on length of stay and antibiotic use in patients in intensive care units. Clin Infect Dis. 2013;56:471–7. 6. Hooton TM, Bradley SF, Cardenas DD, et al. Diagnosis, prevention, and treatment of catheter-associated urinary tract infection in adults: 2009 international clinical practice guidelines from the Infectious Diseases Society of America. Clin Infect Dis. 2010;50:625–63. 7. Yokoe DS, Anderson DJ, Berenholtz SM, et al. A compendium of strategies to prevent healthcare-associated infections in acute care hospitals: 2014 updates. Infect Control Hosp Epidemiol. 2014;35:967–77. 8. Stamm AM, Bettacchi CJ. A comparison of 3 metrics to identify health care-associated infections. Am J Infect Control. 2012;40:688–91. 9. Tedja R, Wentink J, O’Horo JC, Thompson R, Sampathkumar P. Catheter-associated urinary tract infections in intensive care unit patients. Infect Control Hosp Epidemiol. 2015;36:1330–4. 10. Brossette SE, Hacek DM, Gavin PJ, et al. A laboratory-based, hospital-wide, electronic marker for nosocomial infection: the future of infection control surveillance? Am J Clin Pathol. 2006;125:34–9. 11. Peterson KE, Hacek DM, Robicsek A, Thomson RB Jr, Peterson LR. Electronic surveillance for infectious disease trend analysis following a quality improvement intervention. Infect Control Hosp Epidemiol. 2012;33:790–5.

123

Neurocrit Care 12. CDC. Urinary tract infection (catheter-associated urinary tract infection [CAUTI] and non-catheter-associated urinary tract infection [UTI]) and other urinary system infection [USI]) events. http://www.cdc.gov/nhsn/pdfs/pscmanual/7psccauticurrent.pdf. Accessed 17 Jan 2016. 13. Magill SS, Klompas M, Balk R, et al. Developing a new, national approach to surveillance for ventilator-associated events: executive summary. Clin Infect Dis. 2013;57:1742â&#x20AC;&#x201C;6.

14. Meddings J, Saint S, Fowler KE, et al. The Ann Arbor criteria for appropriate urinary catheter use in hospitalized medical patients: results obtained by using the RAND/UCLA appropriateness method. Ann Int Med. 2015;162:S1â&#x20AC;&#x201C;34.

123

March/April 2008 volume 30 number 2

The economic annddiniiacal bfits of portable hed/

The Journal of the American Healthcare Radiology Administrators

neck CTimaging inthe intensive care unit

The Impact Leadership Has on Training and Change By Simon Lia, BA(hons), MS, Sally Grady, RT, and Larry Peters, PhD

Personality Assessments as a Workforce Diversity Tool By Sandra K. Collins, MBA, Brandy Sord, Caleigh Griffin, and Lora Borges

Critical Perspectives on Cultural Competence as a Strategic Opportunity for Achieving High Performance in Healthcare Organizations By Philippa Strelitz, PhD, and Kerry Watson, MA

The Economic and Clinical Benefits of Portable Head/Neck CT Imaging in the Intensive Care Unit By Thomas Masaryk, MD, Renee Kolonick, RT(R)(CT), Tracy Painter, RT(R)(CT), and David B. Weinreb, MD

E U

R Y

• There is a 13% morbidity associated with transporting critically-ill patients outside of the ICU. The incidence of adverse events during transport specifically for CT imaging is as high as 71%. The objective of this study was to assess the feasibility and cost-effectiveness of a portable CT scanner designated to perform bedside imaging in the ICU.

• A fully mobile 8-slice head/neck CT scanner was evaluated for efficiency and personnel allocation. The return-on-investment for the purchase of the portable scanner was calculated. • Data demonstrates that the introduction of a portable CT scanner in the ICU is feasible and cost-effective. At the Cleveland Clinic in Mayfield Heights, Ohio, the portable scanner provided a full returnon-investment within the first 6.9 months of its operation, an internal rate of return of 169%, and a 5 year expected economic benefit of $2,619,290.

MARCH/APRIL

2008

T imaging has become an increasingly valuable diagnostic modality for Intensive Care Unit (ICU) patients, and radiographic findings often significantly impact clinical management. However, many clinicians may be reluctant to perform CT imaging on hemodynamically unstable ICU patients, as there are both risks and logistical difficulties associated with the transport of such patients.1–3 One study has demonstrated a 13% morbidity associated with the transport of ICU patients to the operating room.4 Another prospective study demonstrated that adverse events, defined as the disconnection on monitoring equipment, interruption of vasoactive medication drips, or unintentional extubations, occurred during one-third of all transports outside of the ICU.5 More relevantly, transport specifically to the CT suite was associated with a 71% incidence of adverse events.5 To address concerns for patient safety, resident physicians, nurses, and respiratory therapists accompany ICU patients during transport. As many as 5 staff members may be required for transport and the nurse-to-patient ratio in the ICU may increase from 2:1 to 2.5:1 as a consequence. This is burdensome on staff members not involved in the transport, since they need to care for more patients while their colleagues are in the radiology suite. Mean transport times for CT studies as high as 81 minutes have been reported.6 However, at our institution, the time required for transport from the ICU to the completion of scan is, on average, 50 minutes. The CT department at the Cleveland Clinic in Mayfield Heights, Ohio performs more than 6000 inpatient head CT studies annually, of which 2901 were performed on ICU patients. CT imaging of ICU patients is associated with the aforementioned risks and is demanding on the ICU personnel who must assist with patient transport. The

R ADIOLOGY

MANAGEMENT

The Economic and Clinical Benefits of Portable Head/Neck CT Imaging in the ICU

purchase of a portable CT scanner for dedicated use in the ICU was proposed with the intention that it would: • Require fewer personnel to conduct each study. • Result in increased revenues as the conventional scanner would be dedicated for greater number of outpatient studies. • Diminish the risks of adverse events during patient transport. • Provide ICU physicians with clinically relevant imaging data more rapidly, thereby contributing to better patient care. The objective of this study is to assess the feasibility and cost-effectiveness of implementing the use of a portable CT scanner in the ICU.

Methods The CT department purchased an 8-slice portable head/neck scanner. The scanner, together with its associated drive system, cost $359,000. The portable scanner weighs approximately 700 pounds and is fully mobile. A single technician can transport the scanner to the patient’s bedside. The patient’s head is cantilevered off the end of the bed onto a

carbon graphite scan board. The scanner is designed to perform non-contrast CT, CT angiography, CT perfusion, and Xenon perfusion studies (Figure 1). Prior to the purchase of the scanner, a retrospective study was conducted to determine: • The number and type of inpatient head CT scans performed on ICU patients each year. • Mean number of personnel required for patient transport. • Mean time required for patient transport. • Mean times required to complete CT scans of ICU patients. As approximately 1700 (50%) of all ICU head CT studies were performed on patients in the neurology/neurosurgery ICU, the portable scanner would be dedicated for use in this unit. The personnel costs of patient transport were calculated based on the following yearly estimates of the salaries and benefits: $72,000 for ICU nurses; $72,000 for resident physicians; $60,000 for respiratory therapists; $60,000 for radiology technologists; and $24,000 for transport assistants.

(a)

(b)

Figure 1. (a) Neuro-ICU patient with an external ventricular drainage catheter, brain tissue oxygen monitor, jugular venous saturation catheter, endotracheal tube, feeding tube, arterial line, venous lines (for insulin, IV fluids, versed, and fentanyl), and potential cervical spine injury. (b) Portable CT scanning in the ICU takes 18 minutes and requires a single technologist, compared to 4Ð5 personnel required for transport to the radiology suite, which takes, on average, 50 minutes. Potential complications associated with patient transport are eliminated. The nursing staff remains in the ICU and the conventional scanners are available for other inpatient/outpatient purposes. Source: Photos courtesy of San Francisco General Hospital.

R ADIOLOGY

MANAGEMENT

MARCH/APRIL

2008

The Economic and Clinical Benefits of Portable Head/Neck CT Imaging in the ICU

2% Non-contrast CT Angiography CT Neck CT Perfusion 86%

Figure 2. Of the 6145 head CT studies performed on ICU patients prior to the introduction of the portable scanner, 5369 (86%) were non-contrast CT, 440 (7%) were CT angiography, 300 (5%) were CT neck, and 137 (2%) were CT perfusion studies.

Results Prior to the introduction of the portable scanner, 6145 head CT studies were performed. Of these, 5369 (86%) were non-contrast CT, 440 (7%) were CT angiography, 300 (5%) were CT neck, and 137 (2%) were CT perfusion studies. As a first-line implementation, the portable scanner performed only non-contrast studies, as this was the most commonly requested type of study for ICU patients (Figure 2). For studies performed on the conventional scanner, a minimum of 4–5 staff members (1 resident physician and/or 1–2 ICU nurses, 1 respiratory therapist, and 1 transport helper) were required for each transport. Based on the estimates of annual salaries, the mean personnel cost of each transport was $97. The personnel costs for the CT technologists required for conventional imaging was an additional $19 per scan. Between November 2006 and February 2007, 502 studies were performed on ICU patients with the portable scanner. The mean time elapsed from the request for a scan in the ICU to its completion was 18 minutes. The mean labor cost for the CT technologist was $9 per scan. For the 502 scans performed by the portable scanner, there were personnel cost savings achieved by: • Reducing the number of personnel involved in transport (4–5 for conventional imaging compared to 0 for portable imaging). • Reducing the number of CT technologists needed to operate the scanner (2 technologists for the conventional compared to 1 for the portable scanner). • Reducing the overall time required to complete each scan (50 minutes for conventional, 18 minutes for portable imaging). Based on the estimates of annual salaries, it was determined that these 502 scans would have required personnel costs of $58,689 if performed with conventional scanning. 4

MARCH/APRIL

2008

The portable scanner performed these scans with personnel costs of $4,518, resulting in an estimated cost savings of $54,171. Based on the data for the first 4 months, an annual cost savings of $162,512 is projected. Furthermore, the addition of the portable CT allows for the conventional scanner to perform a greater number of outpatient scans. During the 4 month period, an additional 394 outpatient studies were performed (revenue of $390.14 per scan). Thus, as the conventional scanner could be dedicated to perform a greater number of outpatient studies, there was additional revenue of $153,715 for this 4 month period. An annual increase in revenue of $461,145 is anticipated (Table 1). Based on actual data, the calculated break even point on investment is 6.9 months. The internal rate of return is 169%. The total economic benefit for 5 years is expected to be $2,619,290.

Discussion The present study evaluates the clinical utility of a portable CT scanner for head/neck imaging of ICU patients. The portable scanner performed 502 studies in a 4 month period, including CT angiography and perfusion. The risks of transporting critically ill patients to the radiology suite have been well-described in existing literature.1,2,4–7 Briefly, transport presents risks of interruption of vasoactive medication drips, accidental extubations, and, most significantly, the risk that patients may become hemodynamically unstable outside of the ICU where appropriate care cannot be immediately provided. Bedside imaging eliminates these risks of patient transport. Furthermore, as physicians recognize these risks, there are circumstances for which CT imaging may be delayed or deferred. The availability of a portable scanner would allow for neuroimaging in even the most unstable patients, for whom transport to the radiology suite is contraindicated. Bedside imaging may rapidly provide clinically relevant data in patients for whom CT imaging may otherwise have been R ADIOLOGY

MANAGEMENT

The Economic and Clinical Benefits of Portable Head/Neck CT Imaging in the ICU

R ADIOLOGY

MANAGEMENT

MARCH/APRIL

2008

The Economic and Clinical Benefits of Portable Head/Neck CT Imaging in the ICU

Bedside imaging may rapidly provide clinically relevant data in patients for whom CT imaging may otherwise have been unobtainable. unobtainable. Patients too hemodynamically unstable to be transported may be imaged rapidly at their point-of-care in the ICU with a portable scanner. There are several limitations of this study. First, the portable scanner was employed for only 502 of all inpatient head CT studies during the 4 month interval, which is approximately 25% of a 1 year estimate. Ideally, the study should evaluate the cost-effectiveness when the portable scanner is applied to perform a much higher percentage of the inpatient imaging demands. Second, due to space constraints in the ICU, the scanner had to be stored on a different floor of the hospital. Transport of the scanner and its accompanying workstation to the ICU required 2 technologists. Logically, storing the scanner in the ICU would further reduce the time needed to complete each study. The manufacturer has recognized this issue and developed a motorized transport system where the scanner is transported by a battery-operated drive system operated with a joystick controller. Initial trials of the motorized system have demonstrated that the portable scanner can be easily maneuvered by a single technologist. Another limitation of the portable scanner is that it requires manual timing for contrast scans, ie, CT angiography and perfusion. For patients with cardiac dysfunction and vascular stenoses, manual timing may adversely impact image quality. If delivery of contrast to the cerebral vasculature is delayed by cardiac dysfunction or stenoses, then CT images will be obtained before the peak phase of contrast. This would, in turn, result in suboptimal enhancement of the cerebral vasculature. In response to this technological limitation, the manufacturer has recently developed automatic bolus tracking software. The software images a vessel of interest at a low radiation dose to determine the density of the vessel lumen. When peak enhancement is detected in the vessel of interest, the software then triggers the scanner to begin imaging. This approach has been successfully implemented on conventional scanners. This software feature will be particularly useful in ICU patients with vasospasm, as well as other vascular pathologies often seen in cases of subarachnoid hemorrhage. Our institution will begin testing of the bolus tracking software in 2008.

Conclusion

MARCH/APRIL

References 1

Braman S, Dunn S, Amico C. Complications of intrahospital transport in critically ill patients. Ann Intern Med. 1987; 107: 469–473. 2 Weg J, Haas C. Safe intrahospital transport of critically ill ventilator dependent patients. Chest 1989; 96: 631–635. 3 McCunn M, Mirvis S, Reynolds N, Cottingham C. Physician utilization of a portable computed tomography scanner in the intensive care unit. Crit Care Med 2000. 28: 3808–3813. 4 Insel J, Weissman C, Kemper M, et al. Cardiovascular changes during transport of critically ill and postoperative patients. Crit Care Med. 1986; 14: 539–542. 5 Smith I, Fleming S, Cernaianu A. Mishaps during transport from the intensive care unit. Crit Care Med. 18(3): 278–281. 6 Indeck M, Peterson S, Smith J, et al. Risk, cost, and benefit of transporting ICU patients for special studies. J Trauma. 28(7): 1020–1025. 7 Hurst JM, Davis K Jr, Johnson DJ, et al. Cost and complications during in-hospital transport of critically ill patients: a prospective cohort study. J Trauma. 1992; 33(4): 582–5. Thomas J. Masaryk, MD, is the head of the section of neuroradiology at Cleveland Clinic. Renee M. Kolonick, RT(R)(CT), is the administrative director of radiology at Hillcrest Hospital in Mayfield Heights, Ohio. Tracy Painter, RT(R)(CT), is the clinical manager of the CT Imaging Institute of the Cleveland Clinic. David B. Weinreb, MD, is a resident physician in diagnostic radiology in Connecticut. Please address all correspondence to Dr. Weinreb at david.weinreb@mssm.edu.

ahra

In summary, the purchase of additional imaging technology must be considered in the context of the clinical benefits it may offer and the return-on-investment for the department. Clearly, the ability to image patients at their point-of-care in the ICU will facilitate rapid clinical decision-making and reduce the risks associated with transport. In addition, however, it has been demonstrated that, by reducing staff and 6

time requirement for transport, the portable scanner may achieve annual cost savings of $162,512. Equally as significant, having a scanner dedicated for ICU patients allows for the conventional scanner to perform an additional 1182 outpatient studies each year. Together taken, the introduction of the portable scanner may offer a net economic benefit of $264,658 in the first year of its operation and a total benefit over 5 years greater than $2,619,000.

2008

This article was published in the March/April 2008 issue of Radiology Management. It is reprinted with permission. Radiology Management is an official publication of AHRA. For membership and subscription information, please contact: AHRA, 490-B Boston Post Road, Sudbury, MA 01776 Phone: 800-334-2472, Email: info@ahraonline.org www.ahraonline.org

R ADIOLOGY

MANAGEMENT

The use of a portable head CT scanner in the intensive care unit

Volume 42

Number 2

April 2010

109

The Use of a Portable Head CT Scanner in the Intensive Care Unit Kaitlin Peace, Eileen Maloney Wilensky, Suzanne Frangos, Eileen MacMurtrie, Elizabeth Shields, Marianne Hujcs, Joshua Levine, Andrew Kofke, Wei Yang, Peter D. Le Roux

ABSTRACT Transport of critically ill intensive care unit (ICU) patients may be hazardous. In this study, we examined the use of a portable head CT scanner (CereTomA) in the ICU to assess its feasibility, safety, and radiological quality. Two hundred and twenty-five portable head CT scans were obtained from 114 patients (mean age = 57 T 18 years) treated in a neurosurgical intensive care unit at a university-based Level I trauma center. Patient radiological and ICU records were retrospectively reviewed. The vast majority of portable CT scans were performed after an intracranial procedure (24%) due to neurological deterioration (16%) or in routine follow-up (16%). Diagnostic quality was judged to be adequate, and no scans needed to be repeated because of poor quality. No scans were complicated by accidental disconnection of an intravenous line. In ventilated patients, there were no interruptions in mechanical ventilation and no inadvertent extubations. In addition, continuous intracranial monitoring, when in use, remained connected. The average total time to perform a portable head CT scan was 19.5 T 3.5 min. The actual scan time was 2.5 T 0.7 min. These results suggest that the portable CT scanner (CereTomA) is feasible, easy to use, and safe and provides adequate radiological quality for diagnostic decisions.

he management and prevention of secondary neuronal injury is important in modern neurocritical care. Follow-up head CT (HCT) scans are frequently used to identify patients at risk for secondary neuronal injury and have become a necessary part of the care of severely brain-injured patients and patients in intensive care unit (ICUs; Chang, Meeker, & Holland, 2006; Dharap, Khandkar, Pandey, & Sharma, 1990). Follow-up HCTs may be obtained emergently when there is a decline in neurological function or as a part of routine care and generally involve transport of severely ill patients from the ICU to the CT suite in the radiology department. Transport of a patient to and from the ICU is associated with risks, including unexpected events that

may compromise patient safety or outcome (Andrews, Piper, Dearden, & Miller, 1990; Bercault, Wolf, Runge, Fleury, & Boulain, 2005; Papson, Russel, & Taylor, 2007). Complications associated with intrahospital transport (IHT) may occur in as many as two thirds of patients. Approximately half of the incidents appear to be patient related and half are equipment related (Doring, Kerr, Lovasik, & Thayer, 1999; Lovell, Mudaliar, & Klineberg, 2001; Smith, Fleming, & Cernaianu, 2000; Waydhas, 2001). Furthermore, transporting patients out of the ICU may leave critically ill patients at risk in remote hospital locations where there are fewer resources or in circumstances where support needs to be delivered emergently. We have observed that IHT can adversely affect brain oxygen

Kaitlin Peace is a research assistant in the Neurosurgery Clinical Research Division, Department of Neurosurgery, The Hospital of the University of Pennsylvania, Philadelphia, PA.

Marianne Hujcs, MSN, is a clinical nurse specialist in the NeuroTrauma ICU, The Hospital of the University of Pennsylvania, Philadelphia, PA.

Eileen Maloney Wilensky, MSN ACNP-BC, is a director in the Neurosurgery Clinical Research Division and in the Mid-Level Provider Program, Department of Neurosurgery, The Hospital of the University of Pennsylvania, Philadelphia, PA.

Joshua Levine, MD, is an assistant professor of neurology at the Department of Stroke and Neurocritical Care Division, The Hospital of the University of Pennsylvania, University of Pennsylvania School of Medicine, Philadelphia, PA.

Suzanne Frangos, RN, is a research nurse in the Neurosurgery Clinical Research Division, Department of Neurosurgery, The Hospital of the University of Pennsylvania, Philadelphia, PA.

Andrew Kofke, MD, is a professor in the Departments of Neurosurgery, Neurology, and Anesthesiology and Critical Care, The Hospital of the University of Pennsylvania, University of Pennsylvania School of Medicine, Philadelphia, PA.

Questions or comments about this article may be directed to Eileen MacMurtrie, MSN ANP-BC, at macmurte@uphs.upenn. edu. She is a nurse practitioner in the Department of Neurosurgery, The Hospital of the University of Pennsylvania, Philadelphia, PA. Elizabeth Shields, BSRT(R)(CT), is lead CT scan technologist in the Department of Radiology, The Hospital of the University of Pennsylvania, Philadelphia, PA.

Wei Yang, PhD, is an instructor of biostatistics in the Department of Biostatistics and Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania School of Medicine, Philadelphia, PA. Peter D. Le Roux, MD, is an associate professor in the Department of Neurosurgery, The Hospital of the University of Pennsylvania, Philadelphia, PA. Copyright B 2010 American Association of Neuroscience Nurses

110

Journal of Neuroscience Nursing

Portable CT scanners that produce high-resolution, high-quality images comparable to those generated by nonportable machines are now available for routine use in intensive-care settings. (Swanson et al., in press). Transport of patients within the hospital to the radiology suite also is associated with logistical and safety issues and may require a significant amount of care provider time. Together, these various factors can increase the risk of secondary neuronal injury. Although patient risk can be reduced with increased staffing, careful planning, and use of appropriate equipment, unexpected complications remain common during IHT; when they do occur, they may be difficult to treat (Lahner et al., 2005; Warren, Fromm, Orr, Rotello, & Horst, 2004). Consequently, point-of-care testing may be appropriate for some patients because it may decrease the time needed for critical decision making, reduce adverse clinical events, and contribute to economic savings (Kendall, Reeves, & Clancy, 1998; Halpern et al., 1998). Portable HCT scanners allow an HCT to be performed at the patient’s bedside and are an excellent example of point-of-care testing that potentially reduces time delay in critical decision making and intervention and reduces the risks of IHT (Gunnarsson et al., 1999). Newer portable scanners that have become available in North America in recent years can now produce high-resolution, high-quality images that are comparable with standard scanners. In addition, portable scanners can now support nonenhanced and enhanced imaging, CT angiography (CTA), and bolus contrast perfusion as well as Xenon CT blood flow studies, increasing their versatility and use in the neuro ICU (NICU). A portable CT scanner recently was introduced to our ICUs. A descriptive review was performed to examine its use. The goals of this study were to examine feasibility, indications, radiological quality, staff involvement, time to perform a scan, and radiation safety.

Methods Patient Population Patients admitted to the Hospital of the University of Pennsylvania, a Level I trauma center and a certified Joint Commission Primary Stroke Center, who received portable CT scans within the NICU, trauma surgical ICU

(TSICU), or cardiothoracic surgical ICU (CTSICU) were included in this analysis. Approval for the study was obtained from the institutional review board; consent was waived. The patients were studied retrospectively between February 26, 2007, and June 21, 2007, a period that corresponded to the introduction of portable CT scanning to the ICUs and as required by the institution’s New Technology Committee to examine the feasibility of portable CT scanning. Patients underwent a portable HCT scan at the discretion of their treating physician on the basis of clinical presentation and necessity. The treatment protocol for portable HCT scans at our center includes the following: (a) a neurological decline after a craniotomy, (b) a suspected postoperative hemorrhage, (c) follow-up imaging after acute traumatic brain injury, and (d) an ICU patient who experiences acute neurological decline and is a high transport risk, for example, mechanical ventilation, administration of vasopressors, or cerebral and/or hemodynamic monitoring.

Portable HCT Scan Patients were examined using the CereTom portable eight-slice CT scanner (NeuroLogica Corporation, Danvers, MA), a high-speed, lightweight device that runs on batteries and is charged from a standard threeprong (100y) outlet. The following parameters were used for this study: 120 kV, 7 mA, standard sharpness, and standard resolution (4-s scan), CTDIw = 80 mGy. Three 5-mm axial sections were acquired per 4-s scan to a maximum of 46 images total, and an additional reconstruction at 2.5-mm intervals was typically performed after initial data acquisition. Since this study was completed, we now use low-dose resolution (2-s scan) to decrease radiation exposure. Although this increases image noise, it is still sufficient for image interpretation in the ICU. The scanner also is capable of performing CTA and bolus contrast CT perfusion studies as well as Xenon CT perfusion. The scanner was available for use Sunday through Saturday between the hours of 7:00 a.m. and 3:30 p.m., when CT technical support was available in the ICU. A

Bedside Scan Preparation The charge nurse and the CT technologists developed a patient order list for portable CT scans after morning rounds. Two technologists then retrieved the portable CT scanner from the ICU storeroom. The patient’s data were loaded onto the acquisition screen of a laptop connected remotely to the portable scanner, and the scanner was brought to the patient’s room. The ICU nurse assisted with patient set up, which required the portable scanner to be placed according to the furniture arrangement in each individual room to be minimally intrusive. The patient’s head of bed was

Volume 42

oriented toward the ventilator to reduce the risk of extubation. Similarly, intravenous pumps and tubing, multimodality brain monitors, ventilator tubing, Foley catheters, and lower leg compression devices were carefully aligned to prevent inadvertent disconnection. The patient was placed in supine position, and a bed extension with a cradle for the head was attached to the bed. A lift system (GH2; Guldmann Inc., Tampa, FL) was used to lift the patient and lower them onto the bed extension (Figure 1A). The scanner gantry was lined up with the cradle and slid into place up to the cervical spine. The scanner position was checked with a laser, and a lead shield was placed over the open end of the device. Scans were initiated from the laptop outside the patient’s room. The radiologic technicians (RTs) informed all ICU staff in the immediate area that a CT scan was about to begin, and all staff exited the patient’s room. The scanner performed two slices per rotation then automatically slid into place for the next series of slices. Neurointensivists were able to view images in real time from the laptop and make immediate management decisions. The images were then uploaded

FIGURE 1 Portable CT Scanners in Use

Number 2

April 2010

111

directly to the hospital’s picture archiving and communication system after study completion. The scanner was moved out of the room, and the patient’s bed was returned to its normal position. Intravenous lines connected to the patient were checked for entanglement or disconnection. The scanner was readied for the next patient or returned to the storeroom for recharging.

Data Collection Data for each scan were retrieved from the portable CT scan log. Clinical information including gender, age, admission Glasgow Coma Scale (GCS), admission diagnosis, GCS at the time of scan, rationale for the scan, and scan outcome was collected retrospectively from ICU flow sheets completed by nursing staff. For this study, individual patient scans were reviewed in the picture archiving and communication system by the treating physicians, judged for diagnostic quality, and compared with the original report generated by the radiologists at the time of the scan.

ICU Staff Survey ICU nursing staffs, physicians, and RTs from the NICU, CTSICU, and TSICU were surveyed about portable CT scanner use. Each member of the ICU staff and the CT portable technician team were asked to complete a seven-question survey. Healthcare providers were made aware of the survey through a mass e-mail. The survey was conducted over 10 days.

Results Patient Population Between February 26, 2007, and June 21, 2007, 225 portable HCT scans were obtained from 114 patients. This represented 27% of all HCT scans performed on NICU, CTSICU, or TSICU patients during the same period. There were 65 men and 49 women with a mean age of 57 T 18 years. Table 1 lists patients’ admission GCS and diagnosis. One hundred and six patients (93%) had undergone a surgical procedure.

Patients and CT Scans

Note. (a) ICU patient being lifted into the portable CT scanner while he remains connected to his monitors, ventriculostomy, and multiple infusion pumps. (b) A patient with multiple intracranial monitors that fit easily into the portable HCT scanner.

Seventy-three patients (64%) received one portable HCT scan, 18 patients (16%) received two scans, 9 patients (8%) received three scans, and 14 patients (12%) received more than three scans during their ICU stay. Four Xenon CT perfusion scans were performed using the portable scanner during the study period. The clinical indications for the portable HCT scans are listed in Table 2. Intracranial monitors used to measure intracranial pressure including ventriculostomies or Camino intracranial pressure monitors (Integra Neuroscience, Plainsboro NJ), brain oxygen (LICOX, Integra Neuroscience), microdialysis catheters (CMA-70 microdialysis probes and CMA 106

112

Journal of Neuroscience Nursing

TABLE 1.

Admission Clinical and Diagnostic Characteristics of the Patients Included in This Analysis (N = 114)

Admission Diagnosis Neurologic

Cardiac

Patients

GCS 3Y8

Subarachnoid hemorrhage

Tumor

Traumatic brain injury

Subdural hematoma

GCS 9Y12

GCS 13Y15

4 Y

8 14

Stroke

Intracerebral hemorrhage

Intraparenchymal hemorrhage

Encephalitis

Amyloid bleed

1 Y

Y 1

Arterial venous malformation

Cerebral aneurysm

Hydrocephalus

Left hemifacial spasm

Status epilepticus

Vertebral artery dissection

1 1

Y Y

Congestive heart failure

Aortic dissection

Y Y 4

Myocardial infarction

Coronary artery disease

Thoracic aortic aneurysm

Other (cardiac)

114

Other Total

Note. GCS = Glasgow Comal Scale.

perfusion pump; CMA, Stockholm, Sweden), and scalp electrodes for continuous electroencephalogram easily fit into the portable CT scan opening (Figure 1B). During portable HCT scanning, continuous intracranial and systemic monitoring was continued without interruption in each patient. No scans were complicated by accidental disconnection of an intravenous line or disconnection from mechanical ventilation.

CT Scans The average total time to perform a portable HCT scan (from CT scan setup by the technician, patient preparation, to completion of the scan and equipment removal) was 19.5 min and ranged from 17 to 22 min. The median actual scan time was 2.5 min (range = 2Y3 min). The remaining time was required for room and patient setup, and this usually required two RTs and one ICU nurse. Five patients can be scanned consecutively before the portable scanner loses power. However, if the device is plugged into a standard electrical outlet between scans, more patients can be scanned in a given period.

Radiation Radiation was emitted from the portable scanner at an angle of 45- and spread outward 10 ft. During a scan, the radiation dose to the patient’s head was 0.025 2Sv. Measurements taken at 6 and 10 ft away in front of the scanner demonstrated radiation exposure free-inair of 50 and 139 2R, respectively (129 kVp at 7 mA for 3 min). Radiation badges worn the first year demonstrated that staff were not exposed to radiation; thus, staff are no longer required to wear the badges.

Radiological Findings The radiological findings described at the time of the scan are listed in Table 3. Management changes were made after 108 (48%) of the portable HCT scans that were obtained. Post hoc blinded review of all CT scans by an independent neurosurgeon or agreed with the original diagnosis described by the radiologist in 197 scans (88%). The differences were largely semantic, for example, ‘‘subacute versus evolving infarct.’’ None were clinically significant or altered management. Two hundred and sixteen scans (96%) were

Volume 42

TABLE 2.

Rationale Behind Each of the Portable HCT Scans Performed in This Analysis (N = 225)

Number 2

April 2010

was placed. The evolution of traumatic intracranial hemorrhage was easily assessed (Figure 2B).

Case 2

Reason for Scan

No. Patients (%)

After central nervous system surgery

54 (24)

Mental status change

37 (16)

Follow-up

36 (16)

Cerebral edema

24 (11)

Suspected hydrocephalus

18 (8)

Suspected bleed

17 (8)

Cerebral infarct localization

16 (7)

Ventriculostomy placement/ replacement

7 (3)

A 62-year-old woman was admitted to the hospital with an acute loss of consciousness and a left hemiparesis. Admission HCT revealed a large right centrum semiovale intracerebral hematoma and intraventricular hemorrhage (Figure 3A). A craniotomy was performed, the hematoma was partially evacuated, and a right frontal ventriculostomy was placed. A portable HCT scan was performed postoperatively (Figure 3B), documenting the partial evacuation of the hematoma, the location of the ventriculostomy, and the evolution of intraventricular hemorrhage and hydrocephalus.

Head trauma

6 (3)

Discussion

Licox placement/replacement

4 (2)

Suspected subdural hematoma

4 (2)

Aneurysm/after angiogram

1 (0.4)

Seizures

1 (0.4)

judged to be of adequate quality for diagnostic use. No patient required transport to the radiology suite because an image was unclear on portable CT.

Survey on Use of Portable HCT Scan in the ICU Seventy-three surveys were completed. Respondents included 13 RTs (18%) who support the portable CT program, 4 physician providers (5.5%), 4 nurse practitioners (5.5%), 2 respiratory therapists (2.7%), 48 ICU nurses (66%), 1 nurse manager (1.4%), and 1 certified nurseâ&#x20AC;&#x2122;s aid (1.4%). Twenty-one surveys (29%) were returned from the NICU, 19 (26%) were returned from the CTSICU, and 18 (25%) were returned from the TSICU. Two hundred and twentynine nurses work in these three ICUs; that is, 21% of possible nurse respondents replied. Two physicians and 13 RTs circled all three ICUs because they care for patients across the hospitalâ&#x20AC;&#x2122;s ICUs. The results of the survey are listed in Table 4.

In this descriptive review, we examined the initial use of a portable HCT scanner in the ICU. We reviewed the first 225 portable HCT scans and found that the scans could be performed on a routine basis, and in a short time, radiological quality was adequate and complications were rare. Physicians caring for the patients believed that the radiological quality was adequate for bedside clinical decision making. Furthermore, use of the portable CT scanner meant that no patient required disconnection from a ventilator or intracranial monitors to undergo an imaging study. Our experience has encouraged us to increase the time that the portable CT scanner is available for use and to implement its use in other ICUs.

TABLE 3.

Radiological Findings for Each of the Portable HCT Scans Performed (N = 225)

Radiological Findings

No. Scans (%)

Postoperative changes

53 (24)

No interval change

40 (18)

Hemorrhage/hematoma

39 (17)

No new pathology found

26 (12)

Worsening/resolving hydrocephalus

19 (8.4)

Infarct

15 (6.7)

Edema

13 (5.8)

Case Studies Case 1

Ventriculostomy placement

5 (2.2)

Licox placement

4 (1.8)

A 22-year-old man was admitted to the emergency department after a motorcycle accident. His admission GCS was 3 and the initial emergency department HCT scan revealed right frontal and temporal contusions, bilateral subdural hemorrhages, and a subarachnoid hemorrhage (Figure 2A). The patient was admitted to the NICU for observation and follow-up. A portable HCT scan was obtained after an intracranial monitor

Ischemia

3 (1.3)

Ventriculostomy removal

2 (0.9)

Mass

2 (0.9)

Follow-up pathology

2 (0.9)

Hygroma

1 (0.4)

Device placement check

1 (0.4)

113

114

Journal of Neuroscience Nursing

Study Limitations This is a purely descriptive study of our initial experience with a portable CT scanner. Consequently, there was a learning curve associated with its use as we developed protocols. The data do not allow us to conclude that portable CT scanner use is associated with fewer secondary cerebral insults or better patient outcome. However, because we did not need to

TABLE 4.

disconnect patients from their ventilators or intracranial monitors and did not observe any inadvertent disconnection of intravenous lines, we expect that the risk of secondary cerebral insults will be reduced. At the time of the study, we lacked the radiological support required to perform CTA; therefore, we were not able to evaluate the feasibility of this type of scan. This study was conducted over a short period, and scans were

Responses to Survey on Portable HCT Scan Use

Question

No. response (%)

Do you think portable head CT scanning improves patient care? Yes

69 (95)

1 (1.4)

Sometimes

2 (2.7)

Don’t know

1 (1.4)

Do you prefer portable head CT scans or transporting a patient to the CT suite? Portable

54 (74)

Transport

10 (14)

Depends on patient

4 (5.5)

Both

2 (3)

Unsure

2 (3)

No response

1 (1)

What do you like about portable head CT scanning? (n = 125 responses) Safety

24 (19)

Convenience

18 (14)

Promptness of scanning

17 (14)

Less patient movement

17 (13.6)

‘‘Safer’’ for the hemodynamically unstable patient

14 (11.2)

Do not take staff away from the ICU

13 (10.4)

Able to maintain the resources of the critical care environment

12 (9.6)

What do you dislike about portable head CT scanning? (n = 93 responses) No limiting factor Lack of 24/7 portable HCT availability Image quality Arranging the room

9 (10) 18 (19) 8 (9) 17 (18)

Time needed for set up

3 (3)

Concern for radiation exposure

6 (7)

Multimodality monitoring access concerns

1 (1)

Communication between the unit and the RT staff

9 (10)

Maneuvering bulky equipment around an ICU

9 (10)

Risk of injury to staff member

4 (4)

Potential for equipment abuse in locations outside the CT suite

3 (3)

Are you aware of literature about portable head CT scans and/or intrahospital transport? Yes

17 (23)

53 (73)

No response

3 (4)

Note. Percentages represent number of surveys completed (N = 73) unless otherwise noted.

Volume 42

FIGURE 2 Portable CT Scans in Case 1

Number 2

April 2010

115

patients because of disconnection from a ventilator (Marx et al., 1998; Szem et al., 1995; Waydhas, Schneck, & Duswald, 1995). Our own audit process demonstrated an increased risk for hyperventilation associated with excessive manual bagging resulting in a decrease in arterial PCO2, which can then adversely affect brain oxygenation (PbtO2); this deleterious effect is greater when PbtO2 is already compromised (Swanson et al., in press). A recent study compared the incidence of ventilator-associated pneumonia in patients transported outside the ICU with patients that did not undergo IHT. Although both populations had similar mortality rates, the transported group had a higher incidence of ventilator-associated pneumonia (Bercault et al., 2005). Together, these events during IHT may contribute to secondary neuronal injury.

The Potential Benefits of Portable HCT Scans There are several inherent advantages to portable CT scanners that may benefit the patient, the staff, and the hospital. First, point-of-care technology, unlike IHT, means that the patient remains in their bed and care is not interrupted. This includes continuous intracranial

FIGURE 3 Portable CT Scans in Case 2 Note. (a) Unenhanced HCT scan performed in the emergency room (Siemens Sensation 16) showing the right frontal and temporal contusions, the bilateral subdural hemorrhages, and a subarachnoid hemorrhage (Case 1). (b) Unenhanced portable HCT scan showing evolution of right frontal pathology (Case 1).

obtained during regular working hours. Whether the same results would be obtained over a longer period of evaluation or during ‘‘off’’ hours is not known. Finally, this study was conducted at a single center and so lacks external validity.

Intrahospital Transport Technical mishaps, including inadvertent ventilator disconnection and problems with monitoring equipment, have been associated with IHT, and greater than 50% of IHTs involve some difficulty during transport that is either patient or equipment related (Doring et al., 1999; Lovell et al., 2001). Even the shortest transports can have an adverse effect on patient outcome or lead to unexpected events that may compromise patient care (Andrews et al., 1990; Bercault et al., 2005; Papson et al., 2007). In our own center, 157 transports of critically ill patients requiring HCT scans were recorded, and the average time of transport was 47 min with transport times ranging from 20 to 240 min. IHT is a well-known risk factor that can exacerbate the pulmonary function of intubated

Note. (a) Unenhanced scan obtained in the emergency room (Siemens Sensation 16) showing a large right centrum semiovale intracerebral hematoma and an intraventricular hemorrhage (Case 2). (b) Unenhanced portable CT scan showing evolution of intracerebral hematoma, vasogenic edema, and placement of ventricular catheter (Case 2).

116

Journal of Neuroscience Nursing

monitoring and no change in ventilator support. This may be an important factor in reducing secondary cerebral insults because IHT may be associated with a compromise in lung function or oxygenation. Furthermore, the technological advancements in portable HCT scanners mean that high-quality Xenon CT perfusion, contrast enhanced, and CT angiograms can also be obtained at the bedside. This increased diagnostic capability can help guide patient care. Second, new treatments may be initiated sooner because the treating physicians can rapidly diagnose new radiological findings and review these at the bedside. Third, staff satisfaction can be increased because ICU nurses remain in a safe and controlled environment with their patients (Gunnarsson et al., 1999). In addition, there may be fewer staff injuries associated with moving patients and, particularly with critically ill patients, the equipment needed during transport. Fourth, there may be an economic benefit independent of any effect on patient outcome. For example, fewer staff is needed to perform the portable CT scan, ICU nurses do not leave other patients, and there may be fewer staff injuries.

Conclusions Review of our initial experience using a portable CT scanner suggests that it is easy to implement, feasible, and safe. In addition, we found that healthcare providers quickly embraced the technology, and most preferred using a portable HCT scan because they believed that it improved patient safety. There was an initial learning curve of approximately 1 month, and this was facilitated by close communication and teamwork between the radiology staff and technicians, the ICU nurses, and the treating physicians. In addition, the design of standard protocols has facilitated easy implementation of portable HCT scan use. Further study will be necessary to determine if portable CT scans make a difference in patient outcome and in ICU nurse safety. However, it is our belief that portable CT scanning should improve patient care because enhanced patient safety is likely to result from use of this point-of-care technology.

Acknowledgments The authors thank Ronald L. Wolf, MD PhD, and Alexander Mamourian, MD, Department of Radiology, University of Pennsylvania, Philadelphia. This work was supported, in part, by a research grant from Neurologica, Inc.

References Andrews, P. J., Piper, I. R., Dearden, N. M., & Miller, J. D. (1990). Secondary insults during intrahospital transports of head-injured patients. Lancet, 335(8685), 327Y330.

Bercault, N., Wolf, M., Runge, I., Fleury, J. C., & Boulain, T. (2005). Intrahospital transport of critically ill ventilated patients: A risk factor for ventilator-associated pneumoniaV A matched cohort study. Critical Care Medicine, 33(11), 2471Y2478. Chang, E. F., Meeker, M., & Holland, M. C. (2006). Acute traumatic intraparenchymal hemorrhage: Risk factors for progression in the early post-injury period. Pediatric Surgery International, 22(3), 215Y218. Dharap, S. B., Khandkar, A. A., Pandey, A., & Sharma, A. K. (1990). Repeat CT scan in closed head injury. Critical Care Medicine, 18(3), 278Y281. Doring, B. L., Kerr, M. E., Lovasik, D. A., & Thayer, T. (1999). Factors that contribute to complications during intrahospital transport of the critically ill. Journal of Neuroscience Nursing, 31(2), 80Y86. Gunnarsson, T., Theodorsson, A., Karlsson, P., Fridriksson, S., BostrPm, S., Persliden, J., et al. (1999). Mobile computerized tomography scanning in the neurosurgery intensive care unit: Increase in patient safety and reduction in staff workload. Critical Care, 3(5), R83YR89. Halpern, M. T., Palmer, C. S., Simpson, K. N., Chesley, F. D., Luce, B. R., Suyderhoud, J. P., et al. (1998). Economic and clinical efficiency of point-of-care testing for critically ill patients: A decision-analysis model. American Journal of Medical Quality, 13(1), 3Y12. Kendall, J., Reeves, B., & Clancy, M. (1998). Point of care testing: Randomized controlled trial of clinical outcome. British Medical Journal, 316(7137), 1052Y1057. Lahner, D., Nikolic, A., Marhofer, P., Koinig, H., Germann, P., Weinstabl, C., et al. (2005). Incidence of complications in intrahospital transport of critically ill patientsVExperience in an Austrian university hospital. Weiner Klinische Wochenschrift, 119(13Y14), 412Y416. Lovell, M., Mudaliar, M. Y., & Klineberg, P. L. (2001). Intrahospital transport of critically ill patients: Complications and difficulties. Anesthesia Intensive Care, 29(4), 400Y405. Marx, G., Vangerow, B., Hecker, H., Leuwer, M., Jankowski, M., Piepenbrock, S., et al. (1998). Predictors of respiratory function deterioration after transfer of critically ill patients. Intensive Care Medicine, 24(11), 1157Y1162. Papson, J. P. N., Russel, K. L., & Taylor, D. M. (2007). Unexpected events during the transport of critically ill patients. Academic Emergency Medicine, 14(6), 574Y577. Smith, I., Fleming, S., & Cernaianu, A. (2000). Mishaps during transport from the intensive care unit. Journal of Neurosurgery, 93(3), 432Y436. Swanson, E., Mascitelli, J., Stiefel, M., MacMurtrie, E., Levine, J., Kofke, A., et al. (in press). The effect of patient transport on brain oxygen in comatose patients. Neurosurgery. Szem, J. W., Hydo, L. J., Fisher, E., Kapur, S., Klemperer, J., & Barie, P. S. (1995). High-risk intrahospital transport of critically ill patients: Safety and outcome of the necessary ‘‘road trip.’’ Critical Care Medicine, 23(10), 1660Y1666. Warren, J., Fromm, R. E., Orr, R. A., Rotello, L. C., & Horst, H. M. (2004). Guidelines for the inter- and intrahospital transport of critically ill patients. Critical Care Medicine, 31(2), 256Y262. Waydhas, C. (2001). Intrahospital transport of critically ill patients. Anaesthesia and Intensive Care, 29(4), 400Y405. Waydhas, C., Schneck, G., & Duswald, K. H. (1995). Deterioration of respiratory function after intra-hospital transport in critically ill surgical patients. Intensive Care Medicine, 21, 784Y789.

IOP PUBLISHING

JOURNAL OF RADIOLOGICAL PROTECTION

J. Radiol. Prot. 29 (2009) 483–490

doi:10.1088/0952-4746/29/4/002

The use of mobile computed tomography in intensive care: regulatory compliance and

The useradiation of mobile computed tomography in intensive protection care: regulatory compliance and radiation protection G C Stevens1 , N P Rowles1 , R T Foy2 , R Loader1 , N Barua3 , A Williams3 and J D Palmer3 1 2 3

Healthcare Science and Technology Directorate, Plymouth Hospitals NHS Trust, UK Department of Medical Physics, Royal Cornwall NHS Trust, UK South West Neurosurgery Centre, Plymouth Hospitals NHS Trust, UK

E-mail: Gregory.Stevens@phnt.swest.nhs.uk

Received 3 August 2009, in final form 21 September 2009, accepted for publication 2 October 2009 Published 18 November 2009 Online at stacks.iop.org/JRP/29/483 Abstract The use of mobile head computed tomography (CT) equipment in intensive care is of benefit to unstable patients with brain injury. However, ionising radiation in a ward environment presents difficulties due to the necessity to restrict the exposure to staff and members of the public according to regulation 8(1–2) of the Ionising Radiation Regulations 1999. The methodology for enabling the use of a mobile head CT unit in an open ward area is discussed and a practical solution given. This required the reduction in scatter doses through the installation of extra internal and external shielding, and a further reduction in annual scatter dose by restricting the use of the equipment based on a simulation of the annual ward workload.

1. Introduction There are high number of adverse incidents [1] associated with transporting critically-ill patients outside intensive care [2, 3]. Neurologica Corp., (Massachusetts, USA), have developed a small mobile head CT scanner, the Ceretom™, for use within intensive care (ICU). This has the potential to improve patient safety by eliminating the need for the transfer of intubated and ventilated patients to fixed CT facilities. The Ceretom™ is controlled by a laptop via a wireless link; the gantry moving away from the patient on caterpillar tracks to obtain the head image. Both axial and helical modes are possible. To estimate the feasibility of using the Ceretom™ in Plymouth Hospitals NHS Trust, Trust ICU staff collected data, (over a period of two weeks), on patients requiring urgent CT acquisitions. 85% had external ventricular drains or intracranial pressure (ICP) monitors, 95% had arterial lines, 85% had central lines, and 68% required one or more inotropic infusion. The median time for transfer to the fixed CT facility was 55 min (range 35–110 min) with 0952-4746/09/040483+08$30.00 © 2009 IOP Publishing Ltd

Printed in the UK

484

G C Stevens et al

(a)

(b)

Figure 1. Figures showing the scatter maps before and after shielding modification. Both maps were acquired using the maximum tube rating (120 kV, 7 mA) with scatter rates given in µSv h−1 , each square represents 1 m2 . (a) Diagram showing the original scatter map supplied by the manufacturer. (b) Diagram of the scatter map modified by the extra tube shielding and shadow shields.

ICP control being problematic during transfer in 16% of patients. One patient suffered a catastrophic airway event during transfer and ultimately died from chest sepsis. The three commonest causes of delay in obtaining acquisitions were unavailability of the scanner or anaesthetic staff, and portering delays. Following the publication by Masaryk et al [4], clinical staff at Plymouth Hospitals NHS Trust purchased a Ceretom™ for ICU. However, because the radiation protection approach in the UK varies considerably from that taken in the USA; especially with respect to dose constraints, it was not immediately clear whether the Ceretom™ could comply with UK legislation. Examination of figure 1(a) shows the original manufacturer supplied scatter doses; these were considerable and due to the fact that a single acquisition was expected to take upwards of 1.5 min, would make working practices in the ward impracticable if a constraint of 0.3 mSv per annum were to be applied; this value being recommended by the NRPB as a planning constraint to indicate that the likely dose arising from new sources is restricted, and adopted in the Medical and Dental Guidance Notes (MDGN) [7]. Therefore, it was felt we could not comply with the Ionising Radiation Regulations (1999) (IRR99) [5] without either a modification to the Ceretom™, the use of a number of mobile radiation shields or a total evacuation of the ward; the latter two solutions being impractical due to issues over accessibility to patients and the general fragility of the patients in the ICU respectively. The regulations covering the health and safety of workplace exposures from ionising radiation are IRR99; regulations 16 and 18 require the ability to control access and delineate the controlled area respectively: it was not clear how to comply with this without the use of mobile barriers; the use of which, was not conducive to good working practice. Regulation 11(1) clearly gives the radiation employer the responsibility for ensuring compliance with the dose constraint of 6 mSv for non-classified workers, and 1 mSv for members of the public, (excluding trainees). In order to constrain the exposure to workers, regulation 8 requires the radiation employer to restrict exposure as much as reasonably practicable (ALARP) using a hierarchy of control measures (regulation 8(2)). Regulation 31(1) imposes a duty on the 37

Mobile CT in open wards

(a)

485

(b)

Figure 2. Photographs showing the Ceretom™ with (a) the configuration used to measure scatter using a CTDI head phantom and the pelvic phantom in situ, and (b) the shadow shield position during image acquisition, used to minimise scatter.

manufacturer to design equipment to ensure that exposures to patients, staff and members of the public are ALARP, and because of the differing national radiation dose limits, this is clearly not the case. Additional guidance on the application of these regulations is given in the approved code of practice (ACOP) [6] which states: if the device is designed for use in public areas, or where there is continuous access to the working area by employees or other persons not directly involved in the work, the shielding should be designed to reduce dose rates to the lowest level that is reasonably practicable. In this case, the dose rate should be so low that it is unnecessary to designate the area around the device a supervised area. Consequently, it was felt that the Health and Safety Executive (HSE) would have to be consulted on the applicability of ACOP for this specific use as no amount of practical shielding would limit doses so low that a designated area would not be necessary. 2. Approach and discussion In order to satisfy the regulations regarding restriction of exposure, external shielding was designed [8, 9]. The shadow shields are 0.5 mm lead equivalent aprons attached to the side of the Ceretom™ on a swivel that hangs down to limit scatter. The purpose of the swivel is to ensure quick access to the patient in an emergency, and limit the need for separate external mobile shields, while simultaneously still providing the ability to observe the patient during acquisition. In addition, another shadow shield of 1.0 mm lead equivalence was located at the back of the Ceretom™ to limit scatter in the head direction. Photographs of the shadow shields are shown in figure 2. The shadow shields and internal shielding were installed by the manufacturer prior to shipment. The manufacturer produced scatter map was then verified using a CTDI phantom located centrally in the unit’s bore as the attenuation/scatter medium. However, these standard measurement conditions do not account for the scatter and attenuation characteristics that the 38

486

G C Stevens et al

human body would produce. Because the Ceretom™ was to be used in an open ward, it was felt that some account for the effect of the body on the scattered radiation should be included. In order to obtain realistic patient scatter measurements, a CTDI Phantom head phantom was located centrally in the Ceretom™ bore. Body attenuation was obtained by means of a chest phantom (actually an inverted and rotated pelvic phantom) located near the proximity of the chest. The shadow shields were located parallel to the scanner’s z -axis as shown in figure 2(a); this should be the worse case scenario where a radiographer cannot close these scatter shields further in order to be able to observe the patient. It should be noted however, that the more usual location of the shadow shields during a patient acquisition should be as shown in figure 2(b). A matrix of markers were placed 0.5 m apart to 2 m from the iso-centre and 1 m apart further out, centred on the iso-centre to form an 8 m × 8 m matrix around the unit. Because the ward was too small, it was not possible to extend measurements 8 m in all directions. Instead, measurements were made of one quadrant, then the Ceretom™ was rotated by 180◦ so that the opposite quadrant could be measured, consequently, mirror symmetry parallel to the z -axis has been assumed in all scatter measurements. A Radcal 9010 dosimeter with associated 1800 cm3 ionisation chamber, was used to take each measurement at two different heights above the floor level; (the centre of the unit’s bore (97 cm), and 80 cm above the iso-centre; 177 cm), out to a distance of 4 m from the iso-centre. This was done to try to obtain some estimate of the local scatter direction and possibly detect hot spots. As there was less variation with height than with the distance from the iso-centre, the iso-centre measurements were used for subsequent calculations. The dose rates were measured using a 120 kV, 7 mA, 6 s acquisition; the maximum tube rating. The normal acquisition is 4 revolutions per slice at 1 s per revolution giving 4 s per slice, i.e. using a 1 cm slice width will take 100 s assuming a 25 cm acquisition length. The measured doses were corrected using a cross-calibration, obtained from an ionisation chamber calibrated by the Radiation Protection Centre at St George’s Healthcare NHS Trust [10], and traceable to the NPL standard, and converted to dose rate/mAs, allowing the scatter dose to be calculated for any particular scan parameters. The new scatter map was used to designate the controlled area according to the guidance given in appendix 11 of the MDGN. These results are also shown in figure 3(b) and were based on a maximum of 5 acquisitions per day, 7 days per week. For comparison, figure 3(a) gives the same map generated from data obtained using the CTDI head phantom only, i.e. with no body attenuation. It is interesting to note that the presence of the body phantom increases scatter laterally while decreasing scatter towards the foot of the bed due to attenuation by the body. It was decided that the scatter map that included body attenuation would be used to define the designated areas as this represented the actual situation in the ward. The ICU has a bed spacing of 4 m, so it was convenient to use the existing ward furniture to delineate the controlled area. This would satisfy regulations 16 and 18 as the physical bed could act as a simple delineation marker and simultaneously minimise the chance of accidental entrance to the controlled area during an acquisition. Although the main nurses’ workstation lies in the centre of the ward, and outside of the controlled area, the radiographer would have to be located inside the controlled area to observe the patient. Consequently, it was decided to place a mobile screen at the foot of the bed of the patient having the acquisition to ensure the exposure to the radiographer is ALARP. The radiographer would then be able to acquire the image with adequate protection, simultaneously shield the nurses’ workstation, and still be able to observe the patient. Thus, both adjacent patients and staff lie outside the controlled area. The actual annual dose is dependant on the workload. Doses were then further restricted by limiting the use of the Ceretom™ to an allowed number of annual acquisitions. The area 39

Mobile CT in open wards

487

(a)

(b)

Figure 3. Figures showing the extent of the controlled and supervised areas around the unit. Controlled and supervised areas are represented by dark and light grey shaded areas respectively, the middle three squares represent the Ceretom™, (single white shaded area located at the origin), and couch, (the hatched area). Each square is 1 m2 . (a) Data using the CDTI head phantom only. (b) Data using the CTDI head and ‘chest’ phantom.

of the ward simulated was 15 m × 20 m area with bed spacings similar to that of the actual ward. Consequently, it was necessary to extend our scatter data beyond the distances we had physically measured. This was completed by fitting a 6th order polynomial to the dose data with an R 2 0.998 up to 10 m distance from the iso-centre to ensure agreement with experimental data, and assuming an inverse square relationship beyond that. A 1 m × 1 m resolution coordinate system was then overlaid over the entire scatter data giving the dose at each point as D(x, y) with the point D(0, 0) centred on the CT unit itself as shown in figure 3(b). The ward was also given a coordinate system of similar resolution but the origin is located as shown in figure 4; each point being designated P(x � , y � ). Therefore, the accumulated dose at each point in the ward depends on the value of the scattered dose relative to the Ceretom™ at the time, and the total accumulated dose can be calculated at each point in the ward using: �

�

D(x , y ) = where

D(m, n)

(1)

n=1

m = (−1)n+1 x � − x

(2)

n = (−1)n+1 y � − y

(3)

determines the x ordinate and

determines the y ordinate. n is the bed number as given in figure 4, for a maximum of N = 8 beds, using the ward, (x � , y � ), and scatter map, (x, y) coordinate systems. The value (−1)n+1 accounts for the necessity to invert the data for odd numbered beds because the Ceretom™ faces the opposite direction during these acquisitions. In this manner it is possible to obtain the total accumulated annual dose in each part of the ward as shown in figure 4. This data was then modified to calculate the likely dose to the adjacent patients as follows: the maximum patient stay in ICU is 60 days [11]. Therefore, the worst case scenario is when the patient is surrounded by other patients undergoing CT acquisitions. Figure 4 shows that for an equal number of acquisitions per bed, bed 6 incurs the greatest dose. Therefore, one can then use a planning constraint to calculate the maximum number of annual acquisitions per bed allowed before the scatter doses exceed the planning dose constraint, (in this case, the adjacent patient is considered a member of the public). A dose planning constraint of 150 µGy was used to account for all sources such as ward radiography and ensure no patient is exposed to 40

488

G C Stevens et al

Figure 4. Figure showing the accumulated dose in µGy in the simulated ward based on 105 acquisitions per bed per year. Bed spacing is 4 m apart. The increasing shades of grey denote accumulated dose of less than 150 µGy, 150–300 µGy and greater than 300 µGy respectively.

the more usual 300 µSv in any one year. To calculate the patient dose, one simply calculates the accumulated dose due to all other beds around bed 6 ensuring bed 6 itself contributes nothing to the calculation. As may be expected the central beds in the ward accrue the highest accumulated doses due to the contribution from acquisitions in outlying beds. The limiting number of annual acquisitions per bed occurred in bed 6 with a total maximum number of acquisitions per bed per year of 290. This accumulates a maximum dose near bed 6 of 147 µGy, (see the dashed area in figure 5(a)). A similar method may be used to examine the staff dose for ward nurses. Similarly to adjacent patients, ward nurses were not considered radiation workers, and therefore, a planning dose constraint of 150 µGy was again applied. However, although these members of staff are usually a considerable distance from any acquisition, they can spend a significant amount of their working day at, or near the nurses workstation located in the centre of the ward, and marked as the area encompassing the dashed line in figure 5(b). Therefore it was decided to apply the 150 µGy planning dose constraint to the middle 3 m of the ward and assume staff members spend 100% of their time in this area as a worst case scenario. Because the workstation is roughly equidistant to each bed, the accumulated dose limits the maximum number of acquisitions per bed per year substantially to no more than 65. Staff were assumed to work a 40 week year; the results being shown in figure 5(b). In addition to the issues over engineering controls discussed in this paper, there were a number of other issues pertaining to general health and safety concerns of the use of the Ceretom™ to staff. Of these, the most significant were an inadequate audible alarm during exposure, and the possibility of trapping feet under the unit. As a number of other Trusts in the UK had shown interest in the Ceretom™, it was thought appropriate to consult HSE. 41

Mobile CT in open wards

489

(a)

(b)

Figure 5. Figures showing the annual accumulated dose for a adjacent patients and staff. All accumulated doses are quoted in µGy. (a) Figure showing the accumulated dose for a patient staying in bed 6 for a maximum of 60 days, based on 290 acquisitions per bed per year. (b) Figure showing the accumulated dose around the ward for staff working 40 weeks per year, based on 65 acquisitions per bed per year.

Discussions with the manufacturer, UK distributor and HSE resulted in a modification to the Ceretom™; an audible alarm was fitted and labels were attached to the unit warning of its weight and trapping hazard. Discussions with HSE over ACOP paragraph 79 centred around the premise that the Ceretom™ was bought to image patients who were critically ill and that it would be potentially life threatening to move them. Therefore, if it was life threatening to send them to a fixed facility, and the clinical justification was documented in patient notes; only then would use of the Ceretom™ be justified in an open ward. In addition, because the Ceretom™ is considered an emergency piece of equipment, the logical extension to the justification of use is that patients must have access to the Ceretom™ 24 h a day, 7 days a week. Thus the service would have to be available at all times. In addition, the local rules were extended to ensure the radiographer performing the acquisition wore a 0.35 mm lead equivalent apron. This was because in the event of an emergency, and the failure of the wireless connection between laptop and CT unit, the only emergency off button is on the Ceretom™. 3. Conclusion Even with the proposed restrictions of use resulting from the discussions with the HSE, the use of the Ceretom™ in an open ward relies on the large bed spacing, extra shielding and a restriction on the number of annual acquisitions per bed. Therefore, with a 4 m bed separation, a maximum of 65 acquisitions per bed per year could be permitted whilst adhering to the planning dose constraint of a 0.3 mSv dose to the public. It is interesting but not surprising that it is the staff dose and not the patient dose that is the limiting factor in the number of acquisitions that can be permitted in any one year. In addition, although it is not strictly necessary to designate the area out to each adjacent bed a controlled area, it was convenient from the viewpoint of 42

490

G C Stevens et al

ensuring the controlled area was delineated. Finally, because the ICU has already been built to include numerous pieces of ancillary equipment attached to the ceiling near the head of each bed, the ancillary equipment surrounding each bed makes it difficult to enter the controlled area from anywhere except at the foot of the bed nearest the radiographer performing the acquisition. Acknowledgments The authors wish to thank Mr Gareth Thomas of the HSE for discussions on HSE perspective for the use of the Ceretom™ in the NHS, Mr N Powell and Dr R Cranage for supervision of R Foy for his MSc project, Vertec, the UK distributor of the Ceretom™, and Neurologica for the modifications to the Ceretom™ before and after HSE inspections. References [1] Gentleman D and Jennett B 1990 Audit of transfer of unconscious head injured patients to a neurosurgical unit Lancet 335 330–4 [2] Gentleman D, Dearden M, Midgley S and Maclean D 1993 Guidelines for resuscitation and transfer of patients with serious head injuries Br. Med. J. 307 547–52 [3] Wadell G 1975 Movement of critically ill patients within hospital Br. Med. J. 2 417–9 [4] Masaryk T, Kolonick R, Painter T and Weinreb D B 2008 The economic and clinical benefits of portable head/neck CT imaging in the intensive care unit Radiol. Manage. 30 50–4 [5] The Ionising Radiation Regulations 2000 Statutory Instrument 1999 (London: HMSO) Number 3232 [6] Working with Ionising Radiation 1999 Ionising Radiations Regulation 1999 Approved Code of Practice HSE Books [7] The IPEM Working Group 2002 Medical and Dental Guidance Notes The Institute of Physics and Engineering in Medicine Publishing [8] Foy R, Powell N, Cranage R and Rowles N 2007 The development of novel shielding and an improved modelling methodology in the consideration of the radiation protection aspects of mobile computed tomography Symp. 9th Mtg of the CT Users Group (Queen Elizabeth Hospital, Birmingham, 2007) [9] Foy R 2007 The development of novel shielding and an improved methodology in the consideration of radiation protection aspects of mobile computed tomography MSc Thesis University of Exeter [10] URL: http://www.sghrpc.co.uk/ [11] Barua N 2008 personal correspondence, Directorate of Neurosciences, Plymouth Hospitals NHS Trust

Neurocrit Care DOI 10.1007/s12028-011-9627-3

ORIGINAL ARTICLE

Head Computed Tomography Scanning During Pediatric Neurocritical Care: Diagnostic Yield and the Utility of Portable Studies

Head computed tomography scanning

Kerri L. LaRovere • Molly S. Brett • Robert C. Tasker • Keith J. Strauss • Jeffrey P. Burns • The Pediatric Critical Nervous System Program (pCNSp)

during pediatric neurocritical care: Diagnostic yield and the utility of portable studies

Springer Science+Business Media, LLC 2011

Abstract Background We report our use of portable head computed tomography (CT) and the diagnostic yield and radiation dose from head CT in the pediatric intensive care unit (PICU). Methods 204 PICU patients underwent head CT during 2008–2009. Therapeutic interventions and resource intensity during CT were categorized. Severity of illness was summarized using the pediatric risk of mortality (PRISMIII) model. Estimates of patient radiation dose were based on dose measurements made in four anthropomorphic head phantoms. Results 242 (62%) out of 391 head CT studies were portable. New pathology was identified on 80 (40%) scans. CT findings prompted a change in management in 46 (23%) patients; 25 of these resulted in life-extending treatments and 21 had forgoing of life-sustaining treatments within 24 hours. 26 patients with PRISM score greater than 30% underwent CT; 23 (88%) of these were

portable. More portable versus fixed examinations were performed in patients requiring extracorporeal membrane oxygenation, inhaled nitric oxide, high levels of positive end expiratory pressure, and those with high vasopressor scores (P < 0.05). Estimated patient dose from portable CT was 83 ± 6 mGy compared to 72 + 5 mGy for patients imaged on a fixed scanner (P < 0.0001). Conclusion Two-thirds of CT scans obtained in the PICU were portable because of patients’ intensity of therapy and illness severity. Portable CT showed major new pathology in greater than 1/3 and led to a change in management in 1/4 of higher acuity patients scanned. The estimated radiation dose from portable CT is within the current national guidelines. Keywords Computed tomography Bedside Portable Radiation Children Neurocritical care Critical illness Introduction

The members of pCNSp are given in Appendix. K. L. LaRovere (&) R. C. Tasker The Pediatric Critical Nervous System Program (pCNSp) Department of Neurology, Children’s Hospital Boston, 300 Longwood Avenue, Boston, MA 02115, USA e-mail: kerri.larovere@childrens.harvard.edu M. S. Brett R. C. Tasker J. P. Burns The Pediatric Critical Nervous System Program (pCNSp) Division of Critical Care Medicine, Department of Anesthesiology Perioperative and Pain Medicine, Children’s Hospital Boston and Harvard Medical School, 300 Longwood Avenue, Boston, MA 02115, USA K. J. Strauss X-Ray Computations, Inc, Wrentham, MA 02093, USA

Head computed tomography (CT) imaging is a central component of the initial assessment and management of patients with acute brain insult. The identification of new pathology or a lesion requiring surgery is paramount for time-critical interventions. In neurocritical intensive care unit (ICU) practice in adults, the introduction of portable head CT imaging has provided a further dimension to care: the possibility of obtaining a more complete diagnosis by bringing the technology to the patient’s bedside at a time when transfer of critically ill patients with high acuity to the radiology imaging suite may not be feasible [1–4]. Previous reports in the literature have found that portable head CT provided acceptable image quality in the ICU environment [3, 5–8]. An advantage of portable

123

Neurocrit Care

imaging is the reduced need for intrahospital transport. Transport of critically ill patients can be hazardous: in critically ill adults, adverse events are reported between 20 and 70% of transports to radiology [4, 9] compared to less than 4.3% adverse events with portable CT imaging [1, 2, 4, 7, 10]. For critically ill children, studies have found a deterioration in physiologic variables in up to 72% of transports to the radiology suite [10–12]. A disadvantage of CT is the risk of radiation exposure. The recent dramatic increase in CT imaging in the pediatric population has raised concerns that there could be an increased lifetime attributable risk of radiation induced cancer in children compared to adults [13, 14]. To our knowledge, there are no prior reports of the organ dose (i.e., to directly irradiated brain tissue) from portable head CT imaging in children. Portable head CT imaging has become common practice in adult neurointensive care in the past few decades. In children, there are no published reports about the use of this technology. Four years ago, our center introduced portable head CT imaging capability in the pediatric intensive care unit (PICU) setting. We undertook this retrospective study to present our experience; to assess the overall diagnostic yield and effect on patient management of head CT scanning in critically ill patients in the PICU; to examine patient factors that are associated with the method of CT obtained; and to determine the radiation dose to the patient from both portable and fixed head CT imaging.

Methods Design, Setting and Patients The hospital electronic medical records were searched for those individuals who underwent head CT scanning as determined by the critical care attending or consulting neurologists and neurosurgeons during admission to the cardiac or combined medical/surgical PICUs at Children’s Hospital Boston between February 2008 and December 2009 (22 months). The study period represents the second year after introduction of the portable CT scanner, by which time both units were consistently using this technology. Both PICUs are 30 bed, closed panel units in which the critical care physicians are responsible for all patient care. Patients who had a head CT performed while in the operating room, emergency room or on the ward were excluded since the purpose of this study was to evaluate the use of head CT imaging in the PICU and its impact on clinical decisions for these patients. The subjects were divided into two groups, ‘‘portable’’ and ‘‘fixed,’’ based upon the type of CT scan performed during admission. The Institutional Review Board approved this study with waivers for study of retrospective data.

Data Extraction The critical care attending and fellow daily notes were reviewed by an investigator trained in both neurology and pediatric critical care (KLL) for clinical indication for the CT study, radiographic findings and change in patient management that followed the first head CT scan. For clinical indication, the single most relevant new onset neurologic sign, symptom or presenting problem that led to the decision to order a CT study was recorded. The final report of the CT scan was used to confirm the presence or absence of new brain pathology. The radiographic findings were grouped into the following categories: intracranial hemorrhage, cerebral edema, infarction, anoxic injury, ventriculomegaly/ hydrocephalus, and other/nonspecific. Any change in patient management that occurred both as a result of the CT findings and within 24 hours after the CT scan was performed was recorded. A change in patient management consisted of the following: new brain-directed measures, such as the administration of 3% hypertonic saline or 20% mannitol, institution of hypothermia, blood pressure augmentation for maintenance of cerebral perfusion pressure or elevation of the head of bed; a cancelled operation; the decision to discontinue extracorporeal membrane oxygenation (ECMO) support; the documentation of a new resuscitation status; a neurosurgical operative intervention; and the decision to withdraw or withhold treatment. Patient severity of illness was quantified according to the pediatric risk of mortality-III (PRISM-III) model [15] with data recorded from the first 24 hours of admission as part of this research study. Bedside therapies provided at the time of the first head CT study were categorized according to the intensity of respiratory, cardiac and other organ-specific care. For respiratory parameters, the level of positive end expiratory pressure (PEEP) or mean airway pressure (MAP) for those on high frequency oscillatory ventilation (HFOV) as well as the use of inhaled nitric oxide (iNO) were recorded. The cardiovascular parameters of interest were the total inotropic support as described by the Wernovsky vasopressor score [16] and the use of ECMO. In regards to other high-intensity therapies, we noted the use of renal support with continuous venovenous hemofiltration (CVVH) and the presence of unstable intracranial pressure (ICP) as defined by a level above 20 mm Hg requiring treatment. Portable Head CT Imaging The portable head CT device is a commercially available, FDA-approved scanner (CereTom , Neurologica, Danvers, MA) [17]. We have used this device since March 2007. This portable machine is stationed in one of the PICUs, but it is available to all departments. All patients undergoing

123

Neurocrit Care

portable head CT are examined at the bedside, and they are not transported to a special imaging suite on or off the unit. A trained group of radiographers qualified to operate CereTom are available 24 hours/day, 7 days/week. Estimated Patient Dosimetry Patient radiation dose was estimated from direct measurements of the radiation energy emitted by the CT scanner into a standardized phantom. This CT dose index (CTDI) phantom is a cylinder 16 cm in diameter with a length of 15 cm constructed of polymethyl methacrylate. The CTDI100 is a dose index measured at the surface and center of the CTDI phantom with a calibrated pencil ionization chamber 100 mm in length. CTDIvol is the amount of radiation produced by the CT scanner, which allows for comparison of the radiation generated by two different CT scanners; it is not an estimate of patient dose. CTDIvol estimates the average radiation dose to an axial slice of the CTDI phantom, and it has a unit of milliGray (mGy). The effective dose, which has units of milliSieverts (mSv), is defined as the whole body dose delivered to the patient that results in the same risk to the patient as the actual clinical dose that is delivered to only a fraction of the patient’s whole body. While the effective dose allows one to estimate risk to a population of patients, it does not provide radiation dose estimates to the directly irradiated tissues (e.g., brain) of an individual patient undergoing head CT. The estimated patient organ dose to these tissues, which adjusts for the individual size of the patient was based upon dose measurements made with the 100 mm pencil ionization chamber in four anthropomorphic head phantoms ranging in size from a newborn to adult head within the CT scanners used in this study. The anthropomorphic phantoms are elliptically shaped and constructed of tissue-equivalent plastic to represent soft tissues with a different surface plastic selected to mimic the attenuation properties of a patient’s skull. Statistical Analysis The Statistical package for social sciences (SPSS version 19.0 for Windows, SPSS Inc., Cary, NC) was used for descriptive statistics. Categorical variables were compared using one-sample chi-square test, fisher’s exact test or the binomial test for comparing two proportions. Continuous variables were compared using independent samples t tests; equal variances were not assumed. When normality assumptions about the data were suspect, the Mann– Whitney U-test was used to compare the medians of two groups. All P values are two-tailed and considered significant if P < 0.05.

Results PICU Head CT Utilization and Head CT Findings Over the 22 month period, a total of 204 patients out of 6766 ICU admissions received 391 head CT studies; 242 (62%) of these studies were obtained by the portable CT device. Only the first PICU admission was studied for five patients with multiple admissions that met inclusion criteria. Sixty-eight (33%) patients had studies in the radiology suite (‘‘fixed group’’) and 136 (67%) had a portable CT scan (‘‘portable group’’). The median age of patients undergoing imaging was 5 years (range 0.8–15 years). The two groups did not differ in age, gender or indications for CT. The relationship between clinical indication for the CT examination, head CT findings and impact of the study on patient management is shown in Fig. 1. Two hundred patients had data that were able to be evaluated. The other four patients had indeterminate findings on CT. One hundred thirteen of the 200 patients had signs or symptoms warranting imaging, including change in mental status (n = 34), pupillary change (n = 28), seizure (n = 38) or other neurological sign or symptom (n = 13) (Fig. 1a). The other 87 of the 200 patients had a clinical condition that warranted imaging, such as cardiac arrest (n = 29), known previous lesion requiring follow-up imaging (n = 28), other medical condition (n = 20) or recent neurosurgical operative intervention (n = 10) (Fig. 1a). Eighty (40%) of the 200 patients had previously undiagnosed brain pathology on CT scan (Fig. 1b). Since no difference was seen between the portable and fixed groups for signs/symptoms, clinical indication and diagnostic yield, the results from both groups were combined in Fig. 1. Previously undiagnosed radiological findings on head CT included intracranial hemorrhage, cerebral edema, cerebral infarct, features of anoxic brain injury and worsening ventriculomegaly or hydrocephalus. The most common new radiological diagnosis on head CT was intracranial hemorrhage in 24 (12%) patients, followed by cerebral infarct in 15 (8%) and cerebral edema in 14 (7%). There was no difference between portable and fixed groups for radiologic diagnosis. Forty-six (23%) of the 200 patients undergoing CT examination had a new intervention prompted by new CT findings: 25 (54%) of these had life-prolonging therapies and 21 (46%) had forgoing of life-sustaining treatments within 24 hours of the CT study (Fig. 1b). In 4 (2%) out of 200 CT examinations, neurosurgical procedures were performed despite a lack of new or progressive brain pathology (Fig. 1b); three of these resulted in ICP monitor placement for persistent poor neurological status and one of these resulted in

123

Neurocrit Care

Fig. 1 Relationship between clinical indication, head CT findings, and impact of CT findings on patient management. Since no difference was seen between the portable and fixed groups for head CT findings or change in patient management, the results from both groups were combined in this figure. The numbers shown in the circles and squares in (a) represent the number of patients with new pathology and findings that excluded disease progression or new pathology, respectively. In (a), other neurological signs or symptoms

included hypertension and bradycardia, focal weakness, hypertonia, nystagmus, or posturing; other medical conditions included hypertension, hypotension, blood pressure lability, thrombocytopenia, coagulopathy, pre-transplant evaluation, fever, or apnea. Change in med mgmt change in medical management. NS proc neurosurgical procedure. Four patients who had indeterminate findings were not included in either (a) or (b)

ventriculoperitoneal shunt placement for hydrocephalus, which had been planned electively before this patient’s admission. Unstable ICP was present in only 2 out of the 46 patients who had a change in management as a result of the CT findings. Utility of Portable Head CT Scanning It was apparent from the PRISM-III scores that portable CT scans were carried out in more severely ill patients. The raw PRISM-III score at 24 hours was 8.6 (range 1.1–45.8) in the portable group and 3.1 (range 0.51–14.9) in the fixed group (P = 0.038). A total of twenty-eight patients with PRISM-III risk of mortality greater than 30% underwent CT (Fig. 2); 25 (89%) of these had a portable study. On multivariate analysis by logistic regression adjusting for age, gender, and race, the adjusted risk of all cause mortality for a given PRISM-III score was increased by being in the portable group (OR, 2.26; 95% CI, 1.142–4.49; P = 0.019).

Fig. 2 PRISM-III risk of mortality of patients undergoing portable head CT imaging

We surmised that the intensity of interventions or therapies provided to a particular patient impacted the critical care team’s assessment of the relative feasibility of portable versus fixed CT imaging once the team had

123

Neurocrit Care

determined that a head CT was necessary. Both the portable and fixed groups had similar rates of mechanical ventilation (79 vs. 75% in portable and fixed, respectively, P = 0.678). However, when considering the level of PEEP used, a higher level of PEEP was associated with an increasing use of portable CT (Fig. 3a). We found a similar relationship between a higher Wernovsky score and the use of portable head CT imaging (Fig. 3b). Significant differences between groups were also found for patients receiving ECMO and iNO. Twenty-five (18%) patients in the portable group and 3 (4%) patients in the fixed group were on ECMO at the time of the first head CT study (P = 0.005). A significant difference was also found between groups with respect to the use of iNO, with 22

(16%) patients receiving this therapy in the portable group versus one (2%) patient in the fixed group (P = 0.001). No statistical differences were found between groups for unstable ICP or use of CVVH. We also analyzed the subgroup of patients who underwent CT imaging while on ECMO before and after the introduction of the portable device to determine the impact of portable CT on utilization patterns. Six (4%) out of a total of 159 patients were on ECMO and received a head CT during the 22 months before the introduction of portable CT compared to 28 (22%) out of a total of 125 patients on ECMO during this study period, suggesting increased utilization of CT following the introduction of the portable device. Patient Dose Estimates The radiation dose from head CT imaging for patients in the PICU is shown in Table 1. With respect to measurement error, the radiation delivered to the patient from the portable CT scan (CTDIvol), 76 mGy, was within the recommended American College of Radiology guidelines of 75 mGy for adult head CT scans [18]. However, this radiation dose delivered to the patient from the portable techniques used in this study was approximately 15% greater if a fixed CT scan was performed. The estimated dose to brain tissue from the portable scanner (83 mGy) exceeded the estimated dose from the fixed scanner (72 mGy) by a similar estimate of 15% (P < 0.0001). The effective dose from the portable machine exceeded the effective dose of the fixed CT scanner by approximately 65%. This larger discrepancy in effective dose results from longer irradiated scan lengths of the patient’s head with the portable scanner due to patient positioning difficulties at the bedside as opposed to a fixed CT scanner.

Discussion Since the introduction of portable CT imaging to our hospital in 2007, clinicians have utilized this new technology for patients in the PICU with acute neurologic deterioration. The principal findings in our cohort are the Table 1 Radiation dose from head CT imaging CT group Fig. 3 Amount of ICU support provided for patients in the PICU undergoing head CT imaging. The level of positive end expiratory pressure (PEEP) or mean airway pressure (Mawp) for those patients on high frequency oscillatory ventilation (HFOV) is shown in (a), and the vasopressor score is shown in (b). Significant two-tailed P values are indicated as *P < 0.05 as determined by the binomial test for comparing two proportions

CTDIvol (mGy)

Effective dose (mSv)

*Patient dose (mGy)

Portable

75.6

8.9 ± 3.6

83.4 ± 5.9

Fixed

5.4 ± 2

72 ± 5

CTDI CT dose index, mGY milligray, mSv milliSievert * Two-tailed P value < 0.0001 as determined by the independent samples t-test

123

Neurocrit Care

following: any cranial CT imaging has a diagnostic yield of previously undiagnosed pathology of 40% and leads to a change in patient management in 23%; clinical practitioners use portable CT for high severity cases, and there is increased utilization among patients receiving ECMO; and the radiation dose from portable CT is within national guidelines, but the patient dose is 15% greater than that from a standard fixed CT scanner. Head CT imaging provided a new diagnosis of brain pathology in a total of 80 (39%) patients (Fig. 1a), and the diagnostic yield did not differ between groups. Although overall reduced image quality due to increased noise and artifacts has been reported with the use of portable compared to conventional scanners, the diagnostic accuracy and reliability of portable CT is no different from that of fixed CT [6, 8]. In our study, the image quality of every head CT study analyzed was sufficient to resolve the question that prompted imaging. Two main inferences can be made from our findings. First, since the overall diagnostic yield of portable CT scanning in our series is similar to that of head CT imaging in PICU patients carried out in the radiology suite, it seems likely that CT investigation in the most critically ill patients is being used appropriately. This observation is consistent with the diagnostic yield from CT in critically ill adults, which ranges from 26 to 30% [19, 20]. Second, analysis of the ECMO subgroup supports the conclusion that patients who were receiving ECMO in the past and would have been deemed too critical to transport are now benefiting from portable CT investigation. In regards to the clinical value of head CT imaging in critically ill patients, we found that CT investigation prompted a change in management in 23%, with no difference between the groups. This proportion is similar to the results of portable CT in adults in which 22â&#x20AC;&#x201C;57% of cases have a change in therapy after imaging [2, 7, 9, 19â&#x20AC;&#x201C; 21]. We found that a greater proportion of portable versus fixed examinations were performed in patients requiring more intensive or complex therapies (e.g., high PEEP, high vasopressor category C 5, ECMO, and iNO). This result is similar to the reports in adult neurocritical care. In a survey of physicians who ordered portable CT studies, McCunn et al. [1] found that patients who underwent portable CT imaging were those receiving ECMO (93%) and those with instability in hemodynamic (70%), respiratory (57%) and neurologic (40%) systems. In a prospective study in a neurocritical care unit, Stevens et al. [22] found that 85% of adults who had a portable head CT study had external ventricular drains or ICP monitors, 95% had arterial lines, 85% had central lines and 68% required at least one inotropic infusion. Unlike our population, larger proportions of adult patients who had portable head CT examinations had neurologic instability or had an ICP monitor in place.

Any potential benefit of a change in treatment must be weighed against the risk of conducting a CT investigation. Exposure to ionizing radiation is a major disadvantage of any CT imaging. The estimated lifetime risk of death from cancer from one head CT in the United States of America is between one in 1,000 and one in 5,000 pediatric head CT scans, with risk decreasing as age increases [13, 14]. Although the estimated radiation dose from portable CT in our study is within current national guidelines, our organ dose estimate from the portable CT machine is approximately 15% greater than the dose in patients imaged on a fixed standard scanner in the radiology suite. Even though radiation exposure limits the use of CT in children, the benefits of the diagnostic information provided and the management pathway determined as a result of the CT findings outweigh the minimal radiation exposure from a single CT study [23]. In our study, 130 (64%) patients had a single head CT study, and the diagnostic yield from a single CT scan was 40%. Therefore, an increase in organ dose of only 15% over and above a minimal radiation risk from a single CT study justifies the use of portable over fixed CT because of the risk to benefit ratio. Our study has three main limitations. First, the retrospective study design makes an accurate assessment of the impact of CT imaging on real-time patient management difficult to assess. Therapies were often instituted empirically before the CT procedure, and documentation that clearly linked a specific clinical decision with the CT finding could not be found in some cases. Thus, no specific change in patient management could be demonstrated retrospectively for some patients. Furthermore, data on the precise interval between ordering the scan and image availability for portable and fixed CT could not be determined retrospectively. Second, data from all patients undergoing head CT imaging before the introduction of portable head CT imaging in our institution was not studied, so conclusions about whether the introduction of this device led to more or less utilization of CT or effect on patient management cannot be made. For the ECMO subgroup, however, there was an increased use of head CT imaging after the introduction of the portable machine. Third, given the retrospective nature of our study, we could not assess major complications attributable to intrahospital transport or to the use of portable CT. However, no complications were recorded in the electronic medical records. At our institution, a CT study is performed with the following clinical team members present at the bedside: the patientâ&#x20AC;&#x2122;s nurse; a clinical fellow, attending critical care physician, or nurse practitioner; and a respiratory therapist. Furthermore, for every high risk patient who is transported to the radiology suite, the accompanying team carries a transport bag of equipment and emergency medications. In conclusion, head CT imaging is a necessary investigation in pediatric neurocritical care for the diagnostic

123

Neurocrit Care

evaluation and management of acute neurologic deterioration. Head CT imaging has a high diagnostic yield, and portable CT informs clinical decision-making for children receiving neurointensive care. Some patients are too unstable to travel to radiology, and the technology must come to their bedside. Portable head CT imaging can be carried out on the PICU in critically ill children with acceptable radiation exposure. Acknowledgments The authors would like to thank all staff at Children’s Hospital Boston who participated in the care of these patients. The Pediatric Critical Nervous System Program (pCNSp) is comprised of the Neurocritical Care ICU service, the Critical Care Neurology consult service, the neurosurgical service, and neuroradiology. The authors also thank Matthew Gregas for statistical support; Patricia Berry, Charles Hornberger, and Peter Betit for database support; and Alison Clapp for librarian support. This research was funded by the Department of Anesthesia at Children’s Hospital Boston. Conflict of interest

None.

Appendix: pCNSp members Neurocritical Care Program Robert Tasker (Chair, Departments of Neurology and Anesthesia). Critical Care Neurology consult service (CCNS) Omar Khwaja, Janet Soul, David Urion, Mark Libenson, Michael Rivkin, Basil Darras, Mustafa Sahin. Neurocritical Care ICU service From the Division of Critical Care Medicine: Jeffrey Burns (Chief, Division of Critical Care Medicine), John Arnold, Monica Kleinman, Robert Graham, Daniel Kohane, Thomas Mancuso, Michael McManus, Nilesh Mehta, Robert Pascucci, Gregory Priebe, Adrienne Randolph, Robert Truog, Meredith van der Velden, Sally Vitali, Peter Weinstock, Gerhard Wolf, Christopher Weldon. From the division of Cardiovascular Critical Care: Peter Laussen (Chief, Division of Cardiovascular Intensive Care), Melvin Almodovar, Ravi Thiagarajan, James DiNardo, Thomas Kulik, Joshua Salvin, Cindy Barrett, Sarah Teele, Catherine Allan, John Kheir, Vamsi Yarlagadda, Satish Rajagopal. Neurosurgery R. Michael Scott (Chief, Department of Neurosurgery). Neuroradiology Richard Robertson (Chief, Department of Radiology).

References 1. McCunn M, Mirvis S, Reynolds N, Cottingham C. Physician utilization of a portable computed tomography scanner in the intensive care unit. Crit Care Med. 2000;28:3808–13.

2. Teichgraber UK, Pinkernelle J, Jurgensen JS, Ricke J, Kaisers U. Portable computed tomography performed on the intensive care unit. Intensiv Care Med. 2003;29:491–5. 3. Butler WE, Piaggio CM, Constantinou C, et al. A mobile computed tomographic scanner with intraoperative and intensive care unit applications. Neurosurgery. 1998;42:1304–10. Discussion 10–1. 4. Gunnarsson T, Theodorsson A, Karlsson P, et al. Mobile computerized tomography scanning in the neurosurgery intensive care unit: increase in patient safety and reduction of staff workload. J Neurosurg. 2000;93:432–6. 5. Mirvis SE. Use of portable CT in the R Adams Cowley Shock Trauma Center. Experiences in the admitting area, ICU, and operating room. Surg Clin North Am. 1999;79:1317–30. 6. Matson MB, Jarosz JM, Gallacher D, et al. Evaluation of head examinations produced with a mobile CT unit. Br J Radiol. 1999;72:631–6. 7. Peace K, Wilensky EM, Frangos S, et al. The use of a portable head CT scanner in the intensive care unit. J Neurosci Nurs. 2010;42:109–16. 8. Rumboldt Z, Huda W, All JW. Review of portable CT with assessment of a dedicated head CT scanner. AJNR Am J Neuroradiol. 2009;30:1630–6. 9. Waydhas C. Intrahospital transport of critically ill patients. Crit Care. 1999;3:R83–9. 10. Lahner D, Nikolic A, Marhofer P, et al. Incidence of complications in intrahospital transport of critically ill patients— experience in an Austrian university hospital. Wien Klin Wochenschr. 2007;119:412–6. 11. Wallen E, Venkataraman ST, Grosso MJ, Kiene K, Orr RA. Intrahospital transport of critically ill pediatric patients. Crit Care Med. 1995;23:1588–95. 12. Tobias JD, Lynch A, Garrett J. Alterations of end-tidal carbon dioxide during the intrahospital transport of children. Pediatr Emerg Care. 1996;12:249–51. 13. Brenner DJ. Estimating cancer risks from pediatric CT: going from the qualitative to the quantitative. Pediatr Radiol. 2002;32:228–31. Discussion 42–4. 14. Brenner DJ, Hall EJ. Computed tomography—an increasing source of radiation exposure. N Engl J Med. 2007;357:2277–84. 15. Pollack MM, Ruttimann UE, Getson PR. Pediatric risk of mortality (PRISM) score. Crit Care Med. 1988;16:1110–6. 16. Wernovsky G, Wypij D, Jonas RA, et al. Postoperative course and hemodynamic profile after the arterial switch operation in neonates and infants. A comparison of low-flow cardiopulmonary bypass and circulatory arrest. Circulation. 1995;92:2226–35. 17. Mobile computed tomography evaluation of the NeuroLogica CereTom. Health Devices. 2008;37(11):325–42. 18. Fryback DG, Thornbury JR. The efficacy of diagnostic imaging. Med Decis Mak. 1991;11:88–94. 19. Miller WT Jr, Tino G, Friedburg JS. Thoracic CT in the intensive care unit: assessment of clinical usefulness. Radiology. 1998;209:491–8. 20. Maher MM, Hahn PF, Gervais DA, Seoighe B, Ravenscroft JB, Mueller PR. Portable abdominal CT: analysis of quality and clinical impact in more than 100 consecutive cases. AJR Am J Roentgenol. 2004;183:663–70. 21. Indeck M, Peterson S, Smith J, Brotman S. Risk, cost, and benefit of transporting ICU patients for special studies. J Trauma. 1988;28:1020–5. 22. Stevens GC, Rowles NP, Foy RT, et al. The use of mobile computed tomography in intensive care: regulatory compliance and radiation protection. J Radiol Prot. 2009;29:483–90. 23. Brody AS, Frush DP, Huda W, Brent RL. Radiation risk to children from computed tomography. Pediatrics. 2007;120:677–82.

123

Accepted Manuscript Intraoperative CT and nexframe-guided placement of bilateral hippocampal responsive neurostimulation: A technical note based Intraoperative CTbased and Nexframe-guided placement of bilateral hippocampal responsive neurostimulation: A technical note Kunal Gupta, MBBChir (Cantab), PhD, Jeffrey S. Raskin, MD, MS, Ahmed M. Raslan, MD PII:

S1878-8750(17)30132-8

DOI:

10.1016/j.wneu.2017.01.109

Reference:

WNEU 5206

To appear in:

World Neurosurgery

Received Date: 4 November 2016 Revised Date:

25 January 2017

Accepted Date: 26 January 2017

Please cite this article as: Gupta K, Raskin JS, Raslan AM, Intraoperative CT and Nexframe-guided placement of bilateral hippocampal based responsive neurostimulation: A technical note, World Neurosurgery (2017), doi: 10.1016/j.wneu.2017.01.109. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Intraoperative CT and Nexframe-guided placement of bilateral hippocampal

RI PT

based responsive neurostimulation: A technical note

Kunal Gupta, MBBChir (Cantab), PhD, Jeffrey S. Raskin, MD, MS, and Ahmed M. Raslan, MD

Correspondence: Ahmed M. Raslan, MD Department of Neurological Surgery Oregon Health & Science University Mail Code: CH8N 3303 SW Bond Ave. Portland, OR 97239

M AN U

Department of Neurological Surgery, Oregon Health & Science University, Portland, Oregon

AC C

E-mail: raslana@ohsu.edu

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

INTRODUCTION: Surgical resection of a lesion that correlates with seizure onset in patients with epilepsy can dramatically improve seizure burden and quality of life. For bilateral hippocampal lesions, bilateral resection comes with a risk of severe cognitive deficits.

RI PT

Responsive neurostimulation (RNS) devices offers a new modality to treat multifocal lesions in a reversible manner, including bilateral hippocampal stimulation. We describe technical aspects of Nexframe-assisted placement of bilateral NeuroPace mesial temporal electrodes, and case

examples.

METHODS: Retrospective chart review was performed for 4 patients who underwent bilateral

M AN U

mesial temporal RNS placement for medically intractable epilepsy. Operative techniques were assessed and modified. Ambulatory electrocorticographic recordings and a sub-analysis of available data are summarized.

RESULTS: Eight electrodes were placed in 4 patients, who were followed for up to 6 months. 1

out of 8 electrodes was revised due to vector error greater than 3mm; after surgical technique

modification, all subsequent electrodes were reliably placed in a single pass with less than 1.5 mm vector error. Using patientsâ&#x20AC;&#x2122; seizure diaries, seizure semiologies were correlated with

ambulatory ECoG recording patterns and sub-analyzed; 51.4% were left-sided, 15% right-sided, and 33.6% indeterminate.

AC C

CONCLUSIONS: We report herein the technical nuances of adapting Nexframe to hippocampal based depth electrode RNS system placement. Our group has extensive experience with Nexframe for accurate and safe deep brain stimulation electrode placement. Our preliminary data with bitemporal RNS placement suggests similar accuracy and safety. KEY WORDS responsive neurostimulator/neurostimulation, mesial temporal sclerosis, Nexframe, frameless stereotaxy, epilepsy

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

INTRODUCTION Resective epilepsy surgery can dramatically improve patient quality of life and survival, if patients have a resectable lesion that correlates with seizure focus. In the presence of unilateral

RI PT

hippocampal sclerosis, surgical resection has been shown to result in seizure-free outcomes (Engel class 1) in 50-70% of patients at 1-2 years,1-3 with robust prolonged seizure-free

outcomes in 50% of patients reported at 5 years.4 For bilateral hippocampal sclerosis, bilateral

resection is not possible due to severe cognitive, memory and speech deficits associated with bilateral resection.5 Cukiert et al., report 5 patients with bilateral hippocampal sclerosis, all

M AN U

patients demonstrated unilateral onset and thus received unilateral resection; 4 achieved Engel class 1 seizure-freedom, and 1 achieved Engel class 2 seizure-control.6 In a later report, in a small patient cohort with bilateral hippocampal sclerosis, published by the same group; 2 patients had bilateral ictal onset, however, 80% of activity localized to one side and guided resection, and

outcomes were similar.7 Of patients who have recurrent seizures after unilateral mesial temporal

lobe resection, up to 30% are due to contralateral recurrence; furthermore, equivocal lateralizing ictal activity has been associated with poorer outcomes and complicates surgical treatment for

bilateral hippocampal sclerosis.8

Neurostimulation is a viable treatment option and can offer good seizure control in the

AC C

absence of a resective surgical approach, for example in the presence of multiple lesions or seizure foci in eloquent areas. Open loop stimulation devices include vagal nerve stimulation (VNS); this provides regular afferent abortive stimulation that can be triggered by manual activation of the device. Vagal nerve stimulation (VNS) is typically a salvage procedure for patients with persistent seizures after resection or in the absence of a resectable lesion. NeuroPace (NeuroPace, Mountain View, CA) is a novel closed-loop responsive neurostimulation

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

(RNS) device, which has received indications from the United States Federal Drug Administration (FDA) for medically refractory partial onset epilepsy, in the presence of 1 or 2 seizure foci.9 The device monitors real time electrocorticography (ECoG), and responds to

RI PT

abnormal activity with stimulation. The device is considered modulatory, as opposed to ablative, allowing for use in eloquent cerebral lesions and with bilateral temporal pathology. In addition to a stimulatory function, the device records long-term brain activity via ECoG, which is a novel

benefit, and may enable evaluation of changing seizure patterns over time and guide therapy. Long term RNS monitoring has demonstrated that an average of 40 days of monitoring is

M AN U

required prior to final determination of laterality.10 Long term studies of RNS stimulation have demonstrated 44% seizure reduction at 1 year,11 increasing to 66% over 5 years12, irrespective of prior VNS or resective surgery.13 These studies have focused on mesial temporal pathology, with remaining patients having frontotemporal neocortical pathology.12 Neuropsychological studies

have demonstrated no worsening in cognitive deficits with bilateral temporal lobe stimulation,

and improvements in specific domains, including naming and memory. Adverse effects have rarely been reported, and include a 3.5% infection rate with skin flora primarily, a single

incidence of osteomyelitis in the initial study, and 2.7% rate of hemorrhage.14 The manufacturer (NeuroPace) has not developed a proprietary method of implantation

AC C

(personal communication); therefore it is up to each individual neurosurgeon to develop methodologies for implantation. Prior studies by other neurosurgical groups have reported placement of RNS electrodes primarily by framed based stereotactic systems, and the surgical techniques are not well described.15,16 Our group has extensive experience with placement of deep brain stimulation (DBS) electrodes in the pallidum and thalamus, and we have reported on its accuracy and safety.17 The automatic adaptation of the technical nuances of NexFrame use in

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

DBS placement to placement of deep hippocampal electrodes is not possible without modifications. The purpose of this report is to provide current and/or prospective users of NexFrame with the technical nuances needed for safe use of NexFrame in hippocampal depth

RI PT

electrode placement. At our institute, this is combined with intra-operative computed

tomography (CT; CereTom, NeuroLogica, Danvers, MA) for confirmation of electrode

placement. We then describe follow-up recordings obtained from implanted Neuropace systems.

Two case examples are provided to illustrate the technical nuances.

M AN U

METHODS Patient selection

At our institution patients are selected for bilateral mesial temporal lobe stimulation by the epilepsy multi-disciplinary team. In each case, the presence of bilateral independent temporal

TE D

onset on long-term electroencephalogram (EEG) suggested that unilateral hippocampal resection would be unlikely to provide adequate seizure control. Patients were therefore referred for bilateral mesial temporal electrode placement; the procedures were performed by the senior

author (AMR) using the methodology described below.

AC C

Operative technique

Surgical planning: using a high-resolution 3T magnetic resonance imaging (MRI), an

optimal trajectory of a depth electrode to purchase the entire hippocampus and possibly amygdala, avoiding the ventricle if possible, is planned. Planning should consider the 25-degree coronal angle limit on the Nexframe, which in this case applies to the medial trajectory of the plan. We have found that accommodating the 25-degree limit leads to a more medial entry point

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

on the calvarium. The more medial entry point can prohibit simultaneous mounting of the two Nexframe towers, which could lead to repetition of registration. Therefore a minimum of 7 cm is kept between burr holes, to allow bilateral simultaneous Nexframe mounting.

RI PT

Patient positioning: the patient is placed in a prone position with a DORO radiolucent Mayfield head holder, (pro med instruments, Inc., Cape Coral, FL) taking care to retract the shoulders and allow safe entry of the head into the CereTom portable CT scanner (Figure 1A),

the head is slightly rotated to expose more of the side of the prospective RNS battery implant (Figure 1B). The two pins of the DORO frame are placed contralateral to the side of the RNS

M AN U

battery (Figure 1B). Skull-based fiducial screws are placed into the calvarium, taking into account site of the unsterile registration reference and placement of the calvarial generator (Figure 1C). This can be planned, with regards to the incision and fiducial placement, using the generator template (Figure 1D). An intra-operative thin-cuts CT is performed using CereTom

TE D

and fused to the pre-operative MRI (Figure 1E and F), and the fiducials are registered (Figure 1G). The burr hole sites are identified on the skin using the stereotactic probe and a hand drill used through the scalp after sterile preparation to score the calvarium at the center of each entry

site (Figure 1H). Curvilinear incisions are drawn around the planned entry sites, with a larger incision placed at the planned generator implant site (Figure 1H), cutaneous flaps are reflected

AC C

and burr holes created with an Anspach cranial perforator (Figure 1I). The dura is cauterized, the burr hole edges are burred back and Stimloc base anchoring device is screwed into place (Figure 1J); the locking clips are placed and removed to confirm adequate fixation. The width of the burr hole, over a twist drill hole, increases the degrees of freedom available for targeting with the stereotactic probe. The Nexframe tower is assembled and registration performed with the sterile reference array (Figure 1K). The guidance probe is used to establish the trajectory (Figure 1L)

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

and the microTargeting STar Drive system (FHC, Bowdoin, ME) is then assembled (Figure 1M). The electrode depth stop is placed at the appropriate length for the platform, and using the appropriate distance to target the STar Drive measurement is placed at the appropriate depth, as

RI PT

determined by the Medtronic Framelink targeting system. In our experience this challenged the extreme range of the STar Drive system, which is 115 mm, however remained within its

functioning limits. The cannula was placed and the stylet removed, and the electrode placed until

M AN US C

limited by the depth stop. This was secured, and the cannula sheath withdrawn (Figure 1N). In our initial experience, cannula withdrawal was limited by the targeting system and the electrode could not be accessed and secured proximally; as a result, withdrawal of the cannula resulted in partial withdrawal of the electrode, which therefore required replacement. We then moved to initially passing a short-length cannula with a full-length stylet to guide the electrode: upon withdrawal, the short length cannula allows the electrode to be visualized and secured prior to

fully removing the cannula sheath. The cannula is withdrawn, the Stimloc locking clip is applied

(Figure 1O) and fibrin glue (Tisseel, Baxter, Inc., Deerfield, IL) infused into the burr hole. The electrode stylet is withdrawn and the electrode is withdrawn from the cannula into the Nexframe

tower under direct visualization (Figure 1P). The same process is repeated for the contralateral electrode, which could include re-registration if simultaneous mounting is not possible. The

AC C

Nexframe towers are removed, the Stimloc caps placed and the leads protected until they are secured to the implanted generator (Figure 1Q). The contralateral lead is tunneled under the galea to the incision containing the generator. A full-thickness craniotomy to the shape of the NeuroPace generator is then performed (Figure 1R), the generator secured to the calvarium (Figure 1S), and the leads affixed to this (Figure 1T). A CereTom CT is performed and merged with the pre-operative MRI and surgical plan, to confirm placement of the electrodes along the

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

planned trajectory (Figure 2). Recordings are made to confirm detection of ECoG and the incisions are irrigated and closed.

RI PT

RESULTS NeuroPace RNS placement

Four patients underwent bilateral hippocampal NeuroPace electrode placement. Patient

M AN US C

data are presented in Table 1. Three were male and 1 was female. The median age was 44.5 ± 6.94 years. Over a series of 4 cases including 8 electrodes placed using this method, there was a single electrode replacement due to difficulties encountered with electrode depth and the electrode insertion cannula. All other electrodes were accepted within 2 mm of target and did not require revision or replacement. We have previously evaluated the NexFrame targeting system for vector error in thalamic and pallidal DBS electrode placement and found trajectory and

vector errors to be 1.24 ± 0.87 mm and 1.59 ± 1.11 mm respectively;17 similar studies by other

groups utilizing Nexframe for deep brain stimulation have demonstrated vector errors of 2.78 ± 0.25 mm,18 and 2.8 ± 1.3 mm.19 The deviation encountered in the present study for hippocampal

lead placement is similar to these previously reported values (median 1.65 ± 0.7 mm). The initial case duration was 314 minutes, the subsequent cases’ operative duration ranged from 195 to 226

AC C

minutes (median 225.5 ± 45 minutes). The mean intensive care unit stay duration was 2 ± 1.30 days, with total duration of stay 4.5 ± 2.38 days. At 6 months follow-up, one patient reported the same seizure frequency as pre-op however much reduced duration and intensity of ictal events, one patient reported 50% reduction in frequency and duration, one patient reported no change, one patient is deceased due to events unrelated to surgery. Data are presented as median ± standard deviation.

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

CASE EXAMPLES CASE #1 A 44-year-old male, who developed seizures 2 years prior to surgery, with no prior

RI PT

history of seizures. At the time of his original presentation (2 years prior to surgery), he

developed non-specific symptoms of low-grade fever, dizziness and difficulty sleeping; he was seen in urgent care and was prescribed azithromycin. A few days later he experienced worsening

confusion and had a generalized tonic-clonic seizure. He was transferred to our institution for care; at that point he had continued subclinical electrographic seizures despite multiple

M AN U

concurrent anti-epileptic drugs (AEDs) including phenytoin, levetiracetam, topiramate and benzodiazepines. Cerebrospinal fluid (CSF) was unremarkable. Extensive infectious and autoimmune evaluation was negative, and a high dose solumedrol trial had no clear benefit, however his seizures gradually abated, and after 18 days he was discharged. After discharge, he

continued to have daily seizures of multiple semiologies, often triggered by stress. He therefore

underwent out-patient long-term video-EEG monitoring, which revealed bilateral independent temporal sharp waves and multiple seizures of bilateral independent onset. Brain MRI

demonstrated bilateral mesial temporal sclerosis, and cerebral angiogram demonstrated chronic proximal left middle cerebral artery occlusion with reconstitution by lentriculostriate collateral

AC C

vessels. He is verbal and neurologically non-focal on clinical examination, though he reports episodic memory difficulties.

NeuroPace RNS Recordings The ability to record ECoG was confirmed intra-operatively, and recording abilities were activated immediately. Stimulation parameters were held until the first post-operative visit with

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

his neurologist at 4 weeks. A novel and previously unprecedented ability of an implanted closed loop system is the ability to perform long-term ECoG in the patient’s home environment. Coupled with a physical seizure diary maintained by the patient, this enables neurologists to

RI PT

obtain real-time ECoG and correlate it to patient seizure semiology. The patient described 5 different seizure semiologies: 1) simple partial seizures, 2) complex partial seizures, 3) brief episodes he described as “jolts”, 4) brief jolts accompanied by a remote visual aura, and 5)

feeling of being dissociated from reality, which he described as being in the “third person”. As this patient experienced these events he was able to trigger the device, recording ECoG traces for

M AN U

each of these seizure semiologies for review by his neurologist (Figure 3a-f). In the initial 3 weeks after placement, the device recorded 107 events triggered manually by the patient’s magnet, he endorsed at the clinic visit that he was able to swipe his device for almost all events. Laterality of events was identified by the presence of rhythmic high amplitude activity, of which

51.4% were left-sided, 15% were right-sided and 33.6% were indeterminate (Figure 4a). These

data suggest that a considerable number of his events are below detection threshold, and may not have been captured by conventional staged ECoG. Furthermore, just over 51% were left-sided,

well below the 80% threshold recommended by aforementioned methodology describing unilateral temporal resection in bilateral disease.6 Upon evaluation of the patient’s seizure diary,

AC C

he was able to record semiology and date/time, allowing his events to be correlated with magnettriggered recordings made by the device. He clearly identified a single complex partial seizure, 10 simple partial seizures, 9 dissociative or “3rd person” type events, 5 auras, and 3 “jolts”. The device data suggest a bilateral component to the majority of his seizure types, and highlight the difficulty in accurately determining laterality for his auras and jolts (Figure 4b). It is likely that these will be better defined with prolonged recording time.

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Post-RNS outcome The patient tolerated the procedure well, and the electrodes were placed using Nexframe

RI PT

and intra-operative CT methodology with high accuracy. The incisions healed well, and reliable ECoG tracings were obtained. His device was activated at low stimulation parameters, and his neurologist continues to follow him closely to determine his optimal stimulation pattern for

seizure control.

M AN U

CASE #2

A 45-year-old female with intractable bilateral temporal lobe epilepsy secondary to limbic encephalitis. At age 26 years she was diagnosed with temporal lobe epilepsy from simple partial seizures. She then failed medication trials including carbamazepine, oxcarbazepine,

lamotrigine, and tiagabine for lack of efficacy or unwanted side effects.

In 2007, she suffered non-paraneoplastic anti-glutamic acid decarboxylase antibody (GAD-Abs) positive limbic encephalitis (LE) diagnosed by serum and CSF anti-GAD Abs with

treatments including pulse cyclophosphamide, plasmapheresis, rituximab, prednisone, and azathioprine. She suffered from refractory complex partial seizures numbering 12 per day,

AC C

beginning with an aura of a deja vu and an odd dreamy/fearful feeling, followed by diminished responsiveness, facial tightening, vacant staring, aimless arm movements, and automatic verbalization.

At this time video EEG and bilateral foramen ovale electrodes captured 50 seizures and

45 of the seizures were of clear right temporal onset with ictal discharge and variable spread over the right temporal surface electrodes. Five of the seizures were subclinical seizures of left

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

temporal onset. A brain MRI with and without contrast showed no abnormal enhancement, but bilateral amygdala and hippocampal increased T2 signal of FLAIR sequence with the left hippocampus appearing atrophic and the right hippocampus appearing edematous. A vagal nerve

RI PT

stimulator (VNS) was placed in 2012 as palliative therapy, with encouraging responses including shorter durations of seizures when using the magnet during seizures, and quicker recovery thereafter. She reported some mild dysphonia but otherwise tolerated the treatment well.

She was determined to not be a neurosurgical candidate, because of the immunologic etiology and bilateral temporal lobe involvement. When evaluated for RNS placement she

M AN U

suffered three disabling seizures per week and 3-5 seizures per day despite VNS therapy, antiepileptic therapy including lacosamide, levetiracetam, pregabalin and immunotherapy including azathioprine.

Post-RNS outcome

The patient tolerated the implantation procedure well although she experienced an immediate increase in seizure burden, which tapered over postoperative week one. Her incisions

healed well. Her short-term memory is significantly impaired, and she lives in an adult foster home, staying with her mother and daughter 3 days per week. She underwent monitoring for 1

AC C

month after bilateral mesial RNS electrode placement, followed by initiation of stimulation. At 3-month follow-up she had incurred a 50% reduction in her seizure burden.

DISCUSSION

The Nexframe and intraoperative CT method for delivery of DBS electrodes has been previously described by our group.17 One advantage of Nexframe is that it replaces a

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

conventional stereotactic frame, which would allow performance of this surgery at institutions worldwide that for fiscal and/or other reasons do not possess such infrastructure. This in turn allows more patients, who may benefit from the procedure, access to therapy. The absence of a

RI PT

large frame also allows access to the calvarium for generator placement, without requiring

removal of a frame and risking electrode dislodgement. At our institution, we also perform intraoperative CT to confirm electrode placement prior to closure, reducing our requirement for post-

M AN US C

hoc lead revision.

In our operative experience, the methodology for DBS electrode placement using Nexframe was not easily transferred to RNS temporal lobe electrode placement. The differences could be summarized in three main issues and three further subsidiary issues: 1- Hippocampal electrode depth (105-115 mm), which is much greater than conventional DBS depth (75-95 mm), which could challenge the maximum range of the STar Drive delivery system of 115. That could

be mitigated by adjusting the target to second contact with an offset equal to the distance

between second contact and first contact. 2- The need to use of shorter cannula (short of target) fitted with longer stylet (at target) to allow retrieval of the cannula and visualization of electrode

to be able to apply the stimloc clip. This is also caused by the larger electrode depth hence a smaller distance from the skull to the end of the StarDrive where the cannula would be retrieved

AC C

to. 3- The ideal hippocampal depth trajectories may not allowed a 7 cm minimum distance between the two NexFrame towers, which could create the need for sequential placement of towers and dual sterile registration. 4- Finally, the conventional need for 7 cm distance between the two NexFrame towers may not be necessary due to skull geometry that makes occipital placement of NexFrame towers not on the same horizontal plane therefore a smaller linear distance is possible without collision, so we trial the placement of the towers even through the

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

linear measured distance may be less than 7 cm. 5- The procedure is best done prone in DoroFrame if intraoperative CT is contemplated and there is enough clearance using the Ceretom for

placement is possible and allows for easier access to parietal skull.

RI PT

the head and NexFrame towers placed. 6- Slight head tilt to elevate the side chosen for battery

As noted above, the surgical method required extensive adaption from the method used for conventional DBS and is worth reporting to prospective users. The limitations are induced by

M AN US C

depth, skull and trajectory limitations. This technical report provides key details and nuances that permit a stereotactic neurosurgeon to replicate this procedure in their practice, and broaden patient access to these cutting edge technologies.

We have followed up our small series of patients and obtained high fidelity electrocorticographic recordings for epilepsy management. The large amount of data that can be obtained by a closed loop system heralds a new modality of treatment, one that allows rare

seizures to be captured and analysis of patterns for individual seizure semiologies. In the long-

term, this may permit individual stimulation parameters for each seizure pattern and allow truly tailored epilepsy management for individual patients.

We acknowledge the limitations to this study, this is not an outcome study, it is purely a technical note to aid surgeons in their endeavor. We also acknowledge that conclusions regarding

AC C

safety of this method needs to be interpreted in lieu of the small number of subjects therefore is considered a feasibility report as well. We report herein our experience at a single center; evaluation of the long-term efficacy of this technique would benefit from application in multiple centers with analysis of technical and patient outcome.

CONCLUSION

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

We describe in detail the technical nuances and feasibility of adapting the Medtronic Nexframe frameless stereotactic system for the placement of bilateral mesial temporal RNS electrodes. Our group has extensive experience in the use of Nexframe for DBS electrode

RI PT

placement, and we sought to apply the same methodology to RNS. We noted that this required extensive adaptation and knowledge of intricate surgical nuances to perform this with similar accuracy and reliability. We therefore report in detail surgical techniques and nuances that would

allow other practicing neurosurgeon to replicate this procedure. The authors hope this technical report will be of value to neurosurgeons who practice functional stereotactic operative techniques

M AN U

for epilepsy and allows more adult patients who have not been controlled with two or more AEDs access to this relatively new technology.

REFERENCES

Berkovic SF, McIntosh AM, Kalnins RM, et al. Preoperative MRI predicts outcome of

temporal lobectomy: an actuarial analysis. Neurology 1995;45:1358-1363 Gilliam F, Faught E, Martin R, et al. Predictive value of MRI-identified mesial temporal

sclerosis for surgical outcome in temporal lobe epilepsy: an intent-to-treat analysis. Epilepsia 2000;41:963-966

Kumlien E, Doss RC, Gates JR. Treatment outcome in patients with mesial temporal

AC C

sclerosis. Seizure 2002;11:413-417

Wiebe S, Blume WT, Girvin JP, et al. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311-318

Corkin S. What's new with the amnesic patient H.M.? Nat Rev Neurosci 2002;3:153-160

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Cukiert A, Sousa A, Machado E, et al. Results of surgery in patients with bilateral independent temporal lobe spiking (BITLS) with normal MRI or bilateral mesial temporal sclerosis (MTS) investigated with bilateral subdural grids. Arq Neuropsiquiatr

RI PT

2000;58:1009-1013

Cukiert A, Cukiert CM, Argentoni M, et al. Outcome after cortico-amygdalo-

invasive recording. Seizure 2009;18:515-518 8.

hippocampectomy in patients with severe bilateral mesial temporal sclerosis submitted to

Harroud A, Bouthillier A, Weil AG, Nguyen DK. Temporal lobe epilepsy surgery

RNSÂŽ System - P100026. 2013;

M AN U

failures: a review. Epilepsy Res Treat 2012;2012:201651

http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/DeviceApprovalsan dClearances/Recently-ApprovedDevices/ucm376685.htm. Accessed 1 July, 2016. King-Stephens D, Mirro E, Weber PB, et al. Lateralization of mesial temporal lobe

10.

11.

epilepsy with chronic ambulatory electrocorticography. Epilepsia 2015;56:959-967 Heck CN, King-Stephens D, Massey AD, et al. Two-year seizure reduction in adults with

medically intractable partial onset epilepsy treated with responsive neurostimulation: final results of the RNS System Pivotal trial. Epilepsia 2014;55:432-441 Bergey GK, Morrell MJ, Mizrahi EM, et al. Long-term treatment with responsive brain

AC C

12.

stimulation in adults with refractory partial seizures. Neurology 2015;84:810-817

13.

Morrell MJ. Responsive cortical stimulation for the treatment of medically intractable partial epilepsy. Neurology 2011;77:1295-1304

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

14.

Loring DW, Kapur R, Meador KJ, Morrell MJ. Differential neuropsychological outcomes following targeted responsive neurostimulation for partial-onset epilepsy. Epilepsia 2015;56:1836-1844 Lee B, Zubair MN, Marquez YD, et al. A Single-Center Experience with the NeuroPace

RI PT

15.

RNS System: A Review of Techniques and Potential Problems. World Neurosurg 2015;84:719-726

Sillay KA, Rutecki P, Cicora K, et al. Long-term measurement of impedance in

M AN US C

16.

chronically implanted depth and subdural electrodes during responsive neurostimulation in humans. Brain Stimul 2013;6:718-726 17.

Burchiel KJ, McCartney S, Lee A, Raslan AM. Accuracy of deep brain stimulation electrode placement using intraoperative computed tomography without microelectrode recording. Journal of Neurosurgery 2013;119:301-306

Kelman C, Ramakrishnan V, Davies A, Holloway K. Analysis of stereotactic accuracy of

18.

the cosman-robert-wells frame and nexframe frameless systems in deep brain stimulation surgery. Stereotact Funct Neurosurg 2010;88:288-295 Bot M, van den Munckhof P, Bakay R, et al. Analysis of Stereotactic Accuracy in

19.

Patients Undergoing Deep Brain Stimulation Using Nexframe and the Leksell Frame.

AC C

Stereotact Funct Neurosurg 2015;93:316-325

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

FIGURE LEGENDS Figure 1. Intra-operative images recording key aspects of NexFrame assisted frameless

RI PT

stereotactic placement of medial temporal RNS electrodes; a) the patient is placed in a radiolucent Doro head holder, b) bilateral occipital incisions are planned with a larger incision at the generator site, c, d) skull-based fiducials are placed, e, f) intra-operative CT is performed and

M AN US C

the CT merged with the pre-operative MRI, g) fiducials are registered with the intra-operative CT with the non-sterile reference frame, h) occipital entry points are planned, i) the bilateral curvilinear occipital incisions are opened and burr holes placed at the planned entry sites, j, k) the NexFrame towers are assembled and frameless registration performed, l) the trajectories are planned using the targeting system, m, n) electrodes are placed with the assistance of Microdrive targeting towers, and o, p) secured to the skull, q, r) the craniotomy is planned for generator

placement, s, t) the generator is placed and the wires are connected.

Figure 2. Merged image of the post-placement CT and the pre-operative MRI, which

demonstrates placement of the electrode and each contact in the mesial temporal lobe.

AC C

Figure 3. ECoG recordings from the NeuroPace device, correlated to seizure semiologies; a) baseline, b) aura, c) jolt, d) simple partial, e) complex partial, and f) 3rd person.

Figure 4. Laterality of a) total recorded events, and b) events by correlated seizure semiology from patient 1â&#x20AC;&#x2122;s seizure diary. These demonstrate the variation in laterality of ictal onset between differing seizure semiologies.

M F M M

Age at surgery (years) 44 45 48 30

RNS location Hippocampus Hippocampus Hippocampus Hippocampus

A C

Operating time (minutes) 314 195 225 226

C E

Gender

Electrode deviation (mm) 1.9 1.65 1.55 1.95

Intensive care stay (days) 1 3 1 4

Overall in-patient stay (days) 2 8 3 6

Table 1. Patient demographics, case details, and outcome for bilateral hippocampal NeuroPace electrode placement.

ACCEPTED MANUSCRIPT

S C

M A N U E D P T

R I P T

Patient outcome Engel 2 Engel 3 Deceased Engel 4

AC C

M AN U

RI PT

ACCEPTED MANUSCRIPT

AC C

M AN U

RI PT

ACCEPTED MANUSCRIPT

AC C

M AN U

RI PT

ACCEPTED MANUSCRIPT

AC C

M AN U

RI PT

ACCEPTED MANUSCRIPT

AC C

M AN U

RI PT

ACCEPTED MANUSCRIPT

AC C

M AN U

RI PT

ACCEPTED MANUSCRIPT

AC C

M AN U

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Highlights RNS offers a reversible therapy in bilateral mesial temporal sclerosis

•

Conventional frames can limit surgical exposure for device implantation

•

The authors describe adaption of Nexframe for bitemporal RNS placement

•

RNS provides a high volume of clinically relevant ambulatory electrocorticography

•

These technologies herald a new era for treatment of medically refractory epilepsy

AC C

M AN U

RI PT

•

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Abbreviations AEDs – anti-epileptic drugs

CT – computed tomography DBS – deep brain stimulation ECoG – electrocorticography EEG – electroencephalogram

LE – limbic encephalitis MRI – magnetic resonance imaging RNS – responsive neurostimulation

AC C

VNS – vagal nerve stimulation

M AN U

GAD-Abs – glutamic acid decarboxylase antibody

RI PT

CSF – cerebrospinal fluid

Alterations in Surgical Plan Based on Intraoperative Portable Head Computed Tomography Imaging Alterations in surgical plan based on

Andrew P. Carlson, MD, Jeremy Phelps, MD, Howard Yonas, MD From the Department of Neurosurgery, University of New Mexico, Albuquerque, NM.

intraoperative portable head computed tomography imaging

ABSTRACT BACKGROUND

Intraoperative imaging offers potential for utility in many clinical scenarios. Portable computed tomography (CT) offers a versatile potential alternative when immediate imaging may alter the surgical plan and magnetic resonance imaging is not practical. METHODS

The medical records from the University or New Mexico were reviewed for portable head CT scans done in the operating room since the scanner has been available. Operative reports and imaging studies were reviewed to determine changes in surgical decision after the CT scan. FINDINGS

The portable head CT scanner was used in 50 cases from May 2007 through March 2010. Average operative time overall was 121 minutes and for reservoir placement was 54 minutes. Procedures included reservoir placement (28%), tumor resection (24%), cerebrospinal fluid shunting (24%), vascular lesion resection (8%), trauma craniotomy (6%), abscess drainage (4%), stereotactic biopsy (4%), and open reduction internal fixation of facial fractures (2%). Findings on the CT scan lead to alterations in the surgical plan 16 times (32%).

Keywords: Computed tomography, critical care, portable brain imaging, intraoperative imaging. Correspondence: Address corresponding to Howard Yonas, MD, Chairman and Professor of Neurosurgery, Department of Neurosurgery, University of New Mexico, MSC10–56151 Albuquerque, NM 87131–0001. Email: hyonas@salud. unm.edu. Disclosure: Dr. Carlson and Dr. Phelps have no financial or other relationship with the manufacturer of the described technology. Dr. Yonas owns a minor stock position in Neurologica. No financial support from any source funded this study. J Neuroimaging 2011;XX:1-5 DOI: 10.1111/j.1552-6569.2011.00580.x

CONCLUSION

In select cases, intraoperative portable head CT leads to changes in the surgical plan in 32% of cases. This potentially prevents a return to the operating room and offers a cost-effective alternative to fixed intraoperative imaging facilities.

Introduction Advantages of intraoperative imaging have been described in the literature, including magnetic resonance imaging (MRI)1,2 and fixed computed tomography (CT).3 The high cost and limited versatility of purely intraoperative systems have led to the development of “dual use” scanners that can also be used for diagnostic imaging adjacent to the operating suite.4 Still, these systems are often prohibitively expensive. In addition, special precautions must be taken to operate near the high magnetic field. In addition, the setup and dedicated time with the scanner make it less useful for more “routine” neurosurgical cases, where intraoperative imaging may be useful. Portable CT in some form has been available at many centers for several years.5-9 Intraoperative uses including skull base applications8-10 and tumor resection 6 have been reported. With the acquisition of a newer generation portable head scanner (CereTom- NeuroLogica, Danvers, MA), we have performed over 3000 CT scans, mostly at the bedside (unpublished data). The image quality of the CereTom scanner has been carefully validated.11 We have also used the scanner in select operative cases to aid in decision making. This study examines how frequently changes in the plan were made based on these data.

done in the operating room were further selected, and the operative reports, anesthesia records, and CT images were reviewed. Length of surgery, diagnosis, procedure, role of the CT scanner, and incidence of changes to the operative plan were noted.

Results Since the acquisition of the portable CT machine, 50 intraoperative portable head CT scans have been performed (between May of 2007 and March of 2010). Average time of surgery was 121 minutes for all cases (range 31-563 minutes) and 54 minutes for CSF reservoir placement only. The diagnoses and types of procedures are listed in Table 1. Most scans were noncontrasted, and the most common use was for evaluating ventricular catheter position (28 [54%]). The remaining uses are listed in Table 1. In 16 cases (32%), significant changes were made to the operative plan after obtaining the scan due to unexpected findings. Most of these (10) were for repositioning of a ventricular catheter (either for reservoir or for shunting) to ensure optimal positioning. The complete breakdown of changes to surgical plan by procedure type is shown in Figure 1.

Illustrative Cases

Methods A retrospective review of all portable head CT scans done at our institution since the acquisition of the scanner was performed. Local institutional review board approval was obtained. Scans

Case 1 This 55-year-old female had a diagnosis of central nervous system (CNS) leukemia and a cerebrospinal fluid (CSF) reservoir

◦ 2011 by the American Society of Neuroimaging C

Table 1.

Results

Time period Patients Mean OR time (minutes) Mean time for reservoir (minutes) Diagnoses

Procedure

Scan type

Use of scanner (some cases with multiple uses)

Number of times plan altered Type of change

May 2007-March 2010 50 121 (range 31-563) 54 Tumor Lymphoma/Leukemia HCP/ persistent CSF leak Trauma AVM/ cavernoma Pseudotumor cerebrii Abscess CSF/Cyst reservoir Tumor resection VPS/VAS AVM/Cavernoma resection Crani for trauma Abscess drainage Stereotactic biopsy ORIF facial fractures Noncontrasted Contrasted With/without contrast Noncon/CTA Evaluate catheter position Evaluate resection Pre-op/intra-op localization Rule out hematoma or expansion Confirm biopsy of target Confirm fracture reduction Foreign body retrieval 16 (32%) Catheter repositioned Further resection More foreign body removed Bone floated (due to mass effect) Contralateral craniotomy

19 (38%) 10 (20%) 8 (16%) 4 (8%) 4 (8%) 3 (6%) 2 (4%) 14 (28%) 12 (24%) 12 (24%) 4 (8%) 3 (6%) 2 (4%) 2 (4%) 1 (2%) 40 (80%) 4 (8%) 3 (6%) 3 (6%) 28 (54%) 13 (26%) 8 (16%) 4 (8%)

number of cases in which a change to the surgical plan was made after intraoperative portable imaging.

showed a 1-cm right sided subdural hematoma with right to left midline shift and signs of right temporal contusion (Fig 3). In addition, a small epidural hematoma was seen on the left temporal region with an overlying fracture. He was taken to the operating room for a right-sided decompressive craniectomy, and there was concern that decompressing the right side could potentially relieve any tamponade effect stopping the epidural on the left from expanding. A portable head CT was done in the operating room after the decompression that showed significant expansion of the left epidural hematoma, and so a left craniotomy was performed. The intraoperative scan eliminated the need to transport the patient to radiology and back to the operating room, saving significant time in the diagnosis and treatment of the new hematoma.

Case 3

2 (4%) 1 (2%) 1 (2%)

10 3 1

was requested by the oncology service. Given the relatively small ventricular size and the need for accuracy of the catheter holes entirely within the ventricle, intraoperative portable head CT was used. After the first pass, clear CSF was returned, and a CT was performed (Fig 2A). The trajectory was appropriate, but the catheter was thought to be too short, being only visible on two 5 mm slices in the ventricle. A longer catheter was then repassed, and a second CT showed the distal 2 cm of catheter entirely within the ventricle (Fig 2B). This study confirmed positioning, allowing for immediate utilization of the reservoir by the oncology service. The patient would likely have required a revision due to concern of toxicity to the parenchyma if the intraoperative study had not been performed.

Case 2 This 14-year-old male presented with a severe traumatic brain injury after falling off of a moving car. His initial imaging

Fig 1. Chart showing the number of each type of procedure and the

This 51-year-old female presented with worsening headaches and a previous diagnosis of right frontal anaplastic oligodendroglioma. Her new MRI (after 5 years of being lost to followup) showed that the previously small lesion had grown dramatically, occupying most of the right frontal lobe (Fig 4). She was taken for surgical debulking and decompression of the tumor, particularly the severely distorted ventricular system. An intraoperative CT was performed to assess the amount of resection and to determine whether the ventricles were adequately decompressed. The CT showed some residual area of tumor, which was thought to continue to distort the ventricle, as evidenced by slightly different CT attenuation and the presence of calcifications. The remaining calcified tissue was further removed. An MRI 1 month later showed that the ventricular system had returned to midline position and that there was only a small peripheral area of residual contrast enhancement. The intraoperative CT in this case showed some residual tumor that could be safely debulked without risking entering the ventricle.

Journal of Neuroimaging

Fig 2. Axial CT series obtained intraoperatively for ventricular reservoir placement. Series a shows that the ventricular catheter enters the ventricle on only two slices. Series b shows the catheter positioned to the interior frontal horn, confirming that all the ventricular catheter holes are in the ventricle.

Fig 3. Series of axial CT scans from case 2. (A) Initial select CT levels showing the left epidural hematoma and right subdural hematoma with some effacement of the right lateral ventricle and right to left midline shift. The white arrow shows the fracture and the image above shows the associated epidural hematoma at the same level. (B) The intraoperative CT showing decompression of the right hemicranium with expansion of the epidural hematoma seen earlier (black arrow with white rim). (C) Postoperative CT showing hematoma evacuation on the left and temporal decompression on the right.

Discussion It is intuitive that having real time, precise anatomic information intraoperatively may be beneficial in many neurosurgical procedures. Image-guided navigation is one technique, however has some inherent inaccuracy, and becomes increasingly

inaccurate if anatomy changes intraoperatively.12,13 Intraoperative imaging is becoming more common, particularly with availability of dual-use systems,2-4 however remains fixed to a single operating room suite, requires specific instrumentation and is often costprohibitive.

Carlson et al: Alterations in Surgical Plan

Fig 4. Images from case 4. (A) Preoperative T1 MRI with contrast, showing the large right frontal, contrast enhancing lesion. (B) Intraoperative CT, showing continued distortion of the ventricular system and some calcified tumor remaining in the bed (white arrows). (C) One-month postoperative contrasted T1 MRI, showing return to midline of the ventricular system and some peripheral frontal residual enhancement. The CereTom portable-head CT machine is an eight-slice scanner made only for head imagingâ&#x20AC;&#x201D;increasing its portability, but limiting use in the rest of the body or spine. It is capable of performing most standard CT protocols including navigation protocols, contrasted studies, CT angiography, CT perfusion, and xenon/CT.14,15 A special head holder is used for cases where the scanner is to be used, either with a simple carbonfiber headrest, or with a radiolucent pin-fixation system. The field is entirely covered with a c-arm drape to maintain sterile field during scanning. The portability of the scanner has been important for us in that it can be used in any of our operating rooms, rather than in a single room with a fixed scanner. Since many cases where we desired to use the scanner were â&#x20AC;&#x153;smallerâ&#x20AC;? cases such as shunts and biopsies, this versatility is important in being able to use any available room. Accurate placement of ventricular catheters has been our primary utility of the device. Rates of malpositioning of catheters with image-guided systems range from 1.2 to 10%,12,13 and particularly for CSF reservoirs, a misplaced catheter typically requires operative repositioning due to potential for parenchymal toxicity of intrathecal chemotherapy.16 Though fluoroscopy, ventriculography, and endoscopy have been used, it is still recommended that all patients obtain a postoperative CT as the definitive test prior to instilling any agent.16,17 With the availability and relative ease of intraoperative portable-head CT, obtaining this confirmation intraoperatively seems intuitive. In addition, our operative times for reservoirs remained relatively low at less than an hour average. The scanner was used in a wide variety of additional cases, and seems to be useful in any situation where updated knowledge of gross anatomy is desired. It lacks the fine detail of MRI, but can clearly determine if there is gross tumor remaining, if there is still mass effect, and relation of the resection cavity to vital structures that may have shifted. We used CT angiography on several occasions, and this may present a viable alternative to intraoperative angiography, particularly in cases where positioning can be challenging. Since the scanner is capable of interfacing with the image-guided navigation system, the ability to obtain image data intraoperatively and merge to existing

data offers the ability to update navigation accuracy as anatomy changes. Limitations of the system are primarily related to positioning and body habitus. With the scanner gantry positioned at the level of the shoulders when starting the scan, imaging is usually completed starting at the upper cervical region, however in larger patients, accurate skull base imaging can be challenging. Head position must be carefully planned prior to draping if an intraoperative scan is desired. Using portable CT undoubtedly adds some additional operative time to the case as well, however, as the technology becomes more routine this becomes less of a concern.

Conclusions Intraoperative portable-head CT leads to change in operative plans in 32% of selected cases. This potentially avoids return to the operating room and can diagnose or rule out remote lesions in need of additional intervention. The portable configuration allows for increased versatility and cost-effectiveness compared to fixed systems.

References 1. Claus EB, Horlacher A, Hsu L, et al. Survival rates in patients with low-grade glioma after intraoperative magnetic resonance image guidance. Cancer 2005;103(6):1227-1233. 2. Hirschl RA, Wilson J, Miller B, et al. The predictive value of lowfield strength magnetic resonance imaging for intraoperative residual tumor detection. Clinical article. J Neurosurg 2009;111(2):252257. 3. Uhl E, Zausinger S, Morhard D, et al. Intraoperative computed tomography with integrated navigation system in a multidisciplinary operating suite. Neurosurgery 2009;64(5 Suppl 2):231-239; discussion 239-240. 4. Pamir MN, Ozduman K, Dincer A, et al. First intraoperative, shared-resource, ultrahigh-field 3-Tesla magnetic resonance imaging system and its application in low-grade glioma resection. J Neurosurg 2010;112(1):57-69. 5. Butler WE, Piaggio CM, Constantinou C, et al. A mobile computed tomographic scanner with intraoperative and intensive care unit applications. Neurosurgery 1998;42(6):1304-1310; discussion 13101301.

Journal of Neuroimaging

6. Gwinn R, Cleary K, Medlock M. Use of a portable CT scanner during resection of subcortical supratentorial astrocytomas of childhood. Pediatr Neurosurg 2000;32(1):37-43. 7. Mirvis SE. Use of portable CT in the R Adams Cowley Shock Trauma Center. Experiences in the admitting area, ICU, and operating room. Surg Clin North Am 1999;79(6):13171330. 8. Mattox DE, Mirvis SE. Intraoperative portable computed tomography scanning: an adjunct to sinus and skull base surgery. Otolaryngol Head Neck Surg 1999;121(6):776-780. 9. Stanley RB, Jr. Use of intraoperative computed tomography during repair of orbitozygomatic fractures. Arch Facial Plast Surg 1999;1(1):19-24. 10. Das S, Maeso PA, Figueroa RE, et al. The use of portable intraoperative computed tomography scanning for real-time image guidance: a pilot cadaver study. Am J Rhinol 2008;22(2):166-169. 11. Rumboldt Z, Huda W, All JW. Review of Portable CT with Assessment of a Dedicated Head CT Scanner. AJNR Am J Neuroradiol 2009;30(9):1630-1636.

12. Greenfield JP, Schwartz TH. Catheter placement for Ommaya reservoirs with frameless surgical navigation: technical note. Stereotact Funct Neurosurg 2008;86(2):101-105. 13. Takahashi M, Yamada R, Tabei Y, Nakamura O, Shinoura N. Navigation-guided Ommaya reservoir placement: implications for the treatment of leptomeningeal metastases. Minim Invasive Neurosurg 2007;50(6):340-345. 14. Hillman J, Sturnegk P, Yonas H, et al. Bedside monitoring of CBF with xenon-CT and a mobile scanner: a novel method in neurointensive care. Br J Neurosurg 2005;19(5):395-401. 15. Sturnegk P, Mellergard P, Yonas H, et al. Potential use of quantitative bedside CBF monitoring (Xe-CT) for decision making in neurosurgical intensive care. Br J Neurosurg 2007;21(4):332-339. 16. Sandberg DI, Bilsky MH, Souweidane MM, et al. Ommaya reservoirs for the treatment of leptomeningeal metastases. Neurosurgery 2000;47(1):49-54; discussion 54-45. 17. Chamberlain MC, Kormanik PA, Barba D. Complications associated with intraventricular chemotherapy in patients with leptomeningeal metastases. J Neurosurg 1997;87(5):694-699.

Carlson et al: Alterations in Surgical Plan

6/5/2017

Retrograde Partial Migration of Ventriculoperitoneal Shunt with Chamber: Review of Causative Factors and Its Prevention

J Pediatr Neurosci. 2017 JanMar; 12(1): 93–95. doi: 10.4103/18171745.205654

Retrograde partial migration of

PMCID: PMC5437804

ventriculoperitoneal shunt with chambre:

Retrograde Partial Migration of Ventriculoperitoneal Shunt with Review of causative factors and its prevention Chamber: Review of Causative Factors and Its Prevention Harsha A. Huliyappa, Manish Jaiswal,1 Sunil K. Singh,1 Balakrishna Ojha,1 Anil Chandra,1 and Srivastava Chhitij1 Department of Neurosurgery, JSS Medical College and University, Mysore, Karnataka, India 1 Department of Neurosurgery, King George's Medical University, Lucknow, Uttar Pradesh, India Address for correspondence: Dr. Harsha A. Huliyappa, Department of Neurosurgery, JSS Medical College and University, Mahathma Gandhi Road, Mysore 570 004, Karnataka, India. Email: harsha_huliyappa@yahoo.com Copyright : © 2017 Journal of Pediatric Neurosciences This is an open access article distributed under the terms of the Creative Commons AttributionNonCommercialShareAlike 3.0 License, which allows others to remix, tweak, and build upon the work noncommercially, as long as the author is credited and the new creations are licensed under the identical terms.

Abstract Distal migration of shunt is a very common occurrence. Proximal migration of shunt is rare and possible pathophysiological mechanisms to explain this unusual complication is rarely attempted. A 5monthold child shunted for posttraumatic hydrocephalus presented 1.5 years later with raised intracranial pressure and seizures. Imaging showed subdural hygroma, partial intracranial migration of shunt/chamber. On endoscopy, choroid plexus was adherent to shunt tip and some pericranial tissue was found in the anchoring suture (intraventricularly displaced). Shunt was retrieved endoscopically and diversion established by endoscopic third ventriculostomy with symptoms free followup. Hostrelated and surgical factors have been postulated. Tugofwar effect on the anchoring suture and collapsing cortex are the possible mechanisms that explain proximal migration in our case. Threepoint fixation of the chamber to pericranium, small burr hole with a smaller durotomy, can prevent shunt migration. Proximal shunt migrations should be dealt with endoscopy so as to avoid complications. K‫܀‬禁論码碁‫܀‬: Hypothesis of shunt migration, neuroendoscopy, shunt chamber migration

Iោ󃐀ퟌ ퟰោퟌ Distal migration of ventriculoperitoneal (VP) shunt following detachment from the chamber is a common complication, but complete proximal migration of VP shunt into the ventricle is exceptionally rare with a few anecdotal case reports. Proximal migration of VP shunt may present with shunt malfunction as well as additional features such as seizures and subgaleal coiling. Any breach in the continuity of the shunt also adds upon to risk of developing meningitis. We are presenting an endoscopically managed case of ventricular migration of an intact medium pressure VP shunt.

C焄埛漄 R漄洄ퟌ󃐀ោ A 5monthold male child involved in a road traffic accident presented to casualty in E4V5M6 status. Noncontrast computed tomography (NCCT) of head showed minimal 3rd ventricular intraventricular hemorrhage (IVH) with normal ventricles. Repeat NCCT on day 3 showed ventriculomegaly with complete resolution of IVH with features of raised intracranial pressure (ICP). Hence, a medium pressure VP shunt was placed through the right Keen's point with resolution of symptoms. Six months later, the https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5437804/?report=printable

1/5

6/5/2017

Retrograde Partial Migration of Ventriculoperitoneal Shunt with Chamber: Review of Causative Factors and Its Prevention

patient developed features of shunt malfunction, for which the entire shunt assembly including the ventricular end was replaced. One year later, the patient developed an intermittent headache and three episodes of seizures despite on antiepileptic drug (valproate). On examination, the patient had no visual deficits. Head circumference was normal for the age. Shunt chamber was not palpable at its normal position in the region of mastoid. NCCT head showed rightsided subdural hygroma, ventriculomegaly, with an abnormally large loop of ventricular catheter. Chamber was localized in the right occipital horn [Figure 1a and b]. Three dimensional NCCT reconstruction with minimal bone Subtraction (obtained by neurosurgery residents using our portable CT – CERE TOM™) showed chamber and almost 10 cm of shunt migration intracranially [Figure 1c]. Intracranial location of the chamber with the continuity of the entire system was evident even on skiagram [Figure 1d]. Cerebrospinal fluid (CSF) analysis was normal. The presence of subdural hygroma was suggestive of a still functioning shunt, but clinical features of raised ICP lead to a consensus decision of endoscopic exploration. A possibility of complete migration of the unsterile abdominal end into the ventricle was also kept in mind. Through left Kocher's point, neuroendoscope was inserted and showed moderate pressure. Shunt end slits were filled with choroid plexus which was released by saline irrigation and gentle manipulation. It was noted that the suture material used for anchoring had taken a small bit of pericranial tissue along with intraventricularly [Figure 2a]. The shunt was first cut at the parietal incision site (from externally). Endoscopically, the shunt tip was grasped end on and was delivered out under endoscopic guidance along endoscopic tract [Figure 2b]. Entire chamber with the shunt was delivered out. Endoscopic third ventriculostomy (ETV) was performed successfully. Through a small abdominal incision, the distal end of the catheter was removed. Patient recovered well after ETV and was discharged on postoperative day 7. NCCT showed B/L minimal subdural hygroma. Followup at 6 months and 2 years showed arrested ventriculomegaly with minimal subdural hygroma and asymptomatic course.

D埛ퟰ 埛埛ퟌ Hydrocephalus of various etiologies is common in the pediatric age group. VP shunt surgery and ETV are universally accepted procedures with various success rates based on the disease. Nevertheless, shunt surgery is associated with many complications such as shunt malfunction, shunt infection, pseudocyst formation/CSFomas, cutaneous exposure, bowel perforation, and its presentation through aboral, migration of the tube into pleural cavity, liver, heart, scrotum, abdominal wall, and oral cavities.[1,2,3] Dislodgement and migration of the distal portion of the shunt are commonly encountered at least once in 3–5 years of a neurosurgical career and pose no difficulty in their management. Proximal migration, with its rare incidence (0.1%–0.4%), has only anecdotal casebased management decisions.[4,5] Many hypothesis and factors have been postulated for the migration of shunt. Host factors such as younger age, thin cortical mantle, malnutrition, excessive neck movements producing a windlass effect coupled with a large potential subgaleal space or dilated ventricles with negative suctioning pressure or a positive intraabdominal pressure, patient's habit of rubbing the chamber area has been considered in many studies.[4,6,7] Surgical factors such as inadvertently large burr hole, wide durotomy, and inadequate anchorage to the pericranial tissues have been postulated. A large burr hole with a large dural rent may result in a subgaleal pocket with enough CSF acting like a sump sucking the ventricular catheter into the subgaleal pocket.[8] Chabra's shunt which has a cylindrical chamber has been implicated.[9] Short distance between the ventricular and abdominal end in young patients and rapid decompression of larger hydrocephalus are additional events. [4,5,9] In our case, young child, in a rapid growth phase, slackened pericranial anchorage, and a larger burr hole are evident. We also propose two theories for possible mechanism. Tugofwar

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5437804/?report=printable

2/5

6/5/2017

Retrograde Partial Migration of Ventriculoperitoneal Shunt with Chamber: Review of Causative Factors and Its Prevention

A constant pulsatile thrust exerted by the entrapped choroid plexus at the tip of the shunt and a constant dragging force from a larger distal system can have a tugofwar effect at the fixed anchorage point near the burr hole. This repeated to and fro movement may snap the pericranial tissue along with the suture. Even though we follow a threepoint anchorage to pericranium, in this case, only a singlepoint anchorage had been performed. Collapsing cortex Development of subdural hygroma with receding cortex may add on to more negative pull from the ventricular side. This hypothesis can explain the migration of shunt if the stay sutures have given away as might be in or case. However, it is not evident from literature search about an association between proximal migration and development of subdural hygroma. Unguided pulling off a migrated shunt can lead to catastrophic consequences. In the era of endoscopic interventions, the added advantage of visualizing the pathology and guided shunt removal thereby mitigating injury to the cortex, choroid plexus, and veins. Naik et al. described the first case of endoscopically managed total intracranial shunt migration without any complication.[4] It also allows procedures such as adhesiolysis, septostomy, and ETV thus eliminating shunt complications. In literature, most of the cases of proximal shunt migrations are of whole assembly type which allows only endoscopic removal as an option or a craniotomy. However, a partial proximal migration has not been reported. In partial migration with dysfunction, after endoscopic insertion, percutaneous disconnection is to be done followed by removal under endoscopic visualization. Financial support and sponsorship This study was supported by the Department of Neurosurgery, King George's Medical University, Lucknow – 226 003, Uttar Pradesh, India. Conflicts of interest There are no conflicts of interest.

Acknowledgment We thank all the staff in the Department of Neurosurgery for helping in preparing this case report.

R漄ࠄ漄󃐀漄ퟰ漄埛 1. Jindal A, Kansal S, Mahapatra AK. Unusual complication – VP shunt coming out per rectum and brain abscess. Indian J Pediatr. 1999;66:463–5. [PubMed: 10798095] 2. Agarwal A, Kakani A. Shunt malfunction due to proximal migration and subcutaneous coiling of a peritoneal catheter. J Neurosci Rural Pract. 2010;1:120–1. [PMCID: PMC3139341] [PubMed: 21808520] 3. Sridhar K, Karmarkar V. Peroral extrusion of ventriculoperitoneal shunt: Case report and review of literature. Neurol India. 2009;57:334–6. [PubMed: 19587479] 4. Naik V, Phalak M, Chandra PS. Total intracranial shunt migration. J Neurosci Rural Pract. 2013;4:95–6. [PMCID: PMC3579071] [PubMed: 23546373] 5. Villarejo F, AlvarezSastre C, Gimenez D, Gonzalez C. Migration of an entire onepiece shunt into the ventricle. Neurochirurgia (Stuttg) 1979;22:196–8. [PubMed: 388246] 6. Agarwal A, Kakani A. Total migration of a ventriculoperitoneal shunt catheter into the ventricles. J Pediatr Neurosci. 2011;6:88–9. [PMCID: PMC3173929] [PubMed: 21977102] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5437804/?report=printable

3/5

6/5/2017

Retrograde Partial Migration of Ventriculoperitoneal Shunt with Chamber: Review of Causative Factors and Its Prevention

7. Yee GT, Han SR, Choi CY. Migration and coiling of peritoneal catheter into the subgaleal space: A very rare complication of subgaleoperitoneal shunt. J Korean Neurosurg Soc. 2013;54:525–7. [PMCID: PMC3921284] [PubMed: 24527199] 8. Shahsavaran S, Kermani HR, Keikhosravi E, Nejat F, El Khashab M. Ventriculoperitoneal shunt migration and coiling: A report of two cases. J Pediatr Neurosci. 2012;7:114–6. [PMCID: PMC3519067] [PubMed: 23248689] 9. Young HA, Robb PJ, Hardy DG. Complete migration of ventriculoperitoneal shunt into the ventricle: Report of two cases. Neurosurgery. 1983;12:469–71. [PubMed: 6343910]

Figures and Tables Figure 1

(a and b) Noncontrast computed tomography head showing rightsided subdural hygroma, ventriculomegaly, with an abnormally large loop of ventricular catheter. Shunt chamber was localized in the right occipital horn as a thick tube with a lumen. Furthermore, a long segment of shunt ~ 10 cm lied in the subdural fluid. (c) Threedimensional noncontrast computed tomography reconstruction with minimal manual bone subtraction showed chamber (arrow) and almost 10 cm of shunt migration intracranially. (d) An intracranial location of the chamber with the continuity of the entire system was evident even on skiagram

Figure 2 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5437804/?report=printable

4/5

6/5/2017

Retrograde Partial Migration of Ventriculoperitoneal Shunt with Chamber: Review of Causative Factors and Its Prevention

(a) Endoscopic view showing the intraventricular location of connector anchorage suture with pericranial tissue. (b) Endoscopically, the shunt tip was grasped and was delivered out under endoscopic guidance after separating from the adjacent choroid Articles from Journal of Pediatric Neurosciences are provided here courtesy of Medknow Publications

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5437804/?report=printable

5/5

CLINICAL ARTICLE

Early low cerebral blood flow and high cerebral lactate: prediction of delayed cerebral ischemia in subarachnoid

hemorrhage Early low cerebral blood flow and high cerebral lactate: prediction of delayed cerebral ischemia in subarachnoid hemorrhage Elham Rostami, MD, PhD,1 Henrik Engquist, MD,1,2 Timothy Howells, PhD,1 Ulf Johnson, MD, PhD,1,3 Elisabeth Ronne-Engström, MD, PhD,1 Pelle Nilsson, MD, PhD,1 Lars Hillered, MD, PhD,1 Anders Lewén, MD, PhD,1 and Per Enblad, MD, PhD1 1 Section of Neurosurgery, Department of Neuroscience, 2Anesthesiology and Intensive Care, Department of Surgical Sciences, and 3Department of Surgical Sciences and Radiology, Uppsala University, Uppsala, Sweden

OBJECTIVE Delayed cerebral ischemia (DCI) following subarachnoid hemorrhage (SAH) is one of the major contributors to poor outcome. It is crucial to be able to detect early signs of DCI to prevent its occurrence. The objective of this study was to determine if low cerebral blood flow (CBF) measurements and pathological microdialysis parameters measured at the bedside can be observed early in patients with SAH who later developed DCI. METHODS The authors included 30 patients with severe SAH. The CBF measurements were performed at Day 0–3 after disease onset, using bedside xenon-CT. Interstitial glucose, lactate, pyruvate, glycerol, and glutamate were measured using microdialysis. RESULTS Nine of 30 patients developed DCI. Patients with DCI showed significantly lower global and regional CBF, and lactate was significantly increased in these patients. A high lactate/pyruvate ratio was also detected in patients with DCI. CONCLUSIONS Early low CBF measurements and a high lactate and lactate/pyruvate ratio may be early warning signs of the risk of developing DCI. The clinical value of these findings needs to be confirmed in larger studies. https://thejns.org/doi/abs/10.3171/2016.11.JNS161140

KEY WORDS xenon-CT; lactate; CBF; cerebral blood flow; lactate/pyruvate ratio; SAH; subarachnoid hemorrhage; DCI; delayed cerebral ischemia; vasospasm; vascular disorders

atients with aneurysmal subarachnoid hemorrhage (SAH) suffer high rates of mortality and morbidity as a result of the primary hemorrhage and secondary injuries. Whereas the primary hemorrhagic injury is difficult to prevent, the secondary injuries may be prevented if detected in time. Hemodynamic and metabolic disturbances have been detected in the acute stage of SAH in both animals and humans.7,13 These early disturbances make the brain vulnerable to secondary injuries. Secondary ischemic brain injury is one of the major contributors to poor outcome. Although cerebral vasospasm has been considered to be the major instigator of secondary ischemia, the discrepancy between angiographic findings and the development of clinical signs15,27 suggests additional

contributing factors. It is now believed that the causes are multifactorial and may involve endothelial dysfunction, blood-brain barrier disruption, cortical spreading depolarizations, microvascular thrombosis, and failure of cerebral autoregulation. Current guidelines for the critical care management of patients with SAH recommend the use of the term “delayed cerebral ischemia” (DCI) for this secondary brain injury, which is characterized by focal neurological impairment, decrease of Glasgow Coma Scale (GCS) score, and/or radiological signs of ischemia or infarction.5 To prevent DCI it would be ideal to detect early signs of hemodynamic or metabolic disturbances at the bedside. This would also help in identifying patients who are vul-

ABBREVIATIONS CBF = cerebral blood flow; CPP = cerebral perfusion pressure; DCI = delayed cerebral ischemia; GCS, GCS-M = Glasgow Coma Scale, GCS-motor; GOSE = Glasgow Outcome Scale–Extended; ICP = intracranial pressure; L/P = lactate/pyruvate; MD = microdialysis; NIC = neurointensive care; ROI = region of interest; SAH = subarachnoid hemorrhage; Xe-CT = xenon-CT. SUBMITTED May 3, 2016. ACCEPTED November 17, 2016. INCLUDE WHEN CITING Published online June 2, 2017; DOI: 10.3171/2016.11.JNS161140. ©AANS, 2017

J Neurosurg June 2, 2017

E. Rostami et al.

nerable to secondary brain insults in general. Xenon-CT (Xe-CT) has been shown to be feasible and valuable in detecting early signs of vasospasm and/or low cerebral blood flow (CBF) and to predict development of infarction.28,36 Microdialysis (MD) has been used extensively in patients with SAH, with promising results.9 We have previously reported on the feasibility of combining bedside Xe-CT with MD in patients suffering an SAH.17 The objective of the current study was to investigate whether early bedside CBF measurements in combination with MD metabolites could differentiate between patients with SAH who develop DCI and those who do not.

Methods

Study Population and Design Many details of the following protocols were first reported by us elsewhere.16 Patients suffering an SAH were included in the study if they had undergone intracerebral MD studies and were examined with Xe-CT at Day 0–3 following symptom onset. This time period for Xe-CT was chosen because vasospasm and DCI are almost never seen before Day 3 post-SAH.1,30,32 The Xe-CT studies required mechanical ventilation for administration of the xenon gas, and MD was only used in patients who received a ventriculostomy (i.e., patients with more severe SAH were selected). Xenon-CT scanning and insertion of an MD catheter were not always possible for logistical reasons. Patients with a preexisting neurological deficit or an SAH resulting from trauma or arteriovenous malformation were excluded. The SAH was verified by CT scanning, and the aneurysm was visualized by CT angiography or digital subtraction angiography. All patients included in the study were enrolled within 24 hours of the onset of SAH. Clinical data were collected from the electronic records and the neurointensive care (NIC) unit database for patients with SAH. The Fisher grade was assessed by the responsible neurosurgeon at admission. A neuroradiologist examined the CT scans at discharge to determine whether there were signs of ischemia or infarction. A nurse assessed the Glasgow Outcome Scale–Extended (GOSE) score at the 1-year follow-up.10 Neurointensive Care All patients suffering an SAH were managed according to a standardized protocol.18 All unconscious patients and those with clinical and radiological signs of intracranial hypertension receive a ventriculostomy. If the intracranial pressure (ICP) is > 20 mm Hg, the drainage system is opened and CSF is drained against a pressure level of 15 mm Hg. Identified aneurysms are treated early by endovascular coiling or surgical clipping. All patients receive 0.2 mg/ml nimodipine (Nimotop, Bayer AB). If the blood pressure allows it, the goal rate is 10 ml/hour. A daily fluid substitution of 2.5 L is used before treatment of the aneurysm, and 3.0 L is administered after the treatment. The volume status is maintained by fluid administration in the higher normal range, with addition of albumin infusion (Flexbumin, 200 mg/ml; Baxter AG) if needed, and patients are monitored by central venous pressure mea2

surement, clinical evaluation, and rigorous fluid balance calculation. Hypotension is treated first with crystalloid solutions and albumin 20% and second with dobutamine if needed. The goal is to keep the cerebral perfusion pressure (CPP) > 60 mm Hg. As a part of our protocol a wake-up test is performed 6 times per 24 hours.26 The patients received the diagnosis of DCI when delayed ischemic neurological deterioration and/or deficits occurred that could not be explained by other reasons. The DCI was treated with hypertension, hypervolemia, and hemodilution therapy (triple-H therapy) by the administration of 500 ml/day dextran 40 solutions (Rheomacrodex, 100 mg/ml; Meda AB) and albumin 100 ml × 2 (200 mg/ml; Baxter Medical AB). Cerebral Blood Flow Measurements Xenon-CT is performed at the bedside in the NIC unit, by using a mobile CT scanner (CereTom, Neurologica), at Day 0–3 after admission in all patients with SAH who are receiving mechanical ventilation. The CBF measurements are done according to the principles described by Yonas et al.12,33–35 The xenon gas is radiopaque, highly lipid soluble, and capable of crossing the blood-brain barrier. The Kety-Schmidt equation is applied to measure regional and global CBF.11 Stable xenon, at a concentration of 28%, is administered to the patient’s breathing circuit for approximately 4 minutes, using the Enhancer 3000 and the computer software designed for the purpose (Diversified Diagnostic Products, Inc.). During the xenon inhalation, CT scans are obtained. The xenon delivery and the CT scans are synchronized by the computer software, and the resulting radiological tissue enhancement of the xenon wash-in enables CBF (in ml/100 g/min) to be calculated and plotted as colored maps in scans at 4 different levels of the brain (8 scans per level, 2 baseline and 6 enhanced, with 10-mm spacing). Analysis of CBF measurements has been previously described in detail.17 The mean blood flow in each of 20 evenly distributed cortical regions of interest (ROIs) is calculated for each level, and the global CBF is given as a mean of all 4 levels. The tip of the MD catheter was identified on the structural CT scans, and an ROI was drawn manually (diameter 3 cm) for the corresponding area around the MD catheter on the CBF scans (Fig. 1). Regions with CTdefined hematoma or artifact were noted and excluded. Cerebral MD The cerebral MD technique in the NIC unit has previously been extensively used and described.8,29 The intracerebral MD catheter was placed in the right frontal lobe cortex through a separate bur hole, anterior to the ventricular drain. For intracerebral MD monitoring, we used a No. 70 brain MD catheter (M Dialysis AB) with a membrane length of 10 mm and a membrane cutoff of 20,000 D. The catheters were perfused with artificial CSF (NaCl 147 mmol/L, KCl 2.7 mmol/L, CaCl2 1.2 mmol/L, MgCl2 0.85 mmol/L) (Perfusion Fluid CNS, M Dialysis AB). The perfusion rate was 0.3 μl/minute using a microinjection pump (CMA-106, M Dialysis AB). The MD urea was monitored to validate catheter performance.16 The

J Neurosurg June 2, 2017

Low CBF and high lactate: predictors of DCI in SAH

FIG. 1. Following xenon delivery, tissue enhancement of the xenon wash-in enabled CBF to be calculated and plotted as colored maps (panels A and C). Twenty cortical ROIs were used for CBF calculation (A). Conventional CT images were obtained and the location of the MD catheter was identified (B). The CBF around the MD catheter was calculated by drawing an ROI manually (C). Scale for CBF is given on the right side of panel C (in ml/100 g/min).

MD samples were collected on an hourly basis. The MD samples for the 3-hour period before and the 3-hour period after the Xe-CT examination were used for analysis. Interstitial glucose, lactate, pyruvate, glutamate, glycerol, and urea were analyzed enzymatically using a CMA 600 analyzer or ISCUS Clinical Microdialysis Analyzer (M Dialysis AB). Statistical Analysis All analyses were performed using SPSS Statistics for Macintosh, version 23.0 (IBM Corp.). To assess the normality of the data set, the skewness and kurtosis of the distribution were analyzed. Results are expressed as the mean ± SD. A p value of < 0.05 was considered statistically significant. Ethics Approvals Uppsala University’s regional ethical review board for clinical research granted permission to undertake the study. Written informed consent was obtained from all patients or their proxy for study participation. The study was also approved by the local radiation safety authority.

Results

A total of 401 patients with SAH (282 with aneurysmal SAH) were admitted during the period between October 2012 and May 2015. Sixty-four patients underwent Xe-CT examination, of whom 30 also received an MD catheter and were included in this study. There were no adverse effects associated with Xe-CT examinations and no complications were caused by MD catheter insertion. The clinical characteristics of the 30 patients included in the study as well as the aneurysm locations are given in Table 1. The majority of the patients had aneurysms in the anterior circulation. Physiological conditions before and after the Xe-CT examinations are presented in Table 2. Delayed Cerebral Ischemia Nine of 30 patients developed DCI. There was no sig-

nificant difference in age between the DCI and no-DCI groups, and neither were there any differences in CT Fisher scores, Hunt and Hess grades, GCS-motor (GCS-M) scores at admission, or GCS-M scores at discharge (Table 3). The sedation and physiological conditions during XeCT were also similar (see Table 2). Cerebral Blood Flow Measurements The mean global CBF for all patients was 33.3 ± 13.5 ml/100 g/min (range 13.7–73.9 ml/100 g/min). The mean CBF around the MD catheter for all patients was 30.4 ± 12.6 ml/100 g/min (range 11–56.9 ml/100 g/min). There was no significant difference between the global CBF and CBF around the MD catheter (p = 0.39). There was a positive correlation between the global CBF and the MD CBF (r = 0.924, p < 0.0001). Patients who went on to develop DCI had significantly lower global CBF values than patients who did not (23.7 ± 6.7 ml/100 g/min [range 13.7–34.5 ml/100 g/min] and 37.5 ± 13.7 ml/100 g/min [range 21.5–73.9 ml/100 g/min], respectively; p = 0.005). Illustrative cases are presented in Fig. 2. Patients who developed DCI also had significantly lower values for CBF around the MD catheter than did no-DCI patients (21.4 ± 5.3 ml/100 g/min [range 11–29.1 ml/100 g/min] in the DCI group and 34.4 ± 12.8 ml/100 g/ min [range 18.9–56.9 ml/100 g/min] in the no-DCI group; p = 0.007) (Fig. 3). Microdialysis Parameters The mean values and ranges of glucose, lactate, pyruvate, glycerol, glutamate, and the lactate/pyruvate (L/P) ratio for all patients are given in Table 4. The L/P ratio was higher in the DCI group than in the no-DCI group (32.1 ± 16 vs 24.9 ± 8.1), but the difference was not statistically significant (p = 0.065). Lactate was significantly higher in the DCI group than in the no-DCI group (4.8 ± 2.2 mmol/L and 3.4 ± 1.7 mmol/L, respectively; p = 0.029) (Fig. 4). Glucose did not show any significant difference between the DCI and the no-DCI group (2.1 ± 1.2 mmol/L and 2.3 ± 1.1 mmol/L, respectively) (Fig. 5). There also J Neurosurg June 2, 2017

E. Rostami et al.

TABLE 1. Demographic and clinical characteristics in 30 patients with severe SAH Variable

Value

Age in yrs (range) Female/male Coiling/clipping GCS-M score on admission 1 2 3 4 5 6 GCS-M score at discharge 1 2 3 4 5 6 Hunt & Hess grade I II III IV V Fisher grade 1–2 3 4 Aneurysm location Anterior communicating artery Internal carotid artery Posterior communicating artery Middle cerebral artery Anterior choroidal artery Posterior cerebral artery Basilar artery Posterior inferior cerebellar artery Unknown

58.9 (28–84) 25/5 25/4 1 0 2 2 10 15 1 1 0 6 4 18 2 6 9 11 2 0 8 22 9 4 5 6 1 1 1 2 1

Values are number of patients unless otherwise indicated.

were no significant differences between the groups for pyruvate, glycerol, or glutamate. However, in both the DCI and the no-DCI groups there was very large within-group variation in glutamate between the patients. Cerebral Blood Flow and Lactate In the group of patients who later developed DCI, all had global CBF values < 28 ml/100 g/min, except 2 patients with global CBF of 32.9 ml/100 g/min and 34.5 ml/100 g/ min, respectively (CBF-MD 29.1 and 26.8 ml/100 g/min, respectively). One of these 2 patients had a lactate value of 4

TABLE 2. Physiological parameters before and after Xe-CT measurements in all patients, in patients who developed DCI, and in those who did not

Parameter Before Xe-CT Po2 (kPa) Pco2 (kPa) Fio2 (%) MAP (mm Hg) ICP (mm Hg) CPP (mm Hg) After Xe-CT Po2 (kPa) Pco2 (kPa) Fio2 (%) MAP (mm Hg) ICP (mm Hg) CPP (mm Hg)

All

No DCI

DCI

p Value (no DCI vs DCI)

13.3 ± 2.8 5.1 ± 0.4 39.3 ± 10.3 94.1 ± 14.1 17.0 ± 5.0 77.6 ± 15.5

13.4 ± 3.0 13.0 ± 2.2 5.0 ± 0.3 5.1 ± 0.5 40.5 ± 10.3 36.6 ± 10.3 90.9 ± 10.0 101.8 ± 19.6 17.0 ± 5.6 16.9 ± 3.6 73.7 ± 10.4 86.6 ± 21.5

0.74 0.69 0.36 0.07 0.95 0.040

13.6 ± 3.0 5.2 ± 0.5 39.6 ± 10.1 90.5 ± 11.8 17.5 ± 4.7 76.9 ± 13.1

13.9 ± 3.4 5.3 ± 0.5 40.7 ± 10.2 88.4 ± 11.7 18.0 ± 5.2 72.4 ± 10.9

0.42 0.67 0.40 0.17 0.37 0.006

12.9 ± 2.0 5.2 ± 0.5 37.2 ± 10.0 95.3 ± 11.3 16.2 ± 3.2 87.1 ± 12.6

Fio2 = fraction of inspired oxygen; MAP = mean arterial pressure; Pco2 = partial pressure of carbon dioxide; Po2 = partial pressure of oxygen. Values are expressed as the mean ± SD. Boldface type indicates statistical significance.

2.6 mmol/L, and all other patients in the group who developed DCI had values ≥ 3.7 mmol/L (Table 5). In the 21 patients who did not develop DCI, 3 patients had a global CBF of 21.5–26.7 ml/100 g/min (CBF-MD 18.9–23 ml/100 g/min), and all the other patients had values > 28 ml/100 g/min. The lactate levels for these 3 patients were 2.5–3.5 mmol/L. Three other patients in the group who did not develop DCI had high lactate values (5.8–7.6 mmol/L), and these patients had a high global CBF of 35–41 ml/100 g/min. Glasgow Outcome Scale–Extended Scores and CT Scans At 1-year follow-up 5 patients had died, and 2 of these patients were in the DCI group. Both groups had patients with good recovery at 1-year follow-up, and no significant difference was observed between the groups. The frequency of each GOSE score for both groups is given in Table 6. The GOSE scores for patients who later developed DCI are given in Table 5. TABLE 3. Results of clinical data comparing patients who later developed DCI and those who did not Parameter

No DCI

DCI

p Value

Age in yrs GCS-M score at admission GCS-M score at discharge Fisher grade Hunt & Hess grade

58.5 (28–84) 4.9 (1–6) 5.3 (3–6) 3.8 (3–4) 3.4 (1–5)

58.5 (48–75) 5.6 (4–6) 5.3 (4–6) 3.6 (3–4) 2.9 (2–4)

0.81 0.06 0.71 0.22 0.12

Values are expressed as the mean (range).

J Neurosurg June 2, 2017

Low CBF and high lactate: predictors of DCI in SAH

FIG. 2. Xenon-CT scans of illustrative cases. Left: A patient who did not develop DCI. Right: A patient with low global CBF who later developed DCI. Scale for CBF is given on the right side of the right panel (in ml/100 g/min).

The CT scans at discharge were evaluated, and in the no-DCI group, 4 patients had infarction in 1 hemisphere and 2 patients had it in both hemispheres. No signs of infarction were detected at this time point in the remaining patients. In the group of patients who developed DCI, 1 patient had signs of cortical infarct in 1 hemisphere, 4 patients had it in both hemispheres, and no infarction could be detected in the remaining 4 patients in this group (Table 5).

Discussion

We present results in 30 patients suffering an SAH who had MD monitoring of cerebral metabolism and who underwent CBF measurements at bedside, in which Xe-CT was used. There were no complications associated with the monitoring, and we found the combination of these methods feasible. The results represent values obtained in the first days after hemorrhage (Days 0–3), and before any clinical signs appeared pathological values were documented in patients who developed DCI. Nine patients who developed DCI had significantly higher lactate and lower CBF. They also tended to have a higher L/P ratio, although the increase was borderline significant. These results confirm our previous findings that low CBF is associated with

FIG. 3. Box plot of global CBF and CBF in the region around the MD catheter (CBF_MD) in patients who later developed DCI and those who did not. The difference between the groups was significant in both measurements.

disturbed energy metabolism, and that low CBF as well as disturbed energy metabolism according to MD may identify patients who will develop DCI before clinical signs appear.17 Larger studies are required to assess the clinical value of these findings. There are few reported studies that combined CBF measurements with MD in patients with SAH. The first study was performed by Enblad et al., who used PET and MD in 6 patients with SAH. These investigators detected regional ischemia by PET and found that lactate, L/P ratio, and glutamate had the highest sensitivity for detecting ischemia.6 In 2 additional small studies, which included 13 and 15 patients, respectively, PET was combined with MD, and pathological MD parameters were found to be associated with low regional CBF values,21,22 with the strongest association being between glutamate and low CBF, followed by glycerol. The L/P ratio was the most sensitive and specific parameter for detecting a longer period of is-

TABLE 4. Microdialysis parameters in all patients, in patients who developed DCI, and in those who did not Parameter

All

No DCI

DCI

p Value

Glucose (mmol/L) Lactate (mmol/L) Pyruvate (μmol/L) Glycerol (μmol/L) Glutamate (μmol/L) L/P ratio

2.3 (0.5–5.3) 3.9 (1.4–10.4) 143.6 (66.7–249.4) 147.5 (18–577) 32.2 (0.2–254) 27.6 (15.2–71.2)

2.3 (0.5–5.3) 3.4 (1.4–7.6) 136.1 (66.7–249.4) 129 (18–475) 33.5 (0.2–254) 24.9 (15.2–47.8)

2.1 (0.6–4.2) 4.8 (2.5–10.4) 156.4 (108.7–223.1) 167.8 (28–577) 23.9 (1.4–66.8) 32.1 (16.6–71.2)

0.98 0.029 0.20 0.48 0.86 0.065

Microdialysis values are expressed as the mean (range) in all patients. The p values are calculations made to compare the difference between patients who later developed DCI and those who did not. Boldface type indicates statistical significance. J Neurosurg June 2, 2017

E. Rostami et al.

FIG. 4. Box plots of microdialysate lactate and L/P ratio in patients who later developed DCI and those who did not. Lactate was significantly higher in patients who later developed DCI.

chemia. In symptomatic patients significantly higher L/P ratio and levels of lactate and glutamate were seen.22 The results from these studies confirm our findings that a high L/P ratio and high lactate are associated with reduction of CBF, and that patients with signs of DCI show high lactate and L/P ratio. In contrast to our current results, in both studies glutamate was found to be a sensitive marker of ischemia. This could be partly explained by the differences between the studies, such as number of patients, time period of MD analyzed, and the location of the MD catheter. In the current study we included more patients, and MD values at the time of CBF measurements were chosen. Enblad et al. analyzed MD with a flow rate of 2 μl/minute every 10–15 minutes during PET,6 whereas Sarrafzadeh et al. analyzed MD hourly with a flow of 0.3 μl/minute, and used a mean of 3 days for their analysis.23 Furthermore, Sarrafzadeh et al. placed the MD catheter in the vascular territory at risk for developing vasospasms, whereas we placed the MD catheter in the right frontal lobe. The

glutamate levels at the time of CBF measurements in our study were highly variable both in the DCI group and the no-DCI group. There is also a wide variation in the time of CBF measurements in previous studies, with a range of 2–24 days after the hemorrhage. We used the time frame of 0–3 days, the time period before the occurrence of vasospasm.32 One study combined Xe-CT and MD in 16 patients with SAH during Days 3–10 and could not find any correlation between L/P ratio and clinical status.4 Several studies have used MD without CBF measurements in patients with SAH and reported on high lactate levels25 and L/P ratio3,19,20,23,24 indicating ischemia, and this was associated with poor outcome. Sarrafzadeh et al. detected significantly higher lactate concentrations on Days 1–8 after SAH in patients who later developed DCI, and a higher L/P ratio on Days 3–8 than in asymptomatic patients.23 Interestingly, the majority of patients who developed DCI (83%) showed a characteristic metabolic

TABLE 5. Clinical characteristics and CBF and MD values for each patient who developed DCI Case Fisher Hunt & Hess CT No. Grade Grade GOSE* Scan† 1 2 3 4 5 6 7 8 9

4 4 4 3 3 4 3 4 3

4 3 3 2 2 4 2 2 3

1 5 4 7 4 5 1 4 3

2 0 2 0 1 0 2 0 2

CBF Values (ml/100 g/min)

Cerebral MD Values

Global

Glucose (mmol/L)

29.1 20.0 18.0 11.0 22.2 19.0 26.8 22.6 23.5

32.9 20.0 21.3 13.7 28.0 21.0 34.5 20.8 21.3

1.7 2.7 1.7 0.6 4.2 3.2 2.3 1.5 3.0

Lactate (mmol/L)

Pyruvate (μmol/L)

2.6 6.1 10.5 5.7 4.4 3.7 4.3 3.8 5.5

108.7 193.7 146.9 119.8 184.5 223.1 120.0 175.0 172.4

Glycerol Glutamate (μmol/L) (μmol/L) L/P Ratio 73.0 28.0 577.0 140.0 65.0 48.0 72.0 152.3 528.0

2.0 27.1 66.8 28.2 26.2 21.5 60.1 2.3 1.9

23.7 31.6 71.2 47.2 23.7 16.7 35.6 21.9 31.7

* Score at 1-year follow-up. The GOSE scores range from 1 (death) to 8 (excellent recovery). † CT scan scoring: 0 = no infarcts; 1 = infarction in a single hemisphere; 2 = bilateral infarctions. 6

J Neurosurg June 2, 2017

Low CBF and high lactate: predictors of DCI in SAH

FIG. 5. Box plots of microdialysate glucose, pyruvate, glycerol, and glutamate in patients who later developed DCI and those who did not. No significant differences were observed.

derangement, with high lactate and glutamate levels and L/P ratio before DCI symptoms. This is in line with our findings, except for glutamate levels. In patients in whom the characteristic metabolic derangement was not seen, clinical symptoms were related to a vascular territory not monitored with the MD catheter.23 In the current study, patients who later developed DCI had significantly higher lactate despite the MD catheter being placed in the right frontal lobe. Earlier reports have suggested 2 different sources of lactate elevation: one is “hypoxic” and is bad, and the other is “hyperglycolytic” and may be good for the injured brain.2,14 Although this theory is in need of further validation, the simultaneous measurements of CBF and lactate and pyruvate in the current study showing low CBF and high lactate with low pyruvate indicate anaerobic metabolism and energy failure. Pyruvate is an important metabolite that indicates the level of aerobic metabolism of glucose, and low levels have been associated with unfavorable outcome in patients with SAH.2

In the group who did not develop DCI, 3 patients had high lactate levels, but the CBF values were also high in these patients, which could indicate a hyperglycolytic state. There were 3 patients with low CBF in this group TABLE 6. Frequency of GOSE scores at 1-year follow-up in patients who later did and did not develop DCI GOSE Score

No DCI

DCI

1 2 3 4 5 6 7 8

3 (14) 2 (10) 5 (24) 5 (24) 1 (5) 4 (19) 0 (0) 1 (5)

2 (22) 0 (0) 1 (11) 3 (33) 2 (22) 0 (0) 1 (11) 0 (0)

Values are expressed as the number of patients (%). J Neurosurg June 2, 2017

E. Rostami et al.

who did not develop DCI, and they showed low lactate. This could indicate that low CBF or high lactate per se is not a sufficient indicator of tissue at risk for ischemia and at risk of developing DCI. This is in accordance with previous findings on CBF and MD parameters in the injured brain.31 One of the limitations of this study is that the person who performed the statistical analysis was not blinded to which patients developed DCI and which did not. However, the CBF analysis and MD data extractions were performed before any information was obtained regarding DCI development.

Conclusions

Our study shows that bedside monitoring of CBF and brain metabolism in SAH is feasible and highly informative. Low CBF, high lactate, and an increased L/P ratio were seen early after hemorrhage (Days 0–3) in patients who later developed DCI before any clinical signs had appeared. The clinical significance of these findings must be substantiated in larger studies.

Acknowledgments

We express our warm gratitude to Johan Bäckander for his excellent work with the Xe-CT examinations.

References

1. Bergvall U, Steiner L, Forster DM: Early pattern of cerebral circulatory disturbances following subarachnoid haemorrhage. Neuroradiology 5:24–32, 1973 2. Cesarini KG, Enblad P, Ronne-Engström E, Marklund N, Salci K, Nilsson P, et al: Early cerebral hyperglycolysis after subarachnoid haemorrhage correlates with favourable outcome. Acta Neurochir (Wien) 144:1121–1131, 2002 3. Chen HI, Stiefel MF, Oddo M, Milby AH, Maloney-Wilensky E, Frangos S, et al: Detection of cerebral compromise with multimodality monitoring in patients with subarachnoid hemorrhage. Neurosurgery 69:53–63, 2011 4. De Micheli E, Pinna G, Piovan E, Prisco R, Hillered L, Persson L, et al: Monitoring subtle neurometabolic changes in subarachnoid hemorrhage patients using microdialysis: a study on 16 cases. Acta Neurochir Suppl 77:149–153, 2001 5. Diringer MN, Bleck TP, Hemphill JC III, Menon D, Shutter L, Vespa P, et al: Critical care management of patients following aneurysmal subarachnoid hemorrhage: recommendations from the Neurocritical Care Society’s Multidisciplinary Consensus Conference. Neurocrit Care 15:211–240, 2011 6. Enblad P, Valtysson J, Andersson J, Lilja A, Valind S, Antoni G, et al: Simultaneous intracerebral microdialysis and positron emission tomography in the detection of ischemia in patients with subarachnoid hemorrhage. J Cereb Blood Flow Metab 16:637–644, 1996 7. Frykholm P, Andersson JL, Långström B, Persson L, Enblad P: Haemodynamic and metabolic disturbances in the acute stage of subarachnoid haemorrhage demonstrated by PET. Acta Neurol Scand 109:25–32, 2004 8. Hillered L, Vespa PM, Hovda DA: Translational neurochemical research in acute human brain injury: the current status and potential future for cerebral microdialysis. J Neurotrauma 22:3–41, 2005 9. Hutchinson PJ, Jalloh I, Helmy A, Carpenter KL, Rostami E, Bellander BM, et al: Consensus statement from the 2014 International Microdialysis Forum. Intensive Care Med 41:1517–1528, 2015

10. Jennett B, Snoek J, Bond MR, Brooks N: Disability after severe head injury: observations on the use of the Glasgow Outcome Scale. J Neurol Neurosurg Psychiatry 44:285– 293, 1981 11. Kety SS: The theory and applications of the exchange of inert gas at the lungs and tissues. Pharmacol Rev 3:1–41, 1951 12. Meyer JS, Shinohara T, Imai A, Kobari M, Sakai F, Hata T, et al: Imaging local cerebral blood flow by Xenon-enhanced computed tomography—technical optimization procedures. Neuroradiology 30:283–292, 1988 13. Nyberg C, Karlsson T, Hillered L, Ronne Engström E: Metabolic pattern of the acute phase of subarachnoid hemorrhage in a novel porcine model: studies with cerebral microdialysis with high temporal resolution. PLoS One 9:e99904, 2014 14. Oddo M, Levine JM, Frangos S, Maloney-Wilensky E, Carrera E, Daniel RT, et al: Brain lactate metabolism in humans with subarachnoid hemorrhage. Stroke 43:1418–1421, 2012 15. Qureshi AI, Sung GY, Razumovsky AY, Lane K, Straw RN, Ulatowski JA: Early identification of patients at risk for symptomatic vasospasm after aneurysmal subarachnoid hemorrhage. Crit Care Med 28:984–990, 2000 16. Ronne-Engström E, Cesarini KG, Enblad P, Hesselager G, Marklund N, Nilsson P, et al: Intracerebral microdialysis in neurointensive care: the use of urea as an endogenous reference compound. J Neurosurg 94:397–402, 2001 17. Rostami E, Engquist H, Johnson U, Howells T, RonneEngström E, Nilsson P, et al: Monitoring of cerebral blood flow and metabolism bedside in patients with subarachnoid hemorrhage—a Xenon-CT and microdialysis study. Front Neurol 5:89, 2014 18. Ryttlefors M, Howells T, Nilsson P, Ronne-Engström E, Enblad P: Secondary insults in subarachnoid hemorrhage: occurrence and impact on outcome and clinical deterioration. Neurosurgery 61:704–715, 2007 19. Samuelsson C, Hillered L, Enblad P, Ronne-Engström E: Microdialysis patterns in subarachnoid hemorrhage patients with focus on ischemic events and brain interstitial glutamine levels. Acta Neurochir (Wien) 151:437–446, 2009 20. Sarrafzadeh A, Haux D, Küchler I, Lanksch WR, Unterberg AW: Poor-grade aneurysmal subarachnoid hemorrhage: relationship of cerebral metabolism to outcome. J Neurosurg 100:400–406, 2004 21. Sarrafzadeh A, Haux D, Plotkin M, Lüdemann L, Amthauer H, Unterberg A: Bedside microdialysis reflects dysfunction of cerebral energy metabolism in patients with aneurysmal subarachnoid hemorrhage as confirmed by 15O-H2O-PET and 18 F-FDG-PET. J Neuroradiol 32:348–351, 2005 22. Sarrafzadeh AS, Haux D, Lüdemann L, Amthauer H, Plotkin M, Küchler I, et al: Cerebral ischemia in aneurysmal subarachnoid hemorrhage: a correlative microdialysis-PET study. Stroke 35:638–643, 2004 23. Sarrafzadeh AS, Sakowitz OW, Kiening KL, Benndorf G, Lanksch WR, Unterberg AW: Bedside microdialysis: a tool to monitor cerebral metabolism in subarachnoid hemorrhage patients? Crit Care Med 30:1062–1070, 2002 24. Schmidt JM, Ko SB, Helbok R, Kurtz P, Stuart RM, Presciutti M, et al: Cerebral perfusion pressure thresholds for brain tissue hypoxia and metabolic crisis after poor-grade subarachnoid hemorrhage. Stroke 42:1351–1356, 2011 25. Schulz MK, Wang LP, Tange M, Bjerre P: Cerebral microdialysis monitoring: determination of normal and ischemic cerebral metabolisms in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 93:808–814, 2000 26. Skoglund K, Hillered L, Purins K, Tsitsopoulos PP, Flygt J, Engquist H, et al: The neurological wake-up test does not alter cerebral energy metabolism and oxygenation in patients with severe traumatic brain injury. Neurocrit Care 20:413– 426, 2014 27. Solenski NJ, Haley EC Jr, Kassell NF, Kongable G, German-

J Neurosurg June 2, 2017

Low CBF and high lactate: predictors of DCI in SAH

28.

29. 30. 31.

32. 33. 34. 35.

son T, Truskowski L, et al: Medical complications of aneurysmal subarachnoid hemorrhage: a report of the multicenter, cooperative aneurysm study. Participants of the Multicenter Cooperative Aneurysm Study. Crit Care Med 23:1007–1017, 1995 Sturnegk P, Mellergård P, Yonas H, Theodorsson A, Hillman J: Potential use of quantitative bedside CBF monitoring (XeCT) for decision making in neurosurgical intensive care. Br J Neurosurg 21:332–339, 2007 Ungerstedt U, Rostami E: Microdialysis in neurointensive care. Curr Pharm Des 10:2145–2152, 2004 van Gijn J, Kerr RS, Rinkel GJ: Subarachnoid haemorrhage. Lancet 369:306–318, 2007 Vespa P, Bergsneider M, Hattori N, Wu HM, Huang SC, Martin NA, et al: Metabolic crisis without brain ischemia is common after traumatic brain injury: a combined microdialysis and positron emission tomography study. J Cereb Blood Flow Metab 25:763–774, 2005 Weir B, Grace M, Hansen J, Rothberg C: Time course of vasospasm in man. J Neurosurg 48:173–178, 1978 Yonas H, Darby JM, Marks EC, Durham SR, Maxwell C: CBF measured by Xe-CT: approach to analysis and normal values. J Cereb Blood Flow Metab 11:716–725, 1991 Yonas H, Gur D, Wolfson SK Jr, Good WF, Good BC, Latchaw RE: Xenon-enhanced computerised tomographic cerebral blood flow mapping. Lancet 1:1357, 1984 Yonas H, Pindzola RP, Johnson DW: Xenon/computed tomography cerebral blood flow and its use in clinical management. Neurosurg Clin N Am 7:605–616, 1996

36. Yonas H, Sekhar L, Johnson DW, Gur D: Determination of irreversible ischemia by xenon-enhanced computed tomographic monitoring of cerebral blood flow in patients with symptomatic vasospasm. Neurosurgery 24:368–372, 1989

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Rostami, Enblad. Acquisition of data: Rostami, Engquist, Howells. Analysis and interpretation of data: Rostami, Johnson, Enblad. Drafting the article: Rostami. Critically revising the article: Rostami, Engquist, Howells, Ronne-Engström, Nilsson, Hillered, Lewén, Enblad. Reviewed submitted version of manuscript: Rostami, Engquist, Howells, Ronne-Engström, Nilsson, Hillered, Lewén, Enblad. Approved the final version of the manuscript on behalf of all authors: Rostami. Statistical analysis: Rostami. Administrative/technical/material support: Engquist, Howells, Ronne-Engström, Nilsson, Hillered, Lewén, Enblad. Study supervision: Rostami, Enblad.

Correspondence

Elham Rostami, Section of Neurosurgery, Department of Neuroscience, Uppsala University Hospital, SE-751 85 Uppsala, Sweden. email: elham.rostami@neuro.uu.se.

J Neurosurg June 2, 2017

Hov et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2017) 25:21 DOI 10.1186/s13049-017-0365-1

CASE REPORT

Open Access

Pre-hospital ct diagnosis of subarachnoid Pre-hospital CT diagnosis of subarachnoid hemorrhage

hemorrhage Maren Ranhoff Hov1*, Annette Ryen2,3, Katrine Finsnes1,3, Janne Storflor3, Thomas Lindner1, Jostein Gleditsch4, and Christian Georg Lund1,5 on behalf of the NASPP Group Abstract Background: Subarachnoid hemorrhage (SAH) is associated with higher mortality in the acute phase than other stroke types. There is a particular risk of early and devastating re-bleeding. Patients therefore need urgent assessment in a neurosurgical department, and the shorter the time from symptom onset to diagnosis the better. Case presentation: The Norwegian Acute Stroke Pre-hospital Project (NASPP) has developed a Mobile Stroke Unit (MSU) model, which is staffed with anesthesiologists also trained in pre-hospital clinical assessment of acute stroke patients and interpretation of computerized tomography (CT). The MSU was operated on-call from the local dispatch center in a rural area 45–160 km away from a neurosurgical department. Two patients presented with clinical symptoms and signs compatible with SAH. In both cases, the CT examination confirmed the diagnosis of SAH. Both were transported directly from patient location to the regional neurosurgical department, saving at least 2–2.5 h of pre-neurosurgical time. Conclusion: The Norwegian MSU model staffed with anesthesiologists can rapidly establish an exact diagnosis of SAH, which in a rural area significantly reduces time to neurosurgical care. Trial registration: Study data are retrospectively registered in ClinicalTrail.gov. NCT03036020 Unique Protocol ID: NASPP-2 Brief Title: The Norwegian Acute Stroke Prehospital Project Overall Status: Completed Primary Completion Date: January 2016 [Actual] Verification Date: January 2017 Keywords: Pre-hospital, MSU, Cerebral CT, Subarachnoid hemorrhage, Stroke diagnostics, Neurosurgery

Background Subarachnoid haemorrhage (SAH) is a medical emergency, with overall incidence of approximately nine per 100.000 person-years [1]. SAH is associated with higher mortality in the acute phase than other stroke types, but immediate recognition and access to a neurosurgical department may reduce both mortality and morbidity [2]. A non-contrast CT scan will establish the diagnosis of SAH in most cases [3]. The Norwegian Acute Stroke Pre-hospital Project (NASPP) constructed a Mobile Stroke Unit (MSU) to explore the possibilities of pre-hospital diagnostics of acute * Correspondence: maren.ranhoff.hov@norskluftambulanse.no 1 The Norwegian Air Ambulance Foundation, Holterveien 24, 1448 Drøbak, Norway Full list of author information is available at the end of the article

stroke. The MSU was staffed with an anaesthesiologist from the national Helicopter Emergency Medical Service (HEMS), a paramedic and a nurse paramedic. The MSU anaesthesiologists were educated in making a clinical diagnosis of stroke, a National Institute of Stroke Scale (NIHSS) score, and a cerebral CT examination and interpretation [4]. The cerebral CT scan interpretation was primarily focused on identifying contraindications to thrombolysis for acute ischemic stroke, but also on the differentiation between types of intracranial haemorrhage.

Methods NASPP was initiated in October 2014 using the MSU (Mercedes Sprinter) with a cerebral CT scanner (CereTom Neurologica, Samsung), a point of care biochemical

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Hov et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2017) 25:21

laboratory (pocH-100i Automated Hematology Analyzer and HEMOCHRON Jr. Signature±) and a telemedicine system (Meytech). The MSU’s catchment area was Østfold county, a rural and quite sparsely populated part of Norway with approximately 289 000 inhabitants (1720 inhabitants/ sq. km). The distances to the only local acute hospital are up 100 km, and the distances to the neurosurgical department are between 45 and 160 km. The Emergency Medical Communication Centre (EMCC) for dispatch used the Norwegian index of emergency medicine as triage guideline [5]. The dispatch criteria included “acute thunderclap headache”. We have included various time calculations in our study: estimated transit time from patient location to the local hospital, in hospital admission time and time for CT diagnostics, as well as time used for transfer from the local hospital to the neurosurgical department.

Results During the NASPP study inclusion, the MSU performed 68 pre-hospital CT scans. In two cases (2.8%) the patient presented with hyper-acute headache as the main symptom, and subsequent CT scans showed intracranial bleeding compatible with SAH in both instances (Fig. 1). The anaesthesiologist immediately interpreted the CT scan in the MSU and communicated the findings to the on-call radiologist at the local hospital. A joint decision was made to transport the patient directly to the neurosurgical department, and the on-call neurosurgeon was informed. Estimated pre-neurosurgical time saved by bypassing the local hospital was 120 and 150 min respectively. Full patient characteristics, time parameters and administered medications are shown in Table 1.

Page 2 of 4

Both patients were treated at the Department of Neurosurgery at Oslo University Hospital. Patient 1 had a CTangiography at arrival showing no signs of aneurysms. Due to thelarge volume of blood in the subarachnoid space a CT angiography was repeated, confirming the absence of vascular pathology. The patient subsequently developed signs of elevated intracranial pressure and was in need of external ventricular drainage. After 16 days the patient was discharged to the local hospital for further rehabilitation. Patient 2 also had a negative CT angiography and was conservatively treated at the neurosurgical ward for 4 days before discharge to home. Both patients are now living without any neurological dysfunction.

Discussion We have shown that the Norwegian MSU model is able to establish a definite pre-hospital diagnosis of SAH, which can substantially reduce pre-neurosurgical time. In spite of limited scientific evidence [2] there is reason to believe that SAH and its management is highly time-sensitive. SAH is a subtype of stroke that generally affects a younger population. In SAH there is a significant risk of early and devastating re-bleeding, and hydrocephalus with subsequent increased intracranial pressure may develop acutely. Rapid transportation to a neurosurgical department is crucial to handle these complications [6]. In Norway, hardly any patient with acute headache is triaged to a neurosurgical department without an initial CTscan at the local hospital. If a similar study had been performed in the northern part of Norway, where distances to the local hospital can be more than 200 km, the time saved by establishing a pre-hospital SAH diagnosis could be several hours more than in our current study.

Fig. 1 Pre-hospital cerebral CT scans from the MSU. PATIENT 1: Pre-hospital CT scan. There is subarachnoid hemorrhage located around the brainstem, in the supracellar cistern and in the right lateral fissure. PATIENT 2: Pre-hospital CT scan. There is subarachnoid hemorrhage mainly located ventral to the brainstem

100

Hov et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2017) 25:21

Page 3 of 4

Table 1 Clinical patient characteristics, time parameters (min) and pre-hospital medication Patient 1

Patient 2

Age (years)

Sex

GCS (Glasgow Coma Scale)

Time from symptom onset to alarm MSU

171

Time from alarm MSU to MSU at the scene

Time from MSU at the scene to CT completed

Time from CT diagnostics to Neurosurgical Dept.

Time from symptom onset to neurosurgical department

127

312

Pre-hospital medication

Antiemetic

Tranexamic acid, analgesics, antiemetic

Age (years)

Sex

GCS (Glasgow Coma Scale)

Time from symptom onset to alarm MSU

171

Time from alarm MSU to MSU at the scene

Time from MSU at the scene to CT completed

Time from CT diagnostics to Neurosurgical Dept.

Time from symptom onset to neurosurgical department

127

312

Pre-hospital medication

Antiemetic

Tranexamic acid, analgesics, antiemetic

The main reason behind the development of the MSU concept has been to initiate thrombolytic treatment of acute ischemic stroke as early as possible after symptom start, preferably already in the “golden hour” [7]. Different MSU models for pre-hospital stroke diagnosis and treatment have been studied in recent years [8–11]. In the STEMO MSU study in Berlin, the delivery rate of patients with intracranial hemorrhage to hospitals without a neurosurgical department was 43.0% in conventional care and 11.3% in an MSU (p < 0.01) [12], allowing the authors to conclude that pre-hospital diagnostics might reduce time to neurosurgical care. The different MSU-models in published literature are staffed either by a neurologist, a neurologist and a neuroradiologist, or solely by paramedics [8–10, 13]. Such MSU models may be very suitable in large urban areas, but the availability of specific trained individuals and inhospital specialists makes it difficult to apply such MSU models to countries and regions that are less densely populated. In Norway, we have chosen to develop a MSU-concept similar to the national HEMS service in order to deliver stroke services to all parts of the country, rural and geographically very remote areas included. Moreover, when staffed with an anaesthesiologist a MSU may also provide resuscitation and medical emergency services such as endotracheal intubation and other invasive procedures to any unstable or critically ill patient. Especially in rural areas, the concept of pre-hospital diagnostics and triage of acute cerebral stroke or head trauma may be as important for the MSU as providing

early thrombolytic therapy. A positive cerebral radiological diagnosis, and nearly equally important a “negative finding”, will enable urgent pre-hospital medical decisions regarding to where and in which manner a patient with acute cerebral stroke or trauma should be transported. Such pre-hospital cerebral diagnostics can be carried out ambulatory by a MSU or possibly by a helicopter equipped with a cerebral CT-scanner in the future. Stationary CT scanners could by located for example at a district medical centre in rural areas. As demonstrated by our SAH patients, pre-hospital “acute brain” diagnostics will save a lot of time for the patients in a rural area, and probably increase quality adjusted life years (QUALYs).

Conclusion The Norwegian MSU model with anaesthesiologists can rapidly establish an exact diagnosis of SAH, which in a rural area can significantly reduce time to definitive neurosurgical care. Acknowledgements Ann Kristin Wiik, the Norwegian Air Ambulance Foundation, for administrative leadership in the project. Volker Solyga, head of Dept. of Neurology at Østfold Hospital for arranging the in hospital data collection in the study. Funding The Norwegian Air Ambulance Foundation funded this study. Availability of data and materials Please contact author for data requests.

101

Hov et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2017) 25:21

Page 4 of 4

Authors’ contributions ARO, KF and JS was directly involved in data collection and patient care. JG and TL facilitated data collection, and was involved in data analysis. CL was the medical director of the study and helped in drafting the manuscript. All authors read and approved the final manuscript. Competing interest The author declares that he/she has no competing interests. Consent for publication All patients in the NASPP study and presented in this case report have given their written consent to publication. Ethics approval and consent to participate The Norwegian Regional Ethics committee approved the NASPP study with registration number 2012/2298. Author details 1 The Norwegian Air Ambulance Foundation, Holterveien 24, 1448 Drøbak, Norway. 2Department of Anaesthesiology, Østfold Hospital, Sarpsborg, Norway. 3Department of Neurology, Østfold Hospital, Sarpsborg, Norway. 4 Department of Radiology, Østfold Hospital, Sarpsborg, Norway. 5Department of Neurology, Oslo University Hospital, Rikshospitalet, Norway. Received: 2 December 2016 Accepted: 15 February 2017 References 1. de Rooij NK, Linn FH, van der Plas JA, Algra A, Rinkel GJ. Incidence of subarachnoid haemorrhage: a systematic review with emphasis on region, age, gender and time trends. J Neurol Neurosurg Psychiatry. 2007;78:1365–72. 2. Casaubon LK, Boulanger JM, Blacquiere D, Boucher S, Brown K, Goddard T, et al. Canadian stroke best practice recommendations: hyperacute stroke care guidelines, update 2015. Int J Stroke. 2015;10:924–40. 3. Edlow JA, Fisher J. Diagnosis of subarachnoid hemorrhage: time to change the guidelines? Stroke. 2012;43:2031–2. 4. Hov MR, Nome T, Zakariassen E, Russell D, Roislien J, Lossius HM, et al. Assessment of acute stroke cerebral ct examinations by anaesthesiologists. Acta Anaesthesiol Scand. 2015;59:1179–86. 5. Medicine TLFfA. Norwegian medical association: Norwegian index for medical emergency assistance. 2005. http://traumeplan.no/wp-content/uploads/2015/ 03/Norsk-index-for-medisinsk-nødhjelp.pdf. 6. Starke RM, Connolly Jr ES. Participants in the international multi-disciplinary consensus conference on the critical care management of subarachnoid H. Rebleeding after aneurysmal subarachnoid hemorrhage. Neurocrit Care. 2011;15:241–6. 7. Fassbender K, Balucani C, Walter S, Levine SR, Haass A, Grotta J. Streamlining of prehospital stroke management: the golden hour. Lancet Neurol. 2013; 12:585–96. 8. Walter S, Kostopoulos P, Haass A, Keller I, Lesmeister M, Schlechtriemen T, et al. Diagnosis and treatment of patients with stroke in a mobile stroke unit versus in hospital: a randomised controlled trial. Lancet Neurol. 2012;11:397–404. 9. Cerejo R, John S, Buletko AB, Taqui A, Itrat A, Organek N, et al. A mobile stroke treatment unit for field triage of patients for intraarterial revascularization therapy. J Neuroimaging. 2015;25:940–5. 10. Bowry R, Parker S, Rajan SS, Yamal JM, Wu TC, Richardson L, et al. Benefits of stroke treatment using a mobile stroke unit compared with standard management: the best-msu study run-in phase. Stroke. 2015;46:3370–4. 11. Weber JE, Ebinger M, Rozanski M, Waldschmidt C, Wendt M, Winter B, et al. Prehospital thrombolysis in acute stroke: results of the phantom-s pilot study. Neurology. 2013;80:163–8. 12. Wendt M, Ebinger M, Kunz A, Rozanski M, Waldschmidt C, Weber JE, et al. Improved prehospital triage of patients with stroke in a specialized stroke ambulance: Results of the pre-hospital acute neurological therapy and optimization of medical care in stroke study. Stroke. 2015;46:740–5. 13. Ebinger M, Winter B, Wendt M, Weber JE, Waldschmidt C, Rozanski M, et al. Effect of the use of ambulance-based thrombolysis on time to thrombolysis in acute ischemic stroke: a randomized clinical trial. JAMA. 2014;311:1622–31.

Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit

102

Neurocrit Care DOI 10.1007/s12028-016-0303-5

ORIGINAL ARTICLE

Derivation of a Predictive Score for Hemorrhagic Progression of Cerebral Contusions in Moderate and Severe Traumatic Brain Injury Randall Z. Allison1 • Kazuma Nakagawa2,3 • Michael Hayashi4,5 • Daniel J. Donovan2,4 Matthew A. Koenig2,3

•

Derivation of predictive score for hemorrhagic progres of ecrebral contusions in moderate and severe

Springer Science+Business Media New York 2016

Abstract Backgrounds After traumatic brain injury (TBI), hemorrhagic progression of contusions (HPCs) occurs frequently. However, there is no established predictive score to identify high-risk patients for HPC. Methods Consecutive patients who were hospitalized (2008–2013) with non-penetrating moderate or severe TBI were studied. The primary outcome was HPC, defined by both a relative increase in contusion volume by C30 % and an absolute increase by C10 mL on serial imaging. Logistic regression models were created to identify

& Matthew A. Koenig mkoenig@queens.org Randall Z. Allison zainallison@gmail.com Kazuma Nakagawa knakagawa@queens.org Michael Hayashi mhayashi@queens.org Daniel J. Donovan donovand001@hawaii.rr.com 1

Division of Neurosurgery, Department of Surgery, University of Texas Medical Branch, Galveston, TX, USA

Department of Surgery, Neuroscience Institute, The Queen’s Medical Center, 1301 Punchbowl Street- QET5, Honolulu, HI 96813, USA

Department of Medicine, John A. Burns School of Medicine, University of Hawaii, Honolulu, HI, USA

Department of Surgery, John A. Burns School of Medicine, University of Hawaii, Honolulu, HI, USA

Department of Surgery, The Queen’s Medical Center, Honolulu, HI, USA

brainThe injury independent risktraumatic factors for HPC. HPC Score was then derived based on the final model. Results Among a total of 286 eligible patients, 61 (21 %) patients developed HPC. On univariate analyses, HPC was associated with older age, higher initial blood pressure, antiplatelet medications, anticoagulants, subarachnoid hemorrhage (SAH) subdural hematoma (SDH), skull fracture, frontal contusion, larger contusion volume, and shorter interval from injury to initial CT. In the final model, SAH (OR 6.33, 95 % CI, 1.80–22.23), SDH (OR 3.46, 95 % CI, 1.39–8.63), and skull fracture (OR 2.67, 95 % CI, 1.28–5.58) were associated with HPC. Based on these factors, the HPC Score was derived (SAH = 2 points, SDH = 1 point, and skull fracture = 1 point). This score had an area under the receiver operating curve of 0.77. Patients with a score of 0–2 had a 4.0 % incidence of HPC, while patients with a score of 3–4 had a 34.6 % incidence of HPC. Conclusions A simple HPC Score was developed for early risk stratification of HPC in patients with moderate or severe TBI. Keywords Cerebral contusion Traumatic brain injury Triage Critical care Resource allocation

Introduction In blunt traumatic brain injury (TBI), hemorrhagic progression of contusions (HPCs) refers to enlargement of the hemorrhagic component of cerebral contusions after initial imaging [1]. Commonly called contusion ‘‘blossoming,’’ HPC can result in neurological worsening from mass effect, cerebral edema, and parenchymal tissue necrosis [2, 3]. Prior case series have reported an incidence of HPC

123 103

Neurocrit Care

in TBI patients that ranges from 18 to 51 % [4–10]. Early identification of patients at high risk for HPC could lead to interventions that reduce the morbidity and mortality of TBI. Conversely, for patients who are at low risk for HPC, neurological monitoring in the intensive care unit (ICU) with frequent serial neurological examinations and repeat imaging studies substantially adds to the acuity and cost of care for TBI patients [11]. Prior reports have identified that radiographic factors such as the presence of a subdural hematoma (SDH) [4, 6, 12], subarachnoid hemorrhage (SAH) [4, 5, 7], and larger contusion volume [5, 6, 10] are associated with higher risk for HPC, as well as clinical factors such as older age [4, 13], hyperglycemia [14], antiplatelet medication use [15], low platelet count [16], and timing of the initial CT scan [4, 7]. One prior group attempted to develop a predictive score using clinical parameters [14]; but the lowest risk patient group still had a high incidence of HPC (10.3 %) in this study, and the algorithm was complex. A simple predictive score for HPC could provide a useful tool to improve resource allocation, especially in settings where ICU beds or specialty services may be limited. In this retrospective cohort study, we identified common clinical and radiographic factors associated with HPC in patients admitted for moderate or severe TBI. These factors were then used to derive a predictive scoring model that could be easily calculated to stratify risk for HPC. We hypothesized that an accurate predictive score for HPC could be derived from early markers of injury severity.

Methods The Queen’s Medical Center (QMC) Institutional Review Board approved this research project with waiver of informed consent. A prospectively collected trauma database from QMC, the only American College of Surgeonsverified trauma center in Hawaii, was queried to identify consecutive patients admitted for non-penetrating moderate or severe TBI from January 1, 2008, to December 31, 2013. Severe TBI was defined as blunt head trauma with initial Glasgow Coma Scale (GCS) of 3–8, and moderate TBI was defined as initial GCS of 9–12 [17]. The standard practice at QMC is to admit patients with severe TBI to a dedicated Neurosciences ICU staffed by specialized neurocritical care nurses, acute care nurse practitioners, and board-certified neurointensivists. Moderate TBI patients are triaged to either the Neurosciences ICU or a neurological specialty acute medicalsurgical ward by the admitting trauma surgeon and neurosurgeon based on the anticipated clinical course of the patient. Repeat brain CT within 6–12 h after the initial scan

123

104

is a standard institutional practice unless there are reasons not to perform repeat imaging, e.g., rapid neurological improvement or injuries that are perceived to be nonsurvivable. A portable bedside CT scanner (Ceretom) is available in the ICU and is often used for repeat imaging in the ICU patients. Inclusion and Exclusion Criteria The trauma database was queried for all TBI patients admitted during the study period. Consecutive patients with moderate or severe blunt TBI and repeat brain CT within 24 h were included in the analysis. The presence of a cerebral contusion on the initial brain CT was not required for inclusion because development of contusions may occur after the initial CT, and we believed that an accurate predictive score would need to account for delayed contusions. Patients with penetrating brain injury and those who did not undergo repeat brain imaging were excluded. TBI patients who met the initial GCS criteria due to intoxication but had a rapid subsequent neurological improvement to GCS > 12 were excluded from the study. Outcome Variable The primary outcome variable for the study was HPC. HPC was defined as both a relative increase in volume of the lesion by C30 % compared to the initial CT and an absolute increase in the total contusion volume by C10 mL on repeat CT. The C30 % threshold was based on prior reports [6] and was chosen to set a relatively high threshold for HPC due to potential differences in technique between the initial and follow-up CT scan. Enlargement of lesion volume by C10 mL was based on prior reports [4] and was chosen to identify HPC that had the potential to be clinically significant. The first two brain CT scans were retrospectively evaluated by a board-certified neurointensivist for identification of HPC. Parenchymal hematoma volume was estimated using the ABC/2 method, which has been utilized in prior studies of spontaneous and traumatic hematoma volume [2, 3, 6, 18, 19]. Heterogeneous lesions, defined as mixed high and low density, were measured using the entire length of the hyperdense portion of the lesion. In the case of multiple contusions, the volumes of all contusions were summed together in an effort to account for mass effect from multiple lesions. Punctate contusions were estimated at 1 mL each. The secondary outcome measure was HPC associated with neurological decline, which was defined by the following criteria: unplanned craniotomy or ICP monitor placement, unplanned endotracheal intubation, unplanned ICU transfer during the first 24 h of hospitalization, or sustained decline in GCS by C1 point for C24 h that was

Neurocrit Care

not attributable to sedation. The GCS criterion was adopted for high sensitivity identification of patients with neurological decline while avoiding inclusion of patients with fluctuations of GCS due to sedative medications and circadian patterns. Additional secondary outcome measures were hospital length of stay, ICU length of stay, mortality, and discharge modified Rankin Scale (mRS) score. Factors Associated with HPC Clinical data were obtained from the prospectively collected trauma database supplemented with retrospective review of the electronic medical record (Epic). Initial GCS was defined as the highest recorded value in the emergency department (ED) obtained by the attending trauma surgeon, neurosurgeon, and/or neurointensivist prior to sedation. Decline of GCS during the first 24 h was defined as a reduction in GCS by C1 point that was sustained over 24 h and not attributable to sedation. Initial motor examination and pupil size and reactivity were determined by the attending trauma surgeon, neurosurgeon, or neurointensivist on initial examination in the ED. Initial brain CT was defined as the first imaging study available in the QMC computerized radiology system. For patients presenting directly to QMC, the initial CT was obtained in the QMC ED. For patients transferred to QMC from another facility, the initial CT was reviewed from the referring hospital if it was electronically transmitted or scanned into the QMC computerized radiology system. All CT scans were evaluated for Marshall CT classification [20]; Rotterdam CT score [21]; presence, displacement, and location of skull fractures; presence, location, and maximal thickness of SAH, epidural hematoma, and SDH; presence, location, and volume of cerebral contusions; extent of midline shift; and presence of intraventricular hemorrhage and hydrocephalus. Examples of HPC in two patients are shown in Fig. 1.

Statistical Analysis Data were analyzed using commercially available statistical software (SPSS 22.0, Chicago, IL). Patient characteristics were summarized using descriptive statistics appropriate to variable type. Those with HPC were compared to those without HPC using the Chi-square test for categorical data, two-tailed t test for normally distributed, continuous variables, and the Mann–Whitney U test for nonparametric data. Data were reported as mean ± SD except as noted. Variables that could have a biologically plausible association with HPC, could act as potential confounders, or had been reported to be associated with HPC in prior studies were included in the analyses.

A multivariable model using logistic regression was then constructed to identify independent risk factors for HPC. Variables with p B 0.1 in the univariate testing were entered in the model. Odds ratio (OR) and 95 % confidence interval (CI) were calculated from the beta coefficients and their standard errors. Levels of p < 0.05 were considered statistically significant. After performing the initial regression model, a predictive score was derived using a weighted system by dividing the beta coefficients by a common denominator to obtain an integer score proportional to the magnitude of the beta coefficient. The performance of the score was assessed by calculating the area under the receiver operating characteristic (ROC) curve.

Results Based on the GCS criteria, we identified 419 patients with moderate or severe TBI due to blunt head trauma. Of these patients, we excluded 133 patients who did not have repeat imaging within 24 h of the injury. Among the excluded patients, 19 were children (median age 9 years, range 0–16 years) for whom repeat imaging was not recommended in order to minimize radiation exposure. Among the 114 adult patients who were excluded, 53 died from traumatic injuries prior to performance of a repeat head CT or had injuries that were believed to be nonsurvivable by the treating physician so repeat imaging was not obtained; 36 had no apparent intracranial lesion or minor lesions on the initial head CT, and the treating physician believed that a repeat head CT was unnecessary; 12 had advance directives or early family request for comfort measures only; 7 had improvement in neurological examination, and the treating physician believed that a repeat head CT was unnecessary; and 6 had initial imaging at an outside hospital that was not available for retrospective review. A total of 286 patients met the inclusion criteria and were included in the final analyses. Of these patients, 146 had moderate TBI and 140 had severe TBI. Characteristics of the study population are shown in Table 1. The median initial composite GCS was 9 (interquartile range (IQR) 6–12). The mean initial contusion volume was 6.2 ± 11.6 mL among patients who had contusions on the initial brain CT. The mean interval from injury to initial brain CT scan was 2.4 ± 1.6 h, and the mean interval between first and second brain CT scan was 9.4 ± 4.8 h. Twenty-eight patients (9.8 %) underwent a craniotomy or decompressive craniectomy prior to the second brain CT, but no patients had evacuation of cerebral contusions. Of the 286 patients, 193(67 %) were initially admitted to the Neurosciences ICU, 36 (12.6 %) were admitted to other ICUs, and 57 (19.9 %) were admitted to the medical-surgical ward. For the total study population, the mean ICU

123

105

Neurocrit Care Fig. 1 Examples of hemorrhagic progression of contusions in patients a, b on serial brain CT

length of stay was 7.6 ± 9.5 days and mean hospital length of stay was 10.9 ± 14.4 days. In-hospital mortality occurred in 59 patients (21 %), and the discharge mRS was B2 in 102 patients (36 %). HPC occurred in 61 patients (21 %) and was associated with neurological decline in 32 (52 %) of these patients. Of the 32 patients with neurological decline due to HPC, 1 required unplanned ICU transfer from the medical-surgical ward, 3 required unplanned endotracheal intubation, and 4 required unplanned craniotomy or decompressive craniectomy. In univariate analyses (Table 2), HPC was associated with older age, higher initial systolic blood pressure (SBP), use of antiplatelet medications, use of anticoagulants, larger contusion volume, shorter interval from injury to initial brain CT, receipt of mannitol, no packed red blood cell (pRBC) transfusion, and presence of SAH, SDH, skull fracture, and frontal lobe contusion on the initial brain CT. A multivariable regression model was then constructed (Table 3). In the full model, SAH (OR 6.33, 95 % CI, 1.80–22.23), SDH (OR 3.46, 95 % CI, 1.39–8.63), skull fracture (OR 2.67, 95 % CI, 1.28–5.58), and pRBC transfusion (OR 0.31, 95 % CI, 0.10–0.93) were independently associated with HPC. Note that pRBC transfusion was

123

106

negatively associated with HPC, while the other factors were positively associated with HPC. Using the regression model, a predictive score was derived based on presence or absence of SAH, SDH, and skull fracture. We did not include pRBC transfusion in the final predictive score. The predictive score was constructed for HPC based on the odds ratio of factors in the full multivariable regression model (Table 3). This HPC Score has a possible range of 0–4 calculated by adding 2 points for presence of SAH, 1 point for presence of SDH, and 1 point for presence of a skull fracture. The area under the ROC curve of the HPC Score was 0.77. None of the patients with an HPC Score of 0 (n = 38) had HPC. Patients with an HPC Score of 0–2 had a 4.0 % (5/124) incidence of HPC, while patients with an HPC Score of 3–4 had a 34.6 % (56/162) incidence of HPC (Table 4).

Discussion In a cohort of 286 patients with moderate or severe blunt TBI, we identified factors associated with HPC and derived a simple scoring model called the HPC Score to stratify

Neurocrit Care Table 1 Characteristics of the study population (n = 286)

Table 2 Factors associated with hemorrhagic progression of contusion (HPC)

Population

286

Age (mean ± SD)

47.6 ± 26.7

Sex (% male)

202 (71 %)

Population Age (years)

56.7 ± 24.7

45.2 ± 26.7

0.002

Native Hawaiian or Pacific Islander

85 (30 %)

Initial GCS

9±4

0.756

Asian (non-Filipino)

71 (25 %)

Initial SBP (mmHg)

152 ± 35

138 ± 28

0.005

White

72 (25 %)

INR

1.3 ± 0.6

1.2 ± 0.3

0.835

Filipino

40 (14 %)

Platelet count

223.6 ± 65.5

244.3 ± 91.8

Glucose (mg/dL) Time to initial CT (h)

163 ± 52 2.0 ± 0.8

166 ± 78 2.5 ± 1.5

Race

Mechanism of injury (% most common)

HPC

No HPC

225

p value

0.104 0.796 <0.001

Moped/motorcycle crash

36 (12.6 %)

Assault

43 (15 %)

Contusion volume (mL)

10.1 ± 11.2

5.1 ± 11.4

0.036

Fall from height

50 (17.5 %)

SDH width (cm)

0.8 ± 0.5

0.6 ± 0.5

0.057

Fall from standing

88 (30.7 %)

Midline shift (cm)

0.3 ± 0.4

0.2 ± 0.4

0.087

Frontal lobe location

38 (62 %)

77 (34 %)

<0.001

History Hypertension Diabetes mellitus

76 (26.6 %) 36 (12.6 %)

Antiplatelet meds

13 (21 %)

26 (12 %)

0.049

Anticoagulant meds

6 (10 %)

7 (3 %)

0.044

Antiplatelet medications

39 (13.6 %)

Methamphetamine use

5 (8.2 %)

18 (8.0 %)

0.96

32 (11.2 %)

Alcohol use

15 (25 %)

75 (33 %)

0.19

Clopidogrel

5 (1.7 %)

SAH

58 (95 %)

134 (56 %)

<0.001

Aspirin + clopidogrel

2 (0.6 %)

SDH

54 (88 %)

125 (56 %)

<0.001

Anticoagulant medications

13 (4.5 %)

Skull fracture

40 (66 %)

94 (42 %)

0.001

Warfarin

11 (3.8 %)

Mannitol use

23 (38 %)

35 (16 %)

<0.001

pRBC transfusion

5 (8 %)

43 (19 %)

0.043

Aspirin

Other anticoagulant

2 (0.7 %)

Methamphetamine use

23 (8 %)

Alcohol use

90 (31 %)

Physical examination Initial GCS, median (IQR)

9 (6–12)

Initial pupils non-reactive (%)

44 (15.4 %)

Unilateral non-reactive pupil

22 (7.7 %)

Bilateral non-reactive pupils

22 (7.7 %)

Injury Severity Score (mean ± SD) Imaging

24 ± 10

Initial Marshall Scale (median)

Initial Rotterdam CT Score (median)

Initial contusion volume (mean ± SD)

6.2 ± 11.6

SD standard deviations, GCS Glasgow Coma Scale, IQR interquartile range, CT computed tomography

risk for HPC. This HPC Score has a possible range of 0–4 calculated by adding 2 points for presence of SAH, 1 point for presence of SDH, and 1 point for presence of a skull fracture. Scores of 0–4 had an observed HPC incidence of 0, 3.7, 7.7, 28.9, and 39.5 %, respectively. As in previously published reports, we identified presence of SAH [4, 5, 7], skull fracture [7], and SDH [3, 6, 12] as major risk factors for HPC. SAH, skull fracture, and SDH may be markers of the overall severity of head trauma, or they may represent injury to the cortical veins, bridging veins, and venous sinuses that could contribute to

Data shown as mean ± SD or count (percentage) GCS Glasgow Coma Scale, SBP systolic blood pressure, INR international normalized ratio, CT computed tomography, SDH subdural hematoma, SAH subarachnoid hemorrhage, pRBC packed red blood cells

venous congestion of the brain and increased risk of hematoma expansion [1]. The reason for the independence of these three risk factors for HPC is not known, but they may represent a cumulative injury to the brain such that patients with skull fracture and SDH in addition to SAH experienced greater brain trauma than patients with SAH alone. Older age, anticoagulant use, antiplatelet medication use, and frontal contusion location have also been associated with HPC in other studies [4, 12, 13, 15, 22–24] and were identified as risk factors in our univariate analyses; however, they were not independent risk factors for HPC in the multivariable model and were not included in the HPC Score. The lack of an independent association between HPC and anticoagulant or antiplatelet medications may be due to several factors, including incomplete or inaccurate ascertainment of medication use and the small number of patients on these medications. This finding does not necessarily imply that TBI patients on anticoagulant or antiplatelet medications are at low risk for HPC but rather

123

107

Neurocrit Care Table 3 Multivariable models for hemorrhagic progression of contusions (HPCs)

SAH

Model 1 OR (95 % CI)

Model 2 OR (95 % CI)

Model 3 OR (95 % CI)

Full model OR (95 % CI)

8.43 (2.47, 28.87)*

8.00 (2.34, 27.38)*

7.82 (2.27, 26.93)*

6.33 (1.80, 22.23)*

SDH

4.52 (1.92, 10.65)*

3.93 (1.65, 9.40)*

4.06 (1.69, 9.77)*

3.46 (1.39, 8.63)*

Convexity skull fracture

1.77 (0.93, 3.37)

2.08 (1.05, 4.09)*

2.24 (1.12, 4.48)*

2.67 (1.28, 5.58)*

Antiplatelet use

1.75 (0.74, 4.15)

1.60 (0.66, 3.87)

1.44 (0.55, 3.77)

Anticoagulant use

2.42 (0.67, 8.70)

3.42 (0.90, 13.07)

2.85 (0.75, 10.91)

pRBC transfusion

0.30 (0.10, 0.90)*

Age Time to initial CT

0.31 (0.10, 0.93)* 1.01 (1.00, 1.03) 0.91 (0.82, 1.01)

CI confidence interval, SAH subarachnoid hemorrhage, SDH subdural hematoma, pRBC packed red blood cells, CT computed tomography * p < 0.05

Table 4 Incidence of hemorrhagic progression of contusions (HPCs) based on HPC Score of 0â&#x20AC;&#x201C;4 HPC Score

HPC

No Yes

Count (%)

38 100

45 96.3

36 92.3

54 71.1

52 60.5

Count

(%)

3.7

7.7

28.9

39.5

Scoring criteria were: presence of SAH = 2 points; presence of SDH = 1 point; presence of skull fracture = 1 point

that these patients overwhelmingly had SAH or SDH on the initial brain CT. These radiographic findings, rather than clinical factors like age or medication use, were the strongest predictors of HPC and were, therefore, used to construct the HPC Score. An advantage of the HPC Score derived by this study is that it entirely relies on a small number of radiographic factors that can easily be obtained from the initial brain CT without detailed knowledge of the patientâ&#x20AC;&#x2122;s medical history or specific neurological examination findings. If it is externally validated, a predictive score with these features could allow early risk stratification for HPC among TBI patients by emergency physicians, trauma surgeons, and radiologists with limited neurological expertise and in trauma patients with unknown medical history. The HPC Score requires external validation in other cohorts before adoption in clinical practice can be recommended, however. We observed an apparent protective effect of pRBC transfusion on the incidence of HPC on the multivariable analysis. However, we did not include pRBC transfusion in the HPC Score for several reasons. First, the reason for this association is unclear. As blunt trauma can be

123

108

heterogeneous, TBI patients commonly experience concomitant trauma to other body systems with potential for blood loss. These injuries may create competing priorities in the initial management of the trauma patient, including the need for immediate hemorrhage control which may delay the time to initial head CT in TBI. The presence of systemic injuries may also distribute traumatic forces away from the head, decreasing the severity of brain trauma. RBC transfusion may also reflect unmeasured biases such as aggressiveness of the initial resuscitative effort. Second, we did not want to imply that TBI patients should undergo packed red blood cell (pRBC) transfusion in order to lower the chance of HPC. Third, the addition of pRBC transfusion did not significantly increase the accuracy of the model. Limitations of the study include the retrospective design, which could under-identify patients with neurological decline due to reliance on clinical documentation. The focus on radiographic enlargement may also overemphasize changes in lesion size that are not clinically relevant and under-emphasize important clinical deterioration that is unrelated to changes in the hematoma volume, such as per hematoma edema, mass effect, and delayed hydrocephalus. We also excluded 133 (32 %) of the screened patients because they did not undergo repeat head CT within 24 h, which could result in a selection bias excluding patients perceived as low risk for HPC or those with nonsurvivable injuries. Mild TBI patients were not included in the study because repeat brain CT scans are not routinely performed in this group at our institution. This is an important limitation, and the results may not be applicable to patients with mild TBI. The next steps include external validation of the HPC Score to confirm whether this predictive tool is able to accurately identify patients at high and low risk for HPC. If so, the HPC Score could become an important clinical tool

Neurocrit Care

for early HPC risk stratification in patients with TBI. In this paradigm, patients with high risk for HPC could be identified and triaged to a dedicated Neurosciences ICU with more intensive staffing ratios and more frequent neurological examinations, earlier neurosurgical consultation or ICP monitor placement, or clinical trials of antifibrinolytic agents, clotting factor replacement, or intensive blood pressure control. Conversely, early identification of patients at low risk for HPC could allow more efficient allocation of critical care resources. External validation of the HPC Score should also include patients with mild TBI based on an initial GCS of 13–15. Patients with initially mild TBI can also develop significant HPC, particularly those with bifrontal cerebral contusions [25]. The identification of mild TBI patients at high risk for HPC is particularly important because this group is commonly admitted to lower acuity clinical environments and more rapidly discharged from the hospital. Funding Dr. Nakagawa was partially supported by the National Institute on Minority Health and Health Disparities of the National Institutes of Health (P20MD000173).

Compliance with Ethical Standards Conflict of interest The authors report no conflicts of interest related to this manuscript.

References 1. Kurland D, Hong C, Aarabi B, Gerzanich V, Simard JM. Hemorrhagic progression of a contusion after traumatic brain injury: a review. J Neurotrauma. 2012;29:19–31. 2. Narayan RK, Maas AI, Servadei F, Skolnick BE, Tillinger MN, Marshall LF. Progression of traumatic intracerebral hemorrhage: a prospective observational study. J Neurotrauma. 2008;25: 629–39. 3. Iaccarino C, Schiavi P, Picetti E, et al. Patients with brain contusions: predictors of outcome and relationship between radiological and clinical evolution. J Neurosurg. 2014;120: 908–18. 4. Oertel M, Kelly DF, McArthur D, et al. Progressive hemorrhage after head trauma: predictors and consequences of the evolving injury. J Neurosurg. 2002;96:109–16. 5. Chieregato A, Fainardi E, Morselli-Labate AM, et al. Factors associated with neurological outcome and lesion progression in traumatic subarachnoid hemorrhage patients. Neurosurg. 2005;56:671–80. 6. Alahmadi H, Vachhrajani S, Cusimano MD. The natural history of brain contusion: an analysis of radiological and clinical progression. J Neurosurg. 2010;112:1139–45.

7. Tong WS, Zheng P, Xu JF, et al. Early CT signs of progressive hemorrhagic injury following acute traumatic brain injury. Neuroradiol. 2011;53:305–9. 8. Kaups KL, Davis JW, Parks SN. Routinely repeated computed tomography after blunt head trauma: does it benefit patients? J Trauma. 2004;56:475–81. 9. Sanus G, Zihni N, Tanriverdi T, Alver I, Aydin S, Uzan M. Evolving traumatic brain lesions: predictors and results of ninetyeight head-injured patients. Neurosurg Q. 2004;14:97–104. 10. Chang EF, Meeker M, Holland MC. Acute traumatic intraparenchymal hemorrhage: risk factors for progression in the early post-injury period. Neurosurg. 2006;58:647–56. 11. Reed SD, Bough DK, Meyer K, Jarvik JG. Inpatient costs, length of stay, and mortality for cerebrovascular events in community hospitals. Neurology. 2001;57:305–14. 12. Stein SC, Young GS, Talucci RC, Greenbaum BH, Ross SE. Delayed brain injury after head trauma: significance of coagulopathy. Neurosurg. 1992;30:160–5. 13. Givner A, Gurney J, O’Connor D, Kassarjian A, Lamorte WW, Moulton S. Reimaging in pediatric neurotrauma: factors associated with progression of intracranial injury. J Pediatr Surg. 2002;37:381–5. 14. Yuan F, Ding J, Chen H, et al. Predicting progressive hemorrhagic injury after traumatic brain injury: derivation and validation of a risk score based on admission characteristics. J Neurotrauma. 2012;29:2137–42. 15. Engström M, Romner B, Schalén W, Reinstrup P. Thrombocytopenia predicts progressive hemorrhage after head trauma. J Neurotrauma. 2005;22:291–6. 16. Schnüriger B, Inaba K, Abdelsayed GA, et al. The impact of platelets on the progression of traumatic intracranial hemorrhage. J Trauma. 2010;68:881–5. 17. Brain Trauma Foundation. Guidelines for the management of severe traumatic brain injury, 3rd edition. J Neurotrauma. 2007;24(Suppl 1):S1–2. 18. Kothari RU, Brott T, Broderick JP, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke. 1996;27:1304–5. 19. White CL, Griffith S, Caron JL. Early progression of traumatic cerebral contusions: characterization and risk factors. J Trauma. 2009;67:508–15. 20. Marshall LF, Marshall SB, Klauber MR, et al. A new classification of head injury based on computerized tomography. J Neurosurg. 1991;75:S14–20. 21. Maas AI, Hukkelhoven CW, Marshall LF, Steyerberg EW. Prediction of outcome in traumatic brain injury with computed tomographic characteristics: a comparison between the computed tomographic classification and combinations of computed tomographic predictors. Neurosurgery. 2005;57:1173–82. 22. Stein SC, Smith DH. Coagulopathy in traumatic brain injury. Neurocrit Care. 2004;1:479–88. 23. Fabbri A, Servadei F, Marchesini G, Stein SC, Vandelli A. Predicting intracranial lesions by antiplatelet agents in subjects with mild head injury. J Neurol Neurosurg Psychiatry. 2010;81: 1275–9. 24. Hukkelhoven CW, Steyerberg EW, Rampen AJ, et al. Patient age and outcome following severe traumatic brain injury: an analysis of 5600 patients. J Neurosurg. 2003;99:666–73. 25. Peterson EC, Chesnut RM. Talk and die revisited: bifrontal contusions and late deterioration. J Trauma. 2011;71:1588–92.

123

109

Mobile stroke units for prehospital thrombolysis, triage, and

Review

beyond: benefits and challenges

Mobile stroke units for prehospital thrombolysis, triage, and beyond: beneﬁts and challenges Klaus Fassbender, James C Grotta, Silke Walter, Iris Q Grunwald, Andreas Ragoschke-Schumm, Jeﬀrey L Saver

In acute stroke management, time is brain. Bringing swift treatment to the patient, instead of the conventional approach of awaiting the patient’s arrival at the hospital for treatment, is a potential strategy to improve clinical outcomes after stroke. This strategy is based on the use of an ambulance (mobile stroke unit) equipped with an imaging system, a point-of-care laboratory, a telemedicine connection to the hospital, and appropriate medication. Studies of prehospital stroke treatment consistently report a reduction in delays before thrombolysis and cause-based triage in regard to the appropriate target hospital (eg, primary vs comprehensive stroke centre). Moreover, novel medical options for the treatment of stroke patients are also under investigation, such as prehospital differential blood pressure management, reversal of warfarin effects in haemorrhagic stroke, and management of cerebral emergencies other than stroke. However, crucial concerns regarding safety, clinical efficacy, best setting, and costeffectiveness remain to be addressed in further studies. In the future, mobile stroke units might allow the investigation of novel diagnostic (eg, biomarkers and automated imaging evaluation) and therapeutic (eg, neuroprotective drugs and treatments for haemorrhagic stroke) options in the prehospital setting, thus functioning as a tool for research on prehospital stroke management.

Introduction Stroke is one of the most frequent causes of disability and death, and results in enormous societal costs associated with rehabilitation, long-term care, and loss of workforce.1 Safe and effective treatments for acute ischaemic stroke are available, including coordinated physiological care in a stroke unit, aspirin, intravenous thrombolysis with recombinant tissue plasminogen activator (alteplase) within 4·5 h for many acute patients,2 and intra-arterial treatment within the first 6 h for the subset of acute patients with cerebral ischaemia caused by large-vessel occlusion.3 However, the success of intravenous4,5 or intra-arterial3,6 treatment for ischaemic stroke is extremely timedependent. Every minute of delay before recanalisation counts, with an estimated 2 million neurons lost every minute.7 Based on the findings from large intravenous thrombolysis trials, the number needed to treat for one excellent outcome is approximately five in the first 90 min after symptoms onset, nine with treatment between 91 and 180 min, and more than 14 with treatment between 181 and 270 min.8 Other estimates show that for every 30-min delay before reperfusion by intra-arterial treatment, the relative likelihood of a good clinical outcome decreases by approximately 15%.6,9 Despite two decades of substantial efforts to streamline systems of care, reported alteplase treatment rates extracted from hospital-derived databases range from 3·4 to 9·1% for patients with acute ischaemic stroke,10–12 and the rates of delivery of intra-arterial treatment are far lower. The main reason for such undertreatment is that patients do not reach the hospital quickly enough to be assessed and treated within the narrow therapeutic window. Studies have found that only 15–60% of acute stroke patients arrive at the hospital within 3 h after onset of symptoms.13,14 Disappointingly, data from the Get With The Guidelines-Stroke programme (NCT02693223) www.thelancet.com/neurology Vol 16 March 2017

110

show that the proportion of patients with stroke arriving in time did not increase from 2003 to 2009, despite substantial eﬀorts to improve the system.15 This Review aims to describe the mobile stroke unit (MSU) approach to improving the care of patients with acute stroke. We also discuss the risks and opportunities associated with this novel medical option.

Bringing the hospital to the patient: the mobile stroke unit Existing recommendations for prehospital management of stroke, after early stabilisation and initial assessment, include rapid transfer to the nearest hospital for treatment.16,17 However, eﬀective treatments are available for patients with acute stroke that could, in principle, be administered immediately. Thus, by contrast to the approach of awaiting for the patient arrival at the hospital, the approach of administering treatment directly at the emergency site (termed the MSU concept) was developed in 2003 and investigated in clinical reality in 2008,18 adhering to current stroke guidelines and emergency medical services legislations. As a mobile emergency room,19,20 the MSU contains imaging, a point-of-care laboratory, and a telemedicine connection with a hospital, in addition to appropriate medication and assessment tools. Its aim is the delivery of state-of-the-art prehospital diagnosis and treatment, as well as diagnosis-based triage of the patient to the most appropriate target hospital. Treatments include thrombolysis for acute cerebral ischaemia, anticoagulant reversal for acute intracranial haemorrhage, management of physiological variables for ischaemic or haemorrhagic stroke, and management of further emergencies. Thus, the MSU extends specialised stroke care speciﬁcally to the prehospital phase of stroke management and could act synergistically with hospital stroke units to close the existing treatment gap for patients with acute stroke.

Lancet Neurol 2017; 16: 227–37 This online publication has been corrected. The corrected version ﬁrst appeared at thelancet.com/ neurology on February 15, 2017 Department of Neurology, University Hospital of the Saarland, Homburg, Germany (Prof K Fassbender MD, S Walter MD, A Ragoschke-Schumm MD); Department of Neurology, University of Texas Medical School at Houston, Houston, TX, USA (Prof J C Grotta MD); Neuroscience and Vascular Simulation Unit, Faculty of Medical Science, PMI, Anglia Ruskin University, Chelmsford, UK (Prof I Q Grunwald MD); Department of Stroke Medicine, Southend University Hospital, Southend, UK (Prof I Q Grunwald); and the Stroke Center, University of California Los Angeles, Los Angeles, CA, USA (Prof J L Saver MD) Correspondence to: Prof Klaus Fassbender, Department of Neurology, University of the Saarland, Homburg 66424, Germany klaus.fassbender@uks.eu

227

Review

In general, acute stroke management involves complex multistep testing and multidisciplinary cooperation. Different groups of health-care professionals in various locations are sequentially involved and repeatedly collect medical history, perform clinical examinations, and hand off patients and information to others before treatment can begin (figure 1). The multiple interfaces can contribute to errors and delays. The MSU concept,18 apart from reducing prehospital and in-hospital transport times, saves crucial time by substantially reducing those interfaces. At one location, a single, specialised, interdisciplinary team, consisting of paramedics, physicians, nurses, and technicians, performs the complete diagnostic work-up and acute treatment in a parallel workflow (figure 1).

The mobile stroke unit ambulance The MSU is an ambulance that contains all the tools necessary for guideline-adherent initial acute stroke treatment, and standard emergency care equipment.18 Dimensions of existing ambulances vary according to the specific needs of various regions and health-care systems. In a second-generation MSU, the size of the equipment has been substantially reduced so that it can fit into a commercially available standard ambulance, which not only improves speed, accessibility to narrow roads, and acceptance by emergency care personnel, but also reduces cost.21 However, in other settings, larger vehicle solutions might be advantageous— eg, an extra space allowing relatives to accompany the patient might be beneficial to collect a medical history and obtain informed consent for subsequent treatments.22 A more robust vehicle might also be important to cope with bad roads or for transportation of large scanners. Prehospital

MSU and emergency medical services MSU-based

Stabilisation Diagnosis Therapy (ie, thrombolysis) Triage Transport

Emergency medical services Stabilisation Transport

Telecommunication approaches between standard ambulances and the stroke centre via systems that can provide real-time audiovisual conferencing and sharing of images have been studied for more than a decade. Pioneering work using early mobile broadband systems indicated potential benefits when hospital neurologists were able to assess neurological presentation en route by telemedicine, aiding in the best choice of target hospital and accelerating subsequent in-hospital care processes. However, the first studies faced problems with the reliability of telecommunication and were performed in simulated scenarios rather than in real clinical settings. In later studies, telemedicine transmission of videos of neurological examinations in ambulances, using actors mimicking stroke symptoms23–26 or involving stroke patients,27 was considerably better but still not completely reliable because of the suboptimal 3G public network available at the time. Further improvements have been reported with use of 4G mobile systems.28 Similarly, studies have demonstrated the feasibility of en-route transmission of structured patient information to the target hospital via personal digital assistants.29,30 Telemedicine, including telestroke assessment (realtime bidirectional videoconferencing and high-speed transmission of videos) and teleradiology (transmission of high-quality images), is an integral component of the MSU concept.18 Commercially available systems, routinely used for telemedicine between hospitals, allow MSUs to transmit digital imaging and communication data to the picture-archiving and communication system of the associated health-care facility,18,31 acting within the same firewall. As a result of available 4G connectivity and the prioritisation of telecommunication in emergency medical services, a study32 has reported that telemedicine Hospital 1

Large vessel occlusion, intracranial haemorrhage Yes

Conventional

Telecommunication between ambulance and hospital

Hospital 2

Comprehensive stroke centre Specialised treatment (intra-arterial treatment, neurointensive care, neurosurgery) Primary stroke centre Stroke unit treatment

Primary stroke centre Diagnosis Therapy (ie, thrombolysis) Triage

Large vessel occlusion, intracranial haemorrhage

Comprehensive stroke centre Specialised treatment (intra-arterial treatment, neurointensive care, neurosurgery)

Figure 1: MSU-based stroke management compared with conventional stroke management In MSU-based stroke management, patients are diagnosed at the site of emergency, allowing case-speciﬁc treatment and triage to the most appropriate stroke centre, thus avoiding secondary transfers. In conventional stroke management, due to insuﬃcient knowledge about the cause of the symptoms, patients are transported to the primary stroke centre and eventually, by secondary transfer, brought to a comprehensive stroke centre. MSU=mobile stroke unit.

228

www.thelancet.com/neurology Vol 16 March 2017

111

Review

encounters between MSUs and hospitals were successfully completed in the management of 99 (99%) of 100 patients. A study33 of simultaneous independent assessment by a vascular neurologist on board an MSU and a remote telemedicine-based vascular neurologist showed 98% satisfactory connectivity and 88% agreement on the alteplase decision. This level of agreement was the same as that between two vascular neurologists evaluating the same patients face-to-face in the emergency department.34

Staﬃng: who needs to be on board? Although most research projects have included a vascular neurologist on board the ambulance,18,31,35–37 absolutely reliable telecommunication connectivity permits the sole reliance on paramedics, with or without nurses and radiographers, guided by neurologists and neuroradiologists at a remote hospital via telemedicine.32–34 Research in Norway38 is exploring whether a prehospital stroke diagnostic work-up, including neuroradiological assessment, can be performed by trained non-neurologist physicians. Because of legal restrictions, most MSUs operate in addition to conventional emergency medical services. Moreover, in published studies, use of MSUs has been restricted to daytime and evening hours. For example, in the ﬁrst randomised trial,35 the MSU was in operation from 0800 h to 2200 h from Monday to Friday, and from 0800 h to 1800 h on weekends. With further evidence, and the resolution of legal and reimbursement issues, operational models might change.

Prehospital brain imaging Imaging is key in the management of acute stroke. For acute cerebral ischaemia, CT or MRI must exclude haemorrhage before intravenous thrombolysis can be A

provided.16,17 Signs of early infarction can predict a reduced response to alteplase and an increased rate of adverse effects. In addition, vascular imaging (CT angiography or MR angiography) is recommended for identification of large-vessel occlusion, in light of novel endovascular treatment options.39 For acute intracerebral haemorrhage, imaging results such as the so-called spot sign can allow estimation of the risk of early haematoma growth.40 So far, multimodal imaging (non-contrast CT, CT angiography, and CT perfusion)18,31,35 excluding MRI has been integrated into MSUs (figure 2).41 Most scanners used in ambulances were originally designed for use in intensive care units. As such, they are portable, accumulator driven, and radiation shielded.18,35 The images produced by the 8-slice CT scanners most frequently used in MSUs (eg, Ceretom, Neurologica/ Samsung, Boston, MA, USA) are of sufficient quality for brain parenchymal imaging and for CT angiography of the intracranial circulation, including the circle of Willis, although these scanners do not allow assessment of the proximal neck vessels or the aortic arch. A different scanner (Somatom Scope, Siemens, Erlangen, Germany), which is being used in an MSU research project in Memphis (TN, USA), allows not only higher-resolution imaging than does the Ceretom but also assessment of the neck vessels and the aortic arch. However, these improvements come at the expense of bigger size, requiring more space, and on-board power generation.

Prehospital point-of-care laboratory According to guidelines,16,17 a limited number of laboratory tests (eg, platelet, leucocyte, and erythrocyte counts; haemoglobin and glucose levels; activated partial thromboplastin time and international normalised ratio; γ-glutamyltransferase and pancreatic amylase activity) C

Figure 2: Multimodal imaging in a mobile stroke unit Non-contrast CT (A), CT angiography (B), and ASPECTS (C) done in a mobile stroke unit of a 73-year-old woman with acute right hemiparesis. Although the parenchyma shows no signs of infarction (ASPECTS 10), CT angiography allowed prehospital diagnosis of an occlusion of the left middle cerebral artery (B, arrow). Reproduced from Grunwald et al,41 by permission of Cerebrovascular Diseases (Karger). ASPECTS=Alberta Stroke Program Early CT Score.

www.thelancet.com/neurology Vol 16 March 2017

112

229

Review

are recommended, but should not delay the start of thrombolysis. The only laboratory test required in most patients before start of ﬁbrinolytic therapy is glucose level. Platelet count and measures of coagulation function should be obtained before therapy in case of clinical suspicion of bleeding abnormality or thrombocytopenia, or if the patient could have received anticoagulant therapy. Knowledge of creatinine level is considered helpful for concerns regarding renal function. Generally, evidence is poor regarding the relevance of these laboratory tests in decision making in acute stroke management and, to avoid delays, many centres do not await the laboratory work-up (except of glucose levels and international normalised ratio) in the absence of known coagulopathy before administering alteplase. Extended laboratory testing was shown to be more expeditiously conducted with a point-of-care laboratory system than in a hospital laboratory and can be used in MSUs.18,31,35 Pointof-care testing devices must comply with legislative directives to assure patient safety (ie, CE marking, indicating conformity for products sold in the European Community). A preliminary in-hospital proof-of-principle study42 has shown that use of a point-of-care laboratory decreased time from door to therapy decision (end of all diagnostic procedures) from 84 min (SD 26) to 40 min (SD 24; p<0·0001) compared with use of a centralised hospital laboratory. Nonetheless, the relevance of pointof-care laboratory results on treatment decisions is unclear, and requires clariﬁcation in further studies.

Instruments for dispatch of the ambulance For the most efficient use of MSUs, dispatchers need to identify patients with stroke with the highest possible sensitivity and specificity. Generally, identification of stroke through telephone calls is a serious challenge for dispatchers at emergency medical services.43,44 This requirement might be facilitated by the use of stroke checklists during the initial screening.44–46 So far, the Recognition of Stroke in the Emergency Room (ROSIER) scale46 with a reported sensitivity of 93% and specifity of 83% and the Dispatcher Identification Algorithm of Stroke Emergency (DIASE), with a reported sensitivity of 53% and a specificity of 97% for stroke,47 have been used as inventories for dispatch in MSU studies.18,31,35,37,48 For the future development of a dispatch tool, a good balance between sensitivity and specificity is necessary so that strokes will not be missed, but the MSU will not be dispatched too often for conditions mimicking stroke.

Prehospital stroke management After prehospital stroke thrombolysis was shown to be feasible,18 a randomised, single-centre trial,35 in which 361 patients were screened and 100 recruited, was done by Saarland University in Homburg, Germany. In this MSU trial,35 prehospital stroke management achieved a median time from symptom onset to therapy decision of 56 min (IQR 43–103) and a median time from symptom 230

onset to treatment of 72 min (53–108), without safety concerns (table). Such reductions in delays before treatment were confirmed in the PHANTOM–S (Pre– Hospital Acute Neurological Therapy and Optimization of Medical Care in Stroke) study,37 in an observational study50 in Houston (TX, USA), and in a case series51 in Cleveland (OH, USA; table). The times from these studies are much faster than those observed in all earlier studies evaluating the effect of interventions on reduction of delays to treatment in the emergency department,44 with reported time from symptom onset to treatment usually exceeding 120 min. These metrics also exceed those of stroke management observed in clinical practice, with reported median time from symptom onset to treatment of 140 min (110–165) in the Safe Implementation of Thrombolysis in Stroke–Monitoring Study (SITS– MOST),54 which involved 6853 patients treated at 285 European centres between 2002 and 2006, or of 144 min (115–170) in the Get With The Guidelines-Stroke registry, which involved 58 353 patients treated in 1395 US hospitals between 2003 and 2012.5 In addition to accelerating the initiation of therapy for patients who would have been treated with thrombolysis in hospital, MSU treatment will also enable more patients to be treated within the recommended therapeutic window of 4·5 h.16,17 This capacity is also suggested by the improved treatment rates in the MSU trial35 and the significant increase in treatment rates of prehospital stroke treatment in the PHANTOM-S trial37 (table). The question of how many additional patients could be treated with thrombolysis if the MSU concept were widely used is intriguing, and should be answered in future studies done in specific health-care systems and settings.

Breaking the golden hour limit The term golden hour has been attributed to the trauma surgeon R Adams Cowley, who recognised that the sooner trauma patients receive definitive care— particularly within the first hour after trauma—the better their chance of survival. To illustrate the pronounced time sensitivity of acute stroke management,3–9,55 the term has also been applied to this clinical situation. However, only a very small proportion of patients with stroke receive treatment during the golden hour. In the National Institute of Neurological Disorders and Stroke (NINDS) study,56 only two (<1%) of 302 patients were randomly assigned to a study group within 60 min (both were assigned to placebo). In the Safe Implementation of Treatments in Stroke–International Stroke Thrombolysis Registry (SITS–ISTR) observational study,57 only 166 (1·4%) of 11 429 patients were treated within 60 min, and in the Get With The Guidelines-Stroke registry,5,58 750 (1·3%) of 58 353 patients were treated within this time, despite substantial efforts to improve stroke management over the past decade. Even in studies with streamlined protocols leading to door-to-needle times as short as 20 min, no more than 10% of patients were www.thelancet.com/neurology Vol 16 March 2017

113

Review

Site

Study design Condition

Number of patients with intervention vs standard care*

Time from alarm† to CT (min)

Time from alarm† to therapy decision or therapy (min)

Time from symptom onset to therapy decision or therapy (min)

Number (%) of treated patients with ischaemic stroke

Number (%) of patients with time to therapy decision‡ or therapy ≤60 min

Walter et al18 Homburg, Case report Germany

Acute stroke

34, 33

To therapy decision: 35, 33

To therapy decision: 60, 50

··

Kostopoulos Homburg, Case series et al31 Germany

Acute stroke

38, 41, 41, 27

To therapy decision: 38, 41, 41, 27

To therapy decision: 98, ·· 78, 46, 35

··

Walter et al35 Homburg, Randomised Germany

Acute stroke

53 vs 47

34 (30–38) vs 71 (62–87), p<0·0001

To therapy decision: 35 (31–39) vs 76 (63–94), p<0·0001; to alteplase: 38 (34–42) vs 73 (60–93), p<0·0001

To therapy decision: 56 (43–103) vs 104 (80–156), p<0·0001; to alteplase: 72 (53–108) vs 153 (136–198), p=0·0011

12 (41%) of 29 vs 8 To therapy decision: 30 (32%) of 25, p=0·48 (57%) of 53 vs 2 (4%) of 47, p<0·0001

Weber et al48 Berlin, Germany

Observational Acute stroke

45 vs 50 historical controls

··

To alteplase: 58 (50–65) vs 92 (79–112)

To alteplase: 97 (69–156)

23 (51%) of 45

Ebinger et al37,49

Randomised

Acute stroke

1804 vs 2969

35 (30–42) vs 50 (43–59)

To alteplase: 48 (39–56) vs 72 (62–85), p<0·001

To alteplase: 81 (56–129) vs 105 (81–145), p<0·001

200 (33%) of 614 vs To alteplase (subgroup 220 (21%) of 1 041, analysis of treated patients): 62 (31%) of p<0·001 200 vs 16 (5%) of 330, p<0·01

Bowry et al50 Houston, TX, USA

Case series

Acute stroke

··

To alteplase: 98 (47–265)‡; to intra-arterial therapy: 175 (140–224)‡

12 (50%) of 24

To alteplase: 4 (33%) of 12

Parker et al36 Houston, TX, USA

Case series

Acute stroke

··

On-scene time to ·· alteplase: 24 (12–53)‡

··

4 (31%) of 13

Taqui et al51

Cleveland, Case series OH, USA

Acute stroke

23 vs 34 historical controls

41 (33–47) vs 62 (50·5–97·5), p<0·0001

To alteplase: 64 (58·3–72·3) vs 105 (99–115), p=0·008

To alteplase: 115 (77·5–144) vs 125 (97·5–151·5), p=0·52

6 (26%) of 23 vs 5 (15%) of 34

··

Itrat et al32

Cleveland, Observational Acute stroke OH, USA study

100 vs 56 historical controls

From door: 13 (9–21) vs 18 (12–26), p=0·003

Door to alteplase: 32 (24–47) vs 58 (53–68), p<0·001

··

Cerejo et al52 Cleveland, Observational Acute stroke OH, USA study treated with intra-arterial therapy

5 vs 5 historical controls

From door: 12 (9–14) vs 32 (22–375)

CT to intra-arterial therapy: 82 (65–103) vs 165 (150–201)

··

Kunz et al53

305 vs 353

··

To alteplase: 46 (39–53) vs 76 (64–93), p<0·0005

To alteplase: 73 (53–120) vs 112 (85–175), p<0·0005

··

Berlin, Germany

Observational Acute stroke treated with registry alteplase study§

To alteplase: 4 (17%) of 23

Times are given as individual times or median (IQR) unless otherwise stated. Therapy decision deﬁned as end of all stroke diagnostic examinations. *If applicable. †If starting point is not otherwise speciﬁed. ‡Data are mean (range). §Including patients from the PHANTOM-S Trial35 and its pilot study.45

Table: Studies of stroke management in prehospital mobile stroke units

treated within 70 min after symptom onset59—a finding suggesting a possible ceiling effect with regard to the timing of conventional stroke management. MSUs can break this golden hour limit, as shown in the first MSU trial,35 by the increase in rates of patients with therapy decision within 60 min of symptom onset (table). Further studies support this conclusion, with increased rates of treatment within 60 min in the PHANTOM-S trial49 and in the Houston MSU programme36 (table). These unprecedentedly short times to treatment could translate into improved clinical outcomes, although the extent of this improvement is not fully known owing to the small number of patients treated within such time dimensions to date.44,60,61 So far, stroke patients treated within such an early time frame www.thelancet.com/neurology Vol 16 March 2017

114

have been reported to have signiﬁcantly better clinical outcomes than those treated later.5,58

Prehospital treatment of haemorrhagic stroke Because haemorrhage enlargement occurs very early in the course of intracerebral haemorrhage, the time is brain concept probably also applies to haemorrhagic stroke. Differential adjustment of blood pressure might be beneficial for patients with ischaemic or haemorrhagic stroke,62 and recommendations for blood pressure management differ for stroke caused by ischaemia (elevated blood pressure as high as 185/110 mm Hg can be tolerated by patients receiving thrombolysis)17 or by haemorrhage (reduction of systolic blood pressure to 140–160 mm Hg is safe and can be clinically effective).63 231

Review

However, during the prehospital phase in conventional ambulances, the cause of stroke remains unknown, meaning that early differential blood pressure management is impossible. Diagnostic clarification using an MSU allows differential adjustment of physiological variables such as blood pressure before hospital arrival.18 Warfarin-related intracranial haemorrhage composes 15–20% of intracerebral haemorrhages and is associated with a high mortality.64 Guidelines recommend reversal of warfarin’s effects by medication, such as prothrombin complex concentrates.63 With prehospital diagnosis of haemorrhagic stroke, warfarin effects can be reversed directly at the emergency site.65 In the future, such an approach might even be extended to reversal of the effects of novel oral anticoagulants by antidotes.

Triage to the target hospital International guidelines16,17 recommend prenotification of the target hospital about a patient en route as an evidence-based measure to accelerate in-hospital stroke management.44 MSUs offer the unique option to provide the target hospital with detailed information about the cause of the disease and any information needed for subsequent specialised treatment. Guidelines recommend that patients must be transported to the closest available hospital or to a stroke centre or, if no such facility is nearby, to the most appropriate institution that can provide stroke care.16,17 This practice is under discussion in light of the emerging evidence for the safety and efficacy of intraarterial treatment, and the possibility of performing this specialised treatment only at comprehensive stroke centres, but not at primary stroke centres. Indeed, patients with large-vessel occlusion often arrive at endovascular centres too late for effective treatment if they are first transferred to a hospital without endovascular treatment capabilities and later transferred to a comprehensive stroke centre (figure 1). Patients in the Interventional Management of Stroke III study66 who were given alteplase at a primary stroke centre before transfer to a comprehensive stroke centre (ie, the drip and ship method) had significantly longer median times from alteplase to groin puncture (105 min, IQR 47) than did those directly admitted to and treated in the comprehensive stroke centre (ie, the mothership method; 83 min, IQR 31; p<0·0001). Similar delays were observed in patients treated in the Endovascular Treatment for Small Core and Proximal Occlusion Ischemic Stroke (ESCAPE) trial,67 in which time from symptom onset to door of the comprehensive stroke centre was 41% (32 min) longer if patients received alteplase at the referring hospital rather than at the comprehensive stroke centre. It has been estimated that every minute of delay in transfer reduces the probability that patients will receive intra-arterial treatment by 2·5%.68 232

Therefore, accurate triage with regard to the appropriate target hospital would avoid the transfer of patients with large-vessel occlusion to hospitals without endovascular treatment services and, at the same time, prevent the transfer of all patients to comprehensive stroke centres, which prevents efficient use of limited resources. Options being discussed to improve the accuracy of triage under discussion include the use of clinical stroke scales aimed at differentiating patients with or without large-vessel occlusion.69 A retrospective study70 involving 119 patients showed that a score of 4 or higher on the Los Angeles Motor Scale predicts the presence of large-vessel occlusion with a sensitivity of 81% and a specificity of 89%. A prospective evaluation71 involving 357 patients found that a score of 5 or higher on the Rapid Arterial Occlusion Evaluation scale predicts the presence of largevessel occlusion with a sensitivity of 88% and a specificity of 68%. In a prospective study,72 the Field Assessment Stroke Triage for Emergency Destination scale, based on items of the National Institutes of Health Stroke Scale with a cutoff value of 2 or higher, exhibited a sensitivity of 60% and a specificity of 89% for predicting large-vessel occlusion. A retrospective investigation73 found that the Prehospital Acute Stroke Severity scale, with a cutoff of 2 or higher, exhibited a sensitivity of 66% and a specificity of 83% in predicting large-vessel occlusion. However, all studies have found that a sizable number of large-vessel occlusions are missed by clinical scores.74,75 Furthermore, all of these scales have been developed in emergency departments by physicians, and none have been tested by personnel in the prehospital environment. Development of such a scale or validation of existing scales is another area ripe for investigation on the MSU. By contrast, when vascular imaging such as CT angiography is implemented in the MSU,35 correct triage of patients with stroke to the appropriate target hospital becomes possible owing to knowledge about the presence or absence of large-vessel occlusion directly at the emergency site (figure 2).31 Use of a MSU even with noncontrast imaging has been associated with reduction of delay before intra-arterial treatment in patients with large-vessel occlusion.52 Analogously, prehospital imaging has been shown to allow the triage of patients with haemorrhagic stroke to hospitals with neurosurgery services, bypassing hospitals without such capabilities.31 A subgroup analysis76 of the PHANTOM-S study37 found that the rate of patients with haemorrhage delivered to hospitals without neurosurgery services decreased from 65 (43%) of 151 patients in the conventional treatment group to seven (11%) of 62 patients in the intervention group. Finally, although prehospital diagnosis and treatment of stroke are likely to remain the main indications for the use of MSUs, patients with several time-sensitive cerebral conditions other than stroke, such as traumatic brain injury or status epilepticus, could also benefit from the diagnosis-based triage enabled by MSUs.77 www.thelancet.com/neurology Vol 16 March 2017

115

Review

Conclusions and future directions Worldwide, the number of prehospital stroke treatment programmes is rapidly increasing, with more than 20 sites studying their provision (figure 3). Most of these groups cooperate in the context of the Pre-hospital Stroke Treatment Organization (PRESTO), dedicated to improving data exchange and collaboration. However, whether prehospital stroke treatment will be increasingly implemented and sustainable over time remains to be seen; this outcome depends on the resolution of issues on safety, long-term clinical benefit, best setting (eg, rural vs urban, regional emergency medical services configuration), and cost-effectiveness. The MSU strategy cannot replace continuous efforts to improve the quality of the standard prehospital and in-hospital stroke care.44 Because the time from stroke onset to emergency call is beyond the influence of the deployment of an MSU, public education projects are of relevance.44 The MSU, as a sort of rolling billboard in the community, might in itself promote public awareness. Diagnosis and treatment in an MSU can occur with a level of expertise and equipment comparable to those in a hospital;18 thus, treatment in an MSU is expected to be as safe as treatment in a hospital. Technical failure rates are, at present, within the range of those associated with routine emergency ambulances, and the corresponding hospital equipment. One concern might be that earlier evaluation might result in alteplase treatment for an increased number of stroke mimics. This concern applies

to every intervention that aims to decrease time to treatment, which inherently shortens the time to observe the natural disease course. Although previous studies32,35–37,48 of the deployment of an MSU have found no significant differences with regard to indicators of safety, such as haemorrhagic complications or mortality, safety is still a relevant issue for future research. According to the generally accepted concept that time is brain,7–9 the clear reduction of delays before treatment argues for a clinical benefit. However, the first randomised trial35 of prehospital stroke treatment found no differences in 7-day modified Rankin scores and National Institutes of Health Stroke Scale scores—a finding that can be explained by the insufficient power of the trial to detect differences in these endpoints. Also, although the PHANTOM-S trial37 found that patients who received prehospital stroke treatment were more likely to be discharged home, the study did not find significant improvements in short-term outcomes. An evaluation53 of a large registry of patients treated with alteplase between 2011 and 2015 did not find a significant increase in the primary outcome of a 90-day modified Rankin score of 0 or 1 (53% vs 47%; p=0·14), despite the fact that 112 (37%) of 305 patients treated in the MSU received thrombolysis within the golden hour. However, results significantly favoured the MSU cohort when the analysis was adjusted for baseline differences between the non-randomised groups, and the study also found positive dichotomised secondary outcomes, such as a modified Rankin score of

Toledo (OH, USA) Chicago Edmonton (IL, USA) (Canada)

Cleveland (OH, USA)

Phoenix (AZ, USA)

Berlin Drobak (Germany) Helsinki (Finland) (Norway)

Leuven (Belgium) Southend (UK)

Memphis (TN, USA) Denver (CL, USA) Los Angeles (CA, USA)

Marburg (Germany)

Lille (France) New York Paris (France) (NY, USA) Homburg/Saar Jacksonville (Germany) (FL, USA) Aarau (Switzerland) Doha (Qatar)

Houston (TX, USA)

Buenos Aires (Argentina)

Bangkok (Thailand)

Melbourne (Australia)

Figure 3: Mobile stroke unit research projects Blue circles indicate sites with active mobile stroke units and red triangles indicate sites in which such projects are in the state of implementation.

www.thelancet.com/neurology Vol 16 March 2017

116

233

Review

2 or lower and mortality. Thus, direct evidence from controlled trials comparing similar patients managed by the MSU or by standard emergency medical care is still needed. Indeed, the Benefits of Stroke Treatment Delivered Using a Mobile Stroke Unit (BEST-MSU) trial50—a prospective, randomised multicentre trial coordinated in Houston (TX, USA)—was initiated in 2015. This study, which is planned to include 900 patients, is expected to provide information about clinical and costeffectiveness of the MSU approach. Importantly, the generalisability of the previous results on prehospital stroke treatment needs to be demonstrated in various health-care systems with differing configurations of emergency care, legislation requirements, market forces, and demography. The results of the more than 20 ongoing and planned projects worldwide (figure 3) might provide such much-needed information. The configuration of emergency services is highly variable across different countries. For example, in some European countries the presence of a physician is mandatory, whereas in other countries emergency services are exclusively staffed by paramedics.44 Regional differences in conventions and legislation clearly affect the configuration of MSUs and the degree of integration of the MSU concept into the setting of emergency services. Further research is also needed regarding optimised interaction between MSUs and primary or comprehensive stroke centres in different health-care environments. The benefit of this approach strongly depends on the regional medical standard. Thus, in countries with no thrombolytic options available, bottom-up implementation of such treatment options in the hospital takes priority. Moreover, more research is needed to determine the suitability of the MSU concept in urban, suburban, and rural regions. Because the number of dispatches increases with population density, the MSU appears, at first glance, to be especially advantageous in urban regions. However, because rural regions are often highly underserved with regard to stroke expertise,79 the value of this strategy could also be substantial in rural areas.78 MSUs could also meet the regular emergency medical services ambulance en route at a predefined meeting point. This approach is comparable to that of bringing CT scanners to rural hospitals without such imaging facilities.80 Despite the evidence indicating improvements in process management for one of the most expensive diseases,1 important concerns remain with regard to the Search strategy and selection criteria We searched PubMed between Jan 1, 2010, and Dec 31, 2016, for the terms “stroke” combined with “prehospital”, “pre-hospital”, “ambulance”, “emergency medical service”, or “mobile stroke unit” and found 717 publications. We reviewed articles focusing on originality, timeliness, and relevance to the broad scope of this Review.

234

potentially unacceptable costs of the MSU, its staffing, and its deployment. However, two independent preliminary cost-effectiveness analyses of MSU systems based on modelling health-care costs and outcomes had encouraging results. Dietrich and colleagues78 performed a 1-year cost–benefit analysis of MSUs across a number of scenarios, based on the first controlled MSU trial.35 The economic benefits outweighed its costs: the benefit– cost ratio was 1·96, even in the baseline experimental setting and with two physicians on board. Benefit–cost ratios increased substantially with gradual reductions of staff (eg, use of telemedicine) and with higher population density. Maximum benefit–cost ratios between 2·16 and 6·85 were identified at optimum operating distances ranging from 26·73 to 40·32 miles, depending on the staff configuration. Although efficiency is positively related to population density, benefit–cost ratios can also be greater than 1 in rural settings. A further estimate by Gyrd-Hansen and colleagues81 also suggested the cost-effectiveness of prehospital stroke treatment. This estimate considered the results of the PHANTOM-S trial,37 with an annual net cost of €963 954 because of more frequent and earlier administration of alteplase, and calculated an annual expected health gain, related to the avoidance of 18 cases of disability, equal to 29·7 quality-adjusted life-years. This calculation produced an incremental cost-effectiveness ratio of €32 456 per quality-adjusted life-year. This estimate meets the standard threshold to judge a system innovation as a cost-effective intervention. In the future, cost-effectiveness might be considerably improved by various measures—eg, substitution of physically present physicians for remote hospital experts linked via telemedicine,32 use of standard ambulance solutions as the basis of the unit,21 increased usage rates, or improved stroke-identification algorithms for dispatchers.78 Demonstrating the cost-effectiveness of MSU deployment as a precondition for its future reimbursement remains a key research issue. This research should include prospective data on the actual costs for establishing and maintaining MSUs, and for both acute and long-term care of patients managed both on MSUs and by standard emergency services, and should include analysis from the perspective of MSU providers, tax payers, and patients. In conclusion, this Review of prehospital stroke treatment and the data emanating from MSU studies shows that diagnostic clarification in the ambulance allows not only prehospital thrombolysis, comprehensive prenotification, and correct triage to the most appropriate target hospital, but novel options that might also include cause-specific adjustment of physiological parameters, reversal of anticoagulant effects, and prehospital management of other cerebral emergencies. The MSU allows future research on diagnostic and therapeutic options such as blood markers of cerebral damage,82 transcranial duplex ultrasonography83 and automated imaging decision support tools,41 improved www.thelancet.com/neurology Vol 16 March 2017

117

Review

clot-dissolving medications, sonothrombolysis,84 neuroprotectants,85 and haemostatic treatments to reduce haematoma growth.86 MSUs can serve as an ideal tool for research on hyperacute stroke, profiting from the valuable contribution of paramedics.85,87,88 Additional studies are needed to substantiate safety, clinical benefit, best setting, and cost-effectiveness as preconditions for a wide implementation of MSUs in clinical practice. Contributors All authors contributed to the literature search and to writing and reviewing the manuscript. Declaration of interests JCG is an employee of Memorial Hermann Hospital. Genentech supplies alteplase for his MSU research. JCG is a consultant to Stryker, which manufactures equipment used on MSUs and devices for stroke treatment, and to Frazer Ltd, which manufactures MSUs. IQG is cofounder and Medical Director of Brainomix Ltd, an Oxford University spin-out that is not in competition with this Review. She is involved in the running of medical conferences that have received industry support from Penumbra, Medtronic/Covidien, Micrus/Codman, Acandis, Stryker, Phenox, Gore, Siemens, Phillips, Toshiba, GE, Simbionix/3D Systems, Cook, Microvention, Balt, Abbott and Mentice, but does not receive payments for this. She has received consultancy fees or travel support from Micrus, Medtronic, Penumbra, 3D Systems, and Mentice. JLS has served as an unpaid site investigator in multicentre trials run by Medtronic and Stryker for which the University of California Regents received payments on the basis of clinical trial contracts for the number of subjects enrolled. JLS receives funding for services as a scientific consultant regarding trial design and conduct from Medtronic/Covidien, Stryker, Neuravi, BrainsGate, Pfizer, Squibb, Boehringer Ingelheim (prevention only), ZZ Biotech, and St Jude Medical. JLS serves as an unpaid consultant to Genentech, advising on the design and conduct of the PRISMS trial; neither the University of California nor JLS have received any payments for this voluntary service. The University of California has patent rights in retrieval devices for stroke. All remaining authors declare that they have no competing interests. Acknowledgments Research on prehospital stroke management is funded by the Ministry of Health of the Saarland, Germany (KF) and by grants from the American Heart Association and the Patient Centered Outcomes Institute (JCG). References 1 Ovbiagele B, Goldstein LB, Higashida RT, et al. Forecasting the future of stroke in the United States: a policy statement from the American Heart Association and American Stroke Association. Stroke 2013; 44: 2361–75. 2 Demaerschalk BM, Kleindorfer DO, Adeoye OM, et al. Scientific rationale for the inclusion and exclusion criteria for intravenous alteplase in acute ischemic stroke: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2016; 47: 581–641. 3 Goyal M, Menon BK, van Zwam WH, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet 2016; 387: 1723–31. 4 Strbian D, Soinne L, Sairanen T, et al. Ultraearly thrombolysis in acute ischemic stroke is associated with better outcome and lower mortality. Stroke 2010; 41: 712–16. 5 Saver JL, Fonarow GC, Smith EE, et al. Time to treatment with intravenous tissue plasminogen activator and outcome from acute ischemic stroke. JAMA 2013; 309: 2480–88. 6 Sun CH, Ribo M, Goyal M, et al. Door-to-puncture: a practical metric for capturing and enhancing system processes associated with endovascular stroke care, preliminary results from the rapid reperfusion registry. J Am Heart Assoc 2014; 3: e000859. 7 Saver JL. Time is brain—quantified. Stroke 2006; 37: 263–66. 8 Lees KR, Bluhmki E, von Kummer R, et al. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet 2010; 375: 1695–703.

www.thelancet.com/neurology Vol 16 March 2017

118

12 13 14 15

16 17

18 19 20 21 22 23 24 25 26 27 28 29

Khatri P, Yeatts SD, Mazighi M, et al. Time to angiographic reperfusion and clinical outcome after acute ischaemic stroke: an analysis of data from the Interventional Management of Stroke (IMS III) phase 3 trial. Lancet Neurol 2014; 13: 567–74. Adeoye O, Hornung R, Khatri P, Kleindorfer D. Recombinant tissue-type plasminogen activator use for ischemic stroke in the United States: a doubling of treatment rates over the course of 5 years. Stroke 2011; 42: 1952–55. Schwamm LH, Ali SF, Reeves MJ, et al. Temporal trends in patient characteristics and treatment with intravenous thrombolysis among acute ischemic stroke patients at Get With The Guidelines-Stroke hospitals. Circ Cardiovasc Qual Outcomes 2013; 6: 543–49. Scholten N, Pfaff H, Lehmann HC, Fink GR, Karbach U. Thrombolysis for acute stroke—a nationwide analysis of regional medical care. Fortschr Neurol Psychiatr 2013; 81: 579–85. Agyeman O, Nedeltchev K, Arnold M, et al. Time to admission in acute ischemic stroke and transient ischemic attack. Stroke 2006; 37: 963–66. Evenson KR, Foraker RE, Morris DL, Rosamond WD. A comprehensive review of prehospital and in-hospital delay times in acute stroke care. Int J Stroke 2009; 4: 187–99. Tong D, Reeves MJ, Hernandez AF, et al. Times from symptom onset to hospital arrival in the Get with the Guidelines-Stroke Program 2002 to 2009: temporal trends and implications. Stroke 2012; 43: 1912–17. European Stroke Organisation (ESO) Executive Committee, ESO Writing Committee. Guidelines for management of ischaemic stroke and transient ischaemic attack 2008. Cerebrovasc Dis 2008; 25: 457–507. Jauch EC, Saver JL, Adams HP Jr, et al. Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/ American Stroke Association. Stroke 2013; 44: 870–947. Walter S, Kostopoulos P, Haass A, et al. Bringing the hospital to the patient: first treatment of stroke patients at the emergency site. PLoS One 2010; 5: e13758. Balucani C, Levine SR. The “almost magical” mobile stroke unit revolution. Neurology 2012; 78: 1809–10. Yong E. First response: race against time. Nature 2014; 510: S5. Ragoschke-Schumm A, Walter S, Haass A, et al. Translation of the ‘time is brain’ concept into clinical practice: focus on prehospital stroke management. Int J Stroke 2014; 9: 333–40. Ashkenazi L, Toledano R, Novack V, Elluz E, Abu-Salamae I, Ifergane G. Emergency department companions of stroke patients: implications on quality of care. Medicine (Baltimore) 2015; 94: e520. Liman TG, Winter B, Waldschmidt C, et al. Telestroke ambulances in prehospital stroke management: concept and pilot feasibility study. Stroke 2012; 43: 2086–90. Van Hooff RJ, Cambron M, Van Dyck R, et al. Prehospital unassisted assessment of stroke severity using telemedicine: a feasibility study. Stroke 2013; 44: 2907–09. Wu TC, Nguyen C, Ankrom C, et al. Prehospital utility of rapid stroke evaluation using in-ambulance telemedicine: a pilot feasibility study. Stroke 2014; 45: 2342–47. Eadie L, Regan L, Mort A, et al. Telestroke assessment on the move: prehospital streamlining of patient pathways. Stroke 2015; 46: e38–40. Bergrath S, Reich A, Rossaint R, et al. Feasibility of prehospital teleconsultation in acute stroke—a pilot study in clinical routine. PLoS One 2012; 7: e36796. Chapman Smith SN, Govindarajan P, Padrick MM, et al. A low-cost, tablet-based option for prehospital neurologic assessment: the iTREAT Study. Neurology 2016; 87: 19–26. Ziegler V, Rashid A, Müller-Gorchs M, et al. Mobile computing systems in preclinical care of stroke. Results of the Stroke Angel initiative within the BMBF project PerCoMed. Anaesthesist 2008; 57: 677–85 (in German). Gonzalez MA, Hanna N, Rodrigo ME, Satler LF, Waksman R. Reliability of prehospital real-time cellular video phone in assessing the simplified National Institutes Of Health Stroke Scale in patients with acute stroke: a novel telemedicine technology. Stroke 2011; 42: 1522–27. Kostopoulos P, Walter S, Haass A, et al. Mobile stroke unit for diagnosis-based triage of persons with suspected stroke. Neurology 2012; 78: 1849–52.

235

Review

32 33 34

35 36 37 38 39

42 43 44 45 46 47 48 49

50 51 52 53

236

Itrat A, Taqui A, Cerejo R, et al. Telemedicine in prehospital stroke evaluation and thrombolysis: taking stroke treatment to the doorstep. JAMA Neurol 2016; 73: 162–68. Wu TC, Parker S, Jagolino A, et al. Telemedicine can replace the neurologist on a mobile stroke unit. Stroke 2017; published online Jan 12. DOI:10.1161/STROKEAHA.116.015363. Ramadan AR, Denny MC, Vahidy F, et al. Agreement among stroke faculty and fellows in treating ischemic stroke patients with tissue-type plasminogen activator and thrombectomy. Stroke 2017; 48: 222–24. Walter S, Kostopoulos P, Haass A, et al. Diagnosis and treatment of patients with stroke in a mobile stroke unit versus in hospital: a randomised controlled trial. Lancet Neurol 2012; 11: 397–404. Parker SA, Bowry R, Wu TC, et al. Establishing the first mobile stroke unit in the United States. Stroke 2015; 46: 1384–91. Ebinger M, Winter B, Wendt M, et al. Effect of the use of ambulance-based thrombolysis on time to thrombolysis in acute ischemic stroke: a randomized clinical trial. JAMA 2014; 311: 1622–31. Hov MR, Nome T, Zakariassen E, et al. Assessment of acute stroke cerebral CT examinations by anaesthesiologists. Acta Anaesthesiol Scand 2015; 59: 1179–86. Powers WJ, Derdeyn CP, Biller J, et al. 2015 American Heart Association/American Stroke Association focused update of the 2013 guidelines for the early management of patients with acute ischemic stroke regarding endovascular treatment: a guideline for healthcare professionals from the American Heart Association/ American Stroke Association. Stroke 2015; 46: 3020–35. Demchuk AM, Dowlatshahi D, Rodriguez-Luna D, et al. Prediction of haematoma growth and outcome in patients with intracerebral haemorrhage using the CT-angiography spot sign (PREDICT): a prospective observational study. Lancet Neurol 2012; 11: 307–14. Grunwald IQ, Ragoschke-Schumm A, Kettner M, et al. First automated stroke imaging evaluation via electronic Alberta Stroke Program Early CT Score in a mobile stroke unit. Cerebrovasc Dis 2016; 42: 332–38. Walter S, Kostopoulos P, Haass A, et al. Point-of-care laboratory halves door-to-therapy-decision time in acute stroke. Ann Neurol 2011; 69: 581–86. Puolakka T, Strbian D, Harve H, Kuisma M, Lindsberg PJ. Prehospital phase of the stroke chain of survival: a prospective observational study. J Am Heart Assoc 2016; 5: e002808. Fassbender K, Balucani C, Walter S, Levine SR, Haass A, Grotta J. Streamlining of prehospital stroke management: the golden hour. Lancet Neurol 2013; 12: 585–96. Oostema JA, Carle T, Talia N, Reeves M. Dispatcher stroke recognition using a stroke screening tool: a systematic review. Cerebrovasc Dis 2016; 42: 370–77. Brandler ES, Sharma M, Sinert RH, Levine SR. Prehospital stroke scales in urban environments: a systematic review. Neurology 2014; 82: 2241–49. Krebes S, Ebinger M, Baumann AM, et al. Development and validation of a dispatcher identification algorithm for stroke emergencies. Stroke 2012; 43: 776–81. Weber JE, Ebinger M, Rozanski M, et al. Prehospital thrombolysis in acute stroke: results of the PHANTOM-S pilot study. Neurology 2013; 80: 163–68. Ebinger M, Kunz A, Wendt M, et al. Effects of golden hour thrombolysis: a prehospital acute neurological treatment and optimization of medical care in stroke (PHANTOM-S) substudy. JAMA Neurol 2015; 72: 25–30. Bowry R, Parker S, Rajan SS, et al. Benefits of stroke treatment using a mobile stroke unit compared with standard management: the BEST-MSU study run-in phase. Stroke 2015; 46: 3370–74. Taqui A, Cerejo R, Itrat A, et al. Reduction in time to imaging and intravenous thrombolysis by in-field evaluation and treatment in a mobile stroke treatment unit. Stroke 2015; 46: A54. Cerejo R, John S, Buletko AB, et al. A mobile stroke treatment unit for field triage of patients for intraarterial revascularization therapy. J Neuroimaging 2015; 25: 940–45. Kunz A, Ebinger M, Geisler F, et al. Functional outcomes of prehospital thrombolysis in a mobile stroke treatment unit compared with conventional care: an observational registry study. Lancet Neurol 2016; 15: 1035–43.

55 56 57 58

59 60 61 62 63

64 65 66

68 69 70 71

72 73 74 75 76

Wahlgren N, Ahmed N, Davalos A, et al. Thrombolysis with alteplase for acute ischaemic stroke in the safe implementation of thrombolysis in stroke-monitoring study (SITS-MOST): an observational study. Lancet 2007; 369: 275–82. Mazighi M, Chaudhry SA, Ribo M, et al. Impact of onset-toreperfusion time on stroke mortality: a collaborative pooled analysis. Circulation 2013; 127: 1980–85. The National Institute of Neurological Disorders and Stroke rt–PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 1995; 333: 1581–87. Wahlgren N, Ahmed N, Davalos A, et al. Thrombolysis with alteplase 3–4·5 h after acute ischaemic stroke (SITS–ISTR): an observational study. Lancet 2008; 372: 1303–09. Kim JT, Fonarow GC, Smith EE, et al. Treatment with TPA in the “golden hour” and the shape of the 4.5 hour time–benefit curve in the National US Get With The Guidelines-Stroke population. Circulation 2017; 135: 128–39. Meretoja A, Strbian D, Mustanoja S, Tatlisumak T, Lindsberg PJ, Kaste M. Reducing in-hospital delay to 20 minutes in stroke thrombolysis. Neurology 2012; 79: 306–13. Grotta JC. tPA for stroke: important progress in achieving faster treatment. JAMA 2014; 311: 1615–17. Rajan SS, Baraniuk S, Parker S, Wu TC, Bowry R, Grotta JC. Implementing a mobile stroke unit program in the United States: why, how, and how much? JAMA Neurol 2015; 72: 229–34. Anderson CS, Heeley E, Huang Y, et al. Rapid blood-pressure lowering in patients with acute intracerebral hemorrhage. N Engl J Med 2013; 368: 2355–65. Hemphill JC 3rd, Greenberg SM, Anderson CS, et al. Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2015; 46: 2032–60. Steiner T, Rosand J, Diringer M. Intracerebral hemorrhage associated with oral anticoagulant therapy: current practices and unresolved questions. Stroke 2006; 37: 256–62. Gomes JA, Ahrens CL, Hussain MS, et al. Prehospital reversal of warfarin-related coagulopathy in intracerebral hemorrhage in a mobile stroke treatment unit. Stroke 2015; 46: e118–20. Goyal M, Almekhlafi MA, Fan L, et al. Evaluation of interval times from onset to reperfusion in patients undergoing endovascular therapy in the Interventional Management of Stroke III trial. Circulation 2014; 130: 265–72. Menon BK, Sajobi TT, Zhang Y, et al. Analysis of workflow and time to treatment on thrombectomy outcome in the endovascular treatment for small core and proximal occlusion ischemic stroke (ESCAPE) randomized, controlled trial. Circulation 2016; 133: 2279–86. Prabhakaran S, Ward E, John S, et al. Transfer delay is a major factor limiting the use of intra-arterial treatment in acute ischemic stroke. Stroke 2011; 42: 1626–30. Grotta JC, Savitz SI, Persse D. Stroke severity as well as time should determine stroke patient triage. Stroke 2013; 44: 555–57. Nazliel B, Starkman S, Liebeskind DS, et al. A brief prehospital stroke severity scale identifies ischemic stroke patients harboring persisting large arterial occlusions. Stroke 2008; 39: 2264–67. Perez de la Ossa N, Carrera D, Gorchs M, et al. Design and validation of a prehospital stroke scale to predict large arterial occlusion: the rapid arterial occlusion evaluation scale. Stroke 2014; 45: 87–91. Lima FO, Silva GS, Furie KL, et al. Field assessment stroke triage for emergency destination: a simple and accurate prehospital scale to detect large vessel occlusion strokes. Stroke 2016; 47: 1997–2002. Hastrup S, Damgaard D, Johnsen SP, Andersen G. Prehospital acute stroke severity scale to predict large artery occlusion: design and comparison with other scales. Stroke 2016; 47: 1772–76. Heldner MR, Hsieh K, Broeg-Morvay A, et al. Clinical prediction of large vessel occlusion in anterior circulation stroke: mission impossible? J Neurol 2016; 263: 1633–40. Turc G, Maier B, Naggara O, et al. Clinical scales do not reliably identify acute ischemic stroke patients with large-artery occlusion. Stroke 2016; 47: 1466–72. Wendt M, Ebinger M, Kunz A, et al. Improved prehospital triage of patients with stroke in a specialized stroke ambulance: results of the pre-hospital acute neurological therapy and optimization of medical care in stroke study. Stroke 2015; 46: 740–45.

www.thelancet.com/neurology Vol 16 March 2017

119

Review

77 78 79 80

Schwindling L, Ragoschke-Schumm A, Kettner M, et al. Prehospital imaging-based triage of head trauma with a mobile stroke unit: first evidence and literature review. J Neuroimaging 2016; 26: 489–93. Dietrich M, Walter S, Ragoschke-Schumm A, et al. Is prehospital treatment of acute stroke too expensive? An economic evaluation based on the first trial. Cerebrovasc Dis 2014; 38: 457–63. Leira EC, Hess DC, Torner JC, Adams HP Jr. Rural–urban differences in acute stroke management practices: a modifiable disparity. Arch Neurol 2008; 65: 887–91. Shuaib A, Khan K, Whittaker T, Amlani S, Crumley P. Introduction of portable computed tomography scanners, in the treatment of acute stroke patients via telemedicine in remote communities. Int J Stroke 2010; 5: 62–66. Gyrd-Hansen D, Olsen KR, Bollweg K, Kronborg C, Ebinger M, Audebert HJ. Cost-effectiveness estimate of prehospital thrombolysis: results of the PHANTOM-S study. Neurology 2015; 84: 1090–97. Dvorak F, Haberer I, Sitzer M, Foerch C. Characterisation of the diagnostic window of serum glial fibrillary acidic protein for the differentiation of intracerebral haemorrhage and ischaemic stroke. Cerebrovasc Dis 2009; 27: 37–41.

www.thelancet.com/neurology Vol 16 March 2017

120

84 85 86 87 88

Schlachetzki F, Herzberg M, Holscher T, et al. Transcranial ultrasound from diagnosis to early stroke treatment: part 2: prehospital neurosonography in patients with acute stroke: the Regensburg stroke mobile project. Cerebrovasc Dis 2012; 33: 262–71. Barlinn K, Tsivgoulis G, Barreto AD, et al. Outcomes following sonothrombolysis in severe acute ischemic stroke: subgroup analysis of the CLOTBUST trial. Int J Stroke 2014; 9: 1006–10. Saver JL, Starkman S, Eckstein M, et al. Prehospital use of magnesium sulfate as neuroprotection in acute stroke. N Engl J Med 2015; 372: 528–36. Mayer SA, Brun NC, Begtrup K, et al. Efficacy and safety of recombinant activated factor VII for acute intracerebral hemorrhage. N Engl J Med 2008; 358: 2127–37. Burges Watson DL, Sanoff R, Mackintosh JE, et al. Evidence from the scene: paramedic perspectives on involvement in out-of-hospital research. Ann Emerg Med 2012; 60: 641–50. Middleton S, Grimley R, Alexandrov AW. Triage, treatment, and transfer: evidence-based clinical practice recommendations and models of nursing care for the first 72 hours of admission to hospital for acute stroke. Stroke 2015; 46: e18–25.

237

Consultoría integral, la mejor capacitación y entrenamiento para el profesional de la salud y el servicio postventa más completo.

D E LE C C I E NTÍ FI CA U R U G UAYA

DELEC CIENTÍFICA ARGENTINA

Fco. García Cor na 2357 - Piso 1 Montevideo, Uruguay. Tel. (+598) 2711 4466 Cel. (+598) (0) 93 507 500 delecuruguay@delec.com.uy

Aráoz 821 - C1414DPQ C.A.B.A - Argen na Tel. (54-11) 4775 - 5844 Fax. (54-11) 4803 - 2505 consultas@delec.com.ar

www.delec.com.uy

www.delec.com.ar

DELEC CIENTÍFICA

CereTom - Carpeta de Producto

Articles inside

ÁREAS DE APLICACIÓN

CARACTERÍSTICAS PRINCIPALES