McMaster University Medical Journal, Vol. 17, Issue 1 by MUMJ

McMASTER UNIVERSITY MEDICAL JOURNAL VOLUME 17, ISSUE 1 2020

Original Research

A GEOSPATIAL ANALYSIS OF PERTUSSIS AND ITS RISK FACTORS IN SOUTHERN ONTARIO FROM 2005-2016

Reviews

UNIVERSAL VACCINES AGAINST INFLUENZA VIRUSES: OVERVIEW OF THE PAST, PRESENT, AND PROSPECTIVE

Commentary

THE NATURAL HISTORY OF MEDICAL WASTE

__________________________________________________________ McMaster University Medical Journal Editorial Board 2019-20

__________________________________________________________ Editors-in-Chief AADIL BHARWANI AND SARAH PETERS Executive Editors LAURA LOCKAU AND ALI ZHANG Submission Editors JUNE DONG SIMON FARQUHARSON JASPER HO JESSICA JUNG SARAH KIMBER MEGAN LAM

BRUCE LI APRIL LIU EVA LIU JUSTIN LU SIDDHARTH NATH RYAN Oâ&#x20AC;&#x2122;REILLY

VIVEK PATEL SHERIF RAMADAN MYLINI SAPOSAN KENT TANG CAITLIN YEE JENNY ZHU

Reviewers TAKHLIQ AMIR CYNTHIA CHAN JESSICA CHEE CLARELLE GONSALVES JENSINE GRONDIN YAN JENNIFER GU ALYSON HOLLAND KAITLIN LEWIS

BRUCE LI LAUREN LIN IBRAHIM NADEEM KEEAN NANJI JANE NEWMAN DAVID NGUYEN NIVEDH PATRO JAMES PODREBARAC

SHERIF RAMADAN SOPHIE RAMSDEN KEVEN REN KENT TANG MILICA TANIC MURALIE VIGNARAJAH EUGENE WANG

Cover Art CORA MARINOFF

TABLE OF CONTENTS LETTER FROM THE EDITORS…………………………………………………………………v AADIL BHARWANI AND SARAH PETERS

ORIGINAL RESEARCH ARTICLES POSTOPERATIVE PAIN MANAGEMENT EDUCATION DURING THE SURGERY CORE ROTATION AT MCMASTER UNIVERSITY, WATERLOO REGIONAL CAMPUS……...…1 NIVEDH PATRO AND GRAHAM CAMPBELL A GEOSPATIAL ANALYSIS OF PERTUSSIS AND ITS RISK FACTORS IN SOUTHERN ONTARIO FROM 2005-2016…………………………………………………………………...16 TAHA ELGHAMUDI AND OLAF BERKE A SYSTEMATIC REVIEW OF CLINICAL DECISION TOOLS USED FOR DIAGNOSING PULMONARY EMBOLISM IN THE PEDIATRIC POPULATION…………………………..28 YICHEN (APRIL) LIU, LAURA NGUYEN, MOHAMMED HASSAN-ALI, APRIL KAM

CASE REPORTS ATYPICAL EUTHYROID PRESENTATION OF STEROID RESPONSIVE ENCEPHALOPATHY WITH ASSOCIATED THYROIDITIS………………………………...50 VINEETH BHOGADI AND ARLENE KELLY-WIGGINS STATIN-INDUCED RHABDOMYOLYSIS: A CAUTIONARY TALE FOR HIGH-DOSE ROSUVASTATIN……………………………………………………………………………….57 HENRY HE AND GULSHAN ATWAL INDOLENT T-CELL LYMPHOPROLIFERATIVE DISORDER OF THE GASTROINTESTINAL TRACT MANAGED CONSERVATIVELY WITH CORTICOSTEROIDS: A CASE REPORT……………………………………………………..63 TYLER MCKECHNIE, HAROON YOUSUF, STEPHEN SOMERTON

iii

REVIEW ARTICLES UNIVERSAL VACCINES AGAINST INFLUENZA VIRUSES: OVERVIEW OF THE PAST, PRESENT, AND PROSPECTIVE………………………………………………………………71 YONATHAN AGUNG, HANNAH STACEY, MICHAEL D’AGOSTINO, ALI ZHANG LATERAL VS. SUPINE POSITIONING FOR FEMORAL INTRAMEDULLARY NAILING: A SYSTEMATIC REVIEW OF COMPARATIVE STUDIES…………………………………89 MOHAMED SARRAJ, DANIEL E. AXELROD, SARAH ZHU, HERMAN JOHAL THE CRITICAL ROLE OF ASTROGENESIS AND NEURODEVELOPMENT IN FRAGILE X SYNDROME AND RETT SYNDROME………………………………………………...…103 ZHUO JUN LI AND JOHN-PAUL OLIVERIA

COMMENTARIES FLOAT OR SINK: A SOLUTION TO THE RESIDENT BURN-OUT CRISIS?.....................122 CANDICE LUO, GEORGE HU, TONY CHEN CETERIS PARIBUS? – AN EPISTEMOLOGICAL ERROR WITH ETHICAL CONSEQUENCES……………………………………………………………………………..128 CHRIS ARSENAULT THE NATURAL HISTORY OF MEDICAL WASTE……………………………………...…134 YU FEI XIA AND BETTY HUI YU ZHANG

From the Editors

McMaster University Medical Journal, Volume 17 Sarah Peters, MD, MSc, and Aadil Bharwani, PhD, MD candidate Editors-in-Chief Michael G. DeGroote School of Medicine 17 years after its inception, the McMaster University Medical Journal is delighted to continue its proud tradition of showcasing advances in medicine and science that are driven by medical and graduate students from across Canada. Through this humble endeavour, we strive to provide a platform to disseminate the rapid and ever-changing nuances of medical practice, to explore the oft-overlooked social issues that influence and determine health at the level of the person and population, and to create space for a compelling discourse of the legal and ethical questions that pervade our profession. Akin to its predecessors, this 17th volume comprises work that spans the bench, the bedside, public health, and all that lies in between. Our authors describe a unique presentation of encephalopathy and statin-induced rhabdomyolysis, examine the clinical decision rules for paediatric pulmonary emboli, and outline the management of a rare T-cell disorder as well as patient positioning for femoral intramedullary nailing. They scrutinize the resident burnout crisis and medical education for postoperative pain management. They narrate the endeavours for a universal influenza vaccine and study the epidemiology of pertussis in Southern Ontario. They challenge the current practices of medical waste management and the ethical considerations of prescribing placebo in a recent randomized controlled trial. As ever, we remain grateful to the tireless efforts of our executive editors, submission editors, and reviewers, all of whom have generously contributed their valuable time and expertise to bring this issue to fruition. Whilst these efforts comprise a culture that has remained consistent and true across the young life of this journal, we are prouder than ever this year, which has witnessed significant disruptions to the lives and education of medical students amidst a global pandemic. We would also like to thank the authors for their scientific efforts and the opportunity to showcase their work. You provide an important reminder and means to incessantly reflect upon new ideas and challenges and discovery. Finally: thank you, dear reader, for it is only with your support that this journal can continue to strive and exist. Sincerely, Sarah Peters and Aadil Bharwani

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Original Research Article

Postoperative pain management education during the surgery core rotation at McMaster University, Waterloo Regional Campus Nivedh Patro1 and Graham Campbell2 Michael G. DeGroote School of Medicine, McMaster University, Waterloo Regional Campus, Kitchener/Waterloo, Canada 1

WRC Research Department, McMaster University, Waterloo Regional Campus, Kitchener/Waterloo, Canada 2

Abstract Background: Opioid over-prescription continues to be a challenge in the postoperative setting for management of acute pain. Initiatives have been developed to standardize postoperative opioid prescribing with an emphasis on multimodal pain management. However, there is a concern medical education has not remained current on this topic. Objective: The aim of this preliminary study is to explore current teaching around postoperative pain management during the surgery core rotation at McMaster University, Waterloo Regional Campus (WRC), and identify any opportunities for improvement. Methods: A 13-item survey was developed to determine effectiveness of teaching around postoperative pain management during the surgery core and its alignment with current guidelines. The survey was disseminated to third year medical students at the WRC. Results: Seven of nine respondents indicated that teaching on postoperative pain management and opioid reduction strategies was provided during the surgery core. All respondents receiving this teaching also indicated learning about a multimodal pain control approach consistent with current guidelines. However, only three of seven respondents noted receiving teaching on providing patient and caregiver education around the pain management plan, despite a strong recommendation in guidelines in favour of this practice. Conclusions: Most students receive teaching on multimodal postoperative pain management and opioid reduction strategies during the surgery core at the WRC. Opportunities to strengthen the teaching include addressing the role of patient and caregiver education in the pain management plan as well as incorporating the topic into formal teaching such as classroom sessions or learning objectives in the surgery core. Keywords: Postoperative, Opioids, Multimodal pain management, Medical education, Surgery

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Corresponding author: nivedh.patro@medportal.ca

Introduction One of the biggest concerns facing the Canadian healthcare system over the last decade has been the rise in opioid use and abuse, with the issue reaching epidemic levels in 2016. Between 2016 to 2018 alone, the number of opioid-related deaths in Canada rose from 3023 to 4588 (1). A major source of opioids is through prescription, with Canada having the second-highest per capita consumption of prescription opioids in the world after the United States (2). Postoperative pain management represents one of the key reasons for opioid prescription. One study found that among 653 993 new prescription opioid users in Ontario over the span of a year, one in six received opioid prescriptions post-surgically (3). Many of these prescriptions were for durations and doses that exceeded the maximums suggested by guidelines (3). Considering that about one in sixteen opioid-naĂŻve patients prescribed opioids post-surgically become long term users (4), and that the risk of misuse increases by 34% per additional week of use (3), postoperative opioid prescriptions play a substantial role in opioid-related morbidity and mortality events (3â&#x20AC;&#x201C;5). Recent initiatives have aimed to standardize postoperative opioid prescription protocols based on available guidelines in order to minimize the number of extra or unused prescription opioid pills in circulation and reduce the propensity for abuse (5,6). A component of these initiatives is to promote the use of a multimodal pain management approach involving a variety of pharmacological and non-pharmacological treatments that minimize the reliance on opioids (5,6). The multimodal approach, including options like non-steroidal anti-inflammatory drugs (NSAIDs), acetaminophen, and gabapentin, has been found to achieve better overall pain management than opioids alone (5,7,8). The proposed mechanism for this effect is the simultaneous targeting of multiple pathways involved in pain perception, allowing for improved pain control with the use of fewer, if any, opioids (5,7). As a result, guidelines such as those from the American Pain Society (APS) have recommended the use of acetaminophen and/or NSAIDs in the management of postoperative pain barring any contradictions, use of pregabalin or gabapentin as part of a multimodal approach, and education of patients and caregivers on the pain management plan including tapering of analgesics (7,8). Although initiatives to reduce postoperative opioid prescription exist in the hospital setting, few studies investigate efforts at the level of medical education in Canada to instill awareness of this subject in future physicians early in their training. In 2017, a report in response to the opioid crisis by the Association of Faculties of Medicine of Canada (AFMC) called for teaching on opioids and addiction to occur across all stages of medical training, including preclerkship, clerkship, residency, and Continuing Professional Development programs (9). That same year, a study of the pain curricula in three Ontario medical schools found that the schools varied considerably in the organization of content and number of hours of training on this topic

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

(10). This study also found that the curricula across all three schools focused predominantly on the role of opioids in pain management despite acknowledging the importance of a multimodal approach (10). At McMaster University, teaching on opioids occurs at various points in the current UGME curriculum database published online (11), often in the contexts of pharmacology, chronic pain management, and substance use. During clerkship, one of the learning objectives during the anesthesia core rotation specifies the multimodal use of analgesics during the perioperative period. However, no such objective appears within the surgery core curriculum, even though it is often the surgeon who writes postoperative patient orders. The surgery rotation therefore provides a key opportunity to deliver this teaching as learners are directly involved in the care of the surgical patient, including development of postoperative management plans. The role of the present preliminary study is to explore any opportunities for improvement in the teaching of postoperative pain management during the surgery core rotation at McMaster University given the substantial contribution of postoperative opioids to opioid misuse and a concern that medical education has not remained current on this topic (10,12). The Waterloo Regional Campus, a satellite site of McMaster University and the authors’ home campus, was selected as the study focus. Methods A survey was developed using Google Forms with four broad guiding questions in mind: 1) Is teaching being provided on multimodal postoperative pain management and opioid reduction strategies during the surgery core rotation; 2) Is the teaching in alignment with guideline recommendations on postoperative pain management; 3) Do students find that this teaching is effective and highlights the importance of a multimodal approach; and 4) Is teaching on this subject provided during other points in clerkship and pre-clerkship? The APS guidelines on management of postoperative pain were consulted (7) along with a Health Quality Ontario (HQO) report on opioid prescribing for acute pain (8) in order to develop survey questions concerning multimodal pain management. A 5-point Likert scale was used for questions requiring a ranking from respondents (e.g. “Strongly Agree” to “Strongly Disagree”). The survey was reviewed by two colleagues as well as the WRC Research Office after its development to ensure relevance and clarity of the questions posed. Following its development, the survey was disseminated to the third-year medical class at the WRC. Third-year students were the chosen sample population as they would have completed the most clinical rotations at the time of the survey and had the greatest likelihood of completing their surgery core rotation. The survey was sent out to participants in November 2019 and was closed in January 2020. The full survey can be found in Appendix 1.

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Results

Figure 1. Core rotations completed by respondents at the time of the survey. The survey had a 32% response rate with a total of nine responses collected from 28 students in the third-year class. Although only eight of nine students responded as having completed their surgery core rotation at the time of the survey (Figure 1), the remaining student responded as having received teaching on the postoperative pain management strategies during their surgery core rotation and completed the section specific to the surgery core rotation. Since the student would not have been able to complete this section unless they had in fact completed their surgery core, it was presumed that all respondents had completed their surgery core at the time of the survey. Seven students responded that they had received teaching around multimodal pain management and opioid reduction strategies in the postoperative setting during the surgery core rotation, while two students responded that they had not (Figure 2). Eight respondents received teaching on multimodal postoperative pain management in a core rotation other than surgery. The one respondent that did not indicate receiving this teaching during another core rotation did indicate receiving the teaching during the surgery core. All respondents had therefore received this teaching in at least one core rotation during clerkship. Two thirds of participants indicated receiving teaching on the topic of multimodal postoperative pain management and opioid reduction during pre-clerkship. Six respondents indicated they had received this teaching during the anesthesia core rotation. Two respondents indicated receiving this teaching during the family medicine and internal medicine core rotations. One respondent indicated they had also received this teaching during the orthopedic surgery and obstetrics/gynecology core rotations.

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Figure 2. Respondents were asked if they had received teaching on multimodal postoperative pain management and opioid reduction strategies during (A) the surgery core, (B) core rotations other than surgery, (C) pre-clerkship, and (D) specific core rotations other than surgery.

Figure 3. Respondentsâ&#x20AC;&#x2122; (A) comfort with current knowledge of multimodal postoperative pain management options, and (B) perceived benefit of additional teaching on the topic.

Five students were neutral, three students were unconfident, and one student was confident in their knowledge of multimodal postoperative pain control options (Figure 3). Seven respondents felt they would benefit from additional teaching on this topic, while two respondents were neutral to this prospect. No respondents felt that there would be no benefit to additional teaching.

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Figure 4. Setting in which teaching on postoperative pain management was provided during the surgery core.

The environment in which the teaching was provided was evenly split between a formal and focused setting such as a classroom and an informal setting such as in passing while on service (Figure 4). One respondent indicated they had received this teaching in both settings.

Figure 5. Effectiveness of teaching in highlighting importance around multimodal pain management and opioid reduction in the postoperative setting. Four respondents agreed that the teaching effectively highlighted the importance of multimodal pain management and opioid-reduced postoperative prescriptions, while one respondent strongly agreed, another was neutral, and one disagreed (Figure 5). 6

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Figure 6. Respondents were asked (A) what specific non-opioid pain management modalities were discussed during their surgery core, and (B) whether they had received teaching on providing patient and caregiver education around the pain management plan. NSAIDs = non-steroidal anti-inflammatory drugs.

All respondents were introduced to NSAIDs (naproxen, ibuprofen, celecoxib, or ketorolac) and acetaminophen as part of a multimodal pain management strategy (Figure 6). The next most discussed pain management modalities were pregabalin or gabapentin as well as nonpharmacological interventions including heat, ice, massage, stretching, rest, acupuncture, bracing or wrapping, spinal manipulation, passive physical therapy, positioning, splints, transcutaneous electrical nerve stimulation. Six respondents each indicated discussing these modalities during their surgery core rotation. The third most discussed category was psychological interventions including cognitive behavioural therapy, guided imagery, and other relaxation techniques. The single other modality was specified as â&#x20AC;&#x153;sleep hygiene.â&#x20AC;? The majority of respondents did not receive teaching around the impact of patient and caregiver education on the pain management plan. Only one respondent indicated that a multimodal postoperative pain management approach was always employed by their attendings or residents during the rotation (Figure 7). The vast majority indicated that such an approach was only used sometimes. The majority of respondents also indicated their attendings or residents practiced patient education around the pain management plan only rarely or sometimes, with only about of third saying this was practiced often.

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Figure 7. Respondents were asked if they noticed attending staff or residents (A) practicing a multimodal approach to postoperative pain management, and (B) providing patients and caregivers education around the pain management plan.

Discussion A total of nine survey responses were collected from third year medical students at the WRC. Seven of nine respondents indicated that teaching on postoperative pain management and opioid reduction strategies was provided during the surgery core rotation (Figure 2A). Of the seven respondents, three indicated they had received this training in a formal and focused setting such as a classroom or online module (Figure 4). One reported receiving the teaching both formally and informally, such as from an attending staff or resident while on service (Figure 4). These findings are promising as they demonstrate that most students from the sample have received teaching on the topic during the surgery core. However, given that about half of those students received the teaching in informal settings or discussions, this would suggest that a greater part of the onus shifts to attending staff or residents if this teaching is to be provided. Considering the often-hectic nature of a surgical service, it is possible that attendings may not get around to teaching this specific topic and some students may not receive this teaching at all while on service. This finding points to an opportunity to incorporate this topic into lectures or learning objectives in order to standardize and ensure all students going through their core rotation are at least introduced to the subject. Another aim of the survey was to determine whether the teaching is aligned with current guidelines on postoperative pain management, particularly referencing the guidelines published by the APS in conjunction with other Societies in 2016. All respondents that had received teaching on postoperative pain management during the surgery core indicated they had learned about the role of NSAIDs and acetaminophen in pain management (Figure 6A), which is promising given that the effectiveness of these agents is supported by high-quality evidence (5,7,13). In some cases, effective pain control postoperatively can be achieved using these agents alone without a need to fill opioid prescriptions (5). Six of seven respondents also indicated

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

learning about the use of pregabalin or gabapentin as part of postoperative pain control (Figure 6A), again in line with the guideline recommendations (7,8). Contrary to guideline recommendations, most respondents did not receive teaching around the impact of patient and caregiver education on the pain management plan (Figure 6B). This finding might indicate that the importance of this component of pain management is not sufficiently recognized. The guidelines particularly highlight the importance of this component in informing patients how to safely use their analgesics and optimize pain control, manage side effects, prevent combinations that can result in overdose or death, and explain appropriate disposal of unused supplies of opioids and other medications (7). Inappropriate disposal has been recognized as a concerning source for opioid diversion, with one US survey estimating that 65% of those with an opioid misuse disorder obtain their opioids from a non-prescription source (14). Therefore, incorporating this guideline recommendation as part of teaching on postoperative pain management is a worthwhile endeavour. Respondents also indicated learning about other modalities of pain management including non-pharmacologic interventions as well as psychological interventions (Figure 6A). One respondent also indicated learning about the role of sleep hygiene in pain management. The APS guidelines only provided weak recommendations in favour of transcutaneous electrical nerve stimulation (TENS) and cognitive-behavioural modalities, and could not provide a recommendation for other adjunctive interventions such as acupuncture, massage, or cold therapy due to a lack of supporting data (7). Nevertheless, these therapies may still play an important adjunctive role particularly for patients who cannot tolerate or prefer not to use certain pharmacologic treatments and it is encouraging that students are learning about them. Another aim of the survey was to determine the effectiveness of the teaching in highlighting the importance of a multimodal postoperative pain management approach with reduced opioid prescriptions, as expressed by respondents. Four respondents agreed that the teaching effectively highlighted the importance of this approach, while one respondent strongly agreed, another was neutral, and one disagreed (Figure 5). It is encouraging that more than half of the respondents felt the teaching was effective in this regard. This finding might indicate that when the teaching is provided, it is impactful for the most part. However, there may still be some room for improvement of the content considering the finding that three out of nine respondents indicated they were not confident with their knowledge of multimodal postoperative pain management options. A potential factor that could impede effectiveness of the teaching is the finding that attending staff or residents were not consistent with applying multimodal postoperative pain management and opioid reduction strategies themselves (Figure 7A). Only one respondent indicated that a multimodal approach was always employed by their attendings or residents during the rotation. The vast majority indicated that such an approach was only used sometimes. The majority of respondents also indicated their attendings or residents practiced patient education around the pain management plan only rarely or sometimes, with only about a third saying this was practiced often (Figure 7B). The inconsistent application of the multimodal

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

approach by role models in the clinical setting might undermine the importance of this approach and decrease effectiveness around teaching of this topic. This finding could point to the presence of potential barriers to consistently applying the multimodal approach in actual clinical settings, which may merit study on its own and how those factors can be addressed in teaching on this topic. The survey also explored whether teaching on multimodal postoperative pain management was provided during other clerkship core rotations (Figure 2D). Not surprisingly, the majority of respondents indicated they had received this teaching during the anesthesia core rotation. Two of nine participants also indicated receiving this teaching during Family and Internal Medicine core rotations. Given that these specialties are often involved in managing surgical patients, it is important that these specialties are also informed on good practices in postoperative pain management and opioid reduction. Interestingly, only one respondent indicated receiving this teaching during the orthopedic core rotation and the obstetrics/gynecology core, despite seven participants completing each of these core rotations. This finding points to an opportunity to reiterate the importance and strategies around responsible postoperative pain management in these surgical subspecialty rotations. There is also an opportunity for introducing this teaching in other core rotations such as emergency medicine, given that this specialty is also regularly involved in the care of surgical patients. Given that the AFMCâ&#x20AC;&#x2122;s response to the opioid crisis called for teaching on opioids and addiction to occur across all stages of medical training (9), respondents were also asked whether they received this teaching during pre-clerkship. Two thirds of respondents indicated that they had (Figure 2C). It is encouraging that most students are learning about this topic prior to entering clerkship. However, there is still an opportunity to ensure most if not all students complete pre-clerkship with some level of exposure and teaching on this topic, as per the AFMC recommendations. This goal could be achieved for example by including this topic as a learning objective during small group case-based sessions and having group facilitators ensure it is covered as part of a given session. This study is not without its limitations. Given the relatively limited number of responses received, the data may be subject to a voluntary response bias. The degree to which findings can be generalized to the entire WRC class is also uncertain. Additionally, further questions exploring reasons behind responses were omitted for the sake of maintaining brevity of the survey. These questions might include asking why respondents did or did not feel confident in their knowledge of postoperative pain management options, or what if any barriers appeared to prevent consistent application of a multimodal approach by attendings during the core rotation. Such insights would be helpful in determining how current teaching can be enhanced and how barriers can be navigated. The next steps for this project include disseminating the survey in the McMaster Hamilton and Niagara campuses as well to have a better picture of postoperative pain management teaching within the institution as a whole. Inter-campus variability can also be explored if any exist.

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Conclusion The majority of students appear to receive teaching on multimodal postoperative pain management and opioid reduction strategies during the surgery core at the WRC. This teaching is mostly aligned with current guidelines on the topic. However, the role of patient and caregiver education on the pain management plan can be better addressed as this is a strongly recommended component of postoperative pain management. Other avenues for reinforcing teaching on this topic include incorporating the subject into formal classroom sessions or objectives during the surgery core so that all students have a chance to learn about this topic without reliance on the variable nature of teaching while on service. The teaching can also be incorporated into other surgical core rotations including orthopedics and obstetrics/gynecology to reinforce concepts of effective and responsible postoperative pain management.

Conflict of Interest The authors have no conflict of interest to disclose.

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

References 1.

National Report: Apparent Opioid-related Deaths in Canada [Internet]. 2019 [cited 2019 Oct 9]. Available from: https://health-infobase.canada.ca/datalab/national-surveillanceopioid-mortality.html

Belzak L, Halverson J. The opioid crisis in Canada : a national perspective. Heal Promot Chronic Dis Prev Canada. 2018;38(6).

Bai J-W, Bao J, Bhatia A, Chan VWS. A perioperative approach to the opioid crisis. Can Med Assoc J. 2018;190(39):1151–2.

Overton HN, Hanna MN, Bruhn WE, Bicket MC, Makary MA. Opioid-Prescribing Guidelines for Common Surgical Procedures: An Expert Panel Consensus. J Am Coll Surg. 2018;227(4):411–8.

Hartford LB, Koughnett JAM Van, Murphy PB, Vogt KN, Hilsden RJ, Clarke CFM, et al. Standardization of Outpatient Procedure ( STOP ) Narcotics : A Prospective NonInferiority Study to Reduce Opioid Use in Outpatient General Surgical Procedures. 2018;228:81–9.

2019-20 Surgical Quality Improvement Campaign [Internet]. 2019. Available from: https://www.hqontario.ca/Quality-Improvement/Quality-Improvement-in-Action/OntarioSurgical-Quality-Improvement-Network/Cut-the-Count

Chou R, Gordon DB, Leon-casasola OA De, Rosenberg JM, Bickler S, Brennan T, et al. Management of Postoperative Pain: A Clinical Practice Guideline From the American Pain Society, the American Society of Regional Anesthesia and Pain Medicine, and the American Society of Anesthesiologists’ Committee on Regional Anesthesia, Executive Committee, and Administrative Council. J Pain [Internet]. 2016;17(2):131–57. Available from: http://dx.doi.org/10.1016/j.jpain.2015.12.008

Opioid Prescribing for Acute Pain [Internet]. Health Quality Ontario. 2018. Available from: https://www.hqontario.ca/portals/0/documents/evidence/quality-standards/qs-opioidacute-pain-clinician-guide-en.pdf

Verma S, Raegele M. Report on the AFMC Response to the Canadian Opioid Crisis [Internet]. 2017. Available from: https://afmc.ca/sites/default/files/pdf/2018_AFMC_Response_to_OPIOID_Crisis_EN.pdf

10.

Comer L. Content analysis of chronic pain content at three undergraduate medical schools in Ontario schools in Ontario. Can J Pain [Internet]. 2017;1(1):75–83. Available from: https://doi.org/10.1080/24740527.2017.1337467

11.

Medportal Curriculum Database [Internet]. 2020. p. 5–7. Available from:

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

https://cdb.medportal.ca/ 12.

Neuman MD, Bateman BT, Wunsch H. Inappropriate opioid prescription after surgery. Lancet. 2019;393:1547–57.

13.

Mariano E. Management of acute perioperative pain [Internet]. UpToDate. 2019. Available from: https://www-uptodate-com.proxy.lib.uwaterloo.ca/contents/management-of-acuteperioperative-pain/print?search=acute perioperative pain&source=search_result&selectedTitle=1~150&usage_type=default&display_rank=1

14.

Han B, Compton WM, Blanco C, Crane E. Prescription Opioid Use , Misuse, and Use Disorders in U . S . Adults : 2015 National Survey on Drug Use and Health. 2017;167(5):293–302.

MUMJ Volume 17 No. 1, pp. 1-15

June 2020

Appendix 1 Question Did you receive any teaching around multimodal pain management and opioid reduction strategies (such as use of nonopioid pharmacological and nonpharmacological interventions) in the postoperative patient during the surgery core rotation? Have you received teaching on multimodal pain management and opioid reduction strategies in the postoperative setting in a different core rotation (e.g. anesthesia, orthopedics, etc.)? If you answered yes to the previous question, please specify which rotation(s). Please select the core rotations you have completed so far in clerkship.

Have you received teaching on multimodal pain management and opioid reduction strategies in the postoperative setting during preclerkship (i.e. in tutorial sessions, large group sessions, clinical skills, or professional competencies)? How confident do you feel in your knowledge of multimodal postoperative pain management options at this stage of your training? Do you feel additional teaching on this topic would be beneficial for you? Please select the setting in which you received teaching around multimodal pain management and opioid reduction strategies in the post-operative patient during the surgery core. In your opinion, did the teaching effectively highlight the importance of multimodal pain

Options a) Yes b) No

a) Yes b) No

Custom Responses a) b) c) d) e) f) g) h) i) j) a) b)

Medicine Surgery Obstetrics/Gynecology Orthopedic Surgery Anesthesia Medical Subspecialty Selective Family Medicine Emergency Medicine Pediatrics Psychiatry Yes No

a) b) c) d) e) a) b) c) a)

Very Confident Confident Neutral Unconfident Very Unconfident Yes No Neutral In a formal and focused setting (classroom or online module) Informally (such as in passing by an attending or resident while on service) Both Strongly agree Agree

b) c) a) b)

MUMJ Volume 17 No. 1, pp. 1-15 Which if any of the following pain management modalities were discussed during the core rotation.

Did you notice any of the above approaches being put in practice by attendings or residents in the management of postoperative pain in patients? Did you receive teaching on the positive impact of patient and caregiver education around the pain management plan? Did you notice patient education surrounding pain management being practiced by attendings or residents during the core rotation?

June 2020 a) NSAIDs (naproxen, ibuprofen, celecoxib, ketorolac) b) Acetaminophen c) Pregabalin or gabapentin d) Psychological interventions (cognitive behavioural therapy, guided imagery, and other relaxation techniques) e) Other forms of non-pharmacological intervention (heat, ice, massage, stretching, rest, acupuncture, bracing or wrapping, spinal manipulation, passive physical therapy, positioning, splints, transcutaneous electrical nerve stimulation) f) Other modality not already listed (please specify) a) Always b) Often c) Sometimes d) Rarely e) Never f) Not Applicable/Donâ&#x20AC;&#x2122;t Recall a) Yes b) No a) b) c) d) e) f)

Always Often Sometimes Rarely Never Not Applicable/Donâ&#x20AC;&#x2122;t Recall

Appendix 1. List of survey questions and response options included.

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

Original Research Article

A geospatial analysis of pertussis and its risk factors in southern Ontario from 2005â&#x20AC;&#x201C;2016 Taha Elghamudi1 and Olaf Berke2 1 2

Michael G. DeGroote School of Medicine, McMaster University, Hamilton, Canada Department of Population Medicine, Ontario Veterinary College, Guelph, Canada

Abstract Introduction: Pertussis, commonly known as whooping cough, is a bacterial respiratory tract infection caused by Bordetella pertussis. Pertussis affects more than 48 million people worldwide annually, most of whom are under the age of 5. Hypothesis & Objectives: The hypothesis being investigated is that pertussis incidence, between 2005 and 2016, is not equally distributed across public health units in southern Ontario. We aim to identify disease cluster locations and associate geospatial fluctuations in incidence rates with putative risk factors. Materials and Methods: Data was sourced from Public Health Ontario on pertussis incidence in southern Ontario for all ages, specifically for each public health unitâ&#x20AC;&#x2122;s geographical area. A choropleth map was generated using data smoothed by empirical Bayesian estimation in a spatial analysis context. Following the creation of an incidence map for southern Ontario, the spatial scan test was applied to elucidate the existence of any disease clusters at a public health unit level. Moranâ&#x20AC;&#x2122;s I was used to determine whether there was evidence of any spatial dependence in pertussis incidence. Finally, putative risk factors were assessed in Poisson regression models and spatial Poisson regression models as potential predictor variables. Results and Discussion: The flexible spatial scan test identified three spatial clusters where incidence rates of pertussis were higher than expected. A spatial Poisson regression model was fit that included predictor variables of socioeconomic status and population density. For every 100 people/km2 increase in population density there was a significant 6% increase in pertussis incidence (p=0.03). Interestingly, vaccination rates were not found to be predictive of pertussis incidence nor did the variable improve the model. This epidemiological study identifies where pertussis incidence is clustered and what variables it is associated with, both of which are valuable for public health purposes and as a reference for future research into pertussis. Keywords: Pertussis, Spatial cluster, Mapping, Ontario, Canada, Regression, Vaccination rates, Socioeconomic status, Population density.

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

Corresponding author: taha.elghamudi@medportal.ca Introduction

Pertussis, commonly known as whooping cough, is an infectious disease that affects more than 48 million people worldwide and accounts for approximately 300,000 deaths annually (1-4). In 2017, Ontario saw 584 cases (5) while in the US there are over 48,000 cases of pertussis annually, with a high rate of underreporting due to mild symptoms being common (6). Pertussis is caused by the bacteria Bordetella pertussis which infects the respiratory tract and releases toxins, particularly the pertussis exotoxin, that irritate the airways and cause paroxysmal coughing spasms (7,8). Persistent coughing contributes to the highly contagious nature of pertussis. Pertussis is mainly a childhood disease, with 38% of reported cases being infants younger than 6 months, and close to three quarters of cases being children younger than 5 years old (9). Vaccines have been crucial in the decrease of pertussis incidence in the 20th century. In the United States, the incidence of pertussis decreased from 250,000 cases per year before vaccine introduction in the mid-20th century to as low as 1,010 reported cases in 1976. However, pertussis has slowly re-emerged, with pertussis outbreaks now being seen even in vaccinated populations. Reasons for this re-emergence may include a possible decreased effectiveness of the pertussis vaccine (4,10,11), or parents delaying or forgoing vaccination (12– 14). Moreover, the lack of robust reporting in the past contributes to the more recent rise in reported cases. Studies on cases of prolonged coughing suggest that despite the increase in reporting, pertussis is still more prevalent than was previously thought (4). It is likely that up to a million cases of respiratory illness due to pertussis infection exist in the US per year (4).

The spatial pattern of a disease’s incidence is related to the agent, host, and environment, otherwise known as the “epidemiological triangle.” These factors are essential to learn about a disease and how to control it. Spatial epidemiology can provide clues about these factors and their interrelations. For example, testing for the level of clustering helps to determine the nature of the disease agent, finding clusters can identify populations at risk, and disease mapping helps in visualizing the environment (15). Clustering refers to the spatial dependence in the data. A disease can have a high degree of clustering where cases are found in close spatial proximity, a high degree of regularity where the disease is equally spaced out, or it can have neither of these patterns and be spatially

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

independent. Disease clusters are regions where the number of cases is too high to occur randomly and therefore those regions can be identified as high-risk. Thus, clustering and disease clusters refer to different spatial patterns and can be independent of one another. General linear models are a useful approach for describing patterns and relationships in spatial data. One of the assumptions of general linear models is that observations are independent of one another. However, when analyzing spatial data of a contagious disease this assumption may not hold true. If a disease is rare then it is often considered to follow a Poisson distribution. Thus, the analysis of risk factors for rare diseases often employs a spatial Poisson regression model (16). This type of model is part of a larger group called generalized linear mixed models (16). These models are random effect models, where the random effects are spatially correlated (16). In Ontario, pertussis is a “reportable disease,” meaning physicians and other health-care practitioners are required to report cases of pertussis to their Medical Officer of Health. The rates of pertussis incidence are then collected by each Public Health Unit (PHU) and reported on an annual basis. It is expected in this analysis that pertussis incidence and the identified risk factors vary in space, thus prompting the use of spatial epidemiology to relate them. To the best of the authors’ knowledge no research has been conducted on the geospatial fluctuations in pertussis incidence in southern Ontario.

This study seeks to investigate whether pertussis incidence is equally distributed across PHUs in Ontario between 2005 and 2016. We aim to identify possible disease clusters and associate geospatial fluctuations in incidence rates with putative risk factors. Materials and Methods

Incidence rates of pertussis and population size between 2005 and 2016 in each of the PHUs in southern Ontario were retrieved from Public Health Ontario (14). These annual data were aggregated so that each PHU had a cumulative value for both population and pertussis incidence. The base map used to visually represent the data was a boundary file of the 29 PHUs in southern Ontario retrieved from Statistics Canada (17). PHU data on vaccination rates were taken from the Technical Report of Immunization coverage for school pupils in Ontario (18). PHU data on population density and “low income” status were taken from the 2013 Census found on the Statistics Canada website (19). “Low income” was used as a measure for socioeconomic status, representing the percentage of families or individuals spending 20% more than average of their before-tax income on food, shelter, and clothing (19).

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

Before the data was visualized or analyzed, it was adjusted in order to produce more statistically appropriate results. Given that population sizes differ between different PHUs and that pertussis is a relatively rare disease, there was a difference in the level of error in our estimates since the standard error increases with decreasing sample size. Essentially, regions with smaller populations tend to have less accuracy and are likely to have more extreme results. Regional data were therefore standardized with a method called shrinkage estimation prior to visualizing or analyzing the data (20). This is also called “map smoothing”, or empirical Bayesian estimation, in a spatial analysis context. Shrinkage estimation can be done based on iterative likelihood estimates for prior moments (21) or based on methods-of-moments estimators (22). Both methods for producing estimators are valid for rare events, which follow an approximate Poisson distribution. A choropleth map was generated using the smoothed pertussis incidence (23). This was done under many considerations. The smoothed incidence by PHU over the entire study period was classified using the quintiles of the empirical distribution. Five colours were used in the map, as it was found that best differentiation between regions was found with 5 colours (24). The scale of the colours was generated based on the quintiles of the distribution of incidence rates (24). The colour scheme chosen was colour-blind adjusted, to accommodate those with colourblindness (25). After the data was smoothed and the choropleth map generated, clustering was tested for and specific disease clusters were identified. Clustering was tested for using Moran’s I correlation coefficient (25). This test is the spatial equivalent of Pearson’s correlation coefficient and values for it generally range between -1 and +1. Large positive values indicate strong clustering whereas large negative values indicate regularity in the distribution, and values close to zero indicate spatial independence (27). Pertussis clusters, or regions with higher incidence rates than expected through random chance, indicating increased risk in those areas, were identified. This was done with a spatial scan test (28) which is based on scanning windows of fixed area size to identify locations of excess risk. Specifically, flexibly shaped scanning windows were used because the circle, which is normally used as the scanning window in spatial scan tests, does not always represent the practical shape of disease clusters (29,30). The primary cluster (i.e. the most statistically significant) was identified, as well as any secondary clusters with less significance but potentially greater relative risk for pertussis infection. A Poisson regression model was fitted using vaccination rates (at 7 and 17 years of age), low income, and population density as predictors. Since this type of model ignores spatial effects, a spatial Poisson regression model was also fitted to see if it held more explanatory value. These models start with the inclusion of all potential predictor variables and sequentially remove the least explanatory variable. Interaction effects between potential predictor variables were also investigated by testing whether the addition of interaction terms improved the model.

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

The model that was eventually chosen with this method maximized explanatory value while minimizing the number of variables by removing the ones that had the least significant effect. All mapping and data analysis were done using R Statistical Software as a Geographic Information System (31,32). All statistical tests were also done in R and were evaluated at a significance level of Îą=0.05. Results and discussion

The visualization of pertussis incidence rates (Figure 1) is a geospatial representation of the average annual incidence of the disease over 2005â&#x20AC;&#x201C;2016 in southern Ontario. Important to note in the legend is the difference in the length of each quintile because a long 1st and/or 5th quintile would indicate the potential for a geospatial outlier in pertussis incidence.

There were three clusters identified by the flexible spatial scan test (Figure 2). The primary cluster was comprised of Toronto and York PHUs. The high population density of these two regions is the chief reason for this being the primary cluster. The number of new cases of a disease is strongly related to the number of susceptible individuals, with population density being a very important factor (33). Additionally, the large populations in these regions results in the largest power or amount of evidence for the presence of a cluster as identified by the flexible scan test. The other two clusters consisted almost entirely of PHUs made up of rural counties in Southwestern Ontario. Interestingly, the secondary cluster (consisting of Elgin-St Thomas, Huron, Oxford, and Perth PHUs) has a greater relative risk (RR=4.39, p=0.01) and standardized incidence ratio (SIR=4.02, p=0.01) than the other two clusters (Table 1). However, the smaller population means the analysis for this specific cluster is underpowered, which is likely why it is not the primary cluster. Rural southwestern Ontario regions, including Wellington and Dufferin counties, which make up part of the third cluster, have some of the highest incidence rates in southern Ontario. These counties are home to many of Ontarioâ&#x20AC;&#x2122;s Mennonite communities, which tend to have lower vaccination rates than the general population (34). It is possible that there is an association between these lower vaccination rates and the increased incidence of pertussis.

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

Pertussis cases per 100,000 (smoothed)

100km

6.11 - 40.07 3.41 - 6.11 2.67 - 3.41 1.96 - 2.67 1.15 - 1.96

Figure 1. Choropleth map of the smoothed pertussis incidence data in southern Ontario Public Health Units, from 2005 to 2016.

100km

Figure 2. Map of the three clusters of pertussis incidence in southern Ontario Public Health Units, from 2005 to 2016. The primary cluster is represented in red, the secondary cluster is represented in green, and the tertiary cluster is represented in blue.

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

Table 1. Clusters of pertussis incidence in southern Ontario and their supplementary data. Public Health Units Toronto, York

Elgin-St Thomas, Huron, Oxford, Perth

Wellington-DufferinGuelph

Cases

3290

733

246

Expected Cases

2033

182

148

SIR (Standardized Incidence Ratio) RR (Relative Risk)

1.61

4.02

1.65

2.21

4.39

1.68

p-value

0.01

Finally, although three clusters of regions at higher risk of pertussis were detected, there was no evidence for an overall pattern of clustering in the pertussis incidence data for southern Ontario. Moranâ&#x20AC;&#x2122;s I was very close to 0 (-0.0345, p<0.001) indicating spatial independence. This spatial independence may be explained by the large geographic size of the regions (PHUs) used in our spatial analysis relative to the transmission radius of pertussis among the human population. This transmission area is relative to the activity area of humans, which should be on average much smaller than the PHUs.

The model that was used was a spatial Poisson regression model that included low income and population density as predictor variables (Table 2). The presence of overdispersion in the general linear model motivated the use of a spatial model. Overdispersion is when the variance in the data is significantly greater than the mean, when they should be the same in a Poisson model. This was checked by looking at the ratio of deviance residual sum over degrees of freedom, which should be approximately equal to 1. Vaccination rates were not included because no relationship was found between vaccination rates and pertussis incidence. Population density was found to have a statistically significant effect with a relative risk of 1.06 (p=0.03). Given this relative risk, for every 100 people/km2 increase in population density there was a 6% increase in pertussis incidence. Low income was kept in the model despite not being significant (RR=0.35, p=0.08) because upon its removal, the size of the effect (i.e. RR) of population density changes dramatically. Thus, low income was identified as a confounding variable that was necessary to maximize explanatory value of the model. It has a relative risk of 0.35, therefore, for every 10%

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

increase in the proportion of families and individuals categorized as low income, pertussis incidence decreased by a factor of 0.35. Table 2. Spatial Poisson regression model for pertussis incidence in southern Ontario with Low Income and Population Density as predictor variables. Variable

Estimate

Standard Error

Relative Risk

p-value

(Intercept)

-8.936

0.717

N/A

Low Income

-0.105

0.058

0.35

0.08

0.001

0.0002

1.05

0.03

Population Density

The finding that vaccination rates were not significant was unexpected. A possible explanation for this finding is that since pertussis is a rare disease in southern Ontario and rates of vaccination are high, the differences between PHUs in vaccination rates are not substantial enough to cause any significant difference in incidence. Low income, defined as the percentage of families or individuals spending 20% more than average of their before-tax income on food, shelter and clothing, was inversely related to pertussis which was also unexpected. However, this finding was not significant and low income is a confounding variable with population density. This confounding is likely the result of the fact that the Toronto PHU is an outlier with the highest population density, and it is also the PHU with the highest value for low income. Thus, this result should be viewed with caution.

Although pertussis is a reportable disease in Ontario, there is still a significant underreporting problem that affects the data for pertussis incidence (4). Underreporting of pertussis is likely due to pertussis infections sharing symptoms with other respiratory diseases, coupled with the fact that when symptoms are not overly severe patients often elect to not see a doctor for their cough. Another very important limitation of this research is the lack of controlling for age when conducting tests or creating maps. Age is a crucial factor in pertussis incidence, since it is a childhood disease. It is possible that different PHUs have varying age distributions in their populations which influences the overall incidence. Age may have played an essential role in creating the clusters that were found and would be a confounding variable when associating these clusters with external factors such as vaccination rates or socio-economic status. Toronto PHU being an outlier in population density is a limiting factor for making meaningful conclusions based on the regression model. However, removing this PHU from the 23

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

dataset leaves no significant variables in the regression model. Toronto is such a different environment from the rest of southern Ontario that classifying it alongside the other PHUs should restrict how the model is interpreted. Further, more detailed, analysis of Toronto would be necessary, but the respective data are not available for this project.

Pertussis incidence in southern Ontario for the years 2005â&#x20AC;&#x201C;2016 has a spatial pattern that has now been mapped. There are three clusters of higher-than-expected incidence in both rural and highly metropolitan PHUs. Vaccination rate was not found to be a significant predictor variable for pertussis incidence. Population density was identified as a significant predictor variable and was included in the spatial regression model. Socioeconomic status (SES), here represented by low income percentage, was a confounder between population density and pertussis incidence. However, it was included to improve the experimental model. Knowledge of the spatial pattern of pertussis in Ontario as well as its predictor variables is valuable for public health workers who can use it to augment their strategy in reducing the incidence of this infectious disease. Primary care physicians who care for children in the higherrisk clusters can also be better informed of the increased risk their young patients face. Future research should be directed towards other potential predictor variables of pertussis incidence that were not included in this article. These variables include the different strains of Bordetella pertussis, different types of pertussis vaccines, household structure, climate, and other demographic factors. Future research should also be done in other geographic locations (such as another Canadian province, or a US state) to see whether the results here can be replicated elsewhere. Of particular importance is the question of whether vaccination rates have an effect on pertussis incidence in other developed areas.

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

References 1.

Cotter PA, Miller JF. Genetic analysis of the Bordetella infectious cycle. Immunopharmacology. 2000 Jul 25;48(3):253–5.

Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clin Microbiol Rev. 2005 Apr;18(2):326–82.

Crowcroft NS, Stein C, Duclos P, Birmingham M. How best to estimate the global burden of pertussis? Lancet Infect Dis. 2003 Jul 1;3(7):413–8.

Cherry JD. The Present and Future Control of Pertussis. Clin Infect Dis. 2010 Sep 15;51(6):663–7.

Public Health Ontario. Pertussis (Whooping Cough) [Internet]. Toronto ON: Public Health Ontario. Available from: https://www.publichealthontario.ca/en/diseases-andconditions/infectious-diseases/vaccine-preventable-diseases/pertussis

Centers for Disease Control and Prevention. Pertussis (Whooping Cough) [Internet]. Atlanta: Centers for Disease Control and Prevention; 2017. Available from: https://www.cdc.gov/pertussis/outbreaks/about.html.

Cotter PA, Jones AM. Phosphorelay control of virulence gene expression in Bordetella. Trends Microbiol. 2003;11(8):367–73.

Melvin JA, Scheller E V., Miller JF, Cotter PA. Bordetella pertussis pathogenesis: Current and future challenges. Nat Rev Microbiol. 2014;12(4):274–88.

Lauria AM, Zabbo CP. Pertussis (Whooping Cough) [Internet]. StatPearls. Available from: https://www.ncbi.nlm.nih.gov/books/NBK519008/

10.

Andrews R, Herceg A, Roberts C. Pertussis notifications in Australia, 1991 to 1997. Commun Dis Intell. 1997;21(11):145–8.

11.

De Melker HE, Schellekens JFP, Neppelenbroek SE, Mooi FR, Rümke HC, Conyn-Van Spaendonck MAE. Reemergence of pertussis in the highly vaccinated population of the Netherlands: Observations on surveillance data. Emerg Infect Dis. 2000;6(4):348–57.

12.

Hellenbrand W, Beier D, Jensen E, Littmann M, Meyer C, Oppermann H, et al. The epidemiology of pertussis in Germany: Past and present. BMC Infect Dis. 2009 Feb 25;9.

13.

Raguckas SE, VandenBussche HL, Jacobs C, Klepser ME. Pertussis resurgence: Diagnosis, treatment, prevention, and beyond. Pharmacotherapy. 2007 Jan;27(1):41–52.

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

14.

Tanaka M, Vitek CR, Pascual FB, Bisgard KM, Tate JE, Murphy T V. Trends in Pertussis among Infants in the United States, 1980-1999. J Am Med Assoc. 2003 Dec 10;290(22):2968–75.

15.

Berke O. Exploratory spatial relative risk mapping. Prev Vet Med. 2005 Oct 12;71(3– 4):173–82.

16.

Mohebbi M, Wolfe R, Jolley D. A poisson regression approach for modelling spatial autocorrelation between geographically referenced observations. BMC Med Res Methodol. 2011;11(1).

17.

Statistics Canada. Health region boundary files. In: Health Indicators. 2015. Available from https://www150.statcan.gc.ca/n1/pub/82-402-x/2011001/reg-eng.htm.

18.

Public Health Ontario. Immunization Coverage Report for School Pupils in Ontario: 2013– 14, 2014–15 and 2015–16 School Year. Toronto, ON: Queen’s Printer for Ontario; 2017.

19.

Statistics Canada. Health Profile, December 2013 – City of Toronto Health Unit [Internet]. Ottawa, ON: Statistics Canada; 2013 Dec. Available from: www12.statcan.gc.ca/healthsante/82228/details/page.cfm?Lang=E&Tab=1&Geo1=HR&Code1=3595&Geo2=PR&Code2=35 &Data=Rate&SearchText=City of Toronto Health Unit&SearchType=Contains&SearchPR=01&B1=All&Custom=&B2=All&B3=All.

20.

Berke O. Exploratory disease mapping: Kriging the spatial risk function from regional count data. Int J Health Geogr. 2004 Aug 26;3(1).

21.

Clayton D, Kaldor J. Empirical Bayes Estimates of Age-Standardized Relative Risks for Use in Disease Mapping. Biometrics. 1987 Sep;43(3):671.

22.

Marshall RJ. Mapping disease and mortality rates using empirical Bayes estimators. J R Stat Soc Ser C Appl Stat. 1991;40(2):283–94.

23.

Berke O. Choropleth mapping of regional count data of Echinococcus multilocularis among red foxes in Lower Saxony, Germany. Prev Vet Med. 2001 Dec 3;52(2):119–31.

24.

Brewer CA, Pickle L. Evaluation of methods for classifying epidemiological data on choropleth maps in series. Ann Assoc Am Geogr. 2002;92(4):662–81.

25.

COLORBREWER 2.0 [Internet]. 2018. Available from: http://colorbrewer2.org/#type=sequential&scheme=OrRd&n=3

26.

Moran PA. Notes on continuous stochastic phenomena. Biometrika. 1950;37(1–2):17–23.

MUMJ Volume 17 No. 1, pp. 16-27

June 2020

27.

Waller LA, Gotway CA. Applied Spatial Statistics for Public Health Data. Hoboken, NJ, USA: John Wiley & Sons, Inc.; 2004. (Wiley Series in Probability and Statistics).

28.

Openshaw S, Charlton M, Wymer C, Craft A. A mark 1 geographical analysis machine for the automated analysis of point data sets. Int J Geogr Inf Syst. 1987 Jan 1;1(4):335â&#x20AC;&#x201C;58.

29.

Takahashi K, Kulldorff M, Tango T, Yih K. A flexibly shaped space-time scan statistic for disease outbreak detection and monitoring. Int J Health Geogr. 2008 Apr 11;7(1):14.

30.

Yao Z, Tang J, Zhan FB. Detection of arbitrarily-shaped clusters using a neighborexpanding approach: A case study on murine typhus in South Texas. Int J Health Geogr. 2011 Mar 31;10(1):23.

31.

R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria; 2019.

32.

Brunsdon C, Comber L. (2015). Introduction to R for Spatial Analysis and Mapping. London: Sage; 2015. 360 p.

33.

Tarwater PM. The effects of population density on the spread of disease. Texas Med Cent Diss. 1999 May; Available from: https://digitalcommons.library.tmc.edu/dissertations/AAI9929469.

34.

Williamson G, Ahmed B, Kumar PS, Ostrov BE, Ericson JE. Vaccine-preventable diseases requiring hospitalization. Pediatrics. 2017 Sep 1;140(3).

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Original Research Article

A systematic review of clinical decision tools used for diagnosing pulmonary embolism in the pediatric population YiChen (April) Liu1, Laura Nguyen1, Mohammed Hassan-Ali MD, MSc2, and April Kam MD, MScPH FRCPC2,3 Faculty of Health Sciences, McMaster University Department of Pediatrics, McMaster University 3 McMaster Children's Hospital, McMaster University 1 2

Abstract Objective: This review aims to evaluate the diagnostic accuracy of existing, adult clinical decision tools for pulmonary embolism, in the pediatric population. As a secondary objective, this review aims to summarize the diagnostic use of pre-identified risk factors and clinical features of pulmonary embolism in the pediatric population. Methods: A systematic search and screening of the Pubmed, Embase, CINAHL, and Cochrane databases was done in January 2018. Studies evaluating the diagnostic accuracy of clinical decision tools and/or risk factors and clinical features for pulmonary embolism in the pediatric population were included. The measures of diagnostic accuracy of clinical decision tools were calculated. The pooled sensitivity and specificity of risk factors were calculated using a bivariate random effects model. All included studies were assessed for quality using QUADAS-2. Results: Six studies were included: three case-control and three retrospective cohort studies. We found that no standard clinical decision tool for pulmonary embolism has been evaluated in the pediatric population. As well, adult clinical decision tools have low diagnostic utility in pediatrics. Conclusion: Adult clinical decision tools should not be used for pediatric patients. There was no single risk factor or clinical feature displaying reliable sensitivity; however, a central venous line, a recent surgery, or the finding of hemoptysis, all have a positive likelihood ratio greater than two, demonstrating their potential diagnostic utility. Large, prospective cohort studies are needed. Keywords: Pediatric, Pulmonary embolism, Clinical decision making, Clinical decision rules, Clinical decision tools

Corresponding Author: kama@mcmaster.ca 28

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Introduction Pulmonary embolism has long been described and studied in the adult population because of the high mortality and morbidity rates; however, there is a lack of research addressing pulmonary embolism in pediatric patients. The incidence of pulmonary embolism in this younger population is steadily increasing as medical advances allow critically ill children with predisposing conditions, such as congestive heart disease and malignancy, to survive for longer (1-4). Pulmonary embolism in the pediatric population, as in the adult population, is correlated with increased morbidity and a high mortality rate of around 10% (5). As emboli interrupt pulmonary blood flow, cardiac output is obstructed, and consequently hypotension and hypoxia occur (1,3,6,7). Additionally, recurrent thrombosis is a problem in the pediatric population (8). Using information from the Canadian Childhood Thrombophilia Registry, Monagle et. al, have found that over a mean follow up of 2.86 years, 8.1% of pediatric patients with pulmonary embolism had recurrent thrombosis (8). Adding to the difficulty of pediatric pulmonary embolism diagnosis is the invasive and expensive nature of the current diagnostic gold standard, pulmonary angiography (9). Other available methods for diagnosing pulmonary embolism are ventilation perfusion scans and computed tomography pulmonary angiography. While ventilation perfusion scans offer a more accessible method, conditions like congenital heart disease, known to predispose pediatric patients to pulmonary embolism, can interfere with the interpretation of test results (10,11). On the other hand, computed tomography pulmonary angiography requires radiation exposure, which is accompanied by risks that are still undetermined in the pediatric population (12). This further necessitates valid and reliable clinical decision tools to aid physician decision making around when to clinically rule-out pulmonary embolism, and when to subject patients to more invasive and potentially harmful testing. While numerous clinical decision tools for pulmonary embolism are validated in the adult population, such as the Wells criteria, Pulmonary Embolism Rule-out Criteria (PERC), and the revised Geneva score, there is a paucity of information available for the pediatric population (13-15). This can be partially attributed to the low incidence of pediatric pulmonary embolism, and as such, current medical practices for pediatric pulmonary embolism are based off of data from the adult population (1,7,9,16). There are, however, important differences in pulmonary embolism between the pediatric and adult populations. Pediatric pulmonary embolism is often clinically silent and exists primarily in patients with underlying medical conditions, such as congenital heart disease or infection, that mask the acute pulmonary embolism diagnosis (8,17). Thus, clinicians may have a low suspicion of pulmonary embolism in the pediatric population. This makes the acceptance of adult recommendations for the pediatric population less than ideal (1,18). The aim of this systematic review is to synthesize the available literature that evaluates the diagnostic validity of clinical decision tools for pulmonary embolism in the pediatric population. As a secondary objective, this systematic review will summarize the diagnostic

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

validity of risk factors and clinical features for pulmonary embolism in the pediatric population, in order to aid in the development of a clinical decision tool specific to the pediatric population. Methods

Electronic searches were conducted for Pubmed via Medline, Embase via OVID, CINAHL via EBSCO and the Cochrane Controlled Trials registry on January 26th, 2018. The following search terms were used for Medline and modified for each database: (pulmonary embolism[MeSH] OR pulmonary embolism) AND (diagnosis[MeSH] OR diagnosis* OR decision tree[MeSH] OR OR decision trees OR decision support techniques[MeSH] OR decision support techniques OR d-dimer OR clinical prediction OR clinical decision) AND (paediatric OR pediatric OR children OR child). Additionally, reference lists from retrieved publications were screened for missing articles. Publications were restricted to studies published after 2000 and written in English. The year 2000 was chosen as the cut-off point for the year of publication, as the first clinical decision tool for pulmonary embolism was published in 2001.

Two members of the study team independently scrutinized titles and abstracts, and judged articles to be excluded or to undergo full-text article review. Studies were deemed acceptable for full-text review if the title and/or abstract indicated that the paper evaluated the diagnosis of venous thromboembolism or pulmonary embolism in the pediatric population, or if it did not present an age range in the abstract. The full-text article was obtained if it was judged eligible by at least one reviewer. A fulltext screening form was created and piloted. Cohenâ&#x20AC;&#x2122;s Kappa was calculated to determine the interrater reliability prior to conducting the full-text review. These full-text articles were then judged to be included or excluded by two independent reviewers, and consensus for inclusion was reached by discussion mediated by a third reviewer.

Inclusion criteria In order to be included, studies had to meet the following criteria: 1. The study included children 21 years or younger as per the American Academy of Pediatrics definition. The study must also have presented separate information for this age group. 2. Patients had a suspected diagnosis of pulmonary embolism. 3. Medical findings used in the clinical decision tool, including patientâ&#x20AC;&#x2122;s risk factors and 30

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

physical examination details were described in adequate detail. 4. A diagnosis of pulmonary embolism was confirmed by radiography such as computed topography pulmonary angiography or ventilation/perfusion scans. Exclusion criteria Studies with the following characteristics were excluded: 1. Case reports, case series, and systematic reviews. 2. All children in study were diagnosed with pulmonary embolism. 3. Insufficient detail reported on patientâ&#x20AC;&#x2122;s risk factors and physical examination findings.

Two reviewers blinded to the paperâ&#x20AC;&#x2122;s author, journal, and institution independently assessed the risk of bias and applicability of these studies using QUADAS-2, a tool designed specifically for evaluating studies of diagnostic test accuracy (19).

A data extraction form was created and piloted. Two reviewers independently extracted relevant data. The following data was collected from each study: date of publication, journal of publication, geographic location of study, study design, clinical setting (e.g., hospital outpatient, hospital inpatient, or emergency department), type of reference standard applied (e.g., ventilationperfusion lung scan, helical computed tomography, computed tomography pulmonary angiography), demographic characteristics of sample (age range, mean age, sex), prevalence of pulmonary embolism in the study population, clinical decision tool evaluated and corresponding 2x2 tables, risk factors (e.g., oral contraception use, central venous catheterization, malignancy, surgery, and dehydration) and clinical features (e.g., heart rate, SpO2) evaluated, a n d t h e outcome of patients with each risk factor and clinical feature. If key data was missing, article authors were contacted regarding missing information.

Primary outcome analysis The primary outcome was the diagnostic accuracy of the clinical decision tools. This included the sensitivity, specificity, positive predictive values, negative predictive values, positive likelihood ratios, negative likelihood ratios, and their corresponding 95% confidence intervals. These were calculated using the RcmdrPlugin.EZR package in R (R Version 3.5.0) (20).

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Secondary outcome analysis As a secondary outcome, the diagnostic accuracy of each risk factor and clinical feature was calculated. For risk factors and clinical features evaluated in five or more studies, the bivariate random-effects model was used to summarize the sensitivity and specificity using the mada package in R (R Version 3.5.0) (20-22). The positive and negative likelihood ratios were not pooled, but calculated as point estimates from the pooled sensitivity and specificity values (23). Heterogeneity was also quantified using the I2 value which estimated the percentage of total variation that is due to heterogeneity between the studies rather than chance and takes into account the number of included studies. I2 scores range from 0% (no heterogeneity) to 100% (extreme heterogeneity). For findings evaluated in less than five studies, sensitivities, specificities, positive and negative likelihood ratios were calculated using the madad function in the mada package in R (R Version 3.5.0) (20). These were presented along with their corresponding 95% confidence intervals. Results As shown in Figure 1, over 3000 titles and abstracts were screened, and 6 articles met the final inclusion and exclusion criterion. The characteristics of these six studies can be found in Table 1. The risk of bias and applicability of included studies assessed using the QUADAS-2 Tool are displayed in Table 2 and Table 3, respectively.

Out of the included studies, only two studies evaluated the accuracy of preexisting clinical decision tools (Pulmonary Embolism Rule Out Criteria and Wells Criteria); however, five out of six studies created clinical decision tools after data analysis. A brief description of the clinical decision tools and their diagnostic accuracies are displayed in Table 4.

Table 5 displays the pooled measures of sensitivity and specificity of risk factors and clinical features that were evaluated in five studies. No risk factors nor clinical features were evaluated by all six studies. The measures of heterogeneity, positive likelihood ratios, and negative likelihood ratios are also displayed. The appendices display the measures of diagnostic accuracy of risk factors and clinical features evaluated in four or fewer studies.

MUMJ Volume 17 No. 1, pp. 28-49

Records identified through database searching MEDLINE (n = 1504)

June 2020

Records identified through database searching EMBASE (n = 2122)

Records identified through other databases (CINAHL & COCHRANE) (n = 211)

Records after duplicates removed (n = 3551)

Titles and abstracts screened (n = 3551)

Records excluded (n = 3261)

Full-text articles assessed for eligibility (n = 290)

Full-text articles excluded: 1) Case study, case series, systematic review, editorial (n=36) 2) No separate data for pediatric patients (n=155) 3) Data includes all thrombosis patients (n=58) 4) No controls (n=24) 5) Not enough details about risk factors or clinical features (n=11)

Studies included in qualitative synthesis (n = 6 )

Studies included in quantitative synthesis (meta-analysis) (n = 6 )

Figure 1. PRISMA flow diagram.

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Table 1. Characteristics of included studies. Clinical Authors Study design Reference Cases Controls Funding a b setting Standard Median age n % Median age n % (range)c Female (range)c Female Biss 2009 Non-specific Case-Control VQ, CTPA, 13 (0.003- 50 46% 12 (1-17) 25 44% Baxter Bioscience hospital (24) PA or 17) Canada, Heart and ECHO Stroke Foundation of Canada Hennelly Retrospective (VQ OR ED 15.2 (IQR: 36 56% 16.9 (IQR: 525 67% Not specified 2016 (25) Cohort CTPA) 13.9-20) 15-20.8) AND treatment Lee 2011 ED, IP, and Retrospective with CTPA Mean 13.6 36 50% Mean 14.1 191 54% Not specified anticoagulan (SD: 5.4) Cohort (26) OP (SD: 4.0) t Wang 2015 ED Case-Control VQ OR 15 (12-18) 11 91% 15 (3-18) 39 67% No external funding (27) CTPA Kanis 2017 ED, IP and OP Retrospective VQ OR 15 (IQR: 13- 51 49% 15 (14-17) 492 68% The Eli Lilly Foundation Cohort (28) CTPA 16) Physician Scientist Award Victoria Non-specific Case-Control VQ OR Mean 17 13 70% Mean 17 26 69% Not specified 2008 (29) hospital CTPA (13-21; SD: (13-21; SD: 2.6) 2.4) a ED = emergency department; IP = inpatient; OP = outpatient b VQ = ventilationâ&#x20AC;&#x201C;perfusion scan; CTPA = computed tomography pulmonary angiography; PA = conventional pulmonary angiography; ECHO = echocardiogram. c IQR = interquartile range; SD = standard deviation. IQR or SD was included if age range of included cases was unavailable.

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Table 2. Assessment of risk of bias in included studies using the QUADAS-2 tool. Hennelly Biss Lee 201625 200924 201126 Patient Selection: Could the selection of patients High Low Low have introduced bias? Index Test: Could the conduct and interpretation High Unclear Unclear of the index test have introduced bias? Reference Standard: Could the reference High Unclear Low standard, its conduct, or its interpretation have introduced bias? Flow and Timing â&#x20AC;&#x201C; Could the patient flow have High High Low introduced bias? Table 3. Assessment of applicability in included studies using the QUADAS-2 tool. Hennelly Biss Lee 201625 200924 201126 Patient Selection: Are there concerns that the Low Low Low included patients do not match the review question? Index Test: Are there concerns that the index High High High test, its conduct, or its interpretation differ from the review question? Reference Standard: Are there concerns that the Low High Low target condition as defined by the reference standard does not match the review question?

Wang 201527 High

Kanis 201728 High

Victoria 200829 High

Unclear

High

Unclear

High

Wang 201527 High

Kanis 201728 High

Victoria 200829 Low

High

Low

High

Low

MUMJ Volume 17 No. 1, pp. 28-49 Table 4. Diagnostic accuracy of clinical decision tools evaluated in primary studies. Clinical Decision Tool Evaluated

# of disease+

# of disease-

Biss 2009: Wells Simplified Probability Score24

A priori or Post hoc A priori

Sensitivity (95% CI), %

0.720 (0.575 - 0.838)

Biss 2009: Wells Simplified Probability Score & D-Dimer24

A priori

Hennelly 2016: PERC (excluding age <50 years) 25

A priori

Hennelly 2016: Wells (alternate diagnosis is less likely than PULMONARY EMBOLISM = 0) 25

A priori

Hennelly 2016: Wells (alternate diagnosis is less likely than PULMONARY EMBOLISM = 3) 25

June 2020 Specificity (95% CI), % 0.600 (0.387 - 0.789)

Positive predictive value 0.783 (0.636 - 0.891)

Negative predictive value 0.517 (0.325 - 0.706)

0.593 (0.388 - 0.776)

0.667 (0.349 - 0.901)

0.800 (0.563 - 0.943)

525

1.000 (0.858 - 1.000)

0.240 (0.204 - 0.279)

525

0.167 (0.064 - 0.328)

A priori

525

Hennelly 2016: PULMONARY EMBOLISM Model One of: Use of OCPs. Age-specific tachycardia, Hypoxia (SPO2 < 95%)25

A priori

Lee 2011: PULMONARY EMBOLISM Model At least one of: Immobilization, hypercoagulable state, excess estrogen state (OCP), Indwelling CVL, Prior PULMONARY EMBOLISM and/or DVT.26

Post hoc

Lee 2011: PULMONARY EMBOLISM Model At least two of: Immobilization, hypercoagulable state, excess estrogen state (OCP), Indwelling CVL, Prior PULMONARY EMBOLISM and/or DVT. 26

LR+ (95% CI)

LRâ&#x2C6;&#x2019; (95% CI)

1.800 (1.081 - 2.998)

0.467 (0.270 - 0.807)

0.421 (0.203 - 0.665)

1.778 (0.753 - 4.197)

0.611 (0.333 - 1.120)

0.083 (0.059 - 0.113)

1.000 (0.957 - 1.000)

1.316 (1.254 - 1.381)

0.000 (N/A)

0.960 (0.940 - 0.975)

0.222 (0.086 - 0.423)

0.944 (0.921 - 0.962)

4.167 (1.795 - 9.672)

0.868 (0.749 - 1.006)

0.861 (0.705 - 0.953)

0.579 (0.536 - 0.622)

0.123 (0.085 - 0.170)

0.984 (0.963 - 0.995)

2.046 (1.734 - 2.413)

0.240 (0.106 - 0.543)

525

0.889 (0.739 - 0.969)

0.560 (0.516 - 0.603)

0.122 (0.085 - 0.167)

0.987 (0.966 - 0.996)

2.020 (1.738 - 2.348)

0.198 (0.079 - 0.501)

191

0.944 (0.813 - 0.993)

0.634 (0.561 - 0.702)

0.327 (0.238 - 0.426)

0.984 (0.942 - 0.998)

2.577 (2.104 - 3.156)

0.088 (0.023- 0.339)

Post hoc

191

0.889 (0.739 - 0.969)

0.942 (0.899 - 0.971)

0.744 (0.588 - 0.865)

0.978 (0.945 - 0.994)

15.434 (8.597 - 27.710)

0.118 (0.047 - 0.297)

Lee 2011: PULMONARY EMBOLISM Model Three or more of: Immobilization, hypercoagulable state, excess estrogen state (OCP), Indwelling CVL, Prior PULMONARY EMBOLISM and/or DVT. 26

Post hoc

191

0.333 (0.186 - 0.510)

1.000 (0.971 - 1.000)

1.000 (0.640 - 1.000)

0.888 (0.838 - 0.927)

N/A

0.667 (0.529 - 0.840)

Wang 2015: PULMONARY EMBOLISM Model One of: family history of VTE, Obesity, Current or recent OCP use, recent surgery, immobilization, trauma or fracture, CVL, infection, or malignancy 27

Post hoc

1.000 (0.615 - 1.000)

0.308 (0.170 - 0.476)

0.289 (0.154 - 0.459)

1.000 (0.640 - 1.000)

1.444 (1.172 - 1.781)

0.000 (N/A)

Wang 2015: PULMONARY EMBOLISM Model Two of: family history of VTE, Obesity, Current or recent OCP use, recent surgery, immobilization, trauma or fracture, CVL, infection, or malignancy 27

Post hoc

0.818 (0.482 - 0.977)

0.538 (0.372 - 0.699)

0.333 (0.165 - 0.540)

0.913 (0.720 - 0.989)

1.773 (1.143 - 2.749)

0.338 (0.093 - 1.223)

Kanis 2017: PULMONARY EMBOLISM Exclusion Criteria All of: HR < 100bpm, respiratory rate < 22 breaths/min, SaO2% > 94%, no limb swelling, no recent surgery, no active cancer, no limb immobility, no CVL, and no prior VTE 28

Post hoc

492

0.922 (0.811 - 0.978)

0.439 (0.395 - 0.484)

0.146 (0.109 - 0.189)

0.982 (0.954 - 0.995)

1.643 (1.469 - 1.837)

0.179 (0.069 - 0.460)

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Table 5. Diagnostic validity of risk factors and clinical features evaluated in five studies. Central Venous Line Congenital Cardiac Disease Malignancy Surgery Hemoptysis Tachycardia

Sensitivity 0.225 (0.096 - 0.442) 0.133 (0.071 - 0.326) 0.180 (0.128 - 0.247) 0.264 (0.165 - 0.394) 0.067 (0.034 - 0.126) 0.591 (0.509 - 0.668)

Specificity 0.920 (0.721 - 0.981) 0.920 (0.834 - 0.964) 0.887 (0.733 - 0.957) 0.892 (0.872 - 0.910) 0.970 (0.955 - 0.980) 0.627 (0.507 - 0.734)

I^2 92.0% 75.9% 88.5% 42.1% 1.0% 84.5%

LR+ 2.8125 1.6625 1.5929 2.4444 2.2333 1.5845

LR0.8424 0.9424 0.9245 0.8251 0.9619 0.6523

Discussion

It is clear through this review, that there are only a small number of studies evaluating the diagnostic accuracy of clinical decision tools for pulmonary embolism in the pediatric population. All of these studies are retrospective cohort reviews or case-control studies, both of which have a high or unclear potential for risk of bias. From the limited number of studies we were able to review, it is clear that there is no standard clinical decision tool for pulmonary embolism for use in the pediatric population. While we were able to assess two clinical decision rules validated in the adult population, neither was found to provide utility for decision making in children. This is true even when vital signs are adjusted to pediatric ranges. There may be many reasons for this lack of diagnostic accuracy. First of all, these studies are all retrospective, which intervenes with the ability to properly assess the â&#x20AC;&#x153;alternate diagnosis less likely than pulmonary embolismâ&#x20AC;? condition in the Wells Criteria. Since this criterion is worth three points out of 12.5, its interpretation has huge implications on the accuracy of that decision tool. Additionally, pulmonary embolism in pediatric patients is often the result of underlying disease, which may in itself cause clinical features similar to pulmonary embolism. Due to the of the lack of clinical decision tools available for use in the pediatric population, primary study often created clinical decision tools using the data they had collected. Out of these, one by Lee et al. shows favourable results. It proposes further imaging if a patient presents with at least two of the following: immobilization, hypercoagulable state, excess estrogen state, an indwelling central venous line, and/or prior history of pulmonary embolism and/or deep vein thrombosis (26). This tool has a good balance between sensitivity at 88% and specificity at 94%. While this leaves 12% of pulmonary embolism patients with no further imaging, changing the rule threshold to only require one risk factor or clinical finding causes the specificity to fall to 63%. This would have to be balanced with the sensitivity increase to 94%, and further studied to confirm best use of this rule. Another clinical decision tool created by Wang et al. has 100% sensitivity. It recommends further testing if a patient has one of: family history of venous thromboembolism, obesity, current or recent oral contraceptive use, recent surgery, immobilization, trauma or fracture, central venous line, infection, or malignancy (27).

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Nevertheless, neither of these decision tools are studied prospectively, so further evaluation is needed before recommending either for clinical use.

The included studies examined many different risk factors and clinical features of pediatric patients suspected to have pulmonary embolism; six of these features were examined across five studies. Not one single factor or combination of factors displayed a reliable sensitivity using the thresholds established in the primary literature; however, the presence of a central venous line, a history that includes recent surgery, or the finding of hemoptysis, all have a positive likelihood ratio greater than two, demonstrating the potential diagnostic utility of these risk factors and clinical features. It should be noted that there is a large heterogeneity when the studies evaluating the risk of central venous line are pooled. This may stem from the variability between the clinical settings, study designs, and reference standards used, but subgroup analysis could not be performed due to the small number of primary studies available. Regardless, the presence of a central venous line and/or recent surgery are both identified by the International Society of Thrombosis and Hemostasis Pediatric Pulmonary Embolism Working Group as frequent risk factors associated with pulmonary embolism (30). While central venous lines are frequently used as life-saving interventions in critically ill patients, they provide a nidus for thrombus formation, resulting in increased risk of pulmonary embolism in both adults and children (1,31). Similarly, surgery is known to be a provoking etiology for venous thromboembolism in both adults and children (1, 32). Lastly, hemoptysis in a patient with pulmonary embolism is due to pulmonary tissue infarction and the resulting ischemic pulmonary parenchymal necrosis (33).

The limitations of this systematic review include the small number of studies that met the inclusion criteria, and the fact that none of the studies were prospective in nature. Since many components, such as the completeness of data collection, cannot be controlled for in a retrospective design, the accuracy of the diagnostic validity of these clinical decision tools, as well as of the diagnostic use of risk factors and clinical features evaluated, may have been impacted. As well, the studies were all conducted in tertiary care hospitals, and patients presenting to these hospitals may be more acute and/or complex than those that would present to smaller hospitals. Additionally, previous studies have shown a bimodal distribution of pediatric pulmonary embolism incidence, with the first peak in infants less than one year of age and the second in adolescents (1,17). The different risk factors present in the neonatal population, mean that they are likely to require a different clinical decision tool than older children (2,4). None of the included studies reported separate data for the neonatal population, resulting in another limitation of this review and presenting an area for further investigation.

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Through this review, it is evident that large scale prospective studies in varying hospital levels should be completed. We recognize that this is difficult given the low incidence of pulmonary embolism in the pediatric population. A prospective population-based study, in order to develop and validate a clinical decision tool for pulmonary embolism in the pediatric population would be valuable.

Conclusion In summary, it is evident that the clinical decision rules used to evaluate the possibility of PE in the adult population are not recommended for pediatric patients, due to their low diagnostic accuracy. Given the increased susceptibility of pediatric patients to the radiation associated consequences of imaging techniques, such as computed tomography pulmonary angiography, the development of a clinical decision rule to assist in decision making around appropriate use of this imaging technology in necessary. Risk factors and clinical features found to increase the probability of PE in children are the presence of a central venous line, a history that includes recent surgery, and the finding of hemoptysis. Larger prospective studies are needed to assist in the creation of an appropriate clinical decision rule.

Acknowledgements The authors would like to thank Zelalem F. Negeri at McMaster University for his guidance on the statistical analyses of this study. The authors would also like to thank Dr. Anthony K. C. Chan from the Division of Pediatric Hematology/Oncology at McMaster University for lending his expertise in this area.

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

References 1.

Andrew M, David M, Adams M, Ali K, Anderson R, Barnard D et al. Venous thromboembolic complications (VTE) in children: first analyses of the Canadian Registry of VTE. Blood [Internet]. 1994 [cited 2020 Mar 20] ;83:1251–7. Available from: https://www.ncbi.nlm.nih.gov/pubmed/8118029

Thacker PG, Lee EY. Pulmonary embolism in children. AJR Am J Roentgenol [Internet]. 2015 [cited 2020 Mar 20] ;204:1278–88. Available from: https://www.ncbi.nlm.nih.gov/pubmed/26001239

Babyn PS, Gahunia HK, Massicotte P. Pulmonary thromboembolism in children. Pediatr Radiol [Internet]. 2005 [cited 2020 Mar 20];35:258–274. Available from: https://www.ncbi.nlm.nih.gov/pubmed/15635472

Patocka C, Nemeth J. Pulmonary embolism in pediatrics. J Emerg Med [Internet]. 2012 [cited 2020 Mar 20];42:105–16. Available from: https://www.ncbi.nlm.nih.gov/pubmed/21530139

Biss TT, Brandão LR, Kahr WH, Chan AK, Williams S. Clinical features and outcome of pulmonary embolism in children. Br J Haematol [Internet]. 2008 [cited 2020 Mar 20];142:808–18. Available from: https://www.ncbi.nlm.nih.gov/pubmed/18564359

Stein PD, Kayali F, Olson RE. Incidence of venous thromboembolism in infants and children: data from the National Hospital Discharge Survey. J Pediatr [Internet]. 2004 [cited 2020 Mar 20];145:563–5. Available from: https://www.ncbi.nlm.nih.gov/pubmed/15480387

Van Ommen C, Heijboer H, Buller H, Hirasing RA, Heijmans HS, Peters M. Venous thromboembolism in childhood: a prospective two-year registry in the Netherlands. J Pediatr [Internet]. 2001 [cited 2020 Mar 20];139:676–681. Available from: https://www.ncbi.nlm.nih.gov/pubmed/11713446

Monagle P, Adams M, Mahoney M, Ali K, Barnard D, Bernstein M et al. Outcome of pediatric thromboembolic disease: a report from the Canadian Childhood Thrombophilia Registry. Pediatr Res [Internet]. 2000 [cited 2020 Mar 20];47:763– 766. Available from: https://www.ncbi.nlm.nih.gov/pubmed/10832734

Brandao LR, Labarque V, Diab Y, Williams S, Manson DE. Pulmonary embolism in children. Semin Thromb Hemost [Internet]. 2011 [cited 2020 Mar 20];37:772–85. Available from: https://www.ncbi.nlm.nih.gov/pubmed/22187400

10.

Van Ommen CH, Peters M. Acute pulmonary embolism in childhood. Thromb Res [Internet]. 2006 [cited 2020 Mar 20];118:13–25. Available from: https://www-ncbi-nlmnih-gov.libaccess.lib.mcmaster.ca/pubmed/10776809

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

11.

Tayama M, Hirata N, Matsushita T, Sano T, Fukushima N, Sawa Y, et al. Pulmonary blood flow distribution after the total cavopulmonary connection for complex cardiac anomalies. Heart Vessels [Internet]. 1999 [cited 2020 Mar 20];14:154–60. Available from: https://link.springer.com/article/10.1007/BF02482300

12.

Matsushita T, Matsuda H, Ogawa M, et al. Assessment of the intrapulmonary ventilationperfusion distribution after the Fontan procedure for complex cardiac anomalies: relation to pulmonary hemodynamics. J Am Coll Cardiol [Internet]. 1990 [cited 2020 Mar 20];15:842–848. Available from: https://www.ncbi.nlm.nih.gov/pubmed/15992866

13.

Brenner DJ, Hall EJ. Computed tomography—an increasing source of radiation exposure. N Engl J Med [Internet]. 2007 [cited 2020 Mar 20];357:2277–84. Available from: https://www.ncbi.nlm.nih.gov/pubmed/18046031

14.

Wells P, Anderson D, Rodger M, Stiell I, Dreyer JF, Barnes D, et al. Excluding Pulmonary Embolism at the Bedside without Diagnostic Imaging: Management of Patients with Suspected Pulmonary Embolism Presenting to the Emergency Department by Using a Simple Clinical Model and D-Dimer. Ann Intern Med [Internet]. 2001 [cited 2020 Mar 20];135:98-107. Available from: https://www.ncbi.nlm.nih.gov/pubmed/11453709

15.

Kline JA, Mitchell AM, Kabrhel C, Richman PB, Courtney DM. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. J Thromb Haemost [Internet]. 2004 [cited 2020 Mar 20];2:1247–55. Available from: https://www.ncbi.nlm.nih.gov/pubmed/15304025

16.

LeGal G, Righini M, Roy P-M, et al. Prediction of pulmonary embolism in the emergency department: the revised Geneva score. Ann InternMed [Internet]. 2006 [cited 2020 Mar 20];144: 165– 171. Available from: https://www.ncbi.nlm.nih.gov/pubmed/16461960

17.

Buck JR, Connors RH, Coon WW, Weintraub WH, Wesley JR, Coran AG. Pulmonary embolism in children. J Pediatr Surg [Internet]. 1981 [cited 2020 Mar 20];16:385–91. Available from: https://www.ncbi.nlm.nih.gov/pubmed/7252746

18.

David M, Andrew M. Venous thromboembolism complications in children: A critical review of the literature. J Pediatr [Internet]. 1993 [cited 2020 Mar 20];123:337. Available from: https://www.sciencedirect.com/science/article/abs/pii/S0022347605817305

19.

Whiting PF, Rutjes AW, Westwood ME, Mallett S, Deeks JJ, Reitsma JB et al. QUADAS2: A Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies. Ann Intern Med [Internet]. 2011 [cited 2020 Mar 20];155:529–536. Available from: https://www.ncbi.nlm.nih.gov/pubmed/22007046

20.

R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/. 2013. Accessed 2020 Mar 20.

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

21.

Reitsma J, Glas A, Rutjes A, Scholten R, Bossuyt P, Zwinderman A. Bivariate Analysis of Sensitivity and Specificity Produces Informative Summary Measures in Diagnostic Reviews. Journal of Clinical Epidemiology [Internet]. 2005 [cited 2020 Mar 20];58:982990. Available from: https://www.sciencedirect.com/science/article/abs/pii/S0895435605001629

22.

Philipp Doebler. Mada: Meta-Analysis of Diagnostic Accuracy. R package version 0.5.8. https://CRAN.R-project.org/package=mada. 2017. Accessed 2020 Mar 20.

23.

Zwinderman AH, Bossuyt PM. We should not pool diagnostic likelihood ratios in systematic reviews. Stat Med [Internet]. 2008 Feb 28 [cited 2020 Mar 20];27(5):687-97. Available from: https://onlinelibrary.wiley.com/doi/10.1002/sim.2992

24.

Biss TT, BrandĂŁo LR, Kahr WH, Chan AK, Williams S. Clinical probability score and Ddimer estimation lack utility in the diagnosis of childhood pulmonary embolism. J Thromb Haemost [Internet]. 2009 Oct [cited 2020 Mar 20];7(10):1633-1638. Available from: https://www.ncbi.nlm.nih.gov/pubmed/19682234

25.

Hennelly KE, Baskin MN, Monuteuax MC, Hudgins J, Kua E, Commeree A, et al. Detection of Pulmonary Embolism in High-Risk Children. J Pediatr [Internet]. 2016 [cited 2020 Mar 20];178:214-218.e3. Available from: https://www.ncbi.nlm.nih.gov/pubmed/27567411

26.

Y, Tse SK, Zurakowski D, Johnson VM, Lee NJ, Tracy DA, et al. Children suspected of having pulmonary embolism: multidetector CT pulmonary angiography--thromboembolic risk factors and implications for appropriate use. Radiology [Internet]. 2012 [cited 2020 Mar 20];262(1):242-51. Available from: https://www.ncbi.nlm.nih.gov/pubmed/22106353

27.

Wang CY, Ignjatovic V, Francis P, Kowalski R, Cochrane A, Monagle P. Risk factors and clinical features of acute pulmonary embolism in children from the community. Thromb Res [Internet]. 2016 [cited 2020 Mar 20];138:86-90. Available from: http://www.thrombosisresearch.com/article/S0049-3848(15)30220-6/pdf

28.

Kanis J, Pike J, Hall CL, Kline JA. Clinical characteristics of children evaluated for suspected pulmonary embolism with D-dimer testing. Arch Dis Child [Internet]. 2018 [cited 2020 Mar 20]; 103 (9): 835-840. Available from: https://www.ncbi.nlm.nih.gov/pubmed/29117964

29.

Victoria T, Mong A, Altes T, Jawad AF, Hernandez A, Gonzalez L, et al. Evaluation of pulmonary embolism in a pediatric population with high clinical suspicion. Pediatr Radiol [Internet]. 2009 [cited 2020 Mar 20];39(1):35-41. Available from: https://www.ncbi.nlm.nih.gov/pubmed/19005649

30.

Biss TT, Rajpurkar M, Williams S, van Ommen CH, Chan AKC, Goldenberg NA, et al. Recommendations for future research in relation to pediatric pulmonary embolism: communication from the SSC of the ISTH. J Thromb Haemost [Internet]. 2018 [cited 2020

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Mar 20];16(2):405-408. Available from: https://www.ncbi.nlm.nih.gov/pubmed/29197153 31.

Derish MT, Smith DW, Frankel LR. Venous catheter thrombus formation and pulmonary embolism in children. Pediatr Pulmonol [Internet]. 1995 [cited 2020 Mar 20];20(6):349354. Available from: https://www.ncbi.nlm.nih.gov/pubmed/8649913

32.

Spentzouris G, Scriven RJ, Lee TK, Labropoulos N. Pediatric venous thromboembolism in relation to adults. J Vasc Surg [Internet]. 2012 [cited 2020 Mar 20];55(6):1785-1793. Available from: https://www.ncbi.nlm.nih.gov/pubmed/21944920

33.

Corey R. Hemoptysis. In: Walker HK, Hall WD, Hurst JW, editors. Clinical Methods: The History, Physical, and Laboratory Examinations. 3rd edition. Boston: Butterworths; 1990 [cited 2020 Mar 20]. Chapter 39. Available from: https://www.ncbi.nlm.nih.gov/books/NBK360/

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Appendices Appendix 1. Diagnostic validity of individual risk factors. Immobilization (n = 895) Sensitivity

Specificity

LR+

LR-

Biss 2009

0.560 (0.423 - 0.688)

0.480 (0.300 - 0.665)

1.077 (0.687 - 1.688)

0.917 (0.548 - 1.533)

Lee 2011

0.750 (0.589 - 0.862)

0.948 (0.906 - 0.971)

14.325 (7.613 - 26.954)

0.264 (0.150 - 0.465)

Wang 2015

0.273 (0.097 - 0.566)

0.846 (0.703 - 0.928)

1.773 (0.527 - 5.967)

0.860 (0.584 - 1.264)

Kanis 2017

0.333 (0.220 - 0.470)

0.937 (0.912 - 0.955)

5.290 (3.156 - 8.867)

0.711 (0.585 - 0.865)

Injury or Trauma (n = 1192) Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.139 (0.061 - 0.287)

0.907 (0.879 - 0.929)

1.488 (0.632 - 3.502)

0.950 (0.831 - 1.086)

Wang 2015

0.042 (0.004 - 0.301)

0.888 (0.754 - 0.953)

0.370 (0.021 - 6.401)

1.080 (0.919 - 1.269)

Kanis 2017

0.078 (0.031 - 0.185)

0.967 (0.948 - 0.980)

2.412 (0.838 â&#x20AC;&#x201C; 6.941)

0.953 (0.878 - 1.034)

Victoria 2008

0.077 (0.014 - 0.333)

0.920 (0.750 - 0.978)

0.962 (0.096 â&#x20AC;&#x201C; 9.639)

1.003 (0.826 - 1.219)

Oral contraceptive pill (n = 863) Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.306 (0.180 - 0.469)

0.853 (0.821 - 0.881)

2.083 (1.221 - 3.553)

0.814 (0.653 - 1.014)

Lee 2011

0.222 (0.117 - 0.381)

0.937 (0.893 - 0.964)

3.537 (1.557 - 8.036)

0.830 (0.694 - 0.992)

Wang 2015

0.727 (0.434 - 0.903)

0.846 (0.703 - 0.928)

4.727 (2.082 - 10.735)

0.322 (0.122 - 0.854)

Victoria 2008

0.350 (0.137 - 0.646)

0.971 (0.771 - 0.997)

11.900 (0.683 - 207.457)

0.670 (0.422 - 1.063)

Previous DVT and/or PE (n = 1406) Sensitivity

Specificity

LR+

LR-

Biss 2009

0.360 (0.241 - 0.499)

0.600 (0.407 - 0.766)

0.900 (0.491 - 1.650)

1.067 (0.728 - 1.562)

Hennelly 2016

0.250 (0.138 - 0.411)

0.943 (0.920 - 0.960)

4.375 (2.252 - 8.498)

0.795 (0.658 - 0.962)

Lee 2011

0.444 (0.295 - 0.604)

0.885 (0.832 - 0.923)

3.859 (2.256 - 6.599)

0.628 (0.467 - 0.845)

Kanis 2017

0.490 (0.359 - 0.623)

0.970 (0.950 - 0.981)

16.078 (9.079 - 28.474)

0.526 (0.402 - 0.689)

Thrombophilic condition and/or coagulation disorder (n = 822) Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.056 (0.015 - 0.181)

0.981 (0.965 - 0.990)

2.917 (0.664 - 12.815)

0.963 (0.889 - 1.043)

Lee 2011

0.222 (0.117 - 0.381)

0.932 (0.887 - 0.960)

3.265 (1.459 - 7.307)

0.835 (0.698 - 0.998)

Victoria 2008

0.444 (0.189 - 0.733)

0.960 (0.805 - 0.993)

11.111 (1.424 - 86.707)

0.579 (0.321 - 1.044)

Collagen vascular disease and/or connective tissue disease (n = 581) Sensitivity

Specificity

LR+

LR-

Kanis 2017

0.118 (0.055 - 0.234)

0.945 (0.921 - 0.962)

2.144 (0.929 - 4.947)

0.934 (0.843 - 1.034)

Victoria 2008

0.036 (0.004 - 0.268)

0.942 (0.784 - 0.987)

0.619 (0.027 - 14.216)

1.023 (0.891 - 1.175)

Renal disease (n = 1104)

MUMJ Volume 17 No. 1, pp. 28-49

June 2020

Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.028 (0.005 - 0.142)

0.975 (0.958 - 0.985)

1.122 (0.151 - 8.337)

0.997 (0.942 - 1.055)

Kanis 2017

0.020 (0.003 - 0.103)

0.988 (0.974 - 0.994)

1.608 (0.197 - 13.094)

0.992 (0.954 - 1.033)

Family history of PE/DVT (n = 593) Sensitivity

Specificity

LR+

LR-

Wang 2015

0.273 (0.097 - 0.566)

0.897 (0.764 - 0.959)

2.659 (0.697 - 10.146)

0.810 (0.556 - 1.182)

Kanis 2017

0.039 (0.011 - 0.132)

0.963 (0.943 - 0.977)

1.072 (0.256 - 4.489)

0.997 (0.941 - 1.057)

Infection (n = 88) Sensitivity

Specificity

LR+

LR-

Wang 2015

0.273 (0.097 - 0.566)

0.974 (0.868 - 0.995)

10.636 (1.224 - 92.414)

0.746 (0.518 - 1.076)

Victoria 2008

0.231 (0.082 - 0.503)

0.760 (0.566 - 0.885)

0.962 (0.286 - 3.234)

1.012 (0.699 - 1.466)

Asthma (n = 543) Kanis 2017

Sensitivity

Specificity

LR+

LR-

0.078 (0.031 - 0.185)

0.764 (0.725 - 0.800)

0.333 (0.128 - 0.864)

1.206 (1.098 - 1.325)

Diabetes mellitus (n = 543) Kanis 2017

Sensitivity

Specificity

LR+

LR-

0.01 (0.001 - 0.086)

0.989 (0.975 - 0.995)

0.862 (0.048 - 15. 368)

1.002 (0.974 - 1.030)

Neuromuscular disease (n = 38) Victoria 2008

Sensitivity

Specificity

LR+

LR-

0.231 (0.082 - 0.503)

0.760 (0.566 - 0.885)

0.962 (0.286 - 3.234)

1.012 (0.699 - 1.466)

Immobilization or surgery (n = 561) Hennelly 2016

Sensitivity

Specificity

LR+

LR-

0.194 (0.098 - 0.350)

0.874 (0.843 - 0.900)

1.547 (0.766 - 3.121)

0.921 (0.782 - 1.085)

Neoplasm (n = 38) Victoria 2008

Sensitivity

Specificity

LR+

LR-

0.385 (0.177 - 0.645)

0.840 (0.653 - 0.936)

2.404 (0.776 - 7.450)

0.733 (0.461 - 1.163)

Obesity (n = 50) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.364 (0.152 - 0.646)

0.821 (0.673 - 0.910)

2.026 (0.723 - 5.676)

0.776 (0.485 - 1.241)

Post-partum (n = 543) Kanis 2017

Sensitivity

Specificity

LR+

LR-

0.020 (0.003 - 0.103)

0.974 (0.955 - 0.984)

0.742 (0.099 - 5.557)

1.007 (0.966 - 1.050)

Pregnancy (n = 543)

MUMJ Volume 17 No. 1, pp. 28-49

Kanis 2017

June 2020

Sensitivity

Specificity

LR+

LR-

0.039 (0.011 - 0.132)

0.976 (0.958 - 0.986)

1.608 (0.370 - 6.985)

0.985 (0.930 - 1.043)

Appendix 2. Diagnostic validity of individual clinical features. D-Dimer (n = 213) Sensitivity

Specificity

LR+

LR-

Biss 2009

0.852 (0.675 - 0.941)

0.250 (0.089 - 0.532)

1.136 (0.790 - 1.632)

0.593 (0.156 - 2.249)

Lee 2011

0.880 (0.700 - 0.958)

0.131 (0.080 - 0.208)

1.012 (0.861 - 1.191)

0.917 (0.285 - 2.951)

Wang 2015

0.929 (0.561 - 0.992)

0.528 (0.313 - 0.732)

1.966 (1.158 - 3.340)

0.135 (0.009 - 2.027)

Victoria 2008

0.944(0.629 - 0.994)

0.375 (0.165 - 0.646)

1.511 (0.948 - 2.408)

0.148 (0.009 - 2.414)

Fever (n = 1170) Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.222 (0.117 - 0.381)

0.870 (0.839 - 0.897)

1.716 (0.896 - 3.287)

0.894 (0.748 - 1.067)

Wang 2015

0.050 (0.005 - 0.345)

0.921 (0.719 - 0.982)

0.633 (0.028 - 14.167)

1.031 (0.850 - 1.252)

Kanis 2017

0.137 (0.068 - 0.257)

0.900 (0.871 - 0.924)

1.378 (0.659 - 2.882)

0.958 (0.855 - 1.073)

Victoria 2008

0.077 (0.014 - 0.333)

0.846 (0.665 - 0.938)

0.500 (0.062 - 4.033)

1.091 (0.869 - 1.369)

Shortness of breath/dyspnoea (n = 1381) Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.528 (0.370 - 0.680)

0.493 (0.451 - 0.536)

1.042 (0.756 - 1.435)

0.957 (0.670 - 1.367)

Lee 2011

0.472 (0.320 - 0.630)

0.497 (0.427 - 0.568)

0.940 (0.647 - 1.364)

1.061 (0.755 - 1.491)

Wang 2015

0.818 (0.523 - 0.949)

0.436 (0.293 - 0.590)

1.450 (0.980 - 2.147)

0.417 (0.113 - 1.536)

Kanis 2017

0.686 (0.550 - 0.797)

0.354 (0.313 - 0.397)

1.062 (0.872 - 1.293)

0.887 (0.581 - 1.354)

Chest pain pleuritic (n = 352) Sensitivity

Specificity

LR+

LR-

Biss 2009

0.320 (0.208 - 0.458)

0.760 (0.566 - 0.885)

1.333 (0.595 - 2.986)

0.895 (0.669 - 1.197)

Lee 2011

0.417 (0.271 - 0.578)

0.455 (0.386 - 0.526)

0.765 (0.509 - 1.150)

1.281 (0.933 - 1.758)

Wang 2015

0.727 (0.434 - 0.903)

0.385 (0.249 - 0.541)

1.182 (0.762 - 1.833)

0.709 (0.250 - 2.013)

Cough (n = 1154) Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.222 (0.117 - 0.381)

0.771 (0.734 - 0.805)

0.972 (0.517 - 1.827)

1.008 (0.842 - 1.208)

Wang 2015

0.545 (0.280 - 0.787)

0.769 (0.617 - 0.874)

2.364 (1.076 - 5.192)

0.591 (0.302 - 1.155)

Kanis 2017

0.118 (0.055 - 0.234)

0.829 (0.793 - 0.860)

0.689 (0.317 - 1.498)

1.064 (0.955 - 1.185)

Hypoxia (n = 863) Sensitivity

Specificity

LR+

LR-

Biss 2009

0.360 (0.241 - 0.499)

0.520 (0.335 - 0.700)

0.750 (0.432 - 1.301)

1.231 (0.800 - 1.892)

Hennelly 2016

0.250 (0.138 - 0.411)

0.968 (0.949 - 0.980)

7.721 (3.706 - 16.085)

0.775 (0.641 - 0.937)

Lee 2011

0.250 (0.138 - 0.411)

0.749 (0.683 - 0.805)

0.995 (0.537 - 1.843)

1.002 (0.815 - 1.231)

MUMJ Volume 17 No. 1, pp. 28-49

June 2020 S&S of DVT: limb swelling/pain (n = 650)

Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.222 (0.117 - 0.381)

0.933 (0.909 - 0.952)

3.333 (1.672 - 6.645)

0.833 (0.699 - 0.994)

Wang 2015

0.182 (0.051 - 0.477)

0.795 (0.645 - 0.892)

0.886 (0.219 - 3.586)

1.029 (0.747 - 1.419)

Victoria 2008

0.462 (0.232 - 0.709)

0.808 (0.621 - 0.915)

2.400 (0.899 - 6.411)

0.667 (0.390 - 1.141)

Tachycardia age-adjusted (n = 683) Sensitivity

Specificity

LR+

LR-

Biss 2009

0.540 (0.404 - 0.670)

0.640 (0.445 - 0.798)

1.500 (0.838 - 2.684)

0.719 (0.472 - 1.094)

Hennelly 2016

0.611 (0.449 - 0.752)

0.676 (0.635 - 0.715)

1.887 (1.414 - 2.518)

0.575 (0.380 - 0.870)

Wang 2015

0.778 (0.453 - 0.937)

0.737 (0.580 - 0.850)

2.956 (1.564 - 5.585)

0.302 (0.088 - 1.039)

Chest pain (n = 611) Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.694 (0.531 - 0.820)

0.198 (0.166 - 0.234)

0.866 (0.694 - 1.080)

1.542 (0.916 - 2.599)

Wang 2015

0.958 (0.699 - 0.996)

0.312 (0.191 - 0.467)

1.394 (1.097 - 1.772)

0.133 (0.009 - 2.090)

DVT detected by lower limb Doppler Ultrasound (n = 66) Sensitivity

Specificity

LR+

LR-

Wang 2015

0.200 (0.057 - 0.510)

0.833 (0.436 - 0.970)

1.200 (0.136 - 10.580)

0.960 (0.598 - 1.541)

Victoria 2008

0.727 (0.434 - 0.903)

0.744 (0.589 - 0.854)

2.836 (1.487 - 5.409)

0.367 (0.137 - 0.980)

Syncope (n = 593) Sensitivity

Specificity

LR+

LR-

Wang 2015

0.091 (0.016 - 0.377)

0.923 (0.797 - 0.973)

1.182 (0.136 - 10.268)

0.985 (0.800 - 1.212)

Kanis 2017

0.020 (0.003 - 0.103)

0.929 (0.903 - 0.948)

0.276 (0.039 - 1.970)

1.055 (1.008 - 1.105)

S&S of DVT: lower limb - calf swelling (n = 1104) Sensitivity

Specificity

LR+

LR-

Hennelly 2016

0.056 (0.015 - 0.181)

0.975 (0.958 - 0.985)

2.244 (0.526 - 9.564)

0.968 (0.894 - 1.049)

Kanis 2017

0.157 (0.082 - 0.280)

0.965 (0.945 - 0.978)

4.540 (2.062 - 9.996)

0.873 (0.775 - 0.984)

S&S of DVT: upper limb (n = 636) Sensitivity

Specificity

LR+

LR-

Biss 2009

0.049 (0.015 - 0.146)

0.981 (0.840 - 0.998)

2.549 (0.127 - 51.168)

0.970 (0.893 - 1.053)

Hennelly 2016

0.139 (0.061 - 0.287)

0.950 (0.928 - 0.966)

2.804 (1.145 - 6.867)

0.906 (0.793 - 1.034)

S&S of DVT: lower limb - calf pain (n = 561) Hennelly 2016

Sensitivity

Specificity

LR+

LR-

0.194 (0.098 - 0.350)

0.947 (0.924 - 0.963)

3.646 (1.711 - 7.767)

0.851 (0.724 - 1.00)

S&S of DVT: lower limb (n = 75) Sensitivity

Specificity

LR+

LR-

MUMJ Volume 17 No. 1, pp. 28-49 Biss 2009

0.220 (0.128 - 0.352)

June 2020 0.920 (0.750 - 0.978)

2.750 (0.659 - 11.470)

0.848 (0.703 - 1.022)

Abnormal chest X-ray (n = 42) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.222 (0.063 - 0.547)

0.909 (0.764 - 0.969)

2.444 (0.479 - 12.480)

0.856 (0.594 - 1.233)

Cardiac symptoms (n = 39) Victoria 2008

Sensitivity

Specificity

LR+

LR-

0.077 (0.014 - 0.333)

0.615 (0.425 - 0.776)

0.200 (0.029 - 1.398)

1.500 (1.066 - 2.112)

Chest pain/shortness of breath (n = 39) Victoria 2008

Sensitivity

Specificity

LR+

LR-

0.692 (0.424 - 0.873)

0.038 (0.007 - 0.189)

0.720 (0.497 - 1.043)

8.000 (0.992 - 64.532)

Crackles/rales (n = 50) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.042 (0.004 - 0.301)

0.888 (0.754 - 0.953)

0.370 (0.021 - 6.401)

1.080 (0.919 - 1.269)

Elevated C-reaction protein (n = 20) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.857 (0.487 - 0.974)

0.385 (0.177 - 0.645)

1.393 (0.824 - 2.356)

0.371 (0.053 - 2.586)

Hypotension (n = 29) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.071 (0.008 - 0.439)

0.979 (0.828 - 0.998)

3.429 (0.075 - 157.683)

0.948 (0.766 - 1.174)

Increased respiratory effort (n = 50) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.273 (0.097 - 0.566)

0.897 (0.764 - 0.959)

2.659 (0.697 - 10.146)

0.810 (0.556 - 1.182)

Increased white cell count (n = 43) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.364 (0.152 - 0.646)

0.719 (0.546 - 0.844)

1.293 (0.496 - 3.370)

0.885 (0.539 - 1.455)

Palpitations (n = 561) Hennelly 2016

Sensitivity

Specificity

LR+

LR-

0.014 (0.001 - 0.117)

0.889 (0.859 - 0.913)

0.122 (0.008 - 1.927)

1.110 (1.058 - 1.165)

Pulmonary hypertension (n = 227) Lee 2011

Sensitivity

Specificity

LR+

LR-

0.194 (0.098 - 0.350)

0.963 (0.926 - 0.982)

5.306 (1.981 - 14.211)

0.836 (0.711 - 0.984)

S1Q3T3 pattern on ECG (n = 29)

MUMJ Volume 17 No. 1, pp. 28-49

Wang 2015

June 2020

Sensitivity

Specificity

LR+

LR-

0.500 (0.188 - 0.812)

0.783 (0.581 - 0.903)

2.300 (0.755 - 7.009)

0.639 (0.279 - 1.463)

Seizure (n = 543) Kanis 2017

Sensitivity

Specificity

LR+

LR-

0.020 (0.003 - 0.103)

0.990 (0.976 - 0.996)

1.929 (0.230 - 16.197)

0.990 (0.952 - 1.031)

Shock index >1 (n = 28) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.500 (0.188 - 0.812)

0.864 (0.667 - 0.953)

3.667 (0.978 - 13.745)

0.579 (0.256 - 1.311)

Sinus tachycardia: HR > 100 bpm on ECG (n = 29) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.500 (0.188 - 0.812)

0.783 (0.581 - 0.903)

2.300 (0.755 - 7.009)

0.639 (0.279 - 1.463)

Tachypnoea >20 breaths per minute (n = 45) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.444 (0.189 - 0.733)

0.667 (0.503 - 0.798)

1.333 (0.562 - 3.164)

0.833 (0.445 - 1.562)

Tachypnoea age-adjusted (n = 45) Wang 2015

Sensitivity

Specificity

LR+

LR-

0.444 (0.189 - 0.733)

0.806 (0.650 - 0.902)

2.286 (0.851 - 6.137)

0.690 (0.376 - 1.264)

MUMJ Volume 17 No. 1, pp. 50-56

June 2020

Case Report

Atypical euthyroid presentation of steroid responsive encephalopathy with associated thyroiditis Vineeth Bhogadi1 and Arlene Kelly-Wiggins2 1

Michael G. DeGroote School of Medicine, McMaster University

Geriatric and Memory Disability Clinic, Dalhousie University Faculty of Medicine

Abstract Introduction: Steroid-responsive encephalopathy with associated thyroiditis (SREAT) is an autoimmune disease associated with antithyroid antibodies. Its clinical features are variable, ranging from sudden focal neurologic deficits to progressive sub-acute cognitive decline, and thus it can mimic a host of other neurological and psychiatric conditions. Case presentation: We present a case of a previously healthy 62-year-old female with rapid onset neurocognitive and functional decline. EEG and MRI findings were consistent with an encephalopathy of unknown origin. Serologic findings revealed elevated antithyroid antibodies but were otherwise insignificant. Conclusion: Steroid-responsive encephalopathy with associated thyroiditis is typically a diagnosis of exclusion but should always be considered in cases of encephalopathy of unknown origin. The disorder is often underdiagnosed due to its variable presentation and poorly understood pathophysiology, but prognosis can be significantly improved with greater physician awareness and prompt management with corticosteroids.

Keywords: Hashimotoâ&#x20AC;&#x2122;s, Steroid-responsive, Encephalopathy Corresponding author: vineeth.bhogadi@medportal.ca

MUMJ Volume 17 No. 1, pp. 50-56

June 2020

Introduction Steroid-responsive encephalopathy and associated autoimmune thyroiditis (SREAT), previously known as Hashimotoâ&#x20AC;&#x2122;s encephalopathy, was first described by Brain and colleagues in 1966 (1). This condition is most often characterized by a subacute onset of confusion with an altered level of consciousness, myoclonus, and seizures but can have a wide range of clinical presentations (2). This particular case highlights an atypical presentation of SREAT and demonstrates the importance of thyroid function tests and anti-TPO antibody titers in the diagnosis of rapidly progressive encephalopathy. Case presentation A 62-year-old previously healthy woman presented to our geriatric service with a 4-month history of progressive cognitive and functional decline. She had increasing forgetfulness and anhedonia. She also experienced a gradual loss of activities of daily living (ADLs) and instrumental activities of daily living (IADLs) over 30 days following initial presentation. There was no history of previous cognitive decline, delirium, head injury, malignancy, or psychiatric symptoms such as delusions or hallucinations. Her relevant past medical history was significant for type II diabetes, hypertension, and angina. She had no history of substance use or recent travel. On examination, she was ambulating independently without any aids. Her blood pressure was 176/95 mmHg and heart rate was 64 beats per minute. Pupils were equal and reactive to light, although further visual testing was difficult due to patient cooperation. Cogwheel rigidity was present in the upper extremities bilaterally and hypertonicity was present in the lower extremities bilaterally. Babinskiâ&#x20AC;&#x2122;s sign was positive on the right. Slight paratonia was noted in all four extremities, with maintenance of limb position and involuntary resistance. Muscle bulk and power were preserved. Her gait was wide based and uncoordinated with a tendency to extend backwards. Cerebellar testing was normal. Cardiovascular, respiratory, and abdominal examinations were grossly normal. Cognitive testing revealed an MMSE of 14/30 and a MoCA of 8/30 with notable deficits in recall and orientation. The geriatric depression screen (GDS) was positive at 9/15. Investigations Complete blood count, extended electrolytes, liver function tests, renal function, and urinalysis were all within normal limits. HbA1c was elevated at 8.1%. Infectious disease work-up, including Syphilis IgG, HIV, Lyme disease, and Hepatitis B/C antibody titers were all negative. Anti-nuclear, anti-cardiolipin, p-ANCA, and c-ANCA antibodies were negative for an underlying autoimmune or vasculitic process. Paraneoplastic and voltage-gated potassium channel antibodies were all negative. Immunoglobulins G, A, and M were all within normal

MUMJ Volume 17 No. 1, pp. 50-56

June 2020

range. Thyroid function tests showed TSH and free T4 to be within normal limits. Anti-thyroid antibodies (anti-TPO) were elevated at 1709 U/ml. CSF was clear with normal opening pressure. CSF examination demonstrated elevated glucose at 4.62mmol/L, and total protein was normal. CSF serology was negative for herpes simplex virus and enterovirus. Creutzfeldt-Jacob disease workup was negative, including protein 14-3-3, h-Tau, and QuIC. Electroencephalography (EEG) was abnormal with diffuse moderate-to-severe disturbance of the background with some predominance in the left frontotemporal region. Frontal intermittent rhythmic delta activity (FIRDA) was also noted at times (Figure 1). CT and MRI revealed moderate microvascular small vessel ischemia and mild mid-brain atrophy (Figures 2A; 2B). SPECT brain perfusion scan showed diffuse brain atrophy but no specific findings of note (Figure 2C).

Figure 1. Electroencephalography (EEG) showing diffuse moderate-to-severe disturbance of the background with some predominance in the left frontotemporal region. Frontal intermittent rhythmic delta activity (FIRDA) also seen at times.

MUMJ Volume 17 No. 1, pp. 50-56

June 2020

Figure 2. Magnetic resonance images (MRI) and SPECT images of the brain, demonstrating diffuse brain atrophy and small vessel ischemia. (A) T1 Flair sagittal MRI; (B) T2 Flair axial MRI; (C) SPECT brain perfusion imaging.

Treatment Following the extensive work-up which revealed high anti-TPO antibodies, a diagnosis of encephalopathy of unknown origin was initially made, with consideration given to this being SREAT. The patient was started on 50 mg Prednisone PO at 1 mg/kg body weight for 6 days before being transitioned to a 12-week taper.

MUMJ Volume 17 No. 1, pp. 50-56

June 2020

Outcome At follow-up, the patient had dramatic improvement in cognitive status since initiation of steroid therapy, confirming the diagnosis of SREAT. She also had recovery in her functional status, with the patient able to manage some ADLs and IADLs independently. Repeat cognitive testing demonstrated an improvement, with an MSSE of 21/30 and a MoCA of 13/30, but the patient was not yet at her cognitive baseline. There was worsening functional decline with ongoing steroid taper. As such, the decision was made to maintain her on high dose steroid therapy at 1 mg/kg body weight. Discussion SREAT is a controversial disorder with poorly understood etiology and pathogenesis. A systematic review on the subject in 2016 identified only 251 well-documented cases in the literature (2). This may in part be due to the variable presentation of the disorder, leading to it being underdiagnosed. Among the reported cases, the median age of onset is 52 years with females accounting for 73% of cases (2). Clinical manifestations of SREAT are variable but most often include an acute or subacute onset of confusion accompanied by an altered level of consciousness. Two distinct patterns of presentation are noted to predominate: a stroke-like pattern with recurrent episodes of variable degree of cognitive decline and focal neurological deficits in 25% of cases, or a diffuse progressive pattern characterized by slowly progressive cognitive impairment with dementia, confusion, and hallucinations in the remaining 75% (3,4). On laboratory investigations, a defining feature of SREAT is an elevated serum antithyroid peroxidase antibody (antiTPO Ab) or antithyroglobulin antibody (antiTg Ab) (2,5). However, thyroid function is variable amongst patients, with 25-35% of patients having subclinical hypothyroidism, 17-20% with overt hypothyroidism, and approximately 7% being hyperthyroid (4,5). Additionally, both serum and CSF titres of antiTPO/antiTg Ab have not been shown to correlate with the severity of the disease, and their sensitivity and specificity for SREAT is unknown (6). Thus, the exact pathogenesis is poorly understood, and it is unclear as to how the antiTPO/antiTg antibodies affect the brain. Much of the current evidence points towards the underlying pathophysiology being an autoimmune vasculitis, with immune complexes disrupting the cerebral microvasculature (7,8). CSF is usually non-specific in suspected cases of SREAT. Analysis may reveal mononuclear pleocytosis in 20% of patients, whilst an elevated protein concentration occurs in 82% of patients (2). Similarly, EEG findings may be abnormal in 82% of cases with diffuse slowing consistent with encephalopathy as seen in our patient. MRI and CT imaging findings vary, with 49% of patients showing cerebral atrophy and non-specific white matter hyperintensities (2,4) as visualized in the current case. However, other individual case reports also note diffuse changes suggesting primary demyelination, meningeal enhancement, and T2 signal abnormalities in both hippocampi (5). 54

MUMJ Volume 17 No. 1, pp. 50-56

June 2020

Our case presents more atypically with rapidly progressive cognitive impairment and no focal deficits, unlike the two well described patterns of presentation discussed above (3,4). Notably, the patientâ&#x20AC;&#x2122;s manifestations did not concur with other common neurological symptoms associated with SREAT including tonic-clonic seizures (66% of patients), diffuse hyperreflexia (85% of patients), or myoclonus (38% of patients) (4). Additionally, our patient had no identifiable abnormalities on thyroid function tests and was euthyroid despite significantly elevated serum antiTPO Ab. This further highlights that the underlying mechanism of the condition is likely not related to thyroid status. It is important to keep SREAT on the differential diagnosis for patients with focal neurologic deficits and acute cognitive changes given its rather variable clinical presentation. SREAT is a diagnosis of exclusion and, as such, Creutzfeldt-Jakob disease, meningoencephalitis, decompensating mixed dementia (Alzheimerâ&#x20AC;&#x2122;s, Vascular), stroke and transient ischemic attack, cerebral vasculitis, and psychiatric diseases need to be excluded. Essential testing to exclude these diagnoses includes EEG, MRI, CSF analysis, and laboratory screening for usual causes of delirium as well as exclusion of an infectious etiology given the core immunosuppressive treatment. Patients with SREAT respond dramatically to treatment with corticosteroids, with 91% of patients showing a partial or complete clinical response (2). However, there is no consensus as to the appropriate steroid dose, with oral prednisone doses ranging from 50-150mg daily and a median dose of 60mg/day (2). We found 50mg (1mg/kg) to be the optimal dose for our patient, with further dose titration causing a partial decline in cognitive function. Long-term prognosis is variable, and residual cognitive impairment is seen in approximately 20% of patients (2,3). Our patient showed significant improvement in her neurocognitive status following the introduction of steroids, and progression of her cognitive deficits seems to be halted. This patient would be an excellent candidate for close follow-up to understand the dosage and duration of treatment with corticosteroids and to explore alternative immunosuppressant therapies in the future. Conclusion SREAT is a rare but underdiagnosed syndrome with a variety of clinical presentations. Our case highlights an atypical euthyroid presentation of SREAT despite significantly elevated antithyroid antibodies with non-specific EEG and MRI findings. This highlights the importance of testing antithyroid antibodies in patients with unexplained encephalopathy, along with the exclusion of other possible causes via CT, MRI, EEG, and serum and CSF analysis. A timely trial of corticosteroids is warranted in a patient with encephalopathy and elevated antithyroid antibodies without another obvious etiology. Acknowledgements

We would like to thank the case study patient for granting permission for publication.

MUMJ Volume 17 No. 1, pp. 50-56

June 2020

References 1.

Brain L, Jellinek EH, Ball K. Hashimoto's disease and encephalopathy. The Lancet. 1966 Sep 3;288(7462):512-4.

Laurent C, Capron J, Quillerou B, Thomas G, Alamowitch S, Fain O, Mekinian A. Steroidresponsive encephalopathy associated with autoimmune thyroiditis (SREAT): characteristics, treatment and outcome in 251 cases from the literature. Autoimmunity reviews. 2016 Dec 1;15(12):1129-3.

Kothbauer-Margreiter I, Sturzenegger M, Komor J, Baumgartner R, Hess CW. Encephalopathy associated with Hashimoto thyroiditis: diagnosis and treatment. Journal of neurology. 1996 Aug 1;243(8):585-93.

Canelo-Aybar C, Loja-Oropeza D, Cuadra-Urteaga J, Romani-Romani F. Hashimoto's encephalopathy presenting with neurocognitive symptoms: a case report. Journal of medical case reports. 2010 Dec;4(1):337.

Chong JY, Rowland LP, Utiger RD. Hashimoto encephalopathy: syndrome or myth?. Archives of Neurology. 2003 Feb 1;60(2):164-71.

Ferracci F, Bertiato G, Moretto G. Hashimoto's encephalopathy: epidemiologic data and pathogenetic considerations. Journal of the neurological sciences. 2004 Feb 15;217(2):1658.

Takahashi S, Mitamura R, Itoh Y, Suzuki N, Okuno A. Hashimoto encephalopathy: etiologic considerations. Pediatric neurology. 1994 Nov 1;11(4):328-31.

Philip, R., Saran, S., Gutch, M., & Gupta, K. (2014). An unusual presentation of Hashimoto's encephalopathy. Indian journal of endocrinology and metabolism, 18(1), 113.

MUMJ Volume 17 No. 1, pp. 57-62

June 2020

Case Report

Statin-induced rhabdomyolysis: A cautionary tale for high-dose rosuvastatin Henry He, BSc1, and Gulshan Atwal, MD2 1

Michael G. DeGroote School of Medicine, McMaster University, Hamilton, ON, Canada

Department of Internal Medicine, Southlake Regional Health Centre, Newmarket, ON, Canada 2

Abstract Statins are a widely prescribed lipid-lowering agent for preventing adverse cardiovascular events. However, a major side effect is rhabdomyolysis, a breakdown of muscle tissue, which can cause acute kidney injury and death. We present a case of a 77-year-old Chinese woman who was started on 40 mg rosuvastatin following percutaneous coronary intervention and ultimately developed rhabdomyolysis and acute kidney injury one month later. This case highlights the need to consider patient risk factors for developing statin-induced rhabdomyolysis when choosing the right dose of statin to prescribe. Keywords: Statin, Rhabdomyolysis, Asian, Stent

Corresponding Author: henry.he@medportal.ca

MUMJ Volume 17 No. 1, pp. 57-62

June 2020

Introduction Statins are a class of lipid-lowering medications that inhibit HMG-CoA reductase, the enzyme which catalyzes the rate-limiting reaction in cholesterol synthesis. By reducing cholesterol levels, specifically low-density lipoprotein cholesterol (LDL-C), statins reduce the risk of adverse cardiovascular events, including cardiovascular death, nonfatal myocardial infarction (MI), ischemic stroke, revascularization, and acute coronary syndromes hospitalizations. In patients with clinical atherosclerosis, the number needed to treat is 20 patients to prevent one adverse cardiovascular event over five years of treatment per 1 mmol/L reduction in LDL-C (1). However, a major side effect of statins is rhabdomyolysis, a breakdown of muscle tissue, which can cause acute kidney injury and death. The mechanism of statin-induced rhabdomyolysis is not well understood. The incidence of rhabdomyolysis is around 1 in 22,000 patients on statin monotherapy per year of therapy (2). Certain pre-disposing patient factors also increase the risk of developing rhabdomyolysis. Here we present a case of a 77-year-old Chinese woman who was started on 40 mg rosuvastatin post-percutaneous coronary intervention and ultimately developed rhabdomyolysis and acute kidney injury one month later. Case report A 77-year-old Chinese female presented to hospital for a four-day history of severe bilateral lower extremity muscle weakness and pain. The patient had been admitted a month prior at our hospital for ST-elevation MI and an 80% right coronary artery stenosis was found. A drug eluting stent was placed via PCI, and she was started on dual antiplatelet therapy (aspirin 81 mg, clopidogrel 75 mg), and rosuvastatin 40 mg. Around four days earlier, the patient had gone on a family trip and walked significantly more than usual. She subsequently developed acute leg muscle weakness. The patient denied any other significant past medical history or known allergies and denied taking regular medications at home. The patient was a nonsmoker and nondrinker. The patient noted poor appetite and weight loss of approximately 3 kg since her procedure a month prior. On examination, the patient had a small frame and weighed 45 kg. There was muscle weakness with hip flexion, power 3/5 bilaterally. There were no other upper or lower extremity muscle symptoms. Laboratory test results revealed a markedly elevated serum creatine kinase (CK) at 37345 U/L (0-160 U/L), as well as acute kidney injury (AKI): serum creatinine was 104 Âľmol/L (46-92 Âľmol/L), compared to 64 Âľmol/L at discharge one month prior. Serum alanine aminotransferase (ALT) was 313 U/L (14-49 U/L), serum potassium was 2.9 mmol/L (3.5-5.0 mmol/L) and serum calcium was 2.02 mmol/L (2.18-2.58 mmol/L). Calculated serum LDL-C was 1.45 mmol/L (target <2.0 mmol/L for high or intermediate cardiovascular disease risk). Urine was dark brown in colour. The patient was diagnosed with statin-induced rhabdomyolysis and admitted to general internal medicine. Rosuvastatin was held immediately and she was treated with a combination of

MUMJ Volume 17 No. 1, pp. 57-62

June 2020

intravenous normal saline and 100 mEq sodium bicarbonate at 150 mL/hr. Hypokalemia, hypophosphatemia and hypocalcemia were corrected. By day 5, the patient’s serum CK remained above the maximum value reported by our core lab (>82000 U/L); at this point other potential causes were considered including hypothyroidism, inflammatory myositis, and autoimmune myopathy. However, on day 6, the patient’s CK started to decrease. Intravenous normal saline was continued in addition to furosemide 40mg IV twice daily. Cardiology assessed the patient and discontinued her statin as her LDL-C level was well below target. The patient was ultimately discharged on day 12 of hospital stay with a serum CK of 2638 U/L and followed up in the outpatient internal medicine clinic.

Figure 1. Serum creatine kinase over hospital course. Note: * maximum CK value detected by hospital core lab was 82000 U/L

Discussion Rhabdomyolysis is an acute medical condition which results in skeletal muscle cell necrosis. Clinical presentation often involves proximal limb weakness, pain, and swelling, as well as teacoloured gross pigmenturia. It is characterized by the leakage of muscle cell contents—including myoglobin, CK, and ALT—into the bloodstream. Myoglobin-induced AKI is the most significant complication and can increase mortality. However, long term survival is favourable and majority of patients regain renal function when treated (3). Statins are an important nontraumatic, nonexertional cause of rhabdomyolysis, accounting for around 2.5% of the etiologies (4). The mechanism of statin-associated muscle symptoms is not well understood. Different statins also have different pharmacokinetics.

MUMJ Volume 17 No. 1, pp. 57-62

June 2020

Rosuvastatin has maximal effect at 4 weeks, is mainly metabolized by CYP2C9 in the liver and has a half-life elimination time of 19 hours (5). According to the Canadian Cardiovascular Society, a 50% reduction in LDL-C is recommended following PCI, which usually requires a moderate to high dose of a statin (1). For rosuvastatin, these are typically defined as 5-10 mg and 20-40 mg, respectively. However, Canadian pharmaceutical labelling recommends starting with 5 mg in Asian patients regardless of moderate or high intensity dosing, due to studies that show patients of Asian or Asian-Indian origin have a 2-fold increase in median exposure when compared to a Caucasian control group (5,6). The maximum dose of 40 mg is actually contraindicated in these patients (5). Interestingly, although Asians achieve similar benefits compared to European Americans at lower statin doses, both in terms of LDL-C lowering efficacy and decreased adverse cardiovascular events, evidence to date shows no increased rates of adverse events (myopathy, rhabdomyolysis) in Asian patients taking either lower or higher doses of statins (7).

MUMJ Volume 17 No. 1, pp. 57-62

June 2020

Table 1 shows the risk factors that have been associated with statin-associated muscle symptoms, including myalgia and rhabdomyolysis (5,6). In this case, the patient had multiple risk factors including advanced age, frailty, Chinese race, and excessive physical exercise, hence should have been started on a lower dose (5 mg) of rosuvastatin and titrated up as tolerated. Nonstatin lipid-lowering agents, such as proprotein convertase subtilisin-kexin type 9 (PCSK9) inhibitors evolocumab or alirocumab, may also be considered. This case report highlights the need to consider each patientâ&#x20AC;&#x2122;s risk profile for rhabdomyolysis and individualize dosage appropriately when prescribing statins. Educating patients and caregivers about symptoms is also important for prompt medical treatment.

MUMJ Volume 17 No. 1, pp. 57-62

June 2020

Table 1. Risk factors for statin-associated muscle symptoms. Risk factor

Comment

Pre-existing or family history of muscular disorders

Amyotrophic lateral sclerosis, myasthenia gravis, metabolic myopathy (carnitine palmitoyltransferase 2 deficiency or McArdleâ&#x20AC;&#x2122;s disease)

Previous history of muscle toxicity with another statin

Hypersensitivity syndrome that may include anaphylaxis, angioedema, inflammatory conditions such as polymyalgia rheumatica, fever, chills, dyspnea

Concurrent drug therapy

CYP3A4 inhibitors, fibrates, fusidic acid, niacin

Hypothyroidism Hepatic impairment or alcohol abuse

Active liver disease or unexplained persistent elevations of serum transaminases exceeding 3 times the upper limit of normal. In patients with severe liver disease (Child-Pugh 8 and 9), systemic exposure was increased by at least 2-fold compared to patients with lower Child-Pugh scores.

Renal impairment

Creatinine clearance < 30 mL/min/1.73 m2. Patients have a 3-fold increase in plasma concentration compared to healthy volunteers

Advanced age

Elderly patients >70 years are more susceptible to myopathy

Frailty

Small body frame

Excessive physical exercise Diabetes with hepatic fatty change Surgery and trauma Special populations

Filipino, Chinese, Malaysian, Japanese, Korean, Vietnamese or Asian-Indian patients have around a 2-fold increase in serum concentration

Declaration of Interests The authors have no potential conflicts of interest to disclose. Consent Written consent for this case report was given by the patient's daughter on August 27, 2019.

MUMJ Volume 17 No. 1, pp. 57-62

June 2020

References 1.

Anderson TJ, Grégoire J, Pearson GJ, Barry AR, Couture P, Dawes M, et al. 2016 Canadian Cardiovascular Society guidelines for the management of dyslipidemia for the prevention of cardiovascular disease in the adult. Can J Cardiol. 2016 Nov;32(11):1263– 82.

Graham DJ, Staffa JA, Shatin D, Andrade SE, Schech SD, Grenade LL, et al. Incidence of hospitalized rhabdomyolysis in patients treated with lipid-lowering drugs. JAMA. 2004 Dec 1;292(21):2585–90.

Bosch X, Poch E, Grau JM. Rhabdomyolysis and acute kidney injury. N Engl J Med. 2009 Jul 2;361(1):62–72.

McMahon GM, Zeng X, Waikar SS. A risk prediction score for kidney failure or mortality in rhabdomyolysis. JAMA Intern Med. 2013 Oct 28;173(19):1821.

AstraZeneca Canada Inc. Crestor: rosuvastatin calcium product monograph. [Internet]. 2017 [cited 2019 Oct 23]. Available from: https://www.astrazeneca.ca/content/dam/azca/downloads/productinformation/crestor-product-monograph-en.pdf

Lee E, Ryan S, Birmingham B, Zalikowski J, March R, Ambrose H, et al. Rosuvastatin pharmacokinetics and pharmacogenetics in white and Asian subjects residing in the same environment. Clin Pharmacol Ther. 2005 Oct;78(4):330–41.

Liao JK. Safety and efficacy of statins in Asians. Am J Cardiol. 2007 Feb;99(3):410–4.

MUMJ Volume 17 No. 1, pp. 63-70

June 2020

Case Report

Indolent T-cell lymphoproliferative disorder of the gastrointestinal tract managed conservatively with corticosteroids: A case report Tyler McKechnie, BSc1, Haroon Yousuf, MD FRCPC2, and Stephen Somerton, MD FRCPC3 1 Michael G. DeGroote School of Medicine, McMaster University, Hamilton, Ontario, Canada. 2 Department of Medicine, McMaster University, Hamilton, Ontario, Canada. 3 Gastroenterology, Brant Community Healthcare System, Brantford, Ontario, Canada.

Abstract Introduction: Indolent T-cell lymphoproliferative disorder (TCLPD) of the gastrointestinal (GI) tract is recognized as a provisional entity under the World Health Organization. It most often presents with chronic diarrhea. The diagnosis relies on clinical and endoscopic (macroscopic and microscopic) findings. Treatment regimens are variable but often include chemotherapeutic agents. Case: An 82-year-old female presented with a 4-week history of abdominal pain, weight loss, diarrhea, and nausea. A complete infectious workup was negative. Her computed tomography (CT) scan showed no pathologic changes and her esophagogastroduodenoscopy (EGD) showed mucosal erosion in the duodenum. Her duodenal biopsy demonstrated a marked increase in intraepithelial lymphocytes and her immunohistochemistry was consistent with indolent TCLPD of the GI tract. Management: She was started on high dose prednisone three months after the onset of her symptoms. She gradually improved with complete resolution of erosive changes on her repeat EGD. The prednisone was gradually tapered over three months. One month following the completion of the taper, she had recurrent symptoms. She was thus kept on low dose prednisone for two months. Conclusions: She is the oldest known patient to be diagnosed with indolent TCLPD of the GI tract, thus prompting reconsideration of patient populations most at risk of this disease. Moreover, she represents the first case with complete resolution of macroscopic disease with corticosteroid treatment alone. Keywords: Internal medicine, Gastroenterology, Oncology, Geriatrics Corresponding author: tyler.mckechnie@medportal.ca

MUMJ Volume 17 No. 1, pp. 63-70

June 2020

Introduction Lymphoma is the malignant transformation of lymphocytes. Primary gastrointestinal (GI) lymphomas are generally aggressive neoplasms that carry significant morbidity and mortality (1). Recently, however, cases of primary GI indolent lymphomas have been described. These entities are less aggressive and likely do not require the same aggressive chemotherapeutic approaches that primary GI lymphomas do (2). It is now recognized as a provisional entity in the World Health Organization (WHO) under lymphoid neoplasms (3). Indolent T-cell lymphoproliferative disorder (TCLPD) of the GI tract may present with abdominal pain, diarrhea, vomiting, or dyspepsia (2). The exact pathophysiology is yet to be determined but it is thought to be related to immune dysregulation (1). The diagnosis of indolent TCLPD relies on identifying clinical features consistent with the diagnosis (e.g. abdominal pain, diarrhea), macroscopic endoscopic findings (e.g. mucosal erythema or ulceration), and microscopic endoscopic findings (e.g. intra-epithelial lymphocytes (IELs), low rate of proliferation, CD4/8 staining) (1). Current treatment modalities include cytotoxic agents (e.g. cyclophosphamide, vinblastine, gemcitabine) and corticosteroids. Case report Mrs. EA is an 82-year-old female with a 4-week history of diffuse abdominal pain, weight loss, diarrhea, and nausea without vomiting. She described experiencing up to 15-20 clay-coloured stools, with no rectal bleeding, hematochezia, melena stool, or nocturnal diarrhea. She reported decreased oral intake secondary to nausea. Her past medical history was significant for dyslipidemia, hypertension and squamous cell carcinoma of the lip, which was resected in 2013. Her only medications were hydrochlorothiazide and telmisartan for her hypertension. Her family and social history were non-contributory. She was admitted to hospital and given intravenous rehydration. A complete infectious workup including stool Clostridium difficile and norovirus PCR, stool culture and stain (C&S), and stool ova and parasites (O&P) was negative. The only abnormalities on routine blood work were a low potassium of 3.2mmol/L (normal 3.5-5mmol/L), an elevated erythrocyte sedimentation rate (ESR), an elevated C-reactive protein (CRP), and an elevated lactate dehydrogenase (LDH). She was initially trialed on a gluten-free diet for one month with no abatement of symptoms. She was then restarted on a gluten-containing diet and her celiac markers and anti-nuclear antibody (ANA) were measured. They failed to demonstrate any abnormalities. Her computed tomography (CT) scan of the abdomen and pelvis showed no obvious pathology. Due to her weight loss of close to 12lbs, she was started on total parenteral nutrition (TPN). She subsequently had an esophagogastroduodenoscopy (EGD) which showed erosive damage in the duodenum and a colonoscopy which was normal. Her duodenal biopsy showed inflammation with marked increase in intra-epithelial lymphocytes (IEL). The immunohistochemistry demonstrated lymphocytes positive for CD3, CD4, CD8 and negative for CD56 and CD57 with a monoclonal gene rearrangement of the T65

MUMJ Volume 17 No. 1, pp. 63-70

June 2020

cell gamma receptor (Figure 1). The macroscopic and microscopic mucosal abnormalities ruled out the important differential diagnoses of inflammatory bowel disease (IBD) and Celiac disease (CD), while the CD staining ruled peripheral T-cell lymphoma (PTCL). Thus, these findings were most consistent with a diagnosis of indolent TCLPD of the GI tract. She was started on high dose prednisone roughly three months after the onset of her symptoms. She was subsequently prescribed a slow prednisone taper over three months and her symptoms gradually improved. Her repeat EGD displayed grossly normal duodenum and resolution of the previous erosive changes. One month following completion of the taper, she experienced recurrence of her nausea and diarrhea. A low dose prednisone was started, and she again experienced resolution of her symptoms. She was kept on low dose prednisone for two months, after which she tolerated a taper without recurrent symptoms.

Figure 1. Hematoxylin and eosin (H&E) stains demonstrating dense infiltration of atypical lymphocytes in mucosa (A, B) and immunohistochemistry of these lymphocytes (C, D, E).

Discussion Indolent CD4+/CD8+ TCLPD of the GI tract is a rare clinical entity, with 33 cases diagnosed and reported in the literature over the past 26 years (Table 1). The relatively recent description of this entity is likely a result of advancing diagnostic approaches (i.e., high-definition endoscopy,

MUMJ Volume 17 No. 1, pp. 63-70

June 2020

specific immunohistochemistry staining) (1). Currently the pathogenesis is unknown, but it is speculated that this disease is related to immune dysregulation, inflammation, and persistent antigenic stimulation (1). The lymphocytic derangements in this disease are consistent with this, given the immunomodulating effect of CD8 T-cells in the GI tract (4). The diagnosis of indolent TCLPD of the GI tract can be challenging as chronic diarrhea effects up to 5% of North Americans at any given time and has a vast differential diagnosis (5). This was evident in this patientâ&#x20AC;&#x2122;s case, as her diagnosis of indolent TCLPD of the GI tract was made three months after symptom onset. Indolent TCLPD of the GI tract is often initially diagnosed as PTCL, IBD, or CD (2). Distinguishing indolent TCLPD from these entities is important as it can prevent unnecessary therapy and potential adverse effects related to the therapy. For example, 50% of patients with indolent TCLPD of the GI tract in one case series received chemotherapy that did not change their disease course, yet exposed them to potential adverse effects such as hair loss, immunosuppression, and anemia see Table 1 for a full list of treatments (2). This case of indolent TCLPD of the GI tract is unique for a number of reasons. First, she is one of three patients treated with corticosteroids alone (5). The previous two cases did not have complete resolution of their macroscopic GI tract abnormalities. Although corticosteroids carry risk of adverse events, the severity and frequency of these side effects are less than in chemotherapy. For example, cyclophosphamide, a chemotherapeutic agent used in the management of previously reported cases of indolent TCLPD of the GI tract, is associated with severe adverse effects that steroids are not, such as hemorrhagic cystitis, alopecia, and febrile neutropenia (6). Moreover, the clinical response is likely comparable between the two (5). These considerations served as the basis for treating her solely with corticosteroids, and are important discussion points for future physicians when managing this disease. Second, Mrs. EA was 84 years old on presentation, making her the oldest reported patient with indolent TCLPD of the GI tract. The mean age in previously reported cases was 46.9 (range: 15-77). Third, indolent TCLPD of the GI tract affects a larger proportion of males â&#x20AC;&#x201C; of the 33 cases reported, 22 have been male. Taken together, this case serves as a reminder that demographic factors alone (i.e. age, gender) do not preclude the diagnosis of indolent TCLPD of the GI tract. Mrs. EAâ&#x20AC;&#x2122;s immunohistochemistry results were also remarkable. The results demonstrated marked intra-epithelial lymphocytes (IELs), which was only observed in three other patients with indolent TCLPD (2,7,8). Thus, increased IELs on immunohistochemistry, although less common in indolent TCLPD of the GI tract, does not rule out the diagnosis altogether. Whether this clinical entity is truly a rarity or simply underdiagnosed is yet to be determined (5). Regardless, it is necessary that clinicians consider the diagnosis of indolent TCLPD of the GI tract in patients with chronic diarrhea once more common pathologies (e.g., PTCL, IBD, CD) have been ruled out. While endoscopic investigations (i.e., EGD and colonoscopy) are often performed in the initial workup of patients presenting chronic diarrhea, the threshold for biopsy and immunohistochemistry should be low in patients with duodenal mucosal abnormalities. Investigations should include endoscopy (EGD and colonoscopy) with biopsy and immunohistochemistry. Such an approach may reduce the time to diagnosis and

MUMJ Volume 17 No. 1, pp. 63-70

June 2020

decrease patient exposure to unnecessary therapy. Therapy with corticosteroids alone is likely sufficient in managing this disease, thus correct diagnosis early within the course of the disease may limit exposure to more aggressive therapies such as chemotherapy. It is likely, however, that a long course of corticosteroids (i.e., longer than 3 months) is required for sufficient resolution of GI tract abnormalities and symptoms. Further investigation is required to elucidate the pathogenesis of this disease in order to develop more targeted treatment. Table 1. Summary of previous TCLPD case reports Author, Year

Patient(s)

Endoscopy Findings

Histology

Immunohistochemistry

Management

Mendes et al. (2014)(7)

1 Female; 59 years of age; clinical presentation of diarrhea, weight loss

N/A

•

CD4+, CD8-, CD2+, CD3+, CD7+, CD103-

Gemcitabine, prednisone

5 Males, 5 Females; 22-68 years of age; clinical presentation of diarrhea, weight loss

N/A

CD3+, CD8-

Anti-CD52, vinblastine

6 Males, 4 Females; 15-77 years of age; clinical presentation of diarrhea (8/10), abdominal pain (6/10), oral ulcers (4/10)

Numerous small polyps, erosions, and erythema in duodenum

2 Males, 1 Female; 37-53 years of age; clinical presentation of diarrhea (3/3), weight loss (3/3), night sweats (1/3)

Nodular mucosa in duodenum

1 Male; 42 years of age; clinical presentation of gastritis symptoms

Atrophy of small bowel mucosa

Malamut et al. (2014)(9)

Perry et al. (2013)(2)

Margolskee et al. (2013)(5)

Leventaki et al. (2013)(10)

Zivny et al. (2004)(11)

1 Male; 60 years of age; clinical presentation of diarrhea, weight loss

•

• • •

Gastric and duodenal erythema

• •

•

Partial villous atrophy Small lymphocytes in LP of stomach, SI, and colon IELs increased Subtotal villous atrophy Small lymphocytes in LP of stomach, SI, and colon IELs not increased Dense, nondestructive lymphoid infiltrate in LP of stomach, small bowel, and colon IELs increased in 1/10 patients Partial villous atrophy and small lymphocytes in the LP of the stomach, small bowel, and colon IELs not increased Small lymphocytes in the LP of esophagus, stomach, small bowel, and colon IELs not increased Villous blunting, and small lymphocytes in the LP of the stomach and small bowel IELs not increased

CD4+,

10/10 CD3+, CD4-, CD56-; 8/10 CD8+, CD5+; 7/7 CD2+

• • •

CD2+, CD3+, CD4+, CD8-

• • •

Watch and wait (4/10) Multiple bowel resections (1/10) CHOP (5/10)

Budesonide (3/3) Azathioprine (1/3) Prednisone (1/3)

CD2+, CD3+, CD8+, CD4-, CD5-, CD56-

Interferon isotretinoic acid, steroids

CD3+, CD4+, CD8-, CD103-

Prednisone, vincristine, cyclophosphamide

MUMJ Volume 17 No. 1, pp. 63-70

June 2020

(Table 1, continued. Summary of previous TCLPD case reports.) Author, Year

Patient(s)

Endoscopy Findings

Histology

Immunohistochemistry

Management

Tsutsumi et al. (1996)(12)

2 Male; 48-68 years of age; clinical presentation of diarrhea, weight loss, leg edema

Irregular granular mucosa in small bowel

•

Small lymphocytes in LP of small bowel IELs not increased

CD2+, CD3+, CD5+, CD8+, CD4+/(1/2 patients were CD4+, CD8+)

None

1 Male; 47 years of age; asymptomatic

Polypoid gastric and small bowel mucosa, aphthoid lesions in the large bowel

• •

Normal mucosa Small lymphocytes in LP of small bowel, colon, and rectum IELs not increased

CD2+, CD4+, CD103+

CD3+, CD8-,

Cyclophosphamide, vindesine, pirarubicin, prednisone

1 Male: 51 years of age; clinical presentation of relapsing oral and colorectal ulcers, periumbilical abdominal pain

Small ulcers in oropharynx, terminal ileum, and colon

•

Diffuse and dense infiltration of small lymphocytes in LP of oral cavity, colon, and rectum IELs increased

CD3+, CD8-

CD4+,

Prednisone, salicylazosulfapyridine

3 Males and 1 Female; 28-59 years of age; clinical presentation of diarrhea, weight loss

No gross findings

•

Extensive infiltration of small lymphocytes in LP of stomach and small bowel IELs not increased

CD2+, CD4+, CD7+, CD56-, CD103-

CD3+, CD5+, CD8-, CD57-,

MACOP-B, holoxane, etoposide, teneposide, doxorubicin, chlorambucil

Hirakawa et al. (1996)(13)

Egawa et al. (1995)(8)

Carbonnel et al. (1994, 1999)(14,15)

•

(N/A, not applicable; LP, lamina propria; IEL, intra-epithelial lymphocyte; CD, cluster of differentiation; CHOP, cyclophosphamide, doxorubicin, hydrochloride, vincristine, prednisone; MACOP-B, methotrexate, leucovorin, doxorubicin, cyclophosphamide, vincristine, prednisone, bleomycin)

MUMJ Volume 17 No. 1, pp. 63-70

June 2020

References 1.

Matnani R, Ganapathi KA, Lewis SK, Green PH, Alobeid B, Bhagat G. Indolent T- and NK-cell lymphoproliferative disorders of the gastrointestinal tract: A review and update. Hematol Oncol. 2017;35(1):3–16.

Perry AM, Warnke R a, Hu Q, Gaulard P, Copie-bergman C, Alkan S, et al. Indolent T-cell lymphoproliferative disease of the gastrointestinal tract Indolent T-cell lymphoproliferative disease of the gastrointestinal tract. Blood. 2013;122(22):3599–606.

Swerdlow SH, Campo E, Pileri SA, Harris NL, Stein H, Siebert R, et al. The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood. 2017;127(20):453–62.

Grdic D, Hörnquist E, Kjerrulf M, Lycke NY. Lack of local suppression in orally tolerant CD8-deficient mice reveals a critical regulatory role of CD8+ T cells in the normal gut mucosa. J Immunol. 1998;160(2):754–62.

Margolskee E, Jobanputra V, Lewis SK, Alobeid B, Green PHR, Bhagat G. Indolent small intestinal CD4+ T-cell lymphoma is a distinct entity with unique biologic and clinical features. PLoS One. 2013;8(7).

Fraiser L, Kanekal S, Kehrer J. Cyclophosphamide Toxicity. Drugs. 1991;42:781–95.

Mendes L, Attygalle AD, Cunningham D, Benson M, Andreyev J, Gonzales-de-Castro D, et al. CD4-positive small T-cell lymphoma of the intestine presenting with severe bile-acid malabsorption: A supportive symptom control approach. Br J Haematol. 2014;167(2):265– 9.

Egawa N, Fukayama M, Kawaguchi K, Hishima T, Hayashi Y, Funata N, et al. Relapsing oral and colonic ulcers with monoclonal T‐cell infiltration. A low grade mucosal T‐ lymphoproliferative disease of the digestive tract. Cancer. 1995;75(7):1728–33.

Malamut G, Meresse B, Kaltenbach S, Derrieux C, Verkarre V, Macintyre E, et al. Small intestinal CD4+ T-cell lymphoma is a heterogenous entity with common pathology features. Clin Gastroenterol Hepatol [Internet]. 2014;12(4):599-608.e1. Available from: http://dx.doi.org/10.1016/j.cgh.2013.11.028

10.

Leventaki V, Manning JT, Luthra R, Mehta P, Oki Y, Romaguera JE, et al. Indolent peripheral T-cell lymphoma involving the gastrointestinal tract. Hum Pathol [Internet]. 2014;45(2):421–6. Available from: http://dx.doi.org/10.1016/j.humpath.2013.08.003

11.

Zivny J, Banner BF, Agrawal S, Pihan G, Barnard GF. CD4+T-cell lymphoproliferative disorder of the gut clinically mimicking celiac sprue. Dig Dis Sci. 2004;49(4):551–5.

12.

Tsutsumi Y, Inada KI, Morita K, Suzuki T. T-cell lymphomas diffusely involving the intestine: Report of two rare cases. Jpn J Clin Oncol. 1996;26(4):264–72.

13.

Hirakawa K, Fuchigami T, Nakamura S, Daimaru Y, Ohshima K, Sakai Y, et al. Primary gastrointestinal T-cell lymphoma resembling multiple lymphomatous polyposis. Gastroenterology. 1996;111(3):778–82. 70

MUMJ Volume 17 No. 1, pp. 63-70

June 2020

14.

Carbonnel F, D’Almagne H, Lavergne A, Matuchansky C, Brouet JC, Sigaux F, et al. The clinicopathological features of extensive small intestinal CD4 T cell infiltration. Gut. 1999;45(5):662–7.

15.

Carbonnel F, Lavergne A, Messing B, Tsapis A, Berger R, Gralian A, et al. Extensive small intestinal T-cell lymphoma of low-grade malignancy assocaited with a new chromosomal translocation. Cancer. 1994;73(4):1286–91.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

Review Article

Universal vaccines against influenza viruses: Overview of the past, present, and prospective Yonathan Agung, Hannah Stacey, Michael Dâ&#x20AC;&#x2122;Agostino, and Ali Zhang McMaster University, Department of Biochemistry and Biomedical Sciences

Abstract Influenza is a common disease caused by influenza virus infections. There are an estimated 3 to 5 million annual cases of severe illness and 290 000 to 650 000 respiratory deaths caused by influenza viruses worldwide. Although antiviral drugs are available to treat influenza, vaccination remains the best infection prevention modality. However, current influenza vaccines provide a narrow range of protection and limited efficacy against seasonal and pandemic virus strains. Due to these limitations, novel vaccines that bestow broad protection and demonstrate a high level of efficacy against seasonal and pandemic viruses are desperately needed. The development of several universal influenza vaccines which target conserved epitopes such as the hemagglutinin stalk domain, neuraminidase, and the matrix 2 proton channel have made significant strides in this field. This article provides an overview of promising universal influenza virus vaccine designs, as well as current universal influenza vaccine clinical trials. Keywords: Influenza virus, Neuraminidase, Matrix protein

Universal

vaccine,

Seasonal

vaccine,

Hemagglutinin,

Corresponding author: Ali.Zhang@medportal.ca

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

Classification of influenza viruses Influenza viruses are enveloped RNA viruses that are divided into four different groups: A, B, C, and D, based on their antigenic similarity (1). Both group A and B influenza viruses cause yearly seasonal epidemics. Group A viruses are the only viral class known to have caused pandemics thus far (2). Transmission in humans occurs in three ways: direct contact with an infected person, through fomites, or by inhaling aerosolized infectious particles (3). Group C influenza viruses cause mild infections in humans, but do not contribute to the seasonal epidemics (4). Group D influenza viruses primarily infect cattle, and currently evidence demonstrates that these viruses are not able to infect humans (5). Group A viruses are characterized based on the subtypes of the two major surface proteins expressed: hemagglutinin (HA) and neuraminidase (NA). For example, H3N2 viruses express HA subtype 3 and NA subtype 2. In contrast, group B viruses are divided into two lineages: Yamagata-like and Victoria-like based on their sequence similarity to the ancestral B/Yamagata/16/88 or B/Victoria/2/87 strains, respectively (6). Because of their clinical relevance, the remainder of this review will focus on human influenza A and B viruses. Influenza viruses have a negative sense, single-stranded RNA genome consisting of 8 segments. These segments encode one or more viral proteins. In influenza A viruses, RNA segment 4 encodes HA, while RNA segment 6 encodes NA (7). Due to the lack of proofreading function by the viral RNA-dependent RNA polymerase, influenza virus genome is prone to mutations, causing the virus to mutate on average one nucleotide per genome per infectious cycle, which requires only 6 hours for completion (8–11). If mutations result in amino acid substitutions in either HA or NA, binding capacity of pre-existing antibodies may be diminished leading to decreased viral recognition by the host. These viruses tend to have a selective advantage and become the dominant circulating strain in a process referred to as “antigenic drift” (12). Due to the segmented nature of the influenza virus genome, co-infection of a single cell with multiple different strains of influenza viruses may cause the emergence of reassortant viruses. These reassortant viruses arise when the segmented genomes of multiple viruses undergo recombination in progeny - a process called “antigenic shift” (13). This reassortment may result in novel viruses that are well-adapted for infection and transmission in humans, but contain significantly altered glycoproteins that the majority of the human population have not previously encountered (13). Thus, antigenic drift typically results in seasonal epidemic strains, while antigenic shift is responsible for pandemic influenza virus strains capable of causing global pandemics (13). Influenza virus surface proteins and antigens Influenza virus particles consist of a lipid membrane that is studded with viral surface proteins, including the aforementioned HA and NA. HA is also one of the main antigenic targets for protective antibodies generated against the virus following infection or vaccination. HA is composed of head and stalk domains; the globular head is connected to the viral membrane by the stalk (14). The function of the HA head domain is to bind to sialic acid on host cells and initiate infection, while the HA stalk domain undergoes complex conformational changes to mediate the 73

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

viral fusion process within the cell (14) (Figure 1). HA also undergoes enzymatic cleavage by host proteases to reach its active conformation, and thus, allowing progeny virus to become infectious (15). Most antibodies raised by seasonal influenza vaccines are directed towards the HA head

Figure 1. The influenza virus life cycle begins with the attachment to host cells and ends with the release of viral progeny. The first stage of the viral life cycle, binding, is mediated by HA. HA binds to Îą 2-6 sialic acids on host cells in the upper respiratory tract. Interaction with this glycan initiates fusion with the host cell plasma membrane. The virus enters the cell via endocytosis where the endosome containing the viral particle becomes acidified leading to the uncoating and release of the viral ribonucleoproteins (RNP) segments. Following RNP transport into the nucleus, replication of the viral genome begins. As influenza viruses are negative sense RNA viruses, positive-sense mRNA must first be transcribed for the generation of more viral RNPs. The replication of the viral RNA genome also occurs in the nucleus. The viral genome and proteins come together in the cytosol for assembly, which is followed by budding at the plasma membrane. The final stage of the viral life cycle, release, is facilitated by NA. NA cleaves the HA:sialic acid interactions to allow for the release of mature virions which in turn, infect other cells.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

domain. Because the HA head domain is distal to the virus particle, this region is easily accessible by the immune system. However, this domain is also prone to mutation and thus is highly variable between different strains of influenza virus (16). In contrast, the HA stalk undergoes fewer genetic changes. This is perhaps due to decreased immunological selective pressure, and the need for sequence conservation to allow for complex conformational changes; necessary for the membrane fusion process (16). Antibodies targeting the head domain are therefore more likely to be strainspecific, whereas anti-stalk antibodies recognize multiple different strains of influenza virus. Other proteins embedded in the membrane of influenza viruses include NA and the matrix2 (M2) proton channel. NA is an enzyme that cleaves terminal sialic acid residues from glycoproteins. This process is critical to release nascent virus particles, as HA remains bound to sialic acid residues during budding (Figure 1) (17). Other functions of NA include cleavage of mucins to allow the virus particles to access target cells in the respiratory tract, and binding to receptors on host cells to mediate endocytosis and internalization (17). While the virion is trapped in the endosome following receptor-mediated endocytosis, the M2 proton channel acidifies the interior of the virus capsid (18). The decrease in pH due to endosome acidification, mediates dissociation of the viral genome from the viral capsid, allowing for the release of the ribonucleoproteins into the cytosol after membrane fusion (19) (Figure 1). Vaccination and antiviral therapies to prevent and treat influenza in Canada In Canada, antiviral therapy is recommended for individuals belonging to groups with high risk of complications, such as adults 65 years of age and older, pregnant women, and individuals with severe or complicated influenza who require hospital admission or demonstrate severe symptoms. Additionally, it is used to treat or prevent influenza outbreaks in institutional settings (20). NA inhibitors, which primarily prevent influenza virus egress and budding from infected cells, are the only class of antiviral drugs approved for use in Canada (21). NA inhibitors include oral oseltamivir (Tamiflu), inhaled zanamivir (Relenza), and intravenous peramivir (Rapivab). Since 2006, amantadine and rimantadine, which are M2 proton channel antagonists, are no longer recommended due to widespread resistance in clinical isolates (22). A selective cap-dependent endonuclease inhibitor, Baloxivir Marboxil (Xofluza) was recently approved in the United States (but not Canada) for the treatment of acute uncomplicated influenza in individuals 12 years and older, or those with high risk of complications (23,24). NA inhibitors and cap-dependent endonuclease inhibitors are similarly effective at alleviating influenza symptoms approximately 24 hours sooner compared to placebo when administered within 48 hours of symptom onset (23,25â&#x20AC;&#x201C;27). Although antivirals are somewhat effective at both treating and preventing influenza, these drugs can cause considerable side effects, such as nausea, vomiting, and diarrhea. Similar to antibacterial drugs, antiviral medications can also be rendered ineffective by the emergence of resistant strains of influenza virus (21,26). Seasonal influenza vaccination is currently the best way to prevent influenza viral infections. Several formulations of influenza virus vaccines are clinically approved for use in 75

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

Canada. These formulations differ based on the following four variables: number of strains, effective dosage, method of virus inactivation, and the inclusion of an adjuvant (29). The seasonal influenza vaccine includes three or four strains of influenza viruses as recommended by the World Health Organization (WHO) approximately 6 months prior to the beginning of the flu season. In the 2019-2020 season, trivalent vaccines contain an H1N1 (A/Brisbane/02/2018-like), H3N2 (A/Kansas/14/2017-like), and a Victoria-like (B/Colorado/06/2017-like) strain, while quadrivalent vaccines contain the aforementioned trivalent strains along with an additional Yamagata-like (B/Phuket/3073/2013-like) strain (30). Quadrivalent vaccines are generally recommended over trivalent vaccines, if both are available, due to the broader range of protection offered (29). High dose vaccines contain four times the amount of antigen compared to their standard dose counterparts, and are reserved for those who are above the age of (29,31). The viruses found within vaccines can be inactivated or attenuated in one of three ways. In split vaccines, viruses are disrupted by a detergent, while subunit vaccines are further processed to purify the antigens of interest, largely HA and NA (32). Live attenuated vaccines are composed of viruses that are adapted to replicate at a lower temperature (25Ë&#x161;C) and therefore have decreased virulence in humans. Live attenuated vaccines are reserved for children 2-17 years in the form of a nasal spray, while all other vaccines are delivered intramuscularly (29). Adjuvanted vaccines are made with MF59, a proprietary adjuvant that uses squalene, a long hydrophobic molecule, to form an oil-in-water emulsion. This allows for the elicitation of stronger immune responses to vaccination (33). Adjuvanted vaccines are available for children 6-23 months and adults 65 years and older; as a means to improve immunogenicity in those who typically respond poorly to vaccination attributing to an immature immune system or immunosenescence (29). The majority of seasonal influenza viral vaccines are manufactured using embryonated chicken eggs (34). Although this method has been used for over 50 years there are several major drawbacks. For example, in order to produce high titers of the candidate vaccine strain, viruses used in seasonal vaccines must be adapted to grow in chicken eggs (35). This process lengthens the production time of influenza virus vaccines relative to cell-based vaccine production (35). In turn, longer production times reduce the flexibility of manufacturing, necessitating that vaccine production begins long before the influenza season commences (34). Therefore, vaccine developers cannot alter vaccine formulations in response to the most recent mutations in circulating strains that occur after WHO recommendations have been made (34). This inability to adapt virus strains can result in further delays in vaccine production due to the low yield (36). Additionally, some virus strains, especially H3N2, grow poorly in eggs (37). Furthermore, some egg-based adaptations that occur during the manufacturing process may cause epitope mutations, resulting in poor vaccine efficacy due to inadequate congruency between the circulating and vaccine strains (36). With that being said, several shortcomings of egg-based vaccine production can be rectified through cell-based vaccine production. Cell-based vaccine production can be scaled up more quickly. In addition the viruses produced with this method are more similar to the seed strain, reducing the emergence of antigenic variants that can arise as a result of egg-based adaptations (36). Despite the advantages of cell-based production compared to egg-based

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

production, the latter continues to represent the majority of the influenza vaccine market due its cost effectiveness (36). While seasonal influenza vaccination is the best preventative measure against influenza virus infections currently, the protection provided is often transient and ineffective in subsequent influenza seasons. This is largely due to the aforementioned antigenic shift and antigenic drift (38). As a result, seasonal influenza virus vaccines must be reformulated and re-administered yearly for optimal protection against â&#x20AC;&#x153;driftedâ&#x20AC;? strains. In addition, pandemic strains that arise due to antigenic shift render any seasonal vaccines ineffective (38). The strain-specific nature of antibodies raised in seasonal vaccines highlights the need to develop vaccine platforms that offer both prolonged and more broad protection. Recent developments in the field of universal influenza vaccines show great potential in addressing these needs. These universal formulas do not require yearly reformulation due to their ability to elicit antibodies that target numerous influenza viral strains. One especially promising formulation involves generating immune responses targeting the HA stalk domain. Universal influenza vaccines targeting the HA stalk domain Seasonal influenza virus vaccination elicits the production of HA stalk binding antibodies that can neutralize multiple strains of influenza virus (39,40). These antibodies are termed broadly neutralizing antibodies (bNAbs) (40). bNAbs provide protection in vivo largely by activating immune effector cell functions such as antibody dependent cellular cytotoxicity (ADCC) and antibody dependent cellular phagocytosis (ADCP) (41,42). Additionally, bNAbs are capable of providing protection by blocking fusion of the viral envelope with the endosomal membrane, and preventing proteolytic activation of the HA protein (43). By targeting conserved epitopes found within the HA stalk domain, bNAbs elicited by universal influenza vaccines are capable of providing protection against diverse influenza virus strains and subtypes (44). Although the broad range of protection granted by bNAbs is clearly superior to the narrow range of protection offered by seasonal vaccines, bNAbs have notable limitations. For example, bNAbs have reduced neutralization potency compared to HA head binding antibodies (45). In microneutralization assays, multi-log differences in neutralization potency in favour of HA head binding antibodies over bNAbs were present when both were compared in a monoclonal context (45). However, when neutralization potency was measured in a polyclonal context, the difference in potency was reduced from a multi-log difference to only a 3-fold difference in favour of HA head binding antibodies. This ultimately suggests that the interactions between antibodies present in polyclonal serum of influenza-exposed adults boosts the neutralization efficacy of bNAbs (45). Seasonal vaccines and infections do not typically elicit high titers of stalk-binding antibodies (46). This is because the HA head domain sterically shields the stalk from being immunologically accessible, making it more challenging for bNAbs to be produced (47). Fewer antibodies are also produced against the stalk domain due to high affinity interactions restricted to select few immunoglobulin genes (specifically, VH1-69 and VH1-18) (47). This lowers the frequency of precursor B cells capable of making stalk antibodies. Lastly, bNAbs tend to be more 77

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

autoreactive, potentially causing peripheral tolerance mechanisms to inhibit the expansion of stalkspecific B cells (48). Fortunately, it has been well documented that vaccination with or exposure to pandemic-like influenza virus strains can lead to an immune response that preferentially produces bNAbs over HA head binding antibodies (Figure 2) (49â&#x20AC;&#x201C;52). This phenomenon has been recapitulated using sequential vaccination with viruses expressing chimeric HA (cHA) (53,54). Researchers have shown that sequential immunization of mice with cHA vaccine constructs induced bNAbs conferring protection against both group 1 and group 2 influenza A viruses (53,54). cHA constructs feature an HA head domain derived from an exotic influenza subtype not previously exposed to the human population and the HA stalk domain of a circulating influenza virus subtype (49,55). cHA vaccines result in primary immune responses towards both the HA head and stalk (49). Subsequent vaccinations use different cHA constructs with a homologous stalk domain from the first vaccination, but utilize radically different HA head domains (55). In doing so, the immune response shifts from the previously immunodominant HA head domain towards the stalk domain. cHA vaccines represent an effective method in redirecting immune responses towards the HA stalk domain to boost bNAb titers. An alternative method to induce high titers of stalk binding bNAbs is to use recombinant HA proteins that lack the head domain (56). Unfortunately, removal of the HA head domain destabilizes the protein and damages the neutralizing epitopes on the stalk domain (56). To improve stability of the constructs, two research groups experimented with the addition of leucine zipper motifs onto the HA stalk domains (57,58). Yassine et al. designed their recombinant HA stalk protein around the HA ectodomain of the virus A/New Caledonia/20/1999 and used a ferritin nanoparticle antigen-display platform to create the vaccine (58). This vaccine stimulated both antistalk antibody production and provided protection against lethal influenza virus challenge in mice and ferrets (58). Similarly, Impagliazzo et al. based their â&#x20AC;&#x153;mini-HAâ&#x20AC;? recombinant stalk domain around the HA sequence of A/Brisbane/59/2007 (57). This soluble protein stimulated high titers of broadly-reactive anti-HA antibodies in non-human primates, and protected mice from lethal influenza virus challenge (57). Additional strategies to create universal influenza vaccines The other major surface glycoprotein, NA, represents another promising candidate for universal influenza vaccines (59). Naturally acquired NA inhibiting (NAI+) antibodies protect against influenza infection, and NAI+ antibody titers positively correlate with vaccine effectiveness in both live attenuated and inactivated vaccines (60). NAI+ antibodies can function during the later stages of the viral life cycle relative to HA-neutralizing antibodies by mitigating viral infection through the prevention of viral budding from infected cells (Figure 1) (59). Lastly, NA contains contiguous antigenic domains; monoclonal antibodies against NA can recognize antigenic domains conserved between virus strains. This allows for cross reactivity of antibodies between viruses possessing the same NA subtype (61). Identification of these conserved epitopes in NA make it an intriguing target as a prospective universal influenza vaccine.

MUMJ Volume 17 No. 1, pp. 71-88

A )

June 2020

Head Stalk

Hemagglutinin (HA)

Seasonal HA

Drifted Seasonal HA

Pandemic HA

Chimeric HA #2

Chimeric HA #3

B )

Chimeric HA #1

Figure 2. The immune response to hemagglutinin can be redirected to the subdominant HA stalk following exposure to pandemic strains or sequential exposure to chimeric HA proteins. (A) HA is a surface glycoprotein composed of a highly variable head domain and a conserved, membrane proximal stalk domain. When an individual is exposed to an influenza virus, either through infection or vaccination, the immune system generates an antibody response largely targeting the HA head domain. Antibodies directed to the stalk domain are also generated but to a lesser extent. Upon exposure to a drifted virus strain, antibodies targeting novel head epitopes are produced and antibodies that target conserved epitopes between the previous strain and new drifted variant are boosted. However, upon exposure to a divergent HA, as was the case in 2009 during the Swine flu pandemic, researchers observed that individuals who had been exposed to this pandemic strain had high titers of stalk-binding antibodies. This was attributed to the conserved nature of the HA stalk domain. (B) Using this principle, recombinant chimeric HA (cHA) proteins were generated and a sequential vaccination strategy was employed to boost stalk antibody titers. These cHA proteins contained conserved HA stalk domains, but HA heads from avian influenza virus strains to which the human population has no pre-existing immune memory. Following repeat exposures, the response was directed against the sub-dominant stalk domain and only a weak primary immune response against the HA head was observed, generating small titers of head-specific antibodies. This sequential vaccination strategy has been shown to effectively induce high titers of broadly-neutralizing stalk antibodies in a variety of animal models.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

M2 is a transmembrane, homotetrameric proton ion channel involved in viral uncoating following cell entry and in the formation and budding of virus progeny (40,62,63). The extracellular domain of the M2 protein, M2e, is a highly conserved region in all influenza A viruses and, therefore, is a potential target for a universal influenza vaccine (63). The conservation found in M2e is due to its low immune reactivity, which translates to low selective pressure (63). M2especific antibodies, such as 14C2, provide protection by reducing the expression level of M2, which in turn inhibits formation of new viral particles and limits viral spread (63). Despite this, M2 itself is a very poor immunogen due to its small extracellular domain, membrane proximity, and relatively low abundance on the viral surface compared to HA and NA (64). Despite its theoretical ability to offer protection against many influenza A viruses, the aforementioned reasons pose significant barriers that any potential M2-based vaccines would have to overcome. Targeting T cell immunity serves as another potential method of universal influenza virus protection. Studies have shown that T cells mitigate the severity of influenza related illnesses and reduce viral shedding (65). Once infection has taken place, influenza virus-specific CD4+ T helper cells and CD8+ cytotoxic T cells are activated through the recognition of highly conserved epitopes found across influenza virus subtypes encoded in the viral nucleoprotein (NP) and matrix 1 protein (M1) (40). While CD8+ T cell mediated immunity is short lived, a vaccine capable of boosting the cross-reactive T cell responses towards these conserved antigens possesses the potential to provide a broad range of protection against influenza (65). A modified vaccinia virus Ankara (MVA) vector, MVA-NP+M1, expresses the conserved influenza antigens NP and M1 and may serve as a candidate universal influenza vaccine that boosts existing cross-reactive T cell response to these conserved internal antigens (66). T cell mediated immunity should provide a broader range of protection in comparison to antibodies that target the highly variable external glycoproteins (66). Current universal vaccine clinical trials Several universal influenza vaccine candidates are currently being evaluated in clinical trials. Peptide vaccines have shown some success, with the “Multimeric-001” universal vaccine entering phase III in August 2018 (67). This vaccine is composed of a recombinant protein containing 9 conserved linear epitopes of influenza virus proteins: 6 from HA, 3 from nucleoprotein (NP), and 1 from matrix protein (M1) (68). The Multimeric-001 vaccine induces influenza-specific cellular responses, such as IL-2 and IFN-γ secretion by T cells (67–69). Other peptide-based vaccines include Flu-v, which is composed of four equimolar mixtures of four polypeptides in the M1, M2, NP, and PB1 regions of influenza virus, and FP-01.1, which is comprised of six peptide chains of NP, M1, PB1 and PB2 linked to an inert fluorocarbon chain to increase in vivo half-life (70,71). Peptide vaccines are a relatively new and targeted approach towards eliciting a lasting immune response against specific epitopes. Peptides are readily altered and manufactured, making it possible for vaccine manufacturers to quickly adapt the formulation to match the circulating influenza virus strains. However, disadvantages including poor immunogenicity and stability must be overcome before these vaccines are viable alternative to conventional interventions. 80

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

Another class of vaccines currently being tested include M2e-based vaccines. Many M2ebased vaccines have been tested in the past, including VAX-102, which is composed of four M2e peptides linked to a toll like receptor 5 (TLR5) agonist to enhance the immune response (72). Although the vaccine induces high antibody levels against the M2e protein, it has been shown to cause considerable side effects such as fever, diarrhea, and fatigue at high doses (73,74). Currently, â&#x20AC;&#x153;Unifluâ&#x20AC;?, another M2e-based vaccine, comprised of the M2e protein fused to the hepatitis B viral core antigen, is being evaluated in a phase I clinical trial (75). The high degree of conservation of the M2e across influenza A virus subtypes makes this an attractive target for a universal influenza vaccine (74). However, antibodies against M2e are unable to neutralize virus directly, and therefore largely rely on immune effector cell mediated cytotoxicity to protect against infection (76). Additionally, MVA-NP+M1 vaccines have been, and are, currently being evaluated in clinical trials (66). As previously mentioned, MVA-NP+M1 vaccines utilize modified vaccinia virus Ankara to express a fusion protein of M1 and NP, which is used to boost the T cell response to conserved epitopes in these antigens (77). Phase I clinical trials have been conducted comparing the co-administration of seasonal influenza vaccine with the MVA-NP+M1 vaccine to administration of seasonal influenza vaccine solely in patients aged 50 and up (77). Results have indicated that co-administration was safe and tolerated in patients. In addition the T cell response to the conserved epitopes found on the internal antigens were boosted significantly in the group that received the MVA-NP+M1 vaccine when compared to the group that received the seasonal influenza vaccine alone (78). Currently, MVA-NP+M1 is undergoing a phase II clinical trial to assess its efficacy and immunogenicity as an adjunct to a standard, licensed dose of quadrivalent influenza vaccine in adults aged 18 and up (79). Recently, the first HA stalk based universal vaccine was developed and tested. This vaccine strategy is based around sequential vaccination with inactivated viruses expressing cHA, where the stalk domains are from conserved H1 or H3, and the head domains are from influenza viruses not yet exposed to humans. In this clinical trial, H8 and H5 head domains were used with an H1 stalk to create chimeric H8/1 and H5/1 viruses (80). Interim results of the phase 1 clinical trials showed that H1 stalk-specific IgG antibodies were boosted approximately 5-fold over baseline after two doses of the chimeric H8/1 and H5/1 vaccines with AS03, which is another squalene based adjuvant (80). These anti-stalk antibodies were boosted approximately 2-fold in groups receiving a series of vaccines with the adjuvant when compared to unadjuvanted formulations (80). As expected, the antibodies induced by the vaccine were similarly reactive against the stalk domains of H2, H9, and H18 (80). Stalk-based universal vaccines provide great promise at inducing high levels of stalk-reactive antibodies. However, it remains unknown if antibodies at these titers are protective against infection, and if the auto-reactive tendency of HA stalk-binding antibodies will cause any adverse reactions, especially in those with autoimmune diseases.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

Conclusion Although seasonal vaccination is the current gold standard for protection against influenza, influenza related illnesses and mortality rates remain high, placing a significant strain on todayâ&#x20AC;&#x2122;s global healthcare systems. Drawbacks to seasonal vaccines include occasional ineffectiveness against yearly influenza epidemics, inability to provide protection against pandemic strains, costintensive and time-consuming yearly vaccine manufacturing process, and the propensity for eggbased adaptations to occur. The development of several universal influenza vaccines that target conserved influenza virus epitopes provide promise for a more effective and reliable method of influenza prevention. This would eliminate the need for yearly vaccine reformulation and potentially prevent future influenza pandemics. While great strides have been made in the field of universal influenza vaccines, further animal studies and validation of immunogenicity and efficacy in humans through clinical trials are warranted.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

References 1.

Jang YH, Seong BL. The Quest for a Truly Universal Influenza Vaccine. Front Cell Infect Microbiol. 2019 Oct 10;9:344.

Saunders-Hastings P, Krewski D. Reviewing the History of Pandemic Influenza: Understanding Patterns of Emergence and Transmission. Pathogens. 2016 Dec 6;5(4):66.

Bridges CB, Kuehnert MJ, Hall CB. Transmission of influenza: implications for control in health care settings. Clin Infect Dis Off Publ Infect Dis Soc Am. 2003 Oct 15;37(8):1094– 101.

CDC. Types of Influenza Viruses [Internet]. Centers for Disease Control and Prevention. 2019 [cited 2019 Dec 18]. Available from: https://www.cdc.gov/flu/about/viruses/types.htm

Asha K, Kumar B. Emerging Influenza D Virus Threat: What We Know so Far! J Clin Med [Internet]. 2019 Feb 5 [cited 2019 Dec 18];8(2). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6406440/

Rota PA, Wallis TR, Harmon MW, Rota JS, Kendal AP, Nerome K. Cocirculation of two distinct evolutionary lineages of influenza type B virus since 1983. Virology. 1990 Mar;175(1):59–68.

Bouvier NM, Palese P. The Biology of Influenza Viruses. Vaccine. 2008 Sep 12;26(Suppl 4):D49–53.

Choi KH. Viral Polymerases. Adv Exp Med Biol. 2012;726:267–304.

Shao W, Li X, Goraya MU, Wang S, Chen J-L. Evolution of Influenza A Virus by Mutation and Re-Assortment. Int J Mol Sci [Internet]. 2017 Aug 7 [cited 2019 Dec 19];18(8). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5578040/

10. Parvin JD, Moscona A, Pan WT, Leider JM, Palese P. Measurement of the mutation rates of animal viruses: influenza A virus and poliovirus type 1. J Virol. 1986 Aug;59(2):377–83. 11. WHO/Europe. Virology of human influenza [Internet]. WHO/Europe. [cited 2020 Apr 13]. Available from: http://www.euro.who.int/en/health-topics/communicablediseases/influenza/data-and-statistics/virology-of-human-influenza 12. Boni MF. Vaccination and antigenic drift in influenza. Vaccine. 2008 Jul;26:C8–14. 13. Webster RG, Govorkova EA. Continuing challenges in influenza. Ann N Y Acad Sci. 2014 Sep;1323(1):115–39. 14. Skehel JJ, Wiley DC. Receptor binding and membrane fusion in virus entry: the influenza hemagglutinin. Annu Rev Biochem. 2000;69:531–69.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

15. Hamilton BS, Gludish DWJ, Whittaker GR. Cleavage Activation of the Human-Adapted Influenza Virus Subtypes by Matriptase Reveals both Subtype and Strain Specificities. J Virol. 2012 Oct;86(19):10579–86. 16. Kirkpatrick E, Qiu X, Wilson PC, Bahl J, Krammer F. The influenza virus hemagglutinin head evolves faster than the stalk domain. Sci Rep [Internet]. 2018 Jul 11 [cited 2019 Dec 25];8. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6041311/ 17. McAuley JL, Gilbertson BP, Trifkovic S, Brown LE, McKimm-Breschkin JL. Influenza Virus Neuraminidase Structure and Functions. Front Microbiol [Internet]. 2019 Jan 29 [cited 2019 Dec 21];10. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6362415/ 18. Pielak RM, Chou JJ. Influenza M2 proton channels. Biochim Biophys Acta. 2011 Feb;1808(2):522–9. 19. Helenius A. Unpacking the incoming influenza virus. Cell. 1992 May 15;69(4):577–8. 20. Antiviral Medications for Seasonal Influenza: Information for Health Care Providers, 2019 [Internet]. Queen’s Printer for Ontario; 2019. Available from: https://www.publichealthontario.ca/-/media/documents/qa-antiviral-medicationinfluenza.pdf?la=en 21. Moscona A. Neuraminidase Inhibitors for Influenza. N Engl J Med. 2005 Sep 29;353(13):1363–73. 22. Canada PHA of. Recommendation for Use of Amantadine for Treatment and Prevention of Influenza [Internet]. gcnws. 2006 [cited 2019 Dec 15]. Available from: https://www.canada.ca/en/news/archive/2006/11/recommendation-use-amantadinetreatment-prevention-influenza.html 23. Hayden FG, Sugaya N, Hirotsu N, Lee N, de Jong MD, Hurt AC, et al. Baloxavir Marboxil for Uncomplicated Influenza in Adults and Adolescents. N Engl J Med. 2018 06;379(10):913–23. 24. Influenza Antiviral Medications: Summary for Clinicians | CDC [Internet]. 2019 [cited 2019 Dec 15]. Available from: https://www.cdc.gov/flu/professionals/antivirals/summaryclinicians.htm 25. Jefferson T, Jones M, Doshi P, Spencer EA, Onakpoya I, Heneghan CJ. Oseltamivir for influenza in adults and children: systematic review of clinical study reports and summary of regulatory comments. BMJ [Internet]. 2014 Apr 9 [cited 2019 Dec 15];348. Available from: https://www.bmj.com/content/348/bmj.g2545 26. Heneghan CJ, Onakpoya I, Thompson M, Spencer EA, Jones M, Jefferson T. Zanamivir for influenza in adults and children: systematic review of clinical study reports and summary of regulatory comments. BMJ [Internet]. 2014 Apr 9 [cited 2019 Dec 15];348. Available from: https://www.bmj.com/content/348/bmj.g2547 84

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

27. Kohno S, Kida H, Mizuguchi M, Shimada J, S-021812 Clinical Study Group. Efficacy and safety of intravenous peramivir for treatment of seasonal influenza virus infection. Antimicrob Agents Chemother. 2010 Nov;54(11):4568–74. 28. Baz M, Abed Y, Papenburg J, Bouhy X, Hamelin M-È, Boivin G. Emergence of Oseltamivir-Resistant Pandemic H1N1 Virus during Prophylaxis. N Engl J Med. 2009 Dec 3;361(23):2296–7. 29. Canada PHA of. Canadian Immunization Guide Chapter on Influenza and Statement on Seasonal Influenza Vaccine for 2019–2020 [Internet]. aem. 2019 [cited 2019 Dec 25]. Available from: https://www.canada.ca/en/public-health/services/publications/vaccinesimmunization/canadian-immunization-guide-statement-seasonal-influenza-vaccine-20192020.html 30. WHO | Recommended composition of influenza virus vaccines for use in the 2019-2020 northern hemisphere influenza season [Internet]. WHO. [cited 2019 Dec 25]. Available from: http://www.who.int/influenza/vaccines/virus/recommendations/2019_20_north/en/ 31. Keitel WA, Atmar RL, Cate TR, Petersen NJ, Greenberg SB, Ruben F, et al. Safety of high doses of influenza vaccine and effect on antibody responses in elderly persons. Arch Intern Med. 2006 May 22;166(10):1121–7. 32. Wong S-S, Webby RJ. Traditional and New Influenza Vaccines. Clin Microbiol Rev. 2013 Jul 1;26(3):476–92. 33. O’Hagan DT. MF59 is a safe and potent vaccine adjuvant that enhances protection against influenza virus infection. Expert Rev Vaccines. 2007 Oct;6(5):699–710. 34. Harding AT, Heaton NS. Efforts to improve the seasonal influenza vaccine. Vaccines. 2018;6(2). 35. Ping J, Lopes TJS, Neumann G, Kawaoka Y. Development of high-yield influenza B virus vaccine viruses. Proc Natl Acad Sci U S A. 2016;113(51):E8296–305. 36. Ping J, Lopes TJS, Nidom CA, Ghedin E, MacKen CA, Fitch A, et al. Development of high-yield influenza A virus vaccine viruses. Nat Commun. 2015;6:1–15. 37. Skowronski DM, Janjua NZ, De Serres G, Sabaiduc S, Eshaghi A, Dickinson JA, et al. Low 2012-13 influenza vaccine effectiveness associated with mutation in the egg-adapted H3N2 vaccine strain not antigenic drift in circulating viruses. PLoS ONE. 2014;9(3). 38. Wong SS, Webby RJ. Traditional and new influenza vaccines. Clin Microbiol Rev. 2013;26(3):476–92. 39. Sautto GA, Kirchenbaum GA, Ross TM. Towards a universal influenza vaccine: different approaches for one goal. Virol J. 2018 19;15(1):17.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

40. Pica N, Palese P. Toward a universal influenza virus vaccine: prospects and challenges. Annu Rev Med. 2013;64:189–202. 41. DiLillo DJ, Palese P, Wilson PC, Ravetch JV. Broadly neutralizing anti-influenza antibodies require Fc receptor engagement for in vivo protection. J Clin Invest. 2016 Feb;126(2):605–10. 42. He W, Chen C-J, Mullarkey CE, Hamilton JR, Wong CK, Leon PE, et al. Alveolar macrophages are critical for broadly-reactive antibody-mediated protection against influenza A virus in mice. Nat Commun. 2017 10;8(1):846. 43. Brandenburg B, Koudstaal W, Goudsmit J, Klaren V, Tang C, Bujny M V., et al. Mechanisms of hemagglutinin targeted influenza virus neutralization. PLoS ONE. 2013;8(12). 44. Coughlan L, Palese P. Overcoming Barriers in the Path to a Universal Influenza Virus Vaccine. Cell Host Microbe. 2018;24(1):18–24. 45. He W, Mullarkey CE, Duty JA, Moran TM, Palese P, Miller MS. Broadly Neutralizing Anti-Influenza Virus Antibodies: Enhancement of Neutralizing Potency in Polyclonal Mixtures and IgA Backbones. J Virol. 2015;89(7):3610–8. 46. Krammer F, Palese P. Influenza virus hemagglutinin stalk-based antibodies and vaccines. Curr Opin Virol. 2013 Oct;3(5):521–30. 47. Neu KE, Henry Dunand CJ, Wilson PC. Heads, stalks and everything else: how can antibodies eradicate influenza as a human disease? Curr Opin Immunol. 2016;42:48–55. 48. Andrews SF, Huang Y, Kaur K, Popova LI, Ho IY, Pauli NT, et al. Immune history profoundly affects broadly protective B cell responses to influenza. Sci Transl Med. 2015 Dec 2;7(316):316ra192. 49. Pica N, Hai R, Krammer F, Wang TT, Maamary J, Eggink D, et al. Hemagglutinin stalk antibodies elicited by the 2009 pandemic influenza virus as a mechanism for the extinction of seasonal H1N1 viruses. Proc Natl Acad Sci U S A. 2012 Feb 14;109(7):2573–8. 50. Nachbagauer R, Salaun B, Stadlbauer D, Behzadi MA, Friel D, Rajabhathor A, et al. Pandemic influenza virus vaccines boost hemagglutinin stalk-specific antibody responses in primed adult and pediatric cohorts. Npj Vaccines. 2019 Dec;4(1):51. 51. Nachbagauer R, Wohlbold TJ, Hirsh A, Hai R, Sjursen H, Palese P, et al. Induction of Broadly Reactive Anti-Hemagglutinin Stalk Antibodies by an H5N1 Vaccine in Humans. J Virol. 2014 Nov 15;88(22):13260–8. 52. Ellebedy AH, Krammer F, Li G-M, Miller MS, Chiu C, Wrammert J, et al. Induction of broadly cross-reactive antibody responses to the influenza HA stem region following H5N1 vaccination in humans. Proc Natl Acad Sci. 2014 Sep 9;111(36):13133–8.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

53. Krammer F, Pica N, Hai R, Margine I, Palese P. Chimeric Hemagglutinin Influenza Virus Vaccine Constructs Elicit Broadly Protective Stalk-Specific Antibodies. J Virol. 2013 Jun 15;87(12):6542–50. 54. Krammer F, Margine I, Hai R, Flood A, Hirsh A, Tsvetnitsky V, et al. H3 Stalk-Based Chimeric Hemagglutinin Influenza Virus Constructs Protect Mice from H7N9 Challenge. J Virol. 2014 Feb 15;88(4):2340–3. 55. Ermler ME, Kirkpatrick E, Sun W, Hai R, Amanat F, Chromikova V, et al. Chimeric Hemagglutinin Constructs Induce Broad Protection against Influenza B Virus Challenge in the Mouse Model. J Virol [Internet]. 2017 May 26 [cited 2019 Dec 25];91(12). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5446656/ 56. Krammer F. The Quest for a Universal Flu Vaccine: Headless HA 2.0. Cell Host Microbe. 2015;18(4):395–7. 57. Impagliazzo A, Milder F, Kuipers H, Wagner MV, Zhu X, Hoffman RMB, et al. A stable trimeric influenza hemagglutinin stem as a broadly protective immunogen. Science. 2015 Sep 18;349(6254):1301–6. 58. Yassine HM, Boyington JC, McTamney PM, Wei C-J, Kanekiyo M, Kong W-P, et al. Hemagglutinin-stem nanoparticles generate heterosubtypic influenza protection. Nat Med. 2015 Sep;21(9):1065–70. 59. Krammer F, Fouchier RAM, Eichelberger MC, Webby RJ, Shaw-Saliba K, Wan H, et al. NAction! how can neuraminidase-based immunity contribute to better influenza virus vaccines? mBio. 2018;9(2):1–12. 60. Eichelberger MC, Morens DM, Taubenberger JK. Neuraminidase as an influenza vaccine antigen: a low hanging fruit, ready for picking to improve vaccine effectiveness. Curr Opin Immunol. 2018;53:38–44. 61. Wohlbold TJ, Podolsky KA, Chromikova V, Kirkpatrick E, Falconieri V, Meade P, et al. Broadly protective murine monoclonal antibodies against influenza B virus target highly conserved neuraminidase epitopes. Nat Microbiol. 2017 Oct;2(10):1415–24. 62. Fiers W, De Filette M, El Bakkouri K, Schepens B, Roose K, Schotsaert M, et al. M2ebased universal influenza A vaccine. Vaccine. 2009 Oct 23;27(45):6280–3. 63. Deng L, Cho KJ, Fiers W, Saelens X. M2e-Based Universal Influenza A Vaccines. Vaccines. 2015 Feb 13;3(1):105–36. 64. Kim K-H, Kwon Y-M, Lee Y-T, Kim M-C, Hwang H, Ko E-J, et al. Virus-Like Particles Are a Superior Platform for Presenting M2e Epitopes to Prime Humoral and Cellular Immunity against Influenza Virus. Vaccines. 2018 Sep 20;6(4):66. 65. Mullarkey CE, Boyd A, van Laarhoven A, Lefevre EA, Veronica Carr B, Baratelli M, et al. Improved adjuvanting of seasonal influenza vaccines: Preclinical studies of MVA-NP+M1

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

coadministration with inactivated influenza vaccine: Clinical immunology. Eur J Immunol. 2013 Jul;43(7):1940–52. 66. Lillie PJ, Berthoud TK, Powell TJ, Lambe T, Mullarkey C, Spencer AJ, et al. Preliminary Assessment of the Efficacy of a T-Cell–Based Influenza Vaccine, MVA-NP+M1, in Humans. Clin Infect Dis. 2012 Jul 1;55(1):19–25. 67. A Pivotal Trial to Assess the Safety and Clinical Efficacy of the M-001 as a Standalone Universal Flu Vaccine - Full Text View - ClinicalTrials.gov [Internet]. [cited 2019 Dec 23]. Available from: https://clinicaltrials.gov/ct2/show/NCT03450915 68. Atsmon J, Kate-Ilovitz E, Shaikevich D, Singer Y, Volokhov I, Haim KY, et al. Safety and immunogenicity of multimeric-001--a novel universal influenza vaccine. J Clin Immunol. 2012 Jun;32(3):595–603. 69. Evaluating the immunogenicity and safety of a BiondVax-developed universal influenza vaccine (Multimeric-001) either as a standalone vaccine or as a primer to H5N1 influenza vaccine [Internet]. [cited 2019 Dec 24]. Available from: https://www-ncbi-nlm-nihgov.libaccess.lib.mcmaster.ca/pmc/articles/PMC5369918/ 70. Pleguezuelos O, Robinson S, Stoloff GA, Caparrós-Wanderley W. Synthetic Influenza vaccine (FLU-v) stimulates cell mediated immunity in a double-blind, randomised, placebo-controlled Phase I trial. Vaccine. 2012 Jun 29;30(31):4655–60. 71. Francis JN, Bunce CJ, Horlock C, Watson JM, Warrington SJ, Georges B, et al. A novel peptide-based pan-influenza A vaccine: A double blind, randomised clinical trial of immunogenicity and safety. Vaccine. 2015 Jan 3;33(2):396–402. 72. Talbot HK, Rock MT, Johnson C, Tussey L, Kavita U, Shanker A, et al. Immunopotentiation of Trivalent Influenza Vaccine When Given with VAX102, a Recombinant Influenza M2e Vaccine Fused to the TLR5 Ligand Flagellin. PLOS ONE. 2010 Dec 28;5(12):e14442. 73. Turley CB, Rupp RE, Johnson C, Taylor DN, Wolfson J, Tussey L, et al. Safety and immunogenicity of a recombinant M2e-flagellin influenza vaccine (STF2.4xM2e) in healthy adults. Vaccine. 2011 Jul 18;29(32):5145–52. 74. Mezhenskaya D, Isakova-Sivak I, Rudenko L. M2e-based universal influenza vaccines: a historical overview and new approaches to development. J Biomed Sci. 2019 Oct 19;26(1):76. 75. Tsybalova LM, Stepanova LA, Kuprianov VV, Blokhina EA, Potapchuk MV, Korotkov AV, et al. Development of a candidate influenza vaccine based on virus-like particles displaying influenza M2e peptide into the immunodominant region of hepatitis B core antigen: Broad protective efficacy of particles carrying four copies of M2e. Vaccine. 2015 Jun 26;33(29):3398–406.

MUMJ Volume 17 No. 1, pp. 71-88

June 2020

76. Saelens X. The Role of Matrix Protein 2 Ectodomain in the Development of Universal Influenza Vaccines. J Infect Dis. 2019 Apr 8;219(Supplement_1):S68â&#x20AC;&#x201C;74. 77. Antrobus RD, Berthoud TK, Mullarkey CE, Hoschler K, Coughlan L, Zambon M, et al. Coadministration of Seasonal Influenza Vaccine and MVA-NP+M1 Simultaneously Achieves Potent Humoral and Cell-Mediated Responses. Mol Ther. 2014 Jan;22(1):233â&#x20AC;&#x201C;8. 78. Antrobus RD, Lillie PJ, Berthoud TK, Spencer AJ, McLaren JE, Ladell K, et al. A T CellInducing Influenza Vaccine for the Elderly: Safety and Immunogenicity of MVA-NP+M1 in Adults Aged over 50 Years. Doherty TM, editor. PLoS ONE. 2012 Oct 31;7(10):e48322. 79. Efficacy of Candidate Influenza Vaccine MVA-NP+M1 in Adults [Internet]. [cited 2019 Dec 31]. Available from: https://clinicaltrials.gov/ct2/show/NCT03880474 80. Bernstein DI, Guptill J, Naficy A, Nachbagauer R, Berlanda-Scorza F, Feser J, et al. Immunogenicity of chimeric haemagglutinin-based, universal influenza virus vaccine candidates: interim results of a randomised, placebo-controlled, phase 1 clinical trial. Lancet Infect Dis. 2019 Oct 17;

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

Review Article

Lateral vs. supine positioning for femoral intramedullary nailing: A systematic review of comparative studies Mohamed Sarraj, Daniel E Axelrod, Sarah Zhu, and Herman Johal McMaster University

Abstract Femoral shaft fractures are devastating injuries, often resultant from high-energy mechanisms in victims of poly-trauma. Reamed and statically locked intramedullary nailing (IMN) is the definitive treatment modality for femoral shaft fractures. Patients are most commonly positioned either supine or lateral decubitus. There remains considerable concern regarding the safety of lateral positioning in the traumatized patient, particularly in the management of a potentially difficult airway or concomitant C-spine injuries. We therefore undertook a systematic review of intraoperative positioning among patients with femoral shaft fractures following PRISMA guidelines. Title and abstract screening, full text screening, and data abstraction were all completed in duplicate. Methodological Index for Nonrandomized Studies (MINORS) scores were used to evaluate methodological quality. Results: 3018 studies were included in initial screening, with three studies ultimately meeting all inclusion criteria. A total of 1,949 patients were analyzed, with 684 patients treated in lateral positioning and 1,215 patients in supine positioning. Level of agreement was strong across title (κ = 0.872; 95% CI 0.794 to 0.951), abstract (κ = 0.801; 95% CI 0.585 to 1.000), and full-text screening (κ = 1.000). The consensus mean MINORS score of included studies was 17.67 ± 0.58, indicating good to high quality of evidence. Neither patient positioning offered obvious benefits such as fewer complications or shorter operative time. Furthermore, length of admission, days in ICU or on ventilator, and overall morbidity were not found to be significantly different between positions. Lateral positioning for intramedullary nailing of mid-shaft femur fractures appears to be a safe alternative to the standard supine positioning. There is a lack of both prospective and retrospective comparative studies investigating functional clinical outcomes in the literature. Keywords: Trauma, Long bone fracture, Femur fracture Corresponding author: mohamed.sarraj@medportal.ca

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

Introduction Femoral shaft fractures are devastating injuries that can result from high-energy mechanisms in victims of poly-trauma. Identifying these injuries is a critical part of a primary trauma survey (1); they demand urgent management to control morbidity secondary to blood loss, inflammation, and pain (2,3). Less commonly, femoral shaft fractures can result from fracture through an area of bone weakened by metastatic disease, so called ‘pathologic fracture’. Reamed intramedullary nailing (IMN) is the definitive treatment modality for femoral shaft fractures of any etiology (2– 5). There are variations in surgical technique, including antegrade vs. retrograde nailing, variations in optimal entry point, and variations in patient positioning. The predominant positioning for IMN is supine, with the lateral position being an infrequently used as they were considered to be unsafe, particularly in patients with blunt chest trauma and concomitant pulmonary injury (6). Instead, the supine position is often favoured as the optimal positioning for anesthetic care, particularly in the case of a C-spine injury or an otherwise compromised airway (7,8). A fracture or traction table is often used when the supine position is utilized for reduction of a femoral shaft fracture. Traction tables are generally only available in the supine position, so in the absence of a traction table, lateral positioning with manual traction is used. In either case, traction is critical to maintain length and reduction of the fracture. The main advantage of fracture table positioning includes the ability to hold and maintain the reduction for the duration of the procedure without use of an assistant. However, there have been studies that have questioning the superiority of a traction table compared to manual traction. Improved reduction quality has actually been demonstrated with manual traction in supine position compared to fracture table (9,10), likely due to improved ease of manipulation of the fracture fragments. Additionally, some literature suggests that manual traction has a shorter operative time compared to the use of a fracture table, though this has not been definitively shown in direct comparative studies (11,12). Operative time is critical as it may be part of the surgical burden that can provoke the development of systemic complications such as ARDS (Acute Respiratory Distress Syndrome) or SIRS (Systemic Inflammatory Response Syndrome), which are potentially devastating consequences of femoral fractures (2,13). Surgery can be considered an exacerbating ‘second hit’ of pathophysiologic inflammatory response following the initial traumatic injury (2). Manual traction could mitigate the risk of these complications if it did indeed produce shorter operative times for fixation of femoral fractures. Some proposed advantages of manual traction in the lateral positioning for IMN of femoral fractures include obviating the need for a fracture/traction table, which is costly, not universally available, and may cause soft tissue complications (14), easier access to entry point, particularly with obese patients (15), and optimal positioning for lateral radiographs with good visualization of the femoral head and proximal femur (16). To date, the authors are not aware of a review of lateral versus supine positioning for reamed intramedullary fixation of femoral shaft fractures. Hence the aim of this review was to review clinical outcomes of intraoperative supine and lateral positioning for intramedullary 91

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

fixation of femoral shaft fractures in papers directly comparing both groups. The primary hypothesis was that clinically important outcomes are comparable in both positions. The secondary hypothesis was that fewer intraoperative complications and shorter operative times are reported with lateral positioning.

Methods -

Two independent reviewers searched EMBASE, MEDLINE (including Epub Ahead of Print), CINAHL, and Web of Science for titles comparing lateral to supine positioning for intramedullary fixation of mid-shaft femur fractures from data inception to May 25, 2018. The purpose, research question, and eligibility criteria for the search were determined a priori. Study eligibility criteria are presented in Table 1. Table 1. Eligibility criteria. Inclusion Criteria 1. 2. 3. 4. 5.

Comparative studies with supine positioning as control Mid-shaft femur fractures Clinical outcomes reported Human Studies English language papers

Exclusion Criteria 1. 2. 3. 4. 5.

Inverse of inclusion criteria Pediatric or skeletally immature population Results that are not stratified for comparison with supine position outcomes Non-clinical outcomes reported Review, technical, or otherwise non-prognostic articles

The protocol of this systematic review was prospectively registered via the PROSPERO database (ID: CRD42018099373). Key articles were identified by a senior author prior to the search, all of which were screened for relevant keywords, subject headings, and relevant references. The primary author met with a senior librarian in order to verify that the search strategy was neither too broad nor too narrow, and to verify correct database syntax (Appendix 1 Figure 1; Search terms). The gray literature was reviewed through keyword searches of conference abstracts from the Orthopedic

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

Trauma Association (OTA) and the American Association of Orthopedic Surgeons (AAOS) between 2015 and May 2018. Registered clinical trials were screened by duplicate keyword searches of the clinicaltrials.gov website and the online International Standard Registered Clinical/soCial sTudy Number (ISRCTN) database. -

A systematic screening approach was undertaken in accordance with R-AMSTAR and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) (17) criteria from title to full text screening. Screening was performed by two independent reviewers (M.S. and S.Z) in duplicate. In the case of screening disagreement at the title and abstract stage, titles were automatically included in the next stage. At the full text stage, discrepancies were resolved by consensus decision with input from an independent senior reviewer (H.J). 6B Quality and reporting assessment of included non-randomized papers was performed in duplicate using the Methodological Index for Non-Randomized Studies (MINORS) (18), a validated appraisal tool based on study design features such as inclusion of consecutive patients and prospective collection of data. 12 items on the MINORS checklist are each scored 0-2, with maximum scores of 16 for non-comparative studies and 24 for comparative studies. Scoring for MINORS was done via consensus decision. 6

6 6

B 6

Data abstraction was performed independently in duplicate (M.S and S.Z.) for all included studies and recorded in separate Google Docs spreadsheets. The spreadsheets were combined and discrepancies were resolved by consensus decision. Inter-reviewer agreement was calculated for each stage, including MINORS, with a Kappa (κ) statistic. Agreement was categorized a priori as follows; 0.20 or less; poor, 0.21 to 0.40; fair, 0.41 to 0.60; moderate, 0.61 to 0.80; substantial, and 0.81 to 0.99; excellent. Study data were presented descriptively with means, proportions, and measures of variance when provided in the original source papers. Results -

There was substantial agreement amongst reviewers at each screening stage; title (κ = 0.872; 95% CI 0.794 to 0.951), abstract (κ = 0.801; 95% CI 0.585 to 1.000) and full-text (κ = 1.000).

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

Scoring of MINORS criteria demonstrated moderate agreement (κ = 0.663; 95% CI 0.433 to 0.893). Following discussion, the consensus mean MINORS score of included studies was 17.67 ± 0.58. -

6 6

Based on the search strategy, 3,085 papers were identified. Three full texts met inclusion criteria (Table 1). Manual, gray literature, and registered trial searching did not yield any additional papers. All included studies were level III evidence in the form of retrospective cohort studies. One study was conducted in Canada (19), one in Turkey (20), and one in the USA (21). All included studies were published within the last decade, between 2009 and 2018. This review analyzed a total of 1,949 patients, with 684 patients in the study sample (lateral positioning) and 1,215 patients in the control sample (supine positioning). Patient characteristics are presented in Table 2. In the lateral position, 74.4% were male and 67.4% of the patients were male in the supine position. The weighted mean age of the two groups were comparable at 36.7 ± 19.4 (n = 684) for lateral positioning, and 39.5 ± 20.6 (n = 1215) for the supine sample. All included studies reported data on patient demographics. There were no significant differences in the mean age or gender of included patients, save for one study where the lateral group had a younger mean age (35.8 vs. 40.1), which was “not deemed to be clinically significant” (19). Two studies (19,21) reported injury severity score (ISS) and abbreviated injury score (AIS) scores. The only significant difference found in these scores was a higher mean AIS chest score in the lateral (2.2 +/- 1.7) compared to the supine group (1.5 +/- 1.9) (P=0.01).

Table 2. Overview of selected studies. Author

Published

Study design

Total Sample Size (patients)

Number of males (%)

Mean age (years)

Follow up

Mean MINORS score

Apostle et al.

2009

Retrospective Cohort Study

988

(Lateral) 65.4 (Control) 65.8

(Lateral) 35.8 (Control) 40.1

Firat et al.

2012

Retrospective Cohort Study

(Lateral) 63.7 (Control) 70

(Lateral) 37.3 (Control) 38.1

46 months

Reahl et al.

2018

Retrospective Cohort Study

848

(Lateral) 65.3 (Control) 70.5

(Lateral) 37.1 (Control) 38.4

NR = Not reported

MUMJ Volume 17 No. 1, pp. 89-102 -

June 2020

Operative technique was reported to varying degrees of specificity (Table 3). All included studies reported the same surgical technique was used for both lateral and supine groups; and used reamed statically locked intramedullary nailing for both groups. In terms of approach, two studies utilized antegrade nailing (19,20) while Reahl et al. used a combination of both antegrade and retrograde nail placement depending on surgeon preference (Table 3). Lateral positioning involves specific positioning details, such as placing the patient on a radiolucent table with a bean bag, fracture side up. C-arm positioning for a true lateral is obtained by adjusting 10-20 degrees beyond perpendicular to match the anteversion of the femoral neck. After dissecting tissue to the entry point of the nail, the appropriate reduction technique depends on the nature of the fracture. A detailed exploration of surgical technique is beyond the scope of this article and other papers have been published detailing unique technical aspects of the lateral approach to various femoral fractures (6,16). Supine and lateral positions across all studies were stated to be standard, with the exception of the study by Firat et al. (20). In their supine position, the uninjured leg was raised in a semilithotomy position with the knee at 45-90 degrees flexion, and the hip at 45-90 degrees flexion and 30-45 degrees abduction. Table 3. Overview of operative information. Author

Surgical Approach

Reaming and Locking

Site of entry

Positioning (study;control)

Apostle et al.

Antegrade

Reamed and statically locked

Lateral standard; Supine standard

Firat et al.

Antegrade

Reamed and statically locked

Piriform

Lateral standard; Supine contralateral leg elevated (SCLE)*

Reahl et al.

Antegrade and Retrograde

Reamed and statically locked

Lateral standard; Supine standard

NR = Not reported *SCLE positioning has uninjured leg elevated in semilithotomy position with knee at 45-90 degrees flexion and hip at 45-90 degrees flexion and 30-45 degrees abduction. It also uses manual traction on the affected limb (20).

The study by Firat at el. reported a significantly shorter OR time in SCLE position in comparison to the lateral position at 98.4 minutes and 108.2 minutes, respectively (20). No other studies reported operative time. 95

MUMJ Volume 17 No. 1, pp. 89-102 &

June 2020

Indications for supine or lateral positioning were not specified across the included studies. Reahl et al. did not specify reasons for positioning (21), Apostle et al. reported different positioning was due to surgeon preference (19), and Firat et al. reported positioning was due to integration of a newly acquired skill of performing femoral nailing in the SCLE position (20). 6

6 6B

Results for different measures of morbidity were reported heterogeneously across all included studies. However, the studies by Apostle et al. and Reahl et al. reported similar outcomes (19,21). Apostle et al. measured patient morbidity with admission into intensive care unit (ICU) and length of stay (LOS) in the ICU (19) while Reahl et al. collected information regarding patientâ&#x20AC;&#x2122;s the length of stay in the ICU and days of ventilator use as an indirect indicator of pulmonary complications (21). With regards to patient ICU admission, Apostle et al. reported lower admission in the supine position at 11.9% of patients in comparison to 12.8% of the laterally positioned patients (Table 4) (19). This difference was not found to be statistically significant in the study, however a subgroup analysis of patients with AISâ&#x2030;Ľ3 found that lateral positioning was protective against ICU admission (P=0.044). Apostle et al. reported no statistically significant difference in mean LOS when comparing supine and lateral positioning (19). Reahl et al. reported a significantly decreased length of ICU LOS in the lateral position in comparison to supine control (Table 4) (21). They also found a mean 1.29 day shorter postoperative time on a ventilator in the lateral group, though this was not statistically significant. 6

Patient mortality was only reported by Apostle et al., monitored at up to 115 days postoperatively (19). There was a higher rate of patient mortality in the supine positioning, however this was not significant and had an odds ratio near 1 (Table 4). The most common cause of death was progression of metastatic disease, followed by head injury and sepsis leading to multiorgan failure. Two deaths were attributed to fat embolism syndrome (FES), both of which were in the supine sample. 6

Post-operative complications, specifically leg length discrepancy, malalignment, and malrotation were reported in one study by Firat et al. (20). They found no difference in coronal-sagittal 96

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

malalignment between the lateral and supine groups. Mean difference in limb length was +0.4mm (range -14 to +17mm) in the SCLE group and -1.6mm (range -12 to +14mm) in the lateral group, with no statistically significant difference found in subgroups. Of patients who experienced leg length discrepancy, a significantly higher number of patients had leg shortening of less than 10mm in the lateral group compared to the SCLE group (10% SCLE group vs. 30% lateral group, p<0.001). However, the percentage of patients with malrotation was higher in the laterally positioned group, with the mean rotation difference in the lateral group being internal while the SCLE group had an external mean rotational difference (mean +1.2 degrees in SCLE group vs. mean -2.6 degrees in lateral group). This difference has an unknown statistical significance, although there was a significantly higher number of patients with >15o of internal malrotation in the lateral group in comparison to the SCLE group (p<0.001). Table 4. Overview of patient outcomes. Author

Lateral Position Sample

Supine Position Sample

ICU admission (%)

ICU LOS (days)

Ventilator (days)

Mortality (%)

Leg length discrepancy (%)

Malrotation (%)

Apostle et al.

227

761

(Lateral) 12.8 (Control) 11.9

(Lateral) 1.7 (Control) 1.1

(Lateral) 1.8 (Control) 3.0

Firat et al.

(Lateral) 36.7 (Control) 45.5

(Lateral) 57.6 (Control) 50.0

Reahl et al.

424

(Lateral) 1.64 (Control) 3.63

(Lateral) 2.89 (Control) 4.18

NR = Not Reported

Discussion 6

Findings of this review suggest that lateral positioning is a safe alternative to the standard supine approach. One article (19) found that lateral positioning was not associated with an increased risk of mortality or ICU LOS, and may actually be protective against ICU admission in patients with AIS greater than 3. Another large retrospective cohort in this review (21) found that lateral positioning yielded a mean 1.88 days shorter ICU stay, with a mean 1.29 fewer days with ventilator support. Intraoperative advantages, including ease of access to entry point and optimal imaging, were not directly assessed in the included papers. Though shorter operative time may be a proxy 97

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

for these measures, this was only reported in Firat et al. (n=63) and was indeed shorter in the lateral group (20). Additionally, although anesthesia-related complications have been raised as a concern for lateral positioning, no such direct complications were reported in the included papers. Thus, the aforementioned mortality outcomes do demonstrate the overall safety of this positioning for mid-shaft femur fractures. No papers included reported functional outcome or long-term measures such as union rate, re-operation, or pain. Complications of lateral positioning were poorly defined by the included studies. Only one paper (20) reported malrotation and limb length discrepancy, demonstrating decreased limb length discrepancy, yet increased malrotation, in the lateral position. Angular malalignment was found to be similar in both groups. However, this article had a small study sample (n=33) and was comparing lateral positioning to a novel technique (SCLE). No other intraoperative complications were reported. No conclusions could be made about operating time as the measure was only reported in one paper (n=33) which compared lateral positioning to a novel modified (SCLE) supine approach (20). Literature on supine positioning suggests that manual traction reduces operating time compared to the use of a fracture table (11,12), however, there is no definitive literature regarding operative time required for lateral positioning. Apostle et al. was the only paper including pathologic femur fractures in their analysis, and found no association between mortality in this population and surgical positioning, finding mortality to be the result of progression of already existing metastatic disease. -

6 6

This review only included papers that directly compared lateral positioning to the standard supine position. Hence the comparisons yielded in this review are standardized by time, institution, and methods, thereby minimizing variability. Though few papers are included in this review, there is a large total sample size with robust methodology for included papers, as evidenced by the high mean MINORS score. The high heterogeneity in outcome measures among studies precluded a meta-analysis or data pooling of any kind. All included studies were retrospective and thus treatment allocation was not randomized, which introduces a source of bias. Despite large samples, the main outcomes addressed by these papers, such as ICU admission and mortality, have a low incidence, making it difficult to draw statistically significant conclusions from the results. This is compounded by the fact that these outcomes are multifactorial and that all studies included polytraumatized patients who may have had severe concomitant injuries. Lateral positioning appears to be a valuable tool in the surgeonâ&#x20AC;&#x2122;s armamentarium, with reported advantages including circumferential access to the affected limb, ease of conversion to an extensile approach if needed, and increased access to the piriformis fossa (6,16,19). The advantage of easier access to an entry point is particularly relevant in the context of the increasing global prevalence of obesity (6,22â&#x20AC;&#x201C;24). Challenges associated with this technique

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

include difficulty accessing distal third femoral fractures and decreased access for anesthesia. Some contraindications for lateral positioning are unstable spinal injuries and pulmonary pathology (6). The principal findings of this paper are not related to functional outcome, but rather to safety. This paper confirms that, particularly in resource-scarce settings where access to an expensive traction table may not be available, cost is not traded for mortality with lateral positioning. Further comparative research evaluating functional outcomes, especially with subgroup analyses of obese patients and other groups, would be valuable in further elucidating the role for this positioning in treating femoral fractures. Such studies may also clarify whether the proposed intraoperative advantages of lateral positioning with manual traction, for instance decreased operative time, are materially significant. Conclusion Lateral positioning for intramedullary nailing of mid-shaft femur fractures appears to be a safe alternative to the standard supine positioning. There is a lack of both prospective and retrospective comparative studies investigating functional clinical outcomes in the literature.

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

References 1.

Trunkey D. Initial Treatment of Patients with Extensive Trauma. N Engl J Med. 1991;324(18):1259-1263. doi:10.1056/NEJM199105023241806

Gänsslen A, Gösling T, Hildebrand F, Pape HC, Oestern HJ. Femoral shaft fractures in adults: treatment options and controversies. Acta Chir Orthop Traumatol Cech. 2014;81(2):108-117.

Bone LB, Johnson KD, Weigelt J, Scheinberg R. Early versus delayed stabilization of femoral fractures. A prospective randomized study. J Bone Joint Surg Am. 1989;71(3):336340.

The Science and Practice of Intramedullary Nailing. 2nd ed. Baltimore: Williams & Wilkins; 1996.

O’Brien PJ, Meek RN, Powell JN, Blachut PA. Primary intramedullary nailing of open femoral shaft fractures. J Trauma. 1991;31(1):113-116.

Bishop JA, Rodriguez EK. Closed intramedullary nailing of the femur in the lateral decubitus position. J Trauma. 2010;68(1):231-235. doi:10.1097/TA.0b013e3181c488d8

Johnson KD, Greenberg M. Comminuted femoral shaft fractures. Orthop Clin North Am. 1987;18(1):133-147.

Aiyer S, Jagiasi J, Argekar H, Sharan S, Dasgupta B. Closed antegrade interlocked nailing of femoral shaft fractures operated up to 2 weeks postinjury in the absence of a fracture table or C-arm. J Trauma. 2006;61(2):457-460. doi:10.1097/01.ta.0000210269.05305.75

Stephen DJG, Kreder HJ, Schemitsch EH, Conlan LB, Wild L, McKee MD. Femoral intramedullary nailing: comparison of fracture-table and manual traction. a prospective, randomized study. J Bone Joint Surg Am. 2002;84-A(9):1514-1521.

10.

Jaarsma RL, Pakvis DFM, Verdonschot N, Biert J, van Kampen A. Rotational malalignment after intramedullary nailing of femoral fractures. J Orthop Trauma. 2004;18(7):403-409.

11.

Rohilla R, Singh R, Rohilla S, Magu NK, Devgan A, Siwach R. Locked intramedullary femoral nailing without fracture table or image intensifier. Strateg Trauma Limb Reconstr. 2011;6(3):127-135. doi:10.1007/s11751-011-0122-3

12.

Karpos PA, McFerran MA, Johnson KD. Intramedullary nailing of acute femoral shaft fractures using manual traction without a fracture table. J Orthop Trauma. 1995;9(1):5762.

100

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

13.

Pape H-C, Hildebrand F, Pertschy S, et al. Changes in the Management of Femoral Shaft Fractures in Polytrauma Patients: From Early Total Care to Damage Control Orthopedic Surgery. J Trauma Acute Care Surg. 2002;53(3):452.

14.

A Flierl M, Stahel P, Hak D, J Morgan S, Smith W. Traction Table-Related Complications in Orthopaedic Surgery. Vol 18.; 2010. doi:10.5435/00124635-201011000-00004

15.

Winquist RA, Hansen ST. Comminuted fractures of the femoral shaft treated by intramedullary nailing. Orthop Clin North Am. 1980;11(3):633-648.

16.

Carr JB, Williams D, Richards M. Lateral Decubitus Positioning for Intramedullary Nailing of the Femur Without the Use of a Fracture Table. Orthopedics. 2009;32(10). doi:10.3928/01477447-20090818-05

17.

Kung J, Chiappelli F, Cajulis OO, et al. From Systematic Reviews to Clinical Recommendations for Evidence-Based Health Care: Validation of Revised Assessment of Multiple Systematic Reviews (R-AMSTAR) for Grading of Clinical Relevance. Open Dent J. 2010;4:84-91. doi:10.2174/1874210601004020084

18.

Slim K, Nini E, Forestier D, Kwiatkowski F, Panis Y, Chipponi J. Methodological index for non-randomized studies (minors): development and validation of a new instrument. ANZ J Surg. 2003;73(9):712-716.

19.

Apostle KL, Lefaivre KA, Guy P, et al. The Effects of Intraoperative Positioning on Patients Undergoing Early Definitive Care for Femoral Shaft Fractures. J Orthop Trauma. 2009;23(9):615. doi:10.1097/BOT.0b013e3181a6a941

20.

Firat A, Tecimel O, Deveci A, Ocguder A, Bozkurt M. Surgical Technique: Supine Patient Position With the Contralateral Leg Elevated for Femoral Intramedullary Nailing. Clin Orthop. 2013;471(2):640-648. doi:10.1007/s11999-012-2722-8

21.

Reahl GB, OĘźHara NN, Coale M, et al. Is Lateral Femoral Nailing Associated With Increased Intensive Care Unit Days? A Propensity-Matched Analysis of 848 Cases. J Orthop Trauma. 2018;32(1):39-42. doi:10.1097/BOT.0000000000000999

22.

Wild S, Roglic G, Green A, Sicree R, King H. Global Prevalence of Diabetes: Estimates for the year 2000 and projections for 2030. Diabetes Care. 2004;27(5):1047-1053. doi:10.2337/diacare.27.5.1047

23.

Flegal KM, Carroll MD, Kit BK, Ogden CL. Prevalence of Obesity and Trends in the Distribution of Body Mass Index Among US Adults, 1999-2010. JAMA. 2012;307(5):491497. doi:10.1001/jama.2012.39

24.

Caballero B. The Global Epidemic of Obesity: An Overview. Epidemiol Rev. 2007;29(1):1-5. doi:10.1093/epirev/mxm012

101

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

Appendix 1 3085 Studies Identified Embase: 1244 Studies Medline/Epub Ahead of Print: 1266 Studies Web of Science: 476 Studies CINAHL: 97 Studies

Removal of duplicates

Removed: 1195

1890 Studies

1844 Removed: - Non-femoral fracture; 464; - Non clinical, basic science; 440 - Review or non-prognostic; 107 -Non- intramedullary fixation; 88 -Non-femoral shaft fractures; 438 - Non-comparative papers; 114 - Computer navigated; 17 - Additional duplicates; 22 - Pediatric; 154

Title Review Kappa 0.872

46 Studies

35 Removed: - Non-femoral shaft fractures; 22 - Non-comparative papers; 7 - Review or non-prognostic; 8

Abstract Review Kappa 0.801 11 Studies

Full Text Review Kappa 1.0 Ka 3 Studies Included For Qualitative Analysis

8 Removed: -Non-comparative papers; 4 -Technique article; 1 -Newsletter; 1 -Results not stratified; 1 -Non-english paper; 1

Appendix 1 Figure 1. Systematic search flowchart.

102

MUMJ Volume 17 No. 1, pp. 89-102

June 2020

Appendix 1 Table 1. Search terms. Embase (1244 titles) 1. intramedullary nailing/ or ((intramed* or IM) adj4 fix*).ti,kw,ab. or ((intramed* or IM or inter* or Kuntscher or gamma) adj4 (nail* or rod)).ti,kw,ab. or bone nail/ 2. exp femur/ or (femur or femor* or leg).ti,kw,ab. or femur fracture/ 3. (lateral* or decub* or side or slopp* or (free* adj3 leg*)).ti,kw,ab. 4. 1 and 2 and 3 5. 4 not animals/ not (humans/ and animals/) 6. limit 5 to "review" 7. 5 not 6

Web of Science (476 titles) TOPIC ((intramed* OR IM OR inter OR Kuntscher OR gamma) NEAR/2 (fix* OR nail* OR rod*)) AND TOPIC: (femur* OR femor*) AND TOPIC: (lateral* OR decub* OR side OR slopp* OR (free* NEAR/2 leg*)) Refined by: DOCUMENT TYPES: ( ARTICLE OR MEETING ABSTRACT )

Medline1 (1266 titles) 1. Fracture Fixation, Intramedullary/ or ((intramed* or IM) adj4 fix*).ti,kf,ab. or ((intramed* or IM or inter* or Kuntscher or gamma) adj4 (nail* or rod)).ti,kf,ab. or Bone Nails/ 2. exp FEMUR/ or (femur or femor* or leg).ti,kf,ab. or Femoral Fractures/ 3. (SUPINE POSITION/ or flat.mp. or recumbent.mp. or supin*.mp. or reclin*.mp. or prostrat*.mp. or (fracture adj2 table).mp.) and (lateral* or decub* or side or slopp*).mp. 4. 1 and 2 and 3 5. 4 not animals/ not (humans/ and animals/) 6. limit 5 to "review articles" 7. 5 not 6 CINAHL (97 titles) ( (MH "Femur+") OR (MH "Femoral Fractures+") OR fem?r* ) AND ( (intramed* N3 (nail* OR rod*)) OR (im N3 (nail* OR rod*)) OR (gamma N3 nail* OR Kuntscher) OR (inter* N3 (nail* OR rod)) OR (intramed* N3 fix*) ) AND ( (MH "Lateral Position") OR lateral* OR decub* OR side OR slopp* OR (free* N3 leg*) )

Ti, kw, ab = Term appears in title, keywords, or abstract Mp = Term appears in title, abstract, subject heading, author keywords, or other category 1 OVID Medline Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE(R) Daily and Ovid MEDLINE(R) 1946 to Present

103

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

Review Article

The critical role of astrogenesis and neurodevelopment in Fragile X Syndrome and Rett Syndrome Zhuo Jun Li1 and John-Paul Oliveria1,2 1

McMaster University

Stanford University

Abstract Astrocytes play an important role in the development of functional neural circuits in the brain; they are responsible for coordinating synapse formation and function, axon guidance, and ensuring neuronal survival. A better understanding of these mechanisms and their timing will further characterize its role in neurodevelopmental diseases. This paper will discuss the proposed pathways of astrogenesis, and genetic mutations that give rise to Fragile X Syndrome (FXS) and Rett Syndrome (RS). Normal astrogenesis begins during late gestation and is regulated by both cell intrinsic and extrinsic pathways. However, disruption to astrogenesis have been linked to abnormal astrocyte development and results in pathologies such as FXS and RS. FXS and RS are a result of genetic mutations that inhibit astrocyte function in FXS, and disproportionately activate astrocyte function in RS. Both FXS and RS have been associated with the theory that altered gene transcription during neurodevelopment disrupts astrogenesis, and subsequently, the behavior and function of mature astrocytes in the brain. Overall, current research has focused on the impact of the genetic mutation on the developmental pathway of astrocytes, and how the subsequent changed astrocytes play a role in the pathogenesis of FXS and RS. The existing gaps in knowledge around the timing of initial astrogenesis and identifying astrocyte-specific markers indicates more research is needed to discover the extent astrogenesis are affected by genetic mutations, and how scientists can improve upon existing techniques in studying astrocytes. Keywords: Astrocyte, Astroglia, Astrogenesis, Neurodevelopment, Fragile X Syndrome, Rett Syndrome *Corresponding author: oliveria@standford.edu

104

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

List of abbreviations Abbreviation bFGF BMP CNS CT-1 EAAT-1/GLAST EAAT-2/GLT-1 EGF FMR1 FMRP FXS GFAP gp130 HLH IL-6 JAK-STAT LIF MeCP2 NFIA NSC NT-3 RG STAT3 TNC

Definition basic fibroblast growth factor bone morphogenetic protein central nervous system cardiotrophin-1 excitatory amino acid transporter 1/glutamate aspartate transporter 1 excitatory amino acid transporter 1/glutamate transporter 1 epidermal growth factor Fragile X mental retardation gene 1 Fragile X mental retardation protein fragile X syndrome glial fibrillary acidic protein glycoprotein 130 helix-loop-helix interleukin-6 Janus Kinase/Signal Transducer and Activator of Transcription leukemia inhibitory factor methyl-CpG-binding protein 2 nuclear factor 1 A-type neural stem cell neurotrophin-3 radial glia signal transducer and activator of transcription 3 Tenascin C

105

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

Introduction The central nervous system (CNS) consists of neurons and glia. The latter is also subdivided into astrocytes and oligodendrocytes. Astrocytes, the most abundant non-neuronal cells of the CNS, are known as the choreographers of the neural circuit (1). They play a critical role in developing the functional neural circuits of the brain, coordinating synapse formation and function, ensuring neuronal survival, and axon guidance (2,3). As new neurons form in the developing brain, the brain is subsequently populated by astrocytes, which contribute to the formation and maintenance of neural circuits in the early postnatal brain by controlling synapse formation, function, and elimination (2). In the mature brain, astrocytes perform multiple roles like producing and recycling neurotransmitters, regulating extracellular ion concentrations, and providing structural support to neurons and the blood brain barrier (2,4). Neurons, astrocytes, and oligodendrocytes are differentiated from a common neuroepithelial origin in a temporally defined manner (2). First divisions of neural stem cells (NSC) are thought to be exclusively neurogenic in early gestation and primarily gliogenic in late gestation (5,6). Astrocyte development is proposed to begin with NSCs that differentiate into astrocyte precursors (5). Local proliferation then allows these precursors to mature into astrocytes (Figure 1), which are divided into two major subtypes: fibrous and protoplasmic. Fibrous astrocytes, also known as white matter astrocytes, have fewer but thicker processes and

Figure 1. Process of astrocyte development and maturation. (A) Astrocytes begin as a neural stem cell during early pregnancy, and express predominantly markers for neurogenesis, such as Nestin and brain lipid-binding protein (BLBP). (B) During late pregnancy (in the fetal phase), stem cells begin to express astrocyte-related markers such as GLAST, S100B, and Aqp4, which allows them to become astrocyte precursors. (C) Through local proliferation, astrocyte precursors become functionally mature astrocytes, and now express additional astrocyte-specific markers (ex. GFAP). The mature astrocytes can specify into either (D) fibrous astrocytes or (E) protoplasmic astrocytes depending on their location within the central nervous system.

106

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

typically express higher levels of the astrocyte intermediate filament glial fibrillary acidic protein (GFAP) (6,7). They typically have regular cylindrical processes and contours, forming the more classic star-like appearance (7). Their diverse range of function includes regulating the flow of blood through the CNS, maintaining synapses, and storing and releasing nutrients (7). In grey matter, protoplasmic astrocytes have elaborate processes and eventually arrange into spatially segregated astrocyte domains of adult brains (6-8). Protoplasmic astrocytes contain hundreds of fine processes, with the smallest endfeet directly contacting neuronal synapses, creating tripartite synapses (8). These astrocytes participate in synaptic communication in a variety of direct and indirect functional pathways, such as clearing glutamate, modulating synapse functions, and regulating local capillary blood flow (8). Each astrocyte cell forms a well-defined nonoverlapping territory. As mentioned previously, astrogenesis occurs during late-gestation and early postnatal stages. Astrocytes are generated from NSCs through three main proposed pathways: Notch signaling, BMP signaling pathway, and IL-6, in collaboration with the JAK-STAT pathway (9,10). Astrogenesis is regulated by both cell intrinsic programs and cell extrinsic cues; disruption within the pathways result in abnormal astrocyte differentiation and function. Both Fragile X-Syndrome (FSX) and Rett Syndrome (RS) are rooted in genetic mutations that lead to astrocyte dysfunction. Table 1. Summary of astrogenesis. The stages of astrogenesis through early pregnancy, late pregnancy, and postnatally, including the breakdown of the stages of astrocyte development during each time period. Time (In Chronological Order) 1. Embryonic Phase/Early Pregnancy 2. Fetal Phase/Late

Stages of Astrogenesis I. II. III. I.

Pregnancy II. III. IV. 3. Postnatal Phase

I. II. III.

RGs are derived from the lateral wall of the neural tube, and migrate radially along the cerebral wall Methylated astrocyte-specific genes Neurogenesis occurs RGs accelerate the expansion of the neuronal population, and subsequently switch to gliogenesis to produce astrocytes Astrocyte precursor cells migrate throughout the CNS Chromatin remodeling through demethylation of astrocyte-specific promoters begin astrogenesis and silence neurogenesis Exogenously secreted cues (ex. IL-6, BMP family signaling, and Notch signaling) initiates and maintains astrogenesis Differentiated astrocytes begin to display diversity in morphology Astrocyte processes extend and astrocyte domains become recognizable Astrocytes matures into terminally differentiated state, but functional identity is not completely hardwired.

107

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

Initiation of astrogenesis: Intrinsic chromatin remodeling Astrocyte differentiation from NSCs is a temporally regulated phenomenon that relies on exogenously secreted cues and intrinsic chromatin changes (2,11,12). Notch signaling is the master regulator of astrocyte differentiation (13,14). This pathway represses neurogenesis when the fetus reaches late-gestation, and induces astrocytic differentiation through intrinsic chromatin remodeling (10,13,14). Until late gestation, NSCs are insensitive to cytokines related to astrogenesis, like IL-6, because they have not obtained astrogenic potential (11,12,15). The downstream GFAP gene is only activated in neuroepithelial cells in relatively late gestational stages, never in early or mid-gestation (9,15). Acquisition of astrogenic potential is attributed to DNA demethylation at astrocyte-specific genes such as GFAP, S100Î˛, and aquaporin 4 (11,16). Promoters of astrocyte-specific genes are highly methylated prior to late-gestation; CpG dinucleotides exist at the GFAP promoter region, impairing the activation of the GFAP gene and effectively silencing transcription (11,15,16). When Notch signaling is activated, RBP-JĎ° transcribes notch-target genes such as nuclear factor 1 A-type (NFIA) (15-17). As a result, the increasing expression of NFIA induces demethylation of GFAP promoters, like the STAT3 binding site, and leads to GFAP expression in NSCs and signaling the beginning of astrogenesis (15,16,18). Astrogenesis depends on STAT3 activation, as STAT3 is the transcription factor for the promoter region of the GFAP gene (15). Exposure of the promoter allows NSCs to become responsive to external cues (16). During the neurogenic phase, NSCs elongate to become radial glia (RG), and divide asymmetrically for auto-renewal and generation of neurons/neuron-restricted intermediate progenitor cells (19-21). Newborn neurons migrate along parental RG fibers to their destination (20). At late gestation, majority of RG cells have lost their ventricular attachment and migrate towards the cortical plate, subsequently dividing to become astrocytes (19,21). During this gliogenic phase, RG progenitors gain competence to generate astrocytes due to the activity of growth factors like basic fibroblast growth factor (bFGF) and epidermal growth factor (EGF) (19,22). These factors allow RGs to respond to specific gliogenic signals acting at the extracellular level, and to respond to activated mature astrocyte markers like excitatory amino acid transporter 1 (EAAT1)/glutamate aspartate transporter 1 (GLAST), S100Î˛, and aquaporin 4 (19,23) (Figure 2). GLAST is a glutamate transporter that is functionally active in astrocytes, and its expression coincides with the gliogenic switch, making it a specific marker of astrocyte precursors in the spinal cord (23).

108

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

Figure 2. Astrocyte marker expression profile. Five astrocyte-specific markers are often found on the surface of astrocytes, and aid in the role of astrocytes in the CNS. GFAP is thought to maintain astrocyte mechanical strength and determine the shape of the cell. S100B is a calcium binding protein that regulates intracellular activities. Aquaporin-4 is a water channel expressed by astrocytes, used to maintain cerebral water balance and influences synaptic plasticity and memory. EAAT1 and EAAT2 are both functionally active glutamate transporters whose expressions coincide with the gliogenic switch.

The synergistic effect of IL-6, BMP, and Notch signaling in promoting astrogenesis There are three synergistic pro-astrogenesis pathways; IL-6 and cardiotrophin-1 (CT-1), BMP signaling, and Notch signaling. First, CT-1 binds to glycoprotein130 (gp130) and leukemia inhibitory factor (LIF) receptor beta coreceptors, which then signal via the JAKs to phosphorylate and activate the STAT3 transcription factors (10,13,15,24). STAT3 then forms a complex with Smad (Caenorhabditis elegans Sma genes and the Drosophila Mad, Mother against decapentaplegic) proteins, which are downstream of BMP receptors. BMPs belong to a subset of transforming growth factor-B superfamily and includes multiple functional peptides that control proliferation and differentiation in various cell types, including astrocytes (10). They also inhibit neuronal differentiation during late-gestation. This pathway initiates when BMP2 and BMP4 bind to heterotrimeric serine/threonine kinase receptors to signal via the activation of downstream transcription factors Smad1, Smad 5,

109

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

and Smad 8 (10). They form a transcriptional complex with the aforementioned activated STAT3 in the CT-1 induced pathway, which is then able to bind to astrogenic genes like GFAP (18). At the same time, Notch, upon binding to its ligands, is activated and cleaved. The Notch intracellular domain then translocate to the nucleus where it interacts with RBP-JĎ° to form a transcriptionally active complex that binds directly to the GFAP or S100Î˛ promoter and promotes transcription (24). However, this must occur when the JAK-STAT pathway is also activated (18,25). Notch1 mRNA is also upregulated during astrogenesis, suggesting that a feedback loop sustains Notch signaling until astrocytic differentiation is complete (19). Together, these three pathways and a key proastrocytic transcription factor, NFIA, bind to the GFAP promoter (Figure 3). Their coordinated actions determine the timing and number of astrocytes that are ultimately formed (24).

Figure 3. Synergistic pathway of Notch signaling, BMP family signaling, and IL-6 signaling in conjunction with the activated JAK-STAT pathway to induce astrogenesis. First, CT-1 binds to gp130/LIFRB coreceptors, which induces downstream signaling via the JAK-STAT pathway molecules. Together, it forms a complex with downstream BMP signaling molecules. At the same time, Notch is activated and cleaved, and translocate to the nucleus to interact with RBP-J to form another transcriptionally active complex. Along with NF1, the two complexes bind to the GFAP promoter.

IL-6 and BMP families synergistically induce astrocyte differentiation (9), and the JAKSTAT pathway mediates signal transmission into the nucleus (25). In addition to CT-1, leukemia

110

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

inhibitory factor (LIF) can also bind to leukemia inhibitory factor receptor (LIFR)/gp130 and forms a complex of LIF/LIFR/gp130 (26). In either situation, binding of an IL-6 family molecule (i.e., CT-1 or LIF) triggers JAK1 activation, which is constitutively associated with the cytoplasmic region of LIFR and gp130 (26). Subsequent protein to protein interaction and phosphorylation of the cytoplasmic region of gp130 and LIFR by JAK1 leads to recruitment and tyrosine phosphorylation of STAT3 (9,18). The dimerized STAT3 is translocated into the nucleus, where it binds to the promoter of the GFAP gene (18). Other astrocyte inducers are the BMP family cytokines, such as BMP2. Similar to the pathway discussed above, BMP2 binds to a

Figure 4. Overview of the IL-6 BMP family signaling. IL-6 family molecules bind to the LIFR/gp130 coreceptor, triggering the activation of JAK1, leading to tyrosine phosphorylation and recruitment of STAT3. The dimerized STAT3 translocate to the nucleus, where they it binds to the GFAP promoter. At the same time, BMP2 binds to the BMP receptor complex, leading to the phosphorylation of downstream Smad proteins like Smad1, which then translocate to the nucleus. Downstream collaboration creates the P300/STAT3/Smad1/4 complex that binds to the GFAP gene to promote astrogenesis. BMP family molecules can also induce negative HLH factors that inhibit neurogenesis.

receptor complex composed of type I and II BMP2 receptors. Both receptors are membrane spanning serine-threonine kinases (9). Upon binding of the ligands, type II receptor phosphorylates type I, which then goes on to phosphorylate downstream Smad proteins (9). The activated ligands translocate to the nucleus together; this synergistic effect is required to induce

111

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

astrocyte differentiation (25) (Figure 4). In the nucleus, STAT3 and Smad1 bind to their recognition sequences in the GFAP promoter (9,25). Transcriptional coactivator p300 interacts with STAT3 at its amino terminus with Smad1 binding to its carboxyl terminus, leading to efficient expression of the GFAP gene (9). This synergistic effect can be observed with any combination of the IL-6 family and BMP2, BMP4, or BMP7. In addition to astrogenesis-activating pathways, BMP family molecules suppress neurogenesis to allow for the switch to astrogenesis. BMP2 alters the developmental pathway of NSCs from neurogenesis to astrogenesis by inducing negative helix-loop-helix (HLH) factors such as Id1, Id2, and transcription factor hes5 (9,10,27) (Figure 4). These proteins inhibit the neurogenic bHLHs that normally promote neurogenesis and supress gliogenesis (13,16,27,28). BMP2 is also shown to reduce the number of NSCs expressing a marker for undifferentiated neural precursors, like nestin, and increasing the number of cells expressing S100Î˛, an early astrocyte marker (28). This change initiates the switch from neurogenesis to astrogenesis. Polycomb repressive complex 2 (PRC2) also contributes to suppressing neurogenesis by catalyzing H3L27 methylation to silence the Neurogenin1 gene that codes for neurogenesis (28). The silencing both terminates neurogenesis and stops Neurogenin1 from interfering with the P300-Smad1 complex of STAT3, which is essential for astrocyte development (16). As seen from the multiple synergistic pathways and molecules involved with astrogenesis discussed above, astrogenesis is a tightly regulated process. Dysregulation at any point in this intricate dance can give rise to neurodevelopmental disorders such as FXS and RS.

Abnormal Astrogenesis 6

-B

FXS is a neurodevelopmental disorder that affects 1 in 2500 individuals and is the most common cause of inherited brain development abnormalities (29). Children with FXS mild to severe cognitive impairment, attention deficit, anxiety, susceptibility to seizures, motor disorders, and autistic behaviors (30,31). FXS is linked to mutations in the fragile X mental retardation gene 1 (FMR1) (29,30), which codes for the fragile X mental retardation protein (FMRP). FMRP modulates the transcription of a number of mRNAs important for dendritic growth and development of the synapses (30), and is only expressed in the developmental stages of astrogenesis (29). Individuals diagnosed with FXS have an abnormal number (>200) of CGG repeats in the 5â&#x20AC;&#x2122; noncoding region of the FMR1 gene, resulting in hypermethylation and transcriptional silencing of the gene, meaning insufficient production of FMRP (29) (Figure 5). In absence of FMRP production, there is a disruption in the composition of normal protein milieu in astrocytes.

112

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

Figure 5. Pathogenesis of Fragile X Syndrome. Mutation of the FMR1 gene during pregnancy causes insufficient production of FMRPs, which is a regulator of mRNA translation important for dendritic growth and synaptic development. Lack of FMRPs creates a disruption in the composition of normal protein milieu in cells like astrocytes, leading to abnormalities in astrocyte development. This leads to FXS, with symptoms such as cognitive impairment, seizures, and motor disorders.

Research assessing murine models of FXS show that astrocytes are involved in shaping the dendritic arbors of neurons in FXS (29). Jacobs et al. found that neurons exhibited a distinct abnormal morphology when grown with Fmr1 KO astrocytes, compared to neuronal growth with wild-type astrocytes (29). This suggests a loss of fine-tuning in astrocyte-mediated neuronal growth, migration, and pruning in FXS. This results in complex dendritic arbors with increased branch density in neurons grown with astrocytes from FXS mice, while expected levels of branch density was observed with normal astrocytes (29). Astrocytes in FXS have also been found to be deficient in the ability to regulate synapse development, as neurons in FXS mice exhibit a decreased number of presynaptic and postsynaptic protein aggregates (32). Thus, astrocytes could be a major contributor to erroneous synapse development and behavior maladaptation in FXS. The current theory on abnormal astrocyte development speculates that the reduction of FMRP could directly or indirectly contribute to abnormalities in astrocyte development (32). In astrocytes, FMRP is expressed during development and its expression is downregulated as the brain matures (33). It is possible that astrocytes lack FMRP specifically at a time during development when astrocyte support of neuronal growth and synapse formation are vital (29,32). Because FMRP is a key regulator of translation, FMRP could potentially also regulates a subset of mRNAs in astrocytes (29). Loss of FMRP would then result in abnormal protein translation in astrocytes, interfering with astrocyte-mediated neuronal growth and synaptic development, and contributing to the pathogenesis of the disorder (29). Absence of 113

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

FMRP in astrogenesis also contributes to the enhanced excitability of neurons in FXS. A study by Simhal et al. shows Neurotrophin 3 (NT-3) as a possible mediator of this effect (34). Release of NT-3 is elevated in astrocytes from FMR1 knockout (KO) mice, resulting in abnormal dendritic morphology and synaptic protein expression (34). Loss of FMRP from astrocytes also leads to reduced expression of glutamate transporter GLT-1 (EAAT2) and subsequently reducing glutamate uptake by these cells, which contributes to neuronal hyperactivity and excitotoxicity (34). Thus, astrocyte involvement in FXS pathogenesis is likely through impaired glutamate uptake. Another factor that may contribute to FXS pathogenesis is the astrocyte-secreted extracellular matrix glycoprotein Tenascin C (TNC) (35,36). TNC is an endogenous ligand of toll-like receptor 4 (TLR4) that has been shown to induce the expression of pro-inflammatory cytokines such as IL-6 (35,37). Using FMR1 KO astrocytes from murine models, Krasovska et al. found that secreted TNC and IL-6 were significantly increased, stimulating TLR4 (35). TLR4 activation may influence synaptic development, resulting in abnormal formation and maturation of excitatory synapses in FXS (35). According to this body of literature, it can be concluded that astrocytes are impaired by the genetic mutation on the FMR1 gene, contributing to the pathogenesis of FXS. However, much is still to be learned about abnormal astrogenesis in FXS. -B RS is a X-linked neurodevelopmental disorder that affects around 1 in 12,000 girls (41). Symptoms appear after 6-18 months of seemingly normal development (41). These include seizures, cardiac and breathing problems, repetitive hand movements, and communication difficulties. The disease is predominantly explained by mutations of the methyl-CpG-binding protein 2 (MeCP2) gene, on the X chromosome (42,43). During development, one X chromosome in each somatic cell is randomly inactivated, resulting in a blended expression of both mutant and healthy MeCP2 alleles (41,42). Loss of MeCP2 occurs not only in neurons but also in glial cells, like astrocytes. This mutation is deadly in males, as they do not have another normal functioning MeCP2 gene (41). MeCP2 normally binds to methylated portion of chromatin and can recruit factors to remodel the chromatin into an inactive state (41,42). Specifically, MeCP2 inactivates astrocyte-specific genes so they would not be transcribed at an inappropriate time (41). In normal neurogenesis, MeCP2 binds to methylated portions of astrocyte-specific gene promoters such as GFAP to silence transcription. As development progresses, methylation of the gene decreases and MeCP2 can no longer bind to the promoter and the chromatin remodels into an active state, allowing gene transcription for astrogenesis (44). Expression of MeCP2 in astrocytes is important for differentiation and function. Tightly regulated timing allows for the appropriate number of neurons and astrocytes to be generated during development (41,44). In RS however, the switch to astrogenesis occurs prior to the appropriate time, as MeCP2 is mutated and cannot remodel the promoterâ&#x20AC;&#x2122;s chromatin into an inactive state (41,42) (Figure 6).

114

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

Figure 6. Pathogenesis of Rett Syndrome. Mutation of the MeCP2 allele on one of the X chromosomes causes blended MeCP2 expression. Mutated MeCP2s are unable to remodel an astrocyte promoterâ&#x20AC;&#x2122;s chromatin into an inactive state to allow for neurogenesis to occur during early pregnancy, thus the switch to astrogenesis begins too early. Incorrect neuron and astrocyte numbers being made results in RS, leading to cardiac and breathing problems, seizures, and communication difficulties in females who are affected.

Under normal conditions, healthy astrocytes promote dendritic growth when co-cultured with healthy neurons. However, MeCP2 +/1 mice and healthy neurons resulted in shorter dendrites and somas (43). Astrocytes normally increase the frequency of both excitatory glutamatergic and excitatory GABAergic neuronal synaptic currents in a calcium dependent manner (45). However, in MeCP2 null mice, astrocyte-mediated synaptic modulation is absent and calcium signals in astrocytes are severely blunted (45). Growth rate of MeCP2-deficient astrocytes is significantly slower than normal, and more IL-1Î˛ and IL-6 are released (44,46). They also have altered regulatory effects on neuronal morphology and function, due to certain abnormalities in target gene regulation and toxicity to neurons from abnormal glutamate metabolism (46,47). This suggests that abnormal astrocytes are able to deteriorate the function of neurons. GFAP and S100Î˛ expression are significantly higher in MeCP2-null astrocytes than in normal astrocytes (46,48), suggesting that MeCP2 can couple with the Sin3A/HDAC complex to bind to the GFAP promoter and regulate its transcription (46,48). Additionally, glutamate may induce both neuronal and glial death through excitotoxicity (46,49). While extracellular glutamate concentrations are normally maintained by EAATs of astrocytes, astrogenesis dysregulation results in lower glutamate clearance rates (46,50). Thus, the absence of MeCP2 expression during neurodevelopment has been linked to abnormal astrocyte function that may directly contribute to the pathological process of RS. Overall, abnormal astrogenesis is shown to be induced by genetic mutations in both FXS and RS. FXS is attributed to mutations in the FMR1 gene, which codes for FMRP, an mRNA translation regulator that plays a crucial role in modulating transcription for dendritic growth and development of synapses (29). RS is linked to a mutation in the MeCP2 gene, resulting in mutated MeCP2 that is unable silence transcription of astrocyte-specific promoters during 115

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

neurogenesis (44). Genetic mutations in both FXS and RS potentially results in deficient astrocytes that are unable to regulate synaptic development. Future Directions Understanding of the role of astrocytes in human neurological and psychiatric diseases requires a robust knowledge of astrogenesis and their role in neurodevelopment. Current research has shed much light on the factors regulating the gliogenic switch as well as the pathways to astrogenesis (1-28). However, many questions are unanswered due to the difficult nature of astrocyte research. Astrocytes are difficult to study in in vivo systems because many of the key proteins and genes are expressed by multiple cell types (51). It has been difficult to distinguish contributes of astrocytes from surrounding microglia because they reactive collectively, thus many of the pathways of astrogenesis contain missing pieces (51). For example, scientists still lack detailed understanding of how the pathways simultaneously repress neurogenesis and promote gliogenesis (6). Additionally, many early studies utilize only GFAP, a late astrocyte protein, as an indicator for astrogenesis due to a lack of appropriate markers (10). However, GFAP expression doesnâ&#x20AC;&#x2122;t distinguish astrogenesis from terminal astrocyte differentiation, and many neural precursors also express GFAP, making it not as exclusive to astrocytes as previous thought (10). Limitations also exist in the research methodology used to study astrocytes and their pathways. Current models of signaling and pathogenesis are mostly based on murine models, and scientists are unsure as to how similar human and rodent astrocytes behave, especially in the context of complex disorders like autism (35). Future research should be directed towards understanding the timing of the switch from neurogenesis to astrogenesis. Discovery of new markers of astrocyte maturation and diversity will also help to address heterogeneity in the brain. With regards to psychiatric diseases like FXS and RS, it is necessary to understand how astrocytes modulate synaptic development and function in neural circuits, especially with regards to their role in mediating cognition (34,35,45,46). Progress on better characterizing the role of astrocytes in neurodevelopmental disorders require advancement in our understanding of the genetic factors contributing to neurological diseases and more sophisticated research methods and techniques (6). Recent work has shown the feasibility of deriving functional astrocytes from embryonic stem cells. The in vitro differentiation process has been shown to follow in vivo developmental stages, where stem cell derived neuroepithelial cells transition from multipotent neural progenitors into more restricted astroglial progenitors over time (6). This new process may allow an accessible human cellular system for better understanding the role of astrocytes within neurodevelopment in FXS and RS. Furthermore, using human induced pluripotent stem cell derived astrocytes from affected patients of FXS and RS could help to confirm the observations made in the rodent models and lead to potential therapeutic directions (2). The stem cell cultures are advantageous because patterning molecules can be added during the neuroepithelial stage to specify progenitors, generating a wide variety of astrocyte subtypes to study (2). It may provide

116

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

functionally specific astrocytes for region-specific diseases and allow researchers to identify the molecular basis of the abnormalities that arise from neural diseases (6). Considering our current understanding of the function and morphology of astrocytes, further research is needed to consolidate our knowledge of the role of astrocytes within the CNS, specifically with regards to the timing of astrogenesis, and the particular mechanisms by which astrocytes regulate neuronal development. Conclusion Astrocytes play an important role in the CNS. Besides common functions like modulating the neurovascular blood flow and regulating the extracellular ionic milieu, research has shown that astrocytes shape the synaptic environment and generate signaling mechanisms within neural networks. NSCs first begin to differentiate into neurons during early to mid-gestation, switching to gliogenesis in late gestation. This timing is tightly regulated by external cues and internal cell programming. First, chromatin remodeling occurs in astrocyte-specific genes, like GFAP and S100Î˛. Then, through a combination of Notch, BMP and IL-6 signaling pathways in collaboration with the JAK-STAT pathway, NSC differentiation into astrocytes is promoted. However, genetic mutations that disrupt astrogenesis and subsequent astrocyte function facilitates the onset of neurodevelopmental disorders. FXS is implicated by a mutation in the FMR1 gene that results in the absence of FMRP, which is normally expressed in astrocytes. This interferes with astrocyte-mediated neuronal growth and synaptic development, contributing to the autistic symptoms characteristic of FXS. RS is also due to a genetic mutation in the MeCP2 gene, leading to loss of MeCP2 in astrocytes. Absence of MeCP2 contributes to various abnormalities in astrocyte function. Therefore, it is important to characterize the various stages of astrogenesis in order to understand the mechanisms behind neurodevelopmental disorders, and which pathways to target for therapeutic treatment.

117

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

References 1.

Bayraktar OA, Fuentealba LC, Alvarez-Buylla A, Rowitch DH. Astrocyte development and heterogeneity. Cold Spring Harbor perspectives in biology. 2015 Jan 1;7(1):a020362.

Sloan SA, Barres BA. Mechanisms of astrocyte development and their contributions to neurodevelopmental disorders. Current opinion in neurobiology. 2014 Aug 1;27:75-81. DOI: 10.1016/j.conb.2014.03.005

Clarke LE, Barres BA. Emerging roles of astrocytes in neural circuit development. Nature Reviews Neuroscience. 2013 May;14(5):311.

Chung WS, Clarke LE, Wang GX, Stafford BK, Sher A, Chakraborty C, Joung J, Foo LC, Thompson A, Chen C, Smith SJ. Astrocytes mediate synapse elimination through MEGF10 and MERTK pathways. Nature. 2013 Dec;504(7480):394.

Namihira M, Kohyama J, Semi K, Sanosaka T, Deneen B, Taga T, Nakashima K. Committed neuronal precursors confer astrocytic potential on residual neural precursor cells. Developmental cell. 2009 Feb 17;16(2):245-55.

Molofsky AV, Krenick R, Ullian E, Tsai HH, Deneen B, Richardson WD, Barres BA, Rowitch DH. Astrocytes and disease: a neurodevelopmental perspective. Genes & development. 2012 May 1;26(9):891-907. DOI: 10.1101/gad.188326.112

Miller RH, Raff MC. Fibrous and protoplasmic astrocytes are biochemically and developmentally distinct. Journal of Neuroscience. 1984 Feb 1;4(2):585-92.

Bushong EA, Martone ME, Jones YZ, Ellisman MH. Protoplasmic astrocytes in CA1 stratum radiatum occupy separate anatomical domains. Journal of Neuroscience. 2002 Jan 1;22(1):183-92.

Taga T, Fukuda S. Role of IL-6 in the neural stem cell differentiation. Clinical reviews in allergy & immunology. 2005 Jun 1;28(3):249-56. Available from: https://link.springer.com/content/pdf/10.1385/CRIAI:28:3:249.pdf

10.

Miller FD, Gauthier AS. Timing is everything: making neurons versus glia in the developing cortex. Neuron. 2007 May 3;54(3):357-69. DOI: https://doi.org/10.1016/j.neuron.2007.04.019

11.

Takizawa T, Nakashima K, Namihira M, Ochiai W, Uemura A, Yanagisawa M, Fujita N, Nakao M, Taga T. DNA methylation is a critical cell-intrinsic determinant of astrocyte differentiation in the fetal brain. Developmental cell. 2001 Dec 1;1(6):749-58.

12.

Edlund T, Jessell TM. Progression from extrinsic to intrinsic signaling in cell fate specification: a view from the nervous system. Cell. 1999 Jan 22;96(2):211-24.

118

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

13.

Kamakura S, Oishi K, Yoshimatsu T, Nakafuku M, Masuyama N, Gotoh Y. Hes binding to STAT3 mediates crosstalk between Notch and JAK–STAT signalling. Nature cell biology. 2004 Jun;6(6):547.

14.

Gaiano N, Fishell G. The role of notch in promoting glial and neural stem cell fates. Annual review of neuroscience. 2002 Mar;25(1):471-90.

15.

Urayama S, Semi K, Sanosaka T, Hori Y, Namihira M, Kohyama J, Takizawa T, Nakashima K. Chromatin accessibility at a STAT3 target site is altered prior to astrocyte differentiation. Cell structure and function. 2013:12034.

16.

Takouda J, Katada S, Nakashima K. Emerging mechanisms underlying astrogenesis in the developing mammalian brain. Proceedings of the Japan Academy, Series B. 2017 Jun 9;93(6):386-98. DOI: 10.2183/pjab.93.024

17.

Cebolla B, Vallejo M. Nuclear factor‐I regulates glial fibrillary acidic protein gene expression in astrocytes differentiated from cortical precursor cells. Journal of neurochemistry. 2006 May;97(4):1057-70.

18.

Bonni A, Sun Y, Nadal-Vicens M, Bhatt A, Frank DA, Rozovsky I, Stahl N, Yancopoulos GD, Greenberg ME. Regulation of gliogenesis in the central nervous system by the JAKSTAT signaling pathway. Science. 1997 Oct 17;278(5337):477-83.

19.

Esther LB, Carla AR, Bruno LB, Dinorah HM, Leticia R, Arturo O. Notch Signaling in the Astroglial Phenotype: Relevance to Glutamatergic Transmission. GABA And Glutamate: New Developments In Neurotransmission Research. 2018 Mar 21:25. DOI: 10.5772/intechopen.73318

20.

Pinto L, Götz M. Radial glial cell heterogeneity—the source of diverse progeny in the CNS. Progress in neurobiology. 2007 Sep 1;83(1):2-3.

21.

Anthony TE, Klein C, Fishell G, Heintz N. Radial glia serve as neuronal progenitors in all regions of the central nervous system. Neuron. 2004 Mar 25;41(6):881-90.

22.

Kornblum HI, Hussain R, Wiesen J, Miettinen P, Zurcher SD, Chow K, Derynck R, Werb Z. Abnormal astrocyte development and neuronal death in mice lacking the epidermal growth factor receptor. Journal of neuroscience research. 1998 Sep 15;53(6):697-717. DOI: https://doi.org/10.1002/(SICI)1097-4547(19980915)53:6<697::AID-JNR8>3.0.CO;2-0

23.

Shibata T, Yamada K, Watanabe M, Ikenaka K, Wada K, Tanaka K, Inoue Y. Glutamate transporter GLAST is expressed in the radial glia–astrocyte lineage of developing mouse spinal cord. Journal of Neuroscience. 1997 Dec 1;17(23):9212-9.

24.

Barnabé-Heider F, Wasylnka JA, Fernandes KJ, Porsche C, Sendtner M, Kaplan DR, Miller FD. Evidence that embryonic neurons regulate the onset of cortical gliogenesis via cardiotrophin-1. Neuron. 2005 Oct 20;48(2):253-65.

119

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

25.

He F, Ge W, Martinowich K, Becker-Catania S, Coskun V, Zhu W, Wu H, Castro D, Guillemot F, Fan G, De Vellis J. A positive autoregulatory loop of Jak-STAT signaling controls the onset of astrogliogenesis. Nature neuroscience. 2005 May;8(5):616.

26.

Murphy M, Dutton R, Koblar S, Cheema S, Bartlett P. Cytokines which signal through the LIF receptor and their actions in the nervous system. Progress in neurobiology. 1997 Aug 1;52(5):355-78.

27.

Kageyama R, Ohtsuka T, Hatakeyama J, Ohsawa R. Roles of bHLH genes in neural stem cell differentiation. Experimental cell research. 2005 Jun 10;306(2):343-8.

28.

Muroyama Y, Fujiwara Y, Orkin SH, Rowitch DH. Specification of astrocytes by bHLH protein SCL in a restricted region of the neural tube. Nature. 2005 Nov;438(7066):360.

29.

Jacobs S, Doering LC. Astrocytes prevent abnormal neuronal development in the fragile x mouse. Journal of Neuroscience. 2010 Mar 24;30(12):4508-14. DOI: https://doi.org/10.1523/JNEUROSCI.5027-09.2010

30.

Beckelâ&#x20AC;?Mitchener A, Greenough WT. Correlates across the structural, functional, and molecular phenotypes of fragile X syndrome. Mental retardation and developmental disabilities research reviews. 2004 Feb;10(1):53-9.

31.

Hagerman PJ, Stafstrom CE. Origins of epilepsy in fragile X syndrome. Epilepsy Currents. 2009 Jul;9(4):108-12.

32.

Jacobs S, Nathwani M, Doering LC. Fragile X astrocytes induce developmental delays in dendrite maturation and synaptic protein expression. BMC neuroscience. 2010 Dec;11(1):132.

33.

Pacey LK, Doering LC. Developmental expression of FMRP in the astrocyte lineage: implications for fragile X syndrome. Glia. 2007 Nov 15;55(15):1601-9.

34.

Simhal AK, Zuo Y, Perez MM, Madison DV, Sapiro G, Micheva KD. Multifaceted Changes in Synaptic Composition and Astrocytic Involvement in a Mouse Model of Fragile X Syndrome. bioRxiv. 2019 Jan 1:615930. DOI: 10.1038/s41598-019-50240-x

35.

Krasovska V, Doering L. Regulation of IL-6 secretion by astrocytes via TLR4 in the fragile X mouse model. Frontiers in molecular neuroscience. 2018;11:272. DOI: https://doi.org/10.3389/fnmol.2018.00272

36.

Jones EV, Bouvier DS. Astrocyte-secreted matricellular proteins in CNS remodelling during development and disease. Neural plasticity. 2014;2014.

37.

Maqbool A, Spary EJ, Manfield IW, Ruhmann M, Zuliani-Alvarez L, Gamboa-Esteves FO, Porter KE, Drinkhill MJ, Midwood KS, Turner NA. Tenascin C upregulates

120

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

interleukin-6 expression in human cardiac myofibroblasts via toll-like receptor 4. World journal of cardiology. 2016 May 26;8(5):340. 38.

Wei H, Chadman KK, McCloskey DP, Sheikh AM, Malik M, Brown WT, Li X. Brain IL-6 elevation causes neuronal circuitry imbalances and mediates autism-like behaviors. Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease. 2012 Jun 1;1822(6):831-42.

39.

Wei H, Zou H, Sheikh AM, Malik M, Dobkin C, Brown WT, Li X. IL-6 is increased in the cerebellum of autistic brain and alters neural cell adhesion, migration and synaptic formation. Journal of neuroinflammation. 2011 Dec;8(1):52.

40.

Shen Y, Qin H, Chen J, Mou L, He Y, Yan Y, Zhou H, Lv Y, Chen Z, Wang J, Zhou YD. Postnatal activation of TLR4 in astrocytes promotes excitatory synaptogenesis in hippocampal neurons. J Cell Biol. 2016 Dec 5;215(5):719-34.

41.

King M. Astrocyte in Rett Syndrome: why we shouldn’t ignore them. AXOL. 2018. Available from: https://www.axolbio.com/blog/guest-post-astrocytes-in-rett-syndromewhy-we-shouldnt-ignore-them

42.

Ballas N, Lioy DT, Grunseich C, Mandel G. Non–cell autonomous influence of MeCP2deficient glia on neuronal dendritic morphology. Nature neuroscience. 2009 Mar;12(3):311.

43.

Yasui DH, Xu H, Dunaway KW, LaSalle JM, Jin LW, Maezawa I. MeCP2 modulates gene expression pathways in astrocytes. Molecular autism. 2013 Dec;4(1):3.

44.

Lioy DT, Garg SK, Monaghan CE, Raber J, Foust KD, Kaspar BK, Hirrlinger PG, Kirchhoff F, Bissonnette JM, Ballas N, Mandel G. A role for glia in the progression of Rett’s syndrome. Nature. 2011 Jul;475(7357):497.

45.

Rakela B, Brehm P, Mandel G. Astrocytic modulation of excitatory synaptic signaling in a mouse model of Rett syndrome. Elife. 2018 Jan 9;7:e31629. DOI: 10.7554/eLife.31629

46.

Jin XR, Chen XS, Xiao L. MeCP2 deficiency in neuroglia: new progress in the pathogenesis of Rett syndrome. Frontiers in molecular neuroscience. 2017 Oct 4;10:316. DOI: https://doi.org/10.3389/fnmol.2017.00316

47.

Williams EC, Zhong X, Mohamed A, Li R, Liu Y, Dong Q, Ananiev GE, Mok JC, Lin BR, Lu J, Chiao C. Mutant astrocytes differentiated from Rett syndrome patients-specific iPSCs have adverse effects on wild-type neurons. Human molecular genetics. 2014 Jan 12;23(11):2968-80.

48.

Forbes-Lorman RM, Kurian JR, Auger AP. MeCP2 regulates GFAP expression within the developing brain. Brain research. 2014 Jan 16;1543:151-8.

121

MUMJ Volume 17 No. 1, pp. 103-121

June 2020

49.

Lehmann C, Bette S, Engele J. High extracellular glutamate modulates expression of glutamate transporters and glutamine synthetase in cultured astrocytes. Brain research. 2009 Oct 22;1297:1-8.

50.

Okabe Y, Takahashi T, Mitsumasu C, Kosai KI, Tanaka E, Matsuishi T. Alterations of gene expression and glutamate clearance in astrocytes derived from an MeCP2-null mouse model of Rett syndrome. PloS one. 2012 Apr 20;7(4):e35354.

51.

Liddelow SA, Barres BA. Reactive astrocytes: production, function, and therapeutic potential. Immunity. 2017 Jun 20;46(6):957-67.

122

MUMJ Volume 17 No. 1, pp. 122-127

June 2020

Commentary

Float or sink: A solution to the resident burn-out crisis? Candice Luo, George Hu, and Tony Chen Michael G. DeGroote School of Medicine, McMaster University

Keywords: Medical education, Residency, Float system, Burn out Corresponding author: candice.luo@medportal.ca

123

MUMJ Volume 17 No. 1, pp. 122-127

June 2020

Introduction As medical students, we were always told that medicine is a tough road. The training is gruelling with endless hours of studying, patient care, and considerable sacrifices in our personal lives. Many came into this profession ready for long hours and difficult training. However, until residency, it is tough to envision what the experience really entails. Recently, a debate erupted in the medical Twitter universe when Dr. Colleen Farrell, an Internal Medicine resident at Bellevue Hospital in New York, wrote a tweet decrying 27-hour resident call shifts as inhumane. Dr. Farrell argued that residents and staff physicians deserve protection against harsh working hours and conditions on par with workers in unionized professions. As a local example, the Ontario Nurses Association closely regulates how long nurses can work on a given day with the minimum time nurses must receive for breaks (1). The ensuing debate saw numerous residents and staff physicians joining the conversation on either side of the argument. Some physicians argued that extended call shifts are a necessary part of resident training, which equips residents to work effectively and independently in future demanding roles. On the other hand, others suggested that the lack of adequate rest and humane working hours leaves residents ill-prepared to make decisions and may hinder patient care. In the end, Dr. Farrell received heavy backlash which led to her taking a break from Twitter. A similar challenge exists in Canada, with research suggesting that Canadian medical residents also work long hours and struggle with mental wellness (2). But what does the research say about changes to residency hours? Does reducing work hours improve resident wellness and patient safety? What is the impact on resident education? Can float systems be comparable to call systems in domains of resident wellness, competency, and patient safety? System responses In 2003, residents across Canada were mandated to an 80-hour work limit on average over a 4-week period. This was followed in 2011 by a second mandate limiting first-year resident call shifts to a maximum of 16 hours. Soon after, all Quebec medical residents, regardless of their training level, had a similar 16-hour limit implemented for in-house call. McMaster University was one of the first schools to address these concerns at the program level. For instance, the McMaster Anesthesia Program has reduced call shifts to a 16hour maximum for all residents. Taking an alternate approach, McMasterâ&#x20AC;&#x2122;s pediatrics residency program introduced the float system in 2013-2014 (3). The float system replaces conventional 26-hour call shifts with residents working consecutive night-only shifts for a set period, alternating with a period of daytime service. However, the float system was not new at the time. Studies have shown, as early as 1991, that switching to a float system improved staff morale, reduced working hours, and did not have a detrimental effect on patient care (4). Previous research

124

MUMJ Volume 17 No. 1, pp. 122-127

June 2020

Three review articles have examined the effects of residency shifts with defined lengths, night float, or protected sleep times on patient outcomes (5-7). Two of these reviews found no effect on patient mortality with these changes (5-7), while the third cited evidence of improvement in patient mortality (6). In addition, while there was a decrease in medical errors in general upon imposing shift limits, there was increased patient complication rates specifically in specialties with high acuity, suggesting that the effects of shift limits are program-dependent (7). In 2015, a prospective study was published after the McMaster Anesthesia Residency Program transitioned to a 16-hour call system in 2013 (8). The majority of residents supported the change, citing improved quality of life and less fatigue. Most respondents also believed that there was improved or unchanged patient safety and quality of education. Studies conducted on the night float system implemented at Queenâ&#x20AC;&#x2122;s and Dalhousie University likewise found improved resident satisfaction and quality of life (9,10). Although research into Canadian residents participating in a float system found improved learning by self-report, other international studies have found residents reporting equivocal or decreased education quality, possibly related to difficulties in attending daytime teaching while working night shifts (9-11). Two large randomized studies from American internal medicine and general surgery programs compared examination scores between flexible and standard-hour residency programs and found no significant differences between the two groups (12,13). Furthermore, studies have largely found that the total number of hours completed by residents are comparable between the float system and the call system, suggesting the quality of education did not differ between the two systems (9-14). These studies, though difficult to generalize, suggest that the quality of education does not have to be sacrificed when adequate measures are taken in a float system. The downsides to a float system, however, should not be minimized. Previous studies in internal medicine have suggested that a float system might decrease continuity of care, cause inadequate teaching of the night float staff, and increase miscommunication (4). Due to lack of continuity of care, residents on the night float staff do not participate in the decision-making process for their patientâ&#x20AC;&#x2122;s care, thus removing them from potential learning opportunities (4). However, the McMaster Pediatrics Program has attempted to address these issues through having specific night float teaching so that the residents still obtain adequate education (15). Although there may be fewer errors associated with fatigue, night float is also linked to increased errors due to more frequent and inadequate handover, and decrease in continuity of care (16,17). Several authors also point to the potential risk to quality of life due to chronic night-time work and one study in particular demonstrated a decrease in perceived quality of life upon switching to a night float system (18). Possible explanations for this include the change in sleep habits upon switching from regular work hours to night shifts, as well as the loss of post-call rest days. The presence of this conflicting evidence suggests that the purported benefits of the night float system are not universal, and programs should consider, on a case-by-case basis, which model is most suitable for their residents.

125

MUMJ Volume 17 No. 1, pp. 122-127

June 2020

Conclusion The response to these research findings has been a slow process. Currently, most residency programs in Canada still implement the traditional call system, and implementation of night float systems remains a topic of debate between residency programs and residents themselves. Call systems have existed long before float systems and have indisputably produced many exemplary, competent, and compassionate physicians. Yet, it is also known that inappropriate working conditions lead to burn out and apathy. The research suggests that reducing hours or changing to a more flexible schedule may lead to improvements in perceptions of resident wellness. Furthermore, there does not seem to be a significant impact on resident education, though the impact on patient care remains unclear. Future research examining mixed models with aspects of night float and call system may help clarify whether alternate approaches might reduce residency burnout whilst not sacrificing education quality or patient safety. Dr. Colleen Farrell is likely not a lone voice, but presenting a sentiment shared amongst many residents undergoing their training. Although there is not yet a definitive solution, positive steps have been taken. With time, and more research, individualized solutions for different residencies which optimize for resident wellness, education, and patient care can be found. Author biographies Candice Luo is a second-year medical student at McMaster University. She currently attends the Hamilton Campus and has an interest in medical trainee mental health and wellness. George Hu is a second-year medical student at McMaster University. He currently attends the Niagara Regional Campus and has an interest in medical education. Tony Chen is a second-year student in the MD-PhD program at McMaster University. He has an interest in knowledge translation.

126

MUMJ Volume 17 No. 1, pp. 122-127

June 2020

References 1.

ONA Hospital Central Agreement [Internet]. Collective Agreement. Ontario Nurses' Association; 2018 [cited 2019 Dec 29]. Available from: https://www.ona.org/wpcontent/uploads/ona_hospitalcentralagreement_20180331.pdf

Taher A, Hart A, Dattani ND, Poonja Z, Bova C, Bandiera G, et al. Emergency medicine resident wellness: Lessons learned from a national survey. CJEM. 2018 Sep;20(5):721–4.

Excellence in Education. McMaster University Department of Pediatrics 2013 & 2014 Status Report. 2015 Jun 10;:46-7. Available from: https://issuu.com/pedscom/docs/pediatrics_status_report_2013_14_we.

Trontell MC, Carson JL, Taragin MI. The impact of the night float system on internal medicine residency programs. J Gen Intern Med. 1991 Sep;6(5):445–9.

Fletcher KE, Davis SQ, Underwood W, Mangrulkar RS, McMahon LF, Saint S. Systematic review: effects of resident work hours on patient safety. Ann Intern Med 2004 Dec;141(11):851–7.

Reed DA, Fletcher KE, Arora VM. Systematic review: association of shift length, protected sleep time, and night float with patient care, residents’ health, and education. Ann Intern Med. 2010 Dec;153(12):829–42.

Ahmed N, Devitt KS, Keshet I, Spicer J, Imrie K, Feldman L, et al. A systematic review of the effects of resident duty hour restrictions in surgery: impact on resident wellness, training, and patient outcomes. Ann Surg. 2014 Nov;259:1041–53.

Sussman D, Paul JE. The impact of transitioning from a 24-hour to a 16-hour call model amongst a cohort of Canadian anesthesia residents at McMaster University-a survey study. Adv Med Educ Pract. 2015 Aug;6:501–6.

Mann SM, Borschneck DP, Harrison MM. Implementation of a novel night float call system: resident satisfaction and quality of life. Can J Surg. 2014 Feb;57(1):15–20.

10.

Moeller A, Webber J, Epstein I. Resident duty hour modification affects perceptions in medical education, general wellness, and ability to provide patient care. BMC Med Educ. 2016 Dec;16(1):175.

11.

Drolet BC, Spalluto LB, Fischer SA. Residents' perspectives on ACGME regulation of supervision and duty hours -- a national survey. N Engl J Med. 2010 Dec;363(23):e34.

127

MUMJ Volume 17 No. 1, pp. 122-127

June 2020

12.

Desai SV, Asch DA, Bellini LM, Chaiyachati KH, Liu M, Sternberg AL, et al. Education outcomes in a duty-hour flexibility trial in internal medicine. N Engl J Med. 2018 Apr;378(16):1494–508.

13.

Bilimoria KY, Chung JW, Hedges LV, Dahlke AR, Love R, Cohen ME, et al. National cluster-randomized trial of duty-hour flexibility in surgical training. N Engl J Med. 2016 Feb;374(8):713–27.

14.

Ragel BT, Piedra M, Klimo P, Burchiel KJ, Waldo H, McCartney S, et al. An ACGME duty hour compliant 3-person night float system for neurological surgery residency programs. J Grad Med Educ. 2014 Jun;6(2):315–9.

15.

The McMaster at Night Pediatric Curriculum [Internet]. McMaster Pediatrics Residency Program. [cited 2019 Dec 28]. Available from: https://www.macpeds.com/documents/_baks/McMasteratNightDescription.pdf.0001.22a4.bak

16.

Drolet BC, Christopher DA, Fischer SA. Residents' response to duty-hour regulations - a follow-up national survey. N Engl J Med. 2012 Jun;366(24):e35.

17.

Jagsi R, Shapiro J, Weissman JS, Dorer DJ, Weinstein DF. The educational impact of ACGME limits on resident and fellow duty hours: a pre-post survey study. Acad Med. 2006 Dec;81(12):1059–68.

18.

Zahrai A, Chahal J, Stojimirovic D, Schemitsch EH, Yee A, Kraemer W. Quality of life and educational benefit among orthopedic surgery residents: a prospective, multicentre comparison of the night float and the standard call systems. Can J Surg. 2011 Feb;54(1):25–32.

128

MUMJ Volume 17 No. 1, pp. 128-133

June 2020

Commentary

Ceteris Paribus? â&#x20AC;&#x201C; An epistemological error with ethical consequences Chris Arsenault McMaster University

Abstract Herein I call into question a common epistemological justification for placebo controls, and thereby problematize the use of placebo in many modern clinical trials. I demonstrate both the ethical harm and epistemic inferiority of placebo controls in certain knowledge contexts, arguing the standard of care should be the more acceptable comparator for novel treatments in such contexts. Keywords: Placebo, Epistemology Corresponding author: chris.arsenault@medportal.ca

129

MUMJ Volume 17 No. 1, pp. 128-133

June 2020

Placebo is a treatment with no therapeutic effect used as a control in many clinical trials, and research ethics boards (REBs) must ensure its use is justified (1,2). Placebo can provide us with special knowledge about studied treatments, but it carries risks because it can undertreat patients. I will argue a recent study – Croft et al. 2014 (3) – makes unjustified use of placebos. I will then argue that an epistemological error – belief that placebo always yields more useful knowledge than other controls by isolating pharmacologic effect – can cause unjustified use of placebos. This exposes study participants to unnecessary risk and limits usefulness of study results. Vilazodone is a novel antidepressant with selective serotonin reuptake inhibitor activity and partial 5-HT1A receptor agonism (4). Croft et al. 2014 is a randomized, double-anonymized (conventionally “double-blind”, but see Tremain (5,6)), placebo-controlled stage IV clinical trial of vilazodone. I use this example because in December 2019, I attended a continuing medical education talk about vilazodone during a clerkship elective in Ontario. This study was discussed there as evidence that vilazodone may be effective treatment for some major depressive disorder (MDD) cases, so it is a recent study having current impact on clinical practice in Canada. The study selects participants who have moderate to severe MDD, have not failed more than one treatment, and do not have significant medical or psychiatric confounds. In other words, they are suffering from MDD, have not had adequate trials of first-line medications in this episode of illness, and are uncomplicated cases. These patients are ideal candidates for first-line antidepressants by current clinical practice guidelines (7,8). Instead of giving the participants recommended treatments, however, the study gives them either vilazodone or placebo. Being an under-characterized medication at the time, vilazodone’s effects on MDD are uncertain and potentially effective, which is why it is being studied. In research ethics terms, vilazodone is in equipoise: “Clinical equipoise means a genuine uncertainty exists on the part of the relevant expert community about what interventions are most effective for a given condition” (1). Vilazodone is in equipoise during this trial, but placebo is not – it is known to be inferior to first-line antidepressants (7,8). This study selected patients who should, in the clinical and ethical senses, be treated with certain medications, and instead gave some placebo, known to be less effective. This knowingly risks harming this study’s placebo-control group by giving them treatment not in equipoise with the standard of care (undertreatment), resulting in higher risk of persistent and worsening MDD. Some argue that placebo control is not as harmful as it seems because of various safeguards (which mitigate but do not eliminate harms), and because the study participants might receive no treatment if not for their participation in the study (9). Although a study’s potential benefits to participants who would otherwise receive no treatment may justify a study (but see (1,2) on justice), it cannot justify choosing placebo over a different comparator. Potential benefits to those who would otherwise receive nothing cannot justify choosing placebo because researchers are not choosing between giving participants placebo versus not having a study at all – they are choosing between giving participants placebo versus giving them the standard of care. Placebo is undertreatment compared to what the participants could receive had researchers

130

MUMJ Volume 17 No. 1, pp. 128-133

June 2020

chosen a different study design, so researchers have chosen undertreatment, a potential harm. Patients making informed consent to these risks might mitigate the harm to autonomy of being deceived, but not the harm to body/mind of being undertreated. How do we justify this? The Declaration of Helsinki, which many REB guidelines including Canada’s TCPS2 mirror, has specific criteria. Croft et al. 2014 state their study “… was conducted… in full compliance with… the ethical principles of the Declaration of Helsinki” (3). The Declaration’s Article 33 elaborates two criteria by which placebo controls may be appropriate. The first is that “…no proven intervention exists” (2), which is not the case with MDD, as we have evidence of many placebo-superior options. To have passed REB review, it must then meet the second criterion. To do so it must satisfy two conditions: 1. There is “compelling and scientifically sound methodological reason” (2) to use placebo AND 2. Patients receiving placebo “will not be subject to additional risks of serious or irreversible harm” (2) by not receiving the standard of care. It further states “Extreme care must be taken to avoid abuse of this option” (2). Whether the “serious or irreversible” clause of condition 2 is satisfied is beyond my scope. Condition 1 is our focus because it asks why placebo is needed in the first place. Placebo controls are used because they provide special knowledge about the studied treatment. In this study, using placebo control addresses whether vilazodone has more effect than the placebo effect, i.e. whether it is having effect ceteris paribus (all else held equal). It does so by measuring the effect of placebo (which any treatment including vilazodone will have) to subtract from vilazodone’s measured effect. If vilazodone performs better than placebo, it has another effect in addition to the placebo effect – it has pharmacologic effect. Placebo control is a part of the epistemology of holding all other factors equal to determine treatment effect in isolation, like randomization and double-anonymizing. It tells us something special about the drug itself by isolating its effect from other potential effects on the measured outcomes. Thus, there is epistemic value to the placebo control group – it shows whether vilazodone has pharmacologic effect. For conditions without effective treatments placebo is in equipoise, and so can be used without knowingly depriving patients of a better treatment option. In such contexts, placebo will also give useful information about the studied treatment – whether it has some pharmacologic effect. Having pharmacologic effect would make it the most effective known treatment, as no placebo-superior treatments are yet known. In a context where known placebo-superior treatments exist, however, this is not true because placebo is no longer in equipoise. Vilazodone is in this latter context, where both epistemic and ethical considerations change. Ethically, we have superior options for these patients (e.g. citalopram (7,8)) and yet are knowingly depriving them of these which is potentially harmful. Epistemically, a placebo-controlled trial will tell us whether vilazodone is better than placebo but will not tell us how it performs compared to the standard of care.

131

MUMJ Volume 17 No. 1, pp. 128-133

June 2020

Why does this matter? When no known effective treatments exist any effect greater than placebo (given an acceptable safety profile) demonstrates clinical usefulness, but this is not the case when effective treatments are known. The ground under us has shifted – ceteris paribus is no longer placebo, but the standard of care, because this is what the normal patient should receive as treatment. New treatments do not need to be safer and more effective than placebo to be clinically useful now – they need to be safer and more effective, or at least comparable to, existing treatments. When there are known placebo-superior treatment options we must contextualize efficacy relative to these treatments because it shows us when to use the study treatment instead of another established one. Placebo only tells us that there is some amount of pharmacologic effect, which is less clinically useful knowledge. Croft et al. 2014 asks whether vilazodone is superior and its safety profile noninferior to placebo, but the clinician and patient ask whether it is as effective and safe as other options. The study asks whether there is a mechanism of action apart from placebo, but the patient and clinician ask when they should prefer it over other treatments. The study cannot answer our questions, telling us only that there is some pharmacologic effect. This is clinically inferior knowledge, and it comes at the additional cost of knowingly exposing patients to undertreatment. One could object MDD has a high rate of non-responders to existing therapies, so we should explore novel options. Here placebo returns to equipoise as there are, by definition, no known effective therapies. But this study is explicitly not a study of treatment-resistant depression – it sheds no light on what we could expect for non-responders were we to prescribe them vilazodone, because it excludes that population. Another objection is that the highly variable treatment response of MDD requires a placebo comparator in addition to a standard of care comparator to ensure the standard of care is placebo-superior in the study population (10,11). This may be a justification for a trial like Mathews et al. 2015 where vilazodone is compared to both placebo and citalopram (12). Croft et al. 2014, however, fails to use both comparators, thereby failing to contextualize its placebo control relative to the standard of care. Thus, it is not using placebo in this way. Compared to a study using the standard of care, Croft et al. 2014’s use of placebo provides less clinically useful information and risks doing more harm to study participants by undertreating them. This is unjustifiable by accepted research ethics guidelines. I do not have scope to demonstrate how common this is, but I invite the reader to review whichever clinical literatures they are engaged with and ask the same questions posed herein.

132

MUMJ Volume 17 No. 1, pp. 128-133

June 2020

References 1.

Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada. TriCouncil Policy Statement: Ethical Conduct for Research Involving Humans [Internet]. 2018 [cited 2019 Dec 31]. Available from: https://ethics.gc.ca/eng/policy-politique_tcps2eptc2_2018.html

World Medical Association. WMA Declaration of Helsinki - Ethical Principles for Medical Research Involving Human Subjects [Internet]. Bulletin of the World Health Association. 2018 [cited 2019 Dec 31]. Available from: https://www.wma.net/policies-post/wmadeclaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects/

Croft HA, Pomara N, Gommoll C, Chen D, Nunez R, Mathews M. Efficacy and safety of vilazodone in major depressive disorder: a randomized, double-blind, placebo-controlled trial. J Clin Psychiatry. 2014 Nov;75(11):e1291-1298.

Dawson LA, Watson JM. Vilazodone: A 5‐HT1A Receptor Agonist/Serotonin Transporter Inhibitor for the Treatment of Affective Disorders. CNS Neurosci Ther. 2009 Mar 16;15(2):107–17.

Tremain S. Ableist language and philosophical associations [Internet]. New APPS: Art, Politics, Philosophy, Science. 2011 [cited 2019 Dec 31]. Available from: https://www.newappsblog.com/2011/07/ableist-language-and-philosophicalassociations.html

Tremain S. Foucault and Feminist Philosophy of Disability. Ann Arbor: University of Michigan Press; 2017.

Kennedy SH, Lam RW, McIntyre RS, Tourjman SV, Bhat V, Blier P, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 Clinical Guidelines for the Management of Adults with Major Depressive Disorder. Can J Psychiatry. 2016 Sep;61(9):540–60.

Lam RW, Kennedy SH, Grigoriadis S, McIntyre RS, Milev R, Ramasubbu R, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT) Clinical guidelines for the management of major depressive disorder in adults.: III. Pharmacotherapy. Journal of Affective Disorders. 2009 Oct 1;117:S26–43.

Feifel D. The Use of Placebo-Controlled Clinical Trials for the Approval of Psychiatric Drugs. Psychiatry (Edgmont). 2009 Dec;6(12):19–25.

10.

Millum J, Grady C. The ethics of placebo-controlled trials: Methodological justifications. Contemporary Clinical Trials. 2013 Nov 1;36(2):510–4.

11.

Feifel D. The Use of Placebo-Controlled Clinical Trials for the Approval of Psychiatric Drugs. Psychiatry (Edgmont). 2009 Mar;6(3):41–3.

133

MUMJ Volume 17 No. 1, pp. 128-133 12.

June 2020

Mathews M, Gommoll C, Chen D, Nunez R, Khan A. Efficacy and safety of vilazodone 20 and 40â&#x20AC;&#x2030;mg in major depressive disorder: a randomized, double-blind, placebo-controlled trial. Int Clin Psychopharmacol. 2015 Mar;30(2):67â&#x20AC;&#x201C;74.

134

MUMJ Volume 17 No. 1, pp. 134-141

June 2020

Commentary

The Natural History of Medical Waste Yu Fei Xia and Betty Hui Yu Zhang McMaster University

Keywords: Medical waste, Waste minimization, Health policy

Corresponding author: betty.zhang@medportal.ca

135

MUMJ Volume 17 No. 1, pp. 134-141

June 2020

Introduction The healthcare industry is the second largest contributor to landfills after the food industry (1). This is a product of the increase in single-use medical equipment and increasingly stringent recycling protocols (2,3). Medical waste refers to all by-products generated in the care of a patient (4). More than 85% of medical waste is considered general waste and does not pose a direct hazard to human health. The remaining 15% is considered hazardous waste and comprises infectious, pathological, chemical, cytotoxic and radioactive waste. These need further processing prior to disposal (Figure 1). Optimal medical waste management can help limit its impact on the environment and surrounding community. However, there are many factors that play a role in the efficiency of medical waste processing and recycling. There is limited data available on medical waste management in Canada and as such we will refer to international statistics when necessary to contextualize the issues raised. For many, education about medical waste management is the first step towards reducing its harmful impact and implementing environmentally friendly alternatives. The challenges and costs of medical waste The challenges posed by medical waste result from both the sheer volume of waste produced and the necessity of additional processing for hazardous waste. Among common operating room procedures, a two-hour hysterectomy produces 10kg of medical waste, leading to a hefty combination of plastics, packaging, and drapes - roughly five times what an individual generates each day (5,6). At Hamilton Health Sciences, 11kg of medical waste are produced daily for each patient, leading to an annual production of over 500 tonnes of hazardous waste (7). In the United States (US), the cost to dispose of general waste is $0.12/kg compared to $0.79/kg for hazardous waste â&#x20AC;&#x201C; a difference of 560% (8). This is due to the fact that hazardous waste requires expensive technologies such as autoclaving and incineration, compared to the simpler landfill disposal of general waste. However, up to 85% of the products disposed of as hazardous waste are actually appropriate for general waste (4). Thus, segregation of waste to the appropriate disposal pathway is critical and something often overlooked in a busy hospital. When waste is improperly segregated, it can have significant financial consequences at a systems level. A United Kingdom (UK) audit isolated one area with the greatest potential for improvementsâ&#x20AC;&#x201D;anesthetic waste (2). The packaging of syringes and the glass vials for medications are potentially recyclable materials that can accumulate to 950kg per operating room per year. When the auditors examined the sharps bin, which should be one of the most streamlined forms of medical waste disposal, only 4% were truly sharps waste. The audit suggested that starting to recycle anesthetic waste alone would save 30% of the annual hospital budget allocated for disposing of clinical waste. Given the similarities of our system, these analyses suggest that improving medical waste segregation could have similar effects in Canada.

136

MUMJ Volume 17 No. 1, pp. 134-141

June 2020

Figure 1. Breakdown of the different components of medical waste according to World Health Organization (WHO) definitions (4).

Cultural attitudes and legislation Cultural attitudes and local regulations concerning waste management can have an important impact on the production and disposal of medical waste. For instance, despite the fact that both the UK and Germany have similar sterility standards, they produce drastically different amounts of waste (2). In the UK, up to 5.5kg of medical waste is produced daily for each patient 137

MUMJ Volume 17 No. 1, pp. 134-141

June 2020

compared to 1.9kg in Germany (2). This has been attributed in part to a strong cultural and historical emphasis on recycling in Germany, leading it to recycle 55% of all waste - more than any other country (9). However, to be effective, cultural attitudes must permeate to institutional leadership and impact the decision-makers responsible for overseeing waste management. In Canada, lack of support from hospital leadership has been cited as the number one barrier to recycling in a survey of Canadian anesthesiologists (10). Furthermore, successful programs implemented in other jurisdictions have highlighted the importance of institutional support, with educational programs and engaged leadership being crucial to reducing their footprint (1,11). In terms of legislation, Canadian provinces have jurisdiction over medical waste disposal, but few have specific regulations. In Ontario, best practice guidelines were published in 2016 for institutions involved in hazardous waste generation and disposal, to ensure compliance with the Canadian Environmental Protection Act (12). Nationally the standards outlined by the Canadian Council of Ministers of the Environment place the bulk of responsibility for implementing and updating waste management policies on individual institutions (13). Policies are enforced through self-regulation, with institutions responsible for conducting their own audits, further highlighting the importance of institutional buy-in. The process of when and how hospitals are inspected for adherence to these standards is unknown (14). The lack of transparency and accountability surrounding these policies and their enforcement makes it difficult to identify areas of inefficiency or the extent of waste mismanagement in Canada. Current waste processing methods and alternatives The lack of regulations and enforcement in medical waste disposal accrues not only a financial cost, but a cost to community health as well. Incineration is traditionally the main method of medical waste processing (13). It is the only technology that can handle all components of medical waste, reducing waste volume by 90% and weight by 75%. In the US, 49-60% of medical waste is incinerated, 20-37% is autoclaved and 4-5% is treated by other technologies (8). Medical waste contains a higher proportion of plastic and heavy metals, and as a result incineration creates toxic by-products such as polychlorinated dibenzo-p-dioxins, dibenzofurans, and mercury (8). In the US, the release of these toxins contributes to an annual burden of 470,000 disability adjusted life years (10). As such, it is extremely important that we reconsider these incinerators as the primary method of medical waste disposal. Given the drawbacks of incineration, there has been increased focus on alternative strategies including autoclaving and microwaving. Although these methods cannot change the shape of sharps or kill spores and prions, they are overall more cost-effective solutions and should be applied more broadly where possible (Table 1) (13). There are also emerging technologies such as plasma pyrolysis, which can recycle plastics and metals, and do not generate the same toxic by-products (15). There is currently a lack of infrastructure for widespread adaptation of this technique, but it offers a higher standard of safe medical waste

138

MUMJ Volume 17 No. 1, pp. 134-141

June 2020

Table 1. Streamlining of biomedical waste according to Canadian Council of Ministers of the Environment (CCME) 1992 guidelines (13). Waste Type Human Anatomical Waste

Steam Autoclaving No

Chemical Decontamination No

Animal Waste

Anatomical

Non-Anatomical Yes*

Microbiology Laboratory Waste

Yes

Regulatory Approval Required

Human Blood and Body Fluid Waste

Yes

Waste Sharps

Yes

Yes**

New Technology Plasma Pyrolysis (regulatory approval required) (15)

*Only if followed by incineration under strict control **Chemical treatment alone does not render sharps safe for additional handling. This treatment option applies to filled sharps containers that may undergo further treatment after chemical decontamination, as part of a process, e.g. chemical decontamination coupled with mechanical shredding.

disposal (16). Furthermore, with improved waste segregation at the time of disposal, such as separating syringes from needles, excess release of toxins from incineration could be averted. Of the methods mentioned, autoclaving is the most environmentally friendly and is already regularly used in university laboratories (8). With proper protocols and implementation, it is realistic for autoclaving to become more prevalent in hospitals. This would also encourage the adoption of reusable kits and supplies, thereby reducing single-use kits that generate more general waste. For example, Hamilton Health Sciences has successfully reduced their reliance on incineration, and now autoclaves more than half of their hazardous waste (7). Individual changes As a medical student or physician your direct behaviour can help to reduce the environmental impact of medical waste. Consider using oral medications instead of intravenous when possible. For procedures, try taking only the equipment needed and not excess. Consider repurposing unpackaged equipment that's gone unused for teaching. Even separating the needle from the syringe instead of disposing of both in the sharps container helps appropriately segregate waste, reducing incineration. Ultimately, we can all do at least one small thing to reduce medical waste in our clinical encounters, even if it is as simple as recycling the plastic packaging for a procedure kit instead of throwing it all into the hazardous waste bin.

139

MUMJ Volume 17 No. 1, pp. 134-141

June 2020

At the individual level, we must ensure that healthcare workers are informed about appropriate waste segregation and develop a culture of waste minimization. Education and advocacy cannot be emphasized enough. Education of staff and students is the first step to identifying points of action for improvement. Engage with hospital staff and inquire about whether the hospital has a centralized team dedicated to reducing waste. Does your hospital subcontract waste disposal or dispose of it on-site? Conclusion Improving medical waste management involves overcoming the significant challenge of ensuring patient safety while minimizing environmental impact. Our overview suggests three loci for intervention in the stream of waste management: reduced production, appropriate waste segregation and minimized incineration. In Canada where many hospitals face annual deficits, proper streamlining is a fiscally sensible solution to reduce the cost of hazardous waste disposal (1). However, Canadaâ&#x20AC;&#x2122;s medical waste management system relies on self-regulation, with little transparency. Lack of data at the provincial and federal levels about the amount of medical waste produced further limits our ability to raise awareness on the harms and identify areas for intervention. To facilitate change, there should be more enforcement and education about optimal waste management strategies at the institutional and individual level. In our days of reusable straws and StarbucksÂŽ sipping cups, itâ&#x20AC;&#x2122;s hard to justify mindlessly contributing to the mountains of medical waste without understanding the environmental, health, and societal costs.

Author biographies Betty Zhang is a second-year medical student interested in mentorship and promoting environmentally sustainable practices in medicine. She hopes to go into family medicine or anesthesia. Yu Fei Xia is a second-year medical student passionate about arts and humanities in medicine including the human impact of medical waste. She hopes to go into pediatrics or family medicine.

140

MUMJ Volume 17 No. 1, pp. 134-141

June 2020

References 1.

Kagoma Y, Stall N, Rubinstein E, Naudie D. People, planet and profits: the case for greening operating rooms. CMAJ. 2012;184(17):1905-11.

Hutchins DC, White SM. Coming round to recycling. BMJ. 2009;338:b609.

Laustsen G. Reduce–recycle–reuse: guidelines for promoting perioperative waste management. AORN journal. 2007;85(4):717-28.

World Health Organization. Health-care waste: WHO; 2018 [Available from: https://www.who.int/news-room/fact-sheets/detail/health-care-waste.

Thiel CL, Eckelman M, Guido R, Huddleston M, Landis AE, Sherman J, et al. Environmental impacts of surgical procedures: life cycle assessment of hysterectomy in the United States. Environmental science & technology. 2015;49(3):1779-86.

Environmental Protection Agency. Municipal Solid Waste | Wastes | US EPA 2016 [Available from: https://archive.epa.gov/epawaste/nonhaz/municipal/web/html/.

Hamilton Health Sciences. 2018 Sustainability Report: Hamilton Health Sciences; 2018 [Available from: https://issuu.com/hamiltonhealthsciences/docs/hhs_sustainabilityreport_2018.

Windfeld ES, Brooks MS-L. Medical waste management–A review. Journal of environmental management. 2015;163:98-108.

World Economic Forum. Germany recycles more than any other country 2017 [Available from: https://www.weforum.org/agenda/2017/12/germany-recycles-more-than-any-othercountry.

10.

Petre M-A, Bahrey L, Levine M, van Rensburg A, Crawford M, Matava C. A national survey on attitudes and barriers on recycling and environmental sustainability efforts among Canadian anesthesiologists: an opportunity for knowledge translation. Canadian Journal of Anesthesia/Journal canadien d'anesthésie. 2019;66(3):272-86.

11.

Wyssusek KH, Foong WM, Steel C, Gillespie BM. The gold in garbage: implementing a waste segregation and recycling initiative. AORN J. 2016;103(3):316. e1-. e8.

12.

Government of Ontario. C-4: The Management Of Biomedical Waste In Ontario 2016 [Available from: https://www.ontario.ca/page/c-4-management-biomedical-waste-ontario.

13.

Canadian Council of Ministers of the Environment. Guidelines for Management of Biomedical Waste in Canada 1992 [Available from: https://www.ccme.ca/files/Resources/waste/hazardous/pn_1060_e.pdf.

14.

Walkinshaw E. Medical waste-management practices vary across Canada. Can Med Assoc; 2011.

141

MUMJ Volume 17 No. 1, pp. 134-141

June 2020

15.

Nema S, Ganeshprasad K. Plasma pyrolysis of medical waste. Current science. 2002:2718.

16.

Mathur P, Patan S, Shobhawat AS. Need of biomedical waste management system in hospitals-An emerging issue-a review. Current World Environment. 2012;7(1):117.

142

Cambridge Memorial Hospital has immediate openings for:

• Pediatrician • Hospitalist • Hospitalist (with

• • • • •

General Internal Medicine Infectious Disease Respirology Psychiatry interest in Oncology) Dermatology We are a 144-bed acute care hospital undergoing an and growth in services. We are aﬃliated with the Waterloo Campus of the Michael G. DeGroote School of Medicine Ontario and the Region of Waterloo. Our specialist programs include: Cardiology, Respirology, Gastroenterology, Rheumatology, Hospitalist Medicine, General Surgery, Urology, OBS/GYN, Pediatrics, ENT, Oncology, Psychiatry. Access to other specialist programs are available within the Waterloo Wellington LHIN. Cambridge is a family-oriented community of 140,000 in the bustling, high-tech Region of Waterloo. It is conveniently located one hour from Toronto, London, and Hamilton. Our business-friendly environment and appealing lifestyle have made us one of the fastestLouise Shaw, Medical Aﬀairs Coordinator Tel: 519-621-2333, ext. 2370; Fax: 519-740-4934 Email: Lshaw@cmh.org

143

1555 Upper Ottawa Street, Hamilton, ON (905) 575-9955 info@michelangelos.com | www.michelangelos.com

144

ALARM FATIGUE? TAKE BACK CONTROL Spacelabs SafeNSound™ provides a simple, actionable reporting tool designed to help you reduce noise and increase patient safety. • Review alarm data from previous hours, shifts, days, months, or even years. • See alarm counts by department or location, and by parameter. • Access customized reports including alarm events, alarm settings, and user responses.

145

The life you’ve been working for… www.bchsys.org

Family Physicians Live in a community large enough to build a leading-edge facility with colleagues who call you by your first name. A variety of practice options that utilize EMR and function as turn-key operations are available. Situated between Hamilton and London, we are less than an hour drive from each. In fact, Toronto is only an hour away. As a Family Physician in Brant County, you’ll enjoy working in a diverse, under-serviced population of 1 ,000. Located in the picturesque Grand River Valley, our community offers an unmatched natural playground. With Laurier and Nipissing Universities and top-notch private schools, educational opportunities for you and your family abound. Call today to find out more. You’ll be surprised by all that Brantford-Brant has to offer. For a tour of our community, please call Alice Preston at 519-751-5544 ext. 2354 or e-mail a

@bchsys.org.

146

From shins to twins

Family physicians specialize in the whole person. See why family medicine is the right specialty for you. Choosefamilymedicine.ca

THE COLLEGE OF FAMILY PHYSICIANS OF CANADA

De la luxation à la conception

Médecins de famille : spécialistes de la personne dans sa globalité. Vois pourquoi cette spécialité est pour toi! Choisirlamedecinedefamille.ca

LE COLLÈGE DES MÉDECINS DE FAMILLE DU CANADA

18_0332_CFPC_Students_7_5x10_Print_Ad.indd 1

2020-02-19 2:38 PM

THE COLLEGE OF FAMILY PHYSICIANS OF CANADA

LE COLLÈGE DES MÉDECINS DE FAMILLE DU CANADA

147

McMaster University Medical Journal, Vol. 17, Issue 1

Articles inside

X SYNDROME AND RETT SYNDROME

THE NATURAL HISTORY OF MEDICAL WASTE

FLOAT OR SINK: A SOLUTION TO THE RESIDENT BURN OUT CRISIS?