African Journal of Thoracic and Critical Care Medicine - Vol 24, No 1 (2018) by HMPG

African Journal of

THORACIC AND CRITICAL CARE MEDICINE VOLUME 24

NUMBER 1

MARCH 2018

OFFICIAL JOURNAL OF THE SOUTH AFRICAN THORACIC SOCIETY ISSN 2304-0017

AFRICAN JOURNAL OF

THORACIC AND CRITICAL CARE MEDICINE VOLUME 24 | NUMBER 1 | MARCH 2018

CONTENTS EDITORIAL 2

Lung health in Africa: African solutions for African challenges – to infinity and beyond C F N Koegelenberg, K Dheda The epidemiology and risk factors of respiratory syncitial virus and its impact on timing of prophylaxis P M Jeena

Selecting the right patients is the key R van Zyl-Smit, G Calligaro

Does the source of funding matter? R van Zyl-Smit, J Grigg

AJTCCM EDITOR-IN-CHIEF Prof. K Dheda DEPUTY EDITOR Prof. C Koegelenberg SECTION EDITOR Breath-taking News: Prof. E Irusen EDITORIAL BOARD Prof. G Ainslie, Prof. E Bateman, Prof. R Green, Prof. E Irusen, Prof. M Jeebhay, Prof. P Jeena, Prof. A Linegar, Prof. R Masekela, Dr K Nyamande, Dr J O’Brien, Dr R Raine, Prof. G Richards, Dr R van Zyl Smit, Prof. M Wong, Prof. H Zar

SATS NEWS

A patient with weakness and an abnormal chest radiograph: A case report A van Straaten, J A Shaw, C F N Koegelenberg

INTERNATIONAL EDITORIAL BOARD Prof. A Cattamanchi - USA Prof. F Chung - UK Prof. G B Migliori - Italy Prof. S Sharma - India Prof. W W Yew - China Prof B Kirenga, Kampala - Uganda Prof A Katamba, Kampala -Uganda

REVIEW

PRESIDENT SA THORACIC SOCIETY Prof. K Dheda

CASE REPORT

Invasive fungal infections among critically ill children: Epidemiology, risk factors and outcomes S T Hlophe, N P Govender, R Masekela

HMPG

CEO AND PUBLISHER Hannah Kikaya Email: hannahk@hmpg.co.za

Thoracoscopy: The past, the present and the future! A personal journey I Schewitz

Inhaled corticosteroids in COPD: Personalising the therapeutic choice J A Shaw, E M Irusen

MANAGING EDITORS Naadia van der Bergh, Claudia Naidu

ORIGINAL RESEARCH

TECHNICAL EDITORS Naadia van der Bergh Kirsten Morreira

A comparison of the functional parameters of operability in patients with post-inflammatory lung disease and those with lung cancer requiring lung resection M H Amirali, E M Irusen, C F N Koegelenberg

EXECUTIVE EDITOR Bridget Farham

PRODUCTION MANAGER Emma Jane Couzens DTP AND DESIGN Clinton Griffin

The epidemiology of respiratory syncitial virus bronchiolitis: A retrospective review from Steve Biko Academic Hospital, 2013 - 2016 C X Dearden, A C Jeevarathnum, J Havinga, R J Green

CHIEF OPERATING OFFICER Diane Smith | Tel. 012 481 2069 Email: dianes@hmpg.co.za

BREATH-TAKING NEWS

JOURNAL ADVERTISING Reneé Hinze Ladine van Heerden

ABSTRACTS

ONLINE SUPPORT Gertrude Fani | Tel. 021 532 1281 Email: publishing@hmpg.co.za

The Editor African Journal of Thoracic and Critical Care Medicine

FINANCE Tshepiso Mokoena

PO Box 13725, Mowbray, 7705 Tel. 021 650 3050 | Fax. 021 650 2610 | Email. sarj@iafrica.com

HMPG BOARD OF DIRECTORS Prof. M Lukhele (Chair), Dr M R Abbas, Mrs H Kikaya, Dr M Mbokota, Dr G Wolvaardt

The views expressed in individual articles and advertising material are the personal views of the authors and are not necessarily shared by the editors, the advertisers or the publishers. No articles may be reproduced without the written consent of the publishers. The AJTCCM is published by the Health and Medical Publishing Group (Pty) Ltd. Co. registration 2004/0220 32/07, a subsidiary of SAMA. HEAD OFFICE: Block F, Castle Walk Corporate Park, Nossob Street, Erasmuskloof Ext. 3, Pretoria, 0181 EDITORIAL OFFICE: Suite 11, Lonsdale Building, Lonsdale Way, Pinelands, 7405 | Tel. 021 532 1281 All letters and articles for publication must be submitted online at www.AJTCCM.org.za Email: publishing@hmpg.co.za

PRINTED BY TANDYM PRINT

EDITORIAL

Lung health in Africa: African solutions for African challenges – to infinity and beyond Africa is a place of dichotomy. Our continent is crippled by the ‘colliding epidemics’ of tuberculosis, HIV and smoking-related diseases, but it is also a place of exceptional beauty and great abundance, with a communal spirit and a considerable amount of hope. The unique challenges of our resource-limited environment require unique and innovative solutions. Africa must, and can, provide solutions to our own health crises. Our health-related research, whether it be basic science, epidemiological, or clinical, must be relevant to Africa, and must be read, analysed and implemented in Africa. It is only through this type of capacity development that we will create a multiplier effect. Therefore, there is a niche for and an unmet need to provide a highquality pulmonary and critical-care forum on a continental level that speaks to the needs and aspirations of the African continent (and that is not currently covered by other journals, where intense competition for space has ‘crowded’ out Afrocentric views and perspectives). There are also business concerns and reviewer fatigue, such that only a few journals that are that are owned and run by academic organisations, and well supported, are likely to thrive. This year marks the 24th year of publication of the journal of the South African Thoracic Society (SATS), and the year that it transforms from the South African Respiratory Journal (SARJ) to the African Journal of Thoracic and Critical Care Medicine (AJTCCM). With the recent surge in submissions from authors based in other African countries, the journal has the potential to grow into the flagship thoracic journal of the African continent. Expansion is now assured, as leading academics from African countries other than South Africa (SA) are now being incorporated into the journal’s editorial board. Furthermore, this year also marks the first combined SATS - Pan African Thoracic Society congress to be held in Durban, SA in April 2018. The journal will take further leaps forward with its likely accreditation with PubMed in early 2019. Additionally, we are accepting international submissions, which will greatly boost our circulation. The AJTCCM aims to be the ideal platform for African basic scientists, epidemiologists and healthcare workers to disseminate

2 AJTCCM VOL. 24 NO. 1 2018

their knowledge, research and experience. Is this an SA ‘big brother’ move to dominate and take charge? Absolutely not! Instead, we see the AJTCCM as a unifying initiative, and we envision that it could serve as a platform of excellence for African pulmonary science, and a dissemination platform for many stakeholders, including those from the academic, private and government sectors. Indeed, journals like Respirology and Respiration serve as proof of concept of providing a publication platform for stakeholders from many different countries. We call on the thoracic and critical care fraternities to support our new venture by submitting manuscripts to the AJTCCM for publication. Let us join hands and make African excellence known to the world!

Coenie Koegelenberg Deputy editor African Journal of Thoracic and Critical Care Medicine coeniefn@sun.ac.za

Keertan Dheda Editor-in-chief African Journal of Thoracic and Critical Care Medicine keertan.dheda@uct.ac.za

Afr J Thoracic Crit Care Med 2018;24(1):2. DOI:10.7196/ AJTCCM.2018.v24i1.210

EDITORIAL

The epidemiology and risk factors of respiratory syncytial virus and its impact on the timing of immunoprophylaxis Of an estimated 33.8 million episodes of respiratory syncytial virus (RSV) infections occurring in <5-year-olds globally each year, 3.4 million (10%) have severe disease that requires hospitalisation, of whom an estimated 60 000 - 199 000 die, 99% of these in developing countries.[1,2] RSV is the most common cause of respiratory illness in infants worldwide, with an incidence of 30/1 000 child years, and has a mortality rate nine times that of influenza.[1,2] The National Institute of Communicable Diseases of South Africa (SA) has been involved in several programmes related to surveillance of the syndromic recognition of respiratory illnesses (SRI).[3] The pneumonia surveillance programme based at sentinel hospitals enrolled 3 746 patients, and tested 99% for RSV infection, 17% of whom tested positive. The RSV season (defined as >10% prevalence) started in week 7 - 8 of the year and continued through week 29, with the peak detection rate of 53% in week 18. The case fatality rate (CFR) for RSV was <1 % (4/572). In another programme looking at influenza-like illness (ILI) based at two primary health clinics, in 2016, 1 668 patients with ILI were enrolled, and 1 645 (99%) samples were tested for respiratory pathogens. The overall detection rate of RSV was 6% (100/1 645), and detection rose above 10% in week 7 and was sustained at ≥10% until week 17.[3] Similarly, data from a private-sector laboratory also confirmed the seasonality of RSV to be from late February to the end of June. Several factors have been implicated in the acquisition of RSV infection. These include geographic location (latitude and altitude) and climatic factors (temperature, barometric pressure, relative humidity, vapour tension, precipitation). Risk factors for serious RSV disease include prematurity (hospitalisation rate 25% - 30%, CFR 0.6% - 1.0%), congenital cyanotic heart disease (hospitalisation rate 59%, CFR 2% - 37%) and chronic lung disease (hospitalisation rate 60%, CFR 3.5% - 23%).[4] In this issue of the journal, we have an informative article from the paediatrics department at Steve Biko Academic Hospital that describes viral infections identified from nasopharyngeal aspirates over a 4-year period.[5] Of the 288 viral infections identified, RSV was seen in 162 isolates, diagnosed mainly through immunofluorescence (84%). The majority of RSV cases was seen in the <6-month-olds (63.4%), with a typical autumn-to-winter peak. Only 1.5% of the RSV cases who were tested for HIV were infected. There were eight deaths (4.9%) in the cohort, mainly in those aged >6 months, especially among those with the risk factors of prematurity or cardiac lesion. Although prematurity was identified as a risk factor for RSV, the exact gestational age where it poses the highest risk was not defined, although current international studies have shown the highest mortality rate in premature infants of 30 weeks’ gestational age.[6] Thirty percent (n=49) of cases required admission to the intensive care unit (ICU), most of whom were

<6 months of age, with an average length of ICU stay of 9.5 days. The survival rate of 95.1% is skewed compared with district and regional hospitals, as the unit serves as a tertiary referral centre. The high number of paediatric ICU (PICU) admissions for RSV cases <6 months is expected, as this is normally the age that has the highest numbers of PICU admissions. The linear increase in ICU numbers for RSV was not adjusted for confounding variables. Climatic risk factors for the acquisition of RSV could not be confirmed. These new data on RSV in SA support previous studies showing a high burden but relatively low mortality of RSV infection in infancy. They confirm previous observations of low rates of coinfection in HIVinfected infants, with a seasonality that commences in late February and lasts to the middle of June. Risk factors for acquisition and severity include prematurity <30 weeks, chronic lung disease and congenital heart disease in the first year of life. Immunoprophylaxis of these groups at the appropriate time (between mid-February and mid-June) would be the most cost-effective response, providing the best value for money. Prakash Mohan Jeena Department of Paediatrics and Child Health, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa jeena@ukzn.ac.za

1. Saso A, Kampmann B. Vaccination against respiratory syncytial virus in pregnancy: A suitable tool to combat global infant morbidity and mortality? Lancet Infect Dis 2016;16(8):e153-163. https://doi.org/10.1016/S1473-3099(16)00119-5 2. Nair H, Nokes DJ, Gessner BD, et al. Global burden of acute lower respiratory infections due to respiratory syncytial virus in young children: A systematic review and meta-analysis. Lancet 2010;375(9725):1545-1555. https://doi.org/10.1016/S01406736(10)60206-1 3. Cohen C, Walaza S, Treurnicht FK, et al. In- and out-of-hospital mortality associated with seasonal and pandemic influenza and respiratory syncytial virus in South Africa, 2009 - 2013. Clin Infect Dis 2017;66(1). https://doi.org/10.1093/cid/cix740 4. Simões EA, Anderson EJ, Wu X, Ambrose CS. Effects of chronologic age and young child exposure on respiratory syncytial virus disease among US preterm infants born at 32 to 35 weeks gestation. PLoS ONE 2016;11(11):e0166226. https://doi. org/10.1371/journal.pone.0166226 5. Dearden CX, Jeevarathnum AC, Havinga J, Green RJ. The epidemiology of respiratory syncytial virus: A retrospective review from Steve Biko Academic Hospital 2013 2016. Afr J Thoracic Crit Care Med 2018;24(1):30-35. https://doi.org10.7196/ AJTCCM.2018.v24i1.163 6. American Academy of Pediatrics Committee on Infectious Diseases; American Academy of Pediatrics Bronchiolitis Guidelines Committee. Updated guidance for palivizumab prophylaxis among infants and young children at increased risk of hospitalisation for respiratory syncytial virus infection. Pediatrics 2014;134(02):415420. https://doi.org/10.1371/journal.pone.0188223

Afr J Thoracic Crit Care Med 2017;24(1):3. DOI:10.7196/AJTCCM.2018.v24i1.202

AJTCCM VOL. 24 NO. 1 2018

EDITORIAL

Selecting the right patients is the key The pre-operative evaluation of patients undergoing resectional lung surgery is an integral part of assessing the risk of perioperative complications, and estimating the likelihood of sufficient postoperative pulmonary reserve.[1] In emergency situations, such as massive haemoptysis, a full evaluation is not possible and judgement is based on an assessment of pre-morbid effort tolerance and on the radiological and lung function parameters available at the time. For elective surgery, there is the opportunity to perform both resting and dynamic investigations to fully evaluate patients and to identify those with suspected inadequate postoperative pulmonary reserve.[2,3] In this issue of AJTCCM, Amirali et al.[4] compare forced expiratory volume in 1 second (FEV1), transfer factor (DLCO) and aerobic capacity (VO2max) in two groups of patients: those undergoing resection for lung cancer, and those with post-inflammatory lung disease. The lung cancer patients had higher average pre-operative FEV1 values (62% of predicted v. 52%), which translated (based on expected lung to be removed) into a higher predicted post-operative FEV1 (41% v. 34%). There was no difference in the DLCO values between the groups; however, the predicted postoperative VO2max was also higher in the inflammatory lung disease group. The authors suggest that the lung cancer patients would have a better pre-operative functional reserve. The clinical applicability of this descriptive report is limited, as neither postoperative mortality nor complications were recorded in the two groups. This study highlights one important aspect of evaluation of patients undergoing lung surgery; however, there are many additional aspects that need to be taken into account over and above physiology. The age of patients and co-morbid conditions are critical factors influencing postoperative recovery and risk for complication – even simple smoking status affects outcomes and wound healing.[4,5] The experience of the surgeon, the quality of the intensive care, as well as postoperative nutrition, wound care and physiotherapy all affect the outcomes, and although the predicted postoperative functional capacity may be more than adequate, any disruption to these other parameters may result in poor outcomes – not related to the predicted physiology.[6] All patients undergoing major surgery, including pulmonary resection – especially those who have borderline respiratory functional reserve, significant co-morbidities and potential to not withstand any complications – should be carefully evaluated. A functional assessment may support the decision either to opt for surgery when the physiology would suggest they will withstand resection, or to support the decision to not operate when the physiology suggests lack of reserve, despite other parameters being favourable. Physiological evaluation is the cornerstone as postoperative predictive equations will allow robust evaluations of patients who would not tolerate lung resection.[7]

4 AJTCCM VOL. 24 NO. 1 2018

In these groups of patients, a multi-disciplinary approach involving all the stakeholders, including physicians, surgeons, anaesthetists, intensivists, the patient and their family members is required. The alternatives are potentially poor, unpredicted and fatal outcomes. Given the high risks involved, patient involvement is critical in the decision-making process and adequate family counselling of what to expect, what the potential risks are and how the decision to go ahead or not was made, will ensure that all potential outcomes are foreseen and planned for. Richard van Zyl-Smit Head: Lung Clinical Research Unit, University of Cape Town Lung Institute and Division of Pulmonology, Department of Medicine, University of Cape Town and Groote Schuur Hospital, Cape Town, South Africa richard.vanzyl-smit@uct.ac.za

Greg Calligaro Division of Pulmonology, Department of Medicine, University of Cape Town and Groote Schuur Hospital, Cape Town, South Africa greg.calligaro@uct.ac.za

Afr J Thoracic Crit Care Med 2018;24(1):4. DOI:10.7196/AJTCCM2018.v24i1.204

1. Bolliger CT, Koegelenberg CFN, Kendal R. Preoperative assessment for lung cancer surgery. Curr Opin Pulm Med 2005;11(4):301-306. https://doi.org/10.1097/01. mcp.0000166588.01256.9c 2. Keagy BA, Lores ME, Starek PJ, Murray GF, Lucas CL, Wilcox BR. Elective pulmonary lobectomy: Factors associated with morbidity and operative mortality. Ann Thorac Surg 1985;40(4):349-352. https://doi.org/10.1016/s0003-4975(10)60065-3 3. Datta D, Lahiri B. Preoperative evaluation of patients undergoing lung resection surgery. Chest 2003;123(6):2096-2103. https://doi.org/10.1378/chest.123.6.2096 4. Amirali MH, Irusen EM, Koegelenberg CFN. A comparison of the functional parameters of operability in patients with post-inflammatory lung disease and those with lung cancer requiring lung resection. Afr J Thoracic Crit Care Med 2018;24(1):26-29. https://doi.org/10.7196/AJTCCM.2018.v24i1.158 5. Finlayson EV, Birkmeyer JD. Operative mortality with elective surgery in older adults. Eff Clin Pract 2001;4(4):172-177. 6. Eliasen M, Rod MH, Flensborg-Madsen T, Petersen JH, Gronbaek M, Tolstrup JS. The association between blood alcohol content and cheerfulness, focus distraction, and sluggishness among young adults attending high school parties. Alcohol Clin Exp Res 2014;38(3):826-833. https://doi.org/10.1111/acer.12268 7. Park CM, Chun HK, Lee DS, Jeon K, Suh GY, Jeong JC. Impact of a surgical intensivist on the clinical outcomes of patients admitted to a surgical intensive care unit. Ann Surg Treat Res 2014;86(6):319-324. https://doi.org/10.4174/astr.2014.86.6.319 8. Colice GL, Shafazand S, Griffin JP, Keenan R, Bolliger CT. Physiologic evaluation of the patient with lung cancer being considered for resectional surgery: ACCP evidencedbased clinical practice guidelines (2nd edition). Chest 2007;132(3 Suppl):161S-177S. https://doi.org/10.1378/chest.07-1359

EDITORIAL

Does the source of funding matter? In the competitive space of research and clinical medicine, funding is vitally important to maintain services and staff. For many years, the medical profession has battled with conflicts of interest – be they real or perceived, perverse incentives and many cases of downright unethical kickbacks. Ben Goldacres’ book Bad Pharma is an eyeopening and exhaustive account of how ‘who pays the bills’ matters. We have rightly seen a tightening of regulations regarding industry funding of events, and ‘educational talks’ in far-off luxury resorts are a thing of the past. These industry relationships have clearly established boundaries and, for publications and presentations, obligatory declared conflicts of interest allow the audience to decide on the scientific merits and potential undue influence by the funder on the researcher/clinician. The tobacco industry, however, provides a far more complex relationship to negotiate. In addition to producing commercial products that potentially kill the user, it is well recognised that the industry has spent vast sums of money on obstructing tobacco control regulations, producing favourable research outputs and actively marketing tobacco to the youth for many years. This approach is veiled as protecting their intellectual property, but effectively protects and increases their revenue. ‘Harm reduction’ is the new buzz phrase in and around tobacco smoking, with ‘heat-don’t-burn’ products being marketed along with electronic cigarettes as a ‘safer alternative to smoking’. Harm reduction is vitally important in all aspects of medical care – from diabetes to HIV – but the cynic will argue that tobacco companies also believe in harm reduction. Based on their website statements, they do, but this is not out of apparent concern for the public, and is rather to keep their clients alive for longer so that they can spend more money buying cigarettes. So, does the funding source matter? The newly established Foundation for a Smoke-free World, lead by a former World Health Organization (WHO) anti-tobacco champion, Dereck Yach, is offering independent research grants to the medical fraternity. However, the money is primarily provided by Phillip Morris International, which is a leading international tobacco company. Both parties profess to have no influence on the outcomes of the research funding or results. So, does it matter who gave the money?

Both the American Thoracic Society (ATS) and the European Respiratory Society (ERS) have previously instituted a rigid policy that no funding would be accepted from a tobacco company source, and no research funded by such a source would be presented or published in their meetings/journals. The policy also extends to organisations that are primarily funded by tobacco money, such as the Foundation for a Smoke-free World. The WHO opposes this foundation and its funding, as it contravenes the Framework Convention on Tobacco Control (FTCT) through the involvement of the tobacco industry in healthcare matters. The ATS and ERS have taken a further ethical stance – receiving money from a company that is responsible for direct harm to respiratory patients is not acceptable, as the primary mandate of the societies is the promotion of respiratory health. Similarly, the South African Thoracic Society (SATS) does not accept funding from tobacco sources and, although this new funding may be considered independent, its source is the very lives of patients with tobacco-related illness whom we care for. SATS has taken a stance that is aligned with the ERS and ATS and which may be seen by some to be short-sighted and taking a false moral high-ground. As pulmonologists, money that has been generated by causing the very illnesses our patients suffer and die from, is money we should live without. Richard van Zyl-Smit Head: Lung Clinical Research Unit University of Cape Town Lung Institute and Division of Pulmonology, Department of Medicine, University of Cape Town and Groote Schuur Hospital, Cape Town, South Africa richard.vanzyl-smit@uct.ac.za

Jonathan Grigg Professor of Paediatric, Respiratory and Environmental Medicine, Queen Mary’s Hospital London, Centre for Genomics and Child Health, Blizard Institute, Queen Mary University of London, United Kingdom j.grigg@qmul.ac.uk

Afr J Thoracic Crit Care Med 2018;24(1):5. DOI:10.7196/AJTCCM.2018.v24i1.208

AJTCCM VOL. 24 NO. 1 2018

SATS NEWS

Professor Chris Bolliger Award In mid-September 2017, Editor-in-Chief of the Karger journal Respiration Professor Felix Herth presented the Chris Bolliger Award at the European Respiratory Society Congress in Milan, Italy. Dr Daniela Gompelmann, principal investigator from the Translational Lung Research Center Heidelberg, Germany, is the first winner of this award. She has been honoured for the impact of the article ‘Pneumothorax following endobronchial valve therapy and its impact on clinical outcomes in severe emphysema’. The Chris Bolliger Award is granted to authors for outstanding scientific articles in the Karger journal Respiration for the first time. ‘Chris Bolliger was a visionary editor

of the journal Respiration as well as a great researcher and mentor. It felt like a logical step to introduce an award named after him’, explained Felix Herth. The prize will be awarded annually. It is dedicated to the late Chris Bolliger, an outstanding scientist in pulmonology and former Editor-in-Chief of Respiration. In accordance with the work and vision of Chris Bolliger, the prize will be awarded to scientists who present actual science in an evidence-based approach and have published an original article in the journal.

Reproduced with permission from Respiration.

Professor Chris Bolliger.

Research funding from the Foundation for a Smoke-free World: Position statement of the South African Thoracic Society The South African Thoracic Society (SATS) is a professional society dedicated to the promotion of lung health in South Africa (SA) and Africa through education, training, research and advocacy. Tobacco smoking results in over 7 million avoidable deaths annually and ~10% of all deaths in SA are attributed to tobacco smoking. SATS, along with the American Thoracic Society (ATS) and the European Respiratory Society (ERS), has previously established policies to not accept funding from tobacco companies. The Society views organisations/ institutions that are funded by tobacco industry money in an equivalent light. To accept funding, directly or indirectly, from a tobacco industryrelated entity contradicts our mandate to promote lung health.

This position is aligned with that of the ATS and ERS and, as such, the Foundation for a Smoke-free World, established by Phillip Morris with expected tobacco industry funding of over USD80 million, is considered to be tobacco company funded, even though it professes to be free of influence from the tobacco industry. As a Society, we will not accept research funding or other support, directly or indirectly, from any tobacco company-related source and we strongly encourage our members and colleagues to follow suit. SATS, like the ATS and ERS, will not allow publication or presentation of any research funded directly or indirectly by the tobacco industry at its meetings or in its journals.

AJTCCM VOL. 24 NO. 1 2018

CASE REPORT

A patient with weakness and an abnormal chest radiograph: A case report A van Straaten, MB ChB; J A Shaw, MB ChB, MMed (Int), FCP (SA); C F N Koegelenberg, MB ChB, MMed (Int), FCP (SA), FRCP (UK), Cert Pulm (SA), PhD Division of Pulmonology, Department of Medicine, Tygerberg Academic Hospital and Stellenbosch University, Cape Town, South Africa Corresponding author: A Van Straaten (ankevanstraaten@gmail.com)

A 40-year-old black male presented to ICU after intubation for airway protection due to rapid onset of neck weakness and swallowing difficulty. His chest radiograph showed an unusual mediastinal opacity for which a computer tomography (CT) scan was done, confirming a mediastinal mass. Afr J Thoracic Crit Care Med 2018;24(1):8-10. DOI:10.7196/AJTCCM.2018.v24i1.183

Myasthenia gravis (MG) is a disorder of the neuromuscular junction (NMJ) which results in localised or generalised fatigable weakness of skeletal muscles, commonly with ocular involvement.[1] There are paraneoplastic forms (thymoma-associated) and non–paraneoplastic forms, and the disorder is immunologically heterogeneous. [2] Approximately 10% of patients with MG have a thymoma, while up to a third of patients with a thymoma will develop MG. Pyridostigmine is the preferred symptomatic treatment, and if patients do not adequately respond to this therapy, corticosteroids, azathioprine and thymectomy are first-line immunosuppressive treatments. There are further immunomodulatory drugs emerging, but they are limited by the scarcity of controlled studies. Long-term drug treatment is essential for most patients and must be tailored to the particular form of MG.[3] Thymectomy is indicated as first-line therapy in all patients with thymoma or suspected thymoma, regardless of the status of MG.

Case report

A 40-year-old black male, who was employed as a long-distance truck driver, presented with a history of rapid onset of neck weakness, swallowing difficulty and ptosis. The treating physicians intubated the patient for airway protection, initiated antibiotic treatment for a suspicion of aspiration pneumonia and transferred him to an intensive care unit (ICU). On arrival in the ICU the patient had a temperature of 38.6 °C without an obvious source of infection. His other vital signs were within normal limits. He was noted to have bilateral ptosis and a bulbar palsy. Power in all limbs was normal. The rest of the examination was unremarkable. A chest radiograph (Fig. 1) was performed on his arrival at the ICU. During his admission to the ICU, a computed tomography (CT) scan of the patient’s chest was performed to explore a radiological abnormality, which revealed a well-circumscribed lowanterior mediastinal mass (Fig. 2). Anti-acetylcholine receptor (AChR) antibody testing was positive, and nerve conduction studies demonstrated a decremental pattern in keeping with a diagnosis of myasthenia gravis. The patient was treated with plasmapheresis, corticosteroids and neostigmine. He was

8 AJTCCM VOL. 24 NO. 1 2018

Fig. 1. Chest radiograph of patient on arrival in ICU. successfully extubated after 7 days. Subsequent surgical excision of the mediastinal mass confirmed a type B2 thymoma.

Discussion

Myasthenia gravis and the pathogenesis of autoimmunity

Myasthenia gravis (MG) is a disorder of the neuromuscular junction (NMJ) which results in localised or generalised fatigable weakness of skeletal muscles, commonly with ocular involvement.[1] While congenital MG results from gene mutations affecting the NMJ components, most patients who develop MG in adulthood have autoantibodies specific to the postsynaptic AChR or functionally related molecules.[4] These autoantibodies are thought to originate in hyperplastic germinal centres in the thymus where myoid cells expressing AChR are clustered. Autoantibodies are present in the serum in 80% to 90% of cases, most commonly anti-AChR antibodies.[5] However, there is significant immunological heterogeneity, with variation in antibody structure

CASE REPORT Thymectomy in MG

Fig. 2. A computed tomography scan of the patientâ&#x20AC;&#x2122;s chest showing an anterior mediastinal mass. between individuals as well as between different muscles within a single individual. Thymic abnormalities are present in the majority of AChR antibody positive cases, with 60% - 70% of cases having thymic hyperplasia and 10% - 12% having a thymoma. The thymus contains a small number of myoid cells that are the only known cells to express intact AChR outside of muscle.[6] Thymic epithelial cells produce unfolded AChR subunits that are believed to prime helper T cells to autoimmunity. These T cells then attack the AChR on the myoid cells, creating infiltrating germinal centres in the hyperplastic thymus and triggering complement activation and deposition. The autoimmunisation is completed as the antibodies in the germinal centres diversify to recognize intact muscle AChR.

Thymomas and MG

While only approximately 10% of patients with MG will have a thymoma, it is known that up to a third of patients with a thymoma will develop MG.[7] The role of thymoma in autoimmunity is not clear, although it is thought that the histological subtype of thymoma may be important. The development of MG is associated with mixed thymomas, but not with thymomas of the cortical type. Certain antibodies are more commonly associated with the presence of a thymoma.[2] In addition to AChR antibodies, some individuals have muscle autoantibodies directed against titin or the ryanodine receptor as well as other intracellular muscle proteins. Among patients with MG, the presence of anti-titin antibodies is predictive of a thymic epithelial tumour (sensitivity 69% - 80%, specificity 90% - 100%). Anti-low-density lipoprotein receptor-related protein 4 (LRP4) and anti-muscle-specific kinase (MuSK) antibodies are not associated with thymomas. Thymomas are also associated with an increased risk of developing other autoimmune diseases such as thyroiditis, rheumatoid arthritis and systemic lupus erythematosus.[8]

Thymectomy is indicated as first-line therapy in all patients with thymoma or suspected thymoma, regardless of the status of MG.[7] If complete resection is not feasible, then a biopsy is required prior to neoadjuvant chemotherapy or radiotherapy, to improve likelihood of surgical resection. Patients with MG and a thymoma generally show significant improvement in disease after thymectomy. Moreover, there is evidence for the benefit of thymectomy over drug therapy alone in patients with generalised MG and AChR antibodies, even in the absence of a thymoma.[9] The thymectomy trial for non-thymomatous myasthenia gravis patients receiving prednisone (MGTX), was a multicentre, assessor-blinded trial run between 2006 and 2012, which enrolled 126 subjects with generalised AChR antibody-associated MG.[7] It demonstrated significantly lower severity of weakness, lower prednisone and immunosuppressive agent requirements, reduced need for hospitalisation for MG exacerbations and a greater proportion of subjects with minimal manifestations at 12 months in the thymectomy group compared with the prednisone-alone group. Evidence also favours thymectomy in early-onset disease rather than in late-onset disease, as the latter group of patients often have thymic atrophy and derive no benefit from the procedure.[10] A thymectomy should also be considered in children with MG. Certain populations should not undergo thymectomy based on current evidence. Patients with MG and anti-MuSK or LRP4 antibodies should not be offered thymectomy. In patients with pure ocular MG, there is insufficient evidence that surgery prevents generalisation of results in remission; however, it has been argued that thymectomy should be considered in patients with ocular MG when drug treatment has failed if the patients have AChR antibodies and a risk of generalised disease.

MG in South Africa

A retrospective observational study published in 2007 demonstrated that the annual incidence of anti-AChR-positive MG in the Cape Town metropole of South Africa is similar to that of developed countries, without significant differences in the incidence rates among the three predominant racial groups.[11] However, a difference was noted in the clinical phenotype between the racial groups in that black patients were more likely to develop treatment-resistant complete ophthalmoplegia and ptosis than white patients (18% v. 2%; p=0.041). Despite similar-sized cohorts, white patients were more likely to develop generalised myasthenia poorly responsive to therapy (p=0.005) than black patients. There were no significant racial differences in the time between diagnosis to initiation of therapy, or the performance and timing of thymectomy. Acknowledgements. None. Author contributions. All authors contributed equally to the preparation and revision of the manuscript. Funding. None. Conflicts of interest. EMI has received lecture fees from AstraZeneca, Novartis, MSD, Boehringer Ingelheim and Aspen and is currently employed part-time by GlaxoSmithKline. JAS has previously received a travel scholarship from GlaxoSmithKline which was awarded via Medical Practice Consulting.

AJTCCM VOL. 24 NO. 1 2018

CASE REPORT 1. Gilhus NE. Myasthenia gravis. N Engl J Med 2016;375:2570-2581. https://doi. org/10.1056/nejmra1602678 2. Carr AS, Cardwell CR, McCarron PO, McConville J. A systematic review of population based epidemiological studies in myasthenia gravis. BMC Neurol 2010;10:46-54. https:doi.org/10.1186/1471-2377-10-46 3. Gilhus NE, Verschuuren JJ. Myasthenia gravis: Subgroup classification and therapeutic strategies. Lancet Neurol 2015;14(10):1023-1036. https://doi.org/10.1016/s14744422(15)00145-3 4. Querol L, Illa I. Myasthenia gravis and the neuromuscular junction. Curr Opin Neurol 2013;26(5):459-465. https:// doi: 10.1097/WCO.0b013e328364c079 5. Huijbers MG, Lipka AF, Plomp JJ, Niks EH, van der Maarel SM, Verschuuren JJ. Pathogenic immune mechanisms at the neuromuscular synapse: The role of specific antibody-binding epitopes in myasthenia gravis. J Intern Med 2014;275(1):12-26. https://doi/10.1111/joim.12163 6. Romi F, Skeie GO, Gilhus NE, Aarli JA. Striational antibodies in myasthenia gravis: Reactivity and possible clinical significance. Arch Neurol 2005;62(3):442-446. https:// doi.org/10.1001/archneur.62.3.442

10 AJTCCM VOL. 24 NO. 1 2018

7. Wolfe GI, Kaminski HJ, Aban IB. Randomized trial of thymectomy in myasthenia gravis. N Engl J Med 2016;375(6):511-522. https://doi.org/10.1056/nejmoa1602489 8. Nacu A, Andersen JB, Lisnic V, Owe JF, Gilhus NE. Complicating autoimmune diseases in myasthenia gravis: A review. Autoimmunity 2015;48(6):362-368. https:// doi.org/10.3109/08916934.2015.1030614 9. Cea G, Benatar M, Verdugo RJ, Salinas RA. Thymectomy for non-thymomatous myasthenia gravis. Cochrane Database Syst Rev 2013;10:CD008111. https://doi. org/10.1002/14651858.cd008111.pub2 10. Newsom-Davis J, Cutter G, Wolfe GI, et al. Status of the thymectomy trial for nonthymomatous myasthenia gravis patients receiving prednisone. Ann N Y Acad Sci 2008;1132:344-347. https://doi.org/10.1196/annals.1405.014 11. Heckmann J, Owen EP, Little F. Myasthenia gravis in South Africans: Racial differences in clinical manifestations. Neuromuscul Disord 2007;17(11-12):929-934. https://doi. org/10.1016/j.nmd.2007.07.002

Accepted 10 October 2017.

REVIEW

Invasive fungal infections among critically ill children: Epidemiology, risk factors and outcomes S T Hlophe,1 MB ChB, MMed (Paeds), DCH (SA), FC Paed (SA); N P Govender,2 MB BCh, MMed, FC Path; R Masekela,1 MBBCh, MMed (Paeds) Pret, Dip Allergy (SA), Cert Pulm (SA) Paeds, FCCP, PhD Department of Paediatrics and Child Health, Nelson R Mandela School of Clinical Medicine, University of KwaZulu-Natal, Durban, South Africa Centre for Healthcare-associated Infections, Antimicrobial Resistance and Mycoses, National Institute for Communicable Diseases, National Health Laboratory Service; and School of Pathology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

1 2

Corresponding author: S T Hlophe (sbehlophe@gmail.com)

Critically ill children are at high risk of developing invasive fungal infection in a paediatric intensive care unit. This is due to the vulnerability of these children and invasive nature of the care provided. Afr J Thoracic Crit Care Med 2018;24(1):11-14. DOI:10.7196/AJTCCM.2018.v24i1.172

Paediatric intensive care units (PICUs) are often multidisciplinary, admitting both medical and surgical patients. The innate vulnerability of critically ill children and the invasive nature of care provided in a PICU make this a high-risk setting for healthcare-associated infections (HAIs). HAIs are an important cause of morbidity and mortality in this population, yet are potentially preventable. Dramowski et al.[1] reported an HAI prevalence of 16.5% at Tygerberg Children’s Hospital, Cape Town, during a 6-month period in 2015. Similarly, Spicer et al.[2] reported a prevalence of 20.4% and an incidence of 15.3 cases per 100 PICU admissions over a 2-year period at Grey’s Hospital, Pietermaritzburg. The overall mortality attributed to paediatric HAIs has been estimated at 11%.[3] Invasive fungal infections are among the top four causes of paediatric HAIs.[4] This review is aimed at describing the epidemiology, risk factors and mortality associated with invasive fungal infections among critically ill children admitted to PICUs.

Organisms

Fungi are ubiquitous organisms that live as environmental saprophytes or as commensal microorganisms of humans and animals.[5,6] Medically important fungi are commonly opportunists and rarely primary pathogens in exposed immune-competent subjects. However, when encountered in the context of a PICU they can cause opportunistic infections. Species in the genetically diverse genus Candida commonly cause invasive disease among critically ill children; Aspergillus spp., mucocutaneous moulds and other rarer emerging fungi occasionally cause disease.[5,6] Invasive infections caused by Candida spp. and Aspergillus spp. are associated with high mortality and morbidity as well as high healthcare costs. In this review, we therefore focus only on invasive disease caused by these two pathogens.

Candidaemia and invasive candidiasis

Candida spp. are the leading cause of invasive fungal infections in hospitalised children and are the third most common isolates recovered from paediatric cases of healthcare-associated bloodstream infection

in the United States.[7,4] Spicer et al.[2] found that 23.8% of all the HAIs at Grey’s Hospital were bloodstream infections. Dramowski et al.[1] reported that 6% of HAIs were caused by Candida spp. In children, candidaemia is associated with prolonged hospital stay (median 21 days) and increased costs.[4] Candida albicans is the most common invasive species in the paediatric population, causing 55% of cases.[4] Candida parapsilosis and Candida tropicalis are other common species, which contribute to 17.5% and 10% of the burden of invasive disease, respectively.[4] Candida glabrata and Candida krusei are less frequently cultured but may be encountered in specialist units.[8] Distinguishing Candida colonisation from infection is not always straightforward. Invasive fungal infection is defined as a positive culture from either blood or sterile sites, together with clinical or laboratory evidence of a systemic inflammatory host response. In contrast, colonisation typically occurs in the absence of such a response in cultures from non-sterile sites such as the respiratory tract.[9,10] Candida isolation from respiratory secretions alone should never prompt treatment.[11] Invasive candidiasis is distinguished from candidaemia by clinical, radiological, microbiological or histological evidence of disseminated foci of infection, e.g. splenic/hepatic abscesses or endophthalmitis.[11] Invasive Candida infection has a reported attributable mortality in children of between 20% and 30%.[4] In most studies, Candida is one of the predominant causative agents for sepsis in hospitalised children, together with coagulase-negative staphylococci, enterococci and Staphylococcus aureus. [4] Admission to the PICU is a risk factor for invasive Candida infection.[5] In another Tygerberg Hospital study, the pathogens associated with the highest mortality from bloodstream infections were Acinetobacter spp. (38%; n=30/78), followed by Candida spp. (31%; n=20/65) and Escherichia coli (24%; n=23/97). [12] In the same study, all 21 C. albicans isolates were susceptible to fluconazole, whereas 22 of the 44 isolates other than C. albicans were resistant to fluconazole.[12] Among all risk groups, Candida colonisation is an independent risk factor for infection and precedes invasive infection in most cases.[9]

AJTCCM VOL. 24 NO. 1 2018

REVIEW The risk of infection increases with the number of colonised sites and is dependent on the colonising species. Candida colonisation in the gastrointestinal tract has been reported as a significant risk factor for invasive candidiasis in many studies.[13] Other reported risk factors include:[4,12] • the presence of a central venous catheter • total parenteral nutrition • a pre-existing bacterial infection • immunocompromised status of the host • recent surgery • dialysis • prolonged use of vancomycin • administration of antimicrobial agents against Gram-negative bacteria • mechanical ventilation. Various scoring systems have been developed to aid the differentiation between infection and colonisation and these systems aid in identifying patients at risk of developing infection. The system used for the Candida score and index, developed for critically ill adults, allocates points for each of the following criteria:[4,14,15] • total parenteral nutrition = 1 point • surgery on ICU admission = 1 point • multifocal Candida colonisation = 1 point • severe sepsis = 2 points. The Candida score is a useful tool to differentiate critically ill patients, who would benefit from early antifungal treatment (score >3), from those in whom invasive candidiasis is highly improbable (score ≤3). [15] In a large cohort of non-neutropenic, critically ill patients in whom Candida colonisation was prospectively assessed, those with a Candida score >3 were accurately selected to benefit from early antifungal treatment.[13] The Candida colonisation index is the ratio of the number of distinct non-sterile body sites colonised with Candida spp. to the number of body sites from which specimens were cultured.[4] An index >0.5 has been shown to have a predictive value of 66% in determining infection. The sensitivity can be further increased if a semi-quantitative fungal load is simultaneously determined. An increasing fungal load in successive specimens is also predictive of invasive infection.[5] Invasive infection by different Candida spp. may result in variable outcomes, which makes identification to species level essential.[12] In a small case series of 19 patients admitted with fungal bloodstream infections to the PICU at Inkosi Albert Luthuli Central Hospital, Durban, C. albicans was identified as the most common organism, followed by C. parapsilosis (Table 1). Of these 19 isolates, 12 were sensitive to fluconazole (63.2%). All patients were on broad-spectrum antibiotics for at least 7 days. Invasive fungal infection was confirmed 48 hours post ICU admission in all but 3 cases (15.8%).

Aspergillosis

Aspergillus is the most commonly isolated invasive mould, although there are no epidemiological data for this organism in South Africa (SA). Arendrup et al.[9] suggested that the incidence of invasive aspergillosis in children was increasing, similar to trends observed in adults; aspergillosis results in a case fatality rate of more than

12 AJTCCM VOL. 24 NO. 1 2018

Table 1. Prevalence and in-hospital case fatality associated with Candida spp. bloodstream infections in the paediatric intensive care unit, Inkosi Albert Luthuli Central Hospital, 2015 and 2016 (N=19)* Prevalence, In-hospital case Organism n (%) fatality, n/N (%) Candida albicans 7 (36.8) 2/7 (28.6) Candida parapsilosis 6 (31.6) 1/6 (16.7) Candida tropicalis 2 (10.5) 1/2 (50.0) Candida sake 1 (5.3) 1/1 (100) Other Candida spp. 3 (15.8) 2/3 (66.7) (unspecified) *Unpublished data.

50%. There are conflicting results in the literature regarding the species distribution of Aspergillus among paediatric cases of invasive aspergillosis. In both children and adults, Aspergillus fumigatus was the most frequently isolated species, followed by Aspergillus flavus.[4,9] Most children with invasive aspergillosis present with pulmonary aspergillosis but dissemination to other sites is also seen, particularly to the central nervous system. The aspergillosis clinical syndrome depends on the host’s immune status, ranging from invasive aspergillosis to tracheobronchitis, aspergilloma and chronic necrotising aspergillosis; colonisation without infection also occurs.[4] Several underlying diseases and their treatments are risk factors for invasive Aspergillus infection, including haematological malignancies (primary or relapse), allogeneic bone marrow transplantation, granulocytopenia, systemic corticosteroids, immunosuppressive therapies and immunodeficiencies, such as seen in chronic granulomatous disease, severe combined immunodeficiency and organ transplantation (e.g. heart-lung transplantation). [4] The incidence of invasive Aspergillus infection varies according to the underlying disease and is highest in immunocompromised children with either acute myeloid leukaemia (5.35%) or acute lymphoblastic leukaemia (1.5%). [8,16] The numbers of patients with chronic obstructive pulmonary disease, influenza or decompensated cirrhosis are increasing and are understudied populations at risk for invasive pulmonary aspergillosis in the ICU.[17] There are limited data for aspergillosis in PICU patients.[17] The clinical presentation of invasive aspergillosis and the rate at which the disease progresses vary.[16] As immunosuppression increases, so does the rate of disease progression.[16] Paediatric patients with invasive Aspergillus infection have a 20% higher mortality risk and a 13.5-fold increase in relative risk for death compared with children without invasive Aspergillus infection.[4] In one study, the case fatality rate was 53% and multivariable analysis showed that allogeneic haematopoietic stem cell transplantation was a predictor of poor prognosis.[8] The mortality rate from disseminated aspergillosis is very high (up to 80% in those affected). The mortality rate in those with central nervous system involvement, bone marrow transplantation and advanced HIV infection is 88%, 87% and 86%, respectively.[4]

Diagnostic tests for invasive fungal infections

There are limited data on the value of non-culture diagnostic tests in children. Blood cultures are essential diagnostic tests for candidaemia

REVIEW but are not useful for aspergillosis.[10] The galactomannan test has low specificity and sensitivity and is not recommended for diagnosis of invasive candidiasis.[10] The 1,3-β-D-glucan (BDG) test can be used to exclude invasive fungal infections, including candidiasis, with a sensitivity and specificity of 65% and 80%, respectively.[10] The test is not specific for Candida spp. because the antigen is present in many fungal species. The BDG and galactomannan tests are associated with high false-positive rates. The levels of BDG were found to be higher in subjects receiving human blood products, antibiotics and corticosteroid therapy than in those without these treatments.[18] Invasive aspergillosis may be asymptomatic in up to one-third of patients, and diagnostic difficulties are compounded by the lack of characteristic symptoms and accurate diagnostic tests.[16]

Role of prophylaxis

The impact of antifungal therapy on the outcome of invasive fungal infections is affected by the appropriateness and timing of initiation of treatment. Prophylactic antifungal agents are recommended in only a few specific situations and most treatments are administered on an empiric basis.[19] For extremely premature neonates, the use of fluconazole prophylaxis is an attractive option for reducing invasive candidiasis.[9] This is not routinely recommended in the South African setting because of the reported resistance of Candida in paediatric units. Govender et al.[20] reported that more than half of the C. parapsilosis isolates from bloodstream infections in 2009 - 2010 tested resistant to fluconazole. The guidelines of the European Society for Clinical Microbiology and Infectious Diseases no longer recommend fluconazole for treatment of invasive candidiasis and now endorse the use of echinocandins as first-line empiric therapy.[19,20] Reinforcement of the intestinal mucosal barrier by administration of commensal bacteria (probiotics) as supplements may be useful for prevention of nosocomial fungal infections.[12] Probiotics modify the enteric microflora by colonising the gastrointestinal tract and reduce overgrowth of pathogens that could otherwise lead to colonisation and invasive infection.[12] Some trials have reported a beneficial effect of probiotics in the prevention of enteric colonisation by Candida spp. in preterm newborns, but no such trials have been conducted in critically ill paediatric patients.[12] Monitoring for colonisation with Candida spp. in children undergoing treatment for severe sepsis or septic shock in the PICU for longer than 5 days may offer an opportunity for early intervention to prevent candidaemia.[21] Singhi et al.[21] demonstrated that 90% of the patients who developed candidaemia were colonised by the same Candida species. Shorter courses of antibiotic therapy and routine surveillance cultures for Candida spp. are recommended. The routine use of antifungal prophylaxis in the general ICU setting is discouraged.[22] The principal negative aspect of prophylaxis is selection of resistant strains and antifungal agent-related toxicities. This problem can be minimised by having better diagnostic tools for invasive fungal infection. The value of prophylaxis against invasive aspergillosis in the intensive care setting remains uncertain.[16]

Future research priorities

More data are required on predisposing factors for fungal infections in critically ill children and the effect of prophylactic antifungal therapy,

together with an assessment of the impact on morbidity and mortality. More research is required on prediction rules and diagnostic tests to help with early identification and adequate prophylactic therapy or preemptive therapy. Well-controlled prospective trials are required to assess the changing microbial milieu in PICUs and the impact of antibiotic stewardship on reducing fungal infection in the PICU.

Conclusion

Candida and Aspergillus spp. are the most frequently identified fungi in critically ill children, although there are no data on aspergillosis in the SA context. The attributable mortality of these two invasive infections differs mainly because of heterogeneity in the patient populations. The impact of antifungal therapy is affected by the appropriateness and timing of initiation. It is important to identify patients at risk of developing fungal infection early and to draft policies regarding empirical antifungal therapy. There is need for more data to address the role of antifungal chemoprophylaxis in the PICU setting. Acknowledgements. The authors would like to thank Dr Y Mahabeer (UKZN, microbiologist) for providing microbiology results. Author contributions. All 3 authors have contributed towards drafting and the critical revision and approval of the final version. Funding. None. Conflicts of interest. None. 1. Dramowski A, Cotton MF, Whitelaw A. Surveillance of healthcare-associated infection in hospitalised South African children: Which method performs best? S Afr Med J 2017;107(1):56-63. https://doi.org/10.7196/samj.2016.v107.i1.11431 2. Spicer KB, Green J, Dhada B. Hospital-acquired infections in paediatric medical wards at a tertiary hospital in KwaZulu-Natal, South Africa. Paediatr Int Child Health 2017;38(1):53-59. https://doi.org/10.1080/20469047.2017.1299897 3. Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosocomial infections in pediatric intensive care units in the United States. National Nosocomial Infections Surveillance System. Crit Care Med 1999;103(4):e39. https://doi.org/10.1542/peds.103.4.e39 4. Brissaud O, Guichoux J, Harambat J, Tandonnet O, Zaoutis T. Invasive fungal disease in PICU: Epidemiology and risk factors. Ann Intensive Care 2012;2(1):6. https://doi. org/10.1186/2110-5820-2-6 5. De Pauw BE, Patterson TF. Should the consensus guidelines’ specific criteria for the diagnosis of invasive fungal infection be changed? Clin Infect Dis 2005;41(Suppl 6):S377-380. https://doi.org/10.1086/430919 6. Ambasta A, Carson J, Church DL. The use of biomarkers and molecular methods for the earlier diagnosis of invasive aspergillosis in immunocompromised patients. Med Mycol 2015;53(6):531-557. https://doi.org/10.1093/mmy/myv026 7. Zaoutis TE, Prasad PA, Localio AR, et al. Risk factors and predictors for candidemia in pediatric intensive care unit patients: Implications for prevention. Clin Infect Dis 2010;51(5):e38-e45. https://doi.org/10.1086/655698 8. Dornbusch HJ, Manzoni P, Roilides E, Walsh TJ, Groll AH. Invasive fungal infections in children. Pediatr Infect Dis J 2009;28(8):734-737. https://doi.org/10.1097/ INF.0b013e3181b076b1 9. Arendrup M, Fisher B, Zaoutis T. Invasive fungal infections in the paediatric and neonatal population: Diagnostics and management issues. Clin Microbiol Infec 2009;15(7):613-624. https://doi.org/10.1111/j.1469-0691.2009.02909.x 10. Pappas PG, Barnes RA, Warnock, DW. Fungal infection in the intensive care unit. Mycopathologia 2004;157(1):137-138. https://doi.org/10.1023/ B:MYCO.0000012323.33545.0a 11. Cuenca‐Estrella M, Verweij P, Arendrup M, et al. ESCMID guideline for the diagnosis and management of Candida diseases 2012: Diagnostic procedures. Clin Microbiol Infec 2012;18(Suppl 7):9-18. https://doi.org/10.1111/1469-0691.12038 12. Dramowski A, Cotton MF, Rabie H, Whitelaw A. Trends in paediatric bloodstream infections at a South African referral hospital. BMC Pediatr 2015;15:33. https://doi. org/10.1186/s12887-015-0354-3 13. Kumar S, Singhi S, Chakrabarti A, Bansal A, Jayashree M. Probiotic use and prevalence of candidemia and candiduria in a PICU. Pediatr Crit Care Med 2013;14(9):e409-e415. https://doi.org/10.1097/pcc.0b013e31829f5d88

AJTCCM VOL. 24 NO. 1 2018

REVIEW 14. León C, Ruiz-Santana S, Saavedra P, et al. A bedside scoring system (“Candida score”) for early antifungal treatment in non-neutropenic critically ill patients with Candida colonization. Crit Care Med 2006;34(3):730-737. https://doi.org/10.1097/01. ccm.0000202208.37364.7d 15. Leroy G, Lamboitte F, Thevenin D, et al. Evaluation of “Candida score” in critically ill patients: A prospective, multicenter, observational, cohort study. Ann Intensive Care 2011;1(1):50. https://doi.org/10.1186/2110-5820-1-50 16. Enoch D, Ludlam H, Brown N. Invasive fungal infections: A review of epidemiology and management options. J Med Microbiol 2006;55(Pt 7):809-818. https://doi. org/10.1099/jmm.0.46548-0 17. Colombo AL, de Almeida Júnior JN, Slavin MA, Chen SC, Sorrell TC. Candida and invasive mould diseases in non-neutropenic critically ill patients and patients with haematological cancer. Lancet Infect Dis 2017;17(11):e344-e356. https://doi. org/10.1016/s1473-3099(17)30304-3 18. Zheng F, Zha H, Yang D, Deng J, Zhang Z. Diagnostic values and limitations of (1, 3)-β-D-glucans and galactomannan assays for invasive fungal infection in patients admitted to pediatric intensive care unit. Mycopathologia 2017;182(3-4):331-338. https://doi.org/10.1007/s11046-016-0063-y

14 AJTCCM VOL. 24 NO. 1 2018

19. Calandra T, Roberts JA, Antonelli M, Bassetti M, Vincent J-L. Diagnosis and management of invasive candidiasis in the ICU: An updated approach to an old enemy. Crit Care 2016;20(1):125. https://doi.org/10.1186/s13054-016-1313-6 20. Govender NP, Patel J, Magobo RE, et al. Emergence of azole-resistant Candida parapsilosis causing bloodstream infection: Results from laboratory-based sentinel surveillance in South Africa. J Antimicrob Chemother 2016;71(7):1994-2004. https:// doi.org/10.1093/jac/dkw091 21. Singhi S, Rao DSR, Chakrabarti A. Candida colonization and candidemia in a pediatric intensive care unit. Pediatr Crit Care Med 2008;9(1):91-95. https://doi. org/10.1097/01.pcc.0000298643.48547.83 22. Weinstein RA, Rex JH, Sobel JD. Prophylactic antifungal therapy in the intensive care unit. Clin Infect Dis 2001;32(8):1191-1200. https://doi.org/10.1086/319763

Accepted 26 September 2017.

REVIEW

Thoracoscopy: The past, the present and the future! A personal journey I Schewitz, MB ChB, FCS (Cardiothoracic Surg) Department of Cardiothoracic Surgery, Faculty of Health Sciences, University of Pretoria, South Africa Corresponding author: I Schewitz (ivan@schewitz.com)

Thoracoscopy, in my opinion, is underutilised in Africa, for a multiplicity of reasons. These include a lack of expertise, the perceived cost and difficulties in obtaining and maintaining equipment. The benefits, however, in improved surgery and decreased surgical pain and rapid return to productive work outweigh by far the so-called disadvantages. In my opinion, thorascopic techniques should be routine in all our academic departments. Our newly qualified thoracic surgeons should be trained in video-assisted thoracic surgery. Afr J Thoracic Crit Care Med 2018;24(1):15-18. DOI:10.7196/AJTCCM.2018.v24i1.182

Thoracoscopy is over 100 years old. Jacobaeus[1-3] described its use in 1906. He used it mainly for diagnosing tuberculosis pleurisy. With the rise of chemotherapy for tuberculosis, however, it gradually fell into disuse. During the 1970s, gynaecologists started using laparoscopy, as did general surgeons in the late 1980s. The development of fibreoptics, light transmission and image processing has not only made video-assisted surgery more efficient, but has also increased the scope of what is possible. This article is a review of the last 25 years in one South African (SA) practice, and a prediction for the future. In 1991, Schutte, in the small hospital of Sunward Park, Boksburg, SA, introduced laparoscopic cholecystectomy to the country. He was one of the first in the country to practise the new techniques, which he had learnt in his studies overseas. The advantages were soon obvious in the postoperative course of the patients, who were discharged in 24 - 48 hours. At that stage, in 1991, thorascopic surgery was in its infancy, but I had the opportunity of a very helpful colleague (Schutte), an encouraging hospital management who assisted with the equipment, and patients who at all times were fully informed of the procedure and were aware that the accepted open procedure might be used if necessary. A trip visiting prominent units in the USA followed, which stimulated me to continue what I was doing. The 1990s era was a time when we were taught: • Keyhole surgery is bad surgery. • Real surgeons make big cuts. • You must be able to see what you are doing. Surgery still says: You must see what you are doing. The better the vision, the better the surgery. Modern thorascopic equipment gives by far better visualisation than even the best loops and optics of the past. It also allows one to see around corners that you could not with open surgery. The instruments are constantly evolving, allowing the most complex procedures to be performed more safely than ever before.

1991: The beginning

The beginning was the learning stage. The equipment was one-chip technology and the screens were small and of inferior quality, but

the experience was exhilarating. I was just one surgeon learning as I progressed. My trips to various units in the USA and Europe showed me that surgeons were in the learning phase everywhere. I visited the Royal Brompton Chest in London, Stuttgart Clinic in Germany, the Memorial Sloan Kettering Cancer Center, Philadelphia University, Methodist Dallas Medical Center and the Mayo Clinic in the USA, and since have visited many others, always learning something new. The experience convinced me that the future of surgery was minimally invasive. Bob Ginsberg at Sloan Kettering at the time said to me, ‘I do not believe in this new technology but will allow my junior staff to investigate it.’ Goldstraw at the Brompton had similar sentiments. This taught me the importance of a university investigating new procedures, but with, of course, checks and balances. These two men were real leaders who did not stop innovation but rather encouraged their juniors. History teaches us of the great scientists of the past who have been stifled because of the bureaucracy of the time. We are in the process of great change, in not only medicine but also many areas of the modern world. At our peril, we remain mired in the past. Continuing medical education is vital to remain current in our knowledge and in our practice of medicine. During the 1990s, many thorascopic courses were held internationally, and I had the privilege of attending a number of these. At that time, the standard approach was to use three ports, working in a large triangle, with the camera pointing towards the lesion, and with two instruments, one in each hand. The camera pointed away from the surgeon. As more and more academic units adopted the use of thorascopy, the courses were gradually discontinued as residents and registrars were exposed to the techniques in their routine training. My first cases were simple lung biopsies. These were followed by a mediastinal gland biopsy. Soon I had a number of recurrent spontaneous pneumothoraxes for which I performed a bullous resection and an apical pleurodesis. One of my very early patients was a businessman with enlarged mediastinal glands, who felt that he could not take time off work. He waited for me to return from one of

AJTCCM VOL. 24 NO. 1 2018

REVIEW my overseas trips, was admitted on the day of the procedure, had a biopsy at 11h00 h, was discharged at 18h00 and returned to work the following day. The surgery was performed on the Wednesday, and on the Saturday he played a round of golf and reported that he was two below his handicap. The difference between open surgery and minimally invasive thoracic surgery was like night and day. One patient had a lymphoma diagnosed without the trauma of a major operative procedure. An interesting case was the endoscopic division of an anomalous venous drainage in which the right subclavian artery arose distally from the aorta as the third branch winding behind the oesophagus. The vessel was ligated endoscopically, and attached in the neck to the common carotid through a small neck incision. This adult patient had presented with dysphagia that was totally relieved by the surgery. At the time, unfortunately, there was significant resistance to thoracoscopy from a number of my colleagues, who felt that the dangers outweighed its advantages. The medical aid schemes complained about the cost. I was questioned about safety and about any advantages. It was pointed out that even a simple chest drain can cause a post-thoracotomy syndrome. My international trips, however, had convinced me of the benefits, and I continued the work. I soon had experience encompassing over 1 500 cases. Thorascopic surgery is still in its very early stages in most SA academic units. The last few years, however, have seen the increasing introduction of video-assisted thoascopic surgery (VATS) techniques into our university departments. Very little thorascopic surgery, however, is being performed in sub-Saharan Africa outside SA. From my many visits and via personal communication, I can state that no thorascopic surgery is being performed in Nigeria, Kenya, Malawi, Tanzania, Zambia, Zimbabwe or Botswana.

2000: Maturing

By 2000, the terminology had changed. The term VATS had been introduced. The thoracoscope had become an adjunct to open thoracic surgery. It provides an eye and a light source inside the chest. As the surgeon becomes more skilled, the incisions become smaller and smaller. The massive aggressive incisions of the past have become historic techniques, and should only be found in museums and the history books of the great men and women who led our profession 70 - 100 years ago. By 2000 I required thorascopic instruments to be available in my operating room whenever I performed a chest procedure other than a simple bronchoscopy or mediastinoscopy. A lobectomy become a minimally invasive procedure as I gradually decreased the size of my incisions. As the incisions become smaller, so did the degree of rib spreading, until eventually I was performing major resections with no rib spreading whatsoever. Most of the instruments for the resections were the usual thoracic instruments available in most general thoracic instrument trays. Very importantly, the patients were always draped for a thoracotomy. At this stage, over 60% of all my cases that were previously performed through a thoracotomy were now VATS procedures. VATS lobectomy had been proposed in the mid-1990s by McKenna and others.[4] I began by using a video-assisted approach, gradually decreasing the size of my incision. One of the incisions needs to be big enough to remove the specimen, this being the so-called access incision. This incision is about 4 - 6 cm long, sometimes larger,

16 AJTCCM VOL. 24 NO. 1 2018

depending on the size of the lesion to be removed. There is no point struggling with a small incision only to enlarge it to remove the specimen. My standard approach in the early days was a 3-port approach, with the larger anterior incision for the access port. During 2015 I was invited to Prof. Gonzalez-Rivasâ&#x20AC;&#x2122;s unit in La Corona, Spain, where I witnessed the uniportal lobectomy.[5] I subsequently spent 2 weeks in the Shanghai Pulmonary Hospital, observing not only the uniportal approach but highly skilled surgeons using various techniques, including the standard 3-port, a 2-port approach and of course the uniportal. I also observed a subxiphoid lobectomy and a subcostal incision. Thymectomy was observed through a right and a left thoracoscopy, as well as via the subxiphoid approach. These are new attempts to totally prevent any rib spreading. Even minimum spreading of the ribs, in removing the specimen, can cause a debilitating post-thoracotomy syndrome.

2008: Pectus surgery

By 2008 I was attending international cardiothoracic meetings. In 2008 was the World Society of Cardiovascular and Thoracic Surgeons congress on the island of Kos in Greece. At this meeting, Mustafa Yuksel from Istanbul presented his work on pectus excavatum. He was performing the Nuss procedure, which is a minimally invasive repair of the disorder. We became very friendly, and over dinner with our wives he invited me to visit his unit. Later that year that is exactly what I did. Over a week spent in the vibrant city of Istanbul in Turkey, I observed Prof. Yuksel operating, while learning how to do the operation myself. I was collecting cases, and in October 2008 I invited Prof. Yuksel to join me in Johannesburg, where we operated on four cases. Also present at that time was Prof. Pilegaard from Denmark, who had his own modification of the Nuss procedure. The following month I had a further four cases. To my delight, I discovered that Prof. Donald Nuss himself was visiting SA, and he kindly agreed to assist me with my cases. Prof. Nuss is a South African who qualified in Cape Town as a paediatric surgeon under Prof. Jannie Louw. Nuss spent time at the Mayo Clinic in the USA, following which he was offered a position in Norfolk, Virginia, USA. He began his research in minimally invasive repair of pectus excavatum in 1987.[8] After his presentation in 1998, the repair of the disorder was revolutionised. Over the next few years, the VATS repair became the new standard replacing the Ravitch operation, which was the routine operation at the time. I had met Prof. Ravitch as a registrar when he visited our unit in Cape Town in 1980. He was a truly amazing man, and the leading paediatric surgeon of his time. He gave a lecture on his procedure, but also mentioned his work on surgical staplers. He introduced surgical staplers to the West from the USSR (now Russia). His research on staplers was undertaken at the back of a friendâ&#x20AC;&#x2122;s dry-cleaning business, a Mr Hirsch, and this eventually became the company USSC (United States Surgical Corporation), which after many transitions is now Medtronic. A true innovator, I am sure he would have been at the forefront of research in the constantly changing field of surgery: When you think you know everything about your field of study, when there is nothing extra to learn, that is the time to retire. My teachers have been most gracious in their sharing of knowledge. I will always be grateful to Yuksel, Pilegaard, Nuss and so many others who have freely passed on their surgical secrets, and in so doing affected the lives of literally thousands of patients.

REVIEW During this period, I became a member of the Chest Wall Interest Group, which has now been reformatted as a society, the Chest Wall International Group (CWIG). As part of this group I have met many other truly innovative doctors. Prof. Horacio Abramson, after observing the Nuss procedure, said, ‘If the Nuss works for pectus excavatum, why not a reverse Nuss for the carinatum?’ And so, the Abramson procedure for the pectus carinatum was born. External braces for pectus carinatum had been used for some time, but these had their own problems. Marcelo Martinez-Ferro introduced a custom-made adjustable brace with excellent results. My own experience now includes 134 Nuss procedures for pectus excavatum, 8 Abramson procedures for pectus carinatum and 14 custom-made dynamic compression braces for pectus carinatum. My advice is the brace as the first option for carinatum, which has been shown to correct most defects, reserving surgery for cases where the patients are not willing to wear the device, some older patients, patients with a very high compression required to correct the problem and some complex cases where a Ravitch may be required.

2018: The future

We are now in the beginning of what I believe to be a new era in surgery, the era of robotic surgery. The discussion we are now in the middle of echoes almost identically what was said when thorascopic surgery was introduced in 1991: • It is not safe! (Farjah[6]) • It is expensive! (Casali[7]) • There are no advantages! • Complications are potentially disastrous! • It is no better than thoroscopic surgery! (This being the new standard.) The robot now available is the De Vinci. This is being constantly upgraded, the cost is rapidly coming down, many new companies are on the horizon and more and more articles are being published on the advantages of incorporating robotic technology in operations. Handheld robots will soon be available, allowing the surgeon to combine the technology of open, VATS and robotic techniques (personal observation, Israel).

Discussion

Thoracoscopy can be divided into diagnostic procedures and therapeutic procedures. For pleural effusion, the sensitivity of thoracoscopy is 92% - 97%, and its specificity is 99%. Thorascopic lung biopsy has a similar sensitivity of over 90%.[3] The diagnostic procedures can be performed under sedation and local anaesthetic, and in many cases the patients can be discharged on the day of the procedure. Table 1 gives my indications for diagnostic procedures. The therapeutic procedures, more correctly in my view, should be referred to as VATS. If a thoracotomy is indicated, a VATS procedure is usually a consideration. Pleural adhesions are a relative contraindication. Flimsy adhesions can be broken down. Dense adhesions, as are often found in the African scenario, may be more difficult. In my learning phase, I used the thoracoscope as a light source, with the incision becoming smaller and smaller. My indications for VATS are given in Table 2.

Table 1. Diagnostic indications (as performed in the author’s practice) Pleural biopsies Lung biopsies Mediastinal gland biopsies Staging for lung cancer Table 2. Therapeutic indications (as performed in the author’s practice) Bullous ligation Pleurodesis Pleurectomy Decortication Lobectomy Pneumonectomy Mediastinal lymph node clearance Oesophagectomy (chest part; abdominal mobilisation of stomach performed laparoscopically) Ruptured diaphragm Clotted haemothorax Trauma Thymectomy Mediastinal masses Posterior mediastinal masses Sympathectomy First rib resection Pectus excavatum (Nuss procedure) Pectus carinatum (Abrahams procedure) Work is taking place on a subcostal as well as a subxiphoid approach, to further limit postoperative pain. At the Shanghai Pulmonary Hospital in China, I observed these procedures being performed with great skill. Very importantly, the surgeon at all times must be ready to convert to an open thoracotomy. All my patients are draped and prepared for a thoracotomy. Is it ethical to still be doing trials comparing thoracotomy to VATS? Many studies have been performed comparing anterolateral thoracotomy to VATS, and have shown a marked difference in the postoperative pain and quality of life after each type, confirming the superiority of VATS.[10-13] The main reason not to perform VATS is probably the answer to the most important question that should be asked, which can be summarised as ‘unwilling to’ or ‘unable to’.[9] Possible reasons for this are lack of experience, the cost of equipment and the culture and bias of the local unit. There is a learning curve, especially for VATS lobectomy and thymectomies, but the simple procedures should be within the capabilities of all thoracic surgeons.[14,15]

Conclusion

My prediction: We are in the middle of a revolution of new technology. New is not always better, but change is inevitable. As Africans, we

AJTCCM VOL. 24 NO. 1 2018

REVIEW have the capabilities to be equal to the best in the world. Our leading universities have the facilities to train the surgeons, we have the patients, and we most certainly have the young graduates who are more than capable of being the best in the world. I would encourage the newly qualified to spend time in leading international units. The time spent will broaden the mind, and is never wasted. Thorascopic chest surgery is the modern standard. The teaching of thorascopic techniques should be standard in all our university departments. A newly qualified specialist thoracic graduate who is not trained to perform minimally invasive thoracic procedures is inadequately trained. The next generation will slowly move to robotic surgery. We are now in the early phases, but the future is rapidly approaching. The other field that I believe has great potential is the field of 3D printing. This will involve prostheses as well as other equipment. VATS allows the same operation to be performed safely, with less pain, shorter hospital stay and with outcomes that are in many cases better than those of open surgery. The future is exciting. As Africans, we need to be, and can be, at the forefront of development. Acknowledgements. None. Author contributions. Sole author. Funding. None. Conflicts of interest. None. The opinions expressed in this article are my own personal views as modified by the many international experts that I have witnessed. 1. Jacobaeus HC. The cauterisation of adhesions in artificial pneumothorax treatment of pulmonary tuberculosis under thoracoscopic control. Proc R Soc Med 1923;16(Electro Ther Sect):45-62. 2. Colt HG. Thoracoscopy: Window to the pleural space. Chest 1999;116(5):14091415. https://doi.org/10.1378/chest.116.5.1409

18 AJTCCM VOL. 24 NO. 1 2018

3. Boutin C, Loddenkemper R, Astoul. Diagnostic and therapeutic thoracoscopy: Techniques and indications in pulmonary medicine. Tuber Lung Dis 1993;74(4):225239. https://doi.org/10.1016/0962-8479(93)90048-3 4. McKenna RJ, Houck W, Fuller CB. Video-assisted thoracic surgery lobectomy: Experience with 1 100 cases. Ann Thorac Surg 2006;81(2):421-425. https://doi. org/10.1016/j.athoracsur.2005.07.078 5. Gonzalez-Rivas D. Ann Cardiothoracic Surg 2014;3(2): 204-207. https://doi. org/10.3978/j.issn.2225-319X.2014.03.05 6. Farjah F, Wood DE, Mulligan MS, et al. Safety and efficacy of video-assisted versus conventional lung resection for lung cancer. J Thorac Cardiovasc Surg 2009;137:14151421. https://doi.org/10.1016/j.jtcvs.2008.11.035 7. Casali G, Walker WS. Video-assisted thoracic surgery lobectomy: Can we afford it? Eur J Cardiothorac Surg 2009;35(3):423-428. https://doi.org/10.1016/j.ejcts.2008.11.008 8. Nuss D, Kelly RE, Croitoru DP, et al. A 10-year review of minimally invasive technique for the correction of pectus excavatum. J Pediatr Surg 1998;33(4):545-552. https://doi. org/10.1016/s0022-3468(98)90314-1 9. Cheng-Che Tu and Po-Kuei Hsu. The long-waited high level evidence in thoracic surgery. Ann Transl Med 2016;4(19): 383. https://doi.org/10.21037/atm.2016.08.05 10. Bendixen M, Jørgensen OD, Kronborg C, et al. Postoperative pain and quality of life after lobectomy via video-assisted thoracoscopic surgery or anterolateral thoracotomy for early stage lung cancer: A randomised controlled trial. Lancet Oncol 2016;17(6):836-844. https://doi.org/10.1016/s1470-2045(16)00173-x 11. Yan TD, Black D, Bannon PG, et al. Systematic review and meta-analysis of randomised and nonrandomised trials on safety and efficacy of video-assisted thoracic surgery lobectomy for early-stage non-small-cell lung cancer. J Clin Oncol 2009;27(15):25532562. https://doi.org/10.1016/j.hlc.2010.04.127 12. Cao C, Manganas C, Ang SC, et al. Video-assisted thoracic surgery versus open thoracotomy for non-small cell lung cancer: A meta-analysis of propensity scorematched patients. Interact Cardiovasc Thorac Surg 2013;16(3):244-249. https://doi. org/10.1093/icvts/ivs472 13. Long H, Lin ZC, Lin YB, et al. [Quality of life after lobectomy for early stage nonsmall cell lung cancer-video-assisted thoracoscopic surgery versus minimal incision thoracotomy]. Ai Zheng 2007;26(6):624-628. 14. Petersen RH, Hansen HJ. Learning curve associated with VATS lobectomy. Ann Cardiothorac Surg 2012;1(1):47-50. 15. Zhao H, Bu L, Yang F, et al. Video-assisted thoracoscopic surgery lobectomy for lung cancer: The learning curve. World J Surg 2010;34(1):2368-2372. 16. Boutin, C, Cargnino, P, Viallat, PR. Thoracoscopy in the early diagnosis of malignant pleural effusions. Endoscopy 1980;12(4):155-160. https://doi.org/10.1055/s-2007-1021734

Accepted 11 October 2017.

REVIEW

Inhaled corticosteroids in COPD: Personalising the therapeutic choice J A Shaw, MB ChB (UCT), MMed (Int), FCP (SA); E M Irusen, MB ChB , FCP (SA), PhD, FCCP Division of Pulmonology, Department of Medicine, Tygerberg Academic Hospital and Stellenbosch University, Cape Town, South Africa Corresponding author: J A Shaw (janeshaw@sun.ac.za)

There has been a recent surge in interest in the role of inhaled corticosteroids (ICS) in the treatment of COPD, especially regarding patients with high eosinophil counts. Evidence has shown that despite the increase in localised adverse effects and a small increase in non-fatal pneumonia events with ICS use, ICS still have an important role to play in reducing exacerbation rates and addressing the inflammation that is at the heart of the pathogenesis of COPD. Current international guidelines recommend the use of ICS only in patients with severe disease. This review examines the potential role of ICS in all COPD patients. Afr J Thoracic Crit Care Med 2018;24(1):19-25. DOI:10.7196/AJTCCM.2018.v24i1.184

The use of inhaled corticosteroids (ICS) in chronic obstructive pulmonary disease (COPD) remains a topic of contention among doctors and data on the subject are often contradictory.[1] Recently, there has been a trend toward down-playing ICS use in COPD treatment regimens in all but the most severe group of patients.[2,3] The current Global Initiative for Chronic Obstructive Lung Disease (GOLD) strategy document[4] suggests that ICS should only be used in GOLD C and D patients (i.e. those with two or more exacerbations or one exacerbation leading to hospital admission). There are two primary reasons given against ICS use in other categories of COPD. Firstly, the GOLD strategy cites in vitro evidence that the inflammation present in COPD is inherently corticosteroid resistant. The second concern raised is the highly topical increased risk of nonfatal pneumonia in patients with severe COPD who use ICS. Here, we review the evidence for the abovementioned assertions by examining data on the efficacy of ICS in COPD, the pharmacological actions of ICS with relation to the pathogenesis of COPD, as well as examining the strength of the evidence for an increased risk of pneumonia in this population. We conclude with recommendations on the use of ICS.

Known clinical effects of ICS in COPD

In patients with COPD, exacerbations are associated with an increased risk of mortality, poorer quality of life, and accelerated long-term decline in lung function.[5,6] These effects are greater in those who experience such events more frequently.[7,8] There is good evidence to show that long-term use of ICS reduces the rate of exacerbations in patients with both moderate and severe airflow limitation.[9,10] ICS use has also been shown to affect patients’ quality of life (QOL) and symptoms. In a meta-analysis of ICS use for stable COPD, the rate of decline in QOL as measured by the St George’s Respiratory Questionnaire (SGRQ) was reduced, and there was a small, but statistically significant, improvement in patients’ QOL.[9] In this same

meta-analysis it was noted that some studies also showed a reduction in rescue bronchodilator use.[9] The Withdrawal of Inhaled Steroids during Optimized Bronchodilator Management (WISDOM) study by Magnusson et al.[11] suggested a slower rate of decline in the forced expiratory volume in one second (FEV1) in patients who received ICS therapy; however, long-term use of ICS has not consistently been shown to reduce the rate of decline in FEV1, or to have any significant effect on mortality in COPD patients.[9] These observations were most recently corroborated in the Study to Understand Mortality and MorbidITy in COPD (SUMMIT).[12] While ICS are currently recommended for patients with an FEV1 value <60% of predicted and a history of exacerbations, the SUMMIT sub-study suggests that ICS may also have a role in other patient groups, as there were benefits in those with an FEV1 >60% of predicted and in patients with no exacerbation history.[10]

Combination therapy and evidence for synergistic effects

A meta-analysis of treatment options for patients with severe COPD who remained uncontrolled on short-acting muscarinic-antagonists (SAMA) and short-acting beta-agonists (SABA) alone, found that a combination of an ICS and long-acting beta-agonist (LABA) was the highest-ranked intervention for improving QOL compared with placebo at 6 and 12 months.[13] Long-acting muscarinic-antagonist (LAMA) and LABA therapy were independently ranked second and third, and ICS alone was ranked fourth at 6 months. Martinez et al.[10] recently demonstrated that the combination of ICS/LABA reduced exacerbation rates to a greater degree than either component alone. It has been proposed that the mechanism of this synergistic effect is the LABA enhancing glucocorticoid receptor nuclear translocation and efficacy. This was demonstrated in a study of induced-sputum macrophages: the combination of salmeterol and 100 µg fluticasone propionate (FP) significantly increased nuclear glucocorticoid

AJTCCM VOL. 24 NO. 1 2018

REVIEW receptor levels equivalent to that of 500 µg FP, enhanced ICS-induced mitogen-activated protein kinase phosphatase-1 (MAPK1) mRNA copies and doubled glucocorticoid response element-luciferase reporter gene activity.[14] There is also evidence that the budesonide/ formoterol combination enhanced the expression of pro-surfactant protein-B in the lungs of COPD patients – a population in which surfactant expression is decreased and which has also been associated with poor health outcomes.[15]

Triple therapy

Recently, data have emerged regarding the so-called ‘triple therapy’, which includes a combination of ICS/LAMA/LABA treatment. Clinical trials have previously tested the effectiveness of triple therapy delivered by two separate devices, compared to LAMA monotherapy, LAMA/LABA combination therapy using separate inhalers, and combined ICS/LABA treatment. These studies showed a short-term superiority of triple therapy in terms of lung function and patient-reported outcomes when compared with LAMA monotherapy or ICS/LABA treatment.[16] One study that compared triple therapy with LAMA/LABA (tiotropium and salmeterol) combination therapy (the latter group having had the ICS (FP) sequentially decreased and then completely withdrawn from the initial triple regimen) noted no significant difference in the exacerbation rate. However, they did observe a significant decrease in FEV1 in the group in which ICS was withdrawn, as well as a worsening of dyspnoea scores and health status outcomes.[11] It is important to note that this was a non-inferiority study and thus equivalence or superiority cannot be presumed. Two large randomised trials have compared the ICS/LAMA/ LABA combination in a single inhaler device with ICS/LABA, with similar results: in patients with severe COPD, triple therapy was found to be superior to ICS/LABA combination in improvements in FEV1, reduction in exacerbation rate, as well as health-related QOL scores. [16,17] In TRILOGY, there was a 23% reduction in exacerbations with extra-fine beclomethasone dipropionate, formoterol furoate and glycopyrronium bromide (BDP/FF/GB) compared with BDP/ FF.[16] In FULFIL (Lung FUnction and quality of LiFe assessment in COPD with closed trIpLe therapy), the addition of umeclidinium to FF/VI resulted in a net FEV1 gain of 179 mL compared with BDP/ FF at 1 year, with a higher percentage of subjects who were SGRQ responders in the former and a mean SGRQ change of –4.6 units compared with -1.9 U with BDP/FF.[17] Such a clinically significant difference in the SGRQ (–4U is the clinically significant threshold that patients can perceive) has seldom been documented in COPD trials. Currently, there are no good-quality prospective data comparing triple therapy with the LABA/LAMA combination.[18]

Withdrawal of ICS

A recent meta-analysis on the effects of withdrawal of ICS showed that, while ICS withdrawal did not significantly increase the overall rate of COPD exacerbations, a clinically important increased risk of severe exacerbation was detected. ICS withdrawal significantly impaired both lung function and QOL. The time to the first exacerbation was also significantly shorter in the patients who discontinued ICS.[19]

20 AJTCCM VOL. 24 NO. 1 2018

Corticosteroids, inflammation and COPD

It is well known that inflammation of the airways is present even in the early stages of COPD.[20] The dominant inflammatory cells are neutrophils; however, increased numbers of macrophages and CD8+ T lymphocytes are also present, all of which interact to produce chemokines, cytokines, proteases and reactive oxygen species that cause tissue damage and stimulate further inflammation.[21,22] The presence of inflammatory biomarkers in the sputum has been associated with disease progression and an increased risk of exacerbations,[23] while suppression of airway inflammation has been shown to improve lung function[24] and reduce exacerbation rates by up to 30%.[25] Corticosteroids suppress the multiple inflammatory genes that are activated in chronic inflammatory diseases, such as COPD. This is achieved mainly by reversing histone acetylation of activated inflammatory genes through binding of liganded glucocorticoid receptors to coactivators, and recruitment of histone deacetylase-2 (HDAC2) to the activated transcription complex.[26] It has been suggested that the inflammation specific to COPD is resistant to corticosteroid effects, possibly through reduced HDAC2 expression.[27] However, it has been demonstrated that the action of budesonide in suppressing airway inflammation is independent of the HDAC2 pathway.[28] Other postulated mechanisms of ICS resistance in COPD include activation of mitogen-activated protein (MAP) kinase pathways by certain cytokines, excessive activation of the transcription factor activator protein 1, raised macrophage migration inhibitory factor, and increased P-glycoprotein-mediated drug efflux. [29] However, a meta-analysis of studies examining inflammatory biomarkers in sputum, bronchoalveolar lavage fluid and biopsy specimens concluded that ICS were effective in reducing CD4+ and CD8+ T cell counts, as well as neutrophil and lymphocyte counts. It was noted that macrophage counts were increased in the presence of ICS. The authors hypothesised that these important immunomodulatory effects could be the reason for the efficacy of ICS in reducing exacerbations, as well as the mechanism underlying the apparent increase in pneumonia.[29] A subsequent study concluded that even in the presence of smoking, long-term ICS treatment may lead to anti-inflammatory effects in the lung as ICS reduced bronchial mast cells, CD3+, CD4+ and CD8+ cells, as well as sputum neutrophils and lymphocytes.[30] In addition, a recent report has demonstrated that ICS discontinuation in patients on long-term ICS with moderate-tosevere COPD resulted in increased airway inflammation, as reflected by increased numbers of bronchial CD3+, CD4+, and CD8+ T cells and mast cells, as well as increased sputum total cell count, macrophages, neutrophils and lymphocytes.[31] Another study identified an ICS-insensitive macrophage phenotype in COPD. These macrophages showed significantly lower expression of all receptors, and were associated with higher levels of release of active matrix metalloproteinase 9 compared with macrophages of non-smokers and smokers without COPD.[32] A COPD phenotype that is more likely to respond to ICS has not yet been identified, as response is not predicted by oral steroid response, bronchodilator reversibility or bronchial hyper-responsiveness.[32] However, there is evidence that long-term benefits of ICS on lung function decline in patients with moderate-to-severe COPD are most pronounced in

REVIEW patients with fewer pack years smoking history, less severe emphysema (limited hyperinflation and preserved diffusion) and lower sputum inflammatory cell counts.[33]

Eosinophilic inflammation in COPD

A post hoc analysis of two large multinational studies comparing treatment with ICS/LABA (fluticasone fuorate/vilanterol[VI]) to VI monotherapy, found that patients with a blood eosinophil count â&#x2030;Ľ2.4%, responded better to the combination, with a generally linear relationship of further exacerbation reduction with higher eosinophil counts. The inference from their analysis was that, in general, low eosinophil counts coupled with high levels of smoking could predict a poorer response to ICS, with no significant reduction in exacerbation rates.[34] The linear association of blood eosinophils with exacerbation

reduction by ICS was also noted in a further analysis of the WISDOM study using tiotropium, salmeterol and FP.[35] In a post hoc analysis of the INSPIRE (Investigating New Standards for Prophylaxis in Reduction of Exacerbations) study using an eosinophil cut-off of 2%, FP/salmeterol was associated with a 25% relative risk reduction of exacerbations compared with tiotropium alone.[36] This, and other evidence regarding the anti-inflammatory effects of ICS in COPD, is captured in Table 1.[40-45]

The risk of pneumonia

There is no doubt that the use of ICS is associated with an increased risk of localised adverse effects: oropharyngeal candidiasis, dysphonia and hoarseness, as well as an increased risk of cataracts.[32,46] Additionally, ICS increase the risk of non-fatal serious adverse pneumonia

Table 1. Key studies identifying the anti-inflammatory effects of ICS in COPD Study Findings Reduction in bronchoalveolar lavage fluid cellularity, lactoferrin, lyzozyme and albumin levels (markers of Thompson, 1992[37] inflammation). Saetta, 1997;[38] Saetta, The key inflammatory cells mediating inflammation in COPD were CD68+ macrophages, neutrophils and 1998[39] CD8+ cytotoxic lymphocytes. [23] Inflammatory biomarkers in the sputum were associated with disease progression and an increased risk of Bhowmik, 2000 exacerbation. Neutrophils were the dominant airway inflammatory cells in COPD. Cosio, 2002[22] Barnes, 2003[21]

Macrophages and CD8+ T lymphocytes were also increased.

Hattotuwa, 2002[40]

Above cells interacted to cause tissue damage and further inflammation. Reduction in CD8:CD4 ratio.

Sugiura, 2003[24] Sin, 2003[25] Hogg, 2004[22] Ozol, 2005[41] Gan, 2005[42] (metaanalysis) Barnes, 2006[43] Bathoorn, 2008[44] Lapperre, 2009[45] Jen, 2012[29] (metaanalysis) Wang, 2013[28] Hoonhorst, 2014[30] Chana, 2014[32] Snoeck-Stroband, 2015[33] Hinds, 2016[34]

No reduction in CD8+, CD68+ cells or neutrophils observed, suggested that ICS worked on specific aspects of airway inflammation. Suppression of airway inflammation improved lung function. Suppression of airway inflammation reduced exacerbation rates. Inflammation of the airways was present even in the early stages of COPD. Reduction in interleukin (IL)-8 levels in bronchoalveolar lavage fluid mean percentage of neutrophils. Reduction in sputum total cell, neutrophil and lymphocyte counts when given in adequate dose and duration. Reduction in CD8+, CD45+ and CD4+ cells, but no change in CD68+ cells seen. Reduction in sputum eosinophilia. Reduction in counts of mucosal CD3+, CD4+, CD8+ and mast cells, with effects maintained after 30 months. Reduction in CD4+ and CD8+ T cell counts, as well as neutrophil and lymphocyte counts in bronchoalveolar lavage fluid and biopsy specimens. Action of budesonide on airway inflammation was independent of the HDAC2 pathway. Reduction in bronchial mast cells, CD3+, CD4+ and CD8+ cells, as well as sputum neutrophils and lymphocytes. An ICS insensitive macrophage phenotype identified (lower expression of all receptors, higher levels of release of active matrix metalloproteinase 9). Long-term benefits on lung function decline in patients with moderate-to-severe COPD were most pronounced in patients with fewer pack years smoking history, less severe emphysema and lower sputum inflammatory cell counts. Patients with blood eosinophil counts â&#x2030;Ľ2.4% responded better to ICS/LABA. A linear relationship of further exacerbation reduction with higher eosinophil counts was found.

Watz, 2016[35] Pavord, 2016[36] Kunz, 2017[31]

Low eosinophil counts coupled with high levels of smoking could predict a poorer response to ICS, with no significant reduction in exacerbation rates. Higher blood eosinophils associated with reduction in exacerbation rate in a linear relationship. Blood eosinophils of >2% associated with a 25% relative risk reduction of exacerbations with ICS use. Discontinuation resulted in increased numbers of bronchial CD3+, CD4+, and CD8+ T cells and mast cells, as well as increased sputum total cell count, macrophages, neutrophils and lymphocytes.

ICS = inhaled corticosteroids; COPD = chronic obstructive pulmonary disease; LABA = long-acting beta agonist.

AJTCCM VOL. 24 NO. 1 2018

22 AJTCCM VOL. 24 NO. 1 2018

Rossi, 2014 WISDOM

[59]

Wang, 2016[60]

Wedzicha, 2016 IMPACT

[15]

Singh, 2016 FLAME

Vestbo, 2016[58] TRILOGY

SALFORD

Magnussen, 2014[11]

[57]

[56]

Wedzicha, 2014 INSTEAD

Vogelmeier, 2016[55] FORWARD

[54]

Zhong, 2015

Vogelmeier, 2013[53]

ILLUMINATE, LANTERN Pooled analysis

Wedzicha, 2008 Calverly, 2011 reanalysis[51] Dransfield et al., 2013[52]

INSPIRE

[50]

[49]

Tashkin, 2008

SHINE

Calverly, 2007[48]

Moderate to severe obstruction

Usual care (1 403) ICS/LABA/LABA (687)

ICS/LABA (1 396)

LABA/LAMA (1 242)

LABA (246) ICS/LABA/LAMA (1 243)

LABA (591) ICS/LABA (250)

ICS/LABA (595)

Pneumonia controls (74 849)

ICS/LABA (681) Moderate to severe obstruction with LABA/LAMA (1 675) exacerbations ICS/LABA (1 679) COPD with ICS use Pneumonia cases (19 838)

Severe obstruction with exacerbations

COPD with excerbations

Severe obstruction with exacerbations

Moderate obstruction

Severe obstruction with exacerbations

ICS/LABA (3 dose variations 820/806/811) LABA/LAMA (259 and 372) ICS/LABA (264 and 369)

LABA (818)

LAMA (665)

ICS/LABA (845) ICS/LABA (658)

LABA (284)

ICS (275)

ICS/LABA (1 011) Placebo (300)

COPD with exacerbations

Severe obstruction with exacerbations

ICS (947)

LABA (960)

Table 2. The effect of ICS use in COPD on pneumonia risk in important clinical trials Study Population Intervention (N) Moderate to severe obstruction Placebo (851) TORCH

Odds increased with increasing ICS dose.

Odds ratio for pneumonia = 1.25.

continued...

Incidence of pneumonia 3.2% in LABA/LAMA group and 4.8% in ICS/ LABA group (p=0.02).

Incidence of pneumonia in both groups = 3%

Incidence of pneumonia increased from 5.5% to 5.8% in ICS group, not statistically significant. No excess pneumonia risk.

Hazard ratio for time to first exacerbation = 1.06.

Incidence of pneumonia 0% in LABA group and 0.7% in ICS group, not statistically significant.

Incidence of pneumonia 1.8% in LABA group and 3.8% in ICS/LABA group.

Higher rate in more severe COPD.

Incidence of pneumonia 0.5% in LABA/LAMA group and 2.2% in the ICS/ LABA group (p=0.0074).

High-dose ICS discontinued.

Higher risk in those with baseline severe dyspnoea and baseline raised CRP. Incidence of non-fatal pneumonia increased in ICS group.

Hazard ratio of 1.94 for having pneumonia in ICS group (p=0.008); unchanged when analysis restricted only to patients with CXR.

No increase in risk compared with placebo.

No increase in risk of death from pneumonia in ICS groups.

Effect on pneumonia risk Risk of pneumonia over 3 years 18.3% and 19.6% in ICS/LABA and ICS groups v. 12.3% and 13.3% in placebo and LABA groups (p<0.001).

REVIEW

ICS = inhaled corticosteroids; COPD = chronic obstructive pulmonary disease; LABA = long-acting beta agonist, LAMA = long-acting muscarinic antagonist, CXR = chest radiograph, CRP = C-reactive protein.

ICS users had a numerically lower risk of death. At 52 weeks the groups had a risk of pneumonia of 1.9% and 1.8%. Moderate to severe obstruction with ICS/LABA/LAMA (911) exacerbations ICS/LABA (899) Lipson, 2017[17]

Di Martino, 2014

[63]

Cascini, 2017 reanalysis[46]

FULFIL

Relative risk of pneumonia = 1.23 with past ICS use. Pneumonia controls (12 564)

Risk rates increased with increasing dose of ICS and with increasing age.

Relative risk of pneumonia = 2.29 with current ICS use. Other ICS (1 719) Pneumonia cases (3 141) Moderate to severe COPD Morjaria, 2017 reanalysis[62] OUTPUL

Table 2. (continued) The effect of ICS use in COPD on pneumonia risk in important clinical trials Study Population Intervention (N) Effect on pneumonia risk Moderate to severe obstruction No ICS (2 292) Incidence of pneumonia 5.6% in no-ICS group and 6.8% with ICS use, and UPLIFT was higher with fluticasone proprionate than ‘other ICS’ (p=0.012). Fluticasone roprionate (1 981) Tashkin, 2008[61]

REVIEW events, without conferring any difference in overall mortality rate.[47] This effect was first unexpectedly identified in the large prospective TORCH (TOwards a Revolution in COPD Health) study,[48] and has subsequently been shown in numerous large randomised trials (Table 2)[49-63] and smaller studies. Recent studies have reported a stronger association with pneumonia at higher doses of ICS, suggesting a doseresponse relationship, and age older than 65 has been identified as an additional factor which increases risk.[46] However, it is worth noting that the incidence of pneumonia in all the above studies is low (<6% overall and usually between 0% and 2% more than placebo/LABA). The latter is a reminder that COPD itself predisposes patients to pneumonia through an altered microbiome and the toxic effects of cigarette smoking, as well as the fact that the disease occurs in older individuals who may have used oral corticosteroids, which leads to further suppression of the immune response. The prevailing hypothesis for the mechanism of the propensity to pneumonia with ICS is local airway immunosuppression and a diminished innate immune response to pathogens. Paradoxically, this diminished inflammatory effect is also hypothesised to be the mechanism for the lack of severity of these pneumonia events and thus the low fatality rate.[64,65] The overall quality of the studies dedicated to examining the treatment of COPD is high. There are a number of wellconducted randomised controlled trials with large patient numbers available for analysis. However, there have been a few criticisms of the studies from which the association between ICS use and pneumonia is drawn. The first criticism is a methodological one regarding differences in trial design, patient population and the type of interventions, all of which combine to increase differences in the reported rate of pneumonia events between studies.[66] Dransfield et al. [52] reported a twofold increase in the pneumonia events in the combination ICS/LABA (FF/VI) arm compared with the LABA (VI) monotherapy arm in patients with at least one exacerbation in the previous year and severe airflow obstruction. They recruited >800 patients

in each arm of the study, which in another context would be considered a large study. However, the recent SUMMIT trial included >4 000 patients in each arm (FF/VI v. the monotherapy components and placebo), with only moderate airflow obstruction, and reported that the difference in pneumonia events between the combination and the placebo arm was not statistically significant.[10] These two studies highlight the difficulties with interpretation of the existing body of evidence: two large studies reporting on the same outcome, about the same drugs, but with different patient populations, vastly different numbers and opposing conclusions. T h e s e c on d c r it i c i s m i s a l s o methodological in nature, and refers to the case definition (or lack thereof) used in the reporting of pneumonia events in these trials. The analysis of pneumonia events with ICS therapy in COPD is complicated by the overlap in clinical features of COPD exacerbations and pneumonia, and by the fact that pneumonia remains a relatively uncommon event when compared with acute exacerbations. While the majority of these trials were randomised controlled trials and had the highest quality evidence, they did not adhere to any formal definitions of pneumonia events. Rather, they relied on either the investigator’s retrospective assessment of a reported adverse respiratory event or database reporting systems. Chest radiographs were also not routinely performed or assessed at the time of the reported events.[66]

Conclusion

The recent suggestion that LAMA/ LABA should be the baseline treatment of all patients from GOLD B-D raises the following two concerns: (i) if inflammation is at the core of the pathogenesis of COPD, then this important component is not being addressed with bronchodilators alone; and (ii) does combination therapy with LAMA/LABA represent the ceiling effect, that is, can no further therapeutic gain be achieved? The available evidence shows that the use of ICS in patients with COPD does confer additional clinically significant beneficial effects, in particular the reduction of exacerbations. While the inflammation that occurs in COPD may be partially

AJTCCM VOL. 24 NO. 1 2018

REVIEW corticosteroid resistant, there is good evidence that the use of ICS reduces airway inflammation in a meaningful way. Furthermore, the data suggesting that certain COPD phenotypes will derive benefit from ICS, particularly patients with blood eosinophilia ≥2% or 150 cells/µL, is accumulating rapidly.[34] The search for a more reliable biomarker of the phenotype of ICS responsiveness is still underway. An increase in the risk of non-fatal pneumonia events has been documented; however, important methodological differences should be considered when interpreting these trials. It is worth noting that COPD patients in these studies have a baseline risk of pneumonia of up to 5.6% and the increase in risk is <2% (Table 2). It has been pointed out by other authors[67] that, in considering ICS use, the riskbenefit should be carefully weighed. In one report, the exacerbation reduction with ICS use was in the order of 190 events, compared with the minor increase of ~30 pneumonia events.[67] These and other data prompted the European Medicines Agency’s Pharmacovigilance Risk Assessment Committee (PRAC) report to state that, while increased risk of pneumonia remains a common side-effect for all inhaled corticosteroids, the benefits of ICS continue to outweigh the risk.[68] Finally, given the promising nature of the emerging literature on triple therapy, there is the possibility this may become the treatment of choice in GOLD D patients, and may have a role in other categories of severity as well. More research is needed to identify the true ICS-responsive phenotype, as well as to assess the effects of triple therapy in early stage COPD on lung function decline, as well as exacerbation rates. Careful attention will need to be paid to the rates of pneumonia and other adverse events in these patient groups so that an accurate riskbenefit assessment can be made, on an individual patient basis. Acknowledgements. None. Author contributions. JAS compiled the manuscript. EMI edited and approved it. Funding. None. Conflicts of interest. EMI has received lecture fees from AstraZeneca, Novartis, MSD, Boehringer Ingelheim and Aspen and is currently employed part-time by GlaxoSmithKline. JAS has previously received a travel scholarship from GlaxoSmithKline. 1. Ai-Kassimi FA, Alhamad EH. Chronic obstructive pulmonary disease lost in translation: Why are the inhaled corticosteroids sceptics refusing to go? Ann Thorac Med 2013;8(1):8-13. https://doi.org/10.4103/1817-1737.105711 2. Oh Y. Is the Combination of ICS and LABA, a therapeutic option for COPD, fading Away ? Tuberc Respir Dis 2017;80:93-94. https://doi.org/10.4046/trd.2017.80.1.93 3. Tariq S, Thomas E. Maintenance therapy in COPD: Time to phase out ICS and switch to the new LAMA / LABA inhalers? Int J COPD 2017;12(23):1877-1882. https://doi. org/10.2147/copd.s138006 4. Global Initiative for Chronic Obstructive Lung Disease (GOLD). Global Strategy for the Diagnosis, Management and Prevention of COPD. Bethesda: GOLD, 2017. http:// goldcopd.org (accessed 19 July 2017). 5. Rennard SI, Farmer SG. Exacerbations and progression of disease in asthma and chronic obstructive pulmonary disease. Proc Am Thorac Soc 2004;1(2):88-92. https:// doi.org/10.1513/pats.2306026 6. Wang Q, Bourbeau J. Outcomes and health-related quality of life following hospitalization for an acute exacerbation of COPD. Respirology 2005;10(3):334-340. https://doi.org/10.1111/j.1440-1843.2005.00718.x 7. Vestbo J, Edwards LD, Scanlon PD, et al. Changes in forced expiratory volume in 1 second over time in COPD. N Engl J Med 2011;365(13):1184-1192. https://doi. org/10.1056/NEJMoa1105482

24 AJTCCM VOL. 24 NO. 1 2018

8. Halpin DMG, Decramer M, Celli B, Kesten S, Liu D, Tashkin DP. Exacerbation frequency and course of COPD. Int J COPD 2012;7:653-661. https://doi.org/10.2147/ COPD.S34186 9. Yang I, Clarke M, Eha S, Fong K. Inhaled corticosteroids for stable chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2012;7:CD002991. https://doi.org/ 10.1002/14651858.CD002991.pub3 10. Martinez FJ, Vestbo J, Anderson JA, et al. Effect of fluticasone furoate and vilanterol on exacerbations of chronic obstructive pulmonary disease in patients with moderate airflow obstruction. Am J Respir Crit Care Med 2016;195:1-35. https://doi. org/10.1164/rccm.201607-1421OC 11. Magnussen H, Disse B, Rodriguez-Roisin R, et al. Withdrawal of inhaled glucocorticoids and exacerbations of COPD. N Engl J Med 2014;371(14):1285-1294. https://doi/org/10.1056/NEJMoa1407154 12. Vestbo J, Anderson JA, Brook RD, et al. Fluticasone furoate and vilanterol and survival in chronic obstructive pulmonary disease with heightened cardiovascular risk (SUMMIT): A double-blind randomised controlled trial. Lancet 2016;387(10030):1817-1826. https://doi.org/10.1016/S0140-6736(16)30069-1 13. Kew K, Dias S, Cates C. Long-acting inhaled therapy (beta-agonists, anticholinergics and steroids) for COPD: A network meta-analysis. Cochrane Database Syst Rev 2014;(3):CD010844. https://doi/org/10.1002/14651858.CD010844.pub2 14. Haque R, Hakim A, Moodley T, et al. Inhaled long-acting β2 agonists enhance glucocorticoid receptor nuclear translocation and efficacy in sputum macrophages in COPD. J Allergy Clin Immunol 2013;132(5):1166-1173. https://doi/org/10.1016/j. jaci.2013.07.038 15. Um SJ, Lam S, Coxson H, Man SFP, Sin DD. Budesonide/formoterol enhances the expression of pro surfactant protein-B in lungs of COPD Patients. PLoS ONE 2013;8(12). https://doi.org/10.1371/journal.pone.0083881 16. Singh D, Papi A, Corradi M, et al. Single inhaler triple therapy versus inhaled corticosteroid plus long-acting β2-agonist therapy for chronic obstructive pulmonary disease (TRILOGY): A double-blind, parallel group, randomised controlled trial. Lancet 2016;388(10048):963-973. https://doi.org/10.1016/S0140-6736(16)31354-X 17. Lipson DA, Barnacle H, Birk R, et al. FULFIL Trial: Once-daily triple therapy in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2017;196(4):438-446. https://doi.org/ 10.1164/rccm.201703-0449oc 18. Tan DJ, White CJ, Walters JA, Walters EH. Inhaled corticosteroids with combination inhaled long-acting beta2-agonists and long-acting muscarinic antagonists for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2016;2016(11):CD011600. https://doi.org/10.1002/14651858.CD011600.pub2 19. Calzetta L, Matera MG, Braido F, et al. Withdrawal of inhaled corticosteroids in COPD: A meta-analysis. Pulm Pharmacol Ther 2017;45:148-158. https://doi. org/10.1016/j.pupt.2017.06.002 20. Hogg JC, Chu F, Utokaparch S, et al. The nature of small-airway obstruction in chronic obstructive pulmonary disease. N Engl J Med 2004;350(26):2645-2653. https://doi. org/10.1056/NEJMoa032158 21. Barnes PJ, Shapiro SD, Pauwels RA. Chronic obstructive pulmonary disease: Molecular and cellular mechanisms. Eur Respir J 2003;22(4):672-688. https://doi.org /10.1183/09031936.03.00040703 22. Cosio MG, Majo J. Inflammation of the airways and lung parenchyma in COPD: Role of T cells. Chest 2002;121(5):160S-165S. https://doi.org/10.1378/chest.121.5_ suppl.160S 23. Bhowmik A, Seemungal TA, Sapsford RJ, Wedzicha JA. Relation of sputum inflammatory markers to symptoms and lung function changes in COPD exacerbations. Thorax 2000;55(2):114-120. https://doi.org/10.1136/thorax.55.2.114 24. Sugiura H, Ichinose M, Yamagata S, Koarai A, Shirato K, Hattori T. Correlation between change in pulmonary function and suppression of reactive nitrogen species production following steroid treatment in COPD. Thorax 2003;58(4):299-305. https:// doi.org/10.1136/thorax.58.4.299 25. Sin DD, McAlister F, Man SFP, Anthonisen NR. Contemporary management of chronic obstructive pulmonary disease: Scientific review. JAMA 2003;290(17):23012312. https://doi.org/10.1001/jama.290.17.2301 26. Barnes PJ. How corticosteroids control inflammation: Quintiles Prize Lecture 2005. Br J Pharmacol 2006;148(3):245-254. https://doi.org/10.1038/sj.bjp.0706736 27. Barnes PJ, Adcock IM. Glucocorticoid resistance in inflammatory diseases. Lancet 2009;373(9678):1905–1917. https://doi.org/10.1016/S0140-6736(09)60326-3. 28. Wang X, Nelson A, Weiler ZM, et al. Anti-inflammatory effects of budesonide in human lung fibroblasts are independent of histone deacetylase 2. Am J Respir Crit Care Med 2013;6:109-119. https://doi.org/10.1164/ajrccmconference.2011.183.1_ meetingabstracts.a2135 29. Jen R, Rennard SI, Sin DD. Effects of inhaled corticosteroids on airway inflammation in chronic obstructive pulmonary disease: A systematic review and meta-analysis. Int J Chron Obstruct Pulmon Dis 2012;2012(7):587-595. https://doi.org/10.2147/COPD.S32765

REVIEW 30. Hoonhorst SJM, ten Hacken NHT, Vonk JM, et al. Steroid resistance in COPD? Overlap and differential anti-inflammatory effects in smokers and ex-smokers. PLoS ONE. 2014;9(2):e87443. https://doi.org/10.1371/journal.pone.0087443 31. Kunz LIZ, ten Hacken NHT, Lapperre TS, et al. Airway inflammation in COPD after long-term withdrawal of inhaled corticosteroids. Eur Respir J 2017;49(1):1-9. https:// doi.org/10.1183/13993003.00839-2016 32. Chana KK, Fenwick PS, Nicholson AG, Barnes PJ, Donnelly LE. Identification of a distinct glucocorticosteroid-insensitive pulmonary macrophage phenotype in patients with chronic obstructive pulmonary disease. J Allergy Clin Immunol 2014;133(1):207-216.e11. https://doi.org/10.1016/j.jaci.2013.08.044 33. Snoeck-Stroband JB, Lapperre TS, Sterk PJ, et al. Prediction of long-term benefits of inhaled steroids by phenotypic markers in moderate-to-severe COPD: A randomized controlled trial. PLoS ONE 2015;10(12):1-15. https:/doi.org/10.1371/journal. pone.0143793 34. Hinds DR, DiSantostefano RL, Le HV, Pascoe S. Identification of responders to inhaled corticosteroids in a chronic obstructive pulmonary disease population using cluster analysis. BMJ Open 2016;6(6):e010099. https://doi.org/10.1136/ bmjopen-2015-010099 35. Watz H, Tetzlaff K, Wouters EFM, et al. Blood eosinophil count and exacerbations in severe chronic obstructive pulmonary disease after withdrawal of inhaled corticosteroids: A post-hoc analysis of the WISDOM trial. Lancet Respir Med 2016;4(5):390-398. https://doi.org/10.1016/S2213-2600(16)00100-4 36. Pavord ID, Lettis S, Locantore N, et al. Blood eosinophils and inhaled corticosteroid/ long-acting β-2 agonist efficacy in COPD. Thorax 2016;71(2):118-125. https://doi. org/10.1136/thoraxjnl-2015-207021 37. Thompson AB, Mueller MB, Heires AJ, et al. Aerosolized beclomethasone in chronic bronchitis: Improved pulmonary function and diminished airway inflammation. Am Rev Respir Dis 1992;146(2):389-395. https://doi.org/10.1164/ajrccm/146.2.389 38. Saetta M, Turato G, Facchini FM, et al. Inflammatory cells in the bronchial glands of smokers with chronic bronchitis. Am J Respir Crit Care Med 1997;156(5):1633-1639. https://doi.org/10.1164/ajrccm.156.5.9701081 39. Saetta M, Di Stefano A, Turato G, et al. CD8+ T-lymphocytes in peripheral airways of smokers with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1998;157(3 Pt 1):822-826. https://doi.org/10.1164/ajrccm.157.3.9709027 40. Hattotuwa KL, Gizycki MJ, Ansari TW, Jeffery PK, Barnes NC. The effects of inhaled fluticasone on airway inflammation in chronic obstructive pulmonary disease: A double-blind, placebo-controlled biopsy study. Am J Respir Crit Care Med 2002;165(12):1592-1596. https://doi.org/10.1164/rccm.2105025 41. Ozol D, Aysan T, Solak ZA, Mogulkoc N, Veral A, Sebik F. The effect of inhaled corticosteroids on bronchoalveolar lavage cells and IL-8 levels in stable COPD patients. Respir Med 2005;99(12):1494-1500. https://doi.org/10.1016/j.rmed.2005.04.025 42. Gan WQ, Man SF, Sin DD. Effects of inhaled corticosteroids on sputum cell counts in stable chronic obstructive pulmonary disease: A systematic review and a metaanalysis. BMC Pulm Med 2005;5(1):3. https://doi.org/10.1186/1471-2466-5-3 43. Barnes NC, Qiu YS, Pavord ID, et al. Antiinflammatory effects of salmeterol/ fluticasone propionate in chronic obstructive lung disease. Am J Respir Crit Care Med 2006;173(7):736-743. https://doi.org/10.1164/rccm.200508-1321oc 44. Bathoorn E, Liesker JJ, Postma DS, et al. Anti-inflammatory effects of combined budesonide/formoterol in COPD exacerbations. COPD 2008;5:282–290. https://doi. org/10.1080/15412550802363360 45. Lapperre TS, Snoeck-Stroband JB, Gosman MM, et al. Effect of fluticasone with and without salmeterol on pulmonary outcomes in chronic obstructive pulmonary disease: A randomized trial. Ann Intern Med 2009;151(8):517-527. https://doi. org/10.7326/0003-4819-151-8-200910200-00004 46. Cascini S, Kirchmayer U, Belleudi V, et al. Inhaled corticosteroid use in chronic obstructive pulmonary disease and risk of pneumonia: A nested case-control population-based study in Lazio (Italy) - The OUTPUL Study. COPD 2017;14(3):311317. https://doi.org/10.1080/15412555.2016.1254172 47. Kew KM, Seniukovich A. Inhaled steroids and risk of pneumonia for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2014;2014(3):CD010115. https://doi.org/ 10.1002/14651858.CD010115.pub2 48. Calverley P, Anderson J, Celli B, et al. Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease. New Engl J Med 2007;356(8):775789. https://doi.org/ 10.1056/NEJMoa063070 49. Tashkin DP, Rennard SI, Martin P, et al. Efficacy and safety of budesonide and formoterol in one pressurized metered-dose inhaler in patients with moderate to very severe chronic obstructive pulmonary disease: Results of a 6-month randomized clinical trial. Drugs 2008;68(14):1975-2000. 50. Wedzicha JA, Calverley PM, Seemungal TA, Hagan G, Ansari Z, Stockley RA. The prevention of chronic obstructive pulmonary disease exacerbations by salmeterol/ fluticasone propionate or tiotropium bromide. Am J Respir Crit Care Med 2008;177(1):19-26. https://doi.org/10.1164/rccm.200707-973OC

51. Calverley PMA, Stockley RA, Seemungal TAR, et alReported pneumonia in patients with COPD: Findings from the INSPIRE study. Chest 2011;139(3):505512. https://doi.org/10.1378/chest.09-2992 52. Dransfield MT, Bourbeau J, Jones PW, et al. Once-daily inhaled fluticasone furoate and vilanterol versus vilanterol only for prevention of exacerbations of COPD: Two replicate double-blind, parallel-group, randomised controlled trials. Lancet Respir Med 2013;1(3):210–223. https://doi.org/10.1016/S2213-2600(13)70040-7 53. Vogelmeier CF, Bateman ED, Pallante J, et al. Efficacy and safety of once-daily QVA149 compared with twice-daily salmeterol–fluticasone in patients with chronic obstructive pulmonary disease (ILLUMINATE): A randomised, doubleblind, parallel group study. Lancet Respir Med 2013;1(1):51-60 https://doi. org/10.1016/S2213-2600(12)70052-8 54. Zhong N, Wang C, Zhou X, et al. LANTERN: A randomized study of QVA149 versus salmeterol/fluticasone combination in patients with COPD. Int J Chron Obstruct Pulmon Dis 2015;5(10):1015-1026. https://doi.org/10.2147/COPD. S84436. 55. Vogelmeier C, Zhong N, Humphries MJ, et al. Indacaterol/glycopyrronium in symptomatic patients with COPD (GOLD B and GOLD D) versus salmeterol/ fluticasone: ILLUMINATE/LANTERN pooled analysis. Int J Chron Obstruct Pulmon Dis 2016;11(1):3189-3197. https://doi.org/10.2147/COPD.S116786. 56. Wedzicha JA, Singh D, Vestbo J, et al. Extrafine beclomethasone/formoterol in severe COPD patients with history of exacerbations. Respir Med 2014;108(8):11531162. https://doi.org/10.1016/j.rmed.2014.05.013. 57. Rossi A, van der Molen T, del Olmo R, et al. INSTEAD: A randomised switch trial of indacaterol versus salmeterol/fluticasone in moderate COPD. Eur Respir J 2014;44(6):1548-1556. https://doi.org/10.1183/09031936.00126814. 58. Vestbo J, Leather D, Diar Bakerly N, et al. Effectiveness of fluticasone furoatevilanterol for COPD in clinical practice. N Engl J Med 2016 9;375(13):1253-1260. https://doi.org/10.1056/NEJMoa1608033. 59. Wedzicha JA, Banerji D, Chapman KR, et al. Indacaterol-glycopyrronium versus salmeterol-fluticasone for COPD. N Engl J Med 2016;374(23):2222-2234. https:// doi.org/10.1056/NEJMoa1516385. 60. Wang CY, Lai CC, Yang WC, et al. The association between inhaled corticosteroid and pneumonia in COPD patients: The improvement of patients’ life quality with COPD in Taiwan (IMPACT) study. Int J Chron Obstruct Pulmon Dis 2016;11(1):2775-2783. https://doi.org/10.2147/COPD.S116750 61. Tashkin DP, Celli B, Senn S, et al. A 4-year trial of tiotropium in chronic obstructive pulmonary disease. N Engl J Med. 2008;359(15):1543-1554. https:// doi.org/10.1056/NEJMoa0805800. 62. Morjaria JB, Rigby A, Morice AH. Inhaled Corticosteroid use and the risk of pneumonia and COPD exacerbations in the UPLIFT Study. Lung 2017;195(3):281288. doi: 10.1007/s00408-017-9990-8 63. Di Martino M, Agabiti N, Bauleo L, et al. OUTPUL Study Group. Use patterns of long-acting bronchodilators in routine COPD care: the OUTPUL study. COPD 2014;11(4):414-423. https://doi.org/10.3109/15412555.2013.839646 64. Iannella H, Luna C, Waterer G. Inhaled corticosteroids and the increased risk of pneumonia: What’s new? A 2015 updated review. Ther Adv Respir Dis. 2016;10(3):235-255. https://doi.org/10.1177/1753465816630208 65. Drummond MB, Dasenbrook EC, Pitz MW, Murphy DJ, Fan E. Inhaled corticosteroids in patients with stable chronic obstructive pulmonary disease: A systematic review and meta-analysis. JAMA 2008;300(20):2407-2416. https://doi. org/10.1001/jama.2008.717 66. Bourbeau J, Aaron SD, Barnes NC, Davis KJ, Lacasse Y, Nadeau G. Evaluating the risk of pneumonia with inhaled corticosteroids in COPD: Retrospective database studies have their limitations. Respir Med 2017;123:94-97. https://doi. org/10.1016/j.rmed.2016.12.01 67. Crim C, Dransfield MT, Bourbeau J, et al. Pneumonia risk with inhaled fluticasone furoate and vilanterol compared with vilanterol alone in patients with COPD. Ann Am Thorac Soc 2015;12(1):27-34. https://doi.org/10.1513/AnnalsATS.201409413OC 68. European Medicines Agency (EMA). EMA completes review of inhaled corticosteroids for chronic obstructive pulmonary disease. London: EMA, 2016. http://www.ema.europa.eu/docs/en_GB/document_library/Referrals_document/ Inhaled_corticosteroids_Article_31/European_Commission_final_decision/ WC500210489.pdf (accessed 2 March 2018)

Accepted 29 November 2017.

AJTCCM VOL. 24 NO. 1 2018

RESEARCH

A comparison of the functional parameters of operability in patients with post-inflammatory lung disease and those with lung cancer requiring lung resection M H Amirali, MD, MMed (Int), FCP (SA); E M Irusen, MB ChB, FCP (SA), FCCP, PhD; C F N Koegelenberg, MB ChB, MMed (Int), FCP (SA), FRCP (UK), Cert Pulm (SA), PhD Division of Pulmonology, Department of Medicine, Stellenbosch University and Tygerberg Academic Hospital, Cape Town, South Africa Corresponding author: M H Amirali (mazheramirali@gmail.com)

Background. It is a common, yet unproven, belief that patients with post-inflammatory lung disease have a better functional reserve than patients with lung cancer when compared with their respective functional parameters of operability â&#x20AC;&#x201C; forced expiratory volume in one second (FEV1), maximum oxygen uptake in litres per minute (VO2max) and the diffusion capacity for carbon monoxide (DLCO). Objectives. The aim of this study was to compare a group of patients with lung cancer with a group with post-inflammatory lung disease according to their respective functional parameters of operability. We also aimed to investigate any associations of FEV1 and/or DLCO with VO2max within the two groups. Methods. We retrospectively included 100 adult patients considered for lung resection. All patients were worked up using a validated algorithm and were then sub-analysed according to their parameters of functional operability. Results. Two-thirds of patients had post-inflammatory lung diseases whilst the rest had lung cancer. The majority of the patients in the lung cancer group had coexistent chronic obstructive pulmonary disease (COPD) (n=18). Most (n=47) of the patients in the post-inflammatory group were diagnosed with a form of pulmonary TB (active or previous). Among the two groups, the lung cancer group had a higher median %FEV1 value (62.0%; interquartile range (IQR) 51.0 - 76.0) compared with the post-inflammatory group (52%; IQR 42.0 - 63.0; p=0.01). There was no difference for the %DLCO and %VO2max values. The lung cancer group also had higher predicted postoperative (ppo) values for %FEV1 (41.0%; IQR 31.0 - 58.0 v. 34.0%; IQR 23.0 - 46.0; p=0.03, respectively) and %VO2max (58.0%; IQR 44.0 - 68.0 v. 46.0%; IQR 35.0 - 60.0; p=0.02). There was no difference in the %DLCO ppo values between the groups. Conclusion. Patients with lung cancer had higher percentage values for FEV1 and ppo parameters for %FEV1 and %VO2max compared with those who had post-inflammatory lung disease. Our findings suggest that lung cancer patients have a better functional reserve. Afr J Thoracic Crit Care Med 2018;24(1):26-29. DOI:10.7196/AJTCCM2018.v24i1.158

Cancer is one of the leading causes of mortality worldwide. Lung cancer is the leading cause of cancer-related mortality globally, causing 1.6 million deaths in 2012.[1] However, in southern Africa, the relationship between lung cancer and its mortality rate remains low in comparison with other cancers and respiratory diseases.[2-5] According to the World Health Organization (WHO), an estimated 7.7 million cases of pulmonary tuberculosis (PTB) occurred worldwide in 2007[6] and South Africa (SA) had the third highest tuberculosis (TB) burden.[7,8] Treated PTB can lead to complications, including progressive loss of lung function, persistent pulmonary symptoms[9] and chronic pulmonary aspergillosis.[10-12] These complications frequently necessitate surgery. A study by Rizzi et al.[13] reported that patients with post tuberculous chronic haemoptysis (10.0%), lung destruction (8.1%), chest wall involvement (1.9%), suspected cancer (24.2%), cavitatory lung disease (21.9%) and bronchiectasis (16.1%) required elective surgery, whereas those with massive bleeding (5.4%) or a bronchopleural fistula (3.1%) required emergency surgery. Lung resection can be a high-risk procedure, especially in patients with underlying cardiopulmonary disease. Predictors of mortality

26 AJTCCM VOL. 24 NO. 1 2018

include the extent of resection, comorbidities and cardiopulmonary reserve.[14,15] Ninety percent of lung cancer patients are current or past smokers, which is frequently associated with varying degrees of concomitant chronic obstructive pulmonary disease and/or ischaemic heart disease. Furthermore, many of these patients are of advanced age and this places them at an increased risk of post-operative complications and mortality.[16,17] A number of prospective studies have validated a percentage-predicted forced expiratory volume in one second predicted postoperative value (%FEV1 ppo) of <40% as a prohibitive threshold for pulmonary resection, with mortality rates as high as 50% in such patients. Ferguson et al.[18] demonstrated that a diffusion capacity for carbon monoxide (DLCO) of <60% of the predicted value was a cut-off value for major pulmonary resection. The maximum oxygen uptake in litres per minute predicted postoperative (VO2 max ppo) value of <10 ml/kg/min, obtained from either formal cardiopulmonary exercise testing (CPET) or low-technology (minimal achievement) exercise tests, is associated with a high risk of post-operative complications and death. Regarding the cardiac

RESEARCH risk assessment, the Revised Cardiac Risk Index (RCRI)[19] is used by many authorities. The criteria contain six independent variables that correlate with post-operative cardiac complications - these include a high-risk type of surgery, a history of ischaemic heart disease, cardiac failure, cerebrovascular disease, diabetes requiring treatment with insulin and pre-operative serum creatinine of >177 µmol/L. Patients with more than two variables have a postoperative cardiac complication rate >10% and are considered to be at high risk.[17] The validated algorithms used to assess candidates for lung resection are based on spirometry, the DLCO and the VO2 max.[14] One such algorithm proposed by Bolliger and Perruchoud[15] has been used widely as a tool for evaluating cardiorespiratory reserves of lung resection candidates. The algorithm proposes that patients undergo successive steps of functional testing, the results of which qualify them for varying extents of resection or alternatively preclude them from any surgery.[15] Apart from the underlying cardiopulmonary disease and other comorbidities, the calculated predicted postoperative (ppo) values for FEV1, VO2max and DLCO are directly proportional to postoperative functional state and mortality.[21] It is a commonly held belief by various experts in the field of pulmonology that patients with post-inflammatory lung disease have a better functional reserve postoperatively than patients with lung cancer, when comparing their respective FEV1, VO2max and DLCO values; however, there is limited evidence to support the belief.[16] The aim of the present study was to compare two groups of patients (i.e. patients with lung cancer v. patients with post-inflammatory lung disease), and to investigate the association of functional parameters of operability within these two groups of patients.

Methods

Study design and population

We retrospectively enrolled adult patients who had been considered for lung resection and were referred to the Division of Pulmonology at Tygerberg Academic Hospital, Cape Town, with either lung cancer or post-inflammatory lung disease. Ethical approval for this retrospective analysis was obtained from the Stellenbosch University Research Ethics Committee (ref. no. S15/04/074). The application included a waiver of consent due to the retrospective nature and anonymity of the study design. Cases were identified from existing medical records; they were stratified into two groups, namely ‘A’ and ‘B’, where ‘A’ comprised patients with non-small-cell lung cancer while ‘B’ comprised patients with post-inflammatory lung disease (bronchiectasis, active/post tuberculous haemoptysis, and aspergilloma). After obtaining permission from the chief medical superintendent, the original medical records of all cases identified were requested and data were collected anonymously. The data collected included the demographics (age, gender), comorbidities of patients, indications for lung resection, extent of lung resection, and their pulmonary function test values (i.e. FEV1, FVC, DLCO and VO2max). The ppo value for these parameters can be calculated by the equation in Fig. 2, where the pulmonary function test (PFT) can either be %FEV1, %VO2max or %DLCO. We used three validated ways of estimating the relative

Diagnosis • Stress ECG • Echo • Perfusion scan • Angiogram

Positive Negative

Positive Treatment • Medical • Surgical

Heart • History • ECH

Negative

Yes

Lungs • FEV1 • DLCO

Both >80%

Either one <80%

40 - 75% and 10 - 20 mL.kg–1.min–1

Exercise testing VO2max

>75% or >20 mL.kg–1.min–1

<40% or <10 mL.kg–1.min–1 Both <40%

Split function • FEV1, ppo • DLCO, ppo Either one >40%

<35% or <10 mL.kg–1.min–1

Split function VO2max, ppo >35% and >10 mL.kg–1.min–1

High risk

Resection up to calculated extent

Pneumonectomy

Fig. 1. Algorithm proposed by Bolliger et al.,[15] adapted by Koegelenberg et al. [17] (ECG = electrocardiogram ; FEV 1 = forced expiratory volume in one second ; DLCO = diffusion capacity for carbon monoxide; VO2max = maximum oxygen uptake in litres per minute; mL = millilitres; kg = kilograms; )

%PFT ppo = [%PFT – ((a/n) × %PFT)] × 100 where PFT = pulmonary function test a = number of segments to be resected n = total number of segments Fig. 2. Equation used to calculate %PFT ppo value. (ppo = predicted postoperative, PFT = pulmonary function test.) functional contribution or split function, i.e. anatomical calculation, split radionucleotide perfusion scanning and quantitative computer tomography scanning and dynamic perfusion magnetic resonance imaging (MRI). Anatomical calculations of ppo values were performed on all patients who required pre-operative estimation of post-operative lung function. Patients who required further evaluation underwent either radionucleotide perfusion scanning or quantitative CT scanning. All patients were worked up for lung resection using the algorithm for the assessment of their cardiorespiratory reserves (functional operability).[17] Patients were generally followed up as outpatients and CPET was only performed once the risk of haemoptysis was

AJTCCM VOL. 24 NO. 1 2018

RESEARCH Table 1. Demographic and clinical data of study population (N=100) n (%)* Male 66 (66.0) Female 34 (34.0) Age (years), mean (range) 46.7 (17 - 72) Medical condition Lung cancer Male 15 (62.5) Female 9 (37.5) Comorbidities Hypertension 8 (19.0) HIV 0 (0.0) Pulmonary TB 1 (2.4) COPD 18 (42.9) Smoking 11 (26.2) CAD 2 (4.8) None 2 (4.8) Post-inflammatory Male 51 (67.1) Female 25 (32.9) Diagnoses Post-TB bronchiectasis 14 (19.7) Bronchiectasis 18 (25.3) Aspergillomata 18 (25.3) Destroyed lung 14 (19.7) Echinococcal cysts 3 (4.2) Empyema 1 (1.4) Adenomatoid malformation 1 (1.4) Post-TB upper-lobe changes 1 (1.4) MDR-TB 1 (1.4) Comorbidities Hypertension 6 (4.30) HIV 12 (8.70) Pulmonary TB (active and previous) 47 (34.0) COPD 30 (21.7) Smoking 23 (16.7) CAD 2 (1.4) Bronchiectasis 1 (0.7) None 17 (12.3) TB = tuberculosis; COPD = chronic obstructive pulmonary disease; CAD = coronary artery disease; MDR-TB = multidrug-resistant tuberculosis. *Unless otherwise specified.

evaluated (i. e. no haemoptysis for 2 weeks). Patients included in the study were then evaluated for their respective functional operability parameters.

Statistical analysis

χ2 comparisons and Pearson product-moment correlation coefficient (Pearson’s r or ‘r-squared’) of proportional data were performed. We did not make any assumptions for normality; hence, these nonparametric inferences were used for statistical analysis. A p-value <0.05 in a two-tailed test of proportions (χ2) was considered statistically significant. Unless stated otherwise, data are displayed as median with interquartile range (IQR) values.

Results

We included 100 patients in our study. The demographic data, primary diagnoses and comorbidities of the patients are summarised in Table 1.The majority of our patients were male (n=66/100); 51 were diagnosed with a post-inflammatory lung disease, while the rest had lung cancer. The most common diagnosis in the post-inflammatory group was that of haemoptysis (n=47). Bronchiectasis and aspergilloma were the second most common diagnoses, followed by post-TB bronchiectasis and destroyed lung. The majority of the patients in the lung cancer group had COPD (n=18), 11 of them were either active or previous smokers. Two of the patients had ischaemic heart disease. Most (n=47) of the patients in the post inflammatory group were diagnosed with some form of pulmonary TB (active or previous). COPD and smoking had the second and third highest prevalence, and 17 patients had no associated comorbidities. When comparing the various functional parameters of operability between the two groups, the lung cancer group had higher %FEV1 values (62.0%; IQR 51.0 - 76.0; p=0.01), there were no differences between the %DLCO (56.0%; IQR 44.0 - 75.0; p=0.509), and %VO 2max values (80.0%; IQR 66.0 - 89.0; p=0.105). The lung cancer group also had higher ppo values for %FEV1 (41.0%; IQR 31.0 - 58.0; p=0.03), and %VO2max (58.0%; IQR 44.0 - 68.0; p=0.02); there was ,however, no difference for %DLCO ppo values 40.0% (IQR 23.0 - 51.0; p=0.849). The values for the post-inflammatory group were: %FEV 1 52.0% (IQR 42.0 - 63.0); %DLCO 63.0% (IQR 51.0 - 75.0); and %VO2max 72.0% (IQR 59.0 - 82.0). The ppo values were: %FEV1 34.0% (IQR 23.0 - 46.0); %VO2max 46.0% (IQR 35.0 - 60.0); and %DLCO 39.0% (IQR 26.0 - 55.0). Correlation analysis did not show any correlation between the two groups.

Table 2. Comparison of functional parameters of operability among the two groups All, median (IQR) A,* median (IQR) 55 (43 - 65) 62 (51 - 76) %FEV1 35 (26 - 48) 41 (31 - 58) %FEV1 ppo 73 (60 - 84) 80 (66 - 89) %VO2max 49 (38 - 63) 58 (44 - 68) %VO2max ppo %DLCO 62 (50 - 75) 56 (44 - 75) %DLCO ppo 40 (26 - 54) 40 (23 - 51)

B,† median (IQR) 52 (42 - 63) 34 (23 - 46) 72 (59 - 82) 46 (35 - 60) 63 (51 - 75) 39 (26 - 55)

p-value 0.01 0.03 0.105 0.02 0.509 0.849

IQR = interquartile range; %FEV1 = percentage predicted for forced expiratory volume in one second; %FEV1 ppo = percentage predicted for forced expiratory volume in one second predicted postoperative; %VO2max = percentage predicted for maximum oxygen uptake in litres per minute; %VO2max ppo = percentage predicted for maximum oxygen uptake in litres per minute predicted postoperative; %DLCO = percentage predicted for diffusion capacity for carbon monoxide; %DLCO ppo = percentage predicted for diffusion capacity for carbon monoxide predicted postoperative. *Non-small-cell lung cancer group. † Post-inflammatory group (bronchiectasis, post tuberculous haemoptysis, aspergilloma).

28 AJTCCM VOL. 24 NO. 1 2018

RESEARCH

Discussion

We found statistically significant differences between the two groups when comparing the %FEV1, %FEV1 ppo, and %VO2max ppo; the lung cancer group had a higher %FEV1 (p=0.01), and higher ppo values for %FEV1 and %VO2max (p=0.03 and p=0.02, respectively). We found no statistically significant differences between the two groups when we compared the %DLCO, %DLCO ppo and %VO2max. No genderbased differences were observed. There was no correlation between the variables in either group. Therefore, both FEV1 and DLCO did not predict VO2max in either group. It is well-known that the pre-operative assessment predicts postoperative functional reserve, morbidity and mortality. Usually, a FEV1 ppo, DLCO ppo, and VO2max ppo <40% of normal values have all been found to indicate increased mortality.[22] We have shown that patients with lung cancer have a better functional reserve when compared with those who have post-inflammatory lung disease, and that neither FEV1 nor DLCO predicted VO2max in either group. There was also no predilection of the functional reserve towards the sex or age of our patients. We believe that these findings will have implications for the surgical management of patients with lung cancer, in that they may now be more readily considered for lung resection. Depending on the extent and the time elapsed from the operation, lung resections determine a variable reduction in functional reserve. A study by Brunelli et al.[23] showed that at one month after lobectomy, the FEV1, DLCO, and VO2max values were 79.5%, 81.5%, and 96% of preoperative values, respectively. These recovered to 84%, 88.5% and 97%, respectively, after 3 months. Regarding pneumonectomy, the %FEV1, %DLCO, and VO2max values were 65%, 75%, and 87% of preoperative values at 1 month, respectively; at 3 months postoperatively, the values were 66%, 80%, and 89%, respectively. Other studies have shown similar results.[24-26] Inferring from these data, the lung cancer group in our study would most likely have a better overall functional reserve postoperatively. Therefore, the assumption that lung cancer patients have a worse functional reserve postoperatively when compared with patients who have post-inflammatory lung disease is untrue.

Study strengths and limitations

This was a single-centre study, which benefits from strict adherence to a validated algorithm. The retrospective nature of the study, as well as the potential selection bias, could be limiting as only patients who were deemed clinically fit were recruited as study participants. We did not collect data on postoperative complications and mortality.

Conclusion

We found that patients with lung cancer had higher percentagepredicted values for FEV1 and predicted postoperative values for %FEV1 and %VO2 compared with those who had post-inflammatory lung disease. Future prospective studies should preferably include the postoperative outcomes among the two groups to provide a comprehensive analysis. Acknowledgements. We would like to thank all members of the pulmonary function laboratory team of Tygerberg Academic Hospital for their assistance and Mr Maxwell Chirehwa and Ms Tonya Esterhuizen for help with the statistical analysis.

Author contributions. MHA was the principal investigator, who collected the data and wrote the manuscript. CFNK assisted with data analysis and reviewed the manuscript. EMI reviewed the final manuscript. Funding. None. Conflicts of interest. None. 1. World Health Organization. The 10 leading Causes of Death by Broad Income Group. Geneva: WHO, 2011. 2. Steen TW, Aruwa JE, Hone NM. The epidemiology of adult lung disease in Botswana. Int J Tuberc Lung Dis 2001;5(5):775-782. 3. Groenewald P, Vos T, Norman R, et al. Estimating the burden of disease attributable to smoking in South Africa in 2000. S Afr Med J 2007;97(8 Pt 2):674-681. https:// doi:10.7196/SAMJ.661 4. Sitas F, Urban M, Bradshaw D, et al. Tobacco attributable deaths in South Africa. Tob Control 2004;13(4):396-399. https:// 10.1136/tc.2004.007682 5. Willcox PA, Oâ&#x20AC;&#x2122;Brien JA, Abratt RP. Lung cancer at Groote Schuur Hospital â&#x20AC;&#x201C; a local perspective. S Afr Med J 1990;78(12):716-720. https://doi.org/10.1016/01695002(91)90384-I 6. United Nations. World Population Prospects. The 2008 Revision. New York: UN, 2009. 7. World Health Organization. Global Tuberculosis Control. A Short Update to the 2009 Report. Geneva: WHO, 2009. 8. World Health Organization. Global Tuberculosis Control 2009. Epidemiology, Strategy, Financing. Geneva: WHO, 2009. 9. J Ross, R I Ehrlich, E Hnizdo, N White, G J Churchyard. Excess lung function decline in gold miners following pulmonary tuberculosis. Thorax 2010;65(11):1010-1015. https://doi.org/10.1136/thx.2009.129999 10. Denning DW. Chronic aspergillosis. Washington: ASM Press, 2009. 11. Jewkes J, Kay PH, Paneth M, Citron KM. Pulmonary aspergilloma: Analysis of cavitating invasive pulmonary aspergillosis in immunocompromised patients. Thorax 1983;38(8):572-578. https://doi.org/10.1136/thx.38.8.572 12. Nam HS, Jeon K, Um SW, et al. Clinical characteristics and treatment outcomes of chronic necrotizing pulmonary aspergillosis: A review of 43 cases. Int J Infect Dis 2010;14(6):e479-e482. https://doi.org/10.1016/j.ijid.2009.07.011 13. Rizzi A, Rocco G, Massera F. Results of surgical management of tuberculosis: Experience in 206 patients undergoing operation. Ann Thorac Surg 1995; 59(4):896900. https://doi.org/10.1016/0003-4975(95)00011-9 14. Koegelenberg CFN, Diacon AH, Irani S, Bolliger CT. Stair climbing in the functional assessment of lung resection candidates. Respiration 2008;75(4):374-379. https://doi. org/10.1159/000116873 15. Bolliger CT, Perruchoud AP. Functional evaluation of the lung resection candidate. Eur Respir J 1998;11(1):198-212. https://doi.org/10.1183/09031936.98.11010198 16. Bello B, Fadahun O, Kielkowski K, Nelson G. Trends in lung cancer mortality in South Africa: 1995 - 2006. BMC Public Health 2011;11(1):209. https://doi:10.1186/14712458-11-209. 17. Koegelenberg CFN, Plekker D, Bolliger CT. Functional evaluation for treatment. Eur Respir Monogr 2009;44:169-186. https://doi.org/10.1183/1025448x.00044010 18. Ferguson MK, Little L, Rizzo L, et al. Diffusing capacity predicts morbidity and mortality after pulmonary resection. J Thorac Cardiovasc Surg 1988;96(4):894-900 19. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100(10):1043-1049. https://doi.org/10.1161/01.cir.100.10.1043. 20. Bolliger CT, Koegelenberg CFN, Kendal R. Preoperative assessment for lung cancer surgery. Curr Opin Pulm Med 2005;11(4):301-306. https://doi.org/10.1097/01. mcp.0000166588.01256.9c 21. Bolliger CT, Wyser C, Roser H, Solar M, Perruchoud AP. Lung scanning and exercise testing for the prediction of postoperative performance in lung resection candidates at increased risk for complications. Chest 1995;108(2):341-348. https://doi.org/10.1378/ chest.108.2.341 22. Algar FJ, Antonio A, Salvatierra A, et al. Predicting pulmonary complications after pneumonectomy for lung cancer. Eur J Cardiothorac Surg 2003;23(2):201-208. https:// doi.org/10.1016/s1010-7940(02)00719-4 23. Brunelli A, Xiume F, Refai M, et al. Evaluation of expiratory volume, diffusion capacity, and exercise tolerance following major lung resection. A prospective followup analysis. Chest 2007;131(1):141-147. https://doi.org/10.1378/chest.06-1345 24. Bolliger CT, Jordan P, Soler M, et al. Pulmonary function and exercise capacity after lung resection. Eur Respir J 1996;9(3):415-421. https://doi.org/10.1183/09031936.9 6.09030415 25. Nezu K, Kushibe K, Tojo T, Takahama M, Kitamura S. Recovery and limitation of exercise capacity after lung resection for lung cancer. Chest 1998;113(6):1511-1516. https://doi.org/10.1378/chest.113.6.1511 26. Bolliger CT, Guckel C, Engel H, et al. Prediction of functional reserves after lung resection: Comparison between quantitative computed tomography, scintigraphy, and anatomy. Respiration 2002;69(6):482-489. https://doi.org/10.1159/000066474 Accepted 10 October 2017.

AJTCCM VOL. 24 NO. 1 2018

RESEARCH

The epidemiology of respiratory syncytial virus: A retrospective review from Steve Biko Academic Hospital 2013 - 2016 C X Dearden,1 MB ChB, FC Paed, MMed (Paed), Dip Allerg; A C Jeevarathnum,1 MB BCh, FC Paed, MMed (Paed), Dip Allerg, Cert Pulm Paed; J Havinga,2 Dip Medical Technology; R J Green,1 PhD, DSc 1 2

Department of Paediatrics and Child Health, Faculty of Health Sciences, University of Pretoria and Steve Biko Academic Hospital, Pretoria, South Africa Department of Medical Virology, Faculty of Health Sciences, University of Pretoria and Steve Biko Academic Hospital, Pretoria, South Africa

Corresponding author: C X Dearden (xandrevdh@gmail.com)

Background. Respiratory syncytial virus (RSV) bronchiolitis is a seasonal disease that has an enormous burden on health systems across the world. RSV disease manifestations in children range from mild upper respiratory tract infections to severe lower respiratory tract infections, including pneumonia and bronchiolitis. In South Africa, the seasonality of RSV disease causing both upper and lower respiratory tract illness is well documented. Objectives. To describe the incidence of RSV bronchiolitis among patients ≤24 months of age who presented to a tertiary institution with a diagnosed viral bronchiolitis over a 4-year period. Secondary aims included determining: (i) the risk factors for the development of RSV bronchiolitis; (ii) the fatality rates and risk factors associated with mortality; (iii) the correlation with c-reactive protein values and risk of comorbid bacterial infection; and (iv) the impact of seasonality on RSV incidence. Methods. A retrospective chart-based analysis of laboratory-confirmed RSV cases in children ≤24 months, presenting to Steve Biko Academic Hospital from January 2013 to December 2016, was undertaken. Epidemiology, risk factors and local weather data were collected as part of the analysis. Results. During the 4-year period, a total of 1 127 nasopharyngeal aspirates (NPAs) was collected. RSV was isolated from 162 NPAs by either immunofluorescence (84%) or polymerase chain reaction (16%). Of the 162 patients with RSV bronchiolitis, 131 (80.9%) had a known HIV status. Only 2 (1.5%) of the patients whose status was known were HIV-infected; 26 (19.8%) were HIV-exposed and confirmed negative; and 103 (78.6%) HIV-unexposed. Forty-nine patients (30.2%) with RSV required intensive care unit (ICU, either paediatric or neonatal) admission. Thirty-four (69.4%) of these were <6 months old. Prematurity (27.8%) and cardiac lesions (13%) were the most common risk factors for acquiring the disease identified in patients with RSV bronchiolitis. Conclusions. RSV is still a commonly detected virus among infants who are admitted for bronchiolitis. Significant risk factors associated with admission due to RSV bronchiolitis were prematurity, being <6 months of age and congenital cardiac disease. Male gender and HIV status did not appear to increase the risk of RSV bronchiolitis. In fact, HIV seems to have a protective effect against specifically RSV bronchiolitis in children <2 years of age. Young babies, especially premature infants with RSV bronchiolitis, are at considerable risk of requiring ICU admission, which leads to a significant increase in admission costs. Afr J Thoracic Crit Care Med 2017;24(1):30-35. DOI:10.7196/AJTCCM.2017.v24i1.163

Bronchiolitis is a clinically diagnosed lower respiratory tract viral infection characterised by wheezing and tachypnoea. The highest incidence is among children <2 years old. The pathophysiology of bronchiolitis is that of acute inflammation, oedema and necrosis of the small airway epithelial cells, increased mucus production and bronchospasm.[1] Bronchiolitis can be caused by a host of viruses, all leading to a similar clinical syndrome. The viruses commonly isolated include respiratory syncytial virus (RSV), rhinovirus, influenza, parainfluenza, adenovirus, human metapneumovirus, coronavirus and bocavirus.[2] High-risk children with RSV-associated lower respiratory tract infections (LRTIs) are more likely to be admitted to an intensive care unit (ICU) and have a longer hospital stay than otherwise-healthy children.[3] HIV-infected children are more likely to be diagnosed with pneumonia than with bronchiolitis (p<0.01).[4] In a South African (SA)

30 AJTCCM VOL. 24 NO. 1 2018

study reported in 2012 that considered 105 hospitalised children <2 years of age with LRTIs, RSV was not identified in any HIV-infected cases (n=15) compared with 30.6% of HIV-uninfected cases (n=85; p=0.013), and was identified more frequently in bronchiolitis than in pneumonia cases (43.8% v. 12.3%; p<0.01). This might indicate that HIV infection is protective against RSV and bronchiolitis.[4] Prematurity, low birth weight, being male, maternal smoking, having siblings, a history of atopy, a lack of breastfeeding and household overcrowding (>7 persons) have been observed to be significantly associated with RSV-associated acute LRTIs.[5] In Pretoria, SA, the RSV season peaks in autumn (April - May).[3] RSV follows a temporal trend, while other viruses are more equally distributed over the year.[6] The role of the environment in the spread of respiratory infections is poorly understood. An environmental influence on RSV transmission is required to maintain this seasonality, and to dictate the timing of seasonal epidemics.[7]

RESEARCH Bronchiolitis is a viral disease. Bacterial coinfection is rare in true viral bronchiolitis.[11] Blood tests are not needed routinely.[8] In a study by Korppi,[9] using a c-reactive protein (CRP) value of 40 mg/L as a screening limit seemed to be the most reliable method in differentiating between bacterial and viral respiratory infection. The routine use of antibiotics for mildly and moderately ill children with bronchiolitis is discouraged, because significant bacterial coinfection is rare.[2] The long-term outcome of patients who have had RSV bronchiolitis is currently a topic of much debate. There is evidence that RSV bronchiolitis may predispose patients to recurrent episodes of wheezing, and possibly even asthma.[8] The estimated global case fatality rate among children <1 year of age with severe RSV acute respiratory infection is 6.6 %.[5]

Objective

The aim of this study was to describe the incidence of RSV bronchiolitis among patients ≤24 months of age who presented to a tertiary institution with a diagnosed viral bronchiolitis over a 4-year period. Secondary aims included determining: (i) the risk factors for development of RSV bronchiolitis; (ii) the fatality rates and risk factors associated with mortality; (iii) the correlation with CRP values and risk of comorbid bacterial infection; and (iv) the impact of seasonality on RSV incidence.

Methods

This was a retrospective study of all children ≤24 months old who were seen in an outpatient department, or admitted to the Steve Biko Academic Hospital (SBAH), with proven viral bronchiolitis from January 2013 to December 2016. At the SBAH there is a well-established guideline on how to diagnose and investigate children with bronchiolitis. Every child with a clinical diagnosis of bronchiolitis has a nasopharyngeal aspirate (NPA) collected and sent for investigation. This guideline remained unchanged for the duration of the study. Demographics (age, sex), known risk factors, HIV status, laboratory data (CRP, procalcitonin, full blood count and differential, blood culture), viral coinfection and length of stay in neonatal or paediatric ICU (PICU) and/or the paediatric wards were recorded. All positive RSV NPAs were obtained from the Department of Medical Virology at SBAH. Files were then traced through the records department to gather the study information retrospectively. CRP and PCT cut-offs of 40 mg/L and 1 ng/mL were used, respectively.[9] Either immunofluorescence or polymerase chain reaction (PCR) testing was conducted on NPA samples. HIV ELISA or PCR results were obtained from either the patients’ files or from the national laboratory website, Labtrak. Local weather data were obtained from the weather bureau to identify any possible association between RSV incidence and weather patterns (humidity, rainfall and temperature). Children from both surgical and medical wards were included in this study. Statistical analysis was done by means of descriptive statistics utilising Stata version 15 (StataCorp, USA) software. Ethics approval to conduct the current study was obtained from the University of Pretoria’s Ethics Committee (ref. no. 78/2017), as was consent from SBAH.

Results

Over the 4 years studied (2013 - 2016), a total of 1 127 NPAs were conducted in children ≤24 months. A total of 288 viruses were isolated from 271 positive NPAs – in 17 NPAs, more than one virus was isolated. The most commonly identified viruses were RSV, adenovirus, parainfluenza 1, 2, 3 and 4, human metapneumovirus, influenza, rhinovirus and bocavirus. RSV was by far the most common, isolated in 162 (14.4%) children with bronchiolitis, followed by adenovirus (4.1%) and parainfluenza 3 virus (2.1%). Table 1 depicts the distribution of viruses isolated over the 4-year period, while Table 2 demonstrates the distribution of RSV bronchiolitis over 4 years. Multiplex PCR was only used frequently in the latter 2 years of the study (2015 and 2016; 3 uses of PCR were outsourced to private practice in 2013). Of the 162 RSV-confirmed bronchiolitis NPAs undertaken, 136 (84%) were conducted by immunofluorescence and 26 (16%) by PCR. As seen in Fig. 1, the total number of RSV isolates in 2015 did not increase, but rather decreased compared with previous years, despite the introduction of PCR testing. Of the RSV cases there were 82 male and 80 female patients, giving a male-to-female ratio of 1.03:1.00. The median age was 3.7 months (range 9 days - 2 years), with 43.8% being <3 months and 63.4% <6 months (Fig. 2). A total of 131 (80.9%) patients had a known HIV status. Only 2 (1.5%) of those whose status was known were HIV infected, 26 (19.8%) HIV exposed and confirmed negative and 103 (78.6%) HIV unexposed. Forty-nine (30.2%) of the total number of RSV-confirmed bronchiolitis patients required PICU admission. There were 34 (69.4%) <6 months old (Fig. 3). There was a linear increase in the percentage of patients needing PICU every year, from 19.6% in 2013 to 42.9% in 2016 (Fig. 4). There was no change in PICU bed availability during this time. Table 1. Distribution of viruses isolated over the study period Virus Positive results, n NPAs performed, % RSV 162 14.4 Adenovirus 46 4.1 Parainfluenza 3 30 2.7 HMPV 16 1.4 Influenza 13 1.2 Parainfluenza 1 7 0.6 Parainfluenza 4 5 0.4 Rhinovirus 5 0.4 Parainfluenza 2 3 0.3 Bocavirus 1 0.1 Total 288 100 HMPV = Human metapneumovirus; RSV = respiratory syncytial virus.

Table 2. RSV isolates per year Year NPAs, n 2013 348 2014 275 2015 222 2016 282 Total 1127

RSV, % 51, 14.7 42, 15.3 20, 9.0 49, 17.3 162, 14.3

RSV = respiratory syncytial virus; NPAs = nasopharyngeal aspirates.

AJTCCM VOL. 24 NO. 1 2018

RESEARCH 60 50

40 30 20

0 15 48

10 8

10 0 2013

2014 IF

2015

2016

PCR

Fig. 1. Immunofluorescence (IF) and polymerase chain reaction (PCR) use in detecting respiratory syncytial virus bronchiolitis, 2013 - 2016.

>12 months ≤24 months 19%

30 20

≤28 days 11%

>6 months ≤12 months 18%

>28 days ≤3 months 32%

RSV in ICU, % 2013

2014

2015

2016

Fig. 4. Percentage of patients with respiratory syncytial virus (RSV) admitted to intensive care unit (ICU) each year. The time patients with RSV bronchiolitis spent in PICU ranged from 1 - 32 days, with an average of 9.5 days per patient. Patients who spent time in the paediatric wards stayed for an average of 6.7 days. Of the 162 patients, there were 8 deaths (4.9%). Seven of these patients were <6 months old. Six of the patients who died were HIVnegative, while the other 2 patients had an unknown HIV status. Of the 8 deaths, 2 of the patients had both Down’s syndrome and an atrioventricular septal defect (AVSD), 1 only had an AVSD, 2 were premature and 1 patient had holoprosencephaly.

Risk profiles >3 months ≤6 months 20%

Fig. 2. Age distribution of respiratory syncytial virus-confirmed bronchiolitis patients.

Within the group of 162 patients with RSV bronchiolitis, prematurity, followed by cardiac lesions, were the most common risk factors identified (Table 3). Chronic lung disease included bronchopulmonary dysplasia and bronchiolitis obliterance in this study. Of the 49 patients who required PICU admission, 18 (36.7%) were premature babies (Table 4). HIV did not appear to be a risk factor to contracting RSV disease as only 2 (1.5%) of the 131 patients with known status were HIV-positive.

Laboratory data

Five RSV bronchiolitis patients had positive blood cultures for potential pathogens: Two Escherichia coli, one Candida lypolytica, one Staphylococcus hominis and one Enterococcus faecalis. Three of these five patients had a positive CRP. Three were in PICU. Two patients had additional risk factors, including gastroschisis and Hirschsprung disease. Four of the five patients were <3 months old. A total of 64 patients had negative blood cultures, but only 61 of these also had CRPs conducted. Twelve (19.7%) of the 61 patients had a positive CRP (>40 mg/L). Of these 12 patients with positive CRPs but negative blood cultures, 7 were ventilated and had additional risk factors. Of the 64 patients with negative blood cultures, 27 had a PCT test done. Ten of the 27 PCTs were positive (≥1 ng/mL). Of these 10 patients with positive PCTs, 8 were ventilated and 6 had additional risk factors. 6 - 24 months

<6 months

Fig. 3. Age distribution of respiratory syncytial virus patients admitted to intensive care unit.

32 AJTCCM VOL. 24 NO. 1 2018

Weather influence

As shown in Fig. 5, RSV bronchiolitis at SBAH revealed an autumn and winter predominance, with an increased incidence

80 60 40 20 0 0 Ja 13 n2 0 Ju 14 ly 20 Ja 14 n2 0 Ju 15 ly 20 Ja 15 n2 0 Ju 16 ly 20 16

Ju l

RSV = respiratory syncytial virus.

RSV

6 01

Jul

Jan

Jul

Jan

Jul

Jan

Minimum temperature

Jul

PICU = paediatric intensive care unit.

Humidity

Fig. 6. Influence of humidity on respiratory syncytial virus (RSV) incidence.

Patients, n (%) 18 (36.7) 7 (14.3) 5 (10.2) 2 (4.1)

Jan

in patients requiring PICU

RSV incidence

Table 4. Risk factors identified admission Risk factor Premature Cardiac disease Chronic lung disease Down’s syndrome

Humidity

100

20 15 10 5 0 13

Ja n

Table 3. Risk factors identified in RSV-infected patients Risk factors Patients, n (%) Premature 45 (27.8) Cardiac disease 21 (13.0) Chronic lung disease 12 (7.4) Down’s syndrome 8 (4.9) Malignancy 3 (1.9) Mother/caregiver smoking 3 (1.9) Sickle cell anaemia 1 (0.6)

RSV incidence

RESEARCH

Minimum temperature

Fig. 7. Influence of minimum temperature on respiratory syncytial virus (RSV) incidence.

Discussion

RSV was the most common virus detected during the study, and was found in 162 (14.4%) of the NPAs undertaken at SBAH during 2013 - 2016. RSV detection frequency was followed by adenovirus (4.1%) and parainfluenza 3 virus (2.1%). These results probably reflect

20 15 10 5

Jan

Jul

20 Jul 14 y2 0 Jan 14 20 Jul 15 y2 0 Jan 15 20 Jul 16 y2 01 6

in February - March 2013, March - April 2014, March - July 2015 and May - July 2016. Therefore the higher incidences appear to be slightly later in the year in each year of this study. In 2013, March had the most RSV NPAs isolated, while by 2016 the highest number of RSV isolates was identified in June. Figs 6 - 9 show the effects of humidity, rainfall and temperature on RSV seasonality. Across the 4 years, there was a weak negative correlation between RSV incidence and rainfall (r=–0.1), minimum temperature (r=–0.3) and maximum temperature (r=–0.5), and a weak positive correlation between RSV cases and humidity (r=0.3)

40 35 30 25 20 15 10 5 0

Fig. 5. Respiratory syncytial virus (RSV) incidence compared with total nasopharyngeal aspirates (NPAs).

Jan

RSV incidence

NPAs

Maximum temperature

n2 0 Ap 13 r2 0 Ju 1 ly 3 2 Oc 013 t2 Ja 013 n2 Ap 014 r2 Ju 014 ly 2 Oc 014 t2 Ja 014 n2 Ap 015 r2 Ju 015 ly 2 Oc 015 t2 Ja 015 n2 Ap 016 r2 Ju 016 ly 2 Oc 016 t2 01 6

RSV

Maximum temperature

Fig. 8. Influence of maximum temperature on respiratory syncytial virus (RSV) incidence. a true viral incidence, as NPAs are carried out on all children with clinically suspected bronchiolitis. In 2013 and 2014, virus detection was almost exclusively done by immunofluorescence. PCR testing commenced in 2015, and continued in 2016. It had been expected that detection

AJTCCM VOL. 24 NO. 1 2018

25 20 15 10 5 0

n Ja

300 200 100

Rainfall, mm

RSV incidence

RESEARCH

0 0 0 0 v 2 pt 2 y 2 ay 2 l o u J N M Se RSV

Rainfall

Fig. 9. Influence of rainfall on respiratory syncytial virus (RSV) incidence. (mm = millimetres.) rates would surge in 2015 and 2016, but in contrast, 2015 had markedly lower rates of RSV compared with other years, while 2016 was within the average detection rate. This suggests that the yearly changes in RSV detection cannot be attributed to the introduction of PCR testing. There were 82 male and 80 female patients, an almost equal male to female ratio of 1.03:1.00. Forty-five of these patients were premature or ex-premature babies, of whom the male: female ratio was 0.80:1.00. This finding does not support the belief that male gender is a risk factor for bronchiolitis. Forty-three percent of patients with RSV bronchiolitis were <3 months and 63.4% <6 months old. This corresponds with global data that suggest that infants <6 months old are at increased risk of contracting RSV bronchiolitis. Children within this age range should be targeted as part of prevention strategies. Of the 131 patients with known HIV status, only 2 (1.5%) were HIV infected. Examining the results from a study undertaken by Annamalay et al. RSV was not identified in any HIV-infected children <2 years old with bronchiolitis. One could even conclude from this that HIV protects against RSV and bronchiolitis. Thirty percent of RSV-confirmed bronchiolitis patients required PICU admission, of whom 69.4% were <6 months old. Young babies with RSV bronchiolitis are at considerable risk of requiring PICU admission, which leads to a significant increase in admission costs. Although the admission criteria and number of beds at Steve Biko Paediatric PICU has remained unchanged, there was a linear increase in the percentage of patients requiring PICU over the course of the study, from 19.6% in 2013 up to 42.9% in 2016. The case fatality rate of the 162 patients with RSV bronchiolitis was 4.9% (8 deaths). Seven of these patients were <6 months old. Of the 8 deaths, 6 patients had significant risk factors, re-emphasising the importance of the prevention of RSV in high-risk children. It is well documented in many studies that prematurity, cardiac lesions and chronic lung disease are risk factors for contracting severe RSV disease. This was once again reaffirmed in this study. A total of 27% of all RSV-infected patients were premature, whilst 36% of all patients admitted to ICU were premature. This highlights the need for better prevention of RSV disease in high-risk babies by means of immunoglobulins (palivuzimab) or vaccination.

34 AJTCCM VOL. 24 NO. 1 2018

It is well documented that routine CRP and blood cultures do not contribute to the routine management of bronchiolitis, nor assist with the need for antibiotics. This study was not designed to study this effect; however, in severely ill patients that require PICU admission, it might be beneficial to use a CRP value of >40 mg/mL or PCT >1 ng/ mL to help guide the need for antibiotic treatment. RSV bronchiolitis at SBAH revealed an autumn and winter predominance, with the peak incidence occurring later each year: in 2013, March had the most RSV NPAs isolated, and by 2016 the most RSV isolates were found in June. Across the 4 years, there were only weak correlations between RSV incidence and rainfall, minimum temperature, maximum temperature and humidity. An environmental influence on RSV transmission is needed to maintain its seasonality, but the exact mechanisms involved (whether host, pathogen, or environment) are still poorly understood. Some strengths of this study include the large number of patients’ files that was analysed. Good record keeping facilitated adequate interpretation of the data. The availability of weather-pattern records during the study period allowed correlations to be drawn with seasonality. Unfortunately, the study is limited by the disparity between the lab specimens sent, i.e. NPAs and immunofluorescences. The study would have been strengthened if there was uniformity in this aspect. Rhinovirus bronchiolitis is probably under-represented because of this disparity in testing.

Conclusion

It is beyond doubt that premature babies and infants <6 months old with RSV bronchiolitis are at increased risk for hospital and ICU admission. High risk of mortality and the severe cost implications in these patients necessitates the implementation of prevention strategies. With RSV vaccinations still in the trail phases, the only currently available prevention method is that of palivizumab, currently not available to the vast majority of high-risk patients who desperately need it. Even though palivizumab is not seen as a costeffective solution, its implementation will prevent the deaths of an uncountable number of precious infants. RSV has a seasonal pattern, but the mechanisms involved are not completely understood. Acknowledgments. The authors would like to thank Prof. Ameena Goga for assistance during the writing of this article. Author contributions. CXD: protocol, data collection, write-up; JH: supplied the raw data; ACJ and RJG: oversight, advice and editing. Funding. None. Conflicts of interest. None. 1. Ali S, Plint A, Klassen TP. Bronchiolitis. In: Wilmots RW, Boat TF, Bush A, editors. Kendig and Chernick’s Disorders of the Respiratory Tract in Children. 8th ed. Philadelphia: Elsevier Saunders, 2012. 2. Venter M, Lassaunière R, Kresfelder TL, Westerberg Y, Visser A. Contribution of common and recently described respiratory viruses to annual hospitalisations in children in South Africa. J Med Virol 2011;83(8):1458-1468. https://doi.org/10.1002/jmv.22120 3. Madhi SA, Venter M, Alexandra R, et al. Respiratory syncytial virus associated illness in high-risk children and national characterisation of the circulating virus genotype in South Africa. J Clin Virol 2003;27(2):180-189. https://doi.org/10.1016/s13866532(02)00174-9 4. Annamalay AA, Abbott S, Sikazwe C, et al. Respiratory viruses in young South African children with acute lower respiratory infections and interactions with HIV. J Clin Virol 2016;81(Aug.):58-63. https://doi.org/10.1016/j.jcv.2016.06.002

RESEARCH 5. Stein RT, Bont LJ, Zar H, et al. Respiratory syncytial virus hospitalisation and mortality: Systematic review and meta-analysis. Ped Pulmonol 2016;51(1):1-14. https://doi.org/10.1002/ppul.23570 6. Cangiano G, Nenna R, Frassanito A, et al. Bronchiolitis: Analysis of 10 consecutive epidemic seasons. Ped Pulmonol 2016;51(12):1330-1335. https://doi.org/10.1002/ ppul.23476 7. Paynter S, Sly PD, Ware RS, Williams G, Weinstein P. The importance of the local environment in the transmission of respiratory syncytial virus. Sci Tot Environ 2014;493(15):521-525. https://doi.org/10.1016/j.scitotenv.2014.06.021

8. White DA, Zar HJ, Madhi SA, et al. Acute viral bronchiolitis in South Africa: Diagnostic flow, management and prevention. S Afr Med J 2016;106(4):328-329. https://doi.org/10.7196/samj.2016.v106i4.10441 9. Korppi M, Kröger L. C-reactive protein in viral and bacterial respiratory infection in children. Scand J Infect Dis 1993;25(2):207-213. https://doi. org/10.3109/00365549309008486

Accepted 5 December 2017.

AJTCCM VOL. 24 NO. 1 2018

BREATH-TAKING NEWS

Statin use is associated with a lower risk of tuberculosis Statins are a widely used class of medication that reduce the risk of coronary artery disorders and hypercholesterolaemia by targeting 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase.[1] Studies have shown that statins lower the production of pro-inflammatory cytokines, reduce platelet aggregability, prevent coagulation and reduce injury caused by oxidative stress.[2] The relationship between statin use and development of infection has been investigated for decades. Several meta-analyses have found that the use of statins may improve clinical outcome in patients with sepsis and pneumonia. [3] According to the World Health Organization (WHO), tuberculosis (TB) is one of the most important infectious diseases and in 2015, 10.4 million individuals developed the disease, resulting in 1.8 million deaths worldwide. [4] It has been shown that statin-mediated reduction in cholesterol levels within intracellular phagosomal membranes augments host protection against tuberculosis. [5] Between 2000 and 2013, Su et al. [6] used data from the Taiwan National Health Insurance Research Database to conduct a retrospective cohort study of 102 424 patients who were taking statins, and 202 718 matched controls. The two cohorts were monitored for incident TB disease and the statin and matched cohorts were observed for 571 568 and 1Â 027 385 person-years, respectively. Of the 305 142 subjects, 126 (0.41%) developed TB disease. A multivariate analysis revealed a reduced risk of TB disease among the statin cohort (hazard ratio (HR) 0.53; 95% confidence interval (CI) 0.47 - 0.61; p<0.001). In addition, statin use showed a dose-response relationship with the incident TB disease risk (<180 cumulative defined daily doses (cDDDs): HR 1.06; 95% CI 0.91 - 1.24; p=0.477; 180 - 365 cDDDs: HR 0.57;

36 AJTCCM VOL. 24 NO. 1 2018

95% CI 0.45 - 0.72; p<0.001; and >365 cDDDs: HR 0.27; 95% CI 0.22 - 0.33; p<0.001). The study showed that statin use was associated with a lower risk of incident TB disease compared with non-users, with a dose-dependent benefit on the risk. Tarig H Ahmed Division of Pulmonology, Department of Medicine, Groote Schuur Hospital and University of Cape Town, South Africa

Afr J Thoracic Crit Care Med 2018;24(1):36. DOI:10.7196/AJTCCM.2018.v24i1.205

1. Ray KK, Rao S, Seshasai K, et al. Statins and all-cause mortality in high-risk primary prevention: A meta-analysis of 11 randomized controlled trials involving 65,229 participants. Arch Intern Med 2010;170(12):1024-1031. https://doi.org/10.1001/ archinternmed.2010.182 2. Jain MK, Ridker PM. Anti-inflammatory effects of statins: Clinical evidence and basic mechanisms. Nat Rev Drug Discov 2005;4(12):977-987. https://doi.org/10.1038/ nrd1901 3. Janda S, Young A, FitzGerald JM, Etminan M, Swiston J. The effect of statins on mortality from severe infections and sepsis: A systematic review and meta-analysis. J Crit Care 2010;25(4):656.e7-656.e22. https://doi.org/10.1016/j.jcrc.2010.02.013 4. World Health Organization (WHO). WHO Global Tuberculosis Report 2016. Geneva: WHO, 2016. 5. Parihar SP, Guler R, Khutlang R, et al. Statin therapy reduces the Mycobacterium tuberculosis burden in human macrophages and in mice by enhancing autophagy and phagosome maturation. J Infect Dis 2016;209(5):754-763. https://doi.org/10.1093/ infdis/jit550 6. Su VY, Su W, Yen Y, et al. Statin use is associated with a lower risk of tuberculosis. Chest 2017;152(3):598-606. https://doi.org/10.1016/j.chest.2017.04.170

BREATH-TAKING NEWS

The impact of statin drug use on all-cause mortality in patients with COPD Chronic obstructive pulmonary disease (COPD) affects 380 million people worldwide. Co-morbidities associated with COPD are an important aspect of the disease and cardiovascular disease, in particular, has been shown to be more than twofold more prevalent in patients with COPD relative to the general population.[1] Statin drugs have been shown to be effective in reducing all-cause mortality in patients with risk factors for cardiovascular disease.[2] In patients with COPD, observational evidence has shown that statin drug use may reduce the risk of acute exacerbations of COPD, and reduce both respiratory-related mortality and all-cause mortality.[3] The proposed mechanism for this protective effect is a reduction of underlying systemic inflammation, which is often present in COPD patients and is associated with increased mortality. The STATCOPE (Simvastatin Therapy for Moderate and Severe COPD) study found no significant differences in exacerbation rates or time to exacerbation between statin drug users and control subjects.[4] However, the results of this trial were controversial due to one in three patients being excluded (previous statin use and elevated glycated haemoglobin or cholesterol levels) and being underpowered to assess mortality in COPD. Raymakers et al.[5] have recently published a population-based study using administrative data from British Columbia, Canada, to evaluate the association between all-cause and lung-specific mortality and statin drug use in a cohort of patients with COPD. In this study, 39 678 patients with COPD met the inclusion criteria. Of them, 7 775 patients (19.6%) had received at least one statin drug that was dispensed during the exposure ascertainment window. The mean (standard deviation) age of the patients was 71.0 (11.6) years; 54.7% were women. There were 1 446 all-cause deaths recorded in the cohort in the 1-year period after exposure ascertainment. In a univariate analysis, the hazard ratio (HR) for statin drug use associated with all-cause mortality was 0.79 (95% confidence interval (CI) 0.69 - 0.91; p=0.0012), which suggests a protective effect. In a multivariate analysis, the estimated HR for statin drug exposure was

also 0.79 (95% CI 0.68- 0.92; p=0.0016), suggesting a 21% reduction in the risk of all-cause mortality. For lung-related mortality, there was also a considerable reduction in the risk for all-cause mortality owing to the use of statins (HR 0.55; 95% CI 0.32 - 0.93; p=0.025). These results were robust to different specifications of the exposure ascertainment window. This study suggests that statin drug use is associated with a significant reduction in both all-cause and lung-related mortality in patients with COPD. The strength of this study is its generalisability owing to the use of population-based administrative data from a large Canadian province. David Masuku Division of Pulmonology, Department of Medicine, Groote Schuur Hospital and University of Cape Town, South Africa

Afr J Thoracic Crit Care Med 2018;24(1):37. DOI:10.7196/AJTCCM.2018.v24i1.207 1. Chen W, Thomas J, Sadatsafavi M, FitzGerald JM. Risk of cardiovascular comorbidity in patients with chronic obstructive pulmonary disease: A systematic review and meta-analysis. Lancet Respir Med 2015;3(8):631-639. https://doi.org/10.1016/s22132600(15)00241-6 2. Brugts JJ, Yetgin T, Hoeks SE, et al. The benefits of statins in people without established cardiovascular disease but with cardiovascular risk factors: Meta-analysis of randomised controlled trials. BMJ 2009;338:b2376. https://doi.org/10.1136/bmj. b2376 3. Lawes CM, Thornley S, Young R, et al. Statin use in COPD patients is associated with a reduction in mortality: A national cohort study. Prim Care Respir J 2012;21(1):35-40. https://doi.org/10.4104/pcrj.2011.00095 4. Criner GJ, Connett JE, Aaron SD, et al. Simvastatin for the prevention of exacerbations in moderate-to-severe COPD. N Engl J Med 2014;370(23):2201-2210. https://doi. org/10.1056/nejmoa1403086 5. Raymakers AJN, Sadatsafavi M, Sin DD, De Vera, MA, Lynd LD. The impact of statin drug use on all-cause mortality in patients with COPD. A population-based cohort study. Chest 2017;152(3):486-493: https://doi.org/10.1016/j.chest.2017.02.002

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS

Abstracts of the combined SATS & PATS congress in Durban, 12-15 April 2018 ORAL PRESENTATIONS Sirtuin 1 gene rs2273773 C>T single nucleotide polymorphism and protein oxidation markers in asthmatic patients A Ali, A Mahmoud

Department of Respiratory Medicine, Faculty of Medicine, Sohag University, Egypt abdellahhamed@yahoo.com

Introduction. Sirtuin-1 (SIRT-1) is a protein that has been found to protect the cells against oxidative stress due to its deacetylase activity. Objective. In this investigation, we aimed to study SIRT-1 gene rs2273773 C>T single nucleotide polymorphism (SNP) and markers of serum protein oxidation (protein carbonyl and sulfhydryl groups) in asthmatic patients. Methods. A total of 120 asthmatic patients and 120 healthy controls were genotyped for SIRT- 1 gene rs2273773 C>T SNP using the polymerase chain reaction-confronting two-pair primer method (PCRCTPP). Serum protein carbonyl and sulfhydryl groups were measured using colorimetric methods. Results. SIRT-1 gene rs2273773 C>T SNP genotyping revealed that the TT genotype was significantly higher in the patients compared with the controls (p=0.05). The T allele was significantly higher in the patients compared with the controls (p=0.017). The distribution of the genotypes did not differ among the atopic and the non-atopic asthmatic patients, also no difference was found in the genotype distribution according to the severity of asthma (p>0.05). Serum protein carbonyl group concentration was significantly higher in the patients compared with the controls (p=0.001), while serum protein sulfhydryl group content decreased significantly in the patients compared with the controls (p< 0.0001). No differences in markers of protein oxidation according to SIRT-1 gene rs2273773 C>T genotype were found. Conclusion. In the Egyptian population, SIRT-1 gene rs2273773 C>T SNP was associated with asthma, but not with protein oxidation markers.

Introduction. Chronic obstructive pulmonary disease (COPD) is highly prevalent in Africa, yet its prevalence and associated factors remain almost unknown, especially among people living with HIV/ AIDS (PLWHA). In Uganda, there is a paucity of evidence about COPD among PLWHA, which negatively impacts efforts to develop integrated non-communicable diseases (NCD) care programmes among PLWHA. Much remains to be learnt about its risk factors and the mechanisms involved in its pathogenesis. Our aim was to determine the prevalence and risk factors among PLWHA. Methods. We conducted a cross-sectional study and screened 1 000 PLWHA (≥35 years old) attending the antiretroviral treatment clinic at Nakaseke Hospital for COPD using the modified BOLD questionnaire and spirometry. Spirometry was performed on the participants before and after bronchodilator therapy (400 µg of salbutamol using a spacer) following standardised guidelines. COPD was defined as the ratio of post-bronchodilator forced expiratory volume in 1 minute (FEV1) to forced vital capacity (FVC) less than the lower limit of normal. Results. Of the 719 participants, 58.9% (n=424) were female, 60.2% (n=433) were older than 44 years and 89.2% (n=641) were on ART. A total of 88.5% (n=625) reported exposure to biomass fuel (wood) and 84.5% (n=599) reported no history of smoking cigarettes. The prevalence of COPD among PLWHA was 7.2% and COPD was significantly associated with being male (odds ratio (OR) 2.5), smoking cigarettes (OR 3.2), and history of TB disease. There is also a strong association between COPD and duration of ART treatment, CD4 cell count, using biomass as fuel (wood) and age. Conclusion. COPD is prevalent among PLWHA. Given its chronicity and impact on the quality of life of a patient, it has the potential to reverse the achievements of the global community in the fight against HIV/AIDS. It is thus imperative to extensively study the prevalence of COPD in this population, as well as its associated factors, to inform policy formulation and design of interventions intended to address the ongoing double burden of comorbidity with NCDs and HIV/AIDS.

Prevalence and associated factors of chronic obstructive pulmonary disease Release of interleukin-26 from alveolar among people living with HIV/AIDs in the type II cells in response to bacterial rural communities of Nakaseke, Uganda extracellular vesicles from Haemophilus influenzae and Pseudomonas C Batte, A Kayongo, C Batte, T Naz, B Morgan, F Nassali, aeruginosa D Mawanda, R Kalyesubula, B Kirenga, T Siddharthan, 1

1,2

1,5

3,4

1,2

1,4

3,4

W Checkley3,4

K Che,1 M Paulsson,2 K Riesbeck,2 A Lindén3

Makerere University, College of Health Sciences, Kampala, Uganda Makerere University Lung Institute, Kampala Uganda 3 Bloomberg School of Public Health, Johns Hopkins University, Baltimore, USA 4 Division of Pulmonary and Critical Care Medicine, Johns Hopkins University, Baltimore, USA 4 African Community Center for Social Sustainability (ACCESS), Nakaseke, Uganda batchaux@gmail.com

1 2

38 AJTCCM VOL. 24 NO. 1 2018

Unit for Lung and irway Research, Intitute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden 2 Clinical Microbiology, Department of Translational Medicine, Lund University, Malmö Sweden, 3 Lung Allergy Clinic, KArolinska University Hospital, New Karolinska Solna, Stockholm, Sweden karlhans.che@ki.se

ABSTRACTS Introduction. To date, known cellular sources of the presumed Th17- cytokine interleukin (IL)-26 include T-lymphocytes, bronchial epithelial cells and alveolar macrophages in human airways. This IL10-related cytokine is involved in the innate immune response to bacterial endotoxins in healthy human airways. However, alveolar type II cells, which also play a role in the innate immune responses to bacteria, have not been investigated in terms of IL-26 release. Moreover, Haemophilus influenzae and Pseudomonas aeruginosa are two common causes of pneumonia that may outmanoeuvre innate immune responses while invading human airways, in part through the actions of extracellular vesicles (EV). The IL-26-stimulating properties of EV from these pathogens have not previously been investigated. We hypothesised that alveolar type II epithelial cells release IL-26 protein in response to EV from H. influenzae and P. aeruginosa. Methods. The EV from H. influenzae and P. aeruginosa were isolated and separated after overnight culture of the bacteria. A model of alveolar type II epithelial cells (A549 cell line) was utilised. Cells were cultured and stimulated in triplicates with different concentrations (0.1 and 1µg/mL) of the respective EV during several incubation times (1, 3, 6 and 24 h). IL-26 protein concentrations were thereafter measured in the cell-free conditioned media using an enzyme-linked immunosorbent assay (ELISA) Results. We found that A549 cells released substantial amounts of IL-26 protein in response to the EV from H. influenzae (p=0.03; n=6), as well as from P. aeruginosa (p=0.03; n=6). Notably, we found a concentration-dependent (n=3) as well as time-dependent (n=6) increase in IL-26 concentrations. Conclusion. Our results suggest that human alveolar type II cells produced IL-26 in response to EV from H. influenzae and P. aeruginosa in vitro. Thus, given the documented involvement of IL26 in bacterial responses in human airways, and our current finding suggests that IL-26 is involved in the innate immune responses during bacterial infections. This data also argues for the further exploration of the mechanisms through which alveolar type II cells and IL-26 are involved in the pathogenesis of pneumonia as well as other bacterial infections in the airways.

Treatment outcome of tuberculosis at a specialist hospital in North-Western Nigeria

B I Garba,1 A S Muhammad,2 I Yusuf,3 B A Mohammed,4 U Abdullahi5 Department of Paediatrics, Usmanu Danfodiyo University Teaching Hospital, Sokoto, Nigeria 2 Department of Medicine, Usmanu Danfodiyo University Teaching Hospital, Sokoto, Nigeria 3 Department of Paediatrics, Ahmad Sani Yariman Bakura Specialist Hospital, Gusau, Nigeria 4 Department of Obstetrics and Gynaecology, Ahmad Sani Yariman Bakura Specialist Hospital, Gusau, Nigeria 5 Department of Medicine, Ahmad Sani Yariman Bakura Specialist Hospital, Gusau, Nigeria bgilah@yahoo.com 1

Introduction. Tuberculosis (TB) is a chronic infectious disease that is preventable, treatable and curable, yet it is a major cause of morbidity and mortality. The prevalence and mortality of TB are under-estimated

in many high burden countries, including Nigeria. In order to decrease transmission of the disease, effective identification, diagnosis and treatment of infectious TB patients is required. A proven strategy to ensure patients’ adherence to anti-tuberculosis medication is the use of Directly Observed Treatment Short course (DOTS) therapy. DOTS lead to improved treatment outcome with overall reduction in morbidity and the development of multidrug-resistant TB. Objective. We aimed to determine the form of TB and treatment outcome of patients managed at our DOTS clinic. Methods. A retrospective study of patients managed for tuberculosis at the DOTS clinic of Ahmad Sani Yariman Bakura Specialist Hospital, Gusau, Zamfara State, Nigeria over a 30 month period (Jan 2015 to June 2017). All patients that were treated for TB over the study period were included in the study. Relevant information from the register was reviewed. Tuberculosis treatment outcomes were assessed according to World Health Organisation (WHO) and National TB and Leprosy Control Programme guidelines. ‘Cured’ and ‘treatment completed’ outcomes were referred to as successful treatment. Results. Of the 415 patients, 18.3% (n=76) were children, 81.7% (n=339) were adults and 61.2% (n=254) were male; the male:female ratio was 1.6:1. Most of the patients (83.9%; n=348) had pulmonary TB, while 16.1% (n=67) had extra-pulmonary TB. More males had pulmonary TB than females, which was not significant (χ2 =0.678; p=0.410). There were more male adults than females or children which was significant (χ2=18.504; p=0.000). The majority (97.6%) of the patients were new cases, 4 (1.0%) relapsed, while 3 (0.7%) were TB patients with unknown previous TB treatment history. Completed treatment was observed in 44.2% (n=184), 38.8% (n=161) were cured, 8.4% (n=35) were transferred out, and 7.0% (n=29) died; 2 (0.5%) were removed from the register and 2 (0.5%) were lost to follow-up. The treatment success rate was 83.0%. Conclusion. Treatment outcome of patients treated for TB in our centre was good, with a success rate close to the WHO benchmark of 85%. However, it shows that childhood TB is still under diagnosed and under treated as the number of paediatric cases were low.

Longitudinal changes in lung function in HIV- infected adolescents on antiretroviral therapy in Cape Town, South Africa

L Githinji,1 D Gray,1 S Hlengwa,1 T Machemedze,1 L Myer,2 H Zar1 Department of Paediatrics and Child Health, Red Cross Children’s Hospital and MRC Unit on Child and Adolescent Health, University of Cape Town, South Africa 2 Division of Epidemiology and Biostatistics, School of Public Health and Family Medicine, University of Cape Town, South Africa gthlea001@myuct.ac.za 1

Introduction. Over 90% of HIV-infected children live in subSaharan Africa. Despite increased access to antiretroviral therapy, respiratory illness remains common in HIV-infected youth. There is limited information on progression of lung function in HIV-infected adolescents on antiretroviral therapy. Objective. The aim of this study was to investigate progression of lung function over 2 years in HIV-infected adolescents on antiretroviral therapy in a prospective cohort, the Cape Town Adolescent Antiretroviral cohort (CTAAC).

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS Methods. HIV-infected adolescents aged 9 - 14 years, with at least 6 months of antiretroviral therapy, enrolled on CTAAC, underwent lung function testing. Spirometry and bronchodilator testing was done at enrolment and annually for two years. Healthy HIV-uninfected, age, sex and ethnically matched controls were also tested. Linear mixedmodels were used to compute longitudinal changes in lung function outcomes. Results. A total of 428 HIV-infected and 90 HIV-uninfected adolescents were tested at baseline and at 24 months. The mean (standard deviation) age was 12.0 (1.6) years and 50.4% were male. Median (interquartile range (IQR)) viral load and CD4 cell count at baseline were 2.2 (IQR 1.6 - 3.3) log copies and 731 (IQR 580 959) cells/µL, respectively. HIV-infected adolescents had lower lung function compared to the uninfected at all time points, p<0.05 FEV1 and FVC Z-scores showed similar change over two years in both groups. Obstructive and mixed spirometry patterns were more common in HIV-infected adolescents compared to the uninfected, p<0.05 for both time points. Previous pulmonary tuberculosis and previous lower respiratory tract infections were significantly associated with lung function (p<0.05 for both). Conclusion. HIV-infected adolescents had significantly lower lung function and more obstructive and mixed spirometry patterns than HIV-uninfected at all time points. Lung function tracked similarly over 2 years between groups, suggesting no catch-up growth or lung function deterioration over time. This study informs the importance of lung function surveillance in HIV-infected adolescents.

Exhaled nitric oxide in the first two years of life in African infants: Impact of maternal smoking and indoor air pollution C Jacobs,1 A Vanker,2 L Willemse,1 R MacGinty,1 L Turkovic3

Department of Pediatrics and Child Health, University of Cape Town, South Africa South African MRC Unit on Child and Adolescent Health, University of Cape Town 3 Telethon Kids Institute, Perth, Australia carvern.jacobs@gmail.com 1 2

Introduction. Environmental pollution and maternal smoking may increase risk of infant respiratory disease. Exhaled nitric oxide (eNo), a marker of airway inflammation, is increased in respiratory disease. The impact of indoor air pollution on eNO in early life is poorly understood. Objective. To asses changes in eNO in the first two years of life in African infants and the impact of tobacco smoke exposure and environmental pollutants on eNO at birth and two years. Methods. Infants enrolled in the Drakenstein Child Health Study underwent lung function testing at 6 weeks and again at 23 - 25 months. Lung function testing, completed during unsedated quiet sleep, included eNO and tidal breathing (tbfvl) measures. Antenatal smoke exposure was confirmed on maternal urine cotinine. Postnatal smoke exposure was self-reported by questionnaire. Benzene, particulate matter (PM10) and nitric dioxide (NO2), were measured durring antenatal home visits over a two-week period. The association between exposure and lung function outcomes was investigated using multiple linear regression, with statistical significance set at p=0.05. Results. At 6 weeks and at 2 years, 839 and 640 infants, respectively,

40 AJTCCM VOL. 24 NO. 1 2018

had eNO levels measured. Exhaled NO2 increased from an average (standard deviation (SD)) of 10.4 (7.6) ppb at 6 weeks to 17.5 (12.6) ppb at 2 years. At 6 weeks, infants of mothers who smoked during pregnancy, had lower eNO (10.2 (6.7) v. 11.5 (7.9) ppb; p=0.016) and NO2 output (32.2 (22.1) v. 38.6 (25.4) nL.sec-1,; p=0.001); and lower tidal volumes (33.2 (6.1) mL v. 35.5 (6.5) mL; p=0.001) compared with unexposed infants. After adjusting for size, sex and race, males and HIV-exposed infants had higher tidal volumes, average 2.1 mL (95% CI 1.0 - 3.1; p=0.000) and 1.7 ml higher (95% CI 0.2 - 3.2; p=0.024), respectively. Household smoke exposure was not associated with lung function. Adjusted Results showed that high benzene levels were associated with reduced eNO (average 1.47 ppb lower; CI 2.77 - 0.17; p=0.027) and NO output (average 5.5 nL/sec lower; 95% CI 9.6 - 1.3; p=0.009) at 6 weeks but not at 2 years. Other air pollutants were not associated with eNO levels at 6 weeks or 2 years of age. Conclusion. Maternal smoking during pregnancy and high benzene levels were associated with lower eNO and NO output at birth in African infants. The effects did not persist to 2 years. Further investigation on the impact of eNO levels on disease risk in early life is needed.

Exposure to indoor air pollutants and childhood pulmonary tuberculosis N Jafta,1 L Barregard,2 P M Jeena,3 R N Naidoo1

Discipline of Occupational and Environmental Health, School of Nursing and Public Health, University of KwaZulu-Natal, Durban, South Africa 2 Department of Occupational and Environmental Medicine, Sahlgrenska University Hospital and University of Gothenburg, Gothenburg, Sweden 3 Discipline of Paediatrics and Child Health, School of Clinical Medicine, University of KwaZulu-Natal, Durban, South Africa jaftan@ukzn.ac.za 1

Introduction. There is increasing evidence in studies that have used proxy measures of exposure that indoor air pollution increases the risk for tuberculosis. Objective. To determine the association between exposure to indoor air pollution (IAP) in the homes and childhood pulmonary tuberculosis (PTB). Methods. In this age- and sex-matched case-control study, cases were children diagnosed with PTB and controls were children without PTB. Questionnaires about children’s health; and house characteristics and activities (including household air pollution) and secondhand smoke (SHS) were administered to all the caregivers of the participants. A subset of the participants’ homes was sampled for measurements of PM10 over a 24-hour period (n=105), and NO2 and SO2 over a period of 2 to 3 weeks (n=82). IAP concentrations of PM10 and NO2 were estimated in unsampled homes using predictive models. Logistic regression was used to look for association between IAP concentrations, crude measures of IAP and PTB. Results. Of the 134 participants, 107 were cases and 127 were controls. Pollutants concentrations (µg/m3) for were PM10 GM.50.50 (47.50 53.70) and NO2 GM.15.74 (14.99 - 16.53) and SO2 GM.0.2 (95% CI 0.2 - 0.3). Day-to-day variability was large. All multivariate models were adjusted for age, sex, socioeconomic status, TB contact and HIV status. No significant association was observed between pollutant concentrations and PTB in children for PM10 and NO2. When using

ABSTRACTS the crude exposure measure of pollution, i.e. fuel type (clean or dirty fuel) and SHS, the association was positive but not significant. The presence of dampness in the household was a surprising significant risk factor for childhood TB with aOR ranging from 2.10 - 2.17 for different models. The crude predictors are less influenced by day-today variability. Conclusion. Our study suggests a risk of childhood tuberculosis disease when children are exposed to SHS and dirty cooking fuel but this is not supported by objective measurement of air pollution in the homes. HIV status and TB contact are important factors of childhood PTB in this population.

Impact of a primary care-led asthma clinic model on asthma control in a resource-limited setting I Kimuli, A Muhofa, W Katagira, L Mugenyi, B Kirenga Makerere University Lung Institute, Kampala, Uganda ikimuli@yahoo.com

Introduction. Recent advances in healthcare have dedicated efforts to reduce unnecessary healthcare spending while improving the quality of patient care and outcomes. Despite asthma being a very common disease with immense social impact, data on strategies for its control is very limited especially in resource limited settings. Objective. We assessed asthma control and its evolution over time among asthmatics treated using a primary care-led asthma clinic model which included evidence-based asthma diagnosis and stepwise management. Methods. Asthma patients aged ≥5years presenting at Mulago Hospital between August 2013 and April 2017 were enrolled and followed up every 6 months for 2 years. A separate clinic manned by a medical officer, a nurse and a counsellor was set up within the pulmonary unit. These individuals were trained in evidence-based asthma diagnosis and step-wise asthma management. At each visit, clinical assessments, including history, physical exam, asthma control test (a short tool to assess symptoms, use of rescue medications, and activity), and spirometry were performed. In addition to recording clinical assessment, providers recorded modification of treatment as ‘stepped down’, ‘no change’, or ‘stepped up’ and classified patients as ‘controlled’, ‘partially controlled’, or ‘uncontrolled’ based on the Asthma Control Test. Results. A total of 449 patients, with a mean age of 32 years were enrolled; 33.2% (n=149) were classified as controlled, 38.8% (n=176) were partly controlled, and 28.1% (n=124) were uncontrolled. At six months of follow-up, 69% (n=306) patients were seen, and 64.7% (n=198) were classified as controlled. The prevalence ratio (PR) compared with the baseline was 1.9 (95% confidence interval (CI) 1.6 - 2.2; p=0.001). At 12 months of follow-up, 303 (67%) patients were seen; 230 (75.9%) were classified as controlled (PR 2.9 95% CI 1.9 - 2.6; p=0.001). Conclusion. A primary care-led asthma clinic model more than doubled the proportion of controlled asthma patients. Capacity development through medical education and training of primary health care providers on evidence-based diagnosis and management of asthma significantly resulted in the control of asthma among patients. These results suggest that the primary care provider-led

model on asthma control could be used to improve asthma outcomes in low resource settings.

A case series of minimally invasive lobectomy for inflammatory lung disease J J Koshy, A Ogunromb

Department of Surgery, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa jiths20@gmail.com

Introduction. Minimally invasive lung resection is the standard of care for the management of surgically resectable malignant lung disease. However, its role in inflammatory lung disease with significant adhesions is still debatable. Objective. We present a case series of the use of multiport video-assisted thorascopic surgery (VATS) lung resection for patients presenting with various inflammatory lung pathologies, including recurrent haemoptysis due to mycetoma, destroyed lobe due to tuberculosis and infected bronchogenic cyst. Methods. Surgical approach. Double lumen intubation and right lung isolation, left lateral decubitus position, anterior approach via 3 ports, including the camera port the right upper lobe is approached through the pleural space. However, initial finger extra-pleural dissection for port placement may be required if thick adhesions are encountered at commencement. Subsequently, the upper lobe is freed by extra-pleural dissection. Once the entire upper lobe is freed, the hilar structures are approached with arterial then venous and lobar bronchial division. The order is altered depending on the anatomical difficulty in isolation or divided and stapled in combination depending on the difficulty in separation. Suturing or stapling is used depending on the safety of passing the stapling device around the structures without avulsion. Vascular rubber tapes are used for traction and control. The lung is completely freed from the diaphragm to allow for adequate reexpansion. Results. 1. 40-year-old female with right upper lobe mycetoma and recurrent massive haemoptysis 2. 20 year old female with right upper lobe mycetoma and recurrent minor haemoptysis 3. 54-year-old male with a destroyed right upper lobe due to recurrent tuberculosis. 4. 30-year-old male with a right-upper-lobe infected bronchogenic cyst. Conclusion. The application of minimally invasive lobectomy for inflammatory lung disease requires a number of strategies to minimise blood loss and ease dissection. The use of a combination of extrapleural dissection, maintaining a lower limit of mean arterial blood pressure, diathermy, ligasure and blunt swab on stick dissection makes it a technically feasible approach.

Knowledge, attitude and practice of primary care physicians in Kenya, Nigeria and South Africa regarding obstructive sleep apnoea in children D Marangu1,2 J-W Chang,3 F Akemokwe,4 B Chisunkha5

Department of Paediatrics and Child Health, University of Nairobi, Nairobi, Kenya Division of Pulmonology1 2 Division of Pulmonology, Department of Paediatrics and Child Health, University of Cape Town, Cape Town, South Africa

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS Department of Medicine, Lung Infection and Immunity Unit, University of Cape Town, Cape Town, South Africa 4 Department of Internal Medicine, University of Benin Teaching Hospital, Benin, Nigeria 5 The Malawi Liverpool Wellcome Trust Clinical Research Programme, Blantyre, Malawi dmarangu@uonbi.ac.ke 3

Introduction. Obstructive sleep apnoea (OSA) significantly impacts the morbidity of children globally, resulting in decreased quality of life and increased healthcare utilisation costs. Primary care physicians (PCPs) should be familiar and confident in their abilities to diagnose and manage OSA in children, for an effective impact on public health. Objective. Our study aims to describe the knowledge, attitude and practice of PCPs in Kenya, Nigeria and South Africa (SA) regarding OSA in children. Methods. Between April 2016 and July 2017, we conducted a multicentre cross-sectional survey in Cape Town (SA), Edo State (Nigeria) and Nairobi (Kenya). A minimum of 40 participants were randomly selected from a register of PCPs at each site. Following ethical approval, potential participants were contacted telephonically to obtain permission to email them the link to the online/paper-based validated OSA Knowledge and Attitudes in Children (OSAKA-KIDS) questionnaire. PCPs were excluded from the study if they had retired from medical practice, were not currently practicing in the study site regions, or declined to provide informed consent. Results. The median OSAKA-KIDS knowledge score among 184 participants was 67% (interquartile range (IQR) 56 - 72) -58% (IQR 44 - 67) among 80 Nigerian physicians, 61% (IQR 50 - 72) among 41 South African physicians and 67% (IQR 56 - 72) among 63 Kenyan physicians. Importance of paediatric OSA and confidence in diagnosis and management were rated by PCPs as 4.5 (IQR 4 - 5) and 2.3 (IQR 1.8 - 3.0), respectively, on a 5 point Likert scale. The overall median OSAKA-KIDS attitude score comprising both importance and confidence parameters among PCPs was 3.2 (IQR 2.8 - 3.8). 161 (88%) PCPs referred children with suspected OSA to a sub-specialist, mainly otolaryngologists (87%). A residency training (46 of 184 participants) was associated with OSAKA-KIDS confidence scores independent of respondent age, sex, country and time since graduation (OR 2.39; 95% CI 1.14 - 5.00; p=0.02). Conclusion. PCPs in Cape Town, Edo State and Nairobi have good knowledge and attitude regarding OSA in children; however their perceived confidence in the diagnosis and management is low. Increased emphasis on paediatric OSA management during undergraduate medical training, and through continuing professional development programmes, may be beneficial.

Complete CFTR gene mutation analysis in non-white patients with cystic fibrosis R Mphahlele,1 K de Boeck,1 H Cuppens,1 R Masekela1 University of Kwazulu-Natal University Hospital of Leuven mphahleler@ukzn.ac.za

1 2

42 AJTCCM VOL. 24 NO. 1 2018

Introduction. There is currently a lack of data on cystic fibrosis (CF)causing mutations in non-white African populations. Objective. To identify mutations and characterise phenotypic presentations of non-white patients diagnosed with CF in whom a disease-causing mutation was not found on both CFTR (cystic fibrosis transmembrane conductance regulator) genes using the Elucigene 30-mutation panel screen. Methods. The complete CFTR gene, including introns and flanking intergenic regions (including the promoter region), was sequenced with a highly parallel sequencing assay on blood extracted DNA from 5 non-white patients from KwaZulu-Natal, South Africa. Results. DNA samples of 5 patients with a mean age of 121 months (range 68 - 181) were sequenced. Four were black African and 1 was of Black/Mauritian/Asian (mixed) ancestry. Four CF-causing mutations, including 1 novel mutation were identified in 2 patients. S1255P/R709X (black African female) and L218X/c.2788G>5, the latter novel, (mixed ancestry male). The 3 remaining patients were found to have variants of unknown significance (VUS). Age at clinical diagnosis was lower in the subjects with CF-causing mutations compared with those with VUS: 54 months (range 48 - 60) v. 128 months (range 84 -156; p=0.08), respectively. The subjects with CF-causing mutations had better nutritional status and higher mean sweat test concentrations than those with VUS: weight-for-age Z-scores were -1.71 (range 1.49 - 1.93) and –4.8 (range –4.2 - 5.8; p=0.05), respectively, and the mean sweat test concentrations were 127 mmol/L (range 104 - 151) and 75 mmol/L (range 65 - 83; p=0.07), respectively. All the subjects had severe lung disease with a mean FEV1% of 41 (range 16 - 56). Commonly cultured organisms from sputum, were Staphylococcus aureus and Haemophilus influenzae. C onclusion. A gene sequencing approach can inform phenotypically supported CF molecular data assisting in compilation of an African population specific registry and mutational panel.

Infant lung function and exposure to oxides of nitrogen in a South African birth cohort S Muttoo,1 R Naidoo,1 P Jeena,2 K Ramcharan,1 A Mitku1

Discipline of Occupational and Environmental Health, School of Nursing and Public Health, University of KwaZulu-Natal, Durban, South Africa 2 Discipline of Paediatrics, School of Clinical Medicine, University of KwaZuluNatal, Durban, South Africa muttoos@ukzn.ac.za 1

Introduction. The relationship between air pollution exposure and infant lung function is poorly understood. Exposure to oxides of nitrogen NOx has been considered as an important determinant of lung growth both in utero and in childhood. Objective. The aim of this study was to determine the relationship between antenatal and early postnatal exposure to NOx and infant lung function (ILF). Methods. Newborns of mothers in the ‘Mother and Child in the Environment’ birth cohort, had ILF tests performed at 6 weeks, 6 months, 12 months and 24 months of age. Women in their first trimester of pregnancy were selected from public sector antenatal clinics in both the industrialised south and less-polluted north, in

ABSTRACTS Durban, South Africa, and were followed to delivery. NOx exposure was determined by land use regression (LUR) for the individual regional areas. ILF was conducted in unsedated sleeping infants, according to the ERS standards of ILF testing, with an ultrasonic flowmeter and 5% SF6 tracer gas. Multiple breath washout techniques were used to measure functional residual capacity (FRC) and lung clearance index (LCI). Results. The mean predicted antenatal exposures for NOx, determined by LUR, was 35.3 ug/m3 (range 17.74 - 40.86). One-hundred-andeight infants were tested (68% (n=74) valid ILF tests), with a mean gestational age of 38.95 weeks and mean birth weight of 3.2 kg. A mean FRC of 178.45 mL (range 57.65 - 435.93 mL) and LCI of 7.28 (range 5.71 - 9.73) was obtained. Tidal breathing and LCI were not associated with exposure. FRC showed a 2-unit (95% CI –4.10 - –0.04) decline with each unit increase in NOx, after adjusting for birth weight, gestational age, current child age and weight. Conclusion. The dose response in FRC decline with NOx is promising to understand the determination of lung growth but needs to be viewed cautiously; given the limited sample size and percentage of acceptable tests.

Agricultural burning practice and acute respiratory symptoms among rural farmers in Cameroon: A three-point cohort study M M Nganda,1 A Brian,2 M N B Hugo2

Ntam Sub-divisional Medical Centre, Kumba Iii, Cameroon Respiratory and Occupational Medicine, University of Douala, Cameroon 3 MPH, London School of Hygiene and Tropical Medicine maleakings@yahoo.com 1 2

Introduction. Rural farmers in Cameroon practice controlled burning of agricultural fields and household gardens during the dry pre-planting season, exposing themselves to increased immediate ambient air pollution. Objective. To measure ambient PM2.5 exposure and assess the change in acute respiratory symptoms in these farmers during the preburning, burning and post-burning periods. Methods. We conducted a three-point cohort study in January (preburning), March (burning) and May (post-burning) 2016 among adult rural farmers in Central Yabassi, Cameroon. A one-stage random cluster sampling of 4 on 15 communities and consecutive sampling of all farmers in clusters literate to French and English was applied. Questionnaires regarding demography, burning practice, and acute respiratory symptoms were investigated on same participants all three periods. Two 24-hour ambient PM2.5 measurements were taken in burning areas with a Dylos DC1700 Air Quality Monitor, and a mean value was calculated per study period. The prevalence of symptoms over time was compared using the McNemar test. Results. We studied 251 farmers with a mean (standard deviation (SD)) age of 47.79 (16.52) years; 64.5% (n=162) were exclusive farmers, while 70.1% (n=176) were permanent residents. The average (SD) number of burning days was 14.41 (5.29) with a mean (SD) burning time of 7.51 (2.41) hours per day. With a mean ambient PM2.5 exposure of 10.7 µm, 205.2 µm, and 7.0 µm for the pre-burning, burning and post-burning periods, respectively, the McNemar test revealed a

significant increase (p=0.001) in all symptoms investigated during burning compared with the pre-burning period, including sneezing (95% confidence interval (CI) 16.76 - 27.86), runny nose (95% CI 05.7 - 14.98), dry cough (95% CI 13.86 - 25.18), phlegm (95% CI 10.17 - 20.11), nasal congestion (95% CI 08.18 - 18.14), scratchy throat (95% CI 15.67 - 6.56), wheezing (95% CI 09.69 - 19.79), chest tightness (95% CI 14.94 - 25.70), shortness of breath (95% CI 11.71 - 23.35), and eye irritations (95% CI 24.57 - 36.78). With the exception of runny nose, nasal congestion and chest tightness, the post-burning period was characterised by a significant decrease (p=0.001) in symptoms. Conclusion. Rural farmers in Cameroon who engage in agricultural burning suffer from acute respiratory symptoms and are exposed to high levels of PM2.5. A general population survey involving lung functions will better evaluate health effects.

Women and girls in resource-poor countries experience much greater exposure to household air pollutants than men: Results from Uganda and Ethiopia Gabriel Okello, S Semple

Respiratory Group, Institute of Applied Health Sciences, University of Aberdeen, UK r01go15@abdn.ac.uk

Introduction. Household air pollution (HAP) generated from burning biomass fuels is a major cause of mortality and morbidity in low-income settings worldwide. Little is known about differences in exposure to HAP by age and gender in homes using biomass fuels. Objective. This study examined personal exposure to HAP across six population groups defined by age and gender (infants, young males, young females, adult males, adult females, and elderly) in rural households in Uganda and Ethiopia. Methods. Personal exposure to HAP in each group was assessed by measuring carbon monoxide (CO) and/or fine particulate matter (PM2.5) concentrations in households using biomass fuels. Measurements were made for ~24 hours for each individual with some participants wearing more than one instrument to provide data on comparability. Demographics including household, kitchen characteristics, biomass fuel use assessment and socioeconomic status were recorded. PM2.5 concentrations were measured using a TSI Sidepak AM510 Aerosol Monitor, a Particle and Temperature Sensor (PATS+), as well as Dylos and RTI MicroPersonal Exposure monitors. CO concentrations were measured using EL USB CO Lascar loggers. Results. Data were collected from 215 participants from 85 households. There was a difference in exposure to HAP between males and females (p=0.001). The 24 h PM2.5 exposures were highest among adult females with geometric mean (GM) and geometric standard deviation (GSD) 24 h concentrations of 194 µg/m3 (1.6) in Ethiopia and 156 µg/m3 (1.6) in Uganda. The lowest PM2.5 exposures were recorded among the ‘young male’ grouping with GM (GSD) of 25.2 µg/m3 (1.49) in Uganda and 26.4 µg/m3 (2.30) in Ethiopia. The ‘young female’ group had exposures about two-thirds the value of adult female group. Adult males and the elderly group experienced more moderate exposures. Conclusion. There are substantial differences in personal exposure to air pollutants depending on age and gender in the rural households. Adult and young females were exposed to the highest concentrations

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS in both countries. We found a ~5-fold difference in PM2.5 exposure between adult males and adult females. Future research ought to consider differences in exposure to HAP across the life-course and characterise age and gender differences when implementing exposure reduction interventions.

Treatment outcomes in patients with extensively drug-resistant tuberculosis from South Africa who received bedaquiline

O Olayanju,1 J Limberis,1 A Esmail,1 S Oelofse,1 P Gina,1 E Pietersen,1 R Warren,2 K Dheda1 Lung Infection and Immunity Unit, Division of Pulmonology, Department of Medicine, University of Cape Town, South Africa 2 DST/NRF Centre of Excellence for Biomedical Tuberculosis Research, US/ SAMRC Centre for Tuberculosis Research, Division of Molecular Biology and Human Genetics, Departments of Biomedical Sciences, Faculty of Medicine and Health Sciences, Stellenbosch University, Cape Town, South Africa olyola002@myuct.ac.za 1

Introduction. Optimal treatment regimens for patients with extensively drug-resistant TB (XDR-TB) remain unclear. Long-term prospective data comparing XDR-TB regimens, with and without bedaquiline (Bdq), from an endemic setting are lacking. Objective. Our aim was to compare treatment outcomes between XDR-TB patients who received Bdq-based regimen and those who did not. Methods. We prospectively followed up 272 patients with laboratoryconfirmed XDR-TB (49.3% HIV-infected; median CD4 169 cells/µL) in a 24-month programmatic setting. All patients were admitted to Brooklyn Chest Hospital, Cape Town, which is the designated XDRTB treatment centre in the Western Cape province of South Africa. A total of 204 patients received a non-Bdq-based anti-TB regimen, while 68 received a Bdq-based regimen. The background treatment regimen was prescribed following the Results of individual patient’s drug susceptibility testing. XDR-TB patients in the non-Bdq group were treated with a backbone of clofazimine and para-aminosalicylic acid (PAS), while those in the Bdq group also received clofazimine, linezolid and levofloxacin as major components. Demographic and clinical information was obtained from the patients’ records. Ethical approval was obtained from the University of Cape Town Human Research Ethics Committee. Results. The overall favourable outcome rates were substantially better in the Bdq v. the non-Bdq group (66.2%; (n=45/68) v. 12.8% (n=26/204); p<0.001). Those in the Bdq group also had overall reduced rates of treatment failure (5.9% (n=4/68) v. 25.6% (n=52/204); p=0.001), default (1.5% (n=1/68) v. 15.7% (n=32/204); p=0.003) and mortality (14.7% (n=10/68) v. 33.8% (n=69/204); p=0.004). Admission weight >50 kg, an increasing number of anti-TB drugs, and Bdq were independent predictors of survival. Bdq remained significant in HIVinfected persons, irrespective of CD4 cell count. A negative sputum culture at 6-months of treatment had a sensitivity of 97.2% and 81.0% to predict survival in the BDQ and non-BDQ groups, respectively. Conclusion. XDR-TB patients receiving a backbone of Bdq and linezolid had a substantial improvement in favourable outcome rates compared with those who were not using the drugs. These data inform

44 AJTCCM VOL. 24 NO. 1 2018

the selection of XDR-TB treatment regimens and roll-out of newer drugs in TB-endemic countries.

Very Brief Advice on smoking saves lives, but does it work in low- and middle-income countries? J Pooler,1 A McEwen,2 R Jones1

Plymouth University, Peninsula Schools of Medicine and Dentistry National Centre for Smoking Cessation and Training, UK jillian.pooler@plymouth.ac.uk

1 2

Introduction. The FRESH AIR project aims to improve the prevention, diagnosis and treatment of chronic lung diseases in low- and middle income countries (LMICs). Smoking is the single most preventable cause of death, resulting in an estimated 6 million premature deaths globally per year. Stopping smoking can reduce risk of premature death and improve current and future health. Very Brief Advice on smoking (VBA) is a proven clinical intervention, which identifies smokers, advises them on the best method of quitting and supports subsequent quit attempts. VBA comprises three elements, i.e. ASK, ADVISE and ACT, and is designed to be used opportunistically with patients by healthcare workers in almost any situation with a smoker. Objectives. To determine whether the VBA intervention can be adapted to low- and middle-income countries (LMICs). To investigate whether training healthcare workers in delivering VBA results in a change to their clinical or professional practice. Methods. Mixed-methods implementation science study. Stakeholders in Crete, Vietnam and Kyrgyzstan reviewed and adapted the standard UK model for the delivery of VBA, to ensure suitability to the local context. Training was provided for healthcare workers in the knowledge and skills needed to deliver VBA. Trainees’ self-efficacy and self-reported practice behaviours related to VBA were assessed through questionnaires before, immediately after and one month following the training. Interviews were conducted. Results. The original model for the delivery of VBA required minor local adaptation before implementation. Concern about effective delivery of VBA training by English speakers to healthcare workers via a translator as well as long-term sustainability, were addressed by adopting the train-the-trainer model. Local healthcare workers (N=126) were trained to deliver VBA training to other healthcare workers in the local language: n=29 in Crete; n=60 in Vietnam and n=37 Kyrgyzstan). Initial findings suggest that VBA training improved the skills of the majority of participants and they would recommend the training to others. Significant increases in selfefficacy in advising patients on the best methods of quitting and providing support to smokers were reported between the pre- and post-assessment periods. Conclusion. VBA training is a low-resource training which is acceptable, practicable and feasible in LMICs.

A proteomic atlas of human pulmonary tuberculosis M A Rahman,1 P K Ramdial,2 R Madansein,3 G Alexander3,4

Africa Health Research Institute, Nelson R. Mandela School of Medicine, University of KwaZulu-Natal, South Africa 2 University of KwaZulu-Natal 1

ABSTRACTS Department of Anatomical Pathology, National Health Laboratory Service Inkosi Albert Luthuli Central Hospital, Durban, South Africa Aejaz.Rahman@ahri.org

3 4

Introduction. Tuberculosis (TB) is a severe global health threat caused by Mycobacterium tuberculosis ( Mtb ) that forms granulomatous lesions in the lung. Host cells in these lesions vary in abundance, identity, activation status and the presence of detectable bacilli. The physiological and biochemical features of these lesions are poorly defined. Although animal models infected with Mtb have greatly expanded our understanding of TB, the paucity of TB human tissues for study has made it impossible to validate animal models. As a result, there is a scarcity of studies on fundamental mechanisms of human pulmonary TB. We tested the hypothesis that TB restructures the normal pulmonary architecture, forming heterogeneous granulomatous lesions that vary in cellular composition and arrangement, which is reflected in their proteomic composition. Methods. We appraised the immunohistochemistry of TB granulomas and used proteomics to examine resected pulmonary tissue from 12 TB and 6 healthy (non-TB) patients. The cellular composition of the lung was characterised using flow cytometry. Global proteomic analyses of the lung tissue were examined using a Surveyor HPLC in-line with a Thermo-Orbitrap Velos Pro-hybrid mass spectrometer. Results. MetaCore pathway analyses allowed us to identify statistically significant pathways, including: regulation of biological and developmental process (4.6e-303); extracellular component biogenesis (8.1e-281); cytoskeleton and cell junction organisation (9.1e-267); phosphatidylinositol and fibroblast growth-factor signaling (1.9e-246); response to tgf-β and system development (3.5e-195); apoptotic signaling via death receptors (8.4e-79); regulation of fatty-acid metabolic process (3.3e-65); morphogenesis of epithelium tissue development (5.1e-61); response to wounding and stress signal transduction (1.8e-40); and Fcreceptor signalling in phagocytosis (2.6e-13). Ingenuity Pathway Analysis showed upregulation of pathways involved in: (i) mitochondrial dysfunction; (ii) production of reactive nitrogen and oxygen species; and (iii) glycolysis in the TB-diseased lungs compared with healthy samples. Moreover, our histology results demonstrate discrete zonation of metabolic markers and cell types within the granuloma, which may play a role in restricting Mtb growth and/or dissemination. Conclusion. These pathology and proteomic data provide new insights into the mechanism of human pulmonary TB, which may lead to new paradigms of disease progression; especially how essential host bioenergetic pathways may be subverted by Mtb.

How well do the 2012 GLI multi-ethnic spirometry reference equations fit healthy South African school children?

S-J Smith,1,2 L Zurba,1 G Hall,3 S Stanojevic,4 D Gray,5 R Masekela1 Department of Paediatrics and Child Health, Nelson R Mandela School of Clinical Medicine, College of Health Sciences, University of KwaZulu-Natal, Durban, South Africa 2 National Heart and Lung Institute, Imperial College, London, United Kingdom 3 Children’s Lung Health, Telethon Kids Institute and School of Physiotherapy and Exercise Science, Curtin University, Perth, Australia 4 Division of Respiratory Medicine, Hospital for Sick Children, Toronto, Ontario, Canada 1

Department of Paediatrics and Child Health, Red Cross War Memorial Children’s Hospital and MRC Unit on Child and Adolescent Health, University of Cape Town, South Africa sara-jane.smith@imperial.ac.uk

Introduction. Spirometry is the most widely used lung function test for the diagnosis and management of patients with chronic lung disease. Interpretation requires accurate, ethnically appropriate reference equations. The Global Lung Function Initiative reference equations published in 2012 (GLI2012) provide reference data for caucasian, black, southeast Asian, northeast Asian, and ‘other’. However, the ‘black’ equation was derived from African-American population data and we hypothesise that this may not predict lung function well in black South Africans. Objective. We analysed spirometry results from healthy black South African (SA) children using the GLI2012 ‘black’, ‘caucasian’ and ‘other’ reference equations. Methods. We collected data from school-going children aged 5 - 16 years in KwaZulu-Natal, SA. A minimum of three spirometry measurements were performed to obtain values for FVC, FEV1 and FEV1/FVC. The GLI2012 online calculator was used to generate Z-scores for each prediction equation. Results. From 443 children screened for the study, 99 were excluded and 344 included, of which 327 (95%) were of black African ethnicity and included in this analysis. Of the included children, 208 (63.1%) were female and 236 (72.2%) lived in a rural area. The mean (standard deviation (SD)) height-for-age Z-score was –0.46 (1.06) and the weight-for-age Z-score was 0.15 (1.63). Black SA children did not match the GLI2012 ‘black’ equations, with a mean (SD) Z-score of 1.04 (1.05) for FVC and 1.06 (1.10) for FEV1. The GLI2012 ‘other’ and ‘Caucasian’ equations showed a closer fit, with a mean (SD) FVC Z-score of 0.45 (1.12) and –0.27 (0.99), and mean FEV1 Z-score of 0.41 (1.10) and –0.20 (1.03) respectively. All reference equations were more accurate for FEV1/FVC values, with a maximum mean Z-score of 0.11 (0.86) observed in the caucasian equation analysis. Conclusion. Our initial data suggest that GLI2012 ‘black’ reference equations are not valid for estimating FEV1 and FVC in black SA children. A new reference equation for black African populations is needed to aid the interpretation of pulmonary function testing in this group.

Impact of admission for RSV infection on lung function at one year of age: A case-control study

C Verwey,1,2 A Ghoor,1 L Ramocha,2 D Gray,3 Z Dangor,1,2 S Madhi2 Department of Paediatrics and Child Health, Faulty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa 2 Respiratory and Meningeal Pathogens Research Unit, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa 3 Department of Paediatrics and Child Health, University of Cape Town, South Africa charl.verwey@wits.ac.za 1

Introduction. Respiratory syncytial virus (RSV) infection is the commonest cause of acute lower respiratory tract infection (aLRTI) in children <5 years of age. While RSV aLRTI mortality is relatively low, the global impact of RSV hospitalisation has shifted towards the long-term sequelae associated with RSV aLRTI in early life.

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS Methods. We performed tidal breathing flow-volume loops and SF6 multiple breath wash-out at one year of age in children who were previously hospitalised for RSV aLRTI (n=83) and compared them with healthy, non-hospitalised controls (n=92) during natural sleep as per ERS/ATS guidelines. Results. There were no differences in the reported history of parental or sibling-diagnosed asthma, smoke exposure, method of infant feeding, overcrowding and creche attendance between cases and controls. Cases had more exposure to household pets (p=0.046). A larger proportion of cases reported any wheeze (p=0.001), nocturnal cough when well (p=0.003), hospitalisation for subsequent wheezing episodes (p=0.001) or recurrent aLRTI (p=0.001). Similarly, a larger proportion of cases had an increased respiratory rate (p=0.004), lower oxygen saturations (p=0.03) and lower tidal volumes (p=0.002). The lung clearance index was significantly increased in cases (7.7; interquartile range (IQR) 6.0 - 11.0) compared with controls (7.1; IQR 5.3 - 12.8; p=0.007). Conclusion. Children hospitalised with RSV aLRTI in infancy had more respiratory sequelae at one year of age compared with controls. In addition, these infants had more ventilation inhomogeneity and an increased work of breathing. Longer follow-up of these cases is required to evaluate the impact of RSV aLRTI on respiratory trajectories.

A pilot study assessing knowledge of women about carbon monoxide (CO) and 24-hour CO levels in selected homes in Lagos, South West Nigeria O O Adeyeye, Y E Kuyinu, S I Adewale

Lagos State University College of Medicine, Ikeja, Lagos, Nigeria Olufunkeadeyeye@yahoo.com 1

Introduction. Indoor air pollution remains poorly studied and reported in Nigeria. There is poor power supply and large use of gasoline powered generators by the citizens. Objective. To determine the knowledge of women in selected homes in Lagos about carbon monoxide (CO) as well as to measure the 24hour CO level in their homes. Methods. Convenience sampling of three local government areas was conducted in Lagos, South West Nigeria. A descriptive crosssectional study was carried out within selected households involving women 18 years and above in charge of each household who were interviewed with an interviewer-administered questionnaire. A 17-item questions was used to assess their knowledge about CO. The 24-hour CO level was monitored using the Easy Log USB CO Monitor. Results. The mean (standard deviation (SD)) age of respondents was 46.18 (14.6) years (range 21 - 80 years). A total of 62 (62%) were married, 68% had at least secondary education, more than 70% had lived at the address for >5 years, while 98% were non-smokers and only 58% of the household possessed a power-generating set (56% used petrol and 2% used diesel). Primary fuel use for cooking in household was kerosene in 65%, LPG in 32%. Generators were inappropriately placed in 27 (43.1%) households. The majority (69%) of the women had not heard of CO; 25 of the 31 heard about

46 AJTCCM VOL. 24 NO. 1 2018

it at school. The knowledge score was poor in 02 (6.5%), fair in 11 (35.5%) and good in 18 (58.1). Those who had heard about CO were much younger, with a mean (SD) age 41 (11.5) years compared with 48.66 (15.4) in those who had not (t=–2.448; p=0.012). The range of CO level 0 - 124.5 ppm, mean maximum CO of 19.81 (26.67). This is not affected by age. Only 53 households had safe CO levels, while values greater than 25 ppm were recorded in 26% of the households. None of the households had a CO detector or alarm. Conclusion. Despite the large use of portable gasoline generators among Nigerians, the knowledge of the women regarding the hazards remain poor. There is need for massive education to increase awareness about CO and other pollutants.

Introduction of pentavalent, pneumococcal conjugate and inactivated polio vaccines into routine immunisation schedule: Caregivers’ experiences in Edo State, Nigeria. I Alenoghena,1 A R Isara,2 S Ameh3

Owan East Local Government Council/Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria 2 Department of Community Health, University of Benin, Edo State, Nigeria 3 College of Medical Sciences, University of Calabar, Cross River State, Nigeria alendoc@yahoo.com 1

Introduction. Pentavalent vaccine; pneumococcal conjugate vaccines (PCV) and inactivated polio vaccine (IPV) were recently introduced into the EPI schedule of routine immunisations in Nigeria. Objective. To assess caregivers’ awareness, sources of information, uptake and satisfaction with the new vaccines in the Etsako Central local government area of Edo State, Nigeria. Methods. This descriptive cross-sectional study involved care-giver/ baby pair, receiving immunisation services at the heath facilities. The data collected using a combination of questionnaire and review of records were analysed using IBM SPSS version 21. Results. Of the 110 mother-baby pairs, 50 (45.5%) babies were aged 10 - 19 weeks, and 42 (38.2%) caregivers were aged 35 - 44 years. Most (90.9%) caregivers were aware of new vaccines introduced into the routine immunisation schedule. The major source of information was from health workers, 91 (82.4%). Only 10% were able to correctly identify the three newly introduced vaccines. Pentavalent vaccine was the most frequently obtained antigen, 94 (85.5%), followed by PCV, 40 (36.4%) and IPV, 38 (34.5%). Most (90.9%) caregivers were satisfied with services received regarding the new vaccines. Adverse event following immunisation with the new vaccines was reported in only 28.6% of the babies. Caregivers’ level of education (p=0.021) and satisfaction (p=0.043) with the new vaccines were significantly associated with up to date uptake of the vaccines Conclusion. Caregivers’ awareness about the newly introduced vaccines was generally high in this study, but in-depth knowledge of any of these vaccines was low. Only one third of the children had been immunised up to date with the new vaccines. The caregivers expressed a high level of satisfaction with the new vaccines.

ABSTRACTS

Comparative evaluation of obstructive sleep apnoea between persons living with HIV and general out patients in Federal Teaching Hospital Abakaliki, Ebonyi State, South East Nigeria C Alo, I N N Okedo

Federal Teaching Hospital Abakaliki, Nigeria chihurumnanyalo@gmail.com

Introduction. Obstructive sleep apnoea (OSA) is under-recognised and under-diagnosed, with lots of negative consequences on patients’ health. Undetected OSA can lead to hypertension, heart disease, depression, and death. OSA is also commonly encountered in the care of persons living with HIV (PLHIV). Objective. To compare OSA experience between HIV negative patients and persons living with HIV and to determine the association between OSA and HIV status. Methods. This cross-sectional comparative study was carried out among patients attending General outpatient clinic and HIV clinic of the Federal Teaching Hospital Abakaliki. The intervieweradministered Berlin Questionnaire (BQ) and Epworth Sleepiness Scale was used to assess the risk of OSA and EDS, respectively, among 151 PLHIV and 167 HIV-negative patients attending the GOP clinic. These were selected using a systematic random sampling technique after securing ethical clearance. Data was analysed using SPSS version 20. Frequencies and percentages were calculated and analytical components were compared at 95% level of significance. Results. Of the 167 GOP respondents, most (n=79; 47.3%) were between 20 and 29 years of age, while among the 151 PLHIV, most 52 (34.4%) were between 30 - 39 years of age. There were 88 (52.7%) males and 79 (47.3%) females among the GOP, while among the PLHIV 58 (38.4%) were male and 93 (61.6%) were female. Mean BMI and standard deviation among GOP and PLHIV were 25 (5.5) and 23 (4.6) respectively. Positive snoring experience was significantly higher (χ2=24.62, p=0.000) in GOP 58 (34.7%) than in PLHIV 23 (15.2%). There was no significant difference (χ2=3.645, p=0.07) in positive daytime tiredness between GOP and PLHIV (58 (34.7%), 23 (15.2%)) respectively. Positive daytime sleepiness was significantly higher (χ2=7.797, p=0.007) among PLHIV than GOP (57(37.7%), 39(23.4%)) respectively. Significantly higher proportion (χ2=24.629, p=0.000) of GOP had high risk of experiencing sleep apnoea than PLHIV (43 (25.7%) and 8 (5.3%)). Conclusion. General out patients experience OSA significantly more than PLHIV. However PLHIV experience more daytime sleepiness than GOPs.

Mortality and associated factors among patients with chronic obstructive pulmonary disease in Uganda P Alupo, L Mugenyi, W Katagira, B Kirenga

Makerere University Lung Insitute, Kampala, Uganda alupopat@gmail.com

Introduction. Chronic obstructive pulmonary disease (COPD) is the third leading cause of death globally, with 90% of its attributable deaths occurring in low- and middle-income countries like Uganda.

There is a lack of longitudinal data on the mortality and predictors of COPD among patients in Uganda. Objective. We set up the Uganda Registry for Asthma and COPD (URAC) to fill this data gap. Methods. COPD patients presenting at 6 tertiary hospitals in Uganda were enrolled into the URAC registry and followed up for 2 years to determine the incidence and predictors of mortality. The hospitals were Mulago National Referral, and five other regional referral Hospitals (Mbale, Mbarara, Hoima, Arua, and Gulu) located in different regions of Uganda. Results. We recruited and evaluated a total of 296 COPD patients; 57.6% were male. A total of 33 (11.2%) died. Higher mortality was observed in males IR=116, compared to females IR=76, IRR=1.51 (95% CI 0.72 - 3.26). By age group, incidence rates were 59.9, 27.4, 137.7, 89.9 and 113.6 for age groups 35, 35 - 44, 45 - 54, 55 - 64 and 65+, respectively. COPD stage by FEV1 was significantly associated with mortality, GOLD 4 compared to1 IRR 12.7 (95% CI 2.71 - 119.2; p=0.001), GOLD 3 compared 1 IRR 5.8 (95% CI 1.3 - 54.1; p=0.005) and GOLD 2 compared 1 IRR 2.8 (95% CI 0.6 - 25.9; p=0.090). Smoking history was borderline significant IRR 1.7 (95% CI 0.8 - 3.7; p=0.061). Other non-significant factors were HIV positive IRR 1.9 (95% CI 0.6 - 4.7; p=0.090), biomass smoke exposure IRR 0.6 (95% CI 0.2 - 2.8; p=0.172), and exacerbation h/o IRR 1.2 (95% CI 0.6 - 3.0; p=0.296). Age groups, hypertension and obesity were also not significantly associated with mortality. Conclusion. Among COPD patients in this Ugandan population, GOLD stage by FEV1 was the only prognostic factor for mortality. Co-morbidities did not significantly affect mortality outcomes in this patient population.

Effects of open and closed endotracheal suctioning on oxygenation and ventilator-associated events incidences: A systematic review P Assanga

The Nairobi Hospital, Cicely MacDonnel Nursing School, Nairobi, Kenya ppamessa@yahoo.com

Introduction. To compare the effects of open and closed endotracheal suction systems on oxygenation and ventilator associated events incidences in adult patients on mechanical ventilation. Intubation is the placement of a flexible tube into a patient’s trachea to maintain an open airway and assist with ventilation and oxygenation. Its presence causes increased secretions hence the need for suctioning to clear the airway. This is done through the endotracheal tube. Endotracheal suctioning is performed through use of open or closed system. These systems have various advantages and disadvantages as well as complications. Some of the complications are ventilator associated events (VAE), decreased oxygenation and hemodynamic disturbances. Methods. A systematic literature search was done from PUBMED, MEDLINE, CINAHL, EMBASE and Cochrane library databases to identify randomised controlled trials comparing open and closed endotracheal suction systems. The search was from studies conducted from 2007 - 2017. The search aimed to identify the effects of open and closed endotracheal suction systems on oxygenation and VAE

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS incidences: search was also interested in the studies that had been done on adult patients. Was done manually by going through the studies and critically analysing the findings then making inferences in relation to the aim of the study. Results. The search yielded 118 articles of which 36 articles were incorporated into this study. The results from these articles had the following findings. All the articles reviewed showed that there was a decrease in oxygen levels and saturation when the open suction was used compared to the closed system. On further analysis, the patients whom closed suction system was used there was either no changes in the oxygenation level or had an increase. This was from comparison of blood gas analysis done before and after suctioning. On VAE incidences, there was no difference between the two systems. The laboratory tests done on patients had the same incidence level despite the type of suction system used. Conclusion. The closed endotracheal suction system is better than the open one in terms of oxygenation levels in that from the results there was either an increase or no change in oxygenation. Although there was no difference in the incidences of VAE, the closed suction system has various advantages over the open one.

Atopy, disease control and quality of life in asthma: A cross-sectional study of Nigerian asthmatics O F Awopeju, O Salami, G Erhabor

Obafemi Awolowo University, Ile-Ife, Nigeria yemijide@yahoo.com

Introduction. Atopic sensitisation is one of the strongest risk factors for developing asthma, while quality of life and asthma control are the most common patientsâ&#x20AC;&#x2122; reported outcome measures (PROM) in clinical practice and research. Asthma control remains the central focus of treatment regimens and individual care plans irrespective of disease severity or phenotype. Objective. The relationship between atopy and these PROM is complex and not well established, we, therefore, investigated the relationship between these outcomes and sought role of atopy between these outcomes. Methods. In a cross-sectional study, we consecutively recruited 82 stable asthmatics (mean (standard deviation) age 44.3 (16.3) years; 59 females) who were previously diagnosed according to the GINA criteria, attending pulmonology clinic of a tertiary hospital. We assessed atopy by skin prick test reactivity to 6 common inhalant allergens. The asthma control, quality of life and lung function were evaluated using the ACT (asthma control test), mini AQLQ (asthmaspecific quality of life questionnaire), and spirometry. Correlations between asthma control and asthma specific quality of life were determined using Spearmanâ&#x20AC;&#x2122;s Rank correlation coefficient with line of best fit determined using the least-square linear regression. Mann-Whitney U test was used to test the difference between atopic and non-atopic asthmatics. Results. The median (IQR) ACT score was 18.0 (13.0 - 22.0) and median (IQR) AQLQ score was 4.7 (3.7 - 5.9). The ACT scores correlated positively with total AQLQ scores (rho=0.57; 95% CI 0.41 - 0.71; p=0.001) with a one-point increase in AQLQ associated with 2.12 (95% CI 1.40 - 2.85; p=0.001) increase in ACT. Fifty-

48 AJTCCM VOL. 24 NO. 1 2018

six of the 82 participants (68%) were classified as atopic based on sensitisation to at least one aeroallergen. There is no significant difference the between atopic and non-atopic asthmatics in the % predicted forced expiratory volume in one second (77 v. 72; p=0.70), ACT score (18.5 v. 18.0; p=0.91) and total AQLQ score (4.9 v. 4.6; p=0.22). We concluded that better asthma control is associated with better quality of life, however, atopic status does not affect PROM such asthma control or quality of life. We concluded that better asthma control is associated with better quality of life, however, atopic status do not affect PROM such asthma control or quality of life. We concluded that better asthma control is associated with better quality of life, however, atopic status does not affect PROM such asthma control or quality of life. Conclusion. We concluded that better asthma control is associated with better quality of life; however, atopic status does not affect PROM, such as asthma control or quality of life.

Environmental risk factors for current or severe asthma in 13 - 14-year-old African children participating in ISAAC III A Ayuk,1,2 H Zar,1 L Myer,1 R Erhlich,1 J Ramjith1 University of Cape Town, Cape Town, South Africa University of Nigeria, Enugu, Nigeria adaraymond@yahoo.com

1 2

Introduction. Asthma prevalence in Africa is as high as the global average and increasing. There are environmental factors associated with the increase. Comparison of findings across African countries is complicated by lack of consistency in methods Environmental risk factors for current or severe asthma in 13 - 14-year-old children using data obtained from ISAAC III African sites is yet to be estimated. Methods. We conducted a population-based cross-sectional study where children aged 13 - 14 years from 10 African centres who completed both core written questionnaire (WQ) and environmental questionnaires (EQ) of ISAACIII were randomly selected from school clusters. Appropriate ethical clearance was obtained. The dependent variables were current asthma (wheeze in last 12 months) and severe asthma (presence of any 3 defining questions for severity). The independent exposure outcomes were. engaging in physical exercise 3 or more times in a week, television watching 5 or more hours in a day, biomass and ETS exposure, consumption of paracetamol at least once a month, large family sizes and having pets in the home. Univariate and multivariate analyses were done adjusting for centre variation. Odds ratio and respective 95% confidence intervals were calculated. Significant p>0.05% was used. Results. There were 28 490 adolescents in 10 African centres with 4 middle-income and 6 low-income centres. There was a fairly equal M:F distribution. There were 7 groups of factors studied after exclusion of factors with much variability among centres. Maternal smoking was strongly associated with current asthma and severe asthma in adolescents. Exposure to biomass fuels, particularly cooking in the home with firewood predisposed to current asthma. Having a large family protected from current asthma and engaging in heavy exercise is associated with both current and severe asthma. There was a positive association between both current and severe asthma and the use of paracetamol.

ABSTRACTS Conclusion. The study results demonstrated a number of strong and consistent environmental associations, some of which confirm some fairly established or plausible associations with asthma in children.

Pulmonary capillary hemangiomatosis presenting without pulmonary hypertension: A case report

A Ayuk,1,2 A Vanker,1,3 D Gray,1,3 M Zampoli,1,3 S Owusu,1,4 D Marangu,1,5 K Pillay,1 E Bandekker3

Red Cross War Memorial Childrenâ&#x20AC;&#x2122;s Hospital, Cape Town, South Africa College of Medicine, Enugu Campus, University of Nigeria, Enugu State, Nigeria 3 MRC Unit of Child and Adolescent Health, University of Cape Town, South Africa 4 Komfo Anokye Teaching Hospital, Kumasi, Ghana 5 University of Nairobi, Kenya adaraymond@yahoo.com 1 2

Introduction. Pulmonary capillary haemangiomatosis (PCH) is a rare disorder of unknown aetiology and is often misdiagnosed as primary pulmonary hypertension or pulmonary veno-occlusive disease. It is a locally aggressive benign vascular neoplasm of the lung characterised by proliferation of benign thin-walled capillary sized blood vessels in the lung parenchyma with alveolar capillary proliferation. It usually affects children and young adults with symptoms typically overlapping those of primary pulmonary hypertension. Prognosis is poor and strongly related to associated hypertension. Case report. We present a rare case of a three-year-old girl referred with a history of cough followed by episodes of haemoptysis with clotted blood and melena stools (four bouts in one day), with haemoglobin levels dropping from 12.7 to 10.5 g/dL. Her reticulocyte count was 1% and her platelet count was 123 000/ ÎźL. There was no personal or family history of bleeding disorders. She had no clinical or echocardiography signs of pulmonary hypertension nor of portal hypertension. Chest radiography findings were consistent with chronic interstitial lung changes while chest CT showed ground glass opacification, right middle lobe nodular opacities with tree-in-bud appearance areas of pleural fibrosis at lung biopsy site and an absent right pulmonary artery. Cardiac catheterisation is awaited. Investigations for infections including tuberculosis and autoimmune conditions were negative. Broncho-alveolar lavage yielded >50% iron laden macrophages A diagnosis of Pulmonary capillary haemangiomatosis (PCH), was made on histology of the lung biopsy She had intermittent bouts of small quantity haemoptysis, a troublesome cough, worsening opacification on chest x-ray and had developed mild tachypneoa at rest with desaturation on six-minute walk exercise Consequently she was started on the mTOR inhibitor, Rapamycin, for its antiproliferative properties. We continue to monitor her for evidence of pulmonary hypertension. Conclusion. This case highlights the need for extensive investigation, which may include lung biopsy, and follow-up of children who present with haemoptysis of undetermined origin. Prompt and thorough investigation may allow treatment that improves outcome.

Clinical profile and outcome of pulmonary embolism in Douala, Cameroon

M N B Hugo,1 K Felicite,1 C R Nina,1 T Bruno,2 E Serge3 Department of Internal Medicine, Douala General Hospital Centre des Maladies Respiratoires de Douala, Douala Military Hospital 3 Faculty of Medicine and Pharmaceutical Sciences, University of Douala mbatchou.ngahane@yahoo.com 1 2

Introduction. Pulmonary embolism (PE) is one of the manifestations of venous thromboembolism. A limited number of studies report on PE in sub-Saharan Africa. Objective. To determine the clinical profile, and outcome of pulmonary embolism in 3 hospitals in Douala, Cameroon. Methods. A descriptive study carried out at the intensive care unit and the internal medicine department of the Douala general Hospital, at the medical service of the Douala Military hospital and at the Centre des Maladies Respiratoires de Douala. Patients admitted for PE between January 2009 and May 2017. The diagnosis of PE was confirmed by a thoracic CT angiography. Sociodemographic characteristics, clinical data, the results of the work-up and the outcome of patients were collected. Statistical analyses were done using SPSS 20.0. The study was approved by the institutional review board of the University of Douala. Results. In total, 103 patients were included in the study, with a male predominance. The median age was 52 years. Among the risk factors, the most common was obesity with 49.5% followed by hypertension with 35% (36 patients). Dyspnoea was present in in 86 patients (83.3%) followed by chest pain in 81 patients (78.6%) and respiratory distress in 58 patients (57.3%). Chest X-ray showed a pleural effusion in 33 patients (47.15%). Cardiac ultrasonography showed an enlargement of the right heart cavities in 61 patients (61%). Sinus tachycardia was present in 64 patients (63.36%). The initial treatment consisted of heparin therapy followed by antivitamin K. The complications during hospitalisation were pulmonary infection in 31 patients (30.1%) while the mortality rate was 18.4%. Conclusion. PE mostly affects young subjects in Douala. The most frequent risk factor was obesity. Dyspnoea and chest pain are the mains symptoms and the complication rate is high.

Mediastinal cystic teratoma revealed by pleural effusion. Clinico-pathologic characteristics and outcomes after surgery: Report of a case from a lowincome setting and a review of 63 cases in the literature N S Byamungu,1 P de Marie Katoto2

Department of Paeditric, University of KwaZulu-Natal, South Africa Department of Internal Medicine, Faculty of Medicine, Catholic University of Bukavu, Bukavu, Dem. Rep. of Congo and Centre for Environment and Health, Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium byamungu.nsuli@gmail.com 1 2

Introduction. Mediastinal cystic teratoma is a rare diagnosis in adolescence, especially in low-income settings. Pleural effusion as symptom can often lead to the diagnosis of respiratory infections and delay the resection.

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS Objective. We report a case from a low-income setting and analyse data from the literature. Methods. Clinical record of one adolescent patient from the Democratic Republic of Congo reported here and 63 records from other institutions identified via Medline between May and November 2017 analysed. We excluded three patients from analysis for missing of clinical data. We described, from full-text articles, demographic, clinico-pathologic characteristics, and surgery outcomes and followup of patients. Results. Of the 64 reported cases, 59% were females and the mean (standard deviation) age was 26 (12) years. Most of them were from high-income countries (Japan 28%, France 25%, USA 18% following by India 7%). Cough, shortness of breath, fever and chest pain were mostly reported. Majority of cases were misdiagnosed with time to first clinical symptoms ranged from one week to two years. Further, our case was first treated twice as tuberculosis (different regimens). About 91% showed an immature form whilst 16% were malignant. Ovarian and pancreatic tissues were mostly retrieved as content. Duration of hospitalisation varied from four days to six months and septic as well neurologic complications were mostly reported. With a follow-up period (of half of cases) ranged from one week to two years, 5% of patients died after surgery and 26% relapsed. Conclusion. Mediastinal teratoma can be misdiagnosed for a long period and revealed by a fatal pleural effusion. The present report expands the spectrum of our knowledge showing the scarcity of reported mediastinal cystic teratoma from low-income countries although a high frequency of pleural effusion in this region and therefore highlight the need of its integration for differential diagnosis of pleural effusion.

Interruption of anti-tuberculosis drugs among adolescent and young adults aged 10 - 30 years at Kapkatet County Hospital, Kenya S Cheruiyot, C Kirui, P Koech, A Terer

Ministry of Health, Kapkatet County Hospital, Kenya cheruiyotsambu@yahoo.com

Introduction. Adolescents and young people represent a growing share of tuberculosis (TB) worldwide. Adolescents account for over 40% of the total population in Kenya. The UNAIDS report of 2017 flagged Kenyaâ&#x20AC;&#x2122;s population of school-going teenagers of ages between 15 - 19 years as the most likely to stop treatment. Objective. To highlight the interruption of anti-TB treatment among adolescent and young people at Kapkatet County Hospital. Methods. A retrospective study was done on adolescent and young adults aged 10 - 30 years old, who had been diagnosed with TB and started anti-TB treatment at Kapkatet County Hospital between between January 2015 and December 2017. Ministry of Health TB registers were used as source documents. Data were extracted and analysed using SPSS version 21. Relationship-associated factors and drug interruption was established. P-values <0.05 were considered significant. Results. A total of 300 adolescents and young adults were assessed during the review period; 63% (n=190) were male and 37 % (n=110) were female. Out of 300 adolescents who were on anti-TB drugs, 94% (n=282) were school-going. At the end of the review period, only 60%

50 AJTCCM VOL. 24 NO. 1 2018

(n=180) completed treatment successfully; 60% were female and 40% were male. On anti-TB drug interruption, 40% (n=120) missed drugs and clinic appointments; 70% (n=84) were male. There was a positive correlation between male and female and drug interruption (p=0.002). Conclusion. Treatment failure was generally high and more focus should be put on males adolescent and young adults. Adolescents and young adults are likely to interrupt treatment both at intensive phase and at continuation phase of treatment related to stigma, bill burden, and duration of treatment, social support and school linkage. There should be structured friendly adolescence and young adult package of care and treatment to improve on anti-TB adherence.

TB/HIV services reduce resistance of drugs among co-infected patients at Kapkatet County Hospital K C Kipkosgei, C Sambu Kapkatet District Hospital collkirui83@gmail.com

Introduction. TB/HIV co-infection among TB patients tested for HIV stand at 41% in Kenya. At Kapkatet District Hospital is at 26% co-infection rate as at 2016. TB/HIV services at Kapkatet hospital started in 2005 and was supported by PEPFAR Project. TB clinic was approximately 100 m away from HIV clinic. Despite existence of referral mechanism patient got loss in the process due to distance, time and stigma. This contributed to low uptake of ARVS at 9.5% (2014) and increase in number of defaulters while treatment success was at 62% which contributed to resistance of ARVs. In order to address the challenges faced; the hospital management had to set up integrated TB/HIV services under one roof. Methods. TB/HIV collaboration committee was formed in 2014 comprising of TB/HIV stakeholders and hospital management. In 2015, committee identified the challenges and solutions by review of clinic data. One of the major challenge was separation of TB and HIV clinic. TB/HIV drugs was integrated as comprehensive services including OIs management in the same room Results. ART uptake increased from 9.5% in 2013 to 96% in 2014, adherence to anti-TBs was at 70% in 2013 to 94% in 2016. Out of the 51 co-infected patients treatment success was at 98% and viral load suppression was at 90% thus this results was attributed to integration of service and regular follow up of the patients Conclusion. Integration of TB/HIV at public hospital improve treatment success of TB co-infected patients. Setting up a functional mechanism of collaboration facilitate out the integrated services. The integration of this do not necessarily require additional health care workforce but ensure continuation of care to patient by building confidence on existing healthcare providers this reduce stigma thus improving drug adherence for the patient.

Nutrition status among TB/HIV coinfected patients attending Kapkatet County Hospital, Kericho County, Kenya K C Kipkosgei, C Sambu Kapkatet District Hospital collkirui83@gmail.com

ABSTRACTS Introduction. Despite national progress on increased access to care and treatment of patients with tuberculosis, malnutrition remains a challenge and TB/HIV still remain a major cause of mortality in undernourished patients. At Kapkatet County Hospital, 60% of patients with TB/HIV co-infection are undernourished. Nutrition is a core component of care and treatment in Kenya today. Undernutrition lowers immunity, which leads to more complications. Currently, nutrition supplements supporting TB/ HIV patients target only severe acute malnutrition (SAM) with body mass indices (BMI) <16 kg/m2, which excludes others. All the relapse TB cases were undernourished and 50% missed admission criteria for nutrition supplementation. Objective. To evaluate the nutrition status of TB/HIV co-infected patients attending Kapkatet county hospital. Methods. Prospective study of patients who attended TB/HIV coinfected clinic for care and treatment from January 2015 - December 2016 were assessed for nutritional status. Two nutritionists were assigned to assess all co-infected patients seeking care and treatments and document in the nutrition daily activity register. BMI was the main anthropometric assessment used. Nutrition status and gender was established using STATA. Results. Undernutrition with BMI 18.5 was 60% (n=112); 80% of the patients were male. A total of 15% had a BMI of 25. There was a positive correlation between nutrition status and gender (p=0.05) all severe acute malnutrition cases were males while all over nutrition cases were females. Conclusion. Structured nutrition assessment and counselling is critical in addressing malnutrition in TB/HIV co-infected patients. There is need to continuously support all undernourished (BMI 18.5 kg/m2) more emphasis should be put on male clients in terms of supplementation. Regular and timely supply of nutrition supplements is recommended.

Sleep study: An audit

C Daniels,1,2 G Symons,2 C Koegelenberg,1 H A Tariq,2 R Raine2 University of Stellenbosch University of Cape Town chetoivo@gmail.com

Introduction. Obstructive sleep apnoea is a common condition associated with increased cardiovascular risk. The condition is diagnosed using nocturnal polysomnography in a sleep laboratory. Data from the public health care sector in South Africa is lacking. Methods. The study takes the form of a retrospective review of folders of patients attending the Sleep Clinic at Groote Schuur Hospital, Cape Town, who underwent nocturnal polysomnography studies between 2013 and June 2016. Results. 175 sleep studies were included in this review. 116 (60%) participants are male and 78 (40%) are female. The mean age of participants is 49.2 (13.1; range 16.9 - 85) years. The most commonly occurring co-morbid condition in this cohort is hypertension (115). The symptoms most frequently reported by the participants are snoring (153), poor quality sleep (129), excessive somnolence (126), apnoea and gasping (109). The mean Epworth Sleep Score is 14 (5.8; range 0 - 24). 94 patients are obese. 137 patients were diagnosed with OSA. 82 (48.2%) participants have severe sleep apnoea associated with apnoea/

hypoapnoea index greater than 30. Males in this cohort are 3 times more likely to have obstructive sleep apnoea than females (OR 2.9; 95% CI 1.3 - 6.6; p=0.05). In addition, males are also more likely to have severe sleep apnoea when compared with females in this cohort; 53 males (71%) had severe sleep apnoea. Patients with severe sleep apnoea tended to be younger with 49 patients aged between 30 - 59 years. Conclusion. Obstructive sleep apnoea was common in this cohort. Males were at a greater risk of sleep apnoea independent of other comorbid conditions. Patients with severe sleep apnoea tended to be younger. Further analysis is required.

Lung hydatic cyst in children: Our experience at Nelson Mandela Academic Hospital A Delgado,1 E M Loko2

Department of Surgery, Faculty of Health Sciences, Walter Sisulu University and Nelson Mandela Acadaemic Hospital, Umtata, South Africa Nelson Mandela Academic Hospital, Umtata, South Africa delgadoarturo15@yahoo.com

Introduction. Hydatid infestation of the lung can be primary or secondary. Cystic echinococcosis is regarded as endemic in Sub-Saharan Africa, available evidence suggests that several species or strains within the Echinococcus granulosus complex are prevalent in sub-Saharan Africa; however, very little data are available in most countries. Methods. We did a retrospective study of all children suffering from lung hydatic cyst admitted in our paediatric surgery ward, from September 2015 to September 2017. A total of 11 children wee included in the study and we collected and analysed different variables, e.g. age, sex, lung affected, unilateral or bilateral cyst, size of the cysts, treatment and complications, these variables were interrelated each otherâ&#x20AC;&#x2122;s, were created tables for its statistic study Results. From those 11 children suffering from lung hydatic cysts, 8 were females and 3 males; the 6 - 10 years age group was most affected with 5 patients for a 45.5%, followed by the groups of 5 - 6 years and >10 years with 3 children (27.25%) each. The left lung was the most affected in 6 patients by 54.5%, the right lung and bilateral lungs were affected only in 3 patients each (27.25%). The PAIR (punctureaspiration-injection-reaspiration) procedure was performed in 8 children (72.7%). The bronco-pleural fistula as a complication occurred in 7 patients and a cyst ruptured in 1 patient. We did not have any deaths in our group of patients. Conclusion. Lung hydatic cysts in our patients were observed more frequently in the left side. The surgical treatment performed was PAIR mostly in cysts more than 6 cm or with complications. Conservative treatment is an option in some cases.

Leukocyte count and its diagnostic value in newly diagnosed tuberculosis

A T Ebenezer,1 M N B Hugo,1 N D Eveline,1 C V Lum,1 L N Henry2 University of Douala, Cameroon University of Yaounde, Cameroon tazi_ebenezer@yahoo.fr

Introduction. The diagnosis of tuberculosis (TB) is frequently challenging given that the clinical and radiographic features of TB are

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS often nonspecific. Microscopic examination for acid fast bacilli-the most accessible in resource-limited facilities is of low sensitivity (50 - 60%). Recent studies show that a high or low monocyte lymphocyte count ratio (MLR) is predictive of active TB. Objective. To determine the blood leukocyte cell counts and the diagnostic value of the monocyte/lymphocyte (MLR), the neutrophil/ lymphocyte (NLR) count ratios in newly diagnosed TB patients. Methods. This was a cross-sectional study. A total of 204 consecutive case files of newly diagnosed TB patients were recruited over 2 years from registers of the TB treatment centre of the Douala general Hospital. Controls were 204 healthy volunteers age and sex matched, recruited at the blood bank of the hospital. Demographic, clinical and haematological data were collected. The diagnostic value of leukocyte counts was determined using receiver operating characteristics curve analysis. Sensitivity, specificity, negative (NPV) and positive (PPV) predictive values were also calculated. The areas under the curves were determined. Results. Lymphopenia (22.1%), neutrophilia (14.2%) and monocytosis (23.5%) were the most common abnormalities among newly diagnosed TB patients. The monocyte-lymphocyte count ratio (MLR) and neutrophil-lymphocyte count ratio (NLR) were significantly higher in the patient group compared to control group. NLR>1.79 and MLR0.29 were identified as optimal cut-off values for discriminating TB patients from healthy subjects. The areas under the curves were 0.77 and 0.84 for the MLR and NLR respectively. The MLR showed 67.2% sensitivity, 83.3% specificity, a PPV of 80.12% and a NPV of 71.73%, while NLR showed 70.6% sensitivity, 87.3% specificity, a PPV of 84.7% and NPV of 74.79% Conclusion. Lymphopenia, neutrophilia and monocytosis were the most common abnormalities among newly diagnosed tuberculosis patients. The monocyte/lymphocyte and neutrophil/lymphocyte ratios were raised in TB patients and were fairly predictive of active TB. A NLR >1.79 and MLR >0.29 were the cut-off values for discriminating TB patients from healthy subjects. Similar studies with larger sample sizes are needed to confirm or infirm these findings.

Severe asthma in children: Experience in a tertiary health facility in Nigeria J N Eze, A C Ayuk, T Oguonu

Department of Paediatrics, College of Medicine University of Nigeria and University of Nigeria Teaching Hospital, Ituku Ozalla, Enugu, Nigeria joyezedr@yahoo.com

Introduction. Severe asthma is a heterogeneous disease characterised by sustained asthma symptoms, despite treatment with high doses of corticosteroids. Achieving control in this group of patients can be difficult. Objective. Three cases are presented to highlight the management challenges in Nigeria. Methods. Patient information on initial diagnosis of asthma, frequency of flare ups, medications used and current status were obtained from the case notes. Results. Case 1. A 6-year-old girl diagnosed with asthma at age 3 years. Known triggers were aeroallergens. house dust mite from rug, fumes from generators and a nearby paint industry and these were controlled for. She had recurrent (2 to 3 months) asthma exacerbation despite incremental doses of inhaled corticosteroids, and long-acting beta agonist (ICS/LABA), and leukotriene receptor antagonist (LTRA). Her asthma

52 AJTCCM VOL. 24 NO. 1 2018

control test score (ACT) remained on the average 19 or below. She had frequent emergency hospital visits for life-threatening exacerbations. Case 2. A 16-year-old adolescent female diagnosed with asthma 5 years ago with an average of 6 hospital admissions per year; also has exercise-induced bronchospasms (EIB). She was on low dose ICS/ LABA but monitoring was difficult as she defaulted to appointments. She had several medical visits to the private sector for severe exacerbations. Emotional upset was a notable trigger. Her ACT score remained below 19 on most clinic visits despite incremental doses of ICS/LABA, and LTRA. Case 3. A 12-year-old male with asthma symptoms in the last 4 years who had twice-weekly flare-ups which increased to daily attacks. Risk and trigger factors were BMI of 30, biofuel smoke exposure and exercise. He was on low dose ICS/LABA. His ACT score remained below 19 on most clinic visits. He received incremental doses of ICS/ LABA; and LTRA. Adequate inhaler technique and weight reduction plan were instituted. Need for medication adherence was emphasised. He is currently on a medication step-down process. Conclusion. Severe asthma is common in children as demonstrated in our case reports. There is need to ensure proper diagnoses, eliminate co-morbidities and institute targeted therapy in children with difficult asthma.

Prevalence and predictors of tuberculosis treatment default in Abakaliki, Nigeria: Policy implication on its elimination goal N C Eze, B N Azuogu

Department of Community Medicine, Federal Teaching Hospital Abakaliki, Nigeria ezenelson24@gmail.com

Introduction. Tuberculosis (TB) treatment is the most effective strategy for preventing the spread of the disease. Default in tuberculosis treatment remains an important contributor to treatment failure, resurgence of multidrug resistance (MDR-TB), prolonged infectiousness, relapse and death. Patients’ adherence is an important link between medical process and good treatment prognosis. TB treatment default is a threat to its elimination goal and increases the burden of TB. Nigeria ranked fourth among the 22 high-burden countries for TB in the world and first in Africa. Objective. To determine the prevalence and predictors of tuberculosis treatment default in Abakaliki Methods. Data from the TB treatment register (686 patients) of the Federal Teaching Hospital Abakaliki for the period of 2012 - 2016 were analysed. Key informant interviews with nurses were conducted to determine the factors associated with default. Approval was obtained from the hospital management and permission was also sought from nurses. Treatment outcome was grouped as ‘defaulters’ and ‘nondefaulters.’ Data analysis was done using SPSS statistical software version 20. The χ2 test of statistical significance was used in the analysis and level of significance was determined by a p-value <0.05. Results. The mean (standard deviation) age of the respondents was 34.9 (5.7) years. Of the 686 clients, 72 (10.59%) defaulted within the five years and out of 98 currently on treatment 11 defaulted giving a prevalence rate of 10.7%. Majority (73.1%) defaulted within the intensive phase of treatment. Reasons for default were feeling of well

ABSTRACTS within the first month of commencement of drug, distance to health facility and pill burden. Among the defaulters, 25% were â&#x2030;¤ 29 years, 52.8% males and 68.1% lived in rural area. Predictors of TB treatment default were male gender (aOR 2.1; 95% CI 1.4 - 7.5), rural residence (aOR 1.8; 95% CI 1.3 - 5.7). Conclusion. Default rate was high among clients. Therefore this rate of default needs to be further reduced by repeated counselling, reminder system, effective contact tracing, support group and referral. High default rate is a huge threat to TB elimination goal especially now that there is increase in the incidence of MDR-TB. Further decentralisation of treatment centres closer to rural areas would improve adherence to TB treatment.

Assessment of prevalence, knowledge of preventive and control measures of pulmonary tuberculosis among inmates and staff of Abakaliki prisons, Nigeria: An implication for policy implementation N C Eze, B N Azuogu, I Okedo-Alex

Department of Community Medicine, Federal Teaching Hospital Abakaliki, Nigeria ezenelson24@gmail.com

Introduction. Pulmonary Tuberculosis (PTB) is one of the major diseases of public health importance especially in prisons where case finding rate has been low. The WHO established five facts of prisons PTB spread include. Prisons receive TB, Prisons concentrate TB, Prisons disseminate TB, Prisons make TB worse, and Prisons export TB. Poor TB capse finding result in annual TB transmission risks of 90%. Objective. To assess the prevalence, knowledge of preventive and control measures of pulmonary tuberculosis among inmates and staff of Nigerian Prisons, Abakaliki. Methods. A prison-based cross-sectional descriptive study was undertaken among 307 inmates and staff selected using a systematic sampling technique. Informed consent was obtained from the staff and inmates. The respondents were interviewed using a pre-tested interviewer administered semi-structured questionnaire. Good knowledge of pulmonary tuberculosis was assessed by the proportion of respondents who correctly answered 50% of the knowledge questions, and sputum test was done for respondents with cough of two weeks or more. Data analysis was done using SPSS statistical software version 20. The Ď&#x2021;2 test of statistical significance was used in the analysis and the level of significance was determined by a p-value <0.05. Results were treated with strict confidentiality. Results. The mean (standard deviation (SD)) age of inmates was 34.96 (5.7) years, while the mean (SD) age of the staff was 38.43 (3.5) years. The majority of the participants had secondary education. All the staff and 89% of inmate were aware of pulmonary tuberculosis while 63% and 77% of inmates and staff respectively had good knowledge of pulmonary tuberculosis. Knowledge was significantly associated with educational and employment status of inmates but only educational attainment by staff. This study found 2.1% prevalence of PTB by sputum test. Conclusion. Knowledge of presentation, preventive and control measures of PTB was high among respondents. However, this level of knowledge especially by the inmates needs to be improved upon

by awareness creation. High PTB burden and poor control policies within prisons potentiate high attributable risk. Implementation of current national or international cell occupancy recommendations would reduce TB transmission by 50% and 94% respectively especially now that there is increase in the incidence of MDR-TB.

Detection and molecular characterisation of enterovirus D68 in Senegal between July and December 2014 A Fall, D Ndongo, K Ousmane, B S C Bouh, N M Ndiaye Department of Virology, Instiitut Pasteur of Dakar, Senegal amary022@hotmail.com

Introduction. Since it was first isolated from hospitalised children with pneumonia and bronchiolitis in California in 1962, human enterovirus D68 (EV-D68) has been recognised as an emerging respiratory pathogen in the last decade when it caused outbreaks in several countries. The most recent and largest epidemic of EV-D68 associated with severe respiratory disease began in North America and spread everywhere in the worldwide between August 2014 and January 2015. However, in African countries the epidemiology of EVD68 remains largely unexplored. Objective. To investigate the circulation and molecular characterisation of EV-D68 in Senegal between July and December 2014. Methods. A total of 435 swab samples collected through routine influenza-like infection (ILI) or acute respiratory infection (ARI) surveillance activities between August and December 2014. Collected samples were tested using the EVD68-specific real-time RT-PCR. Molecular characterisation was made by amplification of the VP1 regions, followed by nucleotide sequencing. Results. Real time (RT-PCR) was used to confirm EV-D68 infection in 3.45% of the samples (n=15/435). The majority (n=12/15; 80.0%) of the EV-D68 cases were detected in October. Children under 5 years of age were more vulnerable to EV-D68 infection with a frequency of 60%. Phylogenetic analysis of the VP1 sequences of 14 EV-D68 cases, revealed that all sequences belonged to the A2 variant of clade A viruses. The VP1 sequence of these strains displayed sequence similarities between 99 - 98% with strains A2 found in Germany (KP657740.1) and French (LN6813392) in the same period. Conclusion. These results showed the real circulation of EV-D68 in Senegal during this period. However, it is necessary to carry on more investigation to better understand the epidemiology and the impact of EV-D68 in the population.

Comparison of Xpert MTB/RIF assay and the acid-fast bacilli in detecting tuberculosis in Gusau, North West Nigeria B I Garba,1 A S Muhammad,2 I Yusuf,3 B A Mohammed,4 U Abdullahi5

Department Of Paediatrics, Usmanu Danfodiyo University Teaching Hospital, Sokoto, Nigeria Department Of Medicine, Usmanu Danfodiyo University Teaching Hospital, Sokoto, Nigeria Department Of Paediatrics, Ahmad Sani Yariman Bakura Specialist Hospital, Gusau, Nigeria

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS Department Of Obstetrics And Gynaecology, Ahmad Sani Yariman Bakura Specialist Hospital, Gusau, Nigeria Department of Medicine, Ahmad Sani Yariman Bakura Specialist Hospital, Gusau, Nigeria bgilah@yahoo.com

Introduction. Nigeria is among the countries with highest burden of Tuberculosis (TB) in the world. Diagnosis of TB in developing countries is mostly clinical, often relying on chest X-ray without sputum smears. Diagnosing extra pulmonary TB is more difficult due to low number of bacteria in clinical specimens. The GeneXpert MTB/RIF assay is a realtime PCR assay for the rapid diagnosis of Mycobacterium tuberculosis (MTB) and detection of rifampicin (RIF) resistance. Objective. To compare GeneXpert MTB/RIF assay with microscopy (Ziehl-Neelsen staining) in the diagnosis of TB in our hospital. Methods. A retrospective review of patients managed at the DOTS clinic of ASYBSH, Gusau, Nigeria following acquisition of the GeneXpert assay machine. Relevant information of patients managed from October 2016 to June 2017 was reviewed including smear microscopy and GeneXpert MTB RIF assay. Results. Of the 138 patients followed up at the DOTS clinic during the study period, males were 85 (61.6%), with a M:F ratio of 1.6:1. There were 27 (19.6%) children and 111 (80.4%) adults. The majority (n=118; 85.5%) had pulmonary TB and 136 (98.6%) were new cases. Only 38 (27.5%) were smear-positive while GeneXpert MTB/ RIF detected MTB in 54 (39.2%); of which 3 (2.2%) were MTB/RIF resistant, while MTB was not detected in 39 (28.3%). There was no significance between gender and AFB detection (p=0.187) nor with MTB detection (p=0.734). MTB was detected in 37 (26.8%) patients with smear-positive TB, and in 16 (11.6%) with smear-negative TB, this was significant (p=0.001). GeneXpert MTB/Rif had a sensitivity of 97.4% and specificity of 69.2%. Conclusion. Our study showed GeneXpert assay had a good sensitivity but poor specificity in detecting MTB in our centre. This test was found to be helpful in the diagnosis of TB, especially in patients who had smear-negative sputum.

Asthma exacerbations and related risk factors among patients seen in chest clinic at Tikur Anbessa Specialised Hospital in Addis Ababa, Ethiopia T W Gebremariam,1 C B Sherman,2 N W Schluger3 College of Health Sciences, Addis Ababa Univeristy Warren Alpert Medical School of School of Brown University Columbia University College of Physicians and Surgeons drtewodroshaile@gmail.com

Introduction. Asthma exacerbations are associated with increased disease morbidity and increased utilisation of healthcare resources that are often limited in low-resource countries. Identification of asthmatics at greatest risk for future exacerbations may allow a more focused costeffective approach to care for these patients. Objective. To identify risk factors for asthma exacerbations in chest clinic patients seen at the largest public hospital in Addis Ababa, Ethiopia. Methods. In this cross-sectional study, 182 consecutive patients with physician-diagnosis of asthma seen in chest clinic at Tikur

54 AJTCCM VOL. 24 NO. 1 2018

Anbessa Specialised Hospital between July and December 2015 were studied. Demographics, asthma symptoms, medication use, and asthma exacerbation in the past 12 months were obtained from the clinic records. Asthma exacerbation was defined as the self-report of worsening respiratory symptoms for greater than 48 hours. Lung function was measured using a Diagnostic EasyOne Plus model 2001 SN spirometer. The institutional review board approved the study protocol. Results. Among the 182 asthma patients in the study, the mean (standard deviation) age was 52 (12) years; 124 (68%) were female and 93 (51.1%) had at least one asthma exacerbation in the past 12 months. Patients with asthma exacerbations were more likely to be female (11.05; 95% CI 6.54 - 15.55; p=0.001), use biomass fuel (9.59; 95% CI 5.21 - 13.97; p=0.002), and have incorrect inhaler technique (6.96; 95% CI 2.59 - 11.33; p=0.008) than those without exacerbations. In the multiple logistic regression, female gender (7.93; 95% CI 2.93 - 12.94; p=0.005) and incorrect inhaler technique (6.32; 95% CI 1.55 - 11.05; p=0.012) were found to be significant. Age, comorbidities, controller medication use, and FEV1 were not found to be significant. Conclusion. More aggressive management of asthmatics that have characteristics of female gender, use of biomass fuel, and incorrect inhaler technique may lessen asthmatic exacerbations. Further prospective studies are needed to confirm these findings.

Factors associated with mortality among TB patients in Meru County, Kenya E Kanana,1 J M Ngángá2

Ministry of Health, Meru County Government, Kenya Centre for Health Solutions, USAID TB ARC eunicekanana@yahoo.com

1 2

Introduction. Tuberculosis (TB) continues to be a major cause of death in Sub-Saharan Africa despite the concerted efforts between the various players in the region to control the disease. Some of the risk factors identified include malnutrition, TB/HIV co-infection among others. Meru County is ranked 5th among the high-TB burden counties in Kenya and it’s also among the high burden counties for drug-resistant TB in Kenya. Objective. To determine some of the factors associated with mortality among TB patients in Meru County. Methods. Retrospective cohort data of TB patients who started treatment in 2016 were retrieved and analysed to determine factors associated with mortality among the TB patients and draw comparisons between the various categories of patients. Results. A total of 3 201 were started on treatment during this period, 2 307 (72%) were males, while 894 (27%) were females. Overall mortality was 4% (n=127/3 201). Mortality among the HIV coinfected patients was 8.8% (n=49/569) and 2.93% (76/2 599) among HIV-negative patients. Mortality was markedly high among the HIV co-infected patients who were not on ART when compared with those who were on ART at 18.8% v. 8.1%, respectively. Among the extra pulmonary cases, mortality was 8.3% (37/446). Mortality among the severely malnourished patients was extremely high at 23% (n=44/181) Conclusion. Mortality was found to be significantly high (8.8%) among the HIV co-infected patients as compared to HIV negative

ABSTRACTS patients. Mortality was extremely high (18.8%) among HIV-positive patients who were not on ART and among the severely malnourished patients. Special attention should be given to TB/HIV co-infected patients and the malnourished. There is urgent need to ensure that all TB/HIV confected patients are started on ART.

Respiratory health effects of sulphur dioxide air pollution from the Nyiragongo and Nyamulagira volcanoes in the Democratic Republic of Congo: A time series analysis

P De Marie Chimusa Katoto, 1 C Michellier, 2 M Dramaix, 3 B Nemery,1 F Kervyn2 Centre for Environment and Health, Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium and Department of Internal Medicine, Faculty of Medicine, Catholic University of Bukavu, Bukavu, Democratic Republic of the Congo 2 Royal Museum for Central Africa, Earth Sciences Department, Natural hazards service2, Research Centre of Epidemiology, Biostatistics and Clinical Research, School of Public Health, Université Libre de Bruxelles, Brussels, Belgium 3 Centre for Environment and Health, Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium katotopatrick@gmail.com 1

Introduction. Nyamulagira and Nyiragongo volcanoes are located in the east of the Democratic Republic of Congo. Nyiragongo volcano last erupted in 2002, impacting mainly the infrastructures with large lava flows. Regularly erupting, Nyamulagira volcano is during these periods, among the biggest emitters of volcanic sulphur dioxide (SO2) on earth. Objective. We investigated the possible temporal and spatial relationship between eruptive emissions of SO2 and acute respiratory illnesses (ARI) in surrounding populations. Methods. The total flux of SO2 emitted during eruptions since 2000 and the average spatial distribution of SO2 concentrations in the plume (2004 - 2008) were based on publicly available remote sensing data. The monthly numbers of diseases recorded as ARI among adults and children were extracted from health data collected routinely over 10 years (2000 - 2010). The monthly numbers of ARI recorded during or after eruptions were compared with those recorded before eruptions; spatial distribution was investigated according to altitude and distance from the volcanoes. Results. Seven eruptions occurred between 2000 and 2011 with discharges ranging from 0.093 - 4.5 kT SO2. Of 150 health centres, 78 provided reliable data. ARI were the second most frequently diagnosed conditions, after malaria, and their frequency appeared to increase over the years. Peaks of ARI were observed during the rainy season. No consistent temporal associations between the incidence of ARI and the occurrence or intensity of volcanic eruptions was observed, neither when the whole area was considered, nor when areas at different distances from the eruption site were considered. Conclusion. Our investigation did not allow us to link volcanic emissions of SO2 with ARI in the surrounding area. This may be due to the methodological limitations of a retrospective study that relied

on routinely collected health data, to insufficient knowledge of the size of the exposed populations and to the absence of information on ground level concentrations of SO2, which may not be high enough to affect human health. Nevertheless, Kivu volcanoes are still active, thus justifying the need for measuring the exposure of the population to SO2 and for evaluating the possible adverse respiratory effects of eruptive and continuous exposure to SO2.

Ambient air pollution in sub-Saharan Africa: A neglected risk factor for morbidity and mortality? P de Marie Chimusa Katoto,1,2 L Byamungu,3 P De Boever,4 T Nawrot,1 B Nemery1 Centre for Environment and Health, Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium 2 Department of Internal Medicine, Faculty of Medicine, Catholic University of Bukavu, Bukavu, Democratic Republic of Congo 3 University of KwaZulu-Natal, Department of Paediatric 4 Environmental Risk and Health, VITO, Mol, Belgium patrick.katoto@kuleuven.be 1

Introduction. Human development index does not accompany the massive urbanisation in Sub-Saharan Africa (SSA). Biomass fuel combustion, industrialisation and traffic lead to increased exposure to ambient air pollution (AAP). Despite a growing body of epidemiological studies linking AAP’s with deleterious effects to human health, data from SSA are still scarce. Objective. We conducted a systematic review of the literature to map the effects of AAP on the health of SSA’s population. Methods. We comprehensively searched literature in PubMed, Medline-OVID, EMBASE and Scopus databases as well as the grey literature to identify eligible studies from the inception of the electronic databases to February 7, 2017. We excluded studies assessing indoor air pollution or workplace exposures. Two reviewers independently selected studies, extracted data, and appraised studies. Results. Of the 20 studies included, data covered locations in only seven countries (out of more than 40) across SSA and more than half conducted in South Africa (SA). Most studies were communitybased cross-sectional surveys (15/20). Overall, the studies included the general population but 7 studies focused on children, 2 on elderly people, and 1 on mother-child pairs. Urban/industrialised suburbs were compared with rural/non-industrialised suburbs. Residences in close proximity to roads/mine dumps/refineries were compared to residences far away from these pollution sources. Only five studies assessed the ‘criteria air pollutants’ as defined by the USEPA. Exposure was measured via questionnaire or estimated via aggregated data and rarely personalised or measured continuously using monitoring stations. No studies, except from SA, were based on reliable morbidity or mortality statistics at the level of the region or the country. Selfreported respiratory symptoms were mostly reported. Children and the elderly were found to be more susceptible to the deleterious effect of AAP. Only half of the studies clearly defined group comparability. Conclusion. Rapid urbanisation in SSA is associated with rising levels of AAP, the exposure to which is detrimental to the health of people

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS living in the region. Our review indicates that we know little about AAP related-diseases in SSA outside SA. International and national efforts are necessary to address this neglected risk factor to improve quality of life in this region.

Epidemiology of tuberculosis in three prison health facilities in Kenya V K Kirimi

Ministry of Health, Kenya kimathi97@gmail.com

Introduction. Kenya is ranked 13th among 22 high-burden tuberculosis (TB) countries in the world. Prisoners form a group of society with a high risk of tuberculosis. In Kenya, there was an estimate of 55 000 prisoners in 2016 with a TB morbidity rate of 10%. Objective. We analysed surveillance data to describe TB cases, assess the trends, determine the TB/HIV co-infection and treatment outcomes among patients in Kenyan prison health facilities. Methods. Tuberculosis data from prison health facilities, between January 2012 and December 2015 and collected data on patient sociodemographics, treatment information and outcomes. A case of TB was defined as bacteriologically confirmed smear-positive. Data were downloaded from the TB online register and entered into EPI Info 7 for analysis. Descriptive analysis was done where means were calculated for continuous variables and proportions for categorical variables. Results. A total of 4 474 cases were analysed of which male were 3 557 (80%). The mean age was 33 (12.7) years, 1 549 (30%) were in the age group 20 - 30 years and 1 019 (23%) were from Nairobi county. Most had Pulmonary TB cases (n=3 782, 82%). Smear-positive cases were 1 536 (34%) in 2012, 1 397 (31%) in 2013 and 1 493 (33%) in 2014. Out of 2 039 smear-positive cases at month 0, 452 (22%) tested positive for HIV. Overall treatment success rate was 81%, 153 (3%) died, 9 (0.2%) had treatment failure and 475 (11%) transferred to other clinics. Out of the 9 cases with treatment failure, 3 (33%) were diagnosed with multidrug-resistant TB. Conclusion. The TB burden was high in males, there was minimal change in the trend over the three years and most cases were from Nairobi County. Our findings indicated lower treatment success rate than the WHO targets so there is a need to scale up the patient treatment follow ups.

Deploying spirometry for characterising non-communicable chronic respiratory diseases at a National TB Hospital in Tanzania: Patterns and associated factors R Kisonga,1 L Zurba,2 E Shao,1 K Mortmer,3 S Mpagama3 Kibong’oto Infectious Diseases Hospital, Kilimanjaro, Tanzania Spirometry Training Service Africa CC, Durban, South Africa 3 Liverpool School of Medicine and Tropical Diseases, UK kisonga2002@gmail.com 1 2

56 AJTCCM VOL. 24 NO. 1 2018

Introduction. Spirometry is an important tool for assessing pulmonary function in people presumed with respiratory pathology, however the test is not widely available in resource-limited settings. Objective. To describe the pulmonary function of the presumed cases of respiratory diseases after integrating spirometry in the diagnostic algorithm of respiratory diseases at Kibong’oto Infectious Diseases Hospital. Methods. Cross-sectional study in which the medical charts of patients attending the Kibong’oto National TB hospital from March to December 2017 were reviewed from. Those undergoing spirometry at the outpatient department were included, after excluding the pulmonary tuberculosis (PTB) diagnosis. Results. Baseline spirometry tests were performed in 69 patients who had a mean (SD) age of 41 (16) years. A total of 40 (58%) were male, whereas 14 (20%) had co-infection with HIV. Additional co-morbidities were severe and moderate malnutrition in 9 (13%) and 5 (7%) patients, respectively. Pastoralists, peasants and small-scale miners were 36 (52%), 23 (33%) and 4 (6%), respectively. Patients with respiratory symptoms such as cough, chest pain, difficult in breathing were 44 (64%), 39(56%) and 22 (32%), respectively. Thirty-eight had a previous history of PTB; 15 (22%) had one episode of TB treatment while 24 (35%) had twice or more episodes. Patients with normal pre-bronchodilator FEV1/ FVC were 49 (71%) while 19 (28%) had abnormal values. Abnormal radiographic findings for available films were in 58 (84%) patients; interstitial lung diseases, lung collapse, lung infection were found in 14 (20%), 6 (9%) and 14 (20%) patients, respectively. Only occupation in particular peasants and age >65 years were significantly associated with FEV1/FVC 70%; p values were 0.04 each. Conclusion. Comprehensive lung management in the primary health care in countries like Tanzania is urgently required as majority of communities are peasants.

Risk factors for pulmonary tuberculosis treatment failure in rural settings in Benin, West Africa: A cohort study

A A Kpangon, A C Dovonou, S A Amidou, S Dansou, M J Hounnouga, M A Boni Université de Parakou Service de Médecine Interne du Centre Hospitalier Départemental, Universitaire du Borgou lacteur008@yahoo.fr

Introduction. Tuberculosis (TB) remains a public health issue particularly in north-east Benin, with a high frequency of TB treatment failure. Objective. To identify the risks factors of TB treatment failure in rural north-east settings in Bembèrèkè, Benin (West Africa). Methods. This was a retrospective cohort study. We included smearpositive pulmonary TB patients who began TB treatment between 1 January 2007 and 31 January 2011 and extracted data from the TB registry. The outcomes of TB treatment were defined according to the 2007 WHO guidelines. Failure was defined as remaining smearpositive at month 5 or later during TB treatment for smear-positive pulmonary TB cases. Treatment successes were defined as being either smear-negative (cured) at month 5 (or later) of treatment or

ABSTRACTS having completed TB treatment in situations where sputum smear microscopy was not done. The deceased cases were those who died for any reason during TB treatment. For analysis we also defined composite outcomes (failure or death). After univariate analysis, multivariate analysis with 0.05 as the level of significance was done and focused on sociodemographic variables, HIV status, acid fast bacilli score at baseline. Results. A total of 264 of 270 pulmonary TB patients were included in the final analysis. The median (interquartile range (IQR)) age was 35 (28 - 46) years. Twenty-three failed on TB treatment with a frequency of 8.6% (IQR 5.5 - 12.6%). In multivariate model, positive HIV status (OR 10.38; 95% CI 1.77 - 60.91; p=0.01) and male gender (OR 4.34; 95% CI 1.03 - 18.28; p=0.046) were each significantly associated with an increased risk of TB treatment failure. Only positive HIV status (OR 12.86; 95% CI 4.27 - 38.27; p=0.0001) remained significantly associated to composite outcome. Conclusion. Positive HIV status and male gender are the potential risks factors of TB treatment failure. The association between positive HIV status and composite outcome confirmed the deadly association of TB and HIV. We need to really integrate HIV and TB activities at all levels of healthcare.

Hypersensitivity pneumonitis in an 11-year-old girl presenting to a tertiary hospital in Cape Town, South Africa

S K Owusu,1 D Gray,1 I Bandeker,2 S Chaya,1 D Marangu,1 A Ayuk,1 N A N Affendi,1A Vanker,1 K Pillay,3 M Zampoli1 Division of Paediatric Pulmonology and Red Cross War Memorial Children’s Hospital, Department of Paediatrics and Child Health and MRC Unit on Child and Adolescent Health University of Cape Town 2 Division of Radiology and Red Cross War Memorial Children’s Hospital, Department of Paediatrics and Child Health, University of Cape Town 3 Division of Histopathology and Red Cross War Memorial Children’s Hospital, Department of Paediatrics and Child Health, University of Cape Town abenaboamah18@gmail.com 1

Introduction. Hypersensitivity pneumonitis (HP), is an immunemediated inflammatory lung disease caused by inhalation of various antigenic organic particles. HP is a rare interstitial lung disease and is uncommon in children. Repeated exposure provokes an exaggerated immune response of the small airway and lung parenchyma among susceptible individuals. Methods. We present an 11-year-old girl referred to the Red Cross War Memorial Children’s Hospital, with a two-month history of cough, dyspnoea on exertion, weight loss and prolonged exposure to pigeons bred at home. Significant findings on physical examination were digital clubbing and bilateral crepitations. She had a restrictive pattern on spirometry with a vital capacity of 1.12 litres, LLN 1.50 and Z-score –3.3 SD, moderately impaired diffusion capacity of the lungs (8.7 mL/ min/mmHg), LLN 13.5 and desaturated (SaO2) to 86% on exercise testing. A chest radiograph revealed a diffuse interstitial pattern, while high resolution chest tomography(HRCT) showed extensive bilateral centrilobular opacities, honeycomb cysts and emphysematous changes of both upper lobes, as well as fibrosis with traction bronchiolectasis and localised airspace opacification of the right lower lobe . Broncho alveolar lavage showed a mixed cellular profile and positive Gene Xpert.

Lung biopsy confirmed HP and immunoglobulin G (IgG) antibody to pigeon mix was also markedly elevated (>200 mg/l; range 0.02 - 21). Results. She was commenced on oral cortocosteriods , hydroxychloroquine , antigen avoidance and anti TB medications with marked improvement in lung function and symptoms . Conclusion. This case highlights the importance of a comprehensive history, high index of suspicion and thorough investigation in diagnosing interstitial lung diseases in childhood amenable to treatment.

Case management of active tuberculosis at the Chest Clinic of Connaught Hospital in Freetown, Sierra Leone S Lakoh, D F Jiba

Connaught Hospital, University of Sierra Leone Teaching Hospitals Complex, Freetown, Sierra Leone lakoh2009@gmail.com

Introduction. Sierra Leone is among the top thirty high-TB-burden countries in the world. Major factors that contribute to this high TB burden are TB/HIV co-infection, loss to follow-up and late presentation to health facilities. Objective. To assess TB treatment outcomes and HIV coinfection at the chest clinic of Connaught Hospital in Freetown, Sierra Leone. Methods. A retrospective study design was used to collate data from treatment cards of TB patients registered at Chest Clinic in 2016. Information from 1 127 treatment cards were entered into an excel sheet, stored and analysed using SPSS version 21. Variables were summarised using mean, standard deviation, frequencies and proportion. Associations between age, sex, type of TB and HIV status were tested with treatment outcomes using χ2 and the level of significance was at 5%. Results. The predominant age group of patients treated at the centre was 25 - 34 years. Approximately 2.7% (n=31) were children below 15 years of age. Most (n=778; 69%) patients were male and public facilities were the most common (n=572; 50.8%) source of referral. Acid fast bacilli (AFB) smears were done prior to anti-TB therapy in 91.1% (n=1 027) of patients and 53.0% (n=544) of these were AFB smear-positive. Nearly two-thirds (n=340; 62.4%) showed a degree of 1+ positivity on microscopy. A significant proportion (n=993; 88.1%) of the patients were new and predominantly (n=1 024; 90.9%) in category I. Pulmonary TB accounts for 96.8% (n=1 091) of patients treated in this centre. With a treatment success rate of 57.3%, about a quarter (n=279; 25%) of the patients was lost to follow-up. Nearly all (1 105; 98%) registered active TB patients were tested for HIV. About 31.9% (352) of the patients were HIV-positive. Only 14.2 % (n=50) and 35.2% (n=124) of HIV-infected patients had documentation showing commencement of co-trimoxazole preventive therapy and antiretroviral drugs, respectively. Age, sex, type of TB, type of patient and HIV status have significant associations with treatment outcomes of TB patients. Conclusion. With a high proportion of patients lost to follow-up, especially those who were co-infected with HIV, we recommend increasing efforts in TB case management and strengthening TB/HIV collaboration at the Sierra Leone’s largest DOTS centre.

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS

Sleep-disordered breathing in stroke patients in a sub-Saharan African setting

C M Lamadine,1 M N B Hugo,¹,2 M Chamsadine,1 N Y Mapoure,1,2 C Kenmegne,2 S Mbahe,2 N H Luma,2 M S Doualla2 Faculty of Medicine and Pharmaceutical Sciences, University of Douala, Cameroon 2 Department of Internal Medicine, Douala General Hospital, Cameroon chamsadinemht@gmail.com 1

Introduction. Sleep-disordered breathing (SDB) describes a group of disorders characterised by abnormalities in the frequency and/ or depth of breathing while asleep. the most common type is the obstructive sleep apnoea/hypopnoea syndrome. Stroke is a risk factor for SDB and can also be a complication. Objective. To determine the prevalence and risk factors for SDB among patients with a past history of stroke at the Douala General Hospital (DGH). Methods. This was a cross-sectional study of stroke patients who were monitored at Doula Genereal Hospital Neurology Unit for at least 3 months. We included patients aged 15 years and above with a past history of stroke diagnosed by brain imaging. Patients with other chronic respiratory diseases, cerebral venous thrombosis, subarachnoid haemorrhage and those who were bedridden were excluded. Sociodemographic and clinical data, as well asoximetric parameters and Mallampati’s score were collected. Epworth sleepiness scale and Stopbang questionnaires were also administered. SDB was diagnosed with continuous arterial oximetry over one sleep night. SDB was defined by an oxygen desaturation index ≥5 per hour. The χ2 test was used to investigate the risk factors of SDB. Factors with a p-value <0.005 were then integrated into a multivariate logistic regression model to identify the independent factors associated with SDB. Results. A total of 110 patients were recruited. Among them, 72 (65%) were male, giving a sex ratio of 1.8. The median age was 58.50 years (interquartile range 52 - 63). The prevalence of SDB was 71.8%. Factors associated with SDB were. android obesity (p=0.001), high Mallampati score (p=0.003) and high score for the Stopbang questionnaire (p=0.017). After multivariate analysis, only android obesity (OR 3.75; 95% CI 1.41 - 9.99; p=0.008) and a high score of Mallampati (OR 4.37; 95% CI 1.15 - 16.61; p=0.030) appeared as independent associated factors for SDB. Conclusion. Three patients out of four stroke victims had an SDB. Screening and management of android obesity could reduce the burden of post-stroke sleep disordered breathing.

Clinical aspects and outcomes of haemoptysis in the respiratory unit of the Douala General Hospital, Cameroon N Y W Landry,1 M N B Hugo,2 E B C Françoise,1 L N Henry2

Faculty of Medicine and Pharmaceutical Sciences, University of Douala, Cameroon 2 Department of Internal Medicine, Douala General Hospital, Cameroon landrynguemnang@yahoo.fr 1

Introduction. Haemoptysis is the coughing up of blood coming from the lower respiratory tract. Its unpredictable nature requires

58 AJTCCM VOL. 24 NO. 1 2018

an aetiological approach and adequate therapeutic management. Objective. To describe the clinical aspects and the outcome of patients with haemoptysis in the respiratory unit of the Douala General Hospital. Methods. The study was carried out in the pneumology unit of the Douala General Hospital. It was a descriptive study including records of patients with haemoptysis, aged 15 years and above who were admitted in the unit between January 2009 to December 2016. Socio-demographical data, clinical characteristics, causes of haemoptysis and outcome were collected using a structured data collection sheet. Data were analysed using SPSS version 20 software. Ethics approval was obtained from the ethical review board of the University of Douala. Results. A total of 183 records of patients were included during the study period. The male to female sex ratio was 1.65. The mean (standard deviation) age of patients was 43.43 (16.58) years. The 25 - 34 age group was the most affected. The majority of our patients (72.7%) were living in Douala. Haemoptysis was mild in 84.2% of cases, moderate in 9.8% of cases, and massive in 6.0%. Mild anaemia was found in 50 patients (35.97%). Sputum smear for acid fast bacilli was performed in 121 patients and 44 (36.4%) had positive smears. Chest X-ray was performed in all patients and was abnormal in 94.5% of cases. Bronchoscopy was performed in 63 patients (34.4%) – it was normal in 35 patients (55.6%). The main aetiologies were pulmonary tuberculosis (33.88%), sequelae of tuberculosis (27.87%), acute bacterial pneumonia (15.85%), and lung cancers (13.93%). The outcome was marked by death in 6.01% of cases. Conclusion. Our study showed that haemoptysis affects men more than women. The 25 - 34 age group was most affected. Mild haemoptysis was predominant and accompanied mostly by mild anaemia. Pulmonary tuberculosis (active or sequelae) was the most frequent cause followed by bacterial pneumonia.

A case series report of tuberculosis patients with vitamin D deficiency in Zambia P Lungu,1 E Mubiana,1 K Mateyo,1 S Lakhi,1 P Mwaba2

Department of Medicine, University Teaching Hospital, Lusaka, Zambia APEX Medical University, Lusaka, Zambia patrickpj456@yahoo.co.uk

1 2

Introduction. An association of vitamin D deficiency with tuberculosis remains a valid assumption. It has been observed that TB is highly prevalent in certain ethnic groupings and regions of the world. Populations with darker skins are prone to vitamin D deficiency. The regions inhabited by people with darker skin coincides with high TB burden settings. Vitamin D has a key role in immune modulation of the host response to Mycobacterium tuberculosis. Studies have demonstrated early sputum culture conversion to negative, clinical recovery and radiological improvement with vitamin D supplementation. However, there is currently no consensus on the advantages of its supplementation in TB treatment. We present the first case series report of pulmonary TB patients with severe deficiency of vitamin D in Zambia. Method. We enrolled 3 participants who were selected randomly from the TB patients in admission. A questionnaire was administered

ABSTRACTS to collect information on clinical characteristics, diet and outcome at two months. A blood sample was collected for vitamin D level assessment. Results. Serum vitamin D levels for patients 1, 2 and 3 were 20.5 ng/mL (deficient), 14.6 ng/mL (insufficient) and 13.4 ng/mL. Patient 2 and 3 had prolonged hospital stays. The outcome at 2 months for patient 3 was mortality. Conclusion. The evidence so far links vitamin D deficiency with increased susceptibility to TB infection and reactivation. Severe vitamin D deficiency could be associated with poor outcome. Longitudinal studies are needed to demonstrate association, efficacy and safety of its supplementation.

Multidrug-resistant tuberculosis patients presenting with bronchiectasis and usefulness of the six-minute walk test: A case series report and literature review P Lungu,1 G Chongwe,2 V Shichizya,3 S Lakhi1

University Teaching Hospital, Lusaka, Zambia patrickpj456@yahoo.co.uk 1 Department of Medicine, University Teaching Hospital, Lusaka, Zambia 2 School of Public Health, University of Zambia, Lusaka, Zambia 3 Department of Radiology, University Teaching Hospital, Lusaka, Zambia

Introduction. Multi-drug resistant Tuberculosis (MDR-TB) is associated with extensive lung damage which impinges on the quality of life during and post-treatment. The six-minute walk test (6MWT) demonstrates significance in predicting the cardiopulmonary functional status in tuberculosis patients with bronchiectasis. We present a case series of MDR-TB patients with the multifarious manifestation of bronchiectasis and response to the 6-MWT. Methods. The study population were MDR-TB patients in the intensive phase of treatment admitted at the University Teaching Hospital, Zambia at that time. A questionnaire was administered to establish the presence of symptoms and physical examination was done to elicit signs like finger clubbing, crepitations and crackles. All patients had a high-resolution computed tomography (HRCT) of the chest done. 6-MWT was done as outlined by the American Thoracic Society guidelines. Respiratory and heart rates were monitored. A pulse oximeter was used to monitor the pulse and oxygen saturation. Results.We found bronchiectasis can occur in primary MDR-TB, which is attributed to overwhelming inflammatory response and delay in diagnosis. Mycetoma was a common complication. Conclusion. The 6-MWT was found to be useful as a bedside tool for predicting functional status. MDR-TB should be promptly diagnosed to prevent life-limiting sequelae. The findings in this case series challenge the assumption that MDR-TB is less virulent and calls for more studies to understand its pathogenesis.

Impact of Xpert MTB/RIF assay on MDRTB treatment success rates in a health District in South Africa T C Mahwire,1, 2 M Zunza,3 T Marukutira,4,5 P Naidoo6,7 1

HIV and AIDS/STI/TB Department, Port Shepstone Regional Hospital, Port Shepstone, South Africa

Division of Epidemiology and Biostatistics, Faculty of Medicine and Health Sciences, Stellenbosch University, Cape Town, South Africa 3 Centre for Evidence Based Health Care, Faculty of Medicine and Health Sciences, Stellenbosch University, Cape Town 4 Centre for Population Health, Burnet Institute, Melbourne, Australia 5 School of Public Health and Preventive Medicine, Monash University, Melbourne 6 Bill and Melinda Gates Foundation, Pretoria, South Africa 7 Desmond Tutu TB Centre, Department of Paediatrics and Child Health, Faculty of Medicine and Health Sciences, Stellenbosch University, Cape Town tcmahwire@gmail.com 2

Introduction. The Xpert MTB/RIF assay rapidly diagnoses rifampicin resistance, which enables early initiation of second line TB treatment. However, the impact of an earlier MDR-TB diagnosis on treatment outcomes is unknown. Objective. To compare MDR-TB treatment outcomes in cases diagnosed with smear/culture and Xpert®. Methods. A retrospective cohort study with cohorts defined by the diagnostic assay used in presumptive TB cases. Data were extracted from DR-TB registers including cases from January 2012 - April 2014. Treatment outcomes were assessed at recorded clinical endpoints or after two-years for those completing treatment. Results. 718 patients were enrolled into study. Cure rate (n=148) 43.4% in smear/culture group and (n=118) 33.5% in Xpert group (p=0.01). There were no significant differences between the two groups in terms of gender; age and referral facility. The smear/ culture group had a higher proportion of previously-treated TB cases (p<0.01). In the smear/culture cohort 272 of 354 (76.3%) were HIV-positive, while 271 of 345 (78%) in the Xpert cohort were HIV-positive. Treatment success rates were 54.0% (n=195) and 45.2% (n=159) for the smear/culture and Xpert cohorts, respectively (p=0.01). Xpert reduced median time to MDR-TB treatment initiation to 18.7 days from 75 days in the smear/culture group (p=0.01). Xpert diagnosis (adjusted odds ratio (aOR) 0.38; p<0.01) and male gender (aOR 0.57; p=0.02) were associated with treatment success. Xpert increased the risk of being lost to follow-up (aOR 2.55; p <0.01) and time to sputum culture conversion from 4 to 5 months (log rank test p=0.01). Time to treatment initiation was not associated with treatment success in logistic regression analysis. Conclusion. Despite rapid treatment initiation, MDR-TB treatment success rates were poorer in those diagnosed with Xpert MTB/RIF assay and in males. Additional studies are required to assess possible factors influencing DR TB outcomes.

Contribution of community health volunteers in referral of tuberculosis patients in Kenya: A validation E W Mailu,1 B Ulo,2 E Masini,3 J Ong’ang’o,4 M Kamene1 1

Ministry of Health - National Tuberculosis, Leprosy and Lung Disease Program (NTLD-P) 2 Amref Health Africa, Nairobi, Kenya 3 World Health Organization, kenya Country Office 4 Kenya Medical Research Institute (KEMRI) mailu.eunice@gmail.com

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS Introduction. Studies have shown that incorporating the services of Community Health Volunteers (CHVs) in TB control programmes can lead to early detection of TB cases, minimises the number of missed cases and ultimately helps in achieving desired health outcomes. In Kenya, all Presumptive TB cases referred by CHVs, if confirmed with TB should be recorded in the facility TB register as having been referred by a CHV. However, the proportion of TB patients referred by CHVs has remained constantly very low at 4% for the last five years despite the intensive investment under the support of Global Fund. Objective. To determine the actual proportion of notified TB patients that are referrals by CHVs and identify the factors that may be contributing to their incorrect recording and reporting in Kenya. Methods. This was a cross-sectional study of patients who were in the intensive phase of treatment for drug sensitive TB in Kenya conducted between January and April 2017. Data were collected using a pre-designed mobile phone electronic based questionnaire in KOBO collect. Data were analysed using SPSS and descriptive analysis was conducted to get the proportion of patients referred by CHVs. Inferential statistics were also conducted to determine the factors contributing to incorrect documentation of patient referrals, while multivariate analysis was used to control for confounders. Results. The proportion of TB patients referred by CHVs was 18%. The first point of entry at the health facility was found to be a factor for correct or incorrect recording and reporting. At every entry point of the health facility, the proportion of incorrect recording was higher than correct recording with an overall incorrect recording of 72%. The odds of being recorded correctly if the TB clinic was the first entry was (1.859; 95% CI 1.003 - 3.446). Out of the 355 patients who said they were referred by a CHV, only 24.8% (n=88) were notified to the national TB program as referral by CHVs. Conclusion. The proportion of TB patients referred by CHVs in Kenya is higher than what is usually reported. Incorrect reporting is associated to health system related problems contributing to poor documentation at all levels. However, TB clinic being the first point of patient entry when referred by a CHV has emerged to be more beneficial in ensuring correct reporting. The greatest loss of documenting CHV referrals is during notification to the national TB programme.

Predictors of IPT non-completion in HIVinfected patients in two high-volume faith-based health facilities in Kenya J M Manthi,1 S Nakitare,1 L Mwaniki,1 J Mecha,2 C Njigua1 Christian Health Association of Kenya (CHAK), Kenya University of Nairobi, Kenya jmanthi@chak.or.ke

1 2

Introduction. Tuberculosis (TB) is the leading cause of death among people living with HIV (PLHIV). Administration of isoniazid preventive therapy (IPT) reduces the risk of developing active TB among PLHIV in Kenya, which is a high-TB-burden country. In line with the WHO guidelines, Kenya rolled-out the administration of 6-month IPT to all eligible PLHIV in 2014. Completion rates have been sub-optimal, with paucity of studies to ascertain why. CHAP Uzima is a CDC funded HIV care and treatment program working in faith-based and affiliated health facilities in Kenya. Objective. To assess the patient level predictors of IPT non-completion among PLHIV.

60 AJTCCM VOL. 24 NO. 1 2018

Methods. This was a retrospective cohort analysis of routinely collected programme data in two high volume CHAP Uzima supported health facilities. We included all TB-free HIV infected adults and children aged above 2 years who were initiated on IPT between July 2015 and July 2017. Data were extracted from an electronic medical records system. Multivariate analysis was carried out to determine predictors of IPT non-completion. Results. Of the 3 563 patients initiated, 756 (21%) did not complete IPT. Median time to non-completion was 3 months (IQR 3 - 4). Predictors of non-completion were age below 25 years (OR 1.5; 95%CI 1.157 1.907; p=0.002), having detectable viral load (OR 3.5; CI 2.33 - 5.17; p=0.005,) and currently being on second-line antiretroviral therapy (OR 1.2; CI 1.01 - 1.30; p=0.04). Patients aged 15 - 19 years had the highest non-completion rate (31%). There was no statistical difference by baseline CD4 cell count for patients with IPT non-completion (OR 0.951; CI 0.802 - 1.127; p=0.563,). Conclusion. Among PLHIV initiating IPT in Uzima CHAP-supported health facilities, non-completion is more likely in patients who are younger and more so adolescents, patients on second line therapy and those classified as virally unsuppressed. Programs initiating IPT should consider paying closer attention to these patient groups including enactment of patient treatment preparation and enhanced adherence counselling prior to IPT initiation. More studies on the specific reasons for non-completion are recommended.

Peripheral arterial disease among patients with chronic obstructive pulmonary disease attending the chest clinic at a tertiary hospital in Nairobi, Kenya J Mecha, M Wahinya, J O Mecha, G E Oyoo, E N Ogola University of Nairobi School of Medicine jared.mecha@uonbi.ac.ke

Introduction. Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality worldwide. Cardiovascular disease (CVD) has been shown to be the leading causes of death among COPD patients. Peripheral arterial disease (PAD) is an atherosclerotic process that is characterised by an increased risk of coronary and cerebrovascular ischaemic events. PAD has been shown to occur with greater frequency among COPD patients compared to the general population. There are no data on the burden of PAD among COPD patients in Kenya. Objective. To determine the prevalence of PAD and the prevalence of some associated cardiovascular risk factors among COPD patients at a tertiary urban hospital in Nairobi, Kenya. Methods. Patients with a spirometry diagnosis of COPD attending the chest clinic in KNH were consecutively recruited until the minimum desired sample size of 78 patients was achieved. Anklebrachial index (ABI) was measured using a handheld Doppler device, and blood samples were drawn to analyse for fasting lipid profile, fasting blood sugar and high-sensitivity C-reactive protein (hsCRP). Results. The overall prevalence of PAD was 7.5%. Hyperlipidaemia was the most common cardiovascular risk factor at 47.5%, followed by hypertension at 46.3% and diabetes at 8.8%. Twenty-one percent of the patients were obese, and 96.3% of the patients had hsCRP levels >3 mg/L. Only two patients had symptomatic PAD.

ABSTRACTS Conclusion. The study demonstrated a low prevalence of PAD and a high prevalence of PAD-associated cardiovascular risk factors among our COPD patients. Of the patients who had PAD, only two were found to be symptomatic.

Prevalence and associations of non-smoking chronic obstructive pulmonary disease at a tertiary hospital in Nairobi, Kenya

J Mecha,1 J Bwombengi,1 G Nyale,2 J Kayima,1 J O Mecha1 University of Nairobi School of Medicine Kenyatta National Hospital jared.mecha@uonbi.ac.ke

Introduction. Non-smokers comprise a substantial proportion of patients with chronic obstructive pulmonary disease (COPD). Its pathogenesis is poorly understood and documented. No previous studies have been conducted to evaluate its prevalence and associated risk factors in Kenya. Objective. To establish the prevalence of non-smoking COPD among patients on follow-up for COPD at a tertiary level hospital in Kenya. Methods. We conducted a cross-sectional study among patients followed up with a diagnosis of COPD at a chest clinic at a tertiary level facility. Spirometry was performed on consecutive participants with a diagnosis of COPD. Those meeting the criteria for fixed airway disease (FEV 1/FVC 70% predicted) completed an intervieweradministered questionnaire to identify selected risk factors for COPD on the basis of previous studies. We defined non-smoking COPD as a post-bronchodilator FEV1/FVC) 70% in a person who had not smoked tobacco. Results. Between February and May 2016, 84 patients satisfied the inclusion criteria and consented to participate in this study. Males accounted for 72.6% of the participants. The prevalence of non-smoking COPD was 39.3%. The main risk factors in these participants were poorly controlled asthma (66.7%), exposure to biomass fuel (81.8%), history of childhood respiratory tract disease (40.6%) and history of tuberculosis (36.4%). Conclusion. Non-smoking COPD is common and is associated with low socioeconomic background with a strong history of biomass fuel exposure (from poor ventilation in the cooking area), recurrent childhood pulmonary infections, past tuberculosis, and poorly controlled or chronic asthma.

Diagnostic value of a lateral flow-urine lipoarabinomannan assay in adults with suspected pulmonary tuberculosis at two urban hospitals in Nairobi, Kenya J Mecha, C Okoth, J O Oyugi, T M Munyao University of Nairobi School of Medicine jared.mecha@uonbi.ac.ke

Introduction. Tuberculosis (TB) remains a leading cause of morbidity and mortality globally, with low and middle-income countries being disproportionately affected.

Objective. To determine the diagnostic value of the lateral flow urinary lipoarabinomannan (LAM) assay as a point-of-care diagnosis assay among adult patients with active pulmonary TB. Methods. The study population consisted of ambulatory and hospitalised patients being investigated for active pulmonary TB at two urban hospitals in Nairobi, Kenya. A total of 241 consecutively sampled adults presenting with features of pulmonary tuberculosis were included. Urine samples were obtained for the ULAM assay and sputum samples for Ziehl-Nielsen (ZN) microscopy, Xpert MTB/ Rif test and liquid TB culture. All participants received counselling and testing for HIV infection. CD4+ cell counts were determined for HIV positive patients. Sensitivity, specificity, positive and negative predictive values of the TB diagnostic tests were computed; the liquid culture was used as the gold standard. Results. Sensitivity of the urine LAM assay was low when applied to a heterogeneous population, 28.6% (95% CI 20.6 - 38.3%). The sensitivity in the HIV-negative patients was, 12.7% (95% CI 6.4 - 23.5%) compared with 58.8% (95% CI 42.2 - 73.6%) in HIV infection. When combined with sputum ZN microscopy, then with sputum XpertMTB/Rif test, the sensitivity increased to 70.6% and 96.2%, respectively. In HI-infected cases, sensitivity of the urinary LAM assay increased as the CD4+ count decreased being 76.5%, (95% CI 52.1 - 90.8%); 65.4% (95% CI 46.1 - 80.6%) and 33.3% (95% CI 6 . 2 - 7 9 . 5 % ) at C D 4 c e l l c ou nt s of 5 0 an d 2 0 0 cells/μL respectively. Specificity of the urine LAM assay was lower when applied to the HIV-positive patients compared with the HIV-negative population (85.1%; 95% CI 74.4 - 91.8% v. 93.6%; 95% CI 86.4 - 97.3%, respectively). Conclusion. The lateral flow urine LAM assay, as an easy to perform, point-of-care test, can contribute to improvement in case detection of pulmonary TB especially in TB/HIV co-infected cases with severe immunosuppression (CD4+ counts of ≤200 cells/μL).

Assessment of sputum smearpositive but culture-negative results among newly diagnosed pulmonary tuberculosis patients in Tanzania

N Mnyambwa,1,2 E S Ngadaya,2 K Godfather,2 D-J Kim,1 R Kazwala,3 P Petrucka,1,4 S G Mfinanga2 School of Life Sciences and Bioengineering, Nelson Mandela African Institution of Science and Technology, Arusha, Tanzania 2 National Institute for Medical Research, Muhimbili Medical Research Centre, Dar es Salaam, Tanzania 3 Faculty of Veterinary Medicine, Sokoine University of Agriculture, Morogoro, Tanzania 4 College of Nursing, University of Saskatchewan, Saskatoon, Canada lodnicho@gmail.com 1

Introduction. Diagnosis of pulmonary tuberculosis (TB) in technology-limited countries is widely achieved by smear microscopy, which has limited sensitivity and specificity. The frequency and clinical implication of smear-positive but culture-negative among presumptive TB patients remain unclear. Methods. A cross-sectional study was conducted to identify the proportion of nontuberculous mycobacteria (NTM) infections among 94 smear-positive culture-negative patients diagnosed between

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS January 2013 and June 2016 in selected health facilities in Tanzania. Results. Out of 94 sputa, 25 (26.60%) were GeneXpert TB-positive and 11.70% (n=11/94) were repeat-culture positive; 5 were Capilia TBNeo positive and confirmed by GenoType MTBC to be Mycobacterium tuberculosis/Mycobacterium canettii. The remaining 6 Capilia TB-Neo negative samples were genotyped by GenoType CM/AS, identifying 3 (3.19%) NTM, 2 Gram-positive bacteria, and 1 isolate testing negative, which produced a total of 6/94 (6.38%) confirmed false smearpositives. Twenty-eight (29.79%) were confirmed TB cases, while 60 (63.83%) remained unconfirmed cases. Out of 6 (6.38%) patients who were HIV-positive, 2 patients were co-infected with mycobacteria. Conclusion. The isolation of NTM and other bacteria among smearpositive culture-negative samples and the presence of over two-thirds of unconfirmed TB cases emphasise the need of both advanced differential TB diagnostic techniques and good clinical laboratory practices to avoid unnecessary administration of anti-TB drugs.

Management of bronchiolitis in an HIV endemic area: Are Standard Treatment Guidelines being followed? R Mohee,1 K Naidoo,1,2 V Naidoo,2 R Masekela2

Department of Paediatrics and Child Health, Nelson R Mandela School of Medicine, UKZN 2 University of KwaZulu-Natal, Nelson R Mandela School of Medicine rachnadevimohee@gmail.com 1

Introduction. Bronchiolitis is a common cause of lower respiratory tract infection in children. Standard Treatment Guidelines (STG) have been developed to optimise bronchiolitis care at hospital level. The extent of its use and factors influencing guideline adherence is not known in South Africa. Objective. To determine whether the STG for the bronchiolitis management were utilised in the management of children admitted with bronchiolitis at King Edward VIII Hospital, Durban, KwaZuluNatal. Methods. This study was a retrospective chart audit of children admitted between 1 January 2015 and 31 December 2015. Data including demographics and treatment modalities were collected. Guideline adherence assessment was based on the protocolled management of bronchiolitis as stipulated by STG 2013. For categorical variables (HIV exposure, nutritional status), subgroup comparisons according to adherence were made using Ď&#x2021;2 and Fisherâ&#x20AC;&#x2122;s exact tests. The Wilcoxon Rank Sum test was used for testing associations between length of hospital stay and protocol adherence. Results. A total of 192 infants were enrolled in the study. Of these, 66% were HIV-exposed. The majority (92%) were well-nourished. Full adherence to STG was found in 24% of participants. There was no association between adherence to STG and HIV-exposed/unexposed status ot nutritional status respectively(p>0.05). The mean length of hospital stay was not significantly different depending on adherence or non-adherence 5.1 versus 4.3 days, respectively. One death was reported in the non-adherent group. Conclusion. There is poor adherence to the STG for bronchiolitis, though this did not impact morbidity and mortality. Future multi-centre studies with the recently published bronchiolitis recommendations assessing adherence to protocols with pre- and

62 AJTCCM VOL. 24 NO. 1 2018

post-intervention studies after training of personnel, may provide more evidence of the implications of lack of guideline adherence.

A case of haemoptysis in a girl with Noonan syndrome R K Mopeli, C Verwey, M Lebea, P Adams

Faculty of Health Sciences, University of Witwatersrand. Chris Hani Baragwanath Hospital, Johnnesburg, South Africa refiloe.mopeli@wits.ac.za

Introduction. Haemoptysis is the expectoration of blood originating from the lower respiratory tract. It is uncommon in children, but can be life threatening. The most common causes are respiratory tract infections, aspirated foreign bodies and bronchiectasis. Noonan syndrome (NS) is an autosomal dominant condition characterised by distinctive facial features, congenital heart disease and multiple comorbidities including haematological abnormalities. Bleeding disorders have been reported in up to 65% of patients with NS. Perfusion of the lower respiratory system arises from the pulmonary arterial circulation and the bronchial circulation, and bleeding may arise from either. Methods. Case report. We describe a 7-year-old girl with NS who presented with recurrent episodes of haemoptysis from 6 years of age. She was known to have pulmonary stenosis, with transannular patch repair, and post-operative severe pulmonary regurgitation. Initial work-up suggested a diagnosis of Von Willebrand disease, as her Von Willebrand factor activity was 37%, with an acute pneumonia as the cause of haemoptysis. Chest X-ray and a computed tomography scan showed a right lower lobe (RLL) dense consolidation. Flexible bronchoscopy revealed an inflammatory polyp with 75% obstruction of the orifice of the RLL. The inflammatory polyp and lobar pneumonia were then thought to be the cause of the haemoptysis, which was exacerbated by her underlying haematological abnormality. She was treated with antibiotics, however the haemoptysis persisted. She underwent cardiac catheterisation and an angiogram showed a tortuous right bronchial artery forming a confluence with an abnormal vessel arising from the right common carotid artery with extravasation of blood into the RLL. She had a successful right bronchial artery embolisation and was stable post-procedure. After a further rigid bronchoscopy the patient had another episode of haemoptysis. Ultimately a lobectomy was performed and no further bleeding reported. Conclusion. Although this patient had NS with an increased risk of bleeding, the cause of her haemoptysis was an abnormal bronchial artery supplying blood to the right lower lobe. This case illustrates the importance of being systematic when investigating any patient with haemoptysis, as the cause may be unrelated to the underlying diagnosis.

Prevalence of respiratory symptoms among small-scale wood furniture workers in the Zimbabwe informal sector industry S Muteti-Fana, V Chikwasha, J January

Department of Community Medicine, College of Health Science, University of Zimbabwe smuteti@gmail.com

ABSTRACTS Introduction. Occupational exposure to wood dust has been implicated in respiratory health problems ranging from impaired lung function, chronic bronchitis and asthma. In Zimbabwe, there are numerous small-scale wood furniture workers operating under poor working conditions due to economic hardships. These workers are exposed to multiple occupational hazards such as wood dust, cotton dust and environmental pollution. There is no morbidity and mortality data on respiratory health of wood furniture workers in Zimbabwe. Objective. To determine the most common respiratory symptoms among small scale wood workers and also estimate exposure levels. Methods. A descriptive cross-sectional survey was conducted among small-scale wood furniture workers at Glenview Home Industry. A modified British Medical Research Council interviewer administered questionnaire on respiratory symptoms (cough, phlegm, wheezing and breathlessness), employment history and smoking habits was used for data collection. Total inhalable wood dust samples were collected using 3-peice dust collectors with 37 mm glass fibre filters. Five area (static) samples and 5 personal samples were collected at strategic points over a minimum period of 4 hours and analysed by gravimetric methods. Results were analysed using Stata version 13. Results. A total of 161 participants were recruited with median (interquartile range) age of 25.4 (22.5 - 31.4) years. The prevalence of respiratory symptoms was 77.6% (95% confidence interval (CI) 71.1 - 84.1). The prevalence by individual symptom was: cough 52.2% (95% CI 44.4 - 60.0); phlegm 49.7% (95% CI 41.9 - 57.5), wheezing 41.6% (95% CI 33.9 - 49.3), and breathlessness 21.7% (95% CI 15.3 - 28.2). Comparatively, personal wood dusts samples were higher than area wood samples. Out of the 5 personal samples, the minimum total dust concentration was 0.6 mg/m3 and the maximum was 26.2 mg/m3. For the 5 area samples, the total dust concentration ranged from 0.6 - 11.7 mg/m3. Conclusion. Prevalence of respiratory symptoms among smallscale wood furniture workers is very high. Majority of findings from exposure samples (both area and personal) showed levels above the set American Conference of Government Industrial Hygienists Occupational Exposure Limits (2014) for soft wood dust. Results suggest the need for targeted interventions that focus on the reduction of wood dust exposure among wood furniture workers.

Health-seeking behaviour among individuals with cough symptoms at regional referral hospitals in Uganda: Missed opportunity for early tuberculosis diagnosis

W Muttamba,1 W Ssengoba,2 B Kirenga,2 A Katamba,2 M L Joloba2 Lung Institute, Makerere University, Kampala, Uganda Makerere University, Kampala, Uganda muttamba@gmail.com

1 2

Introduction. Studies on delays in the diagnosis of tuberculosis (TB) in Africa have revealed important patient-related factors, as well as health system inefficiencies. Objective. To assess the health-seeking behaviour among individuals who were presumptive TB cases presenting with a cough at regional referral hospitals.

Methods. A cross-sectional study of adult presumptive TB patients conducted from October 2015 - December 2016 at five regional referral hospitals in Uganda. All study participants were interviewed about TB symptoms, health-seeking behaviour following cough symptoms, and had a GeneXpert test done. Results. Of the 1 862 participants interviewed, the majority (n=1 795; 99.9%) reported cough as a symptom, followed by fever (n=1 223; 68%), weight loss (n=1 192; 66.4%), night sweats (n=1 161; 64.6%) and haemoptysis (n=235; 13.1%). Of the respondents, 75% (n=1 352) had sought care for their cough and this was mainly at public health facilities (60%), followed by private health facilities (21.4%), and drug stores/pharmacies (13.5%). Of those that sought care at public health facilities, only 27.5% were asked to provide a sputum sample. Only 13.5% of those that sought care from a private health facility were asked to provide a sputum sample. The estimated crude odds ratio of the association between seeking care at a private health facility and positivity on GeneXpert was 1.5 (95% confidence interval 1.1 - 1.9; p=0.011) Conclusion. Cough is the main symptom for presumptive TB patients at regional referral hospitals. The study revealed there are still health system inefficiencies for patients who make an attempt at seeking care that could result in delayed diagnosis especially in instances where patients are not asked to provide a sputum sample for testing. Improved TB diagnosis at first contact with the healthcare system has the potential to increase TB case finding and break the cycle of transmission in the community.

Early outcomes of minimally invasive video-assisted thoracoscopic decortication: A single-centre initial experience K Naidoo,1 J K Koshy,2 A Ogunrombi,2 V Sididhza1 Department of Surgery, Klerksdorp/Tshepong Hospital University of Witwatersrand kaylin.naidoo.10@gmail.com 1 2

Introduction. The role of video-assisted thoracic surgery (VATS) approach in inflammatory thoracic conditions has not been widely accepted. The adoption of this approach by sporadic units has shown itsâ&#x20AC;&#x2122; effectiveness in the management of advanced stages of empyema. Objective. To show that the aetiology and stage of empyema does not significantly alter the outcomes of minimally invasive decortication. Methods. We retrospectively reviewed 34 patients who underwent minimally invasive clear-out and decortication for empyema at the Klerksdorp/Tshepong Hospital cardiothoracic unit between October 2015 and November 2017. Results. We performed a total of 37 VATS decortications on 34 patients; the male:female ratio was 3.1. Of the cases, 67% were due to trauma, 32% were non-traumatic (66% were due to TB and the rest were post-pneumonic) and there was a case of complicated liver abscess. Of the two aetiologic subgroups, 66% of the non-trauma and 28% of trauma group were HIV-positive. In the non-trauma group, 33% had active TB and 33% had a previous history of TB. The early post mortality rate for the entire cohort was 0%. There was

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS no statistically significant difference between the trauma and nontrauma groups in the operative time (131 (51) min v. 135 (5) min; p=0.44), mean blood loss (732 mL v. 894 mL; p=0.26), conversion rate (8% v. 8.3%), reoperation rate (16% v. 0%; p=0.28) and hospital stay (15 (14) days v. 15 (12) days; p=0.49), respectively. The majority of conversions were in those needing reoperation. All the reoperations were in the trauma group and a subgroup analysis showed that the reoperation rate was not significantly different (p=0.55) in those with stage II empyema (16%) v. stage III (0%). Conclusion. Minimally invasive clear-out of the pleural space and decortication of the lung is a technically feasible operation regardless of the aetiology or stage of empyema. Our sample size constitutes a small cohort and an early experience and we therefore plan to conduct a prospective study to further evaluate this surgical approach.

Health workers᾽ practices in assessment and management of children with respiratory symptoms in primary care facilities in Jinja district Uganda: A descriptive study R Nantanda

Makerere University Lung Institute, Kamapala, Uganda rnantanda@gmail.com

Introduction. Asthma is the most common chronic childhood condition worldwide, with increasing prevalence in middle- and low-income countries. However, asthma is largely under-diagnosed, particularly in children younger than 5 years of age. Diagnosis of childhood asthma is largely reliant on good history and physical examination. Objective. To describe the health workers’ practices in diagnosis and management of respiratory illnesses among children, with emphasis on asthma, in rural primary care centres in Uganda. Methods. Health workers’ clinical practices were observed during consultations with children under 5 years of age, who presented with cough and/or difficult breathing. A short interview with the caregiver, as well as a short health facility survey, was conducted following the consultation. Data were analysed using descriptive statistics. Results. Fifty health workers were observed during 220 consultations at six different health centres. The average consultation time was 4 minutes (interquartile range 3 - 5). The key symptoms of asthma, recurrent cough, difficult breathing and wheezing, were elicited in only 5% of the consultations. The respiratory rate and chest indrawing were assessed in only 10% of consultations. Pneumonia and asthma were diagnosed in 16.5% (n=36) and asthma in 0.5% (n=1) of the consultations. Antibiotics were prescribed to 32% of all the children but to only 39% of the children diagnosed with pneumonia. In the majority (95%) of consultations, health workers did not explain the diagnosis and management plan to the caregivers. Conclusion. Clinical practices among Ugandan health workers in primary care are insufficient to aid correct identification of asthma and other respiratory diseases in young children. Irrational use of antibiotics is widespread. Interventions to improve the health workers’ awareness, knowledge and skills for diagnosis and management of asthma are urgently needed.

64 AJTCCM VOL. 24 NO. 1 2018

Whole-genome sequencing of Mycobacterium tuberculosis directly from sputum identifies more minority variant mutations than sequencing from culture C Nimmo,1,2 R Doyle,1 R Williams,1 J Breuer,1 A Pym2

Division of Infection and Immunity, University College London, UK Africa Health Research Institute, Durban, South Africa camus.nimmo@ahri.org

Introduction. Mycobacterium tuberculosis high-throughput wholegenome sequencing (WGS) reveals the presence of minority genetic variant (MGV) single nucleotide polymorphisms (SNPs) representing mycobacterial subpopulations within individual patients. WGS is usually performed on cultured isolates, even though this can alter the original population structure, because WGS directly from sputum without enrichment yields insufficient DNA for deep genome coverage. SureSelect oligonucleotide enrichment technology (Agilent, USA) can obtain purified M. tuberculosis DNA directly from sputum. Here, we compare MGVs identified after sequencing from enriched sputum and culture. Methods. Paired sputum samples from 36 patients (22 with drugsensitive tuberculosis (DS-TB) and 24 with rifampicin-resistant (RR) TB) were analysed. DNA was extracted directly from one sample, with the other inoculated into MGIT (mycobacterial growth indicator tube) until it flagged positive. DNA from sputum samples underwent SureSelect enrichment. All samples were sequenced on a NextSeq. Bioinformatics analysis was performed using CLC Genomics Workbench v10. Mapping was performed to H37Rv and MGVs called when more than one nucleotide was reported at a genetic location, each with minimum 10 supporting reads including one in each direction. Variants in or near PE/PPE regions were excluded. Results. More than twice as many MGVs were identified in directly sequenced sputum samples than in MGIT samples (1 547 v. 632). Direct sputum samples contained 684 intergenic, 370 synonymous and 493 non-synonymous MGVs, while MGIT samples contained 204, 188 and 240, respectively. The mean coverage was similar between MGIT and sputum samples (191.1 v. 184.3; p=0.89). Sequencing directly from sputum identified median 24 MGVs per sample compared with 10 in MGIT (p<0.001), and more than twice as many MGVs across the dataset. Conclusion. Sequencing directly from sputum identifies more MGVs compared with the MGIT. This contradicts the findings of one other study that has sequenced directly from sputum, which may be due to greater coverage depth and patient numbers in our study. Our findings support data from other studies, which suggest that subculture leads to a loss of MGVs. Direct sputum sequencing may better represent true mycobacterial genetic diversity within patients and have a rolein investigating heteroresistance.

Metastatic thyroid carcinoma of the lung – a case report N A N Affendi,1 S K Owosu,1 K Pillay,2 M Zampoli1 1

Division of Paediatric Pulmonology and Red Cross War Memorial Children’s Hospital, Department of Paediatrics and Child Health, and MRC Unit on Child and Adolescent Health, University of Cape Town, South Africa

ABSTRACTS Division of Histopathology and Red Cross War Memorial Children’s Hospital, Department of Paediatrics and Child Health, University of Cape Town, South Africa noor.ain.noor.affendi@gmail.com

Introduction. Thyroid carcinoma is rare in children and accounts for only 0.5 - 3.0% of childhood malignancies. It commonly presents as a painless neck nodule. The clinical behaviour is aggressive in children with regional nodal metastasis occurring in 60 - 80% and 20% of cases, respectively, at presentation. Methods and results. We discuss a 9 year-old-girl who presented with severe respiratory distress and hypoxia, with 3 months’ history of weight loss and chronic cough. On examination, she was cachexic with proptosis. There was bilateral hard-matted cervical lymphadenopathy. Chest X-ray showed a diffuse reticulonodular infiltrate pattern without lymphadenopathy. Presumptive diagnosis of disseminated tuberculosis was made before lymph-node histology confirmed papillary thyroid carcinoma. Unfortunately, she succumbed after 6 days of admission and mechanical ventilation. Post mortem tru-cut biopsy of both lungs confirmed metastatic thyroid carcinoma of the lung. Conclusion. Lung metastatic disease, although rare, should be considered in the differential diagnosis of a child who presents with a reticulonodular pattern infiltrate on chest radiograph as prognosis will be poor without early diagnosis and treatment.

Effect of wood smoke on the respiratory health of workers in a semi-urban community in Niger Delta, Nigeria I P Obiebi, P G Oyibo

Delta State University Teaching Hospital, Nigeria irikefewhite@gmail.com

Introduction. The process of charcoal production directly exposes workers to wood smoke. Persistent inhalation of wood smoke causes irritation of the respiratory tract, precipitates respiratory diseases and exacerbates symptoms of pre-existing conditions. Objective. To assess the effect of wood smoke on the respiratory health of charcoal workers in a semi-urban community in Niger Delta, Nigeria. Methods. This cross-sectional comparative study involved an equal number of traders who were matched with charcoal workers for age, height and sex with charcoal workers. A modified version of the British MRC questionnaire on chronic work-related respiratory symptoms among workers was employed to assess respiratory symptoms. Indices of lung function capacity were measured with a hand-held spirometer. SPSS was used for analysis. Odds ratio, McNemar’s, paired-t, and χ2 tests were performed to test the significance between exposure and outcome. Logistic regression was used to adjust for smoking, domestic biomass use and age. Results. Charcoal workers had a higher prevalence of respiratory symptoms than controls: chronic cough (9.5% v. 0.0%), productive cough (13.5% v. 3.6%), breathlessness (19.6% v. 13.5%), nasal discharge (34.9% v. 16.2%), chest tightness (8.8% v. 0.0%) and wheeze (8.8% v. 5.4%), respectively. There was no significant association between work duration and job description, and respiratory symptoms (p> 0.05), although workers were more likely to have chronic cough,

productive cough, wheeze, breathlessness, chest tightness and nasal discharge. Only wheeze was significant after adjusting for age, biomass use and cigarette smoking (odds ratio 4.22; 95% confidence interval 1.37 - 12.99). More charcoal workers had COPD (9.5%) than occupational asthma (6.7%); no control had these conditions. Predicted values of FVC, FEV1, FEV1/FVC ratio and PEFR were higher among controls than charcoal workers; however, the difference was not significant (p>0.05). The mean values of FEV1 and FVC were considerably lower for workers, whereas FEV1/FVC ratio and PEFR were higher among workers (p=0.05). Conclusion. Respiratory symptoms and diseases were more prevalent among charcoal workers who also had reduced lung function capacity. Instituting interventions to reduce workers’ exposure would be an all-important course to pursue if they are to remain healthy and in business.

Air quality at charcoal kiln sites in a developing nation in sub-Saharan Africa I P Obiebi, P G Oyibo, G Eze

Delta State University Teaching Hospital, Niger Delta, Nigeria irikefewhite@gmail.com

Introduction. In Nigeria, at least 1 in 10 000 people die from diseases caused or worsened by air pollution each year. Charcoal production is majorly fraught with emission of wood smoke, which significantly pollutes air; even fine particles in the smoke can be persistently suspended in air and be inhaled by unsuspecting persons in the environment. Such pollutants are known to instigate and complicate respiratory diseases culminating in deaths in those affected in Niger Delta. Objective. To estimate the concentration of air pollutants at charcoal production sites in a community in Niger Delta, Nigeria to highlight how far they deviate from WHO air quality standards. Methods. This was a cross-sectional study in which air quality was assessed at charcoal kiln sites with a hand-held air tester (model CW-HAT 200) to measure particulate matter concentration; and gases were measured with environmental sensor kits (Z-1300(SO2), Z-900(H2S), Z- 1200(O3), Z-700(NO), and Z-1500(NH3)). Analysis was done with SPSS and ANOVA compared mean differences in air pollutants. Air quality index was calculated using PM2.5 because it is a pollutant majorly derived from combustion of wood. Results. The maximal PM 2.5 and PM 10 values at the kiln sites ranged from 20 - 1 064 µg/m3, and 23 - 507 µg/m3 respectively. At the majority (83.3%) of the sites, PM2.5 and PM10 were higher than WHO standards. The mean concentrations of PM2.5 and PM10 were 146.58 µg/m3 and 359.33 µg/m3, respectively, and were 5 times more than WHO standards. The air quality at one third (33.3%) of all the sites was very unhealthy but highly hazardous at a site; only two sites had moderately healthy air quality. Average concentrations of ozone, hydrogen sulphide and ammonia at charcoal production sites were significantly higher than within 100 m of the sites and 500 m away. However, nitric oxide was highest within a 100 m of the sites. The average concentration of sulphur dioxide was higher than WHO standard. Conclusion. The concentrations of air pollutants from charcoal

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS production are enormous and deviate from acceptable limits. Charcoal workers would need to use improved devices which emit lesser amounts of toxic substances for their long-term health benefit.

Outcomes of bullectomy in a South African patient cohort C Ofoegbu, N Swai

Chris Barnard Division of Cardiothoracic Surgery, University of Cape Town, South Africa chimaofu@yahoo.com

Introduction. Bullous lung disease is not uncommon in the South African population and commonly presents as a spontaneous pneumothorax (SP). Objective. To ascertain the indications for bullectomy and short-term outcomes of the procedure. Methods. An ongoing retrospective study of patients who underwent bullectomy between 2011 and 2017 at Groote Schuur Hospital. Preoperative data (demographics and symptomatology), intraoperative and postoperative data were collected. Results. Records of 54 patients have been reviewed thus far. The male to female ratio was 2:1 with a mean (standard deviation (SD)) age and mean (SD) body mass index of 44 (13.4) years and 21.8 (4.5) kg/ m2, respectively. Previous pulmonary tuberculosis (TB) and chronic obstructive pulmonary disease were seen in 44.4% and 33.3%, respectively. Most of the patients (57.4%) had a Modified Medical Research Council Scale of 1, while 46.3% were current smokers. Most patients (65%) presented with a bronchopleural fistula (air leaks) with a mean (SD) duration of 10.2 (13.4) days and the most common preoperative diagnosis was secondary SP (50%). Operative approach was by video-assisted surgery (n=32; 59.3%) or thoracotomy (n=22; 40.7%), with 6 patients being converted to thoracotomy (18.7%). Operative finding was stage IV bullae (> 2cm) in 70% of patients. Pleurectomy accompanied bullectomy in 78% of patients. The mean (SD) duration of the operation was 95.6 (39.8) minutes; 121 minutes for thoracotomy and 77.8 minutes for VATS. There was a high complication rate of 50% with air leaks in 92.5% (n=25/27) of patients and a mean (SD) chest tube duration of 7.5 (5) days. The mean duration of intensive care unit stay was 22 hours and this was mainly for postoperative epidural analgesia. None of the patients required postoperative mechanical ventilation. The mean (SD) postoperative hospital stay was 9.7 (6.6) days. There were no postoperative deaths. The mean (SD) follow-up duration was 1.65 (1.68) months. Conclusion. Bullectomy/pleurectomy is a safe and effective treatment for SP secondary to bullous lung disease, with a low recurrence rate.

Prescription costs in two tertiary hospitals in mid-Western Nigeria S Oghuvwu,1 A Isah2

PATS MECOR and University of Benin Teaching Hospital, Benin City, Nigeria Department of Clinical Pharmacology and Therapeutics, University of Benin, Benin City, Nigeria sunnydoc2003@yahoo.co.uk

66 AJTCCM VOL. 24 NO. 1 2018

Introduction. Prescription costs account for a significant proportion of the healthcare budget in developing countries with implications for affordability and access to healthcare. Routine analysis of medicine costs (using the World Health Organization (WHO) complementary drug use indicator) is necessary to generate data for planning. Objective. To assesses the prevailing cost of medicines in tertiary institutions in Mid-Western Nigeria. Methods. In a cross-sectional study, 1 800 prescription orders from January to December 2014 in two tertiary hospitals were analysed using the WHO complementary drug use indicator tool. Data obtained from prescription orders, including medicines prescribed, cost assigned to each medicine and aggregate cost for each prescription were analysed according to the WHO recommended guideline for complementary indicators. All costs were based on the current price list of each hospital. Results. A total of NGN3 667 548 (USD11 461) was spent on the 1 800 patient encounters (USD1 = NGN 200 as at 2014). The cost per prescription ranged from NGN0 to NGN43 950 (USD0 - USD219.75). The average cost per prescription was NGN2 037.53 (2 515.72) (USD10.19). The percentage of total drug cost spent on antibiotics was 16.2%. Injections and antimalarials accounts for 8.2% and 8.1%, respectively, while 75.7% of the total cost was spent on all other therapeutic categories of drugs. Conclusion. Prescription costs remain a major consideration in the Nigerian healthcare system when considering the high poverty level (minimum wage NGN18 000; i.e. USD90 per month). The findings are a reflection of the situation in most other healthcare facilities in the country and calls for determination of the causative factors and institution of a pricing mechanism.

Knowledge and attitude about obstructive sleep apnoea among resident doctors in Ebonyi State, SouthEast Nigeria I N N Okedo, A Chihurumnanya

Department of Community Medicine, Federal Teaching Hospital Abakaliki, Ebonyi State, Nigeria ijeomaninadr@gmail.com

Introduction. The majority of obstructive sleep apnoea (OSA) patients in Nigeria remain undiagnosed and untreated, which speaks to the need for increased awareness and suspicion among clinicians so that patients can receive optimal management. The knowledge and attitude of resident doctors in Nigeria towards obstructive sleep apnoea is not well documented. Objective. To assess the knowledge and attitudes about OSA among resident doctors in Nigeria. Methods. A cross-sectional study surveyed 148 resident doctors selected by systematic random sampling from specialties at the Federal Teaching Hospital, Ebonyi state, South-East Nigeria. Information was collected using a validated self-administered OSA Knowledge and Attitudes (OSAKA) questionnaire. Data were analysed using SPSS version 20 with the significance level set at 95%. Results. Of the 148 respondents (100% response rate), 107 (72.3%) were male and 41 (27.7%) were female. The majority (75%) were aged 30 - 39 years and 64.5% were in the medical specialities. The mean knowledge score score was 9.03 (3.16); 59.5% had above-average

ABSTRACTS knowledge. Ever having managed a patient with OSA symptoms was significantly associated with good knowledge of OSA (p=0.000), but there was no difference by speciality-, gender- or undergraduate training-related differences. The mean positive attitude score was 3.23 (1.16), with 92% and 95% considering OSA and its identification important, respectively, while 63 - 70% were confident in their ability to identify and manage OSA patients. Postgraduate training on OSA, good OSA knowledge and having managed an OSA patient were associated with a confident attitude (p=0.000, p=0.0012, and p=0.001, respectively). On logistic regression, having managed an OSA patient before was a determinant of good knowledge (odds ratio (OR) 4.38; 95% confidence interval (CI) 1.92 - 9.98). Determinants of confidence in identifying and managing high risk patients were postgraduate training (OR 3.17; 95% CI 1.42 - 7.14 and OR 3.48; 95% CI 1.56 - 7.75, respectively) and ever having managed a patient with OSA (OR 2.51; 95% CI 1.05 - 5.99 and OR 2.56; 95% CI 1.18 - 5.55, respectively) Conclusion. There was good knowledge and attitude about OSA among the surveyed resident doctors. Inclusion of training on OSA during residency across all specialties is advocated.

Caregivers᾽ knowledge of antibiotic use in children with upper respiratory tract infections and their willingness to learn proper antibiotic use from child educators in Enugu State T Okwor, C Agunwa, I Okonkwo, C Ochie

University of Nigeria Teaching Hospital Ituku Ozalla, Enugu State, Nigeria okwortochi@yahoo.com

Introduction. Antimicrobial resistance (AMR) is a major public health problem. Incorrect use of antibiotics in upper respiratory tract infection (URTI) in children contributes to AMR. An objective of the Nigerian AMR action plan is to improve awareness and understanding of AMR through effective communication, education and training. Objective. This study, the first of a two-part study, aims to determine the knowledge and practice of caregivers in the use of antibiotics for upper respiratory tract infection and their willingness to learn proper antibiotic use from child educators. Methods. Cross-sectional descriptive study conducted between April and May, 2017 among caregivers who came to immunise children at four primary, secondary and tertiary health facilities in Enugu metropolis. Ethical approval was obtained from the University of Nigeria Teaching Hospital Health Research and Ethics Committee. Data were collected by an interviewer administered questionnaire and were analysed using IBM SPSS version 20. Discrete variables are presented as proportions and continuous variables as mean (standard deviation (SD)). χ2 tests and Fisher’s exact tests were performed on categorical variables. A p-value pf 0.05 was considered significant. Knowledge scores of >80%, 60 - 80% and <60% were considered good, fair and poor knowledge, respectively. Results. There were 292 respondents with a mean (SD) age of 31.5 (9.2) years; 65.4% were mothers, 13% were guardians, 10.6% were fathers and 6.5% were grandparents. The level of knowledge was good, fair and poor among 1.4%, 32.9% and 65.8% of participants, respectively. The majority of the respondents (60%) believed that antibiotics should

always be prescribed in URTI and 53.1% knew that antibiotics are used in the treatment of bacterial infections. Over half 54.5% had received information on proper use of antibiotics with 45.5% having heard about antimicrobial resistance. A high proportion (74.7%) had ever asked doctor to prescribe antibiotics for URTI for the child. The majority of the respondents (66.1%) were willing to learn proper use of antibiotics from their children if taught at school. Conclusion. There is poor knowledge of antibiotic use in URTI in children and a high demand for doctor’s prescription of antibiotics. Caregivers were willing to learn proper antibiotic use from their children if they are taught in school, which indicates that an opportunity exists for using children as educators in Enugu state contributing to the fight against antibiotic resistance.

Effectiveness and safety of long-term v. short-term treatment regimens of multidrug-resistant pulmonary tuberculosis in Burkina Faso

A R Ouedraogo, G Badoum, G Ouedraogo, K Boncoungou, S Maiga Department of Pneumology, Yalgado Ouedraogo Teaching Hospital oarisgou@yahoo.fr

Introduction. The emergence of anti-tuberculosis drug resistance is of big concern in several countries and impede the effectiveness of tuberculosis control worldwide. The treatment protocol in Burkina Faso was a 21-month long-term regimen (LR) and it was costly and burdensome, both for patients and health staff. Following the promising results of a 2010 study in Bangladesh, Burkina Faso decided, in 2013, to be part of a clinical trial aimed at experimenting the effects of a ninemonth short-term regimen (SR), under the aegis of the International Union Against Tuberculosis and Lung Disease. Objective. To compare the efficacy and tolerance of the LR v. the SR multidrug-resistant tuberculosis (MDR-TB) treatment in Burkina Faso. Methods. We retrospectively compared two cohorts of patients who were followed for MDR-TB, from 1 January 2013 until 31 December 2015, in the Pulmonology Services of the University Hospitals Yalgado Ouédraogo and Souro Sanou in Burkina Faso. The first cohort was on the LR, which was based on the following drugs: pyrazinamide, kanamycin, levofloxacin, ethionamide and cycloserine. The second cohort was on the SR, based on kanamycin, moxifloxacin, prothionamide, isoniazid, clofazimine, ethambutol and pyrazinamide. Results. A total of 80 patients were included in the study; 47 patients were on the LR and 33 were on the SR. There were more retreatment failures in patients on the LR than in those on the SR (p=0.00). Also, during follow-up, patients under SR had a higher mean weight than those under LR (p=0.01). Patients with side effects under the SR (51.5%) were not significantly more numerous than those under the RL (51.1%). All SR patients had a smear negative during control of the last three months treatment. There was significantly more therapeutic success in patients on the SR (87.9%) than in those on the LR (51.1%; p=0.00). More deaths were observed in patients on the LR than in those on the SR (p=0.00). Conclusion. The nine-month SR was well tolerated and more effective than the 21-month LR. The SR protocol was endorsed by the WHO in 2016. Nevertheless, we must look at the matter with the wisdom of hindsight to fully appreciate its benefits.

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS

Potential risk factors for pneumonia among children under 5 years. Findings from the Pakistan Demographic Health Survey 2012 - 2013 S Razzaq, Z Fatmi

Aga Khan University, Karachi, Pakistan shama.razzaq@aku.edu

Introduction. About 13 million children under 5 years of age die every year in the world; out of which 95% of these deaths occur in developing nations and one-third of total deaths are due to ARIs. Findings from the Pakistan Demographic Health Survey (PDHS) showed that ~91 000 children die from pneumonia each year. Hence, identifying potential risk factors associated with acute respiratory infections among children is an important research domain to establish evidence-based interventions. Objective. To identify potential risk factors associated with pneumonia among children under 5 years of age from a secondary data analysis of the PDHS 2012 - 13. Methods. The PDHS 2012 - 13 is a nationally representative population-based random cluster survey. A total of 2 429 children under 5 years of age preceding the survey were included in the analysis. Demographic characteristics, potential environmental and socioeconomic risk factors were assessed and association were seen for pneumonia. A cough accompanied by short, rapid breathing that is chest-related was used as the operational definition of pneumonia. Results. In the multivariate logistic regression analysis, it was seen that the risk of having pneumonia increases among children residing in rural areas (adjusted odds ratio (aOR) 1.91; 95% CI 1.31 - 2.77), in KPK (aOR 2.50; 95% CI 1.29 - 4.83) and Gilgit-Baltistan (aOR 4.17; 95% CI 1.86 - 9.35), belong to poorest wealth quintile (aOR 10.84 95% CI 5.60 - 20.99), more among children less than 2 months of age (aOR 2.86; 95% CI 1.21 - 6.74), among males (aOR 1.48; 95% CI 1.11 - 2.01), having low birth weight ((aOR 2.83; 95% CI 1.14 - 7.01), had diarrhea (aOR 1.62; 95% CI 1.20 - 2.20), had mother’s smoking exposure (aOR 1.55; 95% CI 1.02 - 2.36), using biomass fuel for cooking (aOR 2.41; 95% CI 1.62 - 3.55). children were at less risk who ever had vaccination (aOR 0.73; 95% CI 0.51 - 0.99). Conclusion. We identified potential risk factors emphasiing focus on behavioural interventions which target vaccination uptake, hygienic practices and the use of safe cooking fuel.

Household expenditure for tuberculosis care, its determinants and coping strategies among adults 18 years and above in Karachi, Pakistan S Razzaq, A Zahidie, Z Fatmi

Aga Khan University, Karachi, Pakistan shama.razzaq@aku.edu

Introduction. Tuberculosis (TB) remains a major public health burden around the globe. In developing countries, the total cost of TB care often constitutes >50% of the yearly income of patients which leads the poor into catastrophic situations. Objective. To estimate the average household expenditure of TB care,

68 AJTCCM VOL. 24 NO. 1 2018

including pre-diagnostic and diagnostic costs among adults attending a government health facility in Karachi, Pakistan. Methods. A cross-sectional study was conducted in four government hospitals located in an urban setting in Karachi. Non-probability consecutive sampling was used to select the participants who have completed at least 1 month of treatment for the intensive phase. One standardised questionnaire named, tool to estimate patient’s costs was modified and used for getting costs related information. We defined direct medical and non-medical (food and transport) costs as out-ofpocket payment and indirect costs as loss of productivity. Pre-diagnostic and diagnostic costs were asked and estimates were calculated. Results. Out of 375 participants, 52.1% were female with a mean (standard deviation) age of 32 (13.7) years. Patients spent ~USD70.6 on the pre-diagnostic and diagnostic costs of TB care. More than half (51.2%) of the patients opted for private healthcare as the first place of consultation when they fell ill, followed by a government facility (44.2%), pharmacy or drug store (3.2%) and dispensary (1.6%). Only 43.4% of participants were employed, either formally or informally, and 95.5% of participants had arranged or borrowed money for TB treatment. Conclusion. Patients and households often bear the very high costs of TB care despite the fact that free services are available. There is an urgent need to implement strategies for TB care that are affordable for the poor.

Primary cooking fuel choice and respiratory symptoms among women in charge of household cooking in Ouagadougou, Burkina Faso A Sana,1 N Meda,1 C Bouland2

Laboratoire de Santé Publique (LASAP), Université Ouaga1 Pr Joseph KiZerbo, Ouagadougou, Burkina Faso 2 Environmental and Occupational Health Research Center, School of Public Health, Université Libre de Bruxelles, Brussels, Belgium adou_sanette@yahoo.fr 1

Introduction. In developing countries, solid fuels, including biomass fuels, remain the main sources of energy. Approximately 2.5 billion people rely primarily on biomass for cooking or heating. Many studies have highlighted the link between indoor air pollution and the occurrence of various health problems including cardiovascular and respiratory diseases, in the short-, medium- and long-term. As they are often in charge of the household cooking, women are exposed to sometimes high concentrations of pollutants contained in the smoke. Objective. To estimate the prevalence of respiratory symptoms in a subset of women in charge of household cooking and assess the association with the type of fuel used for cooking. Methods. A cross-sectional study was conducted in 3 neighbourhoods of Ouagadougou, involving 1 702 women who were responsible for cooking in their households. Univariate and multivariate logistic regression analyses were performed. Results. Acute dry cough, breathing difficulties, sneeze, nose tingling and throat irritation are the acute symptoms that are statistically associated with biomass fuel use if compared with butane gas use, respectively with p-value of 0.000, 0.002, 0.011, 0.031 and 0.000. It is also the case of some chronic respiratory symptoms such as sputum

ABSTRACTS production (p=0.001), shortness of breath (p=0.035) and wheezing (p=0.005). While self-reported asthma (p=0.140) and chronic cough (p=0.097) are not significantly associated with biomass use as cooking fuel, as it is the case of stuffy nose, runny nose and coughing during effort. When adjusted with age, socioeconomic status and education, dry cough, breathing difficulties, sneeze, throat irritation sputum production and wheezing remain associated with biomass fuel use. Conclusion. This study confirms that exposure to biomass smoke is associated with respiratory symptoms in women in Ouagadougou. The use of clean fuels, improving the efficiency of current fuel stoves and energy user behaviours, e.g. fuel drying, avoiding smoke exposure as much as possible during cooking, improved kitchen ventilation, properly used and maintained stoves and promoting outdoor cooking can reduce smoke emission and exposure. These measures should decrease respiratory health outcomes of these women.

Traffic air pollution and respiratory health effects: A cross-sectional study among bus drivers in Dakar, Senegal F K Sylla,1A Faye,1 M Fall,2 M Diaw,3 A Tal Dia1

Institut de Santé et de Développement, BP 16390 Dakar, Sénégal. Université Cheikh Anta Diop de Dakar, Senegal 2 Université Cheikh Anta Diop de Dakar, Laboratoire de Toxicologie et Hydrologie, BP 25 064 Dakar-Fann, Sénégal 3 Laboratoire de Physiologie et d’explorations fonctionnelles respiratoires, BP 5005 Dakar-Fann, Sénégal.Université Cheikh Anta Diop de Dakar, Senegal f.sylla@outlook.fr 1

Introduction. Traffic-related air pollution is well-documented to be associated with increased risks of airway diseases. Bus drivers are exposed to hazards resulting from the inhalation of pollutants from traffic. Objective. To describe the frequency of chronic respiratory symptoms and illnesses as well as its related factors and to assess lung functions among bus drivers. Methods. This was a cross-sectional study conducted among the bus drivers in HLM, Medina and Petersen districts, Dakar, Senegal. A total of 178 adult men were assessed using a questionnaire inquiring about sociodemographic factors, respiratory symptoms, toxicological medical evaluation and lung function tests. Logistic regression analysis was done to determine the relationship between various sociodemographic, occupational factors, respiratory symptoms and respiratory illnesses (chronic obstructive pulmonary disease (COPD) and asthma). Results. The results of the study showed that 57.9% of bus drivers had a chronic cough, 65.7% had the common cold and 53.4% had recurrent headaches. A predominance of these abnormal symptoms was noted in bus drivers located in the HLM district. Lung function tests showed that 38.8% of bus drivers had asthma, while 30.3% had COPD. Multivariate analysis found that frequent colds increased the risk of having asthma (odds ratio (OR) 6.3; 95% CI 1.12 - 35.79; p=0.03) and COPD (OR 7.7; 95% CI 1.14 - 52.8; p=0.03). The respiratory health status of bus drivers was dependent on the work area (OR 3.2; 95% CI 1.13 - 9.31; p=0.02). Conclusion. Chronic exposure to air pollution from traffic is associated with respiratory symptoms and illnesses, as well as reduced lung function indices among bus drivers.

Improving TB treatment outcomes through active tracing by facility-based health volunteers in Meru County, Imenti South subcounty, Kenya S Waithaka,1 B E Lubanga,2 J M Ng’ang’a,3 E K Nteere1

Ministry of Health, Meru County Government, Kenya Kenya Conference of Catholoc Bishops (KCCB) 3 Centre for Health Solutions (CHS - Kenya), Tuberculosis Accelerated Response and Care (TB ARC) samuelwaithaka24@gmail.com 1 2

Introduction. Kenya is ranked among the high-tuberculosis (TB), HIV- and drug-resistant TB burden countries. Non-adherence to anti-TB treatment adversely affects treatment success rate, contributes significantly to the development of TB drug resistance, increases disease morbidity and mortality. Previous research reported travel expenses, travelling to treatment centres, male sex, poor patient information and communication, alcoholism and homelessness as the major determinants of adherence to anti-TB treatment. According to the National Leprosy and Tuberculosis Program (2010), the TB treatment interruption rate was 9%, which put Kenya among countries with highest treatment interruption rate in the region. Methods. Community health volunteers (CHVs) were recruited and trained on defaulter tracing. The CHVs were assigned to specific villages/units to trace defaulters. Sensitisation of TB clinic staff on defaulters tracing and the need to keep an up-to-date defaulters list was done to ensure support for CHVs at the clinic level. Frequent meetings with the CHVs were carried out to ensure efforts were on track. The defaulters list was introduced and updated on a weekly basis. Monthly defaulters tracing reports were submitted by the CHVs. CHVs were given a moderate allowance based on the number of patients traced and were provided with airtime to call defaulters. Results. A total of 20 CHVs were recruited, trained and assigned different zones in which to carry out defaulter tracing. There was a reduction in the loss to follow-up rate (LTFU) from 7% (n=39/570) in 2013 to 1.2% in 2014, 2.4% in 2015 and 1.8% (n=10/557) in 2016. The treatment success rate increased from 88% in 2013 to 94% in 2014, and up to 93% in 2016. Conclusion. Active defaulter tracing reduced the LTFU from 7% to as low as 1.2%, and led to an increase in treatment success rate of up to 94%. Facility-based CHVs have the capacity to effectively carry out defaulter tracing with minimal support and basic training.

Autologous blood patch for the management of persistent air leak in inoperable patients: A case series E Wilken, J A Shaw, C F N Koegelenberg

Division of Pulmonology, Department of Medicine, Tygerberg Academic Hospital, Stellenbosch University, Cape Town, South Africa elismawilken@gmail.com

Introduction. Persistent air leak (PAL) after intercostal drain insertion delays lung re-expansion, increases complications, and results in longer hospital stay and higher treatment costs. A proportion of patients with PAL will be inoperable due to severe medical comorbidities and underlying lung disease – a safe alternative therapy should be

AJTCCM VOL. 24 NO. 1 2018

ABSTRACTS sought. It is proposed that the instillation of blood into the pleural space irritates the pleural surface, causing obliteration of the fistula by fibrogenic activity and the patch effect, allowing re-expansion of the lung. We present a case series of five inoperable patients with PAL who were successfully treated with autologous blood patch. Methods. Five inoperable patients with PAL were treated with autologous blood patch between October 2017 and January 2018. A total of 80 - 120 mL of the patientâ&#x20AC;&#x2122;s own unheparinised blood was injected into the pleural cavity via a 50 mL bladder syringe which was attached to the intercostal drain. This was clamped for one hour, followed by chest radiography and if bubbling ceased the intercostal drain was removed. Results. All five patients were successfully treated. The causes of most of the PALs included secondary spontaneous pneumothoraces (n=4) and iatrogenic (n=1). The mean duration of the PAL before pleurodesis was 26 days and a mean time to resolution and removal of the intercostal drain was 26 hours. Four patients were discharged within a week of the procedure. One had a prolong hospital stay owing to medical and social complications unrelated to the procedure. Conclusion. Autologous blood patch was a viable, safe and cheap intervention for PAL in inoperable cases.

Adverse respiratory health in the fibreglass reinforcement industry Z Zulu,1 R Naidoo2

Department of Environmental Health, Mangosuthu University of Technology , Durban, South Africa 2 Discipline of Occupational and Environmental Health, University of KwaZuluNatal, Durban, South Africa zanele@mut.ac.za 1

Introduction. Fibreglass reinforcement industry employees are exposed to both fibreglass and the agents used in the reinforcing process, particularly resins and styrene. They are also implicated with adverse respiratory outcomes. Objective. To determine the adverse respiratory outcomes among employees in the reinforced plastic industry. Methods. A cross-sectional study was conducted in the fibreglass reinforcement industry based in KwaZulu-Natal, South Africa. The

70 AJTCCM VOL. 24 NO. 1 2018

254 employees were interviewed by answering a questionnaire based on respiratory health and occupational exposures. Lung function tests were conducted for all employees according to SATS standards. Environmental monitoring was conducted for styrene and respirable dust. Total cumulative exposure was calculated for each participantâ&#x20AC;&#x2122;s lifetime of employment in the company. Results. The sample consisted of 76.38% males, with an average age of 39.5 years. The majority of the sample were never smokers (68.90%), while 25.59% were current smokers. The median styrene exposure levels were 42.81 (range 18.03 - 202.92) and 20.68 (range 5.2 - 47.28) for the general laminating department and fitting department, respectively. The respirable dust exposure level median was 3.28 (range 1.18 - 8.43). The mean exposure duration of employees in the fibreglass industry was 8.06 (range 1 - 39) years. The prevalence of respiratory symptoms was. chronic cough (15.35%), phlegm (14.17%), breathlessness (6.30%) and wheezing (14.96%). The prevalence of doctor-diagnosed respiratory diseases was low: pneumonia, 1.57%; chronic bronchitis, 1.57%; asthma, 2.36%; and pulmonary tuberculosis, 7.87%. The mean forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) of 3.52 L/min and 4.39 L, respectively, among the males and 2.79 L/min and 3.54 L, respectively, among females were adjusted for age, height and sex. There was an increase in the adjusted odds ratio for symptoms for increase in total cumulative exposure to the pollutants with chronic cough, wheeze and breathlessness odds ratios of 1.02, (95% CI 1.00 - 1.04), 1.01 (95% CI 1.00 - 1.02) and 1.03 (95% CI 1.02 - 1.04), respectively. A statistically significant association (p<0.05) between cumulative exposure for styrene and respiratory outcomes such as chronic cough and breathlessness. Similarly, there was a reduction in lung function parameters with exposure. The FEV1/ FVC ratio <70% showed a deficit in the pulmonary lung function of 15.35%. However, these were either marginal or not statistically significant, probably due to sample size. Conclusion. This study provides evidence that exposure in the fibreglass industry increases the prevalence of respiratory symptoms and is associated with reduced lung function. Greater control of environmental exposure is warranted.

NOTES

AJTCCM VOL. 24 NO. 1 2018

NOTES

72 AJTCCM VOL. 24 NO. 1 2018