A research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, in partial fulfilment of the requirements for the degree of Master of Science in Medicine (Child Health Community Paediatrics) Johannesburg 2014 / Study Aim: To identify the criteria used to diagnose childhood tuberculosis (TB) at two paediatric outpatient departments in Johannesburg between 01 November 1997 and 30 June 1998 and to compare the criteria used to those used by the World Health Organization in the diagnosis of childhood TB. Study Design: An observational descriptive, retrospective, hospital outpatient-based study. Setting: Paediatric outpatient departments of Coronation Hospital (CH) and Johannesburg General Hospital (JGH). Patients and Methods: Patients ranging from the ages of 3 months to 14 years who attended the two paediatric outpatient departments and were diagnosed and notified as having TB, comprised the study population. Criteria used to establish the diagnosis of TB for each patient were extracted from patient records. Clinical history and baseline clinical characteristics were analysed using standard statistical methods, and criteria used to make a diagnosis of tuberculosis were compared to those recommended by the World Health Organization. Results: One hundred and one patients were diagnosed with TB at the two outpatient departments during the eight month study period. The combination of symptoms suggestive of active TB, which included persistent cough >1 month associated with fever, weight loss and loss of appetite, was more common in the JGH cohort (32 of 51 patients; 63%) compared to the CH cohort (10 of 50 patients; 20%); Odds Ratio (OR) 6.74 (95% Confidence Interval [CI], 2.54-18.41), P<0.001. One third (32%) of the total group of children had a positive TB exposure history. Tuberculin skin test (TST) reactions were positive in 86% of the total cohort, with a similar result being seen at both hospitals. Submission of specimens for microbiological assessment was not a common practice in either outpatient department, with 95% (96 of 101 participants) not having any specimens collected. Overall, 93% (94 of 101 participants) were classified as having ‘Probable’ TB. Conclusion: Criteria being used to diagnose childhood TB in the two paediatric outpatient departments are comparable to the WHO criteria recommended for the diagnosis of childhood TB. The majority of children diagnosed were classified as ‘Probable’ TB. TST was the main diagnostic tool used in the two outpatient departments at the time of study conduct.

A research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, in fulfillment of the Masters of Medicine in Paediatrics (MMED) / Paediatric HIV Clinic, Harriet Shezi Children's Clinic in an academic hospital, Chris Hani Baragwanath Hospital, Soweto, South Africa. OBJECTIVE: To describe and compare clinical, immunological and virological characteristics of HIV-infected children co-treated for TB, comparing those investigated microbiologically and those not, with a detailed description of microbiological TB investigation results

M.Med. (Radiology), Faculty of Health Sciences, University of the Witwatersrand, 2011 / Lymphobronchial tuberculosis (LBTB) is tuberculous lymphadenopathy involving the airways, which is particularly common in children. AIM: To describe the CT findings of LBTB in children, the parenchymal complications and associated abnormalities. METHOD: CT scans of 98 children with LBTB were retrospectively reviewed. Lymphadenopathy, bronchial narrowing, parenchymal complications and associations were documented. RESULTS: Infants comprised 51% of patients. The commonest lymphadenopathy was subcarinal (97% of patients). Bronchial compressions (259 in total) were present in all patients, of which 23% were severe / complete stenoses and 28% affected bronchus intermedius. Parenchymal complications were present in 94% of patients, including consolidation (88%), breakdown (42%), air trapping (38%), expansile pneumonia (28%), collapse (17%) and bronchiectasis (9%), all predominantly right-sided (63%). Associations included oval focal bodies, miliary nodules, pleural disease and intracavitory bodies. CONCLUSIONS: The most important CT finding of children with LBTB is visible airway compression as a result of lymphadenopathy. CT of children with LBTB showed that airway compressions were more severe in infants and most commonly involved bronchus intermedius. Numerous parenchymal complications were documented, all showing rightsided predominance. Several associations were identified.

There are few reports on the impact of HIV-1 infection on tuberculosis in children. Microbiologic diagnosis of tuberculosis is difficult and much reliance is placed on the tuberculin skin test, as part of a scoring system, in diagnosing tuberculosis in children. A prospective study, enrolling 168 patients with clinical tuberculosis, was performed between July 1996 and January 1997 at the teaching hospitals attached to the Department of Paediatrics and Child Health, University of the Witwatersrand. Forty-two percent of children with tuberculosis were HIV-1 infected. Extrapulmonary tuberculosis was diagnosed more frequently in HIV uninfected children. Progressive pulmonary tuberculosis, based on radiographic findings, and mortality was higher in HIV-1 infected children with tuberculosis. HIV-1 infected children with pulmonary tuberculosis showed marked hyporeactivity to tuberculin skin testing. Both CD4+ lymphocyte counts and concurrent delayed type hypersensitivity testing, using the "CMI Multitest®”, offered little value in interpreting the tuberculin skin test in HIV-1 infected children with tuberculosis. The findings of the study suggest that aggressive microbiologic investigations coupled with a low threshold of clinical suspicion is essential in diagnosing tuberculosis in children, especially in HIV infected children.