Clinical diagnosis of smear negative pulmonary tuberculosis in HIV-positive patients at Athlone Hospital in Botswana

Abstract

Background and aim: Smear-negative pulmonary tuberculosis (SNPTB) has become an increasingly important clinical and public health problem, especially in areas that are affected by the dual infection of TB and human immunodeficiency virus (HIV) (Mello et al, 2006; WHO, 2006; Harries et al, 1998). There are recommended guidelines for diagnosing SNPTB to reduce misdiagnosis in sub-Saharan Africa, but there is little information on whether these guidelines are followed correctly (Harries et al, 1998). The aim of this study was to investigate the clinical diagnosis of SNPTB in HIV-positive patients at Athlone Hospital in Botswana. Methods: This was a quantitative, descriptive study which used two sources of data and data collection methods: a 4 year retrospective records review and questionnaires for clinicians. All clinicians responsible for treating HIV-positive patients (n=8) were asked to complete a questionnaire on self-reported (1) compliance with the guidelines (2) use of other methods to diagnose SNPTB and (3) reasons for not complying with the guidelines. All records on SNPTB in HIV-positive patients from 2006 to 2009 (n=281) were reviewed to establish the compliance and use of other methods to exclude other respiratory infections. Results: The response rate for clinicians was 87.5% (7/8). All clinicians (100% [7/7]) reported (a) always complying with using chest x-rays (CXRs), but (b) only sometimes complying with using 3 sputum results. Most clinicians (a) considered the duration of cough before making a diagnosis of SNPTB (57.1% [4/7]), and (b) placed patients on a trial of broad spectrum antibiotics before starting PTB treatment (85.7% [6/7]). The main reasons for non-compliance were: the inability of patients to submit sputum (100% [7/7]), delays in the laboratory (71.43% [5/7]), and lack of feedback from Botswana National Tuberculosis Program (BNTP) (57.14% [4/7]). Only 2.1% (6/281) of the records showed that other methods were used to rule out other respiratory infections, and overall compliance with the recommended guidelines was only 13.5% (40/281). Conclusion: The compliance with the recommended guidelines in making a diagnosis of SNPTB was very poor in this study. The unavailability of user-friendly and fast diagnostic methods resulted in many cases being treated for SNPTB with inadequate investigations.