Pulmonary tuberculosis and HIV interaction in a setting with a high prevalence of HIV : clinical, diagnostic and epidemiological aspects

Abstract

The overall aim of this study was to elucidate the influence of HIV infection on the occurrence, clinical presentation, diagnostic yield and transmission of M tuberculosis among consecutive out-patients with symptoms suggestive of PTB in a high TB and HIV endemic area.

Among 509 consecutive out-patients attending a university hospital in Ethiopia, 33.0% were culture-verified as having PTB. PTB patients, non-TB patients and controls were HIV-1 positive in 57.1%, 38.5% and 8.3% of cases, respectively. Independent predictors for culture-verified PTB were age <25 years, male gender and the presence of HIV and fever, whereas profound weight loss indicated HIV infection Diagnosis of PM based on clinical symptoms, direct sputum microscopy for acid-fast bacilli (AFB) and chest radiography (CXR) was significantly less sensitive and specific in HIV-positive patients Our findings indicate that, although M is more often under-diagnosed in HIV-positive patients, other HIV-related pulmonary infections am often misinterpreted as smear-negative PM in this setting. HIV screening is warranted as part of the diagnostic work-up m patients with respiratory symptoms suggestive of PM. Also, increased awareness of, and improved diagnostic tools and algorithms for, HIV-related pulmonary infections, including PTB, are required.

A high proportion of direct smear-negative culture-proven PTB (61.5%) was found in HIV-positive patients. In this patient group, sputum concentration after digestion with sodium hypochlorite significantly increased the yield for AFB, suggesting did this method has a place m routine diagnosis of PTB m HIV endemic countries. Molecular strain typing of the IS6110 and direct repeat genetic markets and drug sensitivity testing were performed in 121 M tuberculosis isolates.

In a multivariate logistic regression model HIV-positive serostatus was significantly associated with clustering in both sexes. There was a trend towards increased clustering with isolates obtained from tuberculous women residing in Addis Ababa (OR 2.10; 95% CI 0.85-5.25). In total, 17 bit of 121 isolates (14.0%) were, resistant to one or more of the standard anti-tuberculous drugs. Inpatients 35-44 years old the high rate of drug resistant isolates (29.6%) coincided with the peak HIV prevalence (77.8%). The majority (62.5%) of resistant isolates in this group were found within clusters. Simultaneous accumulation of certain bacterial clones in a patient population is likely to reflect recent transmission. Hence, we conclude that in Ethiopian HIV-positive patients tuberculosis is commonly caused by recent infection. Furthermore, drug resistant tuberculosis is likely to increase Strengthening of classical TB control measures by promoting active case-finding among HIV-positive adults with TB is warranted to reduce transmission.

The highest prevalence of Pneumocystis carinii (P carinii) in expectorated sputum. samples was 10.9% by immunofluorescence assay (IF) and 30.3% by nested polymerase chain reaction (nPCR) among HIV-positive non-TB patients. In HIV-negative non-TB patients, the prevalence of P carinii by nPCR was significantly lower In the IF positive and nPCR positive HIV-positive non-TB patients more than third were interpreted as PTB by CXR whereas only one patient was diagnosed with clinical PCP. The frequent occurrence of P carinii by nPCR suggests that PCP may be an important but not well recognised, differential diagnosis m HIV-positive patients with symptoms suggestive of PTB, implicating the need for treatment and primary prophylaxis for PCP in Ethiopia.