The impact of HIV on infectiousness of pulmonary tuberculosis: a community study in Zambia

Abstract

Objective: To examine the impact of HIV on infectiousness of pulmonary tuberculosis (TB).

Design: A cross-sectional tuberculin survey carried out among household contacts of HIV-1-positive and negative patients with bacteriologically confirmed pulmonary TB. Contacts were also examined for active TB.

Setting: Index cases were recruited from patients attending the University Teaching Hospital in Lusaka, Zambia and household contacts were examined during visits to their homes within Lusaka.

Patients, participants: A total of 207 contacts of 43 HIV-positive patients, and 141 contacts of 28 HIV-negative patients with pulmonary TB were examined.

Main outcome measures: Proportion of contacts of HIV-positive and negative index cases with a positive tuberculin response (diameter of induration > or = 5 mm to a dose of 2 tuberculin units).

Results: Fifty-two per cent of contacts of HIV-positive pulmonary TB patients had a positive tuberculin response compared with 71% of contacts of HIV-negative patients (odds ratio, 0.43; 95% CI, 0.26-0.72; P < 0.001). This difference persisted after allowing for between-household variations in the tuberculin response. Tuberculin response in the contact was related to age of contact, intimacy with the index case and crowding in the household. However, the effect of HIV status of the index case was not confounded by these variables. Tuberculin response in the contact was also related to the number of bacilli seen in the sputum smear of the index case which partially explained the effect of HIV status of the index case. Active TB was diagnosed in 4% of contacts of HIV-positive and 3% of contacts of HIV-negative cases, respectively (P = 0.8).

Conclusions: HIV-positive patients with pulmonary TB may be less infectious than their HIV-negative counterparts and this may partly be explained by lower bacillary load in the sputum.

PIP: Between April and December 1989, the chest clinic of the University Teaching Hospital in Lusaka, Zambia, confirmed pulmonary tuberculosis (TB) in 141 adults, 95 (67%) of whom were HIV-1 seropositive. Health workers made home visits to 71 of the index cases (43 HIV-1 positive and 28 HIV-1 negative) to learn whether the 348 household members would also develop TB, thus allowing researchers to determine the effect of HIV on infectiousness of TB. Contacts of HIV-1 positive patients developed TB at a lower rate than did those of HIV-1 negative patients (52% vs. 71%; odds ratio [OR] = 0.43; p .001). This difference continued even after controlling for between-household variations, indicating that confounding variables did not account for the difference. Age of contact, intimacy with the index case, and crowding in the household were associated with the tuberculin response in the contact, but they did not confound the effect of HIV status. Tuberculin response in the contact was associated with the number of bacilli in the sputum smear (crude OR = 3.13; p = .013, and adjusted OR =1.84; p = .28), suggesting that the number of bacilli somewhat explained the difference in infectiousness between HIV-1 positive and HIV-1 negative patients. 12 contacts (8 of HIV-positive cases and 4 of HIV-negative cases) developed active TB after the TB diagnosis in the index case. These findings clearly demonstrated that infection with Mycobacterium tuberculosis was less likely in household members of HIV-1 positive cases than in those of HIV-1 negative cases. The lower bacillary load in the sputum in HIV- 1 cases may have accounted somewhat for the lower infectiousness of pulmonary TB.