Integrating active tuberculosis case finding in antenatal services in Zambia.

SETTING: Three out-patient antenatal care (ANC) clinics in Lusaka Zambia. OBJECTIVE: To estimate tuberculosis (TB) prevalence in human immunodeficiency virus (HIV) infected and symptomatic non-HIV-infected pregnant women and explore the feasibility of routine TB screening in ANC settings. DESIGN: Peer educators administered TB symptom questionnaires to pregnant women attending their first ANC clinic visit. Presumptive TB patients were defined as all HIV-infected women and symptomatic non-HIV-infected women. Sputum samples were tested using smear microscopy and culture to estimate TB prevalence. RESULTS: All 5033 (100%) women invited to participate in the study agreed and 17% reported one or more TB symptoms. Among 1152 presumed TB patients 17 (1.5%) had previously undiagnosed culture-confirmed TB; 2 (12%) were smear-positive. Stratified by HIV status TB prevalence was 10/664 (1.5% 95%CI 0. 7-2.8) among HIV-infected women and 7/488 (1.4% 95%CI 0.6-2.9) among symptomatic non-HIV-infected women. In HIV-infected women the only symptom significantly associated with TB was productive cough; symptom screening was only 50% sensitive. CONCLUSION: There is a sizable burden of TB in pregnant women in Zambia which may lead to adverse maternal and infant outcomes. TB screening in ANC settings in Zambia is acceptable and feasible. More sensitive diagnostics are needed.