A cross-sectional study to estimate the prevalence of latent tuberculosis (TB) in a group of Zambians at high risk of human immunodeficiency virus type 1 (HIV-1) infection and to examine the effect of HIV-1 infection on the tuberculin response was conducted in the University Teaching Hospital in Lusaka, Zambia during July to September 1990. Patients were selected from those presenting to the out-patient clinic for first referral with either sexually transmitted or skin disease. 268 adults were included in the study; 158 (59%; 95% confidence interval [CI] = 53-65%) were HIV-1 antibody positive. Of 82 HIV-1 negative participants who returned for Mantoux skin test reading, 51 (62%; 95% CI = 57-67%) had a positive test reaction (diameter > or = 10 mm) after receiving 2 units of RT-23 tuberculin. Of 106 HIV-1 positive participants who returned, only 32 (30%; 95% CI = 26-34%) had a diameter > or = 10 mm. Nine (28%) of the HIV-1 positive and Mantoux positive participants had large reactions > or = 30 mm, compared to 4 (8%) of the HIV-1 negative, Mantoux positive participants (P = 0.03). Results in the HIV-1 negative group indicated a prevalence of latent TB of 62% in this population. HIV-1 infection was associated with a much higher frequency of negative response to tuberculin and with a few large skin test responses. Thus, in populations where HIV seropositivity is high, Mantoux skin tests cannot be used to assess those with latent TB who might benefit from chemoprophylaxis.
PIP: A cross-sectional study of the Mantoux response and HIV-1 status of a sample of patients with sexually transmitted diseases and skin diseases in Lusaka, Zambia, sought to estimate the prevalence of latent tuberculous infection. The sample was selected from patients attending the sexually transmitted diseases/dermatology section at the University Teaching Hospital, Lusaka, Zambia, between July and September 1990. A questionnaire regarding socioeconomic factors, history of TB, contact with TB, location and documentation of bacillus Calmette-Guerin (BCG) scar(s) and history of BCG vaccination was completed, and a physical examination for acquired immune deficiency syndrome (AIDS) was carried out. The Mantoux result was recorded as the average diameter of induration, measured in 2 perpendicular directions by the pen and palpation method. A total of 158 patients (59%) were HIV-1 positive. Of the 66 women who took part, 46 (70%) were HIV-1 positive; of the 201 men, 112 (56%) were HIV-1 positive (p = 0.06). 232 patients had sexually transmitted diseases, the commonest being genital ulceration; 123/231 (53%) were HIV-1 positive. The remaining 36 patients had skin diseases, the commonest being herpes zoster; 32/36 (89%) were HIV-1 positive. Of the 267 patients remaining in the study, 193 (72%) returned to have their Mantoux test read, 188 within 48-72 h. 106 (67%) HIV-1 positive patients and 82 (75%) HIV-1 negative patients returned. Of the 82 HIV-1 negative patients, 51 (62%) had a Mantoux reaction or= 10 mm; 55 (67%) had a reaction or= 5 mm. Of the 106 HIV-1 positive patients, only 32 (30%) had a Mantoux reaction or= 10 mm; 35 (33%) had a response or= 5 mm. Comparing HIV-1 negative and HIV-1 positive participants gave a significant odds ratio of 3.85 for a Mantoux response or= 10 mm. Among the individuals with a Mantoux reaction or= 10 mm, 9/32 (28%) of HIV-1 positive participants had a megareaction or= 30 mm, while megareactions occurred in 4/51 (8%) of HIV-1 negative participants (odds ratio 4.6).