59 (18.3%) of 323 patients with tuberculosis (TB) tested for HIV-1 antibody by ELISA technique (Wellcozyme) were seropositive. In the control group selected among the health personnel working in the Arua Hospital, 7.7% were found positive for HIV-1 antibody, thus showing a significantly lower prevalence compared with the TB patients (P < 0.005). The prevalence of HIV infection was 50% in the urban TB patients, 7% in TB patients living in rural areas surrounding Arua town and 1.6% in the peripheral rural setting. Of 27 TB patients with clinical AIDS, 18 died during the course of the study. The AIDS patients' survival rate was 46.4% 6 months after diagnosis, and 21.4% after 16 months, the median period of survival being 5.0 months. Risk factors, sputum conversion rate, clinical and radiological findings were analysed. No significant difference was found between seropositive and seronegative TB patients for clinical drug-related toxicity (P > 0.05).
PIP: Between June 1987 and August 1989, physicians enrolled 323 tuberculosis (TB) patients and 116 health employees at the Arua Regional Hospital in a rural district of northern Uganda in a case control study. They wanted to look at the link between TB and HIV infection. TB patients were more likely to be HIV seropositive than the employees (18.3% vs. 7.7%; p .005). HIV seropositive individuals tended to be men (71.2% vs. 54.9% for controls; p .05) whose mean age was 27.69 years. Most HIV/TB patients lived in the town of Arua (50% vs. 7% in rural areas peripheral to Arua and 1.6% in a rural area near the district border; p .0001). HIV seropositive TB patients were more likely to have a sexually transmitted disease (STD) than HIV seronegative TB patients (47.4% vs. 12.5%; odds ratio [OR] = 6.32; p .0001), especially gonorrhea (p .0001). They also tended to have had more than 5 sexual partners in the past 2 years (mean number of partners among HIV seropositive TB patients = 10.6; 35.6% vs. 9.5%; OR = 9.24; p .0001). HIV seropositive TB patients were more likely to have participated in prostitution and to have had a blood transfusion than HIV seronegative TB patients (33.9% vs. 3.8%; OR = 13.03; p .001 and 6.8% vs. 1.1% OR = 6.33; p .05). Skin piercing, widely practiced in rural areas, appeared to have a protective effect against HIV infection (OR = .33; p .0005). HIV seropositive TB patients were significantly more likely to have a persistent cough of more than 4 months duration (p .001), fever lasting for more than 1 month (p .05), oral thrush (p .0001), lymphadenopathy (p .0005), and amenorrhea (fertile women only, p .005). 27 or 28 TB patients had AIDS. At the time of submission of this study for publication, 18 HIV seropositive TB patients died during treatment. The case fatality rate was indeed higher among HIV seropositive TB patients than among HIV seronegative TB patients (30.5% vs. 8.7%; p .0001). The TB-AIDS survival rate was 46.4% at 6 months, 32.1% at 12 months, and 21.4% at 16 months. Median survival time was 5 months.