Objective: To examine the association between HIV-II infection and tuberculosis.
Design: Cross sectional study comparing the prevalence of HIV-I and HIV-II infections in patients with tuberculosis and in blood donors.
Setting: Abidjan, Ivory Coast, west Africa.
Patients: 2043 consecutive ambulant patients with tuberculosis (confirmed pulmonary, presumed pulmonary, or extrapulmonary) and 2127 volunteer blood donors.
Main outcome measure: Prevalence of HIV-I and HIV-II infections as assessed by presence of serum antibodies.
Results: Overall rates of HIV infection were 40.2% in patients with tuberculosis (26.4% positive for HIV-I, 4.7% for HIV-II, and 9.0% for both); and 10.4% in blood donors (7.2% positive for HIV-I, 1.9% for HIV-II, and 1.3% for both). HIV-II infection was significantly more common in patients with all types of tuberculosis than in blood donors (97/2043, 4.7% v 40/2127, 1.9%; odds ratio 3.8%, 95% confidence interval 2.6 to 5.6).
Conclusion: Both HIV-I and HIV-II infections are associated with tuberculosis in Abidjan. 35% of adult tuberculosis in Abidjan is attributable to HIV infection and 4% specifically to HIV-II.