MAN-TO-WOMAN SEXUAL TRANSMISSION OF HIV - LONGITUDINAL-STUDY OF 343 STEADY PARTNERS OF INFECTED MEN

To study incidence and risk factors of heterosexually transmitted HIV infection, we followed a cohort of 343 seronegative women, stable, mon ogamous partners of infected men whose only risk of acquiring HIV was sexual exposure to the infected partner. Nineteen seroconversions occu rred in 529.6 person years (py) of observation, yielding an incidence rate of 3.6 per 100 py. The incidence rate was 7.2 per 100 py among wo men who did not always use or never used condoms and 1.1 among those w ho always used them [relative risk (RR) 6.6, 95% confidence interval ( CI) 1.9-21.9]. Anal sex was associated with a risk increase in only th ose women not always using condoms (RR 1.4, 95% CI 0.4-4.8). No seroco nversions were observed among 22 women using oral contraceptives. One of the women using intrauterine devices seroconverted. In couples who did not always use condoms, seroconversions occurred more frequently i n partners of men with symptomatic diseases, with a low CD4 + cell num ber (<400 per mm3) or with a detectable p24 antigen. In couples not al ways using condoms and where the man had a low CD4 + cell count, the j oint presence of blood viral antigens and AIDS symptoms conditioned a fivefold increased risk of seroconversion of the woman (RR 5.4, CI 1.4 -20.3). At multivariate analysis, women with longer relationships (gre ater-than-or-equal-to 1 year) showed a lower risk of seroconversion (R R 0.3, CI 0. 1-0.8), and those partners of men positive for p24 antige n in serum had an increased risk of seroconversion (RR = 4.0, CI 0.1-0 .8).