Increasing likelihood of further live births in HIV-infected women in recent years

Objective To examine how the subsequent childbearing of HIV-infected mothers enrolled in the European Collaborative Study (ECS) has changed over time and identify factors predictive of further childbearing. Design Prospective cohort study. Setting Centres in nine European countries included in the ECS, enrolled between end 1986 and November 2003. Population HIV-infected women (3911): 3693 with only one and 218 with subsequent live births. Methods Univariable and multivariable logistic regression analyses to obtain odds ratios (OR) and 95% confidence intervals (CI). Kaplan-Meier (KM) analyses to estimate cumulative proportions of women having a subsequent live birth. Main outcome measures Subsequent live birth. Results In multivariable analysis adjusting for time period, ethnicity, maternal age and parity, black women were more likely [adjusted odds ratio (AOR) 2.45; 95% CI, 1.75-3.43], and women > 30 years less likely (AOR 0.54, 0.37-0.80), to have a subsequent live birth. Time to subsequent live birth significantly shortened over time, with an estimated 2% of women having a subsequent live birth within 24 months of enrolment pre-1989 versus 14% in 2000-2003 (P < 0.001). Estimated time to subsequent live birth was shorter for black than for white women (P < 0.001). Conclusions The likelihood of subsequent live births substantially increased after 1995 and birth intervals were shorter in women on HAART and among black women. Numbers are currently too small to address the issue of advantages and disadvantages in the management of subsequent deliveries.