Background:  Scheduled ART interruption has been associated with accelerated HIV disease progression in some studies. Effects of treatment interruption after ART prophylaxis in pregnancy on maternal health have not been well studied. We evaluated changes in CD4 cell count, CD4 percentage, HIV RNA, and disease progression between women who continued ART after delivery and those who stopped ART at delivery.

Methods:  Women enrolled in the Women and Infant Transmission Study (WITS) after June 1, 1994 who were ART-naive and had a CD4 count >350 cells/µL were included. CD4 count, CD4% and HIV RNA slope between delivery and 12 months postpartum and progression to class B or C diagnoses were compared between women who continued ARV therapy after delivery and women who stopped at delivery. Mix model analysis was used to compare slopes; Cox models were used to evaluate disease progression.

Results:  Of 3297 women enrolled to WITS, 1165 were ART naive, 512 of these received ARV, and 206 women had CD4⁺ cell counts >350 cells/µL. Women included in this analysis were similar to those excluded, except that they had a higher mean CD4 count and lower gestational age at enrollment. Slopes of CD4 count and percentage, and HIV RNA did not differ between women continuing or stopping therapy after delivery between 2 to 6 months and 6 to 12 months post-partum (p = 0.35 to 0.96). No class C events occurred through 1 year post-partum in either group. New class B events (excluding women with previous class B events) occurred in 8 (9.4%) of 85 women continuing therapy versus 8 (18.6%) of 43 women stopping therapy (RR 2.09, 95%CI 0.79 to 5.58, p = 0.14). Evaluating women receiving zidovudine (ZDV) only during pregnancy and those receiving 2 or more drugs, no differences in slope of CD4 count and percentage, or HIV RNA were seen by therapy type comparing women continuing or stopping therapy. Women on ZDV only who stopped therapy had a RR of 1.24 (0.31 to 4.95) of class B events over 1 year compared to those continuing. For combination therapy, the RR was 2.93 (0.64 to 13.36, p = 0.16) for those stopping therapy.

Conclusions:  The changes in CD4 and HIV RNA levels over 1 year postpartum were similar between women continuing or stopping therapy after initiation with CD4 counts >350 cells/µL. No class C events occurred in either group. No significant difference in rates of new class B events was seen, although power was limited. Larger studies with longer follow-up are required to evaluate the risks of stopping ART prophylaxis after delivery.