|
Background:
Recent analyses examining mortality among HIV-infected mothers by child's
feeding modality yielded conflicting results. We used individual patient
data from randomized, placebo-controlled clinical trials to estimate
mortality risk among HIV-infected mothers over an 18-month period after
delivery, by breastfeeding status.
Methods:
Trials conducted in mainly breastfeeding populations and with regular
assessment of children's feeding modality were eligible for inclusion.
Results:
The analysis cohort included 4237 African mothers with available vital
status (death date or last known alive date) and child's feeding modality
data; 162 (3.8%) died within 18 months of delivery (median time to death:
9.8 months). Median CD4+ count (cells/mm³) around time of delivery was 464
(11% <200; 45%=200-499). Overall mortality was 28.7/1000 person-years
follow-up at 12 months and 32.2/1000 at 18 months. 3717 (87.7%) women ever
breastfed (median duration=8.8 months). Univariately, child's feeding
modality was not associated with mothers' mortality (P>0.11). Independent
risk factors for mortality in multivariate analyses were: maternal CD4+
count (low r CD4+ increased risk of 12- and 18-month mortality (P<0.001)); a
child's feeding modality (mothers who ever breastfed had lower risk of
12-month mortality than mothers of never breastfed children (P=0.033), but
not of 18-month mortality (P=0.068)).
Conclusions:
Women with more advanced HIV disease, manifested by lower CD4+ around time
of delivery, were at increased risk of mortality 12-18 months after
delivery. Mothers who ever breastfed were at lower risk of mortality through
12 months after delivery. Ongoing analyses are further assessing the risk of
mortality among HIV-infected mothers who ever breastfed. |
