Objective. To establish serum vitamin A levels in pregnant women from
a low socio-economic population in which the prevalence of human immun
odeficiency virus (HIV) infection is known to be high. Background. Vit
amin A deficiency occurs frequently in asymptomatic HIV infection and
is associated with increased risk of mother-to-child transmission of H
IV during pregnancy, delivery and lactation. Setting. King Edward VIII
Hospital, Durban, South Africa. Study design. Cross-sectional study.
Patients and methods. A total of 248 patients were studied in three gr
oups. Group 1 consisted of HIV antibody-positive pregnant women, group
2 consisted of HIV anti-positive non-pregnant women, and group 3 cons
isted of HIV antibody-positive women within 24 hours of delivery. Cont
rol groups consisted of HIV-uninfected women matched for age, parity a
nd, where applicable, period of gestation. Vitamin A (serum retinol) w
as measured by reverse-phase liquid chromatography. Results. The mean
vitamin A level in all women studied was 36.2 mu g/dl (range 3-99 mu g
/dl); 123 of 248 (49.6%) had inadequate levels of vitamin A (< 30 mu g
/dl), while 68 (27.4%) had levels consistent with vitamin A deficiency
. Vitamin A levels in the HIV-infected non-pregnant women (29.3 vs 47.
1 mu g/dl; P = 0.007) were significantly lower than those in the respe
ctive HIV-uninfected control groups. In addition the HIV-uninfected pr
egnant women had significantly lower vitamin A levels than their uninf
ected non-pregnant counterparts (32.8 v. 47.1 mu g/dl; P = 0.002). No
significant differences in vitamin A levels were observed between HIV-
infected and uninfected pregnant women (30.4 v. 32.8 mu g/dl). Vitamin
A levels of < 20 mu g/dl were observed in 28.4% of all the pregnant w
omen, and 40% of the infected women post-delivery. Conclusion. This st
udy confirms that vitamin A deficiency is a common health problem amon
g black South African women during pregnancy and associated with HIV i
nfection.