Letters to the Editor Vitamin A Deficiency and Genital Tract Infections in Women Living in Central Africa To the Editor: Low vitamin A and carotenoid levels could increase the risk of sexual HIV acquisition by altering the in- tegrity of the genital epithelium or by providing immunologic dysfunction (1). The impact of vitamin A deficiency on HIV-1 acquisition remains, however, conflicting. Mac Donald et al. (2) recently reported the lack of relationship between vitamin A deficiency and an increased risk of HIV-1 infection among men with concurrent sexually transmitted infections (STI) in Nairo- bi, Kenya, suggesting that vitamin A deficiency is not associ- ated with increased susceptibility to HIV. In the same way, a transversal study of HIV-1 seroconverting women in Rwanda (3) showed no difference in the mean serum retinol levels be- tween HIV-1-seroconverting and HIV-1-seronegative women. In contrast, low serum provitamin A carotenoid levels were associated with an increased risk for heterosexual HIV acqui- sition in male or female patients suffering from STI in Pune, India (4). We herein report that vitamin A deficiency may favor the receptivity to genital tract infections, including STI, which are major cofactors for both HIV transmission and acquisition via sexual intercourse (5). We enrolled 275 consecutive and consenting women attend- ing the Centre National de Reference des Maladies Sexuelle- ment Transmissibles et du SIDA, in Bangui, the capital city of the Central African Republic. The center offers multipurpose reproductive health services including provision of STI ser- vices, as well as operating as the main voluntary HIV testing and counseling centre in Bangui, as previously described (6). Women underwent general, genital, and pelvic examination, during which serum, vaginal, and cervical samples were col- lected. A 7-day follow-up appointment was arranged for all women and appropriate treatment was provided free of charge for any treatable STI syndrome or genital pathogen diagnosed. All women were tested for HIV and syphilis serologies. Stan- dard tests were carried out for bacterial vaginosis (BV), Tricho- monas vaginalis (TV), Candida albicans (CA), Neisseria gon- orrhoeae (NG), and Chlamydia trachomatis (CT) by an enzyme immunoassay of an endocervical swab, and Haemophi- lus ducreyi, as previously described (6). Vitamin A concentra- tions in serum was determined by high performance liquid chromatography with UV reading at 325 nm (7). The micro- nutrient levels were compared to international reference values, with vitamin A deficiency defined at levels <30 g/dL (8). Statistical analyses were carried out using InStat software (GraphPad, Inc., San Diego, CA, U.S.A.). The mean age of the study population was 27 years (range 15–48 years). Median age of first sexual intercourse was 16 years, with a median of two reported lifetime partners (range 1–8). The majority of women were married, either in monoga- mous (47%) or polygamous (13%) marriages. The prevalence of the various genital pathogens is shown in Table 1. Nearly 80% of women had an endogenous vaginal infection (CA and/or BV), but only 4 women (1.4%) had TV, and 13 (4.7%) had a cervical infection (NG and/or CT). Overall, 71 women were HIV infected (26%) and 21 (8%) had evidence of active syphilis. The mean serum concentration of vitamin A was 52 g/dL and only 15 (5.5%) women were considered deficient. There was a statistically significant difference in the levels of vitamin A between HIV-1-seropositive and HIV-negative individuals (p < .001), and between syphilis-seropositive and syphilis- seronegative individuals (p < .01) (Table 1). The proportion of HIV-infected patients with serum vitamin A deficiency was higher than that of HIV-negative patients (p < .005). In patients harboring only one genital tract pathogen, the proportion of patients with vitamin A deficiency was higher in HIV-infected patients than in HIV-negative patients (p < .01) (Table 2). The TABLE 1. Vitamin A serum levels according to infection status in 275 women of childbearing age consulting the main STI clinic of Bangui, Central African Republic Infections Vitamin A levels (g/dL) p value Negative Positive N Mean (SD) N Mean (SD) HIV 204 2.02 (0.61) 71 1.72 (0.62) <.001 Syphilis a 251 1.97 (0.64) 24 1.60 (0.43) <.01 Bacterial vaginosis b 110 1.95 (0.61) 165 1.82 (0.49) NS Candida albicans 147 1.95 (0.62) 128 1.94 (0.65) NS Trichomonas vaginalis 271 1.94 (0.63) 4 2.11 (0.38) NS Neisseria gonorrhoeae 271 1.95 (0.63) 4 1.52 (0.66) NS Chlamydia trachomatis 266 1.94 (0.64) 9 2.00 (0.49) NS a Serologic diagnosis of active syphilis was defined by a positive rapid plasma reagin test (VD-25 Murex Diagnostics, Dartford, UK), with positive confirmatory Treponema pallidum hemagglutination as- say (TPHA, Fujirebio, Tokyo, Japan). b Bacterial vaginosis was diagnosed by the Nugent scoring method of a Gram-stained vaginal smear. SD, standard deviation; NS, not significant. TABLE 2. Vitamin A deficiency and HIV serostatus according to the number of genital tract infections a or seropositivity for syphilis in 275 women of childbearing age living in Central Africa Infections, n N (%) Vitamin A deficiency b p value n (%) HIV + (%) HIV - (%) 0 129 (47) 1 (0.7) 1/34 (<1) 0/95 (0) NS 1 128 (46) 6 (6.3) 4/29 (14) 2/99 (2) .008 2 13 (5) 4 (15.4) 2/4 (50) 2/9 (21) NS 3 5 (2) 4 (80.0) 3/4 (75) 1/1 (100) NS Total 275 15 10/71 (14) 5/204 (2.5) .002 a Genital tract infections were Candida albicans, Trichomonas vagi- nalis, Neisseria gonorrhoeae, and Chlamydia trachomatis. b Vitamin A deficiency is defined as levels <30 g/dL. NS, not significant. JAIDS Journal of Acquired Immune Deficiency Syndromes 29:203–206 © 2002 Lippincott Williams & Wilkins, Inc., Philadelphia 203