Subclinical iron deficiency is a strong predictor of bacterial vaginosis in early pregnancy
1 Department of Obstetrics & Gynaecology, Faculty of Medicine and Health Sciences, Ghent University, Ghent University Hospital, De Pintelaan 185, B-9000 Ghent, Belgium
2 Department of Clinical Chemistry, Microbiology and Immunology, Faculty of Medicine and Health Sciences, Ghent University, Ghent University Hospital, De Pintelaan 185, B-9000 Ghent, Belgium
3 Department of Intensive Care, Faculty of Medicine and Health Sciences, Ghent University, Ghent University Hospital, De Pintelaan 185, B-9000 Ghent, Belgium
BMC Infectious Diseases 2005, 5:55 doi:10.1186/1471-2334-5-55Published: 6 July 2005
Bacterial vaginosis (BV) is the single most common vaginal infection in women of childbearing age and associated with a sizeable infectious disease burden among both non-pregnant and pregnant women, including a significantly elevated risk of adverse pregnancy outcome. Overall, little progress has been made in identifying causal factors involved in BV acquisition and persistence. We sought to evaluate maternal iron status in early pregnancy as a putative risk factor for BV, considering that micronutrients, and iron deficiency in particular, affect the host response against bacterial colonization, even in the setting of mild micronutrient deficiencies.
In a nested case-control study, we compared maternal iron status at entry to prenatal care (mean gestational age 9.2 ± 2.6 weeks) between eighty women with healthy vaginal microflora and eighteen women with vaginosis-like microflora. Vaginal microflora status was assessed by assigning a modified Nugent score to a Gram-stained vaginal smear. Maternal iron status was assayed by an array of conventional erythrocyte and serum indicators for iron status assessment, but also by more sensitive and more specific indicators of iron deficiency, including soluble transferrin receptors (sTfR) as an accurate measure of cellular and tissue iron deficiency and the iron deficiency log10[sTfR/ferritin] index as the presently most accurate measure of body storage iron available.
We found no statistically significant correlation between vaginal microflora status and routinely assessed iron parameters. In contrast, a highly significant difference between the healthy and vaginosis-like microflora groups of women was shown in mean values of sTfR concentrations (1.15 ± 0.30 mg/L versus 1.37 ± 0.38 mg/L, p = 0.008) and in mean iron deficiency log10[sTfR/ferritin] index values (1.57 ± 0.30 versus 1.08 ± 0.56, p = 0.003), indicating a strong association between iron deficiency and vaginosis-like microflora. An sTfR concentration >1.45 mg/L was associated with a 3-fold increased risk (95%CI: 1.4–6.7) of vaginosis-like microflora and after controlling for maternal age, gestational length, body mass, parity, and smoking habits with an adjusted odds ratio of 4.5 (95%CI: 1.4–14.2).
We conclude that subclinical iron deficiency, presumably resulting from inadequate preconceptional iron supplies, is strongly and independently associated with vaginosis-like microflora during early pregnancy.