Evaluation of semen detection in vaginal secretions: comparison of four methods.

PROBLEM: To determine the best method to detect semen in human vaginal secretions. METHOD OF STUDY: Vaginal secretions from 302 pregnant women at mean 11.8 weeks' gestation were analyzed. Semen detection was assessed with: (i) measurement of total prostate-specific antigen (PSA), (ii) acid phosphatase activity, (iii) microscopic measurement of spermatozoa on Gram stain, and (iv) self-reported sexual intercourse in the past 2 days. Sensitivity and specificity were calculated for each technique in comparison with PSA levels. RESULTS: A total of 119 (39.4%) women had a detectable PSA. Compared with measurable PSA, the sensitivity and specificity for other methods were: acid phosphatase (26.9%, 98.4%), Gram stain (36.1%, 98.4%), and self-report of intercourse in the past 48 hr (41.9%, 88.8%). CONCLUSION: Compared with PSA levels, commonly used assays for recent semen exposure are inaccurate. This inaccuracy may affect the results of studies, which measure vaginal immune factors like cytokines or retrieve DNA from vaginal specimens.

Evaluation of iron deficiency in young women in relation to oral contraceptive use.

BACKGROUND: The purpose of this study was to identify the optimal measures for diagnosing iron deficiency (ID) in oral contraceptive (OC) users and nonusers, and to estimate ID frequency in relation to OC use. STUDY DESIGN: Conventional biomarkers of iron status - serum ferritin, iron, transferrin (Tf) and transferrin saturation (TfS) - were compared with serum soluble Tf receptor (sTfR) and the sTfR/log ferritin ratio (sTfR-F index). Two hundred two healthy menstruating white Italian women (aged 24+/-4.8 years) were analyzed. Serum ferritin concentrations <12 microg/L were considered as ID. RESULTS: ID was detected in 29.7% (60/202) of the study women. Fifty-nine women were OC users (59/202, 29.2%). OC use did not significantly affect ID prevalence (p=.24). However, OC use markedly increased Tf in OC users, who had an odds ratio (OR) of 9.3 (CI 3.8-22.7, p<.001) for elevated Tf >330 mg/dL. No other iron status measure was affected by OC. Of the markers for ID adjunctive to ferritin, an elevated sTfR-F index >or =1.5 showed the best performance. Specifically in OC users, the elevated sTfR-F index had better sensitivity (81.0% vs. 33.3%), specificity (94.7% vs. 92.1%), efficiency (89.8% vs. 71.2%), positive predictive value (89.5% vs. 70.0%) and negative predictive value (90.0% vs. 71.1%) than a TfS <15%. Additionally, the sTfR-F index allowed the identification of low iron stores in 4.5% (9/202) of women with ferritin > or =12 microg/L. CONCLUSION: Among healthy OC users and non-OC users, the sTfR-F index had the highest performance for diagnosing ID compared with other serum markers adjunctive to ferritin measurements, whereas sTfR by itself had a low sensitivity. We showed that neither the sTfR nor sTfR-F index was affected by third-generation OC use. The sTfR measurement is useful in the diagnosis of ID, especially in women using OC, where Tf and TfS tests may be misleading.

Association of interleukin-1beta and interleukin-1 receptor antagonist polymorphisms with bacterial vaginosis in non-pregnant Italian women.

Bacterial vaginosis (BV) is the most prevalent alteration of vaginal microflora worldwide. BV is a polymicrobial disorder, and its etiology is elusive. Factors predisposing to this recurrent condition are not fully characterized. We aimed to investigate whether interleukin-1beta (IL-1beta) and IL-1 receptor antagonist (IL-1ra) polymorphisms are associated with BV in non-pregnant white Italian women. Genomic DNA was obtained from 164 BV positive, and 406 control women. Two diallelic polymorphisms in the IL-1beta gene (IL-1B) representing C/T base transitions at - 511 and + 3954 positions and a variable number tandem repeats (VNTR) in intron 2 of the IL-1ra gene (IL-1RN) were assessed. We demonstrated that women who were homozygous for - 511 CC or + 3954 TT of the IL-1B gene were at increased risk for BV with an odds ratio (OR) = 1.5 [95% confidence interval (CI) = 1.03-2.14, P = 0.032], and OR = 2.8 (95% CI = 1.37-5.88, P = 0.004), respectively. The haplotype - 511/ + 3954 T-C was protective for BV, with an OR = 0.7 (95% CI = 0.49-0.90, P = 0.009). The IL-1RN VNTR genotype was not associated with BV, although the rare allele 3 showed a trend towards protection (P = 0.049). These data show that host genetic variants at the IL-1beta locus predispose to BV among Caucasian non-pregnant women. Further studies will determine whether these genetic polymorphisms modulate the risk for BV recurrence, and/or BV associated severe adverse outcomes as preterm birth and human immunodeficiency virus transmission.