High sialidase levels increase preterm birth risk among women who are bacterial vaginosis–positive in early gestation




Objective


The purpose of this study was to assess whether vaginal sialidases level in early pregnancy is associated with preterm birth among women who are bacterial vaginosis–positive.


Study Design


Of the 1806 women who were enrolled at <20 weeks of gestation, 800 of the women were bacterial vaginosis–positive (Nugent score, 7-10); 707 of the women had birth outcome data; 109 of the women who were bacterial vaginosis–positive had an adverse preterm outcome, which included 53 spontaneous preterm births (19 births were early at 20-34 weeks, and 34 births were late at 34-37 weeks), and 14 of the women had late miscarriages (12-20 weeks). Sialidase levels were compared with 352 control subjects (term normal birthweight infants).


Results


Sialidase levels at ≥5, ≥10, and ≥14 nmol (odds ratio [OR], 1.56; 95% confidence interval [CI], 1.01–2.41; OR, 2.14; 95% CI, 1.25–3.64; OR, 3.17; 95% CI, 1.64–6.10, respectively) was associated significantly with all adverse preterm outcomes. The ≥10 nmol and ≥14 nmol cut-points were associated strongly with early spontaneous preterm births (OR, 3.79; 95% CI, 1.42–10.10 and OR, 5.36; 95% CI, 1.77–16.23, respectively) and late miscarriages (OR, 4.87; 95% CI, 1.61–14.65; OR, 8.33; 95% CI, 2.57–26.9, respectively).


Conclusion


Elevated sialidase level that is measured at 12 weeks of gestation is associated strongly with early spontaneous preterm births and late miscarriage.


Preterm delivery (<37 weeks of gestation) remains the most common determinant of neonatal death, morbidity, and long-term neurodevelopmental disorders; early preterm births account for most of such adverse outcomes. Preterm birth is a major public health problem >12% of all US pregnancies end at <37 weeks of gestation. Despite extensive research and intensive medical and public health efforts, the rate of preterm birth has not declined recently and may even have increased. Several investigations have shown that bacterial vaginosis (BV), which is an altered vaginal flora condition in which the normal lactobacilli flora is substituted with a mixed and variable flora of microaerophilic and anaerobic microorganisms, is associated generally with preterm birth, early preterm birth, early and late miscarriage, low birthweight (<2500 g), and maternal complications. Although the association between BV and preterm birth is consistent in the epidemiologic literature, only approximately 10-30% of all pregnant women who are BV-positive give birth prematurely. Thus, BV is a nonspecific marker of risk, and more refined predictive markers are needed. Recently, some European studies found that BV-associated enzyme biomarkers are associated more strongly with early preterm birth and low birthweight when compared to BV by itself.


Numerous randomized clinical trials of antibiotic therapy to reduce preterm birth have been conducted in pregnant women who were selected on the basis of various risk factors with mixed results. One explanation for this may be inappropriate inclusion of women with an actual low risk of preterm birth. Inclusion of women with altered vaginal flora, but with very little risk of adverse birth outcomes, not only could mask real benefit within the truly at risk subset but also potentially could cause harm to the “healthy” women. Thus, there is an urgent need to develop strategies to refine criteria for inclusion in antibiotic treatment trials that target abnormal vaginal flora to reduce the risk of preterm birth. We focused our attention on the potential role of sialidase (neuraminidase) activity as such a marker, because recent findings highlight that sialidase activity has complex roles in the modulation of host local response that culminates in impaired lower genital-tract defenses.


Recent research shows that local host response varies widely among women who are BV-positive and that microbial factors seem to modulate the vaginal immunity. Specifically, vaginal sialidase activity was observed to be correlated inversely with the adaptive response against the Gardnerella vaginali s toxin but was correlated positively with vaginal interleukin (IL)-1β. In addition, sialidase activity combined with elevated vaginal pH has been associated with increased risk of adverse pregnancy outcomes, particularly low birthweight, in pregnant Danish women.


The intent of the present study was to explore whether a subgroup of pregnant women who were BV-positive and who were at a truly elevated risk of adverse birth outcomes could be identified from analysis of vaginal secretions that were obtained in early gestation. The ultimate goal was to refine risk assessment to guide the development of more efficacious prevention strategies that would lead ultimately to the reduction of adverse pregnancy outcomes. We explored the association of progressive levels of sialidase activity, with all adverse preterm pregnancy outcomes, and specific subsets of these outcomes among pregnant women who were BV-positive from whom samples were obtained early in gestation.


Materials and Methods


Study design


The primary aim of this study was to ascertain whether a subset of women who were BV-positive who were truly at risk could be identified through vaginal biomarkers in early pregnancy. The study used a prospective design in which all women who enrolled for prenatal care at <20 weeks of gestation were asked to participate. For those women who met the eligibility criteria and provided written informed consent, vaginal secretions were obtained for diagnosis of BV and banking until birth outcome was determined. Vaginal secretions from all women who were BV-positive who delivered before term were analyzed, as were randomly selected samples of control subjects among women who were BV-positive who delivered term, normal birthweight infants (≥37 weeks of gestation and ≥2500 g). All women who attended their first prenatal visit at 3 hospital-based clinics located in Philadelphia between January 2002 and September 2004 were screened for eligibility. Women were considered eligible for participation if they (1) were <20 weeks pregnant at the time of enrollment, (2) spoke either English or Spanish, (3) had a singleton intrauterine pregnancy, (4) were not positive for the human immunodeficiency virus, (5) did not seek a therapeutic abortion, (6) were ≥18 years old, (7) underwent a routine pelvic examination at the enrollment visit, and (8) had no vaginal bleeding at the time of sample collection. We did not exclude women who reported recent vaginal intercourse, based on our own preliminary work. In an analysis of vaginal secretions that were obtained from 302 women who were BV-positive from this cohort, elevated sialidase levels were not correlated with elevated prostate-specific antigen, which is considered the best marker of seminal fluid contamination of vaginal samples. Furthermore, we did not exclude women who entered prenatal care taking antibiotics or women who were prescribed antibiotics at the time of their first prenatal care visit. We based this decision on the fact that women (n = 199) who were taking antibiotics or who were prescribed antibiotics (including penicillin, ampicillin, amoxicillin, aminoglycosides, clindamycin, metronidazole, macrolides, cephalosporins, and others) did not have a different distribution of preterm birth outcomes compared with those women who did not take antibiotics.


Collection of medical data and specimens


At the time of enrollment, 7 Dacron swabs were used to collect vaginal secretions from the posterior wall of the vaginal fornix and were processed as described elsewhere. Vaginal samples were stored frozen in liquid nitrogen soon after collection and shipped on dry ice to a core laboratory, where all measurements were performed. Gram-stained smears from all women who were eligible and consented to participate (n = 1806; Figure 1 ) were evaluated according to the Nugent score by personnel who were blinded to all other data. BV was defined as a Nugent score of 7-10. The BV diagnosis was not disclosed to the women or their care providers. Only women who were BV-positive were examined further. Pregnancy outcome and gestational age was ascertained after delivery by medical records. Records were also abstracted to determine whether early births were spontaneous or medically indicated. To ensure the accuracy of the information, all medical records were abstracted by trained personnel, and all controversial aspects of the chart were flagged and reviewed by 3 consulting physicians. Sociodemographic, health behavior, and medical history were collected in face-to-face interviews by trained study staff.




FIGURE 1


Sample selection for 461 study participants who were bacterial vaginosis ( BV )– positive

Cauci. Sialidase as biomarker for adverse preterm outcomes. Am J Obstet Gynecol 2011.


Definitions of outcome


Estimates of gestational age were based on the first ultrasound scan for most of the 461 study participants (84.4%; n = 389) in the final analytic sample. For those women (n = 72; 15.6%) without an ultrasound scan, gestational age was determined by the date of the last menstrual period and confirmed by obstetric examination. Among all women who were BV-positive, pregnancies that ended in live birth, stillbirth, or miscarriage were classified into categories based on gestational age and medical record information. Adverse pregnancy outcome groups were compared with term, normal birthweight deliveries. Of the 109 adverse preterm outcomes, 53 women had spontaneous preterm births (SPTBs), of which 19 births were early SPTB (20-34 weeks of gestation) and 34 births were late SPTB (34-37 weeks of gestation). The following remaining adverse outcomes were found: 17 early miscarriages (<12 weeks), 14 late miscarriages (12-20 weeks), 13 medically induced preterm births, 3 preterm births with no information about induction (not further examined in detail), and 9 stillbirths. Unadjusted comparisons between the term normal weight infants, all adverse preterm outcomes, and the 6 discrete subsets were performed on all 3 levels of sialidase activity (5, 10, and 14 nmol of converted substrate).


Evaluation of sialidase activity


Sialidase activity was determined in duplicate as described previously and expressed as nanomoles of converted substrate (methoxyphenol). Women with sialidase activity of ≥0.19 nmol were considered positive for any sialidase. The cut point of sialidase ≥5 nmol was chosen on the basis of previous findings and corresponded to the 62nd percentile in the present population. In addition, cutoffs corresponding to the 85th percentile (≥10 nmol) and 90th percentile (≥14 nmol) were evaluated.


Statistical analysis


Bivariate comparisons, with the use of t and χ 2 tests, were conducted to compare women with sialidase activity above and below the 85th percentile (≥10 nmol) across a wide range of sociodemographic, behavioral, and medical characteristics. A univariable multinomial logistic regression model was used to evaluate the odds ratios (ORs) and 95% confidence intervals (CIs) for elevated sialidase levels that compared the mutually exclusive specific adverse outcomes with the outcomes of the control group. Multiple regression models were conducted for adverse outcome categories and were adjusted for the following variables: age, race, education, marital status, US born, annual income, history of sexually transmitted disease, history of BV, and smoking during current pregnancy. Variables were considered to be candidates for control variables in the regression models based on published literature and based on theoretic considerations. Final variables were selected for the regression models if they were associated with elevated sialidase levels at a probability value of < .25 or if the variables are controlled for commonly in the literature. Given that women with a previous preterm birth are more likely to have a subsequent preterm birth and that the retention of this risk may be reflected in stable alterations of vaginal biomarkers that are associated with preterm birth, models were not adjusted for previous preterm birth to avoid over control. However, in an exploratory analyses, we examined the effect of previous preterm birth on the association between elevated sialidase and risk of preterm adverse outcomes. Adjustment for previous preterm birth did not alter any of our findings (data not shown). Alpha was set at .05 (2-sided); Stata software (version 10.0; Stata Corporation, College Station, TX) was used for all analyses. We assessed the sensitivity, specificity, positive predictive value, and negative predictive value of elevated sialidase on the combined early preterm and late miscarriages after adjustment of the sampling fraction. Additionally, we computed Kaplan-Meier estimates of probability of delivery at specific weeks of gestation and compared the low sialidase category at <5 nmol of converted substrate with elevated categories of sialidase at ≥5, ≥10, and ≥14 nmol of converted substrates. The log-rank test was used to evaluate differences in survival functions between the groups. This research was approved by the institutional review boards at Thomas Jefferson University, Drexel University College of Medicine, and the University of Udine.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jun 21, 2017 | Posted by in GYNECOLOGY | Comments Off on High sialidase levels increase preterm birth risk among women who are bacterial vaginosis–positive in early gestation

Full access? Get Clinical Tree

Get Clinical Tree app for offline access