Placental inflammation and its relationship to cervicovaginal fetal fibronectin in preterm birth.

OBJECTIVE: Late miscarriage and preterm birth are frequently thought to be associated with inflammation and infection, although in most cases the underlying cause of early delivery remains unknown. The placenta is the organ that links mother and fetus during pregnancy, and postnatal examination may provide useful information about pathophysiology. The relationship between placental pathological lesions and predictive markers of early delivery has not been explored. We sought to characterize preterm deliveries according to placental pathology and relate these to the performance of reliable predictive markers, fetal fibronectin and cervical length. STUDY DESIGN: This is a retrospective subanalysis from a larger prospective cohort study on sonographic cervical length, quantitative fetal fibronectin and risk of spontaneous preterm birth. Our cohort was comprised of high-risk asymptomatic women attending the Prematurity Surveillance Clinic at St Thomas' Hospital between 2002 and 2015, who went on to have a late miscarriage or preterm delivery (16-36+6 weeks') and who had available placental histology. The placental pathology of these preterm deliveries was characterized according to the lesions identified, and categorized (according to the Redman classification) into inflammatory (e.g. chorioamnionitis) or non-inflammatory (histologically normal or vascular lesions indicating e.g. malperfusion). We sought to relate placental findings to the performance of reliable predictive markers, in women who delivered early. Standard clinical cut offs for cervical length (<25mm) and fetal fibronectin (>50ng/mL) were used to identify the proportion of preterm births that were accurately predicted by the tests or who showed a false negative result, in relation to their placental histology findings. Binomial logistic regression was carried out to evaluate the relationship between placental inflammation, quantitative fFN and cervical length as continuous variables. RESULTS: 105 women who had a late miscarriage or preterm delivery (16-36+6 weeks') and available placenta pathology were identified. 66% (42/64) of those with inflammatory placental pathology had a positive fetal fibronectin swab result compared to 15% (6/41) of those with non-inflammatory placental pathology (chi-squared 25.9, 95% CI 31.5 to 65.6, p<0.0001). A logistic regression model subanalysis of women in whom both CL and quantitative fFN results were available (n=66) revealed a highly statistically significant relationship with inflammatory placental lesions (p=0.003 and p=0.001 respectively). Placental inflammation was found to be associated with both increasing levels of fFN and a shortening cervix. CONCLUSION: There is a significant association between a positive fetal fibronectin result and underlying inflammatory pathology of the placenta, even more so than the recognized relationship with short cervical length. Infective morbidity may be increased in women and neonates with positive fetal fibronectin who deliver preterm.

Prediction of spontaneous preterm birth using fetal fibronectin in women with a low-lying placenta.

OBJECTIVE: To determine the effect of a low-lying placenta on the concentration of quantitative fetal fibronectin (qfFN) in the cervicovaginal fluid (CVF), and predictive accuracy for spontaneous preterm birth in asymptomatic high-risk women (18 + 0-24 + 0 weeks gestation). METHODS: Median concentrations of qfFN were compared in women who had a low-lying placenta, covering the cervical os (n = 61) to matched controls (n= 61) without a low-lying placenta. Proportions of women with raised qfFN concentrations (>10 ng/ml), and false positive and negative rates (FPR and FNR) for spontaneous preterm delivery were also compared. RESULTS: The median concentration of qfFN in women with low-lying placenta was 5.0 ng/mL, compared with 6.0 ng/mL in controls. Proportion of women with raised levels (>10 ng/mL), positive levels (>50 ng/mL) and very high levels (>200 ng/mL) were similar in both groups (62.3% versus 59.0%, 16.3% versus 22.0% and 6.5% versus 4.9%, p > 0.05 for all thresholds). The FPR and FNR rate for delivery before 34 and 37 weeks were also comparable (FPR 90.0% versus 85.7% and 80.0% versus 78.6%; FNR 5.8% versus 4.3% and 9.8% versus 8.5%). CONCLUSIONS: CVF qfFN concentrations in asymptomatic high-risk women are not affected by the presence of a low-lying placenta.