Prediction of spontaneous preterm birth using CCL2 and CXCL10 in maternal serum of symptomatic high-risk pregnant women: a prospective cohort study

Introduction CCL2 and CXCL10 are putative biomarkers for the prediction of spontaneous preterm birth. This study evaluates these markers in a cohort of pregnant high-risk women.Material and methods In our prospective study, we included 109 women with signs of preterm labor between 20 + 0 and 31 + 6 weeks of gestation. Inclusion criteria were regular (< 3/30 min) or painful contractions, cervical length < 25 mm or a history of previous preterm birth (PTB). Blood samples were obtained upon first admission to our clinic. Biomarker concentrations were measured using pre-coated sandwich immunoassays (ELISA). Primary study outcome was spontaneous preterm birth < 34 weeks, secondary outcome was delivery < 37 weeks or within seven days after study inclusion.Results Sixteen women (14.7%) delivered < 34 weeks and twenty women between 34 + 0 and 36 + 6 weeks (18.4%). Six patients (5.5%) gave birth within seven days after study admission. CXCL10 showed higher medium serum levels in women with PTB < 34 weeks (115 pg/ml compared to 61 pg/ml >= 34 weeks; p < 0.001) and < 37 weeks (103 pg/ml vs. 53 pg/ml; p < 0.001). In contrary, lower CCL2 serum levels were associated with PTB < 34 weeks (46 pg/ml vs. 73 pg/ml; p = 0.032) and birth within 7 days (25 pg/ml vs. 73 pg/ml; p = 0.008). The CXCL10/CCL2-ratio further improved the predictive model with a ROC-AUC of 0.83 (95% CI 0.73-0.93, p < 0.001) for delivery < 34 weeks. These corresponds to a sensitivity, specificity and positive predictive value of 0.67, 0.86 and 0.43 at a cut-off of 2.2.Conclusion Low maternal serum CCL2 levels are associated with a higher risk of preterm delivery within seven days. High CXCL10 serum levels are more associated with a high risk for preterm birth < 34 weeks. Elevated CXCL10/CCL2-ratio is showing the best predictive performance.