Maternal serum neuroserpin in late-onset fetal growth restriction: a prospective case-control study

To determine whether maternal serum neuroserpin is independently associated with late-onset fetal growth restriction (LO-FGR) after accounting for maternal body mass index (BMI) and gestational age at blood sampling, and whether it adds discrimination beyond clinical and sonographic variables.

This single-center prospective case-control study included 45 pregnancies with Delphi-defined LO-FGR and 45 prospectively recruited control pregnancies. Maternal blood was obtained at the visit when LO-FGR was first confirmed, before any treatment or delivery-related intervention, and at the corresponding routine antenatal assessment in controls. Serum neuroserpin was measured by sandwich enzyme-linked immunosorbent assay after completion of clinical follow-up. Associations with LO-FGR were evaluated using Firth penalized logistic regression, and discrimination was assessed using receiver operating characteristic analysis and paired DeLong tests. Neuroserpin tertiles and composite adverse perinatal outcome (CAPO) were examined within the LO-FGR group.

Neuroserpin concentrations were lower in LO-FGR than in controls (median, 21.76 vs 37.33 ng/mL; p < 0.001). Each 10-ng/mL lower concentration was associated with higher odds of LO-FGR after adjustment for BMI and gestational age at sampling (adjusted odds ratio, 1.40; 95% confidence interval, 1.13-1.73; p < 0.001). The association remained significant after additional adjustment for cerebroplacental ratio but was attenuated after abdominal circumference percentile was included. Adding neuroserpin increased the clinical-model area under the curve from 0.751 to 0.832 (DeLong p = 0.020), but did not significantly improve models already containing cerebroplacental ratio or abdominal circumference percentile. Across neuroserpin tertiles, no statistically significant differences were observed in fetal biometry, Doppler indices at diagnosis, or adverse perinatal outcomes; these exploratory analyses were underpowered.

Lower maternal serum neuroserpin was independently associated with LO-FGR and added information beyond simple clinical variables. Its incremental value beyond established sonographic variables and its prognostic value after diagnosis remain uncertain. These single-center findings require multicenter validation and direct comparison with established angiogenic biomarkers.