Unlocking insights: Validating a comprehensive model for predicting spontaneous preterm delivery and microbial...

Preterm labor is a multifactorial syndrome with diverse etiologies such as infection, inflammation, uteroplacental dysfunction, stress, and immune dysregulation. Identifying women at high risk remains a challenge, and multivariate prediction models have been developed to enhance risk stratification. The primary objective of this study was to externally validate Cobo's prediction models for spontaneous delivery within 7 days and microbial invasion of the amniotic cavity (MIAC) in a new population.

This retrospective observational study aimed to externally validate Cobo's prediction models for spontaneous delivery within 7 days and MIAC in an independent cohort. A total of 130 pregnant women with threatened preterm labor or suspected intra-amniotic infection were included at Vall d'Hebron University Hospital between October 2016 and September 2019. Amniocentesis was performed, and both amniotic fluid and maternal blood samples were analyzed, including measurement of interleukin-6 levels in amniotic fluid. Model performance was evaluated using discrimination (area under the receiver operating characteristic curve) and calibration.

Among the 130 women included in the study, chorioamnionitis was diagnosed in 43 (33.1%) cases. Positive amniotic fluid and endocervical cultures were found in 15 and 47 cases, respectively. Spontaneous delivery within 7 days occurred in 47 (36.2%) women, and MIAC was confirmed in 15 of 128 (11.7%) cases. The area under the receiver operating characteristic curve was 0.805 (95% confidence interval [CI]: 0.715-0.895) for predicting delivery within 7 days and 0.897 (95% CI: 0.800-0.995) for MIAC. Calibration was poor for delivery within 7 days (intercept 0.32; slope 3.65) but more accurate for MIAC (intercept -0.02; slope 0.91).

Cobo's models demonstrated strong discriminatory capacity but suboptimal calibration for delivery within 7 days. Calibration for MIAC was more accurate, supporting their clinical utility with room for refinement.