BACKGROUND: Cerebral palsy (CP) is the most common severe motor disability and a manifestation of early brain damage.

AIMS: To analyze if abnormal levels of first-trimester biomarkers were associated with CP. Furthermore, to investigate their clinical applicability in early predicting of CP.

STUDY DESIGN: Nationwide cohort study.

SUBJECTS: We included 258.057 singleton live births, born during 2008-2013 with completed first-trimester assessments.

OUTCOME MEASURES: Data on beta subunit of human chorionic gonadotropin (beta-hCG), pregnancy-associated plasma protein-A (PAPP-A), nuchal translucency thickness, and biparietal diameter (BPD) were converted to multiple of the medians (MoM). Associations were analyzed by comparing mean and extreme levels between pregnancies with and without CP. All CP diagnoses were validated by trained neuropediatricians. Logistic regression was used to create an early prediction model.

RESULTS: The mean beta-hCG value was significantly lower in pregnancies with CP (0.96MoM [95% CI 0.91-1.02] vs 1.04MoM [1.04-1.04], p = 0.01) and the mean PAPP-A value tended to be lower (0.96MoM [0.91-1.01] vs 1.01MoM [1.00-1.01], p = 0.07). Moreover, fetuses that developed CP more likely had a BPD measurement below the fifth percentile (7.5% vs 5%, p = 0.045). The final prediction model had poor discrimination.

CONCLUSIONS: Pregnancies with CP tend to have lower values of beta-hCG and PAPP-A in the first trimester, however, the associations are mediated differently. Nonetheless, abnormal levels of the most common first-trimester biomarkers only have weak associations with CP; resulting in inadequate predictive abilities when included in an early prediction model.