Meta-regression analysis to evaluate relationships between maternal blood levels of placentation biomarkers and low delivery weight

Background: Caution is required for women at increased risk of low neonatal delivery weight.

Objective: To evaluate relationships between maternal placentation biomarkers and the odds of low delivery weight.

Search strategy: Databases including PubMed/MEDLINE were searched up to May 2017 using keywords involving biomarker names and "low birthweight."

Selection criteria: English language studies providing true- and false-positive, and true- and false-negative results of low delivery weight classified by maternal blood levels of placentation biomarkers (in units of multiple of the mean [MoM]) were included.

Data collection and analysis: Coefficients representing changes in log odds ratio for low delivery weight per 1 MoM increase in maternal blood placentation biomarkers, and those adjusted for race, sampling period, and/or study quality were calculated.

Main results: Adjusted coefficients representing changes in log odds ratio for low delivery weight per 1 MoM increase in maternal blood levels of α-fetoprotein (AFP) and β-human chorionic gonadotropin (β-hCG) were significantly greater than 0 (both P<0.001), whereas that for pregnancy-associated plasma protein A (PAPP-A) was significantly less than 0 (P=0.028). Adjusted models explained the higher proportion of between-study variance better than non-adjusted models.

Conclusions: Elevated AFP and β-hCG, and reduced PAPP-A in maternal blood were positively associated with odds of low delivery weight.