Maternal C-reactive protein concentrations during pregnancy and birth weight in a prospective cohort in Rio de Janeiro, Brazil.

Objective: To evaluate the association between maternal C-reactive protein (CRP) concentrations during pregnancy and birth weight (BW) Z-score.

Methods: A prospective cohort of pregnant women were followed at 5-13 (n = 203), 20-26 (n = 181), and 30-36 (n = 181) gestational weeks and at 30-45 d postpartum. Maternal CRP concentrations were assessed three times during pregnancy using immunoturbidimetric methods (ultra-sensitive kits). BW Z-score and newborns classified as small for gestational age (SGA) were evaluated according to Intergrowth-21st curves. Statistical analyses included SGA rates, BW Z-score means (SD) and a two-stage procedure: (1) a linear mixed-effect model (LME) to predict CRP intercept (mean exposure level) and slope (trend of change during pregnancy); and (2) a multiple linear regression model with BW Z-score as the outcome and CRP intercept and slope exposures.

Results: A total of 4.4% (n = 9) women delivered SGA newborns. The mean BW was 3282.0 (37.3) g, and the mean gestational age at delivery was 38.8 (0.1) weeks. Women in the third tertile of the CRP rate of change gave birth to infants with a mean BW Z-score that was lower than those in the first/second tertiles (0.226 versus 0.381; p = 0.324). For the adjusted baseline CRP (β = 0.08; 95% CI: 0.03-0.14), the CRP trend of change was inversely associated with the BW Z-score (β= -3.77; 95% CI: -5.45 to -2.10).

Conclusions: The maternal CRP trend of change during pregnancy was negatively associated with BW Z-score.