Design and population

This study was embedded in the Generation R Study, a population-based prospective cohort study from early fetal life onward in the city of Rotterdam, The Netherlands. Enrollment was aimed in early pregnancy but was allowed until birth of the child. All mothers were enrolled from 2001 through 2005. Response rate was 61%. The study was approved by the Medical Ethical Committee of the Erasmus MC, Rotterdam. Written informed consent was obtained from all participants.

Of the total of 8880 mothers who were enrolled during pregnancy, 76% (n = 6748) were enrolled before a gestational age of 18 weeks. Of these mothers, blood plasma samples were collected in 95% (n = 6398) and CRP was successfully measured in 90% (n = 6091). Mothers with extremely high CRP levels (>100 mg/L) (n = 6), and mothers with twin pregnancies (n = 69) were excluded, leaving 6016 mothers with singleton live births for analysis ( Supplementary Figure S1 ).

Flowchart of study participants

Ernst. C-reactive protein levels in pregnancy and fetal growth. Am J Obstet Gynecol 2011.

High-sensitivity CRP levels

Maternal venous blood samples were collected in early pregnancy (median, 13.2; 95% range, 9.6–17.6 weeks) and transported to the regional laboratory (Star-MDC, Rotterdam, The Netherlands) for processing and storage. Blood samples were stored at −80°C. CRP concentrations were measured in EDTA plasma samples at the Department of Clinical Chemistry of the Erasmus MC in 2009. We measured high-sensitivity CRP since traditional clinically used CRP methods lack the sensitivity in low ranges needed for predicting future risk of events in apparently healthy individuals. CRP levels were analyzed using an immunoturbidimetric assay on the Architect System (Abbot Diagnostics B.V., Hoofddorp, The Netherlands). The within run precision for CRP was 1.3% at 12.9 mg/L and 1.2% at 39.9 mg/L. The lowest level of detection was 0.2 mg/L. We created 6 categories of CRP levels (<5.0, 5.0-9.9, 10.0-14.9, 15.0-19.9, 20.0-24.9, and ≥25 mg/L). Levels <5.0 mg/L and ≥25 mg/L were considered as low (reference) and elevated levels, respectively.

Fetal growth characteristics

Fetal ultrasound examinations were performed in 1 of the 2 dedicated research centers in each trimester of pregnancy. Median gestational age for first, second, and third trimester visits were 12.4 weeks (95% range, 10.7–14.5), 20.5 weeks (95% range, 18.7–23.1), and 30.4 weeks (95% range, 28.6–32.8), respectively. In the second and third trimester of pregnancy, we measured fetal head circumference, abdominal circumference, and femur length to the nearest millimeter using standardized ultrasound procedures. Estimated fetal weight was calculated using the formula by Hadlock et al. Longitudinal growth curves and gestational age-adjusted SD scores (SDS) were constructed for all fetal growth measurements.

Information about offspring sex, gestational age, weight, length, and head circumference at birth was obtained from medical records and registries. Since head circumference and length at birth were not routinely measured at birth, missing birth measures were completed with data from the first-month visit at the routine child health center. Preterm birth was defined as a gestational age of <37 weeks at delivery, low birthweight was defined as birthweight <2500 g, and SGA at birth was defined as a sex-specific gestational age-adjusted birthweight below the fifth percentile in the study cohort (SDS ≤1.81 for boys and ≤1.78 for girls).

Covariates

Information about maternal educational level, ethnicity and parity was obtained by a questionnaire at enrollment in the study. Maternal smoking and alcohol consumption habits were assessed by questionnaires in each trimester. Maternal anthropometrics, including height and weight, were measured without shoes and heavy clothing and body mass index (BMI) was calculated (weight/height ² [kg/m ² ]) at enrollment. Maternal systolic and diastolic blood pressure were measured at intake, using standardized methods. For each participant, the mean value of 2 blood pressure readings over a 60-second interval was documented. Folate levels were analyzed from venous samples drawn in the first trimester of pregnancy. Maternal age was registered at enrollment.

Statistical analysis

We assessed the associations of maternal characteristics with CRP levels as outcome levels using multivariate linear regression models. Since CRP levels were not normally distributed, we applied a logarithmic transformation for these analyses. Results are presented as geometric means (95% range) per determinant category and an overall P for trend based on these regression models. Associations of CRP levels with fetal growth characteristics were assessed using linear regression models. These models were adjusted for gestational age at the measurement, fetal sex, and maternal age, BMI, education, ethnicity, parity, smoking, alcohol consumption. BMI is known to be highly correlated with levels of CRP, and with birthweight, and therefore expected to be our main confounder. Further variables were included in these models based on their association with both the CRP levels and pregnancy outcomes, or a 10% change in the effect estimate. Next, we assessed the associations of CRP level with the risks of neonatal complications (preterm birth, low birthweight, and SGA). These models were adjusted for maternal age, BMI, education, ethnicity, parity, smoking, alcohol consumption and folic acid level at intake. Tests for trends were performed using CRP as continuous variable in multivariate linear and logistic regression analyses. The percentages of missing values within the population for analysis were <1% for continuous data and <13% for the categorical data. We applied multiple imputation for covariates. Since there were no major differences in the observed results between analyses with imputed missing data or complete cases only, only results including imputed missing data are presented. All measures of association are presented with their 95% confidence intervals (CIs). All statistical analyses were performed using the SPSS, version 17.0 for Windows (SPSS Inc, Chicago, IL).