Maternal age at first cesarean delivery related to adverse pregnancy outcomes in a second cesarean delivery: a multicenter, historical, cross-sectional cohort study

This was a multicenter, historical, cross-sectional cohort study conducted at 11 public tertiary hospitals covering 7 provinces, municipalities, and autonomous regions within China (Guangdong, Beijing, Xinjiang, Shanxi, Henan, Hubei, and Chongqing). The cohort comprised 14,734 women with uterine scars who delivered again between January 2017 and December 2017. We selected women with singleton pregnancies at ≥28 gestational weeks, and with a CD history who underwent a repeat CD. Antepartum fetal death, major fetal congenital anomalies, a scarred uterus caused by myomectomy, and a history of 2 or more CDs were excluded. Women lacking their essential records—such as delivery mode or severe data loss—were also excluded. Figure 1 shows a flow diagram of the women’s enrollment process.

Data were obtained by chart review based on electronic medical records. Maternal clinical characteristics included maternal ages at the first and second CDs, gestational weeks, gravidity, parity, nationality, mode of conception (natural vs. assisted), source of pregnant women (“referral” meant that pregnant women were referred from low-level hospitals to tertiary hospitals. “hospital” meant that pregnant women delivered in a tertiary hospital from the beginning), body mass index (BMI) before pregnancy, number of months (interval) between the 2 CDs, sex of the offspring, and indications for the 2 CDs.

We reviewed the electronic medical records with regard to complications, and the following details were recorded: premature rupture of membranes (PROM), placenta previa (PP), placenta accreta spectrum disorders (PAS), placental abruption, idiopathic thrombocytopenic purpura (ITP), abnormal amniotic fluid (oligohydramnios and polyhydramnios), hypertension disorders, diabetes mellitus (DM), thyroid diseases (hypothyroidism and hyperthyroidism), and intrahepatic cholestasis of pregnancy (ICP).

We collected adverse maternal outcomes, including any of the following: postpartum hemorrhage (PPH), severe PPH, uterine rupture, disseminated intravascular coagulation (DIC), puerperal infection, hysterectomy, bladder injury, and blood transfusion. We defined PPH as a loss of ≥1000 ml of blood after cesarean delivery and severe PPH as the loss of ≥1500 ml of blood after cesarean delivery.

Adverse neonatal outcomes included prematurity (< 37 weeks), fetal growth restriction (FGR), mild asphyxia (1-min Apgar score < 8), severe asphyxia (1-min Apgar score < 4), neonatal complication, and admission to a neonatal intensive care unit (NICU). FGR is an estimated fetal weight that is less than the 10th percentile for gestational age. Neonatal complication included neonatal asphyxia, neonatal malformation and other complications (hemolysis, jaundice, congenital heart disease, meconium aspiration pneumonia, et al.).

Information regarding maternal and neonatal diseases was classified according to the WHO’s Classification of diseases (ICD)-10.

We performed statistical analyses using SPSS v24.0 for Windows and R software (version 3.6.1). Missing values were imputed using a random forest algorithm. Missing data was listed in Supplementary Table S1. We examined whether quantitative data were normally distributed using the Kolmogorov-Smirnov test. Non-parametric continuous features are presented as medians and their corresponding interquartile ranges (IQR), and the Kruskal-Wallis test was applied for comparisons among multiple groups. Categorical variables are reported as frequencies (percentages), and the differences between groups were compared using the χ2 or Fisher exact-probability test in the case of small numbers, where appropriate. The associations between maternal age at first CD and each outcome were investigated by logistic regression analysis in 2 models: model 1 was adjusted for possible confounders—including gravidity, parity, BMI, assisted reproductive technology, and interval months; and with model 2 we explored possible explanations of the findings by adding mediating factors—i.e., maternal age at the second CD [19]. There is no official international definition of “advanced maternal age,” nor is there an “age interval” or reference group of maternal age [20]. According to most studies and guidelines, adolescent mothers have been excluded as a unique group. The rationale for our choice of a reference group (25–29 years) for this study was that the expected outcomes would be optimal for this age range, which also included the largest number of pregnancies. We also calculated crude odds ratios (ORs) and adjusted odds ratios (aORs), along with their 95% confidence intervals (CIs). Differences with P-values of < 0.05 were considered to be statistically significant.

This study possesses several strengths. First, our study was based upon the multicenter database that encompassed 11 public tertiary hospitals; and it covered 7 provinces, municipalities, and autonomous regions within China in 2017, effectively avoiding the selection bias of single-center and small-sample studies. To some extent, these data can therefore be generalizable to more heterogeneous populations. Second, this was the first study of its kind on the effect of maternal age at first CD on the complications and adverse outcomes of the second CD. However, this study also has several limitations. This was a historical cohort study, and data from different centers were not always ideal and complete; some missing values were therefore imputed using the random forest algorithm. The number of women at their first CD with age ≥ 35 years was relatively small. Further repeated studies are needed.