Pregnancy complications among nulliparous and multiparous women with advanced maternal age: a community-based prospective cohort study in China

Subjects for this study were recruited from multiple maternal and child healthcare centers in Liuyang city, Hunan Province, China from June 2013 to December 2015 [18]. Liuyang is a county-level city located in the central-south region of China. It has four sub-districts and 33 towns under its jurisdiction. In this study, we randomly selected two sub-districts and 11 towns in Liuyang city to represent the population from both urban and rural areas, and the recruitment was conducted in the maternal and child healthcare centers located in each selected sub-district/ town.

Pregnant women were recruited by their gynecologists during their regular prenatal care visits with the following inclusion criteria: a) received the first prenatal care at the recruited healthcare center; b) had no more than 12 weeks of gestation. Subjects were excluded if they a) had pregnancy complications before recruitment in the first trimester of the index pregnancy because it’s unclear whether the disease was started from this pregnancy or before this pregnancy; b) had an induced abortion during follow-up or did not have complete medical records from enrollment through delivery; c) had language communication barriers or were deaf; or d) refused to actively cooperate for this research.

Informed written consent was obtained from all participants. Ethical approval was obtained from the Ethics Committee of Xiangya School of Public Health, Central South University, and all research was performed in accordance with relevant guidelines/regulations.

Baseline information was collected for all subjects by trained research gynecologists through a face-to-face interview. The questionnaire included socio-demographic characteristics (i.e. age, education, occupation, family income, residential area), pre-pregnancy weight and height, and history of gravidity, parity, and abortion. Routine obstetric examinations and laboratory examinations were also conducted afterward. Subjects were followed up at 16 weeks, 24 weeks, 28 weeks, 32 weeks, 38 weeks of pregnancy (or before delivery), and the end of delivery, and routine obstetric examinations were conducted at each follow-up. During the follow-up period, medical records including the records of routine obstetric care, the diagnosis of pregnancy complications, and the time of diagnosis were collected. A supervisory team was assigned to check whether the follow-up was on time, verify the registered complications, and check and rearrange the missing diagnosis.

Pre-pregnancy body mass index (BMI) was calculated using the formula weight (kg)/height (m)². BMI was categorized into three categories using Asian-specific cut-offs [19]: < 18.5 kg/m², 18.5–23 kg/m², 23–27.5 kg/m², and ≥ 27.5 kg/m². Gravidity and parity were categorized as none, once or more pregnancies/births. History of miscarriages was classified into three categories: none, once, twice and more times, and history of induced abortion and history of preterm labor was categorized as none and once or more times.

Gestational hypertension was defined as having high blood pressure after 20 weeks according to the International Society for the Study of Hypertension in Pregnancy guidelines [20]. Chronic hypertension before 20 weeks of gestation was not an outcome of interest in this study. Gestational diabetes mellitus (GDM) was diagnosed by oral glucose tolerance test (OGTT) conducted at 24–28th week of gestation, with fasting plasma glucose > 5.6 mmol/l or with post-75 g glucose load glucose level of > 8.8 mmol/l in 60th min or > 7.8 mmol/l in 120th min [21]. Anemia was defined as hemoglobin < 12.0 g/dL [22]. Polyhydramnios and oligohydramnios were examined using ultrasound, and was defined as the 4-quadrant amniotic fluid index (AFI) > 24 cm or ≤ 5 cm, respectively. Threatened abortion was defined as a small amount of vaginal bleeding, followed by paroxysmal lower abdominal pain or low back pain before the 28th week of pregnancy. The pelvic examination was not open, the membrane was intact, no pregnancy was discharged, and the size of the uterus was consistent with the gestational age. Preterm labor was defined as birth occurs before the start of the 37th week of pregnancy. Low birth weight is defined as the birth weight of an infant of 2499 g or less.

The demographic and pre-pregnancy characteristics of study participants were compared among four maternal age groups. Significant levels were compared using Chi-square tests, and all the tests were two-tailed. Incidence of pregnancy complications and mode of delivery were compared by parity and age groups, and we used subjects age 20 to 29 years as the reference group considering the low incidence of pregnancy complications in lower age groups. Logistic regression was used to calculate the odds ratio (OR) and 95% confidence interval (CI) for pregnancy complications. Univariate analyses were conducted on all socio-demographic and pre-pregnancy characteristics to select the variable for adjustment, and significant variables were included in the multivariable model. Cramer’s V statistics were used to evaluate the correlation between variables included in the final model. We were not able to conduct logistic regression for placenta previa, infectious disease, and threatened abortion among nulliparas due to low incidence rate. In the multivariate analysis, income level and BMI were adjusted for gestational hypertension, preeclampsia/eclampsia, and threatened abortion; occupation was adjusted for GDM and premature rupture of membrane; income level was adjusted for anemia, polyhydramnios, oligohydramnios, placenta previa, and infectious disease; occupation and history of preterm labor was adjusted for preterm labor; education and BMI were adjusted for the mode of delivery and low birth weight. All analyses were conducted using R version 3.5.2.

A total of 12,170 pregnant women were admitted to the hospital from June 2013 to December 2015. Among them, 10,475 pregnant women were recruited in the study (response rate 86.1%), and 10,171 subjects with complete information were included in the final analysis (97.1%) (Fig. 1).

Among the 10,171 subjects, the average age was 26.7 ± 4.2 years. 3274 (32.2%) subjects were 20–24 years old, 4587 (45.1%) subjects were 25–29 years old, 1637 (16.1%) subjects were 30–34 years old, and 572 (5.6%) subjects were 35 years old and older (Table 1). The majority of subjects (85.9%) lived in rural areas, and the median family annual income was 20,000 Chinese Yuan per capita (around 2880 USD). The majority of subjects did not complete high school education (83.7%), and most of them were housewives (54.3%), factory workers (23.5%), and farmers (13.9%). The pre-pregnancy BMI was normal for 61.4% of subjects, 18.0% of subjects were underweight, and 20.5% of subjects were overweight/obese. Overall, around half of the subjects had one or more births previously, and the proportions were much higher for women aged 30–35 years (79.0%) and women aged 35 years and above (88.6%) than younger age groups. Also, around one-third of women aged 35 years and above had two or more times of miscarriages, while the proportion ranged from 6.3 to 19.5% in younger age groups. 3.1% of subjects had a history of induced abortion, and the proportion ranged from 1.7 to 5.9% in different age groups.

The comparison of incidence rates for different pregnancy complications stratified by parity and age groups is shown in Table 2. Overall, the incidence rate of each complication during pregnancy ranged from 0.9 to 4.2%, and the top three pregnancy complications observed included gestational hypertension (4.2%), preeclampsia or eclampsia (3.5%), and anemia (3.7%). The incidence of pregnancy complications showed an increasing pattern when stratified by age groups for gestational hypertension, preeclampsia, polyhydramnios, premature rupture of membrane among nulliparas, and this pattern was observed for GDM, anemia, polyhydramnios, oligohydramnios, placenta previa, premature rupture of membrane among multiparas. The incidence of preterm labor was 9.4%, which ranged from 9.3 to 13.8% and 8.1 to 14.5% among different age groups in nulliparas and multiparas, respectively. The incidence of low birth weight was 2.8%, and the incidence was higher in multiparas (5.2%) than nulliparas (0.7%). In terms of mode of delivery, around half of subjects aged 30 years old and above performed a cesarean delivery, while this proportion was around 30% in the 20–29-year age group.

Adjusted ORs for pregnancy complications calculated using logistic regression models are shown in Table 3. The risks for gestational hypertension (OR 8.44, 95%CI 4.18–15.58) and preeclampsia/eclampsia (OR 9.92, 95%CI 4.87–18.78) were significantly higher for women with AMA above among nulliparas compared with women aged 20–29 years, while this association was not observed among multiparas. Among multiparas, women with AMA were found to have increased risks for GDM (OR 3.29, 95%CI 1.76–5.94), anemia (OR 1.85, 95%CI 1.25–2.69), polyhydramnios (OR 3.29, 95%CI 1.56–6.64), premature rupture of membrane (OR 5.14, 95%CI 2.12–12.29), preterm labor (OR 1.89, 95%CI 1.42–2.50), while similar associations were only observed for premature rupture of membrane (OR 6.84, 95%CI 2.00–17.69) among nulliparas. Odds of performing cesarean delivery over virginal birth were significantly higher for women in older age groups as compared to the reference group both among nulliparas and multiparas.