The role of maternal age on adverse pregnancy outcomes among primiparous women with singleton birth: a retrospective cohort study in urban areas of China

Abstract Background Both young and advanced maternal age pregnancies have strong associations with adverse pregnancy outcomes; however, there is limited understanding of how these associations present in an urban environment in China. This study aimed to analyze the associations between maternal age and pregnancy outcomes among Chinese urban women. Methods We performed a population-based study consisting of 60,209 singleton pregnancies of primiparous women whose newborns were delivered after 20 weeks’ gestation between January 2012 and December 2015 in urban areas of China. Participants were divided into six groups (19 or younger, 20–24, 25–29, 30–34, 35–39, 40 or older). Pregnancy outcomes include gestational diabetes mellitus (GDM), preeclampsia, placental abruption, placenta previa, premature rupture of membrane (PROM), postpartum hemorrhage, preterm birth, low birthweight, small for gestational age (SGA), large for gestational age (LGA), fetal distress, congenital microtia, and fetal death. Logistic regression models were used to assess the role of maternal age on the risk of adverse pregnancy outcomes with women aged 25–29 years as the reference group. Results The risks of GDM, preeclampsia, placenta previa, and postpartum hemorrhage were decreased for women at a young maternal age and increased for women with advanced maternal age. Both young and advanced maternal age increased the risk of preterm birth and low birthweight. Young maternal age was also associated with increased risk of SGA (aOR 1.64, 95% CI 1.46–1.83) and fetal death (aOR 2.08, 95% CI 1.35–3.20). Maternal age over 40 years elevated the odds of placental abruption (aOR 3.44, 95% CI 1.47–8.03), LGA (aOR 1.47, 95% CI 1.09–1.98), fetal death (aOR 2.67, 95% CI 1.16–6.14), and congenital microtia (aOR 13.92, 95% CI 3.91–49.57). There were positive linear associations between maternal age and GDM, preeclampsia, placental abruption, placenta previa, PROM, postpartum hemorrhage, preterm birth, LGA and fetal distress (all P for linear trend < .05), and a negative linear association between maternal age and SGA (P for linear trend < .001). The analysis of the associations between maternal age and adverse fetal outcomes showed U-shape for preterm birth, low birth weight, SGA, fetal death and congenital microtia (all P for quadratic trend < .001). Conclusions Advanced maternal age predisposes women to adverse obstetric outcomes. Young maternal age manifests a bidirectional effect on adverse pregnancy outcomes. The findings may contribute to improving women’s antenatal care and management.


Introduction
With the change in fertility policy and socio-economic development, maternal age at childbearing has dramatically changed worldwide in the past decades.In 1979, China implemented a one-child policy in order to control the increasing population growth rate [1].As a result, this policy has successfully reduced the fertility rate; however, it has also brought about delayed childbearing, sex-ratio imbalance, population aging, and workforce reduction [2][3][4].The policy was relaxed in 2013, allowing married couples to have two children if either parent was a singleton until the twochild policy was launched in 2015 [5], where this transformation inevitably led to an increasing proportion of advanced maternal age (AMA, 35 years).In China, the mean age at delivery increased from 27.1 years in 2010 to 29.7 years in 2017 [6].The proportion of pregnant women with AMA rose by 68.8%, from 12.5% in the one-child policy period (2011-2013) to 21.1% in the two-child policy period (2017-2019) [7].In the United States and Italy, the mean age at childbearing was constantly rising, the proportion of pregnant individuals with AMA was consistently increasing from 2007 to 2013, and the percentage of pregnancies in patients younger than 17 years of age decreased over the same period [8,9].
Childbirth at an AMA is associated with an increased risk of adverse pregnancy outcomes.A community-based prospective cohort study in China showed that a higher risk of gestational diabetes mellitus (GDM), preeclampsia/eclampsia, premature rupture of membrane (PROM), and preterm birth was found in women with AMA [10].A nationwide prospective cohort study consisting of 548,080 singleton pregnancies in South Korea demonstrated that primiparous women at an AMA had a higher risk of placenta previa, placental abruption, and low birth weight [11].The systematic review and meta-analysis published by Lean et al. in 2017 deduced that AMA had an association with stillbirth, neonatal death and several pregnancy complications, and placental dysfunction may mediate these adverse outcomes [12].On the other hand, young maternal age (YMA, <20 years) also has strong associations with adverse pregnancy outcomes.YMA may increase the risk of preterm birth and fetal or neonatal deaths owing to biological immaturity [9,13], and decrease the odds of GDM, placenta previa, and postpartum hemorrhage as well [14].
Owing to the advancing urbanization in China, the number of urban populations has increased greatly.An observational study from Zhejiang province observed that the percentage of pregnant women who lived in urban areas increased from 50.98% in 2013 to 54.40% in 2017 [15].As an improvement of women's education, the problem of delayed childbearing has become more serious in China, especially in urban areas, leading to a higher prevalence of adverse pregnancy outcomes in urban women [16].Meanwhile, environmental pollution and lifestyle changes (e.g.overweight or obesity) caused by urbanization also affect maternal and fetal outcomes [17,18].Given that the information on both advanced and young maternal ages affecting pregnancy outcomes is scarce in urban areas of China, a population-based retrospective cohort study was necessary to evaluate the associations between extremes of maternal ages and subsequent adverse pregnancy outcomes.

Study design and settings
We implemented a retrospective cohort study on 60,209 singleton pregnancies of nulliparous women, where their delivery and household registration (defined as the place where the resident household registration book was registered in the local police station at birth) both came from urban areas of Nanning.Clinical characteristics assessed were: obesity (prepregnancy body mass index 30 kg/m 2 ), gravidity, examination at first trimester (defined as the first antenatal care before 13 gestational weeks), number of prenatal visits, prior abortion, and assisted reproductive technology.Pregnancy outcomes were categorized into maternal outcomes and fetal outcomes.In this study, maternal outcomes included GDM, preeclampsia, placental abruption, placenta previa, PROM, and postpartum hemorrhage; fetal outcomes included preterm birth, low birth weight, small for gestational age (SGA), large for gestational age (LGA), fetal distress, congenital microtia, and fetal death.

Study area
Nanning, the capital of Guangxi province, is located in the southern region of China, close to Southeast Asia, and is one of the commercial and communication centers in these regions.It has 22,099 square kilometers, and is divided into six urban districts (central urban areas: Qingxiu, Xingning, Xixiangtang, Jiangnan, Liangqing, Yongning) and six counties (rural and suburban areas: Wuming, Hengxian, Binyang, Long'an, Mashan, Shanglin), with a population density of 334.96 per square kilometer.Among a total of 3,527,172 of female population, 774,567 (21.96%) aged less than 20 years, 936,111 (26.54%) aged 20-34 years, 1,260,964 (35.75%) aged 35-59 years, and 555,530 (15.75%) aged 60 or older.

Participants
Pregnant women aged 13-54 years were included in this study from 27 hospitals between January 2012 and December 2015.Pregnant women with induced abortion, therapeutic induction of labor, delivery <20 weeks' gestation, birth weight <300 g, pregestational diabetes mellitus (pre-pregnancy serum fasting plasma glucose 7.0 mmol/L), pregestational hypertension (pre-pregnancy blood pressure 140/90 mmHg), and missing clinical characteristics data were excluded from this analysis.Detailed information on the study participant screening is shown in Figure 1.Ethical approval in this study was obtained from the Ethics Committee of Nanning Maternal and Child Health Hospital.Informed written consent was waived due to the retrospective and observational design.

Data collection
The data on clinical characteristics, obstetric complications and infant outcomes was collected from the Guangxi Woman and Child Health Information Management System.Information from antenatal care, delivery, newborns, and child death for all births was abstracted from the medical records into this data system, a provincial database administered by the Guangxi Health Commission.The data entry and quality control methods were carried out per prior regulations [19].

Variables and definitions
We divided all participants into six groups according to maternal age at delivery: 19 or younger, 20-24, 25-29, 30-34, 35-39, and 40 or older.The age group of 25-29 years was regarded as the reference group.AMA was defined as women aged 35 years or older; meanwhile, YMA was considered as individuals younger than 20.
GDM was defined as fasting plasma glucose 5.1 mmol/l or the 75 g oral glucose tolerance test value 10.0 mmol/l at 60 min or 8.5 mmol/l at 120 min when conducted at 24-28 gestational weeks [20].Preeclampsia was defined as a blood pressure greater than 140/90 mmHg with proteinuria after 20 gestational weeks [21].We described placental abruption as a condition where the placenta was partially or entirely dissected from the uterine wall after 20 weeks of pregnancy [22], and placenta previa as placental implantation near or over the opening of the cervix [23].PROM referred to a spontaneous rupture of fetal membranes before the onset of labor [24].Postpartum hemorrhage was defined as a maternal blood loss exceeding 500 ml within 24 h after birth [25].
Preterm birth was defined as delivery before 37 weeks of gestation.Low birthweight referred to a birth weight less than 2,500 g.Small for gestational age (SGA) was described as a birth weight less than the 10 th percentile for its gestational age, and large for gestational age (LGA) greater than the 90 th percentile [19].Fetal death was defined as stillbirth or spontaneous abortion after 20 weeks of gestation.Fetal distress was caused by insufficient uterine oxygen supply [26].Auricle malformation or external auditory canal atresia was diagnosed as congenital microtia.

Statistical analysis
The descriptive data were presented as frequencies, percentage, mean, and standard deviations.Differences in clinical characteristics and adverse pregnancy outcomes between maternal age groups were investigated using the chi-square (v 2 ) test or the Fisher-Freeman-Halton exact test.Associations between maternal age and pregnancy outcomes were explored using logistic regression models after adjustment for obesity (yes/no), gravidity (1-2, 3 or more), examination at first trimester (yes/no), number of prenatal visits (0-4, 5 or more), and assisted reproductive technology (yes/no).The odds ratio (OR) with 95% confidence interval (95% CI) was used to express the association between exposure and outcome.We analyzed the significance of the linear or quadratic trend by treating maternal age as an ordinal variable in the regression model.The linear trend indicates that the outcome variable constantly increases or decreases with advancing age.In contrast, the quadratic trend reflects that the value of the trend changes with age.All statistical tests were 2-sided and p < .05 was considered statistically significant.All statistical analyses were performed using SPSS version 17.0 (SPSS, Chicago, IL, USA).
The prevalence of adverse maternal outcomes among the six groups is shown in Table 2.The six groups differed with regards to all of the selected maternal outcomes (all p < .01).Compared with the reference group, pregnant women with AMA were more likely to show a higher prevalence of GDM, preeclampsia, placenta previa, PROM, and postpartum hemorrhage, whereas individuals with YMA had lower incidences of these outcomes.In addition, the prevalence of placental abruption in women aged 40 or older was higher than those of 25-29-year-old women.

Discussion
The findings of this study suggest that YMA and AMA were respectively associated with adverse pregnancy outcomes after adjustment for potential confounders.Both younger and older pregnant women were at increased risk of preterm birth, low birth weight, and fetal death.However, the risk patterns of some outcomes may differ for younger and older women.The risks of GDM, preeclampsia, placenta previa, PROM, postpartum hemorrhage, and LGA were decreased for younger women and increased for older women.Younger women had a higher risk of SGA while older women had negligible difference in risk for this outcome compared to 25-29 years old.Risks for placental abruption and congenital microtia were elevated for older women.In addition, linear or quadratic trends were observed between maternal age and several adverse pregnancy outcomes.Maternal age is positively correlated with the risk of adverse maternal outcomes.Owing to advancing urbanization and improving education, an increasing number of Chinese women migrate from rural to urban areas and spend more time on education, consequently delaying their childbearing ages.The association between maternal age and adverse obstetric outcomes has drawn people's common attention in China.The incidence of GDM generally increases with maternal age.Several population-based studies showed that when compared to controls, pregnant women at an AMA were more likely to have a higher risk of GDM [10,11], while those with YMA were less likely to develop it [6,27], which was in accordance with our study.This increasing trend may be attributed to the progressive decline of pancreatic b-cell function with aging, leading to a reduced insulin affinity [28].The risk of preeclampsia also increases continuously with increasing maternal age.Consistent with the study of Cavazos-Rehg et al. [29], we noted that the risk of preeclampsia decreased in women with YMA and increased in those with AMA compared to individuals aged 25-29 years.The opposite phenomenon between younger and older women may be explained by biological aging.Proteinuria excretion elevated with increasing age-related glomerular sclerosis, along with hypertension induced by insulin resistance, eventually contributed to preeclampsia.We also observed that the prevalence of PROM and postpartum hemorrhage elevated as the maternal age increased, which was in line with those of studies conducted in Iran [30] and other areas of China [10,14].In addition, the risk of placental abruption and placenta previa continually increased with maternal age [11,4], which may be linked to uteroplacental ischemia [31], infertility treatment and prior abortion.
Both young and advanced maternal ages are associated with adverse infant outcomes.We found that pregnant women with YMA had a lower risk for fetal distress compared to 25-29-year-old women.However, a study of 7,810,762 sample data from the United States Nationwide Inpatient Sample (NIS) showed the opposite result of our finding [29], which needs to further explore the relationship through a prospective study.YMA and AMA were both associated with an elevated risk of preterm birth.As Schummers et al. and Esposito et al. reported that risk of preterm birth was increased at young and advanced maternal ages for Asian/Pacific Islander and white women [13,32].Consistent with the linear trend we found, the population-based data from the Taiwan Birth Notification System showed an L-shaped association between maternal age and the odds of SGA [33].A multicenter retrospective study conducted in mainland China indicated that the prevalence of SGA seemed to be decreased with increasing age [34].However, the relationship between maternal age and the risk of SGA is still controversial.Several retrospective studies conducted in the United States [13,29], Sweden and Norway [35] found that AMA was also associated with a significantly increased risk of SGA, which was different from the studies performed in China [33,34].This discrepancy may be explained by the difference in race between whites and yellows [13].Interestingly, a quadratic trend was also found between maternal age and SGA in this study, although AMA did not significantly elevate the risk of SGA.We observed a linear trend that the incidence of LGA was upward with increasing maternal age, which was consistent with the study of Dongarwar et al. [36].This study also found that the quadratic trends were U-shaped for low birthweight and fetal death.The findings were in line with those of studies conducted in the United States [13,37], mainland China [14,38,39], and Taiwan [33].The birth defects monitoring data from 2011 to 2020 in Shandong province showed that the incidence of congenital microtia was higher among older women [40], which was in accordance with our study.Placental immaturity or placental dysfunction may contribute to the occurrence of adverse infant outcomes among younger or older women, respectively.
The strengths of our study are the quality of data sets and the risk assessment of both young and advanced maternal ages affecting pregnancy outcomes.However, there are several limitations to note in this study.First, as a multi-institution-database-based study, it is difficult to standardize the measuring instruments used by different hospitals and may thus lead to detection bias; however, due to the comparable proportion of these groups in each hospital, the impact of this bias may be minor.Second, some potential confounders, such as maternal educational level and economic status, are necessary to consider when evaluating the risks for maternal and infant outcomes but are unavailable in the dataset; however, a previous study demonstrated that adjustment for maternal education and income only had a slight impact on effect estimates of neonatal mortality [38].Third, the data of our study represent a sample of urban areas in China, and these findings may not be generalizable to rural populations or other regional populations.
Unlike other studies in China [14,33,38,39], this study comprehensively explored the linear or quadratic trends between maternal age and the risk of adverse pregnancy outcomes in urban women.In summary, we found positive linear associations between maternal age and GDM, preeclampsia, placental abruption, placenta previa, PROM, postpartum hemorrhage, preterm birth, LGA and fetal distress, and a negative linear association between maternal age and SGA.In addition, there were U-shaped associations between maternal age and preterm birth, low birth weight, SGA, fetal death and congenital microtia, and a Ո-shaped association between maternal age and fetal distress.Given the health hazards of younger or older pregnancies, extremely young pregnancies should be avoided, and perinatal care for older women should be strengthened to prevent adverse pregnancy outcomes.

Figure 1 .
Figure 1.Flowchart of the study population.

Figure 3 .
Figure 3. Risk of adverse fetal outcomes, by maternal age group.ORs (95% CIs) were adjusted for obesity, gravidity, examination at first trimester, number of prenatal visits, and assisted reproductive technology.SGA: small for gestational age, LGA: large for gestational age, CI: confidence interval, OR: odds ratio, Ref: reference group (25-29 years).
This study was supported by grants from the Key Research and Development Program of Nanning Municipal Science and Technology Department, China (No. 20223053).The funders had no involvement in the study design, information collection, data analysis and interpretation, and preparation of the manuscript. /orcid.org/0000-0001-6186-6455

Table 1 .
Characteristics of the pregnant women, by maternal age group (%).
SGA: small for gestational age; LGA: large for gestational age.

Table 2 .
Prevalence of adverse maternal outcomes, by maternal age group (%).
THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE