Main results
Our research indicated the specific relationships between maternal age change and the change of BW, LBW and macrosomia. BW increased gradually until maternal age 34, then decreased. The risk of LBW decreased gradually until age 36, then increased. The risk of macrosomia increased with the increase of maternal age. The results of this study on BW showed that the relationship between the risk of abnormal BW and age is not based on the traditional 35-year-old boundary. Our findings provided the absolute risks of abnormal BW by maternal age, which would be useful for patient fertility counseling.
Interpretation
A nonlinear relationship between maternal age and BW was observed and two turning points of maternal age at 24 and 34 years old were found in this study. The BW increased faster from 20 to 23 years old than from 24 to 34 years old. The curve was consistent with the findings of the previous studies [12, 13], however, there was no research provided that the threshold maternal age on BW was 34 years old [12, 21, 22]. We observed a marginal significantly negative association between maternal age and BW when maternal age ranged from 35 to 40 years, which was consistent with the findings of the previous studies [23, 24]. In previous studies, the population was grouped according to maternal age into several subgroups, then to investigate the effect of maternal age on the BW. Most studies considered women in age group 20–29 years as reference group. Differently, in our study, the maternal age was analyzed as a continuous variable rather than categorical variable, to find a more accurate age threshold.
The mechanism of the effect of maternal age on BW is still unclear. In the study, we found the risks of LBW and macrosomia after 35 years present the same pattern, increasing at these maternal age period. Relevant researches showed that most of the effects on the offspring of intrauterine exposure to maternal age-related obstetric complications might be induced by epigenetic DNA reprogramming during critical periods of the embryo or fetal development [25]. And it is well known that mitochondria are maternally derived. We also know that mitochondrial DNA is not capable of DNA repair and is therefore at greater risk of acquiring mutations with age
[12]. Similarly, the quality of woman's eggs declined dramatically with increasing age, leading to an increased risk of pregnancy-related complications among older women [12]. Furthermore, older mothers are more likely to suffer from chronic diseases and their complications in pregnancy, including obesity, anemia and diabetes. In addition, ageing process affects the reproductive system similar to other systems in the human body. These factors may contribute to that the risks of LBW and macrosomia after 35 years present the same pattern, increasing at these maternal age period.
A nonlinear relationship between maternal age and the risk of term LBW was found in our study. The curve was the same as the previous study [13, 26]. As LBW newborns may be premature with other risk factors [27], we restricted our study population to term newborns, therefore the etiology of LBW was intrauterine growth restriction [18]. The prevalence of term LBW in our study was 1.5%, which was lower than 2% reported by the other study [28], which suggesting there was a good perinatal care system in Xi’an, China. Although the prevalence of term LBW is low, it is not negligible and term LBW can lead to adverse pregnancy outcomes, as severe neonatal asphyxia [29]. The threshold maternal age on the risk of term LBW was 36 years old, which means that the risk of term LBW will significantly increase when the maternal age is over 36 years old, but fewer studies have reported it [5, 6, 14]. However, the biological mechanisms by which maternal age cause the term LBW are unclear. The increased risk of LBW among younger can be explained by the low socioeconomic conditions and increased nutritional demands of pregnancy [30].
With the increase of maternal age, the risk of macrosomia was increasing and two turning points of maternal age were found at 24 and 33 years. The curve was identical to the previous studies [31]. The incidence of macrosomia was 6.0% in our study, which was approximate with the previous study [14]. The risk of macrosomia increased faster from 20 to 23 years old than from 24 to 33 years old, which was the same as the change of BW. Term macrosomia is influenced by maternal hyperglycemia and endocrine status through placental circulation [32]. The increased risk of term macrosomia is partly explained by the increased prevalence of diabetes with the increased maternal age [13].
Strengths and limitations
Our research has some advantages: first, our study was a large sample research based on four years records. Besides, our study had very strict inclusion and exclusion criteria and the data was also carried on the strict cleaning and so on, which increases the credibility of the results. However, there were some potential limitations in this study. First, some potential confounders, such as fetus gender, parity, medical history, economic condition and paternal age, were not adjusted because of limited data. Although the cutoff for “paternal advanced age” is not clearly defined, there is an increase in genetic risk as men age. Previous studies showed that economic condition and fetus gender were associated with higher BW, but adjustment for socioeconomic factors and fetus gender made little difference to the results [22]. In addition, at any maternal age, higher parity was associated with higher BW [22].
Therefore, the extrapolation of the results requires caution, especially for others ethnicity [21]. However, the results of this study still have a certain degree of reference significance for other regions. Second, the mother-child pairs were not included when the maternal age was younger than 20 years old or older than 40 years old, because of the small proportion of them and limited data. Women delivering before 20 or after age 40 years had a higher incidence of pregnancy complications. Therefore, our research could only reflect the changing trend of BW, LBW and macrosomia when the maternal age ranged from 20 to 40 years old. In order to ensure the accuracy of our research results, we limited the subjects to full term singleton live birth, which could eliminate some potential influencing factors [27], such as serious pregnancy complications. Although the large sample size might increase potential confounding, we could estimate the relationships between maternal age and BW, LBW and macrosomia in detail with generalized additive model and two-piece wise linear regression model.