The relationship between vitamin D and insulin resistance before delivery in advanced maternal age

In November 2013, the one child policy in China was changed to two child policy that only couple from single-child family can have a second child [1]. In October 2015, the two-child policy was further transformed into universal two child policy and all couples were permitted to have two children [2]. After the implementation of universal two child policy, many older couples already had a child begin to plan for a second child [3]. The Chinese government estimated that 60% of women who benefited from the transformation into universal two child policy are older than 35 years, called as advanced maternal age (AMA) [2].

Several researches have demonstrated that AMA is a risk factor of adverse perinatal outcomes including gestational diabetes (GDM), pre-eclampsia, fetal anomaly and preterm delivery [4, 5]. Among all obstetric complications, a rising prevalence of GDM, characterized by insulin resistance (IR) accompanied with a failure of islet β cells to compensate for it, has been reported over years, reaching 10–15% in the world [6]. Furthermore, the research has demonstrated the pregnancy is a diabetogenic state since the steroid hormones increase and IR occur in peripheral tissues, as well as inflammatory cytokines secreted from adipose tissue and placenta, can contribute to IR and pathogens of GDM [7]. The physiologic condition may be exacerbated by well-established risk factor for GDM, including AMA and overweight and obesity before pregnancy [8].

For decades, vitamin D emerged as a controversial nutrients and pro-hormones. The classical action of vitamin D is the modulation of bone and mineral metabolism [9]. However, increasing evidences suggested several extra skeletal action of vitamin D referred to some chronic conditions including cardiovascular diseases [10], obesity [11], metabolic syndrome [12], some kinds of cancer [13, 14] and autoimmune disease [15]. However, there were few studies focused on AMA to explore the state of vitamin D levels before delivery and investigate the association between the level of vitamin D and IR. Thus, this study was aimed to detect the level of vitamin D before delivery and further demonstrate the relationship between vitamin D and IR in AMA.

This study has demonstrated that the levels of serum 25(OH) D levels in the late pregnancy were negatively associated with IR in AMA (r = − 0.25, P = 0.025, Fig. 1a). Furthermore, women with vitamin D deficiency have higher level of fasting blood insulin (14.70(8.76–34.65) and 10.89(7.15–16.12), respectively, P = 0.031, Table 2) and HOMA-IR (1.78(1.07–4.14) and 1.30(0.83–1.89), respectively, P = 0.024, Fig. 1b) compared to those with vitamin D non-deficiency. After adjusted confounder factors, women with vitamin D non-deficiency were negatively related to IR in comparison to those with vitamin D deficiency (β = − 1.289, P = 0.026, Table 3).

In the current study, women with vitamin D deficiency have higher fasting insulin concentration than those with vitamin D non-deficiency. However, there were no significant difference in fasting glucose concentration and lipid level between the two groups. Similarly, several studies have demonstrated that vitamin D was associated with metabolic syndrome including IR [18‐20] and obesity [21]. Lu L et al. suggested that there was significant inverse association of 25(OH) D with fasting insulin and HOMA-IR in overweight and obese but not in normal-weight subjects in China (P = 0.0363 and P = 0.0187, respectively) [22]. Moreover, Chinese individuals with vitamin D deficiency have higher fasting insulin and HOMA-IR compared to those with vitamin D non-deficiency [19]. Furthermore, there were researches identified that pregnant women with low vitamin D in early pregnancy had higher HOMA-IR indices at 28 weeks(r = − 0.32, P = 0.02) [23], but not associated with the risk of GDM [24]. Taken together, previous studies have observed the relationship between vitamin D and IR in adults and those association needed further examination were still conflict. Our study was the first to demonstrate that serum 25(OH) D levels in the late pregnancy was statistically associated with fasting blood insulin and HOMA-IR, but not with fasting blood glucose and lipid metabolism in AMA, a worthy more attention population in China.

At the same time, after adjusted some related risk factors of IR and GDM, serum 25(OH) D concentrations in vitamin D non-deficiency group were negatively correlated with the HOMA-IR levels compared to those in vitamin D deficiency group (β = − 1.289, P = 0.026, Table 3). As Xiao Y et al. demonstrated that serum 25(OH) D concentrations were significant inversely associated with metabolic covariates including fasting insulin and HOMA-IR, after adjusted for age, sex and BMI (β = − 0.39, P < 0.0001 and β = − 1.49, P < 0.0001, respectively) [20]. However, there was another study demonstrated that among males, 25(OH) D was associated with HOMA-IR (β = − 0.011, P = 0.004) after adjustment for BMI, but not women [25]. As for the sex-specific relationship, some potential reasons could be that middle-aged males have been shown to have a higher risk of incident metabolic compared with middle-aged women [26, 27]. Whereas, pregnancy was a physiologic condition with gradually increase of insulin concentration especially in the late pregnancy [7]. Therefore, if indeed low vitamin D levels were associated with IR in early pregnancy, there might be a stronger relationship between vitamin D and IR in late pregnancy.

IR is considered as a physiologic condition during the pregnancy and has also be implicated as main characteristic of GDM [28]. In the present study, the levels of IR determined by HOMA-IR were significant higher in vitamin D deficient subjects compared to those in vitamin D non-deficiency, which might be explained by vitamin D involved in glucose metabolism contributed to facilitate the secretion and action of insulin [29, 30].

The strengths of our study were the first to explore the relationship between vitamin D levels and IR before delivery in AMA. Several studies have demonstrated the relationship between vitamin D and diabetes [20, 29], even GDM [24] in pregnant women with common maternal age. Whereas, few studies focus on the association between vitamin D and IR in AMA.

Although those strengths, there are still some limitations of this study needed to be considered. Those pregnant women were recruited from a single hospital. Besides, multivariable analysis was unable to adjust for outdoor activity and lifestyle difference because those were not collected in our study.

In brief, the study was the first to highlight the association between vitamin D and IR before delivery in AMA. Pregnant women in AMA with vitamin D deficiency might have higher HOMA-IR levels than those with vitamin D non-deficiency.