To the Editor: Gestational diabetes mellitus (GDM) affected approximately 12.8% of pregnancies worldwide in 2019 according to the estimation of the International Diabetes Federation [1]. Growing evidence suggests that women with a history of GDM have a markedly increased risk of diabetes later in life [2], but it is still unknown whether intensive management of GDM has benefits for prevention of maternal diabetes at postpartum and later in life. Our group conducted a population-based RCT testing the effectiveness of intensive lifestyle intervention of GDM on the risk of macrosomia and large-for-gestational-age (LGA) in the three-tier antenatal care network of Tianjin, China. We found that such an intervention reduced the risk of macrosomia and LGA in offspring of women with GDM [3]. In the current analysis, we investigate the effect of lifestyle intervention on maternal postpartum diabetes among women with prior GDM.

Details of the study setting, design and participants are reported elsewhere [3]. Briefly, 19,847 pregnant women were enrolled and offered a 50 g 1 h glucose challenge test (GCT) between the 24th and 28th week of gestation, at primary hospitals. Women who had GCT ≥7.8 mmol/l (n = 2921) were referred to the Tianjin Women and Children’s Health Center (Tianjin, China) to undergo a 75 g 2 h OGTT. Based on the International Association of Diabetes and Pregnancy Study Group’s criteria, GDM was diagnosed as: fasting plasma glucose (PG) ≥5.1 mmol/l; 1 h PG ≥10.0 mmol/l; or 2 h PG ≥8.5 mmol/l. Of the 2921 participants, 1440 pregnant women were diagnosed with GDM. In the previous analysis [3], 734 women were sequentially excluded due to not being invited to participate in the late stage of the fieldwork (n = 52), meeting one or more of exclusion criteria (n = 440) or receiving unintentional intervention (n = 242). Ultimately, 706 women received either intensive care (n = 344) or usual care (n = 362).

The intensive care intervention included three individualised advice sessions on diet and physical activity and three group education sessions. The details of the intensive care intervention have been published previously [3]. Briefly, total energy intake (i.e. low energy intake) was recommended to the women in the intensive care group based on pre-pregnancy BMI classification for Chinese adults. All women were asked to engage in at least 30 min of light- to moderate-intensity physical activity daily. The intensive care group was also asked to perform self-monitoring of blood glucose (SMBG) with blood glucose control targets of ≥3.5–≤5.1 mmol/l for fasting capillary blood glucose, and ≤7.0 mmol/l for 2 h postprandial capillary blood glucose up to the 36th gestational week or ≤8.0 mmol/l from the 36th gestational week onwards [3]. In contrast, the usual care group was only offered routine prenatal care (one group diabetes education session, without asking them to perform SMBG).

At postpartum, all women enrolled in the RCT were invited to participate in a standard 75 g 2 h OGTT. Diabetes, impaired glucose tolerance (IGT) and impaired fasting glucose (IFG) were defined based on the criteria of World Health Organization (1999) [4]. Ethical approval was obtained from the Ethic Committee of Tianjin Women and Children’s Health Care Center and written informed consent was obtained before data collection. The primary trial is registered at ClinicalTrials.gov (ClinicalTrials.gov registration no. NCT01565564).

All analyses were performed using the Statistical Analysis System (release 9.4; SAS Institute, Cary, NC, USA). Continuous variables were compared using Student’s t test or Wilcoxon two-sample test and categorical variables were compared using χ2 test or Fisher’s exact test where appropriate. Analysis of covariance was used to adjust for age, family history of diabetes, and log10-transformed fasting and 2 h PG during OGTT during pregnancy. Generalised logit model was used to obtain unadjusted and adjusted ORs of intensive care vs usual care for diabetes and IGT and/or IFG. The adjustment included age, family history of diabetes, and fasting and 2 h PG during OGTT during pregnancy.

At a median 9.4 (IQR 7.7–11.7) weeks after delivery, 391 (55.4%) women completed the postpartum OGTT. The baseline clinical and biochemical characteristics, including maternal age, pre-pregnancy BMI, systolic/diastolic blood pressure, ethnicity, parity, family history of diabetes in first degree relatives, gestational age at OGTT, and fasting and 2 h PG during OGTT during pregnancy were similar between the intensive care group and the usual care group (Table 1). All these baseline variables were also similar by intervention among women who were lost to follow-up (data not shown). At follow-up, fasting and 2 h PG during OGTT were similar in the intensive care and the usual care groups. In the intensive care group (n = 218), 3.2% of women developed postpartum diabetes and 34.9% developed postpartum IGT and/or IFG. Similarly, 2.9% of women in the usual care group (n = 173) developed postpartum diabetes and 35.3% developed postpartum IGT and/or IFG (p > 0.9). Because there were significant or borderline differences in some baseline variables between women available and those women lost to follow-up either in the intensive care group and/or in the usual care group (data not shown), we performed multivariable analysis to adjust for these baseline variables. The adjusted postpartum fasting and 2 h PG values were still similar after the adjustment; the adjusted ORs were 0.94 (95% CI 0.28, 3.11) for diabetes and 0.91 (95% CI 0.58, 1.43) for IGT and/or IFG.

Table 1 Clinical and biochemical characteristics of participants
Full size table

In this secondary analysis of follow-up data of the RCT, we found that 3.1% of women with GDM developed diabetes at early postpartum, but lifestyle intervention during pregnancy did not reduce the risk of postpartum diabetes. Consistent with our findings, another RCT (n = 457) also reported that treatment with formal nutritional counselling and diet therapy during pregnancy had no significant impact on subsequent diabetes among women with mild GDM [5]. Women with GDM are characterised by a lower pancreatic beta cell function and higher insulin resistance years after delivery [6]. Although intensive lifestyle intervention has been found, to some extent, to increase insulin sensitivity and improve beta cell function [7] and reduce the risk of delivery of a macrosomia and/or LGA infant [3], intensive care may not be enough to safeguard beta cell function from declining during pregnancy or at early postpartum in women with GDM. It is also noted that our antenatal care system did not offer a universal OGTT at early pregnancy to pregnant women and some women with GDM may have had IGT and/or IFG before pregnancy. Although our study found that intervention of GDM did not reduce risk of early postpartum diabetes, its effects on long-term risk of diabetes in women with prior GDM still needs to be tested in future RCTs.

Our study had some limitations. The sample size was not large and the low follow-up rate may introduce potential lost-to-follow-up bias because some baseline variables were different between women available and not available for follow-up. However, the adjustment analysis may have partially removed potential confounding effects.

In conclusion, we found that intensive lifestyle intervention during pregnancy did not reduce the risk of postpartum diabetes among women with GDM. Its effects on long-term risk of diabetes needs to be tested in future studies.

FormalPara Trial registration

ClinicalTrials.gov NCT01565564.