Introduction

Gestational diabetes mellitus (GDM) – glucose intolerance, resulting in new onset in pregnancy–has emerged as a global public health concern1. GDM is associated with potentially harmful health effects for mother and fetus2. According to the World Health Organization (WHO), the prevalence of GDM is rising3, largely attributed to the worldwide epidemic of obesity, older maternal age at first pregnancy, and a change in the diagnostic criteria for GDM4. Currently, the International Association of Diabetes recommends screening for GDM among all pregnant women, at 24 to 28 weeks’ gestation, using a 75-g oral glucose tolerance test (OGTT)5. AS this approach lacks a sufficient sensitivity and specificity6, a method that may improve the detection of GDM, earlier pregnancy, is desirable.

Some studies have shown that abdominal visceral adipose tissue depth (VAD) measured by ultrasound in earlier pregnancy, may predict glucose intolerance, insulin resistance, metabolic syndrome, newborn weight and GDM later in pregnancy7,8,9,10,11,12,13. Recently, Thaware et al.14 showed that ultrasonography‐measured VAD in early pregnancy aided in the early recognition of GDM. They reported that this method, compared with an OGTT alone, could reduce by half number of pregnant women requiring OGTT screening. However, it is not clear whether ultrasonography-measured VAD offers improve discrimination in the detection of GDM compared with simply using pre-pregnancy body mass index (BMI). Herein, we carried out a prospective cohort study among a large sample of pregnant women, to assess the role of VAD in the prediction of GDM.

Methods

Design and study participants

This was a prospective cohort study of pregnant women, followed from the first trimester until 24 to 28 weeks’ gestation. The main outcome was development of GDM.

The study was performed at Instituto de Medicina Integral Prof. Fernando Figueira (IMIP), Recife-Brazil, from March 2016 to September 2018. IMIP is a large tertiary maternity hospital with approximately 6,000 deliveries annually. This project was previously approved by IMIP Ethics Committee, and all participants gave written informed consent. The study was developed in accordance with the relevant local guidelines and regulations

Women with a singleton pregnancy at <20 weeks’ gestation, based on last menstrual period or dating ultrasound, were invited to participate. Those with previous history of GDM or pre-gestational diabetes were excluded.

VAD measurement

VAD was measured by ultrasound by a single sonographer (AR–coordinator of the IMIP service of “Fetal Medicine”, national reference in prenatal care). The Samsung Medison, Accuvix V20 machine was used, according ti the technique by Armellini et al.15, with a slight modification as described by Martin et al.16. Intra-observer agreement for VAD has an intra-class correlation coefficient of 0.965 (95% CI 0.960 to 0.985). Measurements were made in supine position, and the 5.2 MHz curvilinear Array probe was placed on the anterior abdomen in the xipho-umbilical line, 1 cm above the umbilicus. VAD was measured as the perpendicular distance between the posterior aspect of the junction of the two rectus abdominis muscles (i.e., the linea alba) and the anterior aspect of the abdominal aorta.

Diagnosis of GDM

All pregnant women subsequently underwent a 75 g OGTT between 24 and 28 weeks’ gestational week, following a 12-hour fast, with sampling at 1 and 2 hours post-load. Plasma measurements were performed with glucose oxidase methods. A diagnosis of GDM was according to the International Association of Diabetes and Pregnancy Study Group (IADPSG) criteria17: fasting concentration ≥5.3 mmol/L, 1-hour ≥10.6 mmol/L, or 2-hour ≥9.0 mmol/L. Those diagnosed with GDM were referred for high-risk antenatal care at IMIP18.

Data analysis

Baseline characteristics of GDM and no-GDM pregnant women were compared using independent samples t-test for continuous variables and chi-squared tests for categorical variables. Logistic regression was performed to assess the risk for GDM per 1-cm unit rise in VAD. Included in the model were maternal age and pre-pregnancy BMI. Receiver operator characteristic curve analyses assessed the predictive value of VAD for new-onset GDM, and Youden index was calculated. Pearson’s correlation analysis was performed to analyze the correlation between VAD and each OGTT value. Statistical analysis was performed using STATA version 12.1; a 5% significance level was considered (p < 0.05).

Results

Of the 627 pregnant women enrolled, 518 (82.6%) completed an OGTT. Table 1 shows the characteristics of those who did and did not complete the OGTT. VAD was measured at a mean of 14.4 weeks’ gestation.

Table 1 Characteristics of pregnant women who completed the study and those lost to follow-up.
Full size table

The mean (SD; 95% CI) VAD by ultrasound was 5.44 (±1.27) cm; 95% CI 5.33–5.55. 87 (16.8%) pregnant women developed GDM according to IASDPG criteria. VAD, age and BMI pre-pregnancy differed between women who did and did not develop GDM (Table 2).

Table 2 Baseline characteristics of women according to their diagnosis of gestational diabetes mellitus in the index pregnancy.
Full size table

There was a significant correlation between VAD and the OGTT fasting glucose (r = 0.179, 95% CI 0.094 to 0.261; p < 0.001), 1-hour glucose (r = 0.238, 95% CI 0.154 to 0.319; p < 0.001) and 2-hour glucose (r = 0.221, 95% CI 0.136 to 0.303; p < 0.001) concentration.

There was a significant relation between VAD and subsequent development of GDM (OR = 1.99, 95% CI 1.59–2.46; p < 0.01) (Table 3). The association remained statistically significant after adjusting for maternal age and pre-pregnancy BMI (OR 2.00, 95% CI 1.61 to 2.50).

Table 3 Odds ratios for developing gestational diabetes mellitus in relation to visceral adipose tissue depth measured by ultrasonography at about 14 weeks’ gestation.
Full size table

The ROC curve analysis of GDM showed a higher under the curve for VAD (0.70 95%CI 0.63 to 0.75) than pre-pregnancy BMI (0.57 95% CI 0.50 to 0.64) (p < 0.0001 (Fig. 1). The optimal VAD cut off for that maximized Youden’s index was 5.1 cm.

Figure 1
figure 1

Receiver operator curves for visceral adipose tissue depth (VAD) and pre-pregnancy body mass index (BMI), and their respective associations with developing gestational diabetes mellitus.

Full size image

Discussion

An increasing VAD, measured by ultrasonography in early pregnancy, was associated with higher risk of GDM. This risk persisted after adjustment for pre-pregnancy BMI. Furthermore, ROC analysis showed VAD had a significantly higher predictive area under curve for developing GDM than did pre-pregnancy BMI.

The observed association between VAD in early pregnancy and a higher risk of GDM is consistent with other studies. However, our study also shows the superiority of first trimester VAD over pre-pregnancy BMI.

The first description of the association between visceral adiposity in early gestation and glucose intolerance in later pregnancy was made in 2009 by Martin et al.7. In 62 pregnant women, they observed that a VAD above the upper quartile value was associated with an increased risk for a positive glucose challenge test between 24 and 28 weeks’ gestation. De Souza et al.8 completed a prospective cohort study of 79 pregnant women and observed that VAD, measured by ultrasonography at 11 to 14 weeks’ gestation, explained 42% of the variance in HOMA-IR. In another prospective cohort of 485 women, De Souza et al.9 showed that an elevated VAD, assessed by ultrasound at 11 to 14 weeks’ gestation, independently predicted the risk of dysglycemia later in pregnancy. However, Pontual et al. observed that VAD, measured in the first half of pregnancy, was no better than pre-pregnancy BMI in predicting insulin resistance and dyslipidemia later in pregnancy16.

In a cohort of 1048 pregnant women, Bourdages et al.10 observed that first-trimester VAD was associated with a higher likelihood of developing GDM. They also reported a marginal improvement in GDM prediction using VAD, over using BMI alone. Recently, D’Ambrosi et al.11 showed that VAD measured at 24–28 weeks’ gestation, was higher in cases with GDM than controls with no GDM. VAD may be superior to BMI in predicting GDM, as BMI does not reflect the metabolically actively visceral adipose tissue compartment to the same degree as does VAD. Moreover, VAD, measured by ultrasound, is a safe and simple method to carry out in early pregnancy, such as at the time of pregnancy dating or at the determination of fetal nuchal thickness7,8,9.

This study has several strengths. It prospectively included a large sample of women, using a standardized protocol. IADPSG criteria were used to diagnose GDM. All sonographic exams were performed by an single ultrasonographer, previously trained in the technique of Armellini et al.14. This study also has some limitations. Pregnant women entered the study in the end of the first trimester of pregnancy, once some metabolic and habitus changes would have occurred. During follow-up, 17.4% of participants left the study. The worsening economic crisis in Brazil during the data collection phase, along with the onset of the Zika virus epidemic, likely contributed to loss of follow-up. Even so, the clinical characteristics of those who did and did not complete the study were similar.

In conclusion, there was an increased risk of GDM in relation to VAD measured in early pregnancy. This association remained so after adjusting for BMI, and VAD was more predictive of GDM than pre-pregnancy BMI. Further studies are needed to better explore whether a reduction in VAD depth can reduce the subsequent risk of GDM, whole improving perinatal outcomes.