Abdominal subcutaneous fat thickness combined with a 50-g glucose challenge test at 24-28 weeks of pregnancy in predicting gestational diabetes mellitus

Abstract

Background

This investigation aimed to analyse the efficacy of abdominal subcutaneous fat thickness (ASFT) value >18.1 mm combined with a 50-g glucose challenge test (GCT) between 24-28 weeks of gestation in predicting gestational diabetes mellitus (GDM) cases.

Methods

This cross-sectional study was carried out from February 2021 to December 2022. All pregnant women received a 50-g GCT at 24-28 weeks of pregnancy for the GDM screening. Pregnant women with a blood glucose value between 140-190 mg/dl experienced 100 g OGTT. Even if 50-g GCT was normal, 100-g OGTT was offered to patients with an ASFT value above 18.1 mm

Results

Among the 728 pregnant women we enrolled, 154 (21.2%) cases were screened as positive. The number of patients who first screened positive and determined to be GDM after the 100-g oral glucose tolerance test (OGTT) was 43 (5.9%). A total of 67 cases (9.2%) had an ASFT measurement above 18.1 mm. Two cases with a negative 50-g GCT and ASFT <18.1 mm were diagnosed as GDM in the later weeks of pregnancy. A 50-g GCT combined with ASFT measurement above 18.1 mm predicted GDM with a sensitivity of 87.9%, a specificity of 88.7%, a positive predictive value (PPV) of 36.0%, and a negative PV (NPV) of 99.7%.

Conclusions

A 50-g GCT combined with ASFT measurement that can be easily and accurately obtained during routine antenatal care in the second trimester might be a beneficial indicator for predicting GDM cases.

PLAIN LANGUAGE SUMMARY

Screening and diagnosing pregnant women at greater risk of developing gestational diabetes mellitus are crucial to enhancing short- and long-term outcomes of the mother and foetus. An accurate diagnosis could provide proper treatment, which could be dietary or pharmacological, manage the disease, and improve pregnancy outcomes. In the current study, we revealed that gestational diabetes was predicted with high sensitivity and specificity in pregnant women with a 50-gram glucose challenge test and abdominal subcutaneous fat thickness measurement above 18.1 millimetres. Therefore, abdominal subcutaneous fat thickness measurement is anticipated to be extensively used as an indicative variable for predicting gestational diabetes mellitus cases during the second trimester of pregnancy.

Keywords:

• abdominal subcutaneous fat thickness
• gestational diabetes
• glucose challenge test

Introduction

Gestational diabetes mellitus (GDM) is defined as the first detection or the onset of diabetes or variable severity of impaired carbohydrate intolerance during the second or third trimester of pregnancy, which is definitely not a type 1 or type 2 diabetes undiagnosed prior to or began concomitantly during pregnancy (American Diabetes Association Professional Practice Committee 2022, Johns et al. 2018, Oğlak and Obut 2021). The prevalence of GDM varies between 9.3 and 25.5% depending on the diagnostic criteria applied and the features of the population (Sacks et al. 2012). Uncontrolled hyperglycaemia contributes to a substantial risk for unfavourable gestational consequences for both mother and child (Plows et al. 2018, Oğlak et al. 2020). Also, GDM patients have a considerably increased risk for metabolic diseases following pregnancy, while their offspring are at a higher risk of developing obesity and metabolic syndrome (Chiefari et al. 2017, Oğlak et al. 2022). The GDM incidence is increasing globally because of the increasing obesity prevalence, which was the major risk factor for developing GDM by contributing peripheral insulin-resistant state (Chiefari et al. 2017). Screening and diagnosing pregnant women at greater risk of developing GDM are crucial to enhancing short- and long-term outcomes of the mother and foetus. An accurate diagnosis could provide proper treatment, which could be dietary or pharmacological, manage the disease, and improve pregnancy outcomes. However, there is a disagreement among healthcare providers regarding the screening and diagnosis of GDM (McIntyre et al. 2015).

Maternal obesity is strongly linked with various adverse perinatal and maternal outcomes, including insulin resistance, GDM, and metabolic syndrome (Marchi et al. 2015). However, the prevalence of GDM shows significant differences among pregnant women with obesity. Maternal obesity is a heterogeneous complication with the body fat distribution responsible more for metabolic disturbances and related carbohydrate intolerance risks than the total body weight (Nassr et al. 2018). The adipose tissue storage presents itself in two distinct compartments as central adipose tissue, which accounts for 15% of the total, and subcutaneous adipose tissue, which represents the remaining 85% (Foster and Pagliassotti 2012). An increase in fat tissue accumulation in the central compartment is highly associated with an increased risk of complications related to obesity than the amount of peripheral fat (Engin 2017). The measurement of maternal central obesity is essential to identify the percentage of adipose to nonadipose tissue or the fat distribution (Yao et al. 2020). Body mass index (BMI) is widely used in daily medical practice to determine the degree of overweight. Nevertheless, since BMI relies on the measurement of a patient’s height and weight, this index could not discriminate increased adipose tissue mass, muscle, or bone. Moreover, BMI could not reflect the pattern of the fat distribution in the adipose compartments, which is a significant factor reported to correlate with obesity-associated pregnancy complications. Also, diverse anthropometric measures are influenced by the amount of subcutaneous fat accumulation and are related to a significant inter-observer and inter-ethnicity variability. Therefore, these measurements are not considered reliable and accurate measures for central abdominal obesity (Nassr et al. 2018).

Ultrasonography (US) is a low-cost, non-invasive, and effective procedure to assess the amount of maternal central fat throughout gestation (Tunc et al. 2022). Our previous studies reported that an increased abdominal subcutaneous fat thickness (ASFT) in pregnant women (>18.1 mm), which has been demonstrated to be easily measurable by the US and useful for predicting the fat tissue accumulation in the central compartment, was highly associated with the subsequent GDM development (Budak et al. 2019). Some studies reported that ASFT measurement is supposed to be increasingly used as a complementary diagnostic criterion throughout routine pregnancy evaluation for predicting GDM risk during the second trimester of gestation (Yang et al. 2017, Kansu-Celik et al. 2018, Akgöl et al. 2022). However, currently, screening of pregnant women for GDM is performed with a one-step 75-g oral glucose tolerance test (OGTT) or a two-step 50-g glucose challenge test (GCT) followed by a 100 g OGTT at 24–28 weeks of gestation (American Diabetes Association Professional Practice Committee 2022). A comprehensive meta-analysis that included 26 studies reported that 50-g GCT, which was the most commonly used GDM screening test, has a specificity of 77% and a sensitivity of 74% for diagnosing GDM (van Leeuwen et al. 2012). Therefore, additional probabilities of combining the 50-g GCT with other screening procedures should be investigated to warrant higher accuracy measures.

This study aimed to analyse the efficacy of ASFT value >18.1 mm in combined with a normal or abnormal 50-g GCT between 24-28 weeks of gestation in predicting GDM cases.

Methods

This cross-sectional study was performed at Diyarbakır Gazi Yaşargil Training and Research Hospital, Turkey, from February 2021 to December 2022. The study received approval from the Institutional Ethics Committee of Diyarbakır Gazi Yaşargil Training and Research Hospital (15.01.2021/613). We invited singleton pregnant women to participate in the current investigation during their antenatal visits between 24-28 weeks of gestation. The exclusion criteria were pregnant women with multiple gestations, a known pre-gestational diabetes mellitus, a previous GDM pregnancy history, chronic medication with drugs that cause carbohydrate intolerance, GDM diagnosed in the first trimester of pregnancy (fasting plasma glucose ≥92 mg/dL) (International Association of Diabetes and Pregnancy Study Groups Consensus Panel 2010), and foetal major structural and chromosomal abnormalities.

Prior to the US examination, baseline features of pregnant women, including maternal age, parity, pre-gestational body weight, weight gain during pregnancy, BMI and the gestational week at the time of screening were recorded, and morning pre-prandial blood was drawn for biochemical evaluation. The fasting blood glucose level of the participants was measured.

All pregnant women included in this study underwent US measurement of ASFT at 24-28 weeks of pregnancy by an experienced sonographer (S.C.O.) utilising a technique in our previous study (Budak et al. 2019). We performed a transabdominal US utilising an RAB4-8 curvilinear array probe with a range of frequency between 4 and 8 MHz. US examinations were carried out by the same physician (S.C.O.) to create good reproducibility and reliability. The physician decreased the image depth to reduce the error margin to ensure correct measurement. A longitudinal scan was obtained by the sonographer, confirming that the US transducer utilised the lowest possible pressure on the abdominal wall to prevent adipose tissue compression. The landmarks, including skin, linea alba, umbilicus, iliac crest, and rectus abdominis muscle, were identified in every single image. The ASFT measurement point was determined as the cross-section point of the horizontal line drawn across the highest points of both iliac crests and linea alba. We measured the ASFT from the skin surface to the rectus abdominis muscle’s outer border in millimetres (Budak et al. 2019). The same operator carried out the ASFT measurement three times, calculated the mean ASFT value for each pregnant woman, and recorded it.

All pregnant women subsequently received a 50-g GCT at 24-28 weeks of pregnancy for the GDM screening. All cases received 50 g glucose in water without regard to the time of the last meal or time of day. According to this non-fasting test, a 1-h venous blood sample is drawn. A blood glucose level <140 mg/dL (7.8 mmol/L) was considered normal. Cases with a blood glucose level of ≥190 mg/dl were considered as GDM and further testing was not utilised. Pregnant women with a blood glucose level between 140-190 mg/dl experienced 100 g OGTT in the morning after an overnight fast of at least 8 hours. Moreover, even if 50-g GCT was normal, 100-g OGTT was offered to patients with an ASFT value above 18.1 mm. In preparation for this test, all cases were required to consume a rich carbohydrate diet for at least 3 days beforehand, with an overnight fast before the morning test. The 100-g OGTT consisted of ingesting, within 5 minutes, 100 g oral glucose load. Venous blood samples were obtained before the test and 1, 2 and 3 hours after the ingestion (Lurie et al. 1998). Following the 100-g glucose challenge, the GDM diagnosis was done if two or more plasma glucose values at or above the following thresholds: fasting glucose ≥95 mg/dL (5.5 mmol/L), 1-h glucose ≥180 mg/dL (10 mmol/L), 2-h glucose ≥155 mg/dL (8.6 mmol/L), or 3-h glucose ≥140 mg/dL (7.8 mmol/L) (American Diabetes Association Professional Practice Committee 2022). The primary outcome of this investigation was to examine the effectiveness of ASFT value >18.1 mm in a pregnant woman with a normal or abnormal 50-g GCT between 24-28 weeks of gestation as a predictor of GDM.

Quantitative determinations of venous blood glucose measurements were performed from all tubes immediately after centrifugation on the Abbott Architect System Clinical Chemistry Analyser (ARCHITECT c8000®, Abbott Diagnostics, Abbott Park, IL, USA) using the hexokinase method. Glucose was assayed using a commercial glucose reagent kit (CLINICAL CHEMISTRY Concentrated Glucose Reagent Kit®, Abbott Diagnostics, Abbott Park, IL, USA). The measuring range was 5-800 mg/dL (0.28-44.4 mmol/L) and intra- and interassay coefficient of variation (CV) values were 1.98% and 0.84%, respectively.

Statistical analysis

Sample size calculation and power analysis: The prevalence of GDM is between 6.1% and 8.9% in Turkey (Akbay et al. 2010, Ozgu-Erdinc et al. 2019). A total sample size of 550 (which includes 49 subjects with the disease) achieves 81% power to detect a change in sensitivity from 0.5 to 0.7 using a two-sided binomial test and 100% power to detect a change in specificity from 0.5 to 0.7 using a two-sided binomial test. The target significance level is 0.05. The actual significance level achieved by the sensitivity test is 0.0444 and achieved by the specificity test is 0.0476.

IBM SPSS Statistics for Windows, Version 26.0 was utilised for each statistical procedure utilised. Kolmogorov-Smirnov and Shapiro-Wilk tests were utilised to test the normality of data. Continuous variables that meet the normal distribution assumption were compared utilising Student’s t-test and the remaining by utilising the Mann-Whitney U test. The Chi-Square Test was utilised to compare the differences in proportions between groups. According to the ASFT value, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for the GDM prediction were established. A p-value <0.05 was established as statistically significant.

We examined the intraobserver reliability utilising the intraclass correlation (ICC) coefficient with a 95% confidence interval (CI). Based on this estimation, values >0.90, between 0.75-0.90, between 0.5-0.75, and <0.5 indicate excellent, good, moderate, and poor reliability, respectively (Koo and Li 2017).

Results

During the investigation period, 761 pregnant individuals between 24-28 weeks’ gestation were admitted to our outpatient clinic within the weeks of the 50-g GCT screening and recruited for the study. Of these, 7 patients had a history of GDM, 6 patients had a first-trimester GDM diagnosis, 6 patients presented with known pregestational diabetes, three patients had multiple pregnancies, and three cases presented with a foetal abnormality. Also, three patients were lost to follow-up before 28 weeks of gestation, and five patients opted out of the study. Finally, 728 patients were eligible for this investigation and presented to our cohort (Figure 1).

Figure 1. Flowchart of the study.

We performed a reliability analysis according to the pregnant patient’s ASFT measurements. ICC analysis indicated excellent intra-observer (0.928; 95% CI: 0.912-0.951) reliability of ASFT measurement.

We summarised the characteristics of pregnant women with and without gestational diabetes in the gestational diabetes screening program between 24-28 weeks of gestation in Table 1. The mothers were significantly older in the GDM group (29.4 ± 5.6 years) than the pregnant women without GDM group (25.9 ± 4.8 years, p < 0.001). The mean gestational weight gain was similar between the groups. The mean BMI and ASFT measurements were significantly higher in cases with GDM than those of the cases without GDM (p < 0.001).

Table 1. Characteristics of pregnant woman with and without gestational diabetes in the gestational diabetes screening program between 24 and 28 weeks of gestation.

	Gestational diabetes group (n = 58)		Pregnant woman without gestational diabetes (n = 670)
	Mean ± std	Median (Q1-Q3)	Mean ± std	Median (Q1-Q3)	p-value
Maternal age, years	29.4 ± 5.6	32 (26-39)	25.9 ± 4.8	29 (22-36)	<0.001
Gravidity, n	4.7 ± 1.6	4 (2-6)	2.5 ± 1.4	3 (1-5)	<0.001
Parity, n	3.4 ± 1.8	3 (1-5)	2.2 ± 1.3	2 (0-4)	<0.001
History of abortion, n (%)	4 (6.9%)		37 (5.5%)		0.293
Previous caesarean section, n (%)	13 (22.4)		162 (24.1)		0.571
Gestational week at screening	26.1 ± 1.8	26 (24-28)	26.3 ± 1.4	26 (24-28)	0.197
Gestational weight gain, kg	6.2 ± 3.7	6 (5-10)	5.9 ± 2.8	5 (4-10)	0.086
Body mass index, kg/m^2	27.7 ± 4.9	28.2 (26.6-31.3)	25.2 ± 3.1	25.4 (24.1-30.3)	<0.001
Abdominal subcutaneous fat tissue thickness, mm	20.1 ± 1.6	20.9 (16.8-25.9)	16.8 ± 1.1	16.6 (15.2-21.1)	<0.001

Bold values denote statistical significance at the p-value < 0.05 level.

Among the 728 pregnant women we enrolled, 154 (21.2%) cases were screened as positive, and 574 (78.8%) cases were screened as negative during the 50-g GCT (GDM screening test). The number of cases that initially screened positive but proved to be negative following the 100-g OGTT (GDM confirmatory test) was 111, which accounts for 72.1% of 50-g GCT positive cases, and 15.2% of the study population. The number of patients who first screened positive and determined to be GDM after the 100-g OGTT was 43, which accounts for 27.9% of GDM screening test positive cases, and 5.9% of the study cohort. A total of 67 cases (9.2%) had an ASFT measurement above 18.1 mm; of these, 53 cases had a negative, and 14 cases had a positive 50-g GCT. Two cases with a negative 50-g GCT and ASFT <18.1 mm were diagnosed as GDM in the later weeks of pregnancy. A 50-g GCT combined with ASFT measurement above 18.1 mm predicted GDM with a sensitivity of 87.9%, a specificity of 88.7%, a PPV of 36.0%, and an NPV of 99.7% (Table 2).

Table 2. Prediction of gestational diabetes mellitus cases wit a positive 50-g glucose challenge test in combination with abdominal subcutaneous fat thickness value >18.1 mm.

	Sensitivity	Specificity	Positive predictive value	Negative Predictive value
Positive 50-g GCT and/or ASFT >18.1 mm	87.9%	88.7%	36.0%	99.7%

GCT: glucose challenge test; ASFT: abdominal subcutaneous fat thickness.

Discussion

In the current research, we found that GDM was predicted with 88.7% specificity, 87.9% sensitivity, 36.0% PPV, and 99.7% NPV in pregnant women with a 50-g GCT and ASFT measurement above 18.1 mm.

The physiopathological pathways underlying the association between maternal central obesity and the possibility of insulin resistance and diabetes remain unclear. In addition to being a reservoir for energy storage, adipose tissue is gradually accepted as an essential immune organ that releases signals. There is substantial evidence for the effects of TNF-α, IL-6, adiponectin, and leptin on adipocyte metabolism and insulin sensitivity (Stolarczyk 2017). Visceral fat is considered not sensitive to insulin’s antilipolytic effect, and excess fat tissue in the visceral compartment causes a higher release of free fatty acids (FFAs) into the portal system along with pro-inflammatory cytokines (Gur et al. 2014). The arrival of these FFAs and cytokines to the liver induces both insulin resistance in the liver by increasing gluconeogenesis and in the muscle tissue leading to diabetes mellitus (Kansu-Celik et al. 2018). GDM pregnancies have a considerable degree of insulin resistance compared to healthy pregnancies. This mechanism closely resembles that of the type 2 diabetes (T2DM) development (Plows et al. 2018). As a result of this, the prevalence of GDM and T2DM is frequently similar in the same population and these prevalences are often increased by an elevation in the prevalence of obesity (Zhu and Zhang 2016).

Traditionally, BMI is used to evaluate the risk situation of obesity-related complications in maternity care services. But, BMI indicates total body weight rather than body fat (Gába and Přidalová 2016). It was demonstrated that the ratio of visceral to subcutaneous ATT may vary in two cases with identical WC (Gur et al. 2014). Also, gestational weight gain might be related to a growth of the subcutaneous adipose tissue, whereas visceral adipose tissue is a previously existing and predisposing factor in the first weeks of gestation (D'Ambrosi et al. 2020, Yurtcu et al. 2021). Therefore, assessing the association between maternal central obesity and the possibility of insulin resistance and GDM could be favourable for evaluating the impact of fat distribution. In our previous study by Budak et al. we concluded that when 18.1 mm was utilised as the ASFT cut-off level, GDM cases were identified with a high sensitivity and specificity. Also, GDM risk was detected to be increased 3.86-fold in those with ASFT levels above 18.1 mm (Budak et al. 2019).

Previous studies reported numerous methods for the estimation of ASFT and central adiposity. These methods, including CT scan, and bioelectrical impedance measurements, which have been commonly utilised in the non-pregnant population to particularly measure central adiposity, are cost-intensive and also could not be utilised during gestation because of the changes in body water distribution during pregnancy and the foetal risks related to the exposure of radiation (Bray et al. 2008, Gur et al. 2014). Studies indicated that the ASFT as estimated by the US demonstrated a significant correlation with the subcutaneous fat area at the umbilical region as estimated by computed tomography (CT) scan (Hirooka et al. 2005). Also, US-measured ASFT is a simple, non-invasive measurement that can be performed in just a few minutes. Therefore, it is rational to utilise the US to measure second-trimester ASFT in relation to the subsequent development of GDM. Vlachos et al. stated in a review that in all published studies, the coefficient of variations of ASFT between different examiners has been reported to be less than 6% with excellent reproducibility and repeatability (Vlachos et al. 2007). Also, our study demonstrates that ICCs were higher than 0.9 for intra-observer agreement, indicating excellent reliability.

Universal guidelines concerning GDM screening underline the significance of causative factors in the development of this pregnancy complication. However, these causative factors are frequently observed and indiscriminate in the population to guide clinicians to test all pregnant individuals for GDM (Nassr et al. 2018). Screening all pregnant women is assumed to detect higher GDM frequencies than testing specified risk populations. Pregnant women might endure the OGTT as troublesome as the method is long-lasting and the glucose solution might induce vomiting. Moreover, given the absence of cost-effectiveness investigations utilising the novel diagnostic criteria, it remains challenging which screening procedure is more suitable (van Montfort et al. 2021). Whereas numerous guidelines suggest universal screening, optional screening remains a traditional procedure in some developed countries (Hod et al. 2018). Various studies have assessed the 50-GCT accuracy as a screening test for GDM and concluded different findings. According to the test utilisation (diagnostic or screening) and false-negative and false-positive test findings, specific combinations of precision levels are favoured. van Leeuwen et al. reported that the pooled sensitivity and specificity for the threshold value of 140 mg/dL were 74% and 77%, respectively. They recorded that if the 50-g GCT is utilised as a screening test, a greater sensitivity than 74% would possibly be certified to tolerate a false-positive rate of 83%. Also, if this test is used as a diagnostic tool for GDM, greater diagnosing rates are needed because of the relatively low GDM prevalence in the general population. If the 50-g GCT is combined with the presence of risk factors for GDM, the post-test possibility could be improved (van Leeuwen et al. 2012). Cases with a higher risk of GDM might be clarified by a list involving separate risk factors. Central obesity, as evaluated by ASFT, is one of the crucial risk factors for GDM. This measurement might determine a population at greater GDM risk and identify pregnant women who might benefit from testing and possible glycemic control before or during the regular 24-28-week screening period. In the current study, ASFT measurement >18.1 mm in combination with a 50-g GCT had a sensitivity of 87.9%, a specificity of 88.7%, a PPV of 36.0%, and an NPV of 99.7% to predict GDM cases.

Based on our clinical observations, some pregnant women are unwilling to experience OGTT during antenatal care. However, ASFT can be easily measured and might be used in combination with a 50-g GCT in GDM screening. Thus, these patients might be enlightened according to the obtained value and could be convinced to undergo OGTT after a positive 50-g GCT. ASFT could determine cases labelled as low risk for GDM that routine screening might be useless, and the actual guidelines for universal screening might be reviewed.

The major strength of this investigation is its novelty, being the initial study on GDM prediction utilising ASFT measurement by the US combined with 50-g GCT. The main limitation is the single-centre design that the participants who were admitted to a tertiary referral hospital might not represent the entire population of pregnant individuals.

Conclusion

The findings of the current investigation clearly showed that a 50-g GCT combined with ASFT measurement that can be easily and accurately obtained during routine antenatal care in the second trimester might be a beneficial indicator for predicting GDM cases. Therefore, ASFT measurement is anticipated to be extensively used as an indicative variable for predicting GDM cases during the second trimester of pregnancy.

Ethical approval

We obtained a written informed agreement from all subjects before being involved in this research. The study project was approved by the local ethics committee of Diyarbakır Gazi Yaşargil Training and Research Hospital (15.01.2021/613). This study was performed consistent with the Declaration of Helsinki Ethical Principles.

Authors’ contributions

S.C.O., E.Z.Y., and M.Ş.B. performed the conceptualisation and designation of the research study. S.C.O. performed the research, experimental protocols, and data curation. S.C.O., E.Z.Y., and M.Ş.B. analysed the data and statistics. S.C.O. wrote the manuscript. S.C.O. performed the critical review of the manuscript. All authors have participated sufficiently in the work to take public responsibility for appropriate portions of the content and agreed to be accountable for all aspects of the work in ensuring that questions related to its accuracy or integrity. All authors have read the final manuscript and agreed on final approval of the version to be published.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Additional information

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.