During the last decade, the incidence of gestational diabetes mellitus (GDM) has witnessed a global increase from 10 to 26 percent, possibly due to the increase in maternal obesity and maternal age at conception [1]. GDM occurs when insulin resistance is diagnosed for the first time during pregnancy, mostly during 24–28 weeks of gestation [2]. Although the reason that triggers its development is poorly understood, there is a consensus that dysfunction of $\beta$-cells and failure of insulin secretion in response to pregnancy-induced insulin resistance are likely key players leading to GDM [3, 4]. What is unique about this disorder is that it imposes multiple health challenges to the pregnant mother and her growing fetus, such as preeclampsia, large for a date, preterm labor, and increased perinatal morbidity and mortality [5]. Further, GDM lays its dark shadows on materno-fetal health post-delivery. It was reported that affected cases suffer from a 50 percent risk of frank diabetes mellitus (DM) during the subsequent five years. Both the mother and the newborn suffer a higher future risk of metabolic, glycemic, and cardiovascular risk [6]. It is empirical to establish GDM diagnosis to limit its consequences on maternal and neonate well-being [7].

Currently, the gold standard for establishing a GDM diagnosis is the 75 g oral glucose tolerance test (GTT) [8]. The test has good predictive value, allowing for risk stratification for screened cases. However, it requires multiple blood draws, which makes it time-consuming and stressful for the mothers and thus reduces its acceptability [9, 10]. Some researchers have criticized the test, suggesting that feto-maternal harm is already done by the time of the diagnosis [11]. An urgent need arises for earlier, reliable, and comprehensive screening bio-markers for GDM. Adiponectin is a hormone manufactured by adipose cells; it exhibits anti-inflammation and anti-atherogenic and insulin sensitization properties [12, 13]. Physiologically, adiponectin regulates fat metabolism, and it improves overall insulin sensitivity. Reduced maternal levels suggest increased GDM risk among screened moms as it mirrors glucose intolerance before it clinically manifested [14]. The triglyceride-glycemic (TyG) index is an innovative instrument to estimate insulin resistance (IR), especially among obese individuals and adolescents [15]. The TyG index has been tested in Korean and Brazilian studies and showed promising results [16, 17]. Interestingly, it is accessible in many clinics and has outperformed the Homeostatic Model Assessment of Insulin Resistance in estimating IR. Recently, the TyG index was tested as a potential GDM screening marker; however, the recommendations regarding its performance were inconclusive [16, 17, 18].

This study was designed to examine the performance of maternal adiponectin and the TyG index in screening for GDM among Iraqi women for a more effective and cost-benefit analysis.

The basic criteria of the study were showed in Table 1. The maternal age were statistically higher among cases p = 0.001; while all other parameters were statistically insignificant (miscarriage, parity , BMI , and gestational age) as p $>$ 0.05. Fasting blood sugar and 2-hour glucose tolerance test were significantly higher among GDM cases than in healthy pregnant; p 0.0001 and p 0.0001, respectively. The LDL was statistically insignificant among the two groups; while HDL was significantly higher among controls; p 0.001. The total cholesterol and TG were significantly high among the GDM cases , p 0.001, p = 0.0007, respectively described in Table 1. Serum adiponectin was significantly higher among healthy pregnant (8.44 $\pm$ 1.12 ng/mL vs. 5.28 $\pm$ 0.89 ng/mL); p 0.0001. While the TyG index was significantly higher among GDM cases (4.02 $\pm$ 0.04 vs. 3.96 $\pm$ 0.024), p 0.0001, shown in Fig. 2.

Fig. 2.

Box and whisker chart for serum adiponectin and TyG index among healthy pregnant vs. GDM case. TyG index, triglyceride-glycemic index.

Table 2 describes the Pearson correlation between serum adiponectin vs. some of the study parameters. A strong inverse correlation was proved with fasting blood sugar (FBS), 2-hour glucose tolerance test (2hr_GTT) with r = (–0.76, –0.80); p 0.0001, respectively. The TyG index was moderately, inversely, and significantly linked to serum adiponectin as r = –0.58; p 0.0001.

The ROC curve determined the validity of both markers, serum adiponectin and TyG index, in screening for GDM, shown in Fig. 3.

Fig. 3.

The ROC curve showing adiponectin and TyG index cut-off value in discriminating GDM cases. ROC, receiver operator characteristic curve.

In Table 3, a pairwise comparison was shown between adiponectin and TyG index with respect to their respective cut-off values, 95% CI, standard error (SE), sensitivity, specificity, and significant level among the two markers. It was shown that both adiponectin and TyG index reliably predicted GDM with a high area under the curve (AUC) of 0.83 vs. 0.88; p 0.001. The difference in their significant level was 0.46, which is statistically insignificant.

The analysis confirmed reduced levels of serum adiponectin among cases with GDM. It was significantly and inversely linked to FBS and 2hr_GTT. In line with our results, Gao et al.’s [22] meta-analysis showed an inverse relationship between adiponectin and GDM risk; their study confirmed significantly lower levels among pregnant mothers with GDM vs. mothers with normoglycemic pregnancies. They recommended it as a screening biomarker for GDM. In agreement with our analysis, Bao et al. [23] discussed lower levels of GDM among women in the first and second trimesters.

Hedderson et al. [24] study examined an interesting role of adiponectin; they linked pre-pregnancy adiponectin with the subsequent risk of developing GDM, which can be used in practice as a risk stratification tool to spot high-risk women for GDM before they get pregnant. In accordance with Hedderson et al. [24], Iliodromiti et al. [25] meta-analysis confirmed adiponectin prediction for GDM before and at early pregnancy, with a diagnostic odds ratio of 6.4 and 95% CI of 4.1 to 9.9. Adiponectin levels were lower in the placenta of affected rodents, underscoring its role in placenta development. Moreover, adiponectin receptor-2 and not 1 was significantly lower, which may be the way by which the marker exhibits its effect [26]. The current study showed a good prediction power for serum adiponectin at high sensitivity and specificity (93.2%, 84.1%); p 0.001, respectively.

Rodent’s study signified the therapeutic value of adding Adiponectin to diabetic rates, leading to better glucose levels and protecting against fatty liver [27]. While animal study recommend the therapeutic potential of adiponectin administration, its role in humans, especially among pregnant women, is yet to be confirmed [28]. Some suggested its administration to women in the post-natal periods to reduce their progression risk from GDM into frank type 2 diabetes (T2D) [12, 29]. Adiponectin is an adipokine secreted by fatty cells and has a key role in glucose and lipid metabolism. Reduced levels were linked to higher IR and T2D. Lower levels in pregnancy underscore the metabolic imbalance that defines GDM [30]. This was confirmed in the current study by a strong statistically significant correlation of adiponectin to FBS and 2hr_GTT (–0.76, –0.80), p 0.0001, respectively.

The current results showed a significantly high TyG index among GDM cases p 0.0001; it was moderately, inversely, linked to serum adiponectin p 0.0001; moreover, TyG index distinguished GDM cases with significant AUC 0.88; p 0.001 at the second trimester of pregnancy (24–28 weeks).

In accordance with our results, Liu et al. [18] examined the performance of the TyG index in screening for IR among pregnant women. Their result highlighted significant differences in the TyG index levels in GDM cases (4-fold rise) vs. healthy pregnant. They further highlight that the second trimester was the best time for performing screening by the TyG index, which can be a promising tool for GDM prediction [18]. Sánchez-García et al. [31] examined the diagnostic performance of the TyG index compared to a 2-hour 75-g oral glucose tolerance test in screening for GDM in a comparative study at 24–28 weeks. They also examined their correlations with materno-fetal outcomes. TyG index was significantly high among GDM cases, p 0.001, while our results showed a p-value of 0.0001 at the same gestational age. Furthermore, the TyG index discriminated GDM cases with 89% sensitivity and 50% specificity compared to 84.1% sensitivity and 81.8% specificity, p 0.001 in this study. However, no differences were noticed in the outcome. They recommended the TyG index for its feasibility, affordability, and sensitivity for GDM screening [31].

Higher levels of TyG index among GDM cases and its reliability in the prediction of GDM were discussed by many scientists. Salvatori et al. [32] conducted a cohort study for women to see the best association between TyG index and 2 h-75 g oral GTT-derived insulin sensitivity. With the use of machine learning, the data they recruited on 2-visits, early in pregnancy and at twenty-sixth weeks, were analyzed. Their result showed a weak positive association between 2 h-75 g oral GTT and the conventional TyG index. They proposed a newer improved ratio called TyGIS (the linear function of TyG, body weight, lean body mass percentage, and fasting insulin), which yielded a much more efficient ratio in screening for GDM as p 0.0001 [32]. A study by Pazhohan et al. [33] tested first-trimester fasting blood sugar and TyG index in predicting GDM and large-for-date babies; their result confirms the reliability of the TyG index in screening for both. Yoon et al.’s study [34] discussed that the TyG index association with IR among T2D in children and adolescents was strong and outstood the homeostatic model assessment of insulin resistance. They advised its integration into clinical practice as an early screening test, allowing interventional strategies to halt the increased risk.

Interestingly, the TyG index served as a prognostic factor for cardiovascular accidents (CVSA) among T2D cases. Long monitoring of the ratio helped stratify the CVSA risk among the following cases and independently predicted the occurrence of those events [35]. The rationale behind the TyG index association with GDM is that the IR, a defining feature of GDM, is correlated with abnormal lipid metabolism, like high TG concentrations. Once IR rises, it damps insulin’s inhibitory effect on liver neogenesis and the peripheral uptake of glucose, leading to hyperglycemia. The high IR increases lipolysis and thus elevates blood lipids, i.e., TG.

Currently, screening for GDM is via oral GTT, the gold standard. However, the test is time-consuming and sometimes inconvenient for pregnant women. Since the test takes place in a single session, it offers a limited snip of the dynamic changes in blood glucose. Not to mention that factors such as stress and physical activity may trigger some variability in oral GTT test results [11]. The makers used (adiponectin and TyG index) can be calculated in single blood aspiration, and the result mirrors a broader picture of blood glucose and lipid metabolism, allowing a holistic assessment of metabolic health. Not to mention that changes in the 2-markers under the study occur earlier than oral GTT, which gives time for intervention [36]. The ROC showed that adiponectin and TyG index reliably predicted GDM with high AUC 0.83 vs. 0.88; p 0.001. The pairwise comparison failed to detect a statistical difference in their performance; p = 0.46 confirmed their value. The markers tested here are not suggested to replace oral GTT rather than being complementary tests with added benefits:

(1) By passing oral GTT limitations [9, 10].

(2) Earlier detection for GDM [11].

(3) The analysis confirmed their efficacy for screening with the advantage of less patient burden.

Finding newer biomarkers may improve patients’ identification, especially those that may benefit from robust monitoring and lifestyle changes. This approach will optimize care and align with evidence-based, patient-centered holistic health care.

Many biomarkers for GDM are tested in practice; the markers’ affordability, availability, and accuracy are crucial for their implantation in practice. Adiponectin and TyG index performed well in screening for GDM among Iraqi females. They were comparable in performance, and no significant difference appeared upon comparison. Their reliability opens the door for earlier diagnostic and promising therapeutic applications in GDM.

FBS, fasting blood sugar; 2hr_GTT, 2-hour glucose tolerance test; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TG, triglyceride; TyG index, triglyceride-glycemic index; BMI, body mass index; HOMA-IR, Homeostatic Model Assessment for Insulin Resistance; IR, insulin resistance; ROC, receiver operator characteristic curve; AUC, area under the curve.

The data supporting this study is available on reasonable request from the corresponding author.

Conceptualization: HAGA and AHI; methodology: WN and DAS; software: WA; investigations and data curation: HAGA and AHI; writing, reviewing and editing: WN and HAGA; drafting and supervision: WN. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board. The scientific and ethical committees at the College of Medicine/Mustansiriyah University approved its protocol (IRB NO.18 dated 25/2/2021). Informed consent was obtained from all subjects involved in the study.

To our beloved university for continued support.

This research received no external funding.

The authors declare no conflict of interest.