Diabetes is a metabolic disorder characterized by chronic hyperglycemia and is closely associated with increased mortality rates (1). According to data from the International Diabetes Federation, the global prevalence of diabetes was 10.5% in 2021, and it is projected to rise to 12.2% over the next 24 years (2). Given the large and growing number of individuals with diabetes, there is an urgent need to develop effective strategies for the prevention of this disease. Prediabetes is an intermediate stage between normoglycaemia and diabetes, and it is used to identify individuals at high risk for future diabetes (3). Studies have shown that prediabetes represents the optimal stage for intervention in the worsening trajectory of blood glucose levels (4). Several completed randomized controlled trials have demonstrated that lifestyle and pharmacological interventions can effectively control the progression of prediabetes, thereby reducing the incidence of diabetes and cardiovascular complications (5–7). Even a temporary reversion to normoglycaemia in individuals with prediabetes is significantly associated with a reduced incidence of diabetes (8). Therefore, early identification of modifiable factors influencing the reversion of prediabetes to normoglycaemia or its progression to diabetes is crucial for reducing the burden of diseases related to impaired glucose metabolism.

IR is a pathophysiological condition characterized by reduced responsiveness of target organs or tissues to insulin, leading to impaired glucose utilization. It is a significant risk factor for the development of diabetes and cardiovascular diseases (9). Early monitoring and control of IR are vital for the reversion or progression of prediabetes. The Hyperglycemic-Euglycemic Glucose Clamp (HEGC) technique is considered the gold standard for assessing IR, and evaluating insulin sensitivity by measuring the Glucose Disposal Rate (GDR) (10). However, HEGC is invasive, relatively complex, and time-consuming, making it impractical for routine clinical practice (10, 11). To address this limitation, researchers have derived an estimated GDR (eGDR) based on HEGC data, calculated as 21.158 − [0.09 × waist circumference (WC, cm)] − [3.407 × hypertension (yes = 1, no = 0)] − [0.551 × hemoglobin A1c (HbA1c, %)], which can be utilized for assessing insulin sensitivity (12). To date, eGDR as a measure of IR has been validated in numerous related studies across various ethnic groups (13–19). In summary, eGDR can serve as a surrogate marker for IR in assessing various study outcomes in both diabetic and non-diabetic patients. Compared to HEGC, eGDR holds greater potential as a tool for epidemiological studies and clinical practice applications. It should be noted that although a substantial body of research has provided evidence for the use of eGDR as a surrogate for IR in both diabetic and non-diabetic populations, it remains unclear whether eGDR can be utilized for risk assessment of diabetes as a study outcome. Additionally, we lack understanding of the role of eGDR in the transition from hyperglycemia to normoglycemia. Addressing these issues will facilitate a deeper understanding of the associative relationship between eGDR and glucose metabolism, which is crucial for enhancing the application and awareness of eGDR. To address this issue, this study aims to explore the association between eGDR and the reversion or progression of prediabetes using data from the large longitudinal China Health and Retirement Longitudinal Study (CHARLS) cohort in China.

This nationwide prospective study investigated the association between a surrogate marker of insulin resistance (IR)-eGDR-and the reversion or progression of prediabetes among middle-aged and elderly individuals in China. The findings suggest that higher eGDR is positively associated with the reversion of prediabetes to normoglycaemia and negatively associated with the progression to diabetes. Additionally, the study identified a potential threshold effect point in the dose-response curve between eGDR and the reversion of prediabetes to normoglycaemia, with a HR of 1.98 associated with eGDR values less than 6.81.

The natural course of prediabetes includes maintaining the prediabetic state, progressing to diabetes, or reverting to normoglycaemia (28). A significant body of literature has previously reported a close association between poor glycemic metabolism and adverse outcomes (1, 29–31). Therefore, early identification of factors influencing the progression and reversion of diabetes is crucial for disease prevention and prognosis. IR plays a critical role in the progression of diabetes and is also a key factor in the reversion of prediabetes (12, 13, 32, 33). Previous studies have described associations between certain IR surrogates and the reversion of prediabetes, such as the triglyceride glucose index, triglycerides glucose and body mass index, and the metabolic score for IR (34–36). These findings further confirm the important application value of surrogate biomarkers of IR in glucose metabolism disorders. However, it is important to note that most of these previous IR surrogates primarily considered lipid and glucose factors, overlooking other metabolic factors. eGDR is a novel IR assessment indicator that integrates evaluations of obesity, glucose, and blood pressure factors. Previous studies have shown that eGDR has similar accuracy to the HEGC in assessing IR status (14). Recent studies further validate the significant role of eGDR in evaluating disease prognosis and outcomes. However, most eGDR-related studies to date have focused on analyzing complications or severe adverse outcomes in diabetic patients (15–18, 37, 38). Currently, there is limited understanding of the relationship between eGDR and prediabetes, a stage where glucose levels are still modifiable, and the association of eGDR with the reversion of prediabetes to normoglycaemia remains unclear. To address this issue, the current study analyzed data from the Chinese CHARLS cohort, revealing that higher eGDR promotes the reversion of prediabetes to normoglycaemia. After fully adjusting for confounding factors, the probability of reversion to normoglycaemia in the high eGDR group (Q4) was 2.22 times higher than in the low eGDR group (Q1). This finding underscores the important evaluative value of eGDR in the regression of prediabetes and suggests that monitoring eGDR during glucose control or intervention in prediabetes may be significantly beneficial.

The relationship between IR and its surrogates with diabetes onset has been well documented in numerous previous studies (9–11, 39–41). However, the association between the new IR surrogate eGDR and diabetes was unclear. To clarify this relationship, the current study further analyzed the association between eGDR and diabetes onset in prediabetic patients. After a median follow-up of 4 years, 399 participants (15.3%) in our study cohort progressed to diabetes. The association analysis results showed that after controlling for confounding factors, eGDR was significantly negatively associated with the progression of prediabetes to diabetes. Further analysis based on eGDR quartiles revealed that the risk of progressing to diabetes in the high eGDR group (Q4) was 74% lower than in the low eGDR group (Q1). Contrary to the findings of previous studies on IR (9–11, 39–41), where IR is typically positively associated with diabetes, eGDR in this study was negatively associated with diabetes. This finding suggests that high eGDR inhibits the progression of prediabetes. Notably, the negative association between eGDR and disease has also been reported in recent observational studies, including those on vascular-related diseases (17–19, 37), heart failure (42), atherosclerosis (43, 44), and various chronic complications of diabetes (37, 38, 45, 46). Furthermore, high eGDR has been identified as a protective factor against adverse outcomes such as atrial fibrillation recurrence and mortality (15–17, 40, 44, 47). These findings collectively indicate that maintaining low eGDR has detrimental effects on health, and early intervention is recommended for individuals with low eGDR. Although there are no studies yet on the mechanism by which low eGDR leads to the progression of prediabetes to diabetes, the pathophysiology of IR suggests that increased IR during the transition from normoglycaemia to prediabetes leads to β-cell dysfunction; when β-cell function cannot overcome IR, progression from prediabetes to diabetes occurs (32, 33, 48). As eGDR increases, representing a reduction in IR, the progression of prediabetes is suppressed (14). Considering the current study’s context and findings, we recommend dynamic monitoring of eGDR in prediabetic patients and maintaining this value at a high level whenever possible.

In the RCS analysis, we observed an intriguing finding: a nonlinear, L-shaped relationship between eGDR and the reversion of prediabetes to normoglycaemia. Further analysis calculated the eGDR threshold associated with reversion to normoglycaemia as 6.81. Specifically, when the eGDR level was below 6.81, each unit increase in eGDR was associated with a 98% increase in the likelihood of prediabetes reversion (HR: 1.98, 95% CI: 1.38-2.84, P = 0.0002); when the eGDR level was above 6.81, the likelihood of prediabetes reversion significantly decreased and stabilized (HR: 1.10, 95% CI: 1.01-1.19, P = 0.03). Similar L-shaped associations have also been reported in recent studies. In a recent study of non-alcoholic fatty liver disease (NAFLD) patients, Song et al. identified an eGDR threshold of 5.95 as a potential cutoff point for NAFLD patients at risk of developing atherosclerosis. When eGDR was below 5.95, the probability of atherosclerosis accelerated in NAFLD patients (43). Additionally, a study from Switzerland on mortality reported similar results; Nyström et al. found an L-shaped relationship between eGDR and all-cause mortality in patients with type 1 diabetes, with a threshold between 8-10, where the risk of all-cause mortality stabilized when eGDR exceeded the threshold (16). Based on these findings and the current study results, we believe that attention should be paid to evaluating the nonlinear relationship of eGDR in metabolic-related diseases.

“Treating pre-disease” is an important concept in Traditional Chinese Medicine (49, 50), where “pre-disease” refers to a transitional state before the onset of illness. Prediabetes is an early stage of diabetes as well as multiple chronic complications. From both public health and clinical perspectives, identifying and recognizing this intermediate-risk group is essential, as early intervention in the prediabetic phase is most effective in preventing diabetes, cardiovascular diseases, and various chronic complications (3–8). Therefore, timely diagnosis of the prediabetic population and effective management of key modifiable factors are crucial for preventing chronic complications. Against this backdrop, the findings of the current study hold significant implications for clinical practice: (1) As a non-invasive and simple surrogate for IR, eGDR offers notable advantages over the HEGC in clinical practice, substantially reducing both the economic and time costs for patients. (2) According to the findings of the current study, eGDR provides significant assistance in assessing the transition of glucose status; in summary, a low eGDR is detrimental to the restoration of normal glucose levels in prediabetes and may also accelerate the progression to diabetes. Clinicians can utilize eGDR to evaluate the trends and potential for glucose progression and reversal in patients with prediabetes. (3) Although the mechanism underlying the relationship between eGDR and glycemic metabolic transitions remains unclear, analyzing the development background and constituent components of eGDR suggests that preventing hypertension and obesity can be of significant assistance in promoting glycemic control in prediabetes. (4) The findings of the current study can also provide references for the construction of future prediction models for the progression or regression of prediabetes. Incorporating eGDR may further enhance the predictive performance of these models.

Strengths: (1) This study is the first to describe the association between eGDR and the natural course of prediabetes, with novel findings related to both progression to diabetes and reversion to normoglycemia. (2) The study includes a large, nationally representative cohort and validates its findings across different standards and through sensitivity analyses.

Limitations: (1) Although the study controlled for a large number of confounding factors in the analysis, it cannot eliminate the potential bias from unmeasured or residual confounders (51). (2) The data collected in this study partly relied on self-reporting, supported by a review of existing medical records, but did not include systematic adjudication of clinical outcomes, which may introduce some recall bias (52). (3) The results of this study are based on a middle-aged and elderly Chinese population, which may differ in genetic background compared to younger populations and Western adults. In the future, results verification will be required in young people and other races. (4) The study’s broad geographic follow-up led to some loss of follow-up, resulting in missing data. (5) Due to the lack of oral glucose tolerance test data in the CHARLS cohort, in the current study, we primarily relied on the definition of impaired fasting glucose in the American Diabetes Association standards to diagnose prediabetes. This may have resulted in an underestimation of the incidence rates of prediabetes and diabetes, as well as the regression rate of normoglycemia in the study; additionally, the missing data on impaired glucose tolerance prevented further exploration of the differences in the association between eGDR and glycemic metabolism. Future research needs to further differentiate between subtypes of prediabetes to provide more accurate evidence. (6) In the current study, the specific diagnosis time of prediabetes in the study population could not be determined, making it impossible to ascertain the duration of prediabetes in the patients. This may have a certain impact on the assessment of glycemic transition. Future studies are recommended to further analyze the influence of the duration of prediabetes on subsequent glycemic transition. (7) The follow-up period of this study remains relatively short, making it uncertain whether the association between eGDR and glycemic transition in prediabetes remains stable over a longer follow-up duration. It is recommended to conduct further studies with a longer follow-up period and to explore the impact of changes in eGDR during follow-up on glycemic transition. (8) The current study has only established the association between the IR surrogate eGDR and glycemic transition; Considering that multiple non-invasive IR surrogates have already been developed (34–36), it is necessary to conduct comparative studies in the future to further determine which IR surrogate is most suitable for the assessment of glycemic transition.

In this prospective cohort study of middle-aged and elderly Chinese individuals, our results support a close association between higher eGDR and an increased likelihood of prediabetes reversion to normoglycaemia, as well as a decreased likelihood of progression to diabetes.