Obesity-Related Insulin Resistance Indices and CKD Risk in Patients with Diabetes and Coronary Heart Disease: A Multicenter Cohort Analysis

Introduction

Chronic kidney disease (CKD) is a chronic condition characterized by structural or functional abnormalities of the kidney resulting from a persistent and significant decline in renal function.^1,2 Its primary clinical manifestations are impaired glomerular filtration and the presence of proteinuria.² The long-term, progressive deterioration of renal function leads to complications such as fluid and sodium retention and electrolyte imbalances, which in turn precipitate a cascade of disorders affecting the cardiac, skeletal, and hematological systems.^3–6 Notably, cardiovascular mortality is the leading cause of death in patients with end-stage renal disease.^4,5 Diabetes and hypertension are the two principal causes of CKD.^7–9 Patients with diabetes frequently present with concomitant obesity and insulin resistance (IR).^10,11 Crucially, these very factors—diabetes, IR, and obesity—are also major contributors to the pathogenesis of coronary heart disease (CHD).^12,13 Therefore, for patients with both diabetes and CHD, the prevention of CKD is of paramount importance to mitigate disease progression and reduce the overall disease burden.

Traditionally, strategies for preventing CKD in this high-risk population have been predominantly confined to glycemic and blood pressure (BP) control.^8,14 Consequently, the detrimental impact of obesity and IR on renal health has often been underappreciated and inadequately addressed. Prior research has largely framed obesity and IR as underlying causes of glucose and lipid metabolism disorders.^15,16 However, whether they function as independent risk factors for renal injury and the subsequent development of CKD has been a subject of limited investigation.

Emerging evidence is now beginning to illuminate the direct role of obesity and IR in renal pathology.^17–20 For instance, a study from the United Kingdom indicated that obese individuals have a significantly higher incidence of CKD, suggesting a potential benefit of weight loss in reducing CKD risk.¹⁷ Similarly, IR has been implicated as a key driver of CKD, with research demonstrating its association with declining renal function in middle-aged and older adults.¹⁹ It is proposed that IR may mediate the onset of CKD through its interaction with adiposity-related indicators.²¹

Given the complex pathophysiology of obesity and IR, traditional measures like body mass index (BMI) may not accurately reflect an individual’s true metabolic and adiposity status.^22,23 In response, recent studies have increasingly utilized and validated composite indices that better capture visceral obesity and IR. These include the weight-adjusted-waist index (WWI), the triglyceride glucose-body mass index (TyG-BMI), TyG combined with waist circumference (TyG-WC), and the metabolic score for visceral fat (METS-VF).^22,24–26 These novel indices are believed to provide a more precise assessment of an individual’s internal obesity profile and insulin sensitivity and have demonstrated superior diagnostic and predictive value for a variety of diseases.^22,23,26 Notably, studies have confirmed that IR indices are associated with decreased kidney function in the general population and mediate the occurrence of CKD.^19,21 Furthermore, basic research has further revealed that visceral obesity, as represented by METS-VF, may impair renal function through pathways including lipid accumulation, long-term chronic low-grade inflammation, and excessive activation of the renin-angiotensin-aldosterone system (RAAS).^27,28 Collectively, these lines of evidence from both epidemiological and basic research provide an important theoretical basis for our study.

However, within the specific, high-risk population of patients with coexisting diabetes and CHD—who are inherently prone to obesity and IR, which in turn exacerbate poor glycemic control and accelerate vascular damage—the relationship between these factors and CKD risk remains unclear. It is unknown whether obesity and IR in this group are associated with an increased risk of incident CKD.

To address this knowledge gap, this study aims to investigate the association between obesity-related IR indices and the risk of CKD in patients with diabetes and CHD. By doing so, we seek to provide crucial epidemiological evidence to inform future strategies focused on weight management and metabolic improvement for the effective preservation of renal function in this vulnerable population.

Materials and Methods

Study Design and Population

This multicenter cohort study was conducted across three centers, enrolling patients with concomitant diabetes and CHD from the Changzhi People’s Hospital, West Southwest Medical University Affiliated Hospital, and Anyang People’s Hospital. A total of 9395 patients were initially screened. After excluding those with baseline CKD, acute coronary syndrome or coronary revascularization within the preceding three months, missing data for variables required to calculate obesity-related IR indices, malignant tumors, and use of any weight-loss medications, a final cohort of 7744 patients was included in the analysis. The detailed patient selection process is illustrated in Figure 1.

Figure 1 Selection of study population.

This study was conducted in strict accordance with the Declaration of Helsinki. Ethical approval was obtained from the institutional review boards of all three participating hospitals: Changzhi People’s Hospital (Approval No. CZX.20180211), West Southwest Medical University Affiliated Hospital (Approval No. WSXN.20200711), and Anyang People’s Hospital (Approval No. N.20230607). Written informed consent was obtained from all participants prior to study initiation.

Collection of Covariates

Covariate data were systematically collected from multiple sources, including hospital electronic medical records, follow-up records, and medical insurance databases. The collected information encompassed demographic characteristics, physical examination findings, laboratory test results, medical history, and medication use. Demographic information and physical examination included patients’ sex, age, BMI, WC, systolic and diastolic BP, as well as smoking and alcohol consumption status. Laboratory measurements comprised the following: liver function parameters, total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), hemoglobin A1c (HbA1c), and fasting plasma glucose (FPG). Renal function was assessed by estimating the glomerular filtration rate (eGFR), which was calculated using the CKD-EPI equation adapted for the Chinese population.^9,29 Medication history documented the use of various therapeutic agents, including antiplatelet drugs, lipid-lowering medications, beta-blockers, angiotensin-converting enzyme inhibitors (ACEIs)/angiotensin receptor blockers (ARBs), and insulin. Medical history primarily focused on the presence of hypertension and hyperlipidemia.

Calculation of Obesity-Related IR Indices

Obesity-related IR indices included WWI, TyG-BMI, TyG-WC, and METS-VF, and the specific calculation formulas were as follows:^22,24,26

WWI = WC/√BMI.

TyG-BMI = Ln [(TG ×FPG)/2] × BMI.

TyG-WC = Ln [(TG ×FPG)/2] ×WC.

METS-IR = [ln((2 × FPG) + TG) × BMI] / ln(HDL-C).

METS-VF = 4.466 + 0.011 × [ln(METS-IR)]³ + 3.239 × [ln(WHtR)]³ + 0.319 × Sex (Female=0, Male=1) + 0.594 × ln(Age).

Outcomes

The primary endpoint of this study was the incidence of new-onset CKD. The diagnosis of CKD was established in accordance with the latest kidney disease clinical practice guidelines.^30,31 Specifically, CKD was defined as the presence of either of the following criteria persisting for more than three months: (1) a persistently reduced eGFR of less than 60 mL/min/1.73 m², calculated using the CKD-EPI equation; and/or (2) a persistently elevated urinary albumin-to-creatinine ratio of greater than 30 mg/g. To ensure the chronicity required for a definitive CKD diagnosis, at least two consecutive measurements meeting these criteria, obtained at different time points with an interval exceeding three months, were required. Cases of acute kidney injury were excluded from the outcome ascertainment.

Statistical Analysis

Participants were dichotomized based on the occurrence of the endpoint outcome, CKD, and baseline characteristics were compared between the two groups. Multivariable Cox regression analyses with stepwise adjustment were performed to investigate the association between obesity-related IR indices and the risk of CKD in patients with coexisting diabetes and CHD. To examine trend effects, these indices were categorized into four groups according to their quartiles. The Kaplan-Meier (KM) method was utilized to assess the cumulative incidence of CKD across different groups during the follow-up period. Restricted cubic splines (RCS) were plotted to evaluate the dose-response relationship, and a two-piecewise regression analysis was performed using the identified inflection point. Extensive subgroup analyses and sensitivity analyses were conducted to further verify the robustness of the findings. Finally, receiver operating characteristic (ROC) curves, time-dependent ROC curves, decision curve analysis (DCA), and the C-index were employed to assess the predictive ability of various obesity-related IR indices for CKD and their incremental predictive value to the baseline model.

All statistical analyses were performed using R software (version 4.4.1), and a two-tailed P-value < 0.05 was considered statistically significant.

Results

Comparison of Baseline Characteristics Between the Two Groups

A total of 7744 patients with diabetes and coronary heart disease were enrolled from three centers. During a median follow-up of 3.95 years, 1725 patients developed CKD. Table 1 presents the baseline characteristics of participants stratified by CKD occurrence.

Table 1 Baseline Characteristics of CKD Patients and Non-CKD Patients

Compared with those who did not develop CKD, patients who developed CKD were older and had significantly higher BMI, SBP, and DBP. They were more likely to consume alcohol but exhibited a lower smoking rate. Regarding laboratory parameters, the CKD group showed significantly higher levels of liver enzymes, TC, TG, HbA1c, and FPG, whereas HDL-C and eGFR were significantly lower. Additionally, patients who developed CKD had a higher prevalence of hypertension and hyperlipidemia and were more frequently prescribed lipid-lowering drugs, antiplatelet drugs, beta-blockers, ACEIs/ARBs, and insulin.

Impact of Obesity-Related IR Indices on CKD Risk in Patients with Diabetes and CHD

To evaluate the relationship between obesity-related IR indices and CKD, participants were initially stratified into four groups according to the quartiles of these indices, and the incidence of the endpoint outcome—CKD—was compared across the groups. The findings revealed a progressively higher incidence of CKD with increasing levels of the obesity-related IR indices, demonstrating a clear upward trend from the first quartile (Q1) to the fourth quartile (Q4) (Figure 2).

Figure 2 Incidence of CKD across four obesity-related IR indices by group. (A) WWI; (B) TyG-BMI; (C) TyG-WC; (D) METS-VF. Q1: First quartile group; Q2: Second quartile group; Q3: Third quartile group; Q4: Fourth quartile group.

Subsequently, multivariable-adjusted Cox regression analyses were conducted to further explore these associations. In the fully adjusted Model 5, each one-standard-deviation increase in the four obesity-related IR indices—WWI, TyG-BMI, TyG-WC, and METS-VF—was associated with a 32.0% (Hazard Ratio [HR]: 1.320, 95% confidence interval [CI]: 1.264–1.379), 39.4% (HR: 1.394, 95% CI: 1.333–1.457), 37.5% (HR: 1.375, 95% CI: 1.322–1.429), and 44.0% (HR: 1.440, 95% CI: 1.381–1.501) increased risk of CKD, respectively (Table 2).

Table 2 Association Between Obesity-Related IR Indices and CKD Risk in Patients with Diabetes and CHD

When these indices were analyzed as categorical variables based on quartiles, a graded increase in CKD risk was observed across higher quartiles (Q2–Q4) compared with the lowest quartile (Q1), with the highest risk observed in Q4 (Table 2). Specifically, participants in the Q4 groups of the four indices exhibited a 1.897-fold (HR: 1.897, 95% CI: 1.642–2.190), 2.358-fold (HR: 2.358, 95% CI: 2.034–2.735), 2.477-fold (HR: 2.477, 95% CI: 2.113–2.904), and 3.477-fold (HR: 3.477, 95% CI: 2.909–4.157) higher risk of CKD compared with those in the Q1 groups (Table 2). These findings were further corroborated by KM curve analysis, which demonstrated that the Q4 group consistently had the highest cumulative incidence of CKD throughout the follow-up period (Figure 3).

Figure 3 Kaplan-Meier curves for CKD risk across different groups. (A) WWI; (B) TyG-BMI; (C) TyG-WC; (D) METS-VF. Q1: First quartile group; Q2: Second quartile group; Q3: Third quartile group; Q4: Fourth quartile group.

Dose-Response Relationship Between Obesity-Related IR Indices and CKD

Subsequently, based on our findings that elevated levels of all four obesity-related IR indices were associated with an increased risk of CKD, we further evaluated the dose-response relationship by plotting RCS. The results demonstrated that each of the four indices exhibited a progressively increasing nonlinear association with CKD risk in patients with diabetes and CHD, with evidence of threshold effects (Figure 4). Notably, the risk of CKD increased more markedly when WWI, TyG-BMI, TyG-WC, and METS-VF exceeded 11.2, 245, 530, and 6.2, respectively (Figure 4).

Figure 4 Dose-Response relationship between different obesity-related IR indices and CKD risk. (A) WWI; (B) TyG-BMI; (C) TyG-WC; (D) METS-VF. The solid line represents the fitted line, and the dashed line represents the confidence interval.

To further characterize these threshold effects, we conducted a two-piecewise comparative analysis based on these inflection points. The results revealed that individuals with WWI > 11.2, TyG-BMI > 245, TyG-WC > 530, and METS-VF > 6.2 had a 1.525-fold, 1.729-fold, 2.123-fold, and 2.182-fold higher risk of CKD, respectively, compared with those at or below these thresholds (Table 3).

Table 3 Association Between Obesity-Related IR Indices and CKD Risk in Patients with Diabetes and CHD Based on the Turning Point

These findings suggest that for patients with diabetes and CHD, weight management and improvement of IR—thereby maintaining these obesity-related IR indices within an optimal range—may not only reduce cardiovascular events but also lower the future risk of incident CKD.

Subgroup Analysis and Sensitivity Analysis

Considering that variations in disease status and individual characteristics among patients might confound the association between obesity-related IR indices and CKD risk, we conducted subgroup analyses stratified by sex, age, BMI, hypertension, hyperlipidemia, and insulin use. The results consistently demonstrated that elevated levels of obesity-related IR indices were associated with an increased risk of CKD across all subgroups examined (Figure 5).

Figure 5 Association between different obesity-related IR indices (WWI,TyG-BMI, TyG-WC, and METS-VF) and CKD risk across different subgroups.

To address potential reverse causality during follow-up, we performed a sensitivity analysis by excluding individuals with follow-up duration of less than one year; the results remained substantially unchanged (Table S1). Furthermore, to account for the possibility of unmeasured confounding, we conducted E-value analysis, which indicated that the observed associations were robust and unlikely to be explained by unmeasured confounders (Table S2).

Collectively, these analyses consistently reinforce our conclusion that elevated obesity-related IR indices are associated with an increased risk of CKD in patients with diabetes and CHD.

Comparison of Predictive Performance of Obesity-Related IR Indices for CKD

Based on the observed close associations between various obesity-related IR indices and the increased risk of CKD in patients with diabetes and CHD, we further investigated the predictive capacity of these indices. First, we compared the area under the receiver operating characteristic curve (AUC) for each index. The results demonstrated that METS-VF exhibited the largest AUC, reaching 0.7226 (Figure 6 and Table 4).

Table 4 Comparison of the Predictive Performance of Different Obesity-Related IR Indices for the Risk of CKD in Patients with Diabetes and CHD

Figure 6 Comparison of ROC curves for different obesity-related IR indices.

Given the longitudinal nature of this study, we further conducted time-dependent ROC analysis, which consistently showed that METS-VF possessed the highest predictive ability throughout the entire follow-up period (Figure 7). Additionally, DCA revealed that METS-VF provided the greatest clinical net benefit among the four indices (Figure 8).

Figure 7 Comparison of time-dependent ROC curves for different obesity-related IR indices.

Figure 8 Decision Curve Analysis of different obesity-related IR indices.

Finally, to evaluate the incremental predictive value of these indices for CKD risk, we performed C-index analysis by adding WWI, TyG-BMI, TyG-WC, and METS-VF to the fully adjusted Model 5. The C-index increased by 0.013, 0.022, 0.027, and 0.030, respectively, with METS-VF again demonstrating the largest incremental improvement in predictive performance (Table 5).

Table 5 Incremental Predictive Value of Obesity-Related IR Indices for CKD

Collectively, these findings consistently indicate that METS-VF, as a representative obesity-related IR index, possesses the greatest predictive capacity for CKD risk in patients with diabetes and CHD.

Discussion

Traditionally, obesity and IR have been recognized as major contributors to diabetes and metabolic disorders, which in turn exacerbate vascular injury and dyslipidemia, leading to vascular occlusion and consequently increasing the risk of CHD.^12,13 Emerging evidence has further corroborated that obesity and IR also play significant roles in renal impairment.^32,33 Therefore, patients with coexisting diabetes and CHD, as a high-risk population, warrant particular attention.

Conventionally, BMI has been widely used to assess obesity; however, this metric is relatively simplistic and may not accurately reflect actual body fat distribution or insulin resistance status.^33,34 Consequently, more comprehensive composite indices incorporating both obesity and IR—such as WWI, TyG-BMI, TyG-WC, and METS-VF—have been progressively developed and have demonstrated superior predictive capability across various disease conditions.^22,24,26 Against this background, the present study employed a multicenter cohort design to investigate the associations between these composite obesity-related IR indices and the future risk of CKD in patients with coexisting diabetes and CHD. Our findings confirmed that these metabolic indices were strongly associated with an increased risk of CKD, with evidence of threshold effects. Specifically, when WWI, TyG-BMI, TyG-WC, and METS-VF exceeded 11.2, 245, 530, and 6.2, respectively, the risk became more pronounced. These results underscore the importance of weight management and improvement of IR to maintain these indices within an optimal range, thereby mitigating future CKD risk. Further comparative analyses revealed that METS-VF, which reflects visceral adiposity, demonstrated the strongest predictive capacity, suggesting its superior ability to capture visceral obesity and IR status. Our study not only extends the research scope to a high-risk population for CKD—namely, patients with diabetes and CHD—thereby breaking through the limitations of traditional studies, but also identifies a critical threshold effect. This finding has significant clinical implications: as a simple and easily obtainable indicator, METS-VF can serve as an effective tool for CKD risk assessment in patients with diabetes and CHD. Its clinical application may help promptly identify high-risk individuals and highlight the importance of reducing obesity (especially visceral fat accumulation) and improving insulin resistance in this susceptible population, thereby enabling early risk evaluation and stratified management of CKD.

These composite obesity-related IR indices have demonstrated excellent performance across a wide spectrum of diseases, including kidney disorders.^{20,22,24,26,35} A nationwide cohort study in China revealed that TyG-related obesity-IR indices were associated with renal function decline and progression of CKD in middle-aged and older adults, with TyG-WC identified as the optimal predictor of renal function deterioration.²² Similarly, a trajectory analysis demonstrated that individuals with persistently high trajectories of IR exhibited a significantly elevated risk of CKD.³⁶ In the context of cardiovascular disease (CVD), these indices have also shown notable predictive value.^22,24,25 Research has indicated that obesity assessed by WWI is associated with an increased risk of cardiovascular events in patients with hypertension.²² Compared with other indicators, METS-VF has several distinct advantages. First, in terms of indicator composition, METS-VF integrates more comprehensive metabolic detection indicators and obesity indicators, thereby capturing the overall metabolic disorders and obesity status more holistically. Second, this indicator also incorporates sex and age, enabling it to more effectively reflect and distinguish individual differences. Based on these compositional advantages, METS-VF demonstrates higher accuracy and stronger predictive capability. As a more precise indicator of visceral obesity, METS-VF has been shown to be closely associated with a wide range of conditions, including cardiovascular disease and related mortality, renal impairment, bone health, and hyperuricemia. Moreover, it serves as an excellent predictive marker for these clinical outcomes, highlighting its broad clinical applicability.^24,35,37,38

Collectively, these previous investigations have substantiated the significant value of these composite indices and provide important support and rationale for our exploration of the relationship between these indices and future CKD risk in the high-risk population of patients with concomitant diabetes and CHD.

Obesity-related IR drives CKD progression in patients with diabetes and CHD through several interconnected pathophysiological mechanisms. First, ectopic fat deposition leads to lipotoxicity.^27,39,40 As adipose tissue storage capacity is exceeded, triglycerides accumulate in renal sinuses and perirenal fat, causing mechanical vascular compression.^39,41 Concurrently, lipid accumulation within renal cells directly damages tubular cells and podocytes by inducing endoplasmic reticulum stress and mitochondrial dysfunction.^40,42 Second, microcirculatory dysfunction develops.^39,43 IR impairs the PI3K/AKT vasodilatory pathway while simultaneously enhancing MAPK/ET-1-mediated vasoconstriction, resulting in increased endothelin-1 production.^39,43 This imbalance causes sustained renal vasoconstriction, reduced blood flow, and subsequent microcirculatory deficits.^43,44 Third, RAAS overactivation occurs. Hyperinsulinemia upregulates angiotensin II receptors in the kidneys, exacerbating intraglomerular hypertension and proteinuria.^28,45 Fourth, chronic low-grade inflammation emerges.^28,46 Dysfunctional adipose tissue recruits macrophages that secrete pro-inflammatory cytokines such as TNF-α and IL-6, directly damaging renal tissue and amplifying local inflammatory responses.^46–48 Fifth, oxidative stress ensues.^40,49,50 Energy excess promotes overproduction of reactive oxygen species, triggering oxidative stress that particularly damages renal mitochondria, thereby accelerating nephron loss and fibrotic progression.^40,48,51 Collectively, these interconnected mechanisms synergistically drive CKD progression.

The primary strength of this study lies in its comprehensive statistical analyses, which systematically investigated the relationship between obesity-related IR indices and future CKD risk in a high-risk population of patients with diabetes and CHD—a group often characterized by concomitant obesity and IR. These findings extend the current understanding of CKD prevention and management strategies in this vulnerable population. Additionally, we identified METS-VF as the optimal predictive marker for CKD risk, reflecting its superior ability to capture visceral obesity and IR, which may have significant implications for future CKD screening initiatives.

However, several limitations should be acknowledged. First, our study assessed obesity-related IR indices only at baseline, thereby failing to capture their dynamic changes over time. Second, the lack of data on weight-loss medications during follow-up may have introduced potential confounding, and future studies should incorporate this information. Third, despite the longitudinal design, the observational nature of this study precludes causal inferences; multicenter randomized controlled trials focusing on weight reduction are warranted to validate these findings. Fourth, as our study population was exclusively derived from Chinese centers, the generalizability of our results to other regions and ethnicities requires further investigation. Furthermore, although METS-VF demonstrated better performance compared to other indicators, its AUC was relatively low. It may be more suitable as a risk assessment tool rather than a standalone test. Finally, although we adjusted for numerous confounding factors, the possibility of residual unmeasured confounding cannot be entirely ruled out. Nevertheless, E-value analysis suggested that such unmeasured confounders are unlikely to fully explain the observed associations.

Conclusion

In patients with coexisting diabetes and CHD, elevated obesity-related IR indices are closely associated with an increased risk of CKD, with threshold effects observed. These findings suggest that weight management and improvement of insulin control are important for reducing CKD risk in this population. Notably, METS-VF (threshold >6.2) was identified as the optimal predictive marker, which may serve as a simple and accurate tool for CKD risk assessment. However, given the observational design of this study, these findings suggest that METS-VF may help guide clinical interventions, but further validation is needed. Moreover, as these results are derived from a Chinese multicenter cohort, external validation in other populations is warranted to establish generalizability.