Xin Y, Peng NL, Xin CY, Liao JR, Hu XQ, Dong YH, Zhang XY. Metabolic score for insulin resistance is associated with adverse cardiovascular events in patients with type 2 diabetes. World J Diabetes 2025; 16(8): 108671 [DOI: 10.4239/wjd.v16.i8.108671]
Corresponding Author of This Article
Xiang-Yu Zhang, Department of Geriatrics, The Second Xiangya Hospital of Central South University, No. 139 Renmin Middle Road, Changsha 410011, Hunan Province, China. xiangyuzhang@csu.edu.cn
Research Domain of This Article
Endocrinology & Metabolism
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Ying Xin, Xin-Qun Hu, Department of Cardiovascular Medicine, The Second Xiangya Hospital, Central South University, Changsha 410011, Hunan Province, China
Ying Xin, Na-Ling Peng, Jiang-Rong Liao, Yi-Heng Dong, Xiang-Yu Zhang, Department of Geriatrics, The Second Xiangya Hospital of Central South University, Changsha 410011, Hunan Province, China
Cai-Yan Xin, Fuzhou Medical College, Nanchang University, Fuzhou 344000, Jiangxi Province, China
Co-corresponding authors: Yi-Heng Dong and Xiang-Yu Zhang.
Author contributions: Xin Y, Peng NL, Xin CY, and Dong YH analyzed the data; Xin Y and Peng NL wrote the paper, they contributed equally to this article, they are the co-first authors of this manuscript; Xin CY, Liao JR, and Dong YH edited the paper; Hu XQ, Dong YH, and Zhang XY defined the study theme and methods; Dong YH and Zhang XY contributed equally to this article, they are the co-corresponding authors of this manuscript; and all authors read and approved the final manuscript.
Supported by the Key Research and Development Plan of Hunan Province, No. 2022SK2013; and Central South University, No. 2024ZZTS0931.
Institutional review board statement: The study was exempt from ethical review and approval, as no additional institutional review board approval was necessary for the secondary analysis.
Informed consent statement: The data for our study are derived from the Action to Control Cardiovascular Risk in Diabetes. All participants completed informed consent forms before participating in the study.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Data sharing statement: The datasets used and analyzed during the current study are available from the Action to Control Cardiovascular Risk in Diabetes (ACCORD)/ ACCORDION Research Materials obtained from the National Heart, Lung, and Blood Institute Biologic Specimen and Data Repository Information Coordinating Center. The contents of this report do not necessarily reflect the opinions or views of the ACCORD/ ACCORDION authors or the National Heart, Lung, and Blood Institute.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Xiang-Yu Zhang, Department of Geriatrics, The Second Xiangya Hospital of Central South University, No. 139 Renmin Middle Road, Changsha 410011, Hunan Province, China. xiangyuzhang@csu.edu.cn
Received: April 21, 2025 Revised: May 22, 2025 Accepted: July 7, 2025 Published online: August 15, 2025 Processing time: 116 Days and 17.6 Hours
Abstract
BACKGROUND
Cardiovascular disease represents a major complication in patients with type 2 diabetes mellitus (T2DM), with insulin resistance (IR) recognized as a key underlying pathophysiological mechanism. The metabolic score for IR (METS-IR), a simple, non-invasive, and insulin-independent surrogate marker of IR, has been validated for risk stratification and prognostic assessment in conditions such as hypertension, ischemic cardiomyopathy, and T2DM. Monitoring fluctuations in METS-IR levels among individuals with T2DM may facilitate early identification of elevated cardiovascular risk and inform timely therapeutic adjustments.
AIM
To investigate the association between METS-IR and cardiovascular risk in patients with T2DM and to evaluate its potential utility as a predictive biomarker.
METHODS
This study represents a secondary analysis of a multicenter randomized controlled trial, ultimately including 10191 patients with T2DM aged 40 years to 79 years, with a follow-up duration of approximately 10 years. Baseline METS-IR was calculated using triglycerides, body mass index, high-density lipoprotein cholesterol and fasting plasma glucose. The predictive value of METS-IR for major adverse cardiovascular events (MACEs), all-cause mortality, congestive heart failure, and major coronary heart disease events, was assessed using Cox proportional hazards models, restricted cubic spline analysis, and stratified subgroup analyses. Multivariable adjustments were performed to account for potential confounding factors.
RESULTS
The incidence of MACEs increased steadily across higher METS-IR quartiles. After adjusting for multiple confounding factors, hazard ratios comparing the highest to the lowest METS-IR quartile were 1.25 [95% confidence interval (CI): 1.08-1.45] for MACEs, 1.55 (95%CI: 1.23-1.96) for cardiovascular death, 1.39 (95%CI: 1.21-1.59) for all-cause mortality, 2.22 (95%CI: 1.74-2.82) for congestive heart failure, and 1.35 (95%CI: 1.17-1.56) for major coronary heart disease. Restricted cubic spline analysis supported a positive, dose-dependent relationship between rising METS-IR levels and cardiovascular risk. Moreover, adding METS-IR to conventional risk prediction models enhanced their performance, as evidenced by improvements in the C-statistic, net reclassification improvement, and integrated discrimination improvement. Subgroup analyses indicated possible interactions between METS-IR, hemoglobin A1c levels, and aspirin therapy.
CONCLUSION
METS-IR shows a strong correlation with cardiovascular risk in individuals with T2DM. Tracking METS-IR levels could enhance risk assessment and the prediction of cardiovascular events.
Core Tip: This research demonstrates that metabolic score for insulin resistance is significantly associated with adverse cardiovascular events in patients with type 2 diabetes mellitus. As a comprehensive metabolic indicator, metabolic score for insulin resistance may offer valuable insights for cardiovascular risk assessment, stratification, and outcome prediction - particularly in high-risk populations where traditional methods may be insufficient.
Citation: Xin Y, Peng NL, Xin CY, Liao JR, Hu XQ, Dong YH, Zhang XY. Metabolic score for insulin resistance is associated with adverse cardiovascular events in patients with type 2 diabetes. World J Diabetes 2025; 16(8): 108671
In recent years, the worldwide prevalence of diabetes mellitus has been rising consistently[1]. The International Diabetes Federation estimated that in 2021, diabetes affected about 10.5% of people aged 20 to 79, totaling roughly 536.6 million individuals worldwide. This figure is expected to increase to 783.2 million by 2045[2]. In parallel with the increasing diabetes prevalence, worldwide healthcare costs associated with the disease amounted to 966 billion dollars in 2021 and are projected to rise to 1054 billion dollars by 2045[2]. Type 2 diabetes mellitus (T2DM) represents the majority of diabetes cases and is responsible for around 1.5 million deaths each year[3,4]. Cardiovascular disease (CVD) is estimated to be the cause of death in half of all patients with T2DM[4,5]. In addition, adults with T2DM have a two- to fourfold increased risk of developing CVD compared to those without the condition[6]. Consequently, it is crucial to establish reliable strategies for the early detection and prediction of CVD linked to T2DM, aiming to reduce the risk of complications in affected individuals.
Bello-Chavolla et al[7] introduced the metabolic score for insulin resistance (METS-IR) as a new metric that integrates multiple metabolic markers - including fasting plasma glucose, triglycerides (TG), and high-density lipoprotein cholesterol (HDL-C) - to assess the extent of insulin resistance (IR)[7,8]. IR describes the impaired ability of insulin-responsive tissues - mainly muscle, adipose tissue, and liver - to regulate glucose metabolism, resulting in decreased insulin sensitivity and compromised glucose uptake and utilization[9-12]. Evidence highlights IR as a key pathophysiological link connecting the development of T2DM and CVD[12-14]. Individuals with T2DM frequently present with widespread metabolic dysfunction and IR, which markedly elevates their likelihood of atherosclerotic CVD and related death[15].
Hyperinsulinemic-euglycemic clamp test remains the gold standard for evaluating IR. However, METS-IR provides a simpler and more practical alternative, making it well-suited for routine clinical use[11,16]. Elevated METS-IR levels have been independently correlated with an increased risk of hypertension in adult cohorts, as demonstrated by multiple studies[17,18]. Furthermore, METS-IR has emerged as an independent prognostic marker for disease severity and clinical outcomes in idiopathic pulmonary arterial hypertension, offering significant additive value to the European Society of Cardiology risk stratification model in predicting long-term prognosis[19]. METS-IR may serve as a robust biomarker for risk stratification and prognostic evaluation in patients with ischemic cardiomyopathy and T2DM[20]. Besides, METS-IR has consistently exhibited strong predictive capability for IR and associated metabolic risk factors across numerous investigations[21-23]. However, its effectiveness in forecasting cardiovascular outcomes among patients with T2DM requires additional investigation[24]. In the current investigation, we hypothesize that elevated METS-IR levels correlate with a heightened risk of major adverse cardiovascular events (MACEs), all-cause mortality, and additional cardiovascular complications in individuals with T2DM. Moreover, METS-IR may function as a significant biomarker for risk stratification and prognostic evaluation in T2DM patients with ischemic cardiomyopathy.
Given the increasing prevalence of T2DM and its associated CVD burden, exploring the predictive value of METS-IR is of significant clinical importance. Consequently, the present study was designed to examine the association between METS-IR and the incidence of adverse cardiovascular events among T2DM patients, utilizing data from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) clinical trial and the ACCORD Follow-on Study (ACCORDION), thereby providing a basis for the development of more effective risk stratification tools and therapeutic strategies.
MATERIALS AND METHODS
Study population
The post-hoc analysis leverages data from the ACCORD/ACCORDION trials (ClinicalTrials.gov identifier: NCT00000620). The ACCORD trial’s design, inclusion criteria, and primary cardiovascular endpoints were previously defined and published in the original study protocol[25-27]. In summary, the ACCORD trial was a multicenter, randomized controlled study utilizing a 2 × 2 factorial design to assess the impact of intensive glycemic, antihypertensive, and lipid-lowering interventions on cardiovascular outcomes in high-risk individuals with established T2DM. A total of 10251 participants aged 40-79 years were enrolled, with a follow-up duration of approximately 10 years.
Data collection and outcomes
The dataset comprises detailed demographic and clinical parameters, including age, sex, race, educational attainment, lifestyle behaviors (smoking and alcohol consumption), comorbidities, anthropometric indices, laboratory biomarkers [e.g., fasting plasma glucose, TG, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), HDL-C], and pharmacological treatment profiles. Among the 10251 individuals diagnosed with T2DM, 60 participants were excluded due to missing data on key METS-IR components (fasting glucose, TG, or HDL-C). Some references suggest that a 5% missing data rate is the lower threshold, below which the benefits of multiple imputation can be negligible[28-30]. In this study, the missing data rate was 0.585% (60/10251). Therefore, we did not perform multiple imputation or sensitivity analysis but instead conducted a complete case analysis. After rigorous exclusion criteria and quality control procedures, 10191 participants with complete data were included in the final analytic cohort (Figure 1).
Figure 1 Flowchart of this study.
ACCORD: Action to Control Cardiovascular Risk in Diabetes.
Primary outcome of interest was the occurrence of MACEs, defined as a composite of cardiovascular mortality, non-fatal myocardial infarction (MI), and non-fatal stroke. Secondary endpoints encompassed an expanded macrovascular category, including total mortality, major coronary heart disease (CHD) events (comprising CVD death, non-fatal MI, and unstable angina), and incidence of congestive heart failure (CHF).
Definitions
METS-IR was defined as METS-IR = Ln [2 × fasting plasma glucose (mg/dL) + TG (mg/dL)] × body mass index (BMI)/Ln HDL-C (mg/dL)[7].
Statistical analysis
Statistical analyses were conducted using SPSS version 26.0 (IBM Corp., Armonk, NY, United States), R software (R Foundation for Statistical Computing, Vienna, Austria), and EmpowerStats (X&Y Solutions, Inc., Boston, MA, United States). Baseline characteristics were summarized based on distribution type: Continuous variables as mean ± SD or median with interquartile range, and categorical variables as frequencies and percentages. Group comparisons were performed using analysis of variance or Kruskal-Wallis tests for continuous variables, and Pearson’s χ2 tests for categorical variables, as appropriate.
Participants were stratified into METS-IR quartiles with cutoffs: Q1 (24.11-46.69), Q2 (46.70-53.93), Q3 (53.94-62.03), and Q4 (62.04-134.16). Cox proportional hazards models were used to estimate hazard ratios (HRs) for cardiovascular outcomes across METS-IR quartiles and per 1-SD increase. The proportional hazards assumption was verified through visual inspection of Schoenfeld residuals, confirming model validity. Univariate Cox analyses were initially performed to assess the association between each covariate and MACEs; variables with P < 0.10 were subsequently included in multivariable models. Clinically relevant variables (e.g., age, sex, smoking) were also retained regardless of univariate significance to avoid missing important conventional cardiovascular risk factors (e.g., age, sex, smoking). Three hierarchical multivariable models were developed: Model 1 adjusted for age and sex; model 2 further incorporated race, history of CVD, hypertension, heart failure, proteinuria, diabetes duration, depression, educational attainment, smoking status, systolic blood pressure (SBP), diastolic blood pressure (DBP), hemoglobin A1c (HbA1c), TC, LDL-C, estimated glomerular filtration rate, and alanine aminotransferase; model 3 extended adjustments to include medications (calcium channel blockers, beta-blockers, biguanides, thiazolidinediones, insulins, aspirin, statins, and cholesterol absorption inhibitors).
Kaplan-Meier survival curves stratified by METS-IR quartiles were generated to illustrate event-free survival, with log-rank tests used to compare survival distributions. Nonlinear associations between METS-IR and clinical outcomes were examined using multivariable-adjusted restricted cubic spline (RCS) models with four knots placed at the 5th, 35th, 65th, and 90th percentiles. Nonlinearity was formally evaluated using likelihood ratio tests, with P for non-linear < 0.05 considered statistically significant. Effect modification was assessed through stratified analyses and inclusion of multiplicative interaction terms in the Cox models, evaluating heterogeneity by sex, age (< 60 vs ≥ 60 years), race, history of CVD, hypertension, glycemic control (HbA1c < 8.0% vs ≥ 8.0%), and aspirin use. Interaction P values < 0.05 were interpreted as indicative of effect heterogeneity, prompting subgroup-specific interpretation.
To assess the incremental prognostic value of METS-IR over established cardiovascular risk factors (e.g., age, sex, race, CVD history, diabetes duration, SBP, DBP, TC, LDL-C, smoking status, statin, and aspirin use), model discrimination and reclassification were evaluated using the C-statistic (area under the receiver operating characteristic curve), net reclassification improvement (NRI), and integrated discrimination improvement (IDI). NRI quantified the improvement in risk classification accuracy, while IDI captured the enhancement in overall discrimination between models.
RESULTS
Baseline characteristics stratified by METS-IR quartiles
This analysis included 10191 participants at baseline, with 61.47% being male and a mean age of 62.82 ± 6.65 years. The mean METS-IR value was 54.77 ± 11.27. Participants were stratified into quartiles based on baseline METS-IR levels (Table 1). Higher quartiles were associated with a demographic profile marked by younger age, male predominance, and a greater proportion of white individuals. These groups exhibited elevated levels of BMI, DBP, heart rate, fasting glucose, HbA1c, TC, TG, and alanine aminotransferase levels, alongside reduced SBP, LDL-c, and HDL-C levels. Moreover, individuals in the upper METS-IR quartiles reported lower alcohol consumption but higher tobacco use. A greater prevalence of prior CVD, hypertension, proteinuria, depression, and heart failure was observed in these groups. Pharmacologic data revealed higher usage rates of angiotensin converting enzyme inhibitors/angiotensin receptor blocker, beta-blockers, biguanides, thiazolidinediones, insulins and aspirin, whereas statin utilization was comparatively lower.
Table 1 Baseline characteristics of participants by quartiles of metabolic score for insulin resistance, n (%).
Outcome
Q1 (n = 2548)
Q2 (n = 2547)
Q3 (n = 2548)
Q4 (n = 2548)
P value
METS-IR, mean ± SD
41.19 ± 4.23
50.39 ± 2.09
57.78 ± 2.33
69.71 ± 6.66
< 0.001
Age (years), mean ± SD
64.26 ± 7.12
63.39 ± 6.64
62.47 ± 6.37
61.14 ± 5.98
< 0.001
Sex
0.002
Male
1511 (59.30)
1640 (64.39)
1570 (61.62)
1544 (60.60)
Female
1037 (40.70)
907 (35.61)
978 (38.38)
1004 (39.40)
Race
< 0.001
White
1152 (45.21)
1512 (59.36)
1739 (68.25)
1965 (77.12)
Non-white
1396 (54.79)
1035 (40.64)
809 (31.75)
583 (22.88)
Education
< 0.001
Less than high school graduate
434 (17.05)
413 (16.22)
373 (14.64)
281 (11.04)
High school grad
671 (26.37)
667 (26.19)
673 (26.42)
679 (26.68)
Some college or technical school
690 (27.11)
814 (31.96)
869 (34.12)
968 (38.04)
College graduate or more
750 (29.47)
653 (25.64)
632 (24.81)
617 (24.24)
CVD history
825 (32.38)
898 (35.26)
927 (36.38)
933 (36.62)
0.005
Duration of diabetes (years), mean ± SD
12.05 ± 8.14
11.06 ± 7.76
10.26 ± 7.17
9.83 ± 7.07
< 0.001
Previous hypertension
1861 (73.04)
1901 (74.64)
1925 (75.55)
1993 (78.22)
0.008
Proteinuria
429 (16.84)
483 (18.96)
505 (19.83)
610 (23.94)
< 0.001
Heart failure
81 (3.18)
96 (3.77)
126 (4.95)
184 (7.22)
< 0.001
Depression
425 (16.68)
535 (21.01)
628 (24.66)
824 (32.35)
< 0.001
Living alone
2043 (80.18)
2034 (79.89)
2031 (79.71)
2018 (79.23)
0.861
Smoking
< 0.001
Yes
1325 (52.00)
1513 (59.40)
1549 (60.79)
1547 (60.71)
No
1223 (48.00)
1034 (40.60)
999 (39.21)
1001 (39.29)
Alcohol
< 0.001
Yes
659 (25.86)
661 (25.96)
588 (23.10)
524 (20.58)
No
1889 (74.14)
1885 (74.04)
1958 (76.90)
2022 (79.42)
BMI (kg/m2), mean ± SD
26.36 ± 2.76
30.38 ± 2.67
33.76 ± 3.16
38.41 ± 3.68
< 0.001
SBP (mmHg), mean ± SD
137.20 ± 16.95
136.49 ± 17.25
135.99 ± 17.17
135.73 ± 17.03
0.013
DBP (mmHg), mean ± SD
73.20 ± 10.56
74.48 ± 10.58
75.49 ± 10.52
76.38 ± 10.74
< 0.001
Heart rate (bpm), mean ± SD
71.65 ± 11.67
71.78 ± 11.58
73.27 ± 11.66
73.99 ± 11.94
< 0.001
FPG (mg/dL), mean ± SD
157.90 ± 54.45
170.56 ± 52.95
178.76 ± 53.52
193.51 ± 57.66
< 0.001
HbA1C (%), mean ± SD
8.23 ± 1.08
8.23 ± 1.02
8.30 ± 1.02
8.45 ± 1.09
< 0.001
TC (mg/dL), mean ± SD
181.47 ± 39.36
181.10 ± 39.96
183.22 ± 42.08
187.40 ± 45.44
< 0.001
TG (mg/dL), mean ± SD
125.87 ± 72.81
166.08 ± 92.14
201.16 ± 130.08
267.11 ± 215.59
< 0.001
LDL-C (mg/dL), mean ± SD
106.91 ± 33.05
105.91 ± 33.56
104.30 ± 33.92
102.47 ± 34.94
< 0.001
HDL-C (mg/dL), mean ± SD
49.53 ± 13.12
42.51 ± 10.23
39.88 ± 9.46
35.56 ± 8.46
< 0.001
eGFR (mL/minute/1.73 m2), mean ± SD
91.53 ± 30.19
90.70 ± 27.47
90.89 ± 24.57
91.05 ± 26.00
0.727
ALT (mg/dL), mean ± SD
25.12 ± 13.13
27.47 ± 19.14
28.37 ± 15.98
29.38 ± 15.63
< 0.001
Medications
ARB/ACEI
1665 (65.35)
1790 (70.28)
1758 (69.00)
1850 (72.61)
< 0.001
CCB
484 (19.00)
468 (18.37)
484 (19.00)
516 (20.25)
0.382
Bate blockers
604 (23.75)
720 (28.36)
824 (32.44)
917 (36.07)
< 0.001
Biguanides
1563 (61.34)
1646 (64.63)
1686 (66.20)
1623 (63.70)
0.003
Thiazolidinediones
545 (21.39)
507 (19.91)
590 (23.16)
604 (23.70)
0.004
Sulfonylureas
1399 (54.91)
1354 (53.16)
1353 (53.12)
1339 (52.55)
0.363
Insulins
793 (31.12)
882 (34.63)
888 (34.85)
999 (39.21)
< 0.001
Statins
1631 (64.31)
1638 (64.67)
1657 (65.21)
1539 (60.54)
0.002
Aspirin
1326 (52.31)
1381 (54.54)
1440 (56.76)
1402 (55.20)
0.015
Cholesterol absorption inhibitors
51 (2.01)
53 (2.09)
49 (1.93)
54 (2.13)
0.962
MACEs
402 (15.78)
418 (16.41)
482 (18.92)
510 (20.02)
< 0.001
CVD mortality
147 (5.77)
124 (4.87)
176 (6.91)
216 (8.48)
< 0.001
Non-fatal MI
203 (7.97)
228 (8.95)
241 (9.46)
258 (10.13)
0.054
Non-fatal stroke
103 (4.04)
125 (4.91)
130 (5.10)
126 (4.95)
0.276
Total mortality
448 (17.58)
448 (17.59)
491 (19.27)
556 (21.82)
< 0.001
Congestive heart failure
124 (4.87)
132 (5.18)
196 (7.69)
239 (9.38)
< 0.001
Major CHD
379 (14.87)
426 (16.73)
495 (19.43)
544 (21.35)
< 0.001
Relationship between METS-IR and cardiovascular outcomes in patients
To investigate the prognostic significance of METS-IR in individuals with T2DM, we comprehensively assessed its association with major cardiovascular endpoints. Participants stratified into higher METS-IR quartiles exhibited a stepwise increase in the incidence of MACEs, all-cause mortality, CHF, and major CHD (Table 1). Over a median follow-up duration of 8.82 years, 1812 participants (17.78%) experienced MACEs, including 663 (6.51%) cardiovascular deaths, 930 (9.13%) non-fatal MI, and 484 (4.75%) non-fatal strokes. Additional outcomes included 1,943 deaths from all causes (19.07%), 691 CHF cases (6.78%), and 1844 major CHD events (18.09%). The incidence of MACEs progressively increased across METS-IR quartiles, reaching 20.02% in the highest quartile subgroup (Table 1). Similar trends were observed for cardiovascular mortality, total mortality, CHF, and major CHD, whereas no statistically significant associations were identified between METS-IR and non-fatal MI or stroke.
Kaplan-Meier survival analysis revealed significantly higher cumulative event rates among individuals in the highest METS-IR quartile compared to those in the lowest (Figure 2). In multivariable-adjusted Cox proportional hazards models (model 3), where METS-IR was treated as a continuous variable, each SD increment was associated with a 10% increase in MACEs risk (HR = 1.10, 95%CI: 1.04-1.16, P < 0.001; Table 2). When analyzed by quartiles, individuals in the top METS-IR quartile had the greatest MACEs risk (HR = 1.25, 95%CI: 1.08-1.45, P < 0.001, Table 2). Comparable associations were observed for cardiovascular death, all-cause mortality, CHF, and major CHD (Table 2), whereas non-fatal MI and stroke did not demonstrate significant relationships with METS-IR. Of note, in fully adjusted models, METS-IR quartiles Q3 and Q4 were independently associated with elevated CHF risk relative to Q1, with adjusted HRs of 1.76 (95%CI: 1.39-2.23) and 2.22 (95%CI: 1.74-2.82), respectively (both P < 0.001, Table 2 and Figure 2). Additionally, RCS analysis demonstrated a non-linear association between METS-IR and both cardiovascular and all-cause mortality (P for non-linearity < 0.05; Figure 3), suggesting a complex, dose-dependent relationship.
Figure 3 Multivariable-adjusted hazard ratios for primary and secondary outcomes based on restricted cubic spline analysis.
A: Major adverse cardiovascular events; B: Cardiovascular disease mortality; C: Non-fatal myocardial infarction; D: Non-fatal stroke; E: Total mortality; F: Congestive heart failure; G: Major coronary heart disease. MACEs: Major adverse cardiovascular events; CVD: Cardiovascular disease; MI: Myocardial infarction; CHF: Congestive heart failure; CHD: Coronary heart disease; METS-IR: Metabolic score for insulin resistance; HR: Hazard ratio; CI: Confidence interval.
Table 2 Risk of primary and secondary outcomes based on metabolic score for insulin resistance in type 2 diabetes patient.
Outcome
Events/n
Non-adjusted
Model 1
Model 2
Model 3
HR (95%CI)
P value
HR (95%CI)
P value
HR (95%CI)
P value
HR (95%CI)
P value
MACEs
METS-IR
-
1.01 (1.01-1.01)
< 0.00001
1.01 (1.01-1.02)
< 0.00001
1.01 (1.01-1.01)
< 0.00001
1.01 (1.00-1.01)
0.0004
Q1
402/2548
Reference
-
Reference
-
Reference
-
Reference
-
Q2
418/2547
1.05 (0.92-1.20)
0.4837
1.07 (0.94-1.23)
0.3044
1.02 (0.89-1.17)
0.7661
1.00 (0.87-1.15)
0.9647
Q3
482/2548
1.22 (1.07-1.39)
0.0036
1.31 (1.14-1.49)
< 0.0001
1.23 (1.07-1.41)
0.0035
1.20 (1.05-1.38)
0.0090
Q4
510/2548
1.32 (1.15-1.50)
< 0.0001
1.50 (1.31-1.71)
< 0.00001
1.31 (1.14-1.52)
0.0002
1.25 (1.08-1.45)
0.0024
Per 1 SD
-
1.12 (1.07-1.17)
< 0.00001
1.18 (1.12-1.23)
< 0.00001
1.12 (1.06-1.18)
< 0.00001
1.10 (1.04-1.16)
0.0004
P for trend
-
-
< 0.00001
-
< 0.00001
-
< 0.00001
-
0.0003
CVD mortality
METS-IR
-
1.02 (1.01-1.02)
< 0.0001
1.02 (1.02-1.03)
< 0.00001
1.02 (1.01-1.03)
< 0.0001
1.02 (1.01-1.03)
< 0.0001
Q1
147/2548
Reference
-
Reference
-
Reference
-
Reference
Q2
124/2547
0.85 (0.67-1.08)
0.1790
0.88 (0.70-1.12)
0.3102
0.83 (0.65-1.06)
0.1382
0.80 (0.63-1.02)
0.0778
Q3
176/2548
1.22 (0.98-1.52)
0.0764
1.38 (1.10-1.72)
0.0045
1.28 (1.02-1.60)
0.0361
1.23 (0.98-1.54)
0.0808
Q4
216/2548
1.54 (1.25-1.90)
< 0.0001
1.92 (1.55-2.38)
< 0.0001
1.66 (1.32-2.09)
< 0.0001
1.55 (1.23-1.96)
0.0003
Per 1 SD
-
1.20 (1.12-1.29)
< 0.0001
1.32 (1.22-1.42)
< 0.00001
1.25 (1.15-1.36)
< 0.0001
1.22 (1.12-1.32)
< 0.0001
P for trend
-
-
< 0.0001
-
< 0.0001
-
< 0.0001
-
< 0.0001
Non-fatal MI
METS-IR
-
1.01 (1.00-1.01)
0.0020
1.01 (1.01-1.02)
< 0.00001
1.01 (1.00-1.01)
0.0466
1.00 (1.00-1.01)
0.1933
Q1
203/2548
Reference
-
Reference
-
Reference
-
Reference
-
Q2
228/2547
1.13 (0.94-1.37)
0.1978
1.14 (0.95-1.38)
0.1633
1.09 (0.90-1.32)
0.3753
1.07 (0.88-1.30)
0.4836
Q3
241/2548
1.20 (0.99-1.44)
0.0597
1.25 (1.04-1.51)
0.0178
1.17 (0.96-1.42)
0.1152
1.13 (0.93-1.38)
0.2198
Q4
258/2548
1.31 (1.09-1.57)
0.0043
1.43 (1.18-1.72)
0.0002
1.24 (1.01-1.51)
0.0385
1.16 (0.94-1.42)
0.1588
Per 1 SD
-
1.10 (1.04-1.18)
0.0020
1.14 (1.07-1.22)
< 0.00001
1.08 (1.00-1.15)
0.0466
1.05 (0.98-1.13)
0.1933
P for trend
-
-
0.0037
-
0.0001
-
0.0307
-
0.1388
Non-fatal stroke
METS-IR
-
1.01 (1.00-1.01)
0.1472
1.01 (1.00-1.02)
0.0119
1.01 (1.00-1.02)
0.1492
1.01 (1.00-1.02)
0.1576
Q1
103/2548
Reference
-
Reference
-
Reference
-
Reference
-
Q2
125/2547
1.23 (0.94-1.59)
0.1255
1.26 (0.97-1.64)
0.0836
1.20 (0.92-1.56)
0.1818
1.19 (0.91-1.56)
0.1980
Q3
139/2548
1.28 (0.99-1.66)
0.0618
1.37 (1.06-1.78)
0.0169
1.31 (1.00-1.71)
0.0489
1.32 (1.01-1.73)
0.0443
Q4
126/2548
1.27 (0.98-1.65)
0.0703
1.45 (1.11-1.88)
0.0064
1.29 (0.97-1.72)
0.0751
1.29 (0.97-1.72)
0.0835
Per 1 SD
-
1.07 (0.98-1.17)
0.1472
1.12 (1.03-1.23)
0.0119
1.08 (0.97-1.19)
0.1492
1.07 (0.97-1.19)
0.1576
P for trend
-
-
0.0720
-
0.0052
-
0.0620
-
0.0645
Total mortality
METS-IR
-
1.01 (1.00-1.01)
< 0.0001
1.02 (1.01-1.02)
< 0.0001
1.01 (1.01-1.02)
< 0.0001
1.01 (1.01-1.02)
< 0.0001
Q1
448/2548
Reference
-
Reference
-
Reference
-
Reference
Q2
448/2547
1.01 (0.88-1.15)
0.9213
1.07 (0.93-1.22)
0.3417
1.01 (0.88-1.15)
0.909
0.99 (0.87-1.13)
0.8556
Q3
491/2548
1.12 (0.98-1.27)
0.0891
1.30 (1.14-1.47)
< 0.0001
1.20 (1.05-1.37)
0.0082
1.17 (1.02-1.34)
0.0233
Q4
556/2548
1.31 (1.15-1.48)
< 0.0001
1.70 (1.50-1.93)
< 0.0001
1.47 (1.28-1.68)
< 0.0001
1.39 (1.21-1.59)
< 0.0001
Per 1 SD
-
1.10 (1.05-1.15)
< 0.0001
1.22 (1.17-1.28)
< 0.0001
1.15 (1.10-1.21)
< 0.0001
1.13 (1.08-1.19)
< 0.0001
P for trend
-
-
< 0.0001
-
< 0.0001
-
< 0.0001
-
< 0.0001
CHF
METS-IR
-
1.03 (1.02-1.03)
< 0.0001
1.04 (1.03-1.04)
< 0.0001
1.03 (1.02-1.04)
< 0.0001
1.03 (1.02-1.04)
< 0.0001
Q1
124/2548
Reference
-
Reference
-
Reference
-
Reference
-
Q2
132/2547
1.07 (0.84-1.37)
0.5656
1.13 (0.89-1.45)
0.3203
1.06 (0.83-1.36)
0.6457
1.04 (0.80-1.34)
0.7809
Q3
196/2548
1.62 (1.30-2.03)
< 0.0001
1.86 (1.48-2.33)
< 0.0001
1.82 (1.44-2.30)
< 0.0001
1.76 (1.39-2.23)
< 0.0001
Q4
239/2548
2.09 (1.69-2.60)
< 0.0001
2.65 (2.13-3.31)
< 0.0001
2.31 (1.82-2.92)
< 0.0001
2.22 (1.74-2.82)
< 0.0001
Per 1 SD
-
1.35 (1.26-1.44)
< 0.0001
1.48 (1.38-1.59)
< 0.0001
1.42 (1.31-1.53)
< 0.0001
1.40 (1.29-1.52)
< 0.0001
P for trend
-
-
< 0.0001
-
< 0.0001
-
< 0.0001
-
< 0.0001
Major CHD
METS-IR
-
1.01 (1.01-1.02)
< 0.0001
1.02 (1.01-1.02)
< 0.0001
1.01 (1.01-1.02)
< 0.0001
1.01 (1.01-1.02)
< 0.0001
Q1
379/2548
Reference
-
Reference
-
Reference
-
Reference
-
Q2
426/2547
1.14 (0.99-1.31)
0.0596
1.16 (1.01-1.33)
0.0410
1.10 (0.95-1.27)
0.1887
1.07 (0.93-1.24)
0.3298
Q3
495/2548
1.34 (1.17-1.53)
< 0.0001
1.41 (1.23-1.61)
< 0.0001
1.31 (1.14-1.51)
0.0001
1.26 (1.10-1.45)
0.0012
Q4
544/2548
1.50 (1.32-1.72)
< 0.0001
1.65 (1.44-1.89)
< 0.0001
1.43 (1.24-1.66)
< 0.0001
1.35 (1.17-1.56)
< 0.0001
Per 1 SD
-
1.17 (1.12-1.23)
< 0.0001
1.22 (1.17-1.28)
< 0.0001
1.16 (1.10-1.22)
< 0.0001
1.13 (1.08-1.19)
< 0.0001
P for trend
-
-
< 0.0001
-
< 0.0001
-
< 0.0001
-
< 0.0001
Incremental predicted value of METS-IR for cardiovascular outcomes
The prognostic utility of METS-IR for MACEs, all-cause mortality, CHF, and major CHD was assessed using C-statistic, NRI, and IDI, as presented in Table 3. Incorporation of METS-IR into conventional cardiovascular risk models significantly improved discriminative capacity for MACE, total mortality, CHF, and major CHD events in patients with T2DM[31]. Specifically, the addition of METS-IR yielded a C-statistic of 0.674 (95%CI: 0.664-0.683; P = 0.0170), NRI of 0.054 (95%CI: 0.019-0.084; P = 0.0070), and IDI of 0.003 (95%CI: 0.001-0.005; P < 0.0001) for MACEs, indicating enhanced discriminative capacity and risk reclassification. Similar improvements were observed for total and cardiovascular mortality, CHF, and major CHD outcomes (Table 3). Comparative performance analyses were also conducted between METS-IR and other surrogate markers of IR, including triglyceride-glucose index (TyG), TyG-BMI, waist-to-height ratio (WHTR), and TyG-WHTR. In integrated models, METS-IR consistently outperformed alternative indices, demonstrating the greatest incremental predictive value - particularly for CHF, where it achieved the highest increase in area under the curve (Supplementary Table 1).
Table 3 Additional predictive value of metabolic score for insulin resistance for outcome.
Outcome
C-statistic (95%CI)
P value
NRI (95%CI)
P value
IDI (95%CI)
P value
MACEs
Conventional model
0.669 (0.660-0.679)
-
Reference
-
Reference
-
Conventional model + METS-IR
0.674 (0.664-0.683)
0.0170
0.054 (0.019-0.084)
0.0070
0.003 (0.001-0.005)
< 0.0001
CVD mortality
Conventional model
0.706 (0.697-0.715)
-
Reference
-
Reference
-
Conventional model + METS-IR
0.716 (0.708-0.725)
0.0050
0.104 (0.047-0.145)
< 0.0001
0.004 (0.002-0.008)
< 0.0001
Non-fatal MI
Conventional model
0.661 (0.652-0.670)
-
Reference
-
Reference
-
Conventional model + METS-IR
0.663 (0.654-0.672)
0.2705
0.025 (-0.018 to 0.067)
0.1930
0.001 (0.000-0.003)
0.0130
Non-fatal stroke
Conventional model
0.632 (0.623-0.642)
-
Reference
-
Reference
-
Conventional model + METS-IR
0.634 (0.624-0.643)
0.6054
0.052 (-0.015 to 0.102)
0.1660
0.001 (0.000-0.002)
0.0860
Total mortality
Conventional model
0.693 (0.684-0.702)
-
Reference
-
Reference
-
Conventional model + METS-IR
0.701 (0.692-0.710)
0.0001
0.078 (0.048-0.108)
< 0.0001
0.004 (0.002-0.007)
< 0.0001
Congestive heart failure
Conventional model
0.707 (0.698-0.716)
-
Reference
-
Reference
-
Conventional model + METS-IR
0.732 (0.723-0.741)
< 0.0001
0.211 (0.149-0.254)
< 0.0001
0.016 (0.009-0.025)
< 0.0001
Major CHD
Conventional model
0.677 (0.668-0.686)
-
Reference
-
Reference
-
Conventional model + METS-IR
0.684 (0.674-0.693)
0.0018
0.055 (0.024-0.081)
< 0.0001
0.004 (0.002-0.006)
< 0.0001
Subgroup analyses
To further delineate the association between METS-IR and adverse clinical outcomes, exploratory stratified analyses were conducted across key demographic and clinical subgroups. Stratification variables included age (< 60 vs ≥ 60 years), sex, race (White vs non-White), history of CVD, hypertension, HbA1c levels (< 8.0% vs ≥ 8.0%), and aspirin use (Figure 4). These analyses were designed to identify potential effect modifiers of the METS-IR-MACEs association (Figure 4). Significant interaction effects were observed for HbA1c level (P for interaction = 0.0062) and aspirin use (P for interaction = 0.0234), suggesting that the predictive value of METS-IR for MACEs was amplified in individuals with HbA1c < 8.0% and in those not receiving aspirin therapy. In contrast, no significant interaction was detected for age, sex, race, prior CVD, or hypertension status (all P for interaction > 0.05), nor was there evidence of effect modification in the METS-IR-CHF relationship across any subgroup (P for interaction > 0.05). Collectively, these findings suggest that METS-IR may serve as a more robust prognostic indicator of MACEs in T2DM patients with better glycemic control and without concurrent aspirin use.
Figure 4 Subgroup and interaction analyses of the association between metabolic score for insulin resistance and the risk of major adverse cardiovascular events and congestive heart failure.
Participants were stratified by age (< 60 years and ≥ 60 years), sex, race, cardiovascular disease history, previous hypertension, glycated hemoglobin (< 8.0% and ≥ 8.0%), and aspirin use. Non-White participants included individuals of Hispanic, Black, and other ethnic backgrounds. A: Major adverse cardiovascular events; B: Congestive heart failure. MACEs: Major adverse cardiovascular events; CVD: Cardiovascular disease; CHF: Congestive heart failure; HbA1C: Hemoglobin A1c.
To further explore potential non-linear associations, multivariable-adjusted RCS models were applied within stratified subgroups based on HbA1c status and aspirin use (Figures 5 and 6). Among patients with HbA1c < 8.0%, a significant linear association was observed between METS-IR and MACEs (overall P = 0.003; non-linearity P = 0.398). A similar linear trend was evident in aspirin-naïve individuals (overall P < 0.001; non-linearity P = 0.337). In contrast, no statistically significant associations were observed in the subgroups with HbA1c ≥ 8.0% or in those receiving aspirin (all P > 0.05).
Figure 6 Multivariable-adjusted hazard ratios for primary and secondary outcomes in patients not using aspirin based on restricted cubic spline analysis.
A: Major adverse cardiovascular events; B: Cardiovascular disease mortality; C: Non-fatal myocardial infarction; D: Non-fatal stroke; E: Total mortality; F: Congestive heart failure; G: Major coronary heart disease. MACEs: Major adverse cardiovascular events; CVD: Cardiovascular disease; MI: Myocardial infarction; CHF: Congestive heart failure; CHD: Coronary heart disease; METS-IR: Metabolic score for insulin resistance; CI: Confidence interval; HR: Hazard ratio.
Association of METS-IR with MACEs in patients with HbAC1 < 8.0%
Subsequent analyses restricted to patients with HbA1c < 8.0% evaluated the prognostic value of METS-IR for predicting MACEs using a hierarchical modeling approach: Unadjusted (model 1), partially adjusted (model 2), and fully adjusted (model 3) models (Table 4). In the fully adjusted model, individuals in the highest METS-IR quartile exhibited a 47% increased risk of MACEs compared to those in the lowest quartile (HR = 1.47, 95%CI: 1.16-1.86). Moreover, each 1-SD increment in METS-IR was independently associated with a 17% higher risk of MACEs (HR = 1.17, 95%CI: 1.07-1.27; P < 0.001; Table 4), indicating enhanced prognostic utility of METS-IR within this glycemic control subgroup relative to the overall T2DM population.
Table 4 Association between metabolic score for insulin resistance and major adverse cardiovascular events in patients with glycated hemoglobin < 8.0.
Outcome
Events/n
Non-adjusted
Model 4
Model 5
Model 6
HR (95%CI)
P value
HR (95%CI)
P value
HR (95%CI)
P value
HR (95%CI)
P value
MACEs
-
1.02 (1.01-1.02)
< 0.0001
1.02 (1.02-1.03)
< 0.0001
1.02 (1.01-1.02)
< 0.0001
1.01 (1.01-1.02)
0.0005
METS-IR
Q1
160/1219
Reference
-
Reference
-
Reference
-
Reference
-
Q2
168/1154
1.12 (0.90-1.39)
0.2985
1.16 (0.93-1.44)
0.1907
1.08 (0.87-1.36)
0.4755
1.04 (0.83-1.30)
0.7139
Q3
172/1043
1.29 (1.04-1.60)
0.0212
1.43 (1.15-1.78)
0.0013
1.28 (1.02-1.60)
0.0361
1.22 (0.97-1.54)
0.0848
Q4
192/942
1.62 (1.31-2.00)
< 0.0001
1.90 (1.54-2.36)
< 0.0001
1.61 (1.28-2.03)
< 0.0001
1.47 (1.16-1.86)
0.0012
Per 1 SD
-
1.20 (1.11-1.29)
< 0.0001
1.28 (1.19-1.39)
< 0.0001
1.20 (1.11-1.31)
< 0.0001
1.17 (1.07-1.27)
0.0005
P for trend
-
-
< 0.0001
-
< 0.0001
-
< 0.0001
-
0.0005
Additional stratified analyses across demographic and clinical subgroups - including sex, age, race, history of CVD, hypertension, proteinuria, smoking status, diabetes duration (< 10 years and ≥ 10years), and educational attainment - revealed consistent associations between elevated METS-IR and increased MACE risk among individuals with HbA1c < 8.0%. No significant interaction was observed across any subgroup (all P for interaction > 0.05), as shown in Supplementary Figure 1.
Correlation of METS-IR index with MACEs in aspirin-free patients
Further analyses were conducted to assess the prognostic significance of METS-IR for MACEs among aspirin-naïve individuals. In multivariable-adjusted models, participants in the highest METS-IR quartile demonstrated a 24% increased risk of MACEs compared to those in the lowest quartile (HR = 1.24, 95%CI: 1.07-1.43). Additionally, each 1-SD elevation in METS-IR was associated with a 9% greater MACEs risk (HR = 1.09, 95%CI: 1.04-1.15; P < 0.001; Table 5), reinforcing the independent prognostic utility of METS-IR in aspirin-free T2DM populations.
Table 5 Association between metabolic score for insulin resistance and major adverse cardiovascular events in aspirin-free patients.
Outcome
Events/n
Non-adjusted
Model 7
Model 8
Model 9
HR (95%CI)
P value
HR (95%CI)
P value
HR (95%CI)
P value
HR (95%CI)
P value
MACEs
-
1.01 (1.01-1.01)
< 0.0001
1.01 (1.01-1.02)
< 0.0001
1.01 (1.01-1.01)
< 0.0001
1.01 (1.00, 1.01)
0.0008
METS-IR
Q1
171/1209
Reference
-
Reference
-
Reference
-
Reference
-
Q2
168/1151
1.05 (0.92-1.20)
0.4837
1.07 (0.94-1.23)
0.3044
1.02 (0.89-1.17)
0.7661
1.00 (0.87-1.15)
0.9627
Q3
202/1097
1.22 (1.07-1.39)
0.0036
1.31 (1.14-1.49)
< 0.0001
1.23 (1.07-1.41)
0.0035
1.19 (1.04-1.37)
0.0123
Q4
232/1138
1.32 (1.15-1.50)
< 0.0001
1.50 (1.31-1.71)
< 0.0001
1.31 (1.14-1.52)
0.0002
1.24 (1.07-1.43)
0.004
Per 1 SD
-
1.12 (1.07-1.17)
< 0.0001
1.18 (1.12-1.23)
< 0.0001
1.12 (1.06-1.18)
< 0.0001
1.09 (1.04-1.15)
0.0008
P for trend
-
-
< 0.0001
-
< 0.0001
-
< 0.0001
-
0.0005
Subgroup analyses revealed consistent associations across sex, age, race, hypertension status, proteinuria, smoking behavior, diabetes duration (< 10 years and ≥ 10 years), and educational attainment, with no significant interaction effects detected (all P > 0.05). However, a statistically significant interaction was observed with prior CVD history (P for interaction = 0.0179), where METS-IR demonstrated enhanced prognostic discrimination for MACEs in aspirin-free T2DM patients with established CVD (Supplementary Figure 2).
DISCUSSION
The present study examines the association of the METS-IR with cardiovascular outcomes among individuals with established T2DM. Our findings demonstrated that elevated baseline METS-IR predicts subsequent adverse cardiovascular events, with particular significance for CHF, robust to adjustment for conventional cardiovascular risk factors. Specifically, compared to patients in the first METS-IR quartile, those in the fourth quartile had 25% increased incidence of MACEs, 39% higher total mortality, 35% greater risk of CHD and 122% excess CHF incidence. Furthermore, our research observed that the incorporation of METS-IR into traditional prediction models results in a better predictive performance for most cardiovascular events, encompassing improved discrimination ability, accuracy and precision in risk categorization and prediction. However, for non-fatal MI and stroke, the incremental predictive value of METS-IR incorporation did not reach statistical significance. This may be attributed to a weaker association between these two events and METS-IR or the need for a larger sample size to capture such an association. Notably, METS-IR exhibited significantly enhanced predictive capacity for adverse outcomes in T2DM patients compared to conventional IR indices - WHTR, TyG, TyG-BMI, and TyG-WHTR. Especially, METS-IR achieved the greatest increment in area under the curve values in CHF prediction. Besides, the predictive superiority of METS-IR was especially observed in patients with HbA1c < 8.0% and those not taking aspirin for MACEs.
A previous cohort study found that elevated cumulative METS-IR is closely related to an increased risk of CVD and a more pronounced trend over time, which is consistent with our findings and underscores the necessity of keeping optimal METS-IR levels throughout life[32]. Additionally, multicenter cohort studies conducted in patients with idiopathic pulmonary hypertension have shown that non-insulin-based IR indices predict disease severity and adverse outcomes, emphasizing the relevance of IR beyond diabetes[19].
In the Chinese 4C cohort study, METS-IR outperformed other IR indices in predicting MACEs, regardless of diabetes status[24]. Further studies have indicated that biological aging the effects of METS-IR on all-cause and CVD mortality, underscoring the importance of metabolic health for longevity and healthy aging[33]. Moreover, research has demonstrated the varying predictive abilities of different IR indices for hypertension risk, with METS-IR being linked to the onset of isolated diastolic hypertension[17,18,34]. Results from a national perspective cohort study suggest that METS-IR predicts incident T2DM among individuals in midlife and advanced age groups[23]. Other studies suggest IR could predict occurrence and severity of retinopathy in T2DM[35], further highlighting the role of METS-IR in metabolic syndrome.
Additionally, machine learning-based studies have shown the superior predictive performance of alternative IR indices for coronary artery disease[22], supporting the theoretical basis for using METS-IR to predict cardiovascular events. In post-percutaneous coronary intervention populations, non-insulin-based IR indices have demonstrated significant clinical utility for cardiovascular outcomes assessment, particularly highlighting the robust applicability of METS-IR in secondary prevention strategies and major adverse cardiovascular and cerebrovascular events risk stratification[11]. Data from the NHANES study in the general population further confirm uniquely significant top performance of METS-IR in predicting cardiovascular and all-cause mortality against the three other surrogate IR indicators including TyG index, TG/HDL-C and homeostasis model assessment for IR[8]. All in all, these studies support METS-IR as a superior predictor for MACEs in individuals suffered from T2DM.
The biological pathway from METS-IR to MACEs are likely multifactorial. IR is known to contribute to endothelial dysfunction, systemic inflammation, and dyslipidemia - key drivers of atherosclerosis and its related cardiovascular events[4,36-39]. Higher METS-IR scores reflect greater IR, which may exacerbate these pathological processes and promote the development of adverse cardiovascular events[19,40,41].
In conclusion, our findings highlight significance of METS-IR for cardiovascular risk assessment in T2DM patients. Higher METS-IR is linked to a higher possibility of cardiovascular events, indicating its potential usefulness as a predictive tool in clinical settings. Unlike traditional risk assessment methods, which often rely on single metabolic markers, METS-IR offers a more comprehensive evaluation, potentially identifying high-risk patients who may be overlooked by conventional methods[42,43]. Given its simplicity and non-invasive nature, METS-IR can be readily incorporated into everyday clinical practice, helping to identify high-risk individuals early and allowing for prompt interventions to lower cardiovascular risk[8,20,44].
Our research has indicated that, among patients with T2DM, the predictive value of METS-IR for MACEs is greater in individuals with HbA1c levels below 8.0% than in those with levels at or above 8.0%. This superiority may be attributed to the fact that persistently elevated HbA1c levels may not accurately reflect the severity of IR, thus limiting the effectiveness of METS-IR in evaluating IR to predict MACEs in T2DM. Furthermore, patients with high HbA1c levels may experience severely compromised pancreatic β-cell function, potentially to the extent that they are unable to secrete sufficient insulin to lower blood glucose levels[45]. This impairment can lead to increased blood viscosity and more severe endothelial dysfunction, thereby facilitating thrombus formation and heightening the risk of MACEs[46]. Additionally, our study found that METS-IR has a greater predictive value for MACEs in T2DM patients not undergoing aspirin therapy than in those who are. This observation may suggest that aspirin exerts a certain preventive effect on MACEs. However, our research does not elucidate the balance of benefits and risks associated with the prophylactic use of aspirin in T2DM patients. Notably, among aspirin-free patients, METS-IR exhibited significantly stronger predictive capacity for MACEs in those with pre-existing CVD history. This result corresponds well with clinical observations, as T2DM patients with CVD who do not take aspirin are more likely to experience MACEs. Thus, this study also supports the role of aspirin in preventing CVD as recommended in the ADA guidelines[32]. Therefore, ongoing assessment of METS-IR in T2DM patients is advisable, with treatment strategies adapted according to its changes over time to help reduce the likelihood of future cardiovascular complications.
Despite notable strengths such as a large cohort, extended follow-up duration, and thorough control for confounders, our study is not without limitations. First, the retrospective design precludes causal inference. Although numerous confounding factors were accounted for, the possibility of residual confounding remains. Second, METS-IR was calculated at baseline, without accounting for temporal changes in IR. Future longitudinal studies with repeated METS-IR measurements are needed to capture the dynamic nature of IR and its impact on cardiovascular outcomes in T2DM patients. Additionally, the study did not assess the predictive performance of METS-IR for MACEs in comparison with homeostasis model assessment for IR, the established reference for measuring IR. Lastly, since our cohort primarily consisted of individuals from the United States, the applicability of our findings to broader populations may be limited. The conclusions drawn may be most relevant to United States patients with T2DM aged 40-79 years. Caution is warranted when extending these results to different age ranges, clinical conditions, or ethnic groups.
Further investigations are necessary to assess the clinical applications of METS-IR. Extensive prospective studies are required to confirm its predictive accuracy in diverse populations. Combining METS-IR with other metabolic markers could enhance its role in cardiovascular risk assessment. Additionally, mechanistic studies may provide deeper insights into the biological basis of METS-IR, supporting its use in clinical practice. Targeted interventions aimed at improving IR in T2DM patients with elevated METS-IR could provide important information on their effectiveness in lowering cardiovascular risk.
CONCLUSION
This study establishes a strong link between METS-IR and cardiovascular complications in patients with T2DM. Serving as an integrative metabolic marker, METS-IR could offer important insights for evaluating cardiovascular risk, especially in high-risk groups where conventional approaches fall short. Continued research is needed to further investigate the clinical applications of METS-IR and to confirm its utility in cardiovascular risk prediction.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the ACCORD/ACCORDION study group and the NHLBI Biologic Specimen and Data Repository Information Coordinating Center. The opinions expressed in this article are those of the authors and do not necessarily reflect the views of the ACCORD/ACCORDION study authors or NHLBI.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Endocrinology and metabolism
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade A, Grade B, Grade B, Grade D
Novelty: Grade B, Grade D
Creativity or Innovation: Grade B, Grade E
Scientific Significance: Grade B, Grade D
P-Reviewer: Horowitz M; Karonova T; Nakhratova OV; Shen QE S-Editor: Bai Y L-Editor: A P-Editor: Wang WB
Sun H, Saeedi P, Karuranga S, Pinkepank M, Ogurtsova K, Duncan BB, Stein C, Basit A, Chan JCN, Mbanya JC, Pavkov ME, Ramachandaran A, Wild SH, James S, Herman WH, Zhang P, Bommer C, Kuo S, Boyko EJ, Magliano DJ. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045.Diabetes Res Clin Pract. 2022;183:109119.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 3033][Cited by in RCA: 4767][Article Influence: 1589.0][Reference Citation Analysis (36)]
Bello-Chavolla OY, Almeda-Valdes P, Gomez-Velasco D, Viveros-Ruiz T, Cruz-Bautista I, Romo-Romo A, Sánchez-Lázaro D, Meza-Oviedo D, Vargas-Vázquez A, Campos OA, Sevilla-González MDR, Martagón AJ, Hernández LM, Mehta R, Caballeros-Barragán CR, Aguilar-Salinas CA. METS-IR, a novel score to evaluate insulin sensitivity, is predictive of visceral adiposity and incident type 2 diabetes.Eur J Endocrinol. 2018;178:533-544.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 95][Cited by in RCA: 304][Article Influence: 43.4][Reference Citation Analysis (0)]
Ferrannini G, De Bacquer D, Erlund I, Gyberg V, Kotseva K, Mellbin L, Norhammar A, Schnell O, Tuomilehto J, Vihervaara T, Wood D, Rydén L. Measures of Insulin Resistance as a Screening Tool for Dysglycemia in Patients With Coronary Artery Disease: A Report From the EUROASPIRE V Population.Diabetes Care. 2022;45:2111-2117.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 7][Reference Citation Analysis (0)]
Wang J, Tang H, Tian J, Xie Y, Wu Y. Non-insulin-based insulin resistance indices predict early neurological deterioration in elderly and middle-aged acute ischemic stroke patients in Northeast China.Sci Rep. 2024;14:16138.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 4][Reference Citation Analysis (0)]
Pan L, Zou H, Meng X, Li D, Li W, Chen X, Yang Y, Yu X. Predictive values of metabolic score for insulin resistance on risk of major adverse cardiovascular events and comparison with other insulin resistance indices among Chinese with and without diabetes mellitus: Results from the 4C cohort study.J Diabetes Investig. 2023;14:961-972.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in RCA: 15][Reference Citation Analysis (0)]
Lee KJ, Tilling KM, Cornish RP, Little RJA, Bell ML, Goetghebeur E, Hogan JW, Carpenter JR; STRATOS initiative. Framework for the treatment and reporting of missing data in observational studies: The Treatment And Reporting of Missing data in Observational Studies framework.J Clin Epidemiol. 2021;134:79-88.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 130][Cited by in RCA: 196][Article Influence: 49.0][Reference Citation Analysis (0)]
Vasanthakumar A, Chisanga D, Blume J, Gloury R, Britt K, Henstridge DC, Zhan Y, Torres SV, Liene S, Collins N, Cao E, Sidwell T, Li C, Spallanzani RG, Liao Y, Beavis PA, Gebhardt T, Trevaskis N, Nutt SL, Zajac JD, Davey RA, Febbraio MA, Mathis D, Shi W, Kallies A. Sex-specific adipose tissue imprinting of regulatory T cells.Nature. 2020;579:581-585.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 134][Cited by in RCA: 164][Article Influence: 32.8][Reference Citation Analysis (0)]
