Predictive value of combined triglyceride-glucose index and glycated hemoglobin for long-term cardiovascular outcomes in stable coronary artery disease

doi:10.21203/rs.3.rs-3883934/v1

Background: Insulin resistance (IR) affects prognosis in stable coronary artery disease (SCAD). While triglyceride-glucose (TyG) index assesses IR, glycated hemoglobin (HbA1c) provides long-term glycemic level information. However, the combined predictive value of TyG index and HbA1c is unclear.

Methods: This retrospective cohort study included 944 SCAD patients who underwent percutaneous coronary intervention (PCI). Major adverse cardiovascular events (MACE) were followed up for 31 months. Receiver operating characteristic (ROC) curves assessed the predictive ability of TyG index, HbA1c and their combination. Patients were further grouped based on the optimal cutoff value of the TyG index and clinical abnormal ranges of HbA1c to analyze the impact of different groups.

Results: The TyG index was independently predictive of MACE and repeat revascularization (HR: 1.574, P = 0.021; and HR: 1.379, P = 0.027). Combining TyG index with HbA1c significantly enhanced overall MACE prediction, with the AUC for cardiac death increasing to 0.901 (P = 0.016). The high TyG-low HbA1c group had more than double the cumulative incidence of MACE compared to the low TyG-low HbA1c group (HR: 2.069, P < 0.001), with the high TyG-high HbA1c group showing an even greater risk (HR: 2.463, P < 0.001). Diabetic patients in the high TyG-low HbA1c category had the highest risk increase (HR: 3.375, P = 0.038).

Conclusions: The combined use of TyG and HbA1c improves the predictive ability for MACE in all SCAD patients, with predictive value for chronic outcomes, but not for acute events.

stable coronary artery disease

major adverse cardiovascular events

insulin resistance

triglyceride-glucose index

glycated hemoglobin

Insulin resistance (IR), as a key pathophysiological mechanism of type 2 diabetes, significantly influences the development and progression of coronary artery disease (CAD)(1–3). IR promotes atherosclerosis by enhancing inflammatory responses, oxidative stress, and endothelial dysfunction(4, 5). However, clinical trials indicate that lifestyle interventions aimed at improving IR in diabetic patients without macrovascular complications do not necessarily result in cardiovascular benefits(6). This suggests that research into the impact of IR on cardiovascular events should not be confined to populations with uncomplicated diabetes alone. The triglyceride-glucose (TyG) index has been validated as an effective surrogate for IR(7–9) and is associated with CAD(10), carotid atherosclerosis(11), and other metabolic CVDs(12). It also predicts long-term outcomes after percutaneous coronary intervention (PCI), especially in acute coronary syndrome (ACS) patients(13, 14). Since ACS represents an acute manifestation of CAD and IR is a chronic condition, the TyG index may be more relevant for prognosis post-PCI in stable coronary artery disease (SCAD)(15), which reflects a chronic phase. Our real-world study in urban Chinese populations found that CAD is one of the most common comorbidities in diabetic patients, but approximately 83.1% of diabetic patients die with CAD still in a chronic stage, rather than experiencing acute cardiovascular events(16). Therefore, it might be more appropriate to further investigate the impact of IR on cardiovascular outcomes in patients with SCAD. In addition, the TyG index, derived from fasting glucose and triglyceride levels, may provide a transient snapshot of glycemic status(9). In contrast, glycated hemoglobin (HbA1c) levels indicate longer-term glycemic level. Recognizing that HbA1c is a known CAD risk factor and is linked with post-PCI outcomes(17, 18), This study suggests that the TyG-HbA1c combined index may improve the predictive accuracy of chronic outcomes in patients with SCAD. This novel approach aims to fill the research void by thoroughly assessing the prognostic value of integrating HbA1c with the TyG index in SCAD patients post-PCI.

Study population and follow-up

This retrospective cohort study, approved by the Ethics Committee of Zhongshan Hospital, Fudan University, encompassed a population of 944 adult patients who were diagnosed with SCAD via coronary angiography and received following PCI treatment at the Department of Cardiology from May 1, 2019, to April 30, 2020. Participants with 1) severe cardiac, liver, or kidney diseases, 2) a history of hematologic and solid malignancies, 3) concurrent infections or autoimmune diseases, or 4) incomplete medical records were excluded from the study. Patient follow-up was conducted using a combined approach of Hospital Information System (HIS) records and telephone interviews. During the follow-up period, the loss to follow-up rate was 13.67%. The median follow-up duration was 32.18 months (Fig. 1).

This flowchart illustrates the exclusion criteria, follow-up procedures, and outcome events. For the lost to follow-up population, we conducted an Intention-to-Treat (ITT) analysis. SCAD = stable coronary artery disease, PCI = percutaneous coronary intervention, CKD = chronic kidney disease.

Data collection and definition

The demographic information, medical and medication history, echocardiography, and coronary angiography data were Obtained from electronic medical records by trained clinical physicians. Peripheral venous blood samples were collected on the morning of the first day of hospitalization after at least 8 hours of fasting to determine the blood test indices. The TyG index was calculated using the formula TyG index = ln [fasting triglycerides (mg/dl) × fasting glucose (mg/dl)/2](9). SCAD was defined as including the post-stable phase of ACS in addition to common chronic stable effort angina and ischemic cardiomyopathy(19). Diabetes mellitus (DM) was defined as fasting venous blood glucose ≥ 7.0mmol/L or random venous blood glucose ≥ 11.1mmol/L or a 2-hour blood glucose level ≥ 11.1mmol/L during an oral glucose tolerance test (OGTT), or receiving oral hypoglycemic agents or insulin therapy(20).

PCI

Preoperative preparations were meticulously completed in accordance with established procedural guidelines. Experienced physicians performed coronary angiography on the patients, during which the extent of coronary artery disease and the degree of stenosis were documented. Interventional treatment was recommended for cases where the left main coronary artery stenosis exceeded 50%, or stenosis of other major branches exceeded 80% in diameter. Informed consent was obtained from both the patients and their families prior to proceeding with further interventional procedures. The criteria for a successful operation included postoperative TIMI flow grade 3 in the affected vessels, residual stenosis of less than 20%, and the absence of serious complications.

Endpoints

The primary endpoint of this study was the occurrence of MACE, which included cardiac death, non-cardiac death, recurrent myocardial infarction, hospitalization for recurrent angina, heart failure, stroke, repeat revascularization, and stent thrombosis(21). Repeat revascularization encompassed both repeat PCI and coronary artery bypass grafting (CABG). Repeat PCI was defined as coronary angiography following initial PCI or treatment for new vascular lesions in the presence of ischemic signs or symptoms. Stent thrombosis was identified as the formation of a new thrombus at the stent implantation site within the treated vessel, leading to complete or incomplete occlusion(22).

Statistical analyses

Statistical analyses used SPSS v26.0. Categorical data were compared using chi-squared tests, and continuous variables were tested for normality with the Kolmogorov-Smirnov test. Normally distributed variables were analyzed with T-tests, while non-normally distributed ones used non-parametric tests. Cox regression determined factors for MACE, with significant univariate predictors (P < 0.1) included in a multivariate model. X-tile v3.6.1 identified the TyG index's optimal cutoff for MACE prediction(23). Receiver operating characteristic (ROC) curves and the area under the curve (AUC) assessed the predictive value of TyG index, HbA1c, and their combination. Subgroups were analyzed by DM status. Further TyG index was dichotomized into high and low groups based on the optimal cutoff value, and HbA1c levels were categorized into high (≥ 6.5%) and low (< 6.5%) groups according to the clinical normal range. The cumulative risk of MACE across different TyG-HbA1c groups was compared using the one minus survival function in survival analysis. Two-tailed P < 0.05 was deemed statistically significant.

Baseline characteristics

In the final analysis, 944 patients were included, among whom 237 (25.11%) occurred MACE. The MACE group had a higher proportion of patients with history of prior PCI, family history of ASCVD, and more right coronary artery (RCA) lesions compared to the non-MACE group. Additionally, the MACE group exhibited higher levels of white blood cell count, fasting blood glucose (FPG), HbA1c, triglycerides, TyG index, non-high-density lipoprotein cholesterol (N-HDL-C), apolipoprotein B, apolipoprotein E, and troponin T (P < 0.05), as well as lower left ventricular ejection fraction (LVEF), high-density lipoprotein cholesterol (HDL-C), and shorter follow-up time (P < 0.05). No other statistically significant differences were observed between the two groups. Detailed clinical characteristics are presented in (Table 1).

Analysis of risk factors associated with the occurrence of MACE

For the analysis of risk factors associated with MACE occurrence, we first conducted univariate Cox regression analysis (Supplemental Table 1). The analysis indicated that prior myocardial infarction, prior PCI, use of diuretics, family history of ASCVD, left ventricular end-systolic volume (LVESV), white blood cell count, blood urea nitrogen, FPG, HbA1c, triglycerides, TyG index, N-HDL-C, and apolipoprotein E were all risk factors for MACE occurrence (P < 0.05), while LVEF, eosinophil ratio, HDL-C, and apolipoprotein A1 were protective factors for MACE occurrence (P < 0.05). Subsequently, after adjusting for other factors in a multivariable Cox regression analysis, showed that prior PCI, family history of ASCVD, and TyG index were independent risk factors for MACE occurrence, with hazard ratios (HRs) of 1.514 (95% CI: 1.126–2.035), 1.875 (95% CI: 1.040–3.381), and 1.574 (95% CI: 1.192–2.079), respectively (Figure 2).

HbA1c and TyG index on the occurrence of different MACE types

Using single-factor Cox regression analysis, the impact of HbA1c and TyG index on different types of MACE was examined. The results indicated that both HbA1c and TyG index had a significant impact on overall MACE occurrence and repeat revascularization (P < 0.05). HbA1c was found to be also a risk factor for cardiovascular death, heart failure, and in-stent thrombosis, but this significance was not observed after adjusting for multiple factors in the Cox regression analysis. On the other hand, even after multivariate Cox regression adjustment, TyG index remained significantly associated with MACE occurrence and repeat revascularization, with HRs of 1.574 (95% CI: 1.192–2.079, P = 0.001) and 1.379 (95% CI: 1.080–3.557, P = 0.027), respectively (Table 2).

The optimal cutoff value for TyG index

Using X-tile to find the optimal cutoff value, the maximum chi-square value represents the best cutoff value. In this study, the maximum chi-square value was 35.584, with a corresponding P-value of <0.001. The optimal cutoff value for TyG was found to be 9.28(Supplemental Figure 1), when dividing patients into low and high TyG index groups based on the optimal cutoff value, the sensitivity for predicting MACE was 44.73% and the specificity was 73.97% (Supplemental Table 2).

The predictive efficacy of HbA1c, TyG index, and their combination

The ROC curves for HbA1c, TyG index, and their combination as predictors of overall MACE are shown in (Figure 3). The results indicate that HbA1c, TyG index, and their combination have some predictive efficacy for MACE, with AUC values of 0.554, 0.614, and 0.616, respectively (P < 0.05). The combination of the two indices suggests a trend towards improved predictive efficacy. The optimal cutoff for TyG is found to be 9.281 based on the ROC curve, which is consistent with the cutoff identified using X-tile analysis. At this cutoff, the sensitivity is 45.57% and the specificity is 73.40%. The optimal cutoff for HbA1c is 6.45%, with sensitivity and specificity of 40.93% and 67.89%, respectively.

In terms of different MACE types, the combination of HbA1c and TyG index significantly improves the predictive efficacy for cardiac death (AUC = 0.901, P = 0.016) and heart failure (AUC = 0.723, P = 0.002), while HbA1c and TyG index alone cannot predict these outcomes. HbA1c, TyG index, and their combination all demonstrate good predictive efficacy for the need for revascularization, with statistically significant differences, having AUC values of 0.562, 0.639, and 0.640 (P < 0.05), respectively. For stent thrombosis, the predictive efficacy of HbA1c (AUC = 0.683, P = 0.037) is a little better than the combination of the two (AUC = 0.672, P = 0.049). However, HbA1c, TyG index, and their combination cannot predict acute myocardial infarction and stroke (Table 3).

Subgroup analyses

The optimal cutoff value of the TyG index divided patients into high and low TyG index groups, while based on the clinical normal range of HbA1c, patients were divided into high and low HbA1c groups. It was observed that regardless of whether the patients had DM, there was no statistically significant difference in the occurrence of MACE based on the levels of HbA1c. However, risk for high TyG index was significantly higher for patients with DM (HR = 2.504, P < 0.001) compared to those without DM (HR = 1.915, P < 0.001). Patients were further categorized into four groups based on their TyG index and HbA1c levels: low TyG-low HbA1c, low TyG-high HbA1c, high TyG-low HbA1c, and high TyG-high HbA1c. For patients with DM, the high TyG-low HbA1c group faced an even higher risk, with a HR of 3.375 (95%CI: 1.068-10.663, P = 0.038) (Table 4).

Compared to the low TyG-low HbA1c reference group, the high TyG groups showed an increased cumulative incidence of MACE, specifically, the high TyG-low HbA1c group had a 2.069-fold increased risk (95%CI: 1.455-2.941, P < 0.001) (Figure 4B), and and the high TyG index-high HbA1c group had a 2.463-fold increase (95%CI: 1.764-3.440, P < 0.001) (Figure 4C). Regardless of HbA1c levels, the high TyG index-high HbA1c group compared to the low TyG index-high HbA1c group still had a 2.077-fold increase (95%CI: 1.383-3.120, P < 0.001) (Figure 4E), no significant difference was found between the high TyG groups (Figure 4F).

The present study has uncovered several intriguing findings: 1) After adjusting for confounding factors, the TyG index was identified as an independent risk factor for overall MACE and repeat revascularization but not for myocardial infarction or stroke. In contrast, HbA1c did not show a statistically significant impact, though its clinical implications warrant further long-term investigation. 2) The combined use of TyG index and HbA1c significantly enhanced the predictive capability for overall MACE, cardiac death, and heart failure. This suggests potential utility in clinical risk stratification. 3) Elevated TyG index was consistently associated with higher cumulative incidence rates of MACE in all patients with SCAD, notably within the high TyG-high HbA1c cohort, which exhibited the greatest risk. 4) Among diabetic individuals, the high TyG-low HbA1c group showed an increased risk of MACE, challenging the traditional emphasis on strict HbA1c level for post-PCI prognosis in CAD patients.

The TyG index is an independent risk factor for MACE in SCAD patients after PCI, particularly in increasing repeat revascularization, which may be primarily associated with the progression of chronic atherosclerosis(5, 24), consistent with the notion that IR is a chronic cumulative process. This study found that TyG index is not associated with cardiovascular death, acute myocardial infarction, and stroke, which are acute stress events based on chronic diseases. A systematic review has also confirmed a limited association between TyG index and these outcomes in CAD patients (25). These findings may help to explain why interventional studies with acute ASCVD events as the primary endpoint in diabetic patients have not yielded significant results(26). For CAD, IR is likely to be a promoting factor rather than a direct trigger for acute events, emphasizing the need for a more mechanistic approach in the design of clinical study endpoints or real clinical practice to achieve true precision medicine. Delving into the underlying mechanisms linking these metabolic markers to different cardiovascular outcomes could pave the way for new interventions and improve clinical outcomes (Figure 5).

Prior studies have solely compared the predictive efficacy of TyG index and HbA1c for cardiovascular events(27, 28). This study, for the first time, innovatively combined the TyG index and HbA1c, and found that their combined use significantly improved the predictive capability for overall MACE, cardiovascular death, repeat revascularization, and heart failure, particularly demonstrating good predictive value for cardiovascular death. In patients with CAD, elevated HbA1c levels are associated with an increased risk of cardiovascular death, especially non-arrhythmic death, which may make them less suitable candidates for implantable cardioverter-defibrillator prevention(29). More mechanistic studies are needed to explain this phenomenon. This finding may also suggest that SGLT-2 inhibitors, a class of diabetes drugs targeting the reduction of HbA1c levels, significantly decrease the risk of cardiac death and heart failure(30). In addition to the well-established impact on blood volume, this effect also involve metabolic-related mechanisms that require further exploration.

In the subgroup analyses of TyG index-HbA1c, we observed that a high TyG index was a risk factor for cumulative incidence of MACE, while high HbA1c was not strongly associated with MACE occurrence, even high TyG index-low HbA1c group had the highest risk for MACE in diabetic patients. This is contrary to the conventional belief that strict control of HbA1c is beneficial for the post-PCI prognosis of patients with CAD(31, 32). Several studies also confirmed this point, compared the effects of intensified treatment strategies targeting normal HbA1c levels (<6.5%) with those maintaining HbA1c levels between 7.0% and 7.9% on cardiovascular events in patients with DM. The results showed a higher mortality rate in the intensified treatment group(33, 34). Some studies have suggested that patients with low HbA1c levels may be in a prolonged state of hypoglycemia, which is associated with the occurrence of acute cardiovascular events(35). Additionally, there are studies proposing a "U-shaped" relationship for HbA1c, where both excessively high and low levels are associated with adverse outcomes(36, 37). The optimal range for HbA1c control in post-PCI diabetic patients still requires further research for confirmation. Novel indices like HbA1c variability would reflects more comprehensive blood glucose fluctuations, which warrant further exploration(38).

Limitation

Our study's findings should be interpreted considering several limitations. The retrospective single-center design may affect the generalizability of our results. The sample size, while sufficient for our analyses, could be a little small to identify subtle associations or to conduct detailed subgroup analyses. Despite adjustments for known confounders, the potential for residual confounding by unmeasured factors such as lifestyle and socioeconomic status remains. Future studies should employ prospective multicenter designs with larger and more diverse cohorts to enhance the external validity of our findings regarding the TyG index and HbA1c in predicting cardiovascular outcomes post-PCI. Investigating the underlying biological mechanisms and the impact of targeting insulin resistance alongside glycemic control in SCAD patients through interventional studies could also provide valuable insights into optimizing risk reduction strategies.

The combined use of TyG and HbA1c enhances the predictive capability for MACE in all patients with SCAD, demonstrating prognostic value for chronic outcomes but not for acute events. In SCAD patients with DM, those with high TyG and low HbA1c exhibit the highest risk, suggesting that overly strict blood glucose control may not be suitable for SCAD patients.

IR	insulin resistance
CAD	coronary artery disease
SCAD	stable coronary artery disease
PCI	percutaneous coronary intervention
ACS	acute coronary syndrome
MACE	major adverse cardiovascular events
TyG	triglyceride-glucose
HbA1c	glycated hemoglobin
DM	diabetes mellitus
ROC	receiver operating characteristic
AUC	area under the curve
LVEF	left ventricular ejection fraction
LVESV	left ventricular end-systolic volume
N-HDL-C	non-high-density lipoprotein cholesterol
CABG	coronary artery bypass grafting
SGLT-2	sodium-glucose transport Protein 2

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Zhongshan Hospital, Fudan University.

Consent for publication

Not applicable.

Availability of data and materials

The data are available from the corresponding authors upon reasonable request.

Competing interests

The authors declare that they have no competing interests.

Funding

This study was funded by Clinical Research Special Fund of Zhongshan Hospital, Fudan University (2018ZSLC02).

Authors' contributions

RL, HNZ, and ML conceived and designed the overall study. RL, QZW, and YXX collected patient data. RL conducted the statistical analysis. HNZ drafted the manuscript. YXD, SFJ, HJ, and ML provided guidance on data analysis and interpretation, as well as manuscript revision. All authors read and approved the final manuscript.

Acknowledgements

Not applicable.

Mottillo S, Filion KB, Genest J, Joseph L, Pilote L, Poirier P, et al. The metabolic syndrome and cardiovascular risk a systematic review and meta-analysis. J Am Coll Cardiol. 2010;56(14):1113–32.
Figtree GA, Vernon ST, Harmer JA, Gray MP, Arnott C, Bachour E, et al. Clinical Pathway for Coronary Atherosclerosis in Patients Without Conventional Modifiable Risk Factors: JACC State-of-the-Art Review. J Am Coll Cardiol. 2023;82(13):1343–59.
DeFronzo RA. Insulin resistance, lipotoxicity, type 2 diabetes and atherosclerosis: the missing links. The Claude Bernard Lecture 2009. Diabetologia. 2010;53(7):1270–87.
Kim JA, Montagnani M, Koh KK, Quon MJ. Reciprocal relationships between insulin resistance and endothelial dysfunction: molecular and pathophysiological mechanisms. Circulation. 2006;113(15):1888–904.
Mechanick JI, Farkouh ME, Newman JD, Garvey WT. Cardiometabolic-Based Chronic Disease, Adiposity and Dysglycemia Drivers: JACC State-of-the-Art Review. J Am Coll Cardiol. 2020;75(5):525–38.
Gregg EW, Jakicic JM, Blackburn G, Bloomquist P, Bray GA, Clark JM, et al. Association of the magnitude of weight loss and changes in physical fitness with long-term cardiovascular disease outcomes in overweight or obese people with type 2 diabetes: a post-hoc analysis of the Look AHEAD randomised clinical trial. Lancet Diabetes Endocrinol. 2016;4(11):913–21.
Nabipoorashrafi SA, Seyedi SA, Rabizadeh S, Ebrahimi M, Ranjbar SA, Reyhan SK, et al. The accuracy of triglyceride-glucose (TyG) index for the screening of metabolic syndrome in adults: A systematic review and meta-analysis. Nutr Metab Cardiovasc Dis. 2022;32(12):2677–88.
Lopez-Jaramillo P, Gomez-Arbelaez D, Martinez-Bello D, Abat MEM, Alhabib KF, Avezum Á, et al. Association of the triglyceride glucose index as a measure of insulin resistance with mortality and cardiovascular disease in populations from five continents (PURE study): a prospective cohort study. Lancet Healthy Longev. 2023;4(1):e23–e33.
Tao LC, Xu JN, Wang TT, Hua F, Li JJ. Triglyceride-glucose index as a marker in cardiovascular diseases: landscape and limitations. Cardiovasc Diabetol. 2022;21(1):68.
Wang X, Xu W, Song Q, Zhao Z, Meng X, Xia C, et al. Association between the triglyceride-glucose index and severity of coronary artery disease. Cardiovasc Diabetol. 2022;21(1):168.
Liu J, Ye M, Wang Y. Association between Triglyceride-Glucose Index and carotid plaque in Japanese population: a cross-sectional study. BMJ Open. 2023;13(6):e069663.
Cho YK, Han KD, Kim HS, Jung CH, Park JY, Lee WJ. Triglyceride-Glucose Index Is a Useful Marker for Predicting Future Cardiovascular Disease and Mortality in Young Korean Adults: A Nationwide Population-Based Cohort Study. J Lipid Atheroscler. 2022;11(2):178–86.
Zhao Q, Zhang TY, Cheng YJ, Ma Y, Xu YK, Yang JQ, et al. Triglyceride-Glucose Index as a Surrogate Marker of Insulin Resistance for Predicting Cardiovascular Outcomes in Nondiabetic Patients with Non-ST-Segment Elevation Acute Coronary Syndrome Undergoing Percutaneous Coronary Intervention. J Atheroscler Thromb. 2021;28(11):1175–94.
Wang L, Cong HL, Zhang JX, Hu YC, Wei A, Zhang YY, et al. Triglyceride-glucose index predicts adverse cardiovascular events in patients with diabetes and acute coronary syndrome. Cardiovasc Diabetol. 2020;19(1):80.
Tao S, Yu L, Li J, Xie Z, Huang L, Yang D, et al. Prognostic value of triglyceride-glucose index in patients with chronic coronary syndrome undergoing percutaneous coronary intervention. Cardiovasc Diabetol. 2023;22(1):322.
Li X, Liu R, Chen Y, Han Y, Wang Q, Xu Y, et al. Patterns and Trends in Mortality Associated With and Due to Diabetes Mellitus in a Transitioning Region With 3.17 Million People: Observational Study. JMIR Public Health Surveill. 2023;9:e43687.
Liu Y, Yang YM, Zhu J, Tan HQ, Liang Y, Li JD. Prognostic significance of hemoglobin A1c level in patients hospitalized with coronary artery disease. A systematic review and meta-analysis. Cardiovasc Diabetol. 2011;10:98.
Havakuk O, Banai S, Halkin A, Berliner S, Herz I, Revivo M, et al. HbA1c levels are associated with long term mortality rates in patients undergoing coronary angiography. J Am Coll Cardiol. 2014;63(12Supplement):A1703–A.
Montalescot G, Sechtem U, Achenbach S, Andreotti F, Arden C, Budaj A, et al. 2013 ESC guidelines on the management of stable coronary artery disease: the Task Force on the management of stable coronary artery disease of the European Society of Cardiology. Eur Heart J. 2013;34(38):2949–3003.
ElSayed NA, Aleppo G, Aroda VR, Bannuru RR, Brown FM, Bruemmer D, et al. 2. Classification and Diagnosis of Diabetes: Standards of Care in Diabetes-2023. Diabetes Care. 2023;46(Suppl 1):19–s40.
Hicks KA, Tcheng JE, Bozkurt B, Chaitman BR, Cutlip DE, Farb A, et al. J Am Coll Cardiol. 2015;66(4):403–69. 2014 ACC/AHA Key Data Elements and Definitions for Cardiovascular Endpoint Events in Clinical Trials: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards (Writing Committee to Develop Cardiovascular Endpoints Data Standards).
Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, et al. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation. 2007;115(17):2344–51.
Camp RL, Dolled-Filhart M, Rimm DL. X-tile: a new bio-informatics tool for biomarker assessment and outcome-based cut-point optimization. Clin Cancer Res. 2004;10(21):7252–9.
Guo X, Shen R, Yan S, Su Y, Ma L. Triglyceride-glucose index for predicting repeat revascularization and in-stent restenosis in patients with chronic coronary syndrome undergoing percutaneous coronary intervention. Cardiovasc Diabetol. 2023;22(1):43.
Luo JW, Duan WH, Yu YQ, Song L, Shi DZ. Prognostic Significance of Triglyceride-Glucose Index for Adverse Cardiovascular Events in Patients With Coronary Artery Disease: A Systematic Review and Meta-Analysis. Front Cardiovasc Med. 2021;8:774781.
Brancati FL, Evans M, Furberg CD, Geller N, Haffner S, Kahn SE, et al. Midcourse correction to a clinical trial when the event rate is underestimated: the Look AHEAD (Action for Health in Diabetes) Study. Clin Trials. 2012;9(1):113–24.
Su WY, Chen SC, Huang YT, Huang JC, Wu PY, Hsu WH et al. Comparison of the Effects of Fasting Glucose, Hemoglobin A(1c), and Triglyceride-Glucose Index on Cardiovascular Events in Type 2 Diabetes Mellitus. Nutrients. 2019;11(11).
Jin JL, Sun D, Cao YX, Guo YL, Wu NQ, Zhu CG, et al. Triglyceride glucose and haemoglobin glycation index for predicting outcomes in diabetes patients with new-onset, stable coronary artery disease: a nested case-control study. Ann Med. 2018;50(7):576–86.
Venkateswaran RV, Moorthy MV, Chatterjee NA, Pester J, Kadish AH, Lee DC, et al. Diabetes and Risk of Sudden Death in Coronary Artery Disease Patients Without Severe Systolic Dysfunction. JACC Clin Electrophysiol. 2021;7(12):1604–14.
Kyriakos G, Quiles-Sanchez LV, Garmpi A, Farmaki P, Kyre K, Savvanis S, et al. SGLT2 Inhibitors and Cardiovascular Outcomes: Do they Differ or there is a Class Effect? New Insights from the EMPA-REG OUTCOME trial and the CVD-REAL Study. Curr Cardiol Rev. 2020;16(4):258–65.
Smith SC Jr., Benjamin EJ, Bonow RO, Braun LT, Creager MA, Franklin BA, et al. AHA/ACCF Secondary Prevention and Risk Reduction Therapy for Patients with Coronary and other Atherosclerotic Vascular Disease: 2011 update: a guideline from the American Heart Association and American College of Cardiology Foundation. Circulation. 2011;124(22):2458–73.
Chen J, Yin D, Dou K. Intensified glycemic control by HbA1c for patients with coronary heart disease and Type 2 diabetes: a review of findings and conclusions. Cardiovasc Diabetol. 2023;22(1):146.
Funamizu T, Iwata H, Nishida Y, Miyosawa K, Doi S, Chikata Y, et al. Increased risk of cardiovascular mortality by strict glycemic control (pre-procedural HbA1c < 6.5%) in Japanese medically-treated diabetic patients following percutaneous coronary intervention: a 10-year follow-up study. Cardiovasc Diabetol. 2020;19(1):21.
Hayward RA, Reaven PD, Wiitala WL, Bahn GD, Reda DJ, Ge L, et al. Follow-up of glycemic control and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2015;372(23):2197–206.
Mongraw-Chaffin M, Bertoni AG, Golden SH, Mathioudakis N, Sears DD, Szklo M, et al. Association of Low Fasting Glucose and HbA1c With Cardiovascular Disease and Mortality: The MESA Study. J Endocr Soc. 2019;3(5):892–901.
Currie CJ, Peters JR, Tynan A, Evans M, Heine RJ, Bracco OL, et al. Survival as a function of HbA(1c) in people with type 2 diabetes: a retrospective cohort study. Lancet. 2010;375(9713):481–9.
Lemesle G, Bonello L, de Labriolle A, Maluenda G, Syed AI, Collins SD, et al. Prognostic value of hemoglobin A1C levels in patients with diabetes mellitus undergoing percutaneous coronary intervention with stent implantation. Am J Cardiol. 2009;104(1):41–5.
Guo K, Zhao Q, Wang M, Lu Y, Wo M, Zhou X, et al. The Scope of HbA1c Variability and Risk of Vascular Complications Among Patients with Type 2 Diabetes: A Systematic Review and Meta-Analysis of Prospective Studies. Horm Metab Res. 2022;54(2):94–103.

Table 1 | Baseline characteristics of the study population according to MACE occurrence

Variables	Total (n=944)	Non-MACE (n=707)	MACE (n=237)	P
General conditions
Male (%)	77.12	76.66	78.48	0.564
Age (yrs)	65(58-71)	65 (58-71)	66(59-71)	0.487
BMI (kg/m²)	24.89 (23.13-26.89)	24.93 (23.12-26.95)	24.77 (23.14-26.77)	0.527
Systolic blood pressure (mm Hg)	133(121-147)	133(120-147)	135(122-147)	0.895
Diastolic blood pressure (mm Hg)	78(71-86)	79(70-86)	78(72-86)	0.941
Risk factors, n (%)
Current smoking	231(24.47)	175(24.75)	56(23.63)	0.902
Prior MI	68(7.20)	49(6.93)	19(8.02)	0.576
Hypertension	626(66.31)	458(64.78)	168(70.89)	0.085
DM	292(30.93)	210(29.70)	82(34.60)	0.158
Hyperlipidemia	140(14.83)	109(15.42)	31(13.08)	0.381
Prior PCI	323(34.22)	229(32.39)	94(39.66)	0.041
Family history of ASCVD	32(3.39)	19(2.69)	13(5.49)	0.039
Echocardiography
LVEF (%)	65(61-67)	65(62-67)	64(60-67)	0.024
LVESV (mm)	30(28,32)	30.0(28,32)	30(28,33)	0.100
LVEDV (mm)	47(44,50)	47(44,50)	47(44,51)	0.321
Laboratory test
WBC (10^9/L)	6.19 (5.29-7.47)	6.13 (5.21-7.44)	6.32 (5.49-7.53)	0.045
Eosinophil ratio (%)	2.30 (1.40,3.60)	2.30 (1.40,3.70)	2.20 (1.40,3.40)	0.280
eGFR (mL/min/1.73m²)	80(65-91)	81(65-91)	78(65-91)	0.469
BUN (mmol/L)	5.70 (4.80,6.90)	5.70 (4.80,6.90)	5.80 (4.90,7.00)	0.217
UA (μmol/L)	343(290,406)	341(290,402)	353(288,412)	0.251
FPG (mmol/L)	5.90(5.10-7.60)	5.80(5.00-7.30)	6.30(5.40-8.10)	＜0.001
HbA1c (%)	6.00(5.70-6.90)	6.00(5.60-6.80)	6.10(5.70-7.20)	0.012
TC (mmol/L)	3.42(2.89-4.21)	3.41(2.87-4.18)	3.52(2.94-4.40)	0.068
TG (mmol/L)	1.49(1.10-2.14)	1.44(1.07-2.01)	1.74(1.23-2.58)	＜0.001
TyG index	8.93(8.51-9.41)	8.88(8.45-9.30)	9.17(8.73-9.65)	＜0.001
LDL-C (mmol/L)	1.59(1.18-2.23)	1.59(1.18-2.18)	1.61(1.21-2.36)	0.651
N-HDL-C (mmol/L)	2.38(1.83-3.20)	2.33(1.80-3.12)	2.51(1.93-3.36)	0.009
HDL-C (mmol/L)	1.02(0.87-1.20)	1.03(0.89-1.20)	0.98(0.82-1.15)	0.002
Apolipoprotein A1 (g/L)	1.20(1.07-1.36)	1.21(1.07-1.36)	1.19(1.04-1.35)	0.218
Apolipoprotein B (g/L)	0.65(0.53-0.83)	0.64(0.52-0.82)	0.69(0.54-0.86)	0.035
Apolipoprotein E (g/L)	35(28-43)	34(27-42)	36(29-47)	0.006
CRP (mg/L)	0.95(0.40-3.50)	0.90(0.40-3.30)	1.00(0.50-3.70)	0.269
Troponin T (ng/ml)	0.01 (0.01-0.02)	0.01 (0.01-0.02)	0.01 (0.01-0.02)	0.029
NT-proBNP (ng/ml)	88.20 (43.43-212.53)	85.50 (42.80-198.00)	97.70 (45.70-231.00)	0.192
Coronary artery with lesions, n (%)
LM	93(9.85)	69(9.76)	24(10.13)	0.87
LAD	805(85.28)	602(85.15)	203(85.65)	0.849
LCX	577(61.12)	426(60.25)	151(63.71)	0.345
RCA	593(62.82)	431(60.96)	162(68.35)	0.042
Number of diseased coronary branches, n (%)
0	5(0.53)	5(0.71)	0(0.00)	0.341
1	245(25.95)	189(26.73)	56(23.63)
2	317(33.58)	242(34.23)	75(31.65)
3	319(33.79)	229(32.39)	90(37.97)
4	58(6.14)	42(5.94)	16(6.75)
Severity of coronary artery disease
Genisni scores	44.50 (27.00,69.00)	44.00 (26.00,68.00)	48.00 (27.00,72.00)	0.133
Total stent length	33(22,56)	33(22,56)	33(22,57)	0.614
Cardiovascular medications, n (%)
Anti-platelets	799(84.64)	407(86.23)	472(100.00)	0.176
Statins	754(79.87)	564(79.77)	190(80.17)	0.896
ACEI/ARB	465(49.26)	341(48.23)	124(52.32)	0.276
Beta-blockers	454(48.09)	335(47.38)	119(50.21)	0.451
CCB	291(30.83)	214(30.27)	77(32.49)	0.522
Diuretics	83(8.79)	56(7.92)	27(11.39)	0.102
Follow-up
Follow-up time (month)	32.18 (21.84-36.59)	32.40 (22.93-36.63)	31.67 (17.30-36.30)	0.046
Lost of follow-up, n (%)	129(13.67)	94(13.30)	35(14.77)	0.568
Death, n (%)	9(0.95)	0(0.00)	9(3.80)	＜0.001

Data are presented as median and 25% to 75% percentiles or frequency in percentage.

BMI = body mass index; MI = myocardial infarction; DM = diabetes mellitus; PCI = percutaneous coronary intervention; ASCAD = atherosclerosis coronary artery disease; LVEF = left ventricle ejection fraction; LVESV = left ventricular end systolic volume; LVEDV = left ventricular end diastolic volume; WBC = white blood cell count; eGFR = estimated glomerular filtration rate; BUN = blood urea nitrogen; UA = uric acid; FPG = fasting plasma glucose; HbA1c = glycated hemoglobin; TC = total cholesterol; TG = triglyceride; TyG index = triglyceride-glucose index; LDL-C = low-density lipoprotein-cholesterol; N-HDL-C = non-high-density lipoprotein-cholesterol; HDL-C = high-density lipoprotein-cholesterol; CRP = C-reactive protein; NT-proBNP = N-terminal pro B-type natriuretic peptide; LM = left main artery; LAD = left anterior descending artery; LCX = left circumflex artery; RCA = right coronary artery; ACEI = angiotensin-converting enzyme inhibitors; ARB = angiotensin receptor blockers; CCB = calcium channel blocker.

Table 2 | Cox regression of HbA1c and TyG index with different types of MACE

Indices	Before adjustment			After adjustment
Indices	HR	95% CI	P	HR	95% CI	P
HbA1c
Overall MACE	1.144	1.043-1.254	0.004	1.034	0.907-1.178	0.618
Cardiac death	1.755	1.086-2.836	0.022	218.561	0.000-4962554112.695	0.533
Myocardial infarction	1.199	0.824-1.743	0.343
Repeat revascularization	1.155	1.046-1.275	0.004	0.990	0.840-1.168	0.908
Heart failure	1.461	1.037-2.058	0.030	1.345	0.888-2.037	0.161
Stroke	1.214	0.830-1.774	0.317
Stent thrombosis	1.405	1.025-1.925	0.035	1.191	0.772-1.838	0.430
TyG index
Overall MACE	1.703	1.438-2.017	<0.001	1.574	1.192-2.079	0.001
Cardiac death	1.059	0.214-5.248	0.944
Myocardial infarction	1.114	0.486-2.554	0.799
Repeat revascularization	1.908	1.595-2.283	<0.001	1.379	1.080-3.557	0.027
Heart failure	1.214	0.497-2.962	0.670
Stroke	1.654	0.777-3.520	0.192
Stent thrombosis	1.493	0.687-3.243	0.312

MACE = major adverse cardiovascular events; HR = hazard ratio; CI = confidence interval; HbA1c = glycated hemoglobin; TyG index = triglyceride-glucose index.

Table 3 | Predictive efficacy of HbA1c, TyG index, and their combination for different types of MACE

Outcomes	HbA1c			TyG index			Combination
Outcomes	AUC	95% CI	P	AUC	95% CI	P	AUC	95% CI	P
Overall MACE	0.554	0.512-0.597	<0.001	0.614	0.573-0.656	0.012	0.616	0.575-0.657	<0.001
Cardiac death	0.798	0.535-1.000	0.074	0.560	0.114-1.000	0.719	0.901	0.817-0.986	0.016
Myocardial infarction	0.537	0.350-0.725	0.657	0.538	0.351-0.725	0.649	0.556	0.375-0.737	0.504
Repeat revascularization	0.562	0.517-0.606	0.007	0.639	0.597-0.682	<0.001	0.640	0.597-0.682	<0.001
Heart failure	0.698	0.539-0.857	0.031	0.542	0.352-0.733	0.644	0.723	0.572-0.873	0.002
Stroke	0.516	0.356-0.675	0.850	0.590	0.428-0.752	0.285	0.592	0.423-0.761	0.274
Stent thrombosis	0.683	0.520-0.846	0.037	0.575	0.422-0.727	0.394	0.672	0.508-0.836	0.049

MACE = major adverse cardiovascular events; AUC = area under the curve; CI = confidence interval; HbA1c = glycated hemoglobin; TyG index = triglyceride-glucose index.

Table 4 | TyG index-HbA1c grouping on the occurrence of MACE in patients with or without DM

Subgroups	DM			Non-DM
Subgroups	HR	95% CI	P	HR	95% CI	P
Low-HbA1c^†
High-HbA1c	1.107	0.596-2.055	0.748	1.501	0.989-2.279	0.056
Low-TyG index^†
High-TyG index	2.504	1.600-3.918	<0.001	1.915	1.373-2.671	<0.001
TyG index - HbA1c
Low - High / Low - Low	1.187	0.452-3.114	0.728	1.555	0.880-2.746	0.128
High - Low / Low - Low	3.375	1.068-10.663	0.038	1.975	1.359-2.869	<0.001
High - High / Low - Low	2.817	1.112-7.133	0.029	2.196	1.222-3.947	0.009
High - Low / Low - High	2.750	1.184~6.389	0.014	1.258	0.685~2.309	0.458
High - High / Low - High	1.519	1.293~1.98	<0.001	1.213	0.828~1.777	0.317
High - High / High - Low	1.244	0.556~2.782	0.594	1.117	0.599~2.084	0.728

DM = diabetes mellitus; CI = confidence interval; HbA1c = glycated hemoglobin; TyG index = triglyceride-glucose index.

^†Represents the reference subgroup in the comparison.

No competing interests reported.

Appendix.docx
Supplementary Information Supplemental Tables 1-2 and Supplemental Figure 1.

Predictive value of combined triglyceride-glucose index and glycated hemoglobin for long-term cardiovascular outcomes in stable coronary artery disease

Status:

Version 1

Abstract

Figures

Background

Methods

Study population and follow-up

Data collection and definition

PCI

Endpoints

Statistical analyses

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1