High Triglyceride-Glucose Index is Associated with Poor Cardiovascular Outcomes in Non-Diabetic ACS Patients with LDL-C Below 1.8mmol/L

Background: To evaluate the prognostic value of triglyceride glucose (TyG) index in non-diabetic acute coronary syndrome (ACS) patients with low-density lipoprotein cholesterol (LDL-C) below 1.8mmol/L. Methods: A total of 1655 non-diabetic ACS patients with LDL-C below 1.8mmol/L were included in the analysis. Patients were stratied into 2 groups. Incidence of acute myocardial infarction (AMI), infarct size in patients with AMI, and major adverse cardiac and cerebral event (MACCE) during a median of 35.6-month follow-up were determined and compared between the 2 groups. The TyG index was calculated using the following formula: ln [fasting triglycerides (mg/dL) ×fasting plasma glucose (mg/dL)/2]. Results: Compared with the TyG index <8.33 group, the TyG index ≥ 8.33 group had signicantly higher incidence of AMI (21.2% vs. 15.2%, p =0.014) and larger infarct size in patients with AMI (pTNI: 10.4 vs. 4.8 ng/ml, p =0.003; pCKMB: 52.8 vs. 22.0 ng/ml, p =0.006; pMyo: 73.7 vs. 46.0 ng/ml, p =0.038). Although there was no signicant difference in mortality between the 2 groups, the incidence of revascularization of TyG index ≥ 8.33 group was signicantly higher than that of TyG index <8.33 group (8.9% vs. 5.0%, p =0.035). Multivariable Cox regression revealed that the TyG index was positively associated with revascularization [HR (95% CI): 1.67 (1.02,2.75), p=0.043]. Conclusion: In non-diabetic ACS patients with LDL-C below 1.8mmol/L, the high TyG index level was associated with higher incidence of AMI, larger infarct size, and higher incidence of revascularization. The high TyG index level might be a valid predictor of subsequent revascularization. Trial registration: retrospectively registered. [21, , stroke [23] , carotid atherosclerosis [24] , coronary artery calcication [7] and coronary artery stenosis


Introduction
Acute coronary syndrome (ACS) is the leading cause of morbidity and mortality from cardiovascular disease worldwide [1,2] . Therefore, it is crucial to identify patients at high risk of developing major adverse cardiac and cerebral event (MACCE) that may contribute to optimal management. Insulin resistance (IR), a hallmark of metabolic syndrome (MetS), not only associated with an increased risk of cardiovascular disease, but also signi cantly correlated with a higher risk of MACCE [3,4] . However, direct measurement of IR, including the hyperinsulinemic-euglycemic clamp and the homeostasis model assessment of IR (HOMA-IR), are too complex and expensive to be used in large-scale epidemiological studies and clinical practice [5] . Therefore, we urgently need a simple, accessible and reliable index to quantitively evaluate IR.
High levels of triglyceride (TG) and fasting plasma glucose (FPG) are the important components of MetS. Recently, the triglyceride-glucose (TyG) index, which combines TG and FPG levels, has been proved as a reliable surrogate marker of IR in clinical practice [6] . What's more, the TyG index has been found to be well correlated with coronary artery disease (CAD) [7][8][9] . Luo et al. rstly reported a positive correlation between the TyG index level and the incidence of MACCE in patients with ST-elevation myocardial infarction (STEMI) who underwent percutaneous coronary intervention (PCI) [10] . Mao et al. also found that in patients with non-ST-segment elevation acute coronary syndrome (NSTE-ACS), high TyG index group showed signi cantly increased risk of major adverse cardiac event (MACE) compared with the low TyG index group [11] . To the best of our knowledge, the relationship between TyG index and cardiovascular outcomes in non-diabetic ACS patients with LDL-C below 1.8 mmol/L, is still unknown. Our study was to ll this knowledge gap. Here, we aimed to investigate not only the relationship between TyG index and patient characteristics during hospitalization, but also the predictive value of TyG index on the cardiovascular outcomes.

Study population
Patients' records in the Cardiovascular Center of Beijing Friendship Hospital Database Bank were screened. As shown in Figure 1, the records of 10,216 ACS patients undergoing coronary angiography from December 2012 to March 2020 in our center were screened. Of the 11,110 patients, 9455 were excluded according to the exclusion criteria, which were 1) with diabetes mellitus, 2) with LDL-C≥1.8mmol/L, 3) with severe valvulopathy or cardiomyopathy, 4) with acute infections disease, rheumatic disease, hematological disease, or neoplastic disease, and 5) lacking clinical or follow-up data. Finally, 1655 patients were included in this analysis. According to the median value of TyG index level, 1655 patients were strati ed into 2 groups (TyG index <8.33 group, n = 829 and TyG index ≥8.33 group, n = 826). All patients were followed up till April 30, 2020 with a median follow up of 35.6 (IQR: 13.2, 51.8) months.

Data collections and de nitions
The data collection process was approved by the Institutional Review Board of Beijing Friendship Hospital a liated to Capital Medical University and was in accordance with the Declaration of Helsinki.
Patients' demographics, medical and medication history, laboratory test results, echocardiographic and angiographic evaluation results were collected and veri ed using an electronic medical recording system. The outcomes from MACCE were collected and recorded during clinical follow-up visits.
ACS contains unstable angina pectoris (UAP) and acute myocardial infarction (AMI). UAP was diagnosed in patients with unstable chest discomfort (rest, new onset, or worsening of angina) and without the elevation of myocardial necrosis markers. AMI was de ned as chest pain with new ST-segment changes and elevation of myocardial necrosis markers to at least twice of the upper limit of the normal range. MACCEs included all-cause death, non-fatal MI, non-fatal stroke, revascularization, and cardiac rehospitalization (admission because of angina or heart failure). All-cause death was de ned as the incidence of cardiac death or non-cardiac death. Cardiovascular(CV) death was de ned as fatal myocardial infarction, fatal stroke, sudden death, and other cardiovascular death. Non-fatal stroke, including ischemic and hemorrhagic stroke, was de ned as cerebral dysfunction caused by cerebral vascular obstruction or sudden rupture and was diagnosed based on signs of neurological dysfunction or evidence of brain imaging. Any coronary revascularization was de ned as a revascularization of the target vessel or non-target vessels. Cardiac rehospitalization refers to rehospitalization for angina pectoris or heart failure. The TyG index was calculated as ln [fasting TG (mg/dL) × fasting plasm glucose (FPG, mg/dL) /2] [12] .

Statistical analyses
Continuous variables are presented as mean ± standard deviation (SD) or median (IQR). Comparisons between the two study groups were analyzed by Student's t-test or Mann-Whitney U-test. Categorical variables are expressed as number and percentage and compared using the Pearson chi-square test or Fisher's exact test. To control confounding factors, we performed propensity score matching. The cumulative incidence of MACCE was estimated by Kaplan-Meier survival curves. A multivariable Cox regression analysis was performed to identify independent predictors for MACCE. Baseline variables that were signi cantly correlated with outcomes by univariate analysis were used in the multivariate model.
Correlation analysis among variables was also taken into consideration in the multivariate analysis. All analyses were twotailed and P value <0.05 was considered statistically signi cant. Data were analyzed using the statistical analysis software IBM SPSS statistics 24.0.
Propensity score matching Propensity score matching (PSM) was used to reduce selection bias in this study. The matching process was conducted with a minimum-distance scoring method and a 1-to-1 match between the TyG index ≥8.33 group and the TyG index <8.33 group. In this study, propensity scores were calculated through a binary logistic regression model, including covariates of age, sex, body mass index (BMI), systolic blood pressure (SBP), hemoglobin (HGB), albumin, creatinine, glycated hemoglobin (HbA1c), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), history of smoking and stroke, previous medication history including beta-blocker and statins, and statins treatment during hospitalization. Ultimately, 505 TyG index ≥8.33 patients were individually 1:1 matched to 505 TyG index <8.33 controls using nearest available score matching. The statistical analysis software SPSS version 24.0 was used for the matching.

Patient characteristics
As shown in Figure 1, of the 1655 eligible patients, 829 patients with the TyG index <8.33 and 826 patients with the TyG index≥8.33. Comparing with the TyG index <8.33 group, the TyG index≥8.33 group showed signi cantly higher BMI, younger, lower SBP, higher percent of smoker, lower percent of stroke, and signi cantly more likely to receive beta-blocker or statins before the hospital admission for ACS. In-hospital medical and interventional treatments were similar between the 2 groups except that signi cantly higher patients treated with beta-blocker in the TyG index≥8.33 group than in the TyG index <8.33 group during hospitalization. Laboratory values showed that the TyG index≥8.33 group had signi cantly higher white cell count, HGB, FPG, HbA1c, albumin, creatinine, TG and LDL-C than the TyG index <8.33 group. The results of coronary angiography and echocardiography showed no signi cant difference between the 2 groups (Table 1).
Propensity score matching Propensity scores of 505 TyG index ≥8.33 patients were 1:1 matched to 505 TyG index <8.33 patients. There were no signi cant differences in baseline clinical characteristics and medical history between the PSM TyG index ≥8.33 and TyG index <8.33 groups except that the PSM TyG index ≥8.33 group had signi cantly higher FPG and TG ( Table 1).
The Kaplan-Meier curves show that the TyG index ≥8.33 group had signi cantly higher cumulative rate of subsequent revascularization than the TyG index<8.33 group (Figure 3). The cumulative rate of all cause death, CV death, non-fatal MI, non-fatal stroke, cardiac rehospitalization, and composite MACCE were not statistically different between the 2 groups.

Risk factors for subsequent revascularization
Univariate and multivariate analysis results and predictors for revascularization are presented in Supplemental le.
Univariate analysis revealed that TyG index, FPG, multi-vessel/ left main (LM) coronary artery lesions, and PCI/Coronary Artery Bypass Graft (CABG) treatment during hospitalization were risk factors for revascularization in patients with ACS (all p <0.05). Correlation analysis displayed that FPG and TyG index had a high correlation (p <0.001). Therefore, FPG were not included in the multivariate model. In addition, PCI/CABG treatment during hospitalization was signi cantly correlated with multi-vessel/LM coronary artery lesions (p <0.001). Therefore, PCI/CABG treatment during hospitalization was also not included in the multivariate model.

Independent association of TyG index with subsequent revascularization in different subgroups
As shown in Figure 4, the independent predictive effect of TyG index on subsequent revascularization was mainly re ected in the subgroups of male gender, age<65 years, BMI <25kg/m 2 , smoker, eGFR ≥60ml/min/1.73m 2 , HDL-C <1.01mmol/L, and LVEF ≥50%.

Discussion
To the best of our knowledge, the current study was the rst to investigate whether TyG index could be associated with patient characteristics during hospitalization and subsequent cardiovascular outcomes in non-diabetic ACS patients with LDL-C below 1.8 mmol/L. Our main ndings include: in non-diabetic ACS patients with LDL-C below 1.8 mmol/L, (1) the high TyG index group had signi cantly higher incidence of AMI and larger infarct size than the low TyG index group, (2) the incidence of subsequent revascularization of the high TyG index group was signi cantly higher than that of the low TyG index group, (3) the high TyG index was an independent predictor of subsequent revascularization, and (4) Moreover, the independent predictive effect of TyG index on revascularization was mainly re ected in the subgroups of male gender, age < 65 years, BMI < 25 kg/m 2 , smoker, eGFR ≥ 60 ml/min/1.73 m 2 , HDL-C < 1.01 mmol/L, and LVEF ≥ 50%.
IR, a hallmark of MetS, is de ned as a decrease in the e ciency of insulin in promoting glucose uptake and utilization. IR can induce glucose metabolism imbalance, which leads to chronic hyperglycemia and then in turn triggers oxidative stress and causes in ammatory responses. In addition, IR can alter systemic lipid metabolism, including increased TGs levels, decreased HDL-C levels, increased small dense low-density lipoproteins, and excessive postprandial lipemia. Moreover, IR can also cause endothelial dysfunction by decreasing nitric oxide production from endothelial cells and increasing procoagulant factor release. Therefore, it is easy to understand that IR has been proven to contribute to the progression of cardiovascular disease [13,14] due to above mechanisms. However, traditional measurement of IR, including the hyperinsulinemic-euglycemic clamp and HOMA-IR, are too complex and expensive to be used in clinic practice on a large scale. In recent years, researchers have tried to nd a simpler and more valid surrogate marker of IR.
As is known to all, high levels of TG and FPG are the components of MetS. Recently, the TyG index, a composite indicator composed of TG and FPG, has been demonstrated as a reliable marker of IR [6,12,15] and has a high sensitivity and speci city for identifying MetS [16] . Previous studies reported that the TyG index is a simple, cost-effective surrogate marker of IR compared to HOMA-IR [17,18] . It was demonstrated that TyG index was a useful predictor of Type 2 diabetes mellitus (T 2 DM) [19,20] which contributed to cardiovascular disease risk. Studies have also shown an association of TyG index with arterial stiffness [21,22] , stroke [23] , carotid atherosclerosis [24] , coronary artery calci cation [7] and coronary artery stenosis [8] .
Subsequently, several studies were conducted and found a positive relationship between TyG index and cardiovascular disease. It has been reported that the TyG index may contribute to the early identi cation of apparently healthy individuals at high risk for cardiovascular events [25][26][27] . Jin et al. revealed that TyG index was positively associated with cardiovascular events risk [HR(95% CI): 1.36(1.10,1.69), p = 0.005] in patients with stable CAD [28] . The ndings of Luo et al. showed that the TyG index was signi cantly associated with an increased risk of MACCE in STEMI patients within 1 year after PCI [HR(95% CI):1.53(1.01,2.06), p = 0.003] [10] . It was demonstrated that the TyG index was an independent predictor of MACEs [HR(95% CI):1.88(1.13,3.12), p = 0.015] in patients with NSTE-ACS [11] . Jin et al. [29] and Su et al. [30] found that both TyG index and HbA1c could predict cardiovascular outcomes in T 2 DM patients while TyG index might be better.
Unfortunately, no data is currently available with regard to the effects of TyG index on clinical outcomes in non-diabetic ACS patients, especially those with LDL-C lower than 1.8 mmol/L. Recently, Alizargar et al. questioned the conclusion that TyG index can be used to predict cardiovascular events in patients with CAD, considering that it might be in uenced by diabetes and the hyperlipidemic state that led to cardiovascular disease [31] . Therefore, in order to avoid the mixed effects of diabetes and hyperlipemia on cardiovascular events, we only analyzed ACS patients with non-diabetic and LDL-C below 1.8 mmol/L, which was the novelty of this study.
Although we found that higher TyG index levels had no signi cant effect on the incidence of all-cause death, CV death and composite MACCE. However, our study indicated an association between higher TyG index levels and an increased risk of subsequent revascularization for the rst time, and the TyG index might be a valid predictor of subsequent revascularization in non-diabetic ACS patients with LDL-C below 1.8 mmol/L. This nding might be greatly of interest. A high TyG index level still has a signi cant impact on the risk of subsequent revascularization, after all, patients without diabetes and with LDL-C below 1.8 mmol/L seem to be a relatively low-risk group for cardiovascular events in follow-up. In addition, we found that the independent predictive effect of TyG index on subsequent revascularization was mainly re ected in the subgroups of male gender, age < 65 years, BMI < 25 kg/m 2 , smoker, eGFR ≥ 60 ml/min/1.73 m 2 , HDL-C < 1.01 mmol/L, and LVEF ≥ 50%. This nding implied that using TyG index for early risk strati cation in the above subgroups may have more important clinical signi cance. Moreover, we also found that the high TyG index group had signi cantly higher incidence of AMI and larger infarct size compared with the low TyG index group, which were not reported in previous studies.
Luo et al [10] and Mao et al [11] demonstrated that the TyG index was an independent predictor of cardiovascular events in STEMI and NSTE-ACS population, respectively. The cut-off value of the former was TyG ≥ 9.608 and the latter was TyG ≥ 8.805. In this study, we found that the high TyG index level was an independent predictor of subsequent revascularization in non-diabetic ACS patients with LDL-C < 1.8 mmol/L. We proposed the cut-off point of TyG ≥ 8.33, which was signi cantly lower than that in previous studies. The main reason is that patients with diabetes and hyperlipemia have been excluded in this study, whereas previous studies have not. Therefore, the cut-off value of the TyG index in this study might be lower than that in previous studies.

Limitations
There were several limitations in the present study. Firstly, this was a single-center study, the sample size might be not large enough and the follow-up time might be not long enough; thus, generalization of the ndings should be cautious. Secondly, laboratory parameters were only measured once after hospital admission, which could cause potential bias due to measurement error. Thirdly, this was a retrospective observational study. The information on the levels of TyG index during follow-up was limited. Hence, prospective cohort studies with larger sample and longer follow-up time are required to con rm our ndings.

Conclusions
In non-diabetic ACS patients with LDL-C below 1.8 mmol/L, the high TyG index group had signi cantly higher incidence of AMI and larger infarct size than the low TyG index group. In addition, the incidence of subsequent revascularization of the high TyG index group was signi cantly higher than that of the low TyG index group, and the high TyG index level was an independent predictor of subsequent revascularization.

Declarations
Ethics approval and consent to participate The study data collections were approved by the Institutional Review Board of Beijing Friendship Hospital a liated to Capital Medical University, and written informed consent was obtained from all patients.

Consent for publication
Consent to publish from the participant to report individual patient data: not applicable (no patient identi er or personalized data shown).

Competing Interests
The authors declare that they have no competing interests. Authors' contributions YZ performed study, statistical analysis and wrote manuscript. XSD, BH, QBL and HG participated in study data collection. HC contributed discussion and edited manuscript. XQZ designed study and revised manuscript. WPL designed study, performed statistical analysis and edited manuscript. HWL provided funding support, designed study and reviewed manuscript. All authors read and approved the nal manuscript.     TyG, triglyceride-glucose index; CV: cardiovascular ; MI, myocardial infarction; MACCEs, major adverse cardiac and cerebral event; HR, hazard ratio; CI, con dence interval.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.