TYG Index as a Novel Predictor of Clinical Outcomes in Advanced Chronic Heart Failure with Renal Dysfunction Patients

Introduction

Heart failure (HF) is the final stage of various heart diseases. In recent years, with improvements in the treatment of related diseases such as acute myocardial infarction, the mortality rate of patients has significantly decreased, while the incidence of HF has been steadily rising.¹ Due to the reduced circulatory capacity caused by advanced chronic heart failure (acHF), it often leads to secondary organ complications, with renal dysfunction (RD) being one of the most common complications of acHF. The conflicting treatment principles between these two organs, coupled with the propensity of both to cause systemic complications, often result in frequent readmissions for patients with both CHF and RF, greatly impacting their quality of life.^2,3

Metabolic syndrome can be regarded as the root cause of cardiovascular diseases. In the case of the most common cardiovascular disease, coronary heart disease, prolonged abnormalities in glucose and lipid metabolism lead to endothelial dysfunction and lipid accumulation, triggering the formation of atherosclerosis.⁴ Compared to other patients with coronary heart disease, those with glucose and lipid metabolism abnormalities often exhibit more severe coronary artery lesions and a worse clinical prognosis.⁵ Patients with glucose and lipid metabolism abnormalities are also more prone to postoperative complications such as cerebral infarction and in-stent restenosis.^6,7

For patients with chronic RD, although there is currently no term like “metabolic syndrome renal disease”, the association between the two has been widely recognized. Diabetic nephropathy is a condition secondary to diabetes, involving renal damage due to long-term poorly controlled blood sugar levels leading to microvascular changes. Additionally, research has found that patients with prolonged elevation of blood lipids often have accompanying glomerulosclerosis.⁸

To comprehensively reflect the level of glucose and lipid metabolism in the patient’s body, the TyG index has emerged. With the advancement of related research, the TyG index is currently considered to primarily reflect the level of insulin resistance (IR) in the body and is regarded as an excellent alternative assessment indicator. IR is believed to be the result of the combined effects of chronic inflammation, lipid metabolism abnormalities, or glucose abnormalities in the body. Currently, the prognostic predictive ability of the TyG index has been validated in patients with HF or RD. Through this analysis of clinical data from patients admitted to the department of cardiology in first affiliated hospital of Zhengzhou university, we have found that the TyG index is a novel tool for predicting the occurrence of MACEs and ACM in patients with advanced chronic HF combined with RD.

Method

Study and Population

This study included acHF patients admitted to the Department of Cardiology at the First Affiliated Hospital of Zhengzhou University from September 2019 to December 2020.

Inclusion Criteria

(1) Diagnosis of acHF, based on the 2021 ESC criteria for advanced chronic heart failure,⁹ which includes the following criteria even with optimal treatment: (A) Severe and persistent heart failure symptoms, at least NYHA class III or IV; (B) Severe cardiac dysfunction meeting any of the following criteria: (a) LVEF ≤ 30%, (b) isolated right heart failure, (c) severe valve deformities not amenable to surgical treatment, (d) severe congenital abnormalities not amenable to surgical treatment, (e) elevated NT-proBNP with concomitant HFpEF; (C) Patients requiring intravenous diuretics due to pulmonary or systemic congestion or needing positive inotropic or vasopressor agents due to low cardiac output or malignant arrhythmias leading to more than one unscheduled outpatient visit within 12 months; (D) Impaired exercise capacity due to cardiac disease, inability to exercise, 6MWT < 300m, or pVO2 < 12 mL/kg/min, or <50% of predicted value. (2) Diagnosis of renal dysfunction with eGFR < 90 mL/(min/1.73 m²). (3) Patients with at least two hospitalizations due to heart failure.

Exclusion Criteria

(1) Age <18 years old; (2) first episode of acute heart failure; (3) hospitalization duration <2 days; (4) patients with infectious diseases, malignant tumors, primary renal diseases, congenital heart disease or autoimmune diseases; (5) incomplete clinical records.

Based on the aforementioned criteria, a total of 815 patients were included. Out of these, 98 patients were lost to follow-up, resulting in a loss rate of 12.0%. Data were gathered and documented from the Hospital Information System, encompassing information such as gender, age, relevant medical histories, blood test results, and records of medication treatments.

Definitions

According to ESC recommendations,¹⁰ hypertension is defined as having a systolic blood pressure (SBP) of ≥140 mmHg and/or diastolic blood pressure (DBP) of ≥90 mmHg measured on at least three occasions in two separate places, or the use of any antihypertensive medication. Diabetes is defined as having a fasting blood glucose level of ≥7.0 mmol/L or a post-load blood glucose level of ≥11.1 mmol/L, or the use of antidiabetic medications.¹¹ Smoking and alcohol status are defined as current or past tobacco use and any alcohol consumption within the past six months.

Endpoints

The primary endpoint for patients is major adverse cardiovascular events (MACEs), which includes cardiovascular death, non-cardiovascular death, rehospitalization due to worsening HF, heart transplantation, use of mechanical circulatory support, malignant arrhythmias, myocardial infarction, and rehospitalization for unstable angina. The secondary endpoints is all-cause mortality (ACM).

Follow-Up

Clinical data of eligible patients were extracted from inpatient medical records and outpatient records. Patients receive regular follow-up after discharge, which included telephone follow-up and office visits. The final follow-up for this study was in March 2022, with a median follow-up duration of 21.58 (17.98–25.39) months. All data collection and follow-up were carefully assessed and executed by trained clinical physicians (Figure 1).

Figure 1 The flow chart of participants inclusion.

Calculation of TyG Index

The TyG index was calculated by fasting blood glucose and fasting triglyceride at admission:

Statistical Analysis

Data analysis was conducted using SPSS 26.0 (IBM Corp., Armonk, NY, USA). Receiver Operating Characteristic (ROC) curves were constructed for both endpoints as state variables, and the corresponding areas under the curve (AUC) and Youden’s index were calculated to determine the optimal cutoff values for TyG index. Patients were then categorized into high-TyG and low-TyG groups, with the best cutoff value for the primary endpoint is 8.48 (AUC = 0.759, P < 0.001). The best cutoff value for the secondary endpoint is 8.50 (AUC = 0.700, P < 0.001).

Categorical variables were presented as case numbers (percentages) and analyzed using the χ² test. Continuous variables were expressed as means ± standard deviation (SD) or as medians (interquartile range). For continuous variables, the t-test was used if they satisfied both normal distribution and homogeneity of variances; otherwise, the Mann–Whitney U-test was employed. We employed Multivariate Cox regression analysis to identify independent variables influencing prognosis. Subsequently, we constructed sequential models to assess the added prognostic value of these variables. The inclusion of additional factors in each step was considered statistically significant when the change in log likelihood associated with each model reached a significance level of P<0.05. Kaplan-Meier survival analysis was employed to assess long-term survival rates. Two-sided p-values were used, and significance was defined as P<0.05.

Results

Baseline Characteristics

Among the 717 patients included, the TyG cutoff value for the primary endpoint (MACEs) was determined to be 8.48, with an area under the curve (AUC) of 0.759. Sensitivity was 0.836, specificity was 0.588, and the Youden’s index was 0.424 (Figure 2). Patients were categorized into low-TyG (TyG<8.48, n=318) and high-TyG (TyG≥8.48, n=399) groups, and their baseline characteristics are presented in Table 1. Between the two groups, differences were observed in terms of gender, demographic characteristics (coronary heart disease, diabetes, hyperlipidemia and chronic renal disease), white blood cell, platelet, neutrophil, fibrinogen, blood urea nitrogen, creatinine, uric acid, eGFR, albumin, globulin, t total cholesterol, triglyceride, HDL-C, LDL-C, troponin, C-reactive protein, blood glucose, glycated hemoglobin, and medication history (clopidogrel, Beta-blocker, diuretics and statins) (p<0.05). However, certain common risk factors associated with cardiovascular disease, such as advanced age, smoking history, and hypertension, showed no significant differences.

Table 1 Clinical and Laboratory Characteristics According to the TyG (MACEs)

Figure 2 ROC curve of TyG predicting long‐term MACEs in patients with advanced chronic heart.

In the grouping based on ACM as the endpoint, the TyG cutoff value was determined to be 8.50, with an area under the curve (AUC) of 0.700. Sensitivity was 0.821, specificity was 0.538, and the Youden’s index was 0.359 (Figure 3). Patients were divided into low-TyG (TyG<8.50, n=330) and high-TyG (TyG≥8.50, n=387) groups, and their baseline characteristics are presented in Table 2. Between the two groups, differences were observed in terms of gender, demographic characteristics (coronary heart disease, diabetes and hyperlipidemia), white blood cell, platelet, neutrophil, fibrinogen, creatinine, albumin, globulin, t total cholesterol, triglyceride, HDL-C, LDL-C, troponin, blood glucose, glycated hemoglobin, and medication history (clopidogrel, ARNI and statins) (p<0.05). However, certain common risk factors associated with cardiovascular disease, such as advanced age, smoking history, and hypertension, showed no significant differences. Failure with renal dysfunction.

Table 2 Clinical and Laboratory Characteristics According to the TyG (ACM)

Figure 3 ROC curve of TyG predicting long‐term ACM in patients with advanced chronic heart failure with renal dysfunction.

Relationship Between TyG and MACEs

We constructed univariate Cox models for each predictive variable (Table 3) and created a multivariate Cox regression analysis based on the results of the univariate COX regression analysis and traditional clinical prognostic factors. The results showed that age>60, a history of arrhythmia, elevated CRP, and high TyG were risk factors for the occurrence of MACEs in patients with acHF and RD.

Table 3 Univariate COX Regression Analysis Results for MACEs

On the other hand, low-density lipoprotein cholesterol (LDL-C) was identified as a protective factor (Table 4). Patients with high TyG levels had a 440.2% increase in the risk of experiencing MACEs (hazard ratio [HR] = 5.402; 95% confidence interval [CI], 3.771–7.739; P < 0.001).

Table 4 Univariate COX Regression Analysis Results for ACM

Relationship Between TyG and ACM

We constructed univariate Cox models for each predictive variable (Table 5) and created a multivariate Cox regression analysis based on the results of the univariate COX regression analysis and traditional clinical prognostic factors. The results showed that age>60, a history of arrhythmia, elevated CRP, and high TyG were risk factors for the occurrence of ACM in patients with acHF and RD.

Table 5 Multivariate Cox Regression Analysis Results for MACEs

Conversely, a history of hypertension and LDL-C were identified as protective factors (Table 6). Patients with high TyG levels had a 406.2% increase in the risk of experiencing ACM (hazard ratio [HR] = 5.062; 95% confidence interval [CI], 3.268–7.839; P < 0.001).

Table 6 Multivariate Cox Regression Analysis Results for ACM

Kaplan-Meier Survival Analysis

Kaplan-Meier survival analysis was conducted to predict the occurrence rate of MACEs (Figure 4) and ACM (Figure 5) in patients.

Figure 4 Kaplan–Meier curves for survival analysis of MACEs-free survival (Log rang p<0.001).

Figure 5 Kaplan–Meier curves for survival analysis of ACM-free survival (Log rang p<0.001).

Discussion

In recent years, with the advancement of molecular biology technologies, new biomarkers have emerged, such as miRNA and genomic sequencing.^12,13 Although most of these biomarkers are still in the experimental research stage, they often offer advantages of high sensitivity and specificity. However, their adoption is limited by the lack of more in-depth clinical studies or their relatively higher costs. Some ratio-based biological markers, on the other hand, not only have the advantage of being cost-effective but also significantly improve diagnostic efficacy compared to traditional single biomarkers. The use of ratio-based biological markers in clinical practice has a long history, such as BMI. In recent years, some newly discovered biological ratios have also been validated for their clinical utility through multiple clinical studies.¹⁴ These indicators can systematically reflect certain clinically significant conditions that impact disease. For instance, the SII can systematically indicate the overall level of inflammation in the body, and was found to be useful for predicting outcomes in patients with coronary heart disease.¹⁵ Additionally, the Atherogenic Index of Plasma (AIP) can predict plaque burden in coronary artery disease patients undergoing IVUS examination.¹⁶

TyG index, discovered in recent years, is a novel biomarker believed to have the ability to reflect the level of insulin resistance (IR) in the body.¹⁷ IR refers to a pathological condition characterized by a reduced response to insulin and often occurs in patients with type 2 diabetes mellitus (T2DM).¹⁸ Besides diabetic patients, IR is also widely prevalent in conditions such as hyperlipidemia or hypertension and is considered a risk factor and predictor for a variety of cardiovascular diseases and renal diseases.^19,20 Existing research suggests that IR is a significant pathophysiological basis for metabolic syndrome and serves as a hallmark of systemic metabolic dysregulation or inflammatory response. When insulin-responsive cells are exposed to hormones secreted under conditions of hypoxia, excessive glucose, lipids, or obesity, various cellular stress responses coordinate to induce the occurrence of IR.²¹ IR is a critical early pathophysiological process in atherosclerosis, which can promote plaque formation through various mechanisms, such as facilitating migration of vascular smooth muscle cells to the endothelium, exacerbating local inflammatory responses, damaging endothelial function, and subsequently promoting plaque formation.²² In addition to its involvement in cardiovascular disease occurrence, IR is also associated with renal function impairment. Studies have found that insulin receptors are widely present on the surface of cells throughout the kidney. When IR occurs, insulin receptors in the kidney are simultaneously downregulated, affecting blood pressure and local vascular function through multiple pathways such as promoting sodium retention, downregulating eNOS production, and reducing NO bioavailability, thereby adversely affecting renal function.²¹

Currently, the commonly used method for assessing IR in clinical practice is the glucose tolerance test, but its complexity often results in poor patient compliance. HOMA-IR is a relatively convenient and accurate method for evaluating IR.²³ However, its accuracy is significantly reduced in patients receiving insulin treatment or those with complete loss of pancreatic beta-cell function, limiting its clinical applicability. In comparison to the previous two assessment methods, the TyG index is highly accurate, simple, and unaffected by insulin treatment, making it the focus of widespread attention in recent years.²⁴

The association between the TyG index and cardiovascular diseases has been widely recognized. As mentioned earlier, since the TyG index primarily reflects the level of IR in the body, and IR is closely associated not only with the onset of type 2 diabetes mellitus but also with cardiovascular disease risk factors such as hyperlipidemia, hypertension, or coagulation abnormalities.^25,26 Currently, the TyG index has been confirmed to be closely associated with the prognosis of healthy individuals or patients with coronary heart disease. Yang et al studied 592,616 individuals and found an association between the TyG index and ischemic stroke.²⁷ Erdogan et al suggested that the TyG index could serve as a predictor of adverse cardiovascular outcomes in patients with chronic coronary heart disease.²⁸ A meta-analysis involving 10,164 patients found a strong correlation between high TyG index and post-PCI major adverse cardiovascular events (MACEs), non-fatal myocardial infarction, and all-cause mortality.²⁹

Clinically, there are many factors that can influence the TyG index. Theoretically, any disease or medication that affects triglyceride levels and fasting blood glucose may affect the predictive ability of the TyG index. However, Sun et al found that regardless of whether patients have diabetes, the predictive ability of the TyG index for post-PCI MACEs risk remained consistent.²⁹ About 80% of obese patients have hypertriglyceridemia, and almost all cardiovascular diseases are associated with obesity, which adversely affects patient prognosis.³⁰ However, in patients with acHF accompanied by RD, weight gain due to water and sodium retention and worsening nutritional status may lead to a decrease in fat content, making BMI not a true reflection of obesity status. Therefore, for these patients, the potential application of the TyG index is higher because it can more accurately reflect the patient’s glucose and lipid metabolism and insulin resistance capabilities.

The Tehran Lipid and Glucose Study indicated that the TyG index can serve as an alternative marker to HOMA-IR and is associated with an increased risk of chronic renal disease and decreased renal function.³¹ Additionally, researchers have found a close association between elevated TyG index and decreased renal function in patients with diabetes and acute coronary syndrome.^32,33 Since deteriorating renal function is a significant risk factor for poor prognosis in HF patients, with approximately 40% of HF patients experiencing renal impairment,³⁴ we hypothesize that there is also an association between the TyG index and patient prognosis in patients with acHF accompanied by RD. This study, by revealing the association between the TyG index and patient outcomes, found that patients with a high TyG index had significantly worse clinical outcomes. The TyG index can be used as a predictive factor for patient prognosis. We believe that the reasons for the poorer prognosis in patients with high TyG levels may include: (1) Dyslipidemia is a crucial risk factor for atherosclerosis and one of the initiating factors for HF. The increased blood viscosity due to elevated lipid levels imposes an additional burden on the cardiac of HF patients.³⁵ (2) Prolonged hyperlipidemia can induce oxidative stress in renal tubular epithelial cells and lead to glomerulosclerosis, adversely affecting renal function.⁸ (3) Abnormal glucose metabolism resulting in a hyperglycemic state induces oxidative stress, impairs vascular relaxation and contraction, thereby contributing to the occurrence of cardiovascular diseases and renal failure.^36,37

Conclusions

We identified the TyG index as an independent risk factor for MACEs and ACM in patients with acHF and RD, demonstrating significant predictive potential. Given that this study is retrospective and conducted at a single center, further multicenter studies are still needed to validate the conclusions.