Triglyceride glucose index–body mass index as a predictor of coronary artery disease severity in patients with H-type hypertension across different glucose metabolic states

Abstract

Background

The triglyceride glucose-body mass index (TyG-BMI) is considered to be a reliable surrogate marker of insulin resistance (IR). However, limited evidence exists regarding its association with the severity of coronary artery disease (CAD), particularly in hypertensive patients with different glucose metabolic states, including those with H-type hypertension. This study aimed to investigate the relationship between TyG-BMI and CAD severity across different glucose metabolism conditions.

Methods

This retrospective cohort study included 1537 hypertensive patients who underwent coronary angiography. The TyG-BMI was categorized into tertiles and analyzed using logistic regression models and restricted cubic spline (RCS) models to assess its association with multi-vessel CAD. Receiver operating characteristic (ROC) curves were used to evaluate the predictive value of TyG-BMI in detecting the severity of CAD in different glucose metabolism states, including normal glucose regulation (NGR), pre-diabetes mellitus (Pre-DM), and diabetes mellitus (DM). The above method has also been applied to populations of H-type hypertension patients.

Results

The TyG-BMI was significantly associated with the severity of multi-vessel CAD in hypertensive patients (Odds ratio [OR] 1.043, 95% CI 1.032–1.053). In the diabetic subgroup, after adjusting for risk factors, the risk of multi-vessel CAD in the T3 groups was 3.836-fold (95% CI 1.763–8.347; P = 0.001) higher than in the T1 group, with a non-linear dose–response relationship (P for non-linearity = 0.017). In H-type hypertension patients, the TyG-BMI was also significantly correlated with multi-vessel CAD (OR 5.248, 95% CI 1.821–15.126, P = 0.002) in the DM group. The ROC analysis revealed that TyG-BMI had the highest predictive value for multi-vessel CAD in diabetic patients, with an AUC of 0.720 (95% CI 0.661–0.780, P < 0.001).

Conclusions

The TyG-BMI serves as a robust predictor of CAD severity in hypertensive patients, particularly those with diabetes and H-type hypertension. And the non-linear dose–response relationship between TyG-BMI and multi-vessel CAD in diabetic patients underscores its potential clinical utility. This index could serve as a valuable tool for the early identification of individuals at high risk.

Introduction

Coronary artery disease (CAD) represents the preeminent cause of mortality on a global scale, putting a heavy strain on individual well-being and public health [1]. Type 2 diabetes mellitus (DM) is widely considered a prevalent comorbidity that affects the progression of CAD and the treatment choices for these patients [2]. According to estimates, the prevalence of diabetes among adults aged 20 to 79 worldwide in 2021 was 10.5% (536.6 million), and it was projected to reach 12.2% (783.2 million) by 2045 [3]. The risk of developing CAD in patients with T2DM is 2 to 6 times higher than that in non-diabetic individuals [4]. One of the main contributing factors to the onset and progression of diabetes is insulin resistance (IR) [5]. The process of measuring IR is complex and time-consuming, making clinical application difficult. According to research conducted in the last few years, the triglyceride glucose index (TyG) can also be utilized as a quick assessment tool for IR [6]. Obesity, as defined by the body mass index (BMI), plays an important role in the development of IR. As BMI increases, so does the risk of CAD [7]. Therefore, The TyG-BMI was proposed, which is obtained by multiplying TyG index and BMI. Some studies have shown that TyG-BMI has a higher advantage in predicting IR compared to pure TyG index [8]. In addition, TyG-BMI has shown unique advantages in predicting ischemic stroke [9]. However, limited research has been conducted on the relationship between TyG-BMI and the severity of coronary artery stenosis, particularly in patients with different glucose metabolic states, such as normal glucose regulation (NGR), pre-diabetes mellitus (Pre-DM), and diabetes mellitus (DM).

Dyslipidemia and arterial hypertension are typically found to coexist with DM [10]. Homocysteine (Hcy), a sulfur-containing amino acid involved in the metabolism of methionine and cysteine, has been increasingly recognized as a key factor in cardiovascular diseases [11]. Elevated Hcy levels, or hyperhomocysteinemia, have been linked to coronary heart disease, peripheral vascular disease, and cerebrovascular disease [12,13,14]. The coexistence of essential hypertension and hyperhomocysteinemia defines a condition known as H-type hypertension, which is associated with a five-fold increased risk of cardiovascular events compared to hypertension alone [15, 16]. Given this heightened risk, non-invasive biomarkers that can accurately predict CAD severity in patients with H-type hypertension are urgently needed, especially since some patients are unable to undergo invasive coronary angiography due to allergies or intolerance to contrast agents. The TyG-BMI index, which has been correlated with CAD severity [17], offers a promising avenue for further investigation, particularly in the context of H-type hypertension and different glucose metabolic states.

This study aims to explore the combined impact of H-type hypertension and glucose metabolic states on the severity of CAD, using the TyG-BMI index as a key marker. While H-type hypertension and glucose metabolic states each contribute significantly to cardiovascular risk, their combined effects may amplify CAD progression in a manner that differs from either condition alone. For example, while patients with pre-diabetes may benefit from lifestyle interventions to delay the onset of diabetes, those with established diabetes often require more aggressive pharmacological management to control blood glucose, blood pressure, and homocysteine levels. By examining these combined factors, our research seeks to provide personalized treatment recommendations and early intervention strategies that target the specific needs of H-type hypertensive patients across different glucose metabolic states. A clearer understanding of the cardiovascular risk profiles in these patients will enhance personalized treatment approaches and improve long-term clinical outcomes.

Data and methods

Data sources

This retrospective cohort study included 1537 hypertension inpatients who were admitted to the Department of Cardiology, Wuhan Third Hospital & Tongren Hospital of Wuhan University, between January 2021 and March 2024. The study protocol was approved by the Ethics Committee of the Wuhan Third Hospital (KY2021-001), the study complied with the Declaration of Helsinki, and written informed consent was obtained from all participants. Patients who meet the following criteria will be included in this study: (1) Participants were required to be adults aged 18 years or older, not pregnant or breastfeeding, and have not recently used glucocorticoids or immunosuppressants; (2) The subject underwent coronary angiography during hospitalization; (3) Previous history of hypertension or current oral use of antihypertensive drugs, or two consecutive measurements of blood pressure in a quiet state with systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg during hospitalization; (4) After reading the results of the coronary angiography by two professionals, the coronary angiography report indicated that at least one main coronary artery lumen had a stenosis greater than 50%; (5) Patients with complete clinical data and coronary angiography. The patients who meet the following conditions will be excluded: (1) had oncological, infectious, or serious liver or renal diseases; (2) no main stem lumen > 50% in the coronary angiography report; (3) used folic acid and B vitamins recently. Finally, a total of 1537 hypertensive patients who underwent coronary angiography were included in this study. According to the diagnostic criteria provided by the American Heart Association, the collected patients were divided into Single-vessel CAD group (N = 784) and Multi-vessel CAD group (N = 753) [18].

Data collection and measurements

In this study, age, gender, smoking, alcohol consumption, medical history, and medication history of patients were extracted from inpatient medical records. Venous blood specimens were collected by medical professionals early in the morning on the day after admission following fasting for at least 8 h. Fasting plasma glucose (FPG), glycosylated hemoglobin (HbA1c), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) levels were measured using an automatic hematology analyzer. Coronary angiography was performed using a percutaneous femoral arteriography technique with an angiography machine with sufficient right and left anterior oblique views to diagnose all coronary artery lesions [19]. According to glucose metabolism status, study subjects were divided into NGR, Pre-DM and DM groups. NGR was defined as FPG < 5.6 mmol/L or HbA1c < 5.7%, Pre-DM was defined as FPG 5.6–6.9 mmol/L or HbA1c of 5.7–6.4%, DM was defined as FPG ≥ 7.0 mmol/L or HbA1c ≥ 6.5% [20]. The diagnostic criteria for H-type hypertension entail the inclusion of hypertensive patients who exhibit elevated plasma homocysteine concentrations of 15 mol/L [21]. The BMI was calculated as weight (Kg) divided by the square of height (m²) [22]. The TyG-BMI index was defined as ln [TG (mg/dL) × FBG (mg/dL) /2] × BMI [23]. The number of diseased vessels with ≥ 50% stenosis indicated the severity of CAD in patients with CHD. Patients with only one major coronary artery affected were considered to have single-vessel CAD, and those with two or more major coronary arteries affected were considered to have multi-vessel CAD. No other vascular lesions but stenosis of ≥ 50% in the left main coronary artery was regarded as two lesions [19].

Statistical analyses

All data were analyzed using IBM SPSS Statistics version 24.0 (Chicago, IL, USA), GraphPad Prism 9.5.1 (SanDiego, California, USA), and R version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria). The visualization of the results was carried out using the ggplot2 and rms package within the R program. The included study population was stratified based on the presence of single lesion and multiple vessel lesions. The output of normally distributed variables was expressed as mean ± standard deviation, while the output of skewed distributed variables was expressed as median (25th to 75th quartiles). Categorical variables are expressed in frequency and percentage. Perform Mann Whitney U test or independent sample t-test on continuous data. For categorical variables, use chi square test for comparative analysis and express them in proportion. The patients were divided into three groups based on TyG-BMI index tertiles: T1 (119.34 ≤ TyG-BMI index ≤ 209.89), T2 (209.95 ≤ TyG-BMI index ≤ 237.45) and T3 (237.46 ≤ TyG-BMI index ≤ 360.86). Logistic regression was used to analyze the correlation between the TyG-BMI index and the degree of CAD (single lesion and multi-vessel lesion), and the Odds ratio (OR) value and 95% confidence interval (CI) were calculated. In addition, The Restricted Cubic Splines (RCS) are a valuable tool in statistical modeling for capturing complex nonlinear relationships between variables. the dose–response relationships between the TyG-BMI index and the severity of CAD were visualized using restricted cubic splines with 4 knots at the 5th, 35th, 65th, and 95th percentiles. Evaluate the predictive value of TyG-BMI index for the occurrence of multi-vessel disease in hypertensive patients using receiver operating characteristic (ROC) curves. Single factor and multi-element binary logistic regression analysis were performed by substituting the TyG-BMI index as a grouping variable and continuous variable into different glucose metabolism groups. Calculate the dose–response relationship between the TyG-BMI index and severity of CAD in different glucose metabolism groups. Then, ROC area under the curve (AUC) and 95% CI were calculated by grouping different glucose metabolism states to determine the accuracy of TyG-BMI index in detecting multi-vessel lesions in hypertensive patients across different glucose metabolism states. Select patients with Hcy greater than 15 mmol/L (H-type hypertension) from all hypertensive patients, perform univariate and multivariate binary logistic regression analysis on different glucose metabolism states, and determine its dose–response relationship, and validate with ROC again. P < 0.05 was considered statistically significant.

Result

Data filtering process

In the present study, a total of 9244 patients with hypertension who underwent coronary angiography were collected between January 2021 and March 2024. After the initial screening, a total of 2527 patients with complete angiographic reports and baseline data were identified. Patients with cancer, infectious diseases, or severe hepatic or renal diseases were excluded. By reviewing the coronary angiography reports, no main stem lumen ≥ 50% in the report was also excluded. Based on the patients' medication history, those who had recently used folic acid and vitamin B were excluded, with total of 990 patients being excluded from the study. Ultimately, a total of 1537 analyzable patient cases were obtained, consisting of 784 cases with single-vessel CAD and 753 cases with multi-vessel CAD. Selecting hypertensive patients with homocysteine levels higher than 15 mmol/L, a total of 912 H-type hypertensive patients were obtained, including 443 patients with single-vessel CAD and 469 patients with single-vessel CAD. The flow chart of patient selection was shown in Fig. 1.

Fig. 1

The flow chart of patient selection

Baseline characteristics of the study population

The basic characteristics of the study population of 1537 cases are shown in Table 1. The mean age of the patients was 67.69 ± 10.15 years. There were 820 (53.4%) male patients and 912 (59.3%) patients with H-type hypertension. And the average BMI was 24.66 ± 2.97 kg/m². The median plasma homocysteine concentration was determined to be 16.09 (12.92–20.57) μmol/L. there were 444 (28.9%) patients with NGR, 571(37.2%) patients with Pre-DM, and 522 (34.0%) patients with DM. The patients were divided into three groups based on TyG-BMI index tertiles. The patient's age, sex, BMI, DBP, smoking status, drinking status, H-type hypertension, homocysteine, FPG, hemoglobin, platelet, HbA1c, HDL-C, LDL-C, TC, TG, uric acid, creatinine, gensini score, use of calcium channel blocker (CCB) and angiotensin converting enzyme inhibitors (ACEI)/ angiotensin receptor blocker (ARB) were differed between the three groups (all P < 0.05, Table 1).

Table 1 Clinical and biological characteristics according to TyG-BMI index tertiles

The population was divided into a single-vessel lesion group of 784 (51.0%) patients and a multi-vessel lesion group of 753(49.0%) patients. Comparing the two groups, it was found that the patient’s sex, BMI, smoking status, H-type hypertension, homocysteine, FPG, platelet, HbA1c, Alb, HDL-C, LDL-C, TC, TG, uric acid, creatinine, cystatin-C, TyG-BMI index, different stage of glycometabolism were differed between the two groups (all P < 0.05, Table 2).

Table 2 Clinical and biological characteristics according to CAD severity

Association between TyG-BMI index and severity of coronary artery stenosis in hypertensive patients

Binary logistic regressions were used to assess the associations between CAD severity and various risk factors. Multi-vessel CAD was used as the dependent variable (single-vessel CAD was used as a reference). Male, BMI, H-type hypertension, smoking, diabetes mellitus, TC, TG, LDL-C, FPG, and HbA1c were significantly associated with multi-vessel CAD (P < 0.05) when compared to patients with single-vessel CAD (Table 3). When the TyG-BMI index served as a continuous variable, the TyG-BMI index was significantly correlated with the severity of coronary artery stenosis (OR 1.022, 95% CI 1.018–1.025). When the TyG-BMI index served as a classified variable, the risk of developing multi-vessel CAD was 5.049 (95% CI 3.873–6.582). There was statistically significant association between the TyG-BMI index and severity of coronary artery stenosis in hypertensive patients in multivariate logistic regression analyses. When the TyG-BMI index served as a continuous variable, the TyG-BMI index of hypertensive patients was significantly correlated with the severity of coronary artery stenosis (OR 1.043, 95% CI 1.032–1.053) after adjusting for confounding factors. When the TyG-BMI index served as a classified variable, the risk of developing multi-vessel CAD was 3.571 (95% CI 2.242–5.688) (shown in Table 4). Figure 2 shows the findings from RCS assessment, there was a linear dose–response relationship between TyG-BMI and the risk of multi-vessel CAD (P < 0.001, P non-linear = 0.382) after adjusting for confounding factors.

Table 3 Associations between CAD severity and risk factors in patients with hypertension

Table 4 Association between TyG index and CAD severity

Fig. 2

Restricted cubic spline curves by TyG-BMI index after covariate adjustment in hypertensive patients. OR, odds ratios; TyG, triglyceride glucose; BMI, body mass index; OR, odds ratios; TyG, triglyceride glucose; BMI, body mass index; CAD, Coronary artery disease

As shown in Fig. 3, the ROC curve model was constructed to evaluate the predictive ability of TyG-BMI on the severity of coronary artery stenosis in hypertensive patients. At the TyG-BMI index best truncation value of 226.29, the ROC curve yielded an AUC of 0.692 (95% CI 0.666–0.718, P < 0.001). The sensitivity and specificity of the test were determined to be 60.3% and 68.2%, respectively. The Yoden Index has been determined to be 0.285 (shown in Table 7).

Fig. 3

ROC curve for the use of TyG-BMI index in the detection of multi-vessel CAD in hypertensive patients. AUC, Area under the curve; CI, Confidence interval

Associations between TyG-BMI index and severity of coronary artery stenosis in different glucose metabolism states

As shown in Table 5, after multivariate adjustment showed significant associations between the TyG-BMI index and the risk of multi-vessel CAD in hypertensive patients according to different glucose metabolism states. Taking the T1 as a reference, T3 was significantly associated with an increased risk of multi-vessel CAD, with the highest OR value observed for DM (OR 3.836, 95% CI 1.763–8.347, P = 0.001). Figure 4 shows the findings from the RCS assessment, there was a linear dose–response relationship between TyG-BMI and the risk of multi-vessel CAD in NGR group (P < 0.001, P non-linear = 0.604) and Pre-DM group (P < 0.001, P non-linear = 0.801), but there was a non-linear dose–response relationship in DM group (P < 0.001, P non-linear = 0.017). Its cut-off point was 230.49. The receiver operating characteristic (ROC) curve yielded an area under the curve (AUC) of 0.699 (95% CI 0.653–0.744, P < 0.001) (shown in Fig. 5 and Table 7).

Table 5 Associations between TyG-BMI index and CAD severity according to different glucose metabolism states

Fig. 4

Restricted cubic spline curves by TyG-BMI index after covariate adjustment under different glucose metabolic states. (a) NGR group patients, (b) Pre-DM group patients, (c) DM group patients. OR, odds ratios; TyG, triglyceride glucose; BMI, body mass index; NGR, Normal glucose regulation; Pre-DM, Pre-diabetes mellitus; DM, Diabetes mellitus; CAD, Coronary artery disease

Fig. 5

ROC curve for the use of TyG-BMI index in the detection of multi-vessel CAD under different glucose metabolic states. (a) NGR group patients, (b) Pre-DM group patients, (c) DM group patients. AUC, Area under the curve; CI, Confidence interval

Associations between TyG-BMI index and severity of coronary artery stenosis in different glucose metabolism states in patients with H-type hypertension

The 912 patients with H-type hypertension among 1537 hypertension patients were screened for analysis. Supplementary Table S1 shows the baseline characteristics of the study population according to the TyG-BMI tertiles. Higher BMI, HbA1c, FPG, TC, TG, and LDL-C were most frequent in the highest tertile (TyG-BMI T3 group). A significantly great number of patients with glucose metabolism disorders were found in the T3 group (P < 0.001). Supplementary Table S2 shows the baseline characteristics of the single-vessel and multi-vessel CAD groups. Compared with the single-vessel CAD group, the patients in the multi-vessel CAD group had a higher content of TG, uric acid, creatinine, and cystati-C were more male patients and were more prone to glucose metabolism disorders (P < 0.05).

The Supplementary Table S3 shown that the BMI, smoking, diabetes mellitus, TG, LDL-C, FPG, and HbA1c were significantly associated with multi-vessel CAD (P < 0.05) when compared to patients with single-vessel CAD. When the TyG-BMI index served as a classified variable, the risk of developing multi-vessel CAD was 5.972 (95% CI 3.737–9.544)). Multivariate logistic regression analysis showed that TyG-BMI was an independent risk factor for multi-vessel CAD when used as a continuous variable (OR 1.047; 95% CI 1.033–1.061; P < 0.001). After adjusting for risk factors, the risk of multi-vessel CAD in the T3 groups was 3.927-fold (95% CI 2.119–7.280; P < 0.001) higher than in the T1 group shown in Supplementary Table S4.

In different glucose metabolism states, when the TyG-BMI index served as a classified variable, binary logistic regression analysis was conducted, and after adjusting for multiple factors, it was found that the risk of coronary artery multi-vessel disease in patients with H-type hypertension in NGR state was 3.992 (95% CI 1.033–15.425, P = 0.045), the risk in Pre-DM state was 3.376 (95% CI 1.082–10.528, P = 0.036), and the risk in DM state was 5.248 (95% CI 1.821–15.126, P = 0.002) shown in Table 6. Figure 6 shows the findings from the RCS assessment, there was a linear dose–response relationship between TyG-BMI and the risk of multi-vessel CAD in the NGR group (P < 0.001, P non-linear = 0.424) and Pre-DM group (P < 0.001, P non-linear = 0.603), but there was a non-linear dose–response relationship in DM group (P < 0.001, P non-linear = 0.042), and the cut-off point was 237.54.

Table 6 Associations between TyG-BMI index and CAD severity according to different glucose metabolism states in patients with H-type hypertension

Fig. 6

Restricted cubic spline curves by TyG-BMI index after covariate adjustment under different glucose metabolic states in patients with H-type hypertension. (a) NGR group patients, (b) Pre-DM group patients, (c) DM group patients. OR, odds ratios; TyG, triglyceride glucose; BMI, body mass index; NGR, Normal glucose regulation; Pre-DM, Pre-diabetes mellitus; DM, Diabetes mellitus; CAD, Coronary artery disease

The group with the highest AUC value obtained from the ROC curve of the subjects was the DM group (AUC = 0.720, 95% CI 0.661–0.780, P < 0.001). The sensitivity and specificity of the test were determined to be 74.7% and 61.5%, respectively. The Yoden Index has been determined to be 0.285, and the best truncation value was 226.06 (shown in Fig. 7 and Table 7).

Fig. 7

ROC curve for the use of TyG-BMI index in the detection of multi-vessel CAD under different glucose metabolic states in patients with H-type hypertension. (a) NGR group patients, (b) Pre-DM group patients, (c) DM group patients. AUC, Area under the curve; CI, Confidence interval

Table 7 The diagnostic value of TyG-BMI in assessing the risk of coronary artery stenosis in different data groups

Discussion

The objective of this study is to investigate the correlation between the TyG-BMI and the severity of coronary artery stenosis in hypertensive patients across different glucose metabolic states. Furthermore, we delved deeper into the correlation between the TyG-BMI index and the severity of coronary artery stenosis in patients with H-type hypertension. Our findings revealed a significant association between higher TyG-BMI levels and the severity of CAD in hypertensive patients, with the risk of multi-vessel disease being most pronounced in those with diabetes. Stratified analysis indicated that the TyG-BMI index had a stronger predictive value for multi-vessel CAD in the diabetic group compared to the NGR and Pre-DM groups. Notably, we discovered a non-linear relationship between TyG-BMI and CAD risk in the diabetic group, with a sharp increase in risk when the TyG-BMI index exceeded a critical threshold of 230.49. This non-linear pattern challenges traditional cardiovascular risk assessment models, which generally assume a linear relationship between risk factors and outcomes. The discovery of this non-linear risk accumulation highlights the need for early intervention, especially in diabetic patients, to mitigate CAD risk before the TyG-BMI index reaches critical levels.

IR denotes a state of diminished responsiveness to the normal biological actions of insulin in insulin-responsive tissues and organs [24, 25]. This condition is associated with a constellation of metabolic dysregulations, including impaired glucose and lipid homeostasis, as well as heightened oxidative stress [26, 27]. IR is considered as an important mechanism of type 2 diabetes [28].While traditional methods like the hyperinsulinemic-euglycemic clamp and HOMA-IR are precise, they are impractical for large-scale studies due to cost and complexity. The TyG index has emerged as a simpler, reliable marker for IR, and recent evidence suggests that combining the TyG index with BMI to form the TyG-BMI index may provide an even better predictor of IR [8, 29]. Dyslipidemia and glucose metabolism disorders can induce inflammation and cell apoptosis in blood vessels, thereby damaging endothelial function [30, 31]. Moreover, obesity, which frequently coexists with IR, creates a chronic inflammatory state that disrupts the balance between pro-inflammatory and anti-inflammatory immune cells, affecting insulin sensitivity and exacerbating glucose and lipid metabolic disorders [32]. These combined effects contribute to oxidative stress and further damage endothelial function, promoting the development of cardiovascular and metabolic diseases. Despite the potential utility of the TyG-BMI index, few studies have explored its association with the severity of coronary artery stenosis, particularly in relation to glucose metabolic states. In our study, when the TyG-BMI index was used as a continuous variable, it was significantly associated with the risk of multi-vessel CAD (OR 1.043, 95% CI 1.032–1.053). However, when patients were grouped into tertiles, we found that those in the highest tertile (T3) had a markedly increased risk of multi-vessel CAD (OR 3.571; 95% CI 2.242–5.688). The use of a restricted cubic spline model enabled us to precisely describe the dose–response relationship between the TyG-BMI index and multi-vessel CAD risk, reducing the information loss that often results from subjective data segmentation. This finding builds on earlier studies, which predominantly reported linear associations between TyG-BMI and CAD severity [33, 34], by introducing the concept of non-linearity, particularly in diabetic patients, as confirmed through ROC analysis (AUC 0.692, 95% CI 0.666–0.718).There have been multiple studies on different glucose metabolism before this study, and in patients with coronary heart disease across different glucose metabolism states, the degree of coronary stenosis was more severe in the DM group than in the NGR group [35]. The correlation between carotid plaques in patients with coronary heart disease across different glucose metabolism states is highest in the DM group [36, 37]. The systemic immune inflammation level of patients in the DM group upon admission is closely related to all-cause or specific factor mortality from CAD [38]. Our study expands upon these findings by being the first to explore the relationship between TyG-BMI and coronary artery stenosis in hypertensive patients across different glucose metabolic conditions. We also provide novel insights into the association between TyG-BMI and coronary artery stenosis in patients with H-type hypertension.

Homocysteine, as a non-essential amino acid, is crucial for metabolism in the human body, but when its plasma concentration exceeds a certain threshold, it can lead to a variety of diseases. Hyperhomocysteinemia is recognized as an independent risk factor for cardiovascular and cerebrovascular diseases, with a strong correlation. However, due to the unclear pathogenesis, most of the mechanisms are speculative, leading to ongoing academic debates. Hyperhomocysteinemia may exert detrimental effects on endothelial cells, augmenting coagulation capacity. Homocysteine increases the rigidity of coagulation by elevating levels of prothrombin and fibrinogen, and reducing sensitivity to lyocytosis [39]. It may also decrease fibrinogen levels, disrupting the body's equilibrium between coagulation and fibrinolysis, and precipitating a pre-thrombotic state [40]. Furthermore, homocysteine can stimulate the proliferation of smooth muscle cells [41, 42]. The specific damage caused by this process is to the vascular endothelium and the internal and external elastic laminae. Such reactions have been demonstrated to elevate the risk of atherosclerosis [43]. Mild hyperhomocysteinemia can lead to the deterioration of arterial and site-dependent elastic structures [44]. Serum folate increased by 38%, average homocysteine decreased by 21%, and corresponding ischemic heart disease and stroke decreased by 13% and 20%, respectively [45, 46]. Therefore, this study explored the correlation between the severity of coronary artery stenosis in hypertensive patients with hyperhomocysteinemia from a clinical application perspective. It was found that the TyG-BMI index was higher in predicting the risk of multi-vessel disease in H-type hypertensive patients than in hypertensive patients, and its risk in the population of diabetes patients reached 5.248 (95% CI 1.821–15.126), this is consistent with previous research findings [17]. In statistical analysis, The RCS is a flexible tool used to evaluate the relationship between variables, which can help researchers determine whether the relationship between a variable and the response variable is linear or nonlinear. In the population of H-type hypertension patients, a restricted cubic spline model test showed that the cutoff value of the TyG-BMI index for detecting multi-vessel coronary artery stenosis was 237.54. When TyG-BMI was greater than 237.54, the risk of developing multi-vessel CAD increased sharply, and the ROC validation showed a sensitivity of 74.7% and a specificity of 61.5% in DM group. Therefore, detecting TyG-BMI levels and severity of coronary artery stenosis in H-type hypertension patients has certain clinical application value, factors such as TyG-BMI that can predict and diagnose the risk of multi-vessel coronary artery stenosis should be seriously considered by researchers and clinicians. The basic pathological feature of H-type hypertension is hyperhomocysteinemia, and abnormal glucose metabolism leads to metabolic disorders such as IR, blood glucose fluctuations, and oxidative stress. When the two coexist, its role in promoting vascular endothelial damage, vascular inflammation and atherosclerosis is far greater than that of a single metabolic abnormality. This combined effect accelerates the occurrence of cardiovascular complications, especially the progression of multi-vessel CAD. Therefore, studying this special population not only helps to understand the accelerated development mechanism of CAD, but also provides a basis for early intervention.

Strengths and limitations

This study to explore the relationship between TyG-BMI and CAD severity across different glucose metabolic states in patients with H-type hypertension, providing the foothold for the prediction and diagnosis of multi-vessel CAD. TyG BMI, as an emerging metabolic marker, can provide a simple and non-invasive assessment tool for early identification of cardiovascular risk. This innovative application provides new possibilities for the clinical management of H-type hypertension patients. Our findings validate previous studies on TyG-BMI and coronary stenosis and expand the research by identifying non-linear relationships in diabetic states and specific cut-off points for CAD risk. This novel finding suggests that the TyG-BMI index can play a crucial role in identifying high-risk individuals who might otherwise be overlooked by traditional linear risk models, especially in resource-limited settings where more advanced diagnostic tools are unavailable. Additionally, the use of the restricted cubic spline model allowed for a more precise description of the non-linear dose–response relationship between TyG-BMI and multi-vessel CAD, reducing the potential for information loss typically associated with subjective data segmentation. Furthermore, the stratification of glucose metabolism states offers a personalized approach to treatment, as patients with different metabolic profiles may benefit from tailored intervention strategies. However, there are several limitations to acknowledge. As a retrospective analysis, our study is inherently limited in establishing causal relationships between TyG-BMI and CAD severity. The single-center design and modest sample size also restrict the generalizability of our findings. To enhance the external validity of our results, future research should employ larger, multi-center, prospective studies that can validate these findings and evaluate the predictive capacity of the TyG-BMI index across diverse populations. Furthermore, the absence of key variables such as glycated hemoglobin and homocysteine, due to incomplete data, may have introduced biases into our analysis. Subsequent studies should endeavor to mitigate these limitations by ensuring comprehensive data collection and expanding the sample to include a broader and more representative cohort. Notably, the TyG-BMI index, derived from the product of the TyG index and BMI, exhibits a substantial range, with a span of 241.52 units observed in our study. This variation underscores the complexity of interpreting TyG-BMI as a continuous variable, particularly when assessing its association with multi-vessel CAD, where the OR exhibits marginal changes with each additional case.

Conclusions

The TyG-BMI index holds significant clinical value in predicting the severity of coronary artery stenosis, with the highest predictive accuracy observed in patients with H-type hypertension across different glucose metabolism states. The identification of a non-linear risk accumulation in diabetic patients underscores the importance of early screening and intervention before the TyG-BMI index surpasses critical thresholds. This study provides new insights into how glucose metabolic states modify cardiovascular risk in hypertensive patients, offering a simple, non-invasive tool for early risk stratification and guiding personalized treatment strategies.