Relationship between TyG index, TyG–BMI and MACE events after percutaneous coronary intervention Authors: Rong Wang, Hewei Sun, Yuan Lin, Mingchen Liu Published: 2025 ABSTRACT BackgroundTo explore the relationship between TyG index, TyG–BMI and MACE events in patients after percutaneous coronary intervention, and to evaluate their potential as prognostic indicators compared to traditional risk assessment tools.MethodsFrom October 2022 to September of the following year, a total of 105 inpatients who underwent percutaneous coronary intervention (PCI) were carefully selected as research subjects within the scope of the Department of Cardiovascular Medicine at Chengde Central Hospital. MACE events such as recurrent angina pectoris, acute myocardial infarction, severe arrhythmia, in-stent restenosis, and cardiac death were recorded within 12 months after discharge. According to the occurrence of MACE events, the patients were divided into two groups: the MACE group and the non-MACE group. Subgroup analysis was performed for diabetic and non-diabetic patients. Internal validation was performed using the bootstrap resampling method with 1000 iterations to assess model stability.ResultsA total of 105 patients were followed up, and 23 MACE events occurred. The proportion of hypertension history in the MACE group was higher than that in the non-MACE group (82.6% vs. 69.5%). The age, LDL, creatinine, the TyG index and TyG–BMI values observed in the MACE group surpassed those of the non-MACE group, with these disparities reaching statistical significance (P < 0.05). TyG index and TyG–BMI were associated with MACE events in patients with coronary heart disease after PCI (P < 0.05). Bootstrap internal validation yielded optimism-corrected C-statistics of 0.742 for TyG index and 0.768 for TyG–BMI, with correction factors of 0.023 and 0.028, respectively, indicating moderate overfitting. Similar associations were observed in both diabetic and non-diabetic subgroups.ConclusionThis study suggests that TyG index and TyG–BMI may be associated with MACE events in patients with coronary heart disease after PCI. However, the small sample size, evidence of model overfitting, and lack of external validation limit the generalizability of these findings. Independent validation in larger, multicenter cohorts is essential before these indices can be recommended for clinical use. These preliminary findings provide a foundation for future adequately powered validation studies. Introduction Coronary atherosclerotic heart disease, in short, is a kind of heart disease caused by atherosclerotic lesions, inflammation or thrombosis in the inner wall of the coronary artery. These pathological changes lead to narrowing or even occlusion of the vascular lumen, which in turn leads to insufficient blood supply of the myocardium, hypoxia, and even the damage or death of myocardial cells [1]. Percutaneous coronary intervention (PCI) is an efficient and widely used treatment method for coronary heart disease, which can effectively restore coronary blood flow and improve the prognosis of patients. However, despite the restoration of blood flow, the incidence of cardiovascular adverse events such as recurrent angina, acute myocardial infarction, severe arrhythmia, in-stent restenosis, and cardiac death after PCI is still high [2]. Dyslipidemia is one of the main risk factors for coronary heart disease [3], which can be manifested as increased total cholesterol, triglyceride, and low-density lipoprotein cholesterol, and decreased high-density lipoprotein cholesterol. The key pathophysiological process behind diabetes as an important risk factor for coronary heart disease lies in insulin resistance, which is the cornerstone and core driver of the development of type 2 diabetes. In recent years, the medical community has gained a new perspective on the assessment of insulin resistance. Triglyceride-glucose (TyG) index and TyG–BMI combined with body mass index (BMI) have been shown to be valid and reliable surrogate indicators for the assessment of insulin resistance (IR), which can predict and manage health problems related to insulin resistance, including coronary heart disease risk. It provides new tools and ideas [4]. TyG index, which is composed of the product of triglyceride and glucose, and TyG–BMI index combined with body mass index, has been widely recognized by the medical community as effective and reliable measures to evaluate the degree of individual insulin resistance [5]. What distinguishes this study from previous research is our comprehensive evaluation of both TyG index and TyG–BMI in a prospectively followed cohort, with particular attention to their comparative prognostic value and potential integration into existing cardiovascular risk models. This study used a prospective design to explore the relationship between TyG index, TyG–BMI and MACE events in patients after PCI, aiming to achieve more accurate and personalized medical management of STEMI patients, effectively improving their prognosis and reducing the risk of adverse events. Materials and methods Subjects From October 2022 to September of the following year, a total of 105 inpatients who underwent percutaneous coronary intervention (PCI) in the Department of Cardiology of Chengde Central Hospital were carefully selected as the research subjects. Inclusion criteria: (1) STEMI diagnosis and PCI indications met the relevant criteria in the "Guidelines for the Diagnosis and Treatment of Acute ST-segment elevation myocardial Infarction (2019)"; (2) all patients successfully underwent PCI in our hospital, and were followed up for 1 year through regular outpatient review, telephone and other means; (3) all patients had complete clinical follow-up data. Exclusion criteria: (1) myocarditis, valvular heart disease, hypertrophic or dilated cardiomyopathy; (2) acute infection stage, acute exacerbation of COPD, acute attack of bronchial asthma, pulmonary embolism; (3) anemia; (4) severe chronic liver and kidney dysfunction; (5) persistent AF; (6) previous coronary artery bypass grafting; (7) acute and/or chronic infections; (8) autoimmune diseases; (9) known malignant tumors; (10) patients who met the inclusion criteria but were difficult to follow up. The study was approved by the Chengde Central Hospital, Ethical approval time: 13th June, 2023, Ethical code No. CDCHLL2023-424. Written informed consent was obtained from all individuals included in this study. Research method Laboratory index detection: Fasting venous blood samples of patients were collected in the morning of the next day after admission according to the standardized procedure. These samples were used to accurately measure a number of key biochemical indicators, including but not limited to triglyceride level, fasting blood glucose concentration, B-type natriuretic peptide (BNP), creatinine level, uric acid level, low-density lipoprotein cholesterol (LDL) and high-density lipoprotein cholesterol (HDL), to ensure comprehensive and accurate data support for the follow-up diagnosis and treatment of patients. For comparative analysis, HOMA-IR was calculated using the formula: HOMA-IR = fasting insulin (μU/mL) × fasting glucose (mmol/L) / 22.5, although insulin levels were not routinely measured in all patients. Data collection: The basic information and key indicators of a series of laboratory tests of patients participating in the study were systematically collected. Subsequently, patients who were successfully discharged from the hospital were followed up for 12 months to comprehensively understand and record their recovery and prognosis after PCI treatment in this period. According to the follow-up results, the patients were divided into two groups: MACE group, patients with major adverse cardiovascular events; and the non-MACE group, defined as patients who did not experience such events. In this study, MACE events were defined as the occurrence of recurrent angina pectoris, acute myocardial infarction, severe arrhythmia, in-stent restenosis, cardiac death and other events within 12 months after discharge of the enrolled patients [6]. Coronary angiography: The medical team worked out the most appropriate treatment plan based on the latest clinical guidelines and fully respecting the patient’s personal wishes and choices. In the process of coronary angiography and interventional therapy, the patients were led by experienced coronary intervention experts, and the established operating standards and procedures were strictly followed to ensure the safety and effectiveness of the treatment. PCI was selected as the first choice for patients, so as to achieve the best treatment effect. All patients without contraindications were provided with a standardized set of medical treatment regimens for secondary prevention of coronary heart disease as recommended by current medical guidelines. Observation indicators General information: Age, gender, body mass index, past history (hypertension, diabetes, hyperlipidemia, cerebrovascular disease, etc.), smoking, etc. Detection indicators: Triglyceride, total cholesterol, fasting blood glucose, creatinine, uric acid, BNP, LDL, HDL, white blood cell, hemoglobin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), albumin, etc., TyG index and TyG–BMI were calculated. TyG index = ln[fasting blood glucose (mg/dL) × fasting triglyceride (mg/dL)/2] [7]. TyG–BMI = TyG × body mass index (BMI). Prognosis: After the patients were discharged, a special follow-up mechanism was set up to regularly track and understand the health status and prognosis of the patients in the next 12 months through telephone contact or outpatient return visit. Statistical analysis SPSS 27.0 statistical software was used to conduct comprehensive and detailed processing and analysis of the collected data. The mean plus or minus standard deviation (x ± s) was used to describe the results, and the t test was used to compare the differences between different groups. For data that did not meet the characteristics of normal distribution, we used the median and quartile [M (P25, P75)], and used the Mann–Whitney U rank sum test to evaluate the differences between groups. In the processing of count data, the form of frequency and percentage was used to describe, and according to the specific situation of the data, the Chi-square test or Fisher’s exact probability method was used to test the correlation between different groups. To further explore the independent factors affecting poor prognosis, multivariate logistic regression analysis was performed. This analysis process helped to identify those independent variables that were significantly predictive of poor outcomes. Subgroup analyses were performed for diabetic and non-diabetic patients to evaluate whether the associations were consistent across these groups. Statistical power analysis was conducted post hoc to evaluate the adequacy of our sample size. To assess the internal validity and potential overfitting of our prediction models, we performed bootstrap internal validation using 1000 bootstrap resamples. For each bootstrap sample, we refitted the logistic regression models including TyG index and TyG–BMI separately, calculated the C-statistic (area under the receiver operating characteristic curve) in both the bootstrap sample and the original dataset, and computed the optimism (difference between bootstrap and original C-statistics). The optimism-corrected C-statistic was calculated as the apparent C-statistic minus the mean optimism across all bootstrap iterations. Additionally, we calculated calibration slopes to assess the degree of overfitting, where a slope substantially less than 1.0 indicates overfitting. This internal validation approach, while not a substitute for external validation, provides insight into the stability and potential reproducibility of our findings within similar populations. Throughout the statistical analysis, P < 0.05 was set as a threshold to determine whether differences or associations were statistically significant. Results Follow-up results A total of 105 patients with coronary heart disease after PCI completed the follow-up, including 23 patients (21.9%) with MACE events and 82 patients (78.1%) without MACE events. Among the total cohort, 32 patients (30.5%) had diabetes mellitus and 73 patients (69.5%) were non-diabetic. Comparison of baseline data between the two groups Gender, BMI, smoking, diabetes mellitus, hyperlipidemia, cerebrovascular disease, total cholesterol, triglyceride, HDL, fasting blood glucose, uric acid, BNP, white blood cell, hemoglobin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), albumin and other general data showed no significant difference between the two groups (P > 0.05). The proportion of hypertension history in the MACE group was higher than that in the non-MACE group (82.6% vs. 69.5%). The age, LDL, creatinine, TyG index and TyG–BMI of the MACE group were higher than those of the non-MACE group, and the differences were statistically significant (P < 0.05), as shown in Table 1.Table 1Comparison of baseline data between the two groupsWith MACE (n = 23)Non-MACE (n = 82)χ2/tP valueGender Male/female16/755/271.3760.056age59.15 ± 7.5855.68 ± 9.722.0430.040BMI (kg/m2)25.17 ± 2.3923.08 ± 1.210.0060.995Smoking13 (56.5%)46 (56.1%)0.1100.740Hypertension19 (82.6%)57 (69.5%)1.1600.030Diabetes8 (34.7%)24 (29.3%)0.3210.571Hyperlipidemia16 (69.5%)55 (67.1%)0.1580.691Cerebrovascular disease2 (8.6%)6 (7.3%)0.0530.818TC4.35 ± 0.054.23 ± 0.221.4710.143TG1.36 ± 0.801.17 ± 0.760.1900.249LDL2.52 (2.10,3.21)2.27 (1.83,2.56)1.9810.048HDL0.98 (0.78,1.18)1.06 (0.80,1.33)-1.7580.069Fasting blood glucose5.67 ± 0.285.54 ± 0.541.2100.231Uric acid349.35 ± 93.86323.27 ± 92.531.5380.125BNP73.74 ± 100.9688.71 ± 112.61-0.7350.463Cr74.09 ± 12.2869.64 ± 14.521.9600.047White blood cell6.35 ± 0.236.28 ± 0.311.1350.852Hb129.24 (114.53,145.01)130.12 (113.27,146.78)1.3060.768ALT24.14(18.23,30.32)27.45 (19.38,36.32)0.7580.698AST25.36 (17.92,32.47)24.63 (16.31,32.89)1.2620.801ALP68.75 (59.48,77.64)67.26 (58.35,75.69)1.2030.744ALB39.93 (36.72,43.15)40.62 (36.98,43.72)1.5720.769TyG index8.62 (8.41,8.79)7.29 (7.03,7.55)2.614 < 0.001TyG–BMI216.90 (191.57,242.25)143.54 (153.74,183.37)3.781 < 0.001Note MACE was defined as adverse cardiovascular events. BMI was defined as weight (kg)/height 2 (m2), TC was defined as total cholesterol, TG was defined as triglyceride, LDL-C was defined as low-density lipoprotein cholesterol, HDL-C was defined as high-density lipoprotein cholesterol, BNP was defined as B-type natriuretic peptide, Cr was defined as creatinine, Hb was defined as hemoglobin, TyG was defined as triglyceride glucose index, ALB was defined as albumin, TyG = ln[fasting blood glucose (mg/dL) × fasting triglyceride (mg/dL)/2], TyG–BMI is triglyceride glucose-body mass index, TyG–BMI = triglyceride glucose index (TyG) × body mass index (BMI) Table 1Comparison of baseline data between the two groupsWith MACE (n = 23)Non-MACE (n = 82)χ2/tP valueGender Male/female16/755/271.3760.056age59.15 ± 7.5855.68 ± 9.722.0430.040BMI (kg/m2)25.17 ± 2.3923.08 ± 1.210.0060.995Smoking13 (56.5%)46 (56.1%)0.1100.740Hypertension19 (82.6%)57 (69.5%)1.1600.030Diabetes8 (34.7%)24 (29.3%)0.3210.571Hyperlipidemia16 (69.5%)55 (67.1%)0.1580.691Cerebrovascular disease2 (8.6%)6 (7.3%)0.0530.818TC4.35 ± 0.054.23 ± 0.221.4710.143TG1.36 ± 0.801.17 ± 0.760.1900.249LDL2.52 (2.10,3.21)2.27 (1.83,2.56)1.9810.048HDL0.98 (0.78,1.18)1.06 (0.80,1.33)-1.7580.069Fasting blood glucose5.67 ± 0.285.54 ± 0.541.2100.231Uric acid349.35 ± 93.86323.27 ± 92.531.5380.125BNP73.74 ± 100.9688.71 ± 112.61-0.7350.463Cr74.09 ± 12.2869.64 ± 14.521.9600.047White blood cell6.35 ± 0.236.28 ± 0.311.1350.852Hb129.24 (114.53,145.01)130.12 (113.27,146.78)1.3060.768ALT24.14(18.23,30.32)27.45 (19.38,36.32)0.7580.698AST25.36 (17.92,32.47)24.63 (16.31,32.89)1.2620.801ALP68.75 (59.48,77.64)67.26 (58.35,75.69)1.2030.744ALB39.93 (36.72,43.15)40.62 (36.98,43.72)1.5720.769TyG index8.62 (8.41,8.79)7.29 (7.03,7.55)2.614 < 0.001TyG–BMI216.90 (191.57,242.25)143.54 (153.74,183.37)3.781 < 0.001Note MACE was defined as adverse cardiovascular events. BMI was defined as weight (kg)/height 2 (m2), TC was defined as total cholesterol, TG was defined as triglyceride, LDL-C was defined as low-density lipoprotein cholesterol, HDL-C was defined as high-density lipoprotein cholesterol, BNP was defined as B-type natriuretic peptide, Cr was defined as creatinine, Hb was defined as hemoglobin, TyG was defined as triglyceride glucose index, ALB was defined as albumin, TyG = ln[fasting blood glucose (mg/dL) × fasting triglyceride (mg/dL)/2], TyG–BMI is triglyceride glucose-body mass index, TyG–BMI = triglyceride glucose index (TyG) × body mass index (BMI) Comparison of baseline data between the two groups Note MACE was defined as adverse cardiovascular events. BMI was defined as weight (kg)/height 2 (m2), TC was defined as total cholesterol, TG was defined as triglyceride, LDL-C was defined as low-density lipoprotein cholesterol, HDL-C was defined as high-density lipoprotein cholesterol, BNP was defined as B-type natriuretic peptide, Cr was defined as creatinine, Hb was defined as hemoglobin, TyG was defined as triglyceride glucose index, ALB was defined as albumin, TyG = ln[fasting blood glucose (mg/dL) × fasting triglyceride (mg/dL)/2], TyG–BMI is triglyceride glucose-body mass index, TyG–BMI = triglyceride glucose index (TyG) × body mass index (BMI) Multivariate logistic regression analysis of MACE events Taking age, LDL, creatinine, TyG index, TyG–BMI and hypertension (history of hypertension = 1, no hypertension = 0) as independent variables, and whether MACE occurred after PCI as the dependent variable (MACE = 1, non-MACE = 0), multivariate logistic analysis was performed. The results showed that TyG index and TyG–BMI were associated with MACE events in patients with coronary heart disease after PCI (P < 0.05), as shown in Table 2.Table 2Multivariate logistic analysis of the occurrence of MACE after PCI in patients with CHDβWald (χ2)P valueORHR (95% CI)Age0.0463.4300.0541.0530.852–1.210LDL0.4532.5170.2111.6210.725–2.731Hypertension1.2600.7090.2841.0730.065–1.124Cr1.6200.7520.1031.1460.450–1.371TyG index0.3875.9110.0351.4731.078–2.013TyG–BMI0.6770.9980.0311.9671.522–2.417 Table 2Multivariate logistic analysis of the occurrence of MACE after PCI in patients with CHDβWald (χ2)P valueORHR (95% CI)Age0.0463.4300.0541.0530.852–1.210LDL0.4532.5170.2111.6210.725–2.731Hypertension1.2600.7090.2841.0730.065–1.124Cr1.6200.7520.1031.1460.450–1.371TyG index0.3875.9110.0351.4731.078–2.013TyG–BMI0.6770.9980.0311.9671.522–2.417 Multivariate logistic analysis of the occurrence of MACE after PCI in patients with CHD Bootstrap internal validation results Bootstrap internal validation with 1000 iterations was performed to assess model stability and potential overfitting. For the model including TyG index, the apparent C-statistic was 0.765, with a mean optimism of 0.023, yielding an optimism-corrected C-statistic of 0.742. The calibration slope was 0.891, indicating mild overfitting. For the model including TyG–BMI, the apparent C-statistic was 0.796, with a mean optimism of 0.028, yielding an optimism-corrected C-statistic of 0.768. The calibration slope was 0.874, also indicating mild to moderate overfitting. These results suggest that while the models show reasonable discrimination, there is evidence of overfitting that should be addressed through external validation in independent cohorts. The bootstrap 95% confidence intervals for the odds ratios were: TyG index OR 1.473 (bootstrap 95% CI: 1.021–2.156) and TyG–BMI OR 1.967 (bootstrap 95% CI: 1.398–2.683). Subgroup analysis by diabetes status To evaluate whether the predictive value of TyG indices differs between diabetic and non-diabetic patients, we performed subgroup analyses. In the non-diabetic subgroup (n = 73), the TyG index remained significantly associated with MACE events (OR 1.425, 95% CI 1.034–1.964, P = 0.031). Similarly, TyG–BMI showed a significant association in both diabetic (OR 2.134, 95% CI 1.456–3.128, P = 0.001) and non-diabetic (OR 1.876, 95% CI 1.398–2.517, P = 0.001) subgroups. Discussion Although PCI dilates the stenosed/occluded lesion site and restores myocardial reperfusion, atherosclerosis still exists, and reperfusion itself will also aggravate myocardial injury and increase the risk of MACE after PCI, so early warning and identification are particularly important [8]. This study found that the incidence of MACE in patients with coronary heart disease treated by PCI within 12 months after discharge was 21.9%. The proportion of patients with hypertension in the MACE group was higher than that in the non-MACE group, and the age, LDL, creatinine, TyG index and TyG–BMI in the MACE group were higher than those in the non-MACE group. The difference was statistically significant (P < 0.05). In the multivariate logistic regression analysis, it was revealed that TyG index and TyG–BMI combined with body mass index (BMI) were associated with the occurrence of MACE in patients with coronary heart disease after PCI (P < 0.05). However, bootstrap internal validation revealed evidence of mild-to-moderate overfitting, emphasizing the need for external validation. Interpretation of internal validation results The bootstrap internal validation yielded optimism-corrected C-statistics of 0.742 for TyG index and 0.768 for TyG–BMI, indicating reasonable but not excellent discrimination ability. The optimism values of 0.023 and 0.028, along with calibration slopes of 0.891 and 0.874, suggest mild to moderate overfitting of our models. This is not unexpected given the limited number of events (n = 23) relative to the number of predictors examined. The overfitting observed in internal validation further underscores the critical importance of external validation before these indices can be recommended for clinical use. Importantly, internal validation cannot replace external validation, as it only assesses model stability within the same population and does not address generalizability to different settings, populations, or time periods. Therefore, while our internal validation results provide some reassurance about model stability, they do not overcome the fundamental limitation of lacking independent external validation. Novel contributions and clinical implications What distinguishes our study from previous research is the simultaneous evaluation of both TyG index and TyG–BMI in a prospectively followed cohort, with demonstration of their predictive value in both diabetic and non-diabetic populations. This finding suggests that insulin resistance, as measured by these simple and cost-effective indices, plays a crucial role in post-PCI outcomes regardless of diabetes status. The clinical implication is that TyG index and TyG–BMI should be considered not merely as surrogate markers of insulin resistance, but as integral components of comprehensive risk stratification models for patients undergoing PCI. These indices could potentially enhance existing risk scores such as SYNTAX Score II, which has been shown to correlate with glycated hemoglobin levels in diabetic patients [9]. Integration of TyG indices into clinical risk models could provide a more nuanced assessment of metabolic risk beyond traditional factors. Furthermore, patients with elevated TyG index or TyG–BMI might benefit from more intensive metabolic interventions, including targeted insulin-sensitizing therapies, stricter glycemic control, and more aggressive lipid management, even in the absence of overt diabetes. Comparison with traditional insulin resistance markers While HOMA-IR has been the gold standard for assessing insulin resistance in research settings, it requires insulin measurements that are not routinely performed in clinical practice. Recent studies have suggested that the TyG index may outperform HOMA-IR in certain clinical contexts [10]. In our cohort, although insulin levels were not systematically measured, the association between TyG indices and MACE events supports their potential utility as practical alternatives to HOMA-IR in routine clinical assessment. The advantage of the TyG index lies in its simplicity, requiring only fasting glucose and triglyceride measurements, which are routinely available in all PCI patients. However, the comparative performance of TyG indices versus HOMA-IR should be directly evaluated in future validation studies that systematically measure insulin levels. Pathophysiological mechanisms Triglyceride-glucose (TyG) index is a composite index combining TG and FPG concentration, which has been proposed as a marker for predicting the progression and calcification of coronary atherosclerosis. TyG index comprehensively reflects blood lipid and glucose metabolism, and the TyG index has been associated with cardiovascular risk in patients with coronary heart disease [11]. The pathophysiological link between insulin resistance and adverse cardiovascular outcomes involves multiple mechanisms, including endothelial dysfunction, increased oxidative stress, enhanced inflammatory response, and accelerated atherosclerosis progression. Some studies have found that the incidence of MACE in postmenopausal ACS patients undergoing PCI increases with the increase of the TyG index. In ACS patients undergoing PCI, the TyG index level has been correlated with the incidence of MACE [12]. Additionally, insulin resistance may contribute to platelet hyperactivity and a prothrombotic state, potentially increasing the risk of stent thrombosis. ZHU et al. conducted a 12-month follow-up of 1574 ACS patients and found that TyG index elevation was independently and positively correlated with in-stent restenosis in ACS patients undergoing PCI [13]. The preliminary findings of this study are consistent with the results of previous studies in several aspects: first, it was observed that TyG index in patients with MACE was significantly higher than that in the control group without MACE (P < 0.05), which provides additional support for the potential role of TyG index in predicting the prognosis of patients with coronary heart disease. Secondly, logistic regression analysis suggested that TyG index (odds ratio 1.473, 95% confidence interval 1.078–2.013, P = 0.035) was associated with MACE events in patients with coronary heart disease after PCI. Recent work has also identified other simple composite scores as useful prognostic markers in ischemic settings—for example, the HALP score predicted in-hospital mortality among STEMI patients undergoing primary PCI [14], which supports the broader concept that easily derived indices can carry prognostic information. TyG–BMI as a potential enhanced predictive marker TyG–BMI, as an emerging insulin resistance assessment tool in recent years, integrates multiple key biomarker information, including fasting plasma glucose (FPG), serum triglycerides (TG), and body mass index (BMI). It is the combined product of BMI and TyG, and the combination of TyG and BMI has been suggested to improve the ability to identify insulin resistance [15]. The incorporation of BMI into the TyG index acknowledges the important role of adiposity in insulin resistance and cardiovascular risk. Beyond glucose–lipid metrics, related pathobiological markers have also been linked to adverse outcomes in acute coronary settings—for example, malondialdehyde levels at admission predicted in-hospital mortality in ACS [16], and the triglyceride–glucose index was associated with intracoronary thrombus burden in STEMI patients undergoing primary PCI [10]. Song et al. conducted a study on 210 investigators who underwent coronary angiography (CAG), among whom 126 patients were diagnosed with CHD and the other 84 patients were non-CHD. Through correlation analysis, the results showed that the TyG index and TyG–BMI could not only predict the risk of CHD, but also have a certain correlation with the degree of coronary artery lesions [17]. Huang et al. found that TyG–BMI was significantly associated with the risk of atherosclerotic cardiovascular disease [18]. Cheng et al. included 1438 patients who underwent percutaneous coronary intervention (PCI) and drug-eluting stent (DES) implantation. They were followed up for up to 34 months and found that TyG–BMI was significantly lower in women and the elderly. It was established as an independent and significant prognostic indicator of major adverse cardiovascular events after drug-eluting stent implantation. Specifically, a higher TyG–BMI level is significantly associated with an increased incidence of MACE in these two patient groups [19]. Previous research has established important relationships between metabolic markers and coronary disease severity. Karakoyun et al. demonstrated that glycated hemoglobin levels are significantly associated with SYNTAX Score II in patients with type 2 diabetes mellitus, suggesting that metabolic dysfunction directly correlates with the anatomical complexity of coronary disease [20]. Furthermore, traditional biomarkers continue to provide prognostic value, as shown by Karabağ et al., who found that serum uric acid levels were associated with both SYNTAX Score II and long-term mortality in stable angina patients undergoing PCI for multivessel and/or unprotected left main disease [21]. The preliminary results of this pilot study showed that TyG–BMI in the MACE group was higher than that in the non-MACE group (P < 0.05), and logistic regression analysis showed that TyG–BMI (OR1.967, 95 95% CI: 1.522–2.417, P = 0.031) was associated with MACE events in patients with coronary heart disease after PCI. This potentially enhanced predictive value of TyG–BMI compared to TyG index alone suggests that the incorporation of anthropometric measures may provide additional prognostic information beyond metabolic parameters alone, though this hypothesis requires validation in independent cohorts. Clinical implications and need for validation The preliminary identification of elevated TyG index and TyG–BMI as potential risk factors for MACE could have important implications for post-PCI management strategies if validated in future studies. Patients with elevated values might potentially benefit from more frequent assessment of glycemic and lipid profiles with tighter control targets, structured programs focusing on weight reduction and dietary modification, consideration of insulin-sensitizing agents, more frequent clinical follow-up, and more intensive lipid management. However, we must emphasize that these are speculative implications that require validation before clinical implementation. The current evidence from this pilot study is insufficient to recommend changes in clinical practice. Future randomized controlled trials would be needed to determine whether TyG-guided interventions actually improve outcomes compared to standard care. Integration into clinical practice and future directions As a pilot study, our findings suggest potential directions for future research rather than immediate clinical application. If validated in larger studies, the TyG index and TyG–BMI could potentially be routinely calculated at the time of PCI and incorporated into discharge risk stratification. Cut-off values for risk stratification would need to be established through larger studies with appropriate methodology, potentially leading to the development of TyG-enhanced risk scores. Critical next steps include: (1) external validation in large, independent, multicenter cohorts with diverse populations; (2) direct comparison with established risk scores to determine incremental predictive value; (3) determination of optimal cut-off values through appropriate statistical methods including decision curve analysis; (4) prospective evaluation of whether TyG-guided therapy improves outcomes in randomized controlled trials; (5) investigation of dynamic changes in TyG indices over time and their prognostic significance; and (6) cost-effectiveness analysis of TyG-based risk stratification strategies. Without the completion of these validation steps, clinical implementation of TyG-based risk assessment would be premature. Limitations of this study (1) The most critical limitation of this pilot study is the small sample size (n = 105) with only 23 MACE events, which significantly limits the statistical power and robustness of our findings. Post hoc power analysis revealed a power of 0.72 for detecting the observed effect sizes, which is below the conventional threshold of 0.80, indicating that our study may be underpowered to detect true associations with adequate confidence. The multivariate logistic regression analysis with multiple variables and relatively few events raises substantial concerns about potential overfitting, which was confirmed by our bootstrap internal validation showing calibration slopes of 0.891 and 0.874, indicating mild to moderate overfitting. This may reduce the reliability and generalizability of the conclusions. Most importantly, we do not have an independent external validation cohort, which is the gold standard for assessing model performance and generalizability. While we performed bootstrap internal validation to assess model stability, internal validation cannot substitute for external validation in different populations and healthcare settings. Therefore, our results should be interpreted strictly as hypothesis-generating and preliminary, and the clinical utility of TyG index and TyG–BMI for MACE prediction in post-PCI patients requires mandatory confirmation in adequately powered, multicenter external validation studies before any clinical application can be considered. (2) The data collection form is relatively simple, which may not fully capture all relevant information of patients, thus affecting the depth and comprehensiveness of the analysis. Specifically, we did not systematically measure insulin levels to calculate HOMA-IR for direct comparison with TyG indices. (3) In the follow-up by telephone, despite efforts to maintain objectivity and impartiality, there is still the possibility of subjective bias from the experimenter. (4) The effects of socio-economic and psychological factors on prognosis were not included. (5) The single-center nature of this pilot study limits the generalizability of findings to other populations and healthcare settings with different patient characteristics, treatment protocols, and healthcare systems. (6) The 12-month follow-up period may be insufficient to capture long-term cardiovascular outcomes and the full spectrum of late complications after PCI. (7) We did not compare the predictive performance of TyG indices with established risk scores such as GRACE or SYNTAX Score II, which would be essential for determining their incremental value. (8) As a pilot study, this investigation was not designed or powered to provide definitive evidence for clinical decision-making, but rather to generate hypotheses and inform the design of future validation studies. Conclusion In conclusion, TyG index and TyG–BMI are important influencing factors for MACE events in patients with coronary heart disease after PCI, demonstrating predictive value for short-term prognosis in both diabetic and non-diabetic populations. These easily calculated, cost-effective indices could be integrated into existing cardiovascular risk stratification models to enhance prognostic assessment and guide personalized therapeutic strategies. Their incorporation into routine clinical practice may help identify high-risk patients who would benefit from more intensive metabolic interventions and closer follow-up, potentially improving outcomes after PCI.