This was a retrospective observational study. A total of 454 patients diagnosed with ACS in the First Hospital of Fujian Medical University, were enrolled from June 2015 to September 2021 who underwent coronary angiography at 6 to 24 months after successful DES-PCI. The following exclusion criteria were as follows, (1) age less than 18 years; (2) history of coronary artery bypass grafting (CABG) and stent implantation; (3) no DES implantation; (4) critical structural heart disease requiring intervention; (5) severe liver, respiratory and renal dysfunction (estimated glomerular filtration rate [eGFR] < 30 mL/min/1.73 m²); (6) advanced malignant tumors with short life expectancy; (7) incomplete clinical data or death during hospitalization. The study was approved by the local Ethics Committee and adhered to the Declaration of Helsinki. Informed consent was waived due to the retrospective, anonymized nature of the data.

The coronary intervention and perioperative management were performed in accordance with the current guidelines of our centre (11). All patients received guideline-directed medical therapy, including antiplatelet, lipid-lowering, and glucose-lowering agents. Radial access was preferred. DES type, stent length, diameter, and number were at the operator's discretion. Dual antiplatelet therapy (aspirin combined with clopidogrel or Tegretol) for a minimum of 12 months was recommended in all cases afer the implantation of a drug-eluting stent (DES-PCI), and long-term medication management was guided by individual risk.

Patient demographic, medical history, and clinical characteristics, including age, sex, smoking status, left ventricular ejection fraction (LVEF), angiographic evaluation results, interventional parameters such as number of stents, lesion type, and target vessel, as well as discharge medications, were collected from the electronic medical record system. Furthermore, peripheral venous blood was collected at least 8 h of fasting and analyzed for fasting blood glucose(FBG), uric acid, estimated glomerular filtration rate(eGFR), C-reactive protein (CRP), lipid profile, including triglycerides (TG), TC, LDL-C, and HDL-C.

The definition of ACS complied with the current guideline of the European Society of Cardiology (12). Diabetes mellitus was defined as a prior specialist diagnosis, current use of glucose-lowering therapy, or the presence of diabetic symptoms with either random plasma glucose ≥11.1 mmol/L, fasting plasma glucose ≥7.0 mmol/L, or 2 h plasma glucose ≥11.1 mmol/L during a 75-g oral glucose tolerance test (13). Hypertension was defined as a prior diagnosis by a physician, current use of antihypertensive therapy, or systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg on at least three separate occasions (14). Hypercholesterolemia was defined as fasting serum TC ≥ 6.22 mmol/L, fasting LDL-C ≥ 4.14 mmol/L, or current lipid-lowering therapy (15). Hyperuricemia was defined as serum uric acid ≥420 μmol/L in males or ≥360 μmol/L in females (16). Smoking status was categorized as current smoking or non-smoking, and alcohol consumption as current drinking or non-drinking. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared (kg/m²) (17). Based on angiographic assessment, multi-vessel disease was defined as the presence of at least two vessels with significant diameter stenosis (≥50%), while multi-stent implantation was defined as placement of two or more stents (18). The equation used to calculate eGFR (mL/min/1.73 m²) is as follows, 186 × SCr (mg/dL)^(−1.154) × age^(−0.203) × 0.742 (if female) (19). The Gensini score was determined according to both the degree and location of coronary stenosis, stenosis severity of 25%, 50%, 75%, 90%, 99%, and complete occlusion corresponded to scores of 1, 2, 4, 8, 16, and 32, respectively, each multiplied by a segment-specific weighting factor, with all segment scores summed to yield the final score (20). The Atherogenic Index of Plasma (AIP) was calculated as the logarithm of the ratio between triglycerides (TG) and high-density lipoprotein cholesterol (HDL-C). The formula for calculating the metabolic score for insulin resistance(METS-IR) index was: Ln [2 × FBG (mg/dL) +TG (mg/dL)] × BMI/Ln[ HDL-C (mg/dL)] (21). The formula for calculating the TyG index was: Ln [TG (mg/dL) × FBG (mg/dL)/2] (22). The formula for calculating the CHG index was: CHG index = Ln [TC (mg/dL) × FBG (mg/dL)/2 × HDL (mg/dL)] (10).

All patients underwent scheduled outpatient follow-up, with follow-up coronary angiography performed 6 to 24 months after successful DES-PCI. On the basis of angiographic follow-up results, patients were categorized into ISR and non-ISR groups, which was defined as the presence of significant diameter stenosis (≥50%) at the segment inside the stent or involving its 5 mm edges, which is in line with previous studies (23). Of note, the follow-up angiography was determined by experienced interventional cardiologists.

Continuous variables were presented as mean ± standard deviation (x ± s) for normally distributed data or as median and interquartile range for non-normally distributed data. Categorical variables were described as frequency counts and percentages (%). Group differences were assessed using one-way analysis of variance (ANOVA) for normally distributed continuous variables, the Kruskal–Wallis test for non-normally distributed variables, and the chi-square (χ²) or Fisher's exact test for categorical variables.

The predictive performance of the CHG index for drug-eluting stent in-stent restenosis (DES-ISR) was evaluated using receiver operating characteristic (ROC) analysis. The area under the curve (AUC) and the optimal cutoff value were derived from this analysis. Univariate logistic regression identified potential predictors of ISR after successful DES-PCI. Variables with a univariate P-value <0.1 (Supplementary Table S1) or those deemed clinically relevant were entered into multivariate logistic regression models (4, 11, 23). Three models were developed to mitigate confounding and assess the association between the CHG index (analysed continuously or categorically) and DES-ISR: Model 1 (age, sex, BMI); Model 2 (Model 1 + eGFR, LVEF, CRP, smoking, hypertension, diabetes); Model 3 (Model 2 + Gensini score, multi-vessel disease, multi-stent implantation, total stent length, mean stent diameter). Restricted cubic spline analysis explored the dose-response relationship. Subgroup analyses tested consistency across clinical strata.

To address key confounding, a series of sensitivity analyses were performed to evaluate the robustness of the primary association. Based on the primary fully-adjusted model(Model 3), further adjusting for: (1) statin intensity (high vs. non-high intensity), (2) ezetimibe use, (3) P2Y12 inhibitor type (ticagrelor vs. clopidogrel), and (4) statin intensity and type of P2Y12 inhibitor, and (5) all four medication variables combined.

The CHG index's discriminative performance was compared to other indices (TyG, AIP, METS-IR) using DeLong's test. Model calibration was evaluated with the Hosmer-Lemeshow test. Correlations between the CHG index and cardiovascular risk factors were examined using Pearson or Spearman tests.

All statistical analyses were conducted using SPSS version 27.0 (IBM, Armonk, NY, USA) and R programming language version 4.5.0, with two-sided P < 0.05 considered statistically significant.

A total of 454 patients who underwent follow-up coronary angiography 6 to 24 months after successful DES-PCI were enrolled. As shown in Table 1, the mean age of the cohort was 63.01 ± 10.25 years, and 370 (81.50%) participants were male. The prevalence of current smoking, hypertension, Hypercholesteraemia, and diabetes mellitus were determined to be 58.59%, 71.37%, 22.25%, and 45.37%, respectively. Regarding angiographic findings, multivessel lesions (71.59%) and PCI for the left anterior descending artery disease (67.62%) was the most common. More than half (57.93%) of the patients had multiple stents (≥2) implanted. Overall, 70 patients (15.12%) experienced ISR.

According to tertiles of the CHG index, all participants were categorized into three groups (Table 1), T1 (CHG index <5.264, n = 151), T2 (5.264 ≤ CHG index <5.666, n = 151), and T3 (CHG index ≥5.666, n = 152). With increasing tertiles of the CHG index, levels of serum FBG, TG, TC, LDL-C, TyG index, AIP, METS-IR, prevalence of hypercholesteraemia, STEMI, and diabetes, BMI, and use of insulin, hypoglycemic agents, and diuretics all showed significant increases (P < 0.05). Nevertheless, patients with higher CHG indices were relatively younger, had lower HDL-C levels, LVEF, and exhibited lower rates of calcium channel blocker (CCB) use. No significant differences were observed in other parameters among the three groups (P > 0.05). Notably, there was a trend toward higher rates of multi-vessel disease, target vessel in RCA, and Gensini score, as well as a significantly greater incidence of ISR, in the higher CHG index groups.

As shown in Figure 1A, the prevalence of ISR had stepwise increase with the increasing tertile of the CHG index (9.27% vs. 17.22% vs. 19.14%; P = 0.031). Additionally, it is noteworthy that the ISR group also had a significantly higher CHG index than the non-ISR group (5.47 ± 0.46 vs. 5.66 ± 0.47, P = 0.001, Figure 1B).

In the univariate logistic regression analysis, the CHG index, as a continuous variable, was positively associated with the risk of ISR after successful DES-PCI (OR =  2.43, 95% CI 1.42–4.16, P = 0.001; Table 2). When analyzed categorically, the risk of ISR was significantly higher in the second (OR = 2.04; 95% CI, 1.02–4.07; P = 0.045) and third tertiles (OR = 2.41; 95% CI, 1.25–4.75; P = 0.011; Table 2) compared to the first tertile.

In multivariate logistic regression models, with the CHG index initially included as a continuous variable, each 1-unit increase in CHG index remained independently associated with elevated ISR risk in model 1 (OR = 2.45; 95% CI, 1.42–4.23; P = 0.001; Table 2), model 2 (OR = 2.51; 95% CI, 1.26–4.99; P = 0.009), and the fully adjusted model 3 (OR = 2.61; 95% CI, 1.28–5.33; P = 0.008).

This association persisted when the CHG index was evaluated as a categorical variable, in model 3, after comprehensive adjustment, the adjusted odds ratios (95% CI) for the second and third tertiles were 2.33 (1.12–4.85; P = 0.024) and 2.40 (1.05–5.49; P = 0.038), respectively, compared with the reference group (Table 3).

The consistency of the association between the CHG index and DES-ISR was evaluated across key clinical subgroups (Figure 2). In multivariable-adjusted model 3, a positive association was observed in most subgroups. Notably, no significant interactions were identified between the CHG index and any subgroup variable (all P for interaction > 0.05). Sensitivity analyses confirmed the stability of the primary finding (Supplementary Table S4). The positive association between the CHG index and DES-ISR risk remained statistically significant across all models. In the model that additionally adjusted for statin intensity, ezetimibe use, and P2Y12 inhibitor type, each 1-unit increase in the CHG index continued to be independently associated with a higher risk of DES-ISR(OR = 2.61; 95% CI, 1.28–5.32; P = 0.008; Supplementary Table S4). And when the CHG index was evaluated as a categorical variable, the adjusted odds ratios (95% CI) for the second and third tertiles were 2.31 (1.11–4.84; P = 0.026) and 2.40 (1.05–5.49 == 51; P = 0.038 = 9), respectively, compared with the reference group (Supplementary Table S4).

As presented in Figure 3A, the ROC curve analysis indicated that the CHG index provides modest predictive value for DES-ISR in patients with ACS, with an AUC of 0.619 (95% CI, 0.552 to 0.687, P = 0.002). The optimal cutoff value was 5.402 (sensitivity, 71.4%, specificity, 49.7%). The Hosmer-Lemeshow goodness-of-fit test for the fully adjusted model (Model 3) demonstrated excellent calibration (P = 0.729), indicating no significant deviation between the predicted probabilities of DES-ISR and the observed outcomes (Figure 4). The area under the receiver operating characteristic curve (AUC) for the CHG index predicting ISR was significantly higher compared to metabolic score for insulin resistance (METS-IR), (CHG index vs. METS-IR, 0.619 vs. 0.514, P = 0.009). Furthermore, the AUC for the CHG index was greater than that of TyG index (0.619 vs. 0.613) and AIP (0.619 vs. 0.577), but Delong test revealed this difference was not statistically significant (CHG index vs. TyG index, P = 0.811, CHG index vs. AIP, P = 0.236) (Figure 3B) (Additional File, Supplementary Tables S2, S3).

Figure 5 illustrates that, in the unadjusted RCS model, the CHG index exhibited a positive linear association with the risk of ISR after DES-PCI in ACS patients (P = 0.006, P for nonlinear = 0.233). Even after adjusting for confounders, a significant linear relationship persisted between the CHG index and the risk of ISR (P for overall = 0.016; P for nonlinear = 0.118).

To further ascertain the association between CHG index and other risk factors for adverse cardiovascular outcomes, Spearman rank or Pearson's correlation analyses were performed (Table 3). The CHG index demonstrated a positive correlation with the Gensini scores (r = 0.243, P < 0.001), the TyG index (r = 0.766, P < 0.001), the AIP (r = 0.570, P < 0.001), the METS-IR (r = 0.530, P < 0.001), LDL-C (r = 0.365, P < 0.001), TG (r = 0.766, P < 0.001), and total stent lengths (r = 0.133, P = 0.004). In contrast, the CHG index demonstrated a negative correlation with age (r = −0.181, P < 0.001) and LVEF(r = −0.179, P < 0.001).