The prospective registry database of patients who underwent any percutaneous coronary intervention at Juntendo University Hospital, Tokyo, Japan (Juntendo Physicians’ Alliance for Clinical Trial: J-PACT) is publicly registered (University Medical Information Network Japan-Clinical Trials Registry, UMIN-CTR 000035587). In addition, the ethics committee of the Juntendo Clinical Research approved this study (reference number E22-0409), and all participants provided written informed consent or opting-out consent. The study was conducted in accordance with the principles of the Declaration of Helsinki (15).

This was an observational cohort study that included statin-treated patients who participated in three clinical prospective trials that evaluated the impact on coronary plaque burden measured by serial IVUS at baseline and 6 to 12 months later. Overall, this study enrolled 141 patients with stable CAD and acute coronary syndrome (ACS) from the following trials: (1) the ENTERPRISE trial assessed continuous positive airway pressure in patients with stable CAD coexisting with sleep-disordered breathing (16), (2) the ESPECIAL-ACS study assessed dipeptidyl peptitase-4 inhibitors in patients with ACS and diabetes mellitus (DM) (17), and (3) the ZEUS trial assessed the combination of statin and ezetimibe in patients with ACS (18). The following patients were excluded from this study: patients with severe obesity [body mass index (BMI) ≥35.0 kg/m²], patients for whom LDL-C-lowering therapy including statins was not performed, patients without data of the baseline TyG index, and patients for whom sufficient analysis data were not available.

Regarding lipid profiles, high-density lipoprotein cholesterol (HDL-C), LDL-C, and triglyceride levels were assayed using LABOSPECT 008α (Hitachi, Ltd., Tokyo, Japan). The Japan Atherosclerosis Society has recently updated the lipid management for secondary prevention in atherosclerotic cardiovascular disease (19). Thus, lipid control in this study was defined as aggressive LLT if the following target LDL-C values were achieved: on-statin treatment LDL-C value <55 mg/dL in patients with ACS or DM, on-statin treatment LDL-C value <70 mg/dL in patients with stable CAD and non-DM, or on-statin treatment LDL-C value decreased by ≥50% from baseline; on the other hand, lipid management was defined as non-aggressive LLT when these target values of LDL-C were not achieved. Patients with a blood pressure >140/90 mmHg or those receiving antihypertensive drugs were considered hypertensive (20). Patients with a hemoglobin A1c (HbA1c) level of ≥6.5%, those administered with oral hypoglycemic agents, or those receiving insulin injection therapy were defined as having DM (21). A family history of premature CAD was defined as the presence of any first-degree relative with premature CAD (age <55 years for men and <65 years for women) (22). Chronic kidney disease was defined as an estimated glomerular filtration rate of <60 mL/min/1.73 m² based on the Modification of the Diet in Renal Disease equation modified using the baseline serum creatinine level (23).

The formula for calculating the TyG index is TyG index = Ln [fasting triglyceride (mg/dL) × fasting plasma glucose (mg/dL)/2]. The relationship between the TyG index and cardiovascular events in Southeast Asia has been reported in recent years, and the cutoff value of the TyG index predicting cardiovascular events is 8.3 to 8.5 (24–27). In this study, the patients were divided into two groups according to the cutoff value of the baseline TyG index (dichotomized to ≥ or <8.5), based on a large-scale clinical trial using a database of the National Health and Nutrition Examination Survey (28).

A non-culprit lesion segment was defined as the proximal or distal site to the culprit lesion segment except for the 5-mm proximal and distal stent edges, and were imaged using grayscale IVUS and near-infrared spectroscopy and IVUS (NIRS-IVUS) pullback at both baseline and 6–12 months later. Grayscale IVUS and NIRS-IVUS were performed using the commercially available systems Altantis SR Pro2 (Boston Scientific, Natick, MA, USA), ViewIT (Terumo, Tokyo, Japan), and TVC Imaging System or Makoto Imaging System (Infraredx, Bedford, MA, USA). A grayscale IVUS and NIRS-IVUS catheter were inserted distal to the non-culprit lesion and pulled back at a rate of 0.5 mm/s, after intracoronary injection of nitroglycerin.

Quantitative grayscale IVUS measurements were performed using the Netra 3D iVUS system (AcImage, CA USA), VISIATRAS (Terumo, Tokyo, Japan), and QIvus version 2.1 (Medis, Leiden, the Netherlands) to quantify plaque volume according to two clinical expert consensus documents (29, 30). Baseline and follow-up IVUS images were reviewed side-by-side on a display, and target segments were selected. The quantitative IVUS analysis was accurately performed off-line by the cardiologists who were blinded to the clinical data. The lumen cross-sectional area (CSA) (31) and external elastic membrane (EEM) CSA were measured every frame; in addition, this software can measure vessel volume, and percentage atheroma volume (PAV) (%) = [∑ (EEM CSA − lumen CSA)/∑ EEM CSA] × 100.

All data were analyzed using JMP^® Pro, version 16.0.0 for Macintosh (SAS Institute, Cary, NC, USA). All probabilities were expressed as two-tailed values, with statistical significance set at p < 0.05. All confidence intervals (CIs) were computed at 95% level.

Categorical data were presented as number (percentage) and were compared using the chi-square test. Continuous variables were expressed as mean ± standard deviation or median (interquartile range) and compared using a one-way analysis of variance or the Kruskal–Wallis test. The Shapiro–Wilk test was used to examine whether the scores were likely to follow a certain distribution in all patients. If p < 0.05, the variables were not considered normally distributed. Univariable analysis of the relationship between the absolute change in PAV and the baseline TyG index was performed using the Pearson correlation analysis. The comparison of absolute change in PAV between the low-TyG-index group and high-TyG-index group was evaluated using analysis of covariance. Multivariable linear regression analyses of the absolute change in PAV and %change in the TyG index were performed to evaluate the predictors. A sensitivity analysis was performed to analyze how the value of the TyG index has affected the endpoints, and this study investigated whether similar trends were obtained within the TyG index range of 8.3 to 8.5 shown in the aforementioned reports (24–27).

Among 141 patients, 3 had severe obesity, 4 were not receiving LLT at baseline, 1 patient did not have a baseline TyG index, and 2 had insufficient analysis data. Ultimately, 131 patients completed the study, and their data were analyzed. According to the cutoff value of the baseline TyG index, the patients were divided into two groups; 96 patients were allocated to the high-TyG-index group (TyG index ≥8.5) and 35 patients were allocated to the low-TyG-index group (TyG index <8.5), as shown in Figure 1.

The baseline clinical characteristics of the patients are summarized in Table 1. The mean patient age was 63 ± 10 years, and 92% of the patients were men. Overall, hypertension was observed in 78%, DM in 59%, smoking history in 77%, and family history of premature CAD in 22%. Compared to the low-TyG-index group, the high-TyG-index group had higher BMI, triglyceride, fasting plasma glucose, HbA1c, and proportion of men, DM, and smoking history, as well as lower HDL-C, whereas there was no difference in LDL-C between both groups. Overall, the mean TyG index was 8.8 ± 0.6 at baseline and significantly rose to 9.4 ± 0.6 during the follow-up period regardless of the high-TyG-index group and the low-TyG-index group (p < 0.001).

IVUS measurements showed that there was an overall tendency for coronary plaque regression as shown in Table 2; the mean absolute change in PAV was −1.2% in the median non-culprit lesion length of 12 mm. There was no significant difference in the mean absolute change in PAV between the high-TyG-index group and the low-TyG-index group, although the high-TyG-index group had a significantly higher mean PAV at the time of follow-up, compared to the low-TyG-index group.

Figure 2 shows the correlation between absolute change in PAV and the baseline TyG index according to the presence or absence of aggressive LLT. Among the patients who received aggressive LLT, there was significant correlation with absolute change in PAV and the baseline TyG index (p = 0.049), whereas there was no significant correlation between these in patients who did not receive aggressive LLT. In addition, the absolute change in PAV between the low-TyG-index group and high-TyG-index group only in patients who received aggressive LLT was −3.5% ± 1.4% by analysis of covariance (p = 0.017). Furthermore, Table 3 shows the multiple linear regression analysis predicting the absolute change in PAV under aggressive LLT. As a result, baseline TyG index (estimate 2.71; 95% CIs 0.43, 5.00; p = 0.007), male (estimate 2.97; 95% CIs 0.32, 5.62; p = 0.029), and %change in LDL-C (estimate 0.102; 95% CIs 0.006, 0.197; p = 0.037) were significantly associated with the absolute change in PAV.

This study had several limitations that require consideration. First, unknown confounding factors may have affected the study outcomes regardless of the analytical adjustments, and the relatively small number of enrolled patients may have limited the statistical power of the study. Second, this study was limited to patients with stable CAD and ACS who underwent percutaneous coronary intervention for culprit lesions and were analyzed for non-culprit lesions. Thus, the validity of this causal relationship in healthy individuals and patients eligible for primary prevention is unclear. Third, this study only included Japanese patients, and the results and effects may differ between races. Fourth, we need to note heterogeneity as a limitation of this study. (1) Clinical heterogeneity: Supplementary Table 1 shows the comparison of baseline patients’ characteristics in three studies. The ENTERPRISE trial targeting stable patients with CAD had the highest BMI, triglyceride, and proportion of hypertension, and the lowest HDL-C and LDL-C. In the ESPECIAL-ACS study targeting patients with ACS, all patients were men and had DM, and the baseline TyG index was the highest. The ZEUS trial had the highest proportion of achieved aggressive LLT. (2) Methodological heterogeneity: It was unclear whether LLT was initiated recently or had been administered for many years. A total of 21 out of 44 patients in the ENTERPRISE trial and all patients in the ZEUS trial had recently received LLT, whereas details regarding when LLT was initiated were unclear in the ESPECIAL-ACS study. In addition, the follow-up period for the ENTERPRISE trial was 1 year, but for the ESPECIAL-ACS study and ZEUS trial, it was 6 months. (3) Statistical heterogeneity: Despite the above heterogeneity, the outcomes of all studies were standardized as the absolute change in PAV, and no significant variation in outcomes between studies was observed, even in patients who achieved aggressive LLT. Fifth, lifestyle factors and dietary habits are associated with coronary plaques as well as the TyG index; however, there were no data of studies on physical inactivity, heavy alcohol consumption, and diets high in carbohydrates, sugars, and red meat, and low in fruit and vegetable intake.