Abstract
Background
Recurrent ischemic events after acute myocardial infarction (AMI) may arise from treated culprit lesions (CLs) or untreated non-CLs (NCLs). However, temporal trends in CL- and NCL-attributed revascularization in contemporary clinical practice remain unclear.
Methods and Results
We analyzed data from the Mie ACS Registry, a multicenter registry enrolling patients with AMI who underwent primary percutaneous coronary intervention. Patients with previous coronary artery bypass grafting were excluded. The coprimary endpoints were the times to the first CL- and NCL-attributed revascularization over a 2-year follow-up period. Death was treated as a competing event. Temporal trends and predictors were evaluated using competing-risk analyses. Among 6,363 patients, 2,860 were enrolled in 2013–2017 and 3,503 were enrolled in 2018–2022. The cumulative incidence of CL-attributed revascularization was significantly lower in the later era (6.2% vs. 3.5%; P<0.001), whereas NCL-attributed revascularization showed only modest temporal improvement from 2013–2017 to 2018–2022 (6.6% vs. 5.5%, respectively; P=0.091). In multivariable analyses, late-era (2018–2022) enrollment was independently associated with a lower risk of CL-attributed revascularization (subdistribution hazard ratio [sHR] 0.56; 95% confidence interval [CI] 0.44–0.71; P<0.001), whereas multivessel disease was the strongest predictor of NCL-attributed revascularization (sHR 2.31; 95% CI 1.84–2.91; P<0.001).
Conclusions
Despite temporal reductions in CL-attributed revascularization, NCL-attributed events remained relatively unchanged. NCLs continue to represent an important source of residual coronary risk after AMI.
Key Words: Acute myocardial infarction, Culprit lesion, Non-culprit lesion, Revascularization
Central Figure.
Temporal trends in culprit and non-culprit lesion-attributed revascularization after acute myocardial infarction (AMI). Culprit lesion-attributed revascularization decreased significantly over time, whereas non-culprit lesion-attributed revascularization showed only limited temporal improvement. Non-culprit lesions remain an important source of residual coronary risk, particularly in patients with multivessel disease. CI, confidence interval; HR, hazard ratio; MI, myocardial infarction; PCI, percutaneous coronary intervention; UAP, unstable angina pectoris.
Despite considerable advances in reperfusion therapy and secondary prevention, patients with acute myocardial infarction (AMI) continue to experience recurrent ischemic events after primary percutaneous coronary intervention (PCI). These events remain a major determinant of long-term prognosis and a substantial clinical burden. Importantly, post-AMI ischemic events may arise either from failure of the treated culprit lesion or from progression of untreated non-culprit lesions, which reflect different biological substrates and may require distinct preventive strategies.
Growing mechanistic evidence has underscored the clinical relevance of non-culprit coronary disease. Intravascular imaging studies, including Providing Regional Observations to Study Predictors of Events in the Coronary Tree (PROSPECT) I and II, demonstrated that angiographically mild but lipid-rich, high-plaque-burden lesions are not benign and are associated with adverse cardiovascular events.1,2 These observations shifted the paradigm of residual risk after AMI from isolated culprit lesion failure towards the vulnerability of the entire coronary tree.
However, mechanistic insights do not directly translate into population-level clinical expression. Most previous studies were designed to identify high-risk plaque characteristics at the lesion level in select trial populations, rather than to quantify, at the patient level, how often recurrent ischemic events in routine clinical practice are actually attributed to culprit vs. non-culprit lesions. In this context, large-scale registry data have demonstrated that a substantial proportion of recurrent ischemic events originates from non-culprit lesions in real-world settings.3 Nevertheless, these studies did not address whether the relative contribution of culprit and non-culprit lesions has changed over time.
Over the past decade, AMI management has undergone considerable evolution, including the widespread adoption of new-generation drug-eluting stents, optimized antithrombotic regimens, and intensified lipid-lowering strategies. These advances have the potential to preferentially reduce ischemic events related to treated culprit lesions. In contrast, because non-culprit lesions represent diffuse and systemic coronary atherosclerosis, improvements in focal interventional techniques may have a more limited impact on their associated risk. Whether contemporary practice has meaningfully altered the burden of non-culprit lesion-related events in real-world settings remains unclear.
We therefore hypothesized that ischemic events arising from treated culprit lesions have declined in contemporary practice, whereas events attributed to non-culprit lesions persist and constitute a substantial component of residual risk after AMI. To test this hypothesis, we used a large, population-based registry to examine patient-level culprit and non-culprit lesion-attributed revascularization after AMI. Using a competing-risk framework, we examined temporal trends in culprit and non-culprit lesion-attributed revascularization and their clinical determinants in the Mie ACS Registry.
Methods
Study Population
The Mie ACS Registry is an ongoing prospective multicenter registry initiated in 2013 in Mie Prefecture, Japan, enrolling patients with acute coronary syndrome from 17 participating hospitals.
For the present study, patients were eligible for inclusion if they were diagnosed with AMI and underwent primary PCI. Patients with a prior history of coronary artery bypass grafting were excluded. A study flowchart is shown in Figure 1.
Figure 1.
Flowchart of patient selection and study design from the Mie ACS Registry. AMI, acute myocardial infarction; CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention.
The registry protocol was approved by the Institutional Review Board of Mie University Graduate School of Medicine (Reference no. H2023-017) as well as by the ethics committees of all participating institutions. Because of the observational nature of the registry, the requirement for written informed consent was waived, and an opt-out policy was adopted, in accordance with local ethical guidelines.
This study was registered with the University Hospital Medical Information Network (UMIN) Clinical Trials Registry (ID: UMIN000036020) and is being conducted in accordance with the principles of the Declaration of Helsinki.
For temporal trend analyses, patients were categorized into 2 eras according to the year of enrollment (2013–2017 and 2018–2022). The study period was divided into 2 5-year eras to facilitate comparison of temporal trends across the 10-year study period.
Definitions of Variables and Endpoints
The diagnosis of AMI was based on the fourth universal definition of myocardial infarction (MI).4 Multivessel disease was defined angiographically by visual assessment as the presence of ≥75% diameter stenosis in ≥2 major epicardial coronary arteries or ≥50% diameter stenosis in the left main trunk.
The culprit lesion was defined as the coronary lesion responsible for the index MI, as determined by angiographic findings, electrocardiographic territory, and clinical presentation. Non-culprit lesions were defined as angiographically significant coronary lesions other than the culprit lesion identified at the index MI.
Staged or planned revascularization procedures for non-culprit lesions that were prespecified during the index hospitalization were considered part of the initial treatment strategy and were not counted as endpoint events, regardless of whether they were performed during hospitalization or after discharge. Repeat PCI targeting lesions that had already been treated before or during the index hospitalization was also not considered an endpoint event. To distinguish non-culprit lesion-attributed events from restenosis or progression adjacent to the treated culprit lesion, non-culprit lesions were defined as previously untreated coronary segments located at least 5 mm away from the index culprit lesion treatment site. In contrast, PCI for non-culprit lesions that was newly indicated during follow-up and not prespecified at the index hospitalization was adjudicated as a clinical endpoint.
Outcome data were collected through outpatient clinic visits, hospital chart reviews, or telephone interviews with the patient or close relatives. Clinical events and lesion attribution were determined by attending cardiologists at each participating institution on the basis of clinical presentation, angiographic findings, and available medical records. No independent central adjudication committee was used. Follow-up data were available after hospital discharge for 98.1% of patients, and 86.1% of patients completed the planned 2-year follow-up. Patients who were lost to follow-up were censored at the time of last contact.
All endpoints were defined at the patient level rather than the lesion level. The coprimary endpoints were the time to first culprit lesion-attributed revascularization and time to first non-culprit lesion-attributed revascularization over a 2-year follow-up period. Culprit lesion-attributed revascularization was defined as revascularization due to MI (fatal or non-fatal), unstable angina pectoris, progressive angina, or rapid lesion progression. Non-culprit lesion-attributed revascularization was defined using the same clinical criteria but was restricted to unplanned procedures involving previously untreated non-culprit lesions. Fatal MI was included only when the culprit or non-culprit vessel could be identified based on angiography, electrocardiographic territory, imaging findings, or autopsy.
Secondary endpoints were also defined at the patient level. The first secondary endpoint was the time to the first culprit lesion-attributed revascularization due to MI (fatal or non-fatal) or unstable angina pectoris within 2 years. The second secondary endpoint was the time to the first non-culprit lesion-attributed revascularization, defined using the same criteria as for the coprimary endpoint.
Statistical Analysis
Normally distributed continuous variables are presented as the mean ± SD, whereas variables with a skewed distribution are presented as the median with interquartile range. Categorical variables are summarized as numbers and percentages. Between-group comparisons were performed using the Chi-squared test for categorical variables and Student’s t-test or the Wilcoxon rank-sum test for continuous variables, as appropriate.
Because death precludes the occurrence of subsequent revascularization, it was treated as a competing event. Because the primary objective of this study was to evaluate the patient-level cumulative incidence of first culprit or non-culprit lesion-attributed revascularization over a 2-year follow-up period, cumulative incidence functions accounting for death as a competing risk were estimated using the Aalen–Johansen method and compared between eras using Gray’s test.
Multivariable analyses were conducted using Fine–Gray subdistribution hazard models to assess the association between clinical covariates and cumulative incidence in the presence of competing risks. Because assessment of temporal trends was a primary objective of this study, study era (2013–2017 vs. 2018–2022) was included in all models regardless of its significance in unadjusted analyses. Other covariates were selected a priori based on established clinical relevance and previous literature, rather than by univariate screening. Separate models were constructed for culprit lesion-attributed and non-culprit lesion-attributed revascularization, with additional PCI- and lesion-related variables incorporated according to their mechanistic relevance to each endpoint. To avoid model overfitting, the number of covariates was limited to clinically essential variables.
Missing data were not imputed, and multivariable analyses were performed using complete-case data. Missingness was highest for B-type natriuretic peptides (45.7%), followed by HbA1c (18.1%), lipid parameters (11.2–14.1%), and left ventricular ejection fraction (7.9%). The missingness of variables included in multivariable models is provided in the Supplementary Table.
All statistical tests were 2-sided, and P<0.05 was considered statistically significant. Statistical analyses were performed using EZR (Jichi Medical University, Tochigi, Japan), a graphical user interface for R (R Foundation for Statistical Computing, Vienna, Austria).5
Results
Baseline Patient Characteristics
In all, 6,363 patients with AMI who underwent primary PCI were included in this analysis. Of these patients, 2,860 were enrolled in the early era (2013–2017) and 3,503 were enrolled in the late era (2018–2022). Baseline patient characteristics are summarized in Table 1.
Table 1.
Baseline Clinical, Laboratory, Angiographic, and Procedural Characteristics
| All patients (n=6,363) |
2013–2017 (n=2,860) |
2018–2022 (n=3,503) |
P value | |
|---|---|---|---|---|
| Baseline patient characteristics | ||||
| Age (years) | 71.0 (60.0–78.0) | 70.0 (60.0–78.0) | 72.0 (62.0–80.0) | <0.001 |
| Male sex | 4,943 (77.7) | 2,228 (77.9) | 2,715 (77.5) | 0.717 |
| BMI (kg/m2) | 23.3 (21.2–25.6) | 23.1 (21.2–25.5) | 23.3 (21.2–25.7) | 0.049 |
| Cardiovascular risk factors | ||||
| Hypertension | 3,998 (62.8) | 1,829 (64.0) | 2,169 (61.9) | 0.095 |
| Diabetes | 2,072 (32.6) | 905 (31.6) | 1,167 (33.3) | 0.162 |
| Dyslipidemia | 2,988 (47.0) | 1,375 (48.1) | 1,613 (46.0) | 0.112 |
| Current smoking | 1,799 (28.0) | 877 (30.7) | 902 (25.7) | <0.001 |
| Hyperuricemia | 358 (5.6) | 180 (6.3) | 178 (5.1) | 0.038 |
| Family history of CAD | 430 (6.8) | 153 (5.3) | 277 (7.9) | <0.001 |
| Hemodialysis | 109 (1.7) | 38 (1.3) | 71 (2.0) | 0.033 |
| Prior MI | 535 (8.4) | 239 (8.4) | 296 (8.4) | 0.928 |
| Prior PCI | 615 (9.7) | 267 (9.3) | 348 (9.9) | 0.443 |
| Prior cerebral infarction | 331 (5.2) | 147 (5.1) | 184 (5.3) | 0.865 |
| Laboratory data and discharge medication | ||||
| Laboratory data | ||||
| Total cholesterol (mg/dL) | 193.0 (165.0–222.0) | 190.0 (164.0–219.0) | 195.0 (166.0–225.0) | 0.001 |
| Triglycerides (mg/dL) | 111.0 (74.0–168.0) | 104.0 (71.0–161.0) | 115.0 (77.0–173.75) | <0.001 |
| HDL-C (mg/dL) | 48.9 (41.0–57.5) | 48.0 (40.0–57.0) | 49.0 (41.0–58.0) | 0.001 |
| LDL-C (mg/dL) | 120.0 (97.0–145.0) | 118.5 (95.85–143.0) | 121.0 (98.0–146.0) | 0.027 |
| Creatinine (mg/dL) | 0.87 (0.73–1.08) | 0.85 (0.71–1.06) | 0.89 (0.74–1.10) | <0.001 |
| Hemoglobin (g/dL) | 14.2 (12.7–15.4) | 14.2 (12.7–15.5) | 14.1 (12.6–15.4) | 0.277 |
| Glucose (mg/dL) | 154.0 (125.0–203.0) | 153.0 (125.0–203.75) | 155.0 (125.0–202.5) | 0.814 |
| HbA1c (%) | 6.0 (5.7–6.8) | 6.0 (5.6–6.7) | 6.1 (5.7–6.9) | <0.001 |
| Peak CPK (IU/L) | 1,581.0 (600.0–3,260.0) | 1,728.5 (689.5–3,454.75) | 1,457.0 (537.0–3,100.5) | <0.001 |
| BNP (pg/mL) | 80.7 (26.2–255.62) | 84.3 (29.3–251.95) | 78.2 (24.8–261.0) | 0.441 |
| CRP (mg/dL) | 0.18 (0.07–0.60) | 0.18 (0.08–0.58) | 0.18 (0.07–0.62) | 0.891 |
| Medication at discharge | ||||
| Aspirin | 6,187 (97.3) | 2,800 (97.9) | 3,387 (96.7) | 0.004 |
| Clopidogrel | 1,773 (27.9) | 1,374 (48.1) | 399 (11.4) | <0.001 |
| Prasugrel | 4,318 (67.9) | 1,324 (46.3) | 2,994 (85.5) | <0.001 |
| Warfarin | 464 (7.3) | 230 (8.0) | 234 (6.7) | 0.042 |
| DOAC | 420 (6.6) | 100 (3.5) | 320 (9.1) | <0.001 |
| ACEi/ARB/ARNI | 4,893 (77.0) | 2,355 (82.4) | 2,538 (72.5) | <0.001 |
| β-blocker | 3,437 (54.0) | 1,635 (57.2) | 1,802 (51.5) | <0.001 |
| MRA | 630 (9.9) | 226 (7.9) | 404 (11.5) | <0.001 |
| Calcium channel blocker | 941 (14.8) | 445 (15.6) | 496 (14.2) | 0.118 |
| Statin | 5,570 (87.6) | 2,440 (85.3) | 3,130 (89.4) | <0.001 |
| Ezetimibe | 712 (11.2) | 112 (3.9) | 600 (17.1) | <0.001 |
| Fibrate | 36 (0.6) | 13 (0.5) | 23 (0.7) | 0.317 |
| EPA/DHA | 235 (3.7) | 144 (5.0) | 91 (2.6) | <0.001 |
| Insulin | 355 (5.6) | 181 (6.3) | 174 (5.0) | 0.021 |
| Oral antidiabetic | 1,396 (21.9) | 574 (20.1) | 822 (23.5) | 0.001 |
| Angiographic data and PCI procedure | ||||
| Diagnosis | ||||
| STEMI | 4,823 (76.0) | 2,228 (78.3) | 2,595 (74.1) | <0.001 |
| Killip class | ||||
| 1 | 4,935 (77.6) | 2,257 (78.9) | 2,678 (76.4) | 0.002 |
| 2/3 | 864 (13.6) | 392 (13.7) | 472 (13.5) | |
| 4 | 564 (8.9) | 211 (7.4) | 353 (10.1) | |
| Cardiopulmonary arrest at presentation | 276 (4.3) | 124 (4.3) | 152 (4.3) | 1 |
| IABP | 789 (12.4) | 357 (12.5) | 432 (12.3) | 0.819 |
| VA-ECMO | 137 (2.2) | 51 (1.8) | 86 (2.5) | 0.082 |
| Impella | 35 (0.6) | 0 (0) | 35 (1.0) | <0.001 |
| LVEF (%) | 56.9 (48.0–64.6) | 57.0 (48.18–65.0) | 56.1 (48.0–64.0) | 0.059 |
| Angiographic data | ||||
| Type of acute MI | ||||
| 1 | 6,125 (96.6) | 2,732 (96.2) | 3,393 (96.9) | 0.213 |
| 2 | 120 (1.9) | 56 (2.0) | 64 (1.8) | |
| 4b | 83 (1.3) | 42 (1.5) | 41 (1.2) | |
| Culprit artery | ||||
| LMT | 143 (2.2) | 47 (1.6) | 96 (2.7) | 0.004 |
| LAD | 2,980 (46.9) | 1,307 (45.7) | 1,673 (47.8) | 0.101 |
| RCA | 2,278 (35.8) | 1,065 (37.3) | 1,213 (34.7) | 0.031 |
| LCX | 936 (14.7) | 427 (14.9) | 509 (14.5) | 0.669 |
| Others | 20 (0.3) | 12 (0.4) | 8 (0.2) | 0.185 |
| Unknown | 1 (0.0) | 0 (0) | 1 (0.0) | 0.1 |
| Pre-PCI TIMI 0/1 flow grade | 4,316 (67.9) | 1,960 (68.7) | 2,356 (67.3) | 0.245 |
| Multivessel disease | 3,091 (48.7) | 1,371 (48.2) | 1,720 (49.1) | 0.449 |
| Culprit-lesion-only PCI | 5,245 (82.4) | 2,376 (83.1) | 2,869 (81.9) | 0.233 |
| PCI procedure | ||||
| Aspiration | 3,798 (61.2) | 1,869 (66.7) | 1,929 (56.7) | <0.001 |
| Distal protection | 463 (7.6) | 276 (10.0) | 187 (5.5) | <0.001 |
| BMS usage | 328 (5.2) | 304 (10.7) | 24 (0.7) | <0.001 |
| DES usage | 5,288 (83.7) | 2,299 (80.7) | 2,989 (86.1) | <0.001 |
| Final TIMI 3 flow grade | 5,569 (91.1) | 2,509 (90.9) | 3,060 (91.2) | 0.787 |
Unless indicated otherwise, data are given as the median [interquartile range] or n (%). Missingness was highest for B-type natriuretic peptide (BNP; 45.7%), followed by HbA1c (18.1%), lipid parameters (11.2–14.1%), and left ventricular ejection fraction (LVEF; 7.9%). ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; ARNI, angiotensin receptor–neprilysin inhibitor; BMI, body mass index; BMS, bare-metal stent; BNP, B-type natriuretic peptide; CAD, coronary artery disease; CPK, creatine phosphokinase; CRP, C-reactive protein; DES, drug-eluting stent; DOAC, direct oral anticoagulant; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; HDL-C, high-density lipoprotein cholesterol; IABP, intra-aortic balloon pumping; LAD, left anterior descending artery; LCX, left circumflex artery; LDL-C, low-density lipoprotein cholesterol; LMT, left main trunk; LVEF, left ventricular ejection fraction; MI, myocardial infarction; MRA, mineralocorticoid receptor antagonist; PCI, percutaneous coronary intervention; RCA, right coronary artery; STEMI, ST elevation myocardial infarction; TIMI, Thrombolysis in Myocardial Infarction trial; VA-ECMO, venous-arterial extracorporeal membrane.
Patients in the late era were slightly older than those in the early era (median age 72 vs. 70 years, respectively; P<0.001). The proportion of male patients was similar between the 2 eras. Current smoking was less frequent, whereas hemodialysis and a family history of coronary artery disease were more prevalent in the late era. Other major cardiovascular risk factors, including hypertension, diabetes, dyslipidemia, previous MI, and previous PCI, were comparable between the 2 eras.
Laboratory data and discharge medication are presented in Table 1. Regarding discharge medications, the late era was characterized by a marked shift in antiplatelet therapy, with a decreased use of clopidogrel and an increased use of prasugrel. The use of statins and ezetimibe was significantly higher in the late era, reflecting more intensive lipid-lowering strategies.
Angiographic Findings and PCI Procedures
Angiographic characteristics and procedural details are summarized in Table 1. In the late era, ST-elevation MI was less frequent, whereas presentation with Killip Class 4 was more common. The prevalence of left main trunk culprit lesions was higher, whereas that of right coronary artery culprit lesions was lower, in the late era. The prevalence of multivessel disease, defined angiographically based on visual assessment, was similar between the 2 eras. Procedural strategies evolved over time, with decreased use of aspiration thrombectomy and bare-metal stents in the late era.
Clinical Outcomes
Clinical outcomes are summarized in Table 2. In-hospital, 30-day, and all-cause mortality did not differ significantly between the 2 eras. Cardiovascular mortality and the composite of cardiac death and non-fatal MI were also similar between the eras.
Table 2.
Outcome Data
| All patients (n=6,363) |
2013–2017 (n=2,860) |
2018–2022 (n=3,503) |
P value | |
|---|---|---|---|---|
| Duration in hospital (days) | 12.57 [9.0–17.0] | 13.58 [10.0–18.0] | 12.0 [8.51–16.0] | <0.001 |
| In-hospital mortality | 383 (6.0) | 166 (5.8) | 217 (6.2) | 0.525 |
| All-cause mortality | 779 (12.2) | 351 (12.3) | 428 (12.2) | 0.969 |
| Cardiovascular death | 345 (5.4) | 148 (5.2) | 197 (5.6) | 0.437 |
|
Revascularization for fatal MI, non-fatal MI, UAP, or stable angina |
671 (11.5) | 358 (12.5) | 313 (8.9) | <0.001 |
| Fatal MI | 31 (0.5) | 17 (0.6) | 14 (0.4) | 0.282 |
| Non-fatal MI | 122 (1.9) | 51 (1.8) | 71 (2.0) | 0.521 |
| UAP requiring revascularization | 70 (1.1) | 33 (1.2) | 37 (1.1) | 0.719 |
| Stable angina pectoris requiring revascularization | 537 (8.8) | 294 (10.7) | 243 (7.3) | <0.001 |
Unless indicated otherwise, data are given as the median [interquartile range] or n (%). MI, myocardial infarction; UAP, unstable angina pectoris.
In contrast, the incidence of revascularization for ischemic events (fatal and non-fatal MI, unstable or stable angina) was significantly lower in the late than early era (8.9% vs. 12.5%, respectively; P<0.001), driven primarily by a reduction in revascularization for stable angina. Hospitalization for heart failure was also less frequent in the late era.
Cumulative Incidence of Culprit and Non-Culprit Lesion-Attributed Events
Figure 2 shows the cumulative incidence of first culprit and non-culprit lesion-attributed revascularization for all ischemic events. The cumulative incidence of culprit lesion-attributed revascularization was significantly lower in the late era (Gray’s test P<0.001), whereas non-culprit lesion-attributed revascularization showed no significant temporal difference, despite a numerical reduction from the early to late era (Gray’s test P=0.091).
Figure 2.
Cumulative incidence of first culprit (CL; Left) and non-culprit (NCL; Right) lesion-attributed revascularization for all ischemic events, namely fatal (fMI) and non-fatal (nMI) myocardial infarction, unstable angina pectoris (UAP), and stable angina, over a 2-year follow-up period. Comparisons between 2013–2017 and 2018–2022 were performed using Gray’s test, with death treated as a competing risk. CL-attributed revascularization decreased significantly in the contemporary era (2018–2022), whereas NCL-attributed revascularization showed only a modest temporal improvement. CI, confidence interval.
In analyses limited to severe ischemic clinical endpoints (fatal or non-fatal MI and unstable angina), neither culprit nor non-culprit lesion-attributed events differed significantly between eras (Figure 3).
Figure 3.
Cumulative curves for the time to the first culprit lesion (CL)-attributed (Left) and non-culprit lesion (NCL)-attributed (Right) revascularization for severe ischemic events, namely fatal (fMI) and non-fatal (nMI) myocardial infarction, and unstable angina pectoris (UAP), over a 2-year follow-up period. Comparisons between 2013–2017 and 2018–2022 were performed using Gray’s test, with death treated as a competing risk. No significant temporal differences were observed for severe ischemic events attributed to either CLs or NCLs. CI, confidence interval.
Multivariable Competing-Risk Analysis
Results of multivariable Fine–Gray analyses are presented in Table 3. Enrollment in the late era was independently associated with a significantly lower risk of culprit lesion-attributed revascularization (subdistribution hazard ratio [sHR] 0.561; 95% confidence interval [CI] 0.440–0.714; P<0.001). Hemodialysis, multivessel disease, and non-ST segment elevation MI were independently associated with an increased risk of culprit lesion-attributed revascularization, whereas the use of drug-eluting stents was independently associated with a lower risk.
Table 3.
Multivariate Fine–Gray Analysis for Predicting Culprit and Non-Culprit Lesion-Attributed Revascularization During Follow-up
| Covariate | Subdistribution HR (95% CI) |
P value |
|---|---|---|
| Culprit lesion-attributed revascularization | ||
| 2018–2022 (vs. 2013–2017) | 0.561 (0.440–0.714) | <0.001 |
| Age (per 1-year increase) | 0.993 (0.983–1.004) | 0.210 |
| Male sex | 1.170 (0.834–1.641) | 0.360 |
| BMI (per 1-kg/m2 increase) | 1.012 (0.980–1.044) | 0.480 |
| Hypertension | 0.910 (0.708–1.170) | 0.460 |
| Diabetes | 1.002 (0.775–1.294) | 0.990 |
| Dyslipidemia | 1.086 (0.851–1.385) | 0.510 |
| Hemodialysis | 2.780 (1.534–5.037) | <0.001 |
| Current smoking | 1.221 (0.931–1.600) | 0.150 |
| NSTEMI | 1.392 (1.078–1.797) | 0.011 |
| Multivessel disease | 1.687 (1.320–2.155) | <0.001 |
| DES usage | 0.478 (0.370–0.619) | <0.001 |
| Non-culprit lesion-attributed revascularization | ||
| 2018–2022 (vs. 2013–2017) | 0.810 (0.654–1.003) | 0.054 |
| Age (per 1-year increase) | 0.981 (0.971–0.990) | <0.001 |
| Male sex | 0.945 (0.718–1.245) | 0.690 |
| BMI (per 1-kg/m2 increase) | 0.994 (0.964–1.025) | 0.690 |
| Hypertension | 1.101 (0.875–1.385) | 0.410 |
| Diabetes | 1.105 (0.883–1.382) | 0.380 |
| Dyslipidemia | 1.051 (0.848–1.302) | 0.650 |
| Hemodialysis | 0.704 (0.295–1.683) | 0.430 |
| Current smoking | 0.906 (0.705–1.164) | 0.440 |
| NSTEMI | 1.289 (1.022–1.627) | 0.032 |
| Multivessel disease | 2.310 (1.835–2.909) | <0.001 |
| LAD culprit | 0.780 (0.627–0.972) | 0.027 |
CI, confidence interval; HR, hazard ratio; NSTEMI, non-ST segment elevation myocardial infarction. Other abbreviations as in Table 1.
For non-culprit attributed revascularization, multivessel disease emerged as the strongest predictor (sHR 2.310; 95% CI 1.835–2.909; P<0.001). Older age and a left anterior descending artery culprit lesion were associated with lower risk. Era was not independently associated with non-culprit lesion-attributed revascularization.
Discussion
This multicenter registry of patients with AMI undergoing primary PCI has 3 principal findings. First, culprit lesion-attributed revascularization decreased significantly over time. Second, non-culprit lesion-attributed revascularization showed only modest temporal improvement and remained an important source of residual coronary risk. Third, late-era enrollment was independently associated with a lower risk of culprit lesion-attributed revascularization, whereas multivessel disease was the strongest predictor of non-culprit lesion-attributed revascularization. These findings provide contemporary real-world insights into the evolution of culprit and non-culprit lesion-related risk after AMI.
Culprit lesion-attributed revascularization declined significantly in the contemporary era. The event curve began to diverge between 6 months and 1 year after the index PCI. These findings likely reflect cumulative improvements in culprit lesion management. In particular, the transition from bare-metal stents to drug-eluting stents has substantially reduced restenosis and the need for repeat revascularization, although a residual risk of restenosis requiring repeat revascularization remains.6 In parallel, post-PCI follow-up strategies have evolved over time. Routine surveillance coronary angiography is no longer recommended in most patients,7,8 and an increasing use of physiology-guided ischemia assessment may have contributed to more selective repeat revascularization during follow-up.9 These changes may explain why the reduction in culprit lesion-attributed events was driven mainly by decreases in stable angina-related revascularization, whereas more severe ischemic events, such as MI and unstable angina, remained largely unchanged. Further refinements in PCI techniques and optimization of secondary prevention therapies may have further contributed to these improvements.10–13 Together, these findings suggest that contemporary practice, characterized by improved stent performance and more selective follow-up strategies, may reduce repeat revascularization without an apparent increase in severe ischemic events.
Non-culprit lesion-attributed revascularization showed only limited temporal improvement in this study. Although a numerical reduction was noted in the contemporary era, the difference did not reach statistical significance. These findings suggest that residual coronary risk attributable to untreated non-culprit lesions persists even in the contemporary era, despite advances in medical therapy and interventional care. Previous intravascular imaging studies provide potential mechanistic insights into this observation. In the PROSPECT II study, non-culprit lesions with a high lipid content and large plaque burden were associated with subsequent adverse events even when angiographic stenosis was not severe.2 In addition, the Preventive Percutaneous Coronary Intervention Versus Optimal Medical Therapy Alone for the Treatment of Vulnerable Atherosclerotic Coronary Plaques (PREVENT) trial suggested the potential benefit of PCI for vulnerable non-culprit plaques identified by intracoronary imaging despite physiologically non-significant stenosis.14 Together, these observations may help explain the persistence of non-culprit lesion-related ischemic events and highlight the possibility that strategies beyond conventional ischemia-guided revascularization could further mitigate residual risk.
When analyses were restricted to severe ischemic endpoints, including MI and unstable angina, neither culprit nor non-culprit lesion-attributed events differed significantly between eras. This finding suggests that the observed temporal reduction in repeat revascularization was driven mainly by reductions in stable angina-related revascularization, whereas severe ischemic events remained relatively infrequent in both eras. Therefore, the observed temporal reduction in culprit lesion-attributed revascularization should be interpreted cautiously, because it may partly reflect changes in follow-up practice and revascularization decision-making rather than a true reduction in severe ischemic events.
Previous mechanistic studies, particularly PROSPECT I and II,1,2 demonstrated that non-culprit lesions with high-risk plaque characteristics contribute substantially to future ischemic events. However, these studies were conducted in select trial populations and focused primarily on lesion-level mechanisms assessed by intravascular imaging.1,2 In contrast, data evaluating patient-level ischemic events attributed to culprit and non-culprit lesions in contemporary real-world clinical practice remain limited. To our knowledge, large-scale registry studies systematically evaluating culprit vs. non-culprit lesion-attributed events using a competing-risk framework are lacking. In this context, the present study extends prior mechanistic observations by providing population-based, real-world evidence regarding the temporal evolution of culprit and non-culprit lesion-related risk after AMI.
The different temporal trends across the 2 study eras observed for culprit and non-culprit lesions likely reflect distinct clinical and pathophysiological characteristics. Culprit lesions are directly treated during the index procedure and therefore derive the greatest benefit from advances in interventional strategies. In contrast, non-culprit lesions reflect the overall burden of coronary atherosclerosis and long-term systemic risk factors, and therefore may not uniformly benefit from interventional approaches. Consistent with this concept, multivessel disease emerged as the strongest predictor of non-culprit lesion-attributed revascularization in the present study, highlighting the importance of global coronary disease severity rather than focal lesion characteristics alone.
The present findings have important clinical implications for the management of multivessel disease in AMI. Although continued optimization of culprit-lesion PCI remains essential, our findings suggest that residual coronary risk from non-culprit lesions persists even in the contemporary era. Consistent with this concept, multivessel disease emerged as the strongest predictor of non-culprit lesion-attributed revascularization in the present study. In addition, culprit-lesion-only PCI remained common in both eras, which may also have contributed to the persistent burden of non-culprit lesion-attributed events. Further reduction of post-AMI coronary risk may therefore require complementary approaches beyond conventional lesion-based revascularization, including intensified secondary prevention, improved risk stratification, ongoing risk assessment, and selective preventive interventions for carefully identified high-risk non-culprit lesions.15
Although derived from a single prefecture, the Mie ACS Registry provides near-complete capture of AMI cases within Mie Prefecture, which accounts for approximately one-seventieth of the Japanese population. Importantly, the age distribution of the population in Mie Prefecture closely mirrors that of Japan as a whole, supporting the representativeness of this cohort as a model of national demographic trends in AMI. Accordingly, the findings of this study are likely to reflect broader real-world patterns in contemporary AMI care in Japan.
Study Limitations
Several limitations warrant consideration. First, as an observational registry study, residual confounding cannot be excluded despite multivariable adjustment. Second, lesion attribution was based on clinical and angiographic assessment rather than a standardized imaging protocol, which may have introduced some degree of uncertainty in event classification. In addition, follow-up strategies, ischemia assessment, and decisions regarding repeat revascularization were left to the discretion of the attending physicians and were not standardized across participating centers. Third, detailed lesion-level information, including lesion complexity, physiologic significance, surveillance coronary angiography, and completeness of revascularization, was not systematically collected. Therefore, the clinical appropriateness of culprit-lesion-only PCI and the mechanistic basis of temporal changes in repeat revascularization could not be evaluated. Fourth, the number of severe ischemic events was relatively modest, potentially limiting statistical power for secondary analyses. Finally, although the registry is population-based, the findings may not be fully generalizable to healthcare systems outside Japan.
Conclusions
In contemporary practice, culprit lesion-attributed revascularization after AMI has declined, whereas non-culprit lesion-attributed events have shown only limited temporal improvement. These findings highlight the need for strategies addressing both treated culprit lesions and untreated non-culprit disease to further improve long-term outcomes after AMI.
Disclosures
The authors have no conflicts of interest to disclose concerning this study.
IRB Information
This study was approved by the Mie University Hospital Institutional Review Board (Reference no. H2023-017).
Supplementary Files
Supplementary Table.
Acknowledgments
During the preparation of this manuscript, the authors used Google Gemini to generate illustrative elements for the central figure. The scientific content and interpretation were determined by the authors.
Funding Statement
Sources of Funding: This study did not receive any specific funding.
Data Availability
The deidentified participant data will not be shared.
References
- 1. Stone GW, Maehara A, Lansky AJ, Bruyne B, Cristea E, Mintz GS, et al.. A prospective natural-history study of coronary atherosclerosis. N Engl J Med 2011; 364: 226–235, doi:10.1056/nejmoa1002358. [DOI] [PubMed] [Google Scholar]
- 2. Erlinge D, Maehara A, Ben-Yehuda O, Bøtker HE, Maeng M, Kjøller-Hansen L, et al.. Identification of vulnerable plaques and patients by intracoronary near-infrared spectroscopy and ultrasound (PROSPECT II): A prospective natural history study. Lancet 2021; 397: 985–995, doi:10.1016/S0140-6736(21)00249-X. [DOI] [PubMed] [Google Scholar]
- 3. Varenhorst C, Johansson S, Janzon M, Albertsson P, Leosdottir M, Hambraeus K, et al.. Culprit and nonculprit recurrent ischemic events in patients with myocardial infarction: Data from SWEDEHEART (Swedish Web System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies). J Am Heart Assoc 2018; 7: e007174, doi:10.1161/JAHA.117.007174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Thygesen K, Alpert JS, Jaffe AS, Chaitman BR, Bax JJ, Morrow DA, et al.. Fourth universal definition of myocardial infarction (2018). J Am Coll Cardiol 2018; 72: 2231–2264, doi:10.1016/j.jacc.2018.08.1038. [DOI] [PubMed] [Google Scholar]
- 5. Kanda Y.. Investigation of the freely available easy-to-use software “EZR” for medical statistics. Bone Marrow Transplant 2013; 48: 452–458, doi:10.1038/bmt.2012.244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Giustino G, Colombo A, Camaj A, Yasumura K, Mehran R, Stone GW, et al.. Coronary in-stent restenosis: JACC state-of-the-art review. J Am Coll Cardiol 2022; 80: 348–372, doi:10.1016/j.jacc.2022.05.017. [DOI] [PubMed] [Google Scholar]
- 7. Wennberg E, Abualsaud AO, Eisenberg MJ.. Patient management following percutaneous coronary intervention. JACC Adv 2025; 4: 101453, doi:10.1016/j.jacadv.2024.101453. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Mindrescu C, Brener SJ, Guerchicoff A, Fahy M, Parise H, Mehran R, et al.. Impact of scheduled angiographic follow-up in patients treated with primary percutaneous coronary intervention for ST-segment elevation myocardial infarction. J Interv Cardiol 2013; 26: 319–324, doi:10.1111/joic.12038. [DOI] [PubMed] [Google Scholar]
- 9. Tonino PAL, De Bruyne B, Pijls NHJ, Siebert U, Ikeno F, van’t Veer M, et al.. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med 2009; 360: 213–224, doi:10.1056/NEJMoa0807611. [DOI] [PubMed] [Google Scholar]
- 10. Cannon CP, Blazing MA, Giugliano RP, McCagg A, White JA, Theroux P, et al.. Ezetimibe added to statin therapy after acute coronary syndromes. N Engl J Med 2015; 372: 2387–2397, doi:10.1056/nejmoa1410489. [DOI] [PubMed] [Google Scholar]
- 11. Schwartz GG, Steg PG, Szarek M, Bhatt DL, Bittner VA, Diaz R, et al.. Alirocumab and cardiovascular outcomes after acute coronary syndrome. N Engl J Med 2018; 379: 2097–2107, doi:10.1056/NEJMoa1801174. [DOI] [PubMed] [Google Scholar]
- 12. Kurobe M, Baba K, Nunohiro T, Ishizaki M, Furudono S, Nakata T, et al.. Impact of implementation of a region-wide low-density lipoprotein cholesterol management clinical pathway for the secondary prevention of acute myocardial infarction. Circ J 2024; 88: 1825–1832. [DOI] [PubMed] [Google Scholar]
- 13. Nishihira K, Nakai M, Kuriyama N, Kadooka K, Honda Y, Emori H, et al.. Guideline-directed medical therapy for elderly patients with acute myocardial infarction who undergo percutaneous coronary intervention. Circ J 2024; 88: 931–937. [DOI] [PubMed] [Google Scholar]
- 14. Park SJ, Ahn JM, Kang DY, Yun SC, Ahn YK, Kim WJ, et al.. Preventive percutaneous coronary intervention versus optimal medical therapy alone for the treatment of vulnerable atherosclerotic coronary plaques (PREVENT): A multicentre, open-label, randomised controlled trial. Lancet 2024; 403: 1753–1765, doi:10.1016/S0140-6736(24)00413-6. [DOI] [PubMed] [Google Scholar]
- 15. Lawler PR, Bhatt DL, Godoy LC, Lüscher TF, Bonow RO, Verma S, et al.. Targeting cardiovascular inflammation: Next steps in clinical translation. Eur Heart J 2021; 42: 113–131, doi:10.1093/eurheartj/ehaa099. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table.
Data Availability Statement
The deidentified participant data will not be shared.