A comparison of long‐term clinical outcomes between percutaneous coronary intervention (PCI) and medical therapy in patients with chronic total occlusion in noninfarct‐related artery after PCI of acute myocardial infarction

Abstract Background Chronic total occlusion (CTO) in a noninfarct‐related artery (IRA) is one of the risk factors for mortality after acute myocardial infarction (AMI). However, there are limited data comparing the long‐term outcomes of patients undergoing percutaneous coronary intervention (PCI) with patients having medical therapy (MT) in CTO lesion after AMI PCI. Methods We retrospectively enrolled 330 patients (successful CTO PCI in 166 patients, failed CTO PCI in 32 patients, MT in 132 patients) with non‐IRA CTO from a total of 4372 patients who underwent PCI after AMI in our center. Propensity score matching (PSM) was used to adjust for baseline differences. Results The primary analysis is based on the intention‐to‐treat population. During a median follow‐up period of 946 days, patients in the PCI group (n = 198) had significantly higher cardiac death‐free survival (96.6% vs. 82.8%, p = .004) compared with patients in MT group (n = 132). However, no significant difference in the occurrence of cardiac death was observed after PSM. The analysis based on the per‐protocol population demonstrated significantly higher cardiac death‐free survival in the successful CTO PCI group (n = 166) compared with the occluded CTO group (n = 164) both before and after PSM. In subgroup analysis, successful CTO PCI was associated with less cardiac death in patients over 65 years old, with LVEF < 50%, left anterior descending (LAD) IRA, and non‐LAD CTO lesion compared with occluded CTO group. Conclusions Patients undergoing successful revascularization of non‐IRA CTO after AMI might have a better long‐term prognosis. Moreover, patients with LVEF < 50% may benefit from successful non‐IRA CTO PCI after AMI.


| INTRODUCTION
In the contemporary practice, among patients with acute myocardial infarction (AMI) undergoing percutaneous coronary intervention (PCI), around 50%-60% present with multivessel disease (MVD) 1,2 and 8%-13% have concurrent chronic total occlusion (CTO) lesion. 3,4 Previous studies have shown that AMI with MVD was associated with worse clinical outcomes and complete revascularization in these patients will lead to reduced adverse cardiovascular events. 3,5 MVD with a coexisting CTO lesion in a noninfarct-related artery(non-IRA) is an independent predictor for long-term mortality in AMI patients, 3,4,6 and one study even reported that the presence of CTO alone but not MVD is associated with long-term mortality, 7 indicating the strong association of CTO lesion with cardiac mortality in these patients. 8 However, whether revascularization of CTO lesion in non-IRA will lead to improved clinical outcomes is still controversial. Observational studies [9][10][11][12] and meta-analysis 13 15 Therefore, this study aimed to evaluate the long-term impact of CTO revascularization in AMI patients after IRA PCI in the real world.

| Patient population
All consecutive patients diagnosed as AMI (including STEMI and non-ST segment elevation myocardial infarction [NSTEMI]) and who underwent coronary artery angiography (CAG) in Zhongshan Hospital, Shanghai, China, between July 2011 and July 2019 were retrospectively included in this study. Patients with prior coronary artery bypass graft (CABG) were excluded. We then identified patients treated by successful PCI in IRA and had at least one coexisting non-IRA CTO in the major epicardial coronary arteries. Patients who died during hospital stay after IRA PCI and patients treated by CABG after PCI were excluded from the study. The study was approved by the institutional review board of Zhongshan Hospital, Fudan University and all patients signed a general informed consent form.

| Study definitions and endpoints
AMI was diagnosed according to characteristic clinical symptoms, ECGs changes, cardiac enzyme elevations (Fourth Universal Definition 16 ), and was also confirmed by CAG. Periprocedural MIs were not included in the study. IRA was defined as a major coronary artery perfusing an area compatible with the distribution of ST-segment elevation or depression in the 12-lead ECG and the typical angiographic image. CTO was defined as thrombolysis in myocardial infarction (TIMI) Grade 0 flow and duration of coronary occlusion ≥3 months. In addition, the typical appearance of a CTO includes angiographically visible mature collaterals and the absence of thrombus or staining at the proximal cap. 17  Any revascularization was defined as a repeat PCI or CABG excluding the planned staged PCIs of any segment of the coronary artery.

| Procedures
All patients were treated with 300 mg aspirin and a loading dose of 300 mg of clopidogrel or 180 mg of ticagrelor before the procedure.
During the procedure, unfractionated heparin was administered intravenously to achieve a target activated clotting time of 250-350 s.
GPIIb/IIIa inhibitors were administered at the operator's discretion.
IRA stenting was performed using a drug-eluting stent (DES). Successful IRA PCI was defined as residual stenosis of the culprit lesion <30% and a TIMI flow grade ≥ 2. When to perform PCI in non-IRA vessels (CTO or non-CTO lesion) was left to the operator's discretion, usually within 1 year after IRA PCI. For CTO PCI, the choices of antegrade or retrograde approach and devices used were up to the discretion of the operator. DES was used in successfully recanalized CTO vessels.

| Data collection
Demographic, angiographic, procedural, and outcome data were obtained from a review of the catheterization laboratory database and QIN ET AL. | 137 medical chart. Clinical follow-up data were collected through outpatient visits, telephone interviews, and medical chart reviews.

| Statistical analysis
The primary analysis is based on the intention-to-treat (ITT) population. All continuous variables were presented as mean ± standard deviation or the median with interquartile range and were compared by Student's t test or the Mann-Whitney U test, respectively. Categorical variables were presented as counts and percentages and were compared by χ 2 test (or Fisher's exact test when appropriate). To adjust for any potential confounders, propensity score matching (PSM) analysis was performed using the logistic regression model. Variables that could be of potential relevance to the endpoints, including age, male, hypertension, diabetes, dyslipidemia, current smoking, previous MI, previous PCI, IRA, location of CTO, and left ventricular ejection fraction (LVEF), were used. Matching was performed via a 1:1 matching protocol using the nearest neighbor matching algorithm, with a caliper width equal to 0.05 of the standard deviation of the propensity score. The covariate balance of the matched cohort was assessed using the standardized mean difference (SMD). 18 The C-statistics for PSM was 0.757 in the ITT population. Survival curves were plotted using the Kaplan-Meier method, and comparisons between groups were done using the log-rank test. The Cox proportional hazards model was used to identify the independent predictors of cardiac death. The candidate variables for the model were selected based on significant univariate analysis. Prespecified subgroup analyses were performed for the primary endpoint according to the following variables: age, gender, diagnosis, diabetes, LVEF, IRA, and CTO location. All analyses were performed using SPSS, Version 20.0 (IBM Corporation), and a p < .05 was considered statistically significant.

| Baseline characteristics in ITT population
Among 4372 patients who were diagnosed with AMI and treated by IRA PCI during the study period, we identified 362 eligible patients who had non-IRA CTOs. Of these patients, eight were excluded as the CTO lesions were treated by CABG, 15 patients were excluded because they died during hospital stay after IRA PCI, and nine patients were excluded because the CTO lesions were not located in major epicardial coronary arteries. Finally, 330 patients who were treated by either PCI (n = 198) or MT (n = 132) for non-IRA CTOs were included in the study ( Figure 1).
The baseline, angiographic, and procedural characteristics of patients during IRA PCI are listed in Table 1. The patients in the MT group were older, more likely to be diagnosed as STEMI, had lower estimated glomerular filtration rate (eGFR), higher peak troponin T, and creatinine kinase (CK)-MB level during AMI compared with patients in the PCI group. Furthermore, in MT group, the involvement of LAD coronary artery as IRA (53.8% vs. 38.4%, p = .006) and LCX as CTO vessel (45.5% vs. 29.8%, p = .004) was more frequent than in PCI group, and thus the involvement of LAD as CTO vessel was less frequent (14.4% vs. 39.4%, p < .001). The baseline, angiographic, and procedural characteristics of the two groups were balanced after PSM (108 pairs).

| Long-term clinical outcomes in ITT population
Clinical outcomes in the entire cohort and PSM groups are presented in Table 2 and Figure 2A F I G U R E 1 Flow chart of the study. AMI, acute myocardial infarction; CABG, coronary artery bypass graft; CTO, chronic total occlusion; non-IRA, non-infarct related artery; PCI, percutaneous coronary intervention T A B L E 1 Baseline and procedural characteristics during AMI PCI in the intention-to-treat population

| Subgroup analysis in ITT population
In subgroup analysis, PCI was associated with less cardiac death in patients over 65 years old, without diabetes, with LVEF < 50%, LAD IRA, and non-LAD CTO lesion compared with MT ( Figure S1). In patients with LVEF < 50%, the long-term cardiac death-free survival is higher in the PCI group compared with the MT group ( Figure 2C).
However, there was no significant difference in cardiac death after PSM ( Figure 2D). In patients with LVEF ≥ 50%, no difference in cardiac death was observed between two groups (p = .990).

| Per-protocol analysis (successful CTO PCI vs. MT/failed PCI)
As shown in Figure 1, technical success was achieved in 166 patients, who were classified into the successful PCI (s-PCI) group (n = 166).
Thirty-two patients failed in the PCI procedure and 132 patients receiving MT constituted the occluded CTO (o-CTO) group (n = 164).
The baseline characteristics in per-protocol analysis before and after PSM are shown in Table S1. The results of per-protocol analysis showed a significantly higher incidence of cardiac death-free survival in the s-PCI group both before (96.6% vs. 84.8%, p = .017) and after PSM (97.3% vs. 86.4%, p = .040) when compared with those in the o-CTO group (Table S2 and Figure 3A,B).
In subgroup analysis in the per-protocol population, the longterm cardiac death-free survival is higher in the s-PCI group compared with the o-CTO group in patients with LVEF < 50% before and after PSM ( Figure 3C,D). Similarly, no significant difference was noted between the two groups in the occurrence of cardiac death in patients with LVEF ≥ 50% (p = .70).

| DISCUSSION
The main findings of the current study were as follows. (1)  with optimal medical therapy, which makes the treatment strategy for CTO still controversial. [21][22][23][24] In patients with AMI, the presence of concurrent CTO is associated with increased mortality. 3,6 The pathological mechanisms involved include aggravated ischemia caused by occlusion of IRA and subsequent interruption of collateral supply to non-IRA CTO, microvascular ischemia, reperfusion injury, and electrical instability, which may lead to poorer outcomes in these patients compared with patients who suffered from CTO and stable CAD. 25 Theoretically, revascularization of non-IRA CTO after IRA PCI might yield more clinical benefits, as the recovery of blood supply in both the CTO territory and overlapping border of the infarct zone may reduce left ventricular remodeling and improve contractile function.
Based on this theory, several observational studies have been performed in patients with AMI (including STEMI and NSTEMI) and concurrent non-IRA CTO. Park et al., 10  reported in a retrospective study that the 5-year cumulative incidence of the composite of total death or myocardial infarction was significantly lower in patients who underwent non-LAD CTO PCI than patients receiving MT. To our knowledge, myocardial viability is associated with long-term outcomes after CTO PCI, 35 and the prognosis of CTO-PCI may differ due to the amount of myocardium at risk, which is supplied by the CTO vessel. As myocardial viability data was not available in retrospective studies, the benefit of CTO revascularization in a single vessel would be difficult to determine.

| CONCLUSION
In conclusion, patients undergoing successful revascularization of non-IRA CTO after AMI might have a better long-term prognosis compared with patients with o-CTO. LVEF < 50% is an independent predictor of cardiac death and patients with LVEF < 50% may benefit from successful CTO-PCI after AMI.

| LIMITATIONS
There are several limitations to this study. First, this is a retrospective observational study in a single center. Therefore, a limited number of patients were included. Second, the selection of CTO-PCI or not and the timing of CTO-PCI after IRA PCI is left to the preferences of patients and doctors. Therefore, the potential for selection bias cannot be excluded. Third, the myocardial viability test was not routinely performed, and potential imbalances of the amount of viable myocardium may influence the clinical outcomes.