What's new?

This study provides new insights into the prognosis of patients with ST‑segment elevation myocardial infarction and multivessel coronary artery disease. We found that the extent of atherosclerosis, defined as the number of coronary arteries with significant lesions, did not seem to affect overall survival, the incidence of major adverse cardiovascular events, or unplanned cardiac and all‑cause readmissions during 2‑year follow‑up. We also identified the use of more potent antiplatelets instead of clopidogrel along with revascularization of noninfarct‑related arteries during initial hospitalization as important targets that could help minimize readmission rates in this group of patients.

Introduction

Contemporary management in the acute phase of ST‑segment elevation myocardial infarction (STEMI) is based on early primary percutaneous coronary intervention (pPCI) in the infarct‑related artery (IRA) combined with dual antiplatelet therapy (DAPT).¹ Park et al² demonstrated that approximately 30% to 50% of patients enrolled in the Korea Acute Myocardial Infarction Registry had significant atherosclerotic lesions in more than 1 coronary artery. In our recent report, multivessel coronary artery disease (MVD) was present in 32.9% of STEMI patients.³ These individuals have an adverse prognosis, as compared with those with single‑vessel disease (SVD).⁴ According to previous studies, complete revascularization (CR) reduces the risk of major adverse cardiac events (MACEs), such as cardiovascular mortality, new MI, and ischemia‑driven revascularization.^5,6 Therefore, the latest guidelines of the European Society of Cardiology¹ upgraded the recommendation for non‑IRA revascularization during the index procedure or within 45 days after discharge in hemodynamically stable patients. Following our preliminary report,³ our objective was to evaluate the long‑term prognosis after pPCI in STEMI patients with MVD included in the Polish Registry of Acute Coronary Syndromes (PL‑ACS).

Patients and methods

The principles of the PL‑ACS have been presented elsewhere.^3,7 Briefly, this is a national, multicenter, prospective observational registry that includes data on patients with ACS. By the end of 2022, we were able to collect data on 674 461 patients. Follow‑up data were extracted from the Silesian Cardiovascular Database—a digital tool designed to report data on ACS to the National Health Fund. Data on STEMI patients treated with pPCI in 18 centers between 2017 and 2020 were analyzed. During that period, information regarding lesion significance was available for 1513 individuals. After excluding 68 patients (4.5%) who died during hospitalization and 1006 individuals (69.6%) with MI and SVD, the final study group comprised 439 patients (30.4%) with MVD. MVD was defined as a presence of at least 2 significantly stenosed coronary arteries (≥50% diameter stenosis on visual angiographic assessment), including the culprit lesion and at least 1 significant lesion in a non‑IRA territory. The patients were classified into the 2‑vessel disease (2‑VD) and 3‑vessel disease (3‑VD) groups according to the number of coronary arteries with atherosclerotic plaques (Figure 1). Four revascularization strategies were considered: culprit vessel only during the index pPCI, immediate or staged CR during the index hospitalization, and elective CR after discharge following the index hospitalization. All patients provided written informed consent to participate in the study. All procedures were performed according to the current guidelines.^8,9 The study was approved by the Ethics Committee at the Jan Kochanowski University in Kielce (6/2017).

Demographic, clinical, and angiographic characteristics of the patients, together with data on pharmacotherapy and vascular access site were collected. Coronary anatomy was identified as previously described.³ The patients were followed for up to 24 months to assess the risk of MACEs, defined as all‑cause mortality, MI, stroke, coronary revascularization, as well as unplanned cardiac or all‑cause rehospitalization. Bleeding during follow‑up was defined as gastrointestinal bleeding or hemorrhagic stroke requiring hospitalization.

Results

Clinical outcomes

All participants completed the 2‑year follow‑up. In‑hospital complication rates were low and did not differ between the groups. The patients with 3‑VD had a markedly higher rate of repeat PCI than those with 2‑VD, but it included cases of scheduled non‑IRA PCI. Differences in all‑cause mortality, MI, stroke, bleeding, as well as unplanned cardiac and all‑cause readmission rates were insignificant (Table 3, Figure 2). With respect to readmissions, 62.5% of them were due to cardiac causes (STEMI [25.6%] and chronic coronary syndromes [24.5%] as the leading causes) in both groups (Table 3).

Table 3. In‑hospital and 2‑year outcomes of patients with ST‑segment elevation myocardial infarction and 2- or 3‑vessel coronary artery disease

Outcome	In‑hospital			2‑year follow‑up
	3‑VD group (n = 110)	2‑VD group (n = 329)	P value	3‑VD group (n = 110)	2‑VD group (n = 329)	P value
Data are presented as numbers (percentages). Abbreviations: N/A, not applicable; TIA, transient ischemic attack; others, see Table 1
Death	N/A	N/A	N/A	10 (9.1)	39 (11.9)	0.43
MI	N/A	N/A	N/A	7 (6.4)	24 (7.3)	0.74
PCI	N/A	N/A	N/A	59 (53.6)	109 (33.1)	<⁠0.001
Stroke/TIA	1 (0.9)	0	0.08	3 (2.7)	6 (1.8)	0.56
Bleeding	2 (1.8)	4 (1.2)	0.64	4 (3.6)	9 (2.7)	0.63
All‑cause readmission	N/A	N/A	N/A	42 (38.2)	134 (40.7)	0.64
Unplanned cardiac readmission	N/A	N/A	N/A	32 (29.1)	78 (23.7)	0.26

In the univariable Cox regression, the risk of all‑cause readmission was higher in the patients with obesity, heart failure (HF), chronic kidney disease (CKD), and those receiving clopidogrel, whereas unplanned cardiac readmissions were more likely in the patients with HF, CKD, and previous MI. Administration of ticagrelor and CR during the index hospitalization were associated with a lower risk of all‑cause readmission and unplanned cardiac rehospitalization. In the multivariable Cox regression, the strongest predictor of unplanned readmission was clopidogrel administration. Previous MI and the left main coronary artery as the IRA increased the risk of unplanned cardiac rehospitalization. The only multivariable predictor associated with a lower rehospitalization rate during 2‑year follow‑up was staged non‑IRA PCI during the index hospitalization (Table 4).

Table 4. Uni- and multivariable predictors of all‑cause and cardiac readmissions in patients with ST‑segment elevation myocardial infarction and multivessel coronary artery disease

Variable	Univariable HR (95% CI)	P value	Multivariable HR (95% CI)	P value
Abbreviations: CR, complete revascularization; HF, heart failure; HR, hazard ratio; others, see Tables 1 and 2
Predictors of all‑cause readmissions
Obesity	1.54 (1.07–2.23)	0.02	–	–
HF	2.47 (1.37–4.45)	0.002	–	–
CKD	2.42 (1.42–4.11)	0.001	–	–
Clopidogrel	1.43 (1.06–1.93)	0.02	1.45 (1.07–1.95)	0.01
Ticagrelor	0.61 (0.44–0.83)	0.002	–	–
GP IIb/IIIa inhibitor	0.69 (0.5–0.97)	0.03	–	–
Non‑IRA in‑hospital staged PCI	0.62 (0.46–0.83)	0.002	0.61 (0.46–0.82)	0.001
In‑hospital CR	0.69 (0.51–0.94)	0.02	–	–
Predictors of unplanned cardiac readmissions
HF	4.21 (2.31–7.69)	<⁠0.001	–	–
CKD	2.67 (1.43–4.98)	0.002	–	–
Previous MI	1.59 (0.99–2.57)	0.053	1.63 (1.01–2.63)	0.04
Ticagrelor	0.6 (0.4–0.9)	0.01	–	–
GP IIb/IIIa inhibitor	0.64 (0.42–0.99)	0.04	–	–
IRA LM	2.33 (0.74–7.35)	0.15	3.22 (1–10.38)	0.049
Non‑IRA in‑hospital staged PCI	0.63 (0.43–0.92)	0.02	0.61 (0.42–0.9)	0.01
In‑hospital CR	0.66 (0.44–0.97)	0.04	–	–

Discussion

The main finding of our study is that STEMI patients with MVD benefited from non‑IRA PCI during the index hospitalization in terms of a lower rate of rehospitalization due to both cardiac and noncardiac causes.

Presence of a significant atherosclerotic lesion in the non‑IRA territory is associated with unfavorable long‑term and in‑hospital prognosis in MVD, as compared with SVD.^4,10,11 Several studies on the impact of MVD on clinical outcomes in Polish patients with STEMI were published.^12,13 Our analysis showed a lower incidence of MVD in STEMI patients than reported by Park et al² and Toma et al.¹⁴ In the general population included in the PL‑ACS between 2017 and 2020, the prevalence of MVD (28.4%) and the mean age of the patients (65.2 years) were similar to those reported in our study¹⁵; however, we observed a higher proportion of women than other researchers.^16,17 The most common IRAs were the right (44.2%) and left anterior descending arteries (37.8%), with a 10.9% contribution of the circumflex branch, which is less common than previously reported.^5,6,17

The clinical profile of the patients with 2‑VD and 3‑VD was similar. We provide novel data on comorbidities (ie, obesity, HF, atrial fibrillation, and chronic obstructive pulmonary disease) that impact cardiovascular prognosis and were not previously reported. We identified 24.4% of patients with Killip class greater than I on admission, with no significant differences between the 2‑VD and 3‑VD groups; however, this proportion was greater than previously reported.^5,6,17,18

Several studies provided clinical evidence for the superiority of CR over culprit‑only revascularization among STEMI patients with MVD.^5,6,19,20 However, they used different methods to detect the significance of the non‑IRA lesion, that is, angiographic visual assessment^5,20,21 or FFR evaluation.^17,22 The optimal methods for detecting lesion significance in the non‑IRA territory and their impact on the long‑term prognosis remain unclear. Several reports demonstrated a significant benefit of an angiography‑guided approach over an FFR‑guided strategy,^18,23 while Lee et al²⁴ confirmed the superiority of functional lesion assessment prior to PCI. Furthermore, in the COMPLETE trial (Complete vs Culprit‑only Revascularization to Treat Multi‑vessel Disease After Early PCI for STEMI),⁶ FFR measurement was recommended for lesions with 50%–70% diameter stenosis, while stenosis greater than 70% did not require FFR assessment prior to revascularization. Due to the heterogeneity in the definition of lesion significance and the inclusion criteria across studies, extrapolation of the results remains challenging.

Angiographic characteristics of the lesions on diagnostic coronary angiography differed between the analyzed patients groups. In‑stent restenosis was more prevalent in the 2‑VD group, whereas CTO lesions were more frequently observed in the 3‑VD group. The CR rate was similar in both groups. Bangalore et al²⁵ identified the number of diseased vessels and presence of CTO as the strongest predictors of incomplete revascularization (ICR) in patients with MVD.

ICR was associated with a significantly higher risk of MACEs, as compared with CR. However, univariable Cox regression showed that CTO lesions were not associated with an increased risk of readmissions during 2‑year follow‑up.

We observed lower DAPT rates, as compared with the available reports.^5,6,17,21,22 In our study, clopidogrel was the most widely used platelet receptor P2Y₁₂ inhibitor, but the rate of ticagrelor administration was also high. However, Polish patients received ticagrelor or prasugrel less frequently than individuals from other countries. The reason for this may be the lack of reimbursement for the guideline‑preferred platelet receptor P2Y₁₂ inhibitors. Most of our data predate the publication of the ISAR‑REACT 5 (Prospective, Randomized Trial of Ticagrelor Versus Prasugrel in Patients With Acute Coronary Syndrome) results,²⁶ which may explain the very low use of prasugrel. Our findings support the need for a broad implementation of novel antiplatelet agents, as clopidogrel remained a significant predictor of readmission in the multivariable analysis.