Complete versus culprit-only revascularization in patients with ST-segment elevation myocardial infarction and multivessel disease: a meta-analysis of randomized trials

Background The best strategy for the treatment of the non-infarct artery in patients with ST-elevation myocardial infarction (STEMI) and multivessel disease (MVD) undergoing primary percutaneous coronary intervention (PCI) is not yet defined. Methods We searched the literature for randomized controlled trials (RCTs) that compared complete revascularization (CR) with infarct-related coronary artery (IRA) only revascularization in hemodynamically stable patients with STEMI. Random effect risk ratios (RRs) were calculated for clinical outcomes. Results Nine RCTs with 2989 patients were included. No significant difference in all-cause mortality emerged between CR and IRA-only groups (relative risk [RR] = 0.74; 95% confidence interval [CI]: 0.52 to 1.04; p = 0.08). Compared with IRA-only, CR was associated with significantly lower rates of major adverse cardiac events (MACE) (RR = 0.53; 95% CI: 0.41 to 0.68; p < 0.001), cardiac death (RR = 0.48; 95% CI: 0.29 to 0.79; p = 0.004) and repeat revascularization (RR = 0.38; 95% CI: 0.30 to 0.47; p < 0.001). In subgroups analysis, immediate complete revascularization (ICR) reduced the risk of all-cause mortality (RR = 0.62; 95% CI: 0.39 to 0.97; p = 0.04), whereas staged complete revascularization (SCR) did not show any significant benefit in all-cause mortality (RR = 0.92; 95% CI: 0.46 to 1.86; p = 0.82). Stroke, contrast-induced nephropathy and major bleeding were not different between CR and IRA-only. Conclusions For patients with STEMI and multivessel disease undergoing primary PCI, complete revascularization did not decrease the risk of all-cause mortality in current evidence from randomized trials. When feasible, immediate complete revascularization might be considered in patients with STEMI and multivessel disease. Electronic supplementary material The online version of this article (10.1186/s12872-019-1073-8) contains supplementary material, which is available to authorized users.


Background
Patients with acute ST-segment elevation myocardial infarction (STEMI) are effectively treated with primary percutaneous coronary intervention (PCI) of the infarctrelated coronary artery. Approximately, 40-60% of these patients have multivessel disease (MVD) and are associated with worse clinical outcomes compared with those have single-vessel disease [1][2][3]. The American College of Cardiology/American Heart Association have updated their guidelines recommendation from III to IIb for primary PCI of the non-infarct-related coronary artery in hemodynamically stable patients with STEMI and MVD [4]. Current European Society of Cardiology guidelines support intervention of the non-IRA at the time of primary PCI (Class IIa indication) [5].
Most recent randomized controlled trials (RCTs) reported that complete revascularization (CR) for hemodynamically stable patients with STEMI and MVD at the time of primary PCI might have beneficial effects [6,7]. However, these trials are limited by sample sizes and not powered to detect differences in all-cause mortality or myocardial infarction (MI). Moreover, the optimal strategy of complete revascularization, either immediate complete revascularization (ICR) during primary PCI or staged complete revascularization (SCR), and its impact on mortality is still unclear. Therefore, we conducted this meta-analysis of RCTs to assess whether complete revascularization can reduce all-cause mortality in patients with STEMI and MVD and to determine the possible strategy of complete revascularization.

Data sources
We searched PubMed, MEDLINE, EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL), and Clinical Trials.gov for RCTs using the terms "myocardial infarction", "percutaneous coronary intervention", "coronary angioplasty" and "multivessel". All RCTs published until 1 August 2018 were identified. The reference lists of the retrieved articles were manually searched for additional potential articles. The search algorithm is shown in Fig. 1. The meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [8].

Selection criteria
The inclusion criteria were:1) studies of hemodynamically stable patients with STEMI and MVD at the time of primary PCI; 2) randomized clinical trials comparing complete versus IRA-only revascularization; 3) main outcomes of interest including (mortality, MI, and revascularization) reported. Trials including chronic total occlusion were excluded.

Data extraction and quality assessment
Two authors (JJ.L and HL.L) independently assessed trial eligibility and extracted data. Any disagreements were resolved by consensus of the authors. We performed objective assessment of the trials using a standardized data collection form. The data extracted from the trials included the year of publication, sample size, duration of follow-up, definitions of endpoints used in each study, clinical outcomes (including major adverse cardiac events [MACE], all-cause mortality, MI, repeat revascularization, stoke, contrast-induced nephropathy [CIN], and major bleeding). The trial bias risk was assessed by Cochrane Collaboration guidelines [9].

Outcomes measures
The primary outcome was all-cause mortality. Secondary outcomes included composite MACE, cardiac death, nonfatal MI, repeat revascularization, stroke, CIN and major bleeding.

Statistical analysis
The meta-analysis was performed using STATA software version 12 (STATA Corporation; College Station, Texas). We analyzed outcomes in an intention-to-treat analysis. Risk ratios (RRs) and 95% confidence intervals (CIs) were used as summary statistics. Fixed-effect model of Mantel-Haenszel was used to assess the overall estimate; however, a random-effect model was chosen to calculate pooled RRs when heterogeneity existed. Heterogeneity was assessed by I 2 test and its P value. I 2 < 25% was defined as low heterogeneity and > 75% was defined as significant heterogeneity. Pre-specified subgroup analysis was performed according to strategy of complete revascularization (predominantly immediate revascularization or staged revascularization). Serially leave-one-out analysis was used to eliminate sources of heterogeneity in sensitivity analysis. Publication bias was visually assessed using funnel plots. A 2-tailed P < 0.05 was considered statistically significant.

Characteristics of included studies
Our search identified 3126 articles, among which 343 abstracts were screened. Nine randomized studies [6,7,[10][11][12][13][14][15][16] enrolling 2989 patients met the inclusion criteria and were included in this meta-analysis. The characteristics of the included studies are presented in Table 1. A total of 1582 patients underwent IRA-only revascularization. Among the 1407 patients who underwent complete revascularization, 843 patients were assigned to the ICR group and 564 patients were assigned to the SCR group. In the study by Politi et al. [15], 65 patients who underwent immediate complete revascularization during primary PCI were included in the ICR group, whereas the 65 patients who underwent staged complete revascularization were included in SCR group. The timing of staged intervention was from 2 days up to 57 days in the SCR group. In Politi's study [15], the time of patients who underwent staged complete revascularization of the non-infarct-related artery was 56.8 ± 12.9 days after primary PCI. All studies, except for Ghani [14], excluded patients with cardiogenic shock. Most of the studies employed angiography to estimate stenosis in the non-infarct artery. However, fractional flow reserve (FFR) was used to guide PCI for non-infarct-related coronary artery lesions in 3 studies [6,7,14]. The duration of the follow-up ranged from 6 to 38 months (23.0 months). Table 2 summarizes baseline characteristics of the patients.

Cardiac death
Cardiac death was reported in six studies. The incidence of cardiac death showed a significant difference between CR and IRA-only groups (RR = 0.48; 95% CI: 0.29 to 0.79; p = 0.004, Fig. 4 Fig. 4) and no evidence of heterogeneity among studies (I 2 = 0%, p = 0.42).

Recurrent myocardial infarction
Recurrent myocardial infarction was reported in all studies. The incidence of recurrent MI did not show a significant difference between CR and IRA-only groups (RR = 0.68; 95% CI: 0.44 to 1.07; p = 0.09, Fig. 5

Major bleeding
Major bleeding was reported in four studies. The overall incidence of major bleeding was very low (1.4% vs. 1.9%), with no difference between CR and IRA-only groups (RR = 0.56; 95% CI: 0.24 to 1.27; p = 0.17, Additional file 3: Figure S3).

Subgroup analysis
Trials using angiography had a significant risk reduction in risk for all-cause mortality (RR = 0.59; 95% CI: 0.39 to 0.89; p = 0.01), whereas trials using FFR did not report any beneficial effect (RR = 1.23; 95% CI: 0.66 to 2.29; p = 0.51, Additional file 4: Figure S4). The follow-up time of 5 trails [6,11,[13][14][15] was more than 12 months, whereas other 4 trails [7,10,12,16] was not. There was a trend toward a reduction in all-cause mortality with complete revascularization in trails that the follow-up time was not more than 12 months (RR = 0.52; 95% CI: 0.25 to 1.09; p = 0.09, Additional file 5: Figure S5). The incidence of all-cause mortality did not show a significant difference between CR and IRA-only groups in trails that the follow-up time was more than 12 months (RR = 0.81; 95% CI: 0.54 to 1.21; p = 0.30, Additional file 5: Figure S5).

Publication bias
Funnel plots did not suggest publication bias for any of the clinical outcomes (p = 1.00 for all-cause mortality, Additional file 6: Figure S6).

Discussion
In this meta-analysis of nine randomized trials including 2989 patients with STEMI and MVD, we demonstrated that the risk of all-cause mortality was not different between complete revascularization and infarct-related coronary artery only revascularization. The present analysis found that a trend towards decrease in all-cause mortality in CR when compared with IRA-only, however it did not reach statistical significance. Immediate complete revascularization significantly reduced the risk for all-cause mortality compared with IRA-only; whereas staged revascularization did not show any benefit on the outcome.
Most data of observational studies had suggested that complete revascularization for multivessel disease at the time of primary PCI might be harmful [17][18][19]. Recent RCTs and meta-analyses showed that CR reduced the risk of a composite cardiovascular outcome as compared with IRA-only [6,7,20]. However, none of the RCTs demonstrated a significant difference between complete revascularization and IRA-only revascularization in death or MI, but only in MACE. In our analysis, there was no significant difference in all-cause mortality between CR and IRA-only among patients with STEMI and MVD undergoing primary PCI.
Subgroup analyses demonstrated a significant benefit of ICR compared with IRA-only. The risk of MACE, cardiac death, recurrent MI, and repeat revascularization were also significantly lower among those underwent ICR. The findings were consistent with a more recent network meta-analysis [21]. The timing of staged intervention was heterogeneous across the included trials. In this study by Politi and colleagues [15], the time of staged complete revascularization PCI of the non-IRA after primary PCI was 56.8 ± 12.9 days, which was longer than other trails. After exclusion of the study by Politi et al. [15], the results for all-cause mortality did not change. However, staged revascularization did not seem to be associated with any beneficial effect. Our results were consistent with other meta-analyses that suggested that immediate complete revascularization was associated with significant reduction in all-cause mortality [20][21][22]. Only one small trial [15] that compared immediate versus staged revascularization in STEMI patients with MVD, so it is not possible yet to directly compare these two strategies in this meta-analysis. An earlier network meta-analysis [23] reported the risk of all-cause mortality is not different between the two revascularization strategies. However, two more recent network meta-analyses [21,22] that demonstrated the benefit of ICR in reducing all-cause mortality compared to SCR. The optimal revascularization strategy (CR versus IRA-only) in hemodynamically stable patients with STEMI and MVD continues to be debated. Therefore, large randomized trials might be required to clarify the benefit of immediate complete revascularization in STEMI patients with MVD.
The mechanisms by which immediate complete revascularization may improve prognosis in STEMI patients with MVD are probably related to pathophysiology of MI. First, ICR may theoretically decrease infarct size [24] and preserve left ventricular function [25]. Achieving complete revascularization as soon as possible may help reduce the risk for death and MI. Second, unstable plaques in the non-culprit lesion are associated with increased risk for acute coronary events. Hong et al. [26] reported that noninfarct-related artery plaque ruptures occurred in 17% of patients using intravascular ultrasound (IVUS) examination. Besides, there are several advantages of The meta results showed that the risk of MACE was significantly lower among those in the CR group, which was consistent with the prior meta-analyses [22,23,27]. This benefit was derived from the primarily reduced rate of repeat revascularization. Additionally, the incidence of cardiac death was lower with ICR than with IRA-only.
It has been reported that adding FFR measurements to assess ischemia during coronary angiography might reduce cardiac events in patients with stable coronary artery disease [28,29]. However, whether the FFR measurements benefit in patients with STEMI is not clear. In this subgroup analysis of using FFR measurements, FFR-guided CR did not report any beneficial effect (RR = 1.23; 95% CI: 0.66 to 2.29; p = 0.51) compared to IRA-only. Both the DANAMI 3-PRIMULTI trial and COMPARE-ACUTE trail found that no differences in death between FFR-guided staged PCI and IRA-only. It may be that the disturbed microvascular function in the non-IRA territory in the early stage of STEMI affects the reliability of the technique.
Our results indicated that all-cause mortality had a trend toward a reduction in trails performing complete revascularization which the follow-up time were not more than 12 months. The exact mechanisms linking complete revascularization with better mortality maybe elucidated by less rate of recurrent MI arising from non-culprit lesions and more complete recovery of left ventricular function with less hemodynamic instability and fewer arrhythmias in the early stage after myocardial infarction.
The CR might be safe in hemodynamically stable patients with STEMI and MVD undergoing primary PCI. This meta -analysis showed that CR was not associated with an increased risk of stroke, major bleeding, and contrast-induced nephropathy.
The present meta-analysis has several limitations. The included studies with variability in inclusion/exclusion criteria, endpoint definitions, timing of staged intervention and follow-up time. In addition, publication bias is an inherent limitation of meta-analyses. Finally, few endpoints occurred and it is likely that the analyses are underpowered for individual outcomes. The ongoing COMPLETE (Complete vs Culprit-Only Revascularization to Treat Multi-vessel Disease After Primary PCI for STEMI) and FULL REVASC (FFR-Guidance for Complete Non-Culprit Revascularization) trials will provide important data on the optimal strategy for patients with STEMI and MVD.

Conclusions
For patients with STEMI and multivessel disease undergoing primary PCI, complete revascularization did not decrease the risk of all-cause mortality in current evidence from randomized trials. However, it is associated with significant reductions in MACE and cardiac death along with a reduced need for repeat revascularization. Immediate complete revascularization might be feasible in STEMI and multivessel disease patients undergoing primary PCI. More studies are needed to confirm the indications for and timing of non-infarct artery PCI. PCI: Percutaneous coronary intervention; RCTs: Randomized controlled trials; RR: Risk ratios; SCR: Staged complete revascularization; STEMI: ST-elevation myocardial infarction