Recent trials and meta-analysis have indicated that complete revascularization (CR) of multivessel coronary disease is beneficial in patients with ST-segment elevation myocardial infarction (STEMI) compared to culprit-only intervention. However, the optimal timing of CR remains unclear. We aimed to analyze the optimal timing of CR in patients with STEMI and multivessel disease by performing an updated network meta-analysis using the recent largest randomized controlled trial. PUBMED and EMBASE were searched through October 2020 to identify randomized controlled trials comparing CR and culprit-only revascularization. A random-effect network meta-analysis comparing three arms (same-sitting [during the index procedure] CR versus staged CR versus culprit-only) and 4 arms (same-sitting CR versus staged CR [in-hospital] versus staged CR [out-hospital] versus culprit-only) were performed. Eleven studies with a total of 7,015 patients were included in our analysis. There was no significant difference in major adverse cardiovascular event (MACE) (HR 0.82, 95% CI 0.64-1.05), cardiovascular death (HR 0.69, 95%CI 0.35-1.33), myocardial infarction (HR 0.66, 95%CI 0.37-1.16), and revascularization (HR 1.05, 95%CI 0.70-1.58) between same-sitting CR and staged CR. When staged CR was further divided into staged CR during the hospitalization and after discharge, there was no significant difference in these outcomes between staged CR (in-hospital) and staged CR (out-hospital). In conclusion, in patients with multivessel disease presenting with STEMI, complete revascularization at any timing, including same-sitting, staged in-hospital, and staged out-hospital, may have similar benefits.
Primary revascularization of the infarct-related artery in patients with ST-segment elevation myocardial infarction (STEMI) has been shown to reduce future adverse cardiovascular outcomes, including mortality, non-fatal infarction, and stroke. Recently, in patients with multivessel disease, the Complete versus Culprit-Only Revascularization Strategies to Treat Multivessel Disease after Early PCI for STEMI (COMPLETE) trial and meta-analyses have reported complete revascularization of the non-culprit lesion to be associated with a reduction in cardiovascular death, myocardial infarction, and revascularization. Despite the accumulating evidence of promising outcomes in complete revascularization, the optimal timing of non-culprit lesion revascularization (complete revascularization in the same-sitting procedure, staged revascularization during the admission, or staged revascularization after discharge) remains unclear. We performed a network meta-analysis to determine whether a particular timing strategy was inferior in patients with STEMI and multivessel disease.
Methods
All RCTs comparing complete revascularization to culprit-only revascularization or same-sitting complete to staged complete revascularization in patients with STEMI and multivessel disease with at least 6 months or longer follow-up time were considered eligible to be included in our analysis. Trials published in peer-reviewed journals or presented in an academic meeting were included. No restriction on publication language or sample size was applied.
A comprehensive literature search of PUBMED and EMBASE was conducted from inception through October 8, 2020. The search terms included “ST elevation myocardial infarction”, “multivessel disease”, “complete”, “non-culprit”, “revascularization”, “intervention”, “percutaneous coronary intervention”, “angioplasty”, and “randomized” . A complete search algorithm for each search engine are provided in Supplementary Table S1. In addition, a manual search of secondary sources, including references of initially identified articles, reviews, and commentaries, was conducted to identify relevant studies.
The search results were screened for eligibility by two investigators (H.U. and T.K.) independently through title and abstracts. Full texts of the article were retrieved when deemed potentially relevant, and further eligibility was assessed using predefined eligibility criteria. The data extraction was performed individually into a standardized electronic spreadsheet. The third author (H.T.) verified the study selection and data extraction process. Any disagreements among the investigators within the process were resolved by consensus.
The primary efficacy endpoint was study-defined major adverse cardiovascular event (MACE). Other efficacy endpoints included all-cause death, cardiovascular death, myocardial infarction (MI), and revascularization. The safety endpoint included stroke, major bleeding, and contrast-induced acute kidney injury. In studies including both same-sitting and staged multivessel PCI in complete revascularization arm, the corresponding author of the original trial was contacted in an attempt to retrieve detailed data differentiating the two arms.
We used the Cochrane Collaboration’s tool to assess the source of bias of each individual trial. Publication bias was assessed using the funnel plot. The asymmetry was mathematically assessed using the Egger’s linear regression test.
Hazard ratio (HR) and associated 95% confidence interval (CI) or risk ratio (RR) calculated from numbers of events for each pre-defined clinical outcome were extracted from the included studies. HR was used for synthesis when available, and RR was substituted if not. The primary comparison of the analysis was same-sitting complete revascularization of the non-culprit artery during the index procedure (same-sitting complete) versus staged complete revascularization of the non-culprit artery (staged complete) versus culprit-only revascularization. In addition, we performed a sub-analysis further dividing the staged revascularization into revascularization performed during the same admission “in-hospital staged” and after discharge “out-hospital staged”. Studies with a mean time to staged revascularization performed ≥ 21days after the index procedure was considered as staged (after discharge) arm. A random-effect model network meta-analysis was performed using the “netmeta” 3.6.2 package (R Foundation for Statistical Computing, Vienna, Austria). Within the framework, I 2 statistic was used to quantify the heterogeneity. The Q statistics, which represents the variability of treatment effect between direct and indirect comparisons in the network meta-analysis, were used to evaluate the inconsistency. p-score was used to rank between the treatments. A p-score of 100% indicates the treatment to be certainly the best for the studied outcome, whereas 0% implies the treatment to be certainly the worst. This systematic review and network meta-analysis is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.
Results
After screening and detailed review of the search results, 12 trials met the inclusion criteria and were assessed for systematic review (Supplementary Figure S1). Three trials (Compare-Acute, CvLPRIT, Hamza et al . ) comparing complete revascularization versus culprit-only revascularization included both same-sitting and staged revascularization strategies within the complete revascularization arm, and the individual result was unavailable in the main article. After contacting the corresponding author, detailed data differentiating same-sitting and staged complete revascularization was achieved from the Compare-Acute and CvLPRIT trials. , To maintain the validity of our study, a trial by Hamza et al. consisting of 58% in same-sitting and 42% in staged revascularization was excluded since detailed data was unavailable.
Eleven RCTs were deemed eligible to be included in our network meta-analysis. , Outcomes stratified by all three revascularization strategies were available in 3 studies. , , Two trials compared same-sitting complete versus culprit-only revascularization, , and 4 trials compared staged complete versus culprit-only revascularization. , , , Remaining two trials directly compared same-sitting complete versus staged complete revascularization. , A total of 7,015 patients (same-sitting complete: 699 patients, staged complete: 2,762 patients, and culprit-only: 3,554 patients) were included in our analysis. The details of study characteristics and baseline patient characteristics of each trial are available in Table 1 and 2 . The discharge medication and characteristics of the non-culprit lesion are demonstrated in Supplementary Table S2. The summary of exclusion criteria and definitions of MACE, MI, and major bleeding for individual RCTs are provided in Supplementary Table S3. Most of the trials used a composite of all-cause death or cardiovascular death, recurrent MI, and revascularization as the definition of MACE. The Compare-Acute trial added cerebrovascular events, the CvLPRIT trial added heart failure admission, and the PRAMI trial used refractory angina in place of revascularization for the definition of MACE. The diagram of network meta-analysis is demonstrated in Figure 1 .
| Comparison (patient number) | Outcomes | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Study | Author | Year | Follow-up (median, months) | Patient number | Complete (same-sitting) | Complete (staged) | Culprit-only | Interval from 1 st to 2 nd PCI (days) | MACE | All-cause death | CV death | MI | Revascularization | Stroke | Major bleeding | CI-AKI |
| Compare-Acute | Smits | 2017 | 36 | 753 | 136 | 27 | 590 | 2.1±1.0 * | + | + | + | + | + | + | + | – |
| COMPLETE | Mehta | 2019 | 36 | 4041 | – | 2016 | 2025 | 1 [IQR 1-3] or after discharge within 45 [IQR 12.5-33.5] | + | + | + | + | + | + | + | + |
| CvLPRIT | Gershlick | 2015 | 12 | 285 | 97 | 42 | 146 | During index admission | + | + | – | + | + | – | – | – |
| DANAMI-3-PRIMULTI | Engstrøm | 2015 | 27 | 627 | – | 314 | 313 | During index admission | + | + | + | + | + | + | + | + |
| Ghani | Ghani | 2012 | 36 | 119 | – | 79 | 40 | Unclear | + | + | – | + | + | – | – | – |
| HELP AMI | Di Mario | 2004 | 12 | 69 | 52 | – | 17 | – | + | + | – | + | + | – | – | – |
| Politi | Politi | 2010 | 30 * | 214 | 65 | 65 | 84 | 57±13 * | + | + | + | + | + | – | – | – |
| PRAGUE 13 | Hlinomaz | 2015 | 38 | 214 | – | 106 | 108 | 3-40 | + | + | + | + | – | + | – | – |
| PRAMI | Wald | 2013 | 23 * | 465 | 234 | – | 231 | – | + | + | + | + | + | + | + | + |
| PRIMA | Ochala | 2004 | 6 | 92 | 48 | 44 | – | 27.3 * | + | + | + | + | + | – | + | – |
| Tarasov | Tarasov | 2017 | 12 | 136 | 67 | 69 | – | During index admission | + | + | + | + | + | – | – | – |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
