Complete Revascularization in STEMI Seems to Cut CV Mortality: Meta-analysis

Either fractional flow reserve or angiography can be used to guide nonculprit PCI, but a comparative study would be welcome.

Complete Revascularization in STEMI Seems to Cut CV Mortality: Meta-analysis

Patients with STEMI and multivessel disease have better outcomes when they undergo complete revascularization versus PCI of the culprit artery alone, a new meta-analysis affirms.

In pooled results from 10 randomized trials, including the 4,041-patient COMPLETE trial, multivessel PCI was associated with lower rates of CV mortality, new MI, and a composite of the two outcomes, with odds ratios of 0.69, 0.68, and 0.69, respectively, according to researchers led by Kevin Bainey, MD (Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, Canada).

The findings were not significantly affected by the type of guidance used for PCI of the nonculprit lesions—fractional flow reserve (FFR) or angiography—or the timing of the additional procedures, they report in their paper published online recently in JAMA Cardiology.

Guideline recommendations from the United States and Europe to consider PCI of nonculprit lesions in STEMI patients with multivessel disease, issued before publication of COMPLETE, “were limited to small-sample-size RCTs with lower power to detect differences in CV death or new MI. In addition, most trials included revascularization in the primary composite outcome, which is subject to criticism in an open-label trial,” Bainey et al write.

COMPLETE, which showed that angiography-guided, staged PCI of nonculprit lesions reduced the composite of CV death or new MI compared with culprit-only PCI, bolstered the guideline recommendations, but—like the prior trials exploring this question—was not powered to assess the impact on CV mortality alone.

“With this meta-analysis, I think we’re confident [in saying] that we can improve cardiovascular survival with a complete revascularization approach in those patients that are suitable for complete revascularization,” Bainey told TCTMD.

Reduce MI and Lessen CV Mortality?

For the meta-analysis, the investigators identified 10 relevant trials involving a total of 7,030 unique patients with STEMI and multivessel coronary disease. The largest studies were COMPLETE, COMPARE-ACUTE, and DANAMI3-PRIMULTI. In seven of the trials, nonculprit PCI was guided by angiography, with FFR guidance used in the rest. There was variation in the timing of the additional procedures—ie, during the index procedure or staged—across studies.

Through a weighted mean follow-up of 29.5 months, multivessel PCI was associated with lower rates of CV death, new MI, or a composite of the two endpoints compared with culprit-only PCI, with no difference observed in all-cause mortality.

Outcomes With Multivessel vs Culprit-Only PCI


Multivessel PCI

Culprit-Only PCI

OR (95% CI)

CV Death or New MI



0.69 (0.55-0.87)

CV Death



0.69 (0.48-0.99)

New MI



0.68 (0.49-0.96)

All-Cause Death



0.84 (0.67-1.05)

The type of guidance for and the timing of nonculprit PCI did not significantly affect any of the relationships.

Bainey et al note that the lower rate of CV death observed with complete revascularization is concordant with the results of a recent meta-analysis published in the European Heart Journal  that included six trials.

“This reduction in CV mortality is consistent with a robust reduction in new MI observed with complete revascularization,” they write, adding that the results of an optical coherence tomography substudy of COMPLETE “have demonstrated that approximately one-half of obstructive nonculprit lesions contain unstable plaque morphology. Hence, routine nonculprit-lesion PCI as a preventive strategy could reduce subsequent MI and potentially improve CV long-term survival.”

Possible Guideline Changes

Contacted for comment, Mark Eisenberg, MD (McGill University, Montreal, Canada), said that he had done a similar analysis—currently under review—with McGill medical student Jeremy Levett as lead author. “While there is a promising signal that complete revascularization reduces CV mortality, further data is needed to validate the robustness of this measurement,” they told TCTMD in an email, noting that the 95% confidence interval was wide and that the upper bound approached 1.00. In addition, they said, “this estimate is highly dependent on which studies are included.”

Overall, Eisenberg and Levett said, the RCT data suggest that complete revascularization has an advantage over culprit-only PCI in STEMI patients with multivessel disease. “While this alone is likely sufficient to warrant complete revascularization as a first-line therapy, further data is needed to describe this effect on hard outcomes such as CV mortality,” they argued. “Operators should remain sensitive to the timing and strategy (visually or physiologically guided) of complete revascularization, based on the clinical status of their patients and until further randomized data becomes available.”

Though the meta-analysis did not show that the type of guidance or timing of nonculprit PCI made a significant impact on the findings, Eisenberg and Levett pointed out that there were hints of an advantage for angiographic versus FFR guidance.

“Although estimates are wide and overlapping, this meta-analysis opens the possibility that in the context of a STEMI, a physiology-guided complete revascularization strategy may be suboptimal compared to a visual angiography-guided complete revascularization strategy,” they said. “Randomized data is needed to test this hypothesis.”

Bainey cautioned against reading too much into any potential differences between the strategies, but agreed that there is equipoise to support a randomized trial exploring the question.

For now, he said, there is enough evidence to strengthen guideline recommendations. Although there are still some questions about the optimal approach for intervening on the nonculprit lesions, he added, “I do think the guidelines should support complete revascularization and perhaps shift it to a class I indication.”

Todd Neale is the Associate News Editor for TCTMD and a Senior Medical Journalist. He got his start in journalism at …

Read Full Bio
  • Bainey reports receiving grants and personal fees from AstraZeneca, Bayer, Boehringer Ingelheim, and Bristol-Myers Squibb/Pfizer during the conduct of the study.