From Wiki Journal Club
Jump to navigation Jump to search
Smits PC, et al. "Fractional Flow Reserve-Guided Multivessel Angioplasty in Myocardial Infarction". The New England Journal of Medicine. 2017. 376(13):1234-1244.
PubMedFull textPDF

Clinical Question

In patients with multivessel coronary disease presenting with ST elevation MI (STEMI), is complete revascularization using percutaneous coronary intervention (PCI) with fractional flow reserve (FFR) guidance superior to infarct-related artery (IRA) only PCI with regard to major cardiovascular adverse events (death, MI, revascularization, stroke)?

Bottom Line

In patients with multivessel coronary disease presenting with STEMI, FFR-guided complete revascularization is associated with a 12.7% absolute reduction in major cardiovascular adverse events (death, MI, revascularization, stroke). This was driven primarily by an 11.4% absolute reduction in the need for revascularization as well as a trend toward fewer MIs.

Major Points

PCI is standard of care for STEMI and treatment of the infarct-related artery (IRA) has an established mortality benefit in this disease. However, up to 30% of patients presenting with STEMI are found to have multivessel CAD with significant stenosis seen in one or more non-infarct-related arteries (N-IRA). It is unclear whether these lesions are simply incidental, in which the risks of intervening upon them after culprit vessel revascularization may outweigh the benefits, or whether these N-IRA lesions contribute to a pro-thrombotic milieu, for which intervention may reduce the risk of recurrent events. Older observational and registry studies suggest that multivessel revascularization either at the time of index PCI or within a short interval lead to either similar or worse outcomes. [1][2] However, these retrospective analyses are limited by confounding, as sicker patients may be more likely to undergo more aggressive intervention at the time of PCI. In addition, PCI technology has improved significantly since the time of these analyses and thus the risk profile of more complex intervention may be lower than previously suggested.

As a result, the role of upfront or staged complete revascularization using multivessel PCI has been the subject of multiple RCTs including CvLPRIT, PRAMI, and DANAMI-3 PRIMULTI. Despite differences in trial design, these trials generally suggested improved cardiovascular event rates with multivessel rather than IRA only PCI. More recently, a meta-analysis including both CvLPRIT and PRAMI showed a 40% lower odds of MACE with complete revascularization, with a nonsignificant trend towards improved cardiovascular mortality. It is thought that intervention even upon non-culprit lesions may lead to pacification of an inflammatory milieu in the peri-STEMI anatomy that may lead to decreased risk of recurrent thrombosis and possible salutary effects on ischemic/stunned myocardium. On the strength of these findings, the 2015 ACC/AHA/SCAI guidelines on primary PCI for STEMI now include PCI of N-IRAs in "selected patients with STEMI and multivessel disease who are hemodynamically stable, either at the time of primary PCI or as a planned staged procedure" as a class IIb recommendation.

A limitation of existing trials of multivessel PCI in STEMI is their reliance on angiographic severity in order to determine whether treatment of a lesion is indicated. When compared to angiographic severity, functional significance as measured using fractional flow reserve (FFR, a method of determining physiologically significant coronary stenoses by objectively measuring flow reduction caused by an atheroma) has been associated with improved outcomes in stable coronary disease (see FAME 2). Whether FFR-guidance has a role in guiding multivessel PCI in the setting of STEMI remains unproven.

The 2017 Compare-Acute trial randomized 885 patients presenting with acute STEMI found to have multivessel disease on angiography to IRA only PCI or FFR-guided revascularization of N-IRAs and assessed for a primary outcome of death, nonfatal MI, any revascularization, and cerebrovascular events. Patients randomized to the complete revascularization group could undergo immediate PCI during the index procedure (83%) or staged PCI during the same admission (17%). At 12 months, there was a 12.7% absolute reduction in the primary outcome in patients randomized to FFR-guided PCI of N-IRAs. This was driven primarily by an 11.4% absolute reduction in revascularizations, with no difference in mortality and a numerical trend toward fewer nonfatal MIs. As expected, the benefit of N-IRA PCI was accentuated in the subgroup of patients with lesions that were physiologically significant by FFR. In summary, the Compare-Acute trial provides further evidence that treatment of non-infarct related arteries discovered during STEMI is beneficial, although results suggest that the benefit is largely restricted to a reduction in need for future revascularization rather than "harder" outcomes such as death or MI.


As of March 2018, no guidelines have been published that reflect the results of this trial.


  • Multicenter, open-label, randomized, controlled trial
  • N=885
    • Infarct-related artery only PCI (N=295)
    • FFR-guided complete revascularization (N=590)
  • Setting: 24 sites in Europe and Asia
  • Enrollment: July 2011 - October 2015
  • Duration of follow-up: 12 months
  • Analysis: Intention-to-treat
  • Primary Outcome: Death, nonfatal MI, revascularization, or cerebrovascular event


Inclusion Criteria

  • Age 18-85
  • Presenting with STEMI within 12 hours after the onset of symptoms and eligible for primary PCI
  • At least one stenosis ≥ 50% in a non-infarct related artery judged to be suitable for PCI by the operator

Exclusion Criteria

  • Left main disease (> 50% stenosis)
  • STEMI due to in-stent thrombosis
  • Chronic total occlusion of a non-culprit vessel
  • Severe stenosis with TIMI flow < II of non-culprit vessel
  • Non-culprit vessel not amenable to PCI
  • Complicated culprit lesion treatment (at least one of the following):
    • Extravasation
    • Permanent no-reflow after culprit lesion treatment
    • Inability to place a stent
  • Known severe valve dysfunction that will require surgery in the follow-up period
  • Killip class III or IV at presentation or at completion of culprit vessel PCI
  • Life expectancy < 2 years
  • Intolerance to study drug
  • Gastrointestinal or genitourinary bleeding within the prior 3 months
  • Planned elective surgical procedure necessitating interruption of thienopyridines during the first 6 months post enrollment
  • Pregnancy or planning to become pregnant any time after enrollment into the study

Baseline Characteristics

From the complete revascularization group.

  • Demographics: Age 62 years, male 79%, white 89%
  • Comorbidities: BMI 27, DM 15%, HTN 46%, smoker 41%, HLD 32%, stroke 3%, MI 8%, PCI 9%, CKD 1%, PAD 3%, Killip class ≥ 2 5%
  • Infarct: Posterior 18%, anterior 36%, inferior 51%, lateral 14%
  • Symptoms to PCI: < 6 hours 76%, 6-12 hours 16%, > 12 hours 8%
  • Stenosis: Two arteries 69%, three arteries 31%
  • PCI: Treatment of FFR-positive N-IRA 55%
    • during index PCI 83%, staged 17%
  • Stent: DES 99%, mean number of stents 1.6, mean length 34 mm


  • Patients randomized 1:1
    • Complete revascularization using FFR guidance (N=295)
    • IRA only PCI (N=590)
  • Patients randomized after successful PCI of the infarct-related artery
  • In complete revascularization group, FFR guidance used to determine treatment of N-IRAs (with at least 50% stenosis angiographically)
  • Treatment of N-IRAs encouraged during index procedure but could be staged prior to discharge
  • In patients randomized to IRA only PCI, FFR measurements obtained but then procedure was stopped
    • Further investigations and management of care determined by treating cardiologist and could include revascularization of N-IRAs
    • Elective clinically indicated revascularizations in this group performed within 45 days after primary PCI not counted as events
  • Follow-up conducted at 30 days, 12, 24, and 36 months


Comparisons are complete revascularization vs. infarct-related artery only treatment

Primary Outcomes

Death, nonfatal MI, revascularization, or cerebrovascular event
23 (7.8%) vs. 121 (20.5%); HR 0.35 (95% CI 0.22-0.55); p < 0.001

Secondary Outcomes

4 (1.4%) vs. 10 (1.7%); HR 0.80 (95% CI 0.25-2.56); p = 0.70
Nonfatal MI
7 (2.4%) vs. 28 (4.7%); HR 0.50 (95% CI 0.22-1.13); p = 0.10
18 (6.1%) vs. 103 (17.5%); HR 0.32 (95% CI 0.20-0.54); p < 0.001
Hospitalization for heart failure, unstable angina, or chest pain
13 (4.4%) vs. 47 (8.0%); HR 0.54 (95% CI 0.29-0.99); p = 0.04

Adverse Events

Major bleeding
3 (1.0%) vs. 8 (1.4%); HR 0.75 (95% CI 0.20-2.84); p = 0.67
Stent thrombosis
2 (0.7%) vs. 1 (0.2%); HR 0.58 (95% CI 0.12-2.80); p = 0.50

Subgroup Analysis

  • The reduction the primary outcome was greater for patients in the complete revascularization group among patients with at least one lesion with FFR ≤ 0.80 (8.9% vs. 30.7%, p < 0.001)


  • Patients randomized to infarct-related artery only PCI could still undergo revascularization at the discretion of the treating cardiologist (occurring in 59 patients). This may lead to underestimation of the benefit of FFR-guided complete revascularization.
  • Patients could only be randomized after successful PCI and only if hemodynamically stable after index PCI was complete, limiting generalizability to less stable patients.
  • Open label design may lead to bias, particularly in which patients randomized to IRA only revascularization may be more likely to undergo subsequent revascularization due to providers knowing that lesions were not treated previously.


  • Study supported by Maasstad Cardiovascular Research, which received unconditional grants from Abbott Vascular and St. Jude Medical

Further Reading