The review found no difference over three years in mortality or major adverse cardiovascular and cerebrovascular events between percutaneous coronary intervention (PCI) and coronary artery bypass grafting for unprotected left main coronary artery stenosis. PCI patients had significantly higher target vessel revascularisation risk. It is unclear whether the authors’ conclusions are reliable due to potential review process flaws.

To compare outcomes after treatment of unprotected left main coronary artery (ULMCA) stenosis with either coronary artery bypass grafting (CABG) or percutaneous coronary intervention (PCI).

BioMed Central, ClinicalTrials.gov and PubMed databases and Google Scholar were searched from January 2004 to December 2008; search terms were reported. Proceedings of meetings of American Heart Association, The American College of Cardiology and Trans-Catheter Therapeutics were searched over the same time period. Bibliographies of relevant articles and reviews were handsearched

Unrestricted comparisons of cohorts of CABG and PCI for the treatment of ULMCA stenosis with a minimum of one year’s follow-up were eligible for inclusion if they reported survival and major adverse cardiac and cerebrovascular events (MACCE) and had results from at least 30 patients in each cohort. The primary outcome was mortality at up to three years. Secondary outcomes were MACCE and target vessel revascularisation (TVR) after the procedure. Articles that were subgroup analyses were excluded. Where reported, 81% to 99% of CABG patients received left internal mammary artery CABG; the others received left anterior descending CABG. Approximately 80% of PCI patients received drug-eluting stents; others received bare-metal stents. No details of patient age or sex were reported.

The authors did not state how many reviewers performed the study selection.

Criteria relevant to methodological quality reported included: study design, selection bias, performance bias, attrition bias, detection bias, adjustment for confounders, multicentre enrolment, randomisation and matching.

The authors did not state how many reviewers performed validity assessment.

The number of events for each outcome were extracted in order to calculate odds ratios (OR) and 95% confidence intervals (CI) at one, two and three years after treatment.

The authors did not state how many reviewers performed data extraction.

Odds ratios were pooled using a fixed-effect model (Mantel-Haenzsel) and a random-effects model (DerSimonian and Laird). Between-study heterogeneity was determined using Woolf’s test. Sensitivity analyses were performed by omitting one study at a time.

Ten relevant studies were identified (n=3,773): four prospective studies (n=2,205, range 105 to 1,084) and six retrospective studies (n=1,568, range 134 to 287). Overall, internal validity was moderate. All studies had a moderate risk of performance and detection bias. Most studies had a low risk of selection and attrition bias. Two prospective studies were randomised.

The pooled analysis found no significant difference in risk of death for PCI versus CABG at one year (OR 0.97, 95% CI 0.71 to 1.33; 10 studies), two years (OR 1.28, 95% CI 0.84 to 1.94; six studies) and three years (OR 1.11, 95% CI 0.66 to 1.85; four studies); or for MACCE (including death, myocardial infarction and stroke, excluding TVR) at one year (seven studies), two years (three studies) and three years (three studies). For TVR there was a significant benefit for CABG versus PCI at one year (OR 4.36, 95% CI 2.60 to 7.32; eight studies) and at two years (OR 4.20, 95% CI 2.21 to 7.97; four studies); the effect was no longer significant at three years (OR 3.30, 95% CI 0.96 to 11.33; three studies).

There was no evidence for heterogeneity in any meta-analyses.

There was no difference in mortality or major adverse cardiovascular or cerebrovascular events between PCI and CABG for the treatment of ULMCA stenosis for up to 3 years. PCI patients had a significantly higher risk of target vessel revascularisation. In selected patients with ULMCA stenosis, PCI was emerging as an acceptable option.

The review addressed a well-defined question in terms of participants, interventions, study design and relevant outcomes. Relevant databases were searched and unpublished studies were considered, but it was unclear whether any language restrictions were applied and so some relevant studies may have been missed. Publication bias was not assessed. It was not reported how many reviewers performed study selection, data extraction and validity assessment and whether any efforts were made to reduce error and bias in the review process. Study quality was assessed using suitable criteria. Some relevant study details were reported, but no details were given of the age or sex of patients or loss to follow-up. The statistical analysis seemed appropriate and no study heterogeneity was detected. In view of the potential for error and bias in the review process resulting from the lack of reporting of review methods, it is unclear whether the results of the review are reliable.

Practice: The authors suggested a multidisciplinary management approach among surgeons, cardiologists and primary care physicians for patients with left main coronary artery disease. In such cases, based on patient and angiographic factors, PCI could be a reasonable choice in selected patients. They also suggested that the present USA guidelines (American College of Cardiology/American Heart Association) for left main PCI should be revisited.

Research: The authors identified a need for further studies on long-term durability of PCI versus CABG.

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.