The review concluded there was no evidence to recommend routine intravascular ultrasound-guided percutaneous coronary intervention with bare-metal stent implantation; this may be explained by the paucity and variation of the available studies. Despite some review limitations the authors' cautious conclusions appear appropriate, although variation across the included studies means the generalisability of the findings is uncertain.

To evaluate the efficacy of intravascular ultrasound in percutaneous coronary intervention with bare metal stent implantation.

PubMed, EMBASE and Web of Knowledge were searched from inception to April 2011 without language restrictions; search terms were reported. A manual search of reference lists of relevant reviews and meta-analyses was performed.

Eligible studies were randomised controlled trials (RCTs) that compared intravascular ultrasound-guided percutaneous coronary interventions (PCIs) with angiography-guided PCIs with bare metal stent implantation. The primary outcomes were: death; nonfatal myocardial infarction; and the composite outcome of major adverse cardiovascular events (MACE). The need for revascularisation was assessed.

Criteria for optimal stent implantation, MACE definitions, target vessels and lesion lengths varied across the studies. Mean ages ranged from 54 to 63 years. Around three-quarters of patients were male. Some patients had a previous myocardial infarction but the level of reporting of prior disease status varied. Where reported, the proportion of patients with prior myocardial infarction ranged from 32% to 54%, for dyslipidaemia the range was 40% to 67% and for diabetes 8% to 21%. Some studies used both pre- and post-intervention intravascular ultrasound and some used just post-intervention ultrasound.

Two reviewers independently selected studies for inclusion. Disagreements were resolved by discussion or a third reviewer where consensus could not be reached.

The authors did not appear to formally assess study quality. Some aspects of study quality were considered: blinding, duration of follow-up and use of an intention-to-treat analyses.

Intention-to-treat (ITT) data were extracted in order to calculate odds ratios (OR) with 95% confidence intervals (CI).

The authors did not state how many reviewers extracted data.

Meta-analyses (of studies with at least six months of follow-up) were performed to calculate pooled odds ratios with 95% confidence intervals, using a random effects model. Heterogeneity was assessed using the Ι² statistic. Publication bias was assessed using the Duval and Tweedie trim and fill method; Egger's test was used to investigate the impact of certain factors on treatment effect.

Eight studies were eligible but three were excluded from the analyses for using a stenting technique that was no longer performed. The other five studies included 1,754 participants (range 144 to 744). Follow-up ranged from six to 18 months. Four studies used pre-intervention blinding and one used post-intervention blinding. Three used an intention-to-treat analysis.

There were no statistically significant differences between groups for MACE (OR 0.74, 95% CI 0.49 to 1.13; five RCTs; Ι²=57%), death (OR 1.86, 95% CI 0.88 to 3.95; five RCTs; Ι²=0%) and non-fatal myocardial infarction (OR 0.65, 95% CI 0.27 to 1.58; four RCTs; Ι²=49%).

There was evidence of publication bias and small studies effects for MACE.

There was no evidence to recommend routine intravascular ultrasound-guided percutaneous coronary intervention with bare-metal stent implantation. This may be explained by the paucity and heterogeneity of the studies published so far.

The review addressed a clear question and was supported by reproducible eligibility criteria. Three studies were excluded post hoc from the main analyses; the impact of two of these was investigated in sensitivity analyses. Attempts to identify relevant studies in any language were undertaken by searching electronic databases and checking references. There was no specific search for unpublished studies so it was possible some relevant studies were missed.

Two reviewers independently selected the studies, which minimised the risks of reviewer error and bias affecting the process; it was unclear whether similar methods were used during data extraction. Study quality was not formally assessed; no information was provided to allow evaluation of the randomisation and allocation concealment processes and there was no detail on the level of drop-outs. This limited reporting made it difficult to evaluate the reliability of the trial results. Sufficient study details were provided and appropriate methods were used to pool data and to assess heterogeneity; there was no formal investigation of the sources of heterogeneity.

Despite some limitations of the review and the primary studies the authors' cautious conclusions appear appropriate, although the observed heterogeneity means the generalisability of the findings is uncertain.

Practice: The authors did not state any implications for practice.

Research: The authors stated that studies on specific subgroups and a simple large randomised trial may show different results to this review.

National Council for Scientific and Technological Development, Brazil.

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.