This review concluded that, in selected coronary lesions, direct stenting improved outcome (mainly the incidence of myocardial infarction) in patients undergoing percutaneous coronary intervention compared with conventional (balloon pre-dilation) stenting. Although there were some reporting limitations, the authors' conclusion is consistent with the evidence presented and appears likely to be reliable.

To compare the effectiveness of direct stenting with stenting with pre-dilation in patients undergoing percutaneous coronary intervention.

MEDLINE, EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL) were searched from inception to June 2009, with no language restrictions. Search terms were reported. Eight relevant websites and meeting abstracts from four journals were searched for additional studies. Reference lists of retrieved articles were reviewed.

Randomised controlled trials (RCTs) that compared direct stenting with conventional (with balloon pre-dilation) stenting in patients undergoing percutaneous coronary intervention were eligible for inclusion. Trials with more than 10% of patients lost to follow-up were excluded.

The primary outcome was the composite endpoint of death or myocardial infarction. Secondary outcomes were myocardial infarction and target-vessel revascularisation.

In included trials, all the implanted stents were bare-metal stents of varying types (where reported). The mean age of participants ranged from 56 to 66 years. All clinical endpoints were evaluated up to six months.

Two reviewers independently selected titles and abstracts, with discrepancies resolved using a third reviewer. Details on how full papers were selected were not reported.

The quality of the included trials was assessed on randomisation, adequacy of allocation concealment, analysis by intention-to-treat, sample size calculation, and loss to follow-up.

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

Two reviewers extracted outcome data for the calculation of odds ratios (ORs) and 95% confidence intervals (CIs). If target-vessel revascularisation was not available, target-lesion-revascularisation outcome data were extracted.

Odds ratios and 95% confidence intervals were pooled in a meta-analysis using the Peto fixed-effect model. For analyses showing significant differences in effect, a secondary analysis was conducted using the DerSimonian and Laird random-effects model if heterogeneity between the trials was significant (otherwise the Mantel-Haenszel fixed-effect model was used). Heterogeneity between trials was assessed using χ² and I².

Sensitivity analyses were performed in which individual trials were removed one at a time. Meta-regression analysis was performed to evaluate the impact of crossover rates of each trial on the primary outcome.

Publication bias was assessed by a funnel plot and Begg's test.

Twenty-four RCTs were included in the meta-analysis (n= 6,803 patients). The methods of randomisation were described as appropriate in 14 trials that used computer-generated randomisation sequence. Sample-size calculation was reported in 15 trials. All the trials performed intention-to-treat analysis and reported the number of patients lost to follow-up. Crossover occurred in 21 trials, with rates ranging from 2.5 to 29.7%.

There was significant reduction in a composite of death or myocardial infarction in patients undergoing percutaneous coronary intervention who received direct stenting compared with conventional pre-dilation stenting (OR 0.76, 95% CI 0.60 to 0.96; I²=12%; 24 trials; n=6,803 patients). No statistically significant difference was found between direct and conventional stenting for rate of death (p=0.26).

Patients treated with direct stenting were less likely to experience an episode of myocardial infarction (OR 0.77, 95% CI 0.59 to 0.99; I²=16%; 24 studies; n=6,803 patients) than those treated with conventional stenting, but similar rates of target-vessel revascularisation were found between the two groups (24 studies; I²=0%).

Sensitivity analyses did not change any of these findings. The meta-regression analyses did not demonstrate any relationship between crossover rates and the death odds ratios in the two groups.

The funnel plots and Begg's tests did not indicate evidence of publication bias.

In selected coronary lesions, direct stenting improved outcome (mainly the incidence of myocardial infarction) in patients undergoing percutaneous coronary intervention.

This review addressed a well-defined question for participants, interventions, outcomes, and study design. The search included using relevant databases to identify studies in any language, although it was unclear to what extent unpublished material was sought. However, funnel plot and Begg's test did not suggest evidence of publication bias. To minimise bias and errors during the review process, two reviewers independently selected abstracts and extracted data, although it was unclear how many reviewers selected the full papers.

Relevant quality assessment criteria were applied to the different study design components, although the results for allocation concealment were not reported. The characteristics of the individual trials were presented. Potential sources of heterogeneity were explored. The chosen method of synthesis appeared to be appropriate. Sensitivity analysis demonstrated that the results were robust to changes in the factors considered.

Although not all the quality assessment results were explicitly reported, the authors' conclusion was consistent with the evidence presented and appears likely to be reliable.

The authors did not state any explicit implications for practice or research, although they noted that the review results could not be extended to drug-eluting stents.

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.