This review concluded that there was limited evidence to suggest that non-invasive cardiovascular imaging altered primary prevention efforts and further high-quality studies were needed. Clinical variability across the studies made the reliability and generalisability of the pooled results uncertain. Despite the limitations of the review and the data, the authors' conclusions seem appropriate.

To determine the impact of non-invasive cardiovascular imaging on medication prescribing, lifestyle modification, use of invasive angiography and revascularisation.

MEDLINE, EMBASE, Cardiosource Clinical Trials, Cochrane Central Register of Controlled Trials (CENTRAL), HTA database, HealthSTAR, ISRCTN Register, ClinicalTrials.gov, UpToDate, Web of Science with Conference Proceedings and What’s What Online were searched from inception to November 2010; search terms were reported. Reference lists of primary articles, reviews, editorials, imaging guidelines, the researchers' personal files, the Internet, thesis dissertations and abstract listings were searched and experts in the field were contacted.

Randomised controlled trials (RCTs) that evaluated the impact of non-invasive cardiovascular imaging on clinical care compared to usual or standard care were eligible for inclusion. Direct comparisons of imaging technologies without a control group that did not receive imaging were excluded. Studies had to report on medication prescribing, lifestyle modification, subsequent use of invasive (catheter-based) angiography and/or revascularisation.

Imaging technologies evaluated in the included studies were ultrasonography, echocardiography, exercise treadmill, computed tomography and myocardial perfusion imaging. Populations recruited into the studies varied widely and included smokers, people with diabetes or hypertension and otherwise healthy highly selected screening populations. The technologies were used to screen for myocardial ischaemia, coronary calcification, carotid atherosclerosis and left ventricular hypertrophy. Median prevalence of cardiovascular disease was 22% (interquartile range 19% to 28%). Median follow-up was 12 months (interquartile range nine to 39 months). Median age of participants ranged from 42 to 65 years. The proportion of women ranged from 15% to 100%.

Two reviewers assessed each study for inclusion; disagreements were resolved through consensus.

Study quality was assessed in terms of adequate randomisation, concealed allocation, blinding, completeness of follow-up and crossover between groups. It seemed that two reviewers independently assessed study quality.

Two reviewers independently extracted the incidence for each relevant outcome. Odds ratios (OR) with 95% confidence intervals (CI) were calculated; data were cross-checked for discrepancies. Trial authors were contacted for missing/unpublished data.

Pooled odds ratios with 95% confidence intervals were calculated using the Mantel-Haenszel random-effect model using intention-to-treat analyses. Heterogeneity was assessed using I²: I² of less than 30% was considered low heterogeneity, between 30% and 50% was moderate heterogeneity and more than 50% was substantial heterogeneity. The impact of some study quality criteria were investigated through sensitivity analyses. Publication bias was assessed using funnel plots.

Seven trials met the inclusion criteria (n=2,150, range 50 to 1,123). All seven trials used adequate randomisation methods, six reported adequate allocation concealment and blinding of at least one relevant group (patients, clinicians and outcome assessors). Loss to follow-up ranged from 1% to 18%. Two trials reported cross-over in 1% and 19% of patients.

Non-invasive imaging had no statistically significant impact on overall medication prescribing (four RCTs, n=1,500). The only specific drug for which a statistically significant impact was observed was insulin, which significantly increased (OR 1.28, 95% CI 1.00 to 1.64; two RCTs, n=1,267). Trials that reported lifestyle modifications showed no significant impact of non-invasive imaging on smoking cessation (four RCTs, n=198 smokers), dietary improvement (one RCT, n=56) and physical activity (one RCT, n=450). There was no significant impact of non-invasive imaging on subsequent invasive angiography and/or revascularisation (two RCTs, n=1,173).

There was limited evidence to suggest that non-invasive cardiovascular imaging altered primary prevention efforts.

The review addressed a clear research question supported by appropriate inclusion criteria. The search for studies was extensive and included attempts to locate unpublished data. It was unclear whether language restrictions were applied and so language bias could not be ruled out. No results for the assessment of publication bias were reported. Each stage of the review was conducted in duplicate, which reduced potential for error and bias. Study quality was assessed using appropriate criteria and the results were used in the analysis. Statistical heterogeneity was low, but there were few studies in the meta-analyses and clinical heterogeneity was apparent, so the reliability and generalisability of the pooled results was uncertain.

The authors acknowledged a number of limitations of the review, particularly the small sample size on which the analyses were based. Despite the limitations of the review and the data, the authors' conclusions seem appropriate.

Practice: The authors did not state implications for practice.

Research: The authors stated that further high-quality studies were needed.

Canadian Institutes of Health Research; Canadian Stroke Network; individual authors received awards from several organisations.

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.