Despite interventions being mostly successful in preventing development of diabetes, there were no reductions in mortality or myocardial infarction, with the possible exception of stroke. The author's conclusion reflects the evidence on diabetes incidence, but major uncertainties and insufficient data for all other outcomes preclude definitive conclusions on the overall safety and efficacy of these interventions.

To determine whether interventions aimed at prevention of diabetes alter macrovascular outcomes among people with impaired glucose tolerance and impaired fasting glucose.

PubMed and EMBASE were searched until March, 2010; search terms were reported. Bibliographies of retrieved articles were handsearched to locate further studies.

Eligible for inclusion were prospective randomised controlled trials (RCTs), investigating the effects of a lifestyle intervention (diet, exercise, or diet with exercise) or pharmacological intervention in adults with impaired glucose tolerance or impaired fasting glucose. Development of diabetes was a required outcome; all trials had to be published in English. Trials that contained less than 100 participants and follow-up of less than one year were excluded.

Percentage of male participants per trial ranged from 32 to 87% (where reported); mean participant age ranged from 45.3 to 55.7 years. Non-pharmacological interventions included: group education; individual and group counselling sessions; and tailored lifestyle advice. Control arms received routine or limited diet and exercise advice. Pharmacological interventions administered metformin, acarbose, voglibose, rosiglitazone, pioglitazone, nateglinide, ramipril and valsartan; most control arms administered placebo drugs. One trial contained participants with established cardiovascular disease or one or more risk factors for the disease.

Two reviewers selected studies for inclusion; discrepancies were resolved through consensus.

Two reviewers assessed study quality; no other details were provided.

Incidence of diabetes, all-cause mortality, cardiovascular death, fatal and non-fatal myocardial infarction, and fatal and non-fatal stroke were extracted to calculate risk ratios (RRs) with 95% confidence intervals (CIs).

Two reviewers independently extracted the data.

Risk ratios and 95% confidence intervals were pooled for each outcome using a fixed-effect or random-effects model, according to the level of statistical heterogeneity indicated by the Χ² statistical test. Sensitivity analyses were conducted for each outcome by removing trials that did not demonstrate diabetes prevention, and trials of rosiglitazone. Sub-group analyses for pharmacological and non-pharmacological interventions were performed. A funnel plot was constructed to assess publication bias for cardiovascular mortality.

Ten RCTs (23,152 participants) were included in the review; trial sample size ranged from 207 to 9,306 participants. Intervention follow-up ranged from 2.8 to six years.

Risk of diabetes was statistically significantly lower with intervention groups versus control groups (RR 0.66 95% CI 0.55 to 0.80, ten trials), significant statistical heterogeneity was found (p<0.001). The effect was stronger with non-pharmacological interventions than with pharmacological ones (p<0.05).

A lower risk of fatal and non-fatal stroke was found for intervention groups versus control groups, but this was of borderline statistical significance (RR 0.76, 95% CI 0.58 to 0.99, three trials). No significant statistical heterogeneity was found (p<0.872). A similar result was obtained when rosiglitazone was removed; statistical significance was lost when only trials demonstrating diabetes prevention were examined (RR 0.76, 95% CI 0.57 to 1.01).

There were no statistically significant differences between interventions in all-cause mortality; sensitivity and subgroup analyses did not significantly alter the findings.

No evidence for publication bias was found.

Despite interventions being mostly successful in preventing development of diabetes, this did not result in reductions in cardiovascular or all-cause mortality, or myocardial infarction, with the possible exception of stroke.

The review question was clear and relevant data sources were searched. Inclusion criteria seemed sufficiently replicable. All studies were published in English, which increased the likelihood of language bias. Publication bias was assessed but there were too few trials for the results to have been meaningful. Attempts were made to minimise error and bias during study selection and data extraction. Details and results from any quality assessment were not reported so the quality of the studies was unclear. The pooling of pharmacological and non-pharmacological may not have been appropriate given their differences.

The authors stated that individual studies were underpowered for the mortality and cardiovascular outcomes, there were differences in reporting of cardiovascular risk factors, and follow-up periods were short. Studies that assessed pharmacological interventions other than rosaglitazone also appear to be underpowered. The author's conclusion reflects the evidence on the reduction of diabetes incidence, but major uncertainties and insufficient data for all other outcomes preclude definitive conclusions on the overall safety and efficacy of these interventions.

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

Research: The authors stated that further research that examined the effectiveness of prediabetes interventions for reducing stroke and myocardial infarction were needed.

Alfred Health; the National Health and Medical Research Council (Australia).

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.