Quality of conduct and reporting of meta-analyses of surgical interventions.

BACKGROUND: Meta-analyses are useful tools for summarizing surgical evidence as they aim to encompass multiple sources of information on a particular research question, but they may be prone to methodological and reporting biases. We evaluated the conduct and reporting of meta-analyses of surgical interventions.

METHODS AND FINDINGS: We performed a systematic review of 150 meta-analyses of randomized trials of surgical interventions published between January 2010 and June 2011. A comprehensive search strategy was executed using MEDLINE, EMBASE, and the Cochrane Database of Systematic Reviews. Data were independently extracted by 2 authors using the PRISMA statement (preferred reporting items for systematic reviews and meta-analyses, a standardized quality of reporting guideline) and AMSTAR (a tool for methodological quality). Descriptive statistics were used for individual items, and as a measure of overall compliance, PRISMA and AMSTAR scores were calculated as the sum of adequately reported domains. A median of 8 trials (interquartile range = 8) was included in each meta-analysis. One third of all meta-analyses had an author with a background in epidemiology and/or statistics. Forty-four percent were published in PRISMA-endorsing journals with a median impact factor of 3.5. There was moderate compliance with PRISMA, with an average of 71% of items reported, but poorer compliance with AMSTAR, with 48% of items adequately described, on average.

CONCLUSIONS: Substantial gaps in the conduct and reporting of meta-analyses within the surgical literature exist, mainly in the specification of aims and/or objectives, the use of preplanned protocols, and the evaluation of potential bias at the review (rather than trial) level. Editorial insistence on using reporting guidelines would improve this situation.