The methodological quality of randomized controlled trials in plastic surgery needs improvement: a systematic review

Abstract

Background: Our objective was to assess the methodological quality of randomized controlled trials (RCTs) in Plastic Surgery.

Methods: An information specialist searched MEDLINE for the period of 1 January 2009 to 30 June 2011 for the MESH heading "Surgery, Plastic" with limitations for English language, human studies and randomized controlled trials. Results were manually searched for RCTs involving surgical techniques. The papers were then scored with the authors' seven point extended version of the Linde Internal Validity Scale (ELIVS). Secondary scoring was then performed and discrepancies resolved by consensus.

Results: 57 papers met the inclusion criteria. The median ELIVS score was 3.0 with a range of 1.0-6.5. Compliance was poorest with use of intention to treat analysis (4%), blinding of patients (23%) and the handling and reporting of patient withdrawals (25%). There was no statistically significant correlation between journal ELIVS score and 2010 impact factor or number of authors (Spearman rho 0.10 and 0.27 respectively). Multicentre trials had a higher average ELIVS score than single centre ones (3.6 vs 2.7) although this did not reach significance. There was no correlation between the volume of RCTs performed in a particular country and methodological quality.

Conclusion: The methodological quality of RCTs in Plastic Surgery needs improvement.