Analysis of retracted articles in the surgical literature.

Background: Retractions of scientific articles represent attempts to correct the literature. Our goal was to examine retracted surgical papers.

Methods: NCBI PubMed database was queried using the search terms "surgery," "surg," or "surgical" and "retracted" or "retraction." Article details were recorded.

Results: There were 184 retracted surgical articles identified from 1991 through 2015. Average retraction time was 3.6 years. General (26%), Cardiac (22%), and Orthopedic (10%) surgery were most common. Reasons for retraction were duplication (35.3%), Institutional Review Board violations (18.5%), falsified data (14.7%), data errors (9.8%), author dispute (8.2%), plagiarism (7.6%), copyright violations (2.2%), financial disclosure violations (0.5%), and consent (0.5%). No reason for retraction was given in 8.7% of cases. Median IF was higher for administrative than content-related retraction reasons (3.0 vs. 2.0, P < 0.01). A paywall, requiring a subscription to read, restricted access to 23.4% of retraction notices.

Conclusions: Article retractions occur across all fields of surgery for various reasons, both administrative and content-related. The majority of surgical retraction notices have a reason for retraction listed and do not require payment to read.