Statistically Significant Results Favored in Abstracts of Platelet Rich Plasma Treatment of Knee Osteoarthritis: A Systematic Review and Spin Analysis.

Abstract

To assess the prevalence and associated variables of spin bias in the abstracts of systematic reviews and meta-analyses of knee osteoarthritis (KOA) treatment with intra-articular platelet-rich plasma (PRP). PubMed and Embase were queried for peer-reviewed systematic reviews with or without meta-analysis reporting on intra-articular PRP treatment of KOA. Two authors independently analyzed the abstracts for the 15 most common types of spin by comparing details within the abstract to the full manuscript. Methodological quality was appraised using A Measurement Tool to Assess Systematic Reviews, Version 2 scoring. Descriptive statistics were used for spin and A Measurement Tool to Assess Systematic Reviews, Version 2 scores. Logistic regression and categorical association tests (Fisher's and chi-square) assessed the effect of study characteristics on the presence of spin. One or more forms of spin were found in 92% (45/49) of evaluated abstracts. The most common form of spin identified was type 11, "conclusion focuses selectively on statistically significant efficacy outcome," which was identified in 63% (31/49) of studies. A Measurement Tool to Assess Systematic Reviews, Version 2 ratings were high for 10% (5/49) of studies, moderate in 8% (4/49), low in 27% (13/49), and critically low in 55% (27/49). Decreasing level of evidence was associated with inappropriate extrapolation spin (odds ratio = 2.528, 95% confidence interval 1.037-6.161, P = .041). Spin bias is highly prevalent in the abstracts of systematic reviews and meta-analyses of intra-articular PRP to treat KOA. Findings should be interpreted in the context of the subjective scoring system used to assess spin. A lower level in the hierarchy of evidence was associated with the presence of inappropriate extrapolation spin. Identified spin tended to favor the use of PRP in treating KOA. In the present study, abstract conclusions were found to have a high prevalence of spin. Identifying and understanding spin aids in the critical appraisal of research with implications for clinical decision making in the usage of PRP.