Statistical Fragility of Randomized Controlled Trials Evaluating Platelet-Rich Plasma Use for Knee Osteoarthritis: A Systematic Review

Abstract

Background: Numerous studies have been published on the use of platelet-rich plasma (PRP) for knee osteoarthritis (OA), with conflicting results. Purpose: To determine the fragility index (FI) and fragility quotient (FQ) of randomized controlled trials (RCTs) that evaluated the use of PRP to treat knee OA. Study Design: Systematic review. Methods: RCTs evaluating the efficacy of PRP injections for knee OA from 2000 to 2020 were included for analysis according to PRISMA guidelines. The FI was determined by calculating the number of outcome event reversals required to change the statistical significance. The associated FQ was determined by dividing the FI by the sample size. Results: Our initial search resulted in 41,149 studies, of which 8 RCTs (678 patients, 72 outcome events) were included in the analysis. One study failed to report PRP formulation details, whereas 87.5% of studies reported using either leukocyte-rich or leukocyte-poor PRP. The platelet concentration was reported in 25% of the included trials. The overall FI of the 72 outcome events was 8.5. Accounting for sample size, the associated FQ was determined to be 0.14, suggesting that the reversal of 14% of outcome events was required to change outcome significance. There were 51 statistically significant outcomes, of which the FI and FQ were 12 and 0.164, respectively. Conclusion: Comprehensive fragility analysis suggested that the published literature evaluating the efficacy of PRP use for knee OA may lack statistical stability. We recommend the reporting of both an FI and FQ in addition to P value analysis to provide a clear and thorough understanding of the statistical integrity of studies reporting on PRP use for knee OA.