Patient Selection in Randomized Controlled Trials of Total Knee Arthroplasty: A Systematic Review Assessing Inclusion Criteria.
Academic Article
Overview
abstract
BACKGROUND: Patients undergo total knee arthroplasty (TKA) at widely disparate stages of osteoarthritis, raising the possibility of high heterogeneity among patients enrolled in TKA research studies. Obscuration of treatment effectiveness and other problems that may stem from cohort heterogeneity can be controlled in clinical studies by rigorously defining target patients. The purpose of this review was to determine the extent to which randomized controlled trials (RCTs) on TKA use osteoarthritis severity, as defined by radiographic grade or patient-reported outcome measures (PROMs), in their inclusion criteria, and to investigate potential impact on outcome. METHODS: A search of PubMed, Embase, Scopus, Web of Science, and Cochrane databases used a combination of terms involving TKA, PROMs, and radiographic scoring. A total of 1,227 studies were independently reviewed by 2 screeners for the above terms. RCTs with ≥100 patients (236) and with <100 patients (325) undergoing TKA were analyzed with regard to the specific inclusion criteria and recruitment process. RESULTS: Among the identified RCTs with ≥100 patients, 18 (<8%, with a total of 2,952 randomized patients) used specific radiographic scoring or PROM thresholds in their inclusion criteria. Eleven of the 18 studies used specific radiographic scoring, such as the Kellgren-Lawrence or Ahlbäck classifications. Three studies used preoperative PROM thresholds: Knee Society Knee Score of <60, Knee Society Function Score of <60, Oxford Knee Score of <20, and Hospital for Special Surgery Score of <60. Among studies with <100 patients, 48 (<15%) used specific inclusion criteria. CONCLUSIONS: The vast majority of RCTs (>85%) did not enroll patients based on disease severity, as measured by PROM score thresholds or radiographic classifications, in their inclusion criteria. The lack of consistent inclusion criteria likely results in heterogeneous cohorts, potentially undermining the validity of RCTs on TKA. LEVEL OF EVIDENCE: Prognostic Level I. See Instructions for Authors for a complete description of levels of evidence.
