Comparison of the responsiveness of symptomatic outcome measures in knee osteoarthritis.

ABSTRACT

OBJECTIVE: A number of international scientific societies have recommended a core set of domains to be systematically assessed in clinical research studies on osteoarthritis (OA), i.e., pain, function, and patient's overall assessment. This open, longitudinal, observational study compares the responsiveness of different symptomatic variables evaluating these 3 domains in knee OA. METHODS: Patients were individuals with painful knee OA. The collected data were Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale (0-100) and WOMAC function subscale (0-100), Lequesne's index (0-100), pain after physical activities (visual analog scale [VAS] 100 mm), and patient's global assessment (VAS 100 mm). The procedure used was knee joint lovage. Time of collection was before and 1, 3, and 6 months after the lavage. Analysis was by comparison of the standardized response mean (mean of the changes/SD of the changes) in an intent-to-treat strategy after 1, 3, and 6 months using the jackknife method. RESULTS: Improvement in all dimensions of WOMAC subscale scores and VAS scores was observed at month 1. Lequesne's index was not responsive to change. The standardized response mean was moderate, ranging from 0.00 to 0.40. Comparison of the estimates of the standardized response means using the jackknife method showed a statistically significant difference between Lequesne's index and the WOMAC subscale for function, but not between VAS pain and the WOMAC subscale for pain. CONCLUSION: Most of the evaluated variables have a moderate responsiveness. In knee OA, the WOMAC function scale seems to be more sensitive than Lequesne's index for detecting changes after symptomatic therapy.