Rates of medial tibiofemoral joint space narrowing in osteoarthritis studies consistent despite methodological differences.

RATIONALE: Minimum tibiofemoral joint space width in the medial compartment (JSW) is the most well-established structural outcome measure for osteoarthritis (OA) of the knee. Its usefulness as a measure of therapeutic effectiveness in short-term studies is limited by the rate and variability of joint space narrowing (JSN) in the OA population. Microfocal radiography has been shown to improve reproducibility of JSW measurement compared to standard radiography, but measurement of magnification from microfocal knee films has been problematic, and JSN is yet to be investigated in a longitudinal microfocal study. OBJECTIVE: To establish the effect on JSW reproducibility of a new method of magnification measurement in microfocal radiographs. To report on and compare rates of medial tibiofemoral JSN and their variations in the placebo arms of microfocal and standard radiographic clinical trials in OA, using fluoroscopic semi-flexed (SF) knee positioning. To place in the context of published estimates of rates of JSN from comparable studies. METHODS: Using microfocal radiography, 36 patients were followed at a single centre for 2 years. Using standard radiography, 86 patients were followed for 1 year at a single centre, and 549 for 2 years in a multi-centre international study. Computerised JSW measurement was undertaken using enhanced and automated versions of existing algorithms. Rates of JSN were examined in the context of a review of published rates of JSN using a variety of techniques. RESULTS: Reproducibility of JSW measurement from microfocal radiographs was improved by the new magnification measurement. Rates of JSN were similar across the studies, but more variable when using standard radiography. The rates of JSN were also consistent with those from previously published investigations; all estimates since 2000, bar one, being consistent with the value 0.05 mm/year. CONCLUSION: Microfocal radiography using the new method lowered the variability of the rate of JSN, but the high cost and low availability of microfocal equipment remains a barrier to its more widespread use. The consistently low but highly variable rates of JSN seen in the review suggest that continued attempts to improve radiographic and mensural techniques are unlikely to significantly reduce required sample sizes.