Yttrium synovectomy: a meta-analysis of the literature

Background: Yttrium synovectomy for chronic synovitis of the knee enjoys widespread usage in Australia with approximately 400 patients receiving yttrium-90 in 1991. Despite abundant anecdotal evidence of its efficacy there is a paucity of controlled trials and those that have been done have produced conflicting results and have been of insufficient sample size.
Aims: To critically and quantitatively evaluate the published English literature on comparative trials of yttrium-90 therapy for chronic synovitis of the knee.
Methods: The technique of meta-analysis was utilised. The literature search was carried out using the MeSH terms of synovectomy and knee; and yttrium. This was augmented by referring to reviews, current textbooks and back-references. Outcome measures varied between trials but could be grouped as treatment success. The Peto modification of Mantel and Haenszl was used for statistical pooling of data yielding a pooled odds ratio (OR).
Results: The literature search revealed ten controlled trials of which two were excluded from further analysis. Yttrium was superior to placebo (OR 2.42, 95% CI 1.02–5.73) but this result should be interpreted with caution due to possible publication bias. Yttrium was not superior to triamcinolone (OR 1.89, 95% CI 0.81-10.55) or other active modalities (OR 1.04, 95% CI 0.72-1.52). Further research comparing yttrium with other modalities is necessary to properly determine its place in rheumatological practice. (Aust NZ J Med 1993; 23: 272–275.)