Critical analysis of causality between negative ulnar variance and Kienböck disease.

Abstract

Negative ulnar variance has been associated with Kienböck disease, both supporting a causal link and providing a basis for therapeutic recommendations. The aim of this study was to determine whether there is a causal relationship between Kienböck disease and negative ulnar variance. The causal relationship between Kienböck disease and negative ulnar variance was assessed using three methodologies: (1) an analysis of the quantitative and qualitative distribution of ulnar variance in a case-control study of 81 patients with Kienböck disease and a control group of 212 healthy wrists; (2) a systematic literature review and meta-analysis of six case-control studies (including the case-control study presented in this article) to test for an association between Kienböck disease and negative ulnar variance; and (3) a determination of causal relationship by using the Bradford Hill criteria. The case-control study indicated that when Kienböck disease is associated with negative ulnar variance (59 of 81), in 75 percent of the cases negative ulnar variance was equal (29 of 59) or less (15 of 59) pronounced on the contralateral healthy side. The odds ratio generated from the meta-analysis demonstrated a significant association between Kienböck disease and negative ulnar variance (OR, 3.58; 95 percent CI, 1.59 to 8.06; p = 0.002, random effects method). Six of nine Bradford Hill criteria do not support a causal relationship. The significant association between Kienböck disease and negative ulnar variance can be well explained by selection bias because magnetic resonance imaging, which may detect and allow exclusion of a potential ulnar impaction syndrome, was performed in only the Kienböck disease group in all six case-control studies. The application of the Bradford Hill criteria does not provide sufficient scientific evidence to support a causal relationship between Kienböck disease and negative ulnar variance. Risk, III.