RESUMEN
BACKGROUND: Recurrent subluxation or dislocation of the extensor carpi ulnaris (ECU) tendon from the ulnar groove is an important cause of ulnar-sided wrist pain. Demonstration of ECU subluxation on MRI is of unclear clinical significance. Previous studies have suggested wrist positioning can affect the ECU's position relative to the ulnar groove. This study evaluates the relationship between ECU subluxation and wrist positioning on MRI, and separately their association with ulnar-sided symptoms. METHODS: 161 wrist MRI scans of 141 patients from four hospitals were retrospectively evaluated for wrist position (defined by radio-ulnar angle), degree of ECU subluxation and the presence of ulnar-sided symptoms and MRI abnormalities. 30 scans were scored by two different raters to assess interrater reliability. A linear regression model was constructed to assess the relation between wrist positioning and subluxation, accounting for other variables. A logistic regression model was constructed to evaluate which variables are predictive of ulnar-sided symptoms. RESULTS: ECU subluxation was neither significantly correlated to wrist position (p = 0.338) nor predictive of the presence of ulnar-sided symptoms (odds ratio 1.28, 95% CI 0.39-4.18). ECU position varied widely for all wrist positions and subluxation occurred in all wrist positions, both in symptomatic and asymptomatic subjects. No trend was observed towards more frequent subluxation in supination, contrary to previous studies. Interrater reliability for radioulnar angle and ECU displacement was excellent (intraclass correlation coefficient for consistency 0.993 and 0.943, respectively). CONCLUSION: ECU subluxation occurs frequently in all wrist positions, irrespective of ulnar-sided symptoms, and is not associated with ulnar-sided symptoms.