Abstract
Introduction
We reports the accuracy of direct Magnetic Resonance Arthrography (MRA) in detecting Triangular Fibrocartilage Complex (TFCC), Scapho-Lunate Ligament (SLL) and Luno-Triquetral Ligament (LTL) tears using wrist arthroscopy as the gold standard.
Methods
We reviewed the records of all patients who underwent direct wrist MRA and subsequent arthroscopy over a 4-year period between June 2007 and March 2011. Demographic details, MRA findings, arthroscopy findings and the time interval between MRA and arthroscopy were recorded. The scans were performed using a 1.5T scanner and a high resolution wrist coil. All scans were reported by a musculoskeletal radiologist. Sensitivity, specificity, positive and negative predictive values (PPV & NPV) were calculated.
Results
Two hundred and thirty four (234) MRA were performed over the study period. Fifty patients (50), who subsequently underwent 51 wrist arthroscopies (one bilateral), were included. The mean age was 35 years (range 16–64 years). The average delay between MRA and arthroscopy was 4.8 months (median 4 months, range 17 days–18 months). All patients were symptomatic with wrist pain.
At arthroscopy, 26 TFCC tears, 7 SLL tears and 3 LTL tears were found.
For TFCC, sensitivity was 96%, specificity 88%, PPV 89% and NPV 96%. For SLL, the values were 57%, 66%, 21% and 91% respectively. For LTL, 67%, 79%, 17% and 97%, respectively. Receiver Operating Characteristic (ROC) curve analysis showed that MRA only reliably differentiates between patients with and without TFCC tears (Area Under Curve AUC = 0.92, p < 0.0001) but not SLL (AUC = 0.62, p=0.28) or LTL (AUC = 0.73, p=0.17) tears.
Conclusion
MRA is a sensitive and specific imaging modality for diagnosing TFCC tears. However, the diagnostic accuracy for SLL and LTL tears was not satisfactory. Wrist arthroscopy remains the gold standard if there is a clinical suspicion of inter-carpal ligament tears.
