Concerning the radiologist dealing with tennis players, the demand is high for immediate and accurate diagnosis after a shoulder joint injury. Until recently, diagnostic imaging in the acute phase after injury mostly relied on radiography and ultrasound examinations. Ultrasonography is readily available and cheap, but has its limitations because the accuracy is operator-dependent and ultrasonography does not show all shoulder joint structures. In recent years, both professional and recreational tennis players are increasingly seeking magnetic resonance (MR) imaging soon after trauma. MR arthrography (MRA) is a well established and accurate imaging modality for diagnosis of internal derangements of the shoulder joint. However, immediately following trauma patients are often in pain during the MR examination and can only be immobilised for a short time. Applying the standard MR sequences, mainly spin echo T1-weighted fast saturated (T1-FS) with acquisition times of 5 mins for each imaging plane, often results in degraded MR images due to movement artefacts. VIBE (volumetric interpolated breathhold examination) is a gradient echo sequence that only lasts a few seconds, and nevertheless demonstrates a high contrast-to-noise ratio between gadolinium contrast and the rotator cuff. If the accuracy of VIBE to diagnose internal derangements of the shoulder joint is comparable to the accuracy of the standard sequences, it may become the preferred MR arthrography sequence in athletes who cannot immobilise the shoulder for a prolonged time because of pain after recent trauma. As a first project, we performed a study to evaluate the accuracy of fast MRA with VIBE sequences for diagnosing tears of the rotator cuff using the T1-FS sequence as the imaging gold standard. Eighty-two athletes, mainly tennis players, underwent direct MRA of the shoulder joint with both VIBE and T1-FS sequences in the axial and paracoronal plane. The T1-FS sequence took 5 mins for each imaging plane; the VIBE sequence lasted only 13 seconds. To identify and evaluate rotator cuff tears, two radiologists independently performed separate blind scorings of VIBE and T1-FS images. Scores were defined as normal, fraying or small partial thickness undersurface rotator cuff tears, large partial thickness undersurface rotator cuff tears, and full thickness tears with or without extension to the acromioclavicular joint. Accuracy of VIBE was calcuCalculating the results, sensitivity and specificity values over 90% were found for large partial thickness undersurface tears and for full thickness tears with or without contrast extension in the acromioclavicular joint (Figure 1a and b). However, for fraying and small partial thickness tears at the undersurface of the rotator cuff, the sensitivity was only 55%. This was felt to be related to the lower spatial resolution obtained using the VIBE sequence, and was mainly a consequence of the very short acquisition time.lated per score, using the T1-FS sequence as the imaging gold standard. In conclusion, MR arthrography using VIBE sequences was found to be an accurate tool for diagnosing large partial thickness undersurface tears and full thickness tears of the rotator cuff. Due to its short acquisition time, VIBE may be the preferred MR sequence for evaluating the rotator cuff of patients who cannot lie still because of pain after recent injury, or patients suffering from claustrophobia.
© Copyright 2007 Medicine and Science in Tennis. Alle Rechte vorbehalten.