Comparison of the clinical examination, magnetic resonance imaging and intraoperative arthroscopic findings of the shoulder joint: a retrospective study

Abstract

Purpose of the study: Diagnosing shoulder pathologies is rather challenging. The reason is not only the anatomical structure and biomechanics of the joint, but also the concomitant conditions, specifically of cervical spine, which can significantly contribute to the pathophysiology. The aim of this study was to evaluate the diagnostic accuracy of clinical examination and magnetic resonance imaging in relation to the actual intraoperative finding in patients with selected shoulder pathologies.

Material and methods: The data of patients in whom shoulder arthroscopy was performed at the Department of Trauma Surgery, University Hospital Ostrava in the 2018-2019 period were extracted retrospectively from the hospital information system. Our attention was focused on 4 diagnoses, namely rotator cuff tear (RCT), anterior shoulder instability due to lesion of the anterior capsular-labral complex (STA), shoulder impingement syndrome (impingement) and SLAP lesion – superior labral anterior posterior tear (SLAP).

Results: The study included 71 patients, of whom 46 were men (65%) and 25 women (35%). The difference in the median age of men and women was not significant (p = 0.740). High specificity (clinical examination 91.8%, MRI 91.8%) and low sensitivity (clinical examination 58.0%, MRI 61.0%) were evident in all diagnoses. In shoulder instability (STA) the sensitivity was significantly higher than in the other diagnoses. The accuracy of both methods was also high (clinical examination 79.9%, MRI 81.0%).

Discussion: These matters have already been described in literature. Nonetheless, most of the studies compared the MRI and arthroscopy findings. Our study reflects the actual diagnostic procedure, i.e. the examination of the patient, referral for MRI for suspected pathology and subsequent surgery to confirm or rule out the diagnosis. Magnetic resonance imaging can detect also other pathological findings that are not apparent during shoulder arthroscopy.

Conclusions: High specificity of clinical examination as well as MRI was confirmed by this study. The sensitivity of both examination methods was significantly lower, except for the diagnosis of anterior shoulder instability (STA), where it was satisfactorily high for both methods. In the case of a positive clinical finding, in rotator cuff tear (RCT) diagnosis the MRI scan provides higher accuracy and eliminates false positives, which was, however, not confirmed in the other diagnoses. In the case of a negative clinical finding, the MRI helps refine the diagnoses. This does not apply to the STA diagnosis, where we may assume that a negative finding of clinical examination will most likely mean a negative intraoperative finding and this negative predictive value will not be increased by the MRI.