Dor no ombro: paradoxo entre o modelo patoanatômico e medidas clínicas

Abstract

Introduction: The evaluation of patients with shoulder pain is highly influenced by pathoanatomical factors. Special tests and imaging are commonly used to the clinical decision-making. However, some studies have called this model into question suggesting more diversified evaluation systems such as movement-based and psychosocial aspects. Objectives: To verify the association of pathoanatomical factors with the symptom presentation and determine if pathoanatomical factors contribute to the self-reported shoulder function. Methods: Magnetic resonance imaging was used to evaluate in detail pathoanatomical abnormalities in individuals with unilateral shoulder pain. Images in the coronal, sagittal, and axial planes were generated and independently interpreted by a board-certified, orthopedic fellowship trained orthopedic shoulder surgeon and a musculoskeletal radiologist. Frequencies of pathoanatomical abnormalities for both shoulders were compared and the agreement across the evaluators was verified. In order to assess if pathoanatomical factors contributed to the self-reported shoulder function, a multivariate model was built considering the total score of the Disabilities of the Arm, Shoulder, and Hand (DASH) as the dependent variable. Clinical, demographics, pain catastrophizing, and special tests were used as explanatory variables. Results: Both shoulders presented a high prevalence of pathoanatomical abnormalities. Symptomatic shoulders showed the highest frequency of full-thickness tear and glenohumeral osteoarthrosis. Other observed pathoanatomical abnormalities did not show statistically significant differences between both shoulders. Pathoanatomical variables did not contribute to the multivariate model. The group of variables that best explained the selfreported shoulder function were pain intensity at rest, pain catastrophizing level, and acromioclavicular joint tenderness during palpation. Conclusions: Full-thickness rotator cuff tears and the presence of glenohumeral osteoarthrosis seem to be the only pathoanatomical abnormalities associated with the symptoms. Elevated pain catastrophizing levels and pain intensity at rest as well as acromioclavicular joint tenderness were the best explanatory variables to decreased self-reported shoulder function. Health professionals must be aware that most pathoanatomical abnormalities assessed with the magnetic resonance imaging are not related to the symptoms.