Surgical versus Non-surgical Management of Rotator Cuff Tears: Predictors of Treatment Allocation

OBJECTIVES: Rotator cuff tears are a common shoulder disorder resulting in significant disability to patients and strain on the health care system. While both surgical and non-surgical management are accepted treatment options, little data exist to guide the surgeon in treatment allocation. Defining variables to guide treatment allocation may be important for patient education and counseling, as well as to deliver the most efficient care plan at the time of presentation. The objective of this study was to identify patient characteristics at the time of initial clinical presentation that are associated with allocation to surgical versus non-surgical management for patients with known full-thickness rotator cuff tears. METHODS: 185 consecutive adult patients with known full-thickness rotator cuff tears were enrolled into a prospective cohort study. Robust data were collected for each subject at baseline, including age, gender, body mass index (BMI), shoulder activity score, smoking status, size of cuff tear, duration of symptoms, functional comorbidity index, treating surgeon, the American Shoulder and Elbow Society (ASES) score, the Western Ontario Rotator Cuff Index (WORC), and the Veterans Rand 12 Item Health Survey (VR-12). Logistic regression was performed in order to identity variables associated with treatment allocation, and the corresponding odds ratios were calculated. RESULTS: Of the 185 subject enrolled, 100 underwent surgical intervention and 85 non-operative management. While controlling for co-variates, significant baseline patient characteristics predictive of eventual allocation to surgical treatment included the following: non-smoking status [OR .039 (0.005, 0.300) p=0.002], lower functional comorbidity index [OR 0.739 (0.518, 1.055) p=0.096], younger age [OR 0.872 (0.820, 0.927) p<0.001], lower BMI [OR 0.895 (0.826, 0.970) p=0.007], and symptoms present for fewer than 4 months [OR 3.258 (1.070, 9.921) p=0.038]. Factors that were not associated with treatment allocation included gender, tear size, diabetes, treating surgeon, or any of the patient derived outcome scores at presentation (ASES, WORC, VR-12, shoulder activity score). CONCLUSION: This prospective cohort study suggests that the factors predictive of treatment allocation are related more to patient demographics at presentation than patient derived outcome scores or intrinsic characteristics of the rotator cuff such as tear size. Further study is warranted to help define appropriate indications for treatment allocation in patients with rotator cuff tears.