Results of Arthroscopically-Assisted Core Decompression for Avascular Necrosis of the Proximal Humerus

OBJECTIVES: Avascular necrosis (AVN) of the proximal humerus treatment is challenging. Early recognition and treatment prior to collapse may potentially prevent or delay subsequent progression. Arthroscopically-assisted core decompression of the proximal humerus is one treatment option for AVN. However, there is little published data regarding outcomes of this procedure. This study represents the largest-to-date cohort of patients who underwent arthroscopically-assisted core decompression of the proximal humerus. METHODS: A retrospective review was performed to analyze the outcomes of patients who underwent arthroscopically-assisted core decompression for proximal humeral AVN. All patients had associated shoulder pain that had failed non-operative management. A comprehensive review of operative reports and imaging were examined to verify the use of an arthroscopically-assisted surgical technique without violation of the articular surface. Concomitant shoulder pathology was surgically addressed at the time of decompression. Patient charts were retrospectively reviewed for pre- and post-operative pain as well as shoulder range of motion. Pre-operative and post-operative x-rays were examined for evidence of progression of AVN lesions. A review of documentation for post-operative shoulder injections or repeat surgery was completed. Subjective American Shoulder and Elbow (ASES) outcomes were compared pre- and post-operatively RESULTS: Eleven consecutive patients and 13 consecutive shoulders were identified who underwent arthroscopically-assisted core decompression of the proximal humerus AVN by the two senior surgeons from 2004-2013. No patients were excluded. Six males and five females were identified with an average age of 45 years at surgery. Average follow-up length was 2.6 years. Four shoulders underwent synovectomy and core decompression alone. Concomitant procedures included rotator cuff repair in 5, subacromial decompression in 4, distal clavicle excision in 6, and biceps tenodesis in 1 shoulder(s). Radiographically, the majority of shoulders (8) remained stable without progressive sclerosis or collapse. Two shoulders demonstrated progressive sclerosis, and articular collapse was seen in 1 shoulder. The average pre-operative ASES score was 19.3, compared to 44.8 at post-operative follow-up (p=.001). Pre-operative shoulder elevation averaged 156 degrees, while post-operative range of motion averaged 170 degrees (p=.81). The majority of patients had mild or moderate pain pre-operatively, while 4 patients required narcotics for severe pre-operative pain. Visual analogue pain averaged 8.4 pre-operatively, compared to 4.9 at most recent follow-up (p=.0004). While post-operative pain initially improved in all patients, the majority of patients had return of mild, intermittent pain with longer follow-up. Two patients who took narcotics pre-operatively continued to require narcotics at final follow-up. Six shoulders had undergone therapeutic shoulder injections post-operatively. No patient had required repeat surgery. CONCLUSION: Arthroscopically-assisted core decompression is an option for the treatment of proximal humeral AVN. Visual analogue pain and ASES functional outcomes were reliably improved at average follow-up of 2.6 years. Three shoulders demonstrated radiographic AVN progression, while the majority of lesions remained stable. Six shoulders underwent post-operative injections for pain.