Atraumatic osteonecrosis of the humeral head

To define the epidemiology, clinical and radiographic presentation, treatment, and prognosis of atraumatic osteonecrosis of the humeral head.

Of the 1,056 patients managed for osteonecrosis of any joint between July 1, 1974, and December 1, 1996, 127 shoulders in 73 patients were treated for atraumatic osteonecrosis of the proximal humerus. Clinical and radiographic characterization of this patient cohort was performed.

At presentation, there were 47 women and 26 men with a mean age of 41 years (range 20-60). Numerous associated factors were noted: alcohol use (38%), moderate smoking (30%), asthma (8%), and nephrosis (3%). A corticosteroid association was noted in 60 patients (82%) and 42 of the patients (58%) had an immunocompromising disease. The severity of humeral head osteonecrosis did not correlate with dose or duration of corticosteroid therapy. According to the modified Ficat and Arlet radiographic staging system, there were 20 shoulders with Stage I disease, 55 shoulders with stage II disease, and 52 shoulders with Stage ITI or IV disease. Seventy-four of the shoulders treated with core decompression (78%) had good to excellent clinical outcomes at a mean followup of 6 years (range 2-21). Fourteen of the 16 patients (88%) treated with hemiarthroplasty or total shoulder arthroplasty were clinically successful at a mean followup 4 years (range 2-11).

We observed a low incidence of humeral head involvement in the osteonecrosis patient cohort (7% of all osteonecrosis patients), and a high incidence of corticosteroid use (82%). hip involvement (81%), and bilateral disease (74%). Osteonecrosis of the humeral head should be suspected in patients presenting with shoulder pain and a history of osteonecrosis in other joints. Hip screening for osteonecrosis is advocated in patients with shoulder involvement. Early detection of shoulder osteonecrosis may permit a more conservative, joint-sparing approach as an alternative to surgical management.