Primary glenohumeral degenerative joint disease: factors predisposing to arthroplasty

To evaluate the natural history of shoulder osteoarthritis (OA), in particular the requirement for arthroplasty over time, and to determine the potential predisposing factors for such arthroplasty.

In- and out-patients with the diagnosis of OA of the shoulder seen between January 1990 and December 1994 were contacted by mail or telephone in 2000. Evaluation at the time of diagnosis: demographics, clinical and radiological data were evaluated at the time of diagnosis. The follow-up evaluation consisted of a questionnaire sent to each patient inquiring whether they had had recourse to shoulder arthroplasty and, if not, evaluated their willingness regarding this surgical approach. For the statistical analysis the requirement for arthroplasty over time was evaluated using the Kaplan-Meier technique. Potential factors predisposing to arthroplasty were determined using a Cox-model analysis.

The questionnaire was answered by 72 of the 86 contacted patients. No difference was observed in clinical and radiological variables at the time of diagnosis between responders and non-responders. The requirement for arthroplasty was low (respectively 5% and 13%, 5 and 10 years following the onset of the symptoms). Nearly half of the patients who had not undergone surgery thought that arthroplasty would have been the treatment of choice to improve their quality of life since more than one year at the time of completing the questionnaire. Two variables were picked up in the Cox analysis, with a probability of shoulder replacement higher in patients with concomitant osteonecrosis of the humeral head (p = 0.02) and a non-eccentric glenohumeral OA (p = 0.011).

The low percentage of patients with arthroplasty over time, together with patient perception, suggest underuse of this surgical approach in shoulder OA.