Grammont reversed prosthesis for acute complex fracture of the proximal humerus in an elderly population with 5 to 12 years follow-up.

PURPOSE OF THE STUDY: The aim of this retrospective study was to analyze outcome in 23 cases of Delta III reverse ball-and-socket total shoulder prosthesis implantation for acute complex fractures of the proximal humerus in an elderly population with poor bone quality. In this type of population, this procedure could respond to the difficulties of a reliable and efficient re-fixation of the tubercles.
MATERIAL AND METHODS: From 1993 to 2000, 23 Delta III prostheses were implanted by a single operator for acute injury: 18 three-part and four-part fractures and five fracture-dislocations. The study population included two men and 21 women, mean age 75 years, ten on the dominant side. Surgery was performed under general anesthesia in the semi-sitting position via the anterolateral approach without osteotomy of the acromion, with 10-20° retroversion of the humeral stem (except in one shoulder) and cement fixation in two. For five shoulders, the tubercles were re-fixed. Postoperative physiotherapy was not possible in all patients. Outcome was assessed with the Constant score and with AP and lateral Lamy radiographs.
RESULTS: Seven patients died so the series included 16 cases for analysis. Complications were: reflex sympathetic dystrophy (n=2), postoperative Acinetobacter infection (n=1) requiring revision to clean and drain allowing preservation of the prosthesis, and early postoperative anterior dislocation (n=1) (10° stem anteversion) with surgical revision to re-orient the stem. At a mean follow-up of 86 months, the Constant score was 60 points (contralateral shoulder, 83 points). Outcome, influenced in cases of re-fixation of the tubercles, was good for pain (14.1), activity (13.3), strength (16.1), anterior elevation (6.5), and abduction (6.5), but very poor for external (1.1) and internal (2.4) rotation. The radiographs showed aseptic glenoid loosening (n=1) at 12 years with surgical revision in 2005; the Constant score at 6 months follow-up was 48 points, inferior scapular notching (n=11) according the Nerot classification (six stage 1, four stage 2, one stage 3, at 2, 4.3, and 5 years follow-up), inferior spurs (n=9) appearing at a mean 2.5 years follow-up (stable after emergence without clinical impact), proximal humeral resorption (n=4) (medially for three at a mean 8 years follow-up and one laterally at 10 years), and a humeral radiolucent line (n=1) at 5 years follow-up. DISCUSSION: For acute complex fractures of the proximal humerus in elderly subjects with poor bone quality, when effective and reliable re-fixation of the tubercles is difficult or impossible, the reverse ball-and-socket shoulder prosthesis is a possible alternative providing good functional outcome except for rotations, but with the risk of inferior scapular notching. Although not problematic in the medium term, these notches may contribute to glenoid loosening with bone loss in the long term. Nevertheless, this procedure seems to improve the status of patients with such fractures.