Radiographic geometry and clinical glenohumeral range of motion after reverse shoulder athroplasty, a retrospective cohort study

BACKGROUND: The range of motion (ROM) in reverse shoulder arthroplasty (RSA), is mechanically limited by the surrounding bony obstacles especially in abduction and rotation planes. However, the clinical effect of implant positioning, prosthesis design, and individual differences in bone morphology, on ROM is obscure. The aim of this study was to investigate the correlation between radiographic geometry and clinical glenohumeral (GH) ROM after RSA. METHODS: RSA patients operated at Turku University Hospital during 2007–2013 were called for radiological and clinical follow-up. Pre- and postoperative true anteroposterior radiographs were obtained and the positioning of the center of rotation (COR) in relation to the surrounding bony structures was measured. Active and passive shoulder and GH abduction, flexion, internal and external rotation ROM were measured with goniometer. The Constant score (CS) and pain visual analogue scale (VAS) were recorded. The correlation between the radiographically measured parameters and the active and passive ROM and clinical outcome was statistically analyzed. RESULTS: 91 shoulders were available for analyses with a mean follow-up of 38.7 months ± SD 20 (range 12–83) months. 77% of the patients were female, the mean age was 73 (SD 9) years. The mean angle between the line of supraspinatus fossa, and the line between COR and lateral edge of the acromion (α-angle) was 127° (SD 14) and the mean angle between the lines from lateral edge of the acromion to COR, and from there to the superior edge of the greater tubercle (β-angle) was 54° (SD 11). The mean active shoulder flexion at follow-up was 118° (SD 26), abduction 104° (SD 32), external rotation 41° (SD 22), internal rotation 77° (SD 21). The mean passive GH flexion was 80° (SD 19), abduction 67° (SD 15), external rotation 31° (SD 16) and internal rotation 34° (SD 14). The mean Constant score at follow-up was 53 (SD 18) and pain VAS 2 (SD 3). The positioning of the radiographically measured COR did not statistically significantly correlate with the ROM or clinical outcome scores. CONCLUSIONS: Postoperative radiographically measured two-dimensional geometry and positioning of the COR does not significantly correlate with the glenohumeral range of motion or clinical results after RSA. LEVEL OF EVIDENCE: Level 3, retrospective cohort study