Low bone density and post-operative outcomes following revision total shoulder arthroplasty
TL;DR
A documented diagnosis of LBD is associated with higher rates of periprosthetic fracture following rTSA, and sub-group analysis suggests that this association is primarily driven by patients with osteoporosis, whereas osteopenia alone was not associated with increased fracture risk.
Abstract
Background Low bone density (LBD) may increase complication risk following revision total shoulder arthroplasty (rTSA), but the magnitude and time course of this risk in the revision setting have not been well characterized, with few studies reporting short-term or midterm outcomes. This study compared 90-day and 2-year outcomes after rTSA in patients with and without a documented diagnosis of LBD. Methods We performed a retrospective cohort study including patients from January 1, 2005, through February 1, 2024. LBD was defined by a documented diagnosis of osteoporosis or osteopenia (International Classification of Diseases, Tenth Revision M80, M81, or M85.8) present at the time of the index revision. Adult patients undergoing rTSA with a diagnosis of LBD were matched to patients without such a diagnosis on demographics, comorbidities, and revision indication, so that documented LBD was the only variable differing between cohorts. Ninety-day post-operative outcomes and 2-year surgical outcomes were evaluated. Sub-group analyses assessed outcomes separately for osteoporosis and osteopenia cohorts. Relative risks with 95% confidence intervals were calculated. Results After matching, 1,495 patients per cohort were included. At 90 days, a diagnosis of LBD was associated with higher rates of periprosthetic fracture (7.2% vs. 5.2%; relative risk (RR) 1.37, P = .028) and hospital readmission (24.6% vs. 17.9%; RR 1.38, P < .001). At 2 years, periprosthetic fracture remained significantly higher in the LBD cohort (11.2% vs. 8.7%; RR 1.29, P = .021), while mechanical loosening, dislocation, periprosthetic joint infection, and re-revision were not significantly different. In the osteoporosis sub-group (n = 963 per cohort), osteoporosis was associated with increased 2-year periprosthetic fracture risk (12.5% vs. 9.0%; RR 1.38, P = .015). The osteopenia sub-group (n = 640 per cohort) did not demonstrate a significant difference (9.8% vs. 8.9%; RR 1.11, P = .565). Conclusion A documented diagnosis of LBD is associated with higher rates of periprosthetic fracture following rTSA. Sub-group analysis suggests that this association is primarily driven by patients with osteoporosis, whereas osteopenia alone was not associated with increased fracture risk. These findings highlight the importance of bone health characterization when evaluating risk profiles in patients undergoing rTSA.