Revision risk of cross-linked versus non-crosslinked polyethylene in reverse shoulder arthroplasty: an analysis of 63,502 cases from the Australian Orthopaedic Association National Joint Replacement Registry.

BACKGROUND: The effect of increased cross linking of modular polyethylene components in primary total stemmed reverse shoulder replacement (rTSA) upon revision rates is unknown outside biomechanical testing. We compared the rate of revision of primary stemmed reverse shoulder replacement with either a highly crosslinked (XLPE) or non-highly crosslinked polyethylene (non-XLPE) glenosphere or humeral cup.

MATERIALS AND METHODS: Using data from the Australian Orthopaedic Association National Joint Replacement Registry, all primary rTSA with either polyethylene humeral cups or glenospheres were identified for the period 1 January 2006 to 31 December 2024. The procedures were restricted to the three most common primary diagnoses: osteoarthritis (OA), rotator cuff arthropathy (RCA) and fracture. Within each diagnostic category, procedures were stratified by XLPE or non-XLPE polyethylene type. For each implant type (humeral cups and glenospheres), this created six comparison groups: OA (XLPE vs non-XLPE), RCA (XLPE vs non-XLPE), and fracture (XLPE vs non-XLPE). The cumulative percent revision (CPR) was determined using Kaplan-Meier estimates of survivorship and hazard ratios (HR) from Cox proportional hazard models adjusted for age and sex. A sub analysis from 2015 allowed further adjustment for ASA, BMI, glenosphere size and image derived instrumentation.

RESULTS: Of the 28,819 rTSA for OA, 24,779 for RCA and 9,904 for fracture there was an increased risk of revision for non-XLPE compared to XLPE glenospheres for OA from 1.5 years HR=3.21 (95% confidence interval (CI)1.59, 6.47), p=0.001). but not for RCA or fracture. The rate of revision was increased for OA non-XLPE vs XLPE humeral cups, but other diagnoses did not differ (HR=1.30 (95% CI 1.10, 1.54), p=0.002), with the most common reason for revision being infection in the non-XLPE OA cohort. At sub-group analysis, the revision rate of non-XLPE humeral cups was increased compared to XLPE for OA and RCA but not fracture (HR=1.84 (95%CI 1.42, 2.40), p<0.001, and HR=1.47 (95% CI 1.16, 1.86), p=0.001, respectively) by multivariable adjustment.

DISCUSSION/CONCLUSION: Changes to the amount of cross-linking to polyethylene bearing surfaces in rTSA are associated with variations in all-cause revision rates depending on the primary diagnosis. Patients with non-XLPE components are associated with a higher revision rate compared with XLPE components when used in stemmed rTSA for OA and RCA (particularly infection requiring revision). While at sub-group analysis the revision risk of rTSA with polyethylene glenosphere constructs is confounded by patient, implant characteristics, and surgeon assistive tools as covariates, the evidence suggests increased revision of rTSA with non-XLPE humeral cups.