Abstract
Background
Scapular notching causes glenoid bone loss after a reverse total shoulder arthroplasty (rTSA). The goal of this study was to assess the influence of prosthesis design on notching.
Methods
Prospective, single surgeon cohort. Two different rTSA designs were consecutively implanted and compared: 25 Delta III rTSAs and 57 Delta Xtend rTSAs in 80 patients. Notching (Nerot 0–4) was assessed at 24 months follow-up. Patient dependent variables, surgical technique and implant geometry were assessed. Multivariate binary logistic regression was used to select the strongest independent predictors of notching.
Results
The Delta III showed significantly more notching than the Delta Xtend: 72% and 23% respectively, p<0.001. The extent of notching was comparable. One patient (Delta III) needed revision for notching-associated glenoid loosening. Only 3 variables were significantly associated with notching in multivariate analysis: glenosphere overhang (R square 0.65), prosthesis-scapular neck angle (PSNA, R square 0.18) and humeral cup depth (R square 0.05), predicting 88% of notching cases. The corresponding odds ratios were 0.15 (95% CI 0.05–0.44) for 1 mm extra overhang, 8.4 (95% CI 2.0–35.6) for 10 degrees increase in PSNA and 7.6 (95% CI 1.3–43.3) for 1 mm extra cup depth. Surgical technique related variables, including peg-glenoid rim distance and PSNA, were comparable in both design groups.
Conclusion
The key to prevent notching was to utilise the design features that maximise glenosphere overhang. Therefore, as a rule of thumb the baseplate should be positioned as inferior as possible. Minor contributions came from PSNA (patient anatomy/surgical technique) and polyethylene cup depth (also design). One patient required early revision for notching associated baseplate loosening. Long term follow-up is indicated to assess the effect of notching on prosthesis survival and outcome after revision.
