Comparison of dislocation rates of Teflon and Titanium stapes prostheses: a retrospective survival analysis on 855 patients.

Résumé

BACKGROUND: Stapes prosthesis dislocation is the first cause of revision stapes surgery. To our knowledge, there is no data about stability of the incus attachment of manual crimped prosthesis of different materials. This study aimed to compare the dislocation incidence between titanium and fluoroplastic stapes prostheses.

METHOD: A monocentric retrospective cohort study was conducted between January 2013 and June 2022 in a tertiary-care center. All patients that underwent a primary stapes surgery with manually crimped fluoroplastic or titanium prostheses were included. Prosthesis dislocation from the incus was identified intraoperatively or with CT scan. The incidence of stapedial prosthesis dislocation over time was estimated using the Kalbfleisch and Prentice survival analysis method. Other indications for revision surgery prior to prosthesis dislocation were considered as competing events. Differences in the cumulative incidence functions between the fluoroplastic group and the titanium group was assessed using the Gray's test.

RESULTS: Eight hundred and fifty-five patients underwent primary stapes surgery during the study period. Fluoroplastic prosthesis was used in 758 (88.7%) cases and titanium prosthesis in 97 (11.3%) cases. Median follow-up was 51.7 months (28.4-80.1). Dislocation was observed in 23 (3.0%) patients with fluoroplastic prosthesis and none (0.0%) in the titanium group. The probability of prosthesis dislocation at two years after surgery was 3.5% in the Teflon group and 0.0% in the Titanium group. No significant difference was found in the cumulative incidence of prosthesis dislocation between the fluoroplastic group and the titanium group (p = 0.12).

CONCLUSIONS: Despite lack of statistical power, our results suggest a trend in a more stable incus attachment of manually crimped titanium stapes prosthesis compared to fluoroplastic over time. Further prospective randomized studies could be valuable to assess our findings.