This is the first study comparing stability of the attachment of the loop onto the LPI for two kinds of manual crimping stapes prostheses, fluoroplastic (Causse, Medtronic) and titanium (Big Easy, Medtronic). Prosthesis dislocation from its attachment to the LPI is the major cause of revision procedures for recurrent CMHL after primary stapedotomy [17,18,19]. To our knowledge, the incidence of stapes prosthesis dislocation had not been described before.

In this large series of 855 primary stapes procedures over a period of nearly 10 years and with a median follow-up of 51.7 months (28.4–80.1), we found 23 (3.0%) dislocations among the 758 fluoroplastic prostheses and none (0.0%) among the 97 titanium prostheses. The probability of prosthesis dislocation at two years after surgery was 3.5% in the Teflon group and 0.0% in the Titanium group. Although we were unable to show a statistically significant difference between the two groups, these results show a trend toward more stable fixation to the incus over time with titanium.

Several types or brands of stapes prostheses of different materials are available on the market. Most stapes prostheses require manual crimping, which is crucial for the success of the procedure [20, 21]. The choice of the prosthesis is based on multiple parameters: biocompatibility, safety, hearing results, stability over time, surgeon's habits, price, and, probably increasingly, carbon footprint.

Postoperative hearing results have already been extensively studied, regarding material or surgical technique (partial or total stapedectomy, stapedotomy with manual perforator, microdrill or laser…) [3, 6, 7, 22, 23]. To date, no stapes prosthesis exhibited significant better hearing results over others.

In our daily practice, 0.6 mm diameter fluoroplastic stapes prosthesis is our first choice for primary stapedotomy, whereas 0.5 mm diameter titanium prosthesis is used in selected situations, such as narrow oval window niche, thin or eroded LPI, patient having already been operated on with a titanium prosthesis on the contralateral ear.

There are several limitations to our study: retrospective design, risk of loss of follow-up of patients with prosthesis dislocation managed in other center, and routine use of fluoroplastic prostheses versus selected use of titanium prostheses. The risk of loss of follow-up cannot be excluded, but it is in the author’s opinion that it is unlikely to represent a significant number of patients, given that the study was performed at a referral center for stapes surgery and revision stapes surgery. There might be a selection bias on the choice of prosthesis, because the titanium prostheses were implanted in selected cases. However, this is not a favorable bias for the statistical results and could rather support our results, since the selected cases may have had a higher risk of prosthesis dislocation.

There was a difference in the use of laser for footplate fenestration between Teflon and Titanium groups, but it is not in the author’s opinion that it would impact stability of piston attachment to the LPI. Manual perforator was routinely used till 2015, then diode fiber laser was most regularly used, except in cases where laser was not appropriate (mainly, dehiscent facial nerve) or for academic concerns (residents’ teaching). We also have been using increasingly more Titanium prostheses, leading to this difference. In the opinion of the authors, this would not have been a bias because the laser and non-laser fenestrations are calibrated with the same manual perforator and it would not have changed the stability of the shaft of the prosthesis.

To our knowledge, there is no study comparing stability of prosthesis stability over time regarding the material used. Our study does not address hearing outcomes with regard to material used, since this has been thoroughly described in other studies [6, 24]. We observed 20% moderate and severe incus erosion in our study. In comparison, Lundman et al.’s study had 35% erosion in 227 first revision cases, Blijleven et al. 5% erosion in 63 cases, and Lippy et al. 23.5% erosion in 522 revisions [17,18,19]. The variability of those results could be explained by the lack of clear definition of incus erosion in revision stapedotomy. However, status of the LPI is crucial for deciding the most appropriate surgical technique in revision: standard technique in cases with no or mild erosion, standard technique with surgical cement stabilization in case with moderate erosion, malleostapedotomy in case with severe erosion [25].

Lundman et al. [17] reported a longer median delay from primary to revision surgery (7 years 6 months; min = 3 months; max = 50 years) but our results are not directly comparable since revision patients that had undergone surgery before 2013 were excluded from our study. In our series a triggering event causing the dislocation was found in only 17% of the cases (4/23). In comparison, Puxeddu et al. [26] described 2/44 (4.5%) triggering traumatic events.