**3. Results**

#### *3.1. Audiological Outcomes*

The median words recognition test score was better after reimplantation than before: 78% [47–90%] versus 85% [65–92%] for the best score 3 years after reimplantation (median, 1st and 3rd quartile, Wilcoxon's rank test for paired data, *p* = 0.006). The performances improved by over 10% in 46% (*n* = 23) of children, were similar (an increase or a decrease of less than 10% in scores between the implantation and the reimplantation) in 38% (*n* = 20), and showed a deterioration (decrease of more than 10%) in 16% (*n* = 7).

#### *3.2. Factors Associated with Audiological Performance*

We did not observe a statistically significant difference in the audiological outcomes regarding sex, etiology of deafness, or indication of reimplantation (Table 2). However, adherence to the speech rehabilitation program after the reimplantation was statistically associated with better audiological outcomes in the 3 years after reimplantation.


**Table 2.** Percentage decrease or increase in word discrimination scores depending on sex, etiology, indication of reimplantation, and adherence to speech rehabilitation program after cochlear reimplantation (median and 1st and 3rd quartile).

The scores before reimplantation were correlated with the scores after reimplantation, and followed an exponential non-linear curve (Figure 1a, correlation of fit: 0.685). Indeed, the patients with low scores before reimplantation presented a greater gain than the patients with high scores. Conversely, the patients with high scores tended to have

stable audiological performance after reimplantation. However, the angle of reinsertion (Figure 1b), the age at reimplantation (Figure 1c), and the time since the initial implantation (Figure 1d) were not statistically correlated with better audiological outcomes.

**Figure 1.** Correlation between the percentage increase or decrease in word discrimination and different factors: (**a**) Patients with low scores before reimplantation tend to have significantly increased scores in the 3 years after reimplantation, whereas patients with high scores tend to maintain audiological performance. The angle of insertion of the electrode array (**b**), the age at reimplantation (**c**), and the time since first implantation (**d**) were not correlated with the scores after reimplantation. Each patient cross represents a patient. Dotted lines: decrease or increase of 10% in word discrimination; blue line: simple linear regression; grey area: 95% confidence interval; R: Spearman coefficient of correlation.
