Intraoperative quantification of floating mass transducer coupling quality in active middle ear implants: a multicenter study

Abstract

Purpose: Evaluating the effectiveness of intraoperative auditory brainstem responses (ABRs) to stimulation by the Vibrant Soundbridge (VSB) active middle ear implant for quantifying the implant's floating mass transducer (FMT) coupling quality.

Methods: In a diagnostic multicentric study, patients (> 18 years) who received a VSB with different coupling modalities were included. Pre- and postoperative bone conduction thresholds, intraoperative VSB-evoked ABR thresholds (VSB-ABR) using a modified audio processor programmed to preoperative bone conduction thresholds, postoperative vibrogram thresholds, and postoperative VSB-ABR thresholds were measured. Coupling quality was calculated from the difference between the pure tone average at 1000, 2000, and 4000 Hz (3PTA) vibrogram and postoperative 3PTA bone conduction thresholds.

Results: Twenty-three patients (13 males, 10 females, mean age 56.6 (± 12.5) years) were included in the study. Intraoperative VSB-ABR response thresholds could be obtained in all except one patient where the threshold was > 30 dB nHL. Postoperatively, an insufficient coupling of 36.7 dB was confirmed in this patient. In a Bland-Altman analysis of the intraoperative VSB-ABRs and coupling quality, the limits of agreement exceeded ± 10 dB, i.e., the maximum allowed difference considered as not clinically important but the variation was within the general precision of auditory brainstem responses to predict behavioral thresholds. Five outliers were identified. In two patients, the postoperative VSB-ABR thresholds were in agreement with the coupling quality, indicating a change of coupling before the postoperative testing.

Conclusion: The response thresholds recorded in this set-up have the potential to predict the VSB coupling quality and optimize postoperative audiological results.

Keywords: Active middle ear implant; Coupling quality; Floating mass transducer; Intraoperative; Objective measures.

Fig. 1

Output level of the AP404 with the insert earphone sound tube attached to its microphone for increasing ABR stimulus intensities. The input was the broadband CE-Chirp. The AP404 was programmed to arbitrary BC thresholds of 0, 20, and 40 dB HL. The dashed lines illustrate linear increase with a slope of 1 dB re. 1 µV/1 dB nHL. The dotted area represents the noise floor

Fig. 2

Stimulus transmission of the AP404 with the insert earphone sound tube attached to its microphone. a Temporal waveforms of 500, 1000, 2000, and 4000 Hz tone bursts. The back line shows the original signal, the gray line shows the signal at the FMT. b Frequency-specific delay time function for transmitted tone bursts. c Transmission of the broadband CE-Chirp emerging as a click-like stimulus at the FMT

Fig. 3

Comparison of pre- and postoperative BC thresholds. Overlapping data points are displayed larger. The diagonal lines are the lines of equal thresholds as well as the ± 10 dB limits

Fig. 4

Frequency-specific coupling qualities of the included patients as shown by the VIB thresholds plotted against the postoperative BC thresholds. Overlapping data points are displayed larger. The diagonal lines are the lines of equal thresholds, i.e., perfect coupling quality

Fig. 5

Analysis of the intraoperative VSB-ABR thresholds in relation to the 3PTA coupling qualities of all patients. Overlapping data points are displayed larger. For the patient where the intraoperative VSB-ABR measurement did not show a response, the data point is marked by an arrow (assumed threshold > 30 dB nHL). a Scatterplot of 3PTA coupling qualities and intraoperative VSB-ABR thresholds. The diagonal lines are the line of equal thresholds as well as the ± 10 dB limits. b Bland–Altman analysis plotting the differences between VIB and BC thresholds (3PTA coupling qualities) and the intraoperative VSB-ABR thresholds against the averages of these two measures. The solid horizontal lines show the mean of the differences and the limits of agreement (mean difference ± 1.96 standard deviations). The dotted lines show the ± 10 dB which we allowed as a maximum acceptable (i.e., clinically not relevant) difference between the two methods. Limits of agreement exceeding our tolerance level (± 10 dB) indicated disagreement between the two measurements