Radiologically Assessed Stapes Footplate Thickness and Audiologic Outcomes in Patients with Otosclerosis Open Access

Otosclerosis is a progressive bone remodeling disease confined to the otic capsule of the temporal bone. Histopathological studies report an incidence of 2.5%–12%, while clinical manifestation occurs in up to 0.4% of the Caucasian population, with nearly 85% of patients experiencing bilateral involvement [1, 2]. Despite its clinical significance, the exact underlying pathophysiology remains incompletely understood. The diagnosis of otosclerosis remains primarily on clinical evaluation [3]; however, radiological assessment plays a crucial role in the preoperative diagnosis, guiding surgery by identifying anatomical challenges (e.g., an overhanging facial nerve) and notably excluding differential diagnoses of conductive hearing loss, such as superior semicircular canal dehiscence, large vestibular aqueduct or round window atresia [4‒7]. Pathognomonic active otospongiotic foci can be identified by high-resolution computer tomography (HRCT) with a sensitivity and specificity exceeding 90% [8]. Nevertheless, isolated sclerotic foci may present with Hounsfield units similar to regular bone density, rendering them infraradiological and easy to overlook [9‒11]. Visible foci on HRCT and cone beam computer tomography (CBCT) are typically located anterior to the oval window in a region known as the fissula ante fenestram [12, 13]. Despite the integration of imaging as a standard in the preoperative workup, the correlation between radiological foci and perioperative audiometric findings is inconsistent in terms of localization, size, and density [4, 9, 14‒18]. Several radiological grading systems have been proposed to provide normative or predictive value, though none have met consensus [19]. Beyond that, alterations such as round window and pericochlear manifestation, in addition to thickened stapes footplate, have been described in histopathological and radiological studies [4‒20]. Since preoperative radiological assessment of footplate thickness has been confirmed intraoperatively by visualization of footplate alteration with a sensitivity of 85.3% [20] and included in certain grading systems such as Veillon et al. [21], this could serve as a clinical and radiological relevant feature, independently of otospongiotic foci, for perioperative audiometric prediction. Therefore, footplate thickness might affect preoperative hearing by impeded motion and same-wise surgical outcome by elevated resistance to prosthesis movement after stapedotomy [22]. One would also expect footplate opening to be more difficult and potentially cause inner ear damage. So far, only one publication has investigated intraoperative difficulties during footplate penetration regarding radiologically increased footplate thickness, yet missing correlation with audiometric data [23]. To our knowledge, there are no published data on the relationship between radiologically assessed stapes footplate thickness and perioperative audiometric data. We, therefore, aimed to investigate the correlation between radiologically measured footplate thickness in patients with surgically confirmed otosclerosis and variables such as preoperative hearing, audiometric outcome, and surgical challenges, including complication rates.

This retrospective, single-center study was approved by the Ethics Committee of Zurich, Switzerland (BASEC-Nr. 2024-00028). Patients with signed general consent and surgically confirmed otosclerosis during stapedotomy performed by one of three experienced surgeons (A.H., C.R., A.D.) between January 2020 and March 2024 were identified using the ENT statistics database (INNOFORCE©, Ruggell, Lichtenstein) and included in the study. Patients were excluded if no CBCT was present, in the case of revision surgery, concomitant otological diagnosis affecting audiometric measurements (e.g., superior semicircular canal dehiscence), or insufficient audiological follow-up.

CBCT was routinely performed preoperatively using the NewTom VGiEVO (CEFLA S.C., Imola, Italy), 15 × 5 cm HighRes EnhancedScan setting ± boosted dose (8 × 5 cm in single-sided acquisition). Slice thickness ranged from 0.15 to 0.30 mm, with a calculated matrix resolution of 512 × 512 or 1,024 × 1,024 pixels in the axial plane. Two investigators who are either an ENT (J.M.G.) or a neuroradiology (N.N.) specialist analyzed the acquired images independently. The images were exported in “Digital Imaging and Communications in Medicine” format to the open-source Freeware 3D Slicer (Version 5.6.1, Boston, MA, USA) and reconstructed independently by both investigators in the stapes axial plane, according to a standardized protocol, described elsewhere [24]. Footplate thickness was assessed by fitting a line profile perpendicular to the anterior third of the stapes footplate. The distance in millimeters (mm) relative to Hounsfield units was exported as raw data and processed in MATLAB R2022b (MathWorks, Inc., Natick, MA, USA) to identify density peak and perform curve analysis, yielding definitive footplate thickness for statistical calculations, as depicted in Figure 1. A comprehensive subanalysis was conducted on five groups as shown in Table 1, defined according to the measured thickness of the footplate and by referenced literature.

Pre- and postoperative audiometric data were assessed following ISO 8253-1:2010. Data are presented as air-conduction (AC) and bone-conduction (BC) pure-tone averages (PTAs) at 500 Hz, 1 Hz, 2 kHz, and 4 kHz. The air-bone gap (ABG) was calculated as the difference between AC and BC levels. Long-term audiometric improvement was calculated as the difference between the last preoperative and the last audiometric data during the follow-up interval. Longitudinal postoperative audiometric improvement was defined as the improvement of the first to the last postoperative acquired audiogram for PTA of AC, PTA of BC, and ABG, respectively. Audio-vestibular complications, including severe to profound hearing loss were observed until the last consultation, and vertigo up to 6 months of follow-up.

Statistical analysis was conducted using Prism for Windows 64-bit, version 10.2.0 (GraphPad Software, Inc., La Jolla, CA, USA). A paired t-test was used to compare data between matched pairs, while an unpaired t-test was applied to compare values between two independent groups. Simple or multiple linear regression was performed to assess the relationship between one dependent variable and one or more independent variables. The significance level was set to alpha = 0.05. The Bonferroni method was employed for the correction of multiple comparisons.

Sixty-three ears (29 left and 34 right) from 56 patients (31 women and 25 men) with a mean age of 47.8 years (SD 10.9, range 21–71) and surgically confirmed otosclerosis were included. A NiTiBOND 0.6 × 4.5 mm piston (Heinz Kurz GmbH, Dusslingen, Germany) was used in most cases (79%, 48/63). No intraoperative difficulties during footplate perforation were reported. One case showed postoperative inner ear hearing loss (PTA-BC 26.3–37.5 dB HL), resulting in severe combined hearing loss, occurring 40 days after uneventful surgery. Oral steroid therapy was prescribed. Hence, overall postoperative hearing level improved despite the pronounced sensorineural hearing loss. The footplate thickness was 0.79 mm. No further postoperative sensorineural hearing loss occurred. During the inpatient hospital stay, 5 events, and during further follow-up (up to 6 months), 16 events of vertigo were documented. No patient showed pathological findings in clinical head impulse testing or exhibited spontaneous nystagmus. No significant association between footplate thickness of patients with or without vertigo was seen (t(61) = 0.83, p = 0.41), neither when vertigo occurred during admission nor in an outpatient setting (t(19) = 0.87, p = 0.39). CBCT imaging was acquired at close timing to the last preoperative pure-tone audiogram of the initial surgery with a mean interval of 2.3 months (SD 3.8, range 0–14.7). No additional imaging was obtained before contralateral surgery (n = 7). The mean time from the last preoperative pure-tone audiogram to surgery was 2.0 months (SD 2.1, range 0.0–11.7) in unilateral cases and 2.0 months (SD 2.2, range 0.0–5.9) in bilateral cases. The mean postoperative audiometric follow-up after unilateral stapedotomy (n = 56) and bilateral sequential stapedotomy (n = 7) were 13.4 (SD 9.1, range 2.9–37.3) and 6.6 (SD 5.1, range 2.4–13.9) months, respectively, as summarized in Table 2. Considering the last pure-tone audiogram assessed pre- and postoperatively (long-term), the AC and BC PTA as well as the ABG showed a significant (t(62) = 14.52, 5.18, 13.89, respectively, p < 0.0001) improvement. Furthermore, the longitudinal improvement from the first postoperative (2.47 months, SD 1.35) to the last postoperative audiogram showed significant improvement in AC (mean 31.63 dB HL, SD 12.30 to mean 26.29 dB HL, SD 10.24, t(62) = 5.68, p < 0.0001), BC [20.08 dB HL, SD 10.26–18.13 dB HL, SD 20.11, t(62) = 2.88, p = 0.011]) as well as ABG (10,74 dB HL, SD 7.20–7.70 dB HL, SD 5.69, t(60) = 4.12, p = 0.0002) depicted in detail in Table 3. Mean stapes footplate thickness was 0.87 mm (SD 0.25) by investigator 1 (J.M.G.) and 0.84 mm (SD 0.27) by investigator 2 (N.N.), with no significant difference between the two investigators (t(62) = 0.77, p = 0.44) as seen in Figure 2. The mean thickness for both investigators was 0.85 mm (SD 0.21), which was used for further comparison. For detailed analysis, four additional subgroups were defined based on measured footplate thickness, as shown in Table 1. Measured footplate thickness larger than 0.6 mm was considered pathological. No significant correlation with pre- and, first, as well as last, postoperative audiometric data was detected in the overall analysis and within subgroups. Details are shown in online supplement 1 (for all online suppl. material, see https://doi.org/10.1159/000545673). Additionally, no significant correlation was observed between footplate thickness and gender (t(61) = 0.35, p = 0.73), yet with age but without clinical relevance (R² = 0.07, p = 0.03).

Histopathological studies describe footplate thickening as a relevant change in the disease progression of otosclerosis. Healthy temporal bone specimens show a thickness range of 0.15–0.35 mm depending on the location of evaluation [25], owing to their irregular helical shape [26]. Similar values of 0.30 ± 0.11 mm can be visualized by using micro-computed tomography [27]. Consequently, this finding has been integrated into radiological assessments and incorporated into several grading systems of otosclerosis, e.g., Veillon or Rotteveel [28‒30]. Radiologically measured footplate thickening can be correlated with histopathological temporal bone findings and, in accordance with the literature, considered pathological when exceeding 0.6–0.7 mm [28, 31]. The accurate determination of footplate thickness is of critical importance for comparison since it depends on several factors, including the reconstruction plane, the technique employed, and the location of the assessment. Rousset et al. [24] reported significantly different values between the stapes axial plane, used in the present study, and the lateral semicircular canal plane (mean 0.27 vs. 0.42 mm, respectively), when acquired by HRCT in healthy temporal bone. This discrepancy was attributed to the resulting difference in measurement angles relative to the footplate. Furthermore, a comparison of manual caliper analysis conducted by neuroradiologists and peak-width measurements performed on the identical images demonstrated greater values for the latter-named method. The general consensus is that a thickness of greater than 0.5 mm on HRCT is indicative of a pathological condition, in accordance with previously mentioned values. We used 0.6 mm as a cut-off value for pathological FPT, in line with the utilized value in the classification of Veillon et al. [28]. This is due to thickness measurements being conducted by peak-width analysis and in the anterior third of the footplate, which is in contrast to the midpoint calculations in the previous referenced study. Besides, this measurement site is known to be thicker but with the highest interobserver reliability [23, 32]. In addition, given the fact that this region is close to the fissula ante fenestram, which is the most frequent site of pathognomonic changes in histopathology [1, 12] and on HRCT and CBT [3, 4, 10], our measurements were carried out in the anterior third of the stapedial footplate. There was no significant difference in interobserver values in this study (p = 0.44), confirming the reliability of the reported measurements. Akazawa et al. assessed footplate thickness in the lateral semicircular canal plane using UHRCT (0.25 mm slice thickness) to identify potential intraoperative difficulties. In 1 patient (8.3%, 1/12), difficulties during the opening procedure with a footplate thickness of 0.81 mm were noted. This value is below the mean value of FPT observed in this study. In contrast, the utilization of a pulsed laser and a micro drill to achieve definitive perforation, as employed in the present study, was reported to be an uncomplicated process. However, one surgical report mentioned the visualization of a prominently thickened footplate (0.77 mm), and a second one stated a semimobile footplate (0.65 mm). Both findings had no impact on the surgical or postoperative course of treatment. The overall audiometric improvement (AC PTA, BC PTA, and ABG) was significant during long-term follow-up with a maximum interval of 37.3 month, and regarding longitudinal postoperative audiometric observation, as depicted in Table 3. In particular, the mean ABG closure of 7.7 dB (SD 5.69), with 73% (46/63) of the patients measuring 10 dB or less, is considered substantial, related to comparable studies in the current literature [33‒35]. However, no significant correlation was found between FPT and pre-as well as postoperative AC PTA, BC PTA, and ABG in the entire study population. Despite the assumed audiometric and radiologic association, subgroup analysis yielded similar results as shown in online supplement 1. In vitro measurements in healthy temporal bones have indicated that a thickened stapes footplate limits the range of piston movement when exceeding a certain angle between the stapes prosthesis and the footplate. This has a subsequent impact on the sound transmission efficiency [22]. However, this phenomenon was not observed in the present cohort, irrespective of the radiological assessed thickness subgrouping. Another conceivable contributing factor regarding the results of this study might be the reason that not only otosclerotic ears with exclusively thickened footplate have been evaluated, given that this isolated radiologic finding is rarely seen. Besides footplate alteration, additional multifocal foci of the otic capsule are commonly visualized on radiological imaging [12]. In consideration of the heterogeneous and multifocal progressing bone remodeling disease of the entire otic capsule, it appears that the alteration affecting the footplate seems to be an unsatisfactory indicator to serve as an isolated audiological predictor. In our cohort vertigo, without any evidence of spontaneous nystagmus or pathological head impulse testing at any time, and 1 patient with delayed partial inner ear hearing loss in regular FPT (0.79 mm), was described. No further complications occurred. There was no significant association between footplate thickness in patients with or without vertigo (p = 0.41), and whether vertigo occurred during admission or in the outpatient setting (p = 0.39). These results show that stapedotomy is safe and can restore hearing reliably regardless of footplate thickness. As a result, from a surgical point of view, early surgery must not be advocated in cases of otosclerosis. This study is limited by the use of CBCT, despite its high-resolution imaging capabilities. The use of more advanced imaging modalities, such as photon-counting computer tomography, may potentially impact the accuracy of FPT measurements, leading to decreased interobserver variation [36]. To further address this issue, uniform curve analysis was conducted in MATLAB, which contributed to the reduction of further disparity and demonstrated that there was no statistically significant difference between investigators (p = 0.44). It is noteworthy that the potential for selection bias was mitigated by the fact that all CBCT images of qualifying temporal bones could be examined without exclusion due to poor image quality, in contrast to findings reported in the literature [25]. Finally, the retrospective design in a single center and relatively small sample size limit the generalizability of the findings. Besides future studies with larger cohorts, prospects may include previously mentioned photon-counting computer tomography imaging, providing more accurate structural depictions at the same time as reducing radiation exposure. This could facilitate the visualization of infraradiological otosclerotic foci or indirect manifestations, such as annual ligament impingement or a more precise depiction of stapes structure, in addition to established radiological parameters [36]. This is has already been demonstrated in vitro by imaging techniques such as micro-computed tomography or Synchrotron imaging [11, 27, 37].

Stapedotomy represents a viable treatment option for otosclerosis, offering a low-risk profile and significant audiometric improvement irrespective of footplate thickness. There is no evidence suggesting that increased footplate thickness elevates the risk of postoperative severe to profound hearing loss or vertigo. Accordingly, from a surgical standpoint, the timing of surgical intervention can be guided by audiological and patient-specific factors, rather than being advocated early.

The author would like to express gratitude to the research department and the clinic management of the Department of Otorhinolaryngology, Head and Neck Surgery at the University Hospital Zurich for supporting with protected research time to conduct this study and for their invaluable support in making this project a success. Furthermore, the author would also like to thank Mrs. C. Studer for her assistance with the radiological data export.

This retrospective, single-center study was approved by the Ethics Committee of Zurich, Switzerland (BASEC-Nr. 2024-00028). Only patients with prior signed general consent were included in this study. All procedures were in accordance with the World Medical Association Declaration of Helsinki (2013).

The authors declare that there is no conflict of interest regarding the publication of this article.

This study was not supported by any sponsor or funder.

J.M.-G. was the main author and initiator of the manuscript. A.D., A.H., and C.R. are the otologic surgeons who contributed to the writing and revising of the manuscript. A.D. further contributed to conceptualization and supervising. N.N. is the neuroradiologist and second radiologic investigator, further contributing to the manuscript by writing. R.B. contributed to statistical analysis, developing figures, and manuscript writing. D.B. and F.P. contributed to the data interpretation and manuscript.

All data generated or analyzed during this study are included in this article and its online supplementary material files. Further inquiries can be directed to the corresponding author.