Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: Otol Neurotol. 2022 Jul 1;43(6):e605–e612. doi: 10.1097/MAO.0000000000003535

Computed tomography density as a bio-marker for histologic grade of otosclerosis: a human temporal bone pathology study

Alicia M Quesnel 1,2, Reuven Ishai 3, Timothy Meehan 4, Jennifer T O’Malley 1, Renee Mitchell 5, Jennifer J Shin 2,6, Hugh D Curtin 5, Joseph B Nadol Jr 1,2, Michael J McKenna 7, Amy F Juliano 5
PMCID: PMC9245145  NIHMSID: NIHMS1780275  PMID: 35761453

Abstract

Hypothesis:

Computed tomography (CT) density measurement can be used to objectively distinguish otosclerosis from normal bone and to determine histologic grades of otosclerosis.

Background:

Otosclerosis can be seen on CT as subtle radiolucent areas. An objective radiologic measurement that corresponds to known otosclerosis pathology may improve diagnostic accuracy, and could be used as a radiologic biomarker for otosclerosis grade.

Methods:

A blinded, randomized evaluation of both histologic grade on histopathology slides and CT density measurement was performed on 78 human temporal bone specimens (31 with otosclerosis and 47 controls) that had undergone high-resolution multi-detector CT prior to histologic processing. Assessments were performed at 11 regions of interest (ROIs) in the otic capsule for each specimen.

Results:

The CT density measurement mean (Hounsfield Units) ± standard deviation for all ROIs (#1-#9) was 2245 ± 854 for grade 0 (no otosclerosis, n=711), 1896 ±317 for grade 1 (inactive otosclerosis, n=109), and 1632 ±255 for grades 2 and 3 combined (mixed/active otosclerosis, n=35). There was a strong inverse correlation of CT density to histologic grade at ROIs #1–5 (ANOVA, p<0.0001). The inter-rater reliability for CT density was very good (correlation coefficient 0.87, p<0.05). ROC curves suggested a cut-off of 2150 HU to distinguish otosclerosis from normal bone, and 1811 HU to distinguish low grade from mixed/high grade otosclerosis.

Conclusions:

In human temporal bone specimens, CT density may be used to distinguish normal bone from bone involved by otosclerosis. A higher histologic grade (i.e. indicating a more active otosclerotic focus) correlated with lower density.

Introduction:

Otosclerosis is characterized by abnormal bony remodeling of the otic capsule, and is a relatively common cause of progressive, adult-onset hearing loss. It typically results in a conductive hearing loss due to stapedial fixation by a focus of otosclerosis around the oval window1–4. A mixed hearing loss develops in about one third of patients 5, and is often associated with additional peri-cochlear foci of otosclerosis, and exceedingly rarely, otosclerosis presents with a purely sensorineural hearing loss. Although the accuracy for diagnosis of otosclerosis based on otoscopy and audiometry alone is quite high, multi-detector high-resolution computed tomography (CT) of the temporal bone can help confirm the diagnosis, rule out other possible causes of conductive hearing loss, determine the extent of retrofenestral disease, and inform decisions on if/when to consider systemic treatments such as bisphosphonates 6–9.

The radiologic diagnosis of otosclerosis is typically based on identification of radiolucent area(s) on the CT. The area most often involved initially is anterior to the oval window, manifesting as an abnormally hypodense or lucent focus of bone (Figure 1A,B)10–12. Increasingly severe and extensive disease may lead to retrofenestral involvement, including lucency surrounding the cochlea in a “double ring” or “fourth turn” sign as peri-cochlear otosclerosis. Changes in otic capsule contour and increase in thickness of bone in various locations, including stapedial footplate thickening, sclerotic obliteration of the round window niche, and a bulging/convex contour of the otic capsule anterior to the oval window are also radiologic signs of otosclerosis13.

Figure 1.

Figure 1.

Open in a new tab

CT image (A) and corresponding histologic section (B) showing a temporal bone with a focus of otosclerosis anterior to the oval window (arrows). (C) Histologic section demonstrating the multifocal nature of otosclerosis. Within the same temporal bone, there are areas of active otosclerosis (dominated by large pseudovascular spaces), inactive otosclerosis (small and few pseudovascular spaces and lamellar bone), and mixed activity.

Several authors have demonstrated the utility of CT density measurements in the diagnosis of otosclerosis using clinical CTs from operated patients with otosclerosis 14–16; however, pathologic confirmation of disease or histologic activity assessment was not available. Karosi et al. examined patients who underwent total stapedectomy, such that the stapes footplate was available for histologic evaluation. They found that the sensitivity of CT for the diagnosis of otosclerosis based on the identification of hypodense lesions was higher in patients with active otosclerosis in the footplate (76%) compared to those with inactive otosclerosis (62%)17. This suggests that CT density may be able to distinguish different histologic grades or activity levels.

Histologically, active otosclerosis is characterized by large pseudovascular spaces which result from resorption of enchondral bone around blood vessels leading to enlargement of perivascular spaces, as well as high vascularity, high cellularity, osteoclasts, and deposition of woven bone. In contrast, inactive otosclerosis has few pseudovascular spaces (which are small if present), low vascularity, low cellularity, no osteoclasts, and predominantly lamellar bone2,4,18–21. Within a single temporal bone, there are often multiple foci of otosclerosis, which contain areas of active, inactive, and mixed otosclerosis (Figure 1C).

Since the ability to evaluate human otic capsule pathology during life is very limited, defining radiologic-pathologic correlations in otosclerosis is particularly important. For this study, our hypothesis was that CT density measurement could be used to objectively distinguish otosclerosis from normal bone, and, further, to differentiate histologic grades of otosclerosis. We utilized human temporal bone specimens that underwent multi-detector high-resolution temporal bone CT scanning after fixation and prior to decalcification, sectioning, and additional histologic processing. This enabled direct comparison in the same specimen of the histologic grade of otosclerosis to the CT density measurement (in Hounsfield Units) at multiple points around the otic capsule.

Methods:

This study utilized the human temporal bone pathology specimens in the Massachusetts Eye and Ear Otopathology Laboratory, which were donated through the National Temporal Bone Pathology Resource Registry. The study was approved by the Massachusetts General Brigham Internal Review Board (2021P001593). All temporal bone specimens were extracted and fixed, then underwent CT imaging, and then completed standard histologic processing including decalcification, embedment, sectioning, and mounting on slides. Specimens were sectioned into 20 um thick sections, and every 10th section was stained in Hematoxylin and Eosin. The stained sections were used for the histologic analysis (as described below).

Regions of Interest

The locations of the regions of interest for histologic and radiologic analysis were defined as indicated on Figure 2. Briefly, ROI #1 is located posterior to the footplate, and ROI #2 is located anterior to the footplate. ROIs #3–8 mark locations around the cochlea from the lateral aspect of the basal turn to the apex to the medial aspect of the basal turn. ROI #9 is located at the anterior border of the internal auditory canal (IAC), just medial to the cochlear fossette. ROIs #10 and #11 are reference regions, and are located just posterior to the medial basal turn near the cochlear fossette (ROI #10) and posterior to the IAC (ROI #11). The two reference regions were selected because these areas were considered unlikely to be involved with otosclerosis based on prior histopathologic studies, and CT density measurements in these locations could possibly be used for normalization of the other otic capsule measurements. A 1 mm2 region of interest (ROI) circle was used. When the designated 1 mm2 circle extended beyond the otic capsule on histology or CT, then a 0.5 mm2 concentric circle was used. The smaller size was used for some points to avoid overlap with structures beyond the otic capsule on histologic slides or with structures of different density on the CT.

Figure 2.

Figure 2.

Open in a new tab

Maps demonstrating the location of the 9 regions of interest (ROIs 1–9) and 2 reference regions (ROIs 10 and 11) on (A) CT image and (B) histologic section.

Histology

A new, simplified scale for grading otosclerosis histologically was developed specifically for the purpose of this study (Figure 2) based on prior scales, 9,10,12 as there is no uniformly used scale. This scale is based on the percentage of pseudovascular space and vascularity. The scale was designed to capture the essence of active versus inactive otosclerosis, but yet be simple enough to achieve good inter-rater reliability. Two otopathologists (one with 8 years of otopathology experience and one with 3 years of otopathology experience), blinded to the radiology measurements, independently reviewed each selected slide, and recorded a grade for each of the 9 regions of interest (ROIs) and 2 reference regions (Figure 2) based on light microscopy evaluation. When there was disagreement, the slide was reviewed by the two otopathologists in conjunction with a third otopathologist (who had more than 30 years experience), and a consensus grade was determined.

Radiology

High-resolution multi-detector CT scanning was performed on all temporal bone specimens on a 64-slice multidetector CT scanner (Discovery 750 HD; GE Healthcare, Milwaukee, Wisconsin), using standard clinical parameters, including 0.6-mm collimation, 0.625-mm thickness with 0.312 mm overlap, at 240 mA and 120 kV (peak). Scan matrix was 512×512; the FOV varied depending on specimen size, ranging from 55×55 mm2 to 84×84 mm2. Scanner pitch was 0.531:1, speed 10.62 mm/rotation, rotation time 0.5 seconds, exposure time 6–10 seconds depending on specimen size, and CTDIvol approximately 65 mGy. The CT images were reformatted to the exact plane of the selected histologic slide using the Synapse multiplanar reformation (MPR) tool on the radiology picture archiving and communication system (PACS) viewer (Synapse, Fujifilm, Japan), and a single image that best matched the histologic slide was selected. Two radiologists (one with 4 years of general radiology experience and 1 year of neuroradiology experience; one with 8 years of head and neck radiology experience), who were blinded to the pathology results, independently measured the CT density values in Hounsfield Units on the selected CT image at all 9 ROIs and both reference regions (Figure 3). Measurements were performed on the Synapse viewer.

Figure 3.

Figure 3.

Open in a new tab

Otosclerosis Histologic Grading Scale

Statistics

Inter-rater correlation for CT density measurements was analyzed using Pearson correlation. Average CT density measurements between the two raters was used for the comparisons with histologic grades. Student’s t-test was used to determine whether there are statistically significant differences in CT density values between grade 0 (normal) and grade 1 (inactive otosclerosis) at all ROI locations, and between grade 0 (normal) and grade 2 and 3 (mixed and active otosclerosis) at all ROI locations. A one-way ANOVA analysis was performed to determine whether grade 0, grade 1, and grades 2/3 could be distinguished by CT density values. Since there are multiple individual comparisons, the Bonferroni correction was applied, with the adjusted alpha calculated to be 0.0028; therefore, significance was set at p<0.0028, and only comparisons with p<0.0028 were considered significant.

Receiver operating characteristic (ROC) analysis was performed for predicting (1) the diagnosis of otosclerosis, and (2) the grade of otosclerosis, based on the average HU measurements of the 2 CT readers. The optimal operating point or CT density value cut-off (in Hounsfield Units) for each ROC curve was selected by the maximal Youden’s J statistic. Sensitivity and specificity of that optimal operating point was calculated.

Results:

A total of 78 human temporal bone specimens (TBs) were reviewed, including 31 TBs with pathologically proven otosclerosis and 47 TBs without otosclerosis (controls). The mean CT density measurements, grouped by associated histologic grade, are presented for each ROI in Table 1. Overall, when all regions of interest were included, the mean CT density (Hounsfield Units) ± standard deviation was 2245 ± 854 for areas with no otosclerosis (i.e. grade 0 otosclerosis) (n=711), 1896 ±317 for grade 1 otosclerosis (n=109), and 1632 ±255 for grades 2 and 3 combined (n=35). Due to the small number of temporal bones with grades 2 and 3 otosclerosis, these grades were combined for the purpose of this analysis. Areas of grades 2/3 otosclerosis were most commonly found in ROI #2 and #3 (Table 1).

Table 1.

CT densities (in Hounsfield Units) categorized by histologic grade of otosclerosis, for each region of interest (ROI) in the otic capsule.

ROI Otosclerosis Grade on Histology N CT Density Mean ± SD (HU) No otosclerosis vs. Otosclerosis (grade 1 or 2/3) t-test Multivariate Analysis Comparing Otosclerosis Grades 0, 1, and 2/3
1 0 67 2054 ± 301
1 8 1827 ± 318 p=0.048 p=0.0091
2/3 3 1578 ± 486 p=0.01
2 0 48 2198 ± 154
1 20 1697 ± 294 *p<0.0001 *p<0.0001
2/3 10 1506 ± 303 *p<0.0001
3 0 48 2307 ± 98
1 22 2043 ± 198 *p<0.0001 *p<0.0001
2/3 8 1771 ± 355 *p<0.0001
4 0 61 2240 ± 113
1 12 1928 ± 298 *p<0.0001 *p<0.0001
2/3 5 1751 ± 212 *p<0.0001
5 0 72 2461 ± 378
1 3 1978 ± 518 p=0.035 *p<0.0001
2/3 3 1547 ± 55 *p<0.0001
6 0 71 2136 ± 108
1 7 1852 ± 345 *p<0.0001 NA
2/3 0 NA
7 0 68 2287 ± 72
1 7 2071 ± 304 *p<0.0001 NA
2/3 2 1661   NA NA
8 0 67 2171 ± 98
1 20 2105 ± 53 *p=0.005 NA
2/3 2 1858   NA NA
9 0 65 2279 ± 258
1 11 1562 ± 313 *p<0.0001 NA
2/3 2 1645   NA NA
All ROI (1–9) 0 711 2245 ± 854
1 109 1896 ± 317 *p<0.0001 *p<0.0001
2/3 35 1632 ± 255 *p<0.0001
10 (Ref) 0 68 2370 ± 138
1 9 2021 ± 390
2/3 1 1555   NA
11 (Ref) 0 77 2197 ± 99
1 0
2/3 1 1853   NA
Open in a new tab

The Pearson correlation coefficient for CT densities at all ROIs between the two radiologists was 0.87 (p<0.001), consistent with very good inter-rater reproducibility.

There was an inverse correlation of CT density to histologic grade of otosclerosis at all ROIs. This is illustrated at ROI #2 (Figure 4A) and ROI #3 (Figure 4B), which represent the areas of the otic capsule that are most commonly involved with otosclerosis of any grade.

Figure 4.

Figure 4.

Open in a new tab

Histologic grade of otosclerosis is inversely related to CT density. ROI #2 (A) is shown to illustrate the CT density data at the most common site of otosclerosis predilection, anterior to the oval window. ROI#3 (B), adjacent to the anterior portion of the basal turn of the cochlea, demonstrates a similar inverse relationship. Data shown is the averaged CT density measurements from both radiologists, after independent blinded assessment. The mean +/− standard deviation is shown, with all data from each temporal bone plotted as a point (N=78 temporal bones, with 32 otosclerosis temporal bones and 46 control temporal bones).

The mean CT density was significantly different between grade 0 (normal) and grade 1 (inactive otosclerosis) at all ROIs except ROI #1 and #5 (t test, p <0.0001) (Table 1). The mean CT density was also significantly different between grade 0 (normal) and grade 2/3 (mixed and active otosclerosis) for all ROIs except ROI #1 (t test, p <0.0001) (Table 1).

One-way ANOVA analysis was performed to determine whether grade 0, grade 1, and grades 2/3 could be distinguished by CT density values (Table 1). The means of these groups were significantly different at ROI #2 – 5, demonstrating that CT density may be used to distinguish grades of otosclerosis in these locations. It was not significantly different in ROI#1 (posterior to the footplate). The analysis was not performed at ROI 6 – 9 as there was insufficient numbers of specimens with grade 2/3 otosclerosis in these locations.

The means of the two reference regions (ROI #10 and #11) for grade 0 otosclerosis were 2370 and 2197, respectively. These reference regions were selected because these areas were unlikely to be involved with otosclerosis, and could possibly provide values to normalize CT density measurements. However, at both ROIs #10 and #11, there were TBs in which otosclerosis was found. We found that “normalizing” the ROI densities based on (1) dividing by, (2) subtracting, or (3) subtracting the reference value and then dividing by the reference value added more variability and resulted in less correlation with histologic grades when either ROI #10 or ROI #11 reference measurements were used. Therefore, analysis was performed based on the absolute values of the CT densities (in HU) rather than normalized to a region of interest in the otic capsule.

The receiver operating characteristics (ROC) curve for distinguishing between no otosclerosis (grade 0) and otosclerosis (grades 1, 2, and 3) is shown in Figure 5A. The area under the curve (AUC) was 0.8134. If both sensitivity and specificity are considered equally important, then a cut-off CT density value for distinguishing normal bone from otosclerosis of 2150 HU may be used. Using this cut-off value, the sensitivity and specificity of CT detecting otosclerosis was 74% and 73%, respectively. Using a higher CT density value for the cut-off will result in higher sensitivity for diagnosing otosclerosis, but lower specificity. For example, if 2185 HU is used as the cut-off, the sensitivity is 82% but the specificity falls to 63%.

Figure 5.

Figure 5.

Open in a new tab

ROC curves for distinguishing between (A) normal (grade 0) and otosclerosis-involved bone (grades 1/2/3), (B) distinguishing between inactive otosclerosis (grade 1) vs. mixed/active otosclerosis (grades 2/3), and (C) distinguishing between normal (grade 0) and inactive otosclerosis (grade 1) vs. mixed/active otosclerosis (grades 2/3) by CT density values.

The ROC curve for distinguishing between inactive otosclerosis (grade 1) and mixed/active otosclerosis (grades 2 /3) is shown in Figure 5B. The AUC was 0.7125. If sensitivity and specificity are considered equally important, then the optimal CT density cut-off value for predicting high vs. low grade of otosclerosis was 1811 HU. Using this cut-off value, the sensitivity and specificity of CT predicting high vs. low grade of otosclerosis are 67% and 63%, respectively.

The ROC curve for distinguishing between grades 0/1 vs. grades 2/3 is shown in Figure 5C, with an AUC of 0.9152. Using this curve, a cut-off of 2000 HU results in 90% sensitivity and 85% specificity for distinguishing mixed / active otosclerosis from inactive otosclerosis or normal bone.

Discussion:

A direct comparison of CT density to otosclerotic grade was possible in this study due to the utilization of human temporal bone specimens with matched CT images of the specimen formatted to the exact plane of the histologic slide, as well as numerous regions of interest around the otic capsule. This is an advantage compared to studies evaluating CT density in CT scans from live patients, in which the histologic correlate cannot be evaluated. These data demonstrate (1) a significant difference between mean CT density measurements in Hounsfield Units for normal bone compared to otosclerotic bone, and (2) a correlation of decreasing CT density with increasing histologic grade of otosclerosis. Due to small numbers of specimens with grades 2 and 3 otosclerosis, these grades were combined for statistical analyses. The mean CT density measurements were significantly different between controls, grade 1 (inactive) otosclerosis, and grades 2 /3 (mixed and active) otosclerosis in most locations where the sample size was large enough to permit analysis (ROI #2–5). The greater variability at ROI #1 (posterior to the footplate) may be due to the relatively small area of otic capsule bone in that location, so that even a 0.5 mm2 circle may overlap the adjacent middle ear or vestibule, which have contents of different CT density.

In our study, we showed that a cut-off CT density value of 2150 HU can be used as the distinguishing threshold between normal and otosclerosis-involved bone, with a sensitivity and specificity of 74% and 73%, respectively. This cut-off was derived from an ROC curve with area under the curve (AUC) of 0.8134, which generally indicates a test with “excellent” diagnostic accuracy. Furthermore, a cut-off CT density value of 1811 HU can be used as the distinguishing threshold between inactive otosclerosis and mixed/active otosclerosis. Of note, the diagnostic accuracy is less for distinguishing between individual grades of otosclerosis than for the diagnosis of otosclerosis. The AUC for comparison of grade 1 vs. grade 2/3 otosclerosis was 0.7125, which generally indicates a test with “acceptable” diagnostic accuracy.

If the clinical question posed is whether the patient has no otosclerosis/ grade 1 otosclerosis vs. mixed/active otosclerosis (grade 2/3), then the third ROC (Figure 5C) may be used to determine the “best” cut-off based on whether sensitivity or specificity is more important or whether they are considered equally important. This scenario could be a case in which the otolaryngologist recommends treatment with medication (e.g. bisphosphonates (which would be an off-label use) or future potential medical therapies) based in part on whether the patient has advanced (i.e. mixed/active) otosclerosis. In this case, the AUC is “excellent” at 0.9152; a cut-off of 2000 HU can distinguish mixed/active otosclerosis (< 2000 HU) from inactive disease or normal bone (> 2000 HU) with a sensitivity of 90% and specificity of 85%.

The CT density measurements from this study are similar to results obtained by Kawase et al using high-resolution temporal bone CTs performed on clinical patients with otosclerosis22. Kawase et.al performed a receiver operative characteristic analysis and determined a cut-off value of 2187 HU as the threshold between normal and otosclerosis-involved bone in the area anterior to the oval window. This cut-off value is similar to the cut-off value we determined in this study of 2150 HU that equally emphasizes sensitivity and specificity. This suggests that these data can be translated to use in the clinical setting. Unlike the Kawase et al study, the current study has the advantage of having histopathologic assessment of the specimens in order to determine the histologic grade, and correlate that to CT density appearance and values. This has enabled a comparison of radiologic CT density measurements with known histologic grade.

Although our study demonstrated correlation between larger pseudovascular spaces (i.e. higher grade) and lower CT density, a prior study by Sone et al. did not show any correlation between intraoperative blood flow measurement using laser Doppler flowmetry and CT density measurements in otosclerosis patients undergoing stapedectomy23. A prior clinical study on CT density for the diagnosis of otosclerosis demonstrated that the semi-automated CT histogram analysis may be a way to avoid error due to placement of a region of interest24. Several other clinical studies have demonstrated lower CT density measurements, particularly in the area of the fissula antefenestra, in patients with otosclerosis compared to controls25–27. While several prior clinical studies have examined CT density at multiple regions of interest around the cochlea, histologic grades at those locations could be not be assessed for correlation to CT density in live patients.

One limitation of this study was the challenge to exactly match the 0.5 or 1.0 mm2 ROI on the CT with the same 0.5 or 1.0 mm2 area on the histologic section. Exact matching of the area of assessment radiologically and histologically is important because of the multifocal nature of otosclerosis, and the inherent mixture of various grades of otosclerosis throughout even a single otosclerotic focus. We addressed this by reformatting the CT images into a plane matching the plane of the histologic section. Both the radiologists and otopathologists used a reference map of the ROIs (Figure 3) to determine the exact placement of the ROI on the CT and the histologic slide.

Another limitation was the variability in CT density measurements when the perimeter of the ROI approached a structure of a different density. For example, in some specimens the cochlear promontory thickness was less than 1 mm, which did not allow for enough space to place the 1 mm2 region of interest for CT density measurement without including adjacent structures such as air within the middle ear cavity. In an initial pilot study, we found this created significant inter-rater and even intra-rater variability in the CT density measurements. Thus, we revised the methods to include using a smaller concentric 0.5 mm2 circle for the ROI to avoid including obvious unrelated adjacent areas with different density. This revised method resulted in good inter-rater reproducibility in the current study. Radiologists and otologists should be aware of this in clinical practice when utilizing CT density for assessment of otosclerosis.

A third limitation is the relatively low number of active otosclerosis cases in our series. However, this reflects the prevalence of disease and their histologic grades among our collection of human temporal bone specimens. With time, an increase in donated specimens can potentially allow increase in the number of cases in each histologic grade amenable for analyses.

Lastly, further work may be needed prior to translating the CT density measurements for grading of otosclerosis found in this temporal bone study into clinical use in diagnostic CT performed on live patients. There may be differences in bone density of normal bone (as well as otosclerotic bone) when an isolated temporal bone is imaged (as in this study) compared to CT imaging of the whole head in a live patient. We attempted to begin to address this by measuring CT densities at two reference regions on the otic capsule, which were placed in areas where otosclerosis rarely occurs. We analyzed the CT densities normalized to these reference regions in three ways: (1) taking a ratio (dividing the CT density measurement at the ROI in question by the CT density measurement at either reference region ROI #10 or #11), (2) subtracting the CT density at reference region ROI #10 or #11 from the CT density at the ROI in question, and (3) subtracting the reference ROI and then dividing by the reference ROI for #10 or #11. All three ratio and subtraction methods for “normalizing” the CT density values resulted in more variability among the CT density measurements, with larger standard deviations. These “normalization” methods also resulted in a worse correlation coefficient between radiologists. Thus, ultimately we reported the absolute data in this study, rather than using a normalized value based on the reference ROIs. Another way to evaluate the ability to translate these values to diagnostic CTs on live clinical patients may be to analyze cases in which the donor of the temporal bone cadaveric specimen had undergone a high-resolution temporal bone CT during life.

Since pathologic assessment of the extent of otosclerosis is essentially impossible in living patients, a robust radiologic measure of disease activity in otosclerosis may be the best surrogate. This may help to improve clinical care of patients through the ability to study disease progression in response to medical treatments and improve understanding of the natural history of disease progression. Further comparative study could reveal potential radiologic-audiologic correlations that might be associated with predicted pathology. Thus, CT density could become a radiologic biomarker to guide use of medical treatments for otosclerosis, such as bisphosphonates or other future treatments, and may provide an objective measurement to clarify diagnosis and estimate pathologic grade.

Conclusion:

Utilizing postmortem human temporal bone specimens, we demonstrated that CT density measurements (in Hounsfield Units) on a high-resolution multi-detector temporal bone CT scan can be used to objectively distinguish normal bone, which has a higher density, from bone involved by otosclerosis, which has a lower density. We suggest a cutoff threshold of 2150 HU to distinguish between normal bone (2150 HU or greater) and otosclerosis (less than 2150 HU). Furthermore, decreasing density measurements correlated with increasing histologic grade (e.g. indicating a more active otosclerotic focus). A CT density less than 2000 HU was highly sensitive for more active otosclerosis. A radiologic measure of disease activity in otosclerosis may enable more informed application of medical and surgical treatments and assessment of treatment efficacy.

Acknowledgements:

We thank Garyfallia Pagonis for assistance in figure preparation; MengYu Zhu, Diane Jones, and Barbara Burgess for assistance in specimen histologic processing; and Dr. Felipe Santos and Dr. Csilla Haburcakova for assistance in procuring specimens through the NIDCD National Temporal Bone Hearing and Balance Pathology Resource Registry.

Funding: NIH U24DC013983

JJS receives book royalties from Evidence-Based Otolaryngology, Shin JJ, Randolph GW, editors; New York: Springer, 2008 and Otolaryngology Prep and Practice, Shin JJ, Cunningham MJ, editors; Plural Publishing, 2013. JJS is a recipient of funding from American Academy of the Otolaryngology-Head and Neck Surgery Foundation, the Brigham Care Redesign Program Award, the BWell Funds Awards and a Brigham Innovations Award.

Footnotes

Conflicts of Interest:

AMQ had sponsored research agreements and consulting with Frequency Therapeutics, consulting with Alcon Corp., sponsored research agreement and licensing of a patent with Grace Medical, research study participant for Akouos.

MJM is Co-founder and CMO Akouos Inc.

References:

  • 1.Quesnel AM, Ishai R, McKenna MJ. Otosclerosis: Temporal Bone Pathology. Otolaryngol Clin North Am 2018; 51(2): 291–303. [DOI] [PubMed] [Google Scholar]
  • 2.Iyer PV, Gristwood RE. Histopathology of the stapes in otosclerosis. Pathology 1984; 16(1): 30–8. [DOI] [PubMed] [Google Scholar]
  • 3.Altmann F, Woodman D. Otosclerosis; its pathology and treatment. New York state journal of medicine 1948; 48(2): 173–7. [PubMed] [Google Scholar]
  • 4.Schuknecht HF, Barber W. Histologic variants in otosclerosis. The Laryngoscope 1985; 95(11): 1307–17. [DOI] [PubMed] [Google Scholar]
  • 5.Ishai R, Halpin CF, Shin JJ, McKenna MJ, Quesnel AM. Long-term Incidence and Degree of Sensorineural Hearing Loss in Otosclerosis. Otol Neurotol 2016; 37(10): 1489–96. [DOI] [PubMed] [Google Scholar]
  • 6.Virk JS, Singh A, Lingam RK. The role of imaging in the diagnosis and management of otosclerosis. Otology & neurotology : official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology 2013; 34(7): e55–60. [DOI] [PubMed] [Google Scholar]
  • 7.Quesnel AM, Moonis G, Appel J, et al. Correlation of computed tomography with histopathology in otosclerosis. Otology & neurotology : official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology 2013; 34(1): 22–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Zhu MM, Sha Y, Zhuang PY, et al. Relationship between high-resolution computed tomography densitometry and audiometry in otosclerosis. Auris, nasus, larynx 2010; 37(6): 669–75. [DOI] [PubMed] [Google Scholar]
  • 9.Quesnel AM, Seton M, Merchant SN, Halpin C, McKenna MJ. Third-generation bisphosphonates for treatment of sensorineural hearing loss in otosclerosis. Otology & neurotology : official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology 2012; 33(8): 1308–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Juliano AF, Ginat DT, Moonis G. Imaging Review of the Temporal Bone: Part II. Traumatic, Postoperative, and Noninflammatory Nonneoplastic Conditions. Radiology 2015; 276(3): 655–72. [DOI] [PubMed] [Google Scholar]
  • 11.Puac P, Rodriguez A, Lin HC, et al. Cavitary Plaques in Otospongiosis: CT Findings and Clinical Implications. AJNR Am J Neuroradiol 2018; 39(6): 1135–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Yagi C, Morita Y, Takahashi K, et al. Otosclerosis: anatomical distribution of otosclerotic loci analyzed by high-resolution computed tomography. Eur Arch Otorhinolaryngol 2019; 276(5): 1335–40. [DOI] [PubMed] [Google Scholar]
  • 13.Sanghan N, Chansakul T, Kozin ED, Juliano AF, Curtin HD, Reinshagen KL. Retrospective Review of Otic Capsule Contour and Thickness in Patients with Otosclerosis and Individuals with Normal Hearing on CT. AJNR Am J Neuroradiol 2018; 39(12): 2350–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Guneri EA, Ada E, Ceryan K, Guneri A. High-resolution computed tomographic evaluation of the cochlear capsule in otosclerosis: relationship between densitometry and sensorineural hearing loss. The Annals of otology, rhinology, and laryngology 1996; 105(8): 659–64. [DOI] [PubMed] [Google Scholar]
  • 15.Grayeli AB, Yrieix CS, Imauchi Y, Cyna-Gorse F, Ferrary E, Sterkers O. Temporal bone density measurements using CT in otosclerosis. Acta oto-laryngologica 2004; 124(10): 1136–40. [DOI] [PubMed] [Google Scholar]
  • 16.Kawase S, Naganawa S, Sone M, Ikeda M, Ishigaki T. Relationship between CT densitometry with a slice thickness of 0.5 mm and audiometry in otosclerosis. European radiology 2006; 16(6): 1367–73. [DOI] [PubMed] [Google Scholar]
  • 17.Karosi T, Csomor P, Sziklai I. The value of HRCT in stapes fixations corresponding to hearing thresholds and histologic findings. Otology & neurotology : official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology 2012; 33(8): 1300–7. [DOI] [PubMed] [Google Scholar]
  • 18.Csomor P, Liktor B, Szekanecz Z, Sziklai I, Karosi T. Expression of bone morphogenetic protein 2, 4, 5, and 7 correlates with histological activity of otosclerotic foci. Acta oto-laryngologica 2012; 132(6): 624–31. [DOI] [PubMed] [Google Scholar]
  • 19.Karosi T, Csomor P, Szalmas A, Konya J, Petko M, Sziklai I. Osteoprotegerin expression and sensitivity in otosclerosis with different histological activity. European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies 2011; 268(3): 357–65. [DOI] [PubMed] [Google Scholar]
  • 20.Hueb MM, Goycoolea MV, Paparella MM, Oliveira JA. Otosclerosis: the University of Minnesota temporal bone collection. Otolaryngology--head and neck surgery : official journal of American Academy of Otolaryngology-Head and Neck Surgery 1991; 105(3): 396–405. [DOI] [PubMed] [Google Scholar]
  • 21.Linthicum FH, Jr. Histopathology of otosclerosis. Otolaryngologic clinics of North America 1993; 26(3): 335–52. [PubMed] [Google Scholar]
  • 22.Kawase S, Naganawa S, Sone M, Ikeda M, Ishigaki T. Relationship between CT densitometry with a slice thickness of 0.5 mm and audiometry in otosclerosis. European radiology 2006; 16(6): 1367–73. [DOI] [PubMed] [Google Scholar]
  • 23.Sone M, Yoshida T, Otake H, et al. Evaluation of vascular activity in otosclerosis by laser Doppler flowmetry: comparison with computed tomographic densitometry. Otol Neurotol 2013; 34(9): 1559–63. [DOI] [PubMed] [Google Scholar]
  • 24.Yamashita K, Yoshiura T, Hiwatashi A, et al. The radiological diagnosis of fenestral otosclerosis: the utility of histogram analysis using multidetector row CT. Eur Arch Otorhinolaryngol 2014; 271(12): 3277–82. [DOI] [PubMed] [Google Scholar]
  • 25.Kutlar G, Koyuncu M, Elmali M, Basar F, Atmaca S. Are computed tomography and densitometric measurements useful in otosclerosis with mixed hearing loss? A retrospective clinical study. Eur Arch Otorhinolaryngol 2014; 271(9): 2421–5. [DOI] [PubMed] [Google Scholar]
  • 26.Tringali S, Pouget JF, Bertholon P, Dubreuil C, Martin C. Value of temporal bone density measurements in otosclerosis patients with normal-appearing computed tomographic scan. Ann Otol Rhinol Laryngol 2007; 116(3): 195–8. [DOI] [PubMed] [Google Scholar]
  • 27.Viza Puiggros I, Granell Moreno E, Calvo Navarro C, Bohe Rovira M, Orus Dotu C, Quer IAM. Diagnostic utility of labyrinth capsule bone density in the diagnosis of otosclerosis with high resolution tomography. Acta Otorrinolaringol Esp (Engl Ed) 2020; 71(4): 242–8. [DOI] [PubMed] [Google Scholar]

RESOURCES