Here, for the first time, we report the analysis of the volume, porosity, and bone density distribution within the ossicular chain of fresh-frozen human temporal bones.
We have found that the degree of porosity and bone density within the same area appears to have a negative correlation. The incus emerges as the most spongy ossicle and the least mineralized. In addition to the similar pattern of high-density regions along areas of low force, we also found areas of high density at the attachment of the stapedius muscle and the stapes footplate in all samples. In contrast to the malleus and the incus, the stapes is a hollow ossicle, which explains the higher degree of overall porosity.
Our observations regarding the volume analysis lay, on average, within the standard deviation of previously reported values23 (see Table 1). Despite the small sample size, we observed a high anatomical variability in our study. While it is well established that the malleus is the largest of the auditory ossicles and the stapes the smallest, the relative volumes of these ossicles (= total volume of the ossicle – void space within the ossicle) do not always correlate directly with their sizes (see Table 1). Factors such as structural hollowness and the specific anatomical features of each ossicle can result in variations in volume that do not align with their overall dimensions. Consequently, it is important to correlate the geometrical dimensions of the ossicles and the porosity, as is done in this study.
The porosity of the auditory ossicles lies, on average, between 1.92% and 9.85%. In comparison, cortical bone has a porosity of 5–15%, whereas the porosity of trabecular bone ranges from 40–95%25. A lower degree of porosity was expected from the auditory ossicles, as they consist almost entirely of cortical bone, but no remodeling occurs.
Furthermore, we can note that areas with lower bone density correlate with areas with a higher degree of sponginess. In these areas, we also do not see any vascular channels. While we can define a vascular channel well in Sample 1 along the long process of the incus, this is not possible in Samples 2 and 3 (see Fig. 3). A possible explanation could be that the low or absent blood supply leads to these highly porous areas. Bone remodeling can only occur properly if the growth factors and nutrients reach a particular region. If the blood vessels cannot reach a particular area anymore, the remodeling process is disrupted, leading to the erosion of this area. Previous reports have stated that the long process of the incus is indeed the most susceptible area to bone erosion26–29. For the volume and porosity analysis, we segmented the ossicles first and performed the calculations based on the label fields. Accordingly, the segmentation accuracy determines the accuracy of the volume and porosity calculations. The isotropic voxel size of 2.75 µm is the highest level of detail reported so far to define the volume of human auditory ossicles. It determines the limit of differentiable grey levels and, therefore, the accuracy of the calculations. Hence, we are confident that we can discriminate bone from cavities well and that our values are reliable. Further, we found a difference in the overall bone density between Sample 1 and the other samples. However, the bone density distributions within
the three analyzed samples show very similar patterns. Furthermore, we can not know how significant the differences would be if we looked at the absolute values for the bone densities. Therefore, we can not draw meaningful conclusions about the inter-sample variance. However, we believe that the similar pattern observed in all samples is more meaningful.
To our knowledge, Morris et al.4 has been the only one who reported a bone density analysis for human samples. However, the samples were Thiel-fixed. Unger et al.30 and McDougall et al.31 reported degradation and fragmentation of cells and significant change in biomechanical properties of the tissue that underwent Thiel-fixation. The Thiel-fixation could, therefore, affect bone mineralization as well. Nevertheless, Morris et al.4 hypothesized that similar to long bone, bone remodeling in the auditory ossicles occurs where many biomechanical forces are applied. He, therefore, suggested that the incudomalleolar joint (IMJ), which facilitates just biaxial motion32 and, therefore, is not exposed to many forces, should have a higher bone mineral density compared to the incudostapedial joint (ISJ), which allows multiaxial motion33, and therefore exposed to more forces. Our findings only partially support his hypothesis. The IMJ shows a relatively high bone density for sample 1. However, the ISJ of the same sample showed an even higher bone mineral density. For samples 2 and 3, we could also see that the IMJ is highly mineralized. However, we could not see any considerable differences compared to the ISJ. However, further findings are in line with his suggestions that less remodeling happens at sites of lower forces, which shows in higher mineralization patterns at the periphery of the malleus and the incus, around the incudomalleolar joint, and along the lateral side of the long process.
In addition, all samples showed high bone density spots at the stapes footplate, between 140–150 for sample 2 and up to 170 for sample 1 and 3. This agrees to some extent with Morris et al.4, who stated that the stapes footplate is the densest part of the ossicular chain.
We also saw that the stapedius muscle’s attachment site at the stapes’ neck and the stapes and the anterior crus show very high bone mineral density, as seen in Fig. 7. For sample 1, the attachment site shows the highest mineralization of the three samples, which aligns with the overall mineralization pattern of that sample. Sample 2 shows the lowest bone density at the attachment site. However, the overall density of stapes 2 is lower than the other two samples. Therefore, the area of the attachment site differentiates from the rest of the stapes, with a much higher density in the range of 140. Hill et al.8, who investigated long bones, suggested that bone remodeling should not occur at sites of ligamentous attachment in order to provide stability. Thus, these sites should have higher bone mineralization. Therefore, our findings concerning the stapedius muscle’s attachment site support the findings of Hill et al.8.
In summary, we analyzed the bone density distribution in the ossicular chain of fresh-frozen human temporal bones for the first time. Bone density distribution plays a crucial role in the ossicles. It is essential for mass distribution, as the ossicles are suspended in an air-filled cavity. Therefore, the overall density also influences how well sound is transmitted. Our findings showed a higher density along the periphery of the ossicles, which may help to maintain mass balance. Further, we suggest that the high density of the stapes footplate ensures proper coupling to the oval window and, hence, proper sound transmission to the inner ear. In addition, we analyzed the volume and porosity of the ossicles. We found a negative correlation between holey areas within the ossicles and the density within that area. Further, we saw a similar high bone density pattern along areas exposed to lower forces, such as the periphery of the malleus and incus, along the IMJ, and the long process of the incus. In addition, the anterior crus and the parts of the stapes footplate also showed high bone density.