Marchioni et al. [16] stated that the subtympanic sinus might be, as often as sinus tympani, the place of residual cholesteatoma. Through years and consecutive papers, he classified almost all structures vital for endoscopic ear surgery. Marchioni also emphasized that sinus depth and its relations to the facial nerve are essential factors when planning the removal of inflammatory tissues. When a surgeon intends to perform an endoscopic cholesteatoma excision, one should gather as much information as possible. Computed tomography is the perfect tool to collect such data, and we support the opinion that it is obligatory to interpret CT scans before cholesteatoma surgery. In children, the preoperative assessment is also important before cochlear implantation.

Endoscopic visualization of the fundus of the type C tympanic sinus, facial sinus, and subtympanic sinus is very difficult [9]. The deeper the medial sinuses are, the greater the risk of injuring the facial nerve [11, 17]. Until now, it was proven the type C configuration of the tympanic sinus [17, 31], facial sinus [1, 32] and subtympanic [3, 14] is the least common in healthy adults. Also, in healthy children, type C tympanic sinus [30] and facial sinus [32] are the least common. Our work is the first to investigate subtympanic sinus regarding its types in such numbers and children. Our findings are a bit different from the distribution reported by other researchers (Table 3.), probably due to the sample size. As the overall distribution is less favorable for type A (shallow) STS, it is interesting how the percentage of type A STS grows in a group of older children (from 59 to 72%). The distribution of particular types in group of older children is quite similar as in the groups of healthy ears investigated by Anschuetz et al. [3] and Geneci and Ocak [14]. We believe it may be due to the development of pneumatization routes around the facial nerve and the relative position change of the mastoid part of the facial nerve and the otic capsule. The interesting fact is the distribution of types in diseased ears in the group analyzed by Hool et al. [15] is similar to our group of children and adult groups [3, 14]. There is a need for research on healthy and diseased adults on a larger sample size to confirm these findings.

Anschuetz et al. [3] proposed their new classification based on radiological classifications for the tympanic sinus [17] and facial sinus [1]. Geneci and Ocak [14] put in doubt the necessity of having the classification for the STS. They state that most STS do not exceed the level of facial nerve and therefore encourage one to put an effort into further studies. On the other hand, we found statistically significant differences in depth between types of subtympanic sinuses. These results support the qualitative categorization of STS into types A, B, and C.

Only one paper provides the depth measurements written by Geneci and Ocak [14], who analyzed micro CT scans. They described the depth measurement as distance from the STS’s medial border and the FN’s mastoid part. The exact points chosen for measurements are not defined, so the comparison is limited. The means and ranges are lower in their group (Table 4.), and it may be because they used temporal bones of adults after completion of the pneumatization process in healthy conditions and stable spatial relationship between the facial nerve, otic capsule, and air cell tracts around.

As there is no information in the literature on the topic of STS width, we cannot compare it with other research. It seems the STS has a broader entrance (width) than sinus tympani in children [20, 30] and adults [31]. Facial sinus entrance is generally wider [7, 32]. However, its access is not so difficult because its lateral position about the facial nerve and scutum may always be removed. The shallow STS is more accessible than the sinus tympani as it is anterior and has a wider entrance, but both ST and STS type C may be inaccessible endoscopically, no matter the width of the entrance. That is why the aural surgeon must be able to convert from an endoscopic to a microscopic approach. A microscopic alternative for STS is a retrofacial approach via the transmastoid route, as with deep tympanic sinuses [3].

Hool et al. [15] state that in the case of hypopneumatization of the mastoid, the recesses of the retrotympanum may be shallow (type A). On the other hand, Baklaci et al. [4] found that well-pneumatized mastoid in adults was highly related with a deep and posteriorly positioned ST with respect to FN. As there is no consensus on the classification system of pneumatisation in children, we used Allam’s [2] universal proposition. We found the presence of both retrofacial and hypotympanic air cell tracts increased in the group of older children. That proves the ongoing development of pneumatisation with age. However, no statistically significant correlation was demonstrated between the depth and the presence of these air cell tracts. Hool et al. [15] also compared available data on recesses of retrotympanum by adding all sinuses according to type, i.e., type A ST, FS, and STS, and so on, and they presented percentage values in relation to the summed population from four papers. As we know from extensive group studies on adults and children, the particular sinuses are not characterized by normal distribution. Even though we analyze the same factors as depth and relationship to the mastoid facial nerve, the percentage distributions vary. We recommend assessing each sinus separately and not predicting the type of one sinus based on the other. Lack of information in these regards leads to the idea of proposing radiomorphologic profiles using depth, width, and type of particular sinuses as proposed in the model of nonsyndromic sagittal craniosynostosis assessment [26]. There may be a pattern of particular coexistence of retrotympanic recesses. In further studies, we would like to check the configuration and coexistence of the tympanic sinus, facial sinus, subtympanic sinus, and subcochlear canaliculus at the same time. Four significant recesses are essential for surgical planning, and each has three possible configurations, so we assume a square matrix 4 × 4. In each matrix window, there are 3 options. We want to check whether there are any dominant patterns, including all recesses simultaneously in the same subject.

The inferior wall of the tympanic cavity is also called the jugular wall because of its close relationship to the jugular bulb. Many researchers appreciated the role of the jugular bulb in establishing the final shape of the transitional area between retro- and hypotympanum [8, 11,12,13, 16]. There are quite a few distinct classifications of high-riding jugular bulbs, but one of the possible variants is JB protruding to the tympanic cavity. It may occur, mainly when the domination of JB is observed. Savic and Djeric [23] found the dome of the JB protruding into the hypotympanum in about 25% of examined bones. The cavity of the hypotympanum is significantly reduced or even missing in such a situation. In our group, during ongoing pneumatization, we have found JB modeling the STS in up to 22% of older children with no side correlation. It is similar to the results of Savic and Djeric in terms of the general interpretation of the role of JB in this area. Further investigations are needed to check whether there is a side correlation after the JB establishes its final shape in adults.

The radiologic anatomy of the retrotympanum is very difficult to analyze. Burd et al. [10] highlighted that there is a need for close cooperation between otologists and radiologists. We agree that with this remark, there should be consensus meetings. We think radiologists would benefit from operating room visits, operative video explanations, and participation in cadaver labs. On the other hand, otologists would appreciate more hours with image interpretation and three-dimensional reconstructions. One may be advised because the current study from Beckmann et al. [6] suggests an overestimation of cholesteatoma in CT scans compared to intraoperative findings. It is yet another cause for the thorough and long-lasting study of microscopic, endoscopic, and radiological anatomy.

Our next goal is to correlate these findings with specimens or intraoperative endoscopic ear surgery videos to cover healthy and diseased temporal bones. Further studies in CT scans in adults would also be desired to check the distribution of particular STS types after the pneumatisation development is completed. In contrast to the clinical CT examination it would also be beneficial to consider research that examine the morphology of the STS using micro-CT, as this technique offers a huge potential for detailed 2D and 3D imaging of the small compartments in the tympanic cavity [27]. However it is limited to the examination of dry bone samples.