Morphological analysis of the vestibular aqueduct by computerized tomography images

1. Introduction 

The vestibular aqueduct has been studied since the 17th century. The first classical description of its identity was performed by Cotugno [1], which named the structure as “Aqueduct of Cotunnius”. Years later, the first structure enlargement description was reported by Mondini [2] in samples of dissected temporal bones. Bast and Anson [3], showed the vestibular aqueduct morphology and its topographic relation to other external ear structures. Clemis and Valvassori [4], first studied radiographic images of the structure describing its abnormalities. Ten years later, the same authors reported the first clinical description of a congenital disease characterized by a large vestibular aqueduct, visualized in a computerized tomography image.

The vestibular aqueduct is an osseous canal of 10mm of length, localized in the posterior segment of the medial wall of the petrous portion of the temporal bone. It has an orifice that extends to the posterior surface of the structure and it is divided in two segments: (a) the proximal or isthmus, which is narrowed at the internal opening and has 1.5mm of length and 0.3mm diameter; it is located at the anterior medial wall of the vestibule, and (b) the distal, which has a triangular form with its apex joined to the isthmus and enlarges to end at the base with the external opening between 0.5 and 5mm diameter; its morphology may be compared to an inverted “J”, its normal diameter is between 0.4 and 1.0mm, allowing the passage of a small vein that contains a tubular prolongation of the membranous labyrinths and the endolymphatic duct that ends as a sac shape in the cranium cavity, between the dura mater layers [4]. Recent experiences suggest that the enlargement of the vestibular aqueduct (anteroposterior diameter greater than 1.5mm between the external opening and the common crus) is the inner ear abnormality most commonly detected by radiography, and it can be isolated or associated with other malformations of the inner ear and result in hearing loss.

Many clinical and experimental studies analyze the anatomic and physiologic aspects of the vestibular aqueduct in humans. However, recent otology diagnosis techniques provide a summary of studies of the human temporal bone in order to improve the anatomy knowledge due to pathologies and structural variations that may occur.

The present study aims to identify and characterize the morphology of the human vestibular aqueduct by analysis of computerized tomography images. It also comprehends an investigation of the morphometry in order to define its dimensions through a computerized image processing system.

2. Material and method 

The analyzed patients (n=110, aging between 1 and 92 years) were submitted to computerized tomography equipments (Philips Secura™ e Philips Tomoscan AV™) to acquire images from their temporal bone. The results were grouped according to gender and age, where 28 adults and 26 children were males and 32 adults and 24 children were females. Those aging from 19 to 92 years were considered adults and those aging from 1 to 14 years were considered children. The patients were assisted by the Image Diagnosis Department of the Federal University of São Paulo, Paulista School of Medicine. The Wilcoxon and Mann–Whitney non-parametric statistical tests were used to compare the groups of patients. The level of significance adopted was less than or equal to 5%. The inclusion criteria comprehend normal patients in regard to the vestibular aqueduct anatomy, and by extension, normal regarding the osseous labyrinth and the entire inner ear. Individuals who did not present previous diseases of the ear and vestibular–cochlear system and conductive or neurosensory deafness were also included.

The vestibular aqueduct images were classified according to morphology: funnel-shaped, tubular and filiform.

The images were acquired using the following scan parameters: a high-resolution algorithm, with parameters of 120kVp, 180mAs, exposition time of 1.00s and matrix of 512×512. The matrix was used to calibrate all acquired images in order to produce a same pattern of size and resolution. The acquisition protocol used images in the axial plane with patients facing down in the plane parallel to the orbitomeatal line, using a specific head support for this position; slicing images of 1.0/1.0mm of thickness/increment.

The selected morphometric parameters were area, distance from the vestibular aqueduct to the internal acoustic meatus, the length of the descendent segment and the vertical diameter with width of the external opening of the vestibular aqueduct (Fig. 1). The program used for this morphometric analysis was the Computerized Image Processing System (SPCIM), which was developed by the Universidade de São Paulo Institute of Physics.

View Large Image Download to PowerPoint

Fig. 1. (a) Computerized tomography images of the temporal bone in the axial plane showing the vestibular aqueduct (arrow). (b) Vestibular aqueduct (VA) diagram showing the reference points for the morphometric analysis: (1) area; (2) vertical diameter of the external opening width; (3) length of the descendent segment; (4) distance from the vestibular aqueduct to the internal acoustic meatus (IAM).

For the program to automatically identify the morphometric parameter values, an image of interest was selected and a demarcation was done over it. The program identifies the number of pixels (i=i+1) that are inside the demarcation and calculate it on basis of the matrix calibration to give the morphometric parameters in units of mm2 and mm.

This work was approved by the Ethics Committee of the Federal University of São Paulo, Paulista School of Medicine.

3. Results 

The morphology of the vestibular aqueduct is variable (Fig. 2). In adults, we found the filiform form to be the most prevalent (53.3%) followed by the tubular (25%) and funnel-shaped (21.7%) forms. In children, the funnel-shaped (44%) form was the most prevalent and was followed by the filiform (34%) and tubular (22%) forms (Fig. 3).

View Large Image Download to PowerPoint

Fig. 2. Vestibular aqueduct forms observed from computerized tomography image.

View Large Image Download to PowerPoint

Fig. 3. Total frequency of the vestibular aqueduct in relation to form of 100 (n=50 children) and 120 (n=60 adults) temporal bones of both genders visualized in computerized tomography.

The morphometric parameters of the vestibular aqueduct, in children, showed the following means: 4.86mm2 of area, 2.24mm of the external opening, 4.73mm of length and 11.88mm of distance from the vestibular aqueduct to the internal acoustic meatus (Table 1). The comparison of the morphometric parameters between groups of both genders and sides showed that the differences were not statistically significant.

Table 1. Total mean of area, external opening, length and distance of the vestibular aqueduct (VA) to the internal acoustic meatus (IAM) acquired from computerized tomography images of right and left ears, in both sexes of children and adults

	Groups	Area VA (mm2)	Opening external VA (mm)	Length VA (mm)	Distance VA/IAM (mm)
	Children's N	50	50	50	50
	Mean total	4.86	2.24	4.73	11.88
	Adults N	60	60	60	60
	Mean total	4.93	2.09	4.44	11.35

In adults, the morphometric parameters showed the following means: 4.93mm2 of area, 2.09mm of the external opening, 4.44mm of length and 11.35mm of distance between the vestibular aqueduct and the internal acoustic meatus (Table 1). The comparison of the morphometric parameters between groups of both genders showed that the differences were statistically significant, for the length and distance parameters when compared between the groups of genders. The comparison of other morphometric parameters between groups of both genders and sides showed that the differences were not statistically significant.

4. Discussion 

In the vast amount of literature analyzed, spanning the last century to present, the vestibular aqueduct was generically described in the dissected temporal bone, showing high variability in morphology, position and size.

The vestibular aqueduct descriptions were also described by means of computerized tomography images. By the use of radiologic method, Valvassori and Clemis [7], found that the axial plane was the best projection to describe the structure in filiform morphology. In 3000 analyzed cases, these authors found that the vestibular aqueduct morphology was normal in 57% and filiform in 17% of the total analyzed cases. In the present study, the axial plane was also adopted and brought a considerable frequency of filiform feature to the vestibular aqueduct morphology (53% of filiform) [5]. This content gives support to adopt the axial plane as the best projection to the filiform morphology description.

The first radiological description of the large vestibular aqueduct syndrome was done by Valvassori and Clemis [5] who analysed 50 patients (56% females and 44% males; 50% were children and adolescents) and found that the vestibular aqueduct was greater than 1.5mm of diameter. Considering the both genders, our findings are not in agreement with this description [7]. Uma possível explicação poderia ser a miscigenação racial da casuística estudada. Currently, the Brazilian population has a genetic pool resultant from the mix of European, African and Oriental races that occurred throughout the centuries of occupation of its territory. As it is known, this anatomical characteristic is determined genetically from a polygenic mechanism that may have a combination that determines higher normal values for our population.

Takeda et al. [6] provided a description of the vestibular aqueduct morphology in 30 ears by analyzing computerized tomography images and found 67% of funnel-shaped, 30% of tubular and 3% of filiform. We verified that the funnel-shaped morphology is prevalent (44%) in children only. In adults the prevalence was of the filiform form (53%).

Pappas et al. [8] observed that in 29 patients, with large VA syndrome, aging between 4 months and 36 years, the external opening of the vestibular aqueduct presented 0.95mm of width. Sennaroglu et al. [9] observed that in 40 normal patients, aging between 19 and 60 years, the external opening of the vestibular aqueduct presented 1.5mm of width. Despite the criteria used, the mean width of the external opening of the vestibular aqueduct presented in this study is not very close to that quoted literature. The same hypotheses of racial miscegenation and genetic composition of the studied population mentioned earlier could explain the discrepancies between our findings and those of literature.

The analyzed literature data cannot be statistically compared to our data since the statistical tests in literature were other than the ones we used—the Wilcoxon and Mann–Whitney non-parametric statistical tests.

The following authors provided morphometric parameters that are similar to the present study and our findings approach these authors’ findings in relation to the external opening of the vestibular aqueduct.

Lai and Shiao [15] observed in 12 patients (aging between 2 and 14 years) with large vestibular aqueduct syndrome, an external orifice whose mean diameter is of about 3.33mm. Pediatric patients (n=1200; aging between 6 months and 23 years) were found to present the external opening of the vestibular aqueduct with a mean diameter of 2.2mm (0.6–6.5mm, variability) [14].

Okumura et al. [11], observed that in 181 patients (aging between 11 months and 75 years) the external opening of the vestibular aqueduct was found to be less than 4.0mm in 88%. Different from the last authors, Yang et al. [12], showed a high mean width (7.5mm) of the external opening of the vestibular aqueduct in 95 patients (aging between 1 and 39 years) and Chun-Yu et al. descriptions [13] also showed a high mean width of 7.23mm in right ears and of 6.83mm in left ears (in children of 2 years old). However, our findings showed lower values of the mean width of the external opening of the vestibular aqueduct (2.09mm, in adults; and 2.24mm, in children), being closer to the findings described by Okumura et al. [11].

Reference values for the vestibular aqueduct were provided by analysis of the computerized tomography images of 67 normal patients (aging between 18 and 83 years) and showed a variability of 0.28–7.29mm of length, 0.07–1.86mm of opening width; and gives the mean of the reference values that support our findings, which showed the mean value to be about 4.44mm of length and 2.09mm of width (both in adults) [10].

In 250 samples of temporal dissected bones, 90% measured between 2.0 and 6.0mm of distance between the external openings of the vestibular aqueduct to the internal acoustic meatus. Similar samples (n=600) were analyzed and showed that the distance of the external opening of the vestibular aqueduct in relation to the internal acoustic meatus was between 5.0 and 17mm (mean=10mm), and in 95% the external opening of the vestibular aqueduct presented values between 8.0 and 14mm of distance [16]. These numerical relations are in according to descriptions given by Geurkink [17] when studying 100 samples of temporal dissected bones. The distance between the posterior boundaries of the opening of the internal acoustic meatus and the boundaries of the vestibular aqueduct was 10.7mm (variability of 7.0 and 19mm). Ziolkowski and Kurlej [18] also described such as distance (n=47 male and n=36 female samples, aging between 50 and 70 years) to be between 9.1 and 11.4mm. The mean distance was described to be 10.98mm (n=90, variability of 8.0–15mm); the mean external opening of the vestibular aqueduct was 6.7mm of length (variability of 1.0–15mm) [19]. These descriptions support that the present study brings similar findings to the last quoted reference.

5. Conclusion 

The morphology of the vestibular aqueduct varies considerably. Comparisons between the morphometric parameters between groups of genders or laterality showed that the differences were not statistically significant. A broad knowledge about the temporal bone in normal anatomy and abnormalities would avoid misinterpretation of results, bringing a high quality to the radiological results.