Morphometric Analysis of the External Auditory Canal by Computed Tomography in Indian Population

Abstract

Various studies have shown variations in size and shape of different anthropometric measurements of external auditory canal. We conducted an anthropometric study of the three-dimensional anatomy of the osseous external auditory canal (OEAC) using high-resolution computed tomography the temporal bone to identify the variations in subset of Indian population from North India. A retrospective review of high-resolution computed tomography images of the temporal bones of 125 patients (250 external auditory canals) of different ages (mean 28.43 years) acquired from September 2014 to February of 2015 was performed. Using a method, as proposed by Mahboubio et al. (Otol Neurotol 33:715–720, 2012), six defined dimensions of the OEAC in the parasagittal planes were recorded at the level of annulus, midcanal and the outermost point of osseous external auditory canal at bony-cartilaginous junction. The length and shape of the OEAC were also studied and the frequency rate of each was recorded. The most prevalent shape of the OEAC was found to be conical (64%) and the mean osseous external auditory canal length was 9.61 mm. The length of the OEAC was significantly different between ages above and below 12 years while the 6 defined cross sectional dimensions were statistically significant between ages above and below 8 years. The history of chronic suppurative otitis media had a significant bearing on the inferior mid-anteroposterior dimension at the level of bony-cartilaginous junction. There was statistically significant difference in supero-inferior diameter in the posterior half at the level of mid-canal and outer bony-cartilaginous junction between males and females. The comprehensive set of standardized measurements collected in this study provides three-dimensional information on osseous external auditory canal geometry. These measurements and the methodology will contribute to the development of element models of the osseous canal for computational modeling purposes and also provide important measurements for design of in-the-canal hearing aids, specialized earplugs and for defining average sizes for canalplasty procedures, in pre- and postoperative surgical planning and assessment of canal atresia and stenosis in Indian population. No such previous study has been done in North Indian population.

Introduction

Anthropometry has been used for understanding human physical variations and to correlate them with racial and psychological traits. Changes in lifestyles, nutrition and ethnic composition of populations lead to changes in various anthropological measurements, which justifies periodic updating of anthropometric data collections and understand the variations. In the field of Medical Sciences, these are significant for correct diagnosis of congenital hypoplastic or acquired stenotic conditions as well as pre-operative planning of surgeries. Statistical data about the distribution of body dimensions in the population are used to design equipment and devices of optimum size and shape that fit the human body.

The human external auditory canal (EAC) is composed of lateral cartilaginous portion and the medial osseous portion. The contours of the osseous EAC (OEAC) are reliably visualized using temporal bone high resolution computed tomography (HRCT). Few studies have been attempted to quantify certain anatomic characteristics of the OEAC. A standardized method was proposed by Mahboubio et al., using mesh method to evaluate OEAC dimensions in parasagittal planes. These measurements and the methodology contributes in providing the standardized measurements for design of in-the-canal hearing aids, specialized earplugs and for defining average sizes for canalplasty procedures, in pre and postoperative surgical planning and assessment of canal atresia and stenosis. Various studies have been done in different parts of world; however no published data is available for the subset of North Indian population.

Materials and Methods

HRCT images of temporal bone acquired from September 2014 to February 2015 were reviewed. Ears with congenital canal atresia, stenosis and bony exostosis and history of otologic surgery were excluded from the study. A total of 125 patients (250 auditory canals) of all ages comprised the study sample. Age, sex, ear-related symptoms especially the history of chronic suppurative otitis media (CSOM) of each patient were recorded. Patients were categorized into 5 age groups based on the hypothesis that a potential difference might be observed as a part of normal growth and development (less than 5, 5–8, 9–12, 13–18, older than 18 year).

CT Examination Protocol

CT scans were obtained using a 40-Slice Philips Brilliance CT scanner. Helical acquisition was done with slice thicknesses of 0.6 mm (kVp 140, mAs 340, slice increment 0.33 mm) and sagittal, axial and coronal reformats were made.

Image Interpretation

The parasagittal reformats of the temporal bones were evaluated from the osseo-cartilaginous junction of the auditory canal to the annulus (lateral to medial) along its longitudinal axis. The section with complete bony closure of EAC was considered as the beginning of the OEAC. While scrolling images medially, the section within the completely closed OEAC at the level of notch of Rivinus (an opening at the superior border into the tympanic cavity) was considered as the end of the OEAC (annulus section). The length of OEAC in each ear was calculated by subtracting section numbers of the beginning and the annulus and multiplying by section thickness (in millimeters). A midcanal section as the median section between the 2 previously defined ones was also measured (if the number of sections were even, then the two middle sections’ values was averaged). To measure cross-sectional dimensions of the OEAC reproducibly, “mesh” method was used [1] to characterize the elliptical cross-sections of the EAC (Fig. 1).

Fig. 1.

Sagittal section showing the mesh-method (AB = Max SI, CD = Max AP, EF = Mid S, GH = Mid I, IJ = Mid A, KL = Mid P)

The six dimensions were defined in the form of interconnecting segments, generating 12 points on the perimeter of each OEAC cross-section that served as the end points of each interconnecting segments (Table 1). Each dimension was measured in consecutive sections from the beginning to annulus of all patients (Figs. 2, 3).

Table 1.

Descriptions of dimensions on the elliptical parasagittal cross-section of the osseous external auditory canal [1]

Defined dimension	Segment	Description
MaxSI	AB	Maximum superior-inferior height
MaxAP	CD	Maximum anterior–posterior width perpendicular to and intersecting the midpoint of AB
MidS	EF	Maximum anterior–posterior width at the midpoint of the superior half of AB
MidI	GH	Maximum anterior–posterior width at the midpoint of the inferior half of AB
MidA	IJ	Maximum superior-inferior height at the midpoint of the anterior half of CD
MidP	KL	Maximum superior-inferior height at the midpoint of the posterior half of CD

Fig. 2.

Showing 5 different shapes of EAC canal viz. conical (a), ovoid (b), cylindrical (c), reverse conical (d) and hour-glass (e)

Fig. 3.

a Sagittal section at the section of annulus showing the notch of Rivinus (straight arrow). b Coronal section showing the three levels of annulus (straight white arrow), mid section (black arrow) and outermost bony-cartilaginous section (curved arrow) at which measurements were taken

The shape of the OEAC was evaluated based on the maximum superior-inferior diameter (MaxSI). The measurements of MaxSI at the annulus, mid-canal, and beginning sections were compared to assess the trend (i.e., an increase, decrease, or no change between 2 adjacent sections).

Ears with congenital canal atresia, stenosis and exostoses and history otologic surgery were excluded. Age, sex, and history of mastoid sclerosis secondary to chronic otitis media (COM) of each patient were recorded.

Statistical Analysis

Unpaired Student’s t test was used to analyze the differences between male and female subjects, and between the patients with and without history of COM. One-way analysis of variance (ANOVA) was used to compare the mean dimensions across age groups. All statistical analyses were considered significant where the p value was less than 0.05.

Results

The mean age of the examined patient population was 28.43 years, SD 18.59, range 1–72 years. The mean OEAC length was 9.61 mm, SD of 3.74 and range of 1.8–18.2 mm (Table 2). EAC length was significantly different between age groups of 1–12 years and age groups older than 13 years. The means, standard deviations, and ranges of the OEAC measurements in different age groups are shown in Table 3. Results of the one-way ANOVA (Games-Howell post hoc test for unequal variances) are shown in Table 4. The 6 defined cross sectional dimensions had statistically significant differences between the age 1–8 years (age groups of less than 5 years and 5–8 years), and older age groups. These dimensions were not statistically different among the age groups of 9–12, 13–18, and 18 years or older.

Table 2.

OEAC length in various age groups

Age group	Mean (SD)	Range
Up to 5 years	3.58 (2.08)	1.8–7.8
5–8 years	4.64 (1.61)	2.6–8.7
9–12 years	6.56 (1.58)	3.8–9.5
13–18 years	9.12 (1.92)	6.2–14.6
Above 18 years	11.72 (2.78)	6.7–18.2
All Ages	9.61 (3.74)	1.8–18.2

Table 3.

The mean, standard deviations and ranges of OEAC measurements in different age groups

Site	Age Group	Annulus		Outer		Mid
		Mean (SD)	Range	Mean (SD)	Range	Mean (SD)	Range
Max SI	<5 years	8.03 (1.6)	5.4–10.1	8.23 (1.53)	5.5–10.3	8.23 (1.3)	6.3–10.5
	5–8 years	9.01 (1.27)	7.1–11.8	9.87 (1.82)	7.5–14.5	9.28 (1.47)	7.1–12.5
	9–12 years	9.73 (1.09)	6.9–11.9	10.91 (1.52)	8.8–13.7	10.2 (1.13)	8.0–12.5
	13–18 years	10.21 (1.66)	7.6–14.3	11.21 (2.51)	7.4–17.1	10.53 (1.8)	7.4–13.9
	>18 years	9.85 (1.47)	7.1–16.2	12.05 (1.82)	7.4–16.0	10.84 (1.48)	8.0–15.6
	All Ages	9.71 (1.49)	5.4–16.2	11.40 (2.11)	5.5–17.1	10.44 (1.62)	6.3–15.6
MidA	<5 years	6.98 (1.56)	4.2–9.1	6.93 (1.38)	4.3–8.3	7.16 (1.11)	5.1–8.3
	5–8 years	7.61 (1.05)	6.2–10.6	8.38 (1.83)	5.6–13.2	7.92 (1.44)	6.0–11.5
	9–12 years	8.2 (1.06)	5.2–9.5	8.80 (1.23)	6.6–11.5	8.41 (1.18)	5.4–10.4
	13–18 years	8.99 (1.63)	6.1–12.4	9.84 (2.21)	6.3–15.2	9.27 (1.61)	6.2–12.2
	>18 years	8.22 (1.35)	5.9–13.9	9.78 (1.76)	6.2–14.2	8.72 (1.5)	1.2–12.9
	All Ages	8.20 (1.39)	4.2–13.9	9.40 (1.90)	4.3–15.2	8.61 (1.54)	1.2–12.9
MidP	<5 years	6.65 (1.38)	4.1–8.5	6.88 (1.32)	4.4–8.4	6.91 (1.24)	4.8–8.9
	5–8 years	7.60 (1.06)	5.6–10.3	8.27 (1.84)	5.7–14.1	7.82 (1.48)	5.6–11.8
	9–12 years	7.57 (1.07)	5.0–9.6	8.78 (1.29)	6.2–11.4	8.34 (1.09)	5.8–9.8
	13–18 years	8.68 (1.39)	5.9–11.8	9.56 (2.26)	5.8–14.6	9.03 (1.58)	6.3–12.5
	>18 years	7.99 (1.33)	5.2–13.1	9.89 (1.74)	6.2–13.8	8.76 (1.5)	1.8–13.5
	All Ages	7.94 (1.33)	4.1–13.1	9.41 (1.92)	4.4–14.6	8.57 (1.55)	1.8–13.5
MaxAP	<5 years	5.40 (0.94)	3.9–6.5	5.50 (1.09)	3.5–6.6	5.42 (1.32)	3.7–7.0
	5–8 years	5.42 (1.03)	3.8–7.8	5.91 (1.12)	3.8–8.0	5.64 (1.08)	3.5–7.8
	9–12 years	6.40 (0.95)	5.1–8.8	6.83 (0.83)	5.0–8.8	6.54 (0.75)	5.0–8.1
	13–18 years	6.37 (1.04)	4.1–8.2	6.88 (1.23)	4.6–9.6	6.55 (1.04)	4.4–8.3
	>18 years	6.42 (1.01)	3.6–9.1	7.61 (1.06)	4.5–10.5	6.99 (1.07)	4.0–10.6
	All Ages	6.24 (1.07)	3.6–9.1	7.13 (1.20)	3.5–10.5	6.65 (1.15)	3.5–10.6
Mid S	<5 years	4.66 (0.92)	3.2–5.6	4.84 (0.87)	3.1–5.6	4.37 (1.01)	3.2–6.0
	5–8 years	4.69 (1.08)	2.8–7.1	5.24 (1.09)	3.3–7.4	4.97 (1.04)	3.4–7.3
	9–12 years	5.5 (0.89)	3.8–8.3	6.0 (0.84)	4.3–7.8	5.82 (0.81)	3.9–7.6
	13–18 years	5.70 (1.06)	3.3–7.4	6.23 (1.19)	4.1–8.8	5.84 (0.97)	3.9–7.5
	>18 years	5.43 (0.96)	3.1–7.7	6.79 (0.95)	4.1–9.1	6.15 (0.96)	3.7–8.9
	All Ages	5.35 (1.03)	2.8–8.3	6.36 (1.16)	3.1–9.1	5.86 (1.06)	3.2–8.9
Mid I	<5 years	4.48 (0.94)	3.1–5.8	4.72 (1.0)	2.8–5.7	4.42 (1.21)	3.0–6.6
	5–8 years	4.63 (0.92)	3.1–7.0	5.05 (1.07)	3.3–7.1	4.69 (1.05)	3.1–7.5
	9–12 years	5.34 (0.96)	4.1–7.4	5.7 (0.81)	3.6–7.5	5.41 (0.84)	3.5–7.3
	13–18 years	5.57 (0.99)	3.7–7.2	6.10 (1.18)	3.9–8.8	5.75 (0.98)	3.8–7.5
	>18 years	5.52 (0.91)	3.2–8.1	6.39 (1.01)	3.8–9.4	5.89 (0.96)	3.6–9.1
	All Ages	5.35 (0.98)	3.1–8.1	6.06 (1.14)	2.8–9.4	5.81 (1.07)	3.0–9.1

Table 4.

ANOVA (Games-Howell post hoc test for unequal variances) for significance of various dimensions in different age groups

Site	Age Group (years)	Annulus					Outer					Mid
		<5 years	5–8 years	9–12 years	13–18 years	>18 years	<5 years	5–8 years	9–12 years	13–18 years	>18 years	<5 years	5–8 years	9–12 years	13–18 years	>18 years
MaxSI	<5	–		0.032	0.002	0.006	–		0.005	0.001	0.000	–		0.011	0.001	0.000
	5–8		–		0.006	0.020		–		0.032	0.000		–		0.006	0.000
	9–12			–					–		0.041			–
	13–18				–					–					–
	>18					–					–					–
Mid A	<5	–					–					–
	5–8		–					–					–
	9–12			–					–					–
	13–18	0.002	0.000		–	0.023		0.008		–		0.003	0.002		–
	>18					–	0.000	0.001			–	0.033	0.038			–
Mid P	<5	–					–					–
	5–8		–					–					–
	9–12			–					–					–
	13–18	0.001	0.006	0.010	–	0.041	0.002	0.026		–		0.003	0.008		–
	>18	0.036				–	0.000	0.000	0.033		–	0.006	0.009			–
Max AP	<5	–				0.046	–					–				0.001
	5–8		–	0.002	0.001	0.000		–					–	0.009	0.004	0.000
	9–12			–			0.020	0.010	–					–
	13–18				–		0.011	0.002		–					–
	>18					–	0.000	0.000	0.008	0.004	–					–
Mid S	<5	–					–					–		0.002	0.001	0.000
	5–8		–	0.016	0.000	0.001		–					–	0.007	0.003	0.000
	9–12			–			0.035	0.030	–					–
	13–18				–		0.004	0.001		–					–
	>18					–	0.000	0.000	0.002	0.026	–					–
ITR	<5	–			0.027	0.020	–					–			0.006	0.000
	5–8		–	0.031	0.000	0.000		–					–	0.042	0.000	0.000
	9–12			–					–					–
	13–18				–		0.004	0.000		–					–
	>18					–	0.000	0.000	0.013		–					–

Applying unpaired t test between males and females for various dimensions difference was significant in MidP at outer and mid sections (p value 0.021 and 0.037, respectively). Rest dimensions were not significant.

Applying unpaired t test between presence and absence of CSOM, difference was significant in MidI at outer section (p value 0.037).

Evaluation of the MaxSI’s measurements between the annulus and beginning sections revealed five different shapes of the OEAC (Table 5). In 161 ears (64.4%), MaxSI consistently increased from the annulus to the beginning section (conical). In 38 ears (15.2%), MaxSI at midcanal section was greater than that at the annulus and beginning sections (ovoid). In 27 ears (10.8%), MaxSI was approximately constant at all 3 sections (cylindrical). In 15 ears (6%), MaxSI consistently decreased from the annulus to the beginning section (reverse conical). In 9 ears 3.6%), MaxSI at the midcanal section was lower than that at the annulus and beginning sections (hourglass).

Table 5.

Frequency of different shapes of OEAC

Shapes	Frequency	Percentage
Conical	161	64.4
Ovoid	38	15.2
Cylindrical	27	10.8
Reverse Conical	15	6.0
Hour glass	9	3.6
Total	250	100

Discussion

The external ear (pinna and ear canal) plays a major role in transforming acoustic signals from free field to the tympanic membrane in humans [2].The EAC is on average, 2–3 cm in length, with the inner two-third osseous part and outer one-third is cartilaginous [3, 4]. Human factors and ergonomics play a very important role in mass customization of various in-the-canal products (ITC) and for defining average sizes for canalplasty operations. Many studies have been conducted in different parts of the world for geometrical evaluation of external auditory canal.. Ethnic diversity is a very significant factor that affects the anthropometric data and scope of its application [5]. The results from those studies could not be globally applied. No such study has been conducted on North Indian population.

In early studies, EAC of the cadaveric temporal bones was filled with resins to form a cast of the entire external auditory canals and then its dimensions were measured using micrometer [6] or microtomy and microradiography [7, 8]. Eckerdal et al. [7] used resin, microtomy, and microradiography to measure the distance from complete closure of the OEAC to appearance of the first opening into the middle ear cavity at the level of the annulus. A mean OEAC length of 6.53 mm (SD 1.80) was reported in 58 cadaveric temporal bones of male and female subjects aged 20–100 years. In another study by Smelt et al. [9] on 34 cadaveric temporal bones of male and female subjects aged 46–91 years, resin and micrometry were used, and the mean length of the OEAC was reported to be between 6.5 and 12 mm. However, the results obtained from cadaveric study could not be extrapolated to general population.

Micrometry was replaced with the advent of CT. Egolf et al. [10] used CT to evaluate the ear canals of cadavers for the measurement of volume and compared the results with those obtained from injection measurement method, and observed a significant 6.12% difference between these two methods. The early studies of the temporal bones with CT scan suffered from low resolution simply measured the length and cross-sectional area of the EAC and could identify only gross congenital anomalies. The shape and longitudinal cross sections of the OEAC were not described. High resolution Computed Tomography is now the major imaging method in evaluating the temporal bone [3]. Tu et al. devised non-invasive method for establishment of three-dimensional ear canal model and geometric shape of canal using volumetric CT [11]. Mahboubi et al. [1] used mesh method to measure the various cross-sectional diameters, shape and the length of the EAC and described wide range of values for the length of the OEAC (4.0–17.4 mm) and average length 8.5 mm (SD of 3.4 mm). Our study revealed that the mean length of the OEAC was 9.61 mm (SD 3.74; range 1.8–18.2) in all ears, and 11.72 (SD 2.78; range 6.7–18.2) in the ears of the patients older than 18 years. Our study described a wider range between 1.8 and 18.2 mm which represents a greater variation than those reported in the previous anatomic studies, corresponding to a wider age range, large number of subjects and different ethnic background.

In the study by Eckerdal and Ahlqvist [7, 8] three different shapes of the OEAC: cone shaped (64.1% of the subjects), hourglass shaped (32.1%), and ovoid (3.8%) were reported. In the study by Mahboubio, the most common shape of the OEAC was conical, with a prevalence of 64%. This was followed by ovoid (18%), reverse conical (11%), hourglass (6%), and cylindrical (2%) shapes. Our study also revealed conical as the most common shape (64.4%) followed by ovoid (15.2%), cylindrical (10.8%), reverse conical (6%) and hour-glass (3.6%). The variation in shape can be explained by the different race of subjects being studied.

The method can be used to obtain measurements of the cartilaginous EAC or fitting of the invisible CIC hearing aids. The former is however not studied well because of the elastic and viable nature and lesser involvement in canal procedures.

It has been suggested that patients with chronic otitis media have smaller EAC dimensions, and require routine canalplasty [6]. Mahboubio et al. stated no significant difference in the OEAC measurements between ears with and without history of CSOM. However, in our study significant difference was found in the outer Mid I. In the study by H. Mahboubio, OEAC length was statistically different between the age groups of 5–12, and 13 years or older. The 6 defined cross sectional dimensions had statistically significant differences between the age groups of 5–8 and older. These dimensions were not statistically different among the age groups of 9–12, 13–18, and 18 years or older. In our study, the length of the OEAC was significantly different between age groups of 1–12 years and age groups older than 13 years. The 6 defined cross sectional dimensions had statistically significant differences between the age 1–8 years (age groups of less than 5 years and 5–8 years), and older age groups. These dimensions were not statistically different among the age groups of 9–12, 13–18, and 18 years or older. The significance is explainable since the increase in dimension is function of age. The results from various studies have been summarized in Table 6.

Table 6.

Comparison of results from various studies

	Current	Mahoubi	Cadaveric (Eckerdal)	Cadaveric (Smelt)
Length of EAC	9.61 mm (1.8–18.2 mm)	8.5 mm(4.0–17.4 mm)	6.5 mm	6.5–12 mm
Shapes of EAC	Conical > ovoid > cylindrical > rev.conical > hour-glass	Conical > ovoid > rev.conical > hour-glass > cylindrical	NA	NA
Effect of CSOM	Significant difference in the outer Mid I	No significant difference	NA	NA
Gender difference	Significant in MidP at outer and mid sections	No significant difference	NA	NA
EAC length	Significantly different between age groups of 1–12 years and age groups older than 13 years	Statistically different between the age groups of 5–12, and 13 years or older	NA	NA
Six defined dimensions	Statistically significant differences between the age 1–8 years	Statistically significant differences between the age groups of 5–8 and older	NA	NA

The tympanometry [12, 13] was employed to measure the ear canal volume, with the human ear canal cross-sectional area presumed as a fixed value and the volume of the canal would not change as a result of variations in the soft tissue in the canal and changes in the eardrum. Optical methods [14] were used to evaluate the auditory canal length between the opening of the canal and the ear drum. Advanced and more accurate methods like laser scanning [15] and 3-dimensional shape modeling [16] may be used for the same purpose.

Anthropometry of external auditory canal by various studies has highlighted the variation of the geometry depends on the age, sex and ethnicity of the subjects.

The patients with exostosis, congenital canal atresia, stenosis, and those with history of previous otologic surgery were excluded. The aim of the current study was to provide standardized measurements of the OEAC in North Indian population. However, this does not replace the clinical assessment of these patients, including skin and canal bone thickness. The standardized measurements provided by this study may be used to produce reference ranges for different age groups in North Indian population, to diagnose external auditory canal stenosis and in development of various standardized in-the-canal hearing devices for patients with different OEAC dimensions and in pre-operative planning of canalplasty procedures. This method can also be useful to individualize the measurements before canalplasty procedure due to the anatomic variability between different patients.

Conclusion

The comprehensive set of standardized measurements collected in this study provides three-dimensional information on osseous external auditory canal geometry. These measurements in patients of various ages and the methodology will contribute to the development of more accurate designs and assessment of sound attenuating ear molds and ear plugs and to guide the design of in-the-canal hearing aids and devices. Further these dimensions can be helpful in preoperative planning of canalplasty operations and for diagnosis of canal stenosis. A further study with a larger cohort of cases with same methodology could be helpful in establishing a regional anthropometric database for ethnic Indians.