Superior semicircular canal dehiscence: a radiological and clinical study

Abstract

Aim

to assess radiologically the prevalence of SSCD with its clinical presentations and its relationship with age.

Methods

a prospective cohort study carried out on 200 consecutive patients (400 temporal bones). Radiological evaluation was performed using High Resolution Computed Tomography (HRCT) with measurement of thickness of bone covering superior semicircular canal (SCC), height and diameter of SSC.

Results

Two hundred patients (400 temporal bones) were involved. The mean thickness ± SD, the mean diameter ± SD and the mean height ± SD were 1.38 ± 0.80 mm, 0.94 ± 0.26 mm and 10.91 ± 2.39 mm respectively. The prevalence of SSCD and predehiscence were 1% and 14% respectively. The commonest symptom encountered was autophony (48.3%). When the SSC thickness, diameter and height were compared with the age of patients, statistically significant differences were detected. The highest diameter, lowest height and lowest thickness were found in patients aged from 54 to 72 years. Thickness of bony layer covering SSC was found to be the most validated measurement for differentiation between cases with positive and negative symptoms with the highest sensitivity and specificity.

Conclusion

The prevalence of SSCD and predehiscence varied among the studies. Autophony is the commonest symptom usually encountered. The condition is acquired rather than congenital. The thickness of bone covering SCC is the most validated measurement in differentiation between cases with positive and negative symptoms.

Introduction

Superior semicircular canal dehiscence (SSCD) is described as the deficiency of the bony layer overlying the SSC [1]. Clinical presentations include autophony, vertigo and nystagmus induced by loud sounds. Some cases may be asymptomatic [2–4]. Clinical symptoms and vestibular evoked myogenic potential (VEMP) are helpful in prediction of SSCD. However, the ideal way for diagnosis is HRCT of the temporal bone [5]. Our aim in this study was to assess radiologically the prevalence of SSCD in addition to its clinical presentations and its relationship with age.

Patients and Methods

A prospective cohort study carried out between May 2020 and January 2023 over 200 consecutive patients recruited from the otorhinolaryngology outpatient clinic at our tertiary referral center. Institutional ethics committee approval was obtained before the study conduction (code: MS.20.10.1274). Informed written consents were obtained from all patients. All cases had no history of previous middle cranial fossa surgeries. Patients with history of middle cranial fossa surgery either for otologic or neurologic cause were all excluded.

A detailed history was obtained from all patients with special focusing on manifestations of third window defect syndrome including vertigo (sound induced or pressure induced), hearing loss, tinnitus, autophony and hyperacusis. All patients were examined clinically to detect any signs of SSCD as Tullio phenomenon and Hennebert’s sign.

All patients underwent HRCT scan of their temporal bones using Multi-Splice Helical Computed Tomography equipment (Philips Ingenuity) with 0.625 mm slice thickness. Measurements were carried out in Pöschl plane using DXMM measurement software. The Pöschl plane is almost at a 45 angle to both coronal and sagittal planes. It is distinctively aligned parallel to the SSC.

In this study, three major values were measured in each HRCT scan. They were minimal thickness of the bone layer covering SSC, diameter of the SSC and height of the SSC. These values were compared with age groups of patients to determine the etiology of this condition, whether congenital or acquired. These values were also compared with clinical symptoms suggesting third window syndrome and cervical Vestibular Evoked Myogenic Potentials (cVEMP) test to identify if the clinical symptoms and cVEMP test had role in diagnosis of SSCD or not.

After measurement of the thickness of the bony layer covering SCC, the patients were classified into 4 grades according to Klopp-Dutote classification as follows:

• Grade 1: thickness of the bone was more than 2.5 mm (Fig. 1A).

• Grade 2: thickness of the bone was less than 2.5 mm (Fig. 1B).

• Grade 3 where SSC was in close contact with middle cranial fossa (Fig. 1c).

• Grade 4 where SSC was opening directly into middle cranial fossa (Fig. 1D) [6].

Fig. 1.

Grades of SSCD, A: Grade 1: bone thickness more than 2.5 mm (red line); B: Grade 2 with bone thickness less than 2.5 mm (red line); C: Grade 3 where SSC was in contact with middle cranial fossa; D: Grade 4 where SSC was opening into middle cranial fossa (Fig. 1D)

Grades 1 and 2 were described as normal; grades 3 and 4 were considered predehiscent and dehiscent respectively.

Patients with grade 3 and 4 who clinically had symptoms of third window defect syndrome underwent cVEMP to ensure the diagnosis of SSCD. The cVEMP implies an inhibitory neural reflex starting at the saccule and goes to the ipsilateral sternomastoid muscle. In SSCD, this test is usually abnormal, hence the affected ear is often sensitive to the vibratory or auditory stimuli that were used to elicit such potentials. In case of SSCD, patients usually have less than normal thresholds for cVEMP responses.

Analysis of data was carried out utilizing SPSS software program, version 28 (SPSS Inc., PASW statistics for windows version 28. Chicago: SPSS Inc.). Numbers and percents were used to describe qualitative data, while quantitative data were described using mean ± standard deviation and median. Significance was considered when P value was < 0.05. Student t test was used to compare 2 independent groups for normally distributed data. Receiver operating characteristics curve (ROC curve) was applied to calculate validity (sensitivity & specificity) of continuous variables with calculation of best cut off point.

Results

Two hundred patients (400 temporal bones) were involved in our study. The mean age ± SD was 38.06 ± 16.1 years with ages ranged between 6 and 72 years. One hundred and twelve patients (56%) were females while the remaining 88 patients (44%) were males. Radiographic measurements of superior semicircular canal are illustrated in Table 1. The mean thickness of the bone covering SCC ± SD was 1.38 ± 0.80. The median thickness was 1.10. The minimum thickness was 0.40 mm while the maximum was 3.60 mm. The mean diameter of SCC ± SD was 0.94 ± 0.26. The median diameter was 0.90. The minimum diameter was 0.40 mm while the maximum was 1.60 mm. The mean height ± SD was 10.91 ± 2.39. The median height was 10.50 mm. The minimum height was 4.80 mm while the maximum was 21.50 mm. Table 2 shows the distribution of the grades of SSCD among the studied temporal bones. Fifty-two temporal bones (13%) were classified as grade 1. Grade 2 comprised 288 temporal bones (72%). Fifty-six temporal bones (14%) were categorized as grades 3. Finally, grade 4 involved 4 temporal bones (1%). Out of the 200 patients involved in the study, only 29 patients (14.5%) had symptoms suggesting SSCD which was distributed as follows: 14 patients (48.3%) had autophony, 4 patients (13.8%) had sound induced vertigo, 10 patients (34.5%) had pressure induced vertigo and 1 patient (3.4%) had conductive hyperacusis. Between these 29 patients who had symptoms suggesting SSCD, 21 patients had grade 3&4 and those patients underwent cVEMP test where 18 cases (grade 3) had non identifiable cVEMP while the remaining 3 cases (grade 4) had identifiable cVEMP.

Table 1.

Radiographic measurements of superior semicircular canal (SSC)

	N = 400
Thickness  Mean ± SD  Median  min-max	1.38 ± 0.80 mm 1.10 mm 0.40–3.60 mm
Diameter  Mean ± SD  Median  min-max	0.94 ± 0.26 mm 0.90 mm 0.40–1.60 mm
Height  Mean ± SD  Median  min-max	10.91 ± 2.39 mm 10.50 mm 4.80–21.50 mm

Table 2.

Grades of SSCD among studied cases

	Number = 400	%
Grade 1 Grade 2 Grade 3 Grade 4	52 288 56 4	13 72 14 1

When the SSC measurements (thickness, diameter and height) were compared with the age groups of patients, statistically significant differences were detected. The highest diameter (1.22 ± 0.26 mm), lowest height (8.69 ± 1.36 mm) and lowest thickness (0.74 ± 0.47 mm) were found in patients aged from 54 to 72 years (Table 3).

Table 3.

Relation between thickness, diameter and height of the SSC and age groups of the studied cases

	Age group/years				P-value
	6–21 years	22–37 years	38–53 years	54–72 years
Thickness(mm) Mean ± SD	2.02 ± 0.83	1.66 ± 0.77	1.11 ± 0.57	0.74 ± 0.47	P < 0.001*
Diameter (mm) Mean ± SD	0.75 ± 0.18	0.85 ± 0.19	0.91 ± 0.24	1.22 ± 0.26	P < 0.001*
Height (mm) Mean ± SD	12.19 ± 2.57	11.69 ± 2.48	10.65 ± 2.18	8.69 ± 1.36	P < 0.001*

The relation between SSC measurements (thickness, diameter and height) and the presence or absence of clinical symptoms are illustrated in Table 4. A statistically significant lower thickness among cases with positive symptoms than with negative symptoms (P < 0.001). Also, a statistically significant lower mean height among cases with positive symptoms than cases with negative symptoms (P = 0.03).

Table 4.

Distribution of radiographic measurements of the canal according to presence of symptoms suggesting SSCD

	Negative symptoms	Positive symptoms	P value
Thickness (mm)	1.46 ± 0.80	0.84 ± 0.58	P < 0.001
Diameter (mm)	0.94 ± 0.26	0.94 ± 0.29	P = 0.86
Height (mm)	11.02 ± 2.34	10.27 ± 2.59	P = 0.03

Receiver operating characteristics curve (ROC curve) was utilized to calculate validity (sensitivity & specificity) of SSC thickness, height and diameter in differentiation between positive and negative symptoms (Fig. 2). Thickness was found to have the highest sensitivity and specificity in comparison to height and diameter (Table 5).

Fig. 2.

ROC curve of radiographic measurement in differentiating between cases with positive and negative symptoms

Table 5.

validity of radiographic measurement in differentiating between cases with positive and negative symptoms

	AUC (95%CI)	Cut off point	Sensitivity%	Specificity%
Thickness (mm)	0.79	0.95	84.60	65.30
Diameter (mm)	0.53	0.95	65.40	38.80
Height (mm)	0.57	10.75	61.50	50.00

Discussion

Minor et al. described SSCD for the first time as the lack of the bony layer covering SCC [7]. The incidence of such condition in the general population is difficult to estimate since its manifestations are complex. The prevalence of SSCD can only actually be evaluated in symptomatic patients. However, the small numbers of symptomatic SSCD patients do not permit any large-scale anatomic study. Accordingly, cadaver study is considered the most reliable means of SSCD identification, but it cannot include large numbers of variable age groups. The HRCT allowed a large cohort of variable age groups to be involved in the studies [6].

Clinical symptoms, supra threshold bone conduction, and cVEMP are valuable tools in diagnosis of SSCD, but the gold standard method in diagnosis is HRCT of the temporal bone [8].

When HRCT is utilized in SSCD diagnosis, standard and Stenvers planes can give false-positive and false-negative results. Such problems can be overcome by interpretation in Pöschl plane which can significantly increase specificity, sensitivity, positive and negative predictive values for diagnosing SSCD [9]. Therefore, we used; in our study; HRCT of the temporal bone with reconstruction in Pöschl plane to increase sensitivity and specificity for diagnosis of SSCD.

Different cadaver studies were conducted to assess the prevalence of SSCD. Carey et al. examined1000 specimens and reported that 0.5% had SSCD and 1.4% had predehiscence [10]. Another study by Crovetto et al. reported that prevalence of SSCD was 0.6% in 160 cadavers [11].

Radiological studies, on the other hand, reported different variable increase in the prevalence of SSCD. Crovetto et al. found that the prevalence of SSCD was 3.6% in 604 temporal bones [11]. Ceylan et al. evaluated 186 temporal bones using HRCT with 1 mm thickness and found that 12% had SSCD [12]. Nadgir et al. performed their study among 304 patients (612 temporal bones) and reported that 7.89% of the patients had SSCD while 45% showed predehiscence [13]. Cisneros et al. found that the prevalence of SSCD and predehiscence were 1.8% and 14.1% respectively in 163 temporal bones [14]. Kurt et al. used Cone Beam CT and found that the prevalence of SSCD was 6.28% in 350 temporal bones while predehiscent pattern was found in 17.71% [15]. Klopp-Dutote et al. reported that the prevalence of SSCD and predehiscence were 0.8% and 12% respectively in 354 temporal bones [6]. Another study by Akay et al. using Cone Beam CT reported SSCD in 16.5% of 400 temporal bones and predehiscent pattern in 9.5% [16]. Evlice et al. found that the prevalence of SSCD was 12.2% in 450 temporal bones using Cone Beam CT [17].

In our present study of 400 temporal bones, the prevalence of SSCD was 1% and prevalence of predehiscence was 14%. The present study findings are almost comparable to Klopp-Dutote et al. and Cisneros et al. [6, 14]. The differences between the results of all the above-mentioned studies could be attributed to the study populations, age and slice thicknesses used in CT images [17].

According to Tavassolie et al., the most common symptoms suggesting SSCD are autophony, sound induced vertigo, pressure induced vertigo and conductive hyperacusis [18]. In our study, 29 out of the 200 patients had one or more symptoms suggesting SSCD. The most common symptom in cases who had dehiscent and predehiscent patterns was autophony and this agreed with Tavassolie et al. who reported that autophony is the most common symptom in SSCD patients [18].

According to Noij and Rauch, cVEMP test is used to distinguish between dehiscent and predehiscent cases who have symptoms suggesting SSCD [19]. So, we performed; in our study; cVEMP test in all symptomatized patients who had dehiscent and predehiscent patterns (were 21 patients) and we noticed that 3 patients only had identifiable cVEMP while the remaining 19 patients were normal. Such finding was comparable to Noij and Rauch who reported that in patients with symptoms suggesting SSCD with thin bone in HRCT and/or intraoperative finding, cVEMP tends to appear normal and does not show physiological evidence of dehiscence [19].

Although studies on SSCD had increased in the recent literatures, its etiology, whether congenital or acquired was still debatable [17]. Minor et al., Nadgir et al., Davey et al., Crovetto et al. and Klopp-Dutote et al. found increased incidence of bone dehiscence with advancing age supporting the acquired etiology of such condition [3, 6, 11, 13, 20]. This could be explained by skull aging with temporal bone wear secondary to the pressure of both temporal lobe and cerebrospinal fluid (CSF) in addition to demineralization of the skull bone generally seen with aging [10, 13]. Hagiwara et al.; on the other hand; found that dehiscence is more common in children than elderly favoring the congenital etiology od SSCD [21]. On the contrary, Kurt et al., Evlice et al. and Mahulu et al. found no significant relationship between age and thickness of bone overlying SSC [15, 17, 22].

In our study, there was statistically significant relationship between thickness of bony layer covering SSC, diameter and height and age of studied cases with the lowest canal thickness, the lowest height and the highest diameter being recorded in old age. Such results supported the acquired theory for the etiology of SSCD and agreed with Minor et al., Nadgir et al., Davey et al., Crovetto et al. and Klopp-Dutote et al. results [3, 6, 11, 13, 20].

When we estimated the validity of use of thickness, height and diameter of the SSC to differentiate between cases with positive and negative symptoms, the thickness of the bony layer covering SSC was found to be the best validated measurement with the highest sensitivity and specificity.

Conclusion

The prevalence of SSCD and predehiscence varied among the studies which was attributed to the study populations, age and slice thicknesses used in CT images. Autophony is the commonest symptom usually encountered. The condition is acquired rather than congenital. The thickness of bone covering SCC is the most validated measurement in differentiation between cases with positive and negative symptoms.

Funding

None.

Declarations

Research Involving Human Participants and/or Animals

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments.

Informed Consent

Informed consent was obtained from all individual participants involved in the study.

Conflict of Interest

All authors declare that they had no conflicts of interest.