Morphometric analysis of foramen ovale, foramen spinosum, and foramen rotundum of human skull using computed tomography scan: a cross-sectional study

Background: There is limited literature of objective assessments of foramina of skull base using computed tomography (CT) scan. This study was carried out to analyze the dimensions of foramen ovale (FO), foramen spinosum (FS), and foramen rotundum (FR) using CT scan imaging of the human skull and their associations with sex, age, and laterality of the body. Materials and methods: A cross-sectional study was carried out in the Department of Radiodiagnosis and Imaging at BP Koirala Institute of Health Sciences (BPKIHS), Nepal using a purposive sampling method. We included 96 adult patients (≥18 years) who underwent CT scan of the head for any clinical indications. All those participants below 18 years, inadequate visualization or erosions of skull base foramina, and/or not consenting were excluded. Appropriate statistical calculations were done using the statistical package for social sciences (SPSS), version 21. The P-value of less than 0.05 was considered statistically significant. Results: The mean length, width, and area of FO was 7.79±1.10 mm, 3.68±0.64 mm, and 22.80±6.18 mm2, respectively. The mean length, width, and area of FS was 2.38±0.36 mm, 1.94±0.30 mm, and 3.69±0.95 mm2, respectively. Similarly, the mean height, width, and area of FR was 2.41±0.49 mm, 2.40±0.55 mm, and 4.58±1.49 mm2, respectively. The male participants had statistically significant higher mean dimensions of FO and FS (P<0.05) than the female participants. There were statistically insignificant correlations of dimensions of these foramina with age and between the left and right side of each foraminal dimensions (P>0.05). Conclusions: The sex-based difference in dimensions of FO and FS should be clinically considered in evaluating the pathology of these foramina. However, further studies using objective assessment of foraminal dimensions are required to draw obvious inferences.


Introduction
The FO, FS, and FR are located in the greater wing of the sphenoid bone [1] . The FO is more often irregular in shape with the average maximal and minimal lengths in the adult being 7.48 and 4.17 mm, respectively, and transmits the mandibular nerve (V3), the accessory meningeal artery, the lesser superficial petrosal nerve, and the emissary veins [1,2] . The FS is mostly round in shape with an average adult diameter of 2.63 mm and transmits the middle meningeal vessels from the infratemporal to the middle cranial fossa [1] . The FR is round with an average adult diameter of 3.55 mm and transmits the second (maxillary) division of the fifth cranial nerve [1,3] . Many normal variants of these foramina have been well described in the early anatomic and radiologic HIGHLIGHTS • Evaluation of skull base foramina is an important part of diagnostic medicine, especially relevant to anatomists, neuro/head, neck surgeons, and radiologists. • This study objectively assessed the dimensions of three important foramina of the human skull viz., foramen ovale (FO), foramen spinosum (FS), and foramen rotundum (FR) using computed tomography (CT) scan. • The male participants had statistically significant higher mean dimensions of FO and FS than the female participants. • There were statistically insignificant correlations of dimensions of these foramina with age and between left and right side of each foraminal dimensions of FO, FS, and FR.
literature [4] . The inequality of size of FO, FS, and FR was noted in 30.89%, 16.26%, and 0%, respectively in the same study of Ginsberg et al. [4] . The CT scan remains the modality of choice in defining the bony anatomy of the skull base [5] .
There is limited existing data on the symmetry and diameter of skull base foramina on CT, and determination of foramen widening is typically subjective [6] . Evaluation of skull base foramina is becoming an important part of diagnostic medicine, especially relevant to anatomists, neuro/head and neck surgeons, and radiologists. Therefore, this study was carried out with the objective to analyze the dimensions of FO, FS, and FR using CT scan imaging of the human skull and their associations with sex, age, and laterality of the body.

Materials and methods
Study design, sample size, and sampling technique We carried out a cross-sectional study in the Department of Radiodiagnosis and Imaging BP Koirala Institute of Health Sciences (BPKIHS), Nepal. According to a study by Berlis et al. [7] , CT measured mean breadth of the FR was reported to be 3.11 0.78 mm (mean SD). Based on this study, the sample size for our study was calculated using the one-sample mean formula at 95% CI with 5% permissible error. = ( ) × ( ) /( ) = The purposive sampling method was adopted for the collection of the study samples. We included 96 adult patients ( ≥ 18 years) who underwent a CT scan of the head for any clinical indication (except for any pathology involving these foramina) over the period of 6 months (7 October 2019-6 April 2020) and who had optimal visualization of middle cranial fossa foramen in bone window settings. All those participants with age below 18 years, poor or inadequate visualization of skull base foramina, skull base fracture, evidence of bone erosion, or destruction affecting the margins of foramina, history of skull base surgery, and/or not consenting to participate were excluded from the study. This work has been carried out in accordance with the STROCSS criteria [8] .

Measurements of foramina and their area
All the CT examinations were performed with standard protocol on a 16-slice multidetector CT scanner available in the department (NeuViz 16; Neusoft) as per the pathology/clinical indication of the participants. Volumetric acquisition of image data was obtained and the axial slice generated in the bone window was reconstructed by postreconstruction techniques into thinner slices of 0.625 mm. Three foramina (ovale, spinosum, and rotundum) of middle cranial fossa on each side, that is, a total of six foramina, were viewed in thin image sections in bone window settings (4000ww/400wl) after zooming in the area of interest. Each foramen was defined as a round or oval structure with a sclerotic osseous rim and a lucent center identified on at least two adjacent sections on expected anatomical locations for each of these structures. The FO and FS were measured in axial sections of the skull. The FR was viewed in coronal sections obtained via multiplanar reconstruction. The size of each of the foramina was measured via the electronic caliper in those sections where the foramen was adequately visualized. The maximum anteroposterior diameter (length) and transverse diameter (width) on axial sections were recorded for FO and FS (Figs. 1A and B). The maximum craniocaudal diameter (height) and transverse diameter (width) on the coronal section were recorded for FR (Fig. 1C). The three measurements of the same dimensions were made by three different radiologists and the average was taken as the final value. After recording the respective diameters of each of the foramina, calculation of the area of each of these foramina was done using the standard formula for the area of an ellipse, that is, where, D 1 = length of FO/FS, and height of FR, D 2 = width of FO/FS/FR.

Data entry and analysis
The data collected were tabulated in Microsoft Excel 2019, v16.0 (Microsoft) and analysis was carried out using the statistical package for social sciences (SPSS), version 11.5, IBM SPSS, v21 (IBM Corp.). The descriptive statistics like mean, SD, and range were used to present data like diameters and surface areas of each of the foramen. The distribution of data was confirmed by Kolmogorov-Smirnov test before using parametric tests. For comparing the mean of foramen sizes according to body side (left vs. right), paired sample t-test was used. An independent sample ttest was used to compare the mean of foramen sizes in males and females. Pearson's correlation test was used to assess the correlation of foramen size with age. The P-value of less than 0.05 was considered statistically significant.
The male participants had statistically significant higher mean values of length, width, and area of FO and FS (P < 0.05) than the female participants. However, despite higher mean values of height, width, and area of FR in male participants than female participants, they are not statistically significant (P > 0.05) ( Table 3).
The width and area of FO showed a weak positive correlation with age (r = 0.07 and 0.20, respectively), whereas the length of FO showed a weak negative correlation with age (r = − 0.07). However, none of these correlations were statistically significant (P > 0.05) ( Table 4).
The width and area of FS showed a weak positive correlation with age (r = 0.01 and 0.03, respectively), whereas the length of FS showed no correlation with age (r = 0.00). However, none of these correlations were statistically significant (P > 0.05) (Table 4).
The mean value was higher for length/height, width, and area of left FO, left FS, and left FR except for the higher mean value of the length of right FO. However, there was no statistically significant difference in right and left dimensions FO, FS, and FR (P > 0.05) ( Table 5).

Discussion
Adequate anatomical studies have been done where the foramina have been directly measured on cadaveric skull specimens [9][10][11] . However, only limited radiological studies with CT measurements of these foramina have been done [6,7,12] . Modern CT scanners are excellent in providing precise measurements of foramina which hardly differ from the direct measurements on anatomical specimens [4,6,7] . The observations of our study closely resembled previous studies. In the study by Berlis and colleagues, the CT measured mean ( SD) length and width of FO were  The mean dimensions of both the FO and FS were significantly higher in males compared with females in our study. These observations closely resembled the results of study conducted by Berlis et al. [7] and Reymond et al. [13] . However, in our study, there was statistically insignificant difference in the dimensions of FR between males and females despite higher mean values in male than female. The body habitus may account for higher dimensions in males than females [14][15][16] . There is a lack of literature comparing these dimensions in males and females and it requires further studies to draw some concrete conclusions.
In our study, none of the dimensions of the FO, FS, and FR showed significant correlations with the age of the participants. There is a lack of literature to establish this finding.
The dimensions of FO in our study resembled the various studies conducted on dry skulls in the Indian subcontinent. But the difference in dimensions of FO was not statistically significant in the current study, which is the same as multiple other studies, that is, Patil et al. [17] , Ashwini and Venkateshu [18] , and Gupta and Rai [19] .
Moreover, it was also consistently found that the length, width, and area of FO have a wide range, for which the presence and the number of emissary veins passing through it can be assumed to be the main contributor [20] . Sepahdari and Mong [6] reported easy reproducibility of the measurement of FO with high-resolution CT of the skull base and found greater variation between the subjects rather than that of left and right sides within the same subjects. Meanwhile, the study by Zdilla et al. [21] , reported significant differences of major axes and area of paired left and right-sided FO with right dimensions more than left. The study by Ahmed et al. [22] concluded a significantly higher value of length and width of FO of the right side than the left side with a statistically insignificant difference in the area of FO of either side.
Dogan et al. [23] , Kulkarni and Nikade [24] , and Reymond et al. [13] reported longitudinal and transverse diameters values of FS to be almost symmetrical on comparing right and left sides, with the insignificant difference in mean lengths and widths between the right and left FS. Another study by Emeka et al. [25] in Nigerian dry skulls, reported no significant difference between the mean length and width of the right and left FS, which is in accordance with the results of our study. In the current study, similar to several studies [23][24][25] done in the past, there was no statistically significant difference between the length, width, and area of right and left FS. Apart from this, there could be variations in FO and FS as stated in a study of middle cranial fossa using CT scan, where the most common variation of FO and FS were oval shaped and a confluence of FS with FO, respectively [26] .
There was a statistically insignificant difference between the dimensions of right and left FR in the current study. The study by Dogan et al. [23] reported the longitudinal and transverse diameter values of right and left FR to be almost symmetrical and no statistical analysis was done for this. The study by Berge and Bergman [10] reported FR to be consistently symmetrical, but when asymmetry existed, the left was more likely to be wider.
On review of the currently available literature, no plausible biological explanation could be found for the difference noted in this study. However, most of the previous studies were conducted by direct measurement among dry cadaveric skulls and there was a paucity of literature on studies conducted on objective CT measurements of these foramina. The current study also didn't consider the body mass index of the participants, which could have potentially confounded our result, as males usually have a larger body habitus than females. Also, the representation of two sexes was not proportional in our study. This sex-based difference was a unique observation in this study and is an area with a dearth of literature. This could be a new area of interest and makes a firm scope for further radiological studies. In addition,  we have not taken into account the inter-observer reliability which could also bias the findings.

Conclusions
The mean length, width, and area of FO was 7.79 1.10 mm, 3.68 0.64 mm, and 22.80 6.18 mm 2 , respectively. The mean length, width, and area of FS was 2.38 0.36 mm, 1.94 0.30 mm, and 3.69 0.95 mm 2 , respectively. Similarly, the mean height, width, and area of FR was 2.41 0.49 mm, 2.40 0.55 mm, and 4.58 1.49 mm 2 , respectively. Although the dimensions of FR showed no significant sex-based difference, the dimensions of FO and FS were significantly higher in males compared with females. There was an insignificant difference in either of the dimensions of these foramina with age nor between the dimensions of right and left FO, FS, and FR. These findings would reinforce the existing database pertaining to the objective measurements of these foramina.

Ethical approval
The ethical clearance was obtained from the institutional review committee of BPKIHS (Ref no: 199/076/077-IRC). Written informed consent was taken from each participant and the participation was entirely voluntary.

Consent
Witten informed consent was obtained.

Sources of funding
None.