Background

Sphenoid sinus is an important landmark for endoscopic sinus surgery (ESC) as well as endoscopic skull base surgery and is a compelling factor that directly affects the planning and performing of the endoscopic approaches due to the different variations in septation, shape, dimensions, and complex relation with associated vital structures [1,2,3]. Sphenoid sinuses are double cavities formed within the body of the sphenoid bone, communicating with the roof of the nasal cavity through the sphenoethmoidal recess in its anterior wall. A large sinus can show a number of ridges and depressions related to closely adjacent structures. The degree of pneumatization of the sphenoid sinus is highly variable and reaches its mature size by 14 years of age [4]. The sphenoid bone is an important structure of endoscopic approaches because it is located in the center of the skull base and has a key role such as either a door or barrier to dissection. The sphenoid sinus is the most difficult to reach among the paranasal sinuses. The close relationships of the air spaces in the sphenoid sinuses with important peripheral structures such as the internal carotid artery, optic nerve, maxillary nerves (in foramen rotundum) and Vidian nerves have attracted the attention of clinical studies [3]. In the case of diffuse pneumatization, adjacent structures may protrude into the sinus cavities, sometimes without bony structure [5]. In order to reduce the complications that may occur due to transsphenoidal and functional endoscopic sinus surgery, the sphenoid sinus and related regional anatomy should be known. The most useful method for this is the radiological examinations [6]. Evaluation of anatomy and variations with a paranasal sinus computed tomography (PNSCT) in the preoperative period is important in terms of preventing complications.

The extent of pneumatization is highly variable, and as it expands, it encircles the neighboring structures more. It poses a potential hazard to surrounding structures, including the optic, Vidian, oculomotor, trochlear, and abducens nerves, cavernous sinuses, and internal carotid arteries [7].

The aim of this study was to examine relation of the extent of pneumatization of the sphenoid sinus with the adjacent vital structures as internal carotid artery, optic canal, and Vidian nerve by using PNSCT.

Methods

The study was designed retrospectively. PNSCT images obtained between October 2018 and June 2021 in a secondary referential state hospital were reanalyzed. High-resolution PNSCT images (Toshiba, Activion16 Multislice CT, Kyoto, Japan) in axial, coronal, and sagittal planes examined for sphenoid sinus, Vidian canal, optic canal, and carotid canal bilaterally. Participants over the age of 18 without any paranasal sinus disease were included in the study. The patients with prior sinus surgery, maxillo-facial trauma, sinonasal tumors, severe septum deviation, severe concha bullosa, midline defects, nasal polyposis, and chronic rhinosinusitis were excluded.

Outcome parameters

The type of sphenoid sinus in the sagittal plane (pre-sellar, sellar, or post-sellar) and the coronal (pre-Vidian, intercanal, or post-rotundum) plane was defined. Vidian canal was classified according to the definition by Lee et al. [8]; (1) completely within the SS (Fig. 1), (2) on the floor or partially protruding into the SS, and (3) completely embedded in the sphenoid corpus (SC) (Fig. 2). The optic canal was classified according to the relationship with the sphenoid sinus wall; (1) no indentation, (2) indentation to the SS wall, (3) protrudation to the SS wall (Fig. 3), and (4) extends laterally to the sphenoid sinus and posterior ethmoid cell (PEC) [9]. A similar classification was used for the carotid canal; (1) no indentation, (2) indentation to the SS wall (Fig. 4), and (3) protrudation to the SS wall [10]. All classifications were done bilaterally and recorded separately.

Fig. 1
figure 1

The Vidian canal is completely embedded in the sphenoid corpus on the coronal section

Full size image
Fig. 2
figure 2

The bilateral Vidian canal is completely within the sphenoid sinus

Full size image
Fig. 3
figure 3

Protrudation of both optic canals to the sphenoid sinus wall

Full size image
Fig. 4
figure 4

A partial indentation of the left carotid canal to the sphenoid sinus wall

Full size image

Statistical analysis

SPSS 22.0 program (IBM Corp., Armonk, NY, USA) was used for statistical analysis. The distribution of scale variables was evaluated with the Shapiro-Wilk Francia test, while the homogeneity of variance was evaluated with the Levene test. One-way ANOVA (Robust Test: Brown-Forsythe) and the Tukey HSD test for post hoc analysis were used to compare groups for age. The Pearson chi-square test was tested with the Monte Carlo simulation technique in the comparison of groups according to gender. The possibility of co-occurrence of anatomical groups was evaluated with Spearman’s correlation test. While quantitative variables were expressed as mean (standard deviation) and median (minimum-maximum) in the tables, categorical variables were shown as n (%). The variables were analyzed at a 95% confidence level, and a p value of less than 0.05 was considered significant. The study was carried out in accordance with the 1964 Helsinki Declaration and subsequent amendments. The institutional review board was approved by the ethic committee of affiliated university for the study (1118-949/27.06.2022).

Results

Paranasal sinus computed tomography (PNSCT) of 262 patients was evaluated bilaterally (524 sphenoid sinuses). Of the patients, 100 (38.2 %) were females and 162 (61.2%) were males. In the sagittal plane, sphenoid sinuses on the right and left sides (RSSS vs LSSS) were found to be the most postsellar type (n= 208, 79.4% vs. n= 210, 80.2%). At the coronal plane of images, the most common variation of sphenoid sinus was both on the right and left as pre-VIDIAN type (RSSC vs. LSSS) (n= 124, 47.3% vs. n= 116, 44.3%). The Vidian nerve canal on both sides (RVC vs LVC) was detected as completely embedded in the sphenoid corpus (n= 123, 46.9% vs n= 123, 46.9%). Optic canal with partial indentation to sphenoid sinus wall was found both at right and left (ROC vs. LOC) as mostly variation (n= 120, 45.8% vs. n= 119, 45.9%). No indentation or partial indentation of the right and left carotid canal (RCC and LCC) to the sphenoid sinus wall was the mostly structural formation both on the right and left side (Table 1).

Table 1 Frequencies of variations of the sphenoid sinus and associated structures on right and left side
Full size table

The carotid canal did not show identification on the SS wall in females, both on the right (49.0% vs. 29.6%) and on the left (49.0% vs. 30.2%) side, compared to males (p= 0.002 and 0.002). In accordance, the indentation of the carotid canal on the SS wall was detected significantly with a higher rate in men than in women (Table 2).

Table 2 Frequencies of variations of the sphenoid sinus and associated structures according to the gender on the right and left side
Full size table

While the mean age was found to be significantly higher in patients with presellar type of sphenoid sinus compared to the sellar and postsellar type, the mean age was similar in the sellar and postsellar sphenoid sinus groups on right side (RSSS; p=0.003, 0.006, and 0.341, respectively). On the right side, mean age of patients with VC completely within the sphenoid sinus was significantly lower compared to patients with VC completely embedded in the sphenoid corpus (p= 0.005). There was a similar relationship between Vidian canal type and age on the left side (Table 3). The mean age of patients with an optic canal protrudation to SS wall was significantly higher than patients with an optic canal indentation to SS wall or laterally to the SS and PEC on right side (ROC; p= 0.007 and 0.010).

Table 3 Average age in different variations of the sphenoid sinus and associated structures
Full size table

Left optic canal course as protruding to SS wall was detected in patients with higher average age compared to courses as indentation to SS wall or laterally to the SS and PEC (p= 0.008 and 0.005). The mean, standard deviation, and other statistical data of age according to subgroups for anatomical variations were given in Table 3.

SS variations on the sagittal plane were significantly correlated with RVC, RCC, LVC, and LCC variations. Also, SS variations on the coronal plane were significantly related to types of LVC and RVC (Table 4). But, when the sphenoid sinus variations were evaluated in total (no side classification), according to the sagittal and coronal planes, there were no significant correlation was detected between the sphenoid sinus variations and other VC, CC, or OC variations (Pearson correlation, p> 0.050).

Table 4 Spearman correlation analysis of sphenoid sinus variations and adjacent structures
Full size table

Discussion

Main results of study demonstrated that sphenoid sinuses on the right and left sides were mostly postsellar type and pre-Vidian type at the coronal plane. A completely embedded Vidian nerve canal in the sphenoid corpus, a partial indentation of the optic canal to the sphenoid sinus wall, and a structural formation of the carotid canal were the most presented variations on both sides. The carotid canal variations were significantly differentiated between gender. With increasing age, the Vidian canal tended to be embedded in the SC. The rate of optic canal protrudation to the SS wall was significantly higher in older ages. LSSS variations were significantly correlated with RVC, RCC, LVC, and LCC variations.

Current literature supported that sphenoid sinuses show a parallel expansion in antero/posterior and lateral directions until age 10, and there may be a slight global expansion in all directions after age 14 or 15 [11,12,13]. The results in the presented study may indicate an expansion of the sphenoid sinus in the cranial route rather than lateral direction with advancing age.

SS are complex structures due to their deep anatomical localization and high variability in shape and size depending on the degree of pneumatization. It was stated that size and shape of SS show a high variability with a range of 200% or more in both directions [12]. Embryologically, the carotid artery, optic nerve, and Vidian nerve develop before sphenoid sinus development, and sinus enlargement gives these structures their final anatomical design as irregularities in the sinus wall [12]. The anatomical relationships of the various neurovascular structures vary according to the degree of pneumatization of the sphenoid sinus [14]. Hammer and Radberg showed that sellar type being the commonest variation [15]. The incidence of this study was reported as 86%. In similar studies, it has been shown that sellar pneumatization is higher at a rate of 55–77.3% [14]. El Sayed et al. found a higher rate of presellar pneumatization in their population [16]. Abdalla [17] reported postsellar pneumatization at a rate of 74.8% in his study on the Iraqi population. In our study, we found that postsellar pneumatization was at a highest rate. We think that finding similar results with Abdalla is due to the close neighborhood of the Iraqi and Turkish populations and the fact that people of similar ethnic origin (Arabic and Kurdish) live in both regions. Also at the coronal plane, the most common variation of SS was both on the right and left as the pre-Vidian type, and SS variations on the coronal plane were significantly related to the type of Vidian canal variations. Although this actually shows an internal consistency of SS and VC classification in the coronal plane, in fact, the lateral pneumatization of the SS toward to anterior clinoid process correlates with the Vidian canal type and SS variation in the coronal plane [12, 18, 19].

The Vidian nerve runs through a bony tunnel along the floor of the sphenoid sinus. This canal connects the foramen lacerum with the pterygopalatine fossa. The location of this channel is variable and may differ with the degree of SS pneumatization. The pneumatization pattern of the Vidian canal has a number of implications during the operational approach. The fact that neighboring structures are surrounded by air or bone makes it difficult to see the desired structure. The Vidian canal may be completely embedded in the SS corpus. In the literature, Yeğin et al. [6] reported as 39%, Yazar et al. [20] as 36%, and Adin et al. [21] as 48% as embedded in the sphenoid corpus. In our study, we found that 46.9% of them were embedded in the sphenoid corpus, and it was the most common type. The fact that these studies were also conducted in a specific local population, and the results are different may mean a possible ethnic anatomical change. In our study, we detected the second type protruding towards the sphenoid sinus. Lang and Keller [22] reported that the Vidian canal protrudes into the sinus cavity in 18% and Hewaidi and Omami in 27% [23]. Adin et al. [21] reported 44% rate of the Vidian canal protruding into the sphenoid sinus. We found similar results in the study reported by Adin et al. [21]. This is because these studies were conducted in similar regions in Turkey. Arabic and Kurdish populations live in these regions intensively. This highlights the importance of ethnicity in sinus pneumatization. In this study, as in previous studies, there was no significant gender preference [19, 21, 24].

The optic canal is a cylindrical canal running obliquely through the lesser wing of sphenoid bone near the base where it joins the body of sphenoid. It transmits the optic nerve and ophthalmic artery. It can compress the sphenoid sinus in the superior section. In our study, optic canal indentation was 45.8 % on the right side and 45.9% on the left side, and optic canal protrudation was 25.6 % on the right side and 25.9% on the left side. No gender difference was detected. In the study of Asal et al. [10] indentation rate was determined as 11.3 and 9.9% in males and females. Hewaidi and Omami [23] reported that 35.6% of the patients had optic nerve protrusion and 30.6% had optic nerve indentation. In another study, Fuji et al. [25] reported that 4% of the optic nerves were indentation on the sphenoid bone. Previous studies reported a wide range of protrusion rates of 8 to 70% [23]. In our study, the separation rate was found to be relatively high compared to the literature, and the protrusion rate was found to be relatively low. We think that the different results obtained from these studies are based on regional differences and ethnic origins again. Previously, a significant correlation of the anterior clinoid process (lateral expansion of SS) with the OC variations was reported, but lateral or cranial pneumatization of SS in this study did not show a significant relation with OC in this study [5, 12, 18].

In our study, carotid canal protrudation to the sphenoid sinus wall was present 22% in males and 25.3% in females. Dehiscence in carotid canal was detected more in females (42%) compared to males (29%). Asal et al. [10] reported that 32.1% of the patients had internal carotid artery protrusion into the sphenoid on the right and 23.9% on the left. Hewaidi and Omami [23] reported that 41% of patients had internal carotid artery protrusion towards the sphenoid bone. Gibelli et al. [26] reported the rate of internal carotid artery protrusion as 46.2%. Otherwise, the prevalence of ICA protrusion has been reported in a wide range in the literature, generally varying between 3.9 and 41.0% [12]. Our study is compatible with the literature. A preoperative evaluation of the sphenoid sinus with CT will help prevent possible complications. Variations of SS on the sagittal plane was significantly related CC variation in presented study, but there was no data for a valid comparisons in the literature. Also the course of this internal carotid artery adjacent to the SS and the cranial expansion potential of the SS may be related this association. Additionally, as demonstrated in the current literature, SS pneumatization and expansion essentially determine the variation of associated structures such as CC and OC [17, 26].

Although this study predominantly sampled from a relatively single ethnic population, it supports that, as intended, variations of the SS and associated structures are primarily associated with a pattern of SS pneumatization or enlargement. Repeating similar studies in larger samples from different ethnic origins will yield more valid results.

Conclusion

Pneumatization of SS significantly affects the development of surrounding vital structures, especially with increasing age and gender. So, it should be strongly advised a preoperative radiological evaluation of patients. A PNSCT or magnetic resonance imaging must be obtained routinely preoperatively to reduce the risk of morbidity and mortality.