Introduction

The cribriform plate (CP) is enclosed between the lateral lamellae, and it extends anteroposteriorly from the crista galli to the planum sphenoidale (1). It is perforated by numerous small openings, known as olfactory foramina, which act as a passage for olfactory nerve fibers that enter the roof of the nasal cavity to allow olfaction (2). The fovea ethmoidalis attaches to the lateral lamella, and the junction of the fovea ethmoidalis and the lateral lamellae is the canal for the transmission of the anterior ethmoidal artery, which is at a risk of injury during endoscopic sinus surgery (ESS) (3, 4).
This morphological thinness of the lamina lateralis of the lamina cribosa makes it a high-risk area during endonasal surgery, where most surgical complications occur. The risk of iatrogenic injury due to surgery is increased further if one of the lateral lamellae is inferiorly positioned. Therefore, asymmetry in the olfactory fossa depth or the height of the lateral lamella is associated with an increased risk of intraoperative injury during endonasal surgery (5). 
The variant anatomy of the CP has been studied, and several classifications have been developed. These are the Keros, the Gera classification, and more recently the Thailand-Malaysia-Singapore (TMS) classification. The Keros and Gera classifications are widely used to stratify patients who are at risk for ESS. In Africa, the morphometry of the depth of CP has been done among the South African population. There is a dearth of studies of the morphometry of the depth of CP according to the Keros classification. The main aim of this study is to determine the depth of the CP according to the Keros classification.

​

Materials and Methods

Study design and study site
This was a retrospective, radiomorphometric, cross-sectional quantitative study conducted at a single tertiary institution.

​

Study population, sampling, and sample size
The target population consisted of 667 patients. Thereafter, based on the inclusion and exclusion criteria, a sampling frame consisting of 502 patients was derived from the target population (see Table 1). Probabilistic simple random sampling was then used to obtain a sample of 223 patients using the Yamane formula (6) from the sampling frame calculation (see Table 2).

​

Table 1.

Baseline characteristics of patients

​

Table 2.

Taro Yamane formula and sample size calculation

​

Inclusion criteria
All patients above the age of 18 years who underwent paranasal CT scan between January 2021 and December 2023 were included in the study. All the scans had to meet the scan parameters of 5-mm acquisition slice, spiral acquisition mode, artifact free, and free from motion blur.

 

Exclusion criteria
Patients who had undergone endonasal surgery, patients with osseous or neoplastic pathology affecting the sinonasal area, or patients who had facial bone trauma were excluded from the study.

​

CT imaging protocol and measurement
The images were acquired using a GE 128 CT scanner (GE, Boston, MA, USA) and retrieved from the Fuji RIS/PACS software for analysis and measurement. Subsequently, they were reformatted to a coronal plane with a slice thickness of 1 mm and in a sharp kernel filter (bone window). 
The following anatomical points were identified on coronal CT bone window reformats at the level of the ostial-meatal complex:
I.    Cribriform plate (CP)
II.   Lateral lamella of Cribriform plate (LLCP)
III.  Orbital floor (OF)
IV.  Fovea ethmoidalis (FE)
The cribriform was divided into right and left parts using the crista galli as a reference point. The vertical distance from the base of the CP to the level corresponding to the junction between the LLCP and the fovea ethmoidalis was measured on both sides of the midline. The grading of the depth was done according to the Keros classification:
I.    Type I (1–3 mm)
II.   Type II (4–7 mm)
III.  Type III (>8 mm)

 

Statistical analysis
Descriptive measures such as mean, median, and standard deviation were utilized to describe the data. Statistical analysis was done using SPSS version 30 (trial version) (SPSS Inc., Chicago, IL, USA) and a p-value of <0.05 was considered as statistically significant. The Kolmogorov–Smirnov test (KS) was used to check for normality of the data. The effect size was also calculated to check for the significance of the difference between the sample distribution and the normal distribution. Pearson correlation analysis was also utilized to demonstrate the strength of the relationship between the right depth of the CP and the left depth of the CP.

​

Ethical approval
The institutional ethics and review committee approved the research on February 13, 2024 (TNH-ISERC/DMSR/ERC/RP/021/23).

​

Results

Depth of the right CP
The average depth of the right CP was 5.55 mm with a median of 5.43 mm and a standard deviation of 1.7289. The depth of the CP ranged from 1.65 to 11.04 mm. The KS test showed that the depth of the right CP was normally distributed at a p-value of 0.1524 (p-value = 0.05) with a small observed effect (the difference between the sample distribution and the normal distribution) of D=0.0518 (see Figure 1).

 

Figure 1.

Histogram shows the frequency of the depth of the right cribriform plate.

 

Depth of the left CP
The average depth of the left CP was 5.66 mm with a median of 5.61 mm and a standard deviation of 1.8866. The depth of the CP ranged from 1.96 to 11.57 mm. The KS test showed that the depth of the left CP was not normally distributed at a p-value of 0.06816 (p-value = 0.05) with a very small observed effect (the difference between the sample distribution and the normal distribution) of D=0.0721 (see Figure 2).

Figure 2.

Histogram showing the frequency of the depth of the left cribriform plate. 

​

Correlation between the right depth of the CP and the left depth of the CP
Pearson correlation analysis showed that there was a significantly positive relationship between the right depth of the CP and the left depth of the CP. Correlation (r) equals 0.639 and a p-value of <0.001. This means that there is minimal asymmetry between the depth of the CP on both the right and left sides (see Figure 3).

​

Figure 3.

Line graph showing the relationship between the depth of the right cribriform plate and the depth of the left cribriform plate.

​

Keros classification of the depth of the CP

The majority of the patients fell under the Keros type II classification (4–7 mm), with 74% (n=165) on the right side and 68% (n=153) on the left side. This was followed by Keros type I (1–3 mm), which comprised 19% (n=42) on the right and 20% (n=44) on the left. A minority of the patients, 7% (n=16) on the right and 26% (n=12%) on the left, were classified as Keros type I (>8 mm) (see Table 3).

 

Table 3.

Keros Classification according to the depth of the CP

​

Discussion

The accurate and precise knowledge of the morphometry of the ethmoid bone, particularly the CP, is of utmost importance when it comes to ESS preoperative planning (3). ESS remains the preferred therapeutic tool for the treatment of sinonasal pathology. As with all surgical procedures, it still carries a minimal risk of complications due to the tiny spaces that the surgeons must maneuver. Some of these risks include orbital injury, hemorrhage, cerebrospinal fluid leak, or even toxic shock syndrome (7).
The Keros classification is a preoperative risk stratification tool developed in 1962 based on cadaveric studies by a Croatian physician by the name of Prof. Predrag Keros (1933–2018) (5). To date, it has been used by surgeons before for screening and presurgical planning. The Keros classification is based on the depth of the CP. Patients with Keros type I are considered at a low risk of surgical complications since the depth of the CP is shallow. Patients with Keros type II are considered at moderate risk of surgical complications since the depth of the CP is deeper as compared to Keros type I. In Keros type III, the depth of the CP is deeper than in Keros type I and type II. This carries a significant risk during surgery, and therefore patients with Keros type III are considered high risk for ESS (8). 
There is a paucity of local studies on the depth of the CP according to the Keros classification in Africa. To the best of the authors’ knowledge, there was only one study done in 2022 among the South African population. In that study, Keros type II was the most ubiquitous type (right 74.4%, left 76.7%), followed by type I (right 16.3%, left 11.6%). The minority of the patients had Keros type III (right 8.1%, left 4.7%) (9). The results of that study did not differ significantly from the present study, where the majority of the patients fell under the Keros type II classification, followed by Keros type I, and a minority of the patients were classified as Keros type III.
Hence, the majority of the patients in this study are considered to be at a moderate risk, while a minority of the patients are at a high risk of iatrogenic injury because of ESS. Pearson correlation analysis showed a relationship between the depth of the right CP and the depth of the left CP, which suggests that there is a low level of asymmetry between the right and the left sides of the CP.

​

Conclusion

Based on the results of this study, risk stratification according to the Keros classification should be considered for presurgical workup since a majority of patients have a moderate risk of iatrogenic injury associated with ESS. The advent of cross-sectional imaging such as CT has rendered the radiomorphometric study of human anatomy both accurate and precise. Therefore, low-dose CT of the paranasal sinuses can be a viable tool for risk stratification using the Keros classification. It is also imperative for radiologists to include the depth of the CP and its Keros classification in their reports.

​

Study limitations
The study is not without limitations. Age and gender variables were not considered in this study; hence, it is difficult to draw conclusions on the sexual dimorphism of the depth of the CP.

​

Author contributions

MIK led in conceptualization, data curation, formal analysis, investigation, methodology, project administration, and in writing, reviewing & editing of the original draft. JW supported in each.

 

References

1. Cappabianca P, Villa A, Cavallo LM, et al. Endoscopic approaches to skull base lesions. In: Principles of neurological surgery. 4th ed. Elsevier. S Karger Ag,Basel,Switzerland.2018 

2. Gomez J, Pickup S. Cribiform plate fractures. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing. 2022. Available from https://www.ncbi.nlm.nih.gov/books/NBK562192.

3. Rueff-Barroso CR, Ferreira TM, Fernandes-Santos C, et al. Cribriform plate anatomical variations: a computed tomography study. J Morphol Sci. 2021; 38: 169-74. 10.51929/jms.38.30.

4. Baki A, Yıldız M, Cirik AA, et al. Effect of concha bullosa on skull base. Istanbul Med J. 2020; 21: 64-70. 10.4274/imj.galenos.2020.82335.

5. Mahdian M, Karbasi Kheir M. CBCT assessment of ethmoid roof variations through Keros, Gera, and TMS classifications. Int J Otolaryngol. 2022; 1-11. 10.1155/2022/3708851.

6. Oluigbo CU, Ngozi EC, Ohaegbu PMNI. Determination of sample sizes in research using Taro Yamane formula: an overview. 76th Proceedings of the 1st Annual Faculty of Science International Conference, Niger Delta University 2024 (AFSIC-NDU2024). Volume 1. Bayelsa State. Nigeria 2024. 

7. Abdullah B, Chew SC, Aziz ME, et al. A new radiological classification for the risk assessment of anterior skull base injury in endoscopic sinus surgery. Sci Rep. 2020; 10: 4600. 10.1038/s41598-020-61610-1.

8. Erdem G, Erdem T, Miman MC, et al. A radiological anatomic study of the cribriform plate compared with constant structures. Rhinology. 2004; 42: 225-9. 

9. Naidu L, Sibiya LA, Aladeyelu OS, et al. A computed tomography assessment of olfactory fossa depth in relation to functional endoscopic sinus surgery in a South African population. Transl Res Anat. 2022; 28: 100219. 10.1016/j.tria.2022.100219.