Published: 2023-05-16

Spatial analysis of transnasal olfactory cleft access: a computed tomography study

Division of Otolaryngology Head and Neck Surgery, Department of Surgery, University of British Columbia, St. Paul Sinus Center, Vancouver, BC, Canada; Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada
Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada
Division of Otolaryngology Head and Neck Surgery, Department of Surgery, University of British Columbia, St. Paul Sinus Center, Vancouver, BC, Canada
olfactory cleft trans cribriform intranasal drug application computed tomography study


Objective. To our knowledge, the spatial access of naris to olfactory cleft has not been quantified. We aimed to study the relationship and space of middle turbinate, septum, anterior nasal spine and cribriform plate to improve topical medication delivery and drug applicators.
Methods. One hundred CT scans of patients (50 males, 50 females) over the age of 18 were included. Subjects with radiographic sinonasal pathology, previous surgery, or specific variant nasal anatomy were excluded. Scans were independently reviewed and bilateral measurements on bony landmarks were taken by two blinded authors. Inter-rater reliability was analysed with intraclass correlation.
Results. The average age was 46.26 years (σ = 14.0). Average distance from the anterior nasal spine to olfactory cleft was 52.3 mm (σ = 4.2 mm), and the average length of cribriform plate was 18.8 mm (σ = 3.8) with an angle relative to hard palate averaging -8.8 degrees below parallel (σ = 5.5 degree). The widths of the olfactory cleft at anterior and posterior edges of cribriform plate were 2.3 mm (σ = 0.7 mm) and 2.0 mm (σ = 0.7 mm).
Conclusions. The findings suggest a 52.3 mm distance from the naris to the anterior border of cribriform plate. The average width along this path was 3.2 mm, suggesting devices narrower than this could potentiate direct drug delivery access.


Over the past decade, the intranasal (IN) drug delivery route has gained interest in research and development. IN drug administration has historically been used to topically treat symptoms of sinonasal conditions, such as chronic rhinosinusitis (CRS), by delivering drugs directly to the olfactory cleft (OC) via IN squeeze bottles, spray pumps, powered nebulisers and breath-powered bidirectional nasal devices 1. Recently, the trans-cribriform delivery route has become more prominent as an alternative approach to deliver drugs to the OC in the treatment of several conditions, including Parkinson’s Disease, Alzheimer’s Disease and epilepsy 2-4. This route allows medications to bypass the blood-brain barrier and directly enter the brain through olfactory nerves, which results in fewer systemic side effects and increased local drug activity 1,5. Despite these advances, effective deposition of IN drugs to the OC remains challenging due to its anatomic seclusion, and variations in sinonasal anatomy, devices, drug formulations and administration techniques 6,7.

Development of strategies to improve targeted application of IN drugs to the OC may result in trans-cribriform pharmaceuticals becoming more economically and physiologically accessible 12. To improve drug delivery into the OC, we first must understand the spatial relationships of sinonasal structures impeding access to the OC. A literature search on Web of Science, Embase, and PubMed was completed. To our knowledge, a study examining spatial distance and angle to the OC has not been published. Hence, the purpose of this study is to better characterise the OC relative to other sinonasal structures to improve topical medication delivery vehicles for the OC.



One hundred computed tomography (CT) sinus imaging scans of patients (50 males, 50 females) over the age of 18, with imaging cuts less than or equal to 1.3 mm were consecutively included into our study from the principal investigator’s clinical encounters, in reverse chronological fashion. This was done at a single institution in Vancouver, Canada. Scans were also required to have no reported or appreciable sinonasal pathology (i.e. polyps, CRS, tumours, malignancies), congenital deformities, past sinonasal surgery, or variant anatomy (concha bullosa, paradoxical middle turbinates, concha lamella).

CT scans which met eligibility criteria were reviewed by two authors (TC & ML) in a blinded fashion, and discrepancies were resolved through discussion and deliberation with senior author (AT). This was continued until 100 CT scans were selected.

Data extraction

Two authors (TC & ML) obtained two sets of bilateral measurements of the 100 selected CT scans in a blinded fashion in accordance with a pre-established measurement protocol (Appendix 1). Patient scans were reviewed on Phillips IntelliSpace PACS Enterprise (4.4.541.0), the institutional software, and measurements were made utilising the software localisation mode, ruler and angle tool. The coronal and sagittal CT sinus images were analysed to establish spatial relationships of bone structures pertaining to the access to the OC. The anterior nasal spine (ANS) was selected as the anterior bony landmark for the entrance into the nasal cavity. Measurements were made from ANS to anterior edge of bony middle turbinate, ANS to anterior edge of cribriform plate, length of cribriform plate, angle of cribriform plate relative to hard palate, width of anterior/posterior edge of OC and approximated narrowest space between middle turbinate and septum (Figs. 1-3).

Statistical analysis

All analyses were performed using SPSS 27. Inter-rater reliability between the authors TC & ML was assessed using intraclass correlation (ICC). Model selection was 2-way mixed effect model, type selection was mean of k raters. An official guideline was not found regarding interpretation of ICC coefficients, however, general consensus agree coefficients > 0.9 = excellent reliability, 0.75-0.89 = good reliability, 0.5-0.75 = fair reliability, < 0.5 = poor reliability 15. An independent t-test was utilised when testing for significant variations in sex-related average distances and angles pertaining to access to OC.



A total of 323 patient charts were consecutively reviewed, 100 of which were excluded due to history of past sinonasal surgery, on-going radiographic rhinosinusitis, or CT slice thickness > 1.3 mm. In a blinded fashion, two authors (TC and ML) reviewed 232 patient scans to screen for anatomic variants, on-going sinonasal disease, or abnormal masses or lesions.

CT scans that were included were obtained between 2016-2021. Overall, the mean age of patients was 46.3 ± 14.1 (range: 20-87). The baseline demographic of the males and females cohorts was similar, with mean age of men being 46.1 ± 15.0 years (range: 20-87) and females 46.4 ± 13.1 years (range: 24-74). Despite normal CT scan, symptoms most commonly reported by patients included obstruction (55%) and congestion (51%) (Tab. I).

Key findings

Measurements of 100 patients were collected in tandem by two authors TC & ML yielding 200 sets of measurements, two per patient.

The analysis revealed a mean population distance between the anterior nasal spine and anterior cribriform plate of 52.3 mm ± 4.2 with good interrater reliability, Cohen’s Kappa Coefficient (CKC) of 0.846. When performing a subanalysis of the cohorts, male subjects had a mean ANS to anterior CP distance of 54.3 mm while females had a mean distance of 50.4 mm (95% CI, p < 0.001) with a large effect size (Cohen’s D: -1.055). Males also had a significantly greater distance between the ANS and anterior head of the middle turbinate compared to females by 3.1 mm (95% CI, p < 0.001, Cohen’s D -0.9).

The mean length of the cribriform plate was 18.8 mm ± 3.8 mm with good interrater reliability (CKC = 0.730). Females had a mean length of 19.2 mm, while males had mean length of 18.4 mm. The difference of 0.8 mm was found to be significant (95% CI, p = 0.034), but Cohen D’s coefficient illustrated a small effect size at 0.213 as seen in Table II.

Mean population width of OC was 2.2 mm ± 0.6 with fair interrater reliability (CKC of 0.654). Females had a mean width of 2.2 mm and males a mean width of 2.1 mm. Females had a significant greater width of 0.1 mm (95% CI, p = 0.029) with an expected small effect size (Cohen’s D = 0.219).

A comprehensive compilation of other measurements characterising the access to OC can be found in Table III.


In this CT study, we characterised key anatomic distances and relationships impeding drug delivery to the olfactory cleft. The key aspects we wanted to explore were how far was the OC, how large this space is, and how difficult is it to reach the intended space.

When examining the mean distance in the population to the OC, the distance between anterior nasal spine and anterior edge of cribriform plate was 52.3 mm with a standard deviation of 4.2 mm. The sex specific subanalysis revealed that males had a 3.9 mm longer distance between ANS to anterior CP and likely represents more difficult application of drug to this space. This might be overcome with a cannulated drug applicator device with a longer cannula. Regarding the size of the targeted space, we found the average width of the OC to be 2.2 mm with a standard deviation of 0.6 mm. Comparison of our values with existing volumetric studies on the length and width of the OC had consistent findings 16. However, sex specific data was not previously compared, and we found females had a significantly wider OC by 0.1 mm. The clinical significance of this finding is to be determined. Lastly, regarding the difficulty reaching targeted space, we found the mean distance between middle turbinate and septum to be 3.2 mm with a standard deviation of 0.7 mm. Unsurprisingly, we also found a greater distance at the inferior aspects of the bony middle turbinate compared to superiorly near the OC. There was no sex specific difference found upon subanalysis. In our study, we also examined the slope of the OC relative to the hard palate, as we had to control for variability angulation and positioning of patient’s head during the CT study. With the assumption the hard palate represented our landmark for horizontal, we found that there was an average -8.8° angle of the OC in comparison with a standard deviation of 5.5°. Though this implies a convergent relationship in the general population, we found this was not the case in all individuals. A small minority of subjects had naturally divergent sinonasal anatomy.

Current available sinonasal medication delivery vehicles are not effective in OC drug deposition. In a study by Newman et al., 56% of aerosols from a hand-activated pump spray deposited in the anterior portion of the nasal cavity, and 44% deposited in the nasopharynx after 30 minutes 8. While one of the primary functions of the nasal cavity is to filter inhaled particles, it also prevents effective delivery of drugs to the OC 9. In inflammatory sinonasal pathologies, turbinate inflammation can result in distal obstruction, which prevents traditional intranasal products from dispersing medication to the OC 1. Common anatomic variations of the middle turbinate (MT), such as concha bullosa, concha lamella and paradoxical middle turbinates, can also serve as obstructions preventing particle distribution to the OC 10. Beyond the lack of clinical benefit, improper application and deposition of intranasal drugs may result in deleterious effects such as mucosal dryness, epistaxis, nasal crusting and drying which can negatively impact the quality of life 7.

Previous studies have been performed to optimise inhalation devices parameters and administration techniques to improve drug delivery to the OC. A study by Cheng and colleagues reported that larger nasal spray droplets and a wider spray plume angle increased the deposition in the anterior nose 11. Smaller droplet sizes and narrower plume angles yielded larger deposition past the nasal valve, increasing the probability of reaching the OC 11. Using nasal airway casts, Xi et al. showed that the majority of nasal spray droplets distributed in the anterior nose, while less than 4.6% reached the olfactory region 12. Head position has also been shown to affect particle deposition of intranasal medication in the nasal cavity. The Mygind technique recommends administration in the supine position with the head extended off the end of a bed, and turning the head to the right and left for 30 seconds in each direction 7. This technique was shown to distribute particles from the anterior to the middle turbinate in the lateral aspect of the nose, which is the region of maximal inflammatory response in sinonasal conditions 13. The Kaiteki position which recommends the lateral decubitus position with the neck and chin tilted upwards has also been recommended to increase particle distribution to the OC more than the upright position 14. While these positions promote better deposition patterns of IN drugs, they are often difficult to perform and can result in improper application technique.

The results of this study cannot be interpreted without considering its limitations. Sinonasal anatomic variation is highly prevalent among those with sinonasal conditions and this was seen in our study which was as evident by the high number of CT scans excluded due to anatomic variants, on-going sinonasal disease, or abnormal masses or lesions. As such, this presents as a particular challenge in characterising the OC relative to other sinonasal structures and developing effective standard IN drug delivery devices for the OC. Further study is needed to explore spatial access within these anatomic variants and how best to administer IN drugs to the OC. Moreover, while our study protocol included examination of the angle and distance of maximal septal deviation, unfortunately nasal trauma history was not routinely elucidated. As such, we could not delineate natural deviations versus deviations secondary to traumatic events. Realistically, this will have implications of introducing a cannulate drug delivery device to the OC. Lastly, the measurement protocol was based on bony landmarks as these structures are not subject to physiologic changes of mucosa, while it also increases inter-rater reliability. In the future, prospectively collected data set can be performed to account for this variable. In additions, inherent limitations of computed tomography studies are applicable to our study, as our views are limited to axial, coronal and sagittal planes when trajectories often take oblique paths. However, we believe this study is still adequate as a proof of concept to guide a drug delivery prototype design that would require in vivo trials to validate. Lastly, we acknowledge the narrowest distance between middle turbinate and septum is an approximation, as it is based on a single, consistently identified, coronal section. A more accurate approximation would require importing computational analysis of a rendered 3D model of patient’s skull to identify the narrowest spot along the oblique trajectory from ANS to OC.

With the characterisation of the access to OC, the next steps would be to utilise the present findings to create a more efficient drug delivery system to deliver medication to the cribriform plate. To date, squeeze bottles, spray pumps, powered nebulisers and breath-powered bidirectional nasal devices are the primary means of delivering IN medications 17. When focusing on drug deposition in the OC, the delivery methods currently employed are grossly inefficient. Xi et al. compared subsite specific deposition fractions with various market nasal sprays and nebulisers which revealed up to 4.59 and 9% deposition in upper olfactory regions, respectively with perfect applicator technique 12. In the context of trans-cribriform drug delivery, this would equate, at best, to 90% of drug depositing outside of the location of interest. Inefficiencies of this magnitude could impose prohibitively high costs of therapies as nanomedicine are often costly due to research and development cost as well as infrastructure requirements 18. For instance, amphotericin, an antifungal, has a 11-fold greater cost of delivery via a liposomal nanoparticulate drug delivery vehicle compared to traditional parenteral administration 19. As such, we believe optimisation of sinonasal drug delivery to the OC holds significant importance to increasing accessibility and affordability to trans-cribriform drug therapies.


Key findings in this study are the average population spatial measurements pertaining to access to the OC. We demonstrate there are sex specific differences in these dimensions which should be considered in drug deposition fluid dynamic computational models as well as prototypical drug delivery vehicles. As a first in a series of studies, we ultimately aim to design and develop a drug applicator that optimises efficiency of trans-cribriform drug delivery to advance our ability to capitalise on the nose-to-brain gateway.

Conflict of interest statement

The authors declare no conflict of interest.


This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions

TC: conceptualization, methodology, analysis, original draft preparation, review and editing; ML: methodology, analysis, original draft preparation, reviewing and editing; AT: conceptualization, methodology, originaly draft preparation, review and editing. The authors read and approved the final manuscript.

Ethical consideration

This study received ethics clearance by the University of British Columbia Providence Health Care Research Ethics Board (UBC PHC REB), certificate H21-00083.

The research was conducted ethically, with all study procedures being performed in accordance with the requirements of the World Medical Association’s Declaration of Helsinki.

Written informed consent was obtained from each participant/patient for study participation and data publication.

Figures and tables

Figure 1.Protocol for key measurements regarding Step 1.ANS was identified and marked on the midline image. Subsequently, the sagittal slice 1 slice lateral to each side, visualising the olfactory cleft, was selected to measure the distance between anterior CP and ANS. ANS: Anterior Nasal Spine. CP: Cribriform Plate.

Figure 2.Protocol for key measurements regarding Step 2.The coronal series was selected and the most anterior image visualising olfactory cleft was identified and the width was then measured at the superior most aspect.

Figure 3.Protocol for key measurements regarding Step 3.The anterior most image visualising contiguous middle turbinate attachment to skull base was selected. The distance between middle turbinate and septum of most superior, inferior and midline points were measured.(A) width at superior limit of bony middle turbinate attachment; (B) width at inferior limit of bony middle turbinate attachment; (C) total distance AB; (D) half distance between AB; (G) width at half distance between AB.

Presenting symptom
Congestion 51%
Pressure 2%
Facial pain 29%
Dysosmia 26%
Anterior nasal discharge 44%
Posterior nasal discharge 29%
Obstruction 55%
Headaches 2%
Table I.Demographic data.
Male Female Significance (2-tailed) 1 Cohen’s D
ANS to anterior CP 54.3 50.4 < 0.001 -1.055
ANS to anterior MT 36.2 33.1 < 0.001 -0.9
Anterior CP to posterior CP 18.4 19.2 0.034 0.213
Average width of OC 2.1 2.2 0.029 0.219
Average distance between MT & Septum 3.2 3.2 0.937 0.008
1 Independent t-test, equal variances not assumed. ANS: Anterior Nasal Spine; MT: Middle Turbinate; CP: Cribriform Plate.
Table II.Gender comparison for average measurements (mm).
Average Standard deviation Max Min ICC: intraclass correlation
ANS to anterior MT 34.7 mm 3.7 mm 46.9 mm 24.4 mm 0.955
ANS to anterior CP 52.3 mm 4.2 mm 66.9 mm 29.6 mm 0.846
Anterior CP to posterior CP 18.8 mm 3.8 mm 34.5 mm 10.1 mm 0.730
Angle CP relative to hard palate - 8.8° 5.5° 16.2° -24.1° 0.572
Width of anterior CP 2.3 mm 0.7 mm 6.6 mm 0.8 mm 0.654
Width of posterior CP 2.0 mm 0.7 mm 4.8 mm 0.4 mm 0.544
Average width of OC 2.2 mm 0.6 mm 4.0 mm 0.6 mm 0.608
Superior aspect
Distance between MT & septum 2.1 mm 0.6 mm 8.6 mm 0.8 mm
Middle aspect
Distance between MT & septum 2.6 mm 0.9 mm 5.5 mm 0.4 mm
Inferior aspect
Distance between MT & septum 5.0 mm 1.4 mm 10.7 mm 0.8 mm
Average distance between MT & septum 3.2 mm 0.7 mm 5.5 mm 1.7 mm 0.663
ANS: Anterior Nasal Spine; MT: Middle Turbinate; CP: Cribriform Plate; °: degrees.
Table III.Measurements characterising trajectory to olfactory cleft.


  1. Schiebe M, Bethge C, Witt M. Intranasal administration of medications. Orthop Nurs. 1992; 11:46-47. DOI
  2. Espinoza LC, Silva-Abreu M, Clares B. Formulation strategies to improve nose-to-brain delivery of donepezil. Pharmaceutics. 2019; 11:64. DOI
  3. Musumeci T, Bonaccorso A, Puglisi G. Epilepsy disease and nose-to-brain delivery of polymeric nanoparticles: an overview. Pharmaceutics. 2019; 11:118. DOI
  4. Tzeyung AS, Md S, Bhattamisra SK. Fabrication, optimization, and evaluation of rotigotine-loaded chitosan nanoparticles for nose-to-brain delivery. Pharmaceutics. 2019; 11:26. DOI
  5. Crowe TP, Greenlee MHW, Kanthasamy AG. Mechanism of intranasal drug delivery directly to the brain. Life Sci. 2018; 195:44-52. DOI
  6. Dalgorf DM, Harvey RJ. Sinonasal anatomy and function. Am J Rhinol Allergy. 2013; 27:3-6. DOI
  7. Benninger MS, Hadley JA, Osguthorpe JD. Techniques of intranasal steroid use. Otolaryngol Head Neck Surg. 2004; 130:5-24. DOI
  8. Newman SP, Morén F, Clarke SW. Deposition pattern of nasal sprays in man. Rhinology. 1988; 26:111-120.
  9. Sedaghat AR. Chronic rhinosinusitis. Am Fam Physician. 2017; 96:500-506.
  10. Al-Sebeih KH, Bu-Abbas MH. Concha bullosa mucocele and mucopyocele: a series of 4 cases. Ear Nose Throat J. 2014; 93:28-31. DOI
  11. Cheng YS, Holmes TD, Gao J. Characterization of nasal spray pumps and deposition pattern in a replica of the human nasal airway. J Aerosol Med. 2001; 14:267-280. DOI
  12. Xi J, Yuan JE, Zhang Y. Visualization and quantification of nasal and olfactory deposition in a sectional adult nasal airway cast. Pharm Res. 2016; 33:1527-1541. DOI
  13. Raghavan U, Logan BM. New method for the effective instillation of nasal drops. J Laryngol Otol. 2000; 114:456-459. DOI
  14. Milk DG, Khong GC, Çam OH. A comparison between Mygind and Kaiteki positions in administration of drops to the olfactory cleft. Clin Otolaryngol. 2021; 46:406-411. DOI
  15. Koo TK, Li MY. A Guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016; 15:155-163. DOI
  16. Altundag A, Temirbekov D, Haci C. Olfactory cleft width and volume: possible risk factors for postinfectious olfactory dysfunction. Laryngoscope. 2021; 131:5-9. DOI
  17. Maaz A, Blagbrough IS, Bank PA De. In vitro evaluation of nasal aerosol depositions: an insight for direct nose to brain drug delivery. Pharmaceutics. 2021; 13:1079. DOI
  18. Bossetti R, Jones SL. Cost-effectiveness of nanomedicine: estimating the real size of nano-costs. Nanomedicine (Lond). 2019; 14:1367-1370. DOI
  19. Cagnoni PJ, Walsh TJ, Prendergast MM. Pharmacoeconomic analysis of liposomal amphotericin B versus conventional amphotericin B in the empirical treatment of persistently febrile neutropenic patients. J Clin Oncol. 2000; 18:2476-2483. DOI


Teffran Joey Chan

Division of Otolaryngology Head and Neck Surgery, Department of Surgery, University of British Columbia, St. Paul Sinus Center, Vancouver, BC, Canada; Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada

Melissa Lee

Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada

Andrew Vernu Thamboo

Division of Otolaryngology Head and Neck Surgery, Department of Surgery, University of British Columbia, St. Paul Sinus Center, Vancouver, BC, Canada


© Società Italiana di Otorinolaringoiatria e chirurgia cervico facciale , 2023

  • Abstract viewed - 351 times
  • PDF downloaded - 111 times
  • Supplementary file downloaded - 2 times