PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery 2322 PhiliPPine Journal of otolaryngology-head and neck Surgery ORIGINAL ARTICLES Philipp J Otolaryngol Head Neck Surg 2023; 38 (1): 22-27 c Philippine Society of Otolaryngology – Head and Neck Surgery, Inc. Sinonasal Anatomy Variations on CT Scans of a Sample of Filipino Adults With Chronic Rhinosinusitis Catherine P. Policina, MD Giancarla Marie C. Ambrocio, MD Rodante A. Roldan, MD, MHA Precious Eunice R. Grullo, MD, MPH, MSc Department of Otorhinolaryngology Head and Neck Surgery Rizal Medical Center Correspondence: Dr. Giancarla Marie C. Ambrocio Department of Otorhinolaryngology Head and Neck Surgery Rizal Medical Center, Pasig Blvd, Pasig 1600 Philippines Phone: +63 998 550 5409 Email: giancarlamarie.ambrocio@rmc.doh.gov.ph The authors declared that this represents original material, that the manuscript has been read and approved by the authors, that the requirements for authorship have been met by each author, and that the authors believe that the manuscript represents honest work. Disclosure: The authors signed disclosures that there are no financial or other (including personal) relationships, intellectual passion, political or religious beliefs, and institutional affiliations that might lead to a conflict of interest. ABSTRACT Objective: To determine the prevalence of sinonasal anatomic variations seen on paranasal sinus (PNS) CT scans of a sample of Filipino adults with chronic rhinosinusitis. Methods: Design: Cross-sectional study Setting: Tertiary Government Training Hospital Participants: The PNS CT scans of 51 Filipino patients with chronic rhinosinusitis with and without nasal polyposis diagnosed at our outpatient Department of Otorhinolaryngology-Head and Neck Surgery between October 2015 to December 2020 were reviewed for the presence of sinonasal anatomic variants. The prevalence of the identified variants was calculated. Results: The CT scans of 51 patients, 41 (80.4%) men and 10 (19.6%) women, were included. The median age was 48 years (Q25: 35, Q75: 56, IQR:21). The median Lund Mackay Score (LMS) was 15 (Q25: 12, Q75: 20, IQR:8). Majority (94%) had an LMS of ≥5. The most common anatomic variant in the study population was agger nasi (n=46/51, 90.2% present bilaterally) followed by uncinate process attachment to the lamina papyracea (n=90/102, 88.24%). The third to sixth most common findings were Keros type II classification (n=76/102, 74.51%), nasal septal deviation (n=35/51, 68.62%), optic nerve canal type 1 (n=67/102, 65.69%) and anterior ethmoid artery grade 1 (n=46/102, 45.1%), respectively. Less common variants were Onodi cell (n=13/51, 25.49% unilateral and n=10/51, 19.61% bilateral), Haller cell (n=8/51, 15.69% unilateral and n=1/51, 1.96% bilateral), supraorbital cell (n=4/51, 7.84% unilateral and n=4/51, 7.84% bilateral), middle turbinate concha bullosa (n=3/51, 5.88% unilateral and n=6/51, 11.76% bilateral), superior turbinate concha bullosa (n=2/51, 3.92% unilateral and n=1/51, 1.96% bilateral), pneumatized crista galli (n=2/51, 3.92%) and optic nerve dehiscence (n=1/51, 1.96% bilateral). Conclusion: In the adult Filipino population with CRS sampled in this study, the six most common sinonasal anatomic variants were agger nasi, superior attachment of the uncinate process to the lamina papyracea, Keros type II classification, septal deviation, optic nerve canal type 1 and anterior ethmoid artery grade 1. Pre-operatively, the PNS CT scan of every patient must be meticulously evaluated for the sinonasal anatomic variants to avoid surgical complications. Keywords: chronic rhinosinusitis; paranasal sinus CT; anatomic variants Creative Commons (CC BY-NC-ND 4.0) Attribution - NonCommercial - NoDerivatives 4.0 International PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery 2322 PhiliPPine Journal of otolaryngology-head and neck Surgery ORIGINAL ARTICLES The management for chronic rhinosinusitis (CRS) includes medical treatment. However, those who have failed medical therapy will subsequently undergo surgery.1 According to the European Position Paper (EPOS) on Rhinosinusitis and Nasal Polyps 2020, a paranasal sinus (PNS) computed tomography (CT) scan is required prior to undergoing Endoscopic Sinus Surgery (ESS) to determine both the extent of the disease and the sinonasal anatomic variants that may increase the risk of surgical complications.2 Thus, knowledge of these radiologic anatomic variants is important especially because they differ individually and between ethnic groups. A local study that included PNS CT scans of 45 patients with presumably sinonasal symptoms found out that the most common anatomic variation was the agger nasi cell.3 In 2018, the group of Espinosa determined the prevalence of six anatomic variations, namely agger nasi (AN), Haller cell, septal deviation, concha bullosa, everted uncinate process and paradoxical middle turbinate, of the nasal cavity and paranasal sinuses in patients with chronic rhinosinusitis.4 Acosta and Vicente, on the other hand, described other sinonasal variants such as asymmetric ethmoid roof, deep olfactory fossa, Onodi cells and dehiscent lamina papyracea in patients with nasal polyposis only.5 A search in PubMed Central, Medline, Google Scholar, and HERDIN revealed a lack of published local data that investigates all the other possible sinonasal anatomic variants in CRS patients with or without nasal polyposis such as uncinate process attachment, anterior ethmoid artery grading types, presence or absence of frontal cells, turbinate pneumatization, carotid artery and optic nerve dehiscence, which can have an impact in the surgical management. Hence, this study aimed to identify the prevalence of radiologic sinonasal anatomic variants in the PNS CT scans of a sample of adult Filipino patients with CRS with and without nasal polyposis. It is important for the surgeon to determine all the anatomic variants because failure to do such may lead to complications (i.e. injury to critical structures such as the orbit, skull base, internal carotid artery and optic nerve). As part of describing the PNS CT scans, this study also determined the severity of CRS radiologically using the Lund-Mackay staging system. METHODS With Rizal Medical Center Institutional Review Board Approval (2021-ORL-#020-RP-1.IV), this cross-sectional study considered for inclusion all CT scans of adult Filipino patients with CRS with and without nasal polyposis, as defined by the criteria in the Clinical Practice Guidelines of the Philippine Society of Otolaryngology – Head and Neck Surgery.1 Patients were those diagnosed at the outpatient department of Otorhinolaryngology - Head and Neck Surgery (ORL-HNS) from October 2015 to December 2020 whose CT scans were available at the Department of Radiology, Section of Computed Tomography. Scans of those with prior sinus surgery or craniofacial anomalies were excluded. The PNS CT scan images were taken using the CT scan machines GE Healthcare Revolution ACT with slice thickness of 1.25mm and Philips MX 16-slice with slice thickness of 2mm, and were hosted in the PACS system of the Department of Radiology. The slice thickness of images was set at ≤2 mm intervals. A single number code was assigned to each CT scan DICOM file to anonymize the patient’s identity. All CT scans were independently evaluated and interpreted by each author. The multiplanar views (axial, coronal and sagittal planes) of the imaging studies were evaluated in bone window (width 2000, level 350)6 using Horos or Radiant DICOM medical image viewer. The authors used their laptops to evaluate the CT scans. The laptops satisfied the display requirements for clinical review work, which means that the viewing and interpretation of images were carried out to influence clinical management and not to generate a formal radiology report.7 The Lund-Mackay CT Score (LMS) was first obtained by scoring the 1); opacification of 5 bilateral sinuses (frontal, maxillary, anterior ethmoid, posterior ethmoid, sphenoid); and 2) obstruction of the ostiomeatal complex. Possible range of total Lund-Mackay score was between 0 and 24. The presence or absence of the anatomic variants were identified in each imaging study: Keros classification (I, II, III) – height of the lamella or depth of the olfactory fossa;8 frontal sinus cells (K1, K2, K3, K4);9 frontal sinus outflow tract obstruction; hypoplastic frontal sinus; interfrontal sinus septal cell; agger nasi cell; supraorbital ethmoidal cell; suprabullar cell; infraorbital ethmoidal cell or Haller cell; sphenoethmoidal cell or Onodi cell; dehiscence of lamina papyracea (LP); pneumatized crista galli, pneumatization of superior/middle/inferior turbinate, nasal septum; paradoxic middle turbinate; nasal septal deviation (measured by drawing a line from the superior insertion of the nasal septum at the crista galli to the inferior insertion at the level of the anterior nasal spine) classified as absent (nasoseptal angle 4 degrees and below) or present (nasoseptal angle 5 degrees and above);10 septal spur; uncinate process attachment; sinus lateralis; grading of anterior ethmoid artery (AEA): (Grade 1 – AEA courses within skull base, Grade 2 – AEA adjacent but inferior to, described as prominent from, inferior surface of the skull base, Grade 3 – AEA courses freely inferior to skull base);11 types of sphenoid sinus pneumatization: conchal (region below sella completely ossified and consists of a solid block of bone with no air cavity), presellar (air cavity does not penetrate beyond a vertical plane parallel to anterior sellar wall), incomplete sella (sinus is well developed, and pneumatization extends beyond tuberculum sella below the sella and sometimes with bulging of sellar floor into sinus cavity), complete PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery 2524 PhiliPPine Journal of otolaryngology-head and neck Surgery ORIGINAL ARTICLES sella (air cavity extends into body of sphenoid, continues beyond posterior margin of dorsum sella into clivus bone);12 sphenoid sinus wall dehiscence in relation to optic nerve and internal carotid artery; and types of optic nerve canals (ONC): Type 1 (canal superolateral to sphenoid sinus without indentation on sinus wall on coronal CT sections), Type 2 (indentation on sphenoid sinus contour with less than 50% protrusion of nerve circumference on coronal view), Type 3 (optic nerve traversing through sphenoid sinus with coronal view showing more than 50% protrusion of nerve circumference into sinus), Type 4 (canal adjacent to sphenoid and posterior ethmoid sinuses/presence of Onodi cell).13 The sample size was calculated using the formula n = N*X / (X + N – 1), where, X = Z α/2 2 *p*(1-p) / MOE 2, and Z α/2 is the critical value of the normal distribution at α/2, MOE is the margin of error, p is the sample proportion, and N is the population size. Using a margin of error of 5%, confidence level at 99.1%, population set at 1000 and sample proportion of 98% based on a study by Shpilberg,14 the minimum sample size calculated was 51 CT scan images. The results were recorded using a data collection form. In case of disagreements in interpretation, the authors convened, and the decision was made by consensus (i.e. at least 3 out of the 4 authors agreed). If consensus was not reached, the interpretation of the most senior (and most experienced) rhinologist (RAR) was deemed final. All collected data were tabulated in Microsoft Excel version 16.71 (Microsoft Corp., Redmond, WA, USA). The frequencies or proportions were summarized in appropriately labeled tables. The median, 1st quartile, 3rd quartile and interquartile range (IQR) were reported for age and LMS because of non-normal distribution of age and the ordinal nature of LMS data. For statistical analysis, Python version 3.11.0 (Python Software Foundation, Beaverton, OR, USA) was used. RESULTS Out of 53 PNS CT scans initially retrieved, 2 were excluded (1 with craniofacial anomaly, 1 with previous sinus surgery), and the CT scans of 51 patients, 41 (80.4%) men and 10 (19.6%) women, were finally included in this study. The median age was 48 years (Q25: 35, Q75: 56, IQR:21). The Lund Mackay scores (LMS) ranged from 0 to 24 with a median score of 15 (Q25: 12, Q75: 20, IQR:8). Among the participants, 94% had an LMS of ≥5. Table 1 shows the distribution of sinonasal anatomic variants from the right and left side of each patient’s nasal cavity (n=102), while Table 2 shows the anatomic variations that were either present unilaterally or bilaterally (n=51). The most common anatomic variant in this study was AN (n=46/51, 90.2% present bilaterally and n=1/51, 1.96% present unilaterally). The second most common variant was uncinate process attachment to the lamina papyracea (n=90/102, 88.24%). Keros type II classification (n=76/102, 74.51%) was the third most common variant noted, followed by nasal septal deviation (n=35/51, 68.62%), optic nerve canal type 1 (n=67/102, 65.69%) and anterior ethmoid artery grade 1 (n=46/102, 45.1%). Less common variants were Onodi cell (n=13/51, 25.49% unilateral and n=10/51, 19.61% bilateral), Haller cell (n=8/51, 15.69% unilateral and n=1/51, 1.96% bilateral), supraorbital cell (n=4/51, 7.84% unilateral and n=4/51, 7.84% bilateral) and middle turbinate concha bullosa (n=3/51, 5.88% unilateral and n=6/51, 11.76% bilateral). Some anatomic variants were rarely seen in this study. Only three patients had superior turbinate concha bullosa (n=2/51, 3.92% unilateral and n=1/51, 1.96% bilateral). Two patients had pneumatized crista galli (n=2/51, 3.92%), and only one had optic nerve dehiscence (n=1/51, 1.96% present bilaterally). None had lamina papyracea dehiscence, carotid artery dehiscence, paradoxic middle turbinate, inferior turbinate concha bullosa and pneumatized nasal septum. DISCUSSION Our study found that the majority (94%) of patients had a Lund- Mackay Score (LMS) of ≥5. This score has an excellent positive predictive value, which strongly indicates true disease in patients according to EPOS 2020.2 It is interesting to note that 6% had an LMS of less than 5. This finding of having a low LMS in CRS patients is supported by the study of Dietz de Loos et al.2 In their investigation, 26% (n=107) of patients who were diagnosed with CRS based on nasal symptoms only had an LMS of 1 to 3 in their CT scans.2 They also found out that LMS can be 0 in 50% (n=107) of CRS patients.2 Overall, AN was the most common anatomic variation, which was present bilaterally in 90.2% (n=46) patients and unilaterally in 1.96% (n=1) patient. Similarly, in the other three studies on Filipino patients, AN was also the most common variant detected with a prevalence of 86% (n=88), 62% (n=28) and 78% (n=47), respectively.3,4,15 The presence of this variant helps the surgeon to access the frontal recess faster. However, its pneumatization must be carefully evaluated. A well- pneumatized AN may be mistaken for the frontal recess.13 Removal of an extensively pneumatized AN in a patient with a thin lacrimal bone may increase the risk of orbital injury.13 In the case of an overly pneumatized AN displacing the middle turbinate more medially, resection must be executed with caution to prevent destabilization of the middle turbinate.13 More important than determining the presence of the AN is the thorough assessment of its relationship to the surrounding critical structures to prevent complications. PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery 2524 PhiliPPine Journal of otolaryngology-head and neck Surgery ORIGINAL ARTICLES complications like cerebrospinal fluid (CSF) leak. The third most prevalent variant was the presence of a type II Keros classification (n=76/102, 74.51%). This was in contrast with the local study done by Paber et al. where the most common identified classification was type I Keros (n=165, 81.6%), and only 17.9% (n=52) had type II Keros.17 This varying distribution among Filipinos means that surgeons must be attentive to each patient’s depth of olfactory fossa in each side to avoid skull base and intracranial injuries. It was also notable that Keros type III was the least prevalent in this study which was found in only 1.96% (n=2) patients. This very low prevalence was consistent with the local study of Paber17 and other studies conducted among the Saudi, Indian and Egyptian populations18,19 suggesting that although it was the most dangerous classification, it was uncommon. Table 1. Distribution of right and left sinonasal anatomic variants ClassificationAnatomic variant Right n=51 n (%) Left n=51 n (%) Total n=102 for 2 sides n (%) Keros classification Frontal (Kuhn) cell Uncinate process attachment Anterior ethmoid artery Sphenoid sinus pneumatization Optic nerve canal Septal deviation Septal spur Type I Type II Type III None K1 K2 K3 K4 Lamina papyracea Middle turbinate Skull base Grade 1 Grade 2 Grade 3 Conchal Presellar Incomplete Complete Type 1 Type 2 Type 3 Type 4 Present Present 12 (23.5) 39 (76.4) 0 37 (72.55) 4 (7.84) 0 5 (9.8) 5 (9.8) 44 (86.27) 7 (13.73) 0 23 (45.1) 8 (15.69) 20 (39.22) 3 (5.88) 14 (27.45) 23 (45.1) 11 (21.57) 34 (66.67) 1 (1.96) 1 (1.96) 15 (29.41) 30 (58.82) 7 (13.73) 12 (23.53) 37 (72.55) 2 (3.92) 34 (66.67) 7 (13.73) 0 7 (13.73) 3 (5.88) 46 (90.2) 5 (9.8) 0 23 (45.1) 4 (7.84) 24 (47.06) 4 (7.84) 10 (19.61) 21 (41.18) 16 (31.37) 33 (64.71) 3 (5.88) 3 (5.88) 12 (23.53) 5 (9.8) 0 24 (23.53) 76 (74.51) 2 (1.96) 71 (69.61) 11 (10.78) 0 14 (13.73) 8 (7.84) 90 (88.24) 12 (11.76) 0 46 (45.1) 12 (11.76) 44 (43.14) 7 (6.86) 24 (23.53) 44 (43.14) 27 (26.47) 67 (65.69) 4 (3.92) 4 (3.92) 27 (26.47) n=51 n (%) 35 (68.62) 7 (13.73) Next to agger nasi in terms of prevalence was the uncinate process (UP) superior attachment to the lamina papyracea (n=90/102, 88.24%). This finding was consistent with the study by Tuli et al. wherein 79.8% (n=67) of the study participants had the UP attached to the lamina papyracea.16 In this study, none of the UP was attached to the skull base, while the rest was noted to have superior attachment to the middle turbinate (n=12/102, 11.76%). This was in contrast to the study of Tuli et al. where the least common attachment was to the middle turbinate (n=3, 3.57%).16 These findings suggest that careful uncinectomy must be done to avoid inadvertent injury to the lamina papyracea where it most commonly attaches to. However infrequent, the attachment of the uncinate to the skull base and middle turbinate must still be identified pre-operatively to prevent iatrogenic damage that can cause Table 2. Distribution of unilateral or bilateral sinonasal anatomic variants Frontal sinus Frontal sinus outflow tract obstruction Interfrontal sinus septal cell Hypoplastic frontal sinus Ethmoid sinus and related anatomy Haller cell Supraorbital cell Agger nasi Suprabullar cell Sinus lateralis Onodi cell Lamina papyracea dehiscence Sphenoid sinus and related anatomy Intersphenoid septa Carotid dehiscence Optic nerve dehiscence Turbinates Paradoxic middle turbinate Superior turbinate pneumatization Middle turbinate pneumatization Inferior turbinate pneumatization Crista galli Pneumatized crista galli Nasal septum Pneumatized septum 13 (25.49) 6 (11.76) 7 (13.73) 8 (15.69) 4 (7.84) 1 (1.96) 13 (25.49) 8 (15.69) 13 (25.49) 0 13 (25.49) 0 0 0 2 (3.92) 3 (5.88) 0 28 (54.9) 0 1 (1.96) 1 (1.96) 4 (7.84) 46 (90.2) 16 (31.37) 19 (37.25) 10 (19.61) 0 1 (1.96) 0 1 (1.96) 0 1 (1.96) 6 (11.76) 0 Present 2 (3.92) Present 0 Anatomic variant Unilateral n (%) Bilateral n (%) PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery 2726 PhiliPPine Journal of otolaryngology-head and neck Surgery ORIGINAL ARTICLES Next, the fourth most frequently seen variant in this study was septal deviation, which was present in 68.62% (n=35, more commonly found in the right side n=30). This was in contrast to the findings of other anatomic studies.14,20 Kaygusuz et al. reported that in the Turkish population, the most common anatomic variant was nasal septal deviation (72.7%, n=72).20 In addition, Shpilberg et al. reported that among the population in New York in their investigation, nasal septal deviation was also the most common variant with a prevalence of 98.4% (n=189).14 The slightly lower prevalence among Filipinos may mean that challenging surgical exposure in endoscopic sinus surgeries secondary to an obstructing deviated septum may be a less likely problem. Septal spur may also cause obstruction, but it was not that common in our study population (n=7, 13.73%). Type 1 optic nerve canal was the fifth most common anatomic variant. Among the other types of optic nerve canal, the second most common was type 4 (n=27, 26.47%). Only one patient had optic nerve dehiscence. Onodi cells were present in 23 patients (n=13, 25.49% unilateral and n=10, 19.61% bilateral). This study reported a higher prevalence of Onodi cells compared to other studies involving Filipino participants where the prevalence ranged from 8.2%-8.8%.3,5 The presence of these variants warrants vigilance when approaching the posterior ethmoid and sphenoid sinuses to prevent damage to the optic nerve, which can ultimately cause blindness. The sixth most common variant was grade 1 anterior ethmoid artery (AEA). In this study, grade 1 and grade 3 AEA had close prevalence, 45.1% (n=46) and 43.14% (n=44), respectively. The symmetry of the grading on each patient’s side did not largely deviate from each other. These results were similar to the study of Taha et al. wherein 45% of the patients had grade 1 AEA, and another 45% of the patients had grade 3 AEA, with the prevalence of the right and left side not far from each other.21 From these findings, the AEA can commonly hang below the skull base, hence, extra caution when performing ethmoidectomy is recommended. Inability to recognize the AEA particularly when it travels freely from the skull base can cause unintentional transection of the artery. Its subsequent retraction to the orbit can lead to retro- orbital hemorrhage, which may progress to optic nerve compression and blindness. Other anatomic variants had lower prevalence in this study. Only 9 patients (17%) had Haller cells. In the local studies available, the prevalence of Haller cells was higher ranging from 28% to 41.6%.3,4,15 From the analysis of Badia, which tabulated results of 10 studies including Asian and Caucasian populations, the prevalence of Haller cells ranged from 3% to 46%.22 Another infrequent variant in our study was the supraorbital cell (n=8, 15.6%). None of the local publications described its prevalence among Filipinos. From Shpilberg et al.14 and Gouripur et al.,23 supraorbital cells were present in 28.1% and 26% of patients, respectively. Regardless of the lower prevalence observed in this study, the presence of Haller cells and supraorbital cells should be noted prior to the surgery as they can increase the risk of injury to the orbit. A retained Haller cell may also have an impact on the drainage of the maxillary sinus. Dehiscence in the lamina papyracea (LP) and bone through which the carotid artery runs in relation to the sphenoid sinus are other important variants to evaluate. Further violation of a dehisced LP can lead to prolapse of or damage to the orbital contents. Aggressive clearance of secretions from the sphenoid sinus may result in massive hemorrhage for patients with carotid canal dehiscence. Likewise, sphenoid sinus septations that are connected to the bone overlying the carotid may increase the risk for bleeding during removal. In this present study, none of the subjects had a dehiscence in the LP or bony canal of the carotid artery. Meanwhile, in the study of Santos and Jarin, 37% (n=18) of the Filipino participants had LP dehiscence.15 Lastly, pneumatization of the nasal turbinates is also important to assess. The presence of such can either be a contributing factor to the pathophysiology of chronic rhinosinusitis or may affect the field exposure in sinus surgeries. In this analysis, middle turbinate concha bullosa was present in nine patients (17.6%). This prevalence was similar to the study of Santos et al. (n=9, 20%).3 Interestingly, Badia et al. cited a study where the prevalence of concha bullosa among Caucasians was as high as 53%.22 Superior turbinate pneumatization was rare in this present study, only presenting in three patients (n=2, 3.92% unilateral and n=1, 1.96% bilateral). None of the patients had a pneumatized inferior turbinate. To date, this study has the most number of sinonasal anatomic variants evaluated from the PNS CT scans of Filipinos with CRS. However, it also has several limitations. First, the windowing used in this review was fixed to one setting only. Second, the readers only used laptops instead of primary diagnostic workstations. These factors could have affected the visibility of some anatomic variants and interrater variability. The use of different bone window levels and diagnostic display monitors is suggested in the future. Further multicenter studies with a larger sample size is another recommendation to increase the generalizability of the results. The different ethnicities of Filipinos (e.g., Chinese, Malay, Caucasoid and combinations) can also be captured and analyzed. In the Filipino sample included in this study, the six most common sinonasal anatomic variants were agger nasi, superior attachment of the uncinate process to the lamina papyracea, Keros type II classification, septal deviation, optic nerve canal type 1 and anterior ethmoid artery grade 1. The prevalence of some anatomic variants differ among PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery Vol. 38 no. 1 January - June 2023 PhiliPPine Journal of otolaryngology-head and neck Surgery 2726 PhiliPPine Journal of otolaryngology-head and neck Surgery ORIGINAL ARTICLES REFERENCES 1. Caro R, Acuin J, Villegas M, Llanes EG, Calaquian C, Dumlao KJ. Clinical Practice Guidelines on Chronic Rhinosinusitis in Adults. Philippine Society of Otolaryngology – Head and Neck Surgery. 2016. [cited 2021 Jan 10] Retrieved from https://pso-hns.org/clinical-practice-guidelines-v2/. 2. Fokkens WJ, Lund VJ, Hopkins C, Hellings PW, Kern R, Reitsma S. et al. European Position Paper on Rhinosinusitis and Nasal Polyps 2020. Rhinology. 2020 Feb 20;58(Suppl S29):1-464. DOI: 10.4193/Rhin20.600; PubMed PMID: 32077450. 3. Santos DR, Fragante RE, Espina CS, Hernandez JG. A Descriptive Study of Anatomic Variants of Ethmoid Sinus: Agger Nassi Cell, Haller Cell, Onodi Cell, Concha Bullosa, and Paradoxic Middle Turbinate as seen on CT Scans of Paranasal Sinuses in Filipino Subjects. Radiology Journal. 2009 Jul;3(1):47-51. 4. Espinosa W, Genito R, Ramos RZ. Anatomic variations of the nasal cavity and paranasal sinus and their correlation with chronic rhinosinusitis using Harvard staging system. J Otolaryngol ENT Res. 2018 Jul 6;10(4):190-193. DOI: 10.15406/joentr.2018.10.00343. 5. Acosta ER, Vicente G. Risk areas in endoscopic sinus surgery: A review of 175 computed tomography scans of the sinuses among patients with nasal polyposis. Philipp J Otolaryngol Head Neck Surg. 2001;16(3):118-125. [cited 2021 Jan 10] Available from: https://pjohns.pso-hns. org/index.php/pjohns/issue/view/61. 6. Cebula M, Danielak-Nowak M, Modlinska S. Impact of Window Computed Tomography (CT) Parameters on Measurement of Inflammatory Changes in Paranasal Sinuses. Pol J Radiol. 2017 Oct 20;82:567-570. DOI:  10.12659/PJR.901939; PubMed PMID:  29662587; PubMed Central PMCID; PMC5894057. 7. Radiology Informatics Committee. Picture archiving and communication systems (PACS) and guidelines on diagnostic display devices Third Edition. The Royal College of Radiologists. 2019 February. [cited 2022 Jan 17] Available from: https://www.rcr.ac.uk/publication/picture- archiving-and-communication-systems-pacs-and-guidelines-diagnostic-display-0. 8. Babu AC, Nair MRPB, Kuriakose AM. Olfactory fossa depth: CT analysis of 1200 patients. Indian J Radiol Imaging. 2018 Oct-Dec;28(4):395-400. DOI:  10.4103/ijri.IJRI_119_18; PubMed PMID: 30662198 PubMed Central PMCID: PMC6319094. 9. Lal D, Stankiewicz J. Primary Sinus Surgery in Flint P, Francis H, Haughey B, Lesperance M, Lund V, Robbins K, Thomas J. Cummings Otolaryngology Head and Neck Surgery. 7th edition, Philadelphia; Elsevier, 2020. Pp 677-710. 10. Periyasamy V, Bhat S, Sree Ram MN. Classification of Naso Septal Deviation Angle and its Clinical Implications: A CT Scan Imaging Study of Palakkad Population, India.  Indian J Otolaryngol Head Neck Surg. 2019 Nov;71(Suppl 3):2004-2010. DOI:  10.1007/s12070-018-1425-1; PubMed PMID: 31763284; PubMed Central PMCID: PMC6848586. 11. Taha MA, Hall CA, Zylics HE, Westbrook MB, Barham WT, Stevenson MM, et al. Variability of the Anterior Ethmoid Artery in Endoscopic Sinus Surgery. Ear Nose Throat J. 2022 May;101(4):268- 273. DOI: 10.1177/0145561320950488; PubMed PMID: 32845806. 12. Famurewa OC, Ibitoye BO, Ameye SA, Asaleye CM, Ayoola OO, Onigbinde OS. Sphenoid Sinus Pneumatization, Septation, and the Internal Carotid Artery: A Computed Tomography Study.  Niger Med J. 2018 Jan-Feb;59(1):7-18. DOI: 10.4103/nmj.NMJ_138_18; PubMed PMID: 31198272; PubMed Central PMCID: PMC6561078. 13. Yanagisawa E, Joe JK. The surgical significance of the agger nasi cell. Ear Nose Throat J. 1999 May;78(5):328-30. PubMed Central PMID: 10355191. 14. Shpilberg KA, Daniel SC, Doshi AH, Lawson W, Som PM. CT of Anatomic Variants of the Paranasal Sinuses and Nasal Cavity: Poor Correlation With Radiologically Significant Rhinosinusitis but Importance in Surgical Planning. AJR Am J Roentgenol. 2015 Jun;204(6):1255-60. DOI: 10.2214/ AJR.14.13762; PubMed PMID: 26001236. 15. Santos C, Jarin PS. Computed tomographic analysis of paranasal sinus anatomic variations among Filipinos. Philipp J Otolaryngol Head Neck Surg. 2004;19(3-4):155-160. [cited 2021 Jan 10] Available from: https://pjohns.pso-hns.org/index.php/pjohns/issue/view/53/3. 16. Tuli IP, Sengupta S, Munjal S, Kesari SP, Chakraborty S. Anatomical variations of uncinate process observed in chronic sinusitis.  Indian J Otolaryngol Head Neck Surg. 2012 Dec 25;65(2):157-161. DOI: 10.1007/s12070-012-0612-8; PubMed PMID: 24427557; PubMed Central PMCID: PMC3649035. 17. Paber JEL, Cabato MS, Villarta R, Hernandez J. Radiographic Analysis of the Ethmoid Roof based on KEROS Classification among Filipinos. Philipp J Otolaryngol Head Neck Surg. 2008 Jan- Jun;23(1):15–19. DOI: 10.32412/pjohns.v23i1.763. 18. lmushayti Z A, Almutairi A N, Almushayti M A, et al. 2022. Evaluation of the Keros Classification of Olfactory Fossa by CT Scan in Qassim Region. Cureus. 2022 Feb 19;14(2):e22378. DOI: 10.7759/ cureus.22378; PubMed PMID: 35321069; PubMed Central PMCID: PMC8935634. 19. Murthy VA, Santosh B. A Study of Clinical Significance of the Depth of Olfactory Fossa in Patients Undergoing Endoscopic Sinus Surgery.  Indian J Otolaryngol Head Neck Surg. 2017 Dec;69(4):514-522. DOI:  10.1007/s12070-017-1229-8; PubMed PMID:  29238684; PubMed Central PMCID: PMC5714917. 20. Kaygusuz A, Haksever M, Akduman D, Aslan S, Sayar Z. Sinonasal Anatomical Variations: Their Relationship with Chronic Rhinosinusitis and Effect on the Severity of Disease—A Computerized Tomography Assisted Anatomical and Clinical Study. Indian J Otolaryngol Head Neck Surg. 2014 Sep;66(3):260-6. DOI:  10.1007/s12070-013-0678-y; PubMed PMID:  25032111; PubMed Central PMCID: PMC4071417. 21. Taha MA, Hall CA, Zylicz HE, Westbrook MB, Barham WT, Stevenson MM, et al. Variability of the Anterior Ethmoid Artery in Endoscopic Sinus Surgery. Ear Nose Throat J. 2022 May;101(4):268– 273. DOI: 10.1177/0145561320950488; PubMed PMID: 32845806. 22. Badia L, Lund VJ, Wei W, Ho WK. 2005. Ethnic variation in sinonasal anatomy on CT scanning. Rhinology. 2005 Sep;43(3):210-4. PubMed Central PMID: 16218515. 23. Gouripur K, Kumar U, Janagond AB, Elangovan S, Srinivasa V. Incidence of sinonasal anatomical variations associated with chronic sinusitis by CT scan in Karaikal, South India. Int J Otorhinolaryngol Head Neck Surg. 2017 Jul;3(3):576-580. DOI: http://dx.doi.org/10.18203/ issn.2454-5929.ijohns20172291. different ethnicities and even among studies including the same ethnicity. Therefore, the PNS CT scan of each patient who will undergo sinus surgery must be meticulously analyzed. Every possible anatomic variant mentioned in this study should be noted to avoid inadvertent injuries to the critical structures during ESS.