First Olfactory Fiber as an Anatomical Landmark for Frontal Sinus Surgery

Transcription

1 The Laryngoscope VC 2015 The American Laryngological, Rhinological and Otological Society, Inc. First Olfactory Fiber as an Anatomical Landmark for Frontal Sinus Surgery Smita Upadhyay, MD; Lamia Buohliqah, MD; Gerival Vieira Junior, MD; Bradley A. Otto, MD; Daniel M. Prevedello, MD; Ricardo L. Carrau, MD Objectives/Hypothesis: Access to the frontal sinuses is technically challenging owing to their anterosuperior location, diverse anatomy, close proximity to critical structures and the need to work in a relatively narrow space with angled-lens endoscopes and instruments. This study attempts to study the relationship of the first olfactory fiber with the frontal sinus posterior wall, assessing its fidelity as a surgical landmark during frontal sinus surgery. Study Design: Anatomic study. Methods: Fifteen cadaveric specimens were studied. Measurements were obtained bilaterally using the data from individual CT scans. Median A-P was defined as the anteroposterior (A-P) diameter measured just lateral to the intersinus septum, paramedian A-P was measured 5 mm lateral to the septum, and maximum A-P was defined as the maximum anteroposterior diameter on axial images. A surgical navigation device was used to calculate the distance between the first olfactory fiber and the posterior table of the frontal sinus. Results: The mean distance between the first olfactory fiber and the posterior wall of the frontal sinus was ( ) mm on the right side and ( ) mm on the left. This distance strongly correlated with the maximum A-P diameter of the sinus. Conclusions: In a cadaveric model, the first olfactory fiber was found to be an average of 4.0 mm posterior to the frontal sinus. The significant variability of this distance should be considered when using the first olfactory fiber to establish the posterior boundary of a frontal sinusotomy. Drilling no further posterior than 7 mm rostral to the first olfactory fiber would be safe in 91% of patients. Key Words: Cranial base, skull base, nose, paranasal sinuses, frontal sinus, olfactory. Level of Evidence: NA Laryngoscope, 126: , 2016 INTRODUCTION Lesions in the frontal sinus present a surgical challenge owing to their inherent complex anatomy, wide array of anatomical variants, close proximity to the anterior cranial fossa and orbits, and their anterosuperior location, which demands the use of angledlens endoscopes and angled instruments. Increased experience and technological advances in optics, manual and powered instrumentation, and intraoperative imageguidancehavebeenpivotalintheevolutionand success of the surgical management of frontal sinus disease. From the Department of Otolaryngology Head & Neck Surgery (S.U., L.B., B.A.O., D.M.P., R.L.C.), Wexner Medical Center at The Ohio State University, Columbus, Ohio; Department of Neurosurgery (G.V.J., B.A.O., D.M.P., R.L.C.), Wexner Medical Center at The Ohio State University, Columbus, Ohio, U.S.A. Editor s Note: This Manuscript was accepted for publication August 28, The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Ricardo L. Carrau, MD, Professor, Director of the Comprehensive Skull Base Surgery Program, Department of Otolaryngology Head and Neck Surgery, Wexner Medical Center at The Ohio State University, 320 West 10th Avenue, Starling-Loving Hall, B221, Columbus, OH ricardo.carrau@osumc.edu DOI: /lary A basic tenet of minimizing surgical morbidity is to develop comprehensive knowledge of the pertinent anatomy, especially that involving critical surgical landmarks. The first olfactory fiber has often been cited as an important anatomical marker while performing a Draf type III frontal sinusotomy (modified endoscopic Lothrop procedure). Multiple authors recommend identifying the first olfactory fiber to accurately demarcate the position of the olfactory fossa (i.e., cribriform plate or anterior cranial fossa), thus ascertaining the posterior limit of the dissection. 1 5 Although the need to identify the first olfactory fiber has been repeatedly advocated in the literature, we did not find any report offering objective data to support this recommendation. Therefore, the purpose of the present study was to assess the fidelity of the first olfactory fiber as a landmark during extended frontal sinus surgery. Following this objective, we studied the position of the first olfactory fiber to better define the variability in its position and to assess whether or not the morphometry of the frontal sinus and skull base has any bearing on this position. Geometrical measurements of the frontal sinus were done in an attempt to ascertain if any of these measurements could predict the position of the first olfactory fiber relative to the posterior table of the frontal sinus. 1039

2 Fig. 1. Photograph showing the anteroposterior diameter on an axial computed tomography scan. (A) Shows the anteroposterior diameter as measured on axial images at a level just lateral to the intersinus septum (bilaterally). (B) The anteroposterior diameter measured at a level that was 5 mm lateral to the intersinus septum. (C) Maximum anteroposterior diameter on the same specimen. [Color figure can be viewed in the online issue, which is available at MATERIALS AND METHODS We studied 15 human anatomic specimens (30 sides). All dissections were performed in accordance with institutional protocols at the Anatomical Laboratory Towards Visuospatial Innovations in Otolaryngology and Neurosurgery at the Wexner Medical Center of The Ohio State University. All dissections were performed using rod-lens endoscopes (18 cm in length and 4 mm in width) coupled to a high-definition monitor and camera (Karl Storz Endoscopy, Tuttlingen, Germany). All specimens underwent a fine-cut (1 mm) computed tomography (CT) scan prior to their dissection, and the digital datasets were analyzed using Reviewer (Sorna Corporation, Eagan, MN). In addition, digital datasets from the CT scans were imported to a Stryker surgical navigation system (Stryker, Kalamazoo, MI). Specimens with a history (or radiological evidence) of previous trauma or previous surgery involving the skull base were excluded from the study. In addition, we eliminated specimens with frontal sinuses that would not be considered adequate for an endoscopic Lothrop (i.e., Draf III); therefore, based on the radiographic criteria described by Farhat et al. 6 and Burkart and Zimmer, 7 specimens with a nasal beak thickness greater than 10 mm, an access dimension of less than 0.5 cm, or a parasagittal anteroposterior (A-P) dimension of less than 0.7 cm were also excluded. posterior table of the frontal sinus, and line B, drawn from the anterior edge of the cribriform plate to the dorsum sella (Fig. 2). To identify the first olfactory fiber, an inverted U-shaped incision was made, starting slightly anterior to the anterior attachment of the middle turbinate, across the roof of the nasal cavity, and extending to the nasal septum using a number 10 scalpel blade (Fig. 3). A suction-freer elevator was used to Radiological Measurements Three sets of measurements were made on the radiographic data bilaterally. Axial CT scans were studied, and the measurements were made using the first section in which we could identify the attachment of the crista galli to the cribriform plate (moving from superior to inferior) (Fig. 1). The three sets of measurements were: 1. The A-P diameter for median A-P was measured bilaterally at a level that was just lateral to the intersinus septum. 2. The A-P diameter for the paramedian A-P was measured at a level that was 5 mm lateral to the intersinus septum. 3. The maximum A-P diameter A-P (max) was also noted bilaterally. In addition, the angle at which the skull base meets the posterior table of the frontal sinus (at the level of the crista galli) was also measured on a midsagittal scan. This was calculated as the angle between line A, drawn through the sloping edge of the Fig. 2. Photograph showing the skull base-frontal sinus angle (SBA) on a midsagittal scan at the level of crista galli. Line A is drawn through the sloping edge of the posterior table of the frontal sinus touching the skull base. Line B is drawn from the anterior edge of the cribriform plate to the dorsum sella. The angle between these two lines was calculated manually. 1040

3 Fig. 3. Endoscopic view using a 08 rodlens endoscope demonstrating the steps of dissection. (A) An inverted U-shaped incision was made starting slightly anterior to the anterior attachment of the middle turbinate across the roof of the nasal cavity and extending to the nasal septum using a no. 10 scalpel blade. (B) A suction- Freer elevator was used to elevate the mucosa of this area. (C) A mucoperiosteal flap is elevated, and the underlying bone of the roof of the nasal cavity is visualized. (D) First olfactory fiber is visualized as it exits the cribriform plate. L 5 lateral nasal wall; MT 5 middle turbinate; OF 5 olfactory fiber; R 5 roof of nasal cavity; S 5 septum. dissect the mucosa off the bone of the olfactory fossa to reach, identify, and photograph the first olfactory fiber as it exited its foramen at the cribriform plate. The distance between the first olfactory fiber, as it exits from the cribriform plate (Fig. 4, line C), and the anterior surface of the posterior table (air-bone interface of the frontal sinus (Fig. 4, line F) was measured by sliding the stereotactic probe anteriorly (distance C F; Fig. 4). Two investigators independently conducted these measurements with high inter-rater reliability. Statistical Analysis Table I offers a summary of the calculations with means and standard deviations. The Pearson correlation coefficient (r), Fig. 4. Image from the surgical navigation device. (A) A stereotactic probe is pointing at the first olfactory fiber (line C). (B) Distance C F as measured by sliding the navigation probe anteriorly from position (line C) to the inner cortex of the posterior table of the frontal sinus (line F). 1041

4 TABLE I. Summary of Findings. CF-R CF-L AP med-r AP pm-r APmax -R AP med-l AP pm-l APmax-L (5) (15) (7) (7) (5) (0) (12) (21) (10) (5) (8) (9.2) (5) (0) (12) (9.5) (14.5) (13.7) (11.3) (9.8) (5) (14.4) (5) (7) (8.8) (15) All distances are in millimeters. AP max-l 5 maximum anteroposterior diameter on the left (values in parenthesis indicate distance from midline at which the maximum AP diameter was recorded); AP max-r 5 maximum anteroposterior diameter on the right (values in parenthesis indicate distance from midline at which the maximum AP diameter was recorded); AP med-l 5 median anteroposterior diameter on the left; AP med-r 5 median anteroposterior diameter on the right; AP pm-l 5 paramedian anteroposterior diameter on the left (measured 5 mm lateral to the intersinus septum); AP pm-r 5 paramedian anteroposterior diameter on the right (measured 5 mm lateral to the intersinus septum); CF-L 5 distance between the first olfactory fiber and the frontal sinus on the left side; CF-R 5 distance between the first olfactory fiber and the frontal sinus on the right side. to assess for possible associations between continuous variables, was calculated. A value of P <.05 was considered statistically significant. The following correlations were analyzed bilaterally: 1. Median A-P diameter of the frontal sinus and the position of the first olfactory fiber (distance C F) 2. Paramedian A-P diameter of the frontal sinus and the position of the first olfactory fiber (distance C-F) 3. Maximum A-P diameter of the frontal sinus and the position of the first olfactory fiber (distance C F). 4. Skull base-frontal sinus posterior wall angle (as measured by the angle between lines A and B) and the position of the first olfactory fiber (distance C F). RESULTS Two specimens were excluded from the study, as the parasagittal A-P diameter was less than 7 mm; therefore, 13 specimens (26 sides) were included in the study. The average median A-P diameter was mm on the right side and mm on the left. The maximum A-P diameter A-P (max) was mm on the right side and mm on the left. The mean angulation between the posterior table of the frontal sinus and the skull base at the level of the crista galli was The mean distance between the first olfactory fiber and the anterior surface (adjacent to the air space) of the frontal sinus posterior table was mm on the right side and mm on the left. The Pearson correlation coefficient (r) for the median A-P diameter and the position of the first olfactory fiber was 0.41 on the right side (indicating a moderate positive correlation). This difference, however, was not statistically significant (P 5.16). The r value for the left side was 0.64, indicating a strong positive correlation, which approached statistical significance (P 5.05). The maximum A-P diameter of the frontal sinus A-P (max) strongly correlated with the position of the first olfactory fiber bilaterally. The r value on the right side was 0.64, and on the left was 0.66, both of which were statistically significant (P 5.01). The r value for the association between the skull base-frontal sinus angle and the position of the first olfactory fiber was weak on the right side and moderately positive on the left (r on right and r on the left). DISCUSSION The last few decades have witnessed a radical change in the surgical paradigm to tackle lesions in the frontal sinus. Following the seminal work of Messerklinger, there has been a superior understanding of the physiology and the pathophysiology of the sinonasal mucosa. Increased experience led to alternative algorithms for the surgical treatment of frontal sinus disease. Draf proposed a classification of drainage procedures to deal with incrementally severe frontal sinus disease. 2,8 Extended drainage of the frontal sinus is achieved by resecting the floor of the frontal sinus between the lamina papyracea and the middle turbinate (type IIa) or the nasal septum (type IIb) anterior to the ventral margin of the olfactory fossa. The more extensive type III drainage involves bilateral type IIb drainage with the addition of the resection of the superior aspect of the nasal septum in the area adjacent to the frontal sinus floor. 2,9 The posterior limit of the dissection remains anterior to the olfactory fossa; thus, the first olfactory fiber is identified as it exits the cribriform plate as a proxy landmark of the anterior margin of the olfactory fossa. 1042

5 Fig. 5. Photographs illustrating the endoscopic dissection. (A) Endoscopic image using a 308 rod-lens endoscope showing the branch of anterior ethmoidal artery as it courses through the roof of the nasal cavity. (B) Endoscopic image using a 0 lens showing the branch of anterior ethmoidal artery and nerve. (C) Endoscopic image showing the first olfactory fiber as seen in relation to the frontal sinuses. The superior bony septum has been removed, and the frontal sinuses have been opened bilaterally. (D) Endoscopic image showing the first olfactory fiber on the right side as seen in relation to the frontal sinus. A Draf III procedure has been performed. The frontal sinuses are widely open and the intersinus septum has been drilled. *The position of the branch of anterior ethmoidal artery. The black arrows point to the olfactory fiber. R 5 roof of the nasal cavity; S 5 septum; BMT 5 superior bony portion of the middle turbinate; FS 5 frontal sinus; PWFS 5 posterior wall of frontal sinus. The cribriform plate has about 40 needle-pricksized perforations arranged in two sagittal rows, with the smaller round ones being laterally placed and the larger oval ones placed medially. 10 At the anterior cribriform plate are two small fissures; the medial fissure is occupied by a process of dura matter, whereas the lateral fissure (also known as the cribroethmoidal foramen) transmits the branches of the anterior ethmoidal nerve and artery. 11 This branch of the anterior ethmoidal artery is often encountered as one searches for the first olfactory fiber (Fig. 5). This study did not intend to examine the morphometry of the frontal sinus, which has been well described in the past Rather, the present study assessed the relationship of the first olfactory fiber with respect to the frontal sinus (i.e., air space), and thus ascertained the fidelity of the first olfactory fiber as a reliable landmark during the frontal sinus surgery. In addition, the study explored potential anatomical factors, which might affect this relationship, such as the extent of A-P frontal sinus pneumatization and angle at which the skull base meets the frontal sinus. To a great extent, the complexity of frontal sinus surgery stems from its intricate relationship to vital structures such as the orbits and brain. Therefore, surgical precision is paramount to avoid disastrous complications that might arise upon violation of the sinus boundaries such as injury of intracranial or orbital structures. The reported incidence of cerebrospinal fluid (CSF) fistula during or following an endoscopic Lothrop is variable. Wormald reported a single case of CSF leak among 83 patients (1%) undergoing an endoscopic modified Lothrop approach. 3 Similarly, Schlosser et al. reported one case of posterior table dehiscence (2%) and one case of tension pneumocephalus (2%) in a series of 54 patients undergoing modified endoscopic Lothrop. 15 In a series of patients undergoing a frontal sinus drillout, reported by Batra et al., 16 a CSF leak occurred in two out of 30 patients (6.6%), although the authors claimed that in only one patient (3.3%) the CSF leak was directly related to the drill-out procedure. Ulualp et al. 17 reported a similar incidence of CSF leak of 6.7%, and Close 18 reported a higher incidence of 11.9%. A systematic review by Scott et al. 19 concluded that the incidence of CSF leak following an endoscopic modified Lothrop procedure was higher than that of an osteoplastic flap, even in the hands of experienced surgeons. Recognizing that the endoscopic Lothrop is a technically challenging procedure, Scott proposed to undergo sequential training starting with the study of surgical theory, endoscopic anatomy, and cadaveric dissection to be followed by supervised surgery before finally performing the procedure independently. 19 This study found that the position of the first olfactory fiber with respect to the posterior wall of the frontal sinus was more variable than previously inferred by 1043

6 others. There were variations noted among the cadavers, as well as within the same cadaveric specimen. For example, in specimen 1, the exit of the first olfactory fiber correlated well with the posterior table of the frontal sinus; however, in specimen 6 the first olfactory fiber was 10.8 mm from the posterior table of the frontal sinus. There were also differences noted between the two sides of the same specimen (2, 3, 4, 7, 8, 10, and 13). These findings are in agreement with those of Schmidt, 20 who found that the two sieve plates do not lie in neither the same horizontal nor the same sagittal plane. Therefore, the position of the first olfactory fiber and its correlation to the frontal sinus is variable on both the sides. This has significant clinical implications, as it suggests that the first olfactory fiber may not be a reliable surgical landmark in all patients. Extensive literature exists regarding frontal sinus morphometry showing wide variations. This analysis also found wide variations in the extent of the frontal sinus pneumatization in the A-P axis. The median A-P varied from 9 to 19.4 mm (mean mm) on the right, and varied from 10 to 19.4 mm (mean mm) on the left. This was similar to prior observations by Tatlisumak et al., who reported a mean A-P diameter of mm on the right and mm on the left frontal sinuses. 13 Spaeth et al. reported an A-P diameter of mm for adult men and mm for adult women. 14 These variations are due, in great part, to the use of different measurement techniques and the inclusion of different populations. A moderate positive correlation was noted between the median A-P and the position of the first olfactory fiber. However, a strong positive and statistically significant correlation was noted between the maximum A-P diameter and the position of the first olfactory fiber bilaterally. The A-P (max) in the majority of the patients was recorded at a distance of more than 5 mm lateral to the intersinus septum (average of 8 mm lateral to the intersinus septum on the right and 9 mm lateral to the intersinus septum on the left). The specimen with the largest maximum A-P diameter also had the farthest first olfactory fiber from the posterior table of the frontal sinus. Therefore, in clinical scenarios, the exclusive use of the first olfactory fiber as a guiding landmark could lead to a CSF leak and damage of intracranial neurovascular structures. Various skull base angles have been measured in the past; however, the angle between the posterior wall of the frontal sinus and skull base has not been previously studied. 21 The skull base-frontal sinus angle was on the average (ranging from 918 to 1208) and had a weak association with the position of the first olfactory fiber. Recent advances in CT and magnetic resonance imaging, surgical image guidance, powered surgical instruments, and advanced drill technologies potentially have increased the safety of frontal sinus surgery. However, some have raised concerns regarding the possible mismatch of surgical experience and skills with these and other surgical technologies. 22 Furthermore, this study suggests that in a significant number of patients, the use of the first olfactory fiber as the guiding landmark for the posterior limit of the drill out procedure may not be as reliable as previously believed. 1 3 The technique, as advocated by Wormald, is to use an image-guided navigational device to precisely localize the position of the olfactory fossae forward projections, and carefully remove the overlying bone using the straight drill with a 3.2-mm burr. 3 The mucosa overlying the olfactory fossa is elevated with a suction-freer dissector and the position of the first olfactory neuron is identified. These maneuvers allow the accurate identification of the olfactory fossa. Marking the first olfactory fiber with the navigation device allows the surgeon to ascertain how well this landmark matches the posterior wall of the frontal sinus. We concur with Wormald in that combining the use of anatomic landmarks with image guidance is ideal while removing the bone over the olfactory fossa. In addition, the image guidance may be used to estimate the amount of bone that can be removed safely, and the bone removal should be done in small increments. When a navigation device is not available, one should use multiple anatomical landmarks such as the skin of the nasoorbital area, heads of the middle turbinates, and first olfactory fiber. Drilling no further posterior than 6 mm rostral to the first olfactory fiber would be safe in 80% of cases, whereas staying 7 mm rostral to the first olfactory fiber would be safe in 91% of cases. The authors recognize that this study has some significant limitations, as the sample size is small and the findings were not confirmed by visualization on endoscopy in all specimens. We believe, however, that this latter point is of lesser importance because it is very difficult to estimate distances and choose precise points when viewing a convex surface using bidimensional images. Moreover, only three sets of correlations were checked for. A larger sample size with measurements at multiple levels could yield different results. CONCLUSION In a cadaveric model the position of the first olfactory fiber was highly variable, and on average was found 4.0-mm behind the posterior table of the frontal sinuses. The first olfactory fiber position correlated strongly with the extent of pneumatization of the frontal sinus in the A-P direction. Drilling no further posterior than 6 mm rostral to the first olfactory fiber would be safe in 80% of patients, whereas staying 7 mm rostral to the first olfactory fiber would be safe in 91% of patients. These data compel a detailed study with a larger sample size to corroborate and further elaborate on our findings. BIBLIOGRAPHY 1. Wolfgang Draf AM. The frontal T in the refinement of endonasal frontal sinus type III drainage. Oper Tech Otolaryngol 2006;17: Weber R, Draf W, Kratzsch B, Hosemann W, Schaefer SD. Modern concepts of frontal sinus surgery. Laryngoscope 2001;111: Wormald PJ. Salvage frontal sinus surgery: the endoscopic modified Lothrop procedure. Laryngoscope 2003;113: Georgalas C, Hansen F, Videler WJ, Fokkens WJ. Long terms results of Draf type III (modified endoscopic Lothrop) frontal sinus drainage 1044

7 procedure in 122 patients: a single centre experience. Rhinology 2011; 49: Chin D, Snidvongs K, Kalish L, Sacks R, Harvey RJ. The outside-in approach to the modified endoscopic Lothrop procedure. Laryngoscope 2012;122: Farhat FT, Figueroa RE, Kountakis SE. Anatomic measurements for the endoscopic modified Lothrop procedure. Am J Rhinol 2005;19: Burkart CM, Zimmer LA. Endoscopic modified Lothrop procedure: a radiographic anatomic study. Laryngoscope 2011;121: Draf W. Endonasal Frontal Sinus Drainage Type I-III According to Draf. Berlin, Germany: Springer; Draf W. Endonasal micro-endoscopic frontal sinus surgery: the Fulda concept. Oper Tech Otolaryngol Head Neck Surg 1991;2: Kawahara G, Matsuda M, Sugiyama K, Nakazawa R, Shima K. Studies on the Japanese lamina cribrosa statistical observation on its shape, number of pores and area [in Japanese]. Zasshi Tokyo Ika Daigaku 1968;26: Williams PL. Gray s Anatomy. 38th ed. Edinburgh: Churchill Livingston; Lee MK, Sakai O, Spiegel JH. CT measurement of the frontal sinus gender differences and implications for frontal cranioplasty. J Craniomaxillofac Surg 2010;38: Tatlisumak E, Ovali GY, Asirdizer M, et al. CT study on morphometry of frontal sinus. Clin Anat 2008;21: Spaeth J, Krugelstein U, Schlondorff G. The paranasal sinuses in CTimaging: development from birth to age 25. Int J Pediatr Otorhinolaryngol 1997;39: Schlosser RJ, Zachmann G, Harrison S, Gross CW. The endoscopic modified Lothrop: long-term follow-up on 44 patients. Am J Rhinol 2002;16: Batra PS, Cannady SB, Lanza DC. Surgical outcomes of drillout procedures for complex frontal sinus pathology. Laryngoscope 2007;117: Ulualp SO, Carlson TK, Toohill RJ. Osteoplastic flap versus modified endoscopic Lothrop procedure in patients with frontal sinus disease. Am J Rhinol 2000;14: Close LG. Endoscopic Lothrop procedure: when should it be considered? Curr Opin Otolaryngol Head Neck Surg 2005;13: Scott NA, Wormald P, Close D, Gallagher R, Anthony A, Maddern GJ. Endoscopic modified Lothrop procedure for the treatment of chronic frontal sinusitis: a systematic review. Otolaryngol Head Neck Surg 2003;129: Schmidt HM. Measures and level differences in the median structure of the anterior cranial fossa in the man [in German]. Gegenbaurs Morphol Jahrb 1974;120: Koenigsberg RA, Vakil N, Hong TA, et al. Evaluation of platybasia with MR imaging. AJNR Am J Neuroradiol 2005;26: Gross CW, Becker DG. Instrumentation in endoscopic sinus surgery. Curr Opin Otolaryngol Head Neck Surg 1996;4: