- Email: [email protected]
Image-guided functional endoscopic sinus surgery
Image-guided functional endoscopic sinus surgery GARTH OLSON, MD, and MARTIN J. CITARDI, MD, St Louis, Missouri
INTRODUCTION: Computer-aided surgery (CAS) technology in functional endoscopic sinus surgery (FESS) has engendered considerable discussion. OBJECTIVE: The goals of this study were to describe CAS preoperative planning (software-based CT image analysis) and to develop intraoperative CAS strategies for endoscopic sinus surgery. STUDY DESIGN: Between October 1, 1997, and December 31, 1998, the StealthStation (Sofamor Danek, Memphis, TN) was used in 61 FESS cases, and a retrospective review of the findings was performed. The indication for surgery in all instances was chronic rhinosinusitis refractory to medical management. The StealthStation was used to review all CT scans before surgery. Anatomic fiducial registration supplemented by contour mapping was used. RESULTS: Localization accuracy was estimated to be within 2 mm or better. The StealthStation was used for both CT image review and intraoperative localization. CAS was useful in the frontal recess, sphenoethmoid region, posterior ethmoid system, and skull base area. CAS was deemed helpful in situations where the surgical anatomy was altered by previous surgery and extensive inflammatory disease (polyposis, fungal sinusitis, and pansinusitis). CONCLUSION: The paradigm of image-guided FESS surgery, which integrates CAS into FESS, will serve to increase surgical effectiveness and decrease surgical morbidity. (Otolaryngol Head Neck Surg 2000; 123:188-94.)
Modern surgical nasal telescopes provide brilliant images of the intranasal and intrasinus anatomy; however, this visualization still may be compromised by
anatomic complexity and intraoperative bleeding. In addition, specific inflammatory conditions, such as sinonasal polyposis, allergic fungal sinusitis, and even refractory chronic rhinosinusitis, often further distort anatomic landmarks. Consequently, rhinologic surgery is associated with significant complications, including intracranial and orbital injury. It is critical to consider that nasal telescopes provide only a 2-dimensional view of complex 3-dimensional structures. Preoperative CT scans provide essential information that serves to guide the rhinologic surgeon1; however, the surgeon must mentally extrapolate complex 3-dimensional relationships from 2-dimensional CT scan images and then apply this information to the surgical field. This cognitive process is difficult and often imprecise. Recently, computer-aided surgery (CAS) technology has been introduced and popularized. This technology permits a direct comparison of intraoperative anatomy with preoperative imaging information. After a registration process, the surgeon may point to a specific structure with the CAS instrument and then view the location of the instrument tip on the CT images, which are projected in 3 orthogonal views on a computer monitor. Additionally, CAS technology provides software tools for computer-enabled review of preoperative imaging. Although the use of CAS in functional endoscopic sinus surgery (FESS) is becoming more frequent, its specific utility in specific clinical situations has not been established. The CAS literature has mainly focused on the mechanics of its use rather than on specific situations where it may be useful. In this article the specific anatomic situations where CAS is most useful are described. Emphasis is placed on the impact of CAS in the most challenging rhinologic surgical procedures. METHODS
From the Department of Otolaryngology, Saint Louis University, School of Medicine. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, New Orleans, LA, September 26-29, 1999. Reprint requests: Martin J. Citardi, MD, Department of Otolaryngology, Saint Louis University, 3635 Vista Ave at Grand Blvd, St Louis, MO 63110. Copyright © 2000 by the American Academy of Otolaryngology– Head and Neck Surgery Foundation, Inc. 0194-5998/2000/$12.00 + 0 23/1/107453 doi:10.1067/mhn.2000.107453 188
The charts of all patients who underwent FESS with surgical navigation by the senior author (M.J.C.) between August 1, 1997, and December 31, 1999, were reviewed. The review focused on specific details that illustrated the application of CAS. The StealthStation platform (Sofamor Danek, Memphis, TN) was used in all cases. Both the LandMarX 2.6.4 (Xomed, Jacksonville, FL) and Stealth Cranial 2.6.4 (Sofamor Danek) software packages were used. This system permits intraoperative tracking of surgical instruments with an optical tracking system.
Otolaryngology– Head and Neck Surgery Volume 123 Number 3
Before surgery, a standardized CAS sinus CT scan (1-mm discrete axial images) was obtained and transferred to the CAS workstation, where software-based CT review was completed. The CAS software tools for this review included (1) coronal and sagittal image reconstruction, (2) 3-dimensional model reconstruction, (3) simultaneous scrolling through the 3 orthogonal planes (axial, coronal, and sagittal planes), (4) CT window width and level adjustments, (5) 3-dimensional model cut views, and (6) distance measurement tools. Through this review, specific surgical strategies were developed for each case. At the beginning of each case, a specific headset, which permitted attachment of a reference frame with light-emitting diodes for optical tracking, was placed securely on the patient’s head with a padded clamp mechanism. The headset was not bone anchored. Registration was performed with the PointMerge protocol, which relied on anatomic fiducial points (bilateral medial and lateral canthi and the right tragus). In most cases a SurfaceMerge registration, which is based on contour mapping, was also performed. Typical contours included the medial brows, nasal dorsum, and tragi. If necessary, the registration was edited by reselecting points for improved registration accuracy. Registration was verified by localization on known bony landmarks on the skull and in the nasal cavity. Registration was deemed acceptable if the estimated accuracy was 2 mm or better. A sustained accuracy point for monitoring intraoperative accuracy was also chosen. In all instances, FESS was performed as previously described.2,3 The CAS probes were used extensively for confirmation of surgical position. RESULTS
During the study interval, 62 patients underwent FESS with surgical navigation as described. Conservative medical treatment had failed in all patients, and 50 patients had undergone sinus surgery at another institution. In this group indications for the use of CAS included chronic frontal sinusitis (57 patients), previous surgery (50), sinonasal polyposis (30), sinonasal polyposis with asthma (16), asthma–nasal polyps–aspirin intolerance triad (3), allergic fungal sinusitis (2 patients), chronic granulomatous fungal sinusitis (1), cystic fibrosis (1), and previous maxillofacial trauma (1). CAS was not used for primary surgery for anterior ethmoid and maxillary sinusitis. The following procedures with CAS were performed: endoscopic frontal sinusotomy (113 sides), endoscopic ethmoidectomy (108), endoscopic maxillary antrostomy (108), and endoscopic sphenoidotomy (70). The review demonstrated a consistent pattern in which CAS was most helpful in specific anatomic areas. Case presentations, which emphasize the intraoperative findings, illustrate this usefulness.
OLSON and CITARDI
189
Fig 1. Left frontal sinus has pneumatized the right frontal bone. In the endoscopic view, a curved CAS suction has been passed through the left frontal recess; crosshairs indicate the localization of the probe tip in the contralateral extension of the left frontal sinus. R SOE, Right supraorbital ethmoid cell; L FS, left frontal sinus. *Path across the midline in the axial image.
CASE REPORTS Case 1: Frontal sinus. A 46-year-old man with a history of chronic rhinosinusitis, asthma, and sinonasal polyposis had persistent nasal congestion, purulent drainage, and facial pressure, despite sinus surgery that had been performed at another hospital. Conservative therapy with culture-directed antibiotics, topical nasal steroids, systemic steroids, and mucolytics had failed. FESS with surgical navigation was then performed. Initially, the preoperative CT scan was believed to be consistent with a septated right frontal sinus; however, during surgery the right frontal ostium could not be identified. With the CAS (Fig 1), it became apparent that there was no right frontal sinus at all; instead, the left frontal sinus had pneumatized across the midline. A right supraorbital ethmoid cell was also identified. Case 2: Sphenoid and sphenoethmoid region. A 44year-old woman was referred for treatment of sinonasal polyposis, which was producing anosmia, purulent rhinorrhea, and facial pain. Conservative medical treatment had failed. She underwent FESS with CAS. Review of the preoperative CT scan at the computer workstation demonstrated a left sphenoethmoid cell. During surgery, surgical navigation was used to confirm posterior ethmoid landmarks, including the skull base, sphenoid face, lamina papyracea, and sphenoethmoid cell posterior boundary. In
190
Otolaryngology– Head and Neck Surgery September 2000
OLSON and CITARDI
Fig 2. A, Sagittal image clearly demonstrates the presence of a sphenoethmoid cell above the sphenoid sinus. In the endoscopic view, the CAS suction probe is in this cell; crosshairs on the CT images confirm this localization. (Contrast and brightness of the endoscopic image have been adjusted digitally.) B, CAS suction probe has been placed in the sphenoid sinus; crosshairs on the CT images confirm this localization. (Contrast and brightness of the endoscopic image have been adjusted digitally.)
this way, complete posterior ethmoid dissection and sphenoidotomy were achieved safely and expeditiously (Fig 2). Case 3: Residual ethmoid partitions and sinonasal polyposis. A 70-year-old woman with a history of chronic
sinusitis, asthma, and sinonasal polyposis, had symptoms of facial pressure, cough, and purulent nasal drainage, which persisted despite aggressive antibiotic and steroid treatment. Approximately 7 months earlier, she had undergone sinus surgery at another institution. Bilateral FESS with CAS was performed. Intraoperative findings included severely inflamed mucosa and polyposis, and the surgical field was bloody. Review of the sagittal CT facilitated identification of each residual ethmoid partition. Under direct visualization with simultaneous surgical navigation, each ethmoid partition was incised and removed with through-cutting forceps (Fig 3). Case 4: Revision maxillary antrostomy. A 56-year-old woman with a history of chronic sinusitis and polyposis sought treatment for thick purulent drainage and facial pain. She had previously undergone sinus surgeries at another institution in 1994 and 1996. Medical therapy with culture-directed antibiotics and topical and systemic steroid therapy failed. Bilateral FESS with CAS was performed. Polypoid soft tissue and scar tissue filled the previous maxillary antrostomy. Surgical navigation was used to confirm the anterior extent of the maxillary antrostomy and its relation to the nasolacrimal system (Fig 4). Case 5: Skull base identification. A 53-year-old man
whose chronic sinusitis had been treated with surgery 2 years earlier reported nasal discharge, cough, and severe headaches. Because conservative medical treatment had failed, bilateral revision FESS with CAS was performed. During previous surgery, the patient’s middle turbinates had been resected; the loss of this surgical landmark made skull base identification by simple endoscopic examination challenging. Surgical navigation was used to directly localize the skull base boundary (Fig 5). Case 6: Orbital dehiscence. A 60-year-old patient with a history of chronic rhinosinusitis (including previous sinus surgery), asthma, and sinonasal polyposis had periodic nasal congestion, drainage, and facial pressure. Because his symptoms were worsening despite aggressive medical treatment, bilateral revision FESS with CAS was performed. Computer workstation review of the preoperative CT scans demonstrated an area of right orbital dehiscence. This finding was ascribed to the patient’s previous surgery, although there was no history of an orbital complication. During the initial endoscopy, this area of dehiscence had the appearance of an uncinate process. The CAS probe was used to precisely define the area of dehiscence. Under CAS guidance, access to the posterior ethmoid under the lamina dehiscence was achieved (Fig 6). DISCUSSION
This study reports a large series of patients who underwent FESS with frameless stereotactic surgical
Otolaryngology– Head and Neck Surgery Volume 123 Number 3
Fig 3. Sagittal CT facilitates identification of the residual ethmoid partitions. Each partition was identified sequentially and then removed. In the endoscopic view, localization at a particular partition is seen.
navigation. Recently, the surgical literature has featured numerous reports that describe this use of CAS technology for sinus surgery; the authors of these papers have emphasized hardware, accuracy measurements, and so forth.4-10 In contrast, our report highlights the specific utility of CAS for sinus surgery through the presentation of illustrative cases. In all cases, a surgical navigation system that incorporated optical tracking was used. For this type of tracking, the surgeon must maintain a line of sight between the overhead digitizer (camera array) and the reference frame and surgical instruments; fortunately, instruments that include an offset post for the tracking apparatus satisfy this requirement. A comparison of optical tracking and electromagnetic tracking (Intatrak; Visualization Technology, Woburn, MA) is beyond the scope of this report. The objective of this report is to describe the application of CAS, not the technology of CAS. CAS System Accuracy
Intraoperative CAS registration accuracy was estimated to be 2 mm or better. To determine system accuracy, we compared the CAS instrument tip position indicated by the CAS system with its actual location in the operative field at several areas of the operative volume. The estimated accuracy represents the surgeon’s visual estimate of system accuracy.
OLSON and CITARDI
191
Fig 4. In the endoscopic view, the CAS probe is placed just anterior to the maxillary antrostomy; in this way, the anatomic relationship between the antrostomy and the nasolacrimal system is demonstrated. CAS decreased the risk of inadvertent injury to the nasolacrimal system. (Contrast and brightness of the image have been adjusted digitally.)
Registration accuracy must be assessed by the surgeon during the procedure. It is important to realize that the calculated registration errors and accuracy that CAS systems provide after registration are merely values that may or may not represent true system accuracy. Clinical experience has demonstrated that although “tight” calculated registration errors and accuracy are desirable, these calculated values do not necessarily correlate with real-world registration accuracy. Furthermore, CAS registration may be precise in one area of the operating field but not a second area. For instance, registration may be accurate at the right frontal recess but inaccurate at the left sphenoethmoid recess. In addition, registration may drift if instruments are damaged or the CAS reference frame shifts. For these reasons, the operating surgeon must estimate CAS registration accuracy at several points in the operative field throughout the procedure. Even in revision cases, where many of the usual surgical landmarks are absent, identification of known points is critical to verify registration accuracy. The estimated accuracy of 2 mm in this article is believed to be sufficient for rhinologic procedures, although structures such as the lamina papyracea may be considerably thinner than 1 mm. Even at 2 mm of accuracy, the CAS provides sufficient localization
192
OLSON and CITARDI
Otolaryngology– Head and Neck Surgery September 2000
ations. First, severe sinonasal polyposis may obscure all surgical landmarks; CAS has proved invaluable in these cases. Similarly, CAS has special value in the surgical management of allergic fungal sinusitis, chronic invasive fungal sinusitis, and other disease processes that are characterized by extensive remodeling and/or destruction of surgical landmarks. In addition, revision surgery may be limited by previous surgical misadventures (such as lamina papyracea violation) that increase the risk of the secondary procedure. Rhinologic surgeons also understand that revision sinus surgery requires recognition of landmarks that prior surgery may have significantly altered. CAS facilitates the recognition of these problems and permits the development of a surgical plan for their management. Of course, intraoperative surgical navigation then permits direct correlation of preoperative image planning and assessment with the operative field. Fig 5. Skull base landmark was directly identified, even after middle turbinate resection. Localization at the osteitic roof of the posterior ethmoid is seen.
information because the surgeon also incorporates information from CT scan anatomy (from computerenabled CT review) and the endoscopic views. The modalities of CAS localization, CT scan image review, and endoscopic view are all complementary pieces of information that enhance effective system accuracy beyond the accuracy of any one modality alone. In this way, the net accuracy is beyond the 2-mm reported accuracy of the CAS registration alone—and beyond simple inspection of endoscopic images or CT scans. Specific Utility of CAS in FESS
The assessment of new technology (such as CAS) for particular applications can be problematic. Ideally, quantitative variables may be used, but there are limitations. The effect of the new technology on morbidity rates may require large patient numbers that make such a study impractical. Surgical outcome parameters are difficult to quantify objectively and reliably. For these reasons, it may be preferable to look at the impact of the new technology on treatment implementation. CAS seems to facilitate the surgeon’s comprehension of 3-dimensional anatomy in specific regions. In this way, CAS directly enhances widely practiced FESS techniques. The frontal recess, the sphenoethmoid recess, the sphenoid face, the maxillary sinus, and residual ethmoid cells represent anatomic regions where CAS is most useful. CAS also has special utility in specific clinical situ-
Image-guided FESS Paradigm
On the basis of these clinical experiences, we have formulated a strategy for the application of CAS in FESS. The objective of this effort has been to move beyond the novelty of CAS; the goal has been to directly integrate CAS into FESS. This paradigm has been termed image-guided FESS (IG-FESS) because the images provided by surgical nasal endoscopy and CAS together affect the surgical technique, strategy, and philosophy. Principles of IG-FESS include the following: • Precise anatomic dissection and mucosal dissection are paramount, as numerous authors have described.2,3 • Surgical navigation should not be used as a mere “point-and-hunt” device. • Preoperative CT review is performed on a computer workstation because such software-enabled review provides the surgeon with critical anatomic information that the corresponding static films do not. Through workstation review, the surgeon can more readily develop a 3-dimensional understanding of relevant anatomy. With this information, the surgeon may develop a surgical plan. • Surgical navigation allows the surgeon to directly relate preoperative imaging information and the surgical plan to the operative field. • Intraoperative tracking of surgical instruments, including curved aspirators, microdebriders, pointers, and forceps, directly involves the surgical navigation system in the implementation of the surgical plan. As a result, surgical precision is enhanced. In this way, the CAS unit has shifted from a confirmation aid to critical tool that influences surgical planning and execution.
Otolaryngology– Head and Neck Surgery Volume 123 Number 3
OLSON and CITARDI
193
Fig 6. A, Coronal CT clearly demonstrates a dehiscent region of the right lamina papyracea. Diagnostic endoscopy did not suggest a dehiscence. CAS probe was used to map the area of dehiscence in the operative field. B, Endoscopic view of a posterior ethmoid cell is seen. Localization within this cell is depicted by the crosshairs. Note that the CAS probe has been removed.
IG-FESS has the greatest utility for more complex rhinologic procedures. In fact, CAS should be considered whenever it is difficult to determine the 3-dimensional anatomic relationships from CT data. Indications for this approach include sinonasal polyposis, pansinusitis, and revision surgery. In addition, IGFESS is also useful for surgery at complex anatomic configurations in the frontal recess, skull base, and posterior ethmoid regions. The clinical experiences in this report emphasize the importance of CAS for specific anatomic findings, rather than a global condition (such as polyps). The real strength of CAS is that it brings depth to 2dimensional data, that is, CAS allows surgeons to perceive spatial relationships with a sense of depth previously unobtainable by viewing static films. The need for depth perception is especially great for surgical endoscopy because the telescopes provide only 2-dimensional information. CAS is an enabling technology that allows surgeons to comprehend spatial relationships both more directly and more accurately. IG-FESS Challenges
Surgeons must be cognizant of the limitations of CAS. When registration errors occur, localization accuracy can suffer. System accuracy always must be checked against known landmarks. Software bugs can crash computer systems, and hardware may fail. Human error may aggravate or even cause these problems. Surgeons must know the CAS system that they are
using so they can recognize these issues early and act accordingly. The disclaimer that CAS use is not a substitute for anatomic knowledge and surgical skill bears repeating.5 It is important to remember that CAS relies on preoperative imaging data, rather than on intraoperative imaging. The operating surgeon who uses CAS must always keep these challenges in mind. CONCLUSION
CAS technology includes not only intraoperative surgical navigation but also specific software tools for analysis of preoperative imaging. In rhinologic surgery this software-enabled review of preoperative images facilitates comprehension of 3-dimensional relationships. This article describes a large series of sinus surgery procedures that were performed with CAS. After review of these cases, it became clear that CAS has the greatest utility in specific anatomic locations (ie, frontal sinus, skull base, and sphenoethmoid recess) and in specific clinical situations (ie, revision surgery, sinonasal polyposis, and other scenarios characterized by the loss of surgical landmarks). CAS was critical for these procedures, as the representative cases illustrate; therefore CAS should be directly integrated into the preoperative planning and intraoperative execution of FESS for these more complex cases. The resultant paradigm of image-guided FESS will likely serve to increase surgical effectiveness and decrease overall surgical morbidity.
194
Otolaryngology– Head and Neck Surgery September 2000
OLSON and CITARDI
REFERENCES 1. Meyers RM, Valvassori G. Interpretation of anatomic variation of computed tomography scans of the sinuses: a surgeon’s perspective. Laryngoscope 1998;108:422-5. 2. Kennedy DW. Functional endoscopic sinus surgery (technique). Arch Otolaryngol Head Neck Surg 1985;111:643-9. 3. Kennedy DW. Prognostic factors, outcomes and staging in ethmoid sinus surgery. Laryngoscope 1992;102(Suppl):1-18. 4. Anon JB, Lipman SP, Oppenheim D, et al. Computer-assisted endoscopic sinus surgery. Laryngoscope 1994;104:901-5. 5. Roth M, Lanza DC, Zinreich J, et al. Advantages and disadvantages of three-dimensional computed tomography intraoperative localization for endoscopic sinus surgery. Laryngoscope 1995;105:1279-86. 6. Fried MP, Kleefield J, Gopal H, et al. Image-guided endoscopic
7.
8.
9. 10.
sinus surgery: results of accuracy and performance in a multicenter clinical study using an electromagnetic tracking system. Laryngoscope 1997;107:594-601. Anon JB, Klimek L, Mosges R, et al. Computer-assisted endoscopic sinus surgery (an international review). Otolaryngol Clinic North Am 1997;30:389-400. Fried MP, Kleefield J, Taylor R. New armless image-guidance system for endoscopic sinus surgery. Otolaryngol Head Neck Surg 1998;119:528-32. Anon JB. Computer-aided endoscopic sinus surgery. Laryngoscope 1998;108:949-61. Metson R, Gliklich RE, Cosenza M. A comparison of imageguidance systems for sinus surgery. Laryngoscope 1998;108: 1164-70.
AVAILABILITY OF JOURNAL BACK ISSUES As a service to our subscribers, copies of back issues of Otolaryngology–Head and Neck Surgery for the preceding 5 years are maintained and are available for purchase from Mosby until inventory is depleted. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby, Subscription Customer Service, 6277 Sea Harbor Dr, Orlando, FL 32887, or call 800-654-2452 or 407-345-4000 for information on availability of particular issues and prices. If unavailable from the publisher, photocopies of complete issues may be purchased from Bell & Howell Information and Learning, 300 N Zeeb Rd, Ann Arbor, MI 48106-1346; phone, 734-761-4700 or 800-521-0600.
INTRODUCTION: Computer-aided surgery (CAS) technology in functional endoscopic sinus surgery (FESS) has engendered considerable discussion. OBJECTIVE: The goals of this study were to describe CAS preoperative planning (software-based CT image analysis) and to develop intraoperative CAS strategies for endoscopic sinus surgery. STUDY DESIGN: Between October 1, 1997, and December 31, 1998, the StealthStation (Sofamor Danek, Memphis, TN) was used in 61 FESS cases, and a retrospective review of the findings was performed. The indication for surgery in all instances was chronic rhinosinusitis refractory to medical management. The StealthStation was used to review all CT scans before surgery. Anatomic fiducial registration supplemented by contour mapping was used. RESULTS: Localization accuracy was estimated to be within 2 mm or better. The StealthStation was used for both CT image review and intraoperative localization. CAS was useful in the frontal recess, sphenoethmoid region, posterior ethmoid system, and skull base area. CAS was deemed helpful in situations where the surgical anatomy was altered by previous surgery and extensive inflammatory disease (polyposis, fungal sinusitis, and pansinusitis). CONCLUSION: The paradigm of image-guided FESS surgery, which integrates CAS into FESS, will serve to increase surgical effectiveness and decrease surgical morbidity. (Otolaryngol Head Neck Surg 2000; 123:188-94.)
Modern surgical nasal telescopes provide brilliant images of the intranasal and intrasinus anatomy; however, this visualization still may be compromised by
anatomic complexity and intraoperative bleeding. In addition, specific inflammatory conditions, such as sinonasal polyposis, allergic fungal sinusitis, and even refractory chronic rhinosinusitis, often further distort anatomic landmarks. Consequently, rhinologic surgery is associated with significant complications, including intracranial and orbital injury. It is critical to consider that nasal telescopes provide only a 2-dimensional view of complex 3-dimensional structures. Preoperative CT scans provide essential information that serves to guide the rhinologic surgeon1; however, the surgeon must mentally extrapolate complex 3-dimensional relationships from 2-dimensional CT scan images and then apply this information to the surgical field. This cognitive process is difficult and often imprecise. Recently, computer-aided surgery (CAS) technology has been introduced and popularized. This technology permits a direct comparison of intraoperative anatomy with preoperative imaging information. After a registration process, the surgeon may point to a specific structure with the CAS instrument and then view the location of the instrument tip on the CT images, which are projected in 3 orthogonal views on a computer monitor. Additionally, CAS technology provides software tools for computer-enabled review of preoperative imaging. Although the use of CAS in functional endoscopic sinus surgery (FESS) is becoming more frequent, its specific utility in specific clinical situations has not been established. The CAS literature has mainly focused on the mechanics of its use rather than on specific situations where it may be useful. In this article the specific anatomic situations where CAS is most useful are described. Emphasis is placed on the impact of CAS in the most challenging rhinologic surgical procedures. METHODS
From the Department of Otolaryngology, Saint Louis University, School of Medicine. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, New Orleans, LA, September 26-29, 1999. Reprint requests: Martin J. Citardi, MD, Department of Otolaryngology, Saint Louis University, 3635 Vista Ave at Grand Blvd, St Louis, MO 63110. Copyright © 2000 by the American Academy of Otolaryngology– Head and Neck Surgery Foundation, Inc. 0194-5998/2000/$12.00 + 0 23/1/107453 doi:10.1067/mhn.2000.107453 188
The charts of all patients who underwent FESS with surgical navigation by the senior author (M.J.C.) between August 1, 1997, and December 31, 1999, were reviewed. The review focused on specific details that illustrated the application of CAS. The StealthStation platform (Sofamor Danek, Memphis, TN) was used in all cases. Both the LandMarX 2.6.4 (Xomed, Jacksonville, FL) and Stealth Cranial 2.6.4 (Sofamor Danek) software packages were used. This system permits intraoperative tracking of surgical instruments with an optical tracking system.
Otolaryngology– Head and Neck Surgery Volume 123 Number 3
Before surgery, a standardized CAS sinus CT scan (1-mm discrete axial images) was obtained and transferred to the CAS workstation, where software-based CT review was completed. The CAS software tools for this review included (1) coronal and sagittal image reconstruction, (2) 3-dimensional model reconstruction, (3) simultaneous scrolling through the 3 orthogonal planes (axial, coronal, and sagittal planes), (4) CT window width and level adjustments, (5) 3-dimensional model cut views, and (6) distance measurement tools. Through this review, specific surgical strategies were developed for each case. At the beginning of each case, a specific headset, which permitted attachment of a reference frame with light-emitting diodes for optical tracking, was placed securely on the patient’s head with a padded clamp mechanism. The headset was not bone anchored. Registration was performed with the PointMerge protocol, which relied on anatomic fiducial points (bilateral medial and lateral canthi and the right tragus). In most cases a SurfaceMerge registration, which is based on contour mapping, was also performed. Typical contours included the medial brows, nasal dorsum, and tragi. If necessary, the registration was edited by reselecting points for improved registration accuracy. Registration was verified by localization on known bony landmarks on the skull and in the nasal cavity. Registration was deemed acceptable if the estimated accuracy was 2 mm or better. A sustained accuracy point for monitoring intraoperative accuracy was also chosen. In all instances, FESS was performed as previously described.2,3 The CAS probes were used extensively for confirmation of surgical position. RESULTS
During the study interval, 62 patients underwent FESS with surgical navigation as described. Conservative medical treatment had failed in all patients, and 50 patients had undergone sinus surgery at another institution. In this group indications for the use of CAS included chronic frontal sinusitis (57 patients), previous surgery (50), sinonasal polyposis (30), sinonasal polyposis with asthma (16), asthma–nasal polyps–aspirin intolerance triad (3), allergic fungal sinusitis (2 patients), chronic granulomatous fungal sinusitis (1), cystic fibrosis (1), and previous maxillofacial trauma (1). CAS was not used for primary surgery for anterior ethmoid and maxillary sinusitis. The following procedures with CAS were performed: endoscopic frontal sinusotomy (113 sides), endoscopic ethmoidectomy (108), endoscopic maxillary antrostomy (108), and endoscopic sphenoidotomy (70). The review demonstrated a consistent pattern in which CAS was most helpful in specific anatomic areas. Case presentations, which emphasize the intraoperative findings, illustrate this usefulness.
OLSON and CITARDI
189
Fig 1. Left frontal sinus has pneumatized the right frontal bone. In the endoscopic view, a curved CAS suction has been passed through the left frontal recess; crosshairs indicate the localization of the probe tip in the contralateral extension of the left frontal sinus. R SOE, Right supraorbital ethmoid cell; L FS, left frontal sinus. *Path across the midline in the axial image.
CASE REPORTS Case 1: Frontal sinus. A 46-year-old man with a history of chronic rhinosinusitis, asthma, and sinonasal polyposis had persistent nasal congestion, purulent drainage, and facial pressure, despite sinus surgery that had been performed at another hospital. Conservative therapy with culture-directed antibiotics, topical nasal steroids, systemic steroids, and mucolytics had failed. FESS with surgical navigation was then performed. Initially, the preoperative CT scan was believed to be consistent with a septated right frontal sinus; however, during surgery the right frontal ostium could not be identified. With the CAS (Fig 1), it became apparent that there was no right frontal sinus at all; instead, the left frontal sinus had pneumatized across the midline. A right supraorbital ethmoid cell was also identified. Case 2: Sphenoid and sphenoethmoid region. A 44year-old woman was referred for treatment of sinonasal polyposis, which was producing anosmia, purulent rhinorrhea, and facial pain. Conservative medical treatment had failed. She underwent FESS with CAS. Review of the preoperative CT scan at the computer workstation demonstrated a left sphenoethmoid cell. During surgery, surgical navigation was used to confirm posterior ethmoid landmarks, including the skull base, sphenoid face, lamina papyracea, and sphenoethmoid cell posterior boundary. In
190
Otolaryngology– Head and Neck Surgery September 2000
OLSON and CITARDI
Fig 2. A, Sagittal image clearly demonstrates the presence of a sphenoethmoid cell above the sphenoid sinus. In the endoscopic view, the CAS suction probe is in this cell; crosshairs on the CT images confirm this localization. (Contrast and brightness of the endoscopic image have been adjusted digitally.) B, CAS suction probe has been placed in the sphenoid sinus; crosshairs on the CT images confirm this localization. (Contrast and brightness of the endoscopic image have been adjusted digitally.)
this way, complete posterior ethmoid dissection and sphenoidotomy were achieved safely and expeditiously (Fig 2). Case 3: Residual ethmoid partitions and sinonasal polyposis. A 70-year-old woman with a history of chronic
sinusitis, asthma, and sinonasal polyposis, had symptoms of facial pressure, cough, and purulent nasal drainage, which persisted despite aggressive antibiotic and steroid treatment. Approximately 7 months earlier, she had undergone sinus surgery at another institution. Bilateral FESS with CAS was performed. Intraoperative findings included severely inflamed mucosa and polyposis, and the surgical field was bloody. Review of the sagittal CT facilitated identification of each residual ethmoid partition. Under direct visualization with simultaneous surgical navigation, each ethmoid partition was incised and removed with through-cutting forceps (Fig 3). Case 4: Revision maxillary antrostomy. A 56-year-old woman with a history of chronic sinusitis and polyposis sought treatment for thick purulent drainage and facial pain. She had previously undergone sinus surgeries at another institution in 1994 and 1996. Medical therapy with culture-directed antibiotics and topical and systemic steroid therapy failed. Bilateral FESS with CAS was performed. Polypoid soft tissue and scar tissue filled the previous maxillary antrostomy. Surgical navigation was used to confirm the anterior extent of the maxillary antrostomy and its relation to the nasolacrimal system (Fig 4). Case 5: Skull base identification. A 53-year-old man
whose chronic sinusitis had been treated with surgery 2 years earlier reported nasal discharge, cough, and severe headaches. Because conservative medical treatment had failed, bilateral revision FESS with CAS was performed. During previous surgery, the patient’s middle turbinates had been resected; the loss of this surgical landmark made skull base identification by simple endoscopic examination challenging. Surgical navigation was used to directly localize the skull base boundary (Fig 5). Case 6: Orbital dehiscence. A 60-year-old patient with a history of chronic rhinosinusitis (including previous sinus surgery), asthma, and sinonasal polyposis had periodic nasal congestion, drainage, and facial pressure. Because his symptoms were worsening despite aggressive medical treatment, bilateral revision FESS with CAS was performed. Computer workstation review of the preoperative CT scans demonstrated an area of right orbital dehiscence. This finding was ascribed to the patient’s previous surgery, although there was no history of an orbital complication. During the initial endoscopy, this area of dehiscence had the appearance of an uncinate process. The CAS probe was used to precisely define the area of dehiscence. Under CAS guidance, access to the posterior ethmoid under the lamina dehiscence was achieved (Fig 6). DISCUSSION
This study reports a large series of patients who underwent FESS with frameless stereotactic surgical
Otolaryngology– Head and Neck Surgery Volume 123 Number 3
Fig 3. Sagittal CT facilitates identification of the residual ethmoid partitions. Each partition was identified sequentially and then removed. In the endoscopic view, localization at a particular partition is seen.
navigation. Recently, the surgical literature has featured numerous reports that describe this use of CAS technology for sinus surgery; the authors of these papers have emphasized hardware, accuracy measurements, and so forth.4-10 In contrast, our report highlights the specific utility of CAS for sinus surgery through the presentation of illustrative cases. In all cases, a surgical navigation system that incorporated optical tracking was used. For this type of tracking, the surgeon must maintain a line of sight between the overhead digitizer (camera array) and the reference frame and surgical instruments; fortunately, instruments that include an offset post for the tracking apparatus satisfy this requirement. A comparison of optical tracking and electromagnetic tracking (Intatrak; Visualization Technology, Woburn, MA) is beyond the scope of this report. The objective of this report is to describe the application of CAS, not the technology of CAS. CAS System Accuracy
Intraoperative CAS registration accuracy was estimated to be 2 mm or better. To determine system accuracy, we compared the CAS instrument tip position indicated by the CAS system with its actual location in the operative field at several areas of the operative volume. The estimated accuracy represents the surgeon’s visual estimate of system accuracy.
OLSON and CITARDI
191
Fig 4. In the endoscopic view, the CAS probe is placed just anterior to the maxillary antrostomy; in this way, the anatomic relationship between the antrostomy and the nasolacrimal system is demonstrated. CAS decreased the risk of inadvertent injury to the nasolacrimal system. (Contrast and brightness of the image have been adjusted digitally.)
Registration accuracy must be assessed by the surgeon during the procedure. It is important to realize that the calculated registration errors and accuracy that CAS systems provide after registration are merely values that may or may not represent true system accuracy. Clinical experience has demonstrated that although “tight” calculated registration errors and accuracy are desirable, these calculated values do not necessarily correlate with real-world registration accuracy. Furthermore, CAS registration may be precise in one area of the operating field but not a second area. For instance, registration may be accurate at the right frontal recess but inaccurate at the left sphenoethmoid recess. In addition, registration may drift if instruments are damaged or the CAS reference frame shifts. For these reasons, the operating surgeon must estimate CAS registration accuracy at several points in the operative field throughout the procedure. Even in revision cases, where many of the usual surgical landmarks are absent, identification of known points is critical to verify registration accuracy. The estimated accuracy of 2 mm in this article is believed to be sufficient for rhinologic procedures, although structures such as the lamina papyracea may be considerably thinner than 1 mm. Even at 2 mm of accuracy, the CAS provides sufficient localization
192
OLSON and CITARDI
Otolaryngology– Head and Neck Surgery September 2000
ations. First, severe sinonasal polyposis may obscure all surgical landmarks; CAS has proved invaluable in these cases. Similarly, CAS has special value in the surgical management of allergic fungal sinusitis, chronic invasive fungal sinusitis, and other disease processes that are characterized by extensive remodeling and/or destruction of surgical landmarks. In addition, revision surgery may be limited by previous surgical misadventures (such as lamina papyracea violation) that increase the risk of the secondary procedure. Rhinologic surgeons also understand that revision sinus surgery requires recognition of landmarks that prior surgery may have significantly altered. CAS facilitates the recognition of these problems and permits the development of a surgical plan for their management. Of course, intraoperative surgical navigation then permits direct correlation of preoperative image planning and assessment with the operative field. Fig 5. Skull base landmark was directly identified, even after middle turbinate resection. Localization at the osteitic roof of the posterior ethmoid is seen.
information because the surgeon also incorporates information from CT scan anatomy (from computerenabled CT review) and the endoscopic views. The modalities of CAS localization, CT scan image review, and endoscopic view are all complementary pieces of information that enhance effective system accuracy beyond the accuracy of any one modality alone. In this way, the net accuracy is beyond the 2-mm reported accuracy of the CAS registration alone—and beyond simple inspection of endoscopic images or CT scans. Specific Utility of CAS in FESS
The assessment of new technology (such as CAS) for particular applications can be problematic. Ideally, quantitative variables may be used, but there are limitations. The effect of the new technology on morbidity rates may require large patient numbers that make such a study impractical. Surgical outcome parameters are difficult to quantify objectively and reliably. For these reasons, it may be preferable to look at the impact of the new technology on treatment implementation. CAS seems to facilitate the surgeon’s comprehension of 3-dimensional anatomy in specific regions. In this way, CAS directly enhances widely practiced FESS techniques. The frontal recess, the sphenoethmoid recess, the sphenoid face, the maxillary sinus, and residual ethmoid cells represent anatomic regions where CAS is most useful. CAS also has special utility in specific clinical situ-
Image-guided FESS Paradigm
On the basis of these clinical experiences, we have formulated a strategy for the application of CAS in FESS. The objective of this effort has been to move beyond the novelty of CAS; the goal has been to directly integrate CAS into FESS. This paradigm has been termed image-guided FESS (IG-FESS) because the images provided by surgical nasal endoscopy and CAS together affect the surgical technique, strategy, and philosophy. Principles of IG-FESS include the following: • Precise anatomic dissection and mucosal dissection are paramount, as numerous authors have described.2,3 • Surgical navigation should not be used as a mere “point-and-hunt” device. • Preoperative CT review is performed on a computer workstation because such software-enabled review provides the surgeon with critical anatomic information that the corresponding static films do not. Through workstation review, the surgeon can more readily develop a 3-dimensional understanding of relevant anatomy. With this information, the surgeon may develop a surgical plan. • Surgical navigation allows the surgeon to directly relate preoperative imaging information and the surgical plan to the operative field. • Intraoperative tracking of surgical instruments, including curved aspirators, microdebriders, pointers, and forceps, directly involves the surgical navigation system in the implementation of the surgical plan. As a result, surgical precision is enhanced. In this way, the CAS unit has shifted from a confirmation aid to critical tool that influences surgical planning and execution.
Otolaryngology– Head and Neck Surgery Volume 123 Number 3
OLSON and CITARDI
193
Fig 6. A, Coronal CT clearly demonstrates a dehiscent region of the right lamina papyracea. Diagnostic endoscopy did not suggest a dehiscence. CAS probe was used to map the area of dehiscence in the operative field. B, Endoscopic view of a posterior ethmoid cell is seen. Localization within this cell is depicted by the crosshairs. Note that the CAS probe has been removed.
IG-FESS has the greatest utility for more complex rhinologic procedures. In fact, CAS should be considered whenever it is difficult to determine the 3-dimensional anatomic relationships from CT data. Indications for this approach include sinonasal polyposis, pansinusitis, and revision surgery. In addition, IGFESS is also useful for surgery at complex anatomic configurations in the frontal recess, skull base, and posterior ethmoid regions. The clinical experiences in this report emphasize the importance of CAS for specific anatomic findings, rather than a global condition (such as polyps). The real strength of CAS is that it brings depth to 2dimensional data, that is, CAS allows surgeons to perceive spatial relationships with a sense of depth previously unobtainable by viewing static films. The need for depth perception is especially great for surgical endoscopy because the telescopes provide only 2-dimensional information. CAS is an enabling technology that allows surgeons to comprehend spatial relationships both more directly and more accurately. IG-FESS Challenges
Surgeons must be cognizant of the limitations of CAS. When registration errors occur, localization accuracy can suffer. System accuracy always must be checked against known landmarks. Software bugs can crash computer systems, and hardware may fail. Human error may aggravate or even cause these problems. Surgeons must know the CAS system that they are
using so they can recognize these issues early and act accordingly. The disclaimer that CAS use is not a substitute for anatomic knowledge and surgical skill bears repeating.5 It is important to remember that CAS relies on preoperative imaging data, rather than on intraoperative imaging. The operating surgeon who uses CAS must always keep these challenges in mind. CONCLUSION
CAS technology includes not only intraoperative surgical navigation but also specific software tools for analysis of preoperative imaging. In rhinologic surgery this software-enabled review of preoperative images facilitates comprehension of 3-dimensional relationships. This article describes a large series of sinus surgery procedures that were performed with CAS. After review of these cases, it became clear that CAS has the greatest utility in specific anatomic locations (ie, frontal sinus, skull base, and sphenoethmoid recess) and in specific clinical situations (ie, revision surgery, sinonasal polyposis, and other scenarios characterized by the loss of surgical landmarks). CAS was critical for these procedures, as the representative cases illustrate; therefore CAS should be directly integrated into the preoperative planning and intraoperative execution of FESS for these more complex cases. The resultant paradigm of image-guided FESS will likely serve to increase surgical effectiveness and decrease overall surgical morbidity.
194
Otolaryngology– Head and Neck Surgery September 2000
OLSON and CITARDI
REFERENCES 1. Meyers RM, Valvassori G. Interpretation of anatomic variation of computed tomography scans of the sinuses: a surgeon’s perspective. Laryngoscope 1998;108:422-5. 2. Kennedy DW. Functional endoscopic sinus surgery (technique). Arch Otolaryngol Head Neck Surg 1985;111:643-9. 3. Kennedy DW. Prognostic factors, outcomes and staging in ethmoid sinus surgery. Laryngoscope 1992;102(Suppl):1-18. 4. Anon JB, Lipman SP, Oppenheim D, et al. Computer-assisted endoscopic sinus surgery. Laryngoscope 1994;104:901-5. 5. Roth M, Lanza DC, Zinreich J, et al. Advantages and disadvantages of three-dimensional computed tomography intraoperative localization for endoscopic sinus surgery. Laryngoscope 1995;105:1279-86. 6. Fried MP, Kleefield J, Gopal H, et al. Image-guided endoscopic
7.
8.
9. 10.
sinus surgery: results of accuracy and performance in a multicenter clinical study using an electromagnetic tracking system. Laryngoscope 1997;107:594-601. Anon JB, Klimek L, Mosges R, et al. Computer-assisted endoscopic sinus surgery (an international review). Otolaryngol Clinic North Am 1997;30:389-400. Fried MP, Kleefield J, Taylor R. New armless image-guidance system for endoscopic sinus surgery. Otolaryngol Head Neck Surg 1998;119:528-32. Anon JB. Computer-aided endoscopic sinus surgery. Laryngoscope 1998;108:949-61. Metson R, Gliklich RE, Cosenza M. A comparison of imageguidance systems for sinus surgery. Laryngoscope 1998;108: 1164-70.
AVAILABILITY OF JOURNAL BACK ISSUES As a service to our subscribers, copies of back issues of Otolaryngology–Head and Neck Surgery for the preceding 5 years are maintained and are available for purchase from Mosby until inventory is depleted. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby, Subscription Customer Service, 6277 Sea Harbor Dr, Orlando, FL 32887, or call 800-654-2452 or 407-345-4000 for information on availability of particular issues and prices. If unavailable from the publisher, photocopies of complete issues may be purchased from Bell & Howell Information and Learning, 300 N Zeeb Rd, Ann Arbor, MI 48106-1346; phone, 734-761-4700 or 800-521-0600.