Fiducial Marker Placement Using Endobronchial Ultrasound and Navigational Bronchoscopy for Stereotactic Radiosurgery: An Alternative Strategy

Fiducial Marker Placement Using Endobronchial Ultrasound and Navigational Bronchoscopy for Stereotactic Radiosurgery: An Alternative Strategy Daniel P. Harley, MD, William S. Krimsky, MD, Saiyad Sarkar, MD, David Highfield, PhD, Cengis Aygun, MD, and Burak Gurses, MD Divisions of Thoracic Surgery and Radiation Oncology, Franklin Square Hospital, and Division of Pulmonary Medicine, Pulmonary and Critical Care Associates of Baltimore (PCCAB), Baltimore, Maryland Background. Stereotactic radiosurgery is being increasingly used to treat patients with early-stage non-small cell lung cancers (NSCLC) who are not candidates for surgical resection. Stereotactic radiosurgery usually needs fiducial markers (FMs) for the tracking process. FMs have generally been placed using percutaneous computed axial tomography scan guidance. We report the results of FM placement using endobronchial ultrasound (EBUS) in 43 patients. Methods. A multidisciplinary tumor board evaluates NSCLC patients before they are offered stereotactic radiosurgery. In patients selected for stereotactic radiosurgery, FMs were inserted into peripheral, central, and mediastinal tumors using EBUS and, in selected patients, navigational bronchoscopy. Patients underwent repeat computed axial tomography chest scans 2 weeks later to ensure stability of the FMs before beginning stereotactic radiosurgery. Results. Included were 43 consecutive patients (21 men, 22 women; mean age, 74.4 years). Forty-two (98%) had NSC carcinomas (5 recurrences); 1 had a carcinoid tumor. Twenty-two tumors were located in the left lung, 19 in the right lung, 1 at the carina, and 1 pretracheal. Two to 5 FMs were placed in and around all tumor masses using EBUS and, for peripheral lesions, EBUS combined with navigational bronchoscopy. Thirty patients had no displacement of FMs. In the 13 who had displaced 1 or more FMs, the ability to use the remaining FMs for stereotactic radiosurgery was unimpaired. Conclusions. EBUS and navigational bronchoscopy are safe and effective methods to position FMs for preparing patients with both central and peripheral lung cancers for stereotactic radiosurgery. (Ann Thorac Surg 2010;89:368 74) 2010 by The Society of Thoracic Surgeons Lung cancer is a global epidemic. In 2002 there were 1.35 million new cases of lung cancer causing 1.18 million deaths [1]. The United States has been a major contributor to these statistics. In 2004 an estimated 358,128 Americans were living with lung cancer. In 2007 an estimated 213,380 new cases were diagnosed, and about 160,392 Americans were expected to die of lung cancer. Approximately 16% of the patients are diagnosed with early-stage lung cancer [2], and theoretically, most of these patients would be candidates for a potentially curative resection. Many of patients with lung cancer are afflicted with severe smoking-induced morbidities such as chronic obstructive pulmonary disease (COPD), and arteriosclerotic heart disease that will preclude any attempts at curative resection. Some will have severe age-related disabilities, and some who would be physiologically able to tolerate the operation will refuse despite impartial presentation of evidence that surgical resection of lung Accepted for publication Sept 17, 2009. Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26 28, 2009. Address correspondence to Dr Krimsky, 9103 Franklin Square Dr, Ste 300, Baltimore, MD 21237; e-mail: william.krimsky@medstar.net. cancer is the optimal method for achieving long-term survivorship [3]. The evolution of newer radiosurgical and bronchoscopic technologies is offering alternatives for the diagnosis and treatment of thoracic malignancies. Stereotactic radiosurgery (SRS) was originally designed as a neurosurgical tool in the 1950s [4], but because of the need for a rigid compression device to limit respiration and a rigid immobilization frame for the thorax [5, 6], it was not used for treatment of lung tumors. The Cyberknife with Synchrony (Accuray Robotic Radiosurgery Systems, Accuray Inc, Sunnyvale, Ca) overcame these difficulties. It did not need an immobilizing frame and allowed for real time respiratory tracking [7, 8]. However, small metallic markers called fiducials (FMs) need to be inserted in or near the target tumor to ensure the accuracy of the delivered radiation dose [9, 10]. The insertion of FMs with minimal morbidity has been problematic. They have been inserted percutaneously using computed tomography (CT) guidance, which has resulted in a pneumothorax rate of at least 13% [11]. This percentage may be low, because CT-guided biopsies have reported pneumothorax rates of 23% to 38% [12, 13], 2010 by The Society of Thoracic Surgeons 0003-4975/10/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2009.09.048

Ann Thorac Surg HARLEY ET AL 2010;89:368 74 FIDUCIAL MARKER PLACEMENT FOR SRS Abbreviations and Acronyms ASHD arteriosclerotic heart disease CKS cyberknife with synchrony COPD chronic obstructive pulmonary disease CT computerized tomography DICOM digital imaging and communications in medicine EBUS endobronchial ultrasound ENB electromagnetic navigational bronchoscopy FFB flexible fiberoptic bronchoscopy FM fiducial markers NSCLC non-small cell lung cancer SRS stereotactic radiosurgery and a more recent series of FM placements using CT guidance reported an incidence of 48% needing pigtail catheter treatment of the pneumothoraces [14]. Flexible fiberoptic bronchoscopy has been used for the placement of FMs, but can result in FM embolization or inadvertent early deployment of the FM [15]. Although flexible fiberoptic bronchoscopy is useful in accessing central lung lesions, it has severe limitations in accessing lesions at the periphery of the lung [16, 17]. At the onset of developing our SRS program, we elected to use endobronchial ultrasound (EBUS) and electromagnetic navigational bronchoscopy (ENB) to minimize the morbidity associated with positioning FMs while simultaneously increasing the number of usable FMs. Between February 2007 and June 2008, 391 patients with histologically proven bronchogenic carcinomas were evaluated for treatment at the Franklin Square Hospital Center in Baltimore, Maryland. The patients were reviewed at multidisciplinary thoracic tumor board before being considered for thoracic SRS. Patients with a forced expiratory volume in one second of less than 40% and a diffusion capacity of the lung of less than 40% were considered to have severe chronic obstructive pulmonary disease (COPD) and in most instances were not considered candidates for surgical resection. The Institutional Review Board of MedStar Inc, the parent corporation of the Franklin Square Hospital Center, approved the data collection and analysis of these patients and waived patient consent for this review. Material and Methods All procedures were performed by 2 interventional pulmonologists trained in both EBUS and ENB. The methodology used was developed by one of the interventional pulmonologists (W.S.K.) and was followed by both physicians. The success rate for FM placement did not differ between these 2 pulmonologists. All patients had noncontrast CT scans of the chest configured with slices of 0.6-mm thickness at 1-mm intervals in the Digital Imaging and Communications in Medicine (DICOM) format. All FM placements were 369 done with the patient under general anesthesia using the oral route. Before the start of each procedure, the patient was positioned on the electromagnetic location board (superdimension Inc, Plymouth, MN) in anticipation of the possible need for ENB. The bronchoscope was an Olympus BF-1T180, 3.0-mm working channel, adult therapeutic bronchoscope (Olympus, Tokyo, Japan). EBUS was used in all of the patients and ENB was used selectively for peripheral tumors that were difficult to access using EBUS alone. Mean time for procedures was 44 11 minutes. EBUS Procedure A 20-MHz radial EBUS probe (UM-S20 20R; Olympus) was used. The miniature probe was inserted into a guide sheath and was advanced as a unit into the working channel of the bronchoscope. This was guided into the bronchus of interest. The probe was extended until the operator sensed resistance and then slowly withdrawn while scanning. Ultrasound imaging of normal air-filled alveolar tissue produces a snowstorm-like whitish appearance. In contrast, in FM placement using EBUS, solid peripheral lesions are darker and more homogeneous. When such images were seen, the target was considered to have been reached. After the peripheral lesion was localized, the probe was withdrawn, leaving the sheath either within or close to the target lesion. The FMs were then deployed through the guide sheath using fluoroscopy, if needed. In selected cases, if a peripheral lesion was difficult to reach, ENB was performed before the FMs were deployed. ENB Procedure The superdimension inreach System (superdimension Inc) was used for the ENB component of the procedures. The initial planning phase involved importing the CT scan into the software (superdimension) using the standard format. Registration points were marked by identifying five or six prominent anatomic landmarks on the virtual bronchoscopic images. The multiplanar images were used to plan a target pathway, starting from the target lesion and working toward a central airway. Endobronchial mapping was conducted, and the software generated a registration error that represents the radius of the expected difference in location between the tip of the sensor probe in the actual patient and where the tip is expected to be. The registration error could then be reduced by repositioning a misplaced landmark or by eliminating the landmarks with the greatest deviation. Navigation was completed by wedging the bronchoscope in the suspected bronchial segment and steering the sensor probe, together with the extended working channel, to the lesion using the multiplanar CT scan images and the tip-view orientation. The target was considered located when a minimum distance ( 1 cm) between the steerable probe tip and the target lesion was achieved. When navigation was completed, the steerable probe was removed. GENERAL THORACIC

370 HARLEY ET AL Ann Thorac Surg FIDUCIAL MARKER PLACEMENT FOR SRS 2010;89:368 74 FM Placement After EBUS confirmation, the EBUS probe was removed from the sheath and gold fiducials (351-2, 351-1; Best Medical International, Fiducial Marker Placement Using EBUS, Springfield, VA) were deployed under fluoroscopic imaging. The fiducials, sizes 0.8 5mmor0.8 3 mm, were wedged into the tip of a needle brush (NB-120; Conmed, Utica, NY) or transbronchial needle (WANG Histology Needle, MW-319; Conmed) respectively. The FM was sealed into the deployment systems using bone wax. This prevented loss of the marker before placement into the extending working channel. Once the needle brush or transbronchial needle appeared at the end of the extended working channel, the FM was deployed using the stylet from the needle systems. To ensure that the marker was embedded into a distal airway/lung parenchyma, a cytology brush (Cellebrity, 1601; Boston Scientific, Natick, MA) was used under fluoroscopic guidance. Roentgenograms were obtained after the procedure to confirm the positions of the FMs as well as the absence of iatrogenic pneumothorax. Confirmation Planning CTs were performed 10 to 14 days after FM placement and reviewed by a thoracic surgeon, radiation oncologist, and nuclear physicist to determine the feasibility of using the Cyberknife with Synchrony guided by the FMs. Results Forty-three patients (21 men and 22 women) who were a mean age of 74.4 years (range 43 to 88 years) with a histologically proven thoracic malignancy were discussed and evaluated at a multidisciplinary lung cancer tumor board before being offered bronchoscopic FM placement and SRS. FM placement was accomplished in all patients using EBUS and was supplemented by ENB in 12 patients with peripheral lesions. Severe COPD was present in 31 patients (according to criteria outlined by the American Thoracic Society), 14 had arteriosclerotic heart disease, and 12 octogenarians had an age-related disability. No patient was selected to undergo SRS as a result of refusing an operation. Non-small cell lung cancer was present in 42 of the patients, 5 were local recurrences, and 1 patient with debilitating COPD had a carcinoid. Nine tumors were centrally located and 34 were peripheral. Twenty-two cancers were located on the left, 19 were on the right, 1 was at the carina, and 1 was a pretracheal recurrence. A total of 161 FMs (average of 3.7 per patient) were deployed, and 139 (86.7%) were identified radiologically when the SRS planning CT scan was performed 2 weeks later. No loss or appreciable movement of any of the deployed FMs was noted in 30 patients (69.7%). The FMs were in or on the tumor at the time of the SRS planning CT scan in 39 patients (90.6%). The FMs that had become displaced were an average of 1.67 cm (SD, 1.15; range, 0.5 to 4.56 cm) from the targeted tumor. The mean tumor size was 2.78 (SD, 1.46 cm; range, 0.9 to 6.5 cm). No patient described FM expectoration and there was no radiologic evidence for contralateral FM aspiration. No pulmonary artery or systemic arterial embolizations were encountered. Although many of the patients experienced a small amount of blood-streaked sputum after the procedure, there were no episodes of clinically significant hemoptysis. A small pneumothorax in 1 patient necessitated placement of a pigtail catheter and hospitalization for 1 day. None of the remaining patients needed postoperative hospitalization or other endoscopic interventions. There were no anesthetic-related complications and there were no hospital readmissions before SRS began. All patients were able to undergo SRS without additional FM placement or other procedures. Comment The evolution of EBUS, ENB, and SRS represent a confluence of dissimilar but complimentary technologies that are facilitating innovative methods of treating thoracic malignancies. There is a large potential cohort of patients with early-stage lung cancer who, because of significant comorbidities such as COPD, severe cardiacrelated disease, and chronic disability, will not be physiologically capable of undergoing an operation. Historically, the early use of SRS was limited to tumors that did not move with respiration, such as in the brain and spinal cord [4]. Tumor motion in patients with lung cancer was a major obstacle for using SRS to treat thoracic cancers. To be effective, SRS has to be able to accurately deliver highly focused beams of radiation from a large number of different angles to a precisely defined area within the lung. The Cyberknife with Synchrony overcame this difficulty by using real-time respiratory motion tracking (Synchrony, Accuray Robotic Radiosurgery Systems) during SRS [6 8]. To ensure accuracy, FMs have to be placed in or near the targeted tumor so that they move with the tumor [9, 10]. If possible, they should be in separate locations and not overlap. Early in the SRS experience, FMs were placed using percutaneous CT guidance. This required a series of punctures and resulted in a predictable pneumothorax rate of 13% to 48% [12 14], and all of these patients needed pigtail catheters. Patients who have severe COPD have very limited ability to tolerate an induced pneumothorax. A pneumothorax in these patients could necessitate a closed tube thoracostomy, an unplanned hospitalization, and higher procedure costs. In contrast, only one small pneumothorax occurred in our series using ENB and EBUS for placement of FMs. The distribution of FMs using CT guidance is frequently suboptimal. Approaching centrally located lesions using this technique can be hazardous and is generally avoided. Flexible fiberoptic bronchoscopy has also been tried for the placement of FMs. This technique may be adequate for some centrally located lesions but has obvious limitations in accessing peripheral lesions,

Ann Thorac Surg HARLEY ET AL 2010;89:368 74 FIDUCIAL MARKER PLACEMENT FOR SRS even with fluoroscopy [14]. The manual deployment of FMs during flexible fiberoptic bronchoscopy is often difficult because no tools are specifically designed for this purpose. When flexible fiberoptic bronchoscopy was used, 12 FMs were dropped before deployment, a rate of 22% [15]. In this scenario, tumor tracking during SRS may have decreased accuracy because of poorly positioned FMs. One pulmonary artery embolization occurred during FM deployment [15]. Although it was clinically insignificant, it does illustrate one of the limitations and potential dangers of placing FMs with a bronchoscopic system that is incapable of identifying pulmonary vascular structures. Deployment of a FM in a pulmonary vein could, in theory, result in a catastrophic FM embolism to any organ, including the heart and brain. Fig 1. The computed tomography scan shows fiducial placement using endobronchial ultrasound and electromagnetic navigational bronchoscopy in a peripheral lesion in a patient with severe bullous disease. (Top: lung windows. Bottom: mediastinal windows.) 371 EBUS and ENB offer a superior methodology for the placement of FMs in central and peripheral pulmonary tumors. EBUS not only identifies tumors and adjacent lymph nodes but also identifies vascular structures within the lung [18]. It identified lymph nodes 86% of the time in diagnostic procedures and achieved a diagnostic yield of 71% [19]. Tumors in the periphery of the lung can be found and biopsied without the use of fluoroscopy [20] or real-time CT imaging. Real-time CT imaging was not performed in this series, although limited fluoroscopy was used to confirm FM placement. This results in potentially lower operating costs to the institution and less radiation exposure to the bronchoscopist and staff. ENB reached peripheral tumors with a diagnostic success rate of 80.3% [21]. Combining these two modalities is advantageous in accurately identifying and biopsying peripheral lesions. The diagnostic yield when used together is significantly increased compared with using either modality alone [22]. A feasibility study of using ENB for FM placement demonstrated successful deployment in 8 of 9 patients (89%) [23]. Figure 1 demonstrates FM placement in a peripheral lesion in a patient with severe bullous disease that was approached using both EBUS and ENB. Although either is capable of providing an avenue for FM placement, the use of both in patients with peripheral lesions can result in a higher success rate. ENB helps overcome the difficulty of navigating tortuous peripheral bronchi that can be an obstacle when EBUS is used alone. EBUS can identify solid areas within the lung that may serve as FM anchors and prevent FM dislodgement. In addition, EBUS used in conjunction with ENB for placement of FMs also allows for interrogation of N1 and N2 lymph nodes during the procedure. When they are found to be positive for malignancy, an unnecessary SRS procedure on the lung can be avoided. We believe that the use of SRS is not indicated in patients with N1 nodal disease because the radiation field will be too large in patients who already have compromised lung function. Although early in our experience a patient with a tumor of 6.5 cm received SRS, we discourage its use in tumors greater than 4 cm because of potential pulmonary toxicity. The use of both EBUS and ENB results in a superior distribution of FMs and helps avoid colinear relations between FMs that can decrease the accuracy of tumor tracking during SRS. In more than 90% of the patients in this series, at least 1 FM was in the selected tumor or was very close to it. This is additional confirmation that these techniques are accurate in obtaining tissue for biopsy and marking tumors for thoracoscopic identification. The FMs used in this series were gold pellets which need a tissue anchor that is provided by the tumor mass or the adjacent lymph nodes. The technique of deployment may also cause an inflammatory reaction resulting in a small amount of fibrosis that may further stabilize the position of the FM. There was, however, a 13.3% loss rate, and this may reflect the smooth surface of the FM and their inability to achieve tissue adherence when placed in alveolar spaces. Placement of FMs within free spaces in the lung was GENERAL THORACIC

372 HARLEY ET AL Ann Thorac Surg FIDUCIAL MARKER PLACEMENT FOR SRS 2010;89:368 74 avoided in this series. FMs will need to be designed to achieve better tissue adherence, which will improve the ability of physicians implanting the markers to achieve conformational geometry of the implanted FM and ensure greater SRS tracking accuracy. It is important that FMs be positioned in solid structures, such as lymph nodes and the selected tumors, because FMs placed in alveolar spaces will result in a loss of the FM. To have enough FMs to ensure SRS tracking, it is necessary to implant at least 3 FMs in centrally located tumors and 4 to 5 FMs in peripheral tumors. Alternative fiducials are currently being investigated at our institution to minimize the number of attempted deployments. Although there is a very definite loss of FMs, the complication rate using EBUS and ENB for FM deployment is markedly lower than the rate using CT guidance. Although general anesthesia is not a necessity for using EBUS and ENB for FM placement, it has the distinct advantage of increasing patient comfort as well as allowing the endoscopist to communicate more freely with the assistant during a complex endoscopic procedure that lasts 1 hour at our institution. In our institution, EBUS and ENB have been performed exclusively by 2 interventional pulmonologists who were trained in this technology and used it diagnostically in several hundred cases before embarking on the SRS program and deploying a FM. This can potentially be an area of contention between invasive pulmonology and thoracic surgery colleagues and a source of turf battles. Effective use of these technologies necessitates an overall team approach with a blurring of traditional specialty barriers, and this has been facilitated by having a multidisciplinary lung cancer tumor board. The users of these technologies will vary from institution to institution, but these should not be casual procedures performed by every pulmonologist and every thoracic surgeon. Ideally, SRS should be able to be performed without the use of FMs, and there is interest in systems that can track tumors without FMs. Xsight Lung (Accuray Inc) has been designed to work with the Cyberknife with Synchrony, does not need FMs, and can target peripheral tumors with a diameter of 15 mm or greater. There are, however, significant limitations to this technology. The targeted tumors must be denser than the surrounding tissue to be recognized by the tracking system. In addition, tumors cannot be superimposed on the spine or the mediastinum radiologically for tracking to be effective. Our institution has had no patients that met these criteria. There has also been interest in Linac-based systems that do not need FMs for the treatment of lung cancers. These systems are often unable to continuously adapt to tumor motion caused by lung and diaphragm movement. Although respiratory gating helps overcome this problem, it is inadequate in many patients and will result in larger treatment margins and associated lung damage than treatments with CK. To date, a system that does not need FMs to perform stereotactic radiosurgery for lung cancer has still to be developed. The use of EBUS with the adjunctive use of ENB in selected cases for the positioning of FMs within the lung requires a team approach. There is a very definite learning curve that can be accomplished by performing diagnostic procedures with EBUS and ENB before using them to implant FMs for SRS. EBUS and ENB, utilized by experienced bronchoscopists, provide a safe and effective approach to placing pulmonary FMs for SRS and should be considered as the methodology of choice. References 1. Parkin D, Bray F, Ferlay J, Pisani P. Global cancer statistics 2002. Ca Cancer J Clinic 2005;55:74 108. 2. American Lung Association. Lung cancer. Available at www. lungusa.org. 3. Ginsberg RJ, Martini N. Non-small cell lung cancer/surgical management. In: Pearson FG, Cooper JD, Deslauries J, et al, ed. Thoracic surgery. 2nd ed. Philadelphia, PA: Churchill Livingstone; 2002:837 59. 4. Leksill L. The stereotactic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316 9. 5. Timsnerman R, Papiez L, McGarry R, et al. Intracranial stereotactic radioablation: results of a phase I study in medically inoperable stage I non-small cell lung cancer. Chest 2003;124:1946 55. 6. Schweikard A, Shiomi H, Adler J. Respiration tracking in radiosurgery. Med Phys 2004;31:2738 41. 7. Adler JR, Chang SD, Murphy MJ, et al. The CyberKnife: a frameless robotic system for radiosurgery. Stereotactic Funct Neurosurg 1997;69:124 8. 8. Kuo J, Yu C, Petrovich Z, et al. The CyberKnife stereotactic radiosurgery system: description, installation and an initial evaluation of use and functionality. Neurosurgery 2003;53: 1235 9. 9. Murphy MJ. Fiducial based targeting accuracy for external beam radiotherapy. Med Phys 2002;29:334 4. 10. Wu X, Main W, Schwade G, et al. Measurement of total clinical accuracy of an image guided radiotherapy system. Med Phys 2004;31:1901. 11. Whyte R, Crowner R, Murphy MJ, et al. Stereotactic radiosurgery for lung tumors: preliminary report of a phase I trial. Ann Thorac Surg 2003;75:1097 101. 12. Geraghty P, Kee S, McFarlane G, et al. CT-guided transthoracic needle aspiration biopsy of pulmonary nodules: needle size and pneumothorax rate. Radiology 2003;229:475 81. 13. Yeow KM, Su IH, Pan K, et al. Risk factors of pneumothorax and bleeding: multivariate analysis of 660 CT-guided coaxial cutting needles lung biopsies. Chest 2004;126:748 54. 14. Pennathur A, Luketich J, Heron D, et al. Stereotactic radiosurgery for treatment of Stage I non small cell lung cancers in high risk patients. J Thorac Cardiovasc Surg 2009;137:597 604. 15. Reichner C, Collin B, Gagnon G, et al. The placement of gold fiducials for cyberknife stereotactic radiosurgery using a modified transbronchial needle aspiration technique. J Bronchol 2005;12:193 5. 16. Libby DM, Smith JP, Altoski NK, et al. Managing the small pulmonary nodule detected by CT. Chest 2004;125:1522 9. 17. Wallace JM, Deutsch AL. Flexible fiberoptic bronchoscopy and percutaneous needle lung aspiration or evaluating the solitary pulmonary nodule. Chest 1982;81:665 70. 18. Becker HD, Herth F. Endobronchial ultrasound of airway and mediastinum. Bolinger CT, Mathus PN, eds. Interventional bronchoscopy. Basel, Switzerland: Karger; 2000: 80 93. 19. Herth FJ, Becker HD, Ernst A. Ultrasound-guided transbronchial needle aspiration. Chest 2003;123:604 7.

Ann Thorac Surg HARLEY ET AL 2010;89:368 74 FIDUCIAL MARKER PLACEMENT FOR SRS 20. Becker HD, Herth F. Endobronchial ultrasound. In: Braunwald E, Fauci AS, Isselbaches KJ, et al, eds. Harrison s principles of internal medicine monograph online, 1999. http://www.harrisonsonline.com. 21. Gildea TR, Mazzone PJ, Karnak D, et al. Electromagnetic navigation diagnostic bronchoscopy: a prospective study. Am J Respir Crit Care Med 2006;174:982 9. 373 22. Eberhardt R, Anantham D, Ernst A, et al. Multimodality bronchoscopic diagnosis of peripheral lung lesions. Am J Respir Crit Care Med 2007;176:36 41. 23. Anantham D, Feller-Kopman D, Shanmugham L, et al. Electromagnetic navigation bronchoscopy guided fiducial placement for robotic stereotactic radiosurgery of lung tumors, a feasibility study. Chest 2007;132:930 5. GENERAL THORACIC DISCUSSION DR SHANDA H. BLACKMON (Houston, Texas): Dr Krimsky, did you see any tumor seeding from the endobronchial ultrasound [EBUS] placement of the fiducial going through the airway directly into a tumor adjacent to the trachea or the main stem bronchus? It might potentially pull a tumor back into that airway. Did you see any of those patients, and if you did, did they require any stenting? DR KRIMSKY: Actually, Shanda, no, we did not see that. I was able under bronchoscopic ultrasound guidance to actually place the needle in and around these areas with a great deal of specificity. So, no, we were able to avoid that. DR BLACKMON: Was there anything special you did to prevent seeding of the inside of the airway besides putting a piece of maybe the wax to push the fiducial out and prevent anything from coming back? DR KRIMSKY: No, nothing. DR MARK KRASNA (Towson, MD): I congratulate Dan on an excellent presentation. Bill, I think you should be congratulated in getting a tissue diagnosis on every one of these patients first. I think that really actually is a key, but it does lead to my second question, which maybe you and Dan together need to answer. Can you define for us in a little bit more detail what made these patients, what we will call medically inoperable, meaning some detail of the degree of their COPD [chronic obstructive pulmonary disease]. Part of what is going to be important in assessing this technology going forward prospectively with IRBs [Investigational Review Boards], like you have done, is to really come up with the common denominators of who is that patient who is too risky to undergo a resection or some other treatment. DR KRIMSKY: If you don t mind me saying, I think there is actually a second point there, which is that often we get these referrals from an outside facility, so we are pretty disciplined about making sure we restage the mediastinum either with endobronchial ultrasound or via mediastinoscopy, and that is something that actually we discuss at a tumor board ahead of time. I want to be clear about that. Then I think the other way that this is done is to make sure that these are presented at a multidisciplinary tumor board whereby thoracic surgery, pulmonology, radiology, radiation oncology, et cetera, are all present and definitions then are made at that point. So this is not decided in abstract, but rather in a public forum, if you will. DR KRASNA: Do you have PFT [pulmonary function test] data available? DR KRIMSKY: For those patients, yes, for the patients that were felt to be medically inoperable as a result of their underlying lung disabilities, yes. What they are right now, unfortunately, I don t have them off the top of my head; but regardless, as you know, these patients are often complex. I have heard folks already talking about the difference between a 19-year-old and an 86-year-old, what they look like at the door vs what their pulmonary function tests say. So I think it is a complex answer. DR PHILIP A. LINDEN (Cleveland, OH): I saw that the majority of your lesions were peripheral, I think 34 peripheral and 9 central, but you describe using EBUS, which is mainly a more central technology, to position the peripheral fiducials. Did you rely a lot on fluoroscopy as an adjunct, or how did you do it? DR KRIMSKY: No. I think it is a very astute observation. Part of it was that during that time, if you will note, electromagnetic navigational bronchoscopy [ENB] was not available, and that was during a recall period. So what we were forced to do was rely upon both virtual bronchoscopy and then using those data to inform which particular subsegments we would put a guide sheath with a radial ultrasound probe within it and then proceed to confirm position of that once we got radial ultrasound confirmation that we were in or around the lesion, and obviously we then began to use electromagnetic navigational bronchoscopy when that came back online, and for several reasons. One is that it actually allows us to put markers in various places I think more succinctly and quickly. DR LINDEN: I know no patient expectorated any fiducials, but were you able to tell on the planning CT [computed tomography] scan how many remained and how many were lost? DR KRIMSKY: I think Dr Harley actually did that. Yes, I think of the 160-something, I think they coughed out, lost whatever, and I tried to place more, obviously, knowing that there was a likelihood of losing one or two of these. DR ROBERT J. CERFOLIO (Birmingham, AL): I have two questions for you. Number one, wouldn t you agree that ENB really is the future and it is going to replace EBUS? DR KRIMSKY: I think for peripheral lesions. DR CERFOLIO: I mean even for staging the mediastinum. DR KRIMSKY: ENB vs EBUS? DR CERFOLIO: Yes. DR KRIMSKY: Actually, I would disagree with that. DR CERFOLIO: I am talking about the future now. Obviously it is going to replace it for peripheral nodules. DR KRIMSKY: Sure. DR CERFOLIO: But don t you foresee that ENB is going to maybe replace EBUS?

374 HARLEY ET AL Ann Thorac Surg FIDUCIAL MARKER PLACEMENT FOR SRS 2010;89:368 74 DR KRIMSKY: But is there a real-time component to ENB, and I think that is always going to be a problem. DR CERFOLIO: And the second question is the reimbursement for EBUS and ENB. Would you tell the audience the huge difference? DR KRIMSKY: Yes. Right now the reimbursement for ENB is zero, I think. Is that correct? DR CERFOLIO: But they are getting their own CPT [Current Procedural Terminology] code and the reimbursement may be up to $1,200 to $1,400, and EBUS is under $100 isn t it? DR KRIMSKY: It is $70. DR CERFOLIO: $70. So I just want to make sure the audience is listening to that very carefully and the costs difference need to be weighed so we can keep out costs down when we can. disease discovered on either ENB or EBUS who then did not proceed to stereotactic radiosurgery, but the fiducial markers were implanted, keeping in mind that those cost about $25. DR CERFOLIO: How many surgeons here in their centers are using ENB, electromagnetic navigational bronchoscopy? (A show of hands.) DR CERFOLIO: Good. How many are trying to get it in their center? (A show of hands.) DR CERFOLIO: Six, 7, 8 hands. Good. I think that it is going to take the peripheral nodule guys and put it back in our court. So I think we need to be involved in it and control it the best we can. We need to buy it, we need to own it, and we need to market it. I hate to say it, but that s the reality of our specialty and life. DR KRIMSKY: Yes, yes. Thank you very much. DR JESSICA S. DONINGTON (New York, NY): How long do these procedures take? DR KRIMSKY: What we are doing the entire procedure, EBUS and the electromagnetic navigational bronchoscopy about an hour. It is relatively quick. We do a lot of these. You know, we are able to stage the mediastinum. We do put the fiducial markers in. One thing that I don t think has been made clear is we had about 7 patients who actually had N1 or N2 nodal DR KRIMSKY: If you want to learn, please come to Franklin Square. We are happy to teach. There is a lot of folks who have already come. DR DONINGTON: I would like to make one more comment on that paper. I think this work nicely demonstrates the value of collaboration with our peers in the diagnosis and staging of early-stage lung cancer. I commend both of you on the very nice presentation. DR CERFOLIO: I agree.