- Email: [email protected]
A Novel Technique for Localization of Small Pulmonary Nodules
Original Research CHEST IMAGING
A Novel Technique for Localization of Small Pulmonary Nodules* Weisheng Chen, MD; Long Chen, MD; Shengsheng Yang, MD; Ziqian Chen, MD; Gengnian Qian, MD; Suxun Zhang, MD; and Junjie Jing, MD
Background: To show the safety and accuracy of a new marking technique using an image-guided technique for preoperative localization of a small pulmonary nodule. Methods: CT data of a patient with a peripheral pulmonary nodule < 20 mm were transmitted to a surgical navigation system (StealthStation Treon Treatment Guidance System; Medtronic; Louisville, KY). To match preoperative CT image data to the physical space occupied by the patient during surgery, five to six superficial skin fiducials were used for registration. A 16-gauge needle attached by a positioning sensor was advanced into or immediately adjacent to the nodule for injection of methylene blue under guidance of the StealthStation system. Then the lesion marked by the methylene was thoracoscopically resected. Results: Seventeen patients (12 men and 5 women; mean age, 51.3 years) underwent this procedure, and all the nodules were identified due to the precise location of the probe. They were resected with sufficient margins. There were no surgical complications. The average time of registration was 4.8 ⴞ 0.9 min (ⴞ SD). Registration error was on average 2.7 ⴞ 0.2 mm. Conclusions: Image-guided navigation is useful, accurate, and safe in the localization of small peripheral lung lesions. (CHEST 2007; 131:1526 –1531) Key words: image-guided navigation; preoperative localization; small pulmonary nodules; thoracoscopy Abbreviations: DIBH ⫽ deep inspiration breath-hold; IGS ⫽ image-guided surgery; SPN ⫽ small pulmonary nodule
the recent progress of CT examination, small W ithpulmonary nodules (SPNs) have been able to be increasingly detected. Thoracoscopic wedge resection is an appropriate procedure for such lesions because it is both for diagnostic and therapeutic purposes. However, the thoracoscope only provides the surgeon a limited and two-dimensional operative view, and will not give any information beyond the surfaces of the organs. Such small lesions (⬍ 20 mm) may be too small *From the Departments of Thoracic and Cardiovascular Surgery (Drs. W. Chen, L. Chen, Yang, and Zhang), Radiology (Drs. Z. Chen and Qian), and Neurosurgery (Dr. Jing), Fuzhou General Hospital, Fuzhou, China. This project was supported by the Natural Science Foundation of Fujian Province of China (No. 2006J0369, 2006J0041). The authors have no conflicts of interest to disclose. Manuscript received April 25, 2006; revision accepted December 26, 2006. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Long Chen, MD, Department of Thoracic and Cardiovascular Surgery, Fuzhou General Hospital, No. 156# Xierhuan, Fuzhou, China 350025; e-mail: chenlongdirector@ yahoo.com DOI: 10.1378/chest.06-1017 1526
to be found using the videoscope, and too soft to be distinguished with the touch of a finger. We have learned that for small and deep pulmonary nodules, localization techniques are necessary. In these cases, most surgeons generally use several kinds of localization techniques, including injection of contrast media, radiotracer, methylene blue, and preoperative percutaneously placed hooks of various design.1,2 In this article, we will introduce a reliable and new technique using an optical-based, image-guided, surgical navigation system (StealthStation Treon Treatment Guidance System; Medtronic; Louisville, KY) for the localization of small lung nodules and subsequent thoracoscopic excisional biopsy (Fig 1). The StealthStation consists of a computer workstation, image-processing software, a display monitor, a localization system, and specialized, trackable instrumentation. Materials and Methods From November 2004 to January 2006, 17 patients with recently diagnosed peripheral SPNs on CT were enrolled in this pilot study. There were 5 women and 12 men (average age, 51.3 Original Research
Table 1—Patients and Nodule Characteristics* Characteristics
Data
Male/female gender, No. Age (range), yr Maximum tumor diameter at CT, mm Distance from the pleura at CT, mm Registration error, mm Histology, No. Benign lesions Primary lung tumors Metastases Location of lesions, No. Right upper lobe Right lower lobe Left upper lobe Left lower lobe
12/5 51.3 (32–73) 8–20 (14.8 ⫾ 1.1) 10–25 (16.7 ⫾ 1.1) 1.7–4.5 (2.7 ⫾ 0.2) 10 5 2 6 2 6 3
*Data are presented as range (mean ⫾ SD) unless otherwise indicated.
Fuzhou General Hospital. All patients were given detailed descriptions of the examination and were informed that this was a new approach. Informed consent was obtained in all cases. CT Scans
Figure 1. Top: StealthStation Treon Treatment Guidance System with optical tracking system cameras. Bottom: A syringe with a position-tracking device attached.
years) [Table 1]. Nodules in seven patients were discovered during follow-up of other diseases, and six patients were free of symptoms and had chest radiographs done for nononcologic reasons (routine medical checkups, insurance, preoperative for other pathologies). The remaining four patients had symptoms of coughs or fevers. Among the 17 patients, 4 patients presented with a history of previously diagnosed cancer, 10 patients had smoking history, and 3 patients were referred to thoracoscopic surgery after lack of success of less-invasive procedures, such as CT-guided percutaneous fine-needle lung aspirate biopsy. Patient selection was based on the clinical assessment of the anticipated difficulty in thoracoscopically locating the nodule. This assessment included either the nodule size, its depth from the parietal pleural surface, or a combination of these factors. The enrollment criteria of this study were as follows: (1) a peripheral pulmonary nodule ⬍ 1.0 cm and not in contact with the visceral pleura; and (2) a peripheral pulmonary nodule between 1.0 and 2.0 cm in maximum diameter that was ⬎ 10 mm away from the pleural surface. The study was approved by the Institutional Review Board of www.chestjournal.org
The deep inspiration breath-hold (DIBH) technique described by Mah et al3 was used to coach the patient to the same reproducible deep inspiration level during CT scanning. To familiarize with the procedure, the patients were instructed to repeat the maneuver three to four times. The reproducibility of the maneuver as determined by the spirometry level was carefully monitored. The patient underwent a CT scan performed by the radiologist 1 to 3 days before the operation. Before the CT scan, five to six superficial skin fiducials (marker) were placed on the chest wall for later image registration. As best as possible, markers were placed in areas unlikely to change with patient position. Therefore, bony surfaces (eg, clavicles and sternum) were preferred to flexible soft tissue. The CT images were obtained with a multidetector helical CT scanner (Discovery LS16; GE; Milwaukee, WI) with the following parameters: collimation, 2.5 mm; pitch, 1.375:1; and rotation time, 0.5 s. The scans were performed while the patient was in DIBH, and the volume of inhaled air was measured by spirometry. Registration Registration is the process of matching preoperative CT image data (virtual) to the physical space occupied by the patient during surgery. The accurate correlation (ie, matching) of these two data sets subsequently allows localization of the surgical tools within the operative space. Proper registration of the real surgical field to the image data sets in the computer is therefore essential before initiating image-guided surgery (IGS), in order to provide accurate information and effective localization and navigation. On the operation day, the CT images were sent electronically to the navigation computer in the operation room. After the patient was intubated with a dual-lumen endotracheal tube and placed on mechanical ventilation under general anesthesia, registration was performed by touching the markers with the probe while indicating to the surgical workstation where marker was being touched (Fig 2). CHEST / 131 / 5 / MAY, 2007
1527
Figure 2. Registration required touching each of the markers with the probe and then simultaneously indicating the structure to the computer.
Localization After the registration was completed, the surgical plan was developed using the StealthStation system. The entry point, target point, and border of the lesion were defined. CT images were used to guide a 16-gauge needle attached by a positioning sensor (Fig 1) to the appropriate areas for injection of the
methylene blue. As the position sensor-equipped needle was punctured into the lung, crosshairs indicated its distal tip position and moved on the CT display (Fig 3). Coronal, sagittal, axial, and three-dimensional images were displayed, along with a reconstructed image parallel to the orientation of the tip. When the needle was advanced into or immediately adjacent to the lung nodule, 1 to 2 mL of methylene blue was injected by the surgeon.
Figure 3. The computer display of the navigation system provided trajectory views. Crosshairs indicated the real-time position of the tip of the needle on the two-dimensional views (the tip of the needle had reached the area surrounding the nodule). 1528
Original Research
Both the registration and the localization were performed during DIBH, which was simulated by inflating the lung with artificial ventilation under general anesthesia. The volume of the inhalation of air was equal to that in the CT scans. Thoracoscopic Biopsy The patient subsequently was positioned for thoracoscopic biopsy with single-lung ventilation of the contralateral lung. We introduced the first trocar for the videothoracoscope, usually in the sixth or seventh intercostal space along the midaxillary line. The methylene blue in the lung surface allowed rapid identification of the approximate nodule location. Another two ports were placed anteriorly and posteriorly depending on the location of the nodule in the lung parenchyma. The lung surface exhibiting methylene blue was grasped by the endoscopic forceps, and was followed by digital or instrumental palpation to further confirm the exact localization of the nodule. Then excisional biopsy was performed by an articulating endostapler. The depth of the wedge—the inclination of the stapler—was estimated based on the depth of the nodule on CT, in order to include a safety margin of between 5 mm and 10 mm. The resected nodule was extracted in a plastic bag to avoid neoplastic cell dissemination at the port site. Before sent for frozen-section pathologic examination, biopsy specimens were split open in the operating room to ensure the presence of the entire nodule. When the frozen-section pathologic diagnosis was a benign nodule or metastatic disease, a chest tube was inserted and the operation was finished. When the finding was primitive lung cancer, a major pulmonary resections was performed according to oncology standards.
Results The average time of registration was 4.8 ⫾ 0.9 min (⫾ SD). Registration error was on average 2.7 ⫾ 0.2 mm. The time from the localization to the initial surgical skin incision was 20.8 ⫾ 3.9 min. The size of the SPN at the CT scan was 14.8 ⫾ 1.1 mm (range, 8 to 20 mm). The distance from the pleural surface to the SPN was 16.7 ⫾ 1.1 mm (range, 10 to 25 mm). Lesions were found in the right upper lobe (n ⫽ 6), right lower lobe (n ⫽ 2), left upper lobe (n ⫽ 6), and left lower lobe (n ⫽ 3) at the time of operation (Table 1). All the lesions were successfully localized in the thoracoscope image, and the methylene blue-stained area was found overlying the nodule in all cases during careful macroscopic examination of the resected specimens. No spillage of dye occurred during the procedure that could potentially stain a larger amount of the pleural surface, thus making thoracoscopic visualization more difficult. Five lesions were barely palpable as a slight solid mass by endoscopic devices during surgery, but the remaining 12 lesions could not be detected with instrument. All resection margins were microscopically clear confirmed by the pathologic examination. There were no surgical complications. The chest tubes were removed within 2 to www.chestjournal.org
4 days postoperatively. All patients were discharged on postoperative day 6 on average (range, 3 to 11 days). Histologically, 7 nodules were malignant and 10 were benign. Benign diagnoses included acid-fast bacillus (n ⫽ 1), granuloma (n ⫽ 2), hamartoma (n ⫽ 4), aspergilloma (n ⫽ 1), and inflammatory pseudotumor (n ⫽ 2). Malignant diagnoses included primary adenocarcinoma (n ⫽ 3), squamous cell carcinoma (n ⫽ 2), and pulmonary metastasis (n ⫽ 2).
Discussion A major problem of conducting a successful thoracoscopic resection is to locate small or deeply situated target nodules.4 A few localization techniques have been developed, either preoperative or intrathoracoscopic.4,5 Although all of the methods are effective in locating the target lesion, their limitations are obvious. The needle-wire technique is associated with several complications, such as lung hemorrhage, pleuritic pain, and pneumothorax. The hookwire dislodgement rate of 7.5% was reported due to patient’s breath or the deflated lung during single-lung ventilation.5 A failure rate of approximately 13% for preoperative methylene blue injection under the CT fluoroscopic guidance has been reported due to either an excess of liquid injection or an error in nodule localization.6 It is found that the dye frequently dissipates over a large area by the time the surgical procedure is done, making its localization features inadequate. The localization of pulmonary nodules by radio-guided technique also reveals some drawbacks. One problem is the notable and fast diffusion of contrast medium in the pulmonary parenchyma surrounding the nodule, due to the rich vascularization of the lung. A second problem is locating deep and posterior nodules due to the dimension and structure of the probe, which cannot move freely in the thorax.7 Moreover, using special radiation protection precautions in the operating room or in handling the surgical specimens is be required. However, all the techniques are performed under the CT fluoroscopic guidance. As a result, they require significant CT scanner time and use additional radiation. Radiation exposure for both patients and clinician remains a major concern. The intrathoracoscopic ultrasound technique also has some limitations. The chief difficulty was in obtaining an image as long as any air remained in the lung. Recent advances in IGS have been brought about by the development of computer technology to make minimally invasive techniques safer and more accurate. The primary advantage of IGS is that it provides the surgeon with accurate knowledge of the spatial CHEST / 131 / 5 / MAY, 2007
1529
relationship between the surgical instrument or appliance and the surrounding structures not directly visible in the operative field.8 The image-guided navigation technique is popular in many surgical fields, such as neurosurgery and orthopedics. The navigation technique also has been proved safe and effective in the three-dimensional CT-guided bronchoscopy with a real-time electromagnetic positioning sensor.9 In this study, we demonstrated the feasibility of performing thoracoscopic biopsy using image-guided navigation. The results of our study suggest that the technique provides precise localization. The imageguided procedure does not require CT fluoroscopic guidance. One major limitation of CT fluoroscopic guidance is the relatively high radiation exposure to patient and surgeon.10 Consequently, there was no risk of high radiation exposure to patient and surgeon in the image-guided procedure, compared with the CT fluoroscopy-guided localization techniques. The complications of hemorrhaging and pneumothorax may occur during the image-guided procedure, but the surgeon does not has to worry about complications as in other localization procedure. For most of preoperative CT-guided percutaneous localization techniques, the patients had to wait for a long time before undergoing thoracoscopy after the localization procedure, owing to surgical suite or CT room scheduling issues. During the waiting time, the surgeon usually worried about the lung hemorrhage and pneumothorax becoming more serious and serious, which was very dangerous to the patient with poor lung function. However, our image-guided procedure was performed in surgical suite after the patient had been receiving general anesthesia and mechanical ventilation. The patient was under the control of the surgeon and the anesthetist. Moreover, the time from the localization to the initial surgical skin incision was not more than 30 min. As a result, the surgeon’s worries about the hemorrhaging and pneumothorax lessened to a great extent. Most of the localization techniques use preoperative CTguided percutaneous punctures with local anesthesia. Patients who were kept fixed during the procedure suffered from mental strain and anxiety for the pain and surgical complication. They needed sedatives. However, our image-guided localization is performed under general anesthesia. The patient has no sense of fear associated with surgery, while the surgeon does not have to worry about coughing or the body movements of the patient, which will handicap the biopsy procedure. The technology is not complicated. The accompanying equipment for the StealthStation navigation system does not occupy a large space in the operating room. Furthermore, this navigation system—the 1530
surgical global positioning satellite—provides highly accurate real-time information to the surgeon regarding localization, trajectory, and the depth of the instruments end, and therefore is promising in the clinic, such as for localization and biopsy of mediastinal lymph node metastasis. The biggest limitation of our method remained that the movement of the diaphragm and the chest wall during respiration affected the outcome of the localization. We used the DIBH technique to increase the accuracy of the localization. Mah et al3 reported that the DIBH technique could significantly reduce the breathing-induced tumor motion. Moreover, the patient’s volume of inhalation of air during the localization was equal to that of in the CT scans. As a result, consistent lung inflation levels were achieved in both the CT scans and imageguided localization, and the reproducibility of tumor position was satisfactorily accurate. In other word, the respiratory-induced target motion was minimized. In addition, many reports11,12 have shown a variety of methods to improve the accuracy of treatment for moving tumors. Giraud et al12 reported that a commercial gating system using external markers for respiration synchronized CT scans significantly improved the positional reproducibility of thoracic and upper abdominal structures. Accurate intraoperative real-time navigation also is a safe and practicable solution for minimizing respiratory-induced target motion.13 In conclusion, our research has demonstrated that image-guided marking of peripheral lung lesions is a feasible and promising technique that provides appropriate guidance and proves effective in immediately facilitating subsequent thoracoscopic resection. ACKNOWLEDGMENT: We thank Thomas Lang, PhD (Research Scientist) and Frank Lindseth (Research Scientist) from SINTEF Health Research, Trondheim, Norway, for their help. We also thank Xiong Shengchun, MD, Zhuang Congwen, MD, Lin Jinxiang, MD, Xu Chi, MD, and Liu Daoming, MD, from the Department of Thoracic and Cardiovascular Surgery in the Fuzhou General Hospital for their support.
References 1 Saito H, Minamiya Y, Matsuzaki I, et al. Indication for preoperative localization of small peripheral pulmonary nodules in thoracoscopic surgery. J Thorac Cardiovasc Surg 2002; 124:1198 –1202 2 Sortini D, Feo CV, Carcoforo P, et al. Thoracoscopic localization techniques for patients with solitary pulmonary nodule and history of malignancy. Ann Thorac Surg 2005; 79:258 – 262 3 Mah D, Hanley J, Rosenzweig KE, et al. Technical aspects of the deep inspiration breath-hold technique in the treatment of thoracic cancer. Int J Radiat Oncol Biol Phys 2000; 48:1175–1185 4 Partik BL, Leung AN, Muller MR, et al. Using a dedicated Original Research
5
6 7 8 9
lung-marker system for localization of pulmonary nodules before thoracoscopic surgery. AJR Am J Roentgenol 2003; 180:805– 809 Ciriaco P, Negri G, Puglisi A, et al. Video-assisted thoracoscopic surgery for pulmonary nodules: rationale for preoperative computed tomography-guided hookwire localization. Eur J Cardiothorac Surg 2004; 25:429 – 433 Kerrigan DC, Spence PA, Crittenden MD, et al. Methylene blue guidance for simplified resection of a lung lesion. Ann Thorac Surg 1992; 53:163–164 Carcoforo P, Feo C, Sortini D, et al. Localization of pulmonary nodules. Chest 2004; 125:796 –797 Mayberg MR, LaPresto E, Cunningham EJ. Image-guided endoscopy: description of technique and potential applications. Neurosurg Focus 2005; 19:E10 Hautmann H, Schneider A, Pinkau T, et al. Electromagnetic catheter navigation during bronchoscopy: validation of a novel
www.chestjournal.org
10
11
12 13
method by conventional fluoroscopy. Chest 2005; 128:382– 387 Solomon SB, Patriciu A, Bohlman ME, et al. Robotically driven interventions: a method of using CT fluoroscopy without radiation exposure to the physician. Radiology 2002; 225:277–282 Onishi H, Kuriyama K, Komiyama T, et al. CT evaluation of patient deep inspiration self-breath-holding: how precisely can patients reproduce the tumor position in the absence of respiratory monitoring devices? Med Phys 2003; 30:1183– 1187 Giraud P, Yorke E, Ford EC, et al. Reduction of organ motion in lung tumors with respiratory gating. Lung Cancer 2006; 51:41–51 Reittner P, Tillich M, Luxenberger W, et al. Multislice CT-image-guided endoscopic sinus surgery using an electromagnetic tracking system. Eur Radiol 2002; 12:592–596
CHEST / 131 / 5 / MAY, 2007
1531
A Novel Technique for Localization of Small Pulmonary Nodules* Weisheng Chen, MD; Long Chen, MD; Shengsheng Yang, MD; Ziqian Chen, MD; Gengnian Qian, MD; Suxun Zhang, MD; and Junjie Jing, MD
Background: To show the safety and accuracy of a new marking technique using an image-guided technique for preoperative localization of a small pulmonary nodule. Methods: CT data of a patient with a peripheral pulmonary nodule < 20 mm were transmitted to a surgical navigation system (StealthStation Treon Treatment Guidance System; Medtronic; Louisville, KY). To match preoperative CT image data to the physical space occupied by the patient during surgery, five to six superficial skin fiducials were used for registration. A 16-gauge needle attached by a positioning sensor was advanced into or immediately adjacent to the nodule for injection of methylene blue under guidance of the StealthStation system. Then the lesion marked by the methylene was thoracoscopically resected. Results: Seventeen patients (12 men and 5 women; mean age, 51.3 years) underwent this procedure, and all the nodules were identified due to the precise location of the probe. They were resected with sufficient margins. There were no surgical complications. The average time of registration was 4.8 ⴞ 0.9 min (ⴞ SD). Registration error was on average 2.7 ⴞ 0.2 mm. Conclusions: Image-guided navigation is useful, accurate, and safe in the localization of small peripheral lung lesions. (CHEST 2007; 131:1526 –1531) Key words: image-guided navigation; preoperative localization; small pulmonary nodules; thoracoscopy Abbreviations: DIBH ⫽ deep inspiration breath-hold; IGS ⫽ image-guided surgery; SPN ⫽ small pulmonary nodule
the recent progress of CT examination, small W ithpulmonary nodules (SPNs) have been able to be increasingly detected. Thoracoscopic wedge resection is an appropriate procedure for such lesions because it is both for diagnostic and therapeutic purposes. However, the thoracoscope only provides the surgeon a limited and two-dimensional operative view, and will not give any information beyond the surfaces of the organs. Such small lesions (⬍ 20 mm) may be too small *From the Departments of Thoracic and Cardiovascular Surgery (Drs. W. Chen, L. Chen, Yang, and Zhang), Radiology (Drs. Z. Chen and Qian), and Neurosurgery (Dr. Jing), Fuzhou General Hospital, Fuzhou, China. This project was supported by the Natural Science Foundation of Fujian Province of China (No. 2006J0369, 2006J0041). The authors have no conflicts of interest to disclose. Manuscript received April 25, 2006; revision accepted December 26, 2006. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Long Chen, MD, Department of Thoracic and Cardiovascular Surgery, Fuzhou General Hospital, No. 156# Xierhuan, Fuzhou, China 350025; e-mail: chenlongdirector@ yahoo.com DOI: 10.1378/chest.06-1017 1526
to be found using the videoscope, and too soft to be distinguished with the touch of a finger. We have learned that for small and deep pulmonary nodules, localization techniques are necessary. In these cases, most surgeons generally use several kinds of localization techniques, including injection of contrast media, radiotracer, methylene blue, and preoperative percutaneously placed hooks of various design.1,2 In this article, we will introduce a reliable and new technique using an optical-based, image-guided, surgical navigation system (StealthStation Treon Treatment Guidance System; Medtronic; Louisville, KY) for the localization of small lung nodules and subsequent thoracoscopic excisional biopsy (Fig 1). The StealthStation consists of a computer workstation, image-processing software, a display monitor, a localization system, and specialized, trackable instrumentation. Materials and Methods From November 2004 to January 2006, 17 patients with recently diagnosed peripheral SPNs on CT were enrolled in this pilot study. There were 5 women and 12 men (average age, 51.3 Original Research
Table 1—Patients and Nodule Characteristics* Characteristics
Data
Male/female gender, No. Age (range), yr Maximum tumor diameter at CT, mm Distance from the pleura at CT, mm Registration error, mm Histology, No. Benign lesions Primary lung tumors Metastases Location of lesions, No. Right upper lobe Right lower lobe Left upper lobe Left lower lobe
12/5 51.3 (32–73) 8–20 (14.8 ⫾ 1.1) 10–25 (16.7 ⫾ 1.1) 1.7–4.5 (2.7 ⫾ 0.2) 10 5 2 6 2 6 3
*Data are presented as range (mean ⫾ SD) unless otherwise indicated.
Fuzhou General Hospital. All patients were given detailed descriptions of the examination and were informed that this was a new approach. Informed consent was obtained in all cases. CT Scans
Figure 1. Top: StealthStation Treon Treatment Guidance System with optical tracking system cameras. Bottom: A syringe with a position-tracking device attached.
years) [Table 1]. Nodules in seven patients were discovered during follow-up of other diseases, and six patients were free of symptoms and had chest radiographs done for nononcologic reasons (routine medical checkups, insurance, preoperative for other pathologies). The remaining four patients had symptoms of coughs or fevers. Among the 17 patients, 4 patients presented with a history of previously diagnosed cancer, 10 patients had smoking history, and 3 patients were referred to thoracoscopic surgery after lack of success of less-invasive procedures, such as CT-guided percutaneous fine-needle lung aspirate biopsy. Patient selection was based on the clinical assessment of the anticipated difficulty in thoracoscopically locating the nodule. This assessment included either the nodule size, its depth from the parietal pleural surface, or a combination of these factors. The enrollment criteria of this study were as follows: (1) a peripheral pulmonary nodule ⬍ 1.0 cm and not in contact with the visceral pleura; and (2) a peripheral pulmonary nodule between 1.0 and 2.0 cm in maximum diameter that was ⬎ 10 mm away from the pleural surface. The study was approved by the Institutional Review Board of www.chestjournal.org
The deep inspiration breath-hold (DIBH) technique described by Mah et al3 was used to coach the patient to the same reproducible deep inspiration level during CT scanning. To familiarize with the procedure, the patients were instructed to repeat the maneuver three to four times. The reproducibility of the maneuver as determined by the spirometry level was carefully monitored. The patient underwent a CT scan performed by the radiologist 1 to 3 days before the operation. Before the CT scan, five to six superficial skin fiducials (marker) were placed on the chest wall for later image registration. As best as possible, markers were placed in areas unlikely to change with patient position. Therefore, bony surfaces (eg, clavicles and sternum) were preferred to flexible soft tissue. The CT images were obtained with a multidetector helical CT scanner (Discovery LS16; GE; Milwaukee, WI) with the following parameters: collimation, 2.5 mm; pitch, 1.375:1; and rotation time, 0.5 s. The scans were performed while the patient was in DIBH, and the volume of inhaled air was measured by spirometry. Registration Registration is the process of matching preoperative CT image data (virtual) to the physical space occupied by the patient during surgery. The accurate correlation (ie, matching) of these two data sets subsequently allows localization of the surgical tools within the operative space. Proper registration of the real surgical field to the image data sets in the computer is therefore essential before initiating image-guided surgery (IGS), in order to provide accurate information and effective localization and navigation. On the operation day, the CT images were sent electronically to the navigation computer in the operation room. After the patient was intubated with a dual-lumen endotracheal tube and placed on mechanical ventilation under general anesthesia, registration was performed by touching the markers with the probe while indicating to the surgical workstation where marker was being touched (Fig 2). CHEST / 131 / 5 / MAY, 2007
1527
Figure 2. Registration required touching each of the markers with the probe and then simultaneously indicating the structure to the computer.
Localization After the registration was completed, the surgical plan was developed using the StealthStation system. The entry point, target point, and border of the lesion were defined. CT images were used to guide a 16-gauge needle attached by a positioning sensor (Fig 1) to the appropriate areas for injection of the
methylene blue. As the position sensor-equipped needle was punctured into the lung, crosshairs indicated its distal tip position and moved on the CT display (Fig 3). Coronal, sagittal, axial, and three-dimensional images were displayed, along with a reconstructed image parallel to the orientation of the tip. When the needle was advanced into or immediately adjacent to the lung nodule, 1 to 2 mL of methylene blue was injected by the surgeon.
Figure 3. The computer display of the navigation system provided trajectory views. Crosshairs indicated the real-time position of the tip of the needle on the two-dimensional views (the tip of the needle had reached the area surrounding the nodule). 1528
Original Research
Both the registration and the localization were performed during DIBH, which was simulated by inflating the lung with artificial ventilation under general anesthesia. The volume of the inhalation of air was equal to that in the CT scans. Thoracoscopic Biopsy The patient subsequently was positioned for thoracoscopic biopsy with single-lung ventilation of the contralateral lung. We introduced the first trocar for the videothoracoscope, usually in the sixth or seventh intercostal space along the midaxillary line. The methylene blue in the lung surface allowed rapid identification of the approximate nodule location. Another two ports were placed anteriorly and posteriorly depending on the location of the nodule in the lung parenchyma. The lung surface exhibiting methylene blue was grasped by the endoscopic forceps, and was followed by digital or instrumental palpation to further confirm the exact localization of the nodule. Then excisional biopsy was performed by an articulating endostapler. The depth of the wedge—the inclination of the stapler—was estimated based on the depth of the nodule on CT, in order to include a safety margin of between 5 mm and 10 mm. The resected nodule was extracted in a plastic bag to avoid neoplastic cell dissemination at the port site. Before sent for frozen-section pathologic examination, biopsy specimens were split open in the operating room to ensure the presence of the entire nodule. When the frozen-section pathologic diagnosis was a benign nodule or metastatic disease, a chest tube was inserted and the operation was finished. When the finding was primitive lung cancer, a major pulmonary resections was performed according to oncology standards.
Results The average time of registration was 4.8 ⫾ 0.9 min (⫾ SD). Registration error was on average 2.7 ⫾ 0.2 mm. The time from the localization to the initial surgical skin incision was 20.8 ⫾ 3.9 min. The size of the SPN at the CT scan was 14.8 ⫾ 1.1 mm (range, 8 to 20 mm). The distance from the pleural surface to the SPN was 16.7 ⫾ 1.1 mm (range, 10 to 25 mm). Lesions were found in the right upper lobe (n ⫽ 6), right lower lobe (n ⫽ 2), left upper lobe (n ⫽ 6), and left lower lobe (n ⫽ 3) at the time of operation (Table 1). All the lesions were successfully localized in the thoracoscope image, and the methylene blue-stained area was found overlying the nodule in all cases during careful macroscopic examination of the resected specimens. No spillage of dye occurred during the procedure that could potentially stain a larger amount of the pleural surface, thus making thoracoscopic visualization more difficult. Five lesions were barely palpable as a slight solid mass by endoscopic devices during surgery, but the remaining 12 lesions could not be detected with instrument. All resection margins were microscopically clear confirmed by the pathologic examination. There were no surgical complications. The chest tubes were removed within 2 to www.chestjournal.org
4 days postoperatively. All patients were discharged on postoperative day 6 on average (range, 3 to 11 days). Histologically, 7 nodules were malignant and 10 were benign. Benign diagnoses included acid-fast bacillus (n ⫽ 1), granuloma (n ⫽ 2), hamartoma (n ⫽ 4), aspergilloma (n ⫽ 1), and inflammatory pseudotumor (n ⫽ 2). Malignant diagnoses included primary adenocarcinoma (n ⫽ 3), squamous cell carcinoma (n ⫽ 2), and pulmonary metastasis (n ⫽ 2).
Discussion A major problem of conducting a successful thoracoscopic resection is to locate small or deeply situated target nodules.4 A few localization techniques have been developed, either preoperative or intrathoracoscopic.4,5 Although all of the methods are effective in locating the target lesion, their limitations are obvious. The needle-wire technique is associated with several complications, such as lung hemorrhage, pleuritic pain, and pneumothorax. The hookwire dislodgement rate of 7.5% was reported due to patient’s breath or the deflated lung during single-lung ventilation.5 A failure rate of approximately 13% for preoperative methylene blue injection under the CT fluoroscopic guidance has been reported due to either an excess of liquid injection or an error in nodule localization.6 It is found that the dye frequently dissipates over a large area by the time the surgical procedure is done, making its localization features inadequate. The localization of pulmonary nodules by radio-guided technique also reveals some drawbacks. One problem is the notable and fast diffusion of contrast medium in the pulmonary parenchyma surrounding the nodule, due to the rich vascularization of the lung. A second problem is locating deep and posterior nodules due to the dimension and structure of the probe, which cannot move freely in the thorax.7 Moreover, using special radiation protection precautions in the operating room or in handling the surgical specimens is be required. However, all the techniques are performed under the CT fluoroscopic guidance. As a result, they require significant CT scanner time and use additional radiation. Radiation exposure for both patients and clinician remains a major concern. The intrathoracoscopic ultrasound technique also has some limitations. The chief difficulty was in obtaining an image as long as any air remained in the lung. Recent advances in IGS have been brought about by the development of computer technology to make minimally invasive techniques safer and more accurate. The primary advantage of IGS is that it provides the surgeon with accurate knowledge of the spatial CHEST / 131 / 5 / MAY, 2007
1529
relationship between the surgical instrument or appliance and the surrounding structures not directly visible in the operative field.8 The image-guided navigation technique is popular in many surgical fields, such as neurosurgery and orthopedics. The navigation technique also has been proved safe and effective in the three-dimensional CT-guided bronchoscopy with a real-time electromagnetic positioning sensor.9 In this study, we demonstrated the feasibility of performing thoracoscopic biopsy using image-guided navigation. The results of our study suggest that the technique provides precise localization. The imageguided procedure does not require CT fluoroscopic guidance. One major limitation of CT fluoroscopic guidance is the relatively high radiation exposure to patient and surgeon.10 Consequently, there was no risk of high radiation exposure to patient and surgeon in the image-guided procedure, compared with the CT fluoroscopy-guided localization techniques. The complications of hemorrhaging and pneumothorax may occur during the image-guided procedure, but the surgeon does not has to worry about complications as in other localization procedure. For most of preoperative CT-guided percutaneous localization techniques, the patients had to wait for a long time before undergoing thoracoscopy after the localization procedure, owing to surgical suite or CT room scheduling issues. During the waiting time, the surgeon usually worried about the lung hemorrhage and pneumothorax becoming more serious and serious, which was very dangerous to the patient with poor lung function. However, our image-guided procedure was performed in surgical suite after the patient had been receiving general anesthesia and mechanical ventilation. The patient was under the control of the surgeon and the anesthetist. Moreover, the time from the localization to the initial surgical skin incision was not more than 30 min. As a result, the surgeon’s worries about the hemorrhaging and pneumothorax lessened to a great extent. Most of the localization techniques use preoperative CTguided percutaneous punctures with local anesthesia. Patients who were kept fixed during the procedure suffered from mental strain and anxiety for the pain and surgical complication. They needed sedatives. However, our image-guided localization is performed under general anesthesia. The patient has no sense of fear associated with surgery, while the surgeon does not have to worry about coughing or the body movements of the patient, which will handicap the biopsy procedure. The technology is not complicated. The accompanying equipment for the StealthStation navigation system does not occupy a large space in the operating room. Furthermore, this navigation system—the 1530
surgical global positioning satellite—provides highly accurate real-time information to the surgeon regarding localization, trajectory, and the depth of the instruments end, and therefore is promising in the clinic, such as for localization and biopsy of mediastinal lymph node metastasis. The biggest limitation of our method remained that the movement of the diaphragm and the chest wall during respiration affected the outcome of the localization. We used the DIBH technique to increase the accuracy of the localization. Mah et al3 reported that the DIBH technique could significantly reduce the breathing-induced tumor motion. Moreover, the patient’s volume of inhalation of air during the localization was equal to that of in the CT scans. As a result, consistent lung inflation levels were achieved in both the CT scans and imageguided localization, and the reproducibility of tumor position was satisfactorily accurate. In other word, the respiratory-induced target motion was minimized. In addition, many reports11,12 have shown a variety of methods to improve the accuracy of treatment for moving tumors. Giraud et al12 reported that a commercial gating system using external markers for respiration synchronized CT scans significantly improved the positional reproducibility of thoracic and upper abdominal structures. Accurate intraoperative real-time navigation also is a safe and practicable solution for minimizing respiratory-induced target motion.13 In conclusion, our research has demonstrated that image-guided marking of peripheral lung lesions is a feasible and promising technique that provides appropriate guidance and proves effective in immediately facilitating subsequent thoracoscopic resection. ACKNOWLEDGMENT: We thank Thomas Lang, PhD (Research Scientist) and Frank Lindseth (Research Scientist) from SINTEF Health Research, Trondheim, Norway, for their help. We also thank Xiong Shengchun, MD, Zhuang Congwen, MD, Lin Jinxiang, MD, Xu Chi, MD, and Liu Daoming, MD, from the Department of Thoracic and Cardiovascular Surgery in the Fuzhou General Hospital for their support.
References 1 Saito H, Minamiya Y, Matsuzaki I, et al. Indication for preoperative localization of small peripheral pulmonary nodules in thoracoscopic surgery. J Thorac Cardiovasc Surg 2002; 124:1198 –1202 2 Sortini D, Feo CV, Carcoforo P, et al. Thoracoscopic localization techniques for patients with solitary pulmonary nodule and history of malignancy. Ann Thorac Surg 2005; 79:258 – 262 3 Mah D, Hanley J, Rosenzweig KE, et al. Technical aspects of the deep inspiration breath-hold technique in the treatment of thoracic cancer. Int J Radiat Oncol Biol Phys 2000; 48:1175–1185 4 Partik BL, Leung AN, Muller MR, et al. Using a dedicated Original Research
5
6 7 8 9
lung-marker system for localization of pulmonary nodules before thoracoscopic surgery. AJR Am J Roentgenol 2003; 180:805– 809 Ciriaco P, Negri G, Puglisi A, et al. Video-assisted thoracoscopic surgery for pulmonary nodules: rationale for preoperative computed tomography-guided hookwire localization. Eur J Cardiothorac Surg 2004; 25:429 – 433 Kerrigan DC, Spence PA, Crittenden MD, et al. Methylene blue guidance for simplified resection of a lung lesion. Ann Thorac Surg 1992; 53:163–164 Carcoforo P, Feo C, Sortini D, et al. Localization of pulmonary nodules. Chest 2004; 125:796 –797 Mayberg MR, LaPresto E, Cunningham EJ. Image-guided endoscopy: description of technique and potential applications. Neurosurg Focus 2005; 19:E10 Hautmann H, Schneider A, Pinkau T, et al. Electromagnetic catheter navigation during bronchoscopy: validation of a novel
www.chestjournal.org
10
11
12 13
method by conventional fluoroscopy. Chest 2005; 128:382– 387 Solomon SB, Patriciu A, Bohlman ME, et al. Robotically driven interventions: a method of using CT fluoroscopy without radiation exposure to the physician. Radiology 2002; 225:277–282 Onishi H, Kuriyama K, Komiyama T, et al. CT evaluation of patient deep inspiration self-breath-holding: how precisely can patients reproduce the tumor position in the absence of respiratory monitoring devices? Med Phys 2003; 30:1183– 1187 Giraud P, Yorke E, Ford EC, et al. Reduction of organ motion in lung tumors with respiratory gating. Lung Cancer 2006; 51:41–51 Reittner P, Tillich M, Luxenberger W, et al. Multislice CT-image-guided endoscopic sinus surgery using an electromagnetic tracking system. Eur Radiol 2002; 12:592–596
CHEST / 131 / 5 / MAY, 2007
1531