Original Article
Thoracoscopic wedge lung resection using virtual-assisted lung mapping
Asian Cardiovascular & Thoracic Annals 2015, Vol. 23(1) 46–54 ß The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0218492314539332 aan.sagepub.com
Masaaki Sato, Akihiro Aoyama, Tetsu Yamada, Toshi Menjyu, Fengshi Chen, Toshihiko Sato, Makoto Sonobe, Mitsugu Omasa and Hiroshi Date
Abstract Background: Virtual-assisted lung mapping is a novel bronchoscopic preoperative lung marking technique using virtual images to conduct multiple concurrent lung markings with dye. This study analyzed the indications, mapping design, and outcomes of lung wedge resection using virtual-assisted lung mapping. Methods: From August 2012 to October 2013, 35 patients with 59 lesions were planned to undergo thoracoscopic lung wedge resection aided by virtual-assisted lung mapping. The data related to virtual-assisted lung mapping were prospectively collected, with the exception of the mapping design which was retrospectively analyzed. Results: Suspected primary lung cancer (21 lesions in 18 patients) and metastatic lung tumors (38 lesions in 17 patients) were treated by thoracoscopic lung wedge resection with the aid of virtual-assisted lung mapping; 50 wedge resections were conducted with 107 markings. Virtual-assisted lung mapping was most frequently designed to place 2 (n ¼ 15 wedge resections) or 3 (n ¼ 17) markings to both identify the tumor(s) and secure a sufficient resection margin. In 7 wedge resections, anatomical landmarks and/or imaginary auxiliary lines functioned as complementary parts of the lung map when bronchial anatomy did not allow for markings at ideal spots. The resection outcomes were satisfactory without clinically evident complications. Conclusion: Multiple markings of virtual-assisted lung mapping not only enabled tumor identification, but also secured sufficient resection margins. Special techniques using anatomical landmarks and imaginary auxiliary lines were complementary to the lung map when bronchial anatomy did not allow for markings at ideal spots.
Keywords Bronchoscopy, Lung neoplasms, Fluoroscopy, Staining and labeling, Thoracic surgery, Video-assisted thoracic surgery (VATS)
Introduction Lung marking is an important technique for precise wedge resection of hardly palpable small lung cancers and metastatic tumors.1 Virtual-assisted lung mapping (VAL-MAP) is a novel bronchoscopic preoperative lung marking technique that uses virtual images to conduct multiple concurrent markings with dye injection on the lung surface.2 The multiple lung markings (i.e., lung mapping) of VAL-MAP provide geometric information on the lung surface to not only identify tumors, but also to obtain sufficient resection margins in thoracoscopic wedge resection. This new technique appears to have multiple advantages over other marking strategies that have been reported previously. Bronchoscopic
marking techniques are considered more advantageous than computed tomography (CT)-guided percutaneous marking techniques in that they allow access to areas that are not reachable with percutaneous approaches.2–4 Such areas include the lung apex, the area behind the scapula, and even the interlobar fissure.
Department of Thoracic Surgery, Kyoto University Hospital, Kyoto, Japan Corresponding author: Masaaki Sato, MD, PhD, Department of Thoracic Surgery, Kyoto University Hospital, 54 Kawahara-cho, Shyogoin, Sakyo-ku, Kyoto 606-8507, Japan. Email: [email protected]
Downloaded from aan.sagepub.com at UNIV NEBRASKA LIBRARIES on August 24, 2015
Sato et al.
47
Moreover, bronchoscopic marking techniques are associated with a lower risk of complications such as bleeding and pneumothorax, even if they are not completely free of such complications.2 Most importantly, potentially fatal air embolisms, which have been reported in percutaneous marking techniques, are less likely to occur with bronchoscopic approaches.5–10 Moreover, dye markings are generally associated with a lower risk of marker dislodgement and loss than are wire or coil hook methods.1,11 In the VAL-MAP technique described herein, we utilized virtual bronchoscopy and 3-dimensional images constructed from thin-slice CT to avoid the previously reported use of real-time CT for bronchoscopic markings and thus avoid excessive radiation exposure to the bronchoscopist. The bronchoscopist performing VAL-MAP is exposed to regular fluoroscopy radiation for only 20 to 45 seconds per mark, and the total radiation time is limited to 1 to 2 minutes per patient.2 Our initial report on VAL-MAP described the bronchoscopic marking technique and only preliminary clinical data; however, the detailed strategies for lung mapping, including the mapping design and surgical techniques specialized for VAL-MAP, have not been described. The purpose of the present study was to retrospectively review cases in which VAL-MAP was used, particularly focusing on lung wedge resection, and describe the details of the mapping design and surgical techniques for lung wedge resection using VAL-MAP.
Patients and methods Patients who underwent lung wedge resection using VAL-MAP from August 2012 to October 2013 were included in the study. The general selection criteria for VAL-MAP (either wedge resection or segmentectomy) are: a pulmonary lesion expected to be hardly palpable during thoracoscopic surgery, a pulmonary lesion with a resection margin that must be carefully determined because of the tumor location and/or extension of the ground-glass opacity component, and expected pleural adhesion due for example to a redo operation. The selection criteria for wedge resection are: peripherally located (approximately the outer one-third of the lung) pure ground-glass opacity lesions 1.5 cm, and the actual resection was conducted accordingly. The prospectively collected data related to VALMAP included the patient characteristics, characteristics of target lesions, clinical diagnosis, operative plan, number of planned markings, number of markings visible and usable during the operation, complications related to VAL-MAP, final operation, and final pathology. In addition, the mapping design for lung wedge resection was retrospectively reviewed for each case.
Downloaded from aan.sagepub.com at UNIV NEBRASKA LIBRARIES on August 24, 2015
48
Asian Cardiovascular & Thoracic Annals 23(1)
Figure 1. Tumor located close to the pleural surface. (a) A 49-year-old woman had a lung tumor 4 cm in diameter in the right upper lobe and a 7-mm ground-glass opacity lesion in the right middle lobe (arrow). Thoracoscopic right upper lobe resection and concurrent middle lobe wedge resection using virtual-assisted lung mapping was planned. (b) After bronchoscopic marking under local anesthesia, a computed tomography scan was performed. The markings (injected indigo carmine, 1 mL per mark) were identified as areas of ground-glass opacity (arrows). (c) Three-dimensional image constructed from post-virtual-assisted lung mapping computed tomography. The localization of the tumor (red) and 2 markings (blue) are shown. The interrupted line indicates the imaginary resection line. Note that the 2 markings are placed somewhat away from the tumor to establish an adequate resection margin. (d) Intraoperative view showing the 2 markings. The arrow indicates the expected location of the tumor. (e) The lung tissue expected to include the tumor is lifted using ring forceps, and stapler resected according to the operative plan. The right middle lobe lesion was adenocarcinoma in situ in permanent pathology.
The Student’s t test was used to compare two groups. Data are expressed as mean standard deviation when appropriate.
Results In total, 112 lesions in 76 patients underwent resection using VAL-MAP from August 2012 to October 2013. Among these, 35 patients with 59 lesions were planned to have 50 wedge resections using 107 VAL-MAP markings; segmentectomy was planned for the other lesions. The indications for wedge resection using VAL-MAP were suspected primary lung cancer (21 lesions in 18 patients) and metastatic lung tumors (38 lesions in 17 patients). The origins of suspected
metastatic lung tumors included sarcoma (n ¼ 6), colorectal cancer (n ¼ 5), lung cancer (n ¼ 3), thyroid cancer (n ¼ 1), hepatocellular carcinoma (n ¼ 1), and esophageal cancer (n ¼ 1). The characteristics of the patients and target lesions are shown in Table 1. Patients who were suspected to have metastatic tumors had a larger number of target lesions per patient than did those who were suspected to have primary lung cancer. The size of the target lesion was significantly smaller in patients with suspected metastatic tumors than in those with suspected primary lung cancer. Most suspected primary lung cancer lesions showed ground-glass opacities, with the exception of one solid lesion; VAL-MAP was applied in this case because of expected adhesion due to the second operation. In contrast, most of the
Downloaded from aan.sagepub.com at UNIV NEBRASKA LIBRARIES on August 24, 2015
Sato et al.
49
Figure 2. Tumor requiring a deep margin. (a) A 72-year-old woman had a 2-cm lesion in the right lower lobe and a 5-mm groundglass opacity lesion in the right upper lobe (arrow). Thoracoscopic right lower lobectomy and concurrent upper lobe wedge resection using virtual-assisted lung mapping was planned. Note that the 5-mm lesion is localized approximately 1 cm from the lung surface; to obtain an adequate deep margin (1 cm from the tumor), the depth of the lung tissue to be resected was at least 2.5 cm from the surface. (b) Three-dimensional image constructed from post-virtual-assisted lung mapping computed tomography. The location of the middle lobe tumor and 3 markings is shown. The interrupted line indicates the imaginary resection line. Note that the 2 caudal markings provide information regarding the lateral resection lines and thus the ideal depth of lung tissue to be taken. The inset shows a 3-dimensional slab image (thick section) including the 2 caudal markings and the tumor. (c) Intraoperative view showing 3 markings of virtual-assisted lung mapping. The arrow indicates the expected location of the tumor. The 2 caudal markings appear target-shaped, presumably because of the distance of the catheter tip from the lung surface at the time of dye injection. The center of a marking is considered to correspond to the marking location in the 3-dimensional image. (d) The lung tissue expected to include the tumor is lifted using ring forceps, and stapler resected according to the operative plan, paying attention to the lateral and thus the deep margin. The right upper lobe lesion was adenocarcinoma in situ in permanent pathology.
suspected metastatic tumors were solid nodules, while others were cavitary lesions. In the latter cases, preoperative chemotherapy was effective, and resection of the residual lesions was requested. VAL-MAP was designed according to the principle described in the methods section. In 13 wedge resections, 2 markings were placed on both sides of the lesion (Figure 1). In 2 wedge resections, 2 markings were placed to resect 2 lesions together in one specimen. In 14 wedge resections, 3 markings were used to resect the lesion (Figure 2). In the remaining 2 cases, 3 markings were
used to resect 2 lesions in a single wedge resection. Deviating from the principle of the VAL-MAP design, only one marking was used to resect a single lesion in 3 wedge resections. In these cases, dye markings were placed close to the tumor, similar to conventional marking techniques. In 7 wedge resections of 7 lesions, the bronchial anatomy allowed for neither placement of 2 markings at both ends of the tumor nor placement of 3 markings to surround the tumor. To achieve successful resection in such cases, special techniques were developed using an anatomical
Downloaded from aan.sagepub.com at UNIV NEBRASKA LIBRARIES on August 24, 2015
50
Asian Cardiovascular & Thoracic Annals 23(1) Table 1. Characteristics of patients and target lesions. Preoperative diagnosis Age (years) Sex (male/female) Total no. of target lesions Target lesions per patient [range] Lesion size (mm) Lesion characteristics on computed tomography Pure ground-glass opacity Nodule Cavity Planned concurrent operation Lobectomy Segmentectomy
Suspected primary lung cancer (n ¼ 18)
Suspected metastatic lung tumor (n ¼ 17)
p value
64.4 3.2 9:9 21 1.16 0.31 [1–2] 9.6 0.8
59.8 3.2 11:6 37 2.24 0.32 [1–7] 4.2 0.6
Thoracoscopic wedge lung resection using virtual-assisted lung mapping
Asian Cardiovascular & Thoracic Annals 2015, Vol. 23(1) 46–54 ß The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0218492314539332 aan.sagepub.com
Masaaki Sato, Akihiro Aoyama, Tetsu Yamada, Toshi Menjyu, Fengshi Chen, Toshihiko Sato, Makoto Sonobe, Mitsugu Omasa and Hiroshi Date
Abstract Background: Virtual-assisted lung mapping is a novel bronchoscopic preoperative lung marking technique using virtual images to conduct multiple concurrent lung markings with dye. This study analyzed the indications, mapping design, and outcomes of lung wedge resection using virtual-assisted lung mapping. Methods: From August 2012 to October 2013, 35 patients with 59 lesions were planned to undergo thoracoscopic lung wedge resection aided by virtual-assisted lung mapping. The data related to virtual-assisted lung mapping were prospectively collected, with the exception of the mapping design which was retrospectively analyzed. Results: Suspected primary lung cancer (21 lesions in 18 patients) and metastatic lung tumors (38 lesions in 17 patients) were treated by thoracoscopic lung wedge resection with the aid of virtual-assisted lung mapping; 50 wedge resections were conducted with 107 markings. Virtual-assisted lung mapping was most frequently designed to place 2 (n ¼ 15 wedge resections) or 3 (n ¼ 17) markings to both identify the tumor(s) and secure a sufficient resection margin. In 7 wedge resections, anatomical landmarks and/or imaginary auxiliary lines functioned as complementary parts of the lung map when bronchial anatomy did not allow for markings at ideal spots. The resection outcomes were satisfactory without clinically evident complications. Conclusion: Multiple markings of virtual-assisted lung mapping not only enabled tumor identification, but also secured sufficient resection margins. Special techniques using anatomical landmarks and imaginary auxiliary lines were complementary to the lung map when bronchial anatomy did not allow for markings at ideal spots.
Keywords Bronchoscopy, Lung neoplasms, Fluoroscopy, Staining and labeling, Thoracic surgery, Video-assisted thoracic surgery (VATS)
Introduction Lung marking is an important technique for precise wedge resection of hardly palpable small lung cancers and metastatic tumors.1 Virtual-assisted lung mapping (VAL-MAP) is a novel bronchoscopic preoperative lung marking technique that uses virtual images to conduct multiple concurrent markings with dye injection on the lung surface.2 The multiple lung markings (i.e., lung mapping) of VAL-MAP provide geometric information on the lung surface to not only identify tumors, but also to obtain sufficient resection margins in thoracoscopic wedge resection. This new technique appears to have multiple advantages over other marking strategies that have been reported previously. Bronchoscopic
marking techniques are considered more advantageous than computed tomography (CT)-guided percutaneous marking techniques in that they allow access to areas that are not reachable with percutaneous approaches.2–4 Such areas include the lung apex, the area behind the scapula, and even the interlobar fissure.
Department of Thoracic Surgery, Kyoto University Hospital, Kyoto, Japan Corresponding author: Masaaki Sato, MD, PhD, Department of Thoracic Surgery, Kyoto University Hospital, 54 Kawahara-cho, Shyogoin, Sakyo-ku, Kyoto 606-8507, Japan. Email: [email protected]
Downloaded from aan.sagepub.com at UNIV NEBRASKA LIBRARIES on August 24, 2015
Sato et al.
47
Moreover, bronchoscopic marking techniques are associated with a lower risk of complications such as bleeding and pneumothorax, even if they are not completely free of such complications.2 Most importantly, potentially fatal air embolisms, which have been reported in percutaneous marking techniques, are less likely to occur with bronchoscopic approaches.5–10 Moreover, dye markings are generally associated with a lower risk of marker dislodgement and loss than are wire or coil hook methods.1,11 In the VAL-MAP technique described herein, we utilized virtual bronchoscopy and 3-dimensional images constructed from thin-slice CT to avoid the previously reported use of real-time CT for bronchoscopic markings and thus avoid excessive radiation exposure to the bronchoscopist. The bronchoscopist performing VAL-MAP is exposed to regular fluoroscopy radiation for only 20 to 45 seconds per mark, and the total radiation time is limited to 1 to 2 minutes per patient.2 Our initial report on VAL-MAP described the bronchoscopic marking technique and only preliminary clinical data; however, the detailed strategies for lung mapping, including the mapping design and surgical techniques specialized for VAL-MAP, have not been described. The purpose of the present study was to retrospectively review cases in which VAL-MAP was used, particularly focusing on lung wedge resection, and describe the details of the mapping design and surgical techniques for lung wedge resection using VAL-MAP.
Patients and methods Patients who underwent lung wedge resection using VAL-MAP from August 2012 to October 2013 were included in the study. The general selection criteria for VAL-MAP (either wedge resection or segmentectomy) are: a pulmonary lesion expected to be hardly palpable during thoracoscopic surgery, a pulmonary lesion with a resection margin that must be carefully determined because of the tumor location and/or extension of the ground-glass opacity component, and expected pleural adhesion due for example to a redo operation. The selection criteria for wedge resection are: peripherally located (approximately the outer one-third of the lung) pure ground-glass opacity lesions 1.5 cm, and the actual resection was conducted accordingly. The prospectively collected data related to VALMAP included the patient characteristics, characteristics of target lesions, clinical diagnosis, operative plan, number of planned markings, number of markings visible and usable during the operation, complications related to VAL-MAP, final operation, and final pathology. In addition, the mapping design for lung wedge resection was retrospectively reviewed for each case.
Downloaded from aan.sagepub.com at UNIV NEBRASKA LIBRARIES on August 24, 2015
48
Asian Cardiovascular & Thoracic Annals 23(1)
Figure 1. Tumor located close to the pleural surface. (a) A 49-year-old woman had a lung tumor 4 cm in diameter in the right upper lobe and a 7-mm ground-glass opacity lesion in the right middle lobe (arrow). Thoracoscopic right upper lobe resection and concurrent middle lobe wedge resection using virtual-assisted lung mapping was planned. (b) After bronchoscopic marking under local anesthesia, a computed tomography scan was performed. The markings (injected indigo carmine, 1 mL per mark) were identified as areas of ground-glass opacity (arrows). (c) Three-dimensional image constructed from post-virtual-assisted lung mapping computed tomography. The localization of the tumor (red) and 2 markings (blue) are shown. The interrupted line indicates the imaginary resection line. Note that the 2 markings are placed somewhat away from the tumor to establish an adequate resection margin. (d) Intraoperative view showing the 2 markings. The arrow indicates the expected location of the tumor. (e) The lung tissue expected to include the tumor is lifted using ring forceps, and stapler resected according to the operative plan. The right middle lobe lesion was adenocarcinoma in situ in permanent pathology.
The Student’s t test was used to compare two groups. Data are expressed as mean standard deviation when appropriate.
Results In total, 112 lesions in 76 patients underwent resection using VAL-MAP from August 2012 to October 2013. Among these, 35 patients with 59 lesions were planned to have 50 wedge resections using 107 VAL-MAP markings; segmentectomy was planned for the other lesions. The indications for wedge resection using VAL-MAP were suspected primary lung cancer (21 lesions in 18 patients) and metastatic lung tumors (38 lesions in 17 patients). The origins of suspected
metastatic lung tumors included sarcoma (n ¼ 6), colorectal cancer (n ¼ 5), lung cancer (n ¼ 3), thyroid cancer (n ¼ 1), hepatocellular carcinoma (n ¼ 1), and esophageal cancer (n ¼ 1). The characteristics of the patients and target lesions are shown in Table 1. Patients who were suspected to have metastatic tumors had a larger number of target lesions per patient than did those who were suspected to have primary lung cancer. The size of the target lesion was significantly smaller in patients with suspected metastatic tumors than in those with suspected primary lung cancer. Most suspected primary lung cancer lesions showed ground-glass opacities, with the exception of one solid lesion; VAL-MAP was applied in this case because of expected adhesion due to the second operation. In contrast, most of the
Downloaded from aan.sagepub.com at UNIV NEBRASKA LIBRARIES on August 24, 2015
Sato et al.
49
Figure 2. Tumor requiring a deep margin. (a) A 72-year-old woman had a 2-cm lesion in the right lower lobe and a 5-mm groundglass opacity lesion in the right upper lobe (arrow). Thoracoscopic right lower lobectomy and concurrent upper lobe wedge resection using virtual-assisted lung mapping was planned. Note that the 5-mm lesion is localized approximately 1 cm from the lung surface; to obtain an adequate deep margin (1 cm from the tumor), the depth of the lung tissue to be resected was at least 2.5 cm from the surface. (b) Three-dimensional image constructed from post-virtual-assisted lung mapping computed tomography. The location of the middle lobe tumor and 3 markings is shown. The interrupted line indicates the imaginary resection line. Note that the 2 caudal markings provide information regarding the lateral resection lines and thus the ideal depth of lung tissue to be taken. The inset shows a 3-dimensional slab image (thick section) including the 2 caudal markings and the tumor. (c) Intraoperative view showing 3 markings of virtual-assisted lung mapping. The arrow indicates the expected location of the tumor. The 2 caudal markings appear target-shaped, presumably because of the distance of the catheter tip from the lung surface at the time of dye injection. The center of a marking is considered to correspond to the marking location in the 3-dimensional image. (d) The lung tissue expected to include the tumor is lifted using ring forceps, and stapler resected according to the operative plan, paying attention to the lateral and thus the deep margin. The right upper lobe lesion was adenocarcinoma in situ in permanent pathology.
suspected metastatic tumors were solid nodules, while others were cavitary lesions. In the latter cases, preoperative chemotherapy was effective, and resection of the residual lesions was requested. VAL-MAP was designed according to the principle described in the methods section. In 13 wedge resections, 2 markings were placed on both sides of the lesion (Figure 1). In 2 wedge resections, 2 markings were placed to resect 2 lesions together in one specimen. In 14 wedge resections, 3 markings were used to resect the lesion (Figure 2). In the remaining 2 cases, 3 markings were
used to resect 2 lesions in a single wedge resection. Deviating from the principle of the VAL-MAP design, only one marking was used to resect a single lesion in 3 wedge resections. In these cases, dye markings were placed close to the tumor, similar to conventional marking techniques. In 7 wedge resections of 7 lesions, the bronchial anatomy allowed for neither placement of 2 markings at both ends of the tumor nor placement of 3 markings to surround the tumor. To achieve successful resection in such cases, special techniques were developed using an anatomical
Downloaded from aan.sagepub.com at UNIV NEBRASKA LIBRARIES on August 24, 2015
50
Asian Cardiovascular & Thoracic Annals 23(1) Table 1. Characteristics of patients and target lesions. Preoperative diagnosis Age (years) Sex (male/female) Total no. of target lesions Target lesions per patient [range] Lesion size (mm) Lesion characteristics on computed tomography Pure ground-glass opacity Nodule Cavity Planned concurrent operation Lobectomy Segmentectomy
Suspected primary lung cancer (n ¼ 18)
Suspected metastatic lung tumor (n ¼ 17)
p value
64.4 3.2 9:9 21 1.16 0.31 [1–2] 9.6 0.8
59.8 3.2 11:6 37 2.24 0.32 [1–7] 4.2 0.6
