Thoracoscopic Resection of Pulmonary Metastases· Robert D. Dowling, M. D.; Peter F. Ferson, M. D.; and Rodney J. Landreneau, M.D.
Objective: To describe the use of thoracoscopic techniques to achieve parenchymal sparing wedge resection of peripheral lung lesions in patients with a history of malignancy, and to describe the morbidity, mortality, and hospital course associated with this approach. Deaign: Case series. Setting: UDiversity hospital. ParticipGnta: PatieDts with a history of malignancy and lesions on computerized tomography in the outer ODe third of the lung parenchyma. outcome meaaurementB: Histologic analysis of resected IUDI lesions, operative 6ndings, operative time, duration of chest tube drainage and hospital stay, operative morbidity, and mortality. Baulta: Twenty-one thoracoscopic resections ofpulmonary parenchymal lesions were performed OD 15 patieDts. All peripheral lesions identified by computerized tomography were fouod at thoracoscopy and successfully resected with the Nd:YAG laser (n = 7), an endoscopic stapler (D = 10), or
M'"
pulmonary lesions in patients with a history of malignancy raise serious suspicions for the presence of metastatic disease. A tissue diagnosis is frequently required as the presence of metastatic disease will often dictate further therapy. Pulmonary metastasectomy may also favorably influence survival in patients with certain tumor histologies. 1-6 Currently, thoracotomy or median sternotomy are standard surgical approaches to pulmonary metastasectomy. Early operative morbidity has varied between 5 and 14 percent in recent series with these techniques.P" Beyond the acute perioperative difficulties seen with open approaches, chronic pain syndromes are common after thoracotomy 8.9 Recent success with open, laser-assisted pulmonary resections and advances in endoscopic surgical instrumentation have allowed for thoracoscopic management of many intrathoracic conditions.P''! These achievements encouraged us to employ thoracoscopic surgical techniques in the diagnosis and management of peripherally located lesions suspected to be metastatic disease. We report the first series of thoracoscopic resection of pulmonary metastases.
·From the Department of Cardiothoracic Surgery, Section of Thoracic Surgery, University of Pittsburgh, Pittsburgh. Manuscript received February 27; revision accepted April 29.
1410
both (n=4). The mean diameter of the lesions was 0.8 cm (range 0.2 to 1.5 cm). Histologic analysis revealed metastatic disease in 13 patients and benign disease in 2 patients. All resection margins were free of tumor. The mean duration of chest tube drainage and postoperative hospital stay were 1.8:1:0.1 and 3.3:1:0.1 days, respectivel~ Mean operative time was III min (range 45 to 155 min). One patient who underwent a right thoracoscopic resection developed a transient left vocal cord paresis. There were DO other complications and DO deaths. ConcluBion: Thoracoscopy was successful in identifying peripheral lung lesions and allowed Cor parenchymal sparing resection identical in extent to that performed with open approaches. For select patients with peripheral lung nodules felt to be metastases, thoracoscopic resection may result in reduced morbidity, cost, hospital stay and allow for more rapid institution of therapy £or the primary disease. (Chat 1992; 102:14S0-U)
METHODS Ibflents
Patients with a history of malignancy and suspicious lesions found on chest roentgenogram had computerized tomography of the chest and appropriate workup for the presence of extrathoracic metastases. Patients with lesions on computerized tomography of the chest located in the outer one third of the pulmonary parenchyma were considered candidates for thoracoscopic resection (Fig 1). In all cases, the pulmonary nodules were the only sites suspicious for metastatic diseases or were the most accessible lesions for obtaining a tissue diagnosis. Patients with a favorable tumor histologic condition and limited tumor burden underwent thoracoscopic resection of all identifiable disease. Other patients underwent thoracoscopic resection if the lesions were not amenable to percutaneous biopsy or if previous attempts at percutaneous biopsy had been unsuccessful. Operative Procedure
After induction of general anesthesia, 8exible bronchoscopy was performed through a single lumen endotracheal tube to evaluate the presence of endobronchial lesions that would provide tissue for diagnosis. In the absence of endobronchial lesions, exploratory thoracoscopy was performed." Patients were intubated with a left:sided double lumen endotracheal tube to allow for contralateral single lung ventilation and collapse of the ipsilateral lung which was essential to perform thoracoscopic surgery. After institution of single lung ventilation, an II-mm diameter trocar (Autosuture Surgiport Trocars, United States Surgical Corp) was introduced into the thoracic cavity. The site chosen for the introduction of the initial trocar was dictated by the location of the lesion. A wide angle, zero degree thoracoscope (operating wide angle zero degree Hopkin's telescope, 2604OA, Karl Storz Endoscopic America, Inc) with an operating port was introduced through this trocar, and Thoracoscopic Resection of PulmonaryMetastases (Dowling, Ferson, Lsndreneau)
exploration of the thoracic cavity was performed . Any adhesions present were divided using sharp dissection with the endoscopic scissors or blunt dissection with the endoscopic probe . Two additional trocars were introduced to allow manipulation and examination of the entire lung. The lung was examined both visually and by palpation with a blunt endoscopic probe . Collapse of the lung facilitated identification of lesions that became effaced against the atelectatic pulmonary parenchyma . Slight enlargement of one of the trocar sites to allow introduction of a palpating finger was also used to identify lesions that were not found with the other techn iques. Lesions on the Rat surface of the lung were resected with a noncontact neodymium:yttrium-aluminum garnet (Nd:YAC) laser (Laserscope Inc, model 704) that was calibrated and set on a 35 \v, continuous setting . For smoke evacuation during laser resections, a No. 28 French chest tube was placed through one of the trocar sites and connected to a smoke evacuation system (The LASE System II, LASE Inc, model No. SE-I1111-BII). Lesions found on the edge of the lung were resected with an endoscopic stapler (Autosuture Multi-Fire Endo CIA 30-3.5 No. 030813, United States Surgical Corp). Occasionally, lesions on the edge of the lung could not be completely resected with the stapler due to the thickness of the tissue . For these lesions, resection across the base of the specimen was completed with the Nd:YAC laser. All lesions were excised with a thin rim of normal lung parenchyma . During laser resections, hemostasis was obtained along the margin of resection with the defocused Nd:YAC laser for smaller vessels «2 mm)or with the endoscopic clip applier (Autosuture Endoscopic Clip ML, No. 176615, United States Surgical Corp) for larger vessels. After completion of the wedge resection , the specimen was removed from the thoracic cavity under direct visualization. Specimens smaller than 11 mm in diameter were withdrawn directly through one of the trocars. Larger specimens were removed by enlarging one of the trocar incisions to comfortably accommodate the lesion. These larger specimens were placed in a surgical glove that was introduced into the chest through one of the trocar sites. These maneuvers allowed for the specimen to be removed intact and ensured that spillage of tumor or seeding of the trocar site did not occur, The parenchymal bed of the pulmonary resection was examined to ensure that hemostasis was complete . A single No. 28 French chest tube was placed through one of the trocar sites and guided under direct visualization to the apex of the chest . The remaining trocar incisions sites were closed with absorbable suture. The procedure was terminated after establishing 2G-cm water suction to the chest tube evacuation system. All patients received 1 g of ceftriaxone (Rocephin) intravenously in the operating room prior to the start of the procedure and two doses at 12-h intervals after the procedure. RESULTS
lbtient Population From April to September 1991, 21 patients with a history of malignancy and lesions found on chest roentgenogram were seen in consultation. Six of these 21 patients were found on computerized tomography of the chest to have lesions in the inner two thirds of the lung parenchyma. These lesions were not considered amenable to thoracoscopic resection, and these patients were not included in this study. The remaining 15 patients were found to have lesions in the outer one third of the pulmonary parenchyma and were considered candidates for thoracoscopic resection. There were six women and nine men . Mean age of the patients was 49.0 years, with a
FIGURE 1. Computerized tomograms of the chest demonstrate metastatic lesions that proved to be amenable to thoracoscopic resection : A (top) . Metastatic colon cancer on the medial surface of the right upper lobe. B (center). Metastatic renal cell carcinoma of the left upper lobe. C (bottom) . Metastatic melanoma of the left lower lobe.
range of 21 to 78 years. Fourteen patients had no symptom referable to the lung metastases, while one CHEST I 102 I 5 I NOVEMBER, 1992
1451
patient reported a chronic, nonproductive cough. Computerized tomography of the chest revealed a solitary pulmonary nodule in four patients. Primary tumor histologic findings in these patients were one each of hepatocellular carcinoma, renal cell carcinoma, adenocarcinoma of the colon, and Hodgkin's lymphoma. Five patients were found on computerized tomography to have a single nodule in each lung. Primary tumor histologic findings in these patients were one each of adenocarcinoma of the colon, melanoma, bronchoalveolar lung cancer, breast cancer, and Ewing's sarcoma. The remaining six patients had multiple bilateral pulmonary nodules (one with rectal cancer and five with hepatocellular carcinoma). In 12 of the 15 patients, the sole indication for resection was to determine if the lesions present represented metastatic disease. As noted above, other methods of obtaining a tissue diagnosis, including fine . needle aspiration biopsy had not been successful or were felt not to be indicated based on the location and/or the. size of the lesions. Three of the 15 patients underwent thoracoscopic resection both to establish a diagnosis and in an attempt to achieve a survival benefit. Two of these patients had a single lung lesion. Primary tumor histologic findings were one case each of renal cell carcinoma and adenocarcinoma of the colon. One patient with a history of melanoma had a single nodule in each lung. This patient underwent thoracoscopic resection of the lesion on the left, and one week later, had thoracoscopic resection of the lesion on the right.
Operative Findings In all patients, peripheral lesions present on computerized tomography were found at thoracoscopy by a combination of direct visualization, by palpation with a blunt endoscopic probe, or digital palpation through a slightly enlarged trocar site. Collapse of the lung, which is necessary to perform thoracoscopic examination of the pleural cavity, facilitated identification of the metastatic lesions that were effaced against the atelectatic pulmonary parenchyma. Lesions as small as 2 mm in diameter were found without difficult}: No patients were found to have metastatic disease that involved the mediastinum, the chest wall, or the parietal pleura. Fourteen of the lesions were present on the edge of the lung. Ten of these lesions were removed by performing a stapled wedge resection and the other four with a combination of the endoscopic stapler and Nd:YAG laser as described above. Seven of the lesions were on the flat surface of the lung and were removed with the Nd:YAG laser. In all cases, operative blood loss was minimal. No patients had a significant drop (>3 percent) in hematocrit value or required transfusion of blood products. 1452
Average operative time was III min, with a range of 45 to 155 min.
Specimen lbthology Nineteen of the 21 specimens revealed the presence of metastatic disease with margins free of tumor. Five patients had metastatic hepatocellular carcinoma, two patients had metastatic colon cancer, and there was one patient each with metastatic renal cell carcinoma, breast cancer, Ewing's sarcoma, melanoma, Hodgkin's lymphoma, and bronchoalveolar cell carcinoma. Two patients, one with a hepatocellular carcinoma and one with a history of rectal cancer, were found to have benign granulomatous disease. The mean diameter of the resected lesions was 0.8 em, with a range of 0.2 to 1.5 cm. Seventy-six percent of the lesions were 1.0 cm or less in greatest diameter and 47 percent of the resected lesions were less than 0.5 em in diameter.
Postoperative Course All patients were extubated either in the operating room or shortly after their arrival in the recovery room or intensive care unit. Initially, all patients were transferred to the intensive care unit for overnight observation. Presently, patients are transferred to a thoracic surgery ward from the recovery room. The chest tube was removed on the first postoperative day in patients who did not have an air leak (n = 1). Patients who developed an air leak had removal of the chest tube on the day after cessation of the air leak. Four patients had removal of the chest tube on the second postoperative da~ Three patients required chest tube drainage for three days, and one patient required drainage for four days. The mean duration of chest tube drainage was 1.8±0.1 days. No patients developed any clinical or roentgenographic evidence of pneumonia, hemothorax, or pneumothorax. The only abnormality seen on chest roentgenogram was subsegmental atelectasis that occurred in six patients.
Complications One patient who had undergone a right thoracoscopic resection developed a transient hoarseness related to left vocal cord paresis. This was attributed to traumatic intubation with a double lumen endotracheal tube. There were no other complications and no deaths.
ungth of Hospital Stay Twelve patients were discharged from the hospital after recovery from the operative procedure, while three patients remained hospitalized to receive chemotherapy In the 12 patients whose duration of hospital stay was not prolonged by the administration of Thoracoscopic Resection of Pulmonary Metastases (Dowling, Ferson, Landreneau)
chemotherapy, the mean postoperative hospital stay was 3.3±0.1 days with a range of two to five days. Six of these 12 patients were discharged on the third postoperative day. Two of the three patients who remained hospitalized to receive chemotherapy were felt to be ready for discharge on the second postoperative day, and one was ready for discharge on the fourth postoperative day. DISCUSSION
Thoracoscopy is not a new modality This endoscopic approach to diseases of the chest was introduced in 1910 to facilitate collapse therapy of tuberculosis. ie Until recently, thoracoscopy has been primarily relegated to the diagnosis of diseases of the pleura and management of uncomplicated apical bullous disease. I3- 15 ,17 Improvements in endoscopic surgical instrumentation and advances in laser technology have now expanded the role of thoracoscopy to include resection of peripheral pulmonary parenchymal lesions." Pulmonary lesions found in patients with synchronous or previously diagnosed malignancy often require resection as the presence of metastatic disease will provide prognostic information and may dictate further therapy Removal of all metastases may also favorably influence survival in a select group of patients with certain tumor histologic conditions. 1-6,18 Currently thoracotomy or median sternotomy are standard surgical approaches for pulmonary metastasectomy The early operative morbidity varies between 5 and 14 percent in recent series with these open surgical approaches. 3-7 Additionally, chronic pain syndromes are not uncommon in patients who undergo thoracotom~8t9 In a recent study, 45 percent of patients followed for over two years after traditional lateral thoracotomy reported residual postthoracotomy pain. 9 Minimally invasive (thoracoscopic) resection could potentially reduce the morbidity of pulmonary metastasectomy. The majority of pulmonary metastases are located in the periphery of the lung and are frequently immediately subpleural;" Our experience with interventional thoracoscopy led us to believe such lesions could be identified at thoracoscopy and that a parenchymal sparing resection equivalent in extent to that achieved with open approaches could be performed. In this series, exploratory thoracoscopy was successful in identifying all lesions that were located on computerized tomography in the outer one third of the lung parenchyma. We emphasize that preoperative localization with high resolution, thin cut computerized tomography is an essential aid in the thoracoscopic localization of these lesions. Endoscopic visual examination and palpation with endoscopic probes allowed for rapid identification of lesions that were immediately subpleural. Collapse of the lung im-
proved the ability to identify slightly deeper lesions that become effaced against the atelectatic lung. Slight enlargement of one of the trocar sites to allow introduction of a palpating finger was occasionally used to identify lesions that were not found with the other techniques. Additionall~ visualization of the thoracic cavity for other sites of visceral, pleural, or mediastinal metastases was felt to be superior with the thoracoscope compared to most open approaches. The Nd:YAG laser and endoscopic staplers allowed for thoracoscopic resection of peripheral pulmonary metastases that was equivalent in extent to that achieved at open thoracotomy. When frozen section analysis revealed metastatic disease and resection margins were free of tumor, thoracoscopic resection constituted definitive operative therapy. Thoracoscopic resection may result in a shortened hospital stay as exemplified by the mean stay of 3.3±0.1 days seen in this series. This compares favorably with recent series of open resection of peripheral lung lesions that have reported mean hospital stays of eight to ten days.IO.ll,20 Our experience with thoracoscopic resection of peripheral pulmonary nodules has changed our approach to the management of patients with lung lesions that are presumed to be metastases. A thorough evaluation for the presence of extrathoracic metastases is performed. In patients with favorable tumor histologic findings, complete control of the primary malignancy and minimal tumor burden, metastasectomy is performed both for diagnosis and in an attempt to provide a survival benefit. Our initial approach is now thoracoscopic resection of lesions in the outer one third of the lung parenchyma. We emphasize that open approaches are still required for deeper lesions. In patients with unfavorable primary tumor histologic findings or with multiple unresectable metastases, thoracoscopic resection is employed solely as a diagnostic maneuver. Occasionally, patients with extrathoracic metastases may be considered for resection of the lung lesion if the extrathoracic lesions are not amenable to biopsy This approach has been successful in all patients in this series. In summary, with present endoscopic surgical techniques, pulmonary parenchymal lesions in the outer one third of the lung field are readily identified at thoracoscopy. Parenchymal sparing resection equal in extent to that performed with open approaches can be achieved at thoracoscopy with the use of the Nd:YAG laser, endoscopic staplers, or both. Thoracoscopic resection constitutes definitive operative management in patients found to have metastatic lesions. Open procedures with their attendant morbidity can be avoided. Thoracoscopic resection may also result in reduced cost and hospital stay; and allow for more rapid institution of therapy for the primary disease. CHEST I 102 I 5 I NOVEMBER. 1992
1453
REFERENCES 1 Kern KA, Pass HI, Roth JA. Treatment of metastatic cancer to lung. In: Rosenberg SA, ed. Surgical treatment of pulmonary metastases. Philadelphia: JB Lippincott Co, 1987;69-100 2 van Dongen JA, van Slooten EA. The surgical treatment of pulmonary metastases. Cancer Treat Rev 1978;4:29-48 3 Roth JA. Treatment of metastatic cancer to lung. In: DeVita W, Hellman S, Rosenberg SA, eds. Cancer: principles and practice of oncology. Philadelphia: JB Lippincott Co, 1989; 2261-75 4 Mountain CF, McMurtey MJ, Hermes KE. Surgery for pulmonary metastases: a 20 year experience. Ann Thorac Surg 1984; 38:323-30
5 Putnam JB, Roth JA, Wesley MN, Johnston MR, Rosenberg SA. Analysis of prognostic factors in patients undergoing resection of pulmonary metastases from soft tissue sarcoma. J Thorac Cardiovasc Surg 1984;87:260-67 6 Gorenstein LA, Putnam JB, Natarajan MA, Balch CA, Roth JA. Improved survival after resection of pulmonary metastases from malignant melanoma. Ann Thorac Surg 1991;52:204-10 7 Roth JA, Pass HI, Wesely MN, White D, Putnam JB, Seipp C. Comparison of median sternotomy and thoracotomy for resection of pulmonary metastases in patients with soft tissue sarcoma. Ann Thorac Surg 1986; 42:134-38 8 Hazelrigg SR, Landreneau RJ, Boley TM, Priesmeyer M, Schmaltz RA, Nawarawong ~ et ale The effect of muscle-sparing versus standard posterolateral thoracotomy on pulmonary function, muscle strength and postoperative pain. J Thorac Cardiovase Surg 1991; 101:394-401 9 Dajczman E, Gordon A, Kreisman H, Wolkove N. Long-term postthoracotomy pain. Chest 1991;99:270-74 10 Landreneau R, Hazelrigg SR, Johnson JA, Boley TM, Nawara-
1454
wong ~ Curtis JJ. Nd:YAG laser assisted pulmonary resections. Ann Thorac Surg 1991;973-78 11 LoCicero J, Frederiksen ~ Hartz RS, Michaelis LL. Laserassisted parenchyma-sparing pulmonary resection. J Thorac Cardiovasc Surg 1989; 97:732-36 12 Landreneau RJ, Herlan DB, Johnson JA, BoleyTM, Nawarawong ~ Ferson PF. Thoracoscopic neodymium:yttrium-aluminum garnet laser-assisted pulmonary resection. Ann Thorac Surg 1991;52:1075-78 13 WakabayashiA, Brenner M,Kayaleh RA, Berns M~ Barker SJ, Rice SJ, et ale Thoracoscopic carbon dioxide laser treatment of bullous emphysema. Lancet 1991:881-883 14 Page RD, Jeffrey RR, Donnelly RJ. Thoracoscopy: a review of 121 consecutive surgical procedures. Ann Thorac Surg 1989· 48:66-68 ' 15 Brandt H, Loddenkemper R, Mai J. Atlas of diagnostic thoracoscopy: indications-techniques. New York: Thieme Medical Publishers Inc, 1985; 1-13 16 Jacobaeus HC. Uber die Moglichkeit die Zystoskopie bei Untersuchung seroser Hohlungen anzuwenden. Munch Med Wschr 1910; 57:2090-92 17 Menzies R, Charbonneau M. Thoracoscopy for the diagnosis of pleural disease. Ann Intern Med 1991; 114:271-76 18 Pass HI. Resection of pulmonary metastases. In: Roth ]A, Ruckdeschel JC, Weisenburger TH, eds. Thoracic oncology. Philadelphia: WB Saunders, 1989;619-29 19 Coppage L, Shaw C, Curtis AM. Metastatic disease in the chest in patients with extrathoracic malignancy J Thorac Imaging 1982;2:24-37 20 Moghissi K. Local excision of pulmonary nodular (coin) lesion with noncontact yttrium-aluminum-garnet laser. J Thorac Cardiovasc Surg 1989;97:147-51
Thoracoscopic Resection of Pulmonary Metastases (Dowling, Farson, Landreneau)
Objective: To describe the use of thoracoscopic techniques to achieve parenchymal sparing wedge resection of peripheral lung lesions in patients with a history of malignancy, and to describe the morbidity, mortality, and hospital course associated with this approach. Deaign: Case series. Setting: UDiversity hospital. ParticipGnta: PatieDts with a history of malignancy and lesions on computerized tomography in the outer ODe third of the lung parenchyma. outcome meaaurementB: Histologic analysis of resected IUDI lesions, operative 6ndings, operative time, duration of chest tube drainage and hospital stay, operative morbidity, and mortality. Baulta: Twenty-one thoracoscopic resections ofpulmonary parenchymal lesions were performed OD 15 patieDts. All peripheral lesions identified by computerized tomography were fouod at thoracoscopy and successfully resected with the Nd:YAG laser (n = 7), an endoscopic stapler (D = 10), or
M'"
pulmonary lesions in patients with a history of malignancy raise serious suspicions for the presence of metastatic disease. A tissue diagnosis is frequently required as the presence of metastatic disease will often dictate further therapy. Pulmonary metastasectomy may also favorably influence survival in patients with certain tumor histologies. 1-6 Currently, thoracotomy or median sternotomy are standard surgical approaches to pulmonary metastasectomy. Early operative morbidity has varied between 5 and 14 percent in recent series with these techniques.P" Beyond the acute perioperative difficulties seen with open approaches, chronic pain syndromes are common after thoracotomy 8.9 Recent success with open, laser-assisted pulmonary resections and advances in endoscopic surgical instrumentation have allowed for thoracoscopic management of many intrathoracic conditions.P''! These achievements encouraged us to employ thoracoscopic surgical techniques in the diagnosis and management of peripherally located lesions suspected to be metastatic disease. We report the first series of thoracoscopic resection of pulmonary metastases.
·From the Department of Cardiothoracic Surgery, Section of Thoracic Surgery, University of Pittsburgh, Pittsburgh. Manuscript received February 27; revision accepted April 29.
1410
both (n=4). The mean diameter of the lesions was 0.8 cm (range 0.2 to 1.5 cm). Histologic analysis revealed metastatic disease in 13 patients and benign disease in 2 patients. All resection margins were free of tumor. The mean duration of chest tube drainage and postoperative hospital stay were 1.8:1:0.1 and 3.3:1:0.1 days, respectivel~ Mean operative time was III min (range 45 to 155 min). One patient who underwent a right thoracoscopic resection developed a transient left vocal cord paresis. There were DO other complications and DO deaths. ConcluBion: Thoracoscopy was successful in identifying peripheral lung lesions and allowed Cor parenchymal sparing resection identical in extent to that performed with open approaches. For select patients with peripheral lung nodules felt to be metastases, thoracoscopic resection may result in reduced morbidity, cost, hospital stay and allow for more rapid institution of therapy £or the primary disease. (Chat 1992; 102:14S0-U)
METHODS Ibflents
Patients with a history of malignancy and suspicious lesions found on chest roentgenogram had computerized tomography of the chest and appropriate workup for the presence of extrathoracic metastases. Patients with lesions on computerized tomography of the chest located in the outer one third of the pulmonary parenchyma were considered candidates for thoracoscopic resection (Fig 1). In all cases, the pulmonary nodules were the only sites suspicious for metastatic diseases or were the most accessible lesions for obtaining a tissue diagnosis. Patients with a favorable tumor histologic condition and limited tumor burden underwent thoracoscopic resection of all identifiable disease. Other patients underwent thoracoscopic resection if the lesions were not amenable to percutaneous biopsy or if previous attempts at percutaneous biopsy had been unsuccessful. Operative Procedure
After induction of general anesthesia, 8exible bronchoscopy was performed through a single lumen endotracheal tube to evaluate the presence of endobronchial lesions that would provide tissue for diagnosis. In the absence of endobronchial lesions, exploratory thoracoscopy was performed." Patients were intubated with a left:sided double lumen endotracheal tube to allow for contralateral single lung ventilation and collapse of the ipsilateral lung which was essential to perform thoracoscopic surgery. After institution of single lung ventilation, an II-mm diameter trocar (Autosuture Surgiport Trocars, United States Surgical Corp) was introduced into the thoracic cavity. The site chosen for the introduction of the initial trocar was dictated by the location of the lesion. A wide angle, zero degree thoracoscope (operating wide angle zero degree Hopkin's telescope, 2604OA, Karl Storz Endoscopic America, Inc) with an operating port was introduced through this trocar, and Thoracoscopic Resection of PulmonaryMetastases (Dowling, Ferson, Lsndreneau)
exploration of the thoracic cavity was performed . Any adhesions present were divided using sharp dissection with the endoscopic scissors or blunt dissection with the endoscopic probe . Two additional trocars were introduced to allow manipulation and examination of the entire lung. The lung was examined both visually and by palpation with a blunt endoscopic probe . Collapse of the lung facilitated identification of lesions that became effaced against the atelectatic pulmonary parenchyma . Slight enlargement of one of the trocar sites to allow introduction of a palpating finger was also used to identify lesions that were not found with the other techn iques. Lesions on the Rat surface of the lung were resected with a noncontact neodymium:yttrium-aluminum garnet (Nd:YAC) laser (Laserscope Inc, model 704) that was calibrated and set on a 35 \v, continuous setting . For smoke evacuation during laser resections, a No. 28 French chest tube was placed through one of the trocar sites and connected to a smoke evacuation system (The LASE System II, LASE Inc, model No. SE-I1111-BII). Lesions found on the edge of the lung were resected with an endoscopic stapler (Autosuture Multi-Fire Endo CIA 30-3.5 No. 030813, United States Surgical Corp). Occasionally, lesions on the edge of the lung could not be completely resected with the stapler due to the thickness of the tissue . For these lesions, resection across the base of the specimen was completed with the Nd:YAC laser. All lesions were excised with a thin rim of normal lung parenchyma . During laser resections, hemostasis was obtained along the margin of resection with the defocused Nd:YAC laser for smaller vessels «2 mm)or with the endoscopic clip applier (Autosuture Endoscopic Clip ML, No. 176615, United States Surgical Corp) for larger vessels. After completion of the wedge resection , the specimen was removed from the thoracic cavity under direct visualization. Specimens smaller than 11 mm in diameter were withdrawn directly through one of the trocars. Larger specimens were removed by enlarging one of the trocar incisions to comfortably accommodate the lesion. These larger specimens were placed in a surgical glove that was introduced into the chest through one of the trocar sites. These maneuvers allowed for the specimen to be removed intact and ensured that spillage of tumor or seeding of the trocar site did not occur, The parenchymal bed of the pulmonary resection was examined to ensure that hemostasis was complete . A single No. 28 French chest tube was placed through one of the trocar sites and guided under direct visualization to the apex of the chest . The remaining trocar incisions sites were closed with absorbable suture. The procedure was terminated after establishing 2G-cm water suction to the chest tube evacuation system. All patients received 1 g of ceftriaxone (Rocephin) intravenously in the operating room prior to the start of the procedure and two doses at 12-h intervals after the procedure. RESULTS
lbtient Population From April to September 1991, 21 patients with a history of malignancy and lesions found on chest roentgenogram were seen in consultation. Six of these 21 patients were found on computerized tomography of the chest to have lesions in the inner two thirds of the lung parenchyma. These lesions were not considered amenable to thoracoscopic resection, and these patients were not included in this study. The remaining 15 patients were found to have lesions in the outer one third of the pulmonary parenchyma and were considered candidates for thoracoscopic resection. There were six women and nine men . Mean age of the patients was 49.0 years, with a
FIGURE 1. Computerized tomograms of the chest demonstrate metastatic lesions that proved to be amenable to thoracoscopic resection : A (top) . Metastatic colon cancer on the medial surface of the right upper lobe. B (center). Metastatic renal cell carcinoma of the left upper lobe. C (bottom) . Metastatic melanoma of the left lower lobe.
range of 21 to 78 years. Fourteen patients had no symptom referable to the lung metastases, while one CHEST I 102 I 5 I NOVEMBER, 1992
1451
patient reported a chronic, nonproductive cough. Computerized tomography of the chest revealed a solitary pulmonary nodule in four patients. Primary tumor histologic findings in these patients were one each of hepatocellular carcinoma, renal cell carcinoma, adenocarcinoma of the colon, and Hodgkin's lymphoma. Five patients were found on computerized tomography to have a single nodule in each lung. Primary tumor histologic findings in these patients were one each of adenocarcinoma of the colon, melanoma, bronchoalveolar lung cancer, breast cancer, and Ewing's sarcoma. The remaining six patients had multiple bilateral pulmonary nodules (one with rectal cancer and five with hepatocellular carcinoma). In 12 of the 15 patients, the sole indication for resection was to determine if the lesions present represented metastatic disease. As noted above, other methods of obtaining a tissue diagnosis, including fine . needle aspiration biopsy had not been successful or were felt not to be indicated based on the location and/or the. size of the lesions. Three of the 15 patients underwent thoracoscopic resection both to establish a diagnosis and in an attempt to achieve a survival benefit. Two of these patients had a single lung lesion. Primary tumor histologic findings were one case each of renal cell carcinoma and adenocarcinoma of the colon. One patient with a history of melanoma had a single nodule in each lung. This patient underwent thoracoscopic resection of the lesion on the left, and one week later, had thoracoscopic resection of the lesion on the right.
Operative Findings In all patients, peripheral lesions present on computerized tomography were found at thoracoscopy by a combination of direct visualization, by palpation with a blunt endoscopic probe, or digital palpation through a slightly enlarged trocar site. Collapse of the lung, which is necessary to perform thoracoscopic examination of the pleural cavity, facilitated identification of the metastatic lesions that were effaced against the atelectatic pulmonary parenchyma. Lesions as small as 2 mm in diameter were found without difficult}: No patients were found to have metastatic disease that involved the mediastinum, the chest wall, or the parietal pleura. Fourteen of the lesions were present on the edge of the lung. Ten of these lesions were removed by performing a stapled wedge resection and the other four with a combination of the endoscopic stapler and Nd:YAG laser as described above. Seven of the lesions were on the flat surface of the lung and were removed with the Nd:YAG laser. In all cases, operative blood loss was minimal. No patients had a significant drop (>3 percent) in hematocrit value or required transfusion of blood products. 1452
Average operative time was III min, with a range of 45 to 155 min.
Specimen lbthology Nineteen of the 21 specimens revealed the presence of metastatic disease with margins free of tumor. Five patients had metastatic hepatocellular carcinoma, two patients had metastatic colon cancer, and there was one patient each with metastatic renal cell carcinoma, breast cancer, Ewing's sarcoma, melanoma, Hodgkin's lymphoma, and bronchoalveolar cell carcinoma. Two patients, one with a hepatocellular carcinoma and one with a history of rectal cancer, were found to have benign granulomatous disease. The mean diameter of the resected lesions was 0.8 em, with a range of 0.2 to 1.5 cm. Seventy-six percent of the lesions were 1.0 cm or less in greatest diameter and 47 percent of the resected lesions were less than 0.5 em in diameter.
Postoperative Course All patients were extubated either in the operating room or shortly after their arrival in the recovery room or intensive care unit. Initially, all patients were transferred to the intensive care unit for overnight observation. Presently, patients are transferred to a thoracic surgery ward from the recovery room. The chest tube was removed on the first postoperative day in patients who did not have an air leak (n = 1). Patients who developed an air leak had removal of the chest tube on the day after cessation of the air leak. Four patients had removal of the chest tube on the second postoperative da~ Three patients required chest tube drainage for three days, and one patient required drainage for four days. The mean duration of chest tube drainage was 1.8±0.1 days. No patients developed any clinical or roentgenographic evidence of pneumonia, hemothorax, or pneumothorax. The only abnormality seen on chest roentgenogram was subsegmental atelectasis that occurred in six patients.
Complications One patient who had undergone a right thoracoscopic resection developed a transient hoarseness related to left vocal cord paresis. This was attributed to traumatic intubation with a double lumen endotracheal tube. There were no other complications and no deaths.
ungth of Hospital Stay Twelve patients were discharged from the hospital after recovery from the operative procedure, while three patients remained hospitalized to receive chemotherapy In the 12 patients whose duration of hospital stay was not prolonged by the administration of Thoracoscopic Resection of Pulmonary Metastases (Dowling, Ferson, Landreneau)
chemotherapy, the mean postoperative hospital stay was 3.3±0.1 days with a range of two to five days. Six of these 12 patients were discharged on the third postoperative day. Two of the three patients who remained hospitalized to receive chemotherapy were felt to be ready for discharge on the second postoperative day, and one was ready for discharge on the fourth postoperative day. DISCUSSION
Thoracoscopy is not a new modality This endoscopic approach to diseases of the chest was introduced in 1910 to facilitate collapse therapy of tuberculosis. ie Until recently, thoracoscopy has been primarily relegated to the diagnosis of diseases of the pleura and management of uncomplicated apical bullous disease. I3- 15 ,17 Improvements in endoscopic surgical instrumentation and advances in laser technology have now expanded the role of thoracoscopy to include resection of peripheral pulmonary parenchymal lesions." Pulmonary lesions found in patients with synchronous or previously diagnosed malignancy often require resection as the presence of metastatic disease will provide prognostic information and may dictate further therapy Removal of all metastases may also favorably influence survival in a select group of patients with certain tumor histologic conditions. 1-6,18 Currently thoracotomy or median sternotomy are standard surgical approaches for pulmonary metastasectomy The early operative morbidity varies between 5 and 14 percent in recent series with these open surgical approaches. 3-7 Additionally, chronic pain syndromes are not uncommon in patients who undergo thoracotom~8t9 In a recent study, 45 percent of patients followed for over two years after traditional lateral thoracotomy reported residual postthoracotomy pain. 9 Minimally invasive (thoracoscopic) resection could potentially reduce the morbidity of pulmonary metastasectomy. The majority of pulmonary metastases are located in the periphery of the lung and are frequently immediately subpleural;" Our experience with interventional thoracoscopy led us to believe such lesions could be identified at thoracoscopy and that a parenchymal sparing resection equivalent in extent to that achieved with open approaches could be performed. In this series, exploratory thoracoscopy was successful in identifying all lesions that were located on computerized tomography in the outer one third of the lung parenchyma. We emphasize that preoperative localization with high resolution, thin cut computerized tomography is an essential aid in the thoracoscopic localization of these lesions. Endoscopic visual examination and palpation with endoscopic probes allowed for rapid identification of lesions that were immediately subpleural. Collapse of the lung im-
proved the ability to identify slightly deeper lesions that become effaced against the atelectatic lung. Slight enlargement of one of the trocar sites to allow introduction of a palpating finger was occasionally used to identify lesions that were not found with the other techniques. Additionall~ visualization of the thoracic cavity for other sites of visceral, pleural, or mediastinal metastases was felt to be superior with the thoracoscope compared to most open approaches. The Nd:YAG laser and endoscopic staplers allowed for thoracoscopic resection of peripheral pulmonary metastases that was equivalent in extent to that achieved at open thoracotomy. When frozen section analysis revealed metastatic disease and resection margins were free of tumor, thoracoscopic resection constituted definitive operative therapy. Thoracoscopic resection may result in a shortened hospital stay as exemplified by the mean stay of 3.3±0.1 days seen in this series. This compares favorably with recent series of open resection of peripheral lung lesions that have reported mean hospital stays of eight to ten days.IO.ll,20 Our experience with thoracoscopic resection of peripheral pulmonary nodules has changed our approach to the management of patients with lung lesions that are presumed to be metastases. A thorough evaluation for the presence of extrathoracic metastases is performed. In patients with favorable tumor histologic findings, complete control of the primary malignancy and minimal tumor burden, metastasectomy is performed both for diagnosis and in an attempt to provide a survival benefit. Our initial approach is now thoracoscopic resection of lesions in the outer one third of the lung parenchyma. We emphasize that open approaches are still required for deeper lesions. In patients with unfavorable primary tumor histologic findings or with multiple unresectable metastases, thoracoscopic resection is employed solely as a diagnostic maneuver. Occasionally, patients with extrathoracic metastases may be considered for resection of the lung lesion if the extrathoracic lesions are not amenable to biopsy This approach has been successful in all patients in this series. In summary, with present endoscopic surgical techniques, pulmonary parenchymal lesions in the outer one third of the lung field are readily identified at thoracoscopy. Parenchymal sparing resection equal in extent to that performed with open approaches can be achieved at thoracoscopy with the use of the Nd:YAG laser, endoscopic staplers, or both. Thoracoscopic resection constitutes definitive operative management in patients found to have metastatic lesions. Open procedures with their attendant morbidity can be avoided. Thoracoscopic resection may also result in reduced cost and hospital stay; and allow for more rapid institution of therapy for the primary disease. CHEST I 102 I 5 I NOVEMBER. 1992
1453
REFERENCES 1 Kern KA, Pass HI, Roth JA. Treatment of metastatic cancer to lung. In: Rosenberg SA, ed. Surgical treatment of pulmonary metastases. Philadelphia: JB Lippincott Co, 1987;69-100 2 van Dongen JA, van Slooten EA. The surgical treatment of pulmonary metastases. Cancer Treat Rev 1978;4:29-48 3 Roth JA. Treatment of metastatic cancer to lung. In: DeVita W, Hellman S, Rosenberg SA, eds. Cancer: principles and practice of oncology. Philadelphia: JB Lippincott Co, 1989; 2261-75 4 Mountain CF, McMurtey MJ, Hermes KE. Surgery for pulmonary metastases: a 20 year experience. Ann Thorac Surg 1984; 38:323-30
5 Putnam JB, Roth JA, Wesley MN, Johnston MR, Rosenberg SA. Analysis of prognostic factors in patients undergoing resection of pulmonary metastases from soft tissue sarcoma. J Thorac Cardiovasc Surg 1984;87:260-67 6 Gorenstein LA, Putnam JB, Natarajan MA, Balch CA, Roth JA. Improved survival after resection of pulmonary metastases from malignant melanoma. Ann Thorac Surg 1991;52:204-10 7 Roth JA, Pass HI, Wesely MN, White D, Putnam JB, Seipp C. Comparison of median sternotomy and thoracotomy for resection of pulmonary metastases in patients with soft tissue sarcoma. Ann Thorac Surg 1986; 42:134-38 8 Hazelrigg SR, Landreneau RJ, Boley TM, Priesmeyer M, Schmaltz RA, Nawarawong ~ et ale The effect of muscle-sparing versus standard posterolateral thoracotomy on pulmonary function, muscle strength and postoperative pain. J Thorac Cardiovase Surg 1991; 101:394-401 9 Dajczman E, Gordon A, Kreisman H, Wolkove N. Long-term postthoracotomy pain. Chest 1991;99:270-74 10 Landreneau R, Hazelrigg SR, Johnson JA, Boley TM, Nawara-
1454
wong ~ Curtis JJ. Nd:YAG laser assisted pulmonary resections. Ann Thorac Surg 1991;973-78 11 LoCicero J, Frederiksen ~ Hartz RS, Michaelis LL. Laserassisted parenchyma-sparing pulmonary resection. J Thorac Cardiovasc Surg 1989; 97:732-36 12 Landreneau RJ, Herlan DB, Johnson JA, BoleyTM, Nawarawong ~ Ferson PF. Thoracoscopic neodymium:yttrium-aluminum garnet laser-assisted pulmonary resection. Ann Thorac Surg 1991;52:1075-78 13 WakabayashiA, Brenner M,Kayaleh RA, Berns M~ Barker SJ, Rice SJ, et ale Thoracoscopic carbon dioxide laser treatment of bullous emphysema. Lancet 1991:881-883 14 Page RD, Jeffrey RR, Donnelly RJ. Thoracoscopy: a review of 121 consecutive surgical procedures. Ann Thorac Surg 1989· 48:66-68 ' 15 Brandt H, Loddenkemper R, Mai J. Atlas of diagnostic thoracoscopy: indications-techniques. New York: Thieme Medical Publishers Inc, 1985; 1-13 16 Jacobaeus HC. Uber die Moglichkeit die Zystoskopie bei Untersuchung seroser Hohlungen anzuwenden. Munch Med Wschr 1910; 57:2090-92 17 Menzies R, Charbonneau M. Thoracoscopy for the diagnosis of pleural disease. Ann Intern Med 1991; 114:271-76 18 Pass HI. Resection of pulmonary metastases. In: Roth ]A, Ruckdeschel JC, Weisenburger TH, eds. Thoracic oncology. Philadelphia: WB Saunders, 1989;619-29 19 Coppage L, Shaw C, Curtis AM. Metastatic disease in the chest in patients with extrathoracic malignancy J Thorac Imaging 1982;2:24-37 20 Moghissi K. Local excision of pulmonary nodular (coin) lesion with noncontact yttrium-aluminum-garnet laser. J Thorac Cardiovasc Surg 1989;97:147-51
Thoracoscopic Resection of Pulmonary Metastases (Dowling, Farson, Landreneau)
