- Email: [email protected]
Anatomic lobectomy of the lung by means of thoracoscopy
General Thoracic Surgery
Anatomic lobectomy of the lung by means of thoracoscopy An experimental study We conducted this study in an attempt to investigate whether anatomic lobectomy can be performed through a thoracoscope. Twelve lower (seven left and five right) and six left upper lobectomies were done in 17 pigs weighing 60 to 90 kg. With the use of general anesthesia and one-lung ventilation, a thoracoscope and operating instruments were inserted through four or five trocars. First, the incomplete fissure was bluntly dissected with specially designed cotton-tipped dissectors and divided by an electrocautery spatula or sapphire contact-tip neodymium: yttrium aluminum garnet laser probe. The pulmonary artery and its branches were similarly dissected. Two ligatures were passed around the vessels by endoscopic right-angle clamps and tied by means of an extracorporeal tying technique. The vessels were clipped and divided. In two animals, a hidden branch of the pulmonary artery was inadvertently cut. Bleeding was readily controlled with the dissector until metal clips could be applied. The bronchus was similarly dissected and divided and the proximal stump was closed with sutures. The pulmonary vein was dissected bluntly by the same technique and divided. The excised lobe was placed in a plastic bag and pulled out through one of the trocar holes after the hole had been enlarged to 3 em. Blood loss was minimal. In conclusion, anatomic lobectomy of the lung by means of thoracoscopy was feasible in the pig. (J THORAC CARDIOVASC SURG 1993;105:729-31)
Tadasu Kohno, MD, Tatsuya Murakami, MD, and Akio Wakabayashi, MD, Irvine, Calif.
In the early part of the twentieth century, many surgeons attempted lobectomy of the lungs but failed. I, 2 Anatomic dissection and ligation of the pulmonary vessels and bronchus for lobectomy were described, but all attempts were unsuccessful until Churchill 3 succeeded in 1932. The current standard technique of anatomic lobectomy has evolved over the years thanks to the contributions of many pioneers, and it carries a low mortality rate. From the Department of Surgery, University of California Irvine, Irvine. Calif.
However, thoracotomy is painful and is associated with substantial morbidity." On the other hand, as instrumentation has improved, thoracoscopy has emerged as an attractive alternative to thoracotomy.i' Because thoracoscopy is performed through several small holes between the ribs, the postoperative pain is markedly less and the recovery period much shorter than after conventional thoracotomy. We conducted this animal study to investigate whether we could follow in the footsteps of our pioneers and perform anatomic lobectomy through a thoracoscope.
Read at the Eighteenth Annual Meeting of The Western Thoracic Surgical Association, Kauai, Hawaii, June 24-27, 1992.
Methods
Address for reprints: Akio Wakabayashi, MD, Cardiothoracic Surgery Division, University of California Irvine Medical Center, PO Box 14091, Orange, CA 92613-4091. Copyright
1993 by Mosby-Year Book, Inc.
0022-5223/93 $1.00 +.10
12/6/44494
Instruments. A rigid 10 mm thoracoscope (0 or 30 degrees, Olympus Corp., Lake Success, N.Y.) was used with a xenon light source, a direct coupler video camera, and two color monitors. A neodymium-yttrium aluminum garnet (Nd:YAG) laser with a sapphire contact tip (SLT Surgical Laser Technol-
729
730
The Journal of Thoracic and Cardiovascular Surgery April 1993
Kohno, Murakami, Wakabayashi
ogy, Oaks, Pa.) was used at 14 to 17watts, in a continuous wave mode, to divide the incomplete fissures and inferior pulmonary ligament. For the dissection of the pulmonary artery (PA) and vein (PY), a 10 mm endoscopic cotton-tipped dissector and right-angle clamps were developed (Ethicon, Inc., Cincinnati, Ohio). Anesthesia was maintained by servo-controlled anesthesia/ventilator equipment (900C, Siemens Corp., Danvers, Mass.). Physiologic parameters were continuously displayed on a multichannel screen (Sirecust 1281, Siemens). Anesthesia and surgical techniques. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH publication No. 86-23, revised 1985). Seventeen pigs weighing 70 to 90 kg were used. General anesthesia was induced with intramuscular xylazine hydrochloride (0.8 mg/kg), ketamine hydrochloride (6.25 rug/kg), atropine sulfate (0.023 rug/kg), and intravenous sodium pentobarbital (7.8 mg/kg). Through a tracheostomy, a left-sided endobronchial double-lumen tube (Broncho-Cath, Mallinckrodt Medical, Inc., St. Louis, Mo.) was inserted with the aid of a flexible pediatric bronchoscope (Olympus). Anesthesia was maintained by halothane (1.0% to 1.5%). Routine intraoperative monitors included the electrocardiogram, end-tidal carbon dioxide tension, arterial blood pressure, airway pressure, and expiratory tidal volume, which were displayed on the monitor screen. Twelve lower (seven left and five right) and six left upper lobectomies were performed. Because the right upper lobe in the pig takes off directly from the trachea, a right upper lobectomy was not done. Four to five trocars ( 10 mm) were placed in the lateral chest wall, and the thoracoscope and other operating instruments were inserted through them. For lower lobectomy, the incomplete fissure was first bluntly dissected, with two dissectors being used to exposethe PA. Then the lowerlobe PA and its first branch were bluntly dissected by the dissectors in a similar fashion. After dissection, a right-angle clamp was passed under them and 2-0 silk was passed around. A slipknot was made and the knot was pushed in by a knot pusher. After two ties were placed, a metal medium-large clip (Ligaclips, Ethicon) was placed on the proximal side and the vessel was divided by a hook scissors. The left upper lobe PA had three to five branches. They were bluntly dissected individually and doubly ligated in the same fashion. After the PA was divided, the bronchus was dissected bluntly with two dissectors and divided. The proximal stump was closed by interrupted or continuous 3-0 Prolene stitches (Ethicon) with the aid of an elongated needle holder and a needle catcher. The PY was dissected and divided last in a similar fashion. The incomplete fissure and inferior pulmonary ligament were divided by either an electrocautery spatula (Olympus) or sapphire contact Nd:Y AG laser probe (SLT). A prototype 50 mm endoscopic stapler (Ethicon) was used in one animal for stapling of the incomplete fissure. The excised lobe was placed in a plastic bag that was inserted into the chest cavity through the trocar. One trocar was removed and the hole was enlarged to approximately 3 em long. Then the plastic bag was pulled out of the chest. The air leak from the bronchial stump was checked by immersion of the lung under normal saline solution and inflation of the lung to 25 em H 20 . Then the animals were killed by an intravenous overdose of barbiturates.
Results All attempted thoracoscopic lobectomies were completed successfully. The operating time varied from 2 to 4 hours. Upper lobectomy took a longer time than lower lobectomy. In two animals, the superior segmental branch of the lower lobe PA was inadvertently cut. Bleeding was readily controlled by compression with the dissector and the clip (Ligaclip, Ethicon). In both instances, the blood loss was approximately 150 m\. In all others, blood loss was minimal, less than 50 m\. No injury to the blood vessels occurred during the ligation when the extracorporeal tying technique was used. Most blood loss occurred during the dissection and division of the incomplete fissure. A contact Nd:YAG laser scalpel produced less bleeding than an electrocautery spatula. The 50 mm stapler was difficult to maneuver inside the chest. Once it was set, however, it divided the fissure without any blood loss. No air leak was noted from the bronchial stump. Extraction of the excised lobe did not present any problem.
Discussion Our three major concerns before we started the experiments were (I) how to handle the pulmonary vessels, (2) how to control bleeding from the pulmonary vessels, and (3) how to extract the excised lobe. Before the experiment started, the senior author (A.W.) developed the 10 mm cotton-tipped dissector as a dissecting tool of the pulmonary vessels. It was found to be useful and safe for this purpose. We also evaluated a commercially available 5 mm endoscopic Kittner dissector but found that it was too sharp and frequently injured the blood vessels. The large, round, firm cotton tip of our dissector was effective in controlling bleeding from the pulmonary artery. We used two dissectors at the same time. One was used to hold the tissues while the other dissected the tissues. This countertension facilitated the process. In case of bleeding, the left-hand dissector quickly controlled the bleeding by compression until the metal clip was applied to the bleeder by the right hand. In two such incidents, we were able to control bleeding relatively easily. However, it is still uncertain whether we could always control brisk bleeding from the pulmonary artery. Further animal experiments to test the various techniques of controlling brisk bleeding through a thoracoscope must be done before thoracoscopic lobectomy can be routinely performed in human beings. A large 50 mm endoscopic stapler was rather difficult to maneuver inside the pig's chest cavity. This may be related to the pig's anatomy. The pig's chest is flatter than that of human beings and there is less space in which to work. Therefore any instruments that can be used inside the pig chest should be easier to use inside the human chest. Although the contact Nd: YAG laser probe
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 4
was satisfactory in dividing the incomplete fissure, the endoscopic stapler appears to be the instrument of choice. During routine open lobectomy, we prefer the stapler for the bronchus. However, the straight endoscopic stapler could not be applied to the bronchus perpendicularly in this experiment. Unless the angle of the endoscopic stapler could be adjusted inside the chest, it could not be used for the bronchus. An articulated stapler may be useful for bronchial division and will be tested in animals in future. Because the normal lobe of pigs is elastic, it can be pulled out of a small hole. If the specimen contains a rigid tumor that is larger than the intercostal space, however, it cannot be pulled out. A provision should be made to make the tumor smaller without compromising the quality of the pathologic specimen in clinical application. In conclusion, our animal study has demonstrated that individual ligation and division of the pulmonary vessels and bronchus can be performed safely by thoracoscopy. However, further experimental studies and development of better instruments are needed before this technique can be applied to human beings. REFERENCES I. Davies HM. Recent advances in the surgery of the lungs and pleura. Br J Surg 1913;1:228. 2. McElvein RB. Concatenations. Ann Thorac Surg 1987; 43:463-8. 3. Churchill ED. The surgical treatment of carcinoma of the lung. J THORAC SURG 1932;2:254-66. 4. Mann LJ, Young GR, Williams JK, et al. Intrapleural bupivacaine in the control of postthoracotomy pain. Ann Thorac Surg 1992;53:449-54. 5. Wakabayashi A. Expanded applications of diagnosis and therapeutic thoracoscopy. ] THORAC CARDIOVASC SURG 1991;102:721-3.
Discussion Dr. Rodney J. Landreneau (Pittsburgh, Pa.). It is difficult to conclude from these acute experiments in pigs that thoracoscopic lobectomy is a generally safe and feasible alternative to lobectomy performed through open thoracotomy. Although thoracic wall blood loss in this animal preparation is described as minimal, I believe most thoracic surgeons are primarily concerned about the potential for rapid blood loss in more centrally located structures. I am also concerned about the reliance on the use of endoscopic suturing techniques for bronchial closure, which most of us have extreme difficulty performing. At present, I think we can all embrace video-assisted thoracic surgical techniques for examination of the pleural cavity and for conservative wedge resection biopsy of small peripheral indeterminate lung abnormalities. However, present open public scrutiny of endoscopic surgical practices in this country is becoming more intense. It is mandatory that we thoracic
Kohno, Murakami, Wakabayashi
73 I
surgeons move ahead cautiously to avoid unnecessary complications and the recent criticisms being directed toward many general surgeons performing laparoscopic cholecystectomy. In my opinion, the future role of video-assisted lobectomy remains in question. It certainly is not a procedure to be attempted in one's early thoracoscopic experience. We performed over 140 pulmonary wedge resections before attempting our first lobectomy, and that was only after a number of lobectomies in animals, such as Dr. Kohne's group has done. We must credit Dr. Tom Kirby at the Clevland Clinic with bringing video-assisted anatomic lobectomy under clinical investigation in this country. A group of us are now collaborating with Dr. Kirby and have performed 40 video-assisted major pulmonary resections involving all lobes of the lung and one left pneumonectomy. Thorough analysis ofthis experience is underway, but it appears that the greatest benefit from this approach may be reduced perioperative pain and an earlier recuperation for the patient. However, the operative time is significantly longer and the length of hospital stay has been minimally reduced. Dr. Kohno's laboratory work and the limited clinical experience I describe should not be looked on as a mandate for thoracic surgeons to begin attempting video-assisted lobectomy. Rather than feeling compelled to do so, I believe thoracic surgeons should begin with incorporating an initial thoracoscopic exploration as part of their standard open thoracotomy routine to familiarize their operative team with the general setup and principles of this technique. Conservative pulmonary wedge resection biopsy is appropriate for peripherally based lung masses. This approach will help to solidify the thoracic surgeon's position as the thoracoscopist in his community and avoid the quiet internal and external pressures to become involved with this technique. Referring physicians will remain happy and loyal, and the thoracic surgeon can be more comfortable to work within himself as his experience with video-assisted thoracic surgery grows. I congratulate Dr. Kohno, Dr. Murakami, and Dr. Wakabayashi on this diligent work. We must remember that, as we strive to reduce the overall pain and disability after thoracic surgery, we should not substitute the proper emphasis on the therapeutic goals of pulmonary resection with an overenthusiasm with the means that we surgeons use. Dr. Robert L. Mitchell (Mountain View, Calif). What are the goals for thoracoscopy resection? Because this technique cannot improve the safety of the procedure, we have to look to decreasing morbidity, and the first thing that comes to mind is pain. You use five holes in the chest. Will that indeed cause less pain than a limited incision through a thoracic approach as originally described by Doar? I do not think blood loss should be a question. We do not consider crossmatching to be necessary for thoracotomy. In my own experience, about four or five patients in 600 have required transfusion. Last, is there any evidence that this technique will reduce the compromise in pulmonary function that occurs after a thoracotomy? Dr. Kohno. We have not started a clinical trial. As soon as the instruments are approved by the Food and Drug Administration, we will start using the procedure in selected patients. We have not studied pulmonary function of the patients after lung resection (e.g., wedge resection) by thoracoscopy. We will need to do a randomized study in the future.
Anatomic lobectomy of the lung by means of thoracoscopy An experimental study We conducted this study in an attempt to investigate whether anatomic lobectomy can be performed through a thoracoscope. Twelve lower (seven left and five right) and six left upper lobectomies were done in 17 pigs weighing 60 to 90 kg. With the use of general anesthesia and one-lung ventilation, a thoracoscope and operating instruments were inserted through four or five trocars. First, the incomplete fissure was bluntly dissected with specially designed cotton-tipped dissectors and divided by an electrocautery spatula or sapphire contact-tip neodymium: yttrium aluminum garnet laser probe. The pulmonary artery and its branches were similarly dissected. Two ligatures were passed around the vessels by endoscopic right-angle clamps and tied by means of an extracorporeal tying technique. The vessels were clipped and divided. In two animals, a hidden branch of the pulmonary artery was inadvertently cut. Bleeding was readily controlled with the dissector until metal clips could be applied. The bronchus was similarly dissected and divided and the proximal stump was closed with sutures. The pulmonary vein was dissected bluntly by the same technique and divided. The excised lobe was placed in a plastic bag and pulled out through one of the trocar holes after the hole had been enlarged to 3 em. Blood loss was minimal. In conclusion, anatomic lobectomy of the lung by means of thoracoscopy was feasible in the pig. (J THORAC CARDIOVASC SURG 1993;105:729-31)
Tadasu Kohno, MD, Tatsuya Murakami, MD, and Akio Wakabayashi, MD, Irvine, Calif.
In the early part of the twentieth century, many surgeons attempted lobectomy of the lungs but failed. I, 2 Anatomic dissection and ligation of the pulmonary vessels and bronchus for lobectomy were described, but all attempts were unsuccessful until Churchill 3 succeeded in 1932. The current standard technique of anatomic lobectomy has evolved over the years thanks to the contributions of many pioneers, and it carries a low mortality rate. From the Department of Surgery, University of California Irvine, Irvine. Calif.
However, thoracotomy is painful and is associated with substantial morbidity." On the other hand, as instrumentation has improved, thoracoscopy has emerged as an attractive alternative to thoracotomy.i' Because thoracoscopy is performed through several small holes between the ribs, the postoperative pain is markedly less and the recovery period much shorter than after conventional thoracotomy. We conducted this animal study to investigate whether we could follow in the footsteps of our pioneers and perform anatomic lobectomy through a thoracoscope.
Read at the Eighteenth Annual Meeting of The Western Thoracic Surgical Association, Kauai, Hawaii, June 24-27, 1992.
Methods
Address for reprints: Akio Wakabayashi, MD, Cardiothoracic Surgery Division, University of California Irvine Medical Center, PO Box 14091, Orange, CA 92613-4091. Copyright
1993 by Mosby-Year Book, Inc.
0022-5223/93 $1.00 +.10
12/6/44494
Instruments. A rigid 10 mm thoracoscope (0 or 30 degrees, Olympus Corp., Lake Success, N.Y.) was used with a xenon light source, a direct coupler video camera, and two color monitors. A neodymium-yttrium aluminum garnet (Nd:YAG) laser with a sapphire contact tip (SLT Surgical Laser Technol-
729
730
The Journal of Thoracic and Cardiovascular Surgery April 1993
Kohno, Murakami, Wakabayashi
ogy, Oaks, Pa.) was used at 14 to 17watts, in a continuous wave mode, to divide the incomplete fissures and inferior pulmonary ligament. For the dissection of the pulmonary artery (PA) and vein (PY), a 10 mm endoscopic cotton-tipped dissector and right-angle clamps were developed (Ethicon, Inc., Cincinnati, Ohio). Anesthesia was maintained by servo-controlled anesthesia/ventilator equipment (900C, Siemens Corp., Danvers, Mass.). Physiologic parameters were continuously displayed on a multichannel screen (Sirecust 1281, Siemens). Anesthesia and surgical techniques. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH publication No. 86-23, revised 1985). Seventeen pigs weighing 70 to 90 kg were used. General anesthesia was induced with intramuscular xylazine hydrochloride (0.8 mg/kg), ketamine hydrochloride (6.25 rug/kg), atropine sulfate (0.023 rug/kg), and intravenous sodium pentobarbital (7.8 mg/kg). Through a tracheostomy, a left-sided endobronchial double-lumen tube (Broncho-Cath, Mallinckrodt Medical, Inc., St. Louis, Mo.) was inserted with the aid of a flexible pediatric bronchoscope (Olympus). Anesthesia was maintained by halothane (1.0% to 1.5%). Routine intraoperative monitors included the electrocardiogram, end-tidal carbon dioxide tension, arterial blood pressure, airway pressure, and expiratory tidal volume, which were displayed on the monitor screen. Twelve lower (seven left and five right) and six left upper lobectomies were performed. Because the right upper lobe in the pig takes off directly from the trachea, a right upper lobectomy was not done. Four to five trocars ( 10 mm) were placed in the lateral chest wall, and the thoracoscope and other operating instruments were inserted through them. For lower lobectomy, the incomplete fissure was first bluntly dissected, with two dissectors being used to exposethe PA. Then the lowerlobe PA and its first branch were bluntly dissected by the dissectors in a similar fashion. After dissection, a right-angle clamp was passed under them and 2-0 silk was passed around. A slipknot was made and the knot was pushed in by a knot pusher. After two ties were placed, a metal medium-large clip (Ligaclips, Ethicon) was placed on the proximal side and the vessel was divided by a hook scissors. The left upper lobe PA had three to five branches. They were bluntly dissected individually and doubly ligated in the same fashion. After the PA was divided, the bronchus was dissected bluntly with two dissectors and divided. The proximal stump was closed by interrupted or continuous 3-0 Prolene stitches (Ethicon) with the aid of an elongated needle holder and a needle catcher. The PY was dissected and divided last in a similar fashion. The incomplete fissure and inferior pulmonary ligament were divided by either an electrocautery spatula (Olympus) or sapphire contact Nd:Y AG laser probe (SLT). A prototype 50 mm endoscopic stapler (Ethicon) was used in one animal for stapling of the incomplete fissure. The excised lobe was placed in a plastic bag that was inserted into the chest cavity through the trocar. One trocar was removed and the hole was enlarged to approximately 3 em long. Then the plastic bag was pulled out of the chest. The air leak from the bronchial stump was checked by immersion of the lung under normal saline solution and inflation of the lung to 25 em H 20 . Then the animals were killed by an intravenous overdose of barbiturates.
Results All attempted thoracoscopic lobectomies were completed successfully. The operating time varied from 2 to 4 hours. Upper lobectomy took a longer time than lower lobectomy. In two animals, the superior segmental branch of the lower lobe PA was inadvertently cut. Bleeding was readily controlled by compression with the dissector and the clip (Ligaclip, Ethicon). In both instances, the blood loss was approximately 150 m\. In all others, blood loss was minimal, less than 50 m\. No injury to the blood vessels occurred during the ligation when the extracorporeal tying technique was used. Most blood loss occurred during the dissection and division of the incomplete fissure. A contact Nd:YAG laser scalpel produced less bleeding than an electrocautery spatula. The 50 mm stapler was difficult to maneuver inside the chest. Once it was set, however, it divided the fissure without any blood loss. No air leak was noted from the bronchial stump. Extraction of the excised lobe did not present any problem.
Discussion Our three major concerns before we started the experiments were (I) how to handle the pulmonary vessels, (2) how to control bleeding from the pulmonary vessels, and (3) how to extract the excised lobe. Before the experiment started, the senior author (A.W.) developed the 10 mm cotton-tipped dissector as a dissecting tool of the pulmonary vessels. It was found to be useful and safe for this purpose. We also evaluated a commercially available 5 mm endoscopic Kittner dissector but found that it was too sharp and frequently injured the blood vessels. The large, round, firm cotton tip of our dissector was effective in controlling bleeding from the pulmonary artery. We used two dissectors at the same time. One was used to hold the tissues while the other dissected the tissues. This countertension facilitated the process. In case of bleeding, the left-hand dissector quickly controlled the bleeding by compression until the metal clip was applied to the bleeder by the right hand. In two such incidents, we were able to control bleeding relatively easily. However, it is still uncertain whether we could always control brisk bleeding from the pulmonary artery. Further animal experiments to test the various techniques of controlling brisk bleeding through a thoracoscope must be done before thoracoscopic lobectomy can be routinely performed in human beings. A large 50 mm endoscopic stapler was rather difficult to maneuver inside the pig's chest cavity. This may be related to the pig's anatomy. The pig's chest is flatter than that of human beings and there is less space in which to work. Therefore any instruments that can be used inside the pig chest should be easier to use inside the human chest. Although the contact Nd: YAG laser probe
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 4
was satisfactory in dividing the incomplete fissure, the endoscopic stapler appears to be the instrument of choice. During routine open lobectomy, we prefer the stapler for the bronchus. However, the straight endoscopic stapler could not be applied to the bronchus perpendicularly in this experiment. Unless the angle of the endoscopic stapler could be adjusted inside the chest, it could not be used for the bronchus. An articulated stapler may be useful for bronchial division and will be tested in animals in future. Because the normal lobe of pigs is elastic, it can be pulled out of a small hole. If the specimen contains a rigid tumor that is larger than the intercostal space, however, it cannot be pulled out. A provision should be made to make the tumor smaller without compromising the quality of the pathologic specimen in clinical application. In conclusion, our animal study has demonstrated that individual ligation and division of the pulmonary vessels and bronchus can be performed safely by thoracoscopy. However, further experimental studies and development of better instruments are needed before this technique can be applied to human beings. REFERENCES I. Davies HM. Recent advances in the surgery of the lungs and pleura. Br J Surg 1913;1:228. 2. McElvein RB. Concatenations. Ann Thorac Surg 1987; 43:463-8. 3. Churchill ED. The surgical treatment of carcinoma of the lung. J THORAC SURG 1932;2:254-66. 4. Mann LJ, Young GR, Williams JK, et al. Intrapleural bupivacaine in the control of postthoracotomy pain. Ann Thorac Surg 1992;53:449-54. 5. Wakabayashi A. Expanded applications of diagnosis and therapeutic thoracoscopy. ] THORAC CARDIOVASC SURG 1991;102:721-3.
Discussion Dr. Rodney J. Landreneau (Pittsburgh, Pa.). It is difficult to conclude from these acute experiments in pigs that thoracoscopic lobectomy is a generally safe and feasible alternative to lobectomy performed through open thoracotomy. Although thoracic wall blood loss in this animal preparation is described as minimal, I believe most thoracic surgeons are primarily concerned about the potential for rapid blood loss in more centrally located structures. I am also concerned about the reliance on the use of endoscopic suturing techniques for bronchial closure, which most of us have extreme difficulty performing. At present, I think we can all embrace video-assisted thoracic surgical techniques for examination of the pleural cavity and for conservative wedge resection biopsy of small peripheral indeterminate lung abnormalities. However, present open public scrutiny of endoscopic surgical practices in this country is becoming more intense. It is mandatory that we thoracic
Kohno, Murakami, Wakabayashi
73 I
surgeons move ahead cautiously to avoid unnecessary complications and the recent criticisms being directed toward many general surgeons performing laparoscopic cholecystectomy. In my opinion, the future role of video-assisted lobectomy remains in question. It certainly is not a procedure to be attempted in one's early thoracoscopic experience. We performed over 140 pulmonary wedge resections before attempting our first lobectomy, and that was only after a number of lobectomies in animals, such as Dr. Kohne's group has done. We must credit Dr. Tom Kirby at the Clevland Clinic with bringing video-assisted anatomic lobectomy under clinical investigation in this country. A group of us are now collaborating with Dr. Kirby and have performed 40 video-assisted major pulmonary resections involving all lobes of the lung and one left pneumonectomy. Thorough analysis ofthis experience is underway, but it appears that the greatest benefit from this approach may be reduced perioperative pain and an earlier recuperation for the patient. However, the operative time is significantly longer and the length of hospital stay has been minimally reduced. Dr. Kohno's laboratory work and the limited clinical experience I describe should not be looked on as a mandate for thoracic surgeons to begin attempting video-assisted lobectomy. Rather than feeling compelled to do so, I believe thoracic surgeons should begin with incorporating an initial thoracoscopic exploration as part of their standard open thoracotomy routine to familiarize their operative team with the general setup and principles of this technique. Conservative pulmonary wedge resection biopsy is appropriate for peripherally based lung masses. This approach will help to solidify the thoracic surgeon's position as the thoracoscopist in his community and avoid the quiet internal and external pressures to become involved with this technique. Referring physicians will remain happy and loyal, and the thoracic surgeon can be more comfortable to work within himself as his experience with video-assisted thoracic surgery grows. I congratulate Dr. Kohno, Dr. Murakami, and Dr. Wakabayashi on this diligent work. We must remember that, as we strive to reduce the overall pain and disability after thoracic surgery, we should not substitute the proper emphasis on the therapeutic goals of pulmonary resection with an overenthusiasm with the means that we surgeons use. Dr. Robert L. Mitchell (Mountain View, Calif). What are the goals for thoracoscopy resection? Because this technique cannot improve the safety of the procedure, we have to look to decreasing morbidity, and the first thing that comes to mind is pain. You use five holes in the chest. Will that indeed cause less pain than a limited incision through a thoracic approach as originally described by Doar? I do not think blood loss should be a question. We do not consider crossmatching to be necessary for thoracotomy. In my own experience, about four or five patients in 600 have required transfusion. Last, is there any evidence that this technique will reduce the compromise in pulmonary function that occurs after a thoracotomy? Dr. Kohno. We have not started a clinical trial. As soon as the instruments are approved by the Food and Drug Administration, we will start using the procedure in selected patients. We have not studied pulmonary function of the patients after lung resection (e.g., wedge resection) by thoracoscopy. We will need to do a randomized study in the future.