Reduction Pneumonoplasty for Patients With a Forced Expiratory Volume in 1 Second of 500 Milliliters or Less

Reduction Pneumonoplasty for Patients With a Forced Expiratory Volume in 1 Second of 500 Milliliters or Less John Eugene, MD, Amrit Dajee, MD, Raouf Kayaleh, MD, Harmohinder S. Gogia, MD, Clyde Dos Santos, MD, and Alan B. Gazzaniga, MD Departments of Surgery and Medicine and The U.S. Lung Center, Western Medical Center, Anaheim, and Departments of Surgery and Pulmonary Medicine, University of California, Irvine, California Background. Patients with severely impaired pulmonary function are considered at high risk for emphysema operations. We prospectively evaluated 44 patients with a forced expiratory volume in 1 second of 0.5 L or less undergoing reduction pneumonoplasty for dyspnea uncontrolled by medical management (confirmed by Borg and modified Medical Research Council dyspnea scales). Methods. There were 28 men and 16 women (mean age, 66 years) with a mean preoperative forced expiratory volume in 1 second of 0.41 L (range, 0.23 L to 0.50 L). Preoperative therapy consisted of bronchodilators (100% of patients), oxygen (80%), and steroids (72%). Hypercarbia was seen in 80% of patients, and 66% had pulmonary hypertension. Unilateral reduction pneumonoplasty by a video-assisted thoracic surgical approach was performed in 34 patients, 6 patients underwent bilateral reduction pneumonoplasty by a video-assisted thoracic surgical approach, and 4 patients underwent bilateral reduction pneumonoplasty by median sternotomy. Discrete emphysematous regions were resected using staplers with buttressing, and regions of homogeneous emphysema were plicated with KTP or neodymium:yttrium-aluminum garnet laser radiation. Results. There was one death within 30 days, two additional deaths within 60 days, and five additional deaths within 1 year. Hospital stay averaged 12 days. Intensive care unit stay averaged 4 days. Subjective improvement was noted by 89%. Borg and modified dyspnea scores improved from 7.6 to 4.5 (p < 0.01) and from 3.9 to 2.35 (p < 0.01), respectively. Forced expiratory volume in 1 second was 0.62 L at 1 year, a 51% improvement (p < 0.001). Forced vital capacity was 1.32 L preoperatively and 2.05 L at 1 year (a 56% improvement) (p < 0.001). Conclusions. This experience documents that patients with severely impaired lung function can successfully undergo operation for emphysema. To obtain these results one must tailor the operative approach to the patient s disease. (Ann Thorac Surg 1997;63:186 92) 1997 by The Society of Thoracic Surgeons The surgical treatment of emphysema produces a reduction in lung size by anatomic or functional removal of the most diseased areas. This allows the more normal and previously compressed remaining lung to reexpand. The normal elastic recoil properties of the reexpanded lung lead to improved intrapleural negative pressures and less bronchial collapse. In addition, reduction in lung size improves chest wall dynamics [1 4]. Inclusion criteria are lifestyle-limiting dyspnea, reduced pulmonary function (typically a forced expiratory volume in 1 second [FEV 1 ] 20% to 40% of predicted and residual volume greater than 250% of predicted), hyperexpansion, and diffuse bullous emphysema. Exclusion criteria include advanced age, hypercarbia, irreversible pulmonary hypertension, prior operation, thoracic deformities, significant comorbidity, poor patient compliance, and severely impaired pulmonary function. Our interest in Presented at the Thirty-second Annual Meeting of The Society of Thoracic Surgeons, Orlando, FL, Jan 29 31, 1996. Address reprint requests to Dr Eugene, 1107 S Anaheim Blvd, Anaheim, CA 92805. emphysema surgery led us to evaluate one of the most high risk groups for operation, those with markedly impaired pulmonary function. A protocol was developed to enable us to safely operate on these patients. This report presents an approach to the surgical treatment of emphysema in patients with an FEV 1 of 500 ml or less. Patients and Methods From January 1994 through June 1995, we prospectively studied 44 patients with an FEV 1 of 500 ml or less undergoing reduction pneumonoplasty at Western Medical Center, Anaheim, CA. This represents 29% of the emphysema patients we treated surgically during this time. There were 28 men and 16 women, aged 49 to 82 years (mean, 66 years). All patients were incapacitated by their dyspnea and were unable to work or to care for themselves, with 1 patient bedridden and 5 patients wheelchair-bound. Their symptoms were rated by the Borg category scale and the modified Medical Research Council dyspnea scale. All patients were receiving oral and aerosol bronchodilators; 80% of patients received 1997 by The Society of Thoracic Surgeons 0003-4975/97/$17.00 Published by Elsevier Science Inc PII S0003-4975(96)01014-4

Ann Thorac Surg EUGENE ET AL 1997;63:186 92 REDUCTION PNEUMONOPLASTY FOR FEV 1 OF 500 ML 187 continuous oxygen therapy, and the remaining 20% required intermittent oxygen therapy. Three patients had a permanent tracheostomy. Seventy-two percent of the patients required prednisone. Hypercarbia (carbon dioxide tension 45 mm Hg) was identified in 80% of these patients (carbon dioxide tension range, 36 mm Hg to 80 mm Hg; mean, 53 mm Hg). Pulmonary hypertension (pulmonary artery systolic pressure 40 mm Hg) was present in 66% of patients, and 1 patient had antiprotease deficiency. Preoperative pulmonary rehabilitation was not used in this series. Database entries for all patients included complete pulmonary function test results including spirometry, plethysmography, diffusion capacity, and arterial blood gases. Chest roentgenograms, posteroanterior and lateral, typically showed hyperinflation, and high-resolution computed axial tomography was performed in all patients to define the anatomy of their emphysema. Quantitative ventilation-perfusion lung scanning was used to identify the functioning areas of the lung and ventilation-perfusion mismatches. Using the information obtained from the highresolution computed tomographic scan and the ventilation-perfusion scan, we identified target areas of emphysema for resection or laser plication. If the patients had anatomic evidence of emphysema bilaterally with equal perfusion bilaterally, we performed a bilateral procedure. If there was no anatomic evidence of emphysema on one side or unequal perfusion between the lungs, we would choose to operate only on one lung. The side chosen was the side with the lowest percent of perfusion. Pulmonary artery (Swan-Ganz) catheters were used in all patients and were inserted the night before the operation. Operations were performed using a doublelumen tube, pulmonary artery (Swan-Ganz) catheter, and arterial line monitoring with continuous in-line blood gas measurements. Patients were positioned supine for bilateral pneumonoplasty via median sternotomy or in the lateral decubitus position for unilateral pneumonoplasty. When a bilateral procedure was performed by the thoracic approach, patients were positioned in a lateral decubitus position, then repositioned to the opposite lateral decubitus position once the first operation was completed. If there was a problem identified with the first operation such as a large air leak, we elected not to proceed with the second side but delayed the second operation to a later date. Intraoperatively, decisions were made on the basis of the anatomic findings and the location of the disease. Free beam KTP (532 nm) laser radiation (Laserscope, Inc) and/or free beam neodymium:yttrium-aluminum garnet (Nd:YAG) (1,064 nm) laser radiation (Laserscope, Inc) were directed at homogeneous emphysematous areas to plicate and contract these regions of the lung. Laser radiation was also used at the mediastinal and hilar areas of the lung because these areas are not readily accessible for resection. KTP laser radiation was delivered from 6 W to 10 W continuous wave power, and Nd:YAG laser radiation was delivered from 20 W to 30 W continuous wave power. When heterogeneous emphysematous areas of the lung were encountered or when large nonfunctioning areas of the lung required resection, this was performed using both endoscopic staplers and standard gastrointestinal staplers. Most staple lines were buttressed using bovine pericardium (Biovascular, Inc). The goal of the operation was to achieve a 20% to 30% reduction in the size of the lung. Postoperatively, patients were extubated within 24 hours. Chest tubes were maintained at 10 cm H 2 Oof negative pressure if there was an air leak, or underwater seal if there was no air leak. Patients were begun on physical therapy and activity on the first postoperative day. Patients whose lung position remained stable on routine chest roentgenograms were discharged with Heimlich valves. Postoperative pulmonary rehabilitation was undertaken by 40 patients. After discharge, patients were followed up by telephone and letter and postoperative examinations (when they lived locally) every 3 months. Pulmonary function tests were performed at 1 month, 3 months, 6 months, and 1-year intervals postoperatively. If a patient was poorly compliant, we visited the patient s home with a portable spirometer. Operative mortality was reported as death within 30 days of operation or during the initial hospitalization. Prolonged postoperative air leaks were reported as air leaks persisting greater than 7 days. Data are reported as mean standard error of the mean. Statistical comparison was performed between preoperative and postoperative findings using the t test and 2 test. Results There were 4 bilateral operations performed by median sternotomy, 6 bilateral operations performed by videoassisted thoracic surgery (VATS), and 34 unilateral procedures performed by VATS (including 2 planned bilateral procedures that were aborted). All median sternotomy procedures included the use of laser radiation to treat homogeneous emphysema and resection for areas of heterogeneous emphysema or large bullae. Among the thoracic procedures, nine were performed by VATS and laser radiation alone. The remainder of the operations consisted of laser radiation and surgical resection performed by VATS. Of the 44 patients, 32 were extubated in the operating room and another 11 patients were extubated with 24 hours. Only 1 patient remained on prolonged ventilation postoperatively. Intensive care unit stay averaged 4 days and total hospitalization averaged 12 days. One patient died 15 days postoperatively of sepsis secondary to a spontaneous colon perforation. This was considered an operative death. Perforated diverticulitis occurred in 2 patients postoperatively; they had multiple hospital readmissions and ultimately died of sepsis 40 days and 50 days postoperatively. A fourth patient had an acute myocardial infarction 46 days postoperatively and died. A fifth patient died at 90 days of a ruptured abdominal aortic aneurysm. An additional three deaths occurred within the next 9 months because of pneumonia (2 patients) and pulmonary embolism (1 patient). An additional 4 patients died

188 EUGENE ET AL Ann Thorac Surg REDUCTION PNEUMONOPLASTY FOR FEV 1 OF 500 ML 1997;63:186 92 Table 1. Comparison of Preoperative and Postoperative Pulmonary Function Tests Test Preop (44 patients) 1 Month (43 patients) 3 Months (40 patients) 6 Months (38 patients) 12 Months (18 patients) FVC (L) 1.31 0.06 1.7 0.1 1.83 0.13 1.96 0.15 2.05 0.12 (36% of predicted) (48% of predicted) (50% of predicted) (54% of predicted) (56% of predicted) FEV 1 (L) 0.41 0.01 0.57 0.02 0.63 0.04 0.62 0.03 0.62 0.05 (15% of predicted) (21% of predicted) (23% of predicted) (23% of predicted) (23% of predicted) RV 301% 9.2% 211% 5.2%......... TLC 141% 3.1% 108% 5.4%......... FEV 1 forced expiratory volume in 1 second; FVC forced vital capacity; RV residual volume; TLC total lung capacity. of respiratory failure during the second year; 32 patients are still being followed up. Prolonged postoperative air leaks occurred in 36% of patients. Gastrointestinal complications were seen in 18% of cases, and these included ileus (2 cases), pseudomembranous enterocolitis, lower gastrointestinal bleeding, and gangrenous cholecystitis in addition to the colon perforation and diverticulitis (2 cases) already mentioned. There were 38 patients still alive at 6-month follow-up, and 89% (34/38) of patients reported subjective improvement in dyspnea. Borg scores improved from 7.6 to 4.65 (p 0.01). Modified Medical Research Council dyspnea scale improved from 3.9 to 2.35 (p 0.01). The mean postoperative oxygen tension was 60 2 mm Hg, a significant improvement (p 0.01) from the preoperative mean oxygen tension of 49 2 mm Hg, and supplemental oxygen was required by 45% (17/38) of the patients compared with 80% preoperatively (p 0.01). The mean postoperative carbon dioxide tension was 46 1mmHg (p 0.01 compared with the mean preoperative value of 53 2 mm Hg), and 30% (10/38) of the patients had hypercarbia. Prednisone was necessary in 42% (16/38) of the patients compared with 72% steroid use before the operation (p 0.01). The average reduction in lung size was 25% as determined by comparing the preoperative and postoperative chest films of all patients [5]. A comparison of preoperative and postoperative pulmonary function tests is shown in Table 1. The FEV 1 improved from 0.41 L (15% of predicted) preoperatively to 0.57 L (21% of predicted) at 1 month, 0.63 L (23% of predicted) at 3 months, 0.62 L at 6 months, and 0.62 L (23% of predicted) at 1 year. This represents a 51% improvement in FEV 1 (p 0.001). Similarly, forced vital capacity improved from 1.32 L (36% of predicted) preoperatively to 2.04 L 1 year postoperatively, a 55% improvement (p 0.001). Residual volume decreased by 30% (p 0.01), and total lung capacity decreased by 23% (p 0.01) within the first year of follow-up. The pulmonary function tests of the 6-month survivors were also evaluated according to the type of operation performed. Unilateral reduction pneumonoplasty (resection plus laser) showed a 45% improvement in FEV 1 (0.42 L preoperatively and 0.61 L postoperatively). Bilateral reduction pneumonoplasty (resection plus laser) showed an 82% improvement in FEV 1 (0.38 L preoperatively and 0.69 L postoperatively). The operations that were performed by laser radiation alone (one bilateral, eight unilateral) resulted in a 40% improvement in FEV 1 (0.42 L preoperatively and 0.58 L postoperatively). Comment Patients with severely impaired pulmonary function represent one of the highest risk groups for thoracic operations. Standard criteria for operation exclude any patient with an FEV 1 less than 800 ml for a pulmonary resection [6]. Studies have shown an average reduction in FEV 1 of 600 ml after thoracotomy, and many surgeons believe that a patient must not be left with a postoperative FEV 1 less than 800 ml [7, 8]. In the setting of an emphysema operation, a low FEV 1 may be less hazardous than for other types of lung operations, but these patients still represent a very high risk group. Long-term studies of medical management of emphysema have shown that patients with an FEV 1 less than 750 ml have only a 70% 1-year survival and a 25% 5-year survival [9, 10]. When we began to perform operations for emphysema, we initially chose 500 ml as the cut-off level of acceptable pulmonary function. Early success in our program led us to modify this criterion as an absolute contraindication until we had studied the outcomes. The results reported herein show a dramatic response in both subjective and objective outcomes by employing a variety of operative techniques. The operative techniques used have all been previously described [1 4]. Wakabayashi [1] used thoracoscopy and a contact Nd:YAG laser probe to plicate diffuse emphysema and thoracoscopic resection to treat bullae. Little and associates [2] used free-beam Nd:YAG laser radiation via thoracoscopy to plicate diffuse bullous emphysema. Our group [3] used a VATS approach with both Nd:YAG and KTP laser plication of homogeneous emphysema and surgical resection of bullae. Cooper and colleagues [4] performed bilateral stapled resections via median sternotomy to treat heterogeneous emphysema. These techniques all have the common goal of lung reduction to improve compliance, improve airway patency, and allow improved diaphragmatic and chest wall motion. In approaching the very high risk patient, we attempted to individualize the operation for each patient, drawing from all of the surgical techniques currently

Ann Thorac Surg EUGENE ET AL 1997;63:186 92 REDUCTION PNEUMONOPLASTY FOR FEV 1 OF 500 ML 189 available. The preoperative workup was used to determine the surgical approach for each patient. If the highresolution computed tomographic scan showed emphysema bilaterally and there was no prior history of thoracic operation, a bilateral approach was considered. If the high-resolution computed tomographic scan showed predominance of emphysema in one lung or evidence of severe scarring in a hemithorax, then a unilateral operation was considered. If the ventilationperfusion scan showed equal perfusion between the lungs, a bilateral procedure was considered, but a perfusion difference of greater than 10% was an indication for a unilateral procedure. For example, we would consider a patient for a bilateral operation if the perfusion was measured at 55% for one lung and 45% for the other, but would favor a unilateral operation once the ventilationperfusion scan showed 56% perfusion to one lung. Usually the normal split in lung perfusion is 55% for the right lung and 45% for the left lung. In the high-risk emphysema patient, however, respiratory physiology is not normal, and we used the nuclear scan to estimate the pulmonary function available during one-lung ventilation and chose 10% as our limit. If a bilateral operation was planned, the next decision was median sternotomy versus bilateral VATS approaches. Predominance of bilateral upper lobe emphysema as well as normal thoracic configuration favored median sternotomy. Contraindications included prior sternotomy, tracheostomy, severe kyphosis, high-dose steroid use, and predominant location of emphysema in the lower lobes. Intraoperatively, decisions regarding extent of resection and laser plication were made on the basis of anatomic findings and the response of emphysematous lungs to laser radiation. If firm adhesions were encountered or if an extensive resection was planned, we would convert from thoracoscopy to video-assisted thoracotomy. If there was a good response to laser radiation with obvious contracture and shrinking of lung parenchyma, we would use laser radiation alone and not add resection to the operation. Resections were performed using staplers to remove strips of lung along the edges of each lobe. Laser radiation was used after resections to scarify the visceral pleura and to plicate areas not accessible to staplers such as the hilar region of the lung. Lung tissue normal in color usually responded well to KTP laser radiation, and anthracotic lung tissue responded to Nd: YAG laser radiation. Laser radiation can be used diagnostically as well as therapeutically in emphysema patients. Functioning areas of the lung have normal blood vessel architecture, and blood flow provides cooling for the tissues, which obviates the thermal effects of phototherapy. Emphysematous areas of the lung, however, have a paucity of blood vessels. These regions are exquisitely sensitive to the thermal effects of laser radiation and respond by contracting and shrinking to cause plication of the entire emphysematous section of the lung. Little and associates [2] used free-beam Nd:YAG laser radiation to treat emphysema via VATS. With a unilateral approach they reported an 18% improvement in FEV 1, subjective improvement in 80% of patients, and an operative mortality of 5.5%. Severely impaired pulmonary function is one of their exclusion criteria for operation. Wakabayashi [1] has reported using contact probe Nd: YAG laser radiation and bullous resection to treat emphysema via unilateral VATS. He reports an improvement in FEV 1 of 62% in patients with less than 14% of predicted and an improvement in FEV 1 of 28% in patients with more than 15% of predicted. Our prior work [11] with contact laser techniques has shown perforation, tearing, and disruption of tissue at the contact surface. This explains why some groups have obtained unsatisfactory results using a contact laser technique and why we, therefore, recommended free-beam laser radiation. Wakabayashi has a great deal of experience with his technique, and it appears that he may have found a way to overcome the deficiencies of the contact laser probe. We still believe, however, that the free-beam technique holds several advantages over contact laser use in terms of ease of use, directional stability, and improved laser tissue interaction. Cooper and colleagues [4] do not use phototherapy in the treatment of diffuse bullous emphysema. They report an improvement in FEV 1 of 82% using bilateral resectional techniques alone. The basis for their study is the work of Brantigan and associates [12], who theorized that most of the diffuse bullous changes occurred in a subpleural location and that these emphysematous areas could be eliminated by multiple small wedge resections performed along the pleural surface. If this diffuse emphysema is indeed found at the periphery of the lung, then it should be easier to treat this diffuse disease with laser radiation than with localized resection. That is why we have chosen to use laser radiation and reserve resection for areas of heterogeneous emphysema and large bullae. The present series shows that we can operate on patients with an FEV 1 of 500 ml or less, but does this necessarily mean that we should operate on them? Several patients died within a few months after the operation, and although most of these deaths were due to nonpulmonary causes, this early mortality rate shows that this is a group of patients at high risk for other fatal illnesses. Several other patients died within the first year, giving a 1-year survival thus far of 36/44 (82%), with not all of the patients yet followed up at 1 year. Wakabayashi [1] similarly showed that there was a high early mortality rate in the patients who had severely impaired pulmonary function. The patients who survived had a greatly improved quality of life, but there has to be a better way to select the patients so that we can improve on the 1-year survival. Our patients are currently evaluated preoperatively by gastroenterologists as well as pulmonologists, cardiologists, and surgeons in an attempt to avoid gastrointestinal complications. We have also begun to use preoperative pulmonary rehabilitation in our highrisk patients, as recommended by Cooper and colleagues [4], even though our bed-ridden patients and wheelchairbound patients have had good long-term results in this series. The data presented herein show that a very high risk group of patients can undergo emphysema operation

190 EUGENE ET AL Ann Thorac Surg REDUCTION PNEUMONOPLASTY FOR FEV 1 OF 500 ML 1997;63:186 92 Fig 1. Survival of patients with a forced expiratory volume in 1 second of 500 milliliters or less undergoing reduction pneumoplasty. if the operative approach is carefully tailored to the patient. With continued refinement in our program, it is possible that the surgical treatment of these severely impaired patients may become routine. Addendum. Since the submission of this article, all patients have been followed up for at least 1 year. The 1-year survival is 82%, and the 2-year survival is 73%. Figure 1 shows the the survival curve for the patient population. References 1. Wakabayashi A. Thoracoscopic laser pneumoplasty in the treatment of diffuse bullous emphysema. Ann Thorac Surg 1995;60:936 42. 2. Little AG, Swain JA, Nino JJ, Prabhu RD, Schlacter MD, Barcia TC. Reduction pneumonoplasty for emphysema early results. Ann Surg 1995;222:365 75. 3. Eugene J, Ott RA, Gogia HS, Dos Santos C, Zeit R, Kayaleh RA. Video-thoracic surgery for treatment of end-stage bullous emphysema and chronic obstructive pulmonary disease. Am Surg 1995;61:934 6. 4. Cooper JD, Trulock EP, Triantafillou AN, et al. Bilateral pneumectomy (volume reduction) for chronic obstructive pulmonary disease. J Thorac Cardiovasc Surg 1995;109: 106 19. 5. Harris TR, Pratt PC, Kilburn KH. Total lung capacity measured by roentgenograms. Am J Med 1971;50:106 19. 6. Gass GD, Olsen GN. Preoperative pulmonary function testing to predict postoperative morbidity and mortality. Chest 1986;89:127 35. 7. Hallfeldt KKJ, Siebeck M, Thetter O, Schweiberer L. The effect of thoracic surgery on pulmonary function. Am J Crit Care 1995;4:352 4. 8. Miller JI. Physiologic evaluation of pulmonary function in the candidate for lung resection. J Thorac Cardiovasc Surg 1993;105:347 52. 9. Burrows B, Earle RH. Course and prognosis of chronic obstructive lung disease. N Engl J Med 1969;280:397 404. 10. Diener CV, Burrows B. Further observations on the course and prognosis of chronic obstructive lung disease. Am Rev Respir Dis 1975;111:719 24. 11. Baribeau Y, Eugene J, Firestein SL, Hammer-Wilson M, Berns MW. Comparison of contact and free-beam laser endarterectomy. J Surg Res 1990;48:127 33. 12. Brantigan GE, Muellar E, Krest MD. A surgical approach to bullous emphysema. Am Rev Respir Dis 1959;79:194 206. DISCUSSION DR ALEX G. LITTLE (Las Vegas, NV): I think this is another one of the recent healthy signs that we are moving beyond the gee-whiz stage when it comes to lung reduction surgery, where both the presenter and the audience are mildly amazed at what had been pulled off. I think we are moving now into discussion of solid issues, which is very important. One issue raised by this report is the indications, or maybe better, contraindications for this approach. Doctor Eugene, I think you have certainly helped me develop at least some reassurance toward operating on the people at the low end, but it would be more helpful if you had your data, at least with FEV 1, in terms of percent of predicted rather than an absolute number of 500 ml. It is a little bit similar to looking at left ventricular function by the number of milliliters pumped out with each beat rather than the ejection fraction. I notice your range of FEV 1 is from 10% to 26% of predicted. So perhaps it would be helpful to think in terms of percent of predicted rather than just the absolute number. The other area in which I think you have focused is that of operative technique, and personally I think it is a very reasonable suggestion that we should fit operative choices to the disease and to the patient. This is certainly the way in which we are going to learn what is effective and what is not, and also, as the principle we apply to operations elsewhere, it makes a lot of sense. I have just a few questions about some of your statements.

Ann Thorac Surg EUGENE ET AL 1997;63:186 92 REDUCTION PNEUMONOPLASTY FOR FEV 1 OF 500 ML 191 First, if I understood correctly, when the disease that you are going to resect is principally in the upper lobe, that encourages you toward either a thoracotomy or a sternotomy. My own experience is that the upper lobe is just as accessible as the lower lobe with a thoracoscope. I wish you would clarify that a little bit. In the ongoing discussions of the role of laser contraction versus resection, our own group has found that using the laser certainly achieves some lung reduction, but resection, obviously, is unlimited. On the other hand, the lased areas seem to have much less in the way of air leak than the staple line areas. I wonder if you would comment on that as a means of thinking about the laser technique, as we now do, as complementary to resection. Finally, I question whether there is ever really a patient who should be operated on only unilaterally when both lungs are accessible, that is, when the pleural space has not been obliterated by a previous operation or pleural sclerosis. Our experience, which has evolved from a unilateral to a bilateral approach, is that even when the disease is predominantly on one side, both lungs are involved. Moving to a bilateral approach has both improved results and actually reduced hospital morbidity. So I would appreciate you clarifying that issue. I thought this was a very nicely presented report with important data. DR EUGENE: Thank you, Dr Little. Regarding the percent of predicted value of FEV 1 versus the absolute number of FEV 1, surgeons usually prefer the absolute number. In surgical literature and surgical teachings, it is commonly regarded that most patients should not be left after a lung operation with an FEV 1 less than 800 ml. There are studies that show that thoracotomy alone will leave a patient with an FEV 1 reduced by up to 600 ml, and many surgeons will not operate on a patient with an FEV 1 less than 800 ml. Therefore, in the course of this study I decided to choose patients on the basis of the absolute FEV 1, rather than the percent of predicted value of FEV 1. Regarding the surgical approach, I usually choose the sternotomy approach if the patients have predominantly upper lobe disease with nearly equal ventilation-perfusion bilaterally. If the patients have predominantly lower lobe disease with equal perfusion bilaterally, I prefer a bilateral thoracic approach. I also agree with you that from a thoracoscopic point of view, it does not make any difference where the disease is located unless the patient has significant adhesions. Regarding the occurrence of air leaks with free beam laser radiation versus stapling, I can comment on our findings at reoperation. When we have had to reoperate on patients because of prolonged persistent air leaks, we usually find that the air leak comes from just beneath the staple line. Even if the staple line is buttressed, the leak occurs underneath that line where the lung has reexpanded and torn. We now try to reinforce the staple lines and the buttressing material with interrupted sutures. Regarding bilateral pneumonoplasty versus unilateral pneumonoplasty, I agree with published studies that show better improvement in pulmonary function tests with bilateral versus unilateral pneumonoplasty. I am reluctant, however, in a patient with a pulmonary artery pressure of 70/40 mm Hg to collapse the good lung and operate on that good lung. I was trying to minimize the catastrophic results that can occur when operating on the good lung in such a very high risk patient. Unless there is equal perfusion bilaterally, I would only perform a unilateral procedure in these very high risk patients. They did well, however, in terms of their overall improvement, both subjectively and objectively. DR RODNEY J. LANDRENEAU (Pittsburgh, PA): I also enjoyed the presentation by Dr Eugene. We had an opportunity to talk a little bit before the presentation to give him some ammunition to counter my comments. I would like to say first that I have a disagreement with your definition of elastic recoil. I think what you are trying to describe, at least by my interpretation, is a reduction in the dynamic collapse of the airways associated with the emphysematous process achieved through a lung volume reduction operation. Elastic recoil really relates to the inherent ability of the lung to expel the air within it during exhalation, the springy characteristic of the lungs. Emphysema is a disease characterized by the inability to remove air from the lung s terminal bronchioles resulting in its overexpansion. I think this should be clarified in your article. Tomorrow morning there will be a breakfast session describing the present role of VATS in general thoracic surgical practice. We polled the General Thoracic Surgical Club membership to get a feel of their opinion regarding the role of VATS for lung volume reduction. Forty-plus percent thought that it was an acceptable or preferred approach to the management of diffuse pulmonary emphysema. More than half of the surgeons thought that it was an unacceptable or investigational operation. Unfortunately, we did not ask if the surgeons thought lung volume reduction was acceptable or investigational performed through a sternotomy approach. It is important that we do not equate VATS with laser surgery. It is my belief that a resectional approach rather than a laser ablative approach is the preferred primary management of the bullous process. Although laser ablation can augment the lung volume reduction in areas of the lung where stapling cannot be safely applied, the primary effectiveness of the laser in lung volume reduction remains to be clearly substantiated. It has been consistently shown in studies by Wakabayashi, Cooper, McKenna, and our group in Pittsburgh that bilateral lung volume reduction performed via VATS or sternotomy is important for the severely impaired patient with bullous emphysema. Accordingly, I am having difficulty in accepting your formula of a 10% difference in regional perfusion as a discriminator for unilateral versus bilateral lung volume reduction. We know that anatomically the right lung has approximately 10% more blood flow than the left lung. I wonder if there is a better way of determining if a unilateral or a bilateral procedure is indicated. Along these lines, are you doing more bilateral lung volume reduction procedures at this time? Were the majority of your unilateral VATS lung volume reduction surgical procedures done in your early experience with this technique, or do you still use a unilateral approach in selected patients? Although my following inquiries are numerous and potentially laborious to answer, I would appreciate further clarification of the number of patients who were reintubated after operation and the number of patients who were transferred to chronic institutions after the operation rather than directly to home. I would also like an objective analysis of the preoperative and postoperative assessment of functionality among patients you have operated on, ie, the 6-minute walk analysis. I would like to know about the possible occurrence of empyema or pneumonia among your patients. I would also like to know how long it took to perform the laser procedure compared with the purely resective approach. And finally, how many patients undergoing the unilateral approach ultimately underwent a contralateral lung volume reduction operation? DR EUGENE: Thank you, Dr Landreneau. I do not disagree with your comments about elastic recoil, and I accept your interpre-

192 EUGENE ET AL Ann Thorac Surg REDUCTION PNEUMONOPLASTY FOR FEV 1 OF 500 ML 1997;63:186 92 tation. Regarding the use of VATS for the treatment of emphysema, I do not equate this operation with laser. I believe that laser plication is an important part of my procedure, and in some instances the entire procedure can be performed with laser radiation alone. However, surgical resection using staplers is also an important part of the operation, and this can be performed by VATS as well as the laser plication. Regarding the choice of a unilateral versus bilateral operation, I chose 10% as the difference in the ventilation-perfusion scan between the lungs as the arbitrary number I would accept for a bilateral approach when I began this study. I think we have shown at this time that we can achieve good results in the very high risk patient, and therefore I am willing to relax on my 10% standard and try to adopt a more liberal number when deciding on bilateral pneumonoplasty. Regarding complications in this series, 1 patient was reintubated and there were no empyemas. All patients were referred to pulmonary rehabilitation, either inpatient or outpatient, after their hospital discharge, and the 1 patient who returned with gangrenous cholecystitis and subsequently died several weeks later of a myocardial infarction was referred to a chronic institution. The 6-minute walk was determined in the majority of patients, but 1 patient was bed-ridden, and 5 patients were wheelchair-bound preoperatively, so I thought we would be skewing the data to report the preoperative and postoperative 6-minute walk in this group of patients. DR AKIO WAKABAYASHI (Irvine, CA): I congratulate Dr Eugene for an excellent report and presentation. I had an opportunity to read the paper just before the meeting. I concur with Dr Eugene and associates that low FEV 1 is not a risk factor. In our analysis of our own series we found that there is no statistically significant correlation between FEV 1 and 3-month mortality rate. Doctor Eugene and associates reported also a very high incidence of gastrointestinal complications. We also noticed that that is the number two complication next to respiratory complications. The second patient in my series had right-sided free-beam CO 2 laser bullectomy in September 1989, and his FEV 1 increased from 15% to 60% 1 year later. I had the opportunity to see this patient 6 years later, and to my pleasant surprise, the lung that I operated on remained unchanged, in good shape, but on the other side the bullae got quite larger, causing recurrence of dyspnea, but still an FEV 1 of 30%. So this assured me that the free-beam laser has a very long-lasting effect on the bullous disease. However, I had an uncontrollably high rate of perforation in which I could not see the holes right after free-beam CO 2 laser ablation of the bullae. Therefore I switched to a Contact YAG Laser technique, and now these days I never see perforation by the Contact YAG Laser. Always the leak is along the staple line, as I mentioned. I have two questions. First, how many patients went home with a Heimlich valve? Second, do you measure pulmonary artery pressure in all patients? DR EUGENE: Thank you, Dr Wakabayashi. All patients have a Swan-Ganz catheter as part of their operative management. Eight patients went home with a Heimlich valve. DR JOEL D. COOPER (St. Louis, MO): I enjoyed the presentation. I think it is of important historical significance because I think it marks the transition from an older technique, namely, the use of laser, to perhaps more efficacious techniques. Prompted by your abstract I went back and looked at the patients we have operated on whose FEV 1 was less than 500 ml. We had 52 patients whose FEV 1 was less than 500 ml. You lumped everything together. That is the bad part of this report. You cannot draw any conclusions because there are three or four different techniques and several different surgical approaches, and your results combine bilateral and unilateral procedures. Our experience with 52 similar patients includes 40 patients with complete follow-up at 6 months or more. There was one early death, and one late death, and we saw double the improvement in FEV 1 that you reported in this particular series. The total 1-year actuarial mortality rate was 4%. I have several questions. How did your unilateral results compare with your bilateral results, so that we can reanalyze your data? Number two, in how many cases did you not use a stapler, relying only on the laser? You have given us a 1-year mortality of 18%. Doctor McKenna has shown me his data comparing a unilateral versus a bilateral procedure, and clearly he has demonstrated a similar 1-year mortality as yourself with the unilateral procedure compared with double the improvement, without increased morbidity and with no added mortality out to 1 year, when the bilateral procedure is done. That prompts my last question. Why would you propose a lesser procedure, as you have outlined, when another procedure in the same group of patients can produce double the improvement, far greater improvement in the modified Medical Research Council dyspnea index, and far greater improvement in oxygenation, without increased morbidity or mortality in the early phase and probably less morbidity and less mortality at 1 year? DR EUGENE: Thank you, Dr Cooper. In answer to your questions, the patients who underwent bilateral reduction pneumonoplasty had an 82% improvement in their FEV 1, and the patients who underwent unilateral reduction pneumonoplasty had a 45% improvement in their FEV 1. For the overall series, the improvement was 51%. There were 9 cases performed as pure laser plication; that is, using no resectional techniques. In these patients, the improvement in the FEV 1 was 40%. In regard to the selection of unilateral versus bilateral operations, we performed a bilateral procedure if the patients had anatomic evidence of emphysema bilaterally with no greater than a 10% difference in perfusion between the two lungs. If the emphysema was predominantly located in the upper lobes and the patients had a normal thoracic configuration, we favored median sternotomy. Contraindications to median sternotomy included prior sternotomy, tracheostomy, severe kyphosis, highdose steroid use, and predominant location of emphysema in the lower lobes. We thought that the patients in this series represented an extremely high risk. We thought that the patients would sustain greater intraoperative trauma with a bilateral approach rather than a unilateral approach and therefore we were very cautious in the extent of the operation we performed on these patients. Because we have had good results with this series, we will loosen our indications for bilateral versus unilateral procedures and probably change our criterion from a 10% difference in ventilation-perfusion scan between the two lungs to a 20% difference in ventilation-perfusion scan between the two lungs.