R replacement that uses a sequential implantation. Bilateral Sequential Lung Transplantation: The Procedure of Choice for Double-Lung Replacement

Transcription

1 Bilateral Sequential Lung Transplantation: The Procedure of Choice for Double-Lung Replacement Larry R. Kaiser, MD, Michael K. Pasque, MD, Elbert P. Trulock, MD, Donald E. Low, MD, Carolyn M. Dresler, MD, and Joel D. Cooper, MD Division of Cardiothoracic Surgery, Department of Surgery, and Division of Pulmonary Medicine and Critical Care, Department of Medicine, ashington University School of Medicine, St. Louis, Missouri e recently described a technique for bilateral sequential lung transplantation that replaces the en bloc doublelung operation, a procedure that was accompanied by frequent problems with airway healing. Twenty-seven patients have undergone 28 bilateral sequential lung transplantations over the past 14 months. Eighteen patients had transplantation because of end-stage emphysema; 6, cystic fibrosis; and 1 each, obliterative bronchiolitis, usual interstitial pneumonitis with pulmonary fibrosis, and bronchiectasis. Cardiopulmonary bypass was used electively in the first 5 patients until it was recognized that the procedure could be done safely without it, and in only 3 additional recipients has it been employed. Mean ischemic time for the first lung was 276 k 43 minutes and for the second lung, minutes. There have been five deaths, three in the postoperative period (11% operative mortality) and two late. The other patients are alive and well and do not require oxygen 2 to 15 months after transplantation. Mean forced expiratory volume in 1 second rose from 16% 2 8% of predicted to 84% 2 17% at 12 weeks. Six-minute walk values increased from a mean of 251 k 91 m to m at 24 weeks. The excellent exposure afforded to both hemithoraces by the thoracosternotomy incision and the rare need of cardiopulmonary bypass have allowed us to offer the option of transplantation to patients who formerly would have been turned down because of previous pulmonary resection or pleurectomy. On four occasions, ventilator-dependent patients underwent successful transplantation. The applicability of the procedure seemingly will be limited only by donor considerations and questions regarding how much better one lung is than two in patients other than those with chronic infection. Bilateral sequential lung transplantation can be performed with a minimum of early mortality and morbidity, which is comparable with single-lung transplantation. (Ann Thorac Surg 1991;52:43846) ecently we [l] described a method for bilateral lung R replacement that uses a sequential implantation technique through a transverse thoracosternotomy incision. As we have gained more experience with this procedure, its broad applicability and substantial advantages over previous techniques for double-lung transplantation have become apparent. Many patients who previously would not have been considered candidates for double-lung transplantation because of technical factors stand to benefit from this procedure. Bilateral lung replacement is mandated in situations where infection is present, such as cystic fibrosis or bronchiectasis. hether bilateral lung replacement provides a distinct advantage to patients with end-stage emphysema remains to be definitively determined, but it is our current practice to offer it to the younger patients (less than age 5 years) and to those with a large bullous component to their disease. This age limit, which is by no Presented at the Twenty-seventh Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Feb lb2, Address reprint requests to Dr Kaiser, Division of Cardiothoracic Surgery, Hospital of the University of Pennsylvania, 4 Silverstein, 34 Spruce St, Philadelphia, PA means absolute, stems in part from the higher incidence of serious coronary artery disease found in older individuals, which would be expected to contribute to perioperative problems, especially if cardiopulmonary bypass (CPB) were to be required. Any advantage conferred by bilateral lung transplantation for patients with other than infected end-stage processes depends on a comparative analysis of morbidity and mortality of single-lung and bilateral lung replacement measured against differences in overall function between the two groups over a long-term follow-up period. In this report we detail our initial experience with bilateral sequential lung transplantation in 27 patients. Material and Methods Patients Between October 1989 and January 1991, 27 patients underwent 28 bilateral sequential lung replacements at Barnes Hospital, St. Louis, MO. The mean age of the patients was 41 years (range, 18 to 54 years). Seventeen were men and 1 were women. Eight patients had endstage emphysema of unknown etiology, 1 patients had a-antitrypsin deficiency, 6 patients had cystic fibrosis, and 1991 by The Society of Thoracic Surgeons /91/$3.5

2 Ann Thorac Surg 1991;52; KAISER ET AL patient each had obliterative bronchiolitis, usual interstitial pneumonitis with pulmonary fibrosis, and bronchiectasis. Recipients waited an average of 16? 75 days (range, 2 to 262 days) for donor lungs to become available. Selection Criteria To be considered for lung transplantation, patients must demonstrate evidence of progression of disease by decreased exercise tolerance or deteriorating pulmonary function studies, increasing oxygen requirements, increasing complications, or a combination of these factors such that survival beyond 12 to 18 months would not be expected. Patients undergo a structured inpatient evaluation before final acceptance into the program. Approximately 1% of patients referred to our program are actually accepted for transplantation [2]. Reasons for refusal include advanced age, poor cardiac function or coronary artery disease or both, or psychosocial difficulties. Although, as in the past, patients were routinely turned down because of previous thoracic procedures, which might increase technical problems relating to the transplantation, the development of this operation has rendered these problems exceedingly rare contraindications. All of the patients demonstrated marked clinical deterioration with increasing oxygen requirements over the several months before referral. Patients with other systemic disease or with major renal or hepatic disease are not accepted. Absence of serious coronary artery disease is documented in each patient by coronary arteriography, and right ventricular function is assessed by radionuclide scan. Nutritional status has to be adequate (weight within 1% to 2% of ideal body weight), and patients are required to participate in a supervised outpatient pulmonary rehabilitation program. Patients accepted for transplantation are required to move to St. Louis and are followed up as outpatients on a regular basis. The rehabilitation program for pretransplantation patients consists of stationary bicycling, treadmill walking, or both with continuous transcutaneous oximetry. Psychological and psychiatric evaluation of all potential candidates is performed to rule out a major pathological process or drug abuse problems. Patients are required to have a support person accompany them and must demonstrate significant motivation for transplantation and the rigors of lifelong follow-up care. Pulmonary Function Tests Pulmonary function studies were carried out using a Gould-FRL Medical 1 IV pulmonary function system. The system measures both inspiratory and expiratory flow directly and electronically integrates the flow signals to obtain volume measurements. A full set of pulmonary function tests including diffusing capacity was measured at the time of evaluation and at 3, 6, 12, 24, and 52 weeks after transplantation. Perfusion lung scans in the immediate postoperative period were performed with the patient in the supine position after the intravenous administration of 1.5 mci of technetium 99 m-labeled microaggregated albumen. A portable gamma camera was used, and blood flow to each lung was calculated and reported as a percentage of total blood flow. Exercise Testing and Evaluation As part of the pretransplantation evaluation, functional ability is assessed with a 6-minute walk test and treadmillwalking protocol. For the 6-minute walk, patients walk at their own pace around a level, 15 m (5-foot) circuit. They can stop as often as necessary. During testing before transplantation, patients use as much supplemental oxygen as they require to minimize desaturation as determined by transcutaneous oximetry. Total distance traveled during the 6 minutes is recorded, as are heart rate and the number of stops required. Treadmill walking is also assessed. Initially this was done for 3 minutes while the patient walked on a treadmill set at 1.6 km (1 mile) per hour with a 4% grade. Subsequently we have used more standardized exercisetesting protocols and have chosen the low performance protocol (LOP) as being most appropriate for this group of patients [3]. This protocol, if performed in its entirety, requires 2 minutes and consists of ten progressive stages. The speed begins at 1.76 km (1.1 mile) per hour at a 1% elevation and progresses to 7.2 km (4.5 miles) per hour at a 1% elevation. Metabolic equivalents are quantitated based on the stage completed and recorded along with the total time. Patients were tested at the time of their initial evaluation and periodically up until the time of transplantation. This allows for an assessment of the progress made during the rehabilitation phase before transplantation. After transplantation, patients continue a pulmonary rehabilitation program and are reassessed periodically with the 6-minute walk and treadmill protocol. Surgical Technique A detailed description of the technical aspects of the operation has been previously published [l]. Briefly, with the patient supine and the arms secured overhead to an ether screen, bilateral anterolateral thoracotomies are made accompanied by a transverse sternal division. The patient is maintained on one-lung ventilation while the contralateral hemithorax is opened. e purposely do not immediately open the pleural space on the side being ventilated, as this seems to allow for better ventilation and oxygenation, especially in patients with emphysema. This does require that the thoracotomy and hilar mobilization be completed after the first lung has been replaced. Despite the fact that this prolongs the ischemic time of the second lung, the advantage that it provides in ventilating the patient seems to outweigh any potential advantage of a slightly shorter ischemic time. Occasionally we have used high-frequency jet ventilation during the initial phase of the procedure. e continue to use an omental pedicle to wrap each bronchial anastomosis. Through an upper midline incision, the gastrocolic omentum is mobilized off the transverse colon and split into two pedicles. The pedicles are tunneled substernally into each pleural space. e try whenever possible to avoid the use of CPB, usually basing

3 44 KAlSER ET AL BILATERAL LUNG TRANSPLANTATlON Ann Thorac Surg 1991;52;43846 the decision on our ability to adequately oxygenate the patient and by observing changes in pulmonary artery pressure that occur with a trial period of unilateral pulmonary artery clamping. The use of intraoperative transesophageal echocardiography has greatly contributed to the operative management by allowing continuous assessment of right ventricular function, which, if noted to deteriorate with pulmonary artery clamping, is an indication for instituting CPB. If CPB is required for replacement of either lung, a single, two-stage right atrial cannula and a standard aortic cannula are placed, and partial bypass sufficient to unload the right ventricle is employed. The first lung is removed at the time the donor lungs arrive in the operating room. The lung block is split, and the first lung is implanted in the standard fashion by first completing the bronchial anastomosis, followed by the arterial and atrial anastomoses. After the first lung is reperfused and following a short period of equilibration, the opposite hemithorax is opened, the lung collapsed, and the patient ventilated solely on the newly implanted lung. The second recipient pneumonectomy is then performed with subsequent implantation of the second lung. The transverse sternotomy is closed with several heavygauge wires. The anterolateral thoracotomies are closed in a standard fashion using paracostal sutures. An indwelling epidural catheter, placed preoperatively, is used for continuous infusion of narcotic to minimize postoperative pain, hasten extubation, and maximize respiratory effort. Aggressive diuresis is begun immediately after the procedure, as patients commonly are in positive fluid balance. Immunosuppression is instituted with intravenous cyclosporine on the day of transplantation along with azathioprine and Minnesota antilymphocyte globulin. Suspected acute rejection episodes are treated with methylprednisolone by bolus for 3 consecutive days. Maintenance steroids are withheld until day 6 after transplantation. Patients usually leave the intensive care unit 1 day after extubation. Results Pa f ien t Outcome Cardiopulmonary bypass was used in 8 patients. The mean bypass time was 197 minutes (range, 92 to 365 minutes). For the initial five transplant procedures, we elected to use CPB after implantation of the first lung because of our concern of "flooding" the newly implanted lung. e have subsequently found that CPB rarely is necessary and now use it only when indicated by the clinical situation. In 21 patients, the right side was replaced first. The mean ischemic time for the first lung implanted was 276? 43 minutes (range, 18 to 36 minutes). The mean ischemic time for the second lung was 41 * 64 minutes (range, 3 to 565 minutes). Time between implantation of the first and second lungs averaged slightly more than 2 hours (134 minutes), much of this consumed by the performance of the second recipient pneumonectomy. Quantitative perfusion scan data obtained immediately after completion of the transplant operation failed to demonstrate any major predominance Fig I. Early postoperative chest radiograph after bilateral sequential lung transplantation. of flow to the lung with the shorter ischemic time (mean, 53% * 11%, range, 27% to 8%). Even when a substantial dichotomy in perfusion was present on the first scan, the perfusion tended to equalize within 48 hours. The early postoperative course for the majority of the recipients was remarkably uneventful. No patient required reexploration for bleeding. The typical appearance of an early postoperative chest radiograph is demonstrated in Figure 1. Patients were extubated at a mean of 3 * 2 days (range, 1 to 7 days) and had a mean stay in the intensive care unit of 4.1? 3 days (range, 1 to 14 days). Discharge from the hospital occurred at a mean of 24 * 8 days and ranged from 14 to 52 days. The number of days required to achieve a room air oxygen pressure greater than 6 mm Hg averaged 7 * 6 with a range of 2 to 29 days. The diagnosis of rejection depended mainly on clinical criteria, such as development of an infiltrate on chest radiograph, decreased oxygenation, or a slight temperature elevation. Transbronchial lung biopsies are not performed on a strict protocol basis in the early postoperative period but are done as indicated by the clinical situation, often when an infiltrate persists on the chest radiograph after treatment for rejection. Differentiating between infection and rejection remains difficult, but rejection occurs so commonly within the first 7 days after transplantation that we have a low threshold for administering steroids by bolus injection. If the chest radiograph does not improve within 12 to 24 hours after steroid administration, then bronchoalveolar lavage, protected brush specimens, and at times transbronchial lung biopsy are performed. In the early posttransplantation period, the diagnosis of rejection depends more on experience than any single diagnostic test. Transbronchial lung biopsy plays a major role in the diagnosis of rejection later in the course and is used frequently. Recipients in this series sustained a mean of rejection episodes (range, to 4) during the period of hospitalization.

4 Ann Thorac Surg 1991;52;43%46 KAISER ET AL 441 I2Or 1 - Fig 2. Change in percent of predicted forced expiratoy volume in 1 second. Data are shown t the standard deviation. (TX = transplantation.) 8 - c 2 oz a /I I n.19 I n= 14 1 n=4 " PRE TX EEKS There have been five deaths among the 27 patients. Three occurred in the postoperative period, an 11% operative mortality. Actuarial survival at 1 year is 85%. One patient died of a presumed cardiac arrythmia 2 weeks after transplantation as he was making an otherwise uneventful recovery. Two patients died of what was presumably donor lung dysfunction. In 1, there was diffuse necrosis of both airways evident within 5 days after transplantation. He subsequently underwent retransplantation 1 week after the initial operation but ultimately died of overwhelming sepsis and multipleorgan failure. There were two late deaths. One patient died of metastatic pancreatic carcinoma at 7 months, soon after presenting with multiple hepatic metastases. Exploratory laparotomy provided biopsy confirmation. The other died 1 year after transplantation, which was performed because of end-stage obliterative bronchiolitis, which developed after a previous en bloc double-lung transplantation 1.5 years earlier. Despite a change in her immunosuppression regimen, the patient again had changes consistent with obliterative bronchiolitis as well as lymphoma and pulmonary aspergillosis. The immediate cause of death was acute pancreatitis. All other patients are alive and well 2 months to 15 months after transplantation though morbidity from the procedure is not insubstantial. Eight patients sustained ten anastomotic defects or dehiscences. In only 2 patients were these of clinical significance, with 1 requiring two dilations of a bronchial stricture. Cytomegalovirus (CMV) pneumonitis continues to be a problem after lung transplantation, and 8 patients in this series had CMV disease proven by transbronchial lung biopsy. e attempt at transplantation to provide CMV-negative recipients with only CMV-negative lungs, whenever possible, in an effort to keep CMV pneumonitis to a minimum. e do not, at present, employ any routine antiviral prophylaxis against CMV, even in high-risk situations, as no data are available to support this approach. Six patients had culture-proven bacterial pneumonitis, a situation most commonly seen in recipients with cystic fibrosis. Intraabdominal complications have also occurred commonly. Two patients had major intraabdominal bleeds necessitating laparotomy. Both occurred secondary to mesenteric artery aneurysms in recipients with -antitrypsin deficiency, which may prove to be a contributing factor. One patient had clinical and radiographic evidence of angulation at the antrum and required laparotomy to divide the omental pedicle, which had caused this displacement of the stomach. Cardiac arrhythmias also were common, with supraventricular tachycardia noted in 6 patients. Other complications included pneumothorax (5), donor lung dysfunction (4), early sternal instability (3), wound infection (2), gastrointestinal bleeding (2), obliterative bronchiolitis (2), recurrent laryngeal nerve injury (l), and meningoencephalitis (1). Pulmonary Function Results Before transplantation, the forced expiratory volume in 1 second ranged from 8% of predicted to 51% (mean, 16% * 8%). This corresponded to a value of.54 *.21 L. By 3 weeks after transplantation, the mean percent of predicted forced expiratory volume in 1 second for the group rose to 68% * 18% (Fig 2). This rose to a peak of 84% +- 17% of predicted at 12 weeks and then seemed to plateau with a value of 82% * 2% of predicted at 24 weeks and 75% * 34% of predicted at 52 weeks for the 4 patients evaluable for that period. The change in forced vital capacity after transplantation was not quite as marked (Fig 3). It rose from a pretransplantation value of 44%? 17% of predicted to 75% * 14% of predicted at 24 weeks and showed a continued rise at 52 weeks. Functional Testing At the time of evaluation, patients walked a mean of 251 * 91 m in 6 minutes, which improved to 353 & 122 m at the final measurement before transplantation (Fig 4). This dropped to 323 & 11 m at 3 weeks after transplantation, but by 6 weeks, had increased to m and continued to increase at 12 weeks and 24 weeks to values

5 442 KAISER ET AL Ann Thorac Surg 1991;52;43846 Fig 3. Change in percent of predicted forced vital capacity. Data are shown? the standard deviation. (TX = transplantation.) 1-8- n 6-4- Iy: w 2- a,&-i n.21 n.23 / n.14 n.21 n.19 n=4 of 613 * 128 m and 666 * 42 m, respectively. All patients used supplemental oxygen for the pretransplantation exercise studies, whereas postoperative evaluations were performed without supplemental oxygen. The pattern for the 3-minute treadmill-walking evalu- ation was similar. At the time of evaluation, patients covered a mean of 1, m, which increased to 1,87? 417 m at the last measurement just before transplantation (Fig 5). At 3 weeks after operation, this dropped back almost to the evaluation level but increased to 2,149? 529 m by 6 weeks (range, 1,27 to 3,363 m). By 12 weeks, the mean had increased to 2,661 * 23 m. Using a standardized treadmill protocol (LOP) and quantitating metabolic equivalents, Figure 6 shows the same trend with an increase between the time of evaluation and pretransplantation and a subsequent marked increase by 6 weeks and further increase at 12 weeks (8.5 & 1.). By 12 weeks, the mean value of metabolic equivalents translates into stage 8 of the ten-stage protocol. Comment The development of the bilateral sequential lung transplant procedure was prompted by the recognition of Fig 4. Results of 6-minute walk test. Data are shown? the standard deviation. (EVAL = evaluation; TX *O r I = transplantation.) 6 I I I I I I PRE TX EEKS major complications occurring after the original en bloc double-lung operation. Not only were there problems with healing of the tracheal anastomosis, but complications after CPB were also noted. Several deaths early in our experience with the original operation were directly related to myocardial problems. Because of the recognition of difficulties with tracheal healing, we and others [4] initially modified the en bloc operation by performing bilateral bronchial anastomoses. This seemed to have an impact on the incidence of anastomotic problems, and it was thus a logical extension to proceed to a bilateral sequential replacement technique in an attempt to reproduce the high success rate that had already been achieved with single-lung transplantation, as the operation is essentially two single-lung transplantations performed in the same patient. The initial theoretical objection to this approach was the mandatory increase in ischemic time that would accrue before the second lung was in place and the uncertainty regarding the maximum safe cold ischemic time. e believed that it was optimal to keep the time of ischemia less than 6 hours. Based on the current experience, however, we have found that ischemic times up to and T n.17 n= 15 n.18 I I 1 I I I EVAL PRETX EEKS

6 Ann Thorac Surg 1991;52;43846 KAISER ET AL 443 Fig 5. Total distance walked after 3 minutes on the treadmill at 1.6 km (1 mile) per hour with a 4% grade. Data are shown -+ the standard deviation. (EVAL = evaluation; TX = transplantation.) v, 2 cz 15 z Iooo t c 5 I n=14 " n=15 n=ll EVAL PRE TX EEKS beyond 9 hours are well tolerated. In our laboratory [5], routine safe lung preservation for 24 hours has been achieved in a canine model. Ideally the ischemic times should be kept as short as possible, but the recognition that longer periods of ischemia are still compatible with excellent gas exchange in the majority of patients lends further support to the concept of sequential lung replacement. Although we observed four instances of what we characterized as donor lung dysfunction with a histological picture of diffuse alveolar damage, there was no correlation with ischemic time. Even though we are replacing both lungs, we have not routinely used CPB subsequent to our initial 5 patients but have chosen instead to support the patient on the newly implanted lung as we sew in the second lung. e now recognize that despite an increase in permeability known to occur after ischemic injury [6], the newly implanted lung seems to tolerate the entire cardiac output and provides adequate gas exchange, thus allowing implantation of the second lung without CPB. It is only in those situations where clamping the pulmonary artery results in an unacceptable pressure rise or where decomposition of right ventricular function is noted on transesophageal echocardiography that CPB is required. The thoracosternotomy incision provides excellent exposure to the right atrium and aorta, facilitating cannulation. The ability to perform this operation routinely without the need of CPB and the exposure afforded to both hemithoraces have allowed us to use the operation for patients who previously would have been rejected for transplantation because of technical factors related to the operation itself, such as a previous major pulmonary resection or pleurectomy. The bilateral thoracosternotomy approach not only makes a difficult dissection entirely feasible but allows it to be accomplished without great difficulty. It is the transverse sternal division that allows this excellent exposure. In this series, several patients who previously would not have been considered for transplantation were able to undergo the operation. They include 1 patient who had had a previous double-lung transplantation, a patient who had previously undergone a right middle lobectomy and total parietal pleurectomy, and another with recurrent pneumothoraces treated with multiple tube thoracostomies. On four occasions, ventila- 1 - Fig 6. Metabolic equiluvents (METS) generated dur- Ing the lower performance (LOP) standardized treadmill protdcol. Data are shown t the standard devia- 8- tion. (EVAL = evaluation; TX = transplantation.) v, 6- = 4- n=13 2t L n=14 n =9 1 I I I I EVAL PRE TX 6 12

7 444 KAISER ET AL Ann Thorac Surg 1991;52;43&?46 Fig 7. Comparison of change in percent predicted forced expiratory volume in 2 second for patients undergoing bilateral transplantation and those undergoing single-lung replacement for emphysema. Error bars indicate standard deviation. (TX = transplantation.) n - n LT a z LT Q 8 OSINGLE n.21 n=21 I/ 6 / 2 L n =23,,~ n= 19 n.14 n=4 -- *---'T-----y----y n=9 n=7 n =7 I 1 I I I PRE TX EEKS tor-dependent patients had successful transplantation, 3 of whom were able to be weaned quickly from mechanical ventilation and survived; the fourth died after repeat transplantation necessitated by primary donor lung dysfunction accompanied by diffuse airway necrosis. One of these patients had undergone bilateral bullectomy 2 weeks previously, had massive air leaks requiring multiple chest tubes, and had had a major deterioration in condition during the period after bullectomy. In another patient, respiratory failure developed immediately after completion of her initial evaluation for transplantation. Previously we have not considered patients for transplantation who are ventilator dependent. The procedure is particularly ideal for patients with cystic fibrosis, who often have marked pleural symphysis resulting from the recurrent infections that characterize their course. The ability to perform the operation without the routine use of CPB effectively reduces the risk of major hemorrhage as a result of the difficult dissection. ith the original en bloc double-lung operation, airway healing constituted a major problem [7]. The current operation builds on the modification of bilateral bronchial anastomoses originally suggested by Noirclerc and colleagues [4], and airway problems have, for the most part, been minor and of little clinical significance. If not for the performance of routine bronchoscopies, all but two airway defects would have remained inapparent. e have not had to place a bronchial stent in any patient in this series; all airway defects have healed with conservative therapy, although 1 patient required dilation of a bronchial stricture. The thoracosternotomy incision has not proved to be a major problem, either from the standpoint of pain in the initial perioperative course or in long-term healing. e have noted 3 patients with early sternal instability, but this has not proved to be a long-term problem. Pain in the perioperative period is well controlled with epidural narcotic analgesia. Perhaps the biggest question that remains is whether or not bilateral lung replacement offers an advantage over single-lung replacement in patients with end-stage emphysema. Obviously in patients with infection, such as those with cystic fibrosis or bronchiectasis, this is not an issue. Figure 7 compares the change in forced expiratory volume in 1 second for patients undergoing bilateral sequential lung replacement with that for patients undergoing single-lung replacement for emphysema. Although there seems to be a slight advantage for those receiving two lungs, this advantage is amplified somewhat by the inclusion of the patients with cystic fibrosis, most of whom are considerably younger. Likewise, a review of the 6-minute walk and 3-minute treadmill data seems to suggest a trend favoring those who receive two lungs. It remains to be determined if the slightly increased morbidity associated with a bilateral sequential lung replacement is justified on the basis of improved function with two lungs instead of one. If indeed there is very little functional significance to having the second lung, especially in those patients with end-stage emphysema, then perhaps the increase in morbidity and mortality and the less efficient use of donor lungs cannot be justified. Further answers will be provided as long-term follow-up becomes available. Our initial experience indicates that bilateral sequential lung transplantation can be performed safely, with a minimum of early mortality and acceptable morbidity, and compares favorably with single-lung transplantation in spite of being a more complex procedure with twice the number of airway anastomoses at risk. From May 1989 to May 1991, 7 patients had 37 single-lung transplantations with 1 perioperative death (97% operative survival) and 34 bilateral transplantations with 5 perioperative deaths (85% operative survival). Major problems remain especially in regard to infection from both bacterial and CMV pneumonitis, though we report no deaths directly attributable to either. e think that for patients who require double-lung replacement, the development of the bilateral sequential operation represents a major advance. The operation has special benefits for those patients with difficult technical problems and especially patients with cystic fibrosis. Despite initial concerns to the contrary, the added ischemic time for the second lung has not proved to be a problem, as we routinely have ischemic times in the 7-hour range seemingly without ill effect. Currently, we continue to offer bilateral lung replacement to patients with end-stage lung disease with an

8 Ann Thorac Surg 1991;52;43%46 KAISERETAL 445 infectious component as well as to most patients with emphysema who are less than 5 years old, to those with a large bullous component to their disease, or to those with the highest values of total lung capacity. ith the successful development of this operation and its application to a variety of situations, the indications for heartlung replacement continue to dwindle and now seem to be confined to patients with irreparable congenital heart disease accompanied by secondary pulmonary hypertension or patients with poor left ventricular function. References 1. Pasque MK, Cooper JD, Kaiser LR, Haydock DA, Triantafillou A, Trulock EP. Improved technique for bilateral lung transplantation: rationale and initial clinical experience. Ann Thorac Surg 199;49: Egan TM, Trulock EP, Boychuk J, Ochoa L, Cooper JD, and ashington University Transplantation Group. Analysis of referrals for lung transplantation. Chest 1991;99: Marquette treadmill protocols, series 19. Milwaukee, I: Marquette Electronics Inc. 4. Noirclerc MJ, Metras D, Vaillant A, et al. Bilateral bronchial anastomosis in double lung and heart-lung transplantations. Eur J Cardiothorac Surg 199;4:31&7. 5. Date H, Matsumora A, Manchester JK, et al. Successful twenty-four-hour lung preservation with low potassiumdextran-glucose solution and evaluation of lung metabolism. Presented at the Seventy-first Annual Meeting of the American Association for Thoracic Surgery, ashington, DC, May 6-8, Palazzo R, Hamvas A, Shuman T, Kaiser LR, Cooper J, Schuster D. Injury of nonischemic lung after unilateral pulmonary ischemia with pulmonary reperfusion. J Appl Physiol (in press). 7. Patterson GA, Todd TR, Cooper JD, et al. Airway complications after double lung transplantation. J Thorac Cardiovasc Surg 199;99: DISCUSSION DR G. ALEXANDER PATTERSON (Toronto, Ont, Canada): I congratulate Dr Kaiser on an excellent presentation and on a manuscript that will prove to be a valuable addition to the transplantation literature. The importance of this presentation is not in the detail of the postoperative functional results, as we already know that bilateral lung transplantation provides excellent results among operative survivors, but in the success of this technical modification, which the ashington University group has so nicely described. The original double-lung transplantation technique was flawed because of its technical difficulty, its reliance on cardiopulmonary bypass, and the high incidence of airway complications in patients after transplantation. However, the concept of doublelung replacement with preservation of the recipient heart has always been a sound one. hat was required was the technical innovation to resolve some of the problems that have plagued the initial procedure. Bilateral bronchial anastomoses had already been employed, and the Marseille group deserves a large amount of credit for this. This modification had reduced the incidence of airway complications. At the suggestion of Dr Grillo, my colleagues and I had employed a clamshell type of incision previously and demonstrated its outstanding exposure of the pleural space. In September 1989, we performed an inadvertent bilateral sequential lung transplantation in a tall patient (192.5 cm or 6 feet 5 inches) with emphysema who had waited some 21 months for a donor. The planned left lung transplantation was completed; however, the graft looked ridiculously small in the pleural space. I had worked with Drs Pearson, Cooper, Ginsberg, and Todd long enough to know when trouble was about to occur, so we elected to turn the patient and perform a right lung transplantation through another thoracotomy. The postoperative result was excellent, but the technical implementation of this procedure was not correct at all. Drs Michael Pasque and Larry Kaiser and the ashington University group, under Dr Cooper s typical leadership, have evolved a procedure that is elegantly simple in its conduct. Of our 38 bilateral lung transplantations, the most recent 2 have been performed using the sequential single-lung technique, the decision being to replace the lung with the least function first in hopes of avoiding a long cardiopulmonary bypass run. The incidence of major airway complications has diminished dramatically. In fact, since adopting a telescoping bronchial anastomotic technique with routine perioperative steroid administration and no omentopexy, we have had only one minor anastomotic stricture. However, our increase in deaths from sepsis reflects the number of patients with cystic fibrosis in our group and the tendency for these patients to experience septic complications from the resistant gram-negative organisms that continue to contaminate the airway after transplantation. I have a couple of questions for Dr Kaiser. First, how do you determine intraoperatively which lung is to be removed first and transplanted? Second, what are the precise indications in this ashington University experience at the moment for the application of single-lung or bilateral lung replacement in patients with obstructive lung disease? DR ERNST OLNER (Vienna, Austria): I, too, congratulate Drs Kaiser and Cooper for their excellent results. e from Europe are grateful to Dr Cooper and Dr Patterson for teaching us how to do lung transplantation. e cannot compete with their results, but our limited experience in Vienna now comprises 7 patients who have had bilateral sequential lung transplantation. There are 5 long-term survivors. Our ischemic time is very similar to the reported status. The only difference between our technique and the technique described is that we do not use the omentum for wrapping the bronchial anastomosis. e use an intercostal muscle flap. I have three questions for Dr Kaiser. First, what is your policy about using steroids in patients before, during, and immediately after the operation in regard to bronchial problems? Second, can you foresee whether a candidate for double-lung transplantation will need extracorporeal circulation or not? In our series, we have used femoral-femoral partial bypass in 3 of the 7 patients. Last, we recently had 2 patients with severe obliterative bronchiolitis after heart-lung transplantation. e performed single-lung transplantation in both, and 1 is a long-term survivor. hat is your policy in the case of patients with excellent heart function but obliterative bronchiolitis? Do you perform single-lung trans-

9 446 KAISERETAL Ann Thorac Surg 1991;52;43&46 plantation, double-lung transplantation, or heart-lung transplantation? DR THOMAS M. EGAN (Chapel Hill, NC): This was an excellent presentation. Both the University of Toronto and ashington University groups have been instrumental in the development and application of isolated lung transplantation. At the University of North Carolina in Chapel Hill, our experience is somewhat smaller. In the last year, we performed eight double-lung transplantations in 7 patients. Four patients had cystic fibrosis, 1 had chronic obstructive pulmonary disease, and 2 had other infectious end-stage lung disease, including a young man who had been ventilator-dependent for 4 months with end-stage adult respiratory distress syndrome. All 7 patients survived and were discharged, none require supplemental oxygen, and all are ambulatory. Cardiopulmonary bypass was necessary in 3 of the 8 transplant procedures. One patient required retransplantation for graft dysfunction on the seventh postoperative day. The hospital stay for our patients is somewhat longer than that of the ashington University group. Our mean stay is 44 days. Three patients required a long hospitalization. One of them had a Haldol (haldoperido1)-related drug reaction, which resulted in prolonged ventilation; 1 had a prolonged period of ventilation before transplantation and was in a considerably debilitated condition at the time of transplantation; and the patient who had a redo transplantation required ventilation for a long period. The remaining 4 patients, however, had a mean hospital stay similar to that reported by the ashington University group. Fourteen airways that are at risk have been evaluated. There has been one partial dehiscence, which was in the patient having repeat transplantation. That airway is healing. At present, we have no strictures and, to date, have not needed to use stents. I have a couple of questions for Dr Kaiser. First, what are your current contraindications to double-lung transplantation now that we recognize that this procedure can be applied to patients who have had previous thoracotomies and previous invasion of the pleural space? Second, and 1 guess this really echoes Dr Patterson s question, what do you do for a young patient with end-stage chronic obstructive pulmonary disease? DR KAISER: I thank all the discussants for their comments, and I will try to answer the questions. First, Dr Patterson, I very much appreciate your discussing the manuscript and congratulate you on the excellent results of the Toronto group, who have truly been the leaders in this field. In answer to your question as to which side we do first, if there is a major difference on the perfusion scan, we tend to do the side that has the least perfusion first, so that we can try to avoid cardiopulmonary bypass by saving the better lung and doing it second. In terms of indications for single versus bilateral replacementand this will address one of Dr Egan s questions as well-we had initially limited bilateral transplantation to patients less than 5 years old. I think we have extended that somewhat now. e have tended to do bilateral lung transplantation in patients with the larger total lung capacities as well as those with huge bullous disease, our concern being that perhaps single-lung transplantation in patients with emphysema with a big bullous component might lead to more air trapping. Certainly in the younger patients I think we have favored bilateral transplantation as a routine. Dr olner, I congratulate you on your early results. In regard to the question about steroids, we do not use steroids perioperatively, although we recognize the excellent results that have been obtained by Dr Trinkle, Dr Patterson, and others. I guess old habits are hard to break. e have had good results by withholding steroids until postoperative day 6. e are now considering patients and do see patients who are on a regimen of preoperative steroids, so our policy may change. ith reference to cardiopulmonary bypass and how we assess the need, we use transesophageal echocardiography and a period of trial clamping of the pulmonary artery before removing the lung. atching the right ventricular function and pulmonary artery pressures, we think that we can avoid bypass if clamping seems to be well tolerated. I think in the majority of instances we can do that. Concerning obliterative bronchiolitis, we have now seen 3 cases in our series. One of the patients was a previous heart-lung recipient on whom we did a single-lung transplantation, 1 had pulmonary hypertension, and 1 had bilateral lung transplantation. The main contraindications to the bilateral sequential operation are poor left ventricular function and age greater than 5 years, although, as I said, we are somewhat flexible with the age cutoff. Dr Egan, I also congratulate you on your results, which are certainly excellent.