T nance of chronic immunosuppression [l] and improvements

Transcription

1 Reduced-Size Porcine Lung Transplantation: Long- Term Studies of Pulmonary Vascular Resistance John A. Kern, MD, Curtis G. Tribble, MD, Barry B. K. Chan, MD, Terry L. Flanagan, MPH, and Irving L. Kron, MD Department of Surgery, University of Virginia Health Sciences Center, Charlottesville, Virginia The use of isolated adult lobes for pediatric lung transplantation has recently been reported and could potentially help alleviate the profound pediatric donor lung shortage. However, the effect of chronic denervation on pulmonary vasculature of isolated mature lobar transplants is not well understood. Previously, we reported that chronic denervation of the immature porcine lobe results in abnormal pulmonary vascular compliance. We now present studies of long-term pulmonary hemodynamics in young pigs 12 weeks after transplantation of a reduced-size mature left lower lobe. Resting pulmonary vascular resistance of the transplanted mature lobes was similar to that of innervated lobes of age-matched con- trols. In addition, pulmonary vascular resistance of the transplanted mature lobes did not rise abnormally in response to increased flow caused by clamping the contralateral pulmonary artery. We conclude that denervation of the mature porcine lobe does not result in abnormal pulmonary vascular resistance. In addition, vascular compliance of reduced-size mature porcine lobar transplants is superior to that of denervated reimplanted immature lobes. These findings suggest a deleterious effect of denervation on pulmonary vascular development of the growing porcine lung. (Ann Thoruc Surg 1992;53:583-9) he introduction of cyclosporine in 1983 for the mainte- T nance of chronic immunosuppression [l] and improvements in bronchial anastomotic techniques [2] together ushered in a new era of successful lung transplantation. As a result, unilateral single-lung transplantation has become an accepted and increasingly reliable treatment option for a number of end-stage pulmonary diseases. For adults, accepted indications for this life-saving procedure continue to broaden as long-term survival improves [3]. In the neonatal and pediatric population, however, a relative donor shortage caused by frequent recipient to donor size disparity has prevented any large series documenting long-term survival after single-lung transplantation. An alternative to the use of pediatric cadaveric lung allografts is the use of reduced-size lung transplants in which a lobe or segment of an adult lung, either cadaveric or livingrelated, is sculpted to fit the recipient s chest [4]. Although reduced-size lung transplantation has now been reported experimentally as well as clinically [4, 51, longterm functional capacity of these transplants has not been reported. Experimental lung transplantation with its associated hilar denervation has been well investigated and is known to result in physiologic abnormalities in airway mechanics and pulmonary vascular function [6-91. Previously, we reported that chronic denervation of the reimplanted Presented at the Thirty-eighth Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 7-9, Address reprint requests to Dr Kron, Department of Surgery, University of Virginia Health Sciences Center, Box 181, Charlottesville, VA immature porcine left lower lobe results in abnormal pulmonary vascular compliance when studied after a period of substantial somatic growth. Increasing blood flow through the denervated lobe caused by clamping of the contralateral pulmonary artery resulted in an abnormal rise in pulmonary vascular resistance [lo]. Because human postnatal lung development is not complete until early in the second decade [ll], transplanting a lung from such a young donor requires denervation before functional and morphologic maturity and may result in similar pulmonary vascular dysfunction. To investigate the importance of complete lung maturity at the time of transplantation and its possible role in determining functional outcome, we studied pulmonary vascular physiology and gas exchange of reduced-size mature left lower lobe transplants in growing pigs. Material and Methods Animal Model and Animal Care The domestic swine was chosen because of its rapid postnatal lung development [ll-131 and because of our familiarity with its tolerance of thoracotomy and lung transplantation. All experimental protocols were reviewed and approved by an institutional animal use committee. The guidelines used by this committee conform to the Principles of Laboratory Animal Care put forth by the National Society of Medical Research and the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication No , revised 1985) by The Society of Thoracic Surgeons /92/$5.00

2 584 KERNETAL 1992;53:58>9 Three animal groups were defined. The first was a control group of 5- to 7-month-old mature pigs with normally innervated left lower lobes. The experimental group consisted of 8- to 10-week-old piglets who were recipients of a reduced-size mature left lower lobe allograft from 6-month-old donors. (At 6 months, most features of porcine lung development are at adult levels [ll].) The last group of animals consisted of young piglets who underwent left pneumonectomy with reimplantation of a denervated left lower lobe autograft at 8 to 10 weeks of age. This final group of reimplanted animals served essentially as an additional control group for comparison with the reduced-size transplant group. Some data gathered from this group have previously been reported [lo]. All transplant and reimplant procedures in all animals were performed by the same surgeons using the same surgical technique and method of preservation (ice-cold saline flush without prostaglandin infusion) and similar ischemic times. All animals were obtained from a single supplier, and all recipient animals were of similar initial body weight ( kg). Donor Operation Mature left lower lobes used for transplantation were explanted from 6- to 7-month-old sexually mature pigs (n = 10) whose mean body weight was kg. After induction of anesthesia with intramuscular ketamine (10 mg/kg), the animals were orally intubated and maintained on 1.5% to 2% halothane. The pigs were mechanically ventilated with a volume ventilator (Harvard Apparatus) at a tidal volume of 15 ml/kg and a respiratory rate of 12 to 16 per minute. A left lateral thoracotomy was performed after a subperiosteal resection of the fifth rib. The hemiazygos vein was ligated and divided to gain access to the left main pulmonary artery and superior lobe vein. Both structures were dissected free and isolated, as was the left main bronchus. The left lower lobe vein was then dissected free and the inferior pulmonary ligament sharply divided. The incomplete fissure between the upper and lower lobes was sharply divided, with care to preserve lower lobe parenchyma. The animal was anticoagulated with sodium heparin (100 U/kg intravenously) and the left lung was excised with adequate lengths of artery and vein. The lung was then immersed in ice-cold physiologic saline slush at 0 C and perfused with heparinized ice-cold normal saline solution through the pulmonary artery at low pressure until the venous effluent returned clear. The upper lobe arteries were then ligated and divided, and the left main bronchus was transected immediately distal to the upper lobe bronchus. Wet weights of the isolated lobes were obtained, and the upper lobe was used for measurements of extravascular lung water and calculation of dry lung weight of the transplanted lobe at time of implantation [14]. The donor animal was then killed with a lethal dose of pentobarbital. Recipient Operation At approximately 9 weeks of age, a total of 10 immature piglets (weight 20? 2 kg) were designated as transplant recipients of the mature left lower lobes. Beginning 2 days before the transplant procedure all recipient animals received one aspirin (325 mg/day orally) to decrease the risk of pulmonary vein thrombosis. Animals were anesthetized as described in the donor operation and placed on a volume ventilator and maintained on 0.8% halothane. After the intravenous administration of ampicillin (250 mg) and chloramphenicol(500 mg), the recipient pigs underwent a standard left lateral thoracotomy through the fifth intercostal space. A left pneumonectomy was performed after systemic heparinization with sodium heparin (100 U/kg intravenously). The left pulmonary artery was clamped at its takeoff from the main pulmonary artery, the upper lobe veins were ligated, and the lower lobe vein was clamped at its junction with the pyramidal lobe vein and left atrium. The main bronchus was clamped as proximally as possible, and the left lung was excised. The donor lobe was positioned in the chest and the lower lobe vein was anastomosed first with a running 7-0 absorbable monofilament suture (Maxon; Davis & Geck, Danbury, CT). To compensate for the size mismatch, the recipient vein was opened at its first branch point and this area was used for the anastomosis. The donor lower lobe bronchus was then anastomosed to the recipient lower lobe bronchus by telescoping the recipient main bronchus inside the donor lower lobe bronchus approximately 7 mm. A running 4-0 Maxon suture was used for the back wall of the bronchial anastomosis, and a series of interrupted figure-of-8 sutures was used for the front wall. The bronchus was not wrapped with omentum. The bronchial clamp was released, the tidal volume was adjusted, and the anastomosis was examined for air leaks. Finally, the pulmonary artery was anastomosed using a running 6-0 Maxon suture. Again, any size mismatch was compensated for by opening the recipient vessel through a side branch. Before the arterial suture was tied, the venous clamp was removed to allow for back-bleeding and deairing. At this time the recipient animal received 500 mg of intravenous methylprednisolone, 1 mg/kg of intravenous azathioprine, and a second 100 Ukg dose of heparin. The arterial suture was then tied and the clamp removed. The vascular anastomoses were examined, and the chest was closed in layers after placement of a chest tube for drainage. The chest tube was removed after 24 hours. An additional group of 8- to 10-week-old piglets (n = 6) underwent a left pneumonectomy with reimplantation of the native left lower lobe only, after perfusion of the isolated lobe with heparinized ice-cold normal saline solution as described above. This group served as a set of animals whose left lower lobes were denervated at an immature stage of development. All transplant and reimplant animals were allowed free access to food and water for the 12-week holding period. To serve as control innervated left lower lobes, 7 additional pigs aged 5 to 6 months (mean body weight, 109 * 6 kg) underwent left upper lobectomy acutely at the time of final study, followed immediately by measurements of isolated left lower lobe pulmonary hemodynamics (de-

3 1992;53:58%9 KERNETAL 583 scribed below). Four of these 7 study animals survived the entire experimental procedure and represent the control group. These animals were then sacrificed with a pentobarbital overdose. lrn rn u nos up pression On the day before the transplant procedure, recipient animals received cyclosporine (18 mg. kg-. day- orally; Sandoz Pharmaceutical Corp, East Hanover, NJ), which was continued on a daily basis postoperatively. In addition, recipient animals received methylprednisolone (500 mg intravenously) at the time of transplantation and daily thereafter for 5 days, and azathioprine, 1 mgkg intravenously at the time of transplantation and 1 mg/kg orally daily thereafter. Acute rejection was diagnosed when tachypnea, fever, lethargy, and dry cough occurred. These episodes were treated empirically with pulsed methylprednisolone (500 mg intramuscularly) for 3 days. Cyclosporine levels and chest roentgenograms were obtained within the first 2 weeks postoperatively and on a monthly basis thereafter. In Vivo Studies of Pulmonary Vasculature Approximately 12 weeks after the transplant procedure, all animals were again anesthetized with ketamine and orally intubated, and a chest roentgenogram was performed to document normal aeration. Of the 10 original recipients of a reduced-size mature lobe, 3 pigs showed little to no aeration of their transplanted lobe and were excluded from further study. At autopsy these pigs were believed to have thrombosed their grafted lobe secondary to pulmonary vein thrombosis. Of the remaining 7 pigs, 1 animal died of cardiac arrest during the terminal study period and 1 animal had an atelectatic, consolidated segment of the transplanted left lower lobe, and complete data from these animals were not available. The remaining 5 animals make up the mature lobar transplant group. Of the 6 animals that received a reimplanted autograft left lower lobe, 5 survived the entire terminal experimental procedure and make up the reimplanted immature lobe group. Animals were maintained with intermittent intravenous pentobarbital and mechanically ventilated (12 to 15 ml/kg, 10 to 12 breathdmin, inspired oxygen fraction of 1.0). Carotid artery and pulmonary artery catheters were then placed through right common carotid and external jugular venous cutdowns. Carotid artery, central venous, and pulmonary artery pressures were measured continuously with a multichannel recorder (ES1000; Gould Inc, Cleveland, OH). A median sternotomy was performed, and the pulmonary artery catheter was confirmed to be in the main pulmonary artery by palpation. Total pulmonary hemodynamic and systemic hemodynamic measurements were obtained in the resting steady state at end-expiration. Total pulmonary blood flow was measured in triplicate by thermodilution technique using a thermistor-tipped Swan-Ganz catheter. To verify our blood flow measurements, an attempt was made to dissect free the right and left main pulmonary arteries for placement of ultrasonic flow probes. How ever, due to dense scar tissue around the previously dissected pulmonary artery, safe placement of the flow probes could not be accomplished in all animals. (In 2 transplant and 2 control animals studied in this manner, flow measured by ultrasonic probes correlated to within 7% of the thermodilution cardiac output. Thereafter, tu prevent loss of animals, only thermodilution cardiac out puts were used.) After the pulmonary artery catheter wda manually guided into the left main pulmonary artery, resting (before clamping of the right hilum) left lower lobt pulmonary hemodynamic measurements were taken, and the left lower lobe pulmonary vascular resistance was calculated using the entire cardiac output as determined by thermodilution. To increase the cardiac output to tht transplanted lobe, the contralateral right pulmonary a- tery was then totally occluded with an atraumatic vascular clamp and the animals were allowed to stabilize for 1U minutes. Pulmonary hemodynamic measurements of tht isolated left lower lobe were then repeated with tht contralateral pulmonary artery clamped and all flow di rected to the left lower lobe, and systemic arterial blood samples were obtained for blood gas analysis. Animals were sacrificed with a lethal dose of pentobarbital after completion of the hemodynamic measurements. Studies of Gas Exchange All animals underwent systemic arterial blood gas analysis (inspired oxygen fraction of 1.0) during the period of right pulmonary artery occlusion, with the entire cardiac output shunted through the left lower lobe. In addition, animals that received mature left lower lobe transplants underwent blood gas analysis of right (to serve as control) and left lower lobe pulmonary vein blood samples while breathing room air as a more accurate measure of the reduced-size lobar transplant s ability to oxygenate dnd ventilate. Stat is tics Measurements are reported as the mean standard error of the mean. Student s t statistic was calculated for un paired samples to compare differences between experi mental transplant and control groups and for paired samples within the same group when such samples were obtained. A probability value less than or equal to 0.05 was taken to indicate a significant difference between measurements. Comparison of baseline hemodynamics, isolated left lower lobe pulmonary hemodynamic med surements, and blood gas analyses between control innervated left lower lobes, mature lobar transplants, and the denervated reimplanted immature lobes was done by analysis of variance, with a probability value less than or equal to 0.05 indicating significance. Results Technical The mean ischemic time for the transplanted and reimplanted lobes was 167 k 12 minutes. At the time of transplantation, the mature lobes were significantly larger

4 586 KERNETAL 1992;5358>9 Table 1. Resting Hemodynamic Variables" Heart Rate Aortic Pressure CVP Main PAP Total PVR Group (bea ts/min) (mm Hg) (mm Hg) (mm Hg) (mm Hg * L-' * min-i) Control LLL (innervated) 101 f (n = 7) Mature LLL transplants f 6b 9?l 25 f (n = 6) Reimplanted immature LLL 130 f 14b ' ? 3 2.6? 0.4b*' (n = 6) a Data are mean f standard error of the mean. p versus control LLL. p versus mature LLL transplants. CVP = central venous pressure; LLL = left lower lobe; PAP = pulmonary artery pressure; PVR = pulmonary vascular resistance. than the entire immature left lungs that they replaced, as determined by wet weight (141? 10 versus 66? 6 g; p ). Extravascular lung water fractional content was the same between the mature lobes and immature lungs (0.92? 0.01). At time of final study, all animals (controls, mature lobar transplants, and reimplanted immature lobes) were of similar postnatal age (20 to 22 weeks) and weight ( kg). We saw no bronchoscopic evidence of bronchial anastomotic strictures in any of the animals that received a transplanted or reimplanted left lower lobe. In addition, postmortem studies revealed no evidence of pulmonary vascular anastomotic strictures in any of the animals with a viable left lower lobe allograft or autograft. Pulmonary Hernodynamics Resting baseline systemic hemodynamic variables of all animal groups are presented in Table 1. Baseline total pulmonary vascular resistance was similar between the control innervated lobes and mature lobar transplants. However, the total pulmonary vascular resistance of the animals whose left lower lobes were reimplanted at a young age was significantly higher (p ). Isolated left lower lobe pulmonary hemodynamic measurements of innervated controls, reduced-size mature lobar transplants, and reimplanted immature lobes are shown in Table 2. The values are expressed as mean f standard error of the mean before and after occlusion of the right pulmonary hilum. No significant hemodynamic differences were seen between the transplanted mature lobes and normally innervated lobes either before or after occlusion of the contralateral pulmonary artery. Most importantly, clamping of the contralateral pulmonary artery did not cause an abnormal rise in the pulmonary artery pressure or pulmonary vascular resistance of the transplanted mature lobes. However, when comparison was made with the animals who had undergone denervation of their left lower lobe before complete pulmonary development, significant differences in cardiac output and subsequent pulmonary vascular resistance were seen (p ). Most importantly, clamping the contralateral pulmonary artery resulted in an abnormal rise in pulmonary vascular resistance in the lobes denervated at a young age (p ). Although mean pulmonary artery pressure rose more than in the innervated control or mature transplanted lobes, this was not significant. Table 3 represents the observed increase in pulmonary vascular resistance in all three groups that occurred after clamping of the right pulmonary artery. Again, the transplanted mature lobes demonstrated a change very similar to the control innervated lobes, indicating a normal ability to maintain cardiac output and vasodilate in response to the increased lobar flow. However, the lobes that were Table 2. Isolated Left Lower Lobe Pulmona y Hemodynamics" PAP (mm Hg) PCWP (mm Hg) CO (L/min) PVR (mm Hg L-' min-') Group Pre Post Pre Post Pre Post Pre Post Control LLL (innervated) % ? C (n = 4) Mature LLL transplants 31 c c 1 17? 2 9.2? C Reimplanted immature LLL ? ? 3 5.5? 0.4b,' 3.7 C 0.7b,C b br' a Data are mean? standard error of the mean. CO = cardiac output; LLL = left lower lobe; clamping of the contralateral pulmonary artery; p versus control LLL. p versus mature LLL transplants. PAP = pulmonary artery pressure; PCWP = pulmonary capillary artery pressure; Post = after Pre = before clamping of the contralateral pulmonary artery; PVR = pulmonary vascular resistance.

5 1992; KERNETAL 587 denervated and reimplanted at an immature age demonstrated a rise in pulmonary vascular resistance that was significantly higher than the other two groups (p ). Blood Gas Analysis Systemic blood gas analysis with the entire cardiac output shunted through the left lower lobe (after clamping) of all three groups is presented in Table 4. No significant differences were seen in oxygen tension, carbon dioxide tension, or ph between the three groups, indicating adequate long-term gas exchange in the reduced-size lobar transplant. Before termination of the experiment, animals in the mature lobar transplant group were changed to room air ventilation for 15 minutes and blood samples were obtained from the right and left pulmonary veins to further analyze the ability of the transplanted mature lobe to oxygenate and ventilate. These results are presented in Table 5. Again, no significant differences were seen between the transplanted mature lobe and the normally innervated right lung. Comment Transplantation of an isolated pulmonary lobe or segment from a more mature donor into an immature recipient is technically feasible. Although this type of transplant has been reported both experimentally and clinically [4, 51, long-term functional studies of such reduced-size transplants have not been reported. In Crombleholme and associates [4] recent report of reduced-size lung transplantation in neonatal swine, short-term studies of the pulmonary vasculature demonstrated no increase in pulmonary artery pressure and only a slight increase in pulmonary vascular resistance in the more mature lobe immediately after transplantation. Crombleholme and associates suggested that the more mature lobar graft was better able to accommodate the blood flow from the neonatal recipient, despite denervation. These studies in the early posttransplantation period yielded results similar to our own; however, Crombleholme and associates studies were only in the resting state. The response of the transplanted reduced-size lobe to an acute rise in blood Table 3. Changes in Pulmonary Artery Pressure and Pulmonary Vascular Resistance Caused by Clamping of the Contralateral Pulmonary Artenf Group Change in PAP Change in PVR Control LLL (innervated) 18.5 f f 0.4 (n = 4) Mature LLL transplants 17.6 f f 0.5 Reimplanted immature LLL 34.8 f 10.3b 10.3 f 3.6c3d a Data are mean f standard error of the mean. p versus mature LLL transplants. p versus control LLL. p versus mature LLL transplants. LLL = left lower lobe; PAP = pulmonary artery pressure; PVR = pulmonary vascular resistance. Table 4. Systemic Arterial Blood Gas Analysis With Entire Cardiac Output Shunted Through the Left Lower Lobe (After Clamping) PaCO, PaO, Group ph (mmhg) (mmwb Control LLL (innervated) 7.39 f f f 69 (n = 4) Mature LLL transplants 7.34 f f f 37 Reimplanted immature 7.40 f * LLL (n = 4) a Data are mean 2 standard error of the mean. Inspired oxygen fraction = 1.0. LLL = left lower lobe; PaCO, = partial pressure of carbon dioxide in systemic arterial blood sample; PaO, = partial pressure of oxygen in systemic arterial blood sample. flow and its ability to modulate pulmonary vascular resistance chronically were not studied. In clinical singlelung or lobar transplantation, the contralateral lung will always be abnormal. Thus, studies in experimental models in which there is flow through a normal contralateral lung are insufficient. Studies of vascular compliance under conditions of high flow are necessary before reducedsize transplants can be widely offered as a reliable longterm treatment option for children with end-stage pulmonary disease and to help determine whether mature lobes or immature lungs should be used for donor organs. Most previous studies of pulmonary hemodynamics after experimental lung or lobar reimplantation or allotransplantation have been performed in adult animals with fully developed lungs. The issue of abnormal pulmonary hemodynamics remains unsettled. Many authors note an abnormally high or fixed pulmonary vascular resistance in transplanted or reimplanted lungs, perhaps due to an inability of the denervated lung to vasodilate appropriately [%lo, 15-17]. However, other authors have noted completely normal pulmonary vascular resistance after experimental lung transplantation and claim most problems with abnormal pulmonary vascular resistance are caused by an improperly constructed pulmonary venous anastomosis [18, 191. We use absorbable suture material for all bronchial and vascular anastomoses to avoid the complication of fixed strictures and to allow for growth of the anastomoses. Table 5. Pulmonary Venous Blood Gas Analysis of Reduced- Size Lobar Transplants While Breathing Room AiP Lung PVCO, PVO, Native right lung 31 f f 11 Mature LLL transplant 32 f llb a Data are mean f standard error of the mean. No significant difference by paired t test versus native right lung. LLL = left lower lobe; PvCO, = partial pressure of carbon dioxide in pulmonary venous blood sample; PvO, = partial pressure of oxygen in pulmonary venous blood sample.

6 588 KERNETAL 1992;53:583-9 The pulmonary vascular physiology of experimentally transplanted or reimplanted immature lungs has only recently been studied and reported. We reported that reimplantation of the left lower lobe in young piglets results in decreased vascular compliance of that lobe when studied chronically [lo]. The present results show that denervation of the mature porcine lung does not result in long-term abnormal pulmonary vascular physiology. We believe that the functional differences that exist after denervation of the immature as compared with the mature lung are related to the degree of pulmonary maturation at the time of denervation. The degree to which normal lung development is dependent on intact neural stimulation and the effects of denervation on subsequent development and function of an immature lung are not completely understood. Our present results support the hypothesis that denervation before complete maturation of the immature porcine lung has a profound effect on subsequent function. Because the pulmonary vascular functional abnormalities seen in denervated immature lobes are not present in transplanted mature lobes, we believe that intact neural input is required for proper pulmonary vascular development of the porcine lung and interruption of this pathway results in long-term vascular dysfunction. Correlation between porcine and human lung development [ll] would suggest that clinical transplantation of an organ younger than 10 to 12 years may result in similar vascular dysfunction. Although neural regeneration of a transplanted lung may occur [20], denervation during a critical period of lung development may be enough to produce profound long-term functional abnormalities. In the normally innervated, healthy pulmonary vascular bed, an increase in flow results in capillary recruitment with a resultant decrease in pulmonary vascular resistance and maintenance of pulmonary artery pressure [21]. Denervation of the immature pulmonary vasculature seems to impair this normal physiologic response, whereas denervation of the mature pulmonary vasculature does not. It is possible that capillary recruitment was facilitated in the mature lobe due to its larger size at the time of transplantation. However, because airway physiology of the denervated immature porcine lung is also abnormal [6], we believe that denervation of the immature porcine lung impairs subsequent development of that lung and that the manifestations of this denervation include profound long-term functional abnormalities. Further experimental studies and clinical correlation to human lung transplant recipients will be critical. Our data suggest that transplantation of mature lung tissue may be functionally superior to immature allografts when the need for pediatric lung transplantation arises. This work was supported in part by funds from the American Heart Association, Virginia Affiliate. References 1. Goldberg M, Cooper JD, Lima 0, et al. A comparison between cyclosporine-a and methylprednisolone plus azathioprine on bronchial healing following canine lung autotransplantation. J Thorac Cardiovasc Surg 1983;85: Morgan WG, Lima 0, Goldberg M, et al. Improved bronchial healing in canine left lung reimplantation using omental pedicle wrap. J Thorac Cardiovasc Surg 1983;85: Egan TM, Kaiser LR, Cooper JD. Lung transplantation. In: Wells SA, ed. Current problems in surgery. Chicago: Year Book Medical, 1989: Crombleholme TM, Adzick NS, Longaker MT, et al. Reduced-size lung transplantation in neonatal swine: technique and short-term physiologic response. 1990; 49: Goldsmith MF. Mother to child: first living donor lung transplant. JAMA 1990;264: McGahren ED, Teague WG, Flanagan TL, et al. Airway obstruction following autologous reimplantation of the porcine lobe. J Thorac Cardiovasc Surg 1989; Glanville AR, Burke CM, Theodore J, et al. Bronchial hyperresponsiveness after human cardiopulmonary transplantation. Clin Sci 1987;73: Allgood RJ, Ebert PA, Sabiston DC. Immediate changes in pulmonary hemodynamics following lung autotransplantation. Ann Surg 1968; Deasi PM, Nyhan DP, Nishiwaki K, et al. Angiotensin I1 and arginine vasopressin do not mediate chronic pulmonary vasoconstriction following left lung autotransplantation. FASEB J 1991;5(Suppl):A Johnson AM, Teague WG, Flanagan TL, McGahren ED, Kron IL. Decreased vascular compliance after reimplantation of the left lower lobe in young pigs. 1990; Rendas A, Branthwaite M, Reid L. Growth of pulmonary circulation in normal pig-structural analysis and cardiopulmonary function. J Appl Physiol 1978;45: Winkler GC, Cheville NF. The neonatal porcine lung: ultrastructural morphology and postnatal development of the terminal airways and alveolar region. Anat Rec 1984; Winkler GC, Cheville NF. Morphometry of postnatal development in the porcine lung. Anat Rec 1985;211: Cowan GSM, Staub NC, Edmunds LH. Changes in the fluid compartments and dry weights of reimplanted dog lungs. J Appl Physiol 1976;40: Tisi GM, Trummer MJ, Cuomo AJ, Ashburn WL, Moser KM. Long-term autotransplanted canine lungs: base-line ventilatory and hemodynamic function. J Appl Physiol 1972;32: Garzon AA, Dutta SK, Okadigwe C, et al. Functions of canine lung allografts. J Thorac Cardiovasc Surg 1973;65: Wildevuur CRH, Heemstra H, Tammeling GJ, et al. Longterm observation of the changes in pulmonary arterial pressure after reimplantation of the canine lung. J Thorac Cardiovasc Surg 1968;56: Koerner SK, Veith FJ. Hemodynamics of transplanted lungs. Chest 1971;59: Veith FJ, Montefusco CM. Long-term fate of autografts charged with providing total pulmonary function. Ann Surg 1979;190: Edmunds H, Graf PD, Nadel JA. Reinnervation of the reimplanted canine lung. J Appl Physiol 1971;31: Fishman AP. Dynamics of the pulmonary circulation. In: Dow P, ed. Handbook of physiology, circulation; section 2, vol 11. Washington, DC: American Physiologic Society, 1963:

7 1992;5358%9 KERNETAL 589 DISCUSSION DR THOMAS M. EGAN (Chapel Hill, NC): I think this is an exciting approach to a difficult problem, the problem of a very critical shortage of pulmonary donors. I have a couple of questions. Although your pulmonary arterial anastomoses are open at autopsy, did you measure any pressure difference across them? I ask because they still represent a fixed diameter, and when you divert the entire cardiac output, one of the ways that the pulmonary artery compensates is to increase its diameter. The second question is, with respect to comparing a mature lobe transplanted as a single lung to an immature lobe, would it not be more appropriate to have compared that with an intact lung in terms of pulmonary vascular resistance? Were you disappointed that your pulmonary vascular resistance in these immature lobes that functioned as lungs appeared to be elevated? DR KERN: With respect to the first question, all of the vascular anastomoses were done with absorbable suture to avoid fixed stenoses. Nevertheless, the way we measured pulmonary vascular hemodynamics was with the sternum open so we could manually guide the pulmonary artery catheter into the pulmonary artery to be studied. Pressure measurements were made proximal and distal to the anastomoses, and there were no pressure gradients detectable. We agree that chronic studies in an experimental model of single whole lung transplantation are needed, especially to study the effects of chronic denervation on pulmonary growth. We initially began this project to study the effects of denervation on pulmonary vascular and airway physiology after transplantation. Therefore, we did not want to complicate the matter by using a whole lung, which in this animal model is often complicated by superior lobe vein thrombosis. We have since developed a technique of single whole lung transplantation in this immature porcine model, using for the superior lobe vein anastomosis an atrial cuff on the donor vein going into the left atrial appendage of the recipient. We will shortly begin airway and vascular studies and studies of lung growth in these chronic animals. Finally, although we were surprised to find abnormal pulmonary vascular responses in the denervated immature lobes, we believe the finding of normal hemodynamics in the mature lobes suggests that proper innervation is required for normal pulmonary vascular development of the porcine lung.