Human En Bloc Double-Lung Transplantation: Bronchial Artery Revascularization Improves Airway Perfusion

Human En Bloc Double-Lung Transplantation: Bronchial Artery Revascularization Improves Airway Perfusion Arve Sundset, MD, Samad Tadjkarimi, MD, Asghar Khaghani, FRCS, Knut Kvernebo, PhD, and Magdi H. Yacoub, FRCS Department of Thoracic and Cardiovascular Surgery, Harefield Hospital, Harefield, Middlesex, and National Heart and Lung Institute, London, United Kingdom, and Department of Medicine, Aker Hospital, and Department of Cardiothoracic Surgery, Ullevaal Hospital, Oslo, Norway

Background. Ischemic airway complications are common after en bloc double-lung transplantation with tracheal anastomosis. The aim of this study was to evaluate the effects of a direct revascularization of the donor bronchial artery with the recipient internal thoracic artery on airway perfusion. Methods. Seven patients undergoing double-lung transplantation with tracheal anastomosis were investigated intraoperatively and postoperatively (12 to 36 hours) with endoscopic laser Doppler flowmetry. Sixteen patients undergoing coronary artery bypass grafting served as a control group. Results. Two patients who had double-lung transplantation with tracheal anastomosis died of sepsis and multiorgan failure 1 week after transplantation. In the remaining 5 patients healing of the anastomosis was excellent during the observation period of 3 to 52 months. In 5 patients clamping of the attached internal thoracic

artery induced a reduction of the laser Doppler flowmetry signal from 10% to 60%. In the 2 patients with the highest graft perfusion level, no clamping effect could be detected. Compared with the control group, perfusion was significantly higher in the transplanted airways intraoperatively, at 71 versus 55 arbitrary perfusion units (p < 0.01). Postoperative transplant airway perfusion values were not significantly different from the intraoperative value. The coefficient of variation of repeated measurements was higher in the transplanted airways, with a coefficient of variation of 0.22 versus 0.17 in the control group (p < 0.01), indicating heterogeneous transplant airway perfusion. Conclusions. This study has documented that revascularization with the internal thoracic artery supplies the transplanted airway with additional oxygenated blood. (Ann Thorac Surg 1997;63:790 –5) © 1997 by The Society of Thoracic Surgeons

I

airway healing as well [3, 6–9]. The transplanted airways are totally dependent on collateral perfusion from the pulmonary circulation when the bronchial arteries are not reconnected at the time of transplantation. Mills and co-workers [5] showed that reconstruction of the bronchial arteries in dogs undergoing lung transplantation resulted in a considerable reduction of airway complications and suggested that attempts should be made to reconstruct the bronchial arteries in humans. Couraud and co-workers [10] used a saphenous vein graft to the donor bronchial arteries and implanted the vein graft into the recipient aorta. At Harefield Hospital, a technique of direct revascularization of a donor bronchial artery by anastomosis with the recipient left internal thoracic artery (ITA) was developed [11]. Laser Doppler flowmetry (LDF) is a method that allows endoscopic measurements of microvascular perfusion. Previously, we evaluated this method for endoscopic measurements of perfusion in human airways [12]. The aim of this study was to assess the effect of bronchial artery revascularization with ITA on airway perfusion in DLTx, applying endoscopic LDF.

solated lung or combined heart-lung transplantation has become an established treatment for patients with end-stage pulmonary disease. Infections, acute rejections, and ischemic airway complications are presently the most important causes of early death, whereas obliterative bronchiolitis is the major threat to long-term survivors. Airway stenosis, necrosis, and dehiscence are still major causes of early morbidity and mortality [1–3]. Whereas airway complications have become less frequent in single-lung and combined heart-lung transplantation (5% to 20%), they occur in up to 60% of en bloc double-lung transplants with tracheal anastomosis (DLTx), resulting in an early mortality of up to 25% [2, 4, 5]. Ischemia of the donor airway has been suggested to be the most important cause of bronchial complications, although rejection, immunosuppression, and infections may compromise

Accepted for publication Oct 22, 1996. Address reprint requests to Dr Sundset, Department of Medicine, Aker Hospital, N-0514 Oslo, Norway.

© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc

0003-4975/97/$17.00 PII S0003-4975(96)01273-8

Ann Thorac Surg 1997;63:790 –5

Material and Methods Seven patients undergoing en bloc double-lung transplantation at Harefield Hospital were studied. There were 2 women and 5 men with a mean age of 45 years (range, 32 to 53 years). The indications for transplantation were a-1-antitrypsin deficiency in 4 patients, chronic obstructive airways disease in 1, sarcoidosis in 1, and primary pulmonary hypertension in 1 patient. A control group of 16 patients, 3 women and 13 men with a mean age of 58 years (range, 41 to 72 years), operated for coronary heart disease were investigated during general anesthesia before and after cardiopulmonary bypass.

Revascularization Procedure The technique of organ harvesting and procurement has been described previously [11]. Briefly, the lungs were harvested with the donor aorta with the orifices of the bronchial arteries en bloc, together with the mediastinal tissue anteriorly to the spine, leaving the bronchial arteries intact. Further dissection of the organs was performed after arrival at our institution. The aorta was opened longitudinally just anterior to the left intercostal arteries, and the bronchial artery orifices were identified by probing. The largest bronchial vessel passing toward the main carina (usually the right intercostobronchial artery) was marked with a suture. After reestablishment of the pulmonary circulation, back-bleeding through bronchopulmonary communications confirmed the identification of the bronchial artery. The recipient’s pedicled left internal thoracic artery was anastomosed to the origin of the bronchial artery on the donor’s descending aorta. A button of donor aorta was left for protection of the bronchial arteries. No vasodilators or vasopressors were administered intraoperatively to our patients.

Laser Doppler Flowmetry Laser Doppler flowmetry allows continuous measurements of microvascular perfusion [13]. Monochromatic light from a low-energy helium-neon laser is carried by an optical fiber probe and undergoes multiple scattering and absorption in a tissue volume of a few cubic millimeters. When the photons are scattered by moving blood cells, their frequency is changed (Doppler shifted) and the magnitude of this Doppler shift is dependent on the product of the number and mean velocity of the moving blood cells (cell flux). Scattered light is guided back to the instrument for signal processing, where the Doppler shifted part is isolated and converted into a voltage output signal that is linearly correlated to cell flux [13]. Spatial variation, defined as the LDF measurement variation of repeated measurements in adjacent tissue volumes, is mainly caused by the heterogeneity of the microvascular architecture and the small sampling volumes of LDF [14]. The spatial variation was expressed as the coefficient of variation (5 standard deviation/mean). The laser Doppler equipment used in this study consisted of a 2-mW helium-neon laser (PeriFlux PF3, Perimed, Farfalla, Sweden), operating at a wavelength of

SUNDSET ET AL LUNG TRANSPLANTATION: REVASCULARIZATION

791

632.8 nm. The instrument was calibrated as recommended by the manufacturer. All measurements were performed with a 4-kHz bandwidth filter and the time constant was set to 0.2 s. The laser probe PF309s has a fiber separation of 0.7 mm and a diameter of 2.2 mm. Perfusion values were expressed as arbitrary perfusion units (PU, 100 PU defined as 1 V flowmeter output) and recorded with a linear recorder (ABB SE 120, ABB, Sweden).

Measuring Procedures Bronchoscopy (Olympus 1T10 or 1T20) and endoscopic LDF measurements were performed through the tracheal tube intraoperatively after cardiopulmonary bypass and 12 to 36 hours postoperatively before extubation. All measurements were performed during assisted ventilation without the use of positive end-expiratory pressure. At each location, five replicate LDF measurements of at least 20-second duration were recorded with the endoscopic light switched off to avoid interference with the laser photons [12]. The following locations were investigated: recipient trachea (anterior intercartilaginous part 1 to 2 cm above the anastomosis) and the left and right upper lobe bronchus of the graft, less than 5 mm distal to the bifurcation from the main bronchi. Intraoperatively, continuous measurements were performed during repeated clamping of the attached ITA. Healing of the anastomosis was assessed by additional bronchoscopies after 1 to 2 weeks, 4 weeks, and 3 months. In the control group measurements were performed before cardiopulmonary bypass. Additional measurements were performed at the main carina (in the ventral transition to the trachea) before and approximately 15 to 20 minutes after coming off cardiopulmonary bypass to evaluate effects of extracorporeal circulation.

Angiographic Investigations Angiographic assessment of the patency of the ITA graft and bronchial artery reperfusion was performed 2 to 4 weeks postoperatively.

Statistical Methods Results were accepted as significant at a p value of less than 0.05. Comparisons between the DLTx group and the control group were performed with the nonparametric Mann-Whitney test, and the Wilcoxon matched pairs test was used to assess differences within groups. Spatial variation, defined as the measurement variation of replicate measurements within the same measuring region, was expressed as the coefficient of variation of replicate measurements (5 standard deviation/mean).

Results In total, more than 300 recordings were performed in the 7 patients undergoing DLTx. The central airways sometimes had to be cleared of blood and secretions that limited sight by using sterile saline solution for rinsing and suction. In most patients stable recordings with a favorable signal-to-noise ratio were easily obtained.

792

SUNDSET ET AL LUNG TRANSPLANTATION: REVASCULARIZATION

Ann Thorac Surg 1997;63:790 –5

Table 1. Airway Perfusion Before and After Cardiopulmonary Bypassa Variables Before CPB After CPB p Value

Perfusion (PU)

Spatial Variation (Cv)

Hb (g/100 mL)

BP (mm Hg)

60 (55–75) 60 (57.5–70) 0.95

0.10 (0.07– 0.21) 0.08 (0.06 – 0.15) 0.35

13.5 (12.3–14.7) 10.8 (9.9 –11.3) ,0.001

80 (72– 89) 75 (63– 88) 0.06

a Investigations at the main carina before and after cardiopulmonary bypass in 13 patients undergoing coronary artery bypass grafting. Values are given as medians with 95% confidence limits.

BP 5 blood pressure;

CPB 5 cardiopulmonary bypass;

Cv 5 coefficient of variation;

Hb 5 hemoglobin;

PU 5 arbitrary perfusion units.

Thirteen patients in the control group were investigated before and after cardiopulmonary bypass. No change in perfusion level or spatial variation could be detected after cardiopulmonary bypass in spite of significant hemodilution (Table 1). In the trachea (native airway in the DLTx group) perfusion values were similar in the DLTx and control groups (p 5 0.6) (Fig 1). The perfusion was significantly higher in the transplanted airways (upper lobe bronchi) in comparison with the upper lobe bronchi in the control group (p , 0.01). In the control group, perfusion seemed slightly higher in the trachea compared with the upper lobe bronchi, although these data did not quite reach a significant level (p 5 0.09). There was no difference in perfusion level between the trachea and upper lobe bronchi in the DLTx patients (p 5 0.6). The effect of clamping of the attached ITA varied considerably among the patients, with 1 patient showing a reduction of 60% in the left upper lobe (Fig 2). At the main carina, clamping reduced the LDF signal in all 3 patients where clamping was attempted (Fig 3). A clamping effect could only be demonstrated in 2 patients in the right upper lobe bronchus. In the 2 patients with the

highest perfusion values, no clamping effect could be demonstrated. The postoperative investigations 12 to 36 hours after transplantation did not reveal significant changes in airway perfusion level compared with intraoperative measurements (Table 2). Spatial variation in the trachea (native airways) was unaffected by the transplant procedure, cardiopulmonary bypass, and 12 to 36 hours of ventilator treatment (Fig 4). Intraoperative spatial variation was significantly higher in the upper lobe bronchi (transplanted airways) compared with the controls (p , 0.01). The transplanted airways seemed to have a higher spatial variation intraoperatively compared with the native airways, although these data only reached borderline significance (p 5 0.056). Even in the controls, spatial variation was higher in the upper lobe bronchi than in the trachea (p 5 0.05). Postoperatively, spatial variation of the transplanted airways showed a greater diversity among the patients, but was not significantly different from the controls or the intraoperative data. Two of the patients died of sepsis and multiorgan failure after 1 week. In the 5 remaining patients the observation period was from 3 to 52 months. One patient died after 3 months of hepatitis C and sepsis; the remaining 4 patients are alive and well more than 4 years later. All patients had excellent anastomosis healing with no macroscopic signs of necrosis or stenosis at follow-up

Fig 1. Perioperative perfusion values in the double-lung transplant (DLTx) group compared with the group of patients with coronary heart disease, given as arbitrary perfusion units (PU). The horizontal lines indicate median values. Measurements in the trachea were performed in the native airways in the double-lung transplant group.

Fig 2. Laser Doppler recording in the left upper lobe. The recorded curve shows fluctuations synchronous to pulse (small “spikes”) and respiration (larger “waves”). Repeated clamping (arrows) of the attached internal thoracic artery induced a substantial reduction in the laser Doppler flowmetry output signal. (PU 5 arbitrary perfusion units.)

Ann Thorac Surg 1997;63:790 –5

SUNDSET ET AL LUNG TRANSPLANTATION: REVASCULARIZATION

793

Fig 3. Effect of clamping of the attached internal thoracic artery on transplant airway perfusion. Each cross indicates the clamping effect in each patient. The reduction of the laser Doppler flowmetry signal during clamping is given in percent of baseline before clamping.

bronchoscopies. Angiograms performed 2 to 4 weeks postoperatively showed patent ITA grafts in all 5 patients who survived more than 1 week. A distal runoff through the bronchial arteries was evident in all patients, although the degree of bronchial blood flow varied considerably. Autopsy also showed patent ITA grafts in the 2 patients who died 1 week after transplantation.

Comment The first 3 patients undergoing DLTx at Harefield Hospital in 1988 suffered from severe airway complications, which initiated the development of the bronchial artery revascularization technique [11]. The bronchial arteries supply the bronchi and the lower part of the trachea [5]. The bronchial microcirculation also receives blood from the pulmonary circulation through peripheral anastomoses on both sides of the alveolar vessel bed [15, 16]. Barman and colleagues [17 demonstrated in dogs that the main bronchi receives approximately equal portions of blood from the systemic and pulmonary circulation, whereas blood flow to the

upper two-thirds of the trachea is derived mainly from the systemic circulation [16, 17]. Yokomise and colleagues [18] applied LDF to assess retrograde bronchial blood flow from the pulmonary circulation in an isolated in situ model in dogs. In the central airways, a reduction in perfusion of 36% to 55% was observed after systemic flow was interrupted. Our findings confirm that the pulmonary contribution to central airway perfusion is substantial. We were able to demonstrate that in most patients revascularization with the ITA supplied the transplanted airways with additional blood. We found an increased heterogeneity of perfusion in the transplanted airways intraoperatively. Postoperatively, this finding was more diverse, with some patients showing an extreme perfusion heterogeneity. None of our patients showed any macroscopic signs of airway necrosis or development of airway complications at endoscopic investigations during the follow-up period. The validity of the laser Doppler technique has been evaluated in many experimental studies of microvascular perfusion in various organs. Laser Doppler flowmetry correlates well with blood flow measured with other

Table 2. Intraoperative and Postoperative Investigations of Double-Lung Transplant Patients With Internal Thoracic Artery Revascularizationa Variables Intraoperative (n 5 7) Postoperative (n 5 6) p Value

Trachea

Left Upper Lobe Bronchus

Right Upper Lobe Bronchus

Hb (g/100 mL)

BP (mm Hg)

85 (57.5–90) 67.5 (40 –100) NS

72.5 (57.5–112.5) 67.5 (47.5–105) NS

70 (35–95) 57.5 (42.5– 85) NS

10.9 (7.9 –11.5) 12.8 (9.2–17.1) NS

68 (54 – 80) 70 (49 –100) NS

a Central airway perfusion, hemoglobin, and blood pressure intraoperatively and 12 to 36 hours after transplantation, expressed as medians with total range. Perfusion values are given in arbitrary perfusion units (PU).

BP 5 blood pressure;

Hb 5 hemoglobin;

PU 5 arbitrary perfusion units.

794

SUNDSET ET AL LUNG TRANSPLANTATION: REVASCULARIZATION

Fig 4. Spatial variation, expressed as the coefficient of variation of repeated measurements in each region. The horizontal lines indicate median values. (DLTx) 5 double-lung transplant group.)

techniques [13, 19, 20]. The major advantage of the technique is that continuous measurements can be obtained noninvasively. We have evaluated the applicability of this method in human airways [12]. Other investigators have applied laser Doppler for bronchial LDF measurements in animals [18, 21]. Patients operated on with coronary artery bypass grafting for ischemic heart disease were chosen as a control group. For practical reasons, it was not possible to investigate the control group at all locations both before and after cardiopulmonary bypass, but no effects of cardiopulmonary bypass on perfusion could be demonstrated (see Table 1). The DLTx patients were investigated both after cardiopulmonary bypass (intraoperatively) and after 12 to 36 hours of ventilator treatment (postoperatively). The similar perfusion levels and spatial variations in the trachea of the DLTx group both intraoperatively and postoperatively and in the control group indicate that the data from the upper lobe bronchi could also be compared. An intraoperative reperfusion hyperemia in the transplanted airways was indicated by a significantly higher perfusion in the upper lobe bronchi in the DLTx group, whereas perfusion in the trachea was similar in both groups (see Fig 1). The postoperative investigations did not reveal significant alterations in perfusion level compared to the intraoperative measurements, although there was a tendency toward lower perfusion levels at all locations (see Table 2). Development of infections or rejections, the possible influence of 12 to 36 hours of ventilator treatment, and the diversity of the results in this small group of patients call for caution in the interpretation of the postoperative data. In the left upper lobe, clamping induced a reduction of the LDF signal of 10% to 60% in 5 patients. In 3 patients, LDF recordings were performed in the lower trachea at the ventral transition to the main carina during clamping. This location is closer to the airway anastomosis than the upper lobe bronchi and, therefore, probably better suited for assessment of perfusion at the anastomosis. Instrumentation was, however, more difficult in this area as

Ann Thorac Surg 1997;63:790 –5

care had to be taken that the bronchoscope did not bend against and injure the tracheal anastomosis. All 3 patients showed a clamping response, ranging from 10% to 42%. In the right upper lobe bronchus, a clamping effect could only be detected in 2 patients. The investigations in the right upper lobe bronchus were always carried out after the measurements in the left upper lobe and at the carina, and the surgeon frequently observed that the ITA was becoming increasingly spastic after repeated clamping toward the end of the procedure. Therefore, we are unable to conclude whether the lesser clamping effect was attributable to this spastic vasoconstriction of the ITA or to a lesser effect of the revascularization in the right upper lobe bronchus. The 2 patients without detectable effect of revascularization had the highest transplant airway perfusion values of all our patients, indicating an extensive collateral perfusion from the pulmonary vasculature. The pulmonary collateral blood supply is clearly important, but differently developed in individuals. A varying clamping effect could be expected, as not all patients have anastomotic complications after lung transplantation, even when no revascularization is performed. Other investigators have demonstrated that ischemia of both human skin and bowel enhances perfusion heterogeneity [14, 22]. Spatial variation, defined as the variability of replicate LDF recordings performed in adjacent tissue volumes, was significantly higher in the transplanted airways intraoperatively compared with the controls, indicating heterogeneous perfusion in the transplanted airways. Mechanical and ischemic trauma as well as immunologic events may have caused a maximum dilatation of some microvessels and occlusion of others by leukocyte plugs [23]. We also demonstrated a greater diversity in spatial variation postoperatively, with some patients showing double or triple values as compared with intraoperative values, although there was no significant change in the group as a whole (see Fig 4). Developing infections, rejections, and pressor support might contribute to this diversity. Without revascularization, the donor airway is totally dependent on collateral circulation from the pulmonary vasculature, which partly delivers desaturated blood. Cold ischemia of the graft during transport with subsequent release of vasoactive mediators would cause a general vasodilation not only of nutritional vessels, but of nonnutritional sinuses and microvascular shunts as well. The human airways are extremely rich in nonnutritive, deeper-lying venous plexuses in the submucosa and peribronchially. High LDF perfusion values do not necessarily imply an adequate nutritional perfusion, as shunt flow is also recorded with this technique. Although there is a gradual ingrowth of systemic vessels to the anastomosis area during the first 2 to 4 weeks after transplantation, the central airway is vulnerable to ischemia until the systemic bronchial circulation is reestablished [3, 7]. During this vessel proliferation period the lack of systemic blood supply endangers anastomosis healing. Hence, we suggest that it is in this early period of spontaneous revascularization that a systemic graft to the

Ann Thorac Surg 1997;63:790 –5

bronchial arteries is of importance to airway viability. Techniques like wrapping of the airway anastomosis with omentum or an intercostal pedicle flap, telescoping of the anastomosis, and improvements in graft preservation have been introduced to reduce airway ischemia [1, 4, 7, 24]. None of these techniques increases airway perfusion in the first few days after transplantation, although wrapping shortens the vessel proliferation period as an ingrowth of capillaries can be demonstrated as early as 4 days after transplantation [7]. As anastomotic complications are extremely frequent in DLTx, this transplant procedure has been suggested to be replaced with sequential bilateral lung transplantation with bronchial anastomosis or with heart-lung transplantation [1]. It is generally assumed that the lower incidence of airway complications after heart-lung transplantation is attributable to the preservation of the coronary– bronchial anastomoses and other systemic vessels that supply the central airways. Our results indicate that revascularization with the ITA enhances transplant airway perfusion in the first critical period after transplantation. Thus, bronchial artery revascularization might reduce ischemic airway complications after lung transplantation. This study was supported by the Pulmonary Research Fund of Glaxo Norway and the Perimed AB, Sweden.

References 1. Haydock DA, Trulock EP, Kaiser LR, et al. Lung transplantation. Analysis of thirty-six consecutive procedures performed over a twelve-month period. J Thorac Cardiovasc Surg 1992;103:329– 40. 2. Patterson GA, Todd TR, Cooper JD, Pearson FG, Winton TL, Maurer J, Toronto Lung Transplant Group. Airway complications after double lung transplantation. J Thorac Cardiovasc Surg 1990;99:14–22. 3. Scha¨fers HJ, Haverich A, Wagner TOF, Wahlers T, Alken A, Borst HG. Decreased incidence of bronchial complications following lung transplantation. Eur J Cardio-thorac Surg 1992;6:174–9. 4. Griffith BP, Magee MJ, Gonzales IF, et al. Anastomotic pitfalls in lung transplanatation. J Thorac Cardiovasc Surg 1994;107:743–54. 5. Mills NL, Boyd AD, Gheranpong C, Spencer FC. The significance of bronchial circulation in lung transplantation. J Thorac Cardiovasc Surg 1970;60:866–78. 6. Pinsker KL, Koerner SK, Kamholz SL, Hagstrom JWC, Veith FJ. Effect of donor bronchial length on healing. A canine model to evaluate bronchial anastomotic problems in lung transplantation. J Thorac Cardiovasc Surg 1979;77:669–73. 7. Lima O, Goldberg M, Peters WJ, Ayabe H, Townsend E,

SUNDSET ET AL LUNG TRANSPLANTATION: REVASCULARIZATION

8.

9. 10. 11.

12. 13. 14. 15. 16. 17. 18.

19.

20. 21.

22.

23. 24.

795

Cooper JD. Bronchial omentopexy in canine lung transplantation. J Thorac Cardiovasc Surg 1982;83:418–21. Morgan E, Lima O, Goldberg M, Ayabe H, Ferdman A, Cooper JD. Improved bronchial healing in canine left lung reimplantation using omental pedicle wrap. J Thorac Cardiovasc Surg 1983;85:134–9. Scha¨fers HJ, Haydock DA, Cooper JD. The prevalence and management of bronchial anastomotic complications in lung transplantation. J Thorac Cardiovasc Surg 1991;101:1044–52. Couraud L, Baudet E, Martigne C, et al. Bronchial revascularization in double-lung transplantation: a series of 8 patients. Ann Thorac Surg 1992;53:88–94. Daly RC, Tadjkarimi S, Khaghani A, Banner NR, Yacoub MH. Successful double-lung transplantation with direct bronchial artery revascularization. Ann Thorac Surg 1993;56: 885–92. Sundset A, Hansen G, Haanaes OC, Line PD, Kvernebo K. Human bronchial perfusion evaluated with endoscopic laser Doppler flowmetry. Int J Microcirc Clin Exp 1993;13:233– 45. Nilsson GE, Tenland T, Oeberg PA. Evaluation of a laser Doppler flowmeter for measurements of tissue blood flow. IEEE Trans Biomed Eng 1980;27:597– 604. Line PD, Mowinckel P, Lien B, Kvernebo K. Repeated measurement variation and precision of laser Doppler flowmetry measurements. Microvasc Res 1992;43:285–93. Modell HI, Beck K, Butler J. Functional aspects of canine bronchial–pulmonary communications. J Appl Physiol 1981; 50:1045–51. Ladowski JS, Hardesty RL, Griffith BP. The pulmonary artery blood supply to the supracarinal trachea. J Heart Transplant 1984;4:40–2. Barman SA, Ardell JL, Parker JC, Perry ML, Taylor AE. Pulmonary and systemic blood flow contributions to upper airways in canine lungs. Am J Physiol 1988;255:1130–5. Yokomise H, Cardoso PFG, Kato H, et al. The effect of pulmonary arterial flow and positive end-expiratory pressure on retrograde bronchial mucosal blood flow. J Thorac Cardiovasc Surg 1991;101:201– 8. Ahn H, Lindhagen J, Nilsson GE, Salerud G, Jodal M, Lundgren O. Evaluation of laser Doppler flowmetry in the assessment of intestinal blood flow in cat. Gastroenterology 1985;88:951–7. Kvietys PR, Shepherd AP, Granger DN. Laser-Doppler, H2 clearance, and microsphere estimates of mucosal blood flow. Am J Physiol 1985;249:G221–7. Godden DJ, Baile EM, Pare PD. A comparison of laser Doppler flowmetry with the radiolabeled microsphere reference flow technique to measure tracheal blood flow in dogs. Acta Physiol Scand 1991;142:49–57. Krogh-Sørensen K, Line PD, Haaland T, Horn RS, Kvernebo K. Intraoperative prediction of ischaemic injury to the bowel: a comparison of laser Doppler flowmetry and tissue oximetry to histological analysis. Eur Vasc Surg 1992;6:518–24. Engler RL, Scmid-Sho¨nbein GW, Pavelec RS. Leukocyte capillary plugging in myocardial ischemia and reperfusion in the dog. Am J Pathol 1983;111:98 –111. Calhoon JH, Grover FL, Gibbons WJ, et al. Single lung transplantation. Alternative indications and technique. J Thorac Cardiovasc Surg 1991;101:816–25.