Low-Potassium Solution for Lung Preservation in the Setting of High-Flow Reperfusion Nuno F. DeLima, MD, Oliver A. R. Binns, MD, Scott A. Buchanan, MD, Jeffrey T. Cope, MD, Michael C. Mauney, MD, Kimberly S. Shockey, MS, Curtis G. Tribble, MD, and Irving L. Kron, MD Division of Thoracic and Cardiovascular Surgery, Department of Surgery, University of Virginia Health Sciences Center, Charlottesville, Virginia

Background. We previously demonstrated that stan- dard preservation using Euro-Collins solution impairs lung function in the setting of high-flow reperfusion because of potassium-induced vasoconstriction. Preser- vation strategies for single-lung transplantation are an important factor in patients with pulmonary hyperten- sion. This study investigates the hypothesis that low- potassium preservation solution will improve function of lungs subjected to high-flow reperfusion.

Methods. Twenty-one New Zealand white rabbit lungs were harvested and studied on an isolated, blood- perfused model of lung function after 4 hours of cold ischemia at 4°C. Control lungs were preserved with 50 mL/kg of cold saline solution flush (group I). Experi- mental lungs were preserved with low-potassium solu- tion (group II) or Euro-Collins solution (group III) at similar temperatures and volumes.

Results. The pulmonary arteriovenous oxygen gradient at the end of the 30-minute high-flow reperfusion period

was significantly higher in group II compared with group III (121.3 + 19.2 mm Hg versus 31.1 ± 4.2 mm Hg; p < 0.001). The pulmonary vascular resistance was signifi- cantly lower in group II than in group III (46.3 ± 1.8 x 103 dynes • s • cm -~ versus 79.8 +-. 8.4 × 103 dynes • s • cm-5; p < 0.01). The percent decrease in dynamic airway compliance in group III was significantly greater than in groups I and II (-51.0% ± 13.3% versus -10.2% + 3.4% and -11.2% --- 2.8%, respectively; p < 0.001). Similarly, the wet to dry ratio of the lungs in group III was significantly greater than in groups I and II (13.9 ± 2.3 versus 5.9 ± 0.2 and 6.0 + 0.4, respectively; p < 0.001).

Conclusions. These data demonstrate that a low- potassium preservation solution yields improved lung function after high-flow reperfusion in an ex vivo rabbit lung model. Lung preservation should be aimed at the clinical setting.

(Ann Thorac Surg 1996;61:973-6)

D espite a strong body of research demonstrating the superiority of and potential for prolonged clinical

lung preservation with low-potassium solutions [1-6], the intracellular-type Euro-CoUins solution remains the current clinical standard for lung graft preservation. Although Euro-Collins solution has been successful in most instances clinically, it has not been satisfactory for longer ischemic storage and single-lung transplantation for pulmonary hypertension [7, 8]. Therefore, alternative preservation solutions and methods tailored for specific clinical scenarios may be necessary to achieve optimal lung function after transplantation.

We [9] have previously demonstrated that high-flow reperfusion results in substantial impairment of lung function and that the high potassium concentration of Euro-Collins solution further potentiates the lung injury. The potent pulmonary vasoconstriction induced by the high potassium content of this intracellular solution severely increases the pulmonary vascular resistance (PVR) at high-flow reperfusion, thus leading to edema

Accepted for publication Nov 13, 1995.

Address reprint requests to Dr Kron, Division of Thoracic and Cardio- vascular Surgery, Department of Surgery~ University of Virginia Health Sciences Center, Box 310, Charlottesville, VA 22908.

formation and lung dysfunction. We now hypothesize that pulmonary function after high-flow reperfusion would be better preserved with a low-potassium solu- tion, thereby supporting the use of extracellular-type solutions as the preservation method of choice for single- lung transplantation in the setting of pulmonary hyper- tension. This hypothesis was investigated in isolated, blood-perfused rabbit lungs after 4 hours of cold isch- emia.

Material and Methods

Lung-Heart Block Harvesting Twenty-one New Zealand white rabbits weighing 3.0 to 3.5 kg were randomized to three groups of 7 animals each. All experimental protocols were reviewed and approved by an institutional animal use committee. All animals received humane care in compliance with the "Guide for the Care and Use of Laboratory Animals'" published by the National Institutes of Health (NIH publication 85-23, revised 1985).

Each rabbit was anesthetized with intramuscular ad- ministration of ketamine hydrochloride (50 mg/kg) and xylazine (5 mg/kg). A tracheostomy was performed fol- lowed by induction of paralysis with metocurine iodide

© 1996 by The Society of Thoracic Surgeons 0003-4975/961515.00 Published by Elsevier Science Inc SSDI 0003-4975(95)01135-8

974 DELIMA ET AL Ann Thorac Surg LOW-POTASSIUM SOLUTION FOR LUNG PRESERVATION 1996;61:973-6

(0.2 mg/kg). Mechanical venti lat ion was inst i tuted (ven- t i lator model RSP1002; Kent Scientific Corporation, Litch- field, CT) using room air with a t idal volume of 12 mL/kg and a rate of 20 breaths/rnin. Median s ternotomy and thymectomy were then performed. The super ior and inferior venae cavae were loosely encircled with liga- tures, and the per icard ium was opened. Both the pu lmo- nary ar tery (PA) and aorta were dissected free and similarly encircled. A purses t r ing suture was placed in the free wall of the right ventricle, and the rabbi t was hepar in ized (500 U/kg). After injection of 30 ~g of pros- t a g l a n d i n E 1 (Alp ros t ad i l ; The U p j o h n C o m p a n y , Kalamazoo, MI) into the PA, the venae cavae were ligated, and the t ime of onset of ischemia was charted. The PA was then cannula ted through the right ventr icu- lar pursestr ing. The cannula was secured by tying both the right ventr icular and PA ligatures.

After the left ventricle was vented and the aorta was ligated, 50 mL/kg of preservat ion solution at 4°C was infused into the PA from a height of 30 cm. Topical cooling was achieved with cold saline slush. Dur ing PA flush, the left a t r ium was cannula ted through the left ventricle and a second purses t r ing t ied a round the can- nula. After the PA flush, the inflow and outflow cannulas were clamped. Care was taken to leave the p leurae intact until complet ion of the flush to avoid parenchyrnal in- jury. The t r acheos tomy tube was c l amped at end- inspirat ion, and the lung-hear t block was excised, im- mersed in cold normal saline solution, and s tored at 4°C.

Assessment of Lung Function

After 4 hours of storage, the lung-hear t blocks were suspended from a force t ransducer in a warmed, humid- ified t issue chamber. Venti lat ion was rees tabl ished with 95% oxygen and 5% carbon dioxide at a t idal volume of 12 mL/kg and a rate of 20 breaths/rain . The lungs were reperfused with homologous, fresh whole venous blood from a main reservoir. A second venous blood reservoir was used to de te rmine s ingle-pass oxygenation. With care taken to avoid the introduct ion of air bubbles , the inflow and outflow cannulas were connected to the blood-f i l led perfusion circuit. The circuit (Kent Scientific Corporat ion) was des igned to recirculate 200 mL of wa rmed blood through a 270-p,m blood filter (model 2C7600; Baxter, Deerfield, IL) using a roller pump (model 7521-40; Cole Palmer Ins t rument Company, Chicago, IL) at a rate of 120 mL/min in accordance with the high-flow exper imenta l protocol.

Cont inuous recording of PA pressure, pu lmonary ve- nous pressure, lung weight, a irway flow, and airway p r e s s u r e was ca r r i ed out us ing a d y n a m i c da ta - acquisi t ion p rogram (Workbench PC; Strawberry Tree, Inc, Sunnydale , CA) run on a desktop computer (model 470A; Compaq Prolinea, Houston, TX). This p rogram al lowed immedia te calculation of PVR, t idal volume, and dynamic airway compliance. The pu lmonary venous pressure was main ta ined within the physiologic range (4 to 8 m m Hg) by setting the appropr ia te height of a small outflow reservoir in the circuit. Pu lmonary venous blood samples were collected for b lood gas analysis (Corning

Force transducer

Roller Pump

P

Ventilator

_t

Reservoir

Main Blood Reservoir Flow: 60 or 120 ml/min

Blood Gas Sampling Port

Fig 1. Isolated, ventilated, blood-perfused rabbit lung model. (P = pressure transducer.)

178 pH/b lood gas analyzer, Medfield, MA) at 10, 20, and 30 minutes after the start of reperfusion; at each sampl ing time, inflow from the main reservoir was in te r rup ted and the circuit filled with venous blood from the second inflow reservoir. Oxygen contact with exposed blood surfaces inside the reservoir containers was minimized by continuous passive infusion of 100% nitrogen. After 30 minutes of reperfusion, lung samples were taken for histologic analysis and wet to dry weight ratio calculation after passive desiccation (Fig 1).

Experimental Protocol Lungs were f lushed with normal saline solution (group I), low-potass ium solution (group II), or Euro-Coll ins solu- tion (group III) and reperfused at a high flow rate of 120 mL/min. The exper imenta l protocol and the composi t ion of flush solutions are shown in Table 1. An addi t ional control group was des igned to demonst ra te basel ine lung function. These lungs were f lushed with saline solution and s tudied immedia te ly after harvesting, thus not being subjected to 4 hours of cold ischemia. Saline solution was chosen to flush the lungs of the control groups because of its inert endothel ia l effects and lack of preservat ion propert ies. Data were obta ined every 15 seconds and analyzed at the end of the 30-minute reperfusion period.

Table 1. Experimental Protocol and Composition of Flush Solutions

Group I Group II Group III Component (S; n = 7) (LP; n = 7) (EC; n = 7)

Na + (retool/L) 154 95 10 K + (retool/L) . . . 5 115 C1 (mmol/L) 154 5 15 HCO 3 (mmol/L) . . . . . . 10 HPO4 2 (retool/L) .- . 45 42.5 H2PO 4 (mmol/L) .- . 5 15 Glucose (mrnol/L) .- . 214 214 pH (4°C) 6.6 7.55 7.25

EC = Euro-Collins solution; LP = low-potassium solution; S = saline solution.

Ann Thorac Surg DELIMA ET AL 975 1996;61:973-6 LOW-POTASSIUM SOLUTION FOR LUNG PRESERVATION

P A-V Oxygen Gradient in mmHg 250

200

150

100

50

0 10 20 30

Reperfusion time in minutes

Groups are in the following order: (EC) ==Group (S) C3Group II (LP) ==Group III

Fig 2. Pulmonary arteriovenous (P A-V) oxygen gradient at 10, 20, and 30 minutes of reperfusion. Oxygenation declined significantly in group I lungs over the 30 minutes but was already low at 10 min- utes of reperfusion in group Ill lungs. At 30 minutes, the oxygen- ation of group II lungs was significantly higher than that of group III lungs (p < 0.001). Data are shown as the mean ± the standard error of the mean. (EC = Euro-Collins solution; LP = low-potas- sium solution; S = saline solution.)

S ta t i s t i ca l A n a l y s i s

Statistical analysis was performed by analysis of variance and Kruskal-Wallis nonparametr ic analysis of variance to compare the groups of stored lungs. Differences were considered significant if the p value was less than 0.05. All values are expressed as the mean _+ the standard error of the mean.

Results

There were no significant differences in donor weight, total ischemic time, or perfusate hematocrit between the groups. Mean hematocrit for all groups combined was 30.5% _+ 0.4%. The arteriovenous oxygen gradient at the end of the 30-minute reperfusion period was significantly higher in the lungs preserved with low-potassium solu- tion (group II) compared with lungs preserved with Euro-Collins solution (group III) (group II, 121.3 + 19.2 mm Hg, versus group III, 31.1 -Z_ 4.2 mm Hg; p < 0.001). In group I (saline solution), the oxygen gradient was 75.8 + 13 mm Hg, an intermediate value compared with the other groups and not reaching significance.

During reperfusion, lungs in group I showed a progres- sive and steeper decrease in oxygenation compared with group II, a finding probably related to poor lung preser- vation by saline solution. However, in group III, the arteriovenous oxygen gradient was already extremely low at 10 minutes of reperfusion, and this represented early lung damage (Fig 2).

In this model of high-flow reperfusion, all three groups displayed pulmonary hypertension. However, the PA pressure and PVR at the end of the 30-minute reperfu- sion period were significantly lower in group II than in group III (p < 0.01). The percent decrease in dynamic airway compliance was significant in group III (p < 0.001). Similarly, the lung water content at 30 minutes of reperfusion, expressed as the wet to dry ratio, was significantly greater in group III (p < 0.001). These data and the immediate control group results are shown in Table 2.

C o m m e n t

Graft dysfunction after s ingle-lung transplantat ion is more common in patients with pulmonary hypertension [8, 10]. In these patients, perfusion lung scans performed within 48 hours after t ransplantat ion demonstrated that 95% -+ 7% of blood flows through the t ransplanted lung compared with only 60% _+ 20% in patients with obstruc- tive lung disease and 69% -+ 17% in patients with idio- pathic pulmonary fibrosis [10]. Results of experimental single-lung transplantat ion have also been shown to be dependent on the severity of pulmonary hypertension. Graft pulmonary edema resulted as a consequence of severe shunt ing of blood through the t ransplanted lung in the setting of pulmonary hypertension [11]. We [9] have shown that high-flow reperfusion results in sub- stantial impai rment of lung function and that Euro- Collins preservation further potentiates this lung dys- function. Previous experimental evidence suggests that the intracellular-type Euro-Collins solution causes dam- age during the flush and storage phase of lung preserva- tion and during reperfusion [2, 9]. The high potassium concentration causes vasoconstriction, thereby causing nonhomogeneous flushing of the graft and leaving areas of parenchyrna poorly preserved. In addition, this potas-

Table 2. Results Af ter 30 Minutes o f Reperfusion ~

PVR Group PAP (ram Hg) (x103 dynes • s • cm -s) CPL (%A) Wet/Dry Ratio AV 02 (mm Hg)

I 110.5 ± 11.6 68.2 + 7.6 -10.2 ± 3.4 5.9 ± 0.2 75.8 ± 13 II 77.7 ± 2.3 b 46.3 ± 1.8 b -11.2 ± 2.8 6.0 ± 0.4 121.3 ± 19.2 d III 127.1 ± 19.9 79.8 ± 8.4 -51.0 _+ 13.3 c 13.9 ± 2.3 c 31.1 ± 4.2 IMM 87.0 ± 5.0 53.6 ± 3.4 -10.3 + 3.8 6.0 ± 0.2 243.7 + 40.2

a Data are shown as the mean -+ the standard error of the mean. b Significance: p < 0.01, II versus III. c Significance: p < 0.001, III versus II and I. d Significance: p < 0.001, II versus Ill.

AV 02 = arteriovenous oxygen gradient; CPL = dynamic airway compliance; IMM = immediate control group; PAP = pulmonary artery pressure; PVR = pulmonary vascular resistance; %A = percent change.

976 DELIMA ET AL Ann Thorac Surg LOW-POTASSIUM SOLUTION FOR LUNG PRESERVATION 1996;61:973-6

s ium- induced vasoconstrict ion results in elevated PVR dur ing early reperfusion. We [9] demons t ra ted that de- spite the use of pros taglandin E 1 at the t ime of preserva- tion, increased PVR was seen at normal and high-flow reperfusion. This increase in PVR was responsible for impa i rment of lung function in lungs preserved with Euro-Coll ins solution compared with those f lushed with saline solution. In the present study, we tes ted a low- potass ium solution, an extracel lular- type solution differ- ing from Euro-Coll ins solution only in sodium and po- tassium concentrat ions (see Table 1).

After a relatively short s torage time of 4 hours, the lungs preserved with low-potass ium solution demon- s t ra ted be t t e r lung funct ion c o m p a r e d with Euro- Col l ins-preserved and saline so lu t ion-preserved lungs. The increased oxygenat ion capacity, lower PA pressure and PVR, lower percent decrease in dynamic airway compliance, and lower wet to dry ratio seen with low- potass ium solution preservat ion were significant com- pared with Euro-Collins solution preservat ion. These results confirm the impor tance of low potass ium concen- trations in lung preservat ion solutions as previously sugges ted [12-14]; however, they also suggest that the det r imenta l effects of a high potass ium concentrat ion are due to direct effects on the pu lmonary vasculature. A potent ial l imitat ion of this model may be the inherent sensitivity of the rabbi t vasculature to high concentra- tions of potassium. As always, results of these animal studies must be app l ied cautiously to the human situa- tion.

The lungs preserved in low-potass ium solution dem- ons t ra ted an improved ar ter iovenous oxygen gradient along with lower PA pressure and PVR compared with saline so lu t ion-preserved lungs, but wi thout reaching significance. The improved lung function seen late in the reperfusion per iod with low-potass ium solution over saline solution may be due to advantageous glucose metabol ism, osmotic effects, and the opt imal pH of the phosphate-buffered low-potass ium solution [6, 15-17]. The addi t ion of dextran in low-potass ium preservat ion solutions has also been shown to be advantageous be- cause of its actions as an oncotic agent, oxygen-der ived free radical scavenger, and enhancer of microvascular flow [13]. The beneficial effects of dextran were not tested in this model of high-flow reperfusion to emphasize the role of potass ium in preservat ion solutions.

In conclusion, the low-potass ium preservat ion solution provided super ior protect ion of lungs subjected to high- flow reperfusion compared with those preserved with Euro-Collins solution. In an effort to adapt preservat ion methods to the clinical situation, the results of single- lung t ransplanta t ion for pu lmonary hyper tens ion may be improved with the use of such an extracel lular- type preservat ion solution.

This work was supported by the National Institutes of Health under grant HL 48242 and National Research Service Award fellowship F32HL09115-01A1. Additional support came from CNPq-Conselho Nacional de Desenvolimento Cientifico Tecno- logico, Brazil.

The technical advice of Anthony J. Herring is acknowledged.

R e f e r e n c e s

1. Keshavjee SH, Yamazaki F, Cardoso PF, et al. A method for safe twelve-hour pulmonary preservation. J Thorac Cardio- vasc Surg 1989;98:529-34.

2. Yamazaki F, Yokomise H, Keshavjee SH, et al. The superi- ority of an extracellular fluid solution over Euro-Collins' solution for pulmonary preservation. Transplantation 1990; 49:690-4.

3. Sundaresan S, Lima O, Date H, et al. Lung preservation with low-potassium dextran flush in a primate bilateral trans- plant model. Ann Thorac Surg 1993;56:1129-35.

4. Steen S, Kimblad PO, SjSberg T, Lindberg L, Ingemansson R, Massa G. Safe lung preservation for twenty-four hours with Perfadex. Ann Thorac Surg 1994;57:450-7.

5. Date H, Izumi S, Miyade Y, Andou A, Shimizu N, Teramoto S. Successful canine bilateral single-lung transplantation after 21-hour lung preservation. Ann Thorac Surg 1995;59: 336-41.

6. Wisser W, Ringl H, Wekerle T, Wolner E, Klepetko W. A new flush solution for extended lung preservation. J Heart Lung Transplant 1995;14:289-95.

7. Keenan RJ, Griffith BP, Kormos RL, Armitage JM, Hardesty RL. Increased perioperative lung preservation injury with lung procurement by Euro-Collins solution flush. J Heart Lung Transplant 1991;10:650-5.

8. Bando K, Keenan RJ, Paradis IL, et al. Impact of pulmonary hypertension on outcome after single-lung transplantation. Ann Thorac Surg 1994;58:1336-42.

9. DeLima NF, Binns OAR, Buchanan SA, et al. Euro-Collins solution exacerbates lung injury in the setting of high-flow reperfusion. J Thorac Cardiovasc Surg (in press).

10. Davis RD Jr, Trulock EP, Manley J, et al. Differences in early results after single-lung transplantation. Ann Thorac Surg 1994;58:1327-35.

11. Kawaguchi AT, Kawashima Y, Mizuta T, et al. Single lung transplantation in rats with fatal pulmonary hypertension. J Thorac Cardiovasc Surg 1992;104:825-9.

12. Oka T, Puskas JD, Mayer E, et al. Low-potassium UW solution for lung preservation. Transplantation 1991;52: 984-8.

13. Keshavjee SH, Yamazaki F, Yokomise H, et al. The role of dextran 40 and potassium in extended hypothermic lung preservation for transplantation. J Thorac Cardiovasc Surg 1992;103:314-25.

14. Xiong L, Mazmanian M, Chapelier AR, et al. Lung preser- vation with Euro-Collins, University of Wisconsin, Wall- work, and low-potassium-dextran solutions. Ann Thorac Surg 1994;58:845-50.

15. Date H, Matsumura A, Manchester JK, et al. Evaluation of lung metabolism during successful twenty-four hour canine lung preservation. J Thorac Cardiovasc Surg 1993;105:480-91.

16. Shiraishi T, Igisu H, Shirakusa T. Effects of pH and temper- ature on lung preservation: a study with an isolated rat lung reperfusion model. Ann Thorac Surg 1994;57:639-43.

17. Hiramatsu Y, Muraoka R, Chiba Y, Sasaki M. Influence of pH of preservation solution on lung viability. Ann Thorac Surg 1994;58:1083-6.