Cardiac preservation in patients undergoing transplantation

J

THORAC CARDIOVASC SURG

1991;102:657-65

Original Communications

Cardiac preservation in patients undergoing transplantation A clinical trial comparing University of Wisconsin solution and Stanford solution Recent laboratory investigations have shown significantly improved donor heart preservation and function when the University of Wisconsin solution (UW) is used for arrest and storage. These findings prompted us to compare UW to Stanford solution in a clinical triaL After giving informed consent, patients were blindly randomized to receive a beart arrested and stored in UW or a beart arrested in Stanford solution and stored in normal saline. Orthotopic transplants were performed in a routine manner. Fourteen patients with a mean age of 54 years were randomized to UW, and 15 patients with a mean age of 51 years were randomized to Stanford solution. Mean donor ages (UW 27 years, Stanford 24 years) and ischemic times (UW 150 minutes, Stanford 135 minutes) were similar. Several differences were observed intraoperatively. At end ischemia, mean adenosine triphosphate (UW 5.87 mmoljgm wet weight, Stanford 4.75 mmol/gm) and creatine phosphate (UW 9.26 mmol/gm, Stanford 4.75 mmol/gm) levels were higher in the UW bearts (p < 0.05). Defibrillation requirements (UW 14% [2/14~ Stanford 53% [8/15]) were significantly less in the UW group (p = 0.05). The numher of patients requiring temporary intraoperative pacing also sbowed a significant difference with 7% (1/14) of UW patients versus 47% (7/15) of Stanford patients requiring pacing (p < 0.05). Intraoperative requirement for inotropic support showed a trend in favor of the UW group. End-ischemic and postreperfusion histologic characteristics were similar hetween the two groups. No differences in bemodynamics or ejection fractions were noted postoperatively, but trends toward improved rbythm and decreased inotropic support were present in the UW group. Overall 6-month survival rates were similar (UW 86% [12/14], Stanford 93% [14/15]). No preservation-related deaths occurred. We conclude: (1) UW is a safe and effective preservation solution for buman cardiac transplantation; (2) considering the improved end-iscbemic adenosine tripbosphate and creatine phosphate levels, decreased defibrillations, decreased intraoperative pacing, and trend toward decreased requirement for inotropic support in the UW group, UW appears to be superior to Stanford solution for donor beart preservation.

Darryl G. Stein, MDa (by invitation), Davis C. Drinkwater, Jr., MDa (by invitation), Hillel Laks, MD,a Lester C. Permut, MDa (by invitation), Susheela Sangwan, MDb (by invitation), Howard I. Chait, MDb (by invitation), John S. Child? (by invitation), and Sunita Bhuta, MDd (by invitation), Los Angeles, Calif. From the Division of Cardiothoracic Surgery," Department of Anesthesiology," Divisionof Cardiology,?and Department of Pathology," Universityof California at Los Angeles Medical Center, Los Angeles, Calif. Read at the Seventy-first Annual Meeting of The American Association for Thoracic Surgery, Washington, D.C., May 6-8, 1991.

Supported by a grant from Dupont Pharmaceuticals. Address for reprints: Davis C. Drinkwater, Jr., MD, Associate Professor of Surgery, Division of Cardiothoracic Surgery, UCLA Medical Center, 82-375 CHS, Los Angeles, CA 90024-1741.

12/6/31771

657

The Journal of Thoracic and Cardiovascular

6 5 8 Stein et al.

In the early 1980s, improved immunosuppression and postoperative management led to the acceptance of cardiac transplantation as a therapeutic optionfor end-stage cardiomyopathy. The number of cardiac transplants performed each year has increased from fewer than 100 in 1980 to more than 2500 in 1990.1 Survival has also improved, with more than 80% of patients surviving the first posttransplantation year.2 The primary cause of death in the first 30 days after transplantation, however, remains cardiac in origin rather than from rejection or infection.' Furthermore, increased 30-day mortality has been shown to correlate with increased ischemictimes.' These findings suggest that myocardial preservation remains suboptimal and that better donor heart preservation may improvethese results and safelyallowlonger ischemic periods. Recent advances in preservation solutions for other organs have led to their investigation for use in cardiac graft preservation. Wahlberg, Southard, and Belzer" reported that a solutioncharacterized by an ioniccomposition similar to that of intracellular fluid and containing antioxidants and high-molecular-weight molecules improvedpreservation of pancreas allograftsin laboratory animals. Application of this solution, University of Wisconsin solution (UW), rapidly expanded to kidney and liver preservation in both the laboratory and clinical settings. Decreased cellular injury, improved graft survival, and the safe extension of ischemictimes weredemonstrated."? Laboratory investigations in which UW was used as a perfusate and storage solution for cardiac graft preservation followed. We8 and others9- 13 examined myocardial structure, biochemistry, and function after prolonged periods of ischemic storage in UW. In each comparison, the preservation characteristicsof UW were better than those of other clinically employed arrest and storage solutions. Becauseof the excellent overall results in other organ systems, particularly liver transplantation, and the efficacy demonstrated in laboratory investigations, we initiated a clinical trial with UW. The objective of this prospective, randomized,double-blind study was to compare the efficacy and safety of UW for coldcardioplegic arrest and storage of orthotopically transplanteddonorhearts to our current protocol, in which Stanford solution is used for arrest and normal saline for storage. Methods After obtaining the approval of our institutional review board, we began enrolling patients in February 1990. All patients awaiting transplantation between the ages of 2 and 65 years were considered eligible for the study, with the exception of those considered at high risk of death from factors unrelated to donor heart preservation. These factors included multiorgan system dysfunction, pulmonary hypertension with pulmonary

Surgery

Table I. Recipient and donor profile No. of patients Recipient sex Male Female Recipient age (yr) Mean ± SEM Range Recipient diagnosis Ischemic Idiopathic Viral Donor age (yr) Mean ± SEM Range Donor events Arrest Hypoxia Inotropic agents Ischemic time Mean ± SEM Range

UW

Stanford

14

15

11 3

13 2

54 ± 3 32-64

51 ± 3 20-66

9

12 2 0

5 1

27 ± 3 16-46 4 7 1

24 ± 2 13-35 4 11 2

150 ± 8 103-225

135 ± 7 94-181

Donor events refers to pre-harvest events including cardiac arrest, significant hypoxia or respiratory arrest, and the need for inotropic support. SEM. Standard error of the mean.

Table II. Composition of cardioplegic solution UW pH (room temperature) Na (mEq/L) K (mEq/L) HC03- (mfiq/L) Glucose (gm/dl) Mannitol (gm/dl) Pentafraction (gm/L) Lactobionate (grn/L) Raffinose (grn/L) Allopurinol (gm/L) Adenosine (grn/L) Glutathione (gm/L) Regular insulin (U/L) Dexamethasone (mg/L) Osmolality (mOsm)

7.4 20 140

50 35.83 17.83 0.136 1.34 0.922 40 16 320

Stanford 8.0 .28 30 28 50 25

450

artery systolic pressure greater than two-thirds systemic pressure, and the need for ventricular assist devices. Patients were also excluded if the anticipated ischemic time was greater than 4 hours. The recipient's informed consent was required for participation in the study. On enrollment, patients were stratified with regard to their risk for postoperative renal failure. The objective of this stratification was to accommodate an ongoing immunosuppression/ renal failure trial. This stratification assured similar immunosuppressive regimens between the groups in our study. Recipients were considered at high risk if they met one or more of the following criteria: serum creatinine level greater than 1.5

Volume 102 Number 5 November 1991

Cardiac preservation 6 5 9

1'1

UW

12

ST ANFORD

~

s:

;:"

3

'Q; 3 00

10

8 6

<,

0

E

'1

=>

2 0 UJfJ IS(HFMIA

• 3 HI

15 MII-JUTF5

t~LJTF:;

TIME FROM REPERiUSION

Fig. 1. Intraoperative high-energy phosphates. ADP and AMP levels, not shown here, were similar between the groups at all sample times. Error bars depict the standard error of the mean.

mg/dt or need for inotropic support within 30 days of transplantation. All other patients were considered at low risk. After stratification, patients were randomized to receive a donor heart arrested and stored in UW or a heart arrested in Stanford solution and stored in normal saline. Fourteen patients were randomized to the UW group (four high risk and 10 low risk) and 15 patients to the Stanford/saline group (six high risk and nine low risk). The surgical implantation team, anesthesiologists, cardiologists, and the pathologist managing these patients or conducting evaluative tests were blinded to the identity of the study solution. The harvest team was not blinded but did not participate in recipient evaluation or management. An interim evaluation was conducted after the enrollment of 12 patients to assure patient safety.!" Recipient and donor profiles are shown in Table I. Additional donor evaluation included a comprehensive evaluation of the pre-harvest twelve-lead electrocardiogram. No significant differences were noted between the two groups with regard to rhythm, heart rate, PR interval, QRS interval, QT interval, or T-wave morphology. Echocardiograms were performed in 11 of 14donors in the UW group and 12 of 15 donors in the Stanford group. All were read at the donor hospitals and were reported as showing "normal" wall motion and ejection fractions. The donor organ harvest procedure was conducted in a standard fashion. All donors received 50 ml of 50% dextrose, 5 to 10 units of regular insulin, and 10 to 20 mEq of potassium before organ harvest. At the time of arrest, hearts received 1 L of cold UW or Stanford solution (Table II). Topical cooling was accomplished with normal saline slush. After excision, the heart was placed in a basin containing either cold UW or cold normal saline. The heart was then prepared for implantation, placed in sterile bags containing the appropriate storage solution, packaged in a protective container on ice, and transported to the University of California at Los Angeles. The recipient operation was also performed in a standardized fashion. All hearts remained topically cooled during implantation and, after completion of the left atrial anastomosis, the left ventricular cavity was slowly and continuously flushed with electrolyte solution (Plasma-Lyte). After sequential completion of the right atrial septal, posterior pulmonary artery, and aortic anastomoses, hearts received a 3-minute infusion of a modified reperfusate, aspartate/glutamate-enriched warm blood cardioplegic solution (Table III). This was delivered at a pressure of 50 mm Hg with each group receiving a similar volume:

Table III. Aspartate/glutamate-enriched blood cardioplegic solution Cardioplegic additive KCl (mEqjL) Tham (0.3 mol/L) CPD (mmoljL Ca" ") Aspartate (rnrnol/L) Glutamate (rnmol/L) Glucose (rng/dl) Osmolarity (mOsm)

Concentration delivered 8-10 pH 7.5-7.6

0.15-0.25 13 13

>400 380-400

Concentration delivered includes final composition after mixing one part cardioplegic solution with four parts blood. Tharn, Tromethamine; CPD. citratephosphate-dextrose.

663 ± 34 ml in the UW group and 643 ± 31 ml in the Stanford group. Biochemical sampling was facilitated by unmodified blood perfusion delivered to.the heart via the cardioplegia cannulas for 7 minutes before aortic crossclamp removal. The left ventricle was decompressed with a vent placed through the left atrium. After rewarming and before termination of cardiopulmonary bypass, all patients received a standardized intravenous regimen consisting of dobutamine 5 /Lg/kg/min, dopamine 5 /Lg/kg/min, and prostaglandin E) 0.03 /Lg/kg/min. The hearts were assessed biochemically, pathologically, and functionally during the intraoperative period. Right ventricular endomyocardial biopsy specimens were taken just before reperfusion and at 3 and 15 minutes after reperfusion. Samples were immediately frozen in liquid nitrogen and subsequently analyzed for adenosine triphosphate, diphosphate, and monophosphate (ATP, ADP, AMP) and creatine phosphate (CP) by means of high-pressure liquid chromatography. Lactate and creatine kinase (CK) samples were taken from the coronary sinus and the aortic root at 1,5, and 10 minutes after reperfusion. Simultaneous coronary flows were recorded to allow calculation of lactate flux and CK release at each time point. To account for the effect of the heart's mechanical state on these values, we recorded the time from reperfusion until onset of spontaneous rhythm. Pathologic assessment was made from additional biopsy samples taken just before reperfusion and 15 minutes after reperfusion. Biopsy specimens were immediately fixed in 2.5%

The Journal of Thoracic and Cardiovascular

6 6 0 Stein et al.

Surgery

Table IV. Grading system for myocardial ultrastructure Myocardial ultrastructure

Grade 1

Nucleus

No changes

Mitochondria

Mild enlargement; amorphous matrix densities scarce or absent

Myofilaments

No changes

Interstitium Hemorrhage Edema

Mild Mild

Grade 2

Grade 3

Grade 4

Minimal margination and clumping of chromatin Mild to moderate enlargement; decreased matrix density; disruption of cristae; few amorphous matrix densities Mild to moderate contraction bands; blurred Z lines

Moderate margination and clumping of chromatin Moderate to marked enlargement; decreased matrix density; disruption of cristae; few amorphous matrix densities Moderate to severe contraction bands; blurred Z lines

Marked margination and clumping of chromatin

Mild to moderate Mild to moderate

Moderate to severe Moderate to severe

Table V. Intraoperative evaluation offunction UW

14% (2/14) Defibrillation 7% (I/14) Temporary pacing 14% (2/14) Additional inotropic agents Fractional shortening 56% ± 4% 5 min after CPB (mean ± SEM) 45 min after CPB (mean ± SEM) 59% ± 5%

Stanford

53% (8/15)* 47% (7/15)* 40% (6/15)t 46% ± 4%* 52% ± 4%

Additional inotropic agents refers to the number of patients requiring inotropic support above our standardized regimen (see text). CPB. Cardiopulmonary bypass. 'p < 0.05. tp = 0.13. :j:p = 0.16.

glutaraldehyde and 2% paraformaldehyde buffered to a pH of 7.2 with sodium cacodylate 0.1 mol/L, Samples were subsequently washed with cold buffer, postfixed in 1% osmium tetroxide, dehydrated, and embedded in epoxy resin. Semi thin sections stained with toluidine blue dye were made for screening and selection. Ten to 15 electron micrographs of each specimen were subsequently developed with a Zeiss EM 109 microscope (Carl Zeiss Inc., Thornwood, N.Y.). Samples were then evaluated by one pathologist in a blinded fashion. An overall injury score was assigned to each biopsy sample using a scale with a range from 0 to 4 (0 = normal heart without injury, 4 = severe injury). The injury score was based on nuclear, mitochondrial, myofibril, and interstitial changes (Table IV). Functional assessment included an evaluation of both the e1ectrophysiologic and mechanical function of the heart. The conduction system was assessed by recording the number of hearts requiring defibrillation and the number requiring temporary intraoperative pacing for bradycardia or heart block. Assessment of mechanical function included notation of inotropic requirement at higher doses than our standardized protocol or the addition of inotropic agents not included in the standardized protocol. In addition, left ventricular fractional shortening

Megamitochondria; loss of integrity of membranes; many amorphous matrix densities Severe contraction bands; myofibrillarlysis

Severe Severe

was assessed by transesophageal echocardiography at 5 and 45 minutes after the termination of cardiopulmonary bypass. Postoperative functional assessment was conducted during the first week after transplantation. Twelve-lead electrocardiograms were obtained on each of the first 3 postoperative days, and requirement for temporary pacing was noted. The mean right atrial pressure, mean pulmonary artery pressure, and cardiac index were recorded for the first 12 hours after the operation. The mean systemic arterial pressure was recorded for 72 hours after operation. The total dose of dobutamine and dopamine administered over each of the first 3 days was recorded. Doses of additional inotropic agents were also noted. Left ventricular ejection fractions were calculated frorntransthoracic echocardiograms performed on postoperative days 1,3, and 5. Cardiac index was measured 1 week after transplantation at the time of right ventricular biopsy. Statistical methods. Quantitative data are summarized by mean ± standard error of the mean (SEM). Comparisons were made, generally, by independent t tests. In some cases, in which the variable was highly abnormal (e.g., EM biopsy grades and number of defibrillation attempts), comparisons were made by the Wilcoxon procedure. Categorical data are summarized by frequency counts and percentages. Because of the small number of patients in the study, categorical comparisons were made by Fisher's exact test. All reported p values are two-tailed. No adjustment was made for multiple comparisons.

Results Intraoperative evaluation. The biochemical assessment showed significant differences between the two groups. ATP and CP levels were significantly higher in the UW groupbefore reperfusion and remained higher 3 minutes after reperfusion (Fig. 1).ADP and AMP levels were similar between the groups at all sample times. Decreased lactate release wasnotedin the UW groupat 1and 5 minutes, becoming significantly less at 10minutes (Fig.2). However, therewerenosignificant differences in

Volume 102 Number 5 November 1991

Cardiac preservation 6 6 1

Table VI. Postoperative hemodynamics

uw Mean RA pressure (mm Hg) o hr 10 ± 1 12 hr 8± I 24 hr 12 ± 2 Mean PA pressure (mm Hg) o hr 21 ± 2 12 hr 19 ± 3 24 hr 24 ± 2 Mean arterial pressure (mm Hg) o hr 74 ± 3 12 hr 82 ± 3 24 hr 82 ± 2 Cardiac index (Ljminjm 2) o hr 3.2 ± 0.2 4 hr 3.2 ± 0.2 8 hr 2.9 ± 0.2

100

Stanford w

Ul

8± 1 8± 1 9 ± 2


w

-.J

W

UN

-

.~

E f- E
STANFOf1D

w 0, 50

P
f-

20 ± 1 22 ± 1 20 ± 3 78 ± 4 80 ± 2 81 ± 3 3.2 ± 0.3 3.7 ± 0.2 3.2 ± 0.3

Allresults arc expressed as mean ± standarderror of the mean.Zero hours(0 hr)

represents the first measurement on arrival in the intensivecare unit. Note: Pulmonary artery catheters wereoften removed after 10 hours. The n value in the

pulmonary artery pressure results fell to 5 in each groupat 12 hours and to 3 in the UW groupand 5 in the Stanford group at 24 hours. RA, Right atrial; PA, puImonary artery.

CK release between the two groups at any of the sample times. The time until first rhythm was also similar in the two groups, with a mean time of?8 ± 0.9 minutes in the UW group and 7.3 ± 1.3 minutes in the Stanford group. Electron microscopic assessment showed no differences between the groups. End-ischemic biopsy specimens received similar injury scores, with the UW group scoring 1.3 ± 0.1 and the Stanford group scoring 1.4 ± 0.2. Although no differences were noted between these groups 15 minutes after reperfusion (UW 2.8 ± 0.2, Stanford 2.5 ± 0.2), these scores were significantly worse than their prereperfusion scores (p < 0.05). Differenceswere also noted in the functional evaluation of the two groups (Table V). Defibrillation and temporary pacing was required in significantly fewer UW hearts than Stanford hearts. In addition, a trend toward a decreased need for inotropic agents and improved fractional shortening were noted in the UW group. Postoperative evaluation. Evaluation of postoperative electrocardiograms obtained on days I, 2, and 3 revealed no significant differences between the two groups with regard to heart rate, PR interval, QRS interval, QT interval, and T-wave morphology. A trend toward improved early rhythm was noted, however, in the UW hearts. Seventy-nine percent (11/14) were in normal sinus rhythm 1 day after the operation compared with only53% (8/15) of the Stanford hearts. By postoperative day 3, 86% (12/14) ofUW hearts and 87% (13/15) of Stanford hearts were in normal sinus rhythm. Twenty-

o


". • • • • j 5

•

10

TIME FROM REPERFUSION (min)

Fig. 2. Intraoperative lactate release. Error bars depict the standard error of the mean.

one percent (3/14) of the UW group required temporary pacing and 27% (4/15) of Stanford hearts. No patient in either group required permanent pacemaker implantation. Hemodynamic evaluation revealed no significant differences between the two groups. An abbreviated summary of the results is shown in Table VI. The level of inotropic support required by each group during the first 3 days after operation also showed no significant differences, although a trend toward higher dobutamine requirement in the Stanford group was noted (Fig. 3). All other inotropic agents were administered in similar doses. Dopamine doses were nearly identical between the groups over the first 3 days. A total dose of 7996 ± 1255 /Jg/kg per patient in the UW group and 7731 ± 549 /Jg/kg per patient in the Stanford group was administered on the first postoperative day. By the third day after operation, the dose had fallen to 3521 ± I224/Jg/kg per patient and 3536 ± 511 /Jg/kg per patient in the UW and Stanford groups, respectively. Additional inotropic agents were administered to two patients from each group during the first 24 hours after operation. One patient from each group required catecholamine support and one patient from each group received isoproterenol (lsuprel). These inotropic agents were discontinued within 24 hours in all but one patient from the UW group. This patient required isoproterenol until the third postoperative day. Transthoracic echocardiograms performed on days 1, 3, and 5 after the operation showed no significant differences. Left ventricular ejection fractions were 63% ± 3%, 65% ± 4%, and 64% ± 3% in the UW groups and 61% ± 3%,63% ± 2%, and 68% ± 2% in the Stanford group on days 1, 3, and 5, respectively. Right ventricular dysfunction as evidenced by decreased ejection fraction or wall motion abnormality was present in four UW hearts and six Stanford hearts.

662

The Journal of Thoracic and Cardiovascular

Stein et al.

Surgery

8000 . - - - - - - - - - , - - - - - - , - - - - - - - - , 7000 +---+UJ

Z

C

~

6000

D

III

-L-.----

uw STANFORD

~-r ~0- 5000

1-"

~ ~

o

4000

E 3000

00'

2000 1000 0+-'----

2

3

POSTOPERATIVE DAY

Fig. 3. Postoperative dobutaminerequirements. Differences approachedsignificance on days 2 and 3 with p values of 0.09 and 0.12, respectively.

Table VII. Patient survival Survival I rno

6 rno

uw

Stanford

93% (13/14) 86% (12/14)

100% (15/15) 93% (14/15)

There are no statistical differences, UW versus Stanford, in either I-month or 6-month survival rates.

Follow-up. Mean cardiac index measured I week after transplantation was similar between the groups (UW 2.8 ± 0.1 L/min/m2, Stanford 3.0 ± 0.11 L/min/m 2) . Biopsy evidence of allograft rejection was similar at 1 week, with one patient in the UW group and two patients in the Stanford group having mild rejection. Moderate rejection was present in one UW patient. The remaining patients in both groups showed no evidence of rejection. Patient survival at I month and 6 months was similar between groups (Table VII). One early death occurred in the UW group and no early deaths in the Stanford group. The one death occurred 24 days after transplantation when life support was withdrawn from a 57-year-old patient with irreversible brain injury and sepsis. One late death occurred in each group. A 51-year-old patient from the UW group died from infection 72 days after transplantation. The death in the Stanford group occurred in a 20-year-old patient 102 days after transplantation as a result of severe rejection. Discussion Improving myocardial preservation for transplantation would provide several benefits to recipients, the most obvious being increased survival. Despite overall good results, early patient survival continues to be significantly influenced by preservation, as evidenced by increasing mortality rates as ischemic time is extended.' With

improved preservation, regionally procured hearts would have a better margin of safety and potentially better perioperative function. Increased survival would also be expected. Additionally, ischemic times could be safely extended, which would allow an expanded procurement area. Status 1 patients and patients who are difficult to crossmatch would potentially benefit by having an expanded donor pool. Other benefits include the potential for HLA matching and the ability to perform semielective transplantations. The present study was conducted to determine the efficacy of UW for human donor heart preservation. UW has been shown to be beneficial in the laboratory setting, providing superior preservation after extended ischemic times when compared with other clinically used preservation solutions.v" UW's ability to preserve the human heart was previously unknown; therefore this study was designed to determine the safety ofUW for human heart preservation and to determine if, at relatively short ischemic times, any benefit would be provided by the use ofUW. In each assessment UW was at least equivalent to, and in several instances superior to, our routine preservation protocol utilizing Stanford solution for arrest and normal saline for storage. Several laboratory investigations examined high-energy phosphates in hearts preserved with UW. Trends toward better maintenance of high-energy phosphates at end-ischemia when compared with other preservation solutions were demonstrated.': 9. I \ These trends increased as the ischemic times were prolonged. Swanson and colleagues? compared UW preservation with Stanford preservation in a canine model. After 5 hours of ischemic storage, ATP levels were nearly identical between the groups, whereas after 12 hours ATP levels were significantly higher in the UW group. Differences in high-energy phosphate levels were therefore not expected

Volume 102 Number 5 November 1991

in the present study, as ischemic times averaged only 150 minutes in the UW group and 135 minutes in the Stanford group. The significantly higher ATP and CP levels we observed at end-ischemia in the UW group were intriguing. Although the mechanism is uncertain, the adenosine in UW may provide substrate for increased ATP production. Alternatively, UW may result in decreased ATP use during storage. Regardless of the mechanism, these findings suggest that ischemic times could be safely extended in this group. After reperfusion, high-energy phosphate levels fell in both groups. The UW group again showed an advantage, with significantly higher CP levels 3 minutes after reperfusion. There was no difference in the time to onset of rhythm between the groups, indicating that the contractile state of the heart could have influenced these results. Lactate release during the first 10 minutes after reperfusion also showed a trend in favor of the UW group. The decreased release in the UW group, becoming significant at 10 minutes, suggests an earlier return to aerobic metabolism in these hearts. Release of CK, however, did not differ between the UW and Stanford groups over this time period. Histologic findings were similar in the two groups. End-ischemic biopsy specimens showed only mild interstitial edema and minimal enlargement of mitochondria. Fifteen minutes after reperfusion, both groups in our study showed significant injury. Although no difference was noted between the two groups, the injury was significantly worse than that seen before reperfusion. In 1980, Billingham and associates" described ultrastructural evidence of reperfusion injury after human heart transplantation. Numerous recent investigations have focused on understanding this reperfusion phenomenon and preventing its detrimental effects on preservation.ls-" These investigations have implicated oxygen-derived free radicals as a primary mediator of this injury. 19·21 In an attempt to reduce injury caused by oxygen free radicals and other agents, allopurinol and reduced glutathione were added to UW. Allopurinol inhibits the enzyme xanthine oxidase, which has been implicated in the formation of oxygen free radicals.F Glutathione is a reducing agent for several cytotoxic agents including hydrogen peroxide, lipid peroxides, and free radicals.P Despite these additives, the UW hearts, from a histologic standpoint, fared no better than the Stanford hearts. The concentration ofxanthine oxidase in human hearts, however, has been reported as undetectable." which suggests that the addition of allopurinol to human heart preservation solutions provides no added benefits.The conclusion that UW fails to provide histologic preservation benefits at the time of reperfusion, however, is drawn with some reservations. The biochem-

Cardiac preservation 6 6 3

ical and functional benefits demonstrated with UW are not reflected in the injury scores given to that group. This raises the possibility that our histologic assessment may have lacked the sensitivity to distinguish subtle histologic differences that may have shown a benefit to UW preservation. This information, however, does suggest that reperfusion injury continues to limit optimal preservation and that additional measures to reduce this injury would improve overall preservation. Assessment of the electrocardiogram and cardiac rhythm demonstrated several advantages in the UW group. These patients required significantly fewer defibrillations and significantly less intraoperative pacing when compared with the Stanford group. Although not significant, a trend toward earlier resumption of a normal sinus rhythm in the postoperative period was evident in the UW group. Jacquet and associates-" recently reviewed the prevalence of cardiac rhythm disturbances early after transplantation. Using continuous telemetry, they found bradyarrhythmias in 40% (l0/25) of their patients, nine of whom required permanent postoperative pacing. Longer ischemic time was the only factor found to correlate with postoperative bradyarrhythmias, which suggests that improved preservation might reduce the prevalence of these arrhythmias. Although continuous telemetry was not used in our study, the reduced intraoperative pacing and the trend toward improved postoperative rhythm suggests a preservation advantage in the UW group. No significant differences were noted in the mechanical function of the hearts in the two groups. Again, however, trends were present in favor of the UW group. Intraoperatively, 14% of the UW hearts required inotropic support above our standardized protocol whereas 40% of the Stanford hearts required additional inotropic agents. Postoperatively, there was a trend toward decreased need for inotropic agents in the UW group. Larger numbers of patients may have led to these evaluations becoming significant. One concern with the use of UW for human heart preservation was not directly addressed in this study. The ionic composition of UW is similar to that of intracellular fluid, with a potassium concentration of 140 mfiq/L, The application of cardioplegic solutions with high levels of potassium for human heart preservation has been a concern since the high-potassium (400 mlsq/L) Melrose solution was introduced in 1955 and later found to be associated with myocyte and endothelial cell injury." Concerns over potassium concentrations continued despite subsequent investigations showing that the injury was likely due to citrate and not potassium." Injury to endothelial cells by UW was addressed by Fremes and associates." They hypothermically stored cultured

6 6 4 Stein et al.

human endothelial cells for 36 hours in UW, Stanford solution, and other preservation solutions. The survival and morphology of the UW -stored cells was significantly better than that of cells stored in the other solutions. Although our histologic assessment did not directly evaluate endothelial cell preservation, no specific injury could be ascribed to UW when compared with Stanford solution. Long-term follow-up, however, is necessary to determine if subtle injury did occur and possibly affect the prevalence of rejection or accelerated graft atherosclerosis.

Conclusions UW is a safe and effective preservation solution for human heart transplantation. Considering the improved maintenance of high-energy phosphates during storage, decreased lactate release, decreased need for defibrillation, decreased intraoperative pacing, and trend toward decreased inotropic support in the UW group, UW appears to be superior to Stanford solution for donor heart preservation. Long-term follow-up with increased numbers of patients is necessary to determine if these differences result in improved patient survival. Additionally, limited trials may be warranted to determine the safety of extending ischemic times. We thank Tim Breen, PhD, for the statistical analysis of our results. In addition, we thank Therese Hills, Steve Barthel, our transplant coordinators, perfusionists, operating room nurses, and intensive care unit nurses for their assistance in completing this study. REFERENCES 1. Kaye MP. Heart Registry Report. The international society for heart transplantation. Eleventh annual meeting and scientific sessions. 1991. 2. Kriett 1M, Kaye MP. The registry of the international society for heart transplantation: seventh official report1990. 1 Heart Transplant 1990;9:323-30. 3. Heck CF, Shumway Sl, Kaye MP. The registry of the international society for heart transplantation: sixth official report-1989. 1 Heart Transplant 1989;8:271-6. 4. Wahlberg JA, Southard JH, Belzer Fa. Development of a cold storage solution for pancreas preservation. Cryobiology 1986;23:477-82. 5. Todo S, Nery 1, Yanaga K, Podesta L, Gordon RD, Starzl TS. Extended preservation of human liver grafts with UW solution. JAMA 1989;261:711-4. 6. Kalayoglu M, Sollinger HW, Belzer Fa, et al. Extended preservation of the liver for clinical transplantation. Lancet 1988;1:617-9. 7. Henery ML, Sommer BG, Ferguson RM. Improved immediate function of renal allografts with Belzer perfusate. Transplantation 1988;45:73-5. 8. Breda MA, Drinkwater DC, Laks H, Buhta S, Chang P. Successful long-term preservation of the neonatal heart

The Journal of Thoracic and Cardiovascular Surgery

with a modified intracellular solution [Abstract]. Fifteenth Annual Meeting of The Western Thoracic Surgical Association, 1989. 9. Swanson DK, Pasaoglu I, Berkoff HA, Southard JA, Hegge 10. Improved heart preservation with UW preservation solution. 1 Heart Transplant 1988;7:456-67. 10. Makowka L, Zerbe TR, Chapman F, et al. Prolonged rat cardiac preservation with UW lactobionate solution.Transplant Proc 1989;21(Pt 2):1350-2. 11. Yeh T Jr, Hanan SA, Johnson DE, et al. Superior myocardial preservation with modified UW solution after prolonged ischemia in the rat heart. Ann Thorac Surg 1990;49:932-9. 12. Okouchi Y, Shimizu K, Yamaguchi A, Kamada N. Effectiveness of modified University of Wisconsin solution for heart preservation as assessed in heterotopic rat heart transplant model. 1 THORAC CARDIOVASC SURG 1990; 99:1104-8. 13. Ledingham SlM, Katayama 0, Lachno DR, Yacoub M. Prolonged cardiac preservation: evaluation of the University of Wisconsin solution by comparison with the St. Thomas' Hospital cardioplegic solutions in the rat. Circulation 1990;82(Pt 2):lV351-8. 14. Drinkwater DC Jr, Stein DG, Permut LC, Laks H. Clinical trial of University of Wisconsin solution for cardiac transplantation: preliminary results. J THORAC CARDIOVASC SURG [In press]. 15. Billingham ME, Baumgartner WA, Watson DC, et al. Distant heart procurement for human transplantation: ultrastructural studies. Circulation I980;62(Pt 2)11 1-9. 16. Vinten-Johansen 1, lohnston WE, Mills SA, et al. Reperfusion injury after temporary coronary occlusion. 1 THoRAC CARDIOVASC SURG 1988;95:960-8. 17. Mehta lL, Nichols WW, Mehta P. Neutrophils as potential participants in acute myocardial ischemia: relevanceto reperfusion. J Am Coli Cardiol 1988;11:1309-16. 18. Follette DM, Fey K, Buckberg GD, et al. Reducing postischemic damage by temporary modification of reperfusate calcium, potassium, pH, and osmolarity. 1 THoRAc CARDIOVASC SURG 1981;82:221-38. 19. McCord 1M. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med 1985;312:159-63. 20. Hammond B, Hess ML. The oxygen free radical system: potential mediator of myocardial injury. 1 Am Coli Cardi011985;6:215-20. 21. Stewart lR, Gerhardt EB, Wehr Cl, et al. Free radical scavengers and myocardial preservation during transplantation. Ann Thorac Surg 1986;42:390-3. 22. McCord 1M, Roy RS, Schaffer SW. Free radicals and myocardial ischemia: the role of xanthine oxidase. Adv Myocardiol 1985;5:183-8. 23. Freeman BA, Crapo lD. Biology of disease: free radicals and tissue injury. Lab Invest 1982;47:412-26. 24. Eddy LJ, Stewart lR, Jones HP, Engerson TD, McCord 1M, Downey1M. Free radical producing enzyme, xanthine oxidase, is undetectable in human hearts. Am 1 Physiol 1987;253:H709-11. 25. Jacquet L, Kiadt G, Stein K, et al. Cardiac rhythm distur-

Volume 102 Number 5 November 1991

bances early after orthotopic heart transplantation: prevalenceand clinical importance of the observed abnormalities. J Am Coli CardioI1990;16:832-7. 26. Melrose D, Dreyer B, Bentall H, et al. Elective cardiac arrest. Lancet 1955;2:21-2. 27. Tyers G, Todd G, Niebauer I, et al. The mechanism of myocardial damage following potassium citrate (Melrose) cardioplegia. Surgery 1975;78:45-53. 28. Fremes SE, Li RK, Weisel RD, Mickle DAG, Tumiati LC. Improved hypothermic storage with University of Wisconsin solution. Surg Forum 1989;40:246-8.

Discussion Dr. Henry M. Spotnitz (New York. N.y.). Drs. Weng, Hsu, and Detweiler in our laboratory, stimulated by work by Foglia and Buckberg, developed a model for studying edema in the isolated, arrested, hypothermic pig heart. We found that increasing edema could be quantitatively related to decreasing left ventricular compliance. This was indicated by displacement of the diastolic pressure-volume curve to the left. With 1 L coronary perfusions of solutions of varying composition, UW was the only solution that stabilized left ventricular compliance. UW was also the only perfusate that did not produce progressive increases in heart weight. UW and blood manifested advantages greater than anticipated from osmolarity alone. In freshly excised baboon hearts that otherwise would have been discarded in xenografting experiments, we further examined the effects of serial perfusion on heart weight. Results with this model also support advantages of UW in limiting edema formation. We interpret these results to indicate that UW would be most beneficial under conditions promoting edema, particularly transport runs of long duration or myocardial protection in operations on previously injured hearts or hearts subjected to long crossclamp times. For that reason, I would like to ask you whether you have any relevant experience either in particularly long transplant runs or in other clinical settings. Dr. Axel Haverich (Hannover, Germany). This report mimics our clinical results in a randomized study using UW in some 40 patients. In our clinical experience with heart transplantation, we believe that left ventricular systolic function is not the major problem. We do believe that right ventricular function and especially both sided diastolic ventricular function may be the topic of future investigations to improve cardiac preservation. To evaluate biventricular systolic and diastolic function, we

Cardiac preservation 6 6 5

developed an experimental model of a paracorporeal pig heart and studied the effect of the three different preservation solutions on biventricular systolic and diastolic function after an 8-hour period of global ischemia. We compared the difference between UW and St. Thomas' Hospital solution and St. Thomas' Hospital solution combined with warm blood, substrateenriched reperfusion, which is our current clinical practice. Looking at the data on left ventricular and right ventricular systolic function in the UW group, we could see that after 20 minutes of reperfusion there was a significantly increased systolic function on the right side, whereas the differences in the left ventricle were not very impressive. Looking at diastolic function using the relaxation time constant, we could also see that for the right ventricle, specifically after 20 minutes and after 80 minutes of reperfusion, UW did produce superior results when compared with St. Thomas' Hospital solution or St. Thomas' Hospital plus hot shot reperfusion. My question refers to your data on right ventricular function and dysfunction. Did you see differences between the two groups and at what point after the transplantation did you see right ventricular failure? Dr. Stein. We have used UW for extended preservation in one patient on a compassionate basis. This patient was a 6-year-old child who had left ventricular failure after a Fontan procedure and required support with a left ventricular assist device. After 1 week of support, a heart became available for him from across the country. With no other recipients waiting for this heart, we decided to proceed with the transplant. After arrest and storage in UW, the heart was transported to the University of California at Los Angeles, where it was initially reperfused with an aspartate/glutamate-enriched blood cardioplegic solution for 5 minutes. The heart functioned flawlessly, requiring minimal inotropic support. The ischemic time for this heart was 8 hours. In addition to this heart, we have used UW to arrest and store several hearts with 5-hour ischemic times and found excellent postischemic function. Regarding assessment of right ventricular function, we analyzed right ventricular pressures and echocardiographic evidence of right ventricular dysfunction. Right ventricular pressures were measured for the first 72 hours after transplantation and did not differ significantly between the two groups. Right ventricular dysfunction as evidenced by wall motion abnormalities or decreased right ventricular ejection fractions was present in 29% of the UW hearts versus 40% of the Stanford hearts. This difference suggests an advantage in the UW group, but this difference did not reach statistical significance. Additional studies with increased numbers of patients may show this trend to be significant.