Cardiac transplantation after mechanical circulatory support: A Canadian perspective

Cardiac Transplantation After Mechanical Circulatory Support: A Canadian Perspective Roy G. Masters, MD, Paul J. Hendry, MD, Ross A. Davies, MD, Stuart Smith, MD, Christine Struthers, BScN, Virginia M. Walley, MD, John P. Veinot, MD, Tofy V. Mussivand, PhD, and Wilbert J. Keon, MD Divisions of Surgery, Cardiology, and Pathology, University of Ottawa Heart Institute, Ottawa Civic Hospital, Ottawa, Ontario, Canada

Background. To assess the relative efficacy of cardiac transplantation after mechanical circulatory support with a variety of support systems, we analyzed our consecutive series of patients who had and did not have mechanical support before transplantation. Methods. A review of 209 patients undergoing cardiac transplantation from 1984 to May 1995 was performed. Group 1 consisted of 110 patients who were maintained on oral medications while awaiting transplantation, and group 2 consisted of 60 patients who required intravenous inotropic support. Group 3 included 39 patients who had transplantation after mechanical circulatory support for cardiogenic shock. The indication for device implantation was acute onset of cardiogenic shock in 38 patients and deterioration while awaiting transplantation in I patient. The support systems were an intraaortic

balloon pump in 13 (subgroup 3A), a ventricular assist device in 7 (subgroup 3B), and a total artificial heart in 19 patients (subgroup 3C). Results. After transplantation, infection was more common in group 3 (56%) than in group 1 (28%) or group 2 (32%) (p = 0.005). Survival to discharge was lower for group 3 (71.7%) than for group 1 (90.9%) or 2 (88.3%) (p = 0.009). For mechanically supported patients, survival to discharge was 84.6% in subgroup 3A, 71.4% in subgroup 3B, and 63.1% in subgroup 3C (p = not significant). Conclusions. Transplantation after mechanical support offers acceptable results in this group of patients for w h o m the only alternative is certain death. Patient selection and perioperative management remain the challenge to improving these results.

n the past decade, cardiac transplantation has become a well-established option for the treatment of endstage heart disease. For patients in cardiogenic shock to benefit from this therapy, temporary circulatory support is required until a donor heart becomes available. Staged cardiac transplantation after mechanical support continues to evolve, and many problems remain to be solved. To assess the relative efficacy of circulatory support systems, we compared our results with transplantation after support with the intraaortic balloon pump (IABP), the total artificial heart (TAH), and the ventricular assist device (VAD).

the highest priority, consisted of patients in cardiogenic shock who required mechanical circulatory support. Group 3 patients were further divided into subgroups according to the type of mechanical support used: subgroup 3A, the IABP; subgroup 3B, the VAD; and subgroup 3C, the TAH. We defined cardiogenic shock using the following criteria: cardiac index lower than 2.0 L. min 1 - m 2, pulmonary artery wedge pressure higher than 20 mm Hg, and systolic arterial blood pressure lower than 90 mm Hg, despite maximal inotropic drug support with sympathomimetic amines, phosphodiesterase inhibitors, or a combination of both. In general, our approach to the treatment of patients in cardiogenic shock is to correct, where possible, any lesion amenable to surgical intervention. For patients who are considered to have an inoperable lesion or who remain in shock despite operation, IV inotropic support is initiated. Mechanical support, usually with the IABP, is introduced after failure of inotropic drugs to support the circulation. More advanced circulatory support with either the VAD or the TAH is considered after failure of the IABP and one or two inotropic agents to support the circulation and when the patient has been assessed as an acceptable transplant recipient. The VAD used in this study was the Thoratec device (Thoratec Laboratories, Berkeley, CA). The TAH used is currently manufactured by CardioWest

I

Patients a n d M e t h o d s Data from consecutive patients undergoing cardiac transplantation at the University of Ottawa Heart Institute to April 30, 1995, have been prospectively compiled in a database. For the purpose of this study, patients were grouped according to priority in receiving a donor heart: group 1, the lowest priority, consisted of patients maintained on oral medications either in the hospital or at home; group 2 comprised patients who required intravenous (IV) inotropic agents preoperatively; and group 3, Accepted for publication Jan 26, 1996. Address reprint requests to Dr Masters, Ottawa Heart Institute, 1053 Carling Ave, Ottawa ON KIY4E9,Canada. © 1996 by The Society of Thoracic Surgeons Published by Elsevier Science Inc

(Ann Thorac Surg 1996;61:1734-9)

0003-4975/96/$15.00 PII S0003-4975(96)00138-5

Ann Thorac Surg 1996;61:1734-9

MASTERSET AL TRANSPLANTATIONAFTERMECHANICALSUPPORT

( C a r d i o W e s t Laboratories, R i c h m o n d , BC, C a n a d a ) though previously it was produced by Symbion (Symbion Inc, Salt Lake City, UT). The choice of device for biventricular support was at the discretion of the operating surgeon. After device implantation, patients were monitored in the cardiac surgical intensive care unit. Anticoagulation d u r i n g mechanical support consisted of intermittent lowdose heparin sodium for IABP patients, a continuous IV infusion of heparin (partial thromboplastin time, 1.5 to 2.0 times normal) for VAD patients, a n d a continuous IV infusion of heparin (partial thromboplastin time, 1.5 to 2.0 times normal) in conjunction with IV administration of dipyridamole (10 mg/h) for TAH patients. Warfarin sodium was not used in this study because of the relatively short duration of mechanical support. All donor hearts were preserved with St. Thomas' Hospital crystalloid cardioplegia a n d topical cold saline solution. Donor hearts retrieved prior to 1986 were transported in a cooler at 4°C; since then, the temperature has b e e n m a i n t a i n e d b e t w e e n 8° and 10°C [1]. All transplantations were performed using a standard technique as previously described [2]. For all patients, induction of i m m u n o s u p p r e s s i o n was with IV steroids in conjunction with either W cyclospofine prior to 1988 or 1V i m m u n o g l o b u l i n since that time. The i m m u n o g l o b u l i n used for induction was either antilymphocyte globulin (15 mg • kg 1 . d - l ) or antithymocyte globulin (10 mg • kg -1 • d 1). Maintenance of i m m u nosuppression in all patients consisted of triple therapy based on steroids (1 mg • kg 1 d 1 in tapering doses), azathioprine (2 mg • kg- 1 . d 1), a n d cyclosporine (2.5 to 5.0 mg/kg twice daily to achieve a trough plasma level of 200 to 400 ng/mL). We defined hospital survival as survival to discharge home from the hospital. Comparisons b e t w e e n groups were performed using Student's t test or analysis of variance for continuous variables a n d a 9(2 test or Fisher's exact test for discrete variables. Actuarial survival was calculated using the life-table technique of G r u n k e m e i e r and Starr [3] a n d was reported with 95% confidence limits. Comparisons b e t w e e n groups were considered significant at a p value of less than 0.05. •

Results From May 1984 through April 1995, 213 transplantations were performed in 209 patients with complete follow-up. Four patients u n d e r w e n t retransplantation, 3 emergently within 30 days of the initial procedure a n d before discharge from the hospital a n d I late, 9 years after the first transplantation. Group 1 consisted of 110 patients (53%), 1 of w h o m required late retransplantation. In group 2, there were 60 patients (29%), I of whom required retransplantation on an emergency basis. There were 39 patients (19%) who required mechanical support prior to transplantation (group 3) as follows: subgroup 3A (IABP), 13 patients; subgroup 3B (VAD), 7 patients; a n d subgroup 3C (TAH), 19 patients. Two patients in subgroup 3C required retransplantation emergently.

1735

Table 1. Baseline Data and Early Survival for the Three Study Groups a-c Variable

Group 1 (n - 110)

Recipient age (y) 49.6 -+ 1.0 Recipient sex Male 94 (85) Female 16 (15) Recipient diagnosis CAD 72 (65) Cardiomyopathy 32 (29) Other 6 (5) Donor age (y) 31.1 _+1.1 Anoxia time 155 + 7 (min) Hospital survival 90.9 (%)

Group 2 (n = 60)

Group 3 (n = 39)

p Value

42.0 -+ 2.4 47.9 -4-10.5 0.30 0.87 51 (85) 32 (82) 9 (15) 7 (18)

22 (37) 33 (85) 29 (48) 5 (13) 9 (15) 1 (3) 27.8 + 1.8 33.7 + 1.9 174 + 10 175 _+10 88.3

71.7

0.003

0.06 0.16 0.009

Group 1 received oral medications;group 2, intravenousmedications; and group 3, mechanicalsupport, b Where applicable,data are shown as the mean + the standard error of the mean. c Numbers in parentheses are percentages. CAD - coronaryartery disease.

In addition to these patients, 3 patients had VAD implantation as a bridge to transplantation but subsequently were not offered transplantation. The identification of a p u l m o n a r y malignancy, the d e v e l o p m e n t of right ventricular failure a n d intraoperative hemorrhage precluded transplantation in these 3 patients, a n d all died. The clinical characteristics of groups 1, 2, a n d 3 are shown in Table 1. The recipient ages (p = 0.30) a n d sex ratios (p 0.87), the donor ages (p = 0.06), and the total anoxia times (p = 0.16) were similar in the three groups. However, a greater proportion of recipients in group 3 had coronary artery disease as the primary diagnosis (p = O.OO3). Table 2 details the characteristics of recipients who u n d e r w e n t transplantation after mechanical circulatory support (group 3). Recipient subgroups were similar in the sex ratio (p 0.72), the incidence of CAD as the primary diagnosis (p = 0.94), the incidence of previous coronary artery bypass grafting (p - 0.79), a n d the duration of mechanical circulatory support (p = 0.72). There were no significant differences in donor ages (p = 0.13) a n d total anoxia times (p = 0.43). However, recipients in subgroup 3C were younger than those in either subgroup 3A or 3B (p = 0.005). All of the subgroup 3A patients were seen in the hospital emergently in cardiogenic shock, and n o n e showed deterioration while awaiting transplantation. In addition to circulatory support with the IABP, all patients in this subgroup were m a i n t a i n e d on at least one IV inotropic agent. Coronary artery bypass grafting had b e e n performed in 7 patients, on a previous admission in 4 a n d on the same admission in 3. In subgroup 3B, there were 6 patients with CAD who were seen in shock emergently and 1 patient with car-

1736

MASTERSET AL TRANSPLANTATIONAFTERMECHANICALSUPPORT

Table 2. Baseline Data and Early Survival for Patients Having Transplantation After Mechanical Circulatory Supportp,b

Variable Recipient age (y) Recipient sex Male Female Recipient diagnosis CAD Cardiomyopathy Other Previous CABG Support time (d) Donor age (y) Anoxia time (min) Hospital survival

Subgroup Subgroup 3A (IABP) 3B WAD) (n = 13) (n = 7)

Subgroup 3C (TAH) (n = 19)

54.8 _+1.5

41.8 z 2.6

11 (85) 2 (15)

51.7_+0.7 5 (71) 2 (29)

p Value 0.005 0.72

16 (84) 3 (16) 0.94

12 (92) 1 (8) 0 7 (54) 6 _+1 36.4 _+2.7 175 ,+ 21

6 (86) 1 (14) 0 3 (43) 7 _+2 32.7_+4.1 200 .+ 32

15 (79) 3 (16) 1 (5) 8 (42) 8 _+2 34.1 _+2.9 162 _+11

0.79 0.72 0.13 0.43

11 (84.6)

5 (71.4)

12 (63.1)

0.42

a Where applicable,data are shown as the mean _~the standard error of the mean. b Numbers in parentheses are percentages. CABG - coronary artery, bypass grafting; CAD = coronary artery disease; IABP= intraaorficballoon pump; TAH - total artificial heart; VAD = ventricularassist device.

diomyopathy whose condition deteriorated while he was hospitalized and awaiting transplantation. Biventricular support was used in 6 patients a n d left ventricular support, in 1 patient. Previous coronary artery bypass grafting had been performed in 3 patients, in all on the same admission. Two patients with inoperable coronary disease sustained preoperative cardiac arrest with cardiac massage, a n d 4 patients had the IABP preoperatively. In subgroup 3C, all patients were seen acutely in

Table 3. Complications During Mechanical Support and After Transplantation in Group 3~

Complications During support Infection Neurologic event Reoperation Renal failure After transplantation Infection Rejection b

Subgroup 3A (IABP) (n = 13)

Subgroup 3B (VAD) (n = 7)

Subgroup 3C (TAH) (n = 19)

p Value

4 (31) 0

3 (43) 1 (14)

8 (42) 2 (11)

0.78 1.00

0 0

3 (43) 1 (14)

3 (16) 2 (11)

0.29 0.10

6 (46) 0 (0)

6 (86) 1 (0.4)

10 (53) 6 (0.8)

0.21 0.29

Data are shown as number of patients with percentages in parentheses unless otherwise indicated, b Numbersin parentheses show rejection episodes per 100 hospital days. IABP intraaorticballoon pump; TAH total artificial heart; VAD - ventricularassist device.

Ann Thorac Surg 1996;61:1734-9

cardiogenic shock, a n d the condition of none deteriorated while he or she was on the transplant list. Previous coronary artery bypass grafting had been performed in 8 patients, remotely in 1 patient a n d on the same admission in 7 patients. Cardiac arrest with massage occurred preoperatively in 4 patients, and 11 patients had an IABP preoperatively. Two patients required implantation of a second TAH after transplantation because of rejection and failure of the donor heart. Therefore, 21 artificial hearts were used in subgroup 3C. Both patients received a second transplant but s u b s e q u e n t l y died. Infection was the most c o m m o n complication during mechanical support, a n d there was little difference between subgroups 3A, 3B, a n d 3C (p = 0.78) (Table 3). In subgroup 3A, this complication included IV line sepsis in 3 patients a n d infection at the IABP insertion site in 1 patient. There was one case each of cellulitis, p n e u m o n i a , a n d IV line sepsis in subgroup 3B. In subgroup 3C, 8 patients experienced 12 infections: pneurnonitis (7), IV line sepsis (3), oral Candida (1), a n d disseminated Candida

(1). Neurologic events occurred in 3 patients during circulatory support (p = 1.0). One patient in subgroup 3B sustained a stroke (cerebrovascular accident) with mild arm a n d leg weakness b u t recovered after discharge from the hospital. In subgroup 3C, 1 patient had a self-limited seizure with no residual deficit or recurrence, a n d 1 had a cerebrovascular accident with arm and leg weakness. In the latter patient, the stroke occurred after implantation of a second TAH because of donor heart failure. Reoperation for hemorrhage or t a m p o n a d e was required in 3 patients each in subgroups 3B and 3C (p = 0.29). Renal failure r e q u i t i n g dialysis or hemofiltration occurred during circulatory support in 3 patients, 1 patient in subgroup 3B a n d 2 patients in subgroup 3C (p = 0.10). Two of the patients, 1 in each subgroup, were anuric at the time of device implantation after a cardiac arrest. Hemofiltration was required in both until after transplantation, at which time the renal failure resolved. In the third patient, renal failure developed after i m p l a n tation of a second TAH because of donor heart failure and was treated with hemodialysis. Other complications during circulatory support included transient hepatic insufficiency with an elevated p r o t h r o m b i n time in 2 patients, one each in subgroups 3B a n d 3C, with resolution in both after transplantation. The patient in subgroup 3B who had left ventricular support required IV inotropic support for tight ventricular dysfunction. After transplantation, rejection a n d infection were the most c o m m o n complications e n c o u n t e r e d in the hospital in groups 1, 2, a n d 3. Rejection requiting treatment occurred in 17 patients (0.6 episode/100 hospital days) in group 1, 5 patients (0.4 episode/100 hospital days) in group 2, and 7 patients (0.5 episode/100 hospital days) in group 3 (p = 0.58). Infection, however, was more c o m m o n in group 3. Infection occurred in 31 patients (28%) in group 1, 19 patients (32%) in group 2, and 22 patients (56%) in group 3 (p = 0.005).

A n n Thorac S u r g 1996;61:1734-9

In group 3, infection after transplantation occurred in 6 patients in subgroup 3A, 6 patients in subgroup 3B, and 10 patients in subgroup 3C (p = 0.21) (see Table 3). In subgroup 3A, there were eight infections in 6 patients: urinary tract infection (four), colitis (two), and pneumonitis (two). In subgroup 3B, there were 14 infections in 6 patients: sternal w o u n d infection (three), infection at the driveline exit site (two), pneumonitis (three), urinary tract infection (two), oral herpes (three), and cytomegalovirus infection (one). In subgroup 3C, there were 16 infections in 10 patients: sternal w o u n d infection (four), W line sepsis (four), pneumonitis (six), oral herpes (one), and driveline infection (one). Sepsis in conjunction with multiorgan failure accounted for two deaths in subgroup 3C. Rejection in the hospital occurred in I patient in subgroup 3B (0.4 episode/100 hospital days) and resolved with medical therapy. Rejection occurred in 6 patients in subgroup 3C (0.8 episode/100 hospital days) (p = 0.29). There was resolution with medical treatment in 3 of these patients, but in 1 patient, rejection caused sudden death before therapy could be initiated. In 2 patients, rejection caused profound donor heart dysfunction, and implantation of another TAH was performed after failure of medical treatment. Both patients subsequently died after retransplantation; 1 sustained rejection again and died in the operating room, and the other died of sepsis and multiorgan failure in the hospital. Other complications after transplantation in subgroup 3A included leg ischemia after IABP removal in 2 patients, which resolved with surgical exploration, a seizure possibly related to cyclosporine in 1 patient, and heart block requiring a permanent pacemaker in 1. Intraoperative hemorrhage and intraoperative graft failure each occurred in 1 patient, and both patients died. Other complications in subgroup 3B included hemorrhage, deep venous thrombosis, right ventricular dysfunction, and transient ischemic attack in 1 patient each. The patient with a transient ischemic attack awoke immediately after transplantation with a left arm deficit that completely resolved within hours. This was considered likely to represent an air embolism during VAD removal. One patient in subgroup 3B died intraoperatively of graft failure of undetermined etiology, and 1 patient had development of a tracheoarterial fistula after prolonged oral intubation and died of massive hemoptysis. In subgroup 3C, 2 patients died, 1 of graft failure of undetermined etiology and 1 with normal coronary arteries, of myocardial infarction. Overall, the hospital survival rate was less for patients who required mechanical support before transplantation. The survival rates for groups 1 and 2, were 90.9% and 88.3%, respectively, whereas for group 3, the rate was 71.7% (p = 0.009). Survival to discharge after transplantation was 84.6% for subgroup 3A, 71.4% for subgroup 3B, and 63.1% for subgroup 3C (p = 0.42). The actuarial 5-year survival rate was 78% _+ 5% for group 1, 74% + 7% for group 2, and 57% +_ 9% for group 3 (Fig 1). The difference in 5-year survival estimates was not significant between groups I and 2 (p = 0.62) and between groups 2

MASTERS ET AL TRANSPLANTATION AFTER M E C H A N I C A L SUPPORT

1737

100 80

~

"

"~"----_.

78+5% Group 1 -

..........

74+7% Group 2 60

"-- . . . . . . . . .

%

~'~"

~'~

57_+9% Group 3

40

20

o I 1

2

3

4

s

110

88

72

66

52

60

40

33

27

22

39

20

20

16

13

Year At Risk

Fig 1. Actuarial 5-year survival.

and 3 (p = 0.12). However, the actuarial 5-year survival was significantly lower for patients in group 3 compared with group 1 (p - 0.04).

Comment Cardiac transplantation has evolved as an effective treatment for patients with terminal heart failure. As of 1994, the Registry of the International Society for Heart and Lung Transplantation [4] included 26,704 heart transplantations from 251 centers. In a subset of 9,389 patients from the United States, a n u m b e r of risk factors for early mortality were identified by multivariate analysis. Included a m o n g these was preoperative use of mechanical circulatory support. Similarly, in a smaller series of 179 heart transplant recipients, we [5] previously reported that use of mechanical support was an independent risk factor for hospital mortality in a multivariate analysis. In that study, survival to discharge was 67% for patients undergoing transplantation after mechanical circulatory or ventilatory support compared with 90% for patients without mechanical support. In the present study, we found that patients who required mechanical circulatory support presented a higher operative risk with a lower 5-year survival compared with those who did not. For this group of patients in profound acute cardiovascular collapse, the only alternative was certain death. That more than 70% of patients in this group survived and were discharged home, however, deserves emphasis. A n u m b e r of patients in this study who required mechanical support were sustained with an IABP with an obvious survival advantage, a finding demonstrating the enduring effectiveness of this therapy. However, two thirds of our patients required more invasive support systems because of the severity of the impairment of cardiac function. Of this group of critically ill patients who required either a VAD or an artificial heart, all but I were seen on an emergency basis, 40% had postcardiotomy shock, and approximately 25% had sustained preoperative cardiac arrest. The 1993 report of the voluntary combined registry experience with VADs and TAHs [6] contains data on 544 patients from 66 centers with an overall survival to

1738

MASTERS ET AL TRANSPLANTATION AFTER M E C H A N I C A L SUPPORT

discharge after transplantation of 66.2%. Survival was 87.4% for patients s u p p o r t e d with a left VAD, 69.5°/'., for patients s u p p o r t e d with a biventricular assist device, a n d 49.6% for TAH patients. Although the 1-year and 2-year survival estimate was 65% for the group overall, it was 86% and 83%, respectively, for univentricular support. In that report, however, the indication for device i m p l a n t a tion was deterioration while awaiting transplantation in 92% of patients. C o m p a r i n g m e t h o d s of circulatory support, Reedy and associates [71 r e p o r t e d an early survival advantage with the VAD (100%) c o m p a r e d with the IABP (70%). Of the 14 patients s u p p o r t e d with a VAD, 10 had left ventricular assist. Similarly, both Frazier a n d co-workers [8] and Kormos and colleagues [9] r e p o r t e d excellent results with left VAD s u p p o r t with 80% and 100% hospital survival, respectively, a m o n g patients u n d e r g o i n g transplantation. However, it is i m p o r t a n t to note that these results were o b t a i n e d in patients for w h o m the indication for device implantation was deterioration while awaiting transplantation. In contrast to the p r e s e n t study, mechanical support was not u s e d in a resuscitative setting or in patients with acute-onset shock. The survival data in the p r e s e n t study for patients having transplantation after biventricular s u p p o r t or the TAH are similar to those of the c o m b i n e d registry [61. In our study, only 1 patient s h o w e d deterioration while awaiting transplantation. Most patients were seen acutely with cardiogenic shock, and the circulatory support systems were i m p l a n t e d e m e r g e n t l y a n d in a resuscitative setting. W e believe that these patients p r e s e n t an even greater risk for successful device use. Further, we submit that these patients, m a n y of w h o m were u n d e r going cardiac massage, are less a m e n a b l e to univentricular support. Farrar a n d Hill [10] in 1993 r e p o r t e d a rnulticenter evaluation of 151 patients who received the Thoratec VAD prior to transplantation. After transplantation, 83% and 87% were discharged after biventricular and left ventricular assistance, respectively. Of note, these excellent results were achieved in a group of patients of w h o m 42% had sustained cardiac arrest prior to implantation a n d in w h o m device implantation was often used for resuscitation. Farrar a n d colleagues [11] s u b s e q u e n t l y r e p o r t e d that b l o o d urea nitrogen was the only i n d e p e n dent predictor of hospital survival in these patients. In 1992, Johnson a n d coauthors [12] s u m m a r i z e d a multicenter experience with clinical use of the TAH in conjunction with transplantation in 171 patients. For the 118 patients receiving transplantation, survival was 72% at 30 days a n d 57% at I year. More recent results with the TAH have been r e p o r t e d b y C o p e l a n d a n d associates [13] in a multicenter study. Since 1993, the C a r d i o W e s t TAH has been used in 40 patients in the United States, Canada, a n d France with 93% survival to discharge after transplantation. In our experience since 1991, the TAH has b e e n u s e d in 5 patients with 100% survival to discharge after transplantation. In the p r e s e n t study, although both early survival a n d late survival were lower in patients with mechanical

A n n Thorac S u r g 1996;61:1734-9

circulatory support, b e y o n d the hospital period, the actuarial survival curves are parallel. The challenge, therefore, r e m a i n s patient selection a n d perioperative m a n a g e m e n t . The m o s t c o m m o n c o m p l i c a t i o n in o u r experience with mechanical circulatory s u p p o r t was infection, with the respiratory tract being the most frequent site. A l t h o u g h resolution can be expected in most patients with a p p r o p r i a t e antibiotic use, prevention m u s t be the goal. The high incidence of infection in these patients is likely multifactorial, related to i m m u n o s u p p r e s s i o n , implantation of the device, reoperation for bleeding, prolonged extracorporeal circulation, and the e m e r g e n c y circumstances u n d e r which all of these devices were implanted. The high incidence of infection also serves to e m p h a s i z e the i m p o r t a n c e of aggressive perioperative m a n a g e m e n t of these patients with early extubation, ambulation, and physical therapy. During this study, patients were m a i n t a i n e d on prophylactic antibiotics t h r o u g h o u t the s u p p o r t period. Currently, patients receive five doses of cefazolin sodium perioperatively, two intraoperatively a n d three postoperatively. In s u m m a r y , transplantation after mechanical circulatory s u p p o r t can salvage m a n y patients who are seen e m e r g e n t l y in profound cardiogenic shock. Patient selection, choice of device, a n d timing of implantation r e m a i n a challenge in the m a n a g e m e n t of these patients, who are at increased risk for device implantation a n d transplantation. With the current t r e n d toward s o m e w h a t more elective use of circulatory assist devices, we m u s t not lose sight of patients who are seen on an e m e r g e n c y basis a n d r e q u i r e m e c h a n i c a l s u p p o r t for resuscitation before transplantation.

References 1. Hendry PJ, Walley VM, Koshal A, et al. Are myocardial temperatures attained by donor hearts during transport too cold? J Thorac Cardiovasc Surg 1989;98:517-22. 2. Lower RR, Shumway NE. Studies on orthotopic homotransplantation of the canine heart. Surg Forum 1960;11:18-9. 3. Grunkemeier GL, Start A. Actuarial analysis of surgical results: rationale and method. Ann Thorac Surg 1977;24: 404-8. 4. Kaye MP. Registry of the International Society for Heart and Lung Transplantation: tenth official report--1993. J Heart Lung Transplant 1993;4:541-8. 5. Ibrahim M, Masters RG, Hendry PJ, et al. Determinants of hospital survival after cardiac transplantation. Ann Thorac Surg 1995;59:604- 8. 6. Pae WE Jr. Ventricular assist devices and total artificial hearts: a combined registry experience. Ann Thorac Surg 1993;55:295-8. 7. Reedy JE, Pennington DG, Miller LW, et al. Status I heart transplant patients: conventional versus ventricular assist support. J Heart Lung Transplant 1992;11:246-52. 8. Frazier OH, Rose EA, Macmanus Q, et al. Multicenter clinical evaluation of the HeartMate 1000 IP left ventricular assist device. Ann Thorac Surg 1992;53:1080-90. 9. Kormos RL, Borovetz HS, Armitage JM, et al. Evolving experience with mechanical circulatory, support. Ann Surg 1991;214:471-7. 10. Farrar DJ, Hill JD. Univentricular and biventricular Thoratec VAD support as a bridge to transplantation. Ann Thorac Surg 1993;55:276-82.

A n n Thorac S u r g 1996;61:1734-9

11. Farrar DJ, Thoratec ventricular assist device principal investigators. Preoperative predictors of survival in patients with Thoratec ventricular assist devices as a bridge to transplantation. J Heart Lung Transplant 1994;13:93-101. 12. Johnson KE, Prieto M, Joyce LD, et al. Summary of the

MASTERS ET AL TRANSPLANTATION AFTER M E C H A N I C A L SUPPORT

1739

clinical use of the Symbion total artificial heart: a registry report. J Heart Lung Transplant 1992;11:103-6. 13. Copeland JG, Pavie A, Duveau D, et al. Bridge to transplant with the CardioWest total artificial heart: the international experience 1993-1995. J Heart Lung Transplant 1996;15:94-9.

INVITED COMMENTARY A n y o n e seriously involved with heart transplantation after the use of assist devices is certainly very much aware of the t r e m e n d o u s difficulties e n c o u n t e r e d and how m o r i b u n d these patients often are. Therefore the group from Ottawa m u s t be congratulated on the imm e n s e efforts and achievements in this respect. Masters a n d associates have p o i n t e d out that the majority of assist devices installed were in patients presenting with p r o f o u n d acute cardiovascular collapse. In this nonelective situation bridging systems are the only feasible alternative. Certain s t a n d a r d i z e d a n d irrevocable selection criteria should a p p l y such as age, the presence of mulfiorgan failure, neurologic events, or systemic infection. In general these should be the same selection criteria as used for evaluation of prospective heart transplant candidates, as w e a n i n g from s u p p o r t systems is rarely possible. The significantly lower age of patients s u p p o r t e d with total artificial hearts in this analysis reflects these prerogatives. Once the patient is stabilized on the assist system the physician m u s t decide if heart transplantation is feasible and, if it is, w h e n to transplant. In a time of donor organ scarcity one m u s t critically assess if we can afford a hospital survival of 60% or less. A constant reevaluation of patient status and the presence of exclusion criteria for transplantation must be p e r f o r m e d on a daily basis. Infection was p r e s e n t in almost 50% of the patients on a

total artificial heart in this study. As the m e a n s u p p o r t time was extremely short (mean s u p p o r t time, 8 + 2 days), this implies that transplantation was often performed in infected patients, a n d this is u n d e r l i n e d by the fact that infection r e m a i n e d the most serious complication after transplantation. In previous studies Frazier and P e n n i n g t o n have b e a u t i f u l l y s h o w n h o w infectious events, renal failure, or neurologic sequelae can be treated or convalescence awaited in patients on total artificial hearts. S u p p o r t times of weeks to m o n t h s m a y be necessary to convert a poor candidate to one in w h o m hospital survival of nearly 90% after transplantation can be expected. If, despite the use of bridging devices, the patient cannot be stabilized a n d the n u m b e r of complications is increasing, then heart transplantation should not be an option. W e are responsible, not only for this special patient cohort, but for the entire waiting list and must therefore a t t e m p t optimal use of donor organs.

Bernard Hausen, MD Division of Thoracic and Cardiovascular Surgery Surgical Center Hannover Medical School D-30623 Hannover Germany