Monday Afternoon, May 3, 1982

MONDAY AFTERNOON, May 3, 1982

2:00 p.m. Forum Session - Assembly Hall

7. Improved Myocardial Protection With Decreasing Temperature During Cardioplegic Arrest

F.L. ROSENFELDT* and G.R. STIRLING*,

Victoria, Australia

Sponsored by: David C. Sabiston, Jr., Durham,

North Carolina

The relationship between recovery and temperature during cardioplegic arrest was studied in canine hearts perfused by a support dog. Isolated hearts underwent 2 hours of arrest between -3°C and 37°C using the St. Thomas' Hospital (K+, Mg++, procaine) solution. Isovolumic left ventricular (LV) developed pressure (DP) was measured before and after arrest. LV biopsies were taken for lactate assay before and for ATP assay and electron microscopy after arrest. The results showed ultrastructural damage only at -3°C and a decrease in protection around 15-20°C:

Temp. °C	-3	4	15	20	25	30	37
n	4	6	7	5	8	6	2
DP % control	0	100	98	95	89	60	4
ATP<% control	24	96	106	88	86	60	12
Lactate µM/a	1.5	1.8	3.1	6.6	9.2	13.8	33.3

In a further study arrest was prolonged for 6 hours to bring out differences in protection in the lower temperature range:

Temperature	°C	4 (n = 6)	12 (n = 6)	20 (n = 5)
DP (mmHg):	Reperfusion	103 ± 5	96 ± 5	56 ± 5
ATP µM/g:	Control	3.8 ± 0.2	3.9 ± 0.2	3.6 ± 0.2
	2 hours	3.7 ± 0.2	3.9 ± 0.1	3.2 ± 0.1
	4 hours	3.4 ± 0.2	3.3 ± 0.2	2.4 ± 0.2
	6 hours	2.6 ± 0.2	2.1 ± 0.2	0.9 ± 0.3
	Reperfusion	2.6 ± 0.1	2.1 ± 0.1	1.1 ± 0.2

Values given are mean ± SEM. Analysis of variance was used for statistical comparisons.

ATP was depleted progressively with time in all groups but most rapidly at 20°C. After reperfusion, the 20°C group showed significantly lower biochemical and functional recovery than the other 2 groups (P<0.001) and the 12°C group showed lower biochemical recovery than the 4°C group (P<0.01).

Conclusion:Protection varies inversely with myocardial temperature and although this effect is most marked down to 20°C there are further increases in protection with cooling to 4°C.

*By invitation

8. Critical Importance of Topical Hypothermia In Cooling Myocardial Regions Supplied by Collateral Flow

JOHN C. LASCHINGER*, FRANK P. CATINELLA*,

JOSEPH N. CUNNINGHAM, JR., PETER X. ADAMS*,

IRA M. NATHAN* and FRANK C. SPENCER,

New York, New York

The importance of profound hypothermia in protecting the myocardium during cardioplegic arrest is well known. However, as is often the case in pts. with severe coronary artery disease, regions of the myocardium distal to critical stenoses supplied solely by small collateral vessels are underperfused and inadequately cooled. The following study was undertaken in a canine model of chronic progressive coronary occlusion to assess the importance of added topical hypothermia in augmenting myocardial cooling in regions of collateral flow.

Dogs (12) had hydroscopic ameroid constrictors placed at the origin of the circumflex coronary artery and mid-left anterior descending coronary artery. Eight weeks later after angiographic demonstration of vessel occlusion and collateral development, myocardial ischemia was confirmed by atrial pacing (180 BPM) and noting ST changes. Animals underwent 60 min. of aortic crossclamping (AXC). Hearts were protected with topical hypothermia (normal saline, temp. 4°C) and blood cardioplegia (temp. 10°C, pH 8.0, KC1 30 mEqs/L, 1 initially and then 500 cc every 30 min.). Subendocardial temperatures in both normal and collateralized myocardial regions were measured with thermistor probes after each injection and following the addition of 1 liter of topical hypothermia (TH).

The results are as follows (all values expressed as mean ± SEM):

		DISTAL LAD		RIGHT
	SEPTUM	REGION	CX REGION	VENTRICLE
Post Injection	12 ± .6°C	15 ± .9°C	13 ± .6°C	13 ± .7°C
Post TH	12 ± .7°C	11 ± .6°C	12 ± .6°C	9 ± .4°C
P Value	NS	<.001	<.025	<.001

These data demonstrate the importance of adjunctive topical hypothermia in cooling regions of the myocardium distal to critical coronary stenoses with poor distribution of cardioplegia solution. Furthermore, cooling those areas of the myocardium with the greatest amount of surface area exposed to ambient temperatures (e.g. anterior right ventricle) will also be improved.

*By invitation

9. Benefits of Normothermic Blood Cardioplegic Induction of Prolonged, Multidose Cold Blood Cardioplegia in Energy Depleted Hearts

ELIOT R. ROSENKRANZ*,

JAKOB VINTEN-JOHANSEN*,

GERALD D. BUCKBERG and HANNIBAL EDWARDS*

Los Angeles, California

High risk cardiac patients arc more vulnerable to intraopcrative damage because myocardial ATP depletion may be present before aortic clamping and therefore limit myocardial recovery following cardioplegia. To simulate this experimentally, we produced 47%* ATP depletion in dogs by 45 minutes of normothermic ischemic arrest, followed immediately thereafter by 2 more hours of aortic clamping with 4°C multidose blood cardioplegia. We compared the effects of starting the blood cardioplegic infusion at 37 °C (for 5 minutes before cooling it 10 4°C) with those of immediate 4°C blood cardioplegic infusion. The study was designed lo determine if the initial 37 °C blood cardioplegic infusion would enhance energy replenishment by allowing more oxidative metabolism than that possible if the Q₁₀ effect of 4°C hypothermia was immediate.

During the 5 minute period of cardioplegic induction, dogs receiving 37°C initial blood cardioplegia had a 3.3cc/100g/min* (421%)* greater oxygen uptake than when 4°C blood cardioplegia was used. Post-ischemic ventricular function recovered only 33%* in 10 control dogs undergoing only 45 minutes of aortic clamping (no cardioplegia). Better functional recovery, to 74%* of control, occurred in 5 dogs undergoing an additional 2 hours of aortic clamping when 4°C multidoes blood cardioplegia was used. In contrast, complete recovery (100%)* at a 42% lower left atrial pressure was seen in the four dogs receiving the same cardioplegia dose, but where the first 5 minutes of cardioplegic infusion was warmed to 37°C before inducing myocardial hypothermia.

We conclude that, in energy depleted hearts, enhancing myocardial oxidative metabolism by giving the initial blood cardioplegia dose at 37°C (before inducing hypothermia) improves the heart's tolerance to the subsequent aortic clamping required for cardiac repair.

*p < 0.05 compared with control

*By invitation

10. The Control of Myocardial Ca⁺⁺ Sequestration With Nifedipine/Potassium Cardioplegia

STUART L. BOE*, CHARLES M. DIXON*,

TAMARA A. SAKERT* and GEORGE J. MAGOVERN

Pittsburgh, Pennsylvania

AIM: To determine (a) when and to what extent calcium ion (Ca⁺⁺) accumulation occurs in the ischemic and reperfused myocardium; (b) Can Ca⁺⁺ uptake be inhibited by a specific Ca⁺⁺ antagonist, Nifedipine? METHOD: Mongrel dogs were placed on total cardiopulmonary bypass at 32°C. Aortic crossclamp (AXC) was applied for 1 hour and 300 cc. of cardio-plegic solution at 4°C was infused al 30 minute intervals. Fifteen minutes after AXC,⁴⁵CaCl was injected into the femoral artery, and a steady state of ₄5Ca distribution maintained for 130 minutes. Myocardial biopsies were taken after 1 hour of AXC (T₀) and T₅, T,₀, T,₅, T₂₀, T₃₀ minutes of reperfusion. The source of the ⁴⁵Ca⁺⁺ available to the myocardium was the Serum Ca⁺⁺expressed as the Specific Activity (SA = [(cpm X 10⁴)/1.0 gm]). and myocardial uptake expressed as Relative Specific Activity (RSA ={(SA tissue)/ (SA serum)]). Group 1 (N = 10) received standard cardioplegic solution (SCS). Group II (N = 10) received SCS plus Nifedipine, 20 me/kg. T₀ values represent sequestration of ⁴⁵Ca⁺⁺ in the myocardium via collaterals despite one hour AXC. During all time frames of reperfusion, Nifedipine significantly (p < .01) inhibited myocardial uptake. The results show Nifedipine 1) aids maintenance of calcium ion homeostasis during AXC and reperfusion; and 2) blocks calcium influx into the ischemic myocardium during reperfusion. The delay of RSA returning to unity in Group II implies a beneficial effect by inhibiting Ca^{+ +} transport in early reperfusion, preventing excessive energy consumption by mitochondrial Ca' sequestration, best illustrated in Group I at T₂₀ when Ca⁺⁺ uptake was 5 times normal and confirmed by reduced ATP levels in previous studies. This histochemical sludy provides a rationale for the previously shown improved experimental and clinical hemodynamic function when Nifedipine is added to standard hyperkalemic cardioplegia.

RELATIVE SPECIFIC ACTIVITY
TIME	GROUP I	GROUP II
T₀	0.92 ± .09	0.63 ± .03
T₅	1.74 ± .07	0.67 ± .02
T₁₀	1.92 ± .06	0.69 ± .02
T₁₅	2.30 ± .18	0.75 ± .02
T₂₀	5.15 ± .56	0.78 ± .02
T₃₀	2.17 ± .06	1.31 ± .03

*By invitation

11. Experimental Evaluation of Magnesium Cardioplegia

AKIO WAKABAYASHI, TSUNEHIRO NISHI*

AND EDWARD J. GUILMETTE*

Irvine, California

Magnesium is a major active ingredient in all three representative European cardioplegic solutions bul its effectiveness has not been well defined because their data were based on experiments involving multiple components. Therefore, the present study was undertaken to determine the protective effect of magnesium per sc using it as a sole active ingredient. Methods. Thirty-six excised rabbits hearts were perfused with oxygenated blood, while the maximal net developed tension (T_Nmax) of the isometrically contracting posterior papillary muscle was determined. When T_Nmax became stabilized, anoxic normothermic cardioplegia was induced by infusing 10 ml of either control solution (Na+ 135, K+ 5, Cl^- 140 mEq/L) or test solution (Mg+ + 140, SO₄^-- 140 mEq/L) into the aortic cannula. The heart was immersed in a waterbath (37 C) for 30, 45, or 60 min., six experiments in each group. Percent recovery of T_N|1,._IX was detemined after a full recovery. Results: Magnesium solution induced cardiac arrest within one minute which was significantly faster than the control (10.9 ± 1.9 min). The time required to restart the contractions with reperfusion was not significantly different between the control and test groups. After 30 minutes of magnesium anoxia, % recovery was 91.3 ± 8.3, which was significantly (p<0.01) better than the controls 77.7 ± 4.1. When the anoxic time was more than 45 min., however, this protective effect of magnesium could not be demonstrated; after 45 min., 65.1 ± 16 (control) vs. 55.1 ± 27.1 (magnesium), after 60 min. anoxia, 4.5 ± 5.1 (control) vs 18.6 ± 11.1 (magnesium). Conclusion: Magnesium per se has a protective effect on ischemic myocardium. Percent recovery of myocardial contractility following 30 minutes of normothermic anoxia observed in the present study was very similar to the result achieved by potassium (40 mEq/1) or procaine (0.02%) in our earlier studies. As with these two other active cardioplegic agents, however, the protective effect of magnesium rapidly diminishes as an anoxic time exceeds 30 minutes at normothermia.

*By invitation

12. Adequacy of the Reperfusate: Its Influence In Successful Myocardial Protection

IGNACIO Y. CHRISTLIEB* and RICHARD E. CLARK

St. Louis, Missouri

Previous experiments have indicated that the characteristics of the perfusate at the initiation of myocardial reperfusion after long ischemic inervals (180 min) and long hypothermic (22°C) perfusion times (220 ± 10 min) influenced post ischemic myocardial function despite adequate protection during ischemia. 6 dogs (Group A) were perfused with 5% glucose in Ringer's solution (290 mOsmol) and 6 dogs (Group B) were perfused with 5% glucose-hetastarch solution containing mannitol (osm = 360 ± 20 mOsmol). Priming volumes and oxygenators were identical in the two groups. Perfusion times, and base excess (+ 6 - 8 mEq/L) at the initiation of reperfusion after 3 hr. of global ischemia were equal. Both groups had had identical number and composition of cardioplegic infusions. Complete hemodynamic studies demonstrated that the two groups were equal in terms of cardiac index (CI), left ventricular stroke work index (LVSWI), LVEDP, LV dP/dt and response to volume loading prior to perfusion. 120 min. after cessation of perfusion, the group A dogs had 50% of normal LVSWI (10-12 gm-m/m²) versus 20-22 gm-m/m² for group B dogs. CI's were 1.9 ± 2 vs. 3.6 ± 3 for the A and B groups respectively. Group A dogs had a flat response to volume loading 100 min. after initiation of reperfusion in contrast to group B which had a normal response. Group A required more crystalloid and buffer addition during perfusion. It is concluded that low sodium concentrations, hyperosmolarity and onconicity of the perfusate particularly at the onset of reperfusion significantly influences the myocardial performance immediately and for 2 hrs. after cessation of cardiopulmonary bypass in dogs. These data suggest that the reperfusion phase of cardiac operations is as important as the composition and administration of cardioplegic solutions and may acccount for differences in reports between investigators.

*By invitation

13. Platelet Deposition After Surgically Induced Myocardial Ischemia: An Etiologic Factor For Reperfusion Injury

HAROLD FEINBERG*, DAVID ROSENBAUM*,

SIDNEY LEVITSKY and NORMAN A. SILVERMAN*

Chicago, Illinois

Despite meticulous adherence to presently known principles of myocardial preservation, reperfusion after aortic cross-clamping results in a unique injury manifested by decreasing high energy phosphate levels and increased coronary resistance. We hypothesize that platelet deposition (PD) into the coronary microvasculature is a major factor in reperfusion injury. To differentiate PD due to subendocardial hemorrhage from deposition to vascular entrapment, ¹¹¹In-labeled platelets together with ⁵¹Cr-labeled erythrocytes were infused in 15 dogs on normothermic bypass and subjected to 60 minutes of global ischemia followed by 30 minutes of reperfusion. PD is indicated only when the proportion of platelets to erythrocytes in tissue exceeds that measured in peripheral blood. Myocardial biopsies were obtained after 120 minutes of continuous bypass (Group I) and at the end of reperfusion after global ischemia (Group II). In 5 dogs (Group III), dipyridamole (1 mg/kg), a potent anti-platelet activation agent, was administered in the pre-ischemic period. PD was expressed as a ratio of ¹¹¹In before and after ischemia.

Bypass for 120 min. resulted in only a minimal increase in PD. However, normothermic ischemia followed by reperfusion resulted in over a twofold increase in PD compared to controls.. Pre-treatment with dipyridamole appeared to avoid PD. These data indicate that platelet deposition in the coronary microcirculation following surgically induced myocardial ischemia may be associated with reperfusion injury and that anti-platelet drugs alter this sequence.

3:15 p.m. Intermission - Visit Exhibits

*By invitation

4:00 p.m. Scientific Sessions - Assembly Hall

14. A Critical Reassessment of the Performance Characteristics of the Starr-Edwards 6120 Mitral Valve Prosthesis Ten Years Later

D. CRAIG MILLER*, PHILIP E. OYER*,

EDWARD B. STINSON*, BRUCE A. REITZ*,

STUART W. JAMIESON*, WILLIAM A. BAUMGARTNER*

and NORMAN E. SHUMWAY Stanford, California

Hesitancy regarding the long-term (10 yr) durability of porcine bioprostheses and the thrombotic complictions associated with tilting-disc mechanical prostheses have prompted resurgence in use of the Starr-Edwards (S-E) metal strut, silastic poppet (Model 6120) mitral valve. Despite 16 years of clinical availability, comprehensive data pertaining to its 10-15 year performance characteristics are unavailable. Therefore, we assessed the long-term results attained with this valve in 509 patients undergoing isolated MVR between 1965-1974. Folow-up totalled 3,157 patient-years (pt-yrs), extended to 15.5 yr maximum, and averaged (±SD) 9.8 ± 3.7 yrs. 124 pts were followed >10 yrs. Specific emphasis was focused on valve-related morbidity and mortality. Anticoagulant-related hemorrhage (ACH) occurred at a linearized rate of 3.8%/pt-yr, thromboembolism (TE) at 5.7%/pt-yr, MVR reoperation (RE-OP) at 1.6%/pt-yr, valve failure (VF; defined by one or more of the following criteria: new murmur of mitral regurgitation, thrombotic or tissue valvular occlusion or multiple TEs leading to reoperation or death, or prosthetic valve endocarditis causing reoperation or death) at 3.5%/pt-yr, and late death at 7.2%/pt-yr. Actuarial rates (± SEM) of being free of these complications were:

	ACH	TE	RE-OP	VF	Late Death
5 years	81 ± 2%	65 ± 2%	93 ± 1%	79 ± 2%	71 ± 2%
10 years	67 ± 3%	55 ± 3%	84 ± 2%	73 ± 2%	47 ± 3%
14 years	52 ± 5%	48 ± 3%	80 ± 3%	69 ± 3%	38 ± 4%

54% (60/111) were caused by multiple TE episodes. Although the intrinsic structural integrity of the S-E 6120 prosthesis is sound for 10-15 yrs, its overall "durability" is clearly suboptimal due to tissue and thrombotic valvular obstruction, thromboemboli, and other factors. Furthermore, when anticoagulant-related hemorrhage is added to each patient's calculated risk equation, the overall incidence of S-E valve-related mobidity at 10 yrs appears to favor tissue bioprostheses even if 20-30% or more of xenografts were to require replacement by 10 yrs. Only similar large scale comparative analyses of the performance characteristics of xenografts and tilting-dise valves in this time frame will resolve these questions.

*By invitation

15. Early Valve Replacement In Active Infective Endocarditis: Results and Late Survival

RAMON A CUKINGNAN*, JOSEPH S. CAREY,

JOHN H. WITTIG* and GEORGE E. CIMOCHOWSKI*,

Inglewood and Los Angeles, California

In the last 14 years, 42 patients with active infective endocarditis underwent early valve replacements because of severe congestive heart failure, major prosthetic dehiscence, intramyocardial abscesses, or major embolization. Forty patients had positive blood cultures, and in 2 others, the valve tissues were positive. All patients were on antimicrobials for greater than 1 week and under 4 weeks. Drug addiction was noted in 24%, urinary tract manipulation in 7%, dental work in 5%, and unknown cause in 64%. Organisms were predominantly Staphylococcal (43%) and Streptococcal (41%) and 16% were gram negative (9%) or fungal (7%). The aortic valve was involved in 72%, mitral in 14%, tricuspid in 7%, and both aortic and mitral in 7%. By NYHA class, 90% (38/42) were in Class III or IV.

Operative mortality was 10% (4/42), and all had pre-existing renal failure. No predominant organism correlated with early deaths. In aortic valve replacement (30 patients), operative mortality was 7%. Postoperatively, 94% (35/37) are Class I or II with 1 lost to follow-up. Subsequent reoperation was required in 5 patients (13%) for recurrent endocarditis and major valve dehiscence with an operative mortality of 20% (1/5). Late death occurred in 45% (17/38), being 43% for aortic valves, 67% for mitral, 30% for tricuspid, and none for the combined. Major causes of death were sudden death, recurrent endocarditis, and unknown cause. On analyzing early and late mortality, all patients with fungal endocarditis were dead on follow-up. Similarly, Ser-ratia and Pseudomonas infections resulted in a 100% mortality. Staphylococcal endocarditis had a follow-up mortality of 44% compared to 41% for Streptococcal species. Overall survival by life-table method was 0.71 at 1 year and 0.52 at 5 years. Aortic valve endocarditis has a 5-year survival of 0.58.

This study shows that early operation after at least 1 week of antibiotics in patients with active endocarditis has an operative mortality comparable to non-infected valve replacements. Clinical improvement is excellent in 94%. Poor overall late survival may be due to the more virulent fungal, Staphylococcus epidermidis, and gram-negative organisms.

*By invitation

16. Left Ventricular Outflow Enlargement Using the Konno Procedure

GREGORY A. MISBACH*, KEVIN TURLEY*,

DANIEL J. ULLYOT and PAUL A. EBERT

San Francisco, California

The management of patients with small aortic annulus or left ventricular outflow tract obstruction remains unclear. Between 1976 and 1981, 14 patients have undergone enlargement of their left ventricular outflow tract, using the Konno or a modification of the Konno type procedure. Ten of these 14 patients had previous operations for aortic stenosis or tunnel left ventricular outflow tract and two patients had undergone three previous operations. All 14 patients had symptoms of either heart failure or chest pain, or had electrocardiograph evidence of strain. They ranged in age from 5 years to 57 years, with 12 of the 14 patients being less than 25 years of age.

A dacron patch was used to enlarge the left ventricular outflow tract after incising down the ventricular septum. In all patients, at least a 23mm valve was able to be placed with between 50 and 65 percent of the valve annulus being made up of natural tissue. The remaining portion of the valve annulus was constructed from the dacron patch. The patch was extended up the aorta to enlarge the ascending aorta and a pericardial patch was used to close the defect in the right ventricular outflow tract. In all 14 patients the gradient was obliterated at the time of operation. There were no operative deaths, but there has been one late death due to bacterial endocarditis in a child who also had a parachute mitral valve and evidence of pulmonary hypertension. The remaining 13 patients are functioning well after the Konno procedure. Two are receiving Coumadin, and 11 are receiving aspirin. These results suggest that this is an acceptable method of treating patients with small aortic annulus or left ventricular outflow tract obstructions and would appear to have advantages over a left ventricular apical aortic conduit.

*By invitation

17. The Surgical Treatment of Ventricular Tachycardias (Simple Aneurysmectomy Versus Electrophysiologically Guided Surgery)

JORG OSTERMEYER*, RALF KOLVENBACK*,

GUNTER BREITHARDT*, LUDGER SEIPEL*,

HAGEN D. SCHULTE* and WOLFGANG BIRCKS

Dusseldorf, West Germany

Sponsored by: JOHN W. KIRKLIN,

Birmingham, Alabama

Between 1975 and 1981 35 patients had surgery for life threatening recurrent ventricular tachycardia (VT) resistant to antiarrhythmic drug therapy. All except of 3 had severe coronary artery disease with a history of myocardial infarction. In 10 patients (group I) a simple aneurysmectomy was done, 25 (group II) had an electrophysiologically guided (epicardial/endocar-dial mapping) encircling endocardia! ventriculotomy at the earliest electrical activity during VT. There were no significant differences between the groups in age, sex ratio, NYHA class, coronary disease, aneurysm location, concom-mittant bypass grafting and left ventricular function.

All patients have been restudied by means of 24h-ECG-monitoring, programmed stimulation and/or the ECG-signal-averaging-technique 2-4weeks after surgery. The mean follow-up period in group I is 3,8 years, in group II 22,8 weeks. In group I we had 1 early and 4 late deaths after surgery (50%mortality) mainly due to persistent arrhythmias. 1 patient is free of VT (10%), 3 require still antiarrhythmics, 1 patient's post-operative history is unknown. Out of group II we lost 1 patient intraoperatively, there were no late deaths (4% mortality). 19 patients (76%) are free of VT clinically, they have no persisting late potentials or signs of ventricular irritability at programmed stimulation. 5 patients are under antiarrhythmic drugs, 1 patient has an antiarrhythmic pacemaker.

We conclude that an electrophysiologically guided approach improves the efficacy of the surgical treatment of VT significantly in comparison to simple, "blind" aneurysmectomy.

*By invitation

18. Changes in Indications for Heart Transplantation, An Additional Argument For the Preservation of the Recipient's Diseased Heart

JACQUES G. LOSMAN*, HARRY D. LEVINE*,

CHARLES D. CAMPBELL*, ROBERT L. REPLOGLE

and CHRISTIAAN N. BARNARD*

Chicago, Illinois and Cape Town, South Africa

Heart transplantation (HT) one year (y) survival improved since 1967 from ±30% to ±70%, 5 y survival is now ±50%. This brought renewed interest in HT, now done in about 20 centers in 5 countries, and increased confidence has widened the indication to patients (pt) less than terminally ill, to restore quality of life. This trend is illustrated by the Cape Town series, when 1967-74 pt [CT₁ = 10 orthotopic HT (OHT)] are compared to 1974-78 pt [CT₂ = 20 heterotopic HT (HHT)]. CT₂ pt are younger, m = 38.2±2.2 y than CT₁, pt, m = 51.1±2.7 y [p<0.001], CT₂ illness is shorter, m = 3.6 ± 0.7 y, versus CT1 m = 6.6±1.4 y [p<0.019], and CT₂ NYHA Class is less, m = 3.45 ± 0.11 versus CT₁, m = 3.9 ± 1.0, [p<0.006]. The improved survival is linked to pt selection, progress in management and switch to HHT, but not to progress in matching between donor (d) and recipient (r). Since there is no means to predict tolerance of the d heart (dh), HHT limits the risks from unforseeable mismatch. The retained r heart (rh) is a build-in assist device, life saving when dh fails acutely at surgery, during acute [3 pt] or chronic [2 pt] rejection. Had these pt been OHT r, they would have died. Comparing the 10 oldest HHT r with CT₁, no difference in pre-HT parameters was found. However, HHT r survival was longer during the critical post-HT period: at 3 months p<0.011, at 6 months p<0.05. Larger series will separate effects of management progress from HHT intrinsic advantages. Retaining the rh is logical, and brought few complications. Survival of 38 HHT r was 74% at 6, 67% at 12 and 51% at 36 months, 85% of survivors are in NYHA Class 1. In pt less than desperate, but who refuse to remain cripples, HHT eliminates the growing ethical problem of removing a rh that may still support its owner.

*By invitation