Benefits of normothermic induction of blood cardioplegia in energy-depleted hearts, with maintenance of arrest by multidose cold blood cardioplegic infusions

J THoRAc CARDIOVASC SURG 84:667-677, 1982

Benefits of normothermic induction of blood cardioplegia in energy-depleted hearts, with maintenance of arrest by multidose cold blood cardioplegic infusions This study tests the hypothesis that warm induction of cardioplegia prior to prolonged maintenance by multidose infusions of cold blood cardioplegic solution would increase the tolerance of energy-depleted hearts to subsequent aortic clamping. Eighty percent* depletion of subendocardial adenosine triphosphate (ATP) was produced in 30 dogs by 45 minutes of normothermic ischemia. This was followed either by unmodified blood reperfusion or 2 additional hours of aortic clamping with multidose cold blood cardioplegia. We compared a brief (5 minute) period of 37° C cardioplegic induction to standard 4° C blood cardioplegic induction to determine if warm induction would enhance metabolic and functional recovery. Warm cardioplegic induction resulted in more oxygen consumption than cold induction (/6.9 versus 8./ eel /00 gm)*, and lower levels of glucose-6-phosphate (G6P), suggesting better aerobic metabolism (0.97 versus /.87 pM Igm wet weight). * Prompt repletion of creatine phosphate (CP) occurred with warm and cold cardioplegic induction, although ATP levels remained low. Hearts undergoing ischemia and unmodified reperfusion consumed insufficient oxygen to meet basal metabolic needs during reperfusion (7 cc/lOO gm below requirement)* and recovered only 33% ± 5%* of control left ventricular performance. Better function occurred with cold cardioplegic induction (63% ± 5%), * and almost complete recovery (85% ± 5%) occurred when warm induction of cardioplegia was used. We conclude that warm induction followed by prolonged cold multidose blood cardioplegic arrest enhances aerobic metabolism. results in normal left ventricular performance. and improves tolerance of aortic clamping in energy-depleted hearts.

Eliot R. Rosenkranz, M.D. (by invitation), Jakob Vinten-Johansen, Ph.D. (by invitation), Gerald D. Buckberg, M.D., Fumiyuki Okamoto, M.D. (by invitation), Hannibal Edwards, M.D. (by invitation), and Helen Bugyi, Ph.D. (by invitation), Los Angeles, Calif.

COld cardioplegia has lowered operative morbidity and mortality substantially by reducing energy depletion and cardiac metabolic demands during aortic clamping. There remains, however, a subset of patients (i.e., those in New York Heart Association Class IV, cardiogenic shock, etc) in whom significant risk of intraoperative damage still exists, especially when long From UCLA Medical Center. Los Angeles. Calif. Read at the Sixty-second Annual Meeting of The American Association for Thoracic Surgery. Phoenix. Ariz., May 3-5,1982. Address for reprints: Gerald D. Buckberg, M.D .• UCLA Medical Center. Department of Surgery /Thoracic, Los Angeles. Calif. 90024. *p < 0.05. 0022-5223/82/110667+ 11$01.10/0

© 1982 The

C. V. Mosby Co.

periods of aortic clamping are needed. \, 2 Hypertrophied hearts, for example, have reduced tolerance to standard ischemic insults, 3 perhaps because of chronic depletion of energy reserves before aortic clamping. 4. 5 A similar depletion of energy stores may occur in patients who are in cardiogenic shock or who have a cardiac arrest before cardioplegia can be initiated. 6-8 Blood cardioplegia may be particularly well suited for use in hearts with preexisting energy depletion, since blood cardioplegia has the advantage of preventing further energy loss during aortic clamping when the solution is delivered cold.?: 10 Additional benefits include its capacity to avoid and reverse reperfusion damage and improve metabolic recovery when given normothermically to hearts injured by previous ischemia.t":" 667

The Journal of Thoracic and Cardiovascular Surgery

6 6 8 Rosenkranz et al.

Heater - Cooler

CONTROL

I Ischemia

Cordiopleqic reservoir

(45 min)

I i- 4';-C- -SI;od-:

Cold

®

Warm

i

: Cardioplegia : (2 hr) :

Oxygenator : Post : Ischemia

:_~3.9~~n~_1 Fig. 1. Experimental protocol. See text for description.

The present study was undertaken to determine if the concept of using normothermic blood cardioplegia to optimize the rate of repair of ischemically damaged metabolic processes could be applied to energy-depleted hearts that must undergo a subsequent interval of prolonged aortic clamping. Our usual blood cardioplegia technique was used in hearts depleted of energy by 45 minutes of normothermic ischemia. The results will show that a brief period of warm blood cardioplegia prior to prolonged arrest by multidose infusions of a cold blood cardioplegic solution enhances the safety of aortic clamping and increases recovery from ischemic damage. Methods Thirty adult mongrel dogs (16 to 27 kg) were anesthetized with thiamylal (35 mg/kg) and a-chloralose (75 mg/kg) and ventilated by positive-pressure endotracheal ventilation. A median sternotomy was performed and polyethylene catheters were introduced into the distal aorta, inferior vena cava, and left atrium for pressure monitoring and infusion. After systemic heparinization (3 mg/kg), cannulas were placed into the left subclavian artery, superior and inferior venae cavae, and pulmonary artery for performance of either total or right ventricular cardiopulmonary bypass. The left ventricle was vented through an apical stab wound and the right ventricular and coronary sinus effluent was collected via a cannula in the right ventricle. A double-lumen arterial cannula was placed into the aortic root for cardioplegic infusion and monitoring of the infusion pressure. All hearts were kept at 150 beats/ min by atrial pacing.

Fig. 2. Cardioplegic delivery system. See text for description.

Table I. Blood cardioplegic solution Hematocrit (%) Osmolarity (mOsm) pH Potassium (rnEq/L) Ionic calcium (rnEq/L)

28 280

7.8 16

1.0

Intramyocardial temperature was monitored with a thermistor probe in the intraventricular septum (Yellow Springs Instrument Co., Yellow Springs, Ohio). Systemic temperatures were monitored with a rectal temperature probe and arterial and cardioplegic temperatures were monitored by inline thermistor probes (Model 12100, Sarns, Inc., Ann Arbor, Mich.). Mean arterial pressure was kept at 100 mm Hg, hematocrit value at 30% ± 2%, and body temperature at 28° C. The pH was kept at 7.4 while at 37° C and varied appropriately during hypothermia. 14 Measurements. Hemodynamic and metabolic measurements were made approximately 15 minutes after establishing extracorporeal circulation and 30 minutes after release of the aortic clamp. Total coronary blood flow was measured by collecting the right ventricular effluent. Blood oxygen content was measured by the method of Behar and Severinghaus'" and myocardial consumption (MV0 2 ) was calculated from the equation: MV0 2 = CBF x (A - V)

where CBF = coronary blood flow, A = arterial oxygen content, and V = right ventricular venous oxygen content. Values were expressed as cc/loo gm/min to normalize for heart weight. In all studies, ventricular performance was measured by inscribing right heart bypass curves as described previously;" Results were expressed as stroke work index (SWI):

Volume 84

Normothermic induction of blood cardioplegia

Number 5 November, 1982

669

Table ll. Ventricular blood flow and oxygen metabolism during cardioplegic induction* Induction of cardioplegia 2 min

I min Warm MV02 (ee/IOO gm/rnin) Cardioplegic solution flow (ee/IOO gm/min) Extraction (%) No.

6.2 ±0.7t 126.9 ±13.8 39.9 ±4.7t 6

I

I Cold

Cold

Warm

3.4 ±0.6 105.5 ±5.5 25.9 ±4.6 5

3.6 ±O.4t 113.8 ±13.5t 21.5 ±2.3 6

3 min Warm

2.9 1.8 ±0.2 ±0.6t 86.0 125.0 ±5.7 ± 15.4t 16.5 19.3 ±2.4 ±4.8 5 6

I Cold

4 min Warm

1.1 2.2 ±0.2 ±0.3t 82.1 113.2 ±5.0 ±14.7 10.9 15.3 ± 1.7 ±4.2 5 6

I

Cold

5 min Warm

1.1 2.1 ±0.3 ±0.2t 76.6 107.7 ±5.2 ±14.6 10.5 16.2 ±2.9 ±2.4 5 6

I

Cold 0.8 ±0.3 72.0 ±3.2 8.8 ±2.8 5

6-/0 min Warm 0.9 ±O.I 73.1 ±4.1 10.6 ±O.I 30

I Cold 0.6 ±O.I 66.0 ±2.8 7.3 ±0.8 25

Legend: MVO" Myocardial oxygen uptake. Minutes 6-10 are pooled data. Wann, Wann induction, Cold, Cold induction.

'Mean:!: SEM. tp < 0.05 versus cold induction.

SWI (grn-rn /k g) =

(MAP - LAP) CO Heart rate x body weight

1.36 1()()

X --

where MAP = mean aortic pressure (75 mm Hg), LAP = left atrial pressure, and CO = cardiac output. Full-thickness biopsy specimens of the left ventricular free wall were obtained with a high-speed drill and were divided into endocardial and epicardial layers of approximately equal thickness. These specimens were frozen immediately in liquid nitrogen and analyzed subsequently by fluorometry for adenosine triphosphate (ATP) , creatine phosphate (CP), and glucose-e-phosphate (G6P).J7 Parallel biopsy specimens were analyzed for left ventricular water content by drying to a constant weight at 85° C. Statistical analysis. Computations were performed by means of the Statistical Analysis Systems (SAS Institutes, Carey, N. C.) statistical computer package provided by the UCLA Hospital Computing Facility. Paired Student's t test was used in comparing experimental values to control values in individual dogs. Group data were compared by the analysis of variance and Duncan Multiple Range tests. 18 Significant differences were defined as probabilities for each test of p < 0.05. Pilot studies. Basal oxygen requirements were measured in hearts that did not undergo ischemia. Myocardial oxygen uptake was measured in the beating empty state (heart rate 150 beats/min) at 37° C. Repeat measurements of oxygen uptake were then made (I) after arrest with the 37°C blood cardioplegic solution (Table I) to be used in subsequent studies and (2) when the temperature of the blood cardioplegic infusion was lowered to 8° C (myocardial temperature = 10° C). Experimental studies. All 30 hearts underwent 45 minutes of normothermic ischemia in order to deplete

myocardial energy reserves (measured by left ventricular biopsy after 45 minutes of ischemia). Dogs were then placed into one of two groups: Forty-five minutes of normothermic ischemia (no cardioplegia). Ten hearts underwent 45 minutes of normothermic ischemia followed by 30 minutes of blood reperfusion at 37° C. Forty-five minutes of normothermic ischemia plus 2 hours of multidose cardioplegia. In 20 hearts, the 45 minutes of normothermic ischemia was followed by 2 more hours of aortic clamping with multidose infusions of the cold blood cardioplegic solution. The initial cardioplegic infusion was given over 10 minutes at an infusion pressure of 50 mm Hg, and each reinfusion was given over 2 minutes every 20 minutes. The volumes of cardioplegic solution given during the initial 10 minute infusion are listed in Table II. The dogs were subdivided into two groups, which differed only in the temperature of the cardioplegic solution given during the first 5 minutes of cardioplegic infusion (Fig. 1). WARM INDUCTION OF CARDIOPLEGIA. In 10 hearts, the cardioplegic solution was infused at 37° C for the first 5 minutes. Following this warm induction period, the temperature of the solution was lowered to 4° to 8° C by the cardioplegic delivery system described previously'? (Fig. 2), and the cold cardioplegic solution was infused over 5 additional minutes and replenished each 20 minutes. COLD INDUCTION OF CARDIOPLEGIA. In 10 hearts, the entire 10 minute induction of cardioplegia was carried out at 4 to 8° C, and a similar temperature was used for replenishments of the cardioplegic solution. The last cardioplegic dose was given 9° C above septal temperature prior to aortic unclamping. Systemic blood pressure was lowered to 50 mm Hg before un-

670

The Journal of Thoracic and Cardiovascular Surgery

Rosenkranz et al.

Table III. High-energy phosphate content* Cardioplegia Induction Control

Postischemia (45 min)

Warm

4.1 ± 0.2 4.0 ± 0.4

1.6 ± 0.3t 0.8 ± O.lt

2.1 ± 0.3t 0.9 ± 0.2t

1.5 ± 0.2t 1.4 ± 0.3t

9.6 ± 0.8 7.9 ± 0.8

0.6 ± 0.2t 0.4 ± O.lt

15.2 ± l.4t 11.5 ± 1.6t

12.8 ± I.2t 9.7 ± 0.6t

0.15 ± 0.05 0.13 ± 0.05 77.4 ± 0.2

1.33 ± 0.29t 1.95 ± 0.25t 78.0 ± 0.2t

0.48 ± 0.05:j:§ 0.97 ± 0.20:j:§

1.26 ± 0.28 1.87 ± 0.27

I

Cold

ATP

Epi Endo CP

Epi Endo

G6P

Epi Endo

H 20li

Legend: Control and postischemia values are pooled. ATP. Adenosine triphosphate. CPo Creatine phosphate. G6P, Glucose-6-phosphate. Epi, Epicardial. Endo, Endocardial. 'Mean ± SEM (ILM/gm wet weight).

tp < 0.05 from control. +p < 0.05 from cold induction. §p < 0.05 from postischemia. [Percent water content.

clamping and increased to 100 mm Hg over the next 3 minutes as cardiac mechanical activity resumed.

Results Pilot studies (no ischemia). At 37° C, myocardial oxygen uptake in the beating empty heart was 4.4 ± 0.4 cc/100 gm/min. In contrast, myocardial oxygen uptake fell to 1.0 cclloo gm/min when these normothermic hearts were arrested with a warm (37° C) cardioplegic solution and fell further to 0.3 cclloo gmt min or less when the temperature of the cardioplegic solution was reduced to 4° to 8° C. Experimental studies. Forty-five minutes of normothermic ischemia (no cardioplegia). Hearts undergoing normothermic ischemia were characterized by the following: (1) a striking reduction in endocardial ATP from 4.0 ± 0.4 to 0.8 ± 0.1 JLM/gm wet weight (Table III); (2) ventricular fibrillation following unclamping, necessitating between 2 and 4 countershocks; (3) myocardial oxygen uptake 6.6 cc/loo gm below basal requirements (p < 0.05) for the beating empty state during the 10 minutes of reperfusion (Fig. 3); and (4) only 33% ± 5% recovery of left ventricular function 30 minutes after aortic unclamping (p < 0.05, Fig. 4). Forty-five minutes of normothermic ischemia plus 2 hours of multidose blood cardioplegia. The additional 2 hours of aortic clamping with multiple infusions of the blood cardioplegic solution resulted in (1) spontaneous defibrillation in 19 of 20 (95%) hearts following

aortic unclamping and (2) better recovery of ventricular performance than seen after 45 minutes of ischemia alone (Fig. 4). The best metabolic and functional recovery occurred when multidose cardioplegia was given normothermic ally for the first 5 minutes, as described below. OXYGEN UPTAKE. Hearts subjected to warm blood cardioplegic induction consumed 9 cc/loo gm more oxygen during the first 5 minutes of infusion than hearts undergoing cold cardioplegic induction (16.9 versus 8.1 cc/loo gm, p < 0.05, Table II). The oxygen uptake during this period exceeded oxygen requirements by 12 cc, compared with only 6 cc in hearts arrested by cold cardioplegia (p < 0.05, Fig. 3). These calculations are based upon our determination that normothermically arrested hearts need 1.0 cc/100 gm/min and cold arrested hearts require 0.3 cc/loo gm/min (see Pilot studies). Myocardial oxygen uptake was comparable in the two groups when the cardioplegic solution was delivered at 4° to 8° C and myocardial temperature was reduced below 15° C (minutes 6-10, Table II). The oxygen uptake exceeded basal requirements by a total of 28 cc/ 100 gm when values during the last 5 minutes of induction of arrest and each replenishment were added. Consequently, total excess oxygen uptake (over demands) during all cardioplegic infusions was 40 cc/loo gm in hearts undergoing warm cardioplegic induction and 34 cc/loo gm in hearts undergoing cold cardioplegic induction (p < 0.05).

Volume 84

Normothermic induction of blood cardioplegia

Number 5 November. 1982

Blood reperfusion

Cardioplegia Beating empty

End

I

I

Beating working

I

Cold

Warm

2.1 ± 0.4 1.2 ± 0.3

2.1 ± 0.3 1.9 ± 0.3

2.4 ± 0.2 1.5 ± 0.2

1.8 ± 0.3 1.3 ± 0.2

15.9 ± 1.7 13.8 ± 4.1

13.1 ± 1.6 9.7 ± 1.9

14.1 ± 1.3 9.2 ± 2.1

11.7 ± 1.7 13.5 ± 2.0

10.6 ± 3.9 12.1 ± 1.5

0.14 ± 0.02 0.26 ± 0.10

0.12 ± 0.02 0.16 ± 0.02

0.12 ± 0.02 0.19 ± 0.03

0.19±0.02 0.18 ± 0.02 80.0 ± 0.6t

0.18 ± 0.06 0.22 ± 0.05 81.1 ± 0.5t

Cold

Warm

2.1 ± 0.4 1.3 ± 0.2

2.2 ± 0.3 1.6 ± 0.2

16.5 ± 1.0 12.7 ± 1.7 0.17 ± 0.02 0.23 ± 0.10

Warm

67 1

HIGH-ENERGY PHOSPHATE METABOLISM. Forty-five minutes of normothermic ischemia depleted high-energy phosphate levels (ATP and CP) transmurally, with the greatest degree of depletion in subendocardial muscle (Fig. 5, Table III). Subendocardial muscle ATP fell to 20% (0.8 ± 0.1 JLM/gm wet weight) and subendocardial CP fell to 5% (0.4 ± 0.1 JLMI gm wet weight) of control levels. G6P rose 1,500% (1.95 ± 0.25 JLM/gm), indicating a shift from aerobic to anaerobic metabolism. Blood cardioplegic induction produced a prompt repletion of CP to levels significantly greater than control (Table III, p < 0.05). The levels achieved were comparable in hearts undergoing warm or cold cardioplegic induction (9.7 versus 11.5 JLM/gm, p> 0.05). In contrast, cardioplegic induction resulted in only a mild increase in subendocardial ATP, which remained significantly below preischemic levels in all hearts (Table III). There were, however, striking differences in G6P metabolism between groups during cardioplegic induction. With cold induction, G6P remained elevated (1.95 versus 1.87 JLMI gm wet weight), whereas G6P fell significantly (1.95 to 0.97 JLM/gm wet weight) in hearts undergoing warm induction of arrest (Table III, p < 0.05). The 2 added hours of multidose cold blood cardioplegia were associated with (1) maintenance of the high CP levels attained during cardioplegic induction, (2) no further change in subendocardial ATP, and (3) lowering of G6P to normal in all dogs undergoing either warm or cold induction of cardioplegia. Aortic unclamping produced no further changes of subendocardial or subepicardial levels of ATP, CP, or G6P (Table III). VENTRICULAR PERFORMANCE. Right heart bypass

Cold

function curves inscribed before ischemia were comparable in all dogs from each experimental group. As described previously, hearts undergoing 45 minutes of normothermic ischemia (no cardioplegia) showed a 67% depression of postischemic performance 30 minutes after aortic unclamping (Fig. 4, P < 0.05). In contrast, all hearts undergoing 2 additional hours of aortic clamping with multidose blood cardioplegia had significantly better performance than those undergoing ischemic arrest alone (Fig. 4, P < 0.05). With cold cardioplegic induction, postischemic performance recovered to 63% ± 5% of control stroke work. Stroke work index at each level of left atrial pressure was significantly lower than control (p < 0.05). The best recovery of left ventricular performance occurred in hearts subjected to the brief (5 minutes) induction of cardioplegia at 37° C followed by multidose infusions of a cold blood cardioplegic solution. These hearts recovered 85% ± 5% of stroke work index; this degree of recovery was significantly better than that of the other two groups and did not differ significantly from preischemic levels (Fig. 4). Discussion This study shows that a severely energy-depleted heart can undergo a 2 hour period of aortic clamping with nearly complete recovery if treated with multidose cold blood cardioplegia where the cardioplegia is induced normothermically during the first 5 minutes of infusion. The data provide further confirmation of the capacity of blood cardioplegia to prevent and reverse myocardial damage by enhancing oxidative metabolism during cardioplegic infusions. 9. 10, 12, 13. 20 The experimental model of a heart severely depleted of myocardial ATP and CP before cardioplegic protection pro-

The Journal of

672 Rosenkranz et al.

Thoracic and Cardiovascular Surgery

**

15

5

Beating Cardioplegia

10

Endocardium

4

Epicardium

* ATP 3 (fLM/g m)

5

NET O2 (cc/100gm)

2

o

-5

o -10

*

**

* P< 0.05

P<0.05 vs 45min Ischemia P< 0.05 vs Cold Induction

Fig. 3. Oxygen uptake in excess of basal requirements after ischemia. Net O2 refers to oxygen consumed minus oxygen requirements for metabolic state. Note: (I) Oxygen consumption below requirements in hearts allowed to beat on reperfusion, (2) oxygen consumption in excess of requirement during cardioplegic reperfusion, and (3) highest oxygen uptake with warm induction. 1.5 Warm Induction 15 / / 4°C Blood / Cardioplegia / (2 hr )

1.0 SWI (gm-m/kg)

I

0.5

~

I

1../1

"

("'~Ol?

,'! r . . r

D.!

5

*

..

*

Induction _.~ ._.~'

·--{-Normothermic Ischemia (45 min)

10 15 20 LAP (rnrnl-lql

25

P<0.05 vs Control

Fig. 4. Left ventricular performance 30 minutes after blood reperfusion. Note: (I) normal postischemic performance with warm (37 C) cardioplegic induction, (2) moderate depression with cold (4 C) cardioplegic induction, and (3) severe depression after normothermic ischemia with unmodified reperfusion. SWI, Stroke work index. LAP, Left atrial pressure. 0

0

vides useful data. It demonstrates a potential limitation for recovery with cold blood cardioplegia (stroke work index recovered only 63% with 2 hours of aortic clamping versus complete recovery with 4 hours of aortic clamping in hearts without preexisting ischemia"); it

Control

Post ischemia

vs Control

Fig. 5. Myocardial adenosine triphosphate (A TP). Note: (I) Equal ATP in subepicardial and subendocardial muscle before ischemia (control); (2) significantly greater degree of ATP depletion in the subendocardial muscle postischemia.

casts doubt on the value of postischemic ATP levels in predicting eventual functional recovery (85% functional recovery with 80% persistent ATP depletion); it suggests the need for more detailed inquiry into the metabolic correlates of enhanced postischemic oxygen metabolism (i.e., cellular repair and energy replenishment). Most previous studies of cardioplegia have been carried out in normal hearts undergoing a defined interval of aortic clamping. It has been seen clinically': 2 and shown experimentally":" that tolerance to aortic clamping is reduced in hearts which resemble more closely conditions of chronic heart disease (i.e., advanced valvular heart disease or hypertrophy). Acute ischemic events such as cardiogenic shock or cardiac arrest before elective cardioplegia have been associated also with poor clinical results following technically successful operations. 8 Recent studies suggest that the metabolic basis for decreased tolerance to aortic clamping is depletion in the energy reserves of subendocardial muscle.v"- 22 We selected the model of preceding a long interval of aortic clamping (2 hours) with 45 minutes of ischemia because we have shown previously that this injury (I) depletes the heart of more than 40% of transmural ATP, (2) interferes with postischemic oxidative metabolism, (3) produces severe depression of postischemic myocardial performance at a time when bypass would be discontinued, and (4) represents a logical extension of our previous study, which showed that a warm interval of cardioplegia enhanced myocardial recovery after ischemic injury .11. 12 The normal pattern of metabolic recovery during reperfusion following a period of reversible ischemia

Volume 84

Normothermic induction of blood cardioplegia

Number 5 November, 1982

67 3

7 6 c

E

5

<,

@, 4

Cold Induction

Warm Induction

o

Q ~

u

3

(\./

o 2

• 1 min 10 iE--3rC----il)'foEIE--4°C ~

1 iE--4°C

min liE

Fig. 6. Myocardial oxygen uptake during 10 minute cardioplegic induction. Cross-hatched areas represent the basal requirement of normal myocardium at the indicated temperature. Note: (I) Warm induction results in greater oxygen consumption than cold induction; (2) basal oxygen requirements are reached by the fifth minute of both warm and cold induction.

is characterized by a reactive coronary hyperemia, whereby the heart consumes more oxygen than required to meet basal demands. 12, 23 The precise metabolic use of the increased oxygen uptake is uncertain, but we presume that the extra oxygen facilitates recovery by its being used to replenish depleted energy stores and repair cellular processes damaged by ischemia. Conversely, ischemically damaged hearts which do not recover functionally take up less oxygen than required to meet metabolic demands when allowed to resume beating immediately. 12 In this study, hearts subjected to 45 minutes of ischemia with unmodified reperfusion (no cardioplegia) consumed 18% less oxygen (6.6 cc/l00 gm) than needed to meet the basal demands of the beating empty state during the first phase of reperfusion (Fig. 3). Our use of warm blood cardioplegia during reperfusion evolved from our studies showing that reperfusion with a warm blood cardioplegic solution could avoid and reverse reperfusion in the following ways: (I) by lowering myocardial energy demands so that the limited postischemic oxygen-utilizing capacity could be channelled toward repair": (2) by reducing the ionic calcium available to precipitate in mitochondria':'; (3) by reversal of cellular acidosis by buffering to allow more normal resumption of cellular processes'"; and (4) by simultaneously increasing the rate of metabolic recovery by normothermia." We reasoned that a brief period of warm cardioplegic induction would have similar salutory effects on energy-depleted hearts that

must undergo a subsequent period of aortic clamping, whereas cold cardioplegic induction might retard metabolic rate and slow repair because of the 50% reduction in metabolism that occurs with each 10° C fall in temperature. In reality, blood cardioplegic induction is the first phase of blood reperfusion. We used a 5 minute period of cardioplegic induction because this seemed practical clinically. This empirical decision was accompanied by the experimental observation that postischemic myocardial uptake approached basal requirements for temperature (see Pilot studies) at about 5 minutes of reperfusion in both cold and warm hearts (Fig. 6). Five additional minutes of cold cardioplegia were used to provide adequate hypothermic myocardial protection during the subsequent period of aortic clamping. All hearts receiving only cold cardioplegia underwent a similar 10 minute interval of cardioplegic induction to allow comparison. Hearts receiving the warm cardioplegia consumed more total oxygen as well as more oxygen in excess of the basal requirements of arrest than those receiving cold cardioplegia during the first 5 minutes of cardioplegic infusion (Fig. 3). The 50% fall in subendocardial G6P provided added confirmation of better aerobic metabolism during warm cardioplegic induction. The enhanced oxygen uptake during warm and cold cardioplegic induction in both groups (above demands) was parallelled by recovery of subendocardial muscle CP to levels significantly greater than control (23% to 46%, P < 0.05). There was, however, no significant ATP

The Journal of

674 Rosenkranz et al.

repletion in either group, and postinduction CP and ATP were comparable despite substantially greater oxygen uptake in hearts receiving the brief warm infusion. The metabolic benefits of multidose cold blood cardioplegia were especially evident during the 2 subsequent hours of aortic clamping in these energy-depleted hearts. Oxygen uptake exceeded basal requirements by 28 ccl I()() gm during cardioplegic replenishments (each 20 minutes), G6P returned to control levels, a fact that suggests aerobic metabolic activity, CP remained above control, and ATP did not fall either during aortic clamping or following 30 minutes of reperfusion. These data support the basic hypothesis upon which this study was designed-energy-depleted hearts subjected to warm cardioplegic induction before 2 additional hours of aortic clamping would take up more oxygen and show better recovery than those undergoing cold cardioplegic induction or unmodified reperfusion (45 minutes of ischemia without cardioplegia). However, we were incorrect in our anticipation that the extra oxygen consumed would be used to replenish energy stores above that possible with cold cardioplegic induction; ATP and CP levels were comparable in the two groups. The greater oxygen uptake during warm induction was likely channelled toward better cellular repair than was possible when the rate of repair was slowed by hypothermia (cold induction). The near normal functional recovery with warm induction of multidose cardioplegia may have occurred because the extra oxygen was used to generate ATP, which was utilized immediately. Most likely, repair continued during the 2 hours of subsequent aortic clamping where all hearts receiving multidose cardioplegic infusions took up 28 cel 100 gm more oxygen than needed for the demands of arrest. This may account for the superior recovery of hearts subjected to multidose cardioplegia compared to those subjected to unmodified reperfusion after 45 minutes of normothermic ischemia, where increased reactive hyperemic oxygen uptake did not occur. The failure of ATP levels to recover probably reflects a depletion of the adenine nucleotide pool that recovers slowly both by de novo and salvage pathways.t""? Adequate phosphorylating capacity of the mitochondria was present, since there was a supranormal production of CP.28 We are planning to study the role of adenine nucleotide replenishment to see if postischemic ATP repletion is possible, since return to preischemic levels may be desirable physiologically. Serial measurements of myocardial high-energy phosphate levels (ATP and CP) confirmed the degree of

Thoracic and Cardiovascular Surgery

metabolic derangement produced by 45 minutes of normothermic ischemia. Endocardial ATP fell below 1.0 ~M/gm wet weight, a level previously reported to be incompatible with morphologic, metabolic, and functional recovery. 29 This hypothesis was based upon ultrastructural and functional analyses made after unmodified reperfusion of ischemic hearts. We believe that our modification of the conditions and composition of reperfusion prevented reperfusion damage and allowed the excellent functional recovery seen. Our data do not, however, provide a ready explanation of why hearts recovered so well despite reduction of ATP to levels previously thought to be associated with irreversible damage. It has long been recognized that tissue ATP levels are determined by the rates of both production and utilization of ATP. 6 We did not determine either of these rates but suspect that their measurement will be important metabolic indices in the future. We believe that the normal resting ATP level (control) reflects a potential temporary ATP reserve. Our data would suggest that the heart can function quite well with depletion of this reserve provided that the metabolic machinery responsible for ATP regeneration is intact (the mitochondria). Although depletion of ATP reserves was not detrimental immediately after bypass (i.e., during inscription of function curves), it may limit the heart's ability to tolerate later perioperative stress (hypertension, pain, tachycardia). Conceivably, this ATP depletion may result in the low output state seen 4 to 6 hours after operation in some patients. All energy-depleted hearts receiving multidose infusions of a cold blood cardioplegic solution for 2 additional hours of aortic clamping showed better functional recovery than after 45 minutes of normothermic ischemia alone. In this study, our standard multidose cold blood cardioplegia resulted in only 63% recovery in hearts previously ischemic, whereas complete recovery is possible after 4 hours of cold blood cardioplegia when previous ischemia is not present." This "limitation" of multidose cold blood cardioplegia in energy-depleted hearts was overcome by preceding it with a 5 minute period of warm blood cardioplegia. In clinical practice, we'" use this method of cardioplegic induction in high-risk patients by varying the temperature of the heat exchanger used for cardioplegic delivery (Fig. 2). We find that prompt arrest can be induced with a warm cardioplegic solution by keeping the potassium concentration at 25 mEq/L during the 5 minute infusion and lowering the concentration to 10 mEq/L during the subsequent 5 minutes of cold cardioplegic infusion. This relatively long period of cardioplegic induction is not detrimental: (1) The heart is

Volume 84 Number 5 November, 1982

perfused continually with oxygenated blood in a state of decreased metabolic demands; (2) the longer period of cold cardioplegia allows for better myocardial cooling, especially beyond coronary stenoses-"; (3) proximal and distal grafts or aortic valve removal can be carried out while the cardioplegic solution is being given; (4) metabolic recovery is occurring during this period of induction, as shown by this study. Hopefully, application of this technique will allow safer, prolonged aortic clamping in the patients who are still considered to be in the high-risk category. We wish to acknowledge Jerry D. Leaf, M.S., Edward Dolendo, M.S., and Ms. Nanci Birman for their technical assistance in performing this study and Ms. Judith Miller for assistance in preparing the manuscript. REFERENCES Kirklin JW: A letter to Helen. J THORAC CARDIOVASC SURG 78:643-654, 1979 2 Wideman FE, Blackstone EH, Kirklin JW, Karp R, Kouchoukos NT, Pacifico AD: Hospital mortality of rereplacement of the aortic valve. J THORAC CARDIOVASC SURG 82:692-698, 1981 3 Iyengar SRK, Ramchand S, Charrette EJP, Iyengar CKS, Lynn RB: Anoxic cardiac arrest. An experimental and clinical study of its effects. Part I. J THORAC CARDIOVASC SURG 66:722-730, 1973 4 Sink 10, Pellom GL, Currie WD, Hill RC, Olsen CO, Jones RN, Wechsler AS: Response of hypertrophied myocardium to ischemia. J THORAC CARDIOVASC SURG 81:865-872, 1981 5 Jones RN, Currie WD, Olsen CO, Peyton RB, Van Trigt P, Wechsler AS: Recovery of metabolic function in hypertrophied canine hearts following global ischemia. Circulation 62:Suppl 3:30, 1980 6 Feinstein MB: Effects of experimental congestive heart failure, ouabain, and asphyxia on the high-energy phosphate and creatine content of the guinea pig heart. Circ Res 10:333-346, 1962 7 Jones RN, Peyton RB, Sabina RC, Swain JL, Holmes EW, Spray TL, Van Trigt P, Wechsler AS: Transmural gradient in high-energy phosphate content in patients with coronary artery disease. Ann Thorac Surg 32:546-553, 1981 8 Dawson JT, Hall RJ, Hallman GL, Cooley DA: Mortality in patients undergoing coronary artery bypass surgery after myocardial infarction. Am J Cardiol 33:483-486, 1974 9 Follette DM, Mulder DG, Maloney JV Jr, Buckberg GD: Advantages of blood cardioplegia over continuous coronary perfusion or intermittent ischemia. J THORAC CARDIOVASC SURG 76:604-619, 1978 10 Cunningham IN Jr, Adams PX, Knopp EA, Baumann FG, Snively SL, Gross Rl, Nathan 1M, Spencer FC: Preservation of ATP, ultrastructure, and ventricular per-

Normothermic induction of blood cardioplegia

II

12

13

14

15

16

17

18 19

20

21

22

23

24

67 5

formance after aortic cross-clamping and reperfusion. J THORAC CARDIOVASC SURG 78:708-720, 1979 Lazar HL, Buckberg GD, Manganaro A, Becker H, Mulder DG, Maloney JV Jr: Limitation imposed by hypothermia during recovery from ischemia. Surg Forum 31: 312-315, 1980 Lazar HL, Buckberg GD, Manganaro AJ, Foglia RP, Becker H, Mulder DG, Maloney JV Jr: Reversal of ischemic damage with secondary blood cardioplegia. J THORAC CARDIOVASC SURG 78:688-697, 1979 Follette DM, Fey K, Buckberg GD, Helly 11 Jr, Steed DL, Foglia RP, Maloney JV Jr: Reducing postischemic damage by temporary modification of reperfusate calcium, potassium, pH, and osmolarity. J THORAC CARDIOVASC SURG 82:221-238, 1981 Becker H, Vinten-Johansen J, Buckberg GD, Robertson JM, Leaf JD, Lazar HL, Manganaro AJ: Myocardial damage caused by keeping pH 7.40 during systemic deep hypothermia. J THORAC CARDIOVASC SURG 82:810-820, 1981 Behar MG, Severinghaus JW: Calibration and a correction of blood O 2 content measured by P0 2 after CO saturation. J Appl Physiol 29:413, 1970 Hottenrott C, Maloney JV Jr, Buckberg GD: Studies of the effects of ventricular fibrillation on the adequacy of regional myocardial blood flow. I. Electrical versus spontaneous fibrillation. J THORAC CARDIOVASC SURG 68: 615-625, 1974 Lowry OH, Passonneau JV: A Flexible System of Enzymatic Analysis, New York, 1972, Academic Press, Inc., pp 151-156 Duncan DB: t-Tests and intervals for comparisons suggested by the data. Biometrics 31:339-359, 1975 Buckberg GD, Dyson DW, Emerson RC: Techniques for administering clinical cardioplegia, Blood Cardioplegia, RM Engelman, S Levitsky, eds., Mt. Kisco, N. Y., 1982, Futura Publishing Co., pp 305-316 Singh AK, Farrugia R, Teplitz C, Karlson KE: Electrolyte versus blood cardioplegia. Randomized clinical and myocardial ultrastructural study. Ann Thorac Surg 33:218-227, 1981 Robertson JM, Vinten-Johansen J, Buckberg GD, Follette DM, Maloney JV Jr: Prolonged safe aortic clamping (4 hours) with cold glutamate enriched blood cardioplegia. Circulation 64: Suppl 4: 147, 1981 Peyton RB, Pellom GL, Currie WD, Jones RN, Olsen WD, Van Trigt P, Sink JD, Wechsler AS: Preischemic ATP enhancement to improve tolerance to ischemia in hypertrophied myocardium. Surg Forum 31: 315-317, 1980 Coffman JD, Gregg DE: Oxygen metabolism and oxygen debt repayment after myocardial ischemia. Am J Physiol 201:881-887, 1961 Zimmer HG, Trendelenburg C, Kammermeier H, Gerlach E: De novo synthesis of myocardial adenine nucleotides in the rat. Acceleration during recovery from oxygen deficiency. Circ Res 32:635-642, 1973

The Journal of

676 Rosenkranz et ai.

25 Pasque MK, Spray TL, Pellom GL, Van Trigt P, Peyton RB, Currie WD, Wechsler AS: Ribose-enhanced myocardial recovery following ischemia in the isolated working rat heart. J THORAC CARDIOVASC SURG 83:390-398, 1982 26 Foker JE, Einzig S, Wang T, Anderson RW: Adenosine metabolism and myocardial preservation. Consequences of adenosine catabolism on myocardial high-energy compounds and tissue blood flow. J THoRAc CARDIOVASC SURG 80:506-516, 1980 27 Reibel DK, Rovetto MJ: Myocardial adenosine salvage rates and restoration of ATP content following ischemia. Am J Physiol 237:H247-H252, 1979 28 Reibel DK, Rovetto MJ: Myocardial ATP synthesis and mechanical function following oxygen deficiency. Am J PhysioI234:H620-H624, 1978 29 Preusse CJ, Bretschneider HJ, Gebbard MM: Comparison of cardioplegic methods of Kirklin, Bretschneider, and St. Thomas' Hospital by means of biochemical and functional analyses during the postischemic aortic recovery period, Myocardial Protection for Cardiovascular Surgery, WH Isselhard, ed., Cologne, 1980, Pharmazeutische Verlagsgeselschaft, pp 160-169 30 Robertson JM, Buckberg GD, Vinten-Johansen J: Cardioplegic delivery beyond stenoses. The superiority of blood cardioplegia over asanguineous cardioplegia. Surg Forum 32:286-287, 1981

Discussion DR. NIELS M. BLEESE Hamburg. Federal Republic of Germany

Patients with preoperatively energy-depleted hearts are comparable to patients requiring cross-clamp periods exceeding 2 hours or more: Neither group of patients can tolerate any additional ischemia. To minimize ischemia, the Hamburg method of myocardial protection enables the maintenance of oxidative metabolism during hypothermic cardiac arrest by intermittent or continuous coronary perfusion with the cardioplegic solution. Our asanguineous solution contains 2.5 vol % of physically dissolved oxygen and 10 mM glucose. Any perfusioninduced interstitial edema is prevented by iso-onconicity due to hydroxyethyl starch and pressure limitation. Cardioplegia is brought about by procaine and low sodium; K+ is normal! Since 1975 more than 2,000 patients have been operated upon with this method. In 82 complicated cases the crossclamp period exceeded 150 minutes. We are presenting this special subgroup because we believe that only extraordinary stress situations for the myocardium can prove the reliability of the method of myocardial protection. The average cross-clamp time in this subgroup was 169 minutes, with a maximum of 236 minutes. The mean cardioplegic coronary perfusion time was 48 minutes, that is, about 30% of the whole cross-clamp time. In a few cases the cardioplegic perfusion has been extended up to 2 hours. There

Thoracic and Cardiovascular Surgery

were seven deaths, none of which was related to insufficient myocardial protection. We believe that, in borderline cases, minimizing ischemia by hypothermia and maintaining the oxidative metabolism by intermittent or continuous cardioplegic coronary perfusion is absolutely necessary and is beneficial. DR. J. CUNNINGHAM, JR. New York, N. Y.

At New York University, we too feel that preischemic enhancement is an important adjunct to other methods of myocardial preservation. Recently, we have begun to employ this methodology at our institution, in particular for patients with end-stage cardiac disease requiring prolonged crossclamp intervals and more particularly in patients with advanced valvular disease. We have recently obtained intraoperative myocardial biopsy specimens from six patients and determined endocardial and epicardial ATP levels. Following 5 minutes of normothermic preischemic enhancement with our standard blood cardioplegia solution, we have noted approximately a 30% to 35% increase in endocardial ATP, with further increases up to 40% to 45% following the subsequent injection of the cold arrest solution. Levels of ATP were still elevated just prior to unclamping and even 30 minutes after the last perfusion. In addition, they remained above normal 30 minutes after unclamping and reperfusion. Similar trends, although somewhat less significant, were seen in epicardial ATP levels. Therefore, we strongly advocate the use of this methodology in such seriously ill and presumably energydepleted patients. Thus far our clinical studies have been conducted in patients with right ventricular hypertrophy and mitral valve disease. This has been simply related to the ease with which we have been able to obtain right ventricular biopsy specimens and additionally because administration of warm cardioplegic solution into the aortic root allows for simultaneous perfusion of both coronaries. We have not yet used this technique for aortic valve replacement. I have two questions for the authors. First, have you been able to obtain clinically ATP biopsy tissue which would substantiate our observations? Second, can you describe your technical approach for warm and then cold perfusion during aortic valve replacement in the open aortic root? How do you administer your cardioplegic solution? Before the aortotomy is made, simultaneously into both coronaries, or separately into each coronary? DR. SIDNEY LEVITSKY Chicago. 1/1.

Dr. Buckberg's group has very carefully produced an experimental model of severe ischemia by inducing normothermic global ischemia for 45 minutes, which reduced the ATP levels to below I I-tM/gm (normal 4 to 5 I-tM/gm). We know from work previously published over a decade ago by Isselhard and ourselves that over a 50% depletion of ATP values results in a marked decrease in myocardial contractility. We also know that it takes several hours to days to rebuild

Volume 84

677

Number5 November, 1982

Normothermic induction of blood cardioplegia

ATP stores that have been depleted. Therefore, I find it difficult to comprehend how the short period of warm cardioplegic perfusion utilized in these experiments can rebuild the ATP stores. Since biochemical data were not presented, my question to the authors is this: What were the differences in ATP values after their method of cardioplegia? Did they record an enhancement of ATP values from the depressed levels seen after ischemia to normal levels that would be compatible with the elevated contractility measurements that were presented? If not, how do the authors explain the incompatibility between functional and biochemical data?

the aortic root. With aortic insufficiency we open the aorta and rapidly cannulate the coronary ostia and proceed with warm induction of cardioplegia. Dr. Bleese, we find your concept of continuous infusion quite interesting. However, we found it much more practical to use intermittent doses of cardioplegic solution, especially in patients with coronary disease, in whom retraction of the heart makes the aortic valve incompetent. In the past, Dr. Cunningham presented data before this Association showing that safe aortic clamping is possible for as long as 4 hours using intermittent cold blood cardioplegia. In response to Dr. Levitsky, we measured both ATP and CP serially. ATP was not substantially restored but CP did rise to supranormallevels. Our data suggest therefore that normal myocardial ATP is not essential for functional recovery. We believe ATP was depleted during ischemia as was the adenine nucleotide pool which provides the requisite precursors for its resynthesis. We suspect that the ATP-producing capacity of these cells was normal but that ATP was utilized at an accelerated rate to repair ischemic damage. This resulted in the low steady state level of ATP. The restoration of CP immediately after cardioplegic infusion provides evidence that the mitochondria resume function after ischemia, since ATP generation is required for this to occur. This capacity to produce energy likely accounts for the restoration of left ventricular function that resulted from warm induction of cardioplegia.

DR. R 0 SEN K RAN Z (Closing) I would like to thank all of the discussers for their thoughtful comments. In addition, I would like to add one more note. The impetus for this study was really the observation of Dr. Rodewald that one could improve the postsurgical outcome of critically ill patients by allowing the heart a brief period of normothermic bypass prior to cardioplegic arrest. Dr. Cunningham, we greatly appreciate your sharing your results with us. We have not made similar measurements of ATP or CP in our patients, although we would like to do so. We have used this warm induction technique in patients and have also had very encouraging results . Your second question concerns its use in patients with aortic valve disease. With aortic stenosis we induce warm cardioplegia before opening