Studies on prolonged acute regional ischemia

J THORAC CARDIOVASC SURG 1989;98:567-79

Studies on prolonged acute regional ischemia V Metabolic support of remote myocardium during left ventricular power failure This study tests the hypothesis tbat metabolic support of remote "nonischemic" myocardium during acute infarction will reverse the trend toward cardiogenic shock. Thirty-seven dogs underwent ligation of the left anterior descending coronary artery and 50 % stenosis of the circumflex coronary artery. Irreversible ventricular fibrillation developed in 11 of them. The 26 survivors were observed for up to 6 hours; global

and regional left ventricular function (cardiac index, stroke work index, ultrasonic crystals) and regional blood flow (radioactive microspheres) were measured. After 2 hours, eight dogs received an intravenous infusion of glutamate/aspartate, g1ucose-insutin-potassium, coenzyme QUI> and 2-mercapto-propionylglycine for 4 hours. Five dogs received the mannitol infusion to raise serum osmolarity 30 mOsm. Four additional dogs received the intravenous substrate infusions over 4 hours without undergoing ischemia. The substrate infusion for 4 hours caused no cbange in regional or global cardiac function in the four control dogs. Three of nine untreated dogs died of cardiogenic shock, and progressive left ventricular power failure occurred in the six others (40 % decrease in cardiac index, 50 % decrease in stroke work index, p < 0.05) because of persistent dyskinesia in the left anterior descending region (-40 % of systolic shortening, p < 0.05) and hypocontractility in the circumflex region (48 % of control systolic shortening, p < 0.05~ despite normal transmural blood flow in the posterior left ventricular wall (76 m1/100 gm/min). In contrast, in treated dogs, hypercontractility recovered in the circumflex segment (138 % of systolic shortening) and stroke work index rose to control levels (91 %) without a cbange in regional blood flow. Mannitol infusion did not improve hemodynamics or avoid the development of progressive left ventricular power failure. We conclude tbat cardiogenic shock after myocardial infarction is due, in large part, to impaired ability of "nonischemic" myocardium to maintain hypercontractility. This limitation can be prevented by metabolic support of viable muscle, and the data imply tbat intravenous substrate infusions may be helpful before definitive treatment (i,e., coronary artery bypass grafting) is undertaken.

Friedheim Beyersdorf, MD (by invitation), Christophe Acar, MD (by invitation), Gerald D. Buckberg, MD, Marshall T. Partington, MD (by invitation), Fumiyuki Okamoto, MD (by invitation), Bradley S. Allen, MD (by invitation), Helen H. Young, PhD (by invitation), and Helen I. Bugyi, PhD (by invitation), Los Angeles, Calif.

Left ventricular power failure is the principal cause of in-hospital mortality after acute coronary occlusion.!' Acutely ischemic cardiac muscle stops contracting From the Division of Cardiothoracic Surgery, University of California at Los Angeles Medical Center, Los Angeles, Calif. Supported in part by National Institutes of Health Grant HLl6292, the Josephine H. Owens Fund. and the Selden Ring Fund. Read at the Sixty-eighth Annual Meeting of The American Association for Thoracic Surgery, Los Angeles, Calif., April 18-20, 1988. Address for reprints: Gerald D. Buckberg, MD. UCLA Medical Center. Department of Surgery. Los Angeles, CA 90024-1741.

12/1/12167

immediately and never contributes to cardiac output during the early periinfarction period. Cardiogenic shock develops more commonly in patients with multivessel disease and occurs after an otherwise nonfatal myocardial infarction (i.e., <30% loss of muscle) when remote myocardium fails to develop and sustain the compensatory hypercontractility needed to maintain cardiac ouptut." Conventional treatment of left ventricular power failure includes volume expansion and pharmacologic (inotropic drugs) and mechanical (intraaortic balloon counterpulsation) circulatory support to augment the contractile state of remote muscle and improve its balance between oxygen supply and demand.' 567

The Journal of Thoracic and Cardiovascular Surgery

Beyersdorf et al.

56 8

The results will show that an intravenous nutritive solution (1) reduces mortality, (2) provides hemodynamic and biochemical benefits that avoid or reverse the tendency toward development of progressive cardiogenic shock, (3) do not exert their salutary effects by direct inotropic or osmotic action, and (4) may become a valuable adjunct to the management of patients with cardiac disease while they undergo diagnosis and subsequent definitive treatment of acute coronary occlusion.

v

Methods

Sonic Crystals Fig. 1. Experimental model (see text for description) Ischemia is not a prerequisite for impaired remote muscle function after acute coronary occlusion, inasmuch as inadequate compensatory contractility of remote muscle may occur despite normal or increased blood flow.2 a , 4-6 Specific metabolic changes have been detected in remote myocardium responsible for maintaining the increased workload.?'!" and these biochemical alterations imply that a limited substrate supply may impair the capacity of functional muscle to generate enough energy to support the hypercontractility needed to maintain systemic blood flow, These remote metabolic changes include (1) preference for glucose rather than fatty acids as a metabolic fuel and for energy production," II (2) loss of Krebs cycle intermediates responsible for aerobic and anaerobic production of high-energy phosphates,' and (3) depletion of substances (i.e., coenzyme QIO, sulfhydryl enzymes) that may be needed for oxidative adenosine triphosphate (ATP) production and utilization. 10 We 12* have shown previously the benefits of replenishing each of these metabolic substances during controlled regional reperfusion after segmental ischemia. The current study tests the concept that intravenous metabolic support of remote myocardium may improve its capacity to support the circulation in simulated multivessel disease during acute coronary occlusion.

*Beyersdorf F. Avoiding reperfusion injuryafter limbrevascularization by controlof the initial reperfusion: a newsurgicaltechnique. (Unpublished data).

Experimental preparation. Forty-one mongrel dogs (22 to 30 kg) were anesthetized with a I: I mixture of sodium pentobarbital and sodium thiamylal (30 mgjkg intravenously) and their lungs ventilated by positive-pressure intratracheal ventilation with 100% oxygen, All animals received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the National Academy of Sciences and published by the National Institutes of Health (N IH Publication No. 80-23. revised 1978), The chest was opened by a median sternotomy incision and the pericardium was incised and cradled. A lateral thoracotomy in the fifth intercostal space was done to facilitate isolation of the circumflex coronary artery, Polyethylene catheters were placed into the aortic arch (via the right internal mammary artery), the left atrium, and into the left ventricle for measuring pressure and for timing the cardiac cycle to synchronize with segmental crystal recordings, The hemodynamic and functional data were recorded on a Honeywell 15088 Visicorder oscillograph (Honeywell Inc" Minneapolis. Minn.). Catheters were placed also into the left femoral artery. for withdrawing samples for blood gas analyses and for collecting reference samples for microsphere blood flow analyses, and into the left femoral vein for giving infusions, The left anterior descending coronary artery (LAD) was dissected distal to its first diagonal branch for subsequent occlusion to produce an area at risk of approximately 30(!r (Fig, I), The left circumflex artery was dissected approximately 3 em distal to its origin for subsequent placement of an electromagnetic flow probe transducer to produce a stenosis (see Experimental groups). Ultrasonic dimension crystals (2.2 to 2.5 mm in diameter) were inserted into the endocardium parallel to the direction of the meridional fibers in the center of the eventual ischemic zone and in the posterior left ventricular wall in the distribution of the circumflex coronary artery, Instantaneous segmental length was determined by a sonomicrometer (Triton Technology, Inc" San Diego, Calif), Preparations for extracorporeal circulation were made by cannulating the right femoral artery with a 16F USCI cannula (Bard Implants Division, C. R. Bard, Inc" Billerica, Mass.) and the right atrium with a two-stage Sarns cannula (Sarns Inc" Ann Arbor. Mich.). The cannulas were connected to a pump-oxygenator circuit primed with 1500 ml blood and 400 ml hetastarch (Hespan). A Shiley S-IOOA bubble oxygenator (Shiley lnc., Irvine, Calif.) was prepared for use during each experiment. The extracorporeal circuit was used only temporarily (i.e., 2 to 3 minutes) to facilitate defibrillation during ischemia and thereby increase the yield of completed experiments.

Volume 98 Number 4

Prolonged acute regional ischemia. V

October 1989

A lidocaine bolus (100 mg) was administered just before LAD ligation, and a lidocaine infusion (I mg/rnin) was delivered throughout ischemia. Ventricular fibrillation was converted electrically on temporary cardiopulmonary bypass. Blood gases were maintained at pH 7.3 to 7.4., carbon dioxide tension 30 to 40 mm Hg, and oxygen tension greater than 100 mm Hg. Measurements Regional contractility. Segmental systolic shortening (ss) in the myocardium supplied by the LAD and circumflex arteries was calculated as follows: % SS

EDL - ESL EDL

X 100

where EDL and ESL are the end-diastolic and end-systolic length, respectively. Results were expressed as percent systolic shortening (%SS) relative to control values to allow comparison between dogs and to avoid bias relative to differences in heart size and crystal placement. Three successive beats were analyzed and averaged. The onset of diastole was determined from the end-diastolic pressure point on the ventricular tracing. The diastolic time was the dicrotic notch on the aortic tracing. Global myocardial function. Cardiac output was measured with a 5F pediatric Swan-Ganz catheter (Baxter Edwards Divisions, Irvine, Calif.), which was inserted into the main pulmonary artey via the right femoral vein. Ice-cold saline 5 ml was injected for each measurement. Stroke work index (SWI) in gram-meters per kilogram was used to normalize for body weight: (MAP - LAP) X CO SWI = - - - - - - - ' - - - - - - X 1.36 Heart rate X Body weight where MAP is mean aortic pressure, LAP is left atrial pressure, and CO is cardiac output. Regional blood flow. Regional blood flows were measured by injecting 15 ± 3 J.Lm microspheres labeled with 5JCo;5Nb, "Sc, and II)Sn into the left atrium over 20 seconds while withdrawing a reference sample from the femoral arteries as described previously. The microsphere injections were made during the control period and after 2, 4, and 6 hours of ischemia. The heart was removed after the experiment and samples of endocardial, mid-myocardial, and epicardial muscle were obtained and placed into tared vials. Radioactive content of each nuclide was determined by gamma spectrometry and blood flow was calculated as described previously.!' Myocardial metabolism. After each experiment the heart was arrested by intracardiac potassium injection and transmural biopsy specimens were taken by a high-speed drill from the anterior and posterior walls divided into subepicardial and subendocardial segments, and quick-frozen in liquid nitrogen for later analysis of high-energy phosphates (tissue adenosine triphosphate, diphosphate, and monophosphate [ATP, ADP, and AMP], creatine phosphate, and glucose 6-phosphate) on a Farrand spectrofluorometer (Farrand Optical Company, lnc., Valhalla, N.Y.). Myocardial adenine nucleotides pyrimidine (NAD) was determined by the fluorometric enzymatic methods described by Williamson and Corkey." Drill biopsy specimens were taken also before cardioplegic arrest for analysis of tissue water content: (,7, Water =

Wet weight - Dry weight Wet weight

X 100

569

Table I. Intravenous substrate infusion Glutamate (gm/IOO ml) Aspartate (gm/ I00 ml) Glucose (gm/kg/IOO ml) Insulin (IU/kg/IOO ml) Potassium (mEq/kg/lOO ml) MPG (mEq/kg/lOO ml) CoQlO (mg/kg/IOO ml)

2.1 2.0 0.5 0.25 0.25 2.5 2.5

'vIPG. 2-'vIcrcapto-propionyl-glycinc. CoO",. coenzyme Q".

Mitochondrial function. For the oxidative phosphorylation study, mitochondria were isolated by a modified differential centrifugation method of Sordahl and Stewart. IS At the completion of the study, the heart was excised and immediately placed in ice-cold saline. A piece of tissue approximately 5 gm was removed and separated into epicardial and endocardial sections of equal thickness. Tissue was then placed into ice-cold isolation medium consisting of potassium chloride 180 mrnol/L, ethylenediamine tetracetic acid (EDTA) \0 mrnol/ L, 0.5% bovine serum albumin (Sigma fraction Y, fatty acid free), pH 7.4, with Tris base. The tissue was weighed, minced, and homogenized in 12 volumes of isolation medium containing \ mg nagarse per gram tissue and then centrifuged for 10 minutes at 2000 g. The supernatant was poured through ice-cold gauze and centrifuged at 10,000 g for 10 minutes. Mitochondria were washed twice in isolation medium containing no nagarse and finally suspended in the same isolation medium at a concentration of \5 to 20 mg/rnl, Mitochondrial protein was determined by the method of Lowry and associates." The rate of mitochondrial respiration was measured in a Gilson Oxygraph oximeter (Gilson Medical Electronics, Inc., Middleton, Wis.) and Clark electrode (Yellow Springs Instrument Company, Yellow Springs, Ohio) at 30° C. The mitochondria (1.0 mg) were added to a reaction chamber containing a solution of sucrose 250 mrnol/L, potassium phosphate 8 mmol/L, Tris HCI 10 rnmol/L, pH 7.4, glutamate 10 mmol/L. and malate 5 mmol/L. ADP 300 nmol was then added to initiate State 3 respiration after establishment of stable State 4 respiration. ADP stimulation was obtained twice for each mitochondrial sample and the data were averaged to calculate the ADP /0 ratio, State 3 respiration, State 4 respiration, and respiratory control index. ATP generating capacity or oxidative phosphorylation rate was calculated by multiplying State 3 respiration by the ADP /0 ratio. Calcium content was determined from the mitochondria isolated with a medium containing no EDTA or EGTA * to eliminate the calcium loss." We used lanthanum chloride 0.5 mmol/L to inhibit calcium influx and efflux during mitochondrial isolation." These mitochondria were extracted by the method of Reynafarje and Lehninger" and calcium content determined by an atomic absorption spectrophotometer (AA-275D, Varian Techtron Ply. Ltd., Mulgrave, Australia). Experimental groups Control substrate infusion. no ischemia. In four dogs the LAD and circumflex arteries were isolated but were not occluded or stenosed. Each dog was observed for 2 hours without intervention and was then given an intravenous substrate infusion for 4 hours (Table I) to determine the *Ethyleneglycol-bis-({:i-aminoethylethcr)-N.N,N' ,-N' -tetracetic acid.

The Journal of

570

Thoracic and Cardiovascular Surgery

Beyersdorf et al.

140 120 100 % Control

SS

80 60 40 20

1 - - - - - - Rx-------l

2

3

4

5

6

Hours

Mean ± SE

Fig. 2. Segmental shortening (ss) in remote myocardium during LAD occlusion and circumflex stenosis. Note: (1) progressive hypocontractility of remote muscle during the first 2 hours of observation, (2) progressive decline in contractility in dogs receiving no treatment (no Rs, solid line), and (3) progressive improvement while intravenous substrate infusion was delivered (hatched line). Treatment (Rx) was administered over 4 hours. SE, standard error.

Table II. Serum osmolarity (mOsmjL) Mannitol Control l-hr infusion 2-hr infusion 3-hr infusion 4-hr infusion

310 ± 332 ± 335 ± 339 ± 343 ±

5

5*t 7*t 7*t 6*t

Substrate infusion

304 ± 316 ± 321 ± 316 ± 322 ±

4 3* 4*

6* 5*

Values arc mean ± standard error of the mean.

"n < 0.05 versus substrate infusion. tp

< 0.05 versus control.

hemodynamic consequences of the metabolic support solution without ischemia. LAD occlusion and circumflex stenosis. In 37 dogs, a 50% to 60% stenosis of the left circumflex coronary artery was produced by placing a 2.0 to 2.5 mm electromagnetic flowmeter around the proximal 3.0 to 3.5 mm circumflex coronary artery. The flow probe selected for use reduced peak reactive hyperemic flow to 70% to 100% above resting flow after a IO-second occlusion (the normal reactive hyperemic flow with a 3.0 to 3.5 em probe is 500% to 700% in our laboratory). The LAD was occluded after we had ensured that the circumflex stenosis did not reduce either the resting coronary blood flow or segmental contractility in the area supplied by the circumflex coronary artery. Ventricular fibrillation was treated with temporary institution of extracorporeal circulation, electric countershock (10 to 20 W jsec), and antiarrhythmic drugs. No treatment. In 14 dogs, the natural history of LAD occlusion and circumflex stenosis was observed for 6 hours. These dogs received an intravenous infusion of Ringer's lactate

solution at a rate of 100 mljhr throughout the observation period. Intravenous metabolic support. From a group of 16 dogs. an intravenous substrate infusion (Table I) was started in the 12 dogs that survived the first 2 hours of ischemia. The infusion was continued at a rate of 100 mljhr for 4 hours. The delay in starting the intravenous treatment was used to simulate the time interval between onset of symptoms of coronary occlusion and arrival of medical personnel to administer treatment. Mannitol infusion. From a group of seven dogs, an intravenous mannitol infusion was started in the five dogs surviving the first 2 hours of ischemia to raise serum osmolarity approximately 30 mOsm (Table II). The mannitol infusion was continued for 4 hours. Serum osmolarity was adjusted to produce hyperosmolarity without substrate infusion to distinguish between the metabolic and osmotic effects of the intravenous metabolic support treatment. Statistical analysis. Statistical analyses were done with the Statistical Analysis Systems (SAS Institute, Inc., Cary, N.C.) and a program package provided by the University of California at Los Angeles Hospital Computing Facility in consultation with Edward C. Del.and, PhD, a biomathematician. Comparisons between groups were made by Student's t test and test statistics were expressed as mean ± SEM standard error of the mean. Group data were compared by analyses of variance in F (range) or Duncan multiple range tests. Differences were considered significant at the p < 0.05 level.

Results The results are summarized in Tables III to VII and Figs. 2 to 7. Ventricular fibrillation occurred within 3 to

Volume sa Number 4

Prolonged acute regional ischemia, V

October 1989

57 I

100

mmHg

80

iNo R

60

x

40 20

t------

Rx------t

2

4

mean ± SEM

3

6

5

Hours

Fig, 3. Arterial blood pressure in treated (Rx) and untreated (no Rx) dogs survivmg LAD occlusion and circumflex stenosis. Note: (I) progressive fall in blood pressure when no treatment was given (solid line) and (2) return of blood pressure to normal after treatment. SEM. standard error of the mean.

100

Rx

80 ml/kg/min

60 40

~,

....

, ,,

,,

,,6- - __"6_-_:0-.1.-_-<;

'" ~ ---~-~---1>-----.,:---

20

.....

t - - - - - - - R x,----------1

2

3

4

5

6

Hours Fig. 4. Cardiac index after LAD occlusion and circumflex stenosis. Note: (I) progressive fall in cardiac index in dogs receiving no treatment (no Rx, solid line) and (2) recovery of cardiac index (dashed line) after treatment Rx).

10 minutes in 61% (range 57% to 79%) of dogs subjected to LAD occlusion after placement of a noncritical circumflex stenosis (Table III). The prevalence of intractable ventricular fibrillation was 30%, despite temporary institution of extracorporeal circulation, and was comparable in all groups (range 25% to 36%, Table III). Remote muscle failed to develop compensatory hypercontractility during the initial 2hour observation period in any of the 26 survivors of the initial ventricular tachyarrhythmias. Progressive cardiagenic shock (stroke work index < 0.5 gm-m/kg) developed in the 14 survivors of LAD occlusion and circumflex stenosis that were either untreated or received a

mannitol infusion after 2 hours of acute ischemia; five of 14 of these dogs died of left ventricular power failure (36% mortality from cardiogenic shock, 57% overall mortality). In contrast, the hemodynamic state of 10 of the 12 dogs receiving intravenous metabolic support improved progressively; remote muscle began to hypercontract and stroke work index returned to normal levels. Two dogs had profound left ventricular power failure during the first 2 hours after coronary occlusion and died of progressive cardiogenic shock despite the intravenous substrate infusion (13% mortality from cardiogenic shock, 38% overall mortality). Control, no ischemia. Intravenous infusion of the

The Journal of

572

Thoracic and Cardiovascular Surgery

Beyersdorf et al.

0.8

SWI

Rx

0.6

(g- m/kgl

0.4

-- --0 TI

I

/

/

I

/

/>--- __ &-- - _-6-/._-0

0.2 1 - - - - - - - Rx - - - - - - - - l

2

3

4

5

6

Hours Fig. 5. Left ventricular stroke work index (SWI) during LAD occlusion and circumflex stenosis. Note: (I) progressive fall in stroke work index over the first 2 hours of acute ischemia. (2) progressive fall in stroke work index when no treatment (no Rx) was administered (solid line). and (3) recovery of stroke work index to normal levels after treatment. tRx, hatched line).

* 20 umol/q

dry weight

10

Control

No Treatment

Treatment

*P
± SEM

Fig. 6. Transmural ATP in remote myocardium after 6 hours of LAD occlusion and circumflex stenosis. Note: ( I ) fall in remote muscle aTP (solid bar) in animals receiving no treatment and (2) maintenance of normal ATP in dogs receiving intravenous substrate infusion (open bar. p < 0.05 versus no treatment group).

substrate solution (Table I) at a rate of 100 ml /hr increased osmolarity to an average of 15 mOsm/L, raised serum glucose concentration to 50 to 100 mg/dl, lowered potassium concentration slightly (from 3.1 to 2.6 mliq/L, Table IV), did not change sodium content, and increased Ca '" minimally (from 1.3 to 1.5 mliq/L). The same substrate infusion did not cause substantial changes in regional contractility, mean arterial pressure, or stroke work index. LAD occlusion and circumflex stenosis Regional contractility. The ischemic anterior left ventricle became dyskinetic immediately after LAD

occlusion, and passive lengthening persisted throughout the observation period in all hearts. Regional segmental shortening fell to 70% ± 10% of control (p < 0.05) in the remote posterior left ventricle supplied by the stenotic circumflex coronary artey during the first 2 hours of observation in the 24 hearts that either did not fibrillate or could be cardioverted. Remote muscle hypocontractility progressed in untreated hearts (Fig. 2), and this trend was not reversed by mannitol infusion. In contrast, the intravenous substrate infusion produced an immediate and progressive increase in remote muscle contractility, with segmental shortening reaching 131 %

Volume 98 Number 4

Prolonged acute regional ischemia. V

October 1989

573

50 )J

mol /g

*

dry weight

30

10

*

Control

No Treatment

Treatment

P
mean ± SEM

Fig. 7. Transmural creatine phosphate in remote myocardium after 6 hours of LAD occlusion and circumflex stenosis. Note: (I) severe creatine phosphate depiction in remote muscle when no treatment was given (solid bar) and (2) maintenance of more normal creatine phosphate levels in dogs receiving intravenous substrate infusion (open bar. p < 0.05 versus no treatment group).

Table m. Incidence of VF. CS, and mortality among groups

Control (n = 4) LAD ligation and LeA stenosis (n = 14) Treatment (n = 16) Mannitol (n = 7) VJIUI:~ arc

Incidence of VF

Intractable VF deaths

0'"/(

Late CS deaths

Survival

n.6'7r (11/14)

0'1, 357'Ir (5/14)

0'1, 33'1, (3/9)

67'1, (6/9)

62.5'1, (10/16) 57.I'1r (4/7)

25'1, (4/10) 28.6'1r (2/7)

17'1, (2/12) 40'1, (2/5)

83'1, (10/12) 60'1( (3/5)

100o/r

mcan z. standard error of the mean. \T. Ventricular fibrillation: CS. cardiogcnic shock: LAD. left anterior descending coronary artery: LeA. Icft circumflex

coronary artery.

of control values after 4 hours of substrate administration (Fig. 2). Systemic hemodynamics. The failure of remote muscle compensatory hypercontractility resulted in progressive left ventricular power failure in untreated hearts; stroke work index fell to less than 0.5 grn-rn/kg, and three of nine dogs died of cardiogenic shock. Mannitol infusion raised serum osmolarity 27 mOsm/L but failed to reverse this trend; irreversible left ventricular power failure and hypotension developed in two of five dogs receiving intravenous mannitol, and reduced cardiac output persisted in the other three dogs. Conversely, the intravenous substrate infusion restored arterial blood pressure (Fig. 3), cardiac index (Fig. 4), and stroke work index (Fig. 5) to normal levels after the first hour in 10 of 12 dogs receiving the metabolic support solution, and this improved hemodynamic status was sustained during the 4-hour infusion. The two dogs that were in profound left ventricular power failure at the start of infusion died after the first hour of treatment.

Regional blood flow. Transmural and regional blood flows were comparable and in the normal range during the control period in treated and untreated hearts (Table V). Acute coronary occlusion produced a profound reduction in transmural flow to the anterior left ventricular wall «10 ml/100 gm/rnin), and flow became redistributed away from subendocardial muscle (endocardial/epicardial ratio <0.50). Regional flow to remote muscle did not change significantly after 2 hours of LAD occlusion and circumflex stenosis despite development of mild hypocontractility. Transmural regional remote muscle flow was slightly higher (78 ± 8 versus 62 ± 7 ml/IOO gm/rnin) after 2 hours of LAD occlusion in untreated hearts and fell to 51 ± 9 and 55 ± 7 ml/IOO gm/rnin, respectively, after 4 and 6 hours of ischemia. In contrast, institution of the intravenous substrate infusion was associated with an increased blood flow to remote posterior muscle after 4 hours (97 ± 21 ml/l00 gm/rnin, p < 0.05) and 6 hours (74 ± 10 ml/100 gm/rnin).

The Journal of Thoracic and Cardiovascular

Beyersdorf et al.

574

Surgery

Table IV. Changes in serum electrolytes with substrate infusion Glucose (mgfdl;

K+ ImEq/L)

115.4± 10.8 222.1 ± 31.7" 203.9 ± 31.9" 147.0±31.6 166.6 ± 73.1"

3.07±0.10 3.00 ± 0.20 2.86 ± 0.11 2.78 ± 0.25 2.60 ± 0.17"

Control 1 hr infusion 2 hr infusion 3 hr infusion 4 hr infusion

Ca++ ImEq/L)

Na+ (mliqfl.:

1.28 1.35 1.37 1.34

148.2 148.2 151.3 152.3 153.2

± ± ± ± 1A8 ±

0.04 0.05 0.06 0.05 0.09"

± ± ± ± ±

1.2 1.6 1.6 2.3" 2.2"

Values are mean ± standard error of the mean.

'p

< 0.05

versus control.

Table V. Regional myocardial blood flow (remote muscle) (milliliters per 100 gm minute) No treatment In = 3) Control Epi Mid Endo Transmural 2 hr ischemia Epi Mid Endo Transmural 4 hr ischemia Epi Mid Endo Transmural 6 hr ischemia Epi Mid Endo Transmural

Table VI. Tissue biochemistry

Treatment

Control In = 8)

in = 6)

56 62 68 62

± ± ± ±

5 8 7 7

52 55 71 60

± ± ± ±

8 10 10 9

ATP Epi

66 79 89 78

± ± ± ±

9 7 7 8

52 51 63 56

± ± ± ±

7 8 4 8

CP Epi

49 51 53 51

± ± ± ±

9 9 11 9

90 88 112 97

± ± ± ±

19" 22" 23" 21"

G-6-P Epi

41 57 66 55

± ± ± ±

5 18 31 7

82 60 80 74

± ± ± ±

18" 15 13" 10

H,O

Values are mean ± standard error of the mean. Epi. Epicardial: Mid. midmyocardial: Endo, endocardial. *p < 0.05 versus control.

Tissue biochemistry. Remote muscle ATP and creatine phosphate concentrations decreased significantly (61% and 42%, p < 0.05) in untreated hearts subjected to 6 hours of LAD occlusion and mild circumflex stenosis (Table VI, Figs. 6 and 7). This decline in high-energy phosphate compounds occurred despite a normal or increased coronary blood flow. It was associated with signs of anaerobic remote muscle metabolism (glucose-e-phosphate levels rose to 2.5 versus 0.6 mmol/ gm dry weight, p < 0.05), reduced tissue NAD content (1200 versus 1410 nmol/gm dry weight), and mild edema (water content rose from 77.5 to 79.0, P < 0.05). The intravenous substrate infusion prevented the loss of remote muscle ATP (Fig. 6), reduced creatine phosphate depletion to only 25% (Fig. 7), limited anaerobic metabolism (1.5 versus 2.5 Jlmol/gm dry weight) preserved tissue N AD content (1720 versus 1400

No treatment (LAD ligation + LCA stenosis; In = 6)

Endo

Endo

Endo

19.0 ± 0.9

A. Epi Endo P. Epi Endo

1.6 0.0 11.9 11.3

± ± ± ±

1.0" 0.0" 2.9" 3.4"

1.8 ± 0.0 ± 20.5 ± 18.0 ±

1.8" 0.0" 2.0 2.6

42.5 ± 3.1

A. Epi

37.2 ± 2.2

Endo P. Epi Endo

12.3 5.7 15.7 18.9

± ± ± ±

2.6" 2.5" 5.6" 6.1"

5.3 5.1 30.8 32.5

± ± ± ±

4.9" 3.7" 5.5" 3.8

0.72 ± 0.19 A. Epi 1.4! Endo 5.41 0.62 ± 0.10 P. Epi 2.51 Endo 2.50

± ± ± ±

0.62" 2.10" 0.98" 1.01"

2.01 7.45 1.15 1.76

± ± ± ±

1.48" 2.34" 0.48' 0.32"

A. Epi 78.5 ± 0.5 Endo 80.2 ± OA" P. Epi 78.6 ± 0.8 Endo 79.3 ± 0.6"

79.1 81.2 77.3 78.4

± ± ± ±

0.3" 0.3" 0.4 0.1

A. Epi 950 ± Endo 760± P. Epi 1110 ± Endo 1320 ±

1020 700 1950 1550

± ± ± ±

100" 200"

17.7 ± 1.3

Epi

77.4 ± 0.5

Endo

77.6 ± 0.6

NAD Epi Endo

Treatment (LAD ligation + LeA stenosis) (n = 5)

1420 ± 180 1400 ± 90

60 120' 180" 180

ISOt 90

Values arc mean ± standard error of the mean. LAD. Left anterior descending coronary artery; LeX, left circumflex coronary artery: ATP, adenosine triphosphate (mieromoles per gram dry weight): CPo creatine phospahte (micromolcs per gram dry weight): G-6-P. glucosc-o-phosphatc (mieromolcs per gram dry weight): Epi, epicardial: Endo, endocardial: A. anterior: P. posterior: 1\AD. nicotinamide-dinucleotide (nonomolcs per gram dry weight). 'p < 0.05 versus control.

tp < 0.05 versus no treatment.

mmol/grn dry weight, p < 0.05), and avoided edema formation (77.8% water content). Mitochondrial analyses. The harvested mitochondria from remote muscle retained relatively normal function in all groups. Only minor changes in State 3 and State 4 respiration, respiratory control index, ADP /0, and oxidative phosphorylation rate were observed (Table VII). The mitochondrial yield was reduced in mannitol-

Volume 98 Number 4 October 1989

575

Prolonged acute regional ischemia, V

Table VB. Mitochondrial function parameter Control In = 5) Epi Endo Treatment (n = 15) A. Epi Endo P. Epi Endo Mannitol (n = 5) A. Epi Endo P. Epi Endo

ST, 362 ± 23 397 ± 23

ST, 50.0 ± 6.0 56.2 ± 6.0

RCI

7.6 ± 0.9 7.2 ± 0.4

A DPjO

2.43 ± 0.07 2.44 ± 0.03

OPR

890 ± 66 860 ± 58

Yield 65.9 ± 4.0 61.3 ± 6.0

292 230 318 365

± ± ± ±

29 33 29 42

45.6 45.3 45.7 52.6

± ± ± ±

4.1 3.3 3.8 4.5

6.5 5.1 7.0 6.9

± ± ± ±

0.7 0.4 0.2 0.5

2.09 2.04 2.15 2.15

± ± ± ±

0.120.110.09 0.11

610 471 685 791

± ± ± ±

707185 107

31.6 15.2 54.8 54.8

± ± ± ±

10.00.411.2t 10.0

298 359 339 318

± ± ± ±

31 123 16 143

42.1 57.5 47.4 55.4

± ± ± ±

5.7 16.9 0.5 22

7.1 6.2 7.2 5.6

± ± ± ±

0.2 0.40.4 0.3-

2.11 2.23 2.26 2.13

± ± ± ±

0.020.170.20 0.14-

628 821 770 696

± ± ± ±

69336 106 346

32.0 24.4 37.6 23.6

± ± ± ±

4.43.64.07.6-

Values arc mean ± standard error of the mean. ST J and ST" State 3 and State 4 respiration (nanoatoms oxygen per minute per milligram protein); ReI, respiratory control index: ADP /0. nanomoles of ADP phosphorylated to ATP per nanoatom of oxygen consumed); OPR, oxidative phosphorylation rate (nanomoles of ATP per minute per milligram protein); Yield. milligrams of mitochondrial protein per gram of tissue dry weight; Epi, epicardial; Endo, endocardial; A. anterior; P. posterior;

•p < 0.05 versus control. tp

< 0.05

versus mannitol.

treated hearts (30.6 versus 54.8 mg protein per gram dry weight, p < 0.05) but remained normal in hearts receiving intravenous substrate infusion.

Discussion This study of acute coronary occlusion in simulated multivessel disease (LAD occlusion and 50% to 60% circumflex stenosis) shows that metabolic support of remote myocardium by intravenous substrate infusion prevents or reverses progression toward cardiogenic shock. Ischemic muscle stopped contracting immediately after coronary occlusion, remained dyskinetic throughout the observation period, and remote muscle exhibited mild to moderate remote muscle hypocontractility during the first 2 hours after coronary occlusion. Remote muscle in untreated hearts did not have the compensatory hypercontractility required to maintain cardiac output, showed signs of anaerobic metabolism, and became depleted at high-energy phosphate stores despite "normal" blood flow. An intravenous substrate infusion started 2 hours after coronary occlusion (to simulate the time between the initial event and medical attention) restored compensatory hypercontractility, improved stroke work index to within normal limits, maintained high-energy phosphates levels, and improved survival. These benefits were not caused by a direct inotropic or osmotic effect, inasmuch as a 4-hour substrate infusion in nonischemic hearts did not change regional and global hemodynamics, and hyperosmolarity alone (i.e., mannitol infusion) did not reverse left ventricular power failure. The severity of the ischemic model is reflected in the prevalence of intractable ventricular fibrillation (i.e., 26%), despite the use of temporary extracorporeal

circulation, and the progressive development of fatal cardiogenic shock in 29% of dogs surviving the initial ventricular tachyarrhythmias. The circumflex stenosis was limited to 50% to 60% to preserve some vasodilator reserve capacity so that remote muscle blood flow could increase to meet energy demands. A more restrictive circumflex stenosis would have produced immediate and profound cardiogenic shock and prevented longitudinal observations of either the natural history of simulated multivessel disease or the effects of the treatment protocols. The intravenous substrate infusion caused remote muscle to develop compensatory hypocontractility and restored stroke work index, cardiac index, and aortic pressure to normal limits in 10 of 12 dogs. The two nonsurvivors were in profound cardiogenic shock when the intravenous substrate infusion was started. In contrast, untreated hearts of those treated by hyperosmolarity alone showed progressive cardiogenic shock; five of 14 dogs died of left ventricular power failure during the 6-hour observation period (36% mortality versus 12% mortality in treated groups, p < 0.05). The intravenous solution was formulated to address several of the metabolic changes shown by us and others to occur in remote myocardium that must carry the hemodynamic burden after acute coronary occlusion.""!' The following principles underlie the composition of the intravenous infusion: (1) glucose-insulinpotassium to provide the glucose that becomes the principal metabolic fuel during ischemia," 11 (2) glutamate and aspartate to restore Krebs cycle intermediates.I and (3) coenzyme QlO to replenish a respiratory chain component that becomes depleted in damaged muscle'? and 2-mercapto-propionyl-glycine to replace sulfhydryl

The Journal of

576

Beyersdorf et al.

groups" needed in enzymes responsible for translocation of ATP from mitochondria (where it is produced) to the cytoplasm (where it is used) for muscle contraction. We 12* have previously used each of these principles to modify the reperfusate in ischemic muscle treated by controlled reperfusion. The current study was designed to test their effects on remote muscle. Our prior studies document the critical role of remote myocardium in causing periinfarction cardiogenic shock/" and confirmed the reported biochemical changes in adjacent and remote myocardium.I: 9. II Remote muscle ATP and concentration were reduced (Table VI), anaerobic metabolism became evident (increased glucose-6-phosphate, Table VI), and levels of remote tissue glutamate (an amino acid precursor of Krebs cycle intermediates) fell despite normal or increased coronary blood flow. Our previous studies show rearrangement of remote muscle mitochondrial membrane proteins that involve soluble enzymes like those in the Krebs cycle. One interpretation is that limited substrate availability may impair the capacity of remote muscle to produce the energy needed for increased mechanical work. Studies by Schwaiger," Leidtke,' and their associates suggest that the metabolic compensation of remote muscle includes preferential utilization of glucose rather than fatty acids as the primary fuel for energy generation. The report by Corday's group" of depleted glycogen levels in remote muscle after acute coronary occlusion suggests limited glucose availability. Glucose-insulin-potassium administration was introduced in 1965 by Sodi-Pallares and colleagues" to limit electrocardiographic signs in acutely infarcting myocardium. The possible mechanisms of glucose-insulinpotassium action on ischemic muscle include the following: (1) decrease in systemic fatty acid levels to reduce impairment of mitochondrial function," (2) increase in myocardial glycogen content to allow for potential use of glucose as the metabolic fuel," (3) shift of the metabolic pathway in the myocardium from lipolytic to glycolytic energy sources," (4) optimization of the effects of the sodium/potassium pump to stabilize cell membranes," (5) antiarrhythmic effects," and (6) potential oxygen free-radical scavenger effects.2M Ischemic muscle dyskinesia persisted throughout the coronary occlusion, so that glucose-insulin-potassium had no discernible mechanical effect on the anterior ventricle. Glucose-insulin-potassium infusions also normalize creatine phosphate content in remote muscle during acute coronary occlusion and enhance the benefits of mechanical circulatory support (intraaortic balloon 'Beyersdorf F. Avoiding reperfusion injury after limb revasculariza-> tion by control of the initial rcperfusion: a new surgical technique. Unpublished data.

Thoracic and Cardiovascular Surgery

pumping"), improve long-term survival after circumflex ligation." and increase performance and metabolism of perfused hearts." These reports support our contention that the substrate infusion improved mechanical function of remote muscle by increasing its energy production. Depletion of ATP and creatine phosphate in untreated remote muscle (Table VI) was documented after 6 hours of coronary occlusion and circumflex stenosis, but creatine phosphate depletion of remote muscle is reported after 3 hours." We suspect that glucose-insulin-potassium improved ATP and creatine phosphate production through both stimulation of oxidative phosphorylation and glycolytic pathways, as suggested by Taegtrneyer." Glutamate and aspartate were added to replenish some of the Krebs cycle intermediates (a-ketoglutarate and oxaloacetic acid) needed for aerobic ATP production via the tricarboxylic acid cycle. These amino acids are potentially useful for energy production via anaerobic substrate level phosphorylation if remote muscle blood flow was insufficient to meet metabolic demands," and they are precursors of Krebs cycle intermediates that participate in the malate-aspartate shuttle to coordinate mitochondrial redox potential. Glutamate becomes depleted in remote hypercontracting muscle.?" and is essential to bind free ammonia, which inhibits the tricarboxylic acid cycle and decreases the intramitochondrial pool of nucleotides (i.e., NAD).J4 The glutamate/aspartate metabolic solution preserved normal tissue NAD in treated hearts, whereas NAD depletion occurred in untreated hearts (Table VI). This shortterm study did not explore the potential long-term benefits of amino acid supplementation in supporting the protein synthesis that accompanies the hypertrophy characterizing remote muscle which sustains compensatory hypercontractility after acute coronary occlu-

sion." The increased uptake of glutamine and release of alanine in patients with coronary artery disease is reported by Thomassen and co-workers" to correlate with the severity of coronary artery disease. We suspect that the amino acid infusion enhanced the aerobic energy production required to sustain the augmented regional contractility and improved cardiac output that followed its administration (Fig. 2). Burns and Reddy? showed that amino acid stimulation of increased oxygen uptake is a prerequisite of increased energy generation, and we" found previously that exogenous glutamate supplementation improves contractility after global ischemia. Pisarenko, Lepilin, and Ivanov" report improved postoperative myocardial performance after intravenous glutamate infusion is associated with reduced ammonia release and increased lactate consumption.

Volume 98 Number 4 October 1989

The roles of coenzyme QIO and 2-mercapto-propionylglycine in improving remote muscle function are speculative. Their action as oxygen radical scavengers in remote "normally" perfused muscle is possible, inasmuch as remote muscle glucose-6-phosphate concentration increased despite normal or increased coronary blood flow, which suggests that relative ischemia might have occurred. Coenzyme QIO and 2-mercapto-propionyl-glycine may also have counteracted any associated oxygen free radicals that formed and were not degraded by the oxygen radical scavengers present normally in red blood cells and in myocardial tissue." Myocardial coenzyme QIO' a phospholipase inhibitor and naturally occurring component of the respiratory chain, is deficient in patients with chronic congestive heart failure and coronary artery disease.'? Its possible depletion in remote muscle during early ischemia may have reduced the efficiency of respiratory chain activity in our experimental model. 2-Mercapto-propionyl glycine is a sulfhydryl compound essential for normal function of the adenine nucleotide translocase enzyme responsible for transferring ATP from the mitochondria to the cytoplasm." Mitochondrial ATP production (i.e., oxidative phosphorylation rate) capacity was normal in treated and untreated hearts (Table VII), so that the failure of remote muscle to develop and sustain increased cardiac work in untreated hearts may be due to impaired ATP utilization. The 2-mercapto-propionylglycine might have replenished sulfhydryl groups deficient in remote muscle and facilitated ATP translocation to the cytosol, where it could be used to produce hypercontractility. Tissue ATP in treated hearts remained at normal levels, and creatine phosphate was decreased only 25% (versus 40% ATP and 53% creatine phosphate depletion in untreated hearts, p < 0.05). The decreased glucose6-phosphate associated with the intravenous treatment provided additional inferential evidence of more efficient aerobic metabolism. Remote muscle blood flow did not differ significantly between treated and untreated hearts, but subsequent studies must be done to test the role of vasodilation without substrate infusion. The beneficial effects of the intravenous infusion were not caused by volume expansion, a direct inotropic effect, or increased osmolarity, because (I) untreated hearts received the same 100 mljmin infusion rate, (2) the substrate infusion did not change global or regional hemodynamics in nonischemic hearts, and (3) mannitol, which reverses myocardial edema and increases coronary blood flow," raised osmolarity more than the substrate infusion (30 versus 20 mOsm, p < 0.05) but did not improve segmental or global function. Additional studies are needed also to determine if remote hypercontractility can be sustained after the metabolic

Prolonged acute regionalischemia, V

577

infusion is discontinued, to define the action of the individual components of the solution, and to explore ways to restore more normal fat metabolism (i.e., carnitine supplementation)." The intravenous composition we selected only reflects a preliminary effort at evaluating the role of nourishing the failing myocardium to improve its contractility, rather than administering inotropic drugs to expend further its energy stores in support of the circulation. The striking improvement in regional contractility that restored normal cardiac output after intravenous substrate infusion suggests the failure of remote myocardium is, in large part, caused by reversible metabolic factors. Our data do not impugn the use of inotropic drugs to treat the failing myocardium but imply that inotropic drug requirements may be either diminished or avoided if appropriate intravenous fluids are formulated to treat specific metabolic causes of impaired muscle function. Our findings suggest that use of inotropic and unloading drugs to improve the mechanical function of the failing heart must be reevaluated, and consideration must be given to providing metabolic support as plans are being made for definitive diagnosis (cardiac catheterization) and revascularization (coronary artey bypass grafting) after acute coronary occlusion. The intravenous metabolic solutions described herein must be considered only as a temporizing way to support remote muscle supplied by stenotic vessels; restoration of unobstructed blood flow is essential to provide abundant oxygen supply so that normal compensatory hypercontractility can be sustained. The data imply also that subsequent studies are warranted to see if the principles which underlie the treatment of remote muscle that does not hypercontract after acute occlusion may be applied also to muscle reperfused after global ischemia, where temporary malfunction ("globally stunned heart") may cause low output syndrome after technically successful cardiac operations. We gratefully acknowledge the outstanding supportive efforts of our research associates, Mr. Edward Dolendo and Ms. Nanci Stellino, and the organizational efforts of Ms. Judith Miller Becker. REFERENCES I. Swan HJC, Forrester JS, Danzig R, Allen HN. Power

failure in acute myocardial infarction. Progr Cardiovasc Dis 1970; 12:568-600. 2. Scheidt S, Ascheim R, Killip T. Shock after acute myocardial infarction: a clinical and hemodynamic profile. Am J Cardiol 1970;26:556-64. 2a. Beyersdorf F, Acar C, Buckberg GD, et al. Studies on prolonged acute regional ischemia. III. Early natural history of simulated single and multivessel disease with em-

The Journal 01

5 7 8 Beyersdorf et al.

phasis on remote myocardium. J THORAC CARDIOVASC SURG 1989;98:368-80. 3. Rackley CE, Russell RO, Mantle JA, Rogers WJ. Cardiogenic shock. Cardiovasc Clinic 1981;2:15-24. 4. Guth BD, White FC, Gallagher KP, Bloor CM. Decreased systolic wall thickening in myocardium adjacent to ischemic zones in conscious swine during brief coronary artey occlusion. Am Heart J 1984; I 07 :458-64. 5. Kerber RE, Marcus ML, Ehrhardt J, Wilson R, Abboud FM. Correlation between echocardiographically demonstrated segmental dyskinesis and regional myocardial perfusion. Circulation 1975;52:1097-104. 6. Lie KI, Liem KL, Schuilenburg RM, David GK, Durrer D. Early identification of patients developing late inhospital ventricular fibrillation after discharge from the coronary care unit. Am J Cardiol 1978;41:674-7. 7. Liedtke AJ, Nellis SH, Whitesell LF. Effects of regional ischemia on metabolic function in adjacent aerobic myocardium. J Mol Cell Cardiol 1982; 1:195-205. 8. Mudge GH, Mills RM, Taegtmeyer H, Gorlin R, Lesch M. Alterations of myocardial amino acid metabolism in chronic ischemic heart disease. J Clin Invest 1976; 58: 1185-92. 9. Schwaiger M, Schelbert HR, Ellison D, et al. Sustained regional abnormalities in cardiac metabolism after transient ischemia in the chronic dog model. J Am Coll Cardiol 1985;6:336-47. 10. Folkers K, Watanabe T, Kaji M. Critique of co-enzyme QIO in biomedical research and in 10 years of clinical research on cardiovascular disease. J Mol Med 1977;2:431-60. II. Theroux P, Franklin 0, Ross J, Kemper WS. Regional myocardial function during acute coronary artery occlusion and its modification by pharmacologic agents in the dog. Circ Res 1974;35:896-908. 12. Allen BS, Okamoto F, Buckberg GO, et al. Studies of controlled reperfusion after ischemia. XV. Immediate functional recovery after 6 hours of regional ischemia by careful control of conditions of reperfusion and composition of reperfusate. J THORAC CARDIOYASC SLJRG 1986; 92(suppl):636-48. 13. Buckberg GO, Luck JC, Payne DB, Hoffmann JIE, Archie JP, Fixler DE. Some sources of error in measuring regional blood flow with radioactive microspheres. J Appl Physiol 1971:31:598-604. 14. Williamson JR, Corkey BE. Assays of intermediates of the citric acid cycle and related compounds by f1uorometric enzyme methods. Methods Enzymol 1969;13:435513. 15. Sordahl LA, Stewart ML. Mechanisms(s) of altered mitochondrial calcium transport in acutely ischemic canine hearts. Circ Res 1980;47:814-20. 16. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Bioi Chem 1951;193:265-75. 17. Peng CF, Kane JJ, Murphy ML, Strub KD. Abnormal mitochondrial oxidative phosphorylation of ischemic myocardium reversed by Cat'-chelating agent. J Mol Cell Cardiol 1977;9:897-908.

Thoracic and Cardiovascular Surgerj

18. Mela L. Inhibition and activation of calcium transport in mitochondria: effect of lanthanides and local anesthetic drugs. Biochemistry 1969;8:2481-6. 19. Reynafarje B, Lehninger AL. High affinity and low affinity binding of calcium by rat liver mitochondria. J Bioi Chem 1969;244:584-93. 20. Corday E, Kaplan L, Meerbaum S, et al. Consequences of coronary arterial occlusion on remote myocardium: effects of occlusion and reperfusion. Am J Cardiol 1975: 36:385-94. 21. Ferrari R, Ceconi C, Curello S, Cargnoni A, Medici D. Oxygen free radicals and reperfusion injury: the effect of ischemia and reperfusion on the cellular ability to neutralize oxygen toxicity. J Mol Cell Cardiol 1986J8(suppl 4):67-9. 22. Sodi-Pallares 0, Testelli MD, Fisleder BL. et al. Effects of an intravenous infusion of a potassium-glucose-insulin solution on the electrocardiographic signs of myocardial infarction. Am J Cardiol 1965;9:166-81. 23. Opie LH, Bruyneel K, Owen P. Effects of glucose, insulin and potassium infusion on tissue metabolic changes within the first hour of myocardial infarction in the baboon. Circulation 1975;52:49-57. 24. Lolley DM, Myers WO, Ray JF, Sautter RD, Tawksbury DA. Clinical experience with preoperative myocardial nutrition management. J Cardiovasc Surg 1985;26:23643. 25. Myears OW, Sobel BE, Bergmann SR. Substrate use in ischemic and reperfused canine myocardium: quantitative considerations. Am J Physiol 1987;253:H 107-14. 26. Zieler KL. Possible mechanisms of insulin action on membrane potential and ion. Am J Mcd 1966;40:735-9. 27. Rogers WT, Segal PH, McDaniel HG, Mantle JA. Russel RO, Rackley CEo Prospective randomized trial of glucose-insulin-potassium in acute myocardial infarction. Am J Cardiol 1979;43:801-9. 28. Hess ML, Okabe E, Poland J, Warner M, Steward JR. Greenfield LJ. Glucose, insulin, and potassium protection during the course of hypothermic global ischemia and reperfusion: proposed mechanism by scavenging of free radicals. J Cardiovasc Pharmacol 1983;5:35-43. 29. Pissarek M, Goos H, Nohring J, et al. Beneficial effect of combined glucose-insulin-potassium and mechanical support in acute myocardial ischemia. Biomed Biochim Acta 1986;45:629-36. 30. Hiatt N, Sheinkopf JA, Warner NE. Prolongation of survival after circumflex artery ligation by treatment with massive doses of insulin. J Cardiovasc Res 1971;5:4853. 31. Weissler AM, Altschuld RA, Gibb LE, Pollack ME, Kruger FA. Effect of insulin on the performance and metabolism of the anoxic isolated perfused rat heart. Circ Res 1972;32: I08-16. 32. Taegtrneyer H. A role for glycolysis in the oxygenated energy deficient rat heart perfused with ketone bodies [Abstract]. J Am Coll Cardial 1985;5:465. 33. Young HH, Shimizu T, Nishioka K, et al. Effect of hypoxia and rcoxygenation on mitochondrial function in

Volume 98 Number 4 October 1989

neonatal myocardium. Am J Physiol 1983;245:H9981006. 34. Pisarenko 01, Lepilin MG, Ivanov YE. Cardiac metabolism and performance during i.-glutamic acid infusion in postoperative cardiac failure. Clin Sci 1986;70:7-12. 35. Badeer HS. Pathogenesis of cardiac hypertrophy in coronary atherosclerosis and myocardial infarction. Am Heart J 1972;84:256-64. 36. Thomassen A, Wielsen IT, Bagger JP, Henningsen P. Myocardial exchanges of glutamate, alanine and citrate in controls and patients with coronary artery disease. Clin Sci 1983;64:33-40. 37. Burns AH, Reddy WJ. Amino acid stimulation of oxygen and substrate utilization by cardiac myocytes. Am J Physiol 1978;235:E461-6.

Prolonged acute regional ischemia, V

579

38. Lazar HL, Buckberg GD, Manganaro AM, Becker H. Myocardial energy replenishment and reversal of ischemic damage by substrate enhancement of secondary blood cardioplegia with amino acids during reperfusion. J THORAC CARDIOVASC SURG 1980;80:350-9. 39. Werns SW, Shea MJ, Lucchesi BR. Free radicals and myocardial injury: pharmacologic implications: Circulation 1986;74:1:1-5. 40. Willerson JT, Powell WJ, Guiney TE. Improvement in myocardial function and coronary blood flow in ischemic myocardium after mannitol. J Clin Invest 1972;51 :298998. 41. Opie LH. Role of carnitine in fatty acid metabolism of normal and ischemic myocardium. Am Heart J 1979; 97:375-88.

Bound \'olurnes available to subscribers Bound volumes ofTHE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY areavailable tosubscribers (only) forthe 1989 issues from the Publisher. at a costof$55.00($72.00 international) forVol. 97 (January-June) and Vol. 98 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped ingold on the spine. Payment must accompany all orders. Contact The C.V. Mosby Company, Circulation Department, 11830 Westline Industrial Drive. St. Louis, Missouri 63146-3318, USA;phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.