Changes in myocardial high-energy phosphate stores and carbohydrate metabolism during intermittent aortic crossclamping in dogs on cardiopulmonary bypass at 34° and 25° C

J

THORAC CARDIOVASC SURG

1990;100:389-99

.Changes in myocardial high-energy phosphate stores. and carbohydrate metabolism during intermittent aortic crossclamping in dogs on cardiopulmonary bypass at 34° and 25° C The effect of cooUng to 25° C on myocardial metabotism was studied during four periods of global ischemia (10 minutes each) foUowed by 15 minutes of reperfusion in dogs on cardiopulmonary bypass. Systemic and heart temperature at normothermia (group N, 34° C; n ;" 15) was compared with general hypothermia (group H, 25° C; n = 16). Before and at the end of each aorticcrossclamp period in small myocardial biopsy specimens the adenosine triphosphate, creatine phosphate, inorganic phosphate, glycogen, and lactate content was analyzed. Also, lactate and inorganic phosphate were measured in the coronary effluent during the repetitive periods of reperfusion. Hemodynamic function was not different at 60 minutes after cardiopulmonary bypass compared with pre--cardiopulmonary bypass values, and was not different between the groups N and H. The tissue content of adenosine triphosphate and glycogen decreased progressively during the experimental period, resulting in slightly depressed values in both groups at the end of cardiopulmonary bypass. Pronounced effects of ischemia and reperfusion on tissue content of creatine phosphate, inorganic phosphate, and lactate were observed after each period of ischemia. The net decrease in tissue creatine phosphate content was not different between groupS N and H(41 ± 4 versus 38 ± 4 ILmol . gm"! dry weigltt; mean ± standard error of the mean) after 10 minutes of ischemia. However, during ischemia the net inorganic phosphate increase in myocardial tissue was significantly higher in group H (70 ± 7 #Lmol. gm-I) than in group N (44 ± 3 ILmo) . gm-I). These findings do not support the notion that myocardial protection is improved during hypothermia. Moreover, quantitatively the release of inorganic phosphate and lactate did not correlate with the amount accumulated in the myocardial tissue during the preceding periods of ischemia. The release appeared to be temperature dependent, that is, significantly reduced at 25° C. The present data demonstrate why clinical outcome is satisfactory in both surgical procedures, when in general the periods of aortic crossclamping do not exceed 10 minutes each and the reperfusion periods in between the ischemic episodes last about 15 minutes. Besides, the findings indicate that hypothermia is not strictly necessary under these circumstances.

Frederik H. van der Veen, PhD, Ger J. van der Vusse, PhD, Peter Willemsen, Ruud T. I. Kruger, Theo van der Nagel, Will A. Coumans, and Robert S. Reneman, MD, PhD, Maastricht. The Netherlands

From the Department of Physiology, University of Limburg, Maastricht, The Netherlands. Supported by the Netherlands Heart Foundation. Received for publication March 29,1989. Accepted for publication Nov. 30, 1989. Address for reprints: F. H. van der Veen, PhD, Department of Cardiology, Biomedical Center, University of Limburg, P.O. Box 616, 6200 MD Maastricht, The Netherlands.

12/1/18507

Repetitive aortic crossclamping to allow performance of the distal anastomoses of the grafts bypassing the obstructedcoronaryvessels has beencommonlyappliedin extensive aorta-coronary bypassoperations.!" This technique aims at short periods of myocardial ischemia, followed by short periods of reperfusion. In this technique the total duration of aortic crossclampingappeared to be significantly shorter than during cardioplegic arrest procedures.l To reduce the possible detrimental effect of 389

390

The Journal of Thoracic and Cardiovascular

van der Veen et al.

Surgery

Table I. Preoperative and postoperative hemodynamic characteristics Pre-CPB

Nonnothennia (n = 9) Heart rate (beats. min-I) Left ventricular pressure (kPa) Left ventricular dPjdt max (kPa . sec") Myocardial blood flow (ml . min-I. 100 gm") Hypothermia (n = 10) Heart rate (beats. min-I) Left ventricular pressure (kPa) Left ventricular dPjdt max (kPa . sec'") Myocardial blood flow (ml . min-I. 100 gm")

15 min after CPB

60 min after CPB

142 10.7 180 51

± ± ± ±

38 1.6 40 8

149 10.0 193 96

± ± ± ±

30 1.9 80 43*

132 10.9 193 88

± ± ± ±

31 2.9 73 35*

143 11.3 220 61

± ± ± ±

46 1.7 67 18

161 lOA 160 98

± ± ± ±

24 3.2 53* 54

138 10.1 193 94

± ± ± ±

29 1.6 80 24*

CPB, Cardiopulmonary bypass; n, number of animals. *p < 0.05 versus pre-CPB values. No significant differences between normothermia and hypothermia groups could be observed.

the ischemic insult inflicted on the heart during aortic crossclamping, body temperature and thus cardiac temperature is reduced during the cardiopulmonary bypass (CPB) period." Reduction of systemic temperature is assumed to provide protection by loweringthe metabolic needs of the myocardial tissue, resulting in a decreased rate of breakdown of high-energy phosphates and reduced accumulation of degradation products," Evidence for the beneficial effect ofhypothermia is usually obtained from the postoperative course of the patients rather than from intraoperative biochemical evaluation.' Intraoperative pH measurements within myocardial tissue, in casu interstitium, clearly showed the beneficial effect of hypothermia, when combined with cardioplegia.9, 10 Although during aorta-coronary bypassoperations the mortality rate has been reported to be about 3%,11,12 death after aorta-coronary bypass operations is usually due to causes other than intraoperative ischemia.13 Actually, after cardiac operations the amount of irreversibly damaged myocardial tissue, as assessed by means of postoperative enzyme release, appears to be very small, irrespective of whether cardioplegia or intermittent aortic crossclamping with moderate (body temperature 32° to 35° C) or more profound hypothermia (body temperature 23° to 28° C) is used.5, 6, 13-16 These unexpected findings prompted us to investigate the changes in cardiac high-energy phosphate and glycogen contents during the course of the surgical procedure in more detail. To this end, during CPB canine hearts were subjected to four subsequent periods of 10 minutes of global ischemia at either 25° or 34° C. Metabolic markers of ischemia were evaluated in the myocardial tissue during both ischemia and reperfusion, and in arterial and coronary sinus blood during the periods of reperfusion. Tissue adenosine triphosphate (ATP)

and creatine phosphate were assayed because these substances are important high-energy phosphate stores.The tissue content and coronary sinus concentration of inorganic phosphate served as a measure of acute intracellular high-energyphosphate degradation. Besides, myocardial glycogen content was analyzed to determine the extent of anaerobic glycolysis. Lactate was measured in myocardial tissueand in coronary sinus bloodto estimate the extent and the duration of anaerobic metabolism. Methods Experimental preparation. Thirty-one healthy mongrel dogs ranging in weight from 19 to 34 kg (average 23.2 kg)were allotted to two groups. The rectal temperature of the dogs in group N was maintained at 34 0 C during the CPB period. In group H the rectal temperature was lowered to 25 0 C with the use of the heat exchanger of the extracorporeal perfusion apparatus (see below). The animals were premedicated with Hypnorm (I mi. kg-I), containing 10 mg fluanisone and 0.3 mg fentanyl citrate per milliliter, as described by Marsboom and coworkers.l? Anesthesia was induced intravenously.with at least 10 mg sodium pentobarbital (Narcovet) per kilogram of body weight. After endotracheal intubation, anesthesia was maintained with oxygen-nitrous oxide and a continuous infusion of sodium pentobarbital (2 mg . kg-I. hr- l ) . The lungs were ventilated with a positive-pressure respirator (Pulmomat, DragerwerkAG, Lubeck, Federal Republic of Germany). The tidal volume and the rate of the ventilator were adjusted to keep the arterial carbon dioxide pressure between 30 and 40 mm Hg. Base excess was kept around - 2 mmol . L-I by means of sodium hydrogen carbonate. Before sternotomy, succinylcholine (2 mg. kg-I body weight, intramuscularly) was injected to prevent muscle movement. The sternum was split over its total length and the pericardium was opened completely. The electrocardiogram was derived from the limb leads. A catheter placed in the left femoral artery and one in the left femoral vein were connected to Ailtech pressure transducers (Electromedics, Englewood, Colo.) to measure arterial and

Volume 100 Number 3 September 1990

Cardiac metabolism during aortic crossclamping 3 9 1

. - . NORMOTHERMIA

0 0 HYPOTHERMIA

30,-------------------------------,

20

o E

-?-

10

Q

f-

-r

o

IE

10 20

35 45

60 70

85 95

145

- - - - - - - - - - - - - - - - - - - - - - - - - - - -1 Time (min)

Fig. 1. Content of ATP in myocardial biopsy specimens obtained during CPB. Data are presented as mean ± standard error of the mean of six animals. The values between the two groups were not significantly different. ATP, Adenosine triphosphate; AC, aortic crossclamping; E, empty-beating, * Significantly differentfrom the first ~alue obtained during empty-beating (p < 0.05). peripheral venous pressure, respectively. A catheter placed in the leftatrium wasusedfor microsphere injections, and a catheter placedin the left brachialartery wasused to collectarterial reference samples during microsphere injections. After heparinization (250 V . kg-I bodyweight) and sternotomy, a balloon-guided thermistorcatheter (7F, KMA, Boston Scientific Corp., Mansfield Div.,Mansfield, Mass.) was inserted through the left jugular vein for cardiac output measurements by the thermodilution technique according to Snoeckx and co-workers.P An electromagnetic flow probe connectedto an electromagnetic flowmeter (Transflow 600, Skalar, Delft, The Netherlands) was mounted on the left anterior interventricularcoronaryartery to measurevolumebloodflow. A SF or 7F catheter was inserted into the left jugular vein and moved intothe right atrium. Approximately 2 em of the catheter endingwascarefullyplacedin the coronarysinusto collectcardiac venous blood samples. A thermocouple wasplacedin the rectum to recordrectal temperature. In the subgroupsof six dogs used to obtain multiplebiopsy specimens, thermistors (type A-KCI, Ellab,Copenhagen, Denmark) were placedin the anterior and posterior left ventricularfree wall to monitor cardiac temperature throughoutthe experimental period. Just beforethe onset of CPB, a stainless steel cannula, used forthe arterialreturn from the oxygenator, wasinsertedintothe right femoral artery. Subsequentlya tube was insertedinto the right atrium for venous drainage by gravity.Immediatelyafter thestart ofCPB,a leftventricularventwasinsertedthroughthe apex. CPB was carried out with a Harvey (H200, C.R. Bard, Inc., Santa Ana, Calif.) or a Polystan (VT 5000, Polystan, Copenhagen, Denmark) bubble oxygenator and a Polystan rollerpump.The oxygenator wasprimed with 1 L of fresh heparinizeddonorblood, 2 L of electrolytes (Normosol),and 80 ml

of 4.2%sodium bicarbonate,to which 1000 IV of heparin and 1000mg of calciumgluconate(in solution) wereadded.19 Routinely the carbon dioxide percentage of the sweep gas of the oxygenator was kept at 4% to 5%. However, during hypothermic perfusions (25 0 C) the carbon dioxidetensionof the arterial bloodwaslowered to 1.3to 2.7 kPa, resultingin an increase of pH to 7.6 to 7.8, accordingto Beckerand co-workers.PFentanyl (0.3 to 0.5 mg total dose) was administered at regular intervals during hypothermic perfusions to prevent cold sensation." During extracorporeal circulation, fluid was added to the oxygenator on the basisof hematocrit values. Donor bloodwas given when the hematocrit value decreased below 25%. In group N, fibrillation was induced by touching the heart with battery electrodes. In all experiments the hearts were defibrillated with a direct-current defibrillator (B-D, Electrodyne Co., Inc., Norwood, Mass.) delivering 40 joules totally. There was no support with inotropicdrugs, and anesthesia was standardizedas much as possible to minimizeinfluences of different anesthetic protocols on the parameters measured. Only animalsthat did not belongto the subgroupfor biopsyanalysis were weaned from bypass, because most likelythese multiplebiopsy holeswillnegatively influence myocardialfunction after CPB. Hemodynamic recovery was assessed after 15 and 60 minutes. The animals received humane care in compliance with the Netherlands Guide for the care and use of laboratory animals. Biochemical analysis. For biochemical analysisboth arterial and coronary sinus bloodwas collected. During extracorporeal circulation arterial blood was obtained from the oxygenator. For the determinationoflactate, inorganicphosphate,and glucose,S rnl of bloodwasused, and sampleswere immediately

The Journal of Thoracic and Cardiovascular Surgery

3 9 2 van der Veen et al.

'3'o CJ)

<,

90

0

70

2-

60

E

(j)

+J

ro

J:: Q

(J)

0

50 40

J::

30

(j)

20

Q.

f;

T*

'T

ro

U

:l

$,

.'i~ " ",*:' ,

',

'9

-*

, ,,

\i "

:

'It:* ~*

-L

+J

OJ

-

80

10

0

IE

10 20

35 45

60 70

85 95 Time

(min)

Fig. 2. Content of creatine phosphate in myocardial biopsy specimens. For explanation see legend to Fig. I. No differences between the two groups were observed,

centrifuged. The plasma was stored at -80° C until analyzed. Lactate, inorganic phosphate, and glucose in plasma were measured as described by Van der Vusse and co-workers.F Transmural biopsy specimens were taken from the anterior free wall of the left ventricle in the perfusion area of the left anterior interventricular coronary artery (LAICA) with a small biopsy needle (Tru-Cut; Travenol Labs., Inc., Deerfield, Ill.). These biopsy specimens were obtained in group N (n = 6) and group H (n = 6) before the first aortic crossclamping, and at the end of each crossclamping and reperfusion period. The first biopsy specimen, taken when the heart was in the empty beating state, was considered to be the control, that is, preischemic, value. Analysis of ATP, creatine phosphate, and glycogen in the myocardial tissue was performed, as described in detail by Van der Vusse and co-workers.P Shortly, the freeze-dried biopsy specimen was crushed in a 3 mol . L-I perchloric acid solution at -15° C. This mixture was neutralized with potassium carbonate, and after centrifugation the supernatant was used for the determination of ATP, creatine phosphate, inorganic phosphate, and lactate. The residue was used for glycogen analysis. Inorganic phosphate was analyzed by spectrophotometry according to Van Belle's method.l" Lactate was analyzed by fluorimetry at 25° C according to Passonneau's method. 25 Myocardial blood flow. Volume blood flow in the LA1CA was continuously recorded during the experiments. Total left ventricular blood flow was assessed with the use of radioactively labeled microspheres during a stable hemodynamic period before CPB according to the method of Prinzen, Van der Vusse, and Reneman.P The microspheres (15 ~m diameter) were labeled with 113Sn, I03Ru,141Ce, or 95Nb. Because LAICA flow was measured at the same moment, we were able to calculate a factor by which LAICA flow data (expressed as ml . min-I) could be transformed into normalized flow data (ml. min-I. gm"! of tissue) throughout the experiment.F Experimental protocol during CPB. The animal experiments were performed to investigate in more detail cardiac metabolism during aorta coronary bypass operations. Therefore, the protocol was set up according to clinical protocols pre-

sented in the literature.v 5,13 determining, among other things, duration of CPB, duration of aortic crossclamping, number of crossclampings, myocardial and systemic temperature, and electromechanical state of the heart. The total duration of the extracorporeal circulation was 145 minutes in all experiments. Routinely, cooling in group H was initiated after the start of CPB. After approximately 5 minutes, blood samples and tissue biopsy specimens were collected when the heart was in the empty beating state. Next, as a result of cooling, the hearts in group H started fibrillating and remained fibrillating until rewarming, when rectal temperature had reached 34° C. Thereafter the hearts were deliberately defibrillated. In both groups the first aortic crossclamp was placed 15 minutes after the start of CPB. Each crossclamp period lasted 10 minutes, and the intermittent reperfusion periods lasted 15 minutes. Unlike in group H, in group N the heart was set to fibrillate directly after crossclamping of the aorta and defibrillated after release of the crossclamp. Consequently these hearts were in the empty beating state during the intermediate reperfusion periods. To investigate the effect of the separate periods of ischemia, we obtained biopsy specimens before and at the end of each period of aortic crossclamping. Besides, arterial and coronary sinus blood samples were collected during reperfusion at 0.5, 1,2,3,4,5,6,7.5, and 10 minutes after the release ofthe aortic crossclamp. At the end of CPB, a last set of measurements was taken. During CPB, pump flow was adjusted according to the preCPB cardiac output values and increased when mean aortic pressure decreased below 6.0 kPa. As a result, in group N (preCPB cardiac output 2.5 ± 0.7 L . min-I) pump flow during and at the end (2.1 ± 0.8 L . min-I) of the experiment was stable. Also in group H (pre-CPB cardiac output 2.4 ± 0.8 L . min-I), no major changes in pump flow were observed (2.1 ± 0.6 L . min"! at the end of the CPB). Statistical analysis. Intraindividual differences in the release and uptake of metabolites within one group were evaluated for statistical significance with the Wilcoxon matchedpair signed-rank test (two-tailed probability) with use of the

Volume 100

Cardiac metabolismduring aortic crossclamping 3 9 3

Number 3 September 1990

e-e NORMOTHERMIA 150

o E

0------0 HYPOTHERMIA

r-----------------------------,

120

9*

~ (lJ

+--

ell

£

0-

([)

o

£ 0-

U

c

ell

oc OJ

o

IE

10 20

35 45

60 70

145

85 95

I

Fig. 3. Content of inorganicphosphatein myocardialbiopsy specimens. For explanation see legendin Fig. 1. No differences between the two groups were observed.

BMDP 3S program. Intraindividual differences in the biopsy data (n = 6) wereanalyzedwiththe useofthe two-sample t test. Differences between the twogroupswere evaluatedfor statistical significance by applying the one-sample t test. Statistically significant differences were accepted at p < 0.05.

Results Hemodynamic variables. No differences in heart rate, left ventricular pressure, left ventricular peak dP / dt, and myocardial blood flow were observed between the two groups before CPB (Table I). As the large interindividual variation shows, heart rate was deliberately not adjusted by means of pharmacologic interventions. During extracorporeal circulation pump flow was not significantlydifferent between the two groups, but perfusion pressure was higher in group H than in group N (9.5 ± 1.7 versus 6.1 ± 1.7 kPa after 105 minutes on CPB). Hemodynamic recovery revealed adequate left ventricular pressure and left ventricular peak dP / dt in both groups at 60 minutes after extracorporeal circulation, whereas myocardial blood flow remained at significantly higher values. Signs of "no reflow" or myocardial stunning were absent. Temperature of the heart. The temperature of the posterior and the anterior left ventricular wall remained at about 25° C in group H during the periods of aortic crossclamping. During ischemia at normothermia, the temperature of the anterior wall was also not affected, whereas the temperature of the posterior left ventricular

Table II. Arterial concentrations of lactate, inorganic phosphate. and glucose* Group

Lactate Pre-CPB 5 min on CPB 135 min on CPB Inorganic phosphate Pre-CPB 5 min on CPB 135 min on CPB

Glucose

Pre-CPB 60 min on CPB 135 min on CPB

Normothermia

Hypothermia

(n= 15)

(n= 16)

2.4 ± 1.3 3.9 ± 2.3 8.1 ± 2.6

3.0 ± 2.2 4.9 ± 2.4 8.6 ± 2.5

2.0 ± 0.5 1.4 ± 0.5 1.2 ± 0.3

1.6 ± 0.7 1.2 ± 0.5 0.8 ± 0.4

6.3 ± 0.9 3.6 ± 1.4 2.4 ± 1.7

7.2 ± 0.9 4.9 ± 1.2 5.0 ± 1.5

'Values are mean ± standard deviation (mmol . L -I)

wall decreased significantly from 33.4° ± 1.0° to 31.2° ± 1.2° C. Tissue content of biochemical substances. The changes in tissue content of ATP, creatine phosphate, and inorganic phosphate during CPB are shown in Figs. 1,2, and 3, respectively. In both group N and group H the myocardial content of ATP tended to decrease because of the repetitive periods of ischemia. In group N the values measured after the first and the third aortic crossclamping reached the levelof significance (p < 0.05), compared with the preischemic control value (E) (Fig. 1). Differ-

The Journal of

3 9 4 van der Veen et al.

Thoracic and Cardiovascular Surgery

. - . NORMOTHERMIA

0 · 0 HYPOTHERMIA

200 , - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ,

'3 'o 01

<,

0

E 2c

100

())

01

0

U

>,

l5

o

IE

10 20

35

45

'---

lAC]

lAC]

60 70

85 95

145 ---J

Time (min)

Fig. 4. Content of glycogen in myocardial biopsy specimens (umol glucose moieties· grn"! dry weight). For explanation see legend to Fig. 1. Values between the two groups were not statistically different.

Table III. Relation between content of ATP, creatine phosphate, and inorganic phosphate in myocardial biopsy specimens at end ofeach singleperiodof ischemiawith a duration of 10 minutes, and inorganic phosphate release during the subsequent reperfusion period. * Group .. Normothermia

Hypothermia

(n = 24)/

(n = 24jI

4±2 41 ± 4

38 ± 4

44 ± 3

70 ± 7t

4.9 ± 1.3

\.I ± 0.2t

Decrease in tissue content

ATP Creatine phosphate Increase in tissue content Inorganic phosphate Washout of inorganic phosphate into venous blood

2±2

'Inorganic phosphate release was calculated by planimetry with use of measurements obtained after 0.5, I, 2, 3, 4, 5, 6, 7.5, and 10 minutes of reperfusion, Mean values and standard error of the mean are presented (/.Imol . gm"? dry weight). For the sake of comparison the data obtained during the four individual periods of ischemia and reperfusion, respectively, were grouped for all six animals. tSignificantly different from the normothermia group (p < 0.05).

ences between the two groups were not observed at each individual time point. The creatine phosphate content of the myocardium decreasedsignificantly in both groupsafter each periodof ischemia and returned to or exceededthe control values after 15 minutes of reperfusion (Fig. 2). The pattern of changes in creatine phosphate was similar during the

succeeding periods of reperfusion. No significant differences between groups Nand H were observed. In both groupsthe content of inorganicphosphatewas significantly increasedat the end of each ischemic period of 10minutes(Fig. 3). The increasewasmostpronounced in group H, ranging from 61 to 81 JLmol. gm- I dry weight.In both groupsrestorationto the preischemic levels of inorganicphosphateoccurred within 15 minutesof reperfusion. The tissue content of inorganic phosphate was not different betweenthe two groups at any point in time. The median valuesof the glycogen contentin the myocardial biopsyspecimens showeda gradual decreaseduring the periodof CPB (Fig. 4). In group N the maximal decrease (to 74% of the initial preischemic value) was observed after the fourth aortic crossclamping. In group H the maximum decrease (to 63% of the preischemic value) was observed after the third reperfusion period. The changes that occurred during the repetitiveperiods of ischemiaand reperfusion were inconsistent; the values werenot significantly differentbetweenthe twogroupsat any time. The changes in the myocardial content of lactate are depicted in Fig. 5. During the empty beating state (control period) the lactate content was low (mean values varying from 25 to 37 JLmol . gm"! dry weight) in both groups,but substantiallyincreasedduring each periodof aortic crossclamping (mean increase ranging from 65 to 117JLmol . gm-I dry weight).After 15minutesof reper-

Volume 100 Number 3 September 1990

Cardiac metabolism duringaortic crossclamping 3 9 5

.-.NORMOTHERMIA

0--0 HYPOTHERMIA

150 , - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ,

120

,:\/i* T

90

. ·:····.\Ti . . • '. .L...-

60

•

30

o

'(:i

IE

AC 10 20

35 45

60 70

85 95

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _....J

Fig. 5. Contentof lactate in myocardial biopsy specimens. For explanation see legend to Fig. 1. Because of technical problems this biochemical parameterwasmeasured onlyin a limited numberof biopsy specimens in group H (n

= 2).

fusion the lactate contentdecreased significantly in both groups, and the values at the end of CPB wereno longer significantly different from the preischemic levels. Release and uptake of biochemical substances Arterial blood. Duringextracorporeal circulation, the arterial concentration of lactate increased gradually in bothgroups (Table 11). Concomitantly, in both groups a significant decrease in the arterial concentration of inorganic phosphate and glucose was observed. In Figs. 6 and 7 the myocardial release or uptake of inorganic phosphate and lactate,respectively, duringthe repetitive periods of reperfusion after lO minutes of global ischemia are depicted. These data wereobtained by multiplying the arterial-eoronary sinus differences of these substances and myocardial blood flow. Duringthe initial phaseof reperfusion the release of inorganic phosphatewasless in groupH than in groupN. In generalthe peakvalues of inorganic phosphate werenot significantly different from each other within one group. Exceptions are the values measured after the second and third aortic crossclamp periods, being slightly, but significantly, lower than the preceding peak value. In mostcases inorganic phosphate was released from the myocardial tissue only during the first minute of reperfusion, and subsequently this release turned into uptake. In groupN the uptakerate washigherwhenmore inorganic phosphate wasreleased duringthe first30 seeonds of reperfusion.

Table IV. Relationbetween content ofglycogen and

lactate in myocardial tissue at end ofeach single periodof ischemia with a duration of 10 minutes, and lactate release during reperfusion. * Group

Decrease in tissueglycogen content Lactate increasein tissue Washout of lactate into venous blood

Normothermia (n = 24) 15±8

Hypothermia (n = 24) -13±14

68 ± 6 48 ± 11

55t 16 ± 2:1:

'Lactate releasewascalculatedas describedfor inorganicphosphatein Table III. Mean values and standard error of the mean are presented ()LIlIol . gm'" dry weight).For the sake of comparison the data obtained during the four individual periodsof ischemia and reperfusion, respectively, weregroupedfor all sixanimals. tlndicates that tissuelactatewasdeterminedin onlytwodogsof the hypothermia group. tSignificantlydifferentfrom the normothermiagroup (p < 0.05).

Duringthe empty beatingperiod, beforethe first aorticcrossclamping period, nosignificant amountsoflactate wereextractedor released by the heart in either groups (Fig. 7). Immediately after the lo-minute period of aortic crossclamping, significant amounts of lactate were released on reperfusion. The peak values of release were reached within the first 30 seconds after restoration of flow. These peak values were similar after repetitive crossclamping in group N. In group H, the peak release

The Journal of Thoracic and Cardiovascular

3 9 6 van der Veen et al.

Surgery

Inorganic phosphate ~mol.g- I .rnin - I

-2.5 -2.0 * +

-1.5

* +

*

-1.0 -0.5

o

Normothermia

• AC

+0.5 -1.0

Hypothermia

-0.5

o

• i

i'

152540

Time (min)

Fig. 6. Release (-) /uptake ( +) of inorganic phosphate in the normothermia (n = 15) and the hypothermia group (n 16) (umol . gm"" wet weight . min-I). Data are presented as mean ± standard error of the mean. E. Emptybeating; AC, aortic crossclamping; CPR,"cardiopulmonar y bypass. *Significantly different from the first value obtained during empty-beating (p < 0.05). [ Significantly different from the hypothermia group (p < 0.05).

=

was slightly, but significantly, lower during the third reperfusion period, compared with the first peak value. Similar to the releaseof inorganicphosphate(seepreceding), the release of lactate turned into uptake about 3 minutes after the start of reperfusion. Discussion

The present findings indicate that four repetitiveshort periodsof ischemia,lasting 10minutes each, are welltolerated by the heart from a metabolicand functionalpoint of view. When the periods of flow cessation were alternated with periods of reperfusion of 15 minutes, the increased tissuelevels of inorganicphosphateand lactate returned to normal and the creatine phosphate content recovered completely. ATP content showed a minor reduction at the end of the CPB period. Reduction of temperature during the successive periods of ischemia hardly influenced the extent of creatine phosphatedepletion and lactate and inorganic phosphate accumulation during the ischemic insult. This is an unexpectedobser-

vationsincereductionof temperature hascommonly been advocated to mitigate the unwanted effectsof flow deprivation on the metabolic state of cardiac tissue during heart operations. A satisfactoryexplanationfor this finding cannot be offered. Becausein the present canine study the effectof individual periodsof ischemia and reperfusion on tissuelevelsof ATP, creatine phosphate,inorganicphosphate, glycogen,and lactate and the releaseof inorganicphosphate and lactate has beenmonitored,conclusions can bedrawn about the effect of cardiac temperature on ischemiaand reperfusion separately. Comparison of the amount of ATP and creatine phosphatehydrolyzed on the one hand and the amount of inorganic phosphate accumulated on the other (Table III) showsa good quantitative relation between hydrolysis and accumulation in the normothermia group. In contrast, an additional sourceof inorganic phosphateappears to be presentin the hypothermiagroup because the tissue accumulation of inorganic phosphate (onthe average70 JLmol . gm- J) exceeds the reductionof

Volume 100 Number 3

Cardiac metabolism during aortic crossclamping 3 9 7

September 1990

Lactate

-6

*

IJ mol.g.- 1min; I

-5

..

*

-4

*

-3 *

-2 -I

0

•

AC

AC

-5 -4

AC

* + - - -- ---

/ItC -

.

*

..

-2

+1

+

.

-3

0

.

*

---- --- ---- ---

+1

-I

Normothermia

Hypothermia

•

AC

AC

I~ 0 152540 I

I

.

I

AC

CPS

AC

, 100

I

145 Time (min)

Fig. 7. Release (-)/uptake (+) of lactate in the normothermia (n = 15) and the hypotherm ia group (n = 16) (umol . gm"! wet weight . min"). Data are presented as mean ± standard error of the mean. E, Empty-beating; AC, aortic crossclamp ing; CPB, cardiopulmonary bypass. "'Significantly different from the first value obtained during empty-beating (p < 0.05). [Significantly different from the hypothermia group (p < 0.05).

ATP and creatine phosphate content (on the average 40 JoLmol . gm"). The nature of the additional source of inorganic phosphate remains to be revealed. The data in Table III also indicate that about 10% of the inorganic phosphate accumulated at the end of the ischemic period is washed out from the tissue during the subsequent reperfusion period in the normothermia group. In the hypothermia group less than 2% is released from the tissue upon reinstallation of flow. A possible explanation is that the rate of passage th rough cell membranes is reduced at lower temperatures.P These findings indicate that measurement of the amount of inorganic phosphate, as released from previously ischemic cardiac tissue into the blood, provides an incomplete picture of the metabol ic changes that occur in the tissue during the period of flow cessation. Hence data concerning arterial-cardiac venous differences of inorganic phosphate should be interpreted with great care.

The findings, as depicted in Fig. 7, indicate that hypothermia does not prevent the anaerobic production oflactate during repetitive short periods of ischemia. The increase in tissue lactate content during ischemia to three to four times the initial preischemia value and subsequent return to the initial low levels after restoration of flow are in agreement with previous findings.29-31 Because the changes in tissue lactate content during the ischemic period are in the same order of magnitude in hearts of the hypothermia and normothermia groups (Table IV), reduction of temperature from 34 ° C to 25 ° C has an insignificant effect on the accumulation of anaerobically produced lactate. Because tissue levelsjust before and at the end of 10 minutes of flowcessation were analyzed, the present data do not allow the conclusion that the rate of glycolysis is completely insensitive to lower tissue temperatures. It IS possible that immediately after the onset of ischemia lactate production is higher in

The Journal of Thoracic and Cardiovascular

3 9 8 van der Veen et al.

the normothermic than in the hypothermic hearts. The final maximal concentration of lactate is then obviously reached within 10 minutes of flow cessation at both temperatures investigated. As with inorganic phosphate, a significantly lower proportion of lactate is released from the previously ischemic tissue into the vascular compartment after restoration of flow in the hypothermia group. Whether this phenomenon is due to a reduced transport capacity of lactate across the sarcolemma at lower temperatures/" or to increased intracellular consumption during the reperfusion phase is unknown. This finding stresses the notion that extrapolation of changes in the concentration of substances, such as lactate, in local venous blood to changes in the tissue content during the preceding ischemic insult should be done with care. Intramyocardial pH measurements have shown the beneficial effect of hypothermia in combination with icecold cardioplegia," More importantly, the latter authors reported that an arbitrary terminal pH of 6.8 was reached after about 19 minutes of ischemia at normothermia (38 0 C) without cardioplegia. Ten minutes of ischemia at 34 0 C, as presented in our study, therefore appears to be within safe limits. Also, Khuri and co-workers'? observed a significant acidosis during the latter part of aortic crossclamping (120 minutes), which correlated with postischemic functional recovery. These findings cannot be extrapolated to our study, however, because of differences in duration of crossclamping and the additive protection with cardioplegia. Quantitative comparison oftissue glycogen and lactate levels during ischemia and reperfusion indicates that the contribution ofglycogen degradation to the production of the end product of the glycolytic pathway, that is, lactate, is minor under the present experimental conditions. An alternative source of lactate production in the present dog heart model is glucose, trapped in the tissue at the onset of ischemia. Assuming that the cellular content of glucose is in equilibrium with that of plasma, the amount of glucose trapped is sufficient to induce the anaerobic production of 50 ~mollactate . gm- 1 dry weight of tissue. This production can explain the changes, as observed in the hearts at the end of the ischemic period (Table IV). The present animal data confirm the clinical findings that in patients undergoing aorta-coronary bypass operations with intermittent aortic crossclamping, reduction of temperature of the heart from 34 0 C to 25 0 C has only minor effects on the overall metabolic changes in the heart, as measured in those patients before and after all four periods of aortic crossclamping.P The more detailed animal study clearly indicates that generally changes in cardiac high-energy phosphate stores and tissue lactate levels during 10 minutes of ischemia are comparable in

Surgery

hearts kept at either 34 0 C or 25 0 C during the surgical procedure. During the intermittent periods of reperfusion, creatine phosphate, inorganic phosphate, and lactate concentrations appear to return to normal levels at both temperatures. These metabolic findings explain the clinical observation that the functional outcome of patients in whom aorta-coronary anastomoses were performed with the intermittent aortic crossclamping technique, either at 34 0 Cor 25 0 C, is satisfactory when the ischemic periods last on the average of 10 minutes, alternated with reperfusion periods lasting on the average of 15 minutes. 4• 5• 13 We are grateful to Mrs. C. Legdeurand Mrs. V.Lejeune for typing the manuscript. REFERENCES 1. Delva E, Maille JG, Solymoss BC, Chabot M, Groudin CM, Bouorassa MG. Evaluation of myocardial damage during coronaryartery grafting with serial determinations of serum CPK-MB isoenzyme. J THORAC CARDIOVASC SURG 1978;75:467-75. 2. ChitwoodWR, Hill RC, Sink JD, WechslerAS. Diastolic ventricularproperties in patients during coronaryrevascularization: intermittent ischemicarrest versus cardioplegia. J THORAC CARDIOVASC SURG 1983;85:595-605. 3. McGregor CMA, Muir AL, Smith AF, et al. Myocardial infarction related to coronary artery bypassgraft surgery. Br Heart J 1984;51:399-407. 4. Bonchek LI, Burlingame MW. Coronary artery bypass without cardioplegia. J THORAC CARDIOVASC SURG 1987;93:261-7. 5. Flameng W, Van der Vusse GJ, De MeyereR, et al. Intermittent aortic cross-dampingversusSt. Thomas' Hospital cardioplegia in extensive aorta-coronarybypass grafting. J THORAC CARDIOVASC SURG 1984;88:164-73. 6. Pepper JR, LockeyE, Cankovic-Darracott S, Braimbridge MV. Cardioplegia versus intermittent ischaemic arrest in coronary bypass surgery. Thorax 1982;37:887-92. 7. ChitwoodWR, Sink JD, Hill RC, WechslerAS, Sabiston DC. The effects of hypothermia on myocardial oxygen consumption and transmural coronary blood flow in the potassium-arrested heart. Ann Surg 1979;190: 106-16. 8. Stiles QR, Kirklin JW. Myocardial preservation symposium. J THORAC CARDIOVASC SURG 1981;82:870-7. 9. Takach TJ, Glassman LR, Milewicz AL, Clark RE. Continuous measurement of intramyocardial pH: relative importance of hypothermia and cardioplegic perfusion pressureand temperature. Ann Thorac Surg 1986;42:36571.

10. Khuri SF, Warner KG, Josa M, et al. The superiority of continuous coldbloodcardioplegia in the metabolic protection of the hypertrophied human heart J THORAC CARDlOVASC SURG 1988;95:442-54.

Volume 100 Number 3 September 1990

Cardiac metabolism during aortic crossclamping 3 9 9

II. Kaiser Gc. CABG 1984: technical aspects of bypass surgery. Circulation I 985;72(Pt 2):V46-58. 12. Teoh KH, Christakis GT, Weisel RD, et al. Increased risk of urgent revascularization. J THORAC CARDIOVASC SURG 1987;93:261-7. 13. Van der Vusse GJ, Van der Veen FH, Flameng W, et al. A biochemical and ultrastructural study on myocardial changes during aorta-coronary bypass surgery: St. Thomas Hospital cardioplegia versus intermittent aortic crossclamping at 34° and 25° C. Eur Surg Res 1986;18:1-11. 14. Conti VR, Bertranou EG, Blackstone EH, Kirklin JW, Digerness SB. Cold cardioplegia versus hypothermia for myocardial protection. J THORAC CARDIOVASC SURG 1978;78:708-20. 15. Lolley DM, Ray JF, Myers WO, Sautter RD, Sheldon G. Is reperfusion injury from multiple aortic cross-clamping a current myth of cardiac surgery? Ann Thorac Surg 1980;30:110-7. 16. Roberts AJ, Sanders JH, Moran JM, et al. Nonrandomized matched pair analysis of intermittent ischemic arrest versus potassium crystalloid cardioplegia during myocardial revascularization. Ann Thorac Surg 1981;31:502-11. 17. Marsboom RA, Verstraete D, Thienpont D, Mattheeuws D. The use of halo-anisone and fentanyl for neuroleptanalgesia in dogs. Br Vet J 1964;120:466-8. 18. Snoeckx LH, Verheyen JL, Van de Water A, Lewi P, Reneman RS. On-line computation of cardiac output with the thermodilution method using a digital minicomputer. Cardiovasc Res 1976;10:556-64. 19. Van der Veen FH, Van der Vusse GJ, Lelkens JP, Reneman RS. Ionized Ca 2+ in blood during hypothermic cardiopulmonary bypass in dogs. Scand J Clin Lab Invest 1983;43(suppI163):115-6. 20. Becker H, Vinten-Johanson J, Buckberg GD, et al. Myocardial damage caused by keeping pH 7.40 during systemic deep hypothermia. J THORAC CARDIOVASC SURG 1981;82:810-20. 21. Henschel WF, Geldmacher H, Grabow G. Neuroleptanalgesie und kontrollierte Hypothermie. In: Neue klinische Aspekte der Neuroleptanalgesie. Stuttgart: Schattauer, 1970:198-205.

22. Van der Vusse GJ, Van Belle H, Van Gerven W, Kruger R, Renernan RS. Acute effect of fentanyl on hemodynam-' ics and myocardial carbohydrate utilization and phosphate release during ischemia. Br J Anaestb 1979;51:927-35. 23. Van der Vusse GJ, Cournans WA, Van der Veen E, Drake AJ, Flameng W, Suy R. ATP, creatine phosphate and glycogen content in human myocardial biopsies: markers for the efficacy of cardioprotection during aorta-coronary bypass surgery. Vase Surg 1984;18:127-34. 24. Van Belle H. New and sensitive reaction for automatic determination of inorganic phosphate and its application to serum. Anal Biochem 1970;33:132-42. 25. Passonneau JV. Lactate: fluorirnetric method. In: Bergmeyer HU, ed. Methods of enzymatic analysis. New York: Academic Press, 1974:1488-72. 26. Prinzen FW, Van der Vusse GJ, Reneman RS. Blood flow distribution in the left ventricular free wall in open chest dogs. Basic Res CardioI1981;76:431-7. 27. VanderVeenFH, Vander VusseGJ, Kruger RTI, Vander Nager T, Willemsen P, Reneman RS. Metabolic and haemodynamic changes in the heart during the early phase of cardiopulmonary bypass. II. Animal experiments. Cardiovase Res 1989;33:472-7. 28. Pine MB,Bing OHL, Weintraub RM, Abelmann WHo Myocardial volume regulation after normothermic and hypothermic ischemic arrest in dogs. Am J Physiol 1981;240:HI16-26. 29. Schaper J, Scheid HH, Schmidt U, Hehrlein F. Ultrastructural study comparing the efficacy of five different methods of intraoperative myocardial protection in the human heart. J THORAC CARDIOVASC SURG 1986;92:4755. 30. Van der Veen FH, Van der Vusse GJ, Reneman RS. Myocardial blood flow and oxygen consumption after aortic crossclamping. J Surg Res 1989;47:319-24. 31. Flameng W, Borgers M, Daenen W, Stalpaert G. Ultrastructural and cytochemical correlates of myocardial protection by cardiac hypothermia in man. J THORAC CARDIOVASC SURG 1980;79:413-24.