- Email: [email protected]
Effect of initial reperfusion temperature on myocardial preservation
J THoRAc CARDIOVASC SURG 91:545-550, 1986
Effect of initial reperfusion temperature on myocardial preservation The effect of initial postischemic reperfusion temperature on myocardial preservation was studied in the isolated working rat heart model. After baseline measurement of aortic flow rate, coronary flow rate, and heart rate, 40 hearts were subjected to 60 minutes of ischemic arrest at 15° C induced with a single dose of cold potassium cardioplegic solution. Hearts were then revived with a 10 minute period of nonworking repetfuslonat 28°, 31°,34°,or 37° C (10 hearts each), followed by 5 minutes of nonworking reperf~ion at normothermia, foUowed by 30 minutes of working perf~ion. Repeat measurements of function were obtained and postischemic release of creatine kinase into coronary eflluent was determined. Recovery of aortic flow was significantlyreduced at lower initial reperfusion temperatures (75 % at 28° C verses 88 % at 37° C) and the effect was approximately linear throughout the range studied (p < 0.05). Release of creatine kinase into coronary eflluent was greater at lower initial reperf~ion temperatures (421 ImU jminj gm wet weight at 28° C versus 115 ImU jminj gm wet weight at 37° C), alsoin a linear manner (p < 0.05). In this model, initial postischemic hypothermic reperf~ion is deleterious to ceUular integrity and functional recovery of the preserved myocardium. Studies in higher animals and humans are warranted to further evaluate the effect of initial reperfusion temperature on myocardial preservation.
Mark T. Metzdorff, M.D., Gary L. Grunkemeier, Ph.D., and Albert Starr, M.D., Port/and. Ore.
In the continuing quest for improvement in the different aspects of perioperative myocardial preservation, a number of investigations have centered on the period of initial reperfusion after the ischemic insult. A recent excellent review by Rosenkranz and Buckberg' summarizes the current state of knowledge regarding reperfusion injury and the various manipulations that have been employed to minimize or reverse such damage. These techniques have included modifications in the composition of the initial reperfusion solution and variations in the temperature and mode of administration of the reperfusate. Studies on the effect of temperature during initial reperfusion have suggested that hypothermia may be deleterious to the functional recovery of myocardium. Lazar and colleagues- demonstrated that reperfusion
From the Division of Cardiopulmonary Surgery, Oregon Health Sciences University, Portland, Ore. Supported by a grant-in-aid from the Oregon Affiliate of the American Heart Association. Read at the Eleventh Annual Meeting of The Western Thoracic Surgical Association, Incline Village, Nev., June 16-20, 1985. Address for reprints: Mark T. Metzdorff, M.D., Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Rd., Portland, Ore. 97201.
545
with low-temperature blood after normothermic ischemic arrest failed to improve recovery compared with normothermic reperfusion. Rousou and co-authors' showed that low temperature reperfusion after hypothermic cardioplegic arrest resulted in a significant reduction in myocardial high-energy phosphate levels compared with normothermic reperfusion. The present study was designed to further delineate the effect of sub-normothermic initial reperfusion on functional myocardial recovery after hypothermic cardioplegic arrest. Methods Model. The model used is a modification of the isolated working rat heart described by Neely and associates." Male Sprague-Dawley rats, 250 to 350 gm, are anesthetized with diethyl ether and subjected to a median sternotomy. Heparin 200 IV is injected into the inferior vena cava. The chest contents are quickly removed en bloc and placed in iced Krebs-Hense1eit buffer (contents: NaCI 118 mmoljL, NaHC0 3 25 mmoljL, glucose 11.1 mmoljL, KCI 4.75 mmoljL, CaCI 2 • 2H 20 2.54 mmoljL, KH 2P04 1.19 mmoljL, and MgCI 2 • 6H 20 1.19 mmoIjL). The thymus is dissected free, the pulmonary hila are ligated, and incisions are made in the right atrium and pulmonary
The Journal of
546
MetzdorjJ, Grunkemeier, Starr
A
Thoracic and Cardiovascular Surgery
G
Fig. 1. Drawing of isolated rat heart apparatus used in experiments. A, Reperfusion heart chamber. B, Reperfusion aeration reservoir. C, Normothermic aeration reservoir. D, Normothermic heart chamber. E, Elasticity chamber. F, Atrial infusion reservoir. G, Hypothermic heart chamber. H, Cardioplegia reservoir.
artery to permit egress of coronary venous return, The aorta and left atrium are then cannulated and the heart is transferred to the experimental apparatus (Fig, 1), The apparatus allows both retrograde perfusion of the coronary arteries via the aorta under a pressure head of 70 ern H 20 (nonworking or Langendorff mode) and antegrade perfusion via the left atrium (filling pressure 20 mm H 20 ) with ejection of perfusate against the 70 em H 20 aortic column (working mode), Aortic flow and coronary return are collected and recycled, The entire system is placed in a water jacket and temperature is controlled, The perfusate is passed through a 5 Jlm polycarbonate filter and aerated with 95% oxygen and 5% carbon dioxide, which yields a pH of 7.40 at 370 C. A separate 150 C water-jacketed system is employed for administration of the cardioplegic solution and ischemic arrest. The cardioplegic solution was that used clinically at Oregon Health Sciences University (contents: NaCI 102,6 mmol/L, Na lactate 27.2 mmol/L, NaHCO) 25 mmoljL, KCl 24 mmol/L, glucose 11.1 rnmol/L, and CaCl 2 • 2H 20 1.4 mmol/L), and was administered at 70 em H 20 pressure. For the initial reperfusion phase, separate waterjacketed temperature-controlled aeration and heart
chambers were employed in parallel with the normothermic chambers. After initial reperfusion with buffer at the experimental temperature in the nonworking state, perfusion could be continued at normothermia by moving the heart to the normothermic system and redirecting the recycled buffer from the hypothermic to the normothermic aeration chamber, Measurements. Aortic flow rate (in milliliters per minute) was measured by timed collection into a graduate from the top of the aeration chamber 70 cm above the heart. Aortic flow rate was taken as the mean of three measurements. Coronary flow rate (in milliliters per minute) was determined from a timed collection during the last 5 minutes of the working phase. Heart rate (in beats per minute) was determined from chart recorder tracings obtained via a pressure transducer in the aortic outflow line. Creatine kinase (CK) release into the coronary venous effluent was determined spectrophotometrically according to the method of the Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology' with a commercially prepared kit (Sigma Chemical Co., St. Louis, Mo.). Coronary venous effluent was collected into an iced beaker, and all CK determinations were accomplished within 6 hours of
Volume 91 Number 4
Myocardial preservation
April. 1986
547
Table I. Results Percent recovery of cardiac function
Reperfusion temperature 1 C) 0
28 31 34 37
CK leakage Hearl rate 100.5 105.3 108.3 101.2
± 2.2 ± 2.7 ± 3.9 ± 4.4
Aorticflow rate 74.9 81.8 83.5 88.3
± 5.5 ± 5.6
Coronary flow rate 83.1 85.8 81.0 79.9
± 2.3 ± 1.4
± 1.3 ± 2.4 ± 2.5 ± 3.3
III1lU/l1lin/gl1l)
421 ± 178
234 ± 40 187 ± 66 115 ± 28
t.cgcnd: Results for recovery or cardiac function arc expressed as percent or pre-ischemic control value for each heart; results for all parameters arc given as mean ± standard deviation for 10 hearts. ImUjminjgm:= International milliunits per minute per gram wet weight. CK. Creatine kinase.
collection. At the completion of an experiment, the hearts were immediately removed from the apparatus, trimmed of excess tissue, blotted, and weighed. CK release is expressed in international milliunits per minute per gram of wet tissue weight. Experimental protocol. 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. A total of 40 hearts were used in this experiment. After being mounted on the perfusion apparatus, all hearts underwent a 30 minute recovery period of retrograde nonworking perfusion. They were then switched over to the working mode for 30 minutes, after which baseline measurements were taken of aortic flow rate, coronary flow rate, and heart rate. The hearts were then moved to the 15° C hypothermic chamber where arrest was rapidly induced with topical cooling and aortic root infusion of cardioplegic solution. A 2 minute period of cardioplegic infusion at a pressure of 70 em H 20 typically resulted in a delivery of 20 to 25 m1 of cardioplegic solution. After 60 minutes of ischemic cardioplegic arrest at 15° C, hearts were then revived in the reperfusion system with a 10 minute period of nonworking reperfusion at 28 0, 31 0, 34°, or 37° C (10 hearts at each temperature). All hearts then underwent a 5 minute period of normothermic nonworking perfusion followed by 30 minutes of working perfusion. Repeat measurements of aortic flow rate, coronary flow rate, and heart rate were then obtained. Coronary venous effluent was collected for the final 5 minutes of the postischemic working phase for determination of CK. Statistics. Percent recovery of control values for aortic flow rate, coronary flow rate, and heart rate were determined for each heart, and results for these values as well as for CK release are expressed as the mean ± the standard error of the mean for each group of 10 hearts.
Statistical analyses were performed by analysis of variance with a Hewlett-Packard 3000 computer and SPSS software (Medical Data Research Center, Portland, Ore.). Results With rewarming and reperfusion, all hearts spontaneously resumed beating, usually within a few seconds. Although cardiac electrical activity was not specifically monitored, reperfusion arrhythmias did not appear to be a problem in any of the experimental groups. Results for all parameters are presented in Table I. One-way analysis of variance with temperature as the independent variable showed significant reduction in postischemic recovery of aortic flow rate and significant increase in postischemic CK release with decreasing temperature of initial reperfusion (p < 0.05). No significant trends were evident in analysis of recovery of heart rate or coronary flow rate after ischemic insult. Results for percent recovery of aortic flow rate and for CK release are presented in graphic form in Figs. 2 and 3. Discussion That hypothermia is a critical component of optimal myocardial preservation has been fully established in past investigations, including studies from our institution.? Although the period of initial reperfusion is also a widely recognized factor in myocardial preservation, the role played by temperature in moderating or exacerbating reperfusion injury is less well defined. Lazar,' Rousou,' and their colleagues postulated a benefit to postischemic hypothermic reperfusion. They theorized that hypothermia during reperfusion might limit reperfusion injury by reducing myocardial oxygen demands and by preserving high-energy phosphate stores. These authors' experimental studies failed to support the hypothesis. In fact, the latter work revealed lower levels of high-energy phosphates in those hearts
The Journal of
548
Metzdorff, Grunkemeier, Starr
Thoracic and Cardiovascular Surgery
100 P ercen t Recovery 90
80 70
rr
IT
r:r:-
--=-
60
50 40 30 20 10
o
28
31
34
37
Reperfusion Temperature (Degrees Celsius)
Fig. 2. Percent recovery of aortic flow rate compared with preischemic control, after ischemia with reperfusion at indicated temperatures. Value at each temperature represents mean of 10 hearts. Standard error of the mean is indicated at top of each bar.
treated with a period of low-temperature (18° C) reperfusion as compared with hearts treated with normothermic reperfusion. Another study by Swanson and Myerowitz? used an isolated canine heart model to examine the role of reperfusion pressure and temperature on the functional and metabolic recovery of preserved hearts. They found increased tissue edema, decreased tissue high-energy phosphate levels, and reduced functional recovery in those hearts reperfused with blood at high pressure (80 mm Hg). Although temperature was a variable in their study, the experimental design did not allow a conclusion concerning its independent effect. The authors implicated the rate of tissue rewarming as a significant factor in reperfusion injury, but in their study the rate of rewarming was more a function of pressure than of temperature. The genesis of the present study was the clinical impression that hearts reperfused before complete patient rewarming seemed to recover function more slowly and less completely than those hearts reperfused at normothermia. Within the limits of the model used, the present study supports this observation. The temperature range chosen for scrutiny corresponds to the clinical range over which a patient would be rewarmed while being prepared to be removed from hypothermic cardiopulmonary bypass. Our results suggest that unclamping of the aorta before complete rewarming might indeed be detrimental to recovery of the myocar-
dium from the ischemic insult: It might be preferable to maintain hypothermic cardioplegic arrest until such time as reperfusion may proceed at normothermia. The mechanism through which reperfusion hypothermia might exert its detrimental effect is unknown. The studies of Rousou and co-authors indicate that reperfusion hypothermia may inhibit regeneration of or increase relative utilization of high-energy phosphate compounds necessary for cell function. Hypothermia shifts the oxygen-hemoglobin dissociation curve to the left, which reduces tissue oxygen delivery; this effect may be deleterious to myocardial recovery. Reperfusion hypothermia may also slow other metabolic and reparative processes required for improvement in cell function and integrity after ischemic insult. These might include ionic transport, reduction of cell swelling, and repair of damaged organelles and membranes. Caution must be observed in attempting to extrapolate results with the isolated rat heart model to the clinical setting. The preparation is not completely physiologic: Preload and afterload are fixed, normal hormonal and neurologic mechanisms are absent, and there are no other humoral factors impacting upon myocardial function such as are present in the intact animal. In addition, in the present study the relatively short time course of the experimental protocol leaves some room for question concerning the duration of the effects of hypothermia during initial reperfusion. It may be that functional recovery is not reduced, but just delayed.
Volume 91
Myocardial preservation
Number 4
549
April, 1986
mIU/minute/qram wet weiaht 500 r-'-'.;.;;;..,-,-,.;.;.;.;..;;;;.;.",",-",,",-~.;....;.;..;;..;;.....;.:.=.;;o.;.;.:...._----------_
450 400 350 300 250 200
[
150 100
rr
50
OL.-_ _....L..-='-'-_ ___L...,......... 28
~--L---..L..-.L---...J
31 34 Reperfusion Temperature
37
(Degrees Celsius)
Fig. 3. Release of creatine kinase into coronary venous effluent after ischemia with reperfusion at indicated temperatures. Yalue at each temperature represents mean of 10 hearts. Standard error of the mean is indicated at the top of each bar. mIU, International milliunits.
Nevertheless, the isolated rat heart model has been a reliable predictor of results in higher animals and humans, and it remains a valuable tool for preliminary investigation into mechanisms of myocardial injury and preservation. In summary, in this model, initial postischemic hypothermic reperfusion is deleterious to cellular integrity and functional recovery of the preserved myocardium. Additional studies in higher animals and humans are warranted to further evaluate the effect of initial reperfusion temperature on myocardial preservation. If our findings are borne out, the relatively simple maneuver of prolonging hypothermia cardioplegic arrest until rewarming is complete may make a significant, if modest contribution to myocardial preservation in certain clinical situations. We wish to express our appreciation to Mary Benowitz and David Malone for excellent technical assistance and to R. Mark Yetto, M.D., and Dennis Burger, Ph.D., for provision of laboratory space.
3 Rousou JH, Dobbs WA, Meeran MK, Engelman RM:
4
5
6
7
The temperature dependence of recovery of metabolic function following hypothermic potassium cardioplegic arrest. J THoRAc CARDIOVASC SURG 83:117-121,1982 Neely JR, Liebermeister N, Battersby EJ, Morgan HE: Effect of pressure development on oxygen consumption by the isolated rat heart. Am J Physiol 212:804-814, 1967 The Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology: Recommended method for the determination of creatine kinase in blood. Scand J Clin Lab Invest 36:711-723, 1976 Harlan BJ, Ross D, Macmanus Q, Knight R, Luber J, Starr A: Cardioplegic solutions for myocardial preservation. Analysis of hypothermic arrest, potassium arrest, and procaine arrest. Circulation 58:Suppl I: 114-118, 1978 Swanson DK, Myerowitz PD: Effect of reperfusion temperature and pressure on the functional and metabolic recovery of preserved hearts. J THoRAc CARDIOVASC SURG 86:242- 251, 1983
Discussion DR.BRADLEYJ.HARLAN Sacramento. Calif.
REFERENCES Rosenkranz ER, Buckberg GD: Myocardial protection during surgical coronary reperfusion. J Am cen Cardiol 1:1235-1246, 1983 2 Lazar HL, Buckberg GD, Manganaro A, Becker H, Mulder DG, Maloney JY: Limitations imposed by hypothermia during recovery from ischemia. Surg Forum 31:312-313,1980
Dr. Metzdorff concludes that initial postischemic hypothermic perfusion is deleterious to cellular integrity and functional recovery. I wonder whether his data entirely support this conclusion. The experimental design introduced another variable in addition to that of reperfusion temperature-the variable of the duration of normothermic, nonworking recovery. The hearts with the best postwork function, those revived at 37° C, had a total of 15 minutes of nonworking recovery at 37° C, whereas those with the worst postwork recovery, those
5 50 MetzdorjJ, Grunkemeier, Starr
reperfused at 28° C, had only 5 minutes of nonworking recovery at 37° C before their entering the working mode. The reversal of cardioplegic depression in this latter group may simply have been delayed rather than permanently affected. I would have similar reservations about concluding that the rate of CK-MB release as measured in this study, which might also be affected by different temperatures during recovery, is a definite indicator of cellular integrity. I wonder if Dr. Metzdorff might comment on these points. The principle that this study supports, that reperfusion as near normothermia as possible is desirable, can certainly be achieved in most instances. This study reemphasizes the importance of the period of reperfusion and recovery after cardioplegia and points toward further investigation in this important area.
DR. METZDORFF (Closing) I would like to thank Dr. Harlan for his careful analysis. He pioneered the use of this model at Oregon some 8 years ago, and it is through his efforts that I was able to accomplish this study.
The Journal of Thoracic and Cardiovascular Surgery
One of the nice things about the isolated rat heart model is that after various experimental manipulations the hearts reach a steady state very quickly and are quite stable thereafter. In our initial experiments at each reperfusion temperature, we did look at aortic flow rate with time and saw no change over the course of the second 15 minute working phase. Although it is possible that further time might have allowed improved functional recovery, we believe that in this model it would be unlikely. The CK data especially speak against this possibility. Other studies have shown that in this model the bulk of CK release occurs in the first 30 minutes of reperfusion. Because the apparatus is a closed system, we are measuring essentially all the CK released into the system from the moment of normothermic reperfusion. Enzyme release is expressed in terms of time to adjust for differences in coronary flow rate, but in our experiments these differences were slight. Thus we believe that our data demonstrate truly increased cellular disruption with lower initial reperfusion temperatures. That further time might allow improved functional recovery is certainly debatable. Even if this were so, delayed recovery would seem to be less desirable than the rapid, nearly complete, recovery seen with normothermic reperfusion.
Effect of initial reperfusion temperature on myocardial preservation The effect of initial postischemic reperfusion temperature on myocardial preservation was studied in the isolated working rat heart model. After baseline measurement of aortic flow rate, coronary flow rate, and heart rate, 40 hearts were subjected to 60 minutes of ischemic arrest at 15° C induced with a single dose of cold potassium cardioplegic solution. Hearts were then revived with a 10 minute period of nonworking repetfuslonat 28°, 31°,34°,or 37° C (10 hearts each), followed by 5 minutes of nonworking reperf~ion at normothermia, foUowed by 30 minutes of working perf~ion. Repeat measurements of function were obtained and postischemic release of creatine kinase into coronary eflluent was determined. Recovery of aortic flow was significantlyreduced at lower initial reperfusion temperatures (75 % at 28° C verses 88 % at 37° C) and the effect was approximately linear throughout the range studied (p < 0.05). Release of creatine kinase into coronary eflluent was greater at lower initial reperf~ion temperatures (421 ImU jminj gm wet weight at 28° C versus 115 ImU jminj gm wet weight at 37° C), alsoin a linear manner (p < 0.05). In this model, initial postischemic hypothermic reperf~ion is deleterious to ceUular integrity and functional recovery of the preserved myocardium. Studies in higher animals and humans are warranted to further evaluate the effect of initial reperfusion temperature on myocardial preservation.
Mark T. Metzdorff, M.D., Gary L. Grunkemeier, Ph.D., and Albert Starr, M.D., Port/and. Ore.
In the continuing quest for improvement in the different aspects of perioperative myocardial preservation, a number of investigations have centered on the period of initial reperfusion after the ischemic insult. A recent excellent review by Rosenkranz and Buckberg' summarizes the current state of knowledge regarding reperfusion injury and the various manipulations that have been employed to minimize or reverse such damage. These techniques have included modifications in the composition of the initial reperfusion solution and variations in the temperature and mode of administration of the reperfusate. Studies on the effect of temperature during initial reperfusion have suggested that hypothermia may be deleterious to the functional recovery of myocardium. Lazar and colleagues- demonstrated that reperfusion
From the Division of Cardiopulmonary Surgery, Oregon Health Sciences University, Portland, Ore. Supported by a grant-in-aid from the Oregon Affiliate of the American Heart Association. Read at the Eleventh Annual Meeting of The Western Thoracic Surgical Association, Incline Village, Nev., June 16-20, 1985. Address for reprints: Mark T. Metzdorff, M.D., Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Rd., Portland, Ore. 97201.
545
with low-temperature blood after normothermic ischemic arrest failed to improve recovery compared with normothermic reperfusion. Rousou and co-authors' showed that low temperature reperfusion after hypothermic cardioplegic arrest resulted in a significant reduction in myocardial high-energy phosphate levels compared with normothermic reperfusion. The present study was designed to further delineate the effect of sub-normothermic initial reperfusion on functional myocardial recovery after hypothermic cardioplegic arrest. Methods Model. The model used is a modification of the isolated working rat heart described by Neely and associates." Male Sprague-Dawley rats, 250 to 350 gm, are anesthetized with diethyl ether and subjected to a median sternotomy. Heparin 200 IV is injected into the inferior vena cava. The chest contents are quickly removed en bloc and placed in iced Krebs-Hense1eit buffer (contents: NaCI 118 mmoljL, NaHC0 3 25 mmoljL, glucose 11.1 mmoljL, KCI 4.75 mmoljL, CaCI 2 • 2H 20 2.54 mmoljL, KH 2P04 1.19 mmoljL, and MgCI 2 • 6H 20 1.19 mmoIjL). The thymus is dissected free, the pulmonary hila are ligated, and incisions are made in the right atrium and pulmonary
The Journal of
546
MetzdorjJ, Grunkemeier, Starr
A
Thoracic and Cardiovascular Surgery
G
Fig. 1. Drawing of isolated rat heart apparatus used in experiments. A, Reperfusion heart chamber. B, Reperfusion aeration reservoir. C, Normothermic aeration reservoir. D, Normothermic heart chamber. E, Elasticity chamber. F, Atrial infusion reservoir. G, Hypothermic heart chamber. H, Cardioplegia reservoir.
artery to permit egress of coronary venous return, The aorta and left atrium are then cannulated and the heart is transferred to the experimental apparatus (Fig, 1), The apparatus allows both retrograde perfusion of the coronary arteries via the aorta under a pressure head of 70 ern H 20 (nonworking or Langendorff mode) and antegrade perfusion via the left atrium (filling pressure 20 mm H 20 ) with ejection of perfusate against the 70 em H 20 aortic column (working mode), Aortic flow and coronary return are collected and recycled, The entire system is placed in a water jacket and temperature is controlled, The perfusate is passed through a 5 Jlm polycarbonate filter and aerated with 95% oxygen and 5% carbon dioxide, which yields a pH of 7.40 at 370 C. A separate 150 C water-jacketed system is employed for administration of the cardioplegic solution and ischemic arrest. The cardioplegic solution was that used clinically at Oregon Health Sciences University (contents: NaCI 102,6 mmol/L, Na lactate 27.2 mmol/L, NaHCO) 25 mmoljL, KCl 24 mmol/L, glucose 11.1 rnmol/L, and CaCl 2 • 2H 20 1.4 mmol/L), and was administered at 70 em H 20 pressure. For the initial reperfusion phase, separate waterjacketed temperature-controlled aeration and heart
chambers were employed in parallel with the normothermic chambers. After initial reperfusion with buffer at the experimental temperature in the nonworking state, perfusion could be continued at normothermia by moving the heart to the normothermic system and redirecting the recycled buffer from the hypothermic to the normothermic aeration chamber, Measurements. Aortic flow rate (in milliliters per minute) was measured by timed collection into a graduate from the top of the aeration chamber 70 cm above the heart. Aortic flow rate was taken as the mean of three measurements. Coronary flow rate (in milliliters per minute) was determined from a timed collection during the last 5 minutes of the working phase. Heart rate (in beats per minute) was determined from chart recorder tracings obtained via a pressure transducer in the aortic outflow line. Creatine kinase (CK) release into the coronary venous effluent was determined spectrophotometrically according to the method of the Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology' with a commercially prepared kit (Sigma Chemical Co., St. Louis, Mo.). Coronary venous effluent was collected into an iced beaker, and all CK determinations were accomplished within 6 hours of
Volume 91 Number 4
Myocardial preservation
April. 1986
547
Table I. Results Percent recovery of cardiac function
Reperfusion temperature 1 C) 0
28 31 34 37
CK leakage Hearl rate 100.5 105.3 108.3 101.2
± 2.2 ± 2.7 ± 3.9 ± 4.4
Aorticflow rate 74.9 81.8 83.5 88.3
± 5.5 ± 5.6
Coronary flow rate 83.1 85.8 81.0 79.9
± 2.3 ± 1.4
± 1.3 ± 2.4 ± 2.5 ± 3.3
III1lU/l1lin/gl1l)
421 ± 178
234 ± 40 187 ± 66 115 ± 28
t.cgcnd: Results for recovery or cardiac function arc expressed as percent or pre-ischemic control value for each heart; results for all parameters arc given as mean ± standard deviation for 10 hearts. ImUjminjgm:= International milliunits per minute per gram wet weight. CK. Creatine kinase.
collection. At the completion of an experiment, the hearts were immediately removed from the apparatus, trimmed of excess tissue, blotted, and weighed. CK release is expressed in international milliunits per minute per gram of wet tissue weight. Experimental protocol. 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. A total of 40 hearts were used in this experiment. After being mounted on the perfusion apparatus, all hearts underwent a 30 minute recovery period of retrograde nonworking perfusion. They were then switched over to the working mode for 30 minutes, after which baseline measurements were taken of aortic flow rate, coronary flow rate, and heart rate. The hearts were then moved to the 15° C hypothermic chamber where arrest was rapidly induced with topical cooling and aortic root infusion of cardioplegic solution. A 2 minute period of cardioplegic infusion at a pressure of 70 em H 20 typically resulted in a delivery of 20 to 25 m1 of cardioplegic solution. After 60 minutes of ischemic cardioplegic arrest at 15° C, hearts were then revived in the reperfusion system with a 10 minute period of nonworking reperfusion at 28 0, 31 0, 34°, or 37° C (10 hearts at each temperature). All hearts then underwent a 5 minute period of normothermic nonworking perfusion followed by 30 minutes of working perfusion. Repeat measurements of aortic flow rate, coronary flow rate, and heart rate were then obtained. Coronary venous effluent was collected for the final 5 minutes of the postischemic working phase for determination of CK. Statistics. Percent recovery of control values for aortic flow rate, coronary flow rate, and heart rate were determined for each heart, and results for these values as well as for CK release are expressed as the mean ± the standard error of the mean for each group of 10 hearts.
Statistical analyses were performed by analysis of variance with a Hewlett-Packard 3000 computer and SPSS software (Medical Data Research Center, Portland, Ore.). Results With rewarming and reperfusion, all hearts spontaneously resumed beating, usually within a few seconds. Although cardiac electrical activity was not specifically monitored, reperfusion arrhythmias did not appear to be a problem in any of the experimental groups. Results for all parameters are presented in Table I. One-way analysis of variance with temperature as the independent variable showed significant reduction in postischemic recovery of aortic flow rate and significant increase in postischemic CK release with decreasing temperature of initial reperfusion (p < 0.05). No significant trends were evident in analysis of recovery of heart rate or coronary flow rate after ischemic insult. Results for percent recovery of aortic flow rate and for CK release are presented in graphic form in Figs. 2 and 3. Discussion That hypothermia is a critical component of optimal myocardial preservation has been fully established in past investigations, including studies from our institution.? Although the period of initial reperfusion is also a widely recognized factor in myocardial preservation, the role played by temperature in moderating or exacerbating reperfusion injury is less well defined. Lazar,' Rousou,' and their colleagues postulated a benefit to postischemic hypothermic reperfusion. They theorized that hypothermia during reperfusion might limit reperfusion injury by reducing myocardial oxygen demands and by preserving high-energy phosphate stores. These authors' experimental studies failed to support the hypothesis. In fact, the latter work revealed lower levels of high-energy phosphates in those hearts
The Journal of
548
Metzdorff, Grunkemeier, Starr
Thoracic and Cardiovascular Surgery
100 P ercen t Recovery 90
80 70
rr
IT
r:r:-
--=-
60
50 40 30 20 10
o
28
31
34
37
Reperfusion Temperature (Degrees Celsius)
Fig. 2. Percent recovery of aortic flow rate compared with preischemic control, after ischemia with reperfusion at indicated temperatures. Value at each temperature represents mean of 10 hearts. Standard error of the mean is indicated at top of each bar.
treated with a period of low-temperature (18° C) reperfusion as compared with hearts treated with normothermic reperfusion. Another study by Swanson and Myerowitz? used an isolated canine heart model to examine the role of reperfusion pressure and temperature on the functional and metabolic recovery of preserved hearts. They found increased tissue edema, decreased tissue high-energy phosphate levels, and reduced functional recovery in those hearts reperfused with blood at high pressure (80 mm Hg). Although temperature was a variable in their study, the experimental design did not allow a conclusion concerning its independent effect. The authors implicated the rate of tissue rewarming as a significant factor in reperfusion injury, but in their study the rate of rewarming was more a function of pressure than of temperature. The genesis of the present study was the clinical impression that hearts reperfused before complete patient rewarming seemed to recover function more slowly and less completely than those hearts reperfused at normothermia. Within the limits of the model used, the present study supports this observation. The temperature range chosen for scrutiny corresponds to the clinical range over which a patient would be rewarmed while being prepared to be removed from hypothermic cardiopulmonary bypass. Our results suggest that unclamping of the aorta before complete rewarming might indeed be detrimental to recovery of the myocar-
dium from the ischemic insult: It might be preferable to maintain hypothermic cardioplegic arrest until such time as reperfusion may proceed at normothermia. The mechanism through which reperfusion hypothermia might exert its detrimental effect is unknown. The studies of Rousou and co-authors indicate that reperfusion hypothermia may inhibit regeneration of or increase relative utilization of high-energy phosphate compounds necessary for cell function. Hypothermia shifts the oxygen-hemoglobin dissociation curve to the left, which reduces tissue oxygen delivery; this effect may be deleterious to myocardial recovery. Reperfusion hypothermia may also slow other metabolic and reparative processes required for improvement in cell function and integrity after ischemic insult. These might include ionic transport, reduction of cell swelling, and repair of damaged organelles and membranes. Caution must be observed in attempting to extrapolate results with the isolated rat heart model to the clinical setting. The preparation is not completely physiologic: Preload and afterload are fixed, normal hormonal and neurologic mechanisms are absent, and there are no other humoral factors impacting upon myocardial function such as are present in the intact animal. In addition, in the present study the relatively short time course of the experimental protocol leaves some room for question concerning the duration of the effects of hypothermia during initial reperfusion. It may be that functional recovery is not reduced, but just delayed.
Volume 91
Myocardial preservation
Number 4
549
April, 1986
mIU/minute/qram wet weiaht 500 r-'-'.;.;;;..,-,-,.;.;.;.;..;;;;.;.",",-",,",-~.;....;.;..;;..;;.....;.:.=.;;o.;.;.:...._----------_
450 400 350 300 250 200
[
150 100
rr
50
OL.-_ _....L..-='-'-_ ___L...,......... 28
~--L---..L..-.L---...J
31 34 Reperfusion Temperature
37
(Degrees Celsius)
Fig. 3. Release of creatine kinase into coronary venous effluent after ischemia with reperfusion at indicated temperatures. Yalue at each temperature represents mean of 10 hearts. Standard error of the mean is indicated at the top of each bar. mIU, International milliunits.
Nevertheless, the isolated rat heart model has been a reliable predictor of results in higher animals and humans, and it remains a valuable tool for preliminary investigation into mechanisms of myocardial injury and preservation. In summary, in this model, initial postischemic hypothermic reperfusion is deleterious to cellular integrity and functional recovery of the preserved myocardium. Additional studies in higher animals and humans are warranted to further evaluate the effect of initial reperfusion temperature on myocardial preservation. If our findings are borne out, the relatively simple maneuver of prolonging hypothermia cardioplegic arrest until rewarming is complete may make a significant, if modest contribution to myocardial preservation in certain clinical situations. We wish to express our appreciation to Mary Benowitz and David Malone for excellent technical assistance and to R. Mark Yetto, M.D., and Dennis Burger, Ph.D., for provision of laboratory space.
3 Rousou JH, Dobbs WA, Meeran MK, Engelman RM:
4
5
6
7
The temperature dependence of recovery of metabolic function following hypothermic potassium cardioplegic arrest. J THoRAc CARDIOVASC SURG 83:117-121,1982 Neely JR, Liebermeister N, Battersby EJ, Morgan HE: Effect of pressure development on oxygen consumption by the isolated rat heart. Am J Physiol 212:804-814, 1967 The Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology: Recommended method for the determination of creatine kinase in blood. Scand J Clin Lab Invest 36:711-723, 1976 Harlan BJ, Ross D, Macmanus Q, Knight R, Luber J, Starr A: Cardioplegic solutions for myocardial preservation. Analysis of hypothermic arrest, potassium arrest, and procaine arrest. Circulation 58:Suppl I: 114-118, 1978 Swanson DK, Myerowitz PD: Effect of reperfusion temperature and pressure on the functional and metabolic recovery of preserved hearts. J THoRAc CARDIOVASC SURG 86:242- 251, 1983
Discussion DR.BRADLEYJ.HARLAN Sacramento. Calif.
REFERENCES Rosenkranz ER, Buckberg GD: Myocardial protection during surgical coronary reperfusion. J Am cen Cardiol 1:1235-1246, 1983 2 Lazar HL, Buckberg GD, Manganaro A, Becker H, Mulder DG, Maloney JY: Limitations imposed by hypothermia during recovery from ischemia. Surg Forum 31:312-313,1980
Dr. Metzdorff concludes that initial postischemic hypothermic perfusion is deleterious to cellular integrity and functional recovery. I wonder whether his data entirely support this conclusion. The experimental design introduced another variable in addition to that of reperfusion temperature-the variable of the duration of normothermic, nonworking recovery. The hearts with the best postwork function, those revived at 37° C, had a total of 15 minutes of nonworking recovery at 37° C, whereas those with the worst postwork recovery, those
5 50 MetzdorjJ, Grunkemeier, Starr
reperfused at 28° C, had only 5 minutes of nonworking recovery at 37° C before their entering the working mode. The reversal of cardioplegic depression in this latter group may simply have been delayed rather than permanently affected. I would have similar reservations about concluding that the rate of CK-MB release as measured in this study, which might also be affected by different temperatures during recovery, is a definite indicator of cellular integrity. I wonder if Dr. Metzdorff might comment on these points. The principle that this study supports, that reperfusion as near normothermia as possible is desirable, can certainly be achieved in most instances. This study reemphasizes the importance of the period of reperfusion and recovery after cardioplegia and points toward further investigation in this important area.
DR. METZDORFF (Closing) I would like to thank Dr. Harlan for his careful analysis. He pioneered the use of this model at Oregon some 8 years ago, and it is through his efforts that I was able to accomplish this study.
The Journal of Thoracic and Cardiovascular Surgery
One of the nice things about the isolated rat heart model is that after various experimental manipulations the hearts reach a steady state very quickly and are quite stable thereafter. In our initial experiments at each reperfusion temperature, we did look at aortic flow rate with time and saw no change over the course of the second 15 minute working phase. Although it is possible that further time might have allowed improved functional recovery, we believe that in this model it would be unlikely. The CK data especially speak against this possibility. Other studies have shown that in this model the bulk of CK release occurs in the first 30 minutes of reperfusion. Because the apparatus is a closed system, we are measuring essentially all the CK released into the system from the moment of normothermic reperfusion. Enzyme release is expressed in terms of time to adjust for differences in coronary flow rate, but in our experiments these differences were slight. Thus we believe that our data demonstrate truly increased cellular disruption with lower initial reperfusion temperatures. That further time might allow improved functional recovery is certainly debatable. Even if this were so, delayed recovery would seem to be less desirable than the rapid, nearly complete, recovery seen with normothermic reperfusion.