Superior protective effect of low-calcium, magnesium-free potassium cardioplegic solution on ischemic myocardium

J

THoRAc CARDIOVASC SURG

1991;101:695-702

Superior protective effect of low-calcium, magnesium-free potassium cardioplegic solution on ischemic myocardium Clinical study in comparison with St. Thomas' Hospital solution The protective effect of low-calcium, magnesiem-free potassium cardioplegic solution on ischemic myocardium has been assessed in adult patients undergoing heart operations. Postreperfnsion recovery

of cardiac function and electrical activity was evaluated in 34 patients; 16 received low-calcium, magnesium-free potassium cardioplegic solution (group I) and 18 received St. Thomas' Hospital solution, which is enriched with calcium and magnesium (group II). There were no significant differences between the two groups in age, sex, body weight, and New York Heart Association functional class. Aortic occlusion time (107.3 ± 46.8 minutes verses 113.6 ± 44.3 minutes), highest myocardial temperature during elective global ishemia (11.50 C ± 3.10 C versns 9.30 C ± 3.20 C), and total volume of cardioplegic solution (44.2 ± 20.5 mljkg versus 43.4 ± 17.6 mljkg) were also similar in the two groups. On reperfusion, electrical defibrillation was required in four cases (25.5%) in group I and in 15 cases (83.3%) in group II (p < 0.005), and bradyarrhythmias were significantly more prevalent in group II (6.3% versus 44.4%; p < 0.05). Serum creatine kinase MB activity at 15 minutes of reperfusion (12.3 ± 17.0 lUlL versus 42.6 ± 46.1 lUlL; P < 0.05) and the dose of dopamine or dobuta~ required during the early phase of reperfusion (1.8 ± 2.5 ~gjkgjmin versus 6.1 ± 3.3 ~g/kg/min; p < 0.0002) were both significantly greater in group II. Postischemic left ventricular function, as assessed by percent recovery of the left ventricular end-syst06c pressure-volume relationship in patients who underwent aortic valve replacement alone, was significantly better in group I (160.4% ± 45.5% versus 47.8% ± 12.9%; p < 0.05). Serum level of calcium and magnesium ions was significantly lower in group I. Thus low-calcium, magnesium-free potassium cardioplegic solution provided exceUent protection of the ischemic heart, whereas St. Thomas' Hospital solution with calcium and magnesium enabled relatively poor functional and electrical recovery of the. heart during the early reperfnsion period. These results might be related to differing levels of extraceUular calcium and magnesium on reperfnsion.

Kazuhiko Kinoshita, MD, Masahiro Oe, MD, and Kouichi Tokunaga, MD, Fukuoka, Japan

Rassium cardioplegicsolution has proved beneficial in cardiac surgery. The optimal cardioplegic solution is designed to diminishdamage to the myocardium induced Fromthe Division of Cardiovascular Surgery, Institute of Angiocardiology, Kyushu University, Faculty of Medicine, Fukuoka, Japan. Received for publication Aug. 14, 1989. Accepted for publication March I, 1990. Address for reprints: Dr. KazuhikoKinoshita, MD, Division of Cardiovascular Surgery,Instituteof Angiocardiology, KyushuUniversity, Faculty of Medicine, 3-1-1 Maidashi, Higashiku, Fukuoka 812, Japan. 12/1/21160

by ischemia and reperfusion.Such damage is commonly known to be associated with calcium overload.': 2 Therefore calcium ions should be avoided in cardioplegic solutions, and the concentration of the other ions should also be designed to alleviate intracellular calcium accumulation, To this end we have developeda low-calciumpotassium cardioplegicsolution. Our experimental data.' have shown that the maximum recoveryof electromechanical activity after hypoxiccardioplegicarrest is obtained with a calcium-freepotassiumcardioplegicsolutioncontaining some (around 60 mmoljL) sodium. In contrast, Hearse, Braimbridge, and Jynge" have 695

696

The Journal of Thoracic and Cardiovascular

Kinoshita, Oe, Tokunaga

Surgery

Table I. Composition ofcardioplegic solutions Group I

Na (mmoljL) K (mmoljL) Ca (mmol/L) Mg (mmol/L) CI (mmoljL) HC03 (rnmol/L) Glucose (gm/L) Osmolarity (mOsm/L) pH Oxygenation

87 20 0.1

o

97 10 25 353 7.8 Yes

Table

n. Patient information

Group II

120 16 1.2 16 160 10

o

370 7.8 Yes

pointed out the importance of including calcium (1.2 rnrnoljL) in the potassium cardioplegicsolutionand have formulated the 81. Thomas' Hospital solution.The main reason for adding calcium to the solution was to prevent the calcium paradox.S 6 However, this phenomenon can be attenuated by the adjustment of sodium- 7 and the use oflow temperatures," 81. Thomas' Hospital solutionalso includes high levels(16 rnrnoljL) of magnesium, because this divalent cation is important as an enzyme cofactor, is associated with a part of adenosine triphosphate (ATP), and can inhibit calcium inflUX. 9,IO However, hypermagnesemia causes detrimental arrhythmias, such as atrioventricular conduction block and severe bradycardia. II We therefore doubted that magnesium was necessary in cardioplegicsolutions,particularly in solutionswith little or no calcium. Thus the purpose of this study is to clinically compare the protective effect of low-calcium, magnesium-free potassium cardioplegic solution to that of a solution containing high levels of both calcium and magnesium (81. Thomas' Hospital solution) on the ischemic heart. Methods Patients. Thirty-four consecutive adult patients who underwent coronary bypass or valve operations in the Kyushu University Hospital were randomly divided into two groups according to the cardioplegic solution used during their operations. Emergency cases and patients with congenital heart disease were excluded from the study. The surgeon was not informed of the composition of cardioplegic solution until the operation was over. Thus the study was designed in a prospectively randomized way. Cardioplegic techniques. All operations were performed with standard cardiopulmonary bypass technique and moderate hypothermia (25 0 C). Cardiopulmonary bypass was established with two venous cannulas through the right atrium and cannulation of the ascending aorta. After the ascending aorta was clamped, cold cardioplegic solution (l0 ml/kg body weight) was delivered into the aortic root and an aortic pressure of over 50 mm Hg was maintained. Where there was aortic regurgitation, the coronary orifices were perfused directly. Additional car-

Group I

Group II

p Value

Patients(No.) Age (yr) Sex ratio (M:F) Body weight (kg) Diagnosis ratio (IHD:VHD) IHD 2-Vessel 3-Vessel LMT VHD ASR AR MSR MS MR ASR+MSR AAE

16 54.1 ± 13.4 12:4 53.6 ± 8.2 5:11

18 52.5 ± 7.8 13:5 58.2 ± I I.7 6:12

NS NS NS NS

2 2

2 4 0

4 2 1 0 2 1

2 3 0 3 1 2 1

NYHA CTR (%) EF (%) AF(%)

2.5 ± 0.5 55.2 ± 7.9 57.9 ± 15.9 43.8

2.3 ± 0.8 54.6 ± 11.8 55.2 ± 1I.7 12.5

I I

NS NS NS NS

NS, Not significant; M:F, male:female; IHD, ischemicheart disease; VHD. valvular heart disease; LMT, left main trunk disease; ASR, aortic stenosis and regurgitation;AR, aortic regurgitation;MSR, mitral stenosisand regurgitation; MS, mitral stenosis; MR, mitral regurgitation; AAE, annuloaortic ectasia; NYHA, New York Heart Association functionclass;CTR, cardiothdracicratio on preoperativechest x-ray film;EF, left ventricularejectionfractionon preoperative left ventricular angiography;AF, atrial fibrillation.

dioplegic solution (5 ml/kg body weight) was administered every 20 minutes during aortic clamping, and myocardial protection was also obtained by topical cooling with ice slush in the pericardial cradle. Temperature of the ventricular septum was monitored with a thermistor-tipped probe throughout the cardioplegic arrest. During cardiopulmonary bypass a hemoconcentrator was used for fluid management and control of electrolytes and hematocrit. After the aorta was unclamped, systemic rewarming was commenced gradually while the systemic circulation was partially assisted with cardiopulmonary bypass. When postreperfusion ventricular fibrillation occurred and did not revert, electrical defibrillation was used. Any reduced level of serum calcium ion was not corrected for 30 minutes of reperfusion. Calcium was then given just before discontinuation of cardiopulmonary bypass. At that time atrial or ventricular pacing (or both) was initiated if the heart rate was less than 50 beats/ min. Composition of cardioplegic solution. The composition of cardioplegic solution used for each group is shown in Table I. Both cardioplegic solutions were oxygenated and cooled (4 0 C) before delivery. Infused cardioplegic fluid was discarded from the right atrium in group II to avoid excess magnesium in the perfusate, whereas it was recirculated in the cardiopulmonary bypass circuit in group I. The most striking differences in the formulation of the two solutions are (1) calcium concentration (0.1 versus 1.2 mmol/L), (2) magnesium concentration (0 ver-

Volume 101 Number 4 April 1991

Low-calcium cardioplegic solution

Table III. Operative information Operation CABG AVR MVR OMC MAP AVR + MVR Bentall CPB time (min) AoX time (min) Myocardial temperature Minimum (0 C) Maximum (0 C) Minimum rectal temperature (0 C) Cardioplegic solution (rnl/kg)

200

Group I

Group II

5

6

5 2

5

I

1 2

0 2

2

I

I

183.7 ± 61.6 113.6 ± 44.3

o

Group I • Group II

p Value

150 :i <,

* P
:::;)

0

III :E 100 I

I

164.2 ± 47.4 107.3 ± 46.8

697

~

0.

0

50

NS NS

o 3.3 ± 2.1 11.5±3.1 25.0 ± 2.9

2.6 ± 1.7 9.3 ± 3.2 26.0 ± 1.5

NS NS NS

44.2 ± 20.5

43.4 ± 17.6

NS

CABG. Coronary artery bypass grafting; AVR, aortic valve replacement; MVR, mitral valve replacement; OMC, open mitral commissurotomy; MAP, mitral annuloplasty: NS, not significant; CPB, cardiopulmonary bypass; AoX, aortic crossclarnp: Myocardial temperature, temperature of ventricular septum measured with a needle thermistor during cardioplegic arrest.

sus 16 mrnol/L), and (3) glucose concentration (25 versus 0 gm/ L). The concentration of the other ions (sodium, potassium, chlo~lde, bicarbonate), as well as osmolarity and pH, are sirnilar In the two groups. . Assessment of results. The activity of the myocardial Isozymeof creatine kinase (CK-MB) was measured 15 minutes a.fter reperfusion and on the first, second, and third postoperative days for evaluation of ischemia-induced myocardial damage. Postreperfusion arrhythmia, defined as severe bradycardia (heart rat~ <50 beats/min) caused by advanced or complete atrioventricular conduction block or sinus dysfunction was analyzed by continuous electrocardiographic monitoring. These ~hythm disturbances necessitated postoperative electrical pacmg support. A diagnosis of perioperative myocardial infarction was established by postoperative development of a new Q wave and an elevation of serum CK-MB levels in excess of 50 IV /L. The case ofCK- MB elevation alone, in the absence of a new Q wave, was defined as "suspected" perioperative myocardial infarction. Postischemic cardiac function was assessed by the amount of dopamine or dobutamine used and the left ventricular end-systolic pressure-volume relationship (Emax). The dose of dopamine or dobutamine was expressed as an average of the amount (micrograms per kilogram per minute) dripped intravenously for 12 hours from the onset of myocardial reperfusion. The Emax has been shown to be a reliable load-independent index of left ventricular contractility. 12 For evaluation of the protective effect of cardioplegia on ischemic myocardium, Emax was measured and compared before and after operation in patients undergoing isolated aortic valve replacement alone in each group (n = 5 in each). The Emax study was limited to these patients to minimize the diversity of the results. There were no differences in preoperative values of left ventricular ejection

pre.

rep.15mln 1POD

2POD

3POD

Fig. 1. The activity of serum CK- MB after unclamping of the aorta. CK-MB activities levels were obtained before operation, at 15 minutes of reperfusion, and on the first, second, and third postoperative days in group I (0) and group II (e). Values are mean ± standard deviation. *p < 0.05. POD, Postoperative day. fraction (60.6% ± 13.8% versus 58.9% ± 12.0%, NS*) left ventricular end-diastolic volume (148.8 ± 58.2 ml/rn? v~rsus 135.6 ± 53.8 ml/rn'', NS), and presence of left ventricular hypertrophy on electrocardiogram (5/5 versus 5/5, NS) between these two subgroups. Moreover, the type of aortic disease was also similar, as shown in Table II. An electromagnetic flow probe (Nihon Kohden Ltd., Japan) was placed around the ascending aorta and a catheter-tip micromanometer was inserted into the left ventricular cavity, Thus aortic flow and left ventricular press~re were ~ontinuously and simultaneously recorded. The aortic flow signal was time-integrated to determine instantaneous ventricular ejecting volume and thus the pressurevolume loop of left ventricular ejection was obtained. I 3 The isovolumic peak pressure (Pmax) was estimated by means of the single ejecting beat method reported by Sunagawa and associates.!" Thus the left ventricular end-systolic pressure-volume line can be drawn from Pmax to the left upper corner of the loop without changes of contractile state and measurement of absolute volume of the left ventricle. Emax was determined as a slope of this line. The recovery of Emax after cardioplegic arrest was expressed as a percent of a preischemic control value. Statistical analysis. All values obtained were expressed as a ~ean ± the standard deviation of the mean. The significance of differences for these results was determined by the unpaired Stud~nt's t t~st. The prevalences of variables were analyzed by the X test With the Yates correction. A p value less than 0.05 was considered to be statistically significant.

Results Clinical information on the patients is summarized in Table II. There were no significant differences in preoperative characteristics between the patients in the two *NS = No significant difference.

~100 c 0

;: ;: == .a ;:

•

Q

..

'i u

"i:

-•

,-----, **

~

80

80 60

J

40

•

iii 20 0

0

40 20 0

E .c

>.c

-• lit

>-

0

u

c

•

'0

U .5

Fig. 2. Prevalence andseverity ofrhythm disturbances during

the early period of reperfusion. Use of electrical defibrillation wasdefined as the percentage of patients for whom postreperfusion ventricular fibrillation or tachycardia didnotabatespontaneously. Arrhythmias were defined as severe and uncontrollable bradyarrhythmias, suchascomplete atrioventricular block or sinus bradycardia necessitating postoperative electrical pacing. *p < 0.05; **p < 0.005.

* 4.0

i !

E

o Group I • Group II

3.0

:::l

iii c

•m III

*

2.0

* p<0.0001

:IE

.. en•

E

:::l

1.0

0

*

*

pre-CPB rep. 1POD 30mln

2POD

200

**

,-------, 10.0

III

Q

II

II

:::)

60

....

*

,-----,

100 ....

,-------, *

U

lit

The Journal of Thoracic and Cardiovascular Surgery

Kinoshita, Oe, Tokunaga

69 8

3POD

Fig. 3. Changes in serum magnesium level before operation and after reperfusion in groups I and II. Serum magnesium concentration was obtained before cardiopulmonary bypass (CPR), at 30 minutes ofreperfusion, andonthefirst, second, and thirdpostoperative daysinbothgroups. Values areshown as the mean ± standard deviation in each group. *p < 0.0001 between groups or compared with pre-CPB value within each group.

groups. In patients undergoing coronary bypass,the mean number of grafts was 3.0 ± 0.7 in group I and 3.0 ± 0.9 in group II (NS). As shown in Table III, operative parameters were similar in the two groups. During cardioplegic arrest, the lowest myocardial temperature was 3.3 0 ± 2.1 0 and 2.6 0 ± 1.7 0 C (NS) and the high-

lC III

8.0

-..

E 150

W

C

E <,

Ol

..II: <,

Cll

6.0

.:

0

4.0

0


2.0 a. O

0

>-

•> 100 u • 50 a:

III Q

0

Q

0

II

II

0

Fig. 4. Postischemic recovery of cardiac function in groups I and II. Recovery ofEmaxwas assessed inpatients ofeach group who underwent isolated aortic valve replacement (n = 5 ineach group) and was expressed asa percentage of preischemic value. Thedose ofdopamine (DO?) ordobutamine (DOR) was defined as the average amounts (j.Lgjkgjmin) given intravenously for the first 12 hours of reperfusion in all cases. All values are mean ± standarddeviation. *p < 0.05, **p < 0.0002 between the twogroups.

est was 11.5 0 ± 3.1 0 and 9.3 0 ± 3.2 0 C (NS) in groups I and group II, respectively. Myocardial enzyme and perioperative myocardial infarction. The postoperative activity of CK-MB is presented in Fig. 1. The preoperative control value of the enzyme was nil in all cases. At 15 minutes of reperfusion, the CK-MB value was significantly higher in group II than in group I (12.3 ± 17.0 IV jL in group I and 42.6 ± 46.1 IV jL in group II, p < 0.05). There was also a slight elevation of the enzyme in group II on the first postoperative day (20.2 ± 23.7 IVjL in group I and 88.4 ± 136.2 IV jL in group II). Thereafter, enzyme levels were almost normal in both groups. Perioperative myocardial infarction is one of the most important complications of cardiac operations. In this study perioperative myocardial infarction was not prevalent in either groups, although one patient in group II had a new Q wave and a significant elevation of the CK- MB value (Table IV). In this patient the maximum values of postreperfusion magnesium and CK were extraordinarily high (serum M g2+ 4.84 mmoljL and CK-MB 708 IV jL). She had severe sinus dysfunction postoperatively and, after the sternotomy was closed, required cardiac massage by the chest compression technique because of sudden cardiac arrest despite pacing being attempted. Additional atrial and ventricular pacing electrodes were implanted and atrioventricular sequential pacing was initiated. With these procedures, the blood pressure and cardiac output increased. Even when this atypical case

Volume 101 Number 4

Low-calcium cardioplegic solution 6 9 9

April 1991

Table IV. Prevalence ofperioperative myocardial

infarction

PMI suspected* (%) PMlt(%)

Group I

Group /I

p Value

12.5

16.7

NS NS

o

5.6

PM I. Pcriopcrative myocardial infarction; NS. not significant. 'CK-MB> 50 IUjL without new Q wave.

Table V. Serum calcium and magnesium levels at 30

minutesofreperfusion Serum calcium (mmol/L) Serum magnesium (mrnol/L)

Group I

Group /I

p Value

0.89 ± 0.06

1.00 ± 0.08

<0.001

1.31 ± 0.16

3.42 ± 1.01

<0.0001

+CK-MB> 50 IUjL with new Q wave.

was excluded from the analysis, CK-MB activity at 15 minutes of reperfusion was still significantly higher in group II than in group I (12.3 ± 17.0 IV jL in group I and 35.1 ± 34.3 IV jL in group II; p < 0.05). Postreperfusion rhythm disturbances. After unclamping of the aorta, as shown in Fig. 2, electrical defibrillation was applied in four patients (25.0%) in group I and 15 (83.3%) in group II (p < 0.005). The greater prevalence of ventricular fibrillation in group II may indicate poor recovery of the electrical activity of the myocardium during the early period of reperfusion in these patients. In addition, the return of normal cardiac rhythm was delayed after the onset of reperfusion in patients who receivedSt. Thomas' Hospital solution (Fig. 2). The prevalence of bradyarrhythmias, such as complete atrioventricular block or sinus bradycardia, was 6.3% in group I and 44.4% in group II (p < 0.05). Most of these patients had an adequate heart rate after the first postoperative day when serum Mg2+ concentration returned to normal levels (Fig. 3). Recovery of cardiac function. Postischemic cardiac function, as assessed by dosages of catecholamines and recovery of Emax, was superior in group I (Fig. 4). The average amount of dopamine or dobutamine given intravenously for 12 hours after reperfusion of the heart was 1.8 ± 2.5 JLgjkgjmin in group I and 6.1 ± 3.3 JLgj kgjmin in group II (p < 0.0002). Thus St. Thomas' Hospital solution might afford less protection to the ischemic heart. This possibility is also supported by the results of the Emax study. The latter is thought to be a suitable index of left ventricular contractility. Postoperativerecoveryof Emax was analyzed in a subset of patients in both groups (n = 5 in each) who underwent isolated aortic valve replacement. Recovery of Emax, defined as a percent of preischemic control value, was 160.4% ± 45.5%in group I and 47.8% ± 12.9%ingroup II (p < 0.05). These results suggest that high-potassium cardioplegic solution including less calcium and magnesium better protects postischemic cardiac performance. Postreperfusion Ca2+ and Mg2+ levels. Postreperfusion serum concentrations of Ca 2+ and Mg 2 + were measured and compared 30 minutes after unclamping of the

aorta (Table V). In group I, these cations decreased below standard levels because of hemodilution during cardiopulmonary bypass. Although in group II infused cardioplegic solution was discarded from the cardiopulmonary bypass circuit, both Ca 2+ and Mg 2+ levels were significantly higher than those in group I (Ca 2+ 0.89 ± 0.06 versus 1.00 ± 0.08 mmoljL, p < 0.001; Mg2+ 1.31 ± 0.16 versus 3.42 ± 1.01 mmoljL,p < 0.0001). In addition, as shown in Fig. 3, the serum Mg2+ concentration in group II at 30 minutes ofreperfusion was approximately two times that of the preischemic control value. On the first postoperative day, magnesium level had normalized in group II but it was still lower than the control value in group I (1.37 ± 0.17 mmoljL on the first postoperative day and 1.80 ± 0.23 mmoljL before cardiopulmonary bypass, p < 0.0001). Discussion The present study suggests that reduction or omission of divalent cations such as calcium and magnesium in high-potassium cardioplegic solution provided excellent myocardial protection clinically. On the other hand, St. Thomas' Hospital solution, which has been accepted as a useful cardioplegic solution in recent years, exerted less beneficial effects on postischemic recovery of cardiac function and electrical activity in our practice. One possible explanation for the different capabilities of the two cardioplegic solutions is thought to relate to the amount of calcium and magnesium in the solutions. Therefore it is pertinent to discuss the role of these cations in potassium cardioplegic solution. Calcium and cardioplegia. Ischemia-induced myocardial injury is associated with intracellular calcium accumulation. I, 2 Enhancement of intracellular calcium availability reduces ATP stores during ischemia by stimulating ATP hydrolysis of myofilaments.J5 These biochemical changes may produce an irreversible ischemic contracture, "stone heart."16, 17 Thus it may be argued that calcium should not be included in cardioplegic solution. In support of this point, Boggs and coworkers'" demonstrated that calcium in cardioplegic solution reduced myocardial ATP levels and recovery of cardiac function of the rat heart in a dose-dependent fashion.

700 Kinoshita, Oe, Tokunaga

Using similar experimental models, Torchiana and colleagues'? also indicated that high-energy phosphate concentrations and ventricular function decreased with elevating calcium concentration in high-potassium cardioplegic solution. They suggested that these resultswere associated with increasedmyocardial mechanicalactivity by excessive calcium influx. In agreement with these experimentalstudies,our results indicate that low-calcium potassium cardioplegic solution exerted beneficial effects on the preservation of left ventricular function when compared with St. Thomas' Hospital cardioplegic solution (Fig. 4). Postischemic recovery of Emax was extremely poor in the group that received St. Thomas' Hospital cardioplegic solution, whereas it was well preserved in the group with low-calcium cardioplegic solution. This correlated closely with the results of postoperative catecholamine doses (Fig. 4) and CK-MB values (Fig. 1) in the two groups. We therefore believe that high-potassium cardioplegic solutionwith low or no calcium affords excellent myocardial protection during ischemia, at least in clinicaluse. In contrast to our and other results, optimal calcium concentrationwas found to be around 1.2mmol/L in the St. Thomas' Hospital solution." The basicconceptof the St. Thomas' group for formulating cardioplegic solution was to avoid extremes of ionicconcentrations, osmolarity, or pH in the solution." According to the study of Yamamoto, Braimbridge, and Hearse,"maximumrecovery of aortic flow and minimum CK leakage were observed whenthe calciumconcentration was 1.2mmolj L. However, they used Krebs-Henseleit bicarbonate buffer,in whichcalciumconcentration wasextraordinarily high (2.5 mmoljL) as a control perfusate. Therefore their finding ofan optimalconcentrationofcalciumof 1.2 mmoljL in the St. Thomas' Hospitalsolution may not be directlycomparablewith the clinicalsetting. In addition, as Yamamoto, Braimbridge, and Hearse6 discussed, hypothermia may have influenced the results of their study, in which all experiments were conducted under normothermic conditions. Because hypothermia is. an important factor for myocardialprotection during cardiac operations, studieson cardioplegia shouldnot exclude the additionaleffectof cooling on ischemic myocardium. Jynge,Hearse, and Braimbridge'' alsoclaimedthat the complete omission of calcium in St. Thomas' Hospital solutionshouldbe avoidedbecauseof the possible hazard of a calcium paradox, whichalsocausesmyocardial injury. However, the calcium paradox can be diminished by reducing sodium ions5, 6 or lowering the temperature' in calcium-free perfusate. In clinical states, noncoronary collaterals.P hypothermia," and oxygenation of the car-

The Journa] of Thoracic and Cardiovascular Surgery

dioplegic solution-' can also preventthis phenomenon. It therefore seems clear that the calcium paradox is not important clinically and the addition of calcium to cardioplegic solution is unnecessary. Magnesium and cardioplegia. Magnesium has been advocated as a component of cardioplegic solution by Hearse, Stewart, and Braimbridge.? One of the reasons for including this divalent cation in cardioplegic solution is its antagonistic action against calcium." It can reduce transmembranous calcium influx and thereby minimize the intracellular calcium overload induced by ischemia and reperfusion.!? In addition, magnesium acts as a cofactorof vital enzymaticreactionsand as a complex of the ATP molecules." 10 Theseactionsof magnesium may be important to maintain cellular function. Therefore intracellularmagnesium loss inducedby ischemia-' may be detrimental to tissue survival. Hearse, Stewart, and Braimbridge" concluded that the optimal concentration of magnesium is 16 mmoljL for their originalcardioplegic solution (St. Thomas' Hospital solution). In the presentstudy,wehavefailedto obtain beneficial effects ofmagnesium onischemic myocardium, inasmuch as patients who received St. Thomas' Hospital solution tended to have poorer postischemic recovery of cardiac rhythm. Approximately half (44.4%) patients, in this group had severe bradyarrhythmias during the early periodof reperfusion. Increasingserum magnesium causes sinoatrialand atrioventricular blockand alsointraventricular conduction delay.I I, 24 In fact, a prolonged effect of magnesium on the recovery of cardiac rhythm was clearly noticeable, although most of the St. Thomas' Hospital solution was discarded from the cardiopulmonary bypass circuit. Electrophysiologic abnormalities may readily influence postischemic recovery of mechanicalcardiac function. Patients withSt. Thomas' Hospital solution showed inadequate recovery of Emax and requiredhigh doses of catecholamines postoperatively. A recent study by Cavaliereand coworkers I I demonstrated that hypermagnesemia occurredafter unclamping of the aorta in patients who received St. Thomas' Hospitalcardioplegic solution. They suggested that the use of solutions with a high content of magnesium may be dangerous in some clinical conditions. Engelman and colleagues-' also reported that there were no advantages in including magnesiumin cardioplegic solution in a clinical evaluation. Thus we conclude that magnesium-rich cardioplegic solution is not favorable for clinical myocardial protection. Recovery of electrical activity on reperfusion. During surgical reperfusion, cardiac surgeons frequently encounterreperfusion-induced ventriculartachyarrhyth-

Volume 101 Number 4 April 1991

miaswith or without spontaneous resumption of cardiac rhythm.P' In the present study, spontaneous recovery of cardiac rhythm or spontaneous defibrillation was more prevalent in the low-calcium cardioplegia group. In most (83.3%) patients in the St. Thomas' Hospital solution group,reperfusion-induced ventricularfibrillation necessitated electrical defibrillation. We27 haverecentlydemonstratedthat vulnerability of the heart to reperfusion arrhythmias is largelyrelated to intracellular calcium availability. The electrophysiologic basis of this type of arrhythmias might be dependenton the heterogeneity of electricalrecovery in the previously ischemic myocardium. Relief of intracellular calcium accumulation can decreasethe ischemia-induced electrophysiologic alterations, and consequently heterogeneity of electrical recovery during reperfusion can be reduced. Thus prevention of cellular calcium accumulation by hypocalcemia, hypothermia, and calcium antagonists is important to increase the threshold of ventricular tachyarrhythmias during reperfusion. Glucose and cardioplegia. Glucose is another factor that may haveinfluenced the resultsin the presentstudy. We add glucose (25 gm/L) to correct the osmolarity to within normal levels, not to exert a further direct protective effect on the heart. In our experimental study,' glucose is much less important in suppressing ischemic contracture than calcium, sodium, potassium, and temperature'COmbined. However, we have little information concerning the effect of glucose in crystalloid cardioplegic solutions in the present study. Further studies would be required to clarify whether glucose is necessary in oxygenated crystalloid cardioplegic solution and what the optimaldose of glucose is in the solution.

Concluding comments In the presentstudy, we observed a superiorprotective effect of low-calcium and magnesium-free potassium cardioplegic solution on postischemic recovery of cardiac function and rhythm in comparison with St. Thomas' Hospital solution. However, cardioplegic techniques differ between institutions. The degreeof systemic hypothermia, the dose and infusion rate of cardioplegic solution, and the bloodtemperature at the time of reperfusion may influence the protective properties of cardioplegic solution. We therefore conclude that divalent cations such as calcium and magnesium are inadequate for potassium cardioplegia, at least in combination with our cardioplegic method,and the principleof myocardial preservation may be to minimize calcium-related tissue injury.

Low-calcium cardioplegic solution 7 0 I

REFERENCES I. Jennings RB, Ganote CE, Reimer KA. Ischemic tissue injury. Am J PathoI1975;81:179-98. 2. Shen AC, Jennings RB. Kinetics of calcium accumulation in acute myocardial ischemic injury. Am J Pathol 1972; 67:441-52. 3. Kinoshita K, Ehara T. Importance of sodium ions in the protective effects of high-potassium, high-glucose solution on electromechanical activities in the guinea-pig myocardium. J Mol Cell CardioI1984;16:405-19. 4. Hearse DJ, Braimbridge MY, Jynge P. Protection of the ischemic myocardium. Cardioplegia, 1st ed. New York: Raven Press, 1981;210-99. 5. Jynge P, Hearse DJ, Braimbridge MY. Myocardial protection during ischemic cardiac arrest: possible hazard with calcium-free cardioplegic infusates. J THORAC CARDIOVASC SURG 1977;73:848-55. 6. Yamamoto F, Braimbridge MY, Hearse DJ. Calcium and cardioplegia: the optimal calcium content for the St. Thomas' Hospital cardioplegic solution. J THORAC CARDIOVASe SURG 1984;87:908-12. 7. Goshima K, Wakabayashi S, Masuda A. Ionic mechanism of morphological changes of cultured myocardial cells on successive incubation in media without and with Ca 2+. J Mol Cell Cardiol 1980;12:1135-57. 8. Holland CE, Olson R. Prevention by hypothermia of paradoxical calcium necrosis in cardiac muscle. J Mol Cell Cardiol 1975;7:917-28. 9. Hearse DJ, Stewart DA, Braimbridge MY. Myocardial protection during cardiac arrest: the importance of magnesium in cardioplegic infusates. J THORAC CARDIOVASC SURG 1978;75:877-85. 10. Polimeni PI, Page E. Magnesium in heart muscle. Circ Res 1973;33:367-74. II. Cavaliere F, Sciarra M, Zamparelli R, Monaco C, Bonifazi S, Schiavello R. Magnesium metabolism in open-heart surgery. Resuscitation 1986;3:215-21. 12. Asou T, Oe M, Tominaga R, et a!. Optimal timing for application of ventricular assist devices in patients who cannot be weaned from cardiopulmonary bypass: an experimental study. ASAIO Trans 1988;34:465-9. 13. Igarashi Y, Suga H. Assessment of slope of end-systolic pressure-volume line of in situ dog heart. Am J Physiol 1986;250:H685-92. 14. Sunagawa K, Yamada A, Senda Y, eta!' Estimation of the hydromotive source pressure from ejecting beats of the left ventricle. IEEE Trans Biomed Eng 1980;27:299-305. 15. Weber A, Murray JM. Molecular control mechanisms in muscular contraction. Physiol Rev 1973;3:612-72. 16. Cooley DA, Reul GJ, Wukasch DC. Ischemic contracture of the heart: "stone heart." Am J CardioI1972;29:578-80. 17. Hearse DJ, Garlick PB, Humphrey SM. Ischemic contracture of the myocardium: mechanisms and prevention. Am J CardioI1977;39:986-93. 18. Boggs B, Torchiana DF, Geffin GA, et a!. Optimal myo-

The Journal of

7 0 2 Kinoshita, Oe, Tokunaga

19.

20.

21.

22.

23.

cardial preservation with acalcemic crystalloid cardioplegic solution. J THORAC CARDIOVASC SURG 1987;93:838-46. Torchiana DF, Love TR, Hendren WG, et al. Calcium-induced ventricular contraction during cardioplegic arrest. J THORAC CARDIOVASC SURG 1987;94:606-13. Brazier J, Hottenrott C, Buckburg G. Noncoronary collateral myocardial blood flow.Ann Thorac Surg 1975;19:42635. Alto L, Dhalla N. Myocardial cation contents during induction of calcium paradox. Am J Physiol 1979;237: H713-9. Hendren WG, Geffin GA, Love TR, et al. Oxygenation of cardioplegic solution: potential for the calcium paradox. J THORAC CARDIOVASC SURG 1987;94:614-25. Shen AC, Jennings RB. Myocardial calcium and magnesiuminacuteischemicinjury.AmJPathoI1972;67:417-40.

Thoracic and Cardiovascular Surgery

24. Bellet S. Clinical disorders of the heart beat. 3rd ed. Philadelphia: Lea & Febiger, 1971:910-13. 25. Engelman RM, Rousou JH, Lemeshow S, Dobbs WA. A clinical comparison of potassium and magnesium-potassium crystalloid cardioplegia: metabolic considerations. J Cardiovasc Surg 1985;26:343-50. 26. Manning AS, Hearse DJ. Reperfusion-induced arrhythmias: mechanisms and prevention. J Mol Cell Cardiol 1984;16:497-518. 27. Kinoshita K, Mitani A, Hearse DJ, Braimbridge MV, Manning AS. Reperfusion-induced arrhythmias in the conscious rat: a comparative study with three calcium antagonists. J Surg Res 1989;47:166-72.

Botmd volumes available to subscribers Boundvolumesof THEJOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are availableto subscribers(only) for the 1991issues from the Publisher, at a cost of $64.00 for domestic, $86.48 for Canadian, and $82.00 for international for Vol. !OI (January-June) and Vol. 102 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertisingis 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, volumenumber, and year stamped in goldon the spine. Payment must accompany all orders. Contact Mosby-Year Book, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis, Missouri 63146-3318, USA; phone (800) 325-4177, ext. 4351. Subscriptions must be in force to quatify. Bound volumes are not available in place of a regular JOURNAL subscription.