Influence of the pH of cardioplegic solutions on intracellular pH, high-energy phosphates, and postarrest performance

J

THORAC CARDIOVASC SURG

90:235-242, 1985

Influence of the pH of cardioplegic solutions on intracellular pH, high-energy phosphates, and postarrest performance Protective effects of acidotic, glutamate-containing cardioplegic perfusates The common practice of using alkalotic cardioplegic solutionsis not supported by experimental evidence. The presentstudy was conducted(1)to assess the effects of varyingthe pH (7.00,7.40, and 7.70 at 20° C)

of a g1utamate-containing cardioplegic solution on intraceUular pH, higlH!nergy phosphate content, and postarrest functional recovery and (2) to compare the effects of various buffers (glutamate, bicarbonate, TRIS, and histidine) at a given pH (7.00 and 7.40).Isolated perfusedrat hearts were subjected to 2 hours of cardioplegic arrest at 15° C foUowed by 30 minutes of reperfusion. IntraceUuIar pH and high-energy phosphate content were measured at 4 minute intervals by phosphorus 31 nuclear magnetic resonance spectroscopy. 1bese data were correlated with postischemic recovery of function. There was no significant difference between the intraceUular pH values recorded at the end of arrest in the three g1utamate-containing groups.However, the acidotic solution(pH 7.00)resultedin better preservation than the alkalotic solution (pH 7.70). as evidenced by (1) a higher creatine phosphate content at the end of arrest (61 % ± 9 % of controlvalues versus 30 % ± 9 % [mean ± standard error of the mean], p < 0.05). (2) a higher adenosine triphosphate content at the end ofreperfusion (102% ± 5% versus82% ± 6%, p < 0.05). and (3) a faster recovery of aortic flow (at 3 minutes of reperfusion, 91 % ± 11 % versus 51 % ± 11 %, p < 0.05).Subsequentcomparisonof buffers showed that bicarbonate,TRIS, and histidine were equaUy effective in maintainingintraceUular pH closeto control values duringarrest. Conversely, the use of glutamate resulted in a more pronounced faD in intraceUular pH, which correlated with a better preservation of adenosine triphosphate and a better functional recovery than in the other groups. OveraU, the greatest extent of preservation was provided by the pH 7.00 g1utamate-containing cardioplegic solution. We conclude that additional protection can be conferred to the cold, chemicaUy arrested heart by combining (1) mild intraceUular acidosis, which lowers metabolic needs during arrest, most likely through a limitationof calciumoverload, and (2)provision of glutamate, whichmay act as a substrate for anaerobic energy production while aDowing intraceUular pH to be kept within the appropriate range.

Monique Bernard, Ph.D., Philippe Menasche, M.D., Paul Canioni, Ph.D., Eric Fontanarava, Christian Grousset, Ph.D., Armand Piwnica, M.D., and Patrick Cozzone, Ph.D., Paris and Marseille. France

From the Service de Chirurgie Cardio-Vasculaire and INSERM, U-I27, H6pital Lariboisiere, Paris. France, and the Institut de Chimie Biologique, Universite d'Aix-Marseille I, Marseille, France. Supported by CNRS (LA 202) and grants from INSERM (CRL 813006 and GBM 130067), DGRST (TLBM 81-M-0914), the Federation de Cardiologic, and the UER Lariboisiere-SaintLouis. Received for publication May 20, 1984. Accepted for publication Oct. 18, 1984.

O f the numerous experimental studies dealing with myocardial protection during elective ischemic arrest, few have focused on the optimal pH and buffer of cardioplegic solutions. In practice, most of the cardioplegic solutions in clinical use are titrated to an alkaline Address for reprints: Dr. Philippe Menasche, Service de Chirurgie Cardio-Vasculaire, H6pital Lariboisiere; 2, rue Ambroise Pare, 75010 Paris France.

235

The Journal of

2 3 6 Bernard et al.

Thoracic and Cardiovascular Surgery

ISCHEMIA

pH;

E

<)---(> pH 7.00 }

~6 7.5

- - pH 7.40 +--+pH7.70

REPERFUSION

Glutamate cardioplegia

0---0 Hypothermia

4 6.5

3L-

...l-

5

L-

-L

,..-..

pH

Fig. 1. Titration curve for inorganic phosphate. The solid line represents the theoretical titration curve calculated by appropriate equations. The dots represent data obtained on a test solution titrated in the range of pH 5.0 to 9.0. The pH was determined with a Tacussel pH meter (Model ISIS 20 000) linearly calibrated by means of standard buffers.

pH, I although the use of alkalotic perfusates under conditions of cold chemical cardioplegia can be questioned in view of basic studies demonstrating the protective effects of a mild intracellular acidosis upon the hypoxic and ischemic myocardium." 3 The controversy on this issue is primarily due to the methodologic difficulties in assessing the effects of the pH of cardioplegic solutions upon intracellular pH (pHi) and in correlating tissue changes in pHi with the postischemic recovery of metabolism and function. This problem can now be circumvented by the use of phosphorus 31 nuclear magnetic resonance (NMR) spectroscopy. This technique has provided reliable, noninvasive on-line monitoring of pHi and phosphate metabolites in isolated perfused heart whose indices of left ventricular function can be simultaneously recorded.4, 5 Our previous experience with NMR technology" prompted the present study, which was designed (1) to assess the effects of glutamate-buffered cardioplegic solutions titrated to different values of pH (7.00, 7.40, and 7.70 at 20° C) during global myocardial ischemia and (2) to compare the effects of various buffers (glutamate, bicarbonate, TRIS, and histidine) at a given pH (7.00 and 7.40). Material and metbods Perfusion metbods. The preparation used for these studies was the isolated, perfused, working rat heart.v? The perfusion fluid was Krebs-Henseleit bicarbonate buffer (7.40 at 37° C) supplemented with 11 mM glucose and gassed with 95% oxygen plus 5% carbon dioxide. The perfused heart, contained in the NMR

30

60

90

120 10

20

30

min

Fig. 2. Effect of varying the pH of glutamate-containing cardioplegic solutions (7,00, 7.40 and 7.70 at 20° C) on intracellular pH (pHi).

tube, was placed into the magnet. After a control working period of 15 to 30 minutes at 37° C, global ischemia was induced by clamping the aortic and left atrial lines, and the cardioplegic solution was infused over a 2 minute period via a sidearm on the aortic cannula at a pressure of 60 em H 20 . The hearts were subjected to 120 minutes of global ischemia at 15° C, maintained via a thermostatically controlled heart chamber, with reinfusions of the cardioplegic solution at 30 minute intervals. Normothermic reperfusion was then instituted for 30 minutes in the working mode. Perfusate flow was removed from the NMR sample tube by vacuum aspiration throughout the experimental procedure, so that no phosphate signal originated from the external bathing medium." NMR experiments. Phosphorus 31 NMR spectra were obtained on a Nicolet NT 200 WB spectrometer at 4.7 tesla in a wide-bore superconducting magnet. The instrument was operated in the pulsed Fourier transform mode and was interfaced to a Nicolet 1180 computer. Concentrations of phosphorus-containing metabolites were determined by planimetric integration of the phosphorus 31 resonance areas. Resonance assignments for adenosine triphosphate (ATP), creatine phosphate, and inorganic phosphates (Pi) were made from comparison with authentic samples and agreed with data from the literature.v' The pHi was determined from the chemical shift of the Pi peak relative to that of creatine phosphate. The pH to which a given chemical shift corresponds was derived from a standard titration curve (Fig. 1). That this curve was unaffected by the various buffers used in the cardioplegic solutions supports the validity of our technique for assessing pHi.

Volume 90

Acidotic, glutamate-containing cardioplegic solution

Number 2

237

August. 1985

I

~ pH 7,00 Glutamate '4 ATP _ pH 7,40 cardioplegia EZZZZI pH 7,70 c:::::J Hypothermia

ISCHEMIA

I

I

REPERFUSION

•

110 100

90

%PCr

160 140

=pH7.00 _

= =

pH 7,40

I

I REPERFUSION

ISCHEMIA

I I

Glutamate cardioplegia

I I I I I I I I

pH 7.70 Hypothermia

120

80

100

10 60

80

50

60

40

40

30

20 0

o

*

30

60

90

10

30 min

Fig. 3. Effect of the pH of the glutamate-containing cardioplegic solutions on adenosine triphosphate (A TP) levels. The concentration of ATP was determined from nuclear magnetic resonance spectra.Data are displayed as percentage of control (set at 100%). *p < 0.05 compared to pH 7.70 cardioplegic solution. Experimental design. Sixty rats were divided into 10 groups, each group consisting of six hearts. Except for the control group, in which myocardial protection consisted of hypothermia alone, all hearts received multidose cardioplegia. The composition of the cardioplegicsolution was as follows: K+, 4 mM; Mg2+, 13 mM; Ca2+, 0.25 mM; Na+ 100 mM. Osmolarity was adjusted to 370 mOsmJL by the addition of mannitol. The only two variables between groups were the pH and the buffer of the cardioplegic solution. In a first series of experiments (four groups), we assessed the effects of varying the pH (7.00, 7.40, and 7.70 at 200 C) of cardioplegic solutions, all containing glutamate (20 mM), and compared the results with those obtained with hypothermia alone. In a second series of hearts (six groups), the effects of other buffers, i.e., bicarbonate (20 mM), tris (hydroxymethyl) aminomethane (47.5 mM), and histidine (64.4 mM) were compared to those previouslyobtained with glutamate. Because glutamatecontaining cardioplegic solutions at pH 7.00 and 7.40 had yielded the best results (see Results), these two pH values were selected for that part of the study dealing with the comparison of buffers. Data collection and analysis. Aortic flow, coronary flow, aortic pressure, and heart .rate were serially recorded during the control period and at 3, 5, 10, 15, 20, and 30 minutes of reperfusion. NMR measurements were taken at 4 minute intervals throughout the entire experimental time course. All metabolic and functional results are expressed as

r 30

JJ After CP

60

r

-. 0.

After CP

~: r:: ~ 0.

90

After CP

120

10

30 min

Fig. 4. Effect of the pH of the glutamate-containing cardiaplegic solutions on creatine phosphate (PCr) levels. .The concentration of PCr was determined from nuclear magnetic resonance spectra. Data are displayed as percentageof control (set at 100%). The bar graphs above the legend "after CP" refer to the PCr content determined after each reinfusion of the cardioplegic (CP) solution. *p < 0.05 comparedto pH 7.70 cardioplegic solution. a percent recovery of preischemic values. Data are presented as the mean ± the standard error of the mean. Tests of significance were made with the paired and unpaired Student's t tests or the repeated measures analysis of variance, with p = 0.05 considered to be the limit of significance. Results Comparison of glutamate-containing cardioplegic solutions titrated to different pH values (7.00, 7.40, and 7.70) with hypothermia. pHi. In hearts treated with hypothermia alone, pHi decreased from a control value of 7.10 ± 0.04 to 6.13 ± 0.12 at the end of the 120 minute ischemic period (p < 0.001 versus control). The rate of fall in pHi was significantly decreased in the three cardioplegic groups (Fig. 2). In these hearts, limitation of intracellular acidosis was further ensured by the temporary increase in pHi after each reinfusion of the cardioplegic solution. There was no statistical difference among the values of pHi recorded at the end of arrest for the three cardioplegic groups. After reperfusion was started, pHi underwent a slight overshoot and then returned to control values in all hearts. ATP. In hearts protected by hypothermia alone, ATP fell progressively to 24.69% ± 7.42% of control, a content significantly lower than the 103.11% ± 14.67%, 88.32% ± 6.59%, and 95.46% ± 6.35% in hearts that received pH 7.00 (p < 0.01), pH 7.40 (p < 0.001), and

The Journal of Thoracic and Cardiovascular Surgery

2 3 8 Bernard et al.

% recovery of heart rate

% recovery of aortic flow 110 100

90 80 70

--- --/'

60

/'

/'

50 40

30 20

00----<) pH 7.00} . - pH 7.40 Glutamate cardioplegia ............ pH7.70

10

o

20

25

30

o

Post - ischemic time (min)

Fig. S. Postischemic recovery of aortic flow expressed as a percent of preischemic control for hearts that received glutamate-containing cardioplegic solutions adjusted at different pH values. *p < 0.05 compared to pH 7.70 cardioplegic solution. **p < 0.05 compared to pH 7.70 cardioplegic solution.

Fig. 6. Postischemic recovery of heart rate (expressed as a percent of preischemic control) for hearts that received pH 7.40 cardioplegic solution containing various buffers. *p < 0.01 compared to bicarbonate; p < 0.05 compared to TRIS. **p < om compared to bicarbonate. ***p < 0.05 compared to histidine and TRIS.

pH 7.70 (p < 0.001) cardioplegic solutions, respectively (Fig. 3). Although the three cardioplegic groups demonstrated no significant difference in their ATP content at the end of arrest, their postischemic recovery profiles were different. After 30 minutes of reperfusion, the ATP content in hearts having received acidotic (pH 7.00) cardioplegic solution was found to be significantly higher than that of hearts infused with the alkalotic (pH 7.70) solution during ischemia (101.87% ± 5.27% versus 82.21% ± 6.22%, p < 0.05). Creatine phosphate. After the onset of global ischemia, creatine phosphate levels decreased in all hearts. This decline was less pronounced in the three cardioplegic groups, wherein each additional infusion of cardioplegic solution caused creatine phosphate to rise up to or even above control levels (Fig. 4). At the end of arrest, the highest creatine phosphate content was found in hearts protected by the acidotic cardioplegic solution

(61.37% ± 8.95% versus 30.05% ± 8.94% in the pH 7.70 group, p < 0.05). After initiation ofreflow, creatine phosphate levels increased in all four groups. Pi. The pattern of accumulation of Pi closely resembled the mirror image of that of creatine phosphate, i.e., a marked rise throughout each ischemic interval followed by an abrupt decline after each infusion of cardioplegic solution. Although the difference did not reach statistical significance, hearts protected by the acidotic cardioplegic solution tended to accumulate less Pi than those receiving the alkaline solution. During reperfusion, Pi fell rapidly toward, but did not reach, control levels, with hearts perfused with the pH 7.70 solution demonstrating the greatest decrease in Pi during early reflow. Left ventricular function. Upon reoxygenation, only four of the eight hypothermic hearts had a measurable aortic flow, which averaged 50% of the control values.

Volume 90 Number 2

Acidotic, glutamate-containing cardioplegic solution

August, 1985

2 39

Table I. Comparative effects of various buffers added to cardioplegic solutions at pH 7.40 upon intracellular pH and high-energy phosphate content Creatine phosphate (% of control)

Intracellular pH Buffer

Glutamate Bicarbonate

TRIS

Histidine

Before arrest 7.16 7.08 7.21 7.16

± 0.05 ± 0.06 ± 0.05 ± 0.04

I

End of arrest 6.93 7.32 7.23 7.16

± 0.07 ± 0.08* ± 0.05:j: ± 0.09§

End of arrest 55.54 45.96 52.43 63.78

± 6.56

I

± 6.24 ± 12.03 ± 7.20

Adenosine triphosphate (% of control)

End of reperfusion 82.75 111.05 70.37 82.37

± 9.57

± 7.19t ± 3.35 ± 5.78

End of arrest 88.32 92.53 66.26 83.39

± 6.59 ± 6.83 ± 9.10 ± 4.04

I

End of reperfusion 84.87 81.55 79.78 82.86

±

5.88 5.12 ± 10.87 ± 4.05

±

Legend: All results are the mean of six hearts, and the standard error of the mean is indicated. *p < 0.05 versus glutamate. tp < 0.05 versus glutamate, p < 0.05 versus TRIS. p < 0.0 I versus histidine. tp < 0.01 versus glutamate §p < 0.05 versus glutamate.

Conversely, all the cardioplegically protected hearts resumed contractile activity within seconds after the onset of reperfusion. Although the final extent of recoverywas not significantly different among the three cardioplegicgroups, their rates of recovery differed (Fig. 5): At 3 minutes of reperfusion, hearts infused with the pH 7.00 and 7.40 cardioplegic solution had recovered to 91.06% ± 11.48% and 78.65% ± 4.98%, respectively, of their preischemic aortic flows, whereas hearts infused with the Ph 7.70 cardioplegic solution had recovered to only 51.04% ± 10.08% (p < 0.05). In the pH 7.00 and 7.40 groups, postischemic aortic flows subsequently plateaued at approximately 70% (pH 7.00 group) and 60% (pH 7.40 group) of control levels, whereas in the pH 7.70 group a similar level of recovery was attained only after 30 minutes of retlow. Similar patterns of recovery were observed for heart rate and cardiac output, calculated as the sum of aortic and coronary flow. There was no significant difference in the recovery of coronary flow and aortic pressure among the three groups.

Comparison of various buffers at a constant pH. pH 7.40 experimental series. pHI. The best buffering capacity was provided by bicarbonate, since in this group pH rose from 7.08 ± 0.06 before arrest to 7.32 ± 0.08 at the end of arrest. In hearts receiving histidine and TRIS as a buffer, pHi values after 120 minutes of ischemia were unchanged from control values. In contrast, the use of glutamate resulted in a significant fall in pHi (Table I). HIGH-ENERGY PHOSPHATE CONTENT, These different patterns of changes in pHi had no apparent effect upon the preservation of high-energy phosphates, since creatine phosphate and ATP levels were decreased to the same extent in all hearts at the end of the ischemic period. Similarly, at the end of reflow, tissue levels of

creatine phosphate and ATP did not differ significantly among the four groups (Table I). LEFr VENTRICULAR FUNCITON. The ultimate recovery of aortic flow was similar in all groups. However, aortic flow recovered at a faster rate in glutamate-perfused hearts (p < 0.05 versus TRIS- and histidine-treated hearts at 3 minutes of reflow). The latter group was also the only one in which heart rate equaled control levels at the very onset of reperfusion and remained consistently higher than in the three other groups up to 15 to 20 minutes of reflow (Fig. 6). pH 7.00 experimental series. pHI. As in the pH 7.40 series, the use of bicarbonate, histidine, or TRIS during arrest allowed final values of pHi not to differ significantly from preischemic control values. In contrast, the fall in pHi was more pronounced in glutamate-treated hearts (Table 11). HIGH-ENERGY PHOSPHATE CONTENT. Both at the end of ischemia and at the end of reflow, the creatine phosphate content did not differ significantly among the four groups. This was not the case for ATP. After 120 minutes of arrest, the highest ATP content was yielded by glutamate-treated hearts, whereas histidine resulted in a severe depletion of ATP stores. During reflow, the ATP content remained stable in the glutamate group but fell significantly in bicarbonate-treated hearts (Table 11). LEFT VENTRICULAR FUNCTION. There was no significant difference in the recovery of aortic flow among glutamate-, bicarbonate-, and TRIS-treated hearts which, by 30 minutes of reflow, averaged 74.10% ± 7.46%,68.88% ± 5.67%, and 67.13% ± 3.10% of control values, respectively. In close correlation with metabolic data, secovery of aortic flow during the 20 initial minutes of reperfusion was significantly poorer in histidine-treated hearts. As observed in the pH 7.40 series,

The Journal of

240

Bernard et af.

Thoracic and Cardiovascular Surgery

Table II. Comparative effects of various buffers added to cardioplegic solutions at pH 7.00 upon intracellular pH and high-energy phosphate content

Buffer

Before arrest

Glutamate Bicarbonate TRIS Histidine

7.15 ± 0.Q7 7.05 ± 0.03 7.15±0.05 7.19 ± 0.05

I

End of arrest

7.00 7.12 7.14 7.17

Adenosine triphosphate (% of control)

Creatine phosphate (% of control)

Intracellular pH

± ± ± ±

0.06 0.02 0.05 0.06

End of arrest

61.37 ± 47.08 ± 53.65 ± 56.67 ±

8.95 7.58 2.64 10.24

I

End of reperfusion

93.37 ± 97.89± 114.08 ± 90.13 ±

14.70 5.19 9.61 10.53

End of arrest

103.11 93.29 86.37 62.62

± ± ± ±

14.67* 6.60:1: 8.1711 7.84

I

End of reperfusion

101.87 ± 84.85 ± 95.77 ± 75.70 ±

5.27t 5.65§ 10.28 10.65

Legend: All results are the mean of six hearts, and the standard error of the mean is indicated. *p < 0.05 versus histidine. tp < 0.05 versus bicarbonate. < 0.02 versus TRIS. §p < 0.05 versus TRIS, p < 0.05 versus histidine. Ip < 0.05 versus histidine.

+p

glutamate-buffered cardioplegia was the only solution to allow recovery of control heart rate at the onset of reperfusion, the differences with the TRIS and histidine groups attaining statistical significance up to 10 minutes of reflow. Discussion Measurement of pHi by phosphorus 31 NMR. To what extent the pH of cardioplegic solutions affects myocardial metabolism during hypothermic arrest and postischemic recovery has not been fully investigated. The present study was designed to clarify this issue. Results were assessed by phosphorus 31 NMR spectroscopy. This technique allows a measurement of pHi together with phosphate metabolites in perfused hearts in which various indices of left ventricular performance can be simultaneously recorded.i' Additional important features of NMR are its time resolution and its ability to acquire pHi and high-energy phosphates noninvasively,? so that metabolic and functional data can be obtained serially in the same heart during the ischemic and reperfusion periods. The reliability of this technique of measuring pHi has been demonstrated by several studies showing that the value of cardiac pH determined by NMR was in excellent agreement with pH values determined by more conventional methods.v " Protective effects of acidotic cardioplegic solutions. One major finding of our study was that the acidotic cardioplegic solution provided the best myocardial preservation. This result may appear to contradict previous studies that showed a protective effect of alkalosis either during preischemic coolingII or during postischemic reperfusion.'>" However, these studies examined the effects of pH on perfused beating hearts,

whereas the present work addressed a totally different issue, i.e., the effects of extracellular pH on a cold, cardioplegically arrested heart. In this respect,our data are consistent with other studies,2,3.15.16 which showed that acidosis may exert protective effects upon the ischemic or hypoxic myocardium. Of particular relevance is the report by Nugent and associates" that the use of an alkaline cardioplegic solution (pH 7.70) resulted in a significantly greater depression of postarrest left ventricular function than either a normal or acidic (pH 7.00) solution. Since hydrogen ions compete with Ca2+ ions for several membrane" and cellular sites,":" it is likely that acidosis exerts its major protective effect through the limitation of Ca2+ overload/,21.23,24 which is a major feature of myocardial ischemic injury. I Thus, the postischemic decline in ATP levels and the associated impairment in functional recovery yielded by our alkalotic cardioplegic group could be explained by a defective mitochondrial production of ATp I9 resulting from an alkalosis-induced Ca2+ overload." The improved protection afforded by the acidotic solution is consistent with this concept, since an increased extracellular concentration of protons is expected to have reduced the cellular influx of membrane-bound Ca2+ 26 and thereby to have enhanced recovery of function." However, the magnitude of the fall in pHi should be carefully controlled, since mild acidosis may be protective to the ischemic myocardium whereas severe acidosis may lead to cell necrosis.v 28. 29 There is evidence that pHi should not fall below 6.6. 30,31As shown in Fig. 2 and Tables I and II, all the cardioplegic solutions, including those titrated to pH 7.00, allowed pHi to remain above this critical threshold.

Volume 90 Number 2 August, 1985

It can be argued that the increased protection provided by the acidotic solution was observed without concomitant differences in pHi values among the three cardioplegic groups. We think that maintenance of pHi at similarly high levels in all cardioplegic hearts was due to (1) the alkalotic shift in pHi induced by hypothermia," (2) washout of acid metabolites by multidose cardioplegia, and (3) preservation of ATP content, since hydrolysis of ATP is the principal direct means by which protons are generated in the cytoplasm." Clinically, the buffering capacity of the noncoronary collateral blood flow" is also likely to be significant. These considerations imply that other factors than proton-binding capacity should be considered in buffer selection. Comparative effects of various buffers in cardioplegic solutions. Bicarbonate (pK 6.35 at 25° C) is the most commonly used buffer in cardioplegic solutions. Our results support its effectiveness in preventing tissue acidosis, both in pH 7.00 and in pH 7.40 solutions. However, the aforementioned detrimental effects of alkalosis upon the myocardial ischemic cell suggest that the high pHi values yielded by bicarbonate-treated hearts contributed to the postischernic abnormalities observed in these groups (especially at pH 7.40). These abnormalities include an overshoot in creatine phosphate content, a concomitant decline in ATP levels" (Tables I and 11), and a slow recovery of heart rate. TRIS buffer also has a favorable pK value (8.08 at 25° C), and its high temperature coefficient (-0.028 pH unitsrC) ensures that the change in the pH of the buffer is similar to the change in the point of neutrality (pN) of water on cooling." Accordingly, our TRISbuffered solutions allowed pHi to remain stable during arrest, although this stability did not avoid a poor recovery of ATP levels and aortic flow in the pH 7.40 group. These results are in keeping with those reported by Flaherty and Jacobus" and support other studies documenting the toxicity of TRIS in several tissues. 1 The use of histidine in cardioplegic solutions has been advocated because of its favorable pK value (6.04 at 25° C) and its high temperature coefficient (-0.022 pH units;aC).33. 37 Our data confirm that histidine is an effective buffer. However, as with TRIS, the stability of pHi throughout ischemia did not avoid a poor recovery of both ATP levels and mechanical function in the pH 7.00 group. Increasing the pH to 7.40 improved preservation, which remained, however, inferior to that yielded by the pH 7.00 glutamate-containing solution. Our selection of glutamate was based upon studies showing that the delivery of glutamate provided additional protection to the ischemic heart. 14. 38 This protection seems to be related to an enhancement of substrate

Acidotic, glutamate-eontaining cardioplegic solution

24 1

supply through the malate-aspartate shuttle resulting in an increased anaerobic production of ATP.39 That the fall in pHi during arrest was consistently greater in glutamate-treated hearts than in the other groups could be easily anticipated from its low pK value (4.07 at 25 ° C). However, whereas pHi values yielded by glutamatecontaining solutions remained above the critical threshold of 6.6 in all hearts, we reasoned that the metabolic protection provided by glutamate would overcome the potential disadvantage of a weak buffering capacity. Actually, when data of all groups were plotted together, the best overall recovery was found to have been yielded by the pH 7.00 glutamate-containing cardioplegic solution. Although we admit that this superiority was not consistently based on highly significant differences in data, we believe that modest improvements observed under our experimental conditions, that is, an isolated normal rat heart model, may be of greater relevance to the human diseased heart for which any additional protection can be critical for the ultimate outcome.

2 3 4

5

6

7

8

9

REFERENCES Hearse DJ, Braimbridge MV, Jynge P: Protection of the Ischemic myocardium: Cardioplegia, New York, 1981, Raven Press, pp 263-269 Bing OHL, Brooks WW, Messer JV: Heart muscle viability following hypoxia. Protective effect of acidosis. Science 180:1297-1298, 1973 Greene HL, Weisfeldt ML: Determinants of hypoxic and posthypoxic myocardial contracture. Am J Physiol 232:526-533, 1977 Flaherty JT, Weisfeldt ML, Bulkley BH, Gardner TJ, Gott VL, Jacobus WE: Mechanisms of ischemic myocardial celldamage assessed by phosphorus-31 nuclear magnetic resonance. Circulation 65:561-571, 1982 Gadian GD, Radda GK, Dawson MJ, Wilkie DR: pHi measurements of cardiac and skeletal muscle using 31 P-NMR, Intra-cellular pH. Its Measurement, Regulation and Utilization in Cellular Functions, New York, 1982, Alan R. Liss, pp 61-67 Pernot AC, Ingwall JS, Menasche P, Grousset C, Bercot M, Mollet M, Piwnica A, Fossel ET: Limitations of potassium cardioplegia during cardiac ischemic arrest. A phosphorus-31 nuclear magnetic resonance study. Ann Thorac Surg 32:536-545, 1981 Yamamoto F, Manning AS,Braimbridge MV, Hearse DJ: Cardioplegia and slow calcium-channel blockers. Studies with verapamil. J THORAC CARDIOVASC SURG 86:252-261, 1983 Salhany JM, Pieper GM, Wu S, Todd GL, Clayton FC, Eliot RS: 31 P nuclear magnetic resonance measurement of cardiac pH in perfused guinea-pig hearts. J Mol Cell Cardiol 11:601-610, 1979 Walters FJM, Wilson GJ, Steward DJ, Domenech RJ, McGregor DC: Intramyocardial pH as an index of

242

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

The Journal of Thoracic and Cardiovascular Surgery

Bernard et al.

myocardial metabolism during cardiac surgery. J THORAC CARDIOVASC SURG 78:319-330, 1979 Garlick PB, Radda GK, Seeley PJ: Studies of acidosis in the ischaemic heart by phosphorus nuclear magnetic resonance. Biochem J 184:547-554, 1979 McConnell DH, Nelson R, Goldstein SM, Maloney JV, Deland EC, Buckberg GD: Importance of alkalosis in maintenance of "ideal" blood pH during hypothermia. Surg Forum 26:263-264, 1975 Follette D, Fey K, Livesay J, Maloney JV, Buckberg GD: Studies on myocardial reperfusion injury. I. Favorable modification by adjusting reperfusate pH. Surgery 82:149-155, 1977 Bixler TJ, Flaherty JT, Goldman RA, Gott VL: Beneficial effects of alkalotic reperfusion following ischemic cardiac arrest. J Surg Res 24:488-494, 1978 Menasche P, Grousset C, De Boccard G, Piwnica A: Protective effect of an asanguineous reperfusion solution on myocardial performance following cardioplegic arrest. Ann Thorac Surg 37:222-228, 1984 Biel~ki K: The influence of changes in pH of the perfusion fluid on the occurrence of the calcium paradox in the isolated rat heart. Cardiovasc Res 3:268-271, 1969 Ruigrok TJC, De Moes D, Borst C: Bretschneider's histidine-buffered cardioplegic solution and the calcium paradox. J THORAC CARDIOVASC SURG 86:412-417, 1983 Nugent WC, Levine FH, Liapis CD, LaRaia PJ, Tsai CH, Buckley MJ: Effect of the pH of cardioplegic solution on post-arrest myocardial preservation. Circulation 66:Suppl 1:68-72, 1982 Vogel S, Sperelakis N: Blockade of myocardial slow inward current at low pH. Am J Physiol 233:99-103, 1977 Williamson JR, Schaffer SW, Ford C, Safer B: Contribution of tissue acidosis to ischemic injury in the perfused rat heart. Circulation 53:Suppl 1:3-14, 1976 Poole-Wilson PA, Langer GA: Effect of pH on ionic exchange and function in rat and rabbit myocardium. Am J PhysioI 229:570-581, 1975 Janke J, Fleckenstein A, Hein B, Leder 0, Sigel H: Prevention of myocardial Ca overload and necrotization by Mg and K salts or acidosis, Recent Advances in Studies on Cardiac Structure and Metabolism, Vol 6, A Fleckenstein, G Roma, eds., Baltimore, 1975, University Park Press, pp 33-42 Allen DG, Orchard CH: The effects of changes in pH on intracellular calcium transients in mammalian cardiac muscle. J Physiol (Lond) 335:555-567, 1983 Lakatta EG, Nayler WG, Poole-Wilson PA: Calcium overload and mechanical function in posthypoxic myocardium: biphasic effects of pH during hypoxia. Eur J Cardiol 10:77-87, 1979 Weisfeldt ML, Bishop RL, Greene HL: Effects of pH and Pco, on performance of ischemic myocardium, Recent Advances in Studies on Cardiac Structure and Metabo-

25

26

27 28

29

30

31

32

33

34

35

36

37 38

39

lism, Vol 10, PE Roy, GRona, eds., Baltimore, 1975, University Park Press, pp 355-364 Kubler W; Katz AM: Mechanism of early "pump" failure of the ischemic heart. Possible role of adenosine triphosphate depletion and inorganic phosphate accumulation. Am J Cardiol 40:467-471, 1977 Chapman RA, Ellis D: The existence of a Ca2+ for H+ exchange across the sarcolemma of frog cardiac muscle cells. J Physiol (Lond) 265:19-20, 1977 Nayler WG, Poole-Wilson PA, Williams A: Hypoxia and calcium. J Mol Cell Cardiol 11:683-706, 1979 Nayler WG, Ferrari R, Poole-Wilson PA, Yepez CE: A protective effect of a mild acidosis on hypoxic heart muscle. J Mol Cell Cardiol 11:1053-1071, 1979 Cobbe SM, Poole-Wilson PA: The time of onset and severity of acidosis in myocardial ischemia. J Mol Cell 12:745-760, 1980 Lange R, Kloner RA, Zierler M, Carlson N, Seiler M, Khuri SF: Time course of ischemic alterations during normothermic and hypothermic arrest and its reflection by on-line monitoring of tissue pH. J THORAC CARDIOVASC SURG 86:418-434, 1983 Steenbergen C, Deleeuw G, Rich T, Williamson JR: Effects of acidosis and ischemia on contractility and intracellular pH of rat heart. Circulation 41:849-858, 1977 Gevers W: Editorial. Generation of protons by metabolic processes in heart cells. J Mol Cell Cardiol 9:867-874, 197 Tait GA, Booker PD, Wilson GJ, Coles JG, Steward DJ, McGregor DC: Effect of multidose cardioplegia and cardioplegic solution buffering on myocardial tissue acidosis. J THORAC CARDIOVASC SURG 83:824-829, 1982 Saks VA: Creatine kinase isozymes and the control of cardiac contraction, Heart Creatine Kinase. The Integration of Isozymes for Energy Distribution, WE Jacobus, JS Ingwall, eds., Baltimore, 1980, The Williams & Wilkins Company, pp 109-124 Rahn H, Reeves RB, Howell BJ: Hydrogen ion regulation, temperature and evolution. Am Rev Respir Dis 112:165172, 1975 Flaherty JT, Jacobus WE: Use of 31-phosphorus NMR to optimize buffer capacity in cardioplegic solutions (abstr). Circulation 60:Suppl 2:11, 1979 Bretschneider HJ: Myocardial protection. Thorac Cardiovase Surg 28:295-302, 1980 Robertson JM, Vinten-Johansen J, Buckberg GD, Follette DM, Maloney JV: Prolonged safe aortic clamping (4 hours) with cold glutamate-enriched blood cardioplegia (abstr). Circulation 64:Suppl 4:147, 1981 Rau EE, Shine KI, Gervais A, Douglas AM, Amos EC: Enhanced mechanical recovery of anoxic and ischemic myocardium by amino acid perfusion. Am J Physiol 236:873-879, 1979