Effect of sodium aspartate on the recovery of the rat heart from long-term hypothermic storage

Effect of sodium aspartate on the recovery of the rat heart from long-term hypothermic storage We have investigated the reported ability of aspartate to enhance greatly the cardioprotective properties of the St. Thomas' Hospital cardioplegic solution after prolonged hypothermic storage. Rat hearts (n = 8 per group) were excised and subjected to immediate arrest with St. Thomas' Hospital cardioplegic solution (2 minutes at 4° C) with or without addition of monosodium aspartate (20 mmol/L). The hearts were then immersed in the same solution for 8 hours (4° C) before heterotopic transplantation into the abdomen of homozygous rats and reperfusion in vivo for 24 hours. The hearts were then excised and perfused in the Langendorff mode (20 minutes). Addition of aspartate to St. Thomas' Hospital cardioplegic solution gave a small but significant improvement in left ventricular developed pressure, which recovered to 82 ± 3 mm Hg compared with 70 ± 2 mm Hg in control hearts (p < 0.05). However, coronary flow and high-energy phosphate content were similar in both groups. In subsequent experiments hearts (n = 8 per group) were excised, arrested (2 minutes at 4° C) with St. Thomas' Hospital cardioplegic solution containing a 0, 5, 10, 20, 30, 40, or 50 mmoljL concentration of aspartate, stored for 8 hours at 4° C, and then reperfused for 35 minutes. A bell-shaped doseresponse curve was obtained, with maximum recovery in the 20 mmol/L aspartate group (cardiac output, 48 ± 5 m1/min versus 32 ± 5 m1/min in the aspartate-free control group; p < 0.05). However, additional experiments showed that a comparable improvement could be achieved simply by increasing the sodium concentration of St. Thomas' Hospital cardioplegic solution by 20 mmoljL. Similarly, if sodium aspartate (20 mmoljL) was added and the sodium content of the St. Thomas' Hospital cardioplegic solution reduced by 20 mmol/L, no significant protection was observed when recovery was compared with that of unmodified St. Thomas' Hospital cardioplegic solution alone. In still further studies, hearts (n = 8 per group) were perfused in the working mode at either high (>80 m1/min) or low «50 ml/min) left atrial filling rates. Under these conditions, if functional recovery was expressed as a percentage of preischemic function, artifactuaUy high recoveries could be obtained in the low-fillingrate group. In conclusion, assessment of the protective properties of organic additives to cardioplegic solutions requires careful consideration of (1) the consequences of coincident changes in ionic composition and (2) the characteristics of the model used for assessment. (J 'fHORAC CARDIOV ASC SURe 1992; 103:521-31)

Manuel Galiiianes, MD, David J. Chambers, PhD, and David J. Hearse, PhD, FACC, DSc,

London, UnitedKingdom

From the Department of Cardiovascular Research, The Rayne Institute, St Thomas' Hospital, London, United Kingdom. Supported in part by grants from STRUTH and the National Heart, Lung, and Blood Institute (HL 39457). Received for publication June IS, 1990. Accepted for publication Nov. 21, 1990. Address for reprints: Professor David J. Hearse, Cardiovascular Research, The Rayne Institute, St Thomas' Hospital, London SEI 7EH, UK

12/1/28123

Cardiac transplantation is now a successful treatment for irreversible, final stage cardiac failure.' Improved control of organ rejection has undoubtedly played a major role, 1 but problems still remain, in particular, the relatively short time limit for myocardial preservation during explantation and storage. In consequence, a number of laboratories are endeavoring to develop solutions for the long-term storage of the heart. Although continuous blood perfusion of the heart has been shown to be effec-

521

522

The Journal of Thoracic and Cardiovascular Surgery

Galihanes, Chambers, Hearse

Table I. Composition (in mmoljL) of the cardioplegic solutions used in phase III studies Component

Group 1

Group 2

Group 3

Group 4

NaCl Na aspartate NaHC03 KC1 MgC1 z CaCh Total Na content

110.0 0 10.0 16.0 16.0 1.2 120.0

90.0 20.0 10.0 16.0 16.0 1.2 120.0

130.0 0 10.0 16.0 16.0 1.2 140.0

110.0 20.0 10.0 16.0 16.0

tivein experimentaland clinicalstudies.i- 3 the techniques are complex, may increasethe risk of infection, and may promote undesirable effects, such as the activation of various blood components. For these and other reasons, singleinfusion of cold cardioplegic solutions followed by hypothermic storage remains the most widely used procedure. Considerableeffortsare thus being made to identify additivesthat, whenadded to conventional cardioplegicsolutions, willextendthe duration of their efficacy well beyondthe current 4- to 5-hour limit. In this connection, severalstudies"? suggestthat amino acids,such as glutamate and aspartate, can improvethe recovery of cardiac performance after ischemia and reperfusion. Most have used short periods of reperfusion, however, and consequently it has not been possible to be certain whether the improved recovery is transient (i.e., overcomes myocardial stunning) or sustained (i.e., reduces cell death). For this reason, the present studies were designed to exploit the ability to assess the function and protection of the heterotopically transplanted heart 24 hours after a period of ex vivoischemia and storage," We choseaspartate with a view to defining the extent of the protection afforded and its dose-response characteristics. Materials and methods Male homozygous Lewis (phase I studies) or Wistar (phases II, III, and IV studies) rats, weighing 300 to 350 gm were used. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH Publication No. 80-23, revised 1978). Phase I: Studies of the effect of monosodium aspartate on functional and metabolic recovery of the heart after 8 hours of hypothermic storage and 24 hours of transplantation Surgical procedures. The experimental preparation and techniques have been described in detail previously. 8 EXCISION AND STORAGE OF HEARTS. Homozygous donor Lewis rats (n = 8 per group) were anesthetized with sodium pentobarbital (60 mg/kg, intraperitoneally) and their lungs were mechanically ventilated, through a tracheostomy, at a rate of 55 strokes/min with a ventilation pressure of 12 to 14mm Hg.

1.2

140.0

The chest was opened and the venae cavae and pulmonary veins were isolated; heparin sodium (1000 IV/kg) was administered intravenously. After this both venae cavae and the pulmonary veins were ligated and the heart was excised and placed in cold (4 0 C) saline. The aorta was rapidly cannulated, and the St. Thomas' Hospital cardioplegic solution (STH), containing NaCl, llO.Ommol/L; KCl, l6.0mmol/L;MgCh, l6.0mmol/ L; CaCh, 1.2 mmol/L; and NaHC0 3, 10.0 mmol/L (pH 7.8), with or without added aspartate (20 rnmol/L; L-isomer monosodium salt), was immediately infused at a constant pressure (equivalent to 60cm H20) for 2 minutes at4° C. The heart was then stored by immersion in the same infusion solution for 8 hours at 4 0 C. HETEROTOPIC TRANSPLANTATION. The abdomen of an anesthetized recipient rat was opened through a midline incision,and the abdominal aorta and inferior vena cava were exposed. The stored donor heart was removed from the hypothermic chamber, and, with a 9-0 Ethicon suture (Ethicon, Inc., Somerville, N.J.), its aorta and pulmonary artery were anastomosed (end to side) to the abdominal aorta and the inferior vena cava, respectively, of the recipient rat, in the manner described by Ono and Lindsey? The heart was wrapped in a wet swab and regularly irrigated with cold (4 0 C) saline to maintain its temperature at 100 to 14 0 C during implantation. The duration of the implantation period was standardized at 45 minutes for all experiments. Subsequently the hearts were reperfused and the abdominal wall closed in two layers. The animals were allowed to recover and were maintained unrestrained with normal feeding for 24 hours. Posttransplantation in vitro assessment. Twenty-four hours after transplantation the recipient rats were again anesthetized, the right femoral vein was exposed, and heparin sodium (1000 IV /kg) was administered intravenously. The abdomen was then opened and the transplanted heart removed and placed in cold (4 0 C) saline. The aorta was then rapidly cannulated, and each heart was perfused aerobically in the Langendorff mode 10 at 37 0 C with perfusion fluid!' (containing glucose, 11.1 mmol/L; NaC!, 118.5 mrnol/L; KCl, 4.75 mmol/L; MgS04, 1.19 mmol/L; KH 2P04 , 1.18 mmol/L; NaHC0 3, 25.0 mmol/L; CaCh, 1.36 mmol/L; and pH 7.4 when gassed with 95% 02 plus 5% CO 2) at a constant perfusion pressure (lOa cm H20). Before use, all perfusion fluids were filtered through a 5.0 /Lm porosity filter to remove any particulate matter. All hearts were paced with the use of right atrial pacing at a constant rate of 320 beats/min. After 5 minutes of perfusion, a balloon catheter was inserted into the left ventricle via the left atrium; this was then filled with a volume of saline sufficient to produce a constant left ventricular end-diastolic pressure equivalent to 4 mm Hg. After

Volume 103 Number 3 March 1992

20 minutes of perfusion, coronary flow was measured and function curves ofleft ventricular developed pressure were constructed by progressively increasing the volume of the balloon to achieve end-diastolic pressures of 4, 8, 12, 16, and 20 mm Hg. Hearts were then freeze-clamped with stainless steel tongs cooled to the temperature of liquid nitrogen, and the frozen tissue was taken for the analysis of myocardial content of adenosine triphosphate (ATP) and creatine phosphate (CP).'2 For comparative purposes, aerobic measurements of control cardiac function were also made in fresh, nontransplanted hearts (n = 8) that had not been rendered ischemic. Phase II: Studies of the dose-response characteristics of monosodium aspartate as an additive to the cardioplegic solution. Anesthetized Wistar rats (n = 8 per group) were given heparin sodium (1000 IU /kg) via the femoral vein, the chest was opened, and the heart was excised and placed in cold (4° C) saline. The aorta was cannulated and immediately infused with STH containing various doses (0, 5, 10, 20, 30, 40, or 50 mmol/L) of monosodium aspartate. Temperature, perfusion pressure, and time of infusion were identical to those described in the preceding phase I studies. Hearts were stored for 8 hours at 4 ° C in the same infusion solution but were then reperfused in the Langendorff mode for IS minutes, during which time the coronary effluent was collected and taken for the determination of total creatine kinase (CK) leakage. I 3 This was followed by a 2Q..minute period of working (ejecting) perfusion.lv 15 during which time the recoveries of various indices of cardiac function (coronary flow, aortic flow, cardiac output, heart rate, peak aortic pressure, stroke volume, and stroke work) were measured. At the end of each experiment, the hearts were freeze-clamped and taken for metabolic determinations as described earlier. For comparative purposes, a group of hearts (n = 8) that had not been rendered ischemic were perfused aerobically in theLangendorff mode for IS minutes and for 20 minutes in the working mode. Phase III: Studies of the relative effects of sodium and aspartate in the cardioplegic solution. The experimental protocol for these in vitro studies was identical to that used in phase II. Four groups of hearts (n = 8 per group) were investigated: (I) a control group with a preischemic infusion of unmodified STH (i.e., aspartate free and a total sodium concentration of 120 mmol/L); (2) STH with added monosodium aspartate (20 mmol/L), but with sodium chloride reduced by 20 mmol/L, so as to givea final sodium concentration of 120 mrnol/L; (3) STH without aspartate but with an additional 20 mmol/L of sodium chloride to give a final sodium concentration of 140 rnmol/L; (4) STH with added monosodium aspartate (20 mmol/L), i.e., a final sodium concentration of 140 mmol/L. The complete formulation of each of these four solutions is shown in Table I. Phase IV: Studies of the filling rate of the isolated perfused heart. Hearts were excised from anesthetized, heparinized Wistar rats and immediately cannulated via the aorta to and perfused with standard perfusate as described in the previous sections.After 5 minutes of perfusion, during which time the left atrium was cannulated, the heart was converted to the working mode, and preischemic control function was measured for 25 minutes. In one group of hearts (n = 8), atrial filling pressure was set at 20 em H20 and perfusion fluid allowed to flow freely through a 2.5 mm catheter throughout the 25-minute preischemic working period. In a second group of hearts (n = 8), again with the atrial reservoir set to provide a filling pressure of

Sodium aspartate and cardiac preservation 5 2 3

A c; 160 ::J:

- E ZE

140

I-w

_;:!: 0:

120

~~~ ::J:c.. w OwO:

100

I-O:W WI-c.. ~~g Ww 0> c..w

60

Q~

Oz=>

!q~~

~

t==: =:*

80

-

*

*

~

~

- 0 - - AEROBIC CONTROL

40

-0--

20

0

>

-'

4

8

5TH ALONE 5TH+A5PARTATE 12

16

20

LV END-DIASTOLIC PRESSURE (mmHg)

B

z_

OO!: i=E <"" -IE

~Z-

14 12 10

~~~ ::J: w-, ~zLl.. ,<>-

1-0:0:

8~~ O-wo 00: 0-0 0

STH ALONE

c

STH+ASPARTATE

~ ~

STH ALONE

STH+ASPARTATE

Fig. 1. Left ventricular developed pressure (A~ coronary flow (B), and ATP and CP content (C) after 20 minutes of in vitro aerobic perfusion. Before perfusion, hearts had been subjected to cardioplegia (2 minutes) and 8 hours of cold (4° C) ex vivo storage with the use of STH with or without sodium aspartate (20 mmol/L), 45 minutes of transplantation, and 24 hours of reperfusion in vivo. Each point represents the mean of eight measurements in eight hearts, and the bars represent the standard error of the mean. *p < 0.05 when compared with the aspartate-free group.

20 ern H 20, unrestricted perfusion was permitted for 10 minutes after which the atrial cannula was constricted with a screw clamp to reduce the atrial inflow (as measured by the cardiac output) from its usual value (in excess of 80 ml/rnin) to less than 50 ml/min for the remaining 15 minutes of preischemic aero-

524

The Journal of Thoracic and Cardiovascular Surgery

Galihanes, Chambers, Hearse

B

A 24

40

C

:§

3:

16

:§.

u,

12

0

s

ir -c

3:

20

~

u,

z

8

o<..>

4

oa:

30

<..>

~

a:

10

0

-c

o

5

I ASPARTATE

10

20

30

40

50

CONCENTRATION 1m mol/I)

o

I

50

I-

40

C

300

~ e'"

200

E

::;)

c...

w

30

I-

«

20

-c a:

150

<..>

I-

100

a:

10

0

0

-c

a: «w

o

5

I ASPARTATE

10

20

30

40

50

CONCENTRATION (mmolll)

o

I ~ 25

g'"

0.20

~ 20

w

0.15

20

30

40

50

CONCENTRATION (mmolll)

I

15

~

~

'g,

0.10

w

a:

10

E

:3 c:

::;)

:>:: 0

I

'" ~

~

>

50

F

E

0

5

I ASPARTATE

0.25

:::;;

40

50

I

<..>

.s

30

250

::;)

I-

20

CONCENTRATION Immol/I)

350

C

:§.

10

D

C 60

:§

5

I ASPARTATE

10

:>::

~ 3:

0.05

I-

en 0.00

5

w

o

:>:: 5

I ASPARTATE

10

20

30

40

50

CONCENTRATION Immol/I)

I

oa:

I-

en

o

5

I ASPARTATE

10

20

30

40

50

CONCENTRATION 1m mol/I)

I

Fig. 2. Relationship between amount of sodium aspartate added to STH and postischemic recovery of cardiac function. Hearts were subjected to cardioplegia (2 minutes) and 8 hours of cold (4° C) storage with useofSTH containing various concentrations of sodium aspartate (0, 5, 10, 20, 30, 40, or 50 mmol/L), Coronary flow (A), aortic flow (B~) cardiac output (C), heart rate (D~ stroke volume (E~ and stroke work (F). *p < 0.05 when compared with the aspartate-free group. bic perfusion. All hearts were then arrested with a 2-minute infusion of normothermic (37° C) STH (infusion pressure equivalent to 60 ern H20); this was followed by 30 minutes of normothermic global ischemia (37° C), after which the hearts were reperfused in the Langendorff mode (at a pressure of 100 em H20) for 15 minutes, and total CK leakage was measured 13 in the collected coronary effluent. The hearts were then converted back to the working mode, with the restriction to atrial inflow maintained in one group for 20 minutes before the clamp was removed and free atrial inflow permitted for a further 10

minutes of perfusion. The postischemic recovery of function was measured in each group of hearts and expressed as a percent of the preischemic function in the 5-minute period before the onset of ischemia (i.e., during the period in which atrial inflow was restricted in one group). Expression of results and statistical analysis. In Langendorff-perfused hearts, with intraventricular balloons, left ventricular developed pressure was expressed in millimeters of mercury and calculated as the difference between peak systolic pressures and a specified left ventricular end-diastolic pressure.

Volume 103 Number 3

Sodium aspartate and cardiac preservation

March 1992

A ~

e-

"0

~

'E Il)

~

UJ C)

«: >c:: LiS

~

100

~ "0

80 60

en

30

UJ I-

-c

40

:I:

a.

en 0

:I:

20

a.

20 10

>-

.....J

(!)

a:

>c::

U

40

E ~

I1IlI0' ~ ATPI

B 50

~ "0

525

0

5

I ASPARTATE

10

20

30

40

50

CONCENTRATION (mmol/I)

UJ

I

zUJ

:I:

52

0

5

I ASPARTATE

10

20

30

40

50

CONCENTRATION (mmol/I)

I

:I:

Fig. 3. Relationshipbetweenamount of sodium aspartate added to STH and postischemic (A) CK leakage and (D) tissuehigh-energyphosphatecontent. Hearts were subjected to cardioplegia(2 minutes) and 8 hours of cold (4° C) storage with the use of STH containingvariousconcentrationsof sodiumaspartate (0, 5, 10,20,30,40, or 50 mmolj L).

Coronaryflow (mljmin) was measured by timed collection of the coronary effluentand heart rate (paced at 320 beatsjmin) confirmed from the pressure recording. In hearts perfused in the workingmode, coronary flow (ml/ min), aorticflow (ml/rnin), peak aortic systolic pressure (mm Hg), and heart rate (beats/min) were recorded as previously described'"; cardiac output (rnl/rnin) wascalculated by adding aorticflow to coronary flow, stroke volume (rnl/beat) by dividing cardiac output by heart rate, and stroke work (l05 dynes . mm Hg/beat) by multiplying stroke volume by peak aortic systolic pressure. CK leakage was expressed as IV j 15 minigm dry weight,and ATP and CP content as ~moll gm dry weight. All results are expressed as mean ± standard error of the mean.The two-tailed unpaired t test was used for comparison between two means. Analysis of variancewas used for comparison of more than two means; when a significantF value was obtained, comparisonsbetween the untreated and each of the treatedgroups wereconductedby the two-tailedDunnett's test. A difference wasconsidered statisticallysignificant at a p value less than 0.05.

Results Phase I: Effect of monosodium aspartate on the functional and metabolic recovery of the heart after 8 hours of hypothermic storage and 24 hours of transplantation. The addition of monosodium aspartate in a concentration of 20 mmol/L to STH gave a small but

statistically significant (p < 0.05) improvement in recovery of contractile function after 8 hours of hypothermic storage and 24 hours of transplantation (Fig. 1, A). Thus, at a left ventricular end-diastolic pressure of 16 mm Hg, left ventricular developed pressure was 82 ± 3 mm Hg in the STH plus aspartate group, compared with 70 ± 2 mm Hg (p < 0.05) in the STH alone group. Aerobically perfused control hearts (no storage) had a left ventricu-

lar developed pressure of 141 ± 4 mm Hg at a left ventricular end-diastolic pressure of 16 mm Hg. Thus it can be calculated that the addition of aspartate resulted in an improvement in recovery from 50% to 58%. Postischemic coronary flow (Fig. 1, B) was similar in both groups of hearts (l0.7 ± 0.4 ml/rnin in the STH alone group versus 11.6 ± 0.4 ml/ min in the STH plus aspartate group; p = NS*). These values were significantly (p < 0.05) reduced from that of the aerobically perfused (i.e., nonischemic) hearts (14.3 ± 0.5 ml/rnin). Myocardial A TP and CP content (Fig. 1, C) were similar in both the STH alone and the STH plus aspartate groups. However, the A TP content in each group (11.0 ± 0.6 and 11.5 ± 0.9,umol/gmdryweight,respectively) was significantly (p < 0.05) reduced when compared with control hearts that had not been ischemic or transplanted (19.7 ± 0.2 ,£tmol/gm dry weight). CP content (26.8 ± 4.1 and 28.3 ± 2.9 ,umoljgm dry weight, respectively) was, however, similar to that in the aerobic control hearts (31.4 ± 3.7 umol/gm dry weight; p =NS).

Phase II: Studies of the dose-response characteristics of monosodium aspartate as an additive to the cardioplegic solution. The effect of the addition to STH of various concentrations of sodium aspartate is shown in Fig. 2. The postischemic recoveries of several indices of cardiac function exhibited bell-shaped dose-response profiles. For aortic flow, cardiac output, stroke volume, and stroke work, maximum recovery was observed with a 20 mrnol/L concentration of sodium aspartate. Thus postis*NS

= Not significant.

526

The Journal of Thoracic and Cardiovascular

Galihanes, Chambers, Hearse

Surgery

"2

E

"" .§.. 3:

20 16

0

..J

u,

>a:

12

40

0

a:

0

u

"2

"2

:§

:§. 3: 0

..J

u,

~

a:

4

0

-c

"2

'E 300 !!l 250

I-

40

a..

20

~

o u

-c

10

~

-c

0.25

200

1ij

'"

.g 0.20 .§.. w 0.15 :::;: ::::> ..J 0 0.10 >

«

150 100

w

:J:

50

I-

a: -c w

I-

::::>

E

I-

a:

50

30 20 10

u

350

w

:§

:§.

30

D

'" e'"

60

u

-c z

c

B

A 24

~

0

a:

0.05

1ij

F

'"

~

:J:

E

~ U)

c: '"

25 20

e-,

"0

on C>

:s

15

10 5

en

Fig. 4. Relationship between postischemic recovery ofcardiac function andsodium or aspartate (or both) concentration ofSTH. Heartswere subjected to cardioplegia (2 minutes) and8 hours ofcold (4° C)preservation with the use ofSTH containing various concentrations ofsodium or aspartate (or both) (fordetails, seetext). Coronary flow (A), aortic flow (B), cardiac output (C), heartrate (D), stroke volume (E), and stroke work (F). *p < 0.05 when compared with unmodified cardioplegic solution (i.e., 120 mmol/L Na). chemic aortic flow was 14.3 ± 4.3, 14.8 ± 2.3, 22.3 ± 4.2, 28.4 ± 4.6, 19.0 ± 4.7, 12.9 ± 3.3, and 10.5 ± 3.5 mljmin in the 0, 5, 10,20, 30, 40, and 50 mmoljL aspartate groups, respectively. Comparison of postischemic values for aortic flow, cardiac output, stroke volume, and stroke work in the stored hearts with those in the aerobic (nonischemic) control groups (70.0 ± 1.1 ml/rnin, 92.6 ± 1.2 ml/rnin, 0.34 ± O.oI ml/beat, and 43.6 ± 1.2 105 dynes. mm Hg/beat, respectively) revealed comparable percentage recoveries to those observedin the phase I studies. Thus, in the 20 mmol/L aspartate group, aortic flow, cardiac output, stroke volume,and stroke work recoveredto 41%, 52%, 56%, and 47%, respectively, compared with 20%, 34%, 32%, and 29%, respectively, in the aspartate-free group. The recoveryof coronary flow (Fig. 2, A) and heart rate (Fig. 2, D) were little affected by the addition of aspartate. Although CK leakage (Fig. 3, A) showeda reciprocal bell-shaped profile in comparison with the functional results (with hearts in the 20 and 30 mmol/L aspartate

groups having the lower CK release), this reduction was not statistically significant when compared with that in the untreated group (36 ± 5 and 33 ± 5 1U/15 min/gm dry weight versus 54 ± 7 IV /15 minigm dry weight, respectively; p = NS). As observedin phase I studies,there werenosignificant differences in the myocardial ATP content (Fig. 3, B) in any of the groups; however, the values represented 65% to 70% of that in the aerobic control group (19.7 ± 0.3 Ilmol/gm dry weight). CP content was also similar in all groups (Fig. 3, B); these values were greater than that in the aerobic control group (30.4 ± 3.6 Ilmol/gm dry weight). Phase III: Studies of the relative effects of sodium and aspartate in the cardioplegic solution. The results presented in Fig. 4 show the effect of the presence and absence of aspartate (20 mrnol/L) in the STH with and without correction for the sodium that is included when monosodiumaspartate is added to the STH. Taking aortic flow (Fig. 4, B) as an example, addition of sodium aspartate 20 mmoljL (final sodium concentration = 140

Volume 103 Number 3 March 1992

mmoljL) resulted in a significant improvement in postischemic function (28.5 ± 1.8 mljmin versus 19.3 ± 1.6 mljmin in the unmodified STH group). However, if sodium chloride 20 mmoljL was added instead of sodium aspartate (final sodium concentration again = 140 mmoljL), the postischemic recovery of aortic flow was 29.4 ± 4.1 mljmin. Thus it would appear that it is the addition of sodium rather than aspartate that enhances functional recovery. This was confirmed in studies in which a 20 mmoljL concentration of sodium aspartate was added to the STH, but at the same time the sodium chloride concentration was reduced by 20 mmoljL; under these conditions the postischemic recovery of aortic flow was 23.8 ± 3.1 mljmin, a value that did not differ significantly from that achieved with unmodified STH. The results for aortic flow were reflected in other indices of cardiac function, such as cardiac output, stroke volume, and stroke work. Also, in agreement with the results in phase II, we observed that the manipulations had relatively little effect on heart rate or coronary flow. CK leakage (Fig. 5) tended also to reflect the results for aortic flow;however, the differences did not achieve a level of statistical significance. Phase IV: Studies of the filling rate of the isolated perfused heart. Although our results support the claim 7 that the addition of sodium aspartate to STH increases its protective properties, the improvement we observed was relatively small (e.g., Fig. 4, C, shows that postischemic recovery of cardiac output was 35.0 ± 2.2 mljmin with STH alone and 46.0 ± 2.7 mljmin with STH plus aspartate). We were concerned that the degree of benefit observed in our present studies was much less than that described in a very similar model by Choong and Gavin'? These investigators, like ourselves, used the isolated working rat heart, a 20 rnrnol/L concentration of sodium aspartate, and long periods of hypothermic ischemic storage (8 and 10 hours). Whereas we observed only 30% to 40% improvement (in postischemic aortic flow) with aspartate, Choong and Gavin 7 reported a 100% improvement. In an attempt to understand the reason for this difference, we noted that in Choong and Gavin's study, despite similar filling pressures and afterloads, the absolute level of preischemic cardiac function was much lower than ours. Thus mean aortic flow in our preparation was 70.0 ± 1.1 mljmin (aerobic control in phase II studies), whereas Choong and Gavin reported a value of only 27 ± 2 mljmin. A possible reason for the low aortic flowin the Choong and Gavin study could be the presence in their apparatus of some factors (such as the diameter of their cannula) that limited atrial inflow (despite a 20 em H20 filling pressure) and, hence, aortic output. Under such conditions the recovery of cardiac output may be

Sodium aspartate and cardiac preservation

¥

527

100

Fig. 5. Relationship between postischemic CK leakage and sodium or aspartate (or both) concentration of 8TH. Hearts weresubjectedto cardioplegia (2 minutes) and 8 hours of cold (4° C) preservation with the use of STH containing various concentrations of sodiumor aspartate (or both) (for details,see text).

limited, not by the functional capabilities of the heart, but by the ability of the apparatus to deliver sufficient atrial inflow. We explored this possibility by designing a protocol in which, after establishing control cardiac function, a limitation on atrial inflow was imposed, such that aortic flow fell to the levels reported by Choong and Gavin'? Using this preparation, we compared the recovery of cardiac function after 30 minutes of normothermic ischemia in the rat hearts with and without the limitation of inflow. Fig. 6 shows the preischemic and postischemic values for functional capacity in hearts with unrestricted atrial inflow versus those in which flow was limited for the last 15 minutes of the preischemic working period. Taking aortic flow as illustrative of other indices of cardiac function, it can be seen that in the unrestricted inflow group aortic flow was 65.6 ± 3.0 mljmin before ischemia and 40.1 ± 4.1 mljmin after ischemia; thus the postischemic recovery was 61% ± 5%. In the flow-restriction group, aortic flow before constriction was comparable (74.3 ± 1.7 mljmin) to that in the unrestricted group; however, when inflow was restricted, aortic flow fell to 27.1 ± 0.4 mljmin. Postischemic recovery (with the restriction still in place) was 22.9 ± 0.9 mljmin (i.e., a recovery of 85% when compared with the preischemic constricted flow control); however, on release ofthe clamp these hearts recovered to 57.0 ± 4.6 mljmin. As is discussed later, we believe that such an effect could account for the apparently excellent protection afforded by sodium aspartate in the Choong and Gavin study.? Discussion The present studies demonstrate that (1) the addition of monosodium aspartate (20 mmoljL) to the STH can

528

The Journal of Thoracic and Cardiovascular Surgery

Galihanes, Chambers, Hearse

A

-

80

c: "E

--0--

!

60

0

40

:;: -' u,

c:

cc 0

20

zs

CC

10

20

30

40

50

TIME (min)

60

70

80

Q)

~

10

20

30

10

20

30

40

50

60

70

80

90

40

50

60

70

80

90

TIME (min)

D 80

::c E

~

0.3

'" c: e-,

60

Q)

UJ

-0

0.2

on 0

:s ::.:: cc 0 :;:

> 0

0 0

-g-

Q)

cc fen

20

90

-g-

UJ

40

o

C

::.::

:::> 0

-c

0.4

:::;;

60

«

0

:::> -' 0

:::>

c..>

0

oS

f-

0... f-

-c

g

80

~

c..>

i=

~

0.1

40

20

UJ

::.:: 0

0.0 0

10

20

30

40

50

60

70

80

90

TIME (min)

cc fen

0 0

TIME (min)

Fig. 6. Effect of atrial inflow rate on preischemic and postischernic cardiac function. Aortic flow (A),cardiac output (B), stroke volume (C), and stroke work (D).

improve the recovery of function of the rat heart after long-term hypothermic preservation and that this improvement is maintained even after 24 hours of reperfusion; (2) the observed beneficial effect can be explained by the increase in sodium concentration of the cardioplegic solution rather than by the effects of the aspartate anion; (3) the functional improvement observed in the early and late reperfusion periods was not associated with an improvement in high-energy phosphate content; (4) the construction of the isolated working heart apparatus can influence the apparent efficacy of protective interventions. A number of aspects of our results warrant further discussion. Amino acids and tissue protection. There is a growing number of reports in the literature describing the beneficial effects of exogenous amino acids during myocardial ischemia/reperfusion and hypoxia/reoxygenation. Choong and Gavin," for example, reported that sodium aspartate, when added to the STH, improved postischemic functional recovery in the rat heart. Rau and coworkers" have shown that arginine, ornithine, glutamate, and aspartate can improve posthypoxic function in the rabbit heart. Matsuoka and coworkers'> have also shown protection with glutamate in the rabbit heart, and similar results have been reported in other species,

including sheep' and dogs." Our results in the rat are in broad agreement with these studies. . Possible mechanisms. Amino acids may be involved in numerous aspects of intermediary metabolism. In addition to acting as precursors of protein synthesis, some amino acids (e.g., glutamate and aspartate) can provide (by means of transamination and the malate-aspartate shuttle) a-ketoglutarate, a substrate for the energy-producing citric acid cycle. Amino acids can also exert buffering properties, and some investigators'? have suggested that they can promote anaerobic ATP production in the mitochondria. Enhancing energy availability? Although amino acids have been found to be effective when used at substrate concentrations (i.e., in the millimolar range), Rau and coworkers" have argued that they do not act by the simple provision of substrate, protein synthesis, enhanced production of adenine nucleotide, or buffering. Rather, they suggested that they act through anaerobic intermediary metabolic reactions, particularly those involving transamination and the malate-aspartate shuttle." Although evidence for an involvement of this transmitochondrial shuttle is strong, the precise mechanism by which it affords protection remains to be defined and may not necessarily include anaerobic ATP production. Thus,

Volume 103 Number 3 March 1992

although Choong and Gavin 7 showed that aspartate (but not glutamate) increased myocardial ATP content, our results showed no significant protective effect in terms of energy metabolism. Like Choong and Gavin," however, we observed improved postischemic function in hearts that had received aspartate-containing cardioplegic solution. Possible alternative mechanisms? Two aspects of our study have led us to question the aspartate energy hypothesis. First, as stated, we were unable to demonstrate any significant improvement in intracellular energy status. Second, although aspartate was effective at a concentration of 20 mmol/L, it was ineffective at both higher and lower concentrations. This narrow range of efficacy is not consistent with a simple metabolic involvement. Bell-shaped dose-response curves are frequently observed for protective interventions that involve ionic manipulation. Thus we have shown that the protective properties of the STH solution are exquisitely sensitive to changes in the content of ions such as calcium, 17 magnesium, 18 and sodium. 19 In the case of calcium, a change of as little as 0.2 mmol/L on either side of the optimum can reduce by "50% the postischemic recovery of contractile function. Changes in sodium concentration of 20 to 30 mrnol/L can have a dramatic effect on postischemic leakage of CK and function. 19 As far as we can ascertain, all published studies of protection with aspartate have either used the monosodium compound as the source and have failed to take account of the commensurate increase in extracellular sodium, or they have not specified the source at all. In the present studies we have specifically addressed this issue and have demonstrated that adding aspartate without increasing sodium results in the loss of the apparent protection afforded by aspartate, and that increasing sodium in the absence of aspartate does afford protection. Weare therefore forced to conclude that the protective properties of sodium aspartate lie predominantly in its cationic component rather than in the organic anion. It should, however, be conceded that indirect evidence against our hypothesis comes from the studies of Rau and coworkers," who demonstrated that the n-isomers of the amino acids were less efficacious than the L-isomers. Sodium and tissue injury. If sodium is the active principle in the protection afforded by sodium aspartate, the present studies demonstrate a bell-shaped dose-response curve over the range of 120 to 170 mmol/L, with 140 mmol/L being the optimum. In previous studies.!? with cardioplegic solutions of varying compositions, we have shown that changes in extracellular ionic composition during ischemia can dramatically influence the extent of

Sodium aspartate and cardiac preservation

529

tissue injury. Thus modifying sodium in the range of 30 to 150 mmol/L produced a bell-shaped relationship for the leakage of CK, with 120 mmol/L being the optimum concentration. As the sodium concentration was decreased from 150 to 120 mmol/L, the leakage of CK fell from approximately 10 to 7.5 IV/15 min/heart, and when it was reduced from 120 to 30 mmol/L the leakage of enzyme almost tripled (20 IV/15 min/heart). In these previous experiments we did not study concentrations of sodium between 120 and 150 mmol/L, There are a number of possible explanations for the sodium dose-response curve. First, Tani and N eely20 have shown that the intracellular sodium content at the end of ischemia is an important determinant of reperfusion-induced uptake of calcium and ischemia/reperfusion-induced injury. Thus excessive extracellular concentrations of sodium during ischemia may exacerbate sodium overload and hence promote calcium overload on reperfusion. 20-24 Conversely, insufficient extracellular concentrations of sodium during ischemia may promote calcium overload during ischemia by altering the thermodynamic equilibrium of the Na+-Ca 2+ exchange mechanism. However, calcium overload induced by that mechanism may be limited to early ischemia, since it has been suggested 25,26 that later in ischemia the fall in intracellular pH inhibits Na+-Ca 2+ exchange. Whatever may be the optimal concentrations for extracellular sodium and calcium during ischemia and reperfusion, there can be no doubt that a variety of highly sensitive and dynamic ionic mechanisms exist. The delicate balance between these mechanisms is easily disturbed, and, therefore, even minor changes in the ionic composition of cardioplegic solutions can be expected to have a profound effect on their protective properties. We do acknowledge, however, that, although these relatively small changes in sodium concentration appear to be of relevance in myocardial protection with the STH, this may not necessarily apply to other crystalloid or blood-based cardioplegic solutions. Importance of the design of the apparatus. Recently Choong and Gavin 7 demonstrated that, after storage in STH, for 8 hours at 4 0 C, rat hearts exhibited 100% recovery of function (aortic flow). After 10 hours of storage, however, recovery was only 16%; this was improved to 99% by the addition of a 20 mmol/L concentration of aspartate to the cardioplegic solution. These results are in considerable contrast to those obtained in the present study, in which the recovery of aortic flow in hearts subjected to 8 hours of global ischemia at 4 0 C was only 30% and was improved by aspartate to only 40%. In attempting to explain these striking interlaboratory differences, we noted that the preischemic aortic flow in the study of Choong and Gavin? was only 27 ± 2 mljmin, whereas in

530

Galihanes, Chambers, Hearse

our studies, with similar hearts and under comparable conditions, that value was 70.0 ± 1.1 mljmin. However, when the postischemic values for aortic flow in both studies are compared in absolute rather than percentage terms the values are remarkably similar (about 28 mljmin in the Choong and Gavin study and 22.9 ± 0.9 ml/rnin in ours). A significant difference in preischemic aortic flow could arise if there were a restriction to filling from the left atrial reservoir. This would have the effect of creating an artifactually low preischemic value with which to compare postischemic recovery (when ischemic injury may have reduced the filling requirements of the heart and hence effectively eliminated the impact of the restriction to inflow). Under these conditions a poor postischemic level offunction, when expressed in absolute terms, would become a high recovery. We tested this possibility by restricting atrial inflow and found that aortic flow recovered to 85% ± 4% (compared with the 61% ± 5% recovery in control hearts; p < 0.05). The possibility arises that the protective effects of sodium aspartate may have been overestimated in some studies. Concluding comments. Our studies with the isolated perfused heart demonstrate that the beneficial effects of adding monosodium aspartate to a cardioplegic solution are predominantly due to the increase in sodium concentration and not to the addition of aspartate. We have not been able to determine whether this phenomenon applies in vivo. Nonetheless, our results show that it is possible to make a substantial improvement to the protective properties of the STH when used for long-term hypothermic preservation. Of particular importance is that the additional protection is maintained throughout at least 24 hours of reperfusion. We thank Dr. M. J. Shattock for his comments. REFERENCES 1. Heck CF, Shumway SJ, Kaye MP. The registry of the International Society for Heart Transplantation: sixth official report-1989. J Heart Transplant 1989;8:271-6. 2. Wicomb WN, Cooper DKC, Barnard CN. Twenty-fourhour preservation of the pig heart by a portable hypothermic perfusion system. Transplantation 1982;34:246-50. 3. Wicomb WN, Cooper DKC, Novitsky D, Barnard CN. Cardiac transplantation following storage of the donor heart by a portable hypothermic perfusion system. Ann Thorac Surg 1984;37:243-8. 4. 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 1979; 236:H873-9. 5. Gharagozloo F, Melendez FJ, Hein RA, et aL The effect of amino acid L-glutamate on the extended preservation ex

The JOurnal of Thoracic and Cardiovascular Surgery

vivo of the heart for transplantation. Circulation 1987; 76(Pt 2):V65-70. 6. Rosenkranz ER, Okamoto F, Buckberg GD, Robertson JM, Vinten-Johansen J, Bugyi HI. Safety of prolonged aortic clamping with blood cardioplegia. III. Aspartate enrichment of glutamate-blood cardioplegia in energy-depleted hearts after ischemic and reperfusion injury. J THoRAC CARDIOVASC SURG 1986;91:428-35. 7. Choong YS, Gavin JB. L-Aspartate improvesthe functional recovery of explanted hearts stored in St. Thomas' Hospital cardioplegic solution at 4 0 C. J THORAC CARDIOVASC SURG 1990;99:510-7. 8. Galiiianes M, Hearse DJ. Metabolic, functional, and histologic characterization of the heterotopically transplanted rat heart when used as a model for the study of long-term recovery from global ischemia. J Heart Lung Transplant 1991;10:79-91. 9. Ono K, Lindsey ES. Improved technique of heart transplantation in rats. J THORAC CARDIOVASC SURG 1969; 57:225-9. 10. Langendorff O. Untersuchungen am uberlebenden Saugethierherzen. Pflugers Arch 1895;61:291-332. II. Krebs HA, Henseleit K. Untersuchungen uber die Harnstoffbildung im Tierkorper. Hoppe Seylers Z Physiol Chern 1932;210:33-66. 12. Hearse DJ. Microbiopsy metabolite and paired flow analysis:a new rapid procedure for homogenisation, extraction and analysis of high energy phosphates and other intermediates without any errors from tissue loss. Cardiovasc Res 1984;18:384-90. 13. Urdal P, Stromme JH. Effects of Ca, Mg and EGTA on creatine activity in cerebrospinal fluid. Clin Chern 1979; 25:147-50. 14. Hearse DJ, Braimbridge MV, Jynge P. Protection of the ischemic myocardium: cardioplegia. New York: Raven Press, 1981;50-92. 15. Matsuoka S, Jarmakani JM, Young HH, Uemura S, Nakanishi T. The effect of glutamate on hypoxic newborn rabbit heart. J Mol Cell Cardiol 1986;18:897-906. 16. Penny DG, Cascarano J. Anaerobic rat heart: effects of glucose and tricarboxylic acid cycle metabolites on metabolism and physiological performance. Biochem J 1970; 118:211-27. 17. Yamamoto F, Braimbridge MV, Hearse DJ. Calcium and cardioplegia: the optimal calcium content for the St. Thomas' Hospital cardioplegic solution. J THORAC CARDIOVASC SURG 1984;87:908-12. 18. Hearse DJ, Stewart DA, Braimbridge MV. Myocardial protection during ischemiccardiac arrest: the importance of magnesium in cardioplegic infusates. J THORAC CARDIOVASC SURG 1978;75:877-85. 19. Hearse DJ, Braimbridge MV, Jynge P. Protection of the ischemic myocardium: cardioplegia. New York: Raven Press, 1981:209-99. 20. Tani M, Neely RN. Role of intracellular Na+ in Ca2+ overload and depressed recovery of ventricular function of

Volume 103 Number 3 March 1992

reperfused ischemic rat hearts: possible involvement of H+-Na+ and Na+-Ca2+ exchange. CircRes 1989;65:104556. 21. Lazdunski M, Frelin C, Vigne P. The sodium/hydrogen exchange system in cardiac cells:its biochemical and pharmacological properties and its role in regulating internal concentrations of sodium and internal pH. J Mol Cell CardioI1985;17:1029-42. 22. Reagan TJ, Broisman L, Haider B, Eaddy C, Oldewurtel HA. Dissociation of myocardial sodium and potassium alterations in mild versus severe ischemia. Am J Physiol 1980;238:H575-80. 23. Renlund DG, Gerstenblith G, Lakatta EG, Jacobus WE,

Sodium aspartate and cardiac preservation

531

Kallman CH, Weisfeldt ML. Perfusate sodium during ischemia modifies post-ischemic functional and metabolic recovery in the rabbit heart. J Mol Cell Cardiol 1984; 16:795-801. 24. Grinwald PM, Brosnahan C. Sodium imbalance as a cause of calcium overload in post-hypoxic reoxygenation injury. J Mol Cell CardioI1987;19:487-95. 25. Coffe SM, Poole-Wilson PA. The time of onset and severity of acidosis in myocardial ischemia. J Mol Cell Cardiol 1980;12:745-60. 26. Philipson KD, Bershon MM, Nishimoto AY. Effects of pH on Na+-Ca2+ exchange in canine cardiac sarcolemmal vesicles. Circ Res 1982;50:287-93.