- Email: [email protected]
The effect of temperature and hematocrit level of oxygenated cardioplegic solutions on myocardial preservation
J
THoRAc CARDIOVASC SURG
1988;95:625-30
The effect of temperature and hematocrit level of oxygenated cardioplegic solutions on myocardial preservation The ideal temperature and hematocrit level of blood cardioplegia bas not been clearly established. This study was undertaken (a) to determine the optimal temperature of blood cardioplegia and (b)to study the effect of hematocrit levels in blood cardioplegia. A comparison of myocardial preservation was done among seven groups of animals on the basis of variations in hematocrit levels and temperature of oxygenated cardioplegic solution. The experimental protocol consisted of a 2-hour hypothermic cardioplegic arrest followed by 1 hour of normothermic reperfusion. Group 1 received oxygenated crystalloid cardioplegic solution at 10° C. Groups 2 through 7 received oxygenated blood cardioplegic solution with the following hematocrit values and temperatures: (2) 10%, 10° C; (3) 10%, 20° C; (4) 10%, 30° C; (5) 20%,10° C; (6)20%,20° C; and (7) 20%, 30° C. Parameters studied include coronary blood flow, myocardial oxygen extraction, myocardial oxygen consumption, and myocardial high-energy phosphate levels of adenosine triphosphate and creatine phosphate during control (prearrest), arrest, and reperfusion. Myocardial oxygen consumption at 30° C during arrest was significantly higher than at 10° C and 20° C, which indicates continued aerobic metabolic activity at higher temperature. Myocardial oxygen consumption and the levels of adenosine triphosphate and creatine phosphate during reperfusion were similar in all seven groups. Myocardial oxygen extraction (a measure of metabolic function after ischemia) during initial reperfusion was significantly lower in the 30° C blood group than in the 10° C blood group at either hematocrit level and in the oxygenated crystalloid group, which suggests inferior preservation. The hematocrit level of blood cardioplegia did not affect adenosine triphosphate or myocardial oxygen consumption or extraction. It appears from this study that blood cardioplegia at 10° C and oxygenated crystalloid cardioplegia at 10° C are equally effective. Elevating blood cardioplegia temperature to 30° C, however, reduces the ability of the solution to preserve metabolic function regardless of hematocrit level. Therefore, the level of hypothermia is important in blood cardioplegia, whereas hematocrit level bas no detectable impact, and cold oxygenated crystalloid cardioplegia is as effective as hypothermic blood cardioplegia.
John A. Rousou, MD, Richard M. Engelman, MD, Robert H. Breyer, MD, Hajime Otani, MD, PhD, Stanley Lemeshow, PhD, and Dipak K. Das, PhD, Farmington, Conn., and Springfield, Mass.
Hypothermic cardioplegic arrest has been shown to be an effective method of myocardial preservation From the Departments of Surgery, the University of Connecticut School of Medicine, Farmington, Conn., and The Baystate Medical Center, Springfield, Mass. Supported by Grant HL 22559-06 from the National Institutes of Health, Heart, Lung, and Blood Institute. Received for publication Jan. 27, 1987. Accepted for publication May 20, 1987. Address for reprints: John A. Rousou, MD, The Baystate Medical Center, 759 Chestnut St., Springfield, MA 01107.
during cardiac operations both experimentally and clinically.t? However, a variety of cardioplegic solutions have been used. Variations include the cardioplegic vehicle itself (crystalloid versus hemic), the specific ingredients, and the temperature and hematocrit level of blood used in hemic solutions. Furthermore, the oxygen dissociation curve of hemoglobin shifts to the left during hypothermia with less oxygen being available to the tissues at lower temperatures. Reports arguing the effectiveness and relative merits of these various solutions abound in the literature.'? This study was designed to evaluate the relative effect
625
6 26
The Journal of Thoracic and Cardiovascular
Rousou et al.
Surgery
Table I. Metabolic measurements in seven groups of pig hearts
CBF (ml/rnin) EO, (%) MVO, (ml O,jmin) A TP (/Lmol/L/gm) CP (/Lmol/L/gm) Cardio 0, (vol % 0,) Cardio flow rate
Group J (oxygenated crystalloid)
Group 2 (lO%/W)
Group 3 (10%/20' )
(10%/30')
Group 5 (20%/10')
Group 6 (20%/200)
Group 7 (20%/30')
194 ± 20 37 ± 12.3 4.4 ± 0.98 4.6 ± 0.22 10.0 ± 0.75 2.8 ± 0.13 190±25.2
232 34 4.7 4.3 8.9 6.2 276
116 ± 46 ± 3.0 ± 3.6 ± 9.2 ± 4.7 ± 321 ±
26 8.7 0033 0.44 1.61 0.42 71.5
168 ± 23 46 ± 5.4 4.6 ± 0.68 4.4 ± 0.25 9.0 ± 0.59 5.4 ± 0.43 589 ± 82.9*
178 ± 42 ± 4.5 ± 4.7 ± 8.7 ± 9.6 ± 199 ±
168 ± 37 ± 4.4 ± 4.7 ± 9.6 ± 9.0 ± 137 ±
220 56 6.8 4.8 9.0 8.2 446
± ± ± ± ± ± ±
32 9.8 J.2 0.53 1.09 0031 15.4
Group 4
29 8.0 0.91 0.20 0.69 0.45 40.0
13 10.7 1.39 0.18 0.62 0.61 13.1
± ± ± ± ± ± ±
43 10.7 0.41 0.25 0.58 0.28 87.8*
(rnl /rnin)
61 ± 5Jt 3.1 ± 0.19 2.4 ± 0.34
23 ± 3.1 3.8 ± 0.35 2.2 ± 0.34
26 ± 5.0 3.4 ± 0.44 3.2 ± 0.43
27 ± 4.8 7.7 ± 1.07* 2.2 ± 0.22
23 ± 4.8 3.7 ± 0.47 1.7 ± 0.18
35 ± 2.8 4.1 ± 0.30 1.9 ± 0.16
27 ± 6.6 7.2 ± 0.68* 1.7 ± 0.18
4.1 ± 0.50
5.0 ± 0.92
3.1 ± 0.40
3.2 ± 0.48
2.8 ± 0.32
3.6 ± 0.23
2.8 ± 0.55
140 ± 25 36 ± 7.2 2.3 ± 0.23
126 ± 20 50 ± 7.7 2.2 ± 0.39
132 ± 24 29 ± 2.9 2.2 ± 0.31
178 ± 43 20 ± 6.7+ 103 ± 0.44
108 ± 18 38 ± 5.1 1.8 ± 0.24
133 ± 20 31 ± 5.9 2.6 ± 0.69
184 ± 37 25 ± 3.0+ 1.9±0.51
60 min reperf A TP
2.9 ± 0.23
3.0 ± 0.42
3.0 ± 0.41
2.4 ± 0.13
2.4 ± 0.43
2.5 ± 0.17
2.6 ± 0.16
(/Lmol/L/gm) 60 min reperf CP
8.9 ± 0.25
8.6 ± 1.33
12.2 ± 1.34
9.3 ± 0.51
6.9 ± 0.18
8.0 ± 0.78
9.4 ± 0.92
Cardio EO, (o/c) MVO, (ml O,jmin)
120 min arrest A TP (/Lmol/L/gm)
120 min arrest CP (/Lmol/L/gm) 15 min CF (rnl/rnin) 15 min EO, (%)
15 min MVO, (ml O,jmin)
(/Lmol/L/gm) Legend: CBF. Coronary blood flow. 0,. Oxygen. Cardio 0,. Cardioplegic oxygen content. Cardia. Cardioplegic. Repcrf, Reperfusion. All values are mean ± standard error of the mean.
*p < 0.05 tp < 0.05 :j:p< 0.05
compared with JOc C and 20 c C blood cardioplegia. compared with all blood cardioplegia groups. compared with 10' C blood groups only.
of temperature and hematocrit level of oxygenated cardioplegic solutions on myocardial preservation.
Methods Experimental protocol. Thirty-seven Yorkshire pigs of both sexes that weighed between 15 and 20 kg were tranquilized intravenously with pentobarbital (Nembutal 25 mg/kg) and placed on positive-pressure ventilation with room air. The chest was opened with a median sternotomy incision. The animals were placed on cardiopulmonary bypass and the heart isolated in situ from the systemic circulation with its own perfusion pump, as described previously.' All hearts underwent a control period of normothermic hemic perfusion, 2 hours of hypothermic cardioplegic arrest, and 1 hour of normothermic hemic reperfusion. All animals received oxygenated cardioplegic solutions. Fifty milliliters of the solution was given initially and every 15 minutes during arrest at 75 mm Hg perfusion pressure. This suboptimal cardioplegic volume, which is the standard in our laboratory, is given to accentuate any differences in myocardial preservation between groups. In addition, the heart was immersed during arrest in a water bath of the same temperature as the cardioplegic solution administered. Each animal was randomly assigned to receive a cardioplegic solution with one of seven combinations of temperature and hematocrit levels. Animals in group I (n = 5) received oxygenated crystalloid cardioplegia at 10° C. Ani-
mals in groups 2 through 7 received oxygenated blood cardioplegia with the following hematocrit values and temperatures: group 2 (n = 5), 10%, 10° C; group 3 (n = 5), 10%, 20 0 C; group 4 (n = 5), 10%, 30° C; group 5, (n = 6), 20%, 10° C; group 6 (n = 6), 20 0 C; group 7 (n = 5), 20%, 30° C. Parameters measured during normothermic perfusion, cardioplegic arrest, and reperfusion included coronary blood flow or cardioplegic flow, oxygen content of the coronary infusate, and coronary sinus effluent from which myocardial oxygen extraction (EO z) and myocardial oxygen consumption (MVO,) were calculated. These measurements were obtained once during normothermic perfusion and every 15 minutes during arrest and reperfusion. The arrest values were averaged over the 2-hour interval for each animal to obtain a single value for cardioplegic oxygen content, flow rate, EO" and MVO,. All results within each group were then averaged to obtain the data presented for comparison between groups. In addition, left ventricular biopsy specimens were obtained from the anterior apical area of the left ventricle for determination of high-energy phosphate levels measured per gram wet weight. The specimens were obtained during control perfusion, after 60 and 120 minutes of arrest, and after 30 and 60 minutes of reperfusion. Adenosine triphosphate (ATP) and creatine phosphate (CP) assays were performed by the method of Lamprecht and Trautschold." Statistical analysis of data. The basic analytic strategy was parametric and nonparametric analysis of variance fol-
Volume 95 Number 4 April 1988
Temperature and hematocrit of cardioplegia
•
m C
627
Crystauoro 10°(0/10° 10% 1200
CZJ 100/0 I 30°
a
C
MVO,
E?J
20%/10° 20% J 200 20% I 30°
Fig. 1. MV0 2 (milliliters of oxygen per minute) shown as mean ± standard error of the mean during perfusion, arrest, and after 15 minutes of reperfusion in seven groups. MV0 2 is significantly higher only during arrest in 30° C groups at both hematocrit levels. p < 0.05 compared with 10° C and 20° C groups. 70 60
•
m
50
C
EO, 40
C
30
£I
C
a
Crystalloid 10°/0/100 100;0/200 10°10 / 30°
20% 110' 20% 1200 200/0/30 0
20 10
Fig. 2. Percentage of myocardial E0 2 shown as mean ± standard error of the mean during perfusion, arrest, and after 15 minutes of reperfusion in seven groups. Oxygenated crystalloid cardioplegia had highest E02 among seven groups during arrest. During early reperfusion, 30° C blood cardioplegia resulted in significantly lower E0 2 than did 10° C blood cardioplegia at both hematocrit levels and oxygenated crystalloid cardioplegia. Asterisk indicates p < 0.05 compared with 10° C only. Double asterisk indicates p < 0.05 compared with all blood groups. lowed by Tukey's method for pairwise comparisons and the construction of specifically designated contrasts to identify the sources of difference among the groups. For measurements taken during the prearrest period, comparisons were made among all seven groups to determine whether there were any initial differences with respect to coronary blood flow, MV0 2, E0 2, or high-energy phosphates. Subsequent analyses assessed differences between the group receiving crystalloid cardioplegia and the six groups receiving blood cardioplegia. Finally, with the crystalloid group excluded, analyses of the six other groups were done by either temperature or hematocrit level, or both. The ex level for the determination of significance was adjusted accordingly to account for the multiple statistical tests being performed.
Results The results are shown as the mean ± standard error of the mean in Table I and illustrated schematically in four separate graphs (Figs. I to 4). Coronary flow
measurements are given in Table I. During arrest, cardioplegic flow varied even within each animal depending on time of administration in the arrest interval. To reduce variance in comparing the seven groups during cardioplegic delivery, cardioplegic flow was measured every 15 minutes and was averaged for each animal and then within each group to one value. With this method, a significant increase in coronary cardioplegic flow was noted at 30° C compared with 100 C and 20° C during cardioplegic arrest (p < 0.05). The reperfusion flow listed in Table I is the coronary flow after the first 15 minutes of reflow. There is no significant difference between groups during reperfusion, and there was no significant difference at any subsequent time period. Fig. I illustrates MVO z during the three test periods. During cardioplegic arrest, the groups receiving blood
628
The Journal of Thoracic and Cardiovascular
Rousou et ale
Surgery
•
m
o o
Crvstauorc
1O~"
10'·, '0'" 20". CJ 20'J'0 20'0!
e
ATP
m
10 C' 20 30' '0 20: 30"
Fig, 3. ATP levels (/lmol/L/gm) shown as mean ± standard error of the mean during perfusion, after 120 minutes arrest, and after 60 minutes of reperfusion. There is no significant difference between any groups during three test periods.
14
12
• Crystalloid fig 10°,(1,1 10"
C
10
r:zJ
CP
10°0120' 10'')'" 130 u 20- e, 1Dc'
Cl C 20',! 20'
m
20""
30'
Fig. 4. CP levels (/lmol/L/gm) shown as mean ± standard error of the mean during perfusion, after 120 minutes of arrest, and after 60 minutes of reperfusion. There is no significant difference between groups during arrest. During reperfusion differences between groups are inconsistent.
cardioplegia at 30° C (groups 4 and 7) had significantly higher MV0 2 when compared with the other five groups (p < 0.05). During reperfusion, there were no significant differences in MV0 2 among the seven groups. Fig. 2 illustrates E0 2 among the various groups. E0 2 was similar in all groups during initial perfusion. During arrest, E0 2 was significantly higher in the group receiving crystalloid cardioplegia than in any of the blood cardioplegia groups (p < 0.05). With initial reperfusion, there was significant depression of myocardial E0 2 in the groups that had received 30° C blood cardioplegia compared with that of the groups receiving 10° C blood cardioplegia or oxygenated crystalloid cardioplegia. This depression was abolished by 30 minutes of reperfusion when oxygen extraction equalized among groups and remained equal throughout the rest of reperfusion. Fig. 3 depicts graphically the ATP levels by group. No significant differences in ATP levels among the various groups occurred at any time during the study.
CP levels were not significantly different among the seven groups during normothermic perfusion and throughout cardioplegic arrest (Fig. 4). During reperfusion, differences in CP levels were inconclusive. In comparison with the influence of temperature, no significant difference could be found in myocardial preservation when the two hematocrit levels were compared with regard to ATP, MV0 2, and E0 2• Discussion Hypothermic cardioplegic arrest has been shown to retard myocardial metabolic processes and thus reduce significantly the demand for high-energy phosphates and oxygen.' However, although metabolic processes are markedly reduced, they are not suppressed entirely. The theoretic advantage of providing oxygen to support aerobic pathways during arrest and thus meet these reduced metabolic requirements has had broad practical application. to. II This advantage is counteracted by the
Volume 95 Number 4
Temperature and hematocrit of cardioplegia
April 1988
fact that the oxygen dissociation curve of oxyhemoglobin shifts to the left with lower temperatures, which results in decreased release of oxygen to the tissues." The resulting effect of increased oxygen content of blood cardioplegia but decreased release to tissues with increasing hypothermia remains the subject of controversy. In this study, although there was a tendency for greater oxygen release by oxyhemoglobin (higher E02) at higher temperatures, no significant differences in myocardial preservation were noted between 10° and 20° C blood cardioplegia. Additionally, in view of the lack of significant differences in EO z during arrest (Fig. 2), the higher MVO z in the groups receiving 30° C blood cardioplegia (4 and 7) during arrest would appear to be secondary to the higher cardioplegic flow in these groups. In this model, we were unable to demonstrate any clear superiority in the degree of myocardial protection afforded by oxygenated crystalloid cardioplegia or any of the blood cardioplegic solutions. The one parameter that did differentiate between the cardioplegic solutions studied was EO z. We 6 and others" have shown that decreased EOz does occur during the initial reperfusion period, after cardioplegic arrest of 60 minutes. A persistent depression of E0 2 may be a subtle indicator of inadequate myocardial protection. The most significant depression of EO z was seen in the two groups that received blood cardioplegia at 30° C. This depressed EOz suggests inferior myocardial protection with blood cardioplegia (regardless of hematocrit level) administered at 30° C compared with that of any of the solutions administered at 10° or 20° C. With the small sample sizes available in this study, we could not show any significant difference between the solutions administered at 10° versus 20° C. We did not investigate crystalloid cardioplegic solutions at temperatures higher than 10° C because the effects of increased temperature on the oxyhemoglobin dissociation curve do not apply with nonhemic solutions. Additionally, other investigators have previously shown that lower temperatures are necessary for optimal myocardial protection with unoxygenated crystalloid cardioplegic arrest." Comparison of the groups receiving blood cardioplegia at a given temperature (ie, group 2 versus 5, 3 versus 6, 4 versus 7) did not show any effect of hematocrit level on the adequacy of myocardial protection. We tested the hypothesis that a high hematocrit level (20%) in blood cardioplegia at low temperatures (10° C) may cause sludging in the microcirculation and affect coronary blood flow and myocardial preservation. This study did not demonstrate a significant difference in coronary blood flow between hematocrit values of 10% and 20%
629
at 10° C. Further, the flow of cardioplegic solution during blood cardioplegia with a 20% hematocrit level was not significantly depressed at any temperature (10° to 30° C). We conclude that red cell sludging is not a determinant of flow of blood cardioplegic solution at 10° C temperatures if the hematocrit level is 20% or less. It should be noted that flow is measured in normal coronary beds in this study and flow of blood cardioplegia with high hematocrit levels across diseased coronary arteries was not tested. Equivalent reperfusion levels of ATP and CP were seen with oxygenated crystalloid cardioplegic and blood cardioplegic solutions varying between 10% and 20% hematocrit levels and 10° and 30° C temperatures. Administration of blood cardioplegia at 30° C resulted in impaired oxygen extraction during reperfusion compared with administration of blood cardioplegia at 10° C or crystalloid cardioplegia at 10° C. The blood cardioplegic solutions of 10% and 20% hematocrit levels provided equivalent myocardial protection as measured by high-energy phosphate levels, reperfusion MV0 2, and reperfusion myocardial EO z. In conclusion, therefore, the temperature of a blood cardioplegic solution is of importance in determining the level of myocardial protection, but the hematocrit level (between 10% and 20%) does not have any significant impact.
1.
2.
3.
4.
5.
6. 7.
8.
REFERENCES Gay WA Jr. Potassium-induced cardioplegia. Ann Thorae Surg 1975;20:95-100. Todd GJ, Tyers GFO. Amelioration of the effects of ischemic cardiac arrest by the intracoronary administration of cardioplegic solutions. Circulation 1975;52: 1111-7. Engelman RM, Rousou JH, Longo F, Auvil J, Vertrees RA. The time course of myocardial high-energy phosphate degradation during cardioplegic arrest. Surgery 1979;86: 138-47. Ellis RJ, Mangano DT, Van Dyke DC, Ebert PA. Protection of myocardial function not enhanced by high concentrations of potassium during cardioplegic arrest. J THORAC CARDIOVASC SCRG 1979;78:698-707. Follette OM, Steed DL, Foglia R, et al. Advantages of intermittent blood cardioplegia over intermittent ischemia during prolonged hypothermic aortic damping. Circulation 1978;58(Pt 2):1200-9. Engelman RM, Rousou JH, Lemeshow S, Dobbs WA. The metabolic consequences of blood and crystalloid cardioplegia. Circulation 1981 ;64(Pt 2):II67-74. Engelman RM, Auvil J, O'Donoghue MJ, Levitsky S. The significance of multidose cardioplegia and hypothermia in myocardial preservation during ischemic arrest. J THORAC CARDIOVASC SCRG 1978;75:555-63. Lamprecht W, Trautschold J. Determination with hexo-
The Journal of
630
Rousou et al.
kinase and glucose 6-phosphate dehydrogenase. In: Bergmeyer U, ed. Methods of enzymatic analysis. New York: Academic Press, Inc., 1973:543-610. 9. Dobbs WA, Engelman RM, Rousou lH, Pels MA, Alvarez 1M. Residual metabolism of the hypothermicarrested pig heart. 1 Surg Res 1981;31:319-23. 10. Cunningham lN Jr, Catinella FP, Spencer Fe. Blood cardioplegia: experience with prolonged cross-clamping. In: Engelman RM, Levitsky S, eds. A textbook of clinical cardioplegia. Mt. Kisco, New York: Futura Publishing Co. Inc., 1982:241-64. 11. Barner HB, Laks H, Codd lE, et al. Cold blood as the vehicle for potassium cardioplegia. Ann Thorac Surg 1979;28:509-21.
Thoracic and Cardiovascular Surgery
12. Severinghaus lW. Oxyhemoglobin dissociation curve correction for temperature and pH variation in human blood. 1 Appl Physiol 1958;12:485. 13. Lucas SK, Kanter KR, Schaff HV, Elmer EB, Glower DD Jr, Gardner TJ. Reduced oxygen extraction during reperfusion: a consequence of global ischemic arrest. 1 Surg Res 1980;28:434-41. 14. Rosenfeldt FL. Hypothermic preservation techniques: pitfalls. In: Engelman RM, Levitsky S, eds. The textbook of clinical cardioplegia, Mt. Kisco, New York: Futura Publishing Co. Inc., 1982:117-30.
THoRAc CARDIOVASC SURG
1988;95:625-30
The effect of temperature and hematocrit level of oxygenated cardioplegic solutions on myocardial preservation The ideal temperature and hematocrit level of blood cardioplegia bas not been clearly established. This study was undertaken (a) to determine the optimal temperature of blood cardioplegia and (b)to study the effect of hematocrit levels in blood cardioplegia. A comparison of myocardial preservation was done among seven groups of animals on the basis of variations in hematocrit levels and temperature of oxygenated cardioplegic solution. The experimental protocol consisted of a 2-hour hypothermic cardioplegic arrest followed by 1 hour of normothermic reperfusion. Group 1 received oxygenated crystalloid cardioplegic solution at 10° C. Groups 2 through 7 received oxygenated blood cardioplegic solution with the following hematocrit values and temperatures: (2) 10%, 10° C; (3) 10%, 20° C; (4) 10%, 30° C; (5) 20%,10° C; (6)20%,20° C; and (7) 20%, 30° C. Parameters studied include coronary blood flow, myocardial oxygen extraction, myocardial oxygen consumption, and myocardial high-energy phosphate levels of adenosine triphosphate and creatine phosphate during control (prearrest), arrest, and reperfusion. Myocardial oxygen consumption at 30° C during arrest was significantly higher than at 10° C and 20° C, which indicates continued aerobic metabolic activity at higher temperature. Myocardial oxygen consumption and the levels of adenosine triphosphate and creatine phosphate during reperfusion were similar in all seven groups. Myocardial oxygen extraction (a measure of metabolic function after ischemia) during initial reperfusion was significantly lower in the 30° C blood group than in the 10° C blood group at either hematocrit level and in the oxygenated crystalloid group, which suggests inferior preservation. The hematocrit level of blood cardioplegia did not affect adenosine triphosphate or myocardial oxygen consumption or extraction. It appears from this study that blood cardioplegia at 10° C and oxygenated crystalloid cardioplegia at 10° C are equally effective. Elevating blood cardioplegia temperature to 30° C, however, reduces the ability of the solution to preserve metabolic function regardless of hematocrit level. Therefore, the level of hypothermia is important in blood cardioplegia, whereas hematocrit level bas no detectable impact, and cold oxygenated crystalloid cardioplegia is as effective as hypothermic blood cardioplegia.
John A. Rousou, MD, Richard M. Engelman, MD, Robert H. Breyer, MD, Hajime Otani, MD, PhD, Stanley Lemeshow, PhD, and Dipak K. Das, PhD, Farmington, Conn., and Springfield, Mass.
Hypothermic cardioplegic arrest has been shown to be an effective method of myocardial preservation From the Departments of Surgery, the University of Connecticut School of Medicine, Farmington, Conn., and The Baystate Medical Center, Springfield, Mass. Supported by Grant HL 22559-06 from the National Institutes of Health, Heart, Lung, and Blood Institute. Received for publication Jan. 27, 1987. Accepted for publication May 20, 1987. Address for reprints: John A. Rousou, MD, The Baystate Medical Center, 759 Chestnut St., Springfield, MA 01107.
during cardiac operations both experimentally and clinically.t? However, a variety of cardioplegic solutions have been used. Variations include the cardioplegic vehicle itself (crystalloid versus hemic), the specific ingredients, and the temperature and hematocrit level of blood used in hemic solutions. Furthermore, the oxygen dissociation curve of hemoglobin shifts to the left during hypothermia with less oxygen being available to the tissues at lower temperatures. Reports arguing the effectiveness and relative merits of these various solutions abound in the literature.'? This study was designed to evaluate the relative effect
625
6 26
The Journal of Thoracic and Cardiovascular
Rousou et al.
Surgery
Table I. Metabolic measurements in seven groups of pig hearts
CBF (ml/rnin) EO, (%) MVO, (ml O,jmin) A TP (/Lmol/L/gm) CP (/Lmol/L/gm) Cardio 0, (vol % 0,) Cardio flow rate
Group J (oxygenated crystalloid)
Group 2 (lO%/W)
Group 3 (10%/20' )
(10%/30')
Group 5 (20%/10')
Group 6 (20%/200)
Group 7 (20%/30')
194 ± 20 37 ± 12.3 4.4 ± 0.98 4.6 ± 0.22 10.0 ± 0.75 2.8 ± 0.13 190±25.2
232 34 4.7 4.3 8.9 6.2 276
116 ± 46 ± 3.0 ± 3.6 ± 9.2 ± 4.7 ± 321 ±
26 8.7 0033 0.44 1.61 0.42 71.5
168 ± 23 46 ± 5.4 4.6 ± 0.68 4.4 ± 0.25 9.0 ± 0.59 5.4 ± 0.43 589 ± 82.9*
178 ± 42 ± 4.5 ± 4.7 ± 8.7 ± 9.6 ± 199 ±
168 ± 37 ± 4.4 ± 4.7 ± 9.6 ± 9.0 ± 137 ±
220 56 6.8 4.8 9.0 8.2 446
± ± ± ± ± ± ±
32 9.8 J.2 0.53 1.09 0031 15.4
Group 4
29 8.0 0.91 0.20 0.69 0.45 40.0
13 10.7 1.39 0.18 0.62 0.61 13.1
± ± ± ± ± ± ±
43 10.7 0.41 0.25 0.58 0.28 87.8*
(rnl /rnin)
61 ± 5Jt 3.1 ± 0.19 2.4 ± 0.34
23 ± 3.1 3.8 ± 0.35 2.2 ± 0.34
26 ± 5.0 3.4 ± 0.44 3.2 ± 0.43
27 ± 4.8 7.7 ± 1.07* 2.2 ± 0.22
23 ± 4.8 3.7 ± 0.47 1.7 ± 0.18
35 ± 2.8 4.1 ± 0.30 1.9 ± 0.16
27 ± 6.6 7.2 ± 0.68* 1.7 ± 0.18
4.1 ± 0.50
5.0 ± 0.92
3.1 ± 0.40
3.2 ± 0.48
2.8 ± 0.32
3.6 ± 0.23
2.8 ± 0.55
140 ± 25 36 ± 7.2 2.3 ± 0.23
126 ± 20 50 ± 7.7 2.2 ± 0.39
132 ± 24 29 ± 2.9 2.2 ± 0.31
178 ± 43 20 ± 6.7+ 103 ± 0.44
108 ± 18 38 ± 5.1 1.8 ± 0.24
133 ± 20 31 ± 5.9 2.6 ± 0.69
184 ± 37 25 ± 3.0+ 1.9±0.51
60 min reperf A TP
2.9 ± 0.23
3.0 ± 0.42
3.0 ± 0.41
2.4 ± 0.13
2.4 ± 0.43
2.5 ± 0.17
2.6 ± 0.16
(/Lmol/L/gm) 60 min reperf CP
8.9 ± 0.25
8.6 ± 1.33
12.2 ± 1.34
9.3 ± 0.51
6.9 ± 0.18
8.0 ± 0.78
9.4 ± 0.92
Cardio EO, (o/c) MVO, (ml O,jmin)
120 min arrest A TP (/Lmol/L/gm)
120 min arrest CP (/Lmol/L/gm) 15 min CF (rnl/rnin) 15 min EO, (%)
15 min MVO, (ml O,jmin)
(/Lmol/L/gm) Legend: CBF. Coronary blood flow. 0,. Oxygen. Cardio 0,. Cardioplegic oxygen content. Cardia. Cardioplegic. Repcrf, Reperfusion. All values are mean ± standard error of the mean.
*p < 0.05 tp < 0.05 :j:p< 0.05
compared with JOc C and 20 c C blood cardioplegia. compared with all blood cardioplegia groups. compared with 10' C blood groups only.
of temperature and hematocrit level of oxygenated cardioplegic solutions on myocardial preservation.
Methods Experimental protocol. Thirty-seven Yorkshire pigs of both sexes that weighed between 15 and 20 kg were tranquilized intravenously with pentobarbital (Nembutal 25 mg/kg) and placed on positive-pressure ventilation with room air. The chest was opened with a median sternotomy incision. The animals were placed on cardiopulmonary bypass and the heart isolated in situ from the systemic circulation with its own perfusion pump, as described previously.' All hearts underwent a control period of normothermic hemic perfusion, 2 hours of hypothermic cardioplegic arrest, and 1 hour of normothermic hemic reperfusion. All animals received oxygenated cardioplegic solutions. Fifty milliliters of the solution was given initially and every 15 minutes during arrest at 75 mm Hg perfusion pressure. This suboptimal cardioplegic volume, which is the standard in our laboratory, is given to accentuate any differences in myocardial preservation between groups. In addition, the heart was immersed during arrest in a water bath of the same temperature as the cardioplegic solution administered. Each animal was randomly assigned to receive a cardioplegic solution with one of seven combinations of temperature and hematocrit levels. Animals in group I (n = 5) received oxygenated crystalloid cardioplegia at 10° C. Ani-
mals in groups 2 through 7 received oxygenated blood cardioplegia with the following hematocrit values and temperatures: group 2 (n = 5), 10%, 10° C; group 3 (n = 5), 10%, 20 0 C; group 4 (n = 5), 10%, 30° C; group 5, (n = 6), 20%, 10° C; group 6 (n = 6), 20 0 C; group 7 (n = 5), 20%, 30° C. Parameters measured during normothermic perfusion, cardioplegic arrest, and reperfusion included coronary blood flow or cardioplegic flow, oxygen content of the coronary infusate, and coronary sinus effluent from which myocardial oxygen extraction (EO z) and myocardial oxygen consumption (MVO,) were calculated. These measurements were obtained once during normothermic perfusion and every 15 minutes during arrest and reperfusion. The arrest values were averaged over the 2-hour interval for each animal to obtain a single value for cardioplegic oxygen content, flow rate, EO" and MVO,. All results within each group were then averaged to obtain the data presented for comparison between groups. In addition, left ventricular biopsy specimens were obtained from the anterior apical area of the left ventricle for determination of high-energy phosphate levels measured per gram wet weight. The specimens were obtained during control perfusion, after 60 and 120 minutes of arrest, and after 30 and 60 minutes of reperfusion. Adenosine triphosphate (ATP) and creatine phosphate (CP) assays were performed by the method of Lamprecht and Trautschold." Statistical analysis of data. The basic analytic strategy was parametric and nonparametric analysis of variance fol-
Volume 95 Number 4 April 1988
Temperature and hematocrit of cardioplegia
•
m C
627
Crystauoro 10°(0/10° 10% 1200
CZJ 100/0 I 30°
a
C
MVO,
E?J
20%/10° 20% J 200 20% I 30°
Fig. 1. MV0 2 (milliliters of oxygen per minute) shown as mean ± standard error of the mean during perfusion, arrest, and after 15 minutes of reperfusion in seven groups. MV0 2 is significantly higher only during arrest in 30° C groups at both hematocrit levels. p < 0.05 compared with 10° C and 20° C groups. 70 60
•
m
50
C
EO, 40
C
30
£I
C
a
Crystalloid 10°/0/100 100;0/200 10°10 / 30°
20% 110' 20% 1200 200/0/30 0
20 10
Fig. 2. Percentage of myocardial E0 2 shown as mean ± standard error of the mean during perfusion, arrest, and after 15 minutes of reperfusion in seven groups. Oxygenated crystalloid cardioplegia had highest E02 among seven groups during arrest. During early reperfusion, 30° C blood cardioplegia resulted in significantly lower E0 2 than did 10° C blood cardioplegia at both hematocrit levels and oxygenated crystalloid cardioplegia. Asterisk indicates p < 0.05 compared with 10° C only. Double asterisk indicates p < 0.05 compared with all blood groups. lowed by Tukey's method for pairwise comparisons and the construction of specifically designated contrasts to identify the sources of difference among the groups. For measurements taken during the prearrest period, comparisons were made among all seven groups to determine whether there were any initial differences with respect to coronary blood flow, MV0 2, E0 2, or high-energy phosphates. Subsequent analyses assessed differences between the group receiving crystalloid cardioplegia and the six groups receiving blood cardioplegia. Finally, with the crystalloid group excluded, analyses of the six other groups were done by either temperature or hematocrit level, or both. The ex level for the determination of significance was adjusted accordingly to account for the multiple statistical tests being performed.
Results The results are shown as the mean ± standard error of the mean in Table I and illustrated schematically in four separate graphs (Figs. I to 4). Coronary flow
measurements are given in Table I. During arrest, cardioplegic flow varied even within each animal depending on time of administration in the arrest interval. To reduce variance in comparing the seven groups during cardioplegic delivery, cardioplegic flow was measured every 15 minutes and was averaged for each animal and then within each group to one value. With this method, a significant increase in coronary cardioplegic flow was noted at 30° C compared with 100 C and 20° C during cardioplegic arrest (p < 0.05). The reperfusion flow listed in Table I is the coronary flow after the first 15 minutes of reflow. There is no significant difference between groups during reperfusion, and there was no significant difference at any subsequent time period. Fig. I illustrates MVO z during the three test periods. During cardioplegic arrest, the groups receiving blood
628
The Journal of Thoracic and Cardiovascular
Rousou et ale
Surgery
•
m
o o
Crvstauorc
1O~"
10'·, '0'" 20". CJ 20'J'0 20'0!
e
ATP
m
10 C' 20 30' '0 20: 30"
Fig, 3. ATP levels (/lmol/L/gm) shown as mean ± standard error of the mean during perfusion, after 120 minutes arrest, and after 60 minutes of reperfusion. There is no significant difference between any groups during three test periods.
14
12
• Crystalloid fig 10°,(1,1 10"
C
10
r:zJ
CP
10°0120' 10'')'" 130 u 20- e, 1Dc'
Cl C 20',! 20'
m
20""
30'
Fig. 4. CP levels (/lmol/L/gm) shown as mean ± standard error of the mean during perfusion, after 120 minutes of arrest, and after 60 minutes of reperfusion. There is no significant difference between groups during arrest. During reperfusion differences between groups are inconsistent.
cardioplegia at 30° C (groups 4 and 7) had significantly higher MV0 2 when compared with the other five groups (p < 0.05). During reperfusion, there were no significant differences in MV0 2 among the seven groups. Fig. 2 illustrates E0 2 among the various groups. E0 2 was similar in all groups during initial perfusion. During arrest, E0 2 was significantly higher in the group receiving crystalloid cardioplegia than in any of the blood cardioplegia groups (p < 0.05). With initial reperfusion, there was significant depression of myocardial E0 2 in the groups that had received 30° C blood cardioplegia compared with that of the groups receiving 10° C blood cardioplegia or oxygenated crystalloid cardioplegia. This depression was abolished by 30 minutes of reperfusion when oxygen extraction equalized among groups and remained equal throughout the rest of reperfusion. Fig. 3 depicts graphically the ATP levels by group. No significant differences in ATP levels among the various groups occurred at any time during the study.
CP levels were not significantly different among the seven groups during normothermic perfusion and throughout cardioplegic arrest (Fig. 4). During reperfusion, differences in CP levels were inconclusive. In comparison with the influence of temperature, no significant difference could be found in myocardial preservation when the two hematocrit levels were compared with regard to ATP, MV0 2, and E0 2• Discussion Hypothermic cardioplegic arrest has been shown to retard myocardial metabolic processes and thus reduce significantly the demand for high-energy phosphates and oxygen.' However, although metabolic processes are markedly reduced, they are not suppressed entirely. The theoretic advantage of providing oxygen to support aerobic pathways during arrest and thus meet these reduced metabolic requirements has had broad practical application. to. II This advantage is counteracted by the
Volume 95 Number 4
Temperature and hematocrit of cardioplegia
April 1988
fact that the oxygen dissociation curve of oxyhemoglobin shifts to the left with lower temperatures, which results in decreased release of oxygen to the tissues." The resulting effect of increased oxygen content of blood cardioplegia but decreased release to tissues with increasing hypothermia remains the subject of controversy. In this study, although there was a tendency for greater oxygen release by oxyhemoglobin (higher E02) at higher temperatures, no significant differences in myocardial preservation were noted between 10° and 20° C blood cardioplegia. Additionally, in view of the lack of significant differences in EO z during arrest (Fig. 2), the higher MVO z in the groups receiving 30° C blood cardioplegia (4 and 7) during arrest would appear to be secondary to the higher cardioplegic flow in these groups. In this model, we were unable to demonstrate any clear superiority in the degree of myocardial protection afforded by oxygenated crystalloid cardioplegia or any of the blood cardioplegic solutions. The one parameter that did differentiate between the cardioplegic solutions studied was EO z. We 6 and others" have shown that decreased EOz does occur during the initial reperfusion period, after cardioplegic arrest of 60 minutes. A persistent depression of E0 2 may be a subtle indicator of inadequate myocardial protection. The most significant depression of EO z was seen in the two groups that received blood cardioplegia at 30° C. This depressed EOz suggests inferior myocardial protection with blood cardioplegia (regardless of hematocrit level) administered at 30° C compared with that of any of the solutions administered at 10° or 20° C. With the small sample sizes available in this study, we could not show any significant difference between the solutions administered at 10° versus 20° C. We did not investigate crystalloid cardioplegic solutions at temperatures higher than 10° C because the effects of increased temperature on the oxyhemoglobin dissociation curve do not apply with nonhemic solutions. Additionally, other investigators have previously shown that lower temperatures are necessary for optimal myocardial protection with unoxygenated crystalloid cardioplegic arrest." Comparison of the groups receiving blood cardioplegia at a given temperature (ie, group 2 versus 5, 3 versus 6, 4 versus 7) did not show any effect of hematocrit level on the adequacy of myocardial protection. We tested the hypothesis that a high hematocrit level (20%) in blood cardioplegia at low temperatures (10° C) may cause sludging in the microcirculation and affect coronary blood flow and myocardial preservation. This study did not demonstrate a significant difference in coronary blood flow between hematocrit values of 10% and 20%
629
at 10° C. Further, the flow of cardioplegic solution during blood cardioplegia with a 20% hematocrit level was not significantly depressed at any temperature (10° to 30° C). We conclude that red cell sludging is not a determinant of flow of blood cardioplegic solution at 10° C temperatures if the hematocrit level is 20% or less. It should be noted that flow is measured in normal coronary beds in this study and flow of blood cardioplegia with high hematocrit levels across diseased coronary arteries was not tested. Equivalent reperfusion levels of ATP and CP were seen with oxygenated crystalloid cardioplegic and blood cardioplegic solutions varying between 10% and 20% hematocrit levels and 10° and 30° C temperatures. Administration of blood cardioplegia at 30° C resulted in impaired oxygen extraction during reperfusion compared with administration of blood cardioplegia at 10° C or crystalloid cardioplegia at 10° C. The blood cardioplegic solutions of 10% and 20% hematocrit levels provided equivalent myocardial protection as measured by high-energy phosphate levels, reperfusion MV0 2, and reperfusion myocardial EO z. In conclusion, therefore, the temperature of a blood cardioplegic solution is of importance in determining the level of myocardial protection, but the hematocrit level (between 10% and 20%) does not have any significant impact.
1.
2.
3.
4.
5.
6. 7.
8.
REFERENCES Gay WA Jr. Potassium-induced cardioplegia. Ann Thorae Surg 1975;20:95-100. Todd GJ, Tyers GFO. Amelioration of the effects of ischemic cardiac arrest by the intracoronary administration of cardioplegic solutions. Circulation 1975;52: 1111-7. Engelman RM, Rousou JH, Longo F, Auvil J, Vertrees RA. The time course of myocardial high-energy phosphate degradation during cardioplegic arrest. Surgery 1979;86: 138-47. Ellis RJ, Mangano DT, Van Dyke DC, Ebert PA. Protection of myocardial function not enhanced by high concentrations of potassium during cardioplegic arrest. J THORAC CARDIOVASC SCRG 1979;78:698-707. Follette OM, Steed DL, Foglia R, et al. Advantages of intermittent blood cardioplegia over intermittent ischemia during prolonged hypothermic aortic damping. Circulation 1978;58(Pt 2):1200-9. Engelman RM, Rousou JH, Lemeshow S, Dobbs WA. The metabolic consequences of blood and crystalloid cardioplegia. Circulation 1981 ;64(Pt 2):II67-74. Engelman RM, Auvil J, O'Donoghue MJ, Levitsky S. The significance of multidose cardioplegia and hypothermia in myocardial preservation during ischemic arrest. J THORAC CARDIOVASC SCRG 1978;75:555-63. Lamprecht W, Trautschold J. Determination with hexo-
The Journal of
630
Rousou et al.
kinase and glucose 6-phosphate dehydrogenase. In: Bergmeyer U, ed. Methods of enzymatic analysis. New York: Academic Press, Inc., 1973:543-610. 9. Dobbs WA, Engelman RM, Rousou lH, Pels MA, Alvarez 1M. Residual metabolism of the hypothermicarrested pig heart. 1 Surg Res 1981;31:319-23. 10. Cunningham lN Jr, Catinella FP, Spencer Fe. Blood cardioplegia: experience with prolonged cross-clamping. In: Engelman RM, Levitsky S, eds. A textbook of clinical cardioplegia. Mt. Kisco, New York: Futura Publishing Co. Inc., 1982:241-64. 11. Barner HB, Laks H, Codd lE, et al. Cold blood as the vehicle for potassium cardioplegia. Ann Thorac Surg 1979;28:509-21.
Thoracic and Cardiovascular Surgery
12. Severinghaus lW. Oxyhemoglobin dissociation curve correction for temperature and pH variation in human blood. 1 Appl Physiol 1958;12:485. 13. Lucas SK, Kanter KR, Schaff HV, Elmer EB, Glower DD Jr, Gardner TJ. Reduced oxygen extraction during reperfusion: a consequence of global ischemic arrest. 1 Surg Res 1980;28:434-41. 14. Rosenfeldt FL. Hypothermic preservation techniques: pitfalls. In: Engelman RM, Levitsky S, eds. The textbook of clinical cardioplegia, Mt. Kisco, New York: Futura Publishing Co. Inc., 1982:117-30.