- Email: [email protected]
Transient hemodynamic dysfunction after myocardial revascularization
J
THoRAc CARDIOVASC SURG
86:226-234, 1983
Transient hemodynamic dysfunction after myocardial revascularization Temperature dependence We studied hemodynamics and the effects of right atrial pacing (110 beats/min) following complete myocardial revascularization and hypothermic multidose potassium crystalloid cardioplegia in 12 patients with a normal preoperative left ventricular ejection fraction (LVEF). Measurements were made immediately preoperatively, postoperatively at specified temperatures during the rewarming period (90' F, 94' F, and 98' F), and at 24 hours. No patient had a perioperative myocardial infarction. At 90' F, hemodynamics were characterized by significant decreases in cardiac index, stroke volume index, and left ventricular stroke work index (LVSW) and an increase in systemic vascular resistance index (SVRI). compared to preoperative values (p < 0.05). Right atrial pacing significantly increased cardiac index preoperatively and 24 hours postoperatively, but not during the rewarming period. Over the entire rewarming period (90' F to 98' F), each of the following variables correlated with temperature: cardiac index (r = 0.71 in sinus rhythm and r = 0.66 with right atrial pacing); stroke volume index (r = 0.33 and 0.66); SVRI (r = -0.80 and -0.64); LVSW (r = 0.37 and 0.73); and heart rate in sinus rhythm (r = 0.51). During the rewarming period, there was an inverse relationship between cardiac index and SVRI (r = -0.87). In conclusion, after myocardial revascularization: (1) transient hemodynamic dysfunction occurs during the rewarming period (90' F to 98' F); (2) this dysfunction is temperature-dependent; and (3) right atrial pacing at 110 beats/min does not improve hemodynamic function during the rewarming period. Temperature must be considered in the evaluation of left ventricular and hemodynamic function following myocardial revascularization.
Lawrence Czer, M.D., Angas Hamer, M.D., Franklin Murphy, M.D., John Bussell, M.D., Aurelio Chaux, M.D., Timothy Bateman, M.D., Jack Matloff, M.D., and Richard J. Gray, M.D., Los Angeles, Calif.
Transient left ventricular dysfunction has been demonstrated by scintigraphic and hemodynamic measurements immediately following uncomplicated coronary artery bypass grafting (CABG)Y Although well described, its pathophysiological basis remains unclear. The role of hypothermia in this transient left ventricular dysfunction is not known. In a carefully selected group of patients with a normal (>50%) preoperative left From the Division of Cardiology and Departments of Anesthesiology and Cardiovascular Surgery, Cedars-Sinai Medical Center, UCLA School of Medicine, Los Angeles, Calif. Received for publication Sept. 9, 1982. Accepted for publication Nov. 22, 1982. Address for reprints: Lawrence Czer, M.D., Division of Cardiology, Room 5314, Cedars-Sinai Medical Center, 8700 Beverly Blvd., Los Angeles, Calif. 90048.
226
ventricular ejection fraction (LVEF) undergoing complete myocardial revascularization, we investigated whether this transient hemodynamic dysfunction occurs during the hypothermic rewarming period after CABG and whether atrial pacing at 110 beats/min improves hemodynamic function during this period. Methods Patient selection. Twelve adult patients, 11 men and one woman, aged 39 to 74 years (mean 54), were studied prospectively after informed consent was obtained. All patients were in New York Heart Association (NYHA) Functional Class II or III, had undergone cardiac catheterization within 3 months of the study, and were candidates for elective CABG. Fulfillment of all of the following criteria was required for inclusion of a patient into this study: (1) angiographical-
Volume 86 Number 2 August, 1983
ly determined LVEF greater than 50%; (2) left ventricular end-diastolic pressure less than 18 mm Hg at cardiac catheterization and absence of clinical symptoms or signs of congestive heart failure; (3) absence of distal coronary artery disease (which would preclude complete myocardial revascularization); and (4) absence of concomitant valvular or congenital heart disease. Patients receiving cardiovascular medications other than nitrates or low-dose propranolol «120 mg/day, discontinued 24 to 48 hours prior to operation) were not included in the study. At angiography, the LVEF in the study group ranged from 52% to 75% (mean 64%). No patient had a left ventricular aneurysm. One patient had single-vessel coronary artery disease, three had double-vessel disease, and eight had triple-vessel disease. Anesthesia. The same anesthesiologist, anesthetic premedication, and anesthesia were used for all 12 patients. Anesthetic premedication consisted of secobarbital 200 mg and diazepam 5 to 10 mg by mouth I hour before operation. Anesthesia was induced with thiopental sodium (Pentothal) and was maintained with enflurane, oxygen, and fentanyl. Pancuronium was used for intubation and relaxation. A multipurpose balloon flotation pulmonary artery catheter was inserted via the right internal jugular vein prior to induction of anesthesia. This catheter contained three atrial and two ventricular electrodes, with the most proximal atrial electrode positioned fluoroscopically in the high right atrium. This catheter also contained proximal (right atrial) and distal (pulmonary arterial) ports for pressure determination and a thermistor tip for measuring thermodilution cardiac outputs. Radial arterial and central venous pressure monitoring lines were also placed prior to induction. Surgical technique. Extracorporeal circulation was maintained with a bubble oxygenator during systemic hypothermia to 20 0 C, the mean pump time being 135 ± 53 minutes (mean ± SD). Myocardial preservation was maintained with hypothel'rnic multidose potassium crystalloid cardioplegia (St. Thomas' Hospital solution No. 13) . Each patient had been rewarmed to a core temperature of 98 0 F, and then received 1 gm of Calcium chloride at the termination of the pump run. A single group of cardiovascular surgeons performed the CABG procedures, utilizing autologous saphenous vein. The number of bypass grafts inserted per patient ranged from two (for the patient with single-vessel disease) to five (for some patients with triple-vessel disease), and averaged 3.9. The surgical team judged each patient to be completely revascularized. Postoperative support. No patient required mechan-
Hemodynamic dysfunction after CABG
227
- 44~ u
~
~
36 28
: ; 5.0f 20
~~4.0
oS
-
o~
001
E
~
~
-±-
3.0
Ij
':f~ L-I
--Ij
L-I
L-I
L-I
L....J
Preop 9ooF. 94°F. 9soF. 24hr Fig. 1. Temperature (T). hematocrit (Het), potassium (K). and (Ca) during study. Values are mean ± SD.
ical or pharmacologic support, digoxin, beta blockers, calcium-channel blockers, or antiarrhythmic drugs postoperatively. Oxygenation was maintained within normal limits by mechanical ventilation during the early postoperative period and thereafter with supplementary oxygen by mask. Electrolytes were maintained within normal limits. Creatine kinase and lactic dehydrogenase isoenzymes were measured every 8 hours during the first 48 hours postoperatively; a 12-lead electrocardiogram was performed during each rewarming stage and daily thereafter. No patient had clinical, enzymatic, or electrocardiographic evidence of a perioperative myocardial infarction or ischemia. All patients were in NYHA Functional Class I at discharge. Hemodynamic measurements and temperature. Preoperatively and prior to induction of anesthesia (control), the following hemodynamic measurements were recorded during sinus rhythm: heart rate, systemic mean arterial pressure (MAP), right atrial pressure (RAP), pulmonary capillary wedge pressure (PCWP), and four thermodilution cardiac outputs. Hemodynamic measurements were repeated during right atrial pacing at 110 beats/min. Core temperature was measured from
The Journal of Thoracic and Cardiovascular Surgery
228 Czer et al.
Table I. Hemodynamic data Preop.
90' F
94' F
98' F
CI (Lrrnin/m') SR P
2.24 ± 0.07 2.81 ± O.09t
1.81 ± 0.14* 1.83 ± 0.16*
2.21 ± 0.15 2.26 ± 0.18
2.58 ± 0.14* 2.63 ± 0.19
3.08 ± 0.18* 3.64 ± 0.18*t
SI (cc/m') SR P
35.4 ± 1.5 25.5 ± 0.8t
24.6 ± 1.6* 16.7 ± 1.5*t
26.7 ± 1.8* 20.5 ± 1.7t
29.0 ± 2.0* 23.9 ± 1.8t
34.4 ± 2.5 33.0 ± 1.7*
96.2 ± 3.2 94.0 ± 3.9
87.3 ± 3.6 86.9 ± 3.5
86.1 ± 1.9 89.0 ± 5.3
82.8 ± 2.1 86.2 ± 3.0
SVRI (dyne. sec . cm". m') SR 3,357 ± 179 3,018 ± 144t P
4,302 ± 521* 4,145 ± 527*
3,002 ± 265 3,058 ± 425
2,360 ± 136* 2,471 ± 139*
1,967 ± 131* 1,733 ± 108*t
LVSW (gm. m/rn') SR 42.4 ± 2.3 34.0 ± 1.6t P
28.5 ± 1.7* 18.5 ± 2.0*t
27.8 ± 2.2* 21.0 ± 1.8*t
29.9 ± 2.5* 25.7 ± 3.6*
34.2 ± 3.2* 33.8 ± 2.2
6.0 ± 0.4 5.9 ± 0.6
8.2 ± 0.3* 8.5 ± 0.6
9.0 ± 0.7* 9.2 ± 0.8
8.6 ± 0.8* 8.6 ± 0.2
9.2 ± 1.2* 8.9 ± 1.3
PCWP (mm Hg) SR P
10.6 ± 1.0 12.3 ± 0.7
9.9 ± 0.4 12.8 ± l.l t
10.0 ± 0.8 10.7 ± 1.0
10.9 ± 0.7 11.0 ± 0.9
10.8 ± 0.9 10.9 ± 1.6
HR (beats/min) SR
65 ± 4
75 ± 4
84 ± 4*
91 ± 4*
91 ± 4*
MAP (mm Hg) SR P
99.6 ± 4.8 110.6 ± 3.8t
RAP (mm Hg) SR P
24 hr
Legend:SR, Sinus rhythm. P, Paced (110 beats/min). CI, Cardiac index. SI, Stroke volume index. MAP. Mean arterial pressure. SVRI, Systemic vascular resistance index. LVSW. Left ventricular stroke work index. RAP, Right atrial pressure. PCWP, Pulmonary capillary wedge pressure. HR, Heart rate. Data are expressed as mean ± SEM. 'Significantly different from preoperative value (p < 0.05), analysis of variance. tPaced value significantly different from sinus value (p < 0.05).
the thermistor tip of the multipurpose catheter. The accuracy of the thermistor-thermometer (Model 9520A, Edwards Laboratories, Inc., Santa Ana, Calif.) has been shown to be ±0.01 ° C over a wide range of temperatures (27° to 43° C), with a time constant of 1.75 seconds.' Hematocrit, serum electrolytes, calcium, arterial Po., Pco; and pH, and twelve-lead electrocardiograms were recorded during the control period. The cardiac index (CI) and stroke volume index (SI) were obtained by dividing the cardiac output and the stroke volume by the body surface area. The systemic vascular resistance index (SVRI, dyne. sec . em:' . m') and left ventricular stroke work index (LVSW, gm . mjbeatj m-) were calculated by the following formulas: SVRI LVSW
=
MAP-RAP
CI
x 80
= 0.D136 x SI x (MAP -
PCWP)
All measurements were repeated during the rewarming period postoperatively at three stages, defined by the patient's core temperature (90° F, 94° F, and 98° F):
(1) immediately after stabilization in the intensive care unit, at a mean core temperature of 90° ± 1°F (32.2° ± 0.8° C, mean ± SD); (2) at 94° ± 1.5° F (34.3° ± 0.8° C); and (3) at 98° ± 1° F (36.8° ± 0.5° C). Stage 2 measurements were made 2.6 ± 0.6 hours after Stage 1, and Stage 3 measurements were made 8.0 ± 4.0 hours after Stage I. For simplicity, each stage of rewarming will be referred to hereafter by its mean core temperature: 90° F, 94° F, and 98° F. Twenty-four hours postoperatively, these measurements were again repeated, at a mean core temperature of 99.3° ± 0.5° F (37.4° ± 0.3° C). Fig. 1 shows the temporal course of the mean core temperature, hematocrit value, serum potassium, and total calcium during each stage of the study period. Statistical methods. Mean values were calculated for a given hemodynamic variable at each stage during sinus rhythm and with right atrial pacing (Table I). Friedman's nonparametric analysis of variance for repeated measurements was used to test for differences between stages for a given variable; if differences were
Volume 86 Number 2 August, 1983
Hemodynamic dysfunction after CABG
229
• NSR
.-:e
o Atrial pacino (110)
4.0
0
<, C
'E
"'"
• ....-0
3.0
...J
X
L&J
a
z
2.0
•
~
~
I=ll
~
•
i$
C
0::
~
~ 0
1.0
..-.
\
:r
Preop
9ooF.
94°F.
24h
Fig. 2. Response of cardiac index to right atrial pacing. The cardiac index increased consistently in response to atrial pacing preoperatively and 24 hours postoperatively, but not consistently during the rewarming period. NSR. Normal sinus rhythm.
found, the Student-Newman-Keuls nonparametric method was used to identify which group(s) differed. A p value <0.05 was considered statistically significant. Correlation of temperature with specified hemodynamic variables was analyzed by means of the Spearman rank-order correlation coefficient. Because measurements were made at repeated intervals over time, there exists the possibility that the observed changes in hemodynamics may have been due to the time elapsed since the operation rather than to changes in body temperature. In order to investigate this possibility, the simultaneous effects of time and temperature on all hemodynamic variables 'were examined via a multiple linear regression model by the method of least squares.' Results Cardiac index, stroke volume index, and heart rate. At 90° F, both the cardiac index in sinus rhythm and the cardiac index with right atrial pacing were significantly lower than their preoperative (control) values (p < 0.05; Table I). By 98 0 F, the sinus cardiac index exceeded its control value (p < 0.05), and the paced cardiac index was not significantly different from its control value. Twenty-four hours postoperatively, both the sinus and the paced cardiac indices exceeded control values (p < 0.05). The augmentation of the cardiac index seen during the control period with right atrial
pacing (p < 0.05) was lost during the entire rewarming period (90 0 F, 94°F and 98° F), but returned by 24 hours postoperatively (p < paced versus sinus). Similar observations were seen on an individual basis (Fig. 2): Each patient experienced an augmentation of cardiac index with pacing in the control period and 24 hours postoperatively, but the consistency of this change in cardiac index was lost during the hypothermic rewarming period. The sinus and paced stroke volume indices each were significantly lower at 90° F than their control values (p < 0.05). During the remainder of the rewarming period, both the sinus and the paced stroke volume indices increased in comparison to the values at 90° F. By 98 0 F, the sinus stroke volume index was still lower than the control value (p < 0.05). However, because stroke volume index is calculated from the cardiac index and heart rate, the simultaneous changes in heart rate during rewarming confound the analysis. At a fixed rate of atrial pacing (110 beats/min), the paced stroke volume index at 98 ° F is not significantly different from its control value. By 24 hours postoperatively, the sinus stroke volume index (34.4) was very close to the control value (p > 0.05), and the paced stroke volume index (33.0) exceeded its control value (p < 0.05). The mean sinus rate was 65 beats/min during the control period. The mean heart rate increased progressively during rewarming to 75 beats/min (at 90° F,
om,
The Journal of Thoracic and Cardiovascular Surgery
2 3 0 Czer et al.
5
• Sinus rhythm
r·.71
..-
.• .. •
...
~
3
~ , J--_______.~ ~
•
~
""Ec: d"
50
I
90°
86°
3
X
(OF)
~
(.)
~
10
o
o
Atrial pacing
r·.66
CJ)
CD
~8 (J
o
0-
0
o
o 0
0
0
00(j£
0
. ...•. ....• ,1'-:'
•
9ao
86°
...J
~
0 3 «
..--"
~ 50
X
~ 5
~ UJ
..
I
.~.
30
~
~
9ao
81R1s Rhytlwn
r -.33
"Co)
•
94°
•
> ~ 30
o
Atrial pacinQ
r -.66
o
....a: 00
10
0
98° I
31°
31°
33°
35°
TEMPERATlflE
Fig. 3. Correlation of cardiac index with temperature during rewarming.
p = NS), 84 beats/min (at 94° F, P < 0.05),91 beats/ min (98° F, P < 0.05), and remained at 91 beats/min (p < 0.05) 24 hours postoperatively. RAP and PCWP. The mean RAP in sinus rhythm remained unchanged with atrial pacing preoperatively and at all core temperatures postoperatively. Similarly, there was no significant difference between the sinus PCWP and the paced PCWP, except at 90° F, when the mean PCWP rose from 9.9 to 12.8 with pacing (p < 0.05). Postoperatively, the RAP was higher than the control value (p < 0.05), but the PCWP remained unchanged. MAP and SVRI. During the control period, the MAP was 99.6 mm Hg in sinus rhythm and rose to 110.6 mm Hg with atrial pacing (p < 0.05). At each stage postoperatively, the MAP remained unchanged with atrial pacing (p > 0.05). Compared to control values, there was no significant change in MAP during rewarming, either in sinus rhythm or during atrial pacing. The SVRI was 3,357 dyne . sec . em? . m' in sinus rhythm during the control period and decreased to 3,018 dyne. sec . cnr' . m' with pacing (p < 0.05). Postoperatively, the SVRI fell significantly with atrial pacing at
33°
(OF)
35°
TEMPERATURE Fig. 4. Correlation of stroke volume index with temperature during rewarming.
90° F and 24 hours postoperatively (p < 0.05). Compared to control values, the SVRI increased significantly at 90° F and decreased significantly at 98° F and 24 hours postoperatively, both in sinus rhythm and with atrial pacing (p < 0.05). LVSW. The sinus and paced LVSWs were significantly lower than control values at 90° F, 94°F, and 98° F (p < 0.05). By 24 hours postoperatively, the paced LVSW (but not the sinus LVSW) had returned to control levels. The LVSW fell significantly with atrial pacing during the control period and during rewarming at 90° F and 94° F (p < 0.05). Correlation of hemodynamics with temperature. There was a strong correlation between temperature and several hemodynamic variables during the rewarming period (90 0 F to 98° F). The cardiac index in sinus rhythm and the cardiac index during atrial pacing demonstrated a positive correlation with temperature (r = .71 and 0.66, respectively, Fig. 3), as did the sinus and paced stroke volume indices (r =0.33 and 0.66, Fig. 4) and the sinus and paced LVSWs (r = 0.37 and 0.73). The sinus rate correlated with temperature (r = 0.51). The SVRI was inversely correlated with temperature (r = -0.80 and -0.64) and with the cardiac index (r = -0.87; Fig. 5). The use of a uniform pacing rate (110 beats/min) eliminates simultaneous changes in
Volume 86 Number 2 August, 1983
heart rate as a confounding variable: Note that correlations of stroke volume index and LVSW with temperature were better with the paced values than with the sinus values. The simultaneous effects of time and temperature were examined by means of a multiple linear regression model. When multiple regression indicated that a hemodynamic variable could be predicted from temperature and/or time, temperature was consistently a better predictor. Variables strongly associated with temperature independent of time were cardiac index, stroke volume index, SVRI (all in sinus rhythm and with right atrial pacing), and the paced LVSW. No hemodynamic variable was strongly associated with time when temperature was controlled in the analysis. Discussion
Hypothermia exerts a protective effect on the myocardium through two principal mechanisms: a depression of cellular metabolism, resulting in a decrease in myocardial oxygen consumption and sparing of intracellular energy stores (adenosine triphosphate and creatine phosphate), and a reduction in the fluidity of cellular membranes, resulting in preservation of cellular integrity.3,6.12 However, not all of the cellular effects of hypothermia are beneficial; harmful effects of hypothermia include ionic redistribution, cellular swelling, and cold-induced injury.":" Although these harmful effects can be minimized by the use of multidose potassium cardioplegia, they cannot be completely eliminated. During reperfusion and rewarming, additional detrimental effects, especially inward calcium fluxes and additional cellular swelling, may appear. 19, 20 Reperfusion also has been identified as a cause of injury in intermittent ischemia and reperfusion. 21,22 Thus the postoperative rewarming state may be very different from the hypothermic "state occurring during cooling and cardioplegic arrest, in which the hemodynamic and metabolic effects of hypothermia have been extensively documented." The effects of· hypothermia during rewarming have not been adequately studied. An important animal study" has recently demonstrated significantly higher myocardial adenosine triphosphate and 'creatine phosphate levels after normothermic reperfusion than after hypothermic reperfusion. This observation implies that hypothermic reperfusion may adversely affect hemodynamic recovery of the heart following cardioplegic arrest. Wei have previously presented scintigraphic and hemodynamic evidence for transient left ventricular dysfunction after CABG in a group of 30 patients in whom myocardial preservation consisted of moderate
Hemodynamic dysfunction after CABG
9000
•
• Sinus rhythm
•
• .,
~ 5000
en
1.~~
~.
1000
23I
.
•••••
0
2 4 CARDIAC INDEX (L/min/M 2 )
Fig. 5, Inverse correlation between cardiac index and systemic vascular resistance index (SVJU. dyne. sec . cm" . m') during rewarming.
systemic hypothermia (20 0 to 28 0 C) and intermittent aortic cross-clamping without chemical cardioplegia. The LVEF decreased from 58% ± 2% (preoperatively) to 41% ± 1% (1 to 5 hours postoperatively). At the same time, a decrease in cardiac index and LVSW occurred despite an increase in PCWP. The LVEF, cardiac index, and LVSW returned to normal or supernormal values by the second postoperative day. More recently, Roberts and associates' presented similar evidence of transient left ventricular dysfunction in 40 patients who underwent CABG with hypothermic multidose potassium crystalloid cardioplegia. The LVEF fell from 50% ± 3% (preoperatively) to 38% ± 2% (2 hours postoperatively); the cardiac index and LVSW fell also. The LVEF and cardiac index returned to preoperative levels by 24 hours. Importantly, temperature was not noted in these studies. The present study was undertaken to determine whether this transient postoperative hemodynamic dysfunction occurred during the rewarming period following CABG in 12 patients with a normal global LVEF preoperatively. All patients received hypothermic multidose potassium crystalloid cardioplegia with systemic hypothermia (20 0 C). No patient had a myocardial infarction, sustained arrhythmia, or tamponade during the study period, factors which may produce left ventricular dysfunction. Administration of vasopressor, beta blocker, digitalis, and antiarrhythmic medication was strictly avoided during the study period. The present study demonstrates at 90 0 F a significant fall in cardiac index, stroke volume index, and LVSW and an increase in SVRI compared to control values. The subsequent changes in these hemodynamic variables during the rewarming period (90 F to 98 F) were correlated with temperature. By 98 F, these hemodynamic changes were largely corrected: The cardiac index and the paced 0
0
0
2 3 2 Czer et al.
stroke volume index returned to control levels (fixed pacing eliminating changes in heart rate as a confounding variable). Thus the hemodynamic dysfunction after myocardial revascularization occurs predominantly during the rewarming period and is strongly temperaturedependent. Myocardial oxygen debt accumulated during ischemic time because of imperfect myocardial preservation, and the detrimental effects of reperfusion alluded to earlier may contribute to the postoperative hemodynamic dysfunction independent of temperature. These detrimental effects should diminish with time elapsed after operation. Multiple regression analysis demonstrated that temperature was a consistently better predictor of hemodynamic function during the rewarming period than was time. Thus hypothermia exerts the dominant effect on early postoperative hemodynamic dysfunction. However, the fact that the paced LVSW did not return to control levels until 24 hours postoperatively, with only a small change in temperature (to 99° F), suggests that hypothermia alone does not explain all of the hemodynamic dysfunction that occurs postoperatively. Afterload exerts important effects during the rewarming process. There was a strong inverse relationship between cardiac index and SVRI during the rewarming period (Fig. 5). Thus, under the conditions of impaired left ventricular function that occur during rewarming, cardiac output is quite sensitive to changes in afterload. In the absence of changes in filling pressures, increases in SVRI might be due to hypothermia itself, increased catecholamines, or both. Plasma catecholamines, especially norepinephrine, increase markedly after CABG.24-26 There is a poor correlation between SVRI and catecholamine level during and after CABG. 26 During rewarming, the SVRI was inversely correlated with temperature (r = 0.80); this suggests that hypothermia plays a major role in the changes in SVRI that occur during early rewarming. Other factors that might account for transient left ventricular dysfunction include hypocalcemia and hemodilution. Hypocalcemia is present to a variable degree following CABGY We28 have previously shown in a group of patients not given citrate in the pump prime or parenteral calcium during operation that the postoperative decline in total calcium can be explained fully by the degree of hemodilution. Furthermore, ionized calcium is maintained at normal or near normal levels by homeostatic mechanisms, mainly the decrease in binding to albumin and the maintenance of parathyroid hormone secretion." In the present study, the postoperative drop in total calcium was less pronounced than in the previous study, because of the administration
The Journal of Thoracic and Cardiovascular Surgery
of 1 gm of calcium chloride at the end of the pump run; again, the degree of hypocalcemia could be accounted for totally by the degree of hemodilution. In addition, there was no significant difference between the total calcium levels at 90° F and at 24 hours postoperatively (Fig. I); thus the hypocalcemia is unlikely to explain the hemodynamic changes observed during the rewarming period. Hemodilution also occurs to a variable degree following CABG. 3 Hemodilution has been shown .to increase cardiac index. 29, 3o However, in our study, the cardiac index fell rather than increased in association with the early postoperative fall in hematocrit value. Furthermore, the subsequent increases in cardiac index during the rewarming period were not accompanied by any significant change in hematocrit value, which remained steady at 28% (Fig. 1). Thus hemodilution is unlikely to explain the observed hemodynamic changes during the rewarming period. Atrial pacing has been used to measure myocardial reserve following CABG and to induce ischemia during the immediate postoperative period." The pacing rates required to induce ischemia are generally much higher than 110 beats/min, the rate used in this study." There was no clinical or electrocardiographic evidence of pacing-induced ischemia during the present study, nor was there electrocardiographic or creatine kinase enzymatic evidence of a perioperative myocardial infarction. Although augmentation of cardiac index by atrial pacing was lost during the rewarming period, no patient had a clinically significant deterioration of hemodynamics (cardiac index or MAP) with right atrial pacing at 110 beats/min. In conclusion, the transient hemodynamic dysfunction after CABG occurs during the rewarming period despite maintenance of right atrial and pulmonary capillary wedge filling pressures. This dysfunction is characterized by an initial decrease in cardiac index, stroke volume index, LVSW, and LVEF, a loss of augmentation of cardiac index by atrial pacing, and an increase in heart rate and SVRI. The subsequent changes in cardiac index, stroke volume index, LVSW, SVRI, and heart rate correlate with the temperature changes during rewarming. Further, there is a striking inverse relationship between cardiac index and SVRI during the rewarming process. Atrial pacing at 110 beats/min adds no hemodynamic benefit during the rewarming period. In view of these fmdings, temperature must be considered in the evaluation of left ventricular and hemodynamic function following myocardial revascularization. We wish to thank Carolyn Conklin, R.N., Marjorie Raymond, R.N., and Michael DeRobertis, R.N., for their techni-
Volume 86 Number 2 August, 1983
cal assistance; Robert Kass, M.D., and Myles Lee, M.D., of the Department of Cardiovascular Surgery; and the nurses of the Cardiac Surgical Intensive Care Unit of Cedars-Sinai Medical Center. Special thanks to Morgan Stewart, M.S., of the Scientific Data Center, for help in the statistical analysis of the data. REFERENCES Gray R, Maddahi J, Berman D, Raymond M, Waxman A, Ganz W, Matloff J, Swan HJC: Scintigraphic and hemodynamic demonstration of transient left ventricular dysfunction immediately after uncomplicated coronary artery bypass grafting. J THoRAc CARDIOVASC SURG 77:504-510, 1979 2 Roberts AJ, Spies SM, Sanders JH, Morgan JM, Wilkinson CJ, Lichtenthal PR, White RL, Michaelis LL: Serial assessment of left ventricular performance following coronary artery bypass grafting. J THoRAc CARDIOVASC SURG 81:69-84, 1981 3 Hearse DJ, Braimbridge MV, Jynge P: Protection of the Ischemic Myocardium: Cardioplegia, New York, 1981, Raven Press 4 Shellock FG, Rubin SA: Simplified and highly accurate core temperature measurements. Med Progr Technol 8:187-188, 1982 5 Dixon WJ: BMDP Statistical Software, Berkeley, 1981, University of California Press 6 Archie JP, Kirklin JW: Effect of hypothermic perfusion on myocardial oxygen consumption and coronary resistance. Surg Forum 24:186-188, 1973 7 Buckberg GD, Brazier JR, Nelson RH, Goldstein SM, McConnell DH, Cooper N: Studies of the effects of hypothermia on regional myocardial blood flow and metabolism during cardiopulmonary bypass. I. The adequately perfused beating, fibrillating, and arrested heart. J THoRAc CARDIOVASC SURG 73:87-94, 1977 8 Chitwood WR, Sink JD, Hill RC, Wechsler AS, Sabiston DC: The effects of hypothermia on myocardial oxygen consumption and transmural coronary blood flow in the potassium-arrested heart. Ann Surg 190:106-116, 1979 9 Vinas JF, Fewel JG, Arom KV, Trinkle JK, Grover FL: Effects of systemic hypothermia on myocardial metabolism and coronary blood flow in the fibrillating heart. J THoRAc CARDIOVASC SURG 77:900-908, 1979 10 Blok MC, van Deenen LLM, de Gier J, Op den Kamp JAF, Verkleij AJ: Some aspects of lipid-phase transition on membrane permeability and lipid-protein association, Proceedings in Life Sciences. Biochemistry of Membrane Transport, Berlin, 1977, Springer-Verlag, pp 38-46 II Chapman D, Cornell BA, Quinn PJ: Phase transitions, protein aggregation and a new method for modulating membrane fluidity, Proceedings in Life Sciences. Biochemistry of Membrane Transport, Berlin, 1977, Springer-Verlag, pp 72-85 12 Inesi G, Millman M, Eletr S: Temperature-induced transitions of function and structure in sarcoplasmic reticulum membranes. J Mol BioI 81:483-504, 1973
Hemodynamic dysfunction after CABG
233
13 Martin DR, Scott DF, Downes GL, Belzer FO: Primary cause of unsuccessful liver and heart preservation. Cold sensitivity of the ATP-ase system. Ann Surg 175: 111-117, 1972 14 Belzer FO: Kidney preservation. Surg Clin North Am 58:261-272, 1978 15 Leaf A: Cell swelling. A factor in ischemic tissue injury. Circulation 48:455-458, 1973 16 MacKnight AC, Leaf A: Regulation of cellular volume. Physiol Rev 57:510-573, 1977 17 Pine MB, Kahne D, Jaski B, Apstein CS, Thorp K, Abelmann WH: Sodium permeability and myocardial resistance to cell swelling during metabolic blockade. Am J Physiol 239:H31-H39, 1980 18 Meryman HT: Freezing injury and its prevention in living cells. Ann Rev Biophys Bioeng 3:341-363, 1974 19 Hearse DJ: Reperfusion of the ischemic myocardium. J Mol Cell Cardiol 9:605-616, 1977 20 Zimmerman ANE, Daems W, Hulsmann WC, Snijder J, Wisse E, Durrer D: Morphological changes of heart muscle caused by successive perfusion with calcium-free and calcium-containing solutions (calcium paradox). Cardiovasc Res 1:201-209, 1967 21 Engelman RM, Chandra R, Baumann FG, Goldman RA: Myocardial reperfusion, a cause of ischemic injury during cardiopulmonary bypass. Surgery 80:266-276, 1976 22 Engelman RM, Rousou JH, O'Donoghue MJ, Longo F, Dobbs W A: A comparison of intermittent and continuous arrest for prolonged hypothermic cardioplegia. Ann Thorae Surg 29:217-233,1980 23 Rousou JH, Dobbs WA, Meeran MK, Engelman RM: The temperature dependence of recovery of metabolic function following hypothermic potassium cardioplegic arrest. J THORAC CARDIOVASC SURG 83:117-121, 1982 24 Kim YD, Jones M, Hanowell ST, Koch HP, Lees DE, Weise V, Kopin 11: Changes in peripheral vascular and cardiac sympathetic activity before and after coronary artery bypass surgery. Interrelationships with hemodynamic alterations. Am Heart J 102:972-979, 1981 25 Hoar PF, Stone JG, Faltas AN, Bendixen HH, Head RJ, Berkowitz BA: Hemodynamic and adrenergic responses to anesthesia and operation for myocardial revascularization. J THoRAc CARDIOVASC SURG 80:242-248, 1980 26 Reves JG, Karp RB, Buttner EE, Tosone S, Smith LR, Samuelson PN, Kreusch GR, Oparil S: Neuronal and adrenomedullary catecholamine release in response to cardiopulmonary bypass in man. Circulation 66:49-55, 1982 27 Moffitt EA, Tarhan S, Goldsmith RS, Pluth JR, McGoon DC: Patterns of total and ionized calcium and other electrolytes in plasma during and after cardiac surgery. J THoRAc CARDIOVASC SURG 65:751-757, 1973 28 Gray R, Braunstein G, Krutzik S, Conklin C, Matloff J: Calcium homeostasis during coronary bypass surgery. Circulation 62:Suppl 1:57, 1980 29 Fowler NO, Holmes JC: Blood viscosity and cardiac
2 3 4 Czer et al.
output in acute experimental anemia. J Appl Physiol 39:453-456, 1975 30 Clarke TNS, Prys-Roberts C, Biro G, Foex P, Bennett MJ: Aortic input impedance and left ventricular energetics in acute isovolemic anaemia. Cardiovasc Res 12:49-55, 1978 31 Hilton JD, Weisel RD, Baird RJ, Goldman BS, Jablonsky
The Journal of Thoracic and Cardiovascular Surgery
G, Pym J, ScuIly HE, Ivanov J, Mickle DAG, Feiglin DH, Morch JE, Mclaughlin PR: The hemodynamic and metabolic response to pacing after aortocoronary bypass. Circulation 64:(Suppl 2):48, 1981 32 Judgutt BI, Lee SJK, Taylor RF: Abolition of ischemic response to atrial pacing foIlowing aortocoronary bypass surgery. Circulation 54:225-229, 1976
Bound volumes available to subscribers Bound volumes of THE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1983 issues from the Publisher, at a cost of $58.50 ($73.00 international) for Vol. 85 (January-June) and Vol. 86 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 30 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany a/l orders. Contact Mr. Deans Lynch at The C. V. Mosby Company, 11830 Westline Industrial Drive, S1. Louis, Missouri 63146, USA. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.
THoRAc CARDIOVASC SURG
86:226-234, 1983
Transient hemodynamic dysfunction after myocardial revascularization Temperature dependence We studied hemodynamics and the effects of right atrial pacing (110 beats/min) following complete myocardial revascularization and hypothermic multidose potassium crystalloid cardioplegia in 12 patients with a normal preoperative left ventricular ejection fraction (LVEF). Measurements were made immediately preoperatively, postoperatively at specified temperatures during the rewarming period (90' F, 94' F, and 98' F), and at 24 hours. No patient had a perioperative myocardial infarction. At 90' F, hemodynamics were characterized by significant decreases in cardiac index, stroke volume index, and left ventricular stroke work index (LVSW) and an increase in systemic vascular resistance index (SVRI). compared to preoperative values (p < 0.05). Right atrial pacing significantly increased cardiac index preoperatively and 24 hours postoperatively, but not during the rewarming period. Over the entire rewarming period (90' F to 98' F), each of the following variables correlated with temperature: cardiac index (r = 0.71 in sinus rhythm and r = 0.66 with right atrial pacing); stroke volume index (r = 0.33 and 0.66); SVRI (r = -0.80 and -0.64); LVSW (r = 0.37 and 0.73); and heart rate in sinus rhythm (r = 0.51). During the rewarming period, there was an inverse relationship between cardiac index and SVRI (r = -0.87). In conclusion, after myocardial revascularization: (1) transient hemodynamic dysfunction occurs during the rewarming period (90' F to 98' F); (2) this dysfunction is temperature-dependent; and (3) right atrial pacing at 110 beats/min does not improve hemodynamic function during the rewarming period. Temperature must be considered in the evaluation of left ventricular and hemodynamic function following myocardial revascularization.
Lawrence Czer, M.D., Angas Hamer, M.D., Franklin Murphy, M.D., John Bussell, M.D., Aurelio Chaux, M.D., Timothy Bateman, M.D., Jack Matloff, M.D., and Richard J. Gray, M.D., Los Angeles, Calif.
Transient left ventricular dysfunction has been demonstrated by scintigraphic and hemodynamic measurements immediately following uncomplicated coronary artery bypass grafting (CABG)Y Although well described, its pathophysiological basis remains unclear. The role of hypothermia in this transient left ventricular dysfunction is not known. In a carefully selected group of patients with a normal (>50%) preoperative left From the Division of Cardiology and Departments of Anesthesiology and Cardiovascular Surgery, Cedars-Sinai Medical Center, UCLA School of Medicine, Los Angeles, Calif. Received for publication Sept. 9, 1982. Accepted for publication Nov. 22, 1982. Address for reprints: Lawrence Czer, M.D., Division of Cardiology, Room 5314, Cedars-Sinai Medical Center, 8700 Beverly Blvd., Los Angeles, Calif. 90048.
226
ventricular ejection fraction (LVEF) undergoing complete myocardial revascularization, we investigated whether this transient hemodynamic dysfunction occurs during the hypothermic rewarming period after CABG and whether atrial pacing at 110 beats/min improves hemodynamic function during this period. Methods Patient selection. Twelve adult patients, 11 men and one woman, aged 39 to 74 years (mean 54), were studied prospectively after informed consent was obtained. All patients were in New York Heart Association (NYHA) Functional Class II or III, had undergone cardiac catheterization within 3 months of the study, and were candidates for elective CABG. Fulfillment of all of the following criteria was required for inclusion of a patient into this study: (1) angiographical-
Volume 86 Number 2 August, 1983
ly determined LVEF greater than 50%; (2) left ventricular end-diastolic pressure less than 18 mm Hg at cardiac catheterization and absence of clinical symptoms or signs of congestive heart failure; (3) absence of distal coronary artery disease (which would preclude complete myocardial revascularization); and (4) absence of concomitant valvular or congenital heart disease. Patients receiving cardiovascular medications other than nitrates or low-dose propranolol «120 mg/day, discontinued 24 to 48 hours prior to operation) were not included in the study. At angiography, the LVEF in the study group ranged from 52% to 75% (mean 64%). No patient had a left ventricular aneurysm. One patient had single-vessel coronary artery disease, three had double-vessel disease, and eight had triple-vessel disease. Anesthesia. The same anesthesiologist, anesthetic premedication, and anesthesia were used for all 12 patients. Anesthetic premedication consisted of secobarbital 200 mg and diazepam 5 to 10 mg by mouth I hour before operation. Anesthesia was induced with thiopental sodium (Pentothal) and was maintained with enflurane, oxygen, and fentanyl. Pancuronium was used for intubation and relaxation. A multipurpose balloon flotation pulmonary artery catheter was inserted via the right internal jugular vein prior to induction of anesthesia. This catheter contained three atrial and two ventricular electrodes, with the most proximal atrial electrode positioned fluoroscopically in the high right atrium. This catheter also contained proximal (right atrial) and distal (pulmonary arterial) ports for pressure determination and a thermistor tip for measuring thermodilution cardiac outputs. Radial arterial and central venous pressure monitoring lines were also placed prior to induction. Surgical technique. Extracorporeal circulation was maintained with a bubble oxygenator during systemic hypothermia to 20 0 C, the mean pump time being 135 ± 53 minutes (mean ± SD). Myocardial preservation was maintained with hypothel'rnic multidose potassium crystalloid cardioplegia (St. Thomas' Hospital solution No. 13) . Each patient had been rewarmed to a core temperature of 98 0 F, and then received 1 gm of Calcium chloride at the termination of the pump run. A single group of cardiovascular surgeons performed the CABG procedures, utilizing autologous saphenous vein. The number of bypass grafts inserted per patient ranged from two (for the patient with single-vessel disease) to five (for some patients with triple-vessel disease), and averaged 3.9. The surgical team judged each patient to be completely revascularized. Postoperative support. No patient required mechan-
Hemodynamic dysfunction after CABG
227
- 44~ u
~
~
36 28
: ; 5.0f 20
~~4.0
oS
-
o~
001
E
~
~
-±-
3.0
Ij
':f~ L-I
--Ij
L-I
L-I
L-I
L....J
Preop 9ooF. 94°F. 9soF. 24hr Fig. 1. Temperature (T). hematocrit (Het), potassium (K). and (Ca) during study. Values are mean ± SD.
ical or pharmacologic support, digoxin, beta blockers, calcium-channel blockers, or antiarrhythmic drugs postoperatively. Oxygenation was maintained within normal limits by mechanical ventilation during the early postoperative period and thereafter with supplementary oxygen by mask. Electrolytes were maintained within normal limits. Creatine kinase and lactic dehydrogenase isoenzymes were measured every 8 hours during the first 48 hours postoperatively; a 12-lead electrocardiogram was performed during each rewarming stage and daily thereafter. No patient had clinical, enzymatic, or electrocardiographic evidence of a perioperative myocardial infarction or ischemia. All patients were in NYHA Functional Class I at discharge. Hemodynamic measurements and temperature. Preoperatively and prior to induction of anesthesia (control), the following hemodynamic measurements were recorded during sinus rhythm: heart rate, systemic mean arterial pressure (MAP), right atrial pressure (RAP), pulmonary capillary wedge pressure (PCWP), and four thermodilution cardiac outputs. Hemodynamic measurements were repeated during right atrial pacing at 110 beats/min. Core temperature was measured from
The Journal of Thoracic and Cardiovascular Surgery
228 Czer et al.
Table I. Hemodynamic data Preop.
90' F
94' F
98' F
CI (Lrrnin/m') SR P
2.24 ± 0.07 2.81 ± O.09t
1.81 ± 0.14* 1.83 ± 0.16*
2.21 ± 0.15 2.26 ± 0.18
2.58 ± 0.14* 2.63 ± 0.19
3.08 ± 0.18* 3.64 ± 0.18*t
SI (cc/m') SR P
35.4 ± 1.5 25.5 ± 0.8t
24.6 ± 1.6* 16.7 ± 1.5*t
26.7 ± 1.8* 20.5 ± 1.7t
29.0 ± 2.0* 23.9 ± 1.8t
34.4 ± 2.5 33.0 ± 1.7*
96.2 ± 3.2 94.0 ± 3.9
87.3 ± 3.6 86.9 ± 3.5
86.1 ± 1.9 89.0 ± 5.3
82.8 ± 2.1 86.2 ± 3.0
SVRI (dyne. sec . cm". m') SR 3,357 ± 179 3,018 ± 144t P
4,302 ± 521* 4,145 ± 527*
3,002 ± 265 3,058 ± 425
2,360 ± 136* 2,471 ± 139*
1,967 ± 131* 1,733 ± 108*t
LVSW (gm. m/rn') SR 42.4 ± 2.3 34.0 ± 1.6t P
28.5 ± 1.7* 18.5 ± 2.0*t
27.8 ± 2.2* 21.0 ± 1.8*t
29.9 ± 2.5* 25.7 ± 3.6*
34.2 ± 3.2* 33.8 ± 2.2
6.0 ± 0.4 5.9 ± 0.6
8.2 ± 0.3* 8.5 ± 0.6
9.0 ± 0.7* 9.2 ± 0.8
8.6 ± 0.8* 8.6 ± 0.2
9.2 ± 1.2* 8.9 ± 1.3
PCWP (mm Hg) SR P
10.6 ± 1.0 12.3 ± 0.7
9.9 ± 0.4 12.8 ± l.l t
10.0 ± 0.8 10.7 ± 1.0
10.9 ± 0.7 11.0 ± 0.9
10.8 ± 0.9 10.9 ± 1.6
HR (beats/min) SR
65 ± 4
75 ± 4
84 ± 4*
91 ± 4*
91 ± 4*
MAP (mm Hg) SR P
99.6 ± 4.8 110.6 ± 3.8t
RAP (mm Hg) SR P
24 hr
Legend:SR, Sinus rhythm. P, Paced (110 beats/min). CI, Cardiac index. SI, Stroke volume index. MAP. Mean arterial pressure. SVRI, Systemic vascular resistance index. LVSW. Left ventricular stroke work index. RAP, Right atrial pressure. PCWP, Pulmonary capillary wedge pressure. HR, Heart rate. Data are expressed as mean ± SEM. 'Significantly different from preoperative value (p < 0.05), analysis of variance. tPaced value significantly different from sinus value (p < 0.05).
the thermistor tip of the multipurpose catheter. The accuracy of the thermistor-thermometer (Model 9520A, Edwards Laboratories, Inc., Santa Ana, Calif.) has been shown to be ±0.01 ° C over a wide range of temperatures (27° to 43° C), with a time constant of 1.75 seconds.' Hematocrit, serum electrolytes, calcium, arterial Po., Pco; and pH, and twelve-lead electrocardiograms were recorded during the control period. The cardiac index (CI) and stroke volume index (SI) were obtained by dividing the cardiac output and the stroke volume by the body surface area. The systemic vascular resistance index (SVRI, dyne. sec . em:' . m') and left ventricular stroke work index (LVSW, gm . mjbeatj m-) were calculated by the following formulas: SVRI LVSW
=
MAP-RAP
CI
x 80
= 0.D136 x SI x (MAP -
PCWP)
All measurements were repeated during the rewarming period postoperatively at three stages, defined by the patient's core temperature (90° F, 94° F, and 98° F):
(1) immediately after stabilization in the intensive care unit, at a mean core temperature of 90° ± 1°F (32.2° ± 0.8° C, mean ± SD); (2) at 94° ± 1.5° F (34.3° ± 0.8° C); and (3) at 98° ± 1° F (36.8° ± 0.5° C). Stage 2 measurements were made 2.6 ± 0.6 hours after Stage 1, and Stage 3 measurements were made 8.0 ± 4.0 hours after Stage I. For simplicity, each stage of rewarming will be referred to hereafter by its mean core temperature: 90° F, 94° F, and 98° F. Twenty-four hours postoperatively, these measurements were again repeated, at a mean core temperature of 99.3° ± 0.5° F (37.4° ± 0.3° C). Fig. 1 shows the temporal course of the mean core temperature, hematocrit value, serum potassium, and total calcium during each stage of the study period. Statistical methods. Mean values were calculated for a given hemodynamic variable at each stage during sinus rhythm and with right atrial pacing (Table I). Friedman's nonparametric analysis of variance for repeated measurements was used to test for differences between stages for a given variable; if differences were
Volume 86 Number 2 August, 1983
Hemodynamic dysfunction after CABG
229
• NSR
.-:e
o Atrial pacino (110)
4.0
0
<, C
'E
"'"
• ....-0
3.0
...J
X
L&J
a
z
2.0
•
~
~
I=ll
~
•
i$
C
0::
~
~ 0
1.0
..-.
\
:r
Preop
9ooF.
94°F.
24h
Fig. 2. Response of cardiac index to right atrial pacing. The cardiac index increased consistently in response to atrial pacing preoperatively and 24 hours postoperatively, but not consistently during the rewarming period. NSR. Normal sinus rhythm.
found, the Student-Newman-Keuls nonparametric method was used to identify which group(s) differed. A p value <0.05 was considered statistically significant. Correlation of temperature with specified hemodynamic variables was analyzed by means of the Spearman rank-order correlation coefficient. Because measurements were made at repeated intervals over time, there exists the possibility that the observed changes in hemodynamics may have been due to the time elapsed since the operation rather than to changes in body temperature. In order to investigate this possibility, the simultaneous effects of time and temperature on all hemodynamic variables 'were examined via a multiple linear regression model by the method of least squares.' Results Cardiac index, stroke volume index, and heart rate. At 90° F, both the cardiac index in sinus rhythm and the cardiac index with right atrial pacing were significantly lower than their preoperative (control) values (p < 0.05; Table I). By 98 0 F, the sinus cardiac index exceeded its control value (p < 0.05), and the paced cardiac index was not significantly different from its control value. Twenty-four hours postoperatively, both the sinus and the paced cardiac indices exceeded control values (p < 0.05). The augmentation of the cardiac index seen during the control period with right atrial
pacing (p < 0.05) was lost during the entire rewarming period (90 0 F, 94°F and 98° F), but returned by 24 hours postoperatively (p < paced versus sinus). Similar observations were seen on an individual basis (Fig. 2): Each patient experienced an augmentation of cardiac index with pacing in the control period and 24 hours postoperatively, but the consistency of this change in cardiac index was lost during the hypothermic rewarming period. The sinus and paced stroke volume indices each were significantly lower at 90° F than their control values (p < 0.05). During the remainder of the rewarming period, both the sinus and the paced stroke volume indices increased in comparison to the values at 90° F. By 98 0 F, the sinus stroke volume index was still lower than the control value (p < 0.05). However, because stroke volume index is calculated from the cardiac index and heart rate, the simultaneous changes in heart rate during rewarming confound the analysis. At a fixed rate of atrial pacing (110 beats/min), the paced stroke volume index at 98 ° F is not significantly different from its control value. By 24 hours postoperatively, the sinus stroke volume index (34.4) was very close to the control value (p > 0.05), and the paced stroke volume index (33.0) exceeded its control value (p < 0.05). The mean sinus rate was 65 beats/min during the control period. The mean heart rate increased progressively during rewarming to 75 beats/min (at 90° F,
om,
The Journal of Thoracic and Cardiovascular Surgery
2 3 0 Czer et al.
5
• Sinus rhythm
r·.71
..-
.• .. •
...
~
3
~ , J--_______.~ ~
•
~
""Ec: d"
50
I
90°
86°
3
X
(OF)
~
(.)
~
10
o
o
Atrial pacing
r·.66
CJ)
CD
~8 (J
o
0-
0
o
o 0
0
0
00(j£
0
. ...•. ....• ,1'-:'
•
9ao
86°
...J
~
0 3 «
..--"
~ 50
X
~ 5
~ UJ
..
I
.~.
30
~
~
9ao
81R1s Rhytlwn
r -.33
"Co)
•
94°
•
> ~ 30
o
Atrial pacinQ
r -.66
o
....a: 00
10
0
98° I
31°
31°
33°
35°
TEMPERATlflE
Fig. 3. Correlation of cardiac index with temperature during rewarming.
p = NS), 84 beats/min (at 94° F, P < 0.05),91 beats/ min (98° F, P < 0.05), and remained at 91 beats/min (p < 0.05) 24 hours postoperatively. RAP and PCWP. The mean RAP in sinus rhythm remained unchanged with atrial pacing preoperatively and at all core temperatures postoperatively. Similarly, there was no significant difference between the sinus PCWP and the paced PCWP, except at 90° F, when the mean PCWP rose from 9.9 to 12.8 with pacing (p < 0.05). Postoperatively, the RAP was higher than the control value (p < 0.05), but the PCWP remained unchanged. MAP and SVRI. During the control period, the MAP was 99.6 mm Hg in sinus rhythm and rose to 110.6 mm Hg with atrial pacing (p < 0.05). At each stage postoperatively, the MAP remained unchanged with atrial pacing (p > 0.05). Compared to control values, there was no significant change in MAP during rewarming, either in sinus rhythm or during atrial pacing. The SVRI was 3,357 dyne . sec . em? . m' in sinus rhythm during the control period and decreased to 3,018 dyne. sec . cnr' . m' with pacing (p < 0.05). Postoperatively, the SVRI fell significantly with atrial pacing at
33°
(OF)
35°
TEMPERATURE Fig. 4. Correlation of stroke volume index with temperature during rewarming.
90° F and 24 hours postoperatively (p < 0.05). Compared to control values, the SVRI increased significantly at 90° F and decreased significantly at 98° F and 24 hours postoperatively, both in sinus rhythm and with atrial pacing (p < 0.05). LVSW. The sinus and paced LVSWs were significantly lower than control values at 90° F, 94°F, and 98° F (p < 0.05). By 24 hours postoperatively, the paced LVSW (but not the sinus LVSW) had returned to control levels. The LVSW fell significantly with atrial pacing during the control period and during rewarming at 90° F and 94° F (p < 0.05). Correlation of hemodynamics with temperature. There was a strong correlation between temperature and several hemodynamic variables during the rewarming period (90 0 F to 98° F). The cardiac index in sinus rhythm and the cardiac index during atrial pacing demonstrated a positive correlation with temperature (r = .71 and 0.66, respectively, Fig. 3), as did the sinus and paced stroke volume indices (r =0.33 and 0.66, Fig. 4) and the sinus and paced LVSWs (r = 0.37 and 0.73). The sinus rate correlated with temperature (r = 0.51). The SVRI was inversely correlated with temperature (r = -0.80 and -0.64) and with the cardiac index (r = -0.87; Fig. 5). The use of a uniform pacing rate (110 beats/min) eliminates simultaneous changes in
Volume 86 Number 2 August, 1983
heart rate as a confounding variable: Note that correlations of stroke volume index and LVSW with temperature were better with the paced values than with the sinus values. The simultaneous effects of time and temperature were examined by means of a multiple linear regression model. When multiple regression indicated that a hemodynamic variable could be predicted from temperature and/or time, temperature was consistently a better predictor. Variables strongly associated with temperature independent of time were cardiac index, stroke volume index, SVRI (all in sinus rhythm and with right atrial pacing), and the paced LVSW. No hemodynamic variable was strongly associated with time when temperature was controlled in the analysis. Discussion
Hypothermia exerts a protective effect on the myocardium through two principal mechanisms: a depression of cellular metabolism, resulting in a decrease in myocardial oxygen consumption and sparing of intracellular energy stores (adenosine triphosphate and creatine phosphate), and a reduction in the fluidity of cellular membranes, resulting in preservation of cellular integrity.3,6.12 However, not all of the cellular effects of hypothermia are beneficial; harmful effects of hypothermia include ionic redistribution, cellular swelling, and cold-induced injury.":" Although these harmful effects can be minimized by the use of multidose potassium cardioplegia, they cannot be completely eliminated. During reperfusion and rewarming, additional detrimental effects, especially inward calcium fluxes and additional cellular swelling, may appear. 19, 20 Reperfusion also has been identified as a cause of injury in intermittent ischemia and reperfusion. 21,22 Thus the postoperative rewarming state may be very different from the hypothermic "state occurring during cooling and cardioplegic arrest, in which the hemodynamic and metabolic effects of hypothermia have been extensively documented." The effects of· hypothermia during rewarming have not been adequately studied. An important animal study" has recently demonstrated significantly higher myocardial adenosine triphosphate and 'creatine phosphate levels after normothermic reperfusion than after hypothermic reperfusion. This observation implies that hypothermic reperfusion may adversely affect hemodynamic recovery of the heart following cardioplegic arrest. Wei have previously presented scintigraphic and hemodynamic evidence for transient left ventricular dysfunction after CABG in a group of 30 patients in whom myocardial preservation consisted of moderate
Hemodynamic dysfunction after CABG
9000
•
• Sinus rhythm
•
• .,
~ 5000
en
1.~~
~.
1000
23I
.
•••••
0
2 4 CARDIAC INDEX (L/min/M 2 )
Fig. 5, Inverse correlation between cardiac index and systemic vascular resistance index (SVJU. dyne. sec . cm" . m') during rewarming.
systemic hypothermia (20 0 to 28 0 C) and intermittent aortic cross-clamping without chemical cardioplegia. The LVEF decreased from 58% ± 2% (preoperatively) to 41% ± 1% (1 to 5 hours postoperatively). At the same time, a decrease in cardiac index and LVSW occurred despite an increase in PCWP. The LVEF, cardiac index, and LVSW returned to normal or supernormal values by the second postoperative day. More recently, Roberts and associates' presented similar evidence of transient left ventricular dysfunction in 40 patients who underwent CABG with hypothermic multidose potassium crystalloid cardioplegia. The LVEF fell from 50% ± 3% (preoperatively) to 38% ± 2% (2 hours postoperatively); the cardiac index and LVSW fell also. The LVEF and cardiac index returned to preoperative levels by 24 hours. Importantly, temperature was not noted in these studies. The present study was undertaken to determine whether this transient postoperative hemodynamic dysfunction occurred during the rewarming period following CABG in 12 patients with a normal global LVEF preoperatively. All patients received hypothermic multidose potassium crystalloid cardioplegia with systemic hypothermia (20 0 C). No patient had a myocardial infarction, sustained arrhythmia, or tamponade during the study period, factors which may produce left ventricular dysfunction. Administration of vasopressor, beta blocker, digitalis, and antiarrhythmic medication was strictly avoided during the study period. The present study demonstrates at 90 0 F a significant fall in cardiac index, stroke volume index, and LVSW and an increase in SVRI compared to control values. The subsequent changes in these hemodynamic variables during the rewarming period (90 F to 98 F) were correlated with temperature. By 98 F, these hemodynamic changes were largely corrected: The cardiac index and the paced 0
0
0
2 3 2 Czer et al.
stroke volume index returned to control levels (fixed pacing eliminating changes in heart rate as a confounding variable). Thus the hemodynamic dysfunction after myocardial revascularization occurs predominantly during the rewarming period and is strongly temperaturedependent. Myocardial oxygen debt accumulated during ischemic time because of imperfect myocardial preservation, and the detrimental effects of reperfusion alluded to earlier may contribute to the postoperative hemodynamic dysfunction independent of temperature. These detrimental effects should diminish with time elapsed after operation. Multiple regression analysis demonstrated that temperature was a consistently better predictor of hemodynamic function during the rewarming period than was time. Thus hypothermia exerts the dominant effect on early postoperative hemodynamic dysfunction. However, the fact that the paced LVSW did not return to control levels until 24 hours postoperatively, with only a small change in temperature (to 99° F), suggests that hypothermia alone does not explain all of the hemodynamic dysfunction that occurs postoperatively. Afterload exerts important effects during the rewarming process. There was a strong inverse relationship between cardiac index and SVRI during the rewarming period (Fig. 5). Thus, under the conditions of impaired left ventricular function that occur during rewarming, cardiac output is quite sensitive to changes in afterload. In the absence of changes in filling pressures, increases in SVRI might be due to hypothermia itself, increased catecholamines, or both. Plasma catecholamines, especially norepinephrine, increase markedly after CABG.24-26 There is a poor correlation between SVRI and catecholamine level during and after CABG. 26 During rewarming, the SVRI was inversely correlated with temperature (r = 0.80); this suggests that hypothermia plays a major role in the changes in SVRI that occur during early rewarming. Other factors that might account for transient left ventricular dysfunction include hypocalcemia and hemodilution. Hypocalcemia is present to a variable degree following CABGY We28 have previously shown in a group of patients not given citrate in the pump prime or parenteral calcium during operation that the postoperative decline in total calcium can be explained fully by the degree of hemodilution. Furthermore, ionized calcium is maintained at normal or near normal levels by homeostatic mechanisms, mainly the decrease in binding to albumin and the maintenance of parathyroid hormone secretion." In the present study, the postoperative drop in total calcium was less pronounced than in the previous study, because of the administration
The Journal of Thoracic and Cardiovascular Surgery
of 1 gm of calcium chloride at the end of the pump run; again, the degree of hypocalcemia could be accounted for totally by the degree of hemodilution. In addition, there was no significant difference between the total calcium levels at 90° F and at 24 hours postoperatively (Fig. I); thus the hypocalcemia is unlikely to explain the hemodynamic changes observed during the rewarming period. Hemodilution also occurs to a variable degree following CABG. 3 Hemodilution has been shown .to increase cardiac index. 29, 3o However, in our study, the cardiac index fell rather than increased in association with the early postoperative fall in hematocrit value. Furthermore, the subsequent increases in cardiac index during the rewarming period were not accompanied by any significant change in hematocrit value, which remained steady at 28% (Fig. 1). Thus hemodilution is unlikely to explain the observed hemodynamic changes during the rewarming period. Atrial pacing has been used to measure myocardial reserve following CABG and to induce ischemia during the immediate postoperative period." The pacing rates required to induce ischemia are generally much higher than 110 beats/min, the rate used in this study." There was no clinical or electrocardiographic evidence of pacing-induced ischemia during the present study, nor was there electrocardiographic or creatine kinase enzymatic evidence of a perioperative myocardial infarction. Although augmentation of cardiac index by atrial pacing was lost during the rewarming period, no patient had a clinically significant deterioration of hemodynamics (cardiac index or MAP) with right atrial pacing at 110 beats/min. In conclusion, the transient hemodynamic dysfunction after CABG occurs during the rewarming period despite maintenance of right atrial and pulmonary capillary wedge filling pressures. This dysfunction is characterized by an initial decrease in cardiac index, stroke volume index, LVSW, and LVEF, a loss of augmentation of cardiac index by atrial pacing, and an increase in heart rate and SVRI. The subsequent changes in cardiac index, stroke volume index, LVSW, SVRI, and heart rate correlate with the temperature changes during rewarming. Further, there is a striking inverse relationship between cardiac index and SVRI during the rewarming process. Atrial pacing at 110 beats/min adds no hemodynamic benefit during the rewarming period. In view of these fmdings, temperature must be considered in the evaluation of left ventricular and hemodynamic function following myocardial revascularization. We wish to thank Carolyn Conklin, R.N., Marjorie Raymond, R.N., and Michael DeRobertis, R.N., for their techni-
Volume 86 Number 2 August, 1983
cal assistance; Robert Kass, M.D., and Myles Lee, M.D., of the Department of Cardiovascular Surgery; and the nurses of the Cardiac Surgical Intensive Care Unit of Cedars-Sinai Medical Center. Special thanks to Morgan Stewart, M.S., of the Scientific Data Center, for help in the statistical analysis of the data. REFERENCES Gray R, Maddahi J, Berman D, Raymond M, Waxman A, Ganz W, Matloff J, Swan HJC: Scintigraphic and hemodynamic demonstration of transient left ventricular dysfunction immediately after uncomplicated coronary artery bypass grafting. J THoRAc CARDIOVASC SURG 77:504-510, 1979 2 Roberts AJ, Spies SM, Sanders JH, Morgan JM, Wilkinson CJ, Lichtenthal PR, White RL, Michaelis LL: Serial assessment of left ventricular performance following coronary artery bypass grafting. J THoRAc CARDIOVASC SURG 81:69-84, 1981 3 Hearse DJ, Braimbridge MV, Jynge P: Protection of the Ischemic Myocardium: Cardioplegia, New York, 1981, Raven Press 4 Shellock FG, Rubin SA: Simplified and highly accurate core temperature measurements. Med Progr Technol 8:187-188, 1982 5 Dixon WJ: BMDP Statistical Software, Berkeley, 1981, University of California Press 6 Archie JP, Kirklin JW: Effect of hypothermic perfusion on myocardial oxygen consumption and coronary resistance. Surg Forum 24:186-188, 1973 7 Buckberg GD, Brazier JR, Nelson RH, Goldstein SM, McConnell DH, Cooper N: Studies of the effects of hypothermia on regional myocardial blood flow and metabolism during cardiopulmonary bypass. I. The adequately perfused beating, fibrillating, and arrested heart. J THoRAc CARDIOVASC SURG 73:87-94, 1977 8 Chitwood WR, Sink JD, Hill RC, Wechsler AS, Sabiston DC: The effects of hypothermia on myocardial oxygen consumption and transmural coronary blood flow in the potassium-arrested heart. Ann Surg 190:106-116, 1979 9 Vinas JF, Fewel JG, Arom KV, Trinkle JK, Grover FL: Effects of systemic hypothermia on myocardial metabolism and coronary blood flow in the fibrillating heart. J THoRAc CARDIOVASC SURG 77:900-908, 1979 10 Blok MC, van Deenen LLM, de Gier J, Op den Kamp JAF, Verkleij AJ: Some aspects of lipid-phase transition on membrane permeability and lipid-protein association, Proceedings in Life Sciences. Biochemistry of Membrane Transport, Berlin, 1977, Springer-Verlag, pp 38-46 II Chapman D, Cornell BA, Quinn PJ: Phase transitions, protein aggregation and a new method for modulating membrane fluidity, Proceedings in Life Sciences. Biochemistry of Membrane Transport, Berlin, 1977, Springer-Verlag, pp 72-85 12 Inesi G, Millman M, Eletr S: Temperature-induced transitions of function and structure in sarcoplasmic reticulum membranes. J Mol BioI 81:483-504, 1973
Hemodynamic dysfunction after CABG
233
13 Martin DR, Scott DF, Downes GL, Belzer FO: Primary cause of unsuccessful liver and heart preservation. Cold sensitivity of the ATP-ase system. Ann Surg 175: 111-117, 1972 14 Belzer FO: Kidney preservation. Surg Clin North Am 58:261-272, 1978 15 Leaf A: Cell swelling. A factor in ischemic tissue injury. Circulation 48:455-458, 1973 16 MacKnight AC, Leaf A: Regulation of cellular volume. Physiol Rev 57:510-573, 1977 17 Pine MB, Kahne D, Jaski B, Apstein CS, Thorp K, Abelmann WH: Sodium permeability and myocardial resistance to cell swelling during metabolic blockade. Am J Physiol 239:H31-H39, 1980 18 Meryman HT: Freezing injury and its prevention in living cells. Ann Rev Biophys Bioeng 3:341-363, 1974 19 Hearse DJ: Reperfusion of the ischemic myocardium. J Mol Cell Cardiol 9:605-616, 1977 20 Zimmerman ANE, Daems W, Hulsmann WC, Snijder J, Wisse E, Durrer D: Morphological changes of heart muscle caused by successive perfusion with calcium-free and calcium-containing solutions (calcium paradox). Cardiovasc Res 1:201-209, 1967 21 Engelman RM, Chandra R, Baumann FG, Goldman RA: Myocardial reperfusion, a cause of ischemic injury during cardiopulmonary bypass. Surgery 80:266-276, 1976 22 Engelman RM, Rousou JH, O'Donoghue MJ, Longo F, Dobbs W A: A comparison of intermittent and continuous arrest for prolonged hypothermic cardioplegia. Ann Thorae Surg 29:217-233,1980 23 Rousou JH, Dobbs WA, Meeran MK, Engelman RM: The temperature dependence of recovery of metabolic function following hypothermic potassium cardioplegic arrest. J THORAC CARDIOVASC SURG 83:117-121, 1982 24 Kim YD, Jones M, Hanowell ST, Koch HP, Lees DE, Weise V, Kopin 11: Changes in peripheral vascular and cardiac sympathetic activity before and after coronary artery bypass surgery. Interrelationships with hemodynamic alterations. Am Heart J 102:972-979, 1981 25 Hoar PF, Stone JG, Faltas AN, Bendixen HH, Head RJ, Berkowitz BA: Hemodynamic and adrenergic responses to anesthesia and operation for myocardial revascularization. J THoRAc CARDIOVASC SURG 80:242-248, 1980 26 Reves JG, Karp RB, Buttner EE, Tosone S, Smith LR, Samuelson PN, Kreusch GR, Oparil S: Neuronal and adrenomedullary catecholamine release in response to cardiopulmonary bypass in man. Circulation 66:49-55, 1982 27 Moffitt EA, Tarhan S, Goldsmith RS, Pluth JR, McGoon DC: Patterns of total and ionized calcium and other electrolytes in plasma during and after cardiac surgery. J THoRAc CARDIOVASC SURG 65:751-757, 1973 28 Gray R, Braunstein G, Krutzik S, Conklin C, Matloff J: Calcium homeostasis during coronary bypass surgery. Circulation 62:Suppl 1:57, 1980 29 Fowler NO, Holmes JC: Blood viscosity and cardiac
2 3 4 Czer et al.
output in acute experimental anemia. J Appl Physiol 39:453-456, 1975 30 Clarke TNS, Prys-Roberts C, Biro G, Foex P, Bennett MJ: Aortic input impedance and left ventricular energetics in acute isovolemic anaemia. Cardiovasc Res 12:49-55, 1978 31 Hilton JD, Weisel RD, Baird RJ, Goldman BS, Jablonsky
The Journal of Thoracic and Cardiovascular Surgery
G, Pym J, ScuIly HE, Ivanov J, Mickle DAG, Feiglin DH, Morch JE, Mclaughlin PR: The hemodynamic and metabolic response to pacing after aortocoronary bypass. Circulation 64:(Suppl 2):48, 1981 32 Judgutt BI, Lee SJK, Taylor RF: Abolition of ischemic response to atrial pacing foIlowing aortocoronary bypass surgery. Circulation 54:225-229, 1976
Bound volumes available to subscribers Bound volumes of THE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1983 issues from the Publisher, at a cost of $58.50 ($73.00 international) for Vol. 85 (January-June) and Vol. 86 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 30 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany a/l orders. Contact Mr. Deans Lynch at The C. V. Mosby Company, 11830 Westline Industrial Drive, S1. Louis, Missouri 63146, USA. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.