Hemodynamic Jean-Jacques
and Hormonal Responses to Hypothermic Cardiopulmonary Bypass
and Normothermic
Lehot, MD, PhD, Jacques Villard, MD, Hector Piriz, MD, Daniel M. Philbin, MD, Pierre-Yves Carry,
MD, Guillemette Gauquelin, MD, PhD, Bruno Claustrat, MD, PhD, Genevieve Sassolas, MD, Jean Galliot, Perfusionist, Suzanne Estanove, MD Normothermic cardiopulmonary bypass (CPB) is used in cardiac surgery at some institutions. To compare hemodynamic and hormonal responses to hypothermic (29°C) and normothermic nonpulsatile CPB, 20 adults undergoing coronary artery bypass graft and/or aortic valve replacement were studied. Hemodynamic measurements and plasma hormone concentrations were obtained from preinduction to the third postoperative hour. The two groups were given similar amounts of anesthetics and vasodilators. Systemic vascular resistance increased only during hypothermic CPB, and heart rate was higher at the end of hypothermic CPB. Postoperative central venous pressure and pulmonary capillary wedge pressure were lower after hypothermic CPB. Oxygen consumption decreased by 45% during hypothermic CPB, did not change during normothermic CPB, but increased similarly in the two groups after surgery; mixed venous oxygen saturation (SvO,) was significantly lower during normother-
C
ARDIOPULMONARY bypass (CPB) with moderate hypothermia is used by most cardiac surgeons. As cooling to 30°C decreases oxygen consumption by 50%,’ moderate hypothermia affords protection from ischemia to major organs, such as the brain and myocardium, when perfusion flow or gas exchanges are limited by the pump or the oxygenator. However, hypothermic CPB takes longer because of the time required for cooling and rewarming. In addition, nonhomogenous rewarming can lead to a decrease in central temperature after hypothermic CPB with hemodynamic instability and shivering, which can enhance oxygen consumption by 300% to 400%.’ Pumps and oxygenators have become increasingly more reliable and allow high perfusion flow and efficient gas exchange. Because cold cardioplegia can afford satisfactory myocardial protection, some cardiac surgeons use normothermic CPB routinely.3-5 Since 1980, more than 1,500 cardiac operations with normothermic CPB have been performed by the same surgeon at this institution with
This article is accompanied by an editorial. Please see: Lell WA: Hot or Cold, Continuous or Intermittent? What Goes Around Comes Around. J Cardiothorac Vast Anesth 6:125-126,1992
From the Hipital Cardiovasculaire et Pneumologique L. Pradel; the Laboratoire de Physiologic de I’Environnement; and the Centre de Medecine Nucleaire, Lyons, France; the Pedeciba Hospital de Clinicas and Catedra de Fisiopathologia, Montevideo, Uruguay; and the Massachusetts General Hospital and Harvard Medical School, Boston, MA. Address reprint requests to Jean-Jacques Lehot, MD, PhD, Hopital Cardiovasculaire et Pneumologique L. Pradel, B.P. Lyon-Monchat, 69394 Lyon, Cedex 03, France. Copyright 0 1992 by Cc:B. Saunders Company 1053-077019210602-0003$03.00/O 132
mic CPB. Urine output and composition were similar in the two groups. In both groups, plasma epinephrine, norepinephrine, renin activity, and arginine vasopressin concentrations increased during and after CPB. However, epinephrine, norepinephrine, and dopamine were 200%. 202%. and 165% higher during normothermic CPB than during hypothermic CPB, respectively. Dopamine and prolactin increased significantly during normothermic but not hypothermic CPB. Atrial natriuretic peptide increased at the end of CPB and total thyroxine decreased during and after CPB, with no difference between groups. This study suggests that higher systemic vascular resistance during hypothermic CPB is not caused by hormonal changes, but might be caused by other factors such as greater blood viscosity. A higher perfusion index during normothermic CPB might have allowed higher SvO,. Copyright o 1992 by W.B. Saunders Company
satisfactory clinical results. Previous reports have demonstrated that hypothermia could modify hemodynamic and hormonal status.6.7 The present study was carried out to compare the effects of hypothermic and normothermic CPB. MATERIALS
AND METHODS
After obtaining approval of the ethical committee of the institution and written informed patient consent, 20 adults underwent coronary artery bypass graft (CABG) surgery (n = 15) and/or aortic valvular replacement (n = 7). Patients with respiratory failure, endocrinopathy, cerebral disease, amiodarone or clonidine treatment, or requiring emergency surgery were excluded. Because of the diurnal endocrine variations, operations on study patients were started at 8:OO am. All cardiac medications were continued on the day before surgery. Oral diazepam, 0.35 mgikg, was given as premeditation 1 hour before induction of anesthesia. The induction consisted of fentanyl, 50 pg/kg, diazepam, 0.25 mgikg. and pancuronium, 0.1 mgikg. Anesthesia was maintained with increments of fentanyl and diazepam. Artificial ventilation was provided by a Servo ventilator A (Siemens-Elema, Sweden) (Rate = lo/mitt; VT = 10 mL/kg: oxygen = 100%). Radial artery and pulmonary artery catheters (PAC) were inserted before induction under local anesthesia. Central temperatures were monitored through rectal and PAC probes. Ringer’s lactated solution and vasodilators (nitroglycerin. sodium nitroprnsside) were administered as needed. After the injection of heparin, 300 U/kg, the venous cannula was inserted into the right atrium and the arterial cannula into the root of the ascending aorta. Before CPB, the patients were randomly allocated to one of two groups: group 1 (n = 10) underwent hypothermic CPB (injection temperature = 25°C) and group 2 (n = 10) normothermic CPB (injection temperature = 37°C). A CML membrane oxygenator (Cobe, Arvada, CO) and a nonpulsatile pump (Sarns, Ann Arbor, MI) were used. The preparation consisted of Ringer’s lactated solution (1,500 mL), sodium bicarbonate 1.4% (300 mL), and heparin (5,000 U). Saint Thomas’ Hospital crystalloid cardioplegic solution’ was injected at 4°C into the aortic root immediately after aortic cross-clamping and every 20 minutes during clamping. As is standard practice in
Journalof Cardiothoracic and Vascular Anesthesia, Vol6. No 2 (April), 1992: pp 132.139
HYPO VERSUS NORMOTHERMIC
CPB
this institution, mean arterial pressure (MAP) was maintained between 50 and 80 mm Hg by injections of nitroglycerin. Vasopressor agents were not used in these patients. A heating mattress (38°C) was continuously used, except during cooling in group 1. In this group, rewarming did not begin until after 30 minutes of CPB; patients were rewarmed to 37°C (pulmonary artery temperature) and 33°C (rectal temperature). Reversal of heparin was accomplished with protamine sulfate, 4 mg/kg; aprotinin, 2,000,OOO protease inhibitory units (Iniprol, Choay, Paris, France), was simultaneously administered. After transfer to the intensive care unit (ICU) the patients were ventilated overnight with the same type of ventilator adjusted according to blood gases. Inspired oxygen concentration was increased to 100% 15 minutes before measurements to allow comparisons with intraoperative data. Thereafter, inspired oxygen concentration was returned to 50%. Sedation and muscle relaxants were administered when needed (morphine chlorhydrate, diazepam, pancuronium). Dextrose, 10 g/L by infusion, was started on arrival in the ICU. Hemodynamic parameters and body temperatures were measured immediately before drawing blood samples at the following time intervals: Control: (A) 15 minutes after insertion of catheters; (B) 5 minutes after tracheal intubation, before incision; (C) 15 minutes; (D) 30 minutes after starting CPB; (E) at termination of CPB when cardiac mechanical activity had resumed; and (F) immediately; (G) 1 hour, and (H) 3 hours after closure of the chest. Cardiac output was measured by thermodilution (mean of 3 measurements differing less than 10%) and perfusion output was read on the precalibrated pump. Arterial blood was sampled at these same intervals for acid-base status, blood gases (alpha-stat, not corrected for patient’s temperature), hematocrit, and plasma hormone concentrations. At periods B, D, and G additional arterial and mixed venous blood samples were drawn for oxygen content (Oxymeter, Radiometer, Copenhagen, Denmark) to allow calculation of total body oxygen delivery and consumption. Calculated parameters were obtained using standard formulae.’ Blood lactate levels were measured at the onset and completion of CPB. After centrifugation, the plasma was frozen at -40°C for later assay of the following hormones: epinephrine (EPI), norepinephrine (NE), and dopamine (DA) by high-pressure liquid chromatography with electrochemical detection after extraction on alumine’“; arginine-vasopressin (AVP), plasma renin activity (PRA), atria1 natriuretic peptide (ANP), total thyroxin (lT4), and prolactin by radioimmunoassays.“.” Plasma catecholamines were not assayed when patients received exogenous catecholamines. During a 4-hour period following surgery, the episodes of shivering, hypertension (MAP > 20% of control), hypotension (MAP < 20% of control), and tachycardia (heart rate > 20% of control) were recorded when they lasted more than 5 minutes by a nurse unaware of the patient’s group. Urine electrolytes and volumes were measured intraoperatively and urine output was recorded during the 4 postoperative hours. Parametric data were analyzed by two-way analysis of variance and paired t-tests for intragroup comparisons with levels of significance adjusted to control for multiplicity (Bonferroni), and unpaired t-tests for intergroup comparisons. Nonparametric data were analyzed by the Wilcoxon rank sign test. Correlations were
determined cant when
133
tive treatment, type, and duration of surgery (Table 1). However, CPB and aortic clamping duration were significantly longer in group 1. No significant differences in the amount of injected anesthetics and muscle relaxants were found. During or after CPB, 10 patients were given nitroglycerin (5 in each group), 3 were given sodium nitroprusside (2 in group 1,l in group 2) 3 received dobutamine (1 in group 1,2 in group 2), and 1 patient in group 1 received lidocaine because of ventricular premature beats. During the 4 postoperative hours, the number of hypertensive and hypotensive episodes was not significantly different between groups, but the number of tachycardic episodes was significantly greater in group 1, The number of patients given sedation, muscle relaxants, or vasodilators was not significantly different in the two groups. One group-l patient died during the first postoperative hour because of a massive hemorrhage originating from the ascending aorta. Two group-2 patients presented with electrocardiogram (ECG) (new Q Table 1. Data for 20 Patients Undergoing Cardiac Surgeryt Group 1
No. of patients
IO/O
Age (yrs) Weight (kg)
62 k 2 70 2 3
812 60 ? 3 74 * 3
Height (cm)
169 2 1
171 + 2
NYHA class II
4
5
class Ill
6
5
Preoperative treatment Diuretics
2
1
Nitrates
3
3
Beta-Blockers
2
5
Calcium Channel Blockers
6
6
Surgery CABGIAVR Surgery time (min)
715 153 + 9
8/Z 154 + 12
CPB time (min)
85 + 6
58 + 6*
Clamping time (min)
57 k 6
42 2 3*
No. of countershocks
9
5
5.3 2 0.3
5.2 2 0.3
Diazepam (mg)
17 + 4
19 + 4
Pancuronium (mg)
17 * 4
16 ? 4
Fentanyl (mg)
Patients administered Nitroglycerin
7
6
Sodium nitroprusside
2
3
4-hour postoperative episodes Hypertension
17
8
Hypotension
27
19
Tachycardia
39
6*
9
7
Shivering Patients administered Morphine Diazepam Pancuronium
The results were expressed as pH for the reasons stated by Feinstein.‘”
Nitroglycerin
mean f SEM, including
10
Male/female
by least squares regression. Differences were signifiP was less than 0.05.
IO
Group 2
Sodium nitmprusside Abbreviations: NYHA, New York Heart Association; CABG, Coronary
RESULTS
The groups were not significantly different for anthropometric data, New York Heart Association class, preopera-
artery bypass graft; AVR, aortic valve replacement; CPB, cardiopulmonary bypass. *Different from group 1 (P < 0.05). tMean + SEM.
iEHOT
134
Fig 1. Evolution of pulmonary artery (PA) and at all sampling times (A to H, see MATERIALS hypothermic (A-A) and in normothermic (0-O) tics: Intergroup differences: +, P < 0.05; ++. P
0.05). In both groups, plasma EPI and NE concentrations increased from the beginning of CPB throughout the study duration, but were significantly higher in group 2 during CPB (Fig 2). EPI, NE, and DA were 200%, 202%, and 165%, respectively, higher in the normothermic than in the hypothermic group 15 minutes after starting CPB. Plasma DA concentrations increased during normothermic CPB and postoperatively in the two groups. Plasma renin activity increased similarly in both groups during CPB, with a peak at the end of CPB, then decreased and reached a plateau above control levels (Fig 2). Plasma arginine vasopressure concentrations did not change after induction (Fig 3). In
Table 2. Biochemical Data (Arterial Blood) A
Group
Hematocrit PH PaO,(mmHg) PaCO,(mm
Hg)
Plasma bicarbonate (mmol/L)
El
C
D
-. E
F
G
H
1
0.41 k 0.01 0.38 k O.Olt 0.27 k 0.006t
2
0.40 f 0.01 0.36 +-0.01
0.25 -c0.007t.S 0.26 k 0.007t 0.27 f O.lt
0.26 + 0.009t 0.33 t O.Olt 0.29 + 0.009t 0.35 2 O.Olt 0.36 + O.Olt 0.31 -+O.Olt
1
7.41 rt0.01 7.46 k 0.01
7.43 t 0.01
7.42 t 0.001
7.45 + 0.01
7.36 2 0.004
7.30 k 0.04$ 7.33 t 0.04
2
7.41 r 0.01 7.43 k 0.01
7.46 * 0.01
7.45 + 0.02
7.43 + 0.02
7.40 r 0.01
7.40 k 0.02
0.34 k O.Olt 0.33 t O.Olt 7.39 t-0.02
1
76 + 5
383? 23t
338 + 24t
307k 31t
221 2 25t
302 -t44t
292 *30t
309 k 27t
2
87 2 7
359 2 18t
252 + 31t
262 + 47t
317*42t
312 + 51t
345 + 31t
329 2 27t
1
4422
41 + 2
38 f 2
36klt
33 2 2t
42 + 3
47 + 4
46 f 4
2
41 2 0.3
38 2 2
35 k 2t
35 -+2t
37 2 3
38 k 2
39 + 3
39 -t3
1
26.9 + 1.1
25.4 t 1.0
25.2 2 0.9
23.5 k 0.8t
22.1 2 0.8t
23.2 2 0.9t
24.1 -t0.7t
24.7 k 0.7t
2
24.7 -t2.2
24.8 2 0.6
24.4 + 0.9
24.2 k 0.7
24.2 2 1.0
23.5 + 0.5t
23.8 r 0.4t
24.2 k 0.6
?? Differentfromgroupl tDifferentfromcontrol(A) *Differentfrom postinduction(B)(P< 0.05)
HYPO VERSUS NORMOTHERMIC
135
CPB
Table 3. Hemodynamic Data CPB A
B
68 + 3
65 2 4
63 k 3
64 2 4
90 2 4
70 + 36
68 + 5
78 -+ 6
98 + 46
67 k 5t
63 2 5t
64 2 !jt
Group
HR (beats/min) MAP
10125
(mm Hg) MPAP
C
D
E-
96 + 6
86 2 6’t
101 + 4t
93 + 3
96
59 -C 5t
80 2 3
93 2 3
76 f 4
17.5 +- 1
(mm Hg) CVP
17.0 2 1.5
18.9 2 1.0
6 f 0.9
7 t- 0.6
4&l
5 + 0.3
5 -c ‘I
(mm Hg) PCWP
7 2 0.8
9 2 0.1
6 k 0.7
6 + 0.7
622
(dyne.
s
cm5. m’)
15.2 f 1.6
5 +- 0.9
3 f 0.8
61kl
7 r If
7 + 0.89 + 1
11 -tl
13 2 1
12k 1.6 + O.lt
1
2.2 + 0.2
5 + 2t
5+
1t
10 2 1*
9*
1’
2.8 + 0.3
2.9 2 0.3
2.7 k 0.3
1.8 + O.lt
1.8 k 0.2t
2.0 + 0.3
2.9 f 0.2
2389 + 216
2890 ? 217t
3738 f 389t
2574 k 250
2967 ? 295
2046 2 148
2670 -c 383
2985 ? 429
2722 f 410
2491 + 352*
3139 + 445
2320 f 180x
1989 2 138
189 ? 46
236 -c 28
233 ? 38
291 + 28
281 + 61
150 * 35
175 2 44
196 r 68
177 2 43
177 2 26
. s . cm5 t m2)
Abbreviations:
16.4 + 0.9
8 f 0.6
3.2 2 0.4
PVRI (dyne
16.9 ‘- 2.0
11 + 1 1.9 t 0.1t
76 k 5t 13.5 * 1.3
11 + 1 2.2 -c 0.1
82 ? 3*t 15.2 ‘- 1.9
2219 + 202
(L/min/m2) SVRI
79 2 3t
2 5t
17.4 + 0.7
11 r1 3.0 2 0.2
H
100 + 7t
92 + 4t
18.1 2 1.6
(mm Hg) Cl/PI
G
F
107 -e 7t
2.8 + 0.1
HR. heart rate; MAP, mean arterial pressure; MPAP, mean pulmonary artery pressure; CVP, central venous pressure; PCWP,
pulmonary capillary wedge pressure; Cl, cardiac index; PI, perfusion index during CPB; SVRI, systemic vascular resistance index; PVRI, pulmonary vascular resistance index. ‘Differentfrom
group 1 (P < 0.05)
tDifferent from control (A)
both groups, a 32- to 40-fold increase occurred 15 minutes after starting CPB, then remained elevated throughout the study period. Plasma AVP concentrations did not correlate with SVRI during and after CPB. At control (Fig 3), plasma atria1 natriuretic peptide concentrations were above the normal range (20 to 30 pg/mL) and were correlated with pulmonary capillary wedge pressure (PCWP) (r = 0.63, P < O.Ol), but not with right atria1 pressure (r = 0.05, P > 0.05). This correlation disappeared after induction. After a decrease at induction, only significant in group 1, plasma ANP levels were stable in both groups despite hemodilution during CPB. When the heart was filled and had resumed contractions at the end of CPB, plasma ANP increased by 102% (P < 0.01) in both groups, then decreased progressively to control levels. After a decrease at induction in group 1, TT4 decreased Table 4. Oxygen Transport Data 1 h After Group DO,
(mL/min/m’) VO, (mL/min/m7 svo, (%) C(a-v)O, (mL/lOO mL) 0, extraction (%I
Before
CPB
ClOSUre
CPB
(30 mid
of Chest
1
372 f 20
194 + 7t
432 + 38
2
475 f 62
223 f 13t
410 + 31
1
80 f 4
44 + 9t
129 k 14t
2
71 k-9
80 k 7*
122 r 15t
1
78 + 1
76 2 4
68 2 3t
2
80 2 2
61 + 3*t
68 f 2t
1
3.7 2 0.3
2.8 -t- 0.6
4.8 2 0.4t
2
2.7 -t 0.3
4.3 k 0.4*t
4.2 2 0.3t
1
21 2 1
22 2 4
31 It 3t
2
16 + 2
36 2 3”t
30 -c 2t
Abbreviations: DO,, oxygen delivery index; VO,, oxygen consumption index; SvO,, mixed venous oxygen saturation; C(a-v)D,, arterialvenous oxygen content difference. *Different from Group 1 (P < 0.05) tDifferent from control (A)
by 38% and 44% in groups 1 and 2, respectively, 15 minutes after starting CPB (Fig 3); while hematocrit decreased by 34% and 37%, respectively. In both groups, ‘IT4 remained stable at 30 minutes of CPB then increased progressively without reaching control levels at the third postoperative hour. Plasma prolactin concentrations increased significantly in the two groups after induction of anesthesia (Fig 3) and during normothermic CPB, then decreased progressively in both groups. Diuresis and urine composition were not significantly different between the two groups (Table 5). No correlation between plasma ANP concentrations and urine volume or electrolytes was observed. DISCUSSION
The present study was designed to compare hemodynamic and hormonal variations with hypothermic and normothermic CPB in elective cardiac operations performed by the same surgeon. Except for the temperature of CPB, all the patients’ management was similar according to the standard routine in the institution. Aortic cross-clamping and CPB durations were 26% and 32%, respectively, longer in the hypothermic CPB group, most likely caused by the time for rewarming before the completion of CPB and two combined procedures (CABG and AVR) in group 1. The numbers of countershocks, hypertensive, hypotensive, tachycardic, and shivering postoperative episodes were greater in group 1, but the difference was only significant for tachycardia. Despite rewarming during CPB (similar pulmonary artery temperatures in both groups at the end of CPB), rectal temperature was lower in group 1 than in group 2 at the end of CPB, and pulmonary artery temperature was lower in group 1 at the first and third postoperative hours. The postoperative rewarming may induce substantial vasodilatation and, as
136
LEHOT
I- -cp
and Hormonal Responses to Hypothermic Cardiopulmonary Bypass
and Normothermic
Lehot, MD, PhD, Jacques Villard, MD, Hector Piriz, MD, Daniel M. Philbin, MD, Pierre-Yves Carry,
MD, Guillemette Gauquelin, MD, PhD, Bruno Claustrat, MD, PhD, Genevieve Sassolas, MD, Jean Galliot, Perfusionist, Suzanne Estanove, MD Normothermic cardiopulmonary bypass (CPB) is used in cardiac surgery at some institutions. To compare hemodynamic and hormonal responses to hypothermic (29°C) and normothermic nonpulsatile CPB, 20 adults undergoing coronary artery bypass graft and/or aortic valve replacement were studied. Hemodynamic measurements and plasma hormone concentrations were obtained from preinduction to the third postoperative hour. The two groups were given similar amounts of anesthetics and vasodilators. Systemic vascular resistance increased only during hypothermic CPB, and heart rate was higher at the end of hypothermic CPB. Postoperative central venous pressure and pulmonary capillary wedge pressure were lower after hypothermic CPB. Oxygen consumption decreased by 45% during hypothermic CPB, did not change during normothermic CPB, but increased similarly in the two groups after surgery; mixed venous oxygen saturation (SvO,) was significantly lower during normother-
C
ARDIOPULMONARY bypass (CPB) with moderate hypothermia is used by most cardiac surgeons. As cooling to 30°C decreases oxygen consumption by 50%,’ moderate hypothermia affords protection from ischemia to major organs, such as the brain and myocardium, when perfusion flow or gas exchanges are limited by the pump or the oxygenator. However, hypothermic CPB takes longer because of the time required for cooling and rewarming. In addition, nonhomogenous rewarming can lead to a decrease in central temperature after hypothermic CPB with hemodynamic instability and shivering, which can enhance oxygen consumption by 300% to 400%.’ Pumps and oxygenators have become increasingly more reliable and allow high perfusion flow and efficient gas exchange. Because cold cardioplegia can afford satisfactory myocardial protection, some cardiac surgeons use normothermic CPB routinely.3-5 Since 1980, more than 1,500 cardiac operations with normothermic CPB have been performed by the same surgeon at this institution with
This article is accompanied by an editorial. Please see: Lell WA: Hot or Cold, Continuous or Intermittent? What Goes Around Comes Around. J Cardiothorac Vast Anesth 6:125-126,1992
From the Hipital Cardiovasculaire et Pneumologique L. Pradel; the Laboratoire de Physiologic de I’Environnement; and the Centre de Medecine Nucleaire, Lyons, France; the Pedeciba Hospital de Clinicas and Catedra de Fisiopathologia, Montevideo, Uruguay; and the Massachusetts General Hospital and Harvard Medical School, Boston, MA. Address reprint requests to Jean-Jacques Lehot, MD, PhD, Hopital Cardiovasculaire et Pneumologique L. Pradel, B.P. Lyon-Monchat, 69394 Lyon, Cedex 03, France. Copyright 0 1992 by Cc:B. Saunders Company 1053-077019210602-0003$03.00/O 132
mic CPB. Urine output and composition were similar in the two groups. In both groups, plasma epinephrine, norepinephrine, renin activity, and arginine vasopressin concentrations increased during and after CPB. However, epinephrine, norepinephrine, and dopamine were 200%. 202%. and 165% higher during normothermic CPB than during hypothermic CPB, respectively. Dopamine and prolactin increased significantly during normothermic but not hypothermic CPB. Atrial natriuretic peptide increased at the end of CPB and total thyroxine decreased during and after CPB, with no difference between groups. This study suggests that higher systemic vascular resistance during hypothermic CPB is not caused by hormonal changes, but might be caused by other factors such as greater blood viscosity. A higher perfusion index during normothermic CPB might have allowed higher SvO,. Copyright o 1992 by W.B. Saunders Company
satisfactory clinical results. Previous reports have demonstrated that hypothermia could modify hemodynamic and hormonal status.6.7 The present study was carried out to compare the effects of hypothermic and normothermic CPB. MATERIALS
AND METHODS
After obtaining approval of the ethical committee of the institution and written informed patient consent, 20 adults underwent coronary artery bypass graft (CABG) surgery (n = 15) and/or aortic valvular replacement (n = 7). Patients with respiratory failure, endocrinopathy, cerebral disease, amiodarone or clonidine treatment, or requiring emergency surgery were excluded. Because of the diurnal endocrine variations, operations on study patients were started at 8:OO am. All cardiac medications were continued on the day before surgery. Oral diazepam, 0.35 mgikg, was given as premeditation 1 hour before induction of anesthesia. The induction consisted of fentanyl, 50 pg/kg, diazepam, 0.25 mgikg. and pancuronium, 0.1 mgikg. Anesthesia was maintained with increments of fentanyl and diazepam. Artificial ventilation was provided by a Servo ventilator A (Siemens-Elema, Sweden) (Rate = lo/mitt; VT = 10 mL/kg: oxygen = 100%). Radial artery and pulmonary artery catheters (PAC) were inserted before induction under local anesthesia. Central temperatures were monitored through rectal and PAC probes. Ringer’s lactated solution and vasodilators (nitroglycerin. sodium nitroprnsside) were administered as needed. After the injection of heparin, 300 U/kg, the venous cannula was inserted into the right atrium and the arterial cannula into the root of the ascending aorta. Before CPB, the patients were randomly allocated to one of two groups: group 1 (n = 10) underwent hypothermic CPB (injection temperature = 25°C) and group 2 (n = 10) normothermic CPB (injection temperature = 37°C). A CML membrane oxygenator (Cobe, Arvada, CO) and a nonpulsatile pump (Sarns, Ann Arbor, MI) were used. The preparation consisted of Ringer’s lactated solution (1,500 mL), sodium bicarbonate 1.4% (300 mL), and heparin (5,000 U). Saint Thomas’ Hospital crystalloid cardioplegic solution’ was injected at 4°C into the aortic root immediately after aortic cross-clamping and every 20 minutes during clamping. As is standard practice in
Journalof Cardiothoracic and Vascular Anesthesia, Vol6. No 2 (April), 1992: pp 132.139
HYPO VERSUS NORMOTHERMIC
CPB
this institution, mean arterial pressure (MAP) was maintained between 50 and 80 mm Hg by injections of nitroglycerin. Vasopressor agents were not used in these patients. A heating mattress (38°C) was continuously used, except during cooling in group 1. In this group, rewarming did not begin until after 30 minutes of CPB; patients were rewarmed to 37°C (pulmonary artery temperature) and 33°C (rectal temperature). Reversal of heparin was accomplished with protamine sulfate, 4 mg/kg; aprotinin, 2,000,OOO protease inhibitory units (Iniprol, Choay, Paris, France), was simultaneously administered. After transfer to the intensive care unit (ICU) the patients were ventilated overnight with the same type of ventilator adjusted according to blood gases. Inspired oxygen concentration was increased to 100% 15 minutes before measurements to allow comparisons with intraoperative data. Thereafter, inspired oxygen concentration was returned to 50%. Sedation and muscle relaxants were administered when needed (morphine chlorhydrate, diazepam, pancuronium). Dextrose, 10 g/L by infusion, was started on arrival in the ICU. Hemodynamic parameters and body temperatures were measured immediately before drawing blood samples at the following time intervals: Control: (A) 15 minutes after insertion of catheters; (B) 5 minutes after tracheal intubation, before incision; (C) 15 minutes; (D) 30 minutes after starting CPB; (E) at termination of CPB when cardiac mechanical activity had resumed; and (F) immediately; (G) 1 hour, and (H) 3 hours after closure of the chest. Cardiac output was measured by thermodilution (mean of 3 measurements differing less than 10%) and perfusion output was read on the precalibrated pump. Arterial blood was sampled at these same intervals for acid-base status, blood gases (alpha-stat, not corrected for patient’s temperature), hematocrit, and plasma hormone concentrations. At periods B, D, and G additional arterial and mixed venous blood samples were drawn for oxygen content (Oxymeter, Radiometer, Copenhagen, Denmark) to allow calculation of total body oxygen delivery and consumption. Calculated parameters were obtained using standard formulae.’ Blood lactate levels were measured at the onset and completion of CPB. After centrifugation, the plasma was frozen at -40°C for later assay of the following hormones: epinephrine (EPI), norepinephrine (NE), and dopamine (DA) by high-pressure liquid chromatography with electrochemical detection after extraction on alumine’“; arginine-vasopressin (AVP), plasma renin activity (PRA), atria1 natriuretic peptide (ANP), total thyroxin (lT4), and prolactin by radioimmunoassays.“.” Plasma catecholamines were not assayed when patients received exogenous catecholamines. During a 4-hour period following surgery, the episodes of shivering, hypertension (MAP > 20% of control), hypotension (MAP < 20% of control), and tachycardia (heart rate > 20% of control) were recorded when they lasted more than 5 minutes by a nurse unaware of the patient’s group. Urine electrolytes and volumes were measured intraoperatively and urine output was recorded during the 4 postoperative hours. Parametric data were analyzed by two-way analysis of variance and paired t-tests for intragroup comparisons with levels of significance adjusted to control for multiplicity (Bonferroni), and unpaired t-tests for intergroup comparisons. Nonparametric data were analyzed by the Wilcoxon rank sign test. Correlations were
determined cant when
133
tive treatment, type, and duration of surgery (Table 1). However, CPB and aortic clamping duration were significantly longer in group 1. No significant differences in the amount of injected anesthetics and muscle relaxants were found. During or after CPB, 10 patients were given nitroglycerin (5 in each group), 3 were given sodium nitroprusside (2 in group 1,l in group 2) 3 received dobutamine (1 in group 1,2 in group 2), and 1 patient in group 1 received lidocaine because of ventricular premature beats. During the 4 postoperative hours, the number of hypertensive and hypotensive episodes was not significantly different between groups, but the number of tachycardic episodes was significantly greater in group 1, The number of patients given sedation, muscle relaxants, or vasodilators was not significantly different in the two groups. One group-l patient died during the first postoperative hour because of a massive hemorrhage originating from the ascending aorta. Two group-2 patients presented with electrocardiogram (ECG) (new Q Table 1. Data for 20 Patients Undergoing Cardiac Surgeryt Group 1
No. of patients
IO/O
Age (yrs) Weight (kg)
62 k 2 70 2 3
812 60 ? 3 74 * 3
Height (cm)
169 2 1
171 + 2
NYHA class II
4
5
class Ill
6
5
Preoperative treatment Diuretics
2
1
Nitrates
3
3
Beta-Blockers
2
5
Calcium Channel Blockers
6
6
Surgery CABGIAVR Surgery time (min)
715 153 + 9
8/Z 154 + 12
CPB time (min)
85 + 6
58 + 6*
Clamping time (min)
57 k 6
42 2 3*
No. of countershocks
9
5
5.3 2 0.3
5.2 2 0.3
Diazepam (mg)
17 + 4
19 + 4
Pancuronium (mg)
17 * 4
16 ? 4
Fentanyl (mg)
Patients administered Nitroglycerin
7
6
Sodium nitroprusside
2
3
4-hour postoperative episodes Hypertension
17
8
Hypotension
27
19
Tachycardia
39
6*
9
7
Shivering Patients administered Morphine Diazepam Pancuronium
The results were expressed as pH for the reasons stated by Feinstein.‘”
Nitroglycerin
mean f SEM, including
10
Male/female
by least squares regression. Differences were signifiP was less than 0.05.
IO
Group 2
Sodium nitmprusside Abbreviations: NYHA, New York Heart Association; CABG, Coronary
RESULTS
The groups were not significantly different for anthropometric data, New York Heart Association class, preopera-
artery bypass graft; AVR, aortic valve replacement; CPB, cardiopulmonary bypass. *Different from group 1 (P < 0.05). tMean + SEM.
iEHOT
134
Fig 1. Evolution of pulmonary artery (PA) and at all sampling times (A to H, see MATERIALS hypothermic (A-A) and in normothermic (0-O) tics: Intergroup differences: +, P < 0.05; ++. P
0.05). In both groups, plasma EPI and NE concentrations increased from the beginning of CPB throughout the study duration, but were significantly higher in group 2 during CPB (Fig 2). EPI, NE, and DA were 200%, 202%, and 165%, respectively, higher in the normothermic than in the hypothermic group 15 minutes after starting CPB. Plasma DA concentrations increased during normothermic CPB and postoperatively in the two groups. Plasma renin activity increased similarly in both groups during CPB, with a peak at the end of CPB, then decreased and reached a plateau above control levels (Fig 2). Plasma arginine vasopressure concentrations did not change after induction (Fig 3). In
Table 2. Biochemical Data (Arterial Blood) A
Group
Hematocrit PH PaO,(mmHg) PaCO,(mm
Hg)
Plasma bicarbonate (mmol/L)
El
C
D
-. E
F
G
H
1
0.41 k 0.01 0.38 k O.Olt 0.27 k 0.006t
2
0.40 f 0.01 0.36 +-0.01
0.25 -c0.007t.S 0.26 k 0.007t 0.27 f O.lt
0.26 + 0.009t 0.33 t O.Olt 0.29 + 0.009t 0.35 2 O.Olt 0.36 + O.Olt 0.31 -+O.Olt
1
7.41 rt0.01 7.46 k 0.01
7.43 t 0.01
7.42 t 0.001
7.45 + 0.01
7.36 2 0.004
7.30 k 0.04$ 7.33 t 0.04
2
7.41 r 0.01 7.43 k 0.01
7.46 * 0.01
7.45 + 0.02
7.43 + 0.02
7.40 r 0.01
7.40 k 0.02
0.34 k O.Olt 0.33 t O.Olt 7.39 t-0.02
1
76 + 5
383? 23t
338 + 24t
307k 31t
221 2 25t
302 -t44t
292 *30t
309 k 27t
2
87 2 7
359 2 18t
252 + 31t
262 + 47t
317*42t
312 + 51t
345 + 31t
329 2 27t
1
4422
41 + 2
38 f 2
36klt
33 2 2t
42 + 3
47 + 4
46 f 4
2
41 2 0.3
38 2 2
35 k 2t
35 -+2t
37 2 3
38 k 2
39 + 3
39 -t3
1
26.9 + 1.1
25.4 t 1.0
25.2 2 0.9
23.5 k 0.8t
22.1 2 0.8t
23.2 2 0.9t
24.1 -t0.7t
24.7 k 0.7t
2
24.7 -t2.2
24.8 2 0.6
24.4 + 0.9
24.2 k 0.7
24.2 2 1.0
23.5 + 0.5t
23.8 r 0.4t
24.2 k 0.6
?? Differentfromgroupl tDifferentfromcontrol(A) *Differentfrom postinduction(B)(P< 0.05)
HYPO VERSUS NORMOTHERMIC
135
CPB
Table 3. Hemodynamic Data CPB A
B
68 + 3
65 2 4
63 k 3
64 2 4
90 2 4
70 + 36
68 + 5
78 -+ 6
98 + 46
67 k 5t
63 2 5t
64 2 !jt
Group
HR (beats/min) MAP
10125
(mm Hg) MPAP
C
D
E-
96 + 6
86 2 6’t
101 + 4t
93 + 3
96
59 -C 5t
80 2 3
93 2 3
76 f 4
17.5 +- 1
(mm Hg) CVP
17.0 2 1.5
18.9 2 1.0
6 f 0.9
7 t- 0.6
4&l
5 + 0.3
5 -c ‘I
(mm Hg) PCWP
7 2 0.8
9 2 0.1
6 k 0.7
6 + 0.7
622
(dyne.
s
cm5. m’)
15.2 f 1.6
5 +- 0.9
3 f 0.8
61kl
7 r If
7 + 0.89 + 1
11 -tl
13 2 1
12k 1.6 + O.lt
1
2.2 + 0.2
5 + 2t
5+
1t
10 2 1*
9*
1’
2.8 + 0.3
2.9 2 0.3
2.7 k 0.3
1.8 + O.lt
1.8 k 0.2t
2.0 + 0.3
2.9 f 0.2
2389 + 216
2890 ? 217t
3738 f 389t
2574 k 250
2967 ? 295
2046 2 148
2670 -c 383
2985 ? 429
2722 f 410
2491 + 352*
3139 + 445
2320 f 180x
1989 2 138
189 ? 46
236 -c 28
233 ? 38
291 + 28
281 + 61
150 * 35
175 2 44
196 r 68
177 2 43
177 2 26
. s . cm5 t m2)
Abbreviations:
16.4 + 0.9
8 f 0.6
3.2 2 0.4
PVRI (dyne
16.9 ‘- 2.0
11 + 1 1.9 t 0.1t
76 k 5t 13.5 * 1.3
11 + 1 2.2 -c 0.1
82 ? 3*t 15.2 ‘- 1.9
2219 + 202
(L/min/m2) SVRI
79 2 3t
2 5t
17.4 + 0.7
11 r1 3.0 2 0.2
H
100 + 7t
92 + 4t
18.1 2 1.6
(mm Hg) Cl/PI
G
F
107 -e 7t
2.8 + 0.1
HR. heart rate; MAP, mean arterial pressure; MPAP, mean pulmonary artery pressure; CVP, central venous pressure; PCWP,
pulmonary capillary wedge pressure; Cl, cardiac index; PI, perfusion index during CPB; SVRI, systemic vascular resistance index; PVRI, pulmonary vascular resistance index. ‘Differentfrom
group 1 (P < 0.05)
tDifferent from control (A)
both groups, a 32- to 40-fold increase occurred 15 minutes after starting CPB, then remained elevated throughout the study period. Plasma AVP concentrations did not correlate with SVRI during and after CPB. At control (Fig 3), plasma atria1 natriuretic peptide concentrations were above the normal range (20 to 30 pg/mL) and were correlated with pulmonary capillary wedge pressure (PCWP) (r = 0.63, P < O.Ol), but not with right atria1 pressure (r = 0.05, P > 0.05). This correlation disappeared after induction. After a decrease at induction, only significant in group 1, plasma ANP levels were stable in both groups despite hemodilution during CPB. When the heart was filled and had resumed contractions at the end of CPB, plasma ANP increased by 102% (P < 0.01) in both groups, then decreased progressively to control levels. After a decrease at induction in group 1, TT4 decreased Table 4. Oxygen Transport Data 1 h After Group DO,
(mL/min/m’) VO, (mL/min/m7 svo, (%) C(a-v)O, (mL/lOO mL) 0, extraction (%I
Before
CPB
ClOSUre
CPB
(30 mid
of Chest
1
372 f 20
194 + 7t
432 + 38
2
475 f 62
223 f 13t
410 + 31
1
80 f 4
44 + 9t
129 k 14t
2
71 k-9
80 k 7*
122 r 15t
1
78 + 1
76 2 4
68 2 3t
2
80 2 2
61 + 3*t
68 f 2t
1
3.7 2 0.3
2.8 -t- 0.6
4.8 2 0.4t
2
2.7 -t 0.3
4.3 k 0.4*t
4.2 2 0.3t
1
21 2 1
22 2 4
31 It 3t
2
16 + 2
36 2 3”t
30 -c 2t
Abbreviations: DO,, oxygen delivery index; VO,, oxygen consumption index; SvO,, mixed venous oxygen saturation; C(a-v)D,, arterialvenous oxygen content difference. *Different from Group 1 (P < 0.05) tDifferent from control (A)
by 38% and 44% in groups 1 and 2, respectively, 15 minutes after starting CPB (Fig 3); while hematocrit decreased by 34% and 37%, respectively. In both groups, ‘IT4 remained stable at 30 minutes of CPB then increased progressively without reaching control levels at the third postoperative hour. Plasma prolactin concentrations increased significantly in the two groups after induction of anesthesia (Fig 3) and during normothermic CPB, then decreased progressively in both groups. Diuresis and urine composition were not significantly different between the two groups (Table 5). No correlation between plasma ANP concentrations and urine volume or electrolytes was observed. DISCUSSION
The present study was designed to compare hemodynamic and hormonal variations with hypothermic and normothermic CPB in elective cardiac operations performed by the same surgeon. Except for the temperature of CPB, all the patients’ management was similar according to the standard routine in the institution. Aortic cross-clamping and CPB durations were 26% and 32%, respectively, longer in the hypothermic CPB group, most likely caused by the time for rewarming before the completion of CPB and two combined procedures (CABG and AVR) in group 1. The numbers of countershocks, hypertensive, hypotensive, tachycardic, and shivering postoperative episodes were greater in group 1, but the difference was only significant for tachycardia. Despite rewarming during CPB (similar pulmonary artery temperatures in both groups at the end of CPB), rectal temperature was lower in group 1 than in group 2 at the end of CPB, and pulmonary artery temperature was lower in group 1 at the first and third postoperative hours. The postoperative rewarming may induce substantial vasodilatation and, as
136
LEHOT
I- -cp
