Cardiovascular Effects of Volatile Anesthesia in Rabbits: Influence of Chronic Heart Failure and Enalamilat Treatment Duncan W. Blake, MB, FFARACS, Diana Way, and Barry P. McGrath, MD, FRAU
BSC, Lisbeth
Trigg, David Langton,
MB, FRACP,
Monash University Department of Medicine, Monash Medical Centre, Prince Henry's Hospital, Melbourne, Australia
Circulatory responses to isoflurane and halothane anesthesia were studied in eight rabbits with biventricular cardiomyopathy induced by doxorubicin (Adriamycin, 14 mgkg IV over 7 wk) and in eight controls (saline injections). In preliminary operations pulsed-Doppler flow probes were placed on the ascending aorta, left renal artery, and lower abdominal aorta. Each group was studied after 4,6, and 7 wk of treatment. The development of congestive heart failure (CHF) was associated with decreases in mean arterial pressure and cardiac output (CO) of 14% and 16%, respectively, (P < 0.05) and an increase in heart rate. In controls, each anesthetic agent produced dose-related decreases in mean arterial pressure and increases in heart rate, but no significant changes in CO. Renal blood flow was reduced to a similar degree by 1.3 MAC halothane (24% decrease) and 1.3 MAC
A
preexisting history of congestive heart failure (CHF)significantly increases the risk of mor,bidity and mortality after noncardiac surgery ( 1 3 ) . Human CHF may occur with a low, normal, or high cardiac output, -although chronic low-output CHF is most common. Low-output chronic CHF is a complex clinical syndrome in which there is increased afterload, activation of neurohumoral vasoconstrictor systems, altered regional blood flow distribution, changes in the cardiorenal axis, and complex functional and structural abnormalities in the failing heart itself. Anesthesia and surgery add a number of potential cardiovascular stress factors for these patients, including direct and indirect myocardial depression and changes in neurohumoral systems and peripheral vascular tone (4,5). With the administration of Supported by a grant from the National Heart Foundation of Australia. Accepted for publication May 16, 1991. Address correspondence to Dr. Blake, Department of Anaesthesia, Royal Melbourne Hospital, Grattan Street, Parkville, Victoria, Australia 3050. 01991 by the International Anesthesia Research Society 0003-2999/91/$3.50
isoflurane (21% decrease); hindlimb blood flow was reduced only by halothane. As CHF developed there was an attenuation of the heart rate response to anesthesia. Halothane, but not isoflurane, significantly reduced CO in more advanced stages of CHF. The changes in renal blood flow and hindlimb blood flow with each anesthetic in the CHF group were similar to those observed in controls and did not vary with week of treatment. Administration of the angiotensin-converting enzyme inhibitor enalaprilat (0.2 mgkg IV) reversed the CO and renal blood flow effects of halothane except after 7 wk of treatment in the CHF group, when the combination of halothane and enalaprilat resulted in severe circulatory depression. (Anesth Analg 1991;73:44143)
volatile anesthetics, reduction in afterload may predominate over direct cardiac depression (6). Although there are major differences between agents (9, systematic study of their effects in CHF has been limited because of the lack of a suitable animal model. Our previous studies have shown that one model of CHF, doxorubicin-induced cardiomyopathy in the rabbit, exhibits many features in common with chronic low-output CHF in humans (8-10). In the present study the cardiovascular effects of two commonly used volatile anesthetic agents, halothane and isoflurane, were compared at different stages of CHF in doxorubicin-treated rabbits and in control animals. Changes in systemic hemodynamics, in regional blood flows to renal and hindlimb vascular beds, and in plasma renin activity were examined. The rabbits were studied at 4, 6, and 7 wk of treatment during the development of heart failure and at two different levels of anesthesia for each agent. Angiotensin-converting enzyme inhibition is an established treatment of CHF (11,12) and may also have a place in the intraoperative management Anesth Analg 19!91;73441-8
441
442
BLAKEETAL. ANESTHESIA AND CHRONIC HEART FAILURE
of CHF (13,14). We therefore also examined the interaction between acute enalaprilat infusion and halothane anesthesia at different stages of CHF.
Methods Experiments were performed in 16 New Zealand White rabbits weighing 2.3-3.5 kg and aged 14-18 wk. The study was approved by the Prince Henry's Hospital Research Advisory and Ethics Committee and conformed with the guidelines of the National Health and Medical Research Council of Australia. In eight rabbits, heart failure was induced by the intravenous administration of 1 mgkg of doxorubicin (Adriamycin, Farmitalia Carlo Erba) twice weekly for 7 wk. A control group of eight rabbits received only saline injections but were handled and housed in a similar manner. Preliminary surgery was performed after 2 wk of doxorubicin treatment. General anesthesia was induced with 10 mgkg of intravenous methohexitone and maintained with halothane after endotracheal intubation. An abdominal approach was used to implant pulsed-Doppler flow probes (1-mm2crystals) on the left renal artery and on the abdominal aorta just above its bifurcation (15).A left thoracotomy was also performed and a Doppler probe was applied to the ascending aorta. Wires from each crystal were buried subcutaneously. After surgery, doxorubicin treatment was not restarted for at least 1 wk and was interrupted if the rabbit was losing weight. The initial experiments were performed after completion of 4 wk of doxorubicin treatment and were therefore separated from the surgery by at least 3 wk. On each experimental day, catheters were placed in an ear artery and vein and the Doppler wires were exteriorized using 0.5% lignocaine local anesthesia. The rabbits were then placed in a sealed Perspex box supplied with 4 Wmin of a mixture of air and oxygen. Resting observations were made over the 30 min before administration of halothane or isoflurane. Mean arterial pressure (MAP) was measured using a Hewlett-Packardtransducer and was used to trigger a pulse-interval meter (Baker Medical Research Institute, Melbourne, Australia). All signals were continuously recorded on computer (Macintosh SE, Apple Computer Inc., Cupertino, Calif.) using an A/Dconverter (MacLab, Analog-Digital Instruments, Dunedin, New Zealand). The three Doppler crystals were connected to a pulsed-Doppler flowmeter (model 545C-3; Bioengineering, University of Iowa, Iowa City, Iowa). Flows were measured as kilohertz Doppler shift and calibrated with a frequency generator. Doppler signals from the renal artery (renal blood flow, RBF) and lower aorta (hindlimb blood flow,
ANESTH ANALG 1991;73441-8
HBF) were adjusted using the range control to give maximum output and the clearest signal. These Doppler crystals were mounted in Dacron cuffs and required readjustment of the range at each experiment, so that it was not possible to compare resting levels of RBF and HBF over the 3-wk experimental period. The ascending aorta Dopplers were mounted in polystyrene shells and were used to measure cardiac output (CO) after adjustment to detect the axial flow signal. This system was calibrated against thermodilution CO using an aortic thermistor in a se arate series of four rabbits. A linear correlation ( = 0.92) was obtained over the range 0.51.5 Wmin. The volatile anesthetic agents were delivered in oxygen from Fluotec Mark 2 vaporizers at a flow rate of 4 Umin. The anesthetic concentration in the exit gas from the rabbit box was measured continuously with a crystal detector (Servo Gas Monitor 120, Siemens-Elema AB, Sweden). Experiments were performed in three groups: after 4, 6, and 7 wk of doxorubicin treatment, or after similar periods in the control group. The volatile anesthetics were each administered for 30 min at 0.7 MAC (minimum alveolar concentration) for the rabbit and then at 1.3 MAC for 30 min. This corresponds to 1and 2 vol% halothane and 1.5 and 3 vol% isoflurane (16). In each group of experiments, there was a recovery period of at least 4 h between the administration of halothane and isoflurane, and the order of administration was randomized. The two concentrations of each anesthetic were administered for consecutive 30-min periods during which an approximate steady state was reached as indicated by the cardiovascular variables. Mean arterial pressure, CO, heart rate, RBF, and HBF were averaged for the final 10 min at each anesthetic concentration. Arterial blood samples (0.5 mL) were also taken for blood gas analysis and renin assays at this time. The samples for renin assay were collected on ice, immediately centrifuged, and the plasma was stored at -20°C until assayed by radioimmunoassay (17). The effect of enalaprilat on the response to halothane was studied on a separate day. After the initial rest period, enalaprilat was administered as a 0.2-mgkg intravenous bolus followed by an infusion of 0.003 mg-kg-'-min-l. This regimen reduces converting enzyme activity in the rabbit to less than 3% of control (18). Measurements were averaged for 10-min periods: 20 min after starting the enalaprilat, after 30 min of 0.7 MAC halothane, and after 30 min of 1.3 MAC halothane. Levels of the hemodynamic variables were compared by analysis of variance. The factors used were rabbits, anesthetics, enalaprilat treatment, week of study, and doxorubicin treatment. Specific contrasts were made by partitioning of the analysis of variance.
P
BLAKEETAL.
ANESTH ANALG 1991;73:441-9
ANESTHESIA
443
AND CHRONIC HEART FAILURE
doxorubicin-treated group, there was a reduction after 6 wk in M A P (14%,P < 0.05) and CO (l6%,P < 0.05) and a 13% increase in heart rate (P < 0.01) at 7 wk.
Anesthesia
F i e 1. Mean arterial pressure (MAP), heart rate, and cardiac output (CO) in conscious resting rabbits after 4, 6, and 7 wk of treatment. Left: control group (n = S), saline injectiOnS only. Right: heart failure group (n = S), doxombicin(1 mgkg IV twice weekly). Error bars indicate sw( for each experimental grou derived from analysis of variance. sm = (error mean squardtz)'. 'P < 0.05 for orthogonal comparisons of means.
The Bonferroni correction was made to multiple, nonorthogonal contrasts. In the fjgures, the hemodynamic variables are shown as between-rabbit means with error bars indicating 1 standard error of the mean (SEM).
Results Conscious Animals Resting hernodynamic measurements in the two groups are shown in Figure 1. The control group showed no sigruficant change in resting MAP, CO, heart rate, or regional blood flows with time. In the
The rabbits tolerated each anesthetic without apparent distress. At the lower concentration (0.7 MAC) of each agent, they continued to move spontaneouslyin the experiment box and would react to external stimuli. The higher concentration (1.3 MAC) induced general anesthesia and, although ventilation was not controlled, respiratory rate was unchanged. The degree of respiratory depression did not differ between the anesthetic agents or with the development of CHF. With 1.3 MAC anesthesia, the mean partial arterial pressure of carbon dioxide for both groups was only slightly raised from 29.6 2 1.4 to 35.1 2 2 mm Hg and, with an inspired oxygen concentration of 70%, the mean partial arterial pressure of oxygen was 322 lr 19 mm Hg. There was no significant change in the circulatory responses of the control group to either anesthetic over the 3-wk experimental period (Table l), so in Figures 2-5 the average result from the three experiments in normal rabbits is shown in the left panel (control). At 1.3 MAC, halothane and isoflurane reduced MAP in normal rabbits by 17% and 19%, respectiveIy (Figure 2, P < 0.01). This was associated with similar increases in heart rate with each anesthetic (halothane from 230 to 293 beatshin, P < 0.005; isoflurane from 240 to 294 beatdmin, P < 0.00s). Cardiac output was not changed significantly with either anesthetic (Figure 3), but at 1.3 MAC halothane reduced RBF by 24% (P < 0.05) and isof l u m e reduced REP by 21% (P < 0.05). At 1.3 MAC halothane reduced HBF by 21% (P < 0.05), but isofhuane did not significantly alter HBF. In the CHF p u p , resting MAP was lower after 6 and 7 wk of doxorubicin treatment. At the lower concentration (0.7 MAC), halothane did not significantly alter MAP at any stage of doxorubicin treatment, but 0.7 MAC isoflurane reduced MAP by 7% (5 2 2 mm Hg, P < 0.05) after 6 and 7 wk of drug treatment. At this stage of CHF, 1.3 MAC halothane and isoflurane had similar effects on MAP. The average decrease in MAP of 22% (16 2 4 mm Hg) was similar to that in controls, but the hypotension during anesthesia was more pronounced because of the lower resting MAP (P < 0.05). In the CHP group, the heart rate during anesthesia did not & a g e at weeks 4, 6, and 7 of doxorubicin treatment. However, resting heart rate increased with the development of cardiac failure and the responses to each anesthetic
444
BLAUEETAL. ANESTHESIA AND CHRONIC HEART FAILURE
ANESTH ANALC 1991;73441-8
Table 1. Comparison Between the Responses to Anesthesia in Eight Control Rabbits at 4, 6, and 7 Weeks of the Study Change due to anesthesia ~
Week 4
6
MAC
Agent
0.7
H
1.3
1 H
1.3
I H I H
0.7
1 H
0.7
7
HR (beatshin)
(kHz Ds)
2 -3 - 16 - 10 4 -2 - 15 - 16 -3 -11 -10 - 18 1.01
35 50 63 68 39 38 40 38 27 51 52 50 2.05
0.4 0.4 -0.1 0.1 0.1 -0.1 -0.4 -0.3 0.4 0.2 0.7 0.8 2.62
-0.2 -0.4 -1.3 -1.4 -0.8 -0.6 -1.7 -1.1 0.4 -0.8 -1.4 -1.5 2.75
0.4 -0.5 -1 -0.4 -1.1 -0.4 -2.4 0.1 0.4 -0.6 -0.1 -1.1 1.09
NS
NS
NS
NS
NS
I H I
1.3
co
MAP (mm Hg)
F (2,12 df) between weeks
RBF (&
Ds)
HBF (kHz Ds)
kHz Ds, kilohertz Doppler shift; H, halothane; I, isoflurane; MAP, mean arterial pressure; HR, heart rate; CO, cardiac output; RBF, renal blood flow; HBF, hindlimb blood flow; F, withm-animals comparison of response at 4, 6, and 7 wk by analysis of variance; NS, not significant ( P > 0.05).
80
L
o
60
40
40
CONTROL WEEK4
3001
jj
WEEK6
WEEK7
CONTROL WEEK4
3001
T
WEEK6
WEEK7
T
250
250
!I 200
200
CONTROL WEEK4
WEEK6
WEEK7
Figure 2. Mean arterial pressure (MAP) and heart rate (HR) in two groups of rabbits and at two anesthetic concentrations (0.7 and 1.3 MAC) for halothane and isoflurane anesthesia. The averaged responses of the control group (n = 8) for the three experiments are shown at the I+ of each panel. The results for the CHF group are shown as weeks 4, 6, and 7 of treatment. Error bars indicate standard error of mean from analysis of variance
were significantly attenuated after 6 wk of treatment (P < 0.05 for treatments x times interaction, Figure 2). Anesthetic-induced changes in CO, RBF, and HBF with weeks of doxorubicin treatment are shown in Figure 3 together with the mean changes in the control group. In the CHF group, CO was not signif-
icantly changed by isoflurane anesthesia at 0.7 or 1.3 MAC. However, there was a progressively greater decrease in CO due to 1.3 MAC halothane with the development of CHF, reaching 20% at the 7-wk study (P < 0.05). In the CHF group, 0.7 MAC halothane reduced RBF by an average of 8% and 0.7 MAC
BLAKEETAL.
ANESTH ANALC
445
ANESTHESIA AND CHRONIC HEART FAILURE
1991;73.441-8
Figure 3. Change from mean awake values at two concentrations of halothane and isoflurane for canliac output (CO), renal blood flow (RBF), and hindlimb blood flow (HBF)in two groups of rabbits:controls (n = 8, awrage responses) and CHF (n = 8, weeks 4, 6, and 7 of treatment). Blood flows measured as Doppler shift from chronically implanted pulsed-Doppler flowmeters. E m bars indicate standard error of mean change.
0
4
4
1
4 11
- 2 -
coHTRoLWEEK4 WEEK8 - 7
isoflurane reduced RBF by 15% (P < 0.05). At 1.3 MAC, RBF decreased by 25% with halothane and 30%with isoflurane (P c 0.01). These RBF responses did not differ significantly between anesthetics, between the control and CIO; groups, or between weeks within the groups. No sign&ant changes in HBF were found with i s o f l m e anesthesia. After 7 wk of treatment with dombicin, there was a decrease in HBF of 18% during halothane anesthesia (P < 0.05), similar to the 21% decrease observed in the control group. Plasma renin concentrations increased with anesthesia, but there was no significant difference between the responses to halothane and isoflurane. The average renin increase qf 16.1 ng/mL in the CHF group at 7 wk was significantly greater (P < 0.005) than the increase of 9.6 ng/mL in the controls.
Enalapilat lnfusion In conscious controls, enalaprilat caused a small decrease in W and inmase in heart rate (Figure4). During enalaprilat infusion, there was a p p s s i v e reduction in MAP with 0.7 and 1.3 MAC halothane (P < 0.05), and MAP was si@cantly lower than with halothane alone (P< 0.01). The increase in heart rate with enalaprilat and halothane (P C 0.005) was
.
9’ OUWROL-4
wEN(6 WEEK7
similar to that with halothane alone. During enalaprilat idusion, resting CO and RBF were increased in control rabbits by 17% (P C 0.05) and by 20% (P < 0.025), respectively (Figure 5). The decrease in RBF seen during halothane anesthesia was abolished by the enalaprilat infusion. In the CHF group, enalaprilat again caused a further reduction in MAP during halothane anesthesia ( F i i 4). After 7 wk of doxorubicin treatment, MAP decreased by 34% from 70 mm Hg (resting) to 46 mm Hg (standard error of difference, 5 nun Hg) with enalaprilat plus 1.3 MAC halothane. Cardiac output during halothane plus enalaprilat was maintained in the CHF group except after 7 wk of doxorubicin treatment when CO decreased with 1.3 MAC halothane (Figure 5). In CHF rabbits enalaprilat also prevented the decrease in RBF produced by halothane anesthesia at both 0.7 and 1.3MAC, except during the higher anesthetic dose after 7 wk of doxorubicin treatment. Enalaprilat infusion had little effect on the HBF changes during halothane anesthesia (Figure 5).
Discussion Activation of neurohumoral systems in CHF results in peripheral circulatory changes, redistribution of
446
BLAKEETAL. ANESTHESIA AND CHRONIC HEART FAILURE
ANESTH ANALG 1991;73:441-8
100
2
!
0'
B o
8o
8,
8
-1 -2
40
CONTROL
WEEK4
WEEK6
WEEK7 21
350 1
B 1
1
g
250
0
-2 4
fl
'1
150
CONTROLWEEK4
WEEK0
WEEK7
Figure 4. Mean arterial pressure (MAP) and heart rate (HR) in the conscious rabbits (Rest), during enalaprilat infusion (E), enalaprilat with 0.7 MAC halothane ( E + H . 7 ) , and enalaprilat with 1.3 MAC halothane (E + H1.3) . Averaged responses of the control group are shown on the kff, and the responses in the group with heart failure after 4. 6. and 7 wk of treatment are shown on the right. Error bars indicate standard error of mean change from analysg of variance.
B
-1
-3 CONTROL
CO, increased cardiac load, and further deterioration in function. Ventricular irritability is also increased and there is a high incidence of sudden death (19). It is not surprising, therefore, that the Goldman index alone will not identify many patients who are at risk (1,20). Patients with mild-to-moderate CHF frequently require anesthesia, particularly for associated coronary or peripheral vascular disease. Their increased morbidity may be related to alterations in the pattern of cardiovascular responses to individual anesthetic drugs, but this is difficult to determine in humans. The pathological, hemodynamic, and hormonal changes associated with doxorubicin-induced CHF in the rabbit have been described in previous studies (8-10) and reflect the changes associated with CHF in humans. Over a 7-wk period, the rabbits develop a severe cardiomyopathy with interstitial fibrosis, biventricular dilatation, and hypertrophy. Exercise capacity is reduced progressively as the cumulative dose of doxorubicin increases (21). There is a high mortality after 7 wk of treatment, and ascites and
WEEK 4
WEEK6
WEEK7
Figure 5. Changes from resting in cardiac output (CO), renal blood flow (RBF), and hindlimb blood flow (HBF) during enalaprilat infusion (E), enalaprilat with 0.7 MAC halothane (E+H.7), and enalaprilat with 1.3 MAC halothane (E+H1.3). The /eft panel shows the average responses in the control group, and responses in the group with heart failure after 4,6, and 7 wk of treatment are shown on the right. Error bars indicate standard error of mean change from analysis of variance.
pleural effusions are commonly found at postmortem. There is a progressive increase in plasma noradrenaline and in plasma renin activity between 4 and 8 wk despite an increased blood volume. Renal blood flow measurements using either radiolabeled microspheres or [ '251]orthoiodohippurate show an early reduction, before any reduction in CO, although renal histology and mesenteric blood flow are unchanged. Renal sympathetic nerve activity (RSNA) is increased after only 4 wk of treatment and baroreceptor-induced changes in RSNA are increased early but blunted late in this model of CHF (22). In the present study, responses to anesthesia were tested between 4 and 7 wk of doxorubicin treatment to
ANESTH ANALG 1991;73:441-8
correspond with early and established heart failure. In conscious rabbits, the MAP, CO, and heart rate changes associated with the development of CHF were similar to those previously described. With the pulsed-Doppler technique it was not possible to measure absolute levels of RBF and HBF or to compare flows at different stages of CHF. However, the technique did allow changes in regional blood flows to be determined during the acute experiments. The cardiovascular responses to anesthesia of the control rabbits in the present study were similar to responses in humans. There were dose-related decreases in MAP and RBF with both isoflurane and halothane. Only halothane reduced hindlimb (muscle) blood flow. With isoflurane HBF was maintained, consistent with its vasodilator effect in muscle. However, neither anesthetic significantly reduced CO in normal rabbits, which is in contrast to the effect of halothane on CO in humans. Cardiac output was maintained during halothane anesthesia in the rabbit by a marked increase in heart rate not found in humans. The rabbit has a small heart size relative to body weight, and changes in CO are primarily determined by changes in heart rate rather than by stroke volume. Both isoflurane and halothane reduce RBF in humans (23,24), but there are conflicting reports on the RBF effects of isoflurane from animal studies (25,26). The effects seen in this study in rabbits are similar to those observed in humans. The reduction in RBF owing to anesthesia was similar with both volatile anesthetics. Neurohumoral mechanism are likely to be responsible for this, including an increase in RSNA. Volatile anesthetics are potent depressors of arterial baroreceptor responses (7). Therefore, an increase in renal sympathetic tone during anesthesia might be due to removal of an inhibitory effect on the RSNA from baroreceptor-dependent pathways. As CHF developed with doxorubicin treatment, there was an attenuation of the heart rate response to halothane and isoflurane anesthesia. This is consistent with the progressive impairment of arterial baroreceptor function in this model (22). Both anesthetic agents were associated with absolute decreases in MAP and RBF in the animals with heart failure that were similar to those in controls. However, the lower resting values increase the sigruficance of the blood pressure changes. Although absolute resting RBF was not determined in this study, it is known to be reduced by 30% after 6 wk of doxorubicin treatment and decreases further thereafter (10). Therefore, the dose-related decreases observed in RBF with each anestheticin the CHF group are also very significant. Halothane, but not isoflurane, reduced CO in the more advanced stages of CHF in this study. Studies of isolated cardiac muscle suggest that direct cardiac
BLAKEETAL. ANESTHESIA AND CHRONIC HEART FAILURE
447
depression owing to halothane is not potentiated in CHF, but the net reduction in contractility may be more significant (27). Combined with the blunted heart rate response to anesthesia, this could explain the decrease in CO seen in the CHF group, but not in controls. Isoflurane has been reported to have a smaller negative inotropic effect than halothane in animal studies, and this is supported by clinical observations (5,28,29). Hindlimb blood flow was also maintained during isoflurane anesthesia, but not with halothane. The changes in HBF, reflecting predominantly muscle blood flow, were similar in CHF and control rabbits. In this model of CHF, we have previously shown that exercise induces an exaggerated sympathetic vasoconstriction in both the renal and hindlimb vascular beds (30,31). In the present study, no such exaggerated response was found with anesthesia. Plasma renin levels increased with each anesthetic in both the control and CHF rabbits. In the rabbit, renin release occurs even with mild stress such as the insertion of an intravenous catheter (32). This may have accounted for part of the renin release, but not for the difference between the control and CHF groups. Both increased sympathetic activity and reduced RBF would be likely factors contributing to the exaggerated r e d response in the CHF group. Enalaprilat has been shown to improve survival in severe CHF in humans and may also improve exercise capacity (11,33), although the mechanisms are unknown. In patients without cardiac failure, preoperative enalaprilat reduces MAP during anesthesia but does not influence the autonomic responses to postural change or endotracheal intubation (34).The effect of intraoperative intravenous enalaprilat treatment of a patient with CHF has recently been reported (14). After the acute development of leftventricular failure, enalaprilat greatly increased CO and reduced heart rate with only a small reduction in MAP. In the present study, acute administration of intravenous enalaprilatin rabbits with CHF produced an increase in CO and RBF with only minor changes in MAP and heart rate. This favorable response was obtained at each stage of CHF studied. Enalaprilat infusion, commend before halothane administration, prevented the decrease in RBF in both control and CHF rabbits. This occurred despite a greater decrease in MAP than with halothane alone. Only at the 7-wk stage of CHF was RBF reduced by the higher concentration of halothane and enalaprilat. Despite the activation of several vasoconstrictor systems and a reduction in cardiac reserve in this model of CHF, differences in the cardiovascular effects of halothane and isoflurane were not particularly exaggerated, although isoflurane caused less hemodynamic disturbance than halothane. The re-
448
BLAKEETAL. ANESTHESLA AND CHRONIC HEART FAILURE
sults suggest that higher concentrations of halothane can reduce CO to critical levels and that either anesthetic may cause a severe reduction in RBF, even when CO is maintained. Although the use of angiotensin-converting enzyme inhibitors in CHF is well established, further clinical studies are needed to determine their role in the perioperative period. Enalaprilat infusions during anesthesia in patients with CHF may reduce morbidity caused by regional ischemia. The present study found a relatively severe circulatory depression with the combination of halothane and enalaprilat in the presence of late CHF. This was not unexpected as the use of a potent myocardial depressant is likely to block the reflex increase in CO that normally follows the afterload reduction caused by enalaprilat. It may be necessary to avoid higher concentrations of volatile anesthetics during enalaprilat treatment in patients with heart failure.
References 1. Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845-50. 2. Cutler BS, Wheeler HB, Parascos JA, et al. Applicability and interpretation of ECG stress testing in patients with peripheral vascular disease. Am J Surg 1981;141:501-5. 3. Gerson MC, Hurst JM, Hertzburg VS, et al. Cardiac prognosis in noncardiac geriatric surgery. Ann Intem Med 1985;103: 832-7. 4. Robertson D, Mchelakis AM. Effect of anaesthesia and surgery on plasma renin activity in man. J Clin Endocrinol Metab 1972;34:8314. 5. Hickey RF, Eger EI. Circulatory effects of inhaled anaesthetics. In: Prys-Roberts C, ed. The arculation in anaesthesia. Oxford: Blackwell Scientific, 1980441-57. 6. Reiz S, Balfors E, Gustavsson B, et al. Effects of halothane on coronary haemodynamics and myocardial metabolism in patients with ischaemic heart disease and heart failure. Acta Anaesthesiol %and 1982;26:132-8. 7. Quail AW. Modem inhalational anaesthetic agents: a review of halothane, isoflurane and enflurane. Med J Aust 1989;150:95101. 8. Amolda L, McGrath B, Cocks M, Sumithran E, Johnston C. Adriamycin cardiomyopathy in the rabbit: an animal model of low-output cardiac failure with activation of vasoconstrictor mechanisms. Cardiovasc Res 1985;19:37%32. 9. Amolda L, McGrath B, Cocks M, Johnston C. Vasoconstrictor role for vasopressin in experimental heart failure in the rabbit. J CIin Invest 1986;78:674-8. 10. McGrath BP, Jover BF, Trigg L, Arnolda LF. Adriamycininduced cardiomyopathic heart failure in the rabbit. In: Kawai C, Abelmann WH, eds. Pathogenesis of myocarditis and cardiomyopathy. Tokyo: University of Tokyo Press, 1987 121-33. 11. Lipkin DP, Poole-Wilson PA. Treatment of chronic heart failure: a review of recent drug trials. Br Med J 1985;291:99-. 12. Dzau VJ, Coluca WS, Williams GH, Curfrnan G, Meggs L, Hollenberg NK. Sustained effectiveness of converting enzyme inhibition in patients with severe congestive heart failure. N Engl J Med 1980;302:1373-9.
ANESTH ANALG 1991;73:441-8
13. Russell RM, Jones RM. Postoperative hypotension associated with enalapril. Anaesthesia 1989;44:837-8. 14. Acampora GA, Melendez JA, Keefe DL, T u m b d AD, Bedford RF. Intraoperative administration of the intravenous angiotensin-converting enzyme inhibitor, enalaprilat, in a patient with congestive heart failure. Anesth Analg 1989;69:8-9. 15. Haykood JR, Shaffer RA, Fastenow t,Fink GD, Brody MJ. Regional blood flow measurement with pulsed Doppler flowmeter in the conscious rat. Am J Physiol 1981;241:H273-8. 16. Drummond JC. MAC for halothane, enflurane and isoflurane in the New Zealand white rabbit. Anesthesiology 1985;62: 336-8. 17. lohnston CI, Mendelsohn FAO, Doyle AE. Metabolism of angiotensin I1 in sodium depletion and hypertension in humans. Circ Res 197230:II203-13. 18. Elliot JM, Kapoor V, Cain M, West MJ, Chalmers JP. The mechanism of hypertension and bradycardia following lesions of the caudal ventrolateral medulla in the rabbit. Clin Exp Theory Practice 1985;A71059-82. 19. Packer M. Sudden unexpected death in patients with congestive heart failure: a second frontier. Circulation 1985;72:681-5. 20. Jeffrey C1, Kunsman J, Cullen DJ, Brewster DC. A prospective evaluation of cardiac risk index. Anesthesiology 1983;58:4624. 21. Langton D, Jover B, McGrath BP, Ludbrook J. Cardiovascular responses to graded treadmill exercise during the development of doxorubicin induced heart failure in rabbits. Cardiovasc Res 1990;2495948. 22. Sano N, Way D, McGrath BP. Changes in renal norepinephrine spillover rate and baroreflex responses in evolving heart failure. Am J Physiol 1990;258:F151&22. 23. M a z e Rl, Cousins MJ, Barr GA. Renal effects and metabolism of isoflurane in man. Anesthesiology 1974;40:53&42. 24. Bastron RD. Renal haemodynamics and the effects of anaesthesia. In: Prys-Roberts C, ed. The arculation in anaesthesia. Oxford: Blackwell Scientific, 1980227-39. 25. Theye RA, Michenfelder JD. Individual organ contributions to the decrease in whole body VO, with isoflurane. Anesthesiology 1975;42:3W. 26. Lundeen G, Manohar M, Parks C. Systemic distribution of blood flow in swine while awake and during 1.0 and 1.5 MAC isoflurane with or without 50% nitrous oxide. Anesth Analg 1983;62:499-512. 27. Shimosato S, Yasuda I, Kemmotsu 0, Shanks C, Gamble C. Effect of halothane on altered contractility of isolated heart muscle obtained from cats with experimentally produced ventricular hypertrophy or failure. Br J Anaesth 1973;45:2-9. 28. Housmans PR, Murat 1. Comparative effects of halothane, enflurane and isoflurane at equipotent anaesthetic concentrations on isolated ventricular myocardium of the ferret. Anesthesiology 1988;69:451-63. 29. Stevens WC, Cromwell TH, Halsey MJ, Eger El 11, Shakespeare TF, Bahlman SH. The cardiovascular effects of a new inhalational anaesthetic, Forane, in human volunteers at constant arterial carbon dioxide tension. Anesthesiology 1971;35:&16. 30. Langton D, Jover B, McGrath BP, Ludbrook 1. Cardiovascular responses to treadmill exerase during progressive heart failure in rabbits. Cardiovasc Res 1990;24:959-68. 31. Langton D, Way D, Trigg L, Blake D, McGrath BP. Vasoconstriction in the renal vascular bed during exercise: studies in control and heart failure rabbits. Clin Exp Pharmacol Physiol 1990;17:219-23. 32. McKenzie JK, Ryan JW, Lee MR. Effect of laparotomy on plasma renin activity in the rabbit. Nature 1967;215:542-3. 33. The consensus trial study group: effects of enalapril on mortality in severe congestive heart failure. N Eng J Med 1987;316: 1429-35. 34. Murphy ID, Vaughan RS, Rosen M. Intravenous enalaprilat and autonomic reflexes. Anaesthesia 1989;44:816-21.
BSC, Lisbeth
Trigg, David Langton,
MB, FRACP,
Monash University Department of Medicine, Monash Medical Centre, Prince Henry's Hospital, Melbourne, Australia
Circulatory responses to isoflurane and halothane anesthesia were studied in eight rabbits with biventricular cardiomyopathy induced by doxorubicin (Adriamycin, 14 mgkg IV over 7 wk) and in eight controls (saline injections). In preliminary operations pulsed-Doppler flow probes were placed on the ascending aorta, left renal artery, and lower abdominal aorta. Each group was studied after 4,6, and 7 wk of treatment. The development of congestive heart failure (CHF) was associated with decreases in mean arterial pressure and cardiac output (CO) of 14% and 16%, respectively, (P < 0.05) and an increase in heart rate. In controls, each anesthetic agent produced dose-related decreases in mean arterial pressure and increases in heart rate, but no significant changes in CO. Renal blood flow was reduced to a similar degree by 1.3 MAC halothane (24% decrease) and 1.3 MAC
A
preexisting history of congestive heart failure (CHF)significantly increases the risk of mor,bidity and mortality after noncardiac surgery ( 1 3 ) . Human CHF may occur with a low, normal, or high cardiac output, -although chronic low-output CHF is most common. Low-output chronic CHF is a complex clinical syndrome in which there is increased afterload, activation of neurohumoral vasoconstrictor systems, altered regional blood flow distribution, changes in the cardiorenal axis, and complex functional and structural abnormalities in the failing heart itself. Anesthesia and surgery add a number of potential cardiovascular stress factors for these patients, including direct and indirect myocardial depression and changes in neurohumoral systems and peripheral vascular tone (4,5). With the administration of Supported by a grant from the National Heart Foundation of Australia. Accepted for publication May 16, 1991. Address correspondence to Dr. Blake, Department of Anaesthesia, Royal Melbourne Hospital, Grattan Street, Parkville, Victoria, Australia 3050. 01991 by the International Anesthesia Research Society 0003-2999/91/$3.50
isoflurane (21% decrease); hindlimb blood flow was reduced only by halothane. As CHF developed there was an attenuation of the heart rate response to anesthesia. Halothane, but not isoflurane, significantly reduced CO in more advanced stages of CHF. The changes in renal blood flow and hindlimb blood flow with each anesthetic in the CHF group were similar to those observed in controls and did not vary with week of treatment. Administration of the angiotensin-converting enzyme inhibitor enalaprilat (0.2 mgkg IV) reversed the CO and renal blood flow effects of halothane except after 7 wk of treatment in the CHF group, when the combination of halothane and enalaprilat resulted in severe circulatory depression. (Anesth Analg 1991;73:44143)
volatile anesthetics, reduction in afterload may predominate over direct cardiac depression (6). Although there are major differences between agents (9, systematic study of their effects in CHF has been limited because of the lack of a suitable animal model. Our previous studies have shown that one model of CHF, doxorubicin-induced cardiomyopathy in the rabbit, exhibits many features in common with chronic low-output CHF in humans (8-10). In the present study the cardiovascular effects of two commonly used volatile anesthetic agents, halothane and isoflurane, were compared at different stages of CHF in doxorubicin-treated rabbits and in control animals. Changes in systemic hemodynamics, in regional blood flows to renal and hindlimb vascular beds, and in plasma renin activity were examined. The rabbits were studied at 4, 6, and 7 wk of treatment during the development of heart failure and at two different levels of anesthesia for each agent. Angiotensin-converting enzyme inhibition is an established treatment of CHF (11,12) and may also have a place in the intraoperative management Anesth Analg 19!91;73441-8
441
442
BLAKEETAL. ANESTHESIA AND CHRONIC HEART FAILURE
of CHF (13,14). We therefore also examined the interaction between acute enalaprilat infusion and halothane anesthesia at different stages of CHF.
Methods Experiments were performed in 16 New Zealand White rabbits weighing 2.3-3.5 kg and aged 14-18 wk. The study was approved by the Prince Henry's Hospital Research Advisory and Ethics Committee and conformed with the guidelines of the National Health and Medical Research Council of Australia. In eight rabbits, heart failure was induced by the intravenous administration of 1 mgkg of doxorubicin (Adriamycin, Farmitalia Carlo Erba) twice weekly for 7 wk. A control group of eight rabbits received only saline injections but were handled and housed in a similar manner. Preliminary surgery was performed after 2 wk of doxorubicin treatment. General anesthesia was induced with 10 mgkg of intravenous methohexitone and maintained with halothane after endotracheal intubation. An abdominal approach was used to implant pulsed-Doppler flow probes (1-mm2crystals) on the left renal artery and on the abdominal aorta just above its bifurcation (15).A left thoracotomy was also performed and a Doppler probe was applied to the ascending aorta. Wires from each crystal were buried subcutaneously. After surgery, doxorubicin treatment was not restarted for at least 1 wk and was interrupted if the rabbit was losing weight. The initial experiments were performed after completion of 4 wk of doxorubicin treatment and were therefore separated from the surgery by at least 3 wk. On each experimental day, catheters were placed in an ear artery and vein and the Doppler wires were exteriorized using 0.5% lignocaine local anesthesia. The rabbits were then placed in a sealed Perspex box supplied with 4 Wmin of a mixture of air and oxygen. Resting observations were made over the 30 min before administration of halothane or isoflurane. Mean arterial pressure (MAP) was measured using a Hewlett-Packardtransducer and was used to trigger a pulse-interval meter (Baker Medical Research Institute, Melbourne, Australia). All signals were continuously recorded on computer (Macintosh SE, Apple Computer Inc., Cupertino, Calif.) using an A/Dconverter (MacLab, Analog-Digital Instruments, Dunedin, New Zealand). The three Doppler crystals were connected to a pulsed-Doppler flowmeter (model 545C-3; Bioengineering, University of Iowa, Iowa City, Iowa). Flows were measured as kilohertz Doppler shift and calibrated with a frequency generator. Doppler signals from the renal artery (renal blood flow, RBF) and lower aorta (hindlimb blood flow,
ANESTH ANALG 1991;73441-8
HBF) were adjusted using the range control to give maximum output and the clearest signal. These Doppler crystals were mounted in Dacron cuffs and required readjustment of the range at each experiment, so that it was not possible to compare resting levels of RBF and HBF over the 3-wk experimental period. The ascending aorta Dopplers were mounted in polystyrene shells and were used to measure cardiac output (CO) after adjustment to detect the axial flow signal. This system was calibrated against thermodilution CO using an aortic thermistor in a se arate series of four rabbits. A linear correlation ( = 0.92) was obtained over the range 0.51.5 Wmin. The volatile anesthetic agents were delivered in oxygen from Fluotec Mark 2 vaporizers at a flow rate of 4 Umin. The anesthetic concentration in the exit gas from the rabbit box was measured continuously with a crystal detector (Servo Gas Monitor 120, Siemens-Elema AB, Sweden). Experiments were performed in three groups: after 4, 6, and 7 wk of doxorubicin treatment, or after similar periods in the control group. The volatile anesthetics were each administered for 30 min at 0.7 MAC (minimum alveolar concentration) for the rabbit and then at 1.3 MAC for 30 min. This corresponds to 1and 2 vol% halothane and 1.5 and 3 vol% isoflurane (16). In each group of experiments, there was a recovery period of at least 4 h between the administration of halothane and isoflurane, and the order of administration was randomized. The two concentrations of each anesthetic were administered for consecutive 30-min periods during which an approximate steady state was reached as indicated by the cardiovascular variables. Mean arterial pressure, CO, heart rate, RBF, and HBF were averaged for the final 10 min at each anesthetic concentration. Arterial blood samples (0.5 mL) were also taken for blood gas analysis and renin assays at this time. The samples for renin assay were collected on ice, immediately centrifuged, and the plasma was stored at -20°C until assayed by radioimmunoassay (17). The effect of enalaprilat on the response to halothane was studied on a separate day. After the initial rest period, enalaprilat was administered as a 0.2-mgkg intravenous bolus followed by an infusion of 0.003 mg-kg-'-min-l. This regimen reduces converting enzyme activity in the rabbit to less than 3% of control (18). Measurements were averaged for 10-min periods: 20 min after starting the enalaprilat, after 30 min of 0.7 MAC halothane, and after 30 min of 1.3 MAC halothane. Levels of the hemodynamic variables were compared by analysis of variance. The factors used were rabbits, anesthetics, enalaprilat treatment, week of study, and doxorubicin treatment. Specific contrasts were made by partitioning of the analysis of variance.
P
BLAKEETAL.
ANESTH ANALG 1991;73:441-9
ANESTHESIA
443
AND CHRONIC HEART FAILURE
doxorubicin-treated group, there was a reduction after 6 wk in M A P (14%,P < 0.05) and CO (l6%,P < 0.05) and a 13% increase in heart rate (P < 0.01) at 7 wk.
Anesthesia
F i e 1. Mean arterial pressure (MAP), heart rate, and cardiac output (CO) in conscious resting rabbits after 4, 6, and 7 wk of treatment. Left: control group (n = S), saline injectiOnS only. Right: heart failure group (n = S), doxombicin(1 mgkg IV twice weekly). Error bars indicate sw( for each experimental grou derived from analysis of variance. sm = (error mean squardtz)'. 'P < 0.05 for orthogonal comparisons of means.
The Bonferroni correction was made to multiple, nonorthogonal contrasts. In the fjgures, the hemodynamic variables are shown as between-rabbit means with error bars indicating 1 standard error of the mean (SEM).
Results Conscious Animals Resting hernodynamic measurements in the two groups are shown in Figure 1. The control group showed no sigruficant change in resting MAP, CO, heart rate, or regional blood flows with time. In the
The rabbits tolerated each anesthetic without apparent distress. At the lower concentration (0.7 MAC) of each agent, they continued to move spontaneouslyin the experiment box and would react to external stimuli. The higher concentration (1.3 MAC) induced general anesthesia and, although ventilation was not controlled, respiratory rate was unchanged. The degree of respiratory depression did not differ between the anesthetic agents or with the development of CHF. With 1.3 MAC anesthesia, the mean partial arterial pressure of carbon dioxide for both groups was only slightly raised from 29.6 2 1.4 to 35.1 2 2 mm Hg and, with an inspired oxygen concentration of 70%, the mean partial arterial pressure of oxygen was 322 lr 19 mm Hg. There was no significant change in the circulatory responses of the control group to either anesthetic over the 3-wk experimental period (Table l), so in Figures 2-5 the average result from the three experiments in normal rabbits is shown in the left panel (control). At 1.3 MAC, halothane and isoflurane reduced MAP in normal rabbits by 17% and 19%, respectiveIy (Figure 2, P < 0.01). This was associated with similar increases in heart rate with each anesthetic (halothane from 230 to 293 beatshin, P < 0.005; isoflurane from 240 to 294 beatdmin, P < 0.00s). Cardiac output was not changed significantly with either anesthetic (Figure 3), but at 1.3 MAC halothane reduced RBF by 24% (P < 0.05) and isof l u m e reduced REP by 21% (P < 0.05). At 1.3 MAC halothane reduced HBF by 21% (P < 0.05), but isofhuane did not significantly alter HBF. In the CHF p u p , resting MAP was lower after 6 and 7 wk of doxorubicin treatment. At the lower concentration (0.7 MAC), halothane did not significantly alter MAP at any stage of doxorubicin treatment, but 0.7 MAC isoflurane reduced MAP by 7% (5 2 2 mm Hg, P < 0.05) after 6 and 7 wk of drug treatment. At this stage of CHF, 1.3 MAC halothane and isoflurane had similar effects on MAP. The average decrease in MAP of 22% (16 2 4 mm Hg) was similar to that in controls, but the hypotension during anesthesia was more pronounced because of the lower resting MAP (P < 0.05). In the CHP group, the heart rate during anesthesia did not & a g e at weeks 4, 6, and 7 of doxorubicin treatment. However, resting heart rate increased with the development of cardiac failure and the responses to each anesthetic
444
BLAUEETAL. ANESTHESIA AND CHRONIC HEART FAILURE
ANESTH ANALC 1991;73441-8
Table 1. Comparison Between the Responses to Anesthesia in Eight Control Rabbits at 4, 6, and 7 Weeks of the Study Change due to anesthesia ~
Week 4
6
MAC
Agent
0.7
H
1.3
1 H
1.3
I H I H
0.7
1 H
0.7
7
HR (beatshin)
(kHz Ds)
2 -3 - 16 - 10 4 -2 - 15 - 16 -3 -11 -10 - 18 1.01
35 50 63 68 39 38 40 38 27 51 52 50 2.05
0.4 0.4 -0.1 0.1 0.1 -0.1 -0.4 -0.3 0.4 0.2 0.7 0.8 2.62
-0.2 -0.4 -1.3 -1.4 -0.8 -0.6 -1.7 -1.1 0.4 -0.8 -1.4 -1.5 2.75
0.4 -0.5 -1 -0.4 -1.1 -0.4 -2.4 0.1 0.4 -0.6 -0.1 -1.1 1.09
NS
NS
NS
NS
NS
I H I
1.3
co
MAP (mm Hg)
F (2,12 df) between weeks
RBF (&
Ds)
HBF (kHz Ds)
kHz Ds, kilohertz Doppler shift; H, halothane; I, isoflurane; MAP, mean arterial pressure; HR, heart rate; CO, cardiac output; RBF, renal blood flow; HBF, hindlimb blood flow; F, withm-animals comparison of response at 4, 6, and 7 wk by analysis of variance; NS, not significant ( P > 0.05).
80
L
o
60
40
40
CONTROL WEEK4
3001
jj
WEEK6
WEEK7
CONTROL WEEK4
3001
T
WEEK6
WEEK7
T
250
250
!I 200
200
CONTROL WEEK4
WEEK6
WEEK7
Figure 2. Mean arterial pressure (MAP) and heart rate (HR) in two groups of rabbits and at two anesthetic concentrations (0.7 and 1.3 MAC) for halothane and isoflurane anesthesia. The averaged responses of the control group (n = 8) for the three experiments are shown at the I+ of each panel. The results for the CHF group are shown as weeks 4, 6, and 7 of treatment. Error bars indicate standard error of mean from analysis of variance
were significantly attenuated after 6 wk of treatment (P < 0.05 for treatments x times interaction, Figure 2). Anesthetic-induced changes in CO, RBF, and HBF with weeks of doxorubicin treatment are shown in Figure 3 together with the mean changes in the control group. In the CHF group, CO was not signif-
icantly changed by isoflurane anesthesia at 0.7 or 1.3 MAC. However, there was a progressively greater decrease in CO due to 1.3 MAC halothane with the development of CHF, reaching 20% at the 7-wk study (P < 0.05). In the CHF group, 0.7 MAC halothane reduced RBF by an average of 8% and 0.7 MAC
BLAKEETAL.
ANESTH ANALC
445
ANESTHESIA AND CHRONIC HEART FAILURE
1991;73.441-8
Figure 3. Change from mean awake values at two concentrations of halothane and isoflurane for canliac output (CO), renal blood flow (RBF), and hindlimb blood flow (HBF)in two groups of rabbits:controls (n = 8, awrage responses) and CHF (n = 8, weeks 4, 6, and 7 of treatment). Blood flows measured as Doppler shift from chronically implanted pulsed-Doppler flowmeters. E m bars indicate standard error of mean change.
0
4
4
1
4 11
- 2 -
coHTRoLWEEK4 WEEK8 - 7
isoflurane reduced RBF by 15% (P < 0.05). At 1.3 MAC, RBF decreased by 25% with halothane and 30%with isoflurane (P c 0.01). These RBF responses did not differ significantly between anesthetics, between the control and CIO; groups, or between weeks within the groups. No sign&ant changes in HBF were found with i s o f l m e anesthesia. After 7 wk of treatment with dombicin, there was a decrease in HBF of 18% during halothane anesthesia (P < 0.05), similar to the 21% decrease observed in the control group. Plasma renin concentrations increased with anesthesia, but there was no significant difference between the responses to halothane and isoflurane. The average renin increase qf 16.1 ng/mL in the CHF group at 7 wk was significantly greater (P < 0.005) than the increase of 9.6 ng/mL in the controls.
Enalapilat lnfusion In conscious controls, enalaprilat caused a small decrease in W and inmase in heart rate (Figure4). During enalaprilat infusion, there was a p p s s i v e reduction in MAP with 0.7 and 1.3 MAC halothane (P < 0.05), and MAP was si@cantly lower than with halothane alone (P< 0.01). The increase in heart rate with enalaprilat and halothane (P C 0.005) was
.
9’ OUWROL-4
wEN(6 WEEK7
similar to that with halothane alone. During enalaprilat idusion, resting CO and RBF were increased in control rabbits by 17% (P C 0.05) and by 20% (P < 0.025), respectively (Figure 5). The decrease in RBF seen during halothane anesthesia was abolished by the enalaprilat infusion. In the CHF group, enalaprilat again caused a further reduction in MAP during halothane anesthesia ( F i i 4). After 7 wk of doxorubicin treatment, MAP decreased by 34% from 70 mm Hg (resting) to 46 mm Hg (standard error of difference, 5 nun Hg) with enalaprilat plus 1.3 MAC halothane. Cardiac output during halothane plus enalaprilat was maintained in the CHF group except after 7 wk of doxorubicin treatment when CO decreased with 1.3 MAC halothane (Figure 5). In CHF rabbits enalaprilat also prevented the decrease in RBF produced by halothane anesthesia at both 0.7 and 1.3MAC, except during the higher anesthetic dose after 7 wk of doxorubicin treatment. Enalaprilat infusion had little effect on the HBF changes during halothane anesthesia (Figure 5).
Discussion Activation of neurohumoral systems in CHF results in peripheral circulatory changes, redistribution of
446
BLAKEETAL. ANESTHESIA AND CHRONIC HEART FAILURE
ANESTH ANALG 1991;73:441-8
100
2
!
0'
B o
8o
8,
8
-1 -2
40
CONTROL
WEEK4
WEEK6
WEEK7 21
350 1
B 1
1
g
250
0
-2 4
fl
'1
150
CONTROLWEEK4
WEEK0
WEEK7
Figure 4. Mean arterial pressure (MAP) and heart rate (HR) in the conscious rabbits (Rest), during enalaprilat infusion (E), enalaprilat with 0.7 MAC halothane ( E + H . 7 ) , and enalaprilat with 1.3 MAC halothane (E + H1.3) . Averaged responses of the control group are shown on the kff, and the responses in the group with heart failure after 4. 6. and 7 wk of treatment are shown on the right. Error bars indicate standard error of mean change from analysg of variance.
B
-1
-3 CONTROL
CO, increased cardiac load, and further deterioration in function. Ventricular irritability is also increased and there is a high incidence of sudden death (19). It is not surprising, therefore, that the Goldman index alone will not identify many patients who are at risk (1,20). Patients with mild-to-moderate CHF frequently require anesthesia, particularly for associated coronary or peripheral vascular disease. Their increased morbidity may be related to alterations in the pattern of cardiovascular responses to individual anesthetic drugs, but this is difficult to determine in humans. The pathological, hemodynamic, and hormonal changes associated with doxorubicin-induced CHF in the rabbit have been described in previous studies (8-10) and reflect the changes associated with CHF in humans. Over a 7-wk period, the rabbits develop a severe cardiomyopathy with interstitial fibrosis, biventricular dilatation, and hypertrophy. Exercise capacity is reduced progressively as the cumulative dose of doxorubicin increases (21). There is a high mortality after 7 wk of treatment, and ascites and
WEEK 4
WEEK6
WEEK7
Figure 5. Changes from resting in cardiac output (CO), renal blood flow (RBF), and hindlimb blood flow (HBF) during enalaprilat infusion (E), enalaprilat with 0.7 MAC halothane (E+H.7), and enalaprilat with 1.3 MAC halothane (E+H1.3). The /eft panel shows the average responses in the control group, and responses in the group with heart failure after 4,6, and 7 wk of treatment are shown on the right. Error bars indicate standard error of mean change from analysis of variance.
pleural effusions are commonly found at postmortem. There is a progressive increase in plasma noradrenaline and in plasma renin activity between 4 and 8 wk despite an increased blood volume. Renal blood flow measurements using either radiolabeled microspheres or [ '251]orthoiodohippurate show an early reduction, before any reduction in CO, although renal histology and mesenteric blood flow are unchanged. Renal sympathetic nerve activity (RSNA) is increased after only 4 wk of treatment and baroreceptor-induced changes in RSNA are increased early but blunted late in this model of CHF (22). In the present study, responses to anesthesia were tested between 4 and 7 wk of doxorubicin treatment to
ANESTH ANALG 1991;73:441-8
correspond with early and established heart failure. In conscious rabbits, the MAP, CO, and heart rate changes associated with the development of CHF were similar to those previously described. With the pulsed-Doppler technique it was not possible to measure absolute levels of RBF and HBF or to compare flows at different stages of CHF. However, the technique did allow changes in regional blood flows to be determined during the acute experiments. The cardiovascular responses to anesthesia of the control rabbits in the present study were similar to responses in humans. There were dose-related decreases in MAP and RBF with both isoflurane and halothane. Only halothane reduced hindlimb (muscle) blood flow. With isoflurane HBF was maintained, consistent with its vasodilator effect in muscle. However, neither anesthetic significantly reduced CO in normal rabbits, which is in contrast to the effect of halothane on CO in humans. Cardiac output was maintained during halothane anesthesia in the rabbit by a marked increase in heart rate not found in humans. The rabbit has a small heart size relative to body weight, and changes in CO are primarily determined by changes in heart rate rather than by stroke volume. Both isoflurane and halothane reduce RBF in humans (23,24), but there are conflicting reports on the RBF effects of isoflurane from animal studies (25,26). The effects seen in this study in rabbits are similar to those observed in humans. The reduction in RBF owing to anesthesia was similar with both volatile anesthetics. Neurohumoral mechanism are likely to be responsible for this, including an increase in RSNA. Volatile anesthetics are potent depressors of arterial baroreceptor responses (7). Therefore, an increase in renal sympathetic tone during anesthesia might be due to removal of an inhibitory effect on the RSNA from baroreceptor-dependent pathways. As CHF developed with doxorubicin treatment, there was an attenuation of the heart rate response to halothane and isoflurane anesthesia. This is consistent with the progressive impairment of arterial baroreceptor function in this model (22). Both anesthetic agents were associated with absolute decreases in MAP and RBF in the animals with heart failure that were similar to those in controls. However, the lower resting values increase the sigruficance of the blood pressure changes. Although absolute resting RBF was not determined in this study, it is known to be reduced by 30% after 6 wk of doxorubicin treatment and decreases further thereafter (10). Therefore, the dose-related decreases observed in RBF with each anestheticin the CHF group are also very significant. Halothane, but not isoflurane, reduced CO in the more advanced stages of CHF in this study. Studies of isolated cardiac muscle suggest that direct cardiac
BLAKEETAL. ANESTHESIA AND CHRONIC HEART FAILURE
447
depression owing to halothane is not potentiated in CHF, but the net reduction in contractility may be more significant (27). Combined with the blunted heart rate response to anesthesia, this could explain the decrease in CO seen in the CHF group, but not in controls. Isoflurane has been reported to have a smaller negative inotropic effect than halothane in animal studies, and this is supported by clinical observations (5,28,29). Hindlimb blood flow was also maintained during isoflurane anesthesia, but not with halothane. The changes in HBF, reflecting predominantly muscle blood flow, were similar in CHF and control rabbits. In this model of CHF, we have previously shown that exercise induces an exaggerated sympathetic vasoconstriction in both the renal and hindlimb vascular beds (30,31). In the present study, no such exaggerated response was found with anesthesia. Plasma renin levels increased with each anesthetic in both the control and CHF rabbits. In the rabbit, renin release occurs even with mild stress such as the insertion of an intravenous catheter (32). This may have accounted for part of the renin release, but not for the difference between the control and CHF groups. Both increased sympathetic activity and reduced RBF would be likely factors contributing to the exaggerated r e d response in the CHF group. Enalaprilat has been shown to improve survival in severe CHF in humans and may also improve exercise capacity (11,33), although the mechanisms are unknown. In patients without cardiac failure, preoperative enalaprilat reduces MAP during anesthesia but does not influence the autonomic responses to postural change or endotracheal intubation (34).The effect of intraoperative intravenous enalaprilat treatment of a patient with CHF has recently been reported (14). After the acute development of leftventricular failure, enalaprilat greatly increased CO and reduced heart rate with only a small reduction in MAP. In the present study, acute administration of intravenous enalaprilatin rabbits with CHF produced an increase in CO and RBF with only minor changes in MAP and heart rate. This favorable response was obtained at each stage of CHF studied. Enalaprilat infusion, commend before halothane administration, prevented the decrease in RBF in both control and CHF rabbits. This occurred despite a greater decrease in MAP than with halothane alone. Only at the 7-wk stage of CHF was RBF reduced by the higher concentration of halothane and enalaprilat. Despite the activation of several vasoconstrictor systems and a reduction in cardiac reserve in this model of CHF, differences in the cardiovascular effects of halothane and isoflurane were not particularly exaggerated, although isoflurane caused less hemodynamic disturbance than halothane. The re-
448
BLAKEETAL. ANESTHESLA AND CHRONIC HEART FAILURE
sults suggest that higher concentrations of halothane can reduce CO to critical levels and that either anesthetic may cause a severe reduction in RBF, even when CO is maintained. Although the use of angiotensin-converting enzyme inhibitors in CHF is well established, further clinical studies are needed to determine their role in the perioperative period. Enalaprilat infusions during anesthesia in patients with CHF may reduce morbidity caused by regional ischemia. The present study found a relatively severe circulatory depression with the combination of halothane and enalaprilat in the presence of late CHF. This was not unexpected as the use of a potent myocardial depressant is likely to block the reflex increase in CO that normally follows the afterload reduction caused by enalaprilat. It may be necessary to avoid higher concentrations of volatile anesthetics during enalaprilat treatment in patients with heart failure.
References 1. Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845-50. 2. Cutler BS, Wheeler HB, Parascos JA, et al. Applicability and interpretation of ECG stress testing in patients with peripheral vascular disease. Am J Surg 1981;141:501-5. 3. Gerson MC, Hurst JM, Hertzburg VS, et al. Cardiac prognosis in noncardiac geriatric surgery. Ann Intem Med 1985;103: 832-7. 4. Robertson D, Mchelakis AM. Effect of anaesthesia and surgery on plasma renin activity in man. J Clin Endocrinol Metab 1972;34:8314. 5. Hickey RF, Eger EI. Circulatory effects of inhaled anaesthetics. In: Prys-Roberts C, ed. The arculation in anaesthesia. Oxford: Blackwell Scientific, 1980441-57. 6. Reiz S, Balfors E, Gustavsson B, et al. Effects of halothane on coronary haemodynamics and myocardial metabolism in patients with ischaemic heart disease and heart failure. Acta Anaesthesiol %and 1982;26:132-8. 7. Quail AW. Modem inhalational anaesthetic agents: a review of halothane, isoflurane and enflurane. Med J Aust 1989;150:95101. 8. Amolda L, McGrath B, Cocks M, Sumithran E, Johnston C. Adriamycin cardiomyopathy in the rabbit: an animal model of low-output cardiac failure with activation of vasoconstrictor mechanisms. Cardiovasc Res 1985;19:37%32. 9. Amolda L, McGrath B, Cocks M, Johnston C. Vasoconstrictor role for vasopressin in experimental heart failure in the rabbit. J CIin Invest 1986;78:674-8. 10. McGrath BP, Jover BF, Trigg L, Arnolda LF. Adriamycininduced cardiomyopathic heart failure in the rabbit. In: Kawai C, Abelmann WH, eds. Pathogenesis of myocarditis and cardiomyopathy. Tokyo: University of Tokyo Press, 1987 121-33. 11. Lipkin DP, Poole-Wilson PA. Treatment of chronic heart failure: a review of recent drug trials. Br Med J 1985;291:99-. 12. Dzau VJ, Coluca WS, Williams GH, Curfrnan G, Meggs L, Hollenberg NK. Sustained effectiveness of converting enzyme inhibition in patients with severe congestive heart failure. N Engl J Med 1980;302:1373-9.
ANESTH ANALG 1991;73:441-8
13. Russell RM, Jones RM. Postoperative hypotension associated with enalapril. Anaesthesia 1989;44:837-8. 14. Acampora GA, Melendez JA, Keefe DL, T u m b d AD, Bedford RF. Intraoperative administration of the intravenous angiotensin-converting enzyme inhibitor, enalaprilat, in a patient with congestive heart failure. Anesth Analg 1989;69:8-9. 15. Haykood JR, Shaffer RA, Fastenow t,Fink GD, Brody MJ. Regional blood flow measurement with pulsed Doppler flowmeter in the conscious rat. Am J Physiol 1981;241:H273-8. 16. Drummond JC. MAC for halothane, enflurane and isoflurane in the New Zealand white rabbit. Anesthesiology 1985;62: 336-8. 17. lohnston CI, Mendelsohn FAO, Doyle AE. Metabolism of angiotensin I1 in sodium depletion and hypertension in humans. Circ Res 197230:II203-13. 18. Elliot JM, Kapoor V, Cain M, West MJ, Chalmers JP. The mechanism of hypertension and bradycardia following lesions of the caudal ventrolateral medulla in the rabbit. Clin Exp Theory Practice 1985;A71059-82. 19. Packer M. Sudden unexpected death in patients with congestive heart failure: a second frontier. Circulation 1985;72:681-5. 20. Jeffrey C1, Kunsman J, Cullen DJ, Brewster DC. A prospective evaluation of cardiac risk index. Anesthesiology 1983;58:4624. 21. Langton D, Jover B, McGrath BP, Ludbrook J. Cardiovascular responses to graded treadmill exercise during the development of doxorubicin induced heart failure in rabbits. Cardiovasc Res 1990;2495948. 22. Sano N, Way D, McGrath BP. Changes in renal norepinephrine spillover rate and baroreflex responses in evolving heart failure. Am J Physiol 1990;258:F151&22. 23. M a z e Rl, Cousins MJ, Barr GA. Renal effects and metabolism of isoflurane in man. Anesthesiology 1974;40:53&42. 24. Bastron RD. Renal haemodynamics and the effects of anaesthesia. In: Prys-Roberts C, ed. The arculation in anaesthesia. Oxford: Blackwell Scientific, 1980227-39. 25. Theye RA, Michenfelder JD. Individual organ contributions to the decrease in whole body VO, with isoflurane. Anesthesiology 1975;42:3W. 26. Lundeen G, Manohar M, Parks C. Systemic distribution of blood flow in swine while awake and during 1.0 and 1.5 MAC isoflurane with or without 50% nitrous oxide. Anesth Analg 1983;62:499-512. 27. Shimosato S, Yasuda I, Kemmotsu 0, Shanks C, Gamble C. Effect of halothane on altered contractility of isolated heart muscle obtained from cats with experimentally produced ventricular hypertrophy or failure. Br J Anaesth 1973;45:2-9. 28. Housmans PR, Murat 1. Comparative effects of halothane, enflurane and isoflurane at equipotent anaesthetic concentrations on isolated ventricular myocardium of the ferret. Anesthesiology 1988;69:451-63. 29. Stevens WC, Cromwell TH, Halsey MJ, Eger El 11, Shakespeare TF, Bahlman SH. The cardiovascular effects of a new inhalational anaesthetic, Forane, in human volunteers at constant arterial carbon dioxide tension. Anesthesiology 1971;35:&16. 30. Langton D, Jover B, McGrath BP, Ludbrook 1. Cardiovascular responses to treadmill exerase during progressive heart failure in rabbits. Cardiovasc Res 1990;24:959-68. 31. Langton D, Way D, Trigg L, Blake D, McGrath BP. Vasoconstriction in the renal vascular bed during exercise: studies in control and heart failure rabbits. Clin Exp Pharmacol Physiol 1990;17:219-23. 32. McKenzie JK, Ryan JW, Lee MR. Effect of laparotomy on plasma renin activity in the rabbit. Nature 1967;215:542-3. 33. The consensus trial study group: effects of enalapril on mortality in severe congestive heart failure. N Eng J Med 1987;316: 1429-35. 34. Murphy ID, Vaughan RS, Rosen M. Intravenous enalaprilat and autonomic reflexes. Anaesthesia 1989;44:816-21.
