Volatile halogenated anesthetics greatly affect heart rate (HR), blood pressure (BP), and cardiac contractility, as well as cardiovascular homeostatic control systems. Isoflurane, a widely clinically used inhalant anesthetic, has been shown to produce a decrease in arterial BP and total peripheral resistance (TPR) and an increase in HR (1) despite a negative direct action on the sino-atrial (SA) node (2). Taking into account that baroreceptor reflexes remain functional during isoflurane anesthesia to a greater extent than other clinically useful anesthetics (3), it is reasonable to believe that tachycardia is the result of a baroreflex response to the isoflurane-induced hypotension. However, isoflurane anesthesia was reported (4) to produce dissimilar sympathetic nervous activity and HR responses. Furthermore, many other factors that can influence chronotropic effects of inhalational anesthetics must be taken into account, such as surgical procedures, blood-gas changes, blood pH changes, drug interactions, and different anesthetic concentrations. All these variables can greatly complicate the interpretation of data concerning the chronotropic effects of inhalant anesthetics.
Although much data has been obtained on the hemodynamic effects of isoflurane in human beings and larger animals, only a few studies have provided cardiovascular data in isoflurane-anesthetized and nonanesthetized smaller animals, and only two have provided data on the cardiovascular effects of isoflurane in New Zealand white rabbits. Blake and colleagues (5) showed that at 1.3 minimum alveolar concentration (MAC) isoflurane administered by a rabbit box reduced mean arterial pressure (MAP) by 19% in normal rabbits and increased HR from 240 to 294 beats/min. Bell (6) also reported that administration of isoflurane significantly reduced MAP at all levels of isoflurane in normotensive rabbits. However, he also showed that HR and renal sympathetic nervous activity were not affected by isoflurane anesthesia.
Because little information regarding isoflurane-anesthetized rabbits is available, we investigated the effects of isoflurane on the rabbit cardiovascular system at several end-tidal concentrations. Furthermore, because a discrepancy has been reported to exist between chronotropic and sympathetic nervous activity responses to isoflurane in both humans and animals (4,6), we evaluated whether the chronotropic effects of isoflurane could also be the result of a vagal withdrawal in addition to baroreflex activation.
MATERIALS AND METHODS
Twenty-three New Zealand white (NZW) male rabbits weighing 2.5-3.0 kg were individually caged in a room with artificially controlled temperature (21°-24°C) and a 12-h light/12-h dark cycle. Water and food were provided ad libitum.
Methods and procedures for basal cardiovascular variables and plasma catecholamine study
Rabbits were anesthetized with isoflurane (details on induction and maintenance of anesthesia are provided in the Experimental Protocol Section), and an implantable transmitter (model TL11M2-C50-PXT) (Data Sciences, MN, U.S.A.) was implanted to monitor MAP and the ECG. Using aseptic technique, we made a midline abdominal incision and exposed the abdominal aorta below the renal arteries by using gauze pads to retain the intestines to the sides of the aorta. The telemetric catheter (ID 0.7 mm) was placed in the abdominal aorta and secured in place with a cellulose patch and medical-grade tissue adhesive. Sensing electrodes were subcutaneously placed in line with the long axis of the heart. One electrode was placed in the upper right quadrant of the chest; the other was placed in the lower left abdomen. The telemetric transmitter body was sutured to the inside of the abdominal wall. Each rabbit was treated with analgesics such as buprenorphine (0.01 mg/kg/day intramuscularly), and antibiotics such as cefotaxime (20 mg/kg/day), and gentamicin (5 mg/kg/day) for 4 days, and was allowed to recover for a minimum of 7 days before experiments were performed. The telemetric signals were transformed back to a calibrated analog signal by a Data Sciences UA10 Universal Adapter D/A converter (Data Sciences). The analog signals were monitored directly on a Grass model 7D. Before data and blood samples were collected, habituation procedures were performed according to the method of Duan and associates (7). Each animal was taken from its home cage and placed in a restrainer for 30 min to habituate it to the experimental setting. This procedure was followed for 2 consecutive days. Animals were placed in the restraining cage; the fur on the outside of the right ear was clipped, and the skin was cleaned with Betadine. Thereafter, EMLA cream (Astra-Simes, Milan, Italy) was applied to the ear skin ≈1 h before the experimental protocol was performed to reduce pain after venous punctures (8). A 24-gauge needle catheter was inserted percutaneously into the marginal ear vein, and an intravenous flow control system (DIAL-A-FLO, Abbott, Pomezia, Italy) was attached to it. To obtain arterial blood samples, a polyethylene cannula (PE50) was inserted into the central ear artery after local anesthesia with lidocaine. Arterial blood samples were used to evaluate acid-base balance by an acid-base Analyzer (ABL 30, Radiometer A/S, Copenhagen, Denmark), and plasma catecholamine levels were determined by high-performance liquid chromatography (HPLC) (9).
Determination of baroreflex sensitivity
Rabbits received bolus injections of phenylephrine (10, 25, and 50 μg/kg) to increase the systemic arterial pressure to ≈20, 40, and 60 mm Hg. HR was obtained from ECG and plotted as a function of the mean aortic pressure. Baroreflex sensitivity was expressed as the slope of the linear regression line obtained by plotting change in R-R interval (ms) versus change in mean aortic pressure (mm Hg).
In all animals, arterial BP and ECG were continuously recorded, in the conscious unsedated state, by a data acquisition system [LAB PC (hardware) and Labwindows 2.2 (software), National Instruments Italy, Milan]. After 30 min of hemodynamic stability, arterial blood samples were withdrawn for measurements of pH, blood gases, and plasma norepinephrine (NE) and epinephrine (Epi) levels. In 10 animals, baroreflex sensitivity was tested by phenylephrine injections.
Rabbits received inhalational induction with isoflurane delivered by a face mask (3% isoflurane in a gaseous mixture of air and oxygen at 50%), and orotracheal intubation was attempted immediately after loss of consciousness. If the rabbit did not chew when this was attempted, the level of unconsciousness was judged to be sufficient. A cuffed endotracheal tube (ID 3.0-3.5 mm) was inserted in the trachea (10), and was connected to a respiratory pump (model 6025, Ugo Basile, Italy), which was set at a tidal volume of 10 ml/kg. The ventilatory rate was adjusted to 30-35 breaths/min (0.50-0.58 Hz) to maintain end-tidal CO2 between 33 and 38 mm Hg. End-tidal anesthetic and carbon dioxide levels were continuously monitored (gas monitor, 5250 RGM, Ohmeda, England). To facilitate the mechanical ventilation, pancuronium bromide (0.1 mg/kg intravenously, i.v.) was injected to induce muscle relaxation.
After induction of anesthesia, the following protocols were performed. In 10 rabbits, anesthesia was maintained with isoflurane in a mixture of air and oxygen (50:50 vol/vol) at 0.5, 1.0, and 1.5 MAC (11) in random fashion. Thirty minutes after the beginning of recordings, cardiovascular variables, plasma catecholamines, and baroreflex sensitivity were measured at the first MAC setting. The anesthetic depth was adjusted to the next random level, and another 30-min equilibration period was then allowed before additional measurements were made. In the same animals, the end-tidal concentration of isoflurane was returned to 1 MAC; after 30-min equilibration, clonidine, an α2-adrenergic agonist, was administered at increasing doses (10, 30, 100, and 300 μg/kg i.v.) cumulatively to verify whether the effects of isoflurane on cardiovascular parameters were centrally mediated. In 4 rabbits, anesthesia was maintained with isoflurane in a mixture of air and oxygen (50:50 vol/vol) at 1 MAC; atenolol, a selective β1-adrenoceptor antagonist, was administrated at increasing doses (0.5, 1, 2.5, and 5 mg/kg i.v.) cumulatively to evaluate the role of β1-adrenoceptors in the cardiac effects of isoflurane. To ensure the involvement of β1-adrenoceptors in the cardiac effects of isoflurane, another group of 4 rabbits was pretreated with atenolol at a dose of 5 mg/kg i.v., which completely blocked HR responses to isoflurane. Under these conditions, the anesthetic gas was administered at 0.5, 1.0, and 1.5 MAC in random fashion.
In a separate group of rabbits (n = 5), we performed bilateral cervical vagotomy during placement of the telemetric device to rule out the possibility that mechanisms other than vagal withdrawal were involved in cardiac effects of isoflurane. Anesthesia was induced as already described and was maintained with isoflurane in a mixture of air and oxygen (50:50 vol/vol) at 1 MAC.
Spectral analysis of HR variability
ECG was sampled at 1,000 Hz and processed on a PC 80486 DX2-S to obtain the beat-to-beat variability series of cardiac cycle (R-R, tachogram). R-R intervals were determined with an algorithm capable of detecting the maximum point of each R-wave; thus, R-R intervals were estimated with a resolution of 1 ms. One or more stable data segments of 120-s duration were selected after visual inspection of the beat-to-beat series of the R-R- intervals. From these segments, power spectra were calculated by an event series discrete Fourier transform (12,13).
The drugs used were isoflurane (Forane, Abbott); pancuronium bromide (Pavulon, Organon Teknika); phenylephrine hydrochloride, clonidine, and atenolol (Sigma); buprenorphine (Temgesic, Boehringer Biochemia Robin); cefotaxime (Claforan, Roussel Maestretti); gentamicin (Gentalyn, Essex); and prilocaine plus lidocaine (EMLA cream, Astra-Simes).
Data and statistical analysis
Results are mean ± SE. A Student's t test was used to compare differences between two experimental groups. One-way analysis of variance (ANOVA) and Newman-Keuls multiple-range test were used to determine differences in means of multiple groups (14). Linear regression analysis (least-squares method) was used to obtain the slope of the baroreflex curves generated in each experiment.
Cardiovascular variables in non-anesthetized and isoflurane-anesthetized rabbits are shown in Table 1. Isoflurane caused a significant dose-dependent increase in HR and a dose-dependent decrease in SBP, DBP, and MAP. Isoflurane-mediated cardiovascular effects were accompanied by a dose-related decrease in baroreflex sensitivity, indicated by a progressive reduction in the slope of baroreflex response curves (Fig. 1), but with no significant changes in plasma NE and Epi levels (Table 2).
Spectral decomposition of tachograms obtained from nonanesthetized rabbits allowed us to recognize a small but clear low-frequency band (LO-FR) centered at 0.086 ± 0.005 Hz and a high-frequency band (HI-FR) positioned at 0.68± 0.03 Hz, which shifted leftward (0.57 ± 0.01 Hz) in anesthetized animals. Isoflurane induced a marked decrease in R-R variability (Fig. 2) and in LO-FR and HI-FR spectral powers, ≈80 and 90%, respectively (Fig. 3), independent of isoflurane concentrations. In nonanesthetized rabbits, the LO-FR:HI-FR ratio was 0.3 ± 0.01, whereas this ratio significantly increased by ≈300% under isoflurane anesthesia. No difference was noted among LO-FR:HI-FR ratios by varying levels of anesthesia.
Clonidine and atenolol produced a dose-related decrease in HR during isoflurane anesthesia (Figs. 4 and 5), accompanied by a significant and dose-dependent decrease in mean aortic pressure (Figs. 4 and 5). Moreover, in rabbits pretreated with atenolol at a dose of 5 mg/kg i.v., we noted a slight but significant decrease in HR rate in nonanesthetized animals at rest, whereas the arterial BP was not significantly changed (Table 3). In this group, isoflurane induced no increase in HR, although it produced a significant and dose-dependent decrease in arterial BP (Table 3).
In vagotomized anesthetized rabbits, HR and arterial BP measurements were not different from those obtained in intact anesthetized animals at equiMAC concentrations of isoflurane (Table 4). In vagotomized nonanesthetized rabbits, HR was markedly increased and significantly different from that of nonanesthetized control animals. MAP and arterial SBP and DBP were not significantly influenced by vagotomy, (Table 4).
Our results show that the administration of isoflurane in rabbits produces dose-dependent changes in HR and mean aortic pressure. These results are in accordance with data reported in the literature obtained in both humans and experimental animals (15).
At 0.5 MAC, the HR significantly increased also when mean aortic pressure was not different from control values. Although this apparently suggests that the baroreflex function is not involved in the genesis of tachycardia, tachycardia might be the result of a baroreflex response to other stimuli such as changes in aortic flow. However, how important is the contribution of the baroreflex function in cardiac responses to isoflurane? Considering that HR significantly increases in a dose-dependent manner while baroreflex sensitivity progressively decreases to near zero at 1.5 MAC, one can reasonably assume that the role of baroreflex in heart response to isoflurane is marginal, if existent.
To exclude the idea that tachycardia was due to the release of catecholamines from the adrenal medulla, i.e., a direct stimulation by isoflurane, we also determined plasma catecholamines. The plasma catecholamine levels were unchanged by isoflurane, and no significant differences were observed between anesthetized and non-anesthetized animals. These results suggest that tachycardia is not due to the involvement of the adrenal medulla in the heart response to isoflurane.
Next, to determine whether isoflurane-mediated tachycardia is due to a decrease in parasympathetic tone we evaluated spectral components of HR variability, a tool that provides indications of sympathovagal balance. In nonanesthetized rabbits, spectral analysis of tachogram showed two main frequency bands: a small LO-FR band and a clearer HI-FR band reflecting a major influence of parasympathetic modulation of the cardiac cycle. The LO-FR component, which centered at ≈0.086 Hz, compared well with that reported by Moguilevsky and associates (16) in rabbits and has been considered a marker of sympathetic activity in humans as well as in other animal species (17-19). After rabbits were anesthetized, we observed a marked reduction in LO-FR spectral power, suggesting that isoflurane induces a sympathetic depression; this is in agreement with data obtained in other laboratories (4,20). The HI-FR component, synchronous with ventilation, is believed to be mediated by parasympathetic efferents in rabbits (16) and is interpreted as a measure of vagal tone (17-19). In the present study, we observed a marked, nondose-related decrease in HI-FR spectral power, and an increase of ≈300% in the LO-FR:HI-FR ratio. Thus, spectral power data suggest that isoflurane induces both vagal and sympathetic tone decreases and has a more suppressive effect on the parasympathetic tone than the sympathetic tone, one. Despite the overall reduction in autonomic tone, one can speculate, that the residual sympathetic tone becomes predominant with respect to the parasympathetic one and is sufficient to explain heart response to isoflurane.
To clarify the genesis of tachycardia further, we used clonidine and atenolol. Under these conditions, a dose-related decrease in HR during isoflurane anesthesia was observed, which was accompanied by a significant and dose-dependent decrease in mean aortic pressure. Furthermore, pretreatment with atenolol completely blocked HR responses to isoflurane while enhancing the hypotensive effects of isoflurane, which indicates that the increase in HR plays an important role in the maintenance of cardiac output and, thereby, arterial BP under isoflurane. Overall, these results suggest that tachycardia is mediated by a central sympathetic outflow and that sympathetic tone has an important role in the maintenance of arterial BP under isoflurane anesthesia.
Because our results strongly suggested that tachycardia was mainly due to a parasympathetic withdrawal, we also performed experiments in vagotomized animals to exclude the possibility that mechanisms other than vagal withdrawal were involved in the cardiac effects of isoflurane. We could not determine whether vagal tone was still preserved under anesthesia because we made no direct recording of the parasympathetic nervous traffic. However, if the vagal tone was preserved, an increase in HR under isoflurane would have been expected after vagotomy. Instead, vagotomy induced no increase in HR in isoflurane-anesthetized animals, indicating that the effects on HR are mediated by changes in vagal tone.
The present study has some limitations. First, we were unable to measure the efferent sympathetic nervous activity by neural traffic monitoring. However, the effects of isoflurane on the efferent sympathetic nervous activity have already been studied, and isoflurane has been reported to produce a marked decrease in the efferent sympathetic nervous activity (4,20).
Second, some doubts have been raised about the sensitivity and specificity of power spectral analysis as a marker of sympathetic and parasympathetic tone (21,22). Because spectral power analysis does not necessarily give indication of the tone, we considered spectral analysis of heart rate variability as only one of the tools for obtaining information on sympathovagal balance. For this reason, we also measured plasma catecholamines and baroreflex sensitivity and evaluated the effects of clonidine and atenolol on HR responses to isoflurane.
Our results indicate that (a) isoflurane induces a dose-dependent increase in HR and a dose-dependent decrease in MAP similar to those observed in humans and other animal species; (b) tachycardia is mainly due to a parasympathetic withdrawal, whereas baroreflex responses to isoflurane-induced pressor changes appear to play only a marginal role; and (c) tachycardia is maintained by a central sympathetic outflow, and sympathetic tone has an important role in the maintenance of the arterial blood pressure under isoflurane anesthesia.
Acknowledgment: We thank Dr. Ben J. Ten Voorde of Vrije Universiteit (Amsterdam) for supplying the spectral analysis program, Dr. Adriano Laurenzi of Abbott for supplying isoflurane, and Paul Barker for English revision. We especially thank Professor G. Mancia for criticisms and suggestions.
1. Marshall BE, Longnecker DE, General anesthetics. In: Hardman JG, Limbird LE, Molinoff PB, Ruddon RW, Gilman A, eds. The pharmacological basis of therapeutics
, 9th ed. New York: McGraw-Hill, 1996:307-30.
2. Bosnjak ZJ, Kampine JP. Effects of halothane, enflurane, and isoflurane on SA node. Anesthesiology
3. Takeshima R, Dohi S. Comparison of arterial baroreflex function in humans anesthetized with enflurane or isoflurane. Anesth Analg
4. Muzi M, Ebert TJ. A comparison of baroreflex sensitivity during isoflurane and desflurane anesthesia in humans. Anesthesiology
5. Blake DW, Way D, Trigg L, Langton D, McGrath BP. Cardiovascular effects of volatile anesthesis in rabbits: influence of chronic heart failure and enalaprilat treatment. Anesth Analg
6. Bell LB. Baroreflex modulation by isoflurane anesthesia in normotensive and chronically hypertensive rabbits. Adv Pharmacol
7. Duan YF, Winters RW, McCabe PM, Green EJ, Schneiderman N. Basal and reactive plasma catecholamine levels under stress and anesthesia in rabbits. Physiol Behav
8. Flecknell PA, Liles JH, Williamson HA. The use of lignocaineprilocaine local anesthetic cream for pain-free venepuncture in laboratory animals. Lab Animals
9. Lucarelli C, Betto P, Ricciarello G, et al. An improved approach for the determination of plasma-free catecholamines by HPLC-ED: application to normal and hypertensive patients. Chromatographia
10. Kumar RA, Boyer MI, Bowen CVA. A reliable method of anesthesia for surgery in small rabbits. Lab Animal Sci
11. Drummond JC. MAC for halothane, enflurane, and isoflurane in the New Zealand white rabbit: and a test for the validity of MAC determinations. Anesthesiology
12. Ten Voorde BJ, Faes ThJC, Rompelman O. Spectra of data sampled at frequency-modulated rates in application to cardiovascular signals: part I. Analytical derivation of the spectra. Med Biol Eng Comput
13. Ten Voorde BJ, Faes ThJC, Rompelman O. Spectra of data sampled at frequency-modulated rates in application to cardiovascular signals: part 2. Evaluation of Fourier transform algorithms. Med Biol Eng Comput
14. Tallarida RJ, Murray RB. Manual of pharmacologic calculations with computer program.
New York: Springer-Verlag, 1987:117.
15. Warltier DC, Pagel PS. Cardiovascular and respiratory actions of desflurane: is desflurane different from isoflurane? Anesth Analg
16. Moguilevsky V, Oliver J, McGrath BP. Sympathetic regulation in rabbits with heart failure-experience using power spectra analysis of heart rate variability
. Clin Exp Pharmacol Physiol
17. Akserold S, Gordon D, Madwed JB, Snidman DC, Shannon DC, Cohen RJ. Hemodynamic regulation: investigation by spectral analysis. Am J Physiol
18. Pagani M, Lombardi F, Guzzetti S, et al. Power spectral analysis of heart rate and arterial pressure as a marker of sympatho-vagal interaction in man and conscious dog. Circ Res
19. Malliani A, Pagani M, Lombardi F, Cerutti S. Cardiovascular neural regulation explored in the frequency domain. Circulation
20. Sellgren J, Ponten J, Wallin G. Percutaneous recording of muscle nerve sympathetic activity during propofol, nitrous oxide, and isoflurane anesthesia in humans. Anesthesiology
21. Mancia G, Grassi G, Parati G, Daffonchio A. Evaluating sympathetic activity in human hypertension. J Hypertens
22. Goldberger JJ, Ahmed AW, Parker MA, Kadish AH. Dissociation of heart rate variability
from parasympathetic tone. Am J Physiol