Effects of Isoflurane Anesthesia on the Cardiovascular Function of the C57BL/6 Mouse

Abstract

Isoflurane (ISO) is the most commonly used inhalational anesthetic for experimental interventions in mice and is preferred for imaging technologies that require the mouse to remain anesthetized for relatively long time periods. This study compares the stability of mean arterial pressure (MAP), heart rate (HR), and body temperature under ISO concentrations of 1%, 1.5%, and 2% (volume-to-volume, v/v) for up to 90 minutes postinduction. At all three levels of anesthesia, we examined evoked physiological responses to fractional inspiratory ratio variations of oxygen (FiO₂) and nitrous oxide (N₂O). In addition, we determined the hemodynamic effects of anesthesia on pH, glucose, insulin, glucocorticoids, and partial pressure of oxygen and of carbon dioxide in the blood (p_aO₂, p_aCO₂). The results indicate that the most appropriate ISO dose level was 1.5% v/v, yielding stable MAP and HR values comparable to those observed in the animal’s conscious state, with a minute-to-minute variability in MAP and HR of ≤11%. Based on such recordings, the optimal FiO₂ appeared to be 50%. The additional use of N₂O was associated with higher and more stable values of MAP and HR. Arterial pH values were within the physiological range and varied between 7.20 and 7.43. ISO anesthesia at 1.5% v/v was also associated with mild hyperglycemia (+47%), whereas insulin levels and corticosteroids remained unaltered. We conclude that the application of isoflurane as an inhalational anesthetic in the mouse can be optimized to attain stable hemodynamics by administering it at 1.5% v/v and by supplementing it with N₂O.

Introduction

The mouse has become the preferred animal species for experimental cardiovascular research elucidating the effects of transgenic modifications (Gehrmann et al. 2000; Hoit 2001; Janssen and Smits 2002; MacGowan et al. 2001; Milano et al. 1994). Although phenotypic screening is increasingly conducted by telemetric techniques, many conventional invasive and noninvasive technologies that measure hemodynamic function are conducted under inhalational anesthesia (Erhardt et al. 1984; Hart et al. 2001; Kober et al. 2005; Ohnishi et al. 1974; Price and Ohnishi 1980). Yet anesthetics (injectable or inhalational) are known to depress cardiovascular function moderately to severely (Bosnjak et al. 1992; Connelly and Coronado 1994; Hart et al. 2001; Matsuda et al. 2007; Ohnishi et al. 1974; Pagel et al. 1991; Price and Ohnishi 1980; Szczesny et al. 2004). In addition, they have adverse physiological effects on circulating hormones as well as at the cellular level (Cornett et al. 2008; Lambert et al. 2005; Sato et al. 2006), affecting both calcium entry through L-type calcium ion (Ca²⁺) channels (Ohnishi et al. 1974; Orestes et al. 2009; Price and Ohnishi 1980) and the calcium sensitivity of the contractile proteins (Price and Ohnishi 1980). There are also prominent effects on the central and peripheral nervous system (Doursout and Chelly 1988), but most notably on metabolism, perfusion, and the uncoupling of oxidative phosphorylation by inhibition of adenosine triphosphate (ATP) synthesis (Chance 1963; Kohro et al. 2001; Rottenberg 1983).

For surgical intervention and short-term experimentation, isoflurane (ISO¹) is the most frequently used anesthetic in studies with mice, in part because of its moderate cardiodepressive effects in comparison to those of the injectable agents pentobarbital and urethane or the ketamine/xylazine mixture (Janssen and Smits 2002; Janssen et al. 2004).

Advances in imaging modalities—magnetic resonance imaging (MRI) (Ruff et al. 1998), computed tomography (CT) (Badea et al. 2005), ultrasound (Chu et al. 2006; Roth et al. 2001), and micropositron emission tomography (microPET) (Toyama et al. 2004; Woo et al. 2008)—have also enabled the noninvasive study and phenotyping of the mouse cardiovascular system, with emphasis on the C57BL/6 strain, which is widely used for imaging studies. In such studies, it is imperative that the mouse be kept under prolonged anesthesia (up to 90 minutes) at the lowest possible level of ISO to minimize cardiodepressive effects.

Thermoregulation is vital for maintenance of homeostasis in mice, because of their small body size and large surface area. The effect of temperature on pH, blood flow, and peripheral or total arterial resistance and mean arterial pressure (MAP¹) may also become significant (Barbee et al. 1992; Janssen and Smits 2002; Sarin et al. 1990). Previous studies have investigated the effects of thermoregulation and altered temperature patterns on the basal metabolic rate (rate of heat production) and intrinsic tissue sensitivity to β-adrenergic agonists linked to catechol-O-methyltransferase activity and its modulated response at lower body temperatures in the mouse (Ramirez et al. 1975). Other possible major heat loss mechanisms are attributed to metabolic reactions (anabolic or catabolic) (Dobson and Headrick 1995; Else and Hulbert 1981; Hulbert and Else 1989).

Despite the absence of evidence for direct effects between body temperature and glucose use, higher energetic and metabolic rates in small animals are attributed to their higher energetic O₂ needs. Studies have shown that general inhalational anesthetics (isoflurane, halothane) decrease glucose metabolic rates as a result of inhibition of ATP synthesis in the mitochondria in both mice and rats (Kohro et al. 2001; Rottenberg 1983). Researchers have also reported that these effects impair glucose tolerance; raise baseline glucose, insulin, and glucagon levels; and lead to an enhanced plasma insulin response to glucose (attributed to increased hepatic glycogenolysis due to activated glycogen phosphorylase) (Bailey and Flatt 1980). Also prominent has been a rapid hyperglycemic effect (due to sympathetic liver innervation and immediate hormonal release) (Durand et al. 2009), including adrenomedullary catecholamine release. Similar dose-dependent anesthesia effects were noted for cerebral glucose metabolism in rats (Du et al. 2009; Kofke et al. 1987; Lenz et al. 1998; Ori et al. 1986; Stulken et al. 1977), with glucose metabolic rate and oxygen extraction decreases as low as 41–45% in cortical regions, corresponding increases in mean cerebral blood flow and plasma glucose, and reductions in heart rate. PET studies in mice have documented such findings, supporting dose-dependent, temporal uptake increases of ISO in heart and brain regions, indicative of coronary and cerebral blood flow increases and altered metabolism (Toyama et al. 2004).

Studies of the optimal settings of ISO administration for experimental use and for application in imaging studies are limited (Kober et al. 2005). Furthermore, little information is available on the effects of ISO on cardiovascular homeostasis in mice when the inhaled fractional inspiratory ratio of oxygen (FiO₂¹) is altered or when anesthetic nitrous oxide (N₂O¹) is added as a balancing agent. Yet the use in humans of N₂O as a balancing agent (“second anesthetic”) is widespread (Brunson 2008) and may have both cardiac and vascular effects. Price (1976) showed that N₂O enhances sympathetic nervous activity, counteracting ISO-induced cardiodepressive effects. Other authors (e.g., Sanders et al. 2008) have reported that it elicits a vasoconstrictive effect, increasing peripheral vascular resistance and blood pressure. And, reporting a study in C57BL mice, Samarska and colleagues (2009) suggest that N₂O stabilizes blood pressure by preventing downregulation of vascular cyclooxygenase expression.

We present a detailed study of methods to ensure the physiological welfare of mice under ISO inhalational anesthesia and elucidate the temporal effects of FiO₂ and N₂O (up to 90 minutes postinduction) on cardiac function and integrative physiological response. Specifically, we assessed dose-dependent depressive effects of ISO anesthesia on cardiovascular function and physiological response, possible effects of FiO₂, and effects of the evoked response on global cardiovascular function after administration of N₂O. We also monitored corticosteroid levels, metabolic (insulin and glucose) hormones, and respiration and blood oxygenation parameters in order to assess metabolism and stress-induced responses.

Materials and Methods

Ethical Approval

All animal protocols were approved by the Veterinary Services of the Ministry of Agriculture of the Government of Cyprus in accordance with national rules set by the Ministry, European Animal Research directives,² and international guidelines for animal research (NRC 1996). All experiments conformed to the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes.³¹

Animal Handling, Transportation, and Acclimation

Inbred mice were obtained from the Animal Housing Facility of the Cyprus Institute of Neurology and Genetics and housed in polypropylene cages (2–3 mice per cage) with food and water provided ad libitum. The mice were maintained on a 12:12 hour dark-light cycle with lights switched on at 6:00am. All procedures took place during the light phase.

Physiological Studies

The study involved 165 male C57BL/6 mice (weight, 22–32 g; age, 8–14 weeks). Anesthesia was initially induced by placing the mice for 2–3 minutes in a chamber filled with 3–5% volume-to-volume (v/v¹) ISO (Piramal Healthcare Ltd., London, UK). The mice were then positioned on a specialized circuit board (Indus Instruments, Webster, Texas) that enabled controlled heating and ECG recording. Anesthesia was maintained by inhalation of different concentrations of ISO as described in detail in the Experimental Design section below. Mice breathed freely through a nose cone; gas mixtures were delivered via a constant volume ventilator (SAR-830, CWE Inc., Ardmore, Pennsylvania), with ventilation rate and airflow pressure settings that matched normal physiological respiratory conditions. Mouse respiration rates (counted based on visual inspection) led to measured tidal volumes of 0.12–0.15 ml and flows of 33.6–44 nl/min, in accordance with published reports of artificially ventilated (Gilson 2005) and freely breathing C57BL/6J mice (Schwartz 2000). The ventilator was connected to a calibrated continuous flow ISO vaporizer (Harvard Apparatus, Keighley, UK) and to O₂, medical air, and N₂O flow meters (PTFE, Cole-Parmer, Vernon Hills, Illinois). The depth of anesthesia was monitored from heart rate (HR¹), respiration rates, and limb reflexes induced by paw and tail pinches applied periodically.

Heart rate was independently monitored using bipolar electrocardiograph (ECG) (Ambu Blue Sensor BRS, Ambu, Denmark) taped to the animal’s foot pads. Breathing flow rate was monitored using a respiration flow meter (TSD137A, Biopac, Goleta, California) attached to a commercially available Biopac recording system (Acknowledge, v. 3.8.2, Biopac) that allowed adjustment of inflow rates (≤12 ml/s) and gas mixture delivery pressures.

Temperature was monitored via a rectal probe (Physitemp, Clifton, New Jersey; and Biopac) and maintained at body temperature (approximately 37°C), using a feedback-controlled 175 W infrared lamp (Stoelting, Wood Dale, Illinois) via a temperature controller module (Digi-Sense, Cole-Parmer) (Figure 1). At the end of all procedures, mice were euthanized using the cervical dislocation technique.

Figure 1.

Diagram of experimental setup for laboratory mouse research. ADI, ADInstruments; ECG, electrocardiogram; exp, expiratory; insp, inspiratory; temp., temperature.

Microsurgical Procedures

For monitoring of mean arterial pressure, mice underwent an experimental microsurgical protocol under the guidance of a 100x dissecting magnification microscope (SZ61, Olympus, Japan) equipped with a digital camera (E-330, Olympus, Japan), with the help of a light source (KL1500LCD, Olympus, Mainz, Germany). A left carotid cutdown was performed with a 27 gauge needle to facilitate the insertion of an arterial catheter (PE10 polyethylene tubing, Intramedic, Clay Adams, Becton, Dickinson, Sparks, Maryland) attached to a pressure-transducer and connected to the Biopac recording system. Heparinized saline (154 mEq NaCl, 100 units/100 ml heparin) was administered to prevent clotting of the arterial line and to maintain blood volume.

Experimental Design

We performed six sets of studies as follows.

Study I: Effects of ISO Concentration Variation on MAP, HR, and Body Temperature

Twenty-one male C57BL/6 mice (weight, 25.6±7.3 g; age, 7.3±0.3 weeks) underwent the same experimental induction and surgical protocol for implantation of the carotid catheter. We then divided the mice into three groups of seven each and applied three different ISO concentrations—1%, 1.5%, and 2% v/v—for 40 to 90 minutes. Steady-state values of MAP and HR were measured and their variability estimated (as described in detail in the Statistical Analysis section below). In all these studies FiO₂ was set to 100%.

Study II: Physiological Response to Fractional Inspiratory Ratio (FiO₂)

After determining in study I that the optimal anesthetic level of ISO was 1.5% v/v, we conducted similar hemodynamic studies in 18 additional male C57BL/6 mice (weight, 26.7±1.4 g; age, 7.5±0.4; n = 25 including the 7 from study I) to determine the effects of O₂ level variation (based on the variation of the FiO₂ in the inhaled gas mixture). The three FiO₂ levels that were applied were 50%, 75%, and 100%, in addition to separate recordings in pure medical air (FiO₂, 20%). FiO₂ levels were monitored throughout the studies with an oxygen analyzer and an externally mounted oxygen sensor (Hudson RCI, Temecula, California).

Study III: Titration of Nitrous Oxide in the Inspired Gas Mixture

We investigated the effects of anesthetic N₂O as a balancing agent to the intermediate ISO anesthesia level (1.5% v/v) at two separate levels of O₂ concentration—50% (weight, 28.7±1 g; age, 10.5±0.2 weeks) and 75% (weight, 29.5±0.7 g; age, 10.9±0.1 weeks)—in 10 male C57BL/6 mice, 5 in each group. The O₂/N₂O concentrations in the inhaled mixture were adjusted by altering the N₂O flow rate with respect to the maximum O₂ flow rate, using the oxygen analyzer-sensor assembly.

Study IV: Stressor and Metabolic Hormone Assays

We used biochemical assays to compare associated effects of ISO at 1% and 2% v/v on steroid (cortisol, androstenedione) levels, metabolic (insulin) hormones, and glucose. Matched cohorts of male C57BL/6 mice (weight, 26.3±2.8 g; age, 10.2±1.3 weeks, n = 37 total) underwent anesthesia with 1% or 2% v/v ISO (mixed with 100% O₂ postinduction) and the removal of aortic blood aliquots (1 ml), after which they were euthanized (with 3–5% v/v ISO) at t = 0, 20, 40, 60, and 80 minutes.

Study V: Temporal Variation of Blood Glucose

We studied the temporal effects of varying levels of ISO anesthesia (1%, 1.5%, 2% v/v) on blood glucose levels for up to 90 minutes postinduction. Male C57BL/6 mice (weight, 29±2 g; age, 9.8±0.6 weeks) were allowed to breathe freely ISO at 1% (n = 5), 1.5% (n = 7), or 2% (n = 8) mixed with 100% oxygen. The animals’ body temperature was maintained at 37.1±0.4°C and their ECG, HR, and respiratory flow rate values were constantly monitored. A 0.6 μl blood aliquot was extracted at 5-minute intervals (by a tail vein needle or by cutting the tail vein) for glucose measurements using a blood glucose meter (Ascensia Contour Meter, Bayer, Germany). These tail blood glucose levels (and their variation) were compared to those from 1 ml of carotid arterial blood sampled from two separate cohorts of male C57BL/6 mice (of similar age and weight ranges) anesthetized with 1% and 2% v/v ISO.

Study VI: Anesthesia-Dependent Respiration Effects

To assess possible respiratory distress as a result of anesthesia and optimize the ventilator and respiratory parameters, we used a separate mouse cohort (weight, 23–35 g; age, 9–12 weeks; n = 56), under the same induction protocol as in the other samples, to extract periodic 60 μl carotid artery blood samples in heparinized capillary tubes (Multicap, Siemens Healthcare Diagnostics Ltd, Surrey, UK) for analysis. Blood extracts were collected at t = 0 min (n = 7), t = 20–23 min (ISO = 2% v/v, n = 10; ISO = 1% v/v, n = 10), t = 60–64 min (ISO = 2% v/v, n = 9; ISO = 1% v/v, n = 5), t = 80–90 min (ISO = 2% v/v, n = 9; ISO = 1% v/v, n = 6) postinduction and underwent blood-gas (BG) analysis (pH, p_aCO₂, p_aO₂, Na⁺, K⁺, and hematocrit, or Hct¹; Rapidlab 348, Siemens Healthcare Diagnostics, Deerfield, Illinois). In most of these experiments, mice were euthanized immediately after BG measurements.

Physiological Data Acquisition and Processing

Physiological data (MAP, ventilation, temperature, HR) were recorded in real time at 500–1000 samples/s using the Biopac system (Acknowledge v. 3.8.2). Concurrent recording of ECG activity was achieved with a separate, commercially available eight-channel recording module (Powerlab 8/30, ADInstruments, Oxford, UK) at 500–1000 samples/s, controlled via LabChart acquisition software (Lab Chart, v. 7, ADI). All files were saved and exported in text files. They were subsequently read by MATLAB (Natick, Massachusetts) for processing. Custom-written MATLAB routines with a front-end graphical user interface allowed temporal synchronization of recorded data, 1-minute interval analyses, and other signal analyses including computation of descriptive statistical and autocorrelation indices. ECG electrical parameters were quantified separately using the ECG analysis modules of LabChart with build-in algorithms that allowed R-wave detections, ECG signal averaging, and cluster classification.

Statistical Analysis

All results are reported as mean ± standard deviation (SD). Constancy of HR and MAP values is reported as the coefficient of variability¹ (CV_HR = 100. SD_HR/Mean_HR, CV_MAP = 100. SD_MAP/Mean_MAP) of the 1-minute averaged samples. Two-way unbalanced analysis of variance (ANOVA) and repeated-measures, nonparametric two-tailed student t-tests, were also used (XLSTAT, Addinsoft, New York) to determine whether the various ISO doses were associated with significantly different temporal effects on the cardiovascular, metabolic, hormonal, and respiratory parameters (HR, MAP, temperature, glucose, insulin, androstenedione, cortisol, pH, p_aO₂, p_aCO₂, and electrolyte concentrations) at the 1%, 1.5%, and 2% v/v ISO levels.

Results

Study I

Figure 2 presents variations in HR, MAP, and body temperature at 1%, 1.5%, and 2% v/v ISO anesthesia during the 40- to 90-minute postinduction period. Body temperature was maintained at 36.4±2.1°C (for mice under 1% v/v ISO), 36.3±2.6°C (1.5%), and 36.6±1.5°C (2%). Mean heart rates—408±23 (1% v/v), 478±11.4 (1.5%), and 482±10 (2%)—were significantly lower at 1% v/v ISO than at 1.5% and 2% (p < 0.0001). Corresponding CV_HR values were 14±2 (1%), 11±1 (1.5%), and 14±3 (2%) and MAP values were 85±3 (1% v/v), 92±2 (1.5%), and 72±9 millimeters of mercury (mmHg,¹ a unit measuring force per unit area) (2%), differing significantly at 1.5% v/v from those at 1% and 2% (p < 0.0001). The corresponding CV_MAP values were computed to be 13±3 (1% v/v), 8±2 (1.5%), and 33±2 (2%).

Figure 2.

Variation of heart rate (beats per minute, BPM), mean arterial pressure (MAP), and temperature at 1%, 1.5%, and 2% volume-to-volume (v/v) isoflurane (ISO) anesthesia 40–90 minutes postinduction at inspired O₂ = 100%. Values are means ± standard deviation (SD), n = 7 male C57BL/6 mice per group. mmHg, millimeters of mercury

Study II

Fractional inspiratory ratio (FiO₂) titration results at the intermediate anesthetic level of 1.5% v/v are depicted in Figure 3. Mean HR values in the 40- to 90-minute period were 478±11 (n = 7), 486±21 (n = 7), 476±22 (n = 6), and 477±27 (n = 5) for FiO₂ at 100%, 75%, 50%, and 20%, respectively. Such values were not significantly different at the 5% significance. Corresponding CV_HR values were 11±1, 10±2, 9±1, and 8±2, respectively; MAP values were 92±2 (n = 7), 83±4 (n = 7), 90±1 (n = 6), and 84±2 (n = 5); and CV_MAP values were 8±2, 25±6, 6±2, and 10±4.

Figure 3.

Mean arterial pressure (MAP) and heart rate (beats per minute, BPM) fractional oxygen dependence at various levels of oxygen (O₂)—100%, 75%, 50%, 20%—with the optimal isoflurane (ISO) anesthetic dose of 1.5% volume-to-volume (v/v), measured 40–90 minutes postinduction. Values are means ± standard deviation (SD); n = 5–7 male C57BL/6 mice per group. mmHg, millimeters of mercury

We observed that MAP values were significantly different for all FiO₂ challenges and, in particular, for the 45–60 minute interval for FiO₂ at both 75% and 20%. Figure 4 shows the effects of adding N₂O (in the ISO/O₂ gas mixture) on MAP and HR based on two experiments using 1.5% ISO v/v mixed with either N₂O = 50% and O₂ = 50% or N₂O = 25% and O₂ = 75%. Mean HR and MAP values (t = 40–90 min) were 489.5±14.8 (n = 5), 515.8±8.7 (n = 5) bpm (CV_HR = 10±2; 7±2), and 94±3 mmHg for both groups of 5 male C57BL/6 mice (CV_MAP = 6±3; 10±3) for the two mixtures (p = not significant). Mean HR and MAP values were significantly different from corresponding results in the FiO₂ = 50–100% cases (p < 0.001). Mean body temperature was maintained at 37.6–37.8°C.

Figure 4.

Effect on heart rate (HR) and mean arterial pressure (MAP) of administered nitrous oxide (N₂O) at various inspiration gas mixtures: N₂O = 50%, oxygen (O₂) = 50% (top) and N₂O = 25%, O₂ = 75% (bottom) at the optimal isoflurane (ISO) anesthesia level of 1.5% volume-to-volume (v/v). Values are means ± standard deviation (SD); n = 5 male C57BL/6 mice per group. BPM, beats per minute; mmHg, millimeters of mercury

Study III

Figure 5 shows variations in arterial concentrations of glucose, insulin, cortisol, and androstenedione at 0, 20, 40, 60, and 80 minutes postinduction for 1% and 2% v/v ISO levels. No significant differences from baseline (t = 0) were observed among any of these parameters at either 1% or 2% v/v of ISO, suggesting the stability of physiological and metabolic status with the tested anesthesia protocol.

Figure 5.

Temporal variation of corticosteroid (androstenedione, cortisol), insulin, and glucose at 1% and 2% volume-to-volume (v/v) isoflurane (ISO) anesthesia up to 80 minutes postinduction in male C57BL/6 mice. Values are means ± standard deviation (SD); n = 4 per group (but for t = 0 [white bar] n = 5). Temporal variations are nonsignificant. mg/dl, milligrams per deciliter; ng/dl, nanograms per deciliter; nmol/l, nanomoles per liter

Study IV

Figure 6 presents variations in tail vein glucose measurements conducted in C57BL/6J mice at 5-minute intervals at 1%, 1.5%, and 2% v/v ISO anesthesia postinduction. Glucose concentrations (mg/dl) ranged between 154.6±28.1 and 195.7±76.2 (1% v/v ISO, n = 7), between 161.7±42.9 and 239.0±38.3 (1.5%, n = 7), and between 158.7±51.9 and 204.6±48.8 (2%, n = 7). ANOVA analysis yielded significantly different tail vein glucose values at 1% v/v ISO compared to corresponding values at 1.5% or 2% v/v, at the 5% significance level (p < 0.0001). Mean glucose levels at ISO levels of 1.5% and 2% v/v were significantly higher compared to those at 1% (p < 0.0001).

Figure 6.

Temporal variation of tail vein glucose measurements at 5-minute intervals at 1%, 1.5%, and 2% volume-to-volume (v/v) isoflurane (ISO) anesthesia postinduction in male C57BL/6 mice (numbers per group as shown). mg/dl, milligrams per deciliter

Study V

Temporal variation of BG analyses (pH, p_aCO₂, p_aO₂, Na⁺, K⁺, Hct) at 1% and 2% v/v are shown in Figure 7, at 0, 20–43 (t₁), 60–64 (t₂), and 80–90 (t₃) minutes postinduction. No significant changes were noted at the later time intervals for most of these indices either compared to normal baseline values or for different ISO concentrations, indicating adequate respiration and blood oxygenation. Significant change was observed, however, in p_aO₂ values at the later time intervals (p < 0.0001), as expected after inspiration at 100% FiO₂. The lack of significant changes in Hct values from baseline (49±1%) indicates unaltered homeostasis. Hct values for 2% v/v ISO were 46±4% (t₁), 46±3% (t₂), and 45±4% (t₃), and for 1% v/v ISO were 45±5 % (t₁), 46±2 % (t₂), and 41±6% (t₃).

Figure 7.

Temporal variation of respiratory (pH, p_aCO₂, p_aO₂) and electrolyte parameters (Na⁺, K⁺, Hct) at 1% and 2% volume-to-volume (v/v) isoflurane (ISO) anesthesia 0–90 minutes postinduction. Values are means ± standard deviation (SD); n = 5–11 male C57BL/6 mice per group. Temporal variations are nonsignificant (except for p_aO₂). Hct, hematocrit; K⁺, potassium ion; mmHg, millimeters of mercury; mmol/l, millimoles per liter; Na⁺, sodium ion; p_aO₂, partial pressure of oxygen in the blood; p_aCO₂, partial pressure of carbon dioxide in the blood

Discussion

Taken together, the studies described above revealed the optimum level of isoflurane anesthesia to ensure stability of the murine cardiovascular function for up to 90 minutes. ISO effects were also considered in combination with administered oxygen (FiO₂) and N₂O. Based on 90-minute recordings, we observed that the most stable blood pressure and heart rate accompanied an ISO dose level of 1.5% v/v and that higher and less variable values of these measures resulted when FiO₂ was 25–50% and N₂O concentration was 50–75%. Table 1 shows these results alongside the major cardiovascular and biochemical indices measured in our studies.

Table 1.

Cardiovascular and biochemical indices in inhalationally anesthetized male C57BL/6 mice

		HR (bpm)	CVHR	MAP (mmHg)	CVMAP	tRR (ms)	pH	Glucose (mg/dl)	Insulin (ng/ml)
ISO = 1.0%	O2 = 100%	408±23 (n = 6/7) (65.5–90 min avg for 6 mice only)	14±2 (n = 7)	85±3 (n = 6/7) (63.5–90 min avg for 6 mice only)	13±3 (n = 7)		7.3±0.1 – 7.4±0.1 (n = 9/5)	176.3±26.1 – 188.8±11.1 (n = 4/5)	3.4±0.5 – 4.2±0.9 (n = 4/5)
ISO = 1.5%	O2 = 100%	478±11.4 (n = 7)	11±1 (n = 7)	92±2 (n = 7)	8±2 (n = 7)	114–134 (n = 4)
ISO = 2.0%	O2 = 100%	482±10 (n = 5/7) (78–90 mean reflects avg for 5 mice only)	14±3 (n = 7)	72±9 (n = 5–7); (73.6–77.6 mean reflects avg for 6 mice only; 77.6–90 mean avg for 5 mice)	33±2 (n = 7)		7.3±0.1 – 7.3±0.2 (n = 9/6)	139.4±30.1 – 192±11.9 (n = 4)	3.54±0.3 – 4.24±0.6 (n = 4)
ISO = 1.5%	O2 = 75%	486±21 (n = 6/7) (89.4–90.4 mean reflects avg for 6 mice only)	10±2 (n = 7)	83±4 (n = 6/7) (89.4–90.4 mean reflects avg for 6 mice only)	25±6 (n = 6/7) (89.4–90.4 mean reflects avg for 6 mice only)
ISO = 1.5%	O2 = 50%	475±21.6 (n = 6)	9±1 (n = 6)	90±1.3 (n = 6)	6±2 (n = 6)	108–150 (n = 3–6)
ISO = 1.5%	O2 = 20%	477±27 (n = 3–5) (86.9–87.9 mean reflects avg for 4 mice only; 87.9 reflects avg for 3 mice only [last time point])	8±2 (n = 5)	84±2 (n = 3–5) (86.9–87.9 mean reflects avg for 4 mice only; 87.9 reflects avg for 3 mice only [last time point])	10±4 (n = 5)
ISO = 1.5%	O2 = 50% N2 O = 50%	490±15.0 (n = 5)	10±2 (n = 5)	93±3 (n = 5)	6±3 (n = 5)	114–124 (n = 3–4)
ISO = 1.5%	O2= 75% N2 O = 25%	516±8.7 (n = 5)	7±2 (n = 5)	94±3 (n = 5)	10±3 (n = 5)

avg, average; bpm, beats per minute; CV, coefficient of variability; Hct, hematocrit; HR, heart rate; ISO, isoflurane; MAP, mean arterial pressure; mmHg, millimeters of mercury; mg/ml, milligrams per milliliter; ms, milliseconds; ng/ml, nanograms per milliliter; N₂O, nitrous oxide; O₂, oxygen; t_RR, time interval between successive QRS peaks

In theory, lower doses of ISO would closely match conscious physiological conditions, but our results indicated that lower levels of ISO may limit anesthetic depth and prohibit ease of animal handling, thus adversely affecting surgery, blood sampling, or restraint. Doses higher than 1.5% v/v seem to increase cardiodepression and thus are not advisable.

The ISO effects on MAP and HR are best explained through a reduction of sympathetic tone that reduces both cardiac output (CO) and total peripheral resistance (TPR). Direct dependence of MAP variations on beat-to-beat stroke volume (SV) variations is also possible under anesthetized conditions (Janssen and Smits 2002), notwithstanding reports of invasive, MRI, CT, and ultrasound studies that show relative constancy of temporospatial image-averaged SV values within physiological levels (Badea et al. 2005; Ruff et al. 1998). Thus ISO-induced changes in MAP are dependent on CO (the product of HR and SV) and TPR. Changes in the latter are dependent on ISO-related vasomotor effects and blood flow redistribution (Barbee et al. 1992; Rosenblum 1977; Sarin et al. 1990).

Toyama and colleagues (2004) used PET to assess the effects in mice of ISO on fluorodeoxyglucose (FDG) and its redistribution within 45 minutes postinduction. They observed a differential distribution of ISO with significantly increased uptake of FDG in heart and brain regions (91%) in comparison both to other regions and to awake controls. Their findings support the presence of ISO-induced coronary vasodilation and increased blood flow effects as well as altered metabolic patterns.

Researchers have proposed various mechanisms responsible for the ISO-mediated redistribution of blood flow: (1) direct effects on the vascular smooth muscle (induced release of endothelium-derived relaxing factor) (Loeb et al. 1998); (2) release of nitric oxide (NO) (Sarin et al. 1990; Loeb et al. 1998); (3) autoregulation and time-dependent adaptation (MAP-associated decreases as a result of vasodilation) (Crystal et al. 2000); and (4) metabolic changes. TPR changes have also been manifested through ISO-induced vasodilation leading to increases in coronary and cerebral blood flow (CBF). It appears that the CBF changes are the result of several interacting factors: the relaxing action of anesthetics on the vascular smooth muscle, the possible mediatory role of NO, time-dependent blood flow changes and redistribution, the influence of indirect metabolic mechanisms secondary to reduced cardiac work demand, pressure-flow autoregulation, and time-dependent vascular adaptation (Crystal et al. 2000).

Our results extend this ISO anesthesia titration study to FiO₂ titration and administration of medical air in the inhaled mixture (Figure 3). The optimal FiO₂ appears to be >50%, as reflected by the increased MAP values and their decreased variability. Additionally, the results shown in Figure 4 suggest that N₂O acts as a balancing agent by antagonizing ISO-induced vascular vasodilatory responses and thus stabilizing the mouse’s physiological status. The optimal amount of administered N₂O 25–75% results in higher and less variable MAP values than those observed in the absence of N₂O or in the FiO₂ titration study (Figure 3).

Adequate ventilation and oxygenation were also ensured through a careful match of the artificial ventilation parameters, inflow pressure, inspiratory and expiratory rates, and measured blood gas parameters. Slight deviations of p_aCO₂ from normal ranges (>45 mmHg) at late time intervals postinduction are attributed to possible respiratory distress due to anesthesia; administration of 100% O₂ led to a p_aO₂ of 150 mmHg, indicating hyperoxic or nonphysiological responses. All reported values for electrolytes and respiration rates were within expected ranges. Measured pH values were between 7.28 and 7.42 at the ISO levels 1% and 2%) throughout the 90-minute postinduction period. Hematocrit values were between 40.5% and 48.7% and did not vary in the cases of multiple (2–3) serial blood extractions at different intervals.

Reported physiological pH levels range between 7.2 and 7.43 in a number of mouse strains. Specifically, studies (Dalkara et al. 1995; Kapinya et al. 2002; Zuurbier et al. 2002) have indicated normal physiological murine arterial oxygen (p_aO₂) tensions greater than 65 mmHg (for 50% hemoglobin oxygen saturation) and 30–45 mmHg of p_aCO₂, given pressures of 5–20 cmH₂O at the inspiratory site, ventilator breathing rates of 80–280 breaths per minute, and minute volume values of 0.1–0.6 ml in multiple strains under isoflurane and urethane-α-chloralose anesthesia. Earlier publications on pH variation in the mouse (Bernstein 1966; Dalkara et al. 1995; Kapinya et al. 2002; Lahiri 1975; Siemann 1983; Weir 1949; Wolfe 1959) indicate that such variations are attributed to H₂O intake, food consumption, environment, temperature, or metabolic adaptations. Some studies have reported Hct values of 50–53% (John et al. 1996).

To ensure proper physiological conditions, it is important that blood oxygenation, myocardial oxygen consumption, and stress levels (as indicated by measures of hormones such as cortisol and androstenedione) not deviate from expected physiological ranges (Janssen and Smits 2002). Figure 5 shows cortisol and androstenedione levels immediately after induction of anesthesia; the levels suggest that the mice were handled with care and that their stress was not high (Tuli et al. 1995).

Blood glucose and insulin measurements from our studies show anesthesia-induced effects on mouse glucose metabolism (Figures 5 and 6): ISO levels of 1.5% produce a disruption of glucose metabolism upon induction, a mild hyperglycemic anesthesia-induced effect possibly associated with vasodilatory and increased blood flow effects. ISO-induced (1.5%, 2%) glucose increases of 38–47% were observed within the first 5 minutes after anesthesia administration, although the values stabilized for the remainder of the study, as shown in Figure 6. Such glucose changes and percentage increases are consistent with those reported in a recent study of age-, sex-, and strain-matched male mice upon 2% ISO administration (Durand et al. 2009).

From our study it is not possible to determine whether the mechanism of action involves transient sympathetic activation, steroid release, a direct effect of ISO, or a combination of such effects. The overall effects of ISO may be the result of opposing vasodilatory and vasoconstrictive effects, either directly (vasodilation) or secondary to an anesthesia-induced decrease of the heart’s metabolic demand, inferences supported by numerous published studies (Crystal et al. 1995; Kober et al. 2005; Ogushi et al. 1993; Park et al. 1994).

Conclusion

The protocol we have described for physiological studies of mice under anesthesia has the potential for high reproducibility in diagnostic modalities (including MRI, microCT, ultrasound, and microPET). Our results show that the optimal ISO anesthetic regimen for mice is a dose of 1.5% v/v mixed with 25–50% O₂ and 75–50% N₂O. This regimen can be useful in phenotypic screening and pharmacological studies of cardiac function in mice and can facilitate the migration of such work to noninvasive imaging platforms, with tremendous potential for both basic science and translational research.