Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39
Contents lists available at ScienceDirect
Journal of Pharmacological and Toxicological Methods journal homepage: www.elsevier.com/locate/jpharmtox
Original article
Monitoring the stress-level of rats with different types of anesthesia: A tail-artery cannulation protocol Dávid Z. Balla a,⁎, Saskia Schwarz a, Hannes M. Wiesner a, Anita M. Hennige b, Rolf Pohmann a a b
High-Field MR Center, Max-Planck-Institute for Biological Cybernetics, Spemannstr. 41, 72076 Tübingen, Germany Department of Internal Medicine, Eberhard Karls University, Otfried-Müller-Str. 10, 72076 Tübingen, Germany
a r t i c l e
i n f o
Article history: Received 17 November 2013 Accepted 5 March 2014 Available online 14 March 2014 Keywords: Anesthesia Artery Catheterization Cannulation Corticosterone fMRI Methods Rat Stress Tail
a b s t r a c t Introduction: Functional MRI in rats under anesthesia can largely minimize motion artifacts and attenuate the stress of the animal. However, two issues remain to be clarified and improved. First, fMRI results obtained with different types of anesthesia during surgical preparation and imaging show a large variability, which could be caused by the variable stress level of the rodents. Second, the most common surgical procedure used for anesthesia, blood gas analysis and mean arterial blood-pressure (MABP) monitoring is the femoral vein and artery catheterization that makes longitudinal studies difficult. Methods: In order to examine the variability of the stress level with three different anesthesia protocols using isoflurane (Iso), medetomidine–ketamine (MK) or propofol–remifentanil (PR), we measured the plasma corticosterone (CORT) concentration with 125I-radioimmunoassay in blood samples collected prior to, immediately after and 60 min after surgery. Tail-artery and vein catheterization was adapted for long-term monitoring of MABP with periodic blood sampling and is proposed as a less invasive and technically simple alternative to femoral vessel catheterization in fMRI preparation protocols. Results: We show that the CORT concentration depends on the anesthesia protocol with both alternatives providing more efficient stress reduction than the protocol using Iso. However, only the protocol using PR achieved a significant hormone reduction during surgery. Stress was not reliably manifested in changes in heart-rate and breathing-rate. Anesthesia and strain related changes in these two physiological parameters may be assigned to the pharmacological effects of the premedication and anesthetic agents. The results indicate also that MABP can be monitored over a long period of time (e.g. functional imaging session) through an arterial access point in the rat tail after cannulation with the proposed procedure. Discussion and conclusion: Animals can experience stress during fMRI preparation protocols without obvious signs in commonly monitored physiological parameters. Our results challenge the efficiency of surgical protocols using Iso as mono-anesthetic agent, even when extended with topical analgesia. It was demonstrated that the CORT-based stress-level measurement through tail-artery cannulation can be used for developing anesthesia protocols (i.e. the presented PR protocol) when setting up future fMRI studies. The proposed surgical method for the tail is expected to facilitate longitudinal fMRI studies with permanent arterial access. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The analysis of complex physiological processes in the brain of animal models, like stimulus responses in functional studies, relies on extensive monitoring of basic physiological parameters (Sanganahalli, Bailey, Herman, & Hyder, 2009; Yu et al., 2010). The mean arterial blood pressure (MABP) and the partial pressure of blood gases are sensitive indicators of physiological changes in the cardiovascular system. In functional magnetic resonance imaging (fMRI) studies, ⁎ Corresponding author at: Max-Planck-Institute for Biological Cybernetics, Spemannstr. 41, 72076 Tübingen, Germany. Tel.: +49 7071 601724. E-mail address: [email protected] (D.Z. Balla).
http://dx.doi.org/10.1016/j.vascn.2014.03.003 1056-8719/© 2014 Elsevier Inc. All rights reserved.
systemic changes in these parameters can have significant impact on the results (Silva & Stefanovic, 2008). Therefore, continuous monitoring of MABP and, in case of small animal experiments, periodic arterial blood gas analysis are part of the most fMRI protocols (Sanganahalli et al., 2009; Silva & Stefanovic, 2008; Yu et al., 2010, 2012). For ethical and scientific reasons, anesthesia with loss of consciousness, at least during surgical preparation, and sufficient surgical and post-operative analgesia are pivotal for an efficient imaging protocol (Fish, Brown, Danneman, & Karas, 2008; Hildebrandt, Su, & Weber, 2008). Especially in an fMRI study, a reproducible baseline physiological state of the resting brain shortly after surgery is required. However, since homeostasis is a dynamic equilibrium with regulatory processes activated by various stressors, the physiological state of the brain can
36
D.Z. Balla et al. / Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39
be affected by preceding procedures, like the handling of the animal before anesthesia or pain during or after surgery. Especially pain, which enhances the blood corticosterone (CORT) level and induces numerous physiological changes (e.g. in the cardiovascular system and brain metabolism), has direct impact on the physiological state even without any visible sign like muscle contraction, increased heart or breathing rate (Borsook & Becerra, 2011; Ferris & Stolberg, 2010; Fish et al., 2008). Since the half-life of free plasma CORT concentration is approximately 25 min (Sainio, Lehtola, & Roininen, 1988), trauma during surgery or insufficiently suppressed post-operative pain during fMRI can lead to continuously changing or constantly biased stress-hormone level during the entire experiment (Ferris & Stolberg, 2010). If functional imaging itself is performed under anesthesia, the pharmacological effects of the fMRI anesthetic agent on the already modulated or biased physiological state are difficult to predict (Borsook & Becerra, 2011; Sanganahalli et al., 2009). Thus, homeostatic regulatory mechanisms as a reaction to stressors, like unsuppressed pain, and the pharmacological effects of anesthetic agents used during surgical preparation (Lestage et al., 1985) can confound the relevance, reproducibility and comparability of the fMRI results preceded by artery catheterization. In this study, the analgesic efficiencies of three different anesthetic protocols during a minimally invasive tail artery catheterization in rats were investigated. Catheterization of the rat-tail was preferred (Guo & Zhou, 2003), because it is reported to cause less tissue damage and less stress for the animals than femoral and iliac artery cannulation, although the latter methods are established for fMRI studies (Silva & Stefanovic, 2008). As the main parameter for assessing the rat's stresslevel, and thus the analgesic efficiency, plasma CORT concentrations were quantified with 125I-radioimmunoassay (RIA) for plasma collected at different time-points of the three anesthesia protocols. The first investigated agent, isoflurane as inhalation gas, is commonly used in fMRI experiments for induction of the anesthesia and during surgical preparation (Masamoto, Kim, Fukuda, Wang, & Kim, 2007; Silva & Stefanovic, 2008; Yu et al., 2010, 2012). The second, medetomidine–ketamine as intraperitoneal (i.p.) bolus injection, has a broad spectrum of applications for long lasting surgical anesthesia (Fish et al., 2008), but has so far only rarely been used in animal preparation protocols for fMRI in rats. The third alternative, propofol-remifentanil as intravenous (i.v.) infusion, is a modern combination for general anesthesia with surgical analgesia used in clinical practice (Vuyk, Mertens, Olofsen, Burm, & Bovill, 1997). Since several studies report on rat strain differences in physiology and behavior (Staples & McGregor, 2006; Webb, Gowribai, & Muir, 2003), differences in the level of CORT due to differences in adrenal steroid receptor occupancy and corticosteroid-binding globulin levels in plasma during resting and stress conditions (Dhabhar, McEwen, & Spencer, 1993; Dhabhar, Miller, McEwen, & Spencer, 1995) and on strain dependent susceptibility to anesthetic and analgesic drugs (Avsaroglu, Van der Sar, Van Lith, Van Zutphen, & Hellebrekers, 2007; Yoon, Lee, Lee, Chung, & Chung, 1999), we analyzed three rat strains, Sprague–Dawley, Wistar and Fischer 344, which are often involved in fMRI projects running in our lab. 2. Materials and methods 2.1. Animal handling The study protocol was approved by the local authorities and the experiments were in accordance with the European directive for the handling of laboratory animals (86/609/EEC). Six healthy male CD (Sprague–Dawley), seven Wistar and nine Fischer 344 rats were delivered with 220–250 g (Charles River Laboratories, Germany) and housed individually until the experiment, but not longer than 3 weeks. In the animal housing, lights were switched on for 12 h at 20:00, temperature was regulated to 22 °C and humidity to 40–60%. Food and water were provided ad libitum. Body weight of the animals immediately before the experiment was in the range of 250–380 g, depending on the strain.
2.2. Protocol and monitoring 2.2.1. Preparation After induction of anesthesia, which was preceded by premedication for two of the protocols (see Section 2.2.4), the rat was transported to the surgery table and the head was placed in an inhalation mask with continuous supply of a 1/4 v/v O2 and air mixture. During the following interventions, body temperature, breathing rate (BR), heart rate (HR) and blood oxygen saturation (SpO2) were monitored continuously. Body temperature was kept at 37.5 °C by a feedback guided system composed of a rectal thermo-probe, an electric heating blanket and a controlling unit (Harvard Apparatus, UK). Throughout the experiment, plethysmography was performed with an air-filled pneumatic sensor connected to a piezoelectric transducer, and the electrocardiogram was recorded using subcutaneous (s.c.) needle electrodes (13 × 0.4 mm, 27 gauges) in a forepaw and the contra-lateral hind paw (Rapid Biomedical, Germany). SpO2, BR, HR, breath and pulse distention were measured with an infrared sensor clamped on the free hind paw and connected to a receiver system that amplified, filtered and isolated individual physiological signals (MouseOx, Starr Life Sciences, USA). The O2 concentration of the inhaled gas mixture was adjusted to maintain a SpO2 above 90%. Signals from different receivers were digitally recorded and visualized with a computer-based system (PowerLab & LabChart, ADInstruments, Australia). Tail vein cannulation was started after sensors were mounted, physiological parameters were found to be in the normal range and paw withdrawal (PWR) and tail flick reflexes (TFR) were tested negative. The rat was placed into the right lateral decubitus position and a 24 G neonate catheter (BD Insyte-N, Becton Dickinson Infusion Therapy Systems, USA) was introduced into the lateral tail vein. For the first plasma sample, 0.2–0.5 ml blood was collected and fluid loss was immediately compensated by injection of a balanced electrolyte solution (Jonosteril, Fresenius Kabi, Germany). 2.2.2. Surgery The rat was positioned dorsally and the tail was placed on a custommade holder made of a solid foam-substrate with a central strait for the tail, covered with aluminum foil to facilitate surface disinfection and reuse (see Supplementary material). A swab soaked with lidocaine (Licocainhydrochlorid 2%, 20 mg/ml, Bela Pharm, Germany) was placed for a few minutes on the site of surgery for topical pain suppression. A cranial–caudal 1–1.5 cm long incision through the skin was made approximately 2 mm right of the midline of the tail (adapted from (Guo & Zhou, 2003)). The skin was separated from the intact tissue underneath and was fixed with four needles to the foam-substrate. The artery was carefully prepared and two sutures were made, one cranially (open) and another caudally (closed, see Supplementary material for a depictive sketch). Closely below the cranial suture, a small artery clamp (S&T Vascular Clamps, Fine science tools, Germany) was attached to reduce bleeding during the following step. A 24 G neonate catheter was used to puncture the artery between the clamp and the caudal suture. Subsequently, the clamp was opened and the catheter was moved inside the artery 2 cm in the cranial direction. The cranial suture was closed to fixate the catheter in the actual position and the second blood sample was collected. The arterial catheter was then flushed with 0.1 ml heparinized (16 IU/ml) balanced electrolyte solution. Fluid loss was again compensated via the venous catheter. A pressure transducer (MLT0670, ADInstruments, Germany) was connected to the artery catheter through a PE-50 infusion-line filled with heparinized electrolyte solution (16 IU/ml) and the recording of MABP was initiated. Finally, the wound was covered with a swab soaked with saline solution to avoid tissue damage by desiccation. 2.2.3. Rest and euthanasia During the next 60 min post-surgical period, physiological parameters were recorded. After 1 h of rest, the third blood sample was
D.Z. Balla et al. / Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39
collected through the arterial access-point. Subsequently, the rats were euthanized with 0.5 ml pentobarbital i.v. (Narcoren, Merial, Germany). 2.2.4. Anesthesia protocols 2.2.4.1. Isoflurane (Iso). Anesthesia was induced in an induction box with 3% isoflurane in a 2/3 v/v mixture of O2 and air. After PWR and TFR were tested negative, the inhalation gas composition was modified to contain 1.5–2% isoflurane and 80% air. Isoflurane anesthesia was used on nine rats (three Fischer, three Wistar, three CD). Following frequently used practice in fMRI studies on rats, this protocol did not use premedication for sedation before induction. 2.2.4.2. Medetomidine–ketamine (MK). In this protocol, a premedication of 0.05 mg/kg atropine s.c. (Atropinsulfate 0.5 mg/ml, B. Braun, Germany) was given 10 min before the anesthesia induction was started for sedation and to reduce the saliva production allowing subsequent intubation and fMRI with mechanical ventilation. The sedative effect of an atropine bolus lasts for about 30 min, although the saliva production is reduced for hours. Then, a mixture of 0.4 mg/kg medetomidine (Domitor 1 mg/ml, Janssen-Cilag, Germany) and 60 mg/kg ketamine (Ketamin 10%, Bela Pharm, Germany) was injected intraperitoneally. Again, three rats of all three strains were submitted to this anesthesia protocol. 2.2.4.3. Propofol–remifentanil (PR). The premedication consisted again of 0.05 mg/kg atropine. Anesthesia was induced after 10 min with 3% isoflurane. During vein cannulation, anesthesia was maintained with 1.5–2% isoflurane. Subsequently, the venous access-point was used for the infusion of 0.6 mg/kg/min propofol (Propofol 2%, Fresenius Kabi, Germany) and 2 μg/kg/min remifentanil (Ultiva, GlaxoSmithKline, Germany). The two agents were pumped separately with precision pumps (PHD2000 Programmable & 11 plus, Harvard Apparatus, UK) and were mixed in the infusion-line with a three-way-cock. After the fluid mixture entered the venous catheter, isoflurane was successively reduced while carefully monitoring physiological parameters. This anesthesia protocol has not previously been used for the surgical preparation for fMRI experiments with rats and was applied here to four animals (3 Fischer, 1 Wistar). 2.2.5. Plasma sampling and analysis Blood was collected in 0.5 ml caps with 10 IU of heparin (25 000 IU/5 ml, Heparin-Natrium, Ratiopharm, Germany). The samples were centrifuged for 7 min in a small bench-top centrifuge with 2000 ×g (Roth, Germany). The plasma was transferred with a pipette into a new cap and was stored at − 80 °C in aliquots of 10 μl after the experiment. CORT plasma concentrations were analyzed using a 125 I-RIA kit following the manufacturer's instructions (ImmuChem Double Antibody CORT for Rats & Mice, MP Biomedicals, Germany). Measurements were done in triplicate and the averages of the three values served as inputs for the analyses presented. 2.3. Statistics Normality of the CORT level distribution, HR, BR and MABP values was tested in each group (defined by the grouping factors blood sampling time point, strain and anesthesia protocol) with the Shapiro–Wilk test. Where normality was not granted, absolute values of skewness and kurtosis of the distribution were compared to twice the standard error, examining the symmetry properties of the group. If the analyzed parameters exceeded the threshold value, the group was ignored in the statistical analysis (all groups were symmetric, hence this rule did not have to be applied — see Results). Variance analysis of normal groups was performed with one-way ANOVA or twosampled T-test, depending on whether 3 or 2 groups were compared. The variance of not Gaussian, but symmetric groups was investigated
37
with the Kruskal–Wallis-test. Groups were treated as independent variables and tests were performed for each grouping factor separately. For the classification of the corticosterone level as high or low, we used right-tailed or left-tailed T-tests, respectively. High CORT level refers to a hormone concentration of more than 200 ng/ml and low is defined as less than 100 ng/ml. This scaling was adapted from results of a study by Ferris and Stolberg (2010), where awake rats were exposed to modest stress (isolation in a novel cage) and an intense stressor (fox odor) and stress-hormone levels were measured with 125I-RIA. They found basal hormone levels of 31 ± 5 ng/ml, moderate stress-levels of 84.2 ± 30 ng/ml, while intense stress resulted in a hormone concentration of 307 ± 39 ng/ml. Statistical analysis was performed with SPSS (IBM SPSS Statistics, Version 22). The general significance threshold was set to p = 0.05. 3. Results In the group of Fischer-rats treated with MK, all experiments had to be aborted after the first blood sampling, because of fast blood coagulation inside the catheter. The reason for this effect is unknown, but was observed reproducibly in 6 experiments (in this group 3 additional trials on different rats were performed for verification). One additional dataset was rejected from the analysis (Wistar-rat, 250 g, Isoprotocol) because the animal showed signs of bad health (e.g. reduced weight and gasping during anesthesia). The results of the remaining 18 successful experiments are presented in Fig. 1 (hormone concentrations in plasma collected at the time points Prep, Surgery and Rest) and Table 1 (time averaged physiological parameters per experimental period). Here we focus on significant findings only, whereas the discussion section will cover all observations for a comprehensive interpretation of the data. CORT level reduction at Surgery and Rest relative to Prep was only found in the PR-Fischer groups (Fig. 1). A distinct feature of the PR-Fischer-Surgery group was the low stress level. Significant differences were found between the efficiencies of Iso and PR in the Fischer-Surgery and Fischer-Rest groups and between the Iso-CDSurgery and MK-CD-Surgery groups. Notably, the hormone level was high in the Iso-Fischer-Rest and the Iso-CD-Surgery groups. Strain related differences were found between the Iso-Wistar-Rest and Iso-FischerRest groups. Anesthesia effect differences on the HR were found between the Iso and MK protocols in the CD and Wistar groups (Table 1). Strain related differences in HR and BR were found between CD and Wistar rats when using MK. Iso decreased the HR after surgery in Fischer rats. Based on the comparison of Fig. 1 and Table 1, hormonal changes are not reliably manifested in changes of the monitored physiological parameters. Solely the strain related corticosterone concentration difference between the Iso-CD-Surgery and Iso-Wistar-Surgery groups was found concomitant to significant differences in HR. 4. Discussion and conclusion The observation of increased CORT level at Surgery relative to Prep in CD rats treated with Iso, though differences were not significant, comply with the results of Arnold and Langhans (2010) who compared the effect of venous blood sampling through a chronically installed jugularvein catheter in CD rats during different anesthesia protocols. In contrast, reductions in CORT relative to the Prep group were detected in the PR-Fischer groups and in the MK-Wistar groups. These results suggest on one hand that both PR and MK protocols reduce stress. On the other hand, Iso, even together with a local analgesic as described in the method section, did not sufficiently reduce the CORT level of the operated animal under anesthesia. The differences between the Iso-CD-Surgery and the MK-CD-Surgery groups (CORT and HR) are not surprising, owing to the abovementioned CORT enhancement during stress under Iso. Additionally, medetomidine is reported to cause moderate to strong bradycardia in rats (Fish et al., 2008). However, lower HR in MK than Iso groups is a
38
D.Z. Balla et al. / Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39
Rat strain CD
Wistar
Fischer
600 H
500 400
H H
300
Iso
200
0 600
*#
500
Blood coagulation (n = 6)
#
400
#
300
*
200
MK
*
100 0 600 500
n=1
n.a.
H
Anesthesia protocol
Corticosterone level [ng / ml]
100
400 300
PR
200 L
100 0 1
2
3
1
2
3
1
2
3
Blood sample
* *# # H L
One-way ANOVA, Post-hoc Bonferroni, p < 0.05 Two-sample T-test, p < 0.05 Kruskal-Wallis-test, p < 0.05 Suboptimal normality (Shapiro-Wilk-test failed), symmetry granted (skewness and kurtosis tests passed) Significantly high stress level, right-tailed T-test comparing a group and 200 ng / ml, p < 0.05 Significantly low stress level, left-tailed T-test comparing a group and 100 ng /ml, p < 0.05
Fig. 1. CORT concentrations at different blood sampling time-points (group mean ± standard deviation). Bar plots are grouped along the dimensions defined by anesthesia protocols and rat strains.
general observation (Table 1), which does not seem to correlate with differences in CORT level. Moreover, HR differences between Iso and MK groups appear already at Prep, before induction of anesthesia. This finding suggests that HR differences do not indicate a reduction of the stress level, and are not only governed by the pharmacological effect of medetomidine, but rather the mixed pharmacological effect of premedication (atropine) and the medetomidine–ketamine bolus. Anesthesia related differences in CORT level have also been detected between the Iso-Fischer-Surgery and the PR-Fischer-Surgery, as well as the Iso-Fischer-Rest and the PR-Fischer-Rest groups. Differences, although not significant, are also visible between the Iso-Wistar-Surgery and the MK-Wistar-Surgery groups in Fig. 1. These results extend our first conclusion indicating that the hormone responses under Iso and other anesthesia protocols were different, in fact not depending on the strain factor. Strain related differences in CORT level were found between Wistar and Fischer rats under Iso at Rest. The expected difference between CD and Fischer rats due to variations in adrenal steroid receptor occupancy and corticosteroid-binding globulin levels in plasma during resting and stress conditions (Dhabhar et al., 1993, 1995) are visible, though nonsignificant, in the Iso-Prep groups in Fig. 1. It is important to point out again that anesthesia induction took place after the Prep period, hence our observation complies with the abovementioned expectation for
conscious animals. Interestingly, HR and BR show clear differences (Table 1) between MK-CD and MK-Wistar groups at Prep (not treated with MK) and at Surgery (treated with MK). These differences are not present in the CORT level analysis. Therefore, we interpret these differences found in HR and BR as a strain dependent pharmacological effect of the premedication and the medetomidine–ketamine bolus, which supports also our previous conclusion on HR differences. The baseline hormone concentration at the time of measurement (circadian cycle) may affect at least the first blood sample results. Experiments usually started at 10:00, except for the Iso-CD and the PR-Fischer group, where anesthesia was induced in average at 12:40 and 13:55, respectively. We hypothesize that the CORT level in blood decreases continuously during the active period (Dallman, Akana, Bhatnagar, Bell, & Strack, 2000). Indeed, the hormone levels found in the Iso-CD-Prep group were lower than in the other Prep groups, but increased significantly after surgery. Since the bias to other Prep groups should be negative and given the high CORT level in the Iso-CD-Surgery group, the potential circadian effect could be ignored in this case. In the second case, the PR-Fischer groups, average CORT level at Prep was already high with a standard deviation of only 9% and thus, the spectacular reduction at Surgery could be exclusively assigned to the anesthesia protocol. Therefore, we assume that plasma CORT concentrations measured after surgery are not affected by circadian effects.
D.Z. Balla et al. / Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39 Table 1 Time average HR, BR and MABP values over the three main periods of the experiment (group mean ± standard deviation). Preparation
CD
MK
PR
HR
416 ± 45
330 ± 10
–
BR
75 ± 7
98 ± 24
–
79 ± 3
–
–
–
–
MABP [mm Hg] HR Wistar
Surgery
Iso
414 ± 61b 294 ± 20
408a
Iso
MK
422 ± 46 321 ± 22
Rest PR
Iso
MK
PR
–
388 ± 43 319 ± 17
–
109 ± 19
–
83 ± 10 114 ± 17
–
–
–
78 ± 6
–
415 ± 22 271 ± 18
347a
88 ± 3
397 ± 21 289 ± 14
332a
77 ± 9b
77 ± 26
102a
74 ± 5
74 ± 21b
81a
70 ± 1
85 ± 19b
83a
–
–
–
–
–
–
63a
86 ± 6
95a
HR
362 ± 9
–
352 ± 29 369 ± 23
–
–
347 ± 26
BR
66 ± 3
–
59 ± 10
64 ± 8
–
61±3
77 ± 3b
–
63 ± 11
–
–
–
–
–
–
91 ± 8
–
117 ± 23b
BR MABP [mm Hg]
< >
Fischer
MABP [mm Hg]
351±31 326 ± 12
b*N One-way ANOVA, Post-hoc Bonferroni, p b 0.05. ** Two-sample T-test, p b 0.05. *** Kruskal–Wallis-test, p b 0.05. a N = 1. b Suboptimal normality (Shapiro–Wilk-test failed), symmetry granted (skewness and kurtosis tests passed).
The relatively high MABP, the steadily low BR and the moderate and stable HR during the proposed PR-protocol are practical features, which might be very useful for studies of the brain function. The good reproducibility of the MABP measurements suggests that the proposed catheterization procedure yields reliable values and can thus be applied in routine examinations. Acknowledgments The study was supported by the Max-Planck-Society. References Arnold, M., & Langhans, W. (2010). Effects of anesthesia and blood sampling techniques on plasma metabolites and corticosterone in the rat. Physiology & Behavior, 99, 592–598.
39
Avsaroglu, H., Van der Sar, A., Van Lith, H., Van Zutphen, L., & Hellebrekers, L. (2007). Differences in response to anaesthetics and analgesics between inbred rat strains. Laboratory Animals, 41, 337–344. Borsook, D., & Becerra, L. (2011). CNS animal fMRI in pain and analgesia. Neuroscience & Biobehavioral Reviews, 35, 1125–1143. Dallman, M., Akana, S., Bhatnagar, S., Bell, M., & Strack, A. (2000). Bottomed out: Metabolic significance of circadian trough in glucocorticoid concentrations. International Journal of Obesity Related Metabolic Disorders, 24, S40–S46. Dhabhar, F. S., McEwen, B. S., & Spencer, R. L. (1993). Stress response, adrenal steroid receptor levels and corticosteroid-binding globulin levels — A comparison between Sprague–Dawley, Fischer 344 and Lewis rats. Brain Research, 616, 89–98. Dhabhar, F. S., Miller, A. H., McEwen, B. S., & Spencer, R. L. (1995). Differential activation of adrenal steroid receptors in neural and immune tissues of Sprague–Dawley, Fischer 344, and Lewis rats. Journal of Neuroimmunology, 56, 77–90. Ferris, C. F., & Stolberg, T. (2010). Imaging the immediate non-genomic effects of stress hormone on brain activity. Psychoneuroendocrinology, 35, 5–14. Fish, R. E., Brown, M. J., Danneman, P. J., & Karas, A. Z. (2008). Anesthesia and analgesia in laboratory animals (2nd ed.). Academic Press — Elsevier. Guo, Z. K., & Zhou, L. Z. (2003). Dual tail catheters for infusion and sampling in rats as an efficient for metabolic experiments. Lab Animal, 32, 45–48. Hildebrandt, I. J., Su, H., & Weber, W. A. (2008). Anesthesia and other considerations for in vivo imaging of small animals. ILAR Journal, 49, 17–26. Lestage, P., Vitte, P., Rolinat, J., Minot, R., Broussolle, E., & Bobillier, P. (1985). A chronic arterial and venous cannulation method for freely moving rats. Journal of Neuroscience Methods, 13, 213–222. Masamoto, K., Kim, T., Fukuda, M., Wang, P., & Kim, S. -G. (2007). Relationship between neural, vascular, and BOLD signals in isoflurane-anesthetized rat somatosensory cortex. Cerebral Cortex, 17, 942–950. Sainio, E. L., Lehtola, T., & Roininen, P. (1988). Radioimmunoassay of total and free corticosterone in rat plasma measurement of the effect of different doses of corticosterone. Steroids, 51, 609–622. Sanganahalli, B. G., Bailey, C. J., Herman, P., & Hyder, F. (2009). Tactile and non-tactile sensory paradigms for fMRI and neurophysiologic studies in rodents. In F. Hyder (Ed.), Dynamic Brain Imaging, 489. (pp. 213–242). Humana Press. Silva, A. C., & Stefanovic, B. (2008). Animal models in functional magnetic resonance imaging. In P. M. Conn (Ed.), Sourcebook of models for biomedical research (pp. 483–498). Humana Press. Staples, L. G., & McGregor, I. S. (2006). Defensive responses of Wistar and Sprague– Dawley rats to cat odour and TMT. Behavioural Brain Research, 172, 351–354. Vuyk, J., Mertens, M. J., Olofsen, E., Burm, A. G. L., & Bovill, J. G. (1997). Propofol anesthesia and rational opioid selection — Determination of optimal EC50–EC95 propofol-opioid concentrations that assure adequate anesthesia and a rapid return of consciousness. Anesthesiology, 87, 1549–1562. Webb, A. A., Gowribai, K., & Muir, G. D. (2003). Fischer (F-344) rats have different morphology, sensorimotor and locomotor abilities compared to Lewis, Long-Evans, Sprague–Dawley and Wistar rats. Behavioural Brain Research, 144, 143–156. Yoon, Y. W., Lee, D. H., Lee, B. H., Chung, K., & Chung, J. M. (1999). Different strains and substrains of rats show different levels of neuropathic pain behaviors. Experimental Brain Research, 129, 167–171. Yu, X., Glen, D., Wang, S., Dodd, S., Hirano, Y., Saad, Z., et al. (2012). Direct imaging of macrovascular and microvascular contributions to BOLD fMRI in layers IV–V of the rat whisker–barrel cortex. NeuroImage, 59, 1451–1460. Yu, X., Wang, S., Chen, D. -Y., Dodd, S., Goloshevsky, A., & Koretsky, A. P. (2010). 3D mapping of somatotopic reorganization with small animal functional MRI. NeuroImage, 49, 1667–1676.
Contents lists available at ScienceDirect
Journal of Pharmacological and Toxicological Methods journal homepage: www.elsevier.com/locate/jpharmtox
Original article
Monitoring the stress-level of rats with different types of anesthesia: A tail-artery cannulation protocol Dávid Z. Balla a,⁎, Saskia Schwarz a, Hannes M. Wiesner a, Anita M. Hennige b, Rolf Pohmann a a b
High-Field MR Center, Max-Planck-Institute for Biological Cybernetics, Spemannstr. 41, 72076 Tübingen, Germany Department of Internal Medicine, Eberhard Karls University, Otfried-Müller-Str. 10, 72076 Tübingen, Germany
a r t i c l e
i n f o
Article history: Received 17 November 2013 Accepted 5 March 2014 Available online 14 March 2014 Keywords: Anesthesia Artery Catheterization Cannulation Corticosterone fMRI Methods Rat Stress Tail
a b s t r a c t Introduction: Functional MRI in rats under anesthesia can largely minimize motion artifacts and attenuate the stress of the animal. However, two issues remain to be clarified and improved. First, fMRI results obtained with different types of anesthesia during surgical preparation and imaging show a large variability, which could be caused by the variable stress level of the rodents. Second, the most common surgical procedure used for anesthesia, blood gas analysis and mean arterial blood-pressure (MABP) monitoring is the femoral vein and artery catheterization that makes longitudinal studies difficult. Methods: In order to examine the variability of the stress level with three different anesthesia protocols using isoflurane (Iso), medetomidine–ketamine (MK) or propofol–remifentanil (PR), we measured the plasma corticosterone (CORT) concentration with 125I-radioimmunoassay in blood samples collected prior to, immediately after and 60 min after surgery. Tail-artery and vein catheterization was adapted for long-term monitoring of MABP with periodic blood sampling and is proposed as a less invasive and technically simple alternative to femoral vessel catheterization in fMRI preparation protocols. Results: We show that the CORT concentration depends on the anesthesia protocol with both alternatives providing more efficient stress reduction than the protocol using Iso. However, only the protocol using PR achieved a significant hormone reduction during surgery. Stress was not reliably manifested in changes in heart-rate and breathing-rate. Anesthesia and strain related changes in these two physiological parameters may be assigned to the pharmacological effects of the premedication and anesthetic agents. The results indicate also that MABP can be monitored over a long period of time (e.g. functional imaging session) through an arterial access point in the rat tail after cannulation with the proposed procedure. Discussion and conclusion: Animals can experience stress during fMRI preparation protocols without obvious signs in commonly monitored physiological parameters. Our results challenge the efficiency of surgical protocols using Iso as mono-anesthetic agent, even when extended with topical analgesia. It was demonstrated that the CORT-based stress-level measurement through tail-artery cannulation can be used for developing anesthesia protocols (i.e. the presented PR protocol) when setting up future fMRI studies. The proposed surgical method for the tail is expected to facilitate longitudinal fMRI studies with permanent arterial access. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The analysis of complex physiological processes in the brain of animal models, like stimulus responses in functional studies, relies on extensive monitoring of basic physiological parameters (Sanganahalli, Bailey, Herman, & Hyder, 2009; Yu et al., 2010). The mean arterial blood pressure (MABP) and the partial pressure of blood gases are sensitive indicators of physiological changes in the cardiovascular system. In functional magnetic resonance imaging (fMRI) studies, ⁎ Corresponding author at: Max-Planck-Institute for Biological Cybernetics, Spemannstr. 41, 72076 Tübingen, Germany. Tel.: +49 7071 601724. E-mail address: [email protected] (D.Z. Balla).
http://dx.doi.org/10.1016/j.vascn.2014.03.003 1056-8719/© 2014 Elsevier Inc. All rights reserved.
systemic changes in these parameters can have significant impact on the results (Silva & Stefanovic, 2008). Therefore, continuous monitoring of MABP and, in case of small animal experiments, periodic arterial blood gas analysis are part of the most fMRI protocols (Sanganahalli et al., 2009; Silva & Stefanovic, 2008; Yu et al., 2010, 2012). For ethical and scientific reasons, anesthesia with loss of consciousness, at least during surgical preparation, and sufficient surgical and post-operative analgesia are pivotal for an efficient imaging protocol (Fish, Brown, Danneman, & Karas, 2008; Hildebrandt, Su, & Weber, 2008). Especially in an fMRI study, a reproducible baseline physiological state of the resting brain shortly after surgery is required. However, since homeostasis is a dynamic equilibrium with regulatory processes activated by various stressors, the physiological state of the brain can
36
D.Z. Balla et al. / Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39
be affected by preceding procedures, like the handling of the animal before anesthesia or pain during or after surgery. Especially pain, which enhances the blood corticosterone (CORT) level and induces numerous physiological changes (e.g. in the cardiovascular system and brain metabolism), has direct impact on the physiological state even without any visible sign like muscle contraction, increased heart or breathing rate (Borsook & Becerra, 2011; Ferris & Stolberg, 2010; Fish et al., 2008). Since the half-life of free plasma CORT concentration is approximately 25 min (Sainio, Lehtola, & Roininen, 1988), trauma during surgery or insufficiently suppressed post-operative pain during fMRI can lead to continuously changing or constantly biased stress-hormone level during the entire experiment (Ferris & Stolberg, 2010). If functional imaging itself is performed under anesthesia, the pharmacological effects of the fMRI anesthetic agent on the already modulated or biased physiological state are difficult to predict (Borsook & Becerra, 2011; Sanganahalli et al., 2009). Thus, homeostatic regulatory mechanisms as a reaction to stressors, like unsuppressed pain, and the pharmacological effects of anesthetic agents used during surgical preparation (Lestage et al., 1985) can confound the relevance, reproducibility and comparability of the fMRI results preceded by artery catheterization. In this study, the analgesic efficiencies of three different anesthetic protocols during a minimally invasive tail artery catheterization in rats were investigated. Catheterization of the rat-tail was preferred (Guo & Zhou, 2003), because it is reported to cause less tissue damage and less stress for the animals than femoral and iliac artery cannulation, although the latter methods are established for fMRI studies (Silva & Stefanovic, 2008). As the main parameter for assessing the rat's stresslevel, and thus the analgesic efficiency, plasma CORT concentrations were quantified with 125I-radioimmunoassay (RIA) for plasma collected at different time-points of the three anesthesia protocols. The first investigated agent, isoflurane as inhalation gas, is commonly used in fMRI experiments for induction of the anesthesia and during surgical preparation (Masamoto, Kim, Fukuda, Wang, & Kim, 2007; Silva & Stefanovic, 2008; Yu et al., 2010, 2012). The second, medetomidine–ketamine as intraperitoneal (i.p.) bolus injection, has a broad spectrum of applications for long lasting surgical anesthesia (Fish et al., 2008), but has so far only rarely been used in animal preparation protocols for fMRI in rats. The third alternative, propofol-remifentanil as intravenous (i.v.) infusion, is a modern combination for general anesthesia with surgical analgesia used in clinical practice (Vuyk, Mertens, Olofsen, Burm, & Bovill, 1997). Since several studies report on rat strain differences in physiology and behavior (Staples & McGregor, 2006; Webb, Gowribai, & Muir, 2003), differences in the level of CORT due to differences in adrenal steroid receptor occupancy and corticosteroid-binding globulin levels in plasma during resting and stress conditions (Dhabhar, McEwen, & Spencer, 1993; Dhabhar, Miller, McEwen, & Spencer, 1995) and on strain dependent susceptibility to anesthetic and analgesic drugs (Avsaroglu, Van der Sar, Van Lith, Van Zutphen, & Hellebrekers, 2007; Yoon, Lee, Lee, Chung, & Chung, 1999), we analyzed three rat strains, Sprague–Dawley, Wistar and Fischer 344, which are often involved in fMRI projects running in our lab. 2. Materials and methods 2.1. Animal handling The study protocol was approved by the local authorities and the experiments were in accordance with the European directive for the handling of laboratory animals (86/609/EEC). Six healthy male CD (Sprague–Dawley), seven Wistar and nine Fischer 344 rats were delivered with 220–250 g (Charles River Laboratories, Germany) and housed individually until the experiment, but not longer than 3 weeks. In the animal housing, lights were switched on for 12 h at 20:00, temperature was regulated to 22 °C and humidity to 40–60%. Food and water were provided ad libitum. Body weight of the animals immediately before the experiment was in the range of 250–380 g, depending on the strain.
2.2. Protocol and monitoring 2.2.1. Preparation After induction of anesthesia, which was preceded by premedication for two of the protocols (see Section 2.2.4), the rat was transported to the surgery table and the head was placed in an inhalation mask with continuous supply of a 1/4 v/v O2 and air mixture. During the following interventions, body temperature, breathing rate (BR), heart rate (HR) and blood oxygen saturation (SpO2) were monitored continuously. Body temperature was kept at 37.5 °C by a feedback guided system composed of a rectal thermo-probe, an electric heating blanket and a controlling unit (Harvard Apparatus, UK). Throughout the experiment, plethysmography was performed with an air-filled pneumatic sensor connected to a piezoelectric transducer, and the electrocardiogram was recorded using subcutaneous (s.c.) needle electrodes (13 × 0.4 mm, 27 gauges) in a forepaw and the contra-lateral hind paw (Rapid Biomedical, Germany). SpO2, BR, HR, breath and pulse distention were measured with an infrared sensor clamped on the free hind paw and connected to a receiver system that amplified, filtered and isolated individual physiological signals (MouseOx, Starr Life Sciences, USA). The O2 concentration of the inhaled gas mixture was adjusted to maintain a SpO2 above 90%. Signals from different receivers were digitally recorded and visualized with a computer-based system (PowerLab & LabChart, ADInstruments, Australia). Tail vein cannulation was started after sensors were mounted, physiological parameters were found to be in the normal range and paw withdrawal (PWR) and tail flick reflexes (TFR) were tested negative. The rat was placed into the right lateral decubitus position and a 24 G neonate catheter (BD Insyte-N, Becton Dickinson Infusion Therapy Systems, USA) was introduced into the lateral tail vein. For the first plasma sample, 0.2–0.5 ml blood was collected and fluid loss was immediately compensated by injection of a balanced electrolyte solution (Jonosteril, Fresenius Kabi, Germany). 2.2.2. Surgery The rat was positioned dorsally and the tail was placed on a custommade holder made of a solid foam-substrate with a central strait for the tail, covered with aluminum foil to facilitate surface disinfection and reuse (see Supplementary material). A swab soaked with lidocaine (Licocainhydrochlorid 2%, 20 mg/ml, Bela Pharm, Germany) was placed for a few minutes on the site of surgery for topical pain suppression. A cranial–caudal 1–1.5 cm long incision through the skin was made approximately 2 mm right of the midline of the tail (adapted from (Guo & Zhou, 2003)). The skin was separated from the intact tissue underneath and was fixed with four needles to the foam-substrate. The artery was carefully prepared and two sutures were made, one cranially (open) and another caudally (closed, see Supplementary material for a depictive sketch). Closely below the cranial suture, a small artery clamp (S&T Vascular Clamps, Fine science tools, Germany) was attached to reduce bleeding during the following step. A 24 G neonate catheter was used to puncture the artery between the clamp and the caudal suture. Subsequently, the clamp was opened and the catheter was moved inside the artery 2 cm in the cranial direction. The cranial suture was closed to fixate the catheter in the actual position and the second blood sample was collected. The arterial catheter was then flushed with 0.1 ml heparinized (16 IU/ml) balanced electrolyte solution. Fluid loss was again compensated via the venous catheter. A pressure transducer (MLT0670, ADInstruments, Germany) was connected to the artery catheter through a PE-50 infusion-line filled with heparinized electrolyte solution (16 IU/ml) and the recording of MABP was initiated. Finally, the wound was covered with a swab soaked with saline solution to avoid tissue damage by desiccation. 2.2.3. Rest and euthanasia During the next 60 min post-surgical period, physiological parameters were recorded. After 1 h of rest, the third blood sample was
D.Z. Balla et al. / Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39
collected through the arterial access-point. Subsequently, the rats were euthanized with 0.5 ml pentobarbital i.v. (Narcoren, Merial, Germany). 2.2.4. Anesthesia protocols 2.2.4.1. Isoflurane (Iso). Anesthesia was induced in an induction box with 3% isoflurane in a 2/3 v/v mixture of O2 and air. After PWR and TFR were tested negative, the inhalation gas composition was modified to contain 1.5–2% isoflurane and 80% air. Isoflurane anesthesia was used on nine rats (three Fischer, three Wistar, three CD). Following frequently used practice in fMRI studies on rats, this protocol did not use premedication for sedation before induction. 2.2.4.2. Medetomidine–ketamine (MK). In this protocol, a premedication of 0.05 mg/kg atropine s.c. (Atropinsulfate 0.5 mg/ml, B. Braun, Germany) was given 10 min before the anesthesia induction was started for sedation and to reduce the saliva production allowing subsequent intubation and fMRI with mechanical ventilation. The sedative effect of an atropine bolus lasts for about 30 min, although the saliva production is reduced for hours. Then, a mixture of 0.4 mg/kg medetomidine (Domitor 1 mg/ml, Janssen-Cilag, Germany) and 60 mg/kg ketamine (Ketamin 10%, Bela Pharm, Germany) was injected intraperitoneally. Again, three rats of all three strains were submitted to this anesthesia protocol. 2.2.4.3. Propofol–remifentanil (PR). The premedication consisted again of 0.05 mg/kg atropine. Anesthesia was induced after 10 min with 3% isoflurane. During vein cannulation, anesthesia was maintained with 1.5–2% isoflurane. Subsequently, the venous access-point was used for the infusion of 0.6 mg/kg/min propofol (Propofol 2%, Fresenius Kabi, Germany) and 2 μg/kg/min remifentanil (Ultiva, GlaxoSmithKline, Germany). The two agents were pumped separately with precision pumps (PHD2000 Programmable & 11 plus, Harvard Apparatus, UK) and were mixed in the infusion-line with a three-way-cock. After the fluid mixture entered the venous catheter, isoflurane was successively reduced while carefully monitoring physiological parameters. This anesthesia protocol has not previously been used for the surgical preparation for fMRI experiments with rats and was applied here to four animals (3 Fischer, 1 Wistar). 2.2.5. Plasma sampling and analysis Blood was collected in 0.5 ml caps with 10 IU of heparin (25 000 IU/5 ml, Heparin-Natrium, Ratiopharm, Germany). The samples were centrifuged for 7 min in a small bench-top centrifuge with 2000 ×g (Roth, Germany). The plasma was transferred with a pipette into a new cap and was stored at − 80 °C in aliquots of 10 μl after the experiment. CORT plasma concentrations were analyzed using a 125 I-RIA kit following the manufacturer's instructions (ImmuChem Double Antibody CORT for Rats & Mice, MP Biomedicals, Germany). Measurements were done in triplicate and the averages of the three values served as inputs for the analyses presented. 2.3. Statistics Normality of the CORT level distribution, HR, BR and MABP values was tested in each group (defined by the grouping factors blood sampling time point, strain and anesthesia protocol) with the Shapiro–Wilk test. Where normality was not granted, absolute values of skewness and kurtosis of the distribution were compared to twice the standard error, examining the symmetry properties of the group. If the analyzed parameters exceeded the threshold value, the group was ignored in the statistical analysis (all groups were symmetric, hence this rule did not have to be applied — see Results). Variance analysis of normal groups was performed with one-way ANOVA or twosampled T-test, depending on whether 3 or 2 groups were compared. The variance of not Gaussian, but symmetric groups was investigated
37
with the Kruskal–Wallis-test. Groups were treated as independent variables and tests were performed for each grouping factor separately. For the classification of the corticosterone level as high or low, we used right-tailed or left-tailed T-tests, respectively. High CORT level refers to a hormone concentration of more than 200 ng/ml and low is defined as less than 100 ng/ml. This scaling was adapted from results of a study by Ferris and Stolberg (2010), where awake rats were exposed to modest stress (isolation in a novel cage) and an intense stressor (fox odor) and stress-hormone levels were measured with 125I-RIA. They found basal hormone levels of 31 ± 5 ng/ml, moderate stress-levels of 84.2 ± 30 ng/ml, while intense stress resulted in a hormone concentration of 307 ± 39 ng/ml. Statistical analysis was performed with SPSS (IBM SPSS Statistics, Version 22). The general significance threshold was set to p = 0.05. 3. Results In the group of Fischer-rats treated with MK, all experiments had to be aborted after the first blood sampling, because of fast blood coagulation inside the catheter. The reason for this effect is unknown, but was observed reproducibly in 6 experiments (in this group 3 additional trials on different rats were performed for verification). One additional dataset was rejected from the analysis (Wistar-rat, 250 g, Isoprotocol) because the animal showed signs of bad health (e.g. reduced weight and gasping during anesthesia). The results of the remaining 18 successful experiments are presented in Fig. 1 (hormone concentrations in plasma collected at the time points Prep, Surgery and Rest) and Table 1 (time averaged physiological parameters per experimental period). Here we focus on significant findings only, whereas the discussion section will cover all observations for a comprehensive interpretation of the data. CORT level reduction at Surgery and Rest relative to Prep was only found in the PR-Fischer groups (Fig. 1). A distinct feature of the PR-Fischer-Surgery group was the low stress level. Significant differences were found between the efficiencies of Iso and PR in the Fischer-Surgery and Fischer-Rest groups and between the Iso-CDSurgery and MK-CD-Surgery groups. Notably, the hormone level was high in the Iso-Fischer-Rest and the Iso-CD-Surgery groups. Strain related differences were found between the Iso-Wistar-Rest and Iso-FischerRest groups. Anesthesia effect differences on the HR were found between the Iso and MK protocols in the CD and Wistar groups (Table 1). Strain related differences in HR and BR were found between CD and Wistar rats when using MK. Iso decreased the HR after surgery in Fischer rats. Based on the comparison of Fig. 1 and Table 1, hormonal changes are not reliably manifested in changes of the monitored physiological parameters. Solely the strain related corticosterone concentration difference between the Iso-CD-Surgery and Iso-Wistar-Surgery groups was found concomitant to significant differences in HR. 4. Discussion and conclusion The observation of increased CORT level at Surgery relative to Prep in CD rats treated with Iso, though differences were not significant, comply with the results of Arnold and Langhans (2010) who compared the effect of venous blood sampling through a chronically installed jugularvein catheter in CD rats during different anesthesia protocols. In contrast, reductions in CORT relative to the Prep group were detected in the PR-Fischer groups and in the MK-Wistar groups. These results suggest on one hand that both PR and MK protocols reduce stress. On the other hand, Iso, even together with a local analgesic as described in the method section, did not sufficiently reduce the CORT level of the operated animal under anesthesia. The differences between the Iso-CD-Surgery and the MK-CD-Surgery groups (CORT and HR) are not surprising, owing to the abovementioned CORT enhancement during stress under Iso. Additionally, medetomidine is reported to cause moderate to strong bradycardia in rats (Fish et al., 2008). However, lower HR in MK than Iso groups is a
38
D.Z. Balla et al. / Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39
Rat strain CD
Wistar
Fischer
600 H
500 400
H H
300
Iso
200
0 600
*#
500
Blood coagulation (n = 6)
#
400
#
300
*
200
MK
*
100 0 600 500
n=1
n.a.
H
Anesthesia protocol
Corticosterone level [ng / ml]
100
400 300
PR
200 L
100 0 1
2
3
1
2
3
1
2
3
Blood sample
* *# # H L
One-way ANOVA, Post-hoc Bonferroni, p < 0.05 Two-sample T-test, p < 0.05 Kruskal-Wallis-test, p < 0.05 Suboptimal normality (Shapiro-Wilk-test failed), symmetry granted (skewness and kurtosis tests passed) Significantly high stress level, right-tailed T-test comparing a group and 200 ng / ml, p < 0.05 Significantly low stress level, left-tailed T-test comparing a group and 100 ng /ml, p < 0.05
Fig. 1. CORT concentrations at different blood sampling time-points (group mean ± standard deviation). Bar plots are grouped along the dimensions defined by anesthesia protocols and rat strains.
general observation (Table 1), which does not seem to correlate with differences in CORT level. Moreover, HR differences between Iso and MK groups appear already at Prep, before induction of anesthesia. This finding suggests that HR differences do not indicate a reduction of the stress level, and are not only governed by the pharmacological effect of medetomidine, but rather the mixed pharmacological effect of premedication (atropine) and the medetomidine–ketamine bolus. Anesthesia related differences in CORT level have also been detected between the Iso-Fischer-Surgery and the PR-Fischer-Surgery, as well as the Iso-Fischer-Rest and the PR-Fischer-Rest groups. Differences, although not significant, are also visible between the Iso-Wistar-Surgery and the MK-Wistar-Surgery groups in Fig. 1. These results extend our first conclusion indicating that the hormone responses under Iso and other anesthesia protocols were different, in fact not depending on the strain factor. Strain related differences in CORT level were found between Wistar and Fischer rats under Iso at Rest. The expected difference between CD and Fischer rats due to variations in adrenal steroid receptor occupancy and corticosteroid-binding globulin levels in plasma during resting and stress conditions (Dhabhar et al., 1993, 1995) are visible, though nonsignificant, in the Iso-Prep groups in Fig. 1. It is important to point out again that anesthesia induction took place after the Prep period, hence our observation complies with the abovementioned expectation for
conscious animals. Interestingly, HR and BR show clear differences (Table 1) between MK-CD and MK-Wistar groups at Prep (not treated with MK) and at Surgery (treated with MK). These differences are not present in the CORT level analysis. Therefore, we interpret these differences found in HR and BR as a strain dependent pharmacological effect of the premedication and the medetomidine–ketamine bolus, which supports also our previous conclusion on HR differences. The baseline hormone concentration at the time of measurement (circadian cycle) may affect at least the first blood sample results. Experiments usually started at 10:00, except for the Iso-CD and the PR-Fischer group, where anesthesia was induced in average at 12:40 and 13:55, respectively. We hypothesize that the CORT level in blood decreases continuously during the active period (Dallman, Akana, Bhatnagar, Bell, & Strack, 2000). Indeed, the hormone levels found in the Iso-CD-Prep group were lower than in the other Prep groups, but increased significantly after surgery. Since the bias to other Prep groups should be negative and given the high CORT level in the Iso-CD-Surgery group, the potential circadian effect could be ignored in this case. In the second case, the PR-Fischer groups, average CORT level at Prep was already high with a standard deviation of only 9% and thus, the spectacular reduction at Surgery could be exclusively assigned to the anesthesia protocol. Therefore, we assume that plasma CORT concentrations measured after surgery are not affected by circadian effects.
D.Z. Balla et al. / Journal of Pharmacological and Toxicological Methods 70 (2014) 35–39 Table 1 Time average HR, BR and MABP values over the three main periods of the experiment (group mean ± standard deviation). Preparation
CD
MK
PR
HR
416 ± 45
330 ± 10
–
BR
75 ± 7
98 ± 24
–
79 ± 3
–
–
–
–
MABP [mm Hg] HR Wistar
Surgery
Iso
414 ± 61b 294 ± 20
408a
Iso
MK
422 ± 46 321 ± 22
Rest PR
Iso
MK
PR
–
388 ± 43 319 ± 17
–
109 ± 19
–
83 ± 10 114 ± 17
–
–
–
78 ± 6
–
415 ± 22 271 ± 18
347a
88 ± 3
397 ± 21 289 ± 14
332a
77 ± 9b
77 ± 26
102a
74 ± 5
74 ± 21b
81a
70 ± 1
85 ± 19b
83a
–
–
–
–
–
–
63a
86 ± 6
95a
HR
362 ± 9
–
352 ± 29 369 ± 23
–
–
347 ± 26
BR
66 ± 3
–
59 ± 10
64 ± 8
–
61±3
77 ± 3b
–
63 ± 11
–
–
–
–
–
–
91 ± 8
–
117 ± 23b
BR MABP [mm Hg]
< >
Fischer
MABP [mm Hg]
351±31 326 ± 12
b*N One-way ANOVA, Post-hoc Bonferroni, p b 0.05. ** Two-sample T-test, p b 0.05. *** Kruskal–Wallis-test, p b 0.05. a N = 1. b Suboptimal normality (Shapiro–Wilk-test failed), symmetry granted (skewness and kurtosis tests passed).
The relatively high MABP, the steadily low BR and the moderate and stable HR during the proposed PR-protocol are practical features, which might be very useful for studies of the brain function. The good reproducibility of the MABP measurements suggests that the proposed catheterization procedure yields reliable values and can thus be applied in routine examinations. Acknowledgments The study was supported by the Max-Planck-Society. References Arnold, M., & Langhans, W. (2010). Effects of anesthesia and blood sampling techniques on plasma metabolites and corticosterone in the rat. Physiology & Behavior, 99, 592–598.
39
Avsaroglu, H., Van der Sar, A., Van Lith, H., Van Zutphen, L., & Hellebrekers, L. (2007). Differences in response to anaesthetics and analgesics between inbred rat strains. Laboratory Animals, 41, 337–344. Borsook, D., & Becerra, L. (2011). CNS animal fMRI in pain and analgesia. Neuroscience & Biobehavioral Reviews, 35, 1125–1143. Dallman, M., Akana, S., Bhatnagar, S., Bell, M., & Strack, A. (2000). Bottomed out: Metabolic significance of circadian trough in glucocorticoid concentrations. International Journal of Obesity Related Metabolic Disorders, 24, S40–S46. Dhabhar, F. S., McEwen, B. S., & Spencer, R. L. (1993). Stress response, adrenal steroid receptor levels and corticosteroid-binding globulin levels — A comparison between Sprague–Dawley, Fischer 344 and Lewis rats. Brain Research, 616, 89–98. Dhabhar, F. S., Miller, A. H., McEwen, B. S., & Spencer, R. L. (1995). Differential activation of adrenal steroid receptors in neural and immune tissues of Sprague–Dawley, Fischer 344, and Lewis rats. Journal of Neuroimmunology, 56, 77–90. Ferris, C. F., & Stolberg, T. (2010). Imaging the immediate non-genomic effects of stress hormone on brain activity. Psychoneuroendocrinology, 35, 5–14. Fish, R. E., Brown, M. J., Danneman, P. J., & Karas, A. Z. (2008). Anesthesia and analgesia in laboratory animals (2nd ed.). Academic Press — Elsevier. Guo, Z. K., & Zhou, L. Z. (2003). Dual tail catheters for infusion and sampling in rats as an efficient for metabolic experiments. Lab Animal, 32, 45–48. Hildebrandt, I. J., Su, H., & Weber, W. A. (2008). Anesthesia and other considerations for in vivo imaging of small animals. ILAR Journal, 49, 17–26. Lestage, P., Vitte, P., Rolinat, J., Minot, R., Broussolle, E., & Bobillier, P. (1985). A chronic arterial and venous cannulation method for freely moving rats. Journal of Neuroscience Methods, 13, 213–222. Masamoto, K., Kim, T., Fukuda, M., Wang, P., & Kim, S. -G. (2007). Relationship between neural, vascular, and BOLD signals in isoflurane-anesthetized rat somatosensory cortex. Cerebral Cortex, 17, 942–950. Sainio, E. L., Lehtola, T., & Roininen, P. (1988). Radioimmunoassay of total and free corticosterone in rat plasma measurement of the effect of different doses of corticosterone. Steroids, 51, 609–622. Sanganahalli, B. G., Bailey, C. J., Herman, P., & Hyder, F. (2009). Tactile and non-tactile sensory paradigms for fMRI and neurophysiologic studies in rodents. In F. Hyder (Ed.), Dynamic Brain Imaging, 489. (pp. 213–242). Humana Press. Silva, A. C., & Stefanovic, B. (2008). Animal models in functional magnetic resonance imaging. In P. M. Conn (Ed.), Sourcebook of models for biomedical research (pp. 483–498). Humana Press. Staples, L. G., & McGregor, I. S. (2006). Defensive responses of Wistar and Sprague– Dawley rats to cat odour and TMT. Behavioural Brain Research, 172, 351–354. Vuyk, J., Mertens, M. J., Olofsen, E., Burm, A. G. L., & Bovill, J. G. (1997). Propofol anesthesia and rational opioid selection — Determination of optimal EC50–EC95 propofol-opioid concentrations that assure adequate anesthesia and a rapid return of consciousness. Anesthesiology, 87, 1549–1562. Webb, A. A., Gowribai, K., & Muir, G. D. (2003). Fischer (F-344) rats have different morphology, sensorimotor and locomotor abilities compared to Lewis, Long-Evans, Sprague–Dawley and Wistar rats. Behavioural Brain Research, 144, 143–156. Yoon, Y. W., Lee, D. H., Lee, B. H., Chung, K., & Chung, J. M. (1999). Different strains and substrains of rats show different levels of neuropathic pain behaviors. Experimental Brain Research, 129, 167–171. Yu, X., Glen, D., Wang, S., Dodd, S., Hirano, Y., Saad, Z., et al. (2012). Direct imaging of macrovascular and microvascular contributions to BOLD fMRI in layers IV–V of the rat whisker–barrel cortex. NeuroImage, 59, 1451–1460. Yu, X., Wang, S., Chen, D. -Y., Dodd, S., Goloshevsky, A., & Koretsky, A. P. (2010). 3D mapping of somatotopic reorganization with small animal functional MRI. NeuroImage, 49, 1667–1676.
