The Use of Midazolam, Isoflurane, and Nitrous Oxide for Sedation and Anesthesia of Ball Pythons (Python Regius)

Home , Concentration effect, Fluroxene, Halopropane, Radiated ratsnake, Second gas effect

The Use of Midazolam, Isoflurane, and Nitrous Oxide for Sedation and Anesthesia of Ball Pythons (Python regius)

Cédric B. Larouche

A Thesis presented to The University of Guelph

In partial fulfilment of requirements for the degree of Doctor of Veterinary Science in Pathobiology

Guelph, Ontario, Canada

THE USE OF MIDAZOLAM, ISOFLURANE, AND NITROUS OXIDE FOR SEDATION

AND ANESTHESIA OF BALL PYTHONS (PYTHON REGIUS)

Cédric B. Larouche Co-advisors: University of Guelph, 2019 Nicole Nemeth Dorothee Bienzle

The objectives of this project were to characterize the pharmacodynamics and pharmacokinetics of midazolam in the ball python (Python regius), and to evaluate the effects of midazolam and nitrous oxide (N2O) on the minimum anesthetic concentration of isoflurane (MACIso) in this species.

The pharmacodynamics of midazolam were evaluated in a blinded, randomized, crossover study comparing two dosages (1 and 2 mg/kg intramuscular [IM]) in ten ball pythons. There were no significant differences between the two dosages for any parameter evaluated. Midazolam provided moderate sedation and muscle relaxation in all snakes starting 10-15 min post-injection and lasting up to 3-5 days, with a peak effect

60 min post-injection. Heart rates were significantly lower than baseline from 30 min to

5.3 days post-injection with both dosages. In an open trial using 9 ball pythons, the reversal effect of flumazenil was evaluated by administering 0.08 mg/kg IM 60 min after the administration of 1 mg/kg IM of midazolam. Sedation and muscle relaxation were fully reversed within 10 min of administration, but all snakes fully resedated within 3 h.

In a semi-blinded, randomized, crossover trial, nine ball pythons were used to evaluate the effects of 1 mg/kg IM midazolam and N2O mixed 50:50 with oxygen on the

MACIso. Midazolam significantly decreased the MACIso from 1.11% (95% confidence interval [CI], 0.94-1.28%) to 0.48% (95% CI, 0.29-0.67%), and N2O decreased the

MACIso to 0.92% (95% CI, 0.74-1.09%). No clinically relevant differences between treatments were noted regarding blood gases but isoflurane induction with face mask caused marked acidemia that resolved with mechanical ventilation.

Central venous catheters were surgically placed in twelve ball pythons to evaluate the pharmacokinetics of midazolam. Snakes were randomly and equally divided into two groups receiving 1 mg/kg of midazolam either intracardiac (IC; group 1) or IM (group 2). Plasma concentrations of midazolam and its major metabolite 1- hydroxymidazolam were determined using high-performance liquid chromatography tandem-mass spectrometry. Mean bioavailability following IM administration was 89%.

Mean ± SD clearance of midazolam was 0.053 ± 0.008 L/h/kg, leading to slow elimination half-lives (12.04 ± 3.25 h [IC] and 16.54 ± 7.10 h [IM]).

ACKNOWLEDGEMENTS

This thesis has been a series of mostly self-imposed, tight deadlines. I would therefore like to thank all the members of my advisory committee for dealing so brilliantly with all my last-minute questions and revision requests: Chris Dutton, Dorothee Bienzle, Nicole Nemeth, Hugues Beaufrère, Craig Mosley, and Ron Johnson. There are so many people I wish to thank for their contribution and support. If I have forgotten to mention you, please feel free to request a beer any time.

I would like to thank the Toronto Zoo, OVC Pet Trust, and the Andrea Leger Dunbar Summer Research Studentship for providing funding for this DVSc. This project would have never been possible without their financial support.

I would like to thank Chris Dutton and Pauline Delnatte for being my anchor from day 1 until the very end of my residency. With so many people coming and going, you two have been a constant I could always rely on throughout this project and residency. Chris, your enthusiasm is contagious and working with you is such a delight. You are a brilliant veterinarian and leader, and I consider myself extremely lucky to have been your resident for 3 years. Pauline, you are one of the most dedicated and talented veterinarians I know. Your trust in my skills and your desire to do what’s best for every single animal made every second spent with you a learning opportunity.

I am very grateful to Dorothee Bienzle for her constant support. This project would not have been completed as smoothly without your advice and attention to my timeline. I would also like to thank Hugues Beaufrère for his major help in every step of the project, from providing the original idea to the statistical analyses, and the ACZM preparation classes.

This project would have never been successful without the amazing work of the Toronto Zoo’s Wildlife Health Center staff. All of you either took care of my massive ball pythons, helped me in some way with the experimentation, and/or made my whole residency such a delight. Thank you Tasha Long, Rick Vos, Mark Bongelli, Andrea Dada, Dianne Morrison, Nigel Parr, Lydia Attard, Gerri Mintha, Margaret Kolakowski, Brian Telford, Charles Guthrie, Jason Dawes, Kerri Chase, and Julie Digiandomenico. I especially want to mention the invaluable help from the most amazing veterinary technicians I could have hoped for: Michelle Lovering, Dawn Mihailovic, Cassia Devison, and Stephanie Fleming. I would also like to thank Simon Hollamby and David McLelland for all the teaching opportunities.

I would like to thank Kristopher Afshaun Zaman for his huge help and dedication to the project. You made my life so much easier with your exceptional technical and mental support during the summer 2018. These ball pythons are extremely grateful for the amazing care you provided. I am grateful for all the students who provided technical

support for the experimentation or care of these snakes, especially Hannah Bagnall and Sydney Swartz for their help with the pharmacokinetic project, and Malika Ladak for her help with my first side project on tarantulas.

The last chapter of this thesis on the pharmacokinetics of midazolam would have never been possible without the dedication of Yu Gu. I would like to thank her, Ron Johnson, and everyone who helped with this assay. I am grateful to Francis Beaudry for his help with the pharmacokinetic analysis. I contacted you out of nowhere and you immediately helped.

I would like to acknowledge again the support of my whole advisory committee. Craig, you offered me invaluable advices on the planning, execution, and analysis of this project. Nicole, your support during the preparation phase was highly appreciated. Your talent in writing and revising scientific manuscripts significantly improved the quality of this thesis.

I am very grateful to Leonardo Susta for his incredible dedication to the zoo pathology program. With all the major changes that happened to the program during my residency, you took it upon yourself to make sure the quality of the teaching would be maintained, and I cannot express how much I learned screening and discussing with you. I would also like to thank all the other pathologists who taught me everything I know about pathology, especially Emily Brouwer, Nicole Nemeth, Dale Smith, and Rob Foster. My weekly trips to Guelph would have been much harder without the great comradery of the anatomical and clinical pathology residents in the microscope room. Thank you for being such an awesome team!

I cannot mention all these people without naming my co-residents, who made this whole residency such a delight. Ellie Milnes, you are such a strong, confident, and talented person; nothing can stop you from accomplishing your dreams. Your informal help with every step of this project and my residency probably had the biggest influence on my success. Thank you so much for being that amazing! Jessica Aymen, we didn’t get to work very long together as co-residents, but I wanted to thank you as well for being such an amazing person. The young veterinary student I knew back in Saint- Hyacinthe has become a confident and very talented veterinarian in such a small period of time.

This whole residency craziness started with my first partner in crime, Émilie L. Couture. You introduced me to the zoological medicine department and always pushed me to reach further. The whole team in Saint-Hyacinthe made the step into the residency so much smoother. I would like to thank especially Graham Zoller, Guy Fitzgerald, Isabelle Langlois, Stéphane Lair, Édouard Maccolini, Julie Hébert, Marion Desmarchelier, Marie-Josée Limoges, Shannon Ferrell, Valérie Lacasse, and Léa Ahmed.

I am infinitely grateful to Adam McCall for his immense support throughout this residency. I have no idea how I could have gone through all of this without your help.

Finally, I would like to thank my family that has always been there for me, supporting every choice and steps that I made. Without all the opportunities you have given me, all the lessons you’ve taught me, and all the love you’ve provided, none of this would have ever been possible.

Cédric B. Larouche

Scarborough, August 2019

vii

TABLE OF CONTENTS

Abstract ...... ii

Acknowledgements ...... iv

Table of Contents ...... vii

List of Tables ...... xi

List of Figures ...... xii

List of Symbols, Abbreviations, or Nomenclature ...... xiv

List of Appendices ...... xvi

1 Literature review ...... 1

1.1 Ophidian cardiopulmonary anatomy and physiology ...... 1

1.1.1 Ophidian respiratory system ...... 1

1.1.2 Ophidian cardiovascular system ...... 5

1.2 Inhalation anesthesia ...... 11

1.2.1 Introduction ...... 11

1.2.2 Physicochemical properties and potency of anesthetic gases ...... 11

1.2.3 Isoflurane ...... 17

1.2.4 Nitrous oxide ...... 19

1.3 Midazolam ...... 23

1.3.1 History and physicochemical properties ...... 23

1.3.2 Pharmacokinetics ...... 23

1.3.3 Pharmacodynamics ...... 25

1.3.4 MAC-sparing effects ...... 28

1.3.5 Flumazenil: a benzodiazepine antagonist ...... 28

viii

1.4 Anesthesia of reptiles ...... 30

1.4.1 The challenges of reptile anesthesia ...... 30

1.4.2 Inhalation anesthesia ...... 32

1.4.3 Midazolam ...... 34

1.5 Study rationale and objectives ...... 35

2 Evaluation of the effects of midazolam and flumazenil in the ball python (Python regius) ...... 36

2.1 Abstract ...... 37

2.2 Introduction ...... 38

2.3 Material and methods ...... 39

2.3.1 Animals ...... 39

2.3.2 Midazolam trial ...... 39

2.3.3 Flumazenil trial ...... 41

2.3.4 Statistical analysis ...... 42

2.4 Results ...... 42

2.4.1 Animals ...... 42

2.4.2 Midazolam trial ...... 42

2.4.3 Flumazenil trial ...... 44

2.5 Discussion ...... 48

2.6 Conclusions...... 50

3 The effects of midazolam and nitrous oxide on the minimum anesthetic concentration of isoflurane in the ball python (Python regius) ...... 52

3.1 Abstract ...... 53

3.2 Introduction ...... 54

3.3 Material and methods ...... 55

3.3.1 Animals ...... 55

3.3.2 Study design and anesthesia ...... 55

3.3.3 Determination of MAC ...... 56

3.3.4 Statistical analysis ...... 57

3.4 Results ...... 58

3.5 Discussion ...... 61

3.6 Conclusions...... 64

4 Pharmacokinetics of midazolam and its major metabolite 1- hydroxymidazolam in the ball python (Python regius) after intracardiac and intramuscular administrations ...... 65

4.1 Abstract ...... 66

4.2 Introduction ...... 67

4.3 Material and methods ...... 67

4.3.1 Animals ...... 68

4.3.2 Surgical catheter placement ...... 68

4.3.3 Experimental design and blood sampling ...... 69

4.3.4 Midazolam analysis ...... 70

4.3.5 Method validation ...... 72

4.3.6 Pharmacokinetics analysis ...... 72

4.4 Results ...... 73

4.5 Discussion ...... 79

5 Conclusions and future directions ...... 82

References ...... 85

APPENDIX A: Minimum anesthetic concentration of isoflurane in various species .... 113

APPENDIX B: Descriptive data of the animal research collection ...... 117

APPENDIX C: Descriptive data from chapter 2 ...... 119

APPENDIX D: Descriptive data from chapter 3 ...... 126

APPENDIX E: Validation of the methods to quantify the plasma concentration of midazolam and 1-hydroxymidazolam in the ball python and descriptive data from chapter 4 ...... 132

LIST OF TABLES

Table 1.1 – Physicochemical properties of commonly used anesthetic gases.* ...... 12

Table 1.2 – Mean minimum anesthetic concentration of isoflurane (MACIso) in reptiles...... 17

Table 1.3 – Minimum anesthetic concentration of nitrous oxide (MACN2O) in various species...... 21

Table 1.4 – Pharmacokinetic parameters of midazolam in various animal species...... 24

Table 2.1 – Scoring system for the evaluation of the level of sedation associated with an injection of midazolam in ball pythons...... 41

Table 2.2 – Scoring system for the evaluation of the level of muscle relaxation associated with midazolam in ball pythons (scale of 0-2)...... 41

Table 3.1 – Blood gases, hematocrit (Hct), electrolytes and glucose concentration in blood collected by cardiocentesis from nine ball pythons immediately after intubation (T0) and following minimum anesthetic concentration determination (TMAC) during three anesthetic treatments: treatment SAL-O2, administration of saline (0.2 mL/kg) intramuscularly (IM) and anesthesia induced and maintained with isoflurane in oxygen; treatment MID-O2, administration of midazolam (1 mg/kg) IM and anesthesia with isoflurane in oxygen; and treatment SAL-N2O, administration of saline IM and anesthesia induced and maintained with isoflurane in 50:50 nitrous oxide and oxygen...... 60

Table 4.1 – Mean ± SD pharmacokinetic parameters of midazolam and its major metabolite 1-hydroxymidazolam following intracardiac (IC) and intramuscular (IM) administration of 1 mg/kg midazolam in ball pythons (Python regius) at 30.4 ± 0.8°C. . 78

xii

LIST OF FIGURES

Figure 1.1 – General schematic for cardiac blood flow in snakes...... 6

Figure 1.2 – Relationship between the minimum alveolar concentration (MAC) in humans and the partition coefficient of anesthetic gases in olive oil (log scale). As the partition coefficient of an anesthetic gas increases (i.e., the lipid solubility increases), the MAC decreases and therefore the potency (1/MAC) increases. Adapted from Bovill (2008)...... 13

Figure 1.3 – Dose-response relationship between the concentration of an anesthetic gas and the percent of the population reacting to a supramaximal stimulus. The minimum alveolar concentration (MAC) and the EC95 are the concentrations at which 50% and 95% of the population do not react to a supramaximal stimulus, respectively. The EC95 can typically be estimated as MAC + (2 × standard deviation)...... 14

Figure 2.1 – Mean ± SEM sedation scores (scale 0-6) and muscle relaxation scores (scale 0-2) over time by treatment in ten ball pythons sedated with 1 mg/kg (black circles, full line) or 2 mg/kg (white circles, dotted line) intramuscular of midazolam. No statistically significant differences between treatments were detected at any time point for either score but differences over time were observed...... 45

Figure 2.2 – Mean ± SEM heart and respiratory rates over time by treatment in ten ball pythons sedated with 1 mg/kg (black circles, full line) or 2 mg/kg (white circles, dotted line) intramuscular of midazolam. No statistically significant differences between treatments were detected at any time point for either parameter...... 46

Figure 2.3 – Mean ± SEM sedation score (scale 0-6), muscle relaxation score (scale 0- 2), heart rate, and respiratory rate over time in ten ball pythons sedated with 1 mg/kg intramuscular (IM) of midazolam and reversed with 0.08 mg/kg IM of flumazenil administered 60 min later (as indicated by the vertical dotted line)...... 47

Figure 4.1 – Mean ± SD plasma concentration-time profile of midazolam following the administration of 1 mg/kg intracardiac in ball pythons (Python regius)...... 74

Figure 4.2 – Mean ± SD plasma concentration-time profile of midazolam following the administration of 1 mg/kg intramuscular in ball pythons (Python regius). The lower SD of the plasma concentration 48 h after the injection cannot be illustrated on a semi- logarithmic scale because it extends below 0 (mean ± SD = 95.49 ± 107.86 ng/mL). .. 75

Figure 4.3 – Mean ± SD plasma concentration-time profile of 1-hydroxymidazolam following the administration of 1 mg/kg intracardiac midazolam in ball pythons (Python regius)...... 76

xiii

Figure 4.4 – Mean ± SD plasma concentration-time profile of 1-hydroxymidazolam following the administration of 1 mg/kg intramuscular midazolam in ball pythons (Python regius)...... 77

xiv

LIST OF SYMBOLS, ABBREVIATIONS, OR NOMENCLATURE

5-HT 5-hydroxytryptamine λ Partition coefficient λb:g Blood:gas partition coefficient λo:g Olive oil:gas partition coefficient λz Slope of the terminal phase of the concentration-time curve AGC Automatic gain control AMPA α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid AUC Area under the concentration-time curve AUC0-∞ AUC extrapolated to infinity AUC0-t AUC until the last measurable time point AUMC0-∞ Area under the first moment curve BP Boiling point BW Body weight C0 Plasma concentration at time 0 CI Confidence interval Cl Clearance Cmax Maximum plasma concentration CO2 Carbon dioxide Ctlast Plasma concentration at the last measurable time point CRI Constant-rate infusion CV Coefficient of variation CYP3A Cytochrome P450 3A Ebody Body extraction ratio EC95 Effective concentration in 95% of the population F Bioavailability FE′Iso End-tidal concentration of isoflurane FIIso Inhaled concentration of isoflurane FIO2 Fraction of inspired oxygen FWHM Full width at half maximum GABA γ-aminobutyric acid HCD Higher-energy collisional dissociation HPLC-MS/MS High-performance liquid chromatography tandem-mass spectrometry IC Intracardiac IM Intramuscular IN Intranasal IV Intravenous LOQ Limit of quantification L-R Left-to-right

MAC Minimum alveolar concentration or minimum anesthetic concentration

MACIso MAC of isoflurane MACHal MAC of halothane MACN2O MAC of nitrous oxide MACSevo MAC of sevoflurane MAT Mean absorption time MID-O2 Premedication with midazolam and induction with isoflurane-O2 MRT Mean residence time MTBE Methyl tertiary-butyl ether MW Molecular weight m/z Mass-to-charge ratio N2O Nitrous oxide NaCl Sodium chloride NMDA N-Methyl-D-aspartic acid O2 Oxygen PB Partial pressure of the phase dissolved in blood PCO2 Partial pressure of carbon dioxide PG Partial pressure of the gaseous phase PO2 Partial pressure of oxygen Qp Pulmonary blood flow Qs Systemic blood flow Qshunt Shunted blood flow R-L Right-to-left SAL-N2O Premedication with saline and induction with isoflurane-N2O SAL-O2 Premedication with saline and induction with isoflurane-O2 SC Subcutaneous SD Standard deviation SEM Standard error of the mean SVL Snout-to-vent length SVP Saturated vapor pressure t1/2 Terminal half-life tmax Time to maximum plasma concentration Vss Volume of distribution in steady state v/v Volume-to-volume

xvi

LIST OF APPENDICES

APPENDIX A: Minimum anesthetic concentration of isoflurane in various species .... 113

Table A.1 – Mean minimum alveolar concentration of isoflurane (MACIso) in mammals...... 113

Table A.2 – Mean minimum anesthetic concentration of isoflurane (MACIso) in birds...... 116

APPENDIX B: Descriptive data of the animal research collection ...... 117

Table B.1 – Weights of ball pythons used in the current project over the year 2018...... 117

Table B.2 – Usage of ball pythons for the different trials of the current project...... 118

APPENDIX C: Descriptive data from chapter 2 ...... 119

Table C.1 – Heart rates (beats/min) of ball pythons receiving a low (L, 1 mg/kg) or a high (H, 2 mg/kg) dosage of midazolam intramuscularly over time...... 119

Table C.2 – Respiratory rates (breaths/min) of ball pythons receiving a low (L, 1 mg/kg) or a high (H, 2 mg/kg) dosage of midazolam intramuscularly over time...... 120

Table C.3 – Superficial body temperatures (°C) of ball pythons receiving a low (L, 1 mg/kg) or a high (H, 2 mg/kg) dosage of midazolam intramuscularly over time...... 121

Table C.4 – Sedation scores (out of 6; see Table 2.1 for description of the scoring system) of ball pythons receiving a low (L, 1 mg/kg) or a high (H, 2 mg/kg) dosage of midazolam intramuscularly over time...... 122

Table C.5 – Muscle relaxation scores (out of 2; see Table 2.2 for description of the scoring system) of ball pythons receiving a low (L, 1 mg/kg) or a high (H, 2 mg/kg) dosage of midazolam intramuscularly over time...... 123

Table C.6 – Temperatures (T, °C), heart rates (P, beats/min), and respiratory rates (R, breaths/min) of ball pythons sedated with 1 mg/kg intramuscular (IM) midazolam and reversed with 0.08 mg/kg IM flumazenil 60 min later (dotted line)...... 124

Table C.7 – Sedation scores (S, out of 6) and muscle relaxation scores (M, out of 2) of ball pythons sedated with 1 mg/kg intramuscular (IM) midazolam and reversed with 0.08 mg/kg IM flumazenil 60 min later (dotted line). See Tables 2.1 and 2.2 for the description of the scoring systems...... 125

xvii

APPENDIX D: Descriptive data from chapter 3 ...... 126

Figure D.1 – Presence or absence of a purposeful movement in response to an electrical stimulus in nine ball pythons anesthetized with isoflurane in 100% oxygen with 0.2 mL/kg saline intramuscularly (IM) (“Isoflurane”), isoflurane in 100% oxygen with 1 mg/kg IM midazolam (“Midazolam”), or isoflurane in 50% nitrous oxide and 50% oxygen with 0.2 mL/kg saline IM (“Nitrous oxide”)...... 126

Table D.2 – Time from induction to intubation (TInt), equilibration time following changes in isoflurane concentration, and time from induction to determination of the minimum anesthetic concentration of isoflurane (TMAC) in ball pythons...... 128

Table D.3 – Mean ± SD values of blood gases collected from the heart of nine ball pythons immediately after intubation (Time “0”) and immediately following MAC determination (Time “MAC”) when anesthetized with isoflurane in 100% oxygen with 0.2 mL/kg saline intramuscularly (IM) (“Iso”), isoflurane in 100% oxygen with 1 mg/kg IM midazolam (“Mid”), or isoflurane in 50% nitrous oxide and 50% oxygen with 0.2 mL/kg saline IM (“N2O”)...... 129

APPENDIX E: Validation of the methods to quantify the plasma concentration of midazolam and 1-hydroxymidazolam in the ball python and descriptive data from chapter 4 ...... 132

Figure E.1 – Tandem-mass spectrometry spectrum for midazolam in ball python plasma with specific quantifying, qualifying, and confirming ions...... 132

Figure E.2 – Tandem-mass spectrometry spectrum for midazolam-D4 in ball python plasma with specified quantifying, qualifying, and confirming ions...... 132

Figure E.3 – Tandem-mass spectrometry spectrum for 1-hydroxymidazolam in ball python plasma with specific quantifying, qualifying, and confirming ions...... 133

Figure E.4 – Tandem-mass spectrometry spectrum for 1-hydroxymidazolam-D4 in ball python plasma with specific quantifying, qualifying, and confirming ions...... 133

Table E.1 – Validation data for the quantification of midazolam in ball python plasma using high performance liquid chromatography tandem-mass spectrometry...... 134

xviii

Table E.2 – Validation data for the quantification of 1-hydroxymidazolam in ball python plasma using high performance liquid chromatography tandem-mass spectrometry...... 134

Table E.3 – Plasma concentrations of midazolam over time in 6 ball pythons following intracardiac administration of 1 mg/kg of midazolam at 30.4 ± 0.8°C...... 135

Table E.4 – Plasma concentrations of midazolam over time in 6 ball pythons following intramuscular administration of 1 mg/kg of midazolam at 30.4 ± 0.8°C...... 136

Table E.5 – Plasma concentrations of 1-hydroxymidazolam over time in 6 ball pythons following intracardiac administration of 1 mg/kg of midazolam at 30.4 ± 0.8°C...... 137

Table E.6 – Plasma concentrations of 1-hydroxymidazolam over time in 6 ball pythons following intramuscular administration of 1 mg/kg of midazolam at 30.4 ± 0.8°C...... 138

1 Literature review 1.1 Ophidian cardiopulmonary anatomy and physiology

Snakes are unique among reptiles as they are strictly carnivorous, elongated animals devoid of limbs. Modern snakes form a suborder named Serpentes within the order Squamata. The term “Ophidia” refers to the clade encompassing all extinct and extant species of snakes, including the Serpentes suborder. Snakes evolved from lizards, developing countless adaptations that led to their unique traits. Many of these anatomical and physiological differences can have a major impact on the sedation and anesthesia of snakes. The most clinically relevant cardiopulmonary adaptations of ophidian species are described below.

1.1.1 Ophidian respiratory system

The respiratory system of reptiles is in many ways similar to that of other classes such as mammals and birds. However, its anatomy and physiology are adapted to their lifestyle and metabolic requirements. During inspiration, air passes through the external nares, followed by the nasal sinuses, the internal nares (also named choana), the glottis, the larynx, the trachea, and, finally, the lungs. The lungs are the main site for gas exchanges in most reptilian species, except for a few select species of sea snakes in which the skin and some mucosae, such as the gular and cloacal mucosae, have important gas exchange properties (Graham, 1974; Wang, Smits, & Burggren, 1998). Reptiles generally have metabolic rates 12 to 20 times lower than similarly sized mammalian and avian species (Nagy, 2005), which can greatly decrease oxygen (O2) requirements.

1.1.1.1 Respiratory anatomy

Upper respiratory tract

The upper respiratory tract of snakes is composed of the external nares, nasal sinuses, internal nares, glottis, and larynx, which is itself composed of two vertical arytenoid cartilages, an azygous cartilage dorsally, and a cricoid cartilage ventrally. The central opening is called the glottis. Unlike mammals, reptiles do not have an epiglottis. When the oral cavity is closed, the glottis rests in the internal nares to allow the passage of air from the external nares to the lungs. The glottis is supported by the larynx, which is rostrally located and highly mobile in snakes. During deglutition, the geniotrachealis muscle allows for rostral protrusion beyond the rostral tip of the two mandibles or lateral protrusion of the glottis for uninterrupted breathing (Wallach, 1998). Opening and closing of the glottis are performed by the dilatator laryngis and sphincter laryngis muscles, respectively (Wallach, 1998).

Trachea

The trachea of snakes is composed of incomplete, C-shaped, cartilaginous rings separated by a thin membrane of connective tissue (Driggers, 2000; Wallach, 1998). Most snakes have very few complete cartilaginous rings cranially forming the glottal tube, although this structure can be formed by up to 30 complete cartilaginous rings in a few species (Wallach, 1998). The trachea extends from the larynx to the right lung, located caudal to the heart. The tracheal length in relation to the snout-to-vent length (SVL) of the snake is highly variable among species. For instance, the minimum recorded tracheal length is 14% SVL in thread snakes of the genus Leptotyphlops and the maximum is 57% SVL in sea snakes of the genus Hydrophis, although the average range is 25-35% SVL in most ophidian species. The tracheal and cardiac lungs are thin vascularized tissues capable of blood-gas exchanges that expand the trachea cranially and adjacent to the heart, respectively. However, snakes of the family Pythonidae do not possess a tracheal or cardiac lung (Wallach, 1998).

Lungs

All species of snakes possess a right functional lung located just caudal to the heart and dorsal to the liver, which extends caudally to the gallbladder in most terrestrial species and as far as the cloaca in some sea snakes (Driggers, 2000; Jacobson, 2007; Wallach, 1998). Most snakes do not possess a left lung, but some families such as Pythonidae and Boidae possess a vestigial or small functional left lung (Driggers, 2000; Wallach, 1998). This left lung can represent 51% to 82% of the length of the right lung in pythons (Wallach, 1998). The length of the right lung varies from 8% SVL in some thread snakes (Ramphotyphlops spp.) to 82% SVL in some colubrid snakes (Dasypeltis spp.), with a range of 25-60% SVL in most species (Wallach, 1998). Although the volume of ophidian lungs is usually larger than that of similarly-sized mammals, the respiratory surface area is on average only 10% to 20% that of mammals (Schumacher, 2003).

The ophidian lung is typically divided into two poorly demarcated sections: a cranial, thick-walled, and vascularized portion that is lined by a respiratory epithelium, named the vascular lung, and a caudal, thin-walled, and avascular portion that is lined by a non-respiratory epithelium, named the saccular lung (Driggers, 2000; Schumacher, 2003; Wallach, 1998). The transition zone between these sections is named the semisaccular lung (Wallach, 1998). Aquatic snakes have a larger proportion of respiratory epithelium than terrestrial species to increase the efficiency of gas exchanges during extended periods of apnea (Schumacher, 2003).

Snakes possess unicameral lungs, meaning they are composed of a single saccular chamber. However, some species within the family Typhlopidae also possess a paucicameral or multicameral cranial portion (Wallach, 1998). Paucicameral lungs are divided into a few simple chambers by thin septa, while multicameral lungs are more evolved, multi-chambered lungs connected by bronchi. There are numerous advantages to having a unicameral lung over a paucicameral or multicameral lung. The unicameral 2

lung is much more deformable, light-weight, and easier to inflate and deflate than its more complex, multicameral counterpart. These characteristics allow greater freedom of movement while decreasing the effort required for breathing (Perry, 1998). Furthermore, the elongated shape of the ophidian lung increases the vascularized surface area in comparison to the lung volume. In multicameral species, the trachea and bronchi are multifocally interrupted by foramina that communicate with the various chambers. When present, air typically access the left lung through a foramen within the tracheal wall, although boas possess a true tracheal bifurcation into two mainstem bronchi (Wallach, 1998).

The parenchyma is faveolar in the vast majority of snake species (Wallach, 1998). Faveoli are the gas exchange structures equivalent to the mammalian alveoli. An edicular parenchyma is present in some chelonian, crocodilian, and lizard species, which is formed by wide shallow terminal air spaces. Similarly, a few species of chelonians and lizards possess a trabecular, pulmonary parenchyma, which is characterized by the fusion of trabeculae with the inner pulmonary wall and the absence of septa (Perry, 1998). The ophidian pulmonary parenchyma transitions from a cranial, vascularized, faveolar lung into a caudal, membranous, saccular lung. Edicular and trabecular parenchyma are therefore commonly present in the semisaccular lung (Wallach, 1998).

Like alveoli, faveoli are lined by a squamous epithelium (type I cells) and a less prevalent cuboidal epithelium (type II cells), while the saccular lung is lined by a columnar to squamous epithelium (Jacobson, 2007). Ciliated epithelial cells alongside non-ciliated secretory epithelial and basal cells line the luminal and septal surfaces of the lungs and airways (Jacobson, 2007). Snakes lack a diaphragm and cannot cough, leaving the mucociliary apparatus as the only respiratory clearance mechanism (Wang et al., 1998).

Inspiration occurs by the expansion of the ribs through the contraction of muscles such as the intercostal, pectoral, and abdominal muscles (Driggers, 2000; Schumacher, 2003). Although expiration has been occasionally reported to be passive and a result of muscle relaxation (Driggers, 2000), the normal breathing cycle of snakes includes an active expiration phase characterized by increased intrapulmonary pressure and electrical activity of expiratory muscles, as demonstrated in the common garter snake (Thamnophis sirtalis) (Furilla & Bartlett, 1989).

1.1.1.2 Respiratory physiology

Mechanisms and patterns of respiration

Oxygen and carbon dioxide (CO2) gas exchanges occur primarily in the lungs, except in some species of aquatic snakes in which the skin, pharyngeal mucosa, and cloacal mucosa are capable of blood-gas exchanges (Graham, 1974; Wang et al., 1998). In most of these species, cutaneous gas exchange is mainly used for the elimination of CO2 rather than the uptake of O2 (Feder & Burggren, 1985a, 1985b). 3

The breathing cycle of reptiles begins with expiration, followed by inspiration and a non-ventilatory period (Gans & Clark, 1978). Two main patterns have been described: single breaths followed by non-ventilatory periods, and multiple consecutive breaths followed by a prolonged period of apnea. The first pattern is most commonly seen in terrestrial species, while the second is most common in aquatic species (Wang et al., 1998). When increased ventilation is required, most reptiles decrease the duration of non-ventilatory periods to increase the respiratory rate (Glass, Boutilier, & Heisler, 1985; Glass & Johansen, 1976).

Factors influencing respiration

Partial pressures of O2 (PO2) and CO2 (PCO2), blood pH, and body temperature are the most important factors that stimulate or inhibit ventilation in reptiles. As such, both hypercapnia and hypoxia can affect the ventilation of snakes (Wang et al., 1998). However, the response to either of these conditions is highly variable among ophidian species and is characterized by either variation in the tidal volume or the respiratory rate, both of which influence the minute volume (Milsom, 1990, 1995).

Carbon dioxide is detected by chemoreceptors within the lungs, upper airways, and possibly arteries. Central chemoreceptors within the brain also indirectly monitor CO2 levels through the arterial pH, such as in mammals (Milsom, 1995). Hypercapnia has been shown to increase ventilation through a higher tidal volume in snakes (Milsom, 1990, 1995).

Oxygen is primarily detected by arterial chemoreceptors. In the presence of hypoxia, ventilation is increased by decreasing the duration of non-ventilatory periods, which results in an increased respiratory rate (Milsom, 1990, 1995). However, reptiles, especially aquatic species, are highly resistant to hypoxia and can sustain prolonged periods of apnea. Some species can even convert to anaerobic metabolism in hypoxic environments (Schumacher, 2003). The lung capacity of snakes can be substantially larger than O2 requirements. For instance, in the ball python (Python regius), the O2 uptake capacity of the lungs is 12 times greater than the O2 requirements (Starck et al., 2012). The effects of hyperoxia on ventilation have been far less studied in reptiles than hypoxia. Inhalation of markedly high concentrations of O2 commonly leads to decreased tidal volume and respiratory rate in most reptiles (Benchetrit, Armand, & Dejours, 1977; Glass, Burggren, & Johansen, 1978; Glass & Johansen, 1976). This response has been hypothesized to contribute to the apnea commonly observed during the anesthesia and recovery of reptiles (C. I. Mosley & Mosley, 2015).

Snakes, such as with all reptiles, are ectothermic animals, meaning that they rely on external heat sources to maintain their internal body temperature. Most snakes show some level of poikilothermy and, therefore, their metabolic rate is greatly influenced by body temperature. An increased body temperature typically increases aerobic metabolism, leading to increased ventilation. This response is thought to be the result of an increased hypoxia threshold (Dupré, Hicks, & Wood, 1989; Glass, Boutilier, & Heisler, 1983; Jackson, 1971, 1973). Conversely, lower body temperatures have been 4

hypothesized to decrease the hypoxic threshold in turtles (Jackson, 1973) and iguanas (Dupré et al., 1989). However, most of these studies did not consider that the solubility of O2 is inversely proportional to temperature. As such, less O2 can be dissolved in the blood at higher temperatures, which could potentially influence ventilation.

Mechanoreceptors detecting stretching of the pulmonary walls provide feedback regarding the tidal volume of the lungs. Stimulation of these mechanoreceptors serves to inhibit inspiration and stimulate expiration (Wang et al., 1998). Like mammals, the sensitivity of these mechanoreceptors can be decreased or even suppressed by high levels of CO2 in snakes (Wang et al., 1998).

1.1.2 Ophidian cardiovascular system

1.1.2.1 Cardiovascular anatomy and blood flow

The ophidian heart is grossly elongated, narrow, circular, and lies within a pericardial sac. Unlike most reptilian species, snakes do not possess a gubernaculum cordis that attaches the apex of the heart to the pericardium. The heart of terrestrial and arboreal snake species is located more cranially than their aquatic counterparts, which could decrease the hydrostatic blood pressure cranial to the heart when the head is lifted (Farrell, Gamperl, & Francis, 1998). The heart mass compared to the body weight (BW) is variable in reptiles, ranging from 0.16% of BW in the slowworm (Anguis fragilis) (Poupa & Lindström, 1983) to 0.45% BW in the tree snake (Dispholidus typus) (Farrell et al., 1998). This variation is suspected to be due to species-specific differences in gravitational pressure and physical activity.

The reptilian heart is commonly described with three chambers: two atria and a ventricle. However, most archosaurs (crocodilians and birds) possess a four-chambered heart composed of two atria and two ventricles separated by a complete interventricular septum. In snakes, venous blood is transported to the heart cranially via the left and right precaval veins and caudally via the post-caval vein, which fuse at the level of the heart to form the sinus venosus. This cavity openly communicates with the right atrium and its thin wall contains cardiac muscles that act as the pacemaker of the heart. The right atrium is markedly larger and more muscular than the left atrium in all species of snakes, but this difference is more pronounced in species with a large systemic and pulmonary pressure difference, such as pythons (Jensen, Moorman, & Wang, 2014). The left atrium receives blood from the lungs through the anterior and posterior pulmonary veins. Both atria are separated from the ventricle by an atrioventricular valve and a short elongation of the atrioventricular wall into the ventricle, called the atrioventricular funnel (Farrell et al., 1998).

Figure 1.1 – General schematic for cardiac blood flow in snakes. Deoxygenated venous blood (blue arrows) is ejected from the right atrium into the cavum venosum, followed by the cavum pulmonale and into the pulmonary trunk that divides into the right and left pulmonary arteries in species with two lungs. Initially in the left atrium, oxygenated blood (red arrows) is ejected the into the cavum arteriosum, followed by the cavum venosum and into the right and left aortas.

Although three different ventricular regions are described in snakes, the physical separation between these chambers is incomplete. These chambers are named the cavum arteriosum, cavum venosum, and cavum pulmonale. Briefly, during the ventricular diastole, contraction of the right atrium results in the output of deoxygenated blood into the cavum venosum, which openly communicates with the cavum pulmonale (Figure 1.1). At the same moment, oxygenated blood is expelled from the left atrium into the cavum arteriosum. During the ventricular systole, deoxygenated blood is expelled 6

from the cavum pulmonale to the pulmonary circulation via the pulmonary trunk, while oxygenated blood is transferred to the cavum venosum to be expelled into the systemic circulation via the two aortas. The pulmonary trunk bifurcates into a left and a right pulmonary arteries in species that possess two lungs, such as pythons. (Farrell et al., 1998; Starck, 2009). The cava venosum and arteriosum are separated by an incomplete ridge, called the vertical septum. During the ventricular diastole, communication between these two chambers is blocked by the atrioventricular valves. In turn, the cava venosum and pulmonale are divided by an incomplete muscular ridge called the horizontal septum, which separates the two cavities during the ventricular systole (Jensen, Nielsen, et al., 2010). Despite this complex anatomy, ventricular blood flow is highly efficient in snakes and blood mostly follows the path described above (Farrell et al., 1998). Nonetheless, there is a potential for mixing of oxygenated and deoxygenated blood as well as left-to-right (L-R) or right-to-left (R-L) blood shunting, which has been favored by evolution in some species.

The anatomy of the python heart is slightly different from other snakes. Similar to varanid lizards, the separation of oxygenated and deoxygenated blood within the ventricle is characterized by a marked pressure difference (Wang, Altimiras, & Axelsson, 2002). The muscular ridge between the cava venosum and pulmonale is supported by pads of connective tissue to prevent blood shunting during the ventricular systole. Furthermore, the cavum venosum is particularly small in pythons, representing approximately 10% of the cavum arteriosum. The latter is by far the largest chamber, representing approximately 75% of the ventricular mass (Jensen, Nyengaard, Pedersen, & Wang, 2010). These anatomical adaptations allow for a systemic blood pressure nearly 7 times higher than the pulmonary blood pressure (Wang, Altimiras, Klein, & Axelsson, 2003). While some studies suggest that 50% of the venous blood can by-pass the lungs in ball pythons (Starck, 2009), anatomical studies of the python heart suggest that only up to 20% of this blood could, in theory, perform a R-L shunt (Jensen, Nyengaard, et al., 2010).

Interestingly, it has been shown that the cardiac volume of Burmese pythons (Python molurus) increases by 40% over the 48 hours following ingestion of prey (Andersen, Rourke, Caiozzo, Bennett, & Hicks, 2005). It was hypothesized that cardiomyocellular hypertrophy is at least partially stimulated by the post-prandial increase in plasma fatty acids (Riquelme et al., 2011). However, this cardiac hypertrophy was later shown to be facultative in pythons (Jensen, Larsen, Nielsen, Simonsen, & Wang, 2011), and a reduction in blood oxygen levels is more likely to be responsible for this cardiac hypertrophy (Slay, Enok, Hicks, & Wang, 2014).

1.1.2.2 Cardiovascular shunting

The word shunt is defined as the act of diverting a substance or an object from one side to another. Because of the communication between the numerous chambers of the cardiac ventricle, L-R and R-L blood shunting occur in snakes and other reptilian species, except for crocodilians. A L-R shunt is defined as the passage of oxygenated blood from the left atrium back into the pulmonary circulation through the cavum 7

pulmonale. A R-L shunt is defined as the passage of deoxygenated blood from the right atrium into the systemic arterial circulation through the cavum venosum, therefore bypassing the pulmonary circulation. Intracardiac (IC) shunting can be unidirectional (L- R or R-L) or bidirectional (L-R and R-L). Quantification of a cardiac shunt, described as the shunted blood flow (Qshunt), is measured as the net difference between the pulmonary blood flow (Qp) and systemic blood flow (Qs) (Equation 1.1; Hicks, 1998; Hicks & Comeau, 1994). These values can also be expressed as a fraction of the systemic blood flow for L-R shunts (Equation 1.2) and as a fraction of the pulmonary blood flow in R-L shunts (Equation 1.3; Hicks, 1998; Hicks & Comeau, 1994). However, when a bidirectional shunt occurs, the shunted blood flow measurement becomes inaccurate. Instead, the shunt fraction can be expressed by comparing the O2 content of the pulmonary vein and artery, the systemic arteries, and the right atrium, as described elsewhere (Hicks, 1998).

Equation 1.1: 푄푠ℎ푢푛푡 = 푄푝 − 푄푠

Equation 1.2: 푄푠ℎ푢푛푡(%) = (푄푠 − 푄푝) ÷ 푄푠

Equation 1.3: 푄푠ℎ푢푛푡(%) = (푄푝 − 푄푠) ÷ 푄푝

Intrapulmonary shunts have also been described in snakes, which reduce the efficiency of pulmonary gas exchange (Wang et al., 1998). In sea snakes, lung compression during diving increases intrapulmonary shunt, which is thought to be a protective mechanism against the formation of nitrogen bubbles during decompression and against loss of O2 to the water by decreasing the O2 diffusion gradient (Seymour, 1978).

Mechanisms of cardiac shunting

There are four recognized mechanisms by which a cardiac shunt can occur: (1) a difference in hydrostatic pressure between two communicating chambers (pressure difference shunt), (2) a difference in kinetic energy between two communicating chambers (kinetic shunt), (3) mixing of venous and arterial blood within an open interface (open interface shunt), and (4) arterial and venous blood alternately filling a common cavity (washout shunt) (Heisler & Glass, 1985; Hicks, 1998). In snakes, pressure difference shunts and washout shunts are generally thought to be the most common (Heisler & Glass, 1985).

Hydrostatic differences between the ventricle and cardiac vessels dictates the direction of the blood flow (Wang & Hicks, 1996). For instance, a lower diastolic pressure within the pulmonary arteries compared to the systemic circulation leads to the ejection of ventricular blood within the pulmonary trunk prior to the aortas during systole. The timing between the ejection of blood into the systemic and pulmonary circulation also influences the total volume in each circuit and therefore affects the net shunted blood flow (Wang & Hicks, 1996). Finally, the difference in vascular resistance between the systemic and pulmonary circulation also influences the direction of the blood flow toward the path of least vascular resistance (Hicks & Comeau, 1994; Wang & Hicks, 8

1996). A recent meta-analysis suggested that vascular distensibility had little influence on the direction of intracardiac shunts (Filogonio, Costa Leite, & Wang, 2017), although conclusions from this study are debated (Hillman, Hedrick, & Kohl, 2017).

Washout shunts occur when the residual oxygenated blood within the cavum venosum is pushed into the cavum pulmonale, and then into the pulmonary circulation by the deoxygenated blood from the right atrial systole. Alternatively, the residual deoxygenated blood within the cavum venosum can be pushed into systemic circulation by the oxygenated blood ejected from the left atrium during atrial systole (Heisler & Glass, 1985). Because of the small size of the cavum venosum of pythons, washout shunts are generally considered very small (Jensen, Nielsen, et al., 2010).

Overall, the intensity and direction of a shunt is the result of the anatomy and volume of the heart, as well as any factors that influence vascular resistance, myocardial contractility, and heart rate (Hicks, 1998). Most importantly, the ratio of the systemic and pulmonary vascular resistance has the strongest effect on the direction of the cardiac shunt (Overgaard, Stecyk, Farrell, & Wang, 2002). In turtles, it has been shown that cholinergic stimulation increases pulmonary vascular resistance resulting in decreased pulmonary blood flow and a net R-L shunt (Hicks, 1994). The use of atropine, an anticholinergic drug, confirmed this finding by eliminating R-L cardiac shunting in red-footed tortoises (Chelonoidis carbonaria) (Greunz et al., 2018). Alternatively, adrenergic stimulation decreases pulmonary vascular resistance, resulting in increased pulmonary blood flow and a net L-R shunt (Hicks, 1994; Overgaard et al., 2002). For instance, recovery from inhalation anesthesia has been shown to be significantly shorter with the use of epinephrine in common snapping turtles (Chelydra serpentina) and American alligators (Alligator mississippiensis), which is thought to result from an increased pulmonary blood flow and cardiac output (Gatson, Goe, Granone, & Wellehan, 2017; Goe et al., 2016).

Factors that influence cardiac shunting

(1) Ventilation: It has been demonstrated in the red-eared slider (Trachemys scripta elegans) that a strong cholinergic stimulation occurs during periods of apnea, leading to an 80% decrease in heart rate, 150% increase in pulmonary vascular resistance, and an overall 80% decrease in the systemic blood flow, which consequently leads to a R-L shunt (Shelton & Burggren, 1976). Similar findings have been reported during the non-ventilatory period in squamates (Heisler & Glass, 1985). However, a L-R shunt can occur concurrently during apnea, just as a R-L shunt can occur during ventilation (Heisler & Glass, 1985).

(2) Activity: Increased physical activity usually leads to an increased O2 demand, necessitating both a higher cardiac output and an increased blood O2 content. Studies have demonstrated that increased physical activity in squamates and testudines can lessen the R-L shunt and even increase the L-R shunt (G. S. Mitchell, Gleeson, & Bennett, 1981; West, Butler, & Bevan, 1992).

(3) Thermoregulation: Body temperature significantly influences hemodynamics and cardiac shunting. For instance, higher temperatures typically lead to peripheral vasodilatation, which decreases systemic vascular resistance and therefore promotes R-L shunts (Baker & White, 1970). (4) Other factors: Although not clearly demonstrated, other factors such as hypoxia, hypercapnia, and digestion could potentially impact blood shunting. Digestion, for instance, is a highly energy demanding process that increases O2 metabolism and leads to cardiac hypertrophy in Burmese pythons (Andersen et al., 2005). However, there is currently no definitive evidence to suggest that these changes lead to cardiac shunting. Consequences of cardiac shunting Numerous hypotheses have been suggested or validated concerning the consequences of cardiac shunting in reptiles. There are many advantages to a R-L shunt: intrapulmonary O2 can be stored for prolonged periods of apnea; CO2 output can be decreased to facilitate O2 uptake; pulmonary plasma filtration can be decreased, especially in cases of compromised lymphatic drainage; pulmonary heat loss can be decreased, and cardiac energy can be preserved during periods of apnea by bypassing pulmonary circulation (Hicks, 1998). Alternatively, a L-R shunt can facilitate the elimination of CO2, lead to a better matching of the ventilation-perfusion (V/Q) ratio, and improve blood oxygenation (Hicks, 1998).

1.2 Inhalation anesthesia

1.2.1 Introduction

The use of an anesthetic gas for various procedures was first documented in 1846 when a dentist, Dr. William Morton, successfully used ether to anesthetize a man for tumor removal surgery (Bovill, 2008). However, nitrous oxide (N2O) is thought to have been used for at least two years prior to that demonstration. Shortly thereafter, in 1847, chloroform was introduced for use in humans (Bovill, 2008). Inhalation anesthesia is now widely used in human and veterinary medicine due to its convenience, low-cost, predictable side effects, excellent muscle relaxation, fast induction and recovery, and comparable potency across a vast array of species (Bertelsen, 2014; Bovill, 2008; Steffey, Mama, & Brosnan, 2015). Each anesthetic gas has unique characteristics, which are defined by physicochemical properties such as volatility, solubility, and anesthetic potency.

1.2.2 Physicochemical properties and potency of anesthetic gases

1.2.2.1 Volatility and solubility

All modern anesthetic gases are volatile liquids at ambient temperature and atmospheric pressure, except for N2O and xenon, which are found in gas form (Bovill, 2008). The concept of critical temperature, which is the maximum temperature beyond which, regardless of the pressure, a gas cannot be brought to a liquid form, is only relevant to the two latter gases. Since the critical temperatures of N2O and xenon are above typical ambient temperatures, they are commercially available in a pressurized liquid form. Vapor refers to the gaseous form of an anesthetic gas below the critical temperature. The saturated vapor pressure (SVP) represents the gas pressure when there is an equilibrium between the vapor and liquid phases in a closed environment. The SVP is directly proportional to the volatility of a gas. For instance, N2O is much more volatile than isoflurane with a SVP at 20°C of 5,200 kPa for N2O versus 33.2 kPa for isoflurane (Bovill, 2008).

The partition coefficient (λ) is used to measure the solubility of a gas. It is defined as the ratio between the concentration of a substance that is in two immiscible phases at equilibrium: a phase dissolved in blood (PB) or tissue and a gaseous phase (PG) (Equation 1.4) (Bovill, 2008). Highly soluble gases have a high partition coefficient since a large amount of gas can be absorbed by the blood and tissues before the dissolved partial pressure equilibrates with the gaseous partial pressure. This means that a larger amount of gas is required to obtain equilibrium and therefore induction and recovery times tend to be longer. A summary of the physicochemical properties of anesthetic gases is provided in Table 1.1.

푃퐵 Equation 1.4: λ푏:𝑔 = 푃퐺

Table 1.1 – Physicochemical properties of commonly used anesthetic gases.*

AGENT MOLECULE MW BP SVP λb:g λo:g (°C) at 20°C (kPa) at 37°C at 37°C

NITROUS OXIDE N2O 44.02 -88.5 5,200.0 0.47 1.4 XENON Xe 131.3 -108.1 0.114 1.8‡

HALOTHANE C2HBrClF3 197.4 50.2 32.3 2.4 224 † METHOXYFLURANE C3H4Cl2F2O 165.0 105.0 3.3 13.0 825

ISOFLURANE C3H2ClF5O 184.5 48.5 33.2 1.46 90.8

ENFLURANE C3H2ClF5O 184.5 56.5 22.9 1.9 96.5

SEVOFLURANE C4H3F7O 200.5 58.5 21.3 0.65 42

DESFLURANE C3H2F6O 168.0 23.5 88.3 0.42 18.7

*Adapted from Bovill, 2008. †Tomlin, 1965. ‡Lynch, Baum, & Tenbrinck, 2000. MW: molecular weight; BP: boiling point; SVP: saturated vapor pressure; λb:g: blood:gas partition coefficient; λo:g: olive oil:gas partition coefficient

1.2.2.2 Minimum alveolar concentration

With the development of numerous new, volatile anesthetics in the late 19th and early 20th centuries, a standardized way of comparing and quantifying the potency of anesthetic gases was needed. Clinical parameters such as breathing patterns, muscle tone, palpebral and pupillary reflexes, lacrimation, mentation, sensation, and movements were used to evaluate anesthetic depth (Guedel, 1951; Woodbridge, 1957). Introduced in 1963 by Merkel and Eger, the concept of minimum alveolar concentration (MAC) was meant to compare the equipotency of anesthetic gases (Merkel & Eger, 1963). It was initially described as the minimum anesthetic concentration in the alveolus that prevented a “gross purposeful movement” in response to a noxious stimulus in dogs (Merkel & Eger, 1963). The concept was thereafter refined to be applicable to a larger population. It was then defined as the minimum anesthetic concentration in the alveolus at which 50% of the population did not react to a noxious stimulus (Eger, Saidman, & Brandstater, 1965; Saidman & Eger, 1964). The authors also observed that there was a specific intensity of noxious stimulation beyond which the MAC did not increase. This point was called the supramaximal stimulus (Eger, Saidman, et al., 1965). Currently, MAC is defined as the minimum concentration of an anesthetic gas within the alveoli at sea level that, when in equilibrium with the blood, prevents a purposeful movement in response to a supramaximal stimulus in 50% of the population (Eger, Saidman, et al., 1965). This definition assumes that the end-tidal anesthetic gas concentration is similar to the one found in the alveoli, which is meant to reflect the concentration in the target tissue. The potency of an anesthetic gas was hence defined to be inversely proportional to the MAC (Eger, Brandstater, et al., 1965). This concept also allowed the association of an effect with a factor of MAC rather than a

concentration (e.g., 1.2 MAC, 2.0 MAC), which facilitated comparison between different species and different gases (Merkel & Eger, 1963).

The Meyer-Overton relationship, independently demonstrated in 1899 and 1901 by Hans Horst Meyer and Charles Ernest Overton, respectively, refers to the correlation between the lipid solubility of an anesthetic gas and its potency (Meyer, 1899; Overton, 1901). The partition coefficient of anesthetic gases in olive oil (λo:g) was later successfully correlated to the MAC (Figure 1.2) (Campagna, Miller, & Forman, 2003). Therefore, the MAC of an anesthetic gas can be estimated based on its λo:g, since the product of these two factors equals a constant (Equation 1.5) (Eger et al., 1999). In other words, if the solubility of an anesthetic gas “A” is higher than that of an anesthetic gas “B”, meaning that the partition coefficient in olive oil is higher in A than B (λo:g[A] > λo:g [B]), the MAC of A would be lower than B (MACA < MACB). Consequently, A would be more potent than B.

Equation 1.5: 푀퐴퐶 × λ표:𝑔 = 1.82 ± 0.56 atm

In graphical form, the dose-response relationship of an anesthetized patient subjected to a supramaximal stimulus reveals the MAC as the center of the sigmoid curve, which represents the concentration at which 50% of the population does not respond (Aranake, Mashour, & Avidan, 2013; Eger, Saidman, et al., 1965). This curve shows that the MAC is a logistic function with a very steep curve (Figure 1.3) (Aranake et al., 2013). It also illustrates the limited variability of the MAC in which the standard deviation (SD) is typically estimated at 10% in humans. It can therefore be estimated that the concentration needed to prevent a purposeful movement in response to a supramaximal stimulus in 95% of the population (EC95) is 1.2 times MAC (i.e., MAC + 2 × SD), as illustrated in Figure 1.3 (de Jong & Eger, 1975).

Aside from the response to a supramaximal stimulus, a few other parameters have been evaluated, such as loss of consciousness, amnesia, and analgesia. However, the MAC for these parameters is generally lower than the MAC in which the response to a supramaximal stimulus is suppressed (Aranake et al., 2013). Factors affecting the MAC The MAC has been shown to vary with numerous physiological, pharmacological, and pathological factors. A decreased MAC can result from the use of numerous drugs (e.g., opioids, N2O, α2-agonists, benzodiazepines, propofol, barbiturates, and ketamine), hypothermia, metabolic acidosis, hyponatremia, decreased osmolality, hypoxemia, hypercapnia, anemia, pregnancy, increased age, and some intoxications (Aranake et al., 2013; Quasha, Eger, & Tinker, 1980). Factors shown to significantly increase the MAC include hyperthermia, decreased age, acute drug exposure (e.g., amphetamines, cocaine), ephedrine, and chronic alcohol consumption (Aranake et al., 2013; Quasha et al., 1980). Gender, hyperoxia, hypocapnia, obesity, and the duration of anesthesia do not affect the MAC (Aranake et al., 2013; Quasha et al., 1980). However, not all anesthetic parameters (e.g., consciousness, amnesia, analgesia) are consistently affected by the MAC variation (Aranake et al., 2013). Determination of the MAC Different techniques have been used in veterinary and human medicine to determine the MAC. The bracketing technique is still the most commonly used technique in animals since it provides a precise estimation of the MAC (Quasha et al., 1980). The patient is initially anesthetized with the anesthetic gas alone and the trachea is intubated. The end-tidal anesthetic gas concentration is measured and is assumed to reflect the alveolar (or faveolar, edicular, or trabecular) concentration once equilibrium has been achieved. Time is provided to obtain an equilibrium between the alveoli, blood, and target tissue, typically, 15 minutes in mammals (Quasha et al., 1980). The initial gas concentration evaluated is typically aimed to be higher than the expected MAC, which is estimated based on information from previously published or pilot studies. A supramaximal stimulus, either electrical stimulation or clamping of the tail or limbs, is then applied and the presence or absence of a gross purposeful movement is noted. The gas concentration is either increased (i.e., positive response) or decreased (i.e., negative response), typically by a factor of 0.1 to 0.2. The same procedure is then applied until a 10% difference between two consecutive gas concentrations both permitted and prevented purposeful movement. With this technique, the MAC can be estimated as the midpoint between these two concentrations. Then, individual MAC values are averaged to obtain the mean MAC. However, another statistical technique uses modelling of the response through logistic regression instead (Escobar et al., 2017; Hecker et al., 2003), based on the principle that the MAC represents a 50% probability of having no movement (P) in response to a supramaximal stimulus in a population (Equation 1.6). 15

0.5 Equation 1.6: If logit(P = 0.5) = ln = ln 1 = 0 0.5

And if logit(P) = Intercept + ParameterIso × FE′Iso

Then MAC = −Intercept/ParameterIso In humans, a different technique named the up and down method that requires a larger number of subjects but that reduces the length of the procedure per patient is used. Groups of patients (typically a minimum of four per group) are anesthetized in a similar method to that used for animal patients and a 15-min period is provided to obtain an alveoli-blood-tissue equilibrium at a predetermined anesthetic concentration. A single surgical incision is commonly used as the supramaximal stimulation rather than clamping or electrical stimulation. The same procedure is applied to different groups for increasing gas concentrations and, within each group, the proportion of patients that reacted is illustrated as a function of the mean end-tidal concentration for that group. As such, MAC can be identified as the concentration at which 50% of patients did not react by using a visual line of best-fit drawn through these points (Quasha et al., 1980). Unfortunately, this technique does not yield an estimate of the variance of MAC. More advanced techniques have been described to determine the MAC and to provide an estimate of the variance or SD (Litchfield & Wilcoxon, 1949; Waud, 1972). Historically, the supramaximal stimulus used in animals has been electrical stimulation or clamping of the tail (Eger, Saidman, et al., 1965; Merkel & Eger, 1963). Surgical incision, such as used in human medicine, had been demonstrated to yield the same supramaximal effect as tail clamping or electrical stimulation in rabbits and dogs (Valverde, Morey, Hernández, & Davies, 2003). However, clamping the tail yielded MAC values 20% lower than the same procedure on dew claws in pigs (Eger et al., 1988). These findings support the importance of evaluating the best noxious stimulus providing supramaximal stimulation in any species of interest. The duration of the stimulation is also important since Eger, Saidman, et al. (1965) described that in some cases animals may fail to react after 10 sec, but react after 30 to 40 sec. Merkel and Eger (1963) described that a “gross purposeful movement” following a supramaximal stimulus indicates a positive result. Eger, Saidman, et al. (1965) further described that a purposeful muscular movement of the head or limbs represents a positive response, unlike chewing, swallowing, or coughing. Due to the large variation of reactions in different species, care must be taken to determine what is considered a purposeful movement versus a spasm or reflex. The accuracy of MAC determination also relies on the accuracy of the measurement. For instance, the end-tidal partial pressure of CO2 at various locations within the endotracheal tube is closer to the actual arterial partial pressure of CO2 than the end-tidal partial pressure of CO2 at the level of the proximal connector in rabbits, which is more diluted due to the fresh gas flow (Rich, Sullivan, & Adams, 1990).

1.2.3 Isoflurane

Isoflurane is a halogenated methyl ethyl ether and an isomer of enflurane. It was discovered in 1965 by R. C. Terrell two years after the discovery of enflurane, which was easier to synthesize and purify (Vitcha, 1971). Isoflurane is a clear, colorless, and non-flammable liquid. Physicochemical properties are reported in Table 1.1 (section 1.2.2.1).

1.2.3.1 Potency

The potency of isoflurane, as any anesthetic gas, is evaluated by the MAC. In humans, the MAC of isoflurane (MACIso) is 1.15% (Stevens et al., 1975). In animals, there have been numerous studies on various species using different protocols. An extensive, but non-exhaustive list of MACIso is provided in Table 1.2 for reptiles, Appendix A, Table A.1 for mammals, and Appendix A, Table A.2 for birds.

Table 1.2 – Mean minimum anesthetic concentration of isoflurane (MACIso) in reptiles.

SPECIES SCIENTIFIC NAME MACISO REFERENCES (%) Desert iguana Dipsosaurus 3.14 (35°C) Dohm & Brunson (1998) dorsalis 3.10 (30°C) 2.83 (20°C) Dumeril’s monitor Varanus dumerili 1.54 Bertelsen, Mosley, Crawshaw, Dyson, & Smith (2005c) Green iguana Iguana iguana 1.8 Barter, Hawkins, Brosnan, Antognini, & Pypendop (2006) Green iguana Iguana iguana 2.0 Mosley, Dyson, & Smith 2.1 (2003) Radiated rat snake Elaphe radiata 1.68 Maas & Brunson (2002) Red-footed tortoise Chelonoidis 3.2 Greunz et al. (2018) carbonaria 2.2a aFollowing the administration of 1 mg/kg IV of atropine, which eliminates intracardiac blood shunting.

1.2.3.2 Induction and recovery

The solubility of anesthetic gases in blood and tissues influence the speed of gas equilibration and therefore the rapidity of induction and recovery (Eger & Larson, 1964). 17

Consequently, with its relatively low blood solubility (Table 1.1, section 1.2.2.1), isoflurane provides rapid induction and recovery. However, it should be recognized that numerous factors may influence the rate of isoflurane uptake and consequently the induction time (e.g., ventilation rate), gas concentration, and V/Q matching (Steffey & Mama, 2007). Similarly, recovery depends on the elimination of the anesthetic gas from circulation. Some factors that influence the recovery time include the duration of anesthesia and drug metabolism (Steffey & Mama, 2007).

1.2.3.3 Pharmacodynamics

The mechanism of action of inhalational anesthetics has remained a long-time mystery. Numerous hypotheses have been developed to explain the effects, such as immobility, analgesia, amnesia, and loss of consciousness. The work of Meyer and Overton (see section 1.2.2.2) brought attention to the hydrophobic properties of anesthetic gases, but it was not until 1984 that Franks and Lieb suggested that ion channels could serve as potential targets (Franks & Lieb, 1984). It was hypothesized that stimulation of inhibitory neurotransmitter- or voltage-gated channels (e.g., potassium channels) or blockage of excitatory neurotransmitter- or voltage-gated channels (e.g., sodium channels) may be responsible for the effects of inhalation anesthesia. It was later demonstrated that the spinal cord is likely the most important target site providing immobility rather than the brain or brainstem. Rampil et al. demonstrated that decerebration of the precollicular portion of the brain or sectioning of the thoracic spinal cord of rats did not significantly eliminate the capacity of isoflurane to inhibit movement (Rampil, 1994; Rampil, Mason, & Singh, 1993). Another study demonstrated that isoflurane delivered to the whole body in goats provided a MACIso of 1.2%, while gas delivered directly to the brain provided a MACIso of 2.9% (Antognini & Schwartz, 1993). It is still unclear what receptors are bound by anesthetic gas, but the involvement of both afferent (sensory) and efferent (motor) neurons is suspected (Sonner et al., 2003). Numerous receptors have been evaluated in vitro and in vivo. Glycine and potentially 5-hydroxytryptamine 2A (5-HT2A), N-Methyl-D-aspartic acid (NMDA), and sodium channels appear to be major targets of anesthetic gases. Alternatively, neither the receptors γ-aminobutyric acid A (GABAA), opioid, α2- adrenergic, 5-HT3, acetylcholine, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), and kainate, nor potassium seem to play a major role in immobility during inhalation anesthesia (Sonner et al., 2003).

Isoflurane, like most halogenated gases, is a respiratory depressant that causes a dose-dependent alteration of the physiological response to hypoxia and hypercapnia in mammals (Fourcade et al., 1971). Consequently, associated risks of hypoxemia and respiratory acidosis are markedly increased and assisted ventilation is often necessary.

The cardiovascular effects of isoflurane are generally limited, and myocardial contractility is sustained despite the negative inotropic effect reported in numerous mammalian species (Komai & Rusy, 1987; Pagel, Kampine, Schmeling, & Warltier, 1991; Steffey & Howland, 1977; Stevens et al., 1971). Isoflurane can cause tachycardia and hypotension as a result of decreased peripheral vascular resistance (Eger, 1981; 18

Ludders, Mitchell, & Rode, 1990; Ludders, Rode, & Mitchell, 1989; Pagel et al., 1991; Steffey & Howland, 1977; Stevens et al., 1971). However, central venous pressure remains unaffected (Eger, 1981).

Isoflurane also depresses neuromuscular transmission, leading to muscle relaxation (Eger, 1981). Furthermore, isoflurane decreases O2 consumption by the central nervous system and does not directly affect intracranial blood pressure (Eger, 1981).

1.2.4 Nitrous oxide

Nitrous oxide was discovered by the chemist Joseph Priestley in 1774, and a few years later, in the 1780s, the physician Thomas Beddoes attempted to adapt this gas for medical purposes (Wright, 1999). Although mostly ignored by the medical community for decades, the dentist Dr. Horace Wells started using N2O as an anesthetic and pain relief agent in 1844 and thereafter successfully demonstrated the various properties of N2O with the help of Dr. E. E. Marcy in 1847 (Wright, 1999). Since then, N2O, unlike other early anesthetic gases such as ether and chloroform, has been used as a sole or supplemental agent to provide anesthesia or analgesia, despite the discovery of more potent anesthetic gases.

The physicochemical properties of N2O are described in Table 1.1, section 1.2.2.1. The critical temperature of N2O is 36.4°C and consequently, it is inhaled as a vapor that arises from a compressed liquid form. At the body temperature of most mammals and birds, N2O is exhaled as a gas (Bovill, 2008). However, in most ectothermic animals, N2O remains in vapor form since the body temperature is typically lower than 36.4°C. The effects of body temperature on the kinetics of N2O remain unknown. Along with desflurane, the blood solubility of N2O is among the lowest of modern anesthetic gases, meaning equilibrium is quickly reached between the alveoli, blood, and tissues. Nitrous oxide is not metabolized by the body, although biotransformation by the intestinal bacterial flora occurs. It is eliminated unchanged during expiration (Bovill, 2008).

1.2.4.1 Potency, induction, and recovery

Although the analgesic potency of N2O is relatively high (e.g., 20% N2O is as potent as 15 mg of morphine administered subcutaneously [SC] in humans [Bovill, 2008]), its anesthetic potency is very low. According to the Meyer-Overton relationship described in section 1.2.2.2, it is the least potent of modern anesthetic gases (Figure 1.2). In pharmacology, two drugs used in conjunction may have synergistic, antagonistic, or additive effects. An additive effect has been shown to exist between N2O and halogenated gases, while an antagonistic effect has been demonstrated between N2O and cyclopropane (DiFazio, Brown, Ball, Heckel, & Kennedy, 1972). These findings led to the theory of the simple linear additivity relationship between N2O and halogenated gases, stating that the addition of N2O causes a decrease in anesthetic requirements that is directly proportional to the provided concentration of 19

N2O (DiFazio et al., 1972). This relationship can be illustrated by Equation 1.7, where MACA (%) and MACB (%) are the MAC of gases A and B when used alone, respectively; FE′a (%) is a predefined end-tidal concentration of gas A, and MACb (%) is the MAC of gas B when used in combination with gas A at the predefined FE′a. By rearranging this equation, the MACA can be calculated, such as is illustrated in Equation 1.8. Since the MAC of N2O (MACN2O) is typically higher than 100%, the theory of simple linear additivity has been used frequently to determine the MACN2O in various species (Table 1.3). The MACN2O can thereafter be used to estimate the MAC-reducing potential of N2O on a halogenated anesthetic in a specific species. These MACN2O values clearly highlight the weak anesthetic potency of N2O when used as the sole anesthetic agent.

FE′a MAC Equation 1.7: + b = 1 MACA MACB FE′a Equation 1.8: MACA = 1−MACb⁄MACB

In accordance with the simple linear additivity relationship, numerous studies have demonstrated the MAC-reducing properties of N2O on other anesthetic gases. For instance, the addition of 70% N2O to isoflurane led to a decrease of the MACIso by 65% to 72% in humans (Stevens et al., 1975). However, a non-linear relationship between N2O and halogenated gases has been suggested in rats (Cole, Kalichman, Shapiro, & Drummond, 1990; Russell & Graybeal, 1998) and horses (Testa, Raffe, & Robinson, 1990). The validity of these results in horses can be questioned by the use of the ocular reflex as a way to assess anesthetic depth rather than a true supramaximal stimulus. In rats, the subtle variation of the MAC suggested by the authors and the use of increased atmospheric pressure to determine the MACN2O makes it hard to confirm or refute the theory of simple linear additivity.

The use of N2O with halothane has been associated with faster inductions than halothane alone, which is thought to be due to the “second gas effect” (Epstein, Rackow, Salanitre, & Wolf, 1964). This effect is due to the rapid diffusion of N2O from the alveoli to the blood, which results in an increased alveolar concentration of the second anesthetic gas (Epstein et al., 1964; Stoelting & Eger, 1969). Consequently, diffusion of the second gas to the blood and tissue is accelerated. The low solubility of N2O and the second gas effect make N2O an interesting supplemental agent to accelerate induction and recovery while decreasing anesthetic gas requirements.

Table 1.3 – Minimum anesthetic concentration of nitrous oxide (MACN2O) in various species.

SPECIES SCIENTIFIC NAME MACN2O REFERENCES Cat Felis catus 255% Steffey, Gillespie, Berry, Eger, & Munson (1974) Dog Canis lupus familiaris 188% Eger, Brandstater, et al. (1965) 297% DeYoung & Sawyer (1965) Dumeril’s monitor Varanus dumerili 244% Bertelsen, Mosley, Crawshaw, Dyson, & Smith (2005a) Ferret Mustela putorius furo 267% Murat & Housmans (1988) Horse Equus ferus caballus 205% Steffey & Howland (1978) Human Homo sapiens 104% Hornbein et al. (1982) Mouse Mus musculus 150% Miller, Paton, Smith, & Smith (1972) 275% Deady, Koblin, Eger, Heavner, & D’Aoust (1981) Pig Sus scrofa 162% Eisele, Talken, & Eisele (1985) 277% Weiskopf & Bogetz (1984) Pigeon Columba livia 154% Fitzgerald & Blais (1991) Rat Rattus norvegicus 136% DiFazio et al. (1972) 235% Gonsowski & Eger (1994) Red-tailed hawk Buteo jamaicensis 220% Fitzgerald & Blais (1991) Stump-tailed Macaca arctoides 200% Steffey et al. (1974) monkey

1.2.4.2 Pharmacodynamics

The mechanism of action of N2O, as for other anesthetic gases, has only come to light in recent decades. A non-competitive antagonistic action on NMDA receptors is thought to be responsible for the effects of N2O (Jevtović-Todorović et al., 1998), although potassium channels may also be involved (Gruss et al., 2004). Additional details on this topic are provided in a thorough review of the biologic effects of N2O by Sanders, Weimann, & Maze (2008).

The effects of N2O on the respiratory and cardiovascular systems are limited. Although it decreases myocardial contractility, N2O stimulates the sympathetic nervous system, which prevents a significant decrease in cardiac output (Steffey & Mama, 2007). However, these cardiovascular effects may lead to cardiac arrhythmia when used in combination with epinephrine (Lampe et al., 1990; Liu, Wong, Port, & Andriano, 21

1982). Additional adverse effects following prolonged exposure include the interference of red blood cell production in the bone marrow, polyneuropathy, and megaloblastic hematopoiesis (Sanders et al., 2008). By preventing the action of cobalamin (i.e., vitamin B12) on the enzyme methionine synthase, N2O can also provoke clinical signs of cobalamin deficiency, which include a wide variety of neurological, hematological, and gastrointestinal problems (Bovill, 2008; Briani et al., 2013; Sanders et al., 2008). Nausea, vomiting, and postoperative wound infections have also been associated with the use of N2O in humans (Ko, Kaye, & Urman, 2014).

Nitrous oxide is 34 times more soluble than nitrogen, which composes approximately 78% of ambient air. Consequently, N2O may rapidly accumulate in closed air spaces within the body before nitrogen is eliminated, which can result in increased gas pressure in the middle ear and gastrointestinal tract, emphysematous bullae in the lungs, pneumothorax, or blood gas emboli (Bovill, 2008; Steffey & Mama, 2007). This effect may also lead to a phenomenon called “diffusion hypoxia” when the inhaled concentration of N2O is abruptly decreased. This phenomenon is the result of an increased pulmonary N2O concentration secondary to the rapid elimination of large volumes of N2O, which limits the pulmonary space available for O2 and, hence, O2 diffusion into the blood (Steffey & Mama, 2007).

1.3 Midazolam

1.3.1 History and physicochemical properties

The first benzodiazepine drugs described were chlordiazepoxide in 1960 (Randall, Schallek, Heise, Keith, & Bagdon, 1960) and diazepam in 1961 (Randall et al., 1961). Almost two decades later, midazolam was introduced, and was referred to as RO 21-3981 or midazolam maleate by Conner, Katz, Pagano, & Graham (1978) and Reves, Corssen, & Holcomb (1978). Midazolam represented a major improvement over diazepam, in part because of its increased water solubility. Midazolam is now commonly used as a sedative, preanesthetic agent, muscle relaxant, anxiolytic, anticonvulsant, or as part of total intravenous anesthesia in a wide variety of species (Nordt & Clark, 1997; Rankin, 2015).

Midazolam is a benzodiazepine to which an imidazole ring is fused for added water solubility at the formulated pH of 3.5. The chemical formula is 8-chloro-6-(2- fluorophenyl)-1-methyl-4H-imidazol(l,5-a)(l,4)-benzodiazepine. At physiological pH, it is highly liposoluble. Advantages to midazolam include that it is compatible with commonly used physiological saline solutions (e.g., NaCl 0.9%, lactate ringer’s, and dextrose 5% water) and intramuscular (IM) and intravenous (IV) injections are not irritating to tissues (Reves, Fragen, Vinik, & Greenblatt, 1985).

1.3.2 Pharmacokinetics

Midazolam is rapidly absorbed following IM, SC, and oral administration (Nordt & Clark, 1997). Its bioavailability is generally high, although a significant hepatic first-pass effect reduces oral bioavailability (Nordt & Clark, 1997). Up to 96% of plasma midazolam is bound to proteins, especially albumin (Nordt & Clark, 1997). Due to its highly lipophilic nature, midazolam is widely distributed, resulting in a relatively high volume of distribution (Nordt & Clark, 1997). Obesity has been shown to increase the volume of distribution of midazolam in humans (Brill et al., 2014). Hepatic metabolism by the cytochrome P450 3A (CYP3A) results in the formation of the primary metabolite 1-hydroxymidazolam, also named α-hydroxymidazolam, as well as 4-hydroxymidazolam and 1,4-hydroxymidazolam, also named α,4-hydroxymidazolam (Heizmann, Eckert, & Ziegler, 1983). The metabolite 1-hydroxymidazolam is considered pharmacologically active with a potency 2.5 to 10% that of midazolam in humans, while other metabolites have very little to no effect (Pieri, 1983). The hepatic extraction ratio of midazolam is considered intermediate to high in mammals, resulting in a flow-dependent metabolism (Allonen, Ziegler, & Klotz, 1981; McKindley, Hanes, & Boucher, 1998). Glucuronide conjugation subsequently occurs within the liver to further increase hydrosolubility. Elimination occurs mainly through renal excretion of the metabolites or their glucuronide conjugates (Nordt & Clark, 1997). Up to 90% of midazolam is eliminated in urine as a conjugated form within 24 h in humans (Amrein & Hetzel, 1990). The pharmacokinetics of midazolam have been evaluated in numerous species (Table 1.4).

Table 1.4 – Pharmacokinetic parameters of midazolam in various animal species.

SPECIES DOSAGE ROUTE Vd t½ AUC0-∞ Cl C0 Cmax tmax F (mg/kg) (L/kg) (min) (ng*h/mL) (L/h/kg) (ng/mL) (ng/mL) (min) (%) Alpaca1 0.5 IV 0.525ss 98 - 0.678 0.678 - - - (Vicugna pacos) 0.5 IM - 234 - - - 411 - 92 Dog2,3 0.5 IV 3.0z 77 - 1.62 1.62 - - - (Canis lupus familiaris) 0.5 IM - 56 - - - 549 8 115 0.2 IV 0.68ss 63.3 - 0.606 0.60 - - - 0.2 IM - 27.4 - - - 200 7.8 50 Horse4 0.05 IV 2.0937ss 216 81.2 0.636 0.636 - - - (Equus ferus caballus) 0.1 IV 2.8224ss 408 160 0.624 0.624 - - - Pig5 0.4 IV - 158 ------(Sus scrofa) 0.4 IN - 145 - - - 130 7.2 64 Rabbit6,7 1.80- 0.833- 820.56- 0.35 IV 30.03 274-485 - 1.60 - (Oryctolagus cuniculus) 2.99ss 1.6 1,565.78 0.19 IV 1.67z 90.0 267.2 1.02 - - - - 0.19 SL 1.23z 76.2 294.0 0.84 - 263.63 17.5 100 Sheep8 0.5 IV 0.838ss 28.2 423 1.272 - - - - (Ovis aries) 0.5 IM - 56.4 1,396 - - 820 27.6 352 Chicken9 5 IV 1.56ss 112.8 2,623 1.91 - - - - (Gallus gallus domesticus) Turkey9 5 IV 4.83ss 252 4,003 1.25 - - - - (Meleagris gallopavo) Common pheasant9 5 IV 3.82ss 397.2 3,913 1.28 - - - - (Phasianus colchicus) Bobwhite quail9 5 IV 4.17ss 258 2,085 2.40 - - - - (Colinus virginianus)

ss z Vd: volume of distribution; : volume of distribution in steady state; : volume of distribution during the elimination phase; t½: elimination half-life; AUC0-∞: area under the concentration-time curve extrapolated to infinity; Cl: clearance; C0: plasma concentration at time 0; Cmax: maximum plasma concentration; tmax: time to maximum plasma concentration; F: bioavailability; IV: intravenous; IM: intramuscular; IN: intranasal; SL: sublingual. References: 1(Aarnes et al., 2013); 2(Court & Greenblatt, 1992); 3(M. Schwartz, Muñana, Nettifee-Osborne, Messenger, & Papich, 2013); 4(Hubbell et al., 2013); 5(Lacoste et al., 2000); 6(Bienert et al., 2014); 7(Odou et al., 1999); 8(Simon et al., 2017); 9(Cortright, Wetzlich, & Craigmill, 2007). 24

1.3.3 Pharmacodynamics

The effects of midazolam are a result of the inhibitory actions of its target, the GABAA receptor. GABAA is present in numerous tissues, including the heart and skeletal muscles, but the highest concentrations are within the central nervous system (Nordt & Clark, 1997). Midazolam is a positive allosteric modulator of the GABAA receptor, which means that it does not have a direct effect on the target cell, but rather enhances the binding of the neurotransmitter GABA to the receptor, which results in an influx of chloride and membrane hyperpolarization (Nordt & Clark, 1997). The postsynaptic potential is consequently decreased and the neuron is inhibited (Olsen & DeLorey, 1999). This neuronal inhibition is responsible for the desirable effects of midazolam. However, activation of GABAA receptors at the supraspinal level can inhibit the antinociceptive pathways and therefore block the effects of analgesic drugs such as opioids. This inhibition is partially compensated by the antinociceptive effects of activated GABAA receptors at the spinal level (Lemke, 2007). Like other benzodiazepines, midazolam does not have intrinsic agonistic activity, meaning that central nervous system depression requires the presence of GABA.

The effects of midazolam are limited to the action of the GABAA receptor and the importance of these effects is variable among species. For instance, midazolam produces excellent sedation in humans (Reves et al., 1985), rabbits (Robertson & Eberhart, 1994), swine (Lacoste et al., 2000), goats (Stegmann & Bester, 2001) and numerous species of birds (Day & Roge, 1996; Mans, Guzman, Lahner, Paul-Murphy, & Sladky, 2012; Sadegh, 2013; Valverde, Honeyman, Dyson, & Valliant, 1990; Vesal & Eskandari, 2006) and reptiles (Arnett-Chinn, Hadfield, & Clayton, 2016; Oppenheim & Moon, 1995). However, the quality of the sedation can be highly variable in some species of turtles and tortoises (Bienzle & Boyd, 1992; Emery, Parsons, Gerhardt, Schumacher, & Souza, 2014; Harvey-Clark, 1993). The onset of action is typically very rapid and the duration relatively short in most species (Table 1.5). Midazolam, like other benzodiazepines, can interact with glycine receptors from the spinal cord resulting in muscle relaxation and from the mammillary body of the brainstem, resulting in an anxiolytic effect (Reves et al., 1985). Midazolam can also cause hypnosis by interfering with the reuptake of GABA, as well as amnesia, although the process leading to the latter is unknown (Reves et al., 1985).

The most important adverse effect of midazolam is paradoxical excitation, which is more commonly reported in cats (Ilkiw, Suter, Farver, McNeal, & Steffey, 1996) and less commonly in humans (Robin & Trieger, 2002; Tae et al., 2014), dogs (Court & Greenblatt, 1992; Simon et al., 2014), sheep (Simon et al., 2017), and horses (Hubbell et al., 2013). Paradoxical excitation is most common with higher dosages and the effect typically is transient (Court & Greenblatt, 1992; Simon et al., 2017). Multiple theories attempt to explain this reaction: a central anticholinergic effect leading to delirium, disorientation, and hallucinations; production of rage and aggression due to alteration of the serotonergic balance; and mutations of the GABAA receptor (Robin & Trieger, 2002). Hypersalivation also been reported in sheep (Simon et al., 2017) and dogs (Court & Greenblatt, 1992). Midazolam has been reported to mildly increase the heart 25

rate in humans (Reves et al., 1985), dogs (Jones, Stehling, & Zauder, 1979), and swine (Lacoste et al., 2000), and to mildly decrease blood pressure in humans (Reves et al., 1985). However, the opposite effect has been reported in Burmese pythons, in which midazolam decreases the heart rate (Lopes, Armelin, Braga, & Florindo, 2017). Mild respiratory depression has been reported in humans (Reves et al., 1985) and swine (Lacoste et al., 2000). Overall, these adverse effects are minimal, and midazolam is therefore considered a very safe drug in patients with cardiac or respiratory issues.

Table 1.5 – Pharmacodynamic parameters of midazolam in various species.

SPECIES DOSAGE ROUTE ONSET PEAK DURATION (mg/kg) (min) (min) (min) Alpaca1 0.5 IV 0.4 0.4 105 (Vicugna pacos) 0.5 IM 15 22.5 105 Boa constrictor2 1.0 ICe 15 - 279 (Boa constrictor) 2.0 ICe 15 - 372 Budgerigar3 13.2 IN 1.3 - 71.6 (Melopsittacus undulatus) Cat4 0.05-5.0 IV 0.52 - - (Felis catus) 0.05-5.0 IM 2.38 - - Common snapping turtle5 2.0 IM 5 - - (Chelydra serpentina) Goat6,7 0.6 IV - 5 15.7 (Capra aegagrus hircus) 1.2 IV - 5 25.8 0.6 IM - 20 >40 0.4 IM 5 - 34 1.0 IM 2.7 - 60.9 Hispaniolan amazon8 2.0 IN 2.7 - 60.9 (Amazona ventralis) Ring-necked parakeet9 7.3 IN 2.7 - 175.8 (Psittacula krameria) Rabbit10,11 0.35 IV - 15 40.79 (Oryctolagus cuniculus) 2.0 IN 3.0 - 24.6 Red-eared slider12 1.5 IM 5.5 - 82 (Trachemys scripta elegans) Sheep13,14 0.5 IV 3 - 45 (Ovis aries) 0.5 IM 10 - 60 0.05 IV Transient drowsiness 0.1 IV 1.32 - 14.45 0.2 IV 1.05 - 22.93 0.4 IV 1.35 - 19.90 Tegu15 1.0 IM 10 - 170 (Salvator merianae) Various reptiles species16 0.1-1.0 IM 10 20 - IV: intravenous; IM: intramuscular; ICe: intracoelomic; IN: intranasal. References: 1(Aarnes et al., 2013); 2(De Simone, Hirano, & Santos, 2017); 3(Sadegh, 2013); 4(Ilkiw, Suter, Farver, et al., 1996); 5(Bienzle & Boyd, 1992); 6(Stegmann & Bester, 2001); 7(Stegmann, 1998); 8(Mans et al., 2012); 9(Vesal & Eskandari, 2006); 10(Bienert et al., 2014); 11(Robertson & Eberhart, 1994); 12(Oppenheim & Moon, 1995); 13(Simon et al., 2017); 14(Upton, Martinez, & Grant, 2009); 15(Bisetto, Melo, & Carregaro, 2018); 16(Arnett-Chinn et al., 2016).

1.3.4 MAC-sparing effects

Benzodiazepines have variable MAC-sparing effects on halogenated gases. However, studies evaluating the effect of midazolam on the MACIso are relatively uncommon. The IV administration of midazolam as a constant rate-infusion (CRI) has been shown to cause a dose-dependent reduction of the MACIso in dogs ranging from 11%1 to 32%2 (Seddighi, Egger, Rohrbach, Cox, & Doherty, 2011). Similarly, the use of midazolam at a dosage of 0.25 mg/kg IM in dogs decreased the end-tidal concentration of isoflurane necessary to prevent movements during ovariohysterectomy by 14.3% compared to a placebo (Kropf & Hughes, 2018). In rats, the intrathecal administration of 5, 10, 20, and 30 µg of midazolam decreased the MACIso by 16%, 31%, 42%, and 53%, respectively (Schwieger, Jorge-Costa, Pizzolato, Forster, & Morel, 1994). Most studies evaluating multiple doses of midazolam concluded that the MAC-reducing effect is dose-dependent (Inagaki, Sumikawa, & Yoshiya, 1993; Melvin, Johnson, Quasha, & Eger, 1982; Seddighi et al., 2011) and that a ceiling effect is eventually reached (Seddighi et al., 2011).

Midazolam also reduces the MAC of halothane (MACHal) by 5% to 30% in humans when given 0.15 to 0.6 mg/kg IV, respectively (Melvin et al., 1982). In another 3 human study, midazolam administered as a CRI decreased the MACHal by 39.7% to 4 70.5% (Inagaki et al., 1993). Similarly, 1 mg/kg IV decreased the MACHal by 37% in rats (Greiner & Larach, 1989). In dogs, 2.5 mg/kg IV decreased the MAC of enflurane by 56.2% (Hall, Szlam, & Hug, 1988). Other benzodiazepines such as diazepam also have MAC-reducing effects on isoflurane in dogs (Hellyer, Mama, Shafford, Wagner, & Kollias-Baker, 2001), as well as halothane in humans (Tsunoda, Hattori, Takatsuka, Sawa, & Hori, 1973) and horses (Matthews, Dollar, & Shawley, 1990).

1.3.5 Flumazenil: a benzodiazepine antagonist

Benzodiazepine antagonists such as flumazenil can be used to block or reverse the effects of midazolam due to their higher affinity for GABAA receptors (Rankin, 2015). Flumazenil does not exhibit intrinsic activity in humans, making it an ideal reversal agent (Amrein & Hetzel, 1990). The molecular structure of flumazenil is similar to that of midazolam, except that a phenyl group is replaced with a carbonyl group (Amrein & Hetzel, 1990). It is less lipophilic than midazolam, but also less water soluble due to its weak acid nature (Amrein & Hetzel, 1990). Flumazenil is moderately bound to proteins with an average of 50-60% in unbound form in human plasma (Amrein & Hetzel, 1990; Votey, Bosse, Bayer, & Hoffman, 1991). The drug is almost entirely metabolized by hydroxylation and glucuronide conjugation in the liver before elimination in the urine (Hoffman & Warren, 1993). Pharmacokinetic parameters are similar between midazolam and flumazenil, both having a relatively high volume of distribution and high

1 Midazolam administered IV at 0.2 mg/kg followed by a CRI of 2.5 µg/kg/min 2 Midazolam administered IV at 1.6 mg/kg followed by a CRI of 20 µg/kg/min 3 Midazolam administered IV at 0.1 mg/kg followed by a CRI of 1 µg/kg/min 4 Midazolam administered IV at 0.4 mg/kg followed by a CRI of 4 µg/kg/min 28

hepatic extraction ratio. In humans, the volume of distribution in steady state, the clearance, hepatic extraction ratio, and the elimination half-life of flumazenil are 0.95 L/kg, 0.5-1.3 L/min, 0.6, and 0.7-1.3 h, respectively (Amrein & Hetzel, 1990).

1.4 Anesthesia of reptiles

1.4.1 The challenges of reptile anesthesia

Cardiorespiratory adaptations

Anesthesia in non-avian reptilian species involves an additional level of complexity in comparison to mammals and birds due to their unique anatomy and physiology. As described in section 1.1.2.2, reptiles readily shunt blood within their heart and lungs, which results in an uneven equilibrium of gas tension between the blood and lungs. Differences in PO2 between the arterial blood and airways in reptiles breathing ambient air reportedly range from 30 mmHg in monitor lizards (Burggren & Johansen, 1982) to 70 mmHg in sea snakes (Seymour & Webster, 1975). Reptiles are typically highly resistant to hypoxia, which may increase the safety of anesthesia even in conjunction with respiratory depression (Heard, 2001).

Assessment of anesthetic depth

The highly variable response to noxious or non-noxious stimuli in reptiles is another challenge that makes the assessment of anesthetic depth difficult. Loss of reflexes during inhalation anesthesia have been reported to start with the limbs and mid-body in lizards, subsequently extending to the head, and ending at the tail (Bertelsen, Mosley, Crawshaw, Dyson, & Smith, 2005b; Bonath, 1979). The righting reflex is lost after the loss of head reflexes but before the tail reflexes in lizards (Bertelsen et al., 2005b; Bonath, 1979). In snakes, the tongue retraction reflex is usually maintained when a surgical plane of anesthesia is reached and, therefore, loss of this reflex indicates excessive anesthetic depth (Schumacher, Lillywhite, Norman, & Jacobson, 1997).

Body temperature and metabolism

Reptiles are ectothermic animals, meaning that they need to adapt their behavior and metabolism according to the ambient temperature. Therefore, the efficiency of anesthetic agents and, consequently, the induction and recovery times are highly dependent on body temperature (Dohm & Brunson, 1998; Kischinovsky, Duse, Wang, & Bertelsen, 2013; Olsson & Phalen, 2013b; Schumacher & Yelen, 2005). The pharmacokinetics of anesthetic drugs has also been shown to be affected by body temperature in reptiles. In gopher tortoises (Gopherus polyphemus) receiving 5 mg/kg IM of amikacin, the clearance was faster and the mean residence time shorter when kept at 30°C versus 20°C (Caligiuri et al., 1990). In gopher snakes (Pituophis sp.) receiving the same dosage of amikacin, the volume of distribution was larger and the clearance faster, although the elimination half-life was unchanged (Mader, Conzelman, & Baggot, 1985). The MAC of anesthetic gases has been shown to have a positive linear relationship with temperature in mammalian and reptilian species, as shown in Table 1.2 for reptiles (Aranake et al., 2013; Dohm & Brunson, 1998; Vitez, White, & Eger, 1974). The field metabolic rate of reptiles is overall much lower than that of 30

similarly sized mammals or birds, which is hypothesized to be mainly attributed to their ectothermic nature (Nagy, 2005). Maintaining the body of temperature of reptiles within the high-end of their preferred optimal temperature range is hence of high importance during anesthetic procedures to maintain adequate metabolism.

Renal portal system

Almost all non-mammalian species have a renal portal vascular system, meaning that a portion or the entirety of venous blood from the caudal part of the body must pass through the kidneys and the liver before reaching the heart (Holz, 1999). The anatomy of this renal portal system varies between taxonomic orders and species. Most studies of renal portal blood flow in reptiles have been performed in turtles. In the red-eared slider (Trachemys scripta elegans), a large proportion of the blood from the hindlimbs and tail can bypass the kidneys, but all of this blood converges into the post-caval and abdominal veins that pass through the liver (Holz, Barker, Crawshaw, & Dobson, 1997). A strong renal and hepatic first-pass effect has therefore long been expected when drugs are injected in the caudal part of the body of reptiles. However, pharmacokinetic studies in reptiles in the 1990s and early 2000s failed to demonstrate a first-pass effect (Holz, 1999). In red-eared sliders receiving 10 mg/kg IM of gentamicin or 200 mg/kg IM of carbenicillin, as well as in eastern box turtles (Terrapene carolina carolina) receiving 32 mg/kg IM of gentamicin, there was no difference in pharmacokinetic parameters between an injection in the forelimbs and hindlimbs (Beck, Loomis, Lewbart, Spelman, & Papich, 1995; Holz, Barker, Burger, Crawshaw, & Conlon, 1997). The same findings were reported in carpet pythons (Morelia spilota) receiving 200 mg/kg IM of carbenicillin in the cranial and caudal halves of the body (Holz, Burger, Pasloske, Baker, & Young, 2002). In mammals however, aminoglycosides and penicillins are eliminated almost exclusively by the kidneys without undergoing hepatic metabolism (Baggot, Love, Rose, & Raus, 1981; Barza & Weinstein, 1976). The consequences of the hepatic first-pass effect on the pharmacodynamics and pharmacokinetics of drugs undergoing major hepatic metabolism or elimination was therefore not fully evaluated until recently. Numerous studies in the last decade demonstrated a significant effect of the injection site on the pharmacokinetics and pharmacodynamics of numerous analgesic, sedative, and anesthetic drugs in reptiles. Buprenorphine given at a dosage of 0.02 mg/kg IM resulted in a significantly lower maximum plasma concentration and area under the concentration-time curve when injected in the hindlimbs compared to the forelimbs of red-eared sliders (Kummrow, Tseng, Hesse, & Court, 2008). Sedation was also significantly less profound when administering alfaxalone IM in the caudal part of the body of ball pythons (James, Williams, Bertelsen, & Wang, 2018; Yaw, Mans, Johnson, Doss, & Sladky, 2018) or when administering dexmedetomidine and ketamine IM in the hindlimbs of leopard geckos (Eublepharis macularius; Fink, Doss, Sladky, & Mans, 2018) compared to the cranial body or forelimbs, respectively. Although the renal first- pass effect has not yet been identified as problematic in anesthetizing reptiles, serious consideration must be made regarding the injection site of drugs that are metabolized or eliminated by the liver.

1.4.2 Inhalation anesthesia

A survey among the reptilian veterinary community showed that inhalation anesthesia was more commonly used than injectable drugs for the maintenance of anesthesia (Read, 2004). Although halothane has been widely used in reptiles, it has been replaced over recent decades by safer anesthetic gases such as isoflurane and sevoflurane. The current section will, however, focus on the two anesthetic gases studied in this thesis: isoflurane and N2O.

1.4.2.1 Isoflurane

Isoflurane is the most commonly used anesthetic gas in reptiles (Read, 2004). Induction with 5% isoflurane in 100% O2 typically leads to loss of righting reflex and overall relaxation within 4 to 9 min (Hernandez-Divers, Schumacher, Stahl, & Hernandez-Divers, 2005; Spelman, Cambre, Walsh, & Rosscoe, 1996) and to a surgical plane of anesthesia within 13 to 20 min (Bertelsen et al., 2005b; C. A. E. Mosley et al., 2003a) in various reptilian species. Induction using a face mask has been reported to be much slower in snakes compared to other reptiles (Bertelsen, 2014). However, intubation and mechanical ventilation is possible in non-anesthetized snakes following topical anesthesia of the glottis with lidocaine (Bertelsen, 2014). Induction times may also be prolonged in reptiles that can hold their breath and undergo R-L cardiac shunting, such as aquatic species.

Recovery times seem to be correlated to the duration of anesthesia in most reptiles (Bertelsen, 2014). While short recovery times (i.e., 2-12 min) have been reported following brief isoflurane anesthesia (Spelman et al., 1996), recovery periods of up to 70 min have been reported in lizards following prolonged inhalation anesthesia (Bertelsen et al., 2005b; C. A. E. Mosley et al., 2003a). Intracardiac blood shunting is thought to be a major contributor to the prolonged recovery period of reptiles. The main contributor to IC shunting in reptiles is the difference between the systemic and pulmonary vascular resistance (Overgaard et al., 2002). Stimulation of α-adrenergic and β-adrenergic receptors by catecholamines such as epinephrine and norepinephrine increases systemic vasoconstriction and pulmonary vasodilation, resulting in a net L-R IC shunt (Overgaard et al., 2002). Based on this principle, studies have demonstrated faster recovery times in common snapping turtles and American alligators (Alligator mississippiensis) receiving epinephrine (Gatson et al., 2017; Goe et al., 2016). Conversely, parasympathetic stimulation increases pulmonary vascular resistance leading to a net R-L IC shunt. The use of anticholinergics such as atropine eliminates the parasympathetic tone and consequently eliminates R-L IC shunting in turtles (Greunz et al., 2018; Hicks & Comeau, 1994). The use of 100% O2 instead of 21% O2 for the anesthetic recovery of reptiles has long been hypothesized to be associated with prolonged recovery periods. However, no significant differences were found regarding recovery times in Dumeril’s monitor (Bertelsen et al., 2005b) and bearded dragons (Pogona vitticeps; O, Churgin, Sladky, & Smith, 2015) receiving 100% O2 versus 21% O2.

Isoflurane offers a wide safety margin in reptiles. Concentrations up to 9.2% delivered in 100% O2 (4.32 x MACIso) failed to induce cardiac arrest in the green iguana (Iguana iguana; C. A. E. Mosley, Dyson, & Smith, 2003b). However, isoflurane has been reported to cause mild to moderate bradycardia in the green iguana (Hernandez- Divers et al., 2005) and the radiated ratsnake (Elaphe radiata; Maas & Brunson, 2002). Respiratory depression, which often leads to apnea, is the most important side effect of isoflurane in reptiles (Bertelsen, 2014; Maas & Brunson, 2002). However, as a result of mechanical ventilation with 100% O2, hyperoxia has long been hypothesized to contribute to the inhibition of spontaneous breathing by decreasing the sensitivity to hypoxia, although this theory has never been clearly demonstrated (C. A. E. Mosley, 2005; Schumacher, 2003). Finally, isoflurane has been shown to decrease systemic blood pressure in the green iguana (Chinnadurai et al., 2010; C. A. E. Mosley, Dyson, & Smith, 2004).

1.4.2.2 Nitrous oxide

Nitrous oxide is commonly used as a supplemental agent during inhalation anesthesia in reptiles, although data from clinical trials are limited. The main advantages of N2O in reptiles are decreased induction times (Bertelsen et al., 2005b; Custer & Bush, 1980), decreased sevoflurane requirements (Bertelsen et al., 2005a), and an added analgesic effect (Bertelsen, 2014). Nitrous oxide is typically administrated as a mixture with O2 at an O2:N2O concentration ratio of 50:50 to 34:66 (Bertelsen, 2014). As in mammals and birds, the MACN2O is very high in reptiles and therefore N2O cannot be used a sole anesthetic (Table 1.3).

1.4.2.3 Minimum anesthetic concentration

Considering the absence of alveoli in the reptilian lung, the term “minimum alveolar concentration” is inappropriate. Other terms, such as minimum infundibular concentration (Maas & Brunson, 2002), effective dose (Dohm & Brunson, 1998), and median effective dose (Barter et al., 2006) have been used. However, the term minimum anesthetic concentration (MAC) occasionally used in the avian literature (da Rocha et al., 2017; Escobar et al., 2016) facilitates consistency with the mammalian literature.

Currently available MACIso values of various reptilian species are listed in Table 1.2. A single MAC study has been reported in snakes, in which the authors evaluated the MACIso and the MAC of sevoflurane (MACSevo) in the radiated ratsnake. Results from this study indicated a MACIso of 1.68% and a MACSevo of 2.42% (Maas & Brunson, 2002). Two noteworthy challenges to MAC evaluation in reptiles are the accurate evaluation of anesthetic depth, as well as the achievement and confirmation of lung- blood-tissue partial pressure equilibration. The latter is especially challenging considering the presence of IC and intrapulmonary blood shunting. While the response to tissue clamping under anesthesia varies according to anatomic site (Eger, Saidman, et al., 1965), desensitization of tissues at the electrical stimulation site has also been reported to alter measurement of the MAC in reptiles (Barter et al., 2006). The MACIso 33

has been reported to decrease with prolonged anesthesia in the green iguana (Barter et al., 2006). The authors suggested that this phenomenon may be due to variations of the IC blood shunt or of the blood flow to the central nervous system over time, although an increased anesthetic potency over time or neuronal windup could not be discounted (Barter et al., 2006). However, few studies reported variations of the MAC over time and it is therefore likely that this variation in the green iguana was the result of flawed methodology, such as an inadequate equilibration time leading to an overestimation of the MAC during the beginning of the procedure. Intracardiac shunting has also been shown to significantly affect the MAC, likely due to markedly prolonged lung-blood- tissue equilibration times during R-L cardiac shunting. In red-footed tortoises for instance, the administration of atropine eliminated IC blood shunting and hence decreased the MACIso from 3.2% to 2.2% (Greunz et al., 2018). As previously mentioned, MAC also varies with body temperature, likely due to variations in metabolism and cardiorespiratory function. Dohm & Brunson (1998) demonstrated that hypothermia increased the MACIso in desert iguanas (Dipsosaurus dorsalis).

1.4.3 Midazolam

Midazolam is commonly used in reptiles as a sedative, preanesthetic agent, muscle relaxant, or anxiolytic agent in reptilian species (Arnett-Chinn et al., 2016; Olsson & Phalen, 2013a). A recent retrospective study on the use of midazolam given at a dosage of 0.1-1.0 mg/kg IM in numerous reptilian species reported an acceptable sedation in 35 out of 44 (80%) events, with no significant observed influence of the species, taxonomic order, age, sex, or body weight (Arnett-Chinn et al., 2016). This drug also induced effective sedation when given at a dosage of 1.5 mg/kg IM in the forelimbs of red-eared sliders (Oppenheim & Moon, 1995), 1 mg/kg IM in the forelimbs of tegus (Salvator merianae; Bisetto et al., 2018), 1 and 2 mg/kg intracoelomic in boa constrictors (Boa constrictor; Simone et al., 2017), and 5.0 mg/kg IM in the forelimbs of estuarine crocodiles (Crocodylus porosus; Olsson & Phalen, 2013a). In the latter study, physically restrained crocodiles that did not receive sedatives had disrupted basking and feeding behaviors for three days, while midazolam-treated animals exhibited normal behavior following the procedure, suggesting an anxiolytic effect of midazolam in this species (Olsson & Phalen, 2013a). However, it should be noted that midazolam has been shown to inconsistently produce sedation when administered as a sole agent either IM or intranasal in various species of turtles and tortoises (Bienzle & Boyd, 1992; Emery et al., 2014; Harvey-Clark, 1993). The time to initial effect and peak effect, and the duration of action of midazolam in various reptilian species are reported in Table 1.5.

Flumazenil given at a dosage of 0.01-0.02 mg/kg IM in the forelimbs has been reported to successfully reverse the sedative effects of midazolam in 28/31 (90%) events across a wide range of reptile species (Arnett-Chinn et al., 2016). A dosage of 0.05 mg/kg of flumazenil administered IM in the forelimbs of leopard geckos (Doss, Fink, Sladky, & Mans, 2017) and the cranial body of ball pythons (Yaw, Mans, Doss, Johnson, & Sladky, 2017), in combination with atipamezole, successfully reversed the effects of dexmedetomidine and midazolam. 34

1.5 Study rationale and objectives

Although commonly used as a sole sedative or in anesthetic combinations in snakes (Arnett-Chinn et al., 2016; De Simone et al., 2017; Yaw et al., 2017), the pharmacodynamics and pharmacokinetics of IM midazolam have not been characterized in any ophidian species. Dosages used are therefore extrapolated from the mammalian or avian literature despite major differences between these classes of animals. Desirable effects such as sedation, muscle relaxation, or decreased anesthetic gas requirements are difficult to evaluate in snakes, which may have led to reports of overdosages to provide an obvious effect. Similarly, although nitrous oxide has been reported to decrease the MACSevo in Dumeril’s monitors (Bertelsen et al., 2005a), its effects have not yet been reported on the MACIso in reptiles.

The overall objective of this project was to characterize the pharmacodynamics and pharmacokinetics of midazolam in ball pythons with an emphasis on the MAC-reducing effect of midazolam on isoflurane in comparison to nitrous oxide. Hypotheses formulated for this project are as follow:

- Hypothesis 1: Midazolam provides moderate and safe sedation and muscle relaxation in ball pythons, and flumazenil can be used to reverse the effects of midazolam. - Hypothesis 2: The administration of midazolam IM and the delivery of isoflurane in a mixture of 50% N2O and 50% O2 significantly decreases the MACIso in ball pythons. - Hypothesis 3: The pharmacokinetics of midazolam administered IC and IM in ball pythons is similar to those in mammalian species.

To address the following hypotheses, the following objectives were formulated:

- Objective 1: To evaluate the sedative, muscle relaxant, and cardiorespiratory effects of IM midazolam administered at two dosages and to evaluate the reversal effect of IM flumazenil in ball pythons. - Objective 2: To determine the MACIso and to evaluate the effects of midazolam and N2O on the MACIso in ball pythons. - Objective 3: To evaluate the absorption, distribution, and elimination of midazolam and its major metabolite 1-hydroxymidazolam following the administration of 1 mg/kg through two different routes, IC and IM, in ball pythons.

2 Evaluation of the effects of midazolam and flumazenil in the ball python (Python regius)

Cédric B. Larouche, D.M.V., I.P.S.A.V., Hugues Beaufrère, Dr. Med. Vet., Ph.D., Dipl. A.C.Z.M., Dipl. E.C.Z.M. (Avian), Dipl. A.B.V.P. (Avian), Craig Mosley, D.V.M., M.Sc., Dipl. A.C.V.A.A., Nicole M. Nemeth, D.V.M., Ph.D., Dipl. A.C.V.P., and Christopher Dutton, B.V.Sc., M.Sc., Dipl. A.C.Z.M., Dipl. E.C.Z.M. (Zoo Health Management)

From the Departments of Pathobiology (B. Larouche, Nemeth) and Clinical Studies (Beaufrère), Ontario Veterinary College, University of Guelph, 50 Stone Road E., Guelph, Ontario N1G 2W1, Canada; Toronto Zoo, 361A Old Finch Avenue, Toronto, Ontario M1B 5K7, Canada (B. Larouche, Dutton); and VCA Canada, 404 Veterinary Emergency and Referral Hospital, 510 Harry Walker Parkway S., Newmarket, Ontario L3Y 0B3, Canada (Mosley). Present address (Nemeth): Departments of Pathology and Population Health, College of Veterinary Medicine, University of Georgia, 501 D. W. Brooks Drive, Athens, Georgia 30602, United States.

Larouche, C. B., Beaufrère, H., Mosley, C., Nemeth, N. M., & Dutton, C. (2019). Evaluation of the effects of midazolam and flumazenil in the ball python (Python regius). Journal of Zoo and Wildlife Medicine, 50(3), In Press. https://doi.org/10.1638/2019-0024 (Accepted for publication on 2019-05-14)

2.1 Abstract

The study objective was to evaluate the sedative, muscle relaxant, and cardiorespiratory effects of midazolam and flumazenil in the ball python (Python regius). Ten healthy adult female ball pythons were used in a randomized and blinded crossover trial evaluating the effects of two dosages (1 and 2 mg/kg IM in the cranial third of the body). In a subsequent open trial, nine ball pythons received 1 mg/kg IM of midazolam followed by 0.08 mg/kg IM of flumazenil 60 min later. Heart rate, respiratory rate, temperature, and the level of sedation and muscle relaxation (using a semi-objective scoring system) were evaluated. There were no significant differences between midazolam dosages for any of the parameters evaluated. Sedation scores were significantly increased compared to baseline from 15 min (1 mg/kg) and 10 min (2 mg/kg) post-injection up until 56 h (1 mg/kg) and 72 h (2 mg/kg) post-injection. Peak effect was reached 60 min post-injection with 60% of snakes (6/10) being unable to right themselves. One snake developed paradoxical excitation with the 2 mg/kg dosage. Heart rates were significantly lower than baseline from 30 min to 128 h post-injection with both midazolam dosages. Respiratory rates were significantly lower than baseline at four time points with the highest dosage only: 15, 45, 60 min, and 8 h post-injection. Flumazenil resulted in reversal of sedation and muscle relaxation in all snakes within 10 min of administration. However, resedation was evident in all snakes 3 h after reversal. Midazolam administered at 1 and 2 mg/kg IM provides a moderate to profound, although prolonged, sedation and muscle relaxation in ball pythons. Flumazenil reverses the effects of midazolam in ball pythons, but its duration of action at the evaluated dosage is much shorter than midazolam, leading to resedation.

2.2 Introduction

Midazolam is a benzodiazepine with a wide cardiorespiratory safety margin in numerous species (Lemke, 2007). It is a GABAA receptor agonist, producing reliable and reversible sedation and muscle relaxation with additional anxiolytic and anticonvulsive effects in a wide variety of species (Lemke, 2007; Reves et al., 1985). Furthermore, it has been shown to decrease the minimum alveolar concentration of various anesthetic gases in humans, dogs, and rats (Greiner & Larach, 1989; Melvin et al., 1982; Seddighi et al., 2011). Studies of the specific effects of midazolam in reptiles are limited. Effective sedation has been demonstrated when used as a sole agent injected intramuscularly (IM) in the forelimbs at a dosage of 1 mg/kg in tegus (Salvator merianae; Bisetto et al., 2018), 1.5 mg/kg in red-eared sliders (Trachemys scripta elegans; Oppenheim & Moon, 1995), and 5 mg/kg in estuarine crocodiles (Crocodylus porosus; Olsson & Phalen, 2013a). Furthermore, intracoelomic administration in the caudal third of the body with 1 or 2 mg/kg resulted in marked muscle relaxation and reduced movement in boa constrictors (Boa constrictor), although the righting reflex was maintained (De Simone et al., 2017). However, this sedative effect was not consistently observed in common snapping turtles (Chelydra serpentina; Bienzle & Boyd, 1992) receiving 2 mg/kg IM in the forelimbs, and in red-footed tortoises (Chelonoides carbonaria) and Indian star tortoises (Geochelone platynota) receiving 0.5-1.5 mg/kg intranasally (Emery et al., 2014). A retrospective study on the use of midazolam in a wide variety of reptiles from a zoological institution reported adequate sedation in 80% of events when administered at a dosage of 0.1-1 mg/kg IM (Arnett-Chinn et al., 2016). Its use in combination with either ketamine, dexmedetomidine, or alfaxalone has been shown to provide reliable sedation in numerous reptilian species (Bisetto et al., 2018; Doss et al., 2017; Holz & Holz, 1994; Mans & Sladky, 2012; Yaw et al., 2017). Although commonly used in snakes (Arnett-Chinn et al., 2016; Yaw et al., 2017), to the authors’ knowledge, there are no prospective studies evaluating the pharmacodynamics of an IM administration of midazolam in any ophidian species. Therefore, the desirable and adverse effects of this drug in snakes are not objectively available and its contribution to a multidrug protocol is therefore unknown. Flumazenil is a competitive benzodiazepine receptor antagonist that is used to block or reverse the effects of benzodiazepines. Although reported to be an effective antagonist in numerous reptilian species, its effects have not yet been evaluated in a prospective study to reverse midazolam alone (Arnett-Chinn et al., 2016; Doss et al., 2017; Mans & Sladky, 2012; Yaw et al., 2017). The current study was driven by the hypothesis that midazolam provides moderate and safe sedation and muscle relaxation in ball pythons (Python regius), and that flumazenil can be used to reverse the effects of midazolam. The objectives were therefore to evaluate the sedative, muscle relaxant, and cardiorespiratory effects of 38

midazolam when administered IM at two different dosages (i.e., 1 and 2 mg/kg) and to evaluate the reversal effect of flumazenil in ball pythons.

2.3 Material and methods

2.3.1 Animals

Ten adult female captive-bred ball pythons with a mean ± standard deviation (SD) weight of 3.13 ± 0.53 kg (Appendix B, Table B.1) were used for the study. All snakes were individually housed in plastic terrariums with a rubber mat and access to a water bowl, a hiding area, and a heating pad placed underneath the terrarium. The temperature gradient in the enclosures was 26°C to 35°C, humidity was kept between 40% and 60%, and lighting was programmed to provide 12 h of light and 12 h of darkness per day. Each snake was fed a frozen-thawed mouse or rat every other week, according to the snake size. All snakes were fasted at least seven days before any experimentation to allow for complete digestion of the prey as well as defecation. All snakes were deemed healthy based on physical examination, coelomic ultrasonography, and fecal parasite evaluation by flotation and sedimentation. Each animal was identified by a subcutaneous microchip (Eidap Inc., Brampton, Ontario L7A 1B9, Canada) that was scanned before any procedure. Throughout the experimentation, snakes were only taken out of their terrariums for enclosure servicing, weighing, injections, or assessment of muscle relaxation. All experimental protocols were approved by the University of Guelph (#3812) and Toronto Zoo (#2017-11-02) Animal Care and Research Committees. Experimentation was performed in the Wildlife Health Centre of the Toronto Zoo.

2.3.2 Midazolam trial

A blinded, randomized, crossover trial was performed to compare the effects of a low (1 mg/kg) and a high (2 mg/kg) dosage of midazolam administered IM in ball pythons. The order of snakes and treatments were randomly assigned with replicated Latin squares using a statistical software (R package version 3.5.0, R Core Team [2018], R Foundation for Statistical Computing, Vienna 1020, Austria) to balance for carry-over and first order effects. A two-week washout period was allowed between each experiment, which was considered adequate with an expected duration of sedation of < 48 h based on other studies in reptiles (Bisetto et al., 2018; De Simone et al., 2017). Each snake was manually restrained to receive the midazolam injection (50 mg/10 mL, Sandoz, Boucherville, Québec J4B 7K8, Canada) within the epaxial muscles of the cranial third of the body. Researchers observing the snakes and analyzing the data were blinded from the administered treatments. To evaluate the level of sedation, individual scores were attributed to the reaction of the snake to head restraint, and the 39

head and body righting reflexes (Table 2.1). A snake was considered to have successfully righted its head when the head had completely returned to a normal upright position from an upside-down position. Similarly, a successful body righting reflex was defined as the repositioning from dorsal recumbency to a normal upright position with > 90% of the ventral surface of the snake in contact with the mat. Head and body righting reflexes were scored separately since mildly to moderately sedated snakes would right their head independently from their body. Therefore, an individual that maintained its capacity to right its head but not its body was considered more deeply sedated than if both righting reflexes were maintained. Individual scores from Table 2.1 were added for each snake to obtain a composite sedation score with a maximum of 6. Sedation was considered absent (0), mild (1-2), moderate (3-4), or profound (5-6) based on this composite score. To evaluate the level of muscle relaxation, snakes were restrained and removed from their terrarium. A scoring system based on three observations was used for a maximum score of 2: the capacity of the snake to maintain its shape when lifted by the cranial third of the body, resistance to extension of the body, as well as success and rapidity to regain a resting shape following the extension of the body (Table 2.2). The resting shape was defined as the position the snake took just after being touched, which was typically a coiled “ball” shape. Physiological parameters evaluated included heart rate (HR) using a Doppler probe placed over the heart (Model 811-B, Parks Medical Electronics Inc., Aloha, Oregon 97078, United States), respiratory rate (RR) by visual observation, and temperature using an infrared laser thermometer (Fluke 59 MAX Infrared Thermometer, Fluke Electronics Canada, Mississauga, Ontario L4Z 1X9, Canada) aimed at the level of the heart. Sedation and muscle relaxation scores were evaluated immediately prior to the injection (baseline; T = 0) as well as 3, 5, 10, 15, 30, 45, 60, 90, 120, 180, 240, 300, 360, and 480 min post-injection. Scoring was also performed the following mornings and evenings (24, 32, 48, 56, 72, 80, 96, 104, 120, 128, 144, 152, 168, 176, 192, and 200 h post-injection) until the sedation score was the same as the baseline. Parameters were evaluated in the following order to decrease the impact of manipulations on physiological data: temperature, RR, HR, reaction to head manipulation, muscle relaxation score, and righting reflex. The HR, RR, and temperature were evaluated at all the time points listed above except for 3 min post-injection due to time restriction. Sample size was calculated based on similar studies using a sedation score in reptilian species with an alpha and beta error threshold of 0.05 and 0.2, respectively (Bisetto et al., 2018; Fink et al., 2018; Kohn, Jarrett, & Senyak, 2018). A significant effect size was determined at 2 points on the composite sedation score scale described above and the standard error of the mean (SEM) was expected to be up to 25% of the scale used (i.e., 1.5 points).

Table 2.1 – Scoring system for the evaluation of the level of sedation associated with an injection of midazolam in ball pythons.

SCORE RIGHTING REFLEX RIGHTING REFLEX REACTION TO HEAD (HEAD) (BODY) RESTRAINT 0 Righted in ≤ 2 sec Righted < 20 sec Immediate and strong, rolled around the holder’s hand 1 Righted in > 2 sec and Righted in ≥ 20 Slow and weak; did not roll ≤ 10 sec and < 60 sec around the holder’s hand 2 Righted in > 10 sec Did not right in < None 60 sec Each column was scored separately, then the sum of these three scores provided the composite sedation score (scale of 0-6). Sedation was defined as absent (0), mild (1-2), moderate (3-4), or profound (5-6) based on the composite score.

Table 2.2 – Scoring system for the evaluation of the level of muscle relaxation associated with midazolam in ball pythons (scale of 0-2).

SCORE LEVEL OF MUSCLE CRITERIA RELAXATION 0 None Body kept its resting shape when lifted Marked resistance to extension and immediately flexed back to resting shape 1 Moderate Body lost its resting shape when lifted, but did not extend completely Decreased resistance to extension and slow to flex back to resting shape 2 Marked Body lost its resting shape when lifted and extended completely No resistance to extension and did not flex back to resting shape

2.3.3 Flumazenil trial

A second experimental phase was opportunistically performed 2 weeks after the two first trials in which nine snakes from the previous trial received 1 mg/kg IM of midazolam followed 60 min later by 0.08 mg/kg IM of flumazenil (0.5 mg/5 mL, Fresenius Kabi Canada Ltd., Toronto, Ontario M9W 0C8, Canada) in an open trial. Both injections were in the epaxial muscles of the cranial third of the body. Evaluation of the level of sedation and muscle relaxation, HR, RR, and temperature was performed as described in the midazolam trial, including just before the flumazenil injection at T = 60 min. However, these parameters were additionally evaluated at 65, 70, 75, and 105 min

after the midazolam injection (i.e., 5, 10, 15, and 45 min after the flumazenil injection) to better assess the initial effects of flumazenil.

2.3.4 Statistical analysis

Statistical analyses were performed using a statistical software (R version 3.5.0). Scoring data, while being ordinal variables, were considered to reflect a continuous underlying process (e.g., sedation, muscular relaxation) and were also considered to be homogeneous (i.e., equal space between scores in terms of incremental magnitude of sedation) and proportional. As such, these data were analyzed using linear mixed models to perform post-hoc analysis and to investigate interaction effects. Longitudinal data analysis was performed with linear mixed modeling. Outcome variables included sedation score, muscle relaxation score, HR, RR, and temperature. Explanatory fixed variables included treatment (midazolam 1 mg/kg, midazolam 2 mg/kg), time, and treatment*time interaction. Snakes were also added as a random variable. Residual plots were used to assess linearity, homogeneity of variances, normality, and outliers. Quantile plots were also performed on the residuals by treatment groups for normality assessment. Residuals resulting from the fitted model were verified to be normally distributed and had no evidence of heteroscedasticity. A type III analysis of variance was performed on the fixed effects and post-hoc comparisons were performed using a Tukey adjustment. Values of p < 0.05 were considered significant.

2.4 Results

2.4.1 Animals

All snakes had an adequate to moderately increased amount of internal fat stores on coelomic ultrasonography. One snake presented with agitation, restlessness, hypersalivation, and frequent mouth opening behavior from 30 min to 8 h following the 2 mg/kg injection. The median sedation score during that period for this snake was 2 out of 6 (range: 0 to 4), while this score was 5 out of 6 (range: 3 to 5) in the 1 mg/kg trial for the same time period. Hypersalivation and frequent mouth opening were also observed in this snake during the 1 mg/kg trial, but agitation was much less pronounced. This snake was not identified as an outlier during the statistical analysis and was therefore kept in the data analysis. It was, however, not included in the flumazenil open trial since this snake was prone to paradoxical reaction, which was considered an undesirable side effect.

2.4.2 Midazolam trial

Order of treatment did not influence the response for any scoring outcome variables (all p > 0.05). There was a significant interaction effect of time*treatment for 42

both sedation (p = 0.023) and muscle relaxation scores (p = 0.02), suggesting differences between treatments over time. There was no statistically significant difference in sedation and muscle relaxation scores between the 1 and 2 mg/kg treatments at any time point (Figure 2.1). However, the difference from baseline as well as the duration of sedation and muscle relaxation followed a different pattern between the two dosages. When given 1 mg/kg, scores were significantly different from baseline from 15 min to 56 h (2.3 days) post-injection for sedation and from 10 min to 72 h (3 days) post-injection for muscle relaxation (all p < 0.05). When given 2 mg/kg, scores were significantly different from baseline from 10 min to 72 h (3 days) post-injection for sedation and from 10 min to 120 h (5 days) post-injection for muscle relaxation (all p < 0.05). Temperature had a mild effect on the muscle relaxation score (p = 0.0029), which decreased by a factor of 0.06 as the temperature decreased by 1°C. Temperature was therefore left in the model as a covariable. Peak effect was noted 60 min post-injection with 1 mg/kg and 60-90 min post-injection with 2 mg/kg. At this point, the average sedation and muscle relaxation scores were 4 out of 6 (moderate sedation) and 1 out of 2 (moderate muscle relaxation), respectively, with 1 mg/kg, and 4.5 out of 6 (moderate to profound sedation) and 1 out of 2 (moderate muscle relaxation), respectively, with 2 mg/kg. At the time of peak effect, 60% (6/10) of snakes could not right their body with both dosages, while 20% (2/10) and 40% (4/10) of snakes had no head reaction (i.e., head manipulation reaction score of 2 out of 2) with 1 and 2 mg/kg, respectively. The duration of action was highly variable across snakes with the maximum sedation duration (defined as the time from the injection to the first of two consecutive time points with a sedation score of 0 or 1) being 152 h (6.3 days) after the injection of 2 mg/kg of midazolam. As a comparison, the minimum duration of action with 2 mg/kg was 5 h, while the minimum and maximum with 1 mg/kg were 24 h (1 day) and 144 h (6 days), respectively. Individual sedation and muscle relaxation scores over time are provided in Appendix C, Tables C.4 and C.5, respectively. There were no significant HR differences between the two midazolam dosages over time (all p > 0.05) (Figure 2.2). However, the HR was significantly lower than baseline starting at 30 min and up to 128 h (5.3 days) for both dosages (all p < 0.05). Heart rate also significantly increased with increasing temperature by a mean ± SEM of 1.9 ± 0.4 beats per min per 1°C increase (p < 0.001). It was thus kept as a covariable in the model. Regarding the RR, there was a significant time*treatment effect (p = 0.021). While the RR was overall lower with 2 mg/kg versus 1 mg/kg, it was not statistically different over time (all p > 0.05). The RR did not significantly change from baseline with 1 mg/kg, while a few time points with 2 mg/kg had RR significantly lower than baseline (15, 45, and 60 min, as well as 8 h post-injection; all p < 0.05). Temperature had no effect on the RR. Individual HR, RR, and superficial body temperature over time are provided in Appendix C, Tables C.1, C.2, and C.3, respectively.

2.4.3 Flumazenil trial

There was a significant overall effect of time on the sedation and muscle relaxation scores (p < 0.001). Temperature had no effect on the sedation score (p = 0.75), but significantly affected the muscle relaxation score (p = 0.01) and was thus kept as a covariable in the model. Snakes had significantly higher sedation scores than baseline from 10 to 60 min post-injection (all p < 0.001) and higher muscle relaxation scores than baseline from 10 to 65 min post-injection (all p < 0.001) (Figure 2.3). From 5 min (T = 65 min) to 2 h (T = 3 h) after the flumazenil injection for the sedation and from 10 min (T = 70 min) to 6 h (T = 7 h) after the flumazenil injection for the muscle relaxation, scores were not significantly different from baseline (all p > 0.05). However, following these time points, the sedation and muscle relaxation scores were significantly different from baseline up until 80 h (3.3 days) and 104 h (4.3 days), respectively (all p < 0.05). Individual sedation and muscle relaxation scores over time are provided in Appendix C, Table C.7. There was a significant overall effect of time on the HR and RR (p < 0.001). Temperature had no effect on the RR (p = 0.61), but it significantly affected the HR (mean ± SEM increase of 1.4 ± 0.5 beats per min per 1°C increase [p = 0.012]) and was therefore kept as a covariable in the model. Snakes had significantly lower HR than baseline from 30 to 60 min post-injection (all p < 0.002) and significantly lower RR than baseline from 10 to 65 min post-injection (all p < 0.001) (Figure 2.3). From 5 min to 3 h after the flumazenil injection (T = 65 min to 4 h) for the HR and from 10 min after the flumazenil injection (T = 70 min) up until the end of the study for the RR, values were not significantly different from baseline (all p > 0.05), except for T = 105 min, 24 and 32 h when the RR was significantly lower than baseline (all p < 0.05). The HR was thereafter significantly lower than baseline from 4 h to 95 h after the injection of flumazenil (T = 5 h to 96 h) (all p < 0.05). Individual HR, RR, and superficial body temperature over time are provided in Appendix C, Table C.6.

Figure 2.3 – Mean ± SEM sedation score (scale 0-6), muscle relaxation score (scale 0-2), heart rate, and respiratory rate over time in ten ball pythons sedated with 1 mg/kg intramuscular (IM) of midazolam and reversed with 0.08 mg/kg IM of flumazenil administered 60 min later (as indicated by the vertical dotted line).

2.5 Discussion

Midazolam administered at a dosage of 1 or 2 mg/kg IM provided moderate to profound and lengthy sedation and muscle relaxation in ball pythons. Despite doubling the dosage, no statistically significant difference was observed between the two treatments in term of sedation and muscle relaxation scores, HR, and RR. Although studies evaluating high dosages of midazolam are rare, a ceiling effect for its MAC- reducing potential has been proposed (Amrein & Hetzel, 1990; Hall, Schwieger, & Hug, 1988). However, it is likely that the relationship between the observed effect and the dosage used is not linear and therefore a higher number of animals than used in the present study may have been required to detect a statistically significant difference between the two experimental groups. The sedative and muscle relaxant effects of midazolam in ball pythons were significantly different from baseline 10-15 min following an injection of 1 and 2 mg/kg IM, and the peak effect was first noted 60 min post-injection. The median onset of sedation has been reported to be 5 min in common snapping turtles (Bienzle & Boyd, 1992), 5.5 min in red-eared sliders (Oppenheim & Moon, 1995), and 10 min in tegus (Bisetto et al., 2018) and various other species of reptiles (Arnett-Chinn et al., 2016) following an IM injection. Although peak sedation has been described approximately 20 min post- injection in various reptilian species (Arnett-Chinn et al., 2016), this maximum effect was only noted after 15-60 min in estuarine crocodiles receiving 5 mg/kg IM of midazolam (Olsson & Phalen, 2013a), which is similar to ball pythons in the present study. Sedation and muscle relaxation were significantly different than baseline up until 5 days post-injection at both dosages, which is much longer than what has previously been described in boa constrictors (up to 372 ± 142.27 min; Simone et al., 2017), red- eared sliders (3-114 min; Oppenheim & Moon, 1995), and tegus (5-475 min; Bisetto et al., 2018). Midazolam is a highly lipophilic drug and in humans, obesity has been associated with a slower elimination half-life of the drug as a result of an increased volume of distribution (Brill et al., 2014; Greenblatt et al., 1984). Snakes from the current study had an adequate to moderately increased amount of internal fat stores, which could have contributed to the prolonged duration of action. However, a recent pharmacokinetic study evaluating 1 mg/kg IM of midazolam in ball pythons showed a prolonged terminal half-life (16.54 h) due to very slow clearance rather than a large volume of distribution (see chapter 4). Similar scoring systems have been effectively used to evaluate anesthetic drugs in reptiles (Bisetto et al., 2018; Doss et al., 2017). Evaluation of the sedative potential of drugs is challenging in snakes due to their natural low level of activity and highly variable reaction to external stimuli. For instance, prior to the midazolam injection, snakes took from < 1 sec to 30 sec to right their body. Therefore, the scoring system used in this study was adapted prior to the midazolam trial to reflect this variability. Despite the limitations of a semi-objective scoring system, the sedation and muscle 48

relaxation scores accurately reflected the changes in behavior observed by the researchers. The effects of anesthetic drugs have been shown to be significantly affected by temperature in reptiles (Kischinovsky et al., 2013; Olsson & Phalen, 2013b). In the current study, temperature significantly affected the muscle relaxation score and HR and was therefore kept as a covariable in the analysis of these parameters. The temperature varied between snakes and over time throughout the experiment, but was overall maintained within their optimum temperature range. This variation was likely due, at least in part, to an altered thermoregulatory behavior associated with sedation and stress. The HR of ball pythons significantly decreased 30 min post-injection compared to baseline. However, the HR of snakes can increase significantly following manipulation, which can mask the effects of a drug (James, Williams, Wang, & Bertelsen, 2018). In the current study, snakes were handled for every HR measurement, which could have triggered stress-related tachycardia. Although a true bradycardic effect cannot be confirmed, it is likely that midazolam decreased or limited the tachycardic effect of manipulations. In a study evaluating the HR of Burmese pythons (Python molurus) with the use of electrocardiogram adhesive electrodes, which limited snake manipulation to the time of injection, midazolam (1 mg/kg IM) decreased the HR by ~60%, while a saline injection did not affect the HR (Lopes et al., 2017). Respiratory depression associated with the use of IM midazolam has been described in tegus (Bisetto et al., 2018), goats (Stegmann, 1998), pigs (Lacoste et al., 2000), and humans (Reves et al., 1985). In the current study, even though the RR was significantly lower than baseline at four time points in the 2 mg/kg trial, this effect was not maintained over time. Hence, ball pythons do not appear to be especially susceptible to bradypnea associated with the use of midazolam. A single snake showed clinical signs consistent with paradoxical excitation following midazolam injection at both dosages. Agitation and restlessness have been frequently described with the use of midazolam in cats (Ilkiw, Suter, Farver, et al., 1996; Ilkiw, Suter, McNeal, Farver, & Steffey, 1996) and occasionally in humans (Robin & Trieger, 2002; Tae et al., 2014), dogs (Court & Greenblatt, 1992; Simon et al., 2014), sheep (Simon et al., 2017), and horses (Hubbell et al., 2013). However, this effect is usually transient and is less commonly observed with lower dosages, which is consistent with the decreased intensity of clinical signs observed at 1 mg/kg compared to 2 mg/kg in the abovementioned ball python (Court & Greenblatt, 1992; Simon et al., 2017). Hypersalivation has also been associated with the use of midazolam in dogs (Court & Greenblatt, 1992) and sheep (Simon et al., 2017). The pathophysiology of paradoxical excitation is not fully understood, but several mechanisms have been proposed in human patients, including a central anticholinergic effect leading to delirium, hallucinations, and disorientation, alteration of the serotonergic balance 49

leading to rage and aggression, and individual mutations of the GABA receptors (Robin & Trieger, 2002). Although the metabolism of numerous anesthetic drugs can be affected by gender in humans, this does not seem to be the case for midazolam (Holazo, Winkler, & Patel, 1988). Some midazolam studies in humans demonstrated deeper sedation in men than women, but this effect was typically considered minimal (Nishiyama, Matsukawa, & Hanaoka, 1998; Sun, Hsu, Chia, Chen, & Shaw, 2008). There is currently no evidence of such effects in veterinary medicine (Ilkiw, Suter, Farver, et al., 1996). Nevertheless, we did not evaluate the effect of sex in the current study, as all snakes were female due to availability. Thus, care should be taken when extrapolating results from this study to male ball pythons. The administration of flumazenil reversed the sedation, muscle relaxation, and reduced HR associated with the use of midazolam in snakes from the present study. However, all snakes became fully resedated within 3 h and the HR decreased again 4 h following the flumazenil injection. The flumazenil dosage used, 0.08 mg/kg, was based on a 1:13 midazolam-to-flumazenil dosage ratio described in dogs (Lemke, 2007). The elimination half-life of flumazenil is reported to follow a similar pharmacokinetic profile to midazolam in dogs (Lemke, 2007) and humans (Amrein & Hetzel, 1990). In reptiles, the use of flumazenil has been reported at empirical dosages ranging from 0.01 to 0.05 mg/kg IM or subcutaneous and resedation has not yet been reported, although this could be due to a lack of long-term monitoring following reversal (Arnett-Chinn et al., 2016; Doss et al., 2017; Mans & Sladky, 2012; Yaw et al., 2017). Additional studies are needed to evaluate the origin of the discrepancy between the duration of action of flumazenil and midazolam in ball pythons and to determine the effects and safety of a higher flumazenil dosage. No adverse effects were observed with the use of flumazenil in the current study. In humans, adverse reactions are rare, but can include transient facial erythema, anxiety, and vertigo (Votey et al., 1991). Results from the present flumazenil trial should be interpreted with care due to the open nature of the experiment.

2.6 Conclusions

Midazolam administered at a dosage of 1 and 2 mg/kg IM in the cranial third of ball pythons provided moderate to profound sedation and muscle relaxation for 3 to 5 days with a peak effect 60 min post-injection. Side effects included a reduced HR and less commonly, paradoxical excitation. A dosage of 1 mg/kg IM or lower if used in combination with other anesthetic drugs is preferable considering the absence of significantly deeper sedation at higher dosages. Flumazenil can be safely used to reverse the effects of up to 2 mg/kg IM midazolam, but resedation should be expected with a flumazenil dosage of 0.08 mg/kg IM.

Acknowledgements Funding was provided by the Toronto Zoo. The authors acknowledge the veterinary technicians, veterinarians, and animal keepers in the Wildlife Health Branch of the Toronto Zoo for their technical support and for taking care of these ball pythons during the study.

3 The effects of midazolam and nitrous oxide on the minimum anesthetic concentration of isoflurane in the ball python (Python regius)

Cédric B. Larouche,a,b Craig Mosley,c Hugues Beaufrère,d and Christopher Duttona

Affiliations aToronto Zoo, 361A Old Finch Avenue, Scarborough, ON, M1B 5K7, Canada bDepartment of Pathobiology, Ontario Veterinary College, University of Guelph, 50 Stone Road E., Guelph, ON, N1G 2W1, Canada cVCA Canada, 404 Veterinary Emergency and Referral Hospital, 510 Harry Walker Parkway S., Newmarket, ON, L3Y 0B3, Canada dDepartment of Clinical Studies, Ontario Veterinary College, University of Guelph, 50 Stone Road E., Guelph, ON, N1G 2W1, Canada

Author’s contributions

CBL: conception and design, data acquisition, management, and interpretation, review of videos, and preparation of manuscript CM: conception and design, data interpretation, blinded review of videos, and manuscript revision HB: study design, statistical analysis, data interpretation, and manuscript revision CD: study conception, data interpretation, and manuscript revision

Larouche, C. B., Mosley, C., Beaufrère, H., & Dutton, C. The effects of midazolam and nitrous oxide on the minimum anesthetic concentration of isoflurane in the ball python (Python regius). Veterinary Anaesthesia and Analgesia, In Press. http://doi.org/10.1016/j.vaa.2019.08.002 (Accepted for publication on 2019-08-11) 52

3.1 Abstract

Objective: To evaluate the effects of midazolam and nitrous oxide (N2O) on the minimum anesthetic concentration of isoflurane (MACIso) in ball pythons. Study design: Prospective, crossover, randomized, semi-blinded study. Animals: A total of nine healthy adult female ball pythons (Python regius) weighing 2.76 ± 0.73 kg. Methods: Three protocols were evaluated in each snake with 2-week washouts: treatment MID-O2, midazolam (1 mg/kg) administered intramuscularly (IM) and anesthesia induced with isoflurane-oxygen; treatment SAL-O2, saline (0.2 mL/kg) IM and anesthesia with isoflurane-oxygen; and treatment SAL-N2O, saline IM and anesthesia with isoflurane and 50% nitrous oxide (N2O): 50% oxygen. In each treatment, isoflurane was administered by face mask immediately after premedication. Snakes were endotracheally intubated and inspired and end-tidal isoflurane concentrations were monitored. The study design followed a standard bracketing technique and the MACIso was determined by logistic regression. Electrical stimulation using a Grass stimulator connected to the base of the tail (50V, 50 Hz, 6.5 millisecond pulses) was used as the supramaximal stimulus. Blood gas analysis was performed on cardiac blood collected immediately following intubation and after the last stimulation. Blood gas variables were compared over time and between treatments using linear mixed models.

Results: MACIso at a body temperature of 30.1 ± 0.4°C was 1.11% (95% confidence interval, 0.94-1.28%) in SAL-O2 and was significantly decreased to 0.48% (0.29-0.67%) in MID-O2 (p < 0.001) and to 0.92% (0.74-1.09%) in SAL-N2O (p = 0.016). Heart rate (by 2 beats/min) and PO2 were significantly lower in MID-O2 and SAL-N2O than in SAL-O2.

Conclusions and clinical relevance: Midazolam significantly decreased the MACIso by 57% in ball pythons, whereas addition of N2O resulted in a modest, though significant, decrease. The MACIso in ball pythons was lower than those previously reported in reptilian species.

3.2 Introduction

Anatomical and physiological characteristics of reptiles such as cardiovascular shunting, ectothermy, and prolonged breath-holding capacity can significantly impact the predictability and perceived benefits of isoflurane anesthesia. Like mammals, reptiles are susceptible to anesthetic-induced adverse effects, including respiratory depression and apnea, bradycardia, and systemic hypotension (Bertelsen, 2014; Chinnadurai et al., 2010; Hernandez-Divers et al., 2005; C. A. E. Mosley et al., 2004). The potency of an inhalant anesthetic is measured as the minimum concentration of the agent in the alveoli (minimum alveolar concentration) that will prevent gross purposeful movement in response to a supramaximal stimulus in 50% of the population at standard barometric pressure (760 mmHg) (Eger, Saidman, et al., 1965). Snakes have no alveoli; therefore, this measure will be henceforth referred to as the minimum anesthetic concentration (MAC). The apparent MAC of isoflurane (MACIso) is highly variable across reptilian species, ranging from 1.54 ± 0.17% in Dumeril’s monitors (Varanus dumerili; Bertelsen et al., 2005c) to 3.2 ± 0.4% in red-footed tortoises (Chelonoidis carbonaria; Greunz et al., 2018). Midazolam causes moderate to marked sedation and muscle relaxation in ball pythons (Python regius; see Chapter 2) and boa constrictors (Boa constrictor; Simone et al., 2017). Reported adverse effects include bradycardia in Burmese pythons (Python molurus; Lopes et al., 2017) and paradoxical excitation in a ball python (see Chapter 2). Although midazolam has MACIso-sparing effect in dogs and rats (Kropf & Hughes, 2018; Schwieger et al., 1994; Seddighi et al., 2011), evaluation of this effect in non- mammalian species has not been reported in the literature.

The MAC of nitrous oxide (N2O) is too high to achieve induction of anesthesia in veterinary species when administered alone (Bertelsen et al., 2005a; Steffey et al., 1974). However, N2O may be chosen as part of a protocol for two reasons: first, the low blood gas solubility ensures rapid uptake into the pulmonary circulation creating a situation where, for a short time, the uptake of a concurrently administered halogenated gas is increased (second gas effect; Epstein et al., 1964); second, N2O may provide some analgesia when administered as an adjunct agent (Bovill, 2008). In reptiles, the MAC of sevoflurane was decreased by the addition of N2O in Dumeril’s monitors (Bertelsen et al., 2005a).

The objectives of this study were to determine the MACIso in ball pythons and to evaluate the effects of pretreatment with midazolam or concurrent administration of 50% N2O. We hypothesized that the administration of midazolam (1 mg/kg) intramuscularly (IM) and the delivery of isoflurane in a 50:50 mixture of N2O and oxygen (O2) would each result in a significant reduction of MACIso in ball pythons.

3.3 Material and methods

This study was approved by the Animal Care and Research Committees of the University of Guelph (#3812) and Toronto Zoo (#2017-11-02). Experimentation was performed in the Wildlife Health Center of the Toronto Zoo.

3.3.1 Animals

A group of nine adult female captive-bred ball pythons with a mean ± standard deviation (SD) weight of 2.76 ± 0.73 kg were studied. Each snake was individually housed in a plastic terrarium with constant access to a hiding spot, a water bowl, and at a temperature gradient of 26-35°C. Humidity was maintained at 40-60% and a 12:12 hours of light:dark cycle was provided. Each snake was offered a frozen-thawed rat or mouse every other week and a minimum fasting period of 7 days was used before any experimentation to reduce the effects of metabolic changes associated with digestion in snakes (Secor, Hicks, & Bennett, 2000). Snakes were deemed healthy based on physical examination, coelomic ultrasonography, and evaluation for fecal parasites. All snakes had adequate to moderately increased internal fat stores based on ultrasonographic evaluation. Each ball python was individually identified using a subcutaneous microchip (Trovan ISO FDX-B; Eidap Inc., ON, Canada) that was scanned prior to any experimentation. Eight of these ball pythons had been used for a noninvasive study of the pharmacodynamics of midazolam two months before the current study (Appendix B, Table B.2).

3.3.2 Study design and anesthesia

A semi-blinded, randomized, crossover trial was used to determine the MACIso with three drug combinations: treatment MID-O2, midazolam (1 mg/kg; 5 mg/mL; Sandoz, QC, Canada) administered IM immediately before induction of anesthesia with isoflurane (IsoFlo; Abbott Laboratories, IL, USA) in O2; treatment SAL-O2, saline (0.2 mL/kg; 0.9% sodium chloride; ICU Medical Inc., CA, USA) IM followed by induction of anesthesia with isoflurane in O2; and treatment SAL-N2O, administration of saline as described for treatment SAL-O2 and induction of anesthesia with isoflurane in a 50:50 mixture of N2O and O2 using a premixed cylinder (ALnox; Air Liquide Canada Inc., QC, Canada). Each snake was anesthetized three times with 2 weeks between anesthetic episodes. The order of snakes and treatments was randomized using replicated latin squares generated with a statistical software (R Foundation for Statistical Computing, Austria). Snakes were weighed before each experiment. Drugs for injection were prepared by a veterinarian not involved in the project and injected in the right cranial third of the epaxial muscles by an investigator unaware of the treatment assigned. The use of N2O was not concealed from investigators. Isoflurane was delivered to the snake by placing its head in a tight-fitted face mask with a rubber diaphragm and connected to a circle 55

rebreathing system (Anesthesia WorkStation; Hallowell EMC, MA, USA) at a vaporizer (T3ISO; Benson Medical Industries Inc., ON, Canada) setting of 5% isoflurane in a gas flow of 1 L/min (100% O2 or 50% N2O and 50% O2 according to the treatment). Once the snake was sufficiently relaxed, a 10-gauge x 7.6 cm catheter (Angiocath IV catheter; BD Canada, ON, Canada) that had been modified for endotracheal intubation by cutting, beveling, and smoothing the tip, was inserted in the trachea, secured in place with adhesive tape, attached to the anesthesia circuit, and the fit was verified by inflating the lungs with an airway pressure of 12 cmH2O while listening for leaks. The gas flow was then decreased to 200 mL/kg/min. All animals were mechanically ventilated (Anesthesia Workstation) at 5 breaths/min with a tidal volume of 25 mL/kg/breath based on results of a recent study (Jakobsen, Williams, Wang, & Bertelsen, 2017). Maximum pressure was 12 cm H2O. Physiological variables were recorded immediately prior to supramaximal stimulation, and gas concentrations were recorded every 5 minutes. Heart rate (HR) and rhythm were monitored using two devices: a Doppler probe (Model 811-B; Parks Medical Electronics Inc., OR, USA) placed over the ventral aspect of the heart and a three-lead ECG (Waveline Pro; DRE Inc., KY, USA) connected to 25-gauge hypodermic needles placed subcutaneously in a lead II Einthoven’s triangle (M. A. Mitchell, 2009). Temperature was monitored using a probe (Waveline Pro) inserted 30-50 cm into the esophagus. Body temperature was maintained at 30.1 ± 0.4°C throughout the experiment using an electric heating pad and a heating lamp. A 3.5 Fr urinary catheter (Argyle; Covidien, MA, USA) connected to an airway gas analyzer (Waveline Pro with built-in Dräger anesthetic agent module) with a continuous sampling rate of 50 mL/min was inserted through an airway gas sampling adapter into the endotracheal tube and positioned to reach the tip of the endotracheal tube. The Dräger gas module retains the anesthetic gas factory calibration for the lifetime of the module, according to the manufacturer’s instructions. Measurements included the end-tidal partial pressure of carbon dioxide, inspired (FIO2) and end-tidal O2 concentration, inspired (FIIso) and end- tidal (FE′Iso) isoflurane concentration, and inspired and end-tidal N2O concentration.

Immediately after endotracheal intubation and again at completion of MACIso determinations, blood was collected by cardiocentesis using a 25-gauge needle attached to a 1 mL syringe. Analysis was performed within 2 min for pH, partial pressure of carbon dioxide (PCO2) and O2 (PO2), hematocrit, and sodium, potassium, ionized calcium, and glucose concentrations (i-STAT 1 with CG8+ cartridge; Abaxis, CA, USA). Values calculated by the analyzer included base excess (BE), oxygen saturation (SO2), and bicarbonate (HCO3) and hemoglobin concentrations.

3.3.3 Determination of MAC

The trial design was based on a bracketing technique with electrical stimulation as a supramaximal stimulus (Eger, Saidman, et al., 1965). One 25-gauge needle was

inserted in the subcutis of each side of the tail at the junction of the ventral and lateral scales 3-5 cm caudal to the vent. The position of the needles was changed between each stimulation to avoid desensitization (Barter et al., 2006). A Grass stimulator (S48 Stimulator; Astro-Med Inc., RI, USA) was connected to the needles and delivered 50 V at 50 cycles/sec with a duration of 6.5 millisecond/pulse. Five consecutive stimulations separated by 5 sec intervals were applied as follows: two single pulses followed by three continuous 3 sec stimulations. Following intubation, the vaporizer setting was decreased to provide a concentration above the expected MACIso. To avoid excessively long anesthesia, the initial FIIso varied among snakes and treatments (range: 1.00-1.77%), based on MACIso obtained in other snakes from the same study. Isoflurane was considered to be in equilibration between the lungs and blood when the following two criteria were met for 30 min: (1) the FE′Iso / FIIso ratio was >0.9 and (2) the FE′Iso did not vary by more than 0.02%. While and after the stimulus was applied, the snake was monitored and video recorded for 60 sec to identify the presence of a purposeful movement, such as a movement of the tail, body, or head that was progressive and rhythmic during the stimulation or within 30 sec of the last stimulus. A tetanic or spastic movement was not considered purposeful. In the absence of a reaction, the FIIso was decreased by 10%. If a purposeful movement was noted, the FIIso was increased by 10%. The isoflurane concentration was then allowed to re-equilibrate prior to subsequent stimulation. The trial was considered complete when a negative reaction was followed by a positive reaction after an FIIso decrease. Video recordings were reviewed by an investigator (Mosley) that was unaware of the treatment administered after the study to confirm the presence or absence of a purposeful movement.

All snakes were ventilated with 100% O2 during the initial phase of recovery. When a first voluntary body movement was noted, snakes were then ventilated with ambient air until either spontaneous breathing or signs of intolerance to the endotracheal tube were noted. All snakes were administered meloxicam (0.2 mg/kg; Rheumocam Injection; Merck Animal Health, QC, Canada) subcutaneously during recovery to decrease inflammation associated with electrical stimulation.

3.3.4 Statistical analysis

Sample size was calculated using an online tool (Kohn et al., 2018). The SD of th the MACIso was estimated at 0.48%, which represents the 80 percentile of the SD from similar studies in reptiles (Barter et al., 2006; Bertelsen et al., 2005c; Greunz et al., 2018; C. A. E. Mosley et al., 2003a). A significant effect size was determined as a difference of FE′Iso of 0.75% between two treatments. With alpha and beta errors of 0.05 and 0.2, respectively, a sample size of eight animals per treatment was required. An additional individual was added in case an animal had to be excluded from the study.

A statistical software (R Foundation for Statistical Computing) was used for statistical analysis. The probability of no response was modelled using a mixed logistic model with FE′Iso, treatment, temperature, and anesthesia time as fixed effects and snakes as a random effect. The MACIso (FE′Iso providing a 50% probability of no response) was obtained from a simplification of the model as logit (P = 0.5) = 0 using equation 3.1, where P is the probability of no movement. Likewise, the FE′Iso providing a 95% probability of no movement (EC95) was directly obtained from the model using equation 3.2. To calculate 95% confidence intervals (CI), the variance of the MACIso and EC95 were obtained using the formula for the variance of a ratio of random variables illustrated in equation 3.3. Continuous variables (HR and blood-gas values) were analyzed using linear mixed models with treatment, temperature, anesthesia time, and interactions as fixed effects, and snakes as a random effect. Statistical significance was set at p < 0.05.

0.5 Equation 3.1: If logit(P = 0.5) = ln = 0 0.5

And if logit(P) = Intercept + ParameterIso × FE′Iso

Then MAC = −Intercept/ParameterIso 0.95 Equation 3.2: If logit(P = 0.95) = ln = 2.94 0.05

And if logit(P) = Intercept + ParameterIso × FE′Iso

Then EC95 = (2.94 − Intercept)/ParameterIso 2 2 2 2 μA σA Cov(A,B) σB Equation 3.3: σA/B = 2 ( 2 − 2 + 2 ) μB µA µAµB µB

3.4 Results

Body temperature (p = 0.061) and duration of anesthesia (p = 0.9) did not significantly influence the response to the electrical stimulus in the range of the study. The MACIso in SAL-O2 was 1.11% (95% CI, 0.94-1.28%). Midazolam and N2O resulted in statistically significant reductions of the odds of response to the supramaximal stimulus. MACIso was decreased by 57% in MID-O2 (p < 0.001) and 17% in SAL-N2O (p = 0.016). MACIso in MID-O2 and SAL-N2O were 0.48% (95% CI, 0.29-0.67%) and 0.92% (95% CI, 0.74-1.09%), respectively. The EC95 of isoflurane was 1.43% (95% CI, 1.26- 1.60%) in SAL-O2, 0.79% (95% CI, 0.60-0.98%) in MID-O2, and 1.23% (95% CI, 1.06- 1.40%) in SAL–N2O. Individual responses to supramaximal stimulations at different FE′Iso are reported in Appendix D, Figure D.1 and Table D.1. Assuming that N2O and isoflurane follow a relationship of simple linear additivity, the MAC of N2O was estimated at 289% in ball pythons using Equation 1.7 (Steffey et al., 1974).

The mean ± SD time from induction to intubation and from induction to the first supramaximal stimulation were 18.7 ± 4.0 min and 135.5 ± 16.6 min, respectively, for all treatments combined (Appendix D, Table D.2). The subsequent equilibration times, defined as the times between two supramaximal stimulations, were 59.2 ± 11.5 min for all treatments combined. Mean anesthetic time was 389.9 ± 113.6 min for all snakes combined with an average of 5 ± 2 electrical stimulations (range, 2-10) required to determine MACIso. Equilibration time between stimulations was approximately 7.1 min shorter in MID-O2 compared to SAL-O2 (p = 0.01). No major adverse effects were noted throughout the experiment. All snakes recovered uneventfully from the anesthesia and were returned to their respective terrarium.

HR was 22 ± 4 beats/min throughout the procedure in SAL-O2 and was significantly higher in MID-O2 and SAL-N2O, resulting in an increase of 2 beats/min (p < 0.001) on average versus isoflurane alone. There was no significant temperature*HR interaction in the range of this study (p = 0.52). Anesthesia time had a slight negative effect on the HR (–0.3 beats/min/h, p = 0.047). Results from blood gases are reported in Table 3.1, and individual values are reported in Appendix D, Table D.3. No significant time*treatment interaction (p = 0.36) was detected and, therefore, both variables were analyzed separately. The pH (p < 0.001), SO2 (p = 0.032), BE (p < 0.001), HCO3 (p = 0.038), and glucose (p < 0.001) were significantly increased over time, whereas the PCO2 (p < 0.01), hematocrit (p < 0.001), potassium (p < 0.001), and ionized calcium (p = 0.028) were significantly decreased over time. Compared to SAL-O2, the PO2 (p < 0.001), BE (p = 0.019), and HCO3 (p = 0.014) were significantly lower in SAL-N2O, the hematocrit (p = 0.003) was significantly higher in SAL-N2O, and the PO2 (p = 0.037) was significantly lower in MID-O2.

SAL-O2 MID-O2 SAL-N2O VARIABLE UNIT T0 TMAC T0 TMAC T0 TMAC pH* 7.104 ± 0.211 7.652 ± 0.032 7.175 ± 0.089 7.652 ± 0.043 7.056 ± 0.150 7.595 ± 0.076 mmHg 46.4 ± 17.0 16.1 ± 2.0 42.5 ± 8.7 14.4 ± 2.5 49.7 ± 16.3 14.1 ± 2.2 PCO2* kPa 6.2 ± 2.3 2.1 ± 0.3 5.7 ± 1.2 1.9 ± 0.3 6.6 ± 2.2 1.9 ± 0.3 mmHg 403 ± 165 361 ± 214 309 ± 227† 238 ± 196† 138 ± 70† 130 ± 62† PO2 kPa 54 ± 22 48 ± 29 41 ± 30† 32 ± 26† 18 ± 9† 17 ± 8† BE* mmol/L -16 ± 6 -3 ± 3 -13 ± 2 -5 ± 4 -17 ± 4† -8 ± 3†

SO2* % 97 ± 9 99 ± 2 96 ± 5 99 ± 1 90 ± 11 98 ± 3 † † HCO3* mmol/L 13.9 ± 2.8 17.9 ± 2.8 15.4 ± 1.5 16.1 ± 3.6 13.4 ± 2.5 13.6 ± 2.1 Hct* % 24 ± 3 20 ± 2 23 ± 4 20 ± 3 26 ± 3† 22 ± 3† Hb g/L 82 ± 12 68 ± 8 78 ± 13 67 ± 11 88 ± 10 74 ± 9 Na mmol/L 146 ± 4 148 ± 2 145 ± 4 148 ± 3 149 ± 3 147 ± 5 K* mmol/L 3.7 ± 0.2 3.5 ± 2 3.5 ± 0.3 3.4 ± 0.3 3.7 ± 0.2 3.3 ± 0.2 iCa mmol/L 1.65 ± 0.12 1.61 ± 0.08 1.66 ± 0.21 1.64 ± 0.18 1.72 ± 0.09 1.52 ± 0.18 Glucose mmol/L 2.1 ± 0.3 2.6 ± 0.4 2.0 ± 0.2 2.3 ± 0.4 2.3 ± 0.4 2.6 ± 0.4

PCO2: partial pressure of carbon dioxide; PO2: partial pressure of oxygen; BE: base excess; SO2: oxygen saturation; HCO3: bicarbonate; Hb: hemoglobin; Na: sodium; K: potassium; iCa: ionized calcium; no significant time*treatment interaction (p = 0.36) was detected and, therefore, both † variables were analyzed separately; *significant difference between T0 and TMAC for all treatments combined (p < 0.05); significant difference between treatment and SAL-O2 for both times combined (p < 0.05).

3.5 Discussion

The administration of 1 mg/kg IM of midazolam and the delivery of isoflurane in 50% N2O and 50% O2 significantly decreased the MACIso and the EC95 of isoflurane in ball pythons. The MACIso-sparing effect of midazolam was much more pronounced than previous reports in dogs (11-32% decrease; Kropf & Hughes, 2018; Seddighi et al., 2011) and similar to intrathecal administration in rats (up to 53% decrease; Schwieger et al., 1994). The duration of the sedative effect of 1 mg/kg IM of midazolam was markedly longer (2.3 days) in ball pythons compared with other reptiles based on a recent pharmacodynamic study (see chapter 2). Therefore, the effect of midazolam was likely sustained regardless of the duration of anesthesia, which is supported by the absence of a significant effect of the duration of anesthesia on the response.

The MAC-sparing effect of N2O on halogenated gases has been demonstrated in numerous species, including humans, dogs, rats, swine, and Dumeril’s monitors (Bertelsen et al., 2005a; Murray, Mehta, & Forbes, 1991; Russell & Graybeal, 1998; Tranquilli, Thurmon, & Benson, 1985; Voulgaris et al., 2013). In the current study, the use of 50% N2O and 50% O2 resulted in a modest decrease of the MACIso. Higher concentrations of N2O were not studied because of difficulties in adapting a second gas flowmeter to the anesthetic workstation. The calculated MAC of N2O (289%) was slightly higher in ball pythons than in Dumeril’s monitors (244%; Bertelsen et al., 2005a); values that support the use of this gas as an adjunct agent rather than a sole anesthetic in reptiles.

The apparent MACIso in ball pythons was lower than those in other reptilian species such as the green iguana (1.8 ± 0.3%, 2.0 ± 0.6%, and 2.1 ± 0.6%; Barter et al., 2006; C. A. E. Mosley et al., 2003a), the Dumeril’s monitor (1.54 ± 0.17%; Bertelsen et al., 2005c), and the red-footed tortoise (3.2 ± 0.4%; Greunz et al., 2018). Most MAC studies in reptiles assumed that gas equilibration is reached within 20-30 minutes or when the FE′Iso is stable for 15 to 20 min (Barter et al., 2006; Bertelsen et al., 2005c; Greunz et al., 2018; C. A. E. Mosley et al., 2003a). However, right-to-left intracardiac blood shunting significantly limits gas exchanges within the lungs of reptiles. By eliminating this shunt with the use of atropine, the apparent MACIso significantly decreased to a value likely closer to the real MACIso in red-footed tortoises (2.2 ± 0.3%; Greunz et al., 2018). Prolonged anesthesia has been associated with lower apparent MACIso values in a study in green iguanas, which likely resulted from variations in blood shunting and blood flow to the central nervous system according to the authors (Barter et al., 2006). The longer equilibration times used in the current study may have resulted in a more accurate estimate of the actual MACIso of ball pythons by allowing more time for gas equilibration to occur, even in the presence of intracardiac shunting. Therefore, a higher apparent MACIso would be expected with the use of shorter equilibration times in ball pythons.

The use of electrical stimulation has been previously used as a supramaximal stimulus in reptile MAC studies (Barter et al., 2006; Bertelsen et al., 2005c; C. A. E. Mosley et al., 2003a). In a pilot study performed on three ball pythons, we identified clamping of the tail to be a submaximal noxious stimulus since the apparent MACIso was lower with clamping than electrical stimulation, such as previously reported in the green iguana (C. A. E. Mosley et al., 2003a). A 20 V electrical stimulus was used successfully in green iguanas (Barter et al., 2006; C. A. E. Mosley et al., 2003a) but a voltage of 50 V was needed in Dumeril’s monitors to obtain supramaximal stimulation (Bertelsen et al., 2005c). Although 50 V was used in the current study, the effects of various electrical intensities were not compared. The electrical stimulation protocol (two single followed by two continuous stimulations) was adapted from a study in dogs and rabbits (Valverde et al., 2003). This was modified by the addition of a third continuous electrical stimulus because this third stimulation was necessary to elicit a purposeful movement in one snake from the pilot study. Spontaneous movements were rarely observed 1-5 min following stimulation, or at any other moment between stimulations. These movements were not considered a direct reaction to the stimulus as they occurred >1 min after stimulation during FE′Iso ≤ 0.86%, with one exception in the SAL-N2O treatment (FE′Iso = 1.16%). Therefore, although the MACIso is significantly decreased with midazolam and N2O, spontaneous movements can occasionally occur with low isoflurane concentrations, even if maintained above the MACIso. Physiological derangements such as hypercarbia, hypoxia, and metabolic acidosis have been shown to affect the MAC in humans (Aranake et al., 2013). The point-of-care portable blood analyzer used in the current study has not been validated in ball pythons and results are automatically corrected to 37°C. The solubility of carbon dioxide and O2 decreases with increasing temperature, leading to higher partial pressures and a decreased pH (Malte, Jakobsen, & Wang, 2014). Interpretation of blood gases from the current study should therefore be done with care. Methods to correct these values are available (Malte et al., 2014). Blood was collected by intracardiac puncture, which can result in sampling of a mixture of venous and arterial blood. Some values such as pH, PCO2, and PO2 therefore may not reflect actual arterial values. The blood pH increased over time with all treatments. This change may be a reflection of the abnormally low initial value. A study of the mechanical ventilation of ball pythons suggested that manual restraint for anesthetic induction can cause metabolic acidosis (Jakobsen et al., 2017), which is consistent with the marked base deficit (metabolic acidosis) noted in the first sample of the current study. Hypoventilation due to bradypnea or apnea prior to intubation was likely responsible for the initially high PCO2, contributing to the initially low pH (respiratory acidosis). Blood pH and PCO2 at the end of the procedure were more closely related to values reported in non-anesthetized

ball pythons (pH, 7.53 ± 0.07; PCO2, 2.55 ± 0.54 kPa; Jakobsen et al., 2017), and reference to these “normal” values suggests that mild alkalemia and hypocarbia (respiratory alkalosis) were present at the time of MACIso determination in the current study. The PO2 was significantly lower in MID-O2 and SAL-N2O than SAL-O2. A lower FIO2 associated with the use of N2O would decrease the PO2, such as reported in the Dumeril’s monitor (Bertelsen et al., 2005a). By mechanisms increasing cardiac cholinergic tone and, to a lesser extent, decreasing cardiac adrenergic tone, midazolam decreases the HR of Burmese pythons (Lopes et al., 2017). Whereas cholinergic stimulation can increase right-to-left intracardiac blood shunting in turtles (Hicks, 1994), we hypothesize that the indirect effects of midazolam on the cardiac function of ball pythons could have contributed to slightly decrease PO2. However, the difference in PO2 between treatments SAL-O2 and MID-O2 was not clinically relevant since both mean values would have provided full hemoglobin oxygen saturation. SO2 but not PO2 significantly increased over time. However, the equation used by the blood gas analyzer to calculate SO2 may not be accurate in ball pythons. Aberrantly low individual SO2 values were observed after induction of anesthesia and these values were always associated with marked acidemia (e.g., SO2 of 69% reported in one snake with a PO2 of 64 mmHg and a pH of 6.844). Potassium, ionized calcium, glucose, and the hematocrit significantly decreased over time but there was no difference among treatments and the variation was unlikely to be biologically significant. Although statistically significant, the HR difference between SAL-O2 and the two other treatments was not clinically relevant. There is currently no scientific evidence that sex significantly influences the MAC of anesthetic gases (Aranake et al., 2013; Quasha et al., 1980) or the effects of midazolam (Holazo et al., 1988; Ilkiw, Suter, Farver, et al., 1996; Nishiyama et al., 1998; Sun et al., 2008). However, the effect of sex has not been evaluated in snakes and therefore extrapolation of results from the current study to male ball pythons should be done with care. Some snakes from the current study had a high amount of internal fat stores. Although obesity has been shown to result in a larger volume of distribution of midazolam in humans (Brill et al., 2014; Greenblatt et al., 1984), no studies have demonstrated a significant effect of obesity on the MAC (Aranake et al., 2013). There is a positive linear relationship between temperature and MAC in humans, rats, and desert iguanas (Dipsosaurus dorsalis) (Aranake et al., 2013; Dohm & Brunson, 1998; Vitez et al., 1974). Therefore, MACIso values reported in the current study should only be expected in ball pythons within the reported body temperature range (30.1 ± 0.4°C). The absence of blinding regarding the use of N2O is a limitation of the current study. Video recording of supramaximal stimulations were reviewed by an investigator unaware of the treatment administered to decrease this potential source of bias. Finally, the variability of the initial FIIso could have biased the apparent MACIso, although the duration of anesthesia did not significantly influence MACIso and the SD of the MACIso remained relatively small.

3.6 Conclusions

This study demonstrated that midazolam (1 mg/kg) administered IM in the cranial epaxial muscles of ball pythons significantly decreased the MACIso by 57%. The delivery of isoflurane with 50% N2O resulted in a modest, but significant decrease of the MACIso by 17%. No clinically relevant differences in blood gases and HR resulted from the administration of midazolam and 50% N2O compared to saline and 100% O2, respectively. However, anesthetic induction with isoflurane delivered via a face mask resulted in marked acidemia that resolved over time with mechanical ventilation.

Acknowledgements Funding of this project was provided by the Toronto Zoo and the Andrea Leger Dunbar Summer Research Studentship. The authors acknowledge Kristopher Afshaun Zaman for his substantial contribution to the data collection and organization, all the keepers, veterinary technicians, and veterinarians from the Wildlife Health Branch of the Toronto Zoo for taking care of these snakes, as well as Dr. Nicole Nemeth and Dr. Dorothee Bienzle for their support as residency advisors.

4 Pharmacokinetics of midazolam and its major metabolite 1-hydroxymidazolam in the ball python (Python regius) after intracardiac and intramuscular administrations

Cédric B. Larouche,1,2 Ron Johnson,3 Francis Beaudry,4 Craig Mosley,5 Yu Gu,3 Kristopher Afshaun Zaman,1 Hugues Beaufrère,6 and Christopher Dutton2

Affiliations 1Department of Pathobiology, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada 2Toronto Zoo, Toronto, Ontario, Canada 3Department of Biomedical Sciences, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada 4Groupe de Recherche en Pharmacologie Animale du Québec (GREPAQ), Département de Biomédecine Vétérinaire, Faculté de Médecine Vétérinaire, Université de Montréal, Saint-Hyacinthe, Québec, Canada 5VCA Canada, 404 Veterinary Emergency and Referral Hospital, Newmarket, Ontario, Canada 6Department of Clinical Studies, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada

Author’s contributions All listed authors contributed to the data analysis and read, critically revised, and approved the final manuscript. CBL, RJ, YG, HB, and CD contributed to the conception and design. CBL, CM, KAZ, and CD contributed to the acquisition of data. YG, RJ, and FB planned and performed the midazolam assay. FB performed the pharmacokinetic analysis.

Larouche, C. B., Johnson, R., Beaudry, F., Mosley, C., Gu, Y., Zaman, K. A., … Dutton, C. Pharmacokinetics of midazolam and its major metabolite 1-hydroxymidazolam in the ball python (Python regius) after intracardiac and intramuscular administrations. Journal of Veterinary Pharmacology and Therapeutics, In Press. https://doi.org/10.1111/jvp.12806 (Accepted for publication on 2019-07-30)

4.1 Abstract

Midazolam is a benzodiazepine with sedative, muscle relaxant, anxiolytic, and anticonvulsant effects. Twelve ball pythons (Python regius) were used in a parallel study evaluating the pharmacokinetics of 1 mg/kg midazolam following single intracardiac (IC) or intramuscular (IM) administration. Blood was collected from a central venous catheter placed 7 days prior, or by cardiocentesis, at 15 time points starting just prior to and up to 72 h after drug administration. Plasma concentrations of midazolam and 1- hydroxymidazolam were determined by use of high-performance liquid chromatography tandem-mass spectrometry and pharmacokinetic parameters were estimated using noncompartmental analysis. The mean ± SD terminal half-lives of IC and IM midazolam were 12.04 ± 3.25 h and 16.54 ± 7.10 h, respectively. The area under the concentration- time curve extrapolated to infinity, clearance, and apparent volume of distribution in steady state of IC midazolam were 19,112.3 ± 3,095.9 ng*h/ml, 0.053 ± 0.008 L/h/kg, and 0.865 ± 0.289 L/kg, respectively. The bioavailability of IM midazolam was estimated at 89%. Maximum plasma concentrations following an IM administration were reached 2.33 ± 0.98 h and 24.00 ± 14.12 h post-injection for midazolam and 1- hydroxymidazolam, respectively, and 22.33 ± 20.26 h post-injection for 1- hydroxymidazolam following IC administration.

4.2 Introduction

Midazolam is frequently used in veterinary medicine for its sedative, muscle relaxant, anxiolytic, and anticonvulsant effects (Rankin, 2015). It has also been shown to decrease the minimum alveolar concentration of halogenated gases in numerous species, hence reducing the severity of side effects associated with inhalation anesthesia (Greiner & Larach, 1989; Hall, Szlam, et al., 1988; Kropf & Hughes, 2018; Schwieger et al., 1994; Seddighi et al., 2011). In reptiles, it provides reliable sedation in various species (Arnett-Chinn et al., 2016; Bisetto et al., 2018; Olsson & Phalen, 2013a; Oppenheim & Moon, 1995), although this effect appears highly variable in turtles (Bienzle & Boyd, 1992; Emery et al., 2014; Harvey-Clark, 1993). Midazolam is a highly lipophilic benzodiazepine that enhances the binding of the γ-aminobutyric acid (GABA) neurotransmitter to the GABAA receptors of neurons from the central nervous system. In mammals, it is rapidly absorbed when given orally, intranasally, rectally, subcutaneously (SC), and intramuscularly (IM), but its effects can be substantially decreased by first-pass metabolism (Court & Greenblatt, 1992; Nordt & Clark, 1997; Odou et al., 1999). In humans, it is mainly metabolized into the active form 1-hydroxymidazolam by the hepatic cytochrome P450 3A enzyme (CYP3A) and this metabolite is only 2.5-10% as potent as midazolam itself (Pieri, 1983). Elimination is primarily through renal excretion of the metabolites or their glucuronide conjugates in humans (Nordt & Clark, 1997). The duration of sedation is quite variable across reptilian species receiving midazolam, with significant sedation being reported for 3-114 min in red-eared sliders (Trachemys scripta elegans; Oppenheim & Moon, 1995) and 5-475 min in tegus (Salvator merianae; Bisetto et al., 2018) receiving 1.5 and 1 mg/kg IM, respectively. In ophidian species, the duration of action of midazolam appears to be even longer with a significant sedative effect being reported for 2.3 days in ball pythons (Python regius) receiving 1 mg/kg IM (see chapter 2). However, the pharmacokinetics of midazolam have not been reported in the literature for any reptilian species. This study was developed to better characterize the prolonged duration of action of midazolam observed in snakes (see chapter 2). The objective was to evaluate the absorption, distribution, and elimination of midazolam and its major metabolite 1- hydroxymidazolam following the administration of 1 mg/kg through two different routes, intracardiac (IC) and IM in ball pythons.

4.3 Material and methods

All procedures were approved by the Animal Care and Research Committees of the University of Guelph (#3812) and Toronto Zoo (#2017-11-02), in compliance with

the regulations of the Canadian Council on Animal Care. Experiments were performed in the Wildlife Health Center of the Toronto Zoo.

4.3.1 Animals

Twelve healthy adult female ball pythons with a mean ± standard deviation (SD) weight of 2.62 ± 0.52 kg (Appendix B, Table B.1) were used for the study. Snakes were individually housed in plastic or glass terrariums with access to a hiding area and a fresh water bowl. A light/dark cycle of 12/12 h was provided, and humidity was maintained between 40% and 60% throughout the study. A temperature gradient ranging from 27°C to 35°C was provided using an electric heating pad placed underneath a third of the terrarium. The day before the first blood collection, heating pads were turned off and the room temperature was maintained at a constant 30°C to decrease the potential effects of temperature variation on the pharmacokinetics of midazolam. Snakes were fed a frozen-thawed mouse or rat every other week and all animals were fasted starting one week before catheter placement until the end of the experimentation to avoid the marked metabolic variations caused by digestion (Secor et al., 2000). All ball pythons were considered healthy based on physical examination, ultrasound evaluation, and fecal parasite screening. Snakes had adequate to abundant internal fat stores based on coelomic ultrasonography. Each animal was identified using SC transponders (Trovan ISO FDX-B, Eidap Inc., Brampton, ON, Canada) that were scanned prior to each blood sample. Ten out of twelve snakes had been previously used for non-invasive studies that included the use of midazolam one month or more prior to this study (Appendix B, Table B.2).

4.3.2 Surgical catheter placement

Each snake was anesthetized seven days prior to their respective pharmacokinetic trial for the surgical placement of a central venous catheter. Induction was done by delivering 5% isoflurane (IsoFlo, Abbott Laboratories, Chicago, IL, USA) in 100% oxygen through a face mask. The trachea was intubated with an adapted 10- gauge catheter (Angiocath IV catheter, BD Canada, Mississauga, ON, Canada), and the isoflurane concentration was then maintained at 1-2% throughout the surgery. Mechanical ventilation was performed using a circle rebreathing system (Anesthesia WorkStation, Hallowell EMC, Pittsfield, MA, USA) delivering 5 breaths/min with a tidal volume of 25 mL/kg based on results of a recent study (Jakobsen et al., 2017). Body temperature was maintained around 30°C. Local anesthesia was performed using 1 mg/kg SC lidocaine (Lurocaine 20 mg/mL, Vétoquinol, Lavaltrie, QC, Canada) at the incision site and 1 mg/kg bupivacaine (Marcaine 0.50%, Pfizer Canada, Toronto, ON, Canada) as a splash block before skin closure. While each snake was in dorsal recumbency, an incision was made between the first and second most ventral rows of lateral scales, approximately 10-15 cm cranial

from the heart, and the coelom was accessed just ventral to the ribs. A second skin incision was made 3 cm craniodorsally to the first incision and a puncture was made through the exposed intercostal muscles using mosquito forceps. The right jugular vein was dissected through the first incision, elevated, compressed using two cotton-tip swabs placed 2 cm apart, and a 2-3 mm incision was made through its wall, between the swabs, using a 22-gauge needle. A 22-gauge x 10 cm polyurethane long-term catheter with integrated extension set (Milacath, Mila International Inc., Florence, KY, USA) was inserted through the ribcage puncture then through the vein incision in a caudal direction. The catheter was flushed with heparinized saline (1 mL of heparin sodium injection 10,000 IU/mL [Sandoz Canada, Boucherville, QC, Canada] mixed with 1 L of 0.9% sodium chloride [ICU Medical Inc., San Clemente, CA, USA]) to confirm patency and the base of the catheter was sutured to the outer aspect of the ribs. The vein was put back in place and the wound was closed in two layers: a continuous pattern to reattach the ribs and an everting pattern for the skin, which allowed the catheter extension set to protrude from the incision. Polydioxanone (PDS II, Ethicon US LLC, Bridgewater, NJ, USA) was used for suture. All snakes received 0.3 mg/kg SC meloxicam (Rheumocam Injection 5 mg/mL, Merck Santé Animale, Kirkland, QC, Canada) at the end of the procedure. Isoflurane was turned off when the catheter was sutured in place and all snakes recovered uneventfully. Catheters were flushed with heparinized saline following disinfection of the injection port with 70% alcohol twice daily until removal. Three to 7 days following the last blood sample (13 to 17 days post-surgery), snakes were anesthetized using 5 mg/kg IV propofol (Diprivan, Baxter Corporation Inc., Mississauga, ON, Canada) injected in the catheter. Animals were intubated 1 min later and ventilated as described above, then received 20 mL of a balanced fluid solution (Normosol-R, ICU Medical Inc.) IV. The catheter insertion site was opened with scissors, the catheter base was detached from the ribs, and the catheter was removed. The wound was then sutured as described above and all snakes received 0.3 mg/kg SC meloxicam. Mean ± SD time from propofol induction to extubation was 46.2 ± 11.7 min. All snakes recovered uneventfully.

4.3.3 Experimental design and blood sampling

The study was performed using a parallel design comparing two routes of administration, each involving a group of pythons. Randomization of group allocation and order of experimentation was performed using a statistical software (R Foundation for statistical Computing, Vienna, Austria). Each day of experimentation, one snake from each group received 1 mg/kg midazolam (5 mg/mL, Sandoz Canada) IM (IM group) or IC (IC group). Intracardiac injection and cardiocentesis were considered safe based on other studies and the authors’ experience (Isaza, Andrews, Coke, & Hunter, 2004; McFadden, Bennett, Reavill, Ragetly, & Clark-Price, 2011). The mean ± SD volume of drug injected was 0.53 ± 0.11 mL. Whole blood samples (0.6 mL per sample) 69

were collected via the jugular catheter using a two-stage technique considering the 0.155 mL dead space of the catheter: 0.3 mL of a mixture of heparinized saline and blood was suctioned and discarded, then 0.6 mL of blood was collected with a heparinized syringe and immediately transferred into lithium heparin tubes stored on ice. The catheter was then flushed with 0.5 mL of heparinized saline. If blood could not be collected from the jugular catheter due to occlusion, cardiocentesis was performed. Blood samples were collected prior to drug administration followed by 3, 5, 10, 20, 30, and 45 min post-injection, and 1, 2, 4, 8, 12, 24, 48, and 72 h post-injection for the IC group. Blood was sampled prior to drug administration followed by 5, 15, 30, and 45 min post-injection, and 1, 1.5, 2, 3, 4, 8, 12, 24, 48, and 72 h post-injection for the IM group. Plasma was transferred into sterile cryovials following centrifugation of whole blood samples at 4,000 g for 10 min within 2 h of sampling then stored at -80°C until analysis. Superficial body temperature was noted before each blood sample was collected using an infrared laser thermometer (Fluke 59 MAX Infrared Thermometer, Fluke Electronics Canada, Mississauga, ON, Canada). The general behavior of snakes was monitored for potential adverse effects prior to each blood sample. Blank plasma was opportunistically collected for assay validation prior to the experimentation in ball pythons within the research collection at the Toronto Zoo, including snakes used in the current study. A maximum total blood volume per snake equivalent to 0.8% of the body weight was collected throughout the study, which included blood collected prior to the experimentation (blank plasma) and during the experimentation (midazolam pharmacokinetics). Blank plasma was separated and stored as previously described.

4.3.4 Midazolam analysis

Reference standards midazolam, 1-hydroxymidazolam, and their deuterated internal standards midazolam-D4 and 1-hydroxymidazolam-D4 were obtained from Cerilliant (Round Rock, TX, USA). Ammonium formate was obtained from Sigma- Aldrich (St. Louis, MO, USA). Methyl tertiary-butyl ether (MTBE), high-performance liquid chromatography tandem-mass spectrometry (HPLC-MS/MS)-grade formic acid, acetonitrile, and methanol were obtained from Thermo Fischer Scientific (Oakville, ON, Canada). Water used throughout the assay was obtained from a Milli-Q system (Millipore, Bedfore, MA, USA). Plasma concentrations of midazolam were determined using HPLC-MS/MS analysis. HPLC analysis was performed on a Thermo Scientific Vanquish Flex Binary UHPLC system (Waltham, MA, USA). The sample extracts were separated on an ACQUITY UPLC BEH C18 Column (1.7 µm, 2.1 mm x 100 mm, Waters, Ireland) connected with a VanGuard UPLC BEH C18 Pre-Column (1.7 µm, 2.1 mm x 5 mm, Waters, Ireland). The auto sampler was kept at 4°C and column temperature was set at 30°C. The mobile phase consisted of 10 mM ammonium formate with 0.1% formic acid

aqueous and acetonitrile (68:32, v/v), and flow rate was 200 µL/min. A 3 µL sample was injected into the column. Midazolam and midazolam-D4 were eluted at 2.7 min, and 1- hydroxymidazolam and 1-hydroxymidazolam-D4 were eluted at 3.0 min. Mass spectrometry analysis was performed with a Thermo Scientific Q Exactive Focus Orbitrap mass spectrometer equipped with a Thermo Scientific Ion Max source and a heated electrospray source. Data were acquired in parallel-reaction monitoring positive ion mode. In this mode, [M+H]+ precursor ion m/z 326.1 for midazolam, m/z 330.1 for midazolam–D4, m/z 342.1 for 1-hydroxymidazolam, and m/z 346.1 for 1- hydroxymidazolam-D4 were selected in the quadrupole with an isolation width of 2.0 m/z, and then fragmented in the HCD cell at collision energy 32 eV for midazolam and 22 eV for 1-hydroxymidazolam. The resulting MS/MS product ion spectrum was detected in the Orbitrap at a resolution of 17,500 (FWHM at m/z 200) with AGC target set at 1e5. The ionization conditions were optimized using Tee infusion (10 µL/min) of midazolam (5 µg/mL) into HPLC flow (200 µL/min). The Ion source parameters were optimized as: Sweep gas flow rate 2, Spray voltage 3.5 kV, Capillary temp 250°C, S- lens RF level 50.0, Aux gas heater temperature 400°C. Data were acquired and processed using Thermo Scientific TraceFinder software. The most abundant fragment from the MS/MS spectrum m/z 291.1149 and m/z 324.0679 were selected as the quantifying ions for midazolam and 1- hydroxymidazolam, respectively. Other specific fragments m/z 244.0310 and m/z 203.0359 were selected as confirming ions. The resulting chromatograms were extracted and reconstructed with a mass accuracy of 5 ppm for quantification and confirmation. Appendix E, Figures E.1 and E.2 show representative MS/MS spectra for midazolam and midazolam-D4, respectively, and Figures E.3 and E.4 show representative MS/MS spectra for 1-hydroxymidazolam and 1-hydroxymidazolam-D4, respectively, with qualifying, quantifying and confirming ions specified. Stock solutions of midazolam, 1-hydroxymidazolam, midazolam-D4, and 1- hydroxymidazolam-D4 were prepared in methanol at 100 µg/mL and stored at -80°C. Calibration standards (range of 1-1,000 ng/mL) and quality controls (1.5 ng/mL and 800 ng/mL) were prepared on the day of analysis by spiking working solutions in blank ball python plasma. A liquid-liquid extraction was carried out using MTBE. Briefly, 20 µl of internal standard working solution was added to 100 µl of ball python plasma, then 1mL of MTBE was added. The mixture was vortexed followed by centrifugation at 9,600 g for 5 min. The MTBE layer was evaporated under a constant flow of nitrogen, and the residue was reconstituted with 100 µL of the mobile phase. A volume of 3 µL was injected into the column. Samples with concentrations above the calibration curve range were diluted with blank ball python plasma to obtain values within that range.

4.3.5 Method validation

The method validation procedure determined assay specificity, selectivity, linearity, accuracy, and intra- and inter-day precision using the sample preparation procedure and HPLC-MS/MS instrument conditions described above. Validation data are summarized in Appendix E, Tables E.1 and E.2 for midazolam and 1- hydroxymidazolam, respectively. Six different ball python blank plasma samples were extracted and the peak areas at the retention times of analyte and internal standard were compared with the peak areas found in the lower limit of quantification (1 ng/mL) to determine the selectivity of the assay. No interfering peaks were observed in blank ball python plasma at retention times of midazolam and its metabolite 1- hydroxymidazolam. Calibration curves were prepared and assayed on 10 separate days. Standard curves were calculated using weighted (1/x) least squares linear regression. Eight-point standard calibration curves were linear and reproducible in the concentration range from 1 ng/mL to 1000 ng/mL, with the correlation coefficient R2 > 0.99 for all calibration curves. The limits of quantitation (LOQ) were 1 ng/mL for both midazolam and its active metabolite in this assay. Limits of quantification were determined as the lowest concentrations that had a coefficient of variation (CV) no greater than 20%. The method precision was evaluated by running triplicates of calibration standards on three different days. The analytical method was repeatable and reproducible with intra- and inter-day precision within 15% for each calibration standard except for the LOQ, which is within 20%. Accuracy was evaluated by calculating the percent deviation from the true concentration. The accuracy for each calibration standard was within 15% CV for both midazolam and 1-hydroxymidazolam.

4.3.6 Pharmacokinetics analysis

Pharmacokinetic parameters of midazolam following IC and IM administration were calculated using noncompartmental analysis with a pharmacokinetic add-in program (Zhang, Huo, Zhou, & Xie, 2010) for a commercial spreadsheet program (Microsoft Excel, Microsoft Corp., Redmond, WA, USA). The area under the concentration-time curve (AUC) until the last measurable time point (AUC0-t) was calculated using the trapezoidal method and the AUC extrapolated to infinity (AUC0-∞) was calculated by adding Ctlast/λz to the AUC0-t, where Ctlast was the concentration at the last time point and λz was an estimate of the slope of the terminal phase of the concentration-time curve. The AUC0-t ratio was calculated as the ratio of the AUC0-t of 1- hydroxymidazolam and midazolam. The terminal half-life (t1/2) was calculated as ln(2)/λz, the clearance (Cl) as the dose/AUC0-∞, and the mean residence time (MRT) as AUMC0- ∞/AUC0-∞, where AUMC0-∞ was the area under the first moment curve. The apparent volume of distribution in steady state (Vss) was calculated as the product of the Cl and MRT. The mean absorption time (MAT) was calculated as the difference between the mean IM and IC MRT. The bioavailability of IM midazolam was defined as the ratio of the mean IM and IC AUC0-∞. 72

4.4 Results

No major adverse effects were noted. The mean ± SD superficial body temperature of all snakes combined was 30.4 ± 0.8°C throughout the blood collection phase. Leakage of a small amount of drug following IM administration (estimated at <10% of the volume injected) was occasionally observed due to strong muscle contractions at injection sites. Overall, 146 blood samples (81.6%) were collected from the central venous catheter and 33 (18.4%) by cardiocentesis. Midazolam was detected in the first post-injection plasma sample of all snakes from the IC group and 5/6 snakes from the IM group up until the last sample in 4/6 snakes from each group. 1- hydroxymidazolam was detected starting 3 min (IC group) and 5 min (IM group) post- injection in 6/6 and 5/6 snakes, respectively, up until the last sample in 5/6 and 6/6 snakes, respectively. One blood sample from a snake in the IM group was not collected at 72 h post-injection due to the development of surgical wound dehiscence. Surgical wounds of all other snakes remained clean with minimum inflammation. Plasma concentrations following the administration of 1 mg/kg midazolam are reported with semi-logarithmic concentration-time curves in Figure 4.1 for IC midazolam, Figure 4.2 for IM midazolam, Figure 4.3 for IC 1-hydroxymidazolam, and Figure 4.4 for IM 1-hydroxymidazolam. Individual plasma concentrations of midazolam and 1-hydroxymidazolam over time for each ball python are listed in Appendix E, Tables E.3 (IC midazolam), E.4 (IM midazolam), E.5 (IC 1-hydroxymidazolam), and E.6 (IM 1- hydroxymidazolam). Noncompartmental pharmacokinetic parameters are reported in Table 4.1 for both administration routes. Due to insufficient time points in the terminal phase for both IM and IC 1-hydroxymidazolam, only the AUC0-t, maximum plasma concentration (Cmax) , and time to Cmax (tmax) are reported for the metabolite.

Figure 4.1 – Mean ± SD plasma concentration-time profile of midazolam following the administration of 1 mg/kg intracardiac in ball pythons (Python regius).

Figure 4.2 – Mean ± SD plasma concentration-time profile of midazolam following the administration of 1 mg/kg intramuscular in ball pythons (Python regius). The lower SD of the plasma concentration 48 h after the injection cannot be illustrated on a semi-logarithmic scale because it extends below 0 (mean ± SD = 95.49 ± 107.86 ng/mL).

Figure 4.3 – Mean ± SD plasma concentration-time profile of 1-hydroxymidazolam following the administration of 1 mg/kg intracardiac midazolam in ball pythons (Python regius).

Figure 4.4 – Mean ± SD plasma concentration-time profile of 1-hydroxymidazolam following the administration of 1 mg/kg intramuscular midazolam in ball pythons (Python regius).

MIDAZOLAM 1-HYDROXYMIDAZOLAM PARAMETER UNIT IC IM IC IM t1/2 h 12.04 ± 3.25 16.54 ± 7.10 - - tmax h - 2.33 ± 0.98 22.33 ± 20.26 28.67 ± 16.86 Cmax ng/mL - 810.5 ± 431.3 469.6 ± 162.8 207.0 ± 65.2 AUC0-t ng*h/mL 18,492.0 ± 2,806.3 15,402.1 ± 9,766.7 18,259.5 ± 5,615.5 10,695.2 ± 3,779.5 AUC0-∞ ng*h/mL 19,112.3 ± 3,095.9 17,102.4 ± 11,631.9 - - AUC0-t ratio - - 0.99 ± 0.26 0.82 ± 0.35 C0 ng/mL 3,479.0 ± 1,374.9 - - - Cl L/h/kg 0.053 ± 0.008 - - - -1 λz h 0.063 ± 0.025 0.049 ± 0.020 - - Vss L/kg 0.865 ± 0.289 - - - MRT h 16.90 ± 7.24 23.15 ± 9.64 - - MAT h - 6.25 - - F % - 89 - - t1/2: terminal half-life; tmax: time to maximum plasma concentration; Cmax: maximum plasma concentration; AUC0-t: area under the concentration- time curve until the last measurable time point; AUC0-∞: area under the concentration-time curve extrapolated to infinity; C0: plasma concentration at time zero; Cl: clearance; λz: slope of the terminal phase of the concentration-time curve; Vss: apparent volume of distribution in steady state; MRT: mean residence time; MAT: mean absorption time; F: bioavailability.

4.5 Discussion

Midazolam administered at a dosage of 1 mg/kg in healthy female ball pythons was readily absorbed following IM administration. First order elimination was observed with both IC and IM routes. Plasma concentrations were maintained above the minimum plasma concentration necessary to provide sedation in humans (40 ng/mL; Allonen et al., 1981) for at least 48 and 72 h following IC and IM administrations, respectively. However, the authors are not aware of any published studies evaluating the pharmacokinetic-pharmacodynamic relationship of midazolam in any reptiles and, therefore, no assumptions can be made regarding the dose-response relationship.

The tmax of IM midazolam was 2.33 ± 0.98 h, which was much later than reports in dogs (0.13 h; Court & Greenblatt, 1992; M. Schwartz et al., 2013) and sheep (0.46 h; Simon et al., 2017). Maximum sedation has been reported 1 h post-injection in ball pythons (see Chapter 2), which appears to be shorter than the tmax from the current study. This discrepancy could be due to clockwise hysteresis (e.g., development of tolerance to the drug over time, time dependent protein binding, or a higher affinity of the metabolite to the GABAA receptor), inter-individual variability, difficulty objectively evaluating the level of sedation, or a ceiling effect. Clockwise hysteresis has been previously reported with the IV infusion of midazolam in elderly humans (Albrecht et al., 1999). The bioavailability of IM midazolam in ball pythons was high (89%). Leakage of a small amount of drug outside of the injection site in a few snakes did not seem to significantly affect the overall bioavailability. A hepatic first-pass effect has been well- reported in humans following the oral administration of midazolam (Nordt & Clark, 1997). Additionally, several anesthetic drugs undergoing hepatic metabolism have been reported to have a decreased Cmax or observable effect when administered in the caudal body of reptiles (Doss et al., 2017; Kummrow et al., 2008; Yaw et al., 2018). Therefore, the reported bioavailability in ball pythons likely only applies to an IM injection in the cranial epaxial muscles.

The Vss of midazolam in ball pythons (0.865 ± 0.289 L/kg) was larger than the blood volume, which is consistent with the lipophilic nature of this drug and corresponding widespread distribution (Nordt & Clark, 1997). Obesity has been shown to increase the volume of distribution of midazolam in humans (Brill et al., 2014). However, the Vss was not higher in the current study compared to that of mammalian and avian species, with values as high as 2.99 L/kg in rabbits (Oryctolagus cuniculus) receiving 0.35 mg/kg IV (Bienert et al., 2014) and 4.83 L/kg in turkeys (Meleagris gallopavo) receiving 5 mg/kg IV (Cortright et al., 2007). This variability likely reflected the differences in body composition, plasma protein binding, and organ blood flow, which highlight the limitations of extrapolating these values across taxa.

The clearance of midazolam in ball pythons (0.053 ± 0.008 L/h/kg) was much lower than that of mammalian species, with the lowest reported value being 0.606 L/h/kg in dogs (M. Schwartz et al., 2013). To characterize the importance of this clearance, the body extraction ratio (Ebody) can be estimated by the following formula: Ebody = Cl/cardiac output (Toutain & Bousquet-Mélou, 2004). The cardiac output of snakes, which varies with digestion and exercise (Enok et al., 2016; Secor et al., 2000), has been reported to range from 9.6 mL/min/kg in fasted snakes at rest (Wang, Axelsson, Jensen, & Conlon, 2000) up to 54.1 mL/min/kg in fed snakes (Enok et al., 2016). The maximum Ebody of midazolam in ball pythons can therefore be estimated at 0.092. However, this value cannot be directly compared to mammals since the proportion of the cardiac output that reaches the liver has not be estimated in ball pythons. Since this Ebody is not especially low, the slow clearance could result from the low cardiac output of fasted ball pythons at rest compared to mammals, low hepatic blood flow, or a low hepatic extraction ratio. In mammals, midazolam has an intermediate to high hepatic extraction ratio, suggesting a flow-dependent metabolism (Allonen et al., 1981; McKindley et al., 1998), such as demonstrated in hypovolemic dogs in which midazolam clearance is reduced (Adams et al., 1985). The field metabolic rate of reptiles is much lower than that of mammals or birds of similar size (Nagy, 2005), which may also contribute to the slow rate of drug elimination. In humans, to be excreted in the urine, this lipophilic drug needs to be hydroxylated and optionally conjugated to increase its hydrosolubility (Nordt & Clark, 1997). Alterations of the hepatic metabolism in critically ill humans have been shown to significantly affect the clearance of midazolam (Spina & Ensom, 2007). Hence, hepatic metabolism could have a significant effect on the clearance of this drug in ball pythons, although the activity of the CYP3A has not yet been evaluated in this species. Plasma concentrations of the metabolite 1-hydroxymidazolam reached their peak 22.33 ± 20.26 h following IC administration of midazolam. The difference between the IC tmax of midazolam and its major metabolite in ball pythons (22.33 h) was substantially longer than in rabbits (0.28 h; Bienert et al., 2014) and sheep (0.28 h; Simon et al., 2017), which suggests slow hepatic metabolism or low hepatic blood flow. In humans, prolonged sedation can occur with the accumulation of 1-hydroxymidazolam in patients with compromised renal function (Bauer et al., 1995; Spina & Ensom, 2007). The mechanism of elimination and the potency of 1-hydroxymidazolam have not been evaluated in snakes, but differences in renal physiology and blood supply could potentially delay its elimination, which could hypothetically lead to prolonged sedation. The AUC0-t ratio in ball pythons following IC administration was close to 1, indicating important exposure to the metabolite. However, a ratio of the AUC0-∞ would have been necessary to estimate the impact of hepatic metabolism and renal excretion on the exposure to 1-hydroxymidazolam.

Snakes are ectothermic animals that regulate their body temperature by adapting their behavior in relation to their environment. The pharmacokinetic profile of a drug has been shown to be significantly affected by temperature in reptiles (Caligiuri et al., 1990; Mader et al., 1985). Since this effect has not been evaluated in ball pythons receiving midazolam, extrapolation of the reported pharmacokinetic parameters outside of the temperature range of the study (30.4 ± 0.8°C) should be done with care. In conclusion, the elimination of midazolam administered at a dosage of 1 mg/kg IC and IM is slow in ball pythons, which could be due to low hepatic blood flow, slow hepatic metabolism, or low cardiac output. The apparent volume of distribution is consistent with lipophilic properties of this drug that allow for widespread distribution. In addition, the bioavailability of the IM injection is high. Dosages below 1 mg/kg should be considered to reduce the prolonged duration of sedation.

Acknowledgements Funding of this research was provided by a Pet Trust grant from the Ontario Veterinary College, the Toronto Zoo, and the Andrea Leger Dunbar Summer Research Studentship. The authors would like to thank Hannah Bagnall, Sydney Swartz, Michelle Lovering, and Cassia Devison for their technical support during the animal sampling phase, as well as Rick Vos and all the keepers, veterinary technicians, and veterinarians of the Wildlife Health Center of the Toronto Zoo for taking care of these ball pythons. The authors acknowledge the help and guidance provided by Dr. Dorothee Bienzle and Dr. Nicole M. Nemeth throughout this study.

5 Conclusions and future directions

The objectives of this research were to evaluate the pharmacodynamics and pharmacokinetics of midazolam, and to evaluate the effects of midazolam and nitrous oxide (N2O) on the minimum anesthetic concentration of isoflurane (MACIso) in the ball python (Python regius).

The first hypothesis was that midazolam would provide safe and moderate sedation and muscle relaxation in ball pythons. Although these sedative and muscle relaxant effects were variable among snakes, the general premise of this hypothesis was confirmed. No significant differences were observed by doubling the dosage administered from 1 to 2 mg/kg intramuscular (IM), which likely reflected either a nonlinear dose-response relationship, a ceiling effect, or insufficient statistical power. The duration of action of midazolam in ball pythons ranged from 3 to 5 days, which was substantially longer than that reported in other reptilian species. The heart rate was significantly lower than baseline from 30 min to 5.3 day for both dosages. However, tachycardia likely resulted from the manipulations and injections of snakes and, therefore, baseline values were likely falsely increased. A true bradycardic effect of midazolam could not be confirmed. However, bradycardia has been reported in Burmese pythons (Python molurus) receiving midazolam with the use of stress-free heart monitoring methods (Lopes et al., 2017). A similar methodology would be required to accurately evaluate the effects of midazolam on the heart rate of ball pythons. Paradoxical excitation was observed in one snake and was more pronounced at the highest dosage. Although the mechanisms of paradoxical excitation are unknown in snakes, using lower dosages would likely decrease the prevalence and intensity of this adverse effect.

The reversal effects of flumazenil in ball pythons were evaluated in an open trial. Conclusions from this study are limited by the absence of blinded observers and a control group. However, the response to injection of 0.08 mg/kg IM of flumazenil 60 min following the administration of 1 mg/kg IM of midazolam was clear. Sedation and muscle relaxations scores, as well as the heart rate, did not significantly differ from baseline (i.e., before the midazolam injection) within 10 min of the flumazenil injection. This effect lasted for up to 1 h for heart rate, 2 h for sedation, and 6 h for muscle relaxation. However, resedation was evident in all snakes 3 h following the flumazenil injection. Although the pharmacokinetics of flumazenil are similar to those of midazolam in dogs and humans (Amrein & Hetzel, 1990; Lemke, 2007), this is unlikely to be the case in ball pythons. A pharmacokinetic evaluation of flumazenil in this species is needed to confirm this assumption. Studies evaluating higher dosages of flumazenil in a randomized, blinded, controlled trial are also warranted.

The MACIso in ball pythons was 1.11% (95% confidence interval, 0.94-1.28%), which is lower than that of most reptilian species. By providing more time for anesthetic gas equilibration compared to other studies in reptiles, it is likely that the measured end- tidal concentration of isoflurane more accurately reflected the actual target tissue concentration, which would result in a more accurate estimation of the MACIso. 82

However, without measuring the actual blood concentration of isoflurane, it is difficult to determine if a true equilibrium was reached. Intracardiac (IC) blood shunting can provoke a marked ventilation-perfusion (V/Q) mismatch in reptiles, which severely impairs the capacity of anesthetic gases to reach equilibration between the lungs, blood, and target tissue. By eliminating this IC shunt, a more accurate estimation of the MAC could be performed, such as demonstrated in the red-footed tortoise (Chelonoidis carbonaria) (Greunz et al., 2018). A similar study evaluating the capacity of anticholinergic drugs to eliminate IC shunting in ball pythons followed by a measurement of the MACIso without IC shunting would help validate or invalidate the accuracy of the estimations from the current study.

The second hypothesis of this research was that midazolam and N2O would significantly decrease the MACIso in ball pythons. This hypothesis was confirmed by demonstrating a reduction of the MACIso by 57% with the administration of 1 mg/kg IM of midazolam and by 17% with the delivery of isoflurane in 50% N2O and 50% oxygen. Further studies are required to determine if the MACIso reduction is dose dependent with both drugs.

A pharmacokinetic study of midazolam was performed following the observation of an unusually long duration of action in ball pythons compared to other reptiles. Plasma concentrations of midazolam were highly variable between individuals. This finding was not unexpected considering the highly variable effects observed in the pharmacodynamics study. Inter- and intra-individual variation in the absorption, distribution, and elimination of the drug is suspected.

The third hypothesis was that midazolam follows a similar pharmacokinetic profile in ball pythons compared to mammalian species. Plasma concentrations of midazolam following 1 mg/kg IC or IM administration were maintained well above concentrations reported to provide sedation in mammalian species for at least 2 days. The mean ± SD volume of distribution in steady state was 0.865 ± 0.289 L/kg, which was larger than the blood volume of ball pythons, but not sufficiently large to justify the slow clearance (0.053 ± 0.0087 L/h/kg) and prolonged terminal half-life (12.04 ± 3.25 h) compared to mammalian species. The third hypothesis was therefore invalidated. Slow hepatic metabolism, low hepatic blood flow, and low cardiac output are suspected to be, at least in part, responsible for the slow clearance and, hence, for the prolonged duration of action of midazolam in ball pythons. Accumulation of the metabolite 1- hydroxymidazolam over time could also have contributed to the prolonged sedation. Further studies evaluating the activity of the hepatic cytochrome P450 3A in ball pythons are required to better understand the metabolism of this drug. Additionally, hepatic and renal blood flow can significantly limit the elimination of midazolam. Therefore, quantification of these blood flow under various circumstances in ball pythons would allow for a better understanding of the body extraction ratio of midazolam (Toutain & Bousquet-Mélou, 2004).

A dose-response analysis based on results from the pharmacodynamic and pharmacokinetic studies was not undertaken due to the large differences between the 83

two studies. Some of these inter-study differences included variations in body temperature and body weight, seasonality (i.e., winter versus summer), environmental differences (i.e., no temperature gradient in the pharmacokinetic study, variation in humidity levels), history of surgery prior to the pharmacokinetic study, and lack of randomization of the pharmacodynamic versus pharmacokinetic trials. Hence, a study designed to simultaneously evaluate the pharmacodynamics and pharmacokinetics of midazolam in the ball python is warranted to evaluate this dose-response relationship.

Overall, midazolam was a safe, reliable, and potent sedative and muscle relaxant when administered at a dosage of 1 or 2 mg/kg IM in ball pythons. The duration of action at these dosages was substantially longer in ball pythons compared to other reptilian, avian, or mammalian species. Midazolam provided a significant reduction of the MACIso in ball pythons, while the MACIso-reducing properties of N2O were far less pronounced, though significant. Adverse effects of midazolam included paradoxical excitation and decreased heart rate. A dosage of 1 mg/kg IM or lower, when used in combination with other sedative or anesthetic drugs, is therefore recommended to provide moderate to profound sedation and muscle relaxation, and to decrease the MACIso in ball pythons. Resedation should be expected with the administration of 0.08 mg/kg IM of flumazenil following the administration of 1 mg/kg IM of midazolam.

REFERENCES

Aarnes, T. K., Fry, P. R., Hubbell, J. A. E., Bednarski, R. M., Lerche, P., Chen, W., … Lakritz, J. (2013). Pharmacokinetics and pharmacodynamics of midazolam after intravenous and intramuscular administration in alpacas. American Journal of Veterinary Research, 74(2), 294–299. https://doi.org/10.2460/ajvr.74.2.294

Acevedo-Arcique, C. M., Ibancovichi, J. A., Chavez, J. R., Gutierrez-Blanco, E., Moran- Muñoz, R., Victoria-Mora, J. M., … Sanchez-Aparicio, P. (2014). Lidocaine, dexmedetomidine and their combination reduce isoflurane minimum alveolar concentration in dogs. PLoS ONE, 9(9), e106620. https://doi.org/10.1371/journal.pone.0106620

Adams, P., Gelman, S., Reves, J. G., Greenblatt, D. J., Alvis, J. M., & Bradley, E. (1985). Midazolam pharmacodynamics and pharmacokinetics during acute hypovolemia. Anesthesiology, 63(2), 140–146.

Albrecht, S., Ihmsen, H., Hering, W., Geisslinger, G., Dingemanse, J., Schwilden, H., & Schüttler, J. (1999). The effect of age on the pharmacokinetics and pharmacodynamics of midazolam. Clinical Pharmacology & Therapeutics, 65(6), 630–639. https://doi.org/10.1016/S0009-9236(99)90084-X

Allonen, H., Ziegler, G., & Klotz, U. (1981). Midazolam kinetics. Clinical Pharmacology and Therapeutics, 30(5), 653–661.

Amrein, R., & Hetzel, W. (1990). Pharmacology of Dormicum (midazolam) and Anexate (flumazenil). Acta Anaesthesiologica Scandinavica. Supplementum, 92, 6–15; discussion 47.

Andersen, J. B., Rourke, B. C., Caiozzo, V. J., Bennett, A. F., & Hicks, J. W. (2005). Postprandial cardiac hypertrophy in pythons. Nature, 434(7029), 37–38. https://doi.org/10.1038/434037a

Antognini, J. F., & Eisele, P. H. (1993). Anesthetic potency and cardiopulmonary effects of enflurane, halothane, and isoflurane in goats. Laboratory Animal Science, 43(6), 607–610.

Antognini, J. F., & Schwartz, K. (1993). Exaggerated anesthetic requirements in the preferentially anesthetized brain. Anesthesiology, 79(6), 1244–1249.

Aranake, A., Mashour, G. A., & Avidan, M. S. (2013). Minimum alveolar concentration: ongoing relevance and clinical utility. Anaesthesia, 68(5), 512–522. https://doi.org/10.1111/anae.12168

Arnett-Chinn, E. R., Hadfield, C. A., & Clayton, L. A. (2016). Review of intramuscular 85

midazolam for sedation in reptiles at the National Aquarium, Baltimore. Journal of Herpetological Medicine and Surgery, 26(1–2), 59–63. https://doi.org/10.5818/1529-9651-26.1-2.59

Baggot, J. D., Love, D. N., Rose, R. J., & Raus, J. (1981). The pharmacokinetics of some aminoglycoside antibiotics in the horse. Journal of Veterinary Pharmacology and Therapeutics, 4(4), 277–284.

Bailey, R. S., Barter, L. S., & Pypendop, B. H. (2017). Pharmacokinetics of dexmedetomidine in isoflurane-anesthetized New Zealand white rabbits. Veterinary Anaesthesia and Analgesia, 44(4), 876–882. https://doi.org/10.1016/j.vaa.2017.01.003

Baker, L. A., & White, F. N. (1970). Redistribution of cardiac output in response to heating in Iguana iguana. Comparative Biochemistry and Physiology, 35(2), 253– 262. https://doi.org/10.1016/0010-406X(70)90592-X

Barletta, M., Quandt, J., & Hofmeister, E. (2016). Determination of minimum alveolar concentration of isoflurane in dogs and cats using the up-and-down method. A preliminary study. Research in Veterinary Science, 106, 81–83. https://doi.org/10.1016/j.rvsc.2016.03.011

Barter, L. S., Hawkins, M. G., Brosnan, R. J., Antognini, J. F., & Pypendop, B. H. (2006). Median effective dose of isoflurane, sevoflurane, and desflurane in green iguanas. American Journal of Veterinary Research, 67(3), 392–397. https://doi.org/10.2460/ajvr.67.3.392

Barter, L. S., Hawkins, M. G., & Pypendop, B. H. (2015). Effects of fentanyl on isoflurane minimum alveolar concentration in New Zealand white rabbits (Oryctolagus cuniculus). American Journal of Veterinary Research, 76(2), 111–115. https://doi.org/10.2460/ajvr.76.2.111

Barter, L. S., Ilkiw, J. E., Steffey, E. P., Pypendop, B. H., & Imai, A. (2004). Animal dependence of inhaled anaesthetic requirements in cats. British Journal of Anaesthesia, 92(2), 275–277.

Barza, M., & Weinstein, L. (1976). Pharmacokinetics of the penicillins in man. Clinical Pharmacokinetics, 1(4), 297–308. https://doi.org/10.2165/00003088-197601040- 00004

Bauer, T. M., Ritz, R., Haberthür, C., Ha, H. R., Hunkeler, W., Sleight, A. J., … Haefeli, W. E. (1995). Prolonged sedation due to accumulation of conjugated metabolites of midazolam. Lancet, 346(8968), 145–147. https://doi.org/10.1016/s0140- 6736(95)91209-6

Beck, K., Loomis, M., Lewbart, G., Spelman, L., & Papich, M. (1995). Preliminary comparison of plasma concentrations of gentamicin injected into the cranial and 86

caudal limb musculature of the eastern box turtle (Terrapene carolina carolina). Journal of Zoo and Wildlife Medicine, 26(2), 265–268.

Benchetrit, G., Armand, J., & Dejours, P. (1977). Ventilatory chemoreflex drive in the tortoise, Testudo horsfieldi. Respiration Physiology, 31(2), 183–191.

Bertelsen, M. F. (2014). Squamates. In G. West, D. Heard, & N. Caulkett (Eds.), Zoo Animal and Wildlife Immobilization and Anesthesia (2nd ed., pp. 351–363). Ames, IA, USA: John Wiley & Sons, Inc.

Bertelsen, M. F., Mosley, C. A. E., Crawshaw, G. J., Dyson, D. H., & Smith, D. A. (2005a). Anesthetic potency of sevoflurane with and without nitrous oxide in mechanically ventilated Dumeril monitors. Journal of the American Veterinary Medical Association, 227(4), 575–578.

Bertelsen, M. F., Mosley, C. A. E., Crawshaw, G. J., Dyson, D., & Smith, D. A. (2005b). Inhalation anesthesia in Dumeril’s monitor (Varanus dumerili) with isoflurane, sevoflurane, and nitrous oxide: effects of inspired gases on induction and recovery. Journal of Zoo and Wildlife Medicine, 36(1), 62–68. https://doi.org/10.1638/04-033

Bertelsen, M. F., Mosley, C. A. E., Crawshaw, G. J., Dyson, D., & Smith, D. A. (2005c). Minimum alveolar concentration of isoflurane in mechanically ventilated Dumeril monitors. Journal of the American Veterinary Medical Association, 226(7), 1098– 1101.

Bienert, A., Płotek, W., Wiczling, P., Kostrzewski, B., Kamińska, A., Billert, H., … Grześkowiak, E. (2014). The influence of the time of day on midazolam pharmacokinetics and pharmacodynamics in rabbits. Pharmacological Reports, 66(1), 143–152. https://doi.org/10.1016/j.pharep.2013.06.009

Bienzle, D., & Boyd, C. J. (1992). Sedative effects of ketamine and midazolam in snapping turtles (Chelydra serpentina). Journal of Zoo and Wildlife Medicine, 23(2), 201–204.

Bisetto, S. P., Melo, C. F., & Carregaro, A. B. (2018). Evaluation of sedative and antinociceptive effects of dexmedetomidine, midazolam and dexmedetomidine– midazolam in tegus (Salvator merianae). Veterinary Anaesthesia and Analgesia, 45(3), 320–328. https://doi.org/10.1016/j.vaa.2017.12.004

Bonath, K. (1979). Halothane inhalation anaesthesia in reptiles and its clinical control. International Zoo Yearbook, 19, 112–125. https://doi.org/10.1111/j.1748- 1090.1979.tb00544.x

Botman, J., Dugdale, A., Gabriel, F., & Vandeweerd, J.-M. (2016). Cardiorespiratory parameters in the awake pigeon and during anaesthesia with isoflurane. Veterinary Anaesthesia and Analgesia, 43(1), 63–71. https://doi.org/10.1111/vaa.12262

Bovill, J. G. (2008). Inhalation anaesthesia: from diethyl ether to xenon. Handbook of Experimental Pharmacology, 182, 121–142. https://doi.org/10.1007/978-3-540- 74806-9_6

Briani, C., Dalla Torre, C., Citton, V., Manara, R., Pompanin, S., Binotto, G., & Adami, F. (2013). Cobalamin deficiency: clinical picture and radiological findings. Nutrients, 5(11), 4521–4539. https://doi.org/10.3390/nu5114521

Brill, M. J. E., van Rongen, A., Houwink, A. P. I., Burggraaf, J., van Ramshorst, B., Wiezer, R. J., … Knibbe, C. A. J. (2014). Midazolam pharmacokinetics in morbidly obese patients following semi-simultaneous oral and intravenous administration: a comparison with healthy volunteers. Clinical Pharmacokinetics, 53(10), 931–941. https://doi.org/10.1007/s40262-014-0166-x

Burggren, W., & Johansen, K. (1982). Ventricular haemodynamics in the monitor lizard Varanus exanthematicus: pulmonary and systemic pressure separation. Journal of Experimental Biology, 96(1), 343–354.

Caligiuri, R., Kollias, G. V, Jacobson, E., McNab, B., Clark, C. H., & Wilson, R. C. (1990). The effects of ambient temperature on amikacin pharmacokinetics in gopher tortoises. Journal of Veterinary Pharmacology and Therapeutics, 13(3), 287–291.

Campagna, J. A., Miller, K. W., & Forman, S. A. (2003). Mechanisms of actions of inhaled anesthetics. New England Journal of Medicine, 348(21), 2110–2124. https://doi.org/10.1056/NEJMra021261

Cantalapiedra, A. G., Villanueva, B., & Pereira, J. L. (2000). Anaesthetic potency of isoflurane in cattle: determination of the minimum alveolar concentration. Veterinary Anaesthesia and Analgesia, 27(1), 22–26. https://doi.org/10.1046/j.1467- 2995.2000.00002.x

Chan, F.-T., Chang, G.-R., Wang, H.-C., & Hsu, T.-H. (2013). Anesthesia with isoflurane and sevoflurane in the crested serpent eagle (Spilornis cheela hoya): minimum anesthetic concentration, physiological effects, hematocrit, plasma chemistry and behavioral effects. The Journal of Veterinary Medical Science, 75(12), 1591–1600.

Chavez, J. R., Ibancovichi, J. A., Sanchez-Aparicio, P., Acevedo-Arcique, C. M., Moran- Muñoz, R., & Recillas-Morales, S. (2015). Effect of acetaminophen alone and in combination with morphine and tramadol on the minimum alveolar concentration of isoflurane in rats. PLoS ONE, 10(11), e0143710. https://doi.org/10.1371/journal.pone.0143710

Chinnadurai, S. K., Balko, J. A., & Williams, C. V. (2017). Minimum alveolar concentration and cardiopulmonary effects of isoflurane in ring-tailed lemurs (Lemur catta). Journal of the American Association for Laboratory Animal Science, 56(4), 452–456. 88

Chinnadurai, S. K., DeVoe, R., Koenig, A., Gadsen, N., Ardente, A., & Divers, S. J. (2010). Comparison of an implantable telemetry device and an oscillometric monitor for measurement of blood pressure in anaesthetized and unrestrained green iguanas (Iguana iguana). Veterinary Anaesthesia and Analgesia, 37(5), 434– 439. https://doi.org/10.1111/j.1467-2995.2010.00557.x

Cole, D. J., Kalichman, M. W., Shapiro, H. M., & Drummond, J. C. (1990). The nonlinear potency of sub-MAC concentrations of nitrous oxide in decreasing the anesthetic requirement of enflurane, halothane, and isoflurane in rats. Anesthesiology, 73(1), 93–99.

Conner, J. T., Katz, R. L., Pagano, R. R., & Graham, C. W. (1978). RO 21-3981 for intravenous surgical premedication and induction of anesthesia. Anesthesia and Analgesia, 57(1), 1–5.

Cortright, K. A., Wetzlich, S. E., & Craigmill, A. L. (2007). Plasma pharmacokinetics of midazolam in chickens, turkeys, pheasants and bobwhite quail. Journal of Veterinary Pharmacology and Therapeutics, 30(5), 429–436. https://doi.org/10.1111/j.1365-2885.2007.00880.x

Court, M. H., & Greenblatt, D. J. (1992). Pharmacokinetics and preliminary observations of behavioral changes following administration of midazolam to dogs. Journal of Veterinary Pharmacology and Therapeutics, 15(4), 343–350.

Curro, T. G., Brunson, D. B., & Paul-Murphy, J. (1995). Determination of the ED50 of isoflurane and evaluation of the isoflurane-sparing effect of butorphanol in cockatoos (Cacatua spp.). Veterinary Surgery, 23(5), 429–433.

Custer, R. S., & Bush, M. (1980). Physiologic and acid-base measures of gopher snakes during ketamine or halothane-nitrous oxide anesthesia. Journal of the American Veterinary Medical Association, 177(9), 870–874. da Rocha, R. W., Escobar, A., Pypendop, B. H., Zangirolami Filho, D., Thiesen, R., & Gava, F. N. (2017). Effects of a single intravenous bolus of fentanyl on the minimum anesthetic concentration of isoflurane in chickens (Gallus gallus domesticus). Veterinary Anaesthesia and Analgesia, 44(3), 546–554. https://doi.org/10.1016/j.vaa.2016.07.006

Day, T. K., & Roge, C. K. (1996). Evaluation of sedation in quail induced by use of midazolam and reversed by use of flumazenil. Journal of the American Veterinary Medical Association, 209(5), 969–971. de Jong, R. H., & Eger, E. I. 2nd. (1975). MAC expanded: AD50 and AD95 values of common inhalation anesthetics in man. Anesthesiology, 42(4), 384–389. de Segura, I. A. G., de la Víbora, J. B., & Criado, A. (2009). Determination of the minimum alveolar concentration for halothane, isoflurane and sevoflurane in the 89

gerbil. Laboratory Animals, 43(3), 239–242. https://doi.org/10.1258/la.2008.006065

De Simone, S. B. S., Hirano, L. Q. L., & Santos, A. L. Q. (2017). Effects of midazolam in different doses in redtail boa Boa constrictor Linnaeus, 1758 (Squamata: Boidae). Ciência Animal Brasileira, 18, 1–10. https://doi.org/10.1590/1089-6891v18e-22230

Deady, J. E., Koblin, D. D., Eger, E. I. 2nd, Heavner, J. E., & D’Aoust, B. (1981). Anesthetic potencies and the unitary theory of narcosis. Anesthesia and Analgesia, 60(6), 380–384.

DeYoung, D. J., & Sawyer, D. C. (1965). Anesthetic potency of nitrous oxide during halothane anesthesia in the dog. Journal of the American Animal Hospital Association, 16(1), 125–128.

DiFazio, C. A., Brown, R. E., Ball, C. G., Heckel, C. G., & Kennedy, S. S. (1972). Additive effects of anesthetics and theories of anesthesia. Anesthesiology, 36(1), 57–63.

Doherty, T. J., Rohrbach, B. W., & Geiser, D. R. (2002). Effect of acepromazine and butorphanol on isoflurane minimum alveolar concentration in goats. Journal of Veterinary Pharmacology and Therapeutics, 25(1), 65–67.

Doherty, T. J., Rohrbach, B. W., Ross, L., & Schultz, H. (2002). The effect of tiletamine and zolazepam on isoflurane minimum alveolar concentration in goats. Journal of Veterinary Pharmacology and Therapeutics, 25(3), 233–235.

Doherty, T. J., Will, W. A., Rohrbach, B. W., & Geiser, D. R. (2004). Effect of morphine and flunixin meglumine on isoflurane minimum alveolar concentration in goats. Veterinary Anaesthesia and Analgesia, 31(2), 97–101. https://doi.org/10.1111/j.1467-2987.2004.00156.x

Dohm, L., & Brunson, D. (1998). Effective dose of isoflurane for the desert iguana (Dipsosaurus dorsalis) and the effect of hypothermia on effective dose. Proceedings of the American College of Veterinary Anesthetists, 543.

Doss, G. A., Fink, D. M., Sladky, K. K., & Mans, C. (2017). Comparison of subcutaneous dexmedetomidine–midazolam versus alfaxalone–midazolam sedation in leopard geckos (Eublepharis macularius). Veterinary Anaesthesia and Analgesia, 44(5), 1175–1183. https://doi.org/10.1016/j.vaa.2017.03.007

Driggers, T. (2000). Respiratory diseases, diagnostics, and therapy in snakes. Veterinary Clinics of North America: Exotic Animal Practice, 3(2), 519–530.

Drummond, J. C. (1985). MAC for halothane, enflurane, and isoflurane in the New Zealand white rabbit: and a test for the validity of MAC determinations. Anesthesiology, 62(3), 336–338.

Drummond, J. C., Todd, M. M., & Shapiro, H. M. (1983). Minimal alveolar concentrations for halothane, enflurane, and isoflurane in the cat. Journal of the American Veterinary Medical Association, 182(10), 1099–1101.

Dupré, R. K., Hicks, J. W., & Wood, S. C. (1989). Effect of temperature on chemical control of ventilation in Mexican black iguanas. American Journal of Physiology, 257(5 Pt 2), R1258-63. https://doi.org/10.1152/ajpregu.1989.257.5.R1258

Eger, E. I. 2nd. (1981). Isoflurane: a review. Anesthesiology, 55(5), 559–576.

Eger, E. I. 2nd, Brandstater, B., Saidman, L. J., Regan, M. J., Severinghaus, J. W., & Munson, E. S. (1965). Equipotent alveolar concentrations of methoxyflurane, halothane, diethyl ether, fluroxene, cyclopropane, xenon and nitrous oxide in the dog. Anesthesiology, 26(6), 771–777.

Eger, E. I. 2nd, Johnson, B. H., Weiskopf, R. B., Holmes, M. A., Yasuda, N., Targ, A., & Rampil, I. J. (1988). Minimum alveolar concentration of 1-653 and isoflurane in pigs: definition of a supramaximal stimulus. Anesthesia & Analgesia, 67(12), 1174– 1176.

Eger, E. I. 2nd, & Larson, C. P. (1964). Anaesthetic solubility in blood and tissues: values and significance. British Journal of Anaesthesia, 36, 140–144.

Eger, E. I. 2nd, Saidman, L. J., & Brandstater, B. (1965). Minimum alveolar anesthetic concentration: a standard of anesthetic potency. Anesthesiology, 26(6), 756–763.

Eisele, P. H., Talken, L., & Eisele, J. H. (1985). Potency of isoflurane and nitrous oxide in conventional swine. Laboratory Animal Science, 35(1), 76–78.

Emery, L., Parsons, G., Gerhardt, L., Schumacher, J., & Souza, M. (2014). Sedative effects of intranasal midazolam and dexmedetomidine in 2 species of tortoises (Chelonoidis carbonaria and Geochelone platynota). Journal of Exotic Pet Medicine, 23(4), 380–383. https://doi.org/10.1053/j.jepm.2014.07.015

Enok, S., Leite, G. S. P. C., Leite, C. A. C., Gesser, H., Hedrick, M. S., & Wang, T. (2016). Improved cardiac filling facilitates the postprandial elevation of stroke volume in Python regius. The Journal of Experimental Biology, 219(Pt 19), 3009– 3018. https://doi.org/10.1242/jeb.142729

Epstein, R. M., Rackow, H., Salanitre, E., & Wolf, G. L. (1964). Influence of the concentration effect on the uptake of anesthetic mixtures: the second gas effect. Anesthesiology, 25, 364–371.

Escobar, A., da Rocha, R. W., Midon, M., de Almeida, R. M., Filho, D. Z., & Werther, K. (2017). Effects of tramadol on the minimum anesthetic concentration of isoflurane in white-eyed parakeets (Psittacara leucophthalmus). Journal of Zoo and Wildlife Medicine, 48(2), 380–387. https://doi.org/10.1638/2016-0180R1.1 91

Escobar, A., da Rocha, R. W., Pypendop, B. H., Zangirolami Filho, D., Sousa, S. S., & Valadão, C. A. A. (2016). Effects of methadone on the minimum anesthetic concentration of isoflurane, and its effects on heart rate, blood pressure and ventilation during isoflurane anesthesia in hens (Gallus gallus domesticus). PLoS ONE, 11(3), e0152546. https://doi.org/10.1371/journal.pone.0152546

Farrell, A. P., Gamperl, A. K., & Francis, E. T. B. (1998). Comparative aspects of heart morphology. In C. Gans & A. S. Gaunt (Eds.), Biology of the Reptilia, Vol. 19 (Morphology G: Visceral Organs). (pp. 375–424). Ithaca: Society for the Study of Amphibians and Reptiles.

Feder, M. E., & Burggren, W. W. (1985a). Cutaneous gas exchange in vertebrates: design, patterns, control and implications. Biological Reviews of the Cambridge Philosophical Society, 60(1), 1–45.

Feder, M. E., & Burggren, W. W. (1985b). Skin breathing in vertebrates. Scientific American, 253(5), 126–143.

Figueiró, M. R., Soares, J. H., Ascoli, F. O., Werre, S., & Gómez de Segura, I. Á. (2016). Isoflurane MAC determination in dogs using three intensities of constant- current electrical stimulation. Veterinary Anaesthesia and Analgesia, 43(5), 464– 471. https://doi.org/10.1111/vaa.12341

Filogonio, R., Costa Leite, C. A., & Wang, T. (2017). Vascular distensibilities have minor effects on intracardiac shunt patterns in reptiles. Zoology, 122, 46–51. https://doi.org/10.1016/j.zool.2017.02.005

Fink, D. M., Doss, G. A., Sladky, K. K., & Mans, C. (2018). Effect of injection site on dexmedetomidine-ketamine induced sedation in leopard geckos (Eublepharis macularius). Journal of the American Veterinary Medical Association, 253(9), 1146–1150. https://doi.org/10.2460/javma.253.9.1146

Fitzgerald, G., & Blais, D. (1991). Effects of nitrous oxide on the minimum anaesthetic dose of isoflurane in pigeons and red-tailed hawks. Proceedings of the 4th International Congress of Veterinary Anaesthesia, 18(s1), 111–115. https://doi.org/10.1111/j.1467-2995.1991.tb00526.x

Floriano, B. P., Wagatsuma, J. T., Ferreira, J. Z., Abimussi, C. J. X., Menegheti, T. M., Santos, P. S. P., & Oliva, V. N. L. S. (2016). Effects on indicators of tissue perfusion in dogs anesthetized with isoflurane at two multiples of the minimum alveolar concentration. American Journal of Veterinary Research, 77(1), 24–31. https://doi.org/10.2460/ajvr.77.1.24

Fourcade, H. E., Stevens, W. C., Larson, C. P., Cromwell, T. H., Bahlman, S. H., Hickey, R. F., … Eger, E. I. 2nd. (1971). The ventilatory effects of Forane, a new inhaled anesthetic. Anesthesiology, 35(1), 26–31.

Franks, N. P., & Lieb, W. R. (1984). Do general anaesthetics act by competitive binding to specific receptors? Nature, 310(5978), 599–601.

Furilla, R. A., & Bartlett, D. (1989). Intrapulmonary CO2 inhibits inspiration in garter snakes. Respiration Physiology, 78(2), 207–217.

Gans, Carl, & Clark, B. D. (1978). Air flow in reptilian ventilation. Comparative Biochemistry and Physiology Part A: Physiology, 60(4), 453–457. https://doi.org/10.1016/0300-9629(78)90016-6

Gatson, B. J., Goe, A., Granone, T. D., & Wellehan, J. F. X. (2017). Intramuscular epinephrine results in reduced anesthetic recovery time in American alligators (Alligator mississippiensis) undergoing isoflurane anesthesia. Journal of Zoo and Wildlife Medicine, 48(1), 55–61. https://doi.org/10.1638/2015-0293.1

Gianotti, G., Valverde, A., Johnson, R., Sinclair, M., Gibson, T., & Dyson, D. H. (2014). Influence of prior determination of baseline minimum alveolar concentration (MAC) of isoflurane on the effect of ketamine on MAC in dogs. Canadian Journal of Veterinary Research, 78(3), 207–213.

Glass, M. L., Boutilier, R. G., & Heisler, N. (1983). Ventilatory control of arterial PO2 in the turtle Chrysemys picta belli: effects of temperature and hypoxia. Journal of Comparative Physiology B, 151(2), 145–153. https://doi.org/10.1007/BF00689912

Glass, M. L., Boutilier, R. G., & Heisler, N. (1985). Effects of body temperature on respiration, blood gases and acid-base status in the turtle Chrysemys picta belli. Journal of Experimental Biology, 114(1), 37–51.

Glass, M. L., Burggren, W. W., & Johansen, K. (1978). Ventilation in an aquatic and a terrestrial chelonian reptile. The Journal of Experimental Biology, 72, 165–179.

Glass, M. L., & Johansen, K. (1976). Control of breathing in Acrochordus javanicus, an aquatic snake. Physiological Zoology, 49(3), 328–340. https://doi.org/10.1086/physzool.49.3.30155691

Goe, A., Shmalberg, J., Gatson, B., Bartolini, P., Curtiss, J., & Wellehan, J. F. X. (2016). Epinephrine or GV-26 electrical stimulation reduces inhalant anesthetic recovery time in common snapping turtles (Chelydra serpentina). Journal of Zoo and Wildlife Medicine, 47(2), 501–507. https://doi.org/10.1638/2015-0264.1

Gonsowski, C. T., & Eger, E. I. 2nd. (1994). Nitrous oxide minimum alveolar anesthetic concentration in rats is greater than previously reported. Anesthesia and Analgesia, 79(4), 710–712.

Graham, J. B. (1974). Aquatic respiration in the sea snake, Pelamis platurus. Respiration Physiology, 21(1), 1–7.

Greenblatt, D. J., Abernethy, D. R., Locniskar, A., Harmatz, J. S., Limjuco, R. A., & Shader, R. I. (1984). Effect of age, gender, and obesity on midazolam kinetics. Anesthesiology, 61(1), 27–35.

Greiner, A. S., & Larach, D. R. (1989). The effect of benzodiazepine receptor antagonism by flumazenil on the MAC of halothane in the rat. Anesthesiology, 70(4), 644–648.

Greunz, E. M., Williams, C., Ringgaard, S., Hansen, K., Wang, T., & Bertelsen, M. F. (2018). Elimination of intracardiac shunting provides stable gas anesthesia in tortoises. Scientific Reports, 8(1), 17124. https://doi.org/10.1038/s41598-018- 35588-w

Gruss, M., Bushell, T. J., Bright, D. P., Lieb, W. R., Mathie, A., & Franks, N. P. (2004). Two-pore-domain K+ channels are a novel target for the anesthetic gases xenon, nitrous oxide, and cyclopropane. Molecular Pharmacology, 65(2), 443–452. https://doi.org/10.1124/mol.65.2.443

Guedel, A. E. (1951). Inhalation Anesthesia: A Fundamental Guide (2nd ed.). New York: Macmillan.

Hall, R. I., Schwieger, I. M., & Hug, C. C. (1988). The anesthetic efficacy of midazolam in the enflurane-anesthetized dog. Anesthesiology, 68(6), 862–866.

Hall, R. I., Szlam, F., & Hug, C. C. (1988). Pharmacokinetics and pharmacodynamics of midazolam in the enflurane-anesthetized dog. Journal of Pharmacokinetics and Biopharmaceutics, 16(3), 251–262.

Harvey-Clark, C. (1993). Midazolam fails to sedate painted turtles (Chrysemys picta). Bulletin of the Association of Reptilian and Amphibian Veterinarians, 3, 7–8.

Heard, D. J. (2001). Reptile anesthesia. Veterinary Clinics of North America: Exotic Animal Practice, 4(1), 83–117.

Hecker, K. E., Baumert, J. H., Horn, N., Reyle-Hahn, M., Heussen, N., & Rossaint, R. (2003). Minimum anesthetic concentration of sevoflurane with different xenon concentrations in swine. Anesthesia & Analgesia, 97(5), 1364–1369.

Heisler, N., & Glass, M. L. (1985). Mechanisms and regulation of central vascular shunts in reptiles. In K. Johansen & Alfred Benzon Foundation (Eds.), Cardiovascular Shunts: Phylogenetic, Ontogenetic and Clinical Aspects (1st ed., pp. 334–353). New York: Munksgaard.

Heizmann, P., Eckert, M., & Ziegler, W. H. (1983). Pharmacokinetics and bioavailability of midazolam in man. British Journal of Clinical Pharmacology, 16 Suppl 1, 43S- 49S.

Hellyer, P. W., Mama, K. R., Shafford, H. L., Wagner, A. E., & Kollias-Baker, C. (2001). Effects of diazepam and flumazenil on minimum alveolar concentrations for dogs anesthetized with isoflurane or a combination of isoflurane and fentanyl. American Journal of Veterinary Research, 62(4), 555–560.

Hernandez-Divers, S. M., Schumacher, J., Stahl, S., & Hernandez-Divers, S. J. (2005). Comparison of isoflurane and sevoflurane anesthesia after premedication with butorphanol in the green iguana (Iguana iguana). Journal of Zoo and Wildlife Medicine, 36(2), 169–175. https://doi.org/10.1638/04-057.1

Hicks, J. W. (1994). Adrenergic and cholinergic regulation of intracardiac shunting. Physiological Zoology, 67(6), 1325–1346. https://doi.org/10.1086/physzool.67.6.30163900

Hicks, J. W. (1998). Cardiac shunting in reptiles: mechanisms, regulation, and physiological functions. In C. Gans & A. S. Gaunt (Eds.), Biology of the Reptilia, Vol. 19 (Morphology G: Visceral Organs). (pp. 425–483). Ithaca: Society for the Study of Amphibians and Reptiles.

Hicks, J. W., & Comeau, S. (1994). Vagal regulation of intracardiac shunting in the turtle Pseudemys scripta. Journal of Experimental Biology, 186, 109–126.

Hikasa, Y., Okuyama, K., Kakuta, T., Takase, K., & Ogasawara, S. (1998). Anesthetic potency and cardiopulmonary effects of sevoflurane in goats: comparison with isoflurane and halothane. Canadian Journal of Veterinary Research, 62(4), 299– 306.

Hillman, S. S., Hedrick, M. S., & Kohl, Z. F. (2017). Commentary on: “Vascular distensibilities have minor effects on intracardiac shunt patterns in reptiles” by Filogonio et al. (2017). Zoology, 122, 52–54. https://doi.org/10.1016/j.zool.2017.05.004

Hoffman, E. J., & Warren, E. W. (1993). Flumazenil: a benzodiazepine antagonist. Clinical Pharmacy, 12(9), 641–656; quiz 699–701.

Holazo, A. A., Winkler, M. B., & Patel, I. H. (1988). Effects of age, gender and oral contraceptives on intramuscular midazolam pharmacokinetics. Journal of Clinical Pharmacology, 28(11), 1040–1045.

Holz, P. H. (1999). The reptilian renal portal system - A review. Bulletin of the Association of Reptilian and Amphibian Veterinarians, 9(1), 4–14. https://doi.org/10.5818/1076-3139.9.1.4

Holz, P. H., Barker, I. K., Burger, J. P., Crawshaw, G. J., & Conlon, P. D. (1997). The effect of the renal portal system on pharmacokinetic parameters in the red-eared slider (Trachemys scripta elegans). Journal of Zoo and Wildlife Medicine, 28(4), 386–393. 95

Holz, P. H., Barker, I. K., Crawshaw, G. J., & Dobson, H. (1997). The anatomy and perfusion of the renal portal system in the red-eared slider (Trachemys scripta elegans). Journal of Zoo and Wildlife Medicine, 28(4), 378–385.

Holz, P. H., Burger, J. P., Pasloske, K., Baker, R., & Young, S. (2002). Effect of injection site on carbenicillin pharmacokinetics in the carpet python, Morelia spilota. Journal of Herpetological Medicine and Surgery, 12(4), 12–16. https://doi.org/10.5818/1529-9651.12.4.12

Holz, P. H., & Holz, R. M. (1994). Evaluation of ketamine, ketamine/xylazine, and ketamine/midazolam anesthesia in red-eared sliders (Trachemys scripta elegans). Journal of Zoo and Wildlife Medicine, 25(4), 531–537.

Hornbein, T. F., Eger, E. I. 2nd, Winter, P. M., Smith, G., Wetstone, D., & Smith, K. H. (1982). The minimum alveolar concentration of nitrous oxide in man. Anesthesia & Analgesia, 61(7), 553–556.

Hubbell, J. A. E., Kelly, E. M., Aarnes, T. K., Bednarski, R. M., Lerche, P., Liu, Z., & Lakritz, J. (2013). Pharmacokinetics of midazolam after intravenous administration to horses. Equine Veterinary Journal, 45(6), 721–725. https://doi.org/10.1111/evj.12049

Ide, T., Sakurai, Y., Aono, M., & Nishino, T. (1998). Minimum alveolar anesthetic concentrations for airway occlusion in cats: a new concept of minimum alveolar anesthetic concentration-airway occlusion response. Anesthesia & Analgesia, 86(1), 191–197.

Ilkiw, J. E., Pascoe, P. J., & Fisher, L. D. (1997). Effect of alfentanil on the minimum alveolar concentration of isoflurane in cats. American Journal of Veterinary Research, 58(11), 1274–1279.

Ilkiw, J. E., Suter, C. M., Farver, T. B., McNeal, D., & Steffey, E. P. (1996). The behaviour of healthy awake cats following intravenous and intramuscular administration of midazolam. Journal of Veterinary Pharmacology and Therapeutics, 19(3), 205–216.

Ilkiw, J. E., Suter, C. M., McNeal, D., Farver, T. B., & Steffey, E. P. (1996). The effect of intravenous administration of variable-dose midazolam after fixed-dose ketamine in healthy awake cats. Journal of Veterinary Pharmacology and Therapeutics, 19(3), 217–224.

Imai, A., Ilkiw, J. E., Pypendop, B. H., Farver, T. B., & Steffey, E. P. (2002). Nitrous oxide does not consistently reduce isoflurane requirement in cats. Veterinary Anaesthesia and Analgesia, 29(2), 98. https://doi.org/10.1046/j.1467- 2995.2002.00078_4.x

Imai, A., Steffey, E. P., Farver, T. B., & Ilkiw, J. E. (1999). Assessment of isoflurane- 96

induced anesthesia in ferrets and rats. American Journal of Veterinary Research, 60(12), 1577–1583.

Imai, A., Steffey, E. P., Ilkiw, J. E., & Farver, T. B. (1999). Comparison of clinical signs and hemodynamic variables used to monitor rabbits during halothane- and isoflurane-induced anesthesia. American Journal of Veterinary Research, 60(10), 1189–1195.

Inagaki, Y., Sumikawa, K., & Yoshiya, I. (1993). Anesthetic interaction between midazolam and halothane in humans. Anesthesia and Analgesia, 76(3), 613–617.

Isaza, R., Andrews, G., Coke, R., & Hunter, R. (2004). Assessment of multiple cardiocentesis in ball pythons (Python regius). Journal of the American Veterinary Medical Association, 43(6), 35–38.

Jackson, D. C. (1971). The effect of temperature on ventilation in the turtle, Pseudemys scripta elegans. Respiration Physiology, 12(2), 131–140.

Jackson, D. C. (1973). Ventilatory response to hypoxia in turtles at various temperatures. Respiration Physiology, 18(2), 178–187.

Jacobson, E. R. (2007). Overview of reptile biology, anatomy, and histology. In E. R. Jacobson (Ed.), Infectious diseases and pathology of reptiles (1st ed., pp. 1–130). Boca Raton, FL: Taylor & Francis Group.

Jakobsen, S. L., Williams, C. J. A., Wang, T., & Bertelsen, M. F. (2017). The influence of mechanical ventilation on physiological parameters in ball pythons (Python regius). Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 207, 30–35. https://doi.org/10.1016/j.cbpa.2017.02.012

James, L. E., Williams, C. J. A., Wang, T., & Bertelsen, M. F. (2018). Cardiovascular effects of alfaxalone in the ball python (Python regius) and South American rattlesnake (Crotalus durissus). Proceedings of the American Association of Zoo Veterinarians, 225. Prague, Czech Republic.

James, L. E., Williams, C. J., Bertelsen, M. F., & Wang, T. (2018). Anaesthetic induction with alfaxalone in the ball python (Python regius): dose response and effect of injection site. Veterinary Anaesthesia and Analgesia, 45(3), 329–337. https://doi.org/10.1016/j.vaa.2017.12.003

Jensen, B., Larsen, C. K., Nielsen, J. M., Simonsen, L. S., & Wang, T. (2011). Change of cardiac function, but not form, in postprandial pythons. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 160(1), 35–42. https://doi.org/10.1016/j.cbpa.2011.04.018

Jensen, B., Moorman, A. F. M., & Wang, T. (2014). Structure and function of the hearts of lizards and snakes. Biological Reviews, 89(2), 302–336. 97

https://doi.org/10.1111/brv.12056

Jensen, B., Nielsen, J. M., Axelsson, M., Pedersen, M., Lofman, C., & Wang, T. (2010). How the python heart separates pulmonary and systemic blood pressures and blood flows. Journal of Experimental Biology, 213(10), 1611–1617. https://doi.org/10.1242/jeb.030999

Jensen, B., Nyengaard, J. R., Pedersen, M., & Wang, T. (2010). Anatomy of the python heart. Anatomical Science International, 85(4), 194–203. https://doi.org/10.1007/s12565-010-0079-1

Jevtović-Todorović, V., Todorović, S. M., Mennerick, S., Powell, S., Dikranian, K., Benshoff, N., … Olney, J. W. (1998). Nitrous oxide (laughing gas) is an NMDA antagonist, neuroprotectant and neurotoxin. Nature Medicine, 4(4), 460–463.

Jones, D. J., Stehling, L. C., & Zauder, H. L. (1979). Cardiovascular responses to diazepam and midazolam maleate in the dog. Anesthesiology, 51(5), 430–434.

Kim, Y. K., Lee, S. S., Suh, E. H., Lee, L., Lee, H. C., Lee, H. J., & Yeon, S. C. (2011). Minimum anesthetic concentration and cardiovascular dose–response relationship of isoflurane in cinereous vultures (Aegypius monachus). Journal of Zoo and Wildlife Medicine, 42(3), 499–503. https://doi.org/10.1638/2010-0151.1

Kischinovsky, M., Duse, A., Wang, T., & Bertelsen, M. F. (2013). Intramuscular administration of alfaxalone in red-eared sliders (Trachemys scripta elegans) – effects of dose and body temperature. Veterinary Anaesthesia and Analgesia, 40(1), 13–20. https://doi.org/10.1111/j.1467-2995.2012.00745.x

Ko, H., Kaye, A. D., & Urman, R. D. (2014). Nitrous oxide and perioperative outcomes. Journal of Anesthesia, 28(3), 420–428. https://doi.org/10.1007/s00540-013-1729-y

Kohn, M. A., Jarrett, M. S., & Senyak, J. (2018). Sample Size Calculators. Retrieved June 1, 2018, from University of California, San Francisco, Clinical & Translational Science Institute website: http://www.sample-size.net/

Komai, H., & Rusy, B. F. (1987). Negative inotropic effects of isoflurane and halothane in rabbit papillary muscles. Anesthesia and Analgesia, 66(1), 29–33.

Kropf, J., & Hughes, J. M. L. (2018). Effects of midazolam on cardiovascular responses and isoflurane requirement during elective ovariohysterectomy in dogs. Irish Veterinary Journal, 71(1), 26. https://doi.org/10.1186/s13620-018-0136-y

Kummrow, M. S., Tseng, F., Hesse, L., & Court, M. (2008). Pharmacokinetics of buprenorphine after single-dose subcutaneous administration in red-eared sliders (Trachemys scripta elegans). Journal of Zoo and Wildlife Medicine, 39(4), 590–595. https://doi.org/10.1638/2008-0033.1

Lacoste, L., Bouquet, S., Ingrand, P., Caritez, J. C., Carretier, M., & Debaene, B. (2000). Intranasal midazolam in piglets: pharmacodynamics (0.2 vs 0.4 mg/kg) and pharmacokinetics (0.4 mg/kg) with bioavailability determination. Laboratory Animals, 34(1), 29–35. https://doi.org/10.1258/002367700780578073

Lampe, G. H., Donegan, J. H., Rupp, S. M., Wauk, L. Z., Whitendale, P., Fouts, K. E., … Wilson, C. B. (1990). Nitrous oxide and epinephrine-induced arrhythmias. Anesthesia and Analgesia, 71(6), 602–605.

Laster, M. J., Liu, J., Eger, E. I. 2nd, & Taheri, S. (1993). Electrical stimulation as a substitute for the tail clamp in the determination of minimum alveolar concentration. Anesthesia & Analgesia, 76(6), 1310–1312.

Lemke, K. A. (2007). Anticholinergics and Sedatives. In W. J. Tranquilli, J. C. Thurmon, & K. A. Grimm (Eds.), Lumb & Jones’ Veterinary Anesthesia and Analgesia (4th ed., pp. 203–240). Oxford: Blackwell Publishing Ltd.

Lerman, J., Oyston, J. P., Gallagher, T. M., Miyasaka, K., Volgyesi, G. A., & Burrows, F. A. (1990). The minimum alveolar concentration (MAC) and hemodynamic effects of halothane, isoflurane, and sevoflurane in newborn swine. Anesthesiology, 73(4), 717–721.

Litchfield, J. T. J., & Wilcoxon, F. (1949). A simplified method of evaluating dose-effect experiments. Journal of Pharmacology and Experimental Therapeutics, 96(2), 99– 113.

Liu, W. S., Wong, K. C., Port, J. D., & Andriano, K. P. (1982). Epinephrine-induced arrhythmias during halothane anesthesia with the addition of nitrous oxide, nitrogen, or helium in dogs. Anesthesia and Analgesia, 61(5), 414–417.

Lopes, I. G., Armelin, V. A., Braga, V. H. da S., & Florindo, L. H. (2017). The influence of midazolam on heart rate arises from cardiac autonomic tones alterations in Burmese pythons, Python molurus. Autonomic Neuroscience, 208, 103–112. https://doi.org/10.1016/j.autneu.2017.10.008

Ludders, J. W., Mitchell, G. S., & Rode, J. (1990). Minimal anesthetic concentration and cardiopulmonary dose response of isoflurane in ducks. Veterinary Surgery, 19(4), 304–307.

Ludders, J. W., Rode, J., & Mitchell, G. S. (1989). Isoflurane anesthesia in sandhill cranes (Grus canadensis): minimal anesthetic concentration and cardiopulmonary dose-response during spontaneous and controlled breathing. Anesthesia and Analgesia, 68(4), 511–516.

Lundeen, G., Manohar, M., & Parks, C. (1983). Systemic distribution of blood flow in swine while awake and during 1.0 and 1.5 MAC isoflurane anesthesia with or without 50% nitrous oxide. Anesthesia and Analgesia, 62(5), 499–512. 99

Lynch, C., Baum, J., & Tenbrinck, R. (2000). Xenon anesthesia. Anesthesiology, 92(3), 865–868.

Maas, A., & Brunson, D. B. (2002). Comparison of anesthetic potency and cardiopulmonary effects of isoflurane and sevoflurane in colubrid snakes. Proceedings of the American Association of Zoo Veterinarians, 1–3.

Mader, D. R., Conzelman, G. M., & Baggot, J. D. (1985). Effects of ambient temperature on the half-life and dosage regimen of amikacin in the gopher snake. Journal of the American Veterinary Medical Association, 187(11), 1134–1136.

Malte, C. L., Jakobsen, S. L., & Wang, T. (2014). A critical evaluation of automated blood gas measurements in comparative respiratory physiology. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 178, 7– 17. https://doi.org/10.1016/j.cbpa.2014.07.022

Mans, C., Guzman, D. S.-M., Lahner, L. L., Paul-Murphy, J., & Sladky, K. K. (2012). Sedation and physiologic response to manual restraint after intranasal administration of midazolam in Hispaniolan amazon parrots (Amazona ventralis). Journal of Avian Medicine and Surgery, 26(3), 130–139. https://doi.org/10.1647/2011-037R.1

Mans, C., & Sladky, K. K. (2012). Endoscopically guided removal of cloacal calculi in three African spurred tortoises (Geochelone sulcata). Journal of the American Veterinary Medical Association, 240(7), 869–875. https://doi.org/10.2460/javma.240.7.869

Martin-Jurado, O., Vogt, R., Kutter, A. P. N., Bettschart-Wolfensberger, R., & Hatt, J.-M. (2008). Effect of inhalation of isoflurane at end-tidal concentrations greater than, equal to, and less than the minimum anesthetic concentration on bispectral index in chickens. American Journal of Veterinary Research, 69(10), 1254–1261. https://doi.org/10.2460/ajvr.69.10.1254

Matthews, N. S., Dollar, N. S., & Shawley, R. V. (1990). Halothane-sparing effect of benzodiazepines in ponies. Cornell Vet, 80(3), 259–265.

Mazze, R. I., Rice, S. A., & Baden, J. M. (1985). Halothane, isoflurane, and enflurane MAC in pregnant and nonpregnant female and male mice and rats. Anesthesiology, 62(3), 339–341.

McFadden, M. S., Bennett, R. A., Reavill, D. R., Ragetly, G. R., & Clark-Price, S. C. (2011). Clinical and histologic effects of intracardiac administration of propofol for induction of anesthesia in ball pythons (Python regius). Journal of the American Veterinary Medical Association, 239(6), 803–807. https://doi.org/10.2460/javma.239.6.803

McKindley, D. S., Hanes, S., & Boucher, B. A. (1998). Hepatic drug metabolism in 100

critical illness. Pharmacotherapy, 18(4), 759–778.

Melvin, M. A., Johnson, B. H., Quasha, A. L., & Eger, E. I. 2nd. (1982). Induction of anesthesia with midazolam decreases halothane MAC in humans. Anesthesiology, 57(3), 238–241.

Mercado, J. A., Larsen, R. S., Wack, R. F., & Pypendop, B. H. (2008). Minimum anesthetic concentration of isoflurane in captive thick-billed parrots (Rhynchopsitta pachyrhyncha). American Journal of Veterinary Research, 69(2), 189–194. https://doi.org/10.2460/ajvr.69.2.189

Merkel, G., & Eger, E. I. 2nd. (1963). A comparative study of halothane and halopropane anesthesia including method for determining equipotency. Anesthesiology, 24, 346–357.

Meyer, H. H. (1899). Zur theorie der akoholnarkose. Erste mittheilung. Welche eigenschaft der anästhetica bedingt ihre narkotische wirkung? Archiv for Experimentelle Pathologie Und Pharmakologie, 42, 109–118.

Miller, K. W., Paton, W. D., Smith, E. B., & Smith, R. A. (1972). Physicochemical approaches to the mode of action of general anesthetics. Anesthesiology, 36(4), 339–351.

Milsom, W. K. (1990). Control and co-ordination of gas exchange in air breathers. In R. G. Boutilier (Ed.), Advances in Comparative Environmental Physiology (pp. 347– 400). https://doi.org/10.1007/978-3-642-75380-0_10

Milsom, W. K. (1995). Regulation of respiration in lower vertebrates: role of CO2/pH chemoreceptors. In N. Heisler (Ed.), Advances in Comparative and Environmental Physiology (pp. 61–104). https://doi.org/10.1007/978-3-642-79666-1_4

Mitchell, G. S., Gleeson, T. T., & Bennett, A. F. (1981). Pulmonary oxygen transport during activity in lizards. Respiration Physiology, 43(3), 365–375.

Mitchell, M. A. (2009). Reptile cardiology. Veterinary Clinics of North America: Exotic Animal Practice, 12(1), 65–79. https://doi.org/10.1016/j.cvex.2008.10.001

Monteiro, E. R., Coelho, K., Bressan, T. F., Simões, C. R., & Monteiro, B. S. (2016). Effects of acepromazine-morphine and acepromazine-methadone premedication on the minimum alveolar concentration of isoflurane in dogs. Veterinary Anaesthesia and Analgesia, 43(1), 27–34. https://doi.org/10.1111/vaa.12265

Mosley, C. A. E. (2005). Anesthesia and analgesia in reptiles. Seminars in Avian and Exotic Pet Medicine, 14(4), 243–262. https://doi.org/10.1053/j.saep.2005.09.005

Mosley, C. A. E., Dyson, D., & Smith, D. A. (2003a). Minimum alveolar concentration of isoflurane in green iguanas and the effect of butorphanol on minimum alveolar 101

concentration. Journal of the American Veterinary Medical Association, 222(11), 1559–1564.

Mosley, C. A. E., Dyson, D., & Smith, D. A. (2003b). The cardiac anesthetic index of isoflurane in green iguanas. Journal of the American Veterinary Medical Association, 222(11), 1565–1568.

Mosley, C. A. E., Dyson, D., & Smith, D. A. (2004). The cardiovascular dose-response effects of isoflurane alone and combined with butorphanol in the green iguana (Iguana iguana). Veterinary Anaesthesia and Analgesia, 31(1), 64–72.

Mosley, C. I., & Mosley, C. A. E. (2015). Comparative anesthesia and analgesia of reptiles, amphibians, and fishes. In K. A. Grimm, L. A. Lamont, W. J. Tranquilli, S. A. Greene, & S. A. Robertson (Eds.), Veterinary Anaesthesia and Analgesia: The Fifth Edition of Lumb and Jones (5th ed., pp. 784–799). Ames, IA, USA: John Wiley & Sons, Inc.

Murat, I., & Housmans, P. R. (1988). Minimum alveolar concentrations (MAC) of halothane, enflurane, and isoflurane in ferrets. Anesthesiology, 68(5), 783–786.

Murray, D. J., Mehta, M. P., & Forbes, R. B. (1991). The additive contribution of nitrous oxide to isoflurane MAC in infants and children. Anesthesiology, 75(2), 186–190.

Nagy, K. A. (2005). Field metabolic rate and body size. Journal of Experimental Biology, 208(9), 1621–1625. https://doi.org/10.1242/jeb.01553

Nishiyama, T., Matsukawa, T., & Hanaoka, K. (1998). The effects of age and gender on the optimal premedication dose of intramuscular midazolam. Anesthesia and Analgesia, 86(5), 1103–1108.

Nordt, S. P., & Clark, R. F. (1997). Midazolam: a review of therapeutic uses and toxicity. The Journal of Emergency Medicine, 15(3), 357–365.

O, O., Churgin, S. M., Sladky, K. K., & Smith, L. J. (2015). Anesthetic induction and recovery parameters in bearded dragons (Pogona vitticeps): Comparison of isoflurane delivered in 100% oxygen versus 21% oxygen. Journal of Zoo and Wildlife Medicine, 46(3), 534–539. https://doi.org/10.1638/2014-0193.1

Odou, P., Barthélémy, C., Chatelier, D., Luyckx, M., Brunet, C., Dine, T., … Robert, H. (1999). Pharmacokinetics of midazolam: comparison of sublingual and intravenous routes in rabbit. European Journal of Drug Metabolism and Pharmacokinetics, 24(1), 1–7. https://doi.org/10.1007/BF03190004

Olsen, R. W., & DeLorey, T. M. (1999). GABA receptor physiology and pharmacology. In G. J. Siegel, B. W. Agranoff, R. W. Albers, S. K. Fisher, & M. D. Uhler (Eds.), Basic Neurochemistry: Molecular, Cellular and Medical Aspects (6th ed.). Philadelphia: Lippincott-Raven. 102

Olsson, A., & Phalen, D. (2013a). Comparison of biochemical stress indicators in juvenile captive estuarine crocodiles (Crocodylus porosus) following physical restraint or chemical restraint by midazolam injection. Journal of Wildlife Diseases, 49(3), 560–567. https://doi.org/10.7589/2012-06-160

Olsson, A., & Phalen, D. (2013b). The effects of decreased body temperature on the onset, duration and action of medetomidine and its antagonist atipamezole in juvenile farmed estuarine crocodiles (Crocodylus porosus). Veterinary Anaesthesia and Analgesia, 40(3), 272–279. https://doi.org/10.1111/vaa.12008

Oppenheim, Y. C., & Moon, P. F. (1995). Sedative effects of midazolam in red-eared slider turtles (Trachemys scripta elegans). Journal of Zoo and Wildlife Medicine, 26(3), 409–413.

Overgaard, J., Stecyk, J. A. W., Farrell, A. P., & Wang, T. (2002). Adrenergic control of the cardiovascular system in the turtle Trachemys scripta. The Journal of Experimental Biology, 205(Pt 21), 3335–3345.

Overton, C. E. (1901). Studien Über die Narkose: Zugleich ein Beitrag zur Allgemeinen Pharmakologie. Jena: Gustav Fischer.

Pagel, P. S., Kampine, J. P., Schmeling, W. T., & Warltier, D. C. (1991). Influence of volatile anesthetics on myocardial contractility in vivo: desflurane versus isoflurane. Anesthesiology, 74(5), 900–907.

Palahniuk, R. J., Shnider, S. M., & Eger, E. I. 2nd. (1974). Pregnancy decreases the requirement for inhaled anesthetic agents. Anesthesiology, 41(1), 82–83.

Pavez, J. C., Hawkins, M. G., Pascoe, P. J., Knych, H. K. D., & Kass, P. H. (2011). Effect of fentanyl target-controlled infusions on isoflurane minimum anaesthetic concentration and cardiovascular function in red-tailed hawks (Buteo jamaicensis). Veterinary Anaesthesia and Analgesia, 38(4), 344–351. https://doi.org/10.1111/j.1467-2995.2011.00627.x

Perry, S. F. (1998). Lungs: comparative anatomy, functional morphology, and evolution. In C. Gans & A. S. Gaunt (Eds.), Biology of the Reptilia, Vol. 19 (Morphology G: Visceral Organs). (pp. 1–92). Ithaca: Society for the Study of Amphibians and Reptiles.

Pieri, L. (1983). Preclinical pharmacology of midazolam. British Journal of Clinical Pharmacology, 16 Suppl 1, 17S-27S.

Poupa, O., & Lindström, L. (1983). Comparative and scaling aspects of heart and body weights with reference to blood supply of cardiac fibers. Comparative Biochemistry and Physiology Part A: Physiology, 76(3), 413–421.

Pypendop, B. H., & Ilkiw, J. E. (2005). The effects of intravenous lidocaine 103

administration on the minimum alveolar concentration of isoflurane in cats. Anesthesia & Analgesia, 100(1), 97–101. https://doi.org/10.1213/01.ANE.0000139350.88158.38

Quasha, A. L., Eger, E. I. 2nd, & Tinker, J. H. (1980). Determination and applications of MAC. Anesthesiology, 53(4), 315–334.

Rampil, I. J. (1994). Anesthetic potency is not altered after hypothermic spinal cord transection in rats. Anesthesiology, 80(3), 606–610.

Rampil, I. J., & Laster, M. J. (1992). No correlation between quantitative electroencephalographic measurements and movement response to noxious stimuli during isoflurane anesthesia in rats. Anesthesiology, 77(5), 920–925.

Rampil, I. J., Mason, P., & Singh, H. (1993). Anesthetic potency (MAC) is independent of forebrain structures in the rat. Anesthesiology, 78(4), 707–712.

Randall, L. O., Heise, G. A., Schallek, W., Bagdon, R. E., Banziger, R., Boris, A., … Abrams, W. B. (1961). Pharmacological and clinical studies on Valium (T.M.) a new psychotherapeutic agent of the benzodiazepine class. Current Therapeuthic Research, Clinical and Experimental, 3, 405–425.

Randall, L. O., Schallek, W., Heise, G. A., Keith, E. F., & Bagdon, R. E. (1960). The psychosedative properties of methaminodiazepoxide. Journal of Pharmacology and Experimental Therapeutics, 129, 163–171.

Rankin, D. C. (2015). Sedatives and tranquilizers. In K. A. Grimm, L. A. Lamont, W. J. Tranquilli, S. A. Greene, & S. A. Robertson (Eds.), Veterinary Anaesthesia and Analgesia: The Fifth Edition of Lumb and Jones (5th ed., pp. 196–206). Ames, IA: John Wiley & Sons, Inc.

Read, M. R. (2004). Evaluation of the use of anesthesia and analgesia in reptiles. Journal of the American Veterinary Medical Association, 224(4), 547–552. https://doi.org/10.2460/javma.2004.224.547

Reves, J. G., Corssen, G., & Holcomb, C. (1978). Comparison of two benzodiazepines for anaesthesia induction: midazolam and diazepam. Canadian Anaesthetists’ Society Journal, 25(3), 211–214.

Reves, J. G., Fragen, R. J., Vinik, H. R., & Greenblatt, D. J. (1985). Midazolam: pharmacology and uses. Anesthesiology, 62(3), 310–324.

Rich, G. F., Sullivan, M. P., & Adams, J. M. (1990). Is distal sampling of end-tidal CO2 necessary in small subjects? Anesthesiology, 73(2), 265–268.

Riquelme, C. A., Magida, J. A., Harrison, B. C., Wall, C. E., Marr, T. G., Secor, S. M., & Leinwand, L. A. (2011). Fatty acids identified in the Burmese python promote 104

beneficial cardiac growth. Science, 334(6055), 528–531. https://doi.org/10.1126/science.1210558

Robertson, S. A., & Eberhart, S. (1994). Efficacy of the intranasal route for administration of anesthetic agents to adult rabbits. Laboratory Animal Science, 44(2), 159–165.

Robin, C., & Trieger, N. (2002). Paradoxical reactions to benzodiazepines in intravenous sedation: a report of 2 cases and review of the literature. Anesthesia Progress, 49(4), 128–132.

Russell, G. B., & Graybeal, J. M. (1995). Differences in anesthetic potency between Sprague-Dawley and Long-Evans rats for isoflurane but not nitrous oxide. Pharmacology, 50(3), 162–167.

Russell, G. B., & Graybeal, J. M. (1998). Nonlinear additivity of nitrous oxide and isoflurane potencies in rats. Canadian Journal of Anaesthesia, 45(5), 466–470. https://doi.org/10.1007/BF03012583

Sadegh, A. B. (2013). Comparison of intranasal administration of xylazine, diazepam, and midazolam in budgerigars (Melopsittacus undulatus): clinical evaluation. Journal of Zoo and Wildlife Medicine, 44(2), 241–244. https://doi.org/10.1638/2009- 0116R3.1

Saidman, L. J., & Eger, E. I. 2nd. (1964). Effect of nitrous oxide and of narcotic premedication on the alveolar concentration of halothane required for anesthesia. Anesthesiology, 25, 302–306.

Sanders, R. D., Weimann, J., & Maze, M. (2008). Biologic effects of nitrous oxide: a mechanistic and toxicologic review. Anesthesiology, 109(4), 707–722. https://doi.org/10.1097/ALN.0b013e3181870a17

Scheller, M. S., Zornow, M. H., Fleischer, J. E., Shearman, G. T., & Greber, T. F. (1989). The noncompetitive N-methyl-D-aspartate receptor antagonist, MK-801 profoundly reduces volatile anesthetic requirements in rabbits. Neuropharmacology, 28(7), 677–681.

Schnellbacher, R. W., Carpenter, J. W., Mason, D. E., KuKanich, B., Beaufrere, H., & Boysen, C. (2013). Effects of lidocaine administration via continuous rate infusion on the minimum alveolar concentration of isoflurane in New Zealand white rabbits (Oryctolagus cuniculus). American Journal of Veterinary Research, 74(11), 1377– 1384. https://doi.org/10.2460/ajvr.74.11.1377

Schumacher, J. (2003). Reptile respiratory medicine. Veterinary Clinics of North America: Exotic Animal Practice, 6(1), 213–231.

Schumacher, J., Lillywhite, H. B., Norman, W. M., & Jacobson, E. R. (1997). Effects of 105

ketamine HCl on cardiopulmonary function in snakes. Copeia, 1997(2), 395–400. https://doi.org/10.2307/1447760

Schumacher, J., & Yelen, T. (2005). Anesthesia and analgesia. In D. R. Mader (Ed.), Reptile Medicine and Surgery (2nd ed., pp. 442–452). Philadelphia: W.B. Sanders- Elsevier.

Schwartz, A. E., Maneksha, F. R., Kanchuger, M. S., Sidhu, U. S., & Poppers, P. J. (1989). Flumazenil decreases the minimum alveolar concentration isoflurane in dogs. Anesthesiology, 70(5), 764–766.

Schwartz, M., Muñana, K. R., Nettifee-Osborne, J. A., Messenger, K. M., & Papich, M. G. (2013). The pharmacokinetics of midazolam after intravenous, intramuscular, and rectal administration in healthy dogs. Journal of Veterinary Pharmacology and Therapeutics, 36(5), 471–477. https://doi.org/10.1111/jvp.12032

Schwieger, I. M., Jorge-Costa, M., Pizzolato, G. P., Forster, A., & Morel, D. R. (1994). Intrathecal midazolam reduces isoflurane MAC and increases the apnoeic threshold in rats. Canadian Journal of Anaesthesia, 41(2), 144–148. https://doi.org/10.1007/BF03009809

Schwieger, I. M., Szlam, F., & Hug, C. C. (1989). Absence of agonistic or antagonistic effect of flumazenil (Ro 15-1788) in dogs anesthetized with enflurane, isoflurane, or fentanyl-enflurane. Anesthesiology, 70(3), 477–480.

Secor, S. M., Hicks, J. W., & Bennett, A. F. (2000). Ventilatory and cardiovascular responses of a python (Python molurus) to exercise and digestion. Journal of Experimental Biology, 203(16), 2447–2454.

Seddighi, R., Egger, C. M., Rohrbach, B. W., Cox, S. K., & Doherty, T. J. (2011). The effect of midazolam on the end-tidal concentration of isoflurane necessary to prevent movement in dogs. Veterinary Anaesthesia and Analgesia, 38(3), 195–202. https://doi.org/10.1111/j.1467-2995.2011.00615.x

Seymour, R. S. (1978). Gas tensions and blood distribution in sea snakes at surface pressure and at simulated depth. Physiological Zoology, 51(4), 388–407. https://doi.org/10.1086/physzool.51.4.30160964

Seymour, R. S., & Webster, M. E. (1975). Gas transport and blood acid-base balance in diving sea snakes. Journal of Herpetological Medicine and Surgery, 191(2), 169– 181. https://doi.org/10.1002/jez.1401910204

Shaughnessy, M. R., & Hofmeister, E. H. (2014). A systematic review of sevoflurane and isoflurane minimum alveolar concentration in domestic cats. Veterinary Anaesthesia and Analgesia, 41(1), 1–13. https://doi.org/10.1111/vaa.12083

Shelton, G., & Burggren, W. (1976). Cardiovascular dynamics of the chelonia during 106

apnoea and lung ventilation. Journal of Experimental Biology, 64(2), 323–343.

Simões, C. R., Monteiro, E. R., Rangel, J. P., Nunes-Junior, J. S., & Campagnol, D. (2016). Effects of a prolonged infusion of fentanyl, with or without atropine, on the minimum alveolar concentration of isoflurane in dogs. Veterinary Anaesthesia and Analgesia, 43(2), 136–144. https://doi.org/10.1111/vaa.12282

Simon, B. T., Scallan, E. M., O, O., Ebner, L. S., Cerullo, M. N., Follette, C., … Lizarraga, I. (2017). Pharmacokinetics and pharmacodynamics of midazolam following intravenous and intramuscular administration to sheep. American Journal of Veterinary Research, 78(5), 539–549. https://doi.org/10.2460/ajvr.78.5.539

Simon, B. T., Scallan, E. M., Siracusa, C., Henderson, A., Sleeper, M. M., & Larenza Menzies, M. P. (2014). Effects of acepromazine or methadone on midazolam- induced behavioral reactions in dogs. Canadian Veterinary Journal, 55(9), 875– 885.

Slay, C. E., Enok, S., Hicks, J. W., & Wang, T. (2014). Reduction of blood oxygen levels enhances postprandial cardiac hypertrophy in Burmese python (Python bivittatus). Journal of Experimental Biology, 217(10), 1784–1789. https://doi.org/10.1242/jeb.092841

Sonner, J. M., Antognini, J. F., Dutton, R. C., Flood, P., Gray, A. T., Harris, R. A., … Eger, E. I. 2nd. (2003). Inhaled anesthetics and immobility: mechanisms, mysteries, and minimum alveolar anesthetic concentration. Anesthesia and Analgesia, 97(3), 718–740.

Sonner, J. M., Gong, D., & Eger, E. I. 2nd. (2000). Naturally occurring variability in anesthetic potency among inbred mouse strains. Anesthesia and Analgesia, 91(3), 720–726.

Spelman, L. H., Cambre, R. C., Walsh, T., & Rosscoe, R. (1996). Anesthetic techniques in Komodo dragons (Varanus komodoensis). Proceedings of the American Association of Zoo Veterinarians, 247–250.

Spina, S. P., & Ensom, M. H. H. (2007). Clinical pharmacokinetic monitoring of midazolam in critically ill patients. Pharmacotherapy, 27(3), 389–398. https://doi.org/10.1592/phco.27.3.389

Starck, J. M. (2009). Functional morphology and patterns of blood flow in the heart of Python regius. Journal of Morphology, 270(6), 673–687. https://doi.org/10.1002/jmor.10706

Starck, J. M., Aupperle, H., Kiefer, I., Weimer, I., Krautwald-Junghanns, M.-E., & Pees, M. (2012). Morphological respiratory diffusion capacity of the lungs of ball pythons (Python regius). Zoology, 115(4), 245–254. https://doi.org/10.1016/j.zool.2012.02.003 107

Steffey, E. P., Baggot, J. D., Eisele, J. H., Willits, N., Woliner, M. J., Jarvis, K. A., … Tagawa, M. (1994). Morphine-isoflurane interaction in dogs, swine and rhesus monkeys. Journal of Veterinary Pharmacology and Therapeutics, 17(3), 202–210.

Steffey, E. P., Eisele, J. H., & Baggot, J. D. (2003). Interactions of morphine and isoflurane in horses. American Journal of Veterinary Research, 64(2), 166–175.

Steffey, E. P., Gillespie, J. R., Berry, J. D., Eger, E. I. 2nd, & Munson, E. S. (1974). Anesthetic potency (MAC) of nitrous oxide in the dog, cat, and stump-tail monkey. Journal of Applied Physiology, 36(5), 530–532.

Steffey, E. P., & Howland, D. (1977). Isoflurane potency in the dog and cat. American Journal of Veterinary Research, 38(11), 1833–1836.

Steffey, E. P., & Howland, D. (1978). Potency of halothane-N2O in the horse. American Journal of Veterinary Research, 39(7), 1141–1146.

Steffey, E. P., Howland, D., Giri, S., & Eger, E. I. 2nd. (1977). Enflurane, halothane, and isoflurane potency in horses. American Journal of Veterinary Research, 38(7), 1037–1039.

Steffey, E. P., & Mama, K. R. (2007). Inhalation anesthetics. In W. J. Tranquilli, J. C. Thurmon, & K. A. Grimm (Eds.), Lumb & Jones’ Veterinary Anesthesia and Analgesia (4th ed., pp. 355–394). Oxford: Blackwell Publishing Ltd.

Steffey, E. P., Mama, K. R., & Brosnan, R. J. (2015). Inhalation anesthetics. In K. A. Grimm, L. A. Lamont, W. J. Tranquilli, S. A. Green, & S. A. Robertson (Eds.), Veterinary Anesthesia and Analgesia: The Fifth Edition of Lumb and Jones (5th ed., pp. 297–331). Ames, IA, USA: John Wiley & Sons, Inc.

Steffey, E. P., & Pascoe, P. J. (2002). Detomidine reduces isoflurane anesthetic requirement (MAC) in horses. Veterinary Anaesthesia and Analgesia, 29(4), 223– 227. https://doi.org/10.1046/j.1467-2995.2002.00107.x

Steffey, E. P., Pascoe, P. J., Woliner, M. J., & Berryman, E. R. (2000). Effects of xylazine hydrochloride during isoflurane-induced anesthesia in horses. American Journal of Veterinary Research, 61(10), 1225–1231.

Stegmann, G. F. (1998). Observations on the use of midazolam for sedation, and induction of anaesthesia with midazolam in combination with ketamine in the goat. Journal of the South African Veterinary Association, 69(3), 89–92.

Stegmann, G. F., & Bester, L. (2001). Sedative-hypnotic effects of midazolam in goats after intravenous and intramuscular administration. Veterinary Anaesthesia and Analgesia, 28(1), 49–55. https://doi.org/10.1046/j.1467-2987.2000.00034.x

Stevens, W. C., Cromwell, T. H., Halsey, M. J., Eger, E. I. 2nd, Shakespeare, T. F., & 108

Bahlman, S. H. (1971). The cardiovascular effects of a new inhalation anesthetic, Forane, in human volunteers at constant arterial carbon dioxide tension. Anesthesiology, 35(1), 8–16.

Stevens, W. C., Dolan, W. M., Gibbons, R. T., White, A., Eger, E. I. 2nd, Miller, R. D., … Elashoff, R. M. (1975). Minimum alveolar concentrations (MAC) of isoflurane with and without nitrous oxide in patients of various ages. Anesthesiology, 42, 197–200.

Stoelting, R. K., & Eger, E. I. 2nd. (1969). An additional explanation for the second gas effect: a concentrating effect. Anesthesiology, 30(3), 273–277.

Sun, G.-C., Hsu, M.-C., Chia, Y.-Y., Chen, P.-Y., & Shaw, F.-Z. (2008). Effects of age and gender on intravenous midazolam premedication: a randomized double-blind study. British Journal of Anaesthesia, 101(5), 632–639. https://doi.org/10.1093/bja/aen251

Tae, C. H., Kang, K. J., Min, B.-H., Ahn, J. H., Kim, S., Lee, J. H., … Kim, J. J. (2014). Paradoxical reaction to midazolam in patients undergoing endoscopy under sedation: incidence, risk factors and the effect of flumazenil. Digestive and Liver Disease, 46(8), 710–715. https://doi.org/10.1016/j.dld.2014.04.007

Tearney, C. C., Barter, L. S., & Pypendop, B. H. (2015). Cardiovascular effects of equipotent doses of isoflurane alone and isoflurane plus fentanyl in New Zealand white rabbits (Oryctolagus cuniculus). American Journal of Veterinary Research, 76(7), 591–598. https://doi.org/10.2460/ajvr.76.7.591

Testa, M., Raffe, M. R., & Robinson, E. P. (1990). Evaluation of 25%, 50%, and 67% nitrous oxide with halothane-oxygen for general anesthesia in horses. Veterinary Surgery, 19(4), 308–312.

Tinker, J. H., Sharbrough, F. W., & Michenfelder, J. D. (1977). Anterior shift of the dominant EEG rhytham during anesthesia in the Java monkey: correlation with anesthetic potency. Anesthesiology, 46(4), 252–259.

Tomlin, P. J. (1965). Methoxyflurane. British Journal of Anaesthesia, 37(9), 706–709. https://doi.org/10.1093/bja/37.9.706

Toutain, P. L., & Bousquet-Mélou, A. (2004). Plasma clearance. Journal of Veterinary Pharmacology and Therapeutics, 27(6), 415–425. https://doi.org/10.1111/j.1365- 2885.2004.00605.x

Tranquilli, W. J., Thurmon, J. C., & Benson, G. J. (1985). Anesthetic potency of nitrous oxide in young swine (Sus scrofa). American Journal of Veterinary Research, 46(1), 58–60.

Tsukamoto, A., Iimuro, M., Sato, R., Yamazaki, J., & Inomata, T. (2015). Effect of midazolam and butorphanol premedication on inhalant isoflurane anesthesia in 109

mice. Experimental Animals, 64(2), 139–145. https://doi.org/10.1538/expanim.14- 0073

Tsukamoto, A., Uchida, K., Maesato, S., Sato, R., Kanai, E., & Inomata, T. (2016). Combining isoflurane anesthesia with midazolam and butorphanol in rats. Experimental Animals, 65(3), 223–230. https://doi.org/10.1538/expanim.15-0113

Tsunoda, Y., Hattori, Y., Takatsuka, E., Sawa, T., & Hori, T. (1973). Effects of hydroxyzine, diazepam, and pentazocine on halothane minimum alveolar anesthetic concentration. Anesthesia and Analgesia, 52(3), 390–394.

Turner, P. V., Kerr, C. L., Healy, A. J., & Taylor, W. M. (2006). Effect of meloxicam and butorphanol on minimum alveolar concentration of isoflurane in rabbits. American Journal of Veterinary Research, 67(5), 770–774. https://doi.org/10.2460/ajvr.67.5.770

Upton, R. N., Martinez, A. M., & Grant, C. (2009). Comparison of the sedative properties of CNS 7056, midazolam, and propofol in sheep. British Journal of Anaesthesia, 103(6), 848–857. https://doi.org/10.1093/bja/aep269

Valverde, A., Honeyman, V. L., Dyson, D. H., & Valliant, A. E. (1990). Determination of a sedative dose and influence of midazolam on cardiopulmonary function in Canada geese. American Journal of Veterinary Research, 51(7), 1071–1074.

Valverde, A., Morey, T. E., Hernández, J., & Davies, W. (2003). Validation of several types of noxious stimuli for use in determining the minimum alveolar concentration for inhalation anesthetics in dogs and rabbits. American Journal of Veterinary Research, 64(8), 957–962.

Vesal, N., & Eskandari, M. H. (2006). Sedative effects of midazolam and xylazine with or without ketamine and detomidine alone following intranasal administration in ring-necked parakeets. Journal of the American Veterinary Medical Association, 228(3), 383–388. https://doi.org/10.2460/javma.228.3.383

Villalba, M., Santiago, I., & Gómez de Segura, I. A. (2014). Effects of a constant rate infusion of medetomidine–propofol on isoflurane minimum alveolar concentrations in horses. The Veterinary Journal, 202(2), 329–333. https://doi.org/10.1016/j.tvjl.2014.08.007

Vitcha, J. F. (1971). A history of Forane. Anesthesiology, 35(1), 4–7.

Vitez, T. S., White, P. F., & Eger, E. I. 2nd. (1974). Effects of hypothermia on halothane MAC and isoflurane MAC in the rat. Anesthesiology, 41(1), 80–81.

Votey, S. R., Bosse, G. M., Bayer, M. J., & Hoffman, J. R. (1991). Flumazenil: a new benzodiazepine antagonist. Annals of Emergency Medicine, 20(2), 181–188.

110

Voulgaris, D. A., Egger, C. M., Seddighi, M. R., Rohrbach, B. W., Love, L. C., & Doherty, T. J. (2013). The effect of nitrous oxide on the minimum alveolar concentration (MAC) and MAC derivatives of isoflurane in dogs. Canadian Journal of Veterinary Research, 77(2), 131–135.

Wallach, V. (1998). The lungs of snakes. In C. Gans & A. S. Gaunt (Eds.), Biology of the Reptilia, Vol. 19 (Morphology G: Visceral Organs). (pp. 94–295). Ithaca: Society for the Study of Amphibians and Reptiles.

Wang, T., Altimiras, J., & Axelsson, M. (2002). Intracardiac flow separation in an in situ perfused heart from Burmese python Python molurus. The Journal of Experimental Biology, 205(Pt 17), 2715–2723.

Wang, T., Altimiras, J., Klein, W., & Axelsson, M. (2003). Ventricular haemodynamics in Python molurus: separation of pulmonary and systemic pressures. Journal of Experimental Biology, 206(23), 4241–4245. https://doi.org/10.1242/jeb.00681

Wang, T., Axelsson, M., Jensen, J., & Conlon, J. M. (2000). Cardiovascular actions of python bradykinin and substance P in the anesthetized python, Python regius. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 279(2), R531–R538. https://doi.org/10.1152/ajpregu.2000.279.2.R531

Wang, T., & Hicks, J. W. (1996). The interaction of pulmonary ventilation and the right- left shunt on arterial oxygen levels. Journal of Experimental Biology, 199(Pt 10), 2121–2129.

Wang, T., Smits, A. W., & Burggren, W. W. (1998). Pulmonary functions in reptiles. In C. Gans & A. S. Gaunt (Eds.), Biology of the Reptilia, Vol. 19 (Morphology G: Visceral Organs). (pp. 297–374). Ithaca: Society for the Study of Amphibians and Reptiles.

Waud, D. R. (1972). On biological assays involving quantal responses. Journal of Pharmacology and Experimental Therapeutics, 183(3), 577–607.

Weiskopf, R. B., & Bogetz, M. S. (1984). Minimum alveolar concentrations (MAC) of halothane and nitrous oxide in swine. Anesthesia and Analgesia, 63(5), 529–532.

West, N. H., Butler, P. J., & Bevan, R. M. (1992). Pulmonary blood flow at rest and during swimming in the green turtle, Chelonia mydas. Physiological Zoology, 65(2), 287–310.

White, P. F., Johnston, R. R., & Eger, E. I. 2nd. (1974). Determination of anesthetic requirement in rats. Anesthesiology, 40(1), 52–57.

Williamson, A. J., Soares, J. H. N., Pavlisko, N. D., McAlister Council-Troche, R., & Henao-Guerrero, N. (2017). Isoflurane minimum alveolar concentration sparing effects of fentanyl in the dog. Veterinary Anaesthesia and Analgesia, 44(4), 738– 111

745. https://doi.org/10.1016/j.vaa.2017.02.002

Woodbridge, P. D. (1957). Changing concepts concerning depth of anesthesia. Anesthesiology, 18(4), 536–550.

Wright, A. J. (1999). History of anesthesia: early use of nitrous oxide. Bulletin of Anaesthesia History, 17(3), 10–11.

Yaw, T. J., Mans, C., Doss, G., Johnson, S., & Sladky, K. K. (2017). Comparative efficacy of alfaxalone-midazolam and dexmedetomidine-midazolam sedation in ball pythons (Python regius). Proceedings of the Association of Reptilian and Amphibian Veterinarians, 144. Dallas, TX, United States: Association of Reptilian and Amphibian Veterinarians.

Yaw, T. J., Mans, C., Johnson, S. M., Doss, G. A., & Sladky, K. K. (2018). Effect of injection site on alfaxalone-induced sedation in ball pythons (Python regius). Journal of Small Animal Practice, 59(12), 747–751. https://doi.org/10.1111/jsap.12918

Zhang, Y., Huo, M., Zhou, J., & Xie, S. (2010). PKSolver: An add-in program for pharmacokinetic and pharmacodynamic data analysis in Microsoft Excel. Computer Methods and Programs in Biomedicine, 99(3), 306–314. https://doi.org/10.1016/j.cmpb.2010.01.007

112

APPENDIX A: Minimum anesthetic concentration of isoflurane in various species

Table A.1 – Mean minimum alveolar concentration of isoflurane (MACIso) in mammals.

SPECIES SCIENTIFIC MACISO REFERENCES NAME (%) Cat Felis catus 1.20-2.22 Shaughnessy & Hofmeister (2014) 1.28 Ilkiw, Pascoe, & Fisher (1997) 1.50 Ide, Sakurai, Aono, & Nishino (1998) 1.58 Barletta, Quandt, & Hofmeister (2016) 1.61 Drummond, Todd, & Shapiro (1983) 1.63 Steffey & Howland (1977) 1.90 Barter, Ilkiw, Steffey, Pypendop, & Imai (2004) 2.21 Pypendop & Ilkiw (2005) Cow Bos taurus 1.14 Cantalapiedra, Villanueva, & Pereira (2000) Dog Canis lupus 1.18 Floriano et al. (2016) familiaris 1.18 Gianotti et al. (2014) 1.20 Monteiro, Coelho, Bressan, Simões, & Monteiro (2016) 1.23, 1.27 Barletta et al. (2016) 1.28 Steffey & Howland (1977) 1.30 Steffey et al. (1994) 1.30 Williamson, Soares, Pavlisko, McAlister Council-Troche, & Henao-Guerrero (2017) 1.31 Schwieger, Szlam, & Hug (1989) 1.38-1.39 Simões, Monteiro, Rangel, Nunes-Junior, & Campagnol (2016) 1.39 Steffey & Howland (1977) 1.39 Imai, Ilkiw, Pypendop, Farver, & Steffey (2002) 1.39-1.50 A. E. Schwartz, Maneksha, Kanchuger, Sidhu, & Poppers (1989) 1.44 Acevedo-Arcique et al. (2014) 1.69 Figueiró, Soares, Ascoli, Werre, & Gómez de Segura (2016) Ferret Mustela 1.52 Murat & Housmans (1988) putorius furo 1.74 Imai, Steffey, Farver, & Ilkiw (1999) Gerbil Meriones 1.55 de Segura, de la Víbora, & Criado (2009) unguiculatus

113

Table A.1 – (Continued)

SPECIES SCIENTIFIC MACISO REFERENCES NAME (%) Goat Capra 1.2 Antognini & Schwartz (1993) aegagrus 1.23 Doherty, Rohrbach, & Geiser (2002) hircus 1.29 Hikasa, Okuyama, Kakuta, Takase, & Ogasawara (1998) 1.31 Doherty, Rohrbach, Ross, & Schultz (2002) 1.43 Doherty, Will, Rohrbach, & Geiser (2004) 1.5 Antognini & Eisele (1993) Horse Equus ferus 1.23 Villalba, Santiago, & Gómez de Segura (2014) caballus 1.31 Steffey, Howland, Giri, & Eger (1977) 1.43 Steffey, Eisele, & Baggot (2003) 1.44 Steffey & Pascoe (2002) 1.64 Steffey, Pascoe, Woliner, & Berryman (2000) Java monkey Macaca 1.28 Tinker, Sharbrough, & Michenfelder (1977) fascicularis Mouse Mus 1.31-1.77 Sonner, Gong, & Eger (2000) musculus 1.35 Mazze, Rice, & Baden (1985) 1.38 Tsukamoto, Iimuro, Sato, Yamazaki, & Inomata (2015) 1.41 Deady et al. (1981) Pig Sus scrofa 1.45 Lundeen, Manohar, & Parks (1983) 1.48 Lerman et al. (1990) 1.51 Steffey et al. (1994) 1.55 Eisele et al. (1985) 1.75 Tranquilli et al. (1985) 2.04 Eger et al. (1988) Rabbit Oryctolagus 1.92 Barter, Hawkins, & Pypendop (2015) cuniculus 1.95 Bailey, Barter, & Pypendop (2017) 1.95 Tearney, Barter, & Pypendop (2015) 2.05 Drummond (1985) 2.07 Imai, Steffey, Ilkiw, & Farver (1999) 2.09 Schnellbacher et al. (2013) 2.12 Scheller, Zornow, Fleischer, Shearman, & Greber (1989) 2.49 Turner, Kerr, Healy, & Taylor (2006)

114

Table A.1 – (Continued)

SPECIES SCIENTIFIC NAME MACISO REFERENCES (%) Rat Rattus norvegicus 1.17 Vitez et al. (1974) 1.28 Russell & Graybeal (1995) 1.30 Rampil & Laster (1992) 1.30 Tsukamoto et al. (2016) 1.32 Chavez et al. (2015) 1.38 White, Johnston, & Eger (1974) 1.46 Mazze et al. (1985) 1.46 Laster, Liu, Eger, & Taheri (1993) 1.58 Imai, Steffey, Farver, et al. (1999) 1.58 Cole et al. (1990) Rhesus monkey Macaca mulatta 1.46 Steffey et al. (1994) Ring-tailed lemur Lemur catta 1.9 Chinnadurai, Balko, & Williams (2017) Sheep Ovis aries 1.58 Palahniuk, Shnider, & Eger (1974)

115

Table A.2 – Mean minimum anesthetic concentration of isoflurane (MACIso) in birds.

SPECIES SCIENTIFIC NAME MACISO REFERENCES (%) Chicken Gallus gallus 1.1 da Rocha et al. (2017) domesticus 1.1 Escobar et al. (2016) 1.15 Martin-Jurado, Vogt, Kutter, Bettschart- Wolfensberger, & Hatt (2008) Cinereous vulture Aegypius monachus 1.06 Kim et al. (2011) Cockatoo Cacatua spp. 1.44 Curro, Brunson, & Paul-Murphy (1995) Crested serpent Spilornis cheela 1.46 Chan, Chang, Wang, & Hsu (2013) eagle hoya Duck Anas platyrhynchos 1.30 Ludders et al. (1990) domesticus Pigeon Columba livia 1.51 Fitzgerald & Blais (1991) 1.8 Botman, Dugdale, Gabriel, & Vandeweerd (2016) Red-tailed hawk Buteo jamaicensis 1.45 Fitzgerald & Blais (1991) 2.05 Pavez, Hawkins, Pascoe, Knych, & Kass (2011) Sandhill crane Grus canadensis 1.34 Ludders et al. (1989) Thick-billed parrot Rhynchopsitta 1.07 Mercado, Larsen, Wack, & Pypendop pachyrhyncha (2008) White-eyed Psittacara 2.45 Escobar et al. (2017) parakeet leucophthalmus

116

APPENDIX B: Descriptive data of the animal research collection

Table B.1 – Weights of ball pythons used in the current project over the year 2018.

WEIGHT (kg) SNAKES JANUARY FEBRUARY MARCH APRIL MAY JUNE JULY AUGUST #1 3.43 - 3.26 3.11 3.05 2.99 2.96 2.88 #2 2.76 - 2.71 2.53 2.46 2.43 2.35 2.49 #3 3.01 - 2.95 2.76 - 2.62 2.54 2.61 #4 4.17 - 3.94 3.81 3.61 3.58 3.37 3.37 #5 3.92 - 3.81 3.53 3.37 3.38 3.23 3.29 #6 1.74 - 1.60 1.53 1.48 1.50 1.52 1.48 #7 3.35 - 3.28 3.19 3.05 2.96 2.95 2.97 #8 2.35 - 2.33 2.20 2.13 2.09 2.07 2.11 #9 3.79 - 3.58 3.44 3.45 3.26 3.24 2.94 #10 3.59 - 3.06 2.88 - 2.68 2.90 2.71 #11 3.23 - 3.05 2.83 - 2.76 2.61 2.68 #12 4.40 - 3.94 3.80 - 3.65 - 3.91 #13 2.40 - 2.27 2.16 - - 2.03 2.02 #57 - - - - 2.48 2.51 - 2.35 #58 - - - - 2.15 2.15 - 2.15 #59 - - - - 2.23 2.21 2.06 2.10

117

Table B.2 – Usage of ball pythons for the different trials of the current project.

SNAKES PD PD MAC MAC PK PK MIDAZOLAM FLUMAZENIL PILOT PILOT #1 X X - X - X #2 X X - X - X #3 X X - X - X #4 X X - X - X #5 X X - X - X #6 - - X X - X #7 X X - X - X #8 X X - X - X #9 X X - X - X #10 X X - - - X #11 - - X - - X #12 - - X - X - #13 X - - - - - #57 - - - - X - #58 - - - - X - #59 - - - - - X PD: pharmacodynamics; MAC: minimum anesthetic concentration; PK: pharmacokinetics; X: used for this trial; -: not used for this trial.

118

APPENDIX C: Descriptive data from chapter 2 Table C.1 – Heart rates (beats/min) of ball pythons receiving a low (L, 1 mg/kg) or a high (H, 2 mg/kg) dosage of midazolam intramuscularly over time.

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 SNAKE #13 (min) L H L H L H L H L H L H L H L H L H L H 0 32 24 40 56 44 56 40 40 44 40 36 44 42 44 48 30 44 28 40 40 6 40 36 36 52 40 56 48 40 40 40 44 36 42 40 42 42 48 40 44 40 10 36 36 36 48 44 44 44 42 40 44 32 40 40 48 32 52 44 44 40 36 15 36 32 28 44 40 40 44 44 40 40 40 32 36 40 32 32 36 36 24 32 30 24 28 28 36 32 32 36 42 36 36 24 28 40 44 28 20 24 24 40 28 45 24 28 32 32 20 32 32 36 26 36 20 24 40 44 24 32 24 16 16 24 60 20 20 24 28 24 36 40 32 40 32 20 20 52 48 20 28 20 20 24 28 90 20 20 28 28 24 28 36 28 28 32 16 40 44 52 16 20 24 20 24 32 120 16 20 24 56 16 32 28 42 40 32 24 16 48 52 20 16 20 16 44 44 180 16 20 20 36 16 48 24 42 24 40 16 24 52 60 16 16 28 24 20 40 240 16 24 24 20 20 28 24 28 24 36 16 16 48 56 14 20 24 20 40 42 300 24 20 20 56 16 28 24 32 24 24 16 20 56 40 20 12 20 24 28 42 360 28 16 32 24 32 18 36 24 24 20 12 24 56 36 16 12 20 16 20 32 420 20 20 32 20 16 24 28 20 24 28 16 24 42 48 24 12 20 24 24 28 480 24 20 24 24 20 24 28 24 24 22 16 20 44 44 16 20 16 16 32 28 1440 16 16 28 16 12 16 16 16 16 20 16 16 36 36 16 16 36 20 12 16 1920 28 24 24 28 20 24 32 28 20 28 20 16 44 42 20 20 12 8 24 20 2880 20 24 20 16 20 16 48 20 16 28 28 12 28 32 20 12 16 12 16 16 3360 28 24 24 36 24 20 36 32 24 40 20 20 28 32 16 24 20 16 24 28 4320 28 20 24 20 16 24 24 32 20 24 24 16 32 32 20 16 8 12 12 20 4800 28 20 20 28 24 24 36 28 18 52 24 40 28 28 20 20 20 12 20 28 5760 20 20 20 32 - 20 - 28 16 24 - - - - - 16 16 16 - 24 6240 - 20 24 - - - - - 16 36 - - - - - 12 - 12 - - 7200 ------24 28 - - - - - 16 - 20 - - 7680 ------32 32 ------20 - - 8640 ------28 ------16 - - 9120 ------16 - - 10080 ------20 - - 119

Table C.2 – Respiratory rates (breaths/min) of ball pythons receiving a low (L, 1 mg/kg) or a high (H, 2 mg/kg) dosage of midazolam intramuscularly over time.

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 SNAKE #13 (min) L H L H L H L H L H L H L H L H L H L H 0 12 12 32 32 16 24 12 16 24 32 24 16 12 16 12 8 36 32 12 32 6 13 12 16 12 20 8 16 20 8 12 12 8 12 12 8 12 12 28 12 12 10 16 8 12 8 12 8 20 12 12 12 12 8 18 20 12 12 8 24 8 8 15 10 8 16 12 12 10 16 16 8 12 12 8 16 12 8 8 12 8 8 12 30 12 20 8 12 12 10 16 16 20 8 12 16 12 16 8 8 8 12 8 16 45 8 12 12 8 8 8 24 16 20 16 8 16 12 8 8 8 12 4 8 8 60 12 16 20 8 8 8 16 12 16 8 12 12 16 16 8 4 20 8 12 16 90 20 20 28 28 28 16 20 8 12 12 16 24 20 12 8 4 12 8 12 12 120 20 16 32 12 12 16 28 8 12 16 12 12 12 12 12 4 16 8 12 12 180 28 20 24 28 8 12 36 24 8 24 8 12 12 8 12 4 8 8 8 4 240 24 8 28 24 8 16 16 12 12 16 8 24 16 8 12 4 12 6 4 12 300 12 16 20 32 8 16 20 12 8 12 16 8 16 16 8 4 8 8 8 4 360 20 8 16 32 8 16 20 28 12 8 8 4 16 8 8 8 8 12 4 12 420 20 8 24 24 12 24 16 12 20 12 8 8 20 16 8 4 8 4 4 8 480 20 8 12 28 8 16 16 8 20 8 8 8 12 12 8 4 8 4 16 8 1440 24 16 24 20 12 12 32 16 36 16 8 8 24 24 16 4 12 8 4 8 1920 36 20 32 8 12 24 24 16 16 12 12 26 16 16 16 4 20 4 4 8 2880 32 28 28 32 12 8 20 16 16 12 8 12 20 12 8 4 40 8 4 16 3360 40 24 16 32 12 8 12 16 40 12 8 12 12 16 6 8 12 8 4 8 4320 16 16 28 24 16 12 32 28 20 12 12 8 24 24 8 4 4 4 8 8 4800 20 12 28 12 16 12 32 32 28 12 12 12 20 16 12 8 8 8 4 12 5760 32 16 28 32 - 8 - 24 12 8 - - - - - 8 8 8 - 8 6240 - 12 28 - - - - - 20 8 - - - - - 4 - 8 - - 7200 ------8 12 - - - - - 8 - 12 - - 7680 ------16 16 ------8 - - 8640 ------16 ------4 - - 9120 ------4 - - 10080 ------8 - -

120

Table C.3 – Superficial body temperatures (°C) of ball pythons receiving a low (L, 1 mg/kg) or a high (H, 2 mg/kg) dosage of midazolam intramuscularly over time.

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 SNAKE #13 (min) L H L H L H L H L H L H L H L H L H L H 0 28.6 28.1 28.6 30.3 27.1 30.5 26.6 27.2 26.3 27.0 26.7 26.4 26.1 26.7 28.3 28.8 29.7 28.0 27.2 28.2 6 28.6 28.4 27.8 29.9 27.6 30.2 27.6 28.0 26.5 27.0 26.9 27.1 26.3 26.5 28.1 29.6 29.0 28.8 27.8 28.4 10 28.8 28.3 27.5 30.1 27.4 30.0 27.5 28.2 26.0 27.3 27.6 26.9 25.8 27.9 28.3 29.4 28.8 28.6 27.8 28.2 15 28.5 28.1 27.7 29.8 27.9 29.9 27.8 28.3 26.1 27.4 26.8 27.2 25.9 27.8 28.2 29.9 28.9 28.1 27.8 28.4 30 28.5 28.6 27.8 30.0 27.4 29.7 27.3 28.9 26.8 27.2 27.0 27.3 26.2 27.8 28.2 28.7 28.5 28.1 27.5 29.0 45 28.8 28.5 28.1 30.0 27.1 29.6 27.9 28.6 26.9 27.7 26.9 27.1 27.0 28.0 28.1 28.6 28.5 27.8 27.6 28.9 60 28.5 28.4 28.2 29.9 27.8 29.5 27.8 28.5 27.0 27.5 26.5 27.8 27.3 28.6 27.9 28.3 28.3 27.6 27.8 28.7 90 27.9 28.5 28.6 30.5 27.7 29.5 28.0 28.6 27.3 27.7 26.8 27.7 28.0 28.8 27.8 28.3 28.2 27.4 27.6 28.8 120 28.2 28.8 28.8 30.5 27.5 29.9 28.1 29.0 27.8 28.0 26.9 27.7 28.1 29.0 27.9 27.9 28.1 27.1 27.6 29.5 180 28.3 29.7 29.3 30.5 28.1 29.8 28.5 28.9 28.1 28.7 27.3 27.5 28.4 29.6 28.1 27.5 28.0 27.5 28.2 29.3 240 28.0 29.6 29.6 30.5 28.3 29.5 28.7 29.4 28.2 29.1 27.0 27.9 29.1 30.1 28.1 29.0 28.3 27.2 28.4 29.1 300 28.8 29.0 30.0 30.7 28.5 29.5 28.9 28.8 28.5 29.0 27.4 28.4 30.0 29.9 28.0 27.5 28.8 27.3 28.6 29.1 360 28.5 29.5 30.2 30.5 28.8 29.4 29.0 28.8 28.3 28.5 27.0 28.2 29.7 30.5 27.8 27.7 28.1 27.3 28.5 29.3 420 29.2 29.2 30.1 30.5 28.9 29.3 29.1 28.7 28.5 28.2 27.5 28.5 29.9 30.1 27.8 27.7 27.7 27.6 28.6 29.4 480 29.0 29.4 29.8 30.5 29.5 29.5 29.2 28.5 28.5 28.3 27.5 28.3 29.3 30.2 27.8 27.6 28.3 27.3 28.9 29.3 1440 27.4 27.4 26.9 27.5 26.7 26.7 27.2 27.3 26.5 26.8 26.8 27.0 28.5 28.3 27.1 27.3 27.6 27.0 26.7 26.8 1920 31.8 30.8 30.8 30.3 29.6 29.0 28.5 27.7 27.3 30.0 27.5 28.0 28.7 28.5 28.9 28.8 26.5 27.8 31.1 29.0 2880 27.8 28.0 27.1 27.8 28.0 27.0 26.7 27.5 26.3 27.1 26.4 26.7 26.3 26.4 26.8 26.5 26.9 26.4 27.0 26.6 3360 30.1 29.0 30.9 31.2 31.0 29.5 29.9 30.1 28.8 29.7 28.5 29.6 27.3 28.2 26.9 28.3 29.5 29.0 30.8 29.9 4320 28.4 28.5 27.9 28.3 27.4 29.3 27.6 27.2 26.7 26.8 26.3 26.9 26.6 26.9 26.9 26.6 28.0 27.3 27.0 27.4 4800 32.0 28.5 30.5 31.5 30.3 30.3 29.8 29.5 28.7 30.0 28.5 28.6 31.1 29.1 29.0 27.4 27.9 27.4 30.7 31.3 5760 29.5 28.2 27.6 29.3 - 28.5 - 27.1 26.4 27.5 - - - - - 26.6 26.5 27.1 - 26.8 6240 - 28.8 30.4 - - - - - 28.7 30.6 - - - - - 27.6 - 28.3 - - 7200 ------26.5 27.5 - - - - - 26.7 - 27.0 - - 7680 ------27.7 30.6 ------28.5 - - 8640 ------25.8 ------28.3 - - 9120 ------29.2 - - 10080 ------27.3 - -

121

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 SNAKE #13 (min) L H L H L H L H L H L H L H L H L H L H 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 3 0 0 0 0 0 1 0 0 0 0 0 2 1 0 1 0 0 0 1 0 6 2 2 3 1 0 2 0 1 0 1 0 5 1 0 1 2 0 2 5 1 10 2 4 2 2 0 5 2 2 3 4 4 4 2 2 5 4 2 5 2 3 15 1 4 4 4 1 5 1 3 2 5 5 6 1 2 5 4 2 6 5 5 30 5 2 4 5 2 6 1 2 2 5 2 5 3 2 5 4 3 6 4 2 45 6 2 2 6 2 6 2 4 5 4 5 6 1 1 5 2 5 6 5 2 60 5 3 4 6 2 6 2 4 2 5 6 6 4 4 5 3 5 6 5 2 90 5 4 4 6 2 3 1 5 2 5 5 6 2 3 6 4 2 6 4 3 120 2 1 5 5 2 2 2 2 1 3 4 5 2 2 5 4 4 5 5 2 180 6 4 4 5 2 5 1 2 6 4 3 4 1 2 5 2 4 5 3 2 240 3 2 5 4 2 2 1 1 5 3 2 4 1 2 6 5 5 6 5 2 300 4 4 4 4 2 1 2 2 5 5 4 2 2 2 4 2 4 6 5 2 360 3 6 4 4 1 1 1 1 4 4 1 4 2 3 4 4 4 6 5 0 420 4 3 4 4 2 1 2 2 5 5 5 3 2 2 4 4 4 6 3 2 480 2 2 4 4 2 1 1 2 2 5 2 4 2 2 5 1 1 5 4 4 1440 1 2 5 2 1 1 3 2 5 5 4 1 1 2 3 3 5 6 5 4 1920 1 5 2 3 2 1 3 2 5 5 2 1 1 1 5 4 5 6 5 3 2880 0 2 4 2 4 1 2 1 5 4 3 2 2 1 4 4 4 5 5 3 3360 1 2 4 4 1 2 1 2 5 4 1 1 1 1 3 4 5 5 5 4 4320 1 2 5 3 0 1 0 2 5 5 3 1 0 0 1 5 5 5 2 4 4800 1 2 4 1 0 1 0 1 5 5 1 0 0 0 0 5 4 4 2 2 5760 1 1 3 1 - 1 - 1 4 3 - - - - - 4 1 4 - 4 6240 - 1 2 - - - - - 5 2 - - - - - 3 - 4 - - 7200 ------5 2 - - - - - 1 - 2 - - 7680 ------4 2 ------4 - - 8640 ------1 ------3 - - 9120 ------3 - - 10080 ------1 - -

122

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 SNAKE #13 (min) L H L H L H L H L H L H L H L H L H L H 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 3 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 6 1 0 1 1 0 1 0 1 1 1 0 1 1 1 1 1 0 1 1 0 10 1 1 1 1 0 2 1 0 1 1 1 2 1 1 1 1 1 1 1 1 15 1 1 0 1 1 2 1 1 1 1 1 2 1 1 1 1 1 1 1 1 30 1 1 1 2 1 2 1 1 1 1 1 2 1 1 1 1 1 1 1 1 45 2 1 1 2 1 2 1 1 2 1 2 2 1 1 1 2 1 1 1 2 60 2 1 1 2 1 2 1 1 1 1 2 2 1 1 2 1 2 1 1 2 90 2 1 1 2 1 2 1 1 1 1 2 2 1 1 1 1 1 1 1 1 120 1 1 1 1 2 1 1 1 1 2 2 2 1 1 1 1 1 1 1 1 180 1 1 1 1 1 2 1 1 1 1 2 2 1 1 1 1 1 1 2 1 240 1 1 1 1 1 2 1 1 1 2 2 1 1 1 1 1 1 1 1 1 300 1 1 1 1 1 2 1 1 1 2 2 1 1 1 2 1 1 1 1 1 360 1 1 1 1 1 1 1 1 1 2 2 1 1 2 2 1 1 1 1 1 420 1 1 1 1 2 1 1 1 2 2 2 2 1 1 2 2 1 1 1 1 480 1 1 1 1 1 1 1 1 2 2 2 2 1 1 2 1 1 1 1 1 1440 0 1 1 1 2 1 1 1 1 2 2 1 1 1 1 2 1 2 1 1 1920 1 1 1 1 1 1 0 1 1 2 2 1 1 1 1 2 1 1 1 1 2880 0 1 1 1 1 1 1 1 1 2 2 2 0 1 1 1 1 1 1 1 3360 0 1 1 1 2 2 1 1 1 2 2 2 0 1 1 2 1 1 1 1 4320 1 1 1 1 1 1 1 1 1 2 1 1 0 1 1 2 1 1 1 1 4800 1 1 0 1 1 1 0 1 1 2 1 1 0 0 1 1 1 1 0 1 5760 0 1 0 0 - 2 - 0 1 2 - - - - - 1 1 1 - 0 6240 - 1 0 - - - - - 1 2 - - - - - 1 - 1 - - 7200 ------1 2 - - - - - 1 - 1 - - 7680 ------1 1 ------1 - - 8640 ------0 ------1 - - 9120 ------1 - - 10080 ------1 - -

123

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 (min) T P R T P R T P R T P R T P R T P R T P R T P R T P R 0 27.5 32 12 28.2 52 28 28.0 36 32 28.3 44 28 27.7 48 28 26.2 32 24 28.3 48 32 28.2 32 16 28.3 32 24 6 27.5 40 12 28.8 32 12 28.5 44 24 28.5 24 16 27.9 40 12 26.6 28 12 28.8 48 28 28.8 28 16 28.5 40 20 10 27.5 32 12 28.9 32 8 28.1 36 24 28.6 28 16 27.9 40 8 26.9 24 16 29.1 48 16 28.9 40 12 28.6 40 12 15 27.6 36 12 28.6 32 20 28.1 40 12 29.3 28 12 27.9 36 8 26.7 20 8 29.2 52 16 28.7 28 12 28.5 32 8 30 27.4 20 8 29.0 24 4 28.4 20 12 29.3 48 16 27.7 32 12 27.0 20 8 29.3 40 16 28.2 20 8 28.4 28 12 45 27.2 16 8 29.1 20 4 28.5 20 16 29.4 36 12 27.4 28 12 27.0 16 8 29.4 48 20 28.2 16 12 28.1 24 4 60 27.4 18 12 29.1 20 8 28.6 24 16 29.3 40 8 27.7 32 16 27.0 16 4 30.0 52 16 28.2 20 12 27.9 20 4 65 27.6 32 12 29.6 32 8 28.7 28 12 29.3 48 20 27.8 36 12 27.3 28 8 29.8 56 16 28.5 32 8 28.1 32 12 70 27.5 28 20 29.1 32 8 29.1 44 16 29.5 48 16 27.8 40 16 27.3 36 16 30.1 60 16 28.6 48 12 28.1 36 16 75 28.0 32 20 29.1 32 16 29.1 48 16 29.8 60 8 28.1 44 20 27.4 32 12 30.1 56 12 28.7 48 16 28.3 36 16 90 27.5 28 16 29.5 48 16 29.4 44 20 29.7 52 16 28.0 42 16 27.3 28 16 30.0 56 12 28.6 44 16 28.3 28 24 105 28.4 32 16 29.6 32 16 29.3 32 12 29.9 48 12 28.1 44 16 27.4 40 12 30.1 60 12 28.7 36 16 28.3 40 16 120 28.0 20 16 29.3 44 12 29.0 28 12 29.7 44 24 28.4 36 12 27.4 32 12 30.2 44 16 28.7 44 32 28.5 36 20 180 29.4 20 12 30.2 28 28 29.8 28 24 29.8 40 16 28.8 32 16 28.2 24 16 30.0 52 16 29.0 32 20 28.6 36 12 240 29.1 24 20 29.4 54 28 29.1 20 16 29.8 32 24 29.1 28 12 27.9 20 20 29.8 52 20 29.3 28 12 29.1 24 8 300 29.1 20 16 29.8 32 28 29.4 24 8 30.0 32 16 29.3 28 12 27.8 20 20 29.8 52 20 29.5 24 16 29.4 20 12 360 29.3 20 24 29.2 24 24 28.9 20 24 29.5 32 12 30.0 32 12 27.7 20 16 29.9 48 20 30.0 20 8 28.5 20 8 420 29.5 20 16 29.1 40 16 29.8 20 20 29.9 44 16 29.8 36 12 27.6 28 20 29.3 42 20 29.3 20 12 28.1 16 8 480 29.6 20 24 29.6 36 16 29.3 16 16 29.4 40 16 29.1 28 32 27.7 12 28 29.8 52 24 28.8 24 8 26.8 16 12 1440 27.9 24 20 27.1 24 16 27.3 12 16 27.4 28 8 26.2 16 12 26.7 16 20 27.2 42 16 27.2 16 12 30.6 28 8 1920 28.5 20 20 29.8 24 16 29.2 12 12 28.6 28 24 30.2 28 8 28.0 12 8 29.6 48 20 29.2 28 8 27.7 12 4 2880 27.5 16 36 27.9 28 28 26.9 20 16 28.3 36 16 26.2 36 24 26.1 20 20 26.3 42 24 29.2 24 8 30.4 20 8 3360 29.6 24 28 30.5 24 12 28.9 16 16 28.8 28 12 29.8 40 12 28.6 20 16 27.1 44 12 28.9 20 12 30.7 16 12 4320 28.3 28 12 27.5 24 24 26.6 16 16 27.2 24 20 25.5 24 12 26.4 24 8 26.1 40 24 28.5 20 12 26.9 8 12 4800 28.3 20 16 31.9 32 28 28.0 16 16 28.4 36 20 29.5 24 36 28.5 16 12 27.2 44 24 29.6 20 8 26.8 16 8 5760 27.4 20 12 28.6 20 32 26.7 12 28 27.5 40 20 25.4 24 16 26.8 24 12 ------27.3 24 12 6240 - - - 30.2 24 36 29.7 36 20 - - - 27.5 28 28 28.7 20 16 ------30.4 32 16 7200 ------26.1 32 24 27.0 16 24 ------26.6 28 12 7680 ------29.7 36 12 28.8 28 20 ------29.5 32 16 8640 ------26.3 40 24 27.0 24 20 ------29.1 24 12 9120 ------29.9 36 24 27.0 28 24 ------30.5 28 12 10080 ------27.8 28 12 10560 ------27.8 36 12 11520 ------28.3 28 16

124

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 (min) S M S M S M S M S M S M S M S M S M 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 6 0 0 3 0 0 1 0 0 2 1 1 1 2 0 3 1 1 1 10 2 1 4 1 2 1 1 0 5 1 3 1 2 0 4 1 2 1 15 2 1 5 1 4 1 1 1 6 2 5 1 2 1 4 1 4 1 30 2 1 5 1 2 2 1 1 5 2 5 1 2 1 5 1 5 1 45 3 1 5 1 2 2 1 1 5 2 5 2 2 1 5 2 6 2 60 2 1 6 1 2 2 2 1 5 2 5 2 2 1 5 1 5 1 65 2 1 5 1 1 1 1 1 2 2 3 1 2 1 1 1 2 1 70 1 1 5 1 0 0 0 0 0 1 0 1 0 0 0 0 0 0 75 0 0 1 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 90 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 105 0 0 0 0 1 0 0 0 0 0 1 1 0 0 0 0 2 0 120 0 0 0 0 0 0 0 0 3 0 1 1 0 0 0 0 0 1 180 1 0 2 0 0 0 0 0 3 1 1 1 1 1 0 0 0 0 240 0 0 2 0 1 1 0 0 5 1 4 1 1 0 3 1 5 1 300 1 1 4 0 1 0 0 0 5 1 3 1 2 0 2 1 5 1 360 1 0 5 0 1 1 1 0 5 2 4 1 1 0 4 1 6 1 420 2 0 5 1 1 1 1 0 5 1 5 1 1 0 3 1 4 1 480 1 1 3 0 1 1 0 0 4 2 3 1 1 0 4 1 4 1 1440 1 0 3 0 1 1 3 0 5 2 3 2 0 0 3 1 5 1 1920 1 1 3 1 1 1 1 0 5 2 4 2 1 1 4 1 4 1 2880 3 1 5 0 2 1 1 0 5 2 4 2 1 0 1 1 4 1 3360 1 1 5 1 1 1 5 0 5 2 3 1 0 0 1 1 4 1 4320 0 0 5 1 1 1 1 0 5 2 4 1 0 0 1 1 5 1 4800 1 1 3 1 1 1 5 0 5 2 4 1 0 0 0 1 4 1 5760 0 0 2 1 1 1 0 0 5 1 1 1 - - - 2 5 1 6240 - - 1 0 0 1 - - 4 1 3 1 - - - - 5 1 7200 ------5 1 1 1 - - - - 3 1 7680 ------3 1 3 1 - - - - 3 1 8640 ------2 1 0 1 - - - - 5 1 9120 ------1 1 0 1 - - - - 4 1 10080 ------4 1 10560 ------3 1 11520 ------1 0 125

APPENDIX D: Descriptive data from chapter 3

126

Table D.1 – End-tidal concentration of isoflurane in ball pythons immediately prior to each supramaximal stimulation (electrical stimulus at the base of the tail) over time when anesthetized with isoflurane in 100% oxygen with 0.2 mL/kg saline intramuscularly (IM) (“Iso”), isoflurane in 100% oxygen with 1 mg/kg IM midazolam (“Mid”), or isoflurane in 50% nitrous oxide and 50% oxygen with 0.2 mL/kg saline IM (“N2O”). The presence (+) or absence (-) of a purposeful movement in response to the stimulation is reported. Snake Tx End-tidal concentration of isoflurane (%) 1st 2nd 3rd 4th 5th 6th 7th 8th 9th 10th Iso 1.30(-) 1.18(-) 1.07(-) 0.97(-) 0.89(+) - - - - - #1 Mid 0.94(-) 0.86(-) 0.77(-) 0.68(-) 0.60(-) 0.51(-) 0.45(-) 0.38(-) 0.29(-) 0.26(+) N2O 1.06(-) 0.92(+) ------Iso 1.36(-) 1.20(-) 1.11(+) ------#2 Mid 1.04(-) 0.90(-) 0.82(-) 0.74(-) 0.65(-) 0.53(+) - - - - N2O 1.13(-) 1.00(-) 0.84(-) 0.70(-) 0.58(+) - - - - - Iso 1.77(-) 1.59(-) 1.44(-) 1.29(+) ------#3 Mid 1.28(-) 1.14(-) 1.05(-) 0.92(+) ------N2O 1.11(+) 1.18(+) 1.26(+) 1.37(-) 1.25(+) - - - - - Iso 1.55(-) 1.47(-) 1.30(-) 1.17(-) 1.09(-) 0.96(-) 0.84(+) - - - #4 Mid 0.95(-) 0.89(-) 0.80(-) 0.71(-) 0.63(-) 0.55(-) 0.45(-) 0.40(+) - - N2O 1.44(-) 1.28(-) 1.16(-) 1.05(-) 0.93(-) 0.83(-) 0.77(-) 0.67(-) 0.57(+) - Iso 1.32(-) 1.18(-) 1.08(+) ------#5 Mid 0.96(-) 0.84(-) 0.76(-) 0.70(-) 0.59(-) 0.50(-) 0.41(-) 0.31(+) - - N2O 1.18(-) 1.06(-) 0.98(+) ------Iso 1.36(+) 1.44(-) 1.35(-) 1.20(-) 1.06(+) - - - - - #6 Mid 1.00(-) 0.91(-) 0.80(-) 0.73(-) 0.63(+) - - - - - N2O 1.24(-) 1.08(-) 0.98(-) 0.84(+) ------Iso 1.41(+) 1.54(-) 1.38(+) ------#7 Mid 1.40(-) 1.22(-) 1.12(-) 1.01(-) 0.89(-) 0.78(-) 0.63(+) - - - N2O 1.42(-) 1.27(+) ------Iso 1.20(+) 1.30(+) 1.39(+) 1.55(-) 1.41(-) 1.22(-) 1.08(+) - - - #8 Mid 1.03(-) 0.85(-) 0.75(-) 0.70(-) 0.62(+) - - - - - N2O 1.44(-) 1.27(-) 1.16(-) 1.01(+) ------Iso 1.37(-) 1.24(-) 1.17(-) 1.00(-) 0.87(-) 0.73(-) 0.63(+) - - - #9 Mid 1.21(-) 1.11(-) 1.03(-) 0.92(-) 0.80(-) 0.71(-) 0.63(-) 0.44(-) 0.42(+) - N2O 0.91(-) 0.80(-) 0.69(+) ------

127

† Snake Tx TInt Equilibration times (min) TMAC (min) 1st 2nd 3rd 4th 5th 6th 7th 8th 9th 10th (min) Iso 24 172 65 63 64 56 - - - - - 420 #1 Mid 21 120 43 50 45 51 45 50 60 47 54 565 N2O 29 165 64 ------229 Iso 14 140 79 76 ------295 #2 Mid 25 155 63 59 53 49 54 - - - - 433 N2O 17 130 72 58 65 90 - - - - - 415 Iso 13 120 66 60 51 ------297 #3 Mid 20 110 60 50 67 ------287 N2O 15 120 63 67 58 66 - - - - - 374 Iso 17 140 50 62 68 65 60 75 - - - 520 #4 Mid 15 122 63 61 66 61 59 60 62 - - 554 N2O 15 119 58 57 53 49 46 41 43 46 - 512 Iso 16 136 62 68 ------266 #5 Mid 18 128 72 56 46 65 41 47 48 - - 503 N2O 21 141 83 53 ------277 Iso 18 139 85 54 53 91 - - - - - 422 #6 Mid 14 170 52 64 44 47 - - - - - 377 N2O 18 145 57 59 63 ------324 Iso 18 132 104 50 ------286 #7 Mid 25 149 88 72 61 59 60 53 - - - 542 N2O 15 120 63 ------183 Iso 23 113 45 51 64 65 46 61 - - - 445 #8 Mid 20 141 55 63 55 60 - - - - - 374 N2O 23 121 63 60 53 ------297 Iso 18 124 54 58 63 57 65 50 - - - 471 #9 Mid 18 151 60 60 55 68 64 44 77 18 - 597 N2O 14 136 60 65 ------261 Tx: Treatment; Iso: saline 0.2 mL/kg intramuscular (IM) with isoflurane in 100% oxygen; Mid: midazolam 1 mg/kg IM with isoflurane in 100% oxygen; N2O: saline 0.2 mL/kg IM with isoflurane delivered in 50% nitrous oxide and 50% oxygen. †The equilibration time was defined as the time from the moment the isoflurane concentration was changed until the moment the end-tidal concentration of isoflurane did not vary by more than 0.02% for 30 min and the ratio of the end-tidal and inhaled concentrations of isoflurane was >0.9 for 30 min.

128

Snake Tx Order Time pH PCO2 PO2 BE HCO3 SO2 Na K iCa Glu Hct Hb (mmHg) (mmHg) (mmol/L) (mmol/L) (%) (mmol/L) (mmol/L) (mmol/L) (mmol/L) (%) (g/L) 0 7.358 25.5 492 -11 14.3 100 142 4.0 1.58 1.9 28 95 Iso 3 MAC 7.640 15.2 492 -4 16.4 100 147 3.3 1.58 2.8 21 71 0 7.235 33.1 124 -13 14.0 98 148 3.8 1.63 2.0 29 99 #1 Mid 2 MAC 7.582 13.1 183 -9 12.4 100 150 3.4 1.57 2.5 23 78 0 7.282 21.4 78 -17 10.1 94 145 3.8 1.53 1.6 26 88 N2O 1 MAC 7.533 13.4 218 -11 11.3 100 151 3.4 1.57 2.7 24 82 0 7.353 30.8 539 -8 17.1 100 142 3.2 1.54 1.6 18 61 Iso 3 MAC 7.695 15.3 443 -1 18.7 100 145 3.5 1.46 2.2 18 61 0 7.086 48.6 89 -15 14.6 92 136 3.0 1.16 1.9 15 51 #2 Mid 2 MAC 7.659 13.2 137 -6 14.9 100 139 3.3 1.14 1.9 <15 - 0 7.029 57.9 74 -16 15.3 86 149 3.4 1.74 1.9 22 75 N2O 1 MAC 7.715 11.2 88 -5 14.3 99 148 3.3 1.49 2.9 20 68 0 6.834 60.6 487 -24 10.2 100 150 4.0 1.69 2.2 24 82 Iso 3 MAC 7.691 19.9 576 4 24.1 100 149 3.3 1.59 2.3 20 68 0 7.256 34.1 268 -12 15.1 100 148 3.6 1.74 1.6 23 78 #3 Mid 2 MAC 7.608 14.1 94 -7 14.0 99 149 3.4 1.70 2.1 20 68 0 6.844 70.2 64 -22 12.1 69 152 4.0 1.82 2.1 26 88 N2O 1 MAC 7.701 11.0 55 -6 13.7 96 150 3.5 1.63 2.7 20 68 0 6.790 85.1 72 -22 12.9 72 155 3.6 1.86 2.4 28 95 Iso 1 MAC 7.601 15.1 68 -7 14.9 97 151 3.4 1.6 2.7 19 65 0 7.132 52.8 68 -11 17.7 86 148 3.1 1.80 2.0 24 82 #4 Mid 3 MAC 7.731 11.0 70 -5 14.5 98 150 2.9 1.72 2.3 19 65 0 7.010 45.3 227 -20 11.4 99 148 3.4 1.68 2.7 26 88 N2O 2 MAC 7.507 11.8 60 -14 9.3 94 138 2.8 1.14 2.1 21 71

129

Table D.3 – (Continued)

Snake Tx Order Time pH PCO2 PO2 BE HCO3 SO2 Na K iCa Glu Hct Hb (mmHg) (mmHg) (mmol/L) (mmol/L) (%) (mmol/L) (mmol/L) (mmol/L) (mmol/L) (%) (g/L) 0 7.059 41.5 262 -19 11.7 100 142 3.6 1.48 2.4 25 85 Iso 2 MAC 7.675 12.9 58 -5 15.1 96 147 3.6 1.61 2.9 23 78 0 7.320 28.3 633 -12 14.6 100 143 3.6 1.48 2.3 25 85 #5 Mid 1 MAC 7.643 16.4 470 -3 17.8 100 150 3.8 1.69 2.6 24 82 0 7.185 43.1 223 -12 16.3 100 150 3.4 1.69 3.0 29 99 N2O 3 MAC 7.638 15.5 148 -4 16.6 100 149 3.5 1.58 3.2 25 85 0 6.953 48.8 467 -21 10.8 100 144 3.8 1.60 2.1 20 68 Iso 1 MAC 7.659 16.2 487 -2 18.2 100 145 3.5 1.57 3.6 16 54 0 7.148 43.1 608 -14 14.9 100 146 3.7 1.73 2.1 21 71 #6 Mid 3 MAC 7.683 17.7 511 1 21.0 100 148 3.3 1.71 2.2 16 54 0 6.916 47.7 186 -23 9.7 98 148 4.0 1.77 2.1 25 85 N2O 2 MAC 7.505 16.0 175 -11 12.6 100 138 3.1 1.27 1.8 <15 - 0 7.370 32.5 505 -6 18.8 100 145 3.6 1.62 1.9 23 78 Iso 2 MAC 7.625 14.6 61 -6 15.3 96 148 3.3 1.65 2.3 19 65 0 7.255 39.8 75 -10 17.6 92 147 3.6 1.88 1.8 22 75 #7 Mid 1 MAC 7.640 13.3 59 -7 14.3 96 147 3.4 1.7 1.8 18 61 0 7.177 45.3 103 -12 16.8 96 146 3.7 1.66 1.9 23 78 N2O 3 MAC 7.532 17.8 55 -8 15.0 93 148 3.3 1.63 2.4 20 68 0 7.044 47.4 585 -18 12.9 100 146 3.8 1.69 2.4 22 75 Iso 1 MAC 7.667 18.1 561 0 20.7 100 149 3.7 1.74 2.4 19 65 0 7.101 53.4 580 -13 16.6 100 146 3.8 1.74 2.2 21 71 #8 Mid 3 MAC 7.697 18.8 547 3 23.1 100 149 3.9 1.75 2.4 15 51 0 7.192 37.2 223 -14 14.2 100 147 3.9 1.72 2.6 24 82 N2O 2 MAC 7.627 14.9 186 -6 15.6 100 148 3.6 1.67 2.7 18 61

130

Table D.3 – (Continued)

Snake Tx Order Time pH PCO2 PO2 BE HCO3 SO2 Na K iCa Glu Hct Hb (mmHg) (mmHg) (mmol/L) (mmol/L) (%) (mmol/L) (mmol/L) (mmol/L) (mmol/L) (%) (g/L) 0 7.172 45.5 216 -12 16.7 100 149 3.5 1.82 2.1 28 95 Iso 2 MAC 7.614 17.3 500 -4 17.6 100 151 3.9 1.73 2.3 24 82 0 7.039 49.7 335 -17 13.4 100 146 3.5 1.76 2.0 27 92 #9 Mid 1 MAC 7.627 12.0 75 -9 12.6 98 148 3.2 1.74 3.2 23 78 0 6.872 78.9 65 -19 14.5 72 155 3.8 1.85 2.4 32 109 N2O 3 MAC 7.595 14.9 186 -7 14.4 100 152 3.5 1.71 3.0 26 88

Tx: treatment; Order: order of treatment; PCO2: partial pressure of carbon dioxide; PO2: partial pressure of oxygen; BE: base excess; HCO3: bicarbonate; SO2: oxygen saturation; Na: sodium; K: potassium; iCa: ionized calcium; Glu: glucose; Hct: hematocrit; Hb: hemoglobin.

131

APPENDIX E: Validation of the methods to quantify the plasma concentration of midazolam and 1-hydroxymidazolam in the ball python and descriptive data from chapter 4

Quantifying

Qualifying Confirming

Figure E.1 – Tandem-mass spectrometry spectrum for midazolam in ball python plasma with specific quantifying, qualifying, and confirming ions.

Quantifying

Qualifying

Confirming

Figure E.2 – Tandem-mass spectrometry spectrum for midazolam-D4 in ball python plasma with specified quantifying, qualifying, and confirming ions.

132

Quantifying

Qualifying

Confirming

Figure E.3 – Tandem-mass spectrometry spectrum for 1-hydroxymidazolam in ball python plasma with specific quantifying, qualifying, and confirming ions.

Quantifying Qualifying

Confirming

Figure E.4 – Tandem-mass spectrometry spectrum for 1-hydroxymidazolam-D4 in ball python plasma with specific quantifying, qualifying, and confirming ions.

133

Table E.1 – Validation data for the quantification of midazolam in ball python plasma using high performance liquid chromatography tandem-mass spectrometry.

STANDARD INTRA-DAY PRECISION INTER-DAY PRECISION ACCURACY CONCENTRATION (ng/mL) (%) (%) (%) 1 13.65 16.21 13.62 2 8.56 10.23 9.84 10 2.49 5.69 6.90 20 9.43 8.91 8.43 100 1.17 5.36 5.65 200 0.42 6.97 5.08 500 4.19 8.23 2.42 1,000 9.78 14.01 9.18

Table E.2 – Validation data for the quantification of 1-hydroxymidazolam in ball python plasma using high performance liquid chromatography tandem-mass spectrometry.

STANDARD INTRA-DAY PRECISION INTER-DAY PRECISION ACCURACY CONCENTRATION (ng/mL) (%) (%) (%) 1 14.89 15.42 14.90 2 10.36 12.36 8.79 10 9.25 7.69 7.36 20 3.14 5.79 9.13 100 0.98 3.25 6.47 200 1.59 5.96 4.36 500 9.16 10.11 3.69 1,000 7.35 6.98 8.97

134

Table E.3 – Plasma concentrations of midazolam over time in 6 ball pythons following intracardiac administration of 1 mg/kg of midazolam at 30.4 ± 0.8°C.

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #6 MEAN SD CV (min) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (%) 0† ------3 4,992.39 2,442.44 3,319.66 2,401.80 2,500.42 2,466.56 3,020.54 1,026.85 34.0 5 4,519.13 3,006.74 2,704.39 2,482.51 2,265.63 2,256.29 2,872.45 855.31 29.8 10 4,622.26 3,700.29 2,848.80 2,853.18 3,792.81 2,003.89 3,303.54 920.27 27.9 20 3,645.37 2,213.20 2,145.30 2,723.77 2,228.26 2,075.14 2,505.17 604.02 24.1 30 3,373.88 2,285.43 2,182.71 1,740.46 1,451.72 1,824.00 2,143.03 675.07 31.5 45 2,628.16 1,967.59 2,358.90 1,566.53 1,141.20 1,454.28 1,852.78 568.79 30.7 60 1,843.56 1,676.45 2,128.87 1,517.68 1,112.26 1,214.46 1,582.21 383.74 24.3 120 1,416.12 1,158.66 2,084.58 800.17 592.71 780.64 1,138.81 550.08 48.3 240 867.93 859.63 619.04 657.87 550.96 555.50 685.15 144.07 21.0 480 560.96 571.12 N/A 360.16 322.41 309.91 424.91 130.20 30.6 720 393.16 429.61 399.13 572.22 212.82 249.55 376.08 130.25 34.6 1440 194.38 106.88 130.22 298.33 316.20 507.23 258.87 148.55 57.4 2880 46.67 7.48 66.80 44.47 147.15 215.40 88.00 77.80 88.4 4320 11.46 - - 16.20 30.70 57.38 28.93 20.65 71.4 †Sample collected before midazolam injection; SD: standard deviation; CV: coefficient of variation; -: drug not detected; N/A: no value available due to sampling or analysis error.

135

Table E.4 – Plasma concentrations of midazolam over time in 6 ball pythons following intramuscular administration of 1 mg/kg of midazolam at 30.4 ± 0.8°C.

TIME SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 SNAKE #11 SNAKE #59 MEAN SD CV (min) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (%) 0† ------5 3.47 - 301.10 273.97 199.76 55.18 166.70 132.04 79.2 15 370.61 62.59 182.09 335.27 219.95 234.45 234.16 110.63 47.2 30 491.23 193.14 267.32 377.40 607.81 757.36 449.04 212.46 47.3 45 1,530.13 297.78 411.23 375.20 679.76 635.22 654.89 454.35 69.4 60 958.29 289.61 320.13 523.27 624.91 608.64 554.14 243.72 44.0 90 1,611.68 374.44 512.49 591.05 756.53 822.94 778.19 439.50 56.5 120 1,034.31 395.44 414.02 654.83 776.95 751.38 671.16 241.63 36.0 180 914.26 403.71 369.29 556.43 507.85 810.53 593.68 221.35 37.3 240 784.18 378.20 367.54 607.57 436.40 903.36 579.54 225.16 38.9 480 554.56 269.76 241.52 422.58 439.13 720.47 441.34 179.14 40.6 720 484.02 229.61 167.22 196.18 286.53 665.69 338.21 196.15 58.0 1440 402.15 131.64 100.69 161.94 282.29 528.41 267.85 169.62 63.3 2880 231.05 22.51 39.36 17.80 24.49 237.76 95.49 107.86 112.9 4320 110.44 - N/A - 11.75 123.77 81.98 61.19 74.6 †Sample collected before midazolam injection; SD: standard deviation; CV: coefficient of variation; -: drug not detected; N/A: no value available due to sampling or analysis error.

136

Table E.5 – Plasma concentrations of 1-hydroxymidazolam over time in 6 ball pythons following intracardiac administration of 1 mg/kg of midazolam at 30.4 ± 0.8°C.

TIME SNAKE #1 SNAKE #2 SNAKE #3 SNAKE #4 SNAKE #5 SNAKE #6 MEAN SD CV (min) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (%) 0† ------3 41.72 23.24 39.57 14.59 33.59 24.69 29.57 10.51 35.5 5 53.23 27.90 48.94 67.01 13.52 56.84 44.57 19.95 44.8 10 35.60 40.87 231.87 50.40 16.49 107.68 80.49 80.30 99.8 20 55.01 112.09 215.59 74.47 28.65 104.28 98.35 65.23 66.3 30 78.17 165.51 220.02 56.27 57.95 84.22 110.36 67.02 60.7 45 89.16 183.74 241.96 95.85 69.53 94.36 129.10 68.09 52.7 60 86.67 230.88 231.72 94.06 75.29 98.61 136.20 74.08 54.4 120 128.50 737.67 256.94 133.04 66.05 106.32 238.09 252.96 106.2 240 156.13 409.44 263.47 148.57 100.88 133.40 201.98 115.50 57.2 480 222.12 428.34 300.08 169.87 140.15 157.35 236.32 110.49 46.8 720 277.35 486.12 328.95 469.36 159.99 178.94 316.78 139.45 44.0 1440 246.85 269.03 200.57 441.34 455.60 448.03 343.57 116.94 34.0 2880 185.66 119.18 188.15 206.50 472.01 532.16 283.95 172.60 60.8 4320 92.77 30.48 - 52.04 205.41 131.12 102.36 69.35 67.7 †Sample collected before midazolam injection; SD: standard deviation; CV: coefficient of variation; -: drug not detected.

137

Table E.6 – Plasma concentrations of 1-hydroxymidazolam over time in 6 ball pythons following intramuscular administration of 1 mg/kg of midazolam at 30.4 ± 0.8°C.

TIME SNAKE #7 SNAKE #8 SNAKE #9 SNAKE #10 SNAKE #11 SNAKE #59 MEAN SD CV (min) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (ng/mL) (%) 0† ------5 ------15 6.79 - 3.12 4.93 3.58 3.14 4.31 1.57 36.4 30 11.86 5.84 11.16 12.20 13.46 7.98 10.42 2.90 27.8 45 22.07 19.00 31.91 20.51 26.86 10.74 21.85 7.21 33.0 60 28.02 33.94 33.65 35.54 30.88 15.42 29.57 7.43 25.1 90 54.54 63.95 74.13 52.26 49.82 25.21 53.32 16.44 30.8 120 63.47 81.12 66.63 73.55 72.45 30.08 64.55 17.95 27.8 180 76.10 107.03 78.94 77.37 73.59 49.51 77.09 18.30 23.7 240 75.79 121.81 89.59 100.73 160.00 109.23 109.53 29.37 26.8 480 117.97 165.20 88.90 142.96 159.10 106.38 130.08 30.48 23.4 720 153.21 187.52 89.76 112.77 158.07 149.48 141.80 34.92 24.6 1440 223.28 251.54 95.37 232.85 159.98 223.91 197.82 58.94 29.8 2880 254.64 133.31 86.83 150.08 99.00 247.64 161.92 72.80 45.0 4320 237.98 28.81 N/A 45.99 30.24 230.83 114.77 109.45 95.4 †Sample collected before midazolam injection; SD: standard deviation; CV: coefficient of variation; -: drug not detected; N/A: no value available due to sampling or analysis error.

138