Evaluation of Sedative Behavior Associated with Midazolam, Before and After Flumazenil, in Corn Snakes (Pantherophis guttatus) and Mexican Black Kingsnakes (Lampropeltis californiae nigrita)

Introduction

Snakes are becoming increasingly popular as pets, with corn snakes (Pantherophis guttatus) and kingsnakes (Lampropeltis spp.) considered common privately owned species. Veterinarians frequently require chemical restraint to facilitate safe handling, clinical examination, and to perform quality diagnostics. Due to limited studies on sedation and anesthesia in snakes, drug doses are often extrapolated from other reptilian species, resulting in potential complications, and species-specific dosing is necessary for optimal veterinary care (Sadar and Ambros, 2019).

Midazolam is a water-soluble benzodiazepine that results in muscle relaxation, sedation, anxiolysis, and amnesia (Arnett-Chinn et al., 2016; Larouche et al., 2019b; Hong, 2021). It is a reversible gamma-aminobutyric acid (GABA) A receptor agonist with a safe cardiorespiratory profile and minimum anesthetic concentration–reducing effects in various reptile species (Arnett-Chinn et al., 2016; Lopes et al., 2017; Miller et al., 2018; Larouche et al., 2019b; Hong, 2021). The sedative effects of this drug have been evaluated in multiple snake species as the sole sedative agent and in combination with other anesthetic drugs (Sladky and Mans, 2012; Arnett-Chinn et al., 2016; De Simone et al., 2017; Budden et al., 2018; Larouche et al., 2019a; Larouche et al., 2019b; Sadar and Ambros, 2019; Yaw et al., 2020). Mild to moderate sedation is often reported in reptiles when midazolam is used as a sole sedative agent administered via IM, SC, intracardiac, or intracoelomic injection (Arnett Chinn et al., 2016; De Simone et al., 2017; Larouche et al., 2019a, Larouche et al., 2019b).

Sedative effects of midazolam can be reversed with administration of flumazenil, a competitive GABA_A antagonist (Hong, 2021). Flumazenil is a neutral allosteric modulator that does not activate or inhibit GABA function, thus preventing only GABA receptor binding of benzodiazepines (Hong, 2021). Effectiveness of flumazenil in reversing midazolam-induced sedation in reptiles is infrequently reported in sedation studies, potentially from a lack of long-term monitoring post-reversal administration. Interestingly, there is evidence of re-sedation in midazolam-sedated ball pythons 3 h after flumazenil administration, with some snakes remaining sedate for up to 4 days post-reversal (Larouche et al., 2019a). This reinforces the significance of studying the effects of reversal drug administration in recovering reptiles from sedation or anesthesia.

The goal of this study was to assess the sedation quality and cardiorespiratory effects of 2 mg/kg midazolam administered SC in apparently healthy corn snakes and Mexican black kingsnakes (Lampropeltis californiae nigrita). The study then determined whether antagonism with SC flumazenil effectively reverses the sedation and whether re-sedation occurs in these species, as described in ball pythons (Python regius). We hypothesized that midazolam re-sedation would occur within 24 h of flumazenil administration in both snake species, based on the published data in ball pythons (Larouche et al., 2019a).

Materials and Methods

Sedation testing

In a randomized, double-blind, complete crossover study, all snakes were administered 2 mg/kg midazolam (5 mg/mL, Fresenius Kabi Canada Ltd., Richmond Hill, Ontario, Canada) or 0.9% NaCl (volume equivalent to their dose of midazolam, Hospira, Inc, Lake Forest, IL, USA) SC in the cranial third of the body via manual restraint, with a minimum 7 day washout period between treatments. Following midazolam or saline administration, snakes were placed in their respective cages in a dark room for 1 h with ambient temperatures remaining between 25–28°C (77–82.4°F). Dosing was determined based on previous studies in which midazolam was administered as the sole sedative agent and provided safe and adequate sedation (Arnett-Chin et al., 2016; De Simone et al., 2017; Bressan et al., 2019; Larouche et al., 2019a). For each snake, respiratory rate (RR), heart rate (HR), and sedation parameters were evaluated immediately before administering midazolam or saline and 1 h after administration. The RR was assessed by counting body excursions over 1 min, and HR was measured via manual restraint using a Doppler ultrasonographic probe on the ventral body wall in the region of the heart (Parks Medical Electronics, Inc., Las Vegas, NV, USA).

Sedation variables were evaluated by a blinded observer and scored using a three-point scale (Table 1). Sedation variables included spontaneous body movement, tongue flicking, menace response, and righting reflex (Bunke et al., 2018). To assess sedation variables, snakes were placed into a plastic container (93 cm length × 53.3 cm width × 49.5 cm height [36.6 in × 30 in × 19.5 in]) with an open top and recorded with a video camera placed on a platform 20 cm above the container. The container resembled the type of enclosure in which the snakes were housed before acquisition for this study. Snakes were placed in the container and first recorded for 1 min to assess their level of spontaneous body movement and the number of tongue flicks observed. Spontaneous movement was scored based on the length of time a snake demonstrated no movement in the recorded period. Following this 1 min duration, a menace response was assessed where the investigator moved their hand quickly toward the head of the snake. A positive menace response occurred when a snake reacted by flexing the cranial third of the body to move the head quickly away from the hand. If the snakes did not flex the cranial third of the body in response to the gestured hand, a menace response was considered absent. A righting reflex was performed last by gently placing the snake in dorsal recumbency and evaluating the length of time it took to return to ventral recumbency.

Table 1.Sedation scoring for corn snakes (Pantherophis guttatus) and Mexican black kingsnakes (Lampropeltis californiae nigrita) administered 2 mg/kg midazolam SC or volume equivalent saline and reversed with flumazenil 0.05 mg/kg SC according to four behavioral descriptions.

View Fullscreen

Results

Corn snakes

Midazolam sedation effects

Mean sedation scores 1 h post-midazolam administration were significantly different compared with baseline for movement (P < 0.001; Fig. 1A), tongue flicking (P < 0.001; Fig. 1B), and menace response (P = 0.003; Fig. 1C). Comparatively, a statistically significant difference occurred in mean sedation scores 1 h post-midazolam administration compared with saline administration for movement (P < 0.001; Fig. 1A), tongue flicking (P < 0.001; Fig. 1B), and menace response (P < 0.001; Fig. 1C). Mean righting reflex was significantly delayed compared with baseline and saline measurements; however, this reflex was maintained in all snakes (P = 0.003, P < 0.001, respectively; Fig. 1D).

View Figure

• Download as Powerpoint
• Download Figure

Cardiorespiratory effects of midazolam

The mean HR and RR in corn snakes 1 h post-midazolam administration were 68.3 beats per min (range 50–90 beats per min) and 5.7 breaths per min (range 4–12 breaths per min), respectively. No statistically significant differences in HR and RR were observed between baseline measurements and 1 h post-midazolam administration (Fig. 1E,F). The mean HR and RR in corn snakes 1 h post-saline administration were 73.3 beats per min (range 60–100 beats per min) and 7.7 breaths per min (range 4–16 breaths per min), respectively. Similarly, no statistically significant difference was found in mean HR and RR between saline and midazolam 1 h post-administration (Fig. 1E,F).

Flumazenil sedation reversal and cardiorespiratory effects

No statistically significant differences occurred between mean baseline and sedation scores for any of the tested variables in midazolam- and saline-treated corn snakes at 1, 4, and 24 h post-flumazenil administration (Fig. 1A–D). Comparatively, no significant differences were found in mean HR or RR between baseline and 1, 4, and 24 h post-flumazenil administration (Fig. 1E,F).

Mexican black kingsnakes

Midazolam sedation effects

Mean sedation scores in Mexican black king snakes 1 h post-midazolam administration were significantly different compared with baseline for spontaneous body movement (P < 0.001; Fig. 2A), tongue flicking (P < 0.001; Fig. 2B), menace response (P = 0.007; Fig. 2C), and righting reflex (P < 0.001; Fig. 2D). Righting reflex was absent 1 h post-midazolam treatment in six of seven Mexican black kingsnakes. A statistically significant difference was found in mean sedation scores 1 h post-midazolam administration compared with saline administration for menace response (P = 0.002; Fig. 2C) and righting reflex (P = 0.016; Fig. 2D): Three of seven Mexican black kingsnakes had an absent righting reflex 1 h post-saline administration. Contrastingly, no significant differences occurred in mean spontaneous body movement or tongue flicking 1 h post-saline compared with post-midazolam administration. Rather, a significant difference occurred between mean baseline and 1 h post-saline administration for spontaneous body movement (P = 0.015; Fig. 2A) and tongue flicking (P = 0.031; Fig. 2B).

View Figure

• Download as Powerpoint
• Download Figure

Cardiorespiratory effects of midazolam

The mean HR and RR in Mexican black kingsnakes 1 h post-midazolam administration were 51.7 beats per min (range 24–72 beats per min) and 2.6 breaths per min (range 1–4 breaths per min), respectively. No statistically significant differences in mean RR and HR were observed between baseline measurements and 1 h post-midazolam administration. The mean HR and RR in Mexican black kingsnakes 1 h post-saline administration were 47.4 beats per min (range 24–72 beats per min) and four breaths per min (range 2–6 breaths per min), respectively. Similarly, no statistically significant differences were found in mean RR and HR between saline and midazolam 1 h post-administration (Fig. 2E,F).

Flumazenil-sedation reversal and cardiorespiratory effects

Mean sedation scores in midazolam-treated snakes 1 h post-flumazenil administration were significantly different from baseline for spontaneous body movement (P < 0.001; Fig. 2A), tongue flicking (P = 0.005; Fig. 2B), menace response (P = 0.023; Fig. 2C), and righting reflex (P = 0.014; Fig. 2D). Similarly, mean sedation scores for spontaneous body movement and menace response were significantly different compared with saline 1 h post-flumazenil administration in midazolam-treated snakes (Fig. 2A,B). A significant difference also occurred 1 h post-flumazenil administration between baseline scores and saline-treated snakes for spontaneous body movement (P = 0.015; Fig. 2A) and tongue flicking (P = 0.031; Fig. 2B). Mean and median movement-sedation scores for midazolam-treated snakes 1 h post-flumazenil were 2.7 and 3, respectively (range 2–3), compared with 2 and 2, respectively (range 1–3), in saline-treated snakes at this time point. Mean and median tongue-flicking sedation scores for midazolam-treated snakes 1 h post-flumazenil were 1.28 and 2, respectively (range 0–2), whereas the mean and median scores for saline-treated snakes 1 h post-flumazenil were 1 and 1, respectively (range 0–2).

A significant difference was found 4 h post-flumazenil administration between mean baseline scores in midazolam-treated snakes for spontaneous body movement (P < 0.001; Fig. 2A), tongue flicking (P < 0.001; Fig. 2B), and righting reflex (P = 0.007; Fig. 2D); no significant difference was observed in menace response (Fig. 2C). No significant differences were observed in sedation scores between saline and midazolam treatment at this timepoint. However, a significant difference was found in sedation scores for spontaneous body movement in saline-treated snakes compared with baseline (P = 0.009; Fig. 2A) at 4 h post-flumazenil administration.

No significant differences were found between mean baseline and sedation scores for any of the tested variables in midazolam- or saline-treated Mexican black kingsnakes 24 h post-flumazenil administration. Similarly, no significant difference was found between saline- and midazolam-treated snakes at this time point (Fig. 2A–D).

No significant differences occurred in mean HR or RR between baseline and midazolam- or saline-treated snakes at 1, 4, or 24 h post-flumazenil administration (Fig. 2E,F).

Discussion

The objective of this study was to assess the sedative effects of SC midazolam, the quality of reversal with flumazenil, and the associated cardiorespiratory effects in corn snakes and Mexican black kingsnakes. In corn snakes, midazolam resulted in statistically significant and subjectively moderate sedative effects with delayed righting reflex and overall clinically relevant reduction in movement. Relatedly, Mexican black kingsnakes demonstrated similar subjectively moderate sedative effects when comparing midazolam-treated snakes to their baseline and control data measurements. Unlike previously published data with ball pythons, corn snakes did not demonstrate behavioral effects associated with re-sedation within 24 h of flumazenil administration, which was contradictory to our initial hypothesis (Larouche et al., 2019a). However, midazolam-treated kingsnakes did not return to their baseline sedation scores until 24 h post-flumazenil administration. At 1 and 4 h post-flumazenil, sedation scores trended toward baseline, but statistically significant differences remained compared with baseline at these time points. Overall, no increases in sedation scores were found that would suggest re-sedation in this recovery period, which is subjectively considered more clinically relevant. Assessment for re-sedation in this species after 24 h was not carried out in this study and should still be a consideration.

In contrast to the corn snakes, Mexican black kingsnakes were observed to demonstrate statistically significant changes in behavioral sedation scores when comparing saline administration with baseline data in the crossover portion of this study, where some snakes demonstrated apparent sedative effects from saline administration. These effects were observed with spontaneous body movement and tongue flicking; however, mean and median sedation scores for these parameters demonstrated that midazolam-treated snakes had higher scores, suggestive of increased sedation compared with saline-treated snakes. Increased sedation in saline-treated king snakes may be related to individual temperament and response to repeated handling or potentially a behavioral response of feigning death with intentional reduction in spontaneous body movement following handling, which has been reported in other snake species (Gehlbach, 1970).

The predatory response of feigning death in other snake species often involves an initial phase of erratic defensive behavior before cessation of movement (Gehlbach, 1970). Subjective behavioral resistance to manual restraint for HR acquisition was not recorded before sedation scoring of snakes, and thus this defensive behavioral pattern may be a consideration for apparent changes in sedation variables with saline administration alone. Another consideration for this observation in king snakes is that this species may be more likely to freeze in the presence of an observer, as has been described in green iguanas (Iguana iguana; Fleming and Robertson, 2012). An additional consideration is the metabolic status of the Mexican black kingsnakes because they had been completing a reported brumation period before acquisition for this study. Brumation can result in decreased metabolism overall with changes in lipid and protein levels that may or may not have been present in this cohort (Gillooly et al., 2001; Long, 2016). These metabolic changes can have significant effects on drug distribution, unbound drug fraction in the blood, and, therefore, overall clinical drug effects. This could contribute to significant variation of drug effects in the same individual snake when comparing peri-brumation effects and non-brumation drug effects.

Despite housing snakes in the appropriate species-specific optimal temperature zone, king snakes were fasted for the short duration of this study, as instructed by the owner of the snakes, which may have also affected behavior. Given the unique physiology of reptiles, overall husbandry and metabolic status is a key consideration with anesthetic metabolism and drug effects (Sladky and Mans, 2012). Saline, however, is routinely used as a control agent for crossover anesthetic studies; therefore, a behavioral or potential metabolic component is considered more likely. Overall, a statistically significant and clinically relevant reduction in righting reflex and absent menace gesture response were observed in midazolam-treated Mexican black kingsnakes compared with baseline and saline treatment. These variables may be more reliable indicators of sedation-induced central nervous system depression given that these behaviors are frequently used in clinical neurologic examinations of snakes (Hunt, 2015). Thus, despite similarities in sedation scores between midazolam- and saline-treated snakes for spontaneous body movement and tongue flicking, clinically appropriate and subjectively moderate sedation is likely still achieved.

Midazolam is a GABA agonist that potentiates inhibitory neurotransmitters in the central nervous system through increased chloride conductance in postsynaptic neurons, resulting in hyperpolarization and subsequent reduction in neurotransmission (Hong, 2021). Multiple studies have demonstrated variable dose-dependent sedative effects of midazolam in reptiles when used alone. In snakes, midazolam is frequently administered at 1–2 mg/kg IM, with a pharmacokinetic study reported in ball pythons that demonstrated maximum plasma concentrations at 2 to 3 h post-IM injection (Larouche et al., 2019b). The level of sedation achieved in redtail boa constrictors (Boa constrictor) and ball pythons administered midazolam (1 and 2 mg/kg) intracoelomically or IM is comparable with that achieved in the presented study, with reduced locomotion, muscle relaxation, and a reduced righting reflex (De Simone et al., 2017; Larouche et al., 2019b). However, in redtail boa constructors, increased duration of sedation occurred at higher doses of midazolam (De Simone et al., 2017; Larouche et al., 2019b). The duration of sedation with midazolam was not assessed in the present study, and thus further investigation would be necessary to determine whether species-specific clinical duration of sedation is observed in corn snakes and Mexican black kingsnakes.

Moderate to deep sedation should be considered when performing potentially invasive diagnostic procedures (Sladky and Mans, 2012). The moderate sedation achieved in both corn snakes and Mexican black kingsnakes with 2 mg/kg SC midazolam would be appropriate for reducing stress during handling and for noninvasive diagnostics such as imaging acquisition without manual restraint, peripheral blood collection, and intraoral examination. Tracheal intubation was not performed in this study because all animals maintained spontaneous ventilation following sedation administration. However, given that the righting reflex was reduced, the use of midazolam alone may or may not provide sufficient sedation for endotracheal intubation in healthy corn and Mexican black kingsnakes. This result is comparable with other squamates sedated with midazolam as a sole sedative in which marked relaxation, reduced movement, and maintained righting reflex were observed (Arnett-Chinn et al., 2016; Bressan et al., 2018; Larouche et al., 2019a). Although it is currently unknown whether midazolam provides analgesia in these snake species, a multimodal approach should be considered with the addition of concurrent effective analgesics, injectable or inhalant anesthetics, and/or local anesthetics for more invasive procedures such as biopsies, cloacoscopy, or gastroscopy PO (Sladky and Mans, 2012).

The reversibility of midazolam offers increased control over the duration of recovery in reptiles and allows for correction of potential adverse anesthetic effects. However, few studies have evaluated the effects and duration of action of reversal agents in reptiles (Sladky and Mans, 2012). Flumazenil is a competitive GABA antagonist that acts as a neutral allosteric modulator, where it binds GABA receptors without activation or inhibition (Hong, 2021). Subcutaneous and IM flumazenil doses reported in reptiles range from 0.01 mg/kg to 0.08 mg/kg (Sladky and Mans, 2012). The present study demonstrated appropriate antagonistic effects of 0.05 mg/kg SC flumazenil in corn snakes at 1 h following reversal administration and 24 h post-flumazenil administration in Mexican black kingsnakes. Sedation scores in midazolam-treated Mexican black kingsnakes decreased over the reversal evaluation period; however, statistically significant elevation in sedation scores compared with baseline at 1 and 4 h post-flumazenil administration were still observed. No increases in mean sedation scores were observed following reversal administration, and thus no evidence of re-sedation was observed in either snake species within 24 h post-flumazenil administration. In contrast, re-sedation in midazolam-sedated ball pythons antagonized with 0.08 mg/kg IM flumazenil was previously demonstrated, in which initial reversal of sedation and muscle relaxation was observed within 10 to 60 min of flumazenil administration. However, all ball pythons had evidence of re-sedation 3 h after flumazenil administration and observed sedation remained present for up to 2 to 4 days (Larouche et al., 2019a).

Anecdotally, the authors of the current study have observed this delayed re-sedation effect in clinical cases in which midazolam was administered to green anacondas (Eunectes murinus) for routine wellness exams (unpublished data). A consideration for this variability is the rate of flumazenil metabolism compared with midazolam metabolism, which can be influenced by drug dosing, route of administration, and rate of systemic antagonist uptake. This variable interspecies clinical effect of flumazenil further reinforces the significance of species-specific drug dosing and necessity for long-term post-sedation monitoring for species with unknown reversal effectiveness.

Sedation and reversal drugs in this study were administered SC. Intramuscular or IV routes of administration have traditionally been used in reptiles over SC administration given the presumed associated unpredictability of SC anesthetic induction (Schumacher and Yelen, 2006). Time to onset of initial sedative effects and working sedative plane were not measured in this study. However, appropriate sedation depth was achieved within 1 h post-midazolam administration. This is similar to the maximum sedative effect reported in ball pythons sedated with 1 and 2 mg/kg IM midazolam, where peak effects were observed within 60 min of administration (Larouche et al., 2019a). Subcutaneous administration of various analgesic drugs was demonstrated to have relatively rapid behavioral effects, as well as documented elevated plasma concentrations of parent drugs and metabolites (Sladky and Mans, 2012). These data suggest that SC administration of anesthetic and analgesic drugs in reptiles may provide similar results to IM administration. In addition, SC administration provides a less stressful route compared with IM administration, especially in small species with less robust muscle mass (Bertelsen, 2014). Fewer studies assess the effects of administration on sedation reversal agents. Intramuscular administration of flumazenil in midazolam-sedated ball pythons had appropriate reversal after 1 h and subsequent re-sedation within 3 h in all animals. Because flumazenil was administered SC in the present study, it is possible that the snakes in this study could have experienced re-sedation long after the observation period in this study or 7 day washout period. Continued studies assessing the variation in pharmacokinetics and pharmacodynamics of SC vs. IM anesthetic agents should be performed in the future.

For both midazolam and flumazenil administration, no adverse cardiorespiratory effects were observed in this study. All snakes maintained spontaneous ventilation, and no significant changes were detected in either HRs or RRs when compared with baseline measurements. Species-specific cardiorespiratory drug effects should be considered because multiple studies in Boidae species suggest IM 1 and 2 mg/kg midazolam can reduce both HR and RR (Lopes et al., 2017; Miller et al., 2018; Larouche et al., 2019a).

This study had several limitations. A power analysis was not performed, rather the maximum number of snakes available were used, and although no adverse clinical anesthetic effects were observed, monitoring for such effects should be considered in a clinical setting when using these sedation protocols. Snakes used in this study were all apparently healthy, and caution is warranted when using these sedation doses in debilitated snakes. Furthermore, sedation scores were not evaluated between 4 and 23 h or after 24 h post-flumazenil administration, thus it is unknown whether potential changes in sedation levels occurred during these unobserved periods.

In conclusion, this study demonstrated moderate sedation effects 1 h following SC administration of midazolam (2 mg/kg) in both corn snakes and Mexican black kingsnakes. Subsequent SC flumazenil administration (0.05 mg/kg) resulted in appropriate antagonistic effects with no evidence of re-sedation within 24 h in both species, which was contrary to published data in ball pythons (Larouche et al., 2019a). No adverse cardiorespiratory effects were observed with midazolam or flumazenil administration in either snake species.