Edinburgh Research Explorer

Standardization of electrocardiographic examination in corn snakes (Pantherophis guttatus)

Citation for published version: Lewis, M, Bouvard, J, Eatwell, K & Culshaw, G 2020, 'Standardization of electrocardiographic examination in corn snakes (Pantherophis guttatus)', Veterinary Record. https://doi.org/10.1136/vr.105713

Digital Object Identifier (DOI): 10.1136/vr.105713

Link: Link to publication record in Edinburgh Research Explorer

Document Version: Peer reviewed version

Published In: Veterinary Record

General rights Copyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights.

Take down policy The University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer content complies with UK legislation. If you believe that the public display of this file breaches copyright please contact [email protected] providing details, and we will remove access to the work immediately and investigate your claim.

Download date: 11. Oct. 2021 Veterinary Record

Confidential: For Review Only Standardisation of electrocardiographic examination in corn snakes (Pantherophis guttatus)

Journal: Veterinary Record

Manuscript ID vetrec-2019-105713.R1

Article Type: Original research

Date Submitted by the 13-Dec-2019 Author:

Complete List of Authors: Lewis, Martyn; the royal (dick) school of veterinary studies, exotics; University of Edinburgh Bouvard, Jonathan; the royal (dick) school of veterinary studies, Cardiology Eatwell, Kevin; Edinburgh University, Royal (Dick) School for Veterinary Studies, Hospital for Small Culshaw, Geoff; Royal (Dick) School of Veterinary Studies, Hospital for Small Animals

Introduction: Corn snakes are a very common pet , yet there is an absence of evidence-based literature standardising collection of electrocardiographs (ECG) or detailing ECG deflection morphology in the normal . We describe a well-tolerated, reproducible technique and detail the cardiac cycle in terms of lead 2 equivalent waveforms and intervals. Animals: 29 adult corn snakes. Materials and methods: This prospective study evaluated under species- appropriate, standardised conditions, a technique for producing standard six-lead ECG tracings. Lead 2 equivalent cardiac cycles were described in detail and statistically analysed for gender, weight, length, heart rate Abstract: and mean electrical axis. Results: High-quality tracings demonstrated common ECG characteristics for this species, including: no Q, S or SV waves, prolonged PR and RT intervals, rhythmic oscillation of the baseline, short TP segments and a right displaced mean electrical axis. An influence of gender, weight or length on heart rate and mean electrical axis was not identified. Conclusions: To the authors’ knowledge, this is the first study to describe a standardised technique for recording ECG in significant numbers of normal corn snakes. Ranges have been provided that may be of diagnostic value or form the basis for future development of reference intervals for this species

https://mc.manuscriptcentral.com/vetrec Page 1 of 27 Veterinary Record

1 2 3 4 5 6 7 8 9 10 11 Confidential: For Review Only 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 2 of 27

1 2 3 The Royal (Dick) School of Veterinary Studies and the Roslin Institute, The 4 University of Edinburgh, Easter Bush Campus, Midlothian, UK EH25 9RG 5 6 7 8 Standardization of electrocardiographic examination in corn snakes (Pantherophis 9 guttatus) 10 11 Martyn Lewis 1 BVM&S, Jonathan Bouvard 2 DVM, Kevin Eatwell1 BVSc (hons) 12 Confidential: For Review Only DZooMed (Reptilian) Dip ECZM (Herpetology and Small Mammals), Geoff J. 13 2 14 Culshaw BVMS, PhD, DVC 15 16 17 1 Dick Vet Rabbit and Exotic Animal Practice, Hospital for Small Animals, Royal 18 (Dick) School of Veterinary Studies, The University of Edinburgh, Roslin, Midlothian, 19 20 UK 2 21 Cardiopulmonary Service, Hospital for Small Animals, Royal (Dick) School of 22 Veterinary Studies, The University of Edinburgh, Roslin, Midlothian, UK 23 24 E-mail for correspondence: [email protected] 25 26 27 28 An abstract containing preliminary results of this study was presented at the 29 International Conference on Avian, Herpetological and Exotic Mammal medicine 30 (iCARE), May 2019, London, UK 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

https://mc.manuscriptcentral.com/vetrec Page 3 of 27 Veterinary Record

1 2 3 Abstract 4 5 6 Introduction: Corn snakes are a very common pet reptile species, yet there is an 7 8 absence of evidence-based literature standardising collection of electrocardiographs 9 10 (ECG) or detailing ECG deflection morphology in the normal animal. We describe a 11 12 Confidential: For Review Only well-tolerated, reproducible technique and detail the cardiac cycle in terms of lead 2 13 14 15 equivalent waveforms and intervals. 16 17 Animals: 29 adult corn snakes. 18 19 Materials and methods: This prospective study evaluated under species- 20 21 22 appropriate, standardised conditions, a technique for producing standard six-lead 23 24 ECG tracings. Lead 2 equivalent cardiac cycles were described in detail and 25 26 statistically analysed for gender, weight, length, heart rate and mean electrical axis. 27 28 29 Results: High-quality tracings demonstrated common ECG characteristics for this 30 31 species, including: no Q, S or SV waves, prolonged PR and RT intervals, rhythmic 32 33 oscillation of the baseline, short TP segments and a right displaced mean electrical 34 35 axis. An influence of gender, weight or length on heart rate and mean electrical axis 36 37 38 was not identified. 39 40 Conclusions: To the authors’ knowledge, this is the first study to describe a 41 42 standardised technique for recording ECG in significant numbers of normal corn 43 44 45 snakes. Ranges have been provided that may be of diagnostic value or form the 46 47 basis for future development of reference intervals for this species. 48 49 50 51 52 53 54 55 56 57 58 59 60

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 4 of 27

1 2 3 4 Introduction 5 6 7 Evaluation of cardiac performance and diagnosis of cardiac disease in the ophidian 8 9 patient has historically received little attention, even in captive reptile species, such 10 11 as the corn snake that are common both in private and zoological collections. 12 Confidential: For Review Only 13 14 For captive poor husbandry is ubiquitous and pervasive, and could 15 16 predispose to cardiac disease either through immunocompromise or 17 18 nutritional/metabolic derangement.1 2 Indeed, both primary and secondary forms of 19 20 21 cardiac disease are widely described in literature and demonstrate the broad 22 23 spectrum of conditions from which snakes may suffer.3-12 Despite this, the 24 25 prevalence of cardiac disease in snakes is unknown, and our understanding of the 26 27 impact of cardiac disease on morbidity and mortality in even very common pet 28 29 30 species, such as corn snakes, is rudimentary in comparison to more familiar 31 32 companion mammals. 33 34 Identification of cardiac disease by both owner and clinician can be very 35 36 37 challenging. The typically sedentary lifestyles and highly restricted captive 38 39 environments of many pet snakes places little demand on the cardiovascular system. 40 41 A low basal metabolic rate and high tolerance for anaerobic respiration means that 42 43 1 13 44 heart disease is usually very far advanced before it becomes clinically apparent. 45 46 Basic cardiac assessment, such as auscultation, has limited value due to poor 47 48 contact between stethoscope and scales. While more advanced investigations, such 49 50 as radiography, electrocardiography (ECG) and echocardiography, suffer from a 51 52 1 13 14 53 paucity of species-specific reference ranges and standardised methodologies. 54 55 In common with other non-crocodilian reptiles, the ophidian heart is a three- 56 57 chambered structure composed of two atria and a single, common ventricle. A 58 59 60 reptile-specific pre-filling chamber, the sinus venosus, is situated caudal to the right

https://mc.manuscriptcentral.com/vetrec Page 5 of 27 Veterinary Record

1 2 3 atrium15 and contains spontaneously depolarising pacemaker cells under 4 5 2 16 17 6 vagosympathetic control. There is no sino-atrial node, atrial-ventricular node or 7 8 His/Purkinje network that form the specialised electrical conduction system of higher 9 10 vertebrates, and no annulus fibrosis exists to provide an insulating plane between 11 12 Confidential: For Review Only the atria and ventricle.14 15 Instead a canal or “funnel” of atrioventricular myocardium, 13 14 15 which resembles the ill-defined AV node region of embryonic mammals and birds, 16 17 serves as a conduit for electrical conduction into the ventricle. Despite these 18 19 structural differences, the reptilian ECG waveform is morphologically similar to the 20 21 22 mammalian, with a P wave, QRS complex and T wave representing sequential atrial 23 24 depolarisation, ventricular depolarisation and ventricular repolarisation respectively.1 25 26 13 14 27 28 29 Electrocardiography is widely used in traditional companion mammal species 30 31 to aid evaluation of cardiac health and to monitor cardiac rhythm while under general 32 33 anaesthesia, and should also represent an attractive option for common captive 34 35 snakes to both specialists and general practitioners. Unfortunately, very little 36 37 38 literature exists detailing a standardised approach to recording ECGs in snakes or 39 40 describing wave amplitudes or segment durations. Previous research has typically 41 42 involved a variety of different study designs, very small study cohorts or mixed 43 44

45 species collections and data sets cannot be extrapolated from companion mammals 46 47 or even other reptilian species.18-22 We sought to correct this deficiency in a common 48 49 pet species, the corn snake, and provide information useful to cardiac assessment. 50 51 52 We hypothesised that ECGs can be easily recorded from conscious corn 53 54 snakes in a non-invasive and well tolerated way, and that they take on a 55 56 recognisable sequence of PQRST deflections amenable to measurement. To 57 58 59 60

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 6 of 27

1 2 3 investigate this hypothesis a standardised technique for recording and measuring 4 5 6 ECGs was employed. 7 8 9 10 Materials and Methods 11 12 Confidential: For Review Only The study population comprised 29 adult corn snakes (Pantherophis guttatus) (19 13 14 15 males and 10 females) from privately-owned sources or a locally-based charity 16 17 rehoming centre, fed whole prey items on a seven-day cycle. As specific ages were 18 19 unknown snakes were classified as adult based on size (body weight and snout to 20 21 22 vent length). Gender was determined by use of a metal probe passed into the 23 24 hemipenis or musk sac (female if probe depth <5 scutes, male if > 6 scutes). Owners 25 26 were asked to withhold food for at least four days prior to hospitalisation.23 24 None of 27 28 29 the snakes were undergoing ecdysis at the time of ECG recording. Ethical approval 30 31 was obtained from an independent Edinburgh University Veterinary Ethical Review 32 33 Committee (VERC). 34 35 Subjects were deemed healthy based on clinical history and a physical exam 36 37 38 which comprised evaluation of demeanour, body condition, respiratory rate and 39 40 effort, coelomic palpation, and integumentary, oral and cloacal assessments. To 41 42 allow for thermal acclimation and minimise the influence of travel stress, subjects 43 44 45 were housed individually in reptile vivaria for at least four hours prior to ECG 46 47 examination. Ambient vivarium temperature gradients ranged from 26 - 30oC, which 48 49 was considered to be within the preferred optimum temperature zone for this 50 51 25 52 species. 53 54 The ECGs were performed at a room temperature of 22-24oC as soon as the 55 56 snakes were removed from the vivaria. Subjects were supported in ventral 57 58 recumbency by a single assistant without the use of chemical restraint. When 59 60

https://mc.manuscriptcentral.com/vetrec Page 7 of 27 Veterinary Record

1 2 3 physical restraint was required, it was limited to gentle manual restriction of 4 5 6 movement until the subject settled (figure 1). The heart was estimated at around 7 8 25% snout-to-vent length then either identified visually, with ultrasound confirmation, 9 10 or found directly with the probe and coupling gel (figure 1). Non-traumatic adhesive 11 12 Confidential: For Review Only ECG pads (Ambu ® BlueSensor N) were applied to the lateral skin surface in a four- 13 14 15 lead system, 1cm cranial and 1cm caudal to the heart (Figures 1 and 2). Attachment 16 17 of the ECG pads to the subjects’ scales was simple and reliable, and removal post- 18 19 recording was achieved either with gentle manual traction or the use of a proprietary 20 21 TM 22 spray designed for adhesive bandage removal (Eaze-Off , Millpledge Veterinary). 23 24 In both instances, removal was quick and atraumatic. 25 26 Due to the absence of limbs, positioning of the electrodes was based around a 27 28 13 14 29 modified Einthoven triangle. The red electrode was placed on the right side and 30 31 cranial to heart, the yellow on the left and cranial to heart, and the green on the left 32 33 and caudal to heart (Figure 2). The ECG was recorded continuously for three 34 35 minutes using both 25 and 50 mm sec -1, at sensitivities of 10 and 20mV. 36 37 38 Standard ECG measurements were made from the lead 2 equivalent (negative 39 40 electrode right and cranial to heart, positive electrode left and caudal to heart), with 41 42 mean and standard deviations (SD) calculated from a representative run of three 43 44 -1 45 consecutive cardiac cycles, using paper speeds of 50 mm sec and sensitivities of 46 47 20mV. Heart rate was calculated from the number of R waves in a randomly selected 48 49 30-second period. The mean electrical axis (MEA) was extrapolated from the sum of 50 51 52 the amplitudes of positive and negative deflections in leads 1 and 3 (Lead 1 defined 53 54 as negative electrode is right and cranial to heart, positive electrode is left and 55 56 cranial to heart. Lead 3 is defined as negative electrode left cranial to heart, positive 57 58 electrode left and caudal to the heart).26 59 60

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 8 of 27

1 2 3 Study-specific measurement points were established to facilitate interval and 4 5 6 amplitude calculations. An isoelectric baseline was selected at the segment 7 8 immediately following the T wave to address the persistent deviation from horizontal 9 10 at the PQ and ST segments (Figure 3). As no clear Q or S waves were identifiable 11 12 Confidential: For Review Only on any of the tracings, the QRS was referred to as an R wave, and the QT interval 13 14 15 was measured from the start of the R wave to the end of the T wave and referred to 16 17 as the RT interval (figure 3). 18 19 Data are expressed as mean +/- SD; statistical analysis was performed using 20 21 22 commercial software (Minitab 17, Ltd, Coventry UK). Data were tested for normality 23 24 (Anderson-Darling test), and correlations between length or bodyweight and heart 25 26 rate or MEA were determined by Pearson’s (normal) or Spearman’s rho (not normal) 27 28 29 tests. Differences between genders and bodyweight and length were determined by 30 31 a two-sample t-test. For all tests, a P value < 0.05 was considered significant. 32 33 34 Results 35 36 37 The process of cardiac identification using ultrasound probe and coupling gel was 38 39 simple, accurate and rapid. Subject acceptance of handling, scanning and placement 40 41 of self-adhesive ECG pads was typically excellent, with the exception of a sub-set of 42 43 44 individuals who required additional time to settle before commencing the ECG. High- 45 46 quality ECG tracings were obtained in all subjects. 47 48 A sinus rhythm was recorded in all of the snakes, with R wave preceded by a 49 50 P wave and followed by a T wave. P wave morphology varied slightly between leads, 51 52 53 but was typically low in amplitude, short in duration and monophasic. On lead 2, 54 55 15/29 (52%) study subjects had negative P waves, while in 2/29 (7%) bi-phasic P 56 57 waveforms were present (figure 4). Q and S waves were not identified on any lead 58 59 60 on any of the tracings, therefore ventricular depolarisation consisted universally of

https://mc.manuscriptcentral.com/vetrec Page 9 of 27 Veterinary Record

1 2 3 monophasic R waves, with a high amplitude, short duration and positive deflection 4 5 6 on lead 2 (figure 4). T waves were highly variable between individuals, with 20/29 7 8 (69%) snakes exhibiting a negative waveform on lead 2. Otherwise, T waves were 9 10 monophasic and commonly of moderate to low amplitude (figure 3). 11 12 Confidential: For Review Only There was a gradual downward sloping of the PR segment in 24/29 (83%) 13 14 15 snakes, while the RT segment followed an upward slope in 20/29 (69%) snakes, 16 17 making it difficult to establish the starting point of the T wave. PR and RT intervals 18 19 were proportionally long, while TP segments were short. 11/29 (38%) snakes in this 20 21 22 study demonstrated either pronounced shortening of the TP segment or early 23 24 merging of the T and P waves (figure 5). 25 26 On leads other than lead 2, P, R and T waves could be identified on all 27 28 29 tracings, with the exception of 5 subjects for whom T waves could not be identified 30 31 on lead 1, 2 of which also had no identifiable P waves on lead 1. SV waves were not 32 33 identifiable on any of the tracings. 34 35 Lead 2 ECG amplitude and interval duration values are shown in Table 1. 36 37 38 Parameter Amplitude (mV) Duration (sec) 39 P wave 0.107 +/- 0.045 0.064 +/- 0.016 40 T wave 0.180 +/- 0.127 N/A 41 42 R wave 1.061 +/- 0.395 0.107 +/- 0.019 43 PR interval N/A 0.259 +/- 0.059 44 RT interval N/A 0.585 +/- 0.145 45 46 Table 1. Lead 2 waveform in terms of wave amplitude (mV) and interval duration 47 (sec), expressed as the mean +/- standard deviation (SD). 48 49 Statistics 50 51 There was no difference in bodyweight or length measurements between males and 52 53 54 females (P = 0.634 and P = 0.506, respectively. Figure 6 (A and B)). Heart rate was 55 56 always between 20 and 97 bpm and was not found to be significantly different 57 58 between genders (P = 0.094. Figure 6 (D)) or correlated with weight or length (P = 59 60

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 10 of 27

1 2 3 0.972 and 0.396, respectively. Figure 7). MEA was always positive and displaced to 4 5 6 the left, and was neither correlated with bodyweight nor length (P = 0.910 and P = 7 8 0.771, respectively. Figure 8), or influenced by gender (P = 0.233, T-value = 1.24, 9 10 DF = 15. Figure 6 (C)). 11 12 Confidential: For Review Only The mean and SD for weight, length, heart rate (HR) and MEA, by group, are 13 14 15 presented in table 2. 16 17 Parameter Group (n=29) Female (n=10) Male (n=19) Mean +/- 18 mean and SD Mean +/- SD SD 19 Weight (g) 609 +/- 216 583 +/-189 622 +/- 233 20 21 Length (cm) 114 +/- 24 117 +/- 25 113 +/- 24 22 Heart rate (bpm) 60 +/- 18 53 +/- 17 65 +/- 18 23 24 MEA (degrees) 66 +/- 12 64 +/- 13 70 +/- 11 25 Table 2. Weight, length, HR and MEA expressed as mean +/- SD values for group 26 and gender (male and female) 27 28 29 Discussion 30 31 32 To the authors’ knowledge, this is the first comprehensive descriptive study of ECGs 33 34 from fully conscious and unrestrained corn snakes. We used a modified Einthoven’s 35 36 triangle that was atraumatic and well tolerated to generate high quality recordings 37 38 that were suitable for interpretation. 39 40 41 Despite the absence of a specialised electrical conduction network, ECGs 42 43 from all corn snakes in this study consisted of sequential atrial and ventricular 44 45 depolarisations analogous to the P wave, PQ interval, QRS complex and T waves 46 47 48 observed in mammals. In contrast to mammals, however, and as has been 49 50 previously reported in reptiles,13 14 none of the ECGs had recognisable Q or S waves 51 52 at standard sensitivities, suggesting that PR and RT intervals rather than PQ and QT 53 54 55 intervals is more appropriate terminology in this species. 56 57 We identified several key features of the PRT sequence that are relevant to 58 59 future clinical and research use in this and potentially other snake species. 60

https://mc.manuscriptcentral.com/vetrec Page 11 of 27 Veterinary Record

1 2 3 First, P waves have heterogeneous polarity between individuals and can be positive, 4 5 6 negative or, biphasic (figure 3). It is also noteworthy that P waves were not preceded 7 8 by sinus venosus waves, which have been variably identified in other reptile 9 10 studies.19 22 27 Cardiac mass in reptiles is disproportionately small when compared to 11 12 Confidential: For Review Only mammals or birds, representing only 0.2% to 0.3% of body mass.18 We speculate 13 14 15 that action potentials (AP) from this myocardial structure may be masked by 16 17 preceding T waves and/or background APs from skeletal muscle contraction.14 24 27 18 19 Second, although an AV node has not been identified in snakes, the PR interval, and 20 21 22 hence atrioventricular conduction, is prolonged in a similar way to that mediated by 23 24 the AV node in mammals. It has been suggested that a long PR interval in reptiles 25 26 could reflect slow AP conduction along tortious pathways through the AV canal, slow 27 28 29 rates of cardiac myocyte depolarisation, or an absence of fast-propagating gap 30 31 junctions from the AV canal.28 None of these mechanisms have been investigated in 32 33 the corn snake but do represent possible areas for future research. The PQ interval 34 35 in mammals is under autonomic control, and small changes in such a long PR 36 37 38 interval in snakes could represent an easily measurable marker of autonomic status 39 40 and metabolic health. 41 42 Third, R waves were the tallest complexes of the sequence and always positive on 43 44 45 lead 2, findings which suggest homology to mammals by demonstrating a rapid 46 47 systolic phase and action potential (AP) propagation towards lead 2, respectively. 48 49 This feature also provides an easily identifiable waveform when interpreting corn 50 51 52 snake ECG tracings (figure 3). 53 54 Fourth and finally, the T wave was commonly found to be very wide and of variable 55 56 morphology between corn snakes, as has been reported in other reptile species.21 27 57 58 29 This segment was approximately 2-3 times longer than the QT of mammals, but, 59 60

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 12 of 27

1 2 3 similar to mammals, was around twice the duration of the PR (PQ) interval. High 4 5 6 numbers of snakes demonstrated PR depression followed by a curvilinear elevation 7 8 from the end of the R (J-point) to the T wave, which caused difficulties in establishing 9 10 the isoelectric baseline along traditional mammalian lines. This prompted the 11 12 Confidential: For Review Only creation of a new baseline at the early part of the TP segment (figure 4) which was 13 14 15 consistently more level. In human cardiology, PQ elevation and ST depression, 16 17 either separately or concurrently, are clinically significant and associated with 18 19 myocardial infarction and myocarditis.30 31 Given the high number of subjects in our 20 21 22 study demonstrating this feature and the general good health of the cohort, these 23 24 segmental fluctuations are unlikely to have the same pathological importance in corn 25 26 snakes, although we could speculate that PR elevation and ST depression indicate a 27 28 29 normal degree of myocardial ischaemia in the snake heart. The embryonic hearts of 30 31 birds and mammals demonstrate a great inherent tolerance to ischemia32 and as the 32 33 hearts of adult reptiles are structurally very similar it is possible that this is a shared 34 35 characteristic. An alternative explanation could be mechanical and related to the 36 37 38 relatively unique anatomy of the cardiac region in snakes. The attachments of the 39 40 ophidian heart in the cranial coelom are loose when compared to their mammalian 41 42 counterparts and allow for the passage of whole prey items. They could permit a 43 44 45 rhythmic swinging of the heart through the contraction cycle, leading to a regular 46 47 oscillation of the ECG baseline. This could be investigated further by recording an 48 49 ECG, using the arrangement of ECG pads herein described, while simultaneously 50 51 52 performing echocardiography. 53 54 The heart rates we obtained are in general agreement with those reported in 55 56 other ophidian studies and are not dissimilar from other representatives of the class 57 58 reptilia (Tables 3 and 4). 59 60

https://mc.manuscriptcentral.com/vetrec Page 13 of 27 Veterinary Record

1 2 3 4 5 6 7 8 Study Heart rate range (bpm) 9 Clark and Marx, 1960 18 9 -72 10 Mullen, 1967 19 22 – 136 11 20 12 ValentinuzziConfidential: et al. 1969 For13 Review – 62 Only 33 13 Anderson et al. 1999 57 - 73 14 Lewis et al, unpublished data 42 - 78 15 Table 3. Snake heart rate ranges by study. 16 17 18 19 Study Species Heart rate range (bpm) 20 Heaton-Jones and King, American alligator 33 – 42 21 1994 34 22 Holtz and Holtz, 1995 35 Red-eared slider 16 – 36 23 Martinez-Silvestre et al. Gomeran giant lizard 35 – 60 24 2003 36 25 37 26 Tan et al. 2013 Marbled water monitor 26 - 58 27 Hunt, 2015 27 Central bearded dragon 24 – 170 28 Table 4. Reptile heart rates – study, species and range. 29 30 31 They are however much slower than mammals and reflect the approximately 32 33 13 28 34 10-fold lower metabolic rate seen in reptiles. The heart rate of snakes can be 35 36 affected by a wide range of intrinsic and extrinsic factors, but in general it is 37 38 increased with elevated body temperature, small body size, external stimuli (such as 39 40 16 24 38-41 41 handling), recent meal ingestion and gravidity. Previous studies involving 42 43 ECG in snakes have been highly variable in design and have frequently included 44 45 data from small cohort numbers, or large numbers of species restrained by different 46 47 18-22 48 physical and chemical methods. Our data, obtained from only one species, came 49 50 from a larger study cohort and our protocol minimised confounders. Nonetheless, our 51 52 data resemble closely those obtained where confounding factors have not been so 53 54 rigorously controlled,19 42 suggesting that they may have less influence on heart rate 55 56 57 and cardiac conductivity than previously supposed. 58 59 60

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 14 of 27

1 2 3 Typically in mammals, as heart rate increases the period of time between the 4 5 26 6 T wave and the P wave of the next cardiac cycle shortens. In the study here 7 8 reported, shortening of the TP interval was also observed in snakes with heart rates 9 10 at the upper end of the population range. In some cases this shortening was so 11 12 Confidential: For Review Only extreme that there was merging of T and P waves. Germer et al. (2015)43 mention T 13 14 15 and P wave merging in a group of geckos, while Mullen (1967)19 and Jacob and 16 17 McDonald (1975)29 describe wave overlap at the TP interval or “masking” of the P 18 19 wave by the T wave in cohort sub-groups of wild caught snakes. These authors do 20 21 18 22 not speculate on the possible causes of this finding, but Clarke and Marx (1960) 23 24 identify handling stress as an important cause of increased heart rates in wild caught 25 26 snakes. In our study, while all possible efforts were made to reduce stress at the 27 28 29 time of ECG measurement and although all individuals were captive bred pet 30 31 animals, it was clear that some snakes were initially wary of interference and took 32 33 greater time to settle. This group of snakes tended to have higher heart rates and 34 35 more pronounced TP shortening. We conclude that corn snakes have an inherent 36 37 38 variability to handling tolerance but also, and more importantly, that marked 39 40 shortening of the TP segment or TP merging may be an additional useful, 41 42 quantifiable indicator of stress not only for captive corn snakes but for snakes in 43 44 45 general, a collection of animals for which behavioural indicators of stress can be very 46 47 vague. The physiological advantage of this response is not clear. Atrial 48 49 depolarisation so close to ventricular repolarisation in snakes could significantly 50 51 52 decrease time available for passive ventricular filling (which predominates in 53 54 mammals), even at heart rates considered low in mammals, increasing reliance on 55 56 the atrial contribution to ventricular diastole. If confirmed by, for example spectral 57 58 and tissue Doppler echocardiography, it would suggest a profound difference in the 59 60

https://mc.manuscriptcentral.com/vetrec Page 15 of 27 Veterinary Record

1 2 3 mechanics of diastole in corn snakes compared to those in turtles and pythons, for 4 5 28 6 whom ventricular suction is thought to play an important role. 7 8 We also investigated whether gender might have an influence on HR. 9 10 Follicular genesis can occur in female snakes independent of the presence males 11 12 Confidential: For Review Only and sporadically throughout a captive snake’s life. It would be reasonable to expect 13 14 15 that the elevated metabolic rate associated with this condition would lead to an 16 17 increase in HR. We did not attempt to determine the level of ovarian activity in 18 19 individual females through an additional ultrasound examination due to concern of 20 21 22 increased confounding stress effects. However, we did not identify a significant 23 24 increase in HR in our female corn snake population when compared to males. Other 25 26 factors such as weight and snout-to-vent length were also found to have no 27 28 29 significant relationship with HR. Therefore, we believe that the range of HR we 30 31 recorded in both male and female corn snakes are representative of the general corn 32 33 snake population. 34 35 The mean electrical axis (MEA) in reptiles has historically been regarded as very 36 37 38 challenging (if not impossible) to determine and of limited clinical value.14 By 39 40 contrast, in the study here reported identification and measurement of R waves on 41 42 leads 1 and 3 was readily performed and the MEA was easily calculated. 43 44 45 Importantly, the technique we describe produces a MEA through electrode 46 47 placement and draws similarities to companion mammals, rather than MEA as 48 49 defined by cardiac alignment and physical measurements. Yielded values 50 51 52 demonstrated a left axis dominance (table 2), indicating that it is the left side of the 53 54 common ventricle in corn snakes that makes up most of the myocardium. This axial 55 56 dominance is similar to mammals, where the left ventricle is larger than the right, but 57 58 is also similar to the few other reptile species for which a MEA has been calculated 59 60

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 16 of 27

1 2 3 26 36 34, indicating a common asymmetrical distribution of ventricular muscle mass in 4 5 6 reptiles. We also found that MEA values produced in this study were highly 7 8 conserved between individuals, suggesting a predictable direction of electrical flow 9 10 through the myocardium and therefore a broadly uniform positioning of the heart 11 12 Confidential: For Review Only within the coelom. We suggest that the ECG in corn snakes generates a vector of a 13 14 15 MEA that is quantifiable and has the potential to aid diagnosis of cardiac disease in 16 17 snakes, similar to its use in mammals. Further, that the technique described herein 18 19 will permit the calculation of MEA to the benefit of future ophidian studies. 20 21 22 23 24 Limitations 25 26 There were a number of limitations to this study. First, the exact ages of the snakes 27 28 29 involved was not always known. Size (in terms of snout-to-vent length and weight), 30 31 which was used to classify individuals as adult, may be inaccurate as growth can be 32 33 influenced by a number of factors including gender, long-term nutrition and captive 34 35 environment. However, known adult animals were present in our study population 36 37 38 and no difference in body weight or length measurements was identified between 39 40 male and female snakes. It is therefore reasonable to consider the cohort as 41 42 homogeneous in terms of size and, by extension, adult age. Second, despite a 43 44 45 comprehensive clinical examination, the absence of cardiac or non-cardiac disease 46 47 could not be excluded from our study group. Inclusion of routine blood sampling and 48 49 comprehensive echocardiography would have helped to identify systemic disease 50 51 52 and structural or functional cardiac disease, respectively. These investigations, 53 54 however, fell outside the scope of the study hypothesis and may have introduced a 55 56 significant confounding stress effect on our data. Furthermore, the limits of ethical 57 58 approval precluded invasive procedures (such as blood sampling), or undue stress 59 60

https://mc.manuscriptcentral.com/vetrec Page 17 of 27 Veterinary Record

1 2 3 through prolonged or repeated handling. Third, the cohort size was insufficient to 4 5 6 provide reference ranges for this species. However, to the authors’ knowledge, this 7 8 is the largest single-species ophidian study that has been published. Finally, there 9 10 were twice as many males as females in this study which could have introduced a 11 12 Confidential: For Review Only gender bias that was not identifiable on the variables we measured. However, 13 14 15 excluding females from analysis would have reduced the statistical power of the 16 17 study. 18 19 20 21 22 Conclusion 23 24 As hypothesised, ECG waveforms in corn snakes are morphologically recognisable 25 26 and can be characterised and interpreted in a similar way to ECGs obtained from 27 28 29 mammals. However, there are important differences such as heart rate and, 30 31 particularly, TP intervals that may be useful in identifying corn snakes experiencing 32 33 stress. The data we present could help provide reference ranges for ECG 34 35 parameters in this species, aiding the clinical application of ECGs in general and 36 37 38 specialist practice. 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 18 of 27

1 2 3 References 4 5 6 7 1. Mitchell M. Reptile cardiology. Veterinary Clinics of North America - Exotic 8 Animal Practice 2009; 12:65-79 9 2. Bogan JE. Ophidian Cardiology - A Review. Journal of herpetological 10 medicine and surgery 2017;27(1-2):62-77 11 3. Jacobson ER, Seely JC, Novilla MN et al. Heart failure associated with 12 Confidential: For Review Only 13 unusual hepatic inclusions in a Deckert's rat snake. Journal of Wildlife 14 Diseases 1979; 15:75-81 15 4. Jacobson ER, Homer B, Adams W. Endocarditis and congestive heart failure 16 in a Burmese python (Python molurus bivittatus). Journal of Zoo and Wildlife 17 Medicine 1991; 22:245-248 18 5. Catao-Dias JL, Nichols DK. Neoplasia in snakes at the national zoological 19 park, Washington, DC (1978-1997). Journal of Comparative Pathology 1999; 20 21 120:89-95 22 6. Rishniw M, Carmel BP. Atrioventricular valvular insufficiency and congestive 23 heart failure in a carpet python. Australian Veterinary Journal 1999; 77(9); 24 580-583 25 7. Jensen B, Wang T. Hemodynamic consequences of cardiac malformations in 26 two juvenile ball pythons (Python regius). Journal of Zoo and Wildlife Medicine 27 2009; 40:752-756 28 29 8. Schilliger L, Trehiou-Sechi E, Petit AMP, et al. Double valvular insufficiency in 30 a Burmese python (Python molurus bivittatus, Linnaeus, 1758) suffering from 31 concomitant bacterial pneumonia. Journal of Zoo and Wildlife Medicine 2010; 32 41:742-744 33 9. Schroff S, Schmidt V Kiefer I, Krautwald-Junghanns M, Pees M. 34 Ultrasonographic Diagnosis of an Endocarditis Valvularis in a Burmese 35 36 Python (Python molurus bivittatus) with Pneumonia Journal of Zoo and 37 Wildlife Medicine 2010; 41(4):721-724 38 10.Stumpel JBG, del-Pozo J, French A et al. Cardiac haemangioma in a corn 39 snake (Pantherophis guttatus). Journal of Zoo and Wildlife Medicine 2012; 40 43:360-366 41 11.Konell A, Sanson B, Giannico AT, et al. Cardiac thrombus in a Burmese 42 python (Python morlurus bivitattus). The Herpetological Bulletin 2015; 131:13- 43 44 16 45 12.Schilliger L, Chetboul V, Damoiseaux C, et al. Restrictive cardiomyopathy and 46 secondary congestive heart failure in a McDowell’s carpet python (Morelia 47 spilota McDowelli). Journal of Zoo and Wildlife Medicine 2016; 47(4):1101- 48 1104 49 13.Hynes B, Girling SJ. Cardiovascular and haemopoietic systems. In: Girling SJ, 50 Raiti P, eds. BSAVA Manual of Reptiles, third edition. BSAVA, 2019:323-341 51 52 14.Schilliger L, Girling SJ. Cardiology. In: Divers SJ, Stahl, eds. Mader’s Reptile 53 and Amphibian Medicine and Surgery, Third edition. Elsevier, 2019:669-698 54 15.Farrell AP, Gamperl AK, Francis ETB. Comparative aspects of heart 55 morphology. In: Biology of the Reptilia, Vol.19, Morphology G: Visceral 56 Organs. Eds. Gans C, Gaunt AS. Thomas-Shore, Inc., Dexter, Michigan, 57 USA, 1998:375-419 58 59 16.Wyneken J. Normal reptile heart morphology and function. Veterinary Clinics 60 of North America: Exotic Animal Practice 2009; 12:51-63

https://mc.manuscriptcentral.com/vetrec Page 19 of 27 Veterinary Record

1 2 3 17.Hedberg A, Nilsson S. Vago-sympathetic innervation of the heart of the puff 4 adder, Bitis arietans. Comparative Biochemistry of Physiology Part C 1976; 5 6 53(1):3-8 7 18.Clarke GK, Marx TI. Heart rates of unanesthetized snakes by 8 electrocardiography. American Society of Ichthyologists and 9 Herpetologists1960; 3:236-238 10 19.Mullen RK. Comparative electrocardiography of the . Physiological 11 Zoology 1967; 40(2):114-126 12 Confidential: For Review Only 20.Valentinuzzi ME, Hoff HE, Geddes LA. Observations on the electrical activity 13 14 of the snake heart. Journal of Electrocardiology 1969; 2(1):39-50 15 21.Valentinuzzi ME, Hoff HE, Geddes LA. Electrocardiogram of the snake: 16 Effects of the location of the electrodes and cardiac vectors. Journal of 17 Electrocardiology 1969; 2(3):245-252 18 22.Valentinuzzi ME, Hoff HE, Geddes LA. Electrocardiogram of the snake: 19 Intervals and durations. Journal of Electrocardiology 1969; 2(4):343-352 20 21 23.Andersen JB, Rourke BC, Caiozzo VJ, Bennett AF, Hicks JW. Postprandial 22 cardiac hypertrophy in pythons. Nature 2005; 434:37-38 23 24.Enok S, Simonsen LS, Wang T. The contribution of gastric digestion and 24 ingestion of amino acids on postprandial rise in oxygen consumption, heart 25 rate and growth of visceral organs in pythons. Comparative Biochemistry and 26 Physiology, Part A 2013; 165:46-53 27 25.Scheelings TF. Anatomy and physiology In: Girling SJ, Raiti P, eds. BSAVA 28 29 Manual of Reptiles, third edition. BSAVA, 2019:1-25 30 26.Tilly LP. The approach to the electrocardiogram. In: Essentials of canine and 31 feline electrocardiology interpretation and treatment. Third edition. Lea & 32 Febiger, Malvern, USA. 1992:50-55 33 27.Hunt C. Electrocardiography of the Normal Inland Bearded Dragon (Pogona 34 vitticepts). Royal College of Veterinary Surgeons Diploma in Zoological 35 Medicine 2013, DZM 36 37 28.Jensen B, Moorman AFM, Wang T. Structure and function of the hearts of 38 lizards and snakes. Biological Reviews 2014; 89:302-336 39 29.Jacob JS, McDonald HS. Temperature preferences and electrocardiography 40 of Elaphe obsolete (serpents). Comparative Biochemistry and Physiology 41 1975; 52a:591-594 42 30.Porela P, Kyto V, Nikus K, et al. PR depression is useful in the differential 43 44 diagnosis of myopericarditis and ST elevation myocardial infarction. Annals of 45 Noninvasive Electrocardiology 2012:17; 141-145 46 31.de Bliek EC. ST elevation: Differential diagnosis and caveats. A 47 comprehensive review to help distinguish ST elevation myocardial infarction 48 from nonischemic etiologies of ST elevation. Turkish Journal of Emergency 49 50 Medicine 2018; 18:1-10 51 32.Jensen B, Boukens BJD, Postma AV, et al. Identifying the evolutionary 52 building blocks of the cardiac conduction system. PLOS ONE 2012; 7(9):1-13 53 33.Anderson NL, Wack RF, Calloway L, Hetherington TE, Williams JB. 54 Cardiopulmonary effects and efficacy of propofol as an anesthetic agent in 55 brown tree snakes, Boiga irregularis. Bulletin of the Association of Reptilian 56 57 and Amphibian Veterinarians 1999; 9 (2):9-15 58 34.Heaton-Jones TG, King RR. Characterisation of the electrocardiogram of the 59 American alligator (Alligator mississippiensis). Journal of Zoo and Wildlife 60 Medicine 1994; 25(1):40-47

https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 20 of 27

1 2 3 35.Holz RM, Holz P. Electrocardiography in anaesthetised red-eared sliders 4 (Trachemys scripta elegans). Research in Veterinary Science 1995; 58(1):67- 5 6 69 7 36.Martinez-Silvestre A, Mateo JA, Pether J. Electrocardiographic parameters in 8 the Gomeran giant lizard, Gallotia bravoana. Journal of herpetological 9 medicine and surgery 2003; 13(3):22-25 10 37.Tan YFS, Gicana KRB, Lastica EA. Electrocardiographic profile of captive 11 marbled water (Varanus marmoratus, Weigmann, 1834). 12 Confidential: For Review Only Philippine Journal of Veterinary and Animal Sciences 2013; 39(2):269-276 13 14 38.Birchard GF, Black CP, Schuett GW, Black V. Influence of pregnancy on 15 oxygen consumption, heart rate and hematology in the garter snake: 16 implication for the “cost of reproduction” in live bearing reptiles. Comparative 17 Biochemistry and Physiology 1984; 77A (3):519-523 18 39.Seymore RS. Scaling of cardiovascular physiology in snakes. American 19 Zoologist 1987; 27:97-109 20 21 40.Lillywhite HB, Zippel KC, Farrell AP. Resting and maximal rates in ectothermic 22 vertebrates. Comparative Biochemistry and Physiology Part A 1999; 124:369- 23 382 24 41.Rodrigues JFM, Braga RdeR, Ferreira THA, et al. What makes the heart of 25 Boa constrictor (Squamata: Boidae) beat faster? Zoologia 2015; 32(1):83-85 26 42.Jacobson ER, Whitford WG. Physiological responses to temperature in the 27 Patch-Nosed Snake, Salvadora hexalepis. Herpetologica1971; 27(3):289-295 28 29 43.Germer CM, Tomaz JM, Carvalho AF, et al. Electrocardiogram, heart 30 movement and heart rate in the awake gecko (Hemidactylus mabouia). 31 Journal of Comparative Physiology B 2015; 185:111-118 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

https://mc.manuscriptcentral.com/vetrec Page 21 of 27 Veterinary Record

1 2 A 3 4 5 6 7 8 Confidential: For Review Only 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 B 33 34 35 36 37 38 39 https://mc.manuscriptcentral.com/vetrec 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Veterinary Record Page 22 of 27

1 B 2 A 3 4 5 6 7 8 Confidential: For Review Only 9 10 11 12 13 14 15 16 17 18 19 20 21 I 22 23 24 25 III 26 27 28 29 30 II 31 32 33 34 35 36 37 38 39 https://mc.manuscriptcentral.com/vetrec 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 23 of 27 Veterinary Record

1 2 3 4 5 6 RT segment 7 A C 8 Confidential: For Review Only 9 10 11 12 T 13 14 15 16 17 18 B D 19 20 21 22 Baseline 23 T 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 https://mc.manuscriptcentral.com/vetrec 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Veterinary Record Page 24 of 27

1 2 3 4 5 6 7 A R 8 Confidential: For Review Only 9 10 11 12 13 P 14 15 16 17 18 B R 19 20 21 22 23 P 24 25 26 27 28 R 29 C 30 31 32 33 34 P 35 36 37 38 39 https://mc.manuscriptcentral.com/vetrec 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 25 of 27 Veterinary Record

1 2 3 A 4 5 6 7 8 Confidential:P For Review Only 9 T 10 11 12 13 14 15 B 16 17 18 19 20 T P 21 22 23 24 25 C 26 27 28 29 P 30 T 31 32 33 34 35 36 37 38 39 https://mc.manuscriptcentral.com/vetrec 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Veterinary Record Page 26 of 27

1 2 3 4 5 6 7 8 9 10 11 Confidential: For Review Only 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Figure 6. Bar charts comparing male (black bars) and female (grey bars) corn snakes in terms of mean body 40 length (A), body weight (B), MEA (C) and HR (D). 41 42 208x210mm (300 x 300 DPI) 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/vetrec Page 27 of 27 Veterinary Record

1 2 3 4 5 6 7 8 9 10 11 Confidential: For Review Only 12 13 14 15 16 17 18 19 20 Figure 7. Scatter plots for heart rate against body weight (A), and body length (B), indicating no correlation 21 between variables. 22 23 195x78mm (300 x 300 DPI) 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/vetrec Veterinary Record Page 28 of 27

1 2 3 4 5 6 7 8 9 10 11 Confidential: For Review Only 12 13 14 15 16 17 18 19 20 Figure 8. Scatter plots for MEA against body weight (A), and body length (B), indicating no correlation 21 between variables. 22 196x78mm (300 x 300 DPI) 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/vetrec