EVALUATION OF FEEDING BEHAVIOR AS AN

INDICATOR OF PAIN IN SNAKES

Authors: James, Lauren E., Williams, Catherine J. A., Bertelsen, Mads
F. , and Wang, Tobias
Source: Journal of Zoo and Wildlife Medicine, 48(1) : 196-199
Published By: American Association of Zoo Veterinarians
URL: https://doi.org/10.1638/2016-0064.1

BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access titles
in the biological, ecological, and environmental sciences published by nonprofit societies, associations,
museums, institutions, and presses.

Your use of this PDF, the BioOne Complete website, and all posted and associated content indicates your
acceptance of BioOne’s Terms of Use, available at www.bioone.org/terms-of-use.

Usage of BioOne Complete content is strictly limited to personal, educational, and non - commercial use.
Commercial inquiries or rights and permissions requests should be directed to the individual publisher as
copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit
publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to
critical research.

Downloaded From: https://bioone.org/journals/Journal-of-Zoo-and-Wildlife-Medicine on 03 Dec 2023

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidade Estadual Paulista (UNESP)
 Journal of Zoo and Wildlife Medicine 48(1): 196–199, 2017
Copyright 2017 by American Association of Zoo Veterinarians

EVALUATION OF FEEDING BEHAVIOR AS AN INDICATOR OF

PAIN IN SNAKES

Lauren E. James, B.Sc. (Hons), Catherine J. A. Williams, M.A., M.Sc., Vet.M.B., Mads F. Bertelsen,
D.V.M., D.V.Sc., Dipl. A.C.Z.M., Dipl. E.C.Z.M. (Z.H.M.), and Tobias Wang, Ph.D.

Abstract: The necessity to prevent and manage pain in reptiles is becoming increasingly important, as their use
in scientific research and popularity as exotic pets continues to rise. It was hypothesized that feeding behavior
would provide an adequate indicator of pain perception in the ball python (Python regius). Normal feeding was
defined the previous week, where a dead rodent was struck within 12 sec (n ¼ 10). Eighteen pythons were randomly
assigned to one of three treatments: anesthesia only (AO), chemical noxious stimulus (CS; capsaicin injection), or
surgical noxious stimulus (SS; surgical incision). The time to strike was recorded 4 hr after the procedure and
weekly during the subsequent 3 wk. Delayed feeding was observed in animals in the CS and SS groups, and normal
feeding resumed after 1 and 3 wk, respectively. Spontaneous feeding remained uninterrupted for the AO group.
These findings demonstrate feeding behavior as a potential model to assess pain in snakes.
Key words: Ball python (Python regius), behavior, feeding, nociception, pain, reptile.

BRIEF COMMUNICATION cessation in normal behavior and an indicator of

It is generally acknowledged that reptiles, akin
to other vertebrates, possess the appropriate
neural mechanisms to perceive and respond to
noxious stimuli.9 Nevertheless, identifying an
objective method to measure pain in any nonvocal
animal, including snakes and other reptiles,
remains challenging. Pain is defined as a sensory
and emotional experience attributed to an actual
or potentially damaging stimulus,8 whereas noci-
ception merely describes the detection of the
stimulus.1 To provide adequate analgesia, the
level of pain an animal experiences must be
accurately determined, including the emotional
aspect. This cannot be achieved through the sole
reliance on physiologic monitoring, as the dis-
tinction between nociception and perceived pain
cannot be made. Behavioral assessments facilitate
the consideration of emotional experience and
may be used as an adjunct to physiologic mea-
sures. Interrupted feeding behavior has been
observed in red-eared slider turtles (Trachemys
scripta elegans)3 following surgery, reflecting a
From the Section for Zoophysiology, Department of
Bioscience, Aarhus University Building 1131, Universitet-
sparken, 8000 Aarhus C, Denmark (James [present
address], Williams, Wang), the Faculty of Life Sciences,
University of Manchester, Oxford Road, Manchester, M13
6PT, United Kingdom (James), the Centre for Zoo and
Wild Animal Health, Copenhagen Zoo, Roskildevej 32,
2000 Frederiksberg C, Denmark (Bertelsen). Correspon-
dence should be directed to Lauren E. James
(ljames0910@gmail.com).
pain or discomfort.8
Hence, this experiment aimed to investigate
whether resumption of spontaneous feeding be-
havior following a potentially painful stimulus
provides a reliable measure of pain perception in
snakes. Ball pythons (Python regius), popular pets
that are frequently used in physiologic research,
were used in this experiment with two noxious
stimuli: capsaicin injection, a chemical stimulus
(CS) and an incision, a surgical stimulus (SS).
Capsaicin, the active ingredient in chili peppers,
creates a burning sensation via depolarization of
nociceptive C fibers4 and is a well-known model of
nociception, with potential for analgesic efficacy
testing.2,10 Surgical stimuli that emulate more
clinically relevant situations have also been used
to study persistent pain, where pre- or postoper-
ative analgesia is required.3
The experiment was approved by the Danish
Animal Experiments Inspectorate (permit num-
ber: 2013-15-2934-00847). The mice used for
feeding were laboratory mice bred for feeding
stock at Aarhus University, Denmark. First, a
fixed feeding protocol was developed, where a
freshly killed mouse was held by the tail tip in
forceps approximately 2 cm from the head of the
snake. The mouse was alternatingly held station-
ary for 30 sec and then twitched from side to side
for 30 sec. The time to strike was recorded. If no
strike occurred within 3 min, the snake was
considered unwilling to feed voluntarily. Only
snakes consistently feeding within 30 sec for 3
consecutive weeks were included in the study.
Normal feeding was defined as the mean þ 2 SD of

196

Downloaded From: https://bioone.org/journals/Journal-of-Zoo-and-Wildlife-Medicine on 03 Dec 2023

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidade Estadual Paulista (UNESP)
 JAMES ET AL.—BEHAVIORAL INDICATORS OF PAIN IN SNAKES
times for 10 of the pythons 1 wk prior to
experimentation. Eighteen pythons were random-
ly assigned to one of three different treatments (n
¼ 6 per group): anesthesia only (AO), CS, and SS.
All snakes were fasted for 7 days prior to
experimentation.
Anesthesia was induced in an air-tight bag
containing isoflurane-soaked gauze (IsoFlo Vet
100%, Abbot Laboratories Ltd., Berkshire SL6
4XE, United Kingdom) until righting reflexes and
muscle tone subsided, enabling endotracheal
intubation. Anesthesia was maintained at 3%
isoflurane (97% oxygen) with mechanical ventila-
tion (250 ml/min per kg) by using an Anesthesia
Workstation (Halowell EMC, Pitsfield, Massa-
chusetts 01201, USA) ventilator and Fluotec 3
Vaporizer (Simonsen & Weel, 2620 Albertslund,
Denmark). During surgery, cloacal temperature
was monitored by using a rectal thermometer and
maintained at 308C by a heat mat (Melissa, Adexi,
8240 Risskov, Denmark). Once a surgical plane of
anesthesia was reached, one of the following three
treatments was administered:
AO: Animals were maintained under isoflur-
ane anesthesia for 30 min.
CS: While under anesthesia, a polyethylene
catheter (50 cm long and 0.58 mm inner
diameter [Smiths Medical International Ltd.
Kent, CT21 6JL, United Kingdom]) filled with
saline (Baxter ViaFlow, Baxter, 3450 Allerød,
Denmark; 9 mg/ml) was introduced 20 cm
cranial to the tail tip, tunneled 1 cm subcutane-
ously, and anchored with two skin sutures.
Thirty minutes after anesthetic recovery, snakes
received capsaicin (M2028 Capsaicin, Sigma-
Aldrich Chemie GmbH; 2605 Brøndby, Den-
mark; 3 mg in 0.2 ml) dissolved in 7% Tween 80
(P1754-500; Sigma-Aldrich Chemie GmbH)
and saline, remotely via the catheter. Immediate
reactions were observed for the following 30
min before the catheter was removed and
snakes returned to their terraria.
SS: A longitudinal ventrolateral incision (3
cm) was made through the skin, 20 cm cranial to
the tail tip, followed by blunt dissection through
the muscle and connective tissue to expose the
ribs, emulating exposure of the vertebral artery
for catheterization.7 The incision was sutured in
an everting horizontal mattress pattern (3-0
Prolene; Blue Ethilon II, Ethicon, Livingstone
EH54 0AB, Scotland).
Anesthetic duration was fixed at 30 min for all
groups, and snakes were allowed to recover on
mechanical ventilation until voluntary head with-
Downloaded From: https://bioone.org/journals/Journal-of-Zoo-and-Wildlife-Medicine on 03 Dec 2023
Terms of Use: https://bioone.org/terms-of-use Access provided by Universidade Estadual Paulista (UNESP)
197
drawal. Following the return of muscle tone,
spontaneous movements and head support, they
were returned to their terraria. The feeding
response was evaluated according to the protocol
mentioned previously, 4 hr after treatments, and
subsequently once per week for the following 3
wk. If the snake was unwilling to feed, a second
attempt was made 24 hr later.
The feeding times were scored based on the
extent to which normal feeding (mean þ 2 SD
baseline feeding) was delayed. These scores
represented whether the pythons were feeding
normally (0: normal feeding), with a slight delay
(1: normal feeding , x , 30 sec), required
encouragement (2: 30 sec , x , 180 sec), or
refused the meal and ate 24 hr later (3).
The feeding delay scores were used for statisti-
cal analysis, so data were assumed to not be
normally distributed. The effect of treatment on
feeding behavior over time was analyzed by using
a cumulative link mixed model, with individual
snake defined as a random effect and weeks
postsurgery and treatment as fixed effects. A
pairwise analysis with Tukey adjustment was then
conducted as a post hoc test for significant group
differences. Analyses were performed by using
RStudio (v.3.0.2, ordinal [v2015.1-21] for cumu-
lative link mixed model and lsmeans [v2.16] for
pairwise analysis. Data were considered signifi-
cant when P , 0.05. As the time data were not
normally distributed (Shapiro–Wilk test, w ¼
0.586, P , 0.001), they are presented as median
(interquartile range).
Baseline feeding times from 10 of the pythons
ranged from 1 to 12 sec, with a mean of 4.6 sec;
therefore, normal feeding was determined as 12
sec (mean þ 2 SD, rounded to the nearest whole
second). The median times to feed for pythons in
each treatment group are presented in Table 1,
while Figure 1 displays the median feeding delay
score for each group over the 4-wk observation
period. A significant effect of week (P , 0.0001)
and treatment (P ¼ 0.0096) on feeding delay score
was found. Post hoc analysis revealed a significant
difference between the AO and the SS groups (P ¼
0.01), and the time point week 0 was different
from all other time points (P ¼ 0.019, 0.0096, and
0.0009 for weeks 1, 2, and 3, respectively).
The AO group displayed consistently normal
feeding, despite a significant difference between
weeks 0 and 3 for this group (P ¼ 0.0082). On
occasion, a snake exhibited a slight delay in the
time to feed, but encouragement by twitching the
mouse was never required, as feeding times
remained lower than 30 sec (Table 1); thus, this
 198 JOURNAL OF ZOO AND WILDLIFE MEDICINE

Table 1. Time to feed for ball pythons in three treatment groups: anesthesia only (AO), chemical stimulus
(CS), and surgical stimulus (SS).a

AO (sec), n ¼ 6 CS (sec), n ¼ 6 SS (sec), n ¼ 6

Week 0 4 hr (6) ¼ 8.5 (2–19.5) 4 hr (4) ¼ 26.5 (17–33.8) 4 hr (3) ¼ 33 (29.5–34)

24 hr ¼ not applicable 24 hr (2) ¼ 23 (18.5–27.5) 24 hr (2)b ¼ 149.5 (135.8–163.3)
Week 1 1.5 (1–8.8) 4 (2.5–6.3) 33c (8–103)
Week 2 2 (1.3–5) 10 (5–23) 20.5 (7–34)
Week 3 2.5 (1–5.5) 1.5 (1–2) 1 (1–6)
a
Numbers in parentheses following 4- and 24-hr time points indicate the number of snakes feeding at this time. All data are
presented as medians (interquartile range).
b
One snake refused the meal at both the 4- and 24-hr feeds.
c
One snake refused the meal at this time point. Snakes refusing meals at 24-hr feeds were not included in the analysis.

result was not regarded as a biologically signifi- shedding or residual discomfort arising from
cant difference. CS animals showed an immediate temporary instrumentation with a subcutaneous
reaction to capsaicin injection, persisting for 8–12 catheter. Thus, the effect of chemical stimulation
min and consisting of vigorous caudocranial
cannot be separated from that elicited by the
convulsions, increased activity, and elevation of
catheter. In this case, the catheter represents the
the site of capsaicin instillation. Feeding delay
scores increased on the day of the procedure, inevitable skin intervention required to adminis-
returning on average to normal 1 wk later. All ter capsaicin; thus, the effect seen here represents
animals fed normally by week 3. No obvious signs the effect of the capsaicin model of nociception,
of discomfort were observed in SS-treated py- as described for P. regius previously.10
thons, such as elevation of the wound site;
however, all animals displayed a change in feeding
behavior (Fig. 1). There was a gradual, essentially
linear, return to low feeding delay scores over the
3 wk. At that time, only one animal’s feeding
remained interrupted.
Minimal alteration to feeding behavior was
recorded in animals only anesthetized; therefore,
the stress associated with handling and anesthesia
is not sufficient to alter this behavior. It can also
be surmised that the changes in feeding behavior
reported in the two groups administered a nox-
ious stimulus may be attributed to pain or
discomfort. However, note for future work that a
change in individual baseline feeding time may be
a more intuitive measure of feeding behavior than
a score based on generalized baseline feeding
Figure 1. Feeding delay scores for ball pythons
times.
(Python regius) in three treatment groups: anesthesia
Findings for the CS group corroborate previous only (AO), chemical stimulus (capsaicin injection), and
work reporting tachycardia in response to capsa- surgical stimulus (SS; surgical incision). A score of 0
icin9 and supports its use as a model of acute indicated normal feeding (12 sec), 1 represented a
nociception in snakes. However, the physiologic slight delay in feeding (normal feeding , x , 30 sec), 2
response to capsaicin appears to subside within indicated the python required encouragement to feed
30 min,10 yet feeding behavior remains interrupt- (30 , x , 180 sec) and 3 that snakes refused a meal but
ed after 4 hr. This emphasizes the necessity to fed 24 hr later. The box plot displays the median and
interquartile range for the data. A significant effect of
integrate physiologic and behavioral measures to
week (P , 0.0001) and treatment (P ¼ 0.0096) on
monitor pain perception and provide a compre-
feeding delay score was found. Post hoc analysis
hensive assessment of the duration of capsaicin’s revealed significant differences between the AO and
effect on pythons. There was also an apparent SS groups (P ¼ 0.01) and between week 0 and all other
increase in feeding time around week 2, which time points (P ¼ 0.019, 0.0096, and 0.0009 for weeks 1,
may be a random effect attributable to early 2, and 3, respectively).

Downloaded From: https://bioone.org/journals/Journal-of-Zoo-and-Wildlife-Medicine on 03 Dec 2023

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidade Estadual Paulista (UNESP)
 JAMES ET AL.—BEHAVIORAL INDICATORS OF PAIN IN SNAKES
The delay in return to spontaneous feeding
following a surgical procedure agrees with find-
ings in chelonians,3 further supporting the use of
this model for longer-term reptile pain assess-
ment. Postsurgical tachycardia has also been
reported in Python regius,5 and evidence here
supports the notion that they experience a detect-
able level of discomfort following surgery. Future
work should focus on the use of this model to test
the efficacy of analgesia and determine the effect
of analgesics themselves on feeding behavior, as
morphine has decreased appetite in red-eared
sliders3 and rats (Rattus norvegicus).6
The model presented here shows potential for
use in analgesic efficacy testing to supplement
other assessment methods, such as change in
heart rate and limb withdrawal. The integration
of physiologic and behavioral monitoring could
then be achieved and improve the reliability of
research conducted in this field.
Acknowledgments: The authors thank Heidi
Meldgaard Jensen and team for exceptional
animal care, Rasmus Buchanan for technical
assistance, and the advice of anonymous review-
ers for the development of the manuscript.
LITERATURE CITED
1. International Association for the Study of Pain
Task Force on Taxonomy. Part III: pain terms, a
current list with definitions and notes on usage. In:
Merskey H, Bogduk N (eds.). Classification of chronic
pain. Seattle (WA): IASP Press; 1994, p. 209–214.
Downloaded From: https://bioone.org/journals/Journal-of-Zoo-and-Wildlife-Medicine on 03 Dec 2023
Terms of Use: https://bioone.org/terms-of-use Access provided by Universidade Estadual Paulista (UNESP)
199
2. Kanui TI, Hole K, Miaron JC. Nociception in
crocodiles: capsaicin instillation, formalin and hot
plate tests. Zool Sci. 1990;7:537–540.
3. Kinney ME, Johnson SM, Sladky KK. Behavioral
evaluation of red-eared Slider Turtles (Trachemys
scripta elegans) administered either morphine or butor-
phanol following unilateral gonadectomy. J Herpetol
Med Surg. 2011;21(2–3):54–62.
4. Lynn B. Capsaicin: actions on nociceptive C-fibres
and therapeutic potential. Pain. 1990;41:61–69.
5. Olesen MG, Bertelsen MF, Perry SF, Wang T.
Effects of preoperative administration of butorphanol
or meloxicam on physiologic responses to surgery in
ball pythons. J Am Vet Med Assoc. 2008;233:1883–
1888.
6. Sanger DJ, McCarthy PS. Differential effects of
morphine on food and water intake in food deprived
and freely-feeding rats. Psychopharmacology (Berl).
1980;72:103–106.
7. Skovgaard N, Møller K, Gesser H, Wang T.
Histamine induces postprandial tachycardia through a
direct effect on cardiac H2-receptors in pythons. Am J
Phys. 2009;296(3):R774–R785.
8. Sneddon LU. Evolution of nociception in verte-
brates: comparative analysis of lower vertebrates. Brain
Res Rev. 2004;46:123–130.
9. Sneddon LU, Elwood RW, Adamo SA, Leach MC.
Defining and assessing animal pain. Anim Behav. 2014;
97:201–212.
10. Williams CJA, James LE, Bertelsen MF, Wang
T. Tachycardia in response to remote capsaicin injec-
tion as a model for nociception in the ball python
(Python regius). Vet Anaesth Analg. 2016;43:429–434.
Received for publication 21 March 2016