In mammals, it is well documented that perioperative pain protocols facilitate healing and recovery, and decrease morbidity and mortality rates, resulting in a faster return of normal behaviour (Sladky and Mans, 2012). In order to facilitate this, veterinary professionals must have an understanding of normal behaviour for different species. Typically an ethogram is used as an early development of a pain score, however one does not exist for reptiles due to a lack of consensus of the normal behaviours.
In human medicine, a pain scale for young infants which encompasses facial expressions can help to interpret pain when speech is limited or not possible (Bellieni, 2012). Similarly in veterinary medicine, pain scales must have the ability to interpret pain without verbal input, and instead focus on changes in an animal's normal behaviour. The Glasgow Pain Scale (Murrell et al, 2008), and the Colorado pain scoring system for cats (Hellyer et al, 2006) amongst others have been produced to overcome this barrier, with a focus on physiological and behavioural assessment (Barrett, 2014).
Behavioural signs of pain in reptiles
Goldberg and Shaffran (2015) have attempted to catalogue pain behaviours in reptiles, suggesting that acute pain may cause the animal to demonstrate escape or avoidance behaviours, with clinical signs of painful limb withdrawal and increased respiratory rate; whereas in chronic pain, pain behaviours seem only to be exhibited during manipulation or palpation. Eatwell (2010) highlights the general differences between chronic and acute pain in reptiles, but does not summarise pain behaviours in individual species.
The British Small Animal Veterinary Association has also documented general signs of pain in reptiles including anorexia, immobility, abnormal posture/locomotion, dull colouration, and increase in aggression (Brown and Pizzi, 2012).
Some species specific signs of pain have been mentioned; the adoption of an s-shape is highlighted in snakes, and in chelonians having their head down and dragging their plastron along the ground when mobilising is thought to indicate weakness due to pain (Figures 1–3) (Redrobe, 2004; Bays et al, 2006).
Read (2004) produced a survey to target professionals through the Association of Reptile and Amphibian Veterinary Surgeons, with a focus on anaesthesia in reptiles. Read's (2004) study was useful in determining the major drop off points when participants were asked about analgesia. Overall Read (2004) identified an overwhelming interest in reptile pain management, with 361 of 367 participants (98.4%) believing reptiles had the capacity to feel pain.
This author's study aimed to focus on pain behaviours, by collating the opinions of experienced veterinary professionals and animal carers. The rational for including non-veterinary trained personnel stemmed from the author's experience with very knowledgeable herpetoculturists who have kept animals for many years, and are often overlooked due to lack of veterinary training.
Methods
The prospective, cross-sectional study design allowed for the collection and evaluation of expert opinions on reptile pain perception through the use of an online survey, via Survey Gizmo. An initial draft was created using Microsoft Word, focusing on carefully formatted closed-ended questions to prevent avoidance due to length, lack of understanding, presence of a complicated format, or a ‘wordy’ accompanying explanation. Questions were phrased to ensure the participant committed to an answer, and in some cases, provided four options rather than five to avoid participants choosing the middle answer. Combination of multiple single questions into a table format allowed for the collective length of the questionnaire to be minimised, attempting to reduce participant fatigue and boredom.
Stratified sampling spilt the population into veterinary professionals, found using the RCVS list of specialists in exotic species and targeted using online blog forums, and herpetoculturists/pet owners, by posting links to the survey and a description on several reptile forum sites. Participants chose their own classification, which was further defined into level of experience, either time spent working with reptiles (veterinary professional) or number of reptiles owned (animal carers). Contact with the Federation of British Herpetoculturists (FBH) allowed for the survey link to be posted on their Facebook page and shared with the members of the FBH community. The survey was also promoted by BIAZA and The Royal Veterinary College through use of their logos to encourage uptake.
Survey piloting
The questionnaire was piloted with the veterinary staff at the Zoological Society of London (ZSL), London Zoo. The survey opened in January 2015 and closed in the March 2015.
Data excluded from analysis included any participants who: left the participant category blank; failed to answer question 17 or 18 (behaviour focused and deemed of high importance); or did not reach the end of the survey. 23 responses were removed from the final data analysis including: the fastest 6% of responses; responses longer than 72 seconds; patterned responses; gibberish; and questions where all answers had been ticked, making the data indistinguishable.
Results
A total of 365 responses were received from across the world (Figure 4). The spread of veterinary professionals and animal carers was not determined. The survey was able to reach all countries with access to the survey link. The UK and USA provided the most responses (n=65 in the UK, n=59 in the USA).
After data cleaning there were 234 workable responses: 48.7% (n=114) were veterinary professionals and 51.3% (n= 120) were animal carers, making the numbers in each group relatively equal. Of the veterinary professionals responses 94 were from veterinary surgeons and 29 from veterinary nurses.
Application to scenarios
The respondents were asked a series of questions relating to pain in reptiles associated with different scenarios. For each scenario, the participant could choose ‘not painful’, ‘mild pain’, ‘moderate pain’, or ‘severe pain’ (Figure 5).
In all three species, the scenario scoring the highest in ‘severe pain’ was ‘following significant trauma e.g. extensive burn injuries/crushing injuries’; 84.3% of respondents in snakes (n=193), 87.1% of respondents in lizards (n=203), and 81.7% of respondents in chelonians (n=187) believed that this scenario resulted in ‘severe pain’. ‘Suffering from a chronic illness’ scored highest for moderate pain in all three species; 62.9% of respondents in snakes (n=144), 67.7% of respondents in lizards (n=157), and 65.4% of respondents in chelonians (n=149). Participants believed ‘being handled after eating’ was most painful for snakes, with 35.5% (n=82) believing it was associated with mild pain, 6.5% (n=15) with moderate pain, and 1.3% (n=3) with severe pain. For lizards, 14.1% (n=33) believed this scenario caused mild pain, 0.4% (n=1) believed it caused moderate pain, and 1.3% (n=3) believed it caused severe pain. It was believed to cause least pain in chelonians; 8.7% (n=20) believed it caused mild pain, 0.0% (n=0) moderate pain, and 0.9% (n=2) severe pain.
The participants were then asked to give their opinion of what they believed to be the top three indicators of pain in each species, from the list of ten created using current literature (Figure 6).
The top three in each species group were then extracted, and from this the top pain behaviour from each species group was analysed. There was agreement on the top behavioural sign in lizards between the two participant groups and so only one behaviour was tested (Table 1).
| Veterinary professionals | Animal carers | |
|---|---|---|
| Snakes |
|
|
| Lizards |
|
|
| Chelonians |
|
|
A chi-square test indicated no significant associations between participant group and their perception of the above behaviours, as indicators of pain in the three reptilian subgroups. The chi-square results are given in Table 2.
| Behaviour | Number | Percentage | p-value |
|---|---|---|---|
| Increased in aggression | 268 | 2.11 | 0.550 |
| Cachexia | 289 | 1.67 | 0.643 |
| Unusual gait | 258 | 6.93 | 0.0740 |
| Unusual Posture | 228 | 2.29 | 0.515 |
| Keeping eyelids shut | 244 | 2.59 | 0.460 |
Veterinary professionals were asked a series of questions related to clinical practice that animal carers were not, to gain an overview of the current status of reptile care. Over half of participants (58.0% n=65) claimed they assess pain in reptiles multiple times during their stay, compared with only 3.60% of participants (n=4) who answered that they never assess reptilian pain in practice.
They were then questioned over confidence in assessing pain in reptiles. The results for this varied; of the 111 responses to ‘do you feel confident when assessing pain behaviours in reptiles’ 30.6% of participants (n=34) answered ‘yes’, 26.1% of participants (n=29) answered ‘no’, and 43.2% of participants (n=48) answered ‘dependent on species’. The question that followed concluded that 45.4% of participants (n=50) felt least confident in assessing pain in snakes, compared with chelonians and lizards; closely followed by 36.4% of participants (n=40) who felt ‘equal confidence in all three groups’.
Finally, of 229 responses, 74.2% of participants (n=170) believed that pain perception is altered when a reptile is being observed, with 98.8% of these participants (n=168) believing that this is an attempt to disguise pain from a predator.
Free-text boxes
Free-text boxes were designed to provide participants with the opportunity to add additional or more specific comments on pain behaviours in the three species.
‘Inappetance’ and ‘anorexia’ appeared frequently as a comment in all three species. Top scoring comments in snakes were: inappetence, hiding, lethargy, sudden movements (which included flailing, writhing, contortion of body and flinching), anorexia and flattening themselves out. Comments were also made on trouble sloughing skin and self mutilation.
In lizards, top scoring comments were: lethargy, hiding, anorexia, increased respiratory rate, gaping of mouth, inappetence and pulling away from painful stimulus. Other comments included lying flat to the ground, and going limp when handled. Pica was also mentioned, however there is currently a lack of literature to support this finding.
In chelonians, top scoring comments were: lethargy, hiding in shell, inappetence, anorexia and vocalisation (including hissing and huffing). Additional comments included an increase in the susceptibility of bronchial infections.
Discussion
Application to scenarios
Participants were presented with five scenarios extrapolated from companion animal literature, which were suggested as causing a degree of pain in mammals (Saliba, 2001; Rollin, 2011). ‘Burning/crushing injuries’ scored highest for ‘severe pain’. Burn injuries in reptiles are seen relatively commonly in veterinary practice, although there is a lack of understanding as to why. Early research has proven that reptiles do possess the mechanoreceptors, thermoreceptors and chemoreceptors necessary for nociception (Kenton et al, 1970; Bateson, 1991; Liang and Terashima, 1993), however Mosely (2011), suggests that the action of a thermal stimulus on these receptors may be affected by a reptile's thermal status and natural environmental adaptations. This could suggest that due to a reptile's ectothermic nature and ability to draw heat from their surroundings, thermoreceptors in reptiles may be less sensitive than those of mammals.
The scenario, ‘being handled directly after eating’ was added as a red herring. Literature highlights a regurgitation response that may be seen in snakes when handled directly after eating as a method of initiating a rapid escape (Summers, 2014), which may wrongly present as pain. There is no current research to support the evidence of pain associated with this scenario.
Pain behaviours
The second behaviour question asked about key pain behaviours, selected from current literature (Brown and Pizzi, 2012; Goldberg and Shaffran, 2015).
The behaviour ‘reduction in ability to rest’ scored below 10% in all species and in both participant groups, highlighting that this behaviour may not be a valid indictor of pain in reptiles. When housing reptiles in an unfamiliar veterinary environment, an increase in the frequency of observation of this behaviour by veterinary professionals might be expected, not necessarily associated with pain but perhaps anxiety or fear. Interestingly though, a chisquare test found no statistical significance between participant groups for this behaviour.
The biggest difference in opinion lay in that of ‘skin colour change’ as an indicator of pain in lizards. 42.9% (n=42) of veterinary professionals highlighted skin colour change as important, in comparison with just under half of that value in animal carers (22.2%, n=18), highlighting a stark contrast in opinion. Despite this, skin colour change was not highlighted as a top behaviour.
Confidence and pain assessment
It is interesting and reassuring that the majority of veterinary professionals are assessing pain more than once during the hospitalisation of a reptile (Figure 7). Confidence in doing so however appeared to be lacking. The survey concluded that 45.4% of participants (n= 50) felt least confident in assessing pain in snakes, compared with chelonians and lizards; closely followed by 36.4% of participants (n=40) who felt ‘equal confidence in all three groups’. 3.5% of people (n=4) did not answer the latter question, suggesting a possible lack of confidence in all three species. A distinct lack of confidence, particularly in assessing pain in snakes is evident, and highlights the potential requirement for further training of veterinary professionals in this area.
Free text analysis
Anorexia and inappetence were highlighted as potential pain indicators across all three reptile groups; difficulty lies in interpreting anorexia in reptiles due to their ability to fast for long periods of time without detrimental effects (Summers, 2014), therefore not eating for several weeks could be considered a normal finding in reptiles.
An increase in the susceptibility of bronchial infections was highlighted as an effect of pain in reptiles. In the literature, the main cause of an immunocompromised animal has been linked to poor husbandry, namely inadequate humidity and temperature making the animal more susceptible to a variety of respiratory pathogens (Schumacher, 2011). To the author's knowledge, there is no link between pain and immunocompromised reptiles.
Finally, ‘Flinching’ was noted as a sign of pain in reptiles. This has been highlighted in literature as a useful indicator of pain in companion animals (Murrell et al, 2008), however it appears to hold less value in reptiles as most will withdraw from touch regardless of whether they are in pain (Mosley, 2011).
The three behaviours in each species that had the highest frequency in the survey as previously discussed, and the three most frequently mentioned behaviours highlighted in the free-text analysis have been formulated into an ethogram key for each species, which can be applied to clinical practice (Figure 8). The ‘code for ethogram’ could be used to mark the observation of that given behaviour on a graph, set of clinical notes or a hospital sheet of a given animal, to help identify any trends in pain behaviours.
Limitations
There were several limitations to this paper that are worth discussing.
When asked ‘do you feel confident when assessing pain behaviours in reptiles’ three options were given; ‘yes’, ‘no’ or ‘dependent on species’. This last option made it difficult to distinguish whether ‘dependent on species’ meant a low confidence in all but lower in the stated species, or a high confidence in all three but a slightly lower in the stated species. A better option may have been ‘I don't know’, allowing for more accurate analysis and removal of the ‘I don't know’ answers if necessary, to assess the ‘yes’ and ‘no’ answers only.
The survey concluded that 45.4% of participants (n=50) felt least confident in assessing pain in snakes, compared with chelonians and lizards; closely followed by 36.4% of participants (n=40) who felt ‘equal confidence in all three groups’. 3.5% of people (n=4) did not answer the latter question, suggesting a possible lack of confidence in all three species. Analysis could have been repeated to examine the relationship between these figures and the participant group to which they belonged.
The chi square results for the ten behaviours all resulted in a p-value of less than 0.05, the significance level set for this study. Unusual gait however scored 0.0740, which may be described by some authors as ‘approaching significance’. Although in this study with the sample size used this is an insignificant value, there is a chance that with a bigger sample size significance may have been reached. This suggests that there may be a difference, between the opinions of veterinary professionals and experienced animal carers, in the importance of unusual gait as a sign of pain in reptiles.
In the free-text analysis, difficulty in distinguishing certain terms made data cleaning more difficult. The definitions between, ‘anorexia’, inappetance’ and ‘refusing food’ may have differed between individual people; one participant's view on ‘anorexia’ may be considered a normal amount of time to have not eaten, by a different respondent.
Although the survey attempted to target experienced members in both participant groups, the level of knowledge of each individual participant may have varied. Assessment of knowledge was very difficult in question form; specific targeting of experienced personnel with the use of a focus group may have resulted in more valid results overall.
Conclusions
Overall, there is a lack of confidence when assessing pain in reptiles with snakes being highlighted as the species that veterinary professionals are least confident in assessing.
Future research into reptile pain management should focus on the creation of a pain scale using the appropriate behavioural signs, and application to trial animals to determine its viability. The understanding of reptile pain behaviours will inevitably improve over time, however this study has highlighted pain behaviours that can be used as a reference in clinical practice until a reptile specific pain scale has been fully developed.