The Veterinary Journal 200 (2014) 109–115
Contents lists available at ScienceDirect
The Veterinary Journal journal homepage: www.elsevier.com/locate/tvjl
Monitoring equine visceral pain with a composite pain scale score and correlation with survival after emergency gastrointestinal surgery Johannes P.A.M. van Loon a,⇑, Valerie S.M. Jonckheer-Sheehy b, Willem Back a,d, P. René van Weeren a, Ludo J. Hellebrekers a,c,e a
Department of Equine Sciences, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 114, NL-3584 CM Utrecht, The Netherlands Department of Animals in Science and Society, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 102, NL-3584 CM Utrecht, The Netherlands Department of Clinical Sciences of Companion Animals, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 108, NL-3584 CM Utrecht, The Netherlands d Department of Surgery and Anaesthesiology of Domestic Animals, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, B-9820 Merelbeke, Belgium e Rudolf Magnus Institute of Neuroscience, UMC Utrecht, Universiteitsweg 100, NL-3584 CG Utrecht, The Netherlands b c
a r t i c l e
i n f o
Article history: Accepted 3 January 2014
Keywords: Colic Equine Composite Pain Scale
a b s t r a c t Recognition and management of equine pain have been studied extensively in recent decades and this has led to significant advances. However, there is still room for improvement in the ability to identify and treat pain in horses that have undergone emergency gastrointestinal surgery. This study assessed the validity and clinical application of the composite pain scale (CPS) in horses after emergency gastrointestinal surgery. Composite pain scores were determined every 4 h over 3 days following emergency gastrointestinal surgery in 48 horses. Inter-observer reliability was determined and another composite visceral pain score (numerical rating scale, NRS) was determined simultaneously with CPS scores. CPS scores had higher inter-observer reliability (r = 0.87, K = 0.84, P < 0.001), compared to NRS scores (r = 0.68, K = 0.72, P < 0.001). Horses that survived without complications had significantly lower CPS and NRS scores compared to horses that were euthanased or had to undergo re-laparotomy (P < 0.001). Breed and the location in the intestinal tract (small or large intestine) did not influence pain scores. In conclusion, the use of the CPS improved objectivity of pain scoring in horses following emergency gastrointestinal surgery. High inter-observer reliability allows for comparisons between different observers. This will be of great benefit in larger veterinary hospitals where several attending clinicians are often involved in the care of each case. Ó 2014 Elsevier Ltd. All rights reserved.
Introduction Pain recognition and management in animals are important in the optimisation of animal welfare and have received increasing attention in recent decades. As a result, considerable progress ˇ uela-Fernández et al., 2007; Flecknell, 2008; has been made (Vin Lerche, 2009). However, several surveys among veterinarians working with companion animals (Hugonnard et al., 2004; Williams et al., 2005; Hewson et al., 2006), farm animals (Hewson et al., 2007a, 2007b; Laven et al., 2009) and horses (Price et al., 2002; Dujardin and van Loon, 2011) have shown that further improvement is needed. In the study by Dujardin and van Loon (2011), a large proportion (40–60%) of veterinary practitioners classified their own ability to recognise pain and knowledge of analgesic therapy in horses as only moderate. Continuing education and research in pain management will help to improve methods for objective and reproducible pain assessment and so ⇑ Corresponding author. Tel.: +31 30 253 1350. E-mail address: [email protected] (J.P.A.M. van Loon). 1090-0233/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tvjl.2014.01.003
support the development of newer analgesic drugs and more refined analgesic techniques, which will consequently have a positive effect on welfare (Valverde and Gunkel, 2005). In general, pain is an aversive sensory and emotional experience representing awareness by the animal of damage or threat to the integrity of its tissues. The experience of pain may have an effect on several physiological parameters and the animal may change its physiology and behaviour with the aim to reduce or avoid damage, to reduce the likelihood of recurrence and to promote recovery (Molony and Kent, 1997). As pain is a complex multidimensional experience expressing itself in behavioural, physiological, and emotional variables, there is no single parameter that specifically indicates the presence of pain (Büttner and Finke, 2000; Lerche, 2009). Simple descriptive scales used to classify pain as absent, mild, moderate or severe are suboptimal and largely inadequate instruments for pain evaluation in animals, not in the least due to poor inter-observer reliability (Lindegaard et al., 2010). Combined interactive and observational multifactor pain behaviour rating scales, used together with physiological parameters, have been proposed
110
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115
as more sensitive in identifying and documenting changes instigated by pain in animals (Abbott et al., 1995; Dobromylskyj et al., 2000). Such composite pain scales (CPS) have been described in horses for orthopaedic (Bussières et al., 2008; Lindegaard et al., 2010) and for visceral pain (Pritchett et al., 2003; Graubner et al., 2011). The disadvantage of both visceral pain scales is that they contain a subjective pain score element. Further, the multifactorial numerical rating scale (NRS) by Pritchett et al. (2003) has not been subjected to inter-observer reliability assessment. In a previous study (van Loon et al., 2010) the CPS described by Bussières et al. (2008) was used to assess pain in a cross-section of horses in a referral centre. This pain scale, although developed for equine orthopaedic pain, contains various elements that can equally be applied to visceral pain. The data from that preliminary study (van Loon et al., 2010) suggested that the CPS could be of potential use for pain evaluation of horses in intensive care after gastrointestinal (colic) surgery. If this were so, one pain scale could be used both for orthopaedic and postoperative visceral pain assessment. Furthermore, the CPS was assessed in healthy painfree animals to assess the specificity of the scale and in horses after general anaesthesia to assess possible influences of anaesthesia. The aim of the current study was to further investigate interobserver reliability, sensitivity, specificity and clinical applicability of the CPS for pain assessment and related to survival in a larger group of horses suffering from postsurgical gastrointestinal pain. For reasons of comparison, pain was simultaneously scored by means of the composite NRS scale (Pritchett et al., 2003). Materials and methods
For evaluation of composite pain scores over time, horses were categorised according to outcome (survivors versus non-survivors) and the effects of breed and location and type of intestinal lesion on outcome and composite pain scores were assessed.
Composite pain scale (CPS) The CPS (Bussières et al., 2008) is a multifactorial numerical rating scale based on 13 parameters, including physiologic parameters, responses to stimuli, and spontaneous behavioural parameters (Table 2). Total pain scores range from zero (no signs of pain) to 39 (maximal pain score). To compare clinical usefulness and reproducibility of the CPS with another pain scoring system, we concurrently determined composite pain scores using the multifactorial NRS (Pritchett et al., 2003). The NRS contains nine behavioural categories and total pain scores range from 9 (no signs of pain) to 36 (maximal pain score) (Table 3). Pain scoring was performed with the animals in their box stalls.
Experimental design Due to practical constraints, observations were performed by five different observers, which is representative of the working situation in a large clinic. CPS and NRS scores were given by the same observer for each individual horse and all observers assessed equal parts of the study population. This was deemed justifiable given the very high inter-observer reliability (Spearman correlation coefficient 0.92, weighted kappa correlation 0.81, P < 0.001) determined earlier (van Loon et al., 2010). Prior to commencement of the study, all observers were able to familiarise themselves with the parameters that were assessed in the CPS and NRS using pain-free horses. The observers were not blinded for the clinical diagnosis. Horses were evaluated every 4 h for 3 days following colic surgery. The first pain scores were not performed earlier than 4 h after recovery to exclude effects of anaesthetics on composite pain scores (van Loon et al., 2010). To determine inter-observer variability, 10 randomly chosen horses were also scored at 40 time points by a second observer. Pain scoring by the second observer was done independently but at similar time points (within 10 min) as the first observer. The observers did not discuss their findings.
Animals and analgesic treatment The study design was approved by the institutional Ethics Committee on the Care and Use of Experimental Animals in compliance with Dutch legislation on animal experimentation and individual horse owner’s consent was obtained for all horses and ponies participating in this study. Forty-eight colic cases that had been admitted for emergency laparotomy to the Equine Referral Centre between September 2010 and July 2011, and were subsequently hospitalised in the intensive care unit (ICU) (Table 1), were observed. The study population consisted of 22 mares, 21 geldings and five stallions. Foals and mares with foals were excluded from the study, because of possible effects of mare–foal interaction on the assessment of pain scores. Breeds included Warmbloods (30), Friesians (two), Irish Cobs (two), Fjords (two), Haflingers (two), Standardbred (one) and ponies (nine). Standardised analgesic treatment protocol consisted of non-steroidal anti-inflammatory drugs IV (flunixin, Bedozane, Eurovet; 1.1 mg/ kg IV once or twice daily), combined with lidocaine (Lidocaine 20%, University Pharmacy) constant rate infusions (CRI; 50 lg/kg/min IV with loading dose of 1.3 mg/kg IV) for all small intestinal and large intestinal strangulating problems. Additional analgesic treatment was administered to effect at the discretion of the attending veterinarian and was independent of composite pain scores. Pain assessment was independent of analgesic treatment and the observers were not involved with day-to-day care of the horses and were therefore not aware of the analgesic treatments.
Table 1 Data of horses presenting with colic that had been admitted for emergency laparotomy (n = 48) included in the study, comparing survivors and non-survivors.
Number of horses Small intestinal lesion Strangulating lesion Non-strangulating lesion Large intestinal lesion Strangulating lesion Non-strangulating lesion Mean (±SD) weight (kg) Mean (±SD) age (years)
Survivors
Non-survivors
39 19a 11 8 21a 3 18 526 (120) 12 (7)
9 7 6 1 2 0 2 568 (79) 14 (5)
a One horse had both small and large intestines involved. Non-survivors are horses that developed complications after surgery and were euthanased or had relaparotomy.
Data processing and statistical analysis All data are expressed as medians and quartiles. Median area under the curve (AUC) values were calculated and differences in CPS scores between groups of horses were analysed using the Mann Whitney U test. Inter-observer reliability was assessed using Spearman’s correlation coefficients (r) and weighted kappa coefficients (j). Bland–Altman plots were used to visually evaluate correlations and determine limits of agreement (average difference ± 1.96 standard deviation of the difference) were calculated (Bland and Altman, 1986; Myles, 2007). Correlations between different individual CPS variables and total CPS scores were assessed using 261 CPS evaluations from 17 randomly selected animals. Statistical analysis was performed using SPSS version 20.0 (SPSS). Statistical significance was accepted at P < 0.05.
Results Inter-observer reliability CPS scores of two simultaneously scoring independent observers showed significant correlation. (r = 0.87, j = 0.84, P < 0.001) (Fig. 1). Limits of agreement for CPS scores of two independent observers were between 3.3 and +3.1 and the bias between two observers (average discrepancy) was nearly zero. The NRS scoring system also had a significant, but lower inter-observer reliability (r = 0.68, j = 0.72, P < 0.001) and wider ranges for the limits of agreement (between 4.4 and +6.6) with a bias between the two observers of 1.0. CPS and NRS scores over time for survivors and non-survivors Median CPS and NRS scores of horses that survived the ICU period showed a different pattern during the entire period, compared to horses that developed complications and did not survive (Fig. 2). The median AUC of the non-survivors was significantly higher than the median AUC of survivors (P < 0.001), both for CPS and NRS scores.
111
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115 Table 2 Score sheet of the composite pain score (CPS; Bussières et al., 2008). Behaviour Appearance (reluctance to move, restlessness, agitation and anxiety)
Sweating
Kicking at abdomen
Pawing on the floor (pointing, hanging limbs)
Posture (weight distribution, comfort)
Head movement
Appetite
Criteria Bright and alert, lowered head and ears, no reluctance to move
0
Bright, occasional head movements, no reluctance to move Restlessness, pricked up ears, abnormal facial expressions (like teeth grinding, yawning sedated or grimace face), dilated pupils Excited, continuous body movements, abnormal facial expression No obvious signs of sweat Damp to the touch Wet to the touch, beads of sweat are apparent over the horse’s body Excessive sweating, beads of sweat running off the animal Quietly standing, no kicking Occasional kicking at abdomen (1–2 times/5 min) Frequent kicking at abdomen (3–4 times/5 min) Excessive kicking at abdomen (>5 times/5 min), intermittent attempts to lie down and roll Quietly standing, no pawing Occasional pawing (1–2 times/5 min) Frequent pawing (3–4 times/5 min) Excessive pawing (>5 times/5 min) Stands quietly, normal walk Occasional weight shift, slight muscle tremors Non-weight bearing, abnormal weight distribution Stretching out, prostration, muscle tremors No evidence of discomfort, head straight ahead for the most part Intermittent head movements laterally or vertically, looking at flanks (1–2/5 min), lip curling (1–2/ 5 min) Intermittent and rapid head movements laterally or vertically, frequent looking at flank (3–4/5 min), lip curling (3–4/5 min) Continuous head movements, excessively looking at flank (>5 times/5 min), lip curling (>5 times/ 5 min) Eats hay readily or is not allowed to eat hay Hesitates to eat hay Shows little interest in hay, eats very little or takes hay in mouth but does not chew or swallow Neither shows interest in nor eats hay
1 2
Response to observer Interactive behaviour
Response to palpation of the painful area (abdominal incision)
Respiratory rate
Digestive sounds (bowel movements)
Rectal temperature
3 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3
Criteria
Score
Pays attention to people Exaggerated response to auditory stimulus (the observer calling the horse) Excessive-to-aggressive response to auditory stimulus (biting, turning the hindquarters towards the observer to kick) Stupor, prostration, no response to auditory stimulus No reaction to palpation
0 1 2
Mild reaction to palpation Resistance to palpation Violent reaction to palpation
1 2 3
Physiological data Heart rate
Score
Criteria 24–44 bpm 45–52 bpm 53–60 bpm >60 bpm 8–13 breaths pm 14–16 breaths pm 17–18 breaths pm 18 breaths pm Normal motility Decreased motility No motility Hypermotility 36.9–38.5 °C 36.4–36.9 °C or 38.5–39.0 °C 35.9–36.4 °C or 39.0–39.5 °C 35.4–35.9 °C or 39.5–40.0 °C
There were no differences in median AUC scores between Warmblood and non-Warmblood horses (Fig. 3a), nor between horses with small intestinal and large intestinal lesions (Fig. 3b). Horses with strangulating (Fig. 3c) and non-strangulating (Fig. 3d) lesions, if analysed as separate groups, showed
Score 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3
Total composite pain score
Influence of breed and type of gastrointestinal lesion
3 0
0–39
significantly lower median AUC scores in survivors compared to non-survivors (P < 0.05 and P < 0.001, respectively). Contribution of individual CPS components Fig. 4 shows the correlation coefficients between all 13 individual CPS components and the total CPS scores. All individual
112
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115
Table 3 Score sheet of the composite numerical rating scale (NRS; Pritchett et al., 2003). Behavioural category
1
2
3
Gross pain behaviour Head position Ear position Location in stall
None Above withers Forward, frequent movement At door watching environment
Spontaneous locomotion Response to open door Response to approach Lifting feet
Moves freely
Standing in middle, facing stall door Occasional steps
Moves to door
Looks at door
Moves to observer, ears forward Freely lifts feet when asked
Response to grain
Moves to door and reaches for grain
Looks at observer, ears forward Lifts feet after mild encouragement Looks at door
Occasional At withers Slightly back, little movement Standing in middle, facing sides of stall
4 Continuous Below withers Standing in middle, facing back of stall No movement No response
Moves away from observer
Does not move, ears back Extremely unwilling to lift feet No response
Fig. 1. Scatter plot of composite pain scores (CPS) (A), assessed by two different observers at the same time (n = 40, some dots represent two or more combinations of scores), r = 0.84 (P < 0.001). Bland–Altman plot (B) of same 40 CPS combinations (limits of agreement: 3.3 to +3.1).
Fig. 2. Median ± 25–75th percentiles of composite pain scores (CPS) and numerical rating scale (NRS) scores (A + C) of horses that left the intensive care unit (ICU) alive (n = 39) and of horses that did not survive ICU hospitalisation (n = 9). T, hours after recovery from general anaesthesia for emergency gastrointestinal surgery. Median area under the curve (AUC) scores of survivors (B) and non-survivors (D). Bold lines show median scores, boxes show 25–75th percentiles, error bars show 10–90th percentiles. P < 0.001.
parameters, except for ‘sweating’, showed a significant correlation with total CPS score (P < 0.001). ‘Pawing on the floor’, ‘appearance’, ‘head movements’ and ‘interactive behaviour’ all obtained correlation coefficients with total CPS scores of >0.5.
Discussion Simple visual analogue scale (VAS) scores have been described as reliable, sensitive and very suitable for pain assessment in
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115
Fig. 3. Median area under the curve (AUC) values for Warmbloods (n = 30) and nonWarmbloods (n = 18) (A). Median AUC values for horses with small intestinal (n = 25) and large intestinal disorder (n = 23) (B). Median AUC values for horses that had a strangulating intestinal lesion (n = 14 survivors, n = 6 non-survivors) (C). Median AUC values for horses that did not have a strangulating intestinal lesion (n = 26 survivors, n = 3 non-survivors) (D). Bold lines show median scores, boxes show 25–75th percentiles, error bars show 10–90th percentiles. P < 0.05, P < 0.001.
Fig. 4. Correlation coefficients between 13 individual composite pain scores (CPS) parameters and the total composite pain score (261 CPS evaluations from n = 17 animals), P < 0.001.
verbal humans after day surgery, if the patients can score themselves (Coll et al., 2004). In neonates and nonverbal children as well as in older adults with dementia, behavioural and composite pain scales have been assessed and cross-validated for clinical use (Suraseranivongse et al., 2001, 2006; Herr et al., 2006; Ersek et al., 2011). It was hypothesised that such scales might also be applicable to horses. The CPS was developed for orthopaedic pain in horses by Bussières et al. (2008) and later applied in horses with somatic and visceral pain by van Loon et al. (2010). The current study demonstrates the usefulness of the CPS for the objective, reliable and reproducible evaluation of pain in horses after emergency gastrointestinal surgery. This was further supported by the low inter-observer variability and the clinically acceptable limits of agreement found in Bland–Altman analysis, combined with the lack of bias (average discrepancy between two observers). The specificity and sensitivity for differentiating between pain-free healthy horses and ICU horses and between
113
ICU horses that responded well to treatment and ICU horses that developed complications leading to euthanasia or re-laparotomy also adds to the practical usefulness of the CPS for this category of horses. The composite pain scale by Bussières et al. (2008) combines physiological, behavioural and interactive variables. Although originally developed for orthopaedic pain, the scale contains several elements that could also be expressions of acute visceral pain. If this pain scale was applicable to visceral pain, this would be convenient, as one pain scale could then be used for both orthopaedic and visceral pain in horses. In the original study by Bussières et al. (2008) the key specific and most sensitive behavioural indices were response to palpation of the painful area, posture and, to a lesser extent, pawing on the floor, kicking at the abdomen and head movements. For visceral postoperative pain, pawing on the floor, overall appearance, head movements and interactive behaviour appear to be the most important elements. The original study (Bussières et al., 2008) and our current results show that the CPS can be used for the assessment of pain intensity in both orthopaedic and visceral types of pain, for each of which there may be a subset of most sensitive parameters. The specificity and sensitivity of the CPS could along this line potentially be increased by introducing different weighting factors for individual variables or non-specific parameters could be removed from the definite CPS scale. A shortened version of the CPS for clinical use could consequently consist of only the best correlating individual parameters (pawing on the floor, appearance, head movements and interactive behaviour) that all showed correlations with total CPS scores of >0.5. Further analysis of this shortened version of the CPS was outside the scope of the current study. One of the limitations of our study is that a validated ‘gold standard’ pain score for validation of the CPS for postoperative visceral pain is lacking. As the NRS (Pritchett et al., 2003) has not been fully validated, it is debatable whether this pain scale can be used to meet this objective. Because of this lack of full validation and because of only moderate inter-observer reliability for the NRS that we found in our study, we decided not to use the NRS for validation of the CPS. The NRS has been used for assessment of pain in horses after exploratory celiotomy for colic and in healthy pain-free horses and horses that were subjected to general anaesthesia, and was shown to be able to discriminate between healthy painfree horses and those with visceral postoperative pain (Pritchett et al., 2003). For this reason, the NRS was considered the most suitable scale to compare the CPS with, when assessing the suitability of the latter for the assessment of visceral pain. The NRS has been used in two subsequent studies. In a study in horses after emergency celiotomy, Sellon et al. (2004) found decreased NRS scores in animals treated with flunixin and butorphanol compared to horses that were treated with flunixin only. These observations corresponded with lower plasma cortisol levels, reduced weight loss and faster recovery from surgery in the test group. These results show the sensitivity of the NRS to detect differences in horses that underwent different treatment protocols after emergency celiotomy. Sanz et al. (2009) did not find significant differences in NRS scores in horses undergoing routine castration after being treated with different analgesic protocols. Although the NRS has been shown to be useful in assessing pain in horses after exploratory celiotomy, we believe the CPS has several advantages over the NRS. First, the NRS does not include physiological parameters (although correlations between physiological variables and the NRS were found in the earlier mentioned studies). Second, one of the nine parameters of the NRS is ‘gross pain behaviour’, which is largely subjective. The CPS does not contain subjective criteria and its objectivity and more reproducible results are supported by the high inter-observer reliability of the CPS in our study. Although it was not assessed in the original NRS studies,
114
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115
we found only moderate inter-observer reliability for the NRS, possibly caused by the influence of the subjective gross pain behaviour parameter. Finally, the CPS has been shown to be useful for both orthopaedic and visceral pain in horses, making it a widely applicable universal tool in clinical practice. Sweating was the only variable that did not correlate with total CPS score. Bussières et al. (2008) previously concluded that sweating was a non-sensitive parameter for orthopaedic pain either. Recently, Graubner et al. (2011) described a post abdominal surgery pain assessment scale (PASPAS) in horses. This scale combines behavioural (observational and interactional) with physiological parameters, as in the CPS. Inter-observer reliability was sufficient and PASPAS scores were not influenced by general anaesthesia in healthy animals. Due to the lack of a ‘gold standard’ pain score, the authors used the differences in pain scores between horses with painful complications (ileus, laminitis, colitis, early wound infections) and horses without complications (between which the scores significantly differed) to validate the composite pain scores. A disadvantage of this scale compared to the CPS is that, like the NRS, it contains a general subjective assessment of pain. Further, the PASPAS score has specifically been designed for the assessment of abdominal postsurgical pain and has not (yet) been extended to horses with orthopaedic pain. Comparison of the PASPAS-score with both NRS and CPS could possibly lead to further optimisation and fine-tuning of scoring systems for equine (visceral) pain assessment. In surveys among equine veterinarians in the UK (Price et al., 2002) and in The Netherlands and Belgium (Dujardin and van Loon, 2011), heart rate was cited as one of the major criteria used to assess pain. However, no objective study demonstrating a positive correlation between heart rate and pain in horses has been published so far. Moreover, several studies have reported that increased heart and respiratory rates are not specifically indicative of postoperative pain in cats and dogs (Hansen et al., 1997; Holton et al., 1998). Heart rate is likely to be influenced by many factors other than pain, including endotoxaemia and hypovolaemia (Sellon et al., 2004). Despite its limited reliability as a single indicator, heart rate is nevertheless often incorporated in various composite pain scales. Bussières et al. (2008) found moderate specificity and sensitivity for heart rate, which is in accordance with the moderate correlation between heart rate and total CPS scores we found in the current study. In human infants after surgery, similar observations have been made (van Dijk et al., 2001). The ultimate aim of pain scales is to provide a reliable tool for guidance of analgesic treatment. For an appropriate analgesic strategy, the pain scale not only needs to be specific for pain, but ideally should indicate objective cut-off values as well. Gerbershagen et al. (2011) described four different methods for cut-off point analysis in humans. Parameters such as verbal expression of pain state, subjective tolerable pain intensity, satisfaction with pain therapy, pain-related interference with mood and estimation of pain thresholds are used to achieve this (Gerbershagen et al., 2011). In nonverbal animals, determining such cut-off values is, to say the least, challenging.
Conclusions CPS was shown to allow objective discrimination between horses that recovered well after emergency gastrointestinal surgery and horses that developed painful complications and did not survive the ICU period. In other words, pain scores significantly correlated with final clinical outcome. The use of the CPS improved objectivity in pain management and, on account of a high inter-observer reliability over a range of CPS scores, enabled comparisons between different observers. This would be of major benefit in
larger veterinary hospitals where several attending clinicians are often involved in the treatment of each case. Conflict of interest statement None of the authors of this paper has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper. Acknowledgements We would like to thank Judith Weening, Astrid Rijckaert, Esther van Neste, Marjolijn Beugelsdijk and Marjolijn Heemskerk for their help with data collection. References Abbott, F.V., Franklin, K.B.J., Westbrook, R.F., 1995. The formalin test: Scoring properties of the first and second phases of the pain response in rats. Pain 60, 91–102. Bland, J.M., Altman, D.G., 1986. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 8, 307–310. Bussières, G., Jacques, C., Lainay, O., Beauchamp, G., Leblond, A., Cadoré, J.L., Desmaizières, L.M., Cuvelliez, S.G., Troncy, E., 2008. Development of a composite orthopaedic pain scale in horses. Research in Veterinary Science 85, 294–306. Büttner, W., Finke, W., 2000. Analysis of behavioural and physiological parameters for the assessment of postoperative analgesic demand in newborns, infants and young children: A comprehensive report on seven consecutive studies. Pediatric Anesthesia 10, 303–318. Coll, A.M., Ameen, J.R.M., Mead, D., 2004. Postoperative pain assessment tools in day surgery: Literature review. Journal of Advanced Nursing 46, 124–133. Dobromylskyj, P., Flecknell, P.A., Lascelles, B.D., Livingston, A., Taylor, P., WatermanPearson, A., 2000. Pain assessment. In: Flecknell, P.A., Waterman-Pearson, A. (Eds.), Pain Management in Animals. W.B. Saunders, Philadelphia, Pennsylvania, USA, pp. 53–79. Dujardin, C.L.L., van Loon, J.P.A.M., 2011. Pain recognition and treatment in the horse: A survey of equine veterinarians in the Netherlands and Belgium. Tijdschrift voor Diergeneeskunde 136, 715–724. Ersek, M., Polissar, N., Neradilek, M.B., 2011. Development of a composite pain measure for persons with advanced dementia: Exploratory analyses in selfreporting nursing home residents. Journal of Pain and Symptom Management 41, 566–579. Flecknell, P., 2008. Analgesia from a veterinary perspective. British Journal of Anaesthesia 101, 121–124. Gerbershagen, H.J., Rothaug, J., Kalkman, C.J., Meissner, W., 2011. Determination of moderate-to-severe postoperative pain on the numerical rating scale: A cut-off point analysis applying four different methods. British Journal of Anaesthesia 107, 619–626. Graubner, C., Gerber, V., Doherr, M., Spadavecchia, C., 2011. Clinical application and reliability of a post abdominal surgery pain assessment scale (PASPAS) in horses. The Veterinary Journal 188, 178–183. Hansen, B.D., Hardie, E.M., Carroll, G.S., 1997. Physiological measurements after ovariohysterectomy in dogs: What’s normal? Applied Animal Behaviour Science 51, 101–109. Herr, K., Bjoro, K., Decker, S., 2006. Tools for assessment of pain in nonverbal older adults with dementia: A state-of-the-science review. Journal of Pain and Symptom Management 31, 170–192. Hewson, C.J., Dohoo, I.R., Lemke, K.A., 2006. Perioperative use of analgesics in dogs and cats by Canadian veterinarians in 2001. Canadian Veterinary Journal 47, 352–359. Hewson, C.J., Dohoo, I.R., Lemke, K.A., Barkema, H.W., 2007a. Canadian veterinarians’ use of analgesics in cattle, pigs and horses in 2004 and 2005. Canadian Veterinary Journal 48, 155–164. Hewson, C.J., Dohoo, I.R., Lemke, K.A., Barkema, H.W., 2007b. Factors affecting Canadian veterinarians’ use of analgesics when dehorning beef and dairy calves. Canadian Veterinary Journal 48, 1129–11136. Holton, L.L., Scott, E.M., Nolan, A.M., Reid, J., Welsh, E., 1998. Relationship between physiological factors and clinical pain in dogs scored using a numerical rating scale. Journal of Small Animal Practice 39, 469–474. Hugonnard, M., Leblond, A., Keroack, S., Cadore, J., Troncy, E., 2004. Attitudes and concerns of French veterinarians towards pain and analgesia in dogs and cats. Veterinary Anaesthesia and Analgesia 31, 154–163. Laven, R.A., Huxley, J.N., Whay, H.R., Stafford, K.J., 2009. Results of a survey of attitudes of dairy veterinarians in New Zealand regarding painful procedures and conditions in cattle. New Zealand Veterinary Journal 57, 215–220. Lerche, P., 2009. Assessment and treatment of pain in horses. Equine Veterinary Education 21, 44–45. Lindegaard, C., Thomsen, M.H., Larsen, S., Andersen, P.H., 2010. Analgesic efficacy of intra-articular morphine in experimentally induced radiocarpal synovitis in horses. Veterinary Anaesthesia and Analgesia 37, 171–185.
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115 Molony, V., Kent, J.E., 1997. Assessment of acute pain in farm animals using behavioural and physiological measurements. Journal of Animal Science 75, 266–272. Myles, P.S., 2007. Using the Bland–Altman method to measure agreement with repeated measures. British Journal of Anaesthesia 99, 309–311. Price, J., Marques, J.M., Welsh, E.M., Waran, N.K., 2002. Pilot epidemiological study of attitudes towards pain in horses. Veterinary Record 151, 570–575. Pritchett, L.C., Ulibarri, C., Roberts, M.C., Schneider, R.K., Sellon, D.C., 2003. Identification of potential physiological and behavioural indicators of postoperative pain in horses after exploratory celiotomy for colic. Applied Animal Behaviour Science 80, 31–43. Sanz, M.G., Sellon, D.C., Cary, J.A., Hines, M.T., Farnsworth, K.D., 2009. Analgesic effects of butorphanol tartrate and phenylbutazone administered alone and in combination in young horses undergoing routine castration. Journal of American Veterinary Medical Association 235, 1194–1203. Sellon, D.C., Roberts, M.C., Blikslager, A.T., Ulibarri, C., Papich, M.G., 2004. Effects of continuous rate intravenous infusion of butorphanol on physiologic and outcome variables in horses after celiotomy. Journal of Veterinary Internal Medicine 18, 555–563. Suraseranivongse, S., Santawat, U., Kraiprasit, K., Petcharatana, S., Prakkamodom, S., Muntraporn, N., 2001. Cross-validation of a composite pain scale for preschool children within 24 hours of surgery. British Journal of Anaesthesia 87, 400–405.
115
Suraseranivongse, S., Kaosaard, R., Intakong, P., Pornsiriprasert, S., Karnchana, Y., Kaopinpruck, J., Sangjeen, K., 2006. A comparison of postoperative pain scales in neonates. British Journal of Anaesthesia 97, 540–544. Valverde, A., Gunkel, C.I., 2005. Pain management in horses and farm animals. Journal of Veterinary Emergency and Critical Care 15, 295–307. Van Dijk, M., de Boer, J.B., Koot, H.M., Duivenoorden, H.J., Passchier, J., Bouwmeester, N., Tibboel, D., 2001. The association between physiological and behavioural pain measures in 0- to 3-year-old infants after major surgery. Journal of Pain and Symptom Management 22, 600–609. Van Loon, J.P.A.M., Back, W., Hellebrekers, L.J., van Weeren, P.R., 2010. Application of a composite pain scale to objectively monitor horses with somatic and visceral pain under hospital conditions. Journal of Equine Veterinary Science 30, 641– 649. ˇ uela-Fernández, I., Jones, E., Welsh, E.M., Fleetwood-Walker, S.M., 2007. Pain Vin mechanisms and their implication for the management of pain in farm and companion animals. The Veterinary Journal 174, 227–239. Williams, V.M., Lascelles, B.D., Robson, M.C., 2005. Current attitudes to, and use of, peri-operative analgesia in dogs and cats by veterinarians in New Zealand. New Zealand Veterinary Journal 53, 193–202.
Contents lists available at ScienceDirect
The Veterinary Journal journal homepage: www.elsevier.com/locate/tvjl
Monitoring equine visceral pain with a composite pain scale score and correlation with survival after emergency gastrointestinal surgery Johannes P.A.M. van Loon a,⇑, Valerie S.M. Jonckheer-Sheehy b, Willem Back a,d, P. René van Weeren a, Ludo J. Hellebrekers a,c,e a
Department of Equine Sciences, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 114, NL-3584 CM Utrecht, The Netherlands Department of Animals in Science and Society, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 102, NL-3584 CM Utrecht, The Netherlands Department of Clinical Sciences of Companion Animals, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 108, NL-3584 CM Utrecht, The Netherlands d Department of Surgery and Anaesthesiology of Domestic Animals, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, B-9820 Merelbeke, Belgium e Rudolf Magnus Institute of Neuroscience, UMC Utrecht, Universiteitsweg 100, NL-3584 CG Utrecht, The Netherlands b c
a r t i c l e
i n f o
Article history: Accepted 3 January 2014
Keywords: Colic Equine Composite Pain Scale
a b s t r a c t Recognition and management of equine pain have been studied extensively in recent decades and this has led to significant advances. However, there is still room for improvement in the ability to identify and treat pain in horses that have undergone emergency gastrointestinal surgery. This study assessed the validity and clinical application of the composite pain scale (CPS) in horses after emergency gastrointestinal surgery. Composite pain scores were determined every 4 h over 3 days following emergency gastrointestinal surgery in 48 horses. Inter-observer reliability was determined and another composite visceral pain score (numerical rating scale, NRS) was determined simultaneously with CPS scores. CPS scores had higher inter-observer reliability (r = 0.87, K = 0.84, P < 0.001), compared to NRS scores (r = 0.68, K = 0.72, P < 0.001). Horses that survived without complications had significantly lower CPS and NRS scores compared to horses that were euthanased or had to undergo re-laparotomy (P < 0.001). Breed and the location in the intestinal tract (small or large intestine) did not influence pain scores. In conclusion, the use of the CPS improved objectivity of pain scoring in horses following emergency gastrointestinal surgery. High inter-observer reliability allows for comparisons between different observers. This will be of great benefit in larger veterinary hospitals where several attending clinicians are often involved in the care of each case. Ó 2014 Elsevier Ltd. All rights reserved.
Introduction Pain recognition and management in animals are important in the optimisation of animal welfare and have received increasing attention in recent decades. As a result, considerable progress ˇ uela-Fernández et al., 2007; Flecknell, 2008; has been made (Vin Lerche, 2009). However, several surveys among veterinarians working with companion animals (Hugonnard et al., 2004; Williams et al., 2005; Hewson et al., 2006), farm animals (Hewson et al., 2007a, 2007b; Laven et al., 2009) and horses (Price et al., 2002; Dujardin and van Loon, 2011) have shown that further improvement is needed. In the study by Dujardin and van Loon (2011), a large proportion (40–60%) of veterinary practitioners classified their own ability to recognise pain and knowledge of analgesic therapy in horses as only moderate. Continuing education and research in pain management will help to improve methods for objective and reproducible pain assessment and so ⇑ Corresponding author. Tel.: +31 30 253 1350. E-mail address: [email protected] (J.P.A.M. van Loon). 1090-0233/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tvjl.2014.01.003
support the development of newer analgesic drugs and more refined analgesic techniques, which will consequently have a positive effect on welfare (Valverde and Gunkel, 2005). In general, pain is an aversive sensory and emotional experience representing awareness by the animal of damage or threat to the integrity of its tissues. The experience of pain may have an effect on several physiological parameters and the animal may change its physiology and behaviour with the aim to reduce or avoid damage, to reduce the likelihood of recurrence and to promote recovery (Molony and Kent, 1997). As pain is a complex multidimensional experience expressing itself in behavioural, physiological, and emotional variables, there is no single parameter that specifically indicates the presence of pain (Büttner and Finke, 2000; Lerche, 2009). Simple descriptive scales used to classify pain as absent, mild, moderate or severe are suboptimal and largely inadequate instruments for pain evaluation in animals, not in the least due to poor inter-observer reliability (Lindegaard et al., 2010). Combined interactive and observational multifactor pain behaviour rating scales, used together with physiological parameters, have been proposed
110
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115
as more sensitive in identifying and documenting changes instigated by pain in animals (Abbott et al., 1995; Dobromylskyj et al., 2000). Such composite pain scales (CPS) have been described in horses for orthopaedic (Bussières et al., 2008; Lindegaard et al., 2010) and for visceral pain (Pritchett et al., 2003; Graubner et al., 2011). The disadvantage of both visceral pain scales is that they contain a subjective pain score element. Further, the multifactorial numerical rating scale (NRS) by Pritchett et al. (2003) has not been subjected to inter-observer reliability assessment. In a previous study (van Loon et al., 2010) the CPS described by Bussières et al. (2008) was used to assess pain in a cross-section of horses in a referral centre. This pain scale, although developed for equine orthopaedic pain, contains various elements that can equally be applied to visceral pain. The data from that preliminary study (van Loon et al., 2010) suggested that the CPS could be of potential use for pain evaluation of horses in intensive care after gastrointestinal (colic) surgery. If this were so, one pain scale could be used both for orthopaedic and postoperative visceral pain assessment. Furthermore, the CPS was assessed in healthy painfree animals to assess the specificity of the scale and in horses after general anaesthesia to assess possible influences of anaesthesia. The aim of the current study was to further investigate interobserver reliability, sensitivity, specificity and clinical applicability of the CPS for pain assessment and related to survival in a larger group of horses suffering from postsurgical gastrointestinal pain. For reasons of comparison, pain was simultaneously scored by means of the composite NRS scale (Pritchett et al., 2003). Materials and methods
For evaluation of composite pain scores over time, horses were categorised according to outcome (survivors versus non-survivors) and the effects of breed and location and type of intestinal lesion on outcome and composite pain scores were assessed.
Composite pain scale (CPS) The CPS (Bussières et al., 2008) is a multifactorial numerical rating scale based on 13 parameters, including physiologic parameters, responses to stimuli, and spontaneous behavioural parameters (Table 2). Total pain scores range from zero (no signs of pain) to 39 (maximal pain score). To compare clinical usefulness and reproducibility of the CPS with another pain scoring system, we concurrently determined composite pain scores using the multifactorial NRS (Pritchett et al., 2003). The NRS contains nine behavioural categories and total pain scores range from 9 (no signs of pain) to 36 (maximal pain score) (Table 3). Pain scoring was performed with the animals in their box stalls.
Experimental design Due to practical constraints, observations were performed by five different observers, which is representative of the working situation in a large clinic. CPS and NRS scores were given by the same observer for each individual horse and all observers assessed equal parts of the study population. This was deemed justifiable given the very high inter-observer reliability (Spearman correlation coefficient 0.92, weighted kappa correlation 0.81, P < 0.001) determined earlier (van Loon et al., 2010). Prior to commencement of the study, all observers were able to familiarise themselves with the parameters that were assessed in the CPS and NRS using pain-free horses. The observers were not blinded for the clinical diagnosis. Horses were evaluated every 4 h for 3 days following colic surgery. The first pain scores were not performed earlier than 4 h after recovery to exclude effects of anaesthetics on composite pain scores (van Loon et al., 2010). To determine inter-observer variability, 10 randomly chosen horses were also scored at 40 time points by a second observer. Pain scoring by the second observer was done independently but at similar time points (within 10 min) as the first observer. The observers did not discuss their findings.
Animals and analgesic treatment The study design was approved by the institutional Ethics Committee on the Care and Use of Experimental Animals in compliance with Dutch legislation on animal experimentation and individual horse owner’s consent was obtained for all horses and ponies participating in this study. Forty-eight colic cases that had been admitted for emergency laparotomy to the Equine Referral Centre between September 2010 and July 2011, and were subsequently hospitalised in the intensive care unit (ICU) (Table 1), were observed. The study population consisted of 22 mares, 21 geldings and five stallions. Foals and mares with foals were excluded from the study, because of possible effects of mare–foal interaction on the assessment of pain scores. Breeds included Warmbloods (30), Friesians (two), Irish Cobs (two), Fjords (two), Haflingers (two), Standardbred (one) and ponies (nine). Standardised analgesic treatment protocol consisted of non-steroidal anti-inflammatory drugs IV (flunixin, Bedozane, Eurovet; 1.1 mg/ kg IV once or twice daily), combined with lidocaine (Lidocaine 20%, University Pharmacy) constant rate infusions (CRI; 50 lg/kg/min IV with loading dose of 1.3 mg/kg IV) for all small intestinal and large intestinal strangulating problems. Additional analgesic treatment was administered to effect at the discretion of the attending veterinarian and was independent of composite pain scores. Pain assessment was independent of analgesic treatment and the observers were not involved with day-to-day care of the horses and were therefore not aware of the analgesic treatments.
Table 1 Data of horses presenting with colic that had been admitted for emergency laparotomy (n = 48) included in the study, comparing survivors and non-survivors.
Number of horses Small intestinal lesion Strangulating lesion Non-strangulating lesion Large intestinal lesion Strangulating lesion Non-strangulating lesion Mean (±SD) weight (kg) Mean (±SD) age (years)
Survivors
Non-survivors
39 19a 11 8 21a 3 18 526 (120) 12 (7)
9 7 6 1 2 0 2 568 (79) 14 (5)
a One horse had both small and large intestines involved. Non-survivors are horses that developed complications after surgery and were euthanased or had relaparotomy.
Data processing and statistical analysis All data are expressed as medians and quartiles. Median area under the curve (AUC) values were calculated and differences in CPS scores between groups of horses were analysed using the Mann Whitney U test. Inter-observer reliability was assessed using Spearman’s correlation coefficients (r) and weighted kappa coefficients (j). Bland–Altman plots were used to visually evaluate correlations and determine limits of agreement (average difference ± 1.96 standard deviation of the difference) were calculated (Bland and Altman, 1986; Myles, 2007). Correlations between different individual CPS variables and total CPS scores were assessed using 261 CPS evaluations from 17 randomly selected animals. Statistical analysis was performed using SPSS version 20.0 (SPSS). Statistical significance was accepted at P < 0.05.
Results Inter-observer reliability CPS scores of two simultaneously scoring independent observers showed significant correlation. (r = 0.87, j = 0.84, P < 0.001) (Fig. 1). Limits of agreement for CPS scores of two independent observers were between 3.3 and +3.1 and the bias between two observers (average discrepancy) was nearly zero. The NRS scoring system also had a significant, but lower inter-observer reliability (r = 0.68, j = 0.72, P < 0.001) and wider ranges for the limits of agreement (between 4.4 and +6.6) with a bias between the two observers of 1.0. CPS and NRS scores over time for survivors and non-survivors Median CPS and NRS scores of horses that survived the ICU period showed a different pattern during the entire period, compared to horses that developed complications and did not survive (Fig. 2). The median AUC of the non-survivors was significantly higher than the median AUC of survivors (P < 0.001), both for CPS and NRS scores.
111
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115 Table 2 Score sheet of the composite pain score (CPS; Bussières et al., 2008). Behaviour Appearance (reluctance to move, restlessness, agitation and anxiety)
Sweating
Kicking at abdomen
Pawing on the floor (pointing, hanging limbs)
Posture (weight distribution, comfort)
Head movement
Appetite
Criteria Bright and alert, lowered head and ears, no reluctance to move
0
Bright, occasional head movements, no reluctance to move Restlessness, pricked up ears, abnormal facial expressions (like teeth grinding, yawning sedated or grimace face), dilated pupils Excited, continuous body movements, abnormal facial expression No obvious signs of sweat Damp to the touch Wet to the touch, beads of sweat are apparent over the horse’s body Excessive sweating, beads of sweat running off the animal Quietly standing, no kicking Occasional kicking at abdomen (1–2 times/5 min) Frequent kicking at abdomen (3–4 times/5 min) Excessive kicking at abdomen (>5 times/5 min), intermittent attempts to lie down and roll Quietly standing, no pawing Occasional pawing (1–2 times/5 min) Frequent pawing (3–4 times/5 min) Excessive pawing (>5 times/5 min) Stands quietly, normal walk Occasional weight shift, slight muscle tremors Non-weight bearing, abnormal weight distribution Stretching out, prostration, muscle tremors No evidence of discomfort, head straight ahead for the most part Intermittent head movements laterally or vertically, looking at flanks (1–2/5 min), lip curling (1–2/ 5 min) Intermittent and rapid head movements laterally or vertically, frequent looking at flank (3–4/5 min), lip curling (3–4/5 min) Continuous head movements, excessively looking at flank (>5 times/5 min), lip curling (>5 times/ 5 min) Eats hay readily or is not allowed to eat hay Hesitates to eat hay Shows little interest in hay, eats very little or takes hay in mouth but does not chew or swallow Neither shows interest in nor eats hay
1 2
Response to observer Interactive behaviour
Response to palpation of the painful area (abdominal incision)
Respiratory rate
Digestive sounds (bowel movements)
Rectal temperature
3 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3
Criteria
Score
Pays attention to people Exaggerated response to auditory stimulus (the observer calling the horse) Excessive-to-aggressive response to auditory stimulus (biting, turning the hindquarters towards the observer to kick) Stupor, prostration, no response to auditory stimulus No reaction to palpation
0 1 2
Mild reaction to palpation Resistance to palpation Violent reaction to palpation
1 2 3
Physiological data Heart rate
Score
Criteria 24–44 bpm 45–52 bpm 53–60 bpm >60 bpm 8–13 breaths pm 14–16 breaths pm 17–18 breaths pm 18 breaths pm Normal motility Decreased motility No motility Hypermotility 36.9–38.5 °C 36.4–36.9 °C or 38.5–39.0 °C 35.9–36.4 °C or 39.0–39.5 °C 35.4–35.9 °C or 39.5–40.0 °C
There were no differences in median AUC scores between Warmblood and non-Warmblood horses (Fig. 3a), nor between horses with small intestinal and large intestinal lesions (Fig. 3b). Horses with strangulating (Fig. 3c) and non-strangulating (Fig. 3d) lesions, if analysed as separate groups, showed
Score 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3
Total composite pain score
Influence of breed and type of gastrointestinal lesion
3 0
0–39
significantly lower median AUC scores in survivors compared to non-survivors (P < 0.05 and P < 0.001, respectively). Contribution of individual CPS components Fig. 4 shows the correlation coefficients between all 13 individual CPS components and the total CPS scores. All individual
112
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115
Table 3 Score sheet of the composite numerical rating scale (NRS; Pritchett et al., 2003). Behavioural category
1
2
3
Gross pain behaviour Head position Ear position Location in stall
None Above withers Forward, frequent movement At door watching environment
Spontaneous locomotion Response to open door Response to approach Lifting feet
Moves freely
Standing in middle, facing stall door Occasional steps
Moves to door
Looks at door
Moves to observer, ears forward Freely lifts feet when asked
Response to grain
Moves to door and reaches for grain
Looks at observer, ears forward Lifts feet after mild encouragement Looks at door
Occasional At withers Slightly back, little movement Standing in middle, facing sides of stall
4 Continuous Below withers Standing in middle, facing back of stall No movement No response
Moves away from observer
Does not move, ears back Extremely unwilling to lift feet No response
Fig. 1. Scatter plot of composite pain scores (CPS) (A), assessed by two different observers at the same time (n = 40, some dots represent two or more combinations of scores), r = 0.84 (P < 0.001). Bland–Altman plot (B) of same 40 CPS combinations (limits of agreement: 3.3 to +3.1).
Fig. 2. Median ± 25–75th percentiles of composite pain scores (CPS) and numerical rating scale (NRS) scores (A + C) of horses that left the intensive care unit (ICU) alive (n = 39) and of horses that did not survive ICU hospitalisation (n = 9). T, hours after recovery from general anaesthesia for emergency gastrointestinal surgery. Median area under the curve (AUC) scores of survivors (B) and non-survivors (D). Bold lines show median scores, boxes show 25–75th percentiles, error bars show 10–90th percentiles. P < 0.001.
parameters, except for ‘sweating’, showed a significant correlation with total CPS score (P < 0.001). ‘Pawing on the floor’, ‘appearance’, ‘head movements’ and ‘interactive behaviour’ all obtained correlation coefficients with total CPS scores of >0.5.
Discussion Simple visual analogue scale (VAS) scores have been described as reliable, sensitive and very suitable for pain assessment in
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115
Fig. 3. Median area under the curve (AUC) values for Warmbloods (n = 30) and nonWarmbloods (n = 18) (A). Median AUC values for horses with small intestinal (n = 25) and large intestinal disorder (n = 23) (B). Median AUC values for horses that had a strangulating intestinal lesion (n = 14 survivors, n = 6 non-survivors) (C). Median AUC values for horses that did not have a strangulating intestinal lesion (n = 26 survivors, n = 3 non-survivors) (D). Bold lines show median scores, boxes show 25–75th percentiles, error bars show 10–90th percentiles. P < 0.05, P < 0.001.
Fig. 4. Correlation coefficients between 13 individual composite pain scores (CPS) parameters and the total composite pain score (261 CPS evaluations from n = 17 animals), P < 0.001.
verbal humans after day surgery, if the patients can score themselves (Coll et al., 2004). In neonates and nonverbal children as well as in older adults with dementia, behavioural and composite pain scales have been assessed and cross-validated for clinical use (Suraseranivongse et al., 2001, 2006; Herr et al., 2006; Ersek et al., 2011). It was hypothesised that such scales might also be applicable to horses. The CPS was developed for orthopaedic pain in horses by Bussières et al. (2008) and later applied in horses with somatic and visceral pain by van Loon et al. (2010). The current study demonstrates the usefulness of the CPS for the objective, reliable and reproducible evaluation of pain in horses after emergency gastrointestinal surgery. This was further supported by the low inter-observer variability and the clinically acceptable limits of agreement found in Bland–Altman analysis, combined with the lack of bias (average discrepancy between two observers). The specificity and sensitivity for differentiating between pain-free healthy horses and ICU horses and between
113
ICU horses that responded well to treatment and ICU horses that developed complications leading to euthanasia or re-laparotomy also adds to the practical usefulness of the CPS for this category of horses. The composite pain scale by Bussières et al. (2008) combines physiological, behavioural and interactive variables. Although originally developed for orthopaedic pain, the scale contains several elements that could also be expressions of acute visceral pain. If this pain scale was applicable to visceral pain, this would be convenient, as one pain scale could then be used for both orthopaedic and visceral pain in horses. In the original study by Bussières et al. (2008) the key specific and most sensitive behavioural indices were response to palpation of the painful area, posture and, to a lesser extent, pawing on the floor, kicking at the abdomen and head movements. For visceral postoperative pain, pawing on the floor, overall appearance, head movements and interactive behaviour appear to be the most important elements. The original study (Bussières et al., 2008) and our current results show that the CPS can be used for the assessment of pain intensity in both orthopaedic and visceral types of pain, for each of which there may be a subset of most sensitive parameters. The specificity and sensitivity of the CPS could along this line potentially be increased by introducing different weighting factors for individual variables or non-specific parameters could be removed from the definite CPS scale. A shortened version of the CPS for clinical use could consequently consist of only the best correlating individual parameters (pawing on the floor, appearance, head movements and interactive behaviour) that all showed correlations with total CPS scores of >0.5. Further analysis of this shortened version of the CPS was outside the scope of the current study. One of the limitations of our study is that a validated ‘gold standard’ pain score for validation of the CPS for postoperative visceral pain is lacking. As the NRS (Pritchett et al., 2003) has not been fully validated, it is debatable whether this pain scale can be used to meet this objective. Because of this lack of full validation and because of only moderate inter-observer reliability for the NRS that we found in our study, we decided not to use the NRS for validation of the CPS. The NRS has been used for assessment of pain in horses after exploratory celiotomy for colic and in healthy pain-free horses and horses that were subjected to general anaesthesia, and was shown to be able to discriminate between healthy painfree horses and those with visceral postoperative pain (Pritchett et al., 2003). For this reason, the NRS was considered the most suitable scale to compare the CPS with, when assessing the suitability of the latter for the assessment of visceral pain. The NRS has been used in two subsequent studies. In a study in horses after emergency celiotomy, Sellon et al. (2004) found decreased NRS scores in animals treated with flunixin and butorphanol compared to horses that were treated with flunixin only. These observations corresponded with lower plasma cortisol levels, reduced weight loss and faster recovery from surgery in the test group. These results show the sensitivity of the NRS to detect differences in horses that underwent different treatment protocols after emergency celiotomy. Sanz et al. (2009) did not find significant differences in NRS scores in horses undergoing routine castration after being treated with different analgesic protocols. Although the NRS has been shown to be useful in assessing pain in horses after exploratory celiotomy, we believe the CPS has several advantages over the NRS. First, the NRS does not include physiological parameters (although correlations between physiological variables and the NRS were found in the earlier mentioned studies). Second, one of the nine parameters of the NRS is ‘gross pain behaviour’, which is largely subjective. The CPS does not contain subjective criteria and its objectivity and more reproducible results are supported by the high inter-observer reliability of the CPS in our study. Although it was not assessed in the original NRS studies,
114
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115
we found only moderate inter-observer reliability for the NRS, possibly caused by the influence of the subjective gross pain behaviour parameter. Finally, the CPS has been shown to be useful for both orthopaedic and visceral pain in horses, making it a widely applicable universal tool in clinical practice. Sweating was the only variable that did not correlate with total CPS score. Bussières et al. (2008) previously concluded that sweating was a non-sensitive parameter for orthopaedic pain either. Recently, Graubner et al. (2011) described a post abdominal surgery pain assessment scale (PASPAS) in horses. This scale combines behavioural (observational and interactional) with physiological parameters, as in the CPS. Inter-observer reliability was sufficient and PASPAS scores were not influenced by general anaesthesia in healthy animals. Due to the lack of a ‘gold standard’ pain score, the authors used the differences in pain scores between horses with painful complications (ileus, laminitis, colitis, early wound infections) and horses without complications (between which the scores significantly differed) to validate the composite pain scores. A disadvantage of this scale compared to the CPS is that, like the NRS, it contains a general subjective assessment of pain. Further, the PASPAS score has specifically been designed for the assessment of abdominal postsurgical pain and has not (yet) been extended to horses with orthopaedic pain. Comparison of the PASPAS-score with both NRS and CPS could possibly lead to further optimisation and fine-tuning of scoring systems for equine (visceral) pain assessment. In surveys among equine veterinarians in the UK (Price et al., 2002) and in The Netherlands and Belgium (Dujardin and van Loon, 2011), heart rate was cited as one of the major criteria used to assess pain. However, no objective study demonstrating a positive correlation between heart rate and pain in horses has been published so far. Moreover, several studies have reported that increased heart and respiratory rates are not specifically indicative of postoperative pain in cats and dogs (Hansen et al., 1997; Holton et al., 1998). Heart rate is likely to be influenced by many factors other than pain, including endotoxaemia and hypovolaemia (Sellon et al., 2004). Despite its limited reliability as a single indicator, heart rate is nevertheless often incorporated in various composite pain scales. Bussières et al. (2008) found moderate specificity and sensitivity for heart rate, which is in accordance with the moderate correlation between heart rate and total CPS scores we found in the current study. In human infants after surgery, similar observations have been made (van Dijk et al., 2001). The ultimate aim of pain scales is to provide a reliable tool for guidance of analgesic treatment. For an appropriate analgesic strategy, the pain scale not only needs to be specific for pain, but ideally should indicate objective cut-off values as well. Gerbershagen et al. (2011) described four different methods for cut-off point analysis in humans. Parameters such as verbal expression of pain state, subjective tolerable pain intensity, satisfaction with pain therapy, pain-related interference with mood and estimation of pain thresholds are used to achieve this (Gerbershagen et al., 2011). In nonverbal animals, determining such cut-off values is, to say the least, challenging.
Conclusions CPS was shown to allow objective discrimination between horses that recovered well after emergency gastrointestinal surgery and horses that developed painful complications and did not survive the ICU period. In other words, pain scores significantly correlated with final clinical outcome. The use of the CPS improved objectivity in pain management and, on account of a high inter-observer reliability over a range of CPS scores, enabled comparisons between different observers. This would be of major benefit in
larger veterinary hospitals where several attending clinicians are often involved in the treatment of each case. Conflict of interest statement None of the authors of this paper has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper. Acknowledgements We would like to thank Judith Weening, Astrid Rijckaert, Esther van Neste, Marjolijn Beugelsdijk and Marjolijn Heemskerk for their help with data collection. References Abbott, F.V., Franklin, K.B.J., Westbrook, R.F., 1995. The formalin test: Scoring properties of the first and second phases of the pain response in rats. Pain 60, 91–102. Bland, J.M., Altman, D.G., 1986. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 8, 307–310. Bussières, G., Jacques, C., Lainay, O., Beauchamp, G., Leblond, A., Cadoré, J.L., Desmaizières, L.M., Cuvelliez, S.G., Troncy, E., 2008. Development of a composite orthopaedic pain scale in horses. Research in Veterinary Science 85, 294–306. Büttner, W., Finke, W., 2000. Analysis of behavioural and physiological parameters for the assessment of postoperative analgesic demand in newborns, infants and young children: A comprehensive report on seven consecutive studies. Pediatric Anesthesia 10, 303–318. Coll, A.M., Ameen, J.R.M., Mead, D., 2004. Postoperative pain assessment tools in day surgery: Literature review. Journal of Advanced Nursing 46, 124–133. Dobromylskyj, P., Flecknell, P.A., Lascelles, B.D., Livingston, A., Taylor, P., WatermanPearson, A., 2000. Pain assessment. In: Flecknell, P.A., Waterman-Pearson, A. (Eds.), Pain Management in Animals. W.B. Saunders, Philadelphia, Pennsylvania, USA, pp. 53–79. Dujardin, C.L.L., van Loon, J.P.A.M., 2011. Pain recognition and treatment in the horse: A survey of equine veterinarians in the Netherlands and Belgium. Tijdschrift voor Diergeneeskunde 136, 715–724. Ersek, M., Polissar, N., Neradilek, M.B., 2011. Development of a composite pain measure for persons with advanced dementia: Exploratory analyses in selfreporting nursing home residents. Journal of Pain and Symptom Management 41, 566–579. Flecknell, P., 2008. Analgesia from a veterinary perspective. British Journal of Anaesthesia 101, 121–124. Gerbershagen, H.J., Rothaug, J., Kalkman, C.J., Meissner, W., 2011. Determination of moderate-to-severe postoperative pain on the numerical rating scale: A cut-off point analysis applying four different methods. British Journal of Anaesthesia 107, 619–626. Graubner, C., Gerber, V., Doherr, M., Spadavecchia, C., 2011. Clinical application and reliability of a post abdominal surgery pain assessment scale (PASPAS) in horses. The Veterinary Journal 188, 178–183. Hansen, B.D., Hardie, E.M., Carroll, G.S., 1997. Physiological measurements after ovariohysterectomy in dogs: What’s normal? Applied Animal Behaviour Science 51, 101–109. Herr, K., Bjoro, K., Decker, S., 2006. Tools for assessment of pain in nonverbal older adults with dementia: A state-of-the-science review. Journal of Pain and Symptom Management 31, 170–192. Hewson, C.J., Dohoo, I.R., Lemke, K.A., 2006. Perioperative use of analgesics in dogs and cats by Canadian veterinarians in 2001. Canadian Veterinary Journal 47, 352–359. Hewson, C.J., Dohoo, I.R., Lemke, K.A., Barkema, H.W., 2007a. Canadian veterinarians’ use of analgesics in cattle, pigs and horses in 2004 and 2005. Canadian Veterinary Journal 48, 155–164. Hewson, C.J., Dohoo, I.R., Lemke, K.A., Barkema, H.W., 2007b. Factors affecting Canadian veterinarians’ use of analgesics when dehorning beef and dairy calves. Canadian Veterinary Journal 48, 1129–11136. Holton, L.L., Scott, E.M., Nolan, A.M., Reid, J., Welsh, E., 1998. Relationship between physiological factors and clinical pain in dogs scored using a numerical rating scale. Journal of Small Animal Practice 39, 469–474. Hugonnard, M., Leblond, A., Keroack, S., Cadore, J., Troncy, E., 2004. Attitudes and concerns of French veterinarians towards pain and analgesia in dogs and cats. Veterinary Anaesthesia and Analgesia 31, 154–163. Laven, R.A., Huxley, J.N., Whay, H.R., Stafford, K.J., 2009. Results of a survey of attitudes of dairy veterinarians in New Zealand regarding painful procedures and conditions in cattle. New Zealand Veterinary Journal 57, 215–220. Lerche, P., 2009. Assessment and treatment of pain in horses. Equine Veterinary Education 21, 44–45. Lindegaard, C., Thomsen, M.H., Larsen, S., Andersen, P.H., 2010. Analgesic efficacy of intra-articular morphine in experimentally induced radiocarpal synovitis in horses. Veterinary Anaesthesia and Analgesia 37, 171–185.
J.P.A.M. van Loon et al. / The Veterinary Journal 200 (2014) 109–115 Molony, V., Kent, J.E., 1997. Assessment of acute pain in farm animals using behavioural and physiological measurements. Journal of Animal Science 75, 266–272. Myles, P.S., 2007. Using the Bland–Altman method to measure agreement with repeated measures. British Journal of Anaesthesia 99, 309–311. Price, J., Marques, J.M., Welsh, E.M., Waran, N.K., 2002. Pilot epidemiological study of attitudes towards pain in horses. Veterinary Record 151, 570–575. Pritchett, L.C., Ulibarri, C., Roberts, M.C., Schneider, R.K., Sellon, D.C., 2003. Identification of potential physiological and behavioural indicators of postoperative pain in horses after exploratory celiotomy for colic. Applied Animal Behaviour Science 80, 31–43. Sanz, M.G., Sellon, D.C., Cary, J.A., Hines, M.T., Farnsworth, K.D., 2009. Analgesic effects of butorphanol tartrate and phenylbutazone administered alone and in combination in young horses undergoing routine castration. Journal of American Veterinary Medical Association 235, 1194–1203. Sellon, D.C., Roberts, M.C., Blikslager, A.T., Ulibarri, C., Papich, M.G., 2004. Effects of continuous rate intravenous infusion of butorphanol on physiologic and outcome variables in horses after celiotomy. Journal of Veterinary Internal Medicine 18, 555–563. Suraseranivongse, S., Santawat, U., Kraiprasit, K., Petcharatana, S., Prakkamodom, S., Muntraporn, N., 2001. Cross-validation of a composite pain scale for preschool children within 24 hours of surgery. British Journal of Anaesthesia 87, 400–405.
115
Suraseranivongse, S., Kaosaard, R., Intakong, P., Pornsiriprasert, S., Karnchana, Y., Kaopinpruck, J., Sangjeen, K., 2006. A comparison of postoperative pain scales in neonates. British Journal of Anaesthesia 97, 540–544. Valverde, A., Gunkel, C.I., 2005. Pain management in horses and farm animals. Journal of Veterinary Emergency and Critical Care 15, 295–307. Van Dijk, M., de Boer, J.B., Koot, H.M., Duivenoorden, H.J., Passchier, J., Bouwmeester, N., Tibboel, D., 2001. The association between physiological and behavioural pain measures in 0- to 3-year-old infants after major surgery. Journal of Pain and Symptom Management 22, 600–609. Van Loon, J.P.A.M., Back, W., Hellebrekers, L.J., van Weeren, P.R., 2010. Application of a composite pain scale to objectively monitor horses with somatic and visceral pain under hospital conditions. Journal of Equine Veterinary Science 30, 641– 649. ˇ uela-Fernández, I., Jones, E., Welsh, E.M., Fleetwood-Walker, S.M., 2007. Pain Vin mechanisms and their implication for the management of pain in farm and companion animals. The Veterinary Journal 174, 227–239. Williams, V.M., Lascelles, B.D., Robson, M.C., 2005. Current attitudes to, and use of, peri-operative analgesia in dogs and cats by veterinarians in New Zealand. New Zealand Veterinary Journal 53, 193–202.
