The Science of Measuring Perioperative Pain
American Association of Zoo Veterinarians Conference 2012
Mike Conzemius, DVM, PhD, DACVS
University of Minnesota, St. Paul, MN, USA


Pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage.1 If we accept that, like ourselves, our patients experience pain to some degree, then we can move forward with the statement that one goal of veterinarians is to manage pain. However, pain management has consistently been a problem in our profession. In 1993 it was reported that after major surgery in a large referral hospital in the USA, less than 7% of cats received any postoperative analgesia and only 19% of dogs received analgesia for more than eight hours.2 Unfortunately, the limited use of analgesics for veterinary surgical patients is not limited to the U.S. In 1999 30% of British veterinarians agreed with the statement, “A degree of pain is required to stop the animal being too active postsurgery.”3 Similarly, veterinarians in Canada indicated that just less than half administered analgesics to surgical patients.4

Why not provide analgesics? Is it because our patients cannot verbally communicate? Is it because of our attitude towards animals or, as some have suggested, is it because of a lack of understanding and education regarding pain and analgesics? One would hope that the papers published in the 1990s would have stimulated a push towards more education for veterinary students. It may have in some institutions, but in a recent paper two-thirds of health science programs in Canada were unable to identify specific hours designated for the topic of pain in their programs.5

A lack of the understanding of pain is a major obstacle. We cannot specifically measure it and, therefore, there is little positive feedback if we treat it successfully or little negative feedback if we fail to treat it. Perhaps pain cannot be treated scientifically. One veterinarian said, “Pain is an experience that does not lend itself to objective measurements, so the art of medicine should not be overlooked in favor of the science of medicine.”6 I agree, in part. Certainly, we need to utilize our intuition and creativity when it comes to the topic of animal pain. However, instead of using art to diagnose and treat pain we should use it to design scientific experiments that allow us to study pain, analgesia, and analgesics. Frankly, art is open to interpretation; only reproducible, scientific experiments will provide progress.

Science in the field of animal pain has been prevalent. In a search of “animal pain” on PubMed one finds nearly 45,000 scientific papers. So, what is the problem? With that much research, it should be perfectly clear how to measure animal pain and how to test various analgesics. The vast majority of these studies in small animal veterinary patients utilize patient behavior as a mechanism to measure pain. In a paper I wrote comparing analgesic protocols after intercostal thoracotomy we used a numerical rating scale (NRS) that evaluated patient behaviors such as patient crying, agitation, and movement.7 This is nice, but as I reflect on that work I wonder, how accurately can I separate mild from moderate agitation in a dog that is recovering from general anesthesia and major surgery? In an effort to provide some objective outcome measures in this research we combined the NRS with postoperative changes in heart rate and respiratory rate. At first glance this makes sense because we know that these biologic indices increase in the presence of pain. They are also affected by analgesics, anesthetics, stress, and fear. So, did we measure pain in this study? In a later study we evaluated data that compared the relationship between changes in subjective (behaviors) and objective (HR, RR, pain threshold) measures of pain and found no correlation between the two.8 Thus, one of our outcome measures may have been an effective tool for estimating patient pain; but if one was the others were not.

Since this, subjective measures to estimate animal pain have matured. In a series of papers from a research group in Glasgow, the Glasgow Composite Pain Scale (GCPS) was developed.9 This was shortly followed with a modified version. The GCPS is widely accepted as a validated, gold standard for analgesic studies using subjective outcome measures in small animal patients.10 The GCPS was validated in a study of 20 clinically normal dogs, 20 dogs with medical conditions, and 117 dogs undergoing surgery. After a scaling model was applied to the descriptors to establish weights for each and create a continuous scale, five observers independently used the scale to score signs of pain in four groups of dogs (control dogs, dogs with medical conditions, and 40 dogs undergoing soft tissue or orthopedic surgery). Scores from each group and from groups of conditions perceived to cause no, mild, moderate, and severe pain were compared. In addition, the scale was applied to 77 dogs undergoing orthopedic or soft tissue surgery and scores were compared with simultaneously derived numeric rating scale (NRS) scores; comparisons were made between surgical groups and with time after surgery. They found that median pain scores differed significantly among the four study groups, among pain severity groups, and were typically greater with increasing perceived pain severity. From this they concluded that the measurement scale is a valid measure of acute pain in dogs. This is a step in the right direction. Validation using these methods is made by demonstrating that the scale performed as expected when used to evaluate patient pain after surgeries with various degrees of invasiveness. Validation in this sense is good, and this technique is easily applied in every hospital but it seems that there is still room for improvement when it comes to making strides towards the scientific study of analgesia and analgesics.

Another approach if we cannot directly measure pain is to measure something the patient does because of pain that can be objectively measured. One common approach is to measure lameness after inflammation is induced. This is done in a research setting by using a urate crystal synovitis model. Pain can be estimated by measuring limb function. Limb function can be measured by visual observation, but this would defeat the purpose by introducing opinion. Limb function can be precisely and objectively measured using gait analysis by measuring ground reaction forces (GRF). This technique has been widely instituted and is the preclinical gold standard to determine if an analgesic or anti-inflammatory drug performs better than a placebo medication or at least as good as a drug that has already demonstrated efficacy. Numerous evaluations of nonsteroidal anti-inflammatories (NSAID) can be found in the literature. Limb function can also be measured in a clinical setting using gait analysis after surgery. If the methods of the study are set that all groups receive identical treatment (premedication, anesthetic, surgery, and surgeon), perioperative analgesic techniques can be tested. This technique was demonstrated when use of a perioperative NSAID was found to improve patient use of an operated leg after cranial cruciate ligament surgery as compared to identical treatment but no use of a NSAID in the analgesic regimen.11 Another great benefit to this technique is that patient pain can also be estimated in cats. Using client-owned cats, limb function was measured in cats after unilateral declaw to effectively study various analgesic techniques (always remember to use a multimodal approach to the treatment of pain) and surgical techniques.12,13 Although this would not normally be done in a clinical setting, unilateral declaw gave the patient a choice to use the operated leg or to simply walk on three legs. To accept these models, we must accept that patients will use the operated leg more if they are less painful. Intuitively, I think this makes sense. Finally, measuring GRF is a well-documented way to estimate chronic joint pain in patients and has been used to document the efficacy of both medications and surgical procedures.14,15

Beyond gait analysis there are other intuitive, objective measures of pain. When performing a physical exam, we commonly look for and try to semi-quantify patient pain. Examples of this might be checking for back pain by pressing on the spinal column, abdominal pain by pressing on the abdomen, or joint pain by pressing on the joint capsule. These clinical techniques that we use every day in practice can be improved and translated into a more objective technique by using a standardized instrument that measures the amount of pressure applied before the patient’s first negative response (pain threshold). Pain threshold algometers are commercially available and have been used in clinical veterinary studies. In a study evaluating the efficacy of various analgesic techniques after a standardized knee surgery, an algometer was used to compare groups.16 This study allowed for documentation that a single, postoperative injection of intra-articular bupivacaine provided better analgesia than intra-articular morphine or intra-articular saline. One limitation to this technique is that it requires that the observer interpret the patient’s first negative response; having performed this I always wonder, is this threshold when the patient changes their breathing pattern, stops wagging their tail, or when they want to bite you?

Previously I mentioned that the biologic variables HR and RR were ineffective in measuring pain because they were confounded by too many physical and environmental factors. Until recently, I thought this might be the case with all biologic variables. In a recent study of acute pain in dogs induced by urate crystal synovitis, changes in patient serum cortisol was studied.17 All patients were acclimated to the environment, all had cortisol and GRF measured before synovitis, and all had cortisol and GRF after synovitis. Before the synovitis, cortisol and GRF remained unchanged over the course of the 24-hour study. After synovitis, GRF decreased and cortisol proportionally increased. In fact, in this study a patient serum cortisol level greater than 1.9 mg/ml indicated lameness with 90% sensitivity. I would suggest that this validates the use of serum cortisol as an objective measure of pain. Use of cortisol has been used in clinical studies evaluating patient pain in the past. In one study, dogs that had pericardiectomy performed via thoracoscopy had lower cortisol levels than dogs that had it performed via open thoracotomy.18 It seems prudent to pay close attention to studies that utilize validated, objective estimates of pain.

Obvious limitations with verbal communication with our patients leave the interpretation of the signs of pain in animals to the opinion of the observer. In a clinical setting my approach is straightforward—surgery causes pain; therefore, I should provide analgesia. If a patient appears that it may need more analgesic and it is safe to provide one, I do. When it comes to selection of analgesics, I try to cautiously balance the use of my clinical experience with the peer-reviewed scientific literature that was generated using objective outcome measures or subjective measures that had large patient groups.

Literature Cited

1.  Wall PD. Defining “pain in animals.” In: Short CE, Poznak AV, eds. Animal Pain. New York, NY: Churchill Livingstone; 1992;63–79.

2.  Hansen B, Hardie E. Prescription and use of analgesics in dogs and cats in a veterinary teaching hospital: 258 cases (1983–1989). JAVMA. 1993;202:1485–1494.

3.  Capner CA, Lascelles BDX, Water-Pearson AE. Current British veterinary attitudes to perioperative analgesia for dogs. Vet Rec. 1999;145:95–99.

4.  Dohoo SE, Dohoo IR. Postoperative use of analgesics in dogs and cats by Canadian veterinarians. Can Vet J. 1996;37:546–551.

5.  Watt-Wattson J, et al. A survey of prelicensure pain curricula in health science faculties in Canadian universities. Pain Res Manage. 2009;14(6):439–444.

6.  Hellyer PW. Contradictions characterize pain management in companion animals. Accessed January 7, 2011. (VIN editor: link was not accessible as of 12/17/2020.)

7.  Conzemius MG, Brockman DJ, King LG, et al. Analgesia in dogs after intercostal thoracotomy: a clinical trial comparing intravenous buprenorphine and interpleural bupivacaine. Vet Surg. 1994;23:291–298.

8.  Conzemius MG, Sammarco JL, Perkowski SZ, Hill CM. Correlation between subjective and objective measures used to determine severity of postoperative pain in dogs. JAVMA. 1997;210(11):1619–1622.

9.  Morton CM, Reid J, Scott EM, Holton LL, Nolan AM. Application of a scaling model to establish and validate an interval-level pain scale for assessment of acute pain in dogs. AJVR. 2005;66(12):2154–2166.

10.  Murrell JC, Psatha EP, Scott EM, Reid J, Hellebrekers LJ. Application of a modified form of the Glasgow pain scale in a veterinary teaching centre in the Netherlands. Veterinary Record. 2008;162:403–408.

11.  Horstman CL, Conzemius MG, Evans R, Gordon WJ. Assessing the efficacy of perioperative oral carprofen after cranial cruciate surgery using noninvasive, objective pressure platform gait analysis. Vet Surg. 2004;33(3):286–289.

12.  Romans CW, Gordon WJ, Robinson DA, et al. Effect of postoperative analgesic protocol on limb function following onychectomy in cats. JAVMA. 2005;227(1):89–93.

13.  Robinson DA, Romans CW, Gordon-Evans WJ, Evans RB, Conzemius MG. Evaluation of short-term limb function following unilateral carbon dioxide laser or scalpel onychectomy in cats. JAVMA. 2007;230(3):353–358.

14.  Hanson PD, Brooks KC, Case J, Conzemius M, et al. Efficacy and safety of firocoxib in the management of canine osteoarthritis under field conditions. Veterinary Therapeutics. 2006;7(2):127–140.

15.  Conzemius MG, Aper RL, Corti LB. Short-term outcome after total elbow arthroplasty in dogs with severe naturally occurring osteoarthritis. Vet Surg. 2003;32:545–552.

16.  Sammarco JL, Conzemius MG, Perkowski SZ, Weinstein MJ, Gregor TP, Smith GK. Analgesic effect of intra-articular bupivacaine, morphine, or saline after cranial cruciate repair. Vet Surg. 1996:59–69

17. Feldsein JD, Wilke VL, Evans RB, Conzemius MG. Serum cortisol concentration and force plate analysis in the assessment of pain associated with sodium urate-induce acute synovitis. AJVR. 2010;71(8):940–945.

18.  Walsh PJ, et al. Thoracoscopic versus open partial pericardiectomy in dogs: comparison of postoperative pain and morbidity. Vet Surg. 1999;28(6):472–479.


Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Mike Conzemius, DVM, PhD, DACVS
University of Minnesota
St. Paul, MN, USA

MAIN : AAZV Conference : Science of Measuring Perioperative Pain
Powered By VIN