Clinical assessment of pain in dogs and cats--use of pain scales (Proceedings)

It is apparent that the primary, if not only, reason to assess pain is to take action to relieve pain. However, one of the most challenging responsibilities of veterinary professionals is pain assessment. Our patients are nonverbal, non-self-reporting by typical means, and are diverse in degree of domestication, species, breed, and individual characteristics. Assessment must be performed by surrogates or proxy (family members and medical professionals). Surrogate assessment is fraught with challenges: observer subjectivity, inter-observer variability, observer preconceived perception and assumption, lack of standardized intervention points, among others.

Despite this we have a professional and ethical responsibility to strive toward accuracy in assessment for the purposes of minimizing iatrogenic causes of pain, relieving existing pain, and measuring the impact of our interventions.

A few key concepts

In order to begin to assess pain in another, one must have a clear understanding of the definition of pain. A common definition frequently quoted comes from the International Association for the Study of Pain (IASP) IASP “pain is an unpleasant sensory and emotional experience... “

This rather academic definition is not particularly useful on the clinic floor. McCaffery provides an operational definition of pain: “Pain is whatever the experiencing ‘person' say it is, existing whenever he says it does.”  Fine, but what about the nonverbal patient: human babies, human adults suffering from dementia, aphasia, amyotrophic lateral sclerosis (Lou Gehrig's disease), or animals?

When dealing with nonverbal patients, pain is what WE say it is. This is a daunting responsibility. Getting it wrong has profound implications. Imagine that your comfort is in the hands of someone who neither speaks your language nor understands what it is to be human; you are at the mercy of that individual.  That well meaning individual could under treat your pain and leave you in agony or over treat your pain leaving you groggy, dysphoric, nauseous, etc. Assessment is the only way your caregiver can effectively and dynamically manage your comfort.

Pain is complex. Pain physiology occurs in the body; noxious stimuli set off an intricate web of chemical signals and responses. Pain perception occurs in the brain (amygdala, cerebral cortex, cingulate gyrus). So although significant physiologic responses to noxious stimuli are occurring along the pain pathway, an individual does not experience pain without higher centers registering the incidents. Variability in the cognitive and emotional awareness of/response to pain is defined as tolerance. While variability in the physiologic registering of a stimulus as noxious is defined as threshold. Threshold is the least experience of pain, which a subject can recognize (i.e. a noxious stimulus is interpreted as painful). This new definition from the International Association for the Study of Pain (IASP) emphasizes experience of the stimulus rather than the stimulus itself.

Tolerance, on the other hand, is the greatest level of actual pain, which a subject is prepared to endure. The degree to which either can be manipulated is not full understood. It is clear that contrary to common perception, constant exposure to a painful stimulus does not necessarily create tolerance and may in fact lower threshold worsening the current and future pain experiences. It is also clear that tolerance has a genetic link, and therefore is conceivably beyond the control of an individual (i.e. patients are not “whimpy” as a consequence of their own frailty or lack of toughness). Pain management seeks to diminish the painful stimuli or raise an individual's pain threshold or at least improve their ability to tolerate pain.

If pain equaled nociception, then the same procedure would produce the same pain experience from individual to individual. If pain is complex; if pain is a tangled web of physiology, cognition, and emotion; if pain is what WE say it is because our veterinary patients cannot speak; how do we begin to effectively recognize and treat pain. The principle of analogy serves us well in this regard:Most mammals have anatomic and physiologic pain pathways sufficiently similar to humans that we        can assume what produces pain in a human must undoubtedly produce pain in an animal. The amount of pain and coping ability can be argued for any individual, human or nonhuman, but the principle holds. Assume pain and treat accordingly. Use effective assessment tools to monitor response to therapy

Pain scales as assessment tools           

Assessment tools have been used for centuries in evaluating pain in humans. In recent decades emphasis on validating these tools has lead to an increase in awareness and accuracy. Validity is a measure of the quality of a scale or index; it is an evaluation of the tool's ability to measure what it is intended to measure. There are at least 4 types of validity involved in overall validation of a pain scale: 1) construct validity is used to compare individuals with a given type of pain to healthy individuals; 2) criterion validity is used when describing the correlation between a scale and another external measurement of the same phenomenon; 3) face validity is the extent to which the scale, after it is constructed, is subjectively viewed by individuals knowledgeable in the area of pain and pain assessment; 4) content validity is related to face validity and asks whether the scale is comprehensive, covering all of the generally accepted variables.

With improved validation methods it has become clear that one scale does not fit all circumstances: acute pain scales are not adequate for chronic pain; neonatal scales are not effective in assessing adults; post operative pain is not measured the same way as pain associated with a medical condition such as pancreatitis. The same can be said for veterinary patients: any tool used must address species, circumstance, injury or illness. Validation is an arduous process of psychometrics. At the time of this writing there are few validated veterinary scales; fortunately the numbers are growing.

The advantages of using an appropriately specific veterinary pain scale include: 1) elevated awareness of pain and pain inducing procedures; 2) method for quantifying pain; 3) potential for comparison from moment to moment; 4) basis from which to assess the effect of therapeutic interventions. The limitations of veterinary pain scales include: 1) requirement for species specificity; 2) many current acute scales are developed for post operative patients and not appropriate for nonsurgical pain; 3) very little work done thus far on chronic pain and quality of life assessment; 4) subjectivity; 5) inflexibility; 6) over reliance on physiologic measures can be problematic 7) no gold standard (i.e. no biomarkers for pain).

What should we measure?

The use of physiologic parameters in assessing pain seems logical. A body undergoing stress (pain) will respond in a physiologically appropriate way: increased respiratory rate, increased heart rate, increased blood pressure, pupil dilation, increased plasma cortisol and so on. Unfortunately these measures are highly inaccurate because they are generic responses to many conditions: fear, anxiety, sedation, cardiovascular or metabolic conditions.  This is not to say that other conditions do not co-exist with the experience of pain. It is to say that current studies demonstrate that we cannot parse out the specific pain condition and response to therapy through measurement of physiologic parameters except, perhaps, in the very acute setting. These peracute situations only represent the tiniest proportion of pain scenarios we veterinary professionals encounter. What are we to do the rest of the time?

If we are paying attention at all, we, veterinary professionals, consciously or subconsciously register the comfort or discomfort of our patients all day every day. So what precisely leads us, as we walk by the kennel, to determine whether we should provide more analgesia, or less? It seems to be our ability to assess behavioral changes. And indeed, this has been shown to be what guides our human medicine colleagues as they treat nonverbal human patients. This ability to recognize behavioral change is grounded in our knowledge of the species we treat. Over years of experience we have some general notion of what is “normal” behavior for a given species. Our clients have an understanding of what is “normal” behavior for their individual animal. The trick is to recognize, name, define, and capture our experiential knowledge in a validated scale that is repeatable and leads to appropriate and effective therapy.

Types of scales

One-dimensional acute pain scales are based on an instantaneous impression of the patient suffering from an acute condition. They are include the

1.        Simple descriptive scale (SDS), a series of words: no pain, mild, moderate, severe, very severe, and unbearable, sometimes accompanied by corresponding pictures.

2.        Numeric rating scale (NRS), a series of integers on a line from 0 to 10 in which 0 is no pain and 10 is the worst pain imaginable. This provides a total of 11 choices.

3.        Visual analog scale (VAS), a 100mm line delineated at one end with no pain and the other with worst pain imaginable, without numbers. A mark is placed on this line to delineate level of pain. A measure from no pain to the mark is converted to a number out of a possible 100. This scale provides a total of 100 choices.

4.        Dynamic interactive visual analog scale (DIVAS), the same methodology as VAS however it includes interactive and non-interactive assessment period as well as assessment of the patient in motion.

These acute one-dimensional scales have been adapted for use in the chronic pain patient. The scales described are found to have a high degree of inaccuracy for non-self reporting patients. These subjective scales have not been validated in veterinary species. There is a high degree of inter and intra-observer variability, as such they cannot be recommended for use in veterinary patients. It is interesting to consider, however, what guides the choice when these scales are used. It appears to be the critical assessment of various behavior changes.

Multi-dimensional acute pain scales are based on assessment of various behavioral indicators. The only validated scale in this category is

1.        Glasgow composite Measures Pain Scale (GMPS) is a series of questions relating to interactive and non-interactive behaviors. The composite score is obtained based on the numeric responses to these questions. A short form (2 pages) is available; a feline version is in process of validation. The GMPS is currently the most reliable scale for acute pain assessment in dogs only. Disadvantages include being specific for surgical pain and does not include suggested points for analgesic intervention.

Non-validated scales in this category include

1.        University of Melbourne pain scale (UMPS) is a series of behavior-based questions combined with physiologic parameter assessment.  It is clinically practical and designed specifically for use in dogs; disadvantages include the inclusion of physiologic parameters, lack of content validation, and lack of suggested points for analgesic intervention.

2.        4-A-VET postoperative pain scale is an acute pain scale based on subjective overall rating of 6 behavioral categories. Advantages include multispecies application (dogs and cats, also a disadvantage) and suggested points for analgesic intervention. The disadvantages include lack of information on category selection, poor reproducibility, and lack of species specificity.

3.        Colorado State University Acute Pain Assessment teaching tool is designed to help train veterinary professionals in pain assessment by raising awareness of species and pain-specific behavioral changes. It is not a validated assessment scale and was not intended to be such. The best application of this tool is in developing students' observational skills and interpretive capacity based on their growing clinical knowledge.

Multi-dimensional chronic pain scales are based on assessment of various behavioral indicators specific for patients experiencing long-standing pain syndromes. The validated canine scales in this category are

1.        Helsinki Chronic Pain Index (HCPI) is an easy to use tool that can provide owners with a mechanism for participation in care of their animal. The disadvantages are that the score is a simple descriptive numeric rating and therefore not sensitive, and that it is based on the current moment of assessment rather than a composite of recent days.

2.        Canine Brief Pain Index (C-BPI) is based on a composite assessment of a week and therefore can be used to assess response to therapy. Unfortunately, it relies on clients' understanding of pain based behavioral change. Training of the client may obviate this concern.

3.        Cincinnati Orthopedic Disability Index (CODI) can be used to assess therapeutic interventions, evaluates not only chronic pain, but also disability (impaired activity), and can be adapted to other species. A disadvantage is that disability is not weighted in therapeutic response.

 Non-validated multidimensional chronic pain scales include

1.        Colorado State University Canine Chronic Pain teaching tool is similar to the acute pain teaching tool in that it is intended to be used to develop students' clinical evaluation skills.

Feline chronic pain scales are being developed. Some are feline specific and some are adaptations or applications of canine scales

1.        Feline Chronic Pain Index (FCPI) is based on degenerative joint disease-associated pain in cats; it is not yet validated.

Quality of Life (QOL) Scales have been developed as somewhat distinct from chronic pain scales. These scales may be used both in chronic pain and in hospice patients.

1.        Cincinnati Orthopedic Disability Index (CODI): because it assesses pain and disability this scale is considered a QOL scale. An additional QOL advantage is the ability to assess therapeutic intervention and it is validated in dogs.

2.        Heath Related Quality of Life Scale (HRQLS) is easy to use, multifactorial, can distinguish multiple chronic pathologies, can be used in dogs or cats (although only validated in dogs). The simple descriptive numeric rating scale is not sensitive.

3.        5H-2M scale has been developed by Dr. Alice Villalobos as a guide in making the euthanasia decision. The categories considered are hurt, hunger, hydration, hygiene, happiness, mobility, more good days than bad days. It can be used in dogs or cats. It is not validated and is subjective. However, it can be used to help clients focus on the parameters relevant to this decision.

4.        Glasgow University Veterinary-QUEST is a scale under development (2004) and validation (ongoing) by the same group who developed and validated the GMPS. More information is expected.

Guidelines for the use of pain scales

Pain assessment by proxy (surrogate) is by nature a subjective endeavor. However, through the use of appropriate, validated scales and the development of clinical skills such assessment can relieve pain in our veterinary patients. Guidelines include: 1) learn normal behavior for a given species, breed, individual and learn to recognize deviation; 2) look for unrecognized pain sources (housing, procedures, husbandry, pre/co-existing pain); 3) anticipate painful situations and plan intervention prior; 4) be consistent: choose a scale, train staff and clients, practice; 5) log assessments on every patient; 6) establish action points and then use them; 7) reassess frequently.

When developing a pain assessment technique be methodical: 1) determine the patient's identity, signalment, and history 2) consider what your clinical experience tells you about that individual and its signalment and history; 3) find out what your nurses/techs and family members think of the patient's condition and comfort; 4) quietly observe the patient without interacting (response to sound, visual stimulus; posture; activity, etc); 5) evaluate how your observations compare to #2 above. 6) perform an interactive assessment (meet and greet, non-obtrusive petting, palpation of entire body, examination and palpation of wounded area); 7) evaluate how your interactive assessment compares to #2 above.

When determining whether to take action process the data gathered in the methodical assessment. Make a list of differentials for the observed behaviors and responses. Are they caused by pain, anxiety, fear, hunger, medication effect, environment effect, etc. Perform “diagnostic” testing to narrow the differential list: e.g. feed, take patient out to urinate, adjust bedding, comfort, alter medications. If pain is the source of behavior change assign a pain score; treat appropriately and REASSESS, and then treat as needed and REASSESS, and so on until your patient is discharged from your care. Use of “as needed” pain therapy is common; if you write these orders then methodical reassessment is the only way to determine if it is needed.