This chapter explores the fundamentals of acute and chronic pain management in dogs. The reader is directed to more comprehensive resources for discussions of pain neurobiology, neuropharmacology, and most drug doses. An emphasis is placed on practical and evidence-based approaches, with attention to safe and responsible drug use.
We must be mindful that there is relatively little evidence-based veterinary medicine (EBVM), thus guidance with regard to complex clinical choices is in short supply. It is both the blessing and the curse of contemporary medicine that there exists an impressive array of tools in the pain management toolbox.
What drug(s) shall we use? In what order? In which combinations? For which type of pain? At what dose? For how long? For these and many other variables, there are no simple answers in the veterinary literature. The competing values of a best-evidence approach, personal experience, and client values (preferences) will always be in dynamic tension (Figure 19.1; Sackett et al., 1996).
Pain is further complicated for veterinarians, as our patients are nonverbal, mirroring the challenge in human medicine with nonverbal subpopulations: neonates (Schechter, 1989), the cognitively impaired (Cook et al., 1999), and the elderly (Lovheim et al., 2006). In verbal patients, pain is what the patient says it is; in nonverbal patients, pain is what we say it is.
Clinicians often limit pain medications for fear of adverse drug effects (ADEs) or interactions. Yet we must also consider the harm to the patient if pain is inadequately managed. We have an ethical obligation to minimize pain in our patients, and to recognize negative medical, physiological, emotional, and even cognitive consequence that are a direct result of undermanaged pain.
Pain elicits a cascade of debilitating neurohormonal effects that includes hypertension, catabolism, immunosuppression, and worse. With undermanaged pain, surgical patients heal and recover more slowly, and may develop chronic pain states and even severe, life-threatening complications (Anand & Hickey, 1992). Chronic pain in humans is associated with cognitive impairment (Kreitler et al., 2007) and is comorbid with clinical depression. In dogs, underrecognized, underattended, undermanaged pain can become a criterion for euthanasia.
We know much about optimal pain management in animals, but what we think we know is dwarfed by what we do not. Literature must be read critically, and any recommendations about protocols, including those in this chapter, must be considered analytically, with an open mind toward the viewpoints of others and a commitment to continued learning.
The Multimodal Approach to Pain Management
The principle is simple: Relying upon one modality or drug requires higher doses and/or more frequent and/or prolonged administration to achieve the desired effect, while minimizing the potential benefit and maximizing the possibility of ADEs (Figure 19.2). With multiple modalities, each affecting different aspects of pain processing, requirements for each drug are reduced while achieving superior effect and minimizing ADEs. Although difficult to study, there is growing evidence in the veterinary literature that this principle applies in both acute (Brondani et al., 2009) and chronic pain (Fritsch et al., 2010) settings.
Illustration by Marcia Schlehr.
Neuropharmacology: The Tools in the Toolbox
Nonsteroidal Anti-Inflammatory Drugs (NSAIDs)
Nonsteroidal anti-inflammatory drugs (NSAIDs) are the most commonly used modality to manage pain, and for good reason: they are highly effective, commonly available, licensed for use in dogs, and safe with proper use. Because inflammation is a prime pain-generating physiologic mechanism, NSAIDs are among the most important drugs in the veterinarian’s arsenal.
Their primary mode of analgesic action is to inhibit cyclooxygenase-2 (COX-2), the enzyme that metabolizes arachidonic acid and results in the production of proinflammatory and vasoactive prostaglandins. NSAIDs also appear to inhibit central perception of pain by modulating multiple gene expression pathways (Wang et al., 2007). Arachidonic acid is also metabolized by lipooxygenase to produce leukotrienes, which attract PMNs and promote their adherence to endothelium. The relative roles and molecular dynamics of COX and lipooxygenase (LOX) enzyme variants are still being elucidated, and the optimal LOX- and COX-selective/sparing effect that maximizes effectiveness and limits toxicity remains unclear.
All veterinary NSAID products are effective, although individual patient responses vary. Their main limitation is the potential for adverse effects. COX enzymes are crucial to the production of cyto-protective prostaglandins (COX-1 especially in the gastrointestinal (GI) tract and renal tubules, COX-2 in the renal tubules), so the primary ADEs of NSAIDs include gastroduodenal erosion/ulceration and nephrotoxicity. Additional concerns include rare idiosyncratic hepatotoxicity and effects on tissue healing and platelet function. Evidence suggests that NSAID ADEs may be less dependent on longevity of use than on biologic predisposition and improper use. The single greatest variable in preventing NSAID-related ADE is the veterinarian, who must be aware of concurrent drug use and patient risk factors, and ensure proper use, patient monitoring, and client education. Seventy-five percent of individuals reporting adverse NSAID events to the FDA hotline feel that their veterinarian did not inform them adequately of possible side effects and/or failed to give the client the drug information sheets (Hampshire et al., 2004).
Tips for NSAID Use
The edge in efficacy in both human and canine NSAID studies goes to preoperative use (Lascelles et al., 1998). Systematic reviews in humans suggest safety of preoperative NSAIDs (Lee et al., 2007a), as do studies in healthy dogs even with moderate intraoperative hypotension (Bostrom et al., 2002). It is axiomatic that patients undergoing general anesthesia should have the benefit of intravenous fluid support.
New Drugs in Class
- Robenacoxib (Onsior®; Novartis Animal Health, Greensboro, NC) (Sackett et al., 1996) is a unique COX-2 selective (King et al., 2010) approved for canine and feline use in Europe, and approved in the United States for perioperative pain in cats. It has a short plasma half-life yet accumulates in inflammatory exudates for an extended time (King et al., 2009).
- Mavacoxib (Trocoxcil®, Europe only; Pfizer, Paris) is a sustained-release NSAID approved for chronic pain in dogs. Pharmacokinetics are established (Cox et al., 2010), but there is no peer-reviewed literature regarding efficacy.
- Nitronaproxen (Naproxcinod®, Europe for humans only; NiCox, Sophia Antipolis, France) is a cyclooxygenase-inhibiting nitric oxide-donating drug (CINOD) that appears to have the analgesic efficacy of the parent NSAID (Geusens, 2009) but with a greatly reduced incidence of gastropathy (Fritsch et al., 2010)
- Meloxicam: New formulation for dogs, provided as an oral transmucosal spray (OroCAM®; Abbott Animal Health, Abbott Park, IL, labeled in U.S. for chronic pain associated with osteoarthritis; and RevitaCAM®; Abbott Animal Health, labeled in UK/EU for acute and chronic musculoskeletal disorders).
Acetaminophen may elicit some of its analgesic effects by inhibiting one or more centrally acting COX variants (Kuo, 2006). Studies suggest that it may inhibit COX-2-mediated production of PGE2 (Lee et al., 2007b). Its postoperative pain-modifying (and possible anti-inflammatory) effects have been demonstrated in dogs (Mburu et al., 1988). While the drug is cleared more slowly in dogs than in humans (Plumb, 2011), there is no toxicity data to suggest unusual tendency toward adverse effects in dogs with judicious use. Chronic use cannot be advocated in dogs as long-term safety has not been established (and based on experience in humans, is probably suspect).
Tips for Use
Acetaminophen can be alternated with NSAID, but with a maximum frequency of BID, maximum dose of 325 mg even for the largest dogs, and maximum duration of administration of 5 days. Acetaminophen is contraindicated in dogs with liver disease or exposed to other hepatotoxic or methemoglobin-inducing drugs or toxins.
Synthetic opioids are powerful tools to manage pain because receptors for naturally occurring opioids (endorphins, enkephalins) are distributed throughout the body and can be found in both central and peripheral tissues. Several opioid receptor types and subtypes have been isolated, each with a variant effect.
Activation of a mu-opioid receptor inhibits presynaptic release of and postsynaptic response to excitatory neurotransmitters (especially in the dorsal horn of the spinal cord), hyperpolarizing second-order neurons (Barkin et al., 2006). Activation of kappa receptors promotes the release of inhibitory neurotransmitters (predominantly gamma-aminobutyric acid [GABA]).
Pure Mu Agonists
Morphine remains the prototype opioid and the one in widest use; it has no ceiling effect on analgesia or respiratory depression, elicits histamine release, and in nonpainful dogs (e.g., as a premedicant) causes vomiting.
Oxymorphone (Numorphan®; Endo Health Solutions Inc., Chadds Ford, PA), and hydromorphone (Dilaudid®; Purdue Pharma, Stamford, CT, and also available as a generic form) do not elicit histamine release, produce somewhat less nausea, and have a shorter duration of action than morphine.
Methadone may be a parenteral opioid alternative in animals, in part due to its additional effect as an N-methyl-D-aspartate (NMDA) antagonist. Some veterinarians favor it as a premedication due to its effectiveness, nice sedation, and low ADE profile (minimal if any nausea, no histamine release) (Ingvast-Larsson et al., 2010).
Fentanyl (Sublimaze®) is a potent, short-acting opioid preparation most often used as an intravenous constant rate infusion (CRI). A transdermal patch (Duragesic®; Sandoz Corp, East Hanover, NJ, and other manufacturers) has been used in dogs although a number of studies have demonstrated wide kinetic variability (Kyles et al., 1996). The potential liability with human ingestion, especially children (deaths have occurred), must also be considered. More recently, an extended duration (4 days) transdermal fentanyl preparation was FDA-approved for the control of postsurgical pain in dogs (Recuvyra™; Elanco, Greenfield, IN).
Oral opioids experience a robust first-pass effect in dogs (KuKanich et al., 2005), but they are not without utility. Oral hydrocodone in dogs has a bioavailability of approximately 50% of that found in humans (KuKanich & Paul, 2010), and codeine, while it does not convert in dogs to morphine, appears to make another highly bioactive mu-agonist metabolite (KuKanich, 2010). Rectal suppository opioid formulations may also be prescribed but appear to provide little advantage in bioavailability over the oral route in dogs (Barnhart et al., 2000). These drugs are Class III drugs, and are available as separate drugs or combined with acetaminophen.
Partial Mu Agonist
Buprenorphine (generic buprenorphine; American Regent, Shirley, NY, and other manufacturers) is a partial mu agonist with greater affinity than morphine, and it has a ceiling effect. Wide variation of onset and duration appears to exist in the dog (even when administered IV) (Krotscheck et al., 2008) and may be dose-dependent. Compounded sustained-release buprenorphine formulations are available for use in dogs, but they are not FDA-approved products nor are kinetics well established.
Oral transmucosal bioavailability is low in dogs (approx 40%; Abbo et al., 2008). Commercial buprenorphine patches (e.g., BuTrans®; Purdue Pharma, Stamford, CT) have compared favorably in dogs to IV buprenorphine in a thermal threshold model (Pieper et al., 2011) and using pain scores postovariohysterectomy (Moll et al., 2011).
Kappa Agonist/mu Antagonist
Butorphanol will weakly block the mu opioid receptor, but its kappa agonism will promote the release of inhibitory neurotransmitters such as GABA. It has a ceiling effect and a very short duration of, and effect on, visceral analgesia in the dog (approx. 30–40 minutes) (Sawyer et al., 1991; Grimm et al., 2000), making it a poor choice in this species for any kind of significant or prolonged pain states. It does have utility in select situations and as an adjunct with other medications such as alpha-2 agonists.
Naloxone (Narcan®; Amphastar Pharm, Rancho Cucamonga, CA, and other manufacturers) is a potent mu-opioid receptor antagonist, traditionally used to achieve rapid and complete reversal of opioid overdose or severe ADE. However, microdoses (as low as 0.01–0.05 mcg/kg IV) have been used to improve the analgesia provided by buprenorphine (La Vincente et al., 2008) and minimize adverse effects of IV morphine (Cepeda et al., 2004).
New developments in other long-acting opioids including novel formulations and delivery systems may overcome the need for, and limitations of, opioid intravenous CRI, patches, and PO administration. Opioids, despite their effectiveness, may create clinical challenges as well. In the acute setting, opioid-induced dysphoria, hyperalgesia, and respiratory depression may be encountered (McNicol & Carr, 2007). Having strategies for recognizing and counteracting these signs will minimize complications. With chronic use in humans, the most commonly reported opioid ADE is constipation. New peripherally acting mu opioid receptor antagonists alvimopan (Entereg®; Cubist Pharmaceuticals, Inc., Lexington, MA) and methylnatrexone (Relistor®; Salix Pharmaceuticals, Raleigh, NC) permit the central analgesic effect of opioids but block their effect on GI motility (Gevirtz, 2007).
Tips for Use of Opioids
Preoperatively, opioids are combined with an anxiolytic (tranquilizer/sedative) to create a profoundly relaxed, stress-free, and anesthetic-sparing state. The choice of opioid, route, dose, and duration of administration is dependent upon clinical preferences and patients’ individual needs.
Anecdotally, the most commonly encountered opioid adverse effect in dogs is dysphoria (Epstein, 2012). Treatment strategies include using buprenorphine (displaces pure mu agonists off the receptor), butorphanol (antagonizes the mu receptor but activates kappa receptor), or a microdose of dexmedetomidine; all will diminish dysphoria while maintaining some degree of analgesia.
Oral opioids are recommended for many chronic pain conditions in humans including osteoarthritis (OA) (American Pain Society, 2002). Intermittent, judicious use of oral opioids (+/− combined with acetaminophen) can be considered for dogs experiencing breakthrough pain in OA and other chronic pain states, and/or as palliative end-of-life care.
Local anesthetics (LAs) are a principal means to reduce pain, general anesthetic and concurrent analgesic requirements, and even inflammation and microbial activity (Cassuto et al., 2006; Johnson et al., 2008). They are considered quite safe at customary doses. Most techniques are easily mastered and inexpensive. There no longer exists a rationale against local/regional anesthesia as part of every surgical intervention (Jones, 2008). The International Veterinary Academy of Pain Management (IVAPM) adopted this position in a 2012 Consensus Statement.
Local anesthetics bind to a hydrophilic site within sodium channels thus blockading them; without a Na+ influx, neurons may not depolarize. The effect is complete anesthesia to a site rather than analgesia. Different LAs will have variable onsets, durations of action, and toxicities.
This modality is limited only by the clinician’s ability to learn various techniques and anatomic landmarks. Blocks include local, line, or paraincisional, subcutaneous infiltrative, intrapleural, intra-abdominal, retrobulbar, intratesticular, intra-articular, carpal ring, epidural, dental (orofacial), brachial plexus, intercostal, paravertebral, and use of wound diffusion catheters. Ultrasound and nerve electrolocator devices enhance the precision and dose in perineural delivery. Descriptions of the techniques are available (Campoy et al., 2008).
Tips for Use
The edge in efficacy goes to preoperative use (Savvas et al., 2008), but infiltration of LA postincisionally can be effective. To minimize the sting of administration in awake patients, the LA (Hogan et al., 2011) is warmed to body temperature and injected slowly. Diluting the LA (with saline to increase volume for large infiltrative areas) will slow onset and shorten duration of action. The duration of activity may be doubled with small amounts of an opioid (Bazin et al., 1997).
Toxicity is most likely when administration occurs at very large doses and/or intravenously (which must be avoided with the potentially cardiotoxic bupivacaine). LA can result in motor as well as sensory blockade (a special consideration with epidurals). There is little clinical evidence to support the belief that local anesthetics impair wound healing.
To facilitate catheter placement or other minor skin procedures, a transdermal lidocaine/prilocaine ointment formulation (eutectic mixture of local anesthetics, EMLA®; AstraZeneca, London, also available as a generic) is placed on a shaved area and covered with a bioadhesive dressing or other nonporous wrap (e.g., foil). Although penetration has been reported to be time-dependent (Wahlgren & Quiding, 2000), 20 minutes appears sufficient in canine patients.
Commercial 5% lidocaine patches (Lidoderm®; Endo Pharmaceuticals, Inc., Chadds Ford, PA) are labeled for humans with postherpetic neuralgia (shingles) and also have been described for postoperative paraincisional analgesia in dogs (Weil et al., 2007), with minimal systemic absorption noted (Weiland et al., 2006). The adhesive patches can be cut to the desired size and shape. One cautionary note is that an entire patch contains 700 mg of lidocaine, a toxic dose if ingested; therefore, adequate precautions must be taken.
There is evidence in humans for the anesthetic-sparing effect of intravenous lidocaine (IVL) and its ability to speed the return of bowel function, decrease postoperative pain, minimize opioid consumption, and shorten the hospital stay after abdominal surgery (Groudine et al., 1998). Evidence in dogs is somewhat weaker at this point (MacDougall et al., 2009), although there may be a synergistic effect with other drugs. Formulas for a combination morphine, lidocaine, and ketamine IV CRI have been described for dogs (Muir et al., 2003). The combination is profoundly analgesic, fairly sedating, and superior for the most painful postoperative states. IVL also elicits a sustained effect on neuropathic pain in humans (Cahana et al., 1998).
Other Drugs in Class
Mexilitine is an oral sodium-channel blocker, often called oral lidocaine and labeled for use as a cardiac anti-arrhythmic. It has also been used to treat chemotherapy-induced neuropathic pain states in humans (Egashira et al., 2010). Its utility for chronic pain conditions in dogs has not been established.
Subanesthetic Ketamine Constant Rate Infusion
NMDA receptor antagonism remains a research focus for pain in humans (Fisher et al., 2000). Ketamine is a dissociative anesthetic that binds to a phencyclidine receptor inside the NMDA receptor; that is, the calcium channel would already have to be open and active for ketamine to exert its effect. Once bound, it decreases the channel’s opening time and frequency, thus reducing Ca+2 ion influx and dampening secondary intracellular signaling cascades. It appears to be protective against hyperalgesia and central hypersensitization in the postoperative setting in humans and dogs (Slingsby & Waterman-Pearson, 2000; Hocking et al., 2007), and the evidence in humans is strong for its pain-preventive effects when given IV CRI at subanesthetic doses. Ideal subanesthetic ketamine plasma concentrations in the dog have been reported at 2–3 mcg/mL, which can be achieved by administering ketamine IV CRI at 10 mcg/kg/min (Boscan et al., 2005). The IVAPM adopted a position/consensus statement that clinicians should consider the modality as part of a multimodal approach to transoperative pain management.
Tips for Use
The recommended intraoperative rate can be accomplished by placing 60 mg (0.6 mL of 100 mg/mL stock) ketamine in 1 L of fluids administered at intraoperative rates of 10 mL/kg/h. Postoperatively, the rate can be reduced to maintenance rates of 2 mL/kg/h, which administers the ketamine CRI at 2 mcg/kg/min. A loading dose of 0.25–0.5 mg/kg ketamine IV is recommended prior to the initiation of the CRI in order to rapidly achieve plasma levels.
Medetomidine and dexmedetomidine bind to opioid-like receptors on C- and A-delta fibers, especially in the central nervous system (CNS). Binding presynaptically, norepinephrine (NE) production is reduced and sedation occurs; binding postsynaptically, analgesia is produced, and is profoundly synergistic with opioids. Binding also blocks NE receptors on blood vessels, resulting in vasoconstriction; the resulting hypertension parasympathetically induces bradycardia, which is extended by a subsequent direct decrease in sympathetic tone. Cardiac index is decreased; however, central perfusion is maintained. Many uses are described for the perioperative setting, usually in combination with opioids and at doses much lower than suggested by the manufacturer.
Tips for Use
Intravenous microdoses of dexmedetomidine intra- and postoperatively at 0.25–1.0 mcg/kg can address rocky anesthetic episodes, postoperative pain, or dysphoria. This calculates to tiny volumes even in large dogs, and although the effect at these doses may last only 10–15 minutes, it may be re-dosed to effect (constant-rate infusion doses are also described; Valtolina et al., 2009). If a concerning degree of bradycardia occurs, anticholinergics (atropine, glycopyrolate) should be avoided as they will increase heart rate, but against tremendous vascular resistance with potentially serious consequences. Instead, atipamezole is used to reverse.
Other Drugs in Class
- Tizanadine (Zanaflex®; Acorda Therapeutics, Inc, Ardsley, NY, and also available in generic form) is an oral, centrally acting alpha-2 agonist used in humans primarily as a skeletal muscle relaxant to treat muscle spasticity, and the resulting pain, in multiple sclerosis and a variety of other painful conditions. Its utility in dogs is unknown.
- Clonidine is a centrally acting alpha-2 agonist that can be administered systemically via oral, IV, SC, IM, transdermal, and epidural routes. Indicated in humans as antihypertensive agent and to treat ADHD, new uses are being found for its antinociceptive effects (Neil, 2011).
Pain-Modifying Analgesic Drugs (PMADs)
Tramadol (Ultram®; Jansenn Pharmaceuticals, Titusville, NJ, and also available in generic form) is a popular analgesic that in humans combines a highly active mu-agonist opioid (M1) metabolite along with serotonin and NE (inhibitory neurotransmitters) agonism. However, dogs produce very little of the M1 metabolite, and what little they make has a very short half-life (1.7 hours) (KuKanich & Papich, 2004). Nevertheless, recent studies have demonstrated the clinical usefulness of tramadol in dogs (Seddighi et al., 2009; Martins et al., 2010). It has become a popular adjunct to chronic pain management in humans (Wilder-Smith et al., 2001) and dogs, but there is only one unpublished abstract that suggests effectiveness of tramadol as an adjunct to NSAID in canine OA (Lambert et al., 2003). The incidence of dependence in humans may be substantially higher than previously suspected (Shaya & Ke, 2007), so the drug may move to a controlled status in the future.
Tips for Use
ADEs for Tramadol can include GI and extrapyramidal effects.
Other Drugs in Class
- Tapentadol (Nucytna®; PriCara Pharmacology, Jansenn Pharmaceuticals, Titusville, NJ) is a centrally acting analgesic with a dual mode of action similar to tramadol: mu-opioid receptor agonism and inhibition of NE reuptake. It is the parent compound, not a metabolite, that causes both of these effects, so this drug may offer an alternative superior to tramadol in dogs. Unfortunately, recent data from the United States reveal low bioavailability in dogs and poor performance on a tail-flick analgesia model (see Webliography). A more recent study under review reveals a significant analgesic effect in the canine tail-flick model.
See Figure 19.3 for cautions regarding possible serotoninergic drug interactions.