NSAIDs: choices and issues (Proceedings)

General Pharmacology of NSAIDs

As their name implies, nonsteroidal aniti-inflammatory drugs (NSAIDs) are used in the treatment of inflammatory conditions, which are characterized by redness, swelling, heat, pain, and loss of function. Although the inflammatory response can be viewed as essentially protective and beneficial to the body, excessive inflammation in the face of progressive disease can promote the cycle of increasing damage and inflammation. In addition, the pain associated with inflammation can have adverse effects on patient welfare. The major indications for the administration of NSAIDs are analgesic, anti-inflammatory, and reduction of pyrexia. These indications generally involve symptomatic treatment of a primary problem. Indications for the use of NSAIDs as primary therapy in small animal patients include the prevention and treatment of thromboembolic disease and the adjunctive therapy of neoplasia. Of these indications, the greatest area of use and interest in NSAIDs during recent years has been in the reduction of pain. As the need for geriatric medicine has increased in prominence, there has been a commensurate rise in interest in the use of NSAIDs as palliative therapy for osteoarthritis.

Osteoarthritis involves the cycle of inflammation and damage, in which the underlying damage to chondrocytes and synovial cells results in localized inflammation. Inflammatory mediators, such as prostaglandins, leukotrienes, superoxides, and proteolytic enzymes, lead to decreased viscosity of synovial fluid, further damage to the joint, and increased inflammation. The original injury stimulates this process by liberating phospholipids from cellular membranes. Phospholipases act on the liberated phospholipids to form arachidonic acid, which is itself a substrate for two separate enzyme systems. Arachidonic acid can be metabolized by several isoforms of cyclooxygenase to form prostanoid metabolites, including prostaglandin E2 (PGE2), thromboxane (TXA2), prostacyclin (PGI2), and PGF2a. Alternatively, arachidonic acid can be metabolized by lipoxygenase to form leukotrienes, such as LTB4. All of these eicosanoids are inflammatory mediators that differ with respect to their physiological functions. As the various NSAIDs will inhibit different enzymes in the eicosanoid pathway, an understanding of the functions of each eicosanoid provide the foundation for the differential effects of NSAIDs. The prostanoid most commonly measured to assess the efficacy of NSAIDs is PGE2, which is vasodilatory, sensitizes nerves to pain, and is pyretic. Of the other prostanoids, thromboxane and prostacyclin oppose one another in activity, with thromboxane serving to stimulate vasoconstriction and platelet aggregation whereas prostacyclin exerts the opposite effects. The eicosanoid PGF2a primarily functions as a reproductive hormone, although it also exhibits vasoconstrictive effects. Arguably the best known inflammatory leukotriene is LTB4, which stimulates leukocyte chemotaxis, aggregation, and degranulation. In addition, LTB4 also increases vasodilation and permeability of capillaries, thus increasing redness and swelling. Although leukotrienes appear to play a role in the toxicities that can be associated with NSAID administration, they are also important mediators of hypersensitivity, such as in asthmatic diseases. Other inflammatory mediators, such as proteolytic enzymes and superoxides, contribute to localized lipid peroxidation and tissue destruction.

It is of particular pharmacological interest that there are two isoforms of cyclooxygenase that are differentially expressed in the tissues. Cyclooxygenase I (COX-1) is constitutively expressed in numerous tissues, including the platelets, vascular endothelial cells, gastrointestinal tract, and renal tubule. The prostanoid products of COX-1 include PGE2, which is associated with gastroprotective effects and renal homeostasis. Prostacyclin and thromboxane are also produced by COX-1. In contrast, COX-2 is classically considered to be the inducible isoform of cyclooxygenase, with its expression being increased by injury, and its products, such as PGE2, being primarily pro-inflammatory and associated with pathological effects. The activity of an NSAID can be measured by in vitro assays that yield the concentration of drug that inhibits COX enzyme activity by 50% as compared to the absence of drug (IC50). An agent with a low IC50 more potently inhibits COX than does a drug with a higher IC50. In addition to facilitating the comparison of different NSAIDs, the IC50 can be used to test how an NSAID affects different COX isoforms. The ratio of the IC50 for COX-2:COX-1 can be used to classify the activity of NSAIDs as nonselective, preferential, or selective for inhibition of COX-2. An agent with similar activity against both COX-1 and COX-2, such as aspirin, is nonselective for COX-2. A drug that much more potently inhibits COX-2 as compared to COX-1, such as deracoxib, is considered to be COX-2 selective. Agents that are intermediate in their inhibition of COX-2 as compared to COX-1, such as carprofen, are considered to be preferential inhibitors of COX-2. These designations are somewhat arbitrary and are highly dependent on the assay and the species being studied, but are useful for considering the spectrum of indications and side effects that are expected from each NSAID. Most of the classical side effects associated with NSAID use, such as gastrointestinal ulceration and renal disease, are primarily attributed to inhibition of constitutive COX-1. Therefore, COX-2 preferential and selective agents are associated with fewer of these classical side effects than are nonselective agents. This has allowed many dogs and cats that couldn't have tolerated anti-inflammatory doses of older NSAIDs, such as aspirin and ibuprofen, to tolerate the newer COX-2 selective and preferential agents. In addition, considerable research and development of COX-2 selective NSAIDs in human medicine has led to the development of similar compounds for veterinary use. Altogether, the greater safety of newer agents along with the explosion in research interest has resulted in the approval of many new NSAIDs expressly for veterinary use during the past ten years.

Specific NSAIDs

Aspirin, or acetylsalicylic acid, is a weak acid that is primarily absorbed in its unionized form from the stomach. Enteric coating or buffering of aspirin results in similar overall absorption, but allows more of the aspirin to be absorbed in the small intestine and purportedly reduces gastrointestinal irritation. Aspirin is an established anti-inflammatory, analgesic, and antipyretic agent in dogs, although its popularity for these purposes has decreased in recent years with the advent of newer, safer NSAIDs. The use of aspirin in cats for these indications is more likely to result in toxicity due to their relative deficiency in metabolism by glucuronidation. Gastrointestinal side effects are common with aspirin administration, ranging from inappetance to gastrointestinal ulceration and bleeding. More extreme toxicity includes metabolic acidosis and electrolyte disorders. Although aspirin has largely fallen out of favor as an anti-inflammatory agent in small animal medicine, it continues to be an important anticoagulant drug. Its utility in this role follows from its actions on COX-1 and the formation of thromboxane (TXA2). Following an inciting event, the COX-1 prostanoid product TXA2 is a down-stream factor involved in the eventual clot formation. In addition to other release factors, TXA2 amplifies hemostasis by recruiting and activating additional platelets to the growing thrombus. As a nonselective COX inhibitor, aspirin inhibits the formation of both TXA2 and prostacyclin (PGI2) from COX-1. Although other NSAIDs also inhibit the formation of eicosanoid products, aspirin is unique in its irreversible inhibition of COX-1. As platelets do not synthesize new proteins, COX-1 is effectively inhibited for the lifespan of the platelet, thus losing its ability to produce TXA2. This inhibition of platelet COX-1 occurs at low aspirin concentrations. In contrast, the endothelial cell COX-1 is not similarly inhibited or the enzyme is rapidly replaced, such that the vasodilatory and anticoagulant eicosanoid PGI2 is still active. It is the relative ratio of TXA2 to PGI2 concentrations associated with low-doses of aspirin that appears to be responsible for aspirin's anticoagulant effect. This effect may be decreased by high doses of aspirin or by the concomitant administration of additional NSAIDs.

Acetaminophen is not a true NSAID, as it only weakly inhibits COX. Instead, it is a centrally acting antipyretic and analgesic agent that is chemically distinct from the NSAIDs. The liver metabolizes acetaminophen by a series of interconnected pathways that result in acetaminophen's relatively narrow margin of safety. Acetaminophen can enter into any of several metabolic pathways, including activation to a highly toxic metabolite (NAPQI) by cytochrome P450 enzymes. In healthy humans and dogs, only a small fraction of the available acetaminophen enters the P450 pathway, where NAPQI is rapidly complexed with glutathione to make a nontoxic product. However, when acetaminophen is administered to cats, dogs with liver disease, or in high doses, excessive quantities of NAPQI are formed and glutathione is depleted, leading to hepatic cell death, hemolysis, and Heinz body formation. These side effects preclude the use of acetaminophen in cats and have greatly limited the utility of this drug in dogs, especially given the availability of safer products that are approved for use in this species.

Pyralzolone NSAIDs include the nonselective COX inhibitor phenylbutazone, which is well known for its efficacy in horses but also enjoys an historical approval in dogs. Unfortunately, phenylbutazone appears to be associated with a higher incidence of toxicity, including renal toxicity, hepatotoxicity, and aplastic anemia, in dogs than in horses.1 Although the phenylbutazone approval in dogs persists, it is currently seldom used in this species due to concerns about unacceptable toxicity.

Propionic acids include carprofen, which appears to be a preferential inhibitor of COX-2 in dogs. This preferential inhibition of COX-2 appears to give carprofen a considerably safer general profile than older, nonselective agents. Carprofen is approved for both surgical analgesia and for the therapy of osteoarthritis in dogs. As both injectable and oral formulations are available, injectable carprofen can be given subcutaneously as a preoperative analgesic without apparent affects on hemostasis.2 Despite the low incidence of gastrointestinal and renal toxicity with carprofen use in dogs, post-approval use of this drug revealed a low incidence of acute hepatic syndrome, in which Labrador retrievers were overrepresented.3 Acute hepatic syndrome resulting from carprofen administration appears to be idiosyncratic and unpredictable. Owners should observe dogs for signs of toxicity, such as anorexia, vomiting, and icterus, and should discontinue the drug if such signs are observed. Although carprofen is not approved for use in cats in the U.S., it is approved as a single dose in several other countries. Like other NSAIDs, the use of multiple doses of carprofen in cats is expected to be associated with a greater likelihood for adverse effects than is a single, post-operative dose. However, experimental evidence suggests that a short course of oral carprofen is tolerated by healthy cats.4

Oxicams include meloxicam and piroxicam. Meloxicam and carprofen have demonstrated similar efficacies against post-operative pain in both dogs and cats.5,6 Multiple doses of meloxicam are used routinely and safely in dogs, with gastrointestinal signs being the most commonly observed adverse effect. Meloxicam is a preferential COX-2 inhibitor, imparting a favorable safety profile in both dogs and cats. Meloxicam is the only NSAID approved for use in cats in the U.S., as a single subcutaneous injection for post-operative analgesia. Although the U.S. label warns against the administration of multiple doses of meloxicam to cats, it has also been used extralabel for short and long courses of therapy, which experimental evidence seems to support.7 Piroxicam is not approved for use in veterinary medicine, but has been used in dogs as adjunct therapy for the treatment of several neoplasms, such as oral malignant melanoma and squamous cell carcinoma.8 The pathophysiological rationale for the use of piroxicam for the therapy of neoplasia involves inhibition of COX-2, which some tumors overexpress. More selective COX-2 inhibitors, such as deracoxib, might also be useful agents for the therapy of neoplasia.

Etodolac is an indole acetic acid derivative with a fast onset of action that is approved for the management of pain and inflammation associated with osteoarthritis in dogs. Etodolac is COX-1 sparing in its activity and is primarily associated with minor gastrointestinal irritation. One advantage of etodolac over other NSAIDs is that long-term etodolac administration appears to have less effect on coagulation.9 However, long-term etodolac administration is associated with reversible or irreversible keratoconjuncitivis sicca (KCS) in dogs, suggesting that tear production should be monitored before and during therapy.10

Coxib NSAIDs include deracoxib and firocoxib, which both selectively inhibit COX-2. These drugs represent the culmination in the trend towards greater selectivity in NSAID activity against the COX-2 isoform. As a result, coxib agents are well-tolerated in dogs and are approved for long-term use in the palliative therapy of osteoarthritis. Deracoxib and firocoxib are structurally and functionally related to the well known coxib NSAIDs that are used in human medicine, such as celecoxib. This class of drug has come under increasing suspicion in human medicine, so a closer examination of the interplay between COX enzymes and the coxib drugs is useful. Despite the classical division of isoform function described above, COX-2 is also constitutively expressed in multiple tissues, including kidney, CNS, and endothelium. Of the homeostatic products of COX-2, prostacyclin may be the most important due to its inhibition of platelet aggregation and role in maintaining renal blood flow. Indeed, the homeostatic functions of COX-2 appear to be responsible for the increased risk of thromboembolic disease, such as myocardial infarction, in humans that take this class of drugs long-term. Although prostacyclin is produced by both COX-1 and COX-2, it may be that insufficient concentrations of prostacyclin are produced by COX-1 alone to offset the action of thromboxane, which is also produced by COX-1. Therefore, inhibition of COX-2 without concomitant inhibition of COX-1 may leave the pro-coagulant effects of thromboxane unopposed, thus tipping susceptible patients into a pro-coagulant state. Although there does appear to be an increased risk of some thromboembolic diseases, such as myocardial infarction, in humans, it is currently unknown whether this increased risk is also present in small animal patients. Cardiovascular disease is increasingly recognized in veterinary patients as more pets are maintained as geriatric dogs and cats. In addition, geriatric patients are more likely to receive NSAIDs for chronic orthopedic conditions, increasing the likelihood that chronic NSAID administration will occur in a veterinary patient with attendant cardiovascular disease. Owners may also be aware of the issues surrounding the use of COX-2 inhibitors due to the considerable press that this class of drugs has received. Therefore, it is worthwhile to consider the likelihood that veterinary patients will be at an increased risk of thromboembolic disease due to chronic administration of NSAIDs. Clinical trials in dogs conducted to date have not revealed an increase in thromboembolic disease. However, the initial clinical trials in humans also failed to identify the increase in myocardial infarction that would only be reported post-approval. Perhaps more importantly in veterinary medicine, the most common cardiovascular diseases of dogs and cats do not typically result from arteriosclerosis, which is a major risk factor for myocardial infarction in humans. Therefore, it may be that COX-2 inhibitors are unlikely to increase the risk of thromboembolic disease unless other risk factors are present. By this logic, populations that are susceptible to thromboembolism, such as cats with hypertrophic cardiomyopathy, may be at an increased risk for adverse effects with long-term use of selective COX-2 inhibitors. However, since no NSAID is currently approved for multiple-dose use in cats in the U.S., this adverse effect is unlikely to occur in this particular population. Dogs that are predisposed to thromboembolic disease, such as those with adult heartworm infection, are more likely candidates for pro-coagulant adverse effects due to COX-2 inhibitors.

Dual inhibitors are currently represented by a single NSAID, tepoxalin. In addition to inhibition of COX, tepoxalin, is a "dual-inhibitor", in that it inhibits both COX and lipoxygenase (LOX). Interestingly, tepoxalin appears to preferentially inhibit COX-1 as compared to COX-2, a reversal of the increasingly popular COX-2 inhibitor selectivity. Despite the association of COX-1 inhibition with gastrointestinal ulceration in dogs, tepoxalin appears to be well-tolerated with a low incidence of gastrointestinal side effects. This unexpected finding supports the postulate that gastrointestinal ulceration occurs from a combination of COX-1 inhibition and accumulation of leukotrienes. By inhibiting both of the arms of the inflammatory eicosainoid pathway simultaneously, dual inhibitors may avoid classical NSAID side effects.

REFERENCES

1. Weiss DJ, Klausner JS. Drug-associated aplastic anemia in dogs: eight cases (1984-1988). J Am Vet Med Assoc 1990;196(3):472-5.

2. Bergmann HM, Nolte IJ, Kramer S. Effects of preoperative administration of carprofen on renal function and hemostasis in dogs undergoing surgery for fracture repair. Am J Vet Res 2005;66(8):1356-63.

3. MacPhail CM, Lappin MR, Meyer DJ, et al. Hepatocellular toxicosis associated with administration of carprofen in 21 dogs. J Am Vet Med Assoc 1998;212(12):1895-901.

4. Steagall PV, Moutinho FQ, Mantovani FB, et al. Evaluation of the adverse effects of subcutaneous carprofen over six days in healthy cats. Res Vet Sci 2009;86(1):115-20.

5. Laredo FG, Belda E, Murciano J, et al. Comparison of the analgesic effects of meloxicam and carprofen administered preoperatively to dogs undergoing orthopaedic surgery. Vet Rec 2004;155(21):667-71.

6. Slingsby LS, Waterman-Pearson AE. Comparison between meloxicam and carprofen for postoperative analgesia after feline ovariohysterectomy. J Small Anim Pract 2002;43(7):286-9.

7. Gunew MN, Menrath VH, Marshall RD. Long-term safety, efficacy and palatability of oral meloxicam at 0.01-0.03 mg/kg for treatment of osteoarthritic pain in cats. J Feline Med Surg 2008;10(3):235-41.

8. Boria PA, Murry DJ, Bennett PF, et al. Evaluation of cisplatin combined with piroxicam for the treatment of oral malignant melanoma and oral squamous cell carcinoma in dogs. J Am Vet Med Assoc 2004;224(3):388-94.

9. Luna SP, Basilio AC, Steagall PV, et al. Evaluation of adverse effects of long-term oral administration of carprofen, etodolac, flunixin meglumine, ketoprofen, and meloxicam in dogs. Am J Vet Res 2007;68(3):258-64.

10. Klauss G, Giuliano EA, Moore CP, et al. Keratoconjunctivitis sicca associated with administration of etodolac in dogs: 211 cases (1992-2002). J Am Vet Med Assoc 2007;230(4):541-7.