NSAIDs, anesthesia, and the kidneys: What they are not telling you (Proceedings)

Nonsteroidal anti-inflammatory drugs (NSAIDs) are the most widely used analgesic drug class in human and veterinary medicine. NSAIDs are effective due to both peripheral and central mechanisms of analgesia. The widespread use of NSAIDs in human medicine has resulted in a large number of patients experiencing adverse effects including death. It has been estimated that 16,500 human deaths occur annually in the United States due to NSAID use. NSAID – associated deaths would account for the 15th leading cause of death in the U.S. if included separately on the National Vital Statistics report. Only a small number of patients (~19%) with serious gastrointestinal complications reported symptoms prior to presentation for the serious adverse effect. Therefore monitoring patients for adverse effects is a relatively insensitive method of preventing serious adverse effects associated with NSAID use. It is important to realize that severe complications also occur in veterinary medicine, including death, even with short treatment periods.

Mechanisms of Action

The primary mechanism of action associated with NSAIDs is on the arachidonic acid pathway through the inhibition of the cyclooxygenase (COX) enzymes. However inhibition of lipoxygenase (LOX), nuclear factor κ-B (NF-κB), and bradykinin can also contribute to the effects of NSAIDS. At least two isoforms of COX have been identified, COX-1 and COX-2. A variant of COX-1, termed COX-3, has also been identified in the cerebral cortex, but the biologic significance of COX-3 is still debated. COX-1 is often referred to as constitutive as it contributes to routine physiologic processes and homeostasis. COX-1 is also up-regulated in inflammatory conditions and in the cerebral cortex as a component of the ascending pain pathways. COX-1 catalyzes the formation of prostaglandins that result in vasodilation, inflammation and pain, sensitization of nociceptors, fever, gastroprotective effects (increase mucous production, bicarbonate secretion, increase mucosal blood flow, inhibit acid secretion), and regulate renal blood flow in some species. COX-1 products are also converted to thromboxane in platelets that result in platelet aggregation and vasoconstriction.

COX-2 is often referred to as inducible as it is up-regulated in many inflammatory conditions and some neoplastic tissues. Prostaglandins produced by COX-2 result in vasodilation, inflammation and pain, sensitization of nociceptors, and fever. COX-2 is constitutively expressed in the GI tract and renal vasculature of dogs and is primarily responsible for modulation of renal blood flow in dogs. COX-2 is up-regulated in healing GIT ulcers and is necessary for healing in the GIT. COX-2 products are also converted to prostacyclin which is primarily antagonistic to thromboxane. Prostacyclin results in vasodilation and inhibits platelet aggregation. Prostacyclin also sensitizes nociceptors. COX-2 is expressed in the dorsal horn of the spinal cord and is active in the propagation of the ascending pathways. COX-2 is up-regulated in certain neoplastic diseases such as transitional cell carcinoma and osteosarcoma. The mechanisms and reasons for the up-regulation are unclear, but may aid in the growth of blood vessels or be a result of an inflammatory response to the neoplastic growth. Feline neoplasms appear to express COX-2 to less of an extent than canine neoplasms therefore NSAIDs may be less likely to produce desired clinical effects. COX-2 is up-regulated in the uterus during late pregnancy and NSAID use has resulted in prolonged gestation. Maintenance of the patent ductus arteriosis is mediated by COX-2, therefore inhibition results in premature closure and altered fetal hemodynamics.

COX-3 has been identified in the cerebral cortex of dogs and humans. COX-3 is a splice variant of COX-1 with some researchers hypothesizing it has no biologic effect, but in vitro COX-3 catalyzes similar prostaglandins as COX-1. Researchers have shown that acetaminophen and dipyrone are inhibitors of COX-3 with minor effects on COX-1 and COX-2, leading to the hypothesis that the analgesic and antipyretic effects of acetaminophen and dipyrone are due to central effects and not due to peripheral anti-inflammatory effects.

Lipoxygenase catalyzes the formation of leukotrienes as a part of the arachidonic acid pathway. Leukotrienes are antagonistic to prostaglandins in many ways resulting in vasoconstriction and increased gastric acid secretion. Leukotrienes also result in bronchoconstriction and are pro-inflammatory leading to the activation, recruitment, migration, and adhesion of inflammatory cells in addition to increased production of inflammatory cytokines such as TNF-α and IL-1β . Tepoxalin is the only NSAID currently marketed that inhibits LOX, but the duration of inhibition is short; the IC50 is maintained for less the 6 hours. Zileuton (Zyflo) is a specific LOX inhibitor marketed for treatment of asthma in humans. Leukotriene antagonists such as montelukast (Singulair) and zafirlukast (Accolate) are also available for treatment of human asthma. The clinical usefulness of the zileuton and the LT antagonists for the treatment of pain have not been extensively studied in animals, but experimental models have demonstrated antinociceptive (analgesic) effects.

Lipoxins are modulators of inflammation produced from COX-2 products. Lipoxins are inhibitors of leukotrienes, natural killer cells, inhibit leukocyte migration, inhibit TNF-α and NF-κB, and inhibit adhesion molecules and production of interleukins. Aspirin is relatively unique among the NSAIDs as it acetylates COX-2 with lipoxin being preferentially produced, "Aspirin Triggered Lipoxin." COX-2 inhibitors co-administered with aspirin result in increased GIT toxicity than sole administration of either agent.

The simplistic classification of COX-1 is the "good" isoform and COX-2 is the "bad" isoform is an overly simplistic classification system. As described previously both isoforms are constitutively expressed and required for homeostasis and protective responses.

Adverse Effects

Renal toxicity. Renal adverse effects account for ~21% of the reported adverse effects in dogs. The incidence of renal toxicity, particularly in cats, appears to be increasing with increasing NSAID usage. In dogs, COX-2 is constitutively expressed in the renal vasculature and mediates renal vasodilation, and is up-regulated, maintaining renal blood flow during episodes of renal hypotension, such as shock or anesthesia. In other species COX-1 is constitutively expressed in the renal vasculature, whereas COX-2 is up-regulated during hypotensive episodes and is important in the maintenance of renal blood flow. The expression of COX-1 and COX-2 in the feline renal vasculature has not been described.

The incidence of renal adverse effects is increased during conditions of renal hypotension such as anesthesia, hypovolemia, systemic hypotension, cardiovascular disease, shock, or concurrent administration of nephrotoxic drugs. Use of COX-2 selective NSAIDS does not result in a decrease in renal adverse effects. The use of NSAIDs in chronic renal failure is generally not recommended, but have been used occasionally. Strong client warnings are recommended with the inclusion of a signed consent form for cases involving the use of NSAIDs and chronic renal disease.

Studies examining the long-term effects of various NSAIDs (carprofen, etodolac, flunixin, ketoprofen, and meloxicam) on the kidney resulted in no renal abnormalities as assessed by urinalysis and serum chemistry profile. This is not unexpected as toxicology studies of orally administered NSAIDs IN HEALTHY YOUNG BEAGLES is overall well tolerated with minimal renal adverse effects on serum chemistry, urinalysis and post mortem examinations. Meloxicam administered for 42 days resulted in renal papillary necrosis at 5× the label dose in 1 dog and renal papillary degeneration in 2 dogs at the 5× dose. Administration of meloxicam at 5× the label dose for 6 months resulted in increased BUN in 2/6 dogs, but no histologic evidence of renal changes in any of the dogs. Carprofen administered at 5× the label dose for 1 year in Beagles resulted in no renal adverse effects on serum chemistry or histopathology examination. Deracoxib administered for 6 months at 3× the label dose resulted in hypostheuria and polyuria in 2 dogs, but not in the 5× dose. The BUN increased significantly in the 3×, 4×, and 5× group, but no changes occurred in the creatinine. Deracoxib caused dose-dependent renal tubular degeneration and regeneration in the 3×, 4×, and 5× group, and focal renal papillary necrosis in 3/10 dogs the 5× and 1/10 dogs in the 4× dose groups. Tepoxalin administered for 1 year at 5× the label dose resulted in no abnormal renal chemistry parameters or histopathologic evidence of renal changes. Firocoxib administered at 5× the label dose for 6 months resulted in no reported changes in renal chemistry parameters or histopathologic changes in the kidney. Etodolac administration at 5× the label dose for 1 year to dogs resulted in severe GI adverse effects with 6/8 dogs dying from GI perforation, but 1 of the dogs also had histopathologic renal lesions including renal tubular nephrosis. In general, renal toxicity of approved NSAIDs during long-term administration is minimal at the label dose in healthy dogs, but higher than label doses increase the risk of NSAID induced renal damage. However it is important to realize that no study has demonstrated that any of the approved NSAIDs produces less renal adverse effects at the label dose. It is also important to reiterate that the dogs included in the toxicology studies were young healthy Beagles, were not administered concurrent drugs, and were maintained in a lab environment.

Acute studies critically assessing the renal effects of veterinary approved NSAIDs on the kidney during anesthesia and surgery are lacking. Despite numerous publications assessing various NSAIDs on renal parameters, these studies assessed inappropriate parameters for which NSAIDs have been shown not to affect acutely. The serum chemistries (BUN, creatinine), urine specific gravity, and glomerular filtration rate are unaffected acutely from NSAID administration. However these are the parameters that most of the studies have assessed. In contrast, the primary acute changes in renal function produced by NSAID administration are decreased renal blood flow and sodium excretion. Two studies have evaluated either the renal blood flow or sodium excretion in anesthetized dogs administered veterinary approved NSAIDs. One of the studies administered medetomidine to both the NSAID treated and non-NSAID treated dogs undergoing anesthesia. Both groups had significant decreases in renal blood flow. A separate study measured sodium excretion in NSAID treated dogs undergoing anesthesia and compared to dogs receiving morphine and undergoing anesthesia. The dogs that received carprofen (4 mg/kg) and ketoprofen (1 mg/kg) had significant decreases in sodium excretion whereas the morphine group did not have significant decreases in sodium excretion. Therefore the little data that are available indicate that NSAIDs do significantly affect renal function during anesthesia. A reasonable conclusion is in young healthy dogs administration of an NSAID during anesthesia does not result in renal failure, but we cannot rule out subclinical renal effects. Therefore it is still recommended to closely the hemodynamic parameters of a patient on NSAIDs undergoing anesthesia and to provide fluid support during the peri-anesthetic period. It is also important to realize that NSAID associated renal failure is expected to present with clinical signs 2-7 days after the toxic event.

Concurrent Drug Use

In general the concurrent use of NSAIDs and corticosteroids should be avoided due to the increased potential of GIT adverse effects. The effects of diuretics and angiotensin converting enzyme inhibitors (ACE inhibitors), such as furosemide and enalapril, are decreased when concurrent administered with NSAIDs. Combining NSAIDs with other analgesic such as opioids, tramadol, gabapentin, and amantadine result in additive to synergistic analgesic effects and more effective pain control.

Conclusions

In general NSAIDs are well tolerated by veterinary patients, however serious adverse effects including death occur on an infrequent, but regular basis. NSAIDs are not absolutely safe drugs, but the introduction of newer NSAIDs has decreased the incidence of gastrointestinal adverse drug effects. Owners should always be warned of potential adverse effects including gastrointestinal, renal, and hepatic injury, and even death.

Online resources: FDA brochure available online:http://www.fda.gov/cvm/Documents/NSAIDBrochure.pdf

Disclaimer: The information is accurate to the best of the author's knowledge. However recommendations change as new data become available and errors are possible. The author recommends double checking the accuracy of all information including dosages.