Although NSAIDs that are selective for COX or inhibit both COX and LOX activity have been developed, the use of NSAIDs continues to be limited by the potential for adverse gastrointestinal events. The mechanism of these gastrointestinal injuries is not well understood; however, it is now believed the damage develops as a combination of COX-1 and COX-2 inhibition and the interaction between NSAIDs and phospholipids and consequent uncoupling of mitochondrial oxidative phosophorylation.1–4 Ratios of COX-1 to COX-2 selectivity are inversely proportional to the incidence of serious NSAID-related adverse effects in humans.5 Nonsteroidal anti-inflammatory drugs that spare COX-1 are clinically effective and associated with a 50% decrease in toxic effects in humans, compared with the toxic effects of nonselective NSAIDs.5,6 Historically, in dogs, the toxic effects of nonselective NSAIDs are most pronounced in the stomach.7 However, a study8 in which researchers evaluated gastrointestinal toxic effects associated with a COX-1–sparing drug revealed more duodenal ulceration and perforations than have been reported elsewhere. The reason for the increased risk of duodenal ulceration has not been explained. Investigators in that study8 hypothesized that the relative amount of inhibition of COX isoenzyme and LOX enzyme varies in regard to tissue type and may account for the variation in toxic effects according to tissue type.
Studies9–11 in dogs have revealed in vivo effects of several NSAIDs on multiple tissues, including whole blood, gastric mucosa, and synovial fluid. Findings of these studies indicate the selectivities of COX isoenzymes and 5-LOX enzyme are variable. Furthermore, results of 1 study strongly suggest that calculations of in vitro COX selectivity do not correlate with the in vivo ability to inhibit PGs. To the authors' knowledge, no data are available on COX and LOX inhibition in the remaining gastrointestinal tract, including the duodenum. Therefore, the purpose of the study reported here was to evaluate in vivo concentrations of PGs and leukotrienes in whole blood, gastric and duodenal mucosa, and synovial fluid in response to a nonselective COX and LOX inhibitor (tepoxalin), a COX-1–sparing drug (meloxicam), and a purported highly selective COX-1– sparing drug (firocoxib). The primary objective was to investigate activities of COX and LOX in duodenal mucosa of dogs. A second objective was to assess in vivo inhibition of the COX isoenzyme by firocoxib in several tissues in dogs with chronic OA. The final objective of the study was to test the hypothesis that calculations of the degree of COX-1 sparing based on in vitro data do not correlate well with in vivo concentrations of PGs induced by COX isoenzymes.
Materials and Methods
Animals—Eight mixed-breed adult hound-type dogs (3 males and 5 females; body weight range, 19.9 to 35 kg; age range, 3 to 8 years) with chronic, unilateral OA of the stifle joint were used; 1 dog had bilateral OA. Dogs were part of an existing research colony at the University of Georgia. Osteoarthritis was caused by surgically induced injury of the cranial cruciate ligament via a procedure performed several years previously. The protocols for that study were reviewed and approved by the University of Georgia Animal Care and Use Committee. To be included, dogs were required to be healthy (except for observable lameness on the affected pelvic limb) on the basis of findings from a complete physical examination, CBC, serum biochemical analysis, and urinalysis. Dogs received no other medication other than monthly heartworm preventive treatments during the entire experimental trial.
Experimental design—Dogs were allocated to treatment schedules on the basis of a randomized 4-way crossover design. Each dog received no treatment (placebo), firocoxiba (5 mg/kg, PO, q 24 h), meloxicamb (0.2 mg/kg then 0.1 mg/kg, PO, q 24 h), or tepoxalinc (20 mg/kg then 10 mg/kg, PO, q 24 h) for 7 days followed by a washout period of ≥ 3 weeks. All treatments were administered to dogs within a mass of canned dog food. Placebo treatment consisted of the food without additional drug. Meloxicam was chosen as a positive control treatment because of its known COX-1–sparing activity.9,10 The first day of treatment was designated day 0 (baseline); doses were administered after sample collection. The last samples were collected on the day after the last treatment was administered (ie, day 7).
On days 0, 2, 4, and 7, samples of whole blood, biopsy specimens of gastric and duodenal mucosae, and synovial fluid were obtained. Blood samples were collected via jugular venipuncture. Each dog was anesthetized with propofold (4 mg/kg, IV) and maintained on isoflurane. Synovial fluid was collected via arthrocentesis from both stifle joints. Gastroduodenoscopy was performed, and endoscopic biopsy specimens were collected from the gastric antrum near the pylorus and the duodenum proximal to the duodenal papilla. Investigators were unaware of the treatment each dog received, and the same people performed the sample collection (JPP) and sample processing (LRR).
Measurements of concentrations of PGs and leukotrienes in blood samples—To determine the effect of various treatments on concentrations of TxB2, a 6-mL blood sample was collected into an evacuated glass tube, immediately placed into a warm (37°C) water bath, and incubated for 1 hour. Indomethacine at a final concentration of 30μM was then added to stop additional synthesis of thromboxane. Serum was harvested, and serum concentration of TxB2 was measured via an ELISAe as described elsewhere.9
To determine the effect of various treatments on concentrations of PGE2, a 4-mL blood sample was collected into an evacuated heparinized tubef from which 500 μL of heparinized blood was removed and placed in a microcentrifuge tube. Ten microliters of lipopolysaccharide (Escherichia coli serotype 127:B8; 5 μg/mL)g was added to each tube to stimulate production of PGE2. Tubes were incubated in a 37°C water bath for 24 hours, and then the prostaglandins were separated out by use of a methanol extraction method. An ELISAe was used to measure the concentration of PGE2 in plasma.9
To measure the effect of various treatments on the synthesis of LTB4 in whole blood, blood samples were collected into heparinized tubes.f One milliliter of heparinized blood was transferred into microcentrifuge tubes, to which 50 μL of calcium ionophore A23187g was added to stimulate LTB4 production.11 The tubes were incubated in a 37°C water bath for 15 minutes, the reaction was stopped, and plasma was collected for analysis. The concentration of LTB4 in plasma was determined by use of an ELISA.e
Measurements of PGE2 in synovial fluid—Synovial fluid samples were collected from both (affected and unaffected) stifle joints via a standard arthrocentesis technique and placed in a microcentrifuge tube. The PGE2 concentration was measured as described elsewhere.9
Measurements of PG and leukotriene synthesis in the gastric and duodenal mucosa—All biopsy specimens obtained via endoscopy were processed within 8 minutes after collection from the gastrointestinal tract. Specimens weighing < 3 mg were not included in the final analysis. Synthesis of PGE1 and PGE2 was stimulated via mincing.9,12
Biopsy specimens for the LTB4 assay were collected at the same time as those for the assay of PG production. Fresh tissues were finely minced for 15 seconds, vortexed for 3 minutes, and pelleted. The supernatant was removed, and LTB4 concentrations were measured by use of an ELISA.10,e
Statistical analysis—A repeated-measures ANOVA was used to compare data within and among the types of treatments. When significant changes were detected, means of interest were compared by use of Tukey post hoc analysis. Values of P < 0.05 were considered significant.
Results
Adverse effects—Only 1 substantial adverse effect that required intervention was detected among the 8 dogs during the 4 different phases of treatment. Two days after the last dose of meloxicam was administered, a dog developed colitis that lasted for 3 days. That dog recovered after treatment with metronidazole (10 mg/kg, PO, q 12 h) and sulfasalazine (10 mg/kg, PO, q 12 h). Other adverse events that did not require intervention included 1 dog that vomited once during treatment with meloxicam, 4 dogs that vomited once while on firocoxib, and 1 dog that vomited once while being treated with tepoxalin. Of these dogs, 1 dog vomited twice, once while receiving tepoxalin and once while receiving firocoxib. One dog developed diarrhea 2 days after the last day of endoscopy when receiving the placebo treatment.
Concentrations of PGs and leukotrienes in blood—Treatment of dogs with tepoxalin was significantly associated with decreased serum concentrations of TxB2 at days 2, 4, and 7, compared with baseline serum concentrations for that treatment phase (Figure 1). Significant changes in TxB2 concentrations as a result of treatment with placebo, firocoxib, or meloxicam were not detected. At days 2, 4, and 7 of treatment, a significant decrease in serum concentrations of TxB2 was detected among dogs that received tepoxalin, compared with values among dogs that received placebo, firocoxib, or meloxicam.
Mean ± SEM serum concentrations of TxB2 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo (no treatment; black diamonds), firocoxib (5 mg/kg, PO, q 24 h; white squares), meloxicam (0.2 mg/kg then 0.1 mg/kg, PO, q 24 h; black circles), and tepoxalin (20 mg/kg then 10 mg/kg, PO, q 24 h; white triangles). A crossover design was used, with a washout period of ≥ 3 weeks between treatments. A repeated-measures ANOVA was used to compare concentrations among and within treatment categories. Values for treatment with tepoxalin were significantly (P < 0.05) different from those for treatment with placebo, firocoxib, and meloxicam at days 2, 4, and 7. Values for treatment with tepoxalin at days 2, 4, and 7 were significantly (P < 0.05) different from the value at day 0.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM serum concentrations of TxB2 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo (no treatment; black diamonds), firocoxib (5 mg/kg, PO, q 24 h; white squares), meloxicam (0.2 mg/kg then 0.1 mg/kg, PO, q 24 h; black circles), and tepoxalin (20 mg/kg then 10 mg/kg, PO, q 24 h; white triangles). A crossover design was used, with a washout period of ≥ 3 weeks between treatments. A repeated-measures ANOVA was used to compare concentrations among and within treatment categories. Values for treatment with tepoxalin were significantly (P < 0.05) different from those for treatment with placebo, firocoxib, and meloxicam at days 2, 4, and 7. Values for treatment with tepoxalin at days 2, 4, and 7 were significantly (P < 0.05) different from the value at day 0.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM serum concentrations of TxB2 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo (no treatment; black diamonds), firocoxib (5 mg/kg, PO, q 24 h; white squares), meloxicam (0.2 mg/kg then 0.1 mg/kg, PO, q 24 h; black circles), and tepoxalin (20 mg/kg then 10 mg/kg, PO, q 24 h; white triangles). A crossover design was used, with a washout period of ≥ 3 weeks between treatments. A repeated-measures ANOVA was used to compare concentrations among and within treatment categories. Values for treatment with tepoxalin were significantly (P < 0.05) different from those for treatment with placebo, firocoxib, and meloxicam at days 2, 4, and 7. Values for treatment with tepoxalin at days 2, 4, and 7 were significantly (P < 0.05) different from the value at day 0.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Treatment of dogs with firocoxib significantly decreased plasma concentrations of PGE2 at all time points, compared with baseline concentrations for that treatment phase (Figure 2). Treatment with meloxicam also decreased plasma concentrations of PGE2; however, only concentrations at days 4 and 7 were significantly different from baseline values. At days 2, 4, and 7, plasma concentrations of PGE2 were considerably lower for each of the 3 treatments, compared with values when no treatment was administered. In addition, treatment of dogs with firocoxib was associated with a significantly lower plasma concentration of PGE2 at days 2, 4, and 7, compared with corresponding concentrations in tepoxalin-treated dogs at the same time points.
Mean ± SEM plasma concentrations of lipopolysaccharide-stimulated PGE2 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for treatment with firocoxib, meloxicam, and tepoxalin were significantly (P < 0.05) different from values for treatment with placebo at days 2, 4, and 7. *Within the firocoxib treatment phase, value differed significantly (P < 0.05) from value at day 0. †Within the meloxicam treatment phase, value differed significantly (P < 0.05) from value at day 0. ‡Values for firocoxib-treated dogs differed significantly (P < 0.05) from values for tepoxalin-treated dogs at the same time point. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM plasma concentrations of lipopolysaccharide-stimulated PGE2 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for treatment with firocoxib, meloxicam, and tepoxalin were significantly (P < 0.05) different from values for treatment with placebo at days 2, 4, and 7. *Within the firocoxib treatment phase, value differed significantly (P < 0.05) from value at day 0. †Within the meloxicam treatment phase, value differed significantly (P < 0.05) from value at day 0. ‡Values for firocoxib-treated dogs differed significantly (P < 0.05) from values for tepoxalin-treated dogs at the same time point. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM plasma concentrations of lipopolysaccharide-stimulated PGE2 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for treatment with firocoxib, meloxicam, and tepoxalin were significantly (P < 0.05) different from values for treatment with placebo at days 2, 4, and 7. *Within the firocoxib treatment phase, value differed significantly (P < 0.05) from value at day 0. †Within the meloxicam treatment phase, value differed significantly (P < 0.05) from value at day 0. ‡Values for firocoxib-treated dogs differed significantly (P < 0.05) from values for tepoxalin-treated dogs at the same time point. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Plasma concentrations of LTB4 at day 7 were not significantly different from baseline concentrations in any of the 4 treatment groups (Figure 3). An apparent decrease in concentrations of LTB4 after treatment with tepoxalin was not significant (P = 0.07). Additionally, there were no significant differences in plasma concentrations of LTB4 among any of the groups at any given time point.
Mean ± SEM plasma concentrations of calcium ionophore–stimulated LTB4 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. No significant differences were detected among or within treatments. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM plasma concentrations of calcium ionophore–stimulated LTB4 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. No significant differences were detected among or within treatments. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM plasma concentrations of calcium ionophore–stimulated LTB4 in whole blood samples obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. No significant differences were detected among or within treatments. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Concentration of PGE2 in synovial fluid—Treatment of dogs with tepoxalin significantly decreased PGE2 concentrations in synovial fluid obtained from stifle joints unaffected by OA at days 2, 4, and 7, compared with baseline concentrations for that treatment phase (Figure 4). Treatment of dogs with firocoxib significantly decreased PGE2 concentrations in synovial fluid from unaffected stifle joints at days 2 and 7, compared with baseline values. Treatment with each of the 3 drugs was associated with a significantly lower concentration of PGE2 in synovial fluid on day 7, compared with concentrations in dogs treated with placebo.
Mean ± SEM concentrations of PGE2 in the synovial fluid of normal (A) and osteoarthritic (B) stifle joints of 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for treatment with firocoxib, meloxicam, and tepoxalin were significantly (P < 0.05) different from values for treatment with placebo at days 2, 4, and 7. *Within the tepoxalin treatment phase, value differed significantly (P < 0.05) from value at day 0. †Within the firocoxib treatment phase, value differed significantly (P < 0.05) from value at day 0. ‡Within the meloxicam treatment phase, value differed significantly (P < 0.05) from value at day 0. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM concentrations of PGE2 in the synovial fluid of normal (A) and osteoarthritic (B) stifle joints of 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for treatment with firocoxib, meloxicam, and tepoxalin were significantly (P < 0.05) different from values for treatment with placebo at days 2, 4, and 7. *Within the tepoxalin treatment phase, value differed significantly (P < 0.05) from value at day 0. †Within the firocoxib treatment phase, value differed significantly (P < 0.05) from value at day 0. ‡Within the meloxicam treatment phase, value differed significantly (P < 0.05) from value at day 0. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM concentrations of PGE2 in the synovial fluid of normal (A) and osteoarthritic (B) stifle joints of 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for treatment with firocoxib, meloxicam, and tepoxalin were significantly (P < 0.05) different from values for treatment with placebo at days 2, 4, and 7. *Within the tepoxalin treatment phase, value differed significantly (P < 0.05) from value at day 0. †Within the firocoxib treatment phase, value differed significantly (P < 0.05) from value at day 0. ‡Within the meloxicam treatment phase, value differed significantly (P < 0.05) from value at day 0. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Treatment of dogs with firocoxib and tepoxalin was associated with decreased concentrations of PGE2 in synovial fluid of stifle joints affected by OA at days 2, 4, and 7, compared with baseline concentrations. Treatment with meloxicam was associated with decreased concentrations of PGE2 in these stifle joints at days 4 and 7 only. Treatment with each of the 3 drugs resulted in significantly lower concentrations of PGE2 in stifle joints affected by OA on day 7, compared with concentrations in placebo-treated dogs.
Synthesis of PGs in gastric mucosa—Treatment of dogs with tepoxalin significantly decreased the rates of PGE1 and PGE2 synthesis by the gastric mucosa at all 3 time points, compared with baseline rates (Figure 5). In addition, treatment with tepoxalin significantly decreased the rates of PGE1 and PGE2 synthesis at days 2, 4, and 7, compared with rates for the other 2 drugs and placebo at the same time points. Rates of PGE1 and PGE2 synthesis by the gastric mucosa of dogs treated with firocoxib, meloxicam, and placebo did not differ significantly at any time points, compared with baseline rates.
Mean ± SEM concentration of PGE2 (A) and PGE1 (B) in gastric mucosa obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for tepoxalin were significantly (P < 0.05) different from placebo, firocoxib, and meloxicam at days 2, 4, and 7. Values for tepoxalin at days 2, 4, and 7 were significantly (P < 0.05) different from the value at day 0. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM concentration of PGE2 (A) and PGE1 (B) in gastric mucosa obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for tepoxalin were significantly (P < 0.05) different from placebo, firocoxib, and meloxicam at days 2, 4, and 7. Values for tepoxalin at days 2, 4, and 7 were significantly (P < 0.05) different from the value at day 0. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM concentration of PGE2 (A) and PGE1 (B) in gastric mucosa obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Values for tepoxalin were significantly (P < 0.05) different from placebo, firocoxib, and meloxicam at days 2, 4, and 7. Values for tepoxalin at days 2, 4, and 7 were significantly (P < 0.05) different from the value at day 0. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Synthesis of PGs in duodenal mucosa—Treatment of dogs with tepoxalin significantly decreased the rate of PGE2 synthesis by the duodenal mucosa at days 2, 4, and 7, compared with rates measured in duodenal mucosa at the same time points during the placebo phase (Figure 6). Rates of PGE2 synthesis by the duodenal mucosa were not significantly different from baseline values after any of the 4 treatments, nor were there any significant differences between effects of any 2 drugs at any 1 time. Rates of synthesis of PGE1 by the duodenal mucosa were not significantly different within treatments or among treatments at any time point. In general, rates of synthesis of PGE1 and PGE2 were approximately 10-fold as high in gastric versus duodenal mucosae.
Mean ± SEM rates of synthesis of PGE2 (A) and PGE1 (B) in duodenal mucosa specimens obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Rate of PGE2 synthesis for tepoxalin was significantly (P < 0.05) different from the rate for placebo, firocoxib, and meloxicam at days 2, 4, and 7. No other significant differences were detected. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM rates of synthesis of PGE2 (A) and PGE1 (B) in duodenal mucosa specimens obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Rate of PGE2 synthesis for tepoxalin was significantly (P < 0.05) different from the rate for placebo, firocoxib, and meloxicam at days 2, 4, and 7. No other significant differences were detected. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM rates of synthesis of PGE2 (A) and PGE1 (B) in duodenal mucosa specimens obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Rate of PGE2 synthesis for tepoxalin was significantly (P < 0.05) different from the rate for placebo, firocoxib, and meloxicam at days 2, 4, and 7. No other significant differences were detected. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Concentration of LTB4 in the gastric and duodenal mucosa—Concentrations of LTB4 did not change significantly after any treatment and were not significantly different among treatments at any time point (Figure 7).
Mean ± SEM concentrations of LTB4 in specimens of gastric mucosa (A) and duodenal mucosa (B) obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Differences among and within the various treatments were not significant. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM concentrations of LTB4 in specimens of gastric mucosa (A) and duodenal mucosa (B) obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Differences among and within the various treatments were not significant. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Mean ± SEM concentrations of LTB4 in specimens of gastric mucosa (A) and duodenal mucosa (B) obtained from 8 dogs before treatment (day 0) and 2, 4, and 7 days after treatment with placebo, firocoxib, meloxicam, and tepoxalin. Differences among and within the various treatments were not significant. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1203
Discussion
Results of the study reported here support the supposition that tepoxalin nonselectively inhibits COX isoenzymes and that meloxicam and firocoxib spare inhibition of COX-1 in whole blood of dogs. Although treatment of dogs with each of the 3 NSAIDs resulted in decreased concentrations of PGE2, only treatment with firocoxib resulted in a decreased concentration of PGE2 at day 2, compared with the baseline concentration. This significant decrease was not detected after treatment of dogs with meloxicam until day 4, which is interesting because, in another study,10 treatment of dogs with meloxicam (0.1 mg/kg, PO, q 24 h) resulted in significantly lower concentrations of PGE2 in whole blood on day 3, compared with baseline concentrations. In that study, a sample was not obtained between day 0 and day 3. The finding that firocoxib appeared to work faster may reflect a more rapid ability of firocoxib to suppress the production of PGE2 in whole blood, compared with the ability of other NSAIDs evaluated, or it may be attributable to the study design (eg, small sample size and consequent potential type II errors). Treatment of dogs with tepoxalin did not result in a significant decrease in plasma concentrations of LTB4 during the 8-day study period. Although that finding was somewhat surprising, another study10 revealed that treatment of dogs with tepoxalin (10 mg/kg, PO, q 24 h) does not yield significant decreases in whole blood concentrations of LTB4 until day 10.
Evaluation of the synovial fluid of OA stifle joints revealed some interesting findings. Treatment of dogs with firocoxib or meloxicam significantly decreased concentrations of PGE2 in synovial fluid in the same general pattern as was detected for concentrations of PGE2 in plasma; however, treatment with tepoxalin was associated with a different pattern. Tepoxalin did not significantly decrease production of PGE2 in whole blood at any time point, but it did decrease production of PGE2 in stifle joints affected by OA. This finding is most likely attributable to the possibility that PGE2 in synovial fluid is derived from COX-1 and COX-2 activities, which are suppressed by tepoxalin, whereas the PGE2 in the whole blood assay is solely derived from COX-2 activity. Thus, a proportion of the PGE2 in the synovial fluid is derived from COX-1 activity. Data from the unaffected stifle joints appear to confirm these findings because treatment with tepoxalin significantly decreased concentrations of PGE2 in synovial fluid at all time points, compared with baseline concentrations. However, the finding that firocoxib significantly inhibited the production of PGE2 at all but 1 time point in both joints suggests that COX-2 induces production of most PGE2 in the synovial fluid. When considered collectively, these data strongly support the importance of in vivo analysis of the effect of NSAIDs on COX isoenzyme inhibition in target tissues.
Production of PGs by the gastric mucosa revealed similar results. Predictably, treatment of dogs with tepoxalin resulted in decreased rates of production of PGE1 and PGE2 at all time points, compared with production at baseline or after treatment with placebo, meloxicam, or firocoxib. The reduction in the rate of PGE1 production is consistent with the COX-1 inhibitory effects of tepoxalin. These data, along with the findings that treatment of dogs with firocoxib and meloxicam did not alter production of PGE1 or PGE2 in the gastric mucosa, strongly suggest that most of the PGE1 and PGE2 produced in the gastric mucosa is induced by COX-1.
The most striking result of the evaluation of the production of the PGs in the gastrointestinal mucosa was that, in general, amounts of PGE1 and PGE2 produced by gastric mucosa were approximately 10 times as high as amounts produced by duodenal mucosa. The increased production of the PGs in the stomach is most likely attributable to the physiologic need to protect the gastric mucosa from the acidic environment. Conversely, although production of PG in the duodenum plays a role in mediating duodenal blood flow, motility, and mucus secretion, duodenal PGs play no role in mediating the secretion of HCl by parietal cells and a much less important role in secretion of bicarbonate,13–15 compared with gastric PGs. Specific data on the inhibition of PGE1 and PGE2 production in the duodenum were similar to those in the gastric mucosa. Given the relatively low amounts of PGs measured in the duodenum, the lack of significant differences in PGE1 and PGE2 production detected after dogs were treated with tepoxalin is likely attributable to a type II error, particularly with respect to production of PGE1. Still, treatment with tepoxalin resulted in significantly lower rates of PGE2 production and some (albeit insignificant) decline in rates of PGE1 production, compared with baseline rates. Similar to the lack of effect on the gastric mucosa, meloxicam and firocoxib had no effect on PGE1 and PGE2 production in the duodenal mucosa of treated dogs. These findings again suggest that COX-1 has a central role in PG production in the duodenum and that the contribution of COX-2 is comparatively small.
None of the NSAIDs evaluated in our study appeared to influence production of LTB4 by the gastric and duodenal mucosae. In another study,10 treatment of dogs with tepoxalin significantly suppressed production of LTB4 by the gastric mucosa after 10 days but not after 3 days of treatment. Because we evaluated the effects of 7 days of treatment, one can argue that > 1 week of treatment with tepoxalin is required to significantly suppress production of LTB4. Alternatively, LOX is reportedly an inducible enzyme that stimulates production of low concentrations (2.5 pg/mg) of LTB4 by gastric mucosa in rats.16 In the present study, no effort was made to induce the expression of LOX genes by leukocytes or the gastrointestinal mucosa. In future studies, it may be necessary to stimulate the expression of LOX genes via inflammation to obtain suppressible LTB4 levels.
A final question concerns the importance of the in vitro COX-1:COX-2 ratio for predicting degrees of inhibition of various PGs in actual tissues.11 For example, one may speculate that a drug that is a highly selective COX-2 inhibitor will be a greater inhibitor of PGE2 production than will a moderately selective drug. This supposition is not true for synovia of dogs with OA.11 However, results regarding plasma PGE2 concentrations, a measure of COX-2 production, in dogs with OA in the present study suggest that some association may exist between in vitro COX-2 selectivity measurements and in vivo PG production, at least at day 2. By day 2, treatment of dogs with firocoxib (in vitro COX-1:COX-2 ratio of 384.0)17 resulted in significantly lower PGE2 production, compared with baseline values, whereas treatment with meloxicam (in vitro COX-1:COX-2 ratio of 2.72)18 did not. However, a power analysis (β = 0.80) suggested that our study did not have enough power and that 10 dogs would have been necessary to detect a mean difference of 25% between firocoxib and meloxicam. Similar results were obtained for synovial fluid obtained from stifle joints affected by OA. On day 2, treatment of dogs with firocoxib significantly decreased the concentration of PGE2 in synovial fluid, compared with baseline concentrations. On the other hand, treatment with meloxicam failed to yield a significant decrease in the concentration of PGE2 in synovial fluid. These data are not in conflict with results from the other study10 because that study measured PGE2 concentrations on days 3 and 10. Furthermore, in the present study, the abilities of firocoxib and meloxicam to inhibit production of PGs in stifle joints of dogs after 4 and 7 days of treatment were not significantly different, although the 2 NSAIDs have different in vitro COX-1:COX-2 ratios.11 Thus, the importance of the actual difference of in vitro COX-1:COX-2 ratios in predicting in vivo enzyme production is still unclear but may have some usefulness in the period immediately following administration of NSAIDs.
ABBREVIATIONS
| COX | Cyclooxygenase |
| LOX | Lipoxygenase |
| LTB4 | Leukotriene B4 |
| OA | Osteoarthritis |
| PG | Prostaglandin |
| TxB2 | Thromboxane B2 |
Previcox, Merial Ltd, Duluth, Ga.
Metacam, Boehringer Ingelheim Vetmedica Inc, St Joseph, Mo.
Zubrin, Schering-Plough Animal Health Corp, Union, NJ.
Rapinovet, Schering-Plough Animal Health Corp, Union, NJ.
Thromboxane B2, PGE2, and LTB4 enzyme immunoassay kits, Cayman Chemical Co, Ann Arbor, Mich.
BD Vacutainer plasma tubes, Becton-Dickinson, Franklin Lakes, NJ.
Sigma Chemical Co, St Louis, Mo.
References
- 1.
Adebayo D, Bjarnason I. Is non-steroidal anti-inflammatory drug (NSAID) enteropathy clinically more important than NSAID gastropathy? Postgrad Med J 2006;82:186–191.
- 2.
Bergh MS, Budsberg SC. The coxib NSAIDs: potential clinical and pharmacologic importance in veterinary medicine. J Vet Intern Med 2005;19:633–643.
- 3.
Sigthorsson G, Simpson RJ, Walley M, et al. COX-1 and 2, intestinal integrity and pathogenesis of NSAID-enteropathy in mice. Gastroenterology 2002;122:1913–1923.
- 4.
Whittle BJR. Mechanism underlying intestinal injury induced by antiinflammatory COX inhibitors. Eur J Pharmacol 2004;500:427–439.
- 5.↑
Kawai S. Cyclooxygenase selectivity and the risk of gastro-intestinal complications of various non-steroidal anti-inflammatory drugs: a clinical consideration. Inflamm Res 1998;47(suppl 2):S102–S106.
- 6.
Scheiman JM. NSAIDs, gastrointestinal injury, and cytoprotection. Gastroenterol Clin North Am 1996;25:279–298.
- 7.↑
Jones RD, Baynes RE, Nimitz CT. Nonsteroidal anti-inflammatory drug toxicosis in dogs and cats: 240 cases (1989–1990). J Am Vet Med Assoc 1992;201:475–477.
- 8.↑
Lascelles BD, Blikslager AT, Fox SM, et al. Gastrointestinal tract perforation in dogs treated with a selective cyclooxygenase-2 inhibitor: 29 cases (2002–2003). J Am Vet Med Assoc 2005;227:1112–1117.
- 9.↑
Jones CJ, Streppa HK, Harmon BG, et al. In vivo effects of meloxicam and aspirin on blood, gastric mucosal, and synovial fluid prostanoid synthesis in dogs. Am J Vet Res 2002;63:1527–1531.
- 10.↑
Agnello KA, Reynolds LR, Budsberg SC. In vivo effects of tepoxalin, an inhibitor of cyclooxygenase and lipoxygenase, on prostanoid and leukotriene production in dogs with chronic osteoarthritis. Am J Vet Res 2005;66:966–972.
- 11.↑
Sessions JK, Reynolds LR, Budsberg SC. In vivo effects of carprofen, deracoxib, and etodolac on prostanoid production in blood, gastric mucosa, and synovial fluid in dogs with chronic osteoarthritis. Am J Vet Res 2005;66:812–817.
- 12.
Argentieri DC, Ritchie DM, Ferro MP, et al. Tepoxalin: a dual cyclooxygenase/5-lipoxygenase inhibitor of arachidonic acid metabolism with potent anti-inflammatory activity and a favorable gastrointestinal profile. J Pharmacol Exp Ther 1994;271:1399–1408.
- 13.
Takeuchi K, Ukawa H, Kato S, et al. Impaired duodenal bicarbonate secretion and mucosal integrity in mice lacking prostaglandin E-receptor subtype EP3. Gastroenterology 1999;117:1128–1135.
- 14.
Akiba Y, Guth PH, Engel E, et al. Dynamic regulation of mucus gel thickness in rat duodenum. Am J Physiol Gastrointest Liver Physiol 2000;279:G437–G447.
- 15.
Narita T, Okabe N, Hane M, et al. Nonsteroidal antiinflammatory drugs induce hypermotilinemia and disturbance of interdigestive migrating contractions in instrumented dogs. J Vet Pharmacol Ther 2006;29:569–577.
- 16.↑
Tries S, Neupert W, Laufer S. The mechanism of action of the new antiinflammatory compound ML3000: inhibition of 5-LOX and COX-1/2. Inflamm Res 2002;51:135–143.
- 17.↑
McCann ME, Andersen DR, Zhang D, et al. In vitro effects and in vivo efficacy of a novel cyclooxygenase-2 inhibitor in dogs with experimentally induced synovitis. Am J Vet Res 2004;65:503–512.
- 18.↑
Streppa HK, Jones CJ, Budsberg SC. Cyclooxygenase selectivity of nonsteroidal anti-inflammatory drugs in canine blood. Am J Vet Res 2002;63:91–94.
