Carprofen is a widely used medication prescribed to alleviate chronic pain in dogs.1-3 Although all NSAIDs have the potential for causing gastrointestinal, renal, and hepatic problems, carprofen is generally considered to be well tolerated by dogs.1-5 In several studies,6-8 the drug was not found to induce substantial lesions in the gastric mucosa of dogs,and when administration of carprofen is combined with general anesthesia in healthy dogs, reported effects on renal function have typically been minor,9-12 with only a few investigators reporting decreased function.13 However, severe adverse effects associated with carprofen administration, including hepatocellular toxicosis14 and neutrophilic dermatitis with immune-mediated anemia and thrombocytopenia,15 have been described.
Clinical and experimental studies9-13 on the adverse effects of carprofen have concentrated on dogs in the perioperative and perianesthetic periods. In clinical trials of treatment of chronic pain and in tolerance studies,1,2,4,8 treatment or follow-up periods have typically not extended beyond 2 weeks. The number of clinical or experimental reports3,5,7 in which adverse effects of carprofen were evaluated after administration for several weeks or months is limited. To our knowledge, no placebo-controlled clinical trials on the adverse effects of long-term carprofen administration in dogs have been published.
The aim of the present study was to evaluate potential adverse effects, especially those involving the kidneys and liver, of a 2-month course of orally administered carprofen in dogs. Our hypothesis was that long-term administration would alter certain blood or urine biochemical variables and cause clinically detectable adverse effects.
Materials and Methods
Dogs—Twenty-two dogs with clinical signs and a radiographic diagnosis of chronic osteoarthritis in 1 or both hip or elbow joints were included. Mean ± SD weight was 36.5 ± 8.4 kg (80.3 ± 18.5 lb), and mean age was 5.7 ± 2.5 years (range, 1 to 11 years). Ten of the dogs were female, and 12 were male. Seven were German Shepherd Dogs, and the rest represented 10 other breeds.
Samples for this study were collected as part of a larger studya in which the efficacies of different treatments for chronic orthopedic pain were compared. Owners were instructed to refrain from administering NSAIDs or cortico-steroids for at least 30 days and sodium pentosan polysulfate for at least 90 days prior to the initiation of the study (as inclusion criteria for the efficacy studya). During the present trial, no medications other than routine vaccinations and antiparasitic treatments were allowed. However, for ethical reasons, at the beginning of the trial, owners were supplied with a supplementary package of carprofenb in normal packaging to be used as additional pain relief (dosage: 1 tablet [50 mg] once daily for dogs weighing 20 to 30 kg [44 to 66 lb], 2 tablets for dogs weighing 31 to 40 kg [68 to 88 lb], and 3 tablets for dogs weighing 41 to 60 kg [90 to 132 lb]) if the owner judged the dog to be in considerable pain. Use of extra carprofen was recorded by the owner. Dogs that received > 2 extra doses of carprofen during any 1 week or > 5 extra doses during the entire study period were excluded from the present study. Eleven dogs included in the larger study (4 in the carprofen group and 7 in the placebo group) were excluded from our study. All owners signed consent forms, and the study protocol was approved by the Ethics Committee of the University of Helsinki.
Treatment groups—Dogs were given carprofen (approx dose, 4 mg/kg [1.8 mg/lb], PO; n = 13 dogs) or placebo (9) daily for 8 weeks. Tablets were supplied by the manufacturer in 2 varieties (the test product and the placebo), the appearance of which was identical.
Sample collection—Serum and voided urine samples were collected before initiation of treatment and after 4 and 8 weeks of treatment. Samples were centrifuged (1,300 × g for 10 minutes) and stored at 4°C until analysis. Analyses were performed within 24 hours of collection for all measurements except urine protein and creatinine concentrations, which were analyzed from urine samples frozen at −20°C. Dogs' weights were recorded at the time of sample collection. Colorimetric methods were used for determination of serum total protein,16 albumin,17 and creatinine concentrations and for urine creatinine concentration.18,c Urine protein was determined by use of a pyrogallol red-molybdate complex,19,c and serum urea was measured by use of a kinetic enzymatic method.20,c Serum ALT activity was determined according to the European Committee for Clinical Laboratory Standards,21,c and serum and urine ALP actitivities were determined according to recommendations of the Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology.22,c Serum and urine GGT concentrations were determined by use of an International Federation of Clinical Chemistry and Laboratory Medicine method.23,c Urinary enzyme and protein measurements were determined proportionate to urine creatinine concentration (reported as IU/mmol of creatinine for the urinary ALP-to-creatinine and GGT-to-creatinine ratios and as g/g of creatinine for the urinary protein-to-creatinine ratio) to correct for variation in urine production.
Owners observed dogs for adverse effects and summarized observations in a questionnaire administered at the time of initiation of treatment and at 4 and 8 weeks after treatment was started. Vomiting, diarrhea, and skin reactions (itching or eczema) were scored from 1 (none) to 5 (detected almost every day). Appetite was also scored on a scale from 1 (very good) to 5 (very poor).
Statistical analysis—Parametric data were analyzed by use of ANOVA for repeated measures. When differences were detected, the Student t test was used for pairwise comparisons between groups. Differences between groups at different time points concerning appetite, vomiting, diarrhea, and skin reactions were analyzed via the Mann-Whitney U test.d Significance was set at P < 0.05.
Results
Mean serum protein and albumin concentrations were significantly lower in dogs treated with carprofen than in dogs that received the placebo at 4 weeks, but not at 8 weeks. None of the dogs had serum protein concentrations outside reference range (Table 1). One dog in the carprofen group had serum albumin concentrations below reference range at each sampling but had no other abnormal laboratory or clinical findings during the follow-up period. Another dog in the carprofen group had a low serum albumin concentration after 4 weeks of treatment. That dog concurrently had a serum ALT activity greater than reference range and was the only dog to lose > 10% of its original body weight during the treatment period (from a baseline weight of 40.0 to 34.5 kg (88 to 76 lb) after 8 weeks of treatment). That dog's owner detected no adverse clinical signs.
Mean ± SD (range; number of dogs with values outside reference range, if any) of serum biochemistry variables, urinary ratios, and weights of dogs that were treated with carprofen (n = 13 dogs) or placebo (9) for 8 weeks. The dosage of carprofen administered was 4 mg/kg (1.8 mg/lb), PO, once daily. Urine and serum samples were collected before and 4 and 8 weeks after initiation of treatment.
Variable | Baseline | 4 weeks | 8 weeks |
---|---|---|---|
Protein (g/L) | |||
Carprofen | 66.5 ± 3.3 (61.5–72.4) | 62.2 ± 2.8 (59.1–68.6)* | 63.8 ± 3.6 (58.9–73.0) |
Placebo | 66.4 ± 5.3 (58.2–71.8) | 66.2 ± 5.3 (61.4–77.0) | 65.6 ± 4.8 (58.3–73.4) |
Albumin (g/L) | |||
Carprofen | 34.5 ± 2.5 (28.0–37.9; 1/13) | 33.9 ± 2.9 (26.6–37.3; 2/13)* | 33.4 ± 3.0 (25.4–36.4; 1/13) |
Placebo | 34.9 ± 1.1 (33.5–36.7) | 36.2 ± 1.5 (33.6–38.0) | 35.1 ± 1.4 (33.0–38.0) |
Urea (mmol/L) | |||
Carprofen | 6.0 ± 2.0 (3.2–10.5; 1/13) | 5.9 ± 2.5 (3.0–12.5; 1/13) | 5.6 ± 1.6 (2.8–8.3) |
Placebo | 5.9 ± 1.0 (4.0–6.8) | 6.7 ± 1.1 (5.2–8.5) | 6.2 ± 1.7 (3.9–8.5) |
Creatinine (μmol/L) | |||
Carprofen | 95.3 ± 17.0 (66–123; 1/13) | 96.6 ± 17.4 (63–124; 2/13) | 93.7 ± 12.8 (64–113) |
Placebo | 96.4 ± 12.5 (82–114) | 94.2 ± 15.2 (71–119; 1/9) | 93.3 ± 10.7 (80–106) |
ALP activity (U/L) | |||
Carprofen | 146.4 ± 71.0 (61–273; 2/13) | 111.2 ± 72.9 (42–327; 1/13) | 111.5 ± 53.8 (45–245; 1/13) |
Placebo | 109.0 ± 61.0 (50–218; 1/9) | 79.7 ± 50.1 (35–201) | 116.9 ± 101.4 (39–352; 1/9) |
ALT activity (U/L) | |||
Carprofen | 43.3 ± 12.3 (29–67) | 51.0 ± 29.6 (30–136; 2/13) | 48.5 ± 23.2 (23–107; 2/13) |
Placebo | 39.0 ± 15.7 (15–64) | 35.8 ± 12.2 (18–57) | 49.4 ± 46.8 (16–170; 1/9) |
Urinary ALP-to-creatinine (IU/mmol) | |||
Carprofen | 3.2 ± 2.8 (0.4–8.1) | 2.9 ± 4.2 (0.2–15.8) | 2.3 ± 2.0 (0.8–7.6) |
Placebo | 1.5 ± 1.6 (0.3–5.4) | 5.3 ± 6.2 (0.3–16.1) | 2.3 ± 3.4 (0.3–10.6) |
Urinary GGT-to-creatinine (IU/mmol) | |||
Carprofen | 2.4 ± 1.1 (0.5–4.2) | 2.8 ± 1.2 (1.3–5.6) | 2.5 ± 0.8 (1.4–4.0) |
Placebo | 3.55 ± 4.2 (1.5–14.7) | 4.1 ± 4.9 (1.6–16.9) | 3.4 ± 2.9 (1.6–10.5) |
Urninary protein-to-creatinine (g/g) | |||
Carprofen | 0.07 ± 0.04 (0.02–0.13) | 0.09 ± 0.10 (0.02–0.35) | 0.08 ± 0.07 (0.03–0.29) |
Placebo | 0.13 ± 0.15 (0.04–0.43) | 0.12 ± 0.14 (0.03–0.47) | 0.13 ± 0.10 (0.02–0.29) |
Body weight (kg) | |||
Carprofen | 37.8 ± 8.4 (28.9–56.0) | 37.3 ± 8.3 (29.0–55.0) | 37.0 ± 8.3 (29.0–54.7) |
Placebo | 34.5 ± 8.3 (25.5–46.5) | 35.6 ± 10.4 (25.4–56.0) | 35.0 ± 9.8 (24.0–53.2) |
Values are significantly (P < 0.05) different from placebo group.
To convert kilograms to pounds, multiply by 2.2.
Mean values for serum urea and creatinine concentrations, serum ALP and ALT activities; and urinary ALP-to-creatinine, GGT-to-creatinine, and protein-tocreatinine ratios did not change, and no differences in those variables were observed between groups (Table 1). One dog in the carprofen group had a high serum urea concentration before treatment and 4 weeks after initiation of treatment, but the concentration was not high when measured at 8 weeks. That dog also had a high serum creatinine concentration at 4 weeks and was reported by the owner to have had a skin reaction and to have vomited and had diarrhea about once a week during the first 4 weeks of treatment. In addition, another dog in the carprofen group and 1 dog in the placebo group had a high serum creatinine concentration at 4 weeks, but not at 8 weeks.
Baseline values for serum ALT activity did not exceed reference values in any dogs, but ALT activity was high in 1 or both samples taken after initiation of treatment in 3 dogs treated with carprofen and in 1 dog in the placebo group (Table 1). One of the carprofentreated dogs was reported by the owner to have vomited once or twice a month after treatment commenced. Baseline serum ALP activity was higher than reference range in 2 dogs in the carprofen group, 1 of which had high values in all samples, and in 1 dog in the placebo group. The placebo-treated dog with high serum ALT activity at 4 weeks also had high serum ALP activity in the same sample, but not in the baseline sample.
None of the dogs had a high urinary ALP-to-creatinine or GGT-to-creatinine ratio (ie, > 20.0 IU/mmol of creatinine24) in any sample (Table 1). One dog in the placebo group had an intermediate value for the urinary GGT-to-creatinine ratio (10.0 to 20.0 IU/mmol of creatinine) in all samples, and 1 carprofen-treated and 2 placebo-treated dogs had intermediate values for the urinary ALP-to-creatinine ratio (10.0 to 20.0 IU/mmol of creatinine) at 4 weeks. No dogs had a protein-to-creatinine ratio higher than reference range for healthy dogs (< 0.5 g/g of creatinine)25 in any sample.
Dogs' mean weights did not change significantly during the trial, and no differences between groups were detected. Few dogs vomited or had diarrhea during the follow-up period; dogs that did have vomiting did so at a frequency of once a week or less. Four carprofen-treated and 3 placebo-treated dogs had signs (eg, vomiting, diarrhea, and development of skin reactions) that had not been detected by the owner during the month before initiation of treatment. The frequency of vomiting, diarrhea, and development of skin reactions reported by owners did not differ between groups. Dogs' appetites were better in the carprofen group than in the placebo group during the first 4 weeks (P = 0.037) as well as the interval from weeks 4 to 8 (P = 0.006).
Discussion
Renal syndromes that have been associated with NSAID toxicosis include reversible renal insufficiency, acute and chronic renal failure, interstitial nephritis, papillary necrosis, and nephrotic syndrome.26,27 We assessed renal function by measuring serum creatinine and urea concentrations. Although these variables are not highly sensitive indicators of decreased renal function, they have been used as the sole methods1,2,28 or in combination with either fractional clearance of sodium10 or 24-hour endogenous creatinine clearance13 in clinical trials investigating the effects of carprofen in dogs. In experimental studies, more sophisticated detection methods, such as renal scintigraphic imaging9,11 or plasma clearance of 99mTc-diethylenetri-aminepentaacetic acid,12 have been used to evaluate the impact of carprofen on renal function in dogs. In the present study, no findings suggestive of altered renal function resulting from carprofen administration were detected, a finding that was in agreement with those of several earlier reports.1,2,9,11,12,28 We did not perform a 24-hour endogenous creatinine-clearance study, a methodology that has been used previously to reveal decreased renal function in dogs treated with carprofen at the time of anesthetic induction.13
Increases in urinary brush-border enzyme activities (GGT and ALP) have been associated with renal proximal tubular damage in dogs.29 Prolonged administration of carprofen did not change these activities in relation to urine creatinine concentration in our study. This result is consistent with those from an earlier report,11 in which carprofen administered to dogs before or during anesthesia did not alter the urinary ALP-to-creatinine ratio. Proteinuria of renal origin may be caused by glomerular leak, failure of proximal tubular absorption, or both, but in dogs with glomerular proteinuria, the urinary protein-to-creatinine ratio is > 1.0.25 In our study, the protein-to-creatinine ratio was less than reference range for healthy dogs in all samples and administration of carprofen did not affect this. We found no evidence of carprofen-induced proximal tubule damage or glomerular leak.
In general, carprofen did not appear to influence serum liver enzyme activities. In an earlier study14 of clinical cases, all dogs with carprofen-induced hepatocellular toxicosis had increased serum ALT activity, and most also had high serum ALP and AST activities. In our study, 1 carprofen-treated dog with high ALT activity also had low serum albumin concentration and clinically important weight loss. Although no other clinical signs indicative of carprofen-induced hepatocellular toxicosis, such as inappetence, vomiting, or icterus,14 were observed, the possibility that that dog had hepatocellular toxicosis could not be ruled out. This seems improbable, however, because serum ALT activity and albumin concentration had returned to reference range when analyzed again at 8 weeks even though treatment with carprofen had been continued. The prevalence of adverse hepatic reactions associated with carprofen is likely low, considering the small number of instances in which adverse clinical signs or increased serum activities of hepatic enzymes have been reported, relative to the wide use of the drug.1,3,14 Clinical signs indicative of hepatocellular toxicosis were usually noticed within the first month after initiation of treatment, but they could develop even after several months.14 The follow-up time in our study was therefore sufficiently long to reveal most of the susceptible dogs; however, the probability that susceptible dogs were included in the small treatment group was low.
The effects of carprofen on gastric mucosa in dogs have been studied by use of gastroscopy.6-8 The clinical importance of the transient decrease in prostaglandin E2detected in gastric mucosa in association with carprofen administration has been considered negligible8 because no substantial carprofen-induced gross lesions of the mucosa were observed.6-8 In the present study, gastroscopy was not performed on the client-owned dogs. We did not test feces for occult blood because diet can affect the specificity of peroxidase-based tests for detection of occult blood in canine feces30 and we could not insist that owners modify their dog's diets. In an earlier study,2 dogs treated with carprofen for 2 weeks did not have positive results for fecal occult blood tests more frequently than did placebo-treated dogs.2 In general, previous studies2,4,6-8 with shorter follow-up times have revealed that carprofen is well tolerated in the gastrointestinal tract. In our study, no clinical signs of gastrointestinal irritation, such as vomiting, diarrhea, or poor appetite, were observed in the carprofen-treated dogs; in fact, administration of carprofen seemed to enhance appetite. However, the weight loss and hypoalbuminemia detected in 1 dog and the slight decrease in mean serum albumin concentration in the carprofen-treated group may have arisen from gastrointestinal irritation ascribable to effects of carprofen.
Administration of carprofen was not observed to induce skin reactions. Carprofen-related skin reactions are likely rare in dogs, although 1 report15 of a suspected adverse dermatitic reaction has been published.
In conclusion, the slight and transient decreases in mean serum total protein and albumin concentrations in carprofen-treated dogs may have been caused by altered mucosal permeability in the gastrointestinal tract because no signs of renal or hepatic toxicosis were detected that could account for those findings. However, no clinical signs of gastrointestinal irritation were observed, and dogs treated with carprofen had better appetites than dogs in the control group. Results suggest that long-term oral administration of carprofen is generally well tolerated, although adverse effects may develop in some dogs.
NSAID | Nonsteroidal anti-inflammatory drug |
ALP | Alkaline phosphatase |
ALT | Alanine aminotransferase |
GGT | γ-Glutamyltransferase |
Hielm-Björkman A, Tulamo R-M, Raekallio M. Feed supplement Lyproflex (Perna canaliculus) and antihomotoxic Zeel as treatments for chronic pain due to osteoarthritis: a randomized double-blind double-controlled study (abstr), in Proceedings. 3rd Int Symp Rehabil Phys Ther Vet Med 2004;209.
Rimadyl, 50-mg tablets, Pfizer, Espoo, Finland.
KONE Pro Selective chemistry analyzer, Thermo Electron Oy, Vantaa, Finland.
SPSS for Windows, version 13.0, SPSS Inc, Chicago, Ill.
- 1
Holtsinger SA, Parker RB, Beale BS, et al. The therapeutic efficacy of carprofen (Rimadyl-V™) in 209 clinical cases of canine degenerative joint disease. Vet Comp Orthop Traumatol 1992; 5: 140–144.
- 2↑
Vasseur PB, Johnston AL, Budsberg SC, et al. Randomized, controlled trial of the efficacy of carprofen, a non-steroidal antiinflammatory drug, in treatment of osteoarthritis in dogs. J Am Vet Med Assoc 1995; 206: 807–811.
- 3
Lønaas L, Nyman G, Palmér E. Anti-inflammatory and analgesic treatment of osteoarthritis in the dog—experiences from continuous treatment with carprofen [Swedish]. Svensk Veterinärtidning 1999; 51: 707–711.
- 4
McKellar QA, Pearson T, Bogan JA, et al. Pharmacokinetics, tolerance and serum thromboxane inhibition of carprofen in the dog. J Small Anim Pract 1990; 31: 443–448.
- 5
EMEA/MRL/528/98-FINAL, Committee for Veterinary Medicinal Products, Carprofen, Summary report (2), 1999. Available at: www.emea.eu.int/pdfs/vet/mrls/052899en.pdf. Accessed Jan 1999.
- 6
Forsyth SF, Guilford WG, Haslett SJ, et al. Endoscopy of the gastroduodenal mucosa after carprofen, meloxicam and ketoprofen administration in dogs. J Small Anim Pract 1998; 39: 421–424.
- 7
Reimer ME, Johnston SA, Leib MS, et al. The gastroduodenal effects of buffered aspirin, carprofen, and etodolac in healthy dogs. J Vet Intern Med 1999; 13: 472–477.
- 8↑
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.
- 9
Ko JC, Miyabiyashi T, Mandsager RE, et al. Renal effects of carprofen administered to healthy dogs anesthetized with propofol and isoflurane. J Am Vet Med Assoc 2000; 217: 346–349.
- 10↑
Lobetti RG, Joubert KE. Effect of administration of nonsteroidal anti-inflammatory drugs before surgery on renal function in clinically normal dogs. Am J Vet Res 2000; 61: 1501–1507.
- 11↑
Boström IM, Nyman GC, Lord PE, et al. Effects of carprofen on renal function and results of serum biochemical and hematologic analyses in anesthetized dogs that had low blood pressure during anesthesia. Am J Vet Res 2002; 63: 712–721.
- 12↑
Crandell DE, Mathews KA, Dyson DH. Effect of meloxicam and carprofen on renal function when administered to healthy dogs prior to anesthesia and painful stimulation. Am J Vet Res 2004; 65: 1384–1390.
- 13↑
Forsyth SF, Guilford WG, Pfeiffer DU. Effect of NSAID administration on creatinine clearance in healthy dogs undergoing anaesthesia and surgery. J Small Anim Pract 2000; 41: 547–550.
- 14↑
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: 1895–1901.
- 15↑
Mellor PJ, Roulois AJ, Day MJ, et al. Neutrophilic dermatitis and immune-mediated haematological disorders in a dog: suspected adverse reaction to carprofen. J Small Anim Pract 2005; 46: 237–242.
- 16
Weichselbaum TE. An accurate and rapid method for the determination of proteins in small amounts of blood serum and plasma. Am J Clin Pathol 1946; 16: 40–49.
- 17
Doumas BT, Watson WA, Biggs H. Albumin standards and the measurement of serum albumin with bromcresol green. Clin Chim Acta 1971; 31: 87–96.
- 18
Fabiny DL, Ertigshausen G. Automated reaction rate method for determination of serum creatinine with Centrifichem. Clin Chem 1971; 17: 696–700.
- 19
Watanabe N, Kamei S, Ohkubo A, et al. Urinary protein as measured with a pyrogallol red-molybdate complex, manually and in a Hitachi 726 automated analyzer. Clin Chem 1986; 32: 1551–1554.
- 20
Gutmann I, Bergmeyer HU. Determination of urea with glutamate dehydrogenase as indicator enzyme. In: Bergmeyer HU, ed. Methods of enzymatic analysis. Vol 4. 2nd ed. New York: Academic Press Inc, 1974; 1794–1798.
- 21
European Committee for Clinical Laboratory Standards ECCLS Document 3–4, 1988; 34–69.
- 22
Scandinavian Society for Clinical Chemistry and Clinical Physiology. Recommended methods for the determination of four enzymes in blood. Scand J Clin Lab Invest 1974; 33: 291–306.
- 23
IFCC 2002/7: IFCC primary reference procedures for the measurement of catalytic activity concentrations of enzymes at 37°C. International Federation of Clinical Chemistry and Laboratory Medicine. Part 6. Reference procedure for the measurement of catalytic concentrations of γ-glutamyltransferase. Clin Chem Lab Med 2002; 40: 734–738.
- 24↑
Heiene R, Moe L, Mølmen G. Calculation of urinary enzyme excretion, with renal structure and function in dogs with pyometra. Res Vet Sci 2001; 70: 129–137.
- 25↑
Stockham SL, Scott MA. Urinary system. In: Stockham SL, Scott MA, eds. Fundamentals of veterinary clinical pathologym. Ames, Iowa: Iowa State Press, 2002;277–335.
- 26
Clive DM, Stoff JS. Renal syndromes associated with nonsteroidal antiinflammatory drugs. N Engl J Med 1984; 310: 563–572.
- 27
Stillman MT, Napier J, Blackshear JL. Adverse effects of nonsteroidal anti-inflammatory drugs on the kidney. Med Clin North Am 1984; 68: 371–385.
- 28
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: 667–671.
- 29↑
Clemo FA. Urinary enzyme evaluation of nephrotoxicity in the dog. Toxicol Pathol 1998; 26: 29–32.
- 30↑
Cook AK, Gilson SD, Fischer WD, et al. Effect of diet on results obtained by use of two commercial test kits for detection of occult blood in feces of dogs. Am J Vet Res 1992; 53: 1749–1751.