The precorneal tear film is a trilaminate structure composed of a superficial lipid layer secreted by the meibomian glands of the eyelids, an intermediate aqueous layer secreted by the orbital lacrimal gland and gland of the nictitating membrane, and a deep mucous layer secreted by conjunctival goblet cells.1 Keratoconjunctivitis sicca is a deficiency of the aqueous portion of the precorneal tear film that results in drying of the ocular surface and subsequent inflammation of the cornea and conjunctiva.1 Initially, clinical signs of KCS include blepharospasm caused by ocular pain, mucoid to mucopurulent ocular discharge, conjunctival hyperemia, and corneal ulceration.2 With chronicity, corneal epithelial hyperplasia, pigmentation, neovascularization, and secondary bacterial infection may occur.1,2
Keratoconjunctivitis sicca is a common eye disease of dogs, with reported annual incidences of 0.3% to 1.52% from 1964 to 1996.1,3,4 A search of the Veterinary Medical Database was performed for the period from 1997 to 2002, and the annual incidence rate of KCS in dogs from all reporting veterinary teaching hospitals ranged from 0.800% to 1.039%.a
The diagnosis of KCS is based on the presence of consistent clinical signs, positive staining of cornea and conjunctival epithelial cells with Rose Bengal stain, and measurement of decreased aqueous tear production by use of the STT or phenol-red-thread test.5–7 Typical STT values in dogs are (mean ± SD) 19.8 ± 5.3 mm to 21 ± 4.2 mm of strip wetting per minute,7 although test results may be affected by the type of commercially available tear test strip used.7–9
Numerous causes of KCS have been identified in dogs, including congenital alacrima10; immune-mediated destruction of the lacrimal glands11,12; metabolic diseases such as hypothyroidism, hyperadrenocorticism, and diabetes mellitus13–15; neurologic diseases affecting parasympathetic innervation to the lacrimal glands such as otitis media or interna, facial nerve paralysis, and dysautonomia16,17; and infectious agents such as canine distemper virus and Leishmania spp.18,19 Keratoconjunctivitis sicca has been reported following radiation therapy of the head and orbit20 and after surgical removal of the orbital lacrimal glands.21,22 Dry-eye conditions can occur as an adverse reaction following administration of a number of drugs. Transient lowering of tear production has been reported with topical and systemic administration of atropine, topical anesthetics, and general anesthesia.23,24 Keratoconjunctivitis sicca may occur in dogs in association with a variety of systemically administered drugs, including sulfonamides,25–31 5-aminosalicylic acid,32,33 and phenazopyridine hydrochloride.34–36 Nonsteroidal anti-inflammatory drugs have been reported to be associated with development of KCS in humans and dogs.32,33,37
Etodolac is an orally administered, nonsteroidal anti-inflammatory drug that is available as a generic medication and as a veterinary product.b The veterinary product was released on the market in 1998 and is labeled for use in the management of pain and inflammation associated with osteoarthritis in dogs at a dosage of 10 to 15 mg/kg (4.5 to 6.8 mg/lb), PO, once daily. This product is used widely in the United States, and FDAH estimates that approximately 59,000 dogs were administered the product daily during the time frame of the study reported here (1999 to 2002).c
An anecdotal report of KCS developing in dogs after oral administration of the product is present in the literature.38 In 2000, subsequent to the release of the product into the veterinary market, KCS was added to package inserts as an adverse event.39 A mean of 28 cases of KCS was reported to FDAH annually from 1998 to 2002. On the basis of this number of adverse events and the calculated number of dogs that received the product as estimated by FDAH, this equates to an estimated yearly incidence rate of 0.047%. The purpose of the retrospective case series reported here was to characterize the features of KCS in 2 groups of dogs that developed KCS while receiving etodolac orally.
Criteria for Selection of Cases
Data for this study were obtained from a survey of veterinary ophthalmologists in the United States conducted in 2002 (group A) and a review of all adverse reactions to the veterinary product reported to FDAH from January 1999 to August 2002 (group B). Cases were included if dogs developed KCS while receiving oral administration of edotolac.
Procedures
A detailed questionnaire was mailed to all reporting individuals, and telephone follow-up was performed where possible. Data collected and analyzed included breed, sex, age at the time of KCS diagnosis, weight, etodolac dosage administered, duration of drug administration, indication for etodolac administration, concurrent medications, concurrent diseases, laterality of eyes affected with KCS, STT values at the time of diagnosis and at last follow-up, whether etodolac was discontinued after the diagnosis of KCS, treatment for KCS, response to treatment, and length of follow-up time.
Schirmer tear test values were used to categorize severity of KCS as mild (≥10 mm/min), moderate (5 to 9 mm/min), and severe (< 5 mm/min; Table 1). Each eye of each dog was assigned a KCS score (0 to 3) on the basis of the reported STT value at the time of diagnosis of KCS and at the time of last follow-up. The KCS scores for the right and left eye were summed to provide pretreatment and posttreatment composite KCS scores. For group A, clinical improvement was defined as a composite posttreatment KCS score that was less than the pretreatment composite KCS score. Cure was defined as a dog having a composite score of 0 after treatment. For group B, follow-up STT values were not supplied by the reporting individuals in a large number of cases; therefore, the posttreatment KCS score was based on either the reported STT value at the time of last follow-up, when available, or on the subjective assessment of the reporting individual.
Severity of KCS as determined on the basis of reported STT values (mm of wetting/min) and corresponding assigned KCS scores in dogs that developed KCS while receiving oral administration of edotolac.
| Severity of KCS | Schirmer tear test mm/min) | KCS score | |
|---|---|---|---|
| Normal | • •15 | 0 | |
| Mild | 10–15 | 1 | |
| Moderate | 5–9 | 2 | |
| Severe | • •5 | 3 |
Statistical analysis—Forward stepwise logistic regression modelsd were developed to predict the probability of either cure or clinical improvement of KCS as a function of several independent variables. Data from groups A and B were analyzed separately because criteria for inclusion into the models were different for the 2 groups. Because of the large number of cases from group B in which objective data regarding response to treatment were not supplied by the reporting individuals, less stringent criteria for inclusion into the models for the independent variable outcome were used.
The forward stepwise logistic regression models evaluated the following independent variables: breed (German Shepherd Dog, Labrador Retriever, or other); age (< 10 years or ≥10 years); sex (male or female); dose (< 12 mg/kg [5.5 mg/lb] or ≥ 12 mg/kg); duration of treatment prior to onset of clinical signs (< 10 months or ≥10 months); whether etodolac administration was continued after the diagnosis of KCS; and the presence of concurrent hypothyroidism, otitis, or other diseases. These diseases are recognized causes of KCS in dogs and were individually evaluated in the logistic regression model to determine whether they were associated with outcome of the disease (clinical improvement or cure). Variables were permitted to enter the model at P < 0.1. At each step, the variable with smallest P value was selected for entry into the model. For final comparisons, values of P < 0.05 were considered significant.
Results
Clinical characteristics—Age of dogs at the time of diagnosis was reported in 65 of 65 (group A) and 139 of 146 (group B) dogs. Mean age was 10.1 years in both groups and ranged from 3 to 16 years (A) and 1 to 17 years (B). Ninety-four percent of dogs in both groups (62/65 [A]; 131/139 [B]) were 6 years of age or older at the time of diagnosis.
Sex of dogs was recorded in 65 of 65 (A) and 127 of 146 (B) dogs. Both sexes were equally represented in group A (31 female, 34 male), and in group B, females represented 62% of dogs (79 females, 48 males). Neutered dogs were predominant in both groups. Spayed females represented 97% (30/31 [A]) and 92% (73/79 [B]) of all females, and neutered males represented 88% (30/34 [A]) and 92% (44/48 [B]) of all males.
Breed of dog was reported in 65 of 65 (A) and 136 of 146 (B) cases. A variety of large- and small-breed dogs were represented; however, most were large-breed dogs. Breeds represented in the highest frequency in both groups were German Shepherd Dogs (n = 14 [A]; 32 [B]), Labrador Retrievers (9 [A]; 22 [B]), Golden Retrievers (6 [A]; 15 [B]), and mixed-breed dogs (3 [A]; 21 [B]). Additional breeds included Akita (n = 1 [B]), Australian Blue Heeler (1 [B]), Australian Shepherd (3 [A]; 3 [B]), Bichon Frise (2 [B]), Boxer (2 [B]), Belgian Malinois (1 [A]), Bernese Mountain Dog (2 [B]), Border Collie (1 [A]), Chesapeake Bay Retriever (1 [A]; 1 [B]), Clumber Spaniel (1 [A]), Collie (2 [A]; 2 [B]), Corgi (1 [A]; 1 [B]), Dachshund (2 [B]), Doberman Pinscher (1 [B]), English Bulldog (1 [B]), Flat-coated Retriever (1 [B]), Giant Schnauzer (1 [B]), Great Pyrenees (2 [A]; 1 [B]), Irish Setter (1 [B]), Keeshound (1 [B]), Kuvasz (1 [A]), Lhasa Apso (2 [B]), Maltese (1 [A]), Miniature Doberman Pinscher (2 [B]), Miniature Poodle (1 [A]; 1 [B]), Norwegian Elkhound (1 [A]; 1 [B]), Pomeranian (1 [B]), Portuguese Water Dog (1 [A]), Pug (1 [A], 1 [B]), Rat Terrier (1 [B]), Rottweiler (2 [A]; 3 [B]), Samoyed (3 [A]; 1 [B]), Schnauzer (1 [A]), Shar Pei (1 [A]; 1 [B]), Shetland Sheepdog (2 [B]), Shih Tzu (1 [B]), Standard Poodle (1 [A]; 2 [B]), Yorkshire Terrier (1 [B]), Welsh Terrier (1 [B]), and breed not reported (10 [B]).
Body weight of dogs was reported in 61 of 65 (A) and 117 of 146 (B) dogs. Mean body weight was 32.5 kg (77.4 lb [A]) and 30.0 kg (66 lb [B]), and range was 5.2 to 64.5 kg (11.4 to 141.9 lb [A]) and 3 to 52 kg (6.6 to 114.4 lb [B]).
Etodolac administration—The veterinary etodolac productb is labeled for administration in dogs at 10 to 15 mg/kg (4.5 to 6.8 mg/lb), PO, every 24 hours. In group A, mean dose administered was 11.7 mg/kg (5.3 mg/lb), PO, every 24 hours (range, 0.5 to 22 mg/kg [0.23 to 10.0 mg/lb]). Dose was not recorded in 1 dog. Fifty-nine percent (38/64) of dogs received the labeled dose. The reported dose was less than the labeled dose in 31% (20/64) of dogs, whereas 9% (6/64) received more than the labeled dose. In group B, mean dose administered was 12.0 mg/kg (5.4 mg/lb), PO, every 24 hours (range, 3.4 to 50 mg/kg [1.5 to 22.7 mg/lb]). Dose was not recorded in 26 of 146 dogs. Forty-nine percent (72/146) of dogs received the labeled dose. The reported dose was less than the labeled dose in 23% (33/146) of dogs, whereas 9% (15/146) received more than the labeled dose.
Etodolac is indicated for the treatment of osteoarthritis in dogs. Information regarding the indication for etodolac administration in individual dogs was available in 50 of 65 (A) and 133 of 146 (B) dogs. In all dogs, administration was for treatment of musculoskeletal-related disease, including osteoarthritis, hip dysplasia, cruciate ligament rupture, spondylosis deformans, and diskospondylitis.
Mean duration of etodolac administration prior to diagnosis of KCS was 9.1 months (A) and 7.8 months (B [Figure 1]). Keratoconjunctivitis sicca was diagnosed in dogs after 0.5 to 29 months (A) and 6 days to 68 months (B) of etodolac administration. Duration of drug administration was not reported in 6 of 65 (A) and 14 of 146 (B) dogs. Sixty-six dogs in group B were given etodolac for < 6 months, and 22 of these dogs had records sufficiently complete to allow entry into the logistic regression models; duration of etodolac administration was the only independent variable significantly associated with the outcome of KCS. Affected dogs that were administered etodolac for < 6 months were 4.6 times as likely (P = 0.049) to have resolution of clinical signs of KCS than dogs that were given etodolac for ≥ 6 months.
Duration of oral administration of edotolac in dogs that developed KCS as reported in a survey of veterinary ophthalmologists (group A) and by FDAH (group B). nr = Duration not reported.
Citation: Journal of the American Veterinary Medical Association 230, 4; 10.2460/javma.230.4.541
Concurrent diseases and medications—Concurrent diseases were reported in 23 of 65 (A) and 35 of 146 (B) dogs, including hypothyroidism (n = 9 [A]; 7 [B]), seizures (5 [A]; 4 [B]), urinary tract disease (2 [A]; 6 [B]), and otitis (3 [A]; 2 [B]). Other diseases reported in group A dogs included dermatologic disease (n = 2), cardiac disease (2), colitis (1), and a tail base tumor (1). Additional concurrent diseases reported in group B included unspecified dermatologic disease (n = 6), gastrointestinal tract disease (3), liver disease (3), systemic immune-mediated disease (1), hyperlipidemia (1), hypoalbuminemia (1), and pancreatitis (1). Concurrent ocular diseases were reported in 5 dogs in group B, and these included pannus (n = 2), epibulbar melanoma (1), conjunctivitis (1), and Horner syndrome (1). Hypothyroidism, otitis, and immune-mediated diseases are recognized causes of KCS3,11-13,17; however, none of these variables were significantly associated with clinical improvement or resolution of KCS.
Concurrent administration of additional medications was reported in 22 of 65 dogs (A) and 42 of 146 dogs (B). The most commonly reported drugs included levothyroxine (n = 6 [A]; 8 [B]), phenobarbital (5 [A]; 5 [B]), potassium bromide (2 [A]; 1 [B]), hydroxyzine (3 [B]), and enrofloxacin (1 [A]; 5 [B]). A number of animals in both groups received glucosamine-containing nutraceuticals commonly administered for the treatment of osteoarthritis (n = 9 [A]; 14 [B]). Other nonsteroidal anti-inflammatory drugs were reported in a small number of dogs, including carprofen (n = 1 [A]; 1 [B]) and aspirin (n = 1 [A]; 1 [B]). Diethylstilbestrol, methylprednisolone acetate, orbifloxacin, prednisone, stanozolol, and a combination corticosteroid-antibacterial-antifungal ointmente were administered to a single dog each in group A. Amoxicillin, buspirone, cephalexin, chlorpheniramine, diethylstilbesterol, doxycycline, enalapril, famotidine, gentamicin, metronidazole, phenylpropanolamine, prednisolone, trimethoprimsulfadiazine, ursodiol, and a combination fluocinolone acetonide–dimethyl sulfoxidef were administered to a single dog each in group B. Only 1 dog in group B received trimethoprim-sulfadiazine, a sulfonamide that has been associated with a 4.2% to 15.2% incidence of KCS in dogs.29,40 This dog had severe KCS at the time of diagnosis (STT, 0 mm/min OU), and the response to treatment was unknown because the dog was lost to follow-up.
Characterization of KCS—Laterality of eyes affected was recorded in 65 of 65 (A) and 100 of 146 (B) dogs. Keratoconjunctivitis sicca occurred as a bilateral condition in the majority of cases (62/65 [95.4%; A]; 96/100 [96.0%; B]). Unilateral KCS occurred in < 5% of dogs in both groups (3/65 [4.6%; A]; 4/100 [4.0%; B]) and affected the right eye (n = 3 [A]; 1 [B]) or left eye (1 [B]) or was not specified (2 [B]).
Categorization of KCS (mild, moderate, or severe) on the basis of STT values at the time of diagnosis was available for 130 eyes of 65 dogs (A) and for 150 eyes of 146 dogs (B). Mild KCS was present in 11 eyes of 8 dogs (A) and 11 eyes of 8 dogs (B). Moderate KCS was present in 32 eyes of 22 dogs (A) and in 26 eyes of 19 dogs (B). Severe KCS was present in 84 eyes of 50 dogs (A) and in 111 eyes of 62 dogs (B [Figure 2]). Mean STT values of all dogs were 3.1 mm/min OD and 3.2 mm/min OS for group A and 2.6 mm/min OD and 2.7 mm/min OS for group B. The combined mean STT value of both eyes of bilaterally affected dogs was 3.1 mm/ min (A) and 2.7 mm/min (B). Often, eyes were symmetrically affected with the severity of KCS. In group A, 45 of 62 (73%) dogs with bilateral disease were affected with KCS of equal severity in both eyes. In group B, 58 dogs with bilateral KCS had disease of equal severity between the 2 eyes. Severity of KCS was asymmetric between each eye in 17 (A) and 15 dogs (B).
Severity of KCS as determined on the basis of STT values at the time of diagnosis in dogs in groups A and B. WRR = Within reference range. OD = Right eye. OS = Left eye.
Citation: Journal of the American Veterinary Medical Association 230, 4; 10.2460/javma.230.4.541
Treatment—Specific treatment for KCS was reported in 48 of 65 (A) and 116 of 146 (B) dogs. Topically administered cyclosporine, the current standard of KCS treatment, was used in 46 of 48 (A) and 100 of 116 (B) dogs. In the 2 dogs in group A for which cyclosporine was not administered, a parotid duct transposition was reported as the sole treatment for absolute KCS (STT, 0 mm/min). Of the 16 dogs not treated with cyclosporine in group B, KCS resolved in 5 dogs within 1, 2, 4, and 36 months of follow-up and the time to resolution was not reported in 1 dog. Keratoconjunctivitis sicca improved in 2 dogs within 0.5 and 3 months of diagnosis by use of only supplementation of artificial tears. Two dogs were treated with parotid duct transposition for absolute KCS, and the remaining 7 dogs were lost to follow-up.
A variety of additional topically and orally administered medications were reported, including artificial tears (n = 19 [A], 44 [B] dogs), topically administered antimicrobials (14 [A]; 44 [B]), topically administered corticosteroids (8 [A]; 31 [B]), and orally administered pilocarpine (3 [A]; 1 [B]). Ten eyes of 6 dogs (A) and 8 eyes of 4 dogs (B) with nonresponsive, absolute KCS underwent parotid duct transposition. In 2 dogs in group A, tacrolimus was used topically following a lack of response to cyclosporine. Subsequent to the use of tacrolimus in these 2 dogs, KCS improved from STT values of 0 to 10 mm/min (OU) after 2 months in 1 dog, and KCS was resolved at the 1-year follow-up visit in the other dog.
Outcome—Follow-up data were available in 65 of 65 (A) and 89 of 146 (B) dogs. Mean period of followup was 6.9 months with a range of 0.5 to 24 months (A) and 8.1 months with a range of 0.25 to 41 months (B). Follow-up STT measurements were reported in 36 of 65 (A) and 29 of 146 (B) dogs. Mean reported STT at last follow-up was improved to 7.9 mm/min (OD), 9.2 mm/min (OS), and 8.5 mm/min OU in group A and to 8.1 mm/min OD, 8.5 mm/min (OS), and 8.3 mm/min (OU) in group B.
Complete resolution of KCS with or without continued medical treatment occurred in 7 of 65 dogs (A) and 23 of 89 (B) dogs (Table 2). Mean time to resolution of clinical signs following diagnosis of KCS was 6.4 months with a range of 1 to 12 months (A) and 2.6 months with a range of 0.25 to 9 months (B). Duration of etodolac administration prior to KCS diagnosis was a mean of 6.9 months with a range of 1.5 to 18 months (A) and 5.4 months with a range of 1 to 14 months (B). Time from diagnosis to resolution of KCS was not specifically reported in 4 of 23 dogs in group A; however, KCS reportedly resolved upon cessation of etodolac administration in 3 dogs. The duration of etodolac administration in these 3 dogs was 10 months in 1 dog and not reported in 2 dogs.
Outcome of treatment in dogs (No. [%] of dogs) with KCS as reported in a survey of veterinary ophthalmologists (group A) and by FDAH (group B).
| Outcome | Group A | Group B | Total |
|---|---|---|---|
| KCS resolved | 7(10.8) | 23(15.8) | 30 |
| KCS improved | 30(46.2) | 39(26.7) | 69 |
| No response | 28(43.1) | 27(18.5) | 55 |
| Not reported | 0(0) | 57(39.0) | 57 |
| Total | 65 | 146 | 211 |
Improvement of KCS occurred in dogs in which the posttreatment composite KCS score was less than the pretreatment composite KCS score but not equal to 0 (A) or in dogs that were reported to have improved clinically with treatment (B). Improvement of KCS occurred in 30 of 65 (A) and 39 of 89 (B) dogs (Table 2). Duration of administration prior to KCS diagnosis was a mean of 8.6 months with a range of 0.5 to 29 months (A) and 7.8 months with a range of 0.5 to 68 months (B).
The dogs considered nonresponsive to treatment had posttreatment composite KCS scores equal to or lower than pretreatment composite KCS scores (A) or were reported to have no improvement or worsening of KCS during the follow-up period (B). No response to treatment was observed in 26 of 65 (A) and 27 of 89 (B) dogs (Table 2). Duration of etodolac administration prior to KCS diagnosis in these dogs was a mean of 10.7 months with a range of 3 to 24 months (A) and 9.7 months with a range of 0.5 to 34 months (B).
Results of logistic regression model analysis—In group A, 51 of 65 dogs had records provided by veterinary ophthalmologists that were sufficiently complete to permit use in models used to predict cure and clinical improvement. None of the independent variables considered were significantly associated with the probability of either a cure or clinical improvement.
In group B, 52 of 146 dogs had records provided by FDAH that were sufficiently complete to permit use in models used to predict cure and clinical improvement. None of the independent variables considered were significantly associated with the probability of clinical improvement. Of all variables considered, only duration of treatment prior to diagnosis of KCS was associated with the probability of cure. Dogs with treatment intervals < 6 months prior to the onset of KCS were 4.2 times as likely to have complete resolution of KCS (P = 0.049) as were dogs with treatment intervals ≥ 6 months.
Discussion
Etodolac is a pyranocarboxylic acid that undergoes extensive enterohepatic circulation and partial biotrans-formation to glucuronide conjugates in dogs.41 Adverse reactions listed on the package insert for the veterinary product and generic etodolac include a variety of effects associated with cyclooxygenase inhibitory nonsteroidal anti-inflammatory drugs, including gastrointestinal and renal toxicoses. In addition, acute hepatitis, icterus, hemolytic anemia, urticaria, pruritis, neurologic abnormalities, and KCS are also listed. Weight loss, anorexia, anemia, hypoproteinemia, and gastric and small intestinal ulceration were found to occur at daily dosages of 40 and 80 mg/kg (18.2 and 36.4 mg/lb) in a 1-year toxicologic study of Beagles.42 Idiosyncratic reactions reported in association with etodolac administration in humans include fulminant hepatic failure,43 agranulocytosis,44 and immune-mediated hemolytic anemia.45
Sulfadiazine, sulfasalazine, trimethoprim-sulfonamide combinations, phenazopyridine, and 5-aminosalicylic acid have been reported to cause KCS in dogs.28,29,32-36,40 Chemical structures of the sulfonamides and phenazopyridine each possess a common nitrogen-containing pyrimidine or pyridine ring. The exact pathogenesis by which sulfonamides and phenazopyridine induce KCS in dogs is unknown but has been proposed to be a direct toxic effect on lacrimal acinar cells by the drugs' nitrogen-containing pyrimidine or pyridine ring.27,46 Following administration of phenazopyridine to dogs, gross and histologic changes consisting of progressive intracytoplasmic accumulation of a brown (nonmelanin) pigment are observed within lacrimal acinar cells and a mononuclear and neutrophilic inflammatory infiltrate is evident in the interstitium of the lacrimal glands.35,36 A lymphocytic inflammatory infiltrate has been detected in the lacrimal glands of dogs with sulfonamide-related KCS.12
Although the chemical structure of etodolac includes a nitrogen-containing indole ring, the indolering is distinctly different from the pyrimidine and pyridine rings of phenazopyridine and the sulfonamides. The mechanism by which etodolac causes KCS is unknown at this time, but possible mechanisms include an immune-mediated hypersensitivity or an idiosyncratic lacrimotoxic effect of etodolac or a metabolite. Histologic evaluation of lacrimal glands in dogs with etodolac-associated KCS will be a necessary first step towards elucidation of the pathogenesis.
Immune-mediated destruction of the lacrimal glands is considered the most common cause of KCS in dogs.15 Breeds commonly considered predisposed to KCS include West Highland White Terrier, Cavalier King Charles Spaniel, English Bulldog, Lhasa Apso, Shih Tzu, Cocker Spaniel, Miniature Schnauzer, Boston Terrier, Pekingese, Pug, and Yorkshire Terrier.2–4 Increasing age, female sex, and a neutered status of either sex are considered predisposing factors to KCS in dogs.2,15,47,48
Arthritis is common in older, heavier, large-breed dogs, and the 2 groups of dogs in this study reflected the expected population of dogs administered etodolac for treatment of osteoarthritis.49,50 Most dogs in both groups were Labrador Retrievers, German Shepherd Dogs, Golden Retrievers, and mixed-breed dogs. Mean body weight was 30 to 32.5 kg in both groups, as would be expected of a population of dogs receiving treatment for osteoarthritis. Many of the breeds typically predisposed to KCS are smaller,2–4 but these breeds were found in low numbers in the present study. Lower body weight (< 12 kg) was considered an important predictor of development of KCS in trimethoprim sulfadiazine− treated dogs in 1 study,40 and this was attributed to a possible overdose when dogs were medicated on a milligram-per-kilogram basis. Small-breed dogs may have been underrepresented in the present study because there were likely lower numbers of small dogs in the general population receiving etodolac for treatment of osteoarthritis49,50; however, exact data regarding breed and body weight of dogs receiving etodolac during the time frame of the present study were unavailable for comparison. Because of the low numbers of small-breed dogs in this study and an absence of a reference population, additional conclusions regarding the importance of breed and body weight could not be drawn.
Higher risk for the development of KCS has been identified in neutered dogs of either sex, compared with sexually intact dogs.15,47 Although females have been reported to have an increased prevalence, this may be attributable to a historically higher prevalence of neutering in the female population.2,48 In the present study, neutered dogs represented 92% of dogs in both groups. Total numbers of females and males were roughly equivalent in group A, and only slightly more females were present in group B. In the absence of data for the general population from which these dogs were drawn, statements regarding the overall relative percentages of sexes could not be made conclusively.
Drug-induced decreases in tear production have been reported to occur after a variable period of drug administration. Development of KCS may occur within 1 to 2 weeks of initiating daily treatment with phenazopyridine, and spontaneous recovery of tear production following cessation of this drug administration has been reported.34,35 Sulfadiazine has been reported to cause KCS after as little as 3 days of use,27 and sulfasalazine has been associated with KCS after months to years of use.28 Trimethoprim-sulfadiazine caused decreases in STT within 7 days to 7 months in 1 study.40 In the present study, KCS was reported after as little as 6 days of etodolac administration in 2 dogs and as great as 68 months in 1 dog.
In 2 retrospective studies,27,28 3 of 14 and 4 of 14 dogs had improvement in STT values following discontinuation of sulfonamides and improvement in KCS was often associated with shorter duration of sulfonamide administration. In 1 study,29 only 2 of 16 dogs had resolution of KCS following cessation of trimethoprim-sulfamethoxazole administration, and both dogs received the drug for < 30 days. Following discontinuation of etodolac, 10% (A) to 25% (B) of dogs did experience complete resolution of KCS in the present study. Within group B, dogs that were treated with etodolac for < 6 months were 4.2 times as likely to have complete resolution of KCS as those dogs treated for 6 months or longer.
Results presented here did not confirm a causal relationship between etodolac administration and development of KCS. Limitations of this retrospective study included the lack of a control population of dogs administered etodolac that did not develop KCS. A prospective, randomized study with a control population would be necessary to specifically identify risk factors in the development of etodolac-associated KCS. Additionally, the quality of the data collected from the 2 sources was substantially different. Data from veterinary ophthalmologists were typically more detailed and complete with follow-up information. However, within this group, there was no standardization of tear test strip used or the treatment protocol used. There are a variety of tear test strips available for the STT, and variability of wetting has been documented.8
Data provided by FDAH were subject to greater variability and potential error. The cases and their details were reported to FDAH by owners and veterinarians. In those cases reported by owners, the reported details may have been biased if compensation was sought for diagnostic testing or treatment. Follow-up information was limited, and in most instances, follow-up STT values were not available. Despite allowing data in which only subjective reports of response to treatment were available to be included into forward stepwise logistic regression models, only 36% (52/146) of cases could be statistically analyzed.
Future studies will be necessary to identify risk factors in the development of etodolac-associated KCS in dogs, including prospective, case-controlled studies to evaluate effects of dose, duration, breed, body weight, sex, age, or other factors. To determine the mechanism by which KCS develops in dogs administered etodolac, toxicologic studies, immunologic evaluation, and histologic evaluation of lacrimal glands of affected dogs will be necessary.
Although no conclusions can be drawn regarding a causal relationship between etodolac and KCS, veterinarians should be aware of the potential for KCS associated with the use of etodolac, and performance of the STT should be considered prior to and periodically throughout any period of etodolac administration to allow more definitive evaluation of possible effects on aqueous tear production.
ABBREVIATIONS
| KCS | Keratoconjunctivitis sicca |
| STT | Schirmer tear test |
| FDAH | Fort Dodge Animal Health |
| OD | Right eye |
| OS | Left eye |
| OU | Both eyes |
Veterinary Medical Database, Purdue University, West Lafayette, Ind.
Etodolac oral tablets, EtoGesic, Fort Dodge Animal Health, Fort Dodge, Iowa.
Martin SL, Fort Dodge Animal Health, Shawnee Mission, Kan: Personal communication, 2002.
Sigma Stat, version 2.0, Sigma Stat Statistical Software, Systat Software Inc, Point Richmond, Calif.
Otomax, Schering-Plough Animal Health, Union, NJ.
Synotic, Fort Dodge Animal Health, Fort Dodge, Iowa.
References
- 1↑
Moore CP. Diseases and surgery of the lacrimal secretory system. In:Gelatt KN, ed.Veterinary ophthalmology. 3rd ed.Philadelphia: Lea & Febiger, 1999;583–607.
- 2↑
Sansom J, Barnett KC. Keratoconjunctivitis sicca in the dog: a review of two hundred cases. J Small Anim Pract 1985;26:121–131.
- 3
Kaswan RL, Salisbury MA. A new perspective on canine keratoconjunctivitis sicca. Treatment with ophthalmic cyclosporine. Vet Clin North Am Small Anim Pract 1990;20:583–613.
- 4
Helper LC. The tear film in the dog. Anim Eye Res 1996;15:5–11.
- 5
Brown MH, Galland JC, Davidson HJ, et al. The phenol red thread tear test in dogs. Vet Comp Ophthalmol 1996;6:274–277.
- 6
Gelatt KN. Vital staining of the canine cornea and conjunctiva with rose Bengal. J Am Anim Hosp Assoc 1972;8:17–22.
- 7↑
Gelatt KN, Peiffer RL Jr, Erickson JL, et al. Evaluation of tear function in the dog using a modification of the Schirmer tear test. J Am Vet Med Assoc 1975;166:368–370.
- 8↑
Hawkins EC, Murphy CJ. Inconsistencies in the absorptive capacities of Schirmer tear test strips. J Am Vet Med Assoc 1986;188:511–513.
- 9
Wyman M, Gilger BC, Mueller P, et al. Clinical evaluation of a new Schirmer tear test in the dog. Vet Comp Ophthalmol 1995;5:211–214.
- 10↑
Aguirre GD, Rubin LF, Harvey CE. Keratoconjunctivitis sicca in dogs. J Am Vet Med Assoc 1971;158:1566–1579.
- 11
Kaswan RL, Martin CL, Dawe DL. Keratoconjunctivitis sicca: immunological evaluation of 62 canine cases. Am J Vet Res 1985;46:376–383.
- 12↑
Kaswan RL, Martin CL, Chapman WL Jr. Keratoconjunctivitis sicca: histopathologic study of nictitating membrane and lacrimal glands from 28 dogs. Am J Vet Res 1984;45:112–118.
- 13
Peruccio C. Incidence of hypothyroidism in dogs affected by keratoconjunctivitis sicca. Trans Am Coll Vet Ophthalmol 1982:47–48.
- 14
Barrera R, Cinta Mane M, Rodriguez JF, et al. Keratoconjunctivitis sicca and diabetes mellitus in a dog. J Am Vet Med Assoc 1992;200:1967–1968.
- 15↑
Kaswan R, Pappas C Jr, Wall K, et al. Survey of canine tear deficiency in veterinary practice. Adv Exp Med Biol 1998;438:931–939.
- 16
Wise LA, Lappin MR. A syndrome resembling feline dysautonomia (Key-Gaskell syndrome) in a dog. J Am Vet Med Assoc 1991;198:2103–2106.
- 17
Kern TJ, Erb HN. Facial neuropathy in dogs and cats: 95 cases (1975–1985). J Am Vet Med Assoc 1987;191:1604–1609.
- 18
Martin CL. Distemper-associated keratoconjunctivitis sicca. J Am Hosp Assoc 1985;21:355–359.
- 19
Pena MT, Roura X, Davidson MG. Ocular and periocular manifestations of leishmaniasis in dogs: 105 cases. Vet Ophthalmol 2000;3:35–41.
- 20↑
Roberts SM, Lavach JD, Severin GA, et al. Ophthalmic complications following megavoltage irradiation of the nasal and paranasal cavities in dogs. J Am Vet Med Assoc 1987;190:43–47.
- 21
Helper LC. The effect of lacrimal gland removal on the conjunctiva and cornea of the dog. J Am Vet Med Assoc 1970;157:72–75.
- 22
Helper LC, Magrane WG, Koehm J, et al. Surgical induction of keratoconjunctivitis sicca in the dog. J Am Vet Med Assoc 1974;165:172–174.
- 23
Herring IP, Pickett JP, Champagne ES, et al. Evaluation of aqueous tear production in dogs following general anesthesia. J Am Anim Hosp Assoc 2000;36:427–430.
- 24
Hollingsworth SR, Canton DD, Buyukmihci NC, et al. Effect of topically administered atropine on tear production in dogs. J Am Vet Med Assoc 1992;200:1481–1484.
- 25
Sansom J, Barnett KC, Long RD. Keratoconjunctivitis sicca in the dog associated with the administration of salicylazosulphapyridine (sulphasalazine). Vet Record 1985;116:391–393.
- 26
Aguirre GD. Keratoconjunctivitis sicca caused by sulfadiazine. J Am Vet Med Assoc 1973;162: 8.
- 27↑
Slatter DH, Blogg JR. Keratoconjunctivitis sicca in dogs associated with sulphonamide administration. Aust Vet J 1978;54:444–446.
- 28↑
Morgan RV & Bachrach A Jr. Keratoconjunctivitis sicca associated with sulfonamide therapy in dogs. J Am Vet Med Assoc 1982;180:432–434.
- 29↑
Diehl KJ, Roberts SM. Keratoconjunctivitis sicca in dogs associated with sulfonamide therapy: 16 cases (1980–1990). Prog Vet Comp Ophthalmol 1991;1:276–282.
- 30
Todenhoefer H. Nebenwirkungen von Sulfadiazin ber der Anwendung als Geriatrikum fuer Hunde. Dtsch Tieraerztl Wochenschr 1969;76:14–16.
- 31
Bilyard KG, Joseph EC, Metcalf R. Mesalazine: an overview of key preclinical studies. Scand J Gastroenterol Suppl 1990;172:52–55.
- 32
Barnett KC, Joseph EC. Keratoconjunctivitis sicca in the dog following 5-aminosalicylic acid administration. Hum Toxicol 1987;6:377–383.
- 33
Houston DM, Keller CB. Keratoconjunctivitis sicca in the dog following prolonged use of salazopyrin and other 5-ASA containing drugs for treatment of canine colitis. Can Vet J 1989;30: 437.
- 34
Bryan GM, Slatter DH. Keratoconjunctivitis sicca induced by phenazopyridine in dogs. Arch Ophthalmol 1973;90:310–311.
- 35
Slatter DH. Keratoconjunctivitis sicca in the dog produced by oral phenazopyridine hydrochloride. J Small Anim Pract 1973;14:749–771.
- 36
Slatter DH, Davis WC. Toxicity of phenazopyridine. Electron microscopical studies of canine lacrimal and nictitans glands. Arch Ophthalmol 1974;91:484–486.
- 37
Avisar R, Deutsch D, Savir H. Keratoconjunctivitis sicca induced by dipyrone. Harefuah 1988;114:543–545.
- 40↑
Berger SL, Scagliotti RH, Lund EM. A quantitative study of the effects of Tribrissen on canine tear production. J Am Anim Hosp Assoc 1995;31:236–241.
- 41↑
Cayen MN, Kraml M, Ferdinandi ES, et al. The metabolic disposition of etodolac in rats, dogs, and man. Drug Metab Rev 1981;12:339–362.
- 42↑
Wrenn JM, Inhelder JL, Hemm RD, et al. One year chronic toxicity study of etodolac, a nonsteroidal anti-inflammatory agent, in the Beagle dog. Oyo Yakuri 1990;40:599–646.
- 43↑
Mabee CL, Mabee SW, Baker PB, et al. Fulminant hepatic failure associated with etodolac use. Am J Gastroenterol 1995;90:659–661.
- 44↑
Cramer RL, Aboko-Cole VC, Gualtieri RJ. Agranulocytosis associated with etodolac. Ann Pharmacother 1994;28:458–460.
- 45↑
Cunha PD, Lord RS, Johnson ST, et al. Immune hemolytic anemia caused by sensitivity to a metabolite of etodolac, a nonsteroidal anti-inflammatory drug. Transfusion 2000;40:663–668.
- 46
Trepanier LA. Delayed hypersensitivity reactions to sulphonamides: syndromes, pathogenesis and management. Vet Dermatol 1999;10:241–248.
- 47
Kaswan RL, Salisbury MA, Lothrop CD Jr. Interaction of age and gender on occurrence of canine keratoconjunctivitis sicca. Prog Vet Comp Ophthalmol 1991;1:93–97.
- 48
Barnett KC. Keratoconjunctivitis sicca: sex incidence. J Small Anim Pract 1988;29:531–534.
- 50
Smith GK, Mayhew PD, Kapatkin AS, et al. Evaluation of risk factors for degenerative joint disease associated with hip dysplasia in German Shepherd Dogs, Golden Retrievers, Labrador Retrievers, and Rottweilers. J Am Vet Med Assoc 2001;219:1719–1724.
