Epidemiologic evaluation of calcium oxalate urolithiasis in dogs in the United States: 2010‐2015

Abstract Background Positive health implications of early recognition of calcium oxalate (CaOx) urolithiasis include increased opportunity for nonsurgical removal, early dietary modification to minimize urolith growth, early removal to avoid urinary obstruction, and early recognition of genetic and metabolic diseases before they contribute to additional morbidity. Objectives To identify high‐ and low‐risk dog breeds for CaOx uroliths and to determine the relationship of age and sex to the development of CaOx uroliths. Animals Calcium oxalate urolith submissions between 2010 and 2015. Methods A comparative cross‐sectional study was conducted to identify high‐ and low‐risk breeds for CaOx uroliths by comparing cases to multiple comparison groups. At‐risk breeds were identified if odds ratios were significant (P value <.05) across all comparison groups. Results Of 258 898 urolith submissions, 124 285 were CaOx. Calcium oxalate was identified in 212 breeds. Twelve breeds were identified as high‐risk breeds, and 14 breeds were identified as low‐risk breeds. All high‐risk breeds were small dog breeds, and all low‐risk breeds were medium to large dog breeds. Overall, the mean age ± standard deviation of the first CaOx urolith was 8.4 ± 2.8 years. Conclusions and Clinical Importance To achieve the health benefits of preclinical evaluation, breeds at high risk for CaOx urolithiasis should be screened at 5 to 6 years of age, which is 2 to 3 years before likely development of clinical urolithiasis.


| INTRODUCTION
Calcium oxalate (CaOx) is the most frequent urolith submitted for analysis from dogs in the United States. 1,2 The positive health implications of early recognition of CaOx urolithiasis include increased opportunity for nonsurgical removal, early dietary modification to minimize urolith growth, early removal to avoid urinary obstruction, and early recognition of potential genetic and metabolic diseases such as hypercalcemia before these diseases contribute to additional morbidity. To achieve these health benefits, CaOx urolithiasis will need to be recognized early, before development of clinical disease. One method of recognizing which dogs will benefit most from preclinical evaluation is an understanding of which dogs are at greatest risk for CaOx urolithiasis and when in a dog's lifetime risk is highest.
The optimal comparison group in urolithiasis research to determine dog breeds at risk for urolith formation has not been established. In 2 case-control studies, the comparison groups were hospital-based nonurinary tract disease dogs. 3,4 In another study, the comparison group was dogs with struvite uroliths. 5 Although the studies reported breeds from which CaOx uroliths were most frequently submitted for analysis, the majority of these studies did not account for breed popularity and hence the size of the population at risk. The objective of our comparative cross-sectional study was to identify high-and low-risk breeds for CaOx urolith occurrence and to determine the relationship of age and sex to the development of CaOx uroliths by comparing cases to multiple comparison control groups.

| Calcium oxalate case population
Urolith submissions to the Minnesota Urolith Center, University of Minnesota, between January 1, 2010, and December 31, 2015, were reviewed. Records were eligible for inclusion if dogs resided in the United States, and uroliths were ≥70% CaOx as determined by quantitative analysis. 6 Records of dogs with compound uroliths having a central core that was ≥70% CaOx also were eligible for inclusion.

| Comparison populations
Three canine populations were used for comparison: (1) non-CaOx urolith formers uroliths of which were analyzed by the mineral analysis laboratory during the same period as the CaOx cases, (2)

| Dog-level variables
From the urolith populations (CaOx cases and non-CaOx dogs), breed (as categorized by the American Kennel Club, www.akc.org/dogbreeds), sex (male or female) and age at urolith removal were obtained from urolith submission records. From the hospital population, breed, sex (male or female), and age at first visit were extracted from the medical records. Only information regarding the breed was available from the survey population.

| Statistical analyses
Means and standard deviation (SD) were determined for age at first urolith occurrence or first hospital visit; differences in mean age between the CaOx group and the non-CaOx group, and between the CaOx group and the hospital group, were compared by Student's t test for 2 independent samples after ascertaining that age was normally distributed by Q-Q plot and skewness. Age groups were further categorized into quartiles based on the frequency distribution among CaOx cases: <6, 6-8, 8-10, and >10 years.
Logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) for breed using the CaOx urolith group for cases and the non-CaOx urolith group, hospital admission group, or breed survey as control groups. Odds ratios for age categories and sex used the non-CaOx urolith and hospital admissions data as the control group because age and sex were not available for the breed survey.

| Breed
Calcium oxalate uroliths were identified in 212 breeds, including dogs of mixed breed. Twelve dog breeds were classified as high-risk breeds for developing CaOx uroliths. All high-risk breeds were small breed dogs. However, the magnitude of the association differed depending on the comparison population (Table 1).
Fourteen dog breeds were considered as low-risk breeds for CaOx uroliths. Most low-risk breeds were medium and large breeds except for 2 small dog breeds, Beagle and French Bulldog (Table 1).

| Sex
Calcium oxalate uroliths were submitted more often from male dogs than from female dogs (73.1% versus 26.9%) and were more common in neutered (85.5%) than intact dogs (14.5%). Comparing CaOx urolith T A B L E 1 Odds ratios (OR) and 95% confidence intervals (CI) for dog breeds at high and low risk of forming calcium oxalate (CaOx) uroliths (n = 124 285) compared to three groups: non-CaOx uroliths (n = 134 613), hospital admissions (n = 35 658), and a breed survey (n = 12 003) dogs to the non-CaOx urolith dogs, males had 12 times greater odds of developing CaOx uroliths than did female dogs (

| Age
The mean age of dogs with CaOx uroliths was 8.6 ± 2.7 years, which was significantly older than non-CaOx urolith dogs (6.3 ± 2.9 years, P value <.001) and hospital population dogs (

| DISCUSSION
Identifying breeds at high-and low-risk for CaOx uroliths is important for clinicians to determine which breeds should be screened for disease. In our study, 12 breeds were identified to be at high-risk ( than in older dogs, which magnifies the effect for older CaOx dogs. 19,20 Using the hospital comparison group, the risk for CaOx was highest for dogs that were 8-10 years of age; for older dogs, the risk was lower. This observation has been reported in other studies that considered age as continuous 4 and categorical variables. 3 This finding may imply that the risk for CaOx uroliths decreases in dogs >10 years of age. This apparent decrease also may be due to unmeasured factors that confound the association such as diet, medications, and comorbidities.
Knowing the age when dogs develop their first CaOx urolith can be used to determine when to screen high-risk dogs for CaOx uroliths. In our study, the mean age of first-time urolith formers was 8.4 ± 2.8 years.
Results from a multihospital study determined that the mean age of firsttime CaOx urolith formers was 7.5 ± 3.0 years. 4 This difference may be attributable to the different definition of age between the 2 studies. In the multihospital study, age was identified as age at urolith detection. In our study, age at urolith removal was used. If we assume that all uroliths were submitted to the mineral analysis laboratory, the difference of 10 months likely accounted for dietary and medical treatment to manage uroliths before removal.
The positive health implications of early recognition of CaOx urolithiasis includes early intervention to minimize disease morbidity.
In addition, CaOx uroliths detected early often are small. Small uroliths are removed more easily by nonsurgical methods. In 1 study, the optimal urolith size for removal by voiding urohydropropulsion was <3 mm in small male dogs. 21 were approximately a year earlier than the mean age of all high-risk breeds (Table 3). It is logical to assume that these breeds also should be screened at an earlier age.
In 1 study, the median age of CaOx urolith recurrence in miniature Schnauzers was 1.8 years. 23 We interpret these findings to indicate that once enough risk factors are present, it takes approximately 1.8 years before patients develop clinical disease. Therefore, we recommend annual screening to encompass a shorter interval to detect dogs before clinical disease when uroliths are smaller.
Several imaging modalities are available for screening dogs for CaOx uroliths. We prefer survey radiography because of its ability to evaluate all portions of the urinary tract for this radiopaque urolith. 24 Although ultrasonography is more sensitive at detecting urocystoliths, ultrasonography is limited in its ability to detect urethroliths in the distal urethra. 25 Similarly, ultrasonography is inferior at assisting prediction of urolith composition, which is a function of the radiographic density of uroliths.

CONFLICT OF INTEREST DECLARATION
Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
Authors declare no IACUC or other approval was needed.

HUMAN ETHICS APPROVAL DECLARATION
Authors declare human ethics approval was not needed for this study.