Materials and Methods

Animals and uroliths—Medical records of cats for which uroliths were submitted to the Minnesota Urolith Center were reviewed.^a Cases consisted of cats with urate uroliths submitted for quantitative analysis between January 1, 1981, and December 31, 2008. Uroliths were analyzed by means of optical crystallography or infrared spectroscopy.⁴ For the study, all purine uroliths composed of uric acid or salts of uric acid (eg, ammonium urate, sodium urate, or calcium sodium urate) were collectively referred to as urates. Only cats with uroliths composed of at least 70% urate were included. With the exception of calculations of submission rates, data for cats with xanthine uroliths were excluded because of the apparently low frequency with which such uroliths were submitted and the probability that those uroliths were associated with an inherent dysfunction in purine metabolism.

Controls consisted of cats without urinary tract diseases admitted to veterinary teaching hospitals in the United States and Canada between January 1, 1981, and December 31, 2008. These cats were identified by searching the records of the Veterinary Medical Database.^b Cats evaluated at each veterinary teaching hospital were counted only once; data related to read-missions were excluded.

Statistical analysis—Distributions of breed, age, sex, reproductive status, and urolith location were calculated for the cases. Crude ORs and 95% CIs were calculated by use of the Woolf method⁵ to assess whether distributions of breed, age (< 1 year, 1 to < 2 years, 2 to < 4 years, 4 to < 7 years, 7 to < 10 years, 10 to 15 years, or > 15 years), sex (male vs female), and reproductive status (neutered vs sexually intact) differed between case and control cats. When any expected cell value in a contingency table was < 5, the Fisher exact test was performed.⁵ Domestic longhair, domestic medium-hair, domestic shorthair, and mixed-breed cats were grouped and identified as a mixed-breed group. For comparison purposes, data from cats that were mixed breed, < 1 year of age, female, and sexually intact were used as the referent group.

Estimates of ORs were considered significant when the 95% CI for a given OR did not include 1.0.⁶ Significant ORs with values between 1.1 and 1.9 or between 0.6 and 0.9 were considered weak associations.⁷ Likewise, significant ORs ≥ 2 (suggesting a risk factor) or ≤ 0.5 (suggesting a protective factor) were considered clinically (biologically) important, requiring additional research to determine whether the factors were actually causally related to urate urolith development.

The annual percentage of cats with urate uroliths was calculated by dividing the the number of cats with urate uroliths submitted during a given year by the total number of cats with uroliths submitted that year. Linear regression⁸ was performed to determine whether the yearly percentage of cats with urate uroliths increased or decreased over the study period. A value of P < 0.05 was considered significant.

Results

Uroliths—Uroliths from 106,192 cats were submitted to the Minnesota Urolith Center for analysis from 1981 through 2008. Of these, struvite (46,147/106,192 [43.5%]) and calcium oxalate (48,202/106,192 [45.4%]) were the most common. Urate uroliths were identified in 5,072 (4.8%) cats. Of those urate uroliths, 5,046 (99.5%) were composed of ammonium urate, 10 (0.2%) contained sodium urate, 10 (0.2%) contained uric acid, and 6 (0.1%) contained salts of uric acid. The remaining cats had uroliths composed of calcium phosphate (n = 364), cystine (97), xanthine (214), mixed⁹ minerals (1,037), silica (45) compound⁹ minerals (3,625), matrix (1,000), and other materials (eg, blood clots; 389). A total of 437,228 control cats were evaluated at hospitals contributing to the Veterinary Medical Database during the same period.

Crude ORs—Case cats with urate uroliths included 32 breeds (Table 1). Breeds with significantly higher odds of urate urolith submission than mixed breeds included Bengal, Birman, Egyptian Mau, European Short-hair, Havana Brown, Ocicat, Oriental, Ragdoll, Rex, Snowshoe, and Sphynx. Breeds with a significantly lower odds of developing urate uroliths than mixed breeds included Abyssinian, American Shorthair, Himalayan, Manx, and Persian cats.

The mean ± SD age of cats with urate uroliths was 6.2 ± 4.3 years. Cats in 6 age groups had significantly increased odds of urate urolith submission (Table 2). Cats ≥ 4 but < 7 years were 51 times as likely to develop urate uroliths, compared with cats < 1 year old, and constituted the highest-risk age group.

Of 4,943 cats with urate uroliths for which sex was reported, 54.5% were male and 45.5% were female. Male cats were 1.1 times as likely to develop urate uroliths, compared with females (Table 2). Of 4,943 cats with urate uroliths for which reproductive status was reported, 95.5% were neutered and 4.5 % were sexually intact. Neutered cats were 12 times as likely to develop urate uroliths, compared with sexually intact cats.

In 97.0% (4,920/5,072) of cats, urate uroliths had been retrieved from the lower urinary tract (bladder or urethra or voided), and in 0.7% (34/5,072), urate uroliths had been retrieved from the upper urinary tract (kidney or ureter). Four (0.1%) cats had urate uroliths in the lower and upper urinary tracts. For 2.2% (114/5,072) of cats, location of the uroliths was not specified.

Frequency of urate urolith submission—Between 1981 and 2008, no significant difference was evident in the yearly submission rate of feline urate uroliths (Figure 1). In 1986, there was an unexplained increase in urate urolith submissions. The mean submission rate for urate uroliths specifically was 5.4%/y.

Percentage of urate uroliths among all urolith submissions to the Minnesota Urolith Center between 1981 and 2008. The diagonal line represents a nonsignificant linear regression line.

Citation: Journal of the American Veterinary Medical Association 240, 7; 10.2460/javma.240.7.842

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Percentage of urate uroliths among all urolith submissions to the Minnesota Urolith Center between 1981 and 2008. The diagonal line represents a nonsignificant linear regression line.

Citation: Journal of the American Veterinary Medical Association 240, 7; 10.2460/javma.240.7.842

• Download Figure
• Download figure as PowerPoint slide

Percentage of urate uroliths among all urolith submissions to the Minnesota Urolith Center between 1981 and 2008. The diagonal line represents a nonsignificant linear regression line.

Citation: Journal of the American Veterinary Medical Association 240, 7; 10.2460/javma.240.7.842

• Download Figure
• Download figure as PowerPoint slide

Discussion

Purine nucleotide metabolism in cats results in several end products: hypoxanthine, allantoin, xanthine, and uric acid.¹⁰ Uroliths composed of hypoxanthine have not been recognized in cats. Allantoin, which is the most soluble of these products, is excreted in the urine of cats and is not lithogenic. Ammonium urate, which is the monobasic ammonium salt of uric acid, is the most common (approx 95%) naturally occurring purine urolith in cats.¹⁰ Xanthine uroliths were uncommon among specimens submitted to the Minnesota Urolith Center during the period evaluated in the present study, having been diagnosed in only 214 cats (4% of purine uroliths). Xanthine is typically converted to uric acid by the enzyme xanthine oxidase. None of the cats with xanthine uroliths had been treated with allopurinol, which is an inhibitor of xanthine oxidase. Because xanthine is the least soluble end product of the purines excreted in urine, xanthinuria is a consistent risk factor for xanthine urolith formation.¹⁰

Compared with what is known about urate urolith formation in humans and dogs, little is known about urate lithogenesis in cats. In general, risk factors for urate lithogenesis in cats include an increase in renal excretion and urine concentration of uric acid and potential presence of promoters or absence of inhibitors of urate urolith formation.¹¹ Descriptive and case-control studies¹¹ of calcium oxalate and struvite urolithiasis in cats have revealed consistent associations between alkaline urine pH and struvite urolith formation. However, aciduria is a risk factor for urate urolith formation because uric acid is a weak acid with a physiologic dissociation constant of approximately 5.5 for hydrogen ion.¹²

In cats predisposed to developing urate uroliths, dietary components may also contribute to urolith formation because dietary purines may be digested, absorbed, and incorporated into the body's endogenous purine pool.¹³ Eventually, the by-products are excreted in urine. Whereas allantoin is the principal end product of purine (DNA and RNA) catabolism, urea is the principal end product of amino acid catabolism. Cats have a requirement for dietary protein, which they metabolize for energy. Therefore, this species has a limited ability to upregulate or downregulate the production of catabolic enzymes of amino acid metabolism. These enzymes are nonadaptive, so the obligatory nitrogen loss is high even when cats are fed low-protein diets. For this reason, cats require a higher amount of dietary protein for maintenance than do dogs.¹³ It follows that an increase in dietary protein that contains abundant DNA and RNA (eg, liver tissue) is associated with an increase in the formation and excretion of purines, which may result in oversaturation of urine with lithogenic metabolites of purines.

Ammonium urate uroliths have been recovered from cats with portovascular anomalies.^14,15 Such anomalies can be found in male and female cats and typically are detected when cats are < 12 months of age.^14–18 Direct communication between the portal and systemic vasculature allows blood from the intestines to bypass the liver, resulting in hepatic atrophy and diminished hepatic function. Hepatic dysfunction, in turn, is associated with reduced hepatic conversion of uric acid to allantoin and reduced conversion of ammonia to urea.¹³ The predisposition of cats with portovascular anomalies to urate urolithiasis is probably associated with concomitant hyperammonemia, hyperuricemia, and hyperuricuria. Ammonia may combine with uric acid to form ammonium urate crystals and urate uroliths.^12,19

The biological behavior of ammonium urate uroliths in dogs with portovascular anomalies has been eludidated.^20–22 In contrast, the biological behavior of urate uroliths in cats with portovascular anomalies has only been reported in context of observations in other species.^10,23 It is apparent at this time that the pathophysiologic characteristics of most naturally occurring purine uroliths in cats are not related to congenital portovascular anomalies.^3,10,23 Urate uroliths in cats may develop as a consequence of acquired liver disease, but in our clinical experience, most cats with urocystoliths do not have concomitant liver disease. To date, the cause of urate urolithiasis in most cats is unknown.

Concentrations of lithogenic substances in urine are dependent on urine volume.^1,3 Because manufactured dry foods are associated with the production of a lower volume of urine, compared with manufactured canned foods,²⁴ consumption of dry foods is considered a risk factor for all types of uroliths. The present study was designed to evaluate this factor, but the responses received on the urolith submission form were too inconsistent to enable meaningful evaluation.

The solubility of most purines, particularly ammonium urate, is pH dependent.²⁵ In humans and presumably in cats, urine acidity is a risk factor for urate lithogenesis.¹² Therefore, consumption of diets that promote aciduria, such as diets with high quantities of animal-source protein or diets with other acidifying ingredients, is also a risk factor. Whether the type of feeding method (discrete meals vs ad libitum feeding) is important in urate urolith formation is not known. On one hand, mealtime feeding may result in an increase in postprandial urinary uric acid excretion. On the other hand, ad libitum feeding is often associated with a transient rise in urine uric acid concentration and a longer period of aciduria when compared with the physiologic status after discrete meal feeding.²⁴

The biological behavior of urate uroliths in cats does not appear to have been scientifically evaluated. Purine uroliths have the potential to undergo spontaneous dissolution, remain active (grow), or become inactive (remain unchanged). Although spontaneous dissolution of struvite uroliths has occasionally been observed, we have not yet observed this phenomenon in cats with urate uroliths. In a retrospective study,²⁶ we found that 13% of cats with urate uroliths had the first recurrence of stones with the same mineral (urate) type in a mean of 22 months. The second recurrence of urate uroliths was detected in 3% of these cats in a mean of 44 months after the initial removal of uroliths. Age and sex were not associated with the first or second episode of recurrence. Recurrence of urate uroliths may be influenced by several factors, including persistence of underlying causes, incomplete removal of uroliths from the urinary tract at the time of surgery (ie, pseudore-currence), and lack of compliance of clients or patients with therapeutic or prophylactic recommendations.¹⁰

Overrepresentation of particular breeds of cats with urolithiasis reflects the inherent odds of developing a particular urolith type in any given breed and the breed prevalence in the study region. Bengal, Ocicat, and Snowshoe breeds were at risk for urate urolith submission in the present study. As reported in 2000,²⁷ no association was found for these breeds with respect to struvite or calcium oxalate uroliths. In the present study, Ragdoll cats were at risk for urate uroliths (OR, 5.14), and such cats were at risk for struvite (OR, 5) and calcium oxalate (OR, 8) uroliths in a previous study.²⁷ Also, European Shorthair cats were at risk for urate uroliths (OR, 61), and they are reportedly²⁷ at risk for calcium oxalate (OR, 8) but not for struvite uroliths.

Oriental cats were at risk for urate uroliths (OR, 4) in our study, and they are reportedly²⁷ at risk for struvite (OR, 3) but not calcium oxalate uroliths. Rex cats were at risk for urate uroliths (OR, 2.67), but were found elsewhere²⁷ to be protected against struvite uroliths (OR, 0.1) and not affected by calcium oxalate uroliths. We found Birman cats were at risk for urate uroliths (OR, 6.77), but they appear to be protected from struvite (OR, 0.2) and calcium oxalate (OR, 0.3) uroliths.²⁷

Breeds that had lower odds for urate urolith submission than mixed breeds in the present study included Abyssinian, American Shorthair, Manx, Himalayan, and Persian. The Abyssinian breed appeared protected against struvite (OR, 0.2) and calcium oxalate uroliths (OR, 0.4) in another study.²⁷

Comparison of results of the present study with those of a similar study²² revealed differences in demographic factors associated with urate uroliths. Whereas the present investigation found that Siamese cats were not overrepresented or underrepresented, compared with the reference group, with respect to urate urolith submission, the other study found that Siamese cats were at greater odds for urate uroliths, compared with cats in the reference group, whereas Persian cats were protected from urate uroliths. These differences may be related, at least in part, to differences in the method the other researchers used to classify urolith types. In the other study,²² all cats with uroliths containing as little as 1% of a calculogenic molecule (eg, urate) were included in the urate-forming group, whereas in the present study, a urolith had to be comprised of least 70% of a specific molecule for the associated cat to be included in the urate-forming cat category.

Although male cats were more likely to have urate uroliths (OR, 1.14) than were females in the present study, this difference was not considered biologically important (OR, < 2). This finding is in agreement with those of other studies.^23,27 Neutered cats were also more likely to have urate uroliths (OR, 12.17) than were sexually intact cats, and they were also more likely to have struvite (OR, 3.5) and calcium oxalate (OR, 7) uroliths in another study.²⁷

Cats within all age groups were at risk for urate uroliths in the present study; however, other researchers²³ have reported no association between urate uroliths and age. In a 2000 study,²⁷ cats ≥ 7 but < 10 years of age were at highest odds for calcium oxalate uroliths, and cats ≥ 4 but < 7 years of age were at highest odds for struvite uroliths, compared with other cats.

Ninety-seven percent of the urate uroliths evaluated in the present study were obtained from lower urinary tract, and < 1% were from the upper urinary tract. However, the data collected were insufficient to suggest a reason for this difference.

The mean yearly percentage of urate uroliths among submissions to the Minnesota Urolith Center during the 28-year study period was approximately 5.4%. This percentage did not vary significantly with time. The mean yearly percentage in a smaller (n = 5,230) study²³ encompassing uroliths analyzed from 1985 to 2004 was 10% for urate uroliths, and other findings in that study were similar to ours. In contrast, a study²⁸ involving 11,353 uroliths retrieved from cats found a significant increase in urate urolith submissions from 1998 to 2008.

During the past 25 years, there has been a dramatic change in the relationship between the percentages of calcium oxalate uroliths and struvite uroliths submitted for analysis, with large increases in the volume of calcium oxalate uroliths and decreases in the volume of struvite uroliths. In 1981, calcium oxalate was detected in only 2% of feline uroliths submitted to the Minnesota Urolith Center, whereas struvite was detected in 78%. However, beginning in the mid-1980s, the percentage of calcium oxalate uroliths sharply increased, whereas that of struvite uroliths decreased. From 1994 to 2002, approximately 55% of the feline uroliths submitted to the Minnesota Urolith Center were composed of calcium oxalate, whereas only 33% were composed of struvite. During this period, the described pattern may have been associated with widespread use of diets designed to dissolve struvite uroliths, modification of maintenance and prevention diets to minimize struvite crystalluria (some dietary risk factors that decrease the odds of struvite uroliths increase the odds of calcium oxalate uroliths), and inconsistent follow-up evaluation of the efficacy of dietary management protocols by use of urinalysis and radiography.

In 2003, the percentage of calcium oxalate uroliths declined to 47.4%, whereas the percentage of struvite uroliths increased to 42.3%. During 2004, the percentage of struvite uroliths (44.9%) submitted to the Minnesota Urolith Center surpassed the percentage of calcium oxalate uroliths (44.3%). In 2005, struvite uroliths (48.1%) surpassed calcium oxalate uroliths (40.6%) in frequency of submission. Of 10,093 feline uroliths submitted to the Minnesota Urolith Center in 2006, 5,001 (49.5%) were struvite and 3,914 (38.8%) were calcium oxalate. In 2007, of 11,174 uroliths submitted to the Minnesota Urolith Center, 5,432 (48.6%) were struvite and 4,553 (40.7%) were calcium oxalate.

The progressive decrease in submission rates of naturally occurring calcium oxalate uroliths from 2003 to 2007 may have been associated with diet reformulations to minimize the odds of cats developing calcium oxalate crystalluria or uroliths. The increase in the percentage of struvite uroliths submitted during the same period may have been associated with these same diet reformulations. For example, diets that reduce urine acidity and provide adequate quantities of magnesium may reduce the odds of calcium oxalate urolith formation, but also increase the odds of struvite urolith formation. The sustained increase in the percentage of struvite uroliths from 2003 to 2007 may have been associated with a decrease in the use of diets designed to dissolve sterile struvite uroliths as a consequence of the considerable increase in calcium oxalate uroliths in the 1980s and 1990s. In contrast, the pattern of purine urolith submissions did not differ considerably from 1981 to 2008, suggesting diet reformulations did not have an effect on purine urolith development during the same time period.

As with all case-control studies, significant associations do not prove a cause-and-effect relationship. Additional prospective studies are needed to identify the mechanisms involved in urate urolith formation, the biological behavior of urate uroliths in cats, whether cats that form urate uroliths always have hyperuricuria, the similarities and differences between urate urolith–forming humans and urate urolith–forming cats, and the reason certain purebred, neutered, middle-aged cats are at high risk for urate urolith formation.