Hyperthyroidism is not a risk factor for subclinical bacteriuria in cats: A prospective cohort study.

Abstract Background Subclinical bacteriuria is defined as a positive bacterial urine culture in the absence of clinical evidence of urinary tract infection (UTI). Studies have reported that hyperthyroid cats have UTIs (mostly subclinical) with prevalence rates of 12%‐22%. Consequently, clinicians consider hyperthyroidism a risk factor for development of subclinical bacteriuria, and many recommend urine culture when evaluating hyperthyroid cats. Objectives To compare the prevalence of subclinical bacteriuria (based on positive urine culture) in untreated hyperthyroid cats to that in euthyroid cats of similar age and sex. Animals Three hundred and ninety‐three hyperthyroid cats presented for radioiodine treatment and 131 euthyroid cats (≥7 years of age) presented for routine examination. Cats with signs of lower urinary tract disease were excluded. Methods Prospective cohort study. Both hyperthyroid and euthyroid cats had urine collected by cystocentesis for complete urinalysis and culture. Data pertaining to age, sex, body condition, and serum thyroxine and creatinine concentrations also were acquired. Logistic regression was performed to evaluate for potential risk factors for subclinical bacteriuria. Results Hyperthyroid cats showed a low prevalence of subclinical bacteriuria (4.3%), which did not differ from that found in euthyroid cats (4.6%). Of the signalment factors evaluated, only female sex was a significant risk factor (odds ratio [OR], 6.9; P = .002). Furthermore, positive urine cultures were more likely in specimens with dilute urine concentration (<1.035), pyuria, or microscopic bacteriuria. Conclusions and Clinical Importance Hyperthyroid cats are not at risk for subclinical bacteriuria. In the absence of lower urinary tract signs, no clinical benefit exists in routinely performing urine cultures when evaluating hyperthyroid cats.


| INTRODUCTION
Subclinical bacteriuria (also called occult bacteriuria or asymptomatic bacteriuria) is defined as a positive bacterial urine culture in the absence of clinical evidence of urinary tract infection (UTI). 1 Studies have recognized subclinical bacteriuria as a common clinical syndrome in humans, cats, and dogs. [1][2][3][4] Retrospective studies of hyperthyroid cats have reported a prevalence of UTI ranging from 12 to 22%, 5,6 and consisting mostly of subclinical bacteriuria. Consequently, clinicians consider hyperthyroidism a risk factor for development of UTI, 3,7 and many recommend urine culture as a part of the routine evaluation of hyperthyroid cats. 8,9 However, in our experience, the prevalence of subclinical bacteriuria in hyperthyroid cats appears to be much lower than previously reported. Furthermore, none of the reported studies 5,6,10 suggesting hyperthyroidism as a risk factor for subclinical bacteriuria examined the prevalence of subclinical bacteriuria in a similarly aged, control population of euthyroid cats. Because age is a risk factor for subclinical bacteriuria, 3,11,12 the effect of hyperthyroidism (95% of which occurs in cats >9 years old 13 ) on development of subclinical bacteriuria requires an appropriate control group to eliminate the confounding effect of age. Therefore, we sought to better determine the true prevalence of subclinical bacteriuria in cats with hyperthyroidism. To that end, we screened cats with hyperthyroidism for subclinical bacteriuria, and compared the prevalence of bacteriuria (based on positive urine culture) in these cats to a population of euthyroid cats of similar age and sex that presented for routine examination. Our secondary objectives were to investigate potential risk factors for subclinical bacteriuria (female sex, older age, presence of kidney disease) in our 2 prospectively collected populations of middle-aged to older cats, as well as to evaluate urine sediment findings (eg, hematuria, pyuria, bacteriuria) as predictors of a positive urine cultures.

| Hyperthyroid cats
All hyperthyroid cats referred to our clinic for treatment with radioiodine over the 22-month period from January 2018 to October 2019 were evaluated for inclusion in this prospective cohort study.
To be eligible for inclusion, hyperthyroid cats underwent a thorough evaluation that included review of past medical history, complete physical examination (including body weight, body, and muscle condition scoring), 14 routine laboratory testing (CBC, serum biochemical profile, and complete urinalysis), urine culture, determination of serum thyroid hormone concentrations (total T 4 and TSH), 15,16 and qualitative and quantitative thyroid scintigraphy. 17 In cats previously treated with methimazole, the drug was discontinued ≥7 days before evaluation. None of these hyperthyroid cats had been treated recently with antibiotics. We excluded cats with signs of lower urinary tract disease, and those in which cystocentesis was not possible, either because of small bladder size or the fractious nature of the cat.

| Clinically normal, euthyroid cats
These cats were recruited as controls at time of visit for routine evaluation. To be enrolled in the study, cats had to be ≥7 years of age and considered healthy by their owners. None of these cats had any signs of lower urinary tract disease or had been recently treated with antibiotics, and all were normal on physical examination. These cats also were evaluated by routine laboratory testing (CBC, serum biochemical profile, and complete urinalysis) and urine culture, as well as serum total T 4 and TSH concentrations to exclude hyperthyroidism. 15,16 Again, we excluded cats in which urine could not be collected by cystocentesis.
Ethics approval was obtained from the Institution's Animal Care and Use Committee (IACUC) before the study commenced. All owners provided informed consent.

| Collection and processing of urine samples
Cystocentesis was performed without sedation on cats in lateral recumbency using manual palpation of the bladder and alcohol skin preparation.
Urine samples for urinalysis and culture (≥3 mL) were placed in appropriate sterile, plastic collection tubes, stored at 4 C, and transported to the laboratory (IDEXX Reference Laboratories, Westbrook, Maine, Reference Laboratory location, Manhattan, New York), where they remained refrigerated until being processed within 12 hours of collection. All procedures for routine urinalyses and culture were performed by trained laboratory technicians.

| Routine complete urinalysis procedures
For complete urinalysis, samples were allowed to warm to room temperature and then analyzed within 1 hour. Urine specific gravity was measured using a refractometer (Leica Vet 360, Misco Prod-

| Data and statistical analyses
Data were assessed for normality using the D'Agostino-Pearson test and by visual inspection of graphical plots. 22  sex, body condition score (underweight, ideal weight, or overweight), and presence of azotemic kidney disease (defined as serum creatinine concentration >2.1 mg/dL and urine specific gravity <1.035). 15 To select variables that best explained the probability of a cat being culture-positive, we used a backward stepwise approach. Variables associated with a cat being culture-positive at an alpha level <.2 were entered into the model. The significance of each explanatory variable was tested using the Wald test. Biologically plausible, multiplicative 2-way interactions between the remaining variables were assessed for significance. Results of the final model are reported in terms of adjusted odds ratios (OR) with 95% confidence intervals (95% CI) for each explanatory variable.
A separate logistic regression similarly was performed for the complete urinalysis and sediment findings, again using urine culture results (positive versus sterile) as the dependent variable, and urine pH, urine specific gravity (<1.035 or ≥ 1.035), pyrexia, hematuria, and bacteriuria on urine sediment evaluation as the independent variables.

| Urine culture results
Hyperthyroid cats had a prevalence of subclinical bacteriuria of 4.3% (17/393), which did not differ from that found in euthyroid cats (4.6%; 6/131; Table 1). Likewise, hyperthyroid cats treated with methimazole had a similar prevalence of subclinical bacteriuria to those that had not received methimazole ( Table 2).
Ages of the cats with subclinical bacteriuria (hyperthyroid or euthyroid) did not differ from the ages of cats without subclinical bacteriuria (Table 3). More female cats (20 of the combined 23 hyperthyroid and euthyroid cats) had subclinical bacteriuria than did male cats (P = .001; Table 3).
In the euthyroid cats, similar serum T 4 concentrations were found in the cats that had positive or negative urine culture results. Hyperthyroid cats with subclinical bacteriuria had serum T 4 concentrations that were slightly higher but not significantly different from hyperthyroid cats with negative cultures (P = .86; Table 3).
None of the 6 euthyroid cats and only 2 (11.8%) of the 17 hyperthyroid cats with subclinical bacteriuria were azotemic. Accordingly, serum creatinine concentrations in the cats with subclinical bacteriuria (hyperthyroid or euthyroid) did not differ from those of the cats with negative urine cultures (Table 3).
Hyperthyroid cats, but not euthyroid cats with subclinical bacteriuria had lower urine specific gravity than did corresponding cats with negative cultures (P = .05 for hyperthyroid cats; P = .20 for euthyroid cats; Table 3). Because thyroid status was not associated with probability of subclinical bacteriuria, we grouped all hyperthyroid and euthyroid cats and found that the 23 cats with subclinical bacteriuria had lower urine specific gravity than did the 501 cats with negative cultures (P = .02). Of the 6 euthyroid cats and 17 hyperthyroid cats with subclinical bacteriuria, urine specific gravity was <1.035 in 4 (66.7%) and 12 (70.5%) cats, respectively.
The prevalence of pyuria, bacteriuria, and active sediment were higher in both hyperthyroid and euthyroid cats with positive urine culture results (Table 3). However, the likelihood of microscopic hematuria was similar for cats with positive and negative urine culture results (Table 3).

| Organisms
Of the 17 hyperthyroid cats with bacterial growth, the most common isolates were Escherichia coli and Enterococcus faecalis ( In the 6 euthyroid cats, the most common isolates were E. faecalis and S. felis (Table 4). In contrast to the hyperthyroid cats, E. coli was not isolated from any of the euthyroid cats.

| Predictors of subclinical bacteriuria
Hyperthyroidism did not increase the risk for subclinical bacteriuria (Table 5). Among all cats, risk of subclinical bacteriuria was not associated with age, age stage, body condition, or azotemic kidney disease (Table 5). However, female cats had a significantly increased risk of subclinical bacteriuria compared to male cats (Table 5).

T A B L E 3
Comparison of signalment, serum thyroid, and renal function, and urine sediment in hyperthyroid and euthyroid cats, subdivided into cats with negative and positive urine culture results

| DISCUSSION
The results of our prospective study indicate that hyperthyroid cats have a low prevalence (<5%) of subclinical bacteriuria, nearly identical to that found in a cohort of euthyroid cats of similar age and sex pres- The low prevalence of subclinical bacteriuria in our hyperthyroid cats is similar to that reported in human patients with hyperthyroidism, in which cystitis or asymptomatic bacteriuria is very rare. [28][29][30]  In fact, reviews of asymptomatic bacteriuria in humans do not mention thyroid disease as a possible risk factor. 2,31 Common issues with study design of the past reports 5,6,10 that promoted hyperthyroidism as a risk factor for subclinical bacteriuria in cats include their retrospective nature and lack of an appropriate control population of euthyroid cats. If age-and sex-matched controls had been included in these studies, it is possible that a similar prevalence of subclinical bacteriuria also might have been found in those euthyroid cats. In addition, approximately 30% of the hyperthyroid cats in those retrospective studies had signs of cystitis (not subclinical bacteriuria), leading to selection of these cats for culture. [5][6][7]10 In our study, we prospectively evaluated over 400 hyperthyroid cats that presented to our clinic and excluded only 3 cats  Table 2). In addition, no difference in prevalence existed between the euthyroid control cats and the hyperthyroid cats (treated or not treated with methimazole). Serum T 4 concentrations in the cats that had been treated with methimazole (ie, the drug was discontinued ≥7 days before evaluation) were higher than in those that had not been treated with methimazole. This finding was not unexpected because hyperthyroidism is progressive in cats, 32,33 and treating with methimazole over weeks to months allows the underlying thyroid disease to become more severe. Nevertheless, prior treatment with methimazole had no effect on development of subclinical bacteriuria in our study.
A wide range of prevalence rates for subclinical bacteriuria has been reported in cats, varying from low rates of <1.0%-3.3%, 12,34,35 to intermediate rates of 6.1, 7.9, and 10%-13%, 26,27,36 to a high rate of 28.8%. 11 The main difference in these studies was the age of the cats, which tended to be younger in the studies with lower prevalence rates 12,34,35 and older in the studies with prevalence rates >6%. 11,26,36 Despite the fact that our euthyroid cats were older (median age, 12 years; range, 7-20 years), these cats had a prevalence (<5%) of subclinical bacteriuria that was not nearly as high as reported in some studies. 11,36 Although the reason for these differences in prevalence is not clear, most of our euthyroid cats had no evidence for concurrent disease; all were considered clinically normal by their owners and only a few had biochemical evidence of any comorbidities (eg, mild renal disease). In contrast, some of the previous studies included many cats with concurrent disease, 11 which might have increased the prevalence of subclinical bacteriuria.
We found females to be at higher risk for developing subclinical bacteriuria than males (OR, 6.9; in the presence of comorbidities such as diabetes mellitus or chronic kidney disease. 10,43 In our study, none of the affected hyperthyroid cats were treated with antimicrobials, and our study protocol did not require follow-up for repeat urine culture. However, none of the cats have displayed any signs of lower urinary tract disease since the time of urine culture and radioiodine treatment (4-to 24.5-month followup period). Our findings further support the current guidelines 1 that clinicians refrain from treating subclinical bacteriuria in cats (regardless of thyroid status) with rare exceptions.
In conclusion, results of our prospective cohort study indicate that hyperthyroid cats are not at increased risk for developing subclinical bacteriuria. Routine urine culture in hyperthyroid cats incurs an unnecessary expense for owners and need not be included in the routine diagnostic evaluation of hyperthyroid cats unless clinical signs of lower urinary tract disease are present. One unanswered question is whether subclinical bacteriuria resolves or persists after treatment of hyperthyroidism in these cats. Further prospective studies designed to clarify the long-term outcome of treated hyperthyroid cats are warranted.