Abstract
Knowledge of the occurrence of bacteriuria in adult, healthy cats is scarce in the scientific literature. A study was designed to investigate the occurrence of bacteriuria in healthy cats without current or previous signs of lower urinary tract disease. The study included 108 cats, 53 males (49.5%) and 55 females (50.5%). The cats ranged in age between 7 months and 18 years, with a mean age of 4.4 years and a median age of 4.0 years. Urine was obtained by cystocentesis from all the cats, and was submitted for bacteriological analyses. Urine and urine sediment was cultured on separate blood agar plates for quantification and species identification by standard procedures. Detection of ≤103 colony forming units (cfu) per ml urine was defined as significant bacteriuria. Significant bacteriuria exceeding 105 cfu/ml was detected in one sample with a combination of Enterococcus species and Staphylococcus species. There was no bacterial growth in the urine samples from 107 cats (99.1%). Results from our study indicate that the prevalence of bacteriuria in clinically healthy, adult cats is low. Also, that contamination of samples is rare when urine is collected by cystocentesis.
It is commonly stated in major textbooks that bacterial cystitis is uncommon in cats, unless defences are compromised due to concurrent diseases or treatments.1–3 Indeed, a recent text states that; ‘less than 2% of cats below 10 years presenting with signs of feline lower urinary tract disease (FLUTD) are associated with urinary infection’. 3 However, in a study conducted by Sávik et al significant bacteriuria (≤103 cfu/ml) was observed in 15.1% of adult/mature cats with lower urinary tract disease (LUTD) where urine samples were collected by cystocentesis. 4 Other studies have also reported a relatively high occurrence of bacterial cystitis in cats presenting with signs of LUTD: Kraijer et al reported a prevalence of 22.2% [95% confidence intervals (CI): 10.2–39.1] in primary cases (midstream catheter urine, cut off 104 cfu/ml), Gerber et al reported a prevalence of 8% (95% CI: 3–16%) in referred cases (cystocentesis and catheter urine samples, no cut-off reported), whereas Eggertsdóttir et al reported a prevalence of 23% (95% CI: 12–41%) in primary cases (cystocentesis urine, cut-off 104 cfu/ml).5–7 These studies suggest that bacterial cystitis may be more frequent than has previously been assumed at least in primary care practice. There may, therefore, be a need to reevaluate the dogma that cats with LUTD rarely suffer from bacterial cystitis.
Significant bacteriuria refers to the amount of bacteriuria that presumptively represents a true urinary tract infection, and is dependent on the species (dog versus cat) and method of collection.1,3 In cats, if the urine sample is collected by cystocentesis, a cut-off value of 103 cfu/ml for defining bacteriuria has been suggested to avoid false positive results due to contamination.1,3,9 The same cut-off value has been suggested for dogs if the urine sample is collected by cystocentesis. However, the likelihood for contaminating the urine sample if other methods of collection are used (catheterisation, midstream voided or manual compression) seems to be greater for dogs than cats and in cases where urine is collected in these ways the cut-off values for these two species are different. In cats, the cut-off values for catheterisation is 103 cfu/ml and for midstream voiding and manual compression 104 cfu/ml, while in dogs they are 104 cfu/ml and 105 cfu/ml, respectively.1,3 However, the cut-off values suggested for cats are not backed up by substantial publications.
Cystocentesis is, therefore, regarded as the method of choice for urine collection in cats and in contrast to cats with non-obstructive lower urinary tract disorders it is easy to obtain sufficient amount of urine in healthy cats.7,9 By using cystocentesis contamination from the lower genitourinary tract is avoided, but correct handling of the urinary samples is, nevertheless, important in order to get reliable microbiological results.1,9–11 The most common uropathogens in cats with clinical signs of LUTD are Escherichia coli, Enterococcus faecalis and Staphylococcus felis.4,7,8
The aim of this study was to evaluate the prevalence, species and quantity of bacteria in urine specimens from healthy cats, and to compare the results with the occurrence of bacteria found in the urine from cats with LUTD in Norway reported in previous studies.
The study was designed as a prospective, descriptive and analytical study on the occurrence of occult bacteriuria in healthy cats without current or previous signs of LUTD.
One-hundred-and-eight cats were included. They were recruited from patients brought to the Department of Companion Animal Clinical Sciences, Norwegian School of Veterinary Science (NSVS) for health care reasons requiring mild sedation or anaesthesia (castration or spaying, dental problems, minor surgery or trauma, etc) between December 2007 and February 2010. According to the owners none of the cats had shown signs of inappropriate urination or lower urinary tract disorders like periuria, dysuria, stranguria, haematuria, pollakiuria and/or urinary tract obstruction. No breed or gender restrictions were made but only cats 7 months or older were included in the study. Any cats receiving treatment(s) that could interfere with the diagnostics such as antibiotics, steroids and other hormones were excluded.
Based on history, results of physical examination and analysis of blood samples, all the cats included into the study were regarded as healthy cats without diseases or medical problems that could interfere with the results in the study.
Urine samples were collected from all the cats by cystocentesis. The cats were placed in dorsal recumbency and the skin clipped and prepared aseptically. A ventral midline approach was used and cystocentesis performed with a 23 G needle attached to a 10 ml syringe. Urinalysis included commercial urine dipstick analysis (Krulab; Kruuse, Marslev, Denmark), and specific gravity measured with a refractometer (URC-Ne, ATAGO, Tokyo, Japan). Urine sediment was evaluated after centrifugation of approximately 10 ml urine for 10 min at 450 g (1500 rpm) as native samples and after staining with Sternheimer-Malbins. Findings, with respect to epithelial cells, red blood cells, and white blood cells were scored semi-quantitatively.
All the samples were cultured on the same day they were collected. Quantitative urine bacteriology was performed by streaking 1 μl of urine onto blood (5% cattle blood) agar (Blood Agar Base II, Difco) and qualitatively by cultivation on blood agar from the sediments after centrifugation before incubation at 37°C in 5% CO2 aerobic atmosphere and anaerobically. 12 In addition, the sediments were cultivated on selective bromo-thymol agar (Difco). Bacterial isolates from urine samples were identified according to Bergey's Manual of Determinative Bacteriology 13 and when possible isolates were analysed for susceptibility to antibacterial agents on Mueller-Hinton (MH) agar (Difco, Detroit, MI) with antimicrobial discs (NeoSensitabs; Rosco, Taastrup, Denmark).
The results are expressed as prevalence per cent, with 95% CI in brackets constructed by using the theory of simple binomial sequences. 14 The assumed continuously distributed variables are presented as mean values with 95% CIs or median with ranges. Prevalence data were considered not significantly different from each other at a significance level of 5% if there was an overlap in their respective 95% CI. Data assumed to be continuously distributed were compared using analysis of variance (ANOVA). In all cases, differences were considered significant if P-values were less than or equal to 0.05.
The study sample consisted of 108 cats; including 87.0% (95% CI: 79.4–92.1) domestic cats and 13.0% (95% CI: 7.9–20.6) pure breed cats. Further, 29.9% (95% CI: 22.1–39.2) were intact females, 20.6% (95% CI: 14.0–29.2) were neutered females, 12.2% (95% CI: 7.2–19.7) were intact males and 37.4% (95% CI: 28.8–46.8) were neutered males. Significantly more males than females were neutered (P = 0.0002). The mean age was 4.4 years (95% CI: 3.7–5.1), and median age was 4.0 years (range: 7 months—18 years). Table 1 shows the cats categorised into age groupings (life stage classification) according to the Feline Advisory Bureau. 15
Cats in the study sample categorised according to age.
Bacteriuria was detected in one cat, a 12-year-old neutered female domestic shorthair cat (0.9%; 95% CI: 0.2–5.1) with >105 cfu/ml of a mixed growth of Enterococcus species and Staphylococcus species. There was no bacterial growth in the urine or urine sediment samples from any of the remaining 107 cats (99.1%; 95% CI: 94.9–99.8).
Urine dipstick analysis showed mild proteinuria, pH of 5 and sediment evaluation revealed no red or white blood cells, no crystals and only small amounts of epithelial cells in the urine of the cat with bacteriuria. Moderate amounts of bacteria (cocci) were detected in the urine sediment and specific gravity was >1.050. Comparison of the sediment from the bacteriuric cat with the non-bacteriuric cats was not possible because of the small sample size of bacteriuric cats.
In a study conducted in 1984, urine samples were obtained from 12 healthy adult cats (six males and six female cats) by three different methods in order to characterise bacteriuria in the samples collected by these methods. 9 During a 3-week period six samples were obtained from each cat, two by cystocentesis, two by urethral catheterisation and two voided samples. Samples obtained by cystocentesis did not reveal bacteriuria, but bacteria were found in 18 voided samples and in four samples obtained by catheterisation. The cats in the study were adult (probably 3–6 years old), but their exact age was unknown.
To the authors' knowledge, there have been no other studies investigating the occurrence of occult bacteriuria in healthy adult cats and the present study was conducted to obtain such data. Results from this study provides a reference population for previous and future studies on LUTD in cats, and the ages represented in this study are valuable as LUTD occurs most often in middle-aged cats. 16
As in the study sample the gender distribution of healthy male and female cats visiting our clinic for preventative health care reasons was approximately equal. 7 The majority (87%) of the cats were domestic, which is representative of the feline population in Norway. 17 Many of the cats in the study were brought in for preventative health care reasons like castration or spaying, and that helps to explain why approximately 40% of the cats were young (between 7 months and 2 years of age).
This study has demonstrated that there is a minor risk for bacterial contamination when using cystocentesis as the method of urine collection, as there was only one sample that yielded bacterial growth. The cat with bacteriuria was a 12-year-old neutered female cat with >105 cfu/ml. This is appreciably above the quantity that would be expected as a result of contamination, suggesting a genuine urinary tract infection. This cat was, therefore, treated with amoxicillin/clavulanic acid 40 mg tablets q 12 h for 2 weeks based on sensitivity testing. A week after discontinuation of treatment a second urine sample still yielded bacterial growth (>105 cfu/ml E faecalis). Treatment with the same antibiotics was re-instituted (again after sensitivity testing) but was given q 8 h instead of q 12 h and continued for 3 weeks. 18 A further urine sample collected 2 weeks after treatment was sterile on culture.
The one cat with an occult urinary tract infection in this study was an older female. This is consistent with several published studies which have shown that increasing age increases the risk of a positive urine culture irrespective of disease category. 19 Other studies have shown that older (>10 years) cats, especially female cats, with systemic diseases like diabetes mellitus or chronic kidney diseases are more at risk of getting urinary tract infections with or without signs of a lower urinary tract disorder.20,21,22 However, the female cat in this study had no signs of underlying disease and the urine specific gravity was >1.050.
Results from our study indicate that the prevalence of asymptomatic bacterial cystitis in clinically healthy, adult cats is low. The occurrence of contaminated samples is low when cystocentesis is used and, therefore, a cut-off of >103 cfu/ml, as previously suggested, seems appropriate for cats having urine collected by cystocentesis. However, further studies are needed to adequately establish cut-off values for significant bacteriuria in cats when urine is collected by other techniques. Because of comparable study samples with respect to breed, gender and age, the results of this study strengthen the conclusions from previous studies showing that cats with LUTDs frequently have significant bacteriuria.
Footnotes
Acknowledgements
The authors would like to thank the veterinary nurses Cecilia Hals and Britt Bråten for their help and assistance in this project and for taking such good care of the cats. We would also like to thank Professor Stein Istre Thoresen, Norwegian School of Veterinary Science for analysing the blood samples.