Urinary tract infections in cats

Transcription

1 Original Article 227 Urinary tract infections in cats Prevalence of comorbidities and bacterial species, and determination of antimicrobial susceptibility to commonly used antimicrobial agents Roswitha Dorsch 1 ; Clara von Vopelius-Feldt 1 ; Georg Wolf 2 ; Ralf S. Mueller 1 ; Reinhard K. Straubinger 2 ; Katrin Hartmann 1 1 Clinic of Small Animal Medicine, Centre of Clinical Veterinary Medicine, Ludwig-Maximilians University Munich, Germany; 2 Institute for Infectious Diseases and Zoonoses, Department of Veterinary Sciences, Ludwig-Maximilians University Munich, Germany Keywords Urine, feline, bacterial cystitis, antibiotics, systemic disease Summary Objective: To investigate the prevalence of comorbidities (CM) in cats with urinary tract infections (UTIs), as well as the prevalence of bacterial species in cats with different CM and their antimicrobial susceptibility to the commonly used antibacterial agents doxycycline, trimethoprim-sulfamethoxazole (TMS), amoxicillin-clavulanic acid (AMC), cephalothin, and enrofloxacin. Material and methods: A retrospective analysis of cats with positive urine cultures in the years 2003 to 2009 was performed. Cats were assigned to one of four groups: cats with systemic comorbidities (scm), cats with indwelling urinary catheters (iuc), cats with local comorbidities (lcm), or cats without CM. To evaluate the potential effectivity of the antibiotics the antibacterial impact factors were calculated. Results: A total of 194 cats with 219 isolates were included in the study. In 78.4% (152/194) of cats, a CM was identified; 49.5% had a scm and 28.9% (56/194) had an iuc or a lcm. were significantly older than cats in all other groups, and the proportion of female animals was higher in cats with scm than in cats with iuc or lcm. More than half of the cats with scm did not show clinical signs of lower urinary tract disease. The most commonly isolated bacteria species were Escherichia (E.) coli, Streptococcus spp., Staphylococcus spp., and Enterococcus spp. with a significantly higher proportion of E. coli isolates in cats with scm and significantly higher proportions of Streptococcus and Staphylococcus spp. isolates in cats with iuc and other lcm. According to the antimicrobial impact factors bacterial isolates in cats with any CM were most likely susceptible to AMC and TMS. Isolates from cats with iuc and lcm had a lower likelihood to be susceptible to the tested antimicrobials than cats with scm and cats without CM. Conclusion and clinical relevance: Relevant comorbidities for bacterial urinary tract infection were identified in the majority of cats in the present study. often do not show clinical signs of lower urinary tract disease. AMC and TMS were the antimicrobial agents with the highest antimicrobial impact factor in this population of cats. Schlüsselwörter Urin, felin, bakterielle Zystitis, Antibiotika, systemische Erkrankung Zusammenfassung Ziel: Ermittlung der Prävalenz prädisponierender Begleiterkrankungen (BE) bei Katzen mit bakteriellen Harnwegsinfektionen (HWI) und der Prävalenz von Bakterienspezies bei verschiedenen BE sowie deren Sensibilität gegenüber den bei HWI häufig eingesetzten Antibiotika Doxycyclin, Trimethoprim-Sulfamethoxazol (TMS), Amoxicillin-Clavulansäure (AMC), Cephalothin und Enrofloxacin. Material und Methoden: In die retrospektive Studie wurden Katzen mit positiver Urinkultur im Zeitraum eingeschlossen. Basierend auf den Daten der Krankenakten erfolgte eine Einteilung in vier Gruppen: Katzen mit systemischen prädisponierenden BE, Katzen mit lokalen prädisponierenden BE, Katzen mit Harnblasendauerkathetern (HBDK) und Katzen ohne dokumentierte BE. Zur Ermittlung der wahrscheinlichen Effektivität der Antibiotika wurden deren antimikrobielle Impact-Faktoren berechnet. Ergebnisse: In die Studie gingen 194 Katzen mit 219 bakteriellen Isolaten ein. Davon wiesen 78,4% (152/194) eine BE auf. 49,5% (96/194) hatten eine systemische BE und 28,9% (56/194) hatten eine lokale BE oder einen HBDK. Katzen mit systemischen BE waren signifikant älter als Katzen der anderen Gruppen und häufiger weiblich als Katzen mit lokaler BE und HBDK. Mehr als 50% der Katzen mit systemischer BE zeigten keine klinischen Symptome einer Erkrankung des unteren Harntrakts. Escherichia (E.) coli, gefolgt von Streptococcus spp., Staphylococcus spp. und Enterococcus spp. waren die am häufigsten vorkommenden Isolate. Dabei lag der Anteil der E.-coli-Isolate bei Katzen mit systemischer BE signifikant höher, während bei Katzen mit HBDK und Katzen mit anderen lokalen BE der Anteil an Streptococcusund Staphylococcus-spp.-Isolaten signifikant höher war. Bei Katzen mit lokaler BE und Katzen mit HBDK ergaben sich niedrigere antimikrobielle Impact-Faktoren als bei Katzen der anderen Gruppen. Schlussfolgerung und klinische Re le vanz: Ein Großteil der Katzen mit HWI leidet unter einer prädisponierenden BE. Katzen mit systemischer BE zeigen häufig keine Symptome einer Erkrankung des unteren Harntrakts. AMC und TMS waren in dieser Katzenpopulation die Antibiotika mit den höchsten antimikrobiellen Impact-Faktoren. Correspondence to Dr. Roswitha Dorsch Clinic of Small Animal Medicine Center for Clinical Veterinary Medicine Ludwig-Maximilians University Munich Veterinärstraße München Roswitha.Dorsch@lmu.de Bakterielle Harnwegsinfektionen bei Katzen. Prävalenz prädisponierender Erkrankungen und bakterieller Isolate sowie Ermittlung der antimikrobiellen Resistenz gegenüber häufig eingesetzten Antibiotika Tierärztl Prax 2016; 44 (K): Received: July 27, 2015 Accepted after revision: January 18, 2016 Epub ahead of print: June 9, 2016 Schattauer 2016 Tierärztliche Praxis Kleintiere 4/2016

2 228 R. Dorsch et al.: Urinary tract infections in cats Introduction The development of antibiotic resistance is a major concern in public health and demands the careful and well justified use of antimicrobial agents. Antimicrobial agents are commonly used for the treatment of urinary tract infections (UTIs) in dogs and cats. Whereas a UTI was historically considered to be a rare cause of feline lower urinary tract disease (FLUTD) in cats with a prevalence of less than 3% (22), three more recent studies from Europe revealed that 8 19% of cats with FLUTD suffer from a UTI (11, 14, 30). In addition, several studies demonstrated an increased likelihood for UTIs in cats of older ages and those with concurrent diseases (1, 7, 22, 25, 27). UTIs are classified as simple or complicated based on the pre - sence of relevant comorbidities (CM) (31). A simple uncomplicated UTI is defined as a sporadic bacterial infection of the bladder in an otherwise healthy individual with normal urinary tract anatomy and function (31). In the complicated UTIs, the presence of relevant CM (e. g., diabetes mellitus, chronic kidney disease [CKD]) compromises the natural local or systemic defence mechanisms in a more profound and/or permanent way (20, 31). An increased risk for an iatrogenic UTI is associated with transurethral procedures, and placement of indwelling urinary catheters. UTIs can be observed in approximately one third of cats treated with indwelling urinary catheters (5, 18). Urogenital surgery, such as perineal urethrostomy (8, 15), urolithiasis (30), urinary incontinence, and gastrointestinal diseases such as constipation and megacolon as well as systemic neoplasia and treatment with corticosteroids are other risk factors (26, 28). Ideally, the choice of an antimicrobial agent is based on the results of bacterial culture and susceptibility testing. However, clinical signs often warrant treatment before these results are available, and clinical signs in combination with an inflammatory urine sediment and/or bacteriuria observed in stained urine sediments justify the use of antimicrobials agents at this early time point (31). Textbook recommendations for treatment of uncomplicated urinary tract infections include amoxicillin-clavulanic acid (AMC), ampicillin, cephalexin, doxycycline, enrofloxacin, or trimethoprim-sulfamethoxazole (TMS) (6, 20). In 2011, a group of specia - lists in veterinary microbiology, infectious diseases, epidemiology, and internal medicine (Working Group of the International Society for Companion Animal Infectious Diseases) published guidelines for the antimicrobial use in feline and canine urinary tract disease (31). In these guidelines, only amoxicillin and TMS are recommended as first-line drugs and urine culture and susceptibility should be the basis for treatment decisions in any case. It is however not known, how well these guidelines apply to different feline hospital populations. The purpose of this study was to determine the prevalence of relevant CM in cats with UTI and to investigate the prevalence of bacterial species and their susceptibility to recommended and commonly used antimicrobial agents in cats with UTI with and without CM. Materials and methods Animals A retrospective search for cats with positive urine cultures was performed using the database of the clinic for Small Animal Medicine in the years Cats were included, if a quantitative aerobic urine culture revealed significant bacterial growth ( 10 3 colony-forming units per millilitre (CFU/ml in cystocentesis samples, 10 4 CFU/ml in urine samples obtained via catheterisation) (31). If the cat had multiple urine cultures performed, only the results of the initial urine culture were included. The following data were collected: signalment (age, sex, breed, weight), clinical signs at presentation, diagnoses, results of urinalysis, and serum biochemistry. The cats were divided into four groups: (1) cats with systemic CM (scm), (2) cats with local CM without an indwelling urinary catheter (lcm), (3) cats with an indwelling urinary catheter (iuc) present for more than 24 hours, and (4) cats without documented CM. plus an iuc or another lcm were excluded. Medical records of cats were searched for documentation of CKD, diabetes mellitus, hyperthyroidism, other systemic CM (systemic neoplasia, immunosuppressive treatment) and local CM (urocystoliths, previous perineal urethrostomy, previous transurethral procedure, neurogenic voiding disorder, bladder neoplasia, feacal constipation, trauma). CKD was diagnosed based on the presence of azotaemia with creatinine > 1.9 mg/dl (reference range 1.9 mg/dl), concurrent urine specific gravity < Diabetes mellitus was diagnosed based on clinical signs, hyperglycaemia (reference range [RR] mg/dl), and high fructosamine (> 340 µmol/l, RR < 340 µmol/l), and hyperthyroidism was diagnosed based on increased serum total thyroxin (T4) concentration > 4.5 µg/dl (RR µg/dl) or T4 > 2.3 µg/dl and free T4 > 2.3 ng/dl (RR ng/dl) in combination with clinical signs. Infections with feline immunodeficiency virus (FIV) and feline leukemia virus (FeLV) were diagnosed with the FeLV/FIV SNAP test (IDEXX VetMed Laboratory). Urine cultures For quantitative aerobic culture, 0.1 ml of undiluted urine and 0.1 ml of a 10 3 dilution were inoculated on nutrient agar with 5% sheep blood. In addition, Gassner agar and agar without sheep blood were used for the diluted samples. The plates were incubated aerobically at 37 C and examined after 24 and 48 hours. The incubation time was extended to 72 hours if the history on the submission form indicated a chronic disease history and if there was minimal growth after 48 hours. The results were recorded in CFU/ml. Identification of the bacteria that grew was based on the phenotypic appearance of the colonies as well as on their biochemical characteristics (API, biomerieux). Tierärztliche Praxis Kleintiere 4/2016 Schattauer 2016

3 R. Dorsch et al.: Urinary tract infections in cats 229 Antimicrobial susceptibility The antimicrobial susceptibility patterns for all of the isolates were determined using the Kirby-Bauer disk diffusion method. Bacte - rial samples were applied to Mueller Hinton agar (Merck KGaA, Darmstadt, D), and antibiotic discs (Oxoid Deutschland GmbH, Wesel, D) were placed on the surface. Depending on the standardised diameter of the inhibitory zones according to valid recommendations during that period provided by the Clinical and La - boratory Standards Institute (CLSI M31-A2/S1), the bacterial isolates were considered susceptible, intermediate, or resistant (Table S1) 1. Selected reference strains (Staphylococcus aureus ATCC 29213, Enterococcus faecalis ATCC 2912, E. coli ATCC 25922) were used for quality control of disc diffusion diameters according to the CLSI M31-A2/S1 recommendation. Intermediate and susceptible isolates were included into the susceptible group because most antimicrobial agents reach higher minimal inhibitory concentrations (MICs) in the urine than in the serum (19, 21, 23). The in vitro susceptibility of the most common bacterial species to the antimicrobial agents traditionally recommended for the treatment of UTIs in dogs and cats (doxycycline, AMC, TMS, cephalothin, enrofloxacin) were evaluated. Based on the layout proposed for microtiter plates in routine antimicrobial susceptibility testing for bacteria in dogs and cats (32), cephalothin was used as a representative of the first-generation cephalosporins. As ampicillin was only tested in a small number of isolates (n = 13) during this pe - riod, it was not included in the analysis. The impact factors (likelihood that a urinary tract pathogen would be sensitive to an antimicrobial drug based on in vitro culture and sensitivity testing) for individual antimicrobial agents was calculated in all cats and in cats with different types of CM for bacterial isolates of species accounting for prevalence of 2% using the formula to help select rational antimicrobial therapy (FRAT, Equation 1) (10): The total impact factor F s is the sum of the individual impact factors for each pathogen, P is the prevalence of pathogen i (percentage of the total number of that bacterial species cultured) and S antimicrobial is the proportion of isolates susceptible to the antimicrobial in question. Statistical methods GraphPad Prism 5.04 software (GraphPad) was used for statistical analysis of the data. For descriptive analysis of continuous data the mean (± SD) were calculated. For comparison of the age distribution between the groups, an ANOVA and Dunn s multiple comparison tests were performed. The chi-square test and Fisher s exact test were used to compare categorical variables (gender, breed, sampling types, clinical signs, prevalence of Gram-positive and Gram-negative isolates, and the distribution of the most common isolates (Escherichia [E.] coli, Streptococcus spp., Staphylococcus spp., and Enterococcus spp.) between the four different groups of cats. Significance was set at p < For comparison of indivi - dual groups, a Bonferroni correction was applied and p considered significant. Results Patients, comorbidities, and clinical signs One hundred and ninety-four cats with positive urine cultures (219 bacterial isolates) were included into the study. The study population included 153 Domestic Shorthair (DSH) cats, 16 Persians, seven Maine Coons, four Siamese, two Somali, two Angora crossbreed and one cat each of the breeds: British Shorthair, Devon Rex, Norwegian Forrest, Siberian, Abyssinian, Sacred Birman, and Oriental Shorthair. In three cats the breed was not recorded. There was no difference in the number of DSH and pedigreed cats between groups (p = 0.171). In 78.4% (152/194) of cats, a CM was identified ( Table 1). Comorbidities are listed in Table 2. In cats with scm, 74 cats had one CM, 21 cats two and one cat three CM, respectively. Five cats suffered from CKD and hyperthyroidism, six cats from CKD and systemic neoplasia and another six cats with CKD had received concurrent immunosuppressive treatment. A combination of hyperthyroidism and diabetes mellitus, diabetes mellitus and systemic neoplasia, or diabetes mellitus and concurrent immunosuppressive therapy was recorded in one cat each. The one cat with three scm had CKD, diabetes mellitus and a systemic neoplasia. were significantly older than cats in all other groups and the proportion of female cats was higher in cats with scm than in cats with iuc or lcm (p < 0.005). CKD was the most common scm and diagnosed in 57.3% (55/96) of cats with scm ( Table 2). All cats with iuc (26) suffered from obstructive FLUTD. In 16/26 cats, a urine culture before placement of the indwelling UC was negative, in 6/26 cats, the urine sediment was not suggestive of bacterial infection (no pyuria, no bacteria on urine sediment) when the iuc was placed. In 4/26 cats, no information at the time of catheter placement was available. Therefore, the majority of cats in this group had catheter associated UTI (cuti). In cats with scm the proportion of cats without clinical signs of FLUTD and with polyuria/polydipsia was significantly higher than in cats with lcm (p = 0.003) and without CM (p < 0.001) ( Table 3). 1 Table S1 is available at as free Supplementary Material. Schattauer 2016 Tierärztliche Praxis Kleintiere 4/2016

4 230 R. Dorsch et al.: Urinary tract infections in cats Number of cats (%) Age in years, mean (± SD) Weight in kg, mean (± SD) Gender 96 (49.5%) 12.2 (± 4.82) 3.6 (± 1.32) 28 male (6 mi, 22 mc) 64 female (11 fi, 53 fs) 4 NA Cats with iuc 26 (13.4%) 5.1 (± 2.96) 5.4 (± 1.73) 24 male (5 mi, 19 mc) 2 female (1 fi, 1 fs) Cats with other lcm 30 (15.5%) 7.7 (± 4.28) 4.8 (± 2.29) 18 male (5 mi, 13 mc) 11 female (2 fi, 9 fs) 1 NA Cats without CM 42 (21.6%) 9.0 (± 4.93) 5.4 (± 1.73) 18 male (6 mi, 12 mc) 24 female (7 fi, 17 fs) p value < a < b < c CM = cormobidities, scm = systemic CM, iuc = indwelling urinary catheter, lcm = local CM, SD = standard deviation, NA = data not available, mi = male intact, mc = male castrated, fi = female intact, fs = female spayed a were significantly older than cats of other groups (p < 0.05). b had a significantly lower body weight than cats of all other groups (p < 0.05). c The proportion of female cats was significantly higher in cats without CM and in cats with scm than in cats with iuc (p < 0.001) and cats with lcm (p = 0.004). The proportion of male cats in the group of cats with iuc was significantly higher than in cats with scm (p < 0.001) and cats without CM (p < 0.001). Table 1 Signalment of 194 cats with bacterial urinary tract infection. Tab. 1 Signalement von 194 Katzen mit bakterieller Harnwegsinfektion Number of cats (%) CM (n) 96 (49.5%) Chronic kidney disease (55) Diabetes mellitus (16) Systemic neoplasia (22) Hyperthyroidism (11) Current immunosuppressive therapy (15) Cats with iuc 26 (13.4%) Obstructive FLUTD (26) Cats with other lcm 30 (15.5%) Urocystoliths (8) Previous perineal urethrostomy (6) Previous transurethral procedure (6) Neurogenic voiding disorders (9) Bladder neoplasia (3) Faecal constipation (2) Trauma (1) Cats without CM 42 (21.6%) Table 2 Comorbidities in 194 cats with bacterial urinary tract infections. Tab. 2 Begleiterkrankungen von 194 Katzen mit bakteriellen Harnwegsinfektionen CM = cormobidities, scm = systemic comorbidities, iuc = indwelling urinary catheter, lcm = local comorbidities Table 3 Clinical signs of feline lower urinary tract disease (FLUTD) and polyuria/polydipsia in cats with systemic comorbidities, cats with local comorbidities without indwelling urinary catheter, and cats without comorbidities. Tab. 3 Klinische Symptome einer Erkrankung der unteren Harnwege (FLUTD) und Polyurie/Polydipsie bei Katzen mit systemischen Begleiterkrankungen, Katzen mit lokalen Begleiterkrankungen ohne Harnblasendauerkatheter und Katzen ohne Begleiterkrankung Cats with lcm Cats without CM p value Macroscopic heamaturia 35/96 (36.5) 15/29 (51.7) 18/41 (43.9) Stranguria 30/95 (31.6) 20/29 (69.0) 27/41 (65.9) < a No FLUTD signs 51/96 (53.1) 6/29 (20.7) 6/41 (14.6) < b Polyuria/polydipsia 40/96 (41.7) 2/29 (6.9) 2/41 (4.9) < c CM = cormobidities, scm = systemic comorbidities, lcm = local comorbidities a Stranguria was significantly more common in cats with lcm and without CM than in cats with scm (p < 0.001). b The proportion of cats without clinical signs of FLUTD was significantly higher in cats with scm than in cats with lcm (p = 0.003) and cats without CM (p < 0.001). c PU/PD was more often observed in cats with scm than in cats with lcm (p < 0.001) and cats without CM (p < 0.001). Tierärztliche Praxis Kleintiere 4/2016 Schattauer 2016

5 R. Dorsch et al.: Urinary tract infections in cats 231 Samples and bacterial isolate identification Out of 194 samples, 219 bacterial isolates were grown. A total of 171 (88.1%) samples resulted in growth of a single bacterial isolate, 21 (10.8%) samples of two isolates and two (1.0%) samples of three isolates, respectively. Comparison of individual groups did not reveal a difference in the proportion of polymicrobial infections (all p > 0.005) (Table S2 2 ). The most commonly identified organism was E. coli (88/219, 40.2%), followed by Streptococcus species (spp.) (42/219, 19.2%), Staphylococcus spp. (35/219, 16.0%), Enterococcus spp. (15/219, 6.8%), and Micrococcus spp. (14/219, 6.4%) ( Table 4). In the 23 samples containing more than one isolate, Streptococcus spp. was cultured from 16 (69.6%) samples, and E. coli from 14 (60.9%). Enterococcus spp. and Micrococcus spp. were isolated from four (17.4%) and five (21.7%) samples, respectively. Growth of E. coli and Streptococcus spp. was identified in 2 See Supplementary Material online. eight cases (34.9%). E. coli was significantly more common in cats with scm than in cats with iuc or lcm (p < 0.001), and Streptococcus spp. and Staphylococcus spp. were significantly more common in cats with lcm than in cats with scm (p < 0.001). Antimicrobial susceptibility In all cats and in cats with any CM, AMC and TMS had the highest antimicrobial impact factor. The impact factors for all investigated antimicrobial agents were higher in cats with scm and cats without CM than in cats with iuc or lcm ( Table 5). Discussion A total of 194 cats with UTI were included into the study. Of these, 21.2% had no predisposing CM. This is higher than previously reported in another retrospective study that looked for risk factors and clinical features and identified a concurrent disease in 85% of Table 4 Bacterial isolates obtained from 194 cats with urinary tract infections and systemic comorbidities, indwelling urinary catheter, other local comorbi - dities and without documented comorbidities. Tab. 4 Bakterielle Isolate von 194 Katzen mit Harnwegsinfektion und systemischen Begleiterkrankungen, lokalen Begleiterkrankungen, Harnblasendauerkathetern sowie ohne Begleiterkrankung All cats Cats with iuc Cats with other lcm Cats without CM p value Number of isolates Gram-negative bacteria E. coli spp. Proteus spp. Pasteurella multocida Pseudomonas aeruginosa Klebsiella spp. 104 (47.5) 88 (40.2) 8 (3.7) 4 (1.8) 3 (1.4) 1 (0.5) 62 (59.6) 56 (53.8) 5 (4.8) 1 (0.01) 11 (31.4) 6 (17.1) 2 (5.7) 2 (5.7) 8 (22.9) 6 (17.1) 23 (51.1) 20 (44.4) 1 (2.2) 1 (2.2) 1 (2.2) < a < b Gram-positive bacteria Streptococcus spp. Staphylococcus spp. Enterococcus spp. Micrococcus spp. Corynebacterium spp. Lactococcus spp. 115 (52.5) 42 (19.2) 35 (16.0) 15 (6.8) 14 (6.4) 3 (1.4) 1 (0.5) 40 (38.5) 13 (12.5) 9 (8.7) 8 (7.7) 8 (7.7) 2 (1.9) 23 (65.7) 14 (40.0) 4 (11.4) 3 (8.6) 1(2.9) 27 (77.1) 8 (22.9) 13 (37.1) 3 (8.6) 2 (5.7) 20 (44.4) 7 (15.6) 9 (20.0) 3 (6.7) 1 (2.2) < c d < e Aerobic spore formers 5 (2.3) 2 (1.9) 2 (4.4) CM = comorbidities, scm = systemic comorbidities, lcm = local comorbidities, iuc = indwelling urinary catheter : Statistical analysis was not performed due to the low number of samples. a Gram-negative bacteria more common in cats with scm than in cats with lcm (p < 0.001) b E. coli significantly more common in cats with scm than in cats with iuc or lcm (p < 0.001) c Gram-positive isolates significantly more often in cats with lcm than in cats with scm (p < 0.001) d Streptococcus spp. significantly more common in cats with lcm than in cats with scm (p < 0.001) e Staphylococcus spp. significantly more common in cats with lcm than in cats with scm (p < 0.001) Schattauer 2016 Tierärztliche Praxis Kleintiere 4/2016

6 232 R. Dorsch et al.: Urinary tract infections in cats Table 5 Antimicrobial impact factors for all isolates and isolates obtained from cats with systemic comorbidities, cats with indwelling urinary catheters, cats with other local comorbidities and cats without comorbidity. Tab. 5 Antimikrobielle Impact-Faktoren für alle Isolate und Isolate von Katzen mit systemischen Begleiterkrankungen, Harnblasendauerkatheter, lokalen Begleiterkrankungen und ohne Begleiterkrankung Substance Tested isolates (n) Antimicrobial impact factor All cats Cats with iuc Cats with lcm Cats without CM Doxycycline TMS , AMC Cephalothin Enrofloxacin CM = comorbidities, scm = systemic comorbidities, lcm = local comorbidities, iuc = indwelling urinary catheter, AMC = amoxicillin clavulanic acid, TMS = trimethoprim-sulfamethoxazole cats (26). Possible reasons for the higher proportion of cats without CM in the present study are that cats with concurrent local and systemic CM were excluded from the study, cats were not screened prospectively for concurrent predisposing CM and only specific diseases for which studies have shown an increased prevalence of UTI were defined as predisposing CM. Nearly 50% of cats in the present study suffered from one or several predisposing scm, and one third of cats suffered from a predisposing lcm. Looking at the most common scm, 28.1% of all cats study had CKD. This is similar to results obtained in another retrospective study on UTI (26). In that previous study, renal failure, defined as azotaemia with inadequate urine concentrating ability, was seen in 29% of cats with UTI, and anatomical lower urinary tract disorders in 34.8% of cats with UTI (26). The inclusion criteria for CKD used in this previous study and also in the present study account only for cats with CKD that are already azotemic, and it remains unclear how many cats with earlier stage CKD are in the group of cats without CM. In several previous studies, the prevalence of UTI increased with age, and females were more often affected. However, most of the previous studies focused on cats with specific comorbidities such as CKD, hyperthyroidism and diabetes mellitus (3, 27). In contrast, data of the present study and also data from a recent investigation identified a more uniform distribution of UTI among cats of all ages (26). were significantly older (12.2 years, SD ± 4.82) than cats of all other groups (p < 0.05), and this was particularly evident in cats with iuc (5.1 years ± 2.96). This is not surprising because especially CKD and hyperthyroi - dism are diseases of the elderly cat, and obstructive FLUTD with a need for transurethral procedures or iuc is typically seen in young and middle aged male cats. In cats with iuc, UTI was catheter associated in the majority of cases. Therefore, even though bacterial UTI is only responsible for clinical signs of FLUTD in up to 13% of young cats (11, 22), it is nonetheless a regular complication following the management of cats with obstructive FLUTD (18). Strictly speaking, bacterial UTI (bacteriuria in association with clinical signs and/or signs of inflammation) must be distinguished from bacterial colonization or asymptomatic bacteriuria (31). However, in cats with obstructive FLUTD, that are treated with iuc for se - veral days, it is very difficult to assess the impact of bacteriuria on the severity of clinical signs, because most of these cats will have ongoing clinical signs and an inflammatory urine sediment due to mechanical irritation and sterile inflammation of the bladder and urethra caused by the catheter and also due to ongoing primary disease, e. g. idiopathic cystitis. Excluding those cats with iuc, 60% of cats in the present study were female with no significant difference in gender distribution between disease groups. The higher prevalence in older females has also been identified in a previous large epidemiologic study (22). It is assumed that it is easier for the resident gastrointestinal flora from the perineum to ascend the shorter and wider urethra of females than males and to establish a UTI (25). The presence of a scm could contribute to the impaired local and systemic immune defense in these cats. Clinical signs of FLUTD were reported in 85.4% of cats with no CM and in 79.3% of cats with lcm (excluding cats with iuc because clinical signs cannot be assessed while the catheter is in place), opposed to less than half of the cats with scm in the pre - sent study. Even lower proportions of cats with clinical signs have previously been reported in cats with different CM (31.4%) (27) and cats with CKD (28%) (33), which highlights the high pro - portion of occult UTIs particularly in cats with renal disease and endocrinopathies. A total of 171 (88.1%) samples resulted in growth of a single bacterial isolate, 21 (10.8%) samples of two isolates, and two samples (1.0%) of three isolates, respectively. This is lower than previously reported proportions of polymicrobial infections of 16 22% in different populations of cats, e. g. only cats with CKD, hyperthyroidism, and diabetes mellitus (27), only cats with CKD (33) or only cats with clinical signs of FLUTD (24) or cats with occult UTI (25). It is notably lower than in a study with a similar distribution of comorbidities with 22% polymicrobial UTI (26). A Tierärztliche Praxis Kleintiere 4/2016 Schattauer 2016

7 R. Dorsch et al.: Urinary tract infections in cats 233 possible explanation is that a higher proportion of first opinion cases in the present study led to a lower number of polymicrobial UTIs. In cats with iuc, urine samples were more often catheter-derived than in other groups. The proportion of samples with polymicrobial isolates was 26.9% and 20.0% in cats with iuc and cats with lcm, respectively, compared to 8.3% and 7.1% in cats with scm, and cats without CM, respectively. In humans, polymicrobial UTIs are particularly seen in patients with indwelling urinary catheters (9, 29). Even though analysis did not reveal a significant difference in the present study, it seems likely that iuc and transurethral procedures promote UTIs with more than one isolate. The most commonly cultured bacterial species were E. coli, Streptococcus spp., Staphylococcus spp., and Enterococcus spp. E. coli was also the predominant isolate in other studies in which E. coli represented 37.7% (24) to 71% of the isolates in cats with UTI (33). Particularly in cats with diabetes mellitus, hyperthyroi - dism, or CKD, E. coli seems to be predominant (2, 33), while in studies that included only cats with lower urinary tract signs, an outbalance of the Gram-positive organisms has been noted (24). In the present study, the proportion of E. coli isolates was also significantly higher in cats with scm, while Streptococcus and Staphylococcus spp. were significantly more often isolated from cats with iuc and cats with lcm. Systemic immunity could be more important in preventing E. coli urinary tract infection, whereas local defence mechanisms might be more important for protection against Gram-positive bacteria, such as Staphylococcus spp., Streptococcus spp., and Enterococcus spp. AMC and TMS were the antimicrobial agents with the highest antimicrobial impact factors in all groups of cats except for cats without CM, in which enrofloxacin was the drug with the highest impact factor. Antimicrobial impact factors for all antimicrobial agents were lower in cats with lcm and cats with iuc than in other groups of cats. A significant correlation was seen between the occurrence of urinary multiple drug-resistant E. coli and the number of antibiotic families used within the last 3 months and a trend for an increased risk with increasing days of hospitalisation has been shown (17). Unfortunately, in the present study, previous and current antimicrobial treatment was not included in the analysis but cats with FLUTD often suffer from recurrent episodes and recurrent obstruction (13, 14). Cats with perineal urethrostomy and with neurogenic voiding disorder have a permanent impairment of the urinary tract, and cats with idiopathic cystitis, urolithiasis and bladder neoplasia often have a protracted course of disease with recurrent episodes of FLUTD with intermittent improvement to antibiotic treatment due to secondary UTI. In addition, even though it is well known that idiopathic cystitis is the most common cause of FLUTD, these cats are still very often routinely treated with antimicrobial agents. Therefore, it is possible that cats with iuc and cats with local abnormalities received more types and courses of antimicrobial agents which led to the lower susceptibility rates and impact factors. Current guidelines recommend amoxicillin and TMS as firstline drugs for uncomplicated urinary tract infections (31). Despite the limitation of the retrospective design of the present study and the fact that not all of the cats were screened for systemic and local comorbidities to the same extent and some CM may have been missed, cats without CM in the present study resemble the small proportion of cats with simple uncomplicated UTI. For these cats, TMS and amoxicillin without clavulanic acid are recommended as first-line treatments. Unfortunately, ampicillin as the representa - tive of the penicillins without clavulanic acid and assumed crossresistance with amoxicillin, was not included in the analysis because it was only tested in a small number of isolates. For canine uropathogens, antimicrobial impact factors of AMC are notably higher than those of ampicillin (4, 16). Therefore, it could be that AMC is advantageous over amoxicillin without clavulanic acid for urinary tract pathogens in cats as well. Knowing that the increased prevalence of extended spectrum β-lactamase-producing bacteria is a major concern in human and veterinary medicine and that the empiric use of broad spectrum antimicrobial agents such as AMC is a focus of criticism, inclusion of ampicillin in the standard panel is mandatory and could help to identify subgroups of cats with UTIs for which ampicillin is equally effective. First-generation cephalosporins were previously recommended for treatment of UTI, if cocci were identified in alkaline urine, assuming that Staphylococcus spp. was the most likely organism (6). The inherent resistance of Enterococcus spp. to cephalosporins was the reason for removal of cephalosporins from the list of first-line treatments for UTIs in dogs and cats (31). Cephalothin was the antimicrobial agent with the lowest antimicrobial impact factor in the present study in all groups of cats and thus indeed cannot be recommended for empiric treatment. The recommendations for the use of fluoroquinolones have changed in the last few years toward a more reluctant usage (31). Whereas formerly recommended as the initial antibiotic choice for a UTI when rods were identified in acidic urine upon examination of the urinary sediment (6, 20), more recent recommendations confine the use of fluoroquinolones for UTIs with bacteria resistant to other drugs and for the treatment of pyelonephritis (31). Over a 10-year period ( ), decreased effectiveness in antimicrobial impact factors for enrofloxacin, cephalexin, and oxytetracycline were observed in a study that included a high number of canine samples (670 culture-positive urine specimens) (16). These changes were attributed to the increased use of these drugs as first-line antimicrobial agents for suspected urinary tract infections in dogs. A study that looked for changes in antimicrobial resistance over a 10-year period that also included samples of the present study noticed a decrease in antimicrobial impact factors for most of the tested antimicrobials including enrofloxacin as well (12). This emphasises the need for restriction to few first-line agents. The generally lower impact factors of the tested antimicrobial agents in cats with iuc and cats with lcm highlights the importance of bacterial culture and susceptibility in these cases to Schattauer 2016 Tierärztliche Praxis Kleintiere 4/2016

8 234 R. Dorsch et al.: Urinary tract infections in cats Conclusion for practice Relevant comorbidities were identified in almost 80% of cats with urinary tract infections. Clinical signs were more likely to be observed in cats with local comorbidities and without comorbidities than in cats with systemic comorbidities. The prevalence of E. coli, Streptococcus spp. and Staphylococcus spp. differed significantly between disease groups. Amoxicillin-clavulanic acid and trimethoprim-sulfamethoxazole were the antimicrobial agents with the highest antimicrobial impact factor in this population of cats. Particularly in cats with local comorbidities, we recommend to base the choice of the antimicrobial agent always on the results of culture and susceptibility testing. avoid unnecessary use of antimicrobial agents in cats without UTI and the use of appropriate antimicrobials in case of a UTI. Conflict of interest statement None of the authors of this paper has a financial or personal relationship with other people or organizations that could inappropriately influence or bias the content of the paper. References 1. Aakra A, Vebo H, Snipen L, et al. Transcriptional response of Enterococcus faecalis V583 of erythromycin. Antimicrob Agents Chemother 2005; 49: Bailiff NL, Nelson RW, Feldman EC, et al. Frequency and risk factors for urinary tract infection in cats with diabetes mellitus. J Vet Intern Med 2006; 20: Bailiff NL, Westropp JL, Nelson RW, et al. Evaluation of urine specific gra - vity and urine sediment as risk factors for urinary tract infections in cats. Vet Clin Pathol 2008; 37: Ball KR, Rubin JE, Chirino-Trejo M, et al. Antimicrobial resistance and prevalence of canine uropathogens at the Western College of Veterinary Medicine Veterinary Teaching Hospital, Van Vet J 2008; 49: Barsanti JA, Shotts EB, Crowell WA, et al. Effect of therapy on susceptibility to urinary tract infection in male cats with indwelling urethral catheters. J Vet Intern Med 1992; 6: Barsanti JA. Genitourinary infections. In: Infectious Diseases of the Dog and Cat. Greene CE, ed. St. Louis: Elsevier/Saunders 2006; Bartges JW, Barsanti JA. Bacterial urinary tract infections in cats. In: Current Veterinary Therapy XIII. Bonagura JD, ed. Philadelphia: Elsevier Saunders 2000; Bass M, Howard J, Gerber B, et al. Retrospective study of indications for and outcome of perineal urethrostomy in cats. J Small Anim Pract 2005; 46: Biering-Sorensen F, Bagi P, Hoiby N. Urinary tract infections in patients with spinal cord lesions: treatment and prevention. Drugs 2001; 61: Blondeau JM, Tillotson GS. Formula to help select rational antimicrobial therapy (FRAT): its application to community- and hospital-acquired urinary tract infections. Int J Antimicrob Agents 1999; 12: Dorsch R, Remer C, Sauter-Louis C, et al. Feline lower urinary tract disease in a German cat population. A retrospective analysis of demographic data, causes and clinical signs. Tierärztl Prax 2014; 42 (K): Dorsch R, Von Vopelius-Feldt C, Wolf G, et al. Feline urinary tract pathogens: prevalence of bacterial species and antimicrobial resistance over a 10-year period. Vet Rec 2015; 176: Eisenberg BW, Waldrop JE, Allen SE, et al. Evaluation of risk factors associated with recurrent obstruction in cats treated medically for urethral obstruction. J Am Vet Med Assoc 2013; 243: Gerber B, Boretti FS, Kley S, et al. Evaluation of clinical signs and causes of lower urinary tract disease in European cats. J Small Anim Pract 2005; 46: Griffin DW, Gregory CR. Prevalence of bacterial urinary tract infection after perineal urethrostomy in cats. J Am Vet Med Assoc 1992; 200: Hall JL, Holmes MA, Baines SJ. Prevalence and antimicrobial resistance of canine urinary tract pathogens. Vet Rec 2013; 173: Hernandez J, Bota D, Farbos M, et al. Risk factors for urinary tract infection with multiple drug-resistant Escherichia coli in cats. J Feline Med Surg 2013; 15: Hugonnard M, Chalvet-Monfray K, Dernis J, et al. Occurrence of bacteri - uria in 18 catheterised cats with obstructive lower urinary tract disease: a pilot study. J Feline Med Surg 2013; 15: Kietzmann M, Niedorf F, Kramer S, et al. Plasma and urine concentrations of marbofloxacin following single subcutaneous administration to cats. Berl Munch Tierarztl Wochenschr 2011; 124: Labato MA. Uncomplicated urinary tract infection In: Kirk s Current Veterinary Therapy XIV Bonagura JD, Twedt DC, ed. St. Louis: Elsevier Saunders 2009; Lees GE, Rogers KS. Treatment of urinary tract infections in dogs and cats. J Am Vet Med Assoc 1986; 189: Lekcharoensuk C, Osborne CA, Lulich JP. Epidemiologic study of risk factors for lower urinary tract diseases in cats. J Am Vet Med Assoc 2001; 218: Ling GV, Creighton SR, Ruby AL. Tetracycline for oral treatment of canine urinary tract infection caused by Pseudomonas aeruginosa. J Am Vet Med Assoc 1981; 179: Litster A, Moss SM, Honnery M, et al. Prevalence of bacterial species in cats with clinical signs of lower urinary tract disease: recognition of Staphylococcus felis as a possible feline urinary tract pathogen. Vet Microbiol 2007; 121: Litster A, Moss S, Platell J, et al. Occult bacterial lower urinary tract infections in cats urinalysis and culture findings. Vet Microbiol 2009; 136: Martinez-Ruzafa I, Kruger JM, Miller R, et al. Clinical features and risk factors for development of urinary tract infections in cats. J Feline Med Surg 2012; 14: Mayer-Roenne B, Goldstein RE, Erb HN. Urinary tract infections in cats with hyperthyroidism, diabetes mellitus and chronic kidney disease. J Feline Med Surg 2007; 9: Pressler BM, Vaden SL, Lane IF, et al. Candida spp. urinary tract infections in 13 dogs and seven cats: predisposing factors, treatment, and outcome. J Am Anim Hosp Assoc 2003; 39: Ronald A. The etiology of urinary tract infection: traditional and emerging pathogens. Dis Mon 2003; 49: Saevik BK, Trangerud C, Ottesen N, et al. Causes of lower urinary tract disease in Norwegian cats. J Feline Med Surg 2011; 13: Weese JS, Blondeau JM, Boothe D, et al. Antimicrobial use guidelines for treatment of urinary tract disease in dogs and cats: antimicrobial guidelines working group of the international society for companion animal infectious diseases. Vet Med Int 2011; 2011: 1 9. doi /2011/ Werckenthin C, Luhofer G, Bottner A, et al. Layout proposal for a microtitre plate to be used in routine antimicrobial susceptibility testing of bacteria from infections of dogs and cats. Berl Munch Tierarztl Wochenschr 2008; 121: White JD, Stevenson M, Malik R, et al. Urinary tract infections in cats with chronic kidney disease. J Feline Med Surg 2013; 15: Tierärztliche Praxis Kleintiere 4/2016 Schattauer 2016