Urine culture from peed sample not reliable for female dogs

Abstract

OBJECTIVE: To evaluate whether culture of urine collected by clean midstream voiding had acceptable agreement with culture of urine collected by cystocentesis or cystoscopy in female dogs with lower urinary tract signs. METHODS: Female, client-owned dogs with lower urinary tract signs (n = 53) were enrolled in a prospective study between December 15, 2010, and August 1, 2019. Three urine samples were collected and submitted for culture within 24 hours: a standard voided sample, clean voided sample following cleansing with chlorhexidine, and a cystocentesis or cystoscopy sample. RESULTS: One or more cultures were positive for bacterial growth in 37 of 53 patients (70%). Agreement between the number of isolates grown on both voided samples (standard voided and clean voided) was substantial (κ = 0.69), but only fair between both voided samples and the cystocentesis/cystoscopy sample (clean voided and cystocentesis/cystoscopy, κ = 0.40; standard voided and cystocentesis/cystoscopy, κ = 0.34). With a cutoff of ≥ 105 CFU/mL, with cystocentesis/cystoscopy serving as the gold standard for bacteriuria, standard voided urine culture sensitivity was 75% (specificity, 88%; positive predictive value, 71%; negative predictive value, 90%) and clean voided urine culture sensitivity was 64% (specificity, 97%; positive predictive value, 90%; negative predictive value, 86%). CONCLUSIONS: While vulvar cleansing subjectively reduced contamination, no statistical difference was found between midstream samples obtained before (standard voided) or after (clean voided) vulvar cleansing. Neither voided sample had substantial agreement with the cystocentesis/cystoscopy sample, even when quantitative interpretive cutoffs were applied. CLINICAL RELEVANCE: Cystocentesis should remain the gold standard collection method for urine culture in female dogs.