Bloodstream infections are a significant cause of morbidity and mortality in critically ill dogs, but due to cost and difficulties in sample acquisition, blood cultures are infrequently obtained. In ill dogs, urine cultures may be recommended as surrogates for blood cultures. In order to determine the outcome agreement between parallel urine and blood cultures, we retrospectively analyzed parallel blood and urine specimens submitted for culture from dogs at the NC State Veterinary Hospital between 2011 and 2016. Positive cultures were reported from 15% of the submitted blood specimens and 23% of the submitted urine specimens. A total of 295 urine and blood samples were submitted in parallel, with positive growth demonstrated in 14 concordant and five discordant pairs. A kappa statistic comparing blood and urine culture outcomes was 0.266 (fair) when all parallel growth was included, including concordant and discordant results, and 0.170 (poor) when restricted to parallel concordant growth. The sensitivity of urine to reflect concordant bloodstream bacterial organisms was 30%, with a specificity of 87%. The positive and negative predictive values were 30% and 88%, respectively. Of dogs with both specimens positive on bacterial culture, 7 of 7 (100%) with suspected urogenital infection sources were concordant. All dogs with discordant bloodstream and urinary infections were immunosuppressed. Urinary coagulase-positive Staphylococcus isolates were most likely to be concordant with bloodstream infections. In conclusion, we found that urine culture is neither a substitute nor a screen for blood culture. Blood cultures should be performed in any potentially septic animal, especially those that are considered immunosuppressed.
Keywords: blood culture; bloodstream infections; concordance; dogs; urine culture.
Copyright © 2018 American Society for Microbiology.