Evaluation of diagnostic performance of routine automated urinalysis and association between urinary tract infection and leukocytosis

Objective: For diagnosis of urinary tract infection (UTI), urine culture has been accepted as the gold standard test. High numbers of unnecessary cultures have always been the obvious issue in clinical laboratory. Moreover, urine culture is labor- and time- intensive. In the present study we investigated the diagnostic performances of infection-related parameters of urine preliminary analysis (leukocyte esterase, nitrite, bacteria and leukocyte) in comparison to urine culture method and whether the presence of UTI causes leukocytosis.Materials and Methods: 239.029 urinalyses were retrospectively examined. A total of 3427 patients that complete blood count (CBC), urinalysis and urine culture were requested on the same day were included in the study. Leukocyte count of CBC and leukocyte and bacteria in microscopy parameters of urine analysis were compared with urine culture. Diagnostic performance of the parameters for detection of UTI was estimated.Results: 413 patients had positive urine culture results (12.0%). Among culture positive patients, leukocyte esterase and nitrite positivity were 85% (n=352) and 40% (n=166) respectively. Bacteria and leukocyte positivity on microscope were 31% (n=127) and 75% (n=310), respectively. Although negative predictive values were 80%, 62%, 75%, 58%, positive predictive values of leukocyte esterase (LE),nitrite, pyuria and bacteriuria tests were 69%, 97%, 74%, and 91% respectively.The highest specificity rate was estimated for nitrite (99%). Leukocytosis rate in patients with a positive urine culture were 23% (n=96). A strong association was detected between microscopic WBC and LE count (r = 0.827; P 0.001).Conclusion: Considering that most samples from the patients in our study have insignificant or no growth, urine microscopy and dipstick urine analysis can rule out UTI in these patients. We suggest that the investigation and application of a new algorithm in clinical practice could reduce unnecessary antibiotic prescriptions and in the clinical laboratory setting might reduce workload and cost.