Background: Automated urine sediment analysis of white blood cells (WBCs) and bacteria is a promising approach for urinary tract infections (UTIs) screening. However, available data on their screening efficacy is inconsistent.
Methods: English articles from Pubmed, EMBASE, and Web of Science published before December 1, 2012 were analyzed. The Quality Assessment for Studies of Diagnostic Accuracy (QUADAS) tool was used to evaluate the quality of eligible studies. Performance characteristics of WBCs and bacteria (sensitivity, specificity, and other measures of accuracy) were pooled and examined by random-effects models.
Results: Nineteen studies containing 22,305 samples were included. Pooled sensitivities were 0.87 (95% confidence interval [CI], 0.86-0.89) for WBCs and 0.92 (95% CI, 0.91-0.93) for bacteria. Corresponding pooled specificities were 0.67 (95% CI, 0.66-0.68) for WBCs and 0.60 (95% CI, 0.59-0.61) for bacteria. Areas under the summary receiver operating characteristics curves were 0.87 and 0.93 for WBCs and bacteria, respectively. The major limitation of eligible studies was that enrolled subjects were often not representative of clinical patient populations in which UTI would be suspected.
Conclusions: WBC and bacterial measurements by the UF-100 and UF-1000i are useful indicators in UTI screening; however, the performances of these systems should be rigorously evaluated by additional studies.
Keywords: Automated urine sediment analyzer; Meta-analysis; Systematic review; Urinary tract infection.
Copyright © 2013 Elsevier B.V. All rights reserved.