The aim of this study was to evaluate a new urine screening analyser in a H
ospital Bacteriology Laboratory: the Celifacts Urine Screening Analyser (Mi
crobial Systems Limited (MSL), Coventry, England). a cohort of 1036 urine s
pecimens were analysed by both the CellFacts and routine traditional method
s. Using the CDC urinary tract infection decision levels, and compared to r
eference methods, the sensitivity and specificity were respectively 88.9% a
nd 76.2%, and the predictive negative value and predictive positive value r
espectively 93.3% and 64.8%. Compared to the microscopy, the correlations o
f white blood cell count and red blood cell count were good, respectively (
4 = 0.8, p < 0.0001) and (r = 0.6, p < 0.0001). These results indicate that
, although several positive samples were not reported, CellFacts facilitate
s efficient, rapid screening of infected urine specimens. This leads to per
form a cytobacteriological analysis only on those results screened as posit
ive by the automatic analysis system. (C) 2001 Editions scientifiques et me
dicales Elsevier SAS.