Evaluation of the automated urine culture system Alfred60 for the screening of significant bacteriuria

Abstract number: P1370

Huang T.D., Debecker C., Simon A.

Objectives: Analysis of urine samples is the most prescribed bacterial diagnostic test in clinical laboratory. We aimed to evaluate the ability of the new automated incubation system Alfred60 (Alifax, Italy), compared to classic solid media culture, to screen for samples with significant bacterial growth which would be submitted for manual culture.

Methods: 1025 mid-stream urine samples collected from hospitalised and ambulatory patients within a one-month period were plated onto a chromogenic agar Uriselect4 (Uri4, Biorad) using calibrated loop and transferred into the Alfred60 system for automated testing. A turbidity-based growth curve generated by Alfred60 for each sample was converted to an approximate bacterial quantification in CFU/ml. A sample was reported as positive by Alfred60 if it displayed a bacterial count of leqslant R: less-than-or-eq, slant10000 CFU/ml. By Uri4, urine samples were classified taking account the bacterial growth quantification, the number of species and the presence of the contamination flora. A sample was considered positive by Uri4 after interpretation applying the Kass criteria. Samples reported as negative by Alfred60 and positive by Uri4 underwent further discrepancy analysis including reviews of urinary leucocytes count and patient clinical data.

Results: Of the 1025 urine samples tested, 137 were classified as positives (13%) and 733 negatives (72%) by both methods, representing a total agreement of 85%. 136 samples (13%) were positive by Alfred60 and negative by Uri4 while 19 (2%) were negative by Alfred and positive by Uri4. Of these 19 major discrepancies, 8 bacteriuria were considered clinically insignificant, 7 clinically relevant and the signification remain unknown for 3 patients. Positive and negative predictive values after discrepancy analysis were 50% and 99% respectively.

Conclusion: Alfred60 is a promising urine screening system which can correctly and quickly (5 h) render results as negative for the majority (73%) of urine cultures requested and therefore save as many unnecessary solid culture media. However, Alfred60 positive samples still require further confirmation by classic culture method with possible identification and susceptibility testing of pathogens.

Session Details

Date: 19/04/2008
Time: 00:00-00:00
Session name: 18th European Congress of Clinical Microbiology and Infectious Diseases
Location: Barcelona, Spain
Presentation type:
Back to top