Are bacterial culture quantifications reliable? Comparative performance of the WASP automated inoculation instrument in the era of ISO 15189 accreditation

Purpose. Isolating colonies and obtaining accurate colony counts from bacterial cultures are critical steps for the optimal management of infected patients. The uncertainties in the colony count results from the bacterial cultures were evaluated by verifying the performance of the WASP inoculation system according to the International Organization for Standardization (ISO) 15189 standard. Methodology. We first (i) evaluated the cross-contamination and precision of the WASP instrument (Copan Diagnostics, Italy) and (ii) established enumeration reading grids for urine, swab, bronchopulmonary specimens (BPSs) and catheter tip cultures. Subsequently, 72 clinical samples were tested to compare the results of the WASP, PREVI Isola (bioMerieux, France) and manual inoculation methods. Results. The WASP method did not show cross-contamination. The coefficient of variation for the colony counts in the repeatability experiment was evaluated for 10 mu l and 30 mu l loop protocols and determined to be 29 and 14 %, respectively. The agreement between the automated and manual methods and between the automated methods for the colony counts was high (94.4 and 100 %, respectively). The WASP method yielded better isolation quality compared to the manual method (P=0.020) and to the PREVI Isola only when polymicrobial specimens were considered (P=0.014). For quantification evaluation, the measurement uncertainty was evaluated to 1.8 x 10(3) c.f.u. ml(-1) for a suspension of Escherichia coli at 10 4 c.f.u. ml(-1). Conclusion. We report the verification of the performance of the WASP instrument and describe a rapid procedure for achieving semi-quantitative cultures from BPSs and catheter tips. Quantitative interpretation of the bacterial cultures should be performed with caution.