JAC Advance Access published online on January 16, 2004
Journal of Antimicrobial Chemotherapy, doi:10.1093/jac/dkh060
© 2004 by The British Society for Antimicrobial Chemotherapy
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original article
1 The JONES Group/JMI Laboratories,
Inc., 345 Beaver Kreek Centre, Suite A, North Liberty, IA 52317; Tufts University School of Medicine,
Boston, MA, USA
* Corresponding author. E-mail: ronald-jones{at}jmilabs.com.
Received 16 May 2003
; revised 1 November 2003
; accepted 8 November 2003
Background: Laboratories worldwide
carry out MIC/disc diffusion (DD) tests using various national methods, often
published by local organizations. Potentially different results
may complicate drug development or registration between the US methods
and those used in the European Union (EU). This study uses a new
des-F(6)quinolone, garenoxacin, to compare in vitro results
among the most utilized antimicrobial susceptibility testing methods. Methods: Investigators in France, Germany, Spain,
Sweden and the UK tested two bacterial collections designated, challenge
(CC; n = 330) containing quinolone-resistant
strains and national (NC; n = 540) that
were recent clinical isolates (2000-2001). Results were
compared with values derived from the National Committee for Clinical
Laboratory Standards (NCCLS) methods, carried out by the US coordinating
site. Discords (greater than four-fold) between MIC method results
were repeated. Results were analysed for variation from the NCCLS
results (±1-2 dilutions or
greater than 3-6 mm) and by regression statistics. Ciprofloxacin
was used as the control quinolone agent. Results: CC and NC testing compliance averaged
98.4% and 86.4% among participating centres, respectively,
and correlation (r) of the national method to the
NCCLS MIC was: France (0.98), Germany (0.95), Spain (0.98), Sweden
(0.96) and UK (0.95). CC produced MIC results that were identical
to the NCCLS (Spain) to 0.6 x log2 greater
(Germany), but the percentage of strains ±2
dilutions versus NCCLS MIC results was 98% (Germany)-100% (Sweden,
UK). Similar patterns were observed for the NC (r = 0.93-0.98),
and all nations had >99% of results within ±2 dilutions (±6
mm for disc tests) of NCCLS values. MIC results from four national methods
were slightly elevated compared to the NCCLS result (average 0.2 x log2). Control ciprofloxacin
MIC and DD test results showed similar patterns. Conclusions: Inter-method garenoxacin susceptibility
test results indicate that MIC and zone diameter endpoints derived
from five EU methods compare favourably to the NCCLS method results,
and generally were identical or a fraction of a log2 dilution
step higher. In contrast, zone diameters for garenoxacin and ciprofloxacin
were routinely larger for the EU methods. This unique inter-method ‘bridging’ experiment
allows regulatory agencies to better correlate in vitro testing
results derived from procedures that use different national methodologies.
Keywords: fluoroquinolones, NCCLS, inter-method comparisons
Comparisons of the in vitro susceptibility
testing results for garenoxacin using six different national methods:
report from the garenoxacin international bridging study
2 The JONES Group/JMI Laboratories,
Inc., 345 Beaver Kreek Centre, Suite A, North Liberty, IA 52317;
CiteULike 
Connotea 
Del.icio.us What's this?