Levofloxacin showed comparable in vitro susceptibility to ciprofloxacin among Enterobacteriaceae, Pseudomonas aeruginosa, enterococci, and Staphylococcus aureus, while greater susceptibility was observed in Stenotrophomonas maltophilia and Staphylococcus epidermidis, mainly when oxacillin resistant. The susceptibility of Streptococcus pneumoniae to levofloxacin reached 99%.The recent introduction of levofloxacin, a new fluoroquinolone (FQ), into the Italian clinical scenario led to the need to locally confirm its in vitro activity. Nowadays, only qualitative studies on selected bacterial species have been reported with levofloxacin in Italy (1,3,4,5,6,10,13) but larger studies on clinical isolates are missing. We report here the major findings of a survey carried out on the bacterial species most frequently encountered in the routine work of 24 Italian laboratories distributed Italywide (see Acknowledgments).Each center was requested to collect 194 consecutive, no-copy clinical isolates belonging to a predefined list of bacterial species without any limitation on the ward and/or sample of isolation. All the isolates had to be identified according to the laboratory method used to obtain them. Participants were asked to limit the number of collected Escherichia coli, Staphylococcus aureus, and Staphylococcus epidermidis strains to 30 each per center. The MICs of levofloxacin and ciprofloxacin were determined locally by the Etest (AbBiodisk) method according to the manufacturer's instructions. All the assays were performed on MuellerHinton agar except for Streptococcus pneumoniae and Moraxella catarrhalis, which were tested by using MuellerHinton agar plus 5% sheep blood, and Haemophilus influenzae, which was tested by using Haemophilus test medium (all media were from the same batch). Susceptibility breakpoints were Յ2 g/ml for levofloxacin and Յ1 g/ml for ciprofloxacin according to NCCLS (11). To contain intralaboratory variability, quality controls with E. coli strain ATCC 25922, S. pneumoniae strain ATCC 49619, and H. influenzae strain ATCC 49247 were requested of each center before, during, and after the study's conclusion. The Etest method was selected because of its reliability in testing FQs (9), its ease of use, and its definition of the MIC. Between January and November 2000, case report forms for a total of 4,003 clinical isolates were collected. The quality control results confirmed that the Etest assays were carried out properly during the study, and any center results had to be discharged. As reported by others (9), a slight overestimation (1.6%) of the ciprofloxacin MIC for E. coli ATCC 25922 occurred, but the increase never determined a change in the category. Detailed figures on the isolation wards, samples, and the samples' FQ susceptibilities will be published elsewhere. The isolates and their susceptibilities to levofloxacin and ciprofloxacin are illustrated in Table 1. The Enterobacteriaceae were unhomogeneously susceptible to the study drugs, with a percentage of susceptible strains ranging fr...