Abstract
Despite the first-line recommendation of fosfomycin for uncomplicated urinary tract infections (UTIs), there are pressing barriers for optimizing its use for the treatment of non-Escherichia coli Enterobacterales UTI. There are no approved breakpoints for oral use against other Enterobacterales, and the recommended agar dilution (AD) reference method for minimal inhibitory concentration (MIC) determination is largely impractical. Using 160 clinical Klebsiella pneumoniae isolates, we sought to understand rates of skipped wells and MIC imprecision in broth microdilution (BMD) and how that compares to rates of error using AD. Though the Clinical and Laboratory Standards Institute refers to the skipped well phenomena in their recommendation against the use of BMD, there is a paucity of data on its frequency. While AD and BMD produced similar MIC 50/90 values (32/256 µg/mL for AD and 64/256 µg/mL for BMD), essential agreement was poor. No-growth wells at concentrations below the MIC occurred in up to 10.9% of wells at a given concentration, as the most frequent scientific error. Growth in concentrations above the measured MIC occurred in up to 3.3% of wells and was seen within three dilutions of the MIC for BMD. Observation of single colonies either at or beyond the measured MIC for AD was also common and occurred up to 8.3% and 2.5% of the time, respectively. The frequent scientific error in both testing methods should prompt re-evaluation of AD guidelines and expansion of MIC testing methods for fosfomycin susceptibility testing, as poor agreement with another method prone to scientific error should not be the main detractor from BMD use. IMPORTANCE Despite the recommendation of fosfomycin for uncomplicated urinary tract infections (UTIs), there are barriers for optimizing its use. There are no approved breakpoints for oral use against other Enterobacterales, and the recommended agar dilution (AD) reference method for MIC determination is largely impractical. The use of broth microdilution (BMD) for fosfomycin testing is not recommended by the Clinical and Laboratory Standards Institute due to unsatisfactory precision and skipped wellsoccurrence of no-growth in a single well before the minimal inhibitory concentration (MIC)-and trailing endpoints. We sought to understand rates of skipped wells and growth at concentrations above measured MICs in BMD and how that compares to scientific error using AD. No-growth wells at concentrations below the MIC occurred in up to 10.9% of wells for BMD and single colonies at or beyond measured MICs for AD were also common. Frequent scientific error in both methods should prompt re-evaluation of both AD and BMD for fosfomycin susceptibility testing. T here is a current lack of oral antibiotics available to treat urinary tract infections (UTIs) caused by multidrug-resistant (MDR) pathogens. Fosfomycin-a first-line antibiotic for the treatment of uncomplicated UTI-is often used to treat UTI caused by non-Escherichia coli Enterobacterales (1-5) despite the absence of approved indications. The approval of fosfomycin in the United States includes only the oral treatment of E. coli and Enterococcus faecalis, but internationally the IV formulation is used broadly for all Enterobacterales (6) . However, clinical outcomes data for oral treatment have demonstra ted mixed results for treatment of non-E. coli MDR infections and these retrospective analyses are often difficult to interpret due to the absence of follow-up cultures or clinical outcomes (1) (2) (3) (4) (7) (8) (9) (10) (11) . These retrospective studies of patients with K. pneumoniae show that despite clinical microbiology testing showing in vitro susceptibility to fosfomycinbased on extrapolated breakpoints-treatment was successful less than half of the time (1, 4, 7, 8) . These data highlight the discordance between the susceptibility testing results and clinical outcomes and whether current testing methods and interpretation protocols could be contributing factors. Fosfomycin susceptibility testing recommendations are complex. Agar dilution (AD) is the reference method for both the Clinical and Laboratory Standards Institute (CLSI) and the European Committee on Antimicrobial Susceptibility Testing (EUCAST) (6, 12) . Neither CLSI nor EUCAST supports the use of broth microdilution (BMD) for fosfomycin susceptibility testing (6, 13) . The use of BMD for fosfomycin susceptibility testing is not recommended by CLSI due to unsatisfactory precision and skipped wells-occurrence of no-growth in a single well before the minimal inhibitory concentration (MIC)-and trailing endpoints (14, 15) . However, the published data reporting on these phenomena are limited to fosfomycin studies where the only Enterobacterales included was E. coli, or to non-fosfomycin agents against other Enterobacterales (14, 16, 17) . AD is both time-and labor-intensive, and is not a feasible option for many clinical laboratories as a separate testing method for fosfomycin. It is more efficient for clinical microbiology laboratories to utilize BMD or automated systems which typically use a modified BMD method [Vitek 2 (bioMérieux) (18) and Phoenix (Becton Dickinson)] for MIC testing of clinical isolates when they are already in use for most other tested agents. These methods have been published in use globally; Brazil, Israel, Italy, South Africa, Turkey, and Poland utilize BMD/automated systems (19) (20) (21) (22) (23) (24) . Publications utilizing Vitek 2 and Phoenix for determining susceptibility have also extrapolated CLSI breakpoints despite the explicit recommendation against this practice of using BMD for fosfomycin susceptibility testing (12, 14) . However, bioMérieux has successfully added fosfomycin to their Vitek 2 Gram-Negative Susceptibility Card following 510(k) U.S. Food and Drug Administration (FDA) clearance (18) . The use of Vitek 2 in fosfomycin susceptibility testing has a listed indication for "antimicrobial susceptibility testing Gram-negative bacilli" but currently lists E. coli as the only Gram-negative organism for which the FDA recognizes fosfomycin activity (18) . Finally, EUCAST has notably lower MIC breakpoints compared to CLSI's oral fosfomycin breakpoints, with EUCAST lowering the susceptibility breakpoint for IV fosfomycin from 32 to 8 µg/mL in 2024 (Table 1 ). In light of the lack of published data for performance of BMD in fosfomycin susceptibility testing, we sought to study the rates of skipped wells and growth beyond the measured MIC observed in BMD and AD testing among a collection of 160 clinical K. pneumoniae isolates. We also sought to analyze the agreement of BMD to AD MIC.
Showing the abstract — retrieve the full paper via the Exa API.