LAUSR

W read the article by Bitsori et al on treating urinary tract infections caused by extended spectrum β-lactamase (ESBL)producing Enterobacteriaceae (ESBL-E) with great interest. We would like to share our ideas on this report. We agree with many aspects presented by the authors and the impossibility to prescribe empirical oral treatment in countries with high prevalence of ESBL-E responsible for febrile urinary tract infections (FUTIs). However, we think that 3 points were not or were not sufficiently taken into account. The first is the absolute requirement to spare carbapenems to limit the emergence of carbapenemase-producing Enterobacteriaceae. In our opinion, this crucial point is not sufficiently emphasized. In addition, 3 therapeutic options were not considered: (1) the use of amikacin in monotherapy for treating FUTIs; (2) the opportunity to use nonorthodox cefixime—amoxicillin–clavulanate (AC) combination—as relay of antibiotics administered by parenteral routes for FUTIs with resistance to trimethoprim-sulfamethoxazole and fluoroquinolones; and finally (3) the role of AC in cystitis in young girls. We have known for many years that one of the main factors in the emergence of carbapenemase-producing strains is the use of carbapenems. Because of its frequency, UTI is by far the leading cause of ESBL-related infections and thus the first opportunity to prescribe carbapenems. In general, except for in very young children or those with urinary malformation, FUTI is not a severe disease, and several studies have shown that alternatives to carbapenems are efficacious in this setting. Aminoglycoside monotherapy in a single daily dose is safe and effective in children with acute pyelonephritis, which is not surprising in light of its excellent pharmacokinetics and pharmacodynamics in blood, kidney parenchyma, or urine. Recently, Poey et al reported on >150 FUTIs treated successfully with amikacin in initial monotherapy. At least 80%–90% of cases of ESBL Escherichia coli and Klebsiella pneumoniae remain susceptible to this aminoglycoside, with no significant change in minimal inhibitory concentration (MIC) and similar efficacy as nonESBL strains. Furthermore, in the Madhi et al study of 25 children, FUTIs were successfully treated with amikacin as initial monotherapy. Compared with other penem-saving treatment options, it is the only one that can be administered in a single injection per day, allowing simple outpatient treatment. Finally, the long half-life of aminoglycosides in kidney and urine could allow for short treatment duration. The potential interest of the nonorthodox use of a β-lactamase inhibitor associated with cephalosporins for ESBL-E was suggested since many years. Bingen et al showed that clavulanate greatly reduced the MIC of cefixime for ESBL-E, and for 90% of these strains, the MIC for cefixime was <1 mg/L. Since this article, case reports, and study of a cohort of patients have confirmed that using the non-orthodox cefixime and AC combination could be an alternative relay of treatment in cases of resistance to trimethoprim-sulfamethoxazole or fluoroquinolones. In our opinion, this combination in children is studied as frequently as pivmecillinam or fosfomycin. Finally, for cystitis, EUCAST recently changed the breakpoints for amoxicillin-AC and Enterobacteriaceae responsible for uncomplicated UTIs. The breakpoint for complicated UTI with pyelonephritis, for example, remains 8 mg/L, but for uncomplicated UTI, the breakpoint is now 32 mg/L. For most ESBL-E, amoxicillin-AC MICs are under this concentration, which offers the possibility to treat cystitis.