The Frequency of Extended-Spectrum β-Lactamase Genes Harbored by Enterobacterales Isolates at High Risk for Clinically Significant Chromosomal ampC Expression

To theEditor—Enterobacter cloacae, Klebsiella aerogenes, and Citrobacter freundii compose approximately 20% of gram-negative organisms recovered in blood cultures [1]. These Enterobacterales are at moderate to high risk of clinically significant AmpC production [2, 3]. Therefore, even if they test susceptible to ceftriaxone, guidance suggests avoiding the use of ceftriaxone when these organisms are recovered in clinical cultures because of concerns for clinical failure due to ceftriaxone hydrolysis from excess AmpC production [4–6]. Rather, cefepime therapy is suggested [7]. Cefepime, however, remains a suboptimal choice for infections caused by extended-spectrum β-lactamase (ESBL)–producing Enterobacterales (ESBL-E) [8–10]. It is unclear how commonly E cloacae, K aerogenes, or C freundii clinical isolates from the United States (US) produce ESBL enzymes because ESBL testing is not routinely performed for these organisms. An understanding of ESBL-E gene prevalence among E cloacae, K aerogenes, and C freundii is necessary to determine whether optimally dosed cefepime remains a reasonable treatment option for infections caused by these organisms when cefepime minimum inhibitory concentrations (MICs) are in the susceptible dose-dependent (SDD) range (ie, 4–8 µg/mL).

An observational study from Taiwan demonstrated that of 36 patients with E cloacae bacteremia with a cefepime MIC of 4 or 8 µg/mL [11], 89% had isolates containing an ESBL gene, significantly higher than the 44% prevalence of ESBL production in isolates with cefepime MICs of ≤2 µg/mL (ie, susceptible) [12]. While all patients with non-ESBL-producing E cloacae isolates with cefepime MICs of 4–8 µg/mL treated with cefepime survived, all patients with ESBL-producing E cloacae isolates with cefepime MICs of 4–8 µg/mL treated with cefepime died within 30 days [12]. This observation informed the “Infectious Diseases Society of America Guidance on the Treatment of Antimicrobial-Resistant Gram-Negative Infections” suggestion to avoid the use of cefepime for E cloacae, C freundii, and K aerogenes with cefepime MICs of 4–8 µg/mL due to concerns for co-production of ESBLs at these higher cefepime MICs, with an acknowledgment that data from the US are limited [13].

To investigate this issue in a US cohort, we evaluated all index ceftriaxone-resistant (ie, ceftriaxone MICs ≥4 µg/mL) E cloacae, K aerogenes, and C freundii isolates recovered from blood cultures from unique adult and pediatric patients in 3 Maryland hospitals between January 2018 and July 2021. Testing was limited to ceftriaxone-resistant isolates because ceftriaxone-susceptible isolates are unlikely to harbor ESBL genes [14]. Antimicrobial susceptibility testing was confirmed using lyophilized Sensititre broth microdilution GN7F panels (Thermo Fisher Scientific, Waltham, Massachusetts). Illumina MiSeq short-read sequencing (Illumina, San Diego, California) was performed to evaluate for antimicrobial resistance markers using sequencing and bioinformatics methodology as described elsewhere [15].

A total of 77 isolates were evaluated, with the bacterial species and cefepime MIC distributions displayed in Table 1. The prevalence of ESBL genes stratified by cefepime MIC was as follows: ≤2 µg/mL (3/43 [7%]), 4–8 µg/mL (0/22 [0%]), and ≥16 µg/mL (8/12 [67%]). By species, ESBL genes were most commonly identified in K aerogenes (4/18 [22%]) and E cloacae (7/51 [14%]). No ESBL genes were identified in any C freundii isolates.

In our cohort, ESBL enzymes were uncommon in patients infected with E cloacae, K aerogenes, or C freundii isolates that remain susceptible or SDD to cefepime. This is in contrast to Lee and colleagues [12], whose Taiwanese cohort had a significantly higher prevalence of ESBL genes, highlighting that regional diversity of ESBL genes and their frequency exists. Our findings suggest that optimally dosed cefepime (ie, 2 g every 8 hours in adults with normal renal function, preferably as a 3-hour infusion) may be a reasonable treatment option for E cloacae, K aerogenes, and C freundii isolates, even with cefepime MICs of 4–8 µg/mL, as supported by pharmacokinetic/pharmacodynamic [16, 17] and clinical outcomes [18] data. More specifically, in a pharmacokinetic study investigating 100 critically ill patients receiving cefepime therapy, the median percentage free time (%fT) >2 µg/mL was 100% and %fT >8 µg/mL was 85%, despite 62% of patients with normal renal function receiving suboptimal cefepime dosing [19].

Our previous work indicated that ESBL genes were found in approximately 20% of Escherichia coli, Klebsiella pneumoniae, Klebsiella oxytoca, and Proteus mirabilis clinical isolates collected from patients in our hospital, suggesting that the relatively low prevalence of ESBL genes we identified in E cloacae, K aerogenes, and C freundii isolates with cefepime MIC ≤8 µg/mL is likely not indicative of being in a low-prevalence ESBL region [20]. In summary, cefepime may continue to be a reasonable carbapenem-sparing agent for organisms at moderate to high risk of AmpC production with cefepime MICs in the SDD range, as indicated by our ESBL gene prevalence data and supported by pharmacokinetic/pharmacodynamic studies. Our findings may not be generalizable to other US regions or other regions of the world. This work needs to be repeated in a larger, more geographically diverse cohort.

Notes

Disclaimer. The content is solely the responsibility of the authors and does not necessarily represent the official view of the funders.

Financial support. This work was funded by The Willowcroft Foundation and the Prevention Epicenters Program of the Centers for Disease Control and Prevention (grant numbers 6 U54CK000617–01-02 and 5 U54CK000617-02-00). D. H. is funded by the National Institutes of Health (grant number T32-AI007291).

References

Author notes