Carbapenem Susceptibility Discords among Acinetobacter Isolates

SIR— We read with great interest the case report of a fatal Acinetobacter baumannii infection with discordant carbapenem susceptibility by Lesho et al. [1]. This report implies that such cases are uncommon events and that, in fact, it was “the first report of clinically relevant discordant carbapenem susceptibilities” [1, p. 759]. We have 6 years of surveillance experience with the Meropenem Yearly Susceptibility Test Information Collection (MYSTIC) Program (USA) and have consistently documented the greater potency and high susceptibility rates for imipenem, compared with those for meropenem, when Acinetobacter species are being tested [2,3,4–5]. The results in the United States were further validated by reports from other surveillance programs [6] and several MYSTIC Program publications summarizing findings of Acinetobacter species, especially in Europe [7]. In brief, Lesho and colleagues [1] cited MYSTIC Program results in one article [2] that did not compare findings for imipenem and meropenem tested against clonal multidrug-resistant strains of Acinetobacter species, and the second article they cited [3] clearly demonstrated that meropenem was inferior in activity to imipenem (MIC50/90, 1/32 vs. 0.25/16 µ g/mL) and the percentage of strains inhibited at the breakpoint concentration of ⩽ 4 µ g/mL (percentage of isolates susceptible, 84.1% vs. 88.4%) [8, 9]. This “discord” in carbapenem activity against Acinetobacter species, in favor of imipenem, is routinely found to be reversed for Pseudomonas aeruginosa; among P. aeruginosa isolates analyzed in the United States, meropenem has a 2-fold to 4-fold greater potency than imipenem and a percentage susceptibility advantage of ∼4%–5% [3]. Contemporary susceptibility testing standards [8, 9] in the United States categorize imipenem and meropenem together; the 2 drugs appear in the same box in table 1 of the Clinical and Laboratory Standards Institute (previously the NCCLS) document M100-S15 [9]; this table “designates clusters of comparable agents that need not be duplicated in testing, because interpretive results are usually similar and clinical efficacy comparable” [9, p. 94]. However, the word “or” is not present between the carbapenems that “designates a related group of agents that has an identical spectrum of activity and interpretive results, and for which cross-resistance and -susceptibility are nearly complete” [9, p. 94]. This standards document clearly states that although very similar with respect to spectrum and potency against P. aeruginosa and other non-Enterobacteriaceae or Acinetobacter species, findings for imipenem should not be used to predict those for meropenem, or vice-versa [8, 9]. We agree with Lesho et al. [1] that discords between the susceptibilities of carbapenems may be observed when nonfermentative Gram-negative bacilli are being tested, and that the carbapenem selected for therapy should also be tested by the clinical laboratory to guide treatment. However, our experience with carbapenem resistance surveillance worldwide indicates that such discords are not unusual and will occur frequently in the context of contemporary infectious diseases practice. Potential conflicts of interest. All authors: no conflicts.