Ampicillin-Sensitive, Imipenem-Resistant Strains of Enterococcus faecium

Weinstein has in an interesting paper (3) suggested that testing the susceptibilities of isolates of enterococci to penicillin or ampicillin accurately predicted the in vitro activity of imipenem, and also pointed out that there are no NCCLS guidelines for testing the susceptibility of enterococci to imipenem. However, we found three strains of Enterococcus faecium that were sensitive to ampicillin but resistant to imipenem in blood and an abdominal abscess from an elderly patient being treated with imipenem at an intensive care unit (ICU) in a Swedish hospital. MICs of ampicillin were from 0.25 to 1 μg/ml, and those of imipenem were 4 to 16 μg/ml. Resistance to imipenem in these strains was caused by increased production of PBP5 with decreased affinity to imipenem (1). Similar strains have been isolated in Switzerland from the blood of eight hospitalized patients (V. Brandt, A. Wenger, and J. Bille, 10th Eur. Congr. Clin. Microbiol. Infect. Dis., poster WeP14, 2000), and studies on their resistance mechanisms are in progress. MICs of ampicillin ranged from 0.5 to 6 μg/ml, while MICs of carbapenems (imipenem and meropenem) were ≥16 μg/ml. Interestingly, six of the strains were benzyl penicillin resistant by NCCLS standards, making benzyl penicillin a better, though not perfect, indicator of decreased susceptibility to carbapenems. In a recent investigation, hospitals (2), ampicillin-sensitive, imipenem-resistant strains were found in ICUs in Swedish hospitals (2), but since breakpoints for imipenem have not been defined by NCCLS, it is hard to estimate how frequently, if at all, these strains occur outside ICUs. In conclusion, we feel that sensitivity testing with carbapenem should be done whenever treatment of enterococcal infections with this type of drug is considered.