Enterococcus spp. is one of the most common opportunistic pathogens caused nosocomial infection after S. aureus in all gram-positive bacteria (13). It accounted for 8.42% (16043/190610) among all clinical isolates and ranked seventh, according to the results from the China antimicrobial resistance surveillance network (CHINET) in 2017. Severe infection caused by the Enterococcus would be a danger to the patient's life, and combination therapies of penicillin or ampicillin and aminoglycoside are usually used for the treatment of Enterococcus related infections (14). In addition, studies also shown that imipenem is certain clinically effective against Enterococcus infection (15,16). The determination standard of imipenem susceptibility testing against the Enterococcus is absence at present, and CLSI only mentioned that the susceptibility result of ampicillin can be used to predict the susceptibility to imipenem of E. faecalis. According to study, whether ampicillin susceptibility can accurately predict imipenem susceptibility of E. faecalis is associated with the susceptibility result of penicillin.
Grouped based on penicillin susceptibility results, for E. faecalis with opposite susceptibility results of penicillin and ampicillin, the CA, VME and ME of using ampicillin susceptibility to predict imipenem susceptibility was different. For example, for penicillin-resistant, ampicillin-susceptible E. faecalis, the CA of using ampicillin susceptibility to predict imipenem susceptibility tested by broth microdilution and disk dilution was 57.1% and 81.8%, respectively, and imipenem inter-mediate isolates accounted for 42.9% and 18.2%. For penicillin-resistant, ampicillin-resistant or penicillin-susceptible, ampicillin-susceptible E. faecalis, ampicillin susceptibility of broth microdilution could both accurately predict imipenem susceptibility (CA/VME/ME=100%/0%/0%). Conceicao N et al. also reported that E. faecalis isolates susceptible to both penicillin and ampicillin were also susceptible to imipemem tested by broth microdilution, and it was consistent with our study (8). It should be noted that for E. faecalis resistant to both penicillin and ampicillin tested by disc diffusion, ampicillin susceptibility could not be used to predict imipenem susceptibility (the CA and ME of ampicillin-impenem prediction was 77.8% and 22.2%). In recent years, penicillin-resistant, ampicillin-susceptible E. faecalis has been reported frequently (17,18). As Metzidie et al. reported, the detection rate was 31.4%, higher than that in our study (11% tested by broth microdilution and 8.7% tested by disc diffusion), and most strains was resistant to imipenem (18). Weinstein MP et al. found that the CA of using penicillin susceptibility to predict imipenem susceptibility tested by broth microdilution was 95.2%; and the CA of using ampicillin susceptibility to predict imipenem susceptibility tested by disk diffusion was 99.8% (ME=0.2%) (7). The result ignores penicillin-resistant, ampicillin-susceptibile E. faecalis, leading us to mistakenly believe that ampicillin susceptibility can predict imipenem susceptibility of all E. faecalis with different susceptibility phenotypes.
In our study, the penicillin-resistant, ampicillin-susceptible E. faecium strains are rare (1/124), and for penicillin-susceptible, ampicillin-susceptible or penicillin-resistant, ampicillin-resistant E. faecium, the CA of using ampicillin susceptibility to predict imipenem susceptibility of was both 100%. And for all E. faecium, it was 99.2% tested by either broth microdilution or disk diffusion, which was both 98% in the study of Weinstein MP et al. Based on this, we believe that ampicillin susceptibility can predict imipenem susceptibility of E. faecalis, except for the rare penicillin-resistant, ampicillin-susceptible isolates.
This study evaluated the feasibility of using penicillin and ampicillin susceptibility to predict imipenem susceptibility of E. faecalis and E. faecium. In conclusion, the susceptibility result of ampicillin is superior to that of penicillin to predict imipenem in vitro activity for E. faecalis. And susceptibility of both of them can be used as a predictor of imipenem in vitro activity for E. faecium. However, it should be emphasized that ampicillin susceptibility cannot be used to predict imipenem susceptibility of E. faecalis and E. faecium with penicillin-resistant, ampicillin-susceptible susceptibility phenotypes.