Julio Medina

Detection, treatment, and prevention of carbapenemase-producing Enterobacteriaceae: Recommendations from an International Working Group

Abstract The prevalence of carbapenemase-producing Enterobacteriaceae (CPE) has increased during the past 10 years. Its detection is frequently difficult, because they do not always show a minimum inhibitory concentration (MIC) value for carbapenems in the resistance range. Both broth microdilution and agar dilution methods are more sensitive than disk diffusion method, Etest and automated systems. Studies on antimicrobial treatment are based on a limited number of patients; therefore, the optimal treatment is not well established. Combination therapy with two active drugs appears to be more effective than monotherapy. Combination of a carbapenem with another active agent — preferentially an aminoglycoside or colistin — could lower mortality provided that the MIC is ≤4 mg/l and probably ≤8 mg/l, and is administered in a higher-dose/prolonged-infusion regimen. An aggressive infection control and prevention strategy is recommended, including reinforcement of hand hygiene, using contact precautions and early detection of CPE through use of targeted surveillance.

Ciprofloxacin-Resistant Enterobacteria Harboring the aac(6′)-Ib-cr Variant Isolated from Feces of Inpatients in an Intensive Care Unit in Uruguay

The presence of aac(6′)-Ib-cr is associated with decreased susceptibility to aminoglycosides (kanamycin, amikacin, and tobramycin) and to norfloxacin and ciprofloxacin (9). This allelic variant of aac(6′)-Ib was found to be linked to the extended-spectrum β-lactamase (ESBL) gene blaCTX-M-15 in isolates from many countries (4, 6, 7), while association of aac(6′)-Ib with the blaCTX-M-2 ESBL gene has been widely reported in Uruguay and Argentina (3, 11). In this work we looked for the presence of aac(6′)-Ib and the aac(6′)-Ib-cr variant and their putative ESBL coresistance markers in fecal isolates of enterobacteria resistant to ciprofloxacin and/or ceftazidime from inpatients in an intensive care unit (ICU) in Montevideo, Uruguay. From 1 March to 31 October 2006, 106 patients were admitted to this ICU and followed daily until discharge. Rectal swabs obtained at 1, 4, 7, 10, 13, and 16 days after admission were plated on MacConkey agar plus ceftazidime (4 mg/liter) or ciprofloxacin (2 mg/liter). Enterobacterial isolates were identified by classical methods, including only the first isolate of each bacterial species per patient in this study. Antibiotic resistance profiling, screening, and confirmatory testing for ESBL detection were performed by disk diffusion assay, and results were interpreted following the CLSI guidelines (2). A total of 58/106 patients (55.2%) were colonized with ciprofloxacin- and/or ceftazidime-resistant enterobacteria, and 68 isolates were included in this study. Of these, 48 were resistant to gentamicin and 24 to amikacin (Table ​(Table11). TABLE 1. Main characteristics of the 68 studied isolatesa All aminoglycoside-resistant isolates were screened for aac(6′)-Ib by PCR; amplicons were analyzed by restriction with BstF5I, as described by Park et al. (8). PCR products that were not digested by the enzyme [tentatively assigned to aac(6′)-Ib-cr] were confirmed to contain aac(6′)-Ib-cr by double-strand sequencing. Only two Escherichia coli isolates were positive for aac(6′)-Ib-cr detection. Recalling the observed links between blaCTX-M-15 and aac(6′)Ib-cr (4, 6, 7) and between aac(6′)Ib and blaCTX-M-2, the two aac(6′)-Ib-cr-positive isolates were further analyzed by PCR to detect CTX-M-1 and CTX-M-2 group ESBL genes using previously described primers (3, 5). Both isolates were positive only for CTX-M-1 group genes, identified as blaCTX-M-15 after sequencing. Both isolates were obtained at the time of patient admission into the ICU and showed identical pulsed-field gel electrophoresis patterns (10). Both patients were previously hospitalized before ICU admission, suggesting that this strain could be endemic in the hospital, where it could be horizontally transferred. All the other E. coli isolates yielded different pulsotypes (data not shown) compared with these. PCR assays for the detection of class 1 integrons and ISCR1 elements were performed according to the method of Di Conza et al. (3). Both isolates carried a class 1 integron containing the dfr17 and aadA5 gene cassettes, while ISCR1 elements were not detected. So far we have not been able to transfer these resistance genes, either by transformation or by conjugation. This is the first report of aac(6′)-Ib-cr in Uruguay. In accordance with a previous report (6), blaCTX-M-15 and aac(6′)-Ib-cr do not seem to be associated with class 1 integrons. Demonstration of a link to IS26 as previously reported (1) is pending.