Search

Molecular typing was used to investigate an outbreak of infection caused by multidrug-resistant Enterobacter aerogenes (MREA) susceptible only to gentamicin and imipenem in an intensive care unit (ICU). Over a 9-month period, ciprofloxacin-resistant E. aerogenes isolates were isolated from 34 patients, or 4.1% of ICU admissions, compared with a baseline rate of 0.1% in the previous period (P < 0.001). Infection developed in 15 (44%) patients. In vivo emergence of imipenem resistance (MIC, 32 micrograms/ml) of organisms causing deep-seated infection was observed in two (13%) of these patients following prolonged therapy with imipenem and gentamicin. Arbitrarily primed PCR (AP-PCR) analysis with ERIC1R and ERIC2 primers and pulsed-field gel electrophoresis (PFGE) analysis of XbaI macrorestriction patterns concordantly showed that outbreak-associated MREA isolates were clonally related and distinct from epidemiologically unrelated strains. AP-PCR and PFGE showed discrimination indices of 0.88 and 0.98, respectively. Space-time clustering of cases within units suggests that the epidemic-related MREA isolates were transmitted on the hands of the health care personnel. A case-control study and repeated environmental culture surveys failed to identify a common source or procedure associated with transmission. In spite of the early implementation of isolation measures, the incidence of MREA colonization remained stable until all colonized patients were discharged. This study confirms the usefulness of AP-PCR and PFGE analyses for the epidemiological study of E. aerogenes and underscores the difficulty of controlling the spread of multiresistant clones of this organism in the ICU setting. The emergence of imipenem resistance represents a threat because virtually no therapeutic option is available for such strains.

Purpose: An outbreak of gram-negative bacteremia in patients undergoing endoscopic retrograde cholangiopancreatography (ERCP) was investigated to determine the sources of infection and to control transmission., Patients, Methods, and Results: The incidence of post-ERCP bacteremia increased from 1.6% (60 of 3,696) procedures to 3.6% (53 of 1,454) procedures (relative risk 2.3, p < 0.0001) after endoscopes were processed in a new automated disinfector. Bacteremia involved nine species of Pseudomonas and Enterobacteriaceae, which were also isolated from processed endoscopes. Seven epidemic strains with highly related genomic macrorestriction profiles each infected 2 or more patients, accounting for 29 (55%) episodes of post-ERCP bacteremia. Strains recovered from endoscopes and from the disinfector were associated with 22 (42%) and 5 (9%) bacteremic episodes respectively. Effective endoscope disinfection was achieved by cleansing and disinfection of a blind channel not processed in the disinfector, additional isopropanol-air flush of all channels, and auto-disinfection of the disinfector. In the following period, the incidence of post-ERCP bacteremia returned to the pre-epidemic rate (1.7%, p = 0.0001)., Conclusion: Bacterial genome fingerprinting by macrorestriction analysis enabled delineation of a multi-strain outbreak of post-ERCP bacteremia. Cross-contamination, and to a lesser extent, common-source contamination, appeared related to inadequate disinfection of endoscopes processed in an automated disinfector.

An epidemic involving 10 patients who developed a urinary tract infection within a few days after cystometry and/or cystoscopy is reported. A pure culture of Serratia liquefaciens was obtained from the fluid inside the disposable dome of the cystometer. The outbreak ceased when the dome was systematically changed after each examination as required.