Usefulness of an early detection system to identify discharged patients carrying multidrug-resistant microorganisms who re-enter the hospital

Background:

Specific detection programs are necessary to control the emergence of multidrug-resistant (MDR) microorganisms in hospitals. Since May 2007, our Admissions Service has provided the Preventive Medicine Department with a daily report of readmitted patients who previously carried MDR microorganisms. Data have been collected since January 2008 and stored in a computer program (VINI) for the surveillance of nosocomial infection by MDR microorganisms. Consequently, these patients are defined as high-risk, and undergo subsequent monitoring through sampling.

Objective:

To evaluate the results of a screening program for discharged patients carrying MDR microorganisms who re-enter the hospital and to determine the probability of decolonization.

Methods:

A retrospective observational study involving patients registered by VINI was conducted at the Hospital Gregorio Marañón (Madrid, Spain). Analysis was by Kaplan-Meier survival curves, with the event defined as the negative status of the carrier (ie, a negative sample: nasal for methicillin-resistant Staphylococcus aureus [MRSA], rectal for ESBL-producing Enterobacteriaceae, and underarm, groin, or rectal for MDR Acinetobacter baumannii).

Results:

We detected 210 readmissions in whom 34.8% were still colonized. The distribution of patients who remained positive was as follows: methicillin-resistant Staphylococcus aureus (MRSA), 42.2% (49/116); ESBL-producing Enterobacteriaceae, 28% (21/75); MDR Acinetobacter baumannii, 15.8% (3/19); and vancomycin-resistant Enterococcus, 0% (0/0). The average number of samples taken until a positive result was obtained was as follows: 1.1 for MRSA, 1.1 for ESBL-producing Enterobacteriaceae, and 4 for Acinetobacter baumannii. The differences between decolonization of MRSA and the other microorganisms were significant: 133 days (95% CI, 44.2-221.8) and 82 days (95% CI, 35.9-128.1), respectively.

Conclusions:

One sample is enough to determine whether a patient is colonized with MRSA or ESBL-producing Enterobacteriaceae Three are necessary for MDR Acinetobacter baumannii. Therefore, the system is useful. The percentage of patients colonized by MRSA is higher than for other MDR microorganisms, and the time to decolonization is longer than for other microorganisms. Our data can contribute to the design of new strategies for the prevention and control of nosocomial infection.