[Methicillin-resistant Staphylococcus aureus (MRSA) in an intensive care unit: a one-year survey]

Methicillin-resistant Staphylococcus aureus (MRSA) is frequently isolated in nosocomial outbreaks. In our study, we analysed the occurrence of colonisation and infection in an Intensive Care Unit of our hospital during a 12-month period. We also evaluated the possibility of using automated ribotyping as a molecular method in order to type the isolates. Twice a week a nasal swab and a rectal swab were performed on all patients; from ventilator-assisted patients, a sputum culture was also taken. All the MRSA isolated were identified by using commonly phenotypic procedures and on all isolates susceptibility tests were performed. An automated ribotyping using EcoRI was also done. Out of 292 patients enrolled in the study, 205 were never colonised (group N); among the other 87 who were colonised by MRSA (29.8%), 40 patients (group A) were MRSA carriers at the time of admission, while 47 (group B) were colonised in the ICU. Twenty-seven patients (11 from group A, 15 from group B and 1 from group N) developed 31 infections due to MRSA. Patients from group A exhibited, as a rule, worse clinical conditions than those from the other two groups. For the former group, MRSA infection was frequently systemic (sepsis), while in group B pneumonia was the predominant infection. The prevalence of colonisations in our study was 30%, which is a value comparable to those presented by other authors in similar cases. MRSA colonisation is a necessary condition for subsequent infections in almost all cases, with an average lag of 7 days. Susceptibility tests were non-discriminating among the isolates: all the strains were susceptible to glycopeptides; nearly all of them were resistant to erythromycin, clindamycin, ciprofloxacin and gentamicin. Automated ribotyping allowed us to distinguish 12 different ribogroups, the most frequent of which was composed of 146 isolates. In our study, this molecular method was able to define a possible endemic clone that should be better investigated by using methods with a higher discriminatory power, such as RAPD or PFGE. The method that we employed is highly reliable, easy to perform and not time-consuming. In our opinion, it could be the method of choice in the first screening of high numbers of isolates.