Orally administered colistin leads to colistin-resistant intestinal flora and fails to prevent faecal colonisation with extended-spectrum ß-lactamase-producing enterobacteria in hospitalised newborns.

Colonisation and infection with extended-spectrum ß-lactamase-producing Enterobacteriaceae (ESBL-E) is an emerging problem. The aim of this study was to investigate whether colistin, which is reported to be effective against multiresistant enterobacteria, prevents ESBL-E colonisation in neonates. For prophylaxis of necrotising enterocolitis, oral gentamicin (15 mg/kg/day) is routinely used in all neonates hospitalised at the Neonatal Intensive Care Unit of University Hospital Graz (Austria). During the study period from May 2005 to September 2007, three ESBL-E outbreaks (total duration 18 months) occurred. During these outbreaks, gentamicin was immediately replaced by oral colistin (8 mg/kg/day) in all hospitalised neonates. All neonates colonised with ESBL-E during the study period were retrospectively analysed with regard to the influence of colistin on ESBL-E colonisation. Genetic relatedness of isolates was assessed by repetitive sequence-based polymerase chain reaction (rep-PCR). During the study period, 30 (4.5%) of 667 neonates were colonised with ESBL-E. Twelve of twenty-one patients colonised with Klebsiella pneumoniae (ESBL-Kp) and one of nine patients colonised with Klebsiella oxytoca (ESBL-Ko) had received oral colistin at time of colonisation with ESBL-E. Amongst ESBL-Kp, the rate of colistin resistance was significantly higher in the colistin group (P=0.0075). Four different clones of ESBL-Kp and three different clones of ESBL-Ko were isolated, indicating the occurrence of patient-to-patient transmission. Colistin-resistant as well as colistin-susceptible isolates were detected within the same clones, indicating induction of resistance. At the dosage used, oral colistin did not prevent colonisation with ESBL-E and appeared to select colistin-resistant strains or to induce colistin resistance.

Occurrence and genotyping using automated repetitive-sequence-based PCR of methicillin-resistant Staphylococcus aureus ST398 in Southeast Austria.

In this retrospective study, the occurrence and genetic relatedness of methicillin-resistant Staphylococcus aureus (MRSA) ST398 in Austrian MRSA patients was investigated. From 2002 to 2008, 14 MRSA ST398 were detected. First occurrence of MRSA ST398 was already found in 2004. Spa ribotyping assigned 12 isolates to spa type t011 and 1 each to spa type t034 and spa type t1451. Isolated MRSA ST398 was nontypeable by pulsed-field gel electrophoresis (NT-MRSA) using restriction enzyme SmaI; therefore, genotyping was performed using automated repetitive-sequence-based PCR (rep-PCR) on the DiversiLab system. Rep-PCR results assigned 10 (71%) of the 14 MRSA ST398 into 1 cluster with a similarity >95%; there was 1 cluster consisting of 2 isolates with a similarity >99% and 2 unique MRSA ST398 isolates. In conclusion, MRSA ST398 was continuously detected in Southeast Austria; first in 2004 with up to 5 MRSA ST398 isolates in 2008. Automated rep-PCR proved as a reliable technique in determining genetic relatedness of NT-MRSA ST398 and demonstrates clonal spread of MRSA ST398 in the investigated region.