Surveillance cultures for detection of rectal and lower respiratory tract carriage of colistin-resistant Gram-negative bacilli in intensive care unit patients: comparison of direct plating and pre-enrichment step

Increasing consumption of colistin as treatment for infections with multidrug-resistant (MDR) Gram-negative bacilli (GNB) has been accompanied by increasingly frequent reports of colistin-resistant (ColR) MDR GNB. Higher selective pressure creates a favourable environment that can facilitate the spread of ColR isolates. Monitoring of asymptomatic ColR GNB carriage can give us a better understanding of this emerging healthcare problem, particularly in wards with higher polymyxin selective pressure and prevalence of carbapenem-resistant GNB. Our aim was to assess the ColR GNB colonization rate in intensive care unit (ICU) patients and evaluate the performance of two surveillance protocols using a selective medium. A prospective study included 739 surveillance samples (rectal swabs and tracheal aspirates) from 330 patients that were screened for ColR GNB carriage using SuperPolymyxin medium. Two approaches were used: direct sample plating and overnight pre-enrichment of samples followed by plating. Colistin resistance was confirmed with broth microdilution. ColR isolates were molecularly screened for plasmid-mediated mcr genes. A total of 44/739 samples (45 ColR GNB isolates) were positive for ColR GNB, which included 31/330 (9.4 %) colonized patients; mcr genes were not detected. The direct plating method only identified 17/45 (37.8 %) isolates correctly, whereas the pre-enrichment protocol identified all 45 ColR GNB. The colonization rate among our ICU patients was 9.4 %. Based on our findings, the pre-enrichment step is necessary for the determination of ColR GNB carriage - even though the time to result takes an additional day, fewer than half of ColR GNB carriers were detected using the direct plating protocol.