Colistin resistance in gram-negative bacteria during prophylactic topical colistin use in intensive care units

Topical use of colistin as part of selective digestive decontamination (SDD) and selective oropharyngeal decontamination (SOD) has been associated with improved patient outcome in intensive care units (ICU), yet little is known about the risks of colistin resistance. We quantified effects of selective decontamination on acquisition of colistin-resistant gram-negative bacteria (GNB) using data from a cluster-randomized study and a single-centre cohort. Acquisition of colistin-resistant GNB and conversion from susceptible to resistance in GNB was determined in respiratory samples [from patients receiving SDD (n = 455), SOD (n = 476), or standard care (SC) (n = 315)], and in rectal swabs from 1,840 SDD-patients. Genotyping of converting isolates was performed where possible. The respiratory tract acquisition rates of colistin-resistant GNB were comparable during SDD, SOD, and SC and ranged from 0.7 to 1.1/1,000 patient-days at risk. Rectal acquisition rates during SDD were < 3.3/1,000 days at risk. In patients with respiratory tract GNB carriage, conversion rates were 3.6 and 1.1/1,000 patient-days at risk during SDD and SC, respectively, (p > 0.05). In patients with rectal GNB carriage conversion rates during SDD were 5.4 and 3.2/1,000 days at risk and 15.5 and 12.6/1,000 days at risk when colonized with tobramycin-resistant GNB. Acquisition rates with colistin-resistant GNB in the respiratory tract were low and comparable with and without topical use of colistin. Rates of acquisition of colistin-resistant GNB during SDD were-in ICUs with low endemicity of antibiotic resistance-< 2.5/1,000 days at risk, but were fivefold higher during persistent GNB colonization and 15-fold higher during carriage with tobramycin-resistant GNB.