Comparative outcomes of polymyxins in neurocritical care patients with carbapenem-resistant Gram-negative bacterial pneumonia: a retrospective cohort study.

Objectives: The spread of carbapenem-resistant Gram-negative bacteria (CR-GNB) related to nosocomial infections is an important public health challenge, and polymyxins have become the last line of defence against CR-GNB. In this study, we aimed to compare the efficacy and safety of different polymyxins.

Methods: This retrospective cohort study included neurocritical care patients with CR-GNB pneumonia. The efficacy and safety were compared in original and inverse probability of treatment weighting cohorts. A subgroup analysis was further conducted to explore the impact of augmented renal clearance status at baseline on treatment efficacy.

Results: Of the 331 patients included, 90 received colistin sulphate, 187 received polymyxin B, and 54 received colistin methanesulfonate sodium. Compared with colistin sulphate, colistin methanesulfonate sodium significantly reduced the clinical failure rate on Day 7 (20.8% versus 37.2%, P = 0.034) and 28-day mortality rate (6.2% versus 16.9%, P = 0.013) and improved the microbiological eradication rate on Day 28 (94.8% versus 84.2%, P = 0.013). The colistin methanesulfonate sodium group also had a lower mortality rate (6.2% versus 15.7%, P = 0.005) and clinical failure rate on Day 28 (17.1% versus 27.4%, P = 0.020) than the polymyxin B group. However, colistin methanesulfonate sodium was more nephrotoxic than colistin sulphate and polymyxin B (P = 0.001 and P = 0.004, respectively). Subgroup analysis revealed no statistical difference in clinical failure rate and mortality risk between the three groups.

Conclusions: For CR-GNB pneumonia in neurocritical care patients, intravenous colistin methanesulfonate sodium may provide a useful treatment option, but vigilance is warranted for nephrotoxicity.