Open AccessPolymyxin for treatment of ventilator-associated pneumonia in a setting of high carbapenem resistance
Author(s) -
Thalita Bento Talizin,
Cláudia Maria Dantas de Maio Carrilho,
Cíntia Magalhães Carvalho Grion,
Lucienne Tibery Queiroz Cardoso,
Marcos Toshiyuki Tanita,
Karine Maria Boll,
Ivanil Aparecida Moro Kauss,
Josiane Festti,
Camila Ribeiro Lopes,
Leticia Maria Alves da Silva,
Isabella Patruceli de Azevedo,
Késia Paes,
Eduardo Alexandrino Medeiros
Publication year - 2020
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0237880
Subject(s) - medicine , polymyxin , acinetobacter baumannii , colistin , pneumonia , septic shock , ventilator associated pneumonia , sofa score , polymyxin b , mortality rate , intensive care medicine , sepsis , incidence (geometry) , carbapenem , tigecycline , antibiotics , microbiology and biotechnology , pseudomonas aeruginosa , biology , genetics , physics , optics , bacteria
Objectives To analyse the use of polymyxins for the treatment of ventilator-associated pneumonia (VAP) at a teaching hospital where carbapenem-resistant gram-negative bacteria are endemic. Patients and methods This was a historical cohort study of patients receiving polymyxins to treat VAP in ICUs at a public university hospital in southern Brazil between January 1, 2017 and January 31, 2018. Results During the study period, 179 cases of VAP were treated with polymyxins. Of the 179 patients, 158 (88.3%) were classified as having chronic critical illness. Death occurred in 145 cases (81.0%). Multivariate analysis showed that the factors independently associated with mortality were the presence of comorbidities ( P <0.001) and the SOFA score of the day of polymyxin prescription ( P <0.001). Being a burn patient was a protective factor for mortality ( P <0.001). Analysis of the 14-day survival probability showed that mortality was higher among the patients who had sepsis or septic shock at the time of polymyxin prescription ( P = 0.028 and P <0.001, respectively). Acinetobacter baumannii was identified as the etiological agent of VAP in 121 cases (67.6%). In our cohort, polymyxin consumption and the incidence density of VAP were quite high. Conclusions In our study, comprised primarily of chronically critically ill patients, there was a high prevalence of VAP caused by multidrug-resistant bacteria, consistent with healthcare-associated infections in low- and middle-income countries. Presence of comorbidities and the SOFA score at the time of polymyxin prescription were predictors of mortality in this cohort. Despite aggressive antimicrobial treatment, mortality was high, stressing the need for antibiotic stewardship.