Nebulized antibiotics are a promising new treatment option for ventilator-associated pneumonia. However, more evidence of the benefit of this therapy is required.
The Medline, Scopus, EMBASE, Biological Abstracts, CAB Abstracts, Food Science and Technology Abstracts, CENTRAL, Scielo and Lilacs databases were searched to identify randomized controlled trials or matched observational studies that compared nebulized antibiotics with or without intravenous antibiotics to intravenous antibiotics alone for ventilator-associated pneumonia treatment. Two reviewers independently collected data and assessed outcomes and risk of bias. The primary outcome was clinical cure. Secondary outcomes were microbiological cure, ICU and hospital mortality, duration of mechanical ventilation, ICU length of stay and adverse events. A mixed-effect model meta-analysis was performed. Trial sequential analysis was used for the main outcome of interest.
Twelve studies were analyzed, including six randomized controlled trials. For the main outcome analysis, 812 patients were included. Nebulized antibiotics were associated with higher rates of clinical cure (risk ratio (RR) = 1.23; 95% confidence interval (CI), 1.05 to 1.43; I2 = 34%; D2 = 45%). Nebulized antibiotics were not associated with microbiological cure (RR = 1.24; 95% CI, 0.95 to 1.62; I2 = 62.5), mortality (RR = 0.90; CI 95%, 0.76 to 1.08; I2 = 0%), duration of mechanical ventilation (standardized mean difference = −0.10 days; 95% CI, −1.22 to 1.00; I2 = 96.5%), ICU length of stay (standardized mean difference = 0.14 days; 95% CI, −0.46 to 0.73; I2 = 89.2%) or renal toxicity (RR = 1.05; 95% CI, 0.70 to 1.57; I2 = 15.6%). Regarding the primary outcome, the number of patients included was below the information size required for a definitive conclusion by trial sequential analysis; therefore, our results regarding this parameter are inconclusive.
Nebulized antibiotics seem to be associated with higher rates of clinical cure in the treatment of ventilator-associated pneumonia. However, the apparent benefit in the clinical cure rate observed by traditional meta-analysis does not persist after trial sequential analysis. Additional high-quality studies on this subject are highly warranted.
Trial registration number
CRD42014009116. Registered 29 March 2014
Electronic supplementary material
The online version of this article (doi:10.1186/s13054-015-0868-y) contains supplementary material, which is available to authorized users.