Marin H. Kollef, MD; Jean-Damien Ricard, MD; Damien Roux, MD; Bruno Francois, MD; Eleni Ischaki, MD; Zsolt Rozgonyi, MD; Thierry Boulain, MD; Zsolt Ivanyi, MD; Gál János, MD; Denis Garot, MD; Firas Koura, MD; Epaminondas Zakynthinos, MD; George Dimopoulos, MD; Antonio Torres, MD; Wayne Danker, MD; A. Bruce Montgomery, MD CHEST Jun 2017; 151(6): 1239-1246
Background: Clinical failures in ventilator-associated pneumonia (VAP) caused by gram-negative bacteria are common and associated with substantial morbidity, mortality, and resource utilization.
Methods: We assessed the safety and efficacy of the amikacin fosfomycin inhalation system (AFIS) for the treatment of gram-negative bacterial VAP in a randomized double-blind, placebo-controlled, parallel group, phase 2 study between May 2013 and March 2016. We compared standard of care in each arm plus 300 mg amikacin/120 mg fosfomycin or placebo (saline), delivered by aerosol twice daily for 10 days (or to extubation if < 10 days) via the investigational eFlow Inline System (PARI GmbH). The primary efficacy end point was change from baseline in the Clinical Pulmonary Infection Score (CPIS) during the randomized course of AFIS/placebo, using the subset of patients with microbiologically proven baseline infections with gram-negative bacteria.
Results: There were 143 patients randomized: 71 to the AFIS group, and 72 to the placebo group. Comparison of CPIS change from baseline between treatment groups was not different (P = .70). The secondary hierarchical end point of no mortality and clinical cure at day 14 or earlier was also not significant (P = .68) nor was the hierarchical end point of no mortality and ventilator-free days (P = .06). The number of deaths in the AFIS group was 17 (24%) and 12 (17%) in the placebo group (P = .32). The AFIS group had significantly fewer positive tracheal cultures on days 3 and 7 than placebo.
Conclusions: In this trial of adjunctive aerosol therapy compared with standard of care IV antibiotics in patients with gram-negative VAP, the AFIS was ineffective in improving clinical outcomes despite reducing bacterial burden.