• Michael S Niederman.
• Department of Medicine, Winthrop-University Hospital, Mineola, New York, USA. mniederman@winthrop.org
• Clin. Infect. Dis. 2010 Aug 1; 51 Suppl 1: S12-7.

AbstractClinical trials of nosocomial pneumonia can include patients with hospital-acquired pneumonia, ventilator-associated pneumonia, and health care-associated pneumonia. All study participants should meet a clinical definition of infection and have some microbiologic confirmation of infection and its etiology. If the trial is to reflect clinical practice and to be practical to conduct, insistence that all patients have bronchoscopic quantitative cultures performed may not be practical. In designing a clinical trial, patients treated in the intensive care unit are the best group to target for study, including only those with severe pneumonia but allowing those with both ventilator-associated pneumonia and hospital-acquired pneumonia to be enrolled. All trials should include a protocol to control for standards of care, including timing of initial therapy, recent antibiotic use, local microbiology patterns, duration of therapy, and the use of a de-escalation therapy strategy. Blinding of a trial may not be required if studying a new agent that is more active against multidrug-resistant pathogens than against currently available comparators. Any new agent should meet a noninferiority end point for 30-day mortality, but if superiority is a goal of trial design, end points could be microbiologic eradication, time to microbiologic eradication, prolonged duration of therapy, need to modify initial therapy, and serial evaluation of the arterial oxygen tension to fractional inspired oxygen ratio.

14 keywords...

hide…