Mortality, attributable mortality, and clinical events as end points for clinical trials of ventilator-associated pneumonia and hospital-acquired pneumonia

Abstract

Appropriate end points are crucial for the successful interpretation of clinical trials. Choosing end points for therapeutic trials of ventilator-associated pneumonia (VAP) and hospital-acquired pneumonia (HAP) requires careful consideration, because they are complications of critical illness. It may be difficult to distinguish the consequences of VAP and HAP from manifestations of the underlying illnesses, and it is important to determine their incremental magnitude, to plan for possible treatment effects and, thus, sample size calculations. In this article, we discuss mortality, attributable mortality, and time to clinical events as possible end points for HAP and/or VAP trials. Because of the paucity of evidence on HAP, we focus predominantly on VAP. In a systematic review of applicable trials, VAP appears to have slight intensive care unit and low hospital-attributable mortality. VAP is associated with prolonged durations of intensive care unit stay, hospital stay, and mechanical ventilation. Because of these findings, superiority trials of VAP treatment that use mortality as a primary end point are not possible. Equivalency studies are possible, but there are sample size implications. The use of time to clinical event end points, especially when combined with mortality, may be the best option for trial in the future.