How to use articles about harm: The relationship between high tidal volumes, ventilating pressures, and ventilator-induced lung injury

Background: Intensivists commonly encounter patients who may be inadvertently harmed by critical care interventions. This article is designed to guide clinicians in the evaluations of an individual article assessing a question of harm, as well as the sum of multiple pieces of evidence. Objectives: To assess the vaidity of a group of articles about the relationship between high tidal volumes and ventilating pressures on ventilator-induced lung injury; to interpret the results of these studies; and to consider whether they apply in practice. Data Sources: Issues of harm are sometimes measured in randomized trials, but are evaluated more often in myriad observational studies. Data Extraction: We use critical appraisal guides for experimental studies (e.g., randomized trials) and observational studies (e.g., cohort studies, case-control studies and case series) that evaluate the potentially harmful exposure of high tidal volumes and ventilating pressures. This involves assessing the validity of the research, then determining the strength of association between the putative harmful exposure and adverse outcomes, These study designs and their interpretation using relative risks and odds ratios are reviewed. Finally, the relevance of this information (or lack thereof) to clinical practice needs to be determined, Data Synthesis: Examining these studies individually and in totality, there appears to be a relationship between high tidal volumes and ventilating pressures, although the strength of inference from this research is limited by design issues and sample sizes. Conclusions: Critically appraising a body of literature is more challenging than evaluating a single study, but often gives a broader view of the available evidence. Future large, rigorous, randomized trials of different approaches to mechanical ventilation will help to advance our understanding and to better inform our practice.