April 2002 / Volume 47 / Number 4 / Page 496
The Long-Term Outcomes of Mechanical Ventilation: What Are They and How Should They Be Used?
IntroductionIn the critical care setting, usually the most important outcome is survival. However, this is not the only important outcome of critical care. There are increasing data that the patient's quality of life and functional status can be affected long after an intensive care unit stay, and some data suggest that mechanical ventilation strategies could influence those outcomes. Critical care clinicians' decisions regarding mechanical ventilation and related treatments such as level of sedation might have more profound and far-reaching residual effects than has been previously recognized. To deliver effective, cost-effective, and patient-centered care, critical-care clinicians must consider outcomes other than survival. These outcomes include such diverse concepts as quality of life, functional status, and neuropsychological function. This review addresses theoretical and practical challenges to measuring and interpreting those other outcomes.
Measurement of Mortality
Measurement of Patient-Assessed Health Outcomes
Quality of Care at the End of Life and Quality of Death
In the past, research concerning the outcomes of mechanical ventilation often focused on physiologic end points such as oxygenation and ventilation, and clinical end points such as extubation rates, intensive care unit (ICU) survival, and hospital survival. Such outcomes are very appropriate for many critical care studies concerning mechanical ventilation, because of the high short-term mortality and life-threatening physiologic abnormalities in those patients. However, there is increasing evidence that mechanical ventilation strategies and related critical care treatments such as sedation may have important effects, not just on short-term physiology and survival, but also on long-term survival and the quality of that survival. For example, recent studies suggest that survivors of acute respiratory distress syndrome (ARDS) have markedly lower quality of life than individuals with comparable critical illness but no ARDS. Furthermore, patients with sepsis may survive to hospital discharge but have ongoing decrements in survival for years, which seems to result from the critical illness rather than the underlying medical problems. Also, recent studies suggest that some mechanical ventilation strategies may cause the generalized inflammation or "biotrauma" that worsens lung injury and produces multiple organ failure. Finally, there is growing evidence that mechanical ventilation strategies affect long-term cognitive function and influence the incidence of post-traumatic stress disorder. This suggests that mechanical ventilation strategies may influence not just organ function and short-term survival, but also long-term survival and quality of life. These studies, in combination, suggest that clinical research on mechanical ventilation should expand its horizons and look at some of the longer-term outcomes of critical care, if we are to maximize survival and the quality of that survival for our patients.
Survival is, in most situations, the most important outcome for critical care research concerning mechanical ventilation, because the mortality rate of critically ill patients is high and the general goal of most critical care therapy is to have the patient survive a critical illness or injury. However, long-term critical care outcomes must not be limited to survival alone. Other outcomes that are important to patients and their families include quality of life, functional status, and freedom from pain and other symptoms. Much has been written about increasing the focus of medical research on such "patient-centered" outcomes, which became increasingly popular in the 1980s and early 1990s in assessing treatments for chronic diseases. The last decade has seen an explosion of the use of those outcomes in critical care research, although the quality of the studies has been variable.
Clinical researchers assessing the outcomes of mechanical ventilation cannot, and should not, measure every outcome in every study, for 2 reasons. First, a study that measured all possible outcomes would be too cumbersome to conduct. Second, not all outcomes are relevant in all studies. The present review describes the long-term and patient-centered outcomes of mechanical ventilation, identifies important issues and pitfalls in measuring those outcomes, and identifies the situations in which those outcomes may be more or less important. The outcomes addressed include mortality, physiologic outcomes, quality of life, functional status, symptoms, cognitive and neuropsychologic outcomes, and the quality of death.