Incidence and outcome
Reports of the incidence of ARDS in its different grades of severity vary, to some extent due to the lack of precision in the earlier AECC definition. Incidence estimates range 10–58 cases per 100 000 people, depending on geographical location and on the reporting system used. Using data from a prospective multicentre European cohort study that included 6522 patients treated in ICU, the proportion with ALI and ARDS averaged 7.1% of all patients admitted to critical care. This rose to 12.5% when only patients treated for more than 24 hours in ICU were included. Another study reported that patients with ALI represented 4.5% of all those receiving ventilation at the time of admission to intensive care.
In a recent database analysis from a single ICU treating both surgical and medical patients, a decrease in the prevalence of ARDS was reported: when comparing the periods January 1993–February 1996 and January 2006–April 2009, the authors found a prevalence of 2.5% in the first period and 1.7% in the second period. While the length of stay of survivors in the ICU decreased significantly in the second period, from an average of 17 to 13 days, no significant change in mortality was reported. The mortality rates of 52% and 46% in the first and second periods, respectively, are comparable to the actual mortality seen in routine clinical intensive care. However, studies from ARDSnet, an American network that focuses on ARDS, report a decreasing mortality rate over time. In a study reported in 2000, the network found a significantly lower mortality (31% compared with 39.8%) where a ‘protective ventilation’ (i.e. small tidal volume) approach was used compared with conventional mechanical ventilation. In more recent studies, mortality rates close to 20% have been reported. The latter results, however, probably represent the mortality rates of selected patients included in trials, whereas general surveys or databank analyses of mortality in unselected patients range between 27–45%, or even up to 70%, depending on the severity of disease and comorbidity. The most common cause of death is multi-organ failure, and this has not changed over time. The first 7–10 days seem to be the most relevant for determining the ultimate prognosis of ARDS patients. Within this timespan, about 50% of patients are either successfully weaned from the ventilator or have succumbed to the disease. Young patients with ARDS following trauma seem to fare best, with lung function recovering over 6–12 months. Mild abnormalities of respiratory function (obstructive or restrictive spirometric abnormalities or impaired diffusing capacity) may persist in a proportion of patients. Advanced age, pre-existing comorbidities, septic shock and additional organ failure appear to increase mortality.
Overall, the mortality of patients suffering from ARDS remains unacceptably high despite our extensive knowledge of the pathophysiology of the lung injury and the various multicentre studies of treatment reported to date.