Acute respiratory distress syndrome: frequency, clinical course, and costs of care.

OBJECTIVE

To define the occurrence rate of acute respiratory distress syndrome (ARDS) using established criteria in a well-defined general patient population, to study the clinical course of ARDS when patients were ventilated using a "lung-protective" strategy, and to define the total costs of care.

DESIGN

A 3-yr (1993 through 1995) retrospective descriptive analysis of all patients with ARDS treated in Kuopio University Hospital.

SETTING

Intensive care unit in the university hospital.

PATIENTS: Fifty-nine patients fulfilled the definition of ARDS: Pao2/Fio2 <200 mm Hg (33.3 kPa) during mechanical ventilation and bilateral infiltrates on chest radiograph.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN

RESULTS

With a patient data management system, the day-by-day data of hemodynamics, ventilation, respiratory mechanics, gas exchange, and organ failures were collected during the period that Pao2/Fio2 ratio was <200 mm Hg (33.3 kPa). The frequency of ARDS was 4.9 cases/100,000 inhabitants/yr. Pneumonia and sepsis were the most common causes of ARDS. Mean age was 43+/-2 yrs. At the time of lowest Pao2/Fio2, the nonsurvivors had lower arterial and venous oxygen saturations and higher arterial lactate than survivors, whereas there were no differences between the groups in other parameters. Multiple organ dysfunction preceded the worst oxygenation in both the survivors and nonsurvivors. The intensive care mortality was 37%; hospital mortality and mortality after a minimum 8 months of follow-up was 42%. The most frequent cause of death was multiple organ failure. The effective costs of intensive care per survivor were US $73,000.

CONCLUSIONS

The outcome of ARDS is unpredictable at the time of onset and also at the time of the worst oxygenation. Keeping the inspiratory pressures low (30-35 cm H2O [2.94 to 3.43 kPa]) reduces the frequency of pneumothorax, and might lower the mortality. Most patients are young, and therefore the costs per saved year of life are low.