Background: Despite intensive research, there are no universally accepted clinical definitions for acute lung injury (ALI) or the acute respiratory distress syndrome (ARDS). A recent joint American-European Consensus Conference on ARDS formally defined the difference between ALI and ARDS based on the degree of oxygenation impairment. However, this definition may not reflect the true prevalence, severity and prognosis of these syndromes.

Methods: During a 22-month period, 56 consecutive mechanically ventilated patients who met the American-European Consensus definition for ARDS [arterial oxygen tension/fractional inspired oxygen (PaO(2)/FIO(2) </= 200 mmHg regardless of the level of positive end-expiratory pressure (PEEP), bilateral pulmonary infiltrates, and no evidence of left heart failure] were admitted into the intensive care units (ICU) of the Hospital del Pino, Las Palmas, Spain, and prospectively studied. The diagnosis of ALI and ARDS was made by a PEEP-FIO(2) trial, 24 h after patients met the Consensus inclusion criteria. Patients were classified as having ALI(-24 h) if the PaO(2)/FIO(2) was > 150 mmHg with PEEP = 5 cmH(2)O, and ARDS(-24 h) if the PaO(2) /FIO(2) was </= 150 mmHg with PEEP >/= 5 cmH(2)O.

Results: Overall mortality was 43 % (24 of 56). However, 24 h after inclusion, PaO(2) response to PEEP 5 cmH(2)O allowed the separation of our patients into two different groups: 31 patients met our ALI(-24 h) criteria (PaO(2)/FIO(2) > 150 mmHg) and their mortality was 22.6 %; 25 patients met our ARDS(-24 h) criteria (PaO(2)/FIO(2) </= 150 mmHg) and their mortality was 68 % (p = 0.0016). The differences in the respiratory severity index during the first 24 h of inclusion, PaO(2)/FIO(2) ratio at baseline and at 24 h, maximum plateau airway pressure, maximum level of PEEP, and number of organ system failures during the ICU stay were statistically significant.

Conclusions: Since the use of PEEP in the American-European Consensus criteria for ARDS is not mandatory, that definition does not reflect the true severity of lung damage and outcome. Our data support the need for guidelines based on a specific method of evaluating oxygenation status before the American-European Consensus definition is adopted.