Preexisting comorbid diseases can be associated with increased mortality in ARF. In a multivariate analysis, Luhr et al.  reported that immunosuppression was associated with mortality in ARF patients. The SOFA database  identified a history of hematologic or chronic renal or liver failure as independent risk factors
Fig. 2. Flow chart of the study and different subgroups . 1 = description of the differences between ARF and non-ARF patients on ICU admission; 2 = study of risk factors for the development of ARF in the ICU; and 3 = study of the risk factors for death in the ARF patients; *outcome was undefined in four ARF patients and in one non-ARF patient. From  with permission
Fig. 2. Flow chart of the study and different subgroups . 1 = description of the differences between ARF and non-ARF patients on ICU admission; 2 = study of risk factors for the development of ARF in the ICU; and 3 = study of the risk factors for death in the ARF patients; *outcome was undefined in four ARF patients and in one non-ARF patient. From  with permission for death from ARF. Chronic liver disease has been associated with mortality from ARDS in several studies [2, 9, 13]. Zilberberg and Epstein  identified organ transplantation, human immunodeficiency virus (HIV) infection, cirrhosis, active malignancy, and sepsis as independent factors for hospital mortality in patients with ALI. Monchi et al.  reported that the length of mechanical ventilation prior to ARDS, cirrhosis, and the occurrence of right ventricular failure were associated with an increased risk of death.
Many investigators have found death from ARDS to be primarily related to the degree of organ dysfunction [24, 29,46]. Doyle et al.  found that multiple organ failure (MOF), liver disease, and sepsis were the main factors contributing to death. Other important prognostic factors include age [28, 29, 47] and the development of acute renal failure . The prognostic value of the degree of hypoxemia is not well established. Luhr et al.  emphasized that the degree of hypoxemia was unimportant in terms of mortality prediction. Likewise, Valta et al.  reported that the PaO2/FiO2 ratio at the onset of ARDS was similar in survivors and nonsur-vivors.
The cause of death in ARDS patients is usually nonrespiratory, i.e., they die with, rather than from ARDS. Montgomery et al.  showed that only 16 % of deaths were due to refractory respiratory failure; early death (within 72 hours) was due to the underlying illness or injury whereas late death (beyond 72 hours) was due to sepsis. Ferring and Vincent  reported similar findings in 129 patients with ARDS of whom 67 (25 %) died: 50 % from sepsis/MOF, 16 % from respiratoryfailure, 15 % from cardiac failure/arrhythmia, 10 % from neurologic failure, and 8 % from other causes. Bersten et al.  reported that respiratory failure was the only cause of death in 9 % of patients with ARDS and contributed to death in just 24 % ofARDS patients. Recently Estenssoro and colleagues  noted in 217 patients with ARDS that MOF was the major cause of death in 88 patients, sepsis in 84, and refractory hypoxemia in 19; 56% of patients had more than one cause of death with 17 of the 19 patients with refractory hypoxemia also having sepsis or MOF.
Was this article helpful?