The most important predictors related to hospital mortality were the need for invasive ventilation and complications to MV. Adequate metabolic compensation for respiratory acidosis at admittance is associated with better survival. A high APACHE II score and loss of consciousness (low GCS) were independent predictors of a need to intubate patients.

To determine independent risk factors associated with the pathogenesis hypoxic hepatitis (HH) in hypoxemic patients. The prospective study, multivariate analysis, between January 1999 and March 2003 was conducted at University hospital, intensive care unit, Department of Chest Diseases. A total of 445 consecutive patients were included in the study. Patients who were not hypoxemic (PaO(2) > 80 mmHg) were excluded from analysis. The total number of those found eligible for the study was 297. Patients with a >20-fold increase in liver enzyme levels were diagnosed as HH (study group). The other patients were defined as control group. There were no interventions. Mean age was 62, and 191 were males. Twenty-two patients were diagnosed as HH (study group; 7.41%). Cor pulmonale (CP) was 77.3% in the study group and 36.7% in the control group. The need for mechanical ventilation and hospital mortality were 63.6 and 54.5% in the study group and 32 and 28.7% in the control group, respectively. On multivariate analysis, the presence of CP (p = 0.007; OR: 4.418; CI: 1.492-13.083), high-serum BUN level (>37 mg/dl; p = 0.0001; OR: 1.029; CI: 1.015-1.044) and low PaO(2) level (<46 mmHg; p = 0.008; OR: 0.945; CI: 0.906-0.985) were found to be significantly different between the two groups. Patients with severe hypoxemia, CP and higher BUN level have higher risk levels for HH, the most important risk factor being CP.