Evaluation of some new parameters predicting outcome of patients with acute respiratory failure needing invasive mechanical ventilation due to CAP

Many traditional scores such as sequential organ failure assessment score (SOFA score) are used in the ICU to predict outcome and mortality in mechanically ventilated patients with acute respiratory failure. But new simplified parameter now, is used such as oxygen index (OI). Other authors add PEEP to the PaO2/FiO2 ratio called PaFip which is calculated according to the equation = Ln [(PaO2/FiO2)/(PEEP + 12)] (Ln is the natural logarithm) and used it as an outcome predictor. Aim of the work: is to evaluate the values of oxygen index, PaFip and SOFA score as early predictors of outcome of patients with acute respiratory failure caused by severe CAP who underwent invasive mechanical ventilation. Subjects and methods: this study was carried out on 40 patients with acute respiratory failure due to severe community acquired pneumonia (CAP) who were mechanically ventilated and divided into 2 groups, GI: included 18 non-survivors and GII: included 22 survivors. All patients were subjected to full history and clinical examination, plain X-ray chest and culture and sensitivity of tracheal aspirate. 4 hours after the onset of mechanical ventilation: ABG, measurement of mean airway pressure, calculation of oxygen index, measurement of extrinsic PEEP and calculation of PaFip were done. Results: The SOFA score, PaFip, OI, PaO2/FiO2 and PaO2 were related to mortality in patients on mechanical ventilation due to acute respiratory failure caused by severe community acquired pneumonia. The SOFA score was the parameter most closely related to mortality (P = 0.00I) followed by PaFip (P = 0.006) and PaO2/FiO2 (P = 0.008) while PaO2 and OI were the parameters detecting mortality the least (P = 0.020 & 0.025, respectively). Conclusion: PaFip is nearly equal to the SOFA score as an early outcome predictor for mechanically ventilated patients while the value of oxygen index as early outcome predictor is less than the above 2 parameters.