Introduction: In acute respiratory distress syndrome (ARDS), adequatepositive end-expiratory pressure (PEEP) may decrease ventilator-induced lung injury by minimizing overinflation and cyclic recruitment derecruitment of the lung. We compare setting the PEEP using incremental, decremental titration and a table-based PEEP as regard hemodynamics, respiratory mechanics and gas exchange.Methods: sixty patients with ARDS were randomly assigned to threegroups. Each group contains twenty patients. In group A patients received an increasing PEEP, group B patients received a decreasing PEEP and group C PEEP set according to PEEP/Fio2 tables of the ARDS network trial. Patients were assessed at eight specific times, prior to randomization (baseline), Following PEEP setting and twice daily for 72 hours (12, 24,36,48,60 & 72 hours following PEEP setting).Results: The baseline characteristics did not differ significantly between the three groups. Best PEEP was 11.4±2.11 in group A, 15.3±2.17 in group B and 18.5±3.03 in group C. the group subjected to highest PEEP levels (group C) was not associated with better hemodynamic, respiratory, oxygenation and outcome variables.Conclusions: decremental PEEP titration had better oxygenation andcompliance improvement compared with the incremental PEEP titration and table-based PEEP methods.