Abstract： Objective To explore the effect of different positive end expiratory pressures (PEEP) on oxygen delivery (DO2) after recruitment maneuvers (RM) in dogs with acute respiratory distress syndrome (ARDS). Method After ARDS models established in 15 dogs by oleic acid, static P-V (pressure-volume) curves were determined by low flow technique. Lower inflection point (LIP) was set by two-way linear regression methods. RM was operated with the pressure control method. ARDS dog models were randomly divided into three groups, namely PEEP 8 cmH2O group (group A), 12 cmH2O group (group B) and 16cmH2O group (group C) after RM (equivalent to pressure at4 cmH2O under LIP, 4 cmH2O near LIP, and 4 cmH2O above LIP, respectively). Hemodynamics and arterial blood gas analysis were monitored before RM, 0, 5, 10 and 15 min after RM. The recruited volume was measured by P-V curve method 15 min after RM and respiratory mechanics was also observed at the same time. Then DO2 was calculated. The quantitative variables were summarized as the mean and SD. The t-test was used to compare continuous variables between the two independent samples. One-way analysis of variance was used to compare variables among three groups. The level of significance was set at P＜0.05 for all the tests. Results In group A, the levels of PaO2 were significantly reduced 5 min, 10 min and 15 min after RM[(257 ± 23 )mmHg, (253±21)mmHg, and (255±19)mmHg] compared with PaO2 at 0 min [(322 ± 20) mmHg] (P＜0.05).But in group B and group C, the levels of PaO2 5 min, 10 min and 15 min after RM were not lower than level of PaO2 at 0 min after RM (P＞0.05 ). The levels of PaO2 in groups B and C were higher than that in group A at the same time (P＜0.05). The recruited volume distinctly increased with PEEP levels escalated [(50±12 ) mL, (124 ±15) mL, and ( 157 ±10)mL](P＜0.05). However, the increment in the recruited volume from PEEP 8 cmH2O to 12 cmH2O was dramatically greater than that from PEEP 12 cmH2O to 16 cmH2O.There was no significant difference in static compliance between group A and B [(14.3 ± 2.2) mL/cmH2O vs. (17.2±1.4)mL/cmH2O] (P ＞ 0.05 ). But compared with groups A and B, the static compliance in the group C significantly reduced(10.5 ± 0.9) mL/cmH2O ( P ＜ 0.05 ). The ratios of DO2 after RM to DO2 before RM were different at different levels of PEEP. The levels of DO2 after RM[( 1.15 ± 0. 11 ),( 1. 14 ± 0.12), ( 1.14 ± 0. 12) and ( 1.16 ± 0.11 )] increased more greatly than that before RM ( 1.00 ±0.09) in the group B (P ＜ 0.05 ). It did not occurred in the groups A and C. Conclusions The PEEP 12 cmH2O set at near the LIP after RM could be the optimal PEEP. Not only can it improve DO2 and the static compliance, but also maintain oxygenation and the recruited volume after RM.