OBJECTIVE To investigate the effect of posture and different positive end expiratory pressure (PEEP) on central venous pressure (CVP) in critically ill patients with mechanical ventilation (MV). METHODS Twenty-three critically ill patients were enrolled in this study with MV and CVP measurement in intensive care unit (ICU) from December 2005 to March 2006. CVP, heart rate (HR), mean arterial pressure (MAP), oxygen saturation of pulse (SpO(2)), and Riker's sedation-agitation score (SAS) were obtained simultaneously with a backrest elevation of 30 degree angle or with a supine flat position under different PEEP conditions of 0, 3, 6, 9, 12 and 15 cm H(2)O (1 cm H(2)O=0.098 kPa). RESULTS There were no differences in CVP, HR and MAP between the two positions at the same PEEP condition (all P>0.05). CVP increased gradually by the increment of PEEP with patients in both positions (all P<0.05), and significantly higher than other PEEP conditions when PEEP was 15 cm H(2)O. SpO(2) decreased and SAS increased when the position was adjusted from 30 degree angle to supination (both P<0.01), and when the mechanical ventilation was weaned, SpO(2) was the lowest than those at different PEEP conditions with patients in either position (all P<0.05). SAS decreased gradually by the increment of PEEP except at 15 cm H(2)O (all P<0.05). There were no differences in HR and MAP between different positions and in different PEEP conditions. CONCLUSIONS CVP was not influenced by the posture. CVP increased gradually with the increase in PEEP. Changing posture and weaning of mechanical ventilation led to deterioration of oxygenation and increased agitation. Maintaining of posture and PEEP when measuring CVP could improve safety and comfort of patients, and reduce the work of nursing.