Knowledge of central venous pressure (CVP) is considered valuable in the assessment and treatment of various states of critical illness and injury. We tested a noninvasive means of determining CVP (NICVP) by monitoring upper arm blood flow changes in response to externally applied circumferential pressure to the upper arm veins. Thirty-six patients who were undergoing CVP monitoring as part of their care had NICVP determined and compared with CVP. Volume changes were measured in the upper arm using tetra-polar impedance plethysmography underneath a blood pressure cuff. The cuff was inflated over 5 s to a pressure greater than CVP but less than diastolic arterial pressure. After 45 to 60 s, the cuff was rapidly deflated. Noninvasive CVP was determined as the cuff pressure noted at the maximum derivative of the volume increase under the cuff during deflation. Noninvasive CVP was then compared with invasively measured CVP taken during the same period by Bland-Altman analysis. A total of 108 trials (three per subject) were performed on 36 patients. Mean bias was -0.26 mmHg (95% confidence interval [CI]: -0.67, 0.15). Limits of agreement were -2.7 and 2.2 mmHg with the 95% CI for the lower limit of agreement (-3.4, -2.0 mmHg) and for the upper limit of agreement (1.5, 2.9 mmHg). Correlation between CVP and NICVP was 0.95 (95% CI: 0.93 to 0.97; P < 0.0001). Noninvasive CVP as determined in this study may be a clinically useful substitute for traditional CVP measurement and may offer a tool for early diagnosis and treatment of acute states in which knowledge of CVP would be helpful.