Correlation between central venous pressure and peripheral venous pressure in medical intensive care unit patients

Abstract

Introduction: Central venous pressure (CVP) is a hemodynamic variable commonly used in the intensive care setting to estimate right arterial pressure for evaluation and monitoring a patient's volume status. Risks such as infection, arterial puncture, hematoma, and pneumothorax associated with central venous cannulation can outweigh its benefits. This study was undertaken to determine if peripheral venous pressure (PVP) predicts CVP in medical intensive care unit (ICU) patients. Materials and Methods: This study was conducted on patients admitted to the medical ICU at AlJamhoriya Teaching Hospital in the period from January to September 2009. Sixty-six patients (aged 40–70 years) who were required a central venous line (CVL) were included prospectively in the study. CVP measured through internal jugular vein or subclavian vein by three ways CVL set insertion; and CVL placement was confirmed by chest X-ray. We used the manometers for the measurement of PVP; and 66 paired recordings of CVP and PVP were made. The correlation and Bland-Altman analysis of agreement were performed. Results: The mean (standard deviation [SD]; range) CVP was 11.3778 cmH2O (±5.6; −1.0–27.0); the mean PVP was 15.80 cmH2O (±5.9; 0.0–33.0); offset (bias) of PVP > CVP was 4.42 cmH2O with SD ± 3.62. The correlation of PVP on CVP was r = 0.8059, ( r2 = 0.65), P < 0.0001. The 95% confidence intervals for the bias were 3.5352–5.3133 cmH2O. In the Bland-Altman analysis, lower and upper limits of agreement (95% LOA) were 2.7 (4.43–−7.20) and 11.63 (4.4–7.2) cmH2O. Four out of 66 points were outside the LOA. The dashed zero lies between the LOA. Conclusion: Measurement of PVP from both antecubital area and dorsum of the hand correlated with CVP measurement with acceptable agreement. PVP measurement may be a noninvasive alternative way for estimating CVP.