Objective: To investigate the presence and determinants of femoral-radial gradients in mean arterial pressure (MAP) in a critically ill population.
Design: Prospective observational study.
Setting and participants: Critically ill patients who were undergoing simultaneous monitoring of arterial pressure by radial arterial catheterisation and transpulmonary thermodilution (via femoral arterial access) in a Level 3, mixed medical-surgical intensive care unit, December 2007 to May 2008.
Main outcome measures: Level of agreement between simultaneous measurements of MAP via the femoral and radial arteries, determined by Bland-Altman analysis; haemodynamic and demographic factors associated with a MAP gradient, assessed by multiple linear regression.
Results: 131 observations were made in 24 patients. Mean age of patients was 56 (SD, 18) years, and mean APACHE II score was 27 (SD, 8). Overall mean bias between radial and femoral MAP measurements was 4.27 mmHg (limits of agreement, -3.41 to 11.94 mmHg). Fifteen patients (62%) had maximum MAP gradients > 5mmHg, and seven of these (29% of the total) had maximum gradients > 10 mmHg. The largest discrepancy in MAP was 18 mmHg in a patient with septic shock resistant to high-dose catecholamine infusion. Regression analysis failed to identify any statistically significant associations between patient factors and MAP gradient.
Conclusion: A systematic difference in MAP measured at the radial and femoral sites was demonstrated. In some critically ill patients, the femoral artery may be the preferred site for systemic arterial pressure monitoring.