Objective: The severity of sodium sensitivity is quantified or 1) by the difference in mean arterial pressure (MAP) between high- and low-sodium diets (δMAP), or 2) by the sodium-sensitivity index (SSI), i.e. ratio between δMAP and the difference in urinary sodium excretion rates at the end of the two diets. MAP is usually measured with an arm cuff but the use of finger blood pressure monitors is rapidly increasing. Thus, our aim is to evaluate whether finger measures of MAP can be reliably used for assessing sodium sensitivity. Design and method: We enrolled 68 normotensive volunteers who underwent high- and low-sodium diets of 5 days duration. SSI and δMAP were derived from MAP measures taken both at the arm (gold standard) and at the finger (Portapres model-2). First, volunteers were classified as sodium sensitive (SS) or resistant (SR) if δMAP measured at the arm was or not greater than 3 mmHg, and the receiver operator characteristic (ROC) analysis was performed for the SS/SR classification based on finger δMAP. Then volunteers were classified as SS or SR if SSI from arm MAP was or not greater than 20 mmHg/(mol/day), and ROC analysis was performed for the SS/SR classification based on finger SSI. Results: Fourteen individuals were classified as SS on the basis of arm δMAP greater than 3 mmHg. Similarly, 14 individuals were also classified as SS on the basis of arm SSI greater than 20 mmHg/(mol/day). Classifications based on finger measures were substantially different. In particular, finger measures of δMAP performed poorly for the SS/SR classification: the area under the ROC curve (AUC) was 0.65 only; the best threshold for classification was finger δMAP = 2 mmHg, corresponding to sensitivity = 57%, specificity = 67%. Slightly better performances were obtained for finger SSI (see figure), with AUC = 0.71. The best threshold for classification was finger SSI = 23 mmHg/(mol/day), corresponding to sensitivity = 61%, specificity = 72%.Conclusions: The assessment of sodium sensitivity depends strongly on the MAP measurement site, with important discrepancies between brachial and finger measures.