Pneumatic compression devices, used as part of the therapeutic strategy for lymphatic drainage, often have cuffs with multiple chambers that are inflated sequentially. The purpose of this study was to investigate (1) the relationship between cuff chamber pressure (P(chamber)) and the pressure on the cuff-skin interface (P(interface)) and (2) the mechanical interaction of cuff chambers and consequences for device control. In this study, we used 3 cylindrical (60-, 80-, and 100-mm-diameter) model limbs and 1 ellipsoidal model of the arm to test a commercially available pressure controller using "target pressures," indicated by the controller, of 30, 60, 80, and 100 mm Hg. We studied the time course of P(chamber) and P(interface) during the inflation sequence and the effect of local curvature on P(interface). Our data indicated that, overall, P(interface) is of the same order of magnitude as P(chamber). There was some effect of model diameter and shape, with the smaller curvatures yielding the highest P(interface). Cuff chamber interaction led to P(chamber) and P(interface) values in the most distal (first inflated) chamber that were up to 80% higher than the target pressure. For the 80-mm cylindrical model, for instance, pressure in this chamber reached 54, 98, 121, and 141 mm Hg, respectively, instead of the 30, 60, 80, and 100 mm Hg indicated by the controller. The discrepancy between the target pressure, indicated by the controller, and the pressure measured inside the cuff chambers undermines the therapeutic control and efficacy of the pneumatic compression devices. Because the measured pressures were far beyond the pressure level indicated by the controller, it is recommended that pneumatic compression devices be used at much lower target pressures (<30 mm Hg) than those applied in clinical practice.