The purpose of this study was to evaluate the magnitude of possible error in the measurement of systolic blood pressure if disposable, built-in diaphragm, transducer domes or long extension tubes between the patient and pressure transducer are used. Sinusoidal or arterial pressure patterns were generated with specially designed equipment. With a long extension tube or trapped air bubbles, the resonant frequency of the catheter system was reduced so that the arterial pulse was amplified as it acted on the transducer and, thus, gave an erroneously high systolic pressure measurement. The authors found this error to be as much as 20 mm Hg. Trapped air bubbles, not stopcocks or connections, per se, lead to poor fidelity. The utility of a continuous catheter flush system (Sorenson, Intraflow) to estimate the resonant frequency and degree of damping of a catheter-transducer system is described, as are possibly erroneous conclusions. Given a rough estimate of the resonant frequency of a catheter-transducer system and the magnitude of overshoot in response to a pulse, the authors present a table to predict the magnitude of probable error. These studies confirm the variability and unreliability of static calibration that may occur using some safety diaphragm domes and show that the system frequency response is decreased if air bubbles are trapped between the diaphragms. The authors conclude that regular procedures should be established to evaluate the accuracy of the pressure measuring systems in use, the transducer should be placed as close to the patient as possible, the air bubbles should be assiduously eliminated from the system.