The pressure-flow (P-F) relationship for intravenous infusion systems is nonlinear and may be expressed by the quadratic model P = RLF + RTF2. The flow parameters RL and RT may represent the resistance of laminar and turbulent flow, respectively. In this study pressure and flow were measured, and RL and RT were calculated for several infusion tubings, catheter, and system components. We then developed a technique to identify the relative effect of various devices on achieving the higher flows needed for fluid resuscitation. A typical infusion system was chosen, and the experimentally determined flow parameters RL and RT of its components were used in the quadratic P-F relationship at P = 300 mm Hg. Devices in the infusion system were ranked, using a subtractive algorithm, according to their relative impediment to flow as measured by contribution to the pressure drop. The order of devices removed or replaced, from largest to least pressure drop, was as follows: fluid warmer, 16-gauge catheter, check valve, 14-gauge catheter, standard-bore Y tubing, 12-gauge catheter, and standard-bore stopcock, leaving 10-gauge catheter + wide tubing. Devices with large RT, such as fluid warmers and check valves that produce large pressure drops, should generally be avoided during fluid resuscitation when high flows are needed. A similar ordering of device removal or substitution (largest to least pressure drop) was determined using the traditional but incorrect linear P-F model, P = RF, and the order of devices chosen for elimination was different.(ABSTRACT TRUNCATED AT 250 WORDS)