Measurement of Blood Flow in the Femoral Artery in Man at Rest and during Exercise by Local Thermodilution.

IN 1960 we reported a method of measuring blood flow in individual vessels based on the principle of local thermodilution.1 The present communication presents application of the same principle to measurement of blood flow in the femoral artery in man at rest and during exercise. Methods An upstream catheter was used, modified to allow for percutaneous insertion by the Seldinger technic (fig. 1). Tihe catheter had an asymmetrical double lumen, the narrower lumen being occupied by the leads from the thermistor to an external resistance bridge, the large lumen (1.2 mm.) being used to introduce the catheter over a spiral guide and for injection of indicator. The external diameter was 2.5 mm. The tip of the catheter was slightly conically narrowed and slightly curved. The thermistor was located on the concave surface of the curve (to prevent contact with the vessel wall) about 5 to 10 mm. from the tip. The thermistor was slightly raised from the surface of the catheter, resting on its own lead wires in order to limit a thermal influence on it by the intraluminal contents of the catheter. The insulation layer on the thermistor was as thin as possible in order to have a small time constant, 0.3 second or less, measured in cool water. The indicator was 5 ml. of physiologic saline kept at room temperature. The duration of the injection varied from 0.3 to 0.5 second. Just before each injection we fill the catheter with blood by opening the stopcock slightly. This allows us to subtract a precise dead space of the catheter from the known total injectate. Flow is calculated by the arterial formula: m * 60 * r/t,-tt/ * 1.08 F ~~~~ A * f where F =flow in ml./min., m-amount of injectate, corrected for dead space, r = registra-86 tion paper speed in mm./sec., tB-blood temperature in degrees C., and tl-indicator temperature in degrees C., A = area under the thermodilution curve in mm.2, and f = degree of temperature change corresponding to 1 mm. of deflection on the registration paper. Other details have already been described.1 The terminal part of the dilution curve may be distorted by cooling of the thermistor from the intraluminal contents, and this is manifest as a delayed return to baseline (fig. 2). This source of error can be removed by extrapolating the exponential decay of the curve from two points (0 and P) previous …