W. S. TOPHAM 
1235 
duced to an absolute level of cm/H20. To 
achieve this accuracy, a nev^ method v^^as devel- 
oped in our laboratory, and the results have 
been reported. ^^■^'^ 
A Statham SF-1 transducer v^^as selected be- 
cause it had a lumen in the catheter for external 
pressure measurement. This transducer was se- 
lected even though it v^as more sensitive to tem- 
perature than other transducers which are cur- 
rently available because vivo calibration was 
essential without the insertion of other cathet- 
ers. 
A plot of temperature versus voltage output 
was taken at a zero pressure, and it was found 
that there was a linear relationship. The sensi- 
tivity was 7 cm/H20 per degree Centigrade 
change. Since it was assumed that the tempera- 
ture in the exercising dog would be expected to 
vary at least one degree, this variation had to 
be compensated. This was done by attaching a 
thermistor to the tip of the catheter, as is 
shown in Figure 3. With the thermistor giving 
the actual temperature at the catheter tip, the 
correction could then be made in any drift due 
to blood temperature change at the time the 
data was analyzed. 
To check for the second source of zero insta- 
bility, due to internal heating, the catheter was 
submerged in a saline bath and the saline was 
stirred to provide a flow motion past the tip of 
the catheter. With the temperature maintained 
at a constant value, the stirrer was turned on 
and off, and it was found that there was a shift 
in the pressure reading of 5 to 10 cm/HoO. This 
sensitivity of flow was expected because at the 
suggested power input, the metal tip of the 
catheter would get warm to the touch if left in 
air. The obvious way to eliminate this problem 
is to reduce the amount of power input to the 
transducer; however, reductions of the excita- 
tion voltage by a factor of five or ten, which 
would be necessary to effectively eliminate the 
drift, would provide a pressure transducer sen- 
sitivity of 1 to 2 microvolts per cm/H20. This 
low voltage output would negate any increase in 
accuracy because of the noise problem in the 
amplifier. Another way to reduce the power 
input is to apply the same voltage level, but re- 
Figure 3. — A Statham SP-1 catheter with a thermistor attached for temperature measurement of surrounding 
fluid. The thermistor provides a signal for correction in temperature drift of zero pressure. 
