K. B. LARSON AND D. L. SNYDER 
1159 
2. ZiERLER, K. L. The cardiovascular system. In 
Compartments, Pools, and Spaces in Medical Phys- 
iology, Bergner, P.-E.E. and Lushbaugh, C.C., edi- 
tors, Clearinghouse for Federal Scientific and Tech- 
nical Information, NBS, U.S. Dept. of Commerce, 
Springfield, Va., 1967. 
3. ZiERLER, K. L. Circulation times and the theory of 
indicator-dilution methods for determining blood 
flow and volume. In: Handbook of Physiology, 
Vol. 1, Hamilton, W., and Dow, P., editors, Ameri- 
can Physiological Society, Washington, D.C., 1962. 
4. Stephenson, J. L. Theory of the measurement of 
blood flow by the dilution of an indicator. Bull. 
Math. Biophys. 10:117-121, 1948. 
5. Meier, P., and Zierler, K. L. On the theory of 
the indicator-dilution method for measurement of 
blood flow and volume. J. Appl. Physiol. 
6:731-744, 1954. 
6. Perl, W., Effros, R. M., and Chinard, F. P. 
Indicator equivalence theorem for input rates and 
regional masses in multi-inlet steady-state systems 
with partially labelled input. J. Theor. Biol. 25:297- 
316, 1969. 
7. Larson, K. B., and Snyder, D. L. Measurement 
of relative blood flow, transit-time distributions and 
transport-model parameters by residue detection 
when radiotracer recirculates. J. Theor. Biol. 37: 
503-529, 1972. 
8. Stephenson, J. L. Theory of measurement of blood 
flow by dye dilution technique. IRE. Trans, on Med. 
Electronics PGME-12, 82-88, 1958. 
9. Chinard, F. P., Effros, R., Perl, W., and Silver- 
man, M. Organ vascular and extravascular com- 
partments in vivo. In Compartments, Pools, and 
Spaces in Medical Physiology, Bergner, P.-E.E. and 
Lushbaugh, C.C., editors. Clearinghouse for Fed- 
eral Scientific and Technical Information, NBS, 
U.S. Dept. of Commerce, Springfield, Va., 1967. 
10. Perl, W., and Chinard, F. P. A convection-diff'u- 
sion model of indicator transport through an organ. 
Circ. Res. 22:273-298, 1968. 
DISCUSSION 
Questioner (Unidentified) : Does it matter 
what kind of radioactive isotope you use? Do 
you have one that you use regularly? 
Dr. Larson : It has to be a radioisotope that 
ean be detected externally, which means some- 
thing generally not less than about 80 or 90 
keV for the emitted electromagnetic radiation. 
We have used two radioisotopes in our experi- 
ments: and ^^^Xe. The first is a cyclotron- 
produced positron emitter used to label water; 
the annihilation radiation from the attending 
positron-electron interactions has an energy of 
511 keV. The second is dissolved directly in 
saline solution to form the injectate. The gamma 
radiation from ^^sxe has an energy of 83 keV. 
Marvin Bleiberg, Woodard Research Corp., 
Virginia: In your introduction, you mentioned 
the intercept. Could that be accounted for by 
extending your line down to zero? 
Dr. Larson : Well, this of course is what is 
done very often. An arbitrary extrapolation, an 
exponential extrapolation, is usually made, and 
this method is not subject to too much error if 
recirculation is a fairly late event. If it is an 
early event, the kind of extrapolation that you 
are suggesting is very uncertain. 
Dr. Bleiberg : What are some of the equations 
you use when you're calculating the intercept? 
Dr. Larson : May I write them on the black- 
board, Mr. Chairman? 
Chairman Warner : Yes. 
Dr. Larson: First, let me begin by writing 
the conventional equation for mean transit time. 
This integral [Equation (8)] is the area under 
an arterial washout curve, qa (t) , say. The form- 
ula for mean transit time simply says that you 
find the total area under the arterial curve and 
divide by the amount you injected [Equation 
(8)]. This expression holds only if tracer does 
not recirculate, and is an approximation other- 
wise. When there is recirculation of tracer (and 
normally there always is some, to a greater or 
lesser degree), the arterial residue curve may 
not fall to zero. In fact, if there is no loss of 
tracer at all, it will come down to a nonzero 
steady state as I indicated in one of the slides 
(Figure 3). I assume your question, Dr. Blei- 
berg, refers to the method of calculating the 
area under the residue curve when there are trac- 
er-recirculation effects. The conventional ap- 
proximation used in practice is to plot the falling 
portion of the curve on semilogarithmic paper, 
trying to discern a straight line over some por- 
tion of it. The early portion of the curve is 
presumably free of later recirculation effects, 
and so, hopefully, when extrapolated out along 
the time axis, describes the curve that would 
have been observed after very long times if 
there were no tracer recirculation. To get the 
area, one performs the integration analytically, 
using the time constant obtained from the slope 
of the straight-line segment of the semilogarith- 
mic plot. Unfortunately, small errors in the 
