The spatial myocardial blood flow heterogeneity of the normal heart wa
s previously investigated by means of the standard microsphere-defined
regional myocardial blood flow in nonischemic hearts. We determined t
he probability density functions of coronary blood flows in the rabbit
heart at selected macroautoradiographic 20-mu m cross-sections of the
left ventricle in nonischemic as well as infarcted hearts. Macroautor
adiography gave us spatial resolutions of 0.1-0.2 mm. As a tracer we u
sed C-14-iodoantipyrine given into the root of the aorta. We report he
re for the first time a systematic study of the shape of the flow prob
ability density functions during acute regional myocardial necrosis. A
s the hearts became progressively and extensively necrotic, the distri
bution of flows changed its characteristics showing two independent co
mponents. The first component was the peak representing the nonischemi
c regions in the hearts subjected to acute ischemia. The second compon
ent was a monotonically decreasing component associated with very low
flows and necrosis in the severely hypoperfused portion of the hearts.
This monotonically decreasing component became larger as the extent o
f ischemia increased and was well separated from the peak attributable
to the nonischemic regions. We could not demonstrate a leftward shift
of the nonischemic central peak in the ischemic hearts. Our research
shows that in transaxial radionuclide cardiac sections, such as those
that might be obtained and analyzed in clinical SPECT and clinical PET
, variable amounts of myocardial necrosis will result in a composite c
urve of myocardial blood flow heterogeneities. One portion of the curv
e will indicate the distribution of flows in the nonischemic zones. Th
e other portion will vary in magnitude with the extent of ischemia, ex
hibit the shape of a monotonically decreasing curve. Depending upon th
e spatial resolution of the radionuclide imaging technique utilized, a
border zone will exist representing the interface between normally pe
rfused and occluded vascular beds. In our investigation, it was found
that the border zone determined statistically was consistently and sig
nificantly smaller than the border zone determined visually.