A model of hypothermic circulatory arrest has been developed in newbor
n dogs which simulates the procedure used for the operative repair of
congenital cardiac defects in human infants. Hypothermic circulatory a
rrest for 1.0 h causes no brain damage, whereas cardiac arrest for 1.7
5 h results in damage of the cerebral cortex, basal ganglia and to a l
esser extent the claustrum and amygdaloid nucleus. In the present stud
y, we determined regional cerebral blood flow (rCBF) during 24 h of re
covery from hypothermic circulatory arrest. Newborn nitrous oxide anes
thetized and artificially ventilated dogs were cooled to 20-degrees-C
and subjected to cardiac arrest by the i.v. injection of KCl for eithe
r 1.0 or 1.75 h. Thereafter, animals were resuscitated, rewarmed to 37
-degrees-C, and rCBF measured with [C-14]iodoantipyrine at either 2 or
18 h of recovery. Control animals were rendered hypothermic to 20-deg
rees-C without cardiac arrest for 1.0 or 1.75 h prior to rewarming. No
alterations in CBF at either 2 or 18 h of recovery were present in an
y of 16 analyzed structures in animals previously subjected to hypothe
rmic circulatory arrest compared to controls rendered hypothermic alon
e. A direct linear correlation existed between mean arterial blood pre
ssure and blood flow within frontal, parietal and occipital cortex, oc
cipital white matter, hypothalamus and cerebellar vermis in puppies ar
rested for 1.75 h and recovered for 2 h, suggesting a loss of CBF auto
regulation at this interval. No such association between blood pressur
e and CBF was apparent at 18 h of recovery. The findings suggest that
maintenance of adequate systemic blood pressure and hence cerebral per
fusion in the early post-operative period are important to prevent or
minimize ischemic brain damage in newborn infants subjected to hypothe
rmic circulatory arrest.