The sequential changes in systemic and cerebral hemodynamics, systemic and cerebral oxygen transport and consumption rates, and the regional blood flows (measured with 15 micron microspheres) to the cortical and subcortical brain tissues were determined in nine dogs subjected to graded hemorrhage (10 ml/kg X 4 at 15 min intervals). As hemorrhage progressed, both mean arterial pressure and cardiac output decreased progressively. In contrast to the systemic circulation, the mean cerebral blood flow (mCBF) was well maintained by cerebral vasodilation and the cerebral O2 consumption rate (CMRO2) increased during the first three stages of hemorrhage. At 40 ml/kg of hemorrhage, there were significant reduction in mCBF and CMRO2 despite the increase in O2 extraction, suggesting the occurrence of cerebral hypoxia and decompensation of the cerebral circulation. There were remarkable regional variations in the responses of regional cerebral blood flows (rCBF) to hypovolemia, resulting in a significant redistribution of cerebral blood flow. The fractions of cardiac output supplying the diencephalon (thalamus and hypothalamus), the brain stem (pons and medulla oblongata) and the cervical spinal cord increased after hemorrhage up to 40 ml/kg. The redistribution of rCBF favors those areas where neurons related to cardiovascular control are located. These findings have significant implications relating to hemodynamic regulation during hemorrhagic hypotension.

SINCE THE radioactive microsphere method was introduced by Rudolph and Heymann (1967), it has been applied to studies on blood flow in different animal species and organs. Regional blood flow can be measured in small tissue samples containing more than 400 spheres (Buckberg et al. 1971). Comparisons between the microsphere method and the more generally accepted methods such as electromagnetic flowmetry or clearance technique using diffusible indicator or inert gas have been made for the myocardium (Utley et aL, 1974;Domenech et al., 1969;Yipintsoi et al., 1973), kidney (Slotkoff et al., 1971), and stomach (Delaney and Grim, 1964).When cerebral blood flow is determined by the radioactive indicator clearance method (Lessen and Hoedt-Rasmussen, 1966), the washout curve in the dog differs from that in man in that there is a significant involvement of extracerebral tissues. Therefore the cerebral clearance curve in the dog often has more than two components (Haggendal, 1965), and the interpretation and analysis of the curve becomes relatively difficult. The present Xe washout curve in the dog by comparing it with simultaneous cerebral blood flow measurements obtained by the use of microspheres. It was hoped that these investigations also would help to evaluate the validity of the microsphere method in measuring cerebral blood flow and resolve some of the controversies (Harrell et al., 1977;Fung, 1973) regarding the applicability of this technique. MethodsThe experiments were performed on 15 mongrel dogs (15-25 kg) anesthetized with sodium pentobarbital (30 mg/kg injected intravenously, with supplements of 2 mgAg given approximately hourly). Pancuronium bromide (0.08 mgAg) was given intravenously after the trachea had been intubated. The dog was ventilated with a Harvard respirator to maintain the Pco? between 35 and 42 mm Hg. For five dogs, 5% CO2 was administered to raise the PCO2 to 50-60 mm Hg. The esophageal temperature was monitored by a thermistor probe (Yellow Springs Instrument Co.) and maintained between 36.5° and 37.5°C by electric heating pads. The femoral artery was cannulated with a double-lumen polyethylene catheter of original design with two orifices 18 cm apart, and the catheter was inserted to a distance such that the lower orifice was in the abdominal aorta and the upper orifice in the thoracic aorta. The femoral vein also was cannulated. After incision of the scalp to expose the skull, the central ridge of the skull was trephined, and the by guest on May 9, 2018