The effects of different rates of plasmanate infusions upon brain blood flow after asphyxia and hypotension in newborn piglets

May 1982 The Journal o f P E D I A T R I C S

791

The effects of different rates of plasmanate infusions upon brain blood flow after asphyxia and hypotension in newborn piglets Brain blood flow was determined in 21 spontaneously breathing, awake, newborn piglets during control, asphyxia, superimposed hypotension, and subsequent volume expansion (15 ml/kg of plasmanate). The piglets were divided into three groups based upon the rate of volume expansion: rapid infusion group--piglets received plasmanate in three minutes; slow infusion group--piglets received plasmanate in 30 minutes; the noninfused group--piglets did not receive plasmanate. The results showed comparable increases in brain blood flow among each group during asphyxia, and similar reduction to preasphyxia values during superimposed hypotension. Although pressure-passive changes occurred, the rate of volume expansion did not influence the magnitude of change in brain blood flow. Significantly lower arterial blood pressure and brain blood flow were observed in those piglets who did not have a plasmanate infusio n. Intracranial hemorrhages were not observed at autopsy in any of the study subjects. These data indicate that rapid or slow infusion of plasmanate for volume restoration did not influence the pattern of brain blood flow and that in these relatively mature brains, intracranial bleeding was not observed. Both plasmanate infused groups had higher brain blood flows at study completion (when compared to controls), reflecting compensation for anemia to maintain adequate oxygen delivery. Furthermore, regional differences in blood flow were found during asphyxia and superimposed hypotension (brain-stem > cerebellum > cerebrum), probably refleet(ng compensatory protection of vital portions of the central nervous system.

Abbot Laptook, M.D., Barbara S. Stonestreet, M.D., and William Oh, M.D.,* Providence, R.I.

COMMON NEUROLOGIC PROBLEMS in the perinatal period for the term and preterm infant are hypoxic-ischemic encephalopathy and intraventricular hemorrhage, respectively. ~Asphyxia and hypotension during this period have been implicated in pathogenetic models of these neurologic lesions.2 Asphyxia ha s profound effects upon brain blood flow, Hypoxia, 3 hypercarbia,4 and acidosis5 result in vasodilation of the brain vasculature. Brain blood flow therefore increases if the perfusion pressure is not altered. During the neonatal period, asphyxia may be accompanied by hypotension. Based on the principle of cerebral autoreg-

From the Department of Pediatrics, Women and Infants Hospital of Rhode Island, and The Program in Medicine, Brown University. *Reprint address: 50 Maude St., Providence, RI 02908-9976.

0022-3476/82/050791q-06500.60/0 9 1982 The C. V. Mosby Co.

ulatory control of blood flow, once the mean arterial blood pressure is less than a critical value (hypotension), blood flow decreases and passively follows changes in perfusion pressure. 6 See related article, p. 796.

Abbreviation used MABP: mean arterial blood pressure After insults such as asphyxia and hypovolemia-induced hypotension, the restoration of the systemic blood pressure by volume expansion is of critical importance. The rate at which the systemic blood pressure is restored may have beneficial or deleterious effects upon brain blood flow. Certainly a slow restoration of the blood pressure will

Vol. 100, No. 5, pp. 791-796

792

Laptook, Stonestreet, and Oh

prolong the duration of ischemia, whereas rapid reestablishment of a normal blood pressure will expedite metabolic recovery of the tissue. However, it has been speculated that when brain blood flow autoregulation is lost, transmission of rapid pressure changes from the systemic circulation to the brain vasculature may be an etiologic factor in arterial hemorrhagic infarcts in term infants and periventricular subependymal hemorrhages in preterm infants. 7 If the latter is true, the recovery of systemic blood pressure to the normotensive range may not be reflected by appropriate increments in brain blood flow, as in situations i n w h i c h autoregulation is present. We have therefore examined the relationship between the rate of volume expansion and the changes in brain blood flow after asphyxia and hypotension, using newborn piglets as study subjects. MATERIALS

AND METHODS

Twenty-one spontaneously breathing piglets were studied. Animal preparation was performed with the animals under local anesthesia (1% xylocaine) after a single dose (15 mg/kg) of ketamine was administered intramuscularly. A 31/2 Fr. umbilical vascular catheter (Argyle, Sherwood Medical Industries, St. Louis) was placed in the left ventricle via the left common carotid artery. A 19.0-gauge polyethylene catheter (Deseret Pharmaceuticals, Sandy, Utah) was placed into the abdominal aorta from the femoral artery. A second polyethylene catheter was placed into the inferior vena cava via the femoral vein. A tracheostomy was performed with placement of a cuffed 3.0 mm endotracheal tube. Stabilization followed for a period of one hour. All procedures were performed on an infant radiant warmer and the piglet's body temperature was maintained between 38 and 39~ Heart rate and mean arterial blood pressure were monitored continuously using a Bentley Trantec pressure transducer and recorded on a HewlettPackard polygraph (7700 series). Respiratory rates were determined by direct observation. Arterial Poz, PCo2, and pH were measured using a Corning 165 blood gas analyzer. Hematocrit was measured from arterial blood using the microhematocrit method. Brain blood flow was measured by radionuclide-labeled microspheres (l 5~, New England Nuclear, Boston) using the technique of Hey~mann et al. 8 Microspheres labeled with 5~Cr, 46Sc, l~3Sn, 57Co, and 9~Nb were used. Approximately 5 x 105 microspheres were administered at each blood flow determination. The left ventricle was utilized to inject the microspheres and the abdominal aorta served as thereference sample. For a total of two minutes, blood was withdrawn before, during, and after microsphere injections at a constant rate of 1.03 mi/minute using a Harvard

The Journal of Pediatrics May 1982

withdrawal pump. We have previously validated the use of the abdominal aorta as a reference sample for the calculation of brain blood flow in newborn piglets.9 In addition, we have examined the effects of asphyxia on ductal shunting of blood to ensure that a lower body reference sample would not be inaccurate because of a pulmonary-tosystemic shunt. In four newborn piglets exclusive of the study subjects, eight simultaneously obtained preductal and postductal arterial blood samples showed similar arterial oxygen tensioh values, indicating t h e absence of right-to-left Shunting across the ductus arteriosus. We have also examined the effects of unilateral common carotid artery ligation upon brain blood flow in piglets. The distribution of isotope is equal to each side of the brain (r = 0.99) and the absolute blood flows obtained are unchanged before and after carotid artery ligation. A ten-minute control period followed stabilization. The piglets were then subjected to an asphyxial insult by the addition of a dead space to the endotracheal tube. Initially 45 ml/kg was added to each piglet's a!rway, and this was altered throughout asphyxia to achieve the desired arterial blood gas values, which were measured every ten minutes during this period. Asphyxia lasted one hour. After asphyxia, the dead space was maintained and the animals were bled through the venous catheter until the MABP was less than 50 mm Hg; this was done rapidly and hypotension was maintained for eight to ten minutes. At the end of the hypotensive period, the dead space was removed and the animals were placed into one of the three groups. The rapid infusion group received 15 ml/kg of plasmanate (Cutter Laboratories, Berkeley, Calif.) over three minutes. The slow infusion group received 15 ml/kg of plasmanate over 30 minutes, and the no infusion group did not receive plasmanate. Infusions commenced with removal of the dead spac e and were administered via the venous catheter using a Harvard infusion pump. Five measurements of brain blood flow were performed in each study. These were done during control, at one hour of asphyxia, after ten minutes of suPerimposed hypotension, and during the infusion or no infusion period. The latter two measures corresponded to five and 30 minutes after removal of the dead space. At the time of each blood flow determination, arterial blood gas values and hematocrits were measured. At the termination of the study the animals were killed. Autopsy included verification of catheter placement and removal of the brain. After adequate fixation, the brain was dissected into specific sections and counted with the reference blood samples in a gamma counter (Packard Auto Scintillation Spectrameter). Since the reference Sample is a surrogate organ of known flow, blood flow can be calculated from the following formulaS:

Volume 100

Effects of plasmanate infusion on brain blood flow in piglets

79 3

Number 5

MEAN ARTERIAL (ram Hg)

7,00

,

60

,

,

,

,

PaCO2 (mm Hg) = = - R a p i d Inf. n ~ 8 o'~o-Slow Inf. n=8 ~. ,x No Inf n 5

50-

20-

/ lO [~

~

T 30

20-

--20

I0-

--i5

0 5 0 5

65

75

80

105

TIME (Minutes) Fig. 1. Arterial pH, PcO2, Po2, and base excess of each study group plotted at the time of brain blood flow determinations. Values are mean _+ SEM. (N = Number of piglets in each group; C = control period; ~BP = hypotensive period; * = P < 0.05.)

Known organ flow (reference sampte • blood flow (ml/min) CPM of microspheres in the brain CPM of microspheres in the reference sample

The data were analyzed by two way analysis of variance with repeated measures.i~ Post hoc tests were performed by an unpaired t test to compare the means between the groups. A paired t test was used to compare the brain blood flows of individual periods of the study with its respective control. A P < 0.05 was considered statistically significant for both the analysis of variance and unpaired t test. RESULTS At the time of the study, all three groups of piglets were comparable with respect to age and weight, The control values for arterial blood gases were similar among the three groups except for a higher Pao2 in the slow infusion group (Fig. 1). However, all control Pao2 values were within the physiologic range. All study groups had the

I

I

k

50-

T

I

(%)

S_

-

/ /

-',s';

I

HEMATOCRIT

BASE EXCESS (mEq/L)

~r qi,

25-

o-slo~,Infusion n:Sl A-No Infusion n=5I

FOO-T PaO2 (mm Hg)

Unknown organ flow = (ml/min)

BLOOD PRESSURE

65

TIME

75 8 0

105

(Minutes)

Fig. 2. Mean arterial blood pressure and arterial hematocrit of each study group plotted at the time of brain blood flow determinations. Values are mean _+ SEM. (N = Number of piglets in each group; C =control period; IBP =hypotensive period; * = P < 0.025 as compared to both infused groups.)

expected but comparable responses in pH, Paco2, Pao2 and base excess to asphyxia, hypotension, and the plasmanate infusions or no infusion. The amount of dead space added to the piglets of each group after one hour of asphyxia was also simiilar (M + S E M , 70 _+ 8, 75 + 8, and 78 _+ 17 m l / k g for the rapid, slow, and no infusion groups, respectively). The average M A B P for all 21 piglets was 86 m m Hg with a standard deviation of 15 mm Hg. In this study, taking 2 SD as demarcation points, the lower limit of normotension was 56 mm Hg; thus, we have used a M A B P of less than 50 mm Hg as an indication of hypotension. In all groups asphyxia was associated with an increase of M A B P and phlebotomy produced the desired hypotension (Fig. 2). The volume of blood removed was comparable in each study group (M + S E M , 31 _+ 4, 34 _+ 4, and 34 +_ 4 m l / k g for the rapid, slow, and no infusion group, respectively). Plasmanate administration resulted in an increase in M A B P in both the rapid and slow infusion groups. In the noninfused piglets, the M A B P first

Laptook,Stonestreet, and Oh

794

The Journalof Pediatrics May 1982 REGIONAL BRAIN BLOOD FLOW (ReNd Infusion) PERCENT CHANGE FROM CONTROL

TOTAL BRAIN BLOOD FLOW PERCENT CHANGE FROM CONTROL 200

: ~.-Rapid Inf. n=8J ] o - - - O - S l o w Inf. n=8 I

350

.

CONTROL FLOW VALUES (ml/min/lO0 q)

500

150 -CEREBRUM

250

-CEREBELLUM 112 +12

I00

-BRAIN STEM

200-

50

O

112 +15

109+ 9

150

I1(~(--~ ti4 4- II ml/min/IOOg ~l ~,1OO_+[2 I ~ ' 1 0 8 + 12

/

J

t00

-

50-

--50

:1

ASPHYX!A

0 5 TIME

,

I ,,.FUS.O. I

65 75 80 (Minutes)

105

Fig. 3. Total brain blood flow expressed as percent change from control is plotted for each study group. Values are mean + SEM. Control flows are listed. (n = Number of piglets in each group; c = control period; ~BP = hypotensive period; * = P < 0.05 as compared to the noninfused group.)

Table. Comparison of brain blood flow during each phase of the experiment with the control period

Asphyxia Group

and

[ Asphyxia hypotension

(ume expansion

min]I 30 rain i

Rapid infusion Slow infusion No infusion

~'t ~t ~t

--~ ~-

t* t*

tt tt

= NO change from control. *p < 0.05. ~'p < 0.01.

decreased and was significantly different than both infusion groups, but then increased to within the range of the infused animals by the termination of the study (30 minutes following removal of the dead space). Changes in the hematocrit were uniform among the three groups. The control values for total brain blood flow in each group were similar. Each study group had comparable increases in total brain blood flow with asphyxia and a similar return to control values with superimposed hypotension (Fig. 3). Following removal of the dead space, total brain flow had the following patterns: (1) in the rapid

ASPHYXIA

ASPHYXIA 81 HYPOTENSION

INFUSION PERIOD

Fig. 4. Regional brain blood flow of the rapid infusion group, expressed as percent change from control, during each study period. Values are mean _+ SEM. The two determinations of blood flow during the infusion period are similar and therefore are averaged and graphed once. (*P < 0.025 for brainstem compared to cerebrum and cerebellum; +P < 0.025 for cerebellum compared to cerebrum.)

infusion group there was an immediate rise in brain blood flow followed by a slight decline at study completion, (2) in the slow infusion group, there was a gradual and sustained increase in brain blood flow, and (3) in the no infusion group there was an initial further decrease in brain blood flow with a slight increase at the end of the study. During the 30 minutes after dead space removal, a signifiant difference was observed in brain blood flow between the noninfused group and both the rapid and slow infused groups. All three groups had significant increases in brain blood flow during asphyxia (Table). Hypotension resulted in a decrease in brain blood flow to the control range. The plasmanate-infused piglets had significant changes from control with volume expansion, whereas the noninfused piglets remained unchanged. A different distribution of brain blood flow to the cerebrum, cerebellum and brainstem was observed during asphyxia and hypotension in all three groups of animals studied (Fig. 4). The differential distributions were similar among the three groups. Asphyxia resulted in an increase in blood flow to all regions but the largest increment was to the brainstem and the smallest to the cerebrum (P < 0.025 when the magnitude of change among the three regions

Volume 100 Number 5

Effects o f plasmanate infusion on brain blood flow in piglets

are compared). During superimposed hypotension, the pattern of changes observed was as in asphyxia, despite the fall to control values. After restoration of the MABP and removal of the dead space, no regional differences were present. To interpret the observed brain blood flows with respect to the morphologic development of the central nervous system, a three-day-old piglet brain was examined neurohistologically (Courtesy of Dr. Domenic Purpura, Professor and Chairman, Department of Neuroscience, Albert Einstein College of Medicine, New York, N. Y.). The neurons have extensive dendrites with robust spines. A relatively sparse germinal matrix is present. At autopsy, no intracranial hemorrhages were found in any of the piglets studied. DISCUSSION This study has delineated the changes in brain blood flow during hypotensive asphyxia and following different rates of volume expansion in apontaneously breathing, awake newborn piglets. All piglets had increases in brain blood flow during asphyxia; superimposed hypotension resulted in a return of brain blood flow to the control range. During asphyxia, the observed increase in brain blood flow is a compensatory mechanism to maintain oxygen delivery to this vital organ. With hypotension, brain blood flow decreases, probably because further vessel dilation to maintain blood flow cannot be achieved, l~ However, the brain blood flow did not decrease to below control values because the vasodilatory effects of hypercarbia, hypoxia, and acidosis were still present. The brain blood flows during the infusion period were significantly higher than control in both plasmanate infused groups. The brain blood flow pattern may even given the impression that there was an "over correction" once the systemic arterial blood pressure was restored to the normotensive range. Our results differ from those obtianed by Lou et a12 2 In the latter, near term fetal lambs were asphyxiated in utero by umbilical vessel occlusion and MABP was varied by hemorrhage. After severe hypotensire asphyxia, restoration of the MABP by infusion of whole blood failed to raise cerebral blood flow. We cannot explain the discrepancy between Lou's results and ours. However, we speculate that the increase in brain blood flow in our study is dependent upon two processes. (1) It appears that the restoration of the brain blood flow is a pressure-passive phenomenon. A comparison of MABP (Fig. 2) and brain blood flow (Fig. 3) demonstrates parallel changes with the induction of hypotension and the subsequent volume expansion. (2) Acute anemia, is known to result in compensatory increases in brain blood flow. 13 The adaptive response to anemia probably determines the

795

magnitude of the changes in brain blood flow during this study phase. Jones et al TM have demonstrated that changes in the arterial oxygen content (induced by altering the hematocrit or Pao2 of newborn lambs) are accompanied by reciprocal changes in brain blood flow so that effective oxygen delivery to the brain is maintained. Indeed, calculation of the percent decrement in hematocrit ( ~ 50%) is approximately equivalent to the percent increment in brain blood flow. Therefore, what may appear as a pattern of brain blood flow "over correction" actually has a firm physiologic basis. The lower brain blood flow in the noninfused piglets (unchanged from control) reflects less than optimal oxygen delivery to the brain and is probably secondary to the marginal MABP. Although the restoration of brain blood flow after asphyxia and hypotension appeared to be pressure-passive, no evidence of intracranial hemorrhage was observed. The central nervous system pathology resulting from asphyxia is more pronounced in premature newborn subjects and is characterized by intraventricular hemorrhage; since our piglets were relatively mature (comparable to 36 to 38 weeks' gestation in human infants), as indicated by their histologic findings, the lack of intracranial hemorrhage is not surprising. Furthermore, the rapid or slow infusion during volume re-expansion did not result in intracranial hemorrhage, which suggests that volume restoration can be achieved rapidly without concern of significant injury to the central nervous system in term subjects. Since the maturational state of our study piglets is closer to that of term human infants, the data, of course, are not applicable to preterm subjects. The regional blood flow patterns revealed significant differences during asphyxia alone and with superimposed hypotension. These findings agree with fetal lamb studies in which asphyxia was produced by slow partial umbilical cord compression2 s Prolonged intrauterine hypoxia in lambs also resulted in the same patterns of regional blood flow26 The etiology for the regional differences is unknown. We speculate that the mechanisms responsible may represent a "protective" response of the brainstem. In the neonatal period the brainstem represents the region responsible for reflex behavioral responses and therefore may have high metabolic rates for oxygen and glucose. These results and speculation would also be consistent with previous neuropathologic studies of in utero partial asphyxia utilizing monkeys, ~7in which a relative sparing of brainstem structures occurred. The authors thank Mr. Raymond Petit for his skillful technical assistance and Albert S. Most, M.D., F.A.C.C., Chief of Cardiology and the staff of the Cardiac Research Laboratory of Rhode Island Hospital, for their help and use of the facilities.

796

Laptook, Stonestreet, and Oh

REFERENCES 1. Volpe J J: Perinatal hypoxic-ischemic brain injury, Pediatr Clin North Am 23:383, 1976. 2. Wigglesworth JS, and Pape KE: An integrated model for haemorrhagic and isehaemic lesions in the newborn brain, Early Hum Dev 2:179, 1978. 3. J6hannsson H, and Siesj6 BK: Cerebral blood flow and oxygen consumption in the rat in hypoxic hypoxia, Acta Physiol Scand 93:269, 1975. 4. Reivich M: Arterial PCO2 and cerebral hemodynamics, Am J Physiol 206:25, 1964. 5. Bucciarelli RL, and Eitzman DV: Cerebral blood flow during acute acidosis in perinatal goats, Pediatr Res 13:178, 1979. 6. Lassen NA: Autoregulation of cerebral blood flow, Circ Res 14 (Suppl 1):1, 1964. 7. Wigglesworth JS, and Pape KE: Pathophysiology of intracranial haemorrhage in the newborn, J Perinat Med 8:119, 1980. 8. Heymann MA, Payne BD, Hoffman JIE, and Rudolph AM: Blood flow measurements with radionuclide-labeled particles, Prog Cardiovas Dis 20:55, 1977. 9. Laptook A, Stonestreet BS, and Oh W: Autoregulation of brain blood flow in the newborn piglet: Regional differences in flow reduction during hypotension, Early Hum Dev (in press). 10. BMDP2V Statistical Package: Health Science Computing Facility, University of California, Los Angeles.

The Journal o f Pediatrics May 1982

11. Harper MA, and Glass HI: Effect of alterations in the arterial carbon dioxide tension on the blood flow through the cerebral cortex at normal and low arterial blood pressures, J Neurol Neurosurg Psychiat 28:449, 1965. 12. Lou HC, Lassen NA, Tweed WA, Johnson G, Jones M, and Palahniuk R J: Pressure passive cerebral blood flow and breakdown of the blood brain barrier in experimental fetal asphyxia, Acta Pediatr Scand 68:57, 1979. 13. Borgstr6m L, J6hannsson H, and SiesjS, BK: The influence of acute normovolemic anemia on cerebral blood flow and oxygen consumption of anesthetized rats, Acta Physiol Scand 93:505, 1975. 14. Jones MD, Traystman RJ, Simmons MA, and Molteni RA: Effects of changes in arterial 02 content on cerebral blood flow in the lamb, Am J Physiol 240:H209, 1981. 15. Johnson GN, Palahniuk RJ, Tweed WA, Jones MV, and Wade JG: Regional cerebral blood flow changes during severe fetal asphyxia produced by slow partial umbilical cord compression, Am J Obstet Gynecol 135:48, 1979. 16. Ashwal S, Majcher JS, and Longo LD: Patterns of fetal lamb regional cerebral blood flow during and after prolonged hypoxia: Studies during the posthypoxic recovery period, Am J Obstet Gynecol 139:365, 1981. 17. Brann AW, and Myers RE: Central nervous system findings in the newborn monkey following severe in utero partial asphyxia, Neurology 25:327, 1975.

Brief clinical and laboratory observations Intraventricular hemorrhage following volume expansion after hypovolemic hypotension in the newborn beagle Jan Goddard-Finegold,* Dawna Armstrong, and Robert S. Zeller, Houston, Texas

AN EXPERIMENTAL MODEL for intraventricular hemorrhage has been produced in the beagle pup, which has a germinal m a t r i x at term. W e have been able to monitor hemodynamics in this newborn animal at 24 to 72 hours of From the Baylor College of Medicine. Supported by Baylor College o f Medicine Biomedical Research Support Grant No. 520-9100 and by the Departments of Pediatrics and Pediatric Pathology o f the Texas Children's Hospital. *Reprint address: Department of Pediatric Neurology, Texas Children's Hospital, R618, 662l Fannin St., Houston, TX 77030.

age. By changing one physiologic variable at a time in acute experiments, we have tested specific factors t h o u g h t i m p o r t a n t in the pathogenesis of IVH. To date we have produced intraventricular h e m o r r h a g e and h e m o r r h a g e

See related article, p 791. restricted to the periventricular subependymal cell plate, the choroid plexus, and the tela choroidea in this model. 1,2 These h e m o r r h a g e s have been produced in pups after m e a n systemic arterial pressure has been rapidly but moderately

0022-3476/82/050796+04500.40/0 9 1982 The C. V. Mosby Co.