Determination of cerebral cortical blood flow: A thermal technique

ORIGINAL CONTRIBUTION blood flow, cerebral cortical; cerebral cortical blood flow; perfusion, cerebral cortical

Determination of Cerebral Cortical Blood Flow: A Thermal Technique A mathematical model for tissue thermodilution was developed to study cerebral cortical perfusion before and after controlled perfusion arrest. Cerebrad cortical perfusion rates are readily determined by this method. A thermistor was introduced into the subdural space and secured in direct contact with the frontal cortex in 12 dogs on ketamine and gallamine anesthesia. A 22-gauge angiocath was placed in the right superior thyroid artery and directed into the carotid artery on the same side as the thermistor, The dogs were placed on cardiac bypass using a circuit from the right atrium to the pulmonary artery and a second circuit from the left ventricular apex to the left femoral artery. Arterial pressure, central venous pressure (CVP), intracranial pressure (ICP), and left atrial pressure (LAP) were monitored directly. A heat exchanger was used to maintain a constant blood temperature of 37 C in the output of the left side bypass circuit. Thermal flow curves were generated in the cerebral cortex by injecting 2 to 4 cc of cold saline into the common carotid artery through the injection catheter. Preliminary evaluation of this flow method m comparison to radioactive microspheres indicates that this method can be used in a reliable and reproducible fashion to determine cerebral cortical blood flow. [Hoetmer PJ, Krause GS, White BC, Gadzinski DS: Determination of cerebral cortical blood flow: A thermal technique. Arm Emerg Med 12:2-7, January 1983.]

Paul J. Hoehner Gary S. Krause, MD Blaine C. White, MD, FACEP Daniel S. Gadzinski, MD Detroit, -Michigan From the Section of Emergency Medicine, Wayne State University College of Medicine, Detroit, Michigan. Supported by a grant from The Upjohn Company and Organon Pharmaceuticals. Presented at the University Association for Emergency Medicine Annual Meeting in San Antonio, Texas, April 1981. Resubmitted for publication April 1982. Address for reprints: Blaine C. White, MD, FACEP, Associate Professor and Research Coordinator, Section of Emergency Medicine, A-214-D, East Fee Hail, Michigan State University College of Medicine, East Lansing, Michigan 48823.

INTRODUCTION Brain death or major neurologic deficit after successful cardiopulmonary resuscitation is a tragedy that may occur in up to 60% 1 of patients post resuscitation. The early evidence provided by Ames ~ that perfusion of the cerebral capillary network may be sharply reduced after anoxic cerebral insuits, even at good arterial perfusion pressures, has given both the experimental and clinical study of cerebral blood flow a high priority. A number of methods have been developed to study cerebral blood flow, including such inert gas techniques as the hydrogen clearance method, 3,4 isotopic scanning, 59 the use of radioactive microspheres, 1° and the RapelaGreen technique. 11 All these methods have disadvantages. The inert gas techniques do not provide regional flow information and require selective arterial carmulation. The Rapela-Green technique requires cannulation of the cerebral venous confluence and surgical occlusion of the lateral sinuses. The isotopic scanning methods require the use of radioactive compounds (133xenon, 77 krypton, 14carbon labeled microspheres, or 14carbon iodoantipyfine) and of expensive counting equipment. None of these methods lends itself to rapid repetition. Previous studies of post-anoxic cerebral perfusion have produced cerebral anoxia by high pressure circumferential neck compression or by surgical ligation of the arteries. 12 "18 Our model was designed to allow controlled perfusion arrest as the cerebral anoxic insult and to develop a simple and inexpensive technique to measure local cerebral cortical blood flow repetitively. MATERIALS AND METHODS Perfusion Control System Twelve mongrel dogs weighing between 20 kg and 25 kg were anesthetized with 7 mg/kg ketamine and 1 mg/kg gallamine administered by intravenous 12:1 January 1983

Annals of Emergency Medicine

2/19

DETERMINATIONOF CCBF White et al

Fig. 1. Bypass perfusion system.

Right heart pump. (W) bolus. Anesthesia was maintained with an IV solution prepared with 2 g ketamine and 100 mg gallamine in 500 cc 5DW run at an average rate of 75 cc/hour. Immediately after induction of anesthesia, endotracheal intubation was performed and controlled ventilation on room air was begun with tidal volumes of 12 cc/kg body weight. Ventilator rates were adjusted to maintain the arterial PCO2 at 40 torr on an IL-313 blood gas machine (Instruments Laboratory). Bilateral femoral arterial cutdowns, a right external jugular cutdown, and a median sternotomy were performed. At the time the chest was opened, 5 cm PEEP was added to the ventilation circuit to maintain lung inflation. The dogs were anticoagulated with heparin 3 mg/kg IV and placed on cardiac bypass without extracorporeal oxygenation, using a circuit from the right atrium to the pulmonary artery and a second circuit from the left ventricle to the left femoral artery (Figure 1). A catheter was placed in a left pulmonary vein directed at the left atrium and was connected to a pressure transducer. Systemic arterial pressure was monitored using a line in the right femoral artery. The CVP was monitored continuously using a catheter placed in the superior vena cava from the right external jugular cephalic to the bifurcation with the maxillary vein. Bypass was begun when the dog was placed in ventricular fibrillation by injecting 20 mEq potassium chloride through the pulmonary venous line and applying a defibrillator shock of 20 joules using internal paddles. It was maintained with adjustments of pump rates in the right and left circuits as indicated by continuous monitoring of CVP, arterial pressure, and left atrial pressure. Venous preload was augmented as needed with lactated Ringer's solution through a right femoral venous catheter. Arterial blood temperature was kept constant by the heat exchanger in the left circuit.

Intracranial Monitoring The scalp was reflected bilaterally and the skull exposed. An epidural pressure screw 19 was placed over the left parietal area. A 12-mm craniotolny hole was placed over the right parietal area and a 5-mm hole

20/3

I ~ 1

ILAP

CVP

Left heart pump I

MAP

j

moral a . / ~ / / / \

I. femoral a.

7 Heat exchanger

drilled immediately posterior to it. An Elecath Swan-Ganz catheter (ElectroCatheter Corporation, Rahway, NJ) was cut back to within 3 m m of the thermistor and placed into the cranium epidurally through the 5-mm hole. The tip of the catheter was drawn into the 12-mm craniotomy hole. The dura was carefully opened and the catheter passed through it so that the thermistor was in direct contact with the frontal cortex. The catheter was secured in position and the 12-ram hole closed with a seethrough bolt. The 5-mm hole was closed with quick-setting epoxy.

Cerebral Cortical Blood Flow Determination The superior thyroid branch of the right common carotid artery was iso-

Annals of Emergency Medicine

lated and ligated near the midline. An angiocath was secured in the artery directed toward the common carotid, and was heparinized and closed with a three-way stopcock. The thermistor on the cortex was connected through a Columbia Instruments Thermal Cardiac Output Computer (model No. 72-9FX, Columbia Instruments) to the chart recorder and balanced. The 0.20 C deflection was standardized on the chart recorder using the controls of the computer. Cold (0-4 C} saline was injected in a bolus over less than 5 seconds through the arterial catheter, and the thermal curve generated in the cortex was recorded. The volume of cold saline injected was selected as that necessary to achieve a 0.5- to 0.7-C change under the thermistor. The necessary volume

12:1 January I983

Fig. 2. Calculation of cerebral blood flow from the rewarming part of the thermal curves.

For the 1 c c of brain under the thermistor, by the law of specific heat capacity: 1) Heat (calories) = Temperature x mass x K where K = thermal constant for a substance. For water K = 1. 2) Heat in blood flow = Temperature of blood x flow rate x time (C° x c c / s x s = C ° x c c = cal) 3) Let: HBI = heat carried in blood flow in HBO = heat carried in blood flow out HT = heat in the 1 cc of tissue F = blood flow Ts~ =temperature of blood flowing in TBo =temperature of blood flowing out 4) Then by conservation of energy dHT = dHB~ - dHBo (heat change in tissue = heat in - heat out) 5) dHT = dtTaiF - dtTBoF (by substitution from [2]) 6) Assuming instantaneous heat transfer in the 1 cc Tao = HT (temp blood out = temp tissue) 1 cc 7) dHt = dtF x [TBI -

, HT ] (by substitution of [6] in [5]) 1 cc

8) Note that dHT = F x [Tal - HT ] dt 1 cc means that the instantaneous slope of the rewarming curve is directly proportional to the rate of blood flow.

9) After solving [7] for dt we can integrate both sides, because for rewarming curve and equation [7] when t = 0, HT = HTo For any point a on the rewarming curve.

~ j(HTaHT 0 C]HT dt ='~

F x [TB~"---- HT 1CC 10) t = l#_q____qX [ -- In (TBI (lcc) - HT)] is the indefinite integral. 1 CC 11)t = T x ( -

,TBt (1 CC) -- HTa]) definite integral. lnLTB I ( l c c ) Hvo is the

12) For the rewarming curve HT at t = 0 is (Tp) (1 cc) where Tp = tissue temperature at peak cooling. Similarly, HTa = (tissue temp at a x 1 cc) for any point on the rewarming curve. Therefore: F = -ln

[TB~ - Tissue temp at a] TB/ -- Tp

x

1 cc sec at a 2

ranged from 1.5 to 5.0 cc. A mathematical method was developed to calculate cerebral blood flow from the rewarming part of the thermal curves (Figure 2). Projections of rewarming curves at various cerebral flow rates using this equation {Figure 3) demonstrate the sensitivity of the model.

RESULTS The mean rate of cerebral cortical blood flow in initial-single determinations in the 12 dogs was 2.22 cc/g/min ( + 0.47 cc/g/min = 1 SDU). The flow 12:1 January 1983

rates of two animals kept on bypass for 90 minutes without insults were stable (Figure 4). Four rapid repetitions of the flow study in these animals did not significantly alter the determined flow rates (Figure 5).

DISCUSSION The rate of cerebral cortical blood f l o w d e t e r m i n e d by our t h e r m a l method is in the high range reported by investigators using other methods. Rehncrona 2° has reported baseline cortical flow rates of 1.4 co/g/rain in rats using t4carbon aminoantipyrine. Annals of Emergency Medicine

Bemtman et a121 reported a value of 1.2 cc/g/min in rats using a nonregional 133xenon technique. Hamer et a122 reported a value of 0.56 cc/g/min in dogs using the nonregional nitrous oxide method. Using Rapela-Green, also a nonregional technique, Emerson et a123 reported a flow of 0.5 cc/g/min in dogs. These same investigators, using t4carbon microspheres, report a value of 0.22 co/g/rain in dogs. ]° Ohno 7 reported a value of 1.6 cc/g/min for regional cortical flow in rats. Yamamoto 6 found a regional cortical flow of 0.8 mI/g/min in man using ZZkrypton. Freygang and Sokoloff24 reported flow of 1.4 cc/g/min. These studies reflect a wide range of reported flow values that vary with the method used. Generally, nonregional studies give lower flow values. Regionally selective studies have shown the cortex to have flow rates two to three times greater than white matter. 2s Methods determining total brain blood flow yield low values relative to selective cortical studies. In addition to the fact that our technique is specifically for cortical flow, two other factors in the model may contribute to the flow results being in the high range of previously reported values for CBF. First, we have used ketamine anesthesia because the model is developed so that we can produce perfusion arrest by stopping bypass, and resuscitate by reinstituting bypass. Ketamine. has no p r o t e c t i v e effect for cerebral anoxia. 26 It is, however, known to increase CBF.27 Thus the choice of this anesthetic for model development may increase, our baseline flow determinations. Second, the m e a s u r e m e n t s are being done with animals on bypass perfusion, which produces nonpulsatile flow. This may also increase the flow rates. Preliminary comparison of our flow m e t h o d w i t h a radioactive microsphere regional flow technique in animals without bypass has demonstrated good agreement in the data produced by the two techniques. 28 Five technical advantages of our thermal method are evident. 1) It does not require a carefully fixed volume or temperature of the cool saline injected, because all that is necessary to generate the rewarming is a simple thermal gradient between 4/21

DETERMINATION OF CCBF White et al

0.7

F = 0.5 cc/g/min

0.6

F = 10 cc/g/min cO

E"

/

o.5

F = 15 cc/g/min

d.,) ,t'-

O3

~

.Q

0,4

©

__c

O,3

c-

~ o

0.2

O3

I 0

10

20

3 the tissue and the arterial blood. 2) It does not require selective cannulation of a cerebral artery. 3) It is not d e p e n d e n t on flow through any single cerebral artery, but rather measures net tissue flow from any combination of arterial sources. 4) The generated flow data are specifically regional. 5) It lends itself to rapid, repetitive determinations because one determ i n a t i o n does not alter apparent vascular resistance in succeeding trials (Figure 4). T h e m e t h o d m a k e s t w o fundamental assumptions. The first is that the hydrostatic heat transfer is nearly instantaneous. Support for this assumption comes from three facts. First, tissue cellular surface areas are large with respect to both capillary and cellular v o l u m e y ie, there is a large area to exchange heat. Second, molecular movement of water across cell membranes is very fast; it is close to the velocity of xenon movement, which is assumed to be effectively 22/5

30

40

50

60

70

80

90

100

Time (seconds)

instantaneous. 3° Finally, heat transfer, rather than molecular movement, is the critical element in this method. Heat transfer by molecular collision will be far faster than even water molecular movements. 31,32 Therefore, the assumption that the heat transfer is nearly instantaneous reflects the speed of heat transfer, the high velocity of transmembrane exchanges of the water molecule, and the physiologic fact of large surface area to volume relationships. The second implicit assumption is that the rate of endogenous cerebral heat production is small with respect to the caloric load of the inflowing arterial blood. Cerebral temperature closely reflects arterial temperature. 33 Freygang and Sokoloff24 reported cerebral cortical flow rates of 1.4 cc/g/ m i n and cerebral O2 consumption (CMRO2) of about 1.5 ~ M / g / m i n . General rate of heat production per txM/O2 consumed is about 0.0002 cal/ la,MO2.34 This would give an endogenous rate of cerebral heat production Annals of Emergency Medicine

Fig. 3. Theoretical rewarming cortical flow curves generated by the thermal flow equation. of 0.001 cal/g/min. Meanwhile, total heat flow, using their flow value at 37 C blood temperature, would be 434 calories. Moreover, a theoretical rewarming curve calculated using our method and a 0.5 C gradient at t = 0 would be 80% equilibrated within 69 seconds at a flow rate of 1.4 cc/g/min. During this time endogenous cerebral heat production would have been 0.002 calories in the gram under the thermistor. These calculations argue that the error induced in the instantaneous slopes of the rewarming curve by endogenous heat production is negligible. Finally, although this study does not address the question, our thermal method may, with modification, be applicable for clinical determination of regional cerebral cortical blood flow. Because the method does not re12:1 January 1983

quire selective cerebral vascular cannulation, it may be possible to determine cerebral cortical flow rates using cold saline infusion in patients with a head bolt modified to include a thermistor. Investigation of the use of two thermistors, one in a subdural screw and the other in a peripheral artery,

Fig. 4. Cerebral cortical blood flow by the thermal method during cardiac bypass. Fig. 5. Four thermal cortical blood flow curves generated within 20 minutes.

Data projected as mean +_ 1 SDU (The level mean is a least squares fit of all data points from 2 dogs).

4.0

"0 0.2.

3.0

--

1/11iif///////1111111//111111/ I//////////////////////////////

0 "0 0

2.0

m

with flows determined by area difference during cold saline IV infusion is currently being done in our laboratory. CONCLUSION Cerebral cortical blood flow can be determined using cortical rewarming curves following acute low-grade cortical cooling by injection of cold saline into the c o m m o n carotid artery. The equation for the flow rate in the gram of tissue under the thermistor at all points on the r e w a r m i n g curve is shown (Figure 6). Rapid repetitive flow measurements with this method do not distort determined flow values. The method does not require selective cerebral vascular cannulation. Values determined by this method are in agreement with reported flows determined by regional molecular isotopic scanning methods.

0

REFERENCES

0

0

1.0

...0

0 I

I

I

I

I

I

15

30

45

60

75

90

Minutes on Bypass Perfusion

4

j

Repetitive study did not significant!y alter the determined flow rates.

i0.8

t5

F = 2.6

0.6

F = 2.7

S ~0.4 E 0.2 ~-

1. Myerburg RJ, Conde CA, Sung RJ, et al: Clinical electrophysiologic and hemodynamic profile of patients resuscitated from prehospital cardiac arrest. Am J Med 68: 568, 1980. 2. Ames A, Wright ILL, Kowada M, et al: Cerebral ischemia II: The no reflow phenomenon. Am J Pathol 52:437, 1968. 3. Kety SS, Schmidt CF: Determination of cerebral blood flow in man by the use of nitrous oxide in low concentration. Am J Physiol 143:53, 1945.

0

\

~

2.7

10

F = 2.4

15

20

Time (minutes)

12:1 January 1983

Annals of Emergency Medicine

6/23

DETERMINATION OF CCBF White et al

FCC/sec/g =, -

In r

temperature of arterial blood -

/ temperature of arterial blood

4. Kety SS, Schmidt CF: The nitrous oxide method for the quantitative determination of cerebral blood flow in man: Theory, procedure, and normal values. J ClJn Invest 27:476, 1948. 5. Ingvar DH, Lassen NA: Regional blood flow of the cerebral cortex determined by 8Skrypton. Acta Physiol Scand 54:325, 1962. 6. Yamamoto YL, Thompson C, Meyer E, et al: Three dimensional topographical regional cerebral blood flow in man measured with high efficiency MNI-BGO two ring positron device using 77krypton. Acta Neurol Scand 72{suppl):i86, 1979. 7. Ohno K, Inaba Y, Pettigrew KD, et al: Regional cerebral blood flow in the conscious rat, as measured with 14C-iodoantipyrine and 3H-nicotine. Acta Neurol Scand 72(suppl):210, 1979. 8. Eklof B, Lassen NA, Nilsson L, et al: Blood flow and metabolic rate for oxygen consumption in the cerebral cortex of the rat. Acta Physiol Scand 88:587, 1973 9. Lassen NA, Munck O: The cerebral blood flow in man determined by the use of radioactive krypton. Acta Physiol Scand 33:30, 1955. 10. Bryan WJ, Emerson TE: Blood flow in seven regions of the brain during endotoxin shock in the dog. Proc Soc Exp Biol Med 156:205, 1977. 11. Rapela CE, Green HD: Autoregulation of canine cerebral blood flow. Circ Res 1415(suppl I):205, 1964. 12. Nemoto EM, Bleyaert A, Stezoski SW: Global brain ischemia: A reproducible monkey model. Stroke 8:558, 1977. 13. Ginsberg MD, Welsh FA, Budd WW: Abnormalities of brain blood flow and metabolism during recovery from diffuse cerebral ischemia: Dependence upon duration of ischemia. Acta Neurol Scand 56 (suppl 64):210, 1977. 14. Ginsberg MD, Reivich M, Frank S: Pyridine nucleotide redox state and blood flow in the cerebral cortex following mid-

24/7

(tissue temperature at a) ] x 1 cc tissue temperature at peak J sec at a

dle cerebral artery occlusion in the cat. Stroke 7:125, 1976. 15. Nordstrom CH, Rehncrona S, Siesjo BK: Restitution of cerebral energy state after complete and incomplete ischemia of 30 minutes duration. Acta Physiol Scand 97:270, 1976. 16. Ginsberg MD, Meyers RE: The topography of impaired microvascular perfusion in the primate brain following total circulatory arrest. Neurology 22:998, 1972. 17. Kowada M, Ames A, Majno G: Cerebral ischemia I: An improved experimental method for study; cardiovascular effects and demonstration of an early vascular lesion in the rabbit. J Neurosurg 28:150, 1968. 18. Fischer EG: Impaired perfusion following cerebrovascular stasis. Arch Neurol 29:36!, 1973. 19. Vries JK, Becket DE Young HF: A subarachnoid screw for monitoring intracranial pressure. J Neurosurg 39:416, 1973. 20. Rehncrona S, Abdul-Rahman A, Siesjo BK: Local cerebral blood flow in the postischemia period. Acta Neurol Scand 60 (suppl 72):294, 1979. 21. Berntman LI Carlsson C, Siesjo BK: Cerebral O2 consumption and blood flow m hypoxia: Influence of sympathoadrenal activation. Stroke 10:20, 1979. 22. Hamer J, Hoyer S, Alberti E, et al: Cerebral blood flow and oxidative brain metabolism during and after moderate and profound arterial hypoxemia. Acta Neurochir 33:141, 1976. 23. Emerson TE, Diguez GC, Harkema JM, et al: Cerebral and hemodynamic and metabolic abnormalities in hemorrhagic shocked dogs: Comparison of treatment with whole blood alone, dextran alone, or combined with Solu-Medrol®. Read before the Erwin Riesch Symposium on Cerebral Microcirculation and Metabolism, in New York City, July 9-12, 1980. 24. Freygang WH Jr, Sokoloff L: Quantitative measurement of regional circulation in the central nervous system by the use of

Annals of Emergency Medicine

Fig. 6. Finished equation for tissue blood f l o w by thermal dilution. radioactive inert gas, in Tobias CA, Lawrence JH (eds): Advances in Biological Physics, vol 6. New York, Academic Press, 1958, p 263-279. 25. Ingvar DH, Lassen NA: Regulation of cerebral blood flow, in Himwich HE (ed): Brain Metabolism and Cerebral Disorders. New York, Spectrum Publications, 1976, p 181-206. 26. Lighffoot WE; Molinari GF, Chase TN: Modification of cerebral ischemic damage by anesthetics. Stroke 8:657, 1977. 27. Dawson B, Michenfelder JD, Theye RA: Effects of ketamine on canine cerebral blood flow and metabolism: Modification by prior administration of thiopental. Anesth Analg 50:443, 1971. 28. Hoehner PJ, Dean M, Rogers M, et al: Validation of the thermal clearance technique for regional CBF with radioactive microspheres. Submitted to Anesthesiology. 29. Baptista AG: Aspects of cerebral circulation, in Himwich HE (ed): Brain Metabolism and Cerebral Disorders. New York, Spectrum Publications, 1976, p 129-161. 30. Ketty SS: Measurement of local circulation within the brain by means of inert, diffusible tracers, examination of the theory, assumptions, and possible sources of errors. Acta Neurol Scand 14(suppl):20, 1965. 31. Atkins KR: The nature of heat, in Physics. New York, John Wiley and Sons, Inc, 1965, p 191. 32. Handbook of Chemistry and Physics, ed 44. Cleveland, Chemical Rubber Publishing Company, 1963, Thermal Conductivity Table, p 2535. 33. Baker /VIA, Hayward JM: Intracranial heat exchange and regulation of brain temperature in sleep. Life Sci 7:349, 1968. 34. Shepherd RS: Human Physiology. Philadelphia, Lippincott Publisher, 1971, p 467.

12:1 January 1983