CA1188944A - Stroke treatment utilizing extravascular circulation of oxygenated synthetic nutrients to treat tissue hypoxic and ischemic disorders - Google Patents

Abstract

STROKE TREATMENT UTILIZING EXTRAVASCULAR CIRCULATION
OF OXYGENATED SYNTHETIC NUTRIENTS TO TREAT
TISSUE HYPOXIC AND ISCHEMIC DISORDERS

Abstract:

A novel acute care cerebral support system and method for treating severly ischemic brains is disclosed where-in an oxygenated nutrient emulsion is circulated through at least a portion of the ventriculo-subarachnoid spaces. The nutrient emulsion contains an oxygenatable non-aqeuous compo-nent, an aqueous nutrient component, an emulsification compo-nent, and other components which render physiologic acceptabil-ity to the nutrient emulsion. The disclosed system and method have been shown to effectively exchange oxygen, carbon dioxide, glucose, and other metabolites in severely stroked brains. Significant restoration of oxidative metabolism and electrographic activity result from the disclosed treatment.
Methods for producing the nutrient emulsion and a system for delivering that emulsion to the cerebrospinal pathway are also disclosed. Additionally, novel diagnostic methods for diagnosing the physiologic state of hypoxic-ischemic and other diseased neurologic tissue during treatment are provided.

Description

STROKE TREATMENT UTILIZING EXTRAVASCULAR CIRCULATION
OF OXYGENATED SYNTHETIC NUTRIENTS TO TREAT
TISSUE HYPOXII~ AND ISCHEMIC DISORDER

Background of the Invention Cerebrovascular accident, a disease commonly known as 'Istroke'l~ remains the third leading cause of death, and probably cons~itutes the single largest category of long term disability in this country.
In spite of current medical knowledge and available treatments, a major central nervous system vascular occlusion is quickly attended by irreversible damage to the affected brain region(s). A "completed stroke"
is manifest by a fixed and permanent neurological deficit. Millions of dollars have been expended in stroke research and care by Federal and private agencies ~ithout a single substantial gain in our present chemo-therapeutic abili~ies for a completed stroke.
On a clinical level, once vascular flow in any portion of the central nervous sys~em has ceased for longer than a few minutes, a permanent l'stroke" invariably ~ollows.

~ 1 It is not currently possible to recover substantial neural function with clinical ischemia of 5~7 minutes duration. An exquisite neuronal sensi~ivity to oxygen deprivation has been blamed for this ultra-short stroke irrever~ibility.
Neurons do indeed have meager metabolic storage and are unable to meet energy needs by anerobic means. Well accepted concepts hold that such permissible cerebral ischemia times are critical and neurons must quickly be resupplied or meta-bolic and infarction will result. While clinically true, recent laboratory investi~ations have addressed -the problems of ischemic vascular and neuronal reactions separately with considerably different results. Recently reported studies indicate neurons are not as sensitive as previously believed.
Indeed, it has been suggested that neurons can withstand global ischemia for l hour or longer~ R.A. Hossman~ P.
Kleihues, Arch. Neurol. 29, 375-389 (1973). If the clinical and experimental observations are to ~e reconciled,one hypoth-esis is 'chat long term damage result~ from vascular rather than neuronal sensitivity to oxy~3en deprivation. It is kno~n that secondary reactive change~ appear within the microcirculation after sufficient stagnation. A. ATnes III, R. L. ~ight, M. Kowada, JoM~ Thurston, G. Majno, Am.
J. Pathol. 52, 437~448~ (1968). J. Chiagn, M. Rowada, A.
~mes III, Am. J. PathoI. 52, 455-476 (1568). E. G. Fischer/
~5 Arch Neurol. 29t 361-366, ~lg73). E. G. Fischer, A~ Ames III, E. T. ~edly-Whyte, S. O'Gorman, Stroke 8, 36-39, (1977)0 Even if blood is represented to the local tree, the small vessels do not completely reopen. Vnder these circumstances TJU 3-3 ~ 9 ~ ~

ischemic, though potentially recoverable, neurons may be lethalized because they are not adequately resupplied with blood within their metabolically ~olerable limits. This concept shifts the basic fault in stroke from ~ultrasensitive~
neurons to a protracted blood flood ailure. Nonetheless, a long felt need exists to prevent permanent damage and/or reverse neurol~gic deficits resulting from interrupted vascular flow.
One experimental approach which has heen used to investigate the effects vf stroke on neurologic tissue is the perfusion of fluids of known cGmposition through ven-triculocisternal spaces. For example, E. Fritschka, J. L.
Ferguson and J.J. Spitzer have reported increases in free fatty acid turnover in cerebral spinal fluid during hypo-tension in do~s. According to the Fritschka technique, a amocka cerebral spinal fluid containing radio-labelled palmitate was perfused from the lateral ventricle to the cisterna magna of conscious dogs. Arteriovenous glucose and fatty acid concentrations, and ~mock~ CSF fatty acid concentrations were monitored over a period of 6 hours of perfusionO Estimates of the amount of palmita~e recovered from the cisternal effluent and cerebral venous blood lead to the conclusion that a si~eable fraction of free fatty acids may be taken up by tissues ~in the vicini~y of the CSF
spacen. See Fritschka et al, UIncreased Free Fatty Acid Turnover in CSF During Hypotension in Dogs", American Journal of ~y~ y, 2320H802-H807. Xn ~Bu~k Flow and Diffusion in the Cerebral Spinal Fluid System of ~he Goat~, by Heise, Held, and Pappenheimer~ a ventriculo cisternal perfusion TJU 3-3 ~ 4 method was used on chronically prepared, unanaesthized goats.
Measurements were made of steady-state rates at which inulin, fructose, creatinine, urea, potassium, sodium, and labelled water were removed from perfusion fluid at various hydrostatic S and osmotic pressures. The subject perfusions were carried out on female goats provided with implanted ventricular and cisternal guide tubes or cannulas. Each clearance period involved perfusion of 70~120 mills of fluid through the ventricular cisternal system. Inflow rate was maintained constant in the range of 1.50-2.00 ml~min, and outflow was measured continuouslyO The data obtained was used to investi-gate the effects of hydrostatic pressure on inulin clearance, rate of formation of CSF, and the permeability of the ventri-cular system, particularly as eompared with that of the toad bladder This ventriculo cis~ernal perfusion method was first rep~rted by Pappenheimer, Heise, 3Ordan and Downer in "Perusion of the Cerebral Ventricular in Unanaesthei~ed Goats", merican Journal of Ph~siolo~y, Vol. 203~ pp. 763-774 (1962~. Pappenheimer et al reported that goats are anatomi-cally and tempermentally suited for ventricular cisternal perfusions and can tolerate such perfusions for many hours without showing signs of disaE fort. The volume of the ventricular system and rate of production of CSF are at least double corresponding values reported for large dogs, and the thickness of the goat occipital bone and its shape facilitates retrograde placement of cannula~ through the occipital bone into or above the ci~terna mayna without ~JU 3-3 interring with muscles in the neck. The goat's horns provide natural mechanical protection for the cannulas and "are almost indispensable~ for operative procedures. In accordance with the PappenheLmer et al technique, guide tubes are implanted just above the dura over the cisterna magna and just above the ependymal linings of the lateral ventricules. Prior to each perfusion the cisterna and ventricle are punctured with sharp pro~e needles extending a few millimeters beyond the tips of the guide tubes. Alternatively, cannulas were implant-ed in the subarachnoid space over the parietal cortex, thus permitting perfusion of the entire ventriculo cisternal-sub-arachnoid system. Pappenheimer et al followed detailed protocols for implanting the guide tubes, and for preparing sterile, synethetic CSF. The Pappenheimer et al perfusion circuit is reported to comprise a bottle sealed with a rubber cap having two stainless steel tubes extending to the bottom of the bottle. One tube serves as a gas bubbler, the second as a liquid outlet. A third opening connects with atmosphere through a sterile cotton plug~ The bottle is mounted on an indicating balance and the reservoir outflow is connected through tubing to a parastal ic pump with a variable drive Permitting pumping ~ates in the range of 0~5-5 ml/min~ One pump output is lead to a male syringe joint which fits the ventricular probe needles and a second outlet on the joints ~5 connects to a strain gauge manometer. A 5ml empty sterile syringe is placed in par~llel with the output to damp pulsa-tions of the pumpO The cisternal outflow is connected to an enclosed drop counter and wing flask and the output is recorded cumulatively on a polygraph which also gives a vertical record proportional to outflow rate. Pappenheimer et al reports that perfusion with CSF of normal composition can usually be maintained for 4-8 hours before the animal become~
resistive, and if correctly performed, the animal will show no sign of knowing when the perfusion pump is ~n or off.
No attempt is made to regulate the temperature of fluid entering the ven'cricular probe, however at flow rates of 1-2 ml/min it is theorized that the fluid reaches temperature equilibrium with the brain before reachin~ the hypothalmus.
At higher flow rates (4 6 ml/min) the animals are reported to start to shiver. In this regard, see also F.~. Sklar and D.M. Long, Neurosuryery 5, 48-56 (1977).
Over the years, many experiments have been conducted with materials possessing high oxygen-Aissolving properties, many of which have been incorporated as s~onstituents in aartificial blood~. The concept of utilizing materials possessing high oxygen-dissolving properties for the mainten ar.ce of t issue respirat ion was irst reporte~l by Rodn ight in 1954. See Rodnight, R~,, ~ Journal, Vol. 57, p.
661. Rodn ight capital ized upon the cons iderable oxygen solubility found in silicone oil~, and sustained tissue slices by incubation in these oxygen laden oils~ ApproxLmately 12 years later~ Clark reported experiments involving the total inunersion of ~mall animal~ in ~ilicone oil~ and flu3rocarbon liquids.. Rats totally immersed in oxygenated silicone cil survived for one hour with no apparent ill eiEfects, but died several hours after removal, from unknown causes. S~milar experiments using synthe~ic fluorocarbon liquids~ which dissolve about 3 imes more oxygen than do the silicone oils, were performed with some success. Vnder these conditions animals survived immersion in oxygenated synthetic fluorocar-bon liquids and thereafter returned to apparent health. See Clark, L. C. JrO and Gollon F., Science, Yol. 152, p. 17550 ~1966); and Gollon, F., Clark, L. C. JrO, Alabama Journal of Medical Science; Vol. 4~ p. 336, (1967). While arterial oxygenation was reported as excellent for Clark's studies in rats, coincident impairment of carbon dioxide elimination was also reported, as was pulmonary damage from breathing fluorocarbon liquids. One rat, which was observed for five days following liquid breathing, was described as being in respiratory distress and as succumbing within 15 minutes -after the subcutaneous administration of hydrocortisone (50 1~ mg), with copious loss of body fluid from the trachea. In this regard, Clark concluded:
These or~anic liquids should prove to be of value in studies of gas exchange in living tissues in animals. Organic liquids, since th~y can support respiration with oxygen at atmospheric pressure and hav~ other unique quali-ties, may find use in submarine escape, undersea oxygen support facilities, and medical applicationO The pulmonary damage caused by the breathing of the organic liquids availa~le at the present time remains a major complication of their u~e in man. Science, Vol, 152~, p. ~75~O
See also Ro~ Tremper, R. Lapin and E~ Levine, Critical Care Medicine 8: 738 (1980 ); S ~A. Gould; A.L9 Rosen, L~R. Sehgal, -F _ Proc. 40:2038 (1981).
Following these observ~tions, fluorocarbon liquids 3S were used as an incubation medium for isolated rat hearts.

TJU 3-3 ~ 944 See Gollon and Clark, The Physiologist, Vol. 9, pO 191, ~1966). In this work, myocardial oxygen requirements were apparently well met, however these hearts did not flourish without intermittent fluorocarbon removal and washing with oxygenated, diluted blood. This phenomenon has been explained in terms of aqueous phase lack in pure fluorocarbons such that neces~ary ionic exchange is impeded.
More recently~ considerable attention has been di-rected to the use of fluorocarbons as constituents o ar~i-icial blood. Sloviter, in order to overcome the problem of aqueous-metabolite fluorocarbon insolubility, made an emulsion with fluorocarbon and albumin. Sloviterls emulsion sustained the isolated rat brain by a vascular perfusion as well as did an erythro~yte suspension. See Sloviter, H~A. and Kamimoto T., Nature ~London), Vol. 216, p. 458 (1967) J A better emul-sion w~s later developed comprising a detergent, s'Pluronic F
6~W (manufactured by the Wyandotte Chemical CorpO, Wyandotte, ~iichigan), and fluorocarbon liquids which were properly emulsified using sonic energy. This improved emulsion per-mitted the replacement of most of the blood of a rat which was then reported as surviving in an atmosphere of oxygen for five to six hours. See Geyer ~Survival of Rats Totally Perfllsed with ~ Fluorocarbon-Detergent PreparationU, Perfusion and Preservation, edited by V.C. Normen, NoY~
Appelton-Century~Crofts, pp~ 85-96 (1968), Geyer, R~ P., Federa~
ti~ Pr~cee2L~, Vol. 29, No. 5; Septem~er-Qctober~ 1970;
and Geyer, ~. P. Med u ~nohn, Vol~ 11, p. 256 (1970).
~ .
Experiments have also been reported wherein ~luoro~
carbons have been used to perfuse liv2rs~ Ten hours after TJ~ 3-3 in vitro fluorocarbon per:Lusion, the isolated liver Al~P;
AMP; lactate/pyruvate ratio; and a number of other metabolites were found to be as good or better than livers perfused in vitro with whole blood. See Krone W~ Huttner, W. B., R~lpf S~ C. et al., Biochemika et Biophysica Acta, VolO 372, pp.
55-71 (1974). These detailed metabolic studies indicated that the organs perfused with 100% fluorocarbon liquid were redeemed ~intact~; while only 75% of the whole blood infused organs maintained a similar degree of metabolic integrity.
The ability of fluorocarbon perfusion to maintian cellular integrity was confirmed by electron-microscopy studies1 The cells had normal mitochondrial ultra structure af~er ten hours of fluorocarbon support, indicating the persistence of nonmal or adequate aerobic metabolism. In Brown and Hardison, nFluorocarbon Sonicated as a Substitute for Erythocytes in Rat Liver Perfusion~, Sur~ery 71, pp. 388-394 (1972) a fluoro-carbon perfusate preserved organ function and integrity far better than perfusate with much lower oxygen carrying capacity, but was reported as resulting in a decreased ra~e of bile secretion which was probably the earliest sign of hepatic d~mage, tissue edema, and a reproducible rise of portal pressure over a period of 2 1/2 to 3 hours~ Both ti~sue edema and rising portal pressure with fluorocarbon perfusion were a~sociated with progressive vascular occlusion as determined histogolically~ A greatly dimini~hed perfusion of fluorocarbon at the end of experiments was documented by injection of India ink twenty minutes beforé the end of the perfusion. Brown and Hardison hypothesized that ~he fluoro-carbon per~usate may react w.ith amino acids and proteins~

that the oxygen concentration in the fluorocarbon perfusate may affect the perfusion results, and that filtration of the fluorocarbon emulsion through filter paper and differ-ing instrumentation were responsible for the apparently conflicting results in the literature. Brown and Hardison hypothesize that phagocytosis of fluorocarbon particles might completely block reticticuloendotheilial cells in liver or that capillary endotheilial damage may be another reason for late fluorocarbon perfusion problems.
Fluorocarbons have also been used in experiments involving cerebral blood circulation. In Rosenblum's studies, mouse hematocrits were reduced to 10-15 by ex-changing the animal's blood with a fluorocarbon solution.
When the animals wexe respired with 100% oxygen after intravascular fluorocarbon infusions, the brains remained metabolically sound. These organs were able to reverse rising NADH levels and EEG abnormalities induced by short period nitrogen inhalation. The BEG's of fluorocarbon treated animals could be activated by the central nervous system stimulant netrazole. By these criteria, intra vascular fluorocarbon does support the cerebral micro-circulation and provides functions of oxygenation, metabolism and electrical activity which are normally associated with blood transport. Please refer to Rosenblum, W. I., "Fluorocarbon Emulsions and Cerebral Microcirculationl', Federation Proceedings, Vol. 34, No. 6, p. 1~93 (May 1975). See also S. JO Peerless~
R. Ishikawa, I. G. Hunter, and M. J. Peerless, Stroke 12, pp. 558-563 ~1981); B. Derks, J. Creiglstein, H. H~ Lind, H. Reiger, H. Schultz, J. of Pharm. Method 4, ~ ~, TJU 3-3 ~ 9~

pp. 95~108 (1980); J. Suzuk:i, T. Y. Oshomoto, S. Tanaka, K.
Moizoi, S. Ragawa, Current Topics 9, ppO 465-470 (1981).
As reported by Kontos et al, the marked vasodilation of small cerebral surface arteries which occurs in response to acute profound hypoxemia may be locally obviated by per-fusing oxygen equilibrated fluorocarbon into the space under the cranial window. See Kontos, H. A., et al, ~Role of Tissue ~ypoxemia in Local Regulation of Cerebral Microcircu-lation~, American Journal of Physiolo~y, Vol.. 363, pp.
582-591 ~1978). Kon~os et al descri.bed the effect of per-fusions with fluorocarbon with 100~ oxygen as resulting from increa~ed supplies of oxygen to the neural cells and consequent partial or ccmplete relief of hypoxia, ra~her than to a local increase in the oxygen tension in the immediate environment of the vascular smooth muscle of the pial arterioles. Two other potential explanations for the observed action are also suggested in the Rontos et al article.
In 1977~ Doss, Kaufman and Bicher reported an experiment wherein a fluorocarbon emulsion was used to par-tially replace cerebrospinal fluid, with the intention of evaluating its protective effect against acute anoxia. Dos~
et al, Microvascular Research 13, pp. 253 -260 (1977)o Accord-ing to this experiment, systemic hypoxia was produced through one minute of 10Q% nitrogen inhalation. A bolus of oxygenated fluorocar~n p~aced in the cisterna magna immediately prior to nitrogen breathing increased regional cerebrospinal fluid 2 tension by a factor of 5O During the one minute experimental period, the fluorocarbon emulsion provided ~wice as much brain tissue oxygen as was found in saline injected controls. Doss et al found the anticipated regional tissue oxygenation decline attending nitrogen inhalation to be halved by the administration of the oxygen bearing fluorocarbon emulsionO
In spite of the above described experiments, there is yet to be reported any practical therapeutic approach to the treatment of ischemic neurologic tissue, and particularly human ischemic central nervous system tissue resulting from stroke, accident or disease~
Summary of the Invention According to the invention there is provided an oxygenation apparatus for use in treating hypoxic-ischemic neurologic tissue comprising: ~a) reservoir means for containing an oxygenatable fluid which is physiologically acceptable to neurologic tissue (b) oxygenator means for oxygenating said fluid; (c) first flow control means for establishing a circulation of said fluid between said oxygenator means and said reservoir means at a Eirst pre-selected rate, (d) injection means for injecting said fluid into a cerebrospinal pathway; and (e) second flow conkrol means for establishing a flow of said fluid through said injection means at a second preselected rate.
The present invention is closely related to the invention described and claimed in our previous Canadian patent application Serial No. 374,217 filed on March 30, 1981. Some of the disclosure from that application is duplicated in the present application for the sake of completeness. Accordingly~ phrases such as "the present invention" used in this application may be intended to refer ~o the invention of the previous application. Any confusion can be avoided simply by referring to the claims of the present application~, Applicant has recognized that there is a thera-peutic time window through which neuron can be reached and resuscitated. The method of the present invention is designed to bypass obstructed vascular circulation and deliver cerebral metabolic needs through an alternate cerebral spinal fluid (CSF) circulation portal. Since particle size exerts a major influence on brain pene-tration from CSF, the method of the present invention is hypothesized to permit diffusion of oxygen, glucose, electrolytes and essential amino acids into ischemic neural tissue when presented in abundance in the cerebral spinal pathway. Thus, a rapidly exchanging cerebral spinal fluid perfusion system is provided to amply - 12a -supply these materials and, at the same time, remove metabc)lic waste .
~he cerebrospinal fluid (CSF) pathway system, which int~lTately bathes and permeates brain and spinal cord tissues~
constitutes a unique anatomical relationship within the body, Although it has some similarities 'co systemic lymphatics, its anatomical arrangement differs considerably from that of lymph~
Indeed, this system bas been named the ~third circulationa.
Due to the extensive area of CSF-tissue contact over the cere-bral and cord surfaces, in the miniature Virchow-Robins spaces, and cerebral ventricles I the cerebrospinal fluid ~ystem constitut~s a vast, complex and intimate therepeutic avenue for access to central nervous tisslJe . Excepk ing certain infections and neoplasms where the cerebrospinal fluid is now utilized as a treatment conduit, the cerebrospinal fluid system has not been otherwise widely exploited as an easily accessible therepeutic route and has never been used a~ a continuous therepeutic diagnostic circulation system in man. The present invention is predicated on the recognition that9 when regional ~erebral blood flow is interrupted~ such as after major ~troke, or is otherwise ~eriously impeded by proound vaso-spastic states, the cerebrospinal fluid path-way actually represents the only practical and viable anatomi-cal roulte by which these tissues may be readily treated.
This results from the fac~ that ~he usual vascular del ivery system is e ither occluded or non-funtlonal ~ ,and thus tis5ue5 within ~ffected erritories cannot be properly served~

In accordance wit:h the present invention, essential cellular substrates are delivered to beleaguered ischemic brain regions by utilizing the ~back door" cerebrospinal fluid delivery route. Accordingly, the present invention provides a novel nutrient emulsion, circulatory method and system which provide necessary nutrient penetration in~o regions suffering vascular deprivation.
It has been found tha~ the cerebrospinal fluid to brain relationship is not characterized by the rigid and highly selective barrier mechanism which are present at the blood-brain interface. Thus, the penetration rate of materials from cerebrospinal fluid regions to the brain relate largely to molecular size, that is, small substances penetrate deeply while large molecules move more slowly into brain substance.
Aithough entry rates are generally inversely proportional to molecular weight, penetration is also influenced by lipid solubility and the molecular configuration of the penetrating substance. Accordingly, the present invention provides a nutrien~ emulsion containing essential brain nutrients in-~luding selected electrolytes, having a relatively low molec-ular size which~ in accordance with the methods of the present invention, are caused ko relatively freely diffuse from either the ventricular or subarachnoid fluid regions into the brain matter to be treated. Accordingly; the present invention provides a novel nutrient emulsion which has been purified/
balanced, and perfected to fall within narrow physiologic limits while nonetheless providing the desired nutritional characteristic~ referred to above '`1 In accordance with the preferred embodiment of ~he present invention, this nutrii.ent emulsion constitutes "synthe-tic cerebrospinal fluid~ comprising preselected electrolytes, glucose, amino acids, at least one oxygen-carrying component, typically a fluorocarbon, and o~her components which impart to the composition a preselected pH, buffering capability, and osmolarity. This nutrient emulsion is prepared by con-trolling sonication time and by properly dialy~ing the mater-ials to achieve a toxic free emulsion~ The resul~in~ solution may be rapidly oxydated to 2 pressures of 650 mm of mercury by using the herein disclosed modified recirculating pediatric oxygenator. As a result, a novel oxygenated nutrient emulsion is provided which is believed to exhibit exceptional therapeu-tic properties.
The present invention also provides a novel method and apparatus for circulating ~he oxygena~ed nu~rient emulsion through cerebrospinal fluid pathways, particularly those pathways which contact brain and spinal cord tissue. Accord-ing to these methods, treated tissues exhibit a substantially improved ability to resist and/or repair damage which would otherwise result from vascular occlusion. In accordance with ~he preferred method of the present invention, the novel oxygenated nutrient emulsion is circulated through this cerebrospinal fluid route by injec~ing it into brain ventricles and withdrawing it from the cisterna magna or the spinal subarachnoid ~pace to nourish and to treat central TJU 3~3 1~ 88944 nervous tissues. In other instances the fluid may be injected into the subarachnoid space and withdrawn from another subarach-noid position. The preferred embodiment oxygenated nutrient emulsion should be circulated to tissues to be treated in amoun~s sufficient to provide adequate gas exchange, Pure fluorocarbon may contain 50 ml 2 per 100 ml at one atmosphere oxygen while normal blood con~ains only 20 ml O2/100 ml under the s~me conditions. The o~ygen carrying capability per ml of the final emulsion i5 considerably less than that of pure fluorocarbon by reason of its content of other con-situents for normalizing osmotic pressure, buffering, elec-trolytes, and other physiologic balancing materials. Thusl the preferred em~odiment nutrient emulsion may be charged with oxygen (100~ 2 at one atmosphere) to attain PO2 tensions of 640-700 mm of mercury and an 2 content of 20 ml per 100 mlO Under rapid circulation conditions, the integral 2 exchange (fluorocarbon to tissue) has been found to be about 33~. Thus, an oxygen exchange value of about 6~6 ml O2/100 ml n~trient emulsion per minute is provided by the present method.
In accordance with the preferred embodiment of the present invention, sufficient nutrient emulsion should be supplied to counteract oxygen deprivation to the affec~ed tis-sue. For example, the entire supertentorial adult cat brain weights 12 grams (+ 2) and the normal metabolic consumption of oxygen of ~ammalian brain tissue equals 3 4 ml per 100 grams per minute. This total metabolic need may be met wi~h the cir-culation rate of 6-8 mls per minute. Metaboli~ needs necesszry to simply ~ustain ~nd/or salva~e tissue may be achieved by ~ `

perfusion rates of one half or less of this optimum. Within these constraints an easily achieved sustenance flow rate of at least 20-30 ml/minute, optimally 45-60 ml/minute, would be ~nticipated to salvage 100 gms of human brain tissue.
It has been found experimentally that it is possible to supply ~ufficient oxygen to counteract the deprivation of the affected tis~ue through circulation of the nutrient emulsion through the cerebrospinal fluid route. In fact, under carefully controlled conditions, it is believed within the scope of the pre~ent invention to nourish the entire human brain using the preferred embodiment apparatus, method and substance of the present invention. In this manner, central nervous neurons deprived of major blood supply may be sustained without significant damage.
~5 In accordance with the preerred embodiment of the present invention, a novel system is disclosed for administer~
ing and maintaining the oxygenated nutrient emulsion for deliv-ery and circulation through the cerebrospinal route.
~ The preferred embodiment system of the presen~
invention effectlvely carries out the circulation and equili-bration of the nutrient emulsion during treatment. This system, which is diagrammatically illu5trated in Fig~ 1 r generally compri~es a reservoir containing nutrient emulsion;
means for delivering the nutrient emulsion at preselected flow rates, an oxygenati~n mean~ for equilibrating the nu-tri~nt emulsion to desired gaseous tension levels; heat exchanger and/or cooling unit means for selectively control-ling the temperature of the nutrient emulsion; filtering TJU 3-3 ~L ~ 8~

means for cleansing the nutrient emulsion7 and eirculation monitoring means for insuring that desired circulation flows and pressures are maintained within the system.
The present invention also provides a method of diagnosing conditions of neurologic tissue in mammals~ This novel method generally comprises providing an artificial spinal fluid of known compositionl injecting that artificial spinal fluid into at least a first portion of the cerebro-~pinal pathway of a mammal, withdrawing a diagnostic fluid from 8 second portion of that pathway to create a circulation of fluid at least through a portion of said pathway, moni-toring the composition of said diagnostic fluid, and comparing for at least a selected difference in the compositions of said artificial spinal and diagnostic fluids, whereby the detected differences in those compositions are at least diagnostic of neurologic tissue disposed along said portion of the cerebro-spinal pathway. In accordance with the diagnostic methods of the present invention~ the diagnostic fluids may be mon-itored or diferences in oxygen content, lactic acid concen-tration, carbon dioxide concentration, potassium and/or ~odium ion concentration, en~yme concentration, pH diference~
ammonium concentrations, GP~BA (gal[ma~aminobutyric acid3 and other amino acid(s) concentratîons~ microorganism content~
hacterial count, myelin fragments; cellular fragments or organelles, malignant cells, and~or poisonsO
It is also within the scope of ~he present invention ~o provide a novel nutrient liquid and~or diagnostic liquid for treating cerebro~pinal tissue containing various novel specified ~ lB -TJU 3-3 ~ 9 ~ ~

components which is formulated using novel methodology.
It is additionally within the scope of the present invention to provide a novel apparatus for treating patients having ischemic-hypoxic tissues, including novel injection and withdrawal means comprising a novel catheter means which is particularly adapted for injecting oxygenated nutrient liquid into a cerebral ventricle without danger of substan-tially damaging neurologic tissue in the vicinity of that ventricleO
In addition to the methods described above, it is within the scope of the present invention to provide additional therapeutic agents to the nutrient emulsion, such as antineo-plastic agents; antibiotics, and/or other therapeutic agents for use in treating the target tissue(~).
Accordingly, the primary object of the present is the provision of a method, substance, and system for pro-viding early stroke trea~ment.
Other objects of the present invention are to provide treatments for brain and spînal cord in~uries, cerebral hem-orrhage, cerebral vasospasm, senility, after general hypoxiaand other hypoxic-ischemic related neurological disorders~
It is a further Gbject of the present invention ~v provide therapeutic treatment which may sustain the lie of the brain and central nervous system tissues in case of Z5 profound shock and/or temporary cardio-respiratory failure.
It is a further object of the present invention to pro~ide life-sustaining support to the brain a~d/or spinal cord TJU 3 3 ~ 9~ 4 tissues during the conduct lof neurological or cardiovascular surgery~
Other objects of the present ~vention are the provision of methods which may complLment treatments of central ne rvous sy s tem ne opl asms by e i the r e x te rn a l rad iat ion an d chemotherapy by providing local tissue hyperoxygenation or drugs which may enhance drug or radiation tumorocidal effects.
Further ob jects of the present invention include the provision of methods which are useful in treating anoxic states attending birth injury. The present method will also assist in removal of central nervous system poisons.
These and other objec~s of the present invention will become apparent from the following more detailed descrip-tion.

rief Description of the Drawin~s FIG. 1 is a diagrammatic view ~f the preferred embodiment treatment system of the present invention illustrat-ing the circulation of nutrient emulsion from a reservoir, into a cerebral ventricle, such as a lateral ventricle, ~o through a portion of the cerebrospinal fluid pathway for output from the spinal subarachnoid space or from the cisterna magna;
FIG. 2 is a diagrammatic view of a portion of the pre-ferred em~odiment ltreatmerl~ system of FIG. 1 illustrating an alternate circulation route where in oxygenated nutrien~

emulsion i~ injected in~o the spinal subarachnoid space and is collected fro~ the cisterna magna;

i- ~0 --TJU 3-3 ~ 4 FIG. 3 is a diagr.~mmatic view of a portion of the preferred embodimen~ treatment system illustrated in FIG. 1.
showin~ an alternate circula~ion route wherein oxygenated nutrient emulsion is injected into ~he cisterna magna for S passage through the spinal subarachnoid space for withdrawal ~rom a lumbar region;
FIG. 4 is an EEG power recording ~rom the left and right hemi~pheres of a cat showing traces from the time of an initial stroke, at the end of the stroke, and four hours after the stroke,o FIG. 5 is an EEG recording of an animal perfused with oxygenated nutrient emulsion having a P02 level of 400 and showing a 5~ return of EEG at 4 hours;
FIG. 6 is an EEG similar to FIG. 1 for ~n animal perfused with oxygenated nutrient emulsion having a P02 of 645 and showing an 88% return of electrocerebral power within 4 hours;
FIG. 7 is an EEG trace showing the effect on EEG
activity of a temporary cessation in oxygenated nutrient emul-sion circulation;
FIG~ 8 is a graph showing the effect on glucose metabolism ~CM~Gl), lactate and pyruvate before and after stroke of a perfused animal particularly illustrating the effect of a reduction in p~rfusion rate to insubstantial levels~
EIG. 9 is a bar graph showing the mean EEG recovery (percent) for groups of cats subjected tG strokes resulting in 15 minutes of EEG isoelectricity~ and co~paring naive animals to those perfused only with artificial cerebral spinal fluid (lumbar) and oxygenat:ed nutrient emulsion through lum~ar and cisternal routes;
FIG. 10 is a graph of microequivalents of p~tassium 5per minute versus time for ~wo experimental groups of cats subjected to 15 minutes of stroke induced isoelectricity;
FIG. 11 is a graph similar ~o FIG. 10 wherein the data in FIG. 10 is represented as a percent of the base line figure;
FIG, 12 is a glucose me~abolism (CMRG1~ graph plot-10ting milligrams per grams per minute against time for three perfusions using a standard gluccse concentration, one perfusion using twice that glucose concentration, and a control using artificial cerebral spinal fluid withou~ fluoro-carbon;
15FIG. 13 is a diagrammatic view of an alternate em-bodiment oxygenated nutrient emulsion delivery system or use in performiny the methods of the present invention.

escri~tion of the Preferred Embodiments 20In the ollowing more detailed description, numerous examples have been selectea for the purposes of explanation and illustration of the preferred embodiments of the present in-vention. One of ordinary skill in the art will recoynize that various changes may be made in the materials and methods 2Sdisclosed herein without departing from the scope of the present invention~ which is defined more particularly in the appQnded claims, Referring now to Fig. 1~ the preferred system for circulating nutrient ~mulsion through a cer~bro~pinal pathway - ~2 ~

TJU 3-3 1~889`~ 4 is diagrammatically illustrated. As shown in Fig~ 1, a nutrient emulsion reservoir 10 is provided for receiving and retaininy nutrient emulsion, the preparation of which will be described more fully hereinafter. In accordance with the preferred system and method of the present invention, the nutrient emulsion is injected into a cerebrospinal path way following pH adjustment and filtering, temperature adjust-ment, oxygenation, and adjustment of the pressure and flow rate of the nutrient input stream. In Fig. 1, these steps are illustrated diagrammatically at 12, 14, 16 and 18 respec-tively. Preferablyp the nutrient input stream is delivered to a ventricle of the brain, and more particular to a lateral ventricle 20 of the human brain, designated generally 2~
In3ection of the nutrient input stream permits the oxyg~nated nutrient emulsion to come into contact with the subarachnoid spaces, miniature Virchow-Robins spaces, cerebral and cord surfaces, and cerebral ventricles. For the system illustrated in Fig. 1I the nutrient input stream is diagrammatically i~llustrated as being injected into a lateral ventricle 20. Since the lateral ventricle i5 in fluid communication with other portions of the cerebrospinal pathway, withdrawal of fluid from a portion of the pathway which is remote from that ventricle will create a circulation of fluid within the cere-brospinal pathway. ~ore par~icularly circulation of ~he nutrient input stream though at least a portion of the cere-brospinal pathway may be accomplished by withdrawing fluid from the spinal ~ubarachnoid space, dia~rammatically illus-trated as 26 in ~ig~ 1, or alternatiYelyl from the cisterna ~ 23 '` ~

magna 24.
It is not necessary to conduct steps 10-18 in the sequence illustrated in FIG. 1. In FIG. 13 the presently preferred apparatus for delivering oxygenated nutrient emul-sion is diagrammatically illustrated. Thi~ apparatus, which may be easily constructed using a pediatric blood oxygenator such as an ~-800 Pediatric Oxygenator available from The William ~arvey Cardiopulmonary Division of C.R. Bard, Inc., Santa Ana, California 92705, comprises a nutrient emulsion reservoir having oxygenation and temperature adjustment loops for constantly oxygenating and adjusting the ten~pera-ture of the nutrient emulsion contained within the reservoir.
In this manner the flow rates of nutrient emulsion provided from oxygenation by oxygenator 102 or for tempera~ure adjust-ment 104 may be independently varied ~hrough adjusting the flow rate of delivery by variable speed pumps 103 or 105 to optimi~e the temperature and PO2 chara~teristic~ of the oxygenated nutrient emulsion to be delivered for injection by variable speed delivery pumps 106 and 107. As normally used, pediatric blood oxygenators ail to provide a sufficient oxygen transfer rate to fluid flow rate to accommodate the emulsion of the present invention. The minimum blood flow rate of the H-800 oxygenatorf for example, is 0.5 liters per minute, and the oxygQn transfer rate (to blood) at this flow rate is less than about 25 ml~min~ By routing the outpu~
101 of th~ oxyyenator to the re~crvoir~ the oxygenator output pump 103 may operate at flow rates which ea~ily achieve about 7 liters per minute of oxygen transfer to the fluoro-TJ~ 3-3 ~ ~ 8 ~ 9 ~ ~

carbon emulsion contained in the 2000 ml reservoir. At the same time, delivery pumps llD6 and 107 may provide much lower flow rates of nutrient emulsion to the animal undergoing treatmentO In a similar manner, heat exchange may also be optimized. In order to maintain optimal P02 values, each conduit of this system should be composed of an o~ygen imper-meable material to prevent leakage of oxygen from the oxygen-ated nutrient emulsion during processing and deliveryO The filtration and chemical balancing procedures followed in preparing the nutrient emulsion are no~ presently performed non line", however it is anticipated that chemical balancing may be performed as a closed loop process, as illustrated in FIG. 13~ Filtration 108 is performed on line under pressure from pump 106 using a millipore bacterial filter~ Pump 107 establishes the final injection rateO The flow of nutrient emulsion to the chemical balancing system i5 adju~ted using variable speed pump 111. In the embodiment of FIG. 13, pressure monitoring and control is accomplished usiny an open side arm 114 ~earing indicia thereon which correspond to the hydraulic pressure of oxygenated nutrient emulsion within delivery line 19, The hei~ht of the ~ide arm is adjusted so that overflow will occur when the maximum desired intracranial pressure has been obtained.
As shown in FIGo 1 the oxygenated nutri~nt emulsion input stream i~ carried through input ~tream conduit 19 to an injection canula 20a which is coupled thereto by coupling 210 Injec~ion canula ~Oa is rigidly attached to skull 22 by fitting 22a which holds the canula in its proper orientation to permit injection of the oxygenated nutrient ~mu~sion into lateral ventricle 20 ~ 25 TJU 3-3 ~ 94 4 If preferred, a double lumen catheter, such as catheter 120 (FIG. 13), may be utilized in place of input canula 20a. One of the lumens of this catheter should be connected t~ a pressure monitoring means for monitoring the S intracranial pressure within the lateral ~ntricle 20~
This pre~sure monitoring means may comprise an open side armr such as side arm 122 which functions similarly to ~ide arm 114.
The preferred injection means of the present ~0 invention comprises a cerebral catheter means for insertion into a brain ventricle. This injection means comprises means for preventing a portion of the catheter located within a brain ventricle from damagin~ tissues surrounding the ventricle. In the preferred embodiment, an inflatable balloon tip may be provided for this purpose. The actual injection of nutrient emulsion into the brain Yentricle is accomplished ~y provid ing an arrangement of outlet holes disposed as a series of sl its rad ially spaced around the catheter tip. Both the injec~ n means and withdrawal means also further comprise attachment means for attaching the catheter to the body in the vicinity oiE the injection or wit!hdrawal sites~ Thu~ the injection catheter may compri~e a means for fixing at least a portion thereof with respect to the ~kull to in~ure catheter stability. The withdrawal catheter, which may have a tip with ~5 multiple perforation~ disposed therein, :Eurther comprises means for attaching at least a portion thereof to tissue in the r~g i~n of the ~ubarachnoid space . This a~tachment means may include a staple for attach ilng a non ~collaps ible o the ~ 2~ --catheter to a lumbar region of the skinO
In many applicat:ions, the oxygenated nutrie~t emulsion will be delivered under normothermic conditions, that is, at about 37 C. Under these conditions, and under hypot~ermic or hyperthermic condîtions where the delivery temperature o oxygenated nutrient emulsion is higher than ambient temperature, temperature ~djustment is easily accom-plished by providing a ~hermost~tically controlled he~ter coupled to a suitable heat exchanger for adjusting the t~m~
perature of oxygenated nutrient ~mulsion recirculated to the nutrient emulsion reservoirO
The circulation route illustrated in Figure 1 permits the treatment of at least cerebral tissues. It is within the scope of the present invention, however, to focus treatment on selected neural tissue areas, in which case al~ernative points of injection and withdrawal of fluid may be ~elected by the attending physician. For example, in the case of spinal cord injury, it is anticipated that the point of in~ection of oxyg~nated nutrient emulsion may be the lumbar, spinal subarachnoid space, with the point of with-drawal being the cisterna magna. While the above mentioned cerebrospinal pathway injection and withdrawal p~ints are preferred,it is within the scope of the present invention ~o utilize other injection and wi~hdrawal locations, provided

2~ a ~ubstantial circulation of fluid through the area of af~-fected neurologic tissue is established by uti~ izing the selected loci. Such alternate pathways are illustrated in FIGS. 1~3. In FIG. 1, withdrawal ~f ~he nutrien~ emulsion from the cisterna magna is illustrated via conduit 30 in dotted TJU 3~3 outline. In FIG. 2 input conduit 19 injects oxygenated nutrient emulsion into the diagrammatically illustrated subarachno~d space 26. Withdrawal from the cisterna magna is via conduit 30b. In FIG. 3 injection into the cisterna magna is accom-plished via injection ca~heter 3~a. Withdrawal is from thediagrammatically illustrated spinal subarachnoid space 26 via withdrawal catheter 30c.
The fluid which is withdrawn from the cer~brospinal pathway will not be of identical composition ~o the oxygenated nutrient emulsion which is injected at the injec~ion point.
By taking advantage of differences in composition which are ~etected in the withdrawn fluid, which may be considered to be a diagnostic fluid, the attending physician may easily monitor the physiologic condition of the neurologic tissue which is being treated. This diagnostic 1uid may also be monitored to assure ~hat treatment is proceeding according to plan. Accordingly, fluid which is withdra~n from the cerebrospinal pathway i5 directed to an output collection mean~ 28 or collecting diagnostic fluid. Preferably, an output monitor 34 will continuously monitor various chemical and physical characteristics of the diagnostic fluid for such properties as flow ra~.e, hydraulic pres~ure, potassium and sodium ion concentration! ~emperature, .lactic acid concentration, gamma amino butyric acid and other amino acid concentrations, oxygen ooncen~ration, carbon dioxide concentration, en2ymes, and ammonia concentra^
tion. The outpu~ of this ou~put monitor will not only provide the attending physician with information concerniny the TJU 3-3 ~ 4 ~

state of ~he cerebrospinal tissue being treated, but also will be fed back to the mo:nitor, control and alarm systen,s for at least pressure and flow rate, temperature, oxygen-car-bon dioxide and chemical constituency, as described more fully hereinafter. This diagnos~ic system takes advantage of ~he fact that ischemic neurologic tissue produces higher concentrations of such materials as Gamma-aminobutyric acid (GABA), lactate ion (lactic acid~, enzymes and/or LDH (lactic dehydrogenase~, ammonia, and other constituen~s which have been determined by analyzing cerebrospinal fluid of patents subjected by disease to similar anoxic conditions.* In accordance with the system of the present invention, however, a continuous monitoring of the state of neurologic tissue is possible, since the circulation of oxygenated nutrien~ emul-sion will produce a continuous flushing of the affec~ed tissue regions, and thus will result in diagnostic fluid c~mponent variations which are rapidly reflective of the physiologic state of the tissues being treatedO Due to the multipoint injection-withdrawal method of the present inven-tion, dangers which are inherent in sampling natural cerebro spinal fluid at a single location are avoided by utilizing a double venting method wherein the cerebrospinal fluid pressure is at all times carefully controlled, ~See for example, ~Rapid and Sensitive Ion-Exchange Fluori-metric Measurement of ~-Aminobu~yric Acid in Physiological Fluids~, ~are et al, Anal. Biochem. Vol, 101, pp. 34g-355 (1980~ for a preferred GABA measurement methcdD

~. ~9 _ TJ~ 3-3 It is within the scope of the present invention to ~terili~e and r~constitute that diagnostic fluid as shown at step 32, whereupon that reconstituted diagnostic fluid may be provided as nutrient emulsion to the nutrient emulsion reser-voir 10. As shown in FIG~ 1, the output monitor 34 ~ay monitor the diagnostic fluid during ~he sterilization and reconstitution processes and, if desiredr ensure ~ha~ ~he reconstituted fluid satisfies the requirements of the n~ient emulsion reservoir~ As shown in FIG. 1, in order to ensure that appropriate degrees o:E oxygenation ~ filtration and chemical balancing, temperature ad~ustment, and pressure and flow rate are maintained, the nutrient input stream is moni-tored by various monitors, controls, and alarms, which are intended to proYide a fail safe nutrient input stream. In particular, a pressure and flow rate monitor, control and alarm 38 is provided for monitoring the pressure and ~lcjw rate of the nutrient input ~tream, for controlling the press-ure and flow rate adjus~ment 18 to establish desired press-ures and flow rates, and for sounding an ~larm in the event that the nutrient input stream exceeds or falls below pre-selec~ed pressures or flow rates~ Xf desired, this alarm may additionally disable the pu~ping mechanism producing flow of the nutrient input stream ~uch that the unit Nshuts down" upon detection of unacceptable input stream conditions.
Referring now to the temperature monitor, control and alanm~ the tempera~ure characteristics of ~he nutrlent input stream are similarly de~ected, at leas~ to ens~re that hyperthermic sta~es~ except when used as ~her~peutic ~ :~0 --modality~ are avoidedO While in most instances, the nutrient input stream will be adjusted to a 37~ C~ temperature, it may be desired to select hypothermic te~peratures in order to establish certain treatment conditions~ In either event; the temperature monitor will continuously detect the temperature of the input stream, will control the ternpera-ture adjustmen~ 14 to establish a preselected temperature~
and will sound an alarm and/or disable the system in the event that a preselected temperature range is not main-tained in the nutrient input stream.
Referring now to the chemical monitor, control and alarm 42, the nutrient input stream will be continuously monitored for one or more chemical or physical characteris-tiC5 of the nutrient input stream, and will control the chemical balancing, flltration, etcO which is performed by the iltration and chemical balancing unit 12. The chemi-cal ~oni~or, control and alarm may9 for example, monitor the pH, osmolarity~ electrolyte component, carbohydrate com-ponent, amino acid component t or other components o the nutrient emulsion to ensure that the nutrient inp~t stream falls within preselected stream characteristicsO In the event th~t these sharacteristics do not all within the preselected range, the ala-m for unit 4~ may sound and/or may disable the system ~o ~hereby prevent fur~her in~ection of nu~rient input stream into ~he cerebrospinal pa~hwayO
Finally~ an oxygen/carborl diox1de monitor~ control and alarm unit 36 is provided which continuously monitors the oxygerl alld carbon dioxide contents o~ the nutri.ent input stream, which controls ~he oxygenation unit 16, and which sounds an alarm in the e~i~ent that the oxygen or carbon dioxide concentrations do not fall within preselected ranges.
It is anticipated that each of units 36 42 may provide continuous displays of the information monitored from the nu~rien~ input stream, and may~ if desired, enable back-up units which either manually or automatically supplement or replace the functions of units 12-18 in the event that those units are not functioning to produce a nutrient input 1~ stream within the desired ranges. For example, it is anti-cipated that a manual or battery operated pump, oxygenator, filter, and pressure and flow rate adjustments be provided to enable emergency operation of the system/ since continual nutrient flow is lifesaving for the devitalized portion of the treated organ~
The preferred nutrient emulsion of the present invention is comprised of carefully formulated components which~ to the extent possible while maintaining desired therapeutic activity/ mimic the physical and chemical char-acteristics oE natural cerebrospinal fluid~ Generally, tissues and cells will not fair well if exposed to large volumes o~ non physiologic ionic solutionsO Accordingl.y~
it has been recognized that appropriate electrolyte compo-sitions at the ki~sue level are indi~pensable when it is oonsidered that the circulatory me~hod of khe presen~.
invention would otherwise result in the washin~ and the dilu~ion of el2ctrolytes from the region even after short texms of circulation, to the detriment of cell membrane

- 3~ -functions. Accordingly, in accordance with the preferred embodiment of the present inv~n~ion, sodium, potassi~m, calcium, magnesium, and chloride ions ~re carefully bal-anced in the nutrient emulsion of the present invention to thereby create; to the degree possible, no~lal extra-cellular campositions. The present in~ention also provides a non--aqueous oxygen transfer component for selectively ~ombining wi~h oxygen and for transferring oxygen to the tissues to be treated. Numerous compounds are known to the lD art which are chara~terized by having a high solvent property for oxygen, carbon dioxide, and other gases. The preferred non~aqueous oxygen transfer component of the preferred nutrient liquid should exhibit when so charged, oxygen vapor pressure ranges of above 400, and preferably 600, ~orr.
SUCh oxygen transfer components should similarly not have in themselves high vapor pressures which would boil at body temperatures, nor have viscosities which are difficult if not impossible to ~mulsify. Generally, the preferred com--pounds for use as non-aqueou~ oxygen transfer components are 1uorocarbon polymers, such as perfluorocarbons~ perfluorin-ated alkyl polyethers, fluoroethers, fluor~lines, etc.
~ile compounds within these groups range in molecular weight from 250 to 7000, their selection for use as non-aqueous transport ccmponents are based upon the combination of fea~
tures of the proper vapor pressure, molecular weightt vis-cosity, and emulsifiability, emul~ion stability and tissue distribution. One such fluorocarbon which has been found to be particularly suited for the non-aqueous oxygen transport 33 ~

~U 3-3 component of the preferred nutrient liquid is a reagent grade perfluorobutyltetrahydrofuran which has been sold by the 3-M Corporatlon under ~:he trademark ~FC-80n. FC-80 has an oxygen solubility coefficient ScO2 of 0O45 of ml 02/ml at PO2 of 760 Torr, See Navici et al., Res. Exp. Med. 170, pp. 169-180 tl977), which paper is specifically incorporated by reference as if fully set forth herein. It should be noted that whole blood under the same circumstances contains 0.23 ml O2/ml. The FC-80 ScO2 is linear from 760 to 200 Torr but declines quite rapidly below the lower levelO The high oxygen diffusion coefficient (5.71 x 10-5 cm2/sec per second) indicates more than adequate FC-gas in a physio-logic sense. Similar studies concerning CO2 solubility and diffusion indicate that absorption and release are described by a straiyht line function. From these observationst metabolic tissue CO2 accumulations should theoretically be easily removed by fluorocarbon solutions administered through a circulatory method.
~ot only do fluorocarbons possess these unique ~o physical gaseous properties but they are for the most part non-toxic. The main acute toxicity has been found to re-side in free fluoride ion accumulation which occurs mainly from sonication. See, Clark et al. 9 Fed. Proc. 34, pp. 1468-1477 ~1979). The free ion czn, however, ~e removed Dy repetitive dialysis and the emulsion thereby rendered physiologically acceptable. ~ccordingly~ the preferred embodlment nutrien~ liquid of ~he present in~en~ion, which has been dialysized and fil~ered ~hrough a millipore ~ilte~
~ 3 has evidenced no toxicity either in short term or lony term use during circulation through cerebrospinal pathw~ys ~f animals. One chief advantage of th~ CSF circulation route is that most or all the nutrient liquid can be removed by washing at the time of treatment termination. In this way long term cellular retention as previously noted for liver and reticuloendothelial cells in vascular circulations of oxygenating liquids may be avoided.
In the preferred embod iment nutrient 1 iqu id of the present invention, an emulsiication component is provided for permitting the emulsification of the nutrient component wi~h the oxygen transfer component of that liquid. See Clark et al, Triangle II, ppO 115-122 (1972b); Clark et al, Microvasc~
_ Res. 8, pp. 320-340 (1974). The best currently available ma-terial for this purpo~e is believed to be block polymer polyols, which are known to the art as ~pluronics~, of which, pluronic F6B has proven to be a most efficlent emulsifying agent. As used in a nutrient liquid as described more fully hereinafterr the toxicity from such a pluronic detergent is negligible, At the presen~ time, however, it is an~icipated that other emulsification components which will permit the non-aqueous transf~r component of the nutrient liquid to become soluble with respect to the aqueous nutrient component of ~he nutrient liguid may be u~ilized to provide solutions ~5 which have adequate physiologic perimeters. Such other means of solubili~ing 1uorocarbons include3 the fo~mation of micelles, etcO
~n ~he prepara~ion of the preferred nutrient liquid, TJV 3~3 ~ 94 4 an important factor in pr:oducing an acceptable nutrient liquid is the achievement of an acceptable final osmotic pressure. The osmotic pressure of the nutrient liquid will depend upon the amount of the emulsification component, the particle size of the fluorocarbon, and the ionic composition of the aqueous nutrient component. In accordance with the preferred method of preparing the nutrient liquid of the present invention, toxic emulsification components should be removed by dialysis. Fluorocarbon particle size will be lD controlled by sonification time and filtering, while the ionic composition of the aqueous nutrien~ component will be carefully adjusted to produce a nutrient liquid possessing desired osmotic characteristics. If desired, a final osmotic tuning may be acccmplished in accordan~e with the method of the present invention by adding ascorbic acid to the nutrient liquidO
In order to provide fully successful treatment of i~chemic tissues, it is desirable to provide nutrient liquid for circulation around those ti~sues which will compensate for relative or complete deiciencies o blood transport metabolites. In addition to oxygen, other tissue metabolic requirements include glucose, amino acids7 icnsS hormones, vitaminsl etc~ While in temporary treabment conditions~ it may be suitable to temporarily omit one or more vitamin, hormone, ion, or amino acid, for prolonged treatment and to produce the most desirable results~ it is preferred to provide ~ubstantially all of the above mentioned metabo-lites in the pref~rred nukrient liguid. It is at least TJU 3-3 ~ 9~ ~

desirable to provide in the nutrient liquid all components necessary to support aerobic metabolism which will be available within the medium for use at cellular levels.
Glucose deprivation of central nervous system tissue causes a serious cellular metabolic deficiency, as does the same degree of oxygen deficiency. Accordingly, by providing a total and finely ad~usted mixture that has all ~he neces-sary components for total cell survlval~ an extremely efficient and therapeutic liquid material is provided which is ideal for circulation through the cerebrospinal pathwaysO
In order to illustrate the preferred method and composition of such an oxygen-nutrient material, the following example is provided.
~
Vnder conditions of replacing blood borne materials by perfusion all nutrients necessary for aerobic metabolism must be available within the medium for immediate use at cèllular levels. As far as the central nervous system is concerned, glucose deprivation causes as serious a cellular metabolic deficiency as does the equivalency of oxygen lack.
To achieve the desired ends all known essential nutrients have been added to he FC (fluorocarbon) emulsion. FC i~self thereby serves the purpose of a gas transport system while ?5 the aqueous emulsion phase contains an array o cellular metabolic essentials. The total and finally adjusted mixture has all the necessary ingredients for to~al cell survival~
The combination material is referred to as an oxygen-nu~rient formula (Ox-N), or oxygenated nutrient emulsion.
Method and Composition Preparation of Oxygen-Nutrient Material Reagents 1. (A) 5% Commercial grade Pluronic F68 (~asic Wynadotte).
(B) 20% W/V FC-80 (3M Corporation) (C) Synthetic C~S~Fo Sodium Chloride~ 703 gm/L
Potassium Chloride- 300mg/L
Calcium Chloride (dehyd) - 200mg/L
Magnesium Sulfate- 300mg/L
Sodium Phosphate (hepta) - 200mg/L
Sodium Bicarbonate- l90mg/L
Ad~ust the pH to between 7.380-7.420 with 10% Ascorbic Acid (D) Bacitracin Inj~ 50,000 U/vial (Pharmacy) re onstitute with 10 ml saline to give a concentration of 5000 ~/ml. Use 0.2 ml for each liter of perfusate to obtain a concen-tration of 1,000 units per liter of perfu-sate.
(E~ Essential ~mino Acids (Pool) (5igma~
D-Glutamic Acid- 11.8mg L-Glutamine - 730.0mg DL-Serine - 2603mg D-Threonine - 30.Omg D-Lysine ~ 38.8mg D-Valine (optional)~ l9.0mg D-Leucine - 14~0mg DL-Isoleucine ~ 13.Omg D-Phenylalanine 15.Omg DL-Tyrosine - 1400mg D-Methionine - 4~5mg TJU ~-3 Before oxygenating the fluorocarbon em~lsion add 9.8~mg. amino acid and 200mg dextrose for each 100 ml of emulsion.
(F) Steroid (Me~hylprednisolone sodium succinate) 125 mgs. (The Upjohn Company).
Xeconstitute ~he steroid wi~h 2 ml of diluent to obtain a conc~ntration of 62.5 mg/ml. Add 0.5ml of this m~xture to each liter of emulsion before oxygenation ~31.2mg/L)O
(G) 1 N NaOH
2. Materials [A) Sonifier Cell disrupter (Branson) Model tB) Waring Blender for mechanical dispension of Pluronic Acid.
(C) Dialyzer tubing 7/8 in. (22mm) (Thomas).
It is necessary to dialize the emulsion ~o remove 1uoride ions as well as other low molecular weight contaminants.
(D~ Whitman Filter Paper ~1 (46 x 57~ ~Thomas) The ~mulsion should be iltered to remove particle~ originating from disrupted carbon skeletons o flurocarbon during sonication.

(E) 0.8 micron filter unit (Thomas~. Sterili-zation is acc~mplished by filtering the emulsion throu~h a micro filter.

~c ~ asic Neurochemi.stry ~2nd edition~, Littlet Brown and Company, Boston p. 297 TJU 3-3 ~ 8894 4 (F) CO2 tanh (Welders Supply Company) CO2 is used as a defoaming agent while sonicat:ing r (G) 100% 2 tank (Welders Supply Company) C2 is used as a defoaming agent while sonicating.
~) 100% 2 tank (Welders Supply Company) for saturatiny perfusate.
(I) Sterile Culture Flasks (Thomas) for lQ storing perfusate~
~J) Gas Dispersion Tubes (Fisher Scientific Company~ for equilibrating the emulsion with 2 (K) Aspiratory Bottle (Thomas) a. 250 ml capacity-cut off 2 1/2n from the neck with a glass cutter in order to accommodate the macro~ip for sonification.
b. 500 ml capaci~y -for equilibration of the ~mulsion with lOOml capacity - 100~ 0~.
(L) K50 Extension tubing. Capacity approxi~
mately 2~lml length 40.7 centimeters ~20 in.~
(M~ Circulating Pu~p (N~ Sonification Assembly al Fill a container with crushed ice;
one that will allow drainage of the water a5 the ice melts ~a fish tank will do).
b. On the serrated outlet near the bottom of the aspiratory bottle connect seven lengths (140 in.) of K50 extension tubingO Place the bottle in the ice bath and connect ~he tubing to circulating pump~
c. Place the precoo.led Pluronic acid in the aspirator bottleD Drape and return ex~en~ion tubing from the pump over the side of the bottle. Drape the tubes from the C02 tank over the side of the bottle and bubble slowly.
Carefully lower the macrotip into lS the solution and start soniicationO
3~ Method 20% FC-80 (5~ Pluronic (F68)~ tw~v) (A) Place 25 gms of F68 ~ ~50ml of artificial CSF in a Waring blender and blend at a high speed for 2 minutes. The solution will become very foamy~ For best results the solution should be refrigerated overnigh~ before u~ing~ This allows the head of foam to settle and precools the solutiGn to the proper temperature for sonification.
(B) Place ~he precool.ed Pluronic acid solutlon in the a~pirator bottle. Turn on Sonifier~ With a Pa~eur pipe~te add 58~8ml~ ( 100 gm) of FC 8() over a 30 minute period sonifying throughout. Qnce added ~ Jll --TJI~ 3 - 3 allow the mixture to sorlicate for 45 minutes. Be sure that the temperature doe~ not exceed 20 C.
( C ) Cut d ialyze r tub ing that has been pre~
soaked in artificial C.S.F, into 60-inch strips, Fill each strip half :ull with the m ix tu re . Pl ace ~ t r ips in con t a ine rs filled with approximately 1000 ml, of artiicial C.S.~ Refrigerate and allow to dialiæe for 4B hours, The d ialyz ing solut ion should be changed every twelve hours, and the emulsion checked and transferred to additional tubing since the volume is considerably increased during d ialys is ~
~I~) After di~lysis filter the solution through Whitman $~1 filter paper, then take }he total volume3 25gm of Pluronic acid and 58 750ml of emuls ion c The former volume represents 2û~ FC-80 and 5~ F68 w/v ratio.
The emuls ion ~hould be kept in an ice bath wh ile lprocess ing O
(E) Add ~acltracin to ~he emulsion. The pEI
a~ this point should be be~ween 60 and .5 6,~3 (F ) It is necessary to ad jus~. ~he electrolytes at th is 8 tage ~
Unadjusted electrolytes-Na = 127 K = 5 Cl = 126 C2 1.5 Osmolarity = 271 It is necessary to add 696mg Na C1/L
of emulsion in order to normalize the electrolytes.
Adjus~ed electrolytes:
Na = 131 K = 3.8 Cl = 130 CO~ = 3 Osmolarity = 303 (G) Using 1.0N NaOH adjust the pH to between 7.380 and 7.420, then check the osmolarity (Range 298-317~
(H) Sterilize the emulsion by filtering through 0.8 micron filter. The emulsion can be ~rozen at -20C and is stable for several months.

4. Immediately Before Using Emulsion (A) Add: Glucose 0.8 - 20 5 gm/L
Amino Acid 0.098 gm/L
Steroid 31O2 mg/L (optional) (B) Warm the emulsion to 37C and equilibrate with 100% 2 using a gas dispersion tube for 30 minutes to obtain a P02 of between

5~0-660~

,~"

TJU 3~3 ( C ) A typic:al batch of FC-80 emulsion shows the fol:3 owing properties:
Na = 131 meq/L
X = 3 . 8 meq jL
Cl ~ 13 0 meq/L
~2 = 3 mes~/L
Glucose = 186 mg. %
O smol ar i ty = 311 mOsM
(D) A ~ypical batch of oxygenated nutrient emulsion contains:
Fluorocarbon = 78~6 ml/L
Pluronic Acid - 213 ml/L
Na Cl = 7.3 gm/L
Potassium Cl = 300 mg/L
~5 Calcium Cl (dehydrated ~ = 200 mg/L
Mg S ul f a ~e - 3 00 mg/L
Sodium Phospha~e = 200 mg/L
Sodium Bicarbonate = 190 mg/L
Amino Acid Pool (added to fluoro-carbGn) = 0O 098 ~m/L
Mani~ol In~ect.ion USP 259 ~ 50 ml~L
Bacitracin = 5000 units/L
Gentamicin = 80 mg/L
Dex~ro~e = 2 grn/L
A~corbic Acid = 0.5 ml/L
(10~ ) , S~erile ~ater -- remainder per liter - 4~

Gas Characteristics After Ox~gen Equilibrat]o_ Unsaturated Saturated pH 7~231 7.342 pCo2 3.7 5.7 PO2 190 6~0. 5 In order to provide an indication of the efficacy of the preferred treatment methods, the following examples are provided:
Example 2 For reasons of simplicity and reproducability a model continually in use in applicant's laboratory has been employedO Osterholm, J. L., Pa _ ~y~ y_of S~inal Cord Injury, C. C. Thomas, Springfield, Illinois (1978). Extensive experience with spinal cord injury in terms of standardization, quantitative histological studles, regional blood flow and biochemical parameters suggested these procedures. A primary pathophysiologic event in that model has been determined to be discrete regional ischemia. A microcirculatory flow failure within the injured region has been documented by many study techniques including microangiography, distribution of intravascular particulate materials, hydrogen-platinum flow studies, regional istopic techniques and lactate accumulation. Recent C 14 antipyrine microregional blood flow studies conducted in applicant's laboratory have accurately delineated the rnagnitude of ischemia in the injured cord. Within one hour the regional grey rnatter flow drops from the control of 44cc/lOOgm/min to only 2cc/lOOgm/min. The white matter is also ischemic. Blood flows in these regions are depressed from 15cc/lOOgm/min to 1 2cc/lOOgm/rnin.

i, ~

TJI~ 3-3 From these observ~tions, standarized spinal cord in-jury causes a restricted ischemic lesion which can be easily studied and quantitated. In this rigid system therapeutic treatment effects are readily detected by co~nparison with our extensive untreated inj ury data . It should be noted here that the mechanical injury forces used in these experi-ments are su~stantially above saturation and all ~ounded animals are rendered permanently paraplegicO
Circulation Experiments__ Experiments were carried out by continuously inject ing either saline or Ox-N emulsion saturated with 2 a~ 1 atm into tlle distal subarachnoid spinal space. The ou~flow (withdrawal) of the diagnostic fluid ~as at the cisterna magna. Infusions were begun immediately after severe ~ound-ing~ An infusion rate of 3ml/minute was easily achieved, and this rate was maintained for two hours.
Oxy~en Prior to lumbar spi~al infusion we were able to de-velop P02 tensions of 535 ~ 89mm 2 in the O~-N emulsion ~0 by simpl~ bubbling 100~ oxygen through the solution. Upon exit at the cisterna magna after tra~ersing the entire spinal subarachnoid space the P02 had fallen to 243 ~ 63. The oxygen difference between entering and exit was 292 ~ 639 or a 55% decline~ which is statis~ically ~ignificant at ~he P < 0. 001 level . This f inding indicates a rapid P2 e~.change durin~ the ~hir~y seconds vr less transit time. For various technical reasons our initial P02 was lower ~han can be achieved under idealiYed circumstances. More recently it - ~6 -has been possible to regula:rly attain PO2 of about 650 Torr.
Even better exper.imental results might have now been obtained under conditi~ns of higher 2 tension.
Carbon Dioxide . ~ _ FC-80 is an efficient CO2 exchange and transport agent, and the emulsion therefore easily extracts tissue CO2.
This is indicated by an initial emulsion PCO2 of 2.7 Torr which rose to 16.0 Torr after the tissue perfusion contact. This represents a 593% increase in FC-80 CO~ (P < 0.001). The emul-sion also removes other acid metabolites since in some exper-iments the inherent buffering capacities were exceeded as the exit fluid pH exhibited a considerable depression toward the acid side ~original pH 7~4, exit pH 7DO)- This pH change exceeded an~ acid contribution by the collected CO2, and amounted to 0.248 mole lactate/hour.
A. ~
Frozen tissues were sectioned and s~ained ~H & ~, and acid phosphatase). The sections were evaluated by projection to 25X magnification and preselected lesion parameters mea-sured by means of a compensating polar planimeterO There was considerable increase in the untreated injury cord cross sectional area (12~0mm2) which was significantly reduced in the Ox-N experiments, (8g6mm2). We have assumed t~at this substantial cross sec~ional cord area increase is caused by 2S edema fluid. In the course of other experim~nts~ the degree of edema appearance has been quan~ified. It was found ~hat net water accumulation at tho~e post injury times ranged from - ~7 ~

T;IU 3-3 25% to 40~. The absolute re~:luction in cross sec~ional area b~
the Ox-N treatment is significant at the P=0.001 level.
Lesion Size Using our standard sampling methodology which in-cludes skip serial sections throughout the injury region, and analysis by quantification techniques, the degree of injury induced hemorrhagic necrosis can be determined. With the p~rfected injury system the lesion size at any time point san be reliably predicted. The effects of salîne and Ox-N
circulations upon lesion size were compared to each other and to our established untreated values. The results are summarized in Table I:
T~BLE I
~ESION _SI ZE
2 E~our Injuries % Grey ~ hite % ~otal Standard Injury 79.5 ~ 16% 5D 30.1 + ~ SD 39.5 ~ 10~ SD
~No Infusion) 5aline 78.3 + 154 SD 25.0 -~ 14% SD 34.4 ~ 12~ SD
Circulation Ox-N 47~4* ~ 17~ SD 12.8* ~ 2~ SD lg.2* -~ 10% SD
Circulation Table I ~ Percentages are expre~sed in ~erms of total tissue area lesion~d by hemmorhagic necrosis or grey, white or total cord area two hours after Bevere injury with the various treatments. (~Statistical Significance p c < 0.01~ The saline values are not slgnificant~.

~JU 3-3 The data indicate~i a highly significant deyree of protection against injury lee;ions afforded by the Ox-N circula~
tion treatments. The actual lesions are halved by the treat-ment and this remarkable stabilizing efect upon the important white matter tracts would be anticipated to sub~tantially improve the final functional result attending ~evere spinal cord injury.
Anterior Horn Cells A technique of counting the anterior horn cells which contain visible acid phosphatase histomchemical reaction product has been developed in this laboratory, The procedure has ~een previously used to assess ischemic cellular effects in terms of cellular survival and/or lysis time.
From Table II it can be seen that untreated injury has a highly lethal effect upon anterior horn neurones. Within the two hour experimental time period, more than 97~ of all cells at the injury center undergo cytoplasic lysis. Ox-N
infusions stabilized the injured cells as 60~ of all neurones were protected from lysis.
T~BLE II
ANTERIOR HORN CELLS
Control 34 ~ 2 (SD) Injury 2 ~ 1.73*

Injury + Ox-N 21 ~ 5~12**
circulation Ta~le II- Number of anterior horn cells con~aining acid phosphatase reaction product within well defined cytoplasmic borders~ ~statis-tical difference from control P < 0.001, ~Difference from injury alone P < OoOOl )~
_ ~9 TJU 3 3 ~ 9~ ~R

Spinal Cord Adenosine Triphos,phate (ATP~
Biochemical ATP tissue determinations were under-taken to determine the metabolic oxida~ive state of injured spinal tissues. This metabolite was ~elected for study since it reflects the progress of normal oxidative metabolism.
ATP levels fall very rapidly under sufficient hypoxic-ischemic conditions~ Untreated injured cords have a 200% ATP decline in one minute. In the current e~periments ATP levels would be expected to reflect (1) the cellular oxidative capability and (2) functional cellular viability. The latter aspect is especially important in terms of cellular integrity which was discussed in the preceeding sec~ion.
From Table III it can be seen that 2 hour injury causes a four and thr~e fold drop in grey matter and white matter ATP respectively. This information amply supports other observations about the degree of regional cord tissue ischemia after impaction. ATP was found in significantly h~igher concentration in the Ox-N experiments than noted after saline cir~ulation alone. The hiyh energy compound suffered only a 30% all from normal in the oxygenated per-fusion group which contrasts vividly with the 300 400% loss ound with the saline treatmentsO
T~BLE III

In~y & Saline ~ Control Grey Matter 0.46 1.24* 1,88 White Matter 0~40 0O87~ 1.23 TJU 3 -3 ~ 39~ 4L

Table III - ATP ti~sue levels in control, saline and Ox-N injured cords. The difference between saline and Ox-N is significant *(P =0.05). Although not shown in the Table, the Ox-N trea~ments al50 statis-tically increase ATP in spinal cord regions directly above (P < 0.001) the injury site.
Comparison of the above results to those later reported by R.E. Hanseabout, ~. H~ C. Van Der Jagt, S. S. Sohal, and J.
R. Little, Journal of Neurosu~er~ 55, pp. 725-732 (1981) is of interest. Hanseabout et al report the use of a commercial oxygenated fluorocarbon artificial blood perfusate to treat experimental spinal cord injuries. Treated dogs are reported as sho~ing improved motor function more rapidly and as having a better ~inal hind limb functional result than did controls. To some extent, this non-prior art report confirms the spinal cord injury findings reported here.
Exam~le_3 Cerebrovascular Ischemia Initial studies have been conducted to determine the efficiency of Ox-N emulsions in protectin~ the brain against profound ischemia. We employed the cat brain and utilized right hemispherlc regional vascular interruption o that the left cerebral hemisphere migh~ serve as an internal control.
The middle cerebral artery of cat is accesslble through the bony orbit. It lies immediately above the optic nerve after the canal ha~ been opened and can be identiied with certainty in that position. Preliminary experiments determined that ~`3 ~ .
TJU 3-3 ~ 94 4 an inconstant cerebral field was devascularized by occluding the middle cerebral artery. It became apparent that c~llateral blood flow via the anterior and posterior cerebral arteries supplied some retrograde filling into the experimental region.
This phenomenon could be largely preven~ed by concommittantly reducing the mean systemic blood pressure to 70mmHg by external bleeding. ~emorrhagic hypotension plus middle cerebral artery occulsion yielded a reasonably constant ischemic cerebral lesion from animal to animal.
In that model either saline or Ox-N were circulated from the right cerebral ventricle to the cisterna magna at a rate of 3ml/min. Cerebral tissues were harvested one hour after vascular occlusion by immediate immersion in liquid Freon. The tissues were sectioned in the frozen state and réacted with luciferin upon photographic film. A combination of high energy cellular metabolites plus luciferin react to emit visible light, which is recorded upon the film. Tissues removed from saline treated ischemic cerebral regions were uniquely devoid of phospholuminescence, while the opposite hemisphere demonstrated this reaction to a degree similiar to that found in normal animals. Middle cerebral i~chemic tissue samples from Ox-N treated animals contained sufficient high energy materials to de~onstrate a positive histochemical high energy reaction one hour after vascular arre~t.
~ æ~

The combined evidence rom spinal cord injury and middle cerebral artery occlusion model6 demonstrate that the TJU 3~3 ~ 94 4 preferred oxygenated nutrient emulsion can be circulated to maintain cellular integrity and aerobic metabolism under the stress of profound regional ischemia. A third model was utilized to determine if vascular deprived neurones perfused via cerebrospinal fluid pathways with oxygenated-nutrient would continue to perform a physiologic function. A trans-thoracic aortic ligation just distal to the lef~ subclavian effectively devascularizes the cervical, thoracic and lumbar cat spinal cord. In some examples the lower brain stem was also found ischemic by regional flow studies. The mid and lower ~horacic cord are universally and profoundly blood deprived by this vascular interruption. Animals under light Ketamine anesthesia were treated by circulating from the lumbar subarachnoid space to the cisterna magna with either saline or Ox-N solutions. Respiratory movements we~e evalua~ed in these experimen~s~ The lungs were ventilated by positive presence respiration, but the mechanical movements are easily distin~uished from neuromuscular respiratory contractionsJ
This is especially so since for the most part the respiration and neuromuscular drive occur at separate times and are largely asynchronousO Following the aorta ligation all physiologic neuromuscular respirato~y movements progressively diminished to total cessation after 5-lQ minutes in the saline treated cats. The arrest obtains for inter ostal muscles as well as diaphragmatic contractionsO The Ox-N
treated animals, on the other hand, continue to respire in an essentially normal neuromuscular sequence. The respira~
tion~ under those conditions~ were often o:E irregular rates TJU 3-3 ~ 94 4 diminished in amplitude, and showed some individual magnitude variations The singular difference be~ween saline and Ox-N
circulations is the universal persistence of respiration in the latter group. It is also true that Ox-N sustained suffi-cient chest bellow movements ~o that i the chest were closedthe respirations were clinically adequate to support life.
Example 5 Experiments have also been conducted to determine the efficacy of the herein disclosed methods on global cere-bral ischemia induced in cats.
- Although the Ox-N emulsions of the present invention are oxygenatable by bubbling gas through them, perfusate from stroke animals were initially found to have oxygen pressures ~PO2) below those known efficient oxygen exchange values (PO2 less than 200) for the fluorocarbon camponent of the material.
See Navari et al, ~E~. Accordinglyt the pump oxygenation system described above in connec~ion with Figure 13 was developed to optimize fluorocarbon 2 saturation. As mentioned above, this system comprises a heat exchange~oxygena~or which was coupled to r~circulating, warming and delivery pumps. This ~ystem rapidly oxygenates the emulsion (pO~ =
645[mean] Torr) at 37 C with oxygen gas delivered a~
7 L/min.
Global cerebral ischemia experiment~ were ~onducted on cats after brevital inductio~ and nitrous oxide oxygen (70 30~) anesthesia~ A double lumen inflow cannula of the type described above was sterQtactically placed into a lateral cere-bral ventricle while an exit cannula was inserted either - 5~ ~

9~
, into the cisterna magna or lumbar theca. When the conduits are properly installed, the CSF pathways have li~tle resistance and a mean flow perfusion rate of 6 0 cc/min. can be achieved through the animals without in-tracranial pressure alterations. Entry and exit fluid were collected for metabolic studies. Both gases were normalized by respiratory adjustment. Further experi-mental manipulations awaited electroencephalograph (EEG) normalization. Cerebral ischemia was produced by the combined insult of hemorrhagic hypotension (mean arterial blood pressure lowered to 30 + 3 mm EIg) plus simultaneous carotid artery clamping. This method caused a bihemi-spheric isoelectric EEG within 5-8 minutesO After sustained and total cerebral electro silience for 15 minutes~ the carotid arteries were unclamped and the withdrawn blood reinfused.
A well accepted measure of cerebral function, the EEG, was used to assess both the degree of insult and subsequent discovery. A computer based EEG method, com-pressed spectral analysis, was used to determine the brain activity. A Nicolet Instrument Corporation "MED-80"*
computer utilizing frequency analysis package "Super C'~
was used with the following setup parameters:
2 channels r 1()24 SEC. EPOC~I~ 1024/PT5.EPOCEI
2 sweep average/printout.
The total output is expressed in (microvolts2) assuming a constant source impedene of 1 ohm. The data presented here is the total cerebral power 0. 3-25 Hz in picowatts.
Recordings were made from skull electrodes at maximum sensitivity of 1 picowatt~ Since a steady state prestroke EEG was obtained, each animal served as its own control.

, ~ .
* Trade Mark - S5 -Ten animals had cannulas placed and the stroke accomplished without perfusion. A second control group of ten animals were treated similarly, but were also perfused through the ventirculo spinal (lumbar) route with nutrient solution without fluorocarbon. There were no apparent differ-ences found for post-stroke electroencephalographic activity in these groups. As a measure of stroke severity, 13 animals (of 20) had persisting electrocerebral silience. Of the remaining animals, 5 gained only 2% of their base line power while two had 10% power return within the 4 hour experimental period. FIG. 4 is a representative EEG power tracing from the left and right cerebral hemispheres of a cat perfused only with nutrient solution without flurocarbon and which exhibited persisting electro-cerebral silience during the 4 hour experimental period. The tracings are read from bottom upwards. Normal activity is seen in the lowest tracing and is totally arrested by the ischemic insult half way through the first grouping. There is electro-cerebral silience thereafter throughout the experimental period.
Thirteen cats underwent the same experimental procedure, but were perfused immediately after ischemia with bubble oxygenated nutrient solution (pO2 = 400). For these cats, the flow rate was 4 ml/min with withdrawal from the lumbar theca. Five exhibited continued electro-silience whereas 8 demonstrated EEG recovery from 5% (6 animals) up to 34% (2 animals). FIG. 5 is a representative EEG tracing of one of the eight animals demonstrating 5% recovery after perfusion with oxygenated nutrient emulsion (pO2 = 400).

TJU 3-3 `~

A fourth group of 7 cats was perfused with pump oxygenated nutrient solution (P02 = 645) at 6 ml/min, with withdrawal from the cisterna magna. All cats in this group regained some electrocerebral ac~ivity. The final total power which returned ranged from 5 to 88~ of the pre-stroke base line ~average 22~; p < 0.01 compared to all non-oxygen groups). The elee~roencephalographic activity recovered generally throughout the 4 hour recovery period with the returning total cerebral power exhibiting a first order relationship as a function of time. At the observed recovery ra~e all animals should achieve completely normal EEG power spectra wi thin 8 hours. An oxygen dependent EEG
response is seen when non~oxygenated, bubble oxygenated (PO2 = 400), and pump oxygenated (PO2 = 645) groups are compared as electrocerebral activity recovery greater than 5~ was found in 10%, 62% and 100% respectively. FIG9 6 is an EEG
tracing of the animal showing 88~ return o electrocerebral activity within 4 hours after perfusion with oxygenated nutrient emulsion ~PO2 ~ 645~. The asymmetry between hemis-pheres is an individual variation for this animal.
FIG. 7 is a portion of an EEG tracing showing therecorded effect on electro-cerbral aetivity of a temporary peLfusion failure. This animal, which was perfused using the pump-oxygenated (pO~ - 645) nu~rient ~nulsiorl described above, 25 experienced an inte~ruption (pto A) in perfusiorl for a time period of ~pproximately 1 hour, whereupon perfusion was re-sumed (pt. B~. As seen in this tracing a major deterioration of EEG activity occurred following cessation of perfusion, and resumed thereafter, confinming that the present method in fact sustains EEG activity.
In FIG. 8, the ef~ect of a diminished perfusion flow rate of oxygenated nutrient emulsion is shown on the rate of glucose me~abolism, and lactate and pyruvate concentration.
In accordance with the above-described ventriculo lumbar perfusion procedure using bubbled oxygenated (pO~ = 400) nutrient emulsion, flow rate with nutrlent emulsion without fluorocarbon was established at about 5.0 ml/min. A base line cerebral metabolic rate of glucose metabolism (CMRGl) was esta~lished prior to stroke, which was ollowed after 15 minutes with the perfusion of the oxygenated nutrient emulsion.
CMRGl, which has recovered somewhat after 1 hour, is seen to decline rapidly as the flow rate of perfusate declines~
Similarly, lactate levels rise precipitiously with flow rate decay. These results once again confirm that the flow of o~ygenated nutrient emulsion through the cerebral spinal pathway should be maintained at acceptable rates in order to sustain neurologic tissue.
In FIG. ~, the mean recovery percent for the four groups of animals discussed above is presented in the focm of a bar graph. It is presently preferred to insure that the PO2 value of oxygenated nutrient emulsion upon input is great enough to insure that efficient oxygen transfer capabilities are maintained at the selected flow rate. ~or the FC lumbar group, exposure of oxygenated nutrient solution to certain tis-sue regions when i~s oxygen exchange value was below the known efficient oxygen exchange value (PO2 less than 200) for the fluorocarbon component of this material may have occured. This may be true even though the mean oxygen exchange value of the withdrawn emulsion is above 200. Accordingly, it is presently preferred to maintain the PO2 value of withdrawn oxygenated nutrient emulsion at twice this minimum, or at above 400, either by raising the input PO2 value to much higher levels, as with the ventriculo-cisternal animals described above, or by in-creasing the flow rate of oxygenated nutrient emulsion through the animal to maintain those values. In smaller animals, such as cats, the size of the cerebro spinal pathways creates hydraulic resistance which limits the flow rates which may be achieved at atmospheric pressures using certain pathways~ In such animals, higher oxygen exchange values and shorter perfusion routes, such as the ventriculo~cisternal perfusion route, are preferred. In larger animals, such as humans, it is not anticipated that flow rates will be so limited. Nonetheless, high PO2 values (at least 50~ preferably 80 ~ ~ of the maximum obtainable PO2) are peferred to minimize the volume of perfusate necessary to perform a given treatment and to provide an additional margin of safety at the selected flow rate.
Samples of the perfusing fluids for the animals of this example were removed at predetermined times from entry and exit perfusion ports for analysis o lactate and p~ruvate under a single blind condition. The results are summarized in Table IV:

~ - 59 -. .~

-~ ~
~ c c + .~ ~
~ æ ~ ~ ~ c 3 P~ ~ ~ 1:4 . . 0 4-1 O o~ u~ In P~ O
~:h ~
O
a~ ~ 3 t~ O
a) ~ ~ ~ ~ o .,, or~ ~ ~ O~ r~ ~I C u .lJ,r: C ~ ::1-,1 . ~ O
u ~ ~r~ CO a~
.,1 ~ ` ~ ~ ~ ~ ~ U~ . .
GlC~ ~ C ~ ~ ~ Z r~ ta O ~ ~1 4-1 U
tn O O ~
S~ C
^ o o aJ
~ h Q
--X ~ ~ ~ .c ~1 ~
~ X ~ ~ ~ C
aJ .~ ~ ~ ~ U~ ~ O
æ :~~ O ~:
o ~ c ~ ~1 +1 ~ ~: o 7 ~ o ~ ~ ~
Nr-- ~O r I Ll Q ~a ~ /JJ ~ . . UJ ~ 3 a O ~ ~ U _l ~ ~ ~
4~ O
h ~o tQ ~ a~
C JJ_I ~ aJ h a~ O
~a ~ o ~ . ~,p ,, 3 , . ~ ~ a) u ~ o o ::~
s~~ I+ I ~I~1 ~
. . ~ ~ X
c a ~ ~ u~ u~ o o C
~q~ I~ I tn a X ~ .C O
~ m Lr~ ~a 3 C
Ic ~ . . t) ~:
~ ~-~ l o o u~ O
HE~ Ul .~J ~ r~
~ aJ ~ _10 ~ X
Id(~ 4) ~ ~ O o C t~ ~ ~ ~ V O _l r~ 0 ~ ~ ~ O ~ ~ O
~ Q~-~ O ~15 ~ PJ
E~ . ~ o a~ O
)J ~ ~ . . .,1 J~
SJ ' (~ Q qJ ~ ) ~ * U~

h O U~ ~ ~ Z~ ¦ + ¦ C~
1 U a~ O 1 ~ ~ ~ ~ O ~ ~P ~ ~ ~ ~, SJ ~ ~ C ~ o ~ U
3 :~ O ~ 7 O~ ~u:
O ~ ~D ~. . U ~ .C
1 ~ ~CO '~1C ~ O
aJ.~ ~
~ ~ ~ ~a ~a c o ~ ~ _~ .
Rl ~ ~ _~ ~
h ~ t ¦ ~ 1 O ~J Sl) U~ 10 0 Q, ~ æ o ~D æ o~ cn . ~ ~ . . ~ U~
Q ~ o~ ~:5~ Z
C O ~ U ~F OJ N ~~I) n~ w SJ S ~'1 ta~ u~ z c:~
O 1:: h ~ Z
o aJ
C: 3 1 ~ ~ 1~ S C ~:
~o ~ r~ .
O 1` ~ ~ ~ _ _IU , w^t~ ~ O C _ a~ 3 1 a ~ ~ u~
~3 m ~ n3 ~ ~~ ~s ~ r ,_~
o t~ JJ ~ .a~ ~) o~ ~ c + ~ m O ~ ~1 o 0 V u ~ ~ ~ ~ u Z ~~ V .rt O .,1 ~ ~ ~ ~1 0 _I ~ O ~ . SJ ~ ~
C C ~ ~ ~ ~ Q~ X
R~ a1 ~ O ~ ~1 a~ ri ~-~1 aJ ~ ~ X ~ ~ QJ Q
3 ~ ~ 1~ ~ ~ p~ ;

O
In ~ ~ r~

In animals perfused with nu~rient solution without fluorocarbon the concentration oE lactate during the actual stroke (isoelectro) ~as of the normal CSF value.
The lactate level rose percipitiously, an additional 440~, within 5 minutes of restoring the blood pressure and blood flow through the carotid arteries~ Thereafter the level declined during the 4 hour period to 147% of base line.
In contrast to the lactate data, the pyruvate concentra-tion remained constant through the perfusion period.
When animals were perfused with oxygenated nutrient emulsion, on the other hand, the percipitious increase in lactate did not occur; instead there was a modest 52~ rise during the initial 5 minute period, and the level thereafter remained stable. Significantly, in the oxygenated series the concentration of pyruvate more than doubled during the initial 5 minutes and continued to increase gradually during the remainder of the 4 hour period. The net production of lactate and pyruvate are often used as indicators of anarobic and aerobic glycolysis, respectively. Since these compounds change under different circumstances the expression of lactate/
pyruvate (L/P) ratio best illustrates the net metabolic effects. ~ high L/P ratio indicates that anarobic gly-colysis predominates. It is common practice, therefore, to use the L/P ratio as a sensitive indicator of the redox state of cells. Perfusicn oxygenation in accordance with the present inventions signi~icantly (p < 0.01) ]o~ered the L/P ratio when compared to non-oxygenation (8.3 vs.
43.8). ~t is further evident that the oxygenated 4 hour L/P ra~io is additionally lowered, whereas the non-oxygenated values are still 5 times greater than the control.
Although the oxygenated nutrient perfusate transit time through the brain is only a few seconds, significant oxygen extraction does occur. It was determined by the pO~
difference between inflow and outflow fluids that oxygenated nutrient emulsion lost PO2 = 210 (mean) during its intracere-bral passage. Also unique to the oxygenated nutrient emulsion studies was a rising carbon dioxide presence in the exit fluid which did not occur in non-oxygenated e~perimentsO
The PCO2 rose 5 fold in these fluids over the four hour period (PCO2 = 6.0 vs. 3.0). It is considered that the appearance of carbon dioxide is important since it is a normal product of aerobic metabolism.
In FIGS. 10 and 11 levels of potassium in perfusate before (base line~, during (isoelctric) and follcwing (reflow) global cerebral ischemia in cats are represented. Data are expressed in micro equivalents per minute, and the values are means ~ standard error. Five animals were perfused with nutrient emulsion without fluorocarbon~ and si~ with oxygena~ed 1uorocarbon emulsion. After collecting the perfusate in tubes at 4~C, the samples wer~ stored at -80C for analysis.
Potassiums were assayed by atomic absorption spectrophoto metry. During the base line and i~oelectric time periods, all 11 cats were perfused with nutrient emulsion without fluorocarbon. There were no significant difference~ between isoelectric and base line levels of ~otas~ium in the perfusateO
As seen from FIGS. 10 and 11, significan~ differences in the - ~2 -TJU 3-3 ~ 9 ~ ~

values of potassium were ohserved beginning almost immediately with perfusion tat time 0) and extending throughout the 4 hour experimental period.
FIG. 12 discloses the effects on glucose metabolism for ventriculo-cisterna perfused animals subjected to the stroke and reperfusion procedure of this example. In accor-dance with the invention of Dr. John Lewis Alderman, one of these animals was perfused with twice the glucose concentra-tion (372 mg~) of that used for the remaining animals described herein. As seen from FIG. 12, the glucose metabolism of animals provided with oxygenated nutrient emulsion (glucose =
186 mg~) i5 generally superior following ~eperfusion to the metabolism rate of the control receiving that solution without fluorocarbon. In view of the substantial increase in glucose metabolism exhibited by the animal having a adouble glucose~
solution (372 mg%), it is presently preferred to include at least such elevated glucose concentrations in perfusions performed in accordance with the method of the present inven-tion .
These experimental results demonstrate that extra-vascular perfusion of oxygenated nutrient emulsion affects a significant reversal of the adverse cerebral metabolic effects induced by the experimental stroke condition~ Coincident with the improve metabolic state electrocerebral activity returned.
These findings indicate that extravascularly ~upplied oxygen r glucose and other nutrients were taken up and metabolized in amounts sufficient to restore high energy camp3unds and thereby reactivate ~embrane ionic pumps and reinstitute electrocerebral activity.

~ 6~ -~ TJU 3-3 ~ 9 ~ ~

Oxygenated-fluorocarbon-nutrient-emulsion caused no detrimental effects on vital physiologic functions such as heart rate, blood pressure or electrocerebral (EEG) activity when perfused through ~he ven~ricular system for four hours ~f cats not subjected to the stroke paradigm. These animals exhibited no ill effects after 5-8 months9 and were killed for a double blind neuropathologic examination of the brain~
spinal cord and subarachnoid spaces. No gross or micro~copic changes were observed and the specimens were indistinguishable from non-perfused animals.
In view of the above, those of ordinary skill in the art will recognize that various modifications can be made to the methods and apparatus described above without departing from the scope of the present invention. For example, it should b~ understood that, the injection and withdrawal catheters used to perform the herein described method ~hould be sealed with r~spect to the skull so that a water and bacteria tight - seal is created between the~e catheter and skull. Although conventional bone wax has been used for creating this seal in the feline experiments described above, fitting 22(a3 prefer-ably comprises a double threaded sleeve which i5 thre~ded lnto a bone aperture, and in turn receives complimental threads formed on injection catheter 20a. 5uch attachment mean~, particularly when used with a ventricular injection catheter, ~hould eliminate any need for total head i~mobiliza-tion during human treatmentO

TJU 3-3 ~ 9 ~ 4 It should also be understood ~hat the oxygenated nu-trient emulsions of the present invention may contain vari~us therapeutic agents including free fatty acids, prostaglandins, prostacyclins, cyclic nucleotides and hormones.
S As seen from the above, it is desired to maintain the the P02 level in the withdrawn fluid at levels which are sub-stantially above the minimum level of efficient oxygen exchange of the subjec~ fluorocarbon. For the fluorocarbon nutrient emulsion described above, that minimum tunsaturated condition) occurs at a P02 equal to about 190, which is about 30~ of the readily achieved maximum P02 level. (pO2=645) As described above, it is preferred to perform the treatment method of this invention s~ as to maintain the P02 Of the withdrawn oxygenated nutrient emulsion at a P02 above 400, that is, at a P02 level which is about twice the ~inimum level of efficient oxygen 2xchange for the subject fluorocar~
bon. It is presently anticipa~ed that a similar differential - should be maintained in practicing the present invention utilizing oxygenated nutrient emulsions having other oxygen-2~ atable components exhibiting different ranges of efficient oxygen exchangeO
The methodology described requires the formulation of a physiochemical fluid which must be adequately oxygenated, temperature controlled and delivered under well controlled conditions. The p~rfusion sys~em of ~he present invention may be routlnely placed by trained ani~al surgeons~ ~euro-surgeons commonly possess skills necessary to implant treatr ~ 6~ -TJU 3-3 ~ 9 ~ 4 ment ports in accordance with the presen~ invention in humans.
The procedure is relatively simple and can be quickly accom-plished with available instruments. The oxygenated nutrient emulsion treatment delivery system of the present invention has certain similarities to the arterial heart-lung machine.
Major diff4rences, however, include the use of a ccmplex synthetic fluid for cerebral spinal perfusion, the route performed by cerebral spinal per~usion is an extravascular one, and there is no known limita~ion on perfusion time in accordance with the herein disclosed method. Oxygenated fluorocarbon nutrient emulsion tolerates pumping mechanics well and the exit fluid can either be discarded or recir-culated. Formed blood elements, on the other hand, are fragile and lyse under prolonged recirculation conditions.
It is presently contemplated that cerebral-spinal fluid per-fusion support will need to be carried out until the vascular system can once again take over. Surgical revascularization or bypass procedures will in some cases be necessary to accomplish this end. The return of cerebral vascular compen-tency can be assessed by measurements of regional blood flow, electro cerebral activity, and the metabolic conig-uration of the exit perfusion fluid. One foreseeable compli-cation of this t~chnique is bacterial infection, and rigorous attention to ambient sterility, millipore filtering, and antibiotics should reduce this hazard to acceptable levels.
Safeguards have been built into the pumping system to immedi-ately stop delivery if either the inlet or outlet become obstructed.

- ~6 -Conclusion As seen from the above examples, and the foregoi"g description, circulation of the preferred embodiment nutrient liquid is capable of sustaining cellular in~egrity, aerobic metabolism and ongoing neuronal function. Even for neurons deep within the spinal cord (grey matter) the process has been successful i-n nurturing the ischemic neuronsO The ability to sustain the ~entral nervous sy~tem in a lethally ischemic field which persists for longer than a few minutes has never been accomplished before. The extravascular pathway has not been employed as a global nutrient route prior to the present invention, nor has the combined use of oxygen rich emulsion which also contains the other disclosed novel comp~nents been know to the art.
As seen from ~he above experiments, the methods, compositions and ~ystem of the present invention are capable of providing substantial amount~ of oxygen to neurologic tissues to be treated, while at the same time, removing the by-products of aerobic metabolism, including carbon dioxide, which have been found to exist in substantially higher concen-trations in the exit, diagnostic fluid. Similarly, as dis-cussed above, rapid, normally lethal, lyses of anterior horn cells is readily preventable through the treatment of the present invention, protecting a~ least 60~ of ~he cells through this modality. Similarly, high energy phosphate metabolism utilizing both oxygen and glucose is maintained at sub~tantial levels. Accordingly/ the methodology of the present invention represents a substantial advance in the treatment of central nervous system tissue. Prior to this invention there was not a method available ~o sustain central nervous ~issues after a few minutes of profound ischemic in-sult. This invention should revolutionize the therapeutic capabilities by providing therapeutic approaches for stroke, aneurysm, brain injury, vasospasm, senility, tumors, coma, spinal cord injury, ischemia, post shock, post cardiac arrest and central nervous system poisoning.
It is further anticipated that the treatment method of the present invention should make it possible to interrupt the cerebral blood supply with some impunîty or surgical maneuvers not heretofore possible without great attendant rlsk of producing cerebral infarction. Those of ordinary skill in the art will recognize that future development may result in perfection of the oxygenated nutrient emulsion composition, delivery rates, treatment times, the width of the therapeutic window in which treatment may be instituted and the correlation of behavioral functions in surviving animals with normalization of cerebral chemistry and electro-graphic activityO Nonetheless, by any standard, the present invention provides a dramatic, yet clinically acceptable, therapeutic method for treating i~chemic neurologic tissue~

- 6~ -

Claims (10)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. An oxygenation apparatus for use in treating hypoxic-ischemic neurologic tissue comprising: (a) reservoir means for containing an oxygenatable fluid which is physiologically acceptable to neurologic tissue;
(b) oxygenator means for oxygenating said fluid; (c) first flow control means for establishing a circulation of said fluid between said oxygenator means and said reservoir means at a first preselected rate; (d) injection means for injecting said fluid into a cerebrospinal pathway; and (e) second flow control means for establishing a flow of said fluid through said injection means at a second preselected rate.

2. The apparatus of claim 1 wherein said first flow control means establishes said first preselected rate to achieve a PO2 level of at least about 60% of the sat-urated maximum PO2 value of said oxygenatable fluid.

3. The apparatus of claim 2 wherein said first flow control means establishes said first preselected rate to achieve an oxygenation of said oxygenatable fluid of at least 80% of its saturated maximum PO2 value.

4. The apparatus of claim 1 wherein said first flow control means acts in combination with said oxygenator means to provide a fluid in said reservoir means which is oxygenated to a PO2 value above 600.

5. The apparatus of claim 1 wherein said oxygen-atable fluid comprises perfluorobutyltetrahydrofuran exhibiting a range of efficient oxygen exchange which is substantially linear between about 760 and 200 TORR.

6. The appratus of claim 1 further comprising heat exchanger means for adjusting the temperature of said fluid.

7. The apparatus of claim 6 further comprising third flow control means for establishing a separate circulation between said reservoir means and said heat exchanger means.

8. The apparatus of claim 7 wherein said third flow control means establishes a third preselected flow rate.

9. The apparatus of claim 1 further comprising filtration means for filtering said fluid prior to its injection by said injection means.

10. The apparatus of claim 9 wherein said filtration means comprises a bacterial filter.