Turkish Neurosurgery 10: 90 - 102, 2000 Kutlny: HBO & Fetnl SI'iiinl Cord Gmfts

Effeet of Hyperbarie Oxygen Treatment on Fetal SpinalCord Grafts: A Preliminary Experimental Study

HIperbari~. OksIjen TedavIsInIn FötalUzerIne EtkIsI: Deneysel BIr

SpInal KordÖn Çalisma

Greftleri

MURA T KUTLAY, AHMET ÇOLAK, M. NUSRET DEMIRCAN, OSMAN NIYAZI AKIN,

KENAN KIBICI, KADIR DÜNDAR, SÜKRÜ YILDIRIM

GATA Haydarpasa Training Hospital, Department of Neurosurgery (MK, AÇ, ND, ONA, KK)Department of Undersea and Hyperbaric Medicine (KD), Department of Pathology (SY) Istanbul -Türkiye

Received : 15.11.1999 Çô Accepted: 22.12.1999

Abstract: In fetal neural tissue transplantation, lack ofoxygen supply to the graft in the acute stage is an importantproblem. This study evaluated the effects of hyperbaricoxygen treatment on fetal spinal cord grafts. Spinal cordtissues oQtained from 16-day fetal rats were transplantedto the spinal cord of adult Wistar rats (n=30). The hostswere then divided into two groups. Group 1 wereuntreated controls, and Group 2 rats received 100% O, at2.5 atmospheres pressure for 90 minutes twice aday för 7days. All animals were sacrificed for histologicalexamination. Degenerative changes, interface, host spinalcord ederna, and vascularization of the graft were scoredon three-grade scales, and the scores were statisticallyanalyzed. The graft survival rates in Groups 1 and 20 were46.1% and 71.4%, respectively. Animals treated with HBOshowed statistical!y less spinal cord ederna than theuntreated group (P < 0.05). The interface was alsosignificantly better in the treated group (P< 0.05). Thesefindings indicate that HBO therapy can increase thechances of graft survival. We believe that the edema­red ucing effect of HBO prevented the graft from becomingdisplaced and, thus, contributed to the successfulintegration of graft and host.

Key word s: Fetal graft; hyperbaric oxygen; neuraltransplantation; rat; spinal cord

90

Özet: Fötal nöral doku aktarimlarinda, grefte erkendönemde oksijen desteginin olmayisi önemli birproblemdir. Bu çalismada, hiperbarik oksijen tedavisinin(HBOT) fötal spinal kord greftleri üzerine olan etkileridegerlendirilmistir. 16 günlük fötal ratlardan saglananspinal kord dokulari eriskin Wistar ratlarin (n=30) spinalkordlarina nakledildi. Sonrasinda iki guruba ayrildilar.Ikinci gruptaki denekler günde iki kez, 90 dakika olacaksekilde 7 gün, 2.5 atmosfer basinç altinda % 100 oksijentedavisi aldilar. Tüm denekler histolojik incelemeler içinfeda edildiler. Dejeneratif degisiklikler, greft-konakçibütünlügü, konakçi spinal kord ödemi ve greftinvaskillarizasyonu 3-dereceli ölçeklerde puanlandirildi. Bupuanlarin istatiksel analizleri yapildi. Greft yasam oranlari1. Grup ve 2. Grupta sirasiyla % 46.1 ve % 71.4' dü. HBOtedavisi alan deneklerin spinal kordlarindaki ödem, tedaviyapilmayan deneklere göre istatistikselolarak daha azdi(p<0.05). Konakçi-greft bütünlügü yine bu grupdeneklerde daha iyi (p<0.05) idi. Bu bulgular, HBOTningreftin yasama sansini artirabilecegini göstermektedir.HBOT nin ödem azaltici etkisinin, greftin kayitedenatilimini önledigini, böylece konakçi-greft bütünlesmesinekatkisinin oldugunu düsünmekteyiz.

Anahtar kelimeler: Fötal greft, hiperbarik oksijen, nöraltranplantasyon, rat, spinal kord.

Turkish Neiirowrgery 10: 90 - 102, 2000

INTRODUCTION

Attempts to transpIant fetal central nervoussystem (CNS) grafts into the spinal cord have shownthat fetal cortical, spinaL, and tegmental transplantscan survive and grow in the adult rat spinal cord(5,6,14,15,18,34,66,72-74,85).However, experimentalstudies also indicate that the survival rate of

implanted grafts is variable, and the results that havebeen presented by various authors are difficult tocompare. The reported data are controversia1.Trans plan ts ha ve been reported to prod uceimprovement in some experimental models of spinalcord injury (7,14,49,73,82), but other experimentalefforts to foster axonal regeneration in the transectedspinal cord have yielded discouraging results(15,18,71). Although we currently know thattransplants can survive at sites of complete spinalcord transection (28,35,71), the success ra tes for suchprocedures have been very low (18,20,28).

Today, the main question is not whether neuralgrafting to the spinal cord is possible, but how thesurvival of neural transplants can be enhanced, andwhat conditions lead to maximum graft survival andultimate structural and functional restoration of theinjured spinal cord. A number of recent studies haveemphasized the importance of vascularization of fetalCNS grafts (13,47,48,52,53,81,91). In someinvestigations, pockets and lesions have been createdto increase vascular supply to the transpIant(45,46,66). In attempt to create an appropriateenvironment for the fetal CNS graft, some researchershave placed these grafts into resection lesions thatwere made 1-12 weeks earlier (8,34,45, 46,51,66,76).In addition, the use of various growth factors tomodify the microenvironment after transplantationhas been shown to improve graft survival andfunction (29,79,87). Other therapeutic agents ortechniques have also been tried in efforts to promotelong-term graft survival (1,28,31,43,53,70,86).

Hyperbaric oxygenation (HBO) is a widelyrecognized adjunctive therapy. !ts beneficial effectsas a therapeutic modality have led to a broadspectrum of uses, ranging from carbon monoxidepoisoning to a variety neurological disorders(3,4,26,33,39,40,65,84). Lesions involving excessivetissue damage with acute oxygen deficiency, as inwounds that require skin transplants, radiation­induced necrosis, skin-musc1e flaps, crush-syndrome,necrotic wound infections, compartment examined.syndrome, and burns are all further indications forHBO (21-23,56,58,80,83).

Kiltlay: HBO & Felal Spiiial Cord Grafts

In contrast to other areas of hyperbaricmedicine, little research has been done on fetal tissuetransplantation (50). This study was designed toevaluate the possible beneficial effects of HBOtherapy after transplantation of intact fetal rat spinalcord into a completely transected adult rat spinalcord. We histologically examined the viability,vascularization, and parenchymal integration of eachgrafto This is the first study to deseribe theexperimental use of HBO for this purpose.

MATERIALS AND METHODS

Thirty adult male Wistar albino rats, eachweighing 250 to 280 gm, were used as hosts. Thedonors were fetuses from time pregnant femaleWistar rats at 16 days of gestation (E16). (Day O =day of insemination). All the male rats underwentembryonic homotopic transplantation to their spinalcord. The experimental protocol was approved bythe Animal Care Committee at our institution, andwas performed in accordance with the Principles ofthe Helsinki Dec1aration.

Doiior tissiie preparatiol1:Pregnant female rats at 16 days of gestation

were anesthetized with intraperitoneal sodiumpentobarbital (40 mg/kg), and fetuses were obtainedby cesarean section. At the time of utilization, oneuterine hom was placed in ice-cold lactated Ringer'ssolution and the fetuses were removed from theamniotic sacs. Each harvested fetus was checked for

viability by observing color and movement.Harvested fetuses were anesthetized by hypothermiaon a bed of crushed ice. Under magnification, thespinal cord was dissected free, and 1 to 2 mm longsegments of the thoracic region were obtained usingsharp microscissors. Exposure of the grafts to air wasprevented by continuous irrigation with Ringer'ssolution.

Traiisp laiitatioii procediires:The hosts were anesthetized with

intraperitoneal sodium pentobarbital (40 mg/kg) anda laminectomy was performed at the mid-thoracicleveL.With the aid of an operating microscope, thedura mater was opened longitudinally with a 27­gauge needle and the spinal cord was transectedsubpially using a microaspiration technique (71).Theprocedure resulted in a cavity with a 1 to 2 mm gapbetween the stumps of the transected spinal cord.The cavity was then filled with Gelfoam to assist inhemostasis while the donor was prepared. Followingpreparation of the donor tissue, Gelfoam was

91

Turkisli Neiirosurgery 10: 90 - 102, 2000

removed from the lesion site and the cavity wasrinsed with cold sterile saline. Using jeweler'sforceps, one or two fetal spinal cord fragments wereimmediately implanted into the cavity created byaspiration. Special care was taken to ensure that thefetal spinal cord transplant was oriented in therostrocaudal direction. The lesion site was covered

with a narrow strip of Gore Tex surgical membrane(W.L. Gore & Associates, Inc., Arizona) to limitfibroadhesive reactions. The muscle and skin wounds

were then closed in layers, and the animals weremaintained on heating pads until they recovered fullyfrom the anesthetic.

Postoperative care:Procedures were performed under sterile

condi tion s and normothermia was maintained bymeans of warming pads. All operated animalsreceived daily intramuscular injections ofprophylactic antibiotics for 3 consecutive days aftersurgery. None of the rats were immunosuppressed.In the postoperative period, all the rats underwentmanual bladder expressions twice daily, and this wasusually continued to the end of the study.

HBO therapy:After transplantation, the animals were divided

into two groups. Those in Group 2 underwent HBOtherapy, which was started 1 hour af tertransplantation. They were placed in a hyperbaricchamber designed for laboratory animals andreceived 100% oxygen at 2.5 atmospheres absolute(ATA) for 90 minutes, twice a day for 1 week. Aftereach treatment period, all animals were retumed tonormobaria within 10 min, during which time theybreathed compressed air. We used a regimen thathas been found to be safe and effective (42,90). Therats tolerated HBO conditions well, and theydisplayed no signs of discomfort while receivingtreatment. The pressure chamber was made of steel.A cylinder-shaped HBO room 70 cm long and 40 cmin diameter and resistant to 4 atmospheres ofpressure was constructed for the experiment (Figure1). it was connected to an oxygen tube by a three­valve system, and had a do or and two observationwindows. The inside pressure was monitored andadjusted using amanometer connected to the system.

Histological exaniiiiatioii:The rats were maintained on a 12hr /12 hr light/

dark cycle, with food and water available. After thelast treatment, they were re-anesthetized with anoverdose of intraperitoneal sodium pentobarbitaland were perfused intracardiaiiy with 0.9% saline,

92

Kiitlay: HBO & Fetal Spiiiii/ Corn Gmfts

Figure 1:Photograph of the hyperbaric oxygen applkationchamber that was specifically designed for theexperimental studies.

foIlowed by 4% paraformaldehyde in 0.1 Mphosphate buffer at pH 7.4. it was easy to re-exposethe lesion site because of the use of the surgicalmembrane. The membran e minimized the problemsassociated with scarring and hypervascularity interms of locating the graft site. After the lesion sitewas exposed, the cavity was examined visually forthe presence of an obvious transplant. The spinalsegments that included the transplants wereimmediately removed and drop-fixed in 10%buffered formalin. Following fixa tion in formalin,they were embedded in paraffin. Sections 5 pm-thickwere cut in the longitudinal plane on a microtorne,and were then stained with hematoxylin and eosinin order to examine the characteristics of the graftsand their relationship to host tissue. For each sample,seven sections were prepared and four fields wererandomly selected. Aii slides ,,,,ere evaluated andgraded by a blinded observer.

The slides were examined under lightmicroscopy. Host spinal cord edema was evaluatedusing a three-point scale ranging from 1 for normal,to 3 for severe. Within the graft, degenerative changesincIuding micro- andi or macrocyst formation,cavitation, necrosis, gliosis, and composition ofneurons were assessed using a three-point scale of 1for normal (none or mild degenerative changes), 2for moderate, and 3 for severe. The degree ofparenchymal integration (interface) was also scoredon a three-point scale with the foiiowing specific

..

Titrkish NWl'Ositrgery 10: 90 - 102, 2000 Kittlay: HBO & Fdal Spiiial Cord Gm/I,

descriptives: 1 = an interface with large gaps, severeglial reaction (poorly integrated graft); 2 = aninterface with small gaps, moderate glial reaetion;and 3 = complete parenchymal integrity with no glialreaction and gaps. Vascularization of the graft wasalso scored semiquantitatively. Each branch wascounted as a single vessel. A score of 1 to 3 wasderived for vascularization in each seleeted field,with a score of 1 indicating an area containing 5vessels; a score of 2 indicating an area containing 6­10 vessels; and a score of 3 indicating an areacontaining 11 vessels. The scores for these parameterswere averaged and compared between the groups.In addition, the data were statistically analyzed usingChi-Square and Fisher's Exact Probability tests.Differences were considered to be statisticaliysignificant at P < 0.05.

RESULTS

The results of the histological examinations arepresented in Table 1, and the mean scores are shownin Figure 2. None of the animals showed functionalrecovery.

period. Two of the remaining 13 animals developedlocal postoperative infection at the operative site.Only 6 of 13 (46.1%) ho st animals had viable graftsat 7days after transplantation, and the graft volumesin all 6 cases had decreased compared to their initialsize. In the remaining seven cases, the graft hadbecome displaced from the gap at the site oftransection, presumably due to edema in thetraumatized spinal cord. These displaced grafts hadformed a dense degenerated fibroglial mass. Thetransplantation site was characterized bydegenerating cellular debris, and containedaggregations of extravascular blood cells. In general,the embryonic spinal cord tissue in this groupshowed very limited growth and differentiation. Theneurons were randomly dispersed or grouped industers, and most of them were undifferentiated(Figure 3). Although the transplants containedn"umerous areas of microvacuolation and

d~generative neuronal changes, an area of intensenecrosis cavitation and macrocyst formationwas consistently seen in the central region of thegraft. The mean score for the degenerative changeswas 2.3.

Figure 2: Bar graph depieting the mean scores for thecontrol animals and those treated with HBO.

Most of the grafts showed gliosis. Typical spinalcord cytoarchitecture was not observed. Therewas also more histological evidence of hostspinal cord in this group compared than in theGroup 2 animals. Most of the cases in Group 1 hadmoderate to severe edema, and the median scorewas 2.1 (Figure 4).

Figure 3: An example of a fetal spinal cord graft from aGroup 1 anima!. The tissue contains same smailand many atrophied neurons. The internalstructure of the graft is disorganized and thereis obvious degenerating tissue (H&E x 200).

aGroup 1

• Group 2

Vascular'zaton~ei1aceDegeneratiYe

changes

Group 1:Of 15 animals, 2 died from different causes at

different times throughout the posttransplantation

Table 1: Summary of histological findings.

a Fo~ degenerative changes, l=none or mild, 3=severe

b For host spinal cord ede"ma,l=normal, 3=severec For interface, 1= poorly integrated graft,

3=complete integration,d For vascularization, l=an area containing 5

vessels, 3= an area containing > 11 vessels

93

Turkish Neurosurgery 10: 90 - 102, 2000

There were varying degrees of anatomicalintegration between transplant and host. Most of thetransplants were only partially integrated with theparenchyma of the host spinal cord. Intense glialproliferation and cyst formation were commonlyobserved along the interface (Figure 5). The score for

Figure 4: Section of a Group 1 host spinal cord showingGrade 2 edema formation (H&E x 200).

Figure 5: Section from a Group 1 rat showing the host-graftinterface (broken line). Although the grafr (g) istightly juxtaposed to the host parenchyma (h), itis poorly integrated with it. There are many cysts(c) at the interface (H&E x 100).

94

Kullay: HBO & Felal Spiiial Cord Grafts

interface ranged from 1 to 3, and the mean scorewas 1.6.

Blood vessels of larger diameter were presentin some sections of the grafts, but were few innumber. A few vessels could also be iden tified at the

host-graft interface. The mean score forvascularization was 1.6.

The graft site in this group also exhibitedevidence of rejection. On histological examination,there was a dense accumulation of

polymorphonuclear leukocytes and macrophages inthe grafts (Figure 6), but no evidence of any localspread of infection into the host spinal cord.

Groiip2:In this group, one animal died from un known

cause(s) in the early posttransplantation period.None developed infection. The transplant survivalrate, excluding the case of postoperative death was71.4%. Most of the viable grafts had not decreased insize, and had completely filled the original cavity.On macroscopic examina tion, 8 of 10 fetal grafts weregrossly visible, and 3 grafts in this group had als oexpanded dorsally. The surface of the transplant hada whitish appearance that was clearly different fromthe adjacent spinal cord of the host.

Histologically, the transplants appeared tocontain a higher density of neurons than thesurrounding host spinal cord (Figure 7). There were

Figure 6: Light micrograph of a Group 1 animal showingthe graft occupied by polymorphonuclearleukocytes, which contributed to the necrobiosisof many neurons (H&E x 100).

Tiirkish Neiirasiirgery 10: 90 - 102, 2000

Figure 7: Photomicrograph showing the graft (g) withgood parenchymal integration (broken line) in aGroup 2 specimen. The graft is densely populatedwith cells. All the neurons appear healthy andnormal, and there are no signs of obviouspathological reaction (H&E x 50).

many normal-Iooking, healthy, well-differentiatedneurons, and they were homogeneous in compositioncompared to Group 1 grafts. Nevertheless, in a fewcases, some areas of gliosis were present. Althoughregion s of microcyst formation and micronecrosiswere detected, in most cases there were no expansiveareas of necrosis, hemorrhage, or macrocystformation, all of which were routinely observed inthe Group 1 rats. The mean score for degenerativechanges was 1.5. No typical spinal cord architecturecould be identified. Animals in this group showedstatisticaliy less spinal cord edema than theuntreated group (P< 0.05).The mean score for edemawas 1.3.

In general, the transpIant fused well with thehost spinal cord parenchyma. In the best cases, theregions of parenchymal integrity between graft andhost (interface) occurred wi th no interveningconnective tissue scar or cystic cavitation (Figure 8).The mean score for interface was 2.6, and there wasa statistical difference between the two groups(P< 0.05). In addition, small blood vessels wereobserved along the host-transplant interface(Figure 9). Large, dilated, thin-walled vesselsresembling dilated veins were of ten observed withinthe transplants. The grafts had good vascular supply(Figure 10). Although there was no statistical

K/ltlny: HBO & Fetnl Spiiiiil Can! Gmfts

Figure 8: An example of an interface in a Group 2 anirnal.The broken line shows the bounda ry between thetranspIant and host tissue. The transpIant (t)appears to be well integrated with thesurrounding host neuropi! (h) (H&E x 200).

difference between the two groups, the mean scorefor graft vascularization in the second group was 2.1,which was higher than the score for the untreatedgrafts.

Microscopic examination revealed lessinflammatory reaction in the HBO-treated groupthan in the controls. However, in some cases therewere mononuclear cells surrounding the vessels inthe grafts.

DISCVSSION

The results of this preliminary study indicatethe following: 1.The survival of fetal (E16)homotopicrat spinal cord grafts transplanted to the completelytransected spinal cord of adult rats appears to beenhanced by applying HBO; and 2. The parenchymalcontinuity between the graft and the host spinal cord,and the vascularization of the graft can be enhancedby administering HBO therapy. Although theachieved survival rate was not excellent, there weresignificant differences between the two groups. The

95

Turkis/i Neiirosiirgery 10: 90 - 102, 2000

Figure 9: Sectism of a Group 2 specimen showing smailblood vessels (v) along the host (h)-graft (g)interface (broken line) (H&E x 100).

Figure 10: Many smail blood vessels (v) were seen withinthe graft at 7 days after transplantation in theGroup 2 animals (H&E x 100).

transpIant survival rates in our groups, excludingpostoperative deaths, were 46.1 % and 71.4%,respectively. Note that the rate in the HBO-treatedanimals was approximately double the rate observedin the controls.

96

Kiitlay: HBO & Fetal Spiiial Cord Grafts

There are several possible explanations for thisimproved survival and graft-host integration: first,HBO counteracts tissue hypoxia by elevating tissueoxygen tension (Pt02) (25,44,57,60,63,64,77);second,HBO has the abiIity to reduce edema in traumatizedtissues (25,61,77,84);and third, HBO has can faciIitatethe formation of new vessels (44,57,60,63).

Tissiie Oxygeiiatioii:it is weii known that embryonic CNS tissues

contain mitoticaiiy active precursors of aii the neuraland gIial elements. These elements have remarkablegrowth potential, and are also capable ofdifferentiation. Because theyare resistant to traumaand hypoxia (54,81), they continue to develop andattain adult characteristics after transplantation. Inaddition, grafts of fetal CNS tissue have the capacityto maintain a relatively normal fetalmicroenvironment. it has been reported that theyquickly establish an oxygen microenvironment aftertransplantation, and that oxygen transport is aregulated variable in the graft neuropil (81). TheabiIity of a graft to develop and produce ceiis outsideits normal environment has also been demonstrated

in explant culture in vitro (78). Because neurogenesisis known to continue undisturbed for period s of 5 to6 day s in grafts obtained from rat fetuses (E16) (38),one would anticipate that the rate of microvesselproIiferation known to occur in the normal earlypostnatal period would not match the increasedmetaboIic demands of development. Due to thesefeatures, embryonic neural tissue can remain viablefor some time, but this geneticaiiy preprogrammedcapacity gradual1y diminishes as developmentproceeds. We termed this initial period the "no-flowcritical phase" of transplantation. During this stage,it is important to give supportive treatment toprevent graft damage from infarction until bloodflow becomes available. Oxygen is vital for the CNS,and oxygen deficiency, regardless of the cause, startsa vicious circle of pathological changes in CNS tissue.

In the present study, the object of therapy wasto supply the graft tissue with adequate oxygen untilnormal reperfusion was established, and thusinterrupt this vicious circle. Transplantation destroysa graft's blood supply, leaving it anoxic. This impIiesa rapid faii to zero of the oxygen tension of the graft,and leads to pathological changes in the graft tissue.The transpIant may sustain alarger degree of ceiideath in this "no-flow" critical period. Processes ofproliferation and differentia tion may be delayed. Thisexplains why grafts undergo an initial size reductionafter transplantation, a phenomenon that has been

Turkis/i Neiirosurgery 10: 90 - 102, 2000

highlighted previously (29,68,69).

We believe that the amount of ceii necrosis in

the transpIant plays an important role in determiningthe final volume of the developed graftoIn the presentstudy, most of the viable grafts treated with HBOshowed no initial size decrease and grew more thanthose in the untreated animals. Moreover, some ofthe grafts even grew larger than their initial size attransplantation. This may be the result ofbetter graftoxygenation.

We think that HBO helped enhance graftsurvivability and quality by preventing or limitingceii death. it has been shown that administration of

HBO increases oxygen levels in the rat spinal cord(41,67). Measurement of Pt02 levels within theinjured spinal cords of adult rats after transplantationof fetal spinal cord tissue showed that even the moreextensively developed transplants continued to showPtOi levels lower than those taken from normal spinalcord tissue (81). it was also shown that the lowoxygen levels were significant particular1y in thedeeper samples from the graft (81). In our study, thegrafts in Group 1 showed that intense degenerativechanges had occurred in the central portions of thegrafts. This can be explained by the difficulty ofoxygenating these regions, even under conditions of100% oxygen at ambient pressure (24). In contrast,with HBO at 2-2.5 ATA, diffusion of oxygen into thenonvascularized tissue is enhanced (26).

it has also been proven that at 3 ATA, there is a15-fold increase in dissolved oxygen (9,83). This issufficient to meet requirements for tissue oxygenationin the absence of vascularization. The HBO-treatedgroup also showed microcyst formation anddegenerative changes, but we did not observe severepathological changes and macrocyst formation in thecentral portion of the grafts. Hyperbaric oxygenationinvolves applying oxygen under barometric pressuregreater than that found on the Earth's surface at sealeveL.The improvement of graft oxygenation by HBOmay concomitant1y combat the ischemia and hypoxiaassociated with edema.

HBO therapy has been advocated as a methodof improving tissue oxygen delivery, especiaiiy toareas of diminished flow (40). it counteracts tissuehypoxia by elevating tissue Pt02 (10,22,25,44,57,60,63,64,67,84), and also increases the amount ofoxygen dissolved in the plasma, depending on theabsolute pressure used (9,83,88). it has been proventhat tissue oxygenation may be sustained by diffusion

Kul/ay: HBO & Fetnl Spiiiiil Cord Gmfts

of oxygen from surrounding tissues. Underhyperbaric conditions, oxygen dissolved in plasmacan spread into the extravascular spaces, and cannourish the tissues eve n in the absence of red blood

ceiis (9,26,59). In this regard, even if there is nocirculation HBO will be effective. In this study, lackof oxygen supply to the graft was our main reasonfor using this form of therapy. it is our belief thatincreasing the capillary P02 tension by HBO increasesthe amount of oxygen that reaches the developinggraft.

Aiiti-edema ettect:A second important effect of HBO therapy is

reduction of tissue edema. After transplantation, thegraft may be in an unfavorable environment due totissue edema and ischemia, which are not alwaysimmediately resolved (19,78,81). Even with the besttechnique, transplantation of neural tissues, beingtraumatic to both the transpIant and the host spinalcord, produces some degree of pathological reactionin the transpIant and the surrounding host cord.Edema can compress the capillaries and aggravateischemia, starting another type of vicious circle. Theedema that develops may either persist or worsen(10,25,80). Increased edema in the host spinal cordparenchyma may prevent inihal parenchymalapposition between transpIant and host tissue, whichis essential for the eventual integration of the two(19,28,71). Edema can cause grafts to fold over orbecome deformed, and fetal grafts can be easilyejected from the spinal cord. it can also increase thediffusion distance from host to graft (25).HBO createsa favorable diffusion gradient through edema fluidand other barriers, such as blood, to the hypoxic cells(25).

In most of our Group 1 cases, the transplantedembryonic tissue did not stay between the stumpsof the spinal cord, but went extraparenchymal andformed a dense degenerated mass. Similarobservations have also been reported by otherauthors (19,28,71). Although it is not clear whatcaused the migration of the graft, there appears tobe at least one possible explanation. Since the hostspinal cord was traumatized by the aspirationprocedure, it is possible that the migration of the graftmay have been caused by edema in the adjacent hosttissue. However, there was noticeably less edema inthe HBO-treated group compared to the controlanimals, and most of the grafts in the treated groupwere adequately retained. This beneficial effect wasthought to be due to the edema-reducing effect ofHBO.

97

Tiirkish Neiirosiirgery 10: 90 - 102, 2000

The HBO treatment improved hostresponsiveness by making the environment morefavorable, and allowed the graft to make contact withthe stumps of the host spinal cord. Accordingly,previous experimental and clinicalstudies have proventhat HBO is a useful adjunct in the treatment oftraumatic edema (10,32,61,77,80,83,84,90). it reducesedema through vasoconstriction and reduction ofblood flow (25,26,83). The latter is compensated for byhyperoxia (25,61,84). it has also been shown that anyvasoconstriction of the microarterioles that occurs

during HBO exposure does not persist beyond thetreatment period (90). HBO also reduces neuronalswelling by improving the cells' metabolism (16,26,33).

Vascularizatioii:

A third determinant of graft survival is adequatevasculariza tion of the grafto Under normobaricconditions, since tissue oxygenation is regulated byblood flow and metabolic need within very narrowconstraints during development (30,81), interruptionof blood flow means that the amount of oxygenavailable to the cells also falls to zero. it is clear from

previous investigations that adequate vascularizationof the graft is necessary not only to ensure the viabilityof its cells, but also to allow sufficient phenotypicexpression of the graft and achieve a favorable effectin the host (47,48,52,53,87,91). Thus, rapid access tovascular supply is mandatory to ensure survival of anyneural graft. Although, it has be en reported thatvascularization of the graft occurs within the first 24­72 hours after transplantation (13,47,48,52), someauthors have emphasized that the vasculature ofgrafted neural tissue remains poorly defined for thefirst 7 days (2,11,91). In the revaseularization process,whether the graft vessels are indigenous or host­derived is still controversial (2,47,48). In our opinion,the rapidity and the extent of vascularization, whateverits origin, is important. Although there was nostatistical difference between the two groups in ourstudy, the mean score for vascularization was higherin the HBO-treated grafts. In Group 2 animals, all theviable grafts became well-vascularized at the end ofthe first week, and small blood vessels were alsoidentified at the graft-host interface, as observed inprevious studies (11,12,53).

Because most of the grafts in Group 1 failed, itseems reasonable to assume that these tissues probablydid not yaseularize rapidly enough to ensure survival.Marx et aL.(58) proved that HBO treatment producedeight- to ninefold greater vascular density than thevaseularity observed in both normobaric oxygen andair-breathing controls. In accord with this, a number

98

Kiitlny: HBO & Fetnl Spiiinl Cord Grafts

of studies have shown that HBO promotes capillaryproliferation (22,44,57,60,63,64); however, theunderlying mechanism of enhanced capillaryangiogenesis is stilI uncertain. We believe that HBOsupports hypoxic tissue, inc1uding primitiveendothelial cells, by restoring abnormally low tissueoxygen tension levels. it can also be speculated thatHBO and intermittent exposure to this oxygen pressuremay have trophic effects on endothelial cells. The needfor further studies on this subject is obvious.

Iiiterface:Likewise, the survival of fetal neural grafts is

highly dependent on the establishment of parenchymalintegration between the transpIant and host (19,49,76).The region of parenchymal continuity between the twois known as the interface. The presence of a well­established interface indicates the existence of an ideal

neural transplant, and represents histological evidenceof graft survival. Grafts that are not anatomicallyintegrated cannot be considered to have anatonucal orfunctional significance for the host. They may survivefor only a short time, and eventually degenerate(17,19,53,79). Thus, parenchymal integration of a fetalneural graft seems to be an indispensible requirementfor the ultima te survival of the graft. A number ofprevious studies have indicated that considerable glialscarring at the interface prevents parenchymalintegration of graft and host (19,28,29,34,71,74,81). Theglial scar appears to form a mechanical barrier to graft­host neuronal connectivity (71,75). On the other hand,neural transplants that contain a number of survivingcells are able to attach themselves to the host

parenchyma long before glial scar formation sets in(19), and most scarring develops after 7 day s (53,89).Therefore, if the number of surviving cells within thegraft can be increased in this critical period, it may bepossible to produce much more healthy grafts.

In the present study, there were significantdifferences between the two groups with regard toparenchymal integration. Group 1 showed glialscarring and cyst formation at the host spinal cord­graft interface in each viable grafto In contrast, in theHBO-treated group, there were a few small eysticstructures at the interface, and we observed a slightincrease in gliosis, which did not affect the viabilityand anatomical integration of the transpIant.Moreover, in the best cases, the interface was free ofeystic formations and gliosis. In accord with previousstudies, we suggest that absence of or limited gliosisis attributable to the presence of a healthy graft(6,19,34,76,85). The effectiveness of HBO inestablishing parenchymal integration may be due to

Tiirkis/i Neiirasiirgery 10: 90 - 102, 2000

enhanced survival of cells within the graft in the "noflow" critical phase, rather than to adired effed onthe interface.

None of the animals in the study receivedimmunosuppressive therapy. The vast majority ofgrafts treated with HBO survived well, and there waslittle evidence of rejection. In Group 1animals, a largeamount of cell death might have been brought on bythe invasion of polymorphonuclear leukocytes andmacrophages, whose adivity included clearing thegraft. In contrast, stabilizing the graft by providingad equa te oxygen could have limited the death of thesecells in the acute stage, and thereby inhibited thesubsequent inflammatory process.

The adions of HBO are not limited to reducingtissue edema, increasing tissue oxygen tension, andcombating ischemia. Anaerobic metabolism results inacidosis and depletion of cellular energy. As thedemands for energy produdion are no longer met, cellslose their ability to maintain nom1al ionic homeostasis.This abnormal cellular environment is extremelydamaging to cell membranes (36,77). In addition,inadequate oxygen supply to the injured tissue-resultsin the conversion of aerobic glucose metabolism toanaerobic metabolism (62). However, it has be en

shown that improving the oxygen supply to damagedcells results in normalization of the intracellular

electrolyte balance (27,37). HBO has also a favorableeffect on glucose metabolism (16,33). These factssupport the theory that HBO therapy improves thechances for graft survival and integration with hostspinal cord until blood flow hecomes available, in partyby allowing more aerobic glucose metabolism to occurand helping main ta in normal ionic homeostasis duringthis phase.

Infection also adds significant risk to thetransplantation procedure. it has been shown thathypoxia is a fundamental charaderistic of wounds,and that oxygen availability is an important factorin wound repair and resistance to infedion (24). HBOtreatment is used as an adjunct to antibioticmedication in a variety of infectious diseases. itinhibits gram-positive as well as gram-negativeorganisms, and its positive impact in anaerobicinfections is also well known (22,23,55,56). These

effeds may be an another important considerationin transplantation procedures. Although all theprocedures in this study were performed understerile conditions, two animals from Group 1developed infedions. In contrast, none of the HBO­treated rats had problems with infedion.

Kul/ny: HBO & Felnl Spiiinl Cani Grnfts

Lastly, we want to emphasize that HBO cannotrevitalize necrotic tissues, but by virtue of its effedscan increase the chances that marginally viable cellswill survive. Knowing this, there should be minimaldelay between transplantation and the initiation ofHBO treatment. We believe that prompt in treatmentis important, but there are so me specific issues thatstill need to be addressed, including optimal timingfor initial treatment, duration and frequency oftreatment, and others.

CONCLUSION

This is the first investigation that has evaluatedthe effects of HBO on grafted embryonic spinal cordtissue. Although the results of this preliminaryexperimental studyare not dramatic, theydemonstrate the beneficial effects of HBO therapyon feta! spinal grafts. HBO provides an appropriatemilieu for graft growth, and can increase thelikelihood of graft survival by restoring low tissueoxygen tension and reducing edema. Thus, itcontributes to the integration of the graft with thehost parenchyma. We do not view HBO as a solutionto the many problems in this field, but believe that itcan be used as an adjunct to other therapeutic agentsor transplantation techniques used in fetal tissuetransplantation. Although our results areencouraging, further investigation will more clearlydelineate the role of HBO in embryonic neural tissuetransp!antation.

Tlzjs stiidy was preseiited at tlze 1st AiiiilItllMeetiiig of tlze Spiiie Society of Eiirape, Miliiicli,Germaiiy, September 7-11, 1999. (Oral preseiitatioii)

Correspondence: Dr. Murat KUTLA YGATA Haydarpasa Egitim Hastanesi,Beyin ve Sinir Cerrahi KlinigiKadiköy- 81327Istanbul/TürkiyePhone: + 90 216 3450295 int. 2657

Fax: + 90 216 3487880

REFERENCES

1. Aebischer P, Winn SR, Galletti PM: Transplantation ofneural tissue in polymer capsules. Brain Res 448:364­368,1988

2. Akalan N, Seçkin H: Blood-brain barrier alterationsfollowing intracerebral grafting: solid and cellsuspensions of fetal central nervous system tissue. TurkNeurosurg 4: 21-26, 1994

3. Anderson DC, Bottini AG, ]agiella WM, Westphal B,Ford S, Rockswold GL, Loewenson RB: A pilot study

99

Tiirkis/i Neiirosiirgery 10: 90 - 102, 2000

of hyperbaric oxygen in the treatment of human stroke.Stroke 22:1137-1142, 1991 .

4. Barnes MP, Bates D, Cartlidge N, French JM, Shaw DA:Hyperbaric oxygen and multiple selerosis: Final resultsof a placebo-controiied, double-blind tria1. J NeurolNeurosurg Psychiatry 50:1402-1406, 1987

5. Bernstein JJ, Patel U, Kelemen M, Turtil S:Ultrastructure of fetal spinal cord and cortex implantsinto adult rat spinal cord. J Neuroscie Res 11:359-372,1984

6. Bernstein H, Hoovler DW, Turtil S: Initial growth oftransplanted E 11 fetal cortex and spinal cord in adultrat spinal cord. Brain Res 343:336-345, 1985

7. Bernstein H,Goldberg WJ:Fetal spinal cord homograftsameliorate the severity of lesion-induced hind limbbehavioral deficits. Exp Neurol 98:633-644, 1987

8. Björklund A, Stenevi U: Reconstruction of thenigrostriatal dopamine pathway in the intracerebralnigral transplants. Brain Res 177:555-560, 1979

9. Boerema I, Meyne NG, Brummelkamp WK, Bouma S,Mensch MH, Kamermans F, Sternhanf M, van AlderenW: Life without blood. (A study of the influence of highatmospheric pressure and hypothermia on dilution ofthe blood). J Cardiovasc Surg 1:133-146, 1960

10. Bradshaw PO, Nelson AG, Fanton JW, Yates T, Kagan­Haiiet S: Effect of hyperbaric oxygenation on peripheralnerve regeneration in adult male rabbits. UnderseaHyperbaric Med 23(2):107-113, 1996

11. Broadweii RD, Charlton HM, Balin BI, Salcman M:Angioarchitecture of the CNS , pituitary gland andintracerebral grafts revealed with peroxidasecytochemistry. J Comp Neurol 260:47-62, 1987

12. Broadweii RD, Charlton HM,Ganong WF, Salcman M,Sofroniew M: Allografts of CNS tissue possess a blood­brain barrier. i. Grafts of medial preoptic area inhypogonadal mice. Exp NeuroI105:135-151, 1989

13. Broadweii RD, Charlton HM, Ebert PS, Hickey WF,Shirazi Y, Villegas i,Wolf AL: Aiiografts of CNS tissuepossess a blood-brain barrier. II. Angiogenesis in solidtissue and ceii suspension grafts. Exp Neurol112 (1):1­28,1991

14. Buchanan TJ, Nornes HO: Transplants of embryonicbrainstem containing the locus coeruleus into spinalcord enhance the hindlimb flexion reflex in adult rats.Brain Res 381:225-236, 1986

15. Bunge RP, Johnson MI, Thuline D: Spinal cordreconstruction using cultured embryonic spinal cordstrips, in Kao CC, Bunge RP, Reier PJ (eds), Spinal CordReeonstruction, New York: Raven Press,1983: 341-358

16. Contreras FL, Kadekaro M, Eisenberg HM: The effectof hyperbaric oxygen on glucose utilization in a freeze­traumatized rat brain. J Neurosurg 68:137-141, 1988

17. Das GD: Extraparenchymal neural transplants: Theircytology and survivability. Brain Res 241: 182-186,1982

18. Das GD: Neural transplantation in the spinal cord ofthe adult mammal, in Kao CC, Bunge RP, Reier PJ (eds),Spinal Cord Reconstruction, New York: Raven Press,1983: 367-396

19. Das GD: Neural transplantation in mammalian brain:Some conceptual and technical considerations, in

100

Kiitlay: HBO & Fetal SpilZal Cord Grafts

Wallace RB, Das GD (eds), Neural TissueTransplantation Research, New York: Springer­Werlag, 1983: 1- 64

20. Das GD: Neural transplantation in spinal cord underdifferent corditions of lesions and their functional

significance, in Das GD, Wallace RB (eds), NeuralTransplantation and Regeneration, New York:Springer- Verlag, 1986: 1-61

21. Davis JC, Landeen JM, Levine RA: Pyodermagangrenosum: Skin grafting af ter preparation withhyperbaric oxygen. Plast Reconstr Surg 79(2):200-207,1987

22. Davis JC, Gates GA, Lerner C, Davis MG Jr, Mader JT,Dinesman A: Adjuvant hyperbaric oxygen inmalignant external otitis. Arch Otolaryngol Head eckSurg. 118(1):89-93, 1992

23. Esterhai JH, PisareHo i, Brighton CT, Heppenstaii RB,Gelirnan H, Goldstein G: Adjunctive hyperbaricoxygen therapy in treatment of chronic refractoryosteomyelitis. J Trauma 27:763-768, 1987

24. Fischer B, Jain KK, Braun E, Lehrl S: Hyperbaricoxygenation in plastic surgery, in Fischer B, Jain KK,Braun E, Lehrl S (eds), Handbook of HyperbaricOxygen Therapy, Berlin: Springer Verlag, 1988: 103­112

25. Fischer B, Jain KK, Braun E, Lehrl S: Hyperbaricoxygenation in traumatology and orthopedics, inFischer B, Jain KK, Braun E, Lehrl S (eds), Handbookof Hyperbaric Oxygen Therapy, Berlin: SpringerVerlag, 1988: 113-119

26. Fischer B, Jain KK, Braun E, Lehrl S: Hyperbaricoxygenation in disorders of the nervous system, inFiseher B, Jain KK, Braun E, Lehrl S (eds), Handbookof Hyperbaric Oxygen Therapy, Berlin: SpringerVerlag, 1988:134-175

27. Fujita T, Kitani Y: Intracellular electrolytes inerythrocytes during and after shock: Relation toimpaired consciousness. J Trauma 18:345-354, 1978

28. Gelderd JB, Quarles JE: A preliminary study ofhomotopic fetal cortical and spinal cotransplants inadult rats. Brain Res Bu1l25:35-48, 1990

29. Giacobini MMJ, Hoffer BJ, Zerbe G, Olson L: Acidicand basic fibroblast growth factors augment growthof fetal brain tissue grafts. Exp Brain Res 86:73-81, 1991

30. Gonya-McGee T, Stokes BT: Acute modification ofembryonic spinal cord activity induced by hypoxia.Dev Neurosci 3:11-18, 1980

31. Henschen A, Zerbe G, Nadzan A, McKelvy J, Olson L,Hoffer B: Thyrotropin releasing hormone augmentsgrowth of spinal cord transplants in oculo. Exp Neurol102:125-129, 1988

32. Holbach KH, Wassmann H, Hohelucten LK, Linke D,Ziemann B: Clinical course of spinal lesions treatedwith hyperbaric oxygen. Acta Neurochir 31:297-298,1975

33. Holbach KH, Caroli A, Wassmann H: Cerebral energymetabolism in patients with brain lesions of normo­and hyperbaric oxygen pressures. J Neurol 217:17-30,1977

34. Houle JD, Reier PJ: Transplantation of fetal spinal cord

Turkis/i Neiirosiirgenj 10: 90 - 102, 2000

tissue into the chronically injured adult rat spinal cord.J Comp Neurol 269:535-547, 1988

35. Houle JD, Reier PJ:Regrowth of calcitonin gene-relatedpeptide (CGRP) immunoreactive axons from thechronically injured rat spinal cord tissue transplants.Neurosci Lett: 103: 17-23, 1989

36. Ikeda Y, Long DM: The molecular basis of brain injuryand brain edema: The role of oxygen free radicals.Neurosurgery 27:1-11,1990

37. Iwatsuki N, Takahashi M, Katsuhiko O, Tajima T:Hyperbaric oxygen combined with nicardipineadministration accelerates neurologic recovery aftercerebral ischemia in a canine modeL. Critica i CareMedicine 22(5):858-863, 1994

38. Jaeger CB, Lund RO: Transplantation of embryonicoccipital cortex to the tectal region of newborn rats: alight microscopic study of organization andconnectivity of the transplants. J Comp NeuroI194:571­597, 1980

39. Jiang J, Tyssebotn i: Cerebrospinal fluid pressurechanges after acute carbon monoxide poisoning andtherapeutic effects of normobaric oxygen in consciousrats. Undersea Hyper Med 24(4):245-254, 1997

40. Kapp JP, Philips M, Markov A, Smith RR: Hyperbaricoxygen af ter circulatory arrest: modification ofpostischemic encephalopathy. Neurosurgery 11:496­499, 1982

41. Kelly OL, Lassiter KRL, Vongsvivut A, Smith JM:Effects of hyperbaric oxygenation and tissue oxygenstudies in experimental paraplegia. J Neurosurg 36:425­429,1972

42. Kiliara M, Mc Manis PG, Schmelzer JD, Kihara Y, LowPA: Experimental ischemic neuropathy: Salvage withhyperbaric oxygenation. Ann Neurol 37:89-94, 1995

43. Korfali E, Doygun M, Ulus IH, Rakunt C, Aksoy K:Effects of neuronotrophic factors on adrenal medullagrafts implanted into adult rat brains. Neurosurgery22: 994-998, 1988

44. Korn HN, Wheleer ES, MilIer TA: Effect of hyperbaricoxygen on second degree burn wound healing. ArdiSurg 122:732-737, 1977

45. Kromer LF, Björklund A, Stenevi U: Innervation ofembryonic hippocampal implants by regeneratingaxons of cholinergic septal neurons in the adult rat.Brain Res 210:153-171, 1980.

46. Kromer LF, Björklund A, Stenevi U: Regeneration ofthe septohippocampal pathway in adult rats ispromoted by utilizing embryonic hippocampalimplants as bridges. Brain Res 210:173-200, 1981

47. Krum JM, Rosenstein JM: Patterns of angiogenesis inneural transplant models. i. Autonomic tissuetransplants. J Comp Neurol 258:420-434, 1987

48. Krurn JM, Rosenstein JM: Patterns of angiogenesis inneural transplant models. II . Fetal neocorticaltransplants. J Comp Neurol 271:331-345, 1988

49. Kunkel-Bagden E,Bregman BS:Spinal cord transplantsenhance the recovery of locomotor function after spinalcord injury at birth. Exp Brain Res 81:25-34, 1990

50. Kutlay AM, Demircan MN, Elbüken E, Basekim Cç,Baloglu H, Alkan K: Effect of hyperbaric oxygen

Kutlay: HBO & Fetal Spiiiiil Cord Grafts

therapy on fetal cortical grafts. Türk NöroQir Derg 6:81­89, 1996

51. Labbe R, Firl Jr A, Mufson EJ, Stein DG: Fetal braintransplants: Reduction of cognitive deficits in rats withfrontal cortex lesions. Science 221:470-472, 1983

52. Lawrence JM, Huang SK, Raisman LG: Vascular andastrocytic reactions during establishment ofhippocampal transplants in adult host brain.Neuroscience 12:745-760, 1984

53. Leigh K, Elisevich K, Rogers KA: Vascularization andmicrovascular permeability in solid versus cell­suspension embryonic neural grafts. J Neurosurg81:272-283, 1994

54. Maciunas RJ, Tulipan NB, Burns RS, Allen GS:Transplantation of tissue to the brain, in Wikins RH,Rengachary SS (eds), Neurosurgery, New York:McGraw-HilI: 1996: 4199-4204

55. Mader JT, Hicks CA, Calhoun J: Bacterial osteomyelitis.Adjunctive hyperbaric oxygen therapy. Orthop Rev18(5):581-585, 1989

56. Mader JT, Adams KR, Wallace WR, Calhoun JH:Hyperbaric oxygen as adjunctive therapy forosteomyelitis. Infect Dis Clin North Arn 4(3):433-440,1990

57. Manson PN, Im MJ, Myers RA, Hoopes JE: Improvedcapillaries by hyperbaric oxygen in skin flaps. SurgForum 31:564-566, 1980

58. Marx RE, Ehler WJ, Tayapongsak P, Pierce LW:Relationship of oxygen dose to angiogenesis inductionin irradiated tissue. Arn J Surg 160(5):519-524, 1990

59. McManis PG, Low PA: Factors affecting the relativeviability of centrifascicular and subperineurial axonsin acute peripheral nerve ischemia. Exp NeuroI99:84­95,1988

60. Meltzer T, Myers B: The effect of hyperbaric oxygenon the bursting strength and rate of vascularization ofskin wounds in rats. Amer Surgeon 52:659-662, 1986

61. Miller JD, Ledingham IM, Jennett WB: Effects ofhyperbaric oxygen on intracranial pressure andcerebral blood flow in experimental cerebral edema. JNeurol Neurosurg Psychiatry 33:745-755, 1970

62. Muizelaar JP: Cerebral blood flow, cerebral bloodvolume, and cerebral metabolism after severe headinjury, in Becker DP, Gudeman SK (eds), Textbook ofHead Injury, Philadelphia: WB Saunders, 1989: 221­240

63. Nemiroff PM, Lungu AP: The influence of hyperbaricoxygen and irradiation on vascularity in skin flaps: Acontrolled study. Surg Forum 38:565-567, 1987

64. Nemiroff PM: Hyperbaric oxygen in skin grafts andflaps, in Kindwall EP (ed), Hyperbaric MedicinePractice, Arizona: Best Publishing Co., 1995: 565-579

65. Neubauer RA, End E: Hyperbaric oxygenation as anadjunct therapy in strokes d ue to thrombosis: A reviewof 122 patients. Stroke 11:297-300, 1980

66. Nornes H, Björklund A, Stenevi U: Reinnervation ofthe denervated adult spinal cord of rats by intraspinaltransplants of embryonic brainstem neurons. Cell TissRes 230:15-35, 1983

67. Ogilvie RW, Balentine JD: Oxygen tension in the spinal

101

Tiirkish Neiirosiirgery 10: 90 - 102, 2000

gray matter during exposure to hyperbaric oxygen. JNeurosurg 43: 156-161,1975

68.01son L, Seiger A, Strömberg i: Intraoculartransplantation in rodents. A detailed account of theprocedure and examples of its use in neurobiology withspecial reference to brain tissue grafting, in Fedoroff S(ed), Advances in Cellular Neurobiology, New York:Academic Press, 1983: 407-442

69. Palaoglu S, Benli K, Pamir N, Erbengi T, Erbengi A:Examination of autologous and embryonic corticalbrain tissue transplantation to adult brain cortex in rats.Surg Neurol 29:183-190, 1988

70. Palaoglu S, Erbengi A, Sav A, Erbengi T, Türker RK:Cytoprotective effect of iloprost on isolated corticalbrain tissue grafts in rats. Prostog Leukotr Ess 39:203­206,1990

71. Pallini R, Fernandez E, Gangitano C, Olivieri­Sangiacomo AC, Sbriccoli A: Studies on embryonictransplants to the transected spinal cord of adult rats.J Neurosurg 70: 454-462, 1989.

72. Patel U, Bernstein JJ: Growth, differentiation andviability of fetal rat cortical and spinal cord implantsinto adult rat spinal cord. J Neurosci Res 9:303-310,1983

73. Privat A, Mansour H, Rajaofetra N, Geffard M:Intraspinal transplants of serotonergic neurons in theadult rat. Brain Res Bull22:123-129, 1989

74. Reier PJ, Bregman BS, Wujek JR: Intraspinaltransplantation of embryonic spinal cord tissue inneonatal and adult rats. J Comp Neurol 247:275-296,1986

75. Reier PJ, Houle JD: The glial scar: Its bearing on axonalelongation and transplantation approaches to CNSrepair, in Waxman SG (ed), Functional Recovery inNeurological Disease, New York: Raven Press, 1988:87-138

76. Reier PJ, Stokes BT, Thompson FJ, Anderson DK: Fetalcell grafts into resection and contusion i compressioninjuries of the rat and cat spinal cord. Exp Neurol115:177-188, 1992

77. Rockswold GL, Ford SE, Anderson OC, Bergman TA,Sherman RE: Results of a prospective randomized trialfor treatment of severely brain-injured patients withhyperbaric oxygen. J Neurosurg 76:929-934,1992

78. Sensenbrenner M, Wittendorp E, Barakat i,Rechenmann RV: Autoradiographic study ofproliferating brain cells in culture. Dev Biol 75:268-277,1980

79. Sinson G, Voddi M, McIntosh TK: Combined fetalneural transplanta tion and nerve growth factorinfusion: Effects on neurological outcome following

102

Kiitlay: HBO & Fetal Spiiial COl'rl Gmfts

fluid-percussion brain injury in the rat. J Neurosurg84:655-662, 1996

80. Skyhar MJ, Hargens AR, Strauss MB, Hart GB, AkesonWH: Hyperbaric oxygen reduces edema and necrosisof skeletal muscle in compartment syndromesassociated with hemorrhagic hypotension. J Bone JointSurg (Am) 68A:1218-1224, 1986.

81. Stokes BT, Reier PJ: Oxygen transport in intraspinalfetal grafts: Graft-host relations. Exp Neurol 111:312­323, 1991.

82. Stokes BT, Reier PJ:Fetal grafts alter chronic behavioraloutcome after contusion damage to the adult rat spinalcord. Exp NeuroI116:1-12, 1992

83. Strauss MB, Hargens AR, Gerslmni OH, GreenbergDA, Crenshaw AG, Hart GB, Akeson WH: Reductionof skeletal muscle necrosis using intermittenthyperbaric oxygen in a model compartInent syndrome.J Bone Joint Surg 65-A: 656-662, 1983.

84. Sukoff MH, Ragatz RE: Hyperbaric oxygenation forthe treatment of acute cerebral edema. Neurosurgery10:29-38, 1982.

85. Tessler A: Intraspinal transplants. Ann NeuroI29:115­123, 1991.

86. Tessler A, Fischer I, Giszter S, Himes BT, Miya D, MoriF, Murray M: Embryonic spinal cord transplantsenhance locomotor performance in spinalized newbomrats. Adv Neurol 72:291-303, 1997

87. Tuszynski MH, Buzsaki G, Gage FH: Nerve growthfactor infusions combined with fetal hippocampalgrafts enhance reconstruction of the lesionedseptohippocampal projeetion. Neuroscience 36:33-44,1990

88. Weinstein R, Hameroff SR, Johnson PC, Anderson CG:Effect of hyperbaric oxygen therapy or demethylsulfoxide in cerebral ischemia in unanesthetized

gerbils. Neurosurgery 18:528-532, 198689. WujekJR, Reier PJ: Astrocytic membrane morphology:

Differences between mammalian and_amphibianastrocytes after axotomy. J Comp Neuro1222: 607-619,1984

90. Zamboni WA, Roth AC, Russell RC, Graham B, SuchyH, Kucan JO: Morphological analysis of themicrocirculation during reperfusion of ischemicskeletal muscle and the effect of hyperbaric oxygen.Plast Reconstr Surg 91:1110-1123,1993

91. Zhou CF, Lindsay RM, Lawrence JM, Raisman C:Extent of survival and vascularization of adult superiorcervical sympathetic or nodose ganglia transplantedinto the septal nuclei or choroid fissure of adult rats.Neuroscience 17:803-813, 1986