J Neurosurg 81:272-283, 1994 Vascularization and microvascular permeability in solid versus cell-suspension embryonic neural grafts KERRIE LEIGH, B.Sc., KOST ELISEVICH, M.D., PH.D., AND KEM A. ROGERS, PH.D. Departments of Anatomy and Clinical Neurological Sciences (Division of Neurosurgery), University of Western Ontario and Victoria Hospital, London, Ontario, Canada 0," Vascularization and rnicrovascular permeability were assessed in a comparative study of solid (organized) and cell-suspension (dissociated) fetal nigral grafts implanted in the dopamine-deprived striatum of adult rats. Both graft types were analyzed by chromogen detection of intravenously injected horseradish peroxidase (HRP), which outlined vessel walls, and, in cases in which the blood-brain barrier was compromised, permeated the graft and host parenchyma. Survival of graft-derived dopaminergic cells was assessed using tyrosine hydrox- ylase (Ill) immunocytochemistry. Glint reactivity to cell-suspension grafts was similarly assessed with an antibody directed against glial fibrillary acidic protein. Morphome~ry revealed significantly higher mierovessel density in the cell-suspension grafts (p < 0.001), which effectively equaled that found in the contralateral striatum despite rather prominent surrounding glial reactivity. Capillaries in the cell-suspension grafts were not permeable to blood-borne HRP at postimplantation study times of 7, 14, and 30 days whereas, in the solid grafts, per- meability in some cases could be detected for up to 30 days. Large numbers of cells immunoreactive for TH were seen in cell-suspension grafts; in contrast, few if any were found in the majority of solid transplants. The multiple-fragment solid graft implant model used clinically compares poorly with the cell-suspension model because it lacks consistency in early revascularization and shows a greater (albeit temporary) tendency for blood-brain barrier dysfunction. Delayed and inadequate vascularization of the solid graft is likely to account for graft failure more often than in the cell-suspension graft. Similarly, a certain critical number of specific grafted cells are required to achieve sufficient expression to bring about a favorable response in the disabled host, and this expression appears to be achieved less consistently with the solid implant technique. KEY WORDS 9 neural tissue graft ' neovascularization 9 blood-brain barrier 9 microvascular permeability 9 solid graft cell-suspension graft . rat M UCH debate has arisen regarding the success of neural tissue transplantation in reversing cer- tain clinical features of degenerative neuro- logical illnesses in humans (such as Parkinson's dis- ease) and in restoring a disrupted neuronal circuitry in experimental animal models that attempt to simulate this human condition. Regardless of the mechanisms by which such functional recovery is achieved, vas- cular integration of the transplanted tissue to avoid ir- reparable ischemie injury is essential for eventual sur- vival of the transplant. Postmortem studies have not identified surviving transplanted cells in patients who underwent solid implants of adrenal medullary tissue to the head of the caudate nucleus despite exposure to cerebrospinal fluid in the neighboring ventricle. 1923,33,52 A multiple-fragment solid graft technique had been used routinely in a single-stage procedure. These find- ings parallel the rather dismal long-term clinical out- comes that had already been forecast on the basis of prior experimental work with single-fragment grafts, despite the intraparenchymal placement of autologous tissues. 25Similarly, long-term survival of organized au- tonomic tissue transplants has not commonly been seen experimentally 25 despite the demonstration that host central nervous system (CNS) vessels proliferate and anastomose with the graft vasculature in as little as 24 hours postoperatively. 39,73 Increased trauma from im- plantation of solid grafts in the host parenchyma in- evitably leads to delays in graft reperfusion attributable to scar formation, thus lowering the ganglion cell sur- vival rate. 39 Solid neural allografls, usually harvested from em- bryonic neocortex and implanted intraparenchymally, have shown a similar but delayed capacity for neovas- cularization, with patent vessels apparent in the graft by 72 hours and a more extensive vascular network by 5 days. 4~In some cases, adequate graft perfusion failed to occur within 7 days posttransplantation} 4 A menin- 272 J. Neurosurg. / Volume81 / August, 1994