Russell Dilley

Seeding arterial prostheses with vascular endothelium. The nature of the lining.

Arterial prostheses seeded with autogenous vascular endothelium demonstrate a well-organized, cellular, inner lining. To determine the nature of the lining cells, six animals underwent replacement of the infrarenal aorta with Dacron® prostheses. During the preparation of three such grafts, endothelium was scraped from the saphenous vein with a steel wool pledget, suspended in chilled Sacks solution, and mixed with blood used to preclot the graft. This suspension was omitted from the three control grafts. After six weeks, the grafts were removed, rinsed and examined. Fluorescent Factor VIII related antigen (F VIIIRA) strongly stained the lining cells. Silver nitrate Haütchen and electron microscopy preparations revealed a lining pattern characteristic of vascular endothelium. Endothelial cell-specific Weibel-Palade bodies were identified in the lining cell cytoplasm. Massons trichrome staining revealed a relatively collagen-poor connective tissue within the seeded fabric. Transmission electron microscopy disclosed vascular smooth muscle cells between the seeded graft fabric and the lining cells. Vasa vasorum, arising from the outer capsule, penetrated the fabric to supply the inner capsules of the seeded grafts. It is concluded that the cells lining seeded canine arterial prostheses are true vascular endothelium supported by vascular smooth muscle cells, that the lining contains minimal connective tissue, and that vasa vasorum develop. Unseeded control grafts lacked these features.

Seeding endothelium on canine arterial prostheses--the size of the inoculum.

Nineteen dogs were studied to determine the minimum length of saphenous vein which will yield sufficient endothelium to line arterial prostheses interposed in the infrarenal aorta. Knitted Dacron grafts preclotted by the Sauvage technique were seeded by endothelial cells mechanically harvested from lengths of autogenous saphenous vein varying from 3 to 150 mm. After 1 month, the grafts were removed; the percentage of clot-free surface was determined; the inner surface area of the grafts was calculated (Ag), and the linings were examined histologically. The inner surface areas of the donor vein (Av) and the installed graft (Ag) were computed. The percentage of clot-free surface was 87.6 ± 9.4 in dogs in which the ratio of the area of the vein to that of the graft (Av/Ag) was greater than 0.425 as compared to 25.9 ± 25.9 in others (P 0.425 had thinner inner capsules (217 ± 99 versus 480 ± 222 μm, P < 0.0005). We conclude that the amount of venous surface necessary to provide endothelium to line an arterial prosthesis using these techniques is slightly less than half the area of the graft to be lined. The size of undistributed venous endothelial cells was used to calculate the expected cell yield. In nine additional dogs the number of endothelial cells in the suspensions were compared to Av. If all cells were harvested and seeded successfully, the Av/Ag ratio might be lowered from 0.425 to 0.031.

Graft material, length, and diameter determine the patency of small arterial prostheses in dogs

Abstract Thirty-two dogs were studied to determine the effects of graft material, length (L), and diameter (i.d.) on the patency of small arterial prostheses. Knitted Dacron (n = 15) and expanded polytetrafluoroethylene (e-PTFE, n = 17) were installed in the common carotid and common femoral arteries of dogs. The grafts were removed one month after implantation and their patency noted. Two of three Dacron grafts measuring 4 mm i.d. and 20 mm in length were patent, whereas none of the eight measuring 40 mm or more in length remained open (P = 0.01). With 4-mm e-PTFE grafts, 80% of the 40-mm and none of the 60-mm lengths of graft remained patent (n = 8, P = 0.03). When the grafts measured 4 mm i.d. and 40 mm in length, no Dacron grafts (n = 4) and 80% of the e-PTFE grafts (n = 5) remained patent (P = 0.016). With 3-mm grafts, the lengths had to be much shorter to insure that any grafts remained open, and even small differences approached significance: 60% of those 4 mm long were patent compared to 14% of those 8 mm long (P = 0.07). There was no difference between Dacron and e-PTFE. Given the clinical observation that much longer grafts made of both of these materials generally remain patent when i.d. = 6, equations predicting the maximum critical length consonant with a reasonable incidence of graft patency may be derived for each material. For e-PTFE: L c = r 4 (2.41 - 1.05r) , and for knitted Dacron: L c = r 4 (1.60 - 0.53r) , where Lc = maximumcriticallength of a particular graft and r = radius. We conclude that the patency of small arterial prostheses in dogs is a function of their length, the fourth power of their radius, and the material from which they are constructed.

Immunofluorescent staining for factor VIII-related antigen: A tool for study of healing in vascular prostheses

Abstract Localization of Factor VIII-related antigen by indirect immunofluorescent microscopy reliably identifies endothelium in frozen sections of canine blood vessels, including vasa vasorum. Using this technique, endothelium was found to line 6-cm-long porous Dacron arterial grafts in place in dogs for 6 months and 1 year. Grafts seeded with autogenous, venous endothelial cells at implantation showed specific fluorescence in the midportions of their flow surfaces after 1 month. Unseeded grafts were devoid of specific fluorescence beyond the region of pannus ingrowth. Vasa vasorum were very well delineated, even when no more than a row of single cells. We conclude that localization of Factor VIII-related antigen reliably identifies true endothelium in vascular grafts and will be useful in the study of vascular prosthetic healing.