Br. J. Surg. 1992, Vol. 79, November. 1151 -1 153
J. S. Budd, K. Allen, J. Hartley, A. Walsh, R . F. L. James and P. R . F. Bell Department of Surgery, University of Leicester, Leicester, UK Correspondence to: Mr J. S . Budd, Department of Surgery, Clinical Sciences Building, Leicester Royal Infirmary, PO Box 65, Leicester LE2 7LX, UK
Prostacyclin production from seeded prosthetic vascular grafts Endothelial cell seeding has been proposed as a method of iinproving potency rates in small-calibre prosthetic vascular gr-afis. In vivo, endothelid cells norrnallj, produce prostacyclin (PGI,), a potent antiplatelet agent. The aini of’ this stud?’ was to deterniine whether seeded grclfis show signijicctnt PGI, production qfter in vivo iniplantation. Grafts were seeded with either autologous canine venous endothelial cells or autologous naicrovascular endotlielial cells. After 12 weeks, PGI, production was assessed under basal and stiniulated conditions. Seeded grufis nwe compared with non-seeded controls and the corresponding aorta. The overall paienc?, rate in seeded grafts was 80 per cent compared wiih 10 per cent in non-seeded grqfts (P t0.01). Grafts seeded witli cells froni either source produced signijicantlj, more PGI, than unseeded grafis in both basal cind stiniulated states (P < 0.05). The aorto produced significanrlj- inore PG12 tlian seeded grafts under both conditions ( P < 0.01).Endothelial cell seeding produces a .functional yruft and leads to an improved patencji rate.
Endothelial cells produce a n u m b e r of substances with either t h r o m b o t i c or a n t i t h r o m b o t i c activity; t h e balance i t ] vii:o is normally in favour of t h e a n t i t h r o m b o t i c properties. This activity is b r o u g h t a b o u t by a n u m b e r of mechanisms, b u t one t h a t is t h o u g h t to be particularly i m p o r t a n t is t h e p r o d u c t i o n of prostacyclin (PG12)‘. This potent inhibitor of platelet aggregation is released by normal endothelial cells in varying quantities depending on t h e vascular bed in question. It has ,previously been shown t h a t endothelial cell seeding, using a preformed confluent layer of endothelial cells on prosthetic grafts, reduces graft thrombogenicity a n d improves patency rate’. T h e present study w a s designed t o investigate t h e functional properties of grafts seeded with either a u t o l o g o u s venous endothelial or microvascular endothelial cells.
Materials and methods Ten conditioned Beagle dogs weighing 8-14kg were used for graft implantation. Anaesthesia was carried out using sodium thiopentone (0.5-1 mg:kg) for induction, followed by maintenance with a combination of halothane (1.5-2 per cent). oxygen and nitrous oxide. Grafts in five dogs were seeded with autologous venous endothelial cells and in the other five with microvascular endothelial cells obtained from omentum. Htrrrcs t i r q m i n i i s eritiorlirlitrl cel1.s Endothelial cells were harvested from the right external jugular vein of each animal. All side branches were tied while the vein was still iri s i r i i and the main length of vein handled as little as possible. The length of vein removed was about lOcm from each dog. The harvested vessel was placed in ice-cold minimum essential medium ( M E M ) (Flow Laboratories, Irvine, UK). The harvested vein was cannulated and then distended with about 2ml 0.1 per cent collagenase (Worthington CLS 1, New Jersey, USA) at 37-C. It was then incubated at 37 C for 15min. The cell suspension obtained after flushing the vein was spun at 200y for 7 min at 4 C and the cell pellet resuspended in 4 ml complete medium comprising medium M 199. 20 per cent fetal calf serum (FCS). 100 u n i t s h l penicillin, 100pg,’ml streptomycin, 20 mmoljl Hepes buffer. 2 mmol,’l glutamine, I mmol/l pyruvate (all Flow Laboratories), 20pg,’ml endothelial cell growth supplement and YOpgjml heparin (Sigma Chemicals. Poole, UK). Cells were then plated into a 25-cm’ tissue culture flask and grown for 7 days.
Horrcst it i y n i icrorri.~ ciiltir c~ritlotheliulcells
Endothelial cells were obtained from pieces of omentum. A small midline abdominal incision was made under general anaesthesia using sterile operating conditions. The omentum was delivered through the incision and a piece estimated to be about l o g in weight excised. The incision was closed and the animal allowed to recover. The excised omentum was placed in ice-cold MEM until further processed. was used to harvest A modification of the method of Kern er cells. The omentum was processed within 1 h of harvesting and all procedures carried out under sterile conditions. The first step in the processing was introduced as an addition to Kern‘s method to remove as many mesothelial cells as possible before definitive endothelial cell harvest, This entailed incubating the intact specimen of omentum in l O m l O ~ 1per cent collagenase at 37.C for 20min. After this preliminary incubation. the omentum was washed four times in MEM and then minced finely using two scalpel blades. The minced tissue was divided into four equal aliquots. each of which was placed in a 20-ml universal container (Sterilin; Corning. Stone. U K ) containing 7ml 0.1 per cent collagenase prewarmed to 37 C . The collagenase solution for this digestion was made up in MEM with 3mmol,’l glucose and 4 per cent bovine serum albumin (BSA) (APP. Brierley Hill, UK). The tubes were then incubated at 37 C on a r3ller-table rotating at 30rev;min. After 15 min, the tubes were vigorously shaken and the resultant suspension from all four tubes passed through a 25o-pm pore nylon mesh (Lockertex. Warrington. U K ) to separate the undigested pieces. The tubes were then washed out with MEM containing 10 per cent FCS and this was also passed through the mesh. The filtrate was collected and centrifuged at 300y for 7 min; the supernatant was then removed and the cell pellet resuspended in 3ml MEM with 10 per cent FCS. This cell suspension was subsequently filtered through 30-pm pore size nylon mesh (Lockertex). Clumps of endothelial cells were retained on the mesh and recovered by cutting out the mesh, transferring it to a centrifuge tube containing 20ml MEM. and shaking it to dislodge the cells. The resulting suspension was centrifuged at 300~1for 7min and the cell pellet resuspended in 5 ml medium M 199 with 20 per cent FCS. This suspension was then layered on to 15 ml BSA ( 5 per cent in medium M199). After 10min at room temperature, the upper 5ml of the gradient. which should contain any contaminating isolated stromal cells. was discarded. The remainder of the gradient was centrifuged at 3Wq for 7 min and the cell pellet finally resuspended in 4 ml complete medium. The cells were then plated into a 25-cm2 tissue culture flask and grown for 7 days. sfW/~/l(lO f ’ (JI.(I/f.F
After 7 days in tissue culture, the cells were used to seed polytetraHuoroethylene ( P T F E ) grafts. A 15-cm length of PTFE graft 4 m m in internal diameter (a gift from W. L. Gore. Woking, U K ) was first secured in place inside a seeding chamber as previously described’.
0 1992 Butterworth-Heinemann
Seeded prosthetic vascular grafts: J. S. Budd et al. The chamber with the graft i n siru was sterilized and the luminal surface ofthe graft then coated with a 20-pg,'ml fibronectin solution (Sigma). Endothelial cells were released from the tissue culture flask with trypsin and resuspended in 8 ml complete medium. A cell count was made at this point using a haemocytometer. Cells were introduced into the graft in aliquots of 2ml at 20-min intervals and with 90 rotation of the graft between each stage. The final aliquot was left inside the graft. The chamber was then placed on a roller-table rotating at 10 rev/h inside an incubator at 37 C. After incubation overnight, the graft was ready for implantation. GrcIfi impluntatinn Grafts were placed in the femoral region. Each animal received a graft seeded with its own endothelial cells and a contralateral non-seeded control graft. Each dog therefore acted as its own control. The profunda femoris artery was tied in continuity. Just before occlusion of the femoral vessel, the dog was given 100 units,'kg heparin. Once occluded, an 8-cm length of the superficial femoral artery was excised. The grafts were then used to bridge the defect caused by this excision. The control graft consisted of an 8-cm length of plain PTFE cut from the same length of graft as that used for seeding. The unused portion of the seeded graft was stained with toluidine blue to assess the degree of cell coverage at implantation. Grafts were anastomosed end to end using 7,'O polypropylene sutures.