A low-flow adaptation phase improves shear-stress resistance of artificially seeded endothelial cells.

Introduction: The purpose of this study was to evaluate the effect of different adaptation phases on the shear-stress resistance of endothelial cells seeded artificially onto vascular prostheses and biological heart valves.
Material and Methods: Human endothelial cells (EC), fibroblasts (FB), and smooth muscle cells (SMC) were isolated from vena saphena magna pieces and expanded in culture. Group A: 15 polyurethane vascular grafts (20 mm diameter) were seeded with FB and SMC (53 +/- 1.2 million cells), followed by EC seeding (39 +/- 0.9 million cells). Group B: eight stentless porcine valves (Freestyle, Medtronic, USA) were seeded with FB (68 +/- 1.5 million cells) and EC (42 +/- 1.1 million cells). Shear-stress testing was done under pulsatile flow (pulse rate: 80 pulses/min.). Adaptation phase: flow was set to 0.9 +/- 0.3 l/min (systolic pressure: 40 - 50 mm Hg). High flow was 3.2 +/- 0.6 l/min. (systolic pressure: 140 - 160 mm Hg) and lasted over four hours in all groups. The vascular grafts were divided into three groups (n = 5 each): group 1 (high flow immediately), group 2 (adaptation phase of 15 minutes), and group 3 (adaptation phase of 30 minutes). The valves either were given high flow immediately (n = 4) or had an adaptation phase of 30 minutes (n = 4). Specimens were obtained after cell seeding, before, and after perfusion.
Results: A confluent EC layer was achieved on all grafts. After perfusion without adaptation, large defects within the cell layer were found. No FB and SMC were seen at the bottom of these defects. In group B, the defects were largest on the ventricular surface of the leaflets. After an adaptation phase of 15 minutes in group A, only a few defects within the EC layer were detected with a still confluent FB and SMC. After a 30-minute adaptation phase defects within the EC layer were very rare and no interruption of the underlying FB and SMC layer was seen. Immunohistochemical staining for factor VIII and CD31 proved the EC to be viable and staining for collagen IV and laminin revealed the formation of a basement membrane. After perfusion, the specimen also stained positive for eNOS.
Conclusion: An adaptation phase of 30 minutes proved to be sufficient to allow artificially seeded endothelial cells to adapt to shear stress. The formation of a basement membrane was of great importance for the maintenance of a confluent EC layer.