Initial platelet deposition (PD) in and around the region of a small-vessel anastomosis may set the stage for thrombosis and tissue loss. To study this problem, a human vessel model (human placental artery, HPA) has been designed to mimic the vascular injuries attendant on clinical microsurgery. To perform these studies, dissected lengths of human placental artery were treated to provide the following four types of injury: group I: control, dissected but otherwise uninjured (N = 5); group II: distal portion of vessel endothelium removed (N = 5); group III: central anastomosis, distal endothelium intact (N = 7); and group IV: central anastomosis, distal endothelium removed (N = 4). Vessels were perfused with 25 ml human whole blood for 17 +/- 5 s at an average shear rate of 536 s-1. Vessels in groups I to IV were segmented at 2-cm intervals, and the number of In-111-labeled platelets was measured. Data from the following groups of exposure zones were pooled and analyzed: endothelium intact, endothelium absent, anastomosis present, postanastomosis with endothelium intact, and postanastomosis with endothelium absent. Significant numbers of platelets were found to attach to intact endothelium, indicating that ischemia and microsurgical handling may augment platelet deposition to otherwise uninjured vessels. A similar degree of platelet deposition was measured after exposure of the subendothelium and perfusion, indicating that superficial subendothelial exposure in the absence of an additional prothrombotic stimulus may lead to no greater platelet deposition than occurs on slightly injured endothelium alone. Platelet deposition at anastomoses was strikingly elevated, although the anastomosis had no additive effect on platelet deposition to downstream endothelium. In contrast, an upstream anastomosis significantly augmented platelet deposition to exposed downstream subendothelium.