KINETICS OF E-SELECTIN EXPRESSION IN SURGICAL FLAPS

During the ischemia/reperfusion phenomenon, adhesion molecules seem to play a critical role in the recruitment of neutrophils to sites of ev entual tissue injury. E-selectin is an endothelium-derived molecule th at mediates adhesion of neutrophils to activated endothelial cells. In vitro expression of E-selectin, after exposure to stimuli such as end otoxin, interleukin 1, or tumor necrosis factor alpha is maximal at 4 to 6 h, followed by a decline toward basal levels at 24 to 48 h. Chara cterizing the temporal expression of E-selectin in an in vivo model of skin flap ischemia-reperfusion would help to determine the optimal ap proach to eventual pharmacologic blockade. This intervention may prove therapeutically beneficial in attenuating flap injury. This study, us ing the standard porcine buttock skin flap model, was designed to eval uate immunohistochemically the expression of E-selectin in flaps subje cted to (7) arterial ischemia (8 h)-reperfusion (18 h), (2) venous isc hemia (8 h)-reperfusion (18 h), and (3) distal ischemia (26 h). Four f laps were examined per group, with 8 biopsies being collected sequenti ally over the 26-h study period from each nap. Blinded, semi-quantitat ive histologic scoring revealed the following results: (1) E-selectin is absent in normal porcine skin; (2) with arterial ischemia/reperfusi on, E-selectin expression in flaps was maximal at 1 h of reperfusion, declining thereafter ; (3) with venous ischemia/reperfusion, E-selecti n expression peaked during the first hour of ischemia, with subsequent decline: and (4) within a flap designed to sustain distal ischemia, E -selectin expression is relatively more intense in regions of tile fla p distant from the vascular pedicle, and maximal at 6 h after flap ele vation. Our conclusion, therefore, is that tile kinetics of E-selectin expression within the tissues of porcine skin flaps differs depending on the tt-pc of ischemic insult sustained, Interpretation of these fi ndings, correlating possible pathophysiologic differences in tile diff erent models of ischemia, is offered.