A normotensive model of hindlimb ischemia-reperfusion in Wistar rats was used to test the hypothesis that microvascular perfusion deficits contribute to the initiation of remote hepatic injury during a systemic inflammatory response. Animals were randomly assigned to one of three groups: 4 h of ischemia with 6 h of reperfusion (I/R-6; n = 4), 4 h of ischemia with 3 h of reperfusion (I/R-3; n = 5), or no ischemia (naive; n = 5). With intravital fluorescence microscopy, propidium iodide (PI; 0.05 mg/100 g body wt) was injected for the in vivo labeling of lethally injured hepatocytes (number/10(-1) mm(3)). PI-positive hepatocytes increased progressively over the 6-h period (naive 32.9 +/- 7.8 vs. I/R-3 92.8 +/- 11.5 vs. I/R-6 232 +/- 39.2), with no difference between periportal and pericentral regions of the lobule. Additionally, a significant decrease in continuously perfused sinusoids (naive 70.0 +/- 1.5 vs. I/R-3 65.0 +/- 1.0 vs. I/R-6 48.8 +/- 0.9%) was measured. Regional sinusoidal perfusion differences were only observed after 3 h of limb reperfusion. Indirect measures of hepatocellular injury using alanine transaminase levels support the progressive nature of hepatic parenchymal injury (0 h 57.8 +/- 6.5 vs. 3 h 115.3 +/- 20.7 vs. 6 h 125.6 +/- 19.5 U/l). Evidence from this study suggests that remote hepatic parenchymal injury occurs early and progresses after the induction of a systemic inflammatory response and that microvascular perfusion deficits are not essential for the initiation of such injury.