This study investigated the possibility that pericapillary fibrin deposition, found in the calf skin of patients with venous ulceration and lipodermatosclerosis, might already be present in the dermis of the gaiter area of apparently healthy limbs before any skin changes were visible. The apparently healthy limbs of 19 consecutive patients with a healed venous ulcer on one leg and no history of ulceration or clinical evidence of lipodermatosclerosis in the opposite calf, were studied. Bipedal ascending phlebography and foot volume plethysmography were performed, and systemic fibrinolytic activity and fibrinogen levels were calculated. Transcutaneous oxygen measurements were expressed as a ratio of levels from a fixed position in the gaiter skin over a control site on the arm. Biopsies of a standard site in the gaiter skin and the thigh were assessed for the presence of laminin, fibrinogen and fibronectin using immunofluorescent microscopy. The extent of pericapillary fluorescence was expressed as a ratio of the number of capillaries with deposits divided by the total number of capillaries staining with laminin (fibrin and fibronectin scores). Pericapillary fibrin deposits were observed in the dermis in 16 of the biopsies of the gaiter region (median score 0.20), and in eight of the biopsies from the thigh (median score 0.0). This difference was highly significant (P<0.01, Wilcoxon signed rank test). The transcutaneous oxygen ratio correlated negatively with the fibrin score (Spearman rank correlation coefficient -0.62, P<0.01), and there was a weak negative correlation between the half volume refilling time on foot volume plethysmography (an indicator of venous reflux) and the fibrin score (Speraman rank correlation coefficient -0.47, P<0.05). No such correlation could be shown between the fibrin score and the indicators of calf pump function, the euglobulin clot lysis time or the plasma fibrinogen. The presence of significant numbers of pericapillary fibrin deposits within the dermis of the gaiter skin has been demonstrated in 84% of this cohort of 'at risk' limbs before there is any evidence of clinical lipodermatosclerosis.