Capillary bleeding under oral anticoagulation

Oral anticoagulants are routinely used for prevention of thromboembolism in cardiac, arterial or venous diseases. Hemorrhages are serious treatment complications, frequently occurring under long-term and/or high-dose regiments. From animal experiments it is known that coumarin-type anticoagulants may cause capillary dilatation and increased permeability, red blood cell extravasation and punctate bleeding. Controlled human trials are lacking.

31 patients under oral anticoagulation were examined by video capillary microscopy. 52 patients with comparable diseases and treatment but without oral anticoagulation served as controls. Nailfold capillaries of four fingers of each hand were examined and analyzed off-line according to the following criteria: (1) numbers of capillaries investigated, (2) numbers of capillary bleedings, and (3) bleeding incidence (bleedings per 100 capillaries).

In 23 out of 31 patients (74.2%) capillary bleedings were observed. The bleeding incidence ranged from 0.33 to 4.29 per 100 capillaries. In contrast, only 4 out of 52 controls were detected with capillary bleedings (2.1%, p<0.001). The bleeding incidence was 0.34-2.41. In patients on anticoagulation there was no correlation between the number of capillary bleedings and the INR or Quick values. During a two year follow-up of patients on oral anticoagulation no significant difference was found in terms of clinically obvious bleedings in patients with or without capillary bleedings.

This study shows that capillary bleedings can be demonstrated in patients on oral anticoagulation. Bleedings occur independent of the INR-value. Thus, other factors than the vitamin-k-dependent coagulation effect seem to be causal for the damage of microvessels. Further, the evidence of capillary bleedings is not a prognostic indicator for future hemorrhage.