In his review of the medicinal leech, Adams [1] referred to leeching in the prevention of postoperative thromboses. Lillienthal (referenced by Adams) wrote that "In early thrombophlebitis ... application of leeches ... will usually ... be followed by amazingly rapid and complete recovery" [2]. Despite the continued use of leeches in reconstructive surgery, no published data describe the effect of leeching on modern tests of coagulation. We conducted two such experiments.

Two Hirudo medicinalis were attached to the dorsal left hand of one investigator, and ipsilateral and contralateral values of activated clotting time, thrombin time, activated partial thromboplastin time, and prothrombin time were measured from the cephalic or median cubital vein. No effect on clotting times was noted during 2.5 hours. The second experiment, in which nine leeches were attached, was done 3 weeks later (Figure 1). All leeches fed until sated and spontaneously released after 40 to 90 minutes. The triradiate bite marks bled for 5 to 12 hours after detachment and were compressed. No changes in ipsilateral activated partial thromboplastin time or prothrombin time were noted 15, 60, 120, or 180 minutes after the first attachment (Table 1); the last sample was obtained 90 minutes after the last leech detached.

View larger version (111K): [in this window] [in a new window]   Figure 1. Seven of nine Hirudo medicinalis, near the end of the second of two experiments to define alterations in coagulation test results from leeching.

Natural and recombinant hirudin are nearly completely absorbed after subcutaneous injection, with peak plasma levels occurring at 1 to 2 hours [3], making it unlikely that a delayed systemic antithrombin effect was missed in our experiments. It seemed reasonable to conclude that the "dose" was insufficient, and we endeavored to generate an estimate of a "therapeutic dose" of leeches in humans.

In GUSTO II, a randomized trial of heparin compared with intravenous hirudin in the treatment of acute myocardial infarction or unstable angina, a hirudin bolus of 0.6 mg/kg plus an infusion of 0.2 mg/kg per hour was used, producing an approximate two- to threefold elevation of the activated partial thromboplastin time (GUSTO II Protocol. Unpublished data). Leeches weigh approximately 2 g, and one can extract approximately 80 to 140 000 thrombin inhibitory units of crude hirudin from 1 kg of leeches [4]. Recombinant hirudin contains 16 000 thrombin inhibitory units/mg of protein [5]. Thus, the infusion dose in GUSTO II for an 80-kg person is 256 000 thrombin inhibitory units/h, the equivalent of the total extractable hirudin from 1280 H. medicinalis.

In addition to hirudin, leech saliva contains several active substances, including inhibitors of platelet aggregation [6]. Although these substances may enhance the antithrombotic effect of leech hirudin, direct observations and pharmacokinetics suggest that physicians who have previously practiced leeching were not likely to have produced a regional or systemic anticoagulant effect. For modern practitioners of leeching, systemic bleeding is unlikely to complicate the therapy.