Coagulopathy Induced by Hydroxyethyl Starch

Synthetic colloids are commonly administered as an alternative to the limited supplies of natural albumin. Among these solutions, hydroxyethylstarch (HES) solution is widely used for blood volume expansion with limited side effects. Nevertheless, a coagulopathy can occur during HES administration, usually in a dose-related manner, consisting of a prolongation of the activated partial thromboplastin time (APTT) and a decrease in factor VIII [1,2]. Bleeding has occurred in neurosurgical patients after the infusion of HES solutions [2-4], but the amount of solution administered was usually greater than 20 mL [center dot] kg-1 [center dot] d-1. We describe a patient who was treated with a small volume (2500 mL in 3 days) of 6% hetastarch (average molecular weight of 450 kd) for the management of cerebral vasospasm and who developed a significant coagulation disorder. The patient still had a concentration blood HES level two days after discontinuation of the HES solution. Case Report A 45-yr-old woman (60 kg) was admitted for severe and sudden headache occurring at rest and associated with vomiting. Clinical examination revealed neck rigidity. Computed tomography of the brain showed a right frontal subarachnoid hemorrhage with a right hemispheric subdural hematoma. Cerebral angiography was performed, demonstrating an aneurysm of the right anterior communicating artery. Surgical therapy consisting of clipping the aneurysm was uneventful. On the fifth postoperative day (POD), intracranial Doppler studies revealed increased velocities compatible with cerebral vasospasm. Despite the lack of neurological deficit, hypervolemia was induced with crystalloids (dextrose 5% in NaCl 0.45%, 3000 mL/d) and hetastarch HES 6% (average molecular weight of 450 kd, lot 94J28, Fresenius, Bad Homburg, Germany). Before the infusion, prothrombin time and APTT were normal (Figure 1), and the platelet count was 284,000/mm3. The patient received a total dose of 2500 mL HES in 3 days (POD 7 to 10) (Figure 1). At the end of the HES infusion, prothromin time had decreased to 59%, and APTT was prolonged to 78.8 s. D-dimers were normal (<250 mg/mL), the fibrinogen level was normal (135 mg/dL), and an ethanol test was negative. Lupus anticoagulant and anticardiolipin antibodies were negative. Factor VIII was 18% (normal [N] = 50%-200%), Factor XI was 46% (N = 70%-200%), and Factor XII was 38% (N = 50%-200%). After removal of an arterial catheter on the tenth POD, the patient developed a femoral hematoma, which required prolonged (30-min) local compression. On the twelfth POD, the plasma level of HES, determined by a standard procedure after preparation of plasma with trichloracetic acid 50% (Fresenius) and conservation at -18 degrees C was 12.62 mg/mL. Coagulation variables improved after the discontinuation of hetastarch (Figure 1). The patient had normal renal function (creatinine levels 0.7-1.1 mg/dL). The patient died on the sixteenth POD after massive intracranial bleeding due to clip malposition on the anterior communicating artery.Figure 1: Scale of events after clinical subarachnoid bleeding showing vasospasm on day 5, administration of hydroxyethylstarch from day 7 to day 10, coagulation study from day 10 to day 14, and death occurring on day 15.Discussion Coagulation abnormalities have been described after the administration of HES solutions. Although some authors [5,6] suggest that bleeding disorders may be related to a dilutional effect of colloid administration, specific anticoagulant effects of HES have been ascribed to a decreased activity of factor VIII moieties, i.e., procoagulant factor VIII, von Willebrand factor, and factor VIII-related ristocetin cofactor [1,7,8]. The underlying mechanism is the precipitation of factor VIII by HES [1,7], an effect that may not be concentration dependent [1]. HES may decrease platelet adhesion both by reducing plasma levels of von Willebrand factor and by coating platelet membranes [1,9]. HES may also facilitate thrombolysis [1]. The effects on coagulation vary according to the molecular weight of the HES solution. These effects are usually more significant with hetastarch (Mw = 450 kd) than with pentastarch (Mw = 200 kd) solutions [5,9-12], although they can occur after the administration of pentastarch solutions as well [13,14]. Coagulopathies induced by hetastarch solutions are dose dependent and are usually described with doses exceeding 20 mL [center dot] kg-1 [centered dot] d-1. Claes et al. [15] studied the effects of hetastarch on bleeding in patients undergoing isovolemic replacement of 1 L blood during either abdominal hysterectomy or neurosurgery for brain tumors. They observed no difference in bleeding between patients receiving either 6% hetastarch or 5% human albumin. Shatney et al. [16] described the administration of large volumes of hetastarch solution without increased morbidity and mortality in a series of patients with traumatic shock. Four patients were given more than 5,000 mL, and one received 15,000 mL hetastarch solution in a total of 28,000 mL intravenous fluids. However, other investigators describe increased bleeding after hetastarch administration. In a recent series of 85 patients with cerebral vasospasm, Trumble et al. [17] demonstrated that the infusion of a hetastarch solution instead of plasma protein fraction increased the incidence of clinical bleeding and significantly prolonged APTT from 23.9 to 33.1 s (P < 0.001). Other investigators described coagulation abnormalities during the infusion of smaller doses. For instance, Kuitinen et al. [18] observed increased blood loss after coronary artery bypass graft when hetastarch solution was included in the cardiopulmonary bypass priming solution. San Fellippo and Suberviola [19] reported a patient with cerebral vasospasm who developed gingival bleeding after 10 days of hetastarch infusion at a rate of 1000 mL/d. Damon et al. [4] related a serious coagulopathy with intracerebral bleeding and herniation after the infusion of 1750 mL hetastarch over 24 hours. In view of these dose-related effects on coagulation, a maximum dose of 20 mL [center dot] kg-1 [center dot] d-1 of a standard 6% hetastarch solution is usually recommended [20]. Our patient had no initial coagulation abnormality and developed significant coagulation abnormalities after having received only 2500 mL hetastarch in three days. HES is cleared by the kidneys [21], but the patient had no renal failure. Nevertheless, an HES blood level of 12.6 mg/mL two days after the cessation of the HES administration indicates the persistance of significant amounts in the circulation. In volunteers receiving a single administration of 500 mL HES, blood levels reach a maximum of 5.04 mg/mL after two hours and a value of 2.68 mg/mL two days later (data on file, Fresenius). The patient we described is the first example of coagulopathy induced by such a small dose of hetastarch solution. We propose extreme caution visa-vis the repeated administration of hetastarch solutions in neurosurgical patients. We are grateful to G. Richter, Head of Drug Safety Fresenius, for the determination of the blood concentration of HES.