Disseminated intravascular coagulation and status epilepticus

Disseminated intravascular coagulation (DIC) results from excessive activation of hemostatic mechanisms and can lead to extensive bleeding, widespread microthrombus formation, shock, and death. Acute DIC is most commonly caused by overwhelming infections, massive trauma, and obstetric complications, although many clinical entities have been associated.¹ Status epilepticus has been rarely reported as a cause of acute DIC.^2,3 We describe a woman with fatal autopsy-confirmed DIC associated with status epilepticus; this case directs attention to careful management of hyperpyrexia.

Case report. A 41-year-old woman with a 15-year history of seizures subsequent to head trauma was diagnosed at an outside hospital with prolonged, generalized seizures. She had two prior episodes of status epilepticus 1 and 3 years previously and was currently taking phenytoin and gabapentin. She was healthy until the afternoon of the day of admission when she experienced a prolonged, witnessed generalized seizure. She initially received two doses of 10 mg lorazepam intravenously without resolution of the seizure and was treated with acetaminophen and a cooling blanket for hyperthermia with a maximum temperature of 42.5 °C. She became apneic and was intubated. Pentothal (500 mg) was administered, and her motor activity ceased after a total of 75 minutes. A phenytoin level was 0.5 mg/dL (normal, 10-20 mg/dL).

She was transferred to our hospital approximately 4.5 hours after the onset of her seizure. She was nonresponsive, intubated, and clinically in shock on arrival. Her blood pressure was 85/50, heart rate 174, and rectal temperature 40.3 °C. A large hematoma was present in the right groin near a venipuncture site, and blood was oozing from multiple venipuncture sites. She was comatose, her pupils were fixed at 3 mm bilaterally, and oculocephalic, caloric, and corneal reflexes were absent. Her muscle tone was flaccid and she had no spontaneous muscle activity or respirations. The white blood cell count was 44,800/µL, hematocrit was 44%, and platelet count had decreased to 108,000/µL from 388,000 at the outside hospital. The prothrombin time (PT) was greater than 100 seconds, and the activated partial thromboplastin time (aPTT) was more than 150 seconds. The serum creatine kinase was elevated at 3396 IU/mL. Blood smear examination revealed schistocytes. Additional hematologic laboratory data are shown in the table. A chest radiograph was normal, and urine toxicologic screen results were negative.

Treatment included fluid resuscitation, hemodynamic pressor agents, and broad-spectrum antibiotics. She initially responded with improvement in her blood pressure but then began to bleed from multiple mucosal surfaces including her nose, eyes, nasogastric tube, and rectum. At this time repeat laboratory data demonstrated decreased fibrinogen, increased fibrinogen degradation products, and a dramatic decrease in her hematocrit to 15% and platelet count to 50,000/µL. Hematologic parameters did not improve despite multiple transfusions of red blood cells, platelets, cryoprecipitate, and fresh frozen plasma. Her bleeding and hypotension persisted, and she developed asystole and died 17 hours after arrival.

At autopsy there were widespread submucosal hematomas, a retroperitoneal hematoma, and multiple focal areas of hemorrhage of the small and large bowel. The liver also showed areas of hemorrhage and ischemia. The brain had an old scar in the right temporal lobe consistent with previous trauma, but no intraparenchymal or meningeal hemorrhages. Microscopic examination demonstrated bilateral hippocampal sclerosis consistent with long-standing epilepsy but no evidence of ischemic changes in the CA1 sector of the hippocampi or central nervous system or other infection. Blood culture results were negative after 7 days.

Further coagulation studies were performed on frozen plasma specimens (see table) and revealed evidence of a consumptive coagulopathy including prolonged PT and aPTT and decreased fibrinogen and factors VII and VIII. The concentration of soluble fibrin was markedly elevated, indicating systemic thrombin activity. Increased fibrinolytic activity was reflected by elevated concentrations of tissue plasminogen activator, fibrin(ogen) degradation products, and D-dimer.

Discussion. Our patient had a consumptive coagulopathy in the setting of status epilepticus with prolonged motor activity and hyperpyrexia. On presentation to our institution 4.5 hours after the initial episode she was in shock and coagulation studies showed evidence of systemic thrombin activity, fibrinolysis, and depletion of coagulation factors and platelets with bleeding from multiple sites. In addition, high creatinine kinase levels support the additional diagnosis of rhabdomyolysis. The autopsy findings were consistent with consumptive coagulopathy, demonstrating numerous submucosal hemorrhages and endothelial damage. There was no evidence of infection, trauma, brain damage, or toxic ingestion as alternative explanations for the consumptive coagulopathy.

The association of fever and seizures is well established. Postictal fever is common and is most frequently caused by production of excessive heat by sustained muscle activity, but underlying infection must also be considered. In a retrospective study of patients with status epilepticus and postictal fever, 38 of 90 patients had a temperature greater than 37.8°C.⁴ Another study found that 40 of 93 patients had a temperature higher than 37.8 °C and two higher than 39.5 °C after generalized seizure.⁵ In addition, electrical seizure activity by itself can lead to hyperpyrexia, as complex partial seizure activity has been reported to be a cause of fever, possibly because of seizure-related dysfunction of the hypothalamic thermoregulatory centers.⁶

Hyperpyrexia has been associated with DIC most often in heat stroke and malignant hyperthermia.^7,8 The bleeding diathesis in heat stroke was originally thought to be caused by primary fibrinolysis, but more recent studies suggest that fibrinolysis is a secondary phenomenon and that the primary event is a consumptive coagulopathy.⁹ Autopsy studies of heat stroke victims have shown widespread endothelial damage, which can be the initiating event in DIC.¹⁰ The available reports, therefore, suggest that the consumptive coagulopathy associated with heat stroke includes elements of excessive thrombin generation and heightened fibrinolysis with evidence of primary endothelial damage, similar to the findings in the patient presented here.

Our case represents the third report in the English literature of the association between status epilepticus and DIC. Fisher et al² reported a single nonfatal case of status epilepticus in a patient with rhabdomyolysis, renal failure, and hyperthermia and suggested that DIC was caused by muscle damage, hyperpyrexia, or renal failure. Another brief report presented a small series of 31 patients with status epilepticus and DIC collected from published and unpublished sources. However, multiorgan failure including renal involvement made it difficult to determine the cause of DIC in these patients.³ We propose that consumptive coagulopathy associated with status epilepticus is caused by hyperpyrexia resulting in endothelial damage, which leads to activation of coagulation and fibrinolysis. Rhabdomyolysis is likely to be secondary to the hyperpyrexia and widespread endothelial damage. Autopsy findings in our patient failed to identify other explanation such as infection, malignancy, or trauma for the coagulopathy. This association has important implications in the management of seizures because temperature should be monitored closely in patients with status epilepticus and efforts directed toward prompt lowering of body temperature to prevent the development of consumption coagulopathy. These include administration of antipyretics and the use of cooling blankets and ice packs to treat severe hyperthermia.