Neurologic Complications of Neuraxial Analgesia/Anesthesia: Epidural hematoma

General Information

Epidural hematoma is a rare but potentially catastrophic condition that may be associated with a neuraxial procedure.

1. Incidence

The true risk is difficult to determine for multiple reasons, including underreporting and unknown denominators. A recent meta-analysis of 4 studies post-1990 which involved more than 1 million obstetric patients found an incidence of 1 in 168,000, but with only 6 epidural hematomas the accuracy of this estimate cannot be relied upon. Other surveys have found no cases in more than 700,000 obstetric epidural procedures.

The reality is that it is a rare complication in the obstetric population, whereas in the nonobstetric population, the incidence may be as high as 1 in 3,600 for elderly women. The majority of cases are associated with coagulopathic disorders or anticoagulant therapy. The hypercoagulable state of the pregnant woman together with the rapid dispersal of blood out of the epidural space may protect against hematoma, despite engorgement and fragility of the epidural vessels in these patients.

It is more common following epidural than spinal procedures, with 30-50% developing after removal of the epidural catheter.

Epidural hematomas may occur spontaneously in the population as a whole, without epidural vessel trauma, without neuraxial anesthesia, and without coexisting coagulopathy or therapeutic anticoagulation.

2. Symptoms

Presenting symptoms are most commonly bilateral progressive lower extremity weakness and sensory deficit. Pain is often not the first symptom, but back and radicular pain can develop over time, together with bladder and bowel dysfunction. Finally, paraplegia may occur if undiagnosed and untreated.

3. Diagnosis

A full neurological examination to exclude obstetric neuropathies should be followed by emergent MRI if a hematoma is suspected. Laboratory studies to evaluate the patient for a coagulopathy should be done concurrently.

4. Risk Factors

Risk factors include:

1. Coagulopathy or therapeutic anticoagulation

2. Difficult or bloody epidural needle or catheter placement

3. Indwelling epidural catheter during continued anticoagulation

4. Spinal deformity with or without previous surgery

5. Spinal tumor or arteriovenous malformation

5. Treatment and Prognosis

Decompression laminectomy within 8 hours of the onset of symptoms gives the best chance for full neurological recovery and reversal of spinal cord ischemia caused by cord compression. The longer the time between onset of neurological deficit and surgery, the poorer is the prognosis.

6. Prevention

General precautions to minimize risk of epidural hematoma in patients with a low platelet count include:

1. If the platelet count is low, the most experienced provider should place the block to minimize the risk of multiple attempts at any procedure.

2. If appropriate, choose a spinal technique rather than an epidural technique to minimize the risk of vessel puncture.

3. Soft-tipped catheters may reduce the incidence of vessel trauma.

4. The use of a dilute concentration of local anesthetic for continuous epidural analgesia should not result in a dense motor block, aiding diagnosis in patients who develop symptoms of an epidural hematoma.

5. The platelet count and other coagulation studies, if indicated, should be checked before catheter removal to ensure platelet levels are at least 80,000 ×10(9)/L.

6. Regular neurologic assessment for excessive motor or sensory block should take place during and after delivery and after catheter removal to enable early recognition of any complications.

a. Preeclampsia and HELLP

The platelet count can be decreased in preeclampsia, and by definition will be decreased in HELLP syndrome. When considering a neuraxial procedure in a preeclamptic patient, look for trends in platelet count, i.e., are they low (100-150 ×10(9)/L) but stable over time, or is there a rapid downward trend? The platelet count should be checked every 6 hours in these patients if stable, but every 1-3 hours if showing a rapid downward trend.

No further coagulation studies are usually necessary if the platelet count is >100 ×10(9)/L, but they may be useful if the platelet count is lower. However, platelet function may be abnormal even if the count is adequate. Thromboelastography (TEG) can be used to evaluate overall coagulation function, but its ability to reduce risk of epidural hematoma has not been proven.

The lowest platelet count that an anesthesia provider will consider before placing a neuraxial block depends partly on the comfort level of the provider and partly on the risk/benefit ratio to the patient. Most providers will proceed with a platelet count of 80-100 ×10(9)/L if coagulation studies are normal and there are no clinical signs of bleeding, such as hematoma formation at IV or blood draw sites or new gum bleeding when brushing teeth.

Some providers will place a neuraxial block for cesarean section as long as the count is >50 ×10(9)/L with normal coagulation studies, preferentially choosing a spinal technique rather than an epidural to minimize risk of puncturing an epidural vessel. This may be worth the slightly increased risk in order to avoid general anesthesia, especially in a patient with an edematous airway secondary to preeclampsia.

b. Inherited Bleeding Disorders

Women with inherited bleeding disorders are often refused neuraxial analgesia or anesthesia because of the potential risk of epidural hematoma. Depending on the type and severity of the condition, a neuraxial block may be a viable option.

Benign Gestational Thrombocytopenia

The platelet count decreases by 20% in pregnant women (7% have counts <150 ×10(9)/L, 0.5-1% <100 ×10(9)/L), but platelet function in gestational thrombocytopenia is normal. Benign gestational thrombocytopenia is responsible for 75% of pregnancy-related thrombocytopenic cases. As long as other coagulation studies are normal, neuraxial procedures can be performed at the discretion of the anesthesia provider, who will decide what the lowest acceptable platelet count is. (See section on preeclampsia for more details.)

von Willebrand Disease

There are three main phenotypes. In Type 1, Factor VIII and vWF increase during pregnancy, potentially normalizing coagulation and still allowing neuraxial blockade. However, levels may vary and treatment with desmopressin (DDAVP) can help further increase factor VIII and vWF. Catheters should be removed as soon as possible after delivery, as the coagulation factors diminish rapidly at this time. Advance planning and consultation with a hematologist are important to facilitate any necessary treatment and improve the chance of providing appropriate regional anesthesia.

There are many other disorders that are beyond the scope of this chapter.

c. Anticoagulation and Antithrombotic Drugs

Anticoagulation is used prophylactically or therapeutically for several conditions during pregnancy, including inherited thrombophilias, venous thromboembolism, mechanical heart valves, and procoagulant conditions, which predispose to early pregnancy loss.

The American Society of Regional Anesthesia (ASRA) has developed anticoagulation guidelines to prevent epidural hematoma following neuraxial procedures in surgical patients. These are used to guide management in the obstetric population in the absence of targeted research in this population.

ASRA Guidelines with Modifications for Obstetric Patients