Anticoagulants, Antiplatelet Therapy, and Neuraxis Blockade

Abstract

Summary Regional anesthesia can be safely performed in patients receiving anticoagulant or antiplatelet therapy as attested by studies reported by Rao and El-Etr,26 Horlocker et al,11 and Odoom and Sih.21 Therefore, based on the available information, we believe the following recommendations are appropriate: 1. It is generally agreed that in the absence of extraordinary indications, spinal and epidural anesthesia are contraindicated in patients within 24 hours of receiving thrombolytic therapy, having known coagulopathies, or having significant thrombocytopenia. 2. Although the data by Odoom and Sih21 are reassuring, central neuraxis blockade should probably be avoided in fully anticoagulated patients. However, patients who have received only one or two doses of an oral anticoagulant will not have an increased prothrombin time and may safely undergo regional anesthesia. A prothrombin time may be measured before needle placement if necessary. 3. Patients fully anticoagulated with a continuous heparin infusion should have the infusion discontinued 4 to 6 hours before needle or catheter placement. Subcutaneous low-dose heparin should also not be administered within 4 to 6 hours of a spinal or epidural anesthetic to allow for normalization of the activated partial thromboplastin time. Laboratory verification may be performed before initiation of regional blockade. 4. Epidural or spinal anesthesia followed by systemic heparinization is probably safe, provided adequate precautions are taken.18, 26 Heparinization should not be initiated for at least an hour after needle placement.18, 28 In addition, patients also receiving antiplatelet medications who will undergo subsequent heparinization appear to be at increased risk for spinal hematoma and need to be followed closely. Rao and El-Etra26 recommended careful monitoring of the activated clotting time during heparinization. If needle placement is traumatic or difficult, the decision to proceed with surgery should be reevaluated. Removal of an indwelling epidural catheter in a patient receiving intravenous or subcutaneous heparin should occur 4 to 6 hours after the last heparin dose, and anticoagulation should not be initiated for at least 1 hour after catheter removal. 5. Epidural and spinal anesthesia can be safely performed in a patient receiving antiplatelet therapy.11 Patients with a history of bleeding or bruising may be further evaluated with a preoperative bleeding time. 6. The use of small-gauge needles and the midline approach (thereby avoiding the lateral aspect of the epidural venous plexus) allows more atraumatic needle placement. Epidural catheters should not be inserted more than 3 to 4 cm into the epidural space in order to minimize trauma to the epidural venous structures. 7. Short-acting local anesthetics should be used in patients at increased risk in order to promptly evaluate their neurologic status postoperatively. Likewise, an epidural blockade should be allowed to regress sufficiently before initiating a continuous local anesthetic infusion for postoperative analgesia. A narcotic rather than local anesthetic infusion would allow continuous monitoring of neurologic function and may be a more prudent choice in these patients. It is also worth noting that epidural hematomas have also occurred spontaneously in patients on anticoagulant or antiplatelet medications in the absence of trauma or needle placement.15 The decision to perform spinal or epidural anesthesia on a patient receiving these medications should be made on an individual basis, weighing the risks of hemorrhagic complications and benefits of regional anesthesia for a specific patient. The patient's history should be reviewed to discover any underlying medical conditions with associated bleeding tendencies that would increase the risk. Because it is difficult to predict an individual patient's response to antiplatelet or anticoagulant therapy, laboratory verification may be helpful before initiation of spinal or epidural anesthesia (Table 1). It is most useful clinically to evaluate coagulation parameters that are specific to the clotting defect expected such as the prothrombin time in patients on oral anticoagulants or the activated thromboplastin time in heparinized patients. Although the role of tests such as the thromboelastogram, which measure overall coagulation function, has not been defined at this time, they may prove useful in the future. Needle and catheter placement should be accomplished as atraumatically as possible, and subsequent anticoagulation should be carefully monitored to minimize hemorrhagic complications. The patient should be monitored closely in the perioperative period for early signs of cord compression such as backache or neurologic dysfunction, and any clinical or radiologic evidence of cord compression should be followed by an immediate decompressive laminectomy because recovery is unlikely if surgery is delayed for more than 12 hours.8