ANTIPLATELET drugs are widely used in various categories of patients 1; therefore, knowledge of their impact on coagulation is important for the patients’ perioperative care. The American Society of Regional Anesthesia and Pain Medicine does not consider antiplatelet drugs, by themselves, as risk factors for the development of spinal hematoma in patients having neuraxial blocks, but concurrent use of other medications that affect clotting mechanisms may increase the risk of bleeding complications. 2,3We report a case of an epidural hematoma in a patient to whom low-molecular-weight heparin (LMWH) was administered perioperatively as prophylaxis against development of venous thromboembolism and who additionally had taken oral ibuprofen for pain relief on her own.
A 63-yr-old woman (weight, 75 kg; height, 165 cm) with American Society of Anesthesiologists physical status III presented for reimplantation of a prosthesis of the right knee. One year previously, the prosthetic joint had been removed because of an infection. At this time, the patient also had spondylodiscitis of the lumbar vertebrae 4 and 5 and of the thoracic vertebra 10 and a paravertebral abscess. Clinically and radiologically, the patient had recovered from this complication. The patient’s other significant medical history included an ischemic heart disease of grade II (classification of angina of the Canadian Cardiovascular Society), allergies to several antibiotics, and arthritic pain of both shoulders. Her preoperative oral medications included 5 mg propanolol twice daily and 150 mg ranitidine once daily. Additionally, one subcutaneous injection of LMWH (nadroparin, 3,800 IU/day) was administered to the patient as a prophylactic antithrombotic treatment. Preoperatively, the activated partial thromboplastin time (PTT) was 33 s (normal range, 30–40 s), the prothrombin time (PT) was 84% (normal range, 70–120%), and the International Normalized Ratio (INR) was 1.01 (therapeutic level, 2–4.5). The platelet count was 151 × 109/l, at the lower limit of the normal range (150 × 109/l–400 × 109/l).
On the day of surgery, the patient did not receive the subcutaneous injection of LMWH, the last dose of which had been administered 12 h previously. She was taken to the operating room, and an indwelling epidural catheter via an 18-gauge Tuohy-type needle (Medimex Holfeld, Hamburg, Germany) was placed at the L3–L4 interspace using the “loss-of-resistance” technique. Epidural puncture and insertion of the catheter were uneventful, and no blood or spinal fluid were observed. A test dose of 4 ml bupivacaine, 0.5%, was injected. The block was then extended up to the T6 dermatome using 0.5% bupivacaine. General anesthesia was induced, the trachea was intubated, and surgery was performed with a combined regional–general anesthesia. At the end of surgery, the tracheal tube was removed. The patient was awake and free of pain. The sensory level of anesthesia reached T9, and motor block was Bromage scale 2. After reversal of the motor block, a continuous epidural infusion of 0.25% bupivacaine (8 ml/h) was initiated. Thromboprophylactic treatment with LMWH was reestablished 6 h after the end of surgery. Eighteen hours later, the patient reported pain of increasing intensity at the site of the surgical incisions and back pain. The epidural infusion rate was increased to 12 ml/h, and two boluses of 12 ml bupivacaine, 0.25%, were administered, without yielding a sufficient level of analgesia. Additionally, she received 400 mg ibuprofen rectally for the back pain, which decreased in intensity afterward. The epidural bolus administered after this treatment was painful during injection. Therefore, the epidural infusion was stopped and the catheter was removed approximately 7 h after the second postoperative LMWH administration. At this time, the coagulation parameters of activated partial thromboplastin time, International Normalized Ratio, and prothrombin time were within normal ranges. The platelet count was 118 × 109/l. After catheter removal, the patient received patient-controlled intravenous analgesia using piritramid.
Ten hours after removal of the epidural catheter, the intensity of the back pain increased again, and voiding difficulties developed. Neurologic examination revealed decreased muscular strength in both lower limbs and decreased sensation up to the dermatome L1. A magnetic resonance image revealed a spinal–epidural hematoma extending from T10 to L4, with an intradural extension in the caudal section (figs. 1 and 2). Emergency decompressive surgery was performed 6 h after the first neurologic symptoms had been noticed. The intraoperative site showed a partially organized as well as a fresh hematoma. There was no evidence of neoplastic changes or vascular malformations. Postoperatively, the patient recovered slowly from her neurologic deficiencies. Control magnetic resonance imaging 5 days later showed a small, infarcted area within the spinal cord at the level of T10. In a detailed interview, during which the patient was asked about specific drugs taken preoperatively, she mentioned that she had taken 500 mg ibuprofen four times daily on her own for her arthritic pain and that she had taken the last dose approximately 10 h before surgery. Four weeks after the event, muscular strength and sensation of the right leg had recovered. A discrete muscular weakness of the left leg with an area of hypesthesia at the foot persisted for 1 yr.
There are four major classes of antiplatelet drugs: (1) cyclooxygenase inhibitors (e.g. , acetylsalicylic acid [ASA] and nonsteroidal antiinflammatory drugs [NSAIDs]), (2) selective adenosine diphosphate antagonists (e.g. , ticlopidine), (3) direct thrombin inhibitors (e.g. , hirudin), and (4) platelet glycoprotein IIb and IIIa inhibitors (e.g. , tirofiban). Most cases of spinal–epidural hematoma have been reported in patients who are receiving some kind of anticoagulation therapy or who have an otherwise compromised coagulation. 4–6Studies that investigated the use of central neuraxial block in the presence of an antiplatelet treatment with ASA or NSAIDs found no increased risk for spinal bleeding. 3,7Although case reports of spinal bleeding in the presence of antiplatelet drugs are sparse, 8–10the topic is still controversial. Cyclooxygenase inhibitors acetylate the platelet enzyme cyclooxygenase, which is essential for the biosynthesis of thromboxane A2. Thromboxane A2itself is a potent activator of platelet aggregation and adhesion. The acetylation caused by ASA is irreversible; therefore, even a low dose of ASA (75–325 mg daily) compromises platelet function until a sufficient number of new platelets have been released from the bone marrow. 11However, the acetylation caused by NSAIDs is reversible and lasts for 1–3 days. Cronberg et al. 12demonstrated that the different NSAIDs impair platelet function similarly but that the duration of impairment varies. Ibuprofen (10–12 mg/kg administrated once) had a weak but definite effect on the second wave of aggregation, lasting up to 1 day. Preoperatively, the current patient took a much higher dose (27 mg · kg−1· 24 h−1), and the platelet count at the time of catheter placement was near the lower limit of the normal range. Therefore, the effects of ibuprofen on platelet function were probably stronger and lasted longer. Also, the presumed single dose of ibuprofen administered later for back pain eventually contributed to an impaired thrombocytes function. In addition, the platelet count decreased after surgery, and the patient received two subcutaneous injections of LMWH before catheter removal. Because the subcutaneous blood flow varies interindividually and intraindividually, the patient response to subcutaneous heparin may be difficult to predict; therefore, an intermittently therapeutic anticoagulation cannot be excluded. Altogether, a clotting disturbance likely occurred, which was not detectable by conventional coagulation tests. This also explains the unusually large extension of the hematoma, 5which reached caudally beyond the puncture site. Whether small malformations of epidural vessels after paravertebral abscess formation and spondylodiscitis might have been a complicating factor remains speculative. The intraoperative finding of a partially organized as well as a fresh epidural hematoma suggests that bleeding was prolonged and occurred, or at least continued, throughout the postoperative period. It also suggests that not only the placement of the epidural catheter but also its removal may have contributed to the development of the hematoma. Back pain as an early symptom can be absent in the presence of continuous epidural analgesia. 5In the current patient, insufficient analgesia was probably the first clinical sign of the hematoma, which blocked the spread of the local anesthetic to the spinal nerve roots. This emphasizes that patients must be monitored closely when analgesic requirements increase, especially because a significant amount of epidurally administered local anesthetic might further compromise local coagulation. 13
During the past decade, several case reports and reviews discussed the risk factors for development of a spinal or epidural hematoma after central neuraxial blocks. 4,5,14Although the technique is considered to be safe when the clotting function is normal, its use in the presence of coagulation disorders, either caused by a preexisting coagulopathy or by a drug-induced impairment of the coagulation, may be dangerous. The German Society of Anaesthesiology and Intensive Care Medicine, contrary to the American Society of Regional Anesthesia and Pain Medicine guidelines, demands a free interval of 1–2 days after the last administration of NSAIDs and at least a 3-day interval without ASA-containing medications before central neuraxial blocks should be performed or epidural catheters should be removed. 15,16Our case supports this recommendation and also documents how difficult obtaining a thorough list of all medications used by the patient preoperatively can be after recommendation.
Early diagnosis of a hematoma is crucial for the prognosis, and the current patient eventually recovered without sequelae from the neurologic deficiencies. This case emphasizes that patients should be asked specifically during the preanesthetic visit for additional drugs, not routinely taken drugs, especially because the number of antiplatelet drugs is increasing steadily. 9Also, a routinely performed postanesthetic follow-up after removal of an epidural catheter is mandatory to minimize the risk for detrimental sequelae.