Enoxaparin Associated with Psoas Hematoma and Lumbar Plexopathy after Lumbar Plexus Block 

A 67-yr-old woman presented for repair of an open right calcaneal fracture. The patient's history was significant for insulin-dependent diabetes mellitus, coronary artery disease, and hypertension. Five years before admission, the patient suffered a myocardial infarction complicated by pulmonary edema. At that time, she also underwent four vessel coronary artery bypass grafting. Two years before admission, the patient underwent a left below-knee amputation (BKA) for complications of her diabetes. Her current oral medications included aspirin, 325 mg, once a day, amilodipine, 5 mg, once a day, captopril, 25 mg, three times a day, furosemide, 40 mg, twice a day, digoxin, 0.25 mg, once a day, and humulin NPH and regular insulin. Physical examination was remarkable for decreased breath sounds at the lung bases, a well-healed left BKA stump, and an open right calcaneal fracture. No gross motor deficits were noted. A detailed sensory examination was not performed. Laboratory investigations included leukocyte count, 18.3 x 10³; platelets, 355 x 10³; prothrombin time, 13.6 s (international normalized ratio 1.2); partial thromboplastin time, 26.0 s; and serum creatinine level, 1.1 mg/dl. Over the next six days, the patient underwent three operative procedures for treatment of her calcaneal fracture.

On the night of admission, the patient was taken to the operating room for incision and drainage of her right calcaneal fracture. Anesthesia was achieved with a sciatic nerve block (Labat approach) using a nerve stimulator technique and 30 ml of 1.5% mepivacaine with 1:400,000 epinephrine. The saphenous nerve was blocked by injecting 5 ml of the same local anesthetic posterior to the sartorius muscle. Adequate sensory-motor block was achieved. The operative procedure and recovery were uneventful. The patient regained sensory-motor function the morning after the procedure. The surgical team initiated injections of subcutaneous enoxaparin, 30 mg, twice a day on postoperative day 1 for DVT prophylaxis and resumed all of the patients preoperative medications.

On hospital day three, the patient was again taken to the operating room for an open reduction and internal fixation of her calcaneal fracture. Anesthesia was again achieved using a regional technique. The sciatic nerve was blocked using a Raj approach with 15 ml of 0.5% bupivacaine with 1:200,000 epinephrine. [11] The lumbar plexus was anesthetized with a femoral approach, using an inguinal 3-in-1 paravascular block (according to the technique of Winnie) with 20 ml of the same local anesthetic solution. [11] Both blocks were carried out uneventfully using a nerve stimulator. Good sensory-motor anesthesia was achieved. The operative procedure and recovery were uneventful. The patient continued on subcutaneous injections of enoxaparin, 30 mg, twice a day. No tests of coagulation were performed.

Despite the previously described efforts, on hospital day six (enoxaparin day number 5), the patient was scheduled for a right BKA because of poor distal perfusion. The patient reported excellent pain management with her previous regional anesthetics and requested a similar technique. Nineteen and one-half hours after receiving the last dose of enoxaparin, the patient was prepared for surgery. The anesthesia team elected to perform a lumbar plexus block using the technique initially described by Chayen, despite the increased difficulty of this approach. [12] With the patient's neck, back, and hips flexed in the left lateral decubitus position, a point was identified 4 cm lateral from midline along the intercristal line. Using a nerve stimulator, a 21-gauge, 100-cm, blunt-tip insulated needle (Braun Medical, Bethlehem, PA) was advanced perpendicular to the back in all planes. Despite several attempts, the team was unable to elicit an appropriate motor response. During those attempts, continuous aspiration was negative for blood. The patient was returned to the supine position, and satisfactory anesthesia was achieved using a 3-in-1 plexus block and a Raj approach to the sciatic nerve (using 0.5% bupivacaine with 1:400,000 epinephrine, 20 and 25 ml, respectively). The operative procedure was uneventful. By the next day, the patient regained all sensory-motor function.

Because of her debilitated status and bilateral BKAs, the patient was transferred to the rehabilitation ward. She was continued on the same dose of enoxaparin. The patient received her first postoperative dose of enoxaparin 4.5 h after block placement. On hospital day seven, one day after her right BKA, the patient complained of right hip pain. Despite the persistence of this pain, the patient was able to participate in her rehabilitation regimen for the next 6 days. On hospital day 14 (8 days after her right BKA), the severity of her pain precluded her from participation in daily activities. On hospital day 15 (9 days after her right BKA and lumbar plexus block), the patient awoke and alerted nurses that she was unable to move her right leg. Physical examination revealed the inability to move the right quadriceps, adductors, and hip flexors. Because of the presence of a large dressing and the previous left BKA, a detailed sensory examination was unattainable. Muscle strength on the left was normal. A tentative diagnosis of retroperitoneal hematoma was made. A computed tomography scan of the abdomen confirmed the diagnosis. It revealed a large retroperitoneal hematoma involving the right psoas muscle, which displaced the kidney anteriorly (Figure 1). Laboratory investigations revealed a prothrombin time of 13.6 s (International normalized ratio, 1.2), a partial thromboplastin time of 26.0 s, and a platelet count of 431 x 10³. Enoxaparin administration was discontinued. After careful evaluation, it was decided to allow the hematoma to resorb and avoid surgical evacuation. The patient gradually regained partial motor function over the next 5 days. At a 4-month follow-up visit, the patient had no subjective or objective evidence of a sensory-motor deficit.

Epidural and spinal anesthesia are commonly used for lower extremity surgery. Although effective, these techniques offer little selectivity for the operative side and can be associated with hypotension, urinary retention, and the potential for epidural hematoma in anticoagulated patients. Lumbar plexus block combined with a sciatic nerve block is an alternative anesthetic choice, which is specific for the operative extremity and avoids these side effects. The lumbar plexus is formed from the ventral nerve roots of the second, third, and fourth lumbar nerves as they lie between the quadratus lumborum muscle and the psoas muscle. It then extends anteriorly into the body of the psoas muscle to form the lateral femoral cutaneous, femoral, and obturator nerves. Because the plexus lies within the psoas muscle and its fascia, it is possible to puncture blood vessels within this space, which can lead to hematoma formation and a compartment syndrome. Unfortunately, little is known about the incidence of bleeding after lumbar plexus block. There is only one other case report in the literature that describes a retroperitoneal hematoma after this block. [13] In that report, a loss of resistance technique performed at the level of the first and second lumbar vertebra was used on a nonanticoagulated patient, without a nerve stimulator. It resulted in a renal subcapsular hematoma. This occurred shortly after the block and appeared to be an extremely rare complication. In our case, the attempted lumbar plexus block was noted to be difficult with several attempts being required. Because the hematoma appears to have gradually expanded over 9 days, it suggests that the etiology was related to the nerve block and an associated coagulopathy. In our series of 286 lumbar plexus blocks over the past year (unpublished data), this complication has not otherwise been evident.

Traditionally, full anticoagulation using heparin, Coumadin, or thrombolytic agents is considered a contraindication to regional anesthesia. However, antiplatelet agents and subcutaneous administration of unfractionated low-dose heparin are thought to present little risk of epidural hematoma. [14,15] The incidence of epidural hematoma is difficult to quantify in a prospective study because of its infrequent occurrence. Wulf, [14] in a comprehensive case review, estimated the overall incidence at 1:190,000 epidural administrations. Compared with this low incidence is the fact that to date there have been at least six cases of epidural hematoma potentially linked to low molecular weight heparins. [5–10] Based on these reports, patients treated with enoxaparin may be at greater risk from bleeding complications after regional techniques than was previously believed.

Low molecular weight heparins, like enoxaparin, are different from unfractionated heparins in several ways. In addition to having a more predictable bioavailability and a longer biologic half-life, low molecular weight heparins also have a dose-dependent fibrinolytic activity and inhibit the binding of platelets to fibrinogen and endothelium. Increasing the dose can magnify these effects and produce a therapeutic, instead of a prophylactic, plasma level, which can increase the risk of bleeding. In addition, abnormal renal function, advanced age, and the effect of combining nonsteroidal antiinflammatory drugs with enoxaparin may potentiate its effects. [1] It is therefore recommended that only prophylactic doses be used.

The European dosage prescriptions of enoxaprin's manufacturer are 20 mg once a day in low-risk patients and 40 mg once a day in high-risk patients. In Europe, the administration of two doses of enoxaparin (30 or 40 mg each) per day is only advised in the treatment, but not in the prophylaxis, of DVT. In the United States, enoxaparin is only approved for DVT prophylaxis after total knee or hip replacement surgery. However, it is also being used, outside of FDA guidelines as in this case, for DVT prophylaxis, and the dose has yet to be standardized.

It has been shown that low molecular weight heparins can produce therapeutic plasma levels in patients 2–4 h after subcutaneous administration. Peak plasma levels occur 4 h after subcutaneous administration and reach 50% of their peak values 12 h after injection. Based of these findings, it has been suggested that a 10- to 12-h interval occur between the insertion of and removal of anesthetic needles and catheters and the previous low molecular weight heparin dose. [15] In this case, a 19.5-h interval had occurred between the nerve block and the last dose of enoxaparin. However, there was only a 4.5-h delay from insertion of the block and the first postoperative enoxaparin dose.

Enoxaparin used in standard clinical doses does not alter conventional prothrombin times or partial thromboplastin times. [16] Therefore, standard perioperative tests do not assess the extent of a patient's anticoagulation, which makes it difficult to decide which patients undergoing regional anesthesia may be at particularly high risk for hematoma formation with resulting neurologic injury.

To our knowledge, this is the first case report of a hematoma and nerve plexopathy associated with a peripheral nerve block in a patient treated with enoxaparin. This case further illustrates the potential danger of regional anesthesia in a patient treated with this drug. It especially demonstrates the danger of an expanding hematoma in a noncompressible site. It also reemphasizes the inability to assess anticoagulation in patients treated with low molecular weight heparins. A simple perioperative test to determine the extent of anticoagulation would be of value.