Paraplegia following epidural analgesia: A potentially avoidable cause?

Neurological deficit is an uncommon but catastrophic complication of epidural anesthesia. Epidural hematomas and abscesses are the most common causes of such neurological deficit. We report the case of a patient with renal cell carcinoma with lumbar vertebral metastasis who developed paraplegia after receiving thoracic epidural anesthesia for a nephrectomy. Subsequently, on histo-pathological examination of the laminectomy specimen, the patient was found to have previously undiagnosed thoracic vertebral metastases which led to a thoracic epidural hematoma. In addition, delayed reporting of symptoms of neurological deficit by the patient may have impacted his outcome. Careful pre-operative investigation, consideration to using alternative modalities of analgesia, detailed patient counseling and stringent monitoring of patients receiving central neuraxial blockade is essential to prevent such complications.

INTRODUCTION

Permanent neurological deficit is a rare but devastating complication following epidural anesthesia. The incidence of neurological complications following epidural anesthesia varies between 0.0007% and 0.1% across studies,[1234567] with the most common causes being epidural hematomas and abscesses.

We report the case of a patient of renal cell carcinoma with lumbar vertebral metastasis who developed paraplegia after thoracic epidural catheter insertion for nephrectomy.

CASE REPORT

A 47-year-old male patient presented to our hospital with a left-sided renal mass with a biopsy positive for malignancy. On evaluation, the bone scan carried out 1 month prior showed increased tracer uptake in the body of L2 vertebra suggestive of metastasis. The rest of the skeletal system was normal. The patient was recently diagnosed with hypertension, but was not on any medication. All other investigations (including platelet count and coagulation profile) were within normal limits. The patient was posted for a left-sided cytoreductive nephrectomy with retroperitoneal node dissection. The anesthetic plan was general anesthesia with perioperative epidural analgesia.

Before induction of general anesthesia, epidural catheter placement was attempted in the D8-D9 interspace. This led to a dural tap; the catheter was subsequently inserted one interspace above at the D7-D8 level. Following this, general anesthesia was administered.

The surgery was uneventful with blood loss of 1L and the patient was extubated postoperatively. An epidural infusion of 0.125% bupivacaine was given intra- and postoperatively. The patient was observed overnight in the recovery room and was shifted to the ward the next day. Pain relief was adequate and an epidural infusion of 0.05% bupivacaine was started through an elastomeric pump. Sub-cutaneous once-daily low molecular weight heparin (LMWH) -enoxaparin (60 mg in 0.6 ml) - was started from post-operative day 1 as pharmacological prophylaxis for deep vein thrombosis (DVT). The patient was ambulatory with good pain relief and the epidural infusion was continued for 6 days. The epidural catheter was removed on the 7th post-operative day on the morning rounds (18 hours after the last dose of LMWH) and oral analgesics were started.

On the night of the 8th post-operative day, the patient complained of urinary retention, for which he was catheterized. The next morning he could not move both his legs (power of 0/5-1/5 in both lower limbs) and had a complete loss of bladder sensation. At this stage, on direct questioning, he admitted that he had felt some weakness in his left leg the previous evening, which he had neglected, attributing it to fatigue. An urgent magnetic resonance imaging (MRI) of the spine showed cord compression at D7-D8 levels. Injection dexamethasone was started and the patient was taken up for an emergency decompressive laminectomy for evacuation of an epidural hematoma causing compressive myelopathy and paraplegia. D8 laminectomy was performed and the material was sent for histopathology. However, there was no improvement in motor function over the next 24 hours and a repeat MRI performed the next day showed some re-accumulated blood causing cord compression two spaces above the laminectomy segment. A repeat multi-segment laminectomy was carried out and the material was sent for histopathologic examination. The patient was given intravenous methylprednisolone and mechanical prophylaxis for DVT with sequential compression devices. Power continued to be grade 0/5-1/5 with no improvement in paraplegia. Histopathology report of the epidural decompression material revealed metastatic clear cell renal cell carcinoma with morphology similar to the primary renal tumor. A repeat MRI done a week later revealed post-laminectomy changes with resolving cord edema at D6-D9 regions and a new lesion in L2-L3 region compressing rootlets. Palliative high dose radiotherapy was given from L1 to L4 levels. The patient was discharged from the hospital after 3 weeks with no improvement in neurological status.

DISCUSSION

Various risk factors for the development of epidural hematoma after epidural anesthesia have been described: advanced age, female gender, bony spinal pathology, use of anticoagulants or antiplatelets, cancer, multiple attempts at epidural catheter insertion, difficult epidural, indwelling catheters, emergency surgeries, coagulopathy, spinal cord injury, epidural angiomas.[678] Of these, our patient had renal cell carcinoma with known lumbar vertebral metastasis, had two attempts at epidural insertion with a dural puncture on the first attempt and was on anti-coagulants after surgery. The pre-operative diagnosis of neuraxial bone metastases may be difficult because patients with small epidural lesions may be asymptomatic. Furthermore, it has been suggested that bone scan may not be a sensitive tool to identify lytic bone metastases especially in malignancies like renal cell carcinoma.[9] It may be difficult to perform the bone scan immediately before surgery and fresh metastases may appear in the intervening time. There are a few case reports of paraplegia or neurological deficits after central neuraxial blockade in patients with previously undiagnosed neuraxial bone metastases. Cherng et al. described post-operative paraplegia as a rare complication following spinal anesthesia in a patient with a previously unrecognized spinal tumor.[10] Kararmaz et al. reported a case of post-operative paraplegia in a previously undiagnosed case of vertebral metastases originating from an endometrial cancer following administration of combined spinal-epidural anesthesia.[11] Graham et al. described a case of paraplegia following spinal anesthesia for bilateral orchidectomy in a case of prostatic carcinoma with spinal metastasis with no pre-existing neurological deficit.[12] In a case report by de Mιdicis, lumbar epidural analgesia and thoracic vertebral metastasis was associated with reversible paraplegia.-[13] It is hypothesized that compliance of the epidural space is restricted because of the metastasis and that the administration of local anesthetic causes an increase in both the volume and the pressure in the epidural space, leading to symptoms of compressive myelopathy.[13]

In this patient, while we were aware of the presence of spinal metastases at the level of L2, it was thought that epidural anesthesia with epidural puncture and catheter placement at a higher level might be safe. However, in addition to the known bony metastasis at L2 level, there were undiagnosed metastases in the thoracic epidural space which probably bled after removal of the epidural catheter and caused compressive myelopathy and subsequent irreversible paraplegia.

The American Society of Regional Anesthesia and Pain Medicine has laid down guidelines for management of neuraxial catheters in patients receiving anticoagulation.[8] These guidelines were strictly followed in our patient. However, despite counseling and information about potential problems of epidural anesthesia at the pre-operative visits, our patient failed to report early neurological signs. Timely recognition and decompression of epidural hematomas (ideally within 8 hours of developing symptoms) has been shown to improve chances of neurological recovery.[14]

This case serves to emphasize the fact that both patient education and careful monitoring are essential in the management of patients receiving neuraxial anesthesia. In patients with tumors which have a high propensity to spread to bone, the index of suspicion for unknown spinal metastases in addition to previously diagnosed bony metastases should be high. Epidural analgesia should be avoided and other modalities of post-operative analgesia should be considered.