PERIOPERATIVE epidural analgesia has been reported to improve patient outcome after thoracic surgery. Unfortunately, this procedure is not devoid of inherent potential risk. Specifically, numerous case reports describing neurologic deficits resulting from direct needle trauma or epidural mass effect (e.g. , hematoma or abscess) have been published.1–3We report a unique case of transient, yet profound, neurologic deficit that resulted from an unusual epidural mass.A 91 yr-old, 152-cm, 52-kg female was scheduled to undergo elective thoracotomy for repair of a symptomatic type III paraesophageal hiatal hernia using the Belsey Mark intravenous procedure. Her past medical history was remarkable for well-controlled hypertension, primary cancer of the breast, colon, and bladder (all treated without recurrence), chronic anemia, glaucoma, and profound kyphosis (Cobb angle 72°) secondary to osteoporosis. Medications included ranitidine, raloxifene hydrochloride, verapamil, vitamin B12, iron supplements, latanoprost, and timolol maleate.Two hours before surgery she received 5000 U unfractionated heparin subcutaneously for deep venous thrombosis prophylaxis. On arrival in the operating room, sterile placement of a T6–7 epidural catheter was attempted unsuccessfully. The patient was repositioned, and catheter placement attempted at the T7–8 interspace resulted in dural puncture. A third attempt at the T8–9 level proved successful, as there was a discrete identification of the epidural space (i.e. , loss-of-resistance) and the catheter was advanced 4 cm without difficulty. Catheter aspiration was negative for blood, and no discernible neurologic or hemodynamic changes resulted from a 3-ml test dose of lidocaine 1.5% with 1:200,000 epinephrine. Subsequently, the patient was placed supine, general anesthesia was induced, and the epidural catheter was loaded with 4 ml bupivacaine 0.25% plus 0.4 mg hydromorphone. Shortly thereafter, an epidural infusion of 0.075% bupivacaine with hydromorphone 5 μg/ml was initiated (and maintained postoperatively) at 6 ml/h.Postoperatively, she reported good analgesia from the time of arrival in the postanesthesia care unit through the morning hours of the second postoperative day. During the afternoon of postoperative day 2, she developed profound bilateral lower extremity weakness. She denied back pain and was noted to be afebrile. Physical examination confirmed complete bilateral lower extremity paralysis with areflexia and absence of bilateral great toe proprioception. Light touch and temperature sensations were also diminished. Of note, she had received a total volume of approximately 220 ml of epidural infusate since the time of catheter placement. After confirming her coagulation status was normal, the epidural catheter was promptly removed, and emergent neurosurgical consultation and magnetic resonance imaging were obtained. Magnetic resonance imaging revealed an epidural mass in the dorsal left lateral aspect of the spinal canal extending from T6 to T9 with anterolateral spinal cord displacement and compression (fig. 1). Interestingly, the mass was initially thought to be an epidural hematoma. However, upon closer scrutiny of the magnetic resonance scan with our neuroradiologist (Dr. Rydberg), these images revealed features that were more characteristic of a nonheme-containing fluid. Specifically, the T2-weighted image intensity of the mass was consistent with a high water-content containing fluid (e.g. , cerebrospinal fluid, local anesthetic, or opioid) with only a scant amount of blood (fig. 1).The duration from discontinuation of the epidural infusion and completion of magnetic resonance imaging was approximately 90 min. During this time, the patient gradually regained lower extremity motor function. Within 2 h of discontinuing the epidural infusion, the neurologic deficits completely resolved, thereby circumventing the need for emergent surgical decompression. The remainder of her hospital stay was uneventful and she was discharged to an acute rehabilitation facility on postoperative day 7 with no adverse sequelae.We report a unique case of transient profound neurologic deficit in an elderly patient. To our knowledge, this is the first reported case of analgesic infusate and epidural cerebrospinal fluid causing clinically significant spinal cord compression.Complications of epidural anesthesia include but are not limited to back pain, postdural puncture headache, infection, intravascular injection, inadvertent subdural or intrathecal injection resulting in high neuraxial blockade, arachnoiditis, epidural hematoma, anterior spinal artery thrombosis, and transient or persistent neurologic injury.1–3With regard to neurologic injury, spinal cord damage may result from direct needle trauma, epidural hematoma formation, epidural abscess, arachnoiditis, or compromised vascular supply (e.g. , injury or spasm of the spinal arteries).3We believe the etiology of our observation was multifactorial. Specifically, dural puncture likely caused extravasation of cerebrospinal fluid into the epidural space. More importantly, the epidural infusate continuously added further volume to this expanding mass, which was particularly problematic because of the advanced age of the patient and severe kyphosis.In the setting of normal neuraxial anatomy, epidural fluid is readily redistributed longitudinally along the craniocaudal axis, laterally towards the paravertebral region, or anteriorly.4In elderly patients or patients with abnormal anatomy (i.e. , severe kyphosis), dispersion of the fluid longitudinally and laterally may be limited, thereby resulting in anterior dispersion (i.e. , the pathway offering minimal tissue resistance).4In our patient, this resulted in mass effect and spinal cord compression. Motor deficit without sensory deficit likely resulted from disruption of anterior horn cells as the spinal cord was compressed against the thoracic vertebral bodies.When faced with new onset neurologic deficits in patients receiving epidural analgesia, a neuraxial mass must be expeditiously excluded from the differential diagnosis. In addition to making a prompt diagnosis, it has been suggested that surgical decompression should be performed within 8 h of neurologic deficit onset for complete recovery to occur.5In summary, we report a unique case of transient profound neurologic deficit resulting from accumulation of epidural analgesic infusate and epidural cerebral spinal fluid. In our case, we observed early recognition of the epidural mass and discontinuation of the epidural infusion circumvented surgical intervention yet resulted in complete resolution of the profound deficit.