Secondary Logo

Journal Logo


Transient Paraplegia After Epidural Anesthesia in a Parturient

Mack, Patricia Fogarty, MD; Gurvitch, Dana L., MD; Gadalla, Farida, MB, ChB

Author Information
doi: 10.1097/00000539-200001000-00025
  • Free

Both spinal and epidural anesthesias are often used for MacDonald cerclage for incompetent cervix in order to avoid IV medication in the pregnant patient. Although neurologic injury from spinal and epidural anesthesias are extremely rare, if a patient develops a persistent or progressing sensory and/or motor deficit after neuraxial blockade, a spinal or epidural hematoma must be considered. Along with neurologic examination, computed tomography (CT), or magnetic resonance imaging (MRI) of the lumbar spine may be warranted to evaluate persistent deficits. The implications of radiation exposure to the fetus during lumbar spine CT scan make this procedure undesirable, but not every institution that performs cerclage procedures has an MRI scanner immediately available. We report a case of the onset of spinal block in the postanesthesia care unit after an epidural anesthetic performed for cerclage placement.

Case Report

A 178-cm, 70-kg, 38-yr-old woman (gravida 2, para 0) presented for MacDonald cerclage for incompetent cervix at 15 wk gestation. We placed an epidural catheter (Portex nylon epidural catheter [SIMS Portex Inc., Keene, NH], 20-gauge, closed end, three eyes), while the patient was in the sitting position, at the L3-4 interspace using loss of resistance technique. No cerebrospinal fluid (CSF) or blood was aspirated at any point during catheter placement. The result of a test dose of 3 mL of lidocaine 2% with epinephrine 1:200,000 was negative. A total of 18 mL of lidocaine 2% with epinephrine 1:200,000 was administered to produce a T10 level; however, the patient required fentanyl 250 μg IV for discomfort despite the bilateral T10 level. The surgical procedure concluded 1 h after epidural placement, and the catheter was removed in the operating room.

On arrival to the postanesthesia care unit, the patient had a T10 sensory level and was able to move both feet. Thirty minutes after arrival, the patient complained of “getting number.” Physical examination revealed a pulse of 103 bpm and a blood pressure of 88/60 mm Hg. Neurologic examination revealed a T4 sensory level on the right and T5 level on the left to pinprick; lower extremity strength was 0/5 bilaterally. A neurosurgical consultation was obtained. Fifty minutes after the onset of T4 sensory level, the patient reported “tingling” in her right leg and “feeling stronger.” At that time, sensory level was T8, and motor examination was unchanged. Two hours and 30 min after the onset of paralysis (4 h and 10 min after epidural placement), full sensation and strength had returned. The patient was discharged home with no sequelae and delivered a healthy infant at term with uneventful epidural analgesia.


Onset of paraplegia after epidural blockade requires prompt evaluation and intervention if an epidural hematoma is diagnosed. A metaanalysis (performed before the widespread use of low molecular-weight heparin) determined the risk of hematoma after epidural anesthesia to be 0.0007% (1). In this patient who had no history of bleeding problems, use of nonsteroidal antiinflammatory drugs, or unusual bleeding during the procedure, epidural hematoma was unlikely, but a neurosurgical consultation was obtained in case the block persisted.

Previously, “reappearance” of spinal anesthesia has been attributed to sacral (2) or subdural (3) sequestration of local anesthetic later released into the CSF after patient movement. A similar etiology is likely in our patient. It is possible that the closed end of the catheter penetrated the arachnoid, while the holes remained in either the epidural or subdural space, allowing transport of the local anesthetic into the subarachnoid space when the catheter was removed. The presence of a puncture site in the arachnoid allowing local anesthetic to enter the CSF from the epidural space has been implicated in patients with unexpectedly high spinal level from subarachnoid block after failed epidural blockade (4).

A sequestration of lidocaine in the subdural space is suggested by two findings. First, the administration of 18 mL of local anesthetic produced only a T10 level, and the patient reported pain that required the administration of 250 μg of fentanyl IV. Second, one would expect the onset of a spinal block caused by isobaric lidocaine traversing a dural puncture to occur more quickly than occurred in our patient (T4 spinal block more than 30 minutes after catheter removal) if the lidocaine were freely dispersed in the epidural space. Because there is no objective evidence confirming catheter location in this patient, markedly delayed onset of profound epidural block should also be considered as an etiology.

The obstetrician felt that the findings were purely psychological, secondary to the patient’s anxious nature. Laraki et al. (5) reported a case of hysterical paraplegia after epidural anesthesia; however, this is a diagnosis to be considered only after eliminating all others, and the difference in the sensory level between right and left is not consistent with a psychological etiology.

Diagnostic evaluations to eliminate epidural hematoma include lumbar MRI, CT myelogram, and lumbar CT. MRI is the most accurate imaging technique to identify blood in the epidural space. In addition, no detrimental effect of MRI scanning on the developing fetus has been reported (6). Although one study demonstrated a greater incidence of miscarriage in MRI technicians than in other hospital workers or homemakers, the MRI technician cohort was significantly older than the others (7). As fetal exposure to ionizing radiation has been linked to mental retardation, small head size, intrauterine growth retardation, and childhood cancer, MRI would be preferable in the pregnant patient.

Lumbar CT myelogram is more accurate at demonstrating cord compression secondary to hematoma than lumbar CT; however, the total radiation exposure is greater with the addition of a myelogram. With respect to lumbar CT imaging, the potential risk of radiation exposure to the fetus is dependent on conceptual age. At less than two weeks postconceptual age, the main risk is of resorption of the conceptus. Between two and eight weeks postconception, the risk of small head size is 1% per centigray (cGy). From 8–15 weeks postconceptual age, rapid neuron development and migration occur, with the risk of severe mental retardation being 0.4% per cGY. The risk of childhood cancer is thought to be greatest before the 10th week postconceptual age. After 15 weeks postconception, there is a negligible risk of mental retardation below 50 cGy, a dose which would be highly unusual in diagnostic studies (8).

The radiation dose to which the fetus is actually exposed depends on patient size, scan technique, type of scanner, number of sections obtained, and location of the fetus within the pelvis (9). At 15 weeks gestation, the fetus would probably be partially included in a scan centered on L3-4. Fetal doses for an in-field scan, meaning that the fetus is completely in the beam of the study, are estimated at 1.8 and 4.8 cGy (1 cGy = 1 rad) for 10 one-centimeter sections (8). Doses would be lower if the fetus was not entirely within the field.

At 15 weeks gestation, our patient’s fetus was actually 13 weeks postconceptual age; in addition, the fetus would be at least partially in the field of the study. Thus, a lumbar MRI is the preferred diagnostic examination for both visual resolution of any hematoma and minimal risk to the developing fetus. If that is unavailable, then a decision must be made jointly among the neurosurgeon, obstetrician, radiologist, and patient as to whether a lumbar CT, lumbar CT myelogram, or surgical exploration without a radiologic study is the best course.

In conclusion, we present a case of onset of spinal blockade after the removal of an epidural catheter placed for anesthesia for cerclage. Although it is imperative to eliminate an epidural hematoma, which would require prompt intervention for restoration of neurologic function, this case demonstrates that other reversible etiologies must be considered.


1. Renck H. Neurological complications of central nerve blocks. Acta Anaesthesiol Scand 1995; 39:859–68.
2. Donielson DW, Palmen BD. A case of reappearance of spinal anesthesia [letter]. Anesth Analg 1994; 79:811.
3. Bromage PR. A case of reappearance of spinal anesthesia [letter]. Anesth Analg 1995; 81:658.
4. Mets B, Broccoli E, Brown AR. Is spinal anesthesia after failed epidural anesthesia contraindicated for cesarean section? Anesth Analg 1993; 77:629–31.
5. Laraki M, Orliaguet GA, Flandin C, et al. Hysterical paraplegia as a cause of transient paraplegia after epidural anesthesia. Anesth Analg 1996; 83:876–7.
6. LaBan MM, Viola SV, Williams DA, Wang A-M. Magnetic resonance imaging of the lumbar herniated disc in pregnancy. Am J Phys Med Rehabil 1995; 74:59–61.
7. Evans JA, Savitz DA, Kanal E, et al. Infertility and pregnancy outcome in magnetic resonance imaging workers. J Occup Med 1993; 35:1191–5.
8. Wagner LK, Archer BR, Zeck OF. Conceptus dose from two state-of-the-art CT scanners. Radiology 1986; 159:787–92.
9. Wagner LK, Lester RG, Saldana LR. Exposure of the pregnant patient to diagnostic radiations: a guide to medical management. Philadelphia: JB Lippincott, 1985.
© 2000 International Anesthesia Research Society