THE authors report a case of psoas and foraminal abscess as a complication of epidural anesthesia, with secondary foraminal encroachment of the nerve root due to scar.
A 67-yr-old woman was admitted to our hospital for anterior knee pain, which developed after a motorcycle accident. Two months after the injury, the patient's pain became persistent and unbearable. After obtaining informed consent, an epidural catheter was inserted under sterile conditions in the L3–L4 interspace, using a loss of resistance technique, without difficulty. Sterile gloves and syringes were used, and the tubing connection was swabbed with povidone-iodine. The catheter was advanced 3 cm into the epidural space and secured with a sterile transparent dressing. After an initial bolus dose of 0.25% bupivacaine hydrochloride, subsequent administration was maintained by continuous infusion. Disposable equipment and a micropore filter were used. The patient did not receive prophylactic antibiotics. A 70% reduction of pain was observed. The catheter site was examined daily for any sign of inflammation or contamination and was removed after 13 days.
Two days after removal of the catheter, the knee pain recurred. A second epidural catheter was inserted into the L2–L3 interspace without difficulty, using the same technique as before. There was no evidence of infection at the site before catheter insertion, and no pain or paresthesia occurred. The patient's pain partially improved; however, 2 days after the catheter insertion she reported that another low back pain of stronger intensity had appeared, radiating to the left thigh. Four days after the insertion she noted a sudden onset of high fever (39.0°C) and purulent material leaking from the new insertion site. At that time, physical examination showed tenderness of the left paravertebral muscle. The catheter was removed. Her low back and left thigh pain increased. She assumed a hip flexed position, but nuchal rigidity was absent. Cerebrospinal fluid cultures for bacteria were negative and her white blood count was 19.6 × 103after removal of the second catheter. Computed tomography of the lumbar spine was compatible with psoas abscess (fig. 1). Bacterial cultures from the catheter tip were positive for Staphylococcus aureus, therefore, intravenous antibiotics were used for 2 weeks and her temperature gradually returned to normal in the course of 8 days, not exceeding 37.0°C.
Two months after the epidural anesthesia she was afebrile, but refractory thigh pain remained. Magnetic resonance imaging (MRI) showed residual swelling of the left iliopsoas muscle and narrowing of the left L2–L3 foramen (fig. 2). An L2 nerve root block was performed with 2 ml bupivacaine hydrochloride, 0.25%, for left L2–L3 foraminal stenosis. The patient had 70% improvement in thigh pain, but the relief lasted only 3 days. Infiltration of the L1 or L3 nerve root produced no relief of leg pain. In the following 5 months, L2 nerve root block was repeated 5 times.
Seven months after the epidural anesthesia MRI showed complete resolution of the psoas abscess but foraminal narrowing remained (fig. 3). To decompress this foraminal stenosis, surgery under general anesthesia was performed. In a prone position, a posterior midline skin incision was made and after partial facetectomy of the left L2–L3 facet joint, the left L2–L3 foramen was explored. The left L2 dorsal root ganglion was surrounded by a scar without bony impingement. This scar tissue surrounding the dorsal root ganglion was completely removed. The postsurgical course was excellent. The patient reported that the thigh pain disappeared the day after the surgery.
One and a half years after the epidural anesthesia, the patient had a follow-up examination. Although she still had the same knee pain, which had continued to be present through the entire period of treatment and was under treatment in a pain clinic in another hospital, the thigh pain had never recurred.
Pyogenic psoas abscess is often associated with epidural abscess, 1which has been reported as a rare complication of epidural anesthesia. 2–5Until now, however, there has been no report of psoas abscess as a complication of epidural anesthesia.
The patient in our study developed a psoas abscess after epidural catheter infusion and subsequent radiculopathy caused by the foraminal infection. Although her symptoms suggested the possibility of an accompanying epidural abscess, such a diagnosis could not be made because the radiologic findings did not demonstrate evident signs of epidural involvement. The lack of epidural infection might be possible since the catheter could have passed out the intervertebral foramen or bacteria could have been washed out the foramen during injection. Indeed, in his computed tomography studies of epidural catheter tip position, Hogan 6showed that tips were most often found lateral to the dura, in the foramen and sometimes in the paravertebral tissue lateral to the foramen.
Fibrous foraminal stenosis is not a rare etiology of lumbar radicular symptoms. Kunogi and Hasue 7reported that out of their 28 patients with surgically confirmed foraminal nerve root involvement, four had no apparent nerve root compression, but narrowing and adhesion of the nerve root. However, to our knowledge there has been no report of iatrogenic foraminal stenosis.
Several similarities in symptoms of two different clinical entities, i.e. , L2 nerve root irritation and psoas abscess, could be observed. Low back pain with possible thigh or hip radiation is the most common symptoms for both diseases. Hip extension provokes thigh pain in L2, L3, and L4 nerve roots irritation, however, hip flexion position is a classic sign of the psoas abscess. 8,9These similarities have made authors hesitate to perform immediate surgery. Although foraminotomy was performed 7 months after the onset of radicular pain, the pain completely disappeared even when surgery was performed late.
In conclusion, we present the first reported case of foraminal and psoas abscess as a complication of epidural anesthesia, which developed secondary foraminal stenosis. The decompressive surgery was effective for persistent radiculopathy because of secondary foraminal stenosis.
The authors thank Professor Katsuyuki Fujii, M.D., Ph.D., (Chairman, Department of Orthopaedic Surgery, The Jikei University, School of Medicine, Tokyo, Japan) for his kind advice and supervision.