Favorable outcome of long-lasting thoracic spondylodiscitis with spinal epidural abscess induced by Staphylococcus aureus. (Case Report).

Abstract: A favorable outcome in chronic spondylodiscitis with epidural abscess is rare. A 65-year-old woman developed recurrent, localized, thoracic back pain over 2.5 years. Nine months after the onset of the pain, sensory disturbances of the left lower leg occurred. Fourteen months before admission, she developed recurrent fever, bladder dysfunction, and weakness and numbness of both lower legs. An incomplete sensory transverse syndrome with paraparesis was diagnosed. Magnetic resonance imaging of the spine suggested destruction of the T11 vertebral body, with spondylodiscitis of the adjacent discs and an epidural abscess between levels T4 and T9. Laminectomy was immediately performed and the abscess was drained. Culture revealed infection with Staphylococcus aureus. After 10 weeks of therapy, recovery was almost complete. Spondylodiscitis with epidural abscess may have a favorable outcome, even if symptoms start more than 2 years before treatment.

Key Words: epidural abscess, laminectomy, neurosurgery, Staphylococcus aureus infection, spondylodiscitis, thoracic spine

**********

Spondylodiscitis with epidural abscess is a rare but potentially devastating condition. (1,2) If the diagnosis is established late and the treatment is inadequate, severe, long-term disability may develop. (3,4) We report the case of a patient with chronic spondylodiscitis and concomitant epidural abscess whose symptoms progressed over 2.5 years from back pain to acute transverse syndrome, but who still had a favorable outcome with almost complete recovery.

Case Report

A 65-year-old woman reported focal thoracic dorsalgia for the first time in October 1996 after sitting for several hours; treatment with nonsteroidal analgesics relieved her symptoms. In March 1997, thoracic back pain recurred, radiating subcostally toward the abdomen, after another long period of sitting. Analgesics again promptly relieved the symptoms. In July 1997, the patient first noticed short-term numbness of the left lower leg; clinical neurologic examination at that time was normal. Back pain recurred in November 1997 after the patient had been sitting for 4 to 5 hours, but it responded well to nonsteroidal analgesics. In February 1998, she developed fever up to 40[degrees]C, without symptoms or signs of a common cold, bronchitis, pneumonia, or gastrointestinal or urinary tract infection. In August 1998, the patient experienced recurrent, nonsystematic vertigo; clinical neurologic examination was again normal. Bladder incontinence began in October 1998, and she noted numbness of the left foot after s trenuous exercise. Beginning in January 1999, she noted slowly progressive weakness of the left leg and permanent numbness of the left foot. Fever (up to 39[degrees]C) recurred in February 1999. By March 1999, the numbness had extended up to the left knee, and the patient complained of increasing fatigability. In addition, the previously noted back pain recurred with much greater intensity. Conventional xray film of the thoracic spine showed hyperkyphosis, scoliosis, and wedge-shaped (cuneiform) vertebral bodies at T6, T9, and T11, with impression and reactive sclerosis of the corresponding roof and basal plates. Despite treatment with more potent analgesics, back pain worsened and the weakness and numbness spread to the right leg, resulting in gait disturbance. Urine retention occurred and bladder catheterization was necessary.

At admission in April 1999, the patient complained of severe thoracic back pain, bladder and bowel incontinence, numbness of both lower legs, and the inability to stand or walk unaided. On clinical neurologic examination, the thoracic spine was tender, there was a sensory transverse syndrome at L2 on the left side and at T12 on the right side, and there was paraparesis (Medical Research Council Grade 3 on the left and Grade 4 on the right), making unaided standing and walking impossible. Laboratory examinations revealed a C-reactive protein level of 29.3 mg/dl (normal, <0.8 mg/dl). The erythrocyte sedimentation rate was 98/110 mm/h (normal, <10/20 mm/h), and the gamma globulin level was 24.3% (normal, <22%). Plain x-ray of the thoracic spine appeared unchanged when compared with the previous film. Magnetic resonance imaging (MRI) of the spine suggested incomplete destruction of the T11 vertebral body, with spondylodiscitis of the adjacent discs and a dorsolateral epidural abscess from T4 through T9 (Fig. 1). That same day, the patient underwent complete laminectomy of T7 and incomplete laminectomy of T6 and T8. The abscess was emptied and drained over the 8 days following surgery. Postoperative course was uneventful. Bone scintigraphy showed hyperactivity around T11, but the leukocyte scintigraphy showed no indication of local inflammation. The first postoperative MRI showed a small residual abscess, which was no longer visible on a control image. Culture of the drained material revealed infection with Staphylococcus aureus. For the next 3 months, the patient was treated with clindamycin and the spine was stabilized with a bodice. Under this regimen, patient mobilization began on the fourteenth postoperative day, leading to almost complete recovery 10 weeks after surgery. Follow-up MRI of the thoracic and lumbar spine in October 1999 and March 2000 showed a narrowed intervertebral space between T10 and T11, a fishbone-shaped deformity of the T11 vertebra, slight disc protrusion at T6/T7, increased thoracic kyphos is, fractures of the roof plate of L3, and disc protrusion at L3/L4. Follow-up examination in April 2000 showed that the patient was still able to walk unaided, but had developed axonal polyneuropathy of undetermined cause, with typical stocking-type sensory disturbances and ataxia.

Discussion

Spondylodiscitis in adults is a rare inflammatory disorder of the intervertebral discs and the adjacent vertebral bodies. (1,4-8) In 25 to % of the cases, spondylodiscitis is associated with an epidural abscess. (5,9) Spondylodiscitis may occur spontaneously or as a result of spinal surgery. Patients with diabetes mellitus, lung or systemic infection, IV drug abuse, chronic renal failure, alcoholism, malignancy, and those undergoing dialysis, transplantations, or whose immune system is depressed are more commonly affected. The most frequent microbial agents that cause spondylodiscitis are Staphylococcus. Brucella, Salmonella, and Mycobacterium tuberculosis. (2,5) The first clinical manifestation of spondylodiscitis is usually back pain over the infected area of the spine. (6,10) Back pain can be so severe that it is more disabling than other concomitant neurologic deficits. (6) The most frequently encountered neurologic finding of spinal epidural infection is a sensory deficit; (6) less frequently, urinary re tention or loss of rectal sphincter function are noted. The predominant motor sign in spondylodiscitis is paraparesis. (6) Frequently, the diagnosis of epidural infection cannot be made on clinical grounds alone because of the wide range of neurologic deficits and the difficulty in localizing the lesion.

The pathogenesis of spondylodiscitis and epidural abscess in our patient remains speculative. There was no surgery, infection, spinal tap, epidural catheterization, myelography, trauma, or adjacent local infection preceding the onset of thoracic back pain. The patient had no history of diabetes, IV drug abuse, hemodialysis, colospinal fistula, lung or systemic infection, chronic renal failure, alcoholism, malignancy, or immunodeficiency. No septic focus was found. Possible causes for spondylodiscitis and epidural abscess in this patient could have been alcoholism or occult malignancy. Possible entrances for S. aureus could have been dental surgery or genitourinary tract infection. Hematogenic infection was thought to be rather unlikely, because the vascular supply of the disc space is obliterated in adulthood. (6) Toxic effects of an inflammatory process are more likely to have been causative. Vascular compromise of the feeding arteries or draining veins could have been contributory as well. (6) Loss of spina l cord function in addition could have been due to cord encroachment from cord compression. Overall, it remains speculative whether spondylodiscitis in this patient had been present since the onset of low back pain. Arguments for long-term, chronic spondylodiscitis are that the inflammatory markers were high at each previous visit, that the patient reported fever 14 months before the actual deterioration, and that slight neurologic deficits had already occurred 2 years after disease onset.

X-rays of the thoracic spine, bone scans, and computed tomography (CT) usually provide insufficient data in patients with spondylodiscitis. (2,5,11) Plain radiography may show collapse of the intervertebral disc and destruction of the adjacent vertebral bodies; (6) CT may show disc flattening and vertebral endplate destruction. Magnetic resonance imaging is the investigational tool of choice in diagnosing spondylodiscitis, particularly in the early stages of the disease when other investigations still yield negative results. (2,5) Typical findings of spondylodiscitis on MRI are signal loss of the discs and the adjacent vertebral bodies on T1-weighted images and hyper-intensity on T2-weighted images; inhomogeneous enhancement of the discs, the adjacent vertebral bodies, and the epidural and paravertebral soft tissue; (6) narrowing of the intervertebral disc space; and pathologic intraspinal and paravertebral soft tissue. (12) Contrary to these characteristic MRI findings, the T11 vertebral body was hypointense on T1-weighted sequences in our patient, possibly due to an ongoing inflammatory process. A tuberculous spondylodiscitis as the cause of the latter finding was excluded. Histologically, the enhancing tissue represents inflammatory granulation tissue, whereas the nonenhancing areas represent necrotic tissue and pus. (12) Intramedullary contrast-enhancing lesions are rarely found. Because of the high incidence of back pain due to degenerative vertebral changes in adult patients, a general diagnostic workup with MRI in these patients is not warranted. If there is suspicion on clinical grounds of intervertebral infection despite normal findings on MRI, however, repeat MRI examinations are justified, since there is frequently a delay between the onset of clinical symptoms and the occurrence of imaging abnormalities.

The treatment of choice in spondylodiscitis with concomitant space-occupying lesions, such as abscesses or intraspinal or paravertebral soft tissue, is laminectomy to decompress the spinal cord. (5,13) In addition, antibiotics should be given and the spine should be stabilized. Spondylodiscitis alone can be managed by immobilization, treatment with antibiotics or anti-inflammatory drugs, or debridement of the infected (5,6,14) The postoperative course usually goes well, but sometimes surgical decompression is ineffective, particularly if the spinal cord compression is only mild or moderate. (5,16) Differential diagnoses that have to be considered are sarcoidosis; ankylosing spondylitis; inflammatory, noninfectious, degenerative spondylosis; and lymphoma.

Conclusion

This case shows that the outcome of chronic spondylodiscitis and spinal epidural abscess can be excellent, even if symptoms started years before disease onset and deteriorated to transverse syndrome. The diagnosis of chronic spondylodiscitis may be delayed, because initially symptoms are nonspecific. The most appropriate treatment for chronic thoracic spondylodiscitis with concomitant epidural abscess is multi-segmental laminectomy, drainage of the abscess, adequate antibiotic therapy, and spinal stabilization with a bodice.

Accepted January 16, 2002.

References

(1.) Danner RL, Hartman BJ. Update on spinal dural abscess: 35 eases and review of the literature. Rev Infect Dis 1987;9:265-274.

(2.) Mackenzie AR, Laing RB, Smith CC, Kaar GF, Smith FW. Spinal epidural abscess: The importance of early diagnosis and treatment. J Neurol Neurosurg Psychiatry 1998;65:209-212.

(3.) Martin RJ, Yuan HA. Neurosurgical care of spinal epidural, subdural, and intramedullary abscesses and arachnoiditis. Orthop Clin North Ant 1996;27:125-136.

(4.) Maslen DR, Jones SR, Crislip MA, et al. Spinal epidural abscess: Optimizing patient care. Arch Intern Med 1993;153:1713-1721.

(5.) Maiuri F, laconetta G, Gallicchio B, Manto A, Briganti F. Spondylodiscitis: Clinical and magnetic resonance diagnosis. Spine 1997;22:1741-1746.

(6.) Kuker W, Mull M, Mayfrank L, Topper R, Thron A. Epidural spinal infection: Variability of clinical and magnetic resonance imaging findings. Spine 1997;22:544-551.

(7.) Osenbach RK, Gullick RA: Spinal epidural abscess, in Osenbach RK, Zeidman SM (eds): Infections in Neurological Surgery: Diagnosis and Management. Philadelphia, Lippineott-Raven, 1999, pp 265-275.

(8.) Tacconi L, Johnston FG, Symon L. Spinal epidural abscess: Review of 10 cases. Acta Neurochir (Wien) 1996;138:520-523.

(9.) Kapeller P, Fazekas F, Krametter D, et al. Pyogenic infectious spondylitis: Clinical, laboratory and MRI features. Eur Neurol 1997;38:94-98.

(10.) Wong D, Raymond NJ. Spinal epidural abscess. N Z Med J 1998:111: 345-347.

(11.) Mak KH, Au KK, Fung KY, Chan YW. Spinal epidural abscess: A report of nine eases and the use of intra-operative ultrasonography. Aust N Z J Surg 1996;66:287-290.

(12.) Numaguchi Y, Rigamonti D, Rothman MI, et al. Spinal epidural abscess: Evaluation with gadolinium-enhannced MR imaging. Radiographic 1993;13:545-560.

(13.) Sampath P, Rigamonti D. Spinal epidural abscess: A review of epidemiology, diagnosis, and treatment. J Spinal Disord 1999;12:89-93.

(14.) Manfredi PL, Herskovitz S, Folli F, Pigazzi A, Swerdlow ML. Spinal epidumi abscess: Treatment options. Eur Neurol l998;40:58-60.

(15.) Liem LK, Rigamonti D, Wolf AL, et al. Thoracic epidural abscess. J Spinal Disord 1994;7:449-454.

(16.) Khanna RK, Malik GM, Rock JP, Rosenblum ML. Spinal epidural abscess: Evaluation of factors influencing outcome. Neurosurgery 1996;39:958-964.

RELATED ARTICLE: Key Points

* Chronic back pain may be the initial symptom of spondylodiscitis with an epidural abscess.

* Back pain associated with recurrent fever, bladder dysfunction, and paraparesis should be investigated immediately using magnetic resonance imaging of the spine.

* Transverse syndrome due to spondylodiscitis with an epidural abscess should be treated immediately with a polysegmental laminectomy, abscess drainage, antibiotics, and spinal stabilization.

* The outcome of transverse syndrome due to longlasting spondylodiscitis with an epidural abscess may be favorable, if adequately treated.

From the Neurology Department, Neurological Hospital Rosenhugel, the Neurosurgery Department, Sozialmedizinisches Zentrum Ost, and the Radiology Department, Krankenhaus Lainz, Vienna, Austria.

Reprint requests to Dr. Josef Finsterer, Postfach 348, A-1180 Vienna, Austria. Email: [e-mail]

Copyright [c] 2003 by The Southern Medical Association 0038-4348/03/9601-0070