Spinal epidural abscess: prognostic factors and comparison of different surgical treatment strategies

Spinal epidural abscess (SEA) is a rare but potentially devastating disease requiring immediate surgical intervention and appropriate antibiotic treatment. The standard approach to decompress SEA is laminectomy. No report covers comprehensively the indications for the less invasive interlaminar approach, the usefulness of intra-operative ultrasonography and the suspected benefit of inserting a suction-irrigation drainage.

A retrospective evaluation of the medical and radiological data was undertaken in 27 consecutive patients with SEA operated on during a period of 10 years by a dorsal approach. Factors influencing outcome were evaluated with special regard to different surgical strategies concerning the invasiveness of the operative approach, the use of intra-operative ultrasound and the use of different drainage systems.

Outcome was mainly determined by the pre-operative neurological condition and the localization of the abscess. Recurrence rate was dependent on the longitudinal extent of the mass and the intra-operative finding of granulation tissue, but not on the administration of a postoperative suction-irrigation drainage. An interlaminar approach was equally matched to a decompression by laminectomy in lumbar SEA concerning the incidence of residual/recurrent abscess formation. In concomitant spondylodiscitis, laminectomy bore the risk of the formation of a postoperative kyphotic deformity. The use of intra-operative ultrasound allowed the visualization of hidden inflammatory masses and, thus, reduced the rate of residual abscess formation.

An interlaminar approach should be considered instead of laminectomy in lumbar SEA and in impending anterior column instability due to spondylitis. Intra-operative ultrasound is a beneficial aid for the determination of the extent of decompression during surgery and is practicable even through a narrow interlaminar bony window. The insertion of postoperative suction-irrigation drainage had no beneficial effect on outcome but bore the risk of epidural fluid congestion.