Percutaneous computed tomography-guided needle aspiration drainage of spinal epidural abscess.

To the Editor: Spinal epidural abscess is an uncommon disease but one with potentially devastating neurologic consequences if not diagnosed and treated appropriately. Historically, surgical intervention has been widely accepted as the mainstay of treatment in most instances. However, accumulated clinical experience indicates that selected patients with spinal epidural abscess can be managed effectively with nonsurgical treatment alternatives. (1-3) We report our experience with percutaneous computed tomography (CT)-guided needle aspiration drainage in seven patients with spinal epidural abscess and review the literature on this management option.

From 1990 to 2004, we retrospectively reviewed the medical records of all patients admitted to our academic community hospital and identified seven patients with a diagnosis of spinal epidural abscess who underwent percutaneous CT-guided needle aspiration drainage. Mean patient age was 55 years. Six of seven (86%) patients had lumbar abscesses; all abscesses were located posteriorly in the spinal canal, and all masses were [less than or equal to]2 vertebral bodies in length. All but one subject had onset of symptoms <7 days before diagnosis, and only two had neurologic deficits (mild motor weakness) when first seen. Therapeutic needle aspiration was performed a median of 7 days after presentation (range, 1-60 days). All patients received at least 6 to 8 weeks of parenteral antibiotics. All patients had complete neurologic recovery following CT-guided needle aspiration and completion of antibiotic therapy. Subjects were followed as outpatients for at least 8 to 10 months posttherapy.

In recent years, studies of nonsurgical treatment (including prolonged IV antibiotic therapy alone) of selected patients with spinal epidural abscess have been increasingly reported. Diagnostic needle aspiration of spinal epidural abscess is commonly performed for microbiologic sampling to identify causative organisms. However, therapeutic drainage of these abscesses, as an alternative to decompressive surgery, has been only rarely reported. In small case reports and case series where sufficient details were stated, we could find only 11 previously described cases of spinal epidural abscess managed by percutaneous drainage, involving either needle aspiration or catheter drainage and irrigation. Among our seven cases and the 11 literature cases, only four patients had neurologic abnormalities at the time of initial presentation. (4,5) These were all mild in nature and included lower extremity motor weakness (3 patients) and difficulty with urination (1 patient). More severe motor or autonomic impairment or progressive neurologic deterioration should prompt open surgical intervention.

Criteria for consideration of percutaneous drainage of spinal epidural abscess include the following: 1) clear visualization of the abscess on MRI or CT scan; 2) accessibility of the abscess via location in the posterolateral epidural space; 3) radiographic suggestion of the presence of liquid purulent material within the abscess; 4) minimal or absent neurologic signs and symptoms; 5) refractory to medical therapy alone; and 6) poor surgical candidacy.

Magnetic resonance imaging has become the diagnostic modality of choice for spinal epidural abscess. Gadolinium-enhanced MRI is useful in detecting epidural abscesses in the phlegmatous stage, which corresponds to thickened granulomatous tissue with embedded microabscesses, without collection of liquified pus. These masses are more likely to present chronically rather than acutely. The radiographic finding of a significant amount of suspected granulomatous tissue may preclude adequate decompression via percutaneous drainage.

While others have suggested that larger spinal epidural abscesses that involve a considerable length of the vertebral canal may be more amenable to percutaneous drainage, all seven of our patients had abscesses [less than or equal to]2 vertebral bodies in size. Almost 90% of our patients had a subacute onset of symptoms (eg, <7 d), as did 64% of patients previously reported. The optimal timing of percutaneous drainage remains uncertain. In the 18 cases reviewed here, where stated, the time from diagnosis until percutaneous drainage of spinal epidural abscess ranged from 1 day to 2 months, with a median of 7 days. In many cases, especially for patients with stable neurologic examinations, a trial of medical treatment with parenteral antibiotics would seem warranted before percutaneous drainage is considered. (3) With puncture of the dura mater, there is a risk of development of meningitis, but this complication should be minimized when aspiration is performed by an experienced interventional radiologist.

In summary, percutaneous CT-guided needle aspiration can obviate the need for surgical drainage in selected patients with spinal epidural abscess. When combined with prolonged IV antibiotics in patients without significant neurologic deficits, the likelihood of full recovery is excellent.

Farhan Siddiq, MD

Asif R. Malik, MD

Raymond A. Smego Jr, MD, MPH, FACP, FRCP, DTM & H

University of North Dakota School of Medicine and Health Sciences Center

Fargo, ND

References

1. Khan SH, Hussain MS, Griebel RW, et al. Title comparison of primary and secondary spinal epidural abscesses: a retrospective analysis of 29 cases. Surg Neurol 2003;59:22-33.

2. Ahl T, Hedstrom M, von Heijne A, et al. Acute spinal epidural abscess without concurrent spondylodiscitis. Successful closed treatment in 10 cases. Acta Orthop Scand 1999;70:199-202.

3. Siddiq F, Chowfin A, Tight R, et al. Medical vs surgical management of spinal epidural abscess. Arch Intern Med 2004;164:2409-2412.

4. Lyu RK, Chen C, Tang L. Spinal epidural abscess successfully treated with percutaneous, computed tomography-guided, needle aspiration and parenteral antibiotic therapy: case report and review of the literature. Neurosurgery 2002;51:509-512.

5. Cwikiel W. Percutaneous drainage of abscess in psoas compartment and epidural space. Case report and review of the literature. Acta Radiol 1991;32:159-161.