2013 Outstanding Paper Runner-upSpinal epidural abscesses: risk factors, medical versus surgical management, a retrospective review of 128 cases
Introduction
Spinal epidural abscess (SEA) is a rare condition with potentially devastating consequences. Historical rates of SEA range from 0.2 to 1.2 cases per 10,000 hospital admissions [1]. Rates are expectedly higher at referral centers, 12.5 cases per 10,000 [2], and are increasing, having doubled in the past 20 years [3]. Approximately 50% of patients are initially misdiagnosed at time of presentation (range, 11%–75%) [4], [5].
Spinal epidural abscess results from purulent material collecting between the spinal dural covering and osseous-ligamentous structures of the spine. This condition was first described by Morgagai in 1761 [6] and clearly defined by Bergamaschi in 1820 [7]. Barth performed the first known surgical intervention for SEA in 1901 [1] and most of the early reported cases were fatal [2].
Bacteria gain entrance to the epidural space via hematogenous spread (half of cases), contiguous spread (one third of cases), and no identifiable source in the rest [3]. The primary reason for spinal cord injury is unknown. Leading theories include ischemia from direct compression or disruption of vascular supply from septic thrombophlebitis [3]. Clinical improvement after decompression and various animal studies supports the direct compression model [8], whereas others have shown the combination of direct compression and septic thrombophlebitis to be synergistically worse for outcomes [3].
Although urgent/emergent surgical decompression and IV antibiotic therapy is the cornerstone of therapy for SEA, ideal management of this condition remains controversial. Neurologic function at the time of presentation is a key predictor of clinical outcome, but it is difficult to predict who will experience neurologic deterioration, leading many groups to endorse early operative decompression combined with IV antibiotics therapy as the treatment of choice [4], [9], [10], [11]. Still others have reported similar outcomes for those treated with surgery and IV antibiotics and those with IV antibiotics alone [12], [13], [14], [15], supporting the notion that some SEA may be managed successfully medically [16]. Thus, the literature is plagued with contradictory arguments stemming from small studies.
In 2005, 52 cases of medically managed SEA were reported. All patients were neurologically intact at presentation with three patients crossing over to surgery after their neurologic function declined. This group proposed that medical management is reasonable in neurologically intact patients with SEA, but requires close monitoring and urgent surgical decompression when neurologic changes occur [12]. In 1992, 37 cases from 1970 to 1990 of medically managed SEA were reviewed [17]. All SEA successful managed medically presented without neurologic dysfunction [12], [13], with smaller abscess size [14] and required closed monitoring for neurologic deterioration [5], [18]. Failed medical management of SEA has been reported in with minimal explanation for the failures [9], [13], [17], [19], [20].
Although it is largely accepted that decline in neurologic function is an indication for surgical decompression of epidural abscess, the majority of clinical decisions for medical versus surgical management of this condition are based on anecdotal evidence. Risk factors have been identified for development of SEA (elevated erythrocyte sedimentation rate, leukocytosis, intravenous drug use, diabetes mellitus, prior spine surgery) but the prognostic value of these risk factors is unknown [2], [21], [22], [23]. It is not known which patient will respond favorably to medical management and who will require surgical intervention. Owing to the occurrence of severe neurologic deficits after failed medical management of SEA, it is unethical to perform randomized controlled trials to determine the best treatment for this condition [3].
By studying the demographics, motor scores, and medical comorbidities of patients at our institution with SEA, we can identify risk factors to predict who will acquire SEA and prognosticate regarding the therapeutic efficacy of medical versus surgical management.
Section snippets
Materials and methods
We retrospectively reviewed the records of patients diagnosed with spontaneous SEA from a single quaternary referral medical center composed of two major hospitals from January 2005 to December 2011. The electronic medical record was accessed to obtain patient demographics, presenting complaints, radiographic features, pretreatment and post-treatment neurologic status, type of treatment (medical vs. surgical), and clinical follow-up. The pretreatment neurologic status was determined by using
Results
Over an 81-month-period, 128 consecutive patients were diagnosed with a spontaneous SEA. There were 79 males and 49 females with an age range of 22 to 83 years (mean, 52.9). The average follow-up was 241.1 days.
The most common presenting chief complaint was site-specific pain (100%), subjective fevers (50%), and extremity weakness (47%). The lumbar spine was the most commonly involved region; 54.7% patients had some lumbar involvement. The thoracic spine was involved in 39.1% of patients, the
Discussion
Approximately 50% of patients with SEA are initially misdiagnosed at time of presentation (range, 11%–75%) [4], [5]. This can have devastating consequences, because early diagnosis leading to prompt surgical intervention is regularly rewarded by better recovery [24]. Common problems associated with the diagnosis of SEA include ordering imaging of the wrong site where the SEA is not located, identifying only one of many epidural abscesses, ascribing all clinical findings to osteomyelitis, an
Conclusion
Surgical intervention for SEA results in improved motor scores compared with medical management. Surgical intervention should be undertaken as early after diagnosis of SEA as possible as delayed surgical interventions results in lower postoperative motor scores than early surgery. There is a baseline failure rate of 8.3% for medical management of SEA. Diabetes mellitus, leukocytosis greater than 12.5, positive blood cultures, and C-reactive protein greater than 115 are risk factors for failure
Acknowledgments
The authors would like to thank Philip Louise and Juan Ortiz for their assistance in the writing of this article.
References (24)
- et al.
The clinical presentation and impact of diagnostic delays on emergency department patients with spinal epidural abscess
J Emerg Med
(2004) - et al.
Spinal epidural abscess–experience with 46 patients and evaluation of prognostic factors
J Infect
(2002) - et al.
Adult spinal epidural abscess: clinical features and prognostic factors
Clin Neurol Neurosurg
(2002) - et al.
Spinal epidural abscess: clinical presentation, management, and outcome
Surg Neurol
(2005) - et al.
Spinal epidural abscess: a report of 40 cases and review
Surg Neurol
(1992) - et al.
Spinal epidural abscess in adults caused by Staphylococcus aureus: clinical characteristics and prognostic factors
Clin Neurol Neurosurg
(2012) - et al.
Spinal epidural abscess
N Engl J Med
(1975) - et al.
Spinal epidural abscess: contemporary trends in etiology, evaluation, and management
Surg Neurol
(1999) Spinal epidural abscess
N Engl J Med
(2006)- et al.
Experimental spinal epidural abscess: a pathophysiological model in the rabbit
Neurosurgery
(1987)
Sacrococcygeal teratoma
Ann Surg
The pathogenesis of spinal epidural abscess: microangiographic studies in an experimental model
J Neurosurg
Cited by (193)
The utility of vertebral Hounsfield units as a prognostic indicator of adverse events following treatment of spinal epidural abscess
2024, North American Spine Society JournalPredictors of reoperation after surgery for spinal epidural abscess
2022, Spine Journal
FDA device/drug status: Not applicable.
Author disclosures: ARP: Nothing to disclose. TBA: Nothing to disclose. RJB: Speaking/Teaching Arrangements: AOSpine (B). MJL: Research Support: AOSpine (D, Paid directly to institution); Grants: AHRQ (Coinvestigator) (F, Paid directly to institution); Fellowship Support: AOSpine (E, Paid directly to institution); Consulting: Stryker Spine (C), L&K Biomed (None), Lilli (None). CBB: Nothing to disclose. JRC: Speaking/Teaching Arrangements: UCSF (B), University of Utah (B); Trips/Travel: AO Spine (B); Board of Directors: AO Spine NA (C); Endowments: H.J. Wyss Foundation (H, Paid directly to institution); Research Support: AO Spine NA (D, Paid directly to institution); OREF (E, Paid directly to institution); Fellowship Support: AO Spine NA (E, Paid directly to institution), Omega (E, grant, Paid directly to institution/employer).
The disclosure key can be found on the Table of Contents and at www.TheSpineJournalOnline.com.