The Case

A 30-year-old woman with a history of prior spine surgery presented to the emergency department with a few days of progressive low back pain. She had undergone spinal fusion 1 year prior and was generally healthy and functional. She described the pain as moderate-to-severe, radiating down the left leg and buttock. Because of her prior surgery, a magnetic resonance imaging (MRI) scan was ordered and showed lumbar degenerative joint disease as well as a small L5–S1 disc herniation. She was referred for physical therapy and prescribed a muscle relaxant, nonsteroidal anti-inflammatory medications (NSAIDs), and opioids.

Because of her previous spinal fusion, she made an appointment at the orthopedic clinic 1 week after her initial presentation in the emergency department. At this point, she continued to have back pain and was developing progressive difficulty walking due to the pain. The orthopedic surgeon agreed with the emergency department evaluation and made no further changes.

Ten days later, she presented to a community hospital. She was febrile, unable to ambulate, numb from the waist down, and using a wheelchair. She had a white blood cell count of greater than 30,000 (normal range 4,000–11,000) and was found to be in acute renal and liver failure with a creatinine of 3.6 mg/dL (normal range 0.4–1.0 mg/dL) and markedly elevated liver function tests.

She was immediately transferred to a neurosurgery service at an academic hospital where an MRI showed a T6–T10 thoracic epidural abscess. The patient was taken to the operating room for drainage of the abscess. Postoperatively, she had multiple complications including a pulmonary embolism. She only partially recovered function of her legs and was mainly using a wheelchair.

Subsequent review of the initial MRI to determine why the lesion was initially missed found that the MRI was a lumbar view, which captured the levels T11 through S1 but did not include the affected spinal level (T6–T10).

The Commentary

by Yi Lu, MD, PhD, and Doug Salvador, MD, MPH

Low back pain is a very common medical complaint. It is the third most common reason for visits to the doctor's office, behind skin disorders and osteoarthritis. It is estimated that up to 80% of the population will experience back pain at some time in their lives.(1,2)

On the other hand, spinal epidural abscess (SEA) is uncommon, with an estimated incidence of 0.2/10,000–2.0/10,000 hospital admissions.(3) However, recent studies have suggested the trend might be rising.(4) A recent 10-year case series documented an increase in rate from 2.5/10,000 to 8/10,000 hospital admissions between 2005 and 2015.(5) The reason for this possible increase is unclear.

Spinal epidural abscess is a serious illness. Missed diagnosis and delayed treatment of SEA may lead to devastating neurological deficits, and even death. Thoracic spine SEA (compared with cervical or lumbar spine) is of particular concern, as it is generally associated with more rapid clinical deterioration and more severe neurological deficit. This difference is likely due to the smaller canal diameters in the thoracic level, which lead to more severe early compression of the spinal cord.(6) Prompt diagnosis of SEA is not always easy, as clinical symptoms of SEA are not specific (especially at the onset) and can mimic many benign conditions.

Detailed history taking to identify potential risk factors for SEA is important in establishing the early diagnosis. Spinal epidural abscesses can originate from bloodstream infections, local spread from another infection (e.g., psoas muscle abscess), or direct inoculation during a procedure or surgery. The most common risk factors for SEA are diabetes mellitus, intravenous drug use, trauma, and alcohol use disorder.(7) In a patient with back pain, several "red flags" can be identified as part of the history that make SEA more likely. Specifically, the presence of fever in a patient with new back pain is very concerning.(8) Focal neurologic complaints (e.g., leg weakness, bowel incontinence) are also very concerning. However, the classic symptom triad of back pain, fever, and neurological deterioration is present in only 10%–15% of SEA patients at first contact.(9) Some clues may be present on physical examination as well. Local point percussion tenderness might raise the initial suspicion. A severe, circumscribed tenderness is a frequent early finding for patients developing early SEA. Certainly, a focal neurologic finding (e.g., left leg weakness on examination) may be evidence of a SEA or other serious disease.

Magnetic resonance imaging of the spine with and without gadolinium is the gold standard for diagnosis. If there is concern for SEA, for example, if a patient presents with back pain and fever or with focal neurological findings or with local severe point tenderness of the spine, then MRI with and without gadolinium should be obtained. Plain radiography would not be helpful in diagnosing early SEA, as early SEA does not have detectable bony changes. Precise threshold for triggering an MRI evaluation is not known based on the current evidence. Clinicians should use their judgment and just keep a high suspicion for the diagnosis in patients with back pain and any other concerning features. However, MRI is also costly, and it is unrealistic and unnecessary to perform MRI on every single back pain patient presented to a primary care provider's office or emergency department. Thoracic spine SEA is particularly easy to miss, as without a special suspicion, the majority of image studies would be a lumbar spine MRI, instead of a whole spine or thoracic-dedicated MRI. Therefore, vigilance is needed for providers when evaluating patients presented with above symptoms or exam findings. When there is a high suspicion for spinal epidural abscess, a full spine MRI is needed.

In this case, we do not know whether there were any red flags that should have prompted thoracic imaging. We also do not know whether the back pain is more located in the lower back or in the thoracic area, neither do we know whether she has point tenderness on physical examination. Because of the past lumbar spine surgery, a lumbar spine MRI was ordered, which is most likely appropriate. Prior spine surgery 1 year ago should have minimal impact on the possibility of having a distant area spinal epidural abscess. However, when the patient presented to the orthopedic clinic 1 week after, she developed progressive walking difficulty due to the pain. At this time, the progressive neurological decline should have prompted the provider to carefully reassess the patient and to perform a detailed physical examination. Detailed examination at this point might have prompted the treating physician to suspect thoracic spinal issues and order a thoracic or full spine MRI. Unfortunately, this was not done. Ten days later, the patient presented to the community hospital with fever, inability to ambulate, and numbness from the waist down. It is unclear at what point she developed true weakness and sensory changes.

To avoid missing the diagnosis of the potentially devastating SEA among patients presenting with low back pain like this one, ordering excessive imaging is not the answer. Detailed history taking and careful physical examination with a high vigilance from the treating physicians are the key for prompt and accurate diagnosis. A recent study from Department of Veterans Affairs found that 55% of all SEA cases were initially misdiagnosed.(10) Misdiagnosis was most often the result of inadequate recognition of red flag signs (fever, progressive neurological deficits, active infection) and inadequate initial evaluation performed by the treating physician. In addition to these important considerations in the clinical process, applying a quality improvement approach at a systems level can make an impact.

The National Academy of Science, Engineering, and Medicine made recommendations about reducing diagnostic errors in their landmark report, Improving Diagnosis in Health Care.(11) The report encourages organizations to create and implement approaches to identify, learn from, and reduce diagnostic errors in clinical practice. In addition, systems should facilitate more effective teamwork in the diagnostic process as well as establish a culture that supports improvement in diagnostic performance. This case was reviewed using the Diagnosis Error Evaluation and Research (DEER) Taxonomy.(12) Ten of 32 possible contributing factors were identified including failure/delay to follow-up critical piece of history data, ordering of the wrong test, too much weight on competing/coexisting diagnosis, and failure/delay in timely follow-up/rechecking of patient.

A quality improvement approach to reduce harm from cases like this one may include identifying cases of missed or delayed spinal epidural abscess and performing root cause analysis using a DEER Taxonomy framework.(13) This activity at our institution led to the development and implementation of a standard diagnostic pathway for SEA.(14) This lowered the threshold for clinicians to order MRI for suspected SEA, created an operational process that significantly lowered the time from ordering to final reading, and enhanced the teamwork between emergency, radiology, and neurosurgery services in caring for these patients. Additionally, the protocol ensured that the thoracic level was imaged in all cases of suspected SEA. After implementation of the protocol, more patients were diagnosed with SEA each month and fewer had neurologic deficits at the time of diagnosis. The Improving Diagnosis In Medicine Change Package can be used as a framework for setting up organizational structures and processes to achieve robust learning systems, reliable diagnostic process, and effective teamwork locally.(15)

In summary, thoracic SEA is a rare yet serious medical condition that may have a devastating or even fatal outcome with delayed diagnosis and treatment. One of the main challenges is that the initial presentation can be similar to an extremely common everyday medical condition. In such situations, a detailed history taking to look for risk factors and a careful physical examination looking for subtle neurological signs and point tenderness is essential. Recognition of characteristic signs and symptoms through clinical judgment and vigilance can lead to prompt diagnosis, allowing rapid surgical intervention. In addition, it is important to educate patients to recognize signs indicating neurological deficits associated with severe back pain and to seek medical help as soon as such symptoms arise.

Take-Home Points

• Low back pain is a very common complaint. Thoracic spinal epidural abscess is a rare but serious condition that may only present with low back pain.
• Careful history taking to identify the risk factors and meticulous physical examination to detect local tenderness and onset of neurological deficits are essential in the initial evaluation.
• Patients suspected of having spinal epidural abscess should be referred for timely imaging.
• Systems approaches to build robust learning systems, reliable diagnostic process, and effective teamwork should be applied to this and other types of diagnostic error.
• Prompt treatment, generally in the form of surgical evacuation, before the onset of severe neurological deficits, provides a good chance of functional recovery even with thoracic spinal epidural abscess. Delayed diagnosis and treatments frequently lead to permanent neurological deficits or even death.

Yi Lu, MD, PhD
Director, Neurosurgical Trauma
Co-director, Adult Deformity and Scoliosis Surgery
Assistant Professor of Neurosurgery
Brigham and Women's Hospital Harvard Medical School

Douglas Salvador, MD, MPH
Chief Quality Officer, Baystate Health
Chief Medical Officer, Baystate Medical Center
Associate Professor of Medicine
University of Massachusetts Medical School–Baystate

Acknowledgment: This case was produced in cooperation with the Primary-Care Research in Diagnosis Errors (PRIDE) Learning Network.

References

1. Low Back Pain Fact Sheet. Bethesda, MD: National Institute of Neurologic Disorders and Stroke; December 2014. [Available at]

2. St. Sauver JL, Warner DO, Yawn BP, et al. Why patients visit their doctors: assessing the most prevalent conditions in a defined American population. Mayo Clin Proc. 2013;88:56-67. [go to PubMed]

3. Mackenzie AR, Laing RBS, Smith CC, Kaar GF, Smith FW. Spinal epidural abscess: the importance of early diagnosis and treatment. J Neurol Neurosurg Psychiatry. 1998;65:209-212. [go to PubMed]

4. Krishnamohan P, Berger JR. Spinal epidural abscess. Curr Infect Dis Rep. 2014;16:436. [go to PubMed]

5. Artenstein AW, Friderici J, Holers A, Lewis D, Fitzgerald J, Visintainer P. Spinal epidural abscess in adults: a 10-year clinical experience at a tertiary care academic medical center. Open Forum Infect Dis. 2016;3:ofw191. [go to PubMed]

6. Howie BA, Davidson IU, Tanenbaum JE, et al. Thoracic epidural abscesses: a systematic review. Global Spine J. 2018;8(suppl 4):68S-84S. [go to PubMed]

7. Relhsaus E, Waldbaur H, Seeling W. Spinal epidural abscess: a meta-analysis of 915 patients. Neurosurg Rev. 2000;23:175-204. [go to PubMed]

8. Deyo RA, Rainville J, Kent DL. What can the history and physical examination tell us about low back pain? JAMA. 1992;268:760-765. [go to PubMed]

9. Alerhand S, Wood S, Long B, Koyfman A. The time-sensitive challenge of diagnosing spinal epidural abscess in the emergency department. Intern Emerg Med. 2017;12:1179-1183. [go to PubMed]

10. Bhise V, Meyer AND, Singh H, et al. Errors in diagnosis of spinal epidural abscesses in the era of electronic health records. Am J Med. 2017;130:975-981. [go to PubMed]

11. Improving Diagnosis in Health Care. Committee on Diagnostic Error in Health Care, National Academies of Science, Engineering, and Medicine. Washington, DC: National Academies Press; 2015.

12. Schiff GD, Kim S, Abrams R, et al. Diagnosising diagnosis errors: lessons from a multi-institutional collaborative project. In: Henriksen K, Battles JB, Marks ES, et al, eds. Advances in Patient Safety: From Research to Implementation, Vol 2. Rockville, MD: Agency for Healthcare Research and Quality; February 2005. AHRQ Publication No. 050021. [Available at]

13. Reilly JB, Myers JS, Salvador D, Trowbridge RL. Use of a novel, modified fishbone diagram to analyze diagnostic errors. Diagnosis (Berl). 2014;1:167-171. [go to PubMed]

14. Madhuripan N, Hicks RJ, Feldmann E, Rathlev NK, Salvador D, Artenstein AW. A protocol-based approach to spinal epidural abscess imaging improves performance and facilitates early diagnosis. J Am Coll Radiol. 2018;15:648-651. [go to PubMed]

15. Improving Diagnosis in Medicine Change Package. Chicago, IL: Health Research & Educational Trust; 2018. [Available at]