Non-Traumatic Low Back Pain: Avoiding Liability for Missed Cord Compression

By Robert A. Bitterman, MD, JD, FACEP, Contributing Editor

Low back pain (LBP) is a common (more than 3 million ED visits per year in the United States) yet typically benign ED complaint.1 In approaching the back pain patient presenting to the ED, the emergency physician should determine, based on the patient's age and nature of complaints, whether the patient has an abdominal aortic aneurysm (AAA), spinal cord compression syndrome (SCCS), fracture, infection, benign LBP, or some other atypical etiology — an even half dozen possibilities to consider.2,3 The work-up of the patient, if any is indicated, and the documentation in the chart should reflect that thought process. Most patients will fit into the large category of mechanical/non-specific/non-serious LBP. Consider it the category for all patients with the "garden variety" LBP routinely seen in our EDs, but also note that it includes patients with an acute herniated disc if no new or acute neurological impairment is present. The logic for this is that the immediate management from our perspective in the ED, and the outpatient referral for all patients in this category, is initially the same.1,3

Epidural, spinal, or cauda equina compression syndrome, often referred to as spinal cord compression syndrome (SCCS) is one of the few true surgical emergencies that can present as LBP (AAA is another).4,5 (See box, below.) Etiologies of epidural mass that can produce cord compression include tumor, metastasis, epidural hematoma, epidural abscess, recent back procedure or epidural anesthesia, and herniated disc.2,3,5,6

This article will concentrate on the recognition, management, and the malpractice liability issues related to these cord compression syndromes.

Lawsuits for misdiagnosis of spinal cord compression syndrome (SCCS), delay in diagnosis, or delay in consultation with neurosurgery are becoming increasingly common (for example, the insurance company I'm affiliated with has had 5 cases in the last three years alone), and the catastrophic patient damages can lead to correspondingly large financial losses for emergency physicians and/or their insurance companies.7,8

Patient History

The typical elements that should be addressed in the history of an ED patient complaining of non-traumatic LBP, and documented in the medical record, include the following:2,3,9

  • Characterization of the pain, including onset, location, nature, radiation/sciatica, duration, recent changes, ameliorating or exacerbating factors, and severity;
  • Associated symptoms, including fever, chills, night sweats, weight loss, bowel or bladder problems, and numbness-tingling-weakness in perineum or lower extremities;
  • Past back history (previous back pain, injury, surgery, herniated disk(s), etiology, treatments, and response to prior therapy);
  • Past medical history, including malignancy, tuberculosis, immunosuppression, diabetes, recent infections, hypertension (HTN), AAA, urinary tract stones);
  • Medications, particularly warfarin sodium, steroids, and current pain medications; and
  • Social history (IVDA [intravenous drug abuse], smoking).

Risk factors for possible spinal infections, such as epidural abscesses, include fever, IVDA, immunosuppression (including chronic steroid use), and urinary infection. Fever and back pain in the IVDA patient should be considered spinal infection until proven otherwise (epidural abscess, osteomyelitis, or endocarditis).4

Cancer risk factors include age older than 50, prior history of cancer, unexplained weight loss, and failure to improve after 4-6 weeks of conservative LBP therapy (chronic LBP). If all four of these risk factors for cancer are absent, studies suggest that the possibility of cancer causing the patient's LBP is essentially zero.1,10

Documentation of urinary bladder function is crucial. Cauda equina syndrome, by definition, only exists if the patient has bowel or bladder dysfunction (usually urinary retention or urinary incontinence).5,7,8 Urinary retention typically precedes incontinence. "No bowel or bladder problems" is a mandatory, pertinent negative history that must be documented on the chart of any patient discharged from the ED who presented with a complaint of LBP (or radicular pain down one leg).

True sciatica (LBP with radiation of the pain past the knee, not just into the posterior thigh) has such a high sensitivity (95%) that its absence makes nerve root compression from lumbar disc herniation unlikely.1,3,10

Recent back procedure, epidural anesthesia, warfarin sodium, or heparin therapy raise suspicion for an epidural hematoma as a cause of pain or SCCS.2-4

Physical Examination

The typical elements of the physical exam of the LBP patient include the following:2,9 vital signs; palpation of the back for vertebral tenderness, flank tenderness; abdominal palpation for AAA or other entities that refer pain to the back; range of motion of the back and lower extremities; straight leg raise testing; neuromuscular testing of lower extremities (motor, sensory, and reflexes); gait; perineal/perirectal sensation; and rectal tone.

The abdomen of every patient with LBP should be examined, not only in consideration of AAA, but also to identify the many intraabdominal conditions that can refer pain to the back.9

The emergency physician must pay special attention to the neurological exam of the lower extremities, including detailed motor, sensory, and reflexes exam and a check for saddle/perianal anesthesia.

It is not necessary to perform a digital rectal exam on every patient with LBP (though if done routinely it might cut down the number of ED visits). However, it should be done whenever patients present with bowel or bladder complaints, lower extremity neurological symptoms, or perineal paresthesias. Weak rectal tone and saddle anesthesia are indicative of SCCS.1,9,10 (See Table 1.)

Straight leg raise (SLR) should be assessed to evaluate for nerve root impingement, which is usually but not always due to lumbar disc herniation. Positive SLR is defined as pain in the posterior lateral lower extremity that radiates below the knee with the patient lying supine and the hip flexed 60 degrees or less. It is suggestive of disc herniations, which in more than 95% of cases occur at the L4-5 or L5-S1 levels. Increased back pain alone, without radiation below the knee, does not constitute a positive SLR test. A negative SLR rules out surgically significant disc herniation with more than 95% specificity.1-3,10

A positive crossed straight leg test (radicular pain down the symptomatic leg when elevating the asymptomatic leg) is highly specific for a herniated disc.1,10

Approximately 2.5-6% of patients with disk herniations ultimately suffer cord compression syndrome, which is why discharge instructions for the many patients we see with potential disc disease are so important.1,10

Laboratory Studies

The laboratory is not very useful in the evaluation of patients with LBP.6 A CBC and erythrocyte sedimentation rate (ESR) may help if there is a suspicion of cancer, ankylosing spondylitis, or infection as the etiology. Blood cultures are indicated for presumed epidural abscess or osteomyelitis; as they will often identify the infecting organism. Prothrombin time and INR (international normalized ratio) are indicated for patients on warfarin sodium anticoagulation.

Imaging Studies to Evaluate LBP / Possible Cord Compression

Lumbosacral [LS] Spine X-rays (LS-Spine). Most patients with low back pain do not require any imaging.1,10,11 Consider LS x-rays in patients presenting with new LBP and a history of trauma (including relatively minor trauma in the elderly), unexplained fever or weight loss, immunosuppression, cancer, prolonged steroid use, osteoporosis, or age older than 70. The purpose of LS radiographs is to exclude malignancy/metastasis, fracture, or infection (osteomyelitis).1,10,12

Otherwise, x-rays are of no use and are not indicated in the evaluation of patients presenting with LBP, including patients with sciatica, ruptured discs, and/or cauda equina.1,10,11

The ordering of LS x-rays for "medical legal" reasons to avoid litigation also is not recommended. It is better to explain to the patient why they don't need x-rays and the unnecessary potentially harmful radiation, and then document your reasoning in the medical record. (This also mitigates the inevitable subliminal incompetence question from the family of "You mean they didn't even do any x-rays?")

LS-spine AP (anteroposterior) and lateral x-rays deliver a high dose of radiation, which is about 40 times the dose of a routine chest x-ray. The ovarian dose of radiation to a woman is equivalent to having a chest x-ray daily for more than 6 years.

Also, oblique view x-rays are not recommended since they add minimal information in a fraction of cases and more than double the radiation exposure.13

CT Scan of LS-spine. There is no reason to do a CT scan of the LS-spine in an ED patient to rule out an acute herniated disk. The initial management is the same for the first few weeks, regardless of whether the patient has simple undifferentiated LBP or a new ruptured disc as the cause of the acute LBP.1,10 Explain to the patient that imaging studies may be indicated later, at the discretion of their physician, if his/her symptoms don't resolve over the next few weeks.

Furthermore, CT scan is not the modality of choice to evaluate patients presenting with suspected spinal cord compression. An MRI (magnetic resonance imaging) is a much better diagnostic tool to identify cord compression, its etiology, and the extent or disease.2,6,8 CT scan can be used in patients who are not candidates for MRI, such as those with implanted electrical devices, the morbidly obese, or those who cannot lie still or cooperate for MRI. A CT myelogram (dye is injected into the thecal space) can be useful in facilities that do not have MRI capabilities on site or within a reasonable transfer radius.5,11

A CT scan also is not the right choice for a suspected ruptured AAA. Best is a bedside ultrasound. It's fast, easy, cheap, nearly 100% sensitive, and does not take the patient out of the ED. Only truly stable patients should go to CT.

Any time a non-contrast helical CT scan is ordered to rule out urinary tract stone/obstruction in a patient older than age 50, the radiologist should always view the aorta to detect an "unsuspected" AAA.

MRI of the Spine. MRI is the indicated study to evaluate all of the epidural compression syndromes, including cauda equina syndrome, epidural abscess, epidural hematoma, neoplasm, or infection.1,8

The advantages of MRI over CT scanning include the following:1,8

  • Better visualization of soft tissue pathology, such as epidural abscesses or ruptured discs;
  • Direct visualization of spinal cord and nerve roots;
  • Improved sensitivity for cord pathology or intrathecal masses;
  • Better sensitivity for infection and neoplasm;
  • No radiation exposure;
  • Safer for pregnant women, especially in the first trimester, because there is no radiation exposure.

Medical Decision Making, Consultation Issues

Studies indicate malpractice suits related to SCCS are primarily for failure to recognize the disease on a timely basis, failure to obtain a stat MRI to make the diagnosis, and failure to consult an appropriate spine surgeon on a timely basis.7,8 Not surprisingly, most often the delays in imaging or consultation came on nights, weekends, and holidays. It's important that clinicians know in advance if an MRI is available, and if not, that they know where to transfer the patient to obtain one promptly. Clinicians also should be aware of the capabilities of the surgical staff. If there is not a neurosurgeon or spine surgeon with the expertise to operatively relieve a cord compression, clinicians should to know where to transfer the patient.

Whenever the LBP patient is complaining of any urinary troubles, the emergency physician should catheterize the patient to obtain a post-void residual urine measurement. Ultrasound measurement of post void residual is a reasonable alternative. An amount greater than 50-100 cc is indicative of urinary retention and should prompt immediate consideration of SCCS and initiate the process of obtaining the MRI and appropriate consultation.4,5,7

The emergency physician and the nursing staff should be vigilant in observing for changes in the LBP patient's neuoromotor symptoms or findings while the patient is still in the ED, particularly for patients with a prolonged ED stay due to an overwhelmed department, delay in obtaining an imaging study, awaiting consultation, or lack of in-patient bed availability. More than 85% of patients develop SCCS over a period of only a few hours, which may be while the patient is in the ED!5,8 Any significant or progressive neuromotor deficit requires immediate surgical consultation.

The treatment for developing SCCS is surgery, and the speed of surgical decompression directly correlates with outcome (e.g., avoiding paraplegia, permanent bladder or bowel incontinence, or sexual dysfunction).5,7,8 Controversy exists over the relationship between outcomes and the timing of surgery, particularly in the first 24-48 hours, but there is near universal agreement that earlier is better than later.4,5,7,8 Thus, the role of the emergency physician is to avoid delay in recognition and delay in referral to the appropriate specialist.

The use of steroids for acute cord compression also is controversial. It is common practice to initiate high-dose steroids in the ED, particularly if a malignancy is suspected, but there is no significant primary evidence to support the practice.14 Emergency physicians should make the diagnosis, raise the steroid issue with our consultants, and let them make the decision for each individual patient.

Transfers. Whether the patient with suspected SCCS should undergo a diagnostic procedure prior transfer can be a very difficult clinical decision. When in doubt, discuss the question with the accepting specialist and document your interaction and reasoning.

Discharge Instructions, Referral, and Follow-up

Inadequate discharge instructions are only occasionally the source of litigation related to SCCS; usually SCCS lawsuits arise from failure of the emergency physician to diagnose the disease process and/or refer the patient to the proper specialist in a timely manner.7,8 (See Table 2.)

However, proper discharge instructions may be an invaluable defense tool if litigation stems from a case in which the patient develops SCCS after the initial ED visit. As noted above, most patients develop SCCS over a period of only a few hours. Thus, when discharging a patient with LBP or sciatica, the emergency physician must instruct the patient on what symptoms to look for and the proper procedure for emergency follow-up. (See Table 3.) Thereafter, a patient's failure to follow-up as instructed creates a strong defensive position for the emergency physician if the patient suffers an adverse outcome.

Inform all patients with disc disease/LBP that cord compression is a possible complication of their disease and that they should immediately contact their physician or return to the ED if they experience any bowel or bladder problems, numbness or tingling in the perineal area, or weakness in a lower extremity.

The discharge instructions should be given verbally and in writing, and the patient should sign the instructions to document their receipt and understanding. As with all discharge instructions, the goal is to provide the proper instructions, explain the important issues, and set expectations for the patient to avoid surprises or unexpected outcomes.


SCCS must be considered in all patients presenting to the ED with back pain or leg pain. A careful history and neurological examination, as well as rapid consultation and MRI imaging, are the keys to early intervention and liability prevention. Delays in diagnosis and treatment directly correlate to the severity of complications and the likelihood of claims for medical malpractice.


1. Institute for Clinical Systems Improvement (ICSI).Guideline: Adult low back pain. Revised and updated November 2006. Available at the National Guideline Clearinghouse at

2. Hipp A, Sinert R. Clinical assessment of low back pain. Ann Emerg Med 2006;47:283-285.

3. Deyo RA, Weinstein JN. Low back pain. N Engl J Med 2001;344:363-370. More primary care oriented than emergency medicine oriented, but a nice summary of the literature.

4. Darouiche RO. Spinal epidural abscess. N Engl J Med 2006;355:2012-2020. Nice review article.

5. Shapiro S. Medical realities of cauda equina syndrome secondary to lumbar disc herniation. Spine 2000;25:348-352.

6. Della-Giustina D, Kilcline BA, et al. Back pain: cost-effective strategies for distinguishing between benign and life-threatening causes. Emerg Med Practice 2000;2:1-24.

7. Kostuik JP. Cauda equina syndrome: an uncommon entity and medicolegal controversies. Neurosurg Focus 2004;16:39-41.

8. Rawlings CE. Timing is key for cauda equina syndrome diagnosis in the ED. ED Legal Letter 2005;16:1-12. Nice summary of cauda-equina issues.

9. Deyo RA, Rainville J, et al. What can the history and physical examination tell us about low back pain? JAMA 1992;268:760-765.

10. Agency for Health Care Policy and Research. Acute Low Back Pain in Adults: Clinical Practice Guideline No. 14. AHCPR publication No. 95-0642 (1994).

11. Suarez-Almazor ME, Bleseck E, Russell AS, Mackel JV. Use of lumbar radiographs for the early diagnosis of low back pain. Proposed guidelines would increase utilization. JAMA 1997;277:1782-1786.

12. Liang M, Komaroff AL. Roentgenograms in primary care patients with acute low back pain: a cost-effectiveness analysis. Arch Intern Med 1982;142:1108-1112.

13. Whalen JP, Balter S. Radiation risks associated with diagnostic radiology. Dis Mon 1982;28:1-96.

14. Friedman BW, Holden L, et al. Parenteral corticosteroids for emergency department patients with non-radicular low back pain. J Emerg Med 2006;31:365-370.