Sorry, you need to enable JavaScript to visit this website.
Skip to main content

Back Again

Jon D. Lurie, MD | March 1, 2008
View more articles from the same authors.

Case Objectives

  • Understand the evidence-based evaluation for patients presenting with low back pain.
  • Identify important "red flags" for serious systemic illness presenting with low back pain.
  • Recognize potential pitfalls in caring for patients with low back pain.

Case & Commentary: Part 1

A 34-year-old man came to the emergency department for evaluation of back pain. He stated that the pain had been present for about 1 week. His review of systems was unremarkable except for an isolated episode of fever, which resolved with acetaminophen. His past medical history was significant for use of heroin and cocaine until 1 year earlier. His medications included methadone and ibuprofen. He had no allergies. Physical examination revealed tachycardia and tenderness in the lumbosacral region. Straight leg raising test was negative. X-ray of the lumbar spine was normal. The patient was discharged home on ibuprofen and advised to follow up with his primary physician the next day.

This case presents a common problem, low back pain (LBP), which is typically benign and self-limited. However, it is occasionally the presenting symptom of serious systemic disease, such as cancer or infection, or a surgical emergency, such as with cauda equina compression.(1) The major challenge is to distinguish the vast majority of patients who have benign musculoskeletal pain from the small minority with a serious, specific disease process requiring timely intervention. Table 1 summarizes the major causes of LBP.

Identifying Back Pain Associated with Systemic Disease

The initial task is to assess the likelihood of a serious underlying systemic disease without over-testing those with benign musculoskeletal pain. History is usually the key to early detection of serious causes of LBP.(2) Suspicion should be particularly high in patients whose pain is unrelieved in any position.(3-5)


The classic "red flags" for malignancy include age older than 50, previous history of cancer, unexplained weight loss (greater than 4.5 kg over 6 months), failure to improve after 1 month of therapy, and no relief with bedrest.(6) Malignancy accounts for less than 1% of cases seeking care for LBP.(7) The only finding specific enough to significantly increase the odds of malignancy is a previous history of cancer (Table 2 summarizes test characteristics for red flags in evaluating LBP). The other red flags, when present, only modestly raise the odds.(2) Pain worse at night or with recumbency, particularly when patients sleep in a chair to avoid pain, is very worrisome for malignancy or infection, though the precise sensitivity and specificity are unknown.(4,8) The most sensitive red flag is no relief with bedrest; when this is absent (i.e., pain is relieved with lying down), it significantly reduces the odds of malignancy. However, this finding is rather nonspecific; most patients who report a lack of relief with recumbency/rest will still have benign backache.(2)

In patients with none of the red flags, the probability of malignancy approaches zero.(2) LBP patients with one or more "red flags" have a pretest probability of serious systemic disease of between 1%–10%.(9) The physical examination is less helpful than the history for identifying malignancy. The majority of spinal malignancies are metastatic from breast, lung, or prostate, so these areas should be carefully evaluated when malignancy is suspected.(10) Focal bony tenderness in the midline is a moderately reliable finding for malignancy and should be explored as well.(11)


Intravenous drug use, urinary tract infection, indwelling urinary catheters, and skin infections raise the likelihood of infective spondylitis.(6) Fever strongly suggests infection but remains an insensitive marker.(4,10) Thus, while the presence of fever should raise strong concern for infective spondylitis, the absence of fever does not significantly lower the odds of infection.(4) This is particularly true when patients are taking acetaminophen or nonsteroidal anti-inflammatories for pain, as these can mask fever.

Compression Fracture

With an aging population and better treatment options for osteoporosis, compression fractures as a cause of LBP are becoming more important to recognize.(1) Compression fractures make up about 4% of LBP cases.(3) Being less than 50 years of age significantly lowers the odds of compression fracture, while being over 70 increases the odds of compression fracture.(2) A history of trauma is not particularly useful, and it does not markedly alter the odds of compression fracture.(2) Corticosteroid use is a fairly specific risk factor for compression fractures; compression fracture needs to be strongly considered in any patient using corticosteroids who presents with LBP.(2)

Spinal Cord Compression Syndromes

Cauda equina and spinal cord compression syndromes are the most important neurological entities in the differential diagnosis of LBP as they represent surgical emergencies. Cord compression can occur in the setting of spinal tumors or epidural abscesses or with massive midline intervertebral disc herniation (IDH). Fortunately, this entity is quite rare, accounting for an estimated 0.04% of LBP cases. Unilateral or bilateral leg pain, numbness, and/or weakness are common, each occurring in over 80% of cases.(10) Urinary retention is fairly sensitive and specific, with a high positive likelihood ratio and low negative likelihood ratio.(10)

In this case, the patient is young with few or no concerning features for malignancy, although we are not told enough about the features of his pain to know if it was mechanical or not. The report of fever is worrisome, and the lack of elevated temperature on examination should not be reassuring, particularly with his report of taking acetaminophen. The frequent lack of objective physical and imaging findings in patients with mechanical back pain complicates the evaluation. In this case, a significant red flag is the history of probable injection drug use (which might be misinterpreted as a red flag for "drug-seeking behavior" rather than as a clue to serious systemic illness).

Case & Commentary: Part 2

The patient did not see his primary physician the next day. Instead, the day after that, he went to another emergency department with complaints of back pain. He was again advised to use ibuprofen and follow up with his primary physician. The patient returned to the hospital again after 4 days with complaints of worsening back pain and new shortness of breath. Examination revealed the presence of bilateral rales, a systolic murmur in the mitral area, and track marks over flexor aspects of both upper extremities.

Standard recommendations for the work-up of patients with "red flags" include a complete blood count, erythrocyte sedimentation rate (ESR), urinalysis, and plain radiography of the spine.(6,9) Plain radiographs of the spine have high specificity for malignancy but are relatively insensitive.(2) Infective spondylitis can be difficult to diagnose with plain radiography (2), particularly early on in the course of disease. Bone scanning has good sensitivity for infection but modest specificity.(2) Magnetic resonance imaging (MRI) has excellent sensitivity and specificity and is the test of choice in patients with a high clinical suspicion for infective spondylitis.

A detailed cost effectiveness analysis of different diagnostic strategies recommended advanced imaging (MRI or bone scan followed by MRI if the bone scan is abnormal) in patients with one or more red flags if they have either a worrisome radiograph (lytic or blastic lesion seen) or an ESR greater than 50. This strategy found 88% of all the findable cases at a cost of $40 per patient, $9525 per case found, and with only 1.6 false positives per 1000 patients.(12) The notable exception was patients with a personal history of cancer: because their pretest probability is relatively high at 10%–15%, and cord compression is a major concern, moving directly to MRI is probably warranted, even without a worrisome radiograph or an elevated ESR.

Case & Commentary: Part 3

Shortly after admission, the patient developed acute respiratory failure requiring intubation. He became hypotensive, and laboratory results were significant for the presence of bandemia, thrombocytopenia, coagulopathy, acute renal insufficiency, and micro- and macrohematuria. He was treated with fluid resuscitation, antibiotics, fresh frozen plasma, and platelets. Despite these efforts, he developed bleeding from his venipuncture sites, his oral cavity, and his rectum, along with refractory hypotension. Aggressive resuscitation efforts, including red cell transfusion and vasopressor therapy, were initiated without success. The patient died of overwhelming shock.

The patient's cultures subsequently grew methicillin-resistant Staphylococcus aureus. Autopsy revealed a 2x1 inch tricuspid valve vegetation, bilateral patchy pneumonias, and multiple bilateral cortical infarcts in the kidneys. The final cause of death was "complications of infective endocarditis."

The incidence of infective endocarditis among intravenous drug users is estimated at 1%–5% per year with Staphylococcus aureus being the most common organism.(13) Clinical data described as helpful in identifying infective endocarditis include fever, anorexia, weight loss, and back pain.(14) Back pain is present (but may not be the chief complaint or a prominent symptom) in up to 43% of cases of endocarditis—in one case series of Staphylococcus aureus endocarditis, back pain was the chief presenting complaint in almost 10%.(15,16)

The pathogenesis of back pain with infective endocarditis is often not known but can include septic embolization, renal or splenic infarction, myalgias/arthralgias related to the inflammatory response, or infective spondylitis with or without epidural abscess.(15,17) While frank infective spondylitis has been reported to be rare in endocarditis (17), in one recent series it was present in 15% of cases.(15) In this case, the patient rapidly progressed to severe systemic infection, which may or may not have started as infective spondylitis.

The major risk factor contributing to this patient's serious systemic illness may have contributed to the missed diagnosis—his injection drug use. Care of patients with substance abuse is challenging. These patients may be regarded as "sociopaths, a burden to society, manipulative, and not intelligent enough to make a choice or decision."(18) As a result, health care teams may assign little priority to the evaluation and treatment of pain in these patients.(18) Effective principles for engaging drug users in health care relationships include a respectful approach to substance users, understanding the medical and behavioral sequelae of addiction, use of multidisciplinary teams, and refraining from moralistic judgments.(19) More careful attention to the patient's history of fever and injection drug use at the first two visits might well have led to a more timely diagnosis.

Maintaining proper vigilance for potentially "dangerous" causes of back pain without performing unnecessary diagnostic work-ups in the large numbers of patients seeking health care for simple back pain is a difficult task. The Clinical Practice Guideline published by the then Agency for Health Care Policy and Research on "Acute Low Back Problems in Adults" (6) continues to be a helpful and important resource with simple algorithms that remain highly relevant aids for physicians faced with decisions of diagnostic triage in the acute setting.

Take-Home Points

  • Patients with mechanical LBP and without red flags do not require extensive diagnostic work-up.
  • Patients with a personal history of cancer, noted red flags on history, an ESR greater than 50, or with a worrisome lesion on lumbar radiographs should receive further evaluation, typically MRI.
  • Fever, injection drug use, urinary tract infection, or recent skin infection are red flags for infection in patients presenting with LBP.
  • Back pain is a common complaint in infective endocarditis, occurring in up to 43% of cases.
  • Presenting a myriad of challenges for health care providers, injection drug users are at high risk for serious infections; therefore, extra care in the evaluation of their complaints is often warranted.

Jon D. Lurie, MD Associate Professor of Medicine and of Community and Family Medicine Dartmouth Medical School

Faculty Disclosure: Dr. Lurie has declared that neither he, nor any immediate member of his family, has a financial arrangement or other relationship with the manufacturers of any commercial products discussed in this continuing medical education activity. In addition, his commentary does not include information regarding investigational or off-label use of pharmaceutical products or medical devices.


1. Lurie JD. What diagnostic tests are useful for low back pain? Best Pract Res Clin Rheumatol. 2005;19:557-575. [go to PubMed]

2. Jarvik JG, Deyo RA. Diagnostic evaluation of low back pain with emphasis on imaging. Ann Intern Med. 2002;137:586-597. [go to PubMed]

3. Atlas SJ, Deyo RA. Evaluating and managing acute low back pain in the primary care setting. J Gen Intern Med. 2001;16:120-131. [go to PubMed]

4. Lurie JD, Gerber PD, Sox HC. A pain in the back. N Engl J Med. 2000;343:723-726. [go to PubMed]

5. Speed C. Low back pain. BMJ. 2004;328:1119-1121. [go to PubMed]

6. Bigos SJ, Bowyer OR, Braen GR, et al; for the Agency for Health Care Policy and Research. Acute low back problems in adults: assessment and treatment. Clin Pract Guidel Quick Ref Guide Clin. December 1994:iii-iv, 1-25. [go to PubMed]

7. Deyo RA, Weinstein JN. Low back pain. N Engl J Med. 2001;344:363-370. [go to PubMed]

8. Borenstein DG. A clinician's approach to acute low back pain. Am J Med. 1997;102:16S-22S. [go to PubMed]

9. Wipf JE, Deyo RA. Low back pain. Med Clin North Am. 1995;79:231-246. [go to PubMed]

10. 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]

11. van den Hoogen HM, Koes BW, van Eijk JT, Bouter LM. On the accuracy of history, physical examination, and erythrocyte sedimentation rate in diagnosing low back pain in general practice. A criteria-based review of the literature. Spine. 1995;20:318-327. [go to PubMed]

12. Joines JD, McNutt RA, Carey TS, Deyo RA, Rouhani R. Finding cancer in primary care outpatients with low back pain: a comparison of diagnostic strategies. J Gen Intern Med. 2001;16:14-23. [go to PubMed]

13. Beynon RP, Bahl VK, Prendergast BD. Infective endocarditis. BMJ. 2006;333:334-339. [go to PubMed]

14. Hermans PE. The clinical manifestations of infective endocarditis. Mayo Clin Proc. 1982;57:15-21. [go to PubMed]

15. Le Moal G, Roblot F, Paccalin M, et al. Clinical and laboratory characteristics of infective endocarditis when associated with spondylodiscitis. Eur J Clin Microbiol Infect Dis. 2002;21:671-675. [go to PubMed]

16. Watanakunakorn C. Staphylococcus aureus endocarditis at a community teaching hospital, 1980 to 1991. An analysis of 106 cases. Arch Intern Med. 1994;154:2330-2335. [go to PubMed]

17. Cone LA, Hirschberg J, Lopes C, et al. Infective endocarditis associated with spondylodiscitis and frequent secondary epidural abscess. Surg Neurol. 2008;69:121-125. [go to PubMed]

18. Hopper JA, Shafi T. Management of the hospitalized injection drug user. Infect Dis Clin North Am. 2002;16:571-587. [go to PubMed]

19. Edlin BR, Kresina TF, Raymond DB, et al. Overcoming barriers to prevention, care, and treatment of hepatitis C in illicit drug users. Clin Infect Dis. 2005;40(suppl 5):S276-S285. [go to PubMed]

20. Hennekens CH, Buring JE. Epidemiology in Medicine. Philadelphia, PA: Lippincott Williams & Wilkins; 1987. ISBN: 9780316356367.


Table 1. Causes of Low Back Pain

Reprinted with permission from Elsevier. In: Lurie JD. What diagnostic tests are useful for low back pain? Best Pract Res Clin Rheumatol. 2005;19:557-575.

Regional Mechanical Low Back Pain (greater than or equal to 90%)
Nonspecific mechanical low back pain (sprain, strain, lumbago, etc.)
Degenerative changes in discs and/or facet joints
Osteoporotic compression fractures
Traumatic fractures
Deformity (severe scoliosis, kyphosis, spondylolisthesis)
Mechanical Low Back Pain with Neurogenic Leg Pain (7%–10%)
Intervertebral disc herniation
Spinal stenosis
Spinal stenosis associated with degenerative spondylolisthesis
Non-Mechanical Spine Disorders (less than or equal to 1%)
Neoplasia (metastases, lymphoid tumors, spinal cord tumors, etc.)
Infection (infective spondylitis, epidural abscess, endocarditis, herpes zoster, lyme)
Seronegative spondyloarthritides (ankylosing spondylitis, psoriatic arthritis, reactive arthritis, Reiter's syndrome, inflammatory bowel disease)
Other: usually present with other accompanying symptoms
Pelvic (prostatitis, endometriosis, pelvic inflammatory disease)
Renal (nephrolithiasis, pyelonephritis, renal papillary necrosis)
Aortic aneurysm
Gastrointestinal (pancreatitis, cholecystitis, peptic ulcer disease)
Paget's disease
Parathyroid disease

Table 2. Test Characteristics Associated with "Red Flags" in the Causes of Back Pain (2,4,10,11)

  Positive LR Negative LR Sensitivity Specificity
History of cancer 15      
No relief with bedrest 1.7 0.21 90%  
Focal bony tenderness     15%–60% 60%–78%
Plain radiographs     60% 95%–99.5%
Fever 25   50% 98%
Plain radiographs     82% 57%
Bone scan     90% 78%
MRI     96% 92%
Compression fracture        
Age greater than 70 5.5      
Age less than 50   0.26    
History of trauma     30% 85%
Corticosteroid use 12      
Spinal cord compression syndromes        
Urinary retention 18 0.1 90% 95%
Definition of Terms (20)
Sensitivity: the probability of testing positive if the disease is truly present; as the sensitivity of a test increases, the number of persons with the disease who are incorrectly classified as test-negative (false negatives) will decrease.
Specificity: the probability of testing negative if the disease is truly absent; a highly specific test will rarely be positive in the absence of disease and will therefore result in a lower proportion of persons without the disease who are incorrectly classified as test-positive (false positives).
Likelihood ratio (LR): the probability that a particular test result will occur in patients with disease can be compared with the probability that the same test result will occur in patients without the disease. (Likelihood ratios can be used to convert pretest odds that a disease is present into posttest odds.)
This project was funded under contract number 75Q80119C00004 from the Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health and Human Services. The authors are solely responsible for this report’s contents, findings, and conclusions, which do not necessarily represent the views of AHRQ. Readers should not interpret any statement in this report as an official position of AHRQ or of the U.S. Department of Health and Human Services. None of the authors has any affiliation or financial involvement that conflicts with the material presented in this report. View AHRQ Disclaimers
Related Resources From the Same Author(s)
Related Resources