UMEM Educational Pearls

A recent meta-analysis of 14 studies looked at the typical red flags of back pain to see which ones actually truly increase the risk that the patient will have a fracture or malignancy.

The typical historical red flags that are taught are

  • Age under 18 or over 50
  • Pain lasting more than 6 weeks
  • History of cancer
  • Fever and chills
  • Night sweats, unexplained weight loss
  • Recent bacterial infection
  • Unremitting pain despite rest and analgesics
  • Night pain
  • Intravenous drug users
  • Immunocompromised
  • Major trauma
  • Minor trauma in the elderly

And physical exam red flags are

  • Fever
  • Writhing in pain
  • Bowel or bladder incontinence
  • Saddle anesthesia
  • Decreased or absent anal sphincter tone
  • Perianal or perineal sensory loss
  • Severe or progressive neurologic defect
  • Major motor weakness

However, this meta-analysis showed that the only red flags that actually increased the risk of fracture or malignancy were

  • Older Age  Post test Probability 9% (95% CI 3% to 25%)
  • Prolonged corticosteroid use Post test Probability 33% (95% CI 10% to 67%)
  • Severe trauma Post test Probability 11% (95% CI 8 % to 16%)
  • Presence of contusion or abrasion Post test Probability 62% (95% CI 49% to 74%)

So this study highlights that a lot of the red flags that we have learned do not actually increase the risk fracture or malignancy, although some like fever, IVDA, and immunocomproromised increase the risk of epidural abscesses, which was not addressed in this meta-analysis.

The take home point for me is that plain radiographs/CT scans are probably only needed in patients with older age, prolonged corticosteroid use, severe trauma or presence o contusion or abrasion. If you are really worried about others with back pain just proceed directly to MRI as the plain films/CT scans are not going to be very helpful.


Downie A, Williams CM, Henschke N, et al. Red flags to screen for malignancy and fracture in patients with low back pain. British Journal of Sports Medicine. 2014;48(20):1518–1518. doi:10.1136/bjsports-2014-f7095rep.