The New Physician shieldThe New PhysicianThe work of Sonny Saggar, MD

The Shit List

The MRI That Tried to Diagnose a Backache

A scanner can see exquisite detail. For an adult with new, uncomplicated low-back pain and no warning signs, it often cannot tell us what to do next.

Editorial illustration representing The MRI That Tried to Diagnose a Backache
Original editorial illustration · New Physician · Editorial illustration, not the patient or event described

Sonny Saggar, MD · 2026-11-16

The Claim

A patient bends to lift a suitcase, feels a terrible stab in the lower back, and asks for an MRI. The wish is understandable: if the pain is real, surely we ought to take a picture of it. Yet a striking image is not the same thing as an actionable diagnosis. The question is not whether an MRI can find something. It is whether finding it now will improve care. [1] [2]

For acute low-back pain without red flags or previous treatment, the American College of Radiology rates initial lumbar MRI, CT and radiography as usually not appropriate. NICE similarly advises against routine imaging in nonspecialist settings, including when sciatica is present but serious disease is not suspected. This is a targeted recommendation, not a rule against imaging. [1] [2]

What The Trials Found

In a systematic review of randomized trials, Chou and colleagues compared immediate spinal imaging with usual care without immediate imaging in people whose back pain showed no evidence of serious underlying disease. Routine imaging did not yield meaningful improvements in pain, function or other clinical outcomes. The review is most applicable to acute or subacute presentations in primary care; it is not a test of whether MRI helps when a neurological emergency is suspected. [3]

Guidelines also emphasize that initial treatment should depend on the presentation, examination and patient goals. For many people, reassurance, tolerable activity and carefully selected non-drug treatments are more useful starting points than a radiology report. [2] [5]

The Incidental Finding Trap

Imaging finds changes that sound alarming even in people who have no back pain. A review of imaging in 3,110 symptom-free people estimated that disc degeneration was present in 37% of people aged 20 and 96% of those aged 80. These estimates came from heterogeneous observational studies; they are not the probability that any particular painful disc is harmless. They are a warning against treating every line of an MRI report as the cause of someone's suffering. [4]

A report describing a bulge, wear or a narrowed space can be clinically important when it matches symptoms and examination. It can also prompt unnecessary worry or additional interventions when it does not. MRI itself uses no ionizing radiation, but unnecessary testing still has costs and downstream consequences. [1] [4]

The Strongest Objection

What if we miss a cancer, fracture, infection or compression of the cauda equina? That objection demands a better history and examination, not an MRI for every sore back. The imaging decision changes sharply when symptoms, findings or risk factors suggest serious disease. A clinician should actively look for those clues rather than rely on a no-imaging slogan. [1] [2]

The Exceptions

Urgent assessment and often emergency imaging are appropriate for new bladder retention or loss of control, saddle numbness, severe or progressive leg weakness, or other findings concerning for cauda equina syndrome. Fever with back pain and infection risk, major trauma, suspected fracture, a history of cancer with concerning symptoms, or other signs suggesting serious disease also justify a different pathway. New or worsening neurological findings deserve prompt clinical evaluation. [1] [2] [6]

Persistent or progressive pain despite an appropriate period of management can justify further evaluation. The ACR allows consideration of imaging after roughly six weeks of suitable management when symptoms have not improved, especially if surgery or an intervention is being contemplated. Decisions must be individualized. [1]

What We Do About It

First, ask three questions: What did the examination show? Are there any red flags? What result would change treatment today? If none points toward urgent pathology, discuss a time-limited plan of staying gently active as tolerated, appropriate pain relief when safe and a specific follow-up point. Bed rest for days is not the default. [2] [5]

Second, explain uncertainty without dismissing pain. The patient deserves clear return precautions, a reassessment plan and help restoring function. If there is a genuine indication for imaging, order the study that answers the clinical question. If there is not, explain why restraint is active medical care rather than indifference.

What Would Change The Verdict

Evidence that immediate routine MRI improves patient-relevant outcomes in a clearly defined, low-risk subgroup would alter the calculation. Until then, the strongest evidence favors a careful examination, safety-netting and purposeful imaging over reflexive scanning.

Sources and receipts

  1. [1] American College of Radiology. ACR Appropriateness Criteria: Low Back Pain, clinical variants and imaging ratings. — Source cited in the reviewed article
  2. [2] National Institute for Health and Care Excellence. Low back pain and sciatica in over 16s: assessment and management, NG59 recommendations. — Source cited in the reviewed article
  3. [3] Chou R et al. Imaging strategies for low-back pain: systematic review and meta-analysis. The Lancet. 2009. — Source cited in the reviewed article
  4. [4] Brinjikji W et al. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR. 2015. — Source cited in the reviewed article
  5. [5] Qaseem A et al. Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain: ACP Clinical Practice Guideline. Annals of Internal Medicine. 2017. — Source cited in the reviewed article
  6. [6] Academy of Medical Royal Colleges, Evidence-Based Interventions. Low back pain imaging: criteria, risks and exceptions. — Source cited in the reviewed article

From the same professional problem-solving perspective: YouTube · Spotify · Substack.

Corrections policy and editorial standards