Brain

Meningitis: A Normal Scan Doesn't Exclude It

Fellowship-level neuroradiology teaching by E. Brooke Schrickel, MD. Open this topic in the interactive reading room →

Core clinical idea

A normal MRI does not exclude meningitis — it's a CSF diagnosis. The scan's real jobs are catching the few specific signs (non-suppressing CSF on FLAIR, leptomeningeal enhancement), the management-changing complications, and the dangerous mimic (leptomeningeal carcinomatosis).

Bottom line

A normal MRI doesn't exclude meningitis — image for the complications and the mimics, not the diagnosis.

Core workstation questions

  • Is the CSF suppressing on FLAIR — and if not, what besides infection could cause it (SAH, leptomeningeal tumor, supplemental-O2 artifact)?
  • Is there hydrocephalus or early temporal-horn ballooning?
  • Any ependymal enhancement or intraventricular restricted diffusion (ventriculitis)?
  • Cortical/deep restricted diffusion (cerebritis or infarct)?
  • Is the enhancement basal and exudative (TB) or nodular/mass-like with cranial-nerve involvement (carcinomatosis)?
  • Does this patient have a known primary malignancy?

What changes reporting / management

  • Most sensitive sign of leptomeningeal meningitis is loss of CSF suppression on FLAIR, with leptomeningeal enhancement when present.
  • Never report a normal study as excluding meningitis; state that CSF analysis is required.
  • Image primarily to detect complications: hydrocephalus (arachnoid-granulation dysfunction), cerebritis, ventriculitis/intraventricular empyema, and infarcts.
  • TB meningitis: thick exudative basal-cistern enhancement with high hydrocephalus rate; unreliable on DWI (can be positive or negative).
  • Leptomeningeal carcinomatosis mimic: consider in known primary; favor it with nodular/mass-like enhancement, cranial-nerve involvement, and extracranial disease.

Practical traps

  • Letting 'negative MRI' be read as 'no meningitis.'
  • Committing to infectious meningitis when non-suppressing CSF is non-specific (SAH, tumor, O2 artifact).
  • Calling infectious meningitis in an oncology patient instead of considering leptomeningeal carcinomatosis.
  • Trusting DWI in TB the way you would in pyogenic infection.

Teaching pearls

  • MRI cannot rule out meningitis — meningitis is a CSF diagnosis; a negative study excludes nothing.
  • Non-suppressing CSF on FLAIR isn't specific: meningitis, SAH, leptomeningeal tumor, or supplemental-O2 artifact.
  • The scan's real job is the complications: hydrocephalus, cerebritis, ventriculitis, infarct.
  • TB loves the basal cisterns and lies on DWI — don't trust diffusion in TB the way you do in pyogenic infection.
  • Leptomeningeal carcinomatosis mimics meningitis — nodular/mass-like enhancement, cranial nerves, and a known primary should redirect you.

Source lectures

  • Imaging of Meningitis

Educational material for radiology residents and neuroradiology fellows. Nothing here drives individual patient care, and it contains no patient data.