NeuroRad TractsNeuroradiology education
Search→Residents→Fellowship→Explore Research↗Anatomy Lab↗Neuroradiology Oral Boards→
Brain

Cerebral Amyloid: When the Plaque Isn't the Disease

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

Watch · concise explainerVisual summary

Core clinical idea

The molecular truth and the clinical syndrome are not the same case: amyloid can be deposited without causing dementia, and the disease you're reading may be driven by tau or TDP-43. Read distribution, asymmetry, and hemorrhage pattern — never 'amyloid present = Alzheimer's.'

Bottom line

Dementia is a distribution read — map atrophy + FDG to a syndrome, and never let a positive amyloid scan write your diagnosis.

Core workstation questions

  • Is the atrophy/hypometabolism mesial-temporal+parietal (AD), posterior (PCA), left perisylvian (logopenic PPA), frontal (bvFTD/nonfluent PPA), or out-of-proportion hippocampal + amyloid-negative (LATE)?
  • Is there a peripheral hemorrhagic signature (cortical microhemorrhages, lobar bleed, superficial siderosis) = CAA — and does edema ride with it (CAA-related inflammation)?
  • Is the pattern symmetric or asymmetric, and which side dominates?
  • If DLB is suspected, has a DAT/ioflupane scan been considered instead of leaning on amyloid PET?

What changes reporting / management

  • Asymmetric subcortical edema + peripheral microhemorrhages + siderosis should raise CAA-related inflammation, steering toward immunosuppression rather than a tumor/PRES workup.
  • Name the specific syndrome (PCA, logopenic PPA, LATE) instead of generic 'AD' when distribution/asymmetry supports it — it sets the clinical expectation.
  • Flag out-of-proportion, amyloid-negative hippocampal atrophy as a LATE pattern so it isn't mislabeled end-stage AD.
  • When DLB is in question, the discriminating study is the dopamine transporter (DAT) scan, not amyloid PET.

Practical traps

  • CAA-related inflammation transferred/read as PRES, tumor, or encephalitis.
  • Treating a positive amyloid PET as proof of Alzheimer's — DLB is amyloid-positive in the majority of cases.
  • Reading a normal/near-normal MRI as excluding early AD (MRI atrophy appears late).
  • Equating severe hippocampal volume loss with end-stage AD when the LATE pattern (amyloid-negative, out-of-proportion) fits better.

Teaching pearls

  • Amyloid present is not Alzheimer's present — tau and TDP-43 do the damage.
  • Edema that travels with peripheral microhemorrhages is CAA-related inflammation, not PRES.
  • In DLB the amyloid scan lies; the DAT scan tells the truth.
  • Out-of-proportion, amyloid-negative hippocampal loss in the very old — think LATE.
  • 'No tumor, no stroke' is not a dementia report.

Teaching visuals

Atrophy Pattern → Syndrome Where the cortex thins points to the clinical syndrome. Conceptual shading on an anatomical illustration — not a radiology image. Lateral view — left hemisphere frontal at left · occipital at right Inferior frontalnonfluent / agrammatic PPA Anterior temporal polesemantic variant PPA Parieto-occipitalposterior cortical atrophy Mesial temporal+ temporoparietal / precuneus → typical AD Dominant posterior perisylvian / temporallogopenic primary progressive aphasia Coronal schematic — hippocampal level conceptual shading, not a radiology image midbrain normal atrophic Profound asymmetric hippocampal loss shrunken hippocampus + dilated temporal horn vs spared side → out of proportion, very elderly consider LATE Read alongside the work-up Pattern guides the differential; it does not replace clinical & biomarker work-up. Shading is conceptual — it marks the lobe a syndrome favors, not measured volume. Patterns overlap and are read alongside age & biomarkers.
Atrophy Pattern Maps to Syndrome
Amyloid-positive ≠ Alzheimer dementiaA positive amyloid study is a finding, not a diagnosis — read it in context. Positivity alone does not name the cause of the symptomsA finding to be weighed, never a verdict to be read off in isolation.Amyloid PET = deposition, not ADPlaque can be present without a matchingclinical syndrome. Positivity alone does notname the cause of the symptoms.DLB can be amyloid-positiveDementia with Lewy bodies often carriesco-existing amyloid. A DAT scan maydiscriminate more than amyloid PET here.Profound atrophy + negative amyloidIn the very elderly: out-of-proportion medialtemporal loss with a negative amyloid study→ think LATE rather than typical AD.Synthesize — don’t let one test dominateWeigh the clinical syndrome + the MRI atrophypattern + FDG / DAT + patient age together,as one picture.The diagnosis synthesizesno single biomarker settles itsyndrome+MRI pattern+FDG / DAT+agediagnosisTeaching pointNo single biomarker settles the diagnosis. The pattern + age + the right confirmatorytest, taken together, do.DRAFT
Amyloid-positive is not Alzheimer dementia
CAA vs CAA-related InflammationBoth are peripheral / lobar processes of the elderly. Edema separates the inflammatory variant.CAAcerebral amyloid angiopathyImaging signatureLobar / peripheral microhemorrhages (not deep)Superficial siderosis along peripheral sulciLobar hemorrhage — peripheral blotOn SWIPeripheral signal — NOT deep-hypertensiveDistribution is the discriminator, not severityLegendlobar / peripheral microhemorrhagessuperficial siderosis · lobar hemorrhageCAA-ri / ABRAinflammatory variantImaging signatureAsymmetric cortical/subcortical FLAIR edemaOften posterior+ peripheral microhemorrhages± leptomeningeal enhancementMimicsPRES · tumor · encephalitisEdema is what separates it from bland CAALegendasymmetric cortical/subcortical FLAIR edema+ peripheral microhemorrhages · ± enhancementTeaching linePRES-like asymmetric edema + peripheral microhemorrhages in an older patient → think CAA-related inflammation.CAA is not Alzheimer disease.
CAA vs CAA-related inflammation

Source lectures

  • Amyloid Pathology and Mimics

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