Neurology
Report template: Brain perfusion SPECT (HMPAO / ECD) · Dopaminergic imaging (DaT SPECT, F-18 DOPA PET) · Amyloid PET
This chapter's English text is a draft translation (prepared with Claude and checked against a terminology glossary and automatic checks); specialist review is in progress. Please report errors with the feedback button.
Solve questions (26) →Contents
- 19.1 Cerebral Perfusion SPECT (Tc-99m HMPAO, Tc-99m ECD)
- 19.2 FDG PET Brain Imaging
- 19.3 Assessment of Alzheimer Disease and Dementia
- 19.4 Amyloid PET (Florbetapir, Florbetaben, Flutemetamol)
- 19.5 Tau PET (Flortaucipir)
- 19.6 Dopamine Transporter Imaging (DaTscan – I-123 Ioflupane)
- 19.7 Assessment of Epilepsy (Ictal/Interictal SPECT)
- 19.8 Brain Death Scintigraphy
- 19.9 CSF Flow Scintigraphy (Cisternography)
- Key Points for the Board
- Current guidelines on this chapter's topics
Learning objectives and modules4 modules · about 3 h 20 min
Learning objectives
After this chapter you will be able to:
- Compare brain perfusion tracers (HMPAO, ECD) in terms of uptake mechanism and properties.
- Differentiate types of dementia from FDG PET and perfusion SPECT patterns.
- Explain the indications, tracers and interpretation criteria of amyloid and tau PET.
- Interpret dopamine transporter imaging and its place in the differential diagnosis of parkinsonism.
- Explain the principle and timing of ictal and interictal imaging in epilepsy.
- Interpret the technique and findings of brain death scintigraphy and cisternography.
Modules
The chapter is divided into study modules without changing the order of the text: read, watch the animations, then answer the questions linked to those headings.
- Module 119.1–19.2 Cerebral Perfusion SPECT (Tc-99m HMPAO, Tc-99m ECD) · FDG PET Brain Imaging55 min · 1 animation · 22 questionsAnswer the questions →
- Module 219.3–19.5 Assessment of Alzheimer Disease and Dementia · Amyloid PET (Florbetapir, Florbetaben, Flutemetamol) · Tau PET (Flortaucipir)55 min · 25 questionsAnswer the questions →
- Module 319.6–19.7 Dopamine Transporter Imaging (DaTscan – I-123 Ioflupane) · Assessment of Epilepsy (Ictal/Interictal SPECT)50 min · 2 animations · 19 questionsAnswer the questions →
- Module 419.8–19.9 Brain Death Scintigraphy · CSF Flow Scintigraphy (Cisternography) · Key Points for the Board40 min · 14 questionsAnswer the questions →
Times are estimates: 130 words per minute, 3 min per animation, 1.5 min per question.
MOST FREQUENTLY ASKED IN THIS CHAPTER
✓
HMPAO → trapping by glutathione, 30 min stable (without stabiliser), shows luxury perfusion
✓
ECD → trapping by esterases, 6 hours stable, may not show luxury perfusion
✓
Both cross the BBB → assessment of parenchymal perfusion
✓
"Freezes" the perfusion at the moment of injection → snapshot property
✓
Crossed cerebellar diaschisis: reduced perfusion/metabolism in the cerebellar hemisphere contralateral to a supratentorial infarct/tumour (cerebellum normal on CT/MR)
✓
Acetazolamide (1 g IV) reserve test: an area with exhausted reserve falls relative to baseline; contraindications are sulphonamide allergy and stroke within the last 3 days
✓
Most intense uptake: Cerebral cortex, basal ganglia, thalamus
19.1 Cerebral Perfusion SPECT (Tc-99m HMPAO, Tc-99m ECD)15 questions
General Principle:
Cerebral perfusion SPECT is an imaging method that assesses regional cerebral blood flow (rCBF). After IV injection the radiopharmaceutical crosses the blood-brain barrier (BBB) and is retained in brain tissue. The amount of uptake is proportional to regional blood flow.
•
Perfusion agents are small, neutral, lipophilic molecules: they cross the BBB rapidly, are extracted at a high rate on first pass and are retained in the brain. The activity reaching the brain is ~4-6% for HMPAO and ~5-7% for ECD (peak at ~2 minutes)
•
BBB: Tight junctions between endothelial cells restrict the passage of ionic and hydrophilic molecules; lipophilic molecules cross by passive diffusion, but the P-glycoprotein efflux pump and binding to plasma proteins limit this passage
•
Normal CBF in adults is ~50 mL/100 g/min, and in children aged 3-10 years it is ~2 times this; the grey/white matter uptake ratio with HMPAO is ~2.5:1
•
Maturation in childhood: In the newborn uptake is greatest in the sensorimotor cortex, thalamus, brainstem and cerebellar vermis; frontal cortical uptake increases at 6-12 months and the adult pattern is seen at ~1 year of age. Absolute glucose metabolism is above adult values between ~3-9 years of age
Tc-99m HMPAO (Hexamethylpropyleneamine Oxime = Exametazime):
•
Lipophilic complex → crosses the BBB by passive diffusion
•
After entering the brain cell it reacts with glutathione → converted to a hydrophilic form → cannot leave the cell → metabolic trapping
•
First-pass extraction: 70-80%
•
Brain uptake: Fixed within 30-60 seconds after injection
•
Redistribution: None → "freezes" the perfusion state at the moment of injection (snapshot)
•
Stability: Must be used within 30 minutes of preparation (unstable) → can be extended to 4 hours with a stabiliser (methylene blue or cobalt chloride)
•
Dose: 20-25 mCi (740-925 MBq)
•
Imaging: 15-20 minutes after injection (after brain fixation is complete)
BOARD TIP
▸
"Why is HMPAO trapped in the cell?"
→ Reaction with glutathione → conversion from lipophilic to hydrophilic form → cannot leave the cell.
▸
"Most important disadvantage of HMPAO?"
→ Unstable → must be used within 30 min of preparation.
▸
"Agent property for ictal SPECT?"
→ Both HMPAO and ECD show rapid brain fixation (30-60 s) → if injected during a seizure they reflect perfusion at that moment; because ECD is stable for ~6 hours after preparation, it can be kept ready in a syringe while awaiting a seizure (often preferred in practice).
Tc-99m ECD (Ethyl Cysteinate Dimer = Bicisate):
•
Lipophilic complex → crosses the BBB by passive diffusion
•
De-esterification by esterase enzymes in the brain cell → hydrophilic metabolite → trapped in the cell
•
First-pass extraction: 60-70% (slightly lower than HMPAO)
•
Brain uptake: Fixed at a similar speed to HMPAO
•
Redistribution: Minimal (slightly more back-diffusion than HMPAO)
•
Stability: Stable for up to 6 hours after preparation → much more stable than HMPAO
•
Dose: 20-25 mCi (740-925 MBq)
•
Extracerebral activity: Lower than HMPAO → better brain/background ratio
•
ECD clears rapidly from the blood and is excreted by the kidneys; the critical organ is the bladder wall. This property allows early imaging
•
Feature: Grey matter uptake more pronounced, white matter uptake low
BOARD TIP
▸
"Why is ECD trapped in the cell?"
→ De-esterification by esterases → hydrophilic form.
▸
"ECD vs HMPAO stability?"
→ ECD stable 6 hours, HMPAO 30 minutes (without stabiliser).
▸
"Which is preferred in epilepsy?"
→ Both are used; the 6-hour stability of ECD is an advantage for keeping it ready for ictal injection (HMPAO only ~4 hours with a stabiliser).
Left: a lipophilic perfusion agent crossing the brain capillary endothelium and being trapped in a neuron as a hydrophilic form, while hydrophilic DTPA stops at the tight junction. Right: HMPAO-ECD comparison and flow, blood activity and brain uptake curves.
Continuous stream · no seeking
Lipophilic HMPAO and ECD cross the BBB and are trapped in the cell as a hydrophilic form; with no redistribution the image freezes perfusion at the moment of injection.
Source: extraction, trapping and stability from book 19.1; BBB structure from StatPearls (Blood Brain Barrier) and OpenStax A&P 2e §13.3. Molecule counts and flow curve representative.
How to use: a single click on the drawing pauses and resumes; in the second step ↑ ↓ or the buttons below switch the tracer; ← → change steps.
Did this animation help you understand the topic?
Figure 19.1: Passage of Tc-99m HMPAO and ECD across the blood-brain barrier and their trapping mechanisms in the brain cell (schematic).
HMPAO vs ECD Comparison
| Feature | Tc-99m HMPAO | Tc-99m ECD |
|---|---|---|
| Trapping mechanism | Glutathione (hydrophilic conversion) | Esterase (de-esterification) |
| First-pass extraction | 70-80% | 60-70% |
| Stability | 30 min (without stabiliser) | 6 hours |
| Redistribution | None | Minimal |
| Brain/background | Good | Better |
| Luxury perfusion | Shows | May not show |
| Additional use | Leucocyte labelling | — |
Luxury Perfusion:
•
In a subacute infarct there is reperfusion hyperaemia; flow exceeds metabolic demand (flow-metabolism uncoupling). Because HMPAO is retained in proportion to flow, this area appears to have increased perfusion (HMPAO crosses the BBB freely anyway)
•
ECD may not be retained in this situation (esterase activity reduced → no trapping)
•
HMPAO shows luxury perfusion, ECD may not
•
In acute infarction the perfusion defect is seen immediately and is larger than the lesion on CT or MR because of the surrounding ischaemic penumbra. Lacunar infarcts are small and often cannot be detected; a normal SPECT therefore does not exclude multi-infarct dementia
•
In acute stroke, reduced but partly preserved (residual) perfusion indicates salvageable tissue and a good response to reperfusion; complete absence of uptake is associated with a poor prognosis and a risk of haemorrhagic transformation with thrombolysis
•
Crossed cerebellar diaschisis: Reduction of perfusion and metabolism in the cerebellar hemisphere contralateral to a supratentorial infarct, tumour or trauma, due to loss of corticopontocerebellar connections; diaschisis may also be seen in the ipsilateral thalamus
•
Crossed cerebellar diaschisis is common in large supratentorial infarcts and is associated with a worse clinical prognosis; the cerebellum appears normal on CT and MR
•
O-15 PET: In early ischaemia CBF falls and OEF rises (misery perfusion; salvageable tissue). In an established infarct CMRO2 and CMRglc fall, and OEF also falls. In luxury perfusion OEF falls because flow exceeds demand
•
rCBF thresholds (normal ~50 mL/100 g/min): below ~20-22 electrical function is lost; between ~10-12 and 20-22 is the penumbra, which can recover with reperfusion; below ~10-12 is infarction; a mild reduction above these thresholds is oligaemia
•
AVM: Because the arteriovenous shunt bypasses the capillary bed and there is no extraction, the nidus appears photopenic with HMPAO or ECD, although true flow is high; steal-related hypoperfusion around the nidus indicates a risk of ischaemia
•
The gold standard for absolute rCBF measurement is O-15 water PET (Xe-133 was used formerly); HMPAO and ECD SPECT provide semi-quantitative assessment in routine practice. Cerebral blood volume is measured with C-11 or O-15 labelled CO
•
Balloon test occlusion: Before sacrifice of the ICA (ligation or embolisation), HMPAO or ECD is injected while the occlusion is in place; ipsilateral hypoperfusion indicates inadequate collateral circulation and a risk of ischaemia
Clinical Indications:
•
Assessment of epilepsy (ictal/interictal SPECT) → the most important indication
•
Cerebrovascular disease (stroke, TIA)
–
Acetazolamide (Diamox) reserve test: 1 g is given slowly IV (~15 mg/kg in children) and the radiopharmaceutical is injected 15-20 min later. Normally CBF increases markedly; an area with exhausted reserve appears relatively reduced compared with the baseline study. This indicates a high stroke risk and that the patient is a candidate for revascularisation
–
Indications for the acetazolamide test: Carotid stenosis and TIA (revascularisation decision), moyamoya, steal around an AVM; distinguishing vascular from neuronal causes in dementia (defects in AD respond normally to acetazolamide)
–
Contraindications to the acetazolamide test: Sulphonamide allergy, stroke within the last 3 days, renal or hepatic failure. Side effects are perioral and digital paraesthesia, dizziness and facial flushing; an alternative vasodilatory stimulus is CO2 inhalation
–
TIA: The sensitivity of resting SPECT is highest in the first 24 hours (~60%) and decreases over time; acetazolamide increases sensitivity. CT and MR are mostly normal; it can also be used to monitor treatment
•
Assessment of dementia (an alternative to FDG PET)
•
Confirmation of brain death
•
Neuropsychiatric disorders
EANM Protocol Information:
•
Activity: For ECD and HMPAO 740 MBq in adults (child at least 100 MBq); effective dose 7.7 for ECD and 9.3 µSv/MBq for HMPAO, i.e. 5.7-6.9 mSv
•
HMPAO uptake is more closely related to cerebral blood flow and ECD uptake to metabolic activity; they can therefore be used interchangeably in dementia, but this is not always possible in epilepsy and cerebrovascular disease
•
Regional differences: ECD is relatively more retained in the occipital and parietal cortex, HMPAO in the cerebellum, thalamus, basal ganglia and medial temporal lobe; normal databases and follow-up studies should therefore use the same agent
•
Injection environment: A quiet, dimly lit room; after IV access is established, the patient should not talk or read from 10-15 min before injection until 5-10 min after it
•
Visual or auditory stimulation during injection increases uptake in the corresponding cortex (with eyes open occipital flow increases by ~30%); because the agent fixes the distribution at that moment, this effect cannot be corrected afterwards
•
If sedation is needed, it is given at least 5 min after injection; smoking, alcohol, caffeine and xanthines are avoided
•
Time to start imaging: 30-60 min after injection for ECD, 30-90 min for HMPAO; imaging should be completed within 4 hours
•
Acquisition: Multi-detector camera, highest-resolution collimator, at least 5 million counts, rotation radius <15 cm, matrix at least 128×128, angular sampling <3° and 360° rotation; OSEM or MLEM reconstruction and attenuation correction are recommended
•
Drug effects: Phenytoin causes hypoperfusion in the cerebellum, benzodiazepines in the frontal cortex
•
Sensitivity: The sensitivity of ictal SPECT in TLE is 84% (IAEA) and 85-90% in children; in acute stroke sensitivity is 61-74% and specificity 88-98%
•
Contraindications: Inability of the patient to cooperate (absolute); pregnancy (relative, depending on benefit-risk assessment); breastfeeding need not be interrupted
•
Report content (EANM 2009): in addition to the common items (see 27), medication and MRI/CT findings; injection conditions (quiet, dimly lit room, eyes open or closed) and, in ictal studies, the timing of injection relative to seizure onset (video-EEG); if quantitative analysis was used alongside visual assessment, the method and normal database. Findings: regions of altered perfusion, any asymmetry and the pattern (e.g. temporoparietal, frontal, ictal hyperperfusion), correlation with MRI
19.2 FDG PET Brain Imaging7 questions
Principle:
•
The brain is the most glucose-consuming organ in the body → uses 20-25% of basal metabolism
•
FDG → enters neuronal cells via GLUT-1 and GLUT-3 → FDG-6-phosphate by hexokinase → metabolic trapping
•
FDG uptake is proportional to neuronal activity and synaptic density
•
Grey matter >> white matter uptake (higher neuronal density in grey matter)
Normal FDG Distribution:
•
Most intense uptake: Cerebral cortex (especially frontal, parietal, occipital), basal ganglia, thalamus, cerebellar cortex
•
Low uptake: White matter, periventricular region
•
Symmetric uptake is expected (asymmetry may be pathological)
•
Normal ageing: The decline is most marked in the frontal cortex (prefrontal and anterior cingulate); the primary sensory and visual cortex, basal ganglia, thalamus and cerebellum are relatively preserved
Patient Preparation:
•
Fasting for at least 4-6 hours
•
Blood glucose is measured before injection: a level >110 mg/dL progressively reduces brain FDG uptake (competition with glucose at the transporters and hexokinase). Best results in diabetics are obtained when euglycaemic under proper treatment; a diagnostically acceptable scan can be obtained at 160-180 mg/dL. Rapid correction of hyperglycaemia with insulin does not substantially improve brain image quality (SNMMI/EANM 2.0)
•
Quiet, dimly lit room; the patient awake with eyes open (closing the eyes lowers occipital metabolism and can mimic the occipital hypometabolism of DLB); the same condition is kept within a centre and with the normal database. IV access is placed at least 10 min before injection; the patient does not talk or read, and interaction is minimised for at least 30 min after injection (SNMMI/EANM 2.0)
•
Static acquisition starts 30-40 minutes after injection for neurological diagnosis and 60 minutes (or dual-time-point delayed) for neuro-oncology, never earlier than 30 min; it usually lasts 10-15 min (SNMMI/EANM 2.0)
•
Sedatives/anxiolytics → should not be given before injection (alter brain metabolism)
•
A sedative may be given after injection (FDG distribution already fixed)
•
SNMMI/EANM 2.0: conscious sedation (e.g. IV midazolam) is given at least 30-40 min after injection, preferably a few minutes before acquisition; it is avoided in quantitative dynamic studies; pulse oximetry monitoring, reduced dose in the elderly. In autoimmune encephalitis and primary CNS lymphoma, scan before starting corticosteroids where possible. In epilepsy the patient is monitored from arrival so that FDG is not given in the ictal or postictal phase
•
EANM: Adult 150-250 MBq (child at least 14 MBq); 19 µSv/MBq, i.e. 2.9-4.8 mSv. During the uptake period of about 20 minutes the patient should be kept at psychosensory rest
•
SNMMI/EANM 2.0: earlier guidelines recommended 125-740 MBq (typically 150 MBq) for 3D acquisition; on high-sensitivity digital (TOF <400 ps) or long axial FOV systems the activity can probably be halved or reduced further. In children the EANM dosage card (14 × multiple MBq); in the USA 1.85-3.7 MBq/kg (minimum 14, maximum 148 MBq). Visual reading is the first step; voxel-based analysis is used with a normal database acquired on the same camera and under the same conditions (eyes open/closed)
•
The FDG signal largely reflects glutamate-stimulated astrocytic glucose uptake and synaptic activity
BOARD TIP
▸
"Where is uptake most intense on brain FDG PET?"
→ Cerebral cortex, basal ganglia, thalamus.
▸
"When can a sedative be given?"
→ After injection (brain FDG uptake occurs largely in the first 20-30 min).
▸
"Difference between brain FDG PET and oncological FDG PET?"
→ Brain PET uses a longer acquisition time, thinner slices and different reconstruction parameters.
Clinical Indications:
•
Assessment of dementia (the most important indication) → distinguishing Alzheimer, FTD, DLB
•
Localisation of the epileptic focus (interictal hypometabolism)
•
Grading of brain tumours and assessment of recurrence
•
Tl-201 SPECT (historical, if PET is unavailable): Uptake correlates with glioma grade; recurrence takes up Tl-201, radionecrosis does not. In a patient with AIDS, an intracranial lesion that takes up Tl-201 (and FDG) suggests lymphoma, and one that does not suggests toxoplasmosis
•
Tc-99m MIBI can also be used in brain tumours (tumour/background ratio higher than Tl-201); its disadvantage is that physiological uptake in the choroid plexus may obscure adjacent lesions (this is not marked with Tl-201)
•
Encephalitis (especially autoimmune encephalitis)
•
Neurodegenerative diseases
•
In distinguishing Parkinson disease from atypical parkinsonism (MSA, PSP, CBD), the AUC of FDG PET is 0.94 and that of I-123 IBZM SPECT 0.74; sensitivity and specificity 86% and 91%
•
Used in autoimmune encephalitis (especially if MR is negative); if a paraneoplastic cause is being sought, whole-body imaging can be performed in the same session
•
HSV encephalitis: In the acute phase hyperperfusion and hypermetabolism are seen in the affected temporal lobe; this finding may be detected before CT findings
•
The acute hyperperfusion of HSV encephalitis turns into hypoperfusion within weeks; HMPAO uptake may be increased while ECD uptake is normal or reduced (similar to the dissociation in luxury perfusion)
•
Accuracy in the differential diagnosis of AD and FTLD is 87-89%
•
Voxel-based analysis (SPM, 3D-SSP): The image is normalised to stereotactic space and compared voxel by voxel with an age-appropriate normal database (Z-score map); it increases the accuracy of visual reading and inter-reader agreement
•
Sources of error (SNMMI/EANM 2.0): unintended cerebral activation (visual or motor), movement (during acquisition or between PET and CT/MRI) and MR-based attenuation correction biases (missing bone, air cavities, metal), psychotropic drugs or corticosteroids, sedation, incomplete intravenous injection, absent or insufficient attenuation correction, soft-tissue or skull uptake after surgery of the skull or brain, recent radiotherapy or chemotherapy
–
Preparation: intravenous access is placed at least 10 minutes before injection; the patient does not speak, read or make major movements; interaction with the patient is minimised for at least 30 minutes after injection. With hyperglycaemia the study is postponed until acceptable glycaemia is reached; rapid correction with insulin usually does not substantially improve brain image quality. Prolonged or extreme fasting also reduces glucose metabolism
–
Before epilepsy surgery, the patient is monitored from arrival in the department so that FDG is not given in the ictal or postictal state; the seizure history is documented
–
At interpretation, movement and attenuation artefacts are checked first; morphological changes such as cortical and ventricular enlargement are taken into account and fusion with MRI or CT is performed. Increased uptake may be seen in active epileptogenic foci, tumours, inflammation and activated brain areas
•
Report content (SNMMI/EANM 2.0, 2024): in addition to the common items (see 27), blood glucose, uptake interval, conditions at injection and during uptake (quiet, dimly lit room, eyes open), sedation if used with the drug and its timing relative to FDG, EEG recording in epilepsy and the sequences acquired on PET/MR. Findings: site and severity of hypometabolism (anatomical and functional topography), the normal database used for quantitative values and comparison with MRI. If a generally accepted disease pattern is present, the conclusion gives the most probable diagnosis in the clinical context; interpretation not based on such criteria is stated as hypothetical; limitations (unintended cortical activation, motion, psychotropic drugs or corticosteroids, sedation, incomplete injection, postoperative change, recent radiotherapy or chemotherapy) are listed
19.3 Assessment of Alzheimer Disease and Dementia14 questions
Alzheimer Disease (AD):
•
The commonest cause of dementia (60-80%)
•
Pathology: Extracellular amyloid plaques (Aβ) + intracellular neurofibrillary tangles (tau)
•
FDG PET pattern: Bilateral temporoparietal and posterior cingulate/precuneus hypometabolism
•
In advanced disease: The frontal cortex is also involved
•
Spared areas: Primary motor cortex, primary visual cortex, basal ganglia, thalamus, cerebellar cortex
•
Sensitivity: 90-95%, specificity: 70-80% (in distinguishing from other dementias)
•
Perfusion SPECT shows a posterior temporoparietal pattern in AD similar to FDG PET, but with lower accuracy (sensitivity ~60-75%, positive predictive value ~80%)
•
Cholinergic deficit: In AD there is neuronal loss in the nucleus basalis of Meynert and a reduction in cortical nicotinic (α4β2) receptors; acetylcholinesterase PET (C-11 MP4A/PMP) shows reduced cortical activity
BOARD TIP
▸
"FDG PET pattern in Alzheimer disease?"
→ Bilateral temporoparietal + posterior cingulate/precuneus hypometabolism.
▸
"Spared areas?"
→ Primary motor cortex, primary visual cortex, basal ganglia, cerebellum.
▸
"Earliest region involved?"
→ Posterior cingulate/precuneus → the earliest sign of Alzheimer disease.
Frontotemporal Dementia (FTD):
•
After Alzheimer disease, the commonest cause of dementia under 65 years of age
•
Behavioural variant (bvFTD): Frontal lobe (especially orbitofrontal and medial frontal) + anterior temporal hypometabolism
•
Primary progressive aphasia (PPA):
•
Non-fluent/agrammatic: Left inferior frontal (Broca's area) hypometabolism
•
Semantic: Left anterior temporal hypometabolism
•
Logopenic: Left temporoparietal hypometabolism (may resemble Alzheimer disease)
•
FTD vs Alzheimer: FTD → frontal dominant, Alzheimer → posterior dominant (temporoparietal)
Dementia with Lewy Bodies (DLB):
•
FDG PET: Temporoparietal hypometabolism similar to Alzheimer disease + occipital cortex involvement (different from Alzheimer!)
•
Occipital hypometabolism → the distinguishing feature of DLB
•
Additional diagnostic tools: DaTscan (reduced), cardiac MIBG (reduced) → see Chapter 14.4.3
Vascular Dementia:
•
FDG PET: Multiple, asymmetric, scattered hypometabolic areas → conforming to vascular territories
•
Subcortical involvement may be prominent
•
The typical temporoparietal pattern of Alzheimer disease is not seen
Other FDG patterns according to EANM:
•
DLB: Uptake is low in the temporal, parietal and occipital cortex; the basal ganglia and thalamus are normal or increased (preservation of the posterior cingulate cortex produces the "cingulate island" sign)
•
PSP: Uptake is low in the frontal cortex and midbrain (mesencephalon)
•
CBD: Marked asymmetry in the frontal and parietal cortex and the basal ganglia-thalamus
•
MSA-P: Uptake low in the basal ganglia (putamen) and thalamus, MSA-C: low in the cerebellum
•
Schizophrenia: Frontal hypoperfusion and hypometabolism (hypofrontality)
•
OCD: Hypermetabolism in the orbitofrontal cortex and caudate, decreasing with treatment. Major depression: Hypometabolism in the dorsolateral prefrontal cortex and anterior cingulate; correlates with severity and normalises with treatment
•
In schizophrenia hypofrontality is especially marked in chronic patients with negative symptoms (psychomotor poverty); neuroleptic treatment increases metabolism in the basal ganglia (striatum)
•
Addiction: Chronic alcoholism shows predominantly frontal hypometabolism, acute alcohol intake a global reduction in metabolism; chronic cocaine use shows multifocal hypoperfusion in the prefrontal and orbitofrontal cortex and a reduction in striatal D2 receptors
•
Idiopathic PD: On FDG the lentiform nucleus is normal or relatively increased (PD-related pattern); this distinguishes it from the striatal reduction in MSA and PSP
•
Huntington disease: Hypoperfusion and hypometabolism in the caudate (and putamen); seen before caudate atrophy develops on CT. HIV dementia: Patchy cortical and subcortical hypoperfusion, marked subcortical hypometabolism
•
Wilson disease: Hypometabolism in the lentiform nucleus (especially putamen) and cerebellum. When dementia is added to PD, temporoparietal (± occipital) hypometabolism similar to AD and DLB is seen
FDG PET Comparison of Dementia Types:
| Dementia Type | FDG PET Pattern | Distinguishing Feature |
|---|---|---|
| Alzheimer | Bilateral temporoparietal + posterior cingulate cortex | Posterior dominant, motor/visual spared |
| FTD (behavioural) | Frontal + anterior temporal | Frontal dominant |
| DLB | Temporoparietal + occipital | Occipital involvement (distinguishes from Alzheimer) |
| Vascular | Multiple, asymmetric, scattered | Conforms to vascular territories |
| Corticobasal degeneration | Asymmetric frontoparietal | Marked asymmetry |
| Progressive supranuclear palsy | Frontal + mesencephalon | Mesencephalic hypometabolism |
BOARD TIP
▸
"Difference between DLB and Alzheimer on FDG PET?"
→ In DLB the occipital cortex is involved, in Alzheimer the occipital cortex is spared.
▸
"FTD vs Alzheimer?"
→ FTD frontal dominant, Alzheimer posterior dominant.
▸
"If posterior cingulate/precuneus hypometabolism is asked in the exam?"
→ Think of Alzheimer disease (the earliest region involved).
Figure 19.2: FDG PET hypometabolism patterns in dementia types, lateral and medial views (schematic).
Did this figure help you understand the topic?
19.4 Amyloid PET (Florbetapir, Florbetaben, Flutemetamol)8 questions
Principle:
•
Amyloid PET images fibrillar beta-amyloid (Aβ) plaques in the brain
•
Amyloid plaques are an early sign of Alzheimer pathology → begin to accumulate 15-20 years before clinical symptoms
•
Amyloid PET shows the presence or absence of amyloid pathology → supports or excludes a diagnosis of Alzheimer disease
•
C-11 PiB (Pittsburgh compound B, a thioflavin-T analogue) is the prototype amyloid PET agent; because the half-life of C-11 is short (20 min), an on-site cyclotron is needed, so F-18 labelled agents are used in clinical practice
FDA-Approved Amyloid PET Agents:
1. F-18 Florbetapir (Amyvid):
–
T½: 110 minutes (F-18)
–
Injection-to-imaging interval: 30-50 minutes
–
Imaging time: 10 minutes
–
White matter uptake: Moderate non-specific uptake
2. F-18 Florbetaben (Neuraceq):
–
T½: 110 minutes
–
Injection-to-imaging interval: 45-130 minutes (wide window)
–
Imaging time: 15-20 minutes
–
White matter uptake: Similar to florbetapir
3. F-18 Flutemetamol (Vizamyl):
–
T½: 110 minutes
–
Injection-to-imaging interval: 60-120 minutes (90 minutes in the guideline)
–
Imaging time: 10-20 minutes (20 minutes in the guideline)
–
White matter uptake: More pronounced → may make interpretation more difficult
| Feature | Florbetapir (Amyvid) | Florbetaben (Neuraceq) | Flutemetamol (Vizamyl) |
|---|---|---|---|
| Waiting time | 30-50 min | 45-130 min | 60-120 min (guideline: 90 min) |
| Acquisition time | 10 min | 15-20 min | 10-20 min (guideline: 20 min) |
| White matter uptake | Moderate | Moderate | High |
| Colour scale | Grey-white | Grey-white | Colour (orange-yellow) |
Image Interpretation:
•
Negative (no amyloid): No or minimal uptake in cortical grey matter → a clear boundary between white matter uptake and grey matter is preserved → grey-white matter contrast is distinct
•
Positive (amyloid present): Widespread uptake in cortical grey matter → grey-white matter contrast lost → the boundary between cortex and white matter is blurred
•
Assessment: Dichotomous (positive or negative) → not quantitative
–
SNMMI/EANM 2016: because the cerebellar cortex is spared in most amyloid-positive patients too, it is the reference for visual and semi-quantitative assessment; interpretation begins by confirming grey-white differentiation in the cerebellum. The display scale differs by agent (brightest brain region for florbetapir, white matter for florbetaben, pons at 90% for flutemetamol); florbetaben and florbetapir are read in grey or inverse grey scale, flutemetamol with the manufacturer's colour scale
–
Pitfalls: atrophy can overestimate uptake in the remaining cortex through spill-over from white matter and cause false positives; in severe atrophy, a thin positive cortical ribbon cannot be separated from white matter and the scan looks normal (false negative). Noise, head motion (rescan if needed), low counts from dose infiltration, skull bone marrow uptake and poor head positioning impair interpretation. Encephalomalacia from prior stroke, surgery or trauma is correlated with MRI or CT. A positive scan alone does not mean Alzheimer's disease (it may be positive in dementia with Lewy bodies and in cognitively normal elderly people); the report should not state that "the scan is diagnostic of Alzheimer's disease"
Figure 19.3: Negative and positive appearance on amyloid PET (grey-white matter contrast) and imaging timing of the F-18 agents (schematic).
Did this figure help you understand the topic?
•
Activities recommended by EANM: Florbetapir 370 MBq, flutemetamol 185 MBq, florbetaben 300 MBq; 19 µSv/MBq, i.e. 3.5-7 mSv
•
Sensitivity for showing or excluding the presence of neuritic plaques is ~94% and specificity ~97%
•
Pregnancy is a relative contraindication (depending on benefit-risk assessment); breastfeeding is interrupted for 12 hours
–
Preparation (SNMMI/EANM 2016): no interactions with dementia drugs such as donepezil, galantamine or memantine are known, so no drug is withdrawn. If sedation is needed, the tracer is injected before sedation. Because the tracer can adhere to the catheter, a short (≤4 cm) IV catheter is used and flushed with at least 5-15 mL of saline; the whole brain, including the cerebellum, is in the field of view. The guideline recommends either avoiding the scan in a breastfeeding mother or interrupting breastfeeding for 24 hours (the 12 hours above is a shorter recommendation)
Clinical Use and Appropriate Use Criteria (Alzheimer's Association and SNMMI, 2025):
Situations where amyloid PET is appropriate:
•
Mild cognitive impairment (MCI) or dementia in patients younger than 65 in whom Alzheimer's pathology is suspected
•
MCI or dementia with onset at 65 or older that is consistent with Alzheimer's pathology (amnestic presentation)
•
MCI or dementia that could be consistent with Alzheimer's pathology but has atypical features (non-amnestic presentation, rapid or slow progression, mixed aetiology)
•
MCI or dementia with equivocal or inconclusive recent CSF biomarkers
•
To inform prognosis in MCI due to clinically suspected Alzheimer's pathology
•
To determine eligibility for an approved amyloid-targeting therapy and to monitor response in patients who have received one
•
Uncertain: Subjective cognitive decline (normal on testing) in a person at increased risk because of age, APOE4 or family history; prognosis in dementia due to clinically suspected Alzheimer's pathology
Situations where amyloid PET is rarely appropriate:
•
Cognitively unimpaired individuals (whether or not at increased risk of Alzheimer's disease)
•
Subjective cognitive decline (normal on testing) without increased risk
•
To determine severity or track progression in a patient with an established, biomarker-supported diagnosis of Alzheimer's disease
•
Prodromal Lewy body disease or dementia with Lewy bodies
•
MCI or dementia with conclusive recent CSF biomarkers (whether or not consistent with Alzheimer's pathology)
•
In lieu of genotyping in suspected carriers of an autosomal dominant mutation
•
Non-medical uses such as legal, insurance or employment screening
Anti-Amyloid Therapy and Amyloid PET
•
Amyloid-targeting monoclonal antibody therapy (lecanemab) is started in early Alzheimer's disease, that is, mild cognitive impairment or mild dementia due to Alzheimer's disease. Amyloid pathology must be demonstrated by amyloid PET or CSF testing before treatment; when imaging is used for confirmation, amyloid PET is recommended (a positive tau PET indicates Alzheimer's disease, but some patients have a positive amyloid PET and a negative tau PET)
•
ARIA (amyloid-related imaging abnormalities): A common side effect of amyloid-lowering antibodies. ARIA-E is oedema; ARIA-H is haemorrhagic change (microhaemorrhage, superficial siderosis). It is usually radiographically mild to moderate, asymptomatic or mildly symptomatic, and reversible; it is detected and followed by MRI
•
Frequency (CLARITY AD trial): ARIA-E was reported in 12.6%, symptomatic in 2.8%; 5.4% in APOE4 non-carriers, 10.9% in heterozygotes and 32.6% in homozygotes, which is why APOE genotyping is recommended before treatment
•
MRI requirements: An MRI within 12 months before treatment is required; more than 4 microhaemorrhages, a haemorrhage >10 mm, superficial siderosis or vasogenic oedema exclude treatment. Monitoring MRI is recommended before the 5th, 7th and 14th infusions, and at week 52 in APOE4 carriers and in those who have had ARIA. A patient who cannot undergo MRI (for example with a pacemaker) is not eligible
•
EANM Focus Meeting 7 consensus (2026): plasma biomarkers are the first-line diagnostic tool; confirmation is by CSF or molecular imaging, and staging, treatment inclusion and monitoring by molecular imaging. Amyloid and tau biomarkers are offered only to people with objective cognitive impairment, together with MRI and requested by a cognitive impairment specialist. For treatment eligibility amyloid PET is read visually and quantitatively (centiloid) together; quantitative measurement is mandatory for monitoring, and centiloid-based amyloid PET is the only biological method operationalised for stopping treatment (donanemab). A high baseline centiloid (>108 CL) is associated with a higher ARIA risk. Tau PET is the preferred staging biomarker; it complements but does not replace amyloid PET. FDG PET is used for staging neurodegeneration and for differential diagnosis by pattern; about half of patients with Alzheimer's disease have co-pathology contributing to symptoms
•
Report content (SNMMI/EANM 2016): in addition to the common items (see 27), clinical question and cognitive status, MRI findings (atrophy, vascular lesions); tracer used and its tracer-specific read method, uptake interval. The result is given as positive or negative; if positive, the regions with grey matter uptake are stated; if quantitative analysis (SUVR, centiloid) was performed, the reference region and threshold are given. Atrophy may complicate interpretation, and a positive scan alone does not establish a diagnosis of Alzheimer's disease
BOARD TIP
▸
"What does a negative amyloid PET mean?"
→ Alzheimer pathology very unlikely → strong negative predictive value.
▸
"Positive amyloid PET = Alzheimer disease?"
→ NO! Amyloid positivity may also be seen in healthy elderly individuals (10-30%, age-dependent) → clinical correlation is needed.
▸
"If grey-white matter contrast is lost on amyloid PET?"
→ Positive (amyloid present).
BOARD TIP
▸
"How is amyloid shown before anti-amyloid therapy?"
→ Amyloid PET or CSF testing.
▸
"What are ARIA-E and ARIA-H?"
→ Oedema and haemorrhage (microhaemorrhage, superficial siderosis); followed by MRI.
▸
"Which group has the highest risk of ARIA?"
→ APOE4 homozygotes.
19.5 Tau PET (Flortaucipir)3 questions
Principle:
•
Tau PET images the hyperphosphorylated tau protein that forms neurofibrillary tangles (NFT) in the brain
•
Tau accumulation correlates more closely with neuronal damage and cognitive decline in Alzheimer disease (unlike amyloid)
•
Tau distribution reflects Braak staging
F-18 Flortaucipir (Tauvid):
•
The first FDA-approved tau PET agent (2020); F-18 florquinitau (MK-6240) was also approved in 2026
•
T½: 110 minutes (F-18)
•
Injection-to-imaging interval: 80 minutes
•
Imaging time: 20 minutes
•
Target: Paired helical filament (PHF) tau → Alzheimer-specific 3R/4R tau
Braak Staging and Tau Distribution:
•
Stage I-II: Entorhinal cortex and hippocampus (transentorhinal)
•
Stage III-IV: Limbic structures, inferior temporal cortex
•
Stage V-VI: Widespread neocortical spread (frontal, parietal, occipital)
•
Tau PET can assess Braak stage in vivo
Clinical Use:
•
Provides information additional to amyloid PET in the diagnosis of Alzheimer disease
•
Tau burden is directly proportional to the degree of cognitive impairment (unlike amyloid)
•
Distinguishing Alzheimer from non-Alzheimer tauopathies (although flortaucipir shows limited uptake in non-AD tauopathies)
•
Potential to monitor the efficacy of anti-tau therapies
Off-target Uptake:
•
Choroid plexus (common → source of false positives)
•
Basal ganglia (age-related)
•
Meningioma
•
MAO-B enzyme (monoamine oxidase B)
•
Preparation, protocol and reading (EANM/SNMMI 2026): fasting is not required; details of preparation are given in the guideline's supplement. Flortaucipir has high in vitro affinity for MAO-B, but in vivo binding is minimal. Activity and imaging time by tracer: flortaucipir 370 MBq, 20 minutes starting at 80 minutes (23.5 µSv/MBq); MK-6240 (florquinitau, Tauklarify; FDA approval 13 August 2026) 180-250 MBq, 20 minutes at 90-110 minutes (label: 185 MBq, imaging started at about 90 minutes); PI-2620 185 MBq, 30-75 minutes (0-60 minute dynamic acquisition in 4R tauopathies); florzolotau (PM-PBB3) 185-370 MBq, 20 minutes starting at 90 minutes. Early frames can serve as a perfusion proxy and should be acquired when possible
–
Visual read of flortaucipir: the scale is set at 1.65 times mean cerebellar uptake; increased uptake in posterolateral temporal, parietal or occipital cortex (Braak IV-VI) is positive, and uptake confined to the medial temporal lobe is read as negative. For quantification the reference region is the inferior cerebellar cortex; the hippocampus is excluded because choroid plexus uptake can spill into it. With second-generation tracers, off-target uptake in the meninges and skull may spill into adjacent cortex. Non-specific uptake may occur in haemorrhage, post-ischaemic areas and severe neurodegeneration (e.g. posterior putamen in MSA)
EANM – Tau PET Agents and Appropriate Use:
•
First generation: F-18 flortaucipir (AV-1451), THK5317, THK5351, C-11 PBB3; off-target binding is a problem
•
Second generation: F-18 MK-6240, PI-2620, RO-948, JNJ-311/069, GTP1, PM-PBB3; off-target binding is reduced
•
In AD tau accumulates as paired helical filaments (3R/4R), in other tauopathies as straight or twisted filaments; most agents have higher affinity for "AD-type" tau
•
Appropriate use: MCI or dementia patients younger than 65 years with suspected AD; atypical presentations; determining prognosis in MCI or dementia; assessing eligibility for approved amyloid-targeted therapy
•
Rarely appropriate: Cognitively unimpaired individuals, subjective cognitive decline (normal on testing), non-medical use, in lieu of genotyping in suspected autosomal dominant mutation carriers
•
Uncertain: A typical (amnestic) presentation with onset at 65 or older, determining severity or tracking progression in a biomarker-supported diagnosis, Lewy body disease, patients with recent CSF biomarkers, monitoring response to amyloid-targeting therapy
•
Report content (EANM/SNMMI 2026): in addition to the common items (see 27), tracer, activity, uptake interval and system used (PET, PET/CT, PET/MR), issues affecting quality (motion, extravasation). Findings: regions with increased retention (medial temporal lobe, lateral temporal, parietal, occipital, frontal, motor cortex) and the level of increase (low, intermediate, high); non-AD patterns (e.g. striatal uptake in 4R tauopathies) and non-specific uptake in haemorrhage, post-ischaemic areas or severe neurodegeneration; if quantitative analysis was done, the reference region and SUVR. The conclusion follows the tracer-specific read method
BOARD TIP
▸
"Difference between tau PET and amyloid PET?"
→ Amyloid → early accumulation (15-20 years before symptoms), weak correlation with cognitive decline. Tau → strong correlation with cognitive decline, reflects Braak staging.
▸
"Off-target uptake of flortaucipir?"
→ Choroid plexus (most common), basal ganglia, MAO-B.
Figure 19.4: Timing of amyloid and tau accumulation relative to symptom onset and tau distribution by Braak stage (schematic).
Did this figure help you understand the topic?
19.6 Dopamine Transporter Imaging (DaTscan – I-123 Ioflupane)11 questions
Principle:
•
Dopamine transporter (DAT): A membrane protein on presynaptic dopaminergic nerve terminals → takes dopamine back up from the synaptic cleft
•
DAT density reflects the integrity of nigrostriatal dopaminergic neurons
•
In Parkinson disease and other neurodegenerative parkinsonisms nigrostriatal neurons degenerate → DAT falls
•
In essential tremor, drug-induced parkinsonism and psychogenic parkinsonism the nigrostriatal pathway is intact → DAT normal
I-123 Ioflupane (DaTscan):
•
Cocaine analogue → high-affinity binding to DAT
•
Other DAT agents: I-123 β-CIT (because of its slow kinetics imaging is performed ~24 hours later) and Tc-99m TRODAT-1 (labelled with Tc-99m, used in some countries)
•
T½: 13.2 hours (I-123)
•
Energy: 159 keV
•
Dose: 3-5 mCi (111-185 MBq) IV
•
Thyroid blockade: Potassium perchlorate (400 mg orally, 1 hour before) or Lugol's solution or SSKI → prevents thyroid uptake of free I-123
•
Imaging: 3-6 hours after injection → brain SPECT
•
EANM: Adult 150-250 MBq (usually 185 MBq); the thyroid may be blocked with 200 mg sodium perchlorate at least 5 min before injection; breastfeeding is interrupted for 4 days
•
Interacting drugs are stopped at least 5 biological half-lives before. Cocaine, amphetamines, modafinil, antidepressants, opioids and anaesthetics reduce binding; adrenergic agonists increase it
•
Sensitivity is 91.9% and specificity 83.6% (FDA pooled analysis of 764 cases); interobserver agreement κ 0.8-1.0. The most important pitfall is the age-related decline in binding
•
DaTscan cannot distinguish PD, MSA and PSP from one another; nor can it distinguish DLB from Parkinson disease dementia
F-18 FDOPA PET (presynaptic dopaminergic):
•
Crosses the BBB, is converted to fluorodopamine by aromatic amino acid decarboxylase and stored in vesicles
•
Carbidopa increases striatal uptake ~2-fold; haloperidol and MAO inhibitors increase uptake, whereas reserpine blocks it
•
Preparation: Fasting for at least 4 hours; antiparkinsonian treatment is stopped at least 12 hours before. 150-250 MBq, 25 µSv/MBq (critical organ bladder wall); breastfeeding is interrupted for 12 hours
•
When symptoms begin, uptake in the contralateral putamen has fallen by ~30%; it is not reliable for distinguishing atypical parkinsonism from PD
•
Postsynaptic D2 receptor imaging: With I-123 IBZM SPECT and C-11 raclopride PET striatal binding is reduced in MSA and PSP and normal or increased in PD. C-11 carfentanil is a μ-opioid receptor ligand
Normal Image:
•
Bilateral striatum (caudate nucleus + putamen) → symmetric, comma/crescent-shaped intense uptake
•
"Comma" shape → caudate (head) + putamen (tail) seen together
Abnormal Patterns:
•
Early Parkinson disease: Putaminal uptake reduced (starting in the posterior putamen) → caudate preserved → "dot" (period) shape
•
Advanced Parkinson disease: Both putaminal and caudate uptake reduced → "egg" shape or minimal uptake
•
Asymmetric uptake: Parkinson disease usually begins asymmetrically → more marked reduction on the contralateral side (opposite to the clinical symptoms)
•
In atypical parkinsonism (especially PSP) the striatal reduction is more symmetric and also markedly affects the caudate; however, it does not reliably distinguish it from PD in an individual case
Clinical Indications:
1. Parkinson disease vs essential tremor (the most important indication)
–
Parkinson disease → DaTscan abnormal (reduced striatal uptake)
–
Essential tremor → DaTscan normal
2. DLB vs Alzheimer disease
–
DLB → DaTscan abnormal ("indicative biomarker" in the DLB diagnostic criteria)
–
Alzheimer disease → DaTscan normal
3. Drug-induced parkinsonism vs neurodegenerative parkinsonism
–
Drug-induced → DaTscan normal (presynaptic neurons intact)
–
Neurodegenerative → DaTscan abnormal
Conditions with an Abnormal DaTscan (Presynaptic Dopaminergic Loss):
•
Parkinson disease
•
Dementia with Lewy bodies (DLB)
•
Multiple system atrophy (MSA)
•
Progressive supranuclear palsy (PSP)
•
Corticobasal degeneration (CBD)
Conditions with a Normal DaTscan (Presynaptic Neurons Intact):
•
Essential tremor
•
Drug-induced parkinsonism
•
Psychogenic parkinsonism
•
Vascular parkinsonism (usually normal, but may vary)
•
Dopa-responsive dystonia (Segawa disease): DaTscan is normal because presynaptic neurons are intact; this distinguishes it from early-onset PD
•
Alzheimer disease
•
Dystonic tremor
BOARD TIP
▸
"What does a comma → dot change on DaTscan mean?"
→ Putaminal DAT loss → early Parkinson disease.
▸
"Does DaTscan distinguish Parkinson disease from essential tremor?"
→ YES (the most important indication).
▸
"Does DaTscan distinguish Parkinson disease from MSA?"
→ NO! It is abnormal in both → clinical distinction is needed.
▸
"DaTscan in the DLB diagnostic criteria?"
→ Indicative biomarker → strongly supports the diagnosis of DLB.
Figure 19.5: Normal "comma", early Parkinson "dot" and advanced Parkinson patterns on DaTscan; conditions giving abnormal and normal DaTscan (schematic).
Did this figure help you understand the topic?
Drug Interactions:
Drugs to be stopped before DaTscan (they affect imaging by binding to DAT):
•
Cocaine: 2 days
•
Amphetamine: 7 days; methylphenidate: 2 days
•
Bupropion: 8 days (the frequently asked longest; phentermine ~14 days)
•
SSRIs: Usually need not be stopped (minimal effect on DAT)
•
Levodopa, dopamine agonists: Need not be stopped (do not significantly affect DAT binding)
•
Report content (EANM/SNMMI 2020): in addition to the common items (see 27), side and duration of symptoms, drugs that may affect binding (e.g. SSRIs, amphetamine, bupropion), thyroid blockade and uptake interval. Findings: visual pattern (normal comma shape, unilateral or bilateral putaminal reduction); if semi-quantitative analysis was performed, the software used, specific binding ratios for caudate and putamen, their values against an age-matched normal database, asymmetry and caudate/putamen ratio. The conclusion states the presence or absence of a presynaptic dopaminergic deficit
BOARD TIP
▸
"Drug that must be stopped longest before DaTscan?"
→ Bupropion (8 days; phentermine ~14 days).
▸
"Should levodopa be stopped?"
→ NO → levodopa does not significantly affect DAT binding.
▸
"Should SSRIs be stopped?"
→ Usually not (no clinically significant effect on DAT).
19.7 Assessment of Epilepsy (Ictal/Interictal SPECT)9 questions
General Principle:
In candidates for epilepsy surgery, localisation of the seizure focus (epileptogenic zone) is critically important. Nuclear medicine methods provide additional localising information, especially in patients with MRI-negative or discordant findings.
Interictal SPECT (Between Seizures)
•
Hypoperfusion (reduced blood flow) at the seizure focus between seizures
•
Radiopharmaceutical: Tc-99m HMPAO or Tc-99m ECD
•
Sensitivity: 40-50% (low → insufficient on its own)
•
Specificity: Moderate
•
Finding: Focal hypoperfusion in the epileptogenic zone
Ictal SPECT (During a Seizure)
•
Hyperperfusion (increased blood flow) in the epileptogenic zone during a seizure
•
Radiopharmaceutical: Tc-99m ECD (6 hours stable) or stabilised HMPAO (~4 hours)
•
Sensitivity: 90-95% (much higher than interictal)
•
Critical point: the radiopharmaceutical is injected through a previously placed IV line as early as possible after seizure onset, ideally within the first 20 seconds (EANM 2024; textbooks often give 20-30 s) → late injection → shows seizure propagation, not the focus. Continuous video-EEG monitoring is required; the best results are obtained in focal impaired-awareness seizures; prepared syringes are kept in the monitoring unit, and an automatic injector shortens latency and increases single localising foci
•
In ictal SPECT, hyperperfusion at the focus may be accompanied by hyperperfusion in the ipsilateral basal ganglia and contralateral cerebellum; this supports lateralisation
•
Practical difficulty: It is not known in advance when a seizure will occur → the radiopharmaceutical must be kept ready → HMPAO can be kept stable for up to 4 hours with a stabiliser
•
Imaging after injection: Can be performed even after the seizure has ended (brain fixation complete at the moment of injection → "snapshot" property)
•
EANM 2024: HMPAO or ECD typically 740 MBq in adults; acquisition starts 30-90 min after injection and lasts ~20 min. For interictal SPECT, EEG is recorded during injection where possible (especially the first 2 minutes of uptake) to exclude ictal activity. The patient is supervised throughout; benzodiazepines are used with caution
–
Preparation (EANM 2024): ictal studies require continuous video-EEG monitoring, a previously placed IV line and prepared syringes kept in the epilepsy monitoring unit; the best results are obtained with injection during focal impaired awareness seizures. Acquisition starts 30-90 minutes after injection and lasts about 20 minutes. For interictal SPECT, EEG is monitored during injection if available, to exclude ictal activity especially in the first 2 minutes of tracer uptake. Antiseizure medication is not stopped (caution with benzodiazepines); the time of the last seizure is documented. The patient is under continuous supervision throughout; anaesthesia given within 30 minutes before FDG affects the findings
–
Correct interpretation of SISCOM requires knowledge of the injection time and seizure type; the most commonly used thresholds are z = +1.5 and +2
Timing of Ictal SPECT Injection and Findings:
•
Early ictal (0-20 s): Focal hyperperfusion in the epileptogenic zone → the most valuable finding
•
Late ictal (20-60 s): Seizure propagation → ipsilateral temporal + frontal hyperperfusion → localisation of the focus becomes difficult
•
Postictal (after the seizure): Hypoperfusion in the epileptogenic zone (postictal switch) + hyperperfusion in surrounding areas may be present
•
Postictal SPECT: Injection within 1-2 minutes after the seizure → in temporal lobe epilepsy hypoperfusion at the focus + transient hyperperfusion in the lateral temporal cortex (postictal switch pattern)
BOARD TIP
▸
"When should the injection be given in ictal SPECT?"
→ Within the first 20-30 seconds after seizure onset.
▸
"What does a late injection show?"
→ Shows seizure propagation, not the focus → localising value falls.
▸
"Can imaging be performed after the seizure has ended?"
→ YES → HMPAO freezes the perfusion at the moment of injection → the perfusion at that moment is imaged even if the seizure has ended.
Left: transaxial brain slice at the temporal lobe level showing live perfusion changing during a seizure and the SPECT image frozen at injection. Right: perfusion curves of the focus, ipsilateral spread and contralateral cortex. The second step shows the ictal − interictal difference image and MRI fusion.
Continuous stream · no seeking
Ictal SPECT freezes the moment of injection: injection within the first 20 seconds shows the focus hyperperfused; SISCOM fuses the ictal − interictal difference with MRI to localize it.
Source: phase findings and sensitivities from book 19.7. Perfusion time profile representative. Anatomy: OpenStax A&P 2e §13.2.
How to use: a single click on the drawing pauses and resumes; ↑ ↓ or the buttons below change the injection time (interictal, early ictal, late ictal, postictal); ← → change steps.
Did this animation help you understand the topic?
Figure 19.6: Perfusion SPECT findings in epilepsy by injection time: interictal, early ictal, late ictal and postictal (schematic).
SISCOM (Subtraction Ictal SPECT Co-registered to MRI)
•
The interictal SPECT image is subtracted from the ictal SPECT image
•
The difference image is overlaid on the patient's MRI by co-registration
•
Advantages:
–
Localises the epileptogenic zone anatomically
–
Highlights the perfusion difference between ictal and interictal states
–
Very valuable for surgical planning
•
Sensitivity: higher localising accuracy than ictal SPECT; >90% in TLE and ~70% in extratemporal epilepsy. It shows hypoperfusion as well as hyperperfusion; the most used thresholds are z = +1.5 and +2. When the SISCOM focus agrees with other methods, the odds of seizure freedom after surgery are ~3 times higher (EANM 2024). Injection timing and seizure type must be known for correct interpretation
•
Especially valuable in MRI-negative epilepsy
BOARD TIP
▸
"What is SISCOM?"
→ The ictal − interictal SPECT difference image overlaid on MRI by co-registration.
▸
"When is SISCOM most valuable?"
→ In MRI-negative epilepsy → surgical localisation of the focus.
Left: transaxial brain slice at the temporal lobe level showing live perfusion changing during a seizure and the SPECT image frozen at injection. Right: perfusion curves of the focus, ipsilateral spread and contralateral cortex. The second step shows the ictal − interictal difference image and MRI fusion.
Continuous stream · no seeking
Ictal SPECT freezes the moment of injection: injection within the first 20 seconds shows the focus hyperperfused; SISCOM fuses the ictal − interictal difference with MRI to localize it.
Source: phase findings and sensitivities from book 19.7. Perfusion time profile representative. Anatomy: OpenStax A&P 2e §13.2.
How to use: a single click on the drawing pauses and resumes; ↑ ↓ or the buttons below change the injection time (interictal, early ictal, late ictal, postictal); ← → change steps.
Did this animation help you understand the topic?
Figure 19.7: SISCOM: co-registration to MR of the difference image obtained by subtracting interictal SPECT from ictal SPECT (schematic).
Assessment of Epilepsy with FDG PET
•
Interictal FDG PET: Hypometabolism (reduced FDG uptake) in the epileptogenic zone
•
EANM 2024: interictal SPECT and FDG PET are performed, where possible, at least 24 hours after a focal aware seizure and at least 48 hours after a focal impaired-awareness or focal-to-bilateral tonic-clonic seizure; the time of the last seizure is recorded. FDG shows the irritative zone, not only the seizure onset zone, and grey matter metabolism may be globally reduced by 10-25%. Interictal hypermetabolic foci can occur in cortical malformations; in young children cortical hypermetabolism may mark severe epilepsy and the need for early surgery
•
Sensitivity: 80-90% (in temporal lobe epilepsy) → much higher than interictal SPECT
•
In temporal lobe epilepsy: Ipsilateral temporal lobe hypometabolism
•
Marked unilateral temporal hypometabolism predicts seizure freedom after surgery; bilateral or extensive extratemporal hypometabolism is associated with a worse surgical outcome
•
The hypometabolic area is usually larger than the epileptogenic zone → does not define surgical margins precisely
•
In children the focus is often extratemporal (cortical dysplasia) and MR may be normal. In infantile spasms focal cortical hypometabolism identifies surgical candidates; in Lennox-Gastaut syndrome unilateral focal or diffuse, or bilateral diffuse hypometabolism may be seen
•
Lower sensitivity in extratemporal epilepsy (50-60%)
•
In tuberous sclerosis the tubers are hypometabolic on FDG and the epileptogenic tuber cannot be distinguished this way; on C-11 alpha-methyl-L-tryptophan (AMT) PET uptake is increased in the epileptogenic tuber
•
C-11 flumazenil PET / I-123 iomazenil SPECT: Binding to GABA-A/benzodiazepine receptors is reduced at the epileptogenic focus; because the area of reduced binding is narrower than FDG hypometabolism, the margins of the focus are defined more precisely
•
Ictal FDG PET: Practically not feasible (FDG uptake time 30-60 min → seizures do not last this long)
•
Pitfalls in visual interpretation (EANM 2024): movement artefacts are sought because seizures may occur during acquisition; alignment of the attenuation-correction image (CT or MR µ-map) with the emission data is checked, and in young children (incomplete myelination) MR-based µ-maps may be mis-segmented; EEG electrodes can cause attenuation artefacts. Asymmetry is the main criterion, but bilateral symmetric hypometabolism must not be missed. Interictal FDG may show diffuse hypometabolism beyond the focus (grey matter metabolism is globally 10-25% lower); hypermetabolic foci may also be seen (ongoing epileptic activity, cortical malformation in children). A recent MRI (as contemporaneous as possible under 6 years of age) is used for fusion
•
Report content (EANM 2024): in addition to the common items (see 27), type of study (ictal, interictal, ictal subtraction SPECT or interictal FDG PET), timing of the ictal injection relative to seizure onset and the duration of the seizure (video-EEG), whether seizures occurred between injection and imaging, seizures or EEG findings in the 24 hours before FDG and during uptake, antiseizure medication and sedation. Findings: site and extent of ictal hyperperfusion or interictal hypometabolism, the result of SISCOM or voxel-based analysis if used, assessment co-registered with MRI and concordance with EEG and semiology; that propagation areas may be mistaken for the seizure onset zone
Comparison of Nuclear Medicine Methods in the Assessment of Epilepsy:
| Method | Finding | Sensitivity | Advantage | Disadvantage |
|---|---|---|---|---|
| Interictal SPECT | Hypoperfusion | 40-50% | Easy to perform | Low sensitivity |
| Ictal SPECT | Hyperperfusion | 90-95% | High sensitivity | Timing difficulty |
| SISCOM | Difference image | >90% | Anatomical localisation | Ictal + interictal required |
| Interictal FDG PET | Hypometabolism | 80-90% (TLE) | High sensitivity | Large hypometabolic area |
19.8 Brain Death Scintigraphy7 questions
Definition:
Brain death is the irreversible loss of all brain functions (cerebral hemispheres + brainstem). Clinical diagnosis is fundamental, but in some situations a confirmatory test is needed.
Indications for a Confirmatory Test:
•
Situations in which clinical assessment cannot be performed reliably (severe facial trauma, drug intoxication, hypothermia)
•
Unlike EEG, perfusion scintigraphy is not affected by barbiturates and sedatives or by hypothermia; it is therefore the preferred confirmatory test in these situations
•
Legal requirement in some countries
•
To shorten the observation period
•
In paediatric patients (especially <1 year)
Radiopharmaceuticals and Methods
A) Tc-99m HMPAO (or Tc-99m ECD) – SPECT:
•
The preferred method (most reliable)
•
Crosses the BBB → enters the brain parenchyma → directly assesses brain perfusion
•
In brain death: No uptake in the cerebral hemispheres and cerebellum → "hollow skull" or "empty skull" appearance
•
Nasal mucosal uptake may be seen → "hot nose" sign (external carotid perfusion preserved, no internal carotid perfusion → blood is diverted to the external carotid)
•
Advantage: Assesses perfusion of both the hemispheres and the brainstem/cerebellum
B) Tc-99m DTPA or Tc-99m Pertechnetate – Planar Angiography:
•
Does not cross the BBB → assesses only cerebral blood flow
•
Dynamic imaging: IV bolus → anterior planar → flow in the cerebral arteries is assessed
•
In brain death: Because there is no internal carotid flow, the normal "trident" (ACA and bilateral MCA) appearance is lost; the sagittal sinus mostly does not fill. On delayed images faint activity may be seen in the superior sagittal sinus in the absence of intracranial arterial flow, and this does not exclude the diagnosis
•
Disadvantage: Inadequate assessment of the posterior fossa (brainstem, cerebellum) → the vertebral arteries are poorly assessed on planar imaging
•
HMPAO/ECD SPECT is therefore preferred
Findings on Brain Death Scintigraphy:
•
"Hot nose" sign: Increased nasal mucosal uptake → increased external carotid flow (blood is diverted to the external carotid because there is no internal carotid flow). SNMMI/EANM 2025: historically described as supporting the diagnosis, but it may represent flow rerouted to the brainstem or cervical cord region rather than the nose, or be non-specific. The diagnosis rests on the absence of intracranial perfusion
•
"Empty/hollow skull" sign: No uptake in the cerebral hemispheres and cerebellum (with HMPAO/ECD)
•
Loss of the "trident": The three-pronged flow image normally formed by the ACAs and MCAs is not seen → only external carotid branches are seen (on planar angiography)
•
For a planar flow study to be valid, an adequate carotid bolus must be seen in the neck; absence of intracranial arterial flow despite a good bolus is the diagnostic criterion
BOARD TIP
▸
"Preferred radiopharmaceutical for confirming brain death?"
→ Tc-99m HMPAO (or ECD) → crosses the BBB → directly assesses parenchymal perfusion → both hemispheres and posterior fossa.
▸
"Hot nose sign?"
→ External carotid perfusion preserved, internal carotid absent → blood is diverted to the nasal mucosa.
▸
"Why is Tc-99m DTPA/pertechnetate inadequate?"
→ Does not cross the BBB → assesses only flow → posterior fossa assessment inadequate.
Figure 19.8: Brain death scintigraphy: "empty skull" and "hot nose" with HMPAO/ECD, loss of the "trident" on the planar flow study (schematic).
Did this figure help you understand the topic?
Technical Details:
•
Patient position: Supine, head close to the anterior detector
•
Activity (SNMMI/EANM 2025): in adults 1110 MBq (30 mCi) for both hydrophilic and lipophilic agents; in children by weight, but at least 185 MBq (5 mCi) for a diagnostic flow study. A lower activity may help if a repeat study is anticipated. Lipophilic agents are preferred in adults and children; a hydrophilic agent is acceptable when a lipophilic agent is not available
•
Dynamic phase: 1-2 s/frame × 60 s (bolus injection). SNMMI/EANM 2025: 1-s frames start just before injection and last at least 60 s, and can be combined into 2-3-s frames for review; the field of view covers the common carotids to the vertex
•
Static phase: with a hydrophilic agent (DTPA/pertechnetate), blood-pool images are acquired immediately after the flow phase (1-2 min) and contribute little to diagnosis; there is no parenchymal phase. With a lipophilic agent (HMPAO/ECD) the parenchymal phase is acquired at 20 min and is the essential diagnostic component (SNMMI/EANM 2025)
•
Parenchymal-phase images: planar (anterior and at least one lateral; if possible a second lateral and posterior; laterals are needed for the posterior fossa), SPECT or both. SPECT helps with the posterior fossa and with separating scalp activity, but systematic reviews have not shown it to be superior (SNMMI/EANM 2025)
•
Tourniquet: a scalp tourniquet to reduce scalp blood flow is no longer recommended; it may aggravate skull injury (SNMMI/EANM 2025)
•
EANM: HMPAO SPECT can be used as a confirmatory test in the organ donation process; it can show that damage is more extensive than seen on CT
•
Blood pressure must be stable; because inadequate labelling can cause low brain activity, quality control is critical in brain death studies
•
Prerequisites (SNMMI/EANM 2025): the clinical neurological examination and apnoea test should be performed as completely as possible (deep coma, absent cranial nerve reflexes, apnoea); the request states which components were performed and why an ancillary test is needed. Any central nervous system-mediated reflex on recent examination precludes the diagnosis and makes the test superfluous. The cause of the devastating brain injury (hypoxic-ischaemic, traumatic, cerebrovascular) and any shunt, open fontanelle or craniotomy are recorded. Blood pressure and haemodynamics are kept stable by the intensive care team (absent perfusion due to transient hypotension must not be taken as permanent), correctable biochemical abnormalities are corrected and the patient is normally ventilated (hyperventilation alters cerebral blood flow). The test is indicated when hypothermia, metabolic derangement, intoxication, CNS depressants or spinal cord injury make the clinical examination unreliable
•
With lipophilic agents planar images are mostly sufficient; SPECT is helpful but not mandatory. In newborns and young infants cerebral perfusion may be preserved despite clinical brain death, and the sensitivity of the test falls
•
Interpretation and report (SNMMI/EANM 2025): any intracranial activity on either the flow or the parenchymal phase (cerebrum, cerebellum or, as far as visible, brainstem) indicates perfusion; absent perfusion is reported only in a technically adequate and unequivocal study (an adequate bolus must be seen in the neck). Preliminary or provisional reporting is not recommended. 'Absent cerebral perfusion' is insufficient because the posterior fossa must also be addressed; the conclusion is written as 'no evidence of brain (intracranial) perfusion', not 'brain death' (that diagnosis is made by the clinician). If a small amount of perfusion is present, a repeat study after 1-2 days may be recommended in older children and adults (ischaemia is progressive; it may not progress in infants with open fontanelles or after craniotomy). At least 24 hours should pass before repeating with a lipophilic agent; with DTPA and normal renal function a new bolus can be given after 20-30 min, whereas pertechnetate requires waiting until the next day
19.9 CSF Flow Scintigraphy (Cisternography)7 questions
Principle:
•
The radiopharmaceutical is injected into the subarachnoid space (intrathecally) by lumbar puncture
•
Assesses the CSF flow pathway and dynamics
•
Radiopharmaceutical: In-111 DTPA (preferred) or Tc-99m DTPA
•
Advantage of In-111 DTPA: Long T½ (67 hours) → imaging possible at 24-48-72 hours
•
Tc-99m DTPA: Short T½ (6 hours) → early imaging only
•
EANM: Usually 20 MBq In-111 DTPA is used (~2.4 mSv); ≤4 MBq/1 mL for assessing a drainage system. MEAP collimator, 173 and 247 keV windows
•
Head images are acquired at 4, 24 and 48 hours in adults and at 1, 2, 6 and 24 hours in children; normally activity reaches the convexity at ~24 hours
•
Ventricular reflux is transient in atrophy and in children, and persistent in NPH
•
Other indications: Function of an intrathecal or epidural pump, patency of a CSF drainage system
Normal CSF Flow Pattern:
•
2-4 hours: Filling of the basal cisterns (prepontine, ambient, suprasellar cisterns)
•
6 hours: Ascent begins along the Sylvian fissures and interhemispheric fissure
•
12 hours: Widespread distribution over the cerebral convexities
•
24 hours: Widespread uptake over the convexities → absorption via arachnoid granulations → activity begins to decrease
•
48 hours: Activity markedly reduced → absorption continues
•
Ventricular uptake: Under normal conditions there is NO ventricular uptake (or it is very minimal and transient)
Normal Pressure Hydrocephalus (NPH):
•
Clinical triad: Dementia + gait disturbance (ataxia) + urinary incontinence (Hakim triad)
•
Cisternography findings:
–
Early ventricular filling (2-6 hours) → ventricular reflux → abnormal
–
Persistence of ventricular activity at 24-48 hours → impaired CSF absorption
–
Late or no arrival at the convexities → convexity block
•
Normal: No ventricular uptake + widespread distribution over the convexities at 24 hours
•
NPH vs cerebral atrophy:
–
NPH: Ventricular reflux + convexity block → benefits from shunt surgery
–
Cerebral atrophy (hydrocephalus ex vacuo): Ventricles enlarged but CSF flow normal (at most transient ventricular reflux) → does not benefit from a shunt
•
The value of cisternography in predicting shunt response is limited: the classic pattern increases the likelihood of response, but a normal cisternogram does not exclude NPH; the decision is based on the clinical picture, MR and the CSF drainage (tap) test
BOARD TIP
▸
"Cisternography finding in NPH?"
→ Early ventricular reflux + ventricular persistence at 24-48 hours + failure to reach the convexities.
▸
"Clinical triad of NPH?"
→ Dementia + ataxia + urinary incontinence (Hakim triad).
▸
"NPH vs atrophy?"
→ NPH → ventricular reflux present → benefits from a shunt. Atrophy → CSF flow normal → does not benefit from a shunt.
Figure 19.9: Normal CSF flow on radionuclide cisternography and ventricular reflux and persistence in NPH (schematic).
Did this figure help you understand the topic?
Assessment of CSF Leak:
•
In suspected rhinorrhoea (CSF from the nose) or otorrhoea (CSF from the ear)
•
The commonest cause of traumatic CSF rhinorrhoea is a cribriform plate or ethmoid fracture; because the leak may be intermittent, prolonged pledget counting with In-111 DTPA is an advantage for catching small or intermittent leaks
•
Intrathecal In-111 DTPA injection → nasal or auricular cotton pledgets are placed
•
Activity in the pledgets is measured → compared with plasma activity
•
Pledget/plasma ratio >1.5 → CSF leak positive
•
Focal accumulation of activity may be seen at the leak site on planar imaging
•
SPECT/CT → better shows the location of the leak
•
EANM: Nasal and ear pledgets are placed 4 hours after injection and left for 24 hours; a blood sample is taken when the pledgets are removed (0.5 mL plasma is counted); each centre should establish its own pledget reference values
•
Pitfalls: if part of the injection ends up beside or outside the puncture site it is seen in the lumbar region within half an hour, followed by renal and bladder activity, so the injection site and spine are imaged first. Early bladder activity due to lumbar puncture trauma is not a reliable indirect sign of a leak. Acetazolamide may delay distribution and cause ventricular reflux and is stopped at least 2 days before the study. Subarachnoid adhesions can produce asymmetrical images
–
For pledget counting the pledgets must be of adequate size and absorbency, and each department sets its own reference values; early removal of pledgets lowers sensitivity. When rhinorrhoea is suspected an abdominal image is taken at 24 hours: gastric activity indicates swallowed CSF. In shunt studies an accidental subcutaneous injection mimics distal obstruction (see 25.5.2)
•
CT or MR myelography appears less sensitive than radionuclide cisternography
Shunt Assessment (Ventriculoperitoneal Shunt Patency):
•
Injection of Tc-99m DTPA (or Tc-99m pertechnetate) into the shunt reservoir
•
Normal: The radiopharmaceutical progresses along the shunt → reaches the peritoneal cavity
•
Obstruction: Obstruction is localised at the point where the radiopharmaceutical stops progressing
•
Proximal obstruction: No passage from the ventricle into the shunt
•
Distal obstruction: Progresses along the shunt but does not reach the peritoneal cavity
Key Points for the Board
Cerebral Perfusion SPECT:
•
HMPAO → trapping by glutathione, 30 min stable (without stabiliser), shows luxury perfusion
•
ECD → trapping by esterases, 6 hours stable, may not show luxury perfusion
•
Both cross the BBB → assessment of parenchymal perfusion
•
"Freezes" the perfusion at the moment of injection → snapshot property
•
Crossed cerebellar diaschisis: reduced perfusion/metabolism in the cerebellar hemisphere contralateral to a supratentorial infarct/tumour (cerebellum normal on CT/MR)
•
Acetazolamide (1 g IV) reserve test: an area with exhausted reserve falls relative to baseline; contraindications are sulphonamide allergy and stroke within the last 3 days
FDG PET Brain:
•
Most intense uptake: Cerebral cortex, basal ganglia, thalamus
•
Sedative → may be given after injection (should not be given before)
•
Blood glucose should be <150-200 mg/dL (high blood glucose reduces brain FDG uptake)
FDG PET Patterns in Dementia:
•
Alzheimer: Bilateral temporoparietal + posterior cingulate/precuneus (earliest) → motor, visual cortex and basal ganglia spared
•
FTD: Frontal + anterior temporal → frontal dominant
•
DLB: Temporoparietal + occipital → occipital involvement distinguishes it from Alzheimer
•
Vascular: Multiple, asymmetric, scattered
•
CBD: Asymmetric frontoparietal
•
PSP: Frontal + mesencephalon
Amyloid PET:
•
3 FDA-approved agents: Florbetapir (Amyvid), Florbetaben (Neuraceq), Flutemetamol (Vizamyl)
•
Positive: Grey-white matter contrast lost → widespread cortical uptake
•
Negative: Grey-white matter contrast preserved → strong negative predictive value
•
Positive amyloid PET ≠ definite Alzheimer (asymptomatic positivity in 10-30% of the elderly)
•
Appropriate use: Unexplained cognitive impairment, atypical presentation, early-onset dementia
Tau PET:
•
Flortaucipir (Tauvid, 2020) and florquinitau (Tauklarify, 2026) → FDA-approved tau PET agents
•
Tau burden is directly proportional to cognitive decline (unlike amyloid)
•
Reflects Braak staging in vivo
•
Off-target: Choroid plexus (most common), basal ganglia, MAO-B
DaTscan:
•
I-123 Ioflupane → binds DAT → integrity of presynaptic dopaminergic neurons
•
Normal: Bilateral symmetric comma/crescent shape (caudate + putamen)
•
Early Parkinson disease: Putamen ↓, caudate preserved → "dot" shape
•
Abnormal: Parkinson disease, DLB, MSA, PSP, CBD
•
Normal: Essential tremor, drug-induced parkinsonism, psychogenic, Alzheimer
•
Bupropion → stopped 8 days before (phentermine ~14 days)
•
Levodopa, dopamine agonists, SSRIs → need not be stopped
•
Dopa-responsive dystonia (Segawa) → DaTscan normal (distinguishes it from early-onset PD)
Epilepsy:
•
Interictal SPECT: Hypoperfusion → sensitivity 40-50% (low)
•
Ictal SPECT: Hyperperfusion → sensitivity 90-95% → injection within the first 20-30 s is critical
•
SISCOM: Ictal − interictal SPECT → MRI co-registration → best localisation
•
Interictal FDG PET: Hypometabolism → sensitivity 80-90% (in TLE)
•
Ictal FDG PET: Practically not feasible
•
Tuberous sclerosis: tubers are hypometabolic on FDG; the epileptogenic tuber shows increased uptake on C-11 AMT PET
Brain Death:
•
Preferred: Tc-99m HMPAO (or ECD) → crosses the BBB → parenchymal perfusion
•
Findings: Empty skull + hot nose
•
Tc-99m DTPA/pertechnetate → flow only → posterior fossa inadequate
•
Unlike EEG, perfusion scintigraphy is not affected by barbiturates/sedatives or hypothermia
Cisternography:
•
Intrathecal In-111 DTPA → preferred (long T½ → imaging at 24-48-72 hours)
•
Normal: 2-4 hours basal cisterns → 12-24 hours convexities → no ventricular uptake
•
NPH: Early ventricular reflux + ventricular persistence + convexity block → benefits from a shunt
•
CSF leak: Pledget/plasma ratio >1.5 → positive
•
VP shunt patency: Tc-99m DTPA injected into the shunt reservoir → arrival in the peritoneal cavity is assessed
Additional Notes from EANM Guidelines:
•
Perfusion SPECT: HMPAO/ECD 740 MBq; ECD 30-60 min, HMPAO 30-90 min after; phenytoin → cerebellar, benzodiazepines → frontal hypoperfusion
•
FDG brain: 150-250 MBq; MSA-P → putamen-thalamus ↓, MSA-C → cerebellum ↓, PSP → frontal + midbrain ↓, CBD → asymmetry
•
Amyloid PET: Florbetapir 370, flutemetamol 185, florbetaben 300 MBq; sensitivity ~94%, specificity ~97%; breastfeeding 12 hours
•
Tau PET: 1st generation flortaucipir (off-target); 2nd generation MK-6240, PI-2620; rarely appropriate in CU individuals
•
DaTscan: 150-250 MBq; perchlorate 200 mg; breastfeeding 4 days; distinguishes ET, cannot distinguish PD/MSA/PSP; binding declines with age
•
FDOPA: Carbidopa increases uptake ~2-fold; antiparkinsonian drugs stopped ≥12 hours
•
Cisternography: In-111 DTPA 20 MBq; reaches the convexity at ~24 hours; persistent ventricular reflux in NPH
Current guidelines on this chapter's topics
The latest guidelines and consensus documents on the procedures covered in this chapter (official EANM index and primary sources, November 2026). Comparison of the text with these versions is in progress; consult the guideline itself for details.
- 2026European Association of Nuclear Medicine (EANM) Focus Meeting 7: Consensus on the role of molecular imaging and fluid biomarkers in diagnosis, treatment eligibility, and monitoring in Alzheimer’s Disease (opens in a new tab)
- 2025The SNMMI Procedure Standard/EANM Practice Guideline for Radionuclide Brain Perfusion Scintigraphy in Suspected Death by Neurologic Criteria (Brain Death) 3.0 (opens in a new tab)
- 2025Joint EANM/EANO/RANO/SNMMI practice guideline/procedure standard for PET imaging of brain metastases: version 1.0 (opens in a new tab)
- 2024SNMMI Procedure Standard/EANM Practice Guideline for Brain [18F]FDG PET Imaging, Version 2.0 (opens in a new tab)
- 2024EANM Practice Guidelines for an appropriate use of PET and SPECT for patients with epilepsy (opens in a new tab)
- 2024Joint EANM/EANO/RANO/SNMMI practice guideline/procedure standards for diagnostics and therapy (theranostics) of meningiomas using radiolabeled somatostatin receptor ligands: version 1.0 (opens in a new tab)
- 2020EANM Practice Guideline/SNMMI Procedure Standard for Dopaminergic Imaging in Parkinsonian Syndromes 1.0 (opens in a new tab)
- 2018EANM‐EAN recommendations for the use of brain 18F‐Fluorodeoxyglucose Positron Emission Tomography (FDG‐PET) in neurodegenerative cognitive impairment and dementia: Delphi consensus (opens in a new tab)
- 2018Joint EANM/EANO/RANO practice guidelines/SNMMI procedure standards for imaging of gliomas using PET with radiolabelled amino acids and [18F]FDG: version 1.0. (opens in a new tab)
- 2016SNMMI Procedure Standard/EANM Practice Guideline for Amyloid PET Imaging of the Brain 1.0 (opens in a new tab)
- 2009EANM procedure guideline for brain perfusion SPECT using 99mTc-labelled radiopharmaceuticals, version 2 (opens in a new tab)
Finished the chapter?
This chapter has 26 questions. Solve them now to consolidate the topic.
You can also swipe sideways to change chapter