Table 3.

PET/SPECT studies in migraine patients with (MA) and without (MO) aura to detect an inflammatory phenotype

MO/MA studies which reported on dural uptake
Reference	Population (MA/MO)	Modality and Tracer	Assumptions of Technique	Delay after Migraine Onset	Comparisons	Results	Comments
Hadjikhani et al. [14] 2020 Annals of neurology	MA with visual aura N = 11 HC N = 11 Chronic lower back pain N = 11	PET/MRI w/ 11C-PBR28	11C-PBR28 ligand binds to mitochondrial receptor protein on many activated cell types during inflammation.	8 days median (0-18 range) 1 patient ictal during scan	MA with visual aura (interictally) vs. HC vs. Chronic lower back pain	• Uptake increased in MA in parameningeal tissue (meninges and skull bone) overlying occipital cortex • Uptake correlated with total episodes of visual aura in the preceding 4 weeks • Uptake unrelated to total attacks in the preceding 4 weeks	• 11C-PBR28 does not distinguish among cell types activated during inflammation • Total attack duration and intensity was not considered
Knotkova et al. [12] 2007 Pain Medicine and Pappagallo et al. [27] 1999 Neurology (Preliminary results)	MO (spontaneous) N = 2	SPECT w/ 99-Tc human serum albumin	99-Tc albumin to detect disrupted BBB	1st scan: 3 and 12 h after migraine onset 2nd scan: 3 h after 1st scan 3rd scan: Interictal	MO (ictal) vs. MO (interictal)	• Increased extravasation on the second scan at 3 h in frontotemporal and frontal regions ipsilateral to headache	• Preliminary results • Case reports
MO/MA studies which did not report on dural uptake
Reference	Population (MA/MO)	Modality and tracer	Assumptions of technique	Delay after migraine onset	Comparisons	Results	Comments
Albrecht et al. [13] 2019 Neurology	MA N = 13 HC N = 16	PET/MRI w/ 11C-PBR28	11C-PBR28 ligand binds to mitochondrial receptor protein on many activated cell types during inflammation, including microglia.	8.08 ± 5.03 days (mean ± SD)	MA (interictally) vs. HC	• Uptake increased in visual cortex, thalamus, primary and secondary somatosensory cortices, posterior insular cortex, primary motor cortex, auditory cortices, regions of prefrontal cortex, orbitofrontal cortex, putamen, area MT and V3A • Uptake correlated with frequency of migraine attacks in several cortical and subcortical areas	• 11C-PBR28 does not differentiate between cell types activated during inflammation
Schankin et al. [28] 2016 Brain	Mx N = 6 (4 MO, 2 MA) (GTN induced migraine attacks without aura) HC N = 6	PET/CT w/ 11C-DHE	11C-DHE to detect DHE passage of the BBB	Baseline scan and scan at 3 h after GTN infusion	Mx (Ictal) vs. Mx (interictal) HC (baseline) vs. HC (post-GTN)	• For Mx, no difference between ictal and interictal uptake • For HCs, no difference between baseline and post-GTN uptake • For all participants, uptake in choroid plexus of the lateral and 4th ventricles, pituitary fossa, venous sinus and facial tissue, and no uptake in cortical areas, brainstem, or thalamus	• Whether DHE passes a disrupted BBB is unexamined • Sample size may not have allowed detection of minor changes in BBB permeability
Sianard-Gainko et al. [29] 1993 Cephalalgia	MO N = 7 Cluster headache N = 30	SPECT w/ Gallium 67 citrate	Gallium 67 binds to proteins in inflammatory exudates [30].	NA	MO (interictal)Cluster headache (interictal, chronic and episodic in active period)	• High parasellar uptake in 5/7 MO patients • High parasellar uptake similar for MO and cluster headache	• Absence of quantitative data may have overestimated uptake • No HC group

¹¹C-DHE 11-carbon-dihydroergotamine, ¹¹C-PBR28 [O-methyl-¹¹C]-N-acetyl-N-(2-methoxybenzyl)-2-phenoxy-5-pyridinamine; ⁹⁹Tc 99-technetium, BBB blood-brain barrier, CT computed tomography, GTN glyceryl trinitrate, HC healthy control, MA migraine with aura, MO migraine without aura, Mx migraine with and without aura, NA not applicable, PET positron emission tomography, SPECT single-photon emission computerized tomography