From: Imaging the brain and vascular reactions to headache treatments: a systematic review
Reference | Population | Treatment | Imaging modality and timing of scans | Results | Limitations |
---|---|---|---|---|---|
Chugani et al., 1999 [28] | 11 female Mx (5 studied after treatment) 8 female HC | Type: Propranolol (40 to 180Â mg daily) or nadolol (20 to 40Â mg daily) Duration: 12Â weeks | Modality: [11C]AMT PET Time-point: Pretreatment and at follow-up after 12Â weeks of treatment | Baseline: Patients vs HC Whole brain serotonin synthesis was higher in patients compared to HC Baseline-to-follow-up: Patients No difference in whole brain serotonin synthesis from baseline compared to after 12Â weeks of treatment | Small sample size |
Hebestreit et al., 2017 [29] | 19 Mx (4 MA, 6 CM) 26 HC | Type: Metoprolol 75 mg for patients Metoprolol 75 mg or placebo for HC Duration: 2 months minimum | Modality: Task-based fMRI w/ noxious trigeminal stimulus and visual stimulation Time-point: Pretreatment and at follow-up after minimum 2 months of treatment for patients HC underwent only one MRI after 50 min from drug administration | Baseline-to-follow-up: Patients No brain functional changes after treatment with metoprolol ↑ Hypothalamic activity after metoprolol treatment (exploratory analysis) Negative correlation between hypothalamic activity and reduction of headache days on metoprolol Placebo vs Metoprolol in HC: No significant functional brain differences | Exploratory analysis was uncorrected for multiple comparisons Single metoprolol or placebo administration in HC No placebo-controlled design for patients |
Ahmed et al., 2022 [18] | 500 Mx (150 MA, 235 CM) | Type: Topiramate 2–200 mg Duration: At least 2 months Response definition: ≥ 50% reduction in the monthly headache days frequency compared to the baseline frequency | Modality: T2w MRI Time-point: Pretreatment | Baseline: Responders vs non-responders: Treatment responders less frequently had WMHs, had fewer and smaller WMHs | Only T2-w images were used to assess WMHs. FLAIR acquisition would also be recommended The same population was assessed for response to acute treatments. However, the association between patients’ response to acute and preventive medications have not been examined |
Li et al., 2018 [30] | 14 Mx (13 MO, 1 MA) | Type: Levetiracetam 500 mg daily Duration: 12 weeks | Modality:1H-MRS for GABA Time-point: Pretreatment and at follow-up after 12 weeks of treatment | Baseline-to-follow-up: ↓ GABA levels in PCC after treatment No change in ACC/PFC GABA levels | Data available for 11 patients for the PCC and 8 for the ACC/PFC No strong evidence for levetiracetam as a migraine preventive medication Small sample size |
Wöber et al., 1994 [31] | 11 Mx (6 responders, 5 non-responders) 21 HC | Type: Flunarizine 10 mg daily Duration: Ranged from 1 to 32 months Response definition: ≥ 50% reduction in migraine days | Modality:123I-Iodobenazamide SPECT Time-point: Single scan obtained after 1 to 32 months of treatment in patients HC underwent only one SPECT scan | Responders vs non-responders No difference in D2 receptor binding between responders and non-responders Patients vs HC: ↓ D2 receptor binding in flunarizine treated migraine patients compared to HC | Small sample size |
Dominguez et al., 2020 [32] | 62 CM (47 responders to botox, 15 non-responders) | Type: Botox 155 IU every 12 weeks (PREEMPT protocol) Duration: 12 weeks Response definition: ≥ 50% reduction in frequency of headache | Modality: T2w MRI Time-point: Pretreatment | Baseline: Responders vs non-responders ↑ Iron accumulation in the globus pallidus and periaqueductal gray matter in non-responders compared to responders. Significant only for PAG after adjustment for age Differences in iron PAG deposits were associated with poor response to botox after adjustment for clinical and biochemical variables | No quantitative assessment of T2-w signal as a marker of iron accumulation |
Hubbard et al., 2016 [33] | 23 CM (11 responder, 12 non-responders) | Type: Botox 150 IU every 12 weeks Duration: At least 12 weeks (at least 2 cycle of treatments) Response definition: ≥ 50% reduction in frequency of headache | Modality: T1w MRI and RS fMRI Time-point: Pretreatment | Baseline: Responders vs non-responders ↑ CTh of the right primary somatosensory cortex, anterior insula, left superior temporal gyrus and left inferior frontal gyrus (pars opercularis) in responders compared to non-responders ↓ FC between the right primary somatosensory cortex and left lateral occipital cortex and right dorsomedial prefrontal cortex, as well as between the left inferior frontal gyrus and left lateral occipital cortex and inferior supramarginal gyrus in responders vs non-responders | SBM not adjusted for gender Small sample size |
Ziegler et al., 2020 [34] | 27 Mx (9 MA, 18 MO; 15 CM, 12 EM; 9 responders, 8 non-responders) | Type: Erenumab 70 mg every 4 weeks Duration: 3 months Response definition: ≥ 30% reduction in headache frequency after 3-month treatment | Modality: Task-based fMRI w/ noxious trigeminal stimulus and ASL Time-point: Pretreatment and at follow-up after 2–3 weeks of treatment | Baseline-to-follow-up: ↓ Activation in the thalamus, cerebellum, and several pain processing cortical areas after erenumab treatment The absolute reduction in headache days was correlated with reduced activity of the right putamen, hypothalamus, cerebellum, and thalamus, from baseline to follow-up, in patients within the same migraine phase at the two MRI time points (17 patients) ↓ FC between the hypothalamus and bilateral temporal lobe, bilateral cerebellum, left hippocampus, parahippocampus, fusiform gyrus, nucleus ruber and spinal trigeminal nucleus. ↑ FC between the right hypothalamus and the right anterior insula after treatment with erenumab No changes in rCBF Baseline-to-follow-up: Responders vs non-responders: ↓ Activation in the cerebellum, insula and hypothalamus in responders compared to non-responders (analyses including only the subgroup of patients within same migraine phase: 9 responders vs 8 non-responders) | Field of view optimized for brainstem imaging, so areas above the thalamus were not included Heterogenous sample size: patients with CM and EM; patients with MA and MO; patients taking other preventive medications; 10 patients having headache either at baseline or follow-up visit Results uncorrected for multiple comparisons Small sample size of subgroups included in the analysis responders vs non-responders |
Basedau et al., 2022 [35] | Patients treated with galcanezumab: 26 Mx (11 CM, 15 EM; 7 MA, 19 MO; 8 responders, 7 non-responders) Patients treated with erenumab: 17 Mx (same cohort included in Ziegleret al., 2020 [34] | Type: Galcanezumab 240 mg s.c. single dose Erenumab 140 mg every 4 weeks Duration: 3 months Response definition: ≥ 30% reduction in headache frequency after 3-month treatment | Modality: Task-based fMRI w/ painful thermal stimulation of left forearm and RS fMRI Time-point: Pretreatment and at follow-up after 8 weeks of treatment | At follow-up: ↓ FC of the spinal trigeminal nucleus with the hypothalamus and superior temporal gyrus, and increased FC with the cerebellum, middle temporal gyrus, and insula Baseline-to-follow-up: ↓ Activation in the right hypothalamus, right cerebellum, and cerebellar vermis after 2–3 weeks of treatments Correlation between activation in the spinal trigeminal nucleus at baseline and the absolute reduction in headache days No changes in rCBF Baseline-to-follow-up: Responders vs non-responders: ↓ Activation of brain areas including the inferior parietal, precentral, parahippocampal cortex and cerebellum in responders compared to non-responders (analyses including only the subgroup of patients within same migraine phase: 8 responders vs 7 non-responders) Baseline-to-follow-up: erenumab vs galcanezumab: Compared to erenumab, galcanezumab treatment reduced the activity of the pons, right substantia nigra, left thalamus, and right hypothalamus Compared to galcanezumab, erenumab treatment reduced the activity of the insula, thalamus, cerebellum, hippocampus, lingual gyrus, frontal, parietal and temporal brain areas Analyses were of only the subgroup of patients within the same migraine phase: 15 galcanezumab vs 17 erenumab | Results uncorrected for multiple comparisons Heterogenous sample size: patients with CM and EM; patients with MA and MO; patients taking other preventive medications; 11 patients having headache either at baseline or follow-up visit Small sample size of subgroups included in the analysis responders vs non-responders and galcanezumab vs erenumab |
Schwedt et al., 2022 [36] | 32 Mx (21 CM, 11 EM; 18 responders, 14 non-responders) | Type: Erenumab 140 mg every 4 weeks Duration: 8 weeks Response definition: ≥ 50% reduction in the frequency of migraine days during week 5–8 of treatment | Modality: Task-based fMRI w/ painful thermal stimulation of left forearm and RS fMRI Time-point: Pretreatment and at follow-up after 8 weeks of treatment | Baseline: Responders vs non-responders ↓ Pain-induced response in the frontal supplemental motor region in responders compared to non-responders No RS FC differences Follow-up: Responders vs non-responders ↑ Pain-induced response in the left middle and posterior cingulate cortex, right putamen and periaqueductal gray matter in responders compared to non-responders ↑ Global network efficiency in responders compared to non-responders ↑ RS FC of the hypothalamus, fronto-parietal and temporal brain regions in responders compared to non-responders, as well as widespread differences between responders and non-responders in several graph theory metrics | Analyses uncorrected for multiple comparisons Heterogenous sample size: patients with CM and EM; 19 patients with MOH; patients taking other preventive medications; patients having headache either at baseline or follow-up visit No information regarding the presence of aura |
Peek et al., 2021 [37] | 18 CM (8 treated with botox and 10 treated with erenumab) | Type: Botox 155 IU every 3 months (PREEMPT protocol) Or Erenumab 70 or 140 mg s.c. monthly Duration: 3 months | Modality: MRS measuring GABA and Glx levels in ACC and PCC Time-point: Pretreatment and at follow-up after 3 months of treatment | Baseline-to-follow-up: In the mixed treatment group, increased GABA levels in the ACC correlated with decreased migraine frequency, HIT-6, and MIDAS scores after treatments In post-hoc analysis, greater increase in ACC GABA levels and decrease in headache frequency and HIT-6 scores in patients receiving erenumab compared to those treated with botox | Mixed treatment group which did not differentiate effects of botox and erenumab for all comparisons Small sample size of subgroups of patients |
Newman-Norlund et al., 2020 [38] | 12 CM w/ MOH | Type: Spheno-palatine ganglion block with nasal bupivacaine 0.5% Duration: 6 weeks (12 treatments) | Modality: T1w MRI Time-point: Pretreatment and at follow-up after 6 weeks of treatment | Baseline to follow-up ↑ Volume of left nucleus accumbens after treatment ↓ Volume of right hippocampus and pallidum, after treatment ↓ CTh of the left temporal pole and lateral occipital-temporal sulcus, after treatment | Statistical approach not described in detail |
Krebs et al., 2018 [39] | 10 CM w/ MOH | Type: Spheno-palatine ganglion block with nasal bupivacaine 0.5% Duration: 6 weeks (12 treatments) | Modality: RS fMRI Time-point: Pretreatment and at follow-up after 6 weeks of treatment | Baseline to follow-up Salience network: ↑ FC between the left anterior prefrontal cortex and bilateral orbitofrontal insula, ventral striatum, right supplementary motor area and dorsal prefrontal cortex ↑ FC between the right ventral tegmental area/substantia nigra and left dorsolateral prefrontal cortex and right temporal pole ↑ FC between the left superior temporal cortex and right supramarginal gyrus Executive network ↑ FC between the left dorsolateral prefrontal cortex and right dorsolateral prefrontal cortex, ventrolateral prefrontal cortex, right anterior thalamus, and caudate nucleus ↑ FC between the left dorsal prefrontal cortex and right anterior thalamus, and caudate nucleus | Unclear whether the study used an unadjusted t-test or a permutation test |