Skip to main content

Table 1 Detailed findings of pharmacological fMRI studies

From: Effects of pharmacological and nonpharmacological treatments on brain functional magnetic resonance imaging in Alzheimer’s disease and mild cognitive impairment: a critical review

Reference Treatment Design Sample fMRI protocol/scan timing Outcome measures Clinical findings Direction fMRI changes Brain areas involved Clinical-fMRI relationship
Bakker et al., 2015 [40] Levetiracetam (different doses: 62.5 mg twice/day, 125 mg twice/day, and 250 mg twice/day) and placebo RCT double-blind for patients and single-blind for HC 17 HC
54 MCI
Three-choice recognition memory task Pre-/post-treatment Task-related medial temporal, temporal-polar, and hippocampal functional activity changes; performance improvement at fMRI task and cognitive assessment Improvement on recognition memory task in the group on low-dose treatment. No changes at the BSRT, Verbal Pair Associate test, or BVRT Decreased Post-treatment vs placebo: L CA3 and DG of hippocampus Decreased activity; higher memory performance during task
Increased Post-treatment vs placebo:L entorhinal cortex
Bakker et al., 2012 [41] Levetiracetam (125 mg twice/day) and placebo RCT double-blind for patients and single-blind for HC 17 HC
17 MCI
Three-choice recognition memory task Pre-/post-treatment Task-related hippocampal functional activity changes; performance improvement at fMRI task and cognitive assessment Improvement on recognition memory task. No changes at the BSRT, Verbal Pair Associate test, and BVRT Decreased Post-treatment vs placebo: L CA3 and DG of hippocampus Decreased activity; higher memory performance during task
Bentley et al., 2008 [26] Physostigmine (infusion at a rate of 1 mg/1 h) and placebo (an equivalent volume of saline), in both groups 25 min prior to scan NRCT double-blind 17 HC
16 mild AD
Visuo-attentional task Post-treatment Task-related parietal functional activity changes; performance improvement at fMRI task Improvement on RT for the ‘deeper’ task in AD Increased Group X time, treated vs placebo: R precuneus and posterior parahippocampal cortex; R parietal and PFC
Decreased Group X time, treated vs placebo: R fusiform gyrus
Bentley et al., 2009 [27] Physostigmine (infusion at a rate of 1 mg/1 h) and placebo (an equivalent volume of saline), in both groups 25 min prior to scan NRCT double-blind 18 HC
13 mild AD
Face-encoding task Post-treatment Task-related fusiform functional activity changes and their relationship with performance improvement at fMRI task Task-independent (‘shallow’ vs ‘deeper’) improvement in confident memory Increased Group X time, treated vs placebo: Bilateral fusiform cortex Increased activity; higher face recognition post-scanning
Blautzik et al., 2016 [55] Galantamine (6-month treatment: 8 mg/day for the first month; 16 mg/day for the second month; 24 mg/day for the other months) or placebo, followed by 6 months galantamine (24 mg/day) – open label period RCT double-blind and open-label 11 HC
13 mild-moderate AD
RS fMRI At baseline At 6 months At 12 months DMN functional connectivity changes; performance improvement at cognitive assessment No changes at the CEREAD Increased Post-treatment vs HC (12-month follow-up):Posterior DMN (PCC, precuneus, L > R); Post-treatment vs placebo (12-month follow-up):Hippocampal sub-component (anterior division of hippocampus, R > L)
Bokde et al., 2016 [47] Rivastigmine (3-month treatment: 3 mg/day for the first month; 6 mg/day for the second month; 9 mg/day for the third month) or placebo, followed by 9 months rivastigmine (9 mg/day) – open label period RCT double-blind and open-label 12 MCI Face- and location-matching task At baseline At 3 months At 6 months Task-related whole-brain functional activity changes and performance improvement at cognitive assessment After 3 and 6 months: lower performances at the verbal fluency; stable performances at the CERAD and at the task Increased Pre-/post-treatment (3-month follow-up): Face-matching task: bilateral lingual and fusiform gyrus, L angular gyrus and cerebellum. Location matching task: L inferior temporal gyrus, R precuneus, R angular and inferior frontal gyri
Pre-/post-treatment (6-month follow-up): Location matching task: R inferior parietal and supramarginal gyrus, L precuneus and paracentral lobule, L medial frontal gyrus
Bokde et al., 2009 [32] Galantamine (3-month treatment: 8 mg/day for the first month; 16 mg/day for the second month; 24 mg/day for the last month) Case series 5 mild AD Face- and location-matching task Pre-/post-treatment Task-related ventral and dorsal visual pathway changes; performance improvement at fMRI task and cognitive assessment No changes at the task or at the CEREAD Decreased Pre-/post-treatment:Location-matching task: bilateral dorsal pathway (from occipital to parietal and frontal cortices)
Dhanjal et al., 2013 [29] Donepezil (6-week treatment: 5 mg/day for the first 2 weeks; 10 mg until the end of the study) Case series 9 mild AD Auditory sentence encoding and retrieval with auditory working memory suppressors Pre-/post-treatment Task-related primary auditory, ventro-lateral temporal, pars triangularis and angular gyri functional activity changes; performance improvement at fMRI task Increased percentage of retrieved trials during task Increased Pre-/post-treatment: L anterior ventral temporal cortex and pars triangularis
Dhanjal et al., 2014 [30] Donepezil (6-week treatment: 5 mg/day for the first 2 weeks; 10 mg until the end of the study) Cohort study 18 HC
18 mild AD
Auditory sentence encoding and retrieval with auditory working memory suppressors Pre-/post-treatment Task-related functional activity changes within the executive and salience networks; performance improvement at fMRI task Increased percentage of retrieved trials during task Increased Pre-/post-treatment: Fronto-parietal executive network: L lateral posterior parietal cortex and lateral frontal cortex. Higher-order cortex: L parahippocampal gyrus and anterior ventral temporal cortex
Goekoop et al., 2004 [39] Galantamine (oral intake, single dose: 8 mg; and after prolonged exposure: 4 mg day 1, 8 mg next 4 days, 4 mg on day 6). Washout period: 2 days Cross-over 28 MCI Episodic face-encoding and N-letter back task Pre-/post-treatment Task-related whole-brain functional activity changes N-letter back: task accuracy increased and latency decreased, mainly after single dose intake Increased Pre-/post-treatment (prolonged exposure): Face encoding: L middle frontal and occipital cortices, L posterior hippocampus and R anterior cingulate cortex. N-letter back: R precuneus and middle frontal cortex
Goekoop et al., 2006 [31] Galantamine (acute (8 mg) and prolonged 5 days exposure (4 mg the first day, 8 mg the following 4 days, 4 mg the last day)) Cross-over 18 mild AD
28 MCI
Face-recognition task Pre-/post-treatment Task-related whole-brain functional activity changes No changes at the task Increased Pre-/post-treatment (acute exposure): MCI: L PCC, anterior and temporal lobe, L superior parietal, R frontal lobe and cerebellum. AD: vermis of cerebellum, R inferior temporal and parahippocampal gyri
Decreased Pre-/post-treatment (prolonged exposure): MCI: bilateral superior frontal cortices, L PCC, R middle frontal gyrus. AD: R parahippocampal cortex
Goveas et al., 2011 [50] Donepezil (3-month treatment: 5 mg/day for 4 weeks; 10 mg/day until the end of the study) Cohort study 14 HC
18 mild AD
RS fMRI, seed-based (hippocampus) connectivityPre-/post-treatment Hippocampal functional connectivity changes; performance improvement at cognitive assessment Improvement on ADAS-cog but not on MMSE Increased Pre-/post-treatment: Positively correlated hippocampal functional connectivity network: L middle frontal and precentral gyri, L parahippocampus, insula and thalamus, R PCC Increased hippocampal connectivity strength in the L dorsolateral PFC and middle frontal gyrus; improvement on ADAS-cog
Decreased Pre-/post-treatment: Negatively correlated hippocampal functional connectivity network: L inferior parietal cortex/supramarginal gyrus, L posterior middle temporal gyrus, and R dorsolateral PFC
Griffanti et al., 2016 [52] Donepezil (12-week treatment: 5 mg/day for the first 4 weeks, followed by 10 mg/day until the end of the study) Case series 18 mild-moderate AD RS fMRIPre-/post-treatment Relationship between whole-brain functional connectivity changes and performance improvement at cognitive assessment Greater improvement on MMSE and MoCA in responders compared to nonresponders Increased Pre-/post-treatment: Orbitofrontal network: precuneus, PCC and R dorsolateral frontal cortex (responders > nonresponders) Increased connectivity of anterior and posterior cingulate cortices, precuneus, and R dorsolateral frontal regions within the orbitofrontal network; improvement on MoCA
Grön et al., 2006 [38] Galantamine (4 mg twice a day for 7 days) Case series 10 MCI Spatial navigation taskPre-/post-treatment Task-related hippocampal functional activity changes and performance improvement at cognitive assessment Improvement on verbal episodic memory but not at the task Increased Pre-/post-treatment: R middle occipital and temporal gyri, R PCC, R hippocampus and parahippocampal gyrus; L anterior hippocampus
Haller et al., 2014 [45] Caffeine (one capsule containing caffeine 200 mg or placebo) 30 min before testing NRCT double-blind 15 HC
13 MCI
2-back (vs 0-back) working memory taskPre-/post-treatment Task-related whole-brain functional activity changes No effect on task RT neither on accuracy Increased Post-treatment vs placebo: Task-related: bilateral striatum, temporal and parietal cortices. TICA: L working memory network including PFC, supplementary motor area, ventral premotor and parietal cortices
Kircher et al., 2005 [28] Donepezil (10-week treatment: 5 mg/day for the first 4 weeks; 10 mg/day until the end of the study) Cohort study 10 HC
10 mild-moderate AD
Face memory encoding taskPre-/post-treatment Task-related fusiform gyrus functional activity Improvement on ADAS-cog total score and on the memory subscale. No changes at the task Increased Pre-/post-treatment/Post-treatment vs HC: R fusiform gyrus
Li et al., 2012 [51] Donepezil (12-week treatment: 5 mg/day for the first 4 weeks; 10 mg/day until the end of the study) Case series 12 mild AD RS fMRI, seed based (MCC and PCC) connectivityPre-/post-treatment MCC and PCC functional connectivity changes; cerebral blood flow changes; performance improvement at cognitive assessment Improvement on ADAS-cog but not on MMSE, NPI, or IADL Increased Pre-/post-treatment: Middle cingulate and PCC network connectivity Increased connectivity between the middle cingulate cortex and the ventral anterior cingulate cortex and PFC; and between the PCC and the ventral anterior cingulate cortex-changes in ADAS-cog
Lorenzi et al., 2011 [56] Memantine (6-month treatment: 5 mg/day, increasing by 5 mg/day to a final dose of 20 mg/day) or placebo RCT double-blind 15 moderate AD RS fMRI Pre-/post-treatment DMN functional connectivity changes; performance improvement at cognitive assessment No changes at the cognitive assessment Increased Pre-/post-treatment/Group X time, treated vs placebo: R precuneus and calcarine gyrus within DMN
McGeown et al., 2010 [36] Donepezil (20-week treatment: 10 mg/day) Cohort study 9 HC
12 mild AD
Semantic association and N-back (1-back) task Pre-/post-treatment Task-related whole-brain functional activity changes; performance improvement at fMRI task No changes at the cognitive assessment (including ADAS-cog, NPI and ADL) or at the task Decreased Pre-/post-treatment: Semantic association: L superior parietal, middle temporal, medial and inferior frontal gyrus, and R superior temporal gyrus. Working memory: L caudate, L middle and superior temporal gyri, and R inferior frontal gyrus. Post-treatment vs HC: Semantic association: bilateral middle frontal gyrus, R superior occipital, cuneus and anterior cingulate cortex. Working memory: L thalamus, L parahippocampal gyrus, R inferior frontal gyrus Increased activity in non-task relevant regions (such as bilateral inferior parietal lobe, PCC and precuneus); higher accuracy at the semantic association task
McGeown et al., 2008 [34] Rivastigmine (20-week treatment: 6 mg twice/day) Cohort study 9 HC
11 mild AD
Semantic association and N-back (1-back) task Pre-/post-treatment Task-related whole-brain functional activity changes; performance improvement at cognitive assessment Improvement on ADAS-cog. No further changes at the cognitive assessment or at the task Increased Pre-/post-treatment: Semantic association: bilateral middle frontal and paracentral gyri, parahippocampal and fusiform gyri. Working memory: R superior, middle, medial and inferior frontal gyrus, and R precentral gyrus. Post-treatment vs HC: Semantic association: R inferior frontal and L anterior cingulate cortex. Working memory: R middle frontal, postcentral and supramarginal gyri
Decreased Pre-/post-treatment: Working memory: L middle frontal, precentral and cingulate gyrus, L insula and thalamus. Post-treatment vs HC: Working memory: L PCC and angular gyrus
Miettinen et al., 2011 [24] A single oral dose of rivastigmine (3 mg, acute); and 1.5 mg of rivastigmine twice a day for 4 weeks (chronic); a single oral dose of placebo NRCT double-blind 20 mild AD Face recognition memory task Post-treatment Task-related whole-brain functional activity changes and their relationship with baseline cognitive assessment No changes at the task Increased Post-treatment vs placebo (acute): bilateral PFC, R middle and superior temporal gyrus.
Post-treatment vs placebo (chronic): bilateral PFC, L middle temporal and anterior cingulate cortices, and L parietal gyrus
Increased PFC activity after chronic treatment; poorer MMSE at baseline
Pa et al., 2013 [42] Donepezil (3-month treatment: 5 mg/day for 1 month and 10 mg/day for 2 months) or placebo RCT double-blind 27 MCI Face recognition task Pre-/post-treatment Task-related prefrontal, parietal and hippocampal functional activity changes; performance improvement at cognitive assessment Improvement on task RT and accuracy. No changes at the cognitive assessment Increased Group X time, treated vs placebo: L fusiform face area and its connectivity with R hippocampus and inferior frontal junction Increased connectivity between L fusiform face and R hippocampus; reduced RT for face recognition in treated patients
Petrella et al., 2009 [44] Donepezil (12- or 24- week treatment: 5 mg/day for 42 days, followed by 10 mg/day until the end of the study) RCT double-blind 13 MCI Novel face visual memory task Pre-/post-treatment Task-related whole-brain functional activity changes; performance improvement at cognitive assessment No improvement at the cognitive assessment or at the task Increased Post-treatment vs placebo: Bilateral dorsal e ventrolateral PFC.
Group X time,treated vs placebo: L inferior frontal gyrus
Risacher et al., 2013 [43] Donepezil (3-month treatment: 5 mg/day for 4 weeks, 10 mg/day until the end of study) NRCT open-label 20 HC
18 MCI
Verbal episodic encoding task Pre-/post-treatment Task-related whole-brain functional activity changes and their relationship with patient performances at cognitive assessment before and after treatment Improvement on CVLT. Mild accuracy decline during task Increased Group X time/treated vs HC: R hippocampus and parahippocampal gyrus, R middle frontal gyrus. Increased deactivation of the medial parietal lobe Changes on medial parietal lobe activity-changes in CVLT. Increased connectivity of the L frontal lobe and L caudate; improved task accuracy
Rombouts et al., 2002 [25] Single dose (3 mg) of rivastigmine, 3 h before the first vs the second scanning NRCT single-blind 11 mild AD Face encoding and working memory task Pre-/post-treatment Task-related whole-brain functional activity changes No changes at the task Increased Post-treatment vs placebo: Face encoding: bilateral fusiform gyrus. Simple working memory: L middle and superior frontal gyrus. Increased working memory load: L middle frontal gyrus, R inferior and superior frontal gyrus.
Decreased Post-treatment vs placebo: Increased working memory load: R middle and superior frontal gyrus
Saykin et al., 2004 [46] Donepezil (5 mg/day for 4 weeks; 10 mg/day for 5.67 ± 1.66 weeks on average) NRCT open-label 9 HC
9 MCI
Auditory N-back task Pre-/post-treatment Task-related whole-brain functional activity changes; performance improvement at cognitive assessment and fMRI task Improvement on accuracy during task and on TMT-B. Reduction of subjective cognitive concerns Increased Group X time, treated vs HC: L dorsolateral PFC and L superior frontal cortex Increased activity of the L anterior prefrontal-improved task accuracy
Shanks et al., 2007 [35] Galantamine (20-week treatment: 16 mg twice/day) Cohort study 9 HC
9 mild AD
Semantic association and target detection task Pre-/post treatment Task-related frontal and parieto-temporal functional activity changes No improvement at the cognitive assessment or at the tasks. Increased awareness in patient self-assessment with respect to problems during daily activities Increased Pre-/post-treatment: Semantic association: L paracentral lobule, L caudate and R lingual gyrus. Target detection: bilateral postcentral, L inferior parietal lobule.
Post-treatment vs HC: Semantic association: bilateral superior temporal gyri and insula, R medial frontal gyrus, L inferior frontal. Target detection: bilateral middle frontal, L superior temporal and precuneus
Solé-Padullés et al., 2013 [49] Donepezil (3-month treatment: 5 mg/day for 1 month and 10 mg/day for 2 months) or no treatment RCT single-blind 15 mild-moderate AD RS fMRI andvisual scene encoding task Pre-/post-treatment RS whole-brain functional connectivity and task-related activity changes; performance improvement at fMRI task Improvement on semantic fluency. No further changes at the cognitive assessment or at the task Increased Post-treatment vs untreated: R parahippocampal gyrus within the DMN. No task-related changes were observed
Thiyagesh et al., 2010 [33] Donepezil (23-week treatment: 5 mg/day) Cohort study 11 HC
10 mild AD
Visuospatial tasks Pre-/Post-treatment Task-related functional activity changes in brain regions subtending visuospatial abilities Improvement on MMSE, ADAS-cog, and Present Functioning Questionnaire. No changes at the task Increased Pre-/post-treatment: L precuneus Increased activity of the L precuneus-improvement atthe Present Functioning Questionnaire
Venneri et al., 2009 [37] AchEI treatment (20-week treatment: at the maximum guideline-recommended dosage) Cohort study 9 HC
26 mild AD
Semantic association and N-back (1-back) task Pre-/post-treatment Task-related whole-brain functional activity changes and performance improvement at cognitive assessment in responders compared with nonresponders Improvement of the responders on ADAS-cog. No further changes at the cognitive assessment or at the task Increased Pre-/post-treatment/Group X time, responders vs nonresponders: Semantic association: bilateral inferior and medial frontal gyri, L precentral and postcentral gyri, L insula, middle temporal and inferior parietal gyri and anterior cingulate cortex; R inferior temporal gyrus, precuneus and caudate. Working memory: R precentral, precuneus, inferior parietal and thalamus, L inferior and superior frontal gyrus Increased activity of the L frontal cortex during the semantic association task; poorer performance at the baseline semantic fluency
Wang et al., 2014 [54] Stable dose of AchEIs (donepezil, rivastigmine, or galantamine) for at least 15 days and for almost 18 months Case-control 25 mild treated AD
19 mild untreated AD
RS fMRI Post-treatment Functional connectivity changes and interaction with the APOE genotype Increased Pre-/post-treatment/ApoEε4 treated vs ApoEε4 untreated: Greater composite scores in dorsal attention, control and salience networks
Zaidel et al., 2012 [53] Donepezil (8-week treatment: 5 mg/day for 28 days; 10 mg/day until the end of the study) Case series 11 mild AD RS fMRI, L hemisphere seed-based connectivity Pre-/post-treatment RS functional changes in the interhemispheric connectivity Increased Pre-/post-treatment: L-R dorsolateral PFC
Zhang et al., 2016 [57] Bushen capsule (24-month treatment: 4 capsules 3 times a day) or placebo RCT double-blind 60 MCI RS fMRI At baseline At 12 months At 24 months DMN functional connectivity and performance improvement at cognitive assessment, and their relationship Improvement on MMSE, RAVLT, and digit span Increased Group X time, treated vs placebo:R precuneus within the DMN No relationship was observed between connectivity and cognitive changes
Zhang et al., 2014 [48] CCRC (3-month treatment: 3 capsules per day) or placebo RCT double-blind 39 MCI N-back (0-1-and 2 back) working-memory task Pre-/Post-treatment Task-related whole-brain functional activity changes; performance improvement at cognitive assessment Improvement on MMSE and digit span. No further changes on other cognitive scores or on task Increased Group X time, treated vs placebo and vs HC: Increased negative activation of L PCC and R fusiform gyrus Increased negative activity in L PCC; changes on MMSE and digit span scores
  1. AchEI acetyl-cholinesterase inhibitor, AD Alzheimer’s disease, ADAS-cog Alzheimer's Disease Assessment Scale-cognitive subscale, ADL activities of daily living, APOE apolipoprotein E, BSRT Buschke Selective Reminding Test, BVRT Benton Visual Retention Test, CCRC Compound Congrongyizhi Capsule, CEREAD Consortium to Establish a Registry for Alzheimer's Disease, CVLT California Verbal Learning Test, DG dentate gyrus, DMN default mode network, fMRI functional MRI, HC healthy controls, IADL instrumental activities of daily living, L left, MCC middle cingulate cortex, MCI mild cognitive impairment, MMSE Mini mental state examination, MoCA The Montreal Cognitive Assessment, NPI Neuropsychiatric Inventory, NRCT nonrandomized controlled trial, PCC posterior cingulate cortex, PFC prefrontal cortex, R right, RAVLT Rey auditory verbal learning test, RCT randomized controlled trial, RS fMRI resting state functional MRI, RT reaction time, shallow low-demanding, TICA tensorial-independent component analysis, TMT-B Trail Making Test, part B