Quantitative Assessment of Cortical Excitability in Alzheimer’s Dementia and Its Association with Clinical Symptoms: A Systematic Review and Meta-Analyses

Abstract

Background:

Alzheimer’s disease (AD) is characterized by cognitive and neuropsychiatric symptoms (NPS) due to underlying neurodegenerative pathology. Some studies using electroencephalography (EEG) have shown increased epileptiform and epileptic activity in AD.

Objective:

This review and meta-analyses aims to synthesize the existing evidence for quantitative abnormalities of cortical excitability in AD and their relationship with clinical symptoms.

Methods:

We systematically searched and reviewed publications that quantitatively assessed cortical excitability, using transcranial magnetic stimulation (TMS) resting motor threshold (rMT), active motor threshold (aMT), motor evoked potential (MEP) or directly from the cortex using TMS-EEG via TMS-evoked potential (TEP). We meta-analyzed studies that assessed rMT and aMT using random effects model.

Results:

We identified 895 publications out of which 37 were included in the qualitative review and 30 studies using rMT or aMT were included in the meta-analyses. The AD group had reduced rMT (Hedges’ g = –0.99, 95% CI [–1.29, –0.68], p < 0.00001) and aMT (Hedges’ g = –0.87, 95% CI [–1.50, –0.24], p < 0.00001) as compared with control groups, indicative of higher cortical excitability. Qualitative review found some evidence of increased MEP amplitude, whereas findings related to TEP were inconsistent. There was some evidence supporting an inverse association between cortical excitability and global cognition. No publications reported on the relationship between cortical excitability and NPS.

Conclusion:

There is strong evidence of increased motor cortex excitability in AD and some evidence of an inverse association between excitability and cognition. Future studies should assess cortical excitability from non-motor areas using TMS-EEG and examine its relationship with cognition and NPS.

Keywords

Alzheimer’s disease cognition cortical excitability electrophysiology neuropsychiatric symptoms transcranial magnetic stimulation

INTRODUCTION

Alzheimer’s disease (AD) is characterized by progressive deterioration of cognition and the presence of neuropsychiatric symptoms (NPS) [1 –3]. Affected individuals display deficits in one or more cognitive domain that lead to impaired independence of activities of daily living [4]. In addition, many individuals diagnosed with AD develop behavioral disturbances, known as NPS [3, 5]. The NPS happen at all stages of cognitive impairment and may include depression, anxiety, agitation, aggression, irritability, delusions, hallucinations, and sleep disturbances [6]. Further, the presence of NPS in the earlier stages of the disease predicts faster progression of cognitive decline; however, the underlying pathophysiology is not well understood [7].

Current framework for understanding the neuropathological changes of AD is based on the categorical classification of positive biomarkers known as the AT(N) model [8, 9]. The “A” refers to amyloid-β (Aβ) pathology, “T” refers to tau pathology (phosphorylated-tau or p-tau), and “N” refers to neurodegeneration [9]. While the biomarkers of AD have been studied extensively in the context of cognitive impairment, recent evidence also supports their association with NPS [10, 11]. One of the challenges with AT(N) framework is lower specificity to predict clinical illness [12 –15]. Further, these biomarkers may not be cost effective (structural and functional imaging) and may involve invasive procedures [16, 17]. Therefore, biomarkers that provide a measure of neuronal function through non-invasive and cost-effective techniques may be valuable as an add on to the AT(N) model. Neurophysiological markers can be studied non-invasively using techniques such as electroencephalography (EEG) or magnetoencephalography at rest or during a task [18 –22]. Further, non-invasive brain stimulation techniques such as transcranial magnetic stimulation (TMS) combined with electromyography (EMG) or EEG can be used to study neuroplasticity, and cortical excitability and inhibition in vivo, in healthy or disease states [23, 24]. Studies using resting EEG have found increased power in slower -wave delta and theta bands and reduced power in faster-wave alpha and beta bands [25 –32]. These electrophysiological changes correlate with neuropathological abnormalities and clinical symptoms in AD [33, 34]. Cerebrospinal fluid (CSF) Aβ₄₂ levels negatively correlated with slow wave delta and theta EEG activity, while CSF p-tau inversely correlated with faster wave alpha and beta EEG activity [18]. Further, changes in EEG power have been linked with cognitive deficits in memory, attention, and language domains [35 –37].

There is also evidence supporting a state of cortical hyper-excitability in AD. Studies in AD mouse models have shown frequent epileptiform activity (sharp waves and spikes) as well as spontaneous nonconvulsive seizure activity in the neocortex and hippocampal regions [38 –40]. Some evidence indicates recurrent epileptiform discharges or epileptic activity could increase the spread of AD pathology [41, 42]. Studies in clinical populations with AD have also shown increased subclinical epileptiform and seizure activity [43, 44] such that epidemiology studies have reported unprovoked seizures in roughly 10–22% of patients with AD. Epileptiform discharges and seizures may hasten the progression of AD and have been associated with reduced survival [45, 46]. Over time, patients with AD who have displayed subclinical epileptiform demonstrate faster decline on the MMSE and executive functioning [47]. Therefore, the worsening of AD and its clinical symptoms may be partially ascribed to network hyper-excitability leading to epileptiform and epileptic activity.

Several studies examined changes in quantitative neurophysiological parameters (e.g., TMS resting and active motor threshold, motor evoked potential (MEP) amplitude) in patients with AD. TMS resting motor threshold (hereafter referred to as rMT) is quantitatively defined as the lowest stimulus intensity (as a percentage of maximum stimulator output) applied to the scalp that would elicit a response in a relaxed target muscle [48]. There are three commonly used methods to assess rMT. The relative frequency method requires stimulus intensity to be increased in steps of 5% until MEPs > 50μV, are consistently evoked in each trial [49]. Following this, stimulus intensity is gradually decreased by 1% until 50% of ten consecutive trials elicit MEPs > 50μV [49]. Adaptive threshold hunting methods use a software program (such as Parameter Estimation by Sequential Testing and Maximum Likelihood regression) to estimate the stimulus intensity that would have a 50% probability of eliciting an MEP [50, 51]. Another method, the Two Threshold Method, determines rMT by calculating the arithmetic mean of an inferior (highest stimulus intensity with zero probability of evoking MEP) and superior threshold (lowest stimulus intensity with a 100% probability of evoking MEP) [52]. Active motor threshold (aMT) is defined as the minimum stimulus intensity required to elicit an EMG response in the target muscle when contracted at 20% of maximal muscle strength [49] and may be assessed using previously discussed methods for assessing rMT [49]. Higher rMT or aMT indicate lower cortical and corticospinal excitability and lower rMT or aMT indicate higher cortical and corticospinal excitability [53]. Motor evoked potential (MEP) is defined as the brief activation of peripheral muscles elicited by cortico-motor inputs of the stimulated region, recorded with EMG [48, 54]. As opposed to motor threshold, higher MEP amplitude is indicative of higher cortical or corticospinal excitability and vice versa [53]. Cortical excitability can also be assessed directly from the scalp (from both motor and non-motor areas) using a combination of TMS and EEG using TMS-evoked potential (TEP) and calculating amplitude of an individual TEP peak or area under the TEP curve [55, 56]. While some studies have found evidence of cortical hyper-excitability in AD using these parameters [57, 58] others failed to replicate these findings [59, 60]. Further, the relationship between cortical excitability, cognition and NPS in AD is unclear [61, 62]. Few review papers have provided an overview of TMS studies assessing cortical reactivity and changes in plasticity of neuronal networks in AD [58 , 63–65]. These reviews address neurophysiology measures holistically, however, lack in-depth assessment of cortical excitability and its association with clinical symptoms of AD. Further, these reviews did not take into consideration methodological heterogeneity of the included studies.

The objectives of this systematic review and meta-analyses are to assess 1) the current state of evidence regarding cortical excitability changes in AD when compared to healthy individuals, and 2) the association of cortical excitability with cognition and NPS in AD.

METHODS AND MATERIALS

Search strategy

This systematic review and meta-analyses was conducted in accordance with PRISMA guidelines. A comprehensive literature search of EMBASE, MEDLINE, and PsycINFO databases was performed using the following terms and connectors: (Alzheimer’s or dementia or major neurocognitive disorder) AND (cortical excitability or cortical excitation or neuronal activation or cortical activation or cortical evoked activity or epileptiform activity or epileptic activity or motor evoked potential or TMS-evoked potential or event related potential or global mean field power or local mean field power) AND (TMS or transcranial magnetic stimulation or EEG or electroencephalography or EMG or electromyography or MEG or magnetoencephalography) OR (cognitive impairment or cognitive decline) OR (neuropsychiatric symptoms or behavioural symptoms or behavioural and psychological symptoms or behavioural abnormalities or agitation or aggression or irritability or restlessness or depression or anxiety or apathy or delusions or hallucinations or sleep disturbances or aberrant motor behaviour or disinhibition or psychosis or nighttime behavioural disturbance or elation or euphoria or lability or indifference). Titles and abstracts were reviewed by two authors (SJ and RP) to select the relevant articles for full-text review. Differences were resolved by consensus and consultations with the senior author (SK). Additional studies were identified using references from included studies and other relevant literature. Following criteria were applied to select articles for inclusion in the final review: 1) English-language, 2) original research articles, 3) included a population with AD, 4) quantitatively assessed cortical excitability, and 5) compared cortical excitability between AD and healthy cohort or assessed its relationship with cognition or neuropsychiatric symptoms. Studies were excluded if they were non peer-reviewed (such as conference posters or abstracts) or review articles.

Included studies were assessed for quality using National Institute of Health (NIH) study quality assessment criteria [66]. Quality assessment was evaluated on twelve criterion items using a three-point scale (yes, no, or other). Assessments were conducted by two authors independently (SJ and RP) and final decisions were determined by consensus. Articles were categorized into one of three study quality ratings: good, fair, or poor. Publications were deemed of good quality if most NIH quality assessment criteria were fulfilled.

Data analyses

Random effects meta-analyses were performed using Reviewer Manager Version 5.4 (RevMan5, The Cochrane Collaboration, 2020). First, we performed a meta-analysis of all studies comparing rMT in patients with AD and healthy controls. Several studies followed the internationally established relative frequency method to assess rMT [49], other studies employed modification to this procedure or did not specify exact methodology. We separated studies into two groups based on their techniques (relative frequency method or modified and unspecified methods) used to determine rMT, to see if the methodological heterogeneity impacted the results of the meta-analysis. Finally, we meta-analyzed all studies comparing aMT in patients with AD and healthy controls. Data was extracted from cross-sectional studies and only baseline data was used from longitudinal studies. Motor thresholds were presented as mean intensity percentages with standard deviation (SD). Outcome measure calculated was standard mean difference with 95% confidence intervals (CIs). Hedges’ g [67], a suitable standardized mean difference measure for studies of small sample, was calculated for each study using mean, standard deviation and sample sizes of each group. A weighted average was generated using weights based on sample size of each study. Effect sizes representing the difference in the respective measures between patients with AD and healthy controls were expressed using Hedges’ adjusted g standardized mean difference. Negative values were indicative of reduced rMT or aMT.

Heterogeneity among studies was assessed using tau squared (τ²) and Cochran’s Q test, with a p-value < 0.01 indicative of heterogeneity [68]. Further, we calculated I² statistic to quantify the degree of heterogeneity across studies in a percentage estimate [69]. An I² value of 25%, 50% and 75% indicates low, moderate, and high degrees of heterogeneity respectively.

RESULTS

Literature search

As shown in Fig. 1, the search retrieved 878 publications and 25 publications were found from other sources (e.g., review articles and references of other articles). After a total of 230 duplicates were removed, the remaining 673 publications were screened for eligibility. In the first step, titles and abstracts were screened and 552 publications were removed based on title and 53 articles were removed based on abstracts. The remaining publications (n = 68) were reviewed in full text for inclusion and exclusion criteria. Finally, 37 publications were included, out of which 33 were assessed for inclusion in the meta-analyses and eight publications investigated the relationship between cortical excitability and global cognition. No publications examined the association between cortical excitability and NPS in AD.

Fig. 1

PRISMA diagram of the search and selection process for inclusion in the systematic review and meta-analyses.

Characteristics of included studies

All publications reported on clinical AD samples; however, three publications confirmed disease pathology via CSF biomarkers or PET imaging [70 –72]. There were differences in disease severity across the studies. Most publications reported on mild to moderate AD [57 , 73–87], out of which twelve reported reduced rMT and six reported no difference. Four publications reported on mild AD [60 , 88–90], out of which three reported no difference in rMT and one reported increased aMT. Six publications reported on only moderate AD [70 , 91–95], five reported on the mild to severe AD [62 , 96–98] and one reported on moderate to severe AD [99], out of which all reported reduced rMT. Two publications did not report on disease severity of AD samples [72, 100]; however, both showed reduced rMT. Thirty-two publications reported on TMS combined with electromyography (EMG) and five reported on TMS combined with EEG (TMS-EEG) to assess cortical excitability. Quantitative measures of cortical excitability reported across publications were rMT in thirty-three publications, aMT in eleven publications, MEP amplitude in nine publications, and TEP amplitude in three publications. Data from 30 studies using rMT and seven studies using aMT were available for meta-analyses.

Fourteen publications reported on TMS methods to assess rMT following established international standards [49]. Thirty-three studies matched study groups according to age and/or gender and examined levels of education and cognition (e.g., MMSE). Nineteen studies excluded participants taking medications that could interfere with TMS-induced responses. None of the included publications reported sample size justification or performed power analysis. For specific features and quality ratings of individual studies, please see Table 1.

Table 1

Table summarizing findings from studies of cortical excitability and its relationship with clinical symptoms in AD

Authors, Year [ref]	Sample size	Demographics age (y); Gender (Males (M)/ Females (F))	Cognitive assessments	Cortical excitability measures	Main finding(s)	Additional comments	Study quality rating
Khedr et al., 2020 [97]	AD = 15 HC = 25	AD: Mean (SD) Age = 65.7 (5.1); Gender: M = 6, F = 9 HC: Mean (SD) Age = 60.7 (6.8); Gender: M = 13, F = 12	Modified MMSE: AD Mean (SD) = 49.7 (17.8) HC Mean (SD) = 84.5 (12.5)	1. rMT 2. aMT	1. Reduction in rMT and aMT AD as compared with HC	No differences in cortical excitability were found between the three dementia groups (PD, VaD, AD).	Good
Bagattini et al., 2019 [73]	AD = 26	AD: Mean (SD) Age = 77.6 (4.8) Age Range = 64–84; Gender: M = 3, F = 23	MMSE: Mean (SD) = 20.8 (2.3) Range = 17–26	1. TEP amplitude (in time windows: 15–25ms, 25–35ms, 45–55ms and 60–80 ms) 2. Area under the curve (AUC) of TEPs	1. P30 amplitude was predictive of the MMSE score in AD patients (F (1,21) = 5.19, p = 0.03, R² = 0.20) 2. P30 showed good predictive performance in discriminating between mild and moderate AD patients (AUC = 0.78) 3. P30 predicted performance in face name association memory task (FNMT) (F (1,16₎ = 5.40, p = 0.03, R² = 0.25)	P50 and P70, localized in the superior and inferior frontal gyrus respectively, showed no associated with disease severity.	Good
Motta et al., 2018 [72]	AD = 60 HC = 30	AD: Mean (SD) Age = 68.2 (5.9); Gender: M = 7, F = 12 HC: Mean (SD) Age = 69.2 (3.7); Gender: M = 9, F = 10	MMSE: Not available Rey auditory verbal long-term memory Rey Complex Figure Raven’s Colored Progressive Matrices	1. rMT	1. Reduced motor threshold in AD as compared with HC.	This study also assessed SAI to evaluate LTP-like cortical plasticity and found reduced SAI in AD. LTP-like cortical plasticity was a significant predictor of disease progression and cognitive decline.	Good
Benussi et al., 2018 [74]	AD = 63 HC = 39	AD: Mean (SD) Age = 78.1 (7.8); Gender: M = 31, F = 32 HC: Mean (SD) Age = 68.6 (8.1); Gender: M = 13, F = 26	MMSE: AD Mean (SD) = 20.4 (6.2) HC: not available BADL: AD Mean (SD) = 1.2 (1.6) HC: not available IADL: AD Mean (SD) = 3.0 (2.7) HC: not available	1. rMT	1. No difference in rMT between groups.	This study assessed whether TMS parameters could be used to classify atypical parkinsonian disorders and AD. Results showed impairment in SAI of AD and DLB patients, and SICI in DLP, PSPO and CBS patients.	Good
Benussi et al., 2017 [71]	AD = 79 HC = 32	AD: Mean (SD) Age = 71.2 (8.1); Gender: M = 40, F = 39 HC: Mean (SD) Age = 61.5 (10.3); Gender: M = 14, F = 18	CDR: AD Mean (SD) = 0.9 (0.6)	1. rMT	1. Reduced rMT found in AD as compared with HC.	This study also assessed SICI, LICI and SAI to measure intracortical circuit. AD and FTD groups were distinguishable in SICI and SAI parameters.	Good
Wang et al., 2016 [61]	AD = 27	AD: Mean (SD) Age = 72.07 (1.62); Gender: M = 12, F = 15	MMSE: AD Mean (SD) = 19.85 (0.73) MoCA: AD Mean (SD) = 13.93 (0.90)	1. TEP P300 Amplitude 2. rMT 3. aMT	1. rMT (t = –2.818, p = 0.007) and aMT (t = –3.548, p = 0.001) lower in AD group than the bvFTD 2. No correlation between P300 amplitude and cognition in AD. 3. No correlation between motor cortex function and cognition in AD.	This study also assessed CSP. No difference found in CSP between AD and bvFTD groups.	Good
Ferreri et al., 2016 [75]	AD = 12 HC = 12	AD: Mean (SD) Age = 72.4 (5.9); Gender: M = 5, F = 7 HC: Mean (SD) Age = 68.6 (7.1); Gender: M = 6, F = 6	MMSE: AD Mean (SD) = 20.8 (2.7) HC Mean (SD) = 28.8 (1.8)	1. rMT 2. MEP Amplitudes 3. TEP Amplitudes 4. Global Mean Field Power (GMFP)	1. No difference in rMT and MEP amplitude between AD and HC 2. GMFP analysis found group difference at TEP P30 and P60 wave (p < 0.05)	Analysis via sLORETTA found stronger cortical activation in AD as compared to HC, in which responses remained localized in the sensorimotor cortex.	Good
Yang et al., 2014 [76]	AD = 40 HC = 45	AD: Mean (SD) Age = 66.4 (4.9); Gender: M = 17, F = 23 HC: Mean (SD) Age = 66.8 (0.7); Gender: M = 25, F = 20	MMSE: AD Mean (SD) = 16.5 (8.1) HC Mean (SD) = 28.4 (4.2) ADL: AD Mean (SD) = 29.2 (12) HC Mean (SD) = 17.5 (5.6)	1. rMT	1. Reduced rMT in AD as compared with HC.	This study assessed spinal	Good
Balla et al., 2014 [90]	AD = 12 HC = 13	AD: Mean (SD) Age = 72.6 (9.3); Gender: M = 6, F = 6 HC: Mean (SD) Age = 71.5 (8.4); Gender: M = 7, F = 6	MMSE: Mean (SD) = 23.5 (2.47)	1. aMT	1. Higher aMT in AD group compared with HC at baseline (p < 0.05) 2. Weak positive correlation between MMSE and aMT in AD (R² = 0.04, p < 0.05).	Administration of donepezil reduced aMT in which no difference was found between AD and HC groups	Poor Coil placement and measurement technique was not clearly defined (e.g., location of coil was measured on a line between nasion and inion yet failed to record deviations). Lack of clearly defined methods and procedures differing from international established standards may cause discrepancy in results and comparisons.
Di Lorenzo et al., 2013 [70]	AD = 12 HC = 12	AD: Mean (SD) Age = 69.8 (4.9); Gender: not available HC: Mean (SD) Age = 71.7 (4.4); Gender: not available	MMSE: AD Mean (SD) = 21.08 (3.9) CDR: AD Mean (SD) = 1.21 (1.1)	1. rMT 2. aMT	1. Reduced rMT in AD group 2. Difference in aMT was not reported.	SLAI was decreased in AD and was partially restored using cerebral cTBS.
Wegrzyn et al., 2013 [77]	AD = 19 HC = 9	AD: Mean (SD) Age = 71.8 (4.7); Gender: M = 7, F = 12 HC: Mean (SD) Age = 69.2 (3.7); Gender: M = 9, F = 10	MMSE: AD Mean (SD) = 21.7 (4.9) HC Mean (SD) = 28.6 (1.0)	1. rMT	1. Lower rMT in AD group compared with HC (t₃₆ = 3.196, p = 0.003). 2. No association between TMS and DTI parameters	Longer iSP latency in AD group compared with HC. No correlation between iSP latency and MMSE in AD.	Fair Time elapsed b/w DTI and TMS was over 2 months for 5 AD patients, therefore increasing the likelihood of unaccounted for disease progression.
Hoeppner et al., 2012 [78]	AD = 19 HC = 19	AD: Mean (SD) Age = 71.8 (4.7); Gender: M = 7, F = 12 HC: Mean (SD) Age = 69.2 (3.7); Gender: M = 9, F = 10	MMSE: AD Mean (SD) = 21.7(4.9); range = 10–27 HC Mean (SD) = 28.6 (1.0); range = 26–30	1. rMT 2. MEP peak-to-peak amplitude	1. Lower rMT in AD group compared with HC (F = 5.52, p = 0.024). 2. No difference in MEP peak-to-peak amplitudes between groups. 3. No correlation between TMS excitability parameters and clinical measures (i.e., MMSE and clock drawing test).	Ipsilateral silent period (iSP) latency was significantly longer in AD group compared to HC. No difference in iSP duration or intracortical facilitation (ICF) between AD and HC groups.	Good
Trebbastoni et al., 2012 [79]	AD = 11 HC = 11	AD: Mean (SD) Age = 78.0 (1.4); Gender: M = 3, F = 8 HC: Mean (SD) Age = 75.2 (1.3); Gender: M = 4, F = 7	MMSE: AD Mean = HC Mean = 28.7	1. rMT	1. Reduced rMT found in AD patients as compared with HC.	This study also administered rMTS to assess the changes in synaptic plasticity before and after rivastigmine therapy.	Good
Julkunen et al., 2011 [89]	AD = 5 HC = 5	AD: Mean (SD) Age = 73.2 (8.1); Gender: M = 3, F = 2 MCI: Mean (SD) Age = 74 (8.1), Gender: M = 3, F = 2 HC: Mean (SD) Age = 77.8 (2.6): Gender: M = 1, F = 3	MMSE: AD Mean (SD) = 22 (5.1) MCI Mean (SD) = 25.4 (3.2) HC Mean (SD) = 27 (4.1) CDR: AD Mean (SD) = 0.6 (0.2) MCI Mean (SD) = 0.5 (0.0)	1. rMT 2. TEP P30 amplitude (analysis conducted from an electrode closest to site of stimulation) 3. Receiver operating characteristic (ROC) curve analysis was conducted and Area under the ROC curve (AUC) was computed for P30.	1. No difference in rMT between groups. 2. Lower P30 amplitude in AD group compared to HC (p < 0.05) 3. No difference in P30 amplitude between MCI and HC or MCI and AD. 4. Inverse relationship between P30 amplitude and CDR scores (ρ= –0.808, p < 0.001). 5. Positive relationship between P30 amplitude and MMSE (ρ= 0.537, p < 0.05).	Dementia scales were not correlated with N100 in MCI or AD groups.	Poor Small sample size. All AD subjects were taking cholinesterase inhibitors (ChEIs), whereas all other groups were not on medication. Failure to control for medication in statistical analysis.
Koch et al., 2011 [93]	AD = 10 HC = 10	AD: Mean (SD) Age = 72.5 (6.1); Gender: M = 5, F = 5 HC: Mean (SD) Age = 71.7 (4.9); Gender: not avaliable	MMSE: AD Range = 16–24 HC: not available	1. rMT 2. MEP Amplitude	1. Reduced rMT in AD patients than controls. 2. No difference in MEP amplitude between groups.	This study also assessed the effects of low frequency rTMS and single dose of L-dopa. Repetitive TMS did not induce inhibitory effects in AD. L-dopa induced a reversal of plasticity in HC, not AD.	Good
Kheder et al., 2011 [96]	AD = 45 HC = 37	AD: Mean (SD) Age = 68.4 (6.6) Age Range = 55–82; Gender: M = 16, F = 29 HC: Mean (SD) Age = 65.7 (6.7), Age Range = 55–81; Gender: not available	MMSE: Mean (SD) = 12.3 (4.5); range = 4–23 CDR: Mean (SD) = 2.24 (0.74)	1. rMT 2. aMT 3. MEP latency and amplitude	1. Lower rMT in AD group compared to HC (p < 0.0001). 2. Positive association between motor threshold (rMT and aMT) and MMSE (p = 0.03, p = 0.009 respectively) 3. Negative correlation between CDR and average rMT (ρ= –0.38, p = 0.01) and aMT (ρ= –0.46, p = 0.001)	Prolonged duration of contralateral cortical silent period (CSP) in AD group compared with HC. CSP was negatively correlated with MMSE in AD.	Good
Martorana et al., 2009 [95]	AD = 10 HC = 10	AD: Mean (SD) Age = 72.5 (6.1); Gender: not available HC: Mean (SD) Age = 71.7 (4.9); Gender: not available	MMSE: Mean (SD) = 18.23 (3.2) CDR: Mean (SD) = 1.75 (1.6)	1. rMT	1. Lower rMT in AD group compared to HC (t = 3.14; p < 0.05)	SAI was reduced in AD than HC groups. Administration of L-dopa was able to restore normal levels of SAI in AD group.	Good Extensive clinical investigation completed for all patients; recruitment criteria clearly defined.
Nardone et al., 2008 [59]	AD = 17 HC = 22	AD: Mean Age = 68.4 Age Range = 58–74; Gender: M = 10; F = 7 HC: Mean Age = 70.4 Age Range = 62–73; Gender: M = 10, F = 12	CDR: AD participants had a score of 0.5 or 1 (mild or moderate AD)	1. rMT 2. aMT	1. No difference in rMT and aMT between AD and HC groups	Short latency afferent inhibition (SAI) was reduced in AD compared with HC. SAI did not correlate with disease severity.	Good
Martorana et al., 2008 [94]	AD = 11 HC = 12	AD: Mean (SD) Age = 73 (9.2); Gender: not available HC: Mean (SD) Age = 68 (5.8); Gender: not available	MMSE: Mean (SD) = 15.74 (1.6) ^*HC cut off score > 25 CDR: AD cut off > 1.5	1. rMT 2. aMT	1. Reduced rMT in AD group compared with HC (p < 0.001) 2. No difference in aMT between AD and HC	Administration of L-dopa (melevodopa) reversed the reduced intra-cortical inhibition (ICI) in AD group	Good
Julkunen et al., 2008 [88]	AD = 5 HC = 5	AD: Mean (SD) Age = 73.2 (8.1); Gender: M = 3, F = 2 MCI: Mean (SD) Age = 74 (8.1); Gender: M = 3, F = 2 HC: Mean (SD) Age = 77.8 (2.6); Gender: M = 1, F = 3	MMSE: AD Mean (SD) = 22 (5.1) MCI Mean (SD) = 25.4 (3.2) HC Mean (SD) = 27 (4.1) CDR: AD Mean (SD) = 0.6 (0.2) MCI Mean (SD) = 0.5 (0.0)	1. rMT 2. Global Field Power (GFP)	1. Lower rMT in AD group compared with MCI group 2. AD and control groups displayed significant magnitude differences in GFPs between 30–90 ms and 280–400 ms. 3. AD group showed lower P30 amplitude compared with MCI group	TMS evoked N100 was reduced in MCI group compared with HC.	Poor All AD subjects were taking cholinesterase inhibitors (ChEIs), whereas all other groups were not on medications Medication was not controlled for in statistical analysis.
Di Lazzaro et al., 2008 [80]	AD = 12 HC = 12	AD: Mean (SD) = 69.3 (7.3); Gender: not available HC: Mean (SD) = 73.1 (5.4); Gender: not available	MMSE: AD Mean (SD) = 22.7 (2.7) HC≥29 GDS: AD Mean (SD) = 3.63 (1)	1. rMT 2. aMT	1. Reduced rMT found in AD group compared with HC. 2. Slightly reduced aMT in AD compared to HC, but no statistically significance	Reduced rMT in VaD compared with HC. This study also assessed SAI and found reduced SAI in AD as compared with VaD and HC.	Good
Alberici et al., 2008 [81]	AD = 8 HC = 8	AD: Mean (SD) Age = 74.5 (7.3); Gender: M = 5, F = 8 HC: Mean (SD) Age = 63.1 (7.5); Gender: M = 4, F = 4	MMSE: AD Mean (SD) = 20.2 (4.0) HC Mean (SD) = 27.9 (1.2) IDAL: AD Mean (SD) = 3.4 (2.5) HC Mean (SD) = 0	1. rMT	1. No difference in rMT between groups.	This study also assessed ICI and ICF and found no difference between AD, FLTD and HC groups.	Fair Age of AD patients was significantly higher than control and FLTD groups. Age difference was not accounted for as a covariate in statistical analysis of rMT difference.
Sakuma et al., 2007 [82]	AD = 12 HC = 15	Not available	MMSE: AD Mean (SD) = 21.8 (2.8) MCI Mean (SD) = 27.0 (3.8)	1. rMT	1. Trend of lower rMT in patients with AD and MCI than in HC, but no significant difference (F(2,43) = 1.91, p = 0.16)	Short latency afferent inhibition was reduced in AD compared with HC, but was not reduced when compared with MCI	Fair Statistical analysis was not conducted to determine whether control group is age matched. Statistical analysis did not control for age as possible confound
Battaglia et al., 2007 [83]	AD = 10 HC = 10	AD: Mean (SD) Age = 70.1 (7.4) Gender: M = 6, F = 4 HC: Mean (SD) Age = 68.4 (6.1) Gender: M = 6, F = 4	MMSE: AD Mean (SD) = 20.2 (3.9) HC Mean (SD) = 27.9 (1.8) ADAS-cog: AD Mean (SD) = 24.4 (8) HC Mean (SD) = 8.8 (3.5)	1. rMT	1. No difference in rMT between AD and HC.	This study used PAS protocol to investigate LTP in AD via ISI and MEP amplitude measures. The findings demonstrated PAS did not induce significant potentiation in AD.	Good
Di Lazzaro et al., 2007 [91]	AD = 10 HC = 10	AD: Mean (SD) Age = 72.1 (4.4); Gender: M = 6, F = 4 HC: Mean (SD) Age = 72.0 (5.1); Gender: not available	MMSE: AD Mean (SD) = 17.8 (1.8) HC: not available	1. rMT	1. Reduced rMT in AD when compared with HC.	This study also evaluated SAI and found it was reduced in DLB and AD patients.	Good
Di Lazzaro et al., 2006 [92]	AD = 20 HC = 20	AD: Mean (SD) Age = 69.5 (6.5); Gender: M = 10, F = 10 HC: Mean (SD) Age = 71.8 (5.7); Gender: not available	MMSE: AD Mean (SD) = 20.1 (3.5) HC Mean (SD) = 27 (4.1)	1. rMT	2. Lower rMT in AD group compared with HC group (F (2,54) = 3.21, p < 0.05)	Short latency afferent inhibition was reduced in AD, whereas showed no different in FTD when compared with HC.	Good
Inghilleri et al., 2006 [84]	AD = 20 HC = 20	AD: Mean (SD) Age = 71 (2.1); Gender: not available HC: Mean (SD) Age = 69 (1.6); Gender: not available	MMSE: AD Mean (SD) = 16 (1) HC: not available	1. rMT	1. Reduced rMT in AD patients as compared with HC.	This study also assessed MEP and CSP duration evoked by rTMS. MEP facilitation was reduced in AD, whereas no difference in CSP was found between groups.	Good
Pierantozzi et al., 2004 [60]	AD = 12 HC = 12	AD: Mean (SD) Age = 65.2 (3.5); Gender: not available HC: Mean (SD) Age = 64.5 (3.2); Gender: not available	MMSE: AD Mean (SD) = 21.2 (2.1) HC: not available	1. rMT 2. aMT 3. Single MEP amplitude	1. No difference in rMT or aMT between groups. 2. No difference in single MEP amplitude between groups.	This study also investigated ICI and ICF before and after administration of a single dose of galantamine. AD patients showed reduced ICI and no difference in ICF when compared to FTD and HC groups.	Good
Di Lazzaro et al., 2004 [98]	AD = 28 HC = 12	AD: Mean (SD) Age = 71.3 (6.8); Gedner: not available HC: Mean (SD) Age = 73.1 (5.4); Gender: not available	MMSE: Mean (SD) = 19.35 (3.8) ^*HC cut off score > 24 RAVLT: Mean Immediate Recall (SD) = 21.5 (8.1) Mean Delayed Recall (SD) = 1.9 (2.5)	1. rMT	1. Lower rMT in AD group compared with HC (p,0.05, Mann-Whitney test, N1 = 28, N2 = 12)	Short latency afferent inhibition was found to be reduced in AD as compared with HC. Administration of cholinesterase inhibitor rivastigmine did not modify rMT.	Good
Alagona et al., 2004 [100]	AD = 20 HC = 20	AD: Mean (SD) Age = 72.2 (7.53); Age Range = 55–82; Gender: M = 7, F = 13 SIVD: Mean (SD) Age = 71.8(9.37); Age Range = 48–84; Gender: M = 8, F = 12 HC: Mean (SD) Age = 68.55 (7.96); Age Range: 52–80; Gender: M = 8, F = 12	Not available	1. rMT 2. MEP/Mmax	1. Reduced rMT in AD as compared with HC 2. No difference in MEP/Mmax ratio between groups	This study also assessed SP and found no significant difference between all three groups.	Good
Ferreri et al., 2003 [57]	AD = 16 HC = 13	AD: Mean (SD) Age = 75 (6.9); Age Range = 59–82; Gender: M = 1, F = 15 HC: Mean (SD) Age = 72 (2); Age Range = 56–91; Gender: M = 5, F = 8	CDR: Only patients with scores of 1 or 2 (mild to moderate severity included. MMSE: Not available Mental Deterioration Battery: Not available Instrumental of Daily Living Scale: Not available	1. rMT	1. Global increased motor cortex excitability in AD patients, indicated by lower rMT compared with control group F(5,23) = 3.808; p = 0.012 2. No correlation between motor cortex excitability and AD severity	This study also assessed center of gravity of motor cortical output and found a frontal and medial shift in AD	Good Subjects were in supine position during TMS.
Di Lazzaro et al., 2002 [87]	AD = 15 criteria. HC = 12	AD: Mean (SD) Age = 69.3 (5.3); Gender: M = 9, F = 6 HC: Mean (SD) Age = 73.1 (5.4); Gender: not available	MMSE: Mean (SD) = 18.6 (3.5) ^*HC cut off score > 24 RAVLT: Mean Immediate Recall (SD) = 26.6 (8.4) Mean Delayed Recall (SD) = 2.54 (2.4)	1. rMT 2. aMT	1. Lower rMT in AD group compared with HC (p < 0.05). 2. No difference in aMT between AD and HC	Short latency afferent inhibition was found to be reduced in AD as compared with HC	Good Sample size justification not reported.
Liepert et al., 2001 [85]	AD = 11 HC = 10	AD: Mean (SD) Age = 74.8 (9.7); Age Range = 53–86; Gender: M = 2, F = 9 HC: Mean (SD) Age = 70.0 (6.4); Age Range = 56–80; Gender: M = 3, F = 7	CDR: AD Mean (SD) = 1.64 (0.64)	1. rMT	1. No difference in rMT between AD and HC groups.	This study also investigated ICI and ICF before and after treatment with donepezil. Patients with AD had reduced ICI which correlated with CDR scores. Treatment with donepezil increased ICI in AD.	Good
Alagona et al., 2001 [62]	AD = 21 HC = 18	All AD: Median Age = 72; Age Range = 55–81 Gender: M = 9, F = 12 Mild AD: Mean Age = 67 Age Range = 55–77 Gender: M = 5, F = 5 Moderate AD: Mean Age = 71.1 Age Range = 58–80 Gender: M = 3, F = 4 Severe AD: Mean Age = 71.9 Age Range = 55–81; Gender: M = 4, F = 5 HC: Median Age = 69 Age Range = 52–88; Gender: M = 13, F = 5	Blessed Dementia Scale (BDS): Mild: n = 5 Moderate: n = 7 Severe: n = 9	1. rMT 2. MEP peak-to-peak amplitude	1. Lower median rMT in both hemispheres of AD group compared with HC (p < 0.01) 2. Greater MEP amplitudes in both hemispheres of AD group (p < 0.01) 3. Inverse correlation between threshold and stage of cognitive severity (p < 0.01)	This study also assessed silent period (SP) and central motor conduction time (CMCT). Median values of SP duration were reduced in AD No correlation between CMCT and cognitive severity	Fair Authors did not specify whether controls matched (e.g., age, gender). Covariates were not considered in analysis (e.g., cortical atrophy) although discussed as potential contributing factors.
Pepin et al., 1999 [86]	AD = 17 Aged HC = 22 Young HC = 29	AD: Mean (SD) Age = 70.7 (6.6); Age Range = 57–86; Gender: M = 4, F = 13 Aged HC: Mean (SD) Age = 68.5 (9.2); Age Range = 51–81; Gender: M = 6, F = 16 Young HC: Mean (SD)=34.1 (8.1); Range = 22–48; Gender: M = 17, F = 12	Reisberg’s Global Deterioration Scale: Mild: n = 6 Moderate: n = 8 Moderate-Severe: n = 3	1. rMT 2. aMT 3. Ratio of largest MEP and first negative peak (MEP max/M max)	1. Reduced mean rMT and aMT in AD compared with young and aged HC groups (p < 0.001) 2. Higher MEP max/M max ratio in AD group compared with HC (p < 0.001)	This study also assessed paired magnetic stimulation at interstimulus intervals between 1 to 12 ms and found no difference between AD and HC groups.	Fair Failure to age-match aged healthy control population with AD population or control for age in statistical analysis.
De Carvalho et al., 1997 [101]	AD = 14 HC = 11	AD: Mean Age = 67.8 Age Range = 55–78 Gender: M = 4, F = 10 HC: Mean Age = 66.1 Age Range = 55–75; Gender: M = 3, F = 8	BDS: Mean (SD) 9.8±4.2	1. rMT	1. Lower rMT in left and right hemispheres (p < 0.05)	This study also assessed spinal motor neuron excitability and found no difference between AD and HC groups.	Fair Small sample size. Cognitive assessment scores were not listed.
Perretti et al., 1996 [99]	AD = 15 HC = 10	AD: Mean (SD) Age = 67.2 (7.8); Age Range = 58–80; Gender: M = 4, F = 11 HC: Mean (SD) Age = 67.8 (5); Age Range = 57–73; Gender: M = 3, F = 7	MMSE: Moderate AD (n = 7): Mean (SD) = 21.7 (3.0) Severe AD (n = 8): Mean (SD) = 13.7 (5.6)	1. rMT 2. MEP amplitude	1. Higher rMT in AD compared with HC. 2. No difference in MEP amplitude between groups.	This study also assessed CCT and cSP and found a reduced cSP duration in AD. No difference was found in CCT between groups.	Fair Only obtained MT in 6 of 15 patients (<50%)/ Analyzed sample size was less than half the reported sample. Authors did not specify whether adverse events were present.

AD, Alzheimer’s disease; aMT, active motor threshold; CCT, Central Conduction Time; CDR, Clinical Dementia Rating; cSP, cortical silent period; GDS, Global Deterioration Scale; GMF, global mean field; GMFP, global mean field power; HC, healthy control; LTP, long-term potentiation; MEP, motor evoked potential; MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination; MoCA, Montreal-Cognitive Assessment; PD, Parkinson’s disease; rMT, resting motor threshold; RAVLT, Rey Auditory Verbal Test; SAI, Short Afferent Inhibition; TEP, TMS-evoked potential; VaD, vascular dementia.

Comparison of cortical excitability between AD and healthy cohorts

Resting motor threshold

RMT was significantly reduced in AD group (Hedges’ g = –0.99, 95% CI [–1.29, –0.68], p < 0.00001) as compared with healthy cohort. Considerable heterogeneity was found between studies assessing rMT (τ² = 0.53, Q (29) = 85.68, p < 0.00001, I² =81%). Twenty-four publications reported decreased rMT in AD [57 , 100–102], nine reported no difference between groups [59 , 89] and one reported increased rMT [60] in patients with AD.

Subgroup analyses separating studies based on rMT technique showed reduced rMT in AD groups as compared with healthy cohorts for both the relative frequency method (Hedges’ g = –0.82, 95% CI [–1.19, –0.45], p < 0.00001) and modified or unspecified methods (Hedges’ g = –1.11, [95% CI –1.58, –0.65], p < 0.00001). Considerable levels of heterogeneity were found in the relative frequency method (τ² = 0.29, Q (11) = 41.04, p < 0.0001, I² = 73%) and modified or unspecified methods (τ² = 0.80, Q (17) = 103.86, p < 0.0001, I² = 84 %). Most publications reported using the relative frequency method [57 , 101], however others reported modified methods [80 , 95–97] and some did not specify the procedures used [61 , 99].

Fig. 2

Forest plot of standard mean difference (SMD) comparing (A) resting motor threshold (rMT) and (B) active motor threshold (aMT) between Alzheimer’s disease and healthy control groups. SMD for each study was calculated using individual mean and standard deviation (SD) values and are indicated by the green squares. The size of the square is proportional to the weight of the study in the meta-analyses, and the horizontal lines indicate the 95% confidence interval (CI) of each study. Inverse variance (IV) and random effects methods were used to calculate SMDs, 95% CIs, p values, and the test for overall effect. The diamond represents the estimated overall effect based on random effects meta-analyses. The τ² and χ² test was used to calculate heterogeneity. Random, random effects method.

Active motor threshold

Similar to the results of rMT, AD groups showed reduced aMT (Hedges’ g = –0.87, 95% CI [–1.50, –0.24], p = 0.007). Considerable heterogeneity was found between studies assessing aMT (τ² = 0.58, Q (6) = 31.74, p < 0.00001, I² = 81%). Three publications reported decreased aMT in patients with AD [86 , 97], six reported no difference between groups [59 , 94], whereas one reported increased aMT in the AD group [90].

Publication bias

Visual inspection of the funnel plot demonstrated some asymmetry but this was not sufficient to impact the results (Fig. 4). We conducted a sensitivity analysis by excluding the studies that were causing asymmetry in the funnel plots. The results of the remaining studies did not change (Hedges’ g = –0.84, 95% CI [–0.99, –0.70], p < 0.001), and low heterogeneity was detected (τ² = 0.01, Q (22) = 23.06, p = 0.40, I² = 5%). Further, the main effect of a reduced rMT was retained in each subgroup.

Fig. 4

Funnel plot of (A) resting motor threshold and (B) active motor threshold based on studies included in meta-analyses. Dotted lines represent the 95% confidence interval (CI). SMD, standard mean difference; SE, standard error.

Fig. 3

Forest plot of subgroup meta-analyses based on comparison of methodology used to determine resting motor threshold (rMT). Standard mean difference (SMD) comparing rMT for the (A) relative frequency method versus (B) modified or unspecified methods. SMD for each study was calculated using individual mean and standard deviation (SD) values and are indicated by the green squares. The size of the square is proportional to the weight of the study in the meta-analyses, and the horizontal lines indicate the 95% confidence interval (CI) of each study. Inverse variance (IV) and random effects methods were used to calculate SMDs, 95% CIs, p values, and the test for overall effect. The diamond represents the estimated overall effect based on random effects meta-analyses. The τ² and χ² test was used to calculate heterogeneity. Random, random effects method.

Motor evoked potential and TMS-evoked potential

Studies assessing MEP or TEP were not sufficient to conduct meta-analyses. Two studies reported increased MEP amplitude [62, 86] and one reported a trend of higher MEP amplitude [100] in AD, while others reported no difference [60 , 99] between groups. Most studies evaluated MEP amplitude using averaged responses, however one study used a single unconditioned test stimulus [60] and two studies used the ratio of MEP amplitudes to that of the maximal motor response (MEP/Mmax) [62, 99]. Regarding TEPs, one publication [57] reported greater global field power (GFP) at P30 and P60 evoked response potentials in AD (n = 12) compared to healthy controls (n = 12). Another publication from a smaller study [88] reported reduced P30 amplitude and lower GFP (between 280 and 400 ms of the TEP) in AD (n = 5) compared with healthy controls (n = 4).

Associations between cortical excitability and clinical symptoms of AD

Eight publications reported on the relationship between cortical excitability and cognition using different measures and thus were not amenable to conduct meta-analyses [57 , 96]. To evaluate disease severity and global cognition, assessments including Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Clinical Dementia Rating (CDR) scale, or Blessed Dementia Scale (BDS) were used. In addition, one study used the clock-drawing test (CDT) to assess executive and visuospatial functioning [78].

Two publications reported a positive correlation between motor threshold (both rMT and aMT, higher rMT or aMT indicating lower excitability) and MMSE (higher score indicates better cognition) [90, 96]. These findings indicate that lower cortical excitability is associated with better cognition. Other studies found no correlation between motor cortex excitability and cognition as assessed by MMSE, MoCA, or CDT [57 , 78]. One publication reported a negative correlation between P30 amplitude, after dorsolateral prefrontal cortex (DLFPC) stimulation (higher P30 amplitude indicating higher cortical excitability), and MMSE scores, and also found that P30 amplitude was a good predictor of MMSE and face-naming association memory task performance [73]. Therefore, suggesting that greater cortical excitability was predictive of poor cognition. In contrast, another publication from a very small study reported a positive correlation between global P30 amplitude and MMSE scores, and a negative correlation between global P30 amplitude and CDR scores (higher CDR scores indicate worse illness) [89]. Thus, contradicting the findings of the previous study.

Two studies assessing the relationship between disease severity and motor cortex excitability assessed using rMT or aMT reported a positive association (higher disease severity associated with higher motor cortical excitability) [62, 96]. One publication reported an inverse association between BDS (higher scores indicate more cognitive impairment) and rMT in both cerebral hemispheres [62]. Similarly, in another study CDR scores were negatively associated with both rMT and aMT in both cerebral hemisphere of patients with AD [96]. One study reported no association between cortical excitability and disease severity, as assessed via a combination of MMSE, CDR, and Instrumental Activities of Daily Living scale [57]. We did not find any publications reporting on the relationship between cortical excitability and NPS in patients with AD.

DISCUSSION

This systematic review and meta-analyses provide an in-depth assessment of current literature regarding quantitative abnormalities of cortical excitability and their association with clinical symptoms of AD. In line with previous reviews [58 , 63–65], we found evidence of increased motor cortex excitability in AD as compared to healthy cohorts. This was confirmed by the results of meta-analyses of studies using rMT and aMT, and subgroup analyses based on different methods used to obtain rMT. We found significant methodological heterogeneity among published studies; however, it did not impact our results. Findings of corticospinal excitability evaluated using MEP amplitude were inconsistent, with some publications reporting higher excitability and others finding no differences. Further, the reviewed literature was suggestive of an inverse relationship between motor cortex excitability (assessed using motor threshold) and cognition. Few studies used TMS-EEG to assess cortical excitability directly from the scalp (one after stimulation of the DLPFC), rather than through assessing muscle activity, and reported inconsistent results. We did not find any publications reporting the relationship between cortical excitability and NPS.

Our finding of increased cortical excitability is in agreement with previous clinical EEG studies in which patients with AD display greater incidence of unprovoked seizures and epileptiform discharges [43 , 103]. Network hyper-excitability has been reported in AD mouse models that have shown an increase in epileptiform signatures (e.g., discharges, spikes, and sharp waves) and nonconvulsive seizures [39, 40]. Most publications included in our study reported increased motor cortex excitability [57 , 94–98] as indexed by reduced rMT and aMT in AD as compared with healthy older individuals. A reduced motor threshold means that a lower TMS stimulus intensity is required to evoke motor response, and thus represents increased cortical excitability [53, 104]. There was considerable methodological heterogeneity in the determination of motor threshold. rMT was defined as the minimum stimulus required to produce MEPs > 50μV in five of 10 trials in all but four studies. Two studies defined rMT as the minimum stimulus able to produce MEPs > 20μV [62, 99] whereas, two other studies required MEPs > 100μV in 50% of consecutive trials [60, 94]. Similarly, studies showed discrepancy in aMT definition such that all but one study [94] required MEPs > 200μV to be elicited. Further, two studies used a target muscle contracted at 10% of maximum contraction [59, 96] as opposed to 20% of maximum voluntary muscle contraction [87, 90]. Although these methodological discrepancies did not impact our overall findings, it highlights the need for the use of more consistent techniques across studies. Studies using TMS-EEG did not show conclusive results [61 , 89]. While one study showed increased cortical excitability (greater P30 amplitude) in the sensorimotor cortices of mild AD [75], another showed reduced P30 amplitude in the temporoparietal area of patients with early AD [88]. Previous resting EEG studies have reported a decrease in alpha and beta (fast frequency) and increase in delta and theta (slow frequency) amplitude and power in AD [25 , 105–110]. Additional research is needed to understand the relationship between resting EEG changes and cortical excitability in AD.

Decreased motor threshold may be indicative of increased excitatory activity or disrupted excitation/inhibition balance in motor cortical networks [96] due to over-activation of glutamatergic neurotransmission [111]. Excessive glutamate has been proposed to result in increased excitotoxicity [112, 113], leading to further neuronal loss. Increased excitability is also thought to be a direct consequence of Aβ and tauopathy in the brain [40 , 114–118]. Hyper-excitability was found within close proximity of plaques and thought to be a consequence of the plaque-induced increases in intracellular calcium [119]. Aβ levels have been linked to increase presynaptic Ca^2 +, a well-established regulator of excitatory neurotransmitter (e.g., glutamate) release [120, 121] and the subsequent hyperactivity leads to exacerbated neuronal death and cognitive and behavioral deficits of AD [122]. In addition to changes in excitatory neurotransmitters, mouse models also displayed changes in the GABAergic pathways and excitation/inhibition imbalance [39, 123]. It has been proposed that AD pathology may affect excitatory and inhibitory networks distinctively, resulting in an imbalance [124]. The recurrent epileptiform and seizure activity in EEG recordings of patients with AD and mouse models are also suggestive of network hyper-excitability, as epileptic activity is characteristic of disrupted balance in cortical excitation and inhibition, with increased excitation [125]. However, it still remains unresolved if cortical hyper-excitability is a direct reflection of increased excitatory network activity or a consequence of impairment in inhibition. Future studies need to assess cortical excitability and inhibition in conjunction with other biomarkers of core AD pathology and brain metabolites systems (such as GABA, glutamate, and acetylcholine) to clarify these relationships.

The finding of an inverse relationship between motor cortex excitability and cognition in AD is consistent with previous qualitative EEG studies. Subclinical epileptiform activity occurs at elevated rates in more advanced stages of dementia [43 , 127] and is associated with more rapid cognitive decline [46 , 129]. The occurrence of epileptiform activity may be present with faster disease progression because it hastens AD pathology, thereby contributing to worsening clinical impairment [43]. Few studies have also shown patients with MCI to have reduced rMT when compared to healthy controls; however, the effect sizes are small [82 , 130] and one study has also found higher rMT in MCI. When comparing rMT with disease severity across the included studies, discrepancy was found in mild and mild to moderate AD populations. Most publications reporting on mild AD did not show a difference in rMT, whereas only some reporting on mild to moderate AD did not find a difference in rMT. Publications including moderate or severe AD consistently reported reduced rMT. This suggests that abnormal cortical excitability may be more widely prevalent in advance stage of the illness. The relationship between cortical excitability and cognition has primarily been assessed in the motor cortex, region not directly relevant to cognition. Only one TMS-EEG study evaluated cortical excitability, after stimulating DLPFC, in relation to cognitive performance on the MMSE, and found increased cortical excitability (greater P30 amplitude) to be predictive of lower MMSE performance [73]. The authors found no association between other TEP components such as P50 and P70 (which were localized in different areas of the prefrontal cortex) and cognition, therefore indicating that hyper-excitability may be network specific [73]. These findings are in support of previous fMRI studies in AD that have shown altered DLPFC connectivity with disease progression [73 , 132]. In contrast, a smaller study reported reduced P30 amplitude to be positively correlated with cognition [89]. The inconsistency in findings among studies emphasize the need for future well-powered TMS-EEG studies in non-motor regions to reliably quantify cortical excitability and better understand excitability abnormalities in the frontal cortical regions directly relevant for cognition in patients with AD [133]. Moreover, resting EEG abnormalities have shown associations with cognition in AD; alpha power was associated positively with global cognition [134 –137] and both fast wave beta power and slow wave delta power were associated negatively with global cognition [34, 137]. This suggests that there may be abnormalities of both slow and fast rhythms in AD; however, direct relevance of these findings to cortical excitability is not clear and requires further investigation. Some studies have shown that sex and gender may impact cortical excitability; reduced motor cortex excitability showed association with greater cognition in female but not male healthy individuals [138, 139]; however, further evaluation of this relationship in AD is required. Taken together, while there is some evidence for an inverse relationship between motor cortex excitability and cognition, this needs further study in both motor and non-motor areas. Further, studying mediators of the relationship between cortical excitability and cognition using established clinical, demographic, and pathophysiological markers of AD is required.

We did not find evidence regarding association between cortical excitability and NPS. Previously, cortical excitability (lower rMT) has been associated positively with aggression in post-traumatic stress disorder [140]. Lower rMT is also reported in other neuropsychiatric disorders such as schizophrenia [141, 142] and obsessive-compulsive disorder [143] as compared to healthy populations. Studies in AD populations have shown that NPS are positively associated with cognitive impairment [144 –147]. Further, some studies have found positive associations between NPS and AD neuropathology [11 , 148–153]. Finally, medications altering cortical excitability such as GABA agonists, anti-epileptics, and NMDA antagonist are used for symptomatic management of NPS [154 –157]. These lines of evidence suggest that cortical excitability may play a role in NPS related to AD. Thus, future studies should not only assess the relationship between cortical excitability and NPS in AD, but also aim to understand how AD pathology and other markers of the disease (structural and functional imaging, fluid markers, etc.) may affect this relationship. Only three studies collected CSF or PET biomarker data to confirm evidence of disease pathology (Aβ and tau positivity); however, these biomarkers were not studied in relation to cortical excitability [70 –72]. Such research will be a critical step in advancing mechanistic understanding of cortical excitability and its relationship with clinical symptoms which would eventually lead to development of more targeted treatment interventions.

This review and meta-analyses has several limitations. First, this study was not registered a priori; however, this study is unique and has clearly defined objectives separate from the previous such studies. Second, we included only published reports in English language. However, we included publications covering a broad range of neurophysiological techniques to assess cortical excitability and provided a comprehensive summary of the published literature on this topic. Third, although the findings related to the difference in motor threshold between AD and healthy controls were clear, it was not possible to conduct a meta-analysis about the relationship between cortical excitability and cognition, so we reported a qualitative synthesis of these studies. Fourth, we could not assess the relationship between cortical excitability and NPS because no previous studies directly assessed this relationship, indicating a critical need for further research.

CONCLUSIONS AND FUTURE DIRECTIONS

This systematic review and meta-analyses highlight abnormalities of cortical excitability in AD. Results of our meta-analyses showed strong evidence of reduced rMT and aMT, suggestive of increased motor cortex excitability. Further, the literature provides some support for an inverse relationship between motor cortex excitability and cognition in AD. The evidence regarding abnormal cortical excitability from non-motor areas is sparse and inconsistent. Finally, there was no evidence in support or against an association between cortical excitability and NPS in AD. Further research is needed to understand the state of cortical excitability in the non-motor brain regions (such as DLPFC) in AD and to better understand its associations with clinical symptoms. The mechanisms underlying cortical hyper-excitability and its association with AD pathology, brain structure and function, brain metabolites, and other markers of the disease (such as CSF and blood biomarkers) also remains unclear and needs further study. This may have important implications for developing targeted treatment interventions and markers of treatment response.

Footnotes

ACKNOWLEDGMENTS

This work was supported in part by an Academic Scholars Award from the Department of Psychiatry, University of Toronto to S.K., support from Canada Research Chair to T.K.R., and support from Temerty Family through the CAMH Foundation and the Campbell Research Institute to D.M.B.

Authors’ disclosures available online ().

References

Gotovac

, Nikolac Perkovic

, Pivac

, Borovecki

(2016) Biomarkers of aggression in dementia. Prog Neuropsychopharmacol Biol Psychiatry 69, 125–130.

Friedland

(1993) Epidemiology, education, and the ecology of Alzheimer’s disease. Neurology 43, 246–249.

Geda

, Schneider

, Gitlin

, Miller

, Smith

, Bell

, Evans

, Lee

, Porsteinsson

, Lanctot

, Rosenberg

, Sultzer

, Francis

, Brodaty

, Padala

, Onyike

, Ortiz

, Ancoli-Israel

, Bliwise

, Martin

, Vitiello

, Yaffe

, Zee

, Herrmann

, Sweet

, Ballard

, Khin

, Alfaro

, Murray

, Schultz

, Lyketsos

, Neuropsychiatric Syndromes Professional Interest Area of ISTAART (2013) Neuropsychiatric symptoms in Alzheimer’s disease: Past progress and anticipation of the future. Alzheimers Dement 9, 602–608.

Marshall

, Rentz

, Frey

, Locascio

, Johnson

, Sperling

, Alzheimer’s Disease Neuroimaging Initiative (2011) Executive function and instrumental activities of daily living in mild cognitive impairment and Alzheimer’s disease. Alzheimers Dement 7, 300–308.

Lyketsos

, Carrillo

, Ryan

, Khachaturian

, Trzepacz

, Amatniek

, Cedarbaum

, Brashear

, Miller

(2011) Neuropsychiatric symptoms in Alzheimer’s disease. Alzheimers Dement 7, 532–539.

Zhao

, Tan

, Wang

, Jiang

, Tan

, Xu

, Li

, Wang

, Lai

, Yu

(2016) The prevalence of neuropsychiatric symptoms in Alzheimer’s disease: Systematic review and meta-analysis. J Affect Disord 190, 264–271.

, Hu

, Tan

, Yu

, Tan

(2014) Behavioral and psychological symptoms in Alzheimer’s disease. Biomed Res Int 2014, 927804.

Ekman

, Ferreira

, Westman

(2018) The A/T/N biomarker scheme and patterns of brain atrophy assessed in mild cognitive impairment. Sci Rep 8, 8431.

Jack

Jr. , Bennett

, Blennow

, Carrillo

, Feldman

, Frisoni

, Hampel

, Jagust

, Johnson

, Knopman

, Petersen

, Scheltens

, Sperling

, Dubois

(2016) A/T/N: An unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology 87, 539–547.

10.

Bloniecki

, Aarsland

, Cummings

, Blennow

, Freund-Levi

(2014) Agitation in dementia: Relation to core cerebrospinal fluid biomarker levels. Dement Geriatr Cogn Dis Extra 4, 335–343.

11.

Tekin

, Mega

, Masterman

, Chow

, Garakian

, Vinters

, Cummings

(2001) Orbitofrontal and anterior cingulate cortex neurofibrillary tangle burden is associated with agitation in Alzheimer disease. Ann Neurol 49, 355–361.

12.

Altomare

, de Wilde

, Ossenkoppele

, Pelkmans

, Bouwman

, Groot

, van Maurik

, Zwan

, Yaqub

, Barkhof

, van Berckel

, Teunissen

, Frisoni

, Scheltens

, van der Flier

(2019) Applying the ATN scheme in a memory clinic population: The ABIDE project. Neurology 93, e1635–e1646.

13.

Chetelat

(2013) Alzheimer disease: Abeta-independent processes-rethinking preclinical AD. Nat Rev Neurol 9, 123–124.

14.

Edison

, Archer

, Hinz

, Hammers

, Pavese

, Tai

, Hotton

, Cutler

, Fox

, Kennedy

, Rossor

, Brooks

(2007) Amyloid, hypometabolism, and cognition in Alzheimer disease: An [11C]PIB and [18F]FDG PET study. Neurology 68, 501–508.

15.

Sperling

, Laviolette

, O’Keefe

, O’Brien

, Rentz

, Pihlajamaki

, Marshall

, Hyman

, Selkoe

, Hedden

, Buckner

, Becker

, Johnson

(2009) Amyloid deposition is associated with impaired default network function in older persons without dementia. Neuron 63, 178–188.

16.

Johnson

, Fox

, Sperling

, Klunk

(2012) Brain imaging in Alzheimer disease. Cold Spring Harb Perspect Med 2, a006213.

17.

Silverman

, Gambhir

, Huang

, Schwimmer

, Kim

, Small

, Chodosh

, Czernin

, Phelps

(2002) Evaluating early dementia with and without assessment of regional cerebral metabolism by PET: A comparison of predicted costs and benefits. J Nucl Med 43, 253–266.

18.

Smailovic

, Koenig

, Kareholt

, Andersson

, Kramberger

, Winblad

, Jelic

(2018) Quantitative EEG power and synchronization correlate with Alzheimer’s disease CSF biomarkers. Neurobiol Aging 63, 88–95.

19.

Babiloni

, Del Percio

, Caroli

, Salvatore

, Nicolai

, Marzano

, Lizio

, Cavedo

, Landau

, Chen

, Jagust

, Reiman

, Tedeschi

, Montella

, De Stefano

, Gesualdo

, Frisoni

, Soricelli

(2016) Cortical sources of resting state EEG rhythms are related to brain hypometabolism in subjects with Alzheimer’s disease: An EEG-PET study. Neurobiol Aging 48, 122–134.

20.

Kramberger

, Kareholt

, Andersson

, Winblad

, Eriksdotter

, Jelic

(2013) Association between EEG abnormalities and CSF biomarkers in a memory clinic cohort. Dement Geriatr Cogn Disord 36, 319–328.

21.

Fernandez

, Hornero

, Mayo

, Poza

, Gil-Gregorio

, Ortiz

(2006) MEG spectral profile in Alzheimer’s disease and mild cognitive impairment. Clin Neurophysiol 117, 306–314.

22.

Jelic

, Blomberg

, Dierks

, Basun

, Shigeta

, Julin

, Jensen

, Lannfelt

, Winblad

, Wahlund

(1998) EEG slowing and cerebrospinal fluid tau levels in patients with cognitive decline. Neuroreport 9, 157–160.

23.

Rajji

, Sun

, Zomorrodi-Moghaddam

, Farzan

, Blumberger

, Mulsant

, Fitzgerald

, Daskalakis

(2013) PAS-induced potentiation of cortical-evoked activity in the dorsolateral prefrontal cortex. Neuropsychopharmacology 38, 2545–2552.

24.

Daskalakis

, Farzan

, Barr

, Maller

, Chen

, Fitzgerald

(2008) Long-interval cortical inhibition from the dorsolateral prefrontal cortex: A TMS-EEG study. Neuropsychopharmacology 33, 2860–2869.

25.

Koenig

, Prichep

, Dierks

, Hubl

, Wahlund

, John

, Jelic

(2005) Decreased EEG synchronization in Alzheimer’s disease and mild cognitive impairment. Neurobiol Aging 26, 165–171.

26.

Babiloni

, Binetti

, Cassetta

, Cerboneschi

, Dal Forno

, Del Percio

, Ferreri

, Ferri

, Lanuzza

, Miniussi

, Moretti

, Nobili

, Pascual-Marqui

, Rodriguez

, Romani

, Salinari

, Tecchio

, Vitali

, Zanetti

, Zappasodi

, Rossini

(2004) Mapping distributed sources of cortical rhythms in mild Alzheimer’s disease. A multicentric EEG study. Neuroimage 22, 57–67.

27.

Jeong

(2004) EEG dynamics in patients with Alzheimer’s disease. Clin Neurophysiol 115, 1490–1505.

28.

Ponomareva

, Selesneva

, Jarikov

(2003) EEG alterations in subjects at high familial risk for Alzheimer’s disease. Neuropsychobiology 48, 152–159.

29.

Wolf

, Jelic

, Gertz

, Nordberg

, Julin

, Wahlund

(2003) A critical discussion of the role of neuroimaging in mild cognitive impairment. Acta Neurol Scand Suppl 179, 52–76.

30.

Dierks

, Jelic

, Pascual-Marqui

, Wahlund

, Julin

, Linden

, Maurer

, Winblad

, Nordberg

(2000) Spatial pattern of cerebral glucose metabolism (PET) correlates with localization of intracerebral EEG-generators in Alzheimer’s disease. Clin Neurophysiol 111, 1817–1824.

31.

Huang

, Wahlund

, Dierks

, Julin

, Winblad

, Jelic

(2000) Discrimination of Alzheimer’s disease and mild cognitive impairment by equivalent EEG sources: A cross-sectional and longitudinal study. Clin Neurophysiol 111, 1961–1967.

32.

Prichep

, John

, Ferris

, Reisberg

, Almas

, Alper

, Cancro

(1994) Quantitative EEG correlates of cognitive deterioration in the elderly. Neurobiol Aging 15, 85–90.

33.

Park

, Che

, Im

, Jung

, Bae

, Lee

(2008) Decreased EEG synchronization and its correlation with symptom severity in Alzheimer’s disease. Neurosci Res 62, 112–117.

34.

Stam

, van der Made

, Pijnenburg

, Scheltens

(2003) EEG synchronization in mild cognitive impairment and Alzheimer’s disease. Acta Neurol Scand 108, 90–96.

35.

Kim

, Lee

, Park

, Kim

, Bae

, Kim

, Im

(2012) Clinical implications of quantitative electroencephalography and current source density in patients with Alzheimer’s disease. Brain Topogr 25, 461–474.

36.

Jelic

, Shigeta

, Julin

, Almkvist

, Winblad

, Wahlund

(1996) Quantitative electroencephalography power and coherence in Alzheimer’s disease and mild cognitive impairment. Dementia 7, 314–323.

37.

Pozzi

, Golimstock

, Petracchi

, Garcia

, Starkstein

(1995) Quantified electroencephalographic changes in depressed patients with and without dementia. Biol Psychiatry 38, 677–683.

38.

Palop

, Mucke

(2009) Epilepsy and cognitive impairments in Alzheimer disease. Arch Neurol 66, 435–440.

39.

Palop

, Chin

, Roberson

, Wang

, Thwin

, Bien-Ly

, Yoo

, Ho

, Yu

, Kreitzer

, Finkbeiner

, Noebels

, Mucke

(2007) Aberrant excitatory neuronal activity and compensatory remodeling of inhibitory hippocampal circuits in mouse models of Alzheimer’s disease. Neuron 55, 697–711.

40.

Minkeviciene

, Rheims

, Dobszay

, Zilberter

, Hartikainen

, Fulop

, Penke

, Zilberter

, Harkany

, Pitkanen

, Tanila

(2009) Amyloid beta-induced neuronal hyperexcitability triggers progressive epilepsy. J Neurosci 29, 3453–3462.

41.

, Hussaini

, Bastille

, Rodriguez

, Mrejeru

, Rilett

, Sanders

, Cook

, Fu

, Boonen

, Herman

, Nahmani

, Emrani

, Figueroa

, Diamond

, Clelland

, Wray

, Duff

(2016) Neuronal activity enhances tau propagation and tau pathology in vivo. Nat Neurosci 19, 1085–1092.

42.

Yamamoto

, Tanei

, Hashimoto

, Wakabayashi

, Okuno

, Naka

, Yizhar

, Fenno

, Fukayama

, Bito

, Cirrito

, Holtzman

, Deisseroth

, Iwatsubo

(2015) Chronic optogenetic activation augments abeta pathology in a mouse model of Alzheimer disease. Cell Rep 11, 859–865.

43.

Vossel

, Beagle

, Rabinovici

, Shu

, Lee

, Naasan

, Hegde

, Cornes

, Henry

, Nelson

, Seeley

, Geschwind

, Gorno-Tempini

, Shih

, Kirsch

, Garcia

, Miller

, Mucke

(2013) Seizures and epileptiform activity in the early stages of Alzheimer disease. JAMA Neurol 70, 1158–1166.

44.

Lam

, Sarkis

, Pellerin

, Jing

, Dworetzky

, Hoch

, Jacobs

, Lee

, Weisholtz

, Zepeda

, Westover

, Cole

, Cash

(2020) Association of epileptiform abnormalities and seizures in Alzheimer disease. Neurology 95, e2259–e2270.

45.

Samson

, van Duijn

, Hop

, Hofman

(1996) Clinical features and mortality in patients with early-onset Alzheimer’s disease. Eur Neurol 36, 103–106.

46.

Volicer

, Smith

, Volicer

(1995) Effect of seizures on progression of dementia of the Alzheimer type. Dementia 6, 258–263.

47.

Vossel

, Ranasinghe

, Beagle

, Mizuiri

, Honma

, Dowling

, Darwish

, Van Berlo

, Barnes

, Mantle

, Karydas

, Coppola

, Roberson

, Miller

, Garcia

, Kirsch

, Mucke

, Nagarajan

(2016) Incidence and impact of subclinical epileptiform activity in Alzheimer’s disease. Ann Neurol 80, 858–870.

48.

Groppa

, Oliviero

, Eisen

, Quartarone

, Cohen

, Mall

, Kaelin-Lang

, Mima

, Rossi

, Thickbroom

, Rossini

, Ziemann

, Valls-Sole

, Siebner

(2012) A practical guide to diagnostic transcranial magnetic stimulation: Report of an IFCN committee. Clin Neurophysiol 123, 858–882.

49.

Rossini

, Barker

, Berardelli

, Caramia

, Caruso

, Cracco

, Dimitrijevic

, Hallett

, Katayama

, Lucking

, et al. (1994) Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: Basic principles and procedures for routine clinical application. Report of an IFCN committee. Electroencephalogr Clin Neurophysiol 91, 79–92.

50.

Awiszus

(2003) TMS and threshold hunting. Suppl Clin Neurophysiol 56, 13–23.

51.

Mishory

, Molnar

, Koola

, Li

, Kozel

, Myrick

, Stroud

, Nahas

, George

(2004) The maximum-likelihood strategy for determining transcranial magnetic stimulation motor threshold, using parameter estimation by sequential testing is faster than conventional methods with similar precision. J ECT 20, 160–165.

52.

Mills

, Nithi

(1997) Corticomotor threshold to magnetic stimulation: Normal values and repeatability. Muscle Nerve 20, 570–576.

53.

Rossini

, Burke

, Chen

, Cohen

, Daskalakis

, Di Iorio

, Di Lazzaro

, Ferreri

, Fitzgerald

, George

, Hallett

, Lefaucheur

, Langguth

, Matsumoto

, Miniussi

, Nitsche

, Pascual-Leone

, Paulus

, Rossi

, Rothwell

, Siebner

, Ugawa

, Walsh

, Ziemann

(2015) Non-invasive electrical and magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: Basic principles and procedures for routine clinical and research application. An updated report from an I.F.C.N. Committee. Clin Neurophysiol 126, 1071–1107.

54.

Barker

, Jalinous

, Freeston

(1985) Non-invasive magnetic stimulation of human motor cortex. Lancet 1, 1106–1107.

55.

Ilmoniemi

, Kicic

(2010) Methodology for combined TMS and EEG. Brain Topogr 22, 233–248.

56.

Bonato

, Miniussi

, Rossini

(2006) Transcranial magnetic stimulation and cortical evoked potentials: A TMS/EEG co-registration study. Clin Neurophysiol 117, 1699–1707.

57.

Ferreri

, Pauri

, Pasqualetti

, Fini

, Dal Forno

, Rossini

(2003) Motor cortex excitability in Alzheimer’s disease: A transcranial magnetic stimulation study. Ann Neurol 53, 102–108.

58.

Pennisi

, Ferri

, Lanza

, Cantone

, Pennisi

, Puglisi

, Malaguarnera

, Bella

(2011) Transcranial magnetic stimulation in Alzheimer’s disease: A neurophysiological marker of cortical hyperexcitability. J Neural Transm (Vienna) 118, 587–598.

59.

Nardone

, Bergmann

, Kronbichler

, Kunz

, Klein

, Caleri

, Tezzon

, Ladurner

, Golaszewski

(2008) Abnormal short latency afferent inhibition in early Alzheimer’s disease: A transcranial magnetic demonstration. J Neural Transm (Vienna) 115, 1557–1562.

60.

Pierantozzi

, Panella

, Palmieri

, Koch

, Giordano

, Marciani

, Bernardi

, Stanzione

, Stefani

(2004) Different TMS patterns of intracortical inhibition in early onset Alzheimer dementia and frontotemporal dementia. Clin Neurophysiol 115, 2410–2418.

61.

Wang

, Zhang

, Han

, Zhou

(2016) Cortical function in Alzheimer’s disease and frontotemporal dementia. Transl Neurosci 7, 116–125.

62.

Alagona

, Bella

, Ferri

, Carnemolla

, Pappalardo

, Costanzo

, Pennisi

(2001) Transcranial magnetic stimulation in Alzheimer disease: Motor cortex excitability and cognitive severity. Neurosci Lett 314, 57–60.

63.

Cantone

, Di Pino

, Capone

, Piombo

, Chiarello

, Cheeran

, Pennisi

, Di Lazzaro

(2014) The contribution of transcranial magnetic stimulation in the diagnosis and in the management of dementia. Clin Neurophysiol 125, 1509–1532.

64.

Freitas

, Mondragon-Llorca

, Pascual-Leone

(2011) Noninvasive brain stimulation in Alzheimer’s disease: Systematic review and perspectives for the future. Exp Gerontol 46, 611–627.

65.

Mimura

, Nishida

, Nakajima

, Tsugawa

, Morita

, Yoshida

, Tarumi

, Ogyu

, Wada

, Kurose

, Miyazaki

, Blumberger

, Daskalakis

, Chen

, Mimura

, Noda

(2020) Neurophysiological biomarkers using transcranial magnetic stimulation in Alzheimer’s disease and mild cognitive impairment: A systematic review and meta-analysis. Neurosci Biobehav Rev 121, 47–59.

66.

National Heart, Lung, and Blood Institute, Study Quality Assessment Tools, https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools

67.

Hedges

(1981) Distribution theory for Glass’s estimator of effect size and related estimators. J Educ Stat 6, 107–128.

68.

(2019) Cochrane Handbook for Systematic Reviews of Interventions., John Wiley & Sons, Chichester (UK).

69.

Higgins

, Thompson

, Deeks

, Altman

(2003) Measuring inconsistency in meta-analyses. BMJ 327, 557–560.

70.

Di Lorenzo

, Martorana

, Ponzo

, Bonni

, D’Angelo

, Caltagirone

, Koch

(2013) Cerebellar theta burst stimulation modulates short latency afferent inhibition in Alzheimer’s disease patients. Front Aging Neurosci 5, 2.

71.

Benussi

, Di Lorenzo

, Dell’Era

, Cosseddu

, Alberici

, Caratozzolo

, Cotelli

, Micheli

, Rozzini

, Depari

, Flammini

, Ponzo

, Martorana

, Caltagirone

, Padovani

, Koch

, Borroni

(2017) Transcranial magnetic stimulation distinguishes Alzheimer disease from frontotemporal dementia. Neurology 89, 665–672.

72.

Motta

, Di Lorenzo

, Ponzo

, Pellicciari

, Bonni

, Picazio

, Mercuri

, Caltagirone

, Martorana

, Koch

(2018) Transcranial magnetic stimulation predicts cognitive decline in patients with Alzheimer’s disease. J Neurol Neurosurg Psychiatry 89, 1237–1242.

73.

Bagattini

, Mutanen

, Fracassi

, Manenti

, Cotelli

, Ilmoniemi

, Miniussi

, Bortoletto

(2019) Predicting Alzheimer’s disease severity by means of TMS-EEG coregistration. Neurobiol Aging 80, 38–45.

74.

Benussi

, Dell’Era

, Cantoni

, Ferrari

, Caratozzolo

, Rozzini

, Alberici

, Padovani

, Borroni

(2018) Discrimination of atypical parkinsonisms with transcranial magnetic stimulation. Brain Stimul 11, 366–373.

75.

Ferreri

, Vecchio

, Vollero

, Guerra

, Petrichella

, Ponzo

, Maatta

, Mervaala

, Kononen

, Ursini

, Pasqualetti

, Iannello

, Rossini

, Di Lazzaro

(2016) Sensorimotor cortex excitability and connectivity in Alzheimer’s disease: A TMS-EEG Co-registration study. Hum Brain Mapp 37, 2083–2096.

76.

Yang

, Li

, Chen

, Wang

, Gao

, Yang

, Zhao

, Wang

, Du

(2014) Evaluation of spinal cord motor function in Alzheimer’s disease using electrophysiological techniques indicates association of acetylcholine receptors with the disease. Int J Clin Exp Med 7, 5643–5649.

77.

Wegrzyn

, Teipel

, Oltmann

, Bauer

, Thome

, Grossmann

, Hauenstein

, Hoppner

(2013) Structural and functional cortical disconnection in Alzheimer’s disease: A combined study using diffusion tensor imaging and transcranial magnetic stimulation. Psychiatry Res 212, 192–200.

78.

Hoeppner

, Wegrzyn

, Thome

, Bauer

, Oltmann

, Buchmann

, Teipel

(2012) Intra- and inter-cortical motor excitability in Alzheimer’s disease. J Neural Transm (Vienna) 119, 605–612.

79.

Trebbastoni

, Gilio

, D’Antonio

, Cambieri

, Ceccanti

, de Lena

, Inghilleri

(2012) Chronic treatment with rivastigmine in patients with Alzheimer’s disease: A study on primary motor cortex excitability tested by 5 Hz-repetitive transcranial magnetic stimulation. Clin Neurophysiol 123, 902–909.

80.

Di Lazzaro

, Pilato

, Dileone

, Profice

, Marra

, Ranieri

, Quaranta

, Gainotti

, Tonali

(2008) In vivo functional evaluation of central cholinergic circuits in vascular dementia. Clin Neurophysiol 119, 2494–2500.

81.

Alberici

, Bonato

, Calabria

, Agosti

, Zanetti

, Miniussi

, Padovani

, Rossini

, Borroni

(2008) The contribution of TMS to frontotemporal dementia variants. Acta Neurol Scand 118, 275–280.

82.

Sakuma

, Murakami

, Nakashima

(2007) Short latency afferent inhibition is not impaired in mild cognitive impairment. Clin Neurophysiol 118, 1460–1463.

83.

Battaglia

, Wang

, Ghilardi

, Gashi

, Quartarone

, Friedman

, Nixon

(2007) Cortical plasticity in Alzheimer’s disease in humans and rodents. Biol Psychiatry 62, 1405–1412.

84.

Inghilleri

, Conte

, Frasca

, Scaldaferri

, Gilio

, Santini

, Fabbrini

, Prencipe

, Berardelli

(2006) Altered response to rTMS in patients with Alzheimer’s disease. Clin Neurophysiol 117, 103–109.

85.

Liepert

, Bar

, Meske

, Weiller

(2001) Motor cortex disinhibition in Alzheimer’s disease. Clin Neurophysiol 112, 1436–1441.

86.

Pepin

, Bogacz

, de Pasqua

, Delwaide

(1999) Motor cortex inhibition is not impaired in patients with Alzheimer’s disease: Evidence from paired transcranial magnetic stimulation. J Neurol Sci 170, 119–123.

87.

Di Lazzaro

, Oliviero

, Tonali

, Marra

, Daniele

, Profice

, Saturno

, Pilato

, Masullo

, Rothwell

(2002) Noninvasive in vivo assessment of cholinergic cortical circuits in AD using transcranial magnetic stimulation. Neurology 59, 392–397.

88.

Julkunen

, Jauhiainen

, Westeren-Punnonen

, Pirinen

, Soininen

, Kononen

, Paakkonen

, Maatta

, Karhu

(2008) Navigated TMS combined with EEG in mild cognitive impairment and Alzheimer’s disease: A pilot study. J Neurosci Methods 172, 270–276.

89.

Julkunen

, Jauhiainen

, Kononen

, Paakkonen

, Karhu

, Soininen

(2011) Combining transcranial magnetic stimulation and electroencephalography may contribute to assess the severity of Alzheimer’s disease. Int J Alzheimers Dis 2011, 654794.

90.

Balla

, Maertens de Noordhout

, Pepin

(2014) An investigation of the late excitatory potentials in the hand following transcranial magnetic stimulation in early Alzheimer’s disease. Dement Geriatr Cogn Dis Extra 4, 457–464.

91.

Di Lazzaro

, Pilato

, Dileone

, Profice

, Ranieri

, Ricci

, Bria

, Tonali

, Ziemann

(2007) Segregating two inhibitory circuits in human motor cortex at the level of GABAA receptor subtypes: A TMS study. Clin Neurophysiol 118, 2207–2214.

92.

Di Lazzaro

, Pilato

, Dileone

, Saturno

, Oliviero

, Marra

, Daniele

, Ranieri

, Gainotti

, Tonali

(2006) In vivo cholinergic circuit evaluation in frontotemporal and Alzheimer dementias. Neurology 66, 1111–1113.

93.

Koch

, Esposito

, Codeca

, Mori

, Kusayanagi

, Monteleone

, Di Lorenzo

, Bernardi

, Martorana

(2011) Altered dopamine modulation of LTD-like plasticity in Alzheimer’s disease patients. Clin Neurophysiol 122, 703–707.

94.

Martorana

, Stefani

, Palmieri

, Esposito

, Bernardi

, Sancesario

, Pierantozzi

(2008) L-dopa modulates motor cortex excitability in Alzheimer’s disease patients. J Neural Transm (Vienna) 115, 1313–1319.

95.

Martorana

, Mori

, Esposito

, Kusayanagi

, Monteleone

, Codeca

, Sancesario

, Bernardi

, Koch

(2009) Dopamine modulates cholinergic cortical excitability in Alzheimer’s disease patients. Neuropsychopharmacology 34, 2323–2328.

96.

Khedr

, Ahmed

, Darwish

, Ali

(2011) The relationship between motor cortex excitability and severity of Alzheimer’s disease: A transcranial magnetic stimulation study. Neurophysiol Clin 41, 107–113.

97.

Khedr

, Ahmed

, Sayed

, Abo-Elfetoh

, Ali

, Gomaa

(2020) Electrophysiological differences in cortical excitability in different forms of dementia: A transcranial magnetic stimulation and laboratory biomarkers study. Neurophysiol Clin 50, 185–193.

98.

Di Lazzaro

, Oliviero

, Pilato

, Saturno

, Dileone

, Marra

, Daniele

, Ghirlanda

, Gainotti

, Tonali

(2004) Motor cortex hyperexcitability to transcranial magnetic stimulation in Alzheimer’s disease. J Neurol Neurosurg Psychiatry 75, 555–559.

99.

Perretti

, Grossi

, Fragassi

, Lanzillo

, Nolano

, Pisacreta

, Caruso

, Santoro

(1996) Evaluation of the motor cortex by magnetic stimulation in patients with Alzheimer disease. J Neurol Sci 135, 31–37.

100.

Alagona

, Ferri

, Pennisi

, Carnemolla

, Maci

, Domina

, Maertens de Noordhout

, Bella

(2004) Motor cortex excitability in Alzheimer’s disease and in subcortical ischemic vascular dementia. Neurosci Lett 362, 95–98.

101.

de Carvalho

, de Mendonca

, Miranda

, Garcia

, Luis

(1997) Magnetic stimulation in Alzheimer’s disease. J Neurol 244, 304–307.

102.

Benussi

, Grassi

, Palluzzi

, Koch

, Di Lazzaro

, Nardone

, Cantoni

, Dell’Era

, Premi

, Martorana

, di Lorenzo

, Bonni

, Ranieri

, Capone

, Musumeci

, Cotelli

, Padovani

, Borroni

(2020) Classification accuracy of transcranial magnetic stimulation for the diagnosis of neurodegenerative dementias. Ann Neurol 87, 394–404.

103.

Irizarry

, Jin

, He

, Emond

, Raman

, Thomas

, Sano

, Quinn

, Tariot

, Galasko

, Ishihara

, Weil

, Aisen

(2012) Incidence of new-onset seizures in mild to moderate Alzheimer disease. Arch Neurol 69, 368–372.

104.

Komssi

, Kahkonen

, Ilmoniemi

(2004) The effect of stimulus intensity on brain responses evoked by transcranial magnetic stimulation. Hum Brain Mapp 21, 154–164.

105.

Babiloni

, Blinowska

, Bonanni

, Cichocki

, De Haan

, Del Percio

, Dubois

, Escudero

, Fernandez

, Frisoni

, Guntekin

, Hajos

, Hampel

, Ifeachor

, Kilborn

, Kumar

, Johnsen

, Johannsson

, Jeong

, LeBeau

, Lizio

, Lopes da Silva

, Maestu

, McGeown

, McKeith

, Moretti

, Nobili

, Olichney

, Onofrj

, Palop

, Rowan

, Stocchi

, Struzik

, Tanila

, Teipel

, Taylor

, Weiergraber

, Yener

, Young-Pearse

, Drinkenburg

, Randall

(2020) What electrophysiology tells us about Alzheimer’s disease: A window into the synchronization and connectivity of brain neurons. Neurobiol Aging 85, 58–73.

106.

Jelic

, Johansson

, Almkvist

, Shigeta

, Julin

, Nordberg

, Winblad

, Wahlund

(2000) Quanti-breaktative electroencephalography in mild cognitive impairment: Longitudinal changes and possible prediction of Alzheimer’s disease. Neurobiol Aging 21, 533–540.

107.

Soininen

, Partanen

, Paakkonen

, Koivisto

, Riekkinen

(1991) Changes in absolute power values of EEG spectra in the follow-up of Alzheimer’s disease. Acta Neurol Scand 83, 133–136.

108.

Coben

, Danziger

, Storandt

(1985) A longitudinal EEG study of mild senile dementia of Alzheimer type: Changes at 1 year and at 2.5 years. Electroencephalogr Clin Neurophysiol 61, 101–112.

109.

Babiloni

, Frisoni

, Pievani

, Vecchio

, Lizio

, Buttiglione

, Geroldi

, Fracassi

, Eusebi

, Ferri

, Rossini

(2009) Hippocampal volume and cortical sources of EEG alpha rhythms in mild cognitive impairment and Alzheimer disease. Neuroimage 44, 123–135.

110.

Helkala

, Hanninen

, Hallikainen

, Kononen

, Laakso

, Hartikainen

, Soininen

, Partanen

, Vainio

, Riekkinen

Sr. (1996) Slow-wave activity in the spectral analysis of the electroencephalogram and volumes of hippocampus in subgroups of Alzheimer’s disease patients. Behav Neurosci 110, 1235–1243.

111.

Paula-Lima

, Brito-Moreira

, Ferreira

(2013) Deregulation of excitatory neurotransmission underlying synapse failure in Alzheimer’s disease. J Neurochem 126, 191–202.

112.

Gillessen

, Budd

, Lipton

(2002) Excitatory amino acid neurotoxicity. Adv Exp Med Biol 513, 3–40.

113.

Mattson

(2003) Excitotoxic and excitoprotective mechanisms: Abundant targets for the prevention and treatment of neurodegenerative disorders. Neuromolecular Med 3, 65–94.

114.

Tamagnini

, Novelia

, Kerrigan

, Brown

, Tsaneva-Atanasova

, Randall

(2015) Altered intrinsic excitability of hippocampal CA1 pyramidal neurons in aged PDAPP mice. Front Cell Neurosci 9, 372.

115.

Busche

, Eichhoff

, Adelsberger

, Abramowski

, Wiederhold

, Haass

, Staufenbiel

, Konnerth

, Garaschuk

(2008) Clusters of hyperactive neurons near amyloid plaques in a mouse model of Alzheimer’s disease. Science 321, 1686–1689.

116.

Busche

, Chen

, Henning

, Reichwald

, Staufenbiel

, Sakmann

, Konnerth

(2012) Critical role of soluble amyloid-beta for early hippocampal hyperactivity in a mouse model of Alzheimer’s disease. Proc Natl Acad Sci U S A 109, 8740–8745.

117.

Vogt

, Thomas

, Galvan

, Bredesen

, Lamb

, Pimplikar

(2011) Abnormal neuronal networks and seizure susceptibility in mice overexpressing the APP intracellular domain. Neurobiol Aging 32, 1725–1729.

118.

Palop

, Chin

, Mucke

(2006) A network dysfunction perspective on neurodegenerative diseases. Nature 443, 768–773.

119.

Kuchibhotla

, Goldman

, Lattarulo

, Wu

, Hyman

, Bacskai

(2008) Abeta plaques lead to aberrant regulation of calcium homeostasis in vivo resulting in structural and functional disruption of neuronal networks. Neuron 59, 214–225.

120.

Abramov

, Dolev

, Fogel

, Ciccotosto

, Ruff

, Slutsky

(2009) Amyloid-beta as a positive endogenous regulator of release probability at hippocampal synapses. Nat Neurosci 12, 1567–1576.

121.

Cirrito

, Yamada

, Finn

, Sloviter

, Bales

, May

, Schoepp

, Paul

, Mennerick

, Holtzman

(2005) Synaptic activity regulates interstitial fluid amyloid-beta levels in vivo. Neuron 48, 913–922.

122.

Busche

, Konnerth

(2015) Neuronal hyperactivity–A key defect in Alzheimer’s disease? Bioessays 37, 624–632.

123.

, Yarishkin

, Hwang

, Chun

, Park

, Woo

, Bae

, Kim

, Lee

, Chun

, Park

, Lee

, Hong

, Kim

, Oh

, Park

, Lee

, Yoon

, Kim

, Jeong

, Shim

, Bae

, Cho

, Kowall

, Ryu

, Hwang

, Kim

, Lee

(2014) GABA from reactive astrocytes impairs memory in mouse models of Alzheimer’s disease. Nat Med 20, 886–896.

124.

Palop

, Mucke

(2010) Amyloid-beta-induced neuronal dysfunction in Alzheimer’s disease: From synapses toward neural networks. Nat Neurosci 13, 812–818.

125.

Fritschy

(2008) Epilepsy, E/I balance and GABA(A) receptor plasticity. Front Mol Neurosci 1, 5.

126.

Scarmeas

, Honig

, Choi

, Cantero

, Brandt

, Blacker

, Albert

, Amatniek

, Marder

, Bell

, Hauser

, Stern

(2009) Seizures in Alzheimer disease: Who, when, and how common? Arch Neurol 66, 992–997.

127.

Amatniek

, Hauser

, DelCastillo-Castaneda

, Jacobs

, Marder

, Bell

, Albert

, Brandt

, Stern

(2006) Incidence and predictors of seizures in patients with Alzheimer’s disease. Epilepsia 47, 867–872.

128.

Kleen

, Scott

, Holmes

, Roberts

, Rundle

, Testorf

, Lenck-Santini

, Jobst

(2013) Hippocampal interictal epileptiform activity disrupts cognition in humans. Neurology 81, 18–24.

129.

Lott

, Doran

, Nguyen

, Tournay

, Movsesyan

, Gillen

(2012) Down syndrome and dementia: Seizures and cognitive decline. J Alzheimers Dis 29, 177–185.

130.

Trebbastoni

, Pichiorri

, D’Antonio

, Campanelli

, Onesti

, Ceccanti

, de Lena

, Inghilleri

(2015) Altered cortical synaptic plasticity in response to 5-Hz repetitive transcranial magnetic stimulation as a new electrophysiological finding in amnestic mild cognitive impairment converting to Alzheimer’s disease: Results from a 4-year prospective cohort study. Front Aging Neurosci 7, 253.

131.

Zhang

, Yang

, Lu

, Yang

, Zhang

(2017) PET/MR imaging: New frontier in Alzheimer’s disease and other dementias. Front Mol Neurosci 10, 343.

132.

Agosta

, Pievani

, Geroldi

, Copetti

, Frisoni

, Filippi

(2012) Resting state fMRI in Alzheimer’s disease: Beyond the default mode network. Neurobiol Aging 33, 1564–1578.

133.

Kumar

, Zomorrodi

, Ghazala

, Goodman

, Blumberger

, Cheam

, Fischer

, Daskalakis

, Mulsant

, Pollock

, Rajji

(2017) Extent of dorsolateral prefrontal cortex plasticity and its association with working memory in patients with Alzheimer disease. JAMA Psychiatry 74, 1266–1274.

134.

Luckhaus

, Grass-Kapanke

, Blaeser

, Ihl

, Supprian

, Winterer

, Zielasek

, Brinkmeyer

(2008) Quantitative EEG in progressing vs stable mild cognitive impairment (MCI): Results of a 1-year follow-up study. Int J Geriatr Psychiatry 23, 1148–1155.

135.

Babiloni

, Benussi

, Binetti

, Cassetta

, Dal Forno

, Del Percio

, Ferreri

, Ferri

, Frisoni

, Ghidoni

, Miniussi

, Rodriguez

, Romani

, Squitti

, Ventriglia

, Rossini

(2006) Apolipoprotein E and alpha brain rhythms in mild cognitive impairment: A multicentric electroencephalogram study. Ann Neurol 59, 323–334.

136.

Claus

, Ongerboer De Visser

, Bour

, Walstra

, Hijdra

, Verbeeten

Jr. , Van Royen

, Kwa

, van Gool

(2000) Determinants of quantitative spectral electroencephalography in early Alzheimer’s disease: Cognitive function, regional cerebral blood flow, and computed tomography. Dement Geriatr Cogn Disord 11, 81–89.

137.

Lee

, Park

, Kim

, Im

(2010) Global synchronization index as a biological correlate of cognitive decline in Alzheimer’s disease. Neurosci Res 66, 333–339.

138.

Akilan

, Kumar

, Zomorrodi

, Blumberger

, Daskalakis

, Rajji

(2020) Gender impact on transcranial magnetic stimulation-based cortical excitability and cognition relationship in healthy individuals. Neuroreport 31, 287–292.

139.

Inghilleri

, Conte

, Curra

, Frasca

, Lorenzano

, Berardelli

(2004) Ovarian hormones and cortical excitability. An rTMS study in humans. Clin Neurophysiol 115, 1063–1068.

140.

Bolu

, Balikci

, Erdem

, Oznur

, Celik

, Uzun

(2015) Cortical excitability and aggressive behavior in post-traumatic stress disorder.. Noro Psikiyatr Ars 52, 73–77.

141.

Daskalakis

, Christensen

, Chen

, Fitzgerald

, Zipursky

, Kapur

(2002) Evidence for impaired cortical inhibition in schizophrenia using transcranial magnetic stimulation. Arch Gen Psychiatry 59, 347–354.

142.

Fitzgerald

, Brown

, Daskalakis

, Kulkarni

(2002) A transcranial magnetic stimulation study of inhibitory deficits in the motor cortex in patients with schizophrenia. Psychiatry Res 114, 11–22.

143.

Greenberg

, Ziemann

, Cora-Locatelli

, Harmon

, Murphy

, Keel

, Wassermann

(2000) Altered cortical excitability in obsessive-compulsive disorder. Neurology 54, 142–147.

144.

Aalten

, Verhey

, Boziki

, Brugnolo

, Bullock

, Byrne

, Camus

, Caputo

, Collins

, De Deyn

, Elina

, Frisoni

, Holmes

, Hurt

, Marriott

, Mecocci

, Nobili

, Ousset

, Reynish

, Salmon

, Tsolaki

, Vellas

, Robert

(2008) Consistency of neuropsychiatric syndromes across dementias: Results from the European Alzheimer Disease Consortium. Part II. Dement Geriatr Cogn Disord 25, 1–8.

145.

Craig

, Mirakhur

, Hart

, McIlroy

, Passmore

(2005) A cross-sectional study of neuropsychiatric symptoms in 435 patients with Alzheimer’s disease. Am J Geriatr Psychiatry 13, 460–468.

146.

Geda

, Smith

, Knopman

, Boeve

, Tangalos

, Ivnik

, Mrazek

, Edland

, Petersen

(2004) De novo genesis of neuropsychiatric symptoms in mild cognitive impairment (MCI). Int Psychogeriatr 16, 51–60.

147.

Lyketsos

, Steinberg

, Tschanz

, Norton

, Steffens

, Breitner

(2000) Mental and behavioral disturbances in dementia: Findings from the Cache County Study on Memory in Aging. Am J Psychiatry 157, 708–714.

148.

Scaricamazza

, Colonna

, Sancesario

, Assogna

, Orfei

, Franchini

, Sancesario

, Mercuri

, Liguori

(2019) Neuropsychiatric symptoms differently affect mild cognitive impairment and Alzheimer’s disease patients: A retrospective observational study. Neurol Sci 40, 1377–1382.

149.

Koppel

, Sunday

, Goldberg

, Davies

, Christen

, Greenwald

, Alzheimer’s Disease Neuroimaging I (2014) Psychosis in Alzheimer’s disease is associated with frontal metabolic impairment and accelerated decline in working memory: Findings from the Alzheimer’s Disease Neuroimaging Initiative. Am J Geriatr Psychiatry 22, 698–707.

150.

Rapp

, Schnaider-Beeri

, Grossman

, Sano

, Perl

, Purohit

, Gorman

, Haroutunian

(2006) Increased hippocampal plaques and tangles in patients with Alzheimer disease with a lifetime history of major depression. Arch Gen Psychiatry 63, 161–167.

151.

Rapp

, Schnaider-Beeri

, Purohit

, Perl

, Haroutunian

, Sano

(2008) Increased neurofibrillary tangles in patients with Alzheimer disease with comorbid depression. Am J Geriatr Psychiatry 16, 168–174.

152.

Engelborghs

, Maertens

, Marien

, Vloeberghs

, Somers

, Nagels

, De Deyn

(2006) Behavioural and neuropsychological correlates of frontal lobe features in dementia. Psychol Med 36, 1173–1182.

153.

Zubenko

, Moossy

, Martinez

, Rao

, Claassen

, Rosen

, Kopp

(1991) Neuropathologic and neurochemical correlates of psychosis in primary dementia. Arch Neurol 48, 619–624.

154.

Supasitthumrong

, Bolea-Alamanac

, Asmer

, Woo

, Abdool

, Davies

SJC

(2019) Gabapentin and pregabalin to treat aggressivity in dementia: A systematic review and illustrative case report. Br J Clin Pharmacol 85, 690–703.

155.

Cooney

, Murphy

, Tessema

, Freyne

(2013) Use of low-dose gabapentin for aggressive behavior in vascular and mixed vascular/Alzheimer dementia. J Neuropsychiatry Clin Neurosci 25, 120–125.

156.

Lonergan

, Luxenberg

(2009) Valproate preparations for agitation in dementia. Cochrane Database Syst Rev CD003945.

157.

Gauthier

, Wirth

, Mobius

(2005) Effects of memantine on behavioural symptoms in Alzheimer’s disease patients: An analysis of the Neuropsychiatric Inventory (NPI) data of two randomised, controlled studies. Int J Geriatr Psychiatry 20, 459–464.