Cortical excitability changes as a marker of cognitive impairment in Parkinson’s disease

Abstract

Cognitive impairment of different severity with eventual progression to dementia in Parkinson’s disease (PD) appears during the course of the disease. In this study, transcranial magnetic stimulation (TMS) was used to assess cortical excitability changes in PD patients with varying cognitive impairment. We aimed to identify the TMS parameters that could serve as a non-invasive marker of cognitive impairment in patients with PD. Consecutive PD patients were recruited in the study. Detailed neuropsychological assessment was carried out to identify PD without cognitive impairment (PD-nC), PD with mild cognitive impairment (PD-MCI) and PD with dementia (PDD). Twenty patients of PDD (2 females and 18 males), 20 PD-MCI (4 females and 16 males), 18 PD-nC (5 females, 13 males) and 18 healthy controls (4 females, and 14 males) were included in the study. All the participants underwent TMS with recording of resting motor threshold, central motor conduction time, silent period, short interval intracortical inhibition (SICI) and intracortical facilitation (ICF). All the groups were age matched. The SICI was present in all; however, significantly greater inhibition was noted in PDD (Mean±SD; 0.11 ± 0.08) followed by PD-MCI (0.31 ± 0.17), PD-nC (0.49 ± 0.26) and controls (0.61 ± 0.23; p < 0.001). The ICF was significantly reduced in PDD (Mean±SD; 0.15 ± 0.18), PD-MCI (0.55 ± 0.31), PD-nC (0.96 ± 0.59), when compared to healthy controls (1.81 ± 0.83; p < 0.001). Patients with PD-nC, PD-MCI and PDD had graded reduction in ICF and increasing intracortical inhibition as the disease progressed from PD-nC through PD-MCI to PDD. This suggests progressive overactivity of GABAergic transmission, glutaminergic deficiency with consequent reduction of cholinergic transmission leading to dementia.

1. Introduction

Parkinson’s disease (PD) is a common neurodegenerative disorder with early death of dopaminergic neurons in the substantia nigra pars compacta (SNpc). In addition to the motor symptoms, PD is associated with a broad spectrum of non-motor symptoms [1–3]. Among the various non-motor features of PD, cognitive impairment of different severity and the eventual progression to dementia is known to be a common condition appearing at some point along the course of the disease [4]. The high frequency and devastating impact of cognitive deficits in PD has increasingly been recognized in recent years [3]. At the time of diagnosis, nearly all patients with PD present with some degree of cognitive impairment not severe enough to significantly affect functional independence [2,5]. However, mild cognitive changes have a measurable impact on functional capacity in PD [6]. These early signs of cognitive deterioration are, in many cases, difficult to capture with common screening methods [5,7]. It highlights those early mild cognitive deficits associated with PD are, in many cases, not clinically apparent. However, there is heterogeneity in the presentation of PD with mild cognitive impairment (PD-MCI) and dementia (PDD) and quite possibly in their prognostic implications. The deposition of cortical and subcortical Lewy bodies is implicated in the pathophysiology of cognitive impairment in PD [8]. Recent clinicopathological studies have demonstrated a synergistic role for Alzheimer-type pathology in PDD [9]. Alterations in GABAergic and cholinergic circuits contribute to cognitive deficits observed in several psychiatric and neurological diseases [10]. The pathophysiology of PD-MCI is less characterized with an increased risk of progression to dementia. Imaging and post-mortem studies support a cholinergic basis for cognitive impairment, although other neurotransmitters are most likely involved. Hence there is a need to develop non-invasive tools to determine cognitive impairment in these patients that may help in prognosticating and developing appropriate therapeutic strategies. In this context, transcranial magnetic stimulation (TMS) has become a unique tool to assess distinct intracortical circuits in the CNS and indirectly determine the function of GABAergic, and glutamatergic cortical circuits [11].

TMS has translational role in many neurodegenerative disorders including sleep disorders such as restless leg syndrome (RLS). TMS studies have demonstrated a hyper glutamatergic state with deficiency of GABAergic inhibition in patients with RLS [12]. In addition, abnormalities in multiple TMS parameters can differentiate different types of dementia and parkinsonian disorders [13].

Previous studies have shown normal to reduced RMT in PD patients [14–16]. Medication and surgical procedures does not seem to affect the RMT [16]. The SP is found to be reduced in patients with PD [17]. SICI has been found to be reduced in PD which gets normalized post dopaminergic therapy as well as through STN DBS [16]. SICI is thought to be mediated by GABAA receptors while SP is through the GABAB receptors. This suggests that there is an abnormal GABAergic transmission in PD.

In the present study we aimed to study the intracortical inhibition and facilitation in addition to the cortical excitability changes in patients with PD without cognitive impairment (PD-nC), PD-MCI and PDD using TMS and to determine whether these neurophysiological tools can serve as marker in identifying the disease categories.

2. Material and methods

The study was conducted in the departments of Neurology and Clinical Psychology at the National institute of Mental health and Neuro Sciences (NIMHANS), Bengaluru, India. The study was approved by the Institute’s ethics committee (IEC No. NIMHANS/IEC (BS & NS DIV.) 5th MEETING/2017) and written informed consent was obtained from all the participants. Consecutive patients of PD (n = 58) and 18 healthy controls were recruited in the study. The diagnosis of PD was based on the UKPDS brain bank criteria (2015) [18]. All the patients were levodopa responsive and neuroimaging was done to rule out atypical or secondary parkinsonism. The relevant demographic details were recorded. All the patients were clinically examined in detail by the Movement Disorders Specialists. The severity of disease was assessed using the Unified Parkinson’s disease rating scale (UPDRS) in the “OFF” (12 h off medication) and “ON” state (1−2 h after acute levodopa challenge test). Levodopa equivalent daily dose (LEDD) was calculated for each patient [19,20]. Patients with history of epilepsy, presence of any metallic implants, cardiac pacemakers, cochlear implants in the body were excluded. Patients with associated chronic medical illness such as cardiovascular, renal and hepatic diseases, uncontrolled diabetes mellitus were not included in the study. In addition, patients whose UPDRS-part III “OFF” state upper limb rest tremor score of > 2 were also excluded from the study. Handedness of the subjects were determined using Edinburgh’s handedness inventory [21]. PD patients were evaluated with neuropsychological tests and based on the assessment, patients were categorized into PD-nC (n = 18), PD-MCI (n = 20) and PDD (n = 20).

All the participants subsequently underwent single and paired pulse TMS (Fig. 1).

Fig. 1. Study design.

2.1. Neuropsychological assessment

Detailed neuropsychological evaluation was performed by the clinical psychologists (SH, MG, VN) in the medication “ON” state using select tests from the NIMHANS Neuropsychological battery [22] and Wechsler memory scale III (WMS) [23]. The following domains were assessed: speed (motor and mental), attention (focussed and sustained), executive functions (working memory, planning, fluency, set shifting, response inhibition), learning and memory, and visuo-spatial construction. The data was compared with age, education and gender based normative data. A score below the 15th percentile score (1 SD below the mean) was taken as the cut off score [24]. Patients were categorized into 3 groups as: PD-nC, PD-MCI and PDD. The criterion used was adopted from the MDS criteria [25]. The patients were classified as PD-MCI if they were impaired on 2 tests in one cognitive domain or one impaired test in 2 different domains and PDD when they had impairment in 3 domains [26].

Patients with PD-MCI were found to have impairment in learning, memory (verbal and visual), motor and mental speed, whereas patients with PDD had impairment of several cognitive domains such as motor and mental speed, executive functions (fluency, planning and set-shifting ability), learning and memory (verbal and visual).

2.2. Transcranial magnetic stimulation

TMS was administered in the medication OFF state using Magstim 200 stimulator and hand-held butterfly coil (70 mm size). Left motor cortex was stimulated after identifying the optimal scalp position (hot spot). Motor evoked potentials (MEP) was recorded from the contralateral first dorsal interosseus (FDI) muscle using two Ag−AgCl electrodes placed in a belly tendon montage. Audio-visual feedback was used to ensure lack of significant muscle activity. The stimulus intensity was increased in a step wise manner to obtain a satisfactory MEP which was defined as the minimal stimulus intensity that elicited a response of at least 50μV amplitude peak to peak. Ten consecutive responses were recorded and saved. Both single pulse (for resting motor threshold ‘RMT’, central motor conduction time ‘CMCT’, contralateral and ipsilateral silent periods ‘cSP and iSP’) and paired pulse stimulation (for short interval intracortical inhibition ‘SICI’ and intracortical facilitation ‘ICF’) paradigms were used to record the TMS parameters. The details of the TMS experiments were as per the recommendations of the IFCN committee and have been published elsewhere and briefly described below [27,28].

RMT (expressed as %) was defined as the minimum stimulus intensity that elicited an MEP in the resting muscle (FDI) of at least 50μV amplitude peak to peak in 5 out of 10 stimuli. CMCT was calculated as the difference in the MEP latencies obtained after cortical and spinal stimulation. The stimulus intensity used was 120% of RMT. Spinal stimulus was done above the C7 vertebral spinous process. cSP was obtained using suprathreshold stimuli (120% of RMT) over the left motor cortex with voluntary partial contraction (approximately 30% of the maximal contraction) of the right FDI muscle, whereas for iSP the muscle was fully contracted. SICI and ICF were determined using a subthreshold conditioning stimulus (80% of RMT) followed by suprathreshold test stimulus (120%) at an interstimulus interval (ISI) of 2 msec and 10 msec (Fig. 2).

Fig. 2. TMS protocol used in the study.

2.3. Statistical analysis

The statistical analysis was performed in R software (version 3.6.0). The normality of data was confirmed using the Shapiro-Wilk test (p ≤ 0.05). For comparison of means among patients and controls, one way-ANOVA was used along with Tukey HSD post hoc test. A Pearson’s correlations were performed to check the relationship between clinical variables and TMS measures.

3. Results

3.1. Demographic characteristics

The mean age of patients was 59.8 ± 8.1 years for PDD, 56.4 ± 8.4 years for PD-MCI, 55.1 ± 7.8 years for PD-nC, and 58 ± 5.1 years for controls. The mean UPDRS III score (OFF and ON), LEDD, H & Y stage of all the patient groups are given in Table 1. The mean MoCA score of patients with PD-nC was 26.08 ± 3.09, for PD-MCI was 25 ± 2.37, for PDD was 24 ± 2.61 and healthy controls was 29.12 ± 0.87. There was a significant difference in the MoCA scores between PD-nC and PDD (p = 0.05), however there was no significant difference between PD-MCI and PD-nC (p = 0.54) and PDD (p = 0.58).

Table 1. Demographic and clinical characteristics of the study participants.

	HC	PD-nC	PD-MCI	PDD	Significance
	(n = 18)	(n = 18)	(n = 20)	(n = 20)	PD-nC vs PD-MCI vs PDD vs HC
	(Mean±SD)	(Mean±SD)	(Mean±SD)	(Mean±SD)
Age (years)	58 ± 5.1	55.1 ± 7.8	56.4 ± 8.4	59.8 ± 8.1	0.25
Age at onset (years)	—	47.7 ± 8.3	51.3 ± 8.2	51.6 ± 8.7	0.3
Duration of PD (years)	—	6.4 ± 3.9	5 ± 2.9	7.7 ± 3.7	0.06
UPDRS III (OFF)	—	29.36 ± 15.18	31.15 ± 17.54	34.15 ± 11.50	0.60
UPDRS III (ON)	—	11.69 ± 7.43	15.58 ± 14.98	14.90 ± 9.05	0.52
H & Y stage	—	2.36 ± 1.17	2.25 ± 0.62	2.53 ± 0.88	0.63
LEDD(mg/day)	—	772.64 ± 379.96	731.26 ± 300.87	761.40 ± 307.63	0.92

*p < 0.05,

* *p < 0.01 (Tukey-HSD). HC: Healthy controls, H & Y stage: Hoehn and Yahr staging, LEDD: Levodopa equivalent daily dosage, PD-nC: Parkinson’s disease without cognitive impairment, PDD: Parkinson’s disease with dementia, PD-MCI: Parkinson’s disease with mild cognitive impairment, UPDRS III: Unified Parkinson’s disease rating scale

3.2. Transcranial magnetic stimulation

There was no significant difference observed between the groups for the RMT, CMCT, cSP and iSP (Tables 2 and 3). SICI was present in all the study subjects with significantly greater inhibition noted in PDD (Mean±SD; 0.11 ± 0.08) followed by PD-MCI (0.31 ± 0.17) and PD-nC (0.49 ± 0.26) when compared to controls (0.61 ± 0.23; p < 0.001). There was a significant reduction in the ICF in all the PD subjects when compared to controls (p < 0.001). Severe reduction in ICF was seen in patients with PDD (Mean±SD; 0.15 ± 0.18) followed by PD-MCI (0.55 ± 0.31) and PD-nC (0.96 ± 0.59). ICF was preserved in healthy controls (1.81 ± 0.83) (Fig. 3).

Table 2. Comparison of TMS parameters of patients with PDD, PD-MCI, PD and HC.

	HC	PD-nC	PD-MCI	PDD	Significance
	(N = 18)	(N = 18)	(N = 20)	(N = 20)	PDD vs PD-MCI vs PD-nC vs HC
	(Mean±SD)	(Mean±SD)	(Mean±SD)	(Mean±SD)
RMT (%)	40.22 ± 5.15	39.83 ± 9.10	42.75 ± 7.92	39.90 ± 7.32	0.56
CMCT (msec)	8.49 ± 0.99	8.35 ± 2.46	7.12 ± 2.26	7.71 ± 2.46	0.18
cSP (msec)	113.71 ± 13.78	124.70 ± 39.38	128.57 ± 42.89	122.96 ± 31.20	0.59
iSP (msec)	32.58 ± 9.77	33.05 ± 11.16	36.33 ± 12.36	38.54 ± 12.63	0.34
SICI	0.61 ± 0.23	0.49 ± 0.26	0.31 ± 0.17	0.11 ± 0.08	0.00 **
ICF	1.81 ± 0.83	0.96 ± 0.59	0.55 ± 0.31	0.15 ± 0.18	0.00 **

^*p < 0.05,

^**p < 0.01 (Tukey-HSD). CMCT: central motor conduction time, cSP: contralateral silent period, HC: healthy controls, ICF: intracortical facilitation, iSP: ipsilateral silent period, PD-nC: Parkinson’s disease without cognitive impairment, PDD: Parkinson’s disease with dementia, PD-MCI: Parkinson’s disease with mild cognitive impairment, RBD: REM sleep behaviour disorder, RMT: resting motor threshold, SICI: short interval intracortical inhibition

Table 3. p-values obtained on comparing the TMS parameters between the study groups.

	PDD vs	PDD vs	PDD vs	PD-MCI vs	PD-MCI	PD-nC
	PD-MCI	PD-nC	HC	PD-nC	vs HC	vs HC
RMT	1.00	1.00	1.00	1.00	1.00	1.00
CMCT	1.00	1.00	1.00	1.00	1.00	1.00
cSP	1.00	1.00	1.00	1.00	1.00	1.00
iSP	1.00	0.93	0.64	1.00	1.00	1.00
SICI	0.05 *	0.00 **	0.00 **	0.01 *	0.00 **	0.27
ICF	0.05 *	0.00 **	0.00 **	0.05 *	0.00 **	0.00 **

^*p < 0.05,

^**p < 0.01 (Tukey-HSD). CMCT: central motor conduction time, cSP: contralateral silent period, HC: healthy controls, ICF: intracortical facilitation, iSP: ipsilateral silent period, PD-nC: Parkinson’s disease without cognitive impairment, PDD: Parkinson’s disease with dementia, PD-MCI: Parkinson’s disease with mild cognitive impairment, RMT: resting motor threshold, SICI: short interval intracortical inhibition

Fig. 3. Comparison of SICI and ICF in the study participants.

3.3. Correlation results

In the PD group, the mean SICI and ICF positively correlated with the mean UPDRS III OFF (r = 0.41, p < 0.05; r = 0.19, p < 0.05) and ON score (r = 0.66, p < 0.05; r = 0.37, p < 0.05).

In the PD-MCI group, the mean SICI and ICF negatively correlated with the mean UPDRS III OFF (r = −0.04, p < 0.05; (r = −0.18, p < 0.05) and ON score (r = −0.07, p < 0.05; r = −0.23, p < 0.05).

In the PDD group, the mean SICI and ICF positively correlated with the mean UPDRS III OFF (r = 0.17, p < 0.05; r = 0.38, p < 0.05) and ON score (r = 0.20, p < 0.05; r = 0.24, p < 0.05).

4. Discussion

In our study we found a significant enhancement in SICI and reduced ICF in all the PD groups with marked abnormalities noted in patients with PDD which is contrary to the previous studies. There was no difference in the RMT and SP between the patient groups and controls. Patients with PD-nC PD-MCI and PDD had graded reduction in ICF and progressively greater inhibition observed as the disease progresses from PD-nC through PD-MCI and PDD.

Studies of TMS in PD-MCI and PDD are very few and have focussed mainly on RMT and short latency afferent inhibition (SAI) that is reduced [29,30].

Cortical excitability changes in PD-MCI patients have been investigated in few studies that has shown either reduced [31–34] or normal SICI in these patients [35,36]. In another study normalization of reduced SICI was observed after levodopa administration [31].

In a recent study, patients with PD have reduced SICI even in the early stage with no difference in the ICF [37]. The results are contradictory to our results wherein we found enhanced SICI and reduced ICF. The TMS technique employed is similar in both the studies. The difference in the results could be due to various causes such as: small sample size of our study, our PD patients are slightly younger compared to the other study and earlier age at presentation. There could also be racial and ethnic differences in the cortical excitability changes between the two-patient population.

PDD is a late complication of PD and has a cumulative prevalence of 75−90% in those with a disease duration of 10 years or more [38]. This is due to spread of α-synuclein aggregation into limbic and neocortical brain regions.

The striatum is the most important input nucleus of the basal ganglia that receives afferent from the cortex (motor area, supplementary motor area and primary somatosensory areas) that are mainly glutaminergic and cholinergic. These inputs are mainly glutamatergic excitatory synapses. The second major afferents to the striatum are the dopaminergic fibers from the substantia nigra. The striatum is also composed of two populations of medium spiny neurons that mediate GABAergic neurotransmission [39].

In addition, the indirect pathway also uses glutamate as the neurotransmitter. The striatal GABAergic circuits are controlled by the cholinergic interneurons which participate not only in movement but also in attention and reinforcement-related mechanisms [40]. Recently a novel class of fast-adapting GABAergic interneurons and slow GABAergic interneurons have been discovered in the striatum [40].

SICI is believed to be a measure of GABA_A receptors. GABA receptors are localized on the forebrain cholinergic neurones that project towards the cortex [41,42]. Previous studies have shown that GABA antagonists enhance release of acetylcholine at basal forebrain level [43] and the cerebral cortex [44]. Furthermore, GABA, through GABA_A receptors, tonically inhibits cholinergic activity both at cortical and subcortical level. A study done by [45] demonstrated that the enhancement of GABAA activity produced by lorazepam is associated with a reduction of SAI, a form of inhibition that is believed to involve muscarinic cholinergic activity [45].

Besides, the results have been variable with respect to ICF. While some studies have shown reduced ICF in PD [33,34,46], others have found ICF to be normal in these patients [31,47–49]. Studies have demonstrated, ICF to be mediated through the glutamatergic circuits [46–48,50–54].

Studies have demonstrated prolongation of SP in patients with PD a feature that suggests GABAergic overactivity [55]. In patients with isolated RBD which is known to precede the development of neurode-generative disease also has reduced ICF further strengthening our findings [56].

A significant increase in the inhibition and reduction in facilitation in our study suggests that there is overactivity of the GABAergic neurons and reduction of glutamatergic transmission in the cortex and striatum. Hence, the reduced glutamatergic input (reduced ICF) to the striatum causes increased activity of the striatal GABAergic neurons (increased SICI) which reduces the cholinergic transmission in the basal forebrain through GABA_A receptors.

This suggests that the enhancement of SICI and reduction of ICF in our study is associated with cognitive impairment in PDD patients due to cholinergic inhibition (Fig. 4).

Fig. 4. Mechanism of cholinergic deficiency in PD patients.

The measurements of the motor cortex by TMS reflects the cognition status of the patient. This could be due to the involvement of projections fibres between the motor cortex and other regions of the brain and also, there could be a network effect.

In patients with subcortical ischemic diseases, TMS has shown preservation of the cholinergic and transcallosal inhibition suggesting a different pathophysiologic mechanism of cognitive impairment compared to neurodegenerative Parkinson’s disease [57,58].

Our study is limited by small sample size. SAI was not measured in our patients which could have given information about the cholinergic deficiency.

There was no scale used to assess the functional independence and mood status, which could affect the cortical excitability.

A recent study has demonstrated that about 20−26 trials are required to accurately measure the RMT, SICI and ICF [59]. However, in our study we obtained a consistent MEP response and hence only 10 trials were recorded.

Another limitation is that this is a cross sectional analysis of three groups of PD patients with normal to varying degree of cognitive impairment and not a longitudinal follow up study of the same cohort. Future studies should explore the pattern of changes of cognitive and TMS parameters over time in a large cohort of PD patients with baseline normal cognition.

5. Conclusions

Our study has shown that PD patients with increasing cognitive dysfunction have progressively greater short interval intracortical inhibition and reduced intracortical facilitation. Patients with PDD have significantly greater inhibition and reduced facilitation when compared to PD-MCI and PD-nC. This suggests that in addition to cholinergic deficiency, these patients tend to develop enhanced GABAergic and reduced glutamatergic neurotransmission during the disease evolution from PD-nC through PD-MCI and PDD. Hence TMS parameters can be reliably used as a marker of clinical progression in patients with PD.