Cerebrovascular co-pathology and cholinergic white matter pathways along the Lewy body continuum

doi:10.21203/rs.3.rs-4296476/v1

Background:Dementia with Lewy Bodies (DLB) often presents with cholinergic degeneration and varying degrees of cerebrovascular disease (CVD). Radiological methods for assessing cholinergic degeneration in DLB are limited. We investigated the potential of the radiological Cholinergic Pathway Hyperintensities Scale (CHIPS) in identifying CVD-related disruptions in cholinergic white matter pathways. We compared the utility of CHIPS to other radiological and automated methods of evaluation.

Methods: We included 82 individuals (41 patients along the clinical continuum of Lewy body (LB) disease and 41 healthy controls) from the Sant Pau Initiative on Neurodegeneration cohort. We evaluated white matter signal abnormalities (WMSA) and cholinergic pathways integrity on MRI using two clinical radiological methods (CHIPS, Fazekas) and two automated methods (tractography mean diffusivity, FreeSurfer WMSA volume). Between and within study groups, we investigated associations among those measures, as well as regional atrophy, cerebrospinal fluid (CSF) biomarkers of AD co-pathology, core clinical features, and cognitive measures.

Results: Cholinergic white matter methods (CHIPS and tractography) showed a strong correlation, as did global WMSA methods (FreeSurfer WMSA volume and Fazekas). LB patients showed a significantly higher degree of WMSA in the external capsule cholinergic pathway compared to controls, without global WMSA differences. Receiver operating characteristic (ROC) analysis revealed that the CHIPS score in the posterior external capsule and the mean diffusivity in both external capsule and cingulum exceeded the threshold for an optimal biomarker in discriminating LB patients from controls. Higher CHIPS scores, Fazekas scores, and mean diffusivity from tractography were associated with more pronounced frontal atrophy in LB patients. No significant associations were found between the four WMSA and integrity methods and any of the core clinical features, cognitive measures, or CSF biomarkers in the LB group.

Conclusions: Cholinergic WMSA were more pronounced in LB patients than in healthy controls, above and beyond global WMSA. Mechanistically, CVD in cholinergic white matter may be implicated in frontal atrophy along the LB continuum. Clinically, we demonstrate the potential of CHIPS to assess cholinergic WMSA. Our findings show that CVD co-pathology may contribute to the well-known cholinergic degeneration in DLB, allowing for vascular health-focused therapies from prodromal DLB onwards.

Dementia with Lewy bodies

cholinergic white matter pathways

CHIPS

white matter signal abnormalities

cerebrovascular disease.

Dementia with Lewy Bodies (DLB) is the second most common cause of neurodegenerative dementia, accounting for up to 5–20% of dementia cases [1–4]. Although DLB is primarily driven by alpha-synuclein-related pathology, cerebrovascular disease (CVD) is a common co-pathology that influences neurodegeneration and the clinical phenotype in DLB [5–8]. White matter signal abnormalities (WMSA) on magnetic resonance imaging (MRI) are a key marker of CVD. Assessments of WMSA are commonly used in clinical practice and can predict cognitive impairment and the development of dementia in DLB [9–11]. The clinical presentation in DLB might be partially explained by the localization and degree of WMSA, impacting specific cognitive domains [12]. In particular, WMSA affecting cholinergic white matter pathways characteristically impact cognition in Alzheimer’s disease (AD), vascular dementia, and Parkinson’s disease (PD) with dementia [13,14]. However, the interplay between WMSA and the cholinergic system in DLB, a disease with prominent cholinergic dysfunction, is less understood.

The cholinergic system is known to be impaired early in DLB [15]. Recent studies by Schumacher et al. using probabilistic tractography have shown that the integrity of cholinergic white matter pathways is related to global cognition and attention in DLB [16,17]. These findings may have clinical implications, as cholinergic dysfunction can be treated with cholinesterase inhibitors [18–20]. However, probabilistic tractography is a complex MRI technique and is not readily available in the clinical radiological evaluation of dementia. Therefore, an easier and more accessible radiological method to assess the integrity of cholinergic pathways in dementia at large, and DLB in particular, would be of great clinical value.

The Cholinergic Pathway Hyperintensities Scale (CHIPS) is a semiquantitative visual rating scale for the radiological assessment of cholinergic pathways on MRI.[21] CHIPS has previously been used in individuals with AD and PD [13,22–25], and, to our knowledge, in only one previous study including DLB patients [24]. The utility of CHIPS in the prodromal stages of DLB remains to be investigated.

The first aim of this study was to investigate the association of CHIPS with the integrity of cholinergic white matter pathways from tractography as well as with other common MRI markers of WMSA. In particular, we included two markers of global WMSA: one is clinically available and consists of radiological visual ratings (the Fazekas scale), and the other one is research-oriented and consists of fully automated FreeSurfer estimations of global WMSA volume. This approach provided clinically available and research-oriented measures analogous to the regional measures included for the cholinergic system (CHIPS and tractography, respectively). Further, CHIPS served as a regional measure of WMSA. Hence, we mapped global and regional WMSA with both clinically available and research-oriented MRI methods. To gain further mechanistic understanding, the second aim of this study was to investigate associations of CHIPS and the other MRI markers with regional brain atrophy, CSF biomarkers of AD co-pathology, core clinical features of DLB, and cognitive measures. All analyses were conducted in individuals within the clinical continuum of DLB. Specifically, we included DLB patients as well as patients with mild cognitive impairment of Lewy body type (MCI-LB), which is the most frequent presentation of prodromal DLB [26]. For comparison, we included a group of healthy controls.

Study Population

We included 41 patients with a clinical diagnosis of either probable DLB (17 patients) or MCI-LB (24 patients), combined into a single LB group for statistical analysis. All LB patients were recruited at the Sant Pau Memory Unit (Barcelona, Spain), as part of the Sant Pau Initiative on Neurodegeneration (SPIN) cohort [27], between June 2013 and December 2017. Specific details on the evaluation protocol are described elsewhere [27]. In brief, all individuals enrolled in the SPIN cohort underwent evaluation by neurologists with expertise in neurodegenerative diseases using an extensive neurological and neuropsychological protocol. Patients also provided MRI and CSF samples.

Moreover, 41 healthy individuals from the SPIN cohort were selected as the control group. Control individuals underwent a thorough medical history review, physical examination, and a standard neuropsychological evaluation as previously described [27]. To be recognized as cognitively unimpaired, control individuals had to meet the criteria of having no subjective cognitive complaints, a mini-mental state examination (MMSE) score of 27–30, a clinical dementia rate (CDR) global score of 0, an free and cued selective reminding test (FCSRT) total immediate adjusted-score of ≥ 7, and no impairments in daily living activities. Additional details about the inclusion/exclusion criteria and neuropsychological tests used in the SPIN cohort are provided in previous publications [27].

Magnetic resonance imaging (MRI) scanning

A high-resolution 3D T1-weighted magnetization-prepared rapid gradient echo (MPRAGE) sequence, a fluid-attenuated inversion recovery (FLAIR) sequence, and a diffusion-weighted imaging (DWI) sequence were acquired for each individual at the Hospital de la Santa Creu i Sant Pau (Barcelona, Spain). All 41 LB patients and 41 controls underwent MRI scanning using the 3T Philips Achieva scanner (Philips Healthcare). The scanning parameters are provided in the Additional file 1.

Global and Regional Radiological Measures of WMSA

The radiological measures included the Fazekas scale for global WMSA and CHIPS for regional cholinergic WMSA.

The Fazekas scale was assessed on FLAIR images using the modified Fazekas scale for periventricular and deep white matter regions [28]. Periventricular Fazekas scores ranged from 0 to 3: 0, absence; 1, caps or pencil-thin lining; 2, smooth halo; and 3, irregular periventricular WMSA extending into the deep white matter. Deep Fazekas scores ranged from 0–3: 0, absence; 1, punctuated foci; 2, a beginning confluence of foci; and 3, large confluent areas. In case of discrepancies between the scores in periventricular and deep white matter, the Fazekas scale adopts the highest periventricular or deep WMSA score. For statistical analyses, Fazekas scores of 0 or 1 were classified as a low-WMSA group, and Fazekas scores of 2 or 3 were classified as a high-WMSA group, following the approach of previous studies [29].

CHIPS was assessed on axial sections of FLAIR images [21]. CHIPS is based on anatomical landmarks from immunohistochemical studies and evaluates WMSA in the medial (cingulum) and lateral (external capsule) cholinergic pathways. MRIs are rated at four different standardized axial levels: the lower external capsule, the higher external capsule, the corona radiata, and the centrum semiovale with separate ratings for anterior and posterior regions at each level (Figure 1A). Each region is rated using a three-point system (0 = normal; 1 = <50% involvement; 2≥50% involvement). A factor of 4 is attributed to the lower standardized axial level with linearly decreasing factors for each successive higher level, to account for decreasing cholinergic fiber density. The maximum possible score is 50 for each hemisphere, yielding a total CHIPS score of 100 points. In this study, we used the total CHIPS score as well as CHIPS subregions anterior external capsule, posterior external capsule, and cingulum for statistical analyses.

All images were rated by an experienced neuroradiologist (A.T.) and a radiology resident (C.J.), to ensure interrater agreement. The raters performed the ratings separately using the same hardware and software and were blinded to any information about the study individuals. In cases of disagreement between the two raters, a consensus for the final score was reached through discussion. Interrater agreement between the two radiologists was calculated and expressed as quadratic weighted Kappa coefficients. The benchmark proposed by Landis et al. [30] was used to interpret the level of agreement between the raters. Kappa scores close to 0.0 were interpreted as poor agreement, 0.01-0.20 as slight agreement, 0.21-0.40 as fair agreement, 0.41-0.60 as moderate agreement, 0.61-0.80 as substantial agreement, and 0.81-1.0 as almost perfect agreement.

Automated segmentation of global WMSA assessed with FreeSurfer

We used FreeSurfer v7.3.1 Image Analysis software to segment white matter hypointensities on T1-weighted images as a volumetric marker of WMSA [31]. We previously demonstrated the significant and strong association between white matter hypointensities on T1-weighted images and hyperintensities on T2/FLAIR sequences [32]. FreeSurfer employs an algorithm that assigns a label to each voxel based on probabilistic local and intensity‐related information that is automatically estimated from a built‐in training dataset comprising 41 manually segmented images (surfer.nmr.mgh.harvard.edu/fswiki/AtlasSubjects; [31]). This approach has exhibited good sensitivity in evaluating white matter alterations both in healthy individuals and AD patients [33,34]. Additionally, FreeSurfer software was used to estimate the total intracranial volume (TIV) for each individual. Data was organized and processed through the TheHiveDB system [35].

Tractography analysis of cholinergic white matter pathways

The procedure for tractography analysis has been described in a previous study [36]. Briefly, the diffusion-weighted imaging data were preprocessed using FSL (FMRIB Software Library) [37]. This included the removal of non-brain tissue and correction for EPI distortion, eddy currents, and head motion [38,39]. Probabilistic tracking was performed by repeating 5000 random samples from each of the voxels within a mask of the nucleus basalis of Meynert [40]. Only the tracts traversing through mid-way cingulum or external capsule masks were retained (Figure 1B). Next, an unbiased template was created based on preprocessed b0 images, and the cingulum and external capsule cholinergic pathways of all study individuals were non-linearly warped into the space of the unbiased template. Finally, pathway-specific binary masks were created. To characterize the microstructure properties of the tracked cholinergic pathways, we extracted the index of mean diffusivity (MD) from the diffusion tensor model, consistent with previous studies using this tractography method [16,17,36,41,42].

Measures of regional brain atrophy

To assess the degree of atrophy on MRI, we used AVRA (Automatic Visual Ratings of Atrophy), an artificial intelligence-based method trained on ratings from an experienced neuroradiologist [43]. AVRA has been shown to mimic the neuroradiologist’s rating procedure and achieves similar levels of inter-rater agreement to that obtained between two experienced neuroradiologists, with the advantage that intra-rater AVRA agreements are always 100% [43]. AVRA provides fast and automatic ratings for the Scheltens’ scale of medial temporal atrophy (MTA), the frontal subscale of Pasquier’s Global Cortical Atrophy (GCA-F) scale, and the Koedam’s scale of Posterior Atrophy (PA) [43]. Another advantage of AVRA is that it provides both continuous and categorical measures of atrophy. We applied previously published clinical cutoffs to further classify AVRA scores as normal or abnormal [44]. Specifically, MTA scores ≥1.5, ≥1.5, ≥2, and ≥2.5 were considered abnormal for the respective age ranges of 45–64, 65–74, 75–84, and 85–94 years. For PA and GCA-F, a score of ≥1 was deemed abnormal regardless of the age range [44].

CSF biomarkers

Lumbar puncture and CSF analysis procedures are detailed in previous publications [27]. Core AD biomarkers were measured using the Lumipulse G β‐Amyloid 1‐42, β‐Amyloid 1‐40, Total Tau and pTau 181 assays on LUMIPULSE G600II automated platform (Fujirebio) [45]. Cutoff values for these core AD biomarkers were based on previously published 18F‐Florbetapir PET optimized cutoff values [45].

Statistical analysis

Demographic and clinical characteristics were reported as means and standard deviations for continuous variables and counts and percentages for categorical variables.

To test for group differences and associations among CHIPS, Fazekas, FreeSurfer WMSA volume, and tractography MD values, as well as associations with regional atrophy, CSF biomarkers, core clinical features, and cognitive measures, we used Pearson’s correlations, independent samples T-test, and analysis of variance (ANOVA) for normally distributed variables, and Spearman’s correlations and the Mann-Whitney test for non-normally distributed variables. The Chi-square was used for categorical measures. Effect sizes were expressed as Cohen’s d. When within-group analyses for associations revealed statistically significant results in both LB and healthy control groups separately, we conducted ANOVA/ANCOVA with an interaction term between MRI marker ‘X’ and diagnostic group (LB patients vs. controls) in predicting MRI marker ‘Y’, to determine if the association between X and Y was significantly stronger in the LB group than in healthy controls.

As there was a statistically significant difference in mean age and education between the LB group and the healthy controls, comparisons between the two groups were adjusted for age and education using ANCOVA. FreeSurfer WMSA volumes were additionally adjusted for the TIV using ANCOVA.

Receiver operating characteristic (ROC) curves were generated for CHIPS subregions’ scores and Fazekas's scores to discriminate the LB group from the healthy control group. The area under the curve (AUC) value was reported for each measure, along with 95% confidence intervals, cutoff values, and sensitivity and specificity values. Youden’s J statistic was performed to determine the cutoff values with the highest corresponding sensitivity and specificity.

Statistical significance in all analyses was set at P < .05 for all analyses. All statistical analyses were conducted using IBM SPSS statistics v26 (IBM Corp., Armonk, NY, USA)

Cohort characteristics

Individuals’ characteristics are listed in Table 1. The LB and control groups were statistically comparable in sex distribution (p = 0.269), while the mean age was higher in LB patients than in controls (p < 0.001). The LB group also had fewer years of education than the control group (p < 0.001). As expected, compared with the controls, the LB group showed worse performance in most of the clinical and cognitive measures (p < 0.05), except for the direct digits span and reverse digits span (p = 0.19 and p = 0.057, respectively). The data on core clinical features, regional brain atrophy, and CSF biomarkers are shown in Table 2.

Table 1: Cohort characteristics.

	Lewy body group (n = 41)			Healthy controls (n = 41)
	N	Mean (SD)/Count (%)	N	Mean (SD)/Count (%)	p-value
Age, years	41	75.9 (5.4)	41	67.6 (6.2)	<0.001
min-max		58-85		60-86
Sex, female	41	19 (46%)	41	24 (59%)	0.269
Education, years	41	9.6 (5.0)	41	15 (4.4)	<0.001
Visual hallucinations, presence	41	18 (44%)	41	0 (0%)	<0.001
RBD, presence	41	25 (61%)	41	0 (0%)	<0.001
Cognitive fluctuations, presence	41	32 (78%)	41	0 (0%)	<0.001
Parkinsonism, presence	41	40 (98%)	41	0 (0%)	<0.001
Unified Parkinson's disease rating scale	33	18.8 (10.6)	0	-	-
Geriatric depression scale	38	11.7 (6.3)	0	-	-
Neuropsychiatric inventory total score	24	14.8 (12.3)	29	0.4 (0.9)	<0.001
Global deterioration scale	41	3.5 (0.6)	0	-	-
Clinical dementia rate, 0/0.5/1/2/3	38	1 (2.6%)/27 (73.7%)/8 (18.7%)/2 (5.3%)/0 (0%)	37	36 (97.7%)/1 (2.7%)/0 (0%)/0 (0%)/0 (0%)	<0.001
Mini-mental state examination	40	24.9 (3.6)	40	28.9 (1.1)	<0.001
Digits span - direct	38	4.5 (1.0)	40	5.3 (0.9)	0.19
Digits span - reverse	38	3.2 (1.0)	40	4.5 (1.1)	0.057
Boston naming Test	36	43.3 (7.7)	40	55.4 (3.4)	<0.001
Semantic fluency	38	11.3 (4.1)	40	19.6 (4.0)	<0.001
Phonetic fluency	38	7.7 (4.1)	40	15.8 (5.1)	<0.001
FCSRT total free recall	37	10.4 (8.4)	40	27.3 (6.2)	<0.001
FCSRT total recall	37	26.8 (13.6)	40	44.3 (3.5)	<0.001
Visual objects and space perception battery	38	6.5 (2.8)	40	9.2 (1.2)	0.001
DAT scan, positive	26	20 (50%)	41	0 (0%)	<0.001

FCSRT = Free and Cued Selective Reminding Test; RBD = Rapid eye movement sleep behavioral disorder; DAT = Dopamine active transporter.

Table 2: WMSA and CSF biomarkers.

		Lewy body group (n = 41)		Healthy controls (n = 41)	p-value
	N	Mean (SD)/Count (%)	N	Mean (SD)/Count (%)
CHIPS total	41	30.9 (15.9)	41	9.8 (9.3)	<0.001
CHIPS external capsule	41	28.9 (14.2)	41	9.1 (8.9)	<0.001
CHIPS external capsule - anterior	41	15.0 (7.6)	41	6.4 (5.6)	0.021
CHIPS external capsule - posterior	41	13.9 (7.8)	41	2.7 (4.2)	<0.001
CHIPS cingulum	41	2.0 (2.7)	41	0.7 (1.4)	0.377
High WMSA (Fazekas 2-3)	41	25 (61%)	41	11 (26.8%)	0.126
Freesurfer WMSA	40	5855 (5057)	39	2210 (2185)	0.492
Tractography cingulum	37	0.00109 (0.00011)	37	0.00095 (0.00007)	0.040
Tractography external capsule	37	0.00129 (0.00011)	37	0.00112 (0.00009)	0.031
MTA	41	1.61 (0.75)	41	0.65 (0.47)	<0.001
MTA (% abnormal)	41	39.0	41	5.0	<0.001
PA	41	0.64 (0.5)	41	0.50 (0.46)	0.741
PA (% abnormal)	41	24.4	41	7.5	0.795
GCA-F	41	0.22 (0.28)	41	0.02 (0.09)	0.012
GCA-F (% abnormal)	41	2.4	41	0.0	0.888
AB42-40 ratio	40	0.066 (0.03)	31	0.125 (0.19)	0.308
AB42-40 ratio (% abnormal)	40	60%	31	19%	0.048
Total tau	40	471 (253)	31	361 (260)	0.377
Total tau (% abnormal)	40	43%	31	19%	0.784
Phosphorylated tau	40	77.6 (49.7)	31	55.3 (55.1)	0.300
Phosphorylated tau (% abnormal)	40	50%	31	19%	0.225

CHIPS = Cholinergic pathways hyperintensities scale; WMSA = White matter signal abnormalities; MTA = Medial temporal atrophy; PA = Parietal atrophy; GCA-F = Global cortical atrophy - frontal; AB42-40 ratio = Amyloid beta 1-42/Amyloid beta 1-40 ratio.

Interrater agreement

The weighted Kappa for CHIPS was 0.93, while Fazekas's weighted Kappa was 0.88. Radiologists thus showed almost perfect agreement on both scales.

Associations with age, sex, and education

Please see Figure 2 for a summary of the findings and Additional file 2 for p-values and effect sizes.

Higher Fazekas and CHIPS scores were associated with a higher age in LB patients (p = 0.042 and p = 0.038, respectively), while in controls a higher CHIPS score was associated with higher age (p = 0.010) but the Fazekas score was not (p = 0.460). Fazekas and CHIPS scores were not associated with sex or years of education in any of the two groups (p > 0.05).

A higher FreeSurfer global WMSA volume (TIV adjusted) was associated with a higher age both in LB patients and controls (p = 0.002 and p = 0.005, respectively); but it was not associated with sex in LB patients and controls (p = 0.356 and 0.993, respectively). The FreeSurfer global WMSA volume was not associated with years of education in any of the two groups (p > 0.05).

Regarding cholinergic white matter pathways (tractography), in LB patients, a higher age was associated with a higher mean diffusivity in both cingulum and external capsule cholinergic pathways (p = 0.005 and p = 0.002, respectively). Similarly, higher age was associated with a higher mean diffusivity in both cingulum and external capsule cholinergic pathways in controls (p = 0.004 and p = 0.001, respectively). Mean diffusivity in the external capsule cholinergic pathway was higher in males in both the LB group and healthy controls (p = 0.028 and p = 0.045, respectively). There was no statistically significant difference in mean diffusivity of the cingulum between males and females in both the LB group and healthy controls (p > 0.05). In healthy controls, a higher mean diffusivity in the external capsule cholinergic pathway was associated with fewer years of education (p = 0.038). Apart from that finding, mean diffusivity was not associated with years of education in the LB group or healthy controls (p > 0.05).

Group differences in WMSA markers and white matter cholinergic pathways (tractography)

All the analyses in this section were adjusted for age and education (and additionally for TIV for FreeSurfer WMSA).

Fazekas scores were statistically comparable between LB patients and controls (61% vs. 27%, p = 0.126) (Table 2) and there was no statistically significant difference in FreeSurfer global WMSA volume between LB patients and controls (p = 0.492).

In contrast, the mean total CHIPS score was higher in the LB group compared with the controls (30.9 ± 15.9 vs 9.8 ± 9.3; p < 0.001). Further, LB patients presented with higher CHIPS scores in the external capsule (p < 0.001) including anterior (p = 0.012) and posterior (p < 0.001) subregions, but not in cingulum (p = 0.377), when compared with controls (Figure 3A).

LB patients presented with a statistically significantly higher mean diffusivity in the external capsule cholinergic pathway (p = 0.031), and in the cingulum cholinergic pathway (p = 0.04), compared with controls (Figure 3B).

Associations among WMSA markers and white matter cholinergic pathways (tractography)

Overall, we observed more significant associations in LB patients than in controls (see summary of findings in Figure 2).

Higher CHIPS scores were associated with higher Fazekas scores both in LB patients (p < 0.001) and controls (p < 0.001). The test for the statistical interaction showed that the association between CHIPS scores and Fazekas scores was not stronger in LB compared with that in controls (p = 0.093). The same pattern of results was observed for all CHIPS subregions (data not shown). Similarly, higher CHIPS scores were statistically significantly associated with higher FreeSurfer global WMSA volume (TIV adjusted) both in LB patients (p < 0.001) and controls (p = 0.028). The test for the statistical interaction showed that this association was not stronger in the LB group compared with that in controls (p = 0.388).

In LB patients, a higher Fazekas score was statistically significantly associated with a higher FreeSurfer global WMSA volume (p < 0.001), while in controls, there was no significant association between FreeSurfer global WMSA and Fazekas scores (p = 0.414) (Figure 3D).

In LB patients, higher CHIPS scores were associated with a higher mean diffusivity in both cingulum and external capsule cholinergic pathways (p = 0.008 and p = 0.007, respectively). Similarly, higher Fazekas scores were associated with a higher mean diffusivity in both cingulum and external capsule cholinergic pathways (p = 0.002 and p = 0.001, respectively). In contrast, in controls, the mean diffusivity of cholinergic pathways was not associated with CHIPS or Fazekas scores (p > 0.05) (Figure 3C, E).

Diagnostic utility of CHIPS in discriminating LB patients from controls

The results of ROC analysis are summarized in Figure 4 and fully reported in Additional file 3, including sensitivity and specificity values. In the ROC analysis for the discrimination between LB and controls, the CHIPS score in the posterior external capsule showed the greatest area under the curve (AUC = 0.887), followed closely by mean diffusivity in both external capsule (AUC = 0.873). The CHIPS score in the total external capsule had the next highest area under the curve (AUC = 0.868) followed by the total CHIPS score (AUC = 0.861). The lowest AUC was shown by the Fazekas score (AUC = 0.667) and the CHIPS score in the cingulum (AUC = 0.652). The ROC analysis for FreeSurfer global WMSA volume showed an AUC of 0.810, and the CHIPS score in the anterior external capsule showed an AUC of 0.797.

Associations with Regional Brain Atrophy

Table 2 shows continuous atrophy scores for MTA, GCA-F, and PA, as well as abnormal scores after using clinical cutoffs. 39% of LB patients showed abnormal scores in MTA vs 5% in controls, 24.4% showed abnormal scores in PA vs 7.5% in controls, and 2.4% showed abnormal scores in GCA-F vs 0% in controls. Subsequent analyses were performed on continuous values.

Overall, we observed more significant associations in LB patients than in controls (see summary of findings in Figure 2).

CHIPS scores were associated with more frontal atrophy (GCA-F, p = 0.011) in the LB group. CHIPS scores were not significantly associated with medial temporal or posterior atrophy in the LB group (MTA and PA, p > 0.05). In controls, CHIPS scores were not associated with MTA, PA, or GCA-F (p > 0.05).

In LB patients, higher Fazekas scores were associated with more atrophy in the medial temporal lobe (MTA, p = 0.023) and more frontal atrophy (GCA-F, p = 0.008), but not more posterior atrophy (PA, p = 0.810). In controls, Fazekas scores were not associated with MTA, PA, or GCA-F (p > 0.05).

In LB patients, a FreeSurfer global WMSA volume (TIV adjusted) was not associated with medial temporal atrophy (MTA, p = 0.235), frontal atrophy (GCA-F, p = 0.798) or posterior atrophy (PA, p = 0.237). In controls, a higher FreeSurfer global WMSA volume (TIV adjusted) was associated with more medial temporal atrophy (MTA, p = 0.001) and frontal atrophy (GCA-F, p < 0.001), but not more posterior atrophy (PA, p = 0.266).

In LB patients, a higher mean diffusivity in both external capsule and cingulum cholinergic pathways was associated with more medial temporal atrophy (MTA, p < 0.001 and p = 0.001, respectively) and more frontal atrophy (GCA-F, p < 0.001 and p < 0.001, respectively), but not with more posterior atrophy (PA, p = 0.349 and p = 0.749, respectively). In controls, a higher mean diffusivity in the cingulum cholinergic pathway was associated with more medial temporal atrophy (MTA, p = 0.027) and marginally more posterior atrophy (PA, p = 0.05), but not with more frontal atrophy (GCA-F, p = 0.350). A higher mean diffusivity in the external capsule cholinergic pathway was associated with more posterior atrophy (PA, p = 0.013), but not more frontal or medial temporal atrophy (GCA-F, MTA, p > 0.05 for both). See Additional file 2 for full details on these analyses.

Associations with CSF biomarkers, clinical features, and cognitive performance

CHIPS, Fazekas, FreeSurfer global WMSA volume, and cholinergic white matter pathways (tractography) did not show any statistically significant association with any of the CSF biomarkers, core clinical features, nor the clinical or cognitive measures within the LB group or healthy controls (p > 0.05).

In this study, we investigated the association of CHIPS with other MRI measures of global WMSA and cholinergic degeneration, as well as with regional brain atrophy, CSF biomarkers of AD co-pathology, core clinical features, and cognitive measures in patients within the clinical LB continuum.

We found that LB patients exhibited a significantly higher degree of WMSA in cholinergic white matter compared with controls. In particular, LB patients had significantly more WMSA in both the anterior and posterior regions of the external capsule, but not in the cingulum, as indicated by CHIPS subdomains. The CHIPS findings in the external capsule matched the results obtained from the tractography method, which also demonstrated more pronounced cholinergic degeneration in the external capsule when comparing LB patients with controls. Regarding mechanistic interpretations, Barber et al. postulated that WMSA in deep white matter, such as the external capsule, may be related to CVD, whereas WMSA in periventricular white matter could relate to neurodegeneration [46]. Therefore, our findings suggest the preferential vulnerability of cholinergic pathways to CVD co-pathology in the clinical LB continuum and highlight the key role cholinergic degeneration may play in the disease manifestation [7,47,48]. Additionally, our findings align with the only previous study using CHIPS in DLB, which also reported significantly higher CHIPS scores in DLB patients compared to controls, particularly affecting the external capsule [24].

Regarding the cingulum cholinergic pathway, tractography analysis exhibited a significantly higher cingulum mean diffusivity in LB compared with controls, whereas the CHIPS score could not capture any statistically significant difference in cingulum between the two groups. Prior studies suggested that tractography (relying on diffusion tensor imaging) might be more sensitive to subtle white matter changes compared with visual rating methods for WMSA [49–51]. Tractography may therefore capture early microstructural white matter changes that are not yet detectable by the CHIPS. However, the two previous studies investigating white matter cholinergic pathways in DLB have been inconclusive, with one study showing a higher cingulum mean diffusivity in DLB patients compared with controls [17], while the other study showed no difference between the two groups.[16] Therefore, in addition to different sensitivities of CHIPS and tractography, cholinergic pathways may not be uniformly affected along the LB continuum, with overt degeneration of the cingulum possibly occurring later in the disease process. Given the relatively early stage of the disease in our cohort, we may have only captured changes in the cingulum with the tractography method and not with CHIPS. PET studies have shown that the cingulate gyrus metabolism is relatively preserved in DLB, further supporting this notion [52]. The cingulate island sign, which refers to the relative preservation of posterior cingulate metabolism in relation to precuneus and cuneus metabolism, is a supportive biomarker in the diagnostic criteria of DLB [53]. It has been hypothesized that the preservation of the posterior cingulate in DLB is due to relatively preserved blood perfusion in that area, and neuropathological studies reported low alpha-synuclein-related pathology in the posterior cingulate compared to anterior cingulate in DLB [54–56].

Our results pertaining to cholinergic white matter are relevant in the context of our findings on global WMSA. We observed a qualitatively higher degree of WMSA (as indicated by Fazekas and FreeSurfer WMSA) in LB individuals compared with controls. However, these differences were not statistically significant after adjusting for age (and education). Consequently, we infer that global WMSA, as estimated by the Fazekas and the FreeSurfer methods are not specifically associated with the LB pathology and might rather reflect age-related changes. This interpretation is reinforced by the absence of a statistical interaction between age and diagnostic group in predicting global WMSA measures (both Fazekas and FreeSurfer WMSA, data not shown). This suggests that age contributes equally to global WMSA both in controls and individuals with LB in our cohort. Previous studies reported that age is a significant risk factor for WMSA [7,57,58]. In summary, the absence of statistical differences in global WMSA contrasted with greater cholinergic WMSA and cholinergic degeneration in LB compared with controls, which suggests that cholinergic pathways may be particularly vulnerable to WMSA in LB, likely independent of global WMSA.

Our ROC analysis demonstrated that the posterior external capsule CHIPS scores and tractography (external capsule and cingulum) showed the greatest AUC values. The CHIPS score in the posterior external capsule and mean diffusivity in the cingulum surpassed the threshold for an optimal biomarker in discriminating patients from controls, as both achieved sensitivity and specificity values above 80% [59]. Additionally, the total CHIPS score exhibited a high specificity (92%) at a lower sensitivity (72%), further reinforcing the clinical utility of CHIPS in identifying LB-specific findings, here approached through the discrimination between LB patients and controls. However, the lower AUC values for the Fazekas score and CHIPS score in the cingulum suggest that these measures may be less effective for discriminating LB patients from controls in clinical routine, although they may still be useful in characterizing other aspects of white matter damage and neurodegeneration in LB patients.

We also investigated associations among MRI measures to gain deeper mechanistic insights into CHIPS and LB pathology. We found a significant association between the two cholinergic methods, i.e. CHIPS and tractography. This correlation suggests that CHIPS is a reliable clinical method and could facilitate the assessment of WMSA in the cholinergic white matter in clinical routine, especially considering the complexity and challenges of implementing tractography clinically. Additionally, we observed an association between both global WMSA methods, i.e., Fazekas scale and FreeSurfer, in LB patients. This finding aligns with previous research that reported a correlation between these two methods [32,60–62]. Notably, all these associations were more pronounced in LB patients compared to controls, highlighting their potential to signal abnormal processes.

The associations with regional brain atrophy are intriguing, particularly in understanding how the more apparent grey matter neurodegenerative process in LB relates to CVD co-pathology. We found that higher CHIPS scores are associated with increased frontal atrophy, but not with medial temporal or posterior brain atrophy, in the LB group. These results align with previous DLB studies that reported associations between global WMSA and frontal cortex atrophy [7,8]. Our current study expands upon this previous research by incorporating the cholinergic aspect of WMSA and by further characterizing this association using global WMSA measures and tractography. We observed associations with all these measures for frontal atrophy. A similar pattern was observed for medial temporal atrophy, with the addition of a statistically significant association between CHIPS and medial temporal atrophy. This insight is particularly relevant in the context of the recent literature discussing the association of WMSA with medial temporal atrophy in DLB, which could be considered inconsistent. In our earlier study we showed no association between global WMSA and cortical thickness in medial temporal areas [7]. However, a recent publication reported an association between global WMSA and volume in the fusiform cortex [8], which is adjacent to and partially extends into medial temporal areas. We also demonstrated that this association might be influenced not only by CVD but also by AD co-pathology. Specifically, we found that atrophy in the fusiform could be affected by two mechanisms separately involving CVD and AD co-pathologies [8]. We supported this finding with another recent study using a different method for assessing WMSA and medial temporal atrophy [63], where we showed that the association between higher Fazekas scores and scores on the MTA scale was influenced by AD co-pathology in DLB patients [63]. Thus, the variability in the presence and severity of CVD and AD co-pathologies in DLB may explain the differing abilities to capture associations between CVD and medial temporal atrophy across different cohorts. Indeed, we recently reported that the DLB atrophy subtype with the highest degree of WMSA also exhibited more diffuse brain atrophy, including atrophy in the medial temporal lobe [64]. Furthermore, the absence of a significant association between CHIPS and medial temporal atrophy suggests that it is not specifically a CVD mechanism targeting the cholinergic system that may lead to medial temporal atrophy in LB. Instead, overall CVD burden and the presence of AD co-pathology may be necessary to drive neurodegeneration in the medial temporal regions in DLB [8,63]. The impact of AD co-pathology in medial temporal regions is well recognized along the LB continuum [65,66]. All these associations were more pronounced in LB patients compared to controls in our study, highlighting the contributions of CVD to aspects of neurodegeneration that seem to be beyond and above associations found in the normal population.

We found no associations between CHIPS or any of the other MRI methods and CSF biomarkers of AD, suggesting that AD co-pathology in LB may not significantly contribute to cholinergic white matter degeneration. Previous studies have demonstrated that in cognitively unimpaired older individuals, CVD is the most important contributor to the degeneration of the cholinergic pathways [36,41]. Therefore, the cholinergic dysfunction observed in our LB group appears not to be associated with AD co-pathology but may rather be driven by other pathological processes such as CVD, alpha-synuclein-related processes, or the interplay between them.

No statistically significant correlations were found between the four MRI methods and any of the core clinical features, nor the clinical or cognitive measures in the LB group. The literature on the association of WMSA with clinical presentation in DLB is inconclusive. While a systematic review did not find a clear association between WMSA and clinical outcomes in DLB [67], some primary studies reported a higher degree of WMSA volume to be associated with the absence of REM sleep behavior disorder in DLB [7,10,68]. The link with visual hallucinations remains particularly ambiguous, with studies reporting a higher degree of WMSA associated with both increased and decreased frequencies of visual hallucinations [7,69]. Notably, deep WMSA, but not periventricular WMSA, may be associated with autonomic dysfunction, a supportive feature in the diagnosis of DLB [53,70]. Additionally, three studies have found an association between WMSA and lower global cognition scores, with the notion that Park et al.’s findings were observed in a combined group of AD, DLB, and PD dementia patients [7,24,71]. In this context, higher CHIPS scores have been associated with greater cognitive impairment in PD dementia, AD, and even multiple sclerosis [72–74], suggesting that CHIPS might capture nuances in certain cognitive measures in other neurodegenerative diseases. Further studies are thus needed to elucidate the association between cholinergic and global WMSA and clinical presentation and cognition in LB patients.

Our study has some limitations. Firstly, the cross-sectional design prevents us from establishing causal relationships between WMSA and the other measures, which would be better addressed in a longitudinal study. Secondly, while the size of our LB group is comparable to those in previous DLB studies, there may be limitations in statistical power to detect significant differences. However, we believe that the current size was sufficient for our objective of comparing CHIPS with other MRI and clinical measures, as the power was consistent across all analyses. Thirdly, we utilized well-established MRI markers of CVD, which, though valuable, do not account for other possible complexities of CVD in DLB and lack complete vascular specificity. Finally, the diagnosis of DLB in this study was based solely on clinical criteria and definitive confirmation of DLB requires post-mortem neuropathological examination, which was not available in our cohort.

In our study, cholinergic WMSA were significantly more pronounced in the LB group compared to healthy controls above and beyond global WMSA. Mechanistically, we have demonstrated that CVD in cholinergic white matter may be implicated in frontal atrophy along the LB continuum. In contrast, medial temporal atrophy appears to be more related to global CVD and/or AD co-pathology, as suggested in recent studies [8,63,64]. Clinically, our findings underscore the potential of CHIPS for assessing cholinergic WMSA in clinical settings, thereby aiding in the radiological characterization of LB. Altogether, we conclude that CVD co-pathology could be one of the drivers of the well-documented cholinergic degeneration in DLB, highlighting opportunities for therapeutic interventions that target vascular health in the disease continuum, from early stages of MCI-LB extending to full-blown DLB.

AB42-40 = Amyloid beta 1-42/Amyloid beta 1-40

AD = Alzheimer’s disease

ANOVA = Analysis of variance

AUC = Area under the curve

AVRA = Automatic visual ratings of atrophy

CDR = Clinical dementia rate

CHIPS = Cholinergic Pathway Hyperintensities Scale

CVD = Cerebrovascular disease

DLB = Dementia with Lewy Bodies

DWI = Diffusion-weighted imaging

FCSRT = Free and cued selective reminding test

FLAIR = Fluid-attenuated inversion recovery

GCA-F = Global cortical atrophy – frontal

LB = Lewy body

MCI-LB = Mild cognitive impairment of Lewy body type

MD = Mean diffusivity

MMSE = Mini-mental state examination

MPRAGE = Magnetization-prepared rapid gradient echo

MTA = Medial temporal atrophy

PA = Posterior atrophy

PD = Parkinson’s disease

ROC = Receiver operating characteristic

SPIN = Sant Pau Initiative on Neurodegeneration

TIV = Total intracranial volume

WMSA = White matter signal abnormalities

Ethics approval

The study was approved by the local Ethics Committee in Hospital Sant Pau (Barcelona, Spain), following the ethical standards recommended by the Helsinki Declaration. All study individuals provided written informed consent.

Availability of data and material

Raw anonymized data are available upon reasonable request. All requests should be sent to the corresponding author detailing the study hypothesis and statistical analysis plan. The steering committee of this study will decide whether data sharing is appropriate based on the novelty and scientific rigor of the proposal. All applicants will be asked to sign a data access agreement.

Competing interests

DF consults for BioArctic and has received honoraria from Esteve. All other authors declare they have no competing interests.

Funding

This study was funded by the Swedish Research Council (Vetenskapsrådet, grant 2022-00916), the Center for Innovative Medicina (CIMED, grants 20200505 and FoUI-988826), the regional agreement on medical training and clinical research of Stockholm Region (ALF Medicine, grants FoUI-962240 and FoUI-987534), the Swedish Brain Foundation (Hjärnfonden FO2023-0261, FO2022-0175, FO2021-0131), the Swedish Alzheimer Foundation (Alzheimerfonden AF-968032, AF-980580, AF-994058), the Swedish Dementia Foundation (Demensfonden), the Gamla Tjänarinnor Foundation, the Gun och Bertil Stohnes Foundation, Funding for Research from Karolinska Institutet, Neurofonden, and Foundation for Geriatric Diseases at Karolinska Institutet, as well as contributions from private bequests. Research by MN was partially supported by institutional resources of Czech Technical University in Prague.

Authors’ contributions

CJ contributed to the study design, data analysis, and interpretation of data, and drafted and revised the manuscript. DF designed the study, supervised the project, and obtained funding, and revised the manuscript, and approved the final version. MN contributed to data preparation, performed the analysis and interpretation of data, and revised, and approved the final version of the manuscript. AT contributed to the analysis and interpretation of data, revised, and approved the final version of the manuscript. DA, JA, VM, AB, DA, JF, AL contributed to data acquisition and critically revised the manuscript, and approved the final version of the manuscript. EW provided resources for data processing and analysis, critically revised the manuscript, and approved the final version of the manuscript. MGK contributed to the supervision of the project, critically revised the manuscript, and approved the final version of the manuscript. AR critically revised the manuscript and approved the final version of the manuscript.

Acknowledgements

Not applicable.

Aarsland D, Rongve A, Piepenstock Nore S, Skogseth R, Skulstad S, Ehrt U, et al. Frequency and case identification of dementia with Lewy bodies using the revised consensus criteria. Dement Geriatr Cogn Disord [Internet]. 2008 [cited 2022 Oct 11];26:445–52. Available from: https://pubmed-ncbi-nlm-nih-gov.proxy.kib.ki.se/18974647/
Jellinger KA, Attems J. Prevalence and pathology of dementia with lewy bodies in the oldest old: A comparison with other dementing disorders. Dement Geriatr Cogn Disord [Internet]. 2011 [cited 2022 Oct 11];31:309–16. Available from: https://pubmed-ncbi-nlm-nih-gov.proxy.kib.ki.se/21502762/
Kane JPM, Surendranathan A, Bentley A, Barker SAH, Taylor JP, Thomas AJ, et al. Clinical prevalence of Lewy body dementia. Alzheimers Res Ther [Internet]. 2018 [cited 2022 Oct 11];10. Available from: https://pubmed-ncbi-nlm-nih-gov.proxy.kib.ki.se/29448953/
Vann Jones SA, O’Brien JT. The prevalence and incidence of dementia with Lewy bodies: A systematic review of population and clinical studies [Internet]. Psychol Med. Cambridge University Press; 2014 [cited 2022 Oct 11]. p. 673–83. Available from: https://pubmed-ncbi-nlm-nih-gov.proxy.kib.ki.se/23521899/
De Reuck J, Maurage CA, Deramecourt V, Pasquier F, Cordonnier C, Leys D, et al. Aging and cerebrovascular lesions in pure and in mixed neurodegenerative and vascular dementia brains: A neuropathological study. Folia Neuropathol [Internet]. 2018 [cited 2022 Oct 11];56:81–7. Available from: https://pubmed.ncbi.nlm.nih.gov/30509027/
Watson R, Colloby SJ. Imaging in Dementia with Lewy Bodies: An Overview [Internet]. J Geriatr Psychiatry Neurol. SAGE Publications Inc.; 2016 [cited 2022 Oct 11]. p. 254–60. Available from: https://pubmed.ncbi.nlm.nih.gov/27502300/
Ferreira D, Nedelska Z, Graff-Radford J, Przybelski SA, Lesnick TG, Schwarz CG, et al. Cerebrovascular disease, neurodegeneration, and clinical phenotype in dementia with Lewy bodies. Neurobiol Aging [Internet]. 2021;105:252–61. Available from: https://doi.org/10.1016/j.neurobiolaging.2021.04.029
Ferreira D, Przybelski SA, Lesnick TG, Schwarz CG, Diaz-Galvan P, Graff-Radford J, et al. Cross-sectional Associations of β-Amyloid, Tau, and Cerebrovascular Biomarkers With Neurodegeneration in Probable Dementia With Lewy Bodies. Neurology [Internet]. 2023 [cited 2023 Oct 1];100:E846–59. Available from: https://pubmed.ncbi.nlm.nih.gov/36443011/
Sarro L, Senjem ML, Lundt ES, Przybelski SA, Lesnick TG, Graff-Radford J, et al. Amyloid-β deposition and regional grey matter atrophy rates in dementia with Lewy bodies. Brain [Internet]. 2016 [cited 2023 Oct 1];139:2740–50. Available from: https://pubmed.ncbi.nlm.nih.gov/27452602/
Burton EJ, McKeith IG, Burn DJ, Firbank MJ, O’Brien JT. Progression of white matter hyperintensities in Alzheimer disease, dementia with lewy bodies, and Parkinson disease dementia: a comparison with normal aging. Am J Geriatr Psychiatry [Internet]. 2006 [cited 2023 Oct 1];14:842–9. Available from: https://pubmed.ncbi.nlm.nih.gov/17001024/
Prins ND, Van Dijk EJ, Den Heijer T, Vermeer SE, Koudstaal PJ, Oudkerk M, et al. Cerebral white matter lesions and the risk of dementia. Arch Neurol [Internet]. 2004 [cited 2023 Oct 1];61:1531–4. Available from: https://pubmed.ncbi.nlm.nih.gov/15477506/
Smith EE, Salat DH, Jeng J, McCreary CR, Fischl B, Schmahmann JD, et al. Correlations between MRI white matter lesion location and executive function and episodic memory. Neurology [Internet]. 2011 [cited 2022 Oct 11];76:1492–9. Available from: https://pubmed.ncbi.nlm.nih.gov/21518999/
Behl P, Bocti C, Swartz RH, Gao FQ, Sahlas DJ, Lanctot KL, et al. Strategic subcortical hyperintensities in cholinergic pathways and executive function decline in treated Alzheimer patients. Arch Neurol [Internet]. 2007 [cited 2022 Oct 11];64:266–72. Available from: https://pubmed.ncbi.nlm.nih.gov/17296844/
Kim SH, Kang HS, Kim HJ, Moon Y, Ryu HJ, Kim MY, et al. The effect of ischemic cholinergic damage on cognition in patients with subcortical vascular cognitive impairment. J Geriatr Psychiatry Neurol [Internet]. 2012 [cited 2022 Oct 11];25:122–7. Available from: https://pubmed.ncbi.nlm.nih.gov/22689705/
Bohnen NI, Grothe MJ, Ray NJ, Müller MLTM, Teipel SJ. Recent Advances in Cholinergic Imaging and Cognitive Decline—Revisiting the Cholinergic Hypothesis of Dementia [Internet]. Curr Geriatr Rep. Springer New York LLC; 2018 [cited 2023 Jul 2]. Available from: https://pubmed.ncbi.nlm.nih.gov/29503795/
Schumacher J, Ray NJ, Hamilton CA, Donaghy PC, Firbank M, Roberts G, et al. Cholinergic white matter pathways in dementia with Lewy bodies and Alzheimer’s disease. Brain [Internet]. 2022 [cited 2022 Oct 11];145:1773–84. Available from: https://doi.org/10.1093/brain/awab372
Schumacher J, Ray NJ, Hamilton CA, Bergamino M, Donaghy PC, Firbank M, et al. Free water imaging of the cholinergic system in dementia with Lewy bodies and Alzheimer’s disease. Alzheimer’s & Dementia [Internet]. 2023 [cited 2023 May 6]; Available from: https://pubmed.ncbi.nlm.nih.gov/36919460/
Barrett MJ, Cloud LJ, Shah H, Holloway KL. Therapeutic approaches to cholinergic deficiency in Lewy body diseases [Internet]. Expert Rev Neurother. Taylor and Francis Ltd; 2020 [cited 2022 Oct 12]. p. 41–53. Available from: https://pubmed-ncbi-nlm-nih-gov.proxy.kib.ki.se/31577469/
O’Brien JT, Burns A. Clinical practice with anti-dementia drugs: A revised (second) consensus statement from the British Association for Psychopharmacology [Internet]. Journal of Psychopharmacology. J Psychopharmacol; 2011 [cited 2022 Oct 11]. p. 997–1019. Available from: https://pubmed.ncbi.nlm.nih.gov/21088041/
Taylor JP, McKeith IG, Burn DJ, Boeve BF, Weintraub D, Bamford C, et al. New evidence on the management of Lewy body dementia [Internet]. Lancet Neurol. Lancet Publishing Group; 2020 [cited 2022 Oct 11]. p. 157–69. Available from: /pmc/articles/PMC7017451/
Bocti C, Swartz RH, Gao FQ, Sahlas DJ, Behl P, Black SE. A new visual rating scale to assess strategic white matter hyperintensities within cholinergic pathways in dementia. Stroke [Internet]. 2005 [cited 2022 Jun 23];36:2126–31. Available from: http://ahajournals.org
Hanning U, Teuber A, Lang E, Trenkwalder C, Mollenhauer B, Minnerup H. White matter hyperintensities are not associated with cognitive decline in early Parkinson’s disease – The DeNoPa cohort. Parkinsonism Relat Disord [Internet]. 2019;69:61–7. Available from: https://doi.org/10.1016/j.parkreldis.2019.10.016
Kim HJ, Moon WJ, Han SH. Differential cholinergic pathway involvement in alzheimer’s disease and subcortical ischemic vascular dementia. Journal of Alzheimer’s Disease [Internet]. 2013 [cited 2023 Jun 25];35:129–36. Available from: https://pubmed.ncbi.nlm.nih.gov/23364137/
Park HE, Park IS, Oh YS, Yang DW, Lee KS, Choi HS, et al. Subcortical whiter matter hyperintensities within the cholinergic pathways of patients with dementia and parkinsonism. J Neurol Sci [Internet]. 2015;353:44–8. Available from: http://dx.doi.org/10.1016/j.jns.2015.03.046
Shin J, Choi S, Lee JE, Lee HS, Sohn YH, Lee PH. Subcortical white matter hyperintensities within the cholinergic pathways of Parkinson’s disease patients according to cognitive status. J Neurol Neurosurg Psychiatry. 2012;83:315–21.
McKeith IG, Ferman TJ, Thomas AJ, Blanc F, Boeve BF, Fujishiro H, et al. Research criteria for the diagnosis of prodromal dementia with Lewy bodies. Neurology [Internet]. 2020 [cited 2023 Oct 1];94:743–55. Available from: https://pubmed.ncbi.nlm.nih.gov/32241955/
Alcolea D, Clarimón J, Carmona-Iragui M, Illán-Gala I, Morenas-Rodríguez E, Barroeta I, et al. The Sant Pau Initiative on Neurodegeneration (SPIN) cohort: A data set for biomarker discovery and validation in neurodegenerative disorders. Alzheimer’s and Dementia: Translational Research and Clinical Interventions [Internet]. 2019 [cited 2022 Oct 19];5:597–609. Available from: /pmc/articles/PMC6804606/
Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. American Journal of Roentgenology [Internet]. 1987 [cited 2022 Oct 19];149:351–6. Available from: https://pubmed.ncbi.nlm.nih.gov/3496763/
Joki H, Higashiyama Y, Nakae Y, Kugimoto C, Doi H, Kimura K, et al. White matter hyperintensities on MRI in dementia with Lewy bodies, Parkinson’s disease with dementia, and Alzheimer’s disease. J Neurol Sci. 2018;385:99–104.
Landis JR, Koch GG. The Measurement of Observer Agreement for Categorical Data. Biometrics. 1977;33:159.
Fischl B. FreeSurfer. Neuroimage [Internet]. 2012 [cited 2023 Oct 1];62:774. Available from: /pmc/articles/PMC3685476/
Cedres N, Ferreira D, Machado A, Shams S, Sacuiu S, Waern M, et al. Predicting Fazekas scores from automatic segmentations of white matter signal abnormalities. Aging [Internet]. 2020 [cited 2023 Jul 1];12:894–901. Available from: https://pubmed.ncbi.nlm.nih.gov/31927535/
Leritz EC, Shepel J, Williams VJ, Lipsitz LA, Mcglinchey RE, Milberg WP, et al. Associations between T1 white matter lesion volume and regional white matter microstructure in aging. Hum Brain Mapp [Internet]. 2014 [cited 2023 Jun 25];35:1085–100. Available from: /pmc/articles/PMC4356252/
Salat DH, Tuch DS, van der Kouwe AJW, Greve DN, Pappu V, Lee SY, et al. White matter pathology isolates the hippocampal formation in Alzheimer’s disease. Neurobiol Aging. 2010;31:244–56.
Muehlboeck JS, Westman E, Simmons A. TheHiveDB image data management and analysis framework. Front Neuroinform [Internet]. 2013 [cited 2024 Jan 31];7. Available from: /pmc/articles/PMC3880907/
Nemy M, Cedres N, Grothe MJ, Muehlboeck JS, Lindberg O, Nedelska Z, et al. Cholinergic white matter pathways make a stronger contribution to attention and memory in normal aging than cerebrovascular health and nucleus basalis of Meynert. Neuroimage [Internet]. 2020 [cited 2023 Jun 25];211. Available from: https://pubmed.ncbi.nlm.nih.gov/32035186/
Jenkinson M, Beckmann CF, Behrens TEJ, Woolrich MW, Smith SM. FSL. Neuroimage [Internet]. 2012 [cited 2023 Jun 25];62:782–90. Available from: https://linkinghub.elsevier.com/retrieve/pii/S1053811911010603
Reber PJ, Wong EC, Buxton RB, Frank LR. Correction of off resonance-related distortion in echo-planar imaging using EPI-based field maps. Magn Reson Med [Internet]. 1998 [cited 2023 Jun 25];39:328–30. Available from: https://pubmed.ncbi.nlm.nih.gov/9469719/
Smith SM. Fast robust automated brain extraction. Hum Brain Mapp [Internet]. 2002 [cited 2023 Jun 25];17:143–55. Available from: https://pubmed.ncbi.nlm.nih.gov/12391568/
Behrens TEJ, Woolrich MW, Jenkinson M, Johansen-Berg H, Nunes RG, Clare S, et al. Characterization and Propagation of Uncertainty in Diffusion-Weighted MR Imaging. Magn Reson Med [Internet]. 2003 [cited 2023 Jun 25];50:1077–88. Available from: https://pubmed.ncbi.nlm.nih.gov/14587019/
Cedres N, Ferreira D, Nemy M, Machado A, Pereira JB, Shams S, et al. Association of Cerebrovascular and Alzheimer Disease Biomarkers With Cholinergic White Matter Degeneration in Cognitively Unimpaired Individuals. Neurology [Internet]. 2022 [cited 2023 Oct 1];99:e1619–29. Available from: https://n.neurology.org/content/99/15/e1619
Nemy M, Dyrba M, Brosseron F, Buerger K, Dechent P, Dobisch L, et al. Cholinergic white matter pathways along the Alzheimer’s disease continuum. Brain [Internet]. 2023 [cited 2023 Oct 1];146:2075–88. Available from: https://dx.doi.org/10.1093/brain/awac385
Mårtensson G, Ferreira D, Cavallin L, Muehlboeck JS, Wahlund LO, Wang C, et al. AVRA: Automatic visual ratings of atrophy from MRI images using recurrent convolutional neural networks. Neuroimage Clin [Internet]. 2019 [cited 2022 Feb 11];23. Available from: /pmc/articles/PMC6545397/
Ferreira D, Cavallin L, Larsson EM, Muehlboeck JS, Mecocci P, Vellas B, et al. Practical cut-offs for visual rating scales of medial temporal, frontal and posterior atrophy in Alzheimer’s disease and mild cognitive impairment. J Intern Med [Internet]. 2015 [cited 2023 Nov 11];278:277–90. Available from: https://pubmed.ncbi.nlm.nih.gov/25752192/
Alcolea D, Pegueroles J, Muñoz L, Camacho V, López-Mora D, Fernández-León A, et al. Agreement of amyloid PET and CSF biomarkers for Alzheimer’s disease on Lumipulse. Ann Clin Transl Neurol [Internet]. 2019 [cited 2023 Oct 20];6:1815. Available from: /pmc/articles/PMC6764494/
Barber R, Ballard C, McKeith IG, Gholkar A, O’Brien JT. MRI volumetric study of dementia with Lewy bodies: a comparison with AD and vascular dementia. Neurology [Internet]. 2000 [cited 2023 Oct 1];54:1304–9. Available from: https://pubmed.ncbi.nlm.nih.gov/10746602/
Tiraboschi P, Hansen LA, Alford M, Sabbagh MN, Schoos B, Masliah E, et al. Cholinergic dysfunction in diseases with LEWY bodies. Neurology [Internet]. 2000 [cited 2023 Jul 2];54:407–11. Available from: https://pubmed.ncbi.nlm.nih.gov/10668703/
Schumacher J, Gunter JL, Przybelski SA, Jones DT, Graff-Radford J, Savica R, et al. Dementia with Lewy bodies: Association of Alzheimer pathology with functional connectivity networks. Brain. 2021;144:3212–25.
Charlton RA, Schiavone F, Barrick TR, Morris RG, Markus HS. Diffusion tensor imaging detects age related white matter change over a 2 year follow-up which is associated with working memory decline. J Neurol Neurosurg Psychiatry [Internet]. 2010 [cited 2023 Dec 16];81:13–9. Available from: https://pubmed.ncbi.nlm.nih.gov/19710050/
Sullivan E V., Pfefferbaum A. Diffusion tensor imaging and aging. Neurosci Biobehav Rev [Internet]. 2006 [cited 2023 Dec 16];30:749–61. Available from: https://pubmed.ncbi.nlm.nih.gov/16887187/
Teipel SJ, Meindl T, Wagner M, Stieltjes B, Reuter S, Hauenstein KH, et al. Longitudinal changes in fiber tract integrity in healthy aging and mild cognitive impairment: a DTI follow-up study. J Alzheimers Dis [Internet]. 2010 [cited 2023 Dec 16];22:507–22. Available from: https://pubmed.ncbi.nlm.nih.gov/20847446/
Iizuka T, Iizuka R, Kameyama M. Cingulate island sign temporally changes in dementia with Lewy bodies. Sci Rep. 2017;7.
Mckeith IG, Sci M, Boeve BF, Dickson DW, Halliday G, Taylor J-P, et al. Diagnosis and management of dementia with Lewy bodies Fourth consensus report of the DLB Consortium. 2017.
Minoshima S, Foster NL, Sima AAF, Frey KA, Albin RL, Kuhl DE. Alzheimer’s disease versus dementia with Lewy bodies: cerebral metabolic distinction with autopsy confirmation. Ann Neurol [Internet]. 2001 [cited 2023 Oct 8];50:358–65. Available from: https://pubmed-ncbi-nlm-nih-gov.proxy.kib.ki.se/11558792/
Huang SH, Chang CC, Lui CC, Chen NC, Lee CC, Wang PW, et al. Cortical metabolic and nigrostriatal abnormalities associated with clinical stage-specific dementia with Lewy bodies. Clin Nucl Med [Internet]. 2015 [cited 2023 Oct 8];40:26–31. Available from: https://pubmed-ncbi-nlm-nih-gov.proxy.kib.ki.se/25426755/
Patterson L, Firbank MJ, Colloby SJ, Attems J, Thomas AJ, Morris CM. Neuropathological Changes in Dementia With Lewy Bodies and the Cingulate Island Sign. J Neuropathol Exp Neurol [Internet]. 2019 [cited 2023 Oct 8];78:717. Available from: /pmc/articles/PMC6640897/
Debette S, Markus HS. The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: Systematic review and meta-analysis [Internet]. BMJ (Online). BMJ Publishing Group; 2010 [cited 2023 Jun 29]. p. 288. Available from: /pmc/articles/PMC2910261/
Prins ND, Scheltens P. White matter hyperintensities, cognitive impairment and dementia: An update [Internet]. Nat Rev Neurol. Nature Publishing Group; 2015 [cited 2023 Jun 29]. p. 157–65. Available from: https://pubmed.ncbi.nlm.nih.gov/25686760/
Davies P, Resnick J, Resnick B, Gilman S, Growdon JH, Khachaturian ZS, et al. Consensus Report of the Working Group on: “Molecular and Biochemical Markers of Alzheimer’s Disease.” Neurobiol Aging. 1998;19:109–16.
Riphagen JM, Gronenschild EH, Salat DH, Freeze WM, Ivanov D, Clerx L, et al. Shades of white: diffusion properties of T1- and FLAIR-defined white matter signal abnormalities differ in stages from cognitively normal to dementia. Neurobiol Aging. 2018;68:48–58.
Olsson E, Klasson N, Berge J, Eckerström C, Edman Å, Malmgren H, et al. White matter lesion assessment in patients with cognitive impairment and healthy controls: Reliability comparisons between visual rating, a manual, and an automatic volumetrical MRI method - The gothenburg MCI study. J Aging Res [Internet]. 2013 [cited 2023 Jul 5];2013. Available from: https://pubmed.ncbi.nlm.nih.gov/23401776/
Valdés Hernández MDC, Morris Z, Dickie DA, Royle NA, Muñoz Maniega S, Aribisala BS, et al. Close correlation between quantitative and qualitative assessments of white matter lesions. Neuroepidemiology [Internet]. 2012 [cited 2023 Jul 5];40:13–22. Available from: https://pubmed.ncbi.nlm.nih.gov/23075702/
Rennie A, Ekman U, Shams S, Ryden L, Samuelsson J, Kern S, et al. Cerebrovascular and Alzheimer’s disease biomarkers in dementia with Lewy bodies and other dementias. Brain Communications - in review. 2024;
Inguanzo A, Poulakis K, Mohanty R, Schwarz CG, Przybelski SA, Diaz-Galvan P, et al. MRI data-driven clustering reveals different subtypes of Dementia with Lewy bodies. NPJ Parkinsons Dis [Internet]. 2023 [cited 2023 Nov 4];9. Available from: https://pubmed.ncbi.nlm.nih.gov/36670121/
van der Zande JJ, Gouw AA, Steenoven I van, Scheltens P, Stam CJ, Lemstra AW. EEG Characteristics of Dementia With Lewy Bodies, Alzheimer’s Disease and Mixed Pathology. Front Aging Neurosci [Internet]. 2018 [cited 2023 Dec 4];10. Available from: https://pubmed.ncbi.nlm.nih.gov/30018548/
Kantarci K, Lesnick T, Ferman TJ, Przybelski SA, Boeve BF, Smith GE, et al. Hippocampal volumes predict risk of dementia with Lewy bodies in mild cognitive impairment. Neurology [Internet]. 2016 [cited 2023 Dec 4];87:2317–23. Available from: https://n.neurology.org/content/87/22/2317
Hijazi Z, Yassi N, O’Brien JT, Watson R. The influence of cerebrovascular disease in dementia with Lewy bodies and Parkinson’s disease dementia. Eur J Neurol. 2021;1–12.
Sarro L, Tosakulwong N, Schwarz CG, Graff-Radford J, Przybelski SA, Lesnick TG, et al. An investigation of cerebrovascular lesions in dementia with Lewy bodies compared to Alzheimer’s disease. Alzheimers Dement [Internet]. 2017 [cited 2023 Oct 1];13:257. Available from: /pmc/articles/PMC5303194/
Barber R, Gholkar A, Scheltens P, Ballard C, McKeith IG, O’Brien JT. Medial temporal lobe atrophy on MRI in dementia with Lewy bodies. Neurology [Internet]. 1999 [cited 2023 Oct 1];52:1153–8. Available from: https://pubmed.ncbi.nlm.nih.gov/10214736/
Ballard C, O’Brien J, Barber B, Scheltens P, Shaw F, Mckeith I, et al. Neurocardiovascular instability, hypotensive episodes, and MRI lesions in neurodegenerative dementia. Ann N Y Acad Sci [Internet]. New York Academy of Sciences; 2000 [cited 2023 Jul 6]. p. 442–5. Available from: https://pubmed.ncbi.nlm.nih.gov/10818535/
Chen TY, Chan PC, Tsai CF, Wei CY, Chiu PY. White matter hyperintensities in dementia with Lewy bodies are associated with poorer cognitive function and higher dementia stages. Front Aging Neurosci. 2022;14:935652.
Shin J, Choi S, Lee JE, Lee HS, Sohn YH, Lee PH. Subcortical white matter hyperintensities within the cholinergic pathways of Parkinson’s disease patients according to cognitive status. J Neurol Neurosurg Psychiatry [Internet]. 2012 [cited 2023 Oct 8];83:315–21. Available from: https://jnnp.bmj.com/content/83/3/315
Yu MC, Chuang YF, Wu SC, Ho CF, Liu YC, Chou CJ. White matter hyperintensities in cholinergic pathways are associated with dementia severity in e4 carriers but not in non-carriers. Front Neurol [Internet]. 2023 [cited 2023 Oct 8];14. Available from: https://pubmed-ncbi-nlm-nih-gov.proxy.kib.ki.se/36864910/
Kimura Y, Sato N, Ota M, Maikusa N, Maekawa T, Sone D, et al. A structural MRI study of cholinergic pathways and cognition in multiple sclerosis. eNeurologicalSci [Internet]. 2017;8:11–6. Available from: https://doi.org/10.1016/j.ensci.2017.06.008

Competing interest reported. DF consults for BioArctic and has received honoraria from Esteve. All other authors declare they have no competing interests.

Additionalfile1.pdf
Additional file 1.pdf MRI scanning parameters MRI = Magnetic Resonance Imaging; MPRAGE = Magnetization-Prepared Rapid Acquisition Gradient Echo; FLAIR = Fluid-attenuated inversion recovery.
Additionalfile2.pdf
Additional file 2.pdf Associations between demographic data, WMSA measures, and atrophy measures expressed in p-values with effect sizes expressed as Cohen’s d. WMSA = White matter signal abnormalities, CHIPS = Cholinergic pathways hyperintensities scale, MD = Mean Diffusivity; MTA = Medial temporal atrophy; PA = Parietal atrophy; GCA-F = Global Cortical Atrophy – Frontal; p = p-value, d = Cohen’s d.
Additionalfile3.pdf
Additional file 3.pdf Results from ROC analysis with confidence intervals, cutoff values, sensitivity, and specificity values, for WMSA and tractography measures (ordered by AUC values). Excap = External capsule; CHIPS = Cholinergic pathways hyperintensities scale; WMSA = White matter signal abnormalities; AUC = Area under curve; *Equals to or higher means positive outcome.

Cerebrovascular co-pathology and cholinergic white matter pathways along the Lewy body continuum

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1