White Matter Lesions May Aid in Differentiating Idiopathic Normal Pressure Hydrocephalus and Alzheimer’s Disease

Abstract

Background:

Idiopathic normal pressure hydrocephalus (iNPH) is often misdiagnosed as Alzheimer’s disease (AD) due to overlapping pathophysiology and similar imaging characteristics, including ventricular enlargement and increased white matter lesions (WMLs).

Objective:

To compare the extent and distribution of WMLs directly between iNPH and AD and examine the association with underlying pathophysiology.

Methods:

Twelve patients with iNPH (mean age: 78.08 years; 5 females), 20 with AD (mean age: 75.40 years; 13 females), and 10 normal cognition (NC) participants (mean age: 76.60 years; 7 females) were recruited. The extent and distribution of WMLs and the lateral ventricular volume (LV-V) were evaluated on MRI using voxel-based morphometry analysis. Concentrations of cerebrospinal fluid biomarkers, such as amyloid-β protein (Aβ)₄₂, Aβ₄₀, Aβ₃₈, and tau species, were also measured. Risk factors for small vessel disease (SVD) were assessed by blood examination and medical records.

Results:

The periventricular WML volume (PWML-V) and deep WML volume (DWML-V) were significantly larger in iNPH than in AD and NC. The DWML-V was dominant in iNPH, while the PWML-V was dominant in AD and NC. GM-V was significantly smaller in AD than in iNPH and NC. The LV-V positively correlated with WML-V in all participants. There was a significant negative correlation between LV-V and Aβ₃₈ in iNPH. Furthermore, there was no significant difference in SVD risk factors between the groups.

Conclusion:

The differences in the extent and distribution of WMLs between iNPH and AD, especially predominance of DWML-V over PWML-V in iNPH, may reflect decreased fluid and Aβ clearance.

Keywords

Alzheimer disease amyloid cerebrospinal fluid glymphatic system hydrocephalus white matter

INTRODUCTION

Idiopathic normal pressure hydrocephalus (iNPH), first described in 1965, is common in the elderly, with nearly 3%of individuals aged > 65 years affected in Japan [1]. It is a syndrome characterized by impaired gait, cognition, and urinary control; however, only 60%of patients present with this triad [2]. Despite its pervasive prevalence, a vast majority of iNPH patients are misdiagnosed with other neurodegenerative disorders, reflecting its clinical overlap with other neurodegenerative diseases and the widespread unfamiliarity or skepticism of the medical community towards iNPH [3]. Alzheimer’s disease (AD), rather common in people aged ≥65 years, has an estimated prevalence of 13%and is the dementia most commonly misdiagnosed as iNPH due to similar imaging characteristics [4]. Most typical methods to diagnose iNPH include calculating the Evans index (EI) and detecting disproportionately enlarged subarachnoid space hydrocephalus on magnetic resonance imaging (MRI) [5 –7]. However, cerebral atrophy in AD can also cause ventricular enlargement and increase EI [8 –10]. Recently, it has been speculated that a glymphatic system exists in the human brain and is thought to be impaired in both iNPH and AD [11]. The glymphatic system is a sleep-assisted highly polarized cerebrospinal (CSF) and interstitial fluid (ISF) transport system that facilitates extracellular waste removal through a network of astroglia-supported perivascular or perineural channels that drain into the cervical and basal meningeal lymphatic networks or the major dural sinuses [12, 13]. Attenuation of glymphatic function is thought to be associated with decreased clearance and accumulation of neurotoxic proteins, such as amyloid-β protein (Aβ); thus, the coexistence of AD pathology in iNPH patients has been documented and is shown to be of increased prevalence [14]. Confirming CSF biomarker concentrations, such as decreased Aβ₄₂ and elevated tau species, is useful for AD diagnosis [15]. However, decreased clearance lowers CSF Aβ₄₂ levels also in iNPH, which complicates differentiation [16]. From the neuropsychological perspective, iNPH is associated more with the executive frontal lobe and posterior cortical function than with memory impairment, as seen in AD [17]. However, neuropsychological testing is not a definitive diagnostic technique for iNPH. Functional and molecular nuclear medicine imaging with single-photon emission computed tomography and positron emission tomography (PET) techniques provides valuable information for diagnosing AD, but these techniques are not as easily available as MRI in a clinical setting [18].

Cerebral white matter lesions (WMLs) are also common MRI findings in patients with iNPH and AD compared to people with normal cognition [19]. In the general elderly population, WMLs are often found on T2-weighted or fluid-attenuated inversion recovery (FLAIR) images as white matter (WM) hyperintensities, with a prevalence of 10–20%in the seventh decade of life and nearly 100%in the ninth [19]. These lesions are located in the deep hemispheric WM, typically sparing the subcortical U-fibers. The pathological substrates of WMLs include tissue rarefaction associated with the loss of myelin and axons, enlarged perivascular space and mild gliosis, and small vessel disease (SVD), or simply reflect an increase in free tissue water [19, 20]. Currently, WMLs are divided into periventricular white matter lesions (PWMLs) and deep white matter lesions (DWMLs) [21]. PWMLs and DWMLs have different pathogenic mechanisms. According to postmortem studies, PWMLs show discontinuous ependyma (and therefore high extracellular fluid content), gliosis, and myelin loss, while DWMLs show more axonal loss, vacuolation, and increased tissue loss in more severe lesions, suggesting infarction in addition to demyelination and gliosis [21, 22]. In AD patients, WMLs are found in up to almost 90%of cases and are more severe than WMLs seen in the non-demented older adults [23]. However, previous studies suggest a multifactorial etiology of WMLs in AD, including SVD and degenerative axonal loss secondary to the deposition of cortical AD pathologies or as a result of retrograde Wallerian degeneration [24 –27]. In contrast, there is limited information on WMLs in iNPH. Notably, iNPH patients have more frequent and more severe PWMLs and DWMLs than people in age-matched control groups [28]. Although PWMLs in iNPH are sometimes reversible after shunting surgery, they are often detected after fluid (CSF/ISF) stagnation [29, 30]. Recent studies using diffusion tensor imaging (DTI) and gadobutrol injection also support impaired diffusive ability in periventricular WM in iNPH [31 –33]. While PWMLs in iNPH have received attention, there has been little focus on DWMLs.

Although WMLs are a common pathology in iNPH and AD, very few studies have directly compared the extent and distribution of WMLs between iNPH and AD. Herein, we compared the volume and distribution of WMLs between iNPH and AD, differentiating between PWMLs and DWMLs using voxel-based morphometry (VBM) analysis. Our recent study suggested that increased SVD and WMLs presence was related to ventricular enlargement in AD [34]. We additionally examined whether there was an association between WMLs, lateral ventricular volume, CSF biomarkers, and SVD risk factors to determine the underlying pathophysiological differences of WMLs between iNPH and AD and investigate whether differences in the degree and distribution of WMLs could be a clue in differentiating iNPH and AD.

MATERIALS AND METHODS

Patients

We included patients who met the diagnostic criteria of iNPH and AD and normal cognition (NC) participants who consulted at the memory clinic of the Division of Neurology in the Department of Medicine at the Showa University School of Medicine for Memory Loss. Diagnosis of iNPH was based on the second edition of the Japanese Guidelines for Management of Idiopathic Normal Pressure Hydrocephalus, developed by a multidisciplinary expert working group of the Japanese Society of Normal Pressure Hydrocephalus in conjunction with the Japanese Ministry of Health, Labor and Welfare research project on “Studies on the epidemiology, pathophysiology, and treatment of normal pressure hydrocephalus [35].” iNPH was diagnosed as “probable iNPH,” and all patients had disproportionately enlarged subarachnoid space hydrocephalus on MRI and showed improvement in gait disturbance after the lumbar drainage test. The AD diagnosis was based on the diagnostic guidelines of the National Institute on Aging-Alzheimer’s Association (NIA-AA) workgroups for AD [36]. All AD patients were diagnosed as “probable AD dementia with an evidence of AD pathophysiological process” with elevated CSF tau species and decreased Aβ₄₂ concentration. Patients with a history of neurological and neuropsychiatric diseases, including cerebrovascular diseases were excluded. All participants underwent dementia screening, including medical history, neurological and physical examination, cognitive assessment using the Japanese version of the Mini-Mental State Examination (MMSE-J) [37], blood examination, and brain MRI. An MMSE-J score ≥28 was defined as NC. Only iNPH and AD patients underwent lumbar puncture for CSF examination. This study was approved by the Ethics Committee of Showa University School of Medicine. All participants provided written informed consent in accordance with the Declaration of Helsinki.

CSF analysis

CSF was collected by lumbar puncture in iNPH and AD patients. Aβ₄₂, Aβ₄₀, and Aβ₃₈ CSF concentrations were analyzed using V-PLEX Aβ Peptide Panel 1 (6E10) (Meso Scale Discovery, Rockville, MD) on a MESO QuickPlex SQ 120 system (Meso Scale Diagnostics, LLC, Rockville, MD, USA). CSF concentrations of total tau (t-tau) and tau phosphorylated at threonine 181 (p-tau) were measured using commercially available enzyme-linked immunosorbent assay kits: INNOTEST hTAU Ag and PHOSPHO-TAU (181P) (Fujirebio Europe, Inc., Ghent, Belgium) according to the manufacturer’s instructions. The intra- and inter-assay coefficients of variation were < 20%for all assays.

MRI data acquisition

Structural MRI scans were conducted using a 1.5T MR scanner (Magnetom Essenza, Siemens, Germany). The high-resolution T1-weighted 3D images of the whole brain (144 sagittal slices; 1.0×1.0×1.25 mm³; repetition time = 1,600 ms; echo time =4.7 ms; flip angle = 15°; field of view = 256×256) and transverse FLAIR sequence with 6-mm slice thickness were acquired for each patient.

Visual measurements

The EI (maximum width of the frontal horns of the lateral ventricles/maximal internal diameter of the cranium at the same level) was calculated on FLAIR axial images. The width was measured on three consecutive slices in each section, and the slice with the largest diameter at the maximal width of the frontal horns was selected. In the same slice, the largest internal diameter of the cranium was measured [5, 8].

To assess the extent of WMLs on FLAIR axial images, we used the Fazekas scale (FS). FS is a frequently used visual scale to quantify the amount of WM hyperintensity [38, 39]. The scale divides the WML into PWML and DWML, and each region is given a grade from 0 to 3 depending on lesion size and confluence, as follows: PWML: 0 = absent, 1 = caps or pencil-thin lining, 2 = smooth halo, 3 = irregular periventricular signal extending into the deep WM; DWML: 0 = absent, 1 = punctate foci, 2 = beginning confluence, 3 = large confluent areas. We evaluated the PWML (FS-PWML) and DWML (FS-DWML) scores and a total score (FS-total), ranging from 0–6, by summing the PWML and DWML scores. EI and FS were measured by two neurologists, respectively, under blinded conditions.

VBM analysis

We used BAAD (Brain Anatomical Analysis using Diffeomorphic deformation) software to evaluate the lateral ventricular volume and WML extent and distribution. BAAD is a VBM support software developed based on statistical parametric mapping 12 [40]. The brain is extracted from 3D MR images by skull striping, segmented into gray matter (GM), WM, and CSF and warped into Montreal Neurological Institute space using the Diffeomorphic Anatomical Registration Through Exponentiated Lie Algebra algorithm. A voxel volume is converted into a voxel signal by a “modulation” procedure. Finally, data are smoothed to obtain smoothed, modulated, warped, and segmented images. The latest version 4.2 interpolates FLAIR images with 3D T1-weighted images to automatically correct WML contamination in GM and calculates the volume of WMLs and lateral ventricles. Moreover, it is capable of extracting and measuring the volumes of WMLs in the deep WM area, periventricular area, and corpus callosum area. Details of the VBM processing flow in the BAAD software are shown elsewhere [40]. Total intracranial volume (TI-V); gray matter volume (GM-V); white matter volume (WM-V); lateral ventricular volume (LV-V); and volume of the total WML (TWML-V), periventricular WML area (PWML-V), and deep WML area (DWML-V) were assessed.

Risk factors for SVD

Laboratory testing was performed to evaluate the lipid metabolism (low-density lipoprotein [LDL], high-density lipoprotein [HDL], LDL/HDL, and tri-glycerides [TG]), glucose metabolism (hemoglo-bin A1c [HbA1C]), and the presence of heart failure (brain natriuretic peptide [BNP]). Presence of hypertension and current smoking status were detected based on self-reported medical history and/or medication use.

Statistics

An unpaired t test was used for CSF analysis between iNPH and AD. One-way analysis of variance (ANOVA) was used to analyze the visual measurements on MRI data and hypertension among the three groups. One-way analysis of covariance (ANCOVA) with TI-V as covariate was used to examine group differences in VBM on MRI data. Post-hoc t-tests with Bonferroni correction were performed for multiple comparisons in both ANOVA and ANCOVA. The relationship between regions was examined by a Pearson’s coefficient analysis with familywise error (FEW) correction. A chi-squared test was used to examine the male-female ratio of samples and smoking status. All tests were two-tailed. The results are shown as the mean and standard deviation (SD). SPSS version 26 was used for all statistical analyses. Statistical significance was defined as an adjusted p < 0.05.

RESULTS

Patient characteristics

We included 12 iNPH patients (mean age: 78.08 years; SD: 8.43; range: 61–89; 5 females), 20 AD patients (mean age: 75.40 years; SD: 9.99; range: 50–90; 13 females), and 10 normal cognition participants (NCs) (mean age: 76.60 years; SD: 6.47; range: 72–86; 7 females) as controls. Mean age and sex ratio were not significantly different among the groups. ANOVA showed no main effect of group for age (F_2,42 = 0.324, p = 0.725), and chi-squared test showed no difference in the sex ratio among the groups (χ² = 2.293, p = 0.318).

CSF analysis

The unpaired t test showed that the CSF Aβ₄₂ (t₃₀ =2.878, p = 0.007), Aβ_42/38 (t₃₀ = 6.500, p < 0.0001), and Aβ_42/40 ratio (t₃₀ = 7.245, p < 0.0001) were significantly decreased in AD compared to iNPH, while Aβ₄₀ (t₃₀ = 1.472, p = 0.151) and Aβ₃₈ (t₃₀ = 1.762, p = 0.088) tended to be low in iNPH compared to AD. P-tau (t₃₀ = 3.828, p = 0.001) and t-tau (t₃₀ = 3.661, p = 0.001) were significantly elevated in AD compared to iNPH (Table 1).

Table 1

Comparison of CSF and MRI data and SVD risk factors between groups

	iNPH		AD		NC
n (female)	12 (5)		20 (13)		10 (7)
	mean	SD	mean	SD	mean	SD
Age	78.08	8.43	75.40	9.99	76.60	6.47
MMSE-J	22.42	4.79	18.45^*	4.41	28.40^*#	0.80
CSF
Aβ₃₈ (pg/mL)	1,613.25	623.09	2,114.14	821.99
Aβ₄₀ (pg/mL)	3,948.64	1,346.10	4,653.03	1,220.35
Aβ₄₂ (pg/mL)	297.96	127.13	199.45 ^*	58.99
Aβ_42/38	0.19	0.03	0.10 ^*	0.03
Aβ_42/40	0.07	0.01	0.04 ^*	0.01
p-tau (pg/mL)	34.83	13.22	70.18 ^*	29.23
t-tau (pg/mL)	174.53	65.72	428.35 ^*	226.88
MRI
Visual evaluation
Evans index	0.35	0.04	0.31 ^*	0.03	0.28 ^*	0.02
FS-total	4.17	1.52	2.85	1.71	1.40 ^*	1.43
FS-PWML	2.25	0.72	1.40 ^*	0.86	0.80 ^*	0.75
FS-DWML	1.92	0.95	1.45	0.92	0.60 ^*	0.80
VBM analysis
GM-V (mL)	552.26	47.15	484.62 ^*	32.70	525.31^#	46.91
WM-V (mL)	411.87	45.02	434.86	41.45	462.94 ^*	43.71
LV-V (mL)	117.62	26.30	63.01 ^*	24.28	37.48^*#	11.76
TWML-V (mL)	41.39	13.16	22.24 ^*	19.51	7.76 ^*	7.20
PWML-V (mL)	18.29	2.89	11.24 ^*	7.02	5.07^*#	4.34
DWML-V (mL)	25.32	12.64	11.03 ^*	13.40	2.68 ^*	3.10
SVD risk factors
Blood examination
HDL (mg/dL)	57.82	17.17	69.50	19.98	59.60	10.56
LDL (mg/dL)	121.45	28.06	118.50	32.21	116.90	27.69
LDL/HDL	2.20	0.59	1.79	0.48	1.99	0.45
TG (mg/dL)	133.50	54.07	105.10	49.66	134.70	49.35
HbA1c (%)	5.93	0.75	6.33	1.08	6.25	1.07
BNP (pg/mL)	23.11	22.74	42.88	42.94	37.39	31.82
Hypertension (%)	9 (75)		9 (45)		5 (50)
Current smoker (%)	0 (0)		0 (0)		2 (20)^*#

iNPH, idiopathic normal pressure hydrocephalus; AD, Alzheimer’s disease; NC, normal cognition; SD, standard deviation; CSF, cerebral spinal fluid; Aβ, amyloid β-protein; p-tau, tau phosphorylated at threonine 181; t-tau, total tau; FS, Fazekas scale; VBM, voxel-based morphometry; TI-V, Total intracranial volume; GM-V, gray matter volume; WM-V, white matter volume; LV-V, lateral ventricular volume; TWML-V, volume of total WML; PWML-V, periventricular white matter lesion volume; DWML-V, deep white matter lesion volume; CWML-V, callosal white matter lesion volume; SVD, small vessel disease; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TG, triglycerides; HbA1c, hemoglobin A1c; BNP, brain natriuretic peptide; MMSE-J, Japanese version of Mini-Mental State Examination. ^*p < 0.05 (versus iNPH), ^#p < 0.05 (versus AD).

MRI data

Visual measurements

ANOVA showed a main effect of group for EI (F_2,42 = 12.655, p < 0.0001), while the post-hoc test revealed a significantly larger EI in the iNPH group than those in the AD (p = 0.005) and NC (p < 0.0001) groups, while there was no difference between the AD and NC groups (p = 0.090). The main effect of group was significant for FS-total (F_2,42 = 7.643, p = 0.002), with the score being significantly higher in the iNPH group than in the NC group (p = 0.001), while there was no difference between the AD and NC (p = 0.088) or AD and iNPH (p = 0.106) groups. For FS-PWML, the main effect of group was significant (F_2,42 = 8.671, p = 0.001), with significantly higher results in the iNPH group than in the AD (p = 0.023) and NC (p = 0.001) groups and no difference between the AD and NC groups (p = 0.001). For FS-DWML, the main effect of group was significant (F_2,42 = 5.485, p = 0.008), with significantly higher results in the iNPH group than in the NC group (p = 0.007) and no difference between the AD and NC (p = 0.073) or AD and iNPH (p = 0.542) groups (Table 1).

VBM analysis

ANCOVA showed a main effect of group for GM-V (F_2,39 = 10.150, p < 0.0001), and the post-hoc test revealed that GM-V was significantly smaller in the AD group than in the iNPH (p < 0.0001) and NC (p = 0.036) groups. WM-V (ANCOVA: F_2,39 = 3.674, p = 0.036) was smaller in the iNPH group than in the NC group (p = 0.032), but not in the AD group (p = 0.451). LV-V (ANCOVA: F_2,42 = 33.320, p < 0.0001) was significantly larger in the iNPH group than in the AD (p < 0.0001) and NC (p < 0.0001) groups and in the AD group compared to the NC group (p = 0.024). TWML-V (ANCOVA: F_2,42 = 11.630, p < 0.0001) was larger in the iNPH group than in the AD (p = 0.012) and NC (p < 0.0001) groups, but there was no difference between the AD and NC groups (p = 0.063). PWML-V (ANCOVA: F_2,42 = 11.630, p < 0.0001) was larger in the iNPH group than in the AD (p = 0.009) and NC (p < 0.0001) groups and in the AD group compared to the NC group (p = 0.019). DWML-V (ANCOVA: F_2,42 = 0.887, p < 0.0001) was significantly larger in the iNPH group than in the AD (p = 0.012) and NC (p < 0.0001) groups, but there was no difference between the AD and NC groups (p = 0.190) (Table 1). As for the WML distribution, DWML-V was dominant in iNPH, while PWML-V was dominant in AD (Fig. 1). Pearson’s coefficient analysis with FWE correction showed that the LV-V positively correlated with TWML-V (r = 0.719, p < 0.0001), PWML-V (r = 0.763, p < 0.0001), and DWML-V (r = 0.642, p < 0.0001) in all participants (Fig. 2). LV-V showed a significant positive correlation with TWML-V (r = 0.770, p < 0.0001), PWML-V (r = 0.736, p < 0.0001), and DWML-V (r = 0.758, p < 0.0001) in the AD group. However, in the NC and iNPH groups, the LV-V did not correlate with TWML-V, PWML-V, and DWML-V (Fig. 2).

Fig. 1

Comparison of WML distribution between the iNPH, AD, and NC groups. DWML-V was dominant in A) iNPH, while PWML-V was dominant in B) AD and C) NC. Example of WML distribution in MRI FLAIR axial images in D) iNPH, E) AD, and F) NC. DWML-V, deep white matter lesion volume; PWML-V, periventricular white matter volume; iNPH, idiopathic normal pressure hydrocephalus; AD, Alzheimer’s disease; NC, normal cognition.

Fig. 2

Correlation between LV-V and WML-V. Pearson’s coefficient analysis with familywise error correction showing that the LV-V positively correlated with TWML-V (r = 0.719, p < 0.0001), PWML-V (r = 0.763, p < 0.0001), and DWML-V (r = 0.642, p < 0.0001) in all participants (A, B, C). In each group, LV-V was significantly positively correlated with TWML-V (r = 0.770, p < 0.0001), PWML-V (r = 0.736, p < 0.0001), and DWML-V (r = 0.758, p < 0.0001) in AD (G, H, I). However, LV-V did not correlate with TWML-V, PWML-V, and DWML-V in iNPH (D, E, F) and NC (K, L, M). iNPH, idiopathic normal pressure hydrocephalus; AD, Alzheimer’s disease; NC, normal cognition; LV-V, lateral ventricular volume; WML-V, white matter lesion volume; TWML-V, total white matter lesion volume; DWML-V, deep white matter lesion volume; PWML-V, periventricular white matter volume.

Correlation between CSF biomarkers, LV-V, and WML-V

There was a negative correlation between LV-V and Aβ₃₈ in the iNPH group (r = –0.576, p = 0.050), but no other significant correlation between LV-V and Aβ₄₀ (r = –0.533, p = 0.075), Aβ₄₂ (r = –0.344, p = 0.288), t-tau (r = –0.284, p = 0.371), and p-tau (r = –0.431, p = 0.162) was observed (Fig. 3). In the AD group, there was no correlation between LV-V and Aβ₃₈ (r = –0.361, p = 0.117), Aβ₄₀ (r =–0.227, p = 0.336), Aβ₄₂ (r = –0.165, p = 0.487), t-tau (r = –0.255, p = 0.279), and p-tau (r = –0.321, p = 0.167), nor was there any correlation between TWML-V, DWML-V, PWML-V, and CSF biomarkers in both the iNPH and AD groups.

Fig. 3

Correlation between CSF biomarkers and LV-V in iNPH. There was a negative correlation between LV-V and Aβ₃₈ (r = –0.576, p = 0.050) in iNPH (A), but no significant correlations between LV-V and Aβ₄₀ (r = –0.533, p = 0.075), Aβ₄₂ (r = –0.344, p = 0.288), t-tau (r = –0.284, p = 0.371), and p-tau (r = –0.431, p = 0.162) (B–E). CSF, cerebrospinal fluid; iNPH, idiopathic normal pressure hydrocephalus; LV-V, lateral ventricular volume; Aβ, amyloid-β protein; p-tau, tau phosphorylated at threonine; t-tau, total tau.

Neuropsychological examination

Significant differences were observed in the mean MMSE-J scores between the groups (AD: 18.45±4.41, iNPH: 22.42±4.79, and NC: 28.40±0.80), with ANOVA showing a main effect of group (F_2,42 = 19.307, p < 0.0001) and the post-hoc test revealing that the MMSE-J score was significantly lower in the AD group than in the iNPH (p = 0.037) and NC (p < 0.0001) groups and in the iNPH group compared to the NC group (p = 0.005) (Table 1).

Risk factors for SVD

ANOVA showed no main effect of group for HDL (F_2,42 = 1.877, p = 0.167), LDL (F_2,42 = 0.059, p = 0.943), LDL/HDL (F_2,42 = 2.218, p = 0.123), TG (F_2,42 = 1.513, p = 0.234), HbA1c (F_2,42 = 0.555, p = 0.578), BNP (F_2,42 = 1.068, p = 0.354), and hypertension (F_2,42 = 1.417, p = 0.255). There was a significant difference among groups in the ratio of current smokers (F_2,42 = 3.714, p = 0.033). The smoker ratio in the NC group was higher than that in the AD group (p = 0.044), but no difference was observed between the AD and iNPH groups (p =1.000) (Table 1).

DISCUSSION

Our visual measurements and VBM analysis showed that LV-V, TWML-V, PWML-V, and DWML-V were larger in the iNPH, AD, and NC groups, in that order. LV-V positively correlated with TWML-V, PWML-V, and DWML-V in all participants. Although the ratio of current smokers was significantly higher in the NC group, there was no significant difference in age and other SVD risk factors between the groups. Ventriculomegaly in iNPH is thought to be largely due to CSF stasis [8]. Ventricular enlargement in AD is generally interpreted as being associated with cerebral atrophy; however, we previously reported that the difference in lateral ventricle enlargement and WMLs in AD may partially reflect a difference in the degree of CSF/ISF stagnation [34]. In the present study, LV-V positively correlated with TWML-V, PWML-V, and DWML-V in the AD group. From this perspective, LV-V and WML-V in AD may also have a relationship with the degree of CSF/ISF stasis. However, unlike iNPH, WMLs in AD are thought to be associated with neurodegeneration [24]. The association between WMLs, cognitive decline, and GM (cortical) volume in mild cognitive impairment (MCI) and AD has recently attracted attention. Although WML-V tended to be high and GM-V tended to be low in AD, the association between the degree of WML-V and GM-V was rather strong in MCI [41, 42]. Along with GM-V, measuring cortical thickness is a popular method to assess the degree of neurodegeneration in individuals with dementia. However, the association between cortical thickness and WMLs is not uniform. In a previous study, frontal cortex thickness was negatively correlated with the degree of WMLs, while parietal cortical thickness was positively correlated with the degree of WMLs in AD. Additionally, both parietal and frontal cortical thicknesses positively correlated with the degree of WMLs in MCI [43]. These results suggest that a paradoxical relationship between WMLs and cortical thickness could result from neuroinflammatory processes induced by AD pathology and WMLs. Alternatively, it might reflect a region-specific and disease-stage dependent compensatory hypertrophy in response to a compromised network [43]. In the present study, the GM-V in AD was significantly smaller than that in iNPH and NC. Given that AD had a larger WML-V than NC, we can interpret that increased WML-V in AD is strongly related to cortical atrophy. In contrast, while TWML-V, DWML-V, and PWML-V were significantly larger in iNPH than in AD and NC, GM-V was preserved compared to AD and had no significant difference compared to NC. This finding suggests that WMLs in iNPH is pathophysiologically different from that in AD, which is related more to fluid stagnation than neurodegeneration.

Regarding WMLs distribution, DWML-V was predominant in the iNPH group, while PWML-V was predominant in AD and NC groups. To investigate whether fluid stagnation is related to pathophysiology of PWMLs in iNPH, we predicted that PWML-V would be dominant compared to DWML-V in iNPH [29, 30]. However, DWML-V was predominant over PWML-V in iNPH. A recent study examined differences in pattern of WM diffusion alterations by directly comparing iNPH and AD using diffusion tensor imaging [44]. In this study, higher fractional anisotropy was observed in the corona radiata in the periventricular fibers in NPH patients, and ventricular volumes were correlated with diffusivity parameters in the tracts next to the ventricles in both AD and iNPH patients [44]. This indicates that differences in PWML-V between iNPH and AD are mainly affected by the difference in the degree of CSF stasis. In contrast, pathophysiological differences in DWMLs between iNPH and AD remain unclear. Although the pathophysiology of iNPH is not well elucidated, disturbances in CSF/ISF dynamics and resistance, brain parenchyma alterations, and vascular abnormalities are thought to be involved [3]. For a long time, CSF flow was thought to be dependent on the ratio of absorption and production, termed the bulk flow theory, but has recently been thought to co-exist with pulsatile flow, based on cardiac pulse pressure, through the Virchow-Robin (periventricular) spaces surrounding the intracranial vessels in the brain parenchyma [45]. The pathogenesis of DWMLs in iNPH may be associated with the stagnation of the transport system in the brain parenchyma.

CSF concentrations of Aβ₄₂, Aβ_42/38, and Aβ_42/40 were significantly decreased in AD compared to iNPH, while Aβ₄₀ and Aβ₃₈ tended to be low in iNPH. P-tau and t-tau showed a significant elevation in AD. The lower CSF Aβ₄₂ levels in AD than those in iNPH are considered a result of Aβ₄₂ accumulation in the brain, while the lower CSF Aβ₃₈ and Aβ₄₀ levels in iNPH than those in AD are considered a result of decreased CSF clearance in iNPH [16, 46]. Recently, the concentration ratio of Aβ₄₂ to Aβ₄₀ or Aβ₄₂ to Aβ₃₈ has been suggested to be superior to the concentration of Aβ₄₂ alone in discriminating between AD and non-AD patients [46 –48]. The concentrations of CSF tau species are known to be lower in iNPH patients than in AD patients [16]. The results of CSF biomarkers in our study were consistent with previous results. In recent studies, lower CSF Aβ₃₈ and Aβ₄₀ levels were consistently associated with increased WMLs in cognitive healthy participants, patients with subjective cognitive deficits, and those with MCI, while lower CSF Aβ₄₂ levels were associated with WMLs mainly in AD; therefore, WMLs may be associated with alterations in the production or clearance of Aβ species, particularly Aβ₃₈ and Aβ₄₀ [49]. In the present study, while there was no significant correlation between CSF biomarkers (Aβ₃₈, Aβ₄₀, Aβ₄₂, t-tau, and p-tau) and TWML-V, DWML-V, and PWML-V in iNPH or AD, iNPH had larger WML-V and lower CSF Aβ₃₈ and Aβ₄₀ levels than AD. Therefore, the difference in WML-V between iNPH and AD may have some relationship with the different clearance of Aβ₃₈ and Aβ₄₀. A recent systematic review on the association between WMLs and amyloid pathology, as measured by PET, concluded that amyloid and WMLs have independent but additive effects on dementia risk. However, it remains unclear if WMLs accelerate amyloid pathology or exacerbate the effect of amyloid pathology on the risk of developing cognitive impairment [50, 51]. Whether WMLs are more associated with tau or Aβ pathology remains to be further elucidated in upcoming studies. In the present study, as for the correlation between CSF biomarkers and LV-V, a significant negative correlation was observed between CSF Aβ₃₈ and LV-V in iNPH, while CSF Aβ₄₀ and LV-V had a negative correlation tendency. The correlation between CSF Aβ₃₈ values and LV-V supports the usefulness of Aβ₃₈, and maybe Aβ₄₀, as a marker of impaired clearance in iNPH.

In this study, we only included AD patients who fulfilled the biomarker diagnostic criteria and iNPH patients who responded to the CSF drainage test, to make a clear distinction between these diseases. However, the largest limitation of our study is the small number of participants in a single center. Despite this, it is still possible that there is no statistical difference in risk factors for SVD between groups. Additionally, we did not analyze our results from the perspective of arterial pulsations. The degree of arterial stiffness and arteriosclerosis and SVD risk factors between groups need to be evaluated in more detail using imaging and physiological function examinations. Moreover, we did not measure cortical thickness in this study. Evaluating cortical thickness with GM-V would provide more information and insight into the pathophysiological background of WMLs in AD and iNPH. We would like to investigate the relationship between cortical thickness and GM-V, WMLs, LV-V, and CSF biomarkers in AD and iNPH in a future study.

Based on the association between LV-V, WML-V, GM-V, and CSF biomarkers in iNPH and AD according to the present and previous studies, it seems that fluid stagnation and decreased clearance are dominant in iNPH, while neurodegenerative changes are dominant in AD, as a pathophysiological background for WMLs, when risk factors for SVD are comparable. However, these pathophysiological changes co-exist. The differences in the extent and distribution of PWMLs and DWMLs between iNPH and AD, especially predominance of DWML-V over PWML-V in iNPH, may be related to decreased fluid and Aβ clearance. Increasing the number of participants and comparing the relationship between LVV and PWML-V, as well as DWML-V before and after shunting surgery for iNPH, and evaluating changes in the extent and distribution of WMLs with CSF biomarkers and clinical presentation could help clarify the involvement of fluid stagnation in the formation of WMLs. iNPH and AD are sometimes misdiagnosed on MRI, and differences in WMLs extent and distribution may help distinguish them in clinical settings. Since iNPH and AD have similar imaging findings, differential diagnosis should not be based on imaging alone but on a comprehensive review of the patient’s history and clinical symptoms, as well as the results of psychological tests and response to CSF tap tests and shunt procedures. Further research is required to elucidate the pathophysiological differences underlying WMLs in iNPH and AD.

Footnotes

ACKNOWLEDGMENTS

We would like to thank M. Miyanohara for technical assistance. We would also like to thank the staff in the Division of Neurology, Department of Medicine, Showa University School of Medicine for their cooperation during the study. This study was supported by Grants-in-Aid for Scientific Research (Kakenhi) from the Japan Society for the Promotion of Science (JSPS) under Grants JP26461266 and JP19K07965 (K.O.), Research and Development Grants from the Japan Agency for Medical Research and Development (AMED) (16dk0207021h 0001) (K.O.), and AMED under Grants JP21dm0207073 and JP21dk0207057 (T.I.).

Authors’ disclosures available online ().

References

Hiraoka

, Meguro

, Mori

(2008) Prevalence of idiopathic normal-pressure hydrocephalus in the elderly population of a Japanese rural community. Neurol Med Chir (Tokyo) 48, 197–199; discussion 199–200.

Adams

, Fisher

, Hakim

, Ojemann

, Sweet

(1965) Symptomatic occult hydrocephalus with “normal” cerebrospinal-fluid pressure. a treatable syndrome. N Engl J Med 273, 117–126.

Nassar

, Lippa

(2016) Idiopathic normal pressure hydrocephalus: A review for general practitioners. Gerontol Geriatr Med 2, 2333721416643702.

Hebert

, Scherr

, Bienias

, Bennett

, Evans

(2003) Alzheimer disease in the US population: Prevalence estimates using the 2000 census. Arch Neurol 60, 1119–1122.

Evans

(1942) An encephalographic ratio for estimating ventricular enlargement and cerebral atrophy. Arch Neurol Psychiatr 47, 931–937.

Sherman

, Citrin

, Gangarosa

, Bowen

(1987) The MR appearance of CSF flow in patients with ventriculomegaly. AJR Am J Roentgenol 148, 193–199.

Ishikawa

, Oowaki

, Takezawa

, Takenaka

, Yamada

, Yamamoto

, Okamoto

(2016) Disproportionately enlarged subarachnoid space hydrocephalus in idiopathic normal-pressure hydrocephalus and its implication in pathogenesis. Acta Neurochir Suppl 122, 287–290.

Brix

, Westman

, Simmons

, Ringstad

, Eide

, Wagner-Larsen

, Page

, Vitelli

, Beyer

(2017) The Evans’ Index revisited: New cut-off levels for use in radiological assessment of ventricular enlargement in the elderly. Eur J Radiol 95, 28–32.

Williams

, Malm

(2016) Diagnosis and treatment of idiopathic normal pressure hydrocephalus. Continuum (Minneap Minn) 22, 579–599.

10.

Kim

, Seo

, Lee

, Kim

, Lee

, Nam

, Na

(2011) Differential diagnosis of idiopathic normal pressure hydrocephalus from other dementias using diffusion tensor imaging. AJNR Am J Neuroradiol 32, 1496–1503.

11.

Reeves

, Karimy

, Kundishora

, Mestre

, Cerci

, Matouk

, Alper

, Lundgaard

, Nedergaard

, Kahle

(2020) Glymphatic system impairment in Alzheimer’s disease and idiopathic normal pressure hydrocephalus. Trends Mol Med 26, 285–295.

12.

Aspelund

, Antila

, Proulx

, Karlsen

, Karaman

, Detmar

, Wiig

, Alitalo

(2015) A dural lymphatic vascular system that drains brain interstitial fluid and macromolecules. J Exp Med 212, 991–999.

13.

Iliff

, Wang

, Liao

, Plogg

, Peng

, Gundersen

, Benveniste

, Vates

, Deane

, Goldman

, Nagelhus

, Nedergaard

(2012) A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci Transl Med 4, 147ra111.

14.

Tarasoff-Conway

, Carare

, Osorio

, Glodzik

, Butler

, Fieremans

, Axel

, Rusinek

, Nicholson

, Zlokovic

, Frangione

, Blennow

, Ménard

, Zetterberg

, Wisniewski

, de Leon

(2015) Clearance systems in the brain-implications for Alzheimer disease. Nat Rev Neurol 11, 457–470.

15.

Fagan

, Perrin

(2012) Upcoming candidate cerebrospinal fluid biomarkers of Alzheimer’s disease. Biomark Med 6, 455–476.

16.

Chen

, Liu

, Zhang

, Relkin

, Xing

, Li

(2017) Cerebrospinal fluid Aβ42, t-tau, and p-tau levels in the differential diagnosis of idiopathic normal-pressure hydrocephalus: A systematic review and meta-analysis. Fluids Barriers CNS 14, 13.

17.

Saito

, Nishio

, Kanno

, Uchiyama

, Hayashi

, Takagi

, Kikuchi

, Yamasaki

, Shimomura

, Iizuka

, Mori

(2011) Cognitive profile of idiopathic normal pressure hydrocephalus. Dement Geriatr Cogn Dis Extra 1, 202–211.

18.

Valotassiou

, Malamitsi

, Papatriantafyllou

, Dardiotis

, Tsougos

, Psimadas

, Alexiou

, Hadjigeorgiou

, Georgoulias

(2018) SPECT and PET imaging in Alzheimer’s disease. Ann Nucl Med 32, 583–593.

19.

Weller

, Hawkes

, Kalaria

, Werring

, Carare

(2015) White matter changes in dementia: Role of impaired drainage of interstitial fluid. Brain Pathol 25, 63–78.

20.

Gouw

, Seewann

, van der Flier

, Barkhof

, Rozemuller

, Scheltens

, Geurts

(2011) Heterogeneity of small vessel disease: A systematic review of MRI and histopathology correlations. J Neurol Neurosurg Psychiatry 82, 126–135.

21.

Kim

, MacFall

, Payne

(2008) Classification of white matter lesions on magnetic resonance imaging in elderly persons. Biol Psychiatry 64, 273–280.

22.

Wardlaw

, Valdés Hernández

, Muñoz-Maniega

(2015) What are white matter hyperintensitiesmade of? Relevance to vascular cognitive impairment. J Am HeartAssoc 4, 001140.

23.

Barber

, Scheltens

, Gholkar

, Ballard

, McKeith

, Ince

, Perry

, O’Brien

(1999) White matter lesions on magnetic resonance imaging in dementia with Lewy bodies, Alzheimer’s disease, vascular dementia, and normal aging. J Neurol Neurosurg Psychiatry 67, 66–72.

24.

McAleese

, Walker

, Graham

, Moya

ELJ

, Johnson

, Erskine

, Colloby

, Dey

, Martin-Ruiz

, Taylor

, Thomas

, McKeith

, De Carli

, Attems

(2017) Parietal white matter lesions in Alzheimer’s disease are associated with cortical neurodegenerative pathology, but not with small vessel disease. Acta Neuropathol 134, 459–473.

25.

Pietroboni

, Scarioni

, Carandini

, Basilico

, Cadioli

, Giulietti

, Arighi

, Caprioli

, Serra

, Sina

, Fenoglio

, Ghezzi

, Fumagalli

, De Riz

, Calvi

, Triulzi

, Bozzali

, Scarpini

, Galimberti

(2018) CSF β-amyloid and white matter damage: A new perspective on Alzheimer’s disease. J Neurol Neurosurg Psychiatry 89, 352–357.

26.

Kandel

, Avants

, Gee

, McMillan

, Erus

, Doshi

, Davatzikos

, Wolk

(2016) White matter hyperintensities are more highly associated with preclinical Alzheimer’s disease than imaging and cognitive markers of neurodegeneration. Alzheimers Dement (Amst) 4, 18–27.

27.

McAleese

, Firbank

, Dey

, Colloby

, Walker

, Johnson

, Beverley

, Taylor

, Thomas

, O’Brien

, Attems

(2015) Cortical tau load is associated with white matter hyperintensities. Acta Neuropathol Commun 3, 60.

28.

Krauss

, Regel

, Vach

, Orszagh

, Jüngling

, Bohus

, Droste

(1997) White matter lesions in patients with idiopathic normal pressure hydrocephalus and in an age-matched control group: A comparative study. Neurosurgery 40, 491–495.

29.

Murata

, Handa

, Mori

, Nakano

(1981) The significance of periventricular lucency on computed tomography: Experimental study with canine hydrocephalus. Neuroradiology 20, 221–227.

30.

Akiguchi

, Ishii

, Watanabe

, Kawasaki

, Yagi

, Shiino

, Shirakashi

, Kawamoto

(2008) Shunt-responsive parkinsonism and reversible white matter lesions in patients with idiopathic NPH. J Neurol 255, 1392–1399.

31.

Yokota

, Vijayasarathi

, Cekic

, Hirata

, Linetsky

, Ho

, Kim

, Salamon

(2019) Diagnostic performance of glymphatic system evaluation using diffusion tensor imaging in idiopathic normal pressure hydrocephalus and mimickers. Curr Gerontol Geriatr Res 2019, 5675014.

32.

Ringstad

, Vatnehol

SAS

, Eide

(2017) Glymphatic MRI in idiopathic normal pressure hydrocephalus. Brain 140, 2691–2705.

33.

Ringstad

, Valnes

, Dale

, Pripp

, Vatnehol

, Emblem

, Mardal

, Eide

(2018) Brain-wide glymphatic enhancement and clearance in humans assessed with MRI. JCI Insight 3, e121537.

34.

Kuroda

, Honma

, Mori

, Futamura

, Sugimoto

, Yano

, Kinno

, Murakami

, Ono

(2020) Increased presence of cerebral microbleeds correlates with ventricular enlargement and increased white matter hyperintensities in Alzheimer’s disease. Front Aging Neurosci 12, 13.

35.

Mori

, Ishikawa

, Kato

, Kazui

, Miyake

, Miyajima

, Nakajima

, Hashimoto

, Kuriyama

, Tokuda

, Ishii

, Kaijima

, Hirata

, Saito

, Arai

, Japanese

Society of Normal Pressure Hydrocephalus

(2012) Guidelines for management of idiopathic normal pressure hydrocephalus: Second edition. Neurol Med Chir (Tokyo) 52, 775–809.

36.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

Jr. , Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263–269.

37.

Folstein

, Folstein

, McHugh

(1975) ‘Mini-mental state’. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

38.

Fazekas

, Kleinert

, Offenbacher

, Schmidt

, Kleinert

, Payer

, Radner

, Lechner

(1993) Pathologic correlates of incidental MRI white matter signal hyperintensities. Neurology 43, 1683–1689.

39.

Fazekas

, Chawluk

, Alavi

, Hurtig

, Zimmerman

(1987) MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. AJR Am J Roentgenol 149, 351–356.

40.

Syaifullah

, Shiino

, Kitahara

, Ito

, Ishida

, Tanigaki

(2020) Machine learning for diagnosis of AD and prediction of MCI progression from brain MRI using brain anatomical analysis using diffeomorphic deformation. Front Neurol 11, 576029.

41.

Vipin

, Foo

HJL

, Lim

JKW

, Chander

, Yong

, Ng

ASL

, Hameed

, Ting

SKS

, Zhou

, Kandiah

(2018) Regional white matter hyperintensity influences grey matter atrophy in mild cognitive impairment. J Alzheimers Dis 66, 533–549.

42.

Wang

, Chen

, Cai

, Hu

, Xu

, Wang

, Cao

, Tan

, Yu

; Alzheimer’s Disease Neuroimaging Initiative (2020) Associations of white matter hyperintensities with cognitive decline: A longitudinal study. J Alzheimers Dis 73, 759–768.

43.

Jacobs

, Clerx

, Gronenschild

, Aalten

, Verhey

(2014) White matter hyperintensities are positively associated with cortical thickness in Alzheimer’s disease. J Alzheimers Dis 39, 409–422.

44.

Hořínek

, Štěpán-Buksakowska

, Szabó

, Erickson

, Tóth

, Šulc

, Beneš

, Vrána

, Hort

, Nimsky

, Mohapl

, Roček

, Vécsei

, Kincses

(2016) Difference in white mattermicrostructure in differential diagnosis of normal pressurehydrocephalus and Alzheimer’s disease. Clin Neurol Neurosurg 140, 52–59.

45.

Brinker

, Stopa

, Morrison

, Klinge

(2014) A new look atcerebrospinal fluid circulation. Fluids Barriers CNS 11, 10.

46.

Jeppsson

, Wikkelsö

, Blennow

, Zetterberg

, Constantinescu

, Remes

, Herukka

, Rauramaa

, Nagga

, Leinonen

, Tullberg

(2019) CSF biomarkers distinguish idiopathic normal pressurehydrocephalus from its mimics. J Neurol Neurosurg Psychiatry 90, 1117–1123.

47.

Hansson

, Lehmann

, Otto

, Zetterberg

, Lewczuk

(2019) Advantages and disadvantages of the use of the CSF Amyloid β (Aβ) 42/40 ratio in the diagnosis of Alzheimer’s disease. Alzheimers Res Ther 11, 34.

48.

Janelidze

, Zetterberg

, Mattsson

, Palmqvist

, Vanderstichele

, Lindberg

, van Westen

, Stomrud

, Minthon

, Blennow

, Swedish BioFINDER study group, Hansson

(2016) CSF Aβ42/Aβ40 and Aβ42/Aβ38 ratios: Better diagnostic markers of Alzheimer disease. Ann Clin Transl Neurol 3, 154–165.

49.

van Westen

, Lindqvist

, Blennow

, Minthon

, Nägga

, Stomrud

, Zetterberg

, Hansson

(2016) Cerebral white matter lesions - associations with Aβ isoforms and amyloid PET. Sci Rep 6, 20709.

50.

Roseborough

, Ramirez

, Black

, Edwards

(2017) Associations between amyloid β and white matter hyperintensities: A systematic review. Alzheimers Dement 13, 1154–1167.

51.

Soldan

, Pettigrew

, Zhu

, Wang

, Moghekar

, Gottesman

, Singh

, Martinez

, Fletcher

, DeCarli

, Albert

, BIOCARD Research Team (2020) White matter hyperintensities and CSF Alzheimer disease biomarkers in preclinical Alzheimer disease. Neurology 94, e950–e960.