Forget About Memory: Disentangling the Amnestic Syndrome in Idiopathic Normal Pressure Hydrocephalus

Abstract

Background:

Idiopathic normal pressure hydrocephalus (iNPH) can present with both episodic amnestic syndrome and biomarkers of Alzheimer’s disease (AD) pathology.

Objective:

To examine the associations between amnestic syndrome and cerebrospinal fluid (CSF) AD biomarkers in iNPH and the CSF tap test response in iNPH patients with amnestic syndrome.

Methods:

We used the Free and Cued Selective Reminding Test to divide iNPH into amnestic and non-amnestic patients. We compared their clinical, biological, and radiological characteristics and examined the reversibility of gait spatiotemporal parameters and neuropsychological performances after a CSF tap test. Univariate and multiple linear regression models examined the association between memory performance and clinical-biological characteristics.

Results:

Sixty-two non-amnestic patients (mean age 77.0±7.0 years, 38.7% female) and thirty-eight amnestic patients (mean age 77.0±5.9 years, 36.8% female) presented similar levels of AD biomarkers and clinical-radiological profiles. Global cognition and education levels were lower in the amnestic iNPH group. We found no association between AD biomarkers and memory performances (total tau: β= –4.50; 95% CI [–11.96;2.96]; p = 0.236; amyloid-β (1–42): β= 8.60, 95% CI [–6.30;23.50]; p = 0.240). At baseline, amnestic iNPH patients performed worse on executive functions, attention, and gait speed but improved similarly to the non-amnestic iNPH patients after the tap test.

Conclusions:

In our clinical sample of iNPH patients, we confirm the lack of specificity of the amnestic profile for predicting AD pathology. Clinicians should not preclude amnestic iNPH patients from undergoing an invasive procedure of CSF derivation.

Keywords

Alzheimer’s disease amyloid protein cerebrospinal fluid biomarkers diagnosis episodic memory idiopathic normal pressure hydrocephalus memory deficits neuropsychology normal pressure hydrocephalus tau protein

INTRODUCTION

Idiopathic normal pressure hydrocephalus (iNPH) patients typically present with the classical neuropsychological profile of poor executive functions and apathy.1 –3 However, an essential proportion of iNPH patients exhibit typical episodic memory impairment similar to typical Alzheimer’s disease (AD). 4

It remains unclear whether memory decline in iNPH is attributable to a comorbid AD or directly to iNPH. 5 The Free and Cued Selective Reminding Test (FCSRT) was shown to be sensitive to the amnestic syndrome of the hippocampal type, which identifies early-stage of AD 6 and reliably discriminates typical AD among different neurodegenerative diseases. 7 However, its specificity has been questioned in several studies showing severe amnesia in frontotemporal lobar degeneration, Lewy body disease, and vascular dementia. 8 The FCSRT has never been used to examine episodic memory in iNPH.

We aim to compare the clinical and biological profile of amnestic and non-amnestic iNPH as defined with the FCSRT, which distinguishes the memory pathways components, i.e., encoding, consolidation, and retrieval. 9 Given that AD classically presents with an amnestic syndrome of the hippocampal type, and since 25 to 40% of iNPH present with a comorbid AD pathology, 10 we hypothesize that the amnestic syndrome of the hippocampal type in iNPH is associated with CSF biomarkers for AD (i.e., total tau protein and amyloid-β protein (1–42) (Aβ₄₂). Second, we examine if this amnestic syndrome predicts the reversibility of symptoms in iNPH (i.e., gait and cognition). As our group already demonstrated that comorbid AD pathology was associated with a poor response to the CSF tap test on the gait parameters, 11 we hypothesize that iNPH patients with an amnestic syndrome of the hippocampal type would present poorer reversibility of gait and cognition.

METHODS

Patients

In this prospective, cross-sectional study, we initially included 261 consecutive iNPH patients referred to the Department of Neurology of the Geneva University Hospitals from May 2011 to February 2019. All the patients underwent the Geneva’s protocol—a standard protocol for iNPH assessment, as previously described. 12 Briefly, the patients performed a comprehensive quantitative gait and cognitive evaluation before and 24 h after a CSF tap test. All the patients underwent a neuroimaging study (MRI or CT scan) before undergoing the CSF tap test, allowing for the quantification of age-related white matter changes. The patients were asked for their consent to undergo a CSF analysis for AD pathology biomarkers, i.e., total tau, phosphorylated tau, and Aβ₄₂ proteins. The local laboratory cut-offs at the time of the analyses were as follows: total tau, 360 ng/L; phosphorylated tau, 60 ng/L; Aβ₄₂, 450 ng/L. These cut-offs were used to examine the amyloid and tau status of the iNPH according to the NIA-AA research framework for the biological definition of AD. 13 We further defined the ratio of total tau over Aβ₄₂ that has been shown to identify AD from control subjects reliably. 14 The patients were included if they (i) fulfilled international consensus guideline criteria for iNPH; 15 (ii) completed the Free and Cued Selective Reminding Test (FCSRT); (iii) underwent a gait and cognition evaluation before and after a CSF tap test. Exclusion criteria were: (i) an acute medical illness in the past three months; (ii) a CSF tap test 3 months before the evaluation; (iii) an age inferior to 60 years; (iv) a change in usual treatment between the two assessments; (v) an insufficient knowledge of French to perform the neuropsychological evaluation. One hundred patients were included in the analyses (Fig. 1).

Fig. 1

Patient selection.

Procedures

Neuropsychological assessment

A board-certified neuropsychologist assessed all patients at baseline and 24 h after the CSF tap test. 12 The baseline comprehensive neuropsychological assessment focused on cognitive domains frequently impaired in iNPH: global cognition (Mini-Mental State Examination (MMSE) 16 ), executive functions (Color Trails Test 17 ; phonemic and semantic verbal fluencies 18 ; Stroop test 19 ), attention and working memory (Wechsler Adult Intelligence Scale-III (WAIS-III): symbol digit modalities test, verbal digit spans 20 ; Wechsler Memory Scale-III (WMS-III) spatial spans 21 ), verbal episodic memory (French version of the Free and Cued Selective Reminding Test (FCSRT) 22 ) and behavior (Starkstein apathy scale 23 ). A set of tests from the executive functions and attention/working memory domains were repeated 24 hours after the CSF tap test (Color Trails Test, verbal fluency, Stroop Test, WAIS-III, WMS-III). The following formula expressed the change relative to the baseline: delta(parameter) [%] = (post-CSF tapping parameter – pre-CSF tapping parameter)/pre-CSF tapping parameter.

We examined verbal episodic memory with the FCSRT due to its ability to identify the amnestic syndrome of the hippocampal type 6 and to distinguish encoding from retrieval dysfunctions. 9 The procedure’s description is included in the Supplementary Material. This scale provided a total “free recalls” score and a total “total recalls” score (maximum score: 48) by summing each of the three free and total recalls repetitions, excluding the delayed recalls. Previous research identified a cut-off score for the sum of total recalls defining the amnestic syndrome of the hippocampal type suggestive of AD diagnosis 6 . We used this cut-off of 40/48 to divide the patients into two groups: amnestic iNPH (<40) and non-amnestic iNPH (≥40).

Gait assessment

The patients were asked to walk on a 10-meter walkway at a self-selected, comfortable speed while wearing their own shoes. Each patient had to perform the simple walking task to the best of their capacity without any distracting charge. The patients were tested in the morning to minimize fatigue. Certified physical therapists tested all the patients for gait spatiotemporal parameters before and 24 h after the CSF tap test. The quantitative spatiotemporal parameters were computed over 6 meters (following previously described guidelines 24 ) using heel and toe marker trajectories captured by a 12-cameras optoelectronic system (until 2015: Vicon Mx3+, Vicon Peak, Vicon Motion Systems, Oxford, UK; after 2015: Opus7+, Qualisys, Sweden). The following metrics were extracted: walking speed [m/s], stride time [s] and its coefficient of variation — stride time variability (STV) [%] —, stride length [m], swing phase [% of the gait cycle], and step width [m]. The coefficient of variability of stride time was computed as follows: STV = [(standard deviation of stride time)/(mean of stride time)]*100. The long-term improvement in gait—i.e., one year after ventriculoperitoneal shunt surgery—was assessed qualitatively by the neurosurgeon.

Covariates

The disease duration was calculated using the beginning of gait symptoms in months, as gait disturbance usually represents the most prominent symptom of iNPH. 15 Medical comorbidities were assessed with the Global Health Status score on a range of 0–9, according to the presence of the following conditions: diabetes mellitus, hypertension, angina, chronic heart failure, arthritis, history of myocardial infarction, history of stroke, chronic obstructive pulmonary disease, arthritis, and depression. 25 Parkinsonism was defined, according to the UK Parkinson’s Disease Society brain bank clinical diagnostic criteria, 26 as the presence of bradykinesia and at least one of the following signs: rigidity, rest tremor, postural instability. Regarding the neuroimaging characteristics, patients usually underwent MRI with standard protocol. Morphometric analyses were not available in our center. Six patients underwent a CT scan due to the presence of contraindications to MRI. The patients were assessed for white matter changes using the Age-Related White Matter Changes Scale, ranging from 0 to 30. 27

Statistics

The data normality was assessed with frequency distribution histograms and the Kolmogorov-Smirnov statistic. Since some variables were skewed, the following variables were normalized with a logarithmic transformation: disease duration, total tau, and Aβ₄₂. Missing value analysis was performed for some variables, and subsequent multiple imputations with the linear regression method were conducted using five iterations. The following variables are affected: disease duration, MMSE, number of pharmacological treatments, presence of Parkinsonism, total tau, and Aβ₄₂. As appropriate, between-group comparisons for clinical characteristics and differences were performed using independent samples t-tests, Mann Whitney u, Chi-squared tests, or Fisher’s exact tests. We examined the comparison of learning curves for each individual free or total recall trial, as well as delayed recall scores, using a mixed linear regression model with the following parameters: free and total recall scores as dependent variables; each successive trial, the group (i.e., amnestic versus non-amnestic patients) and the interaction trial * group as fixed effects; the patients as random effects. Pairwise comparisons of conditional marginal effects were computed and adjusted according to Bonferroni. The associations between the memory performance and the clinical, demographical, and biological variables were examined using univariate and multiple blockwise linear regressions, adjusted for age, sex, disease duration, education level, MMSE, CSF biomarkers (t-tau and Aβ₄₂), and white matter changes. We omitted p-tau in the regression analyses to respect the “one in ten rule” for keeping the risk of overfitting low and since the number of missing values was high for this variable (37 missing cases). For the imputed variables — i.e., disease duration, total tau, and Aβ₄₂—, the reported R2 represents the mean of the R² for each iteration. Collinearity was not detected: each tolerance value greater than 0.2; largest variance inflation factor (VIF) = 1.179 (smaller than 10); average VIF = 1.046 (not substantially greater than 1). All statistics were performed using the IBM Statistical Package for Social Sciences, version 25 (SPSS Inc., Chicago, IL, USA) and Stata, version 18 (StataCorp, College Station, TX, USA).

Standard protocol approvals, registrations, and patients consents

All the patients gave their free and informed consent. The research protocol was approved by the Geneva University Hospitals Committee on Human Research (CE 09-160R).

RESULTS

Clinical, demographical, and biological characteristics

Most patients presented a non-amnestic profile: sixty-two (62%) were classified as non-amnestic (FCSRT sum of total recalls ≥40). In comparison, thirty-eight patients (38%) were amnestic (FCSRT sum of total recalls <40). The mean sum of free recalls was 10.7±5.2 (range: [2;20]) for amnestic iNPH patients and 21.0±6.5 (range: [1;34]) for non-amnestic iNPH patients; the mean sum of total recalls was 29.2±8.1 (range: [7;39]) for the amnestic group and 44.7±2.6 (range: [40;48]) for the non-amnestic group. For each group, learning curves were derived from each individual trial for free recalls (FR) and total recalls (TR) scores, as well as delayed scores (see Fig. 2). The mean values for each trial are shown in Supplementary Table 1. Between-group comparisons of the slopes were performed using a mixed linear regression model. For free recall scores, the slope between mean the trials FR1 and FR2 was significantly different between amnestic and non-amnestic patients (contrast = 1.33; 95% CI [0.19;2.48], p = 0.013), but not between FR2 and FR3, and FR3 and delayed free recall. No significant difference was found between the learning slopes for each individual trial of total recall scores in each group.

Fig. 2

Learning curves in free and total recall (means) in the FCSRT. FR, free recall; TR, total recall (defined as the sum of free and cued recalls).

Both groups were compared in Table 1. Overall, amnestic and non-amnestic iNPH patients presented a similar clinical profile: mean age 77.0±5.9 years (amnestic) versus 77.0±7.0 years (non-amnestic); female percentage 36.8% (amnestic) versus 38.7% (non-amnestic). The amnestic iNPH were significantly less educated than the non-amnestic iNPH (10.2±3.5 years versus 13.1±3.0 years; t(98) = –4.35, p < 0.001), with a moderate effect size (r = 0.40). Amnestic patients had a poorer global cognition than the non-amnestic group (MMSE: 22.6±3.3 versus 25.8±2.8; t(98) = –4.87, p < 0.001; r = 0.44). Similarly, regarding other cognitive functions, amnestic iNPH performed worse at baseline on several subtests of the executive functions and attention/working memory domains: Color Trails Test, Part 1 and 2; phonemic and semantic verbal fluencies; Stroop test, Dot and Color conditions; Symbol Digit Modalities Test; forward verbal digit span (see Supplementary Table 2). Regarding the gait, although both groups presented a similar motoric profile (Parkinsonism: 34.3% in the amnestic group; 22.0% in the non-amnestic group, p = 0.194), the amnestic patients walked slower at baseline (mean walking speed: 0.7±0.3 [m/s]) than non-amnestic patients (mean walking speed: 0.8±0.3 [m/s]), t(94) = –2.00, p = 0.049, r = 0.20). The other gait parameters – i.e., cadence, stride time variability, stride length, swing phase, and step width – were similar at baseline between groups (see Table 1).

Table 1

Characteristics of the iNPH patients (n = 100)

Characteristics	Amnestic, n = 38	Non-amnestic, n = 62	p
Demographics
Age [y]	77.0 (5.9)	77.0 (7.0)	0.965
Female, n (%)	14.0 (36.8)	24.0 (38.7)	0.852
Education level [y]	10.2 (3.5)	13.1 (3.0)	<0.001
Clinical characteristics
Disease duration [mo]	33.2 (25.4)	42.6 (48.5)	0.263
Parkinsonism, n (%)	12.0 (34.3)	13.0 (22.0)	0.194
Medications, n	5.2 (2.4)	4.6 (3.1)	0.406
Comorbidities [0–10]^*	1.7 (1.0)	1.7 (1.1)	0.975
MMSE [0–30]	22.6 (3.3)	25.8 (2.8)	<0.001
CSF proteins level
Total tau [ng/L]	220.6 (140.0)	197.8 (106.6)	0.914
Phosphorylated tau [ng/L]	40.6 (15.7)	42.2 (16.6)	0.710
Aβ₄₂ [ng/L]	606.7 (180.8)	678.3 (292.6)	0.476
Total tau/Aβ₄₂ ratio	0.40 (0.35)	0.32 (0.20)	0.225
Radiological characteristics
White matter changes^$ [0–30]	7.6 (5.0)	7.0 (5.2)	0.586
Frontal [0–6]	3.1 (1.2)	2.8 (1.5)	0.315
Parieto-occipital [0–6]	2.8 (1.7)	2.8 (2.1)	0.853
Temporal [0–6]	0.7 (1.2)	0.8 (1.3)	0.861
Basal ganglia [0–6]s	0.5 (1.0)	0.5 (1.0)	0.879
Infratentorial [0–6]	0.5 (1.0)	0.3 (0.7)	0.169
Gait parameters
Walking speed [m/s]	0.7 (0.3)	0.8 (0.3)	0.049
Cadence [# steps/min]	96.3 (17.9)	101.2 (13.1)	0.124
Stride time variability [%]	6.4 (10.9)	4.7 (8.6)	0.403
Stride length [m]	0.8 (0.3)	0.9 (0.3)	0.055
Swing phase [%]	67.3 (5.2)	66.2 (3.9)	0.234
Step width [m]	0.1 (0.1)	0.1 (0.0)	0.464

CSF, cerebrospinal fluid; iNPH, idiopathic normal pressure hydrocephalus; MMSE, Mini-Mental State Examination. Values are indicated as means (standard deviation) or numbers (percentages), as appropriate. ^*Comorbidities were evaluated using the Global Health Score. ^$White matter changes were evaluated using the Age-Related White Matter Changes scale.

Memory and its association with biological profiles

Despite a different clinical profile for memory, both groups presented similar CSF biomarkers for AD pathology (see Table 1). As shown in Table 2, associations between the clinical characteristics and the FCSRT sum of total recalls were examined using a post-imputation linear regression model (univariate and multiple analyses). The CSF biomarkers for AD were not associated with memory performances (t-tau: β= –4.50; 95% CI [–11.96;2.96]; p = 0.236; Aβ₄₂: β= 8.60, 95% CI [–6.30;23.50]; p = 0.240). However, the FCSRT sum of total recalls was associated with the level of education (β= 0.71; 95% CI [0.23;1.20]; p = 0.004), as well as the MMSE (β= 1.09; 95% CI [0.59;1.58]; p < 0.001), even after adjusting for other covariates. Taken together, the clinical covariates explained 33.8% of the memory performances (adjusted R² = 0.338).

Table 2

Univariate and multiple associations between FCSRT total recalls (dependent variable) and clinical characteristics (independent variables)

	Univariate analysis				Multiple analysis
n = 100	β	95% CI	p	R²	β	95% CI	p	R²
Age	–0.16	[–0.43;0.11]	0.251	0.016	–0.01	[–0.31;0.30]	0.977	0.338
Sex	1.46	[–2.19;5.10]	0.433	0.070	2.84	[–0.57;6.26]	0.103
Education level	0.71	[0.23;1.20]	0.004	0.085	0.63	[0.15;1.12]	0.011
Disease duration	4.91	[0.66;9.16]	0.024	0.062	3.00	[–1.08;7.05]	0.149
Comorbidities	0.39	[–1.32;2.10]	0.657	0.002	1.22	[–0.45;2.89]	0.153
MMSE	1.09	[0.59;1.58]	<0.001	0.187	0.93	[0.35;1.52]	0.002
CSF AD protein levels
Total tau	–4.50	[–11.96;2.96]	0.236	0.018	–3.34	[–10.91;4.22]	0.386
Aβ₄₂	8.60	[–6.30;23.50]	0.240	0.033	9.59	[–5.40;24.57]	0.191
White matter changes	–0.09	[–0.44;0.26]	0.628	0.003	–0.03	[–0.38;0.33]	0.873

Dependent variable: FCSRT recall scores [0;48]; independent variables: age, sex, education level, disease duration, comorbidities, MMSE, total tau, Aβ₄₂, and white matter changes. AD, Alzheimer’s disease; CSF, cerebrospinal fluid; iNPH, idiopathic normal pressure hydrocephalus; MMSE, Mini-Mental State Examination. The disease duration, MMSE, total tau, and Aβ₄₂ are handled as log-transformed data. The reported R² is based on the average of the five iterations conducted in the multiple imputation model. Significant associations (p < 0.05, two-tailed) are highlighted in bold.

Interestingly, only one patient matched the current biological definition for AD according to NIA-AA 13 and presented with an A+/T+/N+ profile. This single patient was classified as amnestic iNPH.

Reversibility of gait and cognition

The reversibility of gait and cognition to the CSF tap test was examined in Table 3. Regarding the spatiotemporal parameters of gait, both groups walked faster after the CSF tap test, with a similar improvement in walking speed (amnestic iNPH: +28%; non-amnestic iNPH: +20%, p = 0.332) and other gait parameters (see Supplementary Table 2).

Table 3

Comparisons between the clinical changes (gait and cognition) in iNPH patients (n = 100)

Delta change in parameters, %	Amnestic	Non-amnestic	p
	n = 38	n = 62
Gait
Walking speed	+28 (54)	+20 (27)	0.332
Cadence	–46 (9)	–48 (5)	0.213
Stride time variability^#	–2 (70)	+16 (139)	0.501
Stride length	+14 (24)	+13 (19)	0.804
Swing phase	–4 (54)	–8 (54)	0.764
Step width	0 (19)	–2 (18)	0.491
Cognition
Executive functions
Colour Trails Test
Part 1	+1 (29)	–1 (28)	0.778
Part 2	–8 (22)	–5 (22)	0.575
Index^*	+1 (76)	+33 (1.22)	0.280
Verbal fluency (2 min)
Semantic	+15 (50)	+23 (34)	0.420
Phonemic	+70 (151)	+15 (64)	0.084
Stroop Test
Dot	–1 (16)	–1 (16)	0.905
Colour	–8 (15)	–5 (30)	0.615
Word	0 (28)	–4 (23)	0.531
Index^$	+4 (39)	–1 (28)	0.460
Attention, working memory
Wechsler Adult Intelligence Scale-III
Symbol Digit Modalities Test	+6 (20)	+7 (23)	0.878
Forward verbal digit span	+10 (23)	+6 (21)	0.378
Backward verbal digit span	+4 (28)	+10 (25)	0.369
Wechsler Memory Scale-III
Forward visual span	+16 (32)	+6 (26)	0.121
Backward visual span	+13 (38)	+9 (27)	0.625

The change in each parameter was computed using the following formula: Δ(parameter) = (post-CSF tap test value – pre-CSF tap test value)/pre-CSF tap test value. Each delta is expressed in percentage (standard deviation), a positive sign indicating a better performance and a negative sign indicating a worsened performance. Comparisons are based on independent samples t-tests. Significant differences (p < 0.05, two-tailed) are highlighted in bold. ^*The Color Trails Test index is calculated as follows: (Part 2 – Part 1)/Part 1. ^$The Stroop index is calculated as follows: (Color condition)/(Dot condition). ^#The stride time variability is calculated as the coefficient of variation (CV) of stride time, according to the formula: stride time CV = 100*[(standard deviation of stride time)/(mean stride time)].

For cognition, both groups presented a similar cognitive improvement after the CSF tap test for executive functions and attention. There was a tendency to increase phonemic fluency performance in the amnestic iNPH relative to the non-amnestic iNPH (p = 0.084).

The long-term reversibility of symptoms was assessed qualitatively by the neurosurgeon at the one-year post-surgical visit. Overall, 4 (66%) amnestic iNPH and 21 (78%) non-amnestic iNPH improved; both groups had similar outcomes (p = 0.609).

DISCUSSION

Among our iNPH sample, a majority of patients present with a non-amnestic cognitive deficit. This study demonstrates that a storage deficit in verbal episodic memory deficit exists independently of AD pathology in patients with iNPH. This amnestic deficit is associated with education and global cognition. The short-term reversibility of gait and cognition is also similar between amnestic and non-amnestic iNPH patients.

Most of our iNPH patients did not exhibit an amnestic syndrome of the hippocampal type. The learning curves demonstrate that the non-amnestic iNPH patients exhibit a preserved encoding function compared to amnestic iNPH patients, as shown by the increasing slope between the first and the second trials of the free recall subtest. Total recall trials were similar between groups, suggesting a similar sensitivity to cueing. These findings suggest a predominantly retrieval-related deficit in the non-amnestic group, consistent with the classical dysexecutive syndrome of iNPH previously described.5,28,29 , 5,28,29 By contrast, thirty-eight patients presented with an amnestic syndrome of the hippocampal type similar to typical AD, with impaired encoding and retrieval functions, as shown by the learning curves. Recent research highlights that impairment in the FCSRT free recall scores reliably predicts incipient typical AD dementia. 30 To our knowledge, this is the first study to examine the presence of the amnestic syndrome of the hippocampal type—the core clinical marker for typical AD in the International Working Group (IWG-2) criteria 31 —in iNPH patients. However, recent research highlights that the amnestic syndrome of the hippocampal type is not specific to AD and should be considered a common presentation in various non-AD conditions. 8 Some non-AD patients also presented with an amnestic syndrome of the hippocampal type7,32,33 , 7,32,33 that was not associated with CSF AD biomarkers. 7 Therefore, this study confirms the lack of specificity of the amnestic profile for AD diagnosis in iNPH patients.

Contrary to our hypothesis, CSF AD biomarkers were similar between amnestic and non-amnestic iNPH. The CSF biomarkers for AD were not associated with memory performances in iNPH. Previous studies analyzing brain samples obtained during surgical shunting procedures reported a total prevalence of 25–40% of concomitant AD pathology in iNPH.10,34,35 , 10,34,35 In two postmortem studies, this prevalence reached 80%.36,37, 36,37 In AD, it is now well established that neuropathological changes occur many years before the clinical expression of the disease, moving along a pathological-clinical continuum38,39, 38,39. Therefore, in memory clinics, the amnestic syndrome of the hippocampal type strongly predicts the presence of CSF AD pathology, 40 identifying both prodromal and typical AD.6,7, 6,7 In our study, AD signature in the CSF did not contribute to the amnestic syndrome of the hippocampal type in iNPH patients, suggesting an AD-distinct pathogenesis accounting for memory decline.

Global white matter lesions—a marker for potential cerebrovascular comorbidity and a frequent finding in iNPH—did not contribute to memory decline. The burden’s regional distribution was similar between amnestic and non-amnestic iNPH, even in the temporal lobes. Though the exact mechanism is unknown, white matter lesions could impair the connectivity of memory pathways,41,42, 41,42 e.g., within the entorhinal-hippocampal circuit. In AD, FCSRT total recall scores correlated with white matter lesions. 43 However, this was not the case in our iNPH patients, which suggested distinct mechanisms accounting for memory impairment.

The mechanisms underlying a storage deficit in verbal episodic memory in iNPH have not yet been elucidated. Cognitive dysfunction in iNPH could be related to abnormal CSF dynamics, dysfunction of frontostriatal and entorhinal-hippocampal circuits, abnormal neuromodulation, and deep brain structures’ volumes. 44 Previous studies focused on connectivity impairment caused by tangential shear forces on the periventricular projection fibers, disrupting critical pathways, such as the Papez circuit45 –47 or the default mode network. The latter was associated with worse memory performances, particularly in the immediate recall scores. 48 In a previous study, we showed that neither CSF AD pathology nor verbal episodic memory disturbances drive the observed functional network changes in iNPH. 49 Studies focusing on neurotransmitter changes showed that a reduced cholinergic transmission linked with amyloid depositions was correlated with poorer verbal memory performances in iNPH patients.50,51, 50,51 Finally, we hypothesize that hippocampal atrophy—not assessed in the present work—could account for worse memory performances in iNPH since it was significantly smaller than in healthy controls but still more significant than in AD patients.52,53, 52,53 Poor verbal episodic memory performances were linked with wider temporal horns, suggesting a compression effect accounting for smaller hippocampal size in iNPH. 54

Since neither AD pathology nor the white matter changes account significantly for the amnestic syndrome of the hippocampal type in iNPH, what factors could explain worse memory performances in our sample? First, the regression analyses showed that education and global cognition accounted for 8.5% and 18.7% of memory performances, respectively. Since education represents an appropriate proxy for cognitive reserve 55 and MMSE performance was associated with the education level, 56 a lower cognitive reserve may contribute to worse performances in verbal episodic memory tests in iNPH. Second, between-group comparisons showed that amnestic iNPH patients performed worse on executive functions and attention domains at baseline with preserved sensitivity to semantic cueing, as demonstrated by the FCSRT learning curves. Furthermore, amnestic iNPH patients walked slower than non-amnestic patients at baseline but improved at the same rate. Gait control relies specifically on executive functions and attention,57,58, 57,58 but it is also linked with broader cognitive domains 59 and cognitive reserve.60,61, 60,61 Finally, increased disease duration was surprisingly associated with better memory performances, but this effect was not significant when adjusting for the other covariates. Since gait impairment represented the entry point in the disease process in this study, some patients may have presented with cognitive symptoms at first, but this effect was not measured. Overall, these findings suggest a different memory profile in iNPH independent from AD, with some patients expressing a clinical memory profile as severe as classical amnestic AD: amnestic iNPH patients also present with lower baseline executive and attentional performances, a poorer cognitive reserve, and slower gait.

Despite worse baseline cognition and gait in the amnestic group, amnestic and non-amnestic iNPH patients improved similarly in the short term after the CSF tap test. These results contradict the initial hypothesis that an amnestic syndrome of the hippocampal type—as a clinical hint of a comorbid AD pathology40,62, 40,62—would prevent an improvement in motor and cognitive performances. Previous literature yielded contradictory results regarding CSF derivation response in the context of AD pathology. Some studies showed a negative impact of amyloidosis and tauopathy,11,53,63–66 , 11,53,63–66 whereas others suggested iNPH patients could benefit from CSF shunting irrespective of AD pathology.67 –69 Given that uncertainty and the present demonstration, clinicians should not take the shortcut that the amnestic syndrome of the hippocampal type indicates an AD comorbidity. Therefore, its presence should not prevent surgical treatment in those patients, mainly since the surgery should be performed as soon as possible according to the current recommendations.70 –72

Our study has several limitations. First, we could not quantitatively assess the long-term reversibility of gait and particularly cognition after CSF diversion with the same extensive study protocol we applied. Thus, our results only apply in the short term. Further studies should evaluate whether these findings are confirmed long-term after shunt surgery. Second, the FCRST threshold values we used to distinguish amnestic and non-amnestic iNPH are based on the specific setting of a memory clinic, which Sarazin et al. acknowledged in their princeps paper 6 —and not in an iNPH clinic like ours. Nevertheless, the regression analyses are independent of the threshold values since we used the FCSRT sum of total recalls ranging from 0 to 48. Third, we omitted p-tau—a specific AD biomarker—in the regression analyses for statistical reasons. However, in our sample, only one patient matched the biological definition of AD according to NIA-AA, 13 which is negligible. Fourth, we examined only the verbal component of episodic memory, which relies on different pathways than visual episodic memory. Fifth, we may have underestimated the number of patients who presented with a severe cognitive profile since 90 patients who could not perform the FCSRT were not included in this study. This could prevent the generalization of our results and should be limited to the specific setting of our cohort-based study on relatively mildly cognitively impaired. Sixth, we did not assess the hippocampal volume nor the presence of other proteinopathies to test for other putative pathogenic mechanisms. However, this was beyond the scope of the present study and shall be assessed in future works. Seventh, we did not use the new operationalized definitions for the ARWMC scale in patients who underwent a CT scan to improve interrater reliability for this imaging modality. 73 However, only a relatively small number of CT scans were performed in our sample, which may limit the presence of significant bias in evaluating ARWMC in those cases. Finally, the neuropathologic confirmation of amyloidopathy and tauopathy was not available.

Conclusion

Our study showed that AD pathology is not the only cause of hippocampal memory deficit in iNPH. Still, the mechanisms implicated in genuine memory decline in iNPH remain to be elucidated. To our knowledge, this is the first study that analyzed the memory profile in iNPH in conjunction with AD biomarkers. Furthermore, patients with amnestic syndrome similarly improved after the CSF tap test compared to patients with non-amnestic syndrome. Patients shall not be precluded from invasive intervention if clinically diagnosed with iNPH, even with an amnestic syndrome of the hippocampal type. The amnestic profile in iNPH may be related to the cognitive reserve.

AUTHOR CONTRIBUTIONS

Alma Lingenberg (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Visualization; Writing – original draft); François R. Herrmann (Data curation; Formal analysis; Methodology; Software; Supervision; Writing – review & editing); Stéphane Armand (Data curation; Investigation; Resources; Writing – review & editing); Julie Péron (Data curation; Investigation; Resources; Writing – review & editing); Frédéric Assal (Conceptualization; Writing – review & editing); Gilles Allali (Conceptualization; Funding acquisition; Investigation; Methodology; Project administration; Resources; Supervision; Validation; Writing – review & editing).

Footnotes

ACKNOWLEDGMENTS

The authors have no acknowledgments to report.

FUNDING

This project was supported by the Swiss National Science Foundation (project number 320030_173153). Alma Lingenberg was supported by a grant from the Henriette-Meyer Foundation. Gilles Allali was supported by the Baasch-Medicus Foundation.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

DATA AVAILABILITY

The datasets used and analyzed during the current study are available from the corresponding author upon reasonable request.

The supplementary material is available in the electronic version of this article: .

References

Picascia

, Zangaglia

, Bernini

, et al. A review of cognitive impairment and differential diagnosis in idiopathic normal pressure hydrocephalus. Funct Neurol 2015; 30: 217–228.

Peterson

, Housden

, Killikelly

, et al. Apathy, ventriculomegaly and neurocognitive improvement following shunt surgery in normal pressure hydrocephalus. Br J Neurosurg 2016; 30: 38–42.

Allali

, Laidet

, Armand

, et al. Apathy and higher level of gait control in normal pressure hydrocephalus. Int J Psychophysiol 2017; 119: 127–131.

Nerg

, Junkkari

, Hallikainen

, et al. The CERAD neuropsychological battery in patients with idiopathic normal pressure hydrocephalus compared with normal population and patients with mild Alzheimer’s disease. J Alzheimers Dis 2021; 81: 1117–1130.

Saito

, Nishio

, Kanno

, et al. Cognitive profile of idiopathic normal pressure hydrocephalus. Dement Geriatr Cogn Disord Extra 2011; 1: 202–211.

Sarazin

, Berr

, De Rotrou

, et al. Amnestic syndrome of the medial temporal type identifies prodromal AD: a longitudinal study. Neurology 2007; 69: 1859–1867.

Teichmann

, Epelbaum

, Samri

, et al. Free and Cued Selective Reminding Test - accuracy for the differential diagnosis of Alzheimer’s and neurodegenerative diseases: A large-scale biomarker-characterized monocenter cohort study (ClinAD). Alzheimers Dement 2017; 13: 913–923.

Bertoux

, Cassagnaud

, Lebouvier

, et al. Does amnesia specifically predict Alzheimer’s pathology? A neuropathological study. Neurobiol Aging 2020; 95: 123–130.

Grober

and Buschke

. Genuine memory deficits in dementia. Dev Neuropsychol 1987; 3: 13–36.

10.

Malm

, Graff-Radford

, Ishikawa

, et al. Influence of comorbidities in idiopathic normal pressure hydrocephalus - research and clinical care. A report of the ISHCSF task force on comorbidities in INPH. Fluids Barriers CNS 2013; 10: 22.

11.

Allali

, Laidet

, Armand

, et al. Brain comorbidities in normal pressure hydrocephalus. Eur J Neurol 2018; 25: 542–548.

12.

Allali

, Laidet

, Armand

, et al. A combined cognitive and gait quantification to identify normal pressure hydrocephalus from its mimics: The Geneva’s protocol. Clin Neurol Neurosurg 2017; 160: 5–11.

13.

Jack

Jr. , Bennett

, Blennow

, et al. NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 2018; 14: 535–562.

14.

Seeburger

, Holder

, Combrinck

, et al. Cerebrospinal fluid biomarkers distinguish postmortem-confirmed Alzheimer’s disease from other dementias and healthy controls in the OPTIMA Cohort. J Alzheimers Dis 2015; 44: 525–539.

15.

Relkin

, Marmarou

, Klinge

, et al. Diagnosing idiopathic normal-pressure hydrocephalus. Neurosurgery 2005; 57: S4–S16; discussion ii-v.

16.

Folstein

, Folstein

and McHugh

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

17.

D’Elia

, Satz

, Uchiyama

, et al. Color Trails Test Professional Manual. Psychological Assessment Resources.

18.

Cardebat

, Doyon

, Puel

, et al. Formal and semantic lexical evocation in normal subjects. Performance and dynamics of production as a function of sex, age and educational level. Acta Neurol Belg 1990; 90: 207–217.

19.

Perret

. The left frontal lobe of man and the suppression of habitual responses in verbal categorical behaviour. Neuropsychologia 1974; 12: 323–330.

20.

Wechsler

. Wechsler Adult Intelligence Scale–Third Edition (WAIS-III). Psychological Corporation, 1997.

21.

Wechsler

. Wechsler Memory Scale–Third Edition. Psychological Corporation, 1997.

22.

Van der Linden

, Adam

and Agniel

. L’évaluation des troubles de la mémoire: présentation de quatre tests de mémoire épisodique (avec leur étalonnage). Marseille, France: Solal, 2004.

23.

Starkstein

, Mayberg

, Preziosi

, et al. Reliability, validity, and clinical correlates of apathy in Parkinson’s disease. J Neuropsychiatry Clin Neurosci 1992; 4: 134–139.

24.

Beauchet

, Allali

, Sekhon

, et al. Guidelines for assessment of gait and reference values for spatiotemporal gait parameters in older adults: The Biomathics and Canadian Gait Consortiums Initiative. Front Hum Neurosci 2017; 11: 353.

25.

Holtzer

, Verghese

, Xue

, et al. Cognitive processes related to gait velocity: results from the Einstein Aging Study. Neuropsychology 2006; 20: 215–223.

26.

Hughes

, Daniel

, Kilford

, et al. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992; 55: 181–184.

27.

Wahlund

, Barkhof

, Fazekas

, et al. A new rating scale for age-related white matter changes applicable to MRI and CT. Stroke 2001; 32: 1318–1322.

28.

Iddon

, Pickard

, Cross

, et al. Specific patterns of cognitive impairment in patients with idiopathic normal pressure hydrocephalus and Alzheimer’s disease: a pilot study. J Neurol Neurosurg Psychiatry 1999; 67: 723–732.

29.

Walchenbach

, Geiger

, Thomeer

RTWM

, et al. The value of temporary external lumbar CSF drainage in predicting the outcome of shunting on normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 2002; 72: 503–506.

30.

Grober

, Veroff

and Lipton

. Temporal unfolding of declining episodic memory on the Free and Cued Selective Reminding Test in the predementia phase of Alzheimer’s disease: Implications for clinical trials. Alzheimers Dement (Amst) 2018; 10: 161–171.

31.

Dubois

, Feldman

, Jacova

, et al. Advancing research diagnostic criteria for Alzheimer’s disease: the IWG-2 criteria. Lancet Neurol 2014; 13: 614–629.

32.

Bertoux

, de Souza

, Corlier

, et al. Two distinct amnesic profiles in behavioral variant frontotemporal dementia. Biol Psychiatry 2014; 75: 582–588.

33.

Lagarde

, Olivieri

, Tonietto

, et al. Distinct amyloid and tau PET signatures are associated with diverging clinical and imaging trajectories in patients with amnestic syndrome of the hippocampal type. Transl Psychiatry 2021; 11: 498.

34.

Leinonen

, Koivisto

, Alafuzoff

, et al. Cortical brain biopsy in long-term prognostication of 468 patients with possible normal pressure hydrocephalus. Neurodegener Dis 2012; 10: 166–169.

35.

Elobeid

, Laurell

, Cesarini

, et al. Correlations between mini-mental state examination score, cerebrospinal fluid biomarkers, and pathology observed in brain biopsies of patients with normal-pressure hydrocephalus. J Neuropathol Exp Neurol 2015; 74: 470–479.

36.

Cabral

, Beach

, Vedders

, et al. Frequency of Alzheimer’s disease pathology at autopsy in patients with clinical normal pressure hydrocephalus. Alzheimers Dement 2011; 7: 509–513.

37.

Leinonen

, Koivisto

, Savolainen

, et al. Post-mortem findings in 10 patients with presumed normal-pressure hydrocephalus and review of the literature. Neuropathol Appl Neurobiol 2012; 38: 72–86.

38.

Jack

, Knopman

, Jagust

, et al. Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol 2010; 9: 119–128.

39.

Aisen

, Cummings

, Jack

, et al. On the path to understanding the Alzheimer’s disease continuum. Alzheimers Res Ther 2017; 9: 60.

40.

Wagner

, Wolf

, Reischies

, et al. Biomarker validation of a cued recall memory deficit in prodromal Alzheimer disease. Neurology 2012; 78: 379–386.

41.

Pantoni

, Poggesi

and Inzitari

. The relation between white-matter lesions and cognition. Curr Opin Neurol 2007; 20: 390–397.

42.

Black

, Gao

and Bilbao

. Understanding white matter disease: imaging-pathological correlations in vascular cognitive impairment. Stroke 2009; 40: S48–S52.

43.

Capizzano

, Acion

, Bekinschtein

, et al. White matter hyperintensities are significantly associated with cortical atrophy in Alzheimer’s disease. J Neurol Neurosurg Psychiatry 2004; 75: 822–827.

44.

Xiao

, Hu

, Ding

, et al. Cognitive impairment in idiopathic normal pressure hydrocephalus. Neurosci Bull 2022; 38: 1085–1096.

45.

Bonelli

and Cummings

. Frontal-subcortical circuitry and behavior. Dialogues Clin Neurosci 2007; 9: 141–151.

46.

Dombrowski

, Deshpande

, Dingwall

, et al. Chronic hydrocephalus-induced hypoxia: increased expression of VEGFR-2+ and blood vessel density in hippocampus. Neuroscience 2008; 152: 346–359.

47.

Chen

Y-C

, Chiang

S-W

, Chi

C-H

, et al. Early idiopathic normal pressure hydrocephalus patients with neuropsychological impairment are associated with increased fractional anisotropy in the anterior thalamic nucleus. Medicine (Baltimore) 2016; 95: e3636.

48.

Kanno

, Ogawa

K-I

, Kikuchi

, et al. Reduced default mode network connectivity relative to white matter integrity is associated with poor cognitive outcomes in patients with idiopathic normal pressure hydrocephalus. BMC Neurol 2021; 21: 353.

49.

Griffa

, Bommarito

, Assal

, et al. Dynamic functional networks in idiopathic normal pressure hydrocephalus: Alterations and reversibility by CSF tap test. Hum Brain Mapp 2021; 42: 1485–1502.

50.

Nardone

, Golaszewski

, Schwenker

, et al. Cholinergic transmission is impaired in patients with idiopathic normal-pressure hydrocephalus: a TMS study. J Neural Transm Vienna Austria 2019; 126: 1073–1080.

51.

Kondziella

, Sonnewald

, Tullberg

, et al. Brain metabolism in adult chronic hydrocephalus. J Neurochem 2008; 106: 1515–1524.

52.

Ishii

, Kawaguchi

, Shimada

, et al. Voxel-based analysis of gray matter and CSF space in idiopathic normal pressure hydrocephalus. Dement Geriatr Cogn Disord 2008; 25: 329–335.

53.

Savolainen

, Laakso

, Paljärvi

, et al. MR imaging of the hippocampus in normal pressure hydrocephalus: correlations with cortical Alzheimer’s disease confirmed by pathologic analysis. AJNR Am J Neuroradiol 2000; 21: 409–414.

54.

Lilja-Lund

, Kockum

, Hellström

, et al. Wide temporal horns are associated with cognitive dysfunction, as well as impaired gait and incontinence. Sci Rep 2020; 10: 18203.

55.

Stern

. What is cognitive reserve? Theory and research application of the reserve concept. J Int Neuropsychol Soc 2002; 8: 448–460.

56.

Crum

, Anthony

, Bassett

, et al. Population-based norms for the Mini-Mental State Examination by age and educational level. JAMA 1993; 269: 2386–2391.

57.

Yogev-Seligmann

, Hausdorff

and Giladi

. The role of executive function and attention in gait. Mov Disord 2008; 23: 329–342; quiz 472.

58.

Montero-Odasso

, Verghese

, Beauchet

, et al. Gait and cognition: a complementary approach to understanding brain function and the risk of falling. J Am Geriatr Soc 2012; 60: 2127–2136.

59.

Savica

, Wennberg

AMV

, Hagen

, et al. Comparison of gait parameters for predicting cognitive decline: The Mayo Clinic Study of Aging. J Alzheimers Dis 2017; 55: 559–567.

60.

Holtzer

, Wang

, Lipton

, et al. The protective effects of executive functions and episodic memory on gait speed decline in aging defined in the context of cognitive reserve. J Am Geriatr Soc 2012; 60: 2093–2098.

61.

Holtzer

, Zhu

, Rosso

, et al. Cognitive reserve and risk of mobility impairment in older adults. J Am Geriatr Soc 2022; 70: 3096–3104.

62.

Dubois

, Feldman

, Jacova

, et al. Research criteria for the diagnosis of Alzheimer’s disease: revising the NINCDS-ADRDA criteria. Lancet Neurol 2007; 6: 734–746.

63.

Hamilton

, Patel

, Lee

, et al. Lack of shunt response in suspected idiopathic normal pressure hydrocephalus with Alzheimer disease pathology. Ann Neurol 2010; 68: 535–540.

64.

Patel

, Lee

, Xie

, et al. Phosphorylated tau/amyloid beta 1-42 ratio in ventricular cerebrospinal fluid reflects outcome in idiopathic normal pressure hydrocephalus. Fluids Barriers CNS 2012; 9: 7.

65.

Lim

, Choi

, Park

, et al. Evaluation of coexistence of Alzheimer’s disease in idiopathic normal pressure hydrocephalus using ELISA analyses for CSF biomarkers. BMC Neurol 2014; 14: 66.

66.

Hiraoka

, Narita

, Kikuchi

, et al. Amyloid deposits and response to shunt surgery in idiopathic normal-pressure hydrocephalus. J Neurol Sci 2015; 356: 124–128.

67.

Bech

, Waldemar

, Gjerris

, et al. Shunting effects in patients with idiopathic normal pressure hydrocephalus; correlation with cerebral and leptomeningeal biopsy findings. Acta Neurochir (Wien) 1999; 141: 633–639.

68.

Golomb

, Wisoff

, Miller

, et al. Alzheimer’s disease comorbidity in normal pressure hydrocephalus: prevalence and shunt response. J Neurol Neurosurg Psychiatry 2000; 68: 778–781.

69.

Bech-Azeddine

, Høgh

, Juhler

, et al. Idiopathic normal-pressure hydrocephalus: clinical comorbidity correlated with cerebral biopsy findings and outcome of cerebrospinal fluid shunting. J Neurol Neurosurg Psychiatry 2007; 78: 157–161.

70.

Fasano

, Espay

, Tang-Wai

, et al. Gaps, controversies, and proposed roadmap for research in normal pressure hydrocephalus. Mov Disord 2020; 35: 1945–1954.

71.

Tinelli

, Guldemond

and Kehler

. Idiopathic normal-pressure hydrocephalus: the cost-effectiveness of delivering timely and adequate treatment in Germany. Eur J Neurol 2021; 28: 681–690.

72.

Nakajima

, Yamada

, Miyajima

, et al. Guidelines for management of idiopathic normal pressure hydrocephalus (Third Edition): Endorsed by the japanese society of normal pressure hydrocephalus. Neurol Med Chir (Tokyo) 2021; 61: 63–97.

73.

Xiong

, Yang

, Wong

, et al. Operational definitions improve reliability of the age-related white matter changes scale. Eur J Neurol 2011; 18: 744–749.