Psychological Medicine, Page 1 of 9.

© Cambridge University Press 2014 OR I G I N A L A R T I C L E

doi:10.1017/S0033291714002153

Trajectories of cognitive decline in different types

of dementia

L. L. Smits1*, A. C. van Harten1, Y. A. L. Pijnenburg1, E. L. G. E. Koedam1, F. H. Bouwman1,

N. Sistermans1, I. E. W. Reuling2, N. D. Prins1, A. W. Lemstra1, P. Scheltens1
and W. M. van der Flier1,3
1
Department of Neurology and Alzheimer Centre, VU University Medical Centre, Amsterdam, The Netherlands
2
Department of Medical Psychology and Alzheimer Centre, VU University Medical Centre, Amsterdam, The Netherlands
3
Department of Epidemiology and Biostatistics and Alzheimer Centre, VU University Medical Centre, Amsterdam, The Netherlands

Background. To investigate trajectories of cognitive decline in patients with different types of dementia compared to
controls in a longitudinal study.

Method. In 199 patients with Alzheimer’s disease (AD), 10 with vascular dementia (VaD), 26 with dementia with Lewy
bodies (DLB), 20 with behavioural variant frontotemporal dementia (bvFTD), 15 with language variant frontotemporal
dementia (lvFTD) and 112 controls we assessed five cognitive domains: memory, language, attention, executive and
visuospatial functioning, and global cognition (Mini-Mental State Examination, MMSE). All subjects had at least two
neuropsychological assessments (median 2, range 2–7). Neuropsychological data were standardized into z scores
using baseline performance of controls as reference. Linear mixed models (LMMs) were used to estimate baseline cog-
nitive functioning and cognitive decline over time for each group, adjusted for age, gender and education.

Results. At baseline, patients with dementia performed worse than controls in all cognitive domains (p < 0.05) except
visuospatial functioning, which was only impaired in patients with AD and DLB (p < 0.001). During follow-up, patients
with AD declined in all cognitive domains (p < 0.001). DLB showed decline in every cognitive domain except language
and global cognition. bvFTD showed rapid decline in memory, language, attention and executive functioning (all
p < 0.01) whereas visuospatial functioning remained fairly stable. lvFTD declined mostly in attention and executive func-
tioning (p < 0.01). VaD showed decline in attention and executive functioning.

Conclusions. We show cognitive trajectories of different types of dementia. These estimations of natural disease course
have important value for the design of clinical trials as neuropsychological measures are increasingly being used as out-
come measures.

Received 24 February 2014; Revised 1 August 2014; Accepted 4 August 2014

Key words: Alzheimer’s disease, behavioural variant frontotemporal dementia, cognitive decline, cognitive
neuropsychology, dementia with Lewy bodies, language variant frontotemporal dementia, longitudinal design,
vascular dementia.

Introduction Most studies are not longitudinal, which limits in-

terpretation of the results. Studies with a longitudinal
The most common causes of dementia are character-
set-up are frequently limited to the Mini-Mental State
ized by specific profiles of cognitive impairment.
Examination (MMSE) as a cognitive measure (Hanyu
Alzheimer’s disease (AD) is characterized by promi-
et al. 2009; Tan et al. 2013). As the MMSE is a global
nent episodic memory impairment whereas in other
screening test, no domain-specific conclusions can be
types of dementia, other domains are more primarily
drawn. Furthermore, the single focus of most studies
affected, such as executive functioning in vascular
is AD instead of the various types of dementia, making
dementia (VaD) (Karantzoulis & Galvin, 2011).
it difficult to compare these different types. Two pre-
Studies of trajectories of cognitive decline in differ-
vious longitudinal studies investigated decline on sev-
ent types of dementia in a single cohort are scarce.
eral neuropsychological tests in various types of
dementia (Rogers et al. 2006; Libon et al. 2009). One
study focused on semantic memory tests and used
* Address for correspondence: L. L. Smits, M.Sc., Department of
Neurology and Alzheimer Centre, VU University Medical Centre,
patients with semantic dementia as the reference
PO Box 7057, 1007 MB Amsterdam, The Netherlands. group (Rogers et al. 2006). They found that all types
(Email: [email protected]) of dementia performed worse than controls and AD
2 L. L. Smits et al.

showed a similar but milder pattern of impairment of (1998) or Rascovsky et al. (2011), and for language vari-
semantic memory to semantic dementia. The study ant frontotemporal dementia [lvFTD; n = 15, including
by Libon et al. (2009) focused on three domains (execu- non-fluent progressive aphasia (n = 7) and semantic de-
tive functioning, language and visuoconstruction) in mentia (n = 8)], the criteria of Neary et al. (1998) or
four subtypes of frontotemporal lobar degeneration Gorno-Tempini et al. (2011). In addition, we included
(FTLD), and concluded that diverse neuropsychologi- 112 patients with subjective complaints who served as
cal patterns of FTLD subtypes continue to exist. controls. Patients were considered to have subjective
In the current study we investigated decline in glo- cognitive complaints when they had normal laboratory
bal cognition and five cognitive domains for five investigations and did not have other known causes of
types of dementia compared to controls. We aimed cognitive complaints. Their performances on neuropsy-
to provide a better insight into the trajectories of cogni- chological tests, corrected for age, education or sex, were
tive decline over time in different types of dementia. normal. Furthermore, patients with subjective com-
We expected that each type of dementia would show plaints were only included when they did not progress
decline in those cognitive domains that are associated to mild cognitive impairment (MCI) or dementia within
with the brain areas most prominently involved in the study period. Pharmacological treatment for de-
each type of dementia. mentia was registered: 11 patients with AD used galan-
tamine and 16 patients were treated with rivastigmine
(12 AD, three DLB and one bvFTD). Level of education
Method was classified according to the system of Verhage (1964)
ranging from 1 to 7 (low to highly educated). The local
Subjects
medical ethics committee approved the study and all
We included 270 patients with a diagnosis of dementia patients gave written informed consent for their clinical
and at least two visits including neuropsycho- data to be used for research purposes.
logical evaluations (baseline and follow-up) from the
Amsterdam Dementia Cohort at the Alzheimer
Neuropsychological assessment
Centre of the VU University Medical Centre (VUmc)
Amsterdam, between January 2004 and December Cognitive functions were assessed with a standardized
2011 (van der Flier et al. 2014). Patients with available test battery. We used the MMSE as a measure for global
follow-up were younger (65 ± 8 v. 69 ± 9 years), less cognitive decline (Folstein et al. 1975). For memory, we
often female (41% v. 51%), had higher education used the Visual Association Test (VAT) and total im-
(5 ± 1 v. 4.7 ± 1) and higher MMSE scores (22 ± 4 v. mediate recall and delayed recall of the Dutch version
19 ± 6) than the 384 persons who presented during of the Rey Auditory Verbal Learning Test (RAVLT;
the same time period but received no follow-up (all Rey, 1964; Saan & Deelman, 1986; Lindeboom et al.
p < 0.05). 2002). To examine language, we used the VAT naming,
All patients underwent a standardized 1-day assess- category fluency (animals), the Dutch version of the
ment that included medical history, informant-based Controlled Oral Word Association Test (COWAT) (letter
history, physical and neurological examination includ- fluency), and comparative questions and the naming
ing the Clinical Dementia Rating (CDR) scale, neuropsy- condition of the Arizona Battery for Communication
chological assessment and the Geriatric Depression Disorders (ABCD; Luteijn & Ploeg, 1982; Bayles &
Scale (GDS), laboratory tests, magnetic resonance ima- Tomoeda, 1993; Lindeboom et al. 2002; Schmand et al.
ging of the brain and electroencephalography. The dur- 2008). For attention we used the Trail Making Test
ation of the cognitive complaints as reported by the (TMT) Part A and the forward condition of Digit Span
patient and/or caregiver was recorded to estimate the (extended version) (Reitan, 1958; Lindeboom & Matto,
disease duration at the time of diagnosis. Diagnoses 1994). We used TMT Part B, the backwards condition
were made in a multidisciplinary consensus meeting of Digit Span (extended version) and the Frontal
using international diagnostic consensus criteria. For a Assessment Battery (FAB) to examine executive func-
diagnosis of probable AD (n = 199), patients had to tioning (Reitan, 1958; Lindeboom & Matto, 1994;
fulfil the diagnostic criteria of McKhann et al. (1984, Dubois et al. 2000). We used three subtests of the
2011), for vascular dementia (VaD; n = 10) the National Visual Object and Space Perception (VOSP) battery to
Institute of Neurological Disorders and Stroke assess visuospatial functioning, namely (i) incomplete
Association (NINDS) Internationale pour la Recherche letters, (ii) dot counting and (iii) number location
et l’Enseignement en Neurosciences (AIREN) criteria (Warrington & James, 1991). TMT A and B scores were
(Román et al. 1993), for DLB (n = 26) the criteria of log-transformed because they were not normally dis-
McKeith et al. (2005), for behavioural variant frontotem- tributed. TMT A and B scores were inverted by comput-
poral dementia (bvFTD; n = 20) the criteria of Neary et al. ing −1 × score because higher scores imply a worse
 Trajectories of cognitive decline in dementia 3

performance. There was variability in the number of performance whereas the interaction between diag-
completed neuropsychological tests. In total, the num- nosis and time represents the association between di-
ber of completed tests on baseline and follow-up ranged agnosis and cognitive performance over time.
from 931 (Digit Span forward) to 602 (comparative Controls were used as a reference group. Outcome
questions). measures were compound z scores and MMSE scores.
All analyses were adjusted for age, gender and edu-
Follow-up cation. Data are presented as unadjusted β ± S.E. with
p values of the adjusted models. We repeated the
All subjects in this cohort (n = 382) underwent follow-
analyses without the patients who were on pharmaco-
up, including physical and neurological examination,
logical treatment for dementia. For main effects, the
and a repeated neuropsychological evaluation. For
significance level was set at p < 0.05 and for interac-
the total sample, the median number of neuropsycho-
tions at p < 0.10.
logical assessments was 2 (range 2–7) and the mean
duration of follow-up was 1.5 ± 0.8 years.

Statistical analysis Results

PASW Statistics 19.0 for Mac (SPSS Inc., USA) was Table 1 summarizes the baseline demographics of the
used. For baseline demographics and raw neuropsy- diagnostic groups. In general, patients were older
chological data, χ2 tests, an independent-samples (65 ± 8 v. 61 ± 8 years) and less educated (5.0 ± 1 v.
t test and univariate ANOVAs were performed when 5.5 ± 1) than controls (both p < 0.001). On the MMSE,
appropriate. ANOVAs were conducted with diagnosis patients with dementia scored lower than controls
as the between-subject factor and neuropsychological (all p < 0.05). Patients with AD had the lowest MMSE
test as the dependent variable. Sex, age and education score (22 ± 4). Table 2 lists the raw baseline neuropsy-
were entered as covariates. chological test performance of controls and patients
To obtain an unbiased estimation of the cognitive do- with different types of dementia.
main scores, we imputed missing neuropsychological We used LMMs to compare the trajectories of cogni-
test scores by multiple imputation of individual test tive decline as measured by a comprehensive neurop-
scores in PASW. The method we used was predictive sychological test battery between patients with
mean matching because of the non-Gaussian distri- different types of dementia and controls. The estimated
bution of some of the tests. Predictors for imputation effects are shown in Table 3. In general, patients with a
were age (at time of neuropsychological assessment), diagnosis of dementia performed worse than controls
gender, education, diagnosis, CDR score, GDS score at baseline on all cognitive domains except for visuos-
and all available neuropsychological tests. Fifteen patial functioning, which was only impaired in AD
imputed data sets were created. Neuropsychological and DLB.
baseline data for controls were standardized into z Figure 1 visualizes the trajectories of decline for each
scores and, based on these z scores, we calculated com- of the cognitive domains. Regarding global cognition,
pound z scores for memory, language, attention, execu- we found that, over time, patients with AD and
tive functioning and visuospatial functioning for each bvFTD declined faster than controls whereas patients
imputed dataset. Next, we computed z scores relative with VaD, DLB and lvFTD did not. On memory,
to the baseline z scores of controls for follow-up of con- patients with AD, DLB and bvFTD declined faster
trols and for baseline and follow-up for all dementia over time than controls whereas patients with VaD
groups. We report pooled statistics over 15 imputed and lvFTD did not. For language, patients with AD,
data sets. bvFTD and lvFTD declined faster over time than con-
Linear mixed models (LMMs) with an unstructured trols whereas patients with DLB and VaD did not.
covariance pattern were used to assess associations We found a significant decline over time for attention
between diagnosis and baseline cognition and cogni- in all patients with dementia. Regarding executive
tive performance over time. The LMM has increased functioning, patients with dementia declined over
statistical power as it accounts for within-person cor- time compared to controls. On visuospatial function-
relation over time, allows inter-individual differences ing, only patients with AD and DLB declined over
in number of assessments and differences in time be- time. When we repeated the analyses without the
tween assessments. The model included terms for di- patients using pharmacological treatment for de-
agnosis, time, and the interaction between diagnosis mentia, the results remained essentially unchanged.
and time. Random effects were subject ID and time. Table 4 summarizes the estimated annual cognitive
β ± S.E. for diagnosis represents the association be- decline for the different dementia groups compared
tween the diagnosis and baseline cognitive to controls, in a schematic overview.
4 L. L. Smits et al.

Table 1. Baseline demographics of diagnostic groups

Controls AD VaD DLB bvFTD lvFTD

n 112 199 10 26 20 15
Age (years) 61 ± 8 65 ± 8a 67 ± 5a 66 ± 9a 63 ± 8 63 ± 8
Sex, female 63 (56) 98 (49) 4 (40) 0 (0)abcef 6 (30)ab 3 (20)ab
Educationg 5.5 ± 1.1 4.9 ± 1.2a 4.5 ± 2.1a 5.2 ± 1.0 5.0 ± 1.3 4.8 ± 1.0a
Complaintsh (years) 3.9 ± 4.3 3.1 ± 1.8a 3.7 ± 1.9 3.4 ± 1.7 3.9 ± 3.7 3.1 ± 1.4
No. of NPEs 2 (2–7) 2 (2–4) 2 (2–3) 3 (2–5) 3 (2–5) 3 (2–4)
Follow-up duration (years) 1.6 ± 0.9 1.5 ± 0.7 1.4 ± 0.7 1.7 ± 0.9 1.6 ± 0.9 1.5 ± 0.8
CDR 0±0 1 ± 0.4ad 1 ± 0.3a 1 ± 0.3a 1 ± 0.5a 1 ± 0.5a
GDS 2±2 2±2 3±2 4 ± 3abe 2±2 3±3

NPE, Neuropsychological examination; CDR, Clinical Dementia Rating; GDS, Geriatric Depression Scale; AD, Alzheimer’s
disease; VaD, vascular dementia; DLB, dementia with Lewy bodies; bvFTD, behavioural variant frontotemporal dementia;
lvFTD, language variant frontotemporal dementia.
Values are presented as mean ± standard deviation, number (percentage) or median (range). Significant difference p < 0.05 is
compared to acontrols, bAD, cVaD, dDLB, ebvFTD and flvFTD.
g
According to the Verhage system.
h
Years of reported cognitive complaints, an estimation of disease duration.

Table 2. Baseline raw neuropsychological test performance of diagnostic groups

Controls AD VaD DLB bvFTD lvFTD

ng 112 199 10 26 20 15
MMSE 28 ± 1 22 ± 4acdef 25 ± 4a 23 ± 3ae 26 ± 3a 24 ± 3a
Memory
VAT 12 ± 0.5 6 ± 4acdef 11 ± 1 10 ± 2a 11 ± 2 10 ± 3a
RAVLT total immediate recall 40 ± 9 23 ± 8ae 23 ± 10ae 24 ± 8ae 34 ± 11a 25 ± 6ae
RAVLT delayed 8±3 2 ± 3ade 4 ± 4a 4 ± 3a 5 ± 4a 3 ± 2ae
Language
VAT naming 12 ± 0.2 11 ± 2a 12 ± 0.9 12 ± 0.6 11 ± 2ad 8 ± 3abcde
ABCD naming 19 ± 1 16 ± 3a 18 ± 1 17 ± 2a 13 ± 6abcd 10 ± 5abcde
Comparative questions 6 ± 0.4 5 ± 1a 5±1 5 ± 1a 5 ± 1a 6 ± 0.4
Animal fluency 23 ± 6 13 ± 5a 10 ± 6a 13 ± 5a 12 ± 5a 11 ± 7a
Letter fluency 38 ± 11 26 ± 12a 17 ± 8ab 23 ± 11ab 21 ± 10a 18 ± 11ab
Attention
Digit Span forward 13 ± 3 11 ± 3a 10 ± 2a 11 ± 3 12 ± 2 12 ± 3
TMT Ah 38 ± 14 90 ± 68aef 87 ± 64a 87 ± 41aef 53 ± 22 48 ± 24
Executive functioning
Digit Span backwards 9±2 7 ± 3aef 6 ± 3ae 6 ± 2aef 8±3 8±2
TMT Bh 87 ± 28 205 ± 112aef 216 ± 77aef 213 ± 90aef 120 ± 39 138 ± 63
FAB 17 ± 2 12 ± 4a 14 ± 3a 12 ± 4a 14 ± 2a 14 ± 2a
Visuospatial functioning
Letter fragments 19 ± 0.8 16 ± 5af 18 ± 1 15 ± 4aef 18 ± 3 19 ± 1
Dot counting 10 ± 0.4 9 ± 2af 9±1 9±1 10 ± 0.5 10 ± 0.0
Number location 9±1 8 ± 2acef 9±1 8 ± 1acef 9±1 10 ± 0.6

MMSE, Mini-Mental State Examination; VAT, Visual Association Test; RAVLT, Rey Auditory Verbal Learning Test; ABCD,
Arizona Battery for Communication Disorders; TMT, Trail Making Test (Parts A and B); FAB, Frontal Assessment Battery;
AD, Alzheimer’s disease; VaD, vascular dementia; DLB, dementia with Lewy bodies; bvFTD, behavioural variant frontotem-
poral dementia; lvTD, language variant frontotemporal dementia.
Values are presented as mean ± standard deviation. Significant difference p < 0.05 is compared to acontrols, bAD, cVaD,
d
DLB, ebvFTD and flvFTD.
Cognitive profiles are divided into neuropsychological tests. gn differs for every neuropsychological test because raw data
are presented. hHigher scores means worse performance.
 Trajectories of cognitive decline in dementia 5

Table 3. Estimated effect of diagnosis on baseline and annual change in MMSE score and compound scores for different cognitive domains

MMSE Memory Language Attention Executive functioning Visuospatial functioning

Baseline cognitive score

Controlsa 0.02 ± 0.2 −0.02 ± 0.2 0.0 ± 0.2 −0.01 ± 0.1 0.01 ± 0.1 −0.01 ± 0.2
AD −4.9 ± 0.2*** −5.0 ± 0.5*** −1.9 ± 0.2*** −1.4 ± 0.1*** −2.4 ± 0.1*** −1.8 ± 0.3***
VaD −2.9 ± 0.8** −1.5 ± 0.7* −1.9 ± 0.7** −1.5 ± 0.3*** −2.4 ± 0.4*** −1.0 ± 0.7
DLB −3.5 ± 0.5*** −2.4 ± 0.5*** −1.3 ± 0.4** −1.3 ± 0.2*** −2.6 ± 0.3*** −1.5 ± 0.5**
bvFTD −1.6 ± 0.6** −1.2 ± 0.5* −2.6 ± 0.5*** −0.7 ± 0.2** −1.2 ± 0.3*** −0.4 ± 0.4
lvFTD −2.9 ± 0.6*** −2.1 ± 0.6*** −5.7 ± 0.7*** −0.3 ± 0.3 −1.2 ± 0.3** −0.2 ± 0.5
Annual cognitive change
Controlsa 0.1 ± 0.2 0.1 ± 0.1 −0.1 ± 0.1 0.1 ± 0.1 0.04 ± 0.1 0.004 ± 0.1
AD −1.4 ± 0.1*** −0.9 ± 0.1*** −0.8 ± 0.1*** −0.4 ± 0.04*** −0.5 ± 0.1*** −0.8 ± 0.2***
VaD 0.4 ± 0.6 −0.6 ± 0.4 0.03 ± 0.5 −0.2 ± 0.2* −0.4 ± 0.2* −0.2 ± 0.4
DLB −0.1 ± 0.4 −0.8 ± 0.3** −0.3 ± 0.3 −0.3 ± 0.1** −0.3 ± 0.1* −0.6 ± 0.2*
bvFTD −1.8 ± 0.4*** −1.8 ± 0.3*** −2.0 ± 0.4*** −0.3 ± 0.1** −0.7 ± 0.1*** −0.4 ± 0.3
lvFTD −0.6 ± 0.5 −0.4 ± 0.3 −1.0 ± 0.5* −0.4 ± 0.1** −0.6 ± 0.2** −0.4 ± 0.4

MMSE, Mini-Mental State Examination; AD, Alzheimer’s disease; VaD, vascular dementia; DLB, dementia with Lewy
bodies; bvFTD, behavioural variant frontotemporal dementia; lvFTD, language variant frontotemporal dementia.
a
Reference group.
Data are presented as β ± standard error (S.E.). β values represent uncorrected estimated baseline performance or estimated
change over time for the diagnoses compared with performance of controls at baseline. p values for differences with controls
are given for the models corrected for age, gender and education.
*p < 0.05 for main effect; p < 0.10 for interaction, **p < 0.01, ***p < 0.001.

Discussion
2011). Fluency tasks address both language and
In this prospective longitudinal study in a large sample executive functioning and therefore how to interpret
of patients with different types of dementia, we found fluency tasks is open to debate. As expected, lvFTD
that during an average follow-up time of 1.5 years, pat- was, at baseline, most impaired in language (Reilly
terns of cognitive decline differed between patient et al. 2010; Karantzoulis & Galvin, 2011), which
groups according to type of dementia. Patients with declined further over time. In bvFTD and lvFTD,
AD declined in all cognitive domains whereas patients visuospatial functioning was relatively preserved and
with bvFTD showed the fastest decline over time. showed limited decline over time (Reilly et al. 2010;
Patients with VaD showed decline in attention and Karantzoulis & Galvin, 2011). This might be due to
executive functioning. Patients with lvFTD showed the sparing of anatomical regions associated with
most decline in attention and executive functioning visuospatial abilities, such as the parieto-occipital
whereas patients with DLB showed decline over time lobes (McGinnis, 2012). bvFTD showed the fastest de-
in several cognitive domains. cline of all types of dementia investigated, in all cogni-
bvFTD is neuropsychologically characterized by tive domains except visuospatial functioning. A
executive dysfunction with relative sparing of memory previous study showed a faster rate of cognitive de-
and visuospatial functions (Seelaar et al. 2011). The cline and shorter survival in FTD compared to AD
presence of this cognitive profile, however, is not (Rascovsky et al. 2005). In our study, bvFTD patients
required according to the revised diagnostic criteria also showed a faster cognitive decline than AD
for bvFTD (Rascovsky et al. 2011). Although it has patients. However, the duration of illness seems to be
often been reported that memory is relatively spared variable in FTD, ranging from 2 to 20 years (Seelaar
in bvFTD, prominent amnestic symptoms are present et al. 2011).
in a subgroup of elderly bvFTD patients (Rascovsky DLB is characterized by visuospatial and executive
et al. 2011). In our study, executive dysfunction was impairment, which are present early in the disease
not the only hallmark of bvFTD; we showed that base- course (Karantzoulis & Galvin, 2011). We found that,
line performance on almost every cognitive domain in at baseline, patients with DLB showed most severe im-
bvFTD was worse compared to controls, with lan- pairment on memory and executive functioning. Over
guage impairment most pronounced. This striking lan- time, visuospatial functioning declined fastest after
guage impairment might be due to fluency tasks that memory. We did not find pronounced impairment in
were included in the language domain (Seelaar et al. attention in DLB, compared to the other types of
6 L. L. Smits et al.

Fig. 1. Estimated cognitive performance over time compared with baseline z scores of controls. Lines represent β values as
presented in Table 3. Red lines represent Alzheimer’s disease (AD), dark blue lines vascular dementia (VaD), light blue lines
dementia with Lewy bodies (DLB), pink lines behavioural variant frontotemporal dementia (bvFTD), yellow lines language
variant frontotemporal dementia (lvFTD), and green lines controls. MMSE, Mini-Mental State Examination.

dementia. The literature suggests that differences can Galvin, 2011). We also found more frank visuospatial
only be found on more complex attentional tasks impairment in AD. This could be due to the relatively
(Karantzoulis & Galvin, 2011), and it is possible that young age of our patients with AD. Patients with AD
the tasks we used were insufficiently sensitive. It is and an earlier age of onset more often have posterior
often thought that DLB is one of the fastest declining brain damage, including atrophy and functional
types of dementia but our findings do not support brain changes (de Waal et al. 2011; Moller et al. 2013).
this idea and earlier results seem conflicting We found that patients with VaD showed decline in at-
(Olichney et al. 1998; Walker et al. 2012). We observed tention and executive functioning. As far as we know,
decline in most cognitive domains but the rate of only one study has investigated rate of decline in VaD
decline was not much faster than in other types of and AD. However, this study was conducted in the
dementia. An explanation could be that patients who preclinical stage, and persons with preclinical VaD
declined too fast were lost to follow-up, and as a result showed a faster rate of cognitive decline than preclini-
we saw only the milder subgroup. cal AD in episodic memory, visuospatial functioning
Memory impairment and executive dysfunction are and category fluency (Laukka et al. 2012). This former
well-known early features of AD (Karantzoulis & study did not incorporate tests for executive
 Trajectories of cognitive decline in dementia 7

Table 4. Schematic overview of estimated trajectories of cognitive are scarce, and for these more rare types of dementia
decline for patients with dementia, compared to controls our group sizes are in fact not so small. We feel that
even when reported significance levels are limited by
AD VaD DLB bvFTD lvFTD suboptimal power, reporting effect sizes of these
groups is very valuable. Our results need to be repli-
MMSE ; – – ;; – cated in larger studies. The sample described in our
Memory ; – ; ;; – study was relatively young, which might limit general-
Language ; – – ;; ; izability to dementia in general. Nevertheless, we con-
Attention ; ; ; ; ; sider that the relatively young age also has advantages,
Executive functioning ; ; ; ; ;
as non-AD types of dementia often develop at a
Visuospatial ; – ; – –
younger age and hence we were able to include diverse
functioning
types of dementia. In addition, dementia at a younger
age is often thought to be more pure, and less mixed,
MMSE, Mini-Mental State Examination; AD, Alzheimer’s
disease; VaD, vascular dementia; DLB, dementia with Lewy pathology and hence the patterns of decline may be
bodies; bvFTD, behavioural variant frontotemporal de- more specific. Among the strengths of our study is
mentia; lvFTD, language variant frontotemporal dementia. the longitudinal set-up and the comprehensive neurop-
; Significant annual decline p < 0.10. sychological assessment covering five cognitive
;; Significant annual decline p < 0.05 and > 1.5 z score domains.
annual decline. Our findings have considerable impact, as they pro-
This table is based on the estimated annual cognitive vide estimates of the natural disease course in different
change in Table 3. Patients with AD showed decline on all types of dementia. Clinical trials have thus far most
cognitive domains, and global cognition. Patients with VaD
frequently used global outcome measures such as the
showed a decline in attention and executive functioning. For
Alzheimer’s Disease Assessment Scale – Cognitive
DLB, we found a decline over time in memory, visuospatial
subscale (ADAS-Cog; Rosen et al. 1984), the CDR
functioning, attention and executive functioning. bvFTD
showed a fast decline in global cognition and all cognitive scale, and questionnaires about activities of daily living
domains with the exception of visuospatial functioning. For and behaviour. Shortcomings of the ADAS-Cog in-
lvFTD, we found a decline in language, executive function- clude its primary focus on memory and language,
ing and attention. while tasks assessing attention, executive or visuospa-
tial functioning are not incorporated in the test
(Wesnes & Harrison, 2003). More extensive neuropsy-
functioning and attention, the domains in which we chological assessments covering several cognitive
found decline in VaD. Another relevant shortcoming domains, such as the Neuropsychological Test
of the study by Laukka et al. (2012) is that no brain ima- Battery, are increasingly being preferred over the
ging was available and about 70% of patients had a di- ADAS-Cog. These neuropsychological batteries are
agnosis of VaD, which was primarily a diagnosis of better at detecting subtle cognitive changes and there-
post-stroke dementia. By contrast, the diagnoses of fore seem more suitable for cognitive assessments in
all of our VaD patients were supported by brain ima- trials in preclinical/early disease stages. The Food and
ging, and mostly involved small vessel disease. Drug Administration (FDA) recently proposed revis-
The diverse patterns of cognitive decline could not ing the criteria for drug approval because persons in
be explained by differences in disease duration at base- very early disease stages have no obvious deterioration
line or on the CDR scale, although we found differ- in daily functioning. They suggested that, in trials (in
ences on the MMSE. At baseline, patients with AD very early stages of disease), cognitive tests should
performed worse on the MMSE and this might indicate be used as end-points (Kozauer & Katz, 2013). In ad-
that these patients were cognitively more impaired. dition, clinical trials are currently designed for other
However, the MMSE is known to rely heavily on mem- types of dementia than AD. Therefore, estimates of
ory, the domain on which AD most was impaired cognitive trajectories of different types of dementia
(Nieuwenhuis-Mark, 2010). The MMSE also relies on are essential for trial design and power calculations.
language, and patients with bvFTD declined fastest
on language and memory, possibly reflected in their
fast decline on MMSE. As a consequence, the MMSE Acknowledgements
might have some restrictions when it is used to com- The Alzheimer Centre receives unrestricted funding
pare global cognition in different types of dementia. from various sources through the VUmc Fonds. This
A possible limitation of this study is that, except for study received no specific funding. We thank
AD and controls, sample sizes were relatively small. Dr B. M. Tijms and Dr J. Berkhof for their statistical
However, longitudinal studies on non-AD dementias advice. Research undertaken by the VUmc Alzheimer
8 L. L. Smits et al.

Centre is part of the neurodegeneration research pro- Rascovsky K, Patterson K, Miller BL, Knopman DS,
gramme of the Neuroscience Campus Amsterdam. Hodges JR, Mesulam MM, Grossman M (2011).
The Alzheimer Centre VUmc is supported by Classification of primary progressive aphasia and its
Alzheimer Nederland and Stichting VUmc Fonds. variants. Neurology 76, 1006–1014.
Hanyu H, Sato T, Hirao K, Kanetaka H, Sakurai H, Iwamoto T
The clinical database structure was developed with
(2009). Differences in clinical course between dementia with
funding from Stichting Dioraphte. This research
Lewy bodies and Alzheimer’s disease. European Journal of
received no specific grant from any funding agency, Neurology 16, 212–217.
commercial or not-for-profit sectors. Karantzoulis S, Galvin JE (2011). Distinguishing Alzheimer’s
disease from other major forms of dementia. Expert Review
of Neurotherapeutics 11, 1579–1591.
Declaration of Interest Kozauer N, Katz R (2013). Regulatory innovation and drug
development for early-stage Alzheimer’s disease. New
Dr Prins serves on the advisory board of Boehringer
England Journal of Medicine 368, 1169–1171.
Ingelheim and Envivo. He has been a speaker at sym-
Laukka EJ, Macdonald SW, Fratiglioni L, Backman L (2012).
posia organized by Janssen and Novartis. He has a Preclinical cognitive trajectories differ for Alzheimer’s
senior fellowship at the Alzheimer Centre VUmc partly disease and vascular dementia. Journal of the International
supported by Vereniging AEGON and receives re- Neuropsychological Society 18, 191–199.
search support from the Brain Foundation of The Libon DJ, Xie SX, Wang X, Massimo L, Moore P, Vesely L,
Netherlands (project number H07.03) and Alzheimer Khan A, Chatterjee A, Coslett HB, Hurtig HI, Liang TW,
Nederland (project number WE.03-2012-02). Dr Grossman M (2009). Neuropsychological decline in
Scheltens serves/has served on the advisory boards of frontotemporal lobar degeneration: a longitudinal analysis.
Genentech, Novartis, Pfizer, Roche, Danone, Nutricia, Neuropsychology 23, 337–346.
Jansen AI, Baxter and Lundbeck. He has been a Lindeboom J, Matto D (1994). Digit series and Knox cubes as
concentration tests for elderly subjects [in Dutch]. Tijdschrift
speaker at symposia organized by Lundbeck, Lilly,
voor Gerontologie en Geriatrie 25, 63–68.
Merz, Pfizer, Jansen AI, Danone, Novartis, Roche and
Lindeboom J, Schmand B, Tulner L, Walstra G, Jonker C
Genentech. He serves on the editorial board of (2002). Visual association test to detect early dementia of
Alzheimer’s Research and Therapy and Alzheimer’s the Alzheimer type. Journal of Neurology, Neurosurgery, and
Disease and Associated Disorders, is a member of the Psychiatry 73, 126–133.
scientific advisory board of the European Union (EU) Luteijn F, van der Ploeg F (1982). GIT. Groninger Intelligence
Joint Programme on Neurodegenerative Disease Test [in Dutch]. Swets & Zeitlinger: Lisse.
Research (JPND) and the French National Plan McGinnis SM (2012). Neuroimaging in neurodegenerative
Alzheimer. Dr van der Flier has received funding dementias. Seminars in Neurology 32, 347–360.
and speaker honorarium from Boehringer Ingelheim; McKeith IG, Dickson DW, Lowe J, Emre M, O’Brien JT,
all funds were paid to her institution. The other Feldman H, Cummings J, Duda JE, Lippa C, Perry EK,
Aarsland D, Arai H, Ballard CG, Boeve B, Burn DJ, Costa
authors have no conflicts of interest to declare.
D, Del Ser T, Dubois B, Galasko D, Gauthier S, Goetz CG,
Gomez-Tortosa E, Halliday G, Hansen LA, Hardy J,
Iwatsubo T, Kalaria RN, Kaufer D, Kenny RA, Korczyn A,
References
Kosaka K, Lee VM, Lees A, Litvan I, Londos E, Lopez OL,
Bayles CM, Tomoeda CK (1993). Arizona Battery for Minoshima S, Mizuno Y, Molina JA, Mukaetova-Ladinska
Communication Disorders in Dementia. Canyonlands EB, Pasquier F, Perry RH, Schulz JB, Trojanowski JQ,
Publishing: Tucson, AZ. Yamada M (2005). Diagnosis and management of dementia
de Waal H, Stam CJ, Blankenstein MA, Pijnenburg YA, with Lewy bodies: third report of the DLB Consortium.
Scheltens P, van der Flier WM (2011). EEG abnormalities Neurology 65, 1863–1872.
in early and late onset Alzheimer’s disease: understanding McKhann G, Drachman D, Folstein M, Katzman R, Price D,
heterogeneity. Journal of Neurology, Neurosurgery, and Stadlan EM (1984). Clinical diagnosis of Alzheimer’s
Psychiatry 82, 67–71. disease: report of the NINCDS-ADRDA Work Group
Dubois B, Slachevsky A, Litvan I, Pillon B (2000). The FAB: a under the auspices of Department of Health and Human
Frontal Assessment Battery at bedside. Neurology 55, Services Task Force on Alzheimer’s Disease. Neurology 34,
1621–1626. 939–944.
Folstein MF, Folstein SE, McHugh PR (1975). ‘Mini-mental McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack
state’. A practical method for grading the cognitive state of Jr. CR, Kawas CH, Klunk WE, Koroshetz WJ, Manly JJ,
patients for the clinician. Journal of Psychiatric Research 12, Mayeux R, Mohs RC, Morris JC, Rossor MN, Scheltens P,
189–198. Carrillo MC, Thies B, Weintraub S, Phelps CH (2011). The
Gorno-Tempini ML, Hillis AE, Weintraub S, Kertesz A, diagnosis of dementia due to Alzheimer’s disease:
Mendez M, Cappa SF, Ogar JM, Rohrer JD, Black S, recommendations from the National Institute on
Boeve BF, Manes F, Dronkers NF, Vandenberghe R, Aging-Alzheimer’s Association workgroups on diagnostic
 Trajectories of cognitive decline in dementia 9

guidelines for Alzheimer’s disease. Alzheimer’s and Dementia frontotemporal dementias: a longitudinal study of 236
7, 263–269. patients. Neuropsychology 20, 319–335.
Moller C, Vrenken H, Jiskoot L, Versteeg A, Barkhof F, Román GC, Tatemichi TK, Erkinjuntti T, Cummings JL,
Scheltens P, van der Flier WM (2013). Different patterns of Masdeu JC, Garcia JH, Amaducci L, Orgogozo J-M, Brun A,
gray matter atrophy in early- and late-onset Alzheimer’s Hofman A, Moody DM, O’Brien MD, Yamaguchi T,
disease. Neurobiology of Aging 34, 2014–2022. Grafman J, Drayer BP, Bennett DA, Fisher M, Ogata J,
Neary D, Snowden JS, Gustafson L, Passant U, Stuss D, Kokmen E, Bermejo F, Wolf PA, Gorelick PB, Bick KL,
Black S, Freedman M, Kertesz A, Robert PH, Albert M, Pajeau AK, Bell MA, DeCarli C, Culebras A, Korczyn AD,
Boone K, Miller BL, Cummings J, Benson DF (1998). Bogousslavsky J, Hartmann A, Scheinberg P (1993).
Frontotemporal lobar degeneration: a consensus on clinical Vascular dementia: diagnostic criteria for research studies.
diagnostic criteria. Neurology 51, 1546–1554. Report of the NINDS-AIREN International Workshop.
Nieuwenhuis-Mark RE (2010). The death knoll for the Neurology 43, 250–260.
MMSE: has it outlived its purpose? Journal of Geriatric Rosen WG, Mohs RC, Davis KL (1984). A new rating scale
Psychiatry and Neurology 23, 151–157. for Alzheimer’s disease. American Journal of Psychiatry 141,
Olichney JM, Galasko D, Salmon DP, Hofstetter CR, 1356–1364.
Hansen LA, Katzman R, Thal LJ (1998). Cognitive decline Saan RJ, Deelman BG (1986). The 15-Word Test A and B (A
is faster in Lewy body variant than in Alzheimer’s disease. Preliminary Manual) [in Dutch]. afd. Neuropsychologie,
Neurology 51, 351–357. AZG: Groningen.
Rascovsky K, Hodges JR, Knopman D, Mendez MF, Kramer Schmand B, Groenink SC, van den Dungen M (2008). Letter
JH, Neuhaus J, van Swieten JC, Seelaar H, Dopper EG, fluency: psychometric properties and Dutch normative data
Onyike CU, Hillis AE, Josephs KA, Boeve BF, Kertesz A, [in Dutch]. Tijdschrift voor Gerontologie en Geriatrie 39, 64–76.
Seeley WW, Rankin KP, Johnson JK, Gorno-Tempini ML, Seelaar H, Rohrer JD, Pijnenburg YA, Fox NC, van Swieten
Rosen H, Prioleau-Latham CE, Lee A, Kipps CM, Lillo P, JC (2011). Clinical, genetic and pathological heterogeneity
Piguet O, Rohrer JD, Rossor MN, Warren JD, Fox NC, of frontotemporal dementia: a review. Journal of Neurology,
Galasko D, Salmon DP, Black SE, Mesulam M, Neurosurgery, and Psychiatry 82, 476–486.
Weintraub S, Dickerson BC, Diehl-Schmid J, Pasquier F, Tan KS, Libon DJ, Rascovsky K, Grossman M, Xie SX
Deramecourt V, Lebert F, Pijnenburg Y, Chow TW, Manes (2013). Differential longitudinal decline on the Mini-Mental
F, Grafman J, Cappa SF, Freedman M, Grossman M, State Examination in frontotemporal lobar degeneration
Miller BL (2011). Sensitivity of revised diagnostic criteria and Alzheimer disease. Alzheimer Disease and Associated
for the behavioural variant of frontotemporal dementia. Disorders 27, 310–315.
Brain 134, 2456–2477. van der Flier WM, Pijnenburg YA, Prins N, Lemstra AW,
Rascovsky K, Salmon DP, Lipton AM, Leverenz JB, DeCarli Bouwman FH, Teunissen CE, van Berckel BN, Stam CJ,
C, Jagust WJ, Clark CM, Mendez MF, Tang-Wai DF, Barkhof F, Visser PJ, van Egmond E, Scheltens P (2014).
Graff-Radford NR, Galasko D (2005). Rate of progression Optimizing patient care and research: the Amsterdam
differs in frontotemporal dementia and Alzheimer disease. dementia cohort. Journal of Alzheimer’s Disease 41, 313–327.
Neurology 65, 397–403. Verhage F (1964). Intelligence and Age: Study with Dutch People
Reilly J, Rodriguez AD, Lamy M, Neils-Strunjas J (2010). Aged 12–77 [in Dutch]. Van Gorcum: Assen.
Cognition, language, and clinical pathological features of Walker Z, McKeith I, Rodda J, Qassem T, Tatsch K, Booij J,
non-Alzheimer’s dementias: an overview. Journal of Darcourt J, O’Brien J (2012). Comparison of cognitive
Communication Disorders 43, 438–452. decline between dementia with Lewy bodies and
Reitan RM (1958). Validity of the Trail Making Test as an Alzheimer’s disease: a cohort study. BMJ Open 2, e000380.
indicator of organic brain damage. Perceptual and Motor Warrington EK, James M (1991). The Visual Object and Space
Skills 8, 271–276. Perception Battery. Thames Valley Test Company: Bury St
Rey A (1964). The Clinical Assessment in Psychology [in French]. Edmunds, UK.
Presse de Universitaire de France: Paris. Wesnes KA, Harrison JE (2003). The evaluation of cognitive
Rogers TT, Ivanoiu A, Patterson K, Hodges JR (2006). function in the dementias: methodological and regulatory
Semantic memory in Alzheimer’s disease and the considerations. Dialogues in Clinical Neurosciences 5, 77–88.