diffusion tensor imaging - amyotrophic lateral sclerosis - biomarkers
Tensor de difusión - esclerosis lateral amiotrófica - biomarcadores
Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease of
unknown etiology that is characterized mainly by degeneration of upper and lower motor
neurons[1]. Although the motor system is the most affected clinically, it has been reported
that up to 50% of ALS patients have some degree of cognitive impairment in the course
of the disease[2]. Striking advances in the understanding of genetics and neuropathology in ALS have
built the new concept of a spectrum of the disease that includes both ALS, as well
as frontotemporal dementia[3].
In ALS, the spectrum of cognitive impairment is variable and is expressed only in
a subset of affected patients. Therefore, there is currently great interest in developing
biomarkers of cognitive and extra-motor systems damage in ALS. In this regard, new
neuroimaging techniques are postulated as potential tools to obtain accurate data
and measure early activity in non-motor structures[4]. Diffusion tensor imaging (DTI) is a technique of magnetic resonance imaging (MRI)
used to characterize the architecture of the white matter, based on the orientation
properties of diffusion of water molecules in the brain[5]. White matter fibers have a preferential orientation of the movement of water molecules,
being more restricted in a perpendicular, than in a parallel direction (anisotropic
diffusion). The fractional anisotropy (FA) is a measure of the directionality of diffusion,
which is reduced when the white matter is injured. Previous studies with DTI in ALS
have shown a reduction of FA in the corticospinal tract consistent with the known
characteristics of the disease[6]. Recently, studies have shown DTI white matter damage in extra-motor[7],[8],[9],[10] tracts, but the correlation between the variables of diffusion and clinical presentation
have shown inconsistent results to date[11].
The aim of this preliminary study was to investigate the presence and characteristics
of early changes in the DTI in ALS patients and their correlation with the cognitive
and clinical profile.
METHODS
Participants
A case-control study was performed between December 1, 2012 and December 1, 2014.
Patients between 18 and 80 years with a diagnosis of probable or definite ALS according
to the El Escorial criteria[12] were included. Age and sex-matched controls were identified for each ALS patient.
These were selected from the ALS patients’ relatives or friends. To be included, controls
could not present with a cognitive deficit according to the neuropsychological test.
There was only one patient and one control who refused participation in the study.
Clinical data was assessed and collected at the Neuromuscular Disease Center of the
Hospital Italiano de Buenos Aires. The site of onset, phenotype, age at diagnosis,
time to the cognitive assessment and DTI, forced vital capacity, time to gastrostomy,
level of physical disability assessed by the ALS revised functional rating scale –
(ALSFRS-R) score[13], and treatment used was recorded.
The present study was carried out with the approval of the Ethics Committee of the
Institution and all participants gave informed consent.
Cognitive evaluation
A battery of neuropsychological tests adapted to patients with ALS (including dysarthria
and motor impairment) was used. All neuropsychological tests were performed by neuropsychologists
and neurologists of the Cognitive Disorders Center of the Hospital Italiano de Buenos
Aires. The neuropsychological test battery included tests used to assess memory (Rey
Auditory Verbal Learning Test); executive functions (direct and reverse digit number-letter;
Trail Making Test: Motor Speed, Visual Scanning, Number Sequencing, and Letter Sequencing);
social cognition (Key Search - BADS, Eye Task Baron Cohen); visuospatial functions
(block building); and language (vocabulary – abbreviated Boston Naming Test, phonemic
fluency, semantic fluency). In addition, all patients were evaluated by the Addenbrooke’s
Cognitive Examination revised test[14] to determine cognitive deficit. A value below 86 points was considered a cognitive
deficit for patients with a high education level (complete schooling; more than 12
years of study). A score below 68 points was considered to be a cognitive deficit
for patients with low educational level (incomplete schooling). Revised diagnostic
criteria were applied for the diagnosis of the behavioral variant of frontotemporal
dementia[15]. The diagnosis of other dementias was based on the DSM-IV and NINCDS-ADRDA[16]. Phukan criteria were used for categorizing cognitive deficit in patients without
dementia[17].
Diffusion tensor imaging
Brain MRI was performed using a 1.5-T Sigma resonator scanner (Avanto Siemens). The
sequences obtained were conventional (T1-weighted, proton density and T2-weighted
sequences). The DTI sequences were obtained to evaluate the change in white matter
tracts. The following parameters were used: TR between 5000 and 8000 ms, TE 66 ms,
45 cuts, resolution of 2 mm x 2 mm x 4 mm. The DT images were corrected for the movement,
using a nonlinear technique provided by FMRIB Software Library (FSL). The FA was measured
by the tract based spatial statistics software. Spatial reorientation of FA was allowed
by the tract based spatial statistics in a standard space with no systematic effect
of spatial transformation of the direction of white matter fibers and without the
need to select areas of the fibers that could skew the analysis[18]. Data provided by the tract based spatial statistics were analyzed by FA maps using
a linear model. The FA was compared between patients and controls and between patients
with ALS with and without cognitive impairment, adjusting for clinical and demographic
variables. For this analysis, the statistical significance was set at p < 0.05.
Statistical analysis
Baseline characteristics of the cohort were reported as percentages for categorical
data and media with its standard deviation for continuous data. Data were compared
using Fisher’s exact test and the Mann Whitney U test for categorical and continuous
data respectively. Data from FA maps were correlated with clinical findings by a regression
model. The collection and data analysis were performed using STATA 10.1 program.
RESULTS
A total of 24 patients with sporadic ALS and 13 healthy controls (11 ALS friends and
two ALS relatives) were included. Demographic and clinical characteristics of patients
with ALS and controls are shown in the [Table]. The most frequent cognitive profile was the multi-domain executive dysfunction
(36%, n=10). The DTI showed diffuse damage of the white matter in patients with ALS
([Figure 1]). The greatest alteration was observed in the bilateral frontal and parietal white
matter.
Table
Baseline characteristics and multivariate logistic regression analysis between ALS
patients and controls.
|
Variable
|
ALS patients (n = 24)
|
Control group (n = 13)
|
p-value
|
|
Age, years
|
66.9 ± 2
|
66.9 ± 2.3
|
0.99
|
|
Female sex, n (%)
|
15 (62.5)
|
4 (36.3)
|
0.15
|
|
Age at ALS onset, years
|
64.6 ± 4
|
-
|
-
|
|
Disease duration, months
|
14.6 ± 18
|
-
|
-
|
|
Bulbar onset, n (%)
|
4 (16.7)
|
-
|
-
|
|
Spinal onset, n (%)
|
20 (83.3)
|
-
|
-
|
|
Mean ALS FRS-R at analysis, points
|
35.2 ± 8
|
-
|
-
|
|
Riluzol at analysis
|
15 (65.2)
|
-
|
-
|
|
Gastrostomy at analysis
|
0
|
-
|
-
|
|
Education, years
|
11 ± 1
|
10.2 ± 1
|
0.56
|
|
ALS family history, n
|
0
|
0
|
-
|
|
Dementia family history, n (%)
|
2 (8.3)
|
0
|
0.46
|
|
Parkinson’s disease family history, n (%)
|
1 (4.17)
|
0
|
0.64
|
|
Cognitive dysfunction, n (%)
|
17 (68)
|
0
|
< 0.001
|
|
Multi-domain executive dysfunction, n (%)
|
10 (42)
|
0
|
< 0.001
|
|
Single-domain executive dysfunction, n (%)
|
6 (25)
|
0
|
0.02
|
|
Multi-domain non-executive dysfunction, n (%)
|
0
|
0
|
-
|
|
Single-domain non-executive dysfunction, n (%)
|
0
|
0
|
-
|
|
DTI alteration, FA mean value
|
0.713 ± 0.021
|
0.654 ± 0.024
|
< 0.001
|
ALS: amyotrophic lateral sclerosis; FRS-R: functional rating scale - revised; DTI:
diffusion tensor imaging; FA: fractional anisotropy.
Figure 1 DTI showed diffuse damage of the white matter in patients with ALS.
Multivariate logistic regression analysis showed that the mean value of FA was lower
in ALS patients vs. healthy controls (p < 0.001), ([Table]). The greatest damage to white matter was observed in patients with cognitive impairment
([Figure 2]). The multi-domain executive dysfunction profile was associated with significantly
increased white matter damage, after multivariate analysis (p < 0.001) ([Table]).
Figure 2 The red shows increase in FA in patients with ALS and without cognitive impairment
(A) compared with patients with ALS and cognitive impairment (B).
It was also observed that patients with ALS and cognitive impairment showed greater
involvement of diffuse white matter bilaterally vs. ALS patients without cognitive
dysfunction. This difference was significant (p < 0.001), after logistic regression
analysis,
DISCUSSION
Our preliminary study showed a diffuse alteration of the cerebral white matter, measured
by DTI in sporadic ALS patients compared to healthy controls. It was observed that
the greater involvement of white matter (lower FA values in DTI) was significantly
associated with the presence of cognitive deficits. These findings support previous
studies with voxel-based morphometry, which observed that ALS patients showed greater
cognitive dysfunction and atrophy in extra-motor regions[19].
Recently, diverse authors have reported an association between alterations in DT images
and disease severity in patients with ALS[6],[10]. However DTI longitudinal studies in ALS patients have yielded conflicting results
regarding the correlation of progression of brain damage and concomitant worsening
of disability[20]. In our series of patients with ALS, no association between the degree of disability,
as assessed by the ALSFRS-R scale, and alterations in the DTI was found.
Few studies have evaluated the early neuroimaging findings in ALS patients with mild
neuropsychological alteration. The imaging correlation of cognitive impairment in
ALS has been described more frequently in patients who have ALS associated with frontotemporal
dementia, reporting atrophy of the frontal lobe, temporal and frontal hypometabolism
reduced perfusion of the frontotemporal cortex and anterior cingulate gyrus[21]. Even less evidence exists of the imaging alterations observed in non-demented patients
with ALS but with mild cognitive impairment. Recently, Agosta et al, in a study with
DTI, demonstrated the association of the degeneration of white matter tracts with
neuropsychological deficits in patients with ALS[22].
In our study, we highlight that all patients were evaluated at an early stage of the
disease with a full battery of neuropsychological tests adapted to ALS patients, which
included all cognitive areas. Phukan criteria, a useful validated cognitive classification
for ALS patients, were used for evaluating subtle cognitive impairment in non-demented
patients, which is very important in the early stages of the disease. It is relevant
to note that these criteria comprehensively address the commitment of various cognitive
areas and not just the already-recognized frontal-executive dysfunction described
in ALS patients. Although no patients met the criteria for frontotemporal dementia
or other dementias, 66% had some degree of cognitive dysfunction, which is consistent
with a recently-published study[23]. The most frequent cognitive profile was the multi-domain executive dysfunction,
and it was this subgroup of patients who showed greater impairment of white matter
in DTI, showing a pattern of diffuse involvement, predominantly in the frontal lobes.
The neuropsychological profile observed in our group of ALS patients agrees with different
reported series. The frontal executive skills deficits, evidenced by impaired sustained
attention, working memory, concentration and speed of information processing predominated.
Furthermore, comparing the subset of ALS patients with cognitive dysfunction versus
those without cognitive dysfunction, it was observed that the former showed greater
alteration of diffuse white matter in DTI, although it was a minor difference. The
latter supports the usefulness of DTI as a tool for objective measurement of impairment
of extra-motor structures in ALS patients in the early stages of the disease.
A limitation of our study is that genetic testing was not performed to detect mutation
in the C9orf72 gene in the patients who were included. This is relevant because patients
with this mutation have early cognitive manifestations[6]. We believe it would be useful to assess the changes of DTI in patients with mutations
in C9orf72 and cognitive dysfunction. Thus, evaluation with DTI may be useful to characterize
and select the sporadic ALS patients requiring such genetic evaluation.
Considering our preliminary findings, DTI would be a very useful tool to characterize
accurately and early on, the degree and extent of involvement of white matter in patients
with ALS.
Longitudinal studies are needed to confirm this hypothesis and clarify whether DTI
can serve as an effective biomarker to predict ALS patients will progress to develop
dementia.
