Article Text

Research paper
Cognitive correlates in amyotrophic lateral sclerosis: a population-based study in Italy
  1. Anna Montuschi1,
  2. Barbara Iazzolino1,
  3. Andrea Calvo1,2,3,
  4. Cristina Moglia1,
  5. Leonardo Lopiano3,
  6. Gabriella Restagno4,
  7. Maura Brunetti4,
  8. Irene Ossola4,
  9. Anna Lo Presti5,
  10. Stefania Cammarosano1,
  11. Antonio Canosa1,
  12. Adriano Chiò1,2,3
  1. 1‘Rita Levi Montalcini’ Department of Neuroscience, the ALS Center, University of Torino, Italy
  2. 2The Neuroscience Institute of Torino (NIT)
  3. 3Department of Neurology, Azienda Ospedaliera Città della Salute e della Scienza, Torino, Italy
  4. 4Laboratory of Molecular Genetics, Azienda Ospedaliera Città della Salute e della Scienza, Torino, Italy
  5. 5‘Cognetti De Martiis’ Department of Economical and Statistical Science, University of Torino, Italy
  1. Correspondence to Professor Adriano Chiò, ALS Center, ‘Rita Levi Montalcini’ Department of Neuroscience, University of Torino, via Cherasco 15, 10126 Torino, Italy; achio{at}


Background There is less data available regarding the characteristics of cognitive impairment in patients with amyotrophic lateral sclerosis (ALS) in a population-based series.

Methodology Patients with ALS incident in Piemonte, Italy, between 2009 and 2011 underwent an extensive neuropsychological battery. Cognitive status was classified as follows: normal cognition, frontotemporal dementia (ALS-FTD), executive cognitive impairment (ALS-ECI), non-executive cognitive impairment (ALS-NECI), behavioural impairment (ALS-Bi), non-classifiable cognitive impairment. We also assessed 127 age-matched and gender-matched controls identified through patients’ general practitioners.

Results Out of the 281 incident patients, 207 (71.9%) underwent the neuropsychological testing; of these, 19 were excluded from the analysis due previous conditions affecting cognition. Ninety-one (49.7%) patients were cognitively normal, 23 (12.6%) had ALS-FTD, 36 (19.7%) ALS-ECI, 10 (5.5%) ALS-NECI, 11 (6.0%) ALS-Bi and 11 (6.0%) non-classifiable cognitive impairment, 1 had comorbid Alzheimer's disease. Patients with ALS-FTD were older, had a lower education level, and had a shorter survival than any other cognitive group. Of the nine cases with C9ORF72 mutation, six had ALS-FTD, two ALS-ECI and one was cognitively normal; one of the five patients with SOD1 mutations and one of the five patients with TARBDP mutations had ALS-Bi.

Conclusions About 50% of Italian patients with ALS had some degree of cognitive impairment, in keeping with a previous Irish study, despite the largely different genetic background of the two populations. The lower educational attainment in patients with ALS-FTD indicated a possible role of cognitive reserve in ALS-related cognitive impairment. ALS-ECI and ALS-NECI may represent discrete cognitive syndromes in the continuum of ALS and FTD.

  • ALS

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterised by a progressive loss of spinal, bulbar and cortical motor neurons, leading to voluntary muscles weakness and wasting and ultimately to death due to respiratory failure. While in about 90% of patients ALS occurs sporadically, in 10% it is genetically transmitted.1 ,2 Extramotor features in ALS include cognitive changes, which have been described in 10–50% of patients.3 ,4

Frequency and clinical correlates of cognitive impairment in ALS are still poorly understood. With only one exception,4 all studies have been performed on small clinic-based cohorts and did not use standardised methodologies for the evaluation of cognition. Recent consensus criteria proposed a classification of frontotemporal cognitive and behavioural syndromes in ALS5 which includes ALS with frontotemporal dementia (ALS-FTD) and two milder forms of ALS with behavioural impairment (ALS-Bi) and ALS with cognitive impairment.

The aim of this study was to assess the frequency and the clinical pattern of cognitive impairment in a population-based series of patients with ALS, identified through the Piemonte and Valle d'Aosta register for ALS (PARALS), fully characterised from the clinical and genetic point of view.


We invited to participate to the study all patients with ALS resident in the provinces of Torino and Cuneo of Piemonte region, Italy, (n=281), and diagnosed between 1 January 2009 and 31 December 2011, identified through the Piemonte and Valle d'Aosta register for ALS,6 meeting El Escorial revised diagnostic criteria for definite, probable and probable laboratory-supported ALS.7 Disease severity was assessed with the amyotrophic lateral sclerosis—functional rating scale revised (ALSFRS-R) scale.8 All patients underwent pulmonary function tests within 4 weeks before or after the neuropsychological evaluation.

Patients with history of neurological disorders affecting cognition (major stroke, severe head injuries, mental retardation), alcohol-dependence and drug-dependence, severe mental illness and use of high-dose psychoactive medications were tested but not included in data analysis. Patients resident in the area but who were not of Italian mother tongue were assessed only through an unstructured interview and therefore were excluded from the analysis. Patients were invited to participate to the study at the time of the diagnosis or during the first scheduled follow-up visit (∼2 months later) and were interviewed at home or at the ALS clinic. In no case cognitive examination was performed more than 12 months after diagnosis. A total of 127 healthy age-matched, gender-matched and education-matched controls underwent the same neuropsychological battery. Controls were enrolled thought patients’ general practitioners (GPs) and were interviewed at home, at the GP office or at the hospital. Only nine subjects asked to participate as controls denied their participation. Most GPs were willing to collaborate (∼85%). When a GP did not collaborate, another GP practicing in the same area was contacted.

Neuropsychological battery

Patients and controls underwent a battery of neuropsychological tests encompassing executive function, memory, visuospatial function and language, selected according to Clinical Diagnostic Criteria for Frontotemporal Lobar Degeneration,9 and ALS-FTD Consensus Criteria.5 The neuropsychological battery included: Mini Mental State Examination; Wisconsin Card Sorting Test; Trail Making A and B; Stroop Colour-Word Interference Test; letter and category fluency test; Wechsler Memory Scale II—revised (Form 2); Rey-Osterrieth Complex Figure Test; Token test; Wechsler Adult Intelligence Scale revised; Raven's Progressive Colored Matrices; Frontal Assessment Battery. In some cases supplementary tasks were administered for a comprehensive evaluation of language; the following tests were used: semantic systems tests (7 and 8) of the Battery for the Analysis of Aphasic Deficits10 and the Silhouette trial of the Visual Object and Space Perception battery.11

Neurobehavioral dysfunction was determined on the basis of direct observation and patient's history,9 ,12 and with the Frontal Systems Behavior Scale,13 using the Family-form evaluated by a close relative (scores: normal ≤59, borderline 60–64; pathological ≥65). If a subject had scores reflecting a frontal systems abnormality in the premorbid and in the postillness forms, he/she was considered pathological only if there was an increase of ≥10 points at the T score between the two forms.14 Anxiety and depression were assessed with the Hospital Anxiety and Depression Scale; the item ‘I feel slowed down’ was discussed with patients in order not to refer to physical disability. Cognitive impairment was defined as impairment on two tests of executive or non-executive function that was below the 5th centile of healthy controls.

The battery was administered following the same sequence in order to avoid the possible differential interference of the answers of one test over the others. The administration of the battery required ∼2 h, and was usually performed in the morning. If the subject felt too tired, a further session was scheduled to complete the battery, within 2 weeks after the first one. Patients’ and controls’ O2 saturation at the time of the neuropsychological testing was measured with a pulse oximetry; none of the patients and controls had evidence of hypoxaemia (oxygen saturation <92 mm Hg).

Cognitive classification

Clinical diagnosis and cognitive classification were performed with the collaboration of two neurologists specialist in ALS and FTD and two neuropsychologists. Patients’ cognitive status was classified as follows:

  1. ALS with normal cognition.

  2. ALS with frontotemporal dementia (ALS-FTD). The diagnosis of frontotemporal dementia was defined according to Clinical Diagnostic Criteria for Frontotemporal Lobar Degeneration.9

  3. ALS comorbid with non-FTD dementias. The diagnoses of non-FTD dementias were based on the criteria of the Diagnostic and Statistical Manual of Mental Disorders-IV-TR15 and those of the National Institute of Neurological and Communicative Disorders and Stroke—Alzheimer's Disease and Related Disorders Association.16

  4. ALS with executive cognitive impairment (ALS-ECI). Patients with ALS who did not meet criteria for FTD or other types of dementia, but who had an impairment in two tests of executive dysfunction compared with healthy controls, that is, had an executive dysfunction, were classified as ALS with executive cognitive dysfunction. A more conservative cut-off than that proposed by the ALS-FTD Consensus Criteria5 was used (2.3rd centile).4

  5. ALS with non-executive cognitive impairment (ALS-NECI). This group includes patients with ALS with impairment in two non-executive domains, in particular visuopraxic abilities, and no impairment in executive function.

  6. ALS with behavioural impairment (ALS-Bi). This group includes patients with predominant behavioural disturbances and with impairment in none or only one test of executive dysfunction and no impairment in non-executive domains.

  7. ALS with non-classifiable cognitive impairment (ALS-NCCI). This group includes patients with ALS with impairment in one executive and/or one non-executive test, sometimes associated with smooth behavioural changes.

Genetic analysis

All coding exons and 50 bp of the flanking intron-exon boundaries of SOD1, of exon 6 of TARDBP, and of exons 14 and 15 of FUS and exons 5, 9, 12 and 14 of OPTN and the single exon of ANG have been PCR amplified, sequenced using the Big-Dye Terminator v3.1 sequencing kit (Applied Biosystems), and run on an ABIPrism 3130 genetic analyser. In patients with positive family history for ALS or FTD all coding exons of VCP have also been assessed. These exons were selected as the vast majority of known pathogenic variants are known to lie within these mutational hot spots. A repeat-primed PCR assay was used to screen for the presence of the GGGGCC hexanucleotide expansion in the first intron of C9ORF72.17 A cut-off of ≥30 repeats was considered pathological.

Statistical methods

Comparisons between means were made with Student t test or analysis of variance; comparisons between categorical variables were made with χ2 test; for all comparisons, Levene's test was used to confirm the equality of variances.

Survival was calculated from onset to death/tracheostomy or censoring date (30 June 2013), using the Kaplan-Meier method, and compared with the log-rank test. No patients were lost to follow-up. Multivariable analysis was performed with Cox proportional hazards model (stepwise backward) (for details, see table 3). For the analysis of the relationship between cognitive status and disease progression, the progression rate for the ALSFRS-R score, its four subscores (bulbar, fine motor, gross motor and respiratory) and forced vital capacity per cent of predicted (FVC%) was calculated as the mean monthly number of points lost from disease onset to the time of cognitive evaluation. For example, the progression rate for the ALSFRS-R score was calculated as follows: (48-ALSFRS-R at time of cognitive evaluation)/duration from onset to diagnosis (months). In the Cox model, these variables were dichotomised on the basis of their median value. The list of all variables included in the Cox model is reported in table 3.

A p level <0.05 was considered statistically significant. All tests were two-tailed. Statistical analyses were carried out using SPSS V.20.0 (SPSS, Chicago, Illinois, USA).

Standard protocol approvals, registrations and patient consents

The study design was approved by our institutional Ethical Committee. Patients and controls signed a written informed consent. The database was managed according to the Italian law for the protection of privacy.


A flow chart of the sequence of participants’ selection is reported in figure 1. Of the 281 patients diagnosed in the study area in the 2009–2011 period, 202 (71.9%) underwent the neuropsychological battery. Of the 79 non-captured patients, 34 were not able to undergo the battery of tests due to their motor disability (7 patients were tracheostomised or used non-invasive ventilation for more than 16 h; 18 patients had severe difficulties in writing and speaking; 9 patients had a severe fatigue and, although willing to collaborate, could not adequately perform the whole battery), 5 were not of Italian mother tongue, 30 declined participation and 10 died before being tested. Nineteen tested patients were excluded from the analysis due to previous neurological disorders affecting cognition (seven patients), severe mental illness (six), drug or alcohol abuse (two), use of high-dose psychoactive medications (one due to bipolar disorder, one due to paranoid schizophrenia), analphabetism (one) and mental retardation (one).

Figure 1

Flow chart showing capture rate and the sequence of participant selection. ALS, amyotrophic lateral sclerosis.

Non-captured patients did not differ demographically and clinically from those who underwent the examination (table 1). The median time from diagnosis to the neuropsychological assessment was 1.9 months (IQR 1.2–3.8 months).

Table 1

Demographic and clinical characteristics of patients with ALS and controls

Cognitive classification

According to the classification criteria for patients’ cognitive status, 23 (12.6%) had ALS-FTD, 36 (19.7%) ALS-ECI, 10 (5.5%) ALS-NECI, 11 (6.0%) ALS-Bi and 11 (6.0%) ALS-NCCI; 91 (49.7%) patients were cognitively normal (see online E-figure S1). One patient had comorbid Alzheimer disease (AD). Twenty-two out of the 23 patients with ALS-FTD presented with behavioural changes typical of behavioural variant fronto-temporal lobar dementia (FTLD); one patient had semantic dementia. Mean scores of the performed tests for each cognitive group and controls are reported in online E-table 1. Box plots of selected test are reported in online E- figures S2–S6.

Cognitive groups were clinically and demographically different (table 2). Patients with ALS-FTD, ALS-ECI and ALS-Bi had a higher mean age (∼70 years) than those with normal cognition and ALS-NECI. ALS-NCCI had the lowest age at onset. Patients with ALS-FTD and those with ALS-NECI had a higher frequency of bulbar onset than all other groups (p=0.003). The mean number of education years was significantly lower in patients with ALS-FTD than in all other groups. ALSFRS-R score and FVC% at the time of the cognitive examination did not show significant differences. However, the ALSFRS-R bulbar subscore (items 1, 2 and 3 of the ALSFRS-R scale) was significantly lower in the group with ALS-FTD (data not shown). The rate of decline of ALSFRS-R, of its subscores and of FVC% was similar in the various groups (see online E-table S2).

Table 2

Demographic and clinical characteristics of patients with ALS according to cognitive status

Patients’ cognitive status and genetics

Of the nine cases carrying the GGGGCC hexanucleotide repeat expansion in the first intron of the C9ORF72 gene, six had ALS-FTD, two ALS-ECI and one was cognitively normal. One of the five patients with SOD1 mutations and one of the five patients with TARBDP mutation had ALS-Bi. Both patients with FUS and OPTN mutations were cognitively normal. Genetic status was significantly correlated to the presence/absence of cognitive impairment (p=0.0001).

Survival analysis

The overall median survival time was 2.7 years (95% CI 2.4 to 2.9) (figure 2). Patients with ALS-FTD had a significantly shorter survival (1.9 years, 95% CI 1.7 to 2.2) than any other group of patients with cognitive impairment, with the only exception of those with ALS-NECI (2.0 years, 95% CI 1.6 to 2.4). Patients with ALS-Bi (3.0 years, 95% CI 0.8 to 5.3) had a survival time similar to that of cognitively normal patients (3.1 years, 95% CI 2.7 to 3.4). Patients with ALS-ECI had an intermediate survival between the two groups (2.6 years, 95% CI 2.0 to 3.1). Cognitive status remained significant in Cox multivariable analysis (table 3). The presence of FTD significantly increased the risk of death compared with non-demented patients; also ALS-NECI and ALS-ECI resulted to be independently related to a worse outcome.

Table 3

Cox’s multivariable analysis

Figure 2

Survival curves from disease onset to death/tracheostomy of the incident amyotrophic lateral sclerosis (ALS) cohort according to their cognitive classification; p=0.004. Ticks are censored patients. Red, patients with ALS with normal cognition; green, patients with ALS with comorbid frontotemporal dementia (ALS-FTD); yellow, patients with ALS with executive cognitive impairment (ALS-ECI); violet, patients with ALS with non-executive cognitive impairment (ALS-NECI); blue, patients with ALS with behavioural impairment (ALS-Bi); black, patients with ALS with non-classifiable cognitive impairment (ALS-NCCI). The single patient with ALS with comorbid dementia of Alzheimer's type is not included.


We have studied cognitive status in a population-based series of patients with ALS in Italy using an extensive battery of tests evaluating multiple cognitive domains. In our series, 13% of patients had a comorbid FTD, while 50% had normal cognition. The remaining patients who did not meet the criteria for FTD, but otherwise had some clinical significance, including a negative effect on disease outcome, showed various degrees of cognitive impairment.

The frequency of cognitive impairment in our epidemiological series was similar to that described in Irish patients.4 However, differently from that study, according to ALS-FTD Consensus Criteria5 we identified a group of patients with cognitive impairment, that is, patients with isolated behavioural impairment, accounting for 6% of cases. These patients did not show impairment in more than one executive or one non-executive test, but had a behavioural impairment at extensive clinical observation and at the Frontal Systems Behavior Scale test. Interestingly, one control patient also met the criteria for cognitive behavioural impairment.

We also identified a group of patients (6% of our series) (NCCI) with impairment in one executive and/or one non-executive test who did not fulfill the criteria for other cognitive groups. These patients largely differed from cognitively normal patients and from all other cognitive subgroups, being younger, less frequently bulbar, and with a higher mean education level. It is possible that this group includes premorbid FTD cases, that is, patients who did not meet the criteria for other cognitive impairments but who could have developed more severe impairment later in the course of the disease.

In our series, patients with ALS-FTD with full-blown comorbid dementia had a significantly lower educational level, in keeping with another population-based study.4 The lower mean educational level in Italian patients and controls in this series compared with that of the Irish study4 reflects the low level of education in the Italian population born before 1950.18 Educational level, as well as higher occupation attainment, are considered proxies of cognitive reserve.19 The role of cognitive reserve in protecting from AD is widely accepted,19 although the underlying mechanisms are still unclear. Cognitive reserve is also involved as a protective mechanism in several cognitive functions impaired in FTD, in particular speed of processing/executive functioning, visual spatial abilities and verbal memory.20–22 Our finding suggests that either a long-standing frontal dysfunction interferes with learning and might underline the future development of cognitive impairment or low education level puts patients at higher risk of developing FTD. Differently from patients with ALS-FTD, those with ALS-ECI, ALS-NECI and ALS-Bi did not differ from normal controls regarding educational level, and those with ALS-NCCI had a higher educational level than other cognitive groups and controls. This finding may indicate that either cognitive reserve does not have a role in these variants of cognitive impairments in ALS, or that some patients develop cognitive impairment not meeting the full criteria for FTD because they are protected by their cognitive reserve.

Patients with ALS-FTD and ALS-ECI had an older age at onset than controls, ALS-Bi and ALS-NECI, in keeping with various papers,3 ,23 but not all.4 ,24 This difference may be due to the higher mean age of our patients compared with other series.4 ,24

In our series, bulbar onset was significantly more frequent in ALS-FTD and ALS-NECI. Bulbar onset has been found to be more commonly related to FTD features in several series,3 ,25 ,26 but not in all.4 ,23 Supporting our findings, a [18F]2-fluoro-2-deoxy-D-glucose (FDG) Positron Emission Tomography (FDG-PET) study showed a significantly higher relative decrease in metabolism in large frontal and parietal regions in bulbar onset patients compared with spinal ones.27

In this study a genetic characterisation of all patients with ALS was performed. C9ORF72 hexanucleotide repeat expansion was the more frequent mutation, and, as expected,1 ,28 it was also the one significantly associated with FTD compared with other gene mutations or no genetic mutations. However, patients with FTD with C9ORF72 mutation accounted only for a fourth of all cases with ALS-FTD, indicating that other genetic, epigenetic or environmental mechanisms underlie the involvement of prefrontal cortex in ALS. The role of still unknown genes is supported by the fact that ALS-FTD was more commonly related to a positive family history of ALS than all other cognitive conditions.

Cognitive impairment has a strong negative impact on ALS outcome.4 ,14 ,26 ,29 ,30 The survival of our patients with ALS-FTD and ALS-ECI was about 1 year shorter than that of cognitively normal, patients with ALS-Bi and ALS-NECI. The reason of this finding is still not completely understood. The presence of neurobehavioral dysfunction or of isolate dysexecutive behaviour in ALS at diagnosis has been found to be a strong predictor of a poor outcome, partially related to a reduced efficacy of life-prolonging therapies such as non-invasive ventilation and percutaneous endoscopic gastrostomy,14 while the decline in cognitive function was faster in patients who were cognitively impaired at baseline.31 However, we could not find any significant correlation between ALS progression, evaluated with ALSFRS-R at the time of the interview, and patients’ cognitive status, indicating that the shorter survival of patients with ALS with cognitive impairment is not completely explained by the progression rate of their motor impairment. Cox multivariable analysis confirmed that cognitive status was independently related to ALS outcome.

A limitation of this study is that it is based on a single observation shortly after the diagnosis of the disease. However, at least two series with a follow-up cognitive assessment in patients with ALS found that an onset of FTD or other forms of cognitive impairment is rare during the disease course.31 ,32

In this study of cognitive status of incident Italian patients with ALS, the frequency of cognitive impairment was similar to that reported by a population-based study performed in Ireland,4 despite the different genetic backgrounds of the two populations,1 ,2 that is, the higher frequency of C9ORF72 mutations in Ireland, and of SOD1 and TARDBP mutations in Italy. We found that ∼15% of patients had ALS-FTD and another 35% had some degree of cognitive impairment. Comorbid FTD was associated with higher age at onset, bulbar onset and lower educational level, likely to represent a proxy for a reduced cognitive reserve, and has a significantly reduced survival than any other cognitive group. It remains to be understood whether ALS-ECI and ALS-NECI represent incomplete forms of cognitive impairment or discrete cognitive syndromes within the spectrum of ALS and FTD, with strong effect on the disease outcome.


The authors thank patients and controls for having collaborated with this study. This work was in part supported Ministero della Salute (Ricerca Sanitaria Finalizzata, 2010, grant RF-2010-2309849) and European Community's Health Seventh Framework Programme (FP7/2007-2013 under grant agreement 259867).


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Contributors Study concept and design: AM, Andrea C, LL, Antonio C, Adriano C. Acquisition of data: AM, BI, Andrea C, CM, MB, IO, SC, Adriano C. Analysis and interpretation of data: AM, Andrea C, GR, ALP, Adriano C. Drafting of the manuscript: AM, Adriano C. Critical revision of the manuscript for important intellectual content: AM, Andrea C, CM, LL, GR, ALP, Antonio C, Adriano C. Obtained funding: Adriano C. Administrative, technical and material support: BI, CM, MB, IO, SC. Study supervision: AM, Adriano C. Adriano C had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. All authors have approved the submitted version of the paper.

  • Funding Ministero della Salute (Ricerca Sanitaria Finalizzata, 2010, grant RF-2010-2309849) and European Community's Health Seventh Framework Programme (FP7/2007-2013 under grant agreement 259867).

  • Competing interests Andrea C has received research support from Italian Ministry of Health (Ricerca Finalizzata). LL has received research support from Italian Ministry of Health (Ricerca Finalizzata). GR has received research support from Italian Ministry of Health (Ricerca Finalizzata) and Regione Piemonte (Ricerca Finalizzata). Adriano C serves on the editorial advisory board of Amyotrophic Lateral Sclerosis and has received research support from Italian Ministry of Health (Ricerca Finalizzata), Regione Piemonte (Ricerca Finalizzata), University of Torino, Federazione Italiana Giuoco Calcio, Fondazione Vialli e Mauro onlus, and European Commission (Health Seventh Framework Programme); he serves on scientific advisory boards for Biogen Idec and Cytokinetics.

  • Patient consent Obtained.

  • Ethics approval Comitato Etico Ospedale San Giovanni Battista di Torino.

  • Provenance and peer review Not commissioned; externally peer reviewed.