Abstract
Background: The prognostic implications of cognitive impairment in amyotrophic lateral sclerosis (ALS) are not established.
Objectives: To investigate the survival effect of the comorbid frontotemporal dementia (FTD) and to determine whether, in the absence of dementia, impairment in different cognitive domains affects outcome.
Methods: A prospective population-based study of incident cases of ALS in the Republic of Ireland included home-based neuropsychological assessments using age-, sex-, and education-matched controls. Four cognitive domains were evaluated: executive function, memory, language, and visuospatial skills.
Results: Mean age of the participants (n = 139) was 63.3 years; 61.2% were male and 35.3% had bulbar-onset ALS. Factors associated with shorter survival included age more than 60, severe disability at baseline, shorter delay to diagnosis, and early respiratory involvement. Comorbid FTD was associated with significantly shorter survival time (hazard ratio [HR] 2.67, 95% confidence interval [CI] 1.04–6.85, p = 0.041). In patients with ALS without dementia, the presence of executive dysfunction was significantly associated with shorter survival. This was confirmed in a multivariate model that included age, delay to diagnosis, disease severity at baseline, education, and respiratory status (HR 3.44, 95% CI 1.45–8.18, p = 0.005). In the absence of executive dysfunction, single or multi-domain impairment in other cognitive domains had no significant effect on survival.
Conclusion: Comorbid frontotemporal dementia is a negative prognostic indicator. In patients with ALS without dementia, executive dysfunction, but not impairment in other cognitive domains, is an important negative prognostic indicator.
Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disorder that targets the motor neurons, leading to progressive paralysis of the voluntary muscles and ultimately death.1 The rate of disease progression among patients is variable.2,3 Identification of the key factors that can influence outcome is important for effective timing of medical interventions4,5 and for appropriate stratification in clinical trials.4,6 The most frequently reported negative prognostic indicators in ALS include older age at onset, bulbar onset, and short delay to diagnosis.3,7,–,10
It is now recognized that the clinical phenotype of ALS can extend beyond the motor neurons. Cortical involvement in ALS can lead to frontotemporal syndromes that that range from mild cognitive or behavioral changes to frontotemporal dementia (FTD).1,11
There have been relatively few investigations of the effect of cognitive impairment on survival in patients with ALS.5,12,–,14 Although results are conflicting,12,–,14 it is generally accepted that comorbid dementia is associated with poorer prognosis. The prognostic implications of milder cognitive impairment remain to be established. The aims of this study are to examine the survival effect of comorbid FTD in a population-based cohort of patients with ALS and to determine whether in the absence of dementia, impairment in different cognitive domains influences clinical outcome.
METHOD
All incident patients in the Republic of Ireland diagnosed with ALS between November 2006 and May 2010 were identified through the Irish Register for ALS/MND, and invited to participate in a longitudinal population-based study of cognitive status in ALS. The Irish ALS Register has been in operation since 1995 and has been previously described.15,16
All patients included in the study were diagnosed with possible, probable, or definite ALS according to the revised El Escorial criteria.17 Exclusion criteria included history of other neurologic conditions that could affect cognition (major hemispheric stroke, traumatic brain injury, learning disability, severe active epilepsy), alcohol dependence syndrome, type I or uncontrolled type II diabetes mellitus, permanent invasive ventilation, serious active mental illness, or current use of high-dose psychoactive medication.
Home-based semi-structured interviews were conducted to collect the demographic and clinical data and to carry out a detailed neuropsychological assessment lasting approximately 2 hours. Home visits were repeated at 6-month intervals until death of patients. The neuropsychological battery (table e-1 on the Neurology® Web site at www.neurology.org) was designed to evaluate 4 cognitive domains: attention/executive function, memory, and language and visuospatial skills. Premorbid full-scale IQ (FSIQ) was estimated using Wechsler Test of Adult Reading (WTAR).18
Disease severity was assessed at each visit using the revised ALS Functional Rating Scale (ALSFRS-R).19 Mood was assessed using the Hospital Anxiety and Depression Scale (HADS).20 Sniff nasal inspiratory pressure (SNIP) values were used to assess respiratory function. Arterialized capillary blood gas tensions were documented during the visits using a mobile transcutaneous sensor (TOSCA 500, Radiometer Ireland Ltd.).
Neuropsychological performance in patients with ALS was compared to that of a cohort of age- and sex-matched healthy controls (n = 116) comprising spouses of the patients and healthy volunteers recruited through the patients' primary health care providers.
The neuropsychological assessments were as carried out by 4 appropriately trained researchers under the direction of a senior neuropsychologist (N.P.). Assessments of inter-rater reliability were performed on a regular basis. Case ascertainment was supervised by a senior ALS specialist (O.H.) and a senior neuropsychologist (N.P.).
Comorbid frontotemporal dementia was defined as a behavioral disorder that fulfilled the Neary criteria for frontotemporal lobar degeneration.21 Information on behavior was obtained via direct assessment of the patient, an interview with the carer, and use of the Frontal Systems Behavior Scale.22
Patients with ALS in whom the diagnosis of dementia was excluded were categorized into 3 groups according to their neuropsychological performance in various cognitive domains (see table e-2). This domain-based categorization is based on the current literature suggesting early and dominant involvement of executive frontal function in patients with ALS.1,11
Statistical analysis.
Demographic and clinical characteristics of the participants were documented as percentages for categorical variables and means/medians for continuous variables. Comparisons were made using χ2 test, 2-sample t test, one-way analysis of variance, Kruskal Wallis test, or Mann-Whitney test as appropriate.
Survival time was defined as time from symptom onset to time of death. Patients who were alive at time of analysis were censored. Univariate assessment of the survival effect of categorical variables was carried out Kaplan-Meier survival analysis and equality of outcome was assessed using the log-rank test. Cox proportional hazards method was used for univariate survival analysis of continuous variables, estimation of hazard ratios (HR), and 95% confidence intervals (CI). Cox proportional hazards regression was also used for multivariate analysis using backward elimination (Wald test). In the Cox models, cognitive category was entered as a categorical variable and all covariates (e.g., age, ALSFRS-R scores) were entered as continuous variables.
All tests were 2-tailed and statistical significance was set at p < 0.05. Statistical analyses were carried out using SPSS version 17 (SPSS Inc., Chicago, IL).
Standard protocol approvals, registrations, and patient consents.
Written informed consent was obtained from all the patients who participated in the study. The study had full ethical approval from Beaumont Hospital Research Ethics Committee.
RESULTS
In the period extending from November 1, 2006, to May 1, 2010, 307 incident patients with ALS were ascertained in the Republic of Ireland. A total of 154 incident cases were recruited (figure 1). Fifteen patients were subsequently excluded. Reasons for exclusion included history of major hemispheric stroke (n = 6), alcohol dependency (n = 5), major psychiatric illness (n = 2), permanent invasive ventilation (n = 1), and uncontrolled diabetes mellitus (n = 1).
ALS = amyotrophic lateral sclerosis.
Baseline demographics of cohort (n = 139).
The mean age at time of assessment was 63.3 years (range 34.7–84.7 years) and 61.2% were men. Mean number of years of education was 11.9 (range 2–21).
Bulbar onset disease was documented in 49 patients (35.3%) while respiratory onset was documented in 3 patients (2.2%). The remaining patients (n = 87, 62.6%) had spinal onset ALS. Median time from symptom onset to diagnosis was 10 months (range 1–49). The median ALSFRS-R score at time of assessment was 39 (range 12–48). Evidence of familial onset ALS was documented in 9 patients (6.5%), of whom 2 patients were siblings.
The majority of patients (87.1%) were taking riluzole. At time of assessment, 17 patients (12.2%) had an enteral feeding tube in situ and 12 patients (8.6%) were using noninvasive ventilation (NIV). Blood gases levels were documented at time of assessment in 134 patients (96.4%). None of the patients had evidence of hypoxemia (oxygen saturation less than 92 mm Hg).
Initial exploratory survival analysis.
At time of analysis (October 2010), 88 patients had died (63.3%). The median survival time of the cohort from symptom onset was 27 months. Median survival from time of diagnosis was 14 months.
Univariate analysis was undertaken to explore the survival effect of the following demographic and clinical characteristics: age at symptom onset, gender, site of onset, family history of ALS, education (measured in number of years), severity at baseline estimated using ALSFRS-R score, time from symptom onset to diagnosis, carbon dioxide partial pressure at time of assessment, SNIP values, use of NIV (in patients with SNIP values less than 50, n = 30), riluzole use, and presence of an enteral feeding tube.
Factors that were associated with a significantly shorter survival time on univariate analysis included the following:
-
Age at symptom onset of 60 years or more (HR 1.74, 95% CI 1.12–2.69, p = 0.012)
-
Lower ALSFRS-R score at baseline (HR 0.95, 95% CI 0.92–0.97, p < 0.0001)
-
Shorter time delay to diagnosis (HR 0.94, 95% CI 0.91–0.96, p < 00001)
-
Higher carbon dioxide level at time of assessment (HR 1.51, 95% CI 1.15–2.00, p = 0.003)
-
Lower SNIP values (HR 0.97, 95% CI 0.96–0.99, p < 0.0001)
Correlation analysis using Spearman rank test showed only weak–moderate associations between the above variables and thus they were included in all multivariate models.
Factors that had no significant effect on survival on univariate analysis included the following: gender (p = 0.354); family history of ALS (p = 0.718); education (p = 0.676); riluzole use (p = 0.542); enteral feeding tube (p = 0.170); and NIV use (p = 0.121). Although respiratory onset and bulbar onset patients with ALS had shorter median survival time compared to spinal onset patients (17 months and 30 months respectively compared to 36 months), the difference did not reach statistical significance (p = 0.093).
Frontotemporal dementia.
Three patients (2.2%) had evidence of comorbid Alzheimer-type dementia based on DSM-IV criteria. Twenty patients (14.4%) fulfilled the Neary criteria for frontotemporal dementia (ALS-FTD).
When patients with comorbid FTD were compared to patients with ALS without evidence of dementia (n = 116), there was no significant difference in mean age at symptom onset (p = 0.122), gender (p = 0.521), site of onset (p = 0.729), SNIP values (p = 0.414), carbon dioxide level at time of assessment (p = 0.973) or education measured in years (p = 0.333).
On univariate analysis, median survival time was shorter in patients with ALS with comorbid FTD (n = 20, 23 months, 95% CI 14.51–31.49) compared to patients with ALS without dementia (n = 116, 34 months, 95% CI 30.54–37.46, p = 0.026, figure 2).
Log-rank test for equality of survival functions, p = 0.026. Black line: ALS with comorbid frontotemporal dementia; dotted line: ALS with no evidence of dementia; +: censored cases.
The negative effect of comorbid FTD on survival persisted in a multivariate model that included age at symptom onset, delay to diagnosis, ALSFRS-R score at baseline, carbon dioxide levels, and SNIP values (HR 2.67, 95% CI 1.04–6.85, p = 0.041).
Nondementia cognitive impairment.
A total of 116 (83.5%) patients with ALS had no evidence of dementia according to DSM-IV or Neary criteria. Of these, 3 patients underwent fewer than 2 tests of executive function and were thus excluded from further categorization.
The baseline characteristics of the remaining cohort (n = 113) are listed in table 1. This cohort was divided into 3 categories (table e-2) based on neuropsychological performance compared to age- and sex-matched healthy controls:
-
Patients with deficits in executive functioning (29 patients, 20.9%)
-
Patients with no executive dysfunction but impairment in one or more other cognitive domains (21 patients, 15.1%)
-
Patients in whom no abnormality was detected (63 patients, 45.3%)
Baseline demographics of patients with ALS without dementia compared to healthy controls
Patients with executive dysfunction were older at symptom onset (p = 0.021) and had significantly lower SNIP values (p = 0.006). Patients with ALS who had normal cognition had higher number of years of education compared to the remaining 2 groups (p < 0.0001). There was no significant difference among the 3 groups with respect to gender (p = 0.248), site of onset (p = 0.185), ALSFRS-R score at baseline (p = 0.081), or oxygen or carbon dioxide level at time of assessment (p = 0.516 and p = 0.333, respectively).
Median survival time in patients with executive dysfunction (n = 29, 24 months, 95% CI 17.68–30.32) was shorter than that of patients without executive dysfunction (nonexecutive cognitive impairment or no abnormalities, n = 84, 38 months, 95% CI 30.77–45.23, log-rank test p < 0.0001, figure 3). This was confirmed in a multivariate model which included age at symptom onset, ALSFRS at baseline, delay to diagnosis, and carbon dioxide levels and SNIP values (HR 3.44, 95% CI 1.45–8.18, p = 0.005, table e-3).
Log-rank test for equality of survival functions, p < 0.0001. Black line: patients with executive dysfunction (n = 29); dotted line: patients without executive dysfunction (n = 84); +: censored cases.
No significant difference was observed in median survival time between the cohort of patients with nonexecutive cognitive impairment (n = 21, 34 months, 95% CI 26.58–41.42) and the cohort with normal cognition (n = 63, 39 months, 95% CI 27.73–50.28, log-rank test, p = 0.377, figure 4).
Log-rank test for equality of survival functions, p = 0.377. Black line: patients with nonexecutive cognitive impairment (n = 21); dotted line: patients in whom no abnormality was detected (n = 63); +: censored cases.
DISCUSSION
Our findings confirm and extend those of previous survival studies in patients with ALS-FTD. A previous report12 suggested that patients with ALS-FTD had significantly shorter survival compared to patients with normal cognition after the exclusion of patients with milder cognitive impairment. In our study we compared survival in patients with ALS-FTD to patients without dementia with ALS regardless of their cognitive status. The observed negative effect of comorbid FTD on survival was preserved using multivariate model that adjusted for age, disease severity, delay to diagnosis, and respiratory status.
A recent study reported that, in the absence of dementia, moderate but not mild cognitive impairment had a significant negative effect on survival.5 Cognitive status was estimated using a global deficit score, calculated using z scores from multiple neuropsychological tests. None of the patients had severe cognitive impairment or dementia. The study was also limited by absence of a control population, and by the recruitment of patients through tertiary referral centers in contrast to our population-based design.
Our findings suggest that, in patients without dementia, the presence of executive dysfunction is an important negative prognostic indicator. This suggests that the pattern of cognitive impairment is as important as the degree of impairment. The presence of frontotemporal cortical involvement in the form of dementia or executive-frontal deficits may be a marker for a more aggressive, possibly distinct form of ALS.
The concept of ALS-FTD representing a distinct entity with poorer prognosis is supported by a recent report of significantly shorter survival in these patients compared to sporadic behavioral variant frontotemporal dementia, despite the similarities in neurobehavioral profile.23
Our findings seem to contrast with 2 previous studies that reported that cognitive impairment had no effect on disease evolution.13,14 However, the neuropsychological battery utilized in one study13 was not sensitive to executive dysfunction and the diagnosis of dementia was heavily reliant on memory dysfunction. In the second study,14 cognitive impairment was defined as impairment in at least 2 cognitive domains. Cognitive impairment was not subcategorized with respect to executive dysfunction in either study. This may explain the negative results of these studies, as our observations suggest that impairment in 2 or more cognitive domains (such as memory or language), in the absence of executive dysfunction, has no significant effect on survival.
The demographics of our cohort were comparable to those of the incident cohort in our previously published survival study16 based on the Irish ALS register, a finding that is consistent with the population-based design of both studies. The negative effect on survival of older age, more severe disease at diagnosis, and shorter delay to diagnosis have been previously reported by other groups.2–3,7,9,12,14 Although bulbar onset is classically considered an independent negative prognostic indicator, this is not a uniform finding in the literature.7,14,24,–,28
Frontotemporal cortical involvement is likely to be associated with poor compliance with medical intervention.12 While we cannot exclude the possibility that patients with FTD and those with executive dysfunction may have experienced a shorter survival because of reduced compliance with interventions such NIV, our data would suggest that the disease trajectory on these patients was also more rapid, as patients with executive dysfunction had significantly lower SNIP values. It is unlikely that the poor performance on executive tests in these patients was a consequence of poor respiratory function as there was no significant difference among the 3 cognitive groups in oxygen or carbon dioxide partial pressures at time of assessment. Moreover, in a multivariate analysis that included SNIP values, executive dysfunction remained an independent negative prognostic indicator.
Age and education are considered important predictors of cognitive reserve and function including speed processing and executive function.29 This may partially explain the older age and lower educational attainment in patients with executive dysfunction. However, older age is also the single most important risk factor for development of dementia in PD.30,31 It is conceivable that the interlinked pathomechanisms of age-related neurodegenerative disorders combined with the vulnerability of the aging brain increases the susceptibility of older patients to more aggressive forms of the neurodegenerative process.
Differences in phenotype and survival between incident and prevalent populations have previously been reported by our group.16 If, as our data would suggest, executive function is a negative prognostic indicator, it is likely the frequency of executive dysfunction will be higher in an incident population than in a prevalent-based cohort. Accordingly, our observations may account for some of the differences in previously reported rates of cognitive impairment in ALS.
Our study has certain limitations. Although the original study cohort was large, the number of patients in individual groups (for example, patients with enteral feeding tube) was relatively small and this may have limited the power of the study. Our neuropsychological battery was designed to minimize the effects of motor disability. However, 3 tasks were timed. We adjusted for physical disability in the verbal fluency task and Stroop Color Word task but not in the case of category fluency. This may have resulted in lower scores in patients with bulbar dysfunction resulting in spuriously higher levels of impairment on this measure. Due to the limitations imposed by patient fatigue, the Boston Naming Test was the only task used to assess language function. Although this test has been shown to be sensitive to mild language impairment in the healthy elderly population32 and patients with early dementia,33 this may have reduced the scope of language deficits identified in our cohort.
Notwithstanding these limitations, our observation that executive dysfunction is an important prognostic indicator in ALS, if replicated in further population-based studies, should be considered in stratification of future clinical trials.
AUTHOR CONTRIBUTIONS
Statistical analysis was conducted by Dr. Marwa Elamin.
DISCLOSURE
Dr. Elamin receives research support from the ALS Association, the Health Research Board, and Research Motor Neurone. Dr. Phukan, Dr. Bede, N. Jordan, Dr. Byrne, and Dr. Pender report no disclosures. Dr. Hardiman has served on scientific advisory boards for Novartis, Biogen Idec, Allergan, Inc., Sanofi-Aventis, Ono Pharmaceutical Co. Ltd., and the Health Research Board, Ireland; has received speaker honoraria from Janssen, Biogen Idec, Sanofi-Aventis, and Merck Serono; serves as Editor-in-Chief of Amyotrophic Lateral Sclerosis and on the editorial board of Journal of Neurology Neurosurgery & Psychiatry; has patents pending re: Treatment of central nervous system injury and treatment of neurodegenerative disease; and receives research support from the ALS Association, the Health Research Board, and Research Motor Neurone.
Footnotes
-
Study funding: Supported by the ALS Association, the Health Research Board (Ireland), and Research Motor Neurone.
-
Supplemental data at www.neurology.org
-
- ALS
- amyotrophic lateral sclerosis
- ALSFRS-R
- ALS Functional Rating Scale
- CI
- confidence interval
- DSM-IV
- Diagnostic and Statistical Manual of Mental Disorders, 4th edition
- FSIQ
- full-scale IQ
- FTD
- frontotemporal dementia
- HADS
- Hospital Anxiety and Depression Scale
- HR
- hazard ratio
- NIV
- noninvasive ventilation
- SNIP
- sniff nasal inspiratory pressure
- WTAR
- Wechsler Test of Adult Reading
- Received September 16, 2010.
- Accepted December 22, 2010.
- Copyright © 2011 by AAN Enterprises, Inc.
REFERENCES
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
