Extrapyramidal and cognitive signs in amyotrophic lateral sclerosis: A population based cross-sectional study.

Our objective was to assess the association between amyotrophic lateral sclerosis (ALS) and other neurodegenerative diseases such as Alzheimer's disease (AD), frontotemporal dementia (FTD) and Parkinson's disease (PD). From May 2007 through August 2012 we investigated 146 patients with newly diagnosed ALS and 146 age- and gender-matched controls. Each individual was screened for cardinal extrapyramidal signs (neurological examination) and cognitive dysfunction (Mini Mental State Examination, MMSE and Frontal Assessment Battery, FAB). Results demonstrated that rigidity was present in 8.2% of cases and 2.1% of controls (adjusted odds ratio, adjOR 5.7; 95% CI 1.5–22.0). The corresponding percentages for bradykinesia and postural instability were, respectively, 8.2 vs. 2.7% (adjOR 4.8; 95% CI 1.4–16.5) and 2.7 vs. 9.6% (adjOR 0.3; 95% CI 0.1–0.9). FAB ≤ 13.4 was recorded in 24.8 vs. 9.6%; adjOR 2.9; 95% CI 1.5–5.7). Tremor and abnormal FAB score were predicted by an older age at onset while an abnormal FAB score was associated with cramps and family history of neurodegenerative diseases. In conclusion, our data support the notion that newly diagnosed ALS carries a higher than expected risk of extrapyramidal signs and FTD.

and striatum in ALS. However, the possible association with Parkinson's disease (PD), except for the ALS/PDC of Guam (11), is still based on few case reports or small series from referral centres (12)(13)(14), and data are inconsistent on the association with Alzheimer's disease (AD), mostly reflecting the differing validity of the screening instruments (15). With few exceptions (1,4,16), the ALS-plus syndrome has been investigated in prevalent populations and the results cannot be applied to incident ALS cohorts that better represent the full disease spectrum. In addition, in the absence of matched controls, the possibility cannot be excluded that extrapyramidal signs and cognitive dysfunction are present in some of these elderly patients as a reflection of the aging process. Against this background, we planned a population-based matched cross-sectional study in representative samples of incident ALS and normal individuals. The research hypothesis was that newly diagnosed ALS carries a higher than expected risk of cardinal extrapyramidal signs and/or cognitive dysfunction, with special reference to the executive functions.

Material and methods
The study design and reporting was in line with the STROBE guidelines for observational studies (17). Informed consent was obtained from each study participant.

Population
The study population was represented by patients with newly diagnosed ALS included in two population based registries located in northern Italy ( Figure 1). Case ascertainment in both registries is virtually complete (18,19). During the study period (May 2007 through August 2012), a random sample of all registered patients was selected for this study. The baseline characteristics of the sample compared to the origin ALS populations are illustrated in the Supplementary Tables 1 and 2 to be found online (20). For each patient, a control was selected from the lists of local general practitioners (GPs), matched for age ( 2.5 years), gender and province of residence.

Data collection
After obtaining an informed consent, a board certified neurologist or a GP with neurological training performed the medical interviews and neuropsychological testing with the Mini Mental State Examination (MMSE) (21) and the Frontal Assessment Battery (FAB) (22) to each case and control. The GPs (two in total) received a video with the details of a neurological examination and proper instruction on how to perform neuropsychological testing. The neurological examination was performed in search of cardinal extrapyramidal signs (bradykinesia, tremor, rigidity, postural instability), and other motor findings and signs suggesting an ALS mimic. For ALS patients, details were collected on the EE diagnostic category, site of onset of symptoms (bulbar, spinal), duration of symptoms at diagnosis, fasciculations, cramps, other motor findings, and treatment with riluzole. The following data were recorded in a structured questionnaire: date of enrolment, date of birth, gender, years of education, family history of any neurodegenerative disease among AD, PD, FTD and ALS, family history of each disease separately, cardinal extrapyramidal signs, MMSE and FAB scores. Data were entered for analysis in a password protected web-database with personal login. The following analyses were planned in the study protocol: 1) number and proportion of cases and controls with any of the extrapyramidal signs and with each separate sign; 2) number and proportion of cases and controls with abnormal neuropsychological tests (corrected MMSE score,  24; corrected FAB score,  13.4) (23); 3) number and proportion of cases with family history of AD, PD, FTD, ALS; 4) ALS demographics (age and gender) and clinical findings (age at onset, site of onset, EE diagnostic category, family history of neurodegenerative diseases), and extrapyramidal signs; 5) ALS demographics and clinical findings, and MMSE and FAB scores.

Statistical analysis
Descriptive statistics are reported as count and percentage or median and range/interquartile range (IQR). Bivariate analyses were performed using Fisher's exact test and Wilcoxon-Mann Whitney test as appropriate. Five conditional logistic models were used to compare the risk of ALS patients vs. controls in the occurrence of tremor (model I), bradykinesia (model II), rigidity (model III), postural instability (model IV) and at least one of the four cardinal signs (model V). All models were adjusted for age, gender, years of education. MMSE and FAB were also dummied (normal vs. abnormal) and used as dependent variables in conditional logistic models VI and VII. Abnormal scores were based on normative values obtained in the Italian population (24,25). ALS patients were also tested to assess any possible relation between the demographic and clinical variables and extrapyramidal signs, MMSE and FAB scores using multivariable (logistic regression) models. Variables to be maintained in the multivariable models were selected using a stepwise selection method. A 5% significance level was required for a variable to enter and stay in the model. The results of multivariate analyses are presented as adjusted odds ratios (adjOR) with 95% confidence intervals (95% CI). Given the exploratory nature of this investigation, the sample size was not predetermined. All analyses were performed using the SAS statistical analysis system version 9.2 (SAS Institute Inc., Cary, NC, USA).
ALS patients and controls reporting cardinal extrapyramidal signs and abnormal MMSE and FAB scores are illustrated in Table II. None of the cases and controls received treatment for extrapyramidal signs, or for cognitive dysfunction because these signs were mild and did not interfere with daily living activities. Significant differences were observed for rigidity (adjOR 5.7; 95% CI 1.5-22.0), bradykinesia (adjOR 4.8; 95% CI 1.4-16.5), postural instability (inverse correlation: adjOR 0.3; 95% CI 0.1-0.9), and abnormal FAB score (adjOR 2.9; 95% CI 1.5-5.7). An abnormal MMSE score was reported in similar proportion in cases and controls.
In evaluating the ALS group, the occurrence of fasciculations was inversely correlated with having at least one cardinal symptom of PD (adjOR 0.2; 95% CI 0.1-0.7) (Table III). Age at ALS onset was All the other variables were unremarkable.

Discussion
Our population based case-control study suggests that patients with newly diagnosed ALS are at higher than expected chance to report extrapyramidal signs and to present impairment of executive functions. The odds are almost five-fold for bradykinesia and six-fold for rigidity. The odds of having an abnormal FAB score are three-fold. Each additional year of age corresponds to a 10% increased risk of tremor and/ or to perform poorly when tested with the FAB scale. The odds to present an abnormal FAB score was almost five-fold and six-fold in patients with cramps and in those reporting a family history of neurodegenerative diseases, respectively. Our findings on the impairment of executive functions are in keeping with other reports (4,26,27). In a comprehensive review of the literature, Ravits et al. (2) confirmed a partial overlap between ALS and FTD as up to 15% of FTD patients and 30% of ALS patients experience symptoms and signs suggestive of both clinical conditions. The overlap between ALS and FTD has a strong genetic basis, represented by the C9orf72 gene expansion which is known to cause familial ALS-FTD syndrome and accounts for about 10% of sporadic ALS (28).
Our findings on the increased risk of extrapyramidal signs are supported by few other observations. In a large cohort of patients with ALS from an integrated clinical database, 13.6% of cases were diagnosed with ALS-plus syndrome (16). Extrapyramidal signs and cognitive impairment were present in 22.7% and 8.0%, respectively, of cases. The only other study assessing the possible association between ALS and Parkinsonism in a clinical series was that of Wolf Gilbert et al. (12) who reviewed the Columbia Movement Disorder database of about 5500 PD patients and identified 27 patients with both Parkinsonism and ALS. Three of them were also found to have dementia. Bradykinesia, rigidity and postural instability were present in nearly all cases. However, the present study cannot be strictly compared with the U.S. report because our target population was represented by patients with ALS, the number of individuals fulfilling the diagnosis of PD (i.e. at least two of the four cardinal signs) was similar among cases and controls (6.8 vs. 4.8%, p  0.45), and an active search was made of each extrapyramidal sign. A family history of neurodegenerative diseases was associated with an increased risk of reporting abnormal executive functions in our ALS patients. Our data are in keeping with other epidemiologic and genetic studies supporting common pathophysiologic mechanisms for ALS, AD and PD. Involvement of the extrapyramidal system was more prevalent in 134 patients with probable AD than in 167 controls (29). Evidence for a shared genetic susceptibility to the three neurodegenerative diseases was found in a familial aggregation study on a hospital based cohort (30). However, more recent work from a population based case-control family aggregation study did not support this observation (31). Neurodegenerative diseases have also in common a number of exogenous risk or protective factors, including low education, trauma, pesticide exposure, diet, coffee, smoking, occupation, exercise, and drug exposure (32,33).
One of the mechanisms linking ALS to PD is an increased frequency of variants in the gene encoding angiogenin (ANG), a strong neuroprotective factor (34). Isolated TDP-43 disorders were found to be associated with an ALS-plus syndrome with extrapyramidal features (9,35).
Except for age at onset, we found no association between demographic and clinical features in ALS patients and extrapyramidal signs, cognitive dysfunction and family history of AD, PD, FTD or ALS separately. Given the small number of cases and controls, our study may be underpowered to detect minor differences.
In addition, we did not find ALS-emergent abnormal MMSE scores. The use of this screening measure, which is fairly inaccurate in detecting mild cognitive impairment, might lead to under-ascertainment of cases with altered cognition. Surprisingly, we found an inverse correlation between ALS and postural instability, one of the cardinal PD signs. A chance finding cannot be excluded. However, postural instability is generally observed in advanced PD and can be caused by several clinical conditions other than PD in elderly individuals. We did not investigate the cause of balance disorders in our controls, but old age, musculo-skeletal disorders and vertigo were frequently reported in their medical history.
Our study has strengths and limitations. Our major strength is the population base coupled with the representativeness of the inception cohort. The prevalence of family history of ALS in our sample corresponds to that of other Italian population-based studies (36,37). The study of newly diagnosed ALS patients favours the identification of other neurological signs early in the course of the disease and, in this regard, can help assessing comorbid disorders when the disease is not so advanced as to mask independent symptoms and signs. The second strength is the inclusion of strictly matched controls in order to disentangle symptoms and signs attributable to the disease from coincidental age-related findings. The third strength is the direct screening rather than a retrospective search of extrapyramidal signs and cognitive functions. The first limitation is the cross-sectional assessment of neurological signs. In the absence of a follow-up, we cannot exclude the possibility that PD and dementia might occur later in the course of the disease. However, the functional impairment caused by the disease may mask extrapyramidal signs and the ability to comply with cognitive testing. The second limitation is the use of very simple scales to investigate executive functions. However, the similarity of our findings with other, more accurate reports and the validity of FAB, compared to other neuropsychological tests measuring frontotemporal functions (38), supports the value of our screening method. The third limitation is the limited sample size. Some associations may have thus gone undetected.
In conclusion, our data confirm a possible link between ALS, extrapyramidal disorders and impairment of executive functions. Whether this link is the result of a specific genetic susceptibility is a research hypothesis to be tested by future genetic studies. grant from the Emilia Romagna Regional Health Authority. E. Beghi has received payment for board membership by VIROPHARMA and EISAI; has received funding for travel and speaker honoraria from UCB-Pharma, GSK and also for educational presentations from GSK; has received grants for research activities from the Italian Drug Agency, Italian Ministry of Health, Sanofi-Aventis and the American ALS Association.