Editor's Choice Article and Review
Gait patterns in parkinsonian patients with or without mild cognitive impairment†‡§
Marianna Amboni MD, PhD1,2, Paolo Barone MD, PhD3,*, Luigi Iuppariello MS4, Ilaria Lista PhD1, Riccardo Tranfaglia PhD4, Alfonso Fasano MD, PhD5, Marina Picillo MD6, Carmine Vitale MD, PhD1,2, Gabriella Santangelo PhD7, Valeria Agosti PhD1,2, Alessandro Iavarone MD, PhD8, Giuseppe Sorrentino MD, PhD1,2
Article first published online: 2 OCT 2012
1 University of Naples Parthenope, Naples, Italy
2 IDC Hermitage-Capodimonte, Naples, Italy
3 Neurodegenerative Diseases Center, University of Salerno, Salerno, Italy
4 Department of Biomedical Electronic Engineering, University of Naples Federico II, Naples, Italy
5 Department of Neurology, Catholic University of Sacred Heart, Rome and Department of Neuroscience, AFaR-Fatebenefratelli Hospital, Rome, Italy
6 Department of Neurological Sciences, University of Naples Federico II, Naples, Italy
7 Department of Psychology, Second University of Naples, Caserta, Italy
8 Neurological and Stroke Unit, CTO Hospital, AORN Ospedale dei Colli, Naples, Italy
*Neurodegenerative Diseases Center, University of Salerno, Via Allende, Baronissi (Salerno), Italy
†Funding agencies: This work was supported by a grant from MIUR (Progetto FIRB-MERIT, no. RBNE08LN4P).
‡ Relevant conflicts of interest/financial disclosures: Nothing to report.
§Full financial disclosures and author roles may be found in the online version of this article.
Although in recent years the relationship between cognition and gait in Parkinson's disease (PD) has received increasing attention, the specific connections between gait patterns and cognitive features are not fully understood. The objective of this study was to describe the gait patterns in patients affected by PD with or without mild cognitive impairment (MCI+ and MCI−, respectively). We also sought to find an association between gait patterns and specific cognitive profiles. Using a gait analysis system, we compared the gait patterns among MCI+ patients (n = 19), MCI− patients (n − 24), and age- and sex-matched healthy subjects (HS; n = 20) under the following conditions: (1) normal gait, (2) motor dual task, and (3) cognitive dual task. In PD patients, gait parameters were evaluated in both the off and on states. Memory, executive, and visuospatial domains were assessed using an extensive neuropsychological battery. Compared with MCI− PD and HS, MCI+ PD patients displayed reduced step length and swing time and impairment of measures of dynamic stability; these dysfunctions were only partially reversed by levodopa. We also found that dual-task conditions affected several walking parameters in MCI+ PD in the off and on states relative to MCI− PD and HS. Factor analysis revealed 2 independent factors, namely, pace and stability. The latter was strongly and directly correlated to the visuospatial domain. In conclusion, dysfunctions on specific gait parameters, which were poorly responsive to levodopa and highly sensitive to dual-task conditions, were associated with MCI in PD patients. Importantly, visuospatial impairment was strongly associated with the development of instability and more generally with the progression of PD.
© 2012 Movement Disorder Society
Over the last decade, the relationship between cognitive function and gait performances has received increasing attention. Gait is no longer considered merely an automated motor activity but an activity requiring executive function and attention as well as motivation and judgment of external and internal cues.1
Dysfunction in specific gait variables has been associated with an increased risk of cognitive decline and Alzheimer's disease.2 Furthermore, in a large community-based cohort,3 gait dysfunctions were reported to be frequent in older adults diagnosed with mild cognitive impairment (MCI).
Originally, the construct of MCI was conceptualized as the transitional state between normalcy and Alzheimer's disease.4 More recently, it has been used to identify a predementia state in patients with Parkinson's disease (PD).5,6
It has long been recognized that the “postural instability gait disorder” phenotype is associated with cognitive impairment in PD7 and that axial symptoms such as gait disorders, postural instability,8,9 and cognition10 worsen as PD progresses. Moreover, these symptoms respond poorly to dopaminergic treatment, which may reflect the involvement of neurotransmitters systems other than dopamine.11,12 Several studies13–16 have assessed the levodopa effect on locomotion components with relatively inconsistent results, thus suggesting that several factors might play a role.
Previous studies have pointed out the key role played by executive functions and attention in gait performances in PD patients.17,18 We recently showed that freezing of gait is associated with executive dysfunction19 and with a worse progression of cognitive impairment in parkinsonian patients.20 Conceivably, gait impairment in PD may reflect altered motor control and overload of frontal networks. This consistently affects the ability of PD patients to walk while performing another task (ie, dual task).21,22 Rochester et al23 observed that the magnitude of the dual-task effect in PD is related to age, cognition, motor performance, and affective status. Furthermore, PD patients with poor executive functions, especially with set-shifting ability impairment, have been found to be more sensitive to the dual-task effect, thus being at higher risk of falling.24,25
Very recently, PD patients with MCI in different domains have been reported to display higher postural instability and gait disorder subscale scores than cognitively normal PD patients,26 thus suggesting that cognitive domains other than the executive one could be involved in balance and gait control. To our knowledge, no study has yet explored the relationship between quantitative gait variables and MCI in PD.
In the present study we objectively assessed the gait patterns of PD patients with MCI (MCI+) or without MCI (MCI−) in order to confirm the hypotheses that: (1) these 2 subgroups have different gait patterns during off conditions; (2) gait variables of MCI+ have a poorer response to levodopa than MCI− patients; (3) dual-task paradigms have different effects in MCI+ than in MCI− patients; and (4) specific gait components might correlate with specific cognitive domains. PD patients were also compared with healthy subjects (HS).
Patients and Methods
Patients were screened from a series of consecutive outpatients at the Movement Disorders Unit of the University of Naples Federico II who had a diagnosis of PD according to the United Kingdom Parkinson's Disease Brain Bank criteria.27
Patients were classified as MCI+ if they had both (1) a cognitive deficit not causing a significant functional decline and (2) dysfunction in at least 1 cognitive domain as confirmed by a consistent pattern of impairment (at least 1.5 standard deviations below the expected age and education-corrected mean score) in the specific neuropsychological tests.5 The 2 groups of patients were matched for age, sex, and disease duration.
HS were enrolled among volunteers who were age- and sex-matched to the patient population.
Further details on entry criteria are available online.
Clinical and Cognitive Evaluation
All subjects were evaluated using an extensive neuropsychological battery and a detailed clinical evaluation, including demographic and anthropometric data (see Supplemental Material). All neuropsychological tests were administered to patients during the on state. Tests scores were corrected for current normative values.
Gait was assessed with an optokinetic system (Qualisys, Sandvälen, Sweden) equipped with a set of 6 infrared cameras, a ProReflex Motion Capture Unit (MCU, CCD technology, 240-Hz sampling rate), and data acquisition software (Qualisys Track Manager). Patients' gait was assessed in both the off and on states and during 3 experimental conditions (each performed twice): (1) normal gait (normal walking; gait-off and gait-on); (2) motor dual-task (walking while carrying a tray with 2 glasses filled with water; Mot-off and Mot-on); and (3) cognitive dual task (walking while serially subtracting 7s starting from 100; Cog-off and Cog-on). The gait of HS was assessed in the same 3 conditions (each performed twice). Before the trials, all participants were instructed to walk at a normal pace at their usual speed and were not given any specific instruction regarding prioritization (walking or task).
For further details on gait analysis and walking parameters, refer to the online Supplemental Material.
Differences in the distribution of categorical variables among groups were assessed by the chi-square test. To better avoid type 1 errors, all statistical analyses applied were nonparametric tests. Demographic, clinical, and gait continuous variables of MCI+ PD patients, MCI− PD patients, and HS were compared using the Kruskal–Wallis test. Multiple comparisons of gait variables between groups (MCI+ vs MCI− vs HS) were performed with the post hoc Dunnett test.
The effects of dual-task conditions on gait parameters were evaluated by Friedman's ANOVA using the post hoc Dunnett test for multiple comparisons, assuming group (MCI+ PD vs MCI− PD vs HS) as the between factor and dual-task condition (normal gait parameters vs motor dual-task parameters vs cognitive dual-task parameters) as the within factor. To identify an association between the gait variables and any cognitive domain, factor analysis was performed (further details are available online).
Significance was set at P = .05. Computation was supported by SPSS version 16.0 (SPSS, Chicago, IL).
Sixty-three subjects were evaluated: 43 PD patients and 20 HS. Nineteen patients were classified as MCI+ and 24 as MCI− The 3 groups did not differ on demographic and anthropometric variables. MCI+ PD and MCI− PD patients did not differ in any of the clinical variables (Table 1). The 2 groups differed significantly on several tests loading on memory, executive, and visuospatial domains (Table 2). Cognitive scores were normal in HS.
Table 1. Demographic and clinical features of PD patients with (MCI+) and without (MCI−) mild cognitive impairment and healthy subjects (mean ± standard deviation)
|PD MCI+ (n = 19)||PD MCI− (n = 24)||Healthy subjects (n = 20)||P|
|Age (y)||65.10 ± 6.85||64.08 ± 6.44||63.50 ± 3.14||.68|
|Education (y)||9.00 ± 4.63||11.12 ± 4.52||12.00 ± 2.79||.071|
|Disease duration (y)||5.47 ± 2.71||5.42 ± 2.80||—||.947|
|H&Y stage||2.22 ± 0.43||2.23 ± 0.51||—||.962|
|Gait-Q score||11.53 ± 11.10||9.54 ± 9.96||—||.546|
|FOG-Q score||7.26 ± 6.17||5.83 ± 5.71||—||.441|
|UPDRS I||1.58 ± 1.30||1.79 ± 1.18||—||.583|
|UPDRS II||7.68 ± 3.99||7.71 ± 5.09||—||.986|
|UPDRS III in off state||23.63 ± 6.06||23.04 ± 6.91||—||.767|
|UPDRS III in on state||11.42 ± 3.72||10.83 ± 5.40||—||.676|
|UPDRS IV||2.89 ± 2.42||1.58 ± 1.72||—||.055|
|BDI||9.68 ± 4.89||9.79 ± 6.11||8.00 ± 5.79||.527|
|BMI (kg/m2)||26.36 ± 3.74||26.91 ± 4.86||26.71 ± 3.50||.910|
|Trochanter-malleolus (cm)||74.79 ± 3.81||76.25 ± 4.68||75.12 ± 5.44||.571|
|Intermalleolus (cm)||7.47 ± 0.54||7.55 ± 0.64||7.19 ± 0.65||.527|
Table 2. Data (mean ± standard deviation) from cognitive testing in PD patients with (MCI+) and without (MCI−) mild cognitive impairment and between-group comparisons (1-way ANOVA)
|MCI+ PD||MCI− PD||P|
|MMSE||26.57 ± 2.09||27.71 ± 1.72||NS|
|Episodic memory domain|
|Rey 15 words, immediate recall||35.07 ± 6.28||41.31 ± 7.95||.01|
|Rey 15 words, delayed recall||7.27 ± 2.78||8.52 ± 2.33||NS|
|Phonemic fluency||28.10 ± 10.24||36.88 ± 10.01||<.01|
|Frontal Assessment Battery||12.55 ± 2.03||15.14 ± 1.73||<.0001|
|Stroop, part II (color table)||31.63 ± 10.33a||35.59 ± .89a||NS|
|Stroop, part III (color/word table)||18.07 ± 5.10a||21.67 ± 5.36a||.03|
|Spatial span||4.23 ± 0.85||4.67 ± 0.70||NS|
|Constructive apraxia||10.46 ± 1.47||11.23 ± 1.31||NS|
|Raven's PM 47||23.07 ± 3.95||28.37 ± 4.35||.0001|
|Ten Point Clock test||4.10 ± 3.12||7.08 ± 3.12||.001|
Gait variables comparisons among the 3 groups are reported in Table 3a. Post hoc multiple comparisons were performed only on the gait parameters that differed significantly among the 3 groups.
Table 3. Significance of gait variables in different conditions in PD patients with (MCI+) and without (MCI−) mild cognitive impairment and in healthy subjects (HS) (a) and multiple comparisons (b)
|Gait variable||Gait conditions|
|NG-off (P)||Mot-off (P)||Cog-off (P)||NG-on (P)||Mot-on (P)||Cog-on (P)|
|Single/double support T R||.001||.001||.0001||.009||.013||.002|
|Step length variability||.046||.004||NS||NS||.018||NS|
|Swing time variability||NS||NS||NS||NS||NS||NS|
|Gait conditions (off)|
|NG (P)||Mot (P)||Cog (P)|
|Gait variable||MCI+ vs HSa||MCI− vs HSa||MCI+ vs HSa||MCI− vs HSa||MCI+ vs HSa||MCI− vs HSa|
|Single/double support T R||.0001||NS||.0001||NS||.0001||NS|
|Step length variability||NS||NS||.018||NS||0.034||NS|
|Gait conditions (on)|
|Single/double support T R||.003||NS||.017||NS||.001||NS|
|Step length variability||NS||NS||.013||NS||NS||NS|
Gait Patterns during Off State
As expected, step length was shorter in all conditions in PD patients (both MCI+ and MCI−) than in HS (Table 3b and Fig. 1A).
Figure 1. Post hoc multiple comparison analysis of PD patients with (MCI+) and without (MCI−) mild cognitive impairment and healthy subjects (HS) on step length (A), swing time (B), single/double support time ratio (C), and step length variability (D). NG, normal gait; Mot, motor dual task; Cog, cognitive dual task; off, in off state; on, in on state; NS, not significant; *P < .05; **P < .01; ***P < .001.
Swing time was shorter in MCI+ PD patients than in HS in all conditions, whereas it was shorter in MCI− PD patients compared with HS only during the cognitive dual task (Table 3b and Fig. 1B).
The single/double support time ratio was lower in all conditions in MCI+ PD than in HS, whereas it did not differ between MCI− PD and HS in any condition (Table 3b and Fig. 1C). Furthermore, the single/double support time ratio was significantly lower in MCI+ PD than in MCI− PD during the cognitive dual task (P = .022; Fig. 1C).
Multiple comparisons did not reveal any difference between groups on gait velocity in any conditions (Table 3b).
Step length variability was increased in MCI+ PD compared with MCI− PD and HS, with a gradient of MCI+ PD > MCI− PD > HS in all conditions. Furthermore, step length variability was increased in MCI+ PD versus HS in both dual-task conditions, whereas it did not differ in MCI− PD versus HS in any condition (Table 3b and Fig. 1D).
Levodopa effect on Gait Patterns
After levodopa, step length was still shorter in MCI+ PD than in HS in all conditions, whereas it was shorter in MCI− PD than in HS only in the motor dual task (Table 3b and Fig.1A). Swing time was shorter in MCI+ PD than in HS only during normal gait, whereas it did not differ between MCI− PD and HS in any condition (Table 3b and Fig. 1B). The single/double support time ratio was still lower in all conditions in MCI+ PD versus HS and did not differ between MCI− PD and HS in any condition (Table 3b and Fig.1C). Furthermore, the single/double support time ratio was significantly lower in MCI+ PD versus MCI− PD during normal gait (P = .045; Fig. 1C).
Multiple comparisons did not reveal any difference between groups in gait velocity in any condition (Table 3b).
Step length variability was increased in MCI+ PD versus HS only during the motor dual task. Again, it did not differ in MCI− PD versus HS (Table 3b and Fig. 1D).
Dual-Task Effects on Gait Patterns
Dual tasks affected step length in the off state in both MCI+ PD (P = .001) and MCI− PD (P = .002) compared with HS. During the on state, this effect was still significant in MCI+ PD (P = .001) but not in MCI− PD. Dual-task paradigms affected swing time during both off (P = .019) and on (P = .035) states in MCI+ PD but not in MCI− PD with respect to HS. Dual tasks displayed a significant effect on the single/double support time ratio during both off (P = .0001) and on (P = .0001) states in MCI+ PD versus HS. The same effect was detected in MCI− PD patients during off time, with a trend toward significance (P = .05), but not during the on state. Dual-task conditions did not affect velocity in PD patients relative to HSs in either the off or the on states. Finally, dual tasks affected step length variability during both the off (P = .018) and on (P = .008) states in MCI+ PD but not in MCI− PD with respect to HS. We did not find any significant interaction between groups and dual task for any gait variable.
Gait Factors and Their Relationship with Cognitive Domains
The principal-component analysis had a Bartlett's chi square of 346.576 (P < .0001) and generated 2 factors that explained about 75% of the variance (Table 4). The first factor loaded on the following variables: stance phase, swing phase, cadence, and velocity. The second factor loaded on step length, single support/double support time ratio, step length variability, and swing time variability. In line with the reported locomotion components,28 we named “pace” the first factor loading on variables related to initiation and maintenance of step rhythm, and we named “stability” the second factor loading on variables related to balance. The pace factor was not correlated with any cognitive or clinical variable (Table 5a, b). The stability factor was strongly and directly correlated with the visuospatial domain (Table 5a). Furthermore, the stability factor was also inversely correlated with almost all the main clinical measures of disease progression (Table 5b).
Table 4. Factor analysis of the 8 quantitative gait parameters
|Structure matrix of factor loading after rotation||Factor 1 (pace)||Factor 2 (stability)|
|Single support/double support||0.401||0.735|
|Step length variability||0.023||−0.689|
|Swing time variability||0.014||−0.715|
Table 5. Partial correlation of factor scores with cognitive cumulative scores (a) and main clinical measures (b)
|Factor 1 (pace)||Factor 2 (stability)|
|H&Y stage||0.209||NS||−0.430||(< 0.004)|
|UPDRS III on||0.095||NS||−0.288||(0.06 — trend)|
To our knowledge, this is the first study evaluating the relationship between MCI, quantitative walking parameters, and the effect of dual tasks on gait in PD patients during both the off and on states compared with HS. It is also the first exploring the relationship between specific gait variables and the cognitive domains more commonly impaired in PD. Here we showed that, compared with MCI− patients and HS, MCI+ PD patients display specific gait features (ie, both reduced step length and swing time and impairment of dynamic stability), which are only partially reversed by levodopa. We also found that dual-task conditions affect several walking parameters in MCI+ PD in both the off and on states with respect to MCI− and HS. This finding supports evidence that cognitive loading exerts a detrimental effect on gait performance in PD patients,17 the magnitude of which is related to the underlying cognitive dysfunction. Finally, we have shown that instability is specifically associated with both visuospatial impairment and clinical progression in PD .
Gait Patterns during Off State and Levodopa Effect
In agreement with previous findings,14–16,29 during the off state, step length was shorter in PD patients than in HS. During the on state, step length did not improve significantly in MCI+ PD during any condition, whereas it increased in MCI− during both normal gait and the cognitive dual task but not during the motor dual task. The motor dual task could be more demanding in PD patients because it requires not only executive-attentional but also visuospatial skills,30 which could explain the reduced effect of levodopa during performance of this task.
Similarly, during the off state, the single/double support time ratio was lower in MCI+ PD than in MCI− PD and HS and did not improve in the on state. The single/double support time ratio is a direct measure of dynamic stability.31 By means of a quantitative gait assessment, our findings support the well-known relationship between cognitive impairment and increased instability in PD patients,7,32,33 as well as the poor response of axial symptoms to levodopa.11,34 Consistently, in the off state during dual-task conditions, step length variability was greater in MCI+ PD. This supports the finding that a dual task exerts a strong effect on variability measures of gait in PD patients with cognitive impairment.22,24,35 Previous findings have reported inconsistent results on levodopa efficacy in reducing gait variability in PD,13–16 suggesting that co-occurrence of other factors may account for such differences. In the present study, levodopa reduced step length variability during the cognitive dual task but not during the motor one. Again, this supports the concept that the motor dual task might be more demanding. Because increased step length variability is a measure of stability impairment25 as well as both lower single/double support time ratio and shorter step length, our findings show a reduced response of instability to levodopa in MCI+ PD compared with MCI− PD, perhaps suggesting that dopa-resistant gait components and cognitive dysfunction might share common nondopaminergic network dysfunction.11,34
During the off state, swing time was shorter in MCI+ PD than in HS in all conditions and shorter in MCI− PD than in HS only during the cognitive dual task. During the on state, swing time increased in PD patients, although it remained shorter in MCI+ PD than in HS. Surprisingly, this difference was significant only during normal gait, which could indicate a reduced dual-task effect sensitivity of this parameter in patients with an underlying cognitive dysfunction.
Consistent with previous observations,29 velocity was reduced in PD patients in both the off and on states with respect to HS. However, multiple comparisons did not reveal any significant difference in any condition, thus mirroring that velocity is a raw measure underlain by multiple gait adjustments, variably modulated by patients. Furthermore, it has been reported that cognition contributes poorly to gait speed in PD patients while both normal and dual-task walking.23,36
Dual-Task Effects on Gait Patterns
The lack of any significant interaction between group and dual-task conditions is partly in agreement with previous reports17,37 and indicates that healthy elderly subjects and PD patients tend to deteriorate gait performances under dual tasks. Nevertheless, this effect was significantly different in the comparisons of PD subgroups. In fact, dual-task conditions affected step length, swing time, single/double support time ratio, and step length variability in MCI+ PD patients in both the off and on states, whereas they exerted an effect on both step length and single/double support time ratio during the off but never during the on state in MCI− PD patients. These findings support the hypothesis of “wrong prioritization” (first task and then posture) when gait cortical control collapses under the dual-task overloading, thereby increasing the risk of falling in PD patients.30 However, our comparison of PD patients with and without MCI during both the off and on states with HS demonstrates (1) that “wrong prioritization” mostly concerns MCI+ PD patients maybe mirroring impaired hazard estimation in such patients38 and (2) that levodopa does not reduce increased sensitivity to dual tasks in MCI+ PD patients.
Gait Factors and Their Relationship with the Cognitive Domains
Previous studies have highlighted the crucial role of both attention and executive functions in gait control.17,18,36 Nevertheless, these studies focused on frontal lobe–based cognitive abilities and did not assess more posterior cognitive skills as a visuospatial domain.
In our PD population, the stability factor was specifically correlated with visuospatial domain scores, whereas the executive domain did not correlate with any factor, maybe suggesting a more generalized role for executive functions on gait. A growing body of evidence has drawn attention to the importance of visual information processing during the generation of motor plans39 and in the control of locomotion in PD.40 Furthermore, recent studies evaluating the effect of space perception on gait in PD patients41,42 have shown that visuospatial ability is more greatly affected in PD patients with freezing of gait than in those without. Very recently, UPDRS subscores for postural instability and gait disturbances were found to be associated with visuospatial functions in newly diagnosed PD patients.43 On the one hand, our data confirmed the relationship between the visuospatial domain and gait, and on the other hand, they showed that a specific component of locomotion, namely the stability, is highly related with visuospatial processing. Interestingly, a reduced stability factor score was also inversely associated with higher H&Y stage, gait and freezing questionnaire scores, and motor UPDRS, suggesting a specific relationship among motor progression, instability, and visuospatial impairment in PD. Conceivably, the clinical impact of our results could be that PD patients with visuospatial impairment compared with PD patients with other cognitive dysfunction would display more instability and could be in at greater risk of falling.
Our study has several limitations. First, we know that clinical diagnostic criteria for MCI in PD were published very recently,44 but because these criteria were lacking at the time of study enrollment, we referred to the arbitrary definition for MCI proposed by Caviness et al.5 Nevertheless, the MCI criteria we used, based on consistent dysfunction in at least 1 cognitive domain not causing significant decline, are mostly in agreement with the recent PD-MCI criteria. Second, we identified off and on states according to widely used procedures, but we knew that motor states so identified could be biased by individual and pharmacological factors. Third, we are aware that our results could have been influenced by the small sample size, and therefore they should be regarded as preliminary. Fourth, another limitation was the relatively short duration over which gait was recorded. Finally, the main study limitation was probably relying on its cross-sectional design, which does not allow for inference about a causative relation between cognitive dysfunction and gait abnormalities. Further longitudinal studies are needed to elucidate any causative relationship between these aspects.
In conclusion, our data demonstrate that dysfunctions in specific gait parameters that are poorly responsive to levodopa and highly sensitive to dual-task conditions are associated with MCI in PD. Importantly, visuospatial impairment seems strongly associated with the development of instability and more generally with the progression of PD.
We thank Nicky Cecketts for the English editing.
Marianna Amboni contributed to the conception, organization, and execution of the research project and the design and execution of the statistical analysis, wrote the first draft, and reviewed and critiqued the manuscript. Paolo Barone contributed to the conception of the research project and reviewed and critiqued the statistical analysis and the manuscript. Luigi Iuppariello contributed to the organization and execution of the research project. Ilaria Lista contributed to the organization and execution of the research project. Riccardo Tranfaglia contributed to the organization and execution of the research project. Alfonso Fasano contributed to the organization of the research project and reviewed and critiqued the manuscript. Marina Picillo contributed to the organization and execution of the research project. Carmine Vitale contributed to the organization and execution of the research project. Gabriella Santangelo contributed to the organization and execution of the research project. Valeria Agosti contributed to the organization and execution of the research project. Alessandro Iavarone contributed to the design, execution, and review and critique of the statistical analysis and reviewed and critiqued the manuscript. Giuseppe Sorrentino contributed to the conception of the research project and reviewed and critiqued the statistical analysis and the manuscript.
Financial Disclosures: Marianna Amboni has received honoraria for symposia from Boehringer Ingelheim, Lundbeck, and Novartis; and has received salary from IDC Hermitage-Capodimonte and the Department of Neurological Sciences, University of Naples Federico II, Italy. Paolo Barone has received honoraria as a consultant and advisory board member for Novartis, Schwarz Pharma/UCB, Merck-Serono, Eisai, Solvay, General Electric, and Lundbeck; has received research support from Boehringer Ingelheim, Novartis, Schwarz Pharma/UCB, Merck-Serono, Solvay, and Lundbeck; has received salary from the Department of Neurological Sciences, University of Naples Federico II and from the University of Salerno, Italy. Luigi Iuppariello, Ilaria Lista, and Riccardo Tranfaglia report no disclosures. Alfonso Fasano states stock ownership in medically related fields, intellectual property rights, consultancies, expert testimony, advisory boards, employment, partnerships, contracts, honoraria, royalties, and others: none. Marina Picillo has received salary from the Department of Neurological Sciences, University of Naples Federico II, Italy. Carmine Vitale has received honoraria for symposia from Boehringer Ingelheim, Lundbeck, Novartis, and Schwarz Pharma/UCB and has received salary from IDC Hermitage-Capodimonte and the University of Naples Parthenope, Italy. Gabriella Santangelo has received honoraria for symposia from Lundbeck and has received salary from the Department of Psychology, Second University of Naples, Caserta, Italy. Valeria Agosti reports no disclosures. Alessandro Iavarone has received salary from CTO Hospital, AORN Ospedale dei Colli, Naples, Italy. Giuseppe Sorrentino has received salary from IDC Hermitage-Capodimonte and the University of Naples Parthenope, Italy.
- Yogev-Seligmann G, Hausdorff JM, Giladi N. The role of executive function and attention in gait. Mov Disord.2008; 23: 329–342.
- Verghese J, Wang C, Lipton RB, Holtzer R, Xue X. Quantitative gait dysfunction and risk of cognitive decline and dementia. J Neurol Neurosurg Psychiatry.2007; 78: 929–935.
Verghese J, Robbins M, Holtzer R, Zimmerman M, Wang C, Xue X, Lipton RB. Gait Dysfunction in mild cognitive impairment syndromes. J Am Geriatr Soc.2008; 56: 1244–1251.
- Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment: clinical characterization and outcome. Arch Neurol.1999; 56: 303–308.
Caviness JN, Driver-Dunckley E, Connor DJet al. Defining mild cognitive impairment in Parkinson's disease. Mov Disord.2007; 22; 1272–1277.
Barone P, Aarsland D, Burn D, Emre M, Kulisevsky J, Weintraub D. Cognitive impairment in nondemented Parkinson's disease. Mov Disord.2011; 26: 2483–2495.
- Jankovic J, McDermott M, Carter Jet al. Variable expression of Parkinson's disease: a baseline analysis of the DATATOP cohort. The Parkinson Study Group.Neurology.1990; 40: 1529–1534.
Klawans HL. Individual manifestations of Parkinson's disease after ten or more years of levodopa. Mov Disord.1986; 1: 187–192.
- Bonnet AM, Loria Y, Saint-Hilaire MH, Lhermitte F, Agid Y. Does long-term aggravation of Parkinson's disease result from nondopaminergic lesions?Neurology.1987; 37: 1539–1542.
- Dubois B, Pillon B. Cognitive deficit in Parkinson's disease. J Neurol.1997; 244: 2–8.
- Braak H, Del Tredici. Invited article: nervous system pathology in sporadic Parkinson's disease. Neurology.2008; 70: 1916–1925.
- Kehagia AA, Barker RA, Robbins TW. Neuropsychological and clinical heterogeneity of cognitive impairment and dementia in patients with Parkinson's disease. Lancet Neurol.2010; 9: 1200–1213.
- Blin O, Ferrandez AM, Pailhous J, Serratrice G. Dopa-sensitive and dopa-resistant gait parameters in Parkinson's disease. J Neurol Sci.1991; 103: 51–54.
- Schaafsma JD, Giladi N, Balash Y, Bartels AL, Gurevich Y, Hausdorff JM. Gait dynamics in Parkinosn's disease: relationship to Parkinsonian features, falls and response to levodopa. J Neurol Sci.2003; 212: 47–53.
- Rochester L, Baker K, Nieuwboer A, Burn D. Targeting dopa-sensitive and dopa-resistant gait dysfunction in Parkinson's disease. Selective response to internal and external cues. Mov Disord.2011; 26: 430–435.
- Moore ST, MacDougall HG, Gracies J, Ondo WG. Locomotor response to levodopa in fluctuating Parkinson's disease. Exp Brain Res.2008; 184: 469–478.
Yogev G, Giladi N, Peretz C, Springer S, Simon ES, Hausdorff JM. Dual tasking, gait rhythmicity and Parkinson's disease: Which aspects of gait are attention demanding?Eur J Neurosci.2005; 22: 1248–1256.
- Hausdorff JM, Doniger GM, Springer S, Yogev G, Simon ES, Giladi N. A common cognitive profile in elderly fallers and in patients with Parkinson's disease: the prominence of impaired executive function and attention. Exp Aging Res.2006; 32: 411–429.
Amboni M, Cozzolino A, Longo K, Picillo M, Barone P. Freezing of gait and executive functions in patients with Parkinson's disease. Mov Disord.2008; 23: 395–400.
Amboni M, Barone P, Picillo M, et al. A two-year follow-up study of executive dysfunctions in Parkinsonian patients with freezing of gait at on-state. Mov Disord.2010; 25: 793–795.
- Camicioli R, Oken BS, Sexton G, Kaye JA, Nutt JG. Verbal fluency task affects gait in Parkinon's disease with motor freezing. J Geriatr Psychiatry Neurol.1998; 11: 181–185.
- Hausdorff JM, Balash J, Giladi N. Effects of cognitive challenge on gait variability in patients with Parkinson's disease. J Geriatr Psychiatry Neurol.2003; 16: 53–58.
Rochester L, Nieuwboer A, Baker Cet al. Walking speed during single and dual tasks in Parkinson's disease: which characteristics are important?Mov Disord.2008; 23: 2312–2318.
- Plotnik M, Dagan Y, Gurevich T, Giladi N, Hausdorff JM. Effects of cognitive function on gait and dual tasking abilities in patients with Parkinson's disease suffering from motor response fluctuations. Exp Brain Res.2011; 208: 169–179.
- Plotnik M, Giladi N, Dagan Y, Hausdorff JM. Postural instability and fall risk in Parkinson's disease. Impaired dual tasking, pacing, and bilateral coordination of gait during the “ON” medication state. Exp Brain Res.2011; 210: 529–538.
- Sollinger AB, Goldstein FC, Lah JJ, Levey AI, Factor SA. Mild cognitive impairment in Parkinson's disease: subtypes and motor characteristics. Parkinsonism Relat Disord.2010; 16: 177–180.
- Gelb DJ, Oliver E, Gliman S. Diagnostic criteria for Parkinson disease. Arch Neurol.1999; 56: 33–39.
- Snijders AH, van de Warrenburg BP, Giladi N, Bloem BR. Neurological gait disorders in elderly people: clinical approach and classification. Lancet Neurol.2007; 6: 63–74.
- Morris ME, Iansek R, Matyas TA, Summers JJ. Stride length regulation in Parkinson's disease. Normalization strategies and underlying mechanisms. Brain.1996; 119: 551–568.
- Bloem BR, Grimbergen YA, van Dick JG, Munneke M. The “posture second” strategy: a review of wrong priorities in Parkinson's disease. J Neurol Sci.2006; 248: 196–204.
- Whittle M. Gait Analysis: An Introduction. Oxford, UK: Butterworth-Heinemann; 1991.
Alves G, Larsen JP, Emre M, Wentzel-Larsen T, Aarsland D. Changes in motor subtypes and risk for incident dementia in Parkinson's disease. Mov Disord.2006; 21: 1123–1130.
- Verbaan D, Marinus J, Visser M, et al. Cognitive impairment in Parkinson's disease. J Neurol Neurosurg Psychiatry.2007; 78: 1182–1187.
Devos D, Defebvre L, Bordet R. Dopaminergic and non-dopaminergic pharmacological hypotheses for gait disorders in Parkinson's disease. Fundam Clin Pharmacol.2010; 24: 407–421.
- Yogev G, Plotnik M, Peretz C, Giladi N, Hausdorff JM. Gait asymmetry in patients with Parkinson's disease and elderly fallers: when does the bilateral coordination of gait require attention?Exp Brain Res.2007; 177: 336–346.
- Lord S, Rochester L, Hetherington V, Allock LM, Burn D. Executive dysfunction and attention contribute to gait interference in “off” state Parkinson's disease. Gait Posture.2010; 31: 169–174.
- Bloem BR, Valkenburg VV, Slabbekoorn M, Willemsen MD. Multiple task test. Development and normal startegies. Gait and Posture2001; 14: 191–202.
Yogev-Seligmann G, Hausdorff JM, Giladi N. Do we always prioritize balance when walking? Towards an integrated model of task prioritization. Mov Disord.2012; 27: 765–770.
- Helmich RC, de Lange FP, Bloem BR, Toni I. Cerebral compensation during motor imagery in Parkinson's disease. Neuropsychologia.2007; 45: 2201–2215.
- Lewis GN, Byblow WD, Walt SE. Stride length regulation in Parkinson's disease: the use of extrinsic, visual cues. Brain.2000; 123: 2077–2090.
- Almeida QJ, Lebold CA. Freezing of gait in Parkinson's disease: a perceptual cause for a motor impairment?J Neurol Neurosurg Psychiatry.2010; 81: 513–518.
- Cowie D, Limousin P, Peters A, Day BL. Insights into the neural control of locomotion from walking through doorways in Parkinson's disease. Neuropsychologia.2010; 48: 2750–2757.
Domellof ME, Elgh E, Forsgren L. The relation between cognition and motor dysfunction in drug-naive newly diagnosed patients with Parkinson's disease. Mov Disord.2011; 26: 2183–2189.
Litvan I, Goldman JG, Troster AI, et al. Diagnostic criteria for Mild Cognitive Impairment in Parkinson's Disease: Movement Disorder Society Task Force guidelines. Mov Disord.2012; 27: 349–356.