Arbeiten unterschiedlicher Arbeitsgruppen haben die Frage aufgeworfen, ob es sich bei Freezing of Gait um ein systemisches Defizit bei Parkinsonpatienten handeln k ¨onnte, d. h. die Motorblockaden nicht nur w¨ahrend des Gehens auftreten, sondern auch ver-mehrt w¨ahrend anderer Bewegungen. Dies w ¨urde bedeuten, dass die motorischen Defi-zite, welche beim Gehen von Freezern auftreten auch in der oberen Extremit¨at zu finden sind. In einer Studie mit 990 Parkinsonpatienten zeigten 318 Patienten Motorblockaden.
Von diesen Motorblockaden waren 11% nicht im Gang zu finden sondern in anderen K ¨orperpartien [94]. Dieser und andere Artikel weisen daraufhin, dass Motorblockaden bei Parkinsonpatienten w¨ahrend rhythmischer Bewegungen, wie w¨ahrend des Schrei-bens, Sprechens, Fingertappings und bei anderen antiphasischen Bewegungen vorhan-den sind [5, 6, 70, 94, 162, 180, 270]. W¨are FOG ein systemisches Defizit, w ¨urde dies unser Verst¨andnis von FOG ver¨andern. Es w ¨urden sich neue Untersuchungsm ¨oglichkeiten erge-ben, welche Erkl¨arungen zur Pathophysiologie des Freezings liefern k ¨onnten. So k ¨onnten PET und funktionelle Magnetresonanztomographie (fMRT), welche w¨ahrend des Ge-hens unm ¨oglich sind, an anderen Bewegungen, z. B. Fingertapping, durchgef ¨uhrt werden und so die involvierten Hirnregionen ermittelt werden [158, 207]. Letztendlich k ¨onnten sich daraus neue Therapien entwickeln und den Patienten eine bessere Lebensqualit¨at erm ¨oglichen. Dies setzt voraus, dass diese Freezingepisoden der oberen Extremit¨at de-nen der unteren Extremit¨at gleichen und den selben pathophysiologischen Mechanismen
zugrunde liegen.
Bis jetzt gibt es noch wenige Untersuchungen, die Aufschluss ¨uber solche Motorblo-ckaden geben. Diese werden als das Festfrieren der oberen Extremit¨at, wenn sie in den Armen auftreten (aus dem Englischen: freezing of the upper limb (FOUL)) bezeichnet.
Eine Arbeit, welche sich speziell mit unilateralem Fingertapping in Parkinsonpatien-ten besch¨aftigte, konnte Motorblockaden in Fingerbewegungen feststellen. Hier waren FOUL-Episoden vermehrt bei Parkinsonpatienten, im Vergleich zu gesunden Kontrollpa-tienten, vorhanden. Die FOUL-Episoden nahmen mit Erm ¨udung exponentiell zu [270].
Die Motorblockaden waren nicht korreliert mit der Erkrankungsdauer oder den motori-schen Einschr¨ankungen bei Morbus Parkinson, welche durch den aus dem Englimotori-schen:
Unified Parkinson’s Disease Rating Scale III (UPDRS III) ermittelt wurden. Dies ist ein Fra-geboden, welcher speziell die motorischen Einschr¨ankungen von Parkinsonerkrankten erfasst. Patienten mit FOG zeigten mehr FOUL-Episoden als Parkinsonpatienten ohne Freezing. Zus¨atzlich wurde ermittelt, dass sich nicht die Anzahl, allerdings die Dauer dieser Episoden unter dopaminerger Medikation verringerten [270]. In einer alternieren-den Oberarmbewegung konnten diese Motorblockaalternieren-den ebenfalls nachgewiesen weralternieren-den.
So zeigten sich in 13 Parkinsonpatienten FOUL-Episoden in antiphasischen Oberarm-bewegungen, das waren 8,1% aller Durchl¨aufe [5]. Eine weitere Studie untersuchte den Zusammenhang zwischen der Gangasymmetrie und der Asymmetrie der Bewegung der oberen Extremit¨at. Dazu wurde die Asymmetrie der Schwingzeit der Beine (die Dauer, welche ein Fuß w¨ahrend des Gehens nicht den Boden ber ¨uhrt) mit der Asymmetrieskala des CAPIT (aus dem Englischen: core assessment program for intracerebral transplan-tations), ein Zeittest f ¨ur verschiedene Bewegungen der Unterarme und Hand, f ¨ur beide Arme verglichen. Hierbei wurde eine gr ¨oßere Asymmetrie des Ganges f ¨ur Freezer im Vergleich zu Nicht-Freezern nachgewiesen, was allerdings nicht f ¨ur den CAPIT-Score zutraf [195]. Eine andere Studie hat versucht einen Zusammenhang in der Koordination von alternierenden Schreibbewegungen beider Arme und FOG herzustellen. Leider war das Kollektiv an drei Patienten, welche FOG zeigten, sehr gering. Hier konnte kein Defizit der Regulation von Amplitude und Frequenz in den alternierenden Bewegungen gefun-den wergefun-den, allerdings war auch hier Freezing der oberen Extremit¨at messbar [180]. Das Schwellenmodell, welches bereits erw¨ahnt wurde, l¨asst sich ebenfalls auf FOUL anwen-den, da sich FOUL-Episoden besonders bei kleinamplitudigen Bewegungen mit hoher Frequenz generieren ließen [250].
Nur eine einzige bildgebende Studie existiert zur Untersuchung der zerebralen Ak-tivit¨at von FOUL. Hier wurden unterschiedliche Aktivierungsmuster mittels fMRT bei Freezern in Ruhebedingung und w¨ahrend FOUL-Episoden beim Fingertapping nachge-wiesen (Tabelle 3.4.).
Tabelle 3.4: ¨Ubersicht von Artikeln zu funktionalen Bildgebung von FOUL Publikation Methodik Entstandene Pathologie
Vercruysse et al. (2013) [254]
fMRI - BOLD w¨ah-rend erfolgreichen Fingertappings und w¨ahrend FOUL-Epi-soden von Freezern im Vergleich zu ge-sunden Kontrollen
Bewegung ohne FOUL: vermin-derte Aktivit¨at im dorsolatera-len pr¨afrontadorsolatera-len Kortex, linken dorsalen frontalen Kortex und linkem M1, erh ¨ohte Aktivit¨at im dorsalen Putamen, Pallidum, STN; w¨ahrend FOUL: erh ¨ohte Aktivit¨at im M1, SMA, dorso-lateralen pr¨amotorischen Areal, linken pr¨afrontalen Kortex, ver-minderte bilaterale Aktivit¨at im Pallidium und Putamen
Abk ¨urzungen: BOLD (aus dem Englischen: blood oxygenation level dependent), M1 (prim¨ar-motorischer Kortex), fMRI (aus dem Englischen: functional magnetic resonance imagery), SMA (Supplement¨ar motorisches Areal), STN (Nucleus subthalamicus)
Zusammengefasst zeigen diese Studien, dass es Motorblockaden der oberen Extremit¨at gibt. Eine finale Schlussfolgerung, ob FOG und FOUL einer ¨ahnlichen Koordinationsst ¨orung bei Patienten mit Morbus Parkinson und FOG unterliegen, lassen sie nicht zu. Es fehlt eine quantitative Erhebung der motorischen Defizite des Gehens zum direkten Vergleich der motorischen Defizite der oberen Extremit¨at bei Freezern. Dies w ¨urde helfen die Frage zu kl¨aren, ob Freezing ein systemisches Defizit mit einer gemeinsamen pathophysiologi-schen Ursache ist.
Daraus folgen die Hypothesen dieser Arbeit:
1) Parkinsonpatienten mit FOG zeigen eine h ¨ohere zeitliche und ¨ortliche Variabilit¨at des Ganges als Patienten ohne FOG in Bewegungsphasen ohne FOG.
2) Parkinsonpatienten mit FOG zeigen eine h ¨ohere zeitliche und ¨ortliche Variabilit¨at in Armbewegungen, speziell beim Fingertapping und in der Diadochokinese von Pronation und Supination der Unterarme in Bewegungsphasen ohne FOUL.
3) FOG und FOUL treten bei Patienten mit verifiziertem FOG h¨aufiger auf als bei Parkinsonpatienten ohne FOG.
4) Es besteht eine Korrelation der Dysfunktion der zeitlichen und ¨ortlichen Variabilit¨at der Bewegung der oberen und unteren Extremit¨at bei Patienten mit FOG.
5) Es Besteht eine Korrelation zwischen der Dauer und Anzahl von FOG und FOUL bei Patienten mit FOG.
6) Levodopa verbessert die Beweglichkeit der Patienten mit Morbus Parkinson und FOG und verringert die H¨aufigkeit von FOG und FOUL.
Zun¨achst wurde versucht, die Defizite der Gangst ¨orung, welche in anderen Studien nachgewiesen wurden, zu replizieren. Diese Untersuchungen wurden dann auf die obe-re Extobe-remit¨at ausgeweitet. Das Fingertapping und die Diadochokinese der Unterarme wurden ausgew¨ahlt, da sie in Anlehnung an den UPDRS III zur neurologischen Unter-suchung von Morbus Parkinson geh ¨oren [96] und im klinisch-neurologischen Alltag ein wichtiges Werkzeug darstellen. Darauffolgend wurde untersucht, ob Freezing in der un-teren und oberen Extremit¨at nachweisbar ist und analysiert, ob das Bewegungsdefizit der verschiedenen Extremit¨aten miteinander korrelierte. Zuletzt wurde der therapeutische
Effekt von Levodopa auf die Bewegungen in den verschiedenen Extremit¨atenetagen und auf deren spezifisches Freezing untersucht.
5.1 Gait and upper limb variability in Parkinson’s disease patients with and without freezing of gait.
Weitere Details, besonders die Methodik und die Ergebnisse dieser Arbeit, werden nun in der folgenden Ver ¨offentlichung dargelegt.
Reprinted by permission from Springer Nature License: Springer
Journal of Neurology Gait and upper limb variability in Parkinson’s disease patients with and without freezing of gait.
Barbe MT, Amarell M, Snijders AH, Florin E, Quatuor EL, Sch ¨onau E, Fink GR, Bloem BR, Timmermann L..
J Neurol. 2014 Feb;261(2):330-42. doi: 10.1007/s00415-013-7199-1. Epub 2013 Dec 4.
©Springer 2014
O R I G I N A L C O M M U N I C A T I O N
Gait and upper limb variability in Parkinson’s disease patients with and without freezing of gait
Michael T. Barbe•Martin Amarell•Anke H. Snijders• Esther Florin•Eva-Lotte Quatuor•Eckhard Scho¨nau• Gereon R. Fink•Bastiaan R. Bloem•Lars Timmermann
Received: 27 September 2013 / Revised: 8 November 2013 / Accepted: 20 November 2013 ÓSpringer-Verlag Berlin Heidelberg 2013
Abstract Patients with Parkinson’s disease (PD) and freezing of gait (FOG) (freezers) demonstrate high gait var-iability. The objective of this study was to determine whether freezers display a higher variability of upper limb movements and elucidate if these changes correlate with gait. We were the first group to compare directly objectively measured gait and upper limb movement variability of freezers between freezing episodes. Patients with objectively verified FOG (n=11) and PD patients without FOG (non-freezers) (n=11) in a non-randomized medication condition (OFF/
ON) were analyzed. Uncued antiphasic finger tapping and forearm diadochokinetic movements were analyzed via three-dimensional ultrasound kinematic measurements. Gait vari-ability of straight gait was assessed using ground reaction forces. Freezers had shorter stride length (p=0.004) and higher stride length variability (p=0.005) in the medication OFF condition. Movement variability was not different
during finger tapping or diadochokinesia between the groups.
There was a trend towards more freezing of the upper limb during finger tapping for the freezers (p=0.07). Variability in stride length generation and stride timing was not associ-ated with variability of upper limb movement in freezers. Our findings demonstrate that: (1) freezers have a higher spatial gait variability between freezing episodes; (2) freezing-like episodes of the upper limb occur in PD patients, and tend to be more pronounced among freezers than non-freezers for finger tapping; (3) spatial and temporal upper extremity variability is equally affected in freezers and non-freezers in an uncued task. Upper limb freezing is not correlated to lower limb freezing, implicating a different pathophysiology.
Keywords Parkinson’s diseaseFreezing of gait Freezing of the upper limb FOUL Freezing of the upper limb ITI Inter tap interval
MMSE Mini-mental state examination PD Parkinson’s disease
TTD Total travel distance
UPDRS Unified Parkinson’s Disease Rating Scale
Introduction
Freezing of gait (FOG) is common in patients with advanced Parkinson’s disease (PD). It is defined as ‘brief, M. T. Barbe and M. Amarell contributed equally.
M. T. Barbe (&)M. AmarellE. Florin E.-L. QuatuorG. R. FinkL. Timmermann Department of Neurology, University Hospital Cologne, Kerpener Str. 62, 50937 Cologne, Germany
e-mail: michael.barbe@uk-koeln.de M. T. BarbeG. R. Fink
Institute for Neuroscience and Medicine, INM-3, Forschungszentrum Ju¨lich, Ju¨lich, Germany A. H. SnijdersB. R. Bloem
Department of Neurology, Parkinson Center Nijmegen, Radboud University, Nijmegen Medical Center, Nijmegen, The Netherlands
E. Scho¨nau
Department of Pediatrics, University Hospital Cologne, Cologne, Germany
123
J Neurol
DOI 10.1007/s00415-013-7199-1
episodic absence or marked reduction of forward progres-sion of the feet despite the intention to walk’ [26]. Its relevance increases with disease duration, as about 50 % [9] of PD patients experience FOG after 5 years. The pathophysiology of this complex gait disorder is poorly understood.
Gait analysis studies show that patients with PD and FOG (freezers) have different walking patterns even between FOG episodes. These irregularities include increased gait variability in the spatial [18] and the tem-poral dimension [12,18,28,29]. Prior to FOG episodes, a decrease in stride length (sequence effect) and an increase in gait variability occurs [5,16].
Conceptually, it is interesting to know whether this movement variability is specific to gait or whether motor blocks also occur in the upper limb thus reflecting a sys-temic motor control deficit. To answer this question several studies investigated whether upper limb movement is dif-ferent in freezers compared to patients with PD without FOG (non-freezers).
Freezing-like episodes in the upper limb have been identified in PD patients [1] and seem to be associated with FOG severity [25,38,40]. Plotnik and coworkers investi-gated asymmetry in gait and in rhythmic hand movements [28]. They found a higher gait asymmetry in freezers than in non-freezers during the medication OFF and ON state.
However, asymmetry of repetitive hand movements mea-sured with the CAPIT score was not different between the groups [28]. Nieuwboer et al. [25] explored freezers and non-freezers in a repetitive drawing task with simultaneous alternating movements of both hands. They found no dif-ference between the two groups regarding variation of amplitude generation and cycling frequencies in an inter-nally generated movement [25]. Nanhoe-Mahabier et al.
[23] found a relation between ipsi- and contralateral arm swing synchronization and gait in PD patients but could not distinguish between freezer and non-freezer. Vercruysse et al. [37] found that, similar to FOG, constant amplitude reduction occurs preceding a freezing of the upper limb (FOUL) episode and that FOUL episodes are induced through small amplitude movements. Furthermore, they found a deterioration of temporal bimanual hand move-ment control in freezers only after an initial cue stimulus was taken away [38].
We designed a matched-to-sample study comparing movements of verified freezers with non-freezers. We investigated whether direct correlations in upper and lower limb movement patterns and variability could be found.
According to the aforementioned literature, we formed three hypotheses. Firstly, spatial and temporal gait vari-ability between FOG episodes is more disturbed in freezers than in non-freezers. Secondly, we assumed that
freezing-like episodes are present in upper limb movement in all patients and are more common in patients with FOG, although the latter is under debate. Thirdly, we hypothe-sized that gait variability correlates with variation of the upper limb and clinical parameters in patients with FOG.
Additionally, knowledge about the responsiveness of FOG on levodopa is conflicting. It seems to be the most effective therapy for FOG, [30] but different types of freezing being responsive, non-responsive or even being induced by levodopa exist and further research into freez-ing therapy is necessary [8]. Thus, we further elucidated the effects of levodopa on FOG and on FOUL.
Methods
Ethics
The study was approved by the ethics committee of the medical faculty of the University of Cologne. Patients gave written informed consent before study participation.
Subjects
Thirty-four patients with PD were grouped according to the German version of the freezing of gait questionnaire (FOGQ) [10,11] item 3 (0=non-freezers,C1=freezers) into freezers (n=17) and non-freezers (n=17). Pro-spectively, patients were matched according to age, sex and expected Unified Parkinson’s Disease Rating Scale (UP-DRS III) [7] medication-OFF scores according to previous ratings. Inclusion criteria were: (1) idiopathic PD; (2) akinetic-rigid dominance; (3) right handedness (according to Edinburgh inventory [27]); (4) age between 40 and 80 years; (5) improvement of at least 20 % in UPDRS III scores after intake of at least 200 mg soluble L-Dopa (Madopar LT, Roche), as this is a diagnostic criterion for idiopathic PD and to avoid levodopa unresponsive patients with, e.g., atypical PD [15]. Exclusion criteria were: (1) clinically relevant dementia (mini-mental state examina-tion (MMSE)-score\25 [4]) or depression; (2) ON state freezing [6]; (3) operations and/or chronic diseases of the upper and lower limbs influencing movements; (4) intake of medication active on the central nervous system other than dopaminergic. Laterality of parkinsonian symptoms was calculated by subtracting the left hemibody scores of UPDRS III items 20–26 from those of the right side [3,35].
Testing followed first in the medication OFF state (after cessation of anti-parkinsonian medication for at least 12 h) and then in the medication ON state (after intake of 1.5 times the morning dose or at least 200 mg of solubleL -Dopa). Both medication states were performed at the same day and therefore were not randomized.
J Neurol
123
Walking test
All patients performed a walking test introduced by Schaafsma and coworkers [30], in order to avoid mis-classification, i.e. to separate ‘subjective’ FOG (identified by FOGQ item 3) from ‘definite’ (objectively verified) FOG [33]. This test includes FOG provoking situations as gait initiation, narrow spaces, open quarters and turning.
Patients were sitting on a chair. At one auditory signal patients would stand up, walk straight for 10 m, pass two narrow chairs (50 cm apart), walk straight for another 10 m, step on a yellow cloth dot (1 m diameter), turn 360°
to the right, then 540°to the left and return to the chair at the beginning while passing the two chairs again. Patients were instructed to walk at fast but comfortable speed. All patients performed this walking test four times (twice in the medication OFF and ON condition, respectively). Walking tests were videotaped and assessed for number and duration of FOG episodes by two independent raters (MTB and AHS) both blinded for patient group and medication state.
In case of deviations of FOG recognition and/or duration between raters (which was the case for one patient), con-sensus was achieved by a second rating of the videos by both raters. Inter-rater reliability between the two raters was high for the overall recognition of FOG (Cronbach’s Alpha 1.0 for medication OFF and ON). Further subgroup analysis was performed with the data of patients with verified FOG (freezers, n=11). Matched controls (non-freezers,n=11) had neither subjective freezing (FOGQ # 3=0) nor freezing episodes during the walking tests.
Gait analysis
For gait analysis, patients walked across a 6 m long and 0.8 m wide gangway with force sensors calculating center of force data (Leonardo Mechanograph, NovoTec Medical) [36]. For illustration of gait analysis, see Fig.1a, b.
Patients were instructed to walk at a fast but comfortable speed along the straight walking system 16 times during OFF and ON each (total walking distance 96 m for OFF and ON each). All patients performed one practice trial.
After 4 m of the walking system, a wooden doorway (height 2.1 m; width 0.8 m; frame width 0.06 m) was placed as a FOG provoking stressor. Center of force (CoF) was measured at a sampling frequency of 800 Hz and data was further processed with Matlab (version 7.8.0.347 R2009a, The MathWorks, Inc., Natrick, MA). CoF maxima were used to calculate stride length and stride time, as well as their respective coefficient of variation (CV). CV was calculated as the standard deviation divided by the mean. A small CV indicates a small variability in parameters whereas high CV indicates a high variability. Thus, a small CV is a measure for well-coordinated spatial/temporal
movement. Stride length and stride time were defined as the distance and time, respectively, between two CoF maxima. The first and last two strides of each trial were excluded from the data analysis to avoid artificial variation in gait caused by walking initiation and termination. FOG episodes during walking trials were documented and excluded from the analysis to allow interictal gait analysis.
Freezers are known to have shorter stride length and therefore presumably need more strides to pass the gang-way. In order to avoid a selective bias, i.e. the number of steps influencing the standard deviation, we analyzed 15 randomly selected strides per patient to be able to compare the same amount of strides for both groups. This approach was adapted from the study of Hausdorff et al. [12] who deleted the first two strides and freezing episodes of each trial and analyzed the 15 consecutive strides for assessment of dynamic gait.
Movement analysis of the upper extremity
Movements of alternating tapping of both index fingers and alternating diadochokinetic forearm movements were recorded with an ultrasound measurement system (CMS20, Zebris Medical GmbH, Isny, Germany) [14, 34]. For overview of upper limb movement traces, see Fig.1c–f.
These movements were chosen as they are part of stan-dardized clinical PD examination (UPDRS III, item 23, 25). For finger tapping, the first ultrasound marker was placed on the distal interphalangeal joint and the second marker was placed behind the intercarpophalangeal joint on the index finger on both hands. For diadochokinesia, ultrasound markers were placed in the middle of the middle phalanx of finger II and in the middle of the middle phalanx of finger V. Patients followed standardized video instruc-tions. Patients performed at least one practice trial before testing. During testing, patients were sitting in front of a table. For finger tapping, both forearms were resting with the palms of the hands located on the table. For diado-chokinetic movement, patients rested their elbows on the table and performed alternating movements. Patients were instructed to perform both movements between two acoustic signals, which consisted of a single signal at the beginning and end of each trial. No cues for triggering were presented during the movement. For both paradigms antiphasic movements were recorded for 10 s with 20 s breaks in between, for a total of 16 trials each in the OFF and ON. The sampling rate was 75 Hz. The spatial reso-lution was 0.1 mm. All patients were instructed to perform movements at fast but comfortable speed. The goal was to motivate patients for movement performance but not to force them to perform at maximum speed to reduce fatigue.
Trial duration of 10 s was chosen to avoid fatigue inducing
Trial duration of 10 s was chosen to avoid fatigue inducing