- Email: [email protected]
Gait festination in Parkinson's disease
Parkinsonism & Related Disorders Parkinsonism and Related Disorders 7 (2001) 135±138
www.elsevier.com/locate/parkreldis
Gait festination in Parkinson's disease N. Giladi a,*, H. Shabtai a, E. Rozenberg a, E. Shabtai b a
Movement Disorders Unit, Department of Neurology, Tel Aviv Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel b Statistical Services, Tel Aviv Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel Received 10 April 2000; revised 29 May 2000; accepted 30 May 2000
Abstract Background: Festinating gait (FSG) was ®rst associated with parkinsonism by Sir James Parkinson, in his original essay on ªThe Shaking Palsyº. Its frequency and relation to other parkinsonian features have never been assessed. Objective: To study the relationships between gait festination and other parkinsonian clinical features among patients with Parkinson's disease (PD). Method: During an open lecture to patients with PD who are followed at the Movement Disorders Unit (MDU) of Tel-Aviv Sourasky Medical Center one of us explained verbally and imitated festinating gait on stage. All attending patients with the help of their care-givers or family members, were asked to answer two written questions regarding their own experience with FSG as well as the degree of disability it causes. Clinical information about each patient was taken from his/her chart at the MDU and missing data was completed during the next of®ce visit or from the family physician. Statistical analysis was performed using t-tests for comparison between groups, Chochran± Armitage test for trends and logistic regression to assess the contribution of age of onset, disease duration and disease severity to the development of FSG. Results: Eighty-one PD patients (58 males, mean age 67.5 ^ 10.7 years) answered the FSG questionnaire. Our study population's mean disease duration was 8.5 ^ 6.4 years, mean Hoehn and Yahr (H&Y) clinical stage of 2.6 ^ 0.8 and mean levodopa dose of 608 ^ 375 mg/ day (15 patients were not on levodopa). Twenty-six patients (32.1%) experienced FSG during the previous month and 56% of them reported that FSG was a signi®cant and disabling symptom. FSG was strongly associated with higher stage of H&Y p , 0:001 with a signi®cant trend as the disease progresses p 0:001 but not with total score in the motor part of the Uni®ed Parkinson's Disease Rating Scale (UPDRS). Longer disease duration was the only clinical factor, which was found to be associated with FSG in the multivariate model. Thirty seven percent (37%) of the patients with FSG reported frequent falls with association between occasional or frequent falls, as reported on the activity of daily living (ADL) part of the UPDRS, and the presence of FSG p , 0:08: There was no association between signi®cant postural re¯ex abnormalities as rated on the objective part of the UPDRS and the presence of FSG. There was a signi®cant association between the presence of freezing of gait (FOG) as reported in the ADL part of the UPDRS and the presence of FSG p , 0:001 as well as a signi®cant trend towards more frequent FSG in patients with more severe FOG p , 0:001: Conclusion: FSG was clearly associated with longer duration of PD symptoms but not with disease severity as re¯ected in the motor part of the UPDRS. The relationships between FSG and postural re¯exes abnormalities is unclear but it is frequently associated with falls and freezing of gait. q 2001 Elsevier Science Ltd. All rights reserved. Keywords: Parkinson's disease; Gait; Festination
1. Introduction Festinating gait (FSG) is one of the most typical and unique disturbance of locomotion associated with parkinsonism. FSG is described as; rapid, small steps, done in an attempt to keep the center of gravity (COG) in between the feet while the trunk leans forward involuntarily and shift the COG forward. In an attempt to correct balance, instead of making one or two large corrective steps, the patient makes * Corresponding author. Tel.: 1972-3-697-4912; fax: 1972-3-697-4911. E-mail address: [email protected] (N. Giladi).
hypometric steps, which leaves the COG in front of his feet. In order to compensate for the hypokinesia and in an attempt to prevent falling, there is an increase in stepping velocity to the state of running [1]. The clinical picture was best described by Sir J. Parkinson in his ®rst essay on ªThe Shaking Palsyº [2] ªThe propensity to lean forward becomes invincible, and the patient is thereby forced to step on the toes and fore part of the feet, whilst the upper part of the body is thrown so far forward as to render it dif®cult to avoid falling on the face. In some cases, when this state of the malady is attained, the patient can no longer exercise himself by walking in his usual
1353-8020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved. PII: S 1353-802 0(00)00030-4
136
N. Giladi et al. / Parkinsonism and Related Disorders 7 (2001) 135±138
Table 1 Comparison between patients with festination to those without festination of gait. (PD Parkinson's disease, UPDRS Uni®ed Parkinson's Disease Rating Scale) FSG (1) n 26 (32.1%)
FSG (2) n 55 (67.9%)
t/x 2 statistic
P
Mean current age (years) Mean age of symptoms onset (years) Mean disease duration (years) Mean total levodopa dose mg/day
66.7 ^ 11.6 n 24 54 ^ 13.1 n 24 12.7 ^ 7.4 n 24 650.5 ^ 529.3 n 24
67.9 ^ 10.3 n 54 61.2 ^ 10.6 n 54 6.7 ^ 5.0 n 54 454.4 ^ 332.4 n 54
0.446 2.58 24.2 2.58
N.S. 0.012 0.001 0.012
Hoehn and Yahr I II III IV
3.8 ^ 0.8 n 24 21% n 5 13% n 3 21% n 5 46% n 11
2.4 ^ 0.7 n 54 43% n 23 12% n 12 28% n 15 6% n 3
Total motor score of the UPDRS
30.2 ^ 12.8 n 24
manner, but is thrown on the toes and forepart of the feet; being, at the same time, irresistibly impelled to make much quicker and short steps, and thereby to adopt unwillingly a running pace. In some cases it is found necessary entirely to substitute running for walking; since otherwise the patient, on proceeding only a very few paces, would inevitably fallº. Festination is usually associated with gait but the same pattern of speeding up with decreased amplitude of repetitive automatic complex motor tasks have been observed in hand writing and speech to suggest a fundamental motor control abnormality in parkinsonism. The underlying mechanism responsible for this abnormal motor performance is not known and its relationships with the dopaminomimetic treatment is unclear. FSG is considered a manifestation of advanced Parkinson's disease (PD) but its frequency, and clinical association with other parkinsonian features have to the best of our knowledge, never been studied. This study was conducted in order to evaluate the relationships between FSG and other PD clinical features in an attempt to better characterize this unique symptom. 2. Methods During the annual New Year gathering with our PD patients we circulated a short ªFestination of Gait Questionnaireº (FSG-Q) to all participants. Patients were given verbal explanation about FSG. In order to make sure that everybody in the crowd understood exactly what was meant, FSG was imitated twice by one of us (N.G.) on the stage. Immediately following the demonstration, all patients were asked to mark the correct answer in the FSG-Q by selecting the best answer out of ®ve possibilities. Question #1 asked if FSG has ever been experienced, and its frequency over the last month. Question #2 asked whether falls were ever consequences of FSG and how often. Clinical information about all patients who answered the FSG-Q was collected using the Movement Disorders Unit (MDU) charts as well as outside documents. It is the routine at the MDU to
29.5 ^ 11.9 n 48
4.21
x 2 18.2 20.22
0.0001 0.001 N.S.
perform a complete Uni®ed Parkinson's Disease Rating Scale (UPDRS) [3] and staging at ªoffº and ªonº states according to the Hoehn and Yahr (H&Y) staging [4] every 6 months. The results of the last H&Y and UPDRS scores in the ªonº state from the charts were used for clinical correlation. We looked speci®cally for association between FSG and falls (item #13) or freezing of gait (FOG) (item #14) in the Activity of Daily Living (ADL) part of the UPDRS as well as the total score of the motor part (items #18±#31) (motor/UPDRS) and postural instability (item #30). For the analysis of the association between falls (item #13), FOG (#14) or postural instability (#30), we considered ªfallersº and ªfreezersº when the score on items #13 or #14 was > 1 and for the presence of abnormal postural instability if the score on item #30 was > 2 on the UPDRS. For calculation of trends in those three items, we used the actual score on the UPDRS. Patients with parkinsonism other than PD or those with any signi®cant cognitive decline were excluded from the analysis. Statistical analysis for comparison between groups of patients with FSG (FSG 1) and those without FSG (FSG 2 ) was performed using chi square and t-tests. For assessment of independent contribution of disease progression, disease duration or age of symptom onset to the development of FSG the uni-variate and multi-variate models of logistic regression were used. Trend was calculated using Cochran±Armitage test. Significance was considered if p , 0:05: 3. Results Eighty-one patients with PD have completed the questionnaire. Their mean age at the time of the study was 67.5 ^ 10.7 years, with a mean disease duration of 8.5 ^ 6.4 years and mean H&Y clinical stage of 2.6 ^ 0.8. Thirty-two (32%) percent of our study population reported that they experienced FSG at least several times over the last month. Patients with FSG n 26 and those without FSG n 55; were similar in current age (Table 1). However, patients with FSG had signi®cantly younger age
N. Giladi et al. / Parkinsonism and Related Disorders 7 (2001) 135±138
137
Table 2 The association between FSG and clinical features evaluated by univariate and multivariate logistic regression. (H&Y Hoehn and Yahr, UPDRS Uni®ed Parkinson's Disease Rating Scale, C.I. Con®dence interval, Motor/UPDRS Total score on the motor part of the UPDRS, FSG Festinating gait) Clinical features
Motor/UPDRS Age of onset Disease duration
Uni-variate analysis
Multi-variate model
Odds ratio
95% C.I.
P,
Odds ratio
95% C.I.
P,
1.00 0.95 1.17
0.96±1.05 0.91±0.991 1.07±1.28
NS 0.015 0.001
1.01 0.98 1.13
0.96±1.05 0.93±1.04 1.02±1.25
NS NS 0.01
of symptoms onset p 0:01; longer duration of PD symptoms p 0:001; higher stage of H&Y p 0:0001; and they were taking higher mean daily dose of levodopa p 0:01 (Table 1). In contrast, the total motor score on the UPDRS was not different between the group with FSG and the one without FSG. When we calculated the odds ratios for the risk to develop FSG using the uni-variate model, only disease duration and age of symptoms onset had signi®cant contribution to the development of FSG (Table 2). When the same assessment was performed with the multi-variate model, only longer disease duration was found to be a signi®cant risk for the development of FSG (Table 2). In addition, we applied the multi-variate model but instead of the total score on the motor part of the UPDRS we used Hoehn and Yahr staging as the parameter for disease progression. Disease duration lost its signi®cant contributing effect while age of symptoms onset had no effect as well. However, H&Y stage IV was a signi®cant risk factor for FSG in the uni-variate model P , 0:001 and the multi-variate model p , 0:01: The milder stages of H&Y (1±3) were not associated with FSG. There was no association between FSG and the occurrence of falls, as rated on the ADL part of the UPDRS or abnormal postural re¯exes as scored objectively on the motor part of the UPDRS (Table 3). However, 9 out of 26 patients (34.6%) who experienced FSG reported that they often fall because of festination. These patients had signi®cantly poorer postural re¯exes as rated on the UPDRS p 0:004: FSG was highly associated with the presence of freezing of gait (FOG), p 0:001: Interestingly, we also observed a signi®cant association between FSG and the presence of levodopa induced disturbing dyskinesias p 0:001: There was a direct relationships between frequency of FSG and
severity of FOG or dyskinesias as rated on the UPDRS (Table 3). 4. Discussion FSG has long been associated with parkinsonism but never been studied systematically in PD patients. In spite of its frequency and signi®cant effect on patient's mobility and quality of life, it is not included in the UPDRS. Because of its episodic nature and its infrequent appearance, its assessment objectively as part of the neurological examination is unreliable. This study is based on subjective information provided by the patients. We used a method of verbal explanation and actual demonstration of FSG in order to insure that all the participants received identical information at the same time. We believe that by using this method, we insured homogeneity among patients and more reliable information. Because of the lack of individual interaction with each patient, we cannot insure that all patients fully understood their task, but by excluding demented patients from the analysis we decreased this possibility. FSG was associated with stage 4 of the H&Y but not with higher total score on the motor part of the UPDRS. We suggest that the association with H&Y stage 4 is partly because signi®cant FSG can cause a patient to become stage 4 unrelated to progression of other symptoms. In other words, we believe that H&Y is an unreliable tool to assess general disease progression when one evaluates gait disturbances. To support our point, higher total motor score in the UPDRS, which is another way to assess disease severity taking into an account variety of parkinsonian signs, was not associated with the development of FSG. The present study demonstrated that FSG was not in
Table 3 The relationships between FSG and other parkinsonian symptoms. (FSG Festinating Gait, x 2 Chi square, NS Not signi®cant) Clinical features
FSG (1) n 26
FSG (2) n 55
Abnormal postural responses
15 (65.2%) 9 (37%) 18 (75%)
26 (48.1%) 16 (30%) 12 (22%)
Occurrence of falls Freezing of gait
x2
P,
Chochram Armitage trend test
p
0.52
NS
4.35
NS
21.408
0.08
0.001
25.136
0.001
19.5
NS
138
N. Giladi et al. / Parkinsonism and Related Disorders 7 (2001) 135±138
direct relation with the progression of PD but developed in a subgroup of patients who were prone to locomotor gait disturbances, unrelated to postural disturbances. In this subgroup of patients, longer disease duration is playing a signi®cant risk factor. At the present study as in previous one [5], FSG reported more frequently by patients with younger age of disease onset. This relationship was lost in the multi-variate logistic regression to suggest that it was an artifact. We speculate that this relationship is related to our impression that patients with younger age of symptoms onset attend patients' meetings at more advanced stages in comparison with older patients who are less motivated and frequently unable to come to such meetings as their disease is progressing. We were unable to differentiate in the present study between the contribution of the primary degenerative process and the additional effect of antiparkinsonian treatment to the development of FSG. This is a fundamental question because although FSG was described prior to the introduction of levodopa, it might theoretically be exacerbated by long-term levodopa treatment. Only long-term prospective studies will be able to differentiate between the independent contribution of the disease itself and the additional contribution of antiparkinsonian treatment. The mechanism of FSG is unclear. We propose that FSG is associated with the recently described increased stride-tostride variation. This phenomenon, which has been shown to be associated with advanced PD [6], can cause a step in the middle of locomotion to be unexpectedly too short. As a consequence, the trunk leans forward and the COG is placed in front of the legs. An attempt to bring back the COG in between the two legs is ineffective, possibly because of hypokinesia and hypometric compensatory corrective steps. The ®nal result is an initiation of rapid short steps (running) to avoid falls. If stride-to-stride variations is the primary pathology responsible for FSG, its poor relation to levodopa treatment [6] ®ts with our personal impression that FSG poorly responds to levodopa treatment and sometimes seen at the best ªonº. Stride to-stride variations can be the result of abnormal timing mechanisms of automatic sequential motor acts or inability to generate muscle forces in the legs during stepping at constant speed. Another possible explanation for stride-to-stride variation is impaired postural re¯exes and impaired body sway. In order to explore the possibility that disturbed postural re¯exes contribute to the development of FSG, we looked at the relationships between FSG and abnormal postural responses or falls. We could not ®nd an association between the two to suggest that FSG is not a phenomenon of disturbed postural responses. However, we demonstrated that falls when occurred as a result of FSG are more frequent among patients with abnormal postural responses.
FSG was highly associated with freezing of gait (FOG) with signi®cant trend as FOG became more disabling. This tight association suggests a similar pathophysiology for both. Similar association between FSG and FOG was also reported in the pure freezing syndrome [7±9] or primary progression freezing of gait [10] to support common pathophysiology. One possibility is that both are disturbances of central timing mechanism, which affect locomotion. As we already suggested, FSG might be the result of stride-tostride variation while FOG is associated with desynchronized activation of leg muscles and co-activation of antagonist muscles [11]. Whatever the exact mechanism responsible for these unique locomotion abnormalities, both are seen only in parkinsonism. In conclusion, FSG is a common symptom of advanced PD, which is part of locomotion disturbances. FSG is not caused by disturbed postural re¯exes but will lead to falls if associated with it. The close association with FOG suggests abnormality in the central timing mechanism with disturbance in stride-to-stride variation as possible major contributing factor.
References [1] Brown P, Steiger M. Basal ganglia gait disorders. In: Bronstein A, Brandt T, Woolacott M, editors. Balance posture and gait, New York, USA: Arnold with co-publishers Oxford University Press, 1996. pp. 156±67. [2] Parkinson J. An assay on the Shaking Palsy. London: Sherwood, Neely and Jones, 1817. [3] Fahn S, Elton R. Members of the UPDRS development committee. Uni®ed Parkinson's Disease Rating Scale. In: Fahn S, Marsden C, Calne D, Goldstein M, editors. Recent developments in Parkinson's disease, Florham Park, NJ: Macmillan Health Care Information, 1987. pp. 153±63. [4] Hoehn MM, Yahr MD. Parkinsonism: onset, progression, and mortality. Neurology 1967;50:318. [5] Giladi N. Gait disturbances in advanced stages of Parkinson's disease. In: Calne D, editor. Parkinson disease, Philadelphia, USA: Lippincott Williams and Wilkins, 2000 (in press). [6] Hausdorff JM, Cudkowicz ME, Firtion R, Wei JY, Goldberger AL. Gait variability and basal ganglia disorders: stride-to-stride variations of gait cycle timing in Parkinson's disease and Huntington's disease. Mov Disord 1998;13:428±37. [7] Imai H, Narabayashi H, Sakata E. ªPure akinesiaº and the later added supranuclear ophthalmoplegia. Adv Neurol 1987;45:207±12. [8] Imai H. Festination and freezing. Rinsho Shinkeigaku 1987;33:1307± 9. [9] Matsuo H, Takashima H, Kishikawa M, et al. Pure akinesia: an atypical manifestation of progressive supranuclear palsy. J Neurol Neurosurg Psychiat 1991;54:397±400. [10] Achiron A, Ziv I, Goren M, et al. Primary progressive freezing gait. Mov Disord 1993;8:293±7. [11] Andrews CJ. In¯uence of dystonia on the response to long-term Ldopa therapy in Parkinson's disease. J Neurol Neurosurg Psychiatr 1973;36:630±6.
www.elsevier.com/locate/parkreldis
Gait festination in Parkinson's disease N. Giladi a,*, H. Shabtai a, E. Rozenberg a, E. Shabtai b a
Movement Disorders Unit, Department of Neurology, Tel Aviv Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel b Statistical Services, Tel Aviv Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel Received 10 April 2000; revised 29 May 2000; accepted 30 May 2000
Abstract Background: Festinating gait (FSG) was ®rst associated with parkinsonism by Sir James Parkinson, in his original essay on ªThe Shaking Palsyº. Its frequency and relation to other parkinsonian features have never been assessed. Objective: To study the relationships between gait festination and other parkinsonian clinical features among patients with Parkinson's disease (PD). Method: During an open lecture to patients with PD who are followed at the Movement Disorders Unit (MDU) of Tel-Aviv Sourasky Medical Center one of us explained verbally and imitated festinating gait on stage. All attending patients with the help of their care-givers or family members, were asked to answer two written questions regarding their own experience with FSG as well as the degree of disability it causes. Clinical information about each patient was taken from his/her chart at the MDU and missing data was completed during the next of®ce visit or from the family physician. Statistical analysis was performed using t-tests for comparison between groups, Chochran± Armitage test for trends and logistic regression to assess the contribution of age of onset, disease duration and disease severity to the development of FSG. Results: Eighty-one PD patients (58 males, mean age 67.5 ^ 10.7 years) answered the FSG questionnaire. Our study population's mean disease duration was 8.5 ^ 6.4 years, mean Hoehn and Yahr (H&Y) clinical stage of 2.6 ^ 0.8 and mean levodopa dose of 608 ^ 375 mg/ day (15 patients were not on levodopa). Twenty-six patients (32.1%) experienced FSG during the previous month and 56% of them reported that FSG was a signi®cant and disabling symptom. FSG was strongly associated with higher stage of H&Y p , 0:001 with a signi®cant trend as the disease progresses p 0:001 but not with total score in the motor part of the Uni®ed Parkinson's Disease Rating Scale (UPDRS). Longer disease duration was the only clinical factor, which was found to be associated with FSG in the multivariate model. Thirty seven percent (37%) of the patients with FSG reported frequent falls with association between occasional or frequent falls, as reported on the activity of daily living (ADL) part of the UPDRS, and the presence of FSG p , 0:08: There was no association between signi®cant postural re¯ex abnormalities as rated on the objective part of the UPDRS and the presence of FSG. There was a signi®cant association between the presence of freezing of gait (FOG) as reported in the ADL part of the UPDRS and the presence of FSG p , 0:001 as well as a signi®cant trend towards more frequent FSG in patients with more severe FOG p , 0:001: Conclusion: FSG was clearly associated with longer duration of PD symptoms but not with disease severity as re¯ected in the motor part of the UPDRS. The relationships between FSG and postural re¯exes abnormalities is unclear but it is frequently associated with falls and freezing of gait. q 2001 Elsevier Science Ltd. All rights reserved. Keywords: Parkinson's disease; Gait; Festination
1. Introduction Festinating gait (FSG) is one of the most typical and unique disturbance of locomotion associated with parkinsonism. FSG is described as; rapid, small steps, done in an attempt to keep the center of gravity (COG) in between the feet while the trunk leans forward involuntarily and shift the COG forward. In an attempt to correct balance, instead of making one or two large corrective steps, the patient makes * Corresponding author. Tel.: 1972-3-697-4912; fax: 1972-3-697-4911. E-mail address: [email protected] (N. Giladi).
hypometric steps, which leaves the COG in front of his feet. In order to compensate for the hypokinesia and in an attempt to prevent falling, there is an increase in stepping velocity to the state of running [1]. The clinical picture was best described by Sir J. Parkinson in his ®rst essay on ªThe Shaking Palsyº [2] ªThe propensity to lean forward becomes invincible, and the patient is thereby forced to step on the toes and fore part of the feet, whilst the upper part of the body is thrown so far forward as to render it dif®cult to avoid falling on the face. In some cases, when this state of the malady is attained, the patient can no longer exercise himself by walking in his usual
1353-8020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved. PII: S 1353-802 0(00)00030-4
136
N. Giladi et al. / Parkinsonism and Related Disorders 7 (2001) 135±138
Table 1 Comparison between patients with festination to those without festination of gait. (PD Parkinson's disease, UPDRS Uni®ed Parkinson's Disease Rating Scale) FSG (1) n 26 (32.1%)
FSG (2) n 55 (67.9%)
t/x 2 statistic
P
Mean current age (years) Mean age of symptoms onset (years) Mean disease duration (years) Mean total levodopa dose mg/day
66.7 ^ 11.6 n 24 54 ^ 13.1 n 24 12.7 ^ 7.4 n 24 650.5 ^ 529.3 n 24
67.9 ^ 10.3 n 54 61.2 ^ 10.6 n 54 6.7 ^ 5.0 n 54 454.4 ^ 332.4 n 54
0.446 2.58 24.2 2.58
N.S. 0.012 0.001 0.012
Hoehn and Yahr I II III IV
3.8 ^ 0.8 n 24 21% n 5 13% n 3 21% n 5 46% n 11
2.4 ^ 0.7 n 54 43% n 23 12% n 12 28% n 15 6% n 3
Total motor score of the UPDRS
30.2 ^ 12.8 n 24
manner, but is thrown on the toes and forepart of the feet; being, at the same time, irresistibly impelled to make much quicker and short steps, and thereby to adopt unwillingly a running pace. In some cases it is found necessary entirely to substitute running for walking; since otherwise the patient, on proceeding only a very few paces, would inevitably fallº. Festination is usually associated with gait but the same pattern of speeding up with decreased amplitude of repetitive automatic complex motor tasks have been observed in hand writing and speech to suggest a fundamental motor control abnormality in parkinsonism. The underlying mechanism responsible for this abnormal motor performance is not known and its relationships with the dopaminomimetic treatment is unclear. FSG is considered a manifestation of advanced Parkinson's disease (PD) but its frequency, and clinical association with other parkinsonian features have to the best of our knowledge, never been studied. This study was conducted in order to evaluate the relationships between FSG and other PD clinical features in an attempt to better characterize this unique symptom. 2. Methods During the annual New Year gathering with our PD patients we circulated a short ªFestination of Gait Questionnaireº (FSG-Q) to all participants. Patients were given verbal explanation about FSG. In order to make sure that everybody in the crowd understood exactly what was meant, FSG was imitated twice by one of us (N.G.) on the stage. Immediately following the demonstration, all patients were asked to mark the correct answer in the FSG-Q by selecting the best answer out of ®ve possibilities. Question #1 asked if FSG has ever been experienced, and its frequency over the last month. Question #2 asked whether falls were ever consequences of FSG and how often. Clinical information about all patients who answered the FSG-Q was collected using the Movement Disorders Unit (MDU) charts as well as outside documents. It is the routine at the MDU to
29.5 ^ 11.9 n 48
4.21
x 2 18.2 20.22
0.0001 0.001 N.S.
perform a complete Uni®ed Parkinson's Disease Rating Scale (UPDRS) [3] and staging at ªoffº and ªonº states according to the Hoehn and Yahr (H&Y) staging [4] every 6 months. The results of the last H&Y and UPDRS scores in the ªonº state from the charts were used for clinical correlation. We looked speci®cally for association between FSG and falls (item #13) or freezing of gait (FOG) (item #14) in the Activity of Daily Living (ADL) part of the UPDRS as well as the total score of the motor part (items #18±#31) (motor/UPDRS) and postural instability (item #30). For the analysis of the association between falls (item #13), FOG (#14) or postural instability (#30), we considered ªfallersº and ªfreezersº when the score on items #13 or #14 was > 1 and for the presence of abnormal postural instability if the score on item #30 was > 2 on the UPDRS. For calculation of trends in those three items, we used the actual score on the UPDRS. Patients with parkinsonism other than PD or those with any signi®cant cognitive decline were excluded from the analysis. Statistical analysis for comparison between groups of patients with FSG (FSG 1) and those without FSG (FSG 2 ) was performed using chi square and t-tests. For assessment of independent contribution of disease progression, disease duration or age of symptom onset to the development of FSG the uni-variate and multi-variate models of logistic regression were used. Trend was calculated using Cochran±Armitage test. Significance was considered if p , 0:05: 3. Results Eighty-one patients with PD have completed the questionnaire. Their mean age at the time of the study was 67.5 ^ 10.7 years, with a mean disease duration of 8.5 ^ 6.4 years and mean H&Y clinical stage of 2.6 ^ 0.8. Thirty-two (32%) percent of our study population reported that they experienced FSG at least several times over the last month. Patients with FSG n 26 and those without FSG n 55; were similar in current age (Table 1). However, patients with FSG had signi®cantly younger age
N. Giladi et al. / Parkinsonism and Related Disorders 7 (2001) 135±138
137
Table 2 The association between FSG and clinical features evaluated by univariate and multivariate logistic regression. (H&Y Hoehn and Yahr, UPDRS Uni®ed Parkinson's Disease Rating Scale, C.I. Con®dence interval, Motor/UPDRS Total score on the motor part of the UPDRS, FSG Festinating gait) Clinical features
Motor/UPDRS Age of onset Disease duration
Uni-variate analysis
Multi-variate model
Odds ratio
95% C.I.
P,
Odds ratio
95% C.I.
P,
1.00 0.95 1.17
0.96±1.05 0.91±0.991 1.07±1.28
NS 0.015 0.001
1.01 0.98 1.13
0.96±1.05 0.93±1.04 1.02±1.25
NS NS 0.01
of symptoms onset p 0:01; longer duration of PD symptoms p 0:001; higher stage of H&Y p 0:0001; and they were taking higher mean daily dose of levodopa p 0:01 (Table 1). In contrast, the total motor score on the UPDRS was not different between the group with FSG and the one without FSG. When we calculated the odds ratios for the risk to develop FSG using the uni-variate model, only disease duration and age of symptoms onset had signi®cant contribution to the development of FSG (Table 2). When the same assessment was performed with the multi-variate model, only longer disease duration was found to be a signi®cant risk for the development of FSG (Table 2). In addition, we applied the multi-variate model but instead of the total score on the motor part of the UPDRS we used Hoehn and Yahr staging as the parameter for disease progression. Disease duration lost its signi®cant contributing effect while age of symptoms onset had no effect as well. However, H&Y stage IV was a signi®cant risk factor for FSG in the uni-variate model P , 0:001 and the multi-variate model p , 0:01: The milder stages of H&Y (1±3) were not associated with FSG. There was no association between FSG and the occurrence of falls, as rated on the ADL part of the UPDRS or abnormal postural re¯exes as scored objectively on the motor part of the UPDRS (Table 3). However, 9 out of 26 patients (34.6%) who experienced FSG reported that they often fall because of festination. These patients had signi®cantly poorer postural re¯exes as rated on the UPDRS p 0:004: FSG was highly associated with the presence of freezing of gait (FOG), p 0:001: Interestingly, we also observed a signi®cant association between FSG and the presence of levodopa induced disturbing dyskinesias p 0:001: There was a direct relationships between frequency of FSG and
severity of FOG or dyskinesias as rated on the UPDRS (Table 3). 4. Discussion FSG has long been associated with parkinsonism but never been studied systematically in PD patients. In spite of its frequency and signi®cant effect on patient's mobility and quality of life, it is not included in the UPDRS. Because of its episodic nature and its infrequent appearance, its assessment objectively as part of the neurological examination is unreliable. This study is based on subjective information provided by the patients. We used a method of verbal explanation and actual demonstration of FSG in order to insure that all the participants received identical information at the same time. We believe that by using this method, we insured homogeneity among patients and more reliable information. Because of the lack of individual interaction with each patient, we cannot insure that all patients fully understood their task, but by excluding demented patients from the analysis we decreased this possibility. FSG was associated with stage 4 of the H&Y but not with higher total score on the motor part of the UPDRS. We suggest that the association with H&Y stage 4 is partly because signi®cant FSG can cause a patient to become stage 4 unrelated to progression of other symptoms. In other words, we believe that H&Y is an unreliable tool to assess general disease progression when one evaluates gait disturbances. To support our point, higher total motor score in the UPDRS, which is another way to assess disease severity taking into an account variety of parkinsonian signs, was not associated with the development of FSG. The present study demonstrated that FSG was not in
Table 3 The relationships between FSG and other parkinsonian symptoms. (FSG Festinating Gait, x 2 Chi square, NS Not signi®cant) Clinical features
FSG (1) n 26
FSG (2) n 55
Abnormal postural responses
15 (65.2%) 9 (37%) 18 (75%)
26 (48.1%) 16 (30%) 12 (22%)
Occurrence of falls Freezing of gait
x2
P,
Chochram Armitage trend test
p
0.52
NS
4.35
NS
21.408
0.08
0.001
25.136
0.001
19.5
NS
138
N. Giladi et al. / Parkinsonism and Related Disorders 7 (2001) 135±138
direct relation with the progression of PD but developed in a subgroup of patients who were prone to locomotor gait disturbances, unrelated to postural disturbances. In this subgroup of patients, longer disease duration is playing a signi®cant risk factor. At the present study as in previous one [5], FSG reported more frequently by patients with younger age of disease onset. This relationship was lost in the multi-variate logistic regression to suggest that it was an artifact. We speculate that this relationship is related to our impression that patients with younger age of symptoms onset attend patients' meetings at more advanced stages in comparison with older patients who are less motivated and frequently unable to come to such meetings as their disease is progressing. We were unable to differentiate in the present study between the contribution of the primary degenerative process and the additional effect of antiparkinsonian treatment to the development of FSG. This is a fundamental question because although FSG was described prior to the introduction of levodopa, it might theoretically be exacerbated by long-term levodopa treatment. Only long-term prospective studies will be able to differentiate between the independent contribution of the disease itself and the additional contribution of antiparkinsonian treatment. The mechanism of FSG is unclear. We propose that FSG is associated with the recently described increased stride-tostride variation. This phenomenon, which has been shown to be associated with advanced PD [6], can cause a step in the middle of locomotion to be unexpectedly too short. As a consequence, the trunk leans forward and the COG is placed in front of the legs. An attempt to bring back the COG in between the two legs is ineffective, possibly because of hypokinesia and hypometric compensatory corrective steps. The ®nal result is an initiation of rapid short steps (running) to avoid falls. If stride-to-stride variations is the primary pathology responsible for FSG, its poor relation to levodopa treatment [6] ®ts with our personal impression that FSG poorly responds to levodopa treatment and sometimes seen at the best ªonº. Stride to-stride variations can be the result of abnormal timing mechanisms of automatic sequential motor acts or inability to generate muscle forces in the legs during stepping at constant speed. Another possible explanation for stride-to-stride variation is impaired postural re¯exes and impaired body sway. In order to explore the possibility that disturbed postural re¯exes contribute to the development of FSG, we looked at the relationships between FSG and abnormal postural responses or falls. We could not ®nd an association between the two to suggest that FSG is not a phenomenon of disturbed postural responses. However, we demonstrated that falls when occurred as a result of FSG are more frequent among patients with abnormal postural responses.
FSG was highly associated with freezing of gait (FOG) with signi®cant trend as FOG became more disabling. This tight association suggests a similar pathophysiology for both. Similar association between FSG and FOG was also reported in the pure freezing syndrome [7±9] or primary progression freezing of gait [10] to support common pathophysiology. One possibility is that both are disturbances of central timing mechanism, which affect locomotion. As we already suggested, FSG might be the result of stride-tostride variation while FOG is associated with desynchronized activation of leg muscles and co-activation of antagonist muscles [11]. Whatever the exact mechanism responsible for these unique locomotion abnormalities, both are seen only in parkinsonism. In conclusion, FSG is a common symptom of advanced PD, which is part of locomotion disturbances. FSG is not caused by disturbed postural re¯exes but will lead to falls if associated with it. The close association with FOG suggests abnormality in the central timing mechanism with disturbance in stride-to-stride variation as possible major contributing factor.
References [1] Brown P, Steiger M. Basal ganglia gait disorders. In: Bronstein A, Brandt T, Woolacott M, editors. Balance posture and gait, New York, USA: Arnold with co-publishers Oxford University Press, 1996. pp. 156±67. [2] Parkinson J. An assay on the Shaking Palsy. London: Sherwood, Neely and Jones, 1817. [3] Fahn S, Elton R. Members of the UPDRS development committee. Uni®ed Parkinson's Disease Rating Scale. In: Fahn S, Marsden C, Calne D, Goldstein M, editors. Recent developments in Parkinson's disease, Florham Park, NJ: Macmillan Health Care Information, 1987. pp. 153±63. [4] Hoehn MM, Yahr MD. Parkinsonism: onset, progression, and mortality. Neurology 1967;50:318. [5] Giladi N. Gait disturbances in advanced stages of Parkinson's disease. In: Calne D, editor. Parkinson disease, Philadelphia, USA: Lippincott Williams and Wilkins, 2000 (in press). [6] Hausdorff JM, Cudkowicz ME, Firtion R, Wei JY, Goldberger AL. Gait variability and basal ganglia disorders: stride-to-stride variations of gait cycle timing in Parkinson's disease and Huntington's disease. Mov Disord 1998;13:428±37. [7] Imai H, Narabayashi H, Sakata E. ªPure akinesiaº and the later added supranuclear ophthalmoplegia. Adv Neurol 1987;45:207±12. [8] Imai H. Festination and freezing. Rinsho Shinkeigaku 1987;33:1307± 9. [9] Matsuo H, Takashima H, Kishikawa M, et al. Pure akinesia: an atypical manifestation of progressive supranuclear palsy. J Neurol Neurosurg Psychiat 1991;54:397±400. [10] Achiron A, Ziv I, Goren M, et al. Primary progressive freezing gait. Mov Disord 1993;8:293±7. [11] Andrews CJ. In¯uence of dystonia on the response to long-term Ldopa therapy in Parkinson's disease. J Neurol Neurosurg Psychiatr 1973;36:630±6.