Treadmill training for the treatment of gait disturbances in people with Parkinson's disease: a mini-review
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
J Neural Transm (2009) 116:307–318
DOI 10.1007/s00702-008-0139-z
BASIC NEUROSCIENCES, GENETICS AND IMMUNOLOGY - REVIEW ARTICLE
Treadmill training for the treatment of gait disturbances in people
with Parkinson’s disease: a mini-review
T. Herman Æ N. Giladi Æ J. M. Hausdorff
Received: 2 September 2008 / Accepted: 6 October 2008 / Published online: 4 November 2008
Ó Springer-Verlag 2008
Abstract This report reviews recent investigations of the Keywords Parkinson’s disease Gait Treadmill
effects of treadmill training (TT) on the gait of patients Neuroplasticity
with Parkinson’s disease. A literature search identified
14 relevant studies. Three studies reported on the imme-
diate effects of TT; over-ground walking improved (e.g., Introduction
increased speed and stride length) after one treadmill ses-
sion. Effects persisted even 15 min later. Eleven longer- Treadmill training (TT) is only infrequently prescribed as a
term trials demonstrated feasibility, safety and efficacy, treatment option for patients with Parkinson’s disease
reporting positive benefits in gait speed, stride length and (PD). In contrast, bodyweight supported treadmill training
other measures such as disease severity (e.g., Unified (BWSTT) is often used to promote gait training in patients
Parkinson’s Disease Rating Scale) and health-related with spinal cord injuries and post-stroke (Hesse et al. 1995;
quality of life, even several weeks after cessation of the TT. Laufer et al. 2001; Behrman and Harkema 2000; Dietz
Long-term carryover effects also raise the possibility that et al. 1994; Dobkin et al. 2007; Dobkin 2005; van Hedel
TT may elicit positive neural plastic changes. While et al. 2006). The off loading of bodyweight allows for early
encouraging, the work to date is preliminary; none of the intervention and plays a critical role in this rehabilitation
identified studies received a quality rating of Gold or level process, enabling patients who have difficulties standing to
Ia. Additional high quality randomized controlled studies begin gait training. In PD, patients typically are able to
are needed before TT can be recommended with evidence- stand throughout most of the stages of the disease and they
based support. do not suffer from marked muscle weakness. The challenge
in PD is to improve motor control, but not necessarily to
target muscle strength. Perhaps, this explains why BWSTT
T. Herman N. Giladi J. M. Hausdorff (&) has not been widely applied to PD in the past. Nonetheless,
Laboratory for Gait and Neurodynamics, Movement Disorders recent studies have demonstrated the potential of TT
Unit and Parkinson Center, Department of Neurology, training in PD. Here we review this evidence and describe
Tel-Aviv Sourasky Medical Center, 6 Weizman Street,
the rationale for applying TT to PD (and perhaps other
64239 Tel Aviv, Israel
e-mail: jhausdor@bidmc.harvard.edu disorders that share similar symptoms).
Gait disturbances are an integral part of the clinical
J. M. Hausdorff manifestation of PD and among the most disabling symp-
Department of Physical Therapy, Sackler Faculty of Medicine,
toms of the disease. The gait of patients with PD is
Tel-Aviv University, Tel Aviv, Israel
typically marked by reduced speed, shortened stride length,
J. M. Hausdorff and longer double support phase (Ebersbach et al. 1999;
Division on Aging, Harvard Medical School, Boston, MA, USA Sofuwa et al. 2005; Morris et al. 1994). In addition, gait
dynamics are characterized by exaggerated stride-to-stride
N. Giladi
Department of Neurology, Sackler Faculty of Medicine, variability (Blin et al. 1990; Hausdorff et al. 1998;
Tel-Aviv University, Tel Aviv, Israel Schaafsma et al. 2003; Baltadjieva et al. 2006). This high
123308 T. Herman et al. stride-to-stride variability reflects a disturbance in gait and flexibility exercises. Despite its relatively long his- rhythmicity and an inconsistency of the locomotor pattern, tory, the evidence supporting the use of conventional PT and is a marker for increased fall risk (Hausdorff et al. for treatment of gait in PD is not very strong. Meta- 2001; Hausdorff 2005; Nakamura et al. 1996; Schaafsma analyses concluded that there is little evidence to support et al. 2003). The mobility problems related to gait distur- or refute the use of PT, in part because of methodological bances, instability and falls, have a profound negative flaws in published studies (de Goede et al. 2001; Ru- impact on patients’ quality of life and their mental well- benstein et al. 2002; Deane et al. 2001a, b, 2002; Keus being (de Boer et al. 1996; Martinez-Martin 1998). Despite et al. 2006). While traditional PT may help and benefit advances in pharmacological therapy and surgical proce- patients with PD, efficacy is limited in part because it dures, impairment in gait and balance remain common may not directly address the underlying deficits specific to in PD patients and the therapeutic options for treating PD and the nature of the disease. Application of external these gait disturbances and reducing fall risk in PD are not cueing in forms such as visual and rhythmic auditory yet ideal (Bloem et al. 2004; Grimbergen et al. 2004; cueing is an exciting area of research, but it is not Thurman et al. 2008), leaving much room for alternative yet widely used in clinical practice (Lim et al. 2005; interventions. Nieuwboer et al. 2007; Rubenstein et al. 2002; Hausdorff Anti-parkinsonian medications are probably the most et al. 2007). common means of treating parkinsonian symptoms. Briefly then, the three most common approaches to the Levodopa and dopamine agonists ameliorate many of these treatment of parkinsonian symptoms do not fully alleviate symptoms, including balance and gait disturbances, to the gait disturbances associated with PD. As we detail some degree. Certain balance and gait deficits are likely below, a growing body of relatively recent evidence related to the loss of central dopaminergic transmission, suggests that TT may be used as a complementary, however, levodopa is not very effective in treating all gait alternative option for treating these gait disturbances and abnormalities and preventing falls. In a prospective survey enhancing health-related quality of life. Moreover, a few of fall circumstances during daily life (Bloem et al. 2001), studies have demonstrated that TT may not only provide most falls occurred when patients were in their best clinical symptom relief; in addition, it may promote neuroplas- condition (‘‘on’’ state), possibly reflecting their increased ticity. The purpose of this paper is to review the literature mobility when symptoms are well controlled by medica- that supports this idea and the potential utility of TT in tions and/or the incomplete effectiveness of this therapy. PD. Quantitative studies of postural reactions to sudden per- turbations have also shown that dopamine agonists provide insufficient relief of many postural instability symptoms, Methods though some aspects partially improve in response to this anti-parkinsonian therapy (Bloem et al. 1996; Beckley Search strategy et al. 1995; Frank et al. 2000). Furthermore, levodopa and dopamine agonists may sometime worsen gait and provoke A search was conducted of the major electronic databases falls, e.g., in the case of ‘‘on’’ freezing and freezing of gait including MEDLINE, EMBASE, Web of Science, and in general. Cochrane. The key words searched for were: treadmill and Deep brain stimulation (DBS) is another therapeutic PD. We also looked for related studies using search terms option. When targeting the internal globus pallidus or sub- such as gait, rehabilitation, exercise, PT, and falls. Rele- thalamic nucleus, DBS can partially alleviate ‘‘off’’ related vant abstracts were reviewed and references cited were also gait impairment and instability in well-selected patients followed to check for additional related studies. A study with PD (Herzog et al. 2003; Krack et al. 2003; Stolze et al. was included in the present review if it met the following 2001; Maurer et al. 2003; Balash et al. 2005). The effects criteria: 1. The target population was patients with PD; are generally greatest for ‘‘off’’ state axial symptoms. A 2. The effects of TT intervention were tested (as the pri- particular concern is that several years after an initially mary intervention); 3. The paper was available in English; good response, some patients develop severe gait and and 4. The study was available as of August 1, 2008. balance deficits that are resistant to further therapeutic A study was excluded if: 1. The treadmill was used as an intervention (Krack et al. 2003; Ferraye et al. 2008). assessment tool rather than as a therapeutic, intervention Moreover, these invasive techniques are expensive, carry a tool; 2. A case report study (i.e., three subjects or less); certain degree of risk, and are generally recommended only 3. Animal experiments were studied. Because of the for a small subset of patients. relatively small numbers of studies identified, a formal Physical therapy (PT) is often prescribed to PD meta-analysis was not applied. No previous reviews or patients. It may include strength training, gait training, meta-analysis on this topic were identified. 123
Treadmill training on the gait of patients with Parkinson’s disease 309
effect lasted 15 min. Step
All subjects improved over-
ground GS after training;
Quality ranking
improved GS and stride
patients, independent of
prominent in advanced
length, but the control
length increases more
The treadmill decreased
gait variability in the
A single session of TT
intervention did not
Summary of findings
The quality of the identified studies that reported on the
long-term effects of TT were graded using two scales. The
‘‘Level of Evidence’’ was ranked using the standard scale
gait speed
patients
where A Ia is evidence providing the best recommendation
and highest level of evidence, while on the other end of the
spectrum, D 5 represents a low recommendation and low
TT treadmill training, GS gait speed, STT structured speed-dependent treadmill training, LTT limited progressive treadmill training, CGT conventional gait training
level of evidence. In addition, the grading system recom-
mended by the Cochrane Musculoskeletal Review Group
Cross-over design
(Tugwell et al. 2004) was applied as adapted by Bartels
et al. (2007). In this scheme, the ranking ranges from
Platinum (best), to Gold, Silver and Bronze. As summa-
rized by Bartels et al. Platinum is a published systematic
RCT
No
No
review that has at least two individual controlled trials each
satisfying Gold criteria. Gold refers to a randomized clin-
ical trial meeting all of the following criteria for the major
secondary outcomes
GS, gait variability
GS, stride length,
outcome(s) as reported: (a) Sample sizes of at least 50 per
double support
cadence, stride
GS, step length,
group—if these do not find a statistically significant dif-
variability
Primary and
ference, they are adequately powered for a 20% relative
difference in the relevant outcome. (b) Blinding of patients Table 1 Studies that examined the immediate effects of walking on a treadmill in patients with Parkinson’s disease
and assessors for outcomes. (c) Handling of withdrawals;
[80% follow up (imputations based on methods such as
LOCF are acceptable) and (d) Concealment of treatment
Follow-up
allocation. Silver: randomized trials that do not meet the
above criteria. Silver ranking would also refer to a study of
No
No
No
non-randomized cohorts that did and did not receive the
therapy, or evidence from at least one high quality case–
control study. Bronze: The bronze ranking is given to high
days, 30 min each
Duration/frequency
quality case series without controls (including simple
Four consecutive
*10 min TT
Single session
Single session
before/after studies in which patients act as their own
20 min TT
control) or if the conclusion is derived from expert opinion
based on clinical experience without reference to any of
day
the foregoing (for example, argument from physiology,
bench research or first principles).
conditions: over-ground,
control: four conditions,
with a walker, treadmill
Over-ground walking and
Results
STT, LTT, CGT and
TT three groups:
Walking at three
No BW support
No BW support
No BW support
Fourteen studies were identified that met the inclusion and
cross-over
Intervention
2. advanced
1. moderate
exclusion criteria (see Tables 1, 2). Here we briefly sum-
3. control
marize some of the salient features of these investigations.
Immediate effects
Three studies examined the immediate effects of TT (see
Number of PD patients/
Table 1). Using a cross-over, four consecutive-day trial in
Frenkel-Toledo et al.
17 patients with early PD, Pohl et al. (2003) found that gait
30 Healthy controls
8 Healthy controls
Bello et al. (2008)
Pohl et al. (2003)
speed and stride length can be improved through a single
17 PD patients
intervention of TT (even without body weight support), but
(2005b)
not through conventional gait training. Two training modes
reference
30 PD
16 PD
were used: Structured speed-dependent treadmill training
(STT) and limited progressive treadmill training (LTT). In
123Table 2 Studies of long-term effects of treadmill training in patients with Parkinson’s disease
310
Number of PD Intervention Duration/ Follow- Primary and secondary RCT Level of Grading of Summary of findings
patients/reference frequency up outcomes evidence evidence
123
10 PD patients BW support TT vs. PT 1 month No UPDRS, GS, stride length No cross-over design C IIb Bronze TT with different amount of
Miyai et al. (2000) 45 min/day BWS improved ADL,
motor performance, and
3 days/week
ambulation, more than
PT
20 PD in total 20% BW support TT vs. PT 1 month 5 months UPDRS, GS, stride length Yes B IIa Silver TT group had greater
11 intervention 45 min/day improvement in GS than
PT group at 1 month;
9 control 3 days/week
improvement in stride
Miyai et al. (2002) length at 1 and 4 months
post-training
23 PD patients 3 groups: 6 weeks 1 month BBS, UPDRS, strength and No C IIb Bronze Improvements in SOT,
Toole et al. (2005) (1) TT no load 20 min/day ROM, sensory balance, BBS, UPDRS
organization test/ (Motor Exam), and gait
(2) BW -25% 3 days/week
dynamic posturography parameters for the three
(3) BW ?5%
groups regardless of the
body-weight loading
18 PD in total Intervention group had TT 8 weeks No Falls, GS, step length, five- Yes B IIa Silver Training resulted in a
9 intervention at a greater speed then 60 min/day steps test reduction in falls and
over-ground while improved gait
9 control 3 days/week
walking in four performance and
Protas et al. (2005) directions (no BW Not attention dynamic balance
support) vs. control only controlled
pre/post-testing
7 PD Trained without BW 10 sessions 1 month UPDRS, GS, No C III Bronze All walking parameters and
Pelosin et al. (2007) support 30 min/day Timed up and go, QOL improved
(Abstract) significantly.
3 days/week 6-min walk, fatigue
Cardiopulmonary
severity,
capacity increased.
PDQ-39, Improvement persisted
VO2 max after 1 month
31 PD in total Incremental speed- 8 weeks No Treadmill walking distance Yes B IIa Silver Mobility, postural stability,
21 intervention dependent TT and 30 min/day and speed, Not clear how and balance-confidence
stretching and ROM vs. BBS, DGI, FES attention improved significantly
10 control
control (BW support not controlled after the training
Cakit et al. (2007) mentioned) program
9 PD Intensive-progressive 6 weeks 1 month UPDRS, GS, stride length, No C IIb Bronze UPDRS, gait speed and
Herman et al. (2007) training without BW 30 min/day Timed up and go, gait stride length, mobility
support variability, and QOL improved.
4 days/week
Some gains at 5 weeks
PDQ-39
follow up
T. Herman et al.Table 2 continued
Number of PD Intervention Duration/ Follow- Primary and RCT Level of Grading of Summary of findings
patients/reference frequency up secondaryoutcomes evidence evidence
30 PD in total 3 groups: High and low: No UPDRS,GS, stride length, Yes B Ib Silver All improve in UPDRS;
10 per group High intensity-treadmill 8 weeks Sit-to-stand, ground In high-intensity group:
Fisher et al. (2008) (10% BW support); 1 h/day reaction force, improvement in GS,
Low intensity-exercise; Bio-mechanic properties, stride length, joint
3 days/week;
CSP excursion, lower
Zero intensity-education Zero: six extremity symmetry,
education Lengthening CSP
sessions
5 PD 3-month progressive TT as 12 weeks No UPDRS, falls and near- No C III Bronze UPDRS, GS and home
Skidmore et al. AE (no BW support) 10–20 min falls, regular and fast GS, ambulation improved.
(2008) SBP drops, activity Aerobic program is feasible
3 days/week
monitoring in PD. Best to monitor
autonomic function
24 PD in total 2 groups of instructed ROM 6 weeks No Timed lower extremities Yes B Ib Silver Significant improvement in
12 intervention exercise. Intervention 40 min/day tasks, physical tasks, fitness and self-
group also TT (BW not assessment and ergo- report of physical well-
12 control 3 days/week
mentioned) spirometric test being, and ergo-
Treadmill training on the gait of patients with Parkinson’s disease
Kurtais et al. (2008) spirometric parameters
20 PD in total 2 groups: 6 weeks 6 weeks 6-min walk test, PDQ-39, Yes C IIb Bronze Improvement in QOL and
8 completed TT Home-based TT vs. control 30–40 min/day fatigue Not attention reduced fatigue. No
controlled effect on walking
10 controls group maintain activity 3 days/week
level capacity. Home-based
(Abstract) treadmill training is
Canning et al. feasible and safe
(2008)
TT treadmill training, GS gait speed, STT structured speed-dependent treadmill training, LTT limited progressive treadmill training, CGT conventional gait training, PT physical therapy, ROM
range of motion, BBS Berg balance scale, BW body weight, FES falls efficacy scale, PDQ-39 Parkinson’s disease questionnaire, QOL quality of life, UPDRS Unified Parkinson’s Disease Rating
Scale, CSP cortical silent period, SOT sensory organization test, DGI dynamic gait index, SBP systolic blood pressure, AE aerobic exercise, RCT randomized controlled trial
311
123312 T. Herman et al.
STT, the treadmill speed was increased by 10% each time in motor performance compared to conventional PT,
the patient successfully completed 10 s of walking at a set increasing gait speed (from 1.0 ± 0.1 to 1.2 ± 0.1 m/s),
speed. In LTT, the initial self-adapted over-ground speed stride length (from 0.89 ± 0.09 to 1.02 ± 0.11 m) and
was used without increasing treadmill speed. Over-ground reducing parkinsonian symptoms [motor part of the Unified
gait speed improved from 1.37 ± 0.23 to 1.56 ± 0.23 m/s Parkinson’s Disease Rating Scale (UPDRS) decreased
(P \ 0.001) and from 1.35 ± 0.21 to 1.52 ± 0.20 m/s (improved) from 18.2 ± 1.4 to 15.0 ± 1.3] (Miyai et al.
(P \ 0.001) after SST and LTT, respectively. Similarly, 2000). While promising, a key question left open by this
stride length improved from 0.72 ± 0.12 to 0.78 ± 0.14 m study is whether the improvement in gait was due to the
and from 0.71 ± 0.10 to 0.77 ± 0.09 m (P \ 0.001), after load reduction or to the TT itself. A follow-up randomized
SST and LTT, respectively. Similarly, Bello et al. (2008) controlled trial by the same group demonstrated a long-
found that a single session of 20 min of treadmill walking term effect of BWSTT on gait, after the TT was completed,
increased over-ground gait speed and step length, espe- beyond that of conventional PT, which lasted for about
cially in more advanced patients. Interestingly, these 4 months (Miyai et al. 2002). For example, the BWSTT
positive effects persisted as much as 15 min later. Frenkel- group had significantly (P \ 0.005) greater improvement
Toledo et al. (2005b) assessed the influence of a single than the PT group in gait speed immediate post-training. In
session of TT on stride-to-stride variability. While walking the BWSTT group, gait speed was 0.92 ± 0.07 m/s and
on a treadmill, gait variability was reduced, even when increased to 1.18 ± 0.09 m/s, while in the group that
walking at the same over-ground speed. This suggests that received only PT, gait speed was 0.87 ± 0.13 m/s at
the treadmill was able to generate a more consistent gait baseline and 0.92 ± 0.15 m/s post-training..
pattern with a reduced stride-to-stride variability, a marker Several other studies have found positive effects of TT
related to fall risk (Schaafsma et al. 2003; Hausdorff et al. in PD and indicate that the degree of weight bearing may
1997, 2001, 2003; Baltadjieva et al. 2006). not be critical for achieving benefits in PD (Toole et al.
These findings indicate that treadmill walking can pro- 2005; Cakit et al. 2007). Cakit et al. used a TT protocol of
mote a faster and a more stable walking pattern in patients 8 weeks and reported promising effects on multiple aspects
with PD and suggest that perhaps an intervention program of gait and balance, including a reduction in fear of falling
that includes long-term treadmill walking—without using (Cakit et al. 2007), even though body weight was not
body weight support—might enhance gait speed, restore supported. Toole et al. (Toole et al. 2005) studied the
rhythmicity, reduce gait variability and perhaps succeed at effects of 6 weeks of treadmill walking in 23 subjects with
lowering fall risk. As can be seen below, long-term inter- PD. The subjects were divided into three intervention
vention studies support this notion. groups and each group used different amounts of body
weight support including one group that trained without
Long-term intervention studies any body weight support. Muscle strength did not change,
but significant improvements in balance and gait were seen
Eleven studies examined the effects of intensive TT with in all three groups, regardless of the degree of weight
the training durations ranging from 3.5 to 12 weeks (Miyai bearing.
et al. 2000, 2002; Toole et al. 2005; Protas et al. 2005;
Pelosin et al. 2007; Cakit et al. 2007; Herman et al. 2007; Carry over effects from Long-term Studies
Fisher et al. 2008; Skidmore et al. 2008; Kurtais et al.
2008; Canning et al. 2008) (see Table 2). All studies Five studies examined carry over effects after the cessa-
trained at a frequency of three times per week for about 20– tion of the TT. Three studies re-tested subjects about
40 min each session, except for one study which trained 4 weeks after the intervention was completed (Toole et al.
four times a week (Herman et al. 2007). Six studies were 2005; Herman et al. 2007; Pelosin et al. 2007), one study
defined as randomized controlled studies, but in three of conducted a follow up after 6 weeks (Canning et al.
these it was not clear exactly if and how attention was 2008), and one study examined the long-term, carry over
controlled. All studies reported some benefits immediately effect 5 months after termination of the training (Miyai
post-training, compared to pre-training values and/or et al. 2002). These studies all suggest that many of the
compared to changes seen in a control group. effects persisted even after the patients no longer used a
In the first report on TT in PD, Miyai et al. (2000) treadmill. For example, in the study by Miyai et al. pre-
investigated the effects of 20% BWSTT on gait and par- training stride length was 0.85 ± 0.08 m, and immediate
kinsonian symptoms of PD patients. Patients were post-BWSTT stride length increased to 1.00 ± 0.10 m.
supported by a harness (taking partial body weight to Significant gains in stride length were still observed
reduce the gravitational load on the legs). In this 4-week 1 month (1.02 ± 0.09 m) and 2 months (0.99 ± 0.09 m)
cross-over study, BWSTT produced greater improvement after training.
123Treadmill training on the gait of patients with Parkinson’s disease 313
Quality of life and outcomes beyond gait and balance Adverse events were generally not observed. In the
study conducted by Skidmore et al. (2008) in eight patients,
A number of studies have demonstrated effects that extend two patients were referred to a cardiologist for asymp-
beyond motor function, balance and gait. For example, tomatic ST segment depression and one for symptomatic
Herman et al. found that the PD-Q39, a measure of quality of blood pressure drop as part of the screening process. Based
life, was enhanced even 4 weeks after cessation of the on the findings in their intervention study, which included a
intervention (Herman et al. 2007). Herman et al. (Herman high intensity treadmill protocol, Fisher et al. (2008) sug-
et al. 2007), Toole et al. (Toole et al. 2005), Skidmore et al. gested that subjects with PD can be challenged and tolerate
(Skidmore et al. 2008) and Fisher et al. (Fisher et al. 2008) all a very high-intensity TT. Furthermore, even a home-based
reported significant immediate effects of TT on the UPDRS, treadmill walking procedure appeared to be well-tolerated
the most widely used measure of symptom severity in PD by patients with PD with only minimal levels of muscle
(lower scores are better). Compared to pre-training values, soreness and/or stiffness, and no report of adverse events
improvements in the UPDRS were seen weeks after com- (Canning et al. 2008). Cakit et al. (2007) noted that from
pletion of the training. In one study, UPDRS motor scores among the 21 PD patients in the TT group, two showed
were 29.0 ± 9.3 pre-training and 19.7 ± 6.4 about 5 weeks ventricular extrasystole as a cardiac symptom during a
after the intervention was completed (P = 0.027) (Herman training session, but they were not dropped from the study
et al. 2007). TT also apparently enhances walking confi- and completed it without additional side effects. Pohl et al.
dence and reduces fear of falling (Cakit et al. 2007). (2003) also specifically commented on the absence of
cardiac symptoms (e.g., angina pectoris, significant
Safety and feasibility arrhythmias) and other side effects (e.g., dizziness, muscle
or joint trouble) in response to treadmill walking.
Treadmill training may elevate heart rate (HR) and blood The safety of TT in patients with PD is supported by
pressure and expose subjects to cardiovascular stress. studies that examined the cardiovascular response to
Several studies explicitly state that subjects were excluded exercise in PD. In a study of the effects of treadmill
if they exhibited postural hypotension, uncontrolled high exercise, HR, systolic blood pressure, and the Rate of
blood pressure or cardiovascular disorders (Toole et al. Perceived Exertion were measured during a Modified
2005; Cakit et al. 2007; Skidmore et al. 2008; Fisher et al. Bruce Protocol in patients with PD and healthy controls
2008; Herman et al. 2007). For example, all subjects in the (Werner et al. 2006). During sub-maximal exercise, no
study by Cakit et al. were screened with a stress test using significant differences were found between the PD group
the Modified Bruce Protocol to determine cardiovascular and the control group for HR, blood pressure, or rate of
tolerance. In addition, a few studies also monitored HR and perceived exertion. At peak exercise, one half of the sub-
blood pressure during sessions on the TT (Toole et al. jects with PD exhibited blunted cardiovascular responses,
2005; Skidmore et al. 2008; Fisher et al. 2008). In most despite reaching a comparable intensity of exercise during
studies, the explicit goal was not to elevate HR, instead the a Modified Bruce Protocol. This finding of a reduced car-
focus was on gait. In the few trials that measured HR on the diovascular response is consistent with other reports
treadmill, target HR was generally relatively low (e.g., less (Barbic et al. 2007; Oka et al. 2006; Persico et al. 1998)
than 60% of theoretical HR capacity; (Toole et al. 2005)) and supports the relatively low cardiovascular risks from
indicating that the TT was not applied as an aerobic TT in patients with PD.
exercise. Other studies do not specify the steps taken to
reduce cardiovascular risk.
A total of 260 patients with PD were enrolled in these 14 Discussion
studies that examined the effects of TT on walking. Of
those, 63 patients took part in a single session trial. One As shown in Table 2, there tends to have been a positive
hundred and thirty-five patients completed screening and evolution of the investigations that have been performed to
baseline testing for participation in an extended TT program examine the potential utility of TT in patients with PD.
and began training. Of these, 125 patients completed the Although progress was not always linear and much work
intervention, i.e., 93% of the patients who began a TT still needs to be done, maturation of the research can be
program completed it successfully. Reported reasons for seen with respect to the number of subjects studied, the
drop-outs included loss to follow-up and health problems questions posed, and general study design. TT is relatively
not related to TT. This relatively good compliance may in established as a rehabilitation paradigm for stroke patients
part be explained by a comment by Toole et al. (2005) who and for patients with spinal cord injuries (Hesse et al. 1995;
noted that ‘‘there was 99% adherence to the program. Par- Laufer et al. 2001; Behrman and Harkema 2000; Dietz
ticipants attended even when they did not feel their best.’’ et al. 1994; Dobkin et al. 2007; Dobkin 2005; van Hedel
123314 T. Herman et al.
Table 3 Possible mechanisms
Mechanisms Supporting studies Studies suggesting that this is not involved
underlying the efficacy of
treadmill training in Parkinson’s Aerobic exercise Skidmore et al. (2008) All except for the Skidmore study
disease
Strength training Toole et al. (2005)
Pace retraining Skidmore et al. (2008)
Frenkel-Toledo et al. (2005a, b)
Herman et al. (2007)
Body-weight support Fisher et al. (2008) Herman et al. (2007)
Bello et al. (2008)
Miyai et al. (2000) Toole et al. (2005)
Miyai et al. (2002) Pelosin et al. (2007)
Motor learning Fisher et al. (2008)
Protas et al. (2005)
Many of these mechanisms are Corticomotor exitability Fisher et al. (2008)
not mutually exclusive
et al. 2006). A growing body of intriguing preliminary expressed as a more rhythmic generation of gait cycles
work suggests that gait in PD, although a neurodegenera- (Frenkel-Toledo et al. 2005b). Therefore, a treadmill may
tive disease, is also amenable to change via TT. Dopamine be viewed as an external pacemaker (Frenkel-Toledo et al.
deficits and neuronal loss in PD cause a reduced stride 2005a, b). One possible mechanism of gait motor com-
length and gait dysrhythmicity while TT apparently may mand placement is that rhythmic stimulation is introduced
enhance gait speed, augment rhythmicity and boost walk- to the motor system via the somatosensory pathways.
ing steadiness. Moreover, these improvements in gait Treadmill walking rhythmically stimulates pressure load
apparently have widespread benefits for multiple domains receptors of the feet, muscle spindles, and Golgi tendon
of QOL. organs (Toole et al. 2005). Additional sensory inputs which
become more rhythmic include the afferent vestibular input
Putative mechanisms of treadmill training from the otolith organs (because of pitch head movements)
and the input from neck muscle proprioceptors (because of
As summarized in Table 3, a number of mechanisms may head relative to trunk movements) which are rhythmically
explain the observed benefits of TT in PD. Of course, these activated during treadmill walking (Hirasaki et al. 1999).
are not mutually exclusive. Treadmill walking may activate These rhythmic inputs are transferred to neuronal circuits
neuronal circuits that mediate central pattern generators modulating gait throughout different CNS levels and
(Van de Crommert et al. 1998), i.e., the paced, rhythmic facilitate the pacing of gait. Through these processes,
motor commands that activate limb muscles repetitively. neuronal plasticity and reinforcement may occur and aug-
Enhanced motor learning after deficient motor activity may mented pace-making may be preserved in the neuronal
also play a role. Some have suggested that such learning is circuits performing gait, even for long periods of time,
either reinforced at the synaptic connections in the spinal depending in part on the intensity of the training.
cord level or the trigger for re-organization of neural net- Enhanced corticomotor excitability at the completion of
works (Muir and Steeves 1997; Bello et al. 2008). Walking the TT was also observed in patients with PD (Fisher et al.
on a moving walkway (i.e., a treadmill) inherently provides 2008) and this intriguing finding is consistent with studies
external cueing that is mediated through proprioceptive and on animal models of PD that demonstrated neuroplasticity
vestibular receptors; this in turn generates repetitive sen- after TT (Fisher et al. 2004; Petzinger et al. 2007; Yoon
sory input to the central nervous system (Mathiowetz and et al. 2007). Such changes and the observed carry over
Haugen 1994). Interestingly, treadmill walking also acti- effects weeks after the TT intervention is completed are
vated the cerebellar vermis in a SPECT study of PD consistent with the possibility that TT may elicit neuro-
patients (Hanakawa et al. 1999). plastic improvements in motor function in patients with PD
Unlike over-ground walking where ongoing fluctuations and that TT provides more than just simple symptom relief.
in gait speed are unconstrained, on the treadmill, in order to The basal ganglia assist in the regulation and control of
successfully maintain walking, gait speed must follow the over-learned rhythmic movements like gait. Given the
speed of the treadmill. With the treadmill, there is no basal ganglia impairment in PD, it was previously thought
getting around it. The patient must match his/her pace to that training is not capable of producing long-term effects.
the treadmill constantly. As a consequence, gait speed While the theory behind the empirical data is not fully
becomes more uniform and regular. This uniformity is understood, recent evidence from other motor tasks
123Treadmill training on the gait of patients with Parkinson’s disease 315
suggests that this is not the case (Graybiel 2005). Rather, TT could potentially be applied for aerobic exercise, the
patients with PD can learn ‘‘automatic’’ motor tasks, per- results to date support the idea that the observed effects on
haps using other pathways to compensate (Kelly et al. gait and mobility are not achieved as a result of aerobic
2004; Wu and Hallett 2005). Although not always at the exercise.
same level as that of healthy controls, a number of studies
have shown that implicit learning—the type of motor Potential clinical implications
learning that may explain the observed benefits of TT—is
still possible in PD (Forkstam and Petersson 2005; Gua- The present review indicates that walking on a treadmill
dagnoli et al. 2002; Kelly et al. 2004; Perretta et al. 2005; apparently improves gait, mobility, and quality of life of
Smith et al. 2001; Wu and Hallett 2005). For example, van patients with PD. TT may promote a more stable walking
Hedel and colleagues evaluated the acquisition and per- pattern in patients with PD and an intervention program
formance of a high precision locomotor task in patients that includes intensive long-term treadmill walking appears
with PD, compared to healthy subjects (van Hedel et al. to be able to re-store stride length and rhyhmicity, key
2006). Initially, PD patients performed poorer and deficits to the gait of PD patients (Bloem et al. 2004;
improved foot clearance more slowly. However, after task Morris et al. 1994; Schaafsma et al. 2003), even while off
repetition, the groups performed similarly, indicating that the treadmill. In the presence of PD, many see the goal as
adequate training can improve locomotor behavior in PD maintenance care, but do not see the feasibility or oppor-
patients (van Hedel et al. 2006). tunity for rehabilitation. This review contributes to a
The rationale behind intensive and progressive forms of growing body of evidence that demonstrates that rehabili-
TT, despite the presence of neurodegeneration, is sup- tation-like program may be efficacious even in the presence
ported by studies using high intensity voice treatment. of such a devastating neurodegenerative disease like PD.
There are several features that underlie the voice and gait Some of the studies (Miyai et al. 2000, 2002) suggest
disorders common to PD: reduction in amplitude, a prob- that TT is more effective than conventional approaches to
lem with the sensory perception of effort, and deficient improving gait characteristics associated with PD. While
internal cueing causes difficulty in generation of appro- questions remain, if TT is considered, for select appropriate
priate effort. To some extent, parallel deficiencies can be patients, we suggest combining conventional PT with
found in gait (e.g., shortened stride length as an expression intensive TT using a protocol of at least three sessions per
of small amplitude; impaired rhythmicity as a deficient week, 20–30 min each. For safety reasons, it is recom-
internal cueing). Ramig and colleagues have shown that PD mended to exclude patients with uncontrolled cardiac
patients can be trained to work around these deficits in conditions or unstable medications, and patients that are
speech, and that the training effects can last at least two unable to walk (with or without assistance) for moderate
years after cessation of an initial intensive training period distances. Although TT is typically sub-maximal and not
(Liotti and Ramig 2003; Ramig et al. 2001a, b, 2004). In an aerobic form of exercise, standard recommendations
addition, imaging studies have shown that this treatment about safety and screening for cardiac risk should be fol-
produces changes in recruitment patterns and regional lowed before a patient begins a new exercise program. In
cerebral blood flow, consistent with neuroplastic addition, consistent with the reports to date, it is advised
improvements (Liotti et al. 2003). The parallels between that all training sessions take place in the ‘‘on’’ state. It is
voice and gait suggest that re-training of gait may also be probably important that the physical therapist ensures a
achievable. The patient learns to adapt to progressively ‘‘normal’’ gait pattern while walking on the treadmill.
increasing demands—a process that may enhance the However, if any gait deviations occur or if there are signs
automatization of motor control. Motor learning may of pain or fatigue complaints, treadmill speed should be
explain the carry over effect of TT: the treadmill trains a adjusted accordingly. It is advised that patients start the
steady gait speed and paces gait on a subconscious level training while holding onto the parallel bars and gradually
while pacing cannot be ignored. move to walking while holding on with one hand and
A priori, aerobic exercise is another possible mechanism eventually to walking without holding on at all.
whereby TT may produce benefits. If cardiovascular In PD, there appears to be no need to unload the patient,
function is markedly impaired, gait and mobility will be unless specific issues arise. One has to take into account
affected. Conversely, improvement in aerobic capacity can that due to the relatively high cost of a body-weight support
lead to enhanced mobility. While some of the animal treadmill, the need for a relatively large facility and
studies that found positive effects of TT were performed at increased time commitment, a TT program based on a
a high intensity, most of the studies on TT in patients with medical treadmill (with a safety harness) may not be
PD were performed at exertion levels far below those practical for everyone or for everyday use in the clinic.
typically recommended for aerobic exercise. Thus, while However, a harness used just to prevent falls may be less
123316 T. Herman et al.
expensive and more widely available, compared to a spe- of life. Hopefully the results of the present review will set
cial partial weight bearing system. Furthermore, important the stage for larger scale, randomized controlled studies that
safety issues should be considered when thinking about definitively establish efficacy and long-term, carry over
prescribing TT in the home-setting for patients with PD. effects and directly measure the effects of TT on outcomes
such as quality of life and fall risk, a major cause of mor-
Limitations and future work bidity and functional dependence in PD.
The findings of TT to date are very promising. Still, a Acknowledgments This work was supported in part by NIH grants
AG-14100, by the National Parkinson Foundation and by the Euro-
critical look at the existing literature raises many questions. pean Union Sixth Framework Program, FET contract no. 018474-2,
Most investigations included relatively small sample sizes Dynamic Analysis of Physiological Networks (DAPHNet) and
that limit the ability to generalize the findings. This limi- STREP 045622 (SENSing and ACTION to support mobility in
tation is especially poignant when examining long-term, Ambient Assisted Living, SENSACTION-AAL).
carry over effects. Overall, very few patients were moni-
tored more than 1-month post-training. A number of the TT
reports had no control group, did not control for attention References
and the placebo effect (an especially potent effect in PD), or
did not include any active comparison. None of the identi- Balash Y, Peretz C, Leibovich G, Herman T, Hausdorff JM, Giladi N
fied studies reached a grade of Gold and the level of (2005) Falls in outpatients with Parkinson’s disease: frequency,
impact and identifying factors. J Neurol 252:1310–1315
evidence was below A Ia. It is also helpful to keep in mind Baltadjieva R, Giladi N, Gruendlinger L, Peretz C, Hausdorff JM
that the current literature may reflect a publication bias (2006) Marked alterations in the gait timing and rhythmicity of
since studies that reported positive effects are more likely to patients with de novo Parkinson’s disease. Eur J Neurosci
be published. Moreover, before TT can be combined with 24:1815–1820
Barbic F, Perego F, Canesi M, Gianni M, Biagiotti S, Costantino G,
other forms of therapy (e.g., muscle strengthening), likely Pezzoli G, Porta A, Malliani A, Furlan R (2007) Early
the optimal form of therapy, there is a need for larger scale abnormalities of vascular and cardiac autonomic control in
randomized controlled studies that demonstrate the advan- Parkinson’s disease without orthostatic hypotension. Hyperten-
tages of TT compared to carefully designed control groups sion 49:120–126
Bartels EM, Lund H, Hagen KB (2007) Aquatic exercise for
that receive similar amounts of attention. Questions about the treatment of knee and hip osteoarthritis (review). Cochrane
ideal delivery forms, dosage, frequency, intensity and the Libr Database Syst Rev 4:CD005523
intervention prescription and their effects on compliance Beckley DJ, Panzer VP, Remler MP, Ilog LB, Bloem BR (1995)
and outcome measures also should be addressed in follow- Clinical correlates of motor performance during paced postural
tasks in Parkinson’s disease. J Neurol Sci 132:133–138
up studies. For example, at the moment, little is known Behrman AL, Harkema SJ (2000) Locomotor training after human
about the effects of TT on falls. Larger scale trials will also spinal cord injury: a series of case studies. Phys Ther 80:688–700
enable us to examine if there are patients with specific PD Bello O, Sanchez JA, Fernandez-Del-Olmo M (2008) Treadmill
characteristics that are more likely to respond to TT and the walking in Parkinson’s disease patients: adaptation and gener-
alization effect. Mov Disord 23(9):1243–1249
role of disease severity on TT efficacy. In addition, in Blin O, Ferrandez AM, Serratrice G (1990) Quantitative analysis of
the future, it may be helpful to more directly address the gait in Parkinson patients: increased variability of stride length.
question of whether TT provides symptomatic relief only or J Neurol Sci 98:91–97
whether it has the potential to modify disease progression. If Bloem BR, Beckley DJ, van Dijk JG, Zwinderman AH, Remler MP,
Roos RA (1996) Influence of dopaminergic medication on
additional work supports the latter, more remote possibility, automatic postural responses and balance impairment in Parkin-
treatment with TT early in the disease progress may become son’s disease. Mov Disord 11:509–521
the ideal. Bloem BR, Grimbergen YA, Cramer M, Willemsen M, Zwinderman
AH (2001) Prospective assessment of falls in Parkinson’s
disease. J Neurol 248:950–958
Bloem BR, Hausdorff JM, Visser JE, Giladi N (2004) Falls and
Conclusions freezing of gait in Parkinson’s disease: a review of two
interconnected, episodic phenomena. Mov Disord 19:871–884
Despite these limitations and the many unresolved ques- Cakit BD, Saracoglu M, Genc H, Erdem HR, Inan L (2007) The
effects of incremental speed-dependent treadmill training on
tions, the existing studies support the idea that TT in PD is postural instability and fear of falling in Parkinson’s disease.
safe, feasible, and likely to be efficacious and suggest that Clin Rehabil 21:698–705
TT could play an important role for improving gait and Canning CG, Allen NE, Fung VSC, Morris JGL, Dean CM (2008)
mobility in PD, as it does in other patient groups. TT may be Home-based treadmill walking for individuals with Parkinson’s
disease: a pilot randomized controlled trial. Mov Disord 23:S210
used as an adjunct treatment to complement PT, to improve de Boer AG, Wijker W, Speelman JD, de Haes JC (1996) Quality of
physical performance, enhance gait stability and possibly life in patients with Parkinson’s disease: development of a
reduce fall risk in PD patients, potentially improving quality questionnaire. J Neurol Neurosurg Psychiatry 61:70–74
123Treadmill training on the gait of patients with Parkinson’s disease 317
de Goede CJ, Keus SH, Kwakkel G, Wagenaar RC (2001) The effects Hausdorff JM (2005) Gait variability: methods, modeling and
of physical therapy in Parkinson’s disease: a research synthesis. meaning. J NeuroEng Rehabil 2. doi:10.1186/1743-0003-2-19
Arch Phys Med Rehabil 82:509–515 Hausdorff JM, Cudkowicz ME, Firtion R, Wei JY, Goldberger AL
Deane KH, Ellis-Hill C, Jones D, Whurr R, Ben Shlomo Y, Playford (1998) Gait variability and basal ganglia disorders: stride-to-
ED, Clarke CE (2002) Systematic review of paramedical stride variations of gait cycle timing in Parkinson’s disease and
therapies for Parkinson’s disease. Mov Disord 17:984–991 Huntington’s disease. Mov Disord 13:428–437
Deane KH, Jones D, Ellis-Hill C, Clarke CE, Playford ED, Ben Hausdorff JM, Edelberg HK, Mitchell SL, Goldberger AL, Wei JY
Shlomo Y (2001a) A comparison of physiotherapy techniques (1997) Increased gait unsteadiness in community-dwelling
for patients with Parkinson’s disease. Cochrane Database Syst elderly fallers. Arch Phys Med Rehabil 78:278–283
Rev 1:CD002815 Hausdorff JM, Lowenthal J, Herman T, Gruendlinger L, Peretz C,
Deane KH, Jones D, Playford ED, Ben Shlomo Y, Clarke CE (2001b) Giladi N (2007) Rhythmic auditory stimulation modulates gait
Physiotherapy for patients with Parkinson’s disease: a compar- variability in Parkinson’s disease. Eur J Neurosci 26:2369–2375
ison of techniques. Cochrane Database Syst Rev 3:CD002817 Hausdorff JM, Rios D, Edelberg HK (2001) Gait variability and fall
Dietz V, Colombo G, Jensen L (1994) Locomotor activity in spinal risk in community-living older adults: a 1-year prospective
man. Lancet 344:1260–1263 study. Arch Phys Med Rehabil 82:1050–1056
Dobkin B, Barbeau H, Deforge D, Ditunno J, Elashoff R, Apple D, Hausdorff JM, Schaafsma JD, Balash Y, Bartels AL, Gurevich T,
Basso M, Behrman A, Harkema S, Saulino M, Scott M (2007) Giladi N (2003) Impaired regulation of stride variability in
The evolution of walking-related outcomes over the first Parkinson’s disease subjects with freezing of gait. Exp Brain Res
12 weeks of rehabilitation for incomplete traumatic spinal cord 149:187–194
injury: the multicenter randomized Spinal Cord Injury Locomo- Herman T, Giladi N, Gruendlinger L, Hausdorff JM (2007) Six weeks
tor Trial. Neurorehabil Neural Repair 21:25–35 of intensive treadmill training improves gait and quality of life in
Dobkin BH (2005) Clinical practice. Rehabilitation after stroke. patients with Parkinson’s disease: a pilot study. Arch Phys Med
N Engl J Med 352:1677–1684 Rehabil 88:1154–1158
Ebersbach G, Sojer M, Valldeoriola F, Wissel J, Muller J, Tolosa E, Herzog J, Volkmann J, Krack P, Kopper F, Potter M, Lorenz D,
Poewe W (1999) Comparative analysis of gait in Parkinson’s Steinbach M, Klebe S, Hamel W, Schrader B, Weinert D, Muller
disease, cerebellar ataxia and subcortical arteriosclerotic enceph- D, Mehdorn HM, Deuschl G (2003) Two-year follow-up of
alopathy. Brain 122(7):1349–1355 subthalamic deep brain stimulation in Parkinson’s disease. Mov
Ferraye MU, Debu B, Pollak P (2008) Deep brain stimulation effect Disord 18:1332–1337
on freezing of gait. Mov Disord 23(Suppl 2):S489–S494 Hesse S, Bertelt C, Jahnke MT, Schaffrin A, Baake P, Malezic M,
Fisher BE, Petzinger GM, Nixon K, Hogg E, Bremmer S, Meshul CK, Mauritz KH (1995) Treadmill training with partial body weight
Jakowec MW (2004) Exercise-induced behavioral recovery and support compared with physiotherapy in nonambulatory hemi-
neuroplasticity in the 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydro- paretic patients. Stroke 26:976–981
pyridine-lesioned mouse basal ganglia. J Neurosci Res 77:378– Hirasaki E, Moore ST, Raphan T, Cohen B (1999) Effects of walking
390 velocity on vertical head and body movements during locomo-
Fisher BE, Wu AD, Salem GJ, Song J, Lin CH, Yip J, Cen S, Gordon tion. Exp Brain Res 127:117–130
J, Jakowec M, Petzinger G (2008) The effect of exercise training Kelly SW, Jahanshahi M, Dirnberger G (2004) Learning of ambig-
in improving motor performance and corticomotor excitability in uous versus hybrid sequences by patients with Parkinson’s
people with early Parkinson’s disease. Arch Phys Med Rehabil disease. Neuropsychologia 42:1350–1357
89(7):1221–1229 Keus SH, Bloem BR, Hendriks EJ, Bredero-Cohen AB, Munneke M
Forkstam C, Petersson KM (2005) Towards an explicit account of (2006) Evidence-based analysis of physical therapy in Parkin-
implicit learning. Curr Opin Neurol 18:435–441 son’s disease with recommendations for practice and research.
Frank JS, Horak FB, Nutt J (2000) Centrally initiated postural Mov Disord 22:451–460
adjustments in parkinsonian patients on and off levodopa. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C,
J Neurophysiol 84:2440–2448 Koudsie A, Limousin PD, Benazzouz A, LeBas JF, Benabid AL,
Frenkel-Toledo S, Giladi N, Peretz C, Herman T, Gruendlinger L, Pollak P (2003) Five-year follow-up of bilateral stimulation of
Hausdorff JM (2005a) Effect of gait speed on gait rhythmicity the subthalamic nucleus in advanced Parkinson’s disease. N Engl
in Parkinson’s disease: variability of stride time and swing J Med 349:1925–1934
time respond differently. J NeuroEng Rehabil 2. doi:10.1186/ Kurtais Y, Kutlay S, Tur BS, Gok H, Akbostanci C (2008) Does
1743-0003-2-23 treadmill training improve lower-extremity tasks in Parkinson
Frenkel-Toledo S, Giladi N, Peretz C, Herman T, Gruendlinger L, disease? A randomized controlled trial. Clin J Sport Med
Hausdorff JM (2005b) Treadmill walking as an external 18:289–291
pacemaker to improve gait rhythm and stability in Parkinson’s Laufer Y, Dickstein R, Chefez Y, Marcovitz E (2001) The effect of
disease. Mov Disord 20:1109–1114 treadmill training on the ambulation of stroke survivors in the
Graybiel AM (2005) The basal ganglia: learning new tricks and early stages of rehabilitation: a randomized study. J Rehabil Res
loving it. Curr Opin Neurobiol 15:638–644 Dev 38:69–78
Grimbergen YA, Munneke M, Bloem BR (2004) Falls in Parkinson’s Lim I, Van WE, de Goede C, Deutekom M, Nieuwboer A, Willems
disease. Curr Opin Neurol 17:405–415 A, Jones D, Rochester L, Kwakkel G (2005) Effects of external
Guadagnoli MA, Leis B, Van Gemmert AW, Stelmach GE (2002) rhythmical cueing on gait in patients with Parkinson’s disease: a
The relationship between knowledge of results and motor systematic review. Clin Rehabil 19:695–713
learning in Parkinsonian patients. Parkinsonism Relat Disord Liotti M, Ramig L (2003) Improvement of voicing in patients with
9:89–95 Parkinson’s disease by speech therapy. Neurology 61:1316–1317
Hanakawa T, Katsumi Y, Fukuyama H, Honda M, Hayashi T, Kimura Liotti M, Ramig LO, Vogel D, New P, Cook CI, Ingham RJ, Ingham
J, Shibasaki H (1999) Mechanisms underlying gait disturbance JC, Fox PT (2003) Hypophonia in Parkinson’s disease: neural
in Parkinson’s disease: a single photon emission computed correlates of voice treatment revealed by PET. Neurology
tomography study. Brain 122(7):1271–1282 60:432–440
123318 T. Herman et al.
Martinez-Martin P (1998) An introduction to the concept of ‘‘quality Ramig LO, Sapir S, Countryman S, Pawlas AA, O’Brien C, Hoehn M,
of life in Parkinson’s disease’’. J Neurol 245(1):S2–S6 Thompson LL (2001a) Intensive voice treatment (LSVT) for
Mathiowetz V, Haugen JB (1994) Motor behavior research: impli- patients with Parkinson’s disease: a 2 year follow up. J Neurol
cations for therapeutic approaches to central nervous system Neurosurg Psychiatry 71:493–498
dysfunction. Am J Occup Ther 48:733–745 Ramig LO, Sapir S, Fox C, Countryman S (2001b) Changes in vocal
Maurer C, Mergner T, Xie J, Faist M, Pollak P, Lucking CH (2003) loudness following intensive voice treatment (LSVT) in indi-
Effect of chronic bilateral subthalamic nucleus (STN) stimula- viduals with Parkinson’s disease: a comparison with untreated
tion on postural control in Parkinson’s disease. Brain 126:1146– patients and normal age-matched controls. Mov Disord 16:79–83
1163 Rubenstein TC, Giladi N, Hausdorff JM (2002) The power of cueing
Miyai I, Fujimoto Y, Ueda Y, Yamamoto H, Nozaki S, Saito T, Kang to circumvent dopamine deficits: a review of physical therapy
J (2000) Treadmill training with body weight support: its effect treatment of gait disturbances in Parkinson’s disease. Mov
on Parkinson’s disease. Arch Phys Med Rehabil 81:849–852 Disord 17:1148–1160
Miyai I, Fujimoto Y, Yamamoto H, Ueda Y, Saito T, Nozaki S, Kang Schaafsma JD, Giladi N, Balash Y, Bartels AL, Gurevich T,
J (2002) Long-term effect of body weight-supported treadmill Hausdorff JM (2003) Gait dynamics in Parkinson’s disease:
training in Parkinson’s disease: a randomized controlled trial. relationship to Parkinsonian features, falls and response to
Arch Phys Med Rehabil 83:1370–1373 levodopa. J Neurol Sci 212:47–53
Morris ME, Iansek R, Matyas TA, Summers JJ (1994) The Skidmore FM, Patterson SL, Shulman LM, Sorkin JD, Macko RF
pathogenesis of gait hypokinesia in Parkinson’s disease. Brain (2008) Pilot safety and feasibility study of treadmill aerobic
117(Pt 5):1169–1181 exercise in Parkinson disease with gait impairment. J Rehabil
Muir GD, Steeves JD (1997) Sensorimotor stimulation to improve Res Dev 45:117–124
locomotor recovery after spinal cord injury. Trends Neurosci Smith J, Siegert RJ, McDowall J, Abernethy D (2001) Preserved
20:72–77 implicit learning on both the serial reaction time task and
Nakamura T, Meguro K, Sasaki H (1996) Relationship between falls artificial grammar in patients with Parkinson’s disease. Brain
and stride length variability in senile dementia of the Alzheimer Cogn 45:378–391
type. Gerontology 42:108–113 Sofuwa O, Nieuwboer A, Desloovere K, Willems AM, Chavret F,
Nieuwboer A, Kwakkel G, Rochester L, Jones D, Van WE, Willems Jonkers I (2005) Quantitative gait analysis in Parkinson’s
AM, Chavret F, Hetherington V, Baker K, Lim I (2007) Cueing disease: comparison with a healthy control group. Arch Phys
training in the home improves gait-related mobility in Parkin- Med Rehabil 86:1007–1013
son’s disease: the RESCUE trial. J Neurol Neurosurg Psychiatry Stolze H, Klebe S, Poepping M, Lorenz D, Herzog J, Hamel W,
78:134–140 Schrader B, Raethjen J, Wenzelburger R, Mehdorn HM, Deuschl
Oka H, Mochio S, Onouchi K, Morita M, Yoshioka M, Inoue K G, Krack P (2001) Effects of bilateral subthalamic nucleus
(2006) Cardiovascular dysautonomia in de novo Parkinson’s stimulation on parkinsonian gait. Neurology 57:144–146
disease. J Neurol Sci 241:59–65 Thurman DJ, Stevens JA, Rao JK (2008) Practice parameter:
Pelosin E, Faelli E, Lofrano F, Marinelli L, Bove M, Ruggeri P, assessing patients in a neurology practice for risk of falls (an
Abbruzzese G (2007) Treadmill training improves functional evidence-based review): report of the Quality Standards Sub-
ability and cardiopulmonary capacity in stable Parkinson’s committee of the American Academy of Neurology. Neurology
disease. Mov Disord 22:S175 70:473–479
Perretta JG, Pari G, Beninger RJ (2005) Effects of Parkinson disease Toole T, Maitland CG, Warren E, Hubmann MF, Panton L (2005)
on two putative nondeclarative learning tasks: probabilistic The effects of loading and unloading treadmill walking on
classification and gambling. Cogn Behav Neurol 18:185–192 balance, gait, fall risk, and daily function in Parkinsonism.
Persico AM, Reich S, Henningfield JE, Kuhar MJ, Uhl GR (1998) NeuroRehabilitation 20:307–322
Parkinsonian patients report blunted subjective effects of meth- Tugwell P, Shea B, Boers M (2004) Evidence based rheumatology.
ylphenidate. Exp Clin Psychopharmacol 6:54–63 BMJ Books, Boston, USA
Petzinger GM, Walsh JP, Akopian G, Hogg E, Abernathy A, Arevalo Van de Crommert HW, Mulder T, Duysens J (1998) Neural control of
P, Turnquist P, Vuckovic M, Fisher BE, Togasaki DM, Jakowec locomotion: sensory control of the central pattern generator and
MW (2007) Effects of treadmill exercise on dopaminergic its relation to treadmill training. Gait Posture 7:251–263
transmission in the 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyr- van Hedel HJ, Waldvogel D, Dietz V (2006) Learning a high-
idine-lesioned mouse model of basal ganglia injury. J Neurosci precision locomotor task in patients with Parkinson’s disease.
27:5291–5300 Mov Disord 21:406–411
Pohl M, Rockstroh G, Ruckriem S, Mrass G, Mehrholz J (2003) Werner WG, Francisco-Donoghue J, Lamberg EM (2006) Cardio-
Immediate effects of speed-dependent treadmill training on gait vascular response to treadmill testing in Parkinson disease.
parameters in early Parkinson’s disease. Arch Phys Med Rehabil J Neurol Phys Ther 30:68–73
84:1760–1766 Wu T, Hallett M (2005) A functional MRI study of automatic
Protas EJ, Mitchell K, Williams A, Qureshy H, Caroline K, Lai EC movements in patients with Parkinson’s disease. Brain
(2005) Gait and step training to reduce falls in Parkinson’s 128:2250–2259
disease. NeuroRehabilitation 20:183–190 Yoon MC, Shin MS, Kim TS, Kim BK, Ko IG, Sung YH, Kim SE,
Ramig LO, Fox C, Sapir S (2004) Parkinson’s disease: speech and Lee HH, Kim YP, Kim CJ (2007) Treadmill exercise suppresses
voice disorders and their treatment with the Lee Silverman Voice nigrostriatal dopaminergic neuronal loss in 6-hydroxydopamine-
Treatment. Semin Speech Lang 25:169–180 induced Parkinson’s rats. Neurosci Lett 423:12–17
123You can also read
