Exercise Training on Locomotion in Patients with Alzheimer’s Disease: A Feasibility Study

Abstract

Background:

Although current literature has shown that patients with Alzheimer’s disease (AD) have worse locomotion compared with healthy counterparts, no studies have focused on the efficacy of exercise training in improving gait abnormalities including biomechanics and metabolic aspects, in this population.

Objective:

To verify the effectiveness of exercise training (ET) on gait parameters (i.e., speed, step and stride length, single and double support, and energy cost of walking (Cw)) in patients with AD with respect to a standard cognitive treatment (CT).

Methods:

In this study, we included a small portion of data belonging to a larger study (ClinicalTrials.gov number, NCT03034746). Patients with AD (Mini-Mental State Examination 22±5) were included in the study. Gait parameters and Cw were assessed at baseline and after 6 months (72 treatment sessions) of treatment. ET included 90 min of aerobic and strength training. CT included 90 min of cognitive stimuli.

Results:

The 16 patients assigned to ET exhibited significant improvement of Cw (–0.9±0.1 J/kg·m^-1), while differences in gait parameters were negligible. The effect on gait parameters were undetectable in the 18 patients assigned to CT (–0.2±0.5 J/kg·m^-1).

Conclusions:

Data from this study showed that ET program seems effective in improving Cw in patients with AD. Interestingly, the positive effect of ET on Cw was not coupled with ameliorations of patient’s gait parameters, suggesting that the gain of metabolic aspects of locomotion were the main factors responsible for this positive result.

Keywords

Dementia energy cost of walking physical activity spatio-temporal gait parameters

INTRODUCTION

Alzheimer’s disease (AD) is the most common neurodegenerative disorder in the elderly population, and is characterized by intraneuronal fibrillary tangles and extracellular deposit of the protein amyloid-β (Aβ) [1]. Deposits of Aβ in multiple neuronal systems and brain areas lead to cortical dysfunctions which result in a plethora of complex cognitive impairments [2].

Motor dysfunction is often observed in individuals with AD, likely due to AD-related cortical and subcortical impairments [3]. Moreover, it has been observed that patients with AD likely activate adaptive motor strategies in order to control the complicated interactions between cognitive (i.e., executive and attentional function) and motor tasks [4 –6]. Indeed, patients with AD when compared with healthy-counterparts are characterized by gait abnormalities, that may cause an aggravation of the energy cost of walking (Cw) [7]. Cw is the net rate of oxygen uptake (VO₂) per body mass and distance covered, and it reflects the energy cost associated with muscle activation in order to maintain balance, coordination and posture meanwhile the body is moving forward [7]. By definition, Cw is not affected only by its biomechanical component (spatio-temporal gait parameters), but even by its cardio-metabolic component, which in turn reflect central and peripheral limiting factors. The former reflects the efficiency of the cardiovascular system to supply blood through the exercising muscles; the latter reflects the efficiency of skeletal muscles and mitochondria to extract and utilize oxygen [8]. In recent literature it has been underlined that both these components of the VO₂ are reduced in older adults, and further decreased in patients with AD [9]. Consequently, when biomechanical and/or cardio-metabolic components of the Cw are deteriorated, the efficiency of ambulation decreases reflecting a bigger amount of energy needed for walking. However, no studies have proven that this process is exacerbated in patients with AD.

Several studies have aimed to identify the effect of exercise intervention on Cw in older healthy people, and in other populations with neurodegenerative pathologies [10, 11], demonstrating significant improvements in gait parameters and cardiovascular response to exercise. However, the effect of exercise training on Cw in people with AD is currently unknown. Therefore, the aim of this study was to compare the effects of an exercise training (ET) with respect to a conventional cognitive treatment (CT) on Cw in patients with AD. We hypothesized that an ET program may ameliorate Cw in patients with AD.

METHODS

Study overview

In this study, we included some data that are included in a larger study (ClinicalTrials.gov number, NCT03034746).

For this single-blind study, 356 individuals with AD were screened for eligibility. Three-hundred three people were excluded due to presence of medical exclusion factors or because they were not interested. Fifty-three people were included in the study and evaluated for spatio-temporal gait parameters, and Cw. Subsequently, patients with AD were assigned to ET group (n = 26) or CT group (n = 27). Participants were evaluated again after six months of treatment. The post-treatment Cw test was performed at the same speed utilized during the pre-treatment test (Fig. 1). Research team included “outcome evaluators” and “treatment givers”. Outcome evaluators were uninformed about group assignments, and included physicians, neuropsychologist, and other medical personnel who performed all the test and measurements. Treatment givers included neuropsychologist and kinesiologists who administered CT and ET, respectively. None of the treatment givers took part in the assessment procedures.

Fig.1

Participant flow diagram. AD, patients with Alzheimer’s disease.

Subjects

Fifty-three patients with AD were recruited at the Neuromotor and Cognitive Rehabilitation Research Center (Department of Neurosciences, Biomedicine and Movement Sciences), Azienda Ospedaliera Universitaria Integrata of Verona (AOUI Verona), and at the Mons. A. Mazzali Foundation, Mantua, Italy. Clinical diagnosis of AD was established according to the National Institute of Neurological and Communicative Disorders and Stroke – Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) work group criteria for probable AD. Each pair of patients with AD of the same gender, were randomly assigned one to ET and one to CT group, in order to balance the groups (Randomization.com). All subjects underwent laboratory investigations. Exclusion criteria were: a history of depression or psychosis, alcohol or drug abuse, other neurological condition (e.g., multiple sclerosis, Parkinson’s disease, brain injuries, stroke), known cardiac, orthopedic (e.g., osteoarthrosis), or respiratory pathology (e.g., chronic obstructive pulmonary disease). All experiments were conducted after informed consent was obtained from the patients and their relatives in accordance with the Declaration of Helsinki, as part of a protocol approved by the Institutional Review Board of the Azienda Ospedaliera Universitaria Integrata (Verona, Italy – #2389; NIH Clinical trial identification number: NCT03034746).

Spatio-temporal gait parameters

Gait analysis was conducted by means of GaitRite System (GAITRite, MAP/CIR INC, Haverton, USA), an electronic system used for the gathering of the temporal-spatial data of ambulation. Subjects were instructed to walk 4 times on a 10-meter instrumented walkway and the average of the 4 trials was taken as a result. The following parameters were detected: speed, stride length, single support (phase of the gait cycle during which only one foot is in ground contact; higher value = better outcome), and double support (phase of the gait cycle during which both feet are in ground contact; higher value = worse outcome) [11].

Self-selected speed on treadmill

Patients were instructed to find the self-selected speed (SSS, the most comfortable speed) on the treadmill (Run Race, Technogym, Gambettola, Italy). The test started from 0.5 km/h and the speed was increased by 0.1 km/h every 30 seconds until the patients were walking comfortably [12]. Due to the possible difficulty in walking on a treadmill for unexperienced people, a 30-min familiarization session was conducted one week prior the test.

Cost of walking (Cw)

Subjects performed a 3-speed walking test, on a treadmill (Run Race, Technogym, Gambettola, Italy). A metabolimeter (K4b², Cosmed, Rome, Italy) was used for cardiopulmonary measures. The test consisted of two phases: in the first phase, the subjects were asked to stand in resting condition for 2 min meanwhile the resting oxygen uptake (VO_2rest) was recorded. The second phase consisted of three 5-min bouts of walking at 80% (WS1), 100% (WS2), and 120% (WS3) of SSS, respectively. The initial minutes of walking secured a steady state VO₂. For any bout, the average of the last minute VO₂ measurements, respiratory exchange ratio (measured by means of the metabolimeter) [12], and heart rate (HR) were calculated [11].

C_w was then calculated by means of the following formula [13]: $\begin{matrix} C_{w} (J / kg m^{- 1}) = (({VO}_{2 ex} - {VO}_{2 rest}) / v) \\ \times ((4.94 \times RER) + 16.06) / 1000 \end{matrix}$

Exercise and cognitive treatments

ET included 90 min three times a week of moderate intensity endurance and resistance training. Sessions started with 15 min of warm up which included active joint mobilization and walking on treadmill at preferred speed. Then, patients performed two 15-min endurance exercises (either on cycle ergometer, or treadmill, or arm cranking ergometer, aiming to spend the same amount of time on the different ergometers during the week) at 70% of maximal heart rate (calculated using the Karvonen formula: 220-age in years). Then, patients performed 3 sets of 12 repetitions of resistance exercises at 85% of 1 repetition maximum (1RM). 1RM was determined by means of Brzycki method for all the isotonic ergometers included in the training (chest press, lat machine, leg press. Technogym, Gambettola, Italy) during the first training session. ET ended with stretching exercises for all the muscle involved in the training. All training sessions were supervised by kinesiologist, with a ratio of 2:5.

CT, conducted by neuropsychologist, included 90 min of multi-modal stimuli based on the Cognitive Stimulation Therapy (visual, verbal, auditive, tactile) in order to train residual abilities of the patients, in particular attention and memory functions using sensorial material and repeated short-term tasks. Specifically, it began with a warm-up activity, required to introduce each subject to the other members of the group. This was gentle, aiming to provide continuity and orientation by beginning all sessions in the same way. Sessions focusing on themes allowed the natural process of reminiscence but had an additional focus on the current day. Multisensory stimulation was introduced. Sessions encouraged the retrieval of impaired skills and the stimulation of residual skills with the acquisition of external aids. Each session focused on the stimulation of a specific domain, primarily memory, but also other functions such as attention, language, and executive functions, taking into account the group’s cognitive capabilities, interests, and gender mix [14]. Both treatments were scheduled three times a week for 6 months, for an amount of 72 sessions. Moreover, to guarantee a high adherence rate, when participants could not attend a scheduled session for any reason, the missed session could be retrieved another day.

Statistical analysis

As we reported previously, these data are part of a larger study. Consequently, sample size was calculated based on the Mini-Mental State Examination (MMSE). Indeed, to obtain a significant effect size of 2 MMSE points [14], a sample size of 90 participants was chosen to guarantee a statistical power higher than 0.80. Raw data were analyzed using a statistical software package (IBM SPSS Statistics v. 23, Armonk, NY, USA). Between-group differences in demographic variables were tested with a Student’s t-test. The normal distribution of the sampling was checked by a Shapiro-Wilk test. Whether the normal distribution was respected (VO₂ and HR), the parametric Student’s t-test was applied. Whether the normal distribution was not respected (spatio-temporal gait parameters and Cw), the non-parametric Wilcoxon’s test was applied. Comparison between ET and CT included 16 AD of the ET group and 18 AD of the CT group. If not stated otherwise, data are presented as median of the absolute difference between pre- and post- measurements and interquartile range.

RESULTS

Characteristics of the participants

Fifty-three patients with AD were successfully enrolled in the study. Of the 53 patients with AD enrolled in the study, 26 were assigned in the ET group and 27 in the CT group. The groups were well matched concerning baseline characteristics, including age, sex, physical characteristics, and cognitive status. Participants were community dwelling patients with AD, with mild to moderate dementia, exhibiting impairments in one or more cognitive domains, and clear signs of dementia including impaired executive functions and reduced independence. Ten patients of the ET group dropped out: 8 due to caregivers related problems (such as work commitments, impossibility to drive the relative to the treatment sessions, health problems), 1 due to diabetes complication, and 1 due to a heart attack. Nine people of the CT group dropped out: 6 due to caregivers related problems, 1 due to a knee surgery, 1 due to stroke, and 1 due to heart attacks. Data from 34 patients with AD (ET; n = 16) and (CT; n = 18) were utilized for this study (Fig. 1). In Table 1 are displayed demographic and clinical characteristics, baseline measure of participants who completed the treatment, and drugs (cholinesterase inhibitors, benzodiazepines, neuroleptics) taken by the participants. At baseline, the groups showed no significant difference in any characteristics and measurement, except for neuroleptics intake and Cw at WS3. The 34 patients with AD attended the 72 scheduled sessions of treatment and no adverse events occurred either during exercise or during cognitive trainings.

Table 1

Subjects characteristics and baseline measures

	AD_ET	AD_CT
n	16	18
Female – n (%)	10 (63)	11 (61)
Age (y)	80±7	79±6
Weight (kg)	70.2±12.3	66.7±14.8
Height (m)	1.62±0.10	1.64±0.10
BMI (kg/m²)	26.8±4.9	24.4±4.4
MMSE (0–30)	22±5	21±5
ChEl – n (%)	8 (50)	8 (44)
BDZ – n (%)	2 (13)	2 (11)
Neu – n (%)	0 (0)	2 (11)*
Speed – (cm·s^-1)	92.5±10.2	92.3±5.7
Stride – (cm)	111.3±7.2	104.3±4.6
Step – (cm)	55.7±3.6	52.0±2.3
Single Support – (%)	37.2±1.1	36.1±0.4
Double Support – (%)	26.2±1.3	27.5±0.9
VO₂ WS1 – (ml/min·kg^-1)	13.3±0.6	11.0±0.9
VO₂ WS2 – (ml/min·kg^-1)	12.0±0.6	12.0±0.8
VO₂ WS3 – (ml/min·kg^-1)	14.8±0.7	17.1±1.0*
HR WS1 – (bpm)	98.3±2.3	93.0±2.8
HR WS2 – (bpm)	103.9±3.0	99.2±2.8
HR WS3 – (bpm)	108.2±4.6	106.8±3.2
Cw WS1 – (J/kg·m^-1)	6.2±1.1	6.5±2.6
Cw WS2 – (J/kg·m^-1)	5.1±0.5	4.6±1.0
Cw WS3 – (J/kg·m^-1)	6.2±1.6	5.9±0.9
RER WS1	0.79±0.07	0.75±0.05
RER WS2	0.82±0.03	0.84±0.06
RER WS2	0.89±0.04	0.90±0.06

Note: Plus-minus values are means±SD. The Student’s t-test was used for between-group differences for continuous variables, and a Chi-square test was used for categorical variable. There were no significant between-group differences in any baseline characteristics, except for Neu and VO₂ WS3. AD_ET, patients with Alzheimer’s disease taking part in the exercise training; AD_CT, patients with Alzheimer’s disease taking part in the cognitive treatment; BMI, body mass index; MMSE, Mini-Mental Examination; ChEl, cholinesterase inhibitors; BDZ, benzodiazepines; Neu, neuroplectics; VO₂, oxygen consumption; HR, heart rate; Cw, energy cost of walking; RER, respiratory exchange ratio; WS, walking speed; *p < 0.05.

Effects of exercise training and cognitive treatment in patients with AD

After 6 months of ET, patients with AD increased the gait speed measured by means of GaitRite system; however, the change between groups was not statistically significant (p = 0.076) (Table 2; Fig. 2A). Moreover, neither the ET group nor CT group demonstrated any significant change in other spatio-temporal gait parameters (Table 2; Fig. 2B-E). VO₂ decreased consistently in the ET group, while the CT group did not show any difference. Between group differences were statistically significant at every tested speed (WS1 p = 0.030, WS2 p = 0.006, WS3 p = 0.054) (Table 2, Fig. 3A-C). To control for family-wise Type 1 error rate, we performed the Holm’s sequential Bonferroni procedure; with this procedure only the difference at WS2 was significant. HR decreased at WS1 and WS2 speed in the ET group, while CT group did not exhibit any amelioration. Anyhow, statistically significant changes between groups were noticeable only for WS2 (p = 0.011) (Table 2, Fig. 3D-F). The significance was confirmed with the Holm’s sequential Bonferroni procedure. Cw decreased in the ET group at each tested speed, while CT group did not exhibit any amelioration. Significant change between group were detectable for WS1 and WS2 only (p = 0.041, and p = 0.003 respectively) (Table 2, Fig. 3G-I). With the Holm’s sequential Bonferroni procedure, only WS2 was statistically significant.

Table 2

Spatio-temporal gait parameters, energy cost of walking, heart rate, and oxygen consumption for each tested speed

		AD _ET	AD _CT
				z	t	p
Speed	(cm·s^-1)	10.6 (–6.7 to 26.7)	–0.7 (–16.6 to 9.5)	–1.771		0.076
Stride	(cm)	–1.6 (–14.8 to 5.5)	–3.6 (–11.8 to 1.7)	–0.856		0.391
Step	(cm)	–1.4 (–7.3 to 2.8)	–1.6 (–6.1 to 0.9)	–0.720		0.471
Single supp.	(%)	0.7 (–1.1 to 1.9)	0.4 (–0.6 to 1.4)	0.000		1.000
Double supp.	(%)	–0.7 (–3.3 to 0.8)	0.3 (–3.1 to 2.5)	0.852		0.394
HR WS1*	(bpm)	–8 (–12.6 to –0.7)	0.8 (–8.8 to 9.5)		1.577	0.126
HR WS2*	(bpm)	–6.9 (–16.7 to –2.0)	0.8 (–2.6 to 11.75)		2.717	0.011
HR WS3*	(bpm)	1.0 (–4.7 to 7.2)	–7.3 (–23.5 to –1.9)		1.752	0.107
VO₂ WS1*	(ml/min·kg^-1)	–2.2 (–3.9 to –1.0)	–0.3 (–2.5 to 2.2)		2.280	0.030
VO₂ WS2*	(ml/min·kg^-1)	–1.9 (–5.4 to –0.5)	1 (–1.6 to 2.7)		2.941	0.006
VO₂ WS3*	(ml/min·kg^-1)	–4.4 (–5.8 to –1.5)	–0.6 (–2.4 to 0.7)		2.085	0.054
Cw WS1	(J/kg·m^-1)	–1.2 (–1.7 to –0.6)	–0.4 (–1.3 to 0.6)	2.041		0.041
Cw WS2	(J/kg·m^-1)	–0.9 (–1.3 to –0.5)	–0.2 (–0.6 to 0.5)	2.958		0.003
Cw WS3	(J/kg·m^-1)	–0.8 (–1.4 to –0.4)	–0.2 (–1.3 to 0.7)	1.468		0.142

Note: Data are presented as median of the difference between pre- and post- treatment and interquartile range. The Student’s t-test was used for between-group differences for parametric variables, while Wilcoxon’s test was used for between group differences for non-parametric variables. AD, patients with Alzheimer’s disease; ET, exercise training; CT, cognitive treatment. *parametric variable; z, z-score; t, t-test.

Fig.2

Individual changes after 6 months Exercise treatment (ET) and Cognitive treatment (CT). A) Individual change in gait speed (cm·s^-1). B) Individual change in stride (cm). C) Individual change in step (cm). D) Individual change in single support (%). E) Individual change in double support (%).

Fig.3

Individual changes after 6 months of Exercise treatment (ET) and Cognitive treatment (CT). A) Individual change in oxygen consumption at the first walking speed (VO2 WS1; ml·min^-1·kg^-1). B) Individual change in oxygen consumption at the second walking speed (VO2 WS2; ml·min^-1·kg^-1). C) Individual change in oxygen consumption at the third walking speed (VO2 WS3; ml·min^-1·kg^-1). D) Individual change in heart rate at the first walking speed (HR WS1, bpm). E) Individual change in heart rate at the second walking speed (HR WS2, bpm). F) Individual change in heart rate at the third walking speed (HR WS3, bpm). G) Individual change in energy cost of walking at the first walking speed (Cw WS1, J·kg^-1·m^-1). H) Individual change in energy cost of walking at the second walking speed (Cw WS2, J·kg^-1·m^-1). I) Individual change in energy cost of walking at the third walking speed (Cw WS3, J·kg^-1·m^-1).

DISCUSSION

Although alterations of locomotion have been already investigated in patients with AD, the effects of ET on Cw has so far received little attention. In the present study, we measured the effects of an exercise training with respect to a conventional cognitive treatment on Cw and its biomechanical and metabolic components. Partially in accordance with our hypothesis, after 6 months of exercise training, patients with AD exhibited a clear amelioration of Cw, but the positive effects of exercise training was not associated with improvements in gait biomechanics. Contrarily, the effects of standard cognitive training on Cw and gait biomechanics were negligible.

Evidence about the efficacy of ET in patients with AD

Emerging literature has underlined the positive effects of ET in patients with AD such as improvement in the execution of activity of daily living, reduction in behavioral disorders, and a decrease in the natural progression of cognitive impairment [15 –19]. Less clear are the effects of ET on gait parameters in patients with AD. For instance, Schwenk et al. [20] examined speed, stride, and double support in a group of demented people before and after 3 months of ET, which included resistance, functional and specific walking training. Authors reported great improvements in walking speed, as well as in stride and double support. Our results are partially in accordance with Schwenk et al. [20]. Indeed, patients who underwent ET reported a great improvement in walking speed, which was not accompanied by improvements in stride, single and double support after ET. This result may be due to the different training methods used in our study, which did not include specific walking training. However, together with adaptation in walking speed, patients with AD demonstrated reduction in the Cw. In the past, ET effects on Cw have been investigated only in healthy elderly and patients with Parkinson’s disease, but this important outcome is lacking in patients with AD [10, 21]. Our data are in agreement with the results achieved with other populations showing that patients with AD can obtain great benefit on Cw from an ET program.

Physiological considerations

Results of our study let us speculate that improvements in Cw are mainly induced by positive adaptation of its metabolic components rather than improvement in gait biomechanics, which nevertheless is maintained. Additionally, recent literature provides evidence of several exercise-induced ameliorations in patients with AD [22 –26].

It is important to highlight that the energy demanded by the muscles recruited during ambulation is maintained by an efficient cardio-metabolic response, able to sustain this raised energy request. In patients with AD, cardiovascular dysfunctions go hand in hand with neurological and cognitive disorders. Therefore, disorders at cardiovascular level may result in an “overload” of the system and a loss of efficiency during locomotion. Furthermore, it has been demonstrated that patients with AD has several systemic manifestations that concur with the dysfunction at the central nervous system level, such as mitochondrial abnormalities. When mitochondrial dysfunctions happen at muscular level, it may reduce the ability of the muscles to generate an adequate amount of energy required to sustain movement, inducing a further loss of efficiency during ambulation. Therefore, it is possible that patients with AD need a higher cardio-metabolic effort compared with healthy elderly to maintain a similar pattern of walking which is translated in a higher Cw. Consequently, this higher Cw and abnormalities in gait biomechanics are likely associated with increased fatigue. Increased fatigue leads to loss of balance, increase of fall risk, loss of independence, and worsening of quality of life. Consequently, an intervention focused on improving Cw in this population has practical importance because it may contribute in maintaining independence and the quality of life of individuals with AD.

Certainly, as reported in previous other studies [8 , 27], we cannot neglect that the effects of ET on CW can be ascribed to both central and peripheral components of oxygen uptake. Unfortunately, in the present study, the lack of peripheral measurements cannot support these previous findings. However, we can assume that enhancements in heart elasticity, left ventricular remodeling, and likely peripheral increase in concentration and activity of mitochondria, changes in skeletal muscle phenotype, and a bettered capillarization likely occurred also in the current study after the six months of ET.

Noticeably, the physiological mechanisms that induced an amelioration of CW after our ET protocol were different to those involved in standard walking training [20] in terms of peripheral neuromuscular stimulation. Indeed, a general exercise program like the one we included in this study may generate non-specific adaptations including systemic and metabolic pathways rather than specific adaptation, such as improvement in spatio-temporal gait parameters. Perhaps, a specific training (i.e., walking training) should have been chosen in order to reach specific adaptations in gait biomechanics. However, although the most common recommendation regimen of exercise training for patients with AD is 30 minutes of walking 5 days per week, American College of Sport Medicine’s (ACSM) guidelines emphasize that for most health outcomes additional benefits occur as the amount of physical activity increases through higher intensity, and/or longer duration [28]. Furthermore, ACSM’s guidelines stated that a combination of aerobic and resistance exercise seems to be more effective than either form of training alone in counteracting the detrimental effects of a sedentary lifestyle on the health functioning of cardiovascular system and skeletal muscles [28]. In support of that, Wang et al. [29] and others [30] have shown how resistance training leads to improvements in work efficiency during walking contributing in the maintenance in physical function and consequently the quality of life.

Additionally, we have chosen an ET different from the simple walking training and above all different from our outcome measurement to avoid a crosstalk effect. Indeed, if we had trained patients with AD exclusively on walking, we could have experienced an overlap on the outcomes concerning spatio-temporal gait parameters and energy cost of walking. Nonetheless, even though participants underwent mixed ET including both aerobic and resistance training, have significantly improved energy cost of walking and maintained spatio-temporal gait parameter, which gives even more physiological importance to our findings.

General considerations

In spite of the important dropout rate, which is the major limitation of this study, some general considerations can be drawn. First, ET is feasible to perform safely in patients with AD. Indeed, participants who completed the trial (∼65%) attended all the scheduled sessions and did not experienced any adverse events during ET. Second, patients with AD can achieve positive ET-induced adaptations. Even though the proposed ET was not specific on locomotion, participants experienced a stabilization of spatio-temporal gait parameters and ameliorated metabolic components. However, due to the limited number of participants, other studies with wider sample size are needed in order to translate these results into clinical practice and rehabilitation programs.

Limitations

Clear limitation of the current study was the number of patients excluded at the first biomedical screening, implicating that patients included in this study were potentially a specific sub-group of patients with AD and this does not allow to generalize the results to the entire AD population. Nevertheless, for safety, we decided to include only people without severe comorbidities (i.e., cardiac, orthopedic, respiratory pathologies), which could have been exacerbated during exercise. Nonetheless, two participants dropped due to complications that could have been related to this kind of intervention. However, we cannot establish a link between the exercise intervention and these complications. Second, we did not monitor the amount of physical activity of participants outside the trial, which could be a potential confounder. A further limitation of the current study was the elevated drop-out (∼35%) of the subjects, indicating the limited feasibility of the treatments. It is important to note that the main reason for the patient’s drop-out was due to caregiver’s related problem, including the impossibility to drive the patient to ET or CT. A potential solution for decreasing this elevated drop-out rate could be providing treatment directly at a patient’s home, so they would not need to be driven to either to the gym or to the hospital.

Conclusion

Patients with AD have poor locomotion, accompanied by poor biomechanics and elevated Cw. Interestingly, data from the present study revealed that ET can significantly improve Cw. This amelioration seems to be elicited mainly by cardiovascular and metabolic adaptations rather than modification in gait biomechanics. Nonetheless, the continuous worsening of gait pattern exhibited by the CT group seems to have been reversed in patients with AD who underwent ET. Furthermore, results of our study indicate that the effects of CT on patient’s locomotion were negligible. In conclusion, even though the low adherence of patients with AD to an ET, the effectiveness of this non-pharmacological intervention on the patient’s locomotion was significant.

Footnotes

ACKNOWLEDGMENTS

The authors want to thank all the participants of the present study, as well as all the volunteers who supervised the treatments: G. Viscomi, M. Geccherle, D. Borgo, G. Parisi, D. Tosoni, G.V. La Monica, A. Brugnera, Alzheimer’s Association of Verona and Mons. Mazzali foundation for their committed involvement and adherence to the project. No compensation was received for such contributions.

This work was supported by PRIN 2010KL2Y73_004.

Authors’ disclosures available online ().

References

Reitz

, Mayeux

(2015) Alzheimer disease – epidemiology, diagnostic criteria, risk factors and biomarkers. Biochem Pharmacol 88, 640–651.

Harlein

, Dassen

, Halfens

RJG

, Heinze

(2009) Fall risk factors in older people with dementia or cognitive impairment: A systematic review. J Adv Nurs 65, 922–933.

Beauchet

, Launay

, Annweiler

, Allali

(2015) Hippocampal volume, early cognitive decline and gait variability: Which association? Exp Gerontol 61, 98–104.

Orcioli-Silva

, Simieli

, Barbieri

, Stella

, Gobbi

LTB

(2012) Adaptive walking in alzheimer’s disease. Int J Alzheimers Dis 2012, 674589.

Muir

, Speechley

(2012) Dual-task complexity affects gait in people with mild cognitive impairment: The interplay between gait variability, dual tasking, and risk of falls. Arch Phys Med Rehabil 93, 293–299.

Muir

, Speechley

, Wells

, Borrie

, Gopaul

, Montero-odasso

(2012) Gait assessment in mild cognitive impairment and Alzheimer’s disease: The effect of dual-task challenges across the cognitive spectrum. Gait Posture 35, 96–100.

Chung

, Angelo

, van Emmerik

REA

, Kent

(2016) Energy cost of walking, symptomatic fatigue and perceived exertion in persons with multiple sclerosis. Gait Posture 48, 215–219.

Ortega

, Beck

, Roby

, Turney

, Kram

(2014) Running for exercise mitigates age-related deterioration of walking economy. PLoS One 9, e113471.

Billinger

, Vidoni

, Honea

, Burns

(2011) Cardiorespiratory response to exercise testing inindividuals with Alzheimer’s disease. Arch Phys Med Rehabil 92, 2000–2005.

10.

Pelosin

, Faelli

, Lofrano

, Avanzino

, Marinelli

, Bove

, Ruggeri

, Abruzzese

(2009) Effects of treadmill training on walking economy in Parkinson’s disease: A pilot study. Neurol Sci 30, 499–504.

11.

Munari

, Pedrinolla

, Smania

, Picelli

, Gandolfi

, Saltuari

, Schena

(2016) High-intensity treadmill training improves gait ability, VO₂peak and cost of walking in stroke survivors: Preliminary results of a pilot randomized controlled trial. Eur J Phys Rehabil Med, in press.

12.

Almarwani

, VanSwearingen

, Perera

, Sparto

, Brach

(2016) Challenging the motor control of walking: Gait variability during slower and faster pace walking conditions in younger and older adults. Arch Gerontol Geriatr 66, 54–61.

13.

di Prampero

(1986) The energy cost of human locomotion on land and in water. Int J Sports Med 7, 55–72.

14.

Spector

, Thorgrimsen

, Woods

, Royan

, Davies

, Butterworth

, Orrell

(2003) Efficacy of an evidence-based cognitive stimulation therapy programme for people with dementia: Randomised controlled trial. Br J Psychiatry 183, 248–254.

15.

Baker

, Frank

, Foster-Schubert

, Green

, Wilkinson

, McTiernan

, Plymare

, Fishel

, Watson

, Duncan

, Mehta

, Crafts

(2010) Effects of aerobic exercise on mild cognitive impairment: A controlled trial. Arch Neurol 67, 71–79.

16.

Venturelli

, Lanza

, Muti

, Schena

(2010) Positive effects of physical training in activity of daily living-dependent older adults. Exp Aging Res 36, 190–205.

17.

Venturelli

, Scarsini

, Schena

(2011) Six-month walking program changes cognitive and ADL performance in patients with Alzheimer. Am J Alzheimers Dis Other Demen 26, 381–388.

18.

Forbes

, Thiessen

, Blake

, Forbes

(2013) Exercise programs for people with dementia. Cochrane Database Syst Rev 12, CD006489.

19.

Venturelli

, Ce

, Limonta

, Muti

, Scarsini

, Brasioli

, Schena

, Espostio

(2016) Possible predictors of involuntary weight loss in patients with Alzheimer’s disease. PLoS One 1, e0157384.

20.

Schwenk

, Zieschang

, Englert

, Grewal

, Najafi

, Hauer

(2014) Improvements in gait characteristics after intensiveresistance and functional training in people with dementia: Arandomised controlled trial. BMC Geriatr 14, 73.

21.

Malatesta

, Simar

, Saad H

Ben

, Préfaut

, Caillaud

(2010) Effect of an overground walking training on gait performance inhealthy 65- to 80-year-olds. Exp Gerontol 45, 427–434.

22.

Pedrinolla

, Schena

, Venturelli

(2017) Resilience to Alzheimer’s disease: The role of physical activity. Curr Alzheimer Res 14, 546–553.

23.

Lange-Asschenfeldt

, Kojda

(2008) Alzheimer’s disease, cerebrovascular dysfunction and the benefits of exercise: From vessels to neurons. Exp Gerontol 43, 499–504.

24.

Navarro

, Boveris

(2004) Rat brain and liver mitochondria develop oxidative stress and lose enzymatic activities on aging. Am J Physiol Regul Integr Comp Physiol 287, R1244–1249.

25.

Marques-Aleixo

, Oliveira

, Moreira

, Magalhães

, Ascensão

(2012) Physical exercise as a possible strategy for brain protection: Evidence from mitochondrial-mediated mechanisms. Prog Neurobiol 99, 149–162.

26.

Steiner

, Murphy

, McClellan

, Carmichael

, Davis

(2011) Exercise training increases mitochondrial biogenesis in the brain. J Appl Physiol 111, 1066–1071.

27.

Van Swearingen

, Perera

, Brach

, Cham

, Rosano

, Studenski

(2009) A randomized trial of two forms of therapeutic activity to improve walking: Effect on the energy cost of walking. J Gerontol A Biol Sci Med Sci 64, 1190–1198.

28.

Chodzko-Zajko

, Proctor

, Fiatarone Singh

, Minson

, Nigg

, Salem

, Skinner

(2009) Exercise and physical activity for older adults. Med Sci Sports Exerc 41, 1510–1530.

29.

Wang

, Nyberg

, Hoff

, Zhao

, Leivseth

, TØrhaug

, Husby

, Helgerud

, Richardson

(2017) Impact of maximal strength training on work efficiency and muscle fiber type in the elderly: Implications for physical function and fall prevention. Exp Gerontol 91, 64–71.

30.

Barrett-O’Keefe

, Helgerud

, Wagner

, Richardson

(2012) Maximal strength training and increased work efficiency:Contribution from the trained muscle bed. J Appl Physiol 113, 1846–1851.