The relative aerobic load of stroke patients walking with and without obstacles

The relative aerobic load of stroke patients walking with

and without obstacles.

Name: Marlou Ettema

Academic year: 4

Year of study: 2018/2019

Domain: Faculty of Sport and Nutrition Education: Physical Education

Assignment: Graduation thesis

Supervisors: Ilse Blokland & Fabian Broers Student number: 500727697

Deadline: May 10, 2019 Occasion: First occasion

Contents

THE RELATIVE AEROBIC LOAD OF STROKE PATIENTS WALKING WITH AND

WITHOUT OBSTACLES ... 3 ABSTRACT ... 3 Objective ... 3 INTRODUCTION ... 4 METHOD ... 7 Subjects ... 7 Procedure ... 7 Data analysis ... 8 Statistical analysis ... 9

Validity and reliability ...10

RESULTS ... 10

Normality ...10

Preferred walking speed ...12

Relative aerobic load ...13

DISCUSSION ... 14

CONCLUSION ... 16

REFERENCES ... 17

EPHORUS BEWIJS _{... 21}

The relative aerobic load of stroke patients walking with and without

obstacles

Ettema M.

From the Amsterdam University of Applied Sciences, Bachelor thesis, Netherlands (NL). ABSTRACT

Objective

After a stroke event, stroke patients have a reduced ability of walking and their cardiorespiratory fitness, aerobic load and oxygen uptake (V̇O2 peak) is also lowered. A demand to be able to participate in society again for stroke patients is the ability to walk independently whereas these factors make it more difficult. Many studies have focused on walking on treadmills or on even surfaces whereas in daily life you often encounter obstacles during walking. Therefore this study will focus on the relative aerobic load that is used of stroke patients during walking with and without obstacles. The hypothesis is that the stroke patients will have a higher relative aerobic load during the obstacle walk than during the non-obstacle walk compared to abled bodied.

Methods

Twenty stroke patients and ten able-bodied individuals participated in this study and completed two experiments. One 5-minute walk with obstacles and one 5-minute walk without obstacles, both were executed at the participant’s preferred walking speed (PWS). Prior to the experiments all participants underwent a cardiopulmonary exercise test (CPET). During both experiments and the CPET the participants wore a portable gas analyzer to measure the V̇O2. Furthermore the walking speed was calculated and the relative aerobic load was calculated from both experiments.

Results

A significant effect was found of the relative V̇O2 between walking with and without obstacles (p=0,045) and of the relative V̇O2 between the two groups (p=0.004). However, there was no significant interaction for the relative V̇O2 between the two groups and walking with and without obstacles. In addition, both groups had a slower PWS walking with obstacles compared to walking without obstacles.

Conclusion

During the obstacle walk the mean PWS decreased for both stroke patients as controls, while the relative aerobic load increased only for controls. The relative aerobic load stayed the same for stroke patients.

INTRODUCTION

During 2016, 41.047 people were admitted to the hospital in the Netherlands due to stroke (Buddeke, Van Dis, Visseren, Vaartjes, & Bots, 2017). These patients often encounter a major disruption in their daily living due to the stroke event and need to adapt to their changed bodies and abilities (Rittman, Faircloth, Boylstein, & Gubrium, 2004) . Next to that it is paramount for these stroke patients to follow a rehabilitation program to achieve the best recovery results that will enable them to participate in society again (Horgan, O'regan, Cunningham, & Finn, 2009) . One of the requirements to be able to participate in society again is the ability to execute activities of daily living (ADLs). Walking independently for instance, is a description of an ADL (Rittman et al., 2004).

60% to 70% of stroke patients at discharge of rehabilitation regain the ability to walk independent again (Kelly, Kilbreath, Davis, Zeman, & Raymond, 2003) . Beside the fact that the ability of independent walking is reduced for people whom suffered a stroke, there are several other aspects used for walking that are influenced after a stroke event. One is the cardiorespiratory fitness of stroke patients, which is reduced compared to healthy peers (Ivey, Hafer-Macko, & Macko, 2006) .

Cardiovascular fitness and exercise capacity, expressed as maximal aerobic capacity (VO2 max), is considered to be the best way to measure a person’s cardiorespiratory fitness (McArdle, Katch, & Katch, 2014) . The V̇O2 max represents the ability of the body to supply oxygen to the active muscles during exercise and to remove metabolites. A way to determine the V̇O2 max is through a cardiopulmonary exercise test (CPET). However the cardiovascular fitness and exercise capacity is generally referred to as V̇O2 peak instead of V̇O2 max in patient population. The reason for this is that during a CPET the maximal effort can be influenced by other factors that can limit the oxygen uptake capacity (Billinger, Coughenour, Mackay-Lyons, & Ivey, 2012) .

Focusing on the peak aerobic capacity of stroke patients, Ivey et al., (2006) reported a mean V̇O2 peak of 13.6 ± 4 (mL/kg/min), whereas MacKay-Lyons & Makrides, (2002) found a mean V̇O2 peak of 14.4 ± 5.1 (mL/kg/min) in patients post stroke (>1month). These V̇O2 peak results of stroke patients are 60% ± 16% compared to inactive but healthy age-matched individuals (MacKay-Lyons & Makrides, 2002) . Further a systematic review that combined several studies, reported V̇O2 peak results ranging from 8 ml/kg/min to 22 ml/kg/min (Smith, Saunders, & Mead, 2012) .

Apart from a reduced V̇O2 peak amongst stroke patients, their aerobic load during exercise is also affected post stroke. The aerobic load is expressed as the amount of oxygen uptake (V̇O2 mL/kg/min) that is needed to fulfil a given task or exercise. Kramer, Johnson, Bernhardt, & Cumming, (2016) show that stroke patients have a higher aerobic load during steady-state walking at matched speed compared to healthy controls. But when walking at a preferred walking speed (PWS), there is no significant difference in aerobic load between stroke patients and able-bodies (Platts, Rafferty, & Paul, 2006) .

Some studies have been done on the aerobic load of stroke patients, where the majority of these studies have mainly focused on walking on treadmills and straight surfaces. Fewer studies have been done on the aerobic load needed for walking during real life situations. For instance, while walking outside you often encounter obstacles such as curbstones, poles and puddles that you want to avoid. The movements that are made during real life situations are quite different from walking on a treadmill and therefore they could demand a different aerobic load. Because walking independently during daily life is a requirement to be able to participate in society again for stroke patients, it is of interest to see what aerobic load is needed during walking while encountering obstacles (Rittman et al., 2004)

One of the studies that have been done on walking with obstacles show that the energetic cost of walking (ECW mL/kg/m) increases during walking with obstacles in both healthy people and stroke patients when compared to walking without obstacles (Slawinski, Pradon, Bensmail, Roche, & Zory, 2014) . This energetic cost of walking is expressed in mL/kg/m, which means that the oxygen uptake is compared to the distance that is covered instead of time (ml/kg/min) so both stroke patients and healthy controls use more energy on the same distance during the obstacle course compared to the non-obstacle course (Slawinski et al., 2014). The ECW is a measure to see how efficient someone walks a certain distance while the aerobic load is a measure to see how much energy someone uses while executing an activity such as walking for instance. However, the results of Slawinski et al., (2014) did also show that the aerobic load (mL/kg/min) was higher for the healthy group during the obstacle walk than during the non-obstacle walk. This did not apply for the stroke patients. So when the aerobic load is expressed in time (mL/kg/min) the aerobic load increases for healthy controls during the obstacle walk and stays the same for stroke patients. These findings were also similar to the research of Kafri, Myslinski, Gade, & Deutsch, (2014) . Next to that, what is noticeable is that the stroke patients

walked slower than the healthy group during both conditions. If the two groups where to walk the same speed during a walk without obstacles, the aerobic load would be higher for the stroke patients compared to healthy people (Kramer et al., 2016; Platts et al., 2006). The question remains if this also would apply for walking with obstacles, because the obstacle course adds several other objects which can make it more difficult and less economic to walk for stroke patients and that in turn can influence the aerobic load (mL/kg/min) (Slawinski et al., 2014).

Also, during the experiment of Slawinski et al., (2014), the subjects were told to walk as far as possible, therefore they might have chosen a speed higher than their PWS, which in turn might have influenced the aerobic load. Due to the instructions given during this research, the stroke patients could have been walking so fast that they would have reached their peak capacity (V̇O2 peak). One hypothesis is that “subjects with hemiplegia are unable to increase V̇O2 during the

WS_O because they are already close to their maximal oxygen uptake (V̇O2max) during the

WS_wO” where WS_O is walking with obstacles and WS_wO is walking without obstacles (Slawinski et al., 2014, 5). To test this hypothesis whether the stroke patients where reaching their V̇O2 peak, the aerobic load during walking with and without obstacles should be compared to the V̇O2 peak which is measured during a CPET. Because the V̇O2 peak is measured in mL/kg/min the aerobic load also needs to be measured in time (mL/kg/min) instead of distance (mL/kg/m). Then the results during the non-obstacle and obstacle walk would have been the relative aerobic load expressed in % mL/kg/min compared to the peak aerobic load. This however was not done during both studies on the aerobic load during walking with obstacles for stroke patients (Kafri et al., 2014; Slawinski et al., 2014).

Further investigation of the relative aerobic load in stroke patients during obstacle walking can contribute to health promotion and the results could be taken in consideration when designing rehabilitation tracks for improved participation in daily life activities. The purpose of this study is to compare the relative aerobic load of stroke patients walking with and without obstacles at PWS and to compare this with able-bodied individuals. The hypothesis is that the stroke patients will have a higher relative aerobic load during the obstacle walk compared to the non-obstacle walk. This is because the energy cost of walking is higher for the stroke patients during the obstacle walk and therefore an assumption can be made that the relative aerobic load also would be higher during walking with obstacles (Kafri et al., 2014; Slawinski et al., 2014).

METHOD

This study was a cross-sectional study that will assess the relative aerobic load during walking with and without obstacles for stroke patients.

Subjects

The subjects selected for this experiment were subacute stroke patients (<6 months after stroke), (n=20) whom were inpatients of rehabilitation centre Heliomare in Wijk aan Zee (Bernhardt, Hayward, Kwakkel, Ward, Wolf, Borschmann, & Cramer, 2017). 15 of these stroke patients were male and 5 were female. The control group (n=10) existed of healthy age and gender matched abled-bodied individuals, where 8 were male and 2 were female. The participants were matched on age, gender and BMI, these descriptions are displayed in table 1. Prior to the participation every test subject underwent a cardiopulmonary exercise test (CPET) that functioned as a safety screening and which determined the V̇O2 peak. The participants that were included for this research, had first met the inclusion criteria and not the exclusion criteria for the study Fitness and Function After Stroke (FaFaS). This was to make sure that all individuals who participated were healthy enough to participate in this study without any unnecessary risk for damage. Further exclusion criteria for this study were the following:

-Participant is not able to execute the courses walking with and without obstacles for 4 consecutive minutes.

-Participants CPET results do not fulfill the criteria of the V̇O2 peak list.

This study was a small part of the FaFaS study that was given approval by the medical ethical committee of the VUMC, trial nr. NL6917. Prior to participation all participants signed an informed consent.

Procedure

Prior to the experiment each participant underwent a CPET where the V̇O2 peak was determined. The following information was obtained from the CPET: V̇O2 peak, maximal respiratory exchange ratio (RER), respiratory compensation point (RCP) and the maximal heart rate (HRmax). Also the age, gender and BMI were collected.

All participants then participated in the two testing conditions in randomized order: walking with obstacles (with_O) and without obstacles (without_O). During the experiment, all participants wore a mobile gas analysis system (Cosmed, K4b2, Italy) that measured the gas exchange breath-by-breath. Stroke patients were instructed to walk in both conditions at a PWS.

Control subjects however were instructed to walk at a PWS and at ½ of their PWS to get closer to the stroke patients PWS, as this is markedly lower than able-bodied (Platts et al., 2006). In addition, prior to the experiment the oxygen consumption was measured for a resting period of 10 minutes. Between each activity the participants got a seated rest period of minimal 5 minutes to make sure that the metabolic values returned to a resting baseline. This was done by matching the metabolic values to the values obtained in the rest measurement.

Subjects then walked back and forth in a 30 m long corridor during the with_O experiment. The obstacles used for this experiment where the following:

- A jumping rope (R) hanging horizontal approximately 15cm from the floor and laid transversely on the walking direction. Here participant stepped over the rope.

- A foam carpet (C), which was 30cm wide and laid transversely on the walking direction. Here the participants stepped over the carpet.

- A 4-meter long and one meter wide gym mat (M), which was approximately 5 cm high. The gym mat laid parallel to the walking direction.

- 6 cones (CO), which were placed centered in the corridor. Here the participants had to slalom between.

The obstacles were placed on every 2-meter of the corridor as following: alternating 3 ropes and 3 cloths, 1 gym mat and 6 cones. The setting of walking with obstacles is shown in Figure 1. Both conditions were sustained for 5 minutes as literature suggest that a reliable measurement for steady state walking occurs after 2-3 minutes (Danielsson, Willen, & Sunnerhagen, 2007) .

Figure 1. Arrangement of the with_O track (R: rope, C: carpet, M: gym mat, CO: cones)

Data analysis

From the results of the CPET test, it was determined whether the participant reached their V̇O2 peak. For the V̇O2 peak results to be considered valid the participant had to measure up to 2 of the 3 following conditions:

1. RER>1.10 (Guazzi, Adams, Conraads, Halle, Mezzani, & Vanhees, 2012)

R C R C R C M M M CO CO CO CO CO

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 meters

2. RCP is obtained (Wasserman & Whipp, 1975)

3. HR>85% of the predicted maximal value (van de Port, Kwakkel, & Wittink, 2015) When the V̇O2 peak results were considered maximal, the data from the participant was used for further calculations. The obtained V̇O2 peak result of the participants allowed the researcher to determine the relative aerobic load during the experimental part of walking with (with_O) and without (without_O) obstacles for each individual. These calculations were used for both experiments:

𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑣𝑣𝑅𝑅 𝑅𝑅𝑅𝑅𝑎𝑎𝑎𝑎𝑎𝑎𝑅𝑅𝑎𝑎 𝑅𝑅𝑎𝑎𝑅𝑅𝑙𝑙 𝑤𝑤𝑅𝑅𝑅𝑅ℎ_𝑂𝑂 =𝑚𝑚𝑅𝑅𝑅𝑅𝑚𝑚 𝑅𝑅𝑅𝑅𝑎𝑎𝑎𝑎𝑎𝑎𝑅𝑅𝑎𝑎 𝑅𝑅𝑎𝑎𝑅𝑅𝑙𝑙 𝑤𝑤𝑅𝑅𝑅𝑅ℎ_𝑂𝑂_{VO2 𝑝𝑝𝑅𝑅𝑅𝑅𝑝𝑝} 𝑋𝑋 100%

𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑣𝑣𝑅𝑅 𝑅𝑅𝑅𝑅𝑎𝑎𝑎𝑎𝑎𝑎𝑅𝑅𝑎𝑎 𝑅𝑅𝑎𝑎𝑅𝑅𝑙𝑙 𝑤𝑤𝑅𝑅𝑅𝑅ℎ𝑎𝑎𝑜𝑜𝑅𝑅_𝑂𝑂 =𝑚𝑚𝑅𝑅𝑅𝑅𝑚𝑚 𝑅𝑅𝑅𝑅𝑎𝑎𝑎𝑎𝑎𝑎𝑅𝑅𝑎𝑎 𝑅𝑅𝑎𝑎𝑅𝑅𝑙𝑙 𝑤𝑤𝑅𝑅𝑅𝑅ℎ𝑎𝑎𝑜𝑜𝑅𝑅_𝑂𝑂_{VO2 𝑝𝑝𝑅𝑅𝑅𝑅𝑝𝑝} 𝑋𝑋 100%

The aerobic load used for both conditions was the mean aerobic load taken from the last 2 minutes of the experiment when the participants had reached their steady state. A visual inspection via MATLAB_R2018a was done to insure a steady state V̇O2.

Statistical analysis

All data collected was analyzed in SPSS version 24 (Chicago, Illinois). The Shapiro-Wilk test for normality was used to control whether the data was normally distributed. For the statistical analysis of the hypothesis the two way ANOVA with repeated measures was chosen to test the interaction between stroke/controls and With_O/Without_O. Also the relative V̇O2 between walking with and without obstacles and the relative V̇O2 between the two groups was tested for any significant effect. The reason for this choice was: the relative V̇O2 was the only outcome, and it was continuous. Further there were two predictor variables: stroke/controls and With_O/Without_O. At last the two predictors are categorical and they have both entities in each category (Field, 2013). Here a p value <0.05 was considered statistically significant and would indicate an interaction effect. Because the ANOVA was run with 2 factors, a bonferroni post-hoc analysis was not needed. If there was found an interaction effect, an independent t-test would have been run to see where the interaction laid. The same ANOVA test was run to see if there was any significant interaction for the PWS between stroke/controls and With_O/Without_O. Also the PWS between walking with and without obstacles and the PWS between the two groups was tested for any significant effect. By any interaction found an independent t-test was used to see where the interaction laid. The descriptive data of the

participants was run through descriptive statistics to see if the two groups were matched able-bodied individuals.

Validity and reliability

The main focus of this study was to measure the V̇O2 uptake during the With_O and Without_O. This was done by using the mobile gas analysis system (Cosmed, K4b2, Italy) which was calibrated prior to each test round. The mobile gas analyzer is reliable system with good precision (Akkermans, Sillen, Wouters, & Spruit, 2012) . Due to the difference in V̇O2 peak between stroke patients and controls, the choice was to express the oxygen uptake in relative aerobic load. This way it is expressed equally for each participant. Further, because the participants were instructed to walk at their own preferred walking speed, it was expected that the variance for walking speed could be greater than if the participants were instructed to walk the same speed.

RESULTS Normality

The Shapiro-Wilk test showed that the data was normally distributed with exception of stroke patients walking without obstacles. Because only one of the four factors was not normally distributed the decision was made to interpret the total dataset as normally distributed.

Table 1. Descriptive statistics, here the mean (M) values are presented with standard deviation

(sd). With_O describes the results for walking with obstacles and Without_O describes the results for walking without obstacles.

Mean values descriptive statistics

Stroke (n=20) Control (n=10)

BMI (sd) 27.0(4.2) 24.8(3.2)

Days between: stroke-CPET (sd) 33.3(22.9) -

AGE (sd) 55.4(14.0) 58.2(5.5)

Gender (male/female) 15/5 8/2

VO2 peak (sd) (mL/kg/min) 23.5(7.6) 36.0(7.6)

With_O Without_O With_O Without_O VO2 absolute (sd)(mL/kg/min) 12.2 (2.4) 12.0 (1.7) 14.7 (2.4) 13.1 (2.1) VO2 relative (sd)%(mL/kg/min) 55.3 (13.4) 55.2 (14.7) 42.5 (11.1) 37.6 (8.1) PWS (sd) (m/s) 0.9 (0.3) 1.0 (0.3) 1.2 (0.1) 1.3 (0.2) ECW (sd) (mL/m/kg) 0.28 (0.18) 0.23 (0.14) 0.20 (0.02) 0.16 (0.01)

Preferred walking speed

The ANOVA found a significant difference for the preferred walking speed (PWS) between walking with obstacles and walking without obstacles when the two groups (stroke and controls) were added to one group F(1,28)=32.225, p<0.001. There was also a significant effect for the PWS between the stroke patients and controls when both conditions (With_O and Without_O) were added to one group F(1,28)=9.684, p=0.006.The PWS decreases when walking with obstacles compared to walking without obstacles for both groups, displayed in figure 2. Both groups walked at their own PWS where the stroke patients walked Without_O at a speed of 1.03 ± 0.33 m/s and With_O at 0.89 ± 0.34 m/s. The controls walked Without_O at a speed of 1.34 ± 0.18 m/s and With_O at 1.22 ± 0.12 m/s. However there was found no significant interaction effect F(1,28)=0.197, p=0.661.

Figure 2. significant effect of the PWS between with obstacles and without obstacles F(1,28)=32.225, p<0.001.

Relative aerobic load

The ANOVA showed that there was a significant effect of the relative V̇O2 between walking with and without obstacles F(1,28)=4.417, p=0,045 and between the relative V̇O2 between the two groups F(1,28)=9.866, p=0.004. However, there was no significant interaction for the relative V̇O2 between the two groups and walking with and without obstacles F(1,28)=3.869, p=0.059. These findings are represented in figure 3.

Figure 3. There was no main interaction of the relative VO2 uptake between

DISCUSSION

The purpose of this study was to compare the relative aerobic load of stroke patients walking with and without obstacles at PWS, and to compare this to abled-bodied. The addition of obstacles resulted in a higher relative aerobic load for the entire test group. Also the stroke patients had a higher relative aerobic load compared to the able-bodied during both experiments. Further this study showed that there was a significant effect for the PWS between walking with or without obstacles.

What is noticeable is that the PWS decreased for both groups when adding obstacles. However, relatively, the stroke patients reduced their speed more (to 86% of speed without obstacles) than the controls (91%). This may reflect a difference in strategy: the stroke patients reduce their speed considerably to avoid increase in aerobic load while the able-bodied can increase this load because it is low to start with. The reduction of walking speed during the obstacle experiment were similar to the results of another study done with walking with obstacles (Slawinski et al., 2014). On the other side, a factor that should be taken into consideration is the variance in the stroke group. The variance of the able-bodied was smaller than the stroke group and an explanation for this is that the stroke group consists of two sub groups. To be able to reduce the variance amongst stroke patients the two sub groups could be categorized into stroke patients with a good kinetic movement (those who walk faster) and stroke patients with a less good kinetic movement (those who walk slower).

The addition of obstacles tends to result in a higher relative aerobic load amongst the able-bodied, but for the stroke patients the relative aerobic load tends to stay the same. However this study was not able to prove any significant difference in the relative aerobic load for walking with and without obstacles between stroke patients and abled-bodied, although the results were borderline p < 0.05 (p=0.059). On the other hand, the study of Slawinski et al., (2014) indicated a significant difference in the absolute aerobic load. This was also confirmed by Kafri et al., (2014). The lack of any significance in this study could be explained by a lack of power. If the number of participants were increased in each group, the power would increase as result that the interaction could become significant.

Further the results indicate that stroke patients work at a higher level of their V̇O2 peak

compared to able-bodied during both experiments. Therefore it would be harder for the stroke patients to increase their level of relative aerobic load compared to the controls. This could

explain why the able-bodied increased their relative aerobic load during the obstacle walk and the stroke patients’ relative aerobic load stayed the same. Furthermore the stroke patients were operating around 55,19% and 55,35% of their relative aerobic load for both walking courses with a few outliers, but they never exceeded the 70% of their aerobic load. These results indicate that the hypothesis made by Slawinski et al., (2014) about the stroke patients operating near their maximal aerobic load, could not be confirmed during this study. An interesting note are the V̇O2 peak results for stroke patients found in this study. These are on average higher than reported in previous studies (Smith et al., 2012) .

As the results indicated, different strategies were adapted when walking with obstacles. One of these adaptations was the lowered PWS during the experiment with obstacles. Also the ECW was higher for stroke patients compared to able-bodied and for both groups it increased when walking with obstacles (table 1). This suggests that the task economy during walking with obstacles also is adapted compared to walking without obstacles. However it is of interest to see that both groups had a lowered PWS and a higher ECW during walking with obstacles, but the results hint at a different reaction of the relative aerobic load during the walk with obstacles. Namely, that the relative aerobic load increased for able-bodied and stayed the same for stroke patients during walking with obstacles. It was however not confirmed that there was any interaction effect.

At last the age difference of the participants used in the study of Slawinski et al., (2014) could have influenced their results. The stroke patients had an average age of 52.6 ± 9.4 years and the controls had an average age of 25.6 ± 6.1 years. This is a big gap between the two groups so it can be argued that due to age difference the results of this study were influenced (Slawinski et al., 2014). Waters, Hislop, Perry, Thomas, & Campbell, (1983) show that during a PWS seniors (60-80 years old) have a higher ECW that young adults (20-59 years old). The combination of the fact that stroke patients have a reduced aerobic load and that older people have a higher ECW could have affected the results of Slawinski et al., (2014) where an interaction was found for the aerobic load between walking with and without obstacles between stroke patients and controls. On the other hand, the study of Kafri et al., (2014) had matched the age of the controls to the age of the stroke patients and here the result of the aerobic load between walking with and without obstacles between the two groups also were significant. So therefore it is not yet clear if the age difference between stroke patients and controls could influence the results.

CONCLUSION

The stroke patients have a higher relative aerobic load during walking with and without obstacles compared to able-bodied. Also the stroke patients relatively reduce their PWS more that the controls during the obstacle walk and they have a higher ECW than the controls. At last there is a trend indicating that the relative aerobic load reacts differently between stroke patients and controls when walking with obstacles. However these assumptions could not be confirmed during this study.

These results indicate that stroke patients react different to obstacles and use more energy than able-bodied. Rehabilitation tracks should focus more on the cardiorespiratory fitness of stroke patients and on the task economy during obstacle walks, which combined could relatively lower the aerobic load and energy usage during walking with obstacles.

Finally, future studies should focus more on the difference between stroke patients and energy usage, as there is a large variance for this group. This could lead to a better understanding of the energy usage for walking with obstacles of stroke patients and could in turn lead to beter recommendations for health promotion.

References

Akkermans, M. A., Sillen, M. J., Wouters, E. F., & Spruit, M. A. (2012). Validation of the oxycon mobile metabolic system in healthy subjects. Journal of Sports Science & Medicine, 11(1), 182-183.

Bernhardt, J., Hayward, K. S., Kwakkel, G., Ward, N. S., Wolf, S. L., Borschmann, K., & Cramer, S. C. (2017). Agreed definitions and a shared vision for new standards in stroke recovery research: The stroke recovery and rehabilitation roundtable taskforce.

International Journal of Stroke, 12(5), 444-450. doi:10.1177/1747493017711816 Billinger, S. A., Coughenour, E., Mackay-Lyons, M. J., & Ivey, F. M. (2012). Reduced

cardiorespiratory fitness after stroke: Biological consequences and exercise-induced adaptations. Stroke Research and Treatment, 2012, 959120. doi:10.1155/2012/959120 [doi]

Buddeke, J., Van Dis, I., Visseren, F.L.J., Vaartjes, I., & Bots M.L. (2017). Hart- en vaatziekten in nederland 2017, cijfers over leefstijl, risicofactoren, ziekte en sterfte. Retrieved from http://library.wur.nl/WebQuery/groenekennis/59526

Danielsson, A., Willen, C., & Sunnerhagen, K. S. (2007). Measurement of energy cost by the physiological cost index in walking after stroke. Archives of Physical Medicine and Rehabilitation, 88(10), 1298-1303. doi:S0003-9993(07)00440-6 [pii]

Field, A. P. (2013). Discovering statistics using SPSS: (and sex and drugs and rock 'n' roll)(pp.507-542). Los Angeles [i.e. Thousand Oaks, Calif.: SAGE Publications. Guazzi, M., Adams, V., Conraads, V., Halle, M., Mezzani, A., & Vanhees, L. (2012).

recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. Circulation, 126(18), 2261-2274.

doi:10.1161/CIR.0b013e31826fb946 [doi]

Horgan, N. F., O'regan, M., Cunningham, C. J., & Finn, A. M. (2009). Recovery after stroke: A 1-year profile. Disability & Rehabilitation, 31(10), 831-839.

doi:10.1080/09638280802355072

Ivey, F. M., Hafer-Macko, C. E., & Macko, R. F. (2006). Exercise rehabilitation after stroke. NeuroRX, 3(4), 439-450. doi:10.1016/j.nurx.2006.07.011

Kafri, M., Myslinski, M. J., Gade, V. K., & Deutsch, J. E. (2014). High metabolic cost and low energy expenditure for typical motor activities among individuals in the chronic phase after stroke. Journal of Neurologic Physical Therapy : JNPT, 38(4), 226-232. doi:10.1097/NPT.0000000000000053 [doi]

Kelly, J. O., Kilbreath, S. L., Davis, G. M., Zeman, B., & Raymond, J. (2003).

Cardiorespiratory fitness and walking ability in subacute stroke patients. Archives of Physical Medicine and Rehabilitation, 84(12), 1780-1785. doi:10.1016/S0003-9993(03)00376-9

Kramer, S., Johnson, L., Bernhardt, J., & Cumming, T. (2016). Energy expenditure and cost during walking after stroke: A systematic review. Archives of Physical Medicine and Rehabilitation, 97(4), 632.e1. doi:S0003-9993(15)01474-4 [pii]

MacKay-Lyons, M. J., & Makrides, L. (2002). Exercise capacity early after stroke. Archives of Physical Medicine and Rehabilitation, 83(12), 1697-1702.

Rittman, M., Faircloth, C., Boylstein, C., & Gubrium, J. F. (2004). The experience of time in the transition from hospital to home following stroke. Journal of Rehabilitation Research and Development, 41(3), XI. Retrieved from

https://search.proquest.com/docview/215288168

McArdle, W. D., Katch, F. I., & Katch, V. L. (2014). Exercise physiology : Nutrition, energy, and human performance (Eighth edition, international edition. ed.). Philadelphia:

Wolters Kluwer Health. Retrieved from https://vu.on.worldcat.org/oclc/876397852

Platts, M., Rafferty, D., & Paul, L. (2006). Metabolic cost of overground gait in

younger stroke patients and healthy controls doi:10.1249/01.mss.0000222829.34111.9c Slawinski, J., Pradon, D., Bensmail, D., Roche, N., & Zory, R. (2014). Energy cost of

obstacle crossing in stroke patients. American Journal of Physical Medicine & Rehabilitation, 93(12), 1044-1050. doi:10.1097/PHM.0000000000000122 [doi]

Smith, A. C., Saunders, D. H., & Mead, G. (2012). Cardiorespiratory fitness after stroke: A systematic review. London, England: SAGE Publications.

doi:10.1111/j.1747-4949.2012.00791.x

van de Port, I. G. L, Kwakkel, G., & Wittink, H. (2015). Systematic review of cardiopulmonary exercise testing post stroke: Are we adhering to practice recommendations? Journal of Rehabilitation Medicine, 47(10), 881-900. doi:10.2340/16501977-2031

Wasserman, K., & Whipp, B. J. (1975). Excercise physiology in health and disease. The American Review of Respiratory Disease, 112(2), 219-249.

Waters, R. L., Hislop, H. J., Perry, J., Thomas, L., & Campbell, J. (1983). Comparative cost of walking in young and old adults. Journal of Orthopaedic Research, 1(1), 73-76. doi:10.1002/jor.1100010110

Ephorus Bewijs

Beste Marlou Ettema,

Het document is ingeleverd bij Turnitin | Ephorus en je docent Ramon Stuart (r.stuart@hva.nl) is hiervan op de hoogte gesteld.

Het unieke nummer dat aan het document is toegekend is: e7bc1828-156a-439d-96f8-d503b099802d.

We raden je aan deze pagina uit te printen of op te slaan. Inlevercode: 40150AOSPA

Datum: vrijdag 17 mei 2019 18:21:11 uur CEST Jouw gegevens:

Marlou Ettema 500727697

marlou.ettema@hva.nl

Ettema 500727697 Fabian Broers Je docent:

Ramon Stuart r.stuart@hva.nl

SPSS outputs

General Linear Model

Relative VO2

Notes

Output Created 28-APR-2019 13:00:02

Comments

Input Data /Users/marlouettema/Docu

ments/MATLAB/FaFaS/SP SS uitwerkingen_1.sav

Active Dataset DataSet2

Filter <none>

Weight <none>

Split File <none>

N of Rows in Working Data

File 30

Missing Value Handling Definition of Missing User-defined missing values are treated as missing.

Cases Used Statistics are based on all cases with valid data for all variables in the model.

Syntax GLM VO2_Relative_obs VO2_Relative BY Group /WSFACTOR=With_without _obs 2 Polynomial /METHOD=SSTYPE(3) /PLOT=PROFILE(With_wit hout_obs*Group) /CRITERIA=ALPHA(.05) /WSDESIGN=With_without _obs /DESIGN=Group.

Resources Processor Time 00:00:00,33

Within-Subjects Factors

With_without_obs Dependent Variable

1 VO2_Relative_ obs 2 VO2_Relative

Between-Subjects Factors

Multivariate Tests

Effect Value F Hypothesis df Error df S

With_without_obs Pillai's Trace ,136 4,417b _{1,000 28,000}

Wilks' Lambda ,864 4,417b _{1,000 28,000}

Hotelling's Trace ,158 4,417b _{1,000 28,000}

Roy's Largest Root ,158 4,417b _{1,000 28,000}

With_without_obs *

Group Pillai's Trace ,121 3,869

b _{1,000 28,000}

Wilks' Lambda ,879 3,869b _{1,000 28,000}

Hotelling's Trace ,138 3,869b _{1,000 28,000}

Roy's Largest Root ,138 3,869b _{1,000 28,000} a. Design: Intercept + Group

Within Subjects Design: With_without_obs b. Exact statistic

Mauchly's Test of Sphericity

Measure: MEASURE_1

Within Subjects

Effect Mauchly's W Approx. Chi-Square df Sig.

Epsilonb

Greenhouse-Geisser

Mauchly's Test of Sphericity

Measure: MEASURE_1

Within Subjects Effect Lower-bound Epsilon

With_without_obs

Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed depe

variables is proportional to an identity matrix.a

a. Design: Intercept + Group

Within Subjects Design: With_without_obs

b. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected t are displayed in the Tests of Within-Subjects Effects table.

Tests of Within-Subjects Effects

Source Type III Sum of Squares df Square Mean F S

With_without_obs Sphericity Assumed 84,077 1 84,077 4,417

Greenhouse-Geisser 84,077 1,000 84,077 4,417

Huynh-Feldt 84,077 1,000 84,077 4,417

Lower-bound 84,077 1,000 84,077 4,417

With_without_obs *

Group Sphericity Assumed Greenhouse- 73,642 1 73,642 3,869

Geisser 73,642 1,000 73,642 3,869

Huynh-Feldt 73,642 1,000 73,642 3,869

Lower-bound 73,642 1,000 73,642 3,869

Error(With_without_ob

s) Sphericity Assumed Greenhouse- 532,982 28 19,035

Geisser 532,982 28,000 19,035

Huynh-Feldt 532,982 28,000 19,035

Lower-bound 532,982 28,000 19,035

Tests of Within-Subjects Contrasts

Source With_without_obs Type III Sum of Squares df Square Mean F S

With_without_obs *

Group Linear 73,642 1 73,642 3,869

Error(With_without_obs

) Linear 532,982 28 19,035

Tests of Between-Subjects Effects

Transformed Variable: Average

Source Type III Sum of Squares df Mean Square F Sig.

Intercept 121218,911 1 121218,911 388,927 ,000

Group 3074,890 1 3074,890 9,866 ,004

Error 8726,901 28 311,675

General Linear Model

PWS

Notes

Output Created 30-APR-2019 09:18:47

Comments

Input Data /Users/marlouettema/Docu

ments/MATLAB/FaFaS/SP SS uitwerkingen_1.sav

Active Dataset DataSet2

Filter <none>

Weight <none>

Split File <none>

N of Rows in Working Data

File 30

Missing Value Handling Definition of Missing User-defined missing values are treated as missing.

Cases Used Statistics are based on all cases with valid data for all variables in the model.

Syntax GLM V_without_obs V_with_obs BY Group /WSFACTOR=speed 2 Polynomial /METHOD=SSTYPE(3) /POSTHOC=Group(TUKEY BTUKEY) /PLOT=PROFILE(speed*Gr oup) /PRINT=DESCRIPTIVE HOMOGENEITY /CRITERIA=ALPHA(.05) /WSDESIGN=speed /DESIGN=Group.

Resources Processor Time 00:00:00,30

Warnings

Post hoc tests are not performed for Group because there are fewer than three groups.

Within-Subjects

Factors

Measure: MEASURE_1

speed Dependent Variable

1 V_without_obs 2 V_with_obs

Between-Subjects Factors

Descriptive Statistics

Group Mean Std. Deviation N

without_obs 2 1,3380 ,17706 10 1 1,0275 ,33101 20 Total 1,1310 ,32199 30 with_obs 2 1,2170 ,11795 10 1 ,8860 ,33508 20 Total ,9963 ,32104 30

Box's Test of

Equality of

Covariance

Matrices

Sig. ,009 Tests the null hypothesis that the observed

covariance matrices of the dependent variables are equal across

groups.a

a. Design: Intercept + Group

Within Subjects Design: speed

Multivariate Tests

Effect Value F Hypothesis df Error df S

speed Pillai's Trace ,535 32,225b _{1,000 28,000}

Wilks' Lambda ,465 32,225b _{1,000 28,000}

Hotelling's Trace 1,151 32,225b _{1,000 28,000}

Roy's Largest Root 1,151 32,225b _{1,000 28,000}

speed * Group Pillai's Trace ,007 ,197b _{1,000 28,000}

Wilks' Lambda ,993 ,197b _{1,000 28,000}

Hotelling's Trace ,007 ,197b _{1,000 28,000}

Roy's Largest Root ,007 ,197b _{1,000 28,000}

a. Design: Intercept + Group Within Subjects Design: speed b. Exact statistic

Mauchly's Test of Sphericity

Measure: MEASURE_1

Within Subjects

Effect Mauchly's W Approx. Chi-Square df Sig.

Epsilonb

Greenhouse-Geisser

Huynh-speed 1,000 ,000 0 . 1,000

Mauchly's Test of Sphericity

Measure: MEASURE_1

Within Subjects Effect Lower-bound Epsilon

Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed depe

variables is proportional to an identity matrix.a

a. Design: Intercept + Group Within Subjects Design: speed

b. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected t are displayed in the Tests of Within-Subjects Effects table.

Tests of Within-Subjects Effects

Source Type III Sum of Squares df Mean Square F S

speed Sphericity Assumed ,230 1 ,230 32,225

Greenhouse-Geisser ,230 1,000 ,230 32,225

Huynh-Feldt ,230 1,000 ,230 32,225

Lower-bound ,230 1,000 ,230 32,225

speed * Group Sphericity Assumed ,001 1 ,001 ,197

Greenhouse-Geisser ,001 1,000 ,001 ,197

Huynh-Feldt ,001 1,000 ,001 ,197

Lower-bound ,001 1,000 ,001 ,197

Error(speed) Sphericity Assumed ,200 28 ,007

Greenhouse-Geisser ,200 28,000 ,007

Huynh-Feldt ,200 28,000 ,007

Lower-bound ,200 28,000 ,007

Tests of Within-Subjects Contrasts

Source speed Type III Sum of Squares df Mean Square F Sig.

speed Linear ,230 1 ,230 32,225 ,000

speed * Group Linear ,001 1 ,001 ,197 ,661

Error(speed) Linear ,200 28 ,007

Levene's Test of Equality of Error Variances

F df1 df2 Sig.

without_obs 2,523 1 28 ,123

Tests the null hypothesis that the error variance of the

dependent variable is equal across groups.a

a. Design: Intercept + Group Within Subjects Design: speed

Tests of Between-Subjects Effects

Transformed Variable: Average

Source Type III Sum of Squares df Mean Square F Sig.

Intercept 66,558 1 66,558 421,364 ,000

Group 1,372 1 1,372 8,684 ,006

Error 4,423 28 ,158

Explore

Normality check Relative VO2

Notes

Output Created 29-APR-2019 09:32:30

Comments

Input Data /Users/marlouettema/Docu

ments/MATLAB/FaFaS/SP SS uitwerkingen_1.sav

Active Dataset DataSet2

Filter <none>

Weight <none>

Split File Group

N of Rows in Working Data

File 30

Missing Value Handling Definition of Missing User-defined missing values for dependent variables are treated as missing.

Cases Used Statistics are based on cases with no missing values for any dependent variable or factor used.

Syntax EXAMINE VARIABLES=VO2_Relative VO2_Relative_obs /PLOT HISTOGRAM NPPLOT /STATISTICS DESCRIPTIVES /CINTERVAL 95 /MISSING LISTWISE /NOTOTAL.

Resources Processor Time 00:00:02,11

Elapsed Time 00:00:02,00

Case Processing Summary

N Percent N Percent N Percen 2 VO2_Relative 10 100,0% 0 0,0% 10 100,0 VO2_Relative_obs 10 100,0% 0 0,0% 10 100,0 1 VO2_Relative 20 100,0% 0 0,0% 20 100,0 VO2_Relative_obs 20 100,0% 0 0,0% 20 100,0

Descriptives

for Mean Lower Bound Upper Bound 31,857% 43,445%

5% Trimmed Mean 37,648% Median 36,636% Variance 65,598 Std. Deviation 8,0993% Minimum 24,1% Maximum 51,2% Range 27,1% Interquartile Range 11,2% Skewness ,322 , Kurtosis ,040 1, VO2_Relative_obs Mean 42,512% 3,519 95% Confidence Interval

for Mean Lower Bound Upper Bound 34,551% 50,473%

5% Trimmed Mean 41,946% Median 40,208% Variance 123,842 Std. Deviation 11,1284% Minimum 31,0% Maximum 64,2% Range 33,2% Interquartile Range 19,8% Skewness ,808 , Kurtosis -,246 1, 1 VO2_Relative Mean 55,187% 3,297 95% Confidence Interval

for Mean Lower Bound Upper Bound 48,285% 62,089%

5% Trimmed Mean 55,263%

Variance 217,477 Std. Deviation 14,7471% Minimum 31,0% Maximum 78,0% Range 47,0% Interquartile Range 26,0% Skewness ,107 , Kurtosis -1,050 , VO2_Relative_obs Mean 55,348% 3,001 95% Confidence Interval

for Mean Lower Bound Upper Bound 49,066% 61,630%

5% Trimmed Mean 55,553% Median 58,385% Variance 180,150 Std. Deviation 13,4220% Minimum 34,0% Maximum 73,0% Range 39,0% Interquartile Range 26,3% Skewness -,221 , Kurtosis -1,643 ,

Tests of Normality

Statistic df Sig. Statistic df Sig.

2 VO2_Relative ,187 10 ,200* _{,953 10 ,7}

VO2_Relative_obs ,174 10 ,200* _{,904 10 ,2}

1 VO2_Relative ,111 20 ,200* _{,948 20 ,3}

VO2_Relative_obs ,235 20 ,005 ,886 20 ,0

*. This is a lower bound of the true significance. a. Lilliefors Significance Correction

VO2_Relative_obs

Explore Age & BMI

Notes

Output Created 28-APR-2019 12:41:24

Comments

Input Data /Users/marlouettema/Docu

ments/MATLAB/FaFaS/SP SS uitwerkingen_1.sav

Active Dataset DataSet2

Filter <none>

Weight <none>

Split File Group

N of Rows in Working Data

File 30

Missing Value Handling Definition of Missing User-defined missing values for dependent variables are treated as missing.

Cases Used Statistics are based on cases with no missing values for any dependent variable or factor used.

Syntax EXAMINE VARIABLES=BMI AGE /PLOT STEMLEAF HISTOGRAM NPPLOT /STATISTICS DESCRIPTIVES /CINTERVAL 95 /MISSING LISTWISE /NOTOTAL.

Resources Processor Time 00:00:02,63

Elapsed Time 00:00:03,00

Case Processing Summary

Group

Cases

Valid Missing Total

2 BMI 10 100,0% 0 0,0% 10 100,0%

AGE 10 100,0% 0 0,0% 10 100,0%

1 BMI 20 100,0% 0 0,0% 20 100,0%

AGE 20 100,0% 0 0,0% 20 100,0%

Descriptives

Group Statistic Std. Error

2 BMI Mean 24,778 1,0203

95% Confidence Interval

for Mean Lower Bound Upper Bound 22,470 27,086

5% Trimmed Mean 24,898 Median 25,369 Variance 10,410 Std. Deviation 3,2264 Minimum 18,9 Maximum 28,5 Range 9,5 Interquartile Range 4,9 Skewness -,954 ,687 Kurtosis -,042 1,334 AGE Mean 58,20 1,737 95% Confidence Interval

for Mean Lower Bound Upper Bound 54,27 62,13

5% Trimmed Mean 58,06 Median 58,50 Variance 30,178 Std. Deviation 5,493 Minimum 50 Maximum 69 Range 19 Interquartile Range 8 Skewness ,483 ,687 Kurtosis ,382 1,334 1 BMI Mean 27,005 ,9351 95% Confidence Interval

for Mean Lower Bound Upper Bound 25,048 28,962

5% Trimmed Mean 26,993

Median 27,490

Std. Deviation 4,1819 Minimum 18,0 Maximum 36,2 Range 18,3 Interquartile Range 4,7 Skewness -,133 ,512 Kurtosis ,901 ,992 AGE Mean 55,40 3,121 95% Confidence Interval

for Mean Lower Bound Upper Bound 48,87 61,93

5% Trimmed Mean 55,83 Median 59,00 Variance 194,779 Std. Deviation 13,956 Minimum 31 Maximum 72 Range 41 Interquartile Range 20 Skewness -,703 ,512 Kurtosis -,713 ,992

Tests of Normality

Statistic df Sig. Statistic df Sig.

2 BMI ,205 10 ,200* _{,890 10 ,168}

AGE ,145 10 ,200* _{,970 10 ,890}

1 BMI ,114 20 ,200* _{,976 20 ,872}

AGE ,152 20 ,200* _{,886 20 ,023}

*. This is a lower bound of the true significance. a. Lilliefors Significance Correction

AGE