Abstract
Introduction:
Community-based structured exercise training (CB-SET) programs are beneficial for patients with peripheral artery disease (PAD). However, the impact of lower levels of walking activity accumulated separately from formal exercise is unclear. The aim of this study was to determine the relation of non-exercise walking (NEW) activity with exercise performance in PAD.
Methods:
This was a post hoc analysis from twenty patients with PAD enrolled in a 12-week CB-SET program using diaries and accelerometry. Formal exercise (3 sessions·week−1) was detected using patient-reported diary entries that corresponded with accelerometer step data. NEW activity was characterized as steps completed over five days each week, excluding steps achieved during formal exercise sessions. The primary exercise performance outcome was peak walking time (PWT) assessed on a graded treadmill. Secondary performance outcomes included claudication onset time (COT) from the graded treadmill and peak walking distance (PWD) achieved during the six-minute walk test (6MWT). Partial Pearson correlations evaluated the relation of NEW activity (step·week−1) with exercise performance outcomes using exercise session intensity (step·week−1) and duration (min·week−1) as covariates.
Results:
NEW activity demonstrated a moderate, positive correlation with change in PWT (r=0.50, p=0.04). Other exercise performance outcomes were not significantly related to NEW activity (COT: r=0.14; 6MWT PWD: r=0.27).
Conclusions:
A positive association was demonstrated between NEW activity and PWT following 12 weeks of CB-SET. Interventions to increase physical activity levels outside of formal exercise sessions may be beneficial for patients with PAD.
Keywords: claudication, accelerometry, community-based exercise
INTRODUCTION
Peripheral artery disease (PAD) is the result of atherosclerotic plaque accumulation in the arteries of the lower extremities leading to decreased leg blood flow.1 This inadequate arterial perfusion results in claudication, described by patients as pain, cramping, and/or fatigue in the legs that attenuates with rest.2 Because of the sedentary lifestyle of many patients with PAD due to leg pain, grave prognoses such as adverse cardiovascular events and premature death can occur.3
Upright, weight bearing walking is the primary recommended modality for improving exercise performance outcomes of patients with PAD, which is intuitive given that walking is a primary movement for completing activities of daily living.4 Evidence over the past decade suggests that community-based structured exercise training (CB-SET) programs using walking as the principal mode may be beneficial for this population. Results following these programs indicated improved exercise performance, quality of life, and overall increased step counts, thus demonstrating the importance of adhering to walking exercise programs in community settings.5–8
Despite the need to ensure that patients with PAD complete prescribed walking exercise, it remains challenging for patients to complete given the leg pain they experience and high intensity levels that exercise may demand. Recent evidence suggests incorporating an increased amount of overall daily steps may be beneficial for patients thus allowing an alternative for activity above sedentary levels.9 Because monitored and structured walking exercise completed in community settings represents a small fraction of time patients ambulate, it may be useful to separately quantify the non-exercise walking (NEW) activity performed outside of formal, planned exercise sessions. If increased levels of NEW activity are found to have a link to outcomes such as exercise performance, additional opportunities may be available for healthcare providers to promote activity outside of formal exercise sessions.
The aim of this post hoc study was to examine the relation of NEW activity with change in exercise performance outcomes over the course of a 12 week CB-SET intervention time period. We hypothesized that patients with PAD would demonstrate a positive association between NEW activity and change in the primary performance outcome of peak walking time (PWT) assessed on a graded treadmill. A secondary aim determined the relationship between NEW activity and change in patient-reported outcomes assessed by the Walking Impairment Questionnaire (WIQ) and Medical Outcomes Study Short-Form 36-item (SF-36) questionnaire.
METHODS
Study Design
This was a post hoc analysis of data collected from a two site, randomized, controlled trial involving patients with PAD and claudication.10 The purpose of the parent study was to evaluate change in exercise performance and quality of life outcomes after the implementation of a CB-SET program. The intent-to-treat trial enrolled forty-four patients ≥40 years of age with ankle-brachial index (ABI) ≤0.90. Specific inclusion and exclusion criteria have been described previously.10 Relevant institutional review boards approved the trial and patients completed written informed consent prior to participation.
Peak walking time (min) was the primary exercise performance outcome assessed on a graded treadmill using a modified Gardner protocol.8,11 The treadmill began at 2.0 mph (0.89 m·sec−1) and 0% grade and increased by 2% grade every two minutes. If a patient was able to complete stage 6, the grade remained constant at 10% and speed was increased by 0.5 mph (0.22 m·sec−1) every two minutes until PWT was achieved. As a secondary treadmill performance outcome, claudication onset time (COT; min) was assessed during the treadmill test and defined as the point when lower limb symptoms first ensued. Additional performance outcomes of interest in the secondary analysis included peak walking distance (PWD; min) and claudication onset distance (COD; min) assessed from a standard six-minute walk test (6MWT) in a 50 ft (15.24 m) hallway as previously described.12 Subjective perceptions of patients’ walking ability was measured using the WIQ distance, speed, and stair climbing ability subcomponents,13 with the mental and physical components of the SF-36 summarizing quality of life.14
The previous CB-SET trial consisted of 12 weeks of walking exercise using training, monitoring, and coaching components modeled after supervised, hospital-based programs. Briefly, patients assigned to the intervention group (n=22) were instructed to exercise 3 days∙week−1 and record details of sessions in an exercise diary. Patients were also asked to wear piezoelectric accelerometers (Modus Health, LLC) during formal exercise sessions and for 10 hours a day, 5 days a week. Controls (n=22) were given advice to walk at home and not provided accelerometers, thus were not used in the current post hoc analyses. Further details of the CB-SET intervention components are provided in the original study and a previous PAD community-based exercise trial.8,10
Formal Exercise Sessions and NEW Activity Identification
For the current post hoc analysis, formal CB-SET exercise sessions were identified based on previously published procedures15 that used patients’ self-reported exercise dates and times found in their activity logs. The dose of activity that patients completed during the formal exercise sessions were summarized by graphical outputs via the StepWatch Activity Monitor System software program as strides completed (v. 3.4, Modus Health, LLC, Washington, D.C.) Stride counts were bracketed within the graphical program linking the diary-defined exercise start and stop times provided by patients with objective accelerometer data (Figure 1).
Figure 1.
Example of formal exercise session bracketing based on patient-defined start/stop times provided in the diary. Vertical black lines indicate strides in a given minute.
These methods allowed research staff a means to identify stride counts that partitioned an exercise session from NEW activity. The process of bracketing the exercise sessions to capture the strides patients completed was conducted for the three prescribed exercise days each week, if available. Total strides and time of activity in the diary-defined range were extracted, with strides being doubled to quantify steps achieved as the accelerometer only counted steps from one limb. The steps and time detected for each exercise session were then totaled for each patient and normalized to steps∙week−1 and min∙week−1. Additional characterization rules for identifying formal exercise sessions are described elsewhere.15
Non-exercise walking activity was then summarized by first excluding the steps patients completed during formal CB-SET exercise sessions. Any steps completed before and after a formal exercise session were characterized as NEW activity. For days when patients did not have diary-defined formal exercise, the entire day of accelerometer step data were considered NEW activity (normalized to steps·week−1). When seven days of accelerometer confirmed activity was present in a given week, the two days with the lowest number of steps were excluded to ensure congruence with the protocol-defined instructions of wear time (5 days per week).
Statistical Analyses
Partial Pearson correlations were used to assess the relationships between NEW activity (steps∙week−1) with 12 week change in outcomes. Formal exercise session intensity (steps∙week−1) and duration (min∙week−1) were used as covariates. Due to missing data in the original study, baseline observation carryforward and the expectation maximization algorithm16 were used to impute outcomes and accelerometer step data for analyses. Assumptions of parametric testing (e.g., linearity) were evaluated prior to analyses. In the event data were found to not be linear, non-parametric tests were considered (i.e., Spearman rank order correlation) in lieu of transformations. Significance was defined a priori as α=0.05 and analyses were performed with a standard statistical software package (IBM, SPSS, v.24, Armonk, NY).
RESULTS
Five of the 22 patients were excluded from outcome testing prior to the 12-week study end point due to voluntary dropout and experiencing adverse events (e.g., requiring peripheral revascularization) in the parent study.10 Missing data for outcomes that were assessed at the onset of the study were imputed using the baseline observation carried forward approach. To impute missing accelerometer step data, the expectation maximization algorithm was used due to a lack of activity data over the course of 12 weeks. Prior to imputation using the algorithmic approach, data were found to be missing completely at random (Little’s MCAR test: χ2(75)=0.0, p=1.00). Two patients were subsequently excluded from the final analyses as they were deemed extreme outliers (determined from standard Q-Q plots). Thus, the final sample used for the post hoc analysis was n=20. Table 1 depicts baseline characteristics and outcomes for the final analyzed group of patients.
Table 1.
Characteristics of patients with PAD in the current post hoc analysis.
| Characteristic | |
|---|---|
| Sample size, n | 20a |
| Age, y | 70.4±7.4 |
| Gender, % | |
| -male | 65 |
| -female | 35 |
| Race, % | |
| -White | 95 |
| -American Indian/Alaskan Native | 5 |
| Height, cm | 171.5±10.1 |
| Weight, kg | 91.7±30.8 |
| Body mass index, kg·m−2 | 31.0±9.9 |
| Resting heart rate, beats·min−1 | 73.0±13.4 |
| Resting systolic BP, mmHg | 133.2±16.3 |
| Resting diastolic BP, mmHg | 68.8±11.1 |
| Ankle-brachial index | 0.71±0.16 |
| HBA1C, % | 6.1±0.9 |
| Exercise sessions | |
| -total, sessions | 23.3±8.7 |
| -compliance, % | 64.7±24.2b |
| -intensity, steps·week−1 | 4524.0±3112.3 |
| -duration, min·week−1 | 89.8±39.0 |
| NEW activity | |
| -intensity, steps·week−1 | 20336.8±8586.0 |
Continuous data are mean±SD. Categorical data are percent of total sample.
Two patients from original n=22 intervention patients were extreme outliers and removed.
Based on matches between diary and accelerometer defined sessions.
BP, blood pressure; HBA1C, glycated hemoglobin; NEW, non-exercise walking; PAD, peripheral artery disease
Correlation Analyses
Partial Pearson correlations between NEW activity and PWT change at 12 weeks was significantly positively correlated (r=0.50, p=0.04; Figure 2) when controlling for exercise session intensity (steps∙week−1) and duration (min∙week−1).
Figure 2.
Scatterplot demonstrating the association of NEW activity intensity with 12 week change in PWT, with formal exercise session steps and duration as covariates.
*Moderately strong relation between NEW activity and change in PWT, p=0.04.
NEW, Non-exercise walking; PWT, peak walking time
Near significant positive relationships were found between treadmill-derived COT (p=0.09), the 6MWT performance outcome COD (p=0.05), and the WIQ distance score (p=0.07). Table 2 summarizes correlation analyses between NEW activity and change in exercise performance outcomes, as well as with WIQ and SF-36 subcomponent scores.
Table 2.
Relation of NEW activity with change in outcomes using formal exercise session steps and duration as covariates.
| Outcomes | NEW activity (steps·week−1) | |
|---|---|---|
| ra | p value | |
| PWT (min) | 0.50 | 0.04b |
| COT (min) | 0.41 | 0.09 |
| 6MWT PWD (m) | 0.27 | 0.28 |
| 6MWT COD (m) | 0.46 | 0.05 |
| WIQ distance (%) | 0.43 | 0.07 |
| WIQ speed (%) | 0.26 | 0.29 |
| WIQ stair (%) | 0.11 | 0.67 |
| WIQ combo (%) | 0.36 | 0.15 |
| SF-36 Mental (%) | 0.23 | 0.37 |
| SF-36 Physical (%) | −0.15 | 0.56 |
Controlling for exercise session intensity and duration.
Significant positive relation between 12 week change in PWT with NEW activity. 6MWT, Six minute walk test; COD, claudication onset distance; COT, claudication onset time; NEW, non-exercise walking; PWD, peak walking distance; PWT, peak walking time; SF-36, Short-Form 36-item questionnaire; WIQ, Walking Impairment Questionnaire
DISCUSSION
This study sought to determine the relationship of NEW activity with exercise and questionnaire-based outcomes in patients with PAD who were prescribed 12 weeks of community-based exercise. Results indicated a moderate correlation between the activity patients completed outside of formal exercise sessions with the primary outcome of change in PWT. Although change in secondary exercise performance and WIQ and SF-36 scores were not significantly related to NEW activity, strong numerical trends were observed. These data suggest there may be clinical benefit for patients with PAD who perform higher levels of walking activity outside of formal, planned exercise sessions in home and community settings.
Previous studies in other patient populations have demonstrated non-exercise related activity improves health and reduces morbidity and mortality. Hamer et al17 determined that higher levels of physical activity separate from exercise in older adults were associated with a lower risk of cardiovascular death. In a longitudinal study over 12.5 years, older adult men over the age of sixty who participated in greater levels of non-exercise activity were found to have a reduced risk of an initial adverse cardiovascular event.18 In the PAD literature, several studies have concluded that overall greater levels of physical activity during daily life may lead to reduced risk for premature death.3,19 Further, Gardner and colleagues9 recently evaluated the daily total step counts of patients with PAD over 7 days using piezoelectric accelerometry. The authors concluded that patients with PAD had 29% lower daily step counts compared to age-matched controls, thus highlighting the importance of incorporating a greater amount of walking into daily routines to reach step count goals.9 However, to the best of our knowledge, studying the specific relationship of NEW activity over the course of an entire 12 week intervention, when ruling out the potential effects of formal exercise sessions, has yet to be evaluated in a PAD population.
The lack of significant correlations of the other exercise performance outcomes may be due to a number of reasons. The greater levels of NEW activity may have translated to patients being able to tolerate pain more effectively, in addition to becoming more accustomed to walking greater distances, thus significance was found only with PWT change. However, the near significant relation of NEW activity and COT also indicates that additional walking activity outside of exercise sessions may delay leg symptoms at submaximal levels. Improvements in patient-perceived walking ability or quality of life scores were not related to greater levels of NEW activity. Patients may not perceive benefit from overall increased step counts from a quality of life perspective, although objective measures may still be positively improved. The relationship between the WIQ distance score and NEW activity did nearly reach significance indicating patients may recognize that they can walk farther with less difficulty due to claudication. The small sample size may have also been a factor. A larger patient population for analysis may have provided more power to conclude other improvements in key variables were related to increased levels of NEW activity.
We chose to remove the potential impact that formal exercise may have had on the amount of NEW activity patients completed by including time and intensity of exercise steps achieved as covariates. The impact that exercise has on NEW activity is limited and unclear in both short- and long-term studies in other older adult populations.20 Some studies have found that when patients complete formal exercise training, they are less likely to participate in non-exercise activity or even decrease other activities as a result of training.21,22 Conversely, studies have reported that increased amounts of training may in fact increase activity that is not considered exercise occurring outside of training sessions.23,24 Because of the equivocal consensus, we chose specifically to evaluate only physical activity that occurred before and after exercise and on days where a patient did not record a formal exercise session. Future research is needed to examine the impact formal exercise has on non-exercise physical activity in patients with vascular disease.
There were several limitations to the study. We used correlation analyses and although a positive association between PWT and NEW activity was observed, inferences of causation cannot be drawn from the results. Only intervention patients were used in the analysis because they had available accelerometer data, including exercise sessions that could be linked with diary reported physical activity in the parent study. This resulted in a small sample size and thus there is the need for further research into NEW activity in patients not prescribed exercise (controls). Other limitations included the known inherent issues using patients self-reports of exercise achieved including: 1) inaccurate or failure to report dates/times exercised, 2) falsification of session details, and 3) chronographic device failure/inaccuracies. However, we were able to cross-check the exercise sessions over the course of an entire 12-week intervention (rather than the typical 7 days most studies employ) by checking data from both the diary and the accelerometer data for session confirmations. Lastly, we were unable to determine any specific types of non-exercise physical activity, thus all steps outside of formal exercise were considered NEW activity. There is a possibility that additional, unreported exercise sessions could have occurred within the investigator-defined NEW activity windows.
CONCLUSIONS
In conclusion, the significant relationship between NEW activity and PWT change in patients with PAD following a 12-week CB-SET program indicates a potential new opportunity for nurses and other healthcare providers to improve patient outcomes. The vast majority of physical activity research and clinical care focuses on formal exercise interventions as the primary means for improving function in vascular disease. However, the positive results of the current study may be used to direct further research into NEW activity for patients with PAD. Titration and optimization of activity that occurs outside of formal, planned exercise sessions may represent a novel approach to improve the physical activity profiles of patients with vascular disease.
HIGHLIGHTS.
Peak walking relates to non-exercise walking activity in peripheral artery disease
Correlations demonstrated positive numerical trends for other exercise outcomes
Activity occurring outside of exercise sessions should be promoted by clinicians
FUNDING
Research reported in this publication was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under Award Number K01HL115534. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
DECLARATION OF CONFLICTING INTERESTS
The authors have no potential conflicts of interest with respect to the research, authorship, and publication of this article.
REFERENCES
- 1.Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG. Inter-society consensus for the management of peripheral arterial disease (TASC II). J Vasc Surg 2007;45 Suppl S:S5–67. [DOI] [PubMed] [Google Scholar]
- 2.Hiatt WR, Brass EP. Pathophysiology of Peripheral Artery Disease, Intermittent Claudication, and Critical Limb Ischemia. In: Creager MA, Beckman JA, Loscalzo J, eds. Vascular Medicine: A Companion to Braunwald’s Heart Disease. 2nd ed. Philadelphia, PA: Elsevier Inc.; 2013:223–230. [Google Scholar]
- 3.Gardner AW, Montgomery PS, Parker DE. Physical activity is a predictor of all-cause mortality in patients with intermittent claudication. J Vasc Surg 2008;47(1):117–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Treat-Jacobson D, McDermott MM, Bronas UG, et al. Optimal Exercise Programs for Patients With Peripheral Artery Disease: A Scientific Statement From the American Heart Association. Circulation 2019;139(4):e10–e33. [DOI] [PubMed] [Google Scholar]
- 5.Gardner AW, Parker DE, Montgomery PS, Blevins SM. Step-monitored home exercise improves ambulation, vascular function, and inflammation in symptomatic patients with peripheral artery disease: a randomized controlled trial. J Am Heart Assoc 2014;3(5):e001107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McDermott MM, Liu K, Guralnik JM, et al. Home-based walking exercise intervention in peripheral artery disease: a randomized clinical trial. JAMA 2013;310(1):57–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.McDermott MM, Guralnik JM, Criqui MH, et al. Home-based walking exercise in peripheral artery disease: 12-month follow-up of the GOALS randomized trial. J Am Heart Assoc 2014;3(3):e000711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mays RJ, Hiatt WR, Casserly IP, et al. Community-based walking exercise for peripheral artery disease: An exploratory pilot study. Vasc Med 2015;20(4):339–347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gardner AW, Montgomery PS, Wang M, et al. Daily step counts in participants with and without peripheral artery disease. J Cardiopulm Rehabil Prev 2021;41(3):182–187. [DOI] [PubMed] [Google Scholar]
- 10.Mays RJ, Treat-Jacobson D, Casserly IP, Hiatt WR, Regensteiner JG. Community Walking Exercise for Patients With Peripheral Artery Disease (GAIT). ClinicalTrials.gov Identifier: NCT02075502 (https://clinicaltrials.gov/ct2/show/NCT02075502). [Google Scholar]
- 11.Gardner AW, Skinner JS, Cantwell BW, Smith LK. Progressive vs single-stage treadmill tests for evaluation of claudication. Med Sci Sports Exerc 1991;23(4):402–8. [PubMed] [Google Scholar]
- 12.Montgomery PS, Gardner AW. The clinical utility of a six-minute walk test in peripheral arterial occlusive disease patients. J Am Geriatr Soc 1998;46(6):706–11. [DOI] [PubMed] [Google Scholar]
- 13.Regensteiner JG, Steiner JF, Panzer RJ, Hiatt WR. Evaluation of walking impairment by questionniare in patients with peripheral artery disease. J Vasc Med Biol 1990;2:142–52. [Google Scholar]
- 14.Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30(6):473–83. [PubMed] [Google Scholar]
- 15.Mays RJ, Wesselman CW, White R, et al. Automated detection of exercise sessions in patients with peripheral artery disease: Evidence for an exercise dose response to training. J Cardiopulm Rehabil Prev 2021;41(3):176–181. [DOI] [PubMed] [Google Scholar]
- 16.Dempster AP, Laird NM, Rubin DB. Maximum likelihood from incomplete data via the EM algorithm. J Roy Stat Soc 1977;Series B, 39(1):1–38. [Google Scholar]
- 17.Hamer M, de Oliveira C, Demakakos P. Non-exercise physical activity and survival: English longitudinal study of ageing. Am J Prev Med 2014;47(4):452–60. [DOI] [PubMed] [Google Scholar]
- 18.Ekblom-Bak E, Ekblom B, Vikstrom M, de Faire U, Hellenius ML. The importance of non-exercise physical activity for cardiovascular health and longevity. Br J Sports Med 2014;48(3):233–8. [DOI] [PubMed] [Google Scholar]
- 19.Garg PK, Tian L, Criqui MH, et al. Physical activity during daily life and mortality in patients with peripheral arterial disease. Circulation 2006;114(3):242–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Melanson EL. The effect of exercise on non-exercise physical activity and sedentary behavior in adults. Obes Rev 2017;18 Suppl 1:40–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Morio B, Montaurier C, Pickering G, et al. Effects of 14 weeks of progressive endurance training on energy expenditure in elderly people. Br J Nutr 1998;80(6):511–9. [DOI] [PubMed] [Google Scholar]
- 22.Meijer EP, Westerterp KR, Verstappen FT. Effect of exercise training on total daily physical activity in elderly humans. Eur J Appl Physiol Occup Physiol 1999;80(1):16–21. [DOI] [PubMed] [Google Scholar]
- 23.Meijer GA, Janssen GM, Westerterp KR, Verhoeven F, Saris WH, ten Hoor F. The effect of a 5-month endurance-training programme on physical activity: evidence for a sex-difference in the metabolic response to exercise. Eur J Appl Physiol Occup Physiol 1991;62(1):11–7. [DOI] [PubMed] [Google Scholar]
- 24.Hunter GR, Wetzstein CJ, Fields DA, Brown A, Bamman MM. Resistance training increases total energy expenditure and free-living physical activity in older adults. J Appl Physiol (1985) 2000;89(3):977–84. [DOI] [PubMed] [Google Scholar]