Secondary Logo

Journal Logo

Predicting Heart Rate at the Ventilatory Threshold for Aerobic Exercise Prescription in Persons With Chronic Stroke

Boyne, Pierce PT, DPT, NCS; Buhr, Sarah BS; Rockwell, Bradley BS; Khoury, Jane PhD; Carl, Daniel PhD; Gerson, Myron MD; Kissela, Brett MD, MS; Dunning, Kari PT, PhD

Journal of Neurologic Physical Therapy: October 2015 - Volume 39 - Issue 4 - p 233–240
doi: 10.1097/NPT.0000000000000102
Research Articles
Free
SDC
Watch Video Abstract

Background and Purpose: Treadmill aerobic exercise improves gait, aerobic capacity, and cardiovascular health after stroke, but a lack of specificity in current guidelines could lead to underdosing or overdosing of aerobic intensity. The ventilatory threshold (VT) has been recommended as an optimal, specific starting point for continuous aerobic exercise. However, VT measurement is not available in clinical stroke settings. Therefore, the purpose of this study was to identify an accurate method to predict heart rate at the VT (HRVT) for use as a surrogate for VT.

Methods: A cross-sectional design was employed. Using symptom-limited graded exercise test (GXT) data from 17 subjects more than 6 months poststroke, prediction methods for HRVT were derived by traditional target HR calculations (percentage of HRpeak achieved during GXT, percentage of peak HR reserve [HRRpeak], percentage of age-predicted maximal HR, and percentage of age-predicted maximal HR reserve) and by regression analysis. The validity of the prediction methods was then tested among 8 additional subjects.

Results: All prediction methods were validated by the second sample, so data were pooled to calculate refined prediction equations. HRVT was accurately predicted by 80% HRpeak (R2, 0.62; standard deviation of error [SDerror], 7 bpm), 62% HRRpeak (R2, 0.66; SDerror, 7 bpm), and regression models that included HRpeak (R2, 0.62-0.75; SDerror, 5-6 bpm).

Discussion and Conclusions: Derived regression equations, 80% HRpeak and 62% HRRpeak, provide a specific target intensity for initial aerobic exercise prescription that should minimize underdosing and overdosing for persons with chronic stroke. The specificity of these methods may lead to more efficient and effective treatment for poststroke deconditioning.

Video Abstract available for more insights from the authors (see Supplemental Digital Content 1, http://links.lww.com/JNPT/A114).

Department of Rehabilitation Sciences (P.B., S.B., B.R., D.C., K.D.), Department of Environmental Health (P.B., J.K., K.D.), Departments of Internal Medicine and Cardiology (M.G.), and Department of Neurology and Rehabilitation Medicine (B.K.), University of Cincinnati, Cincinnati, Ohio; and Division of Biostatistics and Epidemiology (J.K.), Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio.

Correspondence: Pierce Boyne, PT, DPT, NCS, Department of Rehabilitation Sciences, University of Cincinnati, 3202 Eden Ave, Cincinnati OH 45220 (Pierce.Boyne@uc.edu).

This research was supported in part with a Magistro Family Research Grant and a Promotion of Doctoral Studies Scholarship from the Foundation for Physical Therapy and by an award from the University of Cincinnati Provost's Pilot Research Program. Institutional support was provided by an NIH Clinical and Translational Science Award (8UL1-TR000077). This work was conducted in partial fulfillment of the requirements for a PhD in Epidemiology (PB) in the Department of Environmental Health at the University of Cincinnati College Of Medicine. Parts of this work were previously presented in poster format at the American Physical Therapy Association Combined Sections Meeting 2014 in Las Vegas, Nevada.

The authors declare no conflict of interest.

Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal's Web site (www.jnpt.org).

Back to Top | Article Outline

INTRODUCTION

After stroke, aerobic deconditioning is a major barrier to recovery.1 Inefficient gait patterns due to hemiparesis can double the energy cost for mobility.2 At the same time, persons with stroke have an average aerobic capacity of about 53% of normative values,3 apparently because of reductions in both cardiac output4,5 and peripheral oxygen extraction.6 This low aerobic capacity limits performance of daily activities,7 causes exertional fatigue,1 and contributes to a high risk for additional cardiovascular events.1 For example, impairment in oxygen uptake kinetics has been associated with shorter bouts of home and community ambulation after stroke.8

Treadmill aerobic exercise has been shown to yield significant improvements in gait function, aerobic capacity, and cardiovascular health after stroke,9 and aerobic intensity (eg, target heart rate [HR]) is a critical parameter of aerobic exercise prescription.10 Insufficient aerobic intensity may lead to limited improvement in aerobic fitness,11–13 whereas excessive intensity may lead to higher safety risks,14,15 or limited adherence.16 Aerobic intensity also has additional importance for persons with stroke because it also reflects the underlying neuromuscular demand of the exercise, which is a critical determinant of neuroplasticity and functional gains (eg, gait speed).11

For stroke rehabilitation, guidelines recommend aerobic exercise at 40% to 70% of HR reserve (HRR) or 55% to 80% of maximal HR.1 However, it has not yet been specified whether these should be based on HRpeak achieved during a symptom-limited exercise test or age-predicted maximal HR (AP-HRmax), which tend to differ after stroke.17,18 Furthermore, the wide percentage ranges in these recommendations could also contribute to underdosing or overdosing of aerobic intensity.19,20 Lack of specific guidelines has been cited by therapists as one of the barriers to the application of aerobic exercise in persons with neurologic conditions.21 Therefore, more specific guidance for target HR calculation is needed.

The ventilatory threshold (VT), measured with gas exchange analysis, occurs at a specific aerobic intensity and has been recommended as an optimal starting point for continuous exercise prescription across different populations.10,19,22 The VT represents the intensity limit of prolonged activity,22 above which a transition to anaerobic metabolism begins.19 Because it signifies a challenge to aerobic metabolism, exercising at the VT ensures an adequate intensity for aerobic adaptation. However, continuous exercise above the VT may not be physiologically sustainable for some persons with stroke and has been associated with negative affective responses, which may decrease exercise adherence.16 Hence, the VT is an excellent target intensity option for the initial aerobic exercise prescription.

The metabolic equipment needed to measure VT is expensive and not typically available in stroke rehabilitation settings. In addition, clinicians are not typically trained in VT measurement. Therefore, the HR at which VT usually occurs during exercise testing (HRVT) may be a more clinically feasible measure of target aerobic intensity than individual VT measurement, if HRVT can be accurately predicted. Thus, the purpose of this study was to identify an accurate method to predict HRVT for use when direct VT measurement is not available. We postulated that target HR calculations on the basis of clinical measures would be able to accurately predict HRVT.

Back to Top | Article Outline

METHODS

Design Overview

This was a secondary analysis of baseline data from a study investigating a treadmill-training intervention for persons with chronic stroke.23 A cross-sectional design was employed. Eighteen subjects performed a maximal-effort graded treadmill exercise stress test (GXT). Data from the 17 subjects with an identifiable VT during the test were used to derive HRVT predictive equations. The validity of the prediction equations was then evaluated in a second sample of 8 additional subjects, all of whom had an identifiable VT during the GXT.

Back to Top | Article Outline

Settings and Participants

This study was approved by the University of Cincinnati Institutional Review Board and was performed in a cardiac stress laboratory and a rehabilitation research laboratory within an academic medical center from January 2013 to July 2014. Subjects were recruited from the community and provided written informed consent before participation.

Inclusion criteria were (1) age 40 to 85 years, (2) unilateral stroke experienced more than 6 months before enrollment, (3) residual gait impairment, (4) able to walk 10 m over ground with no physical assistance, (5) able to walk 3 minutes on the treadmill at 0.13 m/s (0.3 mph) or more with no aerobic exercise contraindications,9,24 (6) stable cardiovascular condition (American Heart Association class B,24 allowing for aerobic capacity <6 metabolic equivalents), and (7) able to follow instructions, communicate with investigators, and provide informed consent.

Exclusion criteria were (1) evidence of significant arrhythmia or myocardial ischemia on treadmill electrocardiogram (ECG) stress test,24 (2) hospitalization for cardiac or pulmonary disease within 3 months, (3) pacemaker or implanted defibrillator, (4) lower-extremity claudication, (5) severe lower-extremity spasticity (Ashworth ≥3),25 and (6) lower-extremity weight-bearing pain of more than 4 of 10 on a visual analogue scale.

Back to Top | Article Outline

Clinical Examination

A detailed medical history was taken, medical records were reviewed, and resting HR was recorded after 5 minutes of quiet sitting. Comfortable overground walking speed was measured with the 10-m walk test,26 which was referenced to age and sex-based norms.27

Back to Top | Article Outline

Graded Exercise Testing Protocol

Subjects wore a harness secured to an overhead support system for fall protection (no body weight support was provided) during all treadmill walking. For subjects on β-blocker medication, the medication was not withdrawn for testing. Subjects performed a treadmill screening test, a symptom-limited GXT, and a repeated GXT with gas exchange analysis, each on a separate visit.

The treadmill screening test lasted 3 to 7 minutes and was used to acclimate the subject to treadmill walking and select a speed for the GXT.9 During the GXT, speed was held constant at the predetermined value and incline was increased by 2% to 4% per stage28 until the point of volitional fatigue, severe gait instability, or a cardiovascular safety limit.24,29 For the second GXT, oxygen consumption (VO2), carbon dioxide output (VCO2), and flow volume (VE) were measured from each expired breath with the TrueOne 2400 metabolic system (Parvo Medics, Inc, Sandy, UT) using a facemask interface. Heart rate was measured from an ECG and integrated into the recording on the metabolic system.

Back to Top | Article Outline

GXT Variables

The following variables were captured from the GXT data:

  • Ventilatory threshold was determined using a combination of the V-slope and ventilatory equivalents methods.19 For the V-slope method, VCO2 is plotted against VO2 and the VT is signified by the point where the slope increases its steepness. For the ventilatory equivalents method, VE/VO2 and VE/VCO2 are plotted over time and the VT is signified by a rise in VE/VO2 that occurs without a concomitant rise in VE/VCO2. Two independent raters determined the VT, and discrepancies were resolved by consensus.
  • Peak oxygen consumption rate (VO2-peak) is the highest VO2 measurement recorded during the GXT.
  • Oxygen consumption rate at the ventilatory threshold (VO2-VT) is the VO2 at the time that VT occurs.
  • Peak respiratory exchange ratio (RERpeak) is the highest ratio of carbon dioxide output to oxygen consumption recorded during the GXT, where values of more than 1.1 generally indicate a physiologic maximal test.
  • HRpeak is the maximum HR achieved during the GXT.
  • Heart rate at the ventilatory threshold (HRVT) is the HR at the time that VT occurs.
Back to Top | Article Outline

Traditional Target HR Calculation Methods

The specific percentages of HRpeak, HRRpeak, AP-HRmax, and AP-HRRmax associated with HRVT were determined using the following calculations:

Physiologic maximal HR was estimated from both the symptom-limited exercise test (HRpeak) and age (AP-HRmax) to assess which of these methods provide a better index for aerobic intensity prescription after stroke. This comparison was important because it is unclear whether HRpeak is solely limited by neurologic impairment in chronic stroke (potentially making AP-HRmax a better estimate of maximal HR) or whether HRpeak still represents an aerobic limit.

Back to Top | Article Outline

Statistical Analysis

Variables and residuals were checked for outliers, extreme influences, violation of regression assumptions, and collinearity. HRpeak and AP-HRmax were compared using a 2-sided paired t test. Descriptive statistics were then calculated for the specific percentages of HRpeak, HRRpeak, AP-HRmax, and AP-HRRmax associated with HRVT.

Linear regression analysis was used to identify an accurate predictive equation for HRVT, following the guidelines of Kleinbaum et al.32 Given a derivation sample size of 17, the final model could include no more than n − 11 = 6 independent variables (predictors) and preferably no more than n/5 = 3. The maximal model under consideration included the following 6 independent variables: age, sex, β-blocker use, comfortable overground gait speed, resting HR, and HRpeak. These variables were selected because they are clinical measures known to affect peak exercise performance (age,24 sex,24 gait speed33) and/or target HR calculation (age,24 β-blocker use,30 resting HR,24 HRpeak24). The analysis was then repeated using AP-HRmax instead of HRpeak.

All possible subsets of the independent variables were considered. Models were first organized into sets on the basis of the number of independent variables, and the models with the highest R2 from each set were selected as the initial candidates. Mallow's Cp was then used as the criterion for selecting the optimal model. When a subset of P independent variables from the maximal model is optimal, Mallow's Cp is expected to be close to P + 1.

To evaluate the validity of the different prediction equations derived from the first sample, predicted HRVT was computed for the 8 new subjects in the second sample. The cross-validation R2 was calculated as the squared Pearson correlation between predicted and observed HRVT in this new sample. Shrinkage on cross-validation was then calculated by subtracting the cross-validation R2 from the original R2 of the model in the initial derivation sample. A shrinkage value of less than 0.10 indicates a valid model and allows pooling of the data from both samples to calculate refined estimates for the predictive equations.32

Sensitivity analyses were also conducted to determine whether the results were confounded by the inclusion of subjects taking β-blockers or subjects who did not achieve RERpeak more than 1.1. Confounding is generally considered to be present if the removal of the potential confounder changes the regression coefficients of the independent variable of interest by more than 10%.32 Therefore, we compared the results in the full sample to the results for the subset of subjects not taking β-blockers and the subset of subjects who achieved RERpeak more than 1.1. SAS version 9.3 (SAS Institute, Inc, Cary, NC) was used for analysis.

Back to Top | Article Outline

RESULTS

In the combined sample (n = 25), subjects had a mean age of 60 years and a comfortable overground gait speed of 0.65 m/s (Table 1). Mean AP-HRmax was 156 bpm, adjusted for β-blocker usage30 (n = 6), and was significantly higher (P = 0.002) than the HRpeak of 132 bpm achieved during the GXT.

Table 1

Table 1

Mean observed HRVT was 106 bpm. On average, observed HRVT occurred at 80% HRpeak, 62% HRRpeak, 68% AP-HRmax, and 47% AP-HRRmax (Table 2).

Table 2

Table 2

Back to Top | Article Outline

Regression Model Selection

The regression models that were most predictive of HRVT (highest R2) in each of the sets involving from 1 to 6 independent variables are shown in Table 3. The model with 3 independent variables had the Cp value nearest to P + 1. Model diagnostics showed no extreme outliers or influences on the regression coefficients, no major violations of regression assumptions, and no collinearity problems.

Table 3

Table 3

When attempting to repeat model selection using AP-HRmax instead of HRpeak, R2 indicated different combinations of optimal variables for each of the sets involving from 1 to 6 independent variables. This instability indicated limited validity, so further regression analysis with AP-HRmax was abandoned.

Back to Top | Article Outline

Cross-Validation and Prediction Accuracy

Negative shrinkage on cross-validation values were observed for all of the HRVT prediction methods (Table 4), indicating even better predictive accuracy for the validity sample than for the derivation sample. Therefore, the derivation and validation samples were pooled to calculate refined estimates for the predictive equations. The final 3-variable equation was as follows:

Table 4

Table 4

Accuracy statistics for each of the prediction methods are shown in Table 4 and Figure 1.

Figure 1

Figure 1

Sensitivity analyses found 0% to 5% differences in the results for the percentages of HRpeak, HRRpeak, AP-HRmax, and AP-HRRmax associated with HRVT when only using the subset of subjects not taking β-blockers (n = 19) or the subset of subjects who achieved RERpeak more than 1.1 (n = 9).

Back to Top | Article Outline

DISCUSSION

This study sought to identify an accurate prediction method for HRVT using regression analysis and traditional target HR calculation methods among 17 persons with chronic stroke. All prediction methods derived from this initial sample seemed to be valid, showing even better accuracy when applied to a new validation sample of 8 similar subjects. Therefore, data were pooled to calculate refined prediction equations and accuracy statistics using all 25 subjects.

Back to Top | Article Outline

Accuracy of HRVT Prediction

Previous studies endorsed by the American College of Sports Medicine as providing accurate methods of predicting maximal HR on the basis of age24(p. 155) have reported R2 values of 0.53 to 0.81 and standard deviations of prediction error (SDerror) of 5 to 10 bpm,31,34 indicating that about 68% of subjects would have had observed values within 5 to 10 bpm of predicted values. In the present study, HRpeak, HRRpeak, and the regression models had R2 values of 0.62 to 0.75 and SDerror of 5 to 7 bpm. Therefore, these methods seem to have adequate accuracy for clinical use.

Although the AP-HRmax or AP-HRRmax methods might increase clinical utility by not requiring a maximal GXT, their lower accuracy (R2, 0.42-0.46; mean error, 9-10 bpm; SDerror, 8-9 bpm) would provide limited confidence in the predicted HRVT. These results reinforce the need for exercise testing to inform aerobic exercise prescription among persons with stroke. Moreover, exercise testing is an important part of safety screening for poststroke aerobic exercise because up to 20% to 40% of persons tested may show signs of silent myocardial ischemia.35 Future studies could further increase clinical utility by assessing the accuracy of submaximal GXT measures for predicting HRVT.

It is important to recognize that even the most accurate HRVT prediction method was more than 10 bpm away from the actual HRVT for 7 (28%) subjects. Therefore, we recommend supplementing HRVT predictions with ratings of perceived exertion and monitoring for hyperventilation, as typically recommended for HR-based aerobic exercise prescription.31 It is also possible that some persons with stroke may not tolerate prolonged exercise at HRVT or may need adapted exercise modes because of sensorimotor or cognitive impairments. Consistent with this view, it has been reported that persons with chronic stroke have a high metabolic cost in their typical everyday activities.36

Back to Top | Article Outline

Clinical Implications

Our findings have several important implications for poststroke exercise guidelines.1 First, the significant difference found between HRpeak and AP-HRmax suggests that future aerobic intensity recommendations in this population should differentiate between GXT-based and AP-HRmax-based exercise prescription. Second, although HRpeak may not always represent an aerobic endpoint (eg, the test may be limited by motor function or skeletal muscle fatigue), it seems to provide more accurate information for exercise prescription compared with AP-HRmax. Finally, this study provides more specific guidance to supplement the range-based intensity recommendations in poststroke exercise guidelines.1 Within the recommended 40% to 70% HRR range, 62% HRRpeak or 47% AP-HRRmax seem to provide the best singular targets for continuous exercise in chronic stroke. Likewise, within the recommended 55% to 80% maximum HR range, 80% HRpeak or 68% AP-HRmax seem to provide the best targets.

These recommendations are also consistent with evidence from intervention studies of poststroke aerobic exercise. Using a treadmill exercise protocol that progressed intensity from 49% to 78% HRRpeak over 3 months for 32 people with chronic stroke, Globas et al37 reported no exercise intolerance and perfect adherence. Compared with a similar training regimen with lower intensity (47%-58% HRRpeak) delivered over 6 months for 20 people with chronic stroke,38 the study by Globas et al found significantly greater improvement in aerobic capacity13 and no variable other than aerobic intensity predicted aerobic fitness gains.13 Our recommended initial training intensity (62% HRRpeak) falls in the middle of the progression by Globas et al. Therefore, the specific intensity dosing targets found in this study may lead to more efficient and effective intervention to maximize recovery from poststroke deconditioning.

Back to Top | Article Outline

Limitations, Generalizability, and Future Study

Given the limited sample size, this study had insufficient power to statistically compare the accuracy of different HRVT prediction equations or to test the statistical significance of the contributions of individual independent variables to the regression models. Therefore, the results of this study should be considered preliminary. Future studies should consider pooling data across sites to replicate our analyses. A larger sample size would increase confidence in HRVT predictions and may yield even more accurate prediction methods.

The current sample size also limits the generalizability of our findings to the larger population of persons with chronic stroke. However, the range of age, stroke characteristics, comorbidities, and function of the study sample (Table 1) is fairly large and likely representative of many persons typically seen in outpatient stroke rehabilitation. Further, β-blocker usage and the achievement of RER more than 1.1 on the maximal GXT had minimal effect on HRVT prediction in this study. Therefore, the derived equations seem to be equally applicable to patients who take and do not take β-blockers and to patients who do or do not achieve physiologic maximum intensity on the GXT. Future study could improve the external validity of these analyses by also including subjects with more recent stroke and different exercise modes.

Back to Top | Article Outline

CONCLUSIONS

HRVT was accurately predicted by 80% HRpeak, 62% HRRpeak, and regression models among persons with chronic stroke. These methods provide a specific target intensity for the initial aerobic exercise prescription, which should minimize underdosing or overdosing, potentially leading to more efficient and effective treatment for poststroke deconditioning.

Back to Top | Article Outline

REFERENCES

1. Billinger SA, Arena R, Bernhardt J, et al. Physical activity and exercise recommendations for stroke survivors: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2014;45(8):2532–2553.
2. Danielsson A, Willen C, Sunnerhagen KS. Measurement of energy cost by the physiological cost index in walking after stroke. Arch Phys Med Rehabil. 2007;88(10):1298–1303.
3. Smith AC, Saunders DH, Mead G. Cardiorespiratory fitness after stroke: a systematic review. Int J Stroke. 2012;7(6):499–510.
4. Tomczak CR, Jelani A, Haennel RG, Haykowsky MJ, Welsh R, Manns PJ. Cardiac reserve and pulmonary gas exchange kinetics in patients with stroke. Stroke. 2008;39(11):3102–3106.
5. Tomczak CR, Haykowsky MJ. Letter by Tomczak and Haykowsky regarding article, “discrepancy between cardiac and physical functional reserves in stroke.” Stroke. 2012;43(9):e91; author reply e92.
6. Jakovljevic DG, Moore SA, Tan LB, Rochester L, Ford GA, Trenell MI. Discrepancy between cardiac and physical functional reserves in stroke. Stroke. 2012;43(5):1422–1425.
7. Manns PJ, Tomczak CR, Jelani A, Cress ME, Haennel R. Use of the continuous scale physical functional performance test in stroke survivors. Arch Phys Med Rehabil. 2009;90(3):488–493.
8. Manns PJ, Tomczak CR, Jelani A, Haennel RG. Oxygen uptake kinetics: associations with ambulatory activity and physical functional performance in stroke survivors. J Rehabil Med. 2010;42(3):259–264.
9. Macko RF, Ivey FM, Forrester LW. Task-oriented aerobic exercise in chronic hemiparetic stroke: Training protocols and treatment effects. Top Stroke Rehabil. 2005;12(1):45–57.
10. Marzolini S, Oh P, McIlroy W, Brooks D. The feasibility of cardiopulmonary exercise testing for prescribing exercise to people after stroke. Stroke. 2012;43(4):1075–1081.
11. Hornby TG, Straube DS, Kinnaird CR, et al. Importance of specificity, amount, and intensity of locomotor training to improve ambulatory function in patients poststroke. Top Stroke Rehabil. 2011;18(4):293–307.
12. Garber CE, Blissmer B, Deschenes MR, et al. American College of Sports Medicine position stand. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: Guidance for prescribing exercise. Med Sci Sports Exerc. 2011;43(7):1334–1359.
13. Lam JM, Globas C, Cerny J, et al. Predictors of response to treadmill exercise in stroke survivors. Neurorehabil Neural Repair. 2010;24(6):567–574.
14. Albert CM, Mittleman MA, Chae CU, Lee IM, Hennekens CH, Manson JE. Triggering of sudden death from cardiac causes by vigorous exertion. N Engl J Med. 2000;343(19):1355–1361.
15. Haskell WL, Lee IM, Pate RR, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation. 2007;116(9):1081–1093.
16. Ekkekakis P, Parfitt G, Petruzzello SJ. The pleasure and displeasure people feel when they exercise at different intensities: decennial update and progress towards a tripartite rationale for exercise intensity prescription. Sports Med. 2011;41(8):641–671.
17. MacKay-Lyons MJ, Howlett J. Exercise capacity and cardiovascular adaptations to aerobic training early after stroke. Top Stroke Rehabil. 2005;12(1):31–44.
18. Billinger SA, Taylor JM, Quaney BM. Cardiopulmonary response to exercise testing in people with chronic stroke: a retrospective study. Stroke Res Treat. 2012;2012:987637.
19. Mezzani A, Hamm LF, Jones AM, et al. Aerobic exercise intensity assessment and prescription in cardiac rehabilitation: a joint position statement of the European Association for Cardiovascular Prevention and Rehabilitation, the American Association of Cardiovascular and Pulmonary Rehabilitation, and the Canadian Association of Cardiac Rehabilitation. J Cardiopulm Rehabil Prev. 2012;32(6):327–350.
20. Biasin L, Sage MD, Brunton K, et al. Integrating aerobic training within subacute stroke rehabilitation: a feasibility study. Phys Ther. 2014;94(12):1796–806.
21. Doyle L, Mackay-Lyons M. Utilization of aerobic exercise in adult neurological rehabilitation by physical therapists in Canada. J Neurol Phys Ther. 2013;37(1):20–26.
22. Guazzi M, Adams V, Conraads V, et al. EACPR/AHA scientific statement. clinical recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. Circulation. 2012;126(18):2261–2274.
23. Boyne P, Dunning K, Carl D, Gerson M, Khoury J, Kissela B. Within-session responses to high-intensity interval training in chronic stroke. Med Sci Sports Exerc. 2015;47(3):476–484.
24. American College of Sports Medicine (ACSM). ACSM's Guidelines for Exercise Testing and Prescription. 9th ed. Philadephia, PA: Lippincott Williams & Wilkins; 2014.
25. Ashworth B. Preliminary trial of carisoprodol in multiple sclerosis. Practitioner. 1964;192:540–542.
26. Tilson JK, Sullivan KJ, Cen SY, et al. Meaningful gait speed improvement during the first 60 days poststroke: minimal clinically important difference. Phys Ther. 2010;90(2):196–208.
27. Bohannon RW, Williams Andrews A. Normal walking speed: a descriptive meta-analysis. Physiotherapy. 2011;97(3):182–189.
28. Ivey FM, Hafer-Macko CE, Macko RF. Task-oriented treadmill exercise training in chronic hemiparetic stroke. J Rehabil Res Dev. 2008;45(2):249–259.
29. Fletcher GF, Ades PA, Kligfield P, et al. Exercise standards for testing and training: A scientific statement from the American Heart Association. Circulation. 2013;128(8):873–934.
30. Brawner CA, Ehrman JK, Schairer JR, Cao JJ, Keteyian SJ. Predicting maximum heart rate among patients with coronary heart disease receiving beta-adrenergic blockade therapy. Am Heart J. 2004;148(5):910–914.
31. Gellish RL, Goslin BR, Olson RE, McDonald A, Russi GD, Moudgil VK. Longitudinal modeling of the relationship between age and maximal heart rate. Med Sci Sports Exerc. 2007;39(5):822–829.
32. Kleinbaum D, Kupper L, Nizam A, Rosenberg E. Applied Regression Analysis and Other Multivariable Methods. 5th ed. Boston, MA: Cengage Learning; 2014.
33. Kelly JO, Kilbreath SL, Davis GM, Zeman B, Raymond J. Cardiorespiratory fitness and walking ability in subacute stroke patients. Arch Phys Med Rehabil. 2003;84(12):1780–1785.
34. Tanaka H, Monahan KD, Seals DR. Age-predicted maximal heart rate revisited. J Am Coll Cardiol. 2001;37(1):153–156.
35. Adams RJ, Chimowitz MI, Alpert JS, et al. Coronary risk evaluation in patients with transient ischemic attack and ischemic stroke: a scientific statement for healthcare professionals from the Stroke Council and the Council on Clinical Cardiology of the American Heart Association/American Stroke Association. Circulation. 2003;108(10):1278–1290.
36. Kafri M, Myslinski MJ, Gade VK, Deutsch JE. High metabolic cost and low energy expenditure for typical motor activities among individuals in the chronic phase after stroke. J Neurol Phys Ther. 2014;38(4):226–232.
37. Globas C, Becker C, Cerny J, et al. Chronic stroke survivors benefit from high-intensity aerobic treadmill exercise: a randomized control trial. Neurorehabil Neural Repair. 2012;26(1):85–95.
38. Macko RF, Ivey FM, Forrester LW, et al. Treadmill exercise rehabilitation improves ambulatory function and cardiovascular fitness in patients with chronic stroke: a randomized, controlled trial. Stroke. 2005;36(10):2206–2211.
Figure

Figure

Keywords:

aerobic exercise; deconditioning; exercise testing; exercise training; intensity; stroke

Supplemental Digital Content

Back to Top | Article Outline
© 2015 Neurology Section, APTA