edicine and Rehabilitation

Archives of Physical M journal homepage: www.archives-pmr.org

Archives of Physical Medicine and Rehabilitation 2013;94:1223-9

ORIGINAL ARTICLE

Does Exercise Tolerance Testing at 60 Days Poststroke PredictRehabilitation Performance?

Dorian K. Rose, PhD, PT,a,b Andrea L. Behrman, PhD, PT,c Stephen E. Nadeau, MD,b,d

Anatole D. Martin, PhD, PT,a Richard S. Schofield, MD,b,e Julie K. Tilson, MS, DPT,f

Steven Y. Cen, PhD,g Xiaomin Lu, PhD,h Samuel S. Wu, PhD,b,h Stan Azen, PhD,g

Pamela W. Duncan, PhD, PT,i for the LEAPS Investigative Team

From the aDepartment of Physical Therapy, University of Florida, Gainesville, FL; bMalcom Randall VA Medical Center, Gainesville, FL;cDepartment of Neurological Surgery, University of Louisville, Louisville, KY; dDepartment of Neurology and eDivision of CardiovascularMedicine, University of Florida, Gainesville, FL; fDivision of Biokinesiology and Physical Therapy, University of Southern California, LosAngeles, CA; gDepartment of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA; hDepartmentof Biostatistics, University of Florida, Gainesville, FL; and iWake Forest Baptist Medical Center, Winston-Salem, NC.

Abstract

Objective: To assess the relationship between exercise tolerance test (ETT) performance at 6 weeks poststroke and subsequent performance in

a treadmill and overground locomotor training program (LTP).

Design: Prospective cohort study.

Setting: Exercise testing laboratory in either a primary care hospital or outpatient clinic.

Participants: Community-dwelling individuals (NZ469), 54.9�19.0 days poststroke, enrolled in the Locomotor Experience Applied Post-Stroke

randomized controlled trial.

Interventions: Not applicable.

Main Outcome Measures: For participants randomly assigned to LTP, the number of sessions needed to attain the training goal of 20 minutes of

treadmill stepping was determined. Regression analyses determined the contribution of ETT performance (cycling duration), age, and 6-minute

walk test (6MWT) distance to attainment of the stepping duration goal.

Results: Age, 6MWT, and ETT performance individually accounted for 10.74%, 10.82%, and 10.76%, respectively, of the variance in the number

of sessions needed to attain 20 minutes of stepping. When age and 6MWT were included in the model, the additional contribution of ETT

performance was rendered nonsignificant (PZ.150).

Conclusions: To the extent that ETT performance can be viewed as a measure of cardiovascular fitness rather than neurologic impairment,

cardiovascular fitness at the time of the ETT did not make a significant unique contribution to the number of sessions needed to achieve 20

minutes of stepping. The 6MWT, which involves less intensive exercise than the ETT and therefore is likely to be predominantly affected by

neurologic impairment and muscular condition, appeared to account for as much variance as the ETT.

Archives of Physical Medicine and Rehabilitation 2013;94:1223-9

ª 2013 by the American Congress of Rehabilitation Medicine

Numerous factors contribute to functional recovery after stroke.Age,1-3 cognition,4 stroke type,2,5 motor impairment,1-3,6 and hemi-sensory loss7,8 all impact recovery. Another less overt potential

Supported by the National Institute of Neurological Disorders and Stroke and the National

Center for Medical Rehabilitation Research (grant no. RO1 NS050506).

No commercial party having a direct financial interest in the results of the research supporting

this article has conferred or will confer a benefit on the authors or on any organization with which

the authors are associated.

Clinical Trial Registration Number: NCT0024391.

0003-9993/13/$36 - see front matter ª 2013 by the American Congress of Re

http://dx.doi.org/10.1016/j.apmr.2013.01.031

contributor to rehabilitation response is tolerance for exercise.Practice intensity and repetition are factors known to improve reha-bilitation outcomes and require exercise capacity.9 Comorbidcardiovascular conditions, present in 75% of poststroke patients,represent the leading cause of death in stroke survivors. However,formal exercise testing before the initiation of rehabilitation is notstandard of care.10 Exercise tolerance evaluation before rehabilitationmay bevaluable in creating effective exercise programs, but its role ininforming rehabilitation performance has not been assessed.

habilitation Medicine

1224 D.K. Rose et al

The randomized controlled trial, Locomotor ExperienceApplied Post-Stroke (LEAPS),11,12 compared a locomotor trainingprogram (LTP) that included stepping on a treadmill with partialbody weight support (BWS), followed by overground walkingtraining, to an impairment-based home exercise program (HEP) ofstrength and balance, for improving the functional level ofwalking.13,14 The goal of each LTP session was to walk at .89m/sfor a total of 20 minutes of stepping. Given the prevalence ofcomorbid cardiovascular disease and the rigor of LTP for indi-viduals just 2 months poststroke, an exercise tolerance test (ETT)was requisite to assess exercise capacity and confirm the ability tosafely participate in progressive LTP. We evaluated the relation-ship between ETT performance and participants’ subsequent LTPperformance. We posited that higher ETT performance (longercycling duration) would be associated with greater locomotortraining readiness (number of sessions needed to attain �20min ofstepping on the treadmill).

Methods

Participants

The LEAPS trial was approved by the institutional review boardat each trial site, and written informed consent was obtained.Before trial randomization participants passed a chart reviewand physical and cognitive screen 5 to 45 days poststroke.Participants proceeded to an ETT as the final stage ofscreening.11,12

Inclusion criteria included stroke within the previous 45 days,residual paresis, ability to walk 3.0m with 1-person assist, abilityto follow a 3-step command, and walking speed <0.8m/s. Giventhe presence of comorbid cardiovascular disease and the potentialrigor of the intervention, there were extensive exclusion criteriaspecific to cardiovascular and pulmonary conditions: myocardialinfarction or heart surgery within the previous 3 months; history ofcongestive heart failure; serious and/or unstable cardiac arrhyth-mias; hypertrophic cardiomyopathy; severe aortic stenosis; anginaor dyspnea at rest or during activities of daily living; class 3 or 4heart failure according to the New York Heart Association;coronary artery bypass grafts or valve replacement within theprevious 3 months if participation was not approved by a cardio-thoracic surgeon and either a cardiologist or primary care

List of abbreviations:

APMHR age-predicted maximum heart rate

BWS body weight support

DBP diastolic blood pressure

ECG electrocardiography

e-LTP early locomotor training program

ETT exercise tolerance test

HEP home exercise program

LEAPS Locomotor Experience Applied Post-Stroke

LEFM-M Lower Extremity Fugl-MeyereMotor Assessment

l-LTP late locomotor training program

LTP locomotor training program

MET metabolic equivalent

RPE rate of perceived exertion

RPP rate-pressure product

SBP systolic blood pressure

6MWT 6-minute walk test

10MWT 10-meter walk test_VO2max maximum oxygen consumption

physician; history of serious chronic obstructive pulmonarydisease or use of supplemental oxygen; history of pulmonaryembolism within 6 months; and severe hypertension with systolicblood pressure (SBP) >200mmHg and diastolic blood pressure(DBP) >110mmHg, not reduced to �180/100mmHg withmedical therapy.11

Procedure

Exercise tolerance testAfter completion of a screening questionnaire to assess cardiacmedical history, cardiac disease risk factors, family cardiachistory, and current medications, a physical examination bya cardiologist determined ETT eligibility. If contraindicated,participants were referred to their primary care physician andexcluded from the trial.

The ETT, using a previously established bicyclea ergometryprotocol,15,16 was conducted in cardiologist-supervised assess-ment laboratories with trial staff providing standardized instruc-tions. Electrocardiography (ECG) electrodes (nZ10) in thestandard configuration monitored cardiac status. ECG, bloodpressure, and heart rate were recorded for 1 minute supine, fol-lowed by 1 minute sitting. Participants sat on the bicycle or behindit in a chair (if poor sitting balance) with their hemiparetic legsecured to the ergometer with a boot and their hemiparetic arm tothe handle bar with a glove support. Vital signs were recorded foran additional 2 minutes. Requisite parameters to commence wereas follows: DBP �100mmHg, SBP �180mmHg, and heart rate�100 beats/min. Participants pedaled 40 to 60rpm with 10W/min(from 0W) power increase. Heart rate from the ECG tracing wasrecorded every minute, and blood pressure readings, obtainedmanually, were recorded every 2 minutes. Participants reportedtheir rate of perceived exertion (RPE), using the Borg Scale,17

every minute.Target ETT endpoint was 90% age-predicted maximum heart

rate (APMHR; 220eage). For those receiving beta-blockers, thetarget endpoint was RPE >18. ETT was terminated beforeattaining these target endpoints if 40rpm could not be maintainedor if the participant experienced the onset of preestablishedlimiting symptoms. At test termination, a final RPE score wasrecorded, resistance was removed, and pedaling continued at40rpm for 1 minute. Blood pressure and heart rate were recorded1, 3, and 5 minutes posttest or until vital signs returned to pretestvalues. The cardiologist provided written documentation forETT termination and determined trial eligibility. If a preestab-lished ETT failure point occurred, the participant was managedmedically, referred for follow-up care, and excluded fromthe trial.

ETT outcome measures

Percentage APMHR and absolute rate-pressure product (RPP)at peak exercise intensity (peak RPP) described maximalexercise performance. Percentage APMHR was chosen asa primary outcome measure because of its relation to maximumoxygen consumption ( _VO2max).18 RPP, an index of myocardialoxygen consumption, was calculated at rest and at maximalexercise according to the following formula: RPP Z (heart rate� SBP)/100.19 The percentage increase in RPP from rest tomaximal exercise (%RPP) was also calculated. ETT perfor-mance was defined as cycling duration from the initial

www.archives-pmr.org

Exercise testing 60 days poststroke 1225

downstroke of the foot pedal to attainment of one of the pre-determined test termination criteria. Peak metabolic equivalents(METs) were estimated using participants’ weight, age, andcycling duration.20

Baseline assessments

Those who passed the ETT proceeded to complete baselinephysical, psychosocial, and cognitive assessments.11 For thepurposes of this study, we examined performance on the 10-mwalk test (10MWT), the Lower Extremity Fugl-MeyereMotorAssessment (LEFM-M), and the 6-minute walk test (6MWT).

10-m walk testImpairment severity was defined by gait speed (<0.4m/s, severe;�0.4m/s and <0.8m/s, moderate) on the 10MWT. Trainedassessors used a standardized procedure previously described ina poststroke walking intervention study.21 The time (average of 2trials) to traverse 10m at the participant’s comfortable paceusing their customary assistive device, orthotic device, or bothwas recorded.

Lower Extremity Fugl-MeyereMotor AssessmentThe LEFM-M22 is a test that assesses movements of the paretichip, knee, and ankle in and out of synergistic patterns, as wellas reflexes and coordination. Each movement is graded ona 3-point scale (0, cannot perform; 1, performs partially; 2,performs fully).

Six-minute walk testUsing a previously standardized protocol,23 participants walked asfar as they were able in 6 minutes with their customary assistivedevice, orthotic device, or both.

Group allocation

In the LEAPS trial,12 participants were randomly assigned to earlyLTP (e-LTP at 2mo poststroke), late LTP (l-LTP at 6mo post-stroke), or HEP. Randomization was stratified for walking speedseverity as determined by the 10MWT.

LTP intervention

The overall goal of LTP was to achieve independent communitywalking. For each intervention session, the goal was to attain 20minutes of step training using a BWS system over a treadmill.We felt that 20 minutes of stepping on the treadmill wouldtranslate into functional walking distances over ground. Walkingduration, speed, and BWS were recorded for each trainingsession to monitor progression, and vital signs were exten-sively monitored.

Statistical analyses

Statistical analyses were conducted using SAS 9.2.b To assess theeffect of stroke severity on ETT performance, all randomizedparticipants were analyzed. To assess the relationship of ETTperformance to locomotor training readiness, those randomlyassigned to the 2 LTP groups were included in the analysis. Alinear regression model with ETT performance (cycling duration)as the dependent variable, stroke severity (moderate, severe) as the

www.archives-pmr.org

independent variable, and age and beta-blocker use as covariatesassessed the impact of stroke severity on ETT. A discrete-timeproportional hazard model with the number of sessions as thediscrete time was used to test our hypothesis that individuals withbetter ETT performance (longer cycling duration) would needfewer training sessions to attain �20 minutes of stepping on thetreadmill. In this analysis, for those who did not attain �20minutes of stepping, the number of needed training sessions wastreated as censored at the last observed session, meaning the truenumber of sessions needed was larger than the censored time.Training group (e-LTP vs l-LTP), age, baseline severity, LEFM-M,and 6MWT served as covariates. The McKelvey-Zavonia methodwas used to calculate the proportion of variability in the number ofneeded training sessions explained by the ETT performance andcovariates.24

Results

Participants

For the LEAPS trial, 4909 patients were screened and 3137initially excluded.12 At the secondary screening, an additional1299 were excluded, leaving 473 eligible to proceed to the ETT.In this study of ETT performance at 2 months poststroke, weexamined 3 cohorts of participants: (1) participants whocompleted an ETT (nZ469); (2) participants randomly assignedto the LEAPS12 trial (nZ408); and (3) participants in the 2 LTPcohorts (nZ282) (table 1). There were no differences in demo-graphics, time postonset, or medical comorbidities among the 3study cohorts except that those who completed the ETT but werenot randomly assigned (nZ61) had a lower National Institutes ofHealth Stroke Scale score (PZ.001) and were more likely tohave had a right hemisphere stroke (PZ.037) than thoserandomly assigned to the trial. There were no differencesbetween those randomly assigned to the LTP cohorts and thosewho were not.12

Exercise testing performance

Four of the 473 eligible for the ETT were unable to initiate theETT because heart rate or blood pressure values exceeded theacceptable range or because of orthopedic limitations, resultingin a cohort of 469 (see table 1). Participants pedaled, onaverage, 6.0�2.3 minutes, attained a mean %APMHR of72%�13%, a peak RPP of 198�53, and an increase in RPPfrom rest to maximal exercise of 110%�58%. Estimated meanpeak METs, derived from cycling power settings, was 3.5�1.1(table 2). There were no adverse events associated withthe ETT.

Beta-blockers (prescribed for 42% of the study population)significantly influenced maximal exercise performance. Therewere differences in all exercise responses for those receivingbeta-blockers compared with those who were not except forpeak blood pressure (see table 2). Mean %APMHR was 77%�12% for those not receiving beta-blockers versus 67%�12% forthose who were (P<.001). Mean peak RPP was 208�50for those not taking a beta-blocker compared with 184�52 forthose who were (P<.0001). These results corroborate previ-ously reported effects of beta-blockers on exercise testing10

(see table 2).

Table 1 Characteristics of participants

Characteristic

All ETT Participants

(nZ469)

All Randomized Participants

(nZ408)

LTP Randomized Participants

(nZ282)

Age (y) 62.9�12.7 62.7�12.7 62.4�12.5

Sex (M/F) 259/210 224/184 159/123

Race

Asian 63 (13.4) 54 (13.2) 39 (13.8)

Black 103 (22) 90 (22.1) 66 (23.4)

White 270 (57.67) 236 (57.8) 158 (56.0)

Other 33 (6.8) 28 (6.9) 19 (6.7)

Time poststroke at ETT (d) 54.9�19.0 54.8�20.1 54.8�23.5

Stroke type

Ischemic 383 (81.7) 327 (80.1) 225 (79.8)

Hemorrhagic 81 (17.3) 76 (18.6) 53 (18.8)

Uncertain 5 (1.1) 5 (1.2) 4 (1.4)

Stroke location

Right 233 (49.7) 197 (48.3) 134 (47.5)

Left 159 (33.9) 143 (35.0) 95 (33.7)

Brainstem 70 (14.9) 62 (15.2) 47 (16.7)

Bilateral 6 (1.3) 6 (1.5) 6 (2.1)

Missing 1 (0.2) 0 (0) 0 (0)

NIH Stroke Scale 7.3�3.7 7.5�3.8 7.7�3.8

On b-blocker 198 (42.2) 168 (41.2) 116 (41.1)

Hypertension 344 (81) 331 (81.1) 227 (80.5)

Hypercholesterolemia 187 (39.9) 159 (39.0) 110 (39.0)

Diabetes 169 (36) 140 (34.3) 96 (34)

Cardiovascular disease (angina, CHF, MI,

arrhythmia)

114 (24.3) 109 (26.8) 72 (25.5)

Coronary artery disease 68 (14.5) 56 (13.7) 35 (12.4)

Peripheral vascular disease 39 (8.3) 37 (9.1) 24 (8.5)

Atrial fibrillation 34 (7.2) 31 (7.6) 22 (7.8)

NOTE. Values are mean � SD, n, or n (%).

Abbreviations: CHF, congestive heart failure; F, female; M, male; MI, myocardial infarction; NIH, National Institutes of Health.

1226 D.K. Rose et al

Ninety-seven percent of participants performed the ETT toan a priori specified endpoint without significant cardiovascularfindings of concern; 3% (nZ15) did not. Development ofexercise-induced ST-segment depression was the most preva-lent reason for ETT failure (table 3). The primary reason fortest termination was fatigue. Participants either stoppedpedaling stating they were fatigued (32%) or were unable tomaintain at least 40rpm cycling cadence (22%) (table 4).Overall, 39% of individuals attained the heart rate or Borg RPEtarget endpoint; 61% did not. Of the 454 individuals who

Table 2 ETT performance characteristics

Exercise Response All (nZ469)

Peak heart rate (beats/min) 114�22

% APMHR 72�13

Peak SBP (mmHg) 174�29

Peak DBP (mmHg) 87�15

Peak RPP 198�53

% RPP increase from rest to maximal exercise 110�58

Exercise duration (min) 5.99�2.27

Estimated peak METs 3.46�1.03

Highest RPE achieved 16�3

NOTE. Values are mean � SD or as otherwise indicated.

* Between b-blocker and no b-blocker.

completed the ETT without a safety concern, 46 did notproceed to trial randomization for noncardiovascular reasons(eg, moved from the area, withdrew consent, or no longer mettrial criteria).

ETT performance by stroke severity

There was a significant association between walking impairmentand ETT performance; that is, the moderate group (gait speed,�.04m/s) had significantly longer cycling duration (6.5�2.3min)

b-Blocker (nZ197) No b-Blocker (nZ272) P*

104�20 120�20 <.0001

67�12 77�12 <.0001

176�30 172�28 .144

88�15 86�15 .290

184�52 208�52 <.0001

100�53 117�57 .0008

5.77�2.05 6.15�2.41 .073

3.22�0.93 3.63�1.07 .0001

17�2 16�3 .045

www.archives-pmr.org

Table 3 ETT failures (nZ15)

Reason No. of Participants

ST-segment depression 7

Severe hypertension 2

Hypotension (drop in SBP >20mmHg) 2

Angina 2

Heart block 1

Peripheral claudication 1

Exercise testing 60 days poststroke 1227

than the severe (gait speed, <.04m/s) group (5.5�2.0min)(P<.001). Corresponding estimated peak METs were significantlygreater for the moderate (3.7�1.1) compared with the severegroup (3.3�1.0) (P <.01). In addition, age was a prognostic co-variate for ETT performance, with younger individuals pedalinglonger than older participants (P<.001).

ETT performance and locomotor training readiness

ETT performance was negatively associated with the number ofsessions to first attain a stepping duration of �20 minutes(table 5). The full model accounted for only 11.0% of thevariance in locomotor readiness. Age, 6MWT, and ETTperformance independently accounted for 10.74%, 10.82%, and10.76% of the variance, respectively (table 6). ETT perfor-mance accounted for an additional .05% of the variance(PZ.004) in the age-covaried model, and .04% of the variance(PZ.002) in the 6MWT-adjusted model. There was a highdegree of covariance between ETT cycling duration and the6MWT (Pearson rZ.34; P<.001). When age and 6MWTperformance were included in the model, the additionalcontribution made by ETT performance was rendered nonsig-nificant (PZ.150). The addition of group (e-LTP/l-LTP),severity (moderate/severe), and LEFM-M to age accounted fora small but statistically significant increment in variance rela-tive to age alone (increment, .15%; P<.0001).

Table 4 Reasons for termination for those who passed the ETT

(nZ454)

Reason No. (%) of Participants

General/voluntary fatigue 145 (32)

Borg RPE >18 107 (24)

Unable to maintain cycling cadence of

at least 40rpm

100 (22)

Attained 90% THR 68 (15)

SBP �220mmHg or DBP �120mmHg 26 (6)

ST-segment depression 1* (<1.0)

Hypotension 1* (<1.0)

Dyspnea 2 (<1.0)

Angina 1* (<1.0)

Other 3 (<1.0)

Abbreviation: THR, target heart rate.

* The 3 participants listed in table 4 whose ETTs were terminated

secondary to hypotension, angina, and downsloping ST segment were

determined by clinical judgment of the onsite cardiologist to pass the

ETT despite the presence of an ETT failure indicator.

www.archives-pmr.org

Discussion

ETT performance and locomotor training intervention

To the best of our knowledge, this is the first prospective studywith individuals poststroke to examine the relationship betweena plausible measure of cardiorespiratory fitness and subsequentambulatory performance during a rehabilitation intervention. Astrength of this study is that it examined a large population ofpatients early after stroke onset and followed them prospectivelyas they engaged in an exercise intervention. The most salientfinding of this study was that, although we studied all ourpotentially predictive variables, in total these variables accountedfor only 11.01% of the variance in the number of training sessionsneeded to attain 20 minutes of stepping.

The small amount of variance in locomotor readiness accountedfor by our models is not entirely surprising given the numerouspatient-centered factors that contribute to locomotor trainingprogression, as well as the implicit dynamic between patient andtherapist. Patient-centered factors we could not explicitly measure,such as emotional state, medication influences, muscle forceproduction, interlimb coordination, and dynamic balance, alsoinfluence training progression and may explain the relatively smallcontribution of ETT performance to locomotor training readiness.To the extent that ETT performance can be regarded as a measure ofcardiovascular fitness, our study suggests that cardiovascular fitnessat baseline had little impact on LTP progression.

Previous cross-sectional studies in chronic stroke report no,25

low,26 or moderate27 correlation between cardiorespiratoryfitness ( _VO2max or peak oxygen consumption) and the 6MWT,suggesting that factors other than cardiorespiratory status (musclestrength, balance, spasticity) contributed to ambulatory capacity asmeasured by the 6MWT. In contrast, and in corroboration of ourresults, Kelly et al28 reported a strong correlation between peakcardiorespiratory fitness (peak oxygen consumption) and6MWT distance.

Our results do not suggest that 6MWT and age should replacethe ETT as a screening tool for asymptomatic cardiac disease, butrather that these simple and easily obtained clinical and demo-graphic data can be instructive for LTP intervention planning andgoal setting. The 6MWT is simple to administer with a minimalequipment required, and therapists can use it to establish treatmentplans and to set goals.

This ETT protocol aimed to test exercise capacity and assesscardiovascular response to exercise, not to detect previouslyundiagnosed ischemic heart disease. In this highly screened cohortof poststroke patients, 10 individuals (2.1%) developed ischemicsymptoms or electrocardiographic (ST-segment depression)abnormalities. This low prevalence is likely due to the extensivescreening to exclude those with a significant cardiac historyor symptoms.

Participants in this submaximal exercise test achieved anaverage of 73% APMHR. Although this value is comparable tovalues found in other studies10,15,25,28 of submaximal exercisetesting poststroke, it is less than the standard American College ofSports Medicine guidelines for termination of an exercise stresstest at 90% APMHR. Participants may not have reached a work-load intensity that would provoke signs of myocardial ischemia.However, we would emphasize that the goal was not to exercisepatients to their maximal levels but rather to screen a high-risk,deconditioned patient subgroup for evidence of significant occultcardiac disease.

Table 5 ETT performance by session at which 20 minutes of stepping was attained

Session

All (nZ262) Early-LTP (nZ131) Late-LTP (nZ131)

ETT Performance

(min) No. of Participants

ETT Performance

(min) No. of Participants

ETT Performance

(min) No. of Participants

1 7.95�2.85 21 8.90�2.84 8 7.37�2.81 13

2 6.13�2.32 65 6.19�2.29 28 6.09�2.38 37

3 6.40�2.06 44 6.50�2.30 28 6.22�1.61 16

4 5.86�2.21 30 5.80�2.58 14 5.91�1.90 16

5 5.80�2.34 30 6.36�2.20 17 5.06�2.40 13

6 5.49�1.72 10 5.31�1.94 3 5.56�1.78 7

7 6.06�2.02 9 6.30�2.71 5 5.75�0.97 4

8 5.30�2.29 13 4.82�2.19 8 6.07�2.48 5

9 5.24�1.69 9 5.63�1.95 5 4.75�1.42 4

10 5.03�1.69 9 4.31�2.04 3 5.39�1.55 6

�11 5.19�1.97 22 5.06�2.13 12 5.35�1.87 10

NOTE. Values are mean � SD or as otherwise indicated.

1228 D.K. Rose et al

Study limitations

A limitation is that this study used cycle ergometry instead ofa treadmill to assess cardiorespiratory fitness. However, forminimally ambulatory poststroke participants, unassisted ambu-lation on a treadmill is not feasible. The advantages of cycleergometry testing in this population have been discussed.28

Testing with a cycle ergometer in this population with a limitedwork capacity is advantageous in that workload can be increasedin small, precise increments to prevent overstressing patients withseverely limited exercise tolerances and asymmetric lowerextremity strength. According to the LEAPS trial design, the ETTwas conducted before randomization, creating a 4-month lagbefore the start of LTP intervention for those randomly assigned tol-LTP. We recognize that further deconditioning may haveoccurred in this cohort. Participants were community dwelling andhighly screened for known cardiac conditions, so these resultscannot be generalized to all patients poststroke.

Not all participants attained 90% APMHR during the ETT. Inthese cases, ETT performance may not be a true measure ofcardiorespiratory fitness. We did not directly measure oxygenuptake, carbon dioxide production, respiratory exchange ratio, or

Table 6 Contributions to locomotor training readiness

Variable

% Variance

Explained

Increment in

Variance

Explained by

ETT (%)

Odds Ratio*

(P) for ETT

Age 10.74 NA

6MWT 10.82 NA

ETT 10.76 NA 1.17 (<.0001)

Age þ ETT 10.79 .05 1.14 (.0004)

6MWT þ ETT 10.86 .04 1.12 (.0019)

Age þ 6MWT

þ ETT

10.99 .01 1.06 (.1495)

Abbreviation: NA, not applicable.

* Estimated ratio of conditional odds of attaining locomotor training

readiness (stepping duration of �20min) at a session, given that

locomotor training readiness was not attained up to that session with

1-minute increase in ETT cycling duration.

expired ventilation, but rather used %APMHR as our primaryoutcome measure secondary to its relationship to _VO2max.

18 Wechose number of sessions to attain �20 minutes of stepping as ourindicator of locomotor training readiness although other trainingparameters, such as number of sessions to attain .89m/s or numberof sessions to attain 0% BWS, could also potentially serve asindicators. Because no ETT failures proceeded to trial participation,we cannot comment on its sensitivity or specificity for identifyingthose at risk for subsequent cardiovascular events.

Conclusions

Although ETT cycling performance was correlated with loco-motor training readiness, it was not associated with the attainmentof �20 minutes of stepping when participants’ age and baseline6MWT distance were taken into account. Cardiorespiratoryimpairment, as measured by the ETT, did not make an importantcontribution to the progression of stepping time. Neurologicimpairment may have influenced ETT performance, limiting theextent the ETT can be regarded as a true measure of cardio-respiratory fitness. The 6MWT, conducted pretraining, can be usedto develop a locomotor training intervention plan and set reha-bilitation goals.

Suppliers

a. HealthCare International, Inc., PO Box 1509, Langley, WA98260.

b. SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513.

Keywords

Exercise; Exercise test; Gait; Rehabilitation; Stroke

Corresponding author

Dorian K. Rose, PhD, PT, University of Florida, Health ScienceCenter, PO Box 100154, Gainesville, FL 32610-0154. E-mailaddress: dkrose@phhp.ufl.edu.

www.archives-pmr.org

Exercise testing 60 days poststroke 1229

Acknowledgments

LEAPS Investigative Team includes the following investigators:Wake Forest University, Winston-Salem, NC: Pamela Duncan, PT,PhD. Duke University Administrative Coordinating Center,Durham, NC: Sarah Hayden, Mysha Sissine, Quishi Feng, PhD.Brooks Rehabilitation Hospital, Jacksonville, FL: Deborah Stew-art, MD, Trevor Paris, MD, Joann Gallichio, PT, DSc. FloridaHospital, Orlando, FL: Mitchell Freed, MD, Michelle Dolske,PhD, Craig Moore, PT, Bettina Brutsch, PT. Long BeachMemorial Hospital, Long Beach, CA: H. Richard Adams, MD,Diehma Hoang, MD, Anita Correa, PT. Sharp RehabilitationCenter, San Diego, CA: Jerome Stenehjem, MD, Roxanne Hon,MD, Molly McLeod, PT. UCLA Medical Center, Los Angeles,CA: David Alexander, MD. University of Southern California,Los Angeles, CA: Julie Hershberg, DPT, Samneang Ith-Chang,DPT. Clinical Coordinating CentereUniversity of Florida, Gain-esville, FL: Andrea L. Behrman, PT, PhD, Dorian K. Rose, PT,PhD. Clinical Coordinating CentereUniversity of Southern Cal-ifornia, Los Angeles, CA: Julie K. Tilson, DPT, MS. DataManagement and Analysis CentereUniversity of Southern Cal-ifornia, Los Angeles, CA: Steven Cen, PhD, Chris Hahn, MS,James Gardener. University of Florida, Gainesville, FL: YunfengDai, MS, Xiaomin Lu, PhD. Consultants: Anatole D. Martin, PhD,Richard Schofield, MD, University of Florida, Gainesville, FL.Steering Committee: Pamela Duncan, PT, PhD, Wake ForestUniversity, Winston-Salem, NC; Andrea L. Behrman, PT, PhD,Samuel S. Wu, PhD, Stephen Nadeau, MD, University of Florida,Gainesville, FL; Stanley P. Azen, PhD, University of SouthernCalifornia, Los Angeles, CA; Bruce H. Dobkin, MD, University ofCalifornia, Los Angeles, Los Angeles, CA; Sarah K. Hayden,Duke University, Durham, NC.

References

1. Denti L, Agosti M, Franceschini M. Outcome predictors of rehabilita-

tion for first stroke in the elderly. Eur J Phys Rehabil Med 2008;44:3-11.

2. Kelly PJ, Furie KL, Shafqat S, Rallis N, Chang Y, Stein J. Functional

recovery following rehabilitation after hemorrhagic and ischemic

stroke. Arch Phys Med Rehabil 2003;84:968-72.

3. Lin JH, Hsieh CL, Lo SK, Hsiao SF, Huang MH. Prediction of

functional outcomes in stroke inpatients receiving rehabilitation. J

Formos Med Assoc 2003;102:695-700.

4. Mutai H, Furukawa T, Araki K, Misawa K, Hanihara T. Factors

associated with functional recovery and home discharge in stroke

patients admitted to a convalescent rehabilitation ward. Geriatr Ger-

ontol Int 2012;12:215-22.

5. Katrak PH, Black D, Peeva V. Do stroke patients with intracerebral

hemorrhage have a better functional outcome than patients with

cerebral infarction? PM R 2009;1:427-33.

6. Shelton FD, Volpe BT, Reding M. Motor impairment as a predictor of

functional recovery and guide to rehabilitation treatment after stroke.

Neurorehabil Neural Repair 2001;15:229-37.

7. Prescott RJ, Garraway WM, Akhtar AJ. Predicting functional outcome

following acute stroke using a standard clinical examination. Stroke

1982;13:641-7.

www.archives-pmr.org

8. Reding MJ, Potes E. Rehabilitation outcome following initial unilat-

eral hemispheric stroke. Life table analysis approach. Stroke 1988;19:

1354-8.

9. Rose D, Paris T, Crews E, et al. Feasibility and effectiveness of circuit

training in acute stroke rehabilitation. Neurorehabil Neural Repair

2011;25:140-8.

10. Hill DC, Ethans KD, MacLeod DA, Harrison ER, Matheson JE.

Exercise stress testing in subacute stroke patients using a combined

upper- and lower-limb ergometer. Arch Phys Med Rehabil 2005;86:

1860-6.

11. Duncan PW, Sullivan KJ, Behrman AL, et al. Protocol for the Loco-

motor Experience Applied Post-Stroke (LEAPS) trial: a randomized

controlled trial. BMC Neurol 2007;7:39.

12. Duncan PW, Sullivan KJ, Behrman AL, et al. Body-weight-sup-

ported treadmill rehabilitation after stroke. N Engl J Med 2011;

364:2026-36.

13. Schmid A, Duncan PW, Studenski S, et al. Improvements in speed-

based gait classifications are meaningful. Stroke 2007;38:2096-100.

14. Perry J, Garrett M, Gronley JK, Mulroy SJ. Classification of walking

handicap in the stroke population. Stroke 1995;26:982-9.

15. Yates JS, Studenski S, Gollub S, et al. Bicycle ergometry in subacute-

stroke survivors: feasibility, safety, and exercise performance. J Aging

Phys Act 2004;12:64-74.

16. Potempa K, Lopez M, Braun LT, Szidon JP, Fogg L, Tincknell T.

Physiological outcomes of aerobic exercise training in hemiparetic

stroke patients. Stroke 1995;26:101-5.

17. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports

Exerc 1982;14:377-81.

18. Londeree BR, Ames SA. Trend analysis of the % VO2 max-HR

regression. Med Sci Sports 1976;8:123-5.

19. Kitamura K, Jorgensen CR, Gobel FL, Taylor HL, Wang Y. Hemo-

dynamic correlates of myocardial oxygen consumption during upright

exercise. J Appl Physiol 1972;32:516-22.

20. Thompson WR. American College of Sports Medicine. Guidelines for

exercise testing and prescription. Philadelphia: Lippincott, Williams &

Wilkins; 2010.

21. Sullivan KJ, Brown DA, Klassen T, et al. Effects of task-specific

locomotor and strength training in adults who were ambulatory after

stroke: results of the STEPS randomized clinical trial. Phys Ther 2007;

87:1580-602.

22. Fugl-Meyer AR, Jaasko L, Leyman I, Olsson S, Steglind S. The post-

stroke hemiplegic patient. 1. A method for evaluation of physical

performance. Scand J Rehabil Med 1975;7:13-31.

23. Solway S, Brooks D, Lacasse Y, Thomas S. A qualitative systematic

overview of the measurement properties of functional walk tests used

in the cardiorespiratory domain. Chest 2001;119:256-70.

24. DeMaris A. Explained variance in logistic regressionda Monte

Carlo study of proposed measures. Sociol Methods Res 2002;31:

27-74.

25. Eng JJ, Dawson AS, Chu KS. Submaximal exercise in persons with

stroke: test-retest reliability and concurrent validity with maximal

oxygen consumption. Arch Phys Med Rehabil 2004;85:113-8.

26. Pang MY, Eng JJ, Dawson AS. Relationship between ambulatory

capacity and cardiorespiratory fitness in chronic stroke: influence of

stroke-specific impairments. Chest 2005;127:495-501.

27. Tang A, Sibley KM, Bayley MT, McIlroy WE, Brooks D. Do func-

tional walk tests reflect cardiorespiratory fitness in sub-acute stroke? J

Neuroeng Rehabil 2006;3:23.

28. Kelly JO, Kilbreath SL, Davis GM, Zeman B, Raymond J. Cardio-

respiratory fitness and walking ability in subacute stroke patients.

Arch Phys Med Rehabil 2003;84:1780-5.