Characterizing Heart Rate Response During Upper Extremity Repetitive Task Practice in Chronic Stroke

Approximately 800,000 people experience a stroke each year in the United States (Billinger et al., 2014). The motor consequences of stroke on the performance of activities of daily living (ADLs) are well recognized and documented (Dhamoon et al., 2009). An additional effect of stroke is diminished cardiopulmonary capacity because poststroke aerobic capacity (peak volume of oxygen consumption [VO_2peak]) ranges from 26% to 87% of normal values (Smith et al., 2012). Given that VO_2peak is associated with the ability to independently perform ADLs (Cress & Meyer, 2003), integrating aerobic activity during therapy may enhance recovery.

Engaging in aerobic exercise poststroke can improve cardiovascular function, depressive symptoms, cognition and memory, health-related quality of life, and fatigue (Billinger et al., 2014). Additionally, when coupled with medication (e.g., antihypertensives), aerobic exercise reduces risk of recurrent stroke by 80% (Billinger et al., 2014). Given its potential to facilitate recovery and protect against further cardiovascular complications, aerobic exercise may be a valuable intervention for survivors of stroke.

Sedentary behavior is common during acute stroke management, in which survivors are sedentary approximately 87% of a typical day (Barrett et al., 2018). Given that physical and occupational therapy are foundational in stroke rehabilitation, an opportunity exists for these professions to encourage safe and appropriate levels of physical activity in all phases of recovery. Traditional stroke rehabilitation interventions generally are low intensity with only a few minutes spent within in an aerobic zone in typical inpatient and outpatient sessions (Barrett et al., 2018; Kuys et al., 2006; MacKay-Lyons & Makrides, 2002a; Polese et al., 2014). However, researchers from these studies estimated aerobic intensity using age-predicted maximum heart rate. Survivors of stroke attain on average 77% to 85% of the age-predicted maximum heart rate during cardiopulmonary stress tests (Dobrovolny et al., 2003; MacKay-Lyons & Makrides, 2002b; Yates et al., 2004); thus, it is possible that researchers estimating poststroke aerobic intensity using age-predicted calculations are underestimating aerobic intensity.

It is also possible that even with accurate estimates of aerobic intensity, poststroke rehabilitation still is not eliciting a robust cardiovascular response. Several studies have shown that rehabilitation professionals typically do not monitor heart rate during poststroke rehabilitation (Doyle & MacKay-Lyons, 2013; MacKay-Lyons & Makrides, 2002a; Nathoo et al., 2018), which in part may be because they are prioritizing other aspects of recovery such as quality of movement and ADL independence (Connell et al., 2018; Latham et al., 2006; MacKay-Lyons & Makrides, 2002a). It may be feasible to address these deficits while eliciting an aerobic response. This method would allow clinicians to “kill two birds with one stone,” maximizing the benefits of poststroke rehabilitation.

No study to date has examined the aerobic response with repetitive task practice (RTP), which uses goal-oriented functional movements and emphasizes achieving a high volume of repetitions while still providing appropriate motor challenges (Lang & Birkenmeier, 2013). The primary aim of this analysis was to characterize heart rate response during upper extremity (UE) RTP administered to improve motor function. The secondary aim was to determine factors that predict a higher heart rate reserve (HRR) during UE RTP.

A randomized clinical trial was conducted investigating the effects of aerobic exercise combined with UE RTP (Grant K01HD092556; ClinicalTrials.gov registration number NCT03819764). Data for this analysis were from a subset of participants randomized to the control group (UE RTP only). The study was approved by the Cleveland Clinic institutional review board, and participants completed the informed consent process in accordance with the Declaration of Helsinki.

A total of 19 participants with chronic stroke were included in the analysis. These participants were assessed at baseline for UE impairment with the Fugl-Meyer Assessment (UE–FMA; Singer & Garcia-Vega, 2017), and their scores ranged from 19 to 55. Our pilot studies have shown that people presenting within this range on the UE–FMA have sufficient volitional movement to participant in and make meaningful gains from RTP interventions. Additional inclusion criteria were (1) ≥6 mo since a single ischemic or hemorrhagic stroke confirmed by neuroimaging, (2) ambulatory ≥20 m with no more than contact guard assistance, and (3) ages 18 to 85 yr. Exclusion criteria were (1) hospitalization for myocardial infarction, heart failure, or heart surgery within 3 mo; (2) cardiac arrhythmia; (3) hypertrophic cardiomyopathy; (4) severe aortic stenosis; (5) pulmonary embolus; (6) significant contractures; (7) antispasticity injection to paretic UE within 3 mo of enrollment; and (8) other contraindications to exercise (Linder et al., 2023).

After completion of a baseline cardiopulmonary exercise test and assessment of baseline UE–FMA, participants attended 24 therapy sessions over 8 wk, completing 90 min of UE RTP per session. Each session aimed for 450 repetitions. Heart rate was continuously monitored (Figure 1) with a Wahoo chest strap (Wahoo Fitness, Atlanta, GA). Each task was annotated within the Wahoo app. Mean HRR for each task was calculated with the Karvonen formula (American College of Sports Medicine, 2021). Outcomes obtained included time spent completing RTP (minutes), repetitions completed, mean HRR during each task (%), and the following task descriptors: (1) task position (standing or sitting), (2) target height (below shoulder height or at or above shoulder height), and (3) task type (gross, fine motor, or a combination of gross and fine motor). After completing the intervention, participants completed another cardiopulmonary exercise test and UE–FMA to quantify change in aerobic capacity and UE impairment.

Summary statistics were calculated for demographic characteristics and RTP outcomes. A linear mixed effects model was conducted with HRR as the dependent variable and the following independent variables as fixed effects: task position (standing or sitting), baseline UE–FMA, target height (below shoulder height or at or above shoulder height), percentage of age-predicted VO_2peak achieved at the baseline stress test, and task type (fine motor, gross motor, or a combination). Random intercepts per subject were included in the model. Tests of significance were conducted with the Kenward–Roger approximation on the results of the linear mixed effects model. Relationships between time spent in a moderate or vigorous intensity aerobic zone, change from baseline to end of treatment UE–FMA, and change from baseline to end of treatment percentage of age-predicted VO_2peak were assessed visually with scatterplots.

Demographic, session, and task characteristics are shown in Table 1.

Position, F(1, 2947) = 270.03, p < .0001; target height, F(1, 2946) = 13.80, p < .001; and task type, F(2, 2947) = 8.66, p < .001, were significant predictors of aerobic response. On the basis of coefficients from the model (Table 2), targets at or above shoulder height, standing position, and gross motor tasks were associated with a higher HRR, with a standing position being the strongest predictor of higher HRR. Baseline UE–FMA, F(1, 16) = 0.31, p = .584, and percentage of predicted VO_2peak, F(1, 16) = 1.45, p = .246, were not significant predictors.

Of tasks that were moderate to vigorous intensity, 54.1% were performed in standing, 79.7% were gross motor, and 27.9% had targets at or above shoulder height. Among sessions that elicited an aerobic response, a median of 15.2 (Quartile 1 = 8.7, Quartile 3 = 43.5) total minutes were spent in moderate to vigorous intensities. Visual assessment of scatterplots revealed no relationship between time spent in an aerobic zone and change in aerobic capacity or UE–FMA.

Most RTP in this study was performed at <40% HRR, which is in alignment with previous studies reporting low aerobic response during physical or occupational therapy (Barrett et al., 2018; MacKay-Lyons & Makrides, 2002a; Polese et al., 2014). Although time spent in an aerobic zone was not associated with improvement in UE impairment or aerobic capacity, in this observational study, the goal of RTP was to provide appropriate motor challenge for functional recovery. The administration of RTP was not manipulated to elicit a cardiovascular response. In traditional physical or occupational therapy, therapists similarly prioritize skills such as movement quality and ADL independence over aerobic training (Connell et al., 2018; Latham et al., 2006); additionally, aerobic exercise itself is not reimbursable as skilled therapy. Given that aerobic exercise is beneficial for poststroke recovery, it is important to consider methods that can combine aerobic training with traditional goals for stroke rehabilitation.

Our results suggest that although it is possible to administer UE rehabilitation in a manner that simultaneously taxes the cardiovascular system, therapists must be intentional about the mode of training. In our study, higher aerobic response was elicited by tasks performed in standing, where the target was at or above shoulder height, and with gross motor movements, regardless of the participant’s UE impairment or aerobic capacity. A standing position was the strongest predictor of higher HRR. Standing increases postural demands and utilization of large bilateral muscle groups, contributing to an increased aerobic response. The effects of using large UE muscle groups for gross motor RTP tasks and the added balance challenge of reaching overhead complement the beneficial effects of standing.

These findings can be translated to clinical practice, allowing therapists to intentionally design sessions that elicit higher aerobic response by incorporating standing and gross motor tasks with targets at or above shoulder height. Modifying RTP activities with these characteristics will facilitate efficient poststroke rehabilitation that simultaneously targets improved motor function and elicits a robust cardiopulmonary response.

This study was a secondary analysis of a larger randomized clinical trial that used a subset of the participants. The primary aim of this study was not to elicit an aerobic response during RTP. In future studies, researchers should examine the effects of fusing aerobic activity with traditional therapy on stroke outcomes using a larger sample size.

The results of this study have the following implication for occupational therapy practice:

• ▪ Training in standing, with targets at or above shoulder height and with an emphasis on incorporating gross motor tasks, will maximize aerobic response to poststroke upper extremity repetitive task practice.

Monitoring heart rate response during stroke rehabilitation may increase awareness of training elements that provoke a higher aerobic response. To elicit a cardiovascular response during UE RTP, therapists should consider training in standing, with targets at or above shoulder height, and with an emphasis on incorporating gross motor tasks into UE functional rehabilitation.