Physical Fitness Training in Patients with Subacute Stroke (PHYS-STROKE): multicentre, randomised controlled, endpoint blinded trialBMJ 2019; 366 doi: https://doi.org/10.1136/bmj.l5101 (Published 18 September 2019) Cite this as: BMJ 2019;366:l5101
All rapid responses
Re: Physical Fitness Training in Patients with Subacute Stroke (PHYS-STROKE): multicentre, randomised controlled, endpoint blinded trial
We thank Peters and colleagues for commenting on the previously published results of the PHYS-STROKE trial . Particularly, we highly appreciate further scientific interpretation of the study results and we endorse the approach to initiate an open discussion to derive important clinical implications.
We follow the idea that not all moderately affected stroke survivors performing physical fitness training will be at increased risk of serious adverse events (SAE) and, indeed, several smaller studies reported safe and promising results [2–4]. However, small studies are at high risk of bias  and the assessment of efficacy and safety needs to take place in larger clinical trials. Our new data suggest that aerobic fitness training should not be generally recommended to moderately or severely affected patients with subacute stroke.
The PHYS-STROKE trial filled a gap in previously available evidence in current stroke rehabilitation research. However, the study results will not provide definite answers to all questions in aerobic exercise training after acute stroke, but rather help us now to address further issues surrounding this intervention. In fact, as Peters et al. correctly point out, several novel questions have arisen from the study. One size does not fit all, which is particularly true in stroke rehabilitation. Individual clinical assessment of stroke patients and trying to determine who will, and who will not benefit, remains the decisive question.
The question now is which stroke patients may safely receive early aerobic fitness and who benefits more from a start at a later stage? We fully agree with Peters et al. that we need to address these questions, e.g. what is the role of blood pressure, atrial fibrillation, post-stroke walking ability, or secondary prevention regarding predictive power for SAEs? We will address some of these issues, including follow-up data after six-months, in a dedicated safety report of PHYS-STROKE currently in preparation.
The PHYS-STROKE cohort at large was not capable of performing aerobic fitness without the use of a body-weight supported training (BWST) due to the severity of strokes and the early time of enrolment. We agree that BWST puts limited stress on postural control, which might hamper treatment success and possibly cause more subsequent falls. However, superiority of overground walking compared to BWST in early stroke rehabilitation still needs to be proven.
Only four patients (4%) in the PHYS group and three patient (3%) in the relax group were treated with dual antiplatelet regimen at baseline. According to the AHA guideline recommendations of 2019, dual antiplatelet therapy is recommended for TIA and minor stroke (NIHSS <4) within the first 21 days after stroke . Patients enrolled in PHYS-STROKE do not necessarily fall into this category (median NIHSS = 8, median time of enrolment = 28 days after stroke). Besides, at time of enrolment (2013-2017), dual antiplatelet therapy was not recommended in respective guidelines.
Regarding cardio-embolic stroke, 18 (20%) patients randomized to the PHYS group had diagnosed cardio-embolic stroke and all were treated with anticoagulation. We agree that single ECG could have missed episodes of atrial fibrillation and prolonged ECG monitoring prior to enrolment may increase detection of atrial fibrillation and hence reduce the number of cryptogenic stroke patients. In fact, atrial fibrillation was diagnosed in three patients during follow-up (all in the PHYS group). Further analyses will address the association of cardiac comorbidities including atrial fibrillation and SAE in PHYS-STROKE.
It is correct that the difference in heart rate was calculated after the cool down phase of patients in PHYS-STROKE. We agree that a continuous reporting of heart rate should be performed in future trials and we acknowledge this limitation in our trial. We also support the other suggested directions for future aerobic training studies. Moreover, it is also important to acknowledge and reflect on the lack of treatment effect of aerobic fitness training compared to the control group in PHYS-STROKE. In future trials, we also need to investigate the clinical benefit of relaxation sessions in addition to an extensive in-patient standard treatment plan including physical and occupational therapy.
1 Nave AH, Rackoll T, Grittner U, et al. Physical Fitness Training in Patients with Subacute Stroke (PHYS-STROKE): multicentre, randomised controlled, endpoint blinded trial. BMJ 2019;366. doi:10.1136/bmj.l5101
2 Mackay-Lyons M. Aerobic treadmill training effectively enhances cardiovascular fitness and gait function for older persons with chronic stroke. J Physiother 2012;58:271. doi:10.1016/S1836-9553(12)70131-5
3 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;26:85–95. doi:10.1177/1545968311418675
4 Ivey FM, Ryan AS, Hafer-Macko CE, et al. Treadmill aerobic training improves glucose tolerance and indices of insulin sensitivity in disabled stroke survivors: a preliminary report. Stroke 2007;38:2752–8. doi:10.1161/STROKEAHA.107.490391
5 Button KS, Ioannidis JPA, Mokrysz C, et al. Power failure: Why small sample size undermines the reliability of neuroscience. Nat Rev Neurosci 2013;14:365–76. doi:10.1038/nrn3475
6 Powers WJ, Rabinstein AA, Ackerson T, et al. Guidelines for the Early Management of Patients With Acute Ischemic Stroke: 2019 Update to the 2018 Guidelines for the Early Management of Acute Ischemic Stroke: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke . Stroke 2019;50:344–418. doi:10.1161/STR.0000000000000211
Competing interests: No competing interests
Aerobic intensity matters: Overgeneralization may limit research in individuals post-stroke who need it most. Response to Nave et al. 2019.
Stroke rehabilitation is on the cusp of a new era. Emerging high-intensity aerobic research shows benefits post-stroke for motor priming, neuroplasticity, motor learning, cognition, and cardiovascular improvements(1). Hence, we need to take the time to reflect on the results from Nave et al. (2019) which found that a daily 20-minute bout of aerobic training in addition to usual subacute stroke rehabilitation resulted in higher rates of falls and serious adverse events (SAEs), including recurrent strokes. We commend the authors for a thorough and well done clinical trial termed PHYS-STROKE. However, their clinical implications which state “aerobic physical fitness could still be an invaluable part of stroke rehabilitation – for example, in people with chronic stroke or mildly affected with subacute stroke” effectively removes all moderately impaired stroke patients from aerobic activity. We should be cautious in further examination of aerobic intensity in patients with moderate impairment, yet, multiple fundamental questions remain.
Who gets worse with aerobic training?
A large question surrounding the PHYS-STROKE results is who experienced these SAEs? One of the unique characteristics of PHYS-STROKE is that patients were not required to have standing or stepping ability. Indeed, 10% of the sample was unable to complete a 10 m walk test at baseline. Could the severity of these patients be contributing to the adverse events? Of concern, the aerobic group had more severe strokes at baseline (median NIHSS score of 9, range 5-12) than the relaxation group (median 7, range 5-11). In contrast, other small aerobic studies (total n=159) with mild to moderate stroke with some ambulatory ability, produced gains in walking speed or endurance without SAE(2-4) when training commenced within 4 weeks post-stroke, and outcomes assessed at 3-3, 6- 4, and 12-months(2).
Did patients with SAEs have more severe impairment, worse cardiovascular status, a particular stroke type, or suboptimal antithrombotic strategies? Rates of “undetermined” ischemic stroke mechanisms (usually 20-25%(5)) were high (34%) in the overall cohort, and higher (37% vs. 31%) in the aerobic group. Was the increase in recurrent stroke in the aerobic group due to individuals with occult atrial fibrillation not being on anticoagulation? Or, were individuals in this group not prescribed dual antiplatelet therapy during the first month when the greatest risk of recurrent stroke is present(6)? Furthermore, fall rate differences could be due to demographic imbalances. There were more women, known to have higher rates of self-reported falls, than men in the aerobic group(7). Could patients with abnormal BP responses to exercise be at greater risk? Wijkman et al. (2018) reported that 66% of their sample of ambulatory subacute stroke patients had exaggerated systolic BP response with exercise(3).
The PHYS-STROKE relaxation group had fewer adverse events, and possibly lower heart rate and blood pressure post-session (Table S5). Some mindfulness-based interventions show reduction in systolic blood pressure along with other psychological benefits(8), so potentially these benefits may have extended to the PHYS-STROKE relaxation group participants.
What intensity of aerobic training results in an increased risk for vascular events?
While the authors set exercise targets, no objective data shows actual the heart rate (HR) over the session. A 15 beat/minute change in HR at the end of the session is reported (we assume this occurred after the 2 minute cool-down). Reporting the actual duration patients were able to sustain target HR during the 20-minute sessions (e.g., average HR over the session, number of minutes at 40% HR reserve) is essential information to help interpret results and permit progression of stroke rehabilitation research and clinical translation(9).
Does Body Weight Supported Treadmill (BWST) training need to be formally coupled with overground walking?
Providing the aerobic intervention on BWST may have also impacted the outcomes. A 2017 Cochrane review showed patients who are able to walk at the start of an intervention benefit the most from BWST training for increased walking speed and endurance, and any adverse events were deemed not serious (1504 participants, 125 total non-serious events(10)). A limitation of BWST is its low demand for postural control, whereas overground walking may be more effective in improving dynamic balance, walking speed and distance, and reducing falls risk.
Future directions for Aerobic Training Studies
Future trials should report multiple measures of cardiovascular function throughout training like heart rate, and blood pressure(9). Additionally, PHYS-STROKE followed patients to three months after their event; longer follow up durations are necessary to capture serious and non-serious adverse events, and to assess durability of an intervention’s effect in the context of the natural history of functional improvement. The large LEAPS trial followed patients for 1 year post-stroke where walking function plateaus(11). Ultimately, we should be cautious moving forward with aerobic training of moderate stroke patients; nevertheless, we should move.
1. Hasan SM, Rancourt SN, Austin MW, et al. Defining Optimal Aerobic Exercise Parameters to Affect Complex Motor and Cognitive Outcomes after Stroke: A Systematic Review and Synthesis. Neural Plast 2016;2016:2961573. doi: 10.1155/2016/2961573 [published Online First: 2016/02/18]
2. Mackay-Lyons M, McDonald A, Matheson J, et al. Dual effects of body-weight supported treadmill training on cardiovascular fitness and walking ability early after stroke: a randomized controlled trial. Neurorehabil Neural Repair 2013;27(7):644-53. doi: 10.1177/1545968313484809 [published Online First: 2013/04/20]
3. Wijkman MO, Sandberg K, Kleist M, et al. The exaggerated blood pressure response to exercise in the sub-acute phase after stroke is not affected by aerobic exercise. J Clin Hypertens (Greenwich) 2018;20(1):56-64. doi: 10.1111/jch.13157 [published Online First: 2018/01/18]
4. Sandberg K, Kleist M, Falk L, et al. Effects of Twice-Weekly Intense Aerobic Exercise in Early Subacute Stroke: A Randomized Controlled Trial. Arch Phys Med Rehabil 2016;97(8):1244-53. doi: 10.1016/j.apmr.2016.01.030 [published Online First: 2016/02/24]
5. Hart RG, Diener HC, Coutts SB, et al. Embolic strokes of undetermined source: the case for a new clinical construct. Lancet Neurol 2014;13(4):429-38. doi: 10.1016/s1474-4422(13)70310-7 [published Online First: 2014/03/22]
6. Wang Y, Johnston SC, Bath PM, et al. Acute dual antiplatelet therapy for minor ischaemic stroke or transient ischaemic attack. Bmj 2019;364:l895. doi: 10.1136/bmj.l895 [published Online First: 2019/03/02]
7. Gale CR, Westbury LD, Cooper C, et al. Risk factors for incident falls in older men and women: the English longitudinal study of ageing. BMC Geriatr 2018;18(1):117. doi: 10.1186/s12877-018-0806-3 [published Online First: 2018/05/18]
8. Scott-Sheldon LAJ, Gathright EC, Donahue ML, et al. Mindfulness-Based Interventions for Adults with Cardiovascular Disease: A Systematic Review and Meta-Analysis. Ann Behav Med 2019 doi: 10.1093/abm/kaz020 [published Online First: 2019/06/06]
9. Walker MF, Hoffmann TC, Brady MC, et al. Improving the development, monitoring and reporting of stroke rehabilitation research: Consensus-based core recommendations from the Stroke Recovery and Rehabilitation Roundtable. Int J Stroke 2017;12(5):472-79. doi: 10.1177/1747493017711815 [published Online First: 2017/07/13]
10. Mehrholz J, Thomas S, Elsner B. Treadmill training and body weight support for walking after stroke. Cochrane Database Syst Rev 2017;8:Cd002840. doi: 10.1002/14651858.CD002840.pub4 [published Online First: 2017/08/18]
11. Duncan PW, Sullivan KJ, Behrman AL, et al. Body-weight-supported treadmill rehabilitation after stroke. N Engl J Med 2011;364(21):2026-36. doi: 10.1056/NEJMoa1010790 [published Online First: 2011/05/27]
Competing interests: No competing interests