High-intensity interval training after stroke: evidence & practice

Abstract

[Purpose]

To synthesize clinical, cognitive, safety, and mechanistic evidence on supervised high-intensity interval training (HIIT) after stroke and translate key findings into practice.

[Methods]

We conducted a narrative review of supervised HIIT interventions in adults after stroke. Electronic searches of PubMed and Web of Science identified studies published between January 1, 2014, and September 30, 2025. Eligibility criteria emphasized feasibility, safety, and neurofunctional outcomes (six-min walk distance, gait speed, peak aerobic capacity, activities/participation, and cognition). Quantitative pooling was not performed, and the findings were qualitatively synthesized. To explain biological plausibility, mechanistic and translational sources were reviewed irrespective of the year and summarized separately.

[Results]

HIIT was feasible under guideline-concordant screening and monitoring, with no serious adverse events. Consistent gains were observed in aerobic capacity, walking endurance, and usual gait speed, whereas activity/participation effects were mixed. Cognitive benefits were domain-specific and the clearest for executive functions. Mechanistic signals (e.g., brain-derived neurotrophic factor (BDNF), frontal oxygenation, and endothelial function) support biological plausibility.

[Conclusion]

Supervised HIIT appears safe and clinically useful for augmenting locomotor and aerobic outcomes after stroke and may preferentially enhance executive cognition. Implementation should complement task-specific therapy and follow standard screening and monitoring procedures. Future work should refine dose- and phase-specific protocols to maximize application to daily function.

INTRODUCTION

Post-stroke cognitive impairment (PSCI) is a common condition. Pooled estimates suggest that 39%-47% of survivors exhibit measurable cognitive deficits within the first year, underscoring cognition as a central target for rehabilitation [1]. Exercise is a promising avenue for cognitive recovery after stroke, with meta-analytical evidence indicating that structured training improves overall cognition when performed at least three times a week for 12 weeks or more [2]. In stroke rehabilitation, aerobic exercise is recommended as standard care to address low cardiorespiratory fitness and reduce secondary vascular risk [3]. Among aerobic modalities, higher-intensity paradigms, particularly high-intensity interval training (HIIT), rank as the most effective for improving peak oxygen uptake (VO₂peak), 6-min walk distance, and usual gait speed compared with lower-intensity exercise [4]. Consistent with this, clinical practice guidelines recommend task-specific walking at moderate-to-high intensities to maximize locomotor recovery [5]. Safety concerns regarding vigorous exercise after stroke persist; however, meta-analytic data show that high-intensity training improves cardiorespiratory fitness without excess adverse events when appropriately supervised [6]. More recent syntheses reinforce benefits extending to gait speed and balance alongside fitness without an increased rate of serious events [7]. Biologically, HIIT can upregulate neuroplasticity-related signals, most notably by elevating serum brain-derived neurotrophic factor (BDNF) and enhancing frontal cerebral oxygenation, providing a plausible mechanistic bridge to cognitive gains [8]. These effects align with the contemporary view that the brain retains an extended window of heightened plastic potential beyond the early post-stroke months, making intensity-dependent stimuli attractive [9]. Accordingly, HIIT has been proposed as a modality capable of promoting not only cardiovascular and functional recovery but also neuroplastic changes relevant to cognition [10]. Clinically, a multicenter randomized controlled trial (RCT) revealed that adding treadmill-based HIIT to usual care improved walking capacity and balance and, importantly, enhanced executive function (Trail Making Test-B (TMT-B)) compared with usual care alone [11]. By contrast, in a 6-month program for patients with prior cerebrovascular disease, HIIT plus moderate training achieved fitness and mood benefits comparable to moderate training alone, while global cognition measured by the Montreal Cognitive Assessment (MoCA) improved similarly across the exercise groups, suggesting domain-specific and context-dependent cognitive responses to intensity [12].

Against this background, in this review, we focus on the neurological and cognitive benefits of HIIT in stroke rehabilitation, synthesizing the current evidence on cognitive outcomes, putative mechanisms, and clinical implications.

METHODS

Ethical Considerations

In this narrative review, we synthesized evidence from previously published sources and involved no new interventions, interactions, or access to individual-level data from human participants or animals. No protected health information or other personally identifiable data were handled, and no attempts were made to re-identify individuals. Institutional review board approval and informed consent were not required in accordance with the institutional policies and widely accepted guidelines for secondary analyses of publicly available reports. Ethical approval and consent procedures for the primary studies were performed as stated by the original investigators.

Design

A single-author narrative review of clinical and mechanistic studies was conducted following the SANRA guidelines for the quality assessment of narrative reviews [13]. As no meta-analysis was performed, the Synthesis Without Meta-analysis (SWiM) framework was used to transparently report how outcome directions and effect sizes were compiled and interpreted qualitatively [14]. Study selection and data extraction were undertaken by the author; no independent second reviewer crosschecked these processes, which introduced potential selection and interpretation bias. To mitigate this risk, eligibility criteria were pre-specified, a structured data-extraction template was used, and the extracted values were verified against original reports before synthesis.

Search and Selection

A systematic literature search (PubMed and Web of Science) was conducted for articles published between January 1, 2014, and September 30, 2025, focusing on supervised HIIT or high-intensity aerobic training interventions in adults after stroke. The core search string combined stroke-related terms (“stroke,” “cerebrovascular”) with high-intensity exercise terms (“high-intensity interval training,” “HIIT,” “vigorous”) and safety and feasibility terms (“feasible,” “adverse events”). The database search was supplemented by manual screening of reference lists from key trials, guidelines, and recent reviews [13], to minimize omissions. Inclusion criteria included the following: (1) studies of adults with ischemic or hemorrhagic stroke in subacute or chronic phases; (2) interventions involving supervised HIIT or high-intensity aerobic training (including interval-based walking or stepping programs); and (3) reporting of at least one relevant outcome—feasibility metrics (e.g., adherence and drop-outs), safety outcomes (adverse events or near-misses), or neurofunctional measures (cardiorespiratory fitness, gait speed, endurance, balance, activities of daily living (ADLs), or cognitive performance) [5]. To contextualize biological plausibility, we also considered seminal mechanistic studies and reviews (in human or animal models) irrespective of the publication year, although these were analyzed separately and were not counted among the clinical outcome studies. Animal studies lacking translational context, case series with less than 10 participants, unsupervised home-based training without monitoring, non-English publications, and gray literature (theses and non-peer-reviewed sources) were excluded to maintain evidence quality and relevance [13]. Screening and selection were performed in two stages (title and abstract scanning and full-text review) by the author, using predefined criteria. Borderline cases were resolved by re-reading the full text and re-applying the eligibility rules; no formal duplicate screening was undertaken by an independent reviewer. The reasons for full-text exclusion were logged, and the flow of records through identification, screening, eligibility, and inclusion is summarized in a PRISMA-style diagram (Figure 1).

Figure 1. PRISMA 2020 flow diagram of study identification and selection for supervised post-stroke high-intensity interval training (HIIT) and other high-intensity aerobic interventions.

Mechanistic and translational evidence was identified through backward citation searching and synthesized narratively; it is not represented in this clinical flow diagram.

Appraisal and Data Handling

The risk of bias in the RCTs was evaluated using the revised Cochrane RoB 2 tool [15], which assesses domains such as randomization, deviations from intended interventions, missing data, outcome measurement, and reporting bias. Non-randomized studies were appraised using the ROBINS- I instrument [16], for potential confounding and selection biases. The systematic reviews included in the background or mechanistic insights were assessed using the AMSTAR 2 checklist [17]. We extracted key data from each study, including participant characteristics, HIIT protocol details (frequency, intensity targets, interval structure, and total program duration), and outcomes. In synthesizing the results, we tabulated the direction and magnitude of the HIIT effects on each outcome domain rather than computing pooled estimates, consistent with SWiM guidance [14]. Feasibility and safety data (e.g., session adherence rates and occurrence of adverse events) were compiled to identify any signals of risk. We cross-referenced exercise protocols against established guidelines, such as pre-participation screening criteria, vital sign thresholds, and exercise termination criteria, which were compared with the American Heart Association recommendations for stroke rehabilitation [18]. HIIT intensity and dosing were interpreted in the context of known locomotor training guidelines (e.g., the Locomotor Training clinical practice guideline)5 and a recent European Stroke Organization rehabilitation guideline [19] to judge whether the included interventions reached the recommended thresholds for intensity and volume. Based on these sources and on the prescriptions used in the included stroke HIIT trials and broader HIIT meta-analyses [11,12,20,21], high intensity was operationally defined in this review as supervised, interval-based aerobic training in which the work bouts were targeted to at least approximately ≥ 85% of maximal heart rate (HRmax), ≥ 60% of heart-rate reserve (HRR), or a Borg rating of perceived exertion (RPE) of 15-17 for most participants. Because individual trials varied somewhat in how intensity was monitored (e.g., HRmax vs. HRR vs. RPE) and in the exact target ranges, we classified protocols that met one or more of these thresholds as HIIT, while acknowledging that this variability may limit direct comparability between studies and contribute to heterogeneity in the observed effects [18-20]. Reporting of exercise interventions was checked against the Consensus on Exercise Reporting Template (CERT) to ensure the clarity of training descriptions [22]. All data were synthesized in a narrative format, with an emphasis on consistent findings across studies and notable discrepancies.

RESULTS

Evidence from HIIT Studies on Cognitive and Neurological Outcomes after Stroke

Across these trials and recent meta-analyses, representative effect sizes are consistent with the qualitative impression of capacity gains, and the key HIIT trials and meta-analyses underpinning these estimates are summarized in Table 1. HIIT typically increases VO₂peak by approximately 2-4 mL·kg⁻¹·min⁻¹ relative to those of lower-intensity or usual-care comparators [6,7], improves self-selected gait speed by approximately 0.05-0.15 m/s [23-25], and increases 6-min walk distance by roughly 20-90 m, with many higher-intensity programs exceeding a commonly cited minimal clinically important difference of around 34-70 m in people with slower baseline walking speeds [11,23-26]. Executive function gains on the Trail Making Test Part B (TMT-B), on the order of 20-25 s faster than those of the control group, have also been reported [11], whereas global screening scores such as the MoCA typically improve by only 1-2 points, with little difference between exercise intensities [12].

Table 1.

Selected studies reporting HIIT effects on cognitive and neurological outcomes after stroke

Study (Year)	Participants & HIIT Protocol	Cognitive/Neurological Outcomes	Key Findings
Askim et al. (2014)	n = 10, 3-9 mo post-stroke; 4 × 4 min intervals on treadmill (85% HRmax 3×/week, 4 weeks).	Feasibility; VO2peak; mobility measures.	HIIT feasible in subacute stroke; ↑ VO2peak (11%) with no adverse events.
Boyne et al. (2016)	n = 20, chronic stroke; 30-min HIIT (interval walking) vs moderate continuous training, 3×/week, 4 weeks.	6-Min Walk Test (6MWT); gait speed; adverse events.	No serious adverse events; HIIT group achieved higher intensity and showed non-significant trend of ↑ 6MWT distance vs moderate group.
Gjellesvik et al. (2021)	n = 70, chronic stroke; 8-week supervised treadmill HIIT (4 × 4 min 85–95% HRmax) + standard rehab vs rehab alone	6MWT; Berg Balance Scale (BBS); Trail Making Test-B (TMT-B); Functional Independence Measure (FIM).	HIIT group: +28 m on 6MWT, +1.3 points BBS vs control; TMT-B 24 s faster (executive function improvement). TMT-B gains maintained at 12 mo. No excess adverse events.
Amanzonwé et al. (2024)	n = 30, early subacute stroke (< 2 mo); cycle ergometer HIIT (10 × 1 min intervals 85% HRR) vs usual care, 3×/week, 6 weeks.	Motor Function (FM scale); Functional Ambulation Category; Barthel Index.	Feasible in subacute setting (high adherence). HIIT group showed faster gait attainment and higher functional independence gains than control.
Lapointe et al. (2023)	n = 52, stroke or transient ischaemic attack (TIA) (mean 2.5 yrs post); 6-month program: HIIT + MICT combined vs MICT-only vs control, 4×/week.	VO2peak; MoCA (global cognition); HADS (mood); physical activity level.	Both exercise groups ↑ VO2peak 3 mL/kg/min (vs decline in controls); HADS depression/anxiety scores ↓ similarly in HIIT+MICT and MICT. MoCA improved 1-2 points in both exercise groups with no group difference.
Krawczyk et al. (2019)	n = 48, recent lacunar stroke; home-based HIIT walking (intervals at RPE 15-17) vs usual care, 12 weeks.	Recruitment rate; adherence; 6MWT; blood pressure response.	HIIT at home feasible (95% sessions completed with patientchosen modality). No hypotensive or hypertensive events; similar ↑6MWT in both groups but greater selfreported vigor in HIIT group.
Moore et al. (2020)	n = 50, inpatient rehab (acute stroke); high-intensity stepping training integrated into therapy vs standard therapy (self-selected intensities).	Gait speed (10 m walk); walking distance; balance.	High-intensity group: significantly ↑ gait speed (+0.16 m/s) and ↑ 6MWT distance (+68 m) at discharge vs control. Emphasizing intense task practice improved locomotor outcomes.
Henderson et al. (2022)	n = 30, inpatient rehab; ↑ stepping dosage (additional high-intensity step training) vs standard care.	10 m walk speed; 6MWT; transfer and stair performance.	Greater gains in locomotor outcomes (10 m and 6MWT improvements 1.5–2× minimal clinical difference) with high-intensity protocol. Also improved transfers and stair-climbing (non-locomotor functional tasks).
Ojeda-Manzano et al. (2025)	Systematic review of 10 studies (n = 677) on HIT (60–84% HRR) vs lower intensity in subacute stroke (≤ 6 mo).	Walking speed; walking endurance (6MWT).	HIT led to superior outcomes in selfselected gait speed (significant in 4 of 10 studies) and ↑6MWT (average+60–197 m) versus comparison interventions. Benefits seen immediately and at follow-up.
Baricich et al. (2024)	Systematic review and metaanalysis (13 RCTs, n = 609) of highintensity exercise in stroke rehab.	VO2peak; gait speed; balance (BBS); adverse events.	HIIT significantly ↑ VO2peak (+2.7 mL/kg/min), ↑ gait speed (+0.10 m/s) and ↑ BBS (+1.7 points) versus controls. No increase in serious adverse events; supports efficacy of vigorous training for fitness and function.

Note. Clinical synthesis includes stroke HIIT studies published 2014-2025; HIIT was defined as reported in each study (e.g., intervals at or above 85% maximal heart rate, ≥ 60% heart-rate reserve, or Borg RPE 15-17). Effects are summarized as directionality and exemplar magnitudes from the original trials (no re-analysis). n denotes sample size, and post-stroke phase (e.g., 3-9 months, early subacute, chronic) and supervised program duration (weeks) are indicated in the Participants & HIIT Protocol column. Adverse events refer to serious events unless otherwise noted. 6MWT = 6-Minute Walk Test; BBS = Berg Balance Scale; TMT-B = Trail Making Test, Part B; FIM = Functional Independence Measure; MoCA = Montreal Cognitive Assessment; HADS = Hospital Anxiety and Depression Scale; HRmax = maximal heart rate; HRR = heart-rate reserve; VO₂peak = peak oxygen uptake; RPE = Rating of Perceived Exertion; MCID = minimal clinically important difference.

Over the last decade, post-stroke HIIT research has progressed from small feasibility experiments to multicenter RCTs, broadening the focus from aerobic capacity to cognition and neural health [7]. Early studies established that supervised HIIT is feasible and safe for selected survivors at 3-9 months post-stroke, with meaningful gains in walking capacity and no serious adverse events [27]. Ambulatory individuals in the chronic phase also tolerated HIIT well and were trained at higher intensities than those during moderate continuous exercise, with signals for larger endurance improvements [28]. Its feasibility has been shown in recent lacunar stroke cases using a home-based interval-walking model with high adherence and stable hemodynamics, extending the use case beyond clinics [29]. A multicenter RCT demonstrated that adding 8 weeks of supervised treadmill-based HIIT to u sual care improved walking distance and balance and, critically, accelerated TMT-B performance, consistent with enhanced executive function [11]. This cognitive advantage is clinically credible considering the extended window of neuroplastic potential after stroke, which suggests that well-timed, intensity-dependent stimuli can yield durable benefits [9]. In the early subacute period (≤ 2 months), recumbent-cycle HIIT was also feasible and produced larger gains in mobility and functional independence than standard rehabilitation in a low-resource setting, supporting safe introduction soon after stroke [30]. Not all trials have shown global cognitive superiority with HIIT, and this nuance matters for interpretation and program design [31]. In a 6-month randomized study of participants with prior stroke or transient ischemic attack (TIA), HIIT combined with moderate training improved fitness and mood to a degree similar to that achieved with moderate training alone, whereas MoCA gains were modest and comparable between the exercise groups, indicating domain-specific rather than universal cognitive effects [12]. Taken together, the available trials imply that HIIT preferentially benefits executive functions (e.g., set-shifting and processing speed) over broad screening scores unless cognition is explicitly targeted within the program [11]. Regarding functional outcomes, dose-response findings underscore that vigorous walking sustained for a sufficient duration produces larger immediate gains in walking capacity than those produced by moderate training and helps define practical dosing for clinics [23]. In inpatient rehabilitation, integrating high-intensity stepping into standard therapy yields superior improvements in 10-m walk speed, 6-min walk distance, and even non-locomotor tasks (e.g., transfers and stairs), reinforcing the practicality of intensity at scale [24]. A focused synthesis in patients with subacute stroke further concluded that high-intensity protocols outperformed those of conventional therapy in improving gait speed and endurance, with benefits often maintained at follow-up [25]. These clinical patterns align with modern locomotor practice statements advocating task-specific walking at high intensities to maximize recovery, provided that screening and monitoring are in place [8]. At the mechanistic level, an RCT showed that HIIT can increase serum BDNF and enhance frontal cerebral oxygenation, providing plausible biological correlates that may partly underpin but do not prove the executive function gains observed [6]. Meta-analytic work consistently indicates that higher-intensity paradigms produce greater VO₂peak, faster gait speed, and modestly higher Berg Balance Scale (BBS) scores than those produced by lower-intensity paradigms without serious adverse events when appropriately supervised [7]. In summary, the weight of evidence supports HIIT as a cardiovascular and cognition-adjacent intervention; it reliably improves endurance and mobility, selectively enhances executive domains, and is biologically primed to interact with post-stroke plasticity, particularly when delivered early enough, for a sufficient duration, and with safety protocols that match patient risk [11].

Mechanisms Underlying HIIT-Induced Neuroplasticity and Cognitive Improvement

HIIT is hypothesized to support post-stroke brain recovery through an interlocking set of mechanisms, including neurotrophic signaling, cerebrovascular adaptation, metabolic triggers, inflammation control, cardiometabolic improvements, and task-driven learning; these are the same pathway families summarized in Table 2.

Table 2.

Proposed mechanisms by which HIIT may enhance neuroplasticity and cognitive recovery after stroke, with supporting evidence

Mechanism	Description and Rationale	Supporting Evidence	Study Reference
Brain-Derived Neurotrophic Factor	HIIT triggers greater release of BDNF, a neurotrophin that supports neuron growth, synaptic plasticity, and learning. Higher BDNF may facilitate cognitive improvements and neural repair.	HIIT RCT showed serum BDNF post-training vs moderate exercise. HIIT participants’ serum promoted neurite outgrowth in vitro. Acute intense exercise raises BDNF more than moderate activity.	Hsu et al., 2021; Müller et al., 2020; Schiffer et al., 2011
Cerebral Perfusion & Oxygenation	Repeated high-intensity intervals improve cerebrovascular function, increasing blood flow and oxygen delivery to the brain. This supports metabolic demands of neural tissue and may stimulate angiogenesis.	Stroke RCT showed greater frontal oxygen extraction/total hemoglobin after HIIT vs MICT; vigorous training also improves endothelial function, plausibly supporting cerebrovascular adaptation.	Hsu et al., 2021; Ramos et al., 2015
Neurogenesis & Synaptic Plasticity	Intensive aerobic exercise may activate cellular pathways that promote the birth of new neurons and formation of synapses, especially in the subacute period of heightened plasticity.	Animal studies: post-stroke HIIT upregulated neuroplasticity markers (e.g. synapsin, BDNF) and increased hippocampal neurogenesis when initiated early. Suggests HIIT creates a pro-plastic environment for brain reorganization.	Muller et al., 2020
Lactate Surge and Growth Factor Signaling	HIIT induces high lactate levels, which can cross into the brain and act as a signaling molecule to upregulate growth factors (like BDNF, IGF-1) and enhance neural metabolism.	High lactate from intense exercise is correlated with BDNF elevation and improved motor learning in humans. Proposed as a trigger for exercise-driven brain plasticity (the “lactate shuttle” hypothesis).	Schiffer et al., 2011
Inflammation & Oxidative Stress	Vigorous exercise training reduces chronic inflammation and oxidative damage, which contribute to cognitive impairment post-stroke. Lower systemic inflammation may improve neuronal survival and recovery.	HIIT is linked to reduced pro-inflammatory cytokines and favorable shifts in immune cell profiles in chronic conditions. By mitigating inflammation, HIIT could protect the recovering brain from secondary injury.	Oberlin et al., 2017; Wang et al., 2025
Cardiometabolic Health	HIIT produces greater improvements in blood pressure, glucose control, and lipid profile than moderate exercise. Better vascular health lowers the risk of further cerebrovascular damage and cognitive decline.	HIIT improves hypertension, insulin sensitivity, and cholesterol more efficiently. Stroke guidelines emphasize risk factor control to prevent recurrent strokes and vascular cognitive decline. These systemic benefits of HIIT indirectly support long-term brain health.	Ramos et al., 2015; Jelleyman et al., 2015; Baricich et al., 2024
Enhanced Motor Learning & Engagement	The intermittent, challenging nature of HIIT (e.g. rapid walking bursts) may engage higher-level motor planning, attention, and coordination, potentially driving greater cortical reorganization than steady exercise.	Stroke locomotor trials show greater functional gains with high-intensity, variable stepping practice. The cognitive effort and novelty in HIIT sessions (vs monotonous exercise) might stimulate additional neural circuits (the exact contribution of this remains under study).	Moncion et al., 2024; MacKay-Lyons et al., 2020

Note. Mechanistic/translational sources were included irrespective of year to explain biological plausibility and were not counted in the clinical synthesis (2014–2025). Pre-2014 items are foundational exemplars retrieved via backward citation chaining. Human and animal/physiology studies are presented as exemplars rather than exhaustive listings; proposed links do not imply causality. BDNF = Brain-Derived Neurotrophic Factor; IGF-1 = Insulin-Like Growth Factor-1; VO₂peak = peak oxygen uptake; HRmax = maximal heart rate; HRR = heart-rate reserve; RPE = Rating of Perceived Exertion.

Evidence from human RCTs revealed that HIIT is associated with larger increases in serum BDNF than those associated with moderate continuous training and simultaneously augments aerobic capacity, cardiac output, and cerebral oxygenation indices, indicating that higher intensity is linked to plasticity-relevant biological changes [8]. Serum obtained after HIIT has also been shown to promote neurite growth in vitro, providing biologically plausible but indirect links between training dose and neuronal remodeling [8]. Convergent translational reviews in stroke reinforce this picture, highlighting HIIT as a candidate for functional recovery, cardiovascular health, and neuroplasticity when appropriately supervised [10]. Beyond neurotrophins, repeated high-intensity bouts improve vascular endothelial function more than moderate-intensity programs do, a systemic adaptation with logical spillover to the brain microcirculation [20]. Consistent with this, fNIRS data from patients with stroke show greater frontal oxygen extraction and total hemoglobin after HIIT, implying better cerebral oxygen delivery and utilization during effort [8]. Timing also matters. The window for heightened plastic responsiveness persists well beyond the first few months post-stroke, strengthening the rationale for introducing sufficient intensity once safety screening is satisfied [9]. Metabolically, the lactate surge characteristic of HIIT acts as a neuromodulatory signal. In humans, lactate infusion alone raises circulating BDNF, suggesting a mechanistic lactate-to-BDNF axis that high exercise intensities preferentially engage [32]. Contemporary mechanistic syntheses extend this link by arguing that the lactate-driven cascades interface with the BDNF/insulin-like growth factor-1 (IGF-1) pathways to facilitate synaptic and cellular plasticity [33]. Intensity shapes acute neurotrophic and neurophysiological responses after stroke, supporting the idea that brief, hard intervals can “prime” the brain for subsequent rehabilitation [34]. Regular exercise exerts anti-inflammatory effects at the whole-body level, a recognized route through which training can protect neural tissues from secondary injury and support cognitive trajectories [35]. HIIT also improves glycemic control and insulin sensitivity across clinical populations, thereby addressing small-vessel and metabolic risks that otherwise erode brain health [21]. Finally, the intermittent, demanding nature of HIIT requires attention, planning, and coordination, dovetailing with inpatient evidence that higher-intensity, variable stepping practice yields larger gains in locomotor and non-locomotor functions—an activity-dependent learning effect with clear neurological plausibility [25]. Taken together, these strands suggest a proplastic milieu in which neurotrophic upregulation, improved cerebral hemodynamics, metabolic signaling, and cognitively engaging task practice may work in concert, consistent with the integrated logic mapped in Table 2, although the current mechanistic evidence remains largely correlational rather than causal.

Clinical Effects of HIIT on Cognitive and Neurofunctional Outcomes in Stroke Survivors

The ultimate question for clinicians is how HIIT translates into real-world functional improvements and cognitive benefits for survivors of stroke. Table 3 summarizes the effects of HIIT on cognitive and neurofunctional outcomes reported in various studies. As discussed above, one consistent cognitive finding is improved executive function (mental flexibility and task-switching speed) following HIIT. In a trial conducted by Gjellesvik et al., the HIIT group improved their TMT-B time by 24 s more than that of the control group, representing a meaningful enhancement in executive processing speed [11]. This pattern is consistent with the hypothesis that HIIT may benefit frontal lobe-mediated functions, potentially via increased frontal perfusion and BDNF levels, although such mechanistic links have not yet been confirmed at the individual level [8]. However, global cognitive status measured by tools such as the MoCA has not shown significant improvement with HIIT beyond what is achieved with standard rehabilitation [12]. In a 6-month study in which both groups received extensive aerobic exercise, MoCA scores increased only modestly (1-2 points), and the inclusion of interval training did not confer an extra boost [12]. This finding could indicate a ceiling effect or suggest that more targeted cognitive training is needed to achieve global cognitive gains. Importantly, no study to date has reported worsening of cognitive performance with HIIT. At a minimum, intensive exercise appears to be cognitively safe poststroke, even for those with mild deficits at baseline.

Table 3.

Summary of clinical effects of HIIT on key cognitive and neurofunctional outcomes in stroke survivors (comparisons are HIIT vs baseline or vs control conditions as reported)

Outcome Measure	Effect of HIIT	Study Reference
Executive Function (Trail Making Test Part B)	Improved after HIIT. Example: TMT-B completion time 24 seconds faster with HIIT vs control at post-test (indicating better set-shifting and mental flexibility). Benefit sustained at 12 months in one RCT.	Gjellesvik 2021
Global Cognition (Montreal Cognitive Assessment)	No significant additional improvement with HIIT compared to moderate training in chronic phase. MoCA scores increased modestly (+1–2 points) in both groups over long-term rehab, with no between-group difference.	Lapointe 2023
Depression/Anxiety (HADS)	Reduced with exercise (any intensity). Depression and anxiety scores improved significantly from baseline in HIIT programs, but no difference versus moderate-intensity exercise – both yielded similar mood benefits.	Lapointe 2023
6-Minute Walk Distance (Aerobic Endurance)	Substantial increase with HIIT. Typical gains of +50–100 m after HIIT interventions in subacute/chronic stroke. Often significantly greater improvement than control/ moderate training (+28 m vs control in one study; HIIT gains frequently exceed minimal clinically important difference of 34 m).	Gjellesvik 2021; Ojeda-Manzano 2025
Gait Speed (10-Meter Walk Test)	Faster walking speed with HIIT. High-intensity gait training improves 10 m walk speed by 0.1–0.2 m/s, outperforming lower intensity training. Clinically important gains (≥0.1 m/s) achieved in HIIT groups in multiple studies, enhancing community ambulation potential.	Henderson 2022; Baricich 2024
Balance (Berg Balance Scale)	Improved modestly. Interval training led to small but significant BBS score increases (e.g. +1 to +2 points) relative to control in some trials. Enhanced balance reflects better postural control, possibly from high-intensity dynamic exercises.	Gjellesvik 2021; Baricich 2024
Functional Independence (FIM or Barthel Index)	No clear additional gain from HIIT beyond standard rehab. Overall ADL independence improved in all groups with rehabilitation. In one study, control (no HIIT) had slightly higher FIM at 1-year follow-up, suggesting task-specific training is crucial alongside aerobic exercise.	Gjellesvik 2021
Fatigue (Perceived exertion, endurance)	Improved endurance capacity with HIIT tends to reduce activity-related fatigue. Participants report higher energy and ability to sustain activity post-training. No increase in undue fatigue was noted with HIIT in trials (when properly progressed).	Boyne 2016
Quality of Life (Stroke Impact Scale, etc.)	Improved in physical domains. Stroke Impact Scale physical sub-scores (strength, mobility) improved with HIIT, reflecting better functional status. Cognitive and social domains of QoL showed less change unless targeted specifically.	Ashcroft 2025 - qualitative
Falls or Balance Confidence	Tended to improve. Although fall rates during training were low and similar between groups, HIIT participants often gained confidence in walking and balance (anecdotally reported). Formal balance confidence scales (Activities-specific Balance Confidence) have shown trends toward improvement with vigorous training.	Amanzonwé 2024

Note. Outcomes summarize typical directions and ranges reported in included clinical studies (2014–2025); comparators are baseline or control as specified in each source. Values are study-reported (not meta-analytic estimates). Where helpful, interpretation references common MCID thresholds (e.g., 6MWT +34 m; gait speed ≥0.10 m/s). Lower TMT-B times indicate better performance. 6MWT = 6-Minute Walk Test; TMT-B = Trail Making Test, Part B; MoCA = Montreal Cognitive Assessment; HADS = Hospital Anxiety and Depression Scale; BBS = Berg Balance Scale; FIM = Functional Independence Measure; VO₂peak = peak oxygen uptake; HRmax = maximal heart rate; HRR = heart-rate reserve; RPE = Rating of Perceived Exertion; MCID = minimal clinically important difference.

HIIT yields benefits comparable to those of moderate-intensity exercise in terms of mood and psychological health. Stroke survivors in exercise programs often experience reductions in depression and anxiety symptoms, attributed to the neurochemical effects of exercise (e.g., endorphins and BDNF) and improved self-efficacy. Lapointe et al. observed significant decreases in Hospital Anxiety and Depression Scale (HADS) scores over 6 months in both the HIIT + moderate-intensity continuous training (MICT) and MICTalone groups, with no difference between them [12]. Thus, engaging in moderate- or high-intensity aerobic training can ameliorate post-stroke mood disturbances, which are a critical component of cognitive health. Some patients may find HIIT more enjoyable or motivating because of its time-efficient and varied nature, which can enhance their overall psychological well-being and adherence [10]. A qualitative study by Ashcroft et al. (2025) noted that many stroke survivors and therapists had positive perceptions of HIIT, valuing the challenge and viewing it as a way to “push limits” in a controlled environment [36]. These psychosocial factors, which are difficult to quantify, contribute to cognitive recovery by increasing patient engagement in rehabilitation.

The most pronounced effects of HIIT are seen in neurofunctional outcomes such as mobility and endurance. Nearly every HIIT trial in stroke survivors has reported superior improvements in walking ability, an outcome tightly linked to neurological recovery. As shown in Table 3, 6-min walk test (6MWT) distances tend to increase substantially after HIIT interventions. For example, Moore et al. (2020) documented an average gain of +68 m in the 6MWT in the high-intensity group versus +30 m in the conventional therapy group [24]. Similarly, Henderson et al. (2022) found that patients receiving high-intensity stepping training achieved clinically meaningful gains in gait speed (0.16-0.20 m/s) compared with minimal changes in the standard care group [25]. Improvements of this magnitude can differentiate between household and community ambulation, highlighting how high-intensity rehabilitation directly enhances functional independence. Balance is another domain with notable improvements. Interval training that incorporates dynamic exercises (treadmill inclines and rapid stepping) can yield small but significant increases in balance scores (e.g., BBS) relative to those with usual care [11]. Better balance and gait translate into greater confidence and reduced fall risk, secondary benefits that further reinforce activity levels and cognitive engagement.

Some outcomes showed no differential effects of HIIT. ADLs and overall disability indices (such as the Functional Independence Measure (FIM)) improved similarly between the HIIT and control groups. In fact, in a study by Gjellesvik, the FIM scores at 1 year were slightly higher in the control group than in the HIIT group [11]. The difference was small (2 points) but statistically significant, possibly because the control group may have dedicated more therapy time to task-specific ADL practice, whereas the HIIT group spent more time on aerobic exercise [11]. This highlights an important point: HIIT should be viewed as an adjunct to task-oriented stroke therapy rather than as a replacement. Intensive aerobic training can boost fitness and certain cognitive capacities, creating a readiness to benefit from rehabilitation; however, patients still require focused practice of ADLs, speech, or other skills to translate gains into daily function. In combined programs, careful balance is required to ensure that aerobic work does not displace excessive functional training.

In summary, the clinical profile of HIIT in patients with stroke is characterized by complementary benefits faster walking, improved endurance, and better executive functioning based on a foundation of enhanced cardiovascular health. Table 3 provides a quick reference to the range of outcomes reported. It becomes evident that while HIIT excels at improving capacity (what a person can do under testing conditions, such as walking farther or thinking faster), translation to performance in daily life (such as independent living skills) may require integration with conventional therapies. Ongoing trials are investigating whether combining HIIT with cognitive training or task-specific motor practice can amplify real-world benefits [10, 37]. Notably, a recent clinical guideline on stroke locomotor training strongly recommends high-intensity walking training to improve gait, reflecting accumulated evidence [5]. As research continues, we anticipate more nuanced insights into optimizing HIIT protocols (interval length, frequency, and modality) for different phases of stroke and targeting specific outcomes (e.g., memory versus mobility). For now, the evidence supports that “pushing the intensity” in aerobic exercise is a safe and efficacious strategy to maximize neurological recovery after stroke when applied judiciously.

DISCUSSION

Clinical Implications

Across studies, higher-intensity aerobic paradigms in stroke rehabilitation reliably increase cardiorespiratory fitness and translate to faster gait and greater walking capacity, positioning intensity as a practical lever for mobility recovery in the clinic [38]. Definitions of high intensity varied across studies, with some protocols prescribing 4 × 4-min treadmill intervals at 85%-95% of peak heart rate, others targeting 60%-84% of HRR on cycle ergometers, and home-based programs relying on Borg RPE of 15-17. This heterogeneity in intensity targets, modalities, and total program duration limits direct comparability between trials. Consequently, conclusions should be interpreted as applying to supervised vigorous interval paradigms rather than a single standardized HIIT prescription, which may partly explain the variability in reported clinical and cognitive effects.

Cognitive outcomes appear to be more domain-specific than global; meta-analytic syntheses show small-to-moderate improvements most consistently in attention and processing speed, whereas omnibus screen scores (e.g., global cognition) may shift only modestly, without clear differences between programs [39]. A recent focused meta-analysis restricted to aerobic exercise after stroke similarly reported beneficial effects on cognition but suggested that program structure and dosing, rather than intensity alone, likely determine which cognitive domains change [40]. Timing and dose matter: meta-regression indicates that initiating structured exercise within the first 3-6 months yields clinically meaningful mobility gains, supporting earlier but still screened and supervised uptake at higher intensities [41]. Equally, the AVERT program cautions that very-early high-dose mobilization (< 24 h) can worsen outcomes, underscoring that “early enough” is not the same as initiating mobilization as early as possible [42]. Beyond generic treadmill exposure, the delivery of intensity and variability seems to amplify effects; protocols that explicitly raise stepping intensity and variability produce larger gains in walking ability and symmetry than lower-intensity comparators [43]. Safety signals remain reassuring when programs are screened and monitored. Cochrane analyses of fitness training and treadmill-based paradigms report improvements in fitness, balance, and endurance without excess serious adverse events, supporting the implementation of higher-intensity walking in practice [37]. At the same time, broader HIIT physiology points to adaptations that are stroke-agnostic but mechanistically relevant, such as improved endothelial function, mitochondrial remodeling, and autonomic balance, which likely contribute to the observed capacity gains after stroke [44]. Mood and participation outcomes reinforce the clinical picture. Meta-analyses indicate that exercise training reduces depressive symptoms post-stroke, and higher physical-activity levels are positively associated with participation in daily life, supporting intensity-focused programs as part of holistic recovery [45]. To convert capacity into real-world activity, pairing vigorous walking with behavioral strategies that increase steps/day (e.g., goal setting and feedback) shows promise and may help maintain gains beyond supervised sessions [46].

Future Directions and Methodological Priorities

Several methodological priorities emerge from this synthesis. First, trials should report intervention fidelity using exercise-specific reporting frameworks (CERT) alongside TIDieR so that intensity targets, interval structure, supervision, monitoring, and adherence can be replicated and audited [22]. Second, heterogeneity in responses warrants a priori stratification and responder profiling using impairment, corticospinal integrity, and imaging/physiological markers to individualize intensity prescriptions and interpret domain-specific cognitive effects [47]. Third, guideline-concordant integration with vascular risk management is essential because secondary prevention (BP, lipid levels, and glycemia) interacts with training benefits and shapes cognitive trajectories after stroke [48]. Finally, key research gaps include dose optimization (interval length, work-rest ratio, and total weeks), phase-specific protocols (early subacute vs. chronic), and hybrid designs that embed cognitive dual-tasking within high-intensity walking to target executive functions without displacing task-specific practices [49]. Taken together, current evidence indicates that supervised HIIT after stroke appears feasible and safe when participants are risk-stratified, appropriately screened, and continuously monitored. Compared with lower-intensity comparators, HIIT tends to yield larger gains in aerobic capacity, walking endurance, and gait speed, whereas cognitive benefits are most consistently observed in executive functions (e.g., set-shifting and processing speed), with more modest changes in global screening scores. Emerging mechanistic signals, including increased BDNF, enhanced frontal and cerebral oxygenation, and improved endothelial function, provide biological plausibility for these clinical patterns but should still be interpreted as correlational rather than causal. In practice, HIIT should complement rather than replace task-specific therapy; its implementation is best guided by individualized risk management, phase-appropriate dosing and progression, and rigorous adherence and safety monitoring. Looking ahead, priorities include defining dose-response parameters (interval length, work-rest ratio, and program duration), developing phase-specific protocols across the stroke continuum, standardizing safety and outcome reporting, and testing hybrid designs (e.g., pairing HIIT with cognitive or dual-task training) to maximize transfer to everyday function.