Poor cardiorespiratory fitness is characteristic of individuals experiencing the effects of stroke. Measurements taken during the first 3 months of recovery demonstrate that maximum oxygen uptake values range from 10 to 18 mL min^–1 kg^–1 (Mackay-Lyons and Makrides, 2004; Tang et al., 2006; Mackay-Lyons et al., 2013). Given the inverse association between cardiorespiratory fitness and future risk of stroke (Pandey et al., 2016) it is perhaps not surprising that these values correspond to the “very poor” category based on normative data (Riebe et al., 2018). However, it is striking that fitness levels among stroke survivors are less than 60% of that reported for age and sex-matched coronary heart disease patients upon completion of cardiac rehabilitation (Banks et al., 2019). Such differences in fitness could be explained by the fact that people with cerebrovascular disease are offered lower-intensity task-oriented rehabilitation compared to people having cardiovascular disease. (Hebert et al., 2016). Although relevant to overcome functional limitations, task-oriented therapies offered during stroke rehabilitation provide insufficient cardiovascular stress (MacKay-Lyons and Makrides, 2002; Barrett et al., 2018).

Important advances are taking place in the field of stroke rehabilitation to enhance the provision of aerobic training (Biasin et al., 2014); specifically transitioning eligible patients with stroke to cardiac rehabilitation (Marzolini, 2018). Feasibility and preliminary effectiveness of comprehensive cardiac rehabilitation programs has been confirmed among individuals who have experienced a transient ischemic attack or non-disabling minor stroke (Prior et al., 2011). The challenge, however, is adapting such programs for individuals with more severe stroke-related impairments (Tang et al., 2009). Indeed, stroke survivors with mild-to-moderate levels of motor impairments can meet the minimal aerobic exercise targets (Marzolini et al., 2012) and achieve statistically significant increases in pre-to-post measures of cardiorespiratory fitness upon completion of adapted cardiac rehabilitation programs (Lennon et al., 2008; Tang et al., 2010; Marzolini et al., 2012, 2014). However, cardiorespiratory fitness remains in the “very poor” category upon completion of adapted cardiac rehabilitation and the training-induced improvements in cardiometabolic risk factors have been inconsistent (Lennon et al., 2008; Tang et al., 2010; Marzolini et al., 2014). Accordingly, new approaches are needed to enhance the effects of physical exercise interventions currently offered to stroke survivors.

Normobaric hypoxia conditioning has the potential to enhance the effects of exercise in stroke survivors. Normobaric hypoxia exposure involves reductions in the fraction of inspired oxygen (F_IO₂) from values recorded near sea-level (F_IO₂ = 20.9%) to values that simulate the moderate levels of hypoxia experienced at elevations between 2,000 and 3,000 m above sea-level (F_IO₂ = 17.0 to 15.0%) (Bartsch and Saltin, 2008). During normobaric hypoxia conditioning studies, participants perform submaximal exercise while breathing the normobaric hypoxia gas mixture (Klug et al., 2018). Since maximum aerobic workloads decrease at a rate of about 7% for every 1,000 m increase in altitude (≅ 2% decrease in F_IO₂) (Fulco et al., 1998), the absolute workloads performed in normobaric hypoxia must be reduced to be equivalent to the relative exercise intensity experienced under typical sea-level conditions (Bartsch and Saltin, 2008). In other words, training in normobaric hypoxia is a method to increase the cardiometabolic stress of submaximal exercise without an increase in biomechanical strain (Pramsohler et al., 2017). Interestingly, among overweight/obese participants, exercise completed in normobaric hypoxia conditions resulted in similar (Klug et al., 2018) or even enhanced (Netzer et al., 2008; Wiesner et al., 2010) benefits on cardiometabolic outcomes compared to aerobic workloads matched for relative intensity at sea-level. Clinically, the approximately 15% increase in relative effort during submaximal exercise performed in moderate normobaric hypoxia (Fulco et al., 1998) may provide a unique solution for stroke survivors with diminished ability to perform external work due to hemiparesis. Although preliminary evidence in other clinical populations suggests a synergistic effect of pairing exercise with normobaric hypoxia exposure on metabolic (Mackenzie et al., 2011; Mahler et al., 2018) and functional outcomes (Navarrete-Opazo et al., 2017b), more work is needed to translate these findings to stroke survivors. Accordingly, the objective of the current preliminary study was to determine if exposing chronic hemiparetic stroke survivors to moderate levels of normobaric hypoxia during treadmill aerobic exercise was safe and feasible.

The current observational study was undertaken to examine the preliminary safety and feasibility of performing treadmill aerobic exercise under conditions of moderate normobaric hypoxia among chronic hemiparetic stroke survivors. Other than one participant reporting mild symptoms of nausea upon reducing F_IO₂ during exercise, no adverse events were observed during normobaric hypoxia exposure or the morning after. As anticipated, the addition of moderate normobaric hypoxia to constant workload exercise was associated with reductions in SpO₂ and corresponding increases in HR. However, the increased cardiovascular strain had little effect on BP measurements and participants did not perceive the efforts as more demanding, as indicated by RPE scores. The current study provides initial evidence that supports the use of normobaric hypoxia exposure to increase the cardiovascular demands of submaximal exercise in chronic stroke survivors with walking impairments.

Several research reports have evaluated the safety of exposure to environments with reduced oxygen availability, such as terrestrial high altitudes (i.e., hypobaric hypoxia), in older adults with preexisting cardiovascular conditions (Bartsch and Gibbs, 2007; Levine, 2015; Parati et al., 2018). It is generally accepted that individuals with asymptomatic coronary heart disease who have a sufficiently high exercise capacity (i.e., > 6 metabolic equivalents or 21 ml min^–1 kg^–1) are at minimal increased risk during high altitude trekking (Bartsch and Gibbs, 2007). However, as reflected in the current data, the majority of stroke survivors have cardiorespiratory fitness levels below this threshold (Smith et al., 2012; Saunders et al., 2020). Also, hemiparetic gait is associated with a 1.5 to twofold increased energy demand (Kramer et al., 2016). The net effect of low cardiorespiratory fitness and increased energy cost of ambulatory activities is a diminished capacity to accommodate changes in aerobic demand. With the goal of developing normobaric hypoxia conditioning protocols appropriate for stroke rehabilitation, we investigated whether chronic hemiparetic stroke survivors could tolerate treadmill walking while progressively and intermittently exposed to reductions in F_IO₂. Arterial oxygen saturation, electrocardiography, blood pressure, and rating of perceived exertion were recorded during constant-load exercise to monitor for signs of excessive cardiovascular stress. Unremarkable changes in blood pressure and electrocardiography tracings were recorded during normobaric hypoxia exposure. These outcomes are consistent with data collected in a relatively large cohort (N = 97) of older adults upon initial ascent to high altitude (Erdmann et al., 1998). As anticipated, normobaric hypoxia exposure reduced SpO₂ both at rest and during constant workload exercise. However, SpO₂ values remained above the cutoff score in all participants, including those prescribed beta blockers, which has previously been identified as a potential risk among individuals receiving antihypertensive treatment (Bilo et al., 2011). Mild symptoms of nausea were reported by one participant, which dissipated shortly after stopping the exercise protocol. As reported in the Supplementary Table 1, this participant (#5) selected a treadmill speed above the target workload and it is unclear if the symptoms of nausea were a result of normobaric hypoxia itself or the high-intensity nature of such workloads when paired with normobaric hypoxia. No symptoms of headache, nausea, or dizziness/light-headedness were reported in the remaining participants. Collectively, findings from this preliminary study support that performing submaximal exercise under conditions of moderate normobaric hypoxia provides no additional risk beyond that of exercise in typical environments among chronic hemiparetic stroke survivors.

With respect to feasibility, participants were able to sustain 20-min of constant workload exercise while intermittently exposed to moderate normobaric hypoxia. Although the treadmill speed and incline were maintained throughout AINH, average HR was 7.8 bpm higher during the five 2-min intervals of normobaric hypoxia exposure. This corresponded to a 10% increase in relative effort based on participants’ HRR. Although lower than the anticipated 15% increase in relative effort for the level of F_IO₂ studied (Fulco et al., 1998), the change in relative effort is likely underestimated given that HRR is diminished under conditions of reduced oxygen availability (Bartsch and Gibbs, 2007). It is equally important to point out that HR and SpO₂ began to recover between intervals, which suggest that participants were able accommodate to the increased cardiovascular stain of moderate normobaric hypoxia exposure during submaximal exercise. Furthermore, participants did not perceive the workloads as more demanding when performed at reduced F_IO₂. The lack of effect on RPE is consistent with the results of a previous placebo-controlled trial that reported participants were unable to accurately predict whether or not they performed submaximal exercise in normobaric hypoxia or under typical sea level conditions (Netzer et al., 2008). Accordingly, normobaric hypoxia exposure appears to be a feasible method of increasing the cardiovascular strain of submaximal exercise in chronic hemiparetic stroke survivors.

The focus of the current study was on using normobaric hypoxia exposure to increase the cardiovascular strain of submaximal exercise among chronic hemiparetic stroke survivors. However, normobaric hypoxia exposure may enhance the effects of submaximal exercise independent of changes in relative exercise intensity. For instance, in a recent randomized trial among overweight/obese females, metabolic benefits were enhanced when normobaric hypoxia was combined with 12-weeks of high-intensity interval training (Camacho-Cardenosa et al., 2018). This is consistent with normobaric hypoxia conditioning studies that report similar (Klug et al., 2018) or enhanced (Netzer et al., 2008; Wiesner et al., 2010) effects on cardiometabolic outcomes at lower absolute workloads. Furthermore, intermittent normobaric hypoxia exposure enhanced the effects of locomotor training on walking speed, endurance, and dynamic balance among individuals with incomplete spinal cord injury (Navarrete-Opazo et al., 2017b). The latter studies incorporated more severe levels of normobaric hypoxia (F_IO₂ = 9.0%) immediately prior to bodyweight supported treadmill training, which suggests a priming effect of normobaric hypoxia exposure. Although mechanisms underlying the synergistic effects of pairing normobaric hypoxia exposure with exercise are yet to be elucidated, both acute physiological responses and more prolonged acclamatory processes are likely involved (Hochachka et al., 1999). For example, during constant workload exercise performed in hypoxia there is an increased mobilization of carbohydrate and lipid energy substrates compared to exercise performed in normoxia at the same absolute and relative workloads (Peronnet et al., 2006). Despite the fact that increased mobilization of both energy substrates is characteristic of hypoxia exposure, substrate partitioning is shifted toward carbohydrate energy sources during exercise in hypoxia and lipid oxidation is enhanced during the post exercise recovery period under normoxic conditions (Kelly and Basset, 2017). Such metabolic perturbations during and in-between individual bouts of exercise may help to explain the similar or enhanced effects of normobaric hypoxia exposure on adiposity and metabolic profile among sedentary overweight adults at a lower total energy expenditure of exercise (Haufe et al., 2008; Netzer et al., 2008; Wiesner et al., 2010). Further, hypoxia-induced increases in the expression of transcriptional factors including hypoxia-inducible factor-1alpha (HIF-1 alpha) and proliferator-activated receptor-gamma coactivator-1 alpha (PGC-1 alpha) have been linked to improvements in insulin sensitivity and skeletal muscle lipid metabolism after normobaric hypoxia conditioning, respectively (Haufe et al., 2008; Wiesner et al., 2010). Normobaric hypoxia exposure also increases expression of brain derived neurotrophic factor and its receptor tyrosine kinase receptor-B (TrkB) (Dale et al., 2014), which is reasoned to explain the beneficial effects of intermittent hypoxia on ambulatory function among individuals with incomplete spinal cord injury (Navarrete-Opazo et al., 2017a). Given the benefits reported in other clinical populations and the multiple potential mechanisms underlying acute and more prolonged physiological responses, studies investigating the dose response effects of pairing normobaric hypoxia with task-oriented activity after stroke are warranted.

There are several methodological considerations that must be discussed. Firstly, the current study was conducted in a small convenience sample of chronic stroke survivors. This cohort was chosen because of their previous experience with aerobic exercise and the fact that they were able to achieve moderate intensity workloads without adverse events in a previous trial. In order to prepare for future studies, we needed to demonstrate that the additional risk associated with normobaric hypoxia exposure was proportional to its potential benefits. Secondly, we exposed participants to a narrow range of normobaric hypoxia both at rest and during a single constant workload. This dose of normobaric hypoxia exposure was chosen because of previous intervention studies conducted in sedentary populations that demonstrated favorable outcomes when submaximal aerobic exercise was paired with similar reductions in F_IO₂ (Haufe et al., 2008; Wiesner et al., 2010; Camacho-Cardenosa et al., 2018; Fernandez Menendez et al., 2018). Given the range of HR and SpO₂ responses observed and the fact that perception of effort was not influenced by normobaric hypoxia exposure, this cohort likely could have tolerated even further decreases in F_IO₂ and/or increases in absolute workload. Also, decreases in SpO₂ during exercise performed in normobaric hypoxia were highly variable between participants and individual adjustments in F_IO₂ may be required to ensure similar levels of cardiovascular and metabolic stress between participants (Lee and Thake, 2018).

The current preliminary study provides initial evidence that moderate normobaric hypoxia exposure is a safe and feasible method to increase the cardiovascular strain of submaximal exercise after stroke. Scientific evidence collected over the last three decades indicates that stroke survivors can respond positively to aerobic exercise training (Saunders et al., 2020). However, the doses of aerobic exercise studied thus far are insufficient to enhance cardiorespiratory fitness to more normal and low risk levels. Stroke-related impairments and comorbid health conditions are significant barriers to sustaining workloads associated with moderate intensity aerobic exercise in this population (Biasin et al., 2014; Moncion et al., 2020). Accordingly, normobaric hypoxia exposure may be an effective strategy to enhance the effects of aerobic exercise training on cardiorespiratory fitness, secondary prevention, and functional recovery after stroke. Future studies are needed to confirm the results of the current study and to determine the most appropriate protocols for pairing normobaric hypoxia exposure with physical exercise in large representative samples of stroke survivors at different stages of recovery.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Newfoundland and Labrador Health Research Ethics Board. The patients/participants provided their written informed consent to participate in this study.

LK, FB, JM, and MP contributed to the conception and design of the study. LK and JM collected the data. LK performed the statistical analysis and wrote the first draft of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.