The efficacy of High Intensity Interval Training (HIIT) versus aerobic exercise in the early stage of Parkinson’s Disease: A Systematic Review

Introduction

Parkinson’s Disease (PD) is a neurodegenerative condition caused by a decrease in dopaminergic neurons in the Substantial Nigra Pars Compacta (SNpc) and reduced dopamine levels in the basal nuclei.¹ The prevalence rate of PD is growing faster than any other condition² and is expected to double by 2030.³ The loss of dopamine in the Substantial Nigra (SN) may contribute to neuroinflammation, increasing oxidative stress, leading to intrinsic destruction of the neurons.⁴ These processes are interconnected; therefore, it is difficult to identify which pathological changes are responsible for the disease.⁵

The aetiology of PD evolves from a complex interaction of genetic, environmental, and ageing risk factors.⁶ PD is characterized by subtle non-motor symptoms such as sleep disturbances, mood disorders, depression, anxiety, urinary dysfunction, orthostatic hypotension, excessive daytime sleepiness, cognitive impairment, fatigue, and pain.⁷ As PD progresses, motor symptoms, may affect one side of the body, and gradually extend to the other side. These cardinal motor symptoms include resting tremors, bradykinesia, rigidity, gait dysfunction, and postural instability.⁸ The Hoehn and Yahr scale uses these symptoms to stage the severity of the disease.⁹

Pharmacological management, engagement in physical activity, and lifestyle changes are the main approaches advocated for people with PD. Anti-parkinsonism medications such as dopamine replacement and dopamine agonists are administered as first-choice treatment in PD management.¹⁰ Long-term use of anti-parkinsonism medications may lead to adverse effects (AE) such as wearing off, peak-dose dyskinesia and on-off phenomenon.¹¹ The role of neurorehabilitation as part of management of PD has significantly increased due to developments within the process of neuroplasticity. Potential neurorestorative effects of exercise in PD have been explored with promising outcomes if introduced in the early stage of the disease.¹

Exercise activates the central and peripheral nervous system and improves motor and non-motor symptoms.¹² Those have been associated with production of neuroprotective factors that optimize antioxidant mechanisms, thus slowing down the progression of PD.¹³

Aerobic training is a physical activity aimed at increasing HR and oxygen consumption. Aerobic training enhances neuroplastic changes (angiogenesis, neurogenesis, synaptogenesis),¹ strengthens synaptic force and improves neural networks.¹⁴ American Physical Therapy Association (APTA)¹⁵ suggests that moderate to high intensity aerobic exercises should be implemented to reduce motor disease severity and improve functional outcomes in people with PD in the stages 1-3 of the Hoehn and Yahr scale.

High Intensity Interval Training (HIIT) is an anaerobic type of exercise that reduces the risk of PD and slows down its progression. HIIT incorporates bouts of high-intensity/vigorous training with short rest or low-intensity periods in between.¹⁶ O’Callaghan et al.¹² measured changes in brain-derived neurotrophic factor (BDNF) at the start and the end of 12 weeks. BDNF levels did not rise significantly from the start to the end of individual sessions. Over 12 weeks, BDNF levels significantly increased in the HIIT group but not in the Moderate Intensity Continuous Training (MICT) group or the control group. Kim et al.¹⁷ suggested that HIIT significantly improves various sarcopenia-related parameters, such as lean mass, skeletal muscle mass, and functional performance measures, more than moderate-intensity continuous training (MICT) and control group. Ergun Y. et al.¹⁸ compared aerobic fitness and HIIT on PD over 6 weeks. The authors found improved aerobic fitness (VO2max), motor function, fatigue, mood, executive function, and quality of life, while most cognitive measures remained unchanged. Motor Unified Parkinson’s Disease Rating Scale UPDRS scores were correlated with improved cognitive performance, selective attention and quality of life (p < 0.05). Harvey et al. reported significant improvements (p = 0.02) in maximum heart rate (HRmax) and peak oxygen consumption (VO2max), when HIIT was compared to controls. Fernandes et al.¹⁹ compared HIIT and MICT collecting outcomes on hemodynamic and functional characteristics in PD patients and found improved endothelial reactivity and various hemodynamic measures. Considering all the above we can safely state that engaging in exercises by people living with PD has a vital role in their rehabilitation²⁰ due to exercise-induced positive neuroplastic changes resulting from increased physical activity.

The emerging evidence supports the benefits of aerobic and HIIT training in managing PD symptoms has shown that the comparative effectiveness of these exercise modalities remains to be determined, specifically in the early stage of PD. The absence of a systematic review comparing HIIT versus aerobic training creates uncertainty in clinical decision-making. Despite the growing interest in exercise, there is a need for evidence-based guidelines as there is a lack of consensus regarding the most effective type of exercise.²¹ This systematic review aims to assess the efficacy of HIIT versus aerobic training in early-stage of PD, assess the quality of reporting of studies identified, summarise findings and provide any future research recommendations.

Methodology

Results

Selection of studies

A total of 327 studies were obtained from electronic databases and manual searches, of which 87 were duplicated and eliminated. Following duplicate removal, 238 studies were included in the title and abstract screening. There was 100% agreement with the independent reviewer (PAN) on study selection. From 238 studies, 206 were excluded, and 32 citations were moved to full-text screening, of which 23 were excluded for different reasons: not peer reviewed (1), wrong intervention (16) and wrong study design (6). A final number of 9 studies were included in the final analysis (see Figure 1).

Figure 1. Prisma flow diagram.

Quality of the studies

Overall, four studies (44.4%) had a low risk of bias, indicating high-quality studies.^16,17,24,25 One (11.1%) study²⁶ presented a high risk of bias. Four studies (44.4%) had some concerns across the risk of bias domains. In domain 1, the plot indicated that 55.5% of studies had no bias from the randomisation process. The remaining 44.5% of studies presented some concerns regarding the method used for randomisation. In domain 2, 100% of studies indicated a high risk of bias due to the blinding process and the nature of the intervention. Domain 3, showed that all (100%) studies had a low risk of bias regarding any missing outcome data. Domain 4 presented six studies (66.6%) with a low risk of bias, one (11.1%) study²⁶ had some concerns regarding the measures utilized to assess outcomes at comparable times between intervention groups and two (22.2%) studies reported high risk of bias. Domain 5 indicated a high risk of bias in six studies (66.6%) aroused through the reporting of result which means they did not report analysis of intention/intention to treat and two (22.2%) studies^16,25 reported intention to treat (see Figure 2).

Figure 2. Traffic light plot of risk of bias.

Discussion

This systematic review assessed the efficacy of HIIT versus aerobic exercises or control group in the early stage of PD. We also assessed the quality of reporting of the RCTs that were eligible for inclusion. The efficacy of HIIT over aerobic training or control group with usual care showed improvements in aerobic capacity, motor function, sarcopenia-related parameters, cognitive function, and Brain-Derived Neurotrophic Factor (BDNF) levels at different time points. Therefore, these results may be associated with the overall efficacy of this intervention over time.

A recent systematic review and meta-analysis¹⁴ aimed to assess the feasibility, safety, and clinical effects of HIIT in PD. The authors identified eleven studies and found that HIIT may be a safe and feasible option for people with PD, those with mild to moderate disease severity. This is in line with our findings. However, this study did not include participants in a specific stage of the disease and included any study design considering any modality of HIIT.

The RCT conducted by Ergun et al.¹⁸ observed that HIIT improved aerobic fitness, motor function, fatigue, mood, and overall QoL. There was statistical significance (p < 0.05) on increasing VO2 max, QoL and improvement in the Unified Parkinson’s Disease Rating Scale (UPDRS) score across all participants. Similarly, another RCT conducted by O’Callaghan et al.¹² found that BDNF levels increased significantly in the long term across a 12-week study, but changes were absent in the MICT and control group. The reduction in neuroinflammation biomarkers caused by increased BDNF levels has been highlighted as an important therapeutic strategy for PD.²⁸ This is in line with the study from Harpham et al.¹⁴ that found HIIT improved cardiorespiratory fitness and may increase BDNF levels.

Similarly to the previous two studies, Kathia et al.²⁵ found that both HIIT and MICT significantly increased VO2peak during the study (p < 0.01). However, the improvement was better in the HIIT group (∼3.7 ± 3.7 ml/kg-1/min-1) than MICT (∼1.7 ± 3.2). UPDRS-III symptoms showed statistical improvement (P < 0.001) equally in both groups. This suggesting that HIIT may provide greater cardiovascular benefits which can be clinically important.

Furthermore, the RCT conducted by Fernandes et al.¹⁹ found increase in endothelial reactivity in HIIT (∼8%, P < .01) and aerobic capacity but not in the MICE group after follow-up. Also, this study reported overall increase in sit to stand test for both HIIT (27.2 ± 6.1%, P < .05) and MICE (21.5 ± 5.4%, P < .05). HIIT showed a better outcome in the six-minute walk test (10.4 ± 3.8%, P < .05), but not after MICE.

However, a RCT conducted by Fernandes et al.²⁴ found an increase in HR and systolic blood pressure, with a small statistical significance between HIIE and MICE (P < 0.01) reported. This study had a small sample size (12 participants), and the duration of the study (14 days) was small, compared to the study conducted by Fernandes et al.,¹⁹ which used a similar intervention regime. Therefore, this could influence the detection of true effects and extrapolation of data into clinical settings. Also, the study conducted by Fernandes et al.¹⁹ used a single-blinded design, which can lead to the ascertainment of outcomes and increase the risk of co-intervention. The allocation concealment was affected, leading to exaggerated estimates of treatment effects.

Moreover, a RCT conducted by Kim et al.¹⁷ not only indicated that HIIT was superior (p = 0.024) to MICT and the control group in improving the 6-min walk test but also indicated improvement in sarcopenia-related parameters [Lean mass (p = 0.011), Appendicular skeletal muscle mass (ASM) (p = 0.035), and ASM index (p = 0.025)] after 24-week intervention. In addition, Harvey et al., found that HIIT showed a significant improvement of 0.23% (p = 0.019) per week in mean HRpeak across the intervention period, as well as a significant increase in VO2peak pre and post intervention that was maintained at six-week follow-up in comparison to the control group. Similarly, the meta-analysis conducted by Harpham et al.,¹⁴ which explored the pooled effects of HIIT on VO2 max, revealed a significant improvement in HIIT compared to usual care.

However, significant improvement in HIIT was not only observed in physiological and functional parameters but also in cognitive functions. Fiorelli et al.²⁶ found that an acute bout of HIIT showed an improvement in auditory memory (p < .02), attention (P < .001) and sustained attention (P < .01), whereas MICT improved immediate auditory memory (P < .01) and no effects on cognitive function on the control group.

Feasibility and adherence also was assessed by Harpham et al.²⁷ and found 78.4% adherence rate to the HIIT sessions where only one participant withdrew due to a non-intervention-related back injury and no serious adverse events reported. Exercises intensity was achieved during HIIT (77.2% HRmax) however, 3 of 7 participants did not consistently achieved threshold intensity of ≥75% HRmax, which raises concerns regarding intensity fidelity in unsupervised settings.

Overall, the changes in different physiological and psychological parameters of these nine RCTs indicated that the effects of HIIT were depicted as a long-term effect. Therefore, it is conceivable that a correlation between intensity and dose-response exists.

Although the significant effects of HIIT versus aerobic or control group were transparent, the risk of bias identified potential systematic errors or deviation from the truth in the study, which can lead to misleading results. Across nine studies, only one study presented an overall high risk of bias, four studies presented a low risk of bias, and four studies presented some concerns. However, although the overall risk presented as high-quality papers, all studies presented high-risk bias due to the inadequate blinding and concealment allocation process due to the nature of the intervention. This can lead to exaggerated treatment effects, affecting the validity and reliability of findings. Also, 66.67% of studies^{12,17,19,24,26} indicated bias arousing through the reporting of results as these studies did not report intention to treat. This finding suggests that some studies did not include all participants in their analyses, potentially leading to misleading conclusions due to the exclusion of dropouts and protocol deviations. This can affect the validity and generalizability of the conclusions drawn from the studies.²⁹

The completeness of intervention reporting demonstrated variability across groups. Completeness of the intervention reporting was more accurate in the intervention group (42%) compared to the comparator (36%) and control group (22%). Excellent quality of reporting was observed for the intervention name, rationale, procedures and when and how much in the intervention group and moderate in the comparator/control group. However, results indicate that there was poor-quality reporting of item 5 (who provided intervention), 9 (tailoring of intervention), 11 (how intervention fidelity was assessed) and 12 (actual intervention fidelity), which shows that the validity of studies is affected. The studies fail to provide level of expertise of the professionals delivering the interventions as well as details about specific training provision. Those imposes challenges when considering implementation of quoted interventions. In addition, there was poor reporting of how the intervention was planned to be tailored, titrated, or adapted to participants. Therefore, it is difficult to understand if the intervention was adapted to participants, thus affecting the replicability of the intervention in clinical settings. Moreover, fidelity was poorly reported across studies and interventions, which means that researchers may encounter difficulties in understanding how closely the intervention was delivered as intended. This affects the reliability of the study findings.³⁰ Lastly, the actual intervention fidelity was reported poorly, which indicates that the study may not have followed the intended intervention. If fidelity was affected during the intervention, this could impact the effectiveness of the intervention and affect future researchers’ ability to replicate the interventions provided accurately.³⁰

From the nine RCTs included, there were no serious AE. Seven studies reported minor musculoskeletal AE (knee pain, lower back pain, ankle pain and hip pain) and one participant experienced a drop in blood pressure during HIIT. This could be due to the impact of PD in the autonomic nervous system causing dysfunction such as neurogenic orthostatic hypotension (NOH), characterised by blood pressure drop upon standing causing symptoms such as dizziness, blurred vision, or fainting.³¹ There were 8/65 participants who reported AE during HIIT, 7/77 participants in aerobic training and 0/48 participants in the control group. This indicated that more participants in HIIT reported exercise-related AE than those in aerobic training or the control group. However, it remains unclear whether these events resulted directly from HIIT. Overall, this showed a low AE across all interventions, suggesting minimal harmful effects and highlighting the safety and tolerability of interventions.

Conclusion

In this systematic review, we have provided comprehensive evidence regarding HIIT compared to aerobic exercises or control groups (usual care, GP care, placebo, waiting list) in improving different outcomes that contribute to improving QoL in the early stage of PD. These findings suggest that HIIT is a highly effective intervention compared to aerobic training or usual care. However, different styles of exercises exist offering different options to patient with a long-term condition. Considering patients preferences as well as services being able to provide the same style of exercises. Although this study showed the effectiveness of HIIT, however, it is essential to note the limitations of the studies included, and the bias aroused. Therefore, future research should focus on conducting a meta-analysis to synthesise evidence and provide a quantitative assessment of intervention effectiveness to guide future research and clinical practice.