Umeå University's logo

Background: Age-related cognitive and physiological decline can in part be mitigated by increasing older adults’ physical exercise, but individuals respond differently, for example in cognitive domains such as working memory (WM). This thesis examines how high intensity exercise affects cognitive, neural, and physiological measures in older adults and explores how clinical findings can be translated into real world settings. 

Methods: This thesis includes four papers, and is based on the Umeå HIT study, which compared supramaximal high intensity interval training (HIT) to moderate intensity training (MIT) for older adults. Papers I-III used data from the Umeå HIT randomized controlled trial (RCT), while Paper IV used data from the Umeå HIT Home Study. The Umeå HIT RCT assessed the effects of 12 weeks of twice-weekly supramaximal HIT (20 minutes total, including 10 x 6 second intervals) compared to MIT (40 minutes total, including 3 x 8-minute bouts) among non-exercising older adults (n = 68, 66-79 years old, 56% women). Exercise intensity was individualized and controlled. Specifically, Paper I assessed cognitive, physiological, well-being and adverse event outcomes, of supramaximal HIT vs MIT. Paper II assessed effects of supramaximal HIT vs MIT on a functional magnetic resonance imaging WM task and examined the relationship between improved leg strength and WM manipulation task-related blood oxygen level dependent (BOLD) response and performance in an MRI subsample. Paper III tested baseline- and change-factors related to WM improvement. Paper IV, based on the Umeå HIT Home study (n = 11, 69-74 years old, 55% women) explored how the original supramaximal HIT protocol could be adapted to home use through a co-creation study involving participants who had exercised in the supramaximal HIT-group of the RCT. 

Results: Paper I found that irrespective of group, cardiorespiratory fitness and systolic and diastolic blood pressure were significantly improved, while global cognitive function was not affected. A significant group x time interaction was found in WM performance and isometric leg extensor strength in favor of supramaximal HIT. Paper II found that increased isometric leg extensor strength in the supramaximal HIT group was positively related to dorsolateral prefrontal cortex  BOLD response, which in turn was related to increased WM performance. Paper III showed that the link between increased isometric leg extensor strength and improved WM also applied to a broader WM composite and the relationship was found in both exercise groups. It further showed that baseline white matter lesion load did not limit WM improvements following supramaximal HIT, unlike MIT. Upregulated brain derived neurotrophic factor was related to WM improvements, but differed by group, suggesting a stronger relationship in MIT. Paper IV identified alternative modalities to stationary bicycling and several adaptations to the supramaximal HIT protocol, including extending intervals to ten seconds, using an audio metronome to control intensity, and a mobile application for exercise delivery. Of the suggested modalities, chair stand intervals elicited similar acute physiological responses to supramaximal HIT on a stationary bicycle, intensity could be systematically modulated using a metronome, and the modality was considered safe.  

Conclusion: This thesis found that supramaximal HIT elicits similar- to superior effects on physiological and cognitive outcomes compared to MIT, despite half the exercise time. Furthermore, leg strength improvements were related to increased BOLD response in a key WM area, which in turn was related to improved WM performance in supramaximal HIT. Leg strength gains were further related to broader WM improvements irrespective of exercise group, indicating that muscular adaptations may be an important target for future exercise-cognition studies. Unlike MIT, supramaximal HIT-related WM gains were not limited by baseline white matter lesion load, and future studies should test this hypothesis directly. Adapting HIT for home use, especially with chair stand intervals, appears promising for future implementation, potentially enabling both cardiorespiratory and muscular gains. Future research should test the feasibility and effects of home-based supramaximal chair stand, as a step toward future implementation to real-world settings for older adults.

Physical exercise shows positive effects on cognitive functions such as working memory (WM) for older adults; however, large individual differences in response exist and the underlying mechanisms are not well understood. We tested the hypothesis that exercise-induced changes in cardiorespiratory fitness and leg strength would improve WM-related brain activity, which subsequently would improve WM performance. This study was based on the Umeå HIT study, a randomized controlled trial assessing the effects of watt-controlled supramaximal high-intensity interval training (HIT) versus moderate-intensity training for nonexercising older adults (N=68). A subsample (n=43, 66 to 79 years, 56% females) underwent task-based functional magnetic resonance imaging, testing WM. The outcomes of interest were change in WM performance, WM task activation, cardiorespiratory fitness, and leg strength. For WM performance, we found no significant between-group difference in change; however, there was a significant within-group increase for HIT in WM composites. For HIT, changes in leg strength significantly predicted increased right dorsolateral prefrontal cortex activation, which in turn predicted improved in-scanner WM task performance. Cardiorespiratory fitness did not predict WM-related functional change. These results indicate a specific physiological ingredient, namely leg strength gains, that is a potential mechanism in exercise-induced prefrontal activation and WM performance increases.

This analysis of secondary outcomes investigated the applicability of supramaximal high-intensity interval training (HIT) with individually prescribed external intensity performed on stationary bicycles. Sixty-eight participants with a median (min; max) age of 69 (66; 79), at the time not engaged in regular exercise were randomized to 25 twice-weekly sessions of supramaximal HIT (20-min session with 10 × 6-s intervals) or moderate-intensity training (MIT, 40-min session with 3 × 8-min intervals). The primary aim was outcomes on applicability regarding; adherence to prescribed external interval intensity, participant reported positive and negative events, ratings of perceived exertion (RPE 6–20), and affective state (Feeling Scale, FS -5–5). A secondary aim was to investigate change in exercise-related self-efficacy (Exercise Self-Efficacy Scale) and motivation (Behavioural Regulations in Exercise Questionnaire-2). Total adherence to the prescribed external interval intensity was [median (min; max)] 89 % (56; 100 %) in supramaximal HIT, and 100 % (95; 100 %) in MIT. The supramaximal HIT group reported 60 % of the positive (112 of 186) and 36 % of the negative (52 of 146) events. At the end of the training period, the median (min; max) session RPE was 15 (12; 17) for supramaximal HIT and 14 (9; 15) for MIT. As for FS, the median last within-session rating was 3 (−1; 5) for supramaximal HIT and 3 (1; 5) for MIT. Exercise-related motivation increased (mean difference in Relative Autonomy Index score = 1.54, 95 % CI [0.69; 2.40]), while self-efficacy did not change (mean difference = 0.55, 95 % CI [-0.75; 1.82]), regardless of group. This study provide support for supramaximal HIT in supervised group settings for older adults.

BACKGROUND: This study examined the effects of regulated and controlled supramaximal high-intensity interval training (HIT) adapted for older adults, compared to moderate-intensity training (MIT), on cardiorespiratory fitness; cognitive, cardiovascular, and muscular function; and quality of life.

METHODS: Sixty-eight nonexercising older adults (66-79 years, 44% males) were randomized to 3 months of twice-weekly HIT (20-minute session including 10 × 6-second intervals) or MIT (40-minute session including 3 × 8-minute intervals) on stationary bicycles in an ordinary gym setting. Individualized target intensity was watt controlled with a standardized pedaling cadence and individual adjustment of the resistance load. Primary outcomes were cardiorespiratory fitness (V̇o2peak) and global cognitive function (unit-weighted composite).

RESULTS: V̇o2peak increased significantly (mean 1.38 mL/kg/min, 95% CI [0.77, 1.98]), with no between-group difference (mean difference 0.05 [-1.17, 1.25]). Global cognition did not improve (0.02 [-0.05, 0.09]), nor differed between groups (0.11 [-0.03, 0.24]). Significant between-group differences in change were observed for working memory (0.32 [0.01, 0.64]), and maximal isometric knee extensor muscle strength (0.07 N·m/kg [0.003, 0.137]), both in favor of HIT. Irrespective of the group, there was a negative change in episodic memory (-0.15 [-0.28, -0.02]), a positive change in visuospatial ability (0.26 [0.08, 0.44]), and a decrease in systolic (-2.09 mmHg [-3.54, -0.64]) and diastolic (-1.27 mmHg [-2.31, -0.25]) blood pressure.

CONCLUSIONS: In nonexercising older adults, 3 months of watt-controlled supramaximal HIT improved cardiorespiratory fitness and cardiovascular function to a similar extent as MIT, despite half the training time. In favor of HIT, there was an improvement in muscular function and a potential domain-specific effect on working memory.

CLINICAL TRIAL REGISTRATION: NCT03765385.

Background: Physical exercise can help prolong healthy aging, yet few options enable older adults to exercise at very high intensities at home. A previous gym-based supramaximal High Intensity interval Training (HIT) protocol on stationary bicycles has shown promising results. Core components were supramaximal interval intensity that could be systematically modulated (controlled, individualized, escalated, and de-escalated) and protocol safety. In this exploratory co-creation study with older adults, we aimed to adapt the supramaximal HIT protocol for potential future implementation in older adults’ home settings.

Methods: Eleven older adults (6 females; ages 69-74) with prior supramaximal HIT experience participated in this two-phase co-creation study. In phase one, co-creators engaged in workshops to identify, explore, and discuss available home-based training modalities. In phase two, co-creators took part in lab tests wherein suitable modalities were tested and compared to stationary bicycling regarding acute physiological responses and safety. Results were continuously merged with the protocol’s core components.

Results: Physiological and emotional reactions to HIT, potential exercise modalities, and necessary protocol adaptations were identified in phase one. When merged with core components three modalities - walking up steps, chair stand, and rubber band cross country double poling - were selected for phase two testing. Of these, chair stand elicited physiological responses most comparable to stationary bicycling, while fulfilling all core components. 

Conclusions: We adapted a researcher-supervised watt-controlled supramaximal HIT-protocol on stationary bicycles to a chair stand protocol with audio metronome-controlled intensity for potential implementation at home. This adaptation is the first step towards a safe and scalable implementation of a home-based supramaximal HIT program for older adults.

Background: Physical exercise can improve cognitive functions for older adults, however it is unclear who responds and why. Change to cognitive function may be influenced by factors related to brain integrity, physical fitness, and exercise-induced biochemicals (myokines), which are proposed to affect neuroplasticity. 

Aims: We tested how exercise-induced working memory change was related to baseline- and change in brain integrity-, fitness-, and myokine factors.   

Methods: We analyzed data from the Umeå HIT study, a randomized controlled trial comparing supramaximal high intensity interval training (HIT) with moderate intensity training (MIT) in older adults. The outcome of interest was exercise-induced change in a working memory composite. We first tested working memory change in relation to baseline measures of fitness (VO2peak and isometric leg extensor strength) and whether brain integrity (periventricular and deep white mater lesion load) moderated exercise-induced gains. Next, we examined whether changes in fitness and circulating myokines (brain-derived neurotrophic factor, insulin-like growth factor-1, and vascular endothelial growth factor A) were associated with working memory improvements.

Results: Baseline periventricular white matter lesion load moderated working memory improvement dependent on exercise type. Participants in the supramaximal HIT-group showed working memory improvements despite severe lesions, while those in the MIT-group decreased in performance if lesion load was high. In terms of change-related factors, increased isometric leg strength was positively associated with working memory gains across both exercise groups. Meanwhile, brain-derived neurotrophic factor change was related to working memory gains, and the relationship differed by group, suggesting a stronger relationship in MIT. 

Conclusions: Individual differences in working memory gains following exercise were related to increased leg strength, brain-derived neurotrophic factor. Periventricular lesion load severity at baseline moderated working memory improvements by group, and high intensity exercise may provide sufficient physiological stimulation to overcome common age-related structural vulnerabilities.