Improvements in cognition and associations with measures of aerobic fitness and muscular power following structured exercise

Experimental Gerontology 112 (2018) 76–87

Contents lists available at ScienceDirect

Experimental Gerontology journal homepage: www.elsevier.com/locate/expgero

Improvements in cognition and associations with measures of aerobic fitness and muscular power following structured exercise

T

Nicholas Cherupa, Kirk Robersona, Melanie Potiaumpaia, Kayla Widdowsona, Ann-Marie Jaghaba, ⁎ Sean Chowdharia, Catherine Armitagea, Afton Seeleya, Joseph Signorilea,b, a b

Laboratory of Neuromuscular Research & Active Aging, University of Miami, Department of Kinesiology and Sport Sciences, Coral Gables, FL, USA University of Miami Miller School of Medicine, Center on Aging, Miami, FL, USA

A R T I C LE I N FO

A B S T R A C T

Keywords: Cognition Executive function Circuit training Exercise Resistance training

Objectives: Cognition, along with aerobic and muscular fitness, declines with age. Although research has shown that resistance and aerobic exercise may improve cognition, no consensus exists supporting the use of one approach over the other. The purpose of this study was to compare the effects of steady-state, moderate-intensity treadmill training (TM) and high-velocity circuit resistance training (HVCRT) on cognition, and to examine its relationships to aerobic fitness and neuromuscular power. Methods: Thirty older adults were randomly assigned to one of three groups: HVCRT, TM, or control. Exercise groups attended training 3 days/wk for 12 weeks, following a 2 week adaptation period. The NIH Cognitive Toolbox was used to assess specific components of cognition and provided an overall fluid composite score (FCS). The walking response and inhibition test (WRIT) was specifically used to assess executive function (EF) and provided an accuracy (ACC), reaction time (RT) and global score (GS). Aerobic power (AP) and maximal neuromuscular power (MP) were measured pre- and post-intervention. Relationships between variables using baseline and mean change scores were assessed. Results: Significant increases were seen from baseline in ACC (MD = 14.0, SE = 4.3, p = .01, d = 1.49), GS (MD = 25.6, SE = 8.0, p = .01, d = 1.16), and AP (MD = 1.4, SE = 0.6, p = .046, d = 0.31) for HVCRT. RT showed a trend toward a significant decrease (MD = −0.03, SE = 0.016, p = .068, d = 0.32) for HVCRT. No significant within-group differences were detected for TM or CONT. Significant correlations were seen at baseline between AP and FCS, as well as other cognitive domains; but none were detected among change scores. Although no significant correlation was evident between MP and FCS or GS, there was a trend toward higher MP values being associated with higher FCS and GS scores. Conclusions: Our results support the use of HVCRT over TM for improving cognition in older persons, although the precise mechanisms that underlie this association remain unclear.

1. Introduction Age related cognitive decline is a common process, beginning as early as age 45 and progressing exponentially throughout the human lifespan (Singh-Manoux et al., 2012). Exercise has been shown to ameliorate symptoms associated with senescence in a variety of studies (Tivadar, 2017). Despite this consensus, the type of exercise (aerobic and anaerobic), dosing patterns (intensity, duration and frequency), and supporting physiological changes remain undetermined (Saez de Asteasu et al., 2017). Though many authors suggest aerobic training is optimal (Gregory et al., 2013; Fabre et al., 2002), others support the value of concurrent (endurance and resistance training) or multi-

⁎

component training (aerobic fitness, strength, balance etc.) (Coelho et al., 2013) over mono-targeted designs (Saez de Asteasu et al., 2017). Therefore, there is a need to explore how different forms of exercise can impact cognition, and more specifically, which form of exercise produces the greatest improvements. An important consideration when evaluating cognition, especially if incorporating the executive function (EF) testing components is testing methodology. Typically, EF assessments utilize either paper and pencil questionnaires or digital display systems that require the use of eyekeyboard activity. Although these approaches can be used to accurately assess EF processes, findings may not translate into real world scenarios involving movement and functional abilities. To address this, the

Corresponding author at: 1507 Levante Ave., Coral Gables, FL 33146, USA. E-mail address: [email protected] (J. Signorile).

https://doi.org/10.1016/j.exger.2018.09.007 Received 4 January 2018; Received in revised form 30 August 2018; Accepted 11 September 2018 Available online 14 September 2018 0531-5565/ © 2018 Published by Elsevier Inc.

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

the precise dose-response relationships required to achieve such benefits, once again, remain to be determined (Chang et al., 2012; Kelly et al., 2014; Smith et al., 2010; Young et al., 2015; Etnier et al., 2006). Over the past two decades, there has been a proliferation of research surrounding resistance-training to increase functional activities and cognition. Specifically, studies have demonstrated that high-velocity resistance training can promote greater improvements in balance (Orr et al., 2006) and functional outcome measures (Miszko et al., 2003; Bottaro et al., 2007; Reid and Fielding, 2012) than standard strength training programs. Additionally, our laboratory has demonstrated that high-velocity resistance training can produce meaningful increases in measures of cognition including working memory and processing speed (Strassnig et al., 2015). The use of high-intensity resistance training, where loads are progressively increased rather than velocity, can positively enhance EF (Best et al., 2015; Liu-Ambrose et al., 2010); however, there is a paucity of research examining the effects of highvelocity resistance training on EF. Given that intensity during resistance training can be modulated through changes in load and movement velocity, and most studies examining resistance training and cognition have used loading to increase intensity, research using velocity-specific increases in intensity is warranted. Although physical activity has been shown to improve several components of cognition, including EF, using a variety of modalities, to our knowledge, no study has compared the effects of a steady-state, moderate-intensity treadmill program (TM) to a high-velocity circuit resistance training program (HVCRT) in an elderly population. Furthermore, we could find no study that established a relationship between exercise-induced changes in neuromuscular power and cognition. Therefore, the primary purpose of this study was to compare the effects of TM training to HVCRT on cognitive domains. A secondary goal was to examine the relationships of cognition to aerobic fitness and neuromuscular power. We hypothesized that both groups would make significant improvements in oxygen consumption, neuromuscular power, and cognition; however, the individuals participating in the HVCRT group would demonstrate greater increases in these measures than TM and control (CONT). Further, we postulated that increases in oxygen consumption and neuromuscular power would be positively correlated to improvements in multiple cognitive domains, including EF.

current study utilized the recently developed Walking Response and Inhibition Test (WRIT; Leyva et al., 2017), which integrates gross motor control and a standardized digital display system to assess the EF domains of cognitive speed flexibility, response selection, inhibition, and capacity to initiate voluntary movement. Addressing EF is particularly important when considering the observed decline in cognitive and functional ability associated with aging (Johnson et al., 2007; Raz and Rodrigue, 2006; Bherer et al., 2013). Diminished EF (Royall et al., 2004) and muscular power (Foldvari et al., 2000; Reid et al., 2014; Izquierdo and Cadore, 2014) are primary factors influencing the decline in functional status in elderly populations. According to a 2010 report by the U.S. Census Bureau, 39.4% of adults over the age of 65 report functional limitations that may interfere with ambulation and 12% reported limitations with at least one activity of daily living (ADL; Brault, 2012). Seniors living independently must maintain certain aptitudes (i.e. ambulating, grocery shopping, bathing, cleaning the house, etc.) to care for themselves and sustain their daily routines. Consequently, finding ways to attenuate age-related EF and muscular power decline can have a significant impact on the maintenance of an independent lifestyle. Age-related senescence and cognitive impairment may be explained by progressive deterioration of white matter microstructure and other subcortical nuclei, expansion of cerebral spinal fluid (CSF), abatement of the cerebral parenchyma and volumetric changes in the neostriatum, hippocampus, and cerebellum (Erickson and Kramer, 2009; Raz and Rodrigue, 2006). As the striatum and cerebellum are vital to movement execution, any degeneration of these structures may lead to compromised motor control and declines in EF. Since neuromuscular deterioration is one of the first signs of decreased power output (Reid et al., 2014; McKinnon et al., 2015) and atrophy in areas associated with neuromuscular control accrue with age (Raz and Rodrigue, 2006), a potential relationship may exist between age-related declines in muscular power and EF. Notably, exercise has been found to enhance a variety of functional and cognitive tasks, and research has shown that brain regions exclusively dedicated to EF appear to be especially sensitive to exercise training (Kramer et al., 2005; Hillman et al., 2008; Saez de Asteasu et al., 2017). Currently there is a growing body of literature supporting the use of resistance (Cassilhas et al., 2007; Chang et al., 2012; Liu-Ambrose et al., 2010), aerobic (Albinet et al., 2010; Kramer et al., 1999; Smiley-Oyen et al., 2008; Guiney and Machado, 2013), and multicomponent (i.e. stretching, resistance and aerobic based) training approaches (Nouchi et al., 2014; Forte et al., 2013; Saez de Asteasu et al., 2017) to improve EF in elderly adults. For example, resistance training programs ranging from 3 to 12 months have been shown to improve selective attention, conflict resolution (Liu-Ambrose et al., 2010), inhibitory capacity (Forte et al., 2013), short-term and long-term memory (Cassilhas et al., 2007), and reaction time (Tsai et al., 2015). Alternatively, 6–10 months of aerobic exercise has been shown to increase cortical activity in areas associated with attentional control (Colcombe et al., 2004), task switching abilities (Kramer et al., 1999), inhibition and working memory (Smiley-Oyen et al., 2008). The cardiovascular fitness hypothesis suggests that aerobic capacity is the mediator that explains the relationship between physical exercise and improved cognitive performance (North et al., 1990). Mechanisms that have been postulated to contribute to this observed improvement are: increased cerebral blood flow, alterations in neurotransmitter release, and altered arousal (Gligoroska and Manchevska, 2012). There is, however, insufficient evidence to support this relationship as results differ among studies. A recent meta-analysis conducted by Kelly et al. (2014) revealed that findings between randomized controlled trials are inconsistent and often in stark contrast to epidemiological, cross-sectional, and neuroimaging studies. The authors postulated that baseline physical performance, length of intervention and follow-up, and the efficacy and adherence to any given intervention may contribute to such conclusions. Furthermore, the optimal intervention lengths and

2. Materials & methods 2.1. Study design The study employed a 14-week randomized, controlled, design to assess cognitive performance (including EF), aerobic (AP) and neuromuscular power (MP), and secondarily, body weight and fat-free mass, in independently living older persons. 2.2. Participants Twenty-two women (age: 69.3 ± 8.1 years; height: 2.58 ± 0.13 m; body weight: 81.1 ± 12.5 kg) and eight men (age: 73.0 ± 5.4 years; height: 3.08 ± 0.32 m; body weight: 101.4 ± 5.8 kg) were recruited to the study. Subjects were stratified in descending order by sex, Mini Mental State Exam (MMSE) scores, education level, hypertension, dyslipidemia and depression scores, and randomized among groups. Sample size estimation was based on previous studies using the Stroop test as a primary outcome measure. The NIH Dimensional Change Card Sort (DCCS) test, a variation of the Stroop test, was used as a primary variable in the current study. Therefore, we used the results reported by Anderson-Hanley et al. (2010) for our power analysis. An ANCOVA model was used for the analysis. A minimum power of 0.80, an effect size of 0.54, and an alpha level of 0.05 yielded a total sample size of 30 participants. Participants 77

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

Fig. 1. Graphic representation of participant's flow through the study.

were between the ages of 55 and 85 years and had a score of ≥23 points on the Mini-Mental State Examination. Individuals were excluded if they had regularly participated in a structured physical activity program within the past three months, had any uncontrolled neuromuscular, orthopedic, or cardiovascular disease, were unable to read or speak English, or had significant cognitive impairment. Additionally, persons who were prescribed and were currently taking a beta-blocker were excluded from the study. Information regarding physical and mental health, as well as physical activity, was provided by each participant on a Physical Activity Readiness and Health History Questionnaire. Participants were instructed to refrain from any type of structured or unstructured exercise throughout the course of the experimental protocol. The study was approved by the University Institutional Review Board and all participants were informed of the benefits and risks of the investigation before signing an institutionally approved informed consent document.

2.3. Procedures Prior to the start of the study, all 30 participants were randomized into the TM or HVCRT groups using randomization software (Research Randomizer; www.randomizer.org/) and completed all required consent forms and questionnaires. The first and final weeks of the study were used for testing. Testing began with anthropometric measurements including body weight and height, and fat free mass. The NIH Cognitive Toolbox was used to assess multiple cognitive domains and a calculated composite score was used to assess overall cognition. The WRIT was employed to assess multiple aspects of EF. AP was assessed using a peak oxygen consumption test on a cycle ergometer, and MP was computed using a sit-to-stand test. During weeks 2–13, participants in the TM and HVCRT were provided supervised exercise programs three times per week. Successful adherence was defined as missing no more than six exercise sessions. The CONT group received no intervention and participants were asked not to change their exercise or dietary habits. Following the intervention, CONT group members were 78

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

conducted in a laboratory setting on a 7 m × 5 m course. There was an additional 2.4 m area at the start which was used to determine walking speed. A colored line was placed in the center of the testing area to provide a visual reference for assessing changes in direction throughout the test. Participants were instructed to walk in a “normal” manner at a comfortable speed. An infrared photo-relay was located at the start of the testing area and was used to activate the system after participants completed the 2.4 m approach. Once the relay was tripped by the participant, an image was displayed following a random time interval, to which the participants were instructed to react. A high-speed camera (JVC GC-PX100; JVC Corp., Long Beach, CA) was placed below the screen, perpendicular to the starting line to record each trial. Trials were recorded at 240 frames per second and all video analyses were completed using Kinovea Software. For the purposes of the study, accuracy (ACC) was calculated as the percentage of correct responses (to the displayed image) across all trials. Reaction time (RT) was calculated by measuring the time in seconds between the appearance of the visual cue and the participant's first defining movement that indicated either a change of direction or the intention to stop (change in foot strike). Testing procedures were described in detail to all participants. They were then shown each of the possible visual symbols and were provided a demonstration followed by the opportunity to perform a familiarization trial prior to conducting the actual test. The same researcher conducted all WRIT pretests and post-tests. Approach speed (AS) was determined using the methods detailed by Leyva et al. and was included to account for any speed-accuracy tradeoff (Leyva et al., 2017). Outcome measures obtained from the WRIT included: decision making accuracy (ACC), reaction time (RT), and a composite global score (GS). The GS incorporating each of the aforementioned variables was calculated using the equation:

provided the opportunity to receive 12 weeks of training using the intervention that was determined to be most effective. Each participant was scheduled to attend 42 training sessions over the course of 14 weeks. 2.4. Primary outcome measures 2.4.1. NIH Toolbox The NIH Toolbox is a well-validated measurement tool for assessing neurological and behavioral function (Gershon et al., 2013). Specifically, the cognitive domain of the NIH Toolbox utilized in this study has been validated in multiple populations (Weintraub et al., 2013; Weintraub et al., 2014). Primary outcome measures assessed by the NIH Toolbox included: EF, attention, episodic memory, working memory, processing speed, and overall cognition. For the purposes of this study, five tests from the cognition battery were selected: the Flanker Inhibitory Control and Attention Test (FLNK), the Dimensional Change Card Sort Test (DCCS), the Picture Sequence Memory Test (PS), the List Sorting Working Memory Test (LS), and the Pattern Comparison Processing Speed Test (PAT), which coincide with five components of EF. A Fluid Cognition Composite Score (FCS), which included the scores from all of the tests, was used to assess overall cognition. The FCS is derived by averaging the scores for each of the measures, and developing standard scores based on the new distribution. Age-corrected standard scores were used for comparative analyses. 2.4.2. Walking response and inhibition test (WRIT) To address the need to measure EF using movement through a physical environment rather than keyboard strokes, the current study utilized the recently developed Walking Response and Inhibition Test (WRIT; Leyva et al., 2017). The test uses high-speed analysis of gross movement responses to a standardized digital display to assess the EF domains of cognitive speed flexibility, response selection, inhibition, and initiation of voluntary movement capacity. Fig. 1 provides a graphic representation of the WRIT testing environment; while Table 1 presents the components of the WRIT test and the associated measures of executive function (EF). The WRIT is a validated and reliable tool used to assess EF that positively correlates with three separate computer-based EF tests and the Timed Up & Go test, an ADL-related assessment known to require EF (Leyva et al., 2017). An in-depth description of the testing environment, instructions, and scoring are provided elsewhere (Leyva et al., 2017). The WRIT was used to measure several discrete aspects of EF including cognitive speed flexibility, response selection, inhibition, initiation, and motor movement capacity (Leyva et al., 2017). Briefly, the WRIT was

[GS = (ACC∗AS)/RT]

2.4.3. Aerobic power testing Aerobic power was assessed using a peak oxygen consumption test. Each participant completed the VO2peak test on a Monark electrically braked cycle ergometer (Model 839E, Vansbro, Sweden). Participants completed a two-minute warm-up at 15 W, followed by between two and four, three-minute stages at increasing workloads. Exercise continued until the participant reached 85% of the age-predicted maximum HR, or until they could no longer continue. HR was continuously monitored using a 12‑lead ECG software program (Carefusion: Cardiosoft, USA). A two-way non-rebreathing nasal and mouth face mask was used, and each participant was fitted with a mask size that minimized any sample loss. Expired air was collected and analyzed

Table 1 The components of the WRIT test and the associated measures of executive function (EF). WRIT stage

Symbols

Task

EF and associated measures

Number of trials

1

Stop immediately at the appearance of the red circle.

1

2

Turn immediately in the appropriate direction dictated by the left or right arrow. If left arrow, turn left Turn immediately in the appropriate direction dictated by the left or right arrow, turn in the direction opposite the arrow if the arrow is accompanied by the blue circle. If left arrow, turn left. If left arrow with blue circle, turn right Perform the same task as described in WRIT 3; however, if anytime the red circle appears with other symbols, the participant stops. Red circle override all other symbols. If left arrow, turn left If left arrow with blue circle, turn right If left arrow with blue light and red light, stop.

Screening for walking, stopping, seeing and understanding visual signal, response inhibition Time perception, initiation, motor control, working memory, response inhibition, reactive agility Time perception, initiation, motor control, working memory, response inhibition, reactive agility, task switching, cognitive flexibility

4

Time perception, initiation, motor control, working memory, response inhibition, reactive agility, task switching, cognitive flexibility, response conflict

8

3

4

From: Leyva et al. (2017). 79

2

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

gradually. During week one participants completed one full-rotation of the circuit. Week two consisted of two rotations, and weeks 3–12, three rotations. HR was obtained at the start and midway point of each circuit by palpating the radial artery for 20 s and multiplying the number of beats counted by three. An average HR for the entire exercise session was calculated from the six recorded values. There was no attempt to target any HR values during HVCRT given that it was an interval-based program. The total time for the entire workout was also recorded for each session. Participants were asked to provide an RPE which reflected the difficulty of the entire workout using the Borg Category Ratio (Borg CR-10). The Borg CR-10, rather than the Borg 15-point RPE scale was chosen for this protocol because scores from the Borg 15-point RPE scale increase linearly with HR and oxygen consumption (Borg, 1998) and HR changes suddenly and disproportionately with each set and rest period during resistance training (Sweet et al., 2004). Total time, average HR, average rest-period, and RPE values were only recorded for weeks 3–12 when three full-circuits were completed. Keiser pneumatic resistance machines were interfaced with a laboratory computer, which allowed the transfer and analysis of data including force, velocity, and power. Progression of load was based on 5% increase in the average power output of a particular exercise on at least two of the three sessions in the previous week. When a significant increase in power was observed, a 10% increase in load was applied for subsequent training sessions. All training sessions were conducted at the same time each day ( ± 1 h) and participants were instructed not to consume high amounts of caffeine or any other substances that may alter HR prior to each exercise session.

continuously using indirect calorimetry (Carefusion: Vmax, USA), and data were exported in 10 s intervals. The highest relative VO2 value attained during testing was accepted as the VO2peak. While reliability has not been tested in our laboratory, the Carefusion Vmax system, Monark electrically braked cycle ergometer, and Carefusion Cardiosoft components constitute an automated system for graded exercise testing that minimizes differences among repeated tests. 2.4.4. Neuromuscular power testing To measure MP, a Tendo unit (Tendo Sports Machines, Trencin, Slovak Republic) was used during the performance of a sit-to-stand (STS) movement. The Tendo is a validated and reliable tool used to measure velocity, force, and power (Janssen, 2015), and has been tested under similar conditions in an age-matched population (Gray and Paulson, 2014). During each STS trial, the participants sat on a chair of standard height (0.47 m) with their arms crossed over their chest. Participants were asked to place their feet in the position that they deemed most advantageous for standing. Foot placement was marked, and identical positioning was used for all subsequent measures. Participants were instructed to stand from the chair to a fully-erect position as fast, and with as much force as possible, without lifting their feet from the ground. Between trials a 45 s recovery period was provided. A total of five attempts were performed by each participant, and the highest power output was selected for use in the analysis. In our laboratory the Cronbach's alpha for reliability using this technique was 0.95, which was similar to the 0.98 reported by Gray and Paulson (2014). 2.5. Secondary outcome measures

2.6.2. Treadmill training protocol All exercise training was conducted in a supervised laboratory setting with an instructor-to-participant ratio of 1:1. Unlike the HVCRT group, this was considered necessary since subjects had the potential to misstep or fall off the back of the treadmill if unable to maintain the dictated pace. Once again, all training sessions were conducted at the same time each day ( ± 1 h) and participants were instructed not to consume high amounts of caffeine or any other substances that may alter HR prior to each exercise session. One familiarization day was provided prior to the start of the 12-week intervention. Participants were allowed to ask questions regarding the equipment and procedures, and all assessment outcomes were explained in detail. During this time, participants were instructed on the use of the Borg 15-point rating of perceived exertion (RPE) scale. All treadmill exercises were performed three days per week on a motorized Cybex 790T treadmill (Cybex International, Inc., Medway, MA, USA). In accordance with ACSM guidelines for moderate intensity exercise (American College of Sports Medicine [ACSM], 2013), training was conducted at 55% ( ± 2 bpm) of each participant's heart rate reserve (HRR), as determined by the Karvonen formula using age predicted maximum heart rate values (Camarda et al., 2008). HR was continuously monitored using a Polar portable sensor (H7 Heart Rate Sensor, Lake Success, NY, USA). The first three weeks of training comprised the conditioning phase, in which participants steadily increased exercise duration until the target time of 35 min was achieved. Participants were instructed to perform a self-paced warm-up for the first three minutes of each exercise session. Exercise intensity was modulated by increasing either the speed or percent grade of the treadmill. Participants were instructed to begin increasing the percent grade once they had achieved the fastest walking pace that could be maintained for the duration of the workout. A cool-down was conducted at the end of each 35 min training session. During the cool-down, participants were instructed to decrease speed and grade until their HR reached a value within 10% of resting. At the end of each session, an RPE, average speed, percent grade, and total time were recorded. During weeks 4–12, participants were encouraged to increase the intensity once their exercising HR dropped > 2 bpm below their target (55% HRR).

2.5.1. Body weight and fat-free mass Body weight and fat free mass were measured using a Tanita BC-418 bioelectrical impedance scale (Tanita, Corporation of America, Inc., Illinois, USA), and height was measured using a stadiometer that was a component of a medical dual beam scale (Detecto Corp, Webb City, MO, USA). 2.6. Interventions 2.6.1. High-velocity circuit-resistance training Eleven computerized pneumatic resistance exercise machines (Keiser A420, Keiser Corporation, Fresno, CA, USA) were utilized during the study. A one-repetition maximum (1RM) was determined for each participant on each machine using guidelines established by the National Strength and Conditioning Association (Baechle et al., 2008). Following testing, the optimal training load for the production of muscular power was calculated for each exercise, relative to a participant's 1RM. The percentages of each 1RM corresponding to the optimal loads for each machine were previously identified by Potiaumpai et al., in an age-matched population using identical resistance exercise machines (Potiaumpai et al., 2016). For the circuit training protocol, the 11 exercises were completed at the optimal loads specified for each machine (%1RM) in the following order: chest press (50%), leg press (60%), latissimus dorsi pull-down (40%), hip adduction (70%), overhead press (60%), leg curl (60%), seated row (50%), hip abduction (70%), elbow extension (50%), plantar flexion (60%), and elbow flexion (50%). Each set consisted of 12 repetitions. For the concentric portion of the lift, the participant was asked to “move the load as forcefully and as quickly as possible.” The eccentric phase of the lift consisted of a two-second controlled motion. Rest intervals were self-directed; however, participants were instructed not to begin the next exercise until they were confident that they could complete all assigned repetitions without an intra-set rest period. The duration of each rest-period was measured using a digital stopwatch and the average time for all rest-periods was calculated. In the first three weeks of training, the loading volume increased 80

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

group differences following training were detected among groups. Minimum detectable changes (MDC) for AP in the HVCRT, TM and CONT were 2.06, 2.09, 2.21 ml·kg−1·min−1, respectively. Three participants in the HVCRT group met or exceeded this threshold, while two participants in the TM group and none in the CONT group met or exceeded their MDC values. For MP, MDC values were 161.8, 148.1, and 239.9 for the HVCRT, TM and CONT groups, respectively. One subject in the HVCRT group and 1 subject in the TM group met or exceeded their MDC values.

2.7. Statistical analysis An analysis of covariance (ANCOVA) was used to examine between group differences. Pretest values were used as covariates to improve precision and to control for possible imbalances during the randomization process. Bonferroni post hoc tests were used when appropriate to determine specific pairwise differences. Paired t-tests were used to investigate within-group changes. A Pearson product-moment correlation coefficient was computed to assess the relationships between variables. The strength of each association is interpreted as: weak = 0.1 < |r| < 0.3, moderate = 0.3 < |r| < 0.5, strong = |r| > 0.5. All significance tests were two-tailed and a significance level of α ≤ 0.05 was set a priori. Effects sizes for Cohen's d were listed as absolute values and are interpreted as: 0.80 is considered large, 0.50 is considered medium, and 0.20 is considered small. All statistical analyses were performed using the SPSS, version 24 statistical package (IBM SPSS Statistics, Armonk, NY). Data are expressed as mean ± standard error (SE).

3.5. Correlation analyses

Descriptive data for the 24 subjects are summarized in Table 2. There were no significant differences in baseline characteristics between groups. Body weight and fat free mass did not significantly change from baseline in any group.

Two separate analyses were conducted to determine the associations between variables: an analysis using the combined data from pretest scores of participants in all groups, and an analysis using change scores, where the data for each group were examined separately. Descriptive data for all participants at baseline are presented in Table 4. We examined the relationships between the GS derived from the WRIT and components of the NIH Cognitive Toolbox. A significant positive correlation was seen between GS and FCS (r = 0.481, p = .02), indicating that generally, individuals with a higher GS attained a higher FCS (Fig. 4). There was also a strong significant negative relationship between FCS and RT (r = −0.534, p < .01), indicating that individuals with higher fluid composite scores achieved faster reaction times. The relationships of AP and MP with several components of the WRIT and NIH Toolbox are displayed in Table 5. There were significant positive correlations between AP and FCS (r = 0.494, p = .017), PAT (r = 0.714, p < .01), PS (r = 0.437, p = .037), and DCCS (r = 0.486, p = .016). No significant correlations were found between MP and any computed variables. Fig. 5 shows the relationships of AP and MP with FCS and GS. Data from the correlation analyses using change scores for the WRIT and NIH Cognitive Toolbox are presented in Tables 5 and 6, respectively, with each group represented separately. For TM, there was a strong significant negative correlation between MP and LS (r = −0.788, p = .035), indicating that as MP decreased, LS increased. Additionally, for CONT, there were strong significant negative correlations between MP and LS (r = −0.951, p = .049) and FLNK (r = −0.986, p = .014).

3.3. RPE, average HR, rest intervals, and total exercise time

4. Discussion

RPE was assessed using the BORG CR-10 for HVCRT and the Borg 15-point RPE scale for TM. Average RPE for HVCRT was 6.0 ± 0.2, which corresponds to a work effort that lies between “Hard” and “Very Hard”, while the average RPE for TM was 10.9 ± 0.5, which corresponds to a “Light” level of perceived exertion. The difference in average HR between HVCRT (107 ± 4 bpm) and TM (114 ± 3 bpm) was not statistically significant. Average recovery interval time for all subjects in the HVCRT group for weeks 3–12 of training was 32 ± 3 s. Total exercise session time for the TM group was 33 ± 2 mins for weeks 3–12, and 30 min ± 2 min for HVCRT.

The purpose of this investigation was to compare the effects of TM and HVCRT on components of cognition, including the set-shifting component of EF, visuospatial inhibitory attention, working memory, episodic memory, and processing speed. The distinct relationships of AP and MP to cognition were also examined. To our knowledge, this was the first study to use a circuit resistance training strategy utilizing explosive contractions to effect changes in cognition. Our findings indicate that HVCRT was a more effective training modality for improving AP and WRIT performance (including RT, ACC, and GS). These findings suggest improvements in EF. Additionally, we discovered significant positive correlations between AP and components of the NIH Cognitive Toolbox at baseline, which were not seen using change scores.

3. Results 3.1. Adherence Of the 30 individuals recruited, three participants withdrew due to medical complications unrelated to the study, and one voluntarily withdrew, leaving 26 participants (20 women and six men) who completed the study. Subject data were included for analysis only if criteria for adherence were met. Study adherence was defined as attending 30 of 36 required exercise sessions (83%) and all pretesting and posttesting sessions. Based on these criteria, data from 24 of the 26 subjects were included in the analysis. Fig. 2 presents a chart showing the flow of participants through the study. 3.2. Baseline characteristics and anthropometrics

3.4. Primary outcome measures Within-group data for the six measures from the NIH Cognitive Toolbox and two measures from the WRIT are presented in Table 3. Fig. 3 depicts the within-group differences for GS, AP, and MP. A within-group analysis revealed significant increases from baseline in ACC (MD = 14.0, SE = 4.3, p = .01, d = 1.49), GS (MD = 25.6, SE = 8.0, p = .01, d = 1.16), and AP (MD = 1.4, SE = 0.6, p = .046, d = 0.31) for HVCRT. RT showed a trend toward a significant decrease (MD = −0.03, SE = 0.016, p = .068, d = 0.32) for HVCRT. No significant within-group differences were detected for TM or CONT. A significant difference was detected in MP between CONT and HVCRT at baseline (MD = −553.6, SE = 177.4, p = .020, d = 1.89); however, the ANCOVA accounted for this difference. No significant between-

4.1. Effects of HVCRT on cognition Results supporting the role of exercise for improving cognition in the elderly remain equivocal (Kelly et al., 2014; Snowden et al., 2011). While some research indicates that both multicomponent (balance, agility, coordination training) and progressive resistance training favorably reduce age-related declines in EF (Forte et al., 2013); other studies failed to produce significant findings (Snowden et al., 2011; Smith et al., 2010; Young et al., 2015). Our study affirms the efficacy of resistance training in enhancing cognition, with results showing significant increases in WRIT GS and ACC in the HVCRT group with RT 81

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

Fig. 2. Graphic representation of the WRIT testing environment.

AP change scores in the HVCRT were not significant, we hypothesize that both neuromuscular and cardiovascular training are necessary to maximize exercise-induced increases in cognitive capacity. Increased aerobic capacity has been found to enhance cognitive function (Fabre et al., 2002); therefore, the observed improvements in AP with HVCRT should be considered to have a notable influence on the cognitive variables measured. This is based on the finding that significant improvements were seen in AP, GS, and its subcomponent ACC following HVCRT, while no other significant within-group changes were noted. In animal models, aerobic exercise has been shown to enhance cognition through neural structure modifications such as synaptogenesis, neurogenesis, and increased vascular branching (Churchill et al., 2002). The efficacy of enhanced circulation to brain regions associated with higher order processing is attributed to increases in the profusion of glucose and other growth factors, such as BDNF and IGF-1, that support structural changes in the brain (Cotman et al., 2007; Lojovich, 2010). There is a body of work indicating that aerobic exercise can increase grey (Erickson and Kramer, 2009) and white matter in older populations (Colcombe et al., 2006); however, these increases are not unique to aerobic training. Tsai et al. (2015) reported increased serum IGF-1 levels and improved performance on several cognitive assessments as a result of a high-intensity resistance training.

Table 2 Baseline participant characteristics.

Age (years) Height (cm) Weight (kg) MMSE (score) Education level Hypertension (n) Dyslipidemia (n) Depression (n)

HVCRT

TM

CONT

n=9

n=8

n=7

72.2 ± 2.6 166.8 ± 2.7 91.5 ± 3.0 29.1 ± 0.2 3.0 ± 0.4 5 3 2

68.5 ± 2.8 164.8 ± 2.9 85.6 ± 5.3 28.6 ± 0.5 2.62 ± 0.5 4 3 3

70.3 ± 3.0 158.4 ± 3.1 78.3 ± 7.0 25.1 ± 4.2 2.4 ± 0.4 3 2 2

All values are mean ± standard error. HVCRT = high-velocity circuit resistance training; TM = continuous moderate-intensity treadmill training; CONT = control. Education level was based on the degree completed and stratified into five levels; 1 = high school, 2 = associates, 3 = bachelor's, 4 = master's, 5 = doctorate. Positive hypertension, dyslipidemia and depression values were obtained through a health questionnaire and/or if the individual was currently taking medication to treat their illness.

scores approaching significance. The HVCRT group also achieved significant increases in AP. Given that no significant differences were detected in GS for TM or CONT, and the correlations between GS and 82

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

Table 3 Changes from baseline for primary outcome measures. HVCRT (n = 9); TM (n = 8); CON (n = 7)

WRIT Accuracy (%) TM HVCRT CONT Reaction Time (s) TM HVCRT CONT NIH cognitive battery List Sorting Memory TM HVCRT CONT Pattern Comparison Processing Speed TM HVCRT CONT Picture Sequence Memory TM HVCRT CONT Flanker Inhibitory Control & Attention TM HVCRT CONT Dimensional Change Card Sort TM HVCRT CONT Fluid Cognition Composite Score TM HVCRT CONT

Baseline

Change at post-test

Treatment effect

p

Cohen's d

75.00 ± 3.15 69.18 ± 3.15 78.57 ± 5.71

4.76 (−7.71, 17.24) 13.98 (4.14, 23.83) 4.55 (−8.30, 17.40)

−2.24 (−18.04, 13.55) 2.98 (−12.78, 18.73)

0.39 0.01⁎ 0.42

0.45 1.49 0.33

0.872 ± 0.062 0.783 ± 0.035 0.872 ± 0.063

−0.046 (−0.176, 0.084) −0.033 (−0.069, 0.003) −0.105 (−0.34, 0.13)

0.058 (−0.147, 0.263) 0.023 (−0.178, 0.223)

0.42 0.07 0.32

0.26 0.32 0.62

105.90 ± 4.77 100.06 ± 2.74 101.63 ± 3.68

3.74 (−3.62, 11.09) 4.02–5.52, 13.57) −1.33 (−11.91, 9.25)

6.81 (−7.45, 21.07) 4.71 (−9.01, 18.43)

0.27 0.36 0.79

0.28 0.43 0.12

89.59 ± 3.40 94.19 ± 3.34 91.09 ± 4.09

4.66 (−5.21, 14.53) −0.64–10.41, 9.12) −5.37 (−20.27, 9.53)

10.30 (−8.26, 28.86) 4.17 (−14.02, 22.37)

0.301 0.863 0.421

0.32 0.04 0.36

97.77 ± 3.97 103.96 ± 6.16 101.07 ± 4.21

3.77 (−6.88, 14.4) 7.87 (−2.35, 18.09) 9.33 (−5.02, 23.67)

−8.02 (−19.87, 3.82) 0.69 (−10.47, 11.86)

0.42 0.11 0.16

0.41 0.51 1.05

125.15 ± 7.58 126.76 ± 7.07 122.49 ± 4.33

−0.56 (−11.20, 10.07) −0.50 (−15.84, 14.84) −2.16 -39.42, 35.11)

3.36 (−28.00, 34.73) 4.49 (−26.15, 35.11)

0.90 0.94 0.89

0.03 0.03 0.08

130.74 ± 7.40 141.48 ± 7.03 147.54 ± 7.07

8.06 (−2.61, 18.74) 7.79 (−1.77, 17.35) −4.69 (−34.80, 25.43)

5.16 (−21.55, 31.87) 9.74 (−15.03, 34.51)

0.12 0.10 0.72

0.37 0.42 0.21

117.06 ± 5.36 119.52 ± 3.61 118.90 ± 5.01

3.16 (−4.62, 10.93) 3.17 (−7.02, 13.36) 1.06 (−14.86, 16.97)

1.50 (−16.93, 19.94) 2.31 (−15.04, 19.66)

0.36 0.49 0.88

0.26 0.22 0.07

All values are mean ± standard error. HVCRT = high-velocity circuit resistance training; TM = continuous moderate-intensity treadmill training; CONT = control; Meandiff ± SE = Mean difference with standard error. ⁎ Significant at p < .05.

Our findings are unique, since they show that HVCRT can increase peak oxygen consumption (AP: p = .046) and WRIT scores (ACC: p = .01; GS: p = .01; p = .068]; RT: p = .068) to a greater extent than TM. This was not unexpected since a number of studies have shown that interval training can produce greater improvements in AP compared to low and/or moderate intensity steady state training (Hafstad et al., 2011; Helgerud et al., 2007; Kemi et al., 2005; Tjonna et al., 2008). Helgerud et al. (2007) employed a similar study design in which groups completed either steady-state moderate-intensity treadmill training or interval training. They found that only interval training significantly increased maximum oxygen consumption and cardiac stroke volume. Similar findings were reported by Kemi et al. (2005), who showed that higher intensity interval training was superior to moderate intensity interval training in increasing maximal oxygen consumption, cardiomyocyte dimensions and contractile capacity. Tjonna et al. (2008) also demonstrated 16 weeks of thrice weekly aerobic interval training increased maximal oxygen consumption to a greater extent than moderate continuous treadmill exercise. It should be noted that each of these studies used treadmill walking/running and not resistance exercise as an interval training modality. However, Haennel et al. (1991) demonstrated that circuit resistance exercise, albeit not high-velocity, significantly increased cardiac output, stroke volume, and peak oxygen consumption. To our knowledge, no studies have assessed improvements in cardiovascular measures using high-velocity circuit resistance exercise as an interval training modality compared to steady-state treadmill training. Observable increase in GS following HVCRT in our study may be

explained by the diverse explosive resistance exercises that required individuals to dedicate additional cognitive resources when changing neuromuscular patterns used during training. Repeated activation of these motor pathways may have enhanced performance during the WRIT, which requires purposeful and more complex movements when compared to keyboard entry tasks on the NIH Toolbox and other cognitive assessment programs. This proposition is supported by a number of studies that have shown that movement complexity is a strong stimulus for improved cognition. Oltmanns et al. (2017) demonstrated that adding complexity by changing task patterns in the workplace, could have a positive impact on processing speed and working memory, and increase grey matter volume in cortical areas associated with learning that typically decline with age. Additionally, higher aerobic and motor fitness (i.e. movement speed, balance, fine coordination, and flexibility) can synergistically activate brain regions associated with EF and provide more efficient use of available cognitive resources (Voelcker-Rehage et al., 2010). Ultimately, increased cognitive capacity should translate into activities of daily living that require greater mental focus, thereby maintaining or improving independence in aging individuals. No changes were seen in MP as a result of HVCRT or TM. The lack of impact on MP due to TM training is not without precedent. Bernard et al. (2016) reported no significant improvements in average isokinetic knee extension power of the dominant leg at 3.14 rad·s−1 or at 1.05 rad·s−1 or the non-dominant leg at 3.14 rad·s−1 in sedentary older women following six months of brisk walking. Tabata et al. (1990) examined the impact of high-intensity endurance training on maximal 83

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

Table 4 Baseline values for all participants. Measure

Mean ± SE

Walking response inhibition test GS RT (s) ACC (%)

87.32 ± 4.38 0.83 ± 0.03 73.41 ± 2.31

NIH cognitive battery FCS LS PAT PS FLNK DCCS

118.58 ± 2.53 102.46 ± 2.14 91.75 ± 2.02 101.20 ± 2.91 124.98 ± 3.73 139.67 ± 4.22

Power tests AP (ml·kg−1·min−1) MP (W)

14.64 ± 0.69 1075.79 ± 79.87

GS = global score; RT = reaction time; ACC = accuracy; FCS = fluid cognition composite score; LS = list sorting working memory test; PAT = pattern comparison processing speed test; PS = picture sequence memory test; FLNK = flanker inhibitory control and attention test; DCCS = dimensional change card sort test; AP = aerobic power; MP = neuromuscular power.

Fig. 4. Correlation between the WRIT Global Scores and NIH Cognitive Toolbox Fluid Composite Scores for all subjects. r = 0.481, p = .02. Fig. 3. Primary outcome variable results. Within group changes in (a) Global Score, (b) aerobic power and (c) neuromuscular power. *p < .05. TM = continuous moderate-intensity treadmill training; HVCRT = high velocity circuit resistance training; CONT = control. ■ = pretest; □ = post-test.

Table 5 Correlation analyses using change scores for the WRIT global score and NIH Toolbox fluid cognition composite score in each group.

oxygen uptake and isokinetic power and found that isokinetic power measures increased at the slower testing speeds (0.52, 1.05, 2.10 rad·s−1), but not at higher speeds (3.14, 4.19, 5.24 rad·s−1). Additionally, there is substantial research demonstrating that the mechanical and enzymatic changes associated with endurance training are not optimal for the development of mechanical power (Dubouchaud et al., 2000; Kraemer et al., 1995; Rhea et al., 2008). Numerous studies also indicate the inclusion of aerobic exercise to a resistance training program can reduce improvements in hypertrophy, strength and power (Wilson et al., 2012). In the HVCRT protocol, the work cycle was approximately 30 s, while recovery was 32 ± 3 s, this constituted a 1:1 work: recovery duty cycle, which targeted aerobic conditioning over improvement in strength or power. Further, in order to maximize power output, the optimal recovery should be 2 to 5 min (Robinson et al., 1995; Tan, 1999). Data collected during training, which ranged from a perceived exertion of hard to very hard, showed that the intensity of

Group

Cognitive measures

GS

AP

MP

TM

GS FCS GS FCS GS FCS

– −0.114 – 0.327 – 0.089

−0.341 0.122 −0.097 0.217 0.610 0.316

−0.432 −0.004 0.393 −0.231 0.071 −0.833

HVCRT CONT

Data are Pearson's correlation coefficients. TM = continuous moderate-intensity treadmill training; HVCRT = high-velocity circuit resistance training; CONT = control; GS = global score; FCS = fluid cognition composite score; AP = aerobic power; MP = neuromuscular power.

training declined during each exercise circuit, providing progressively less stimulus for improving power. Finally, it is possible that improvements in power output, seen in individual exercises during HVRT, did not translate into improvements on the sit-to-stand test. Differences in the biomechanics of the training and testing motions may not have allowed the proper adaptations achieved to improve the sit-to-stand 84

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

Fig. 5. Correlations of aerobic power with (a) NIH Cognitive Toolbox Fluid Composite Scores (r = 0.494, p = .017) and (b) WRIT Global Scores (r = 0.355, p = .088). Correlations of neuromuscular power with (c) NIH Cognitive Toolbox Fluid Composite Scores (r = 0.263, p = .293) and (d) WRIT Global Scores (r = 0.292, p = .225).

branching (Churchill et al., 2002) and it has been hypothesized that an increased profusion of glucose and other growth factors (i.e. BDNF and IGF-1) is associated with higher order processing (Cotman et al., 2007; Lojovich, 2010). The body of work that corroborates these propositions shows that aerobic exercise can increase grey (Erickson and Kramer, 2009) and white matter within older populations (Colcombe et al., 2006). Little support for this hypothesis is provided by our findings, since TM training failed to increase AP significantly. A potential explanation for the lack of cognitive improvement with our TM protocol was its moderate intensity level, we suggest that more strenuous aerobic exercise may be required to enhance EF. Failure of the TM program to increase AP may also be attributed to the length of

test. Each of these factors or their collective impact may have been responsible for the lack of power improvements following these training interventions.

4.2. Effects of TM on cognition A recent meta-analysis by Etnier et al. (2006) found a lack of evidence supporting the cardiovascular fitness hypothesis. They reported a significant negative relationship between change scores across their training period for aerobic fitness and cognitive performance. In contrast, animal models have shown that aerobic exercise can enhance cognition through synaptogenesis, neurogenesis, and increased vascular

Table 6 Correlation analyses using change scores for each test of the NIH Toolbox for each group.

TM HVCRT CONT

AP MP AP MP AP MP

LS

PAT

PS

FLNK

DCCS

ACC

RT

−0.377 −0.788⁎ 0.232 −0.029 −0.117 −0.951⁎

−0.529 −0.445 −0.095 0.616 0.610 −0.106

−0.474 0.037 −0.127 0.365 0.659 −0.027

0.027 −0.165 0.068 −0.614 0.134 −0.986⁎

0.270 −0.183 0.294 −0.078 −0.352 −0.725

−0.116 0.059 0.348 0.166 −0.569 −0.705

0.319 0.240 −0.064 −0.028 −0.685 −0.372

Data are Pearson's correlation coefficients. TM = continuous moderate-intensity treadmill training; HVCRT = high-velocity circuit resistance training; CONT = control; LS = list sorting working memory test; PAT = pattern comparison processing speed test; PS = picture sequence memory test; FLNK = flanker inhibitory control and attention test; DCCS = dimensional change card sort test; ACC = accuracy; RT = reaction time; AP = aerobic power; MP = neuromuscular power. ⁎ Correlation significant; p < .05. 85

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

current evidence substantiates the need for continued participation in a structured exercise program, especially in aging adults.

the current TM intervention. Past authors have questioned the use of shorter exercise protocols and have suggested that long term physical activity is more efficacious in addressing age-related cognitive decline and improving cognition (Erickson and Kramer, 2009; Weuve et al., 2004; Kelly et al., 2014).

Acknowledgements We would like to thank all of the loyal study participants of the Laboratory for Neuromuscular Research and Active Aging, and our undergraduate students for their continued dedication and help.

4.3. Correlation between aerobic and neuromuscular power and cognition Baseline measurements of AP correlated positively with FSC, PAT, PS, and DCCS for all groups. This indicates that individuals that were more aerobically fit at the onset of the intervention tended to score better on tasks requiring visuospatial inhibitory attention, working memory, episodic memory, and processing speed. These data partially support the assertion that cognition is influenced by increased aerobic fitness in older populations (Guiney and Machado, 2013) and that aerobic exercise can reduce cortical decay (Erickson and Kramer, 2009; Colcombe and Kramer, 2003) and increase connectivity among higher order circuitry (Voss et al., 2010), thereby preserving functional structures that decline with age. Finally, results from the current study revealed a significant positive correlation between baseline GS and FCS across groups. These findings support the use of movement based measures of EF, such as the WRIT, in tandem with the cognitive tests in the NIH Cognitive Toolbox. Moreover, the use of the WRIT should be given significant clinical consideration since cognitive ability has been found to significantly predict mobility in elderly populations (Gothe et al., 2014). Considering that reduced EF (Royall et al., 2004; Gothe et al., 2014) and decreased power (Foldvari et al., 2000; Reid et al., 2014; Izquierdo and Cadore, 2014) are linked to the loss of functional status in the elderly, the WRIT should be added to the battery of cognitive assessments when ADL execution and mobility are important factors.

References Albinet, C.T., Boucard, G., Bouquet, C.A., Audiffren, M., 2010. Increased heart rate variability and executive performance after aerobic training in the elderly. Eur. J. Appl. Physiol. 109 (4), 617–624. American College of Sports Medicine, 2013. General principles for exercise prescription. In: Pescatello, L., Arena, R., Riebe, D., Thompson, P. (Eds.), ACSM's Guidelines for Exercise Testing and Prescription. Lippincott Williams & Wilkins, Baltimore, MD, pp. 162–167. Anderson-Hanley, C., Nimon, J.P., Westen, S.C., 2010. Cognitive health benefits of strengthening exercise for community-dwelling older adults. J. Clin. Exp. Neuropsychol. 32 (9), 996–1001. Baechle, T.R., Earle, R.W., Wathen, D., 2008. Resistance training. In: Baechle, T.R., Earle, R.W. (Eds.), Essentials of Strength Training and Conditioning, 3rd ed. Human Kinetics, Champaign, IL, pp. 395–396. Bernard, P.L., Tallon, G., Ninot, G., Jaussent, A., Ramdani, S., Coste, O., Picot, M.C., Blain, H., 2016. Influence of a brisk walking program on isokinetic muscular capacities of knee in sedentary older women. Aging Clin. Exp. Res. 28 (6), 1219–1226. Best, J.R., Chiu, B.K., Hsu, C.B.L., Nagamatsu, L.S., Liu-Ambrose, T., 2015. Long-term effects of resistance exercise training on cognition and brain volume in older women: results from a randomized controlled trial. J. Int. Neuropsychol. Soc. 21 (10), 745–756. Bherer, L., Erickson, K.I., Liu-Ambrose, T., 2013. A review of the effects of physical activity and exercise on cognitive and brain functions in older adults. J. Aging Res. 657508. https://doi.org/10.1155/2013/657508. Borg, G., 1998. Borg's Perceived Exertion and Pain Scales. Human Kinetics, Champaign, IL. Bottaro, M., Machado, S.N., Nogueira, W., Scales, R., Veloso, J., 2007. Effect of high versus low-velocity resistance training on muscular fitness and functional performance in older men. Eur. J. Appl. Physiol. 99 (3), 257–264. Brault, M.W., 2012. Americans with Disabilities. US Department of Commerce, Economics and Statistics Administration, US Census Bureau, Washington, DC, pp. 2010. Camarda, S.R., Tebexreni, A.S., Páfaro, C.N., Sasai, F.B., Tambeiro, V.L., Juliano, Y., Barros Neto, T.L., 2008. Comparison of maximal heart rate using the prediction equations proposed by Karvonen and Tanaka. Arq. Bras. Cardiol. 91 (5), 311–314. Cassilhas, R.C., Viana, V.A., Grassmann, V., Santos, R.T., Santos, R.F., Tufik, S., Mello, M.T., 2007. The impact of resistance exercise on the cognitive function of the elderly. Med. Sci. Sports Exerc. 39 (8), 1401–1407. Chang, Y., Pan, C., Chen, F., Tsai, C., Huang, C., 2012. Effect of resistance-exercise training on cognitive function in healthy older adults: a review. J. Aging Phys. Act. 20 (4), 497–517. Churchill, J.D., Galvez, R., Colcombe, S., Swain, R.A., Kramer, A.F., Greenough, W.T., 2002. Exercise, experience and the aging brain. Neurobiol. Aging 23 (5), 941–955. Coelho, F.G.D.M., Andrade, L.P., Pedroso, R.V., Santos-Galduroz, R.F., Gobbi, S., Costa, J.L.R., Gobbi, L.T.B., 2013. Multimodal exercise intervention improves frontal cognitive functions and gait in Alzheimer's disease: a controlled trial. Geriatr Gerontol Int 13 (1), 198–203. Colcombe, S., Kramer, A.F., 2003. Fitness effects on the cognitive function of older adults: a meta-analytic study. Psychol. Sci. 14 (2), 125–130. Colcombe, S.J., Kramer, A.F., Erickson, K.I., Scalf, P., McAuley, E., Cohen, N.J., Webb, A., Jerome, G.J., Marquez, D.X., Elavsky, S., 2004. Cardiovascular fitness, cortical plasticity, and aging. Proc. Natl. Acad. Sci. U. S. A. 101 (9), 3316–3321. Colcombe, S.J., Erickson, K.I., Scalf, P.E., Kim, J.S., Prakash, R., McAuley, E., Elavsky, S., Marquez, D.X., Hu, L., Kramer, A.F., 2006. Aerobic exercise training increases brain volume in aging humans. J. Gerontol. A Biol. Sci. Med. Sci. 61 (11), 1166–1170. Cotman, C.W., Berchtold, N.C., Christie, L., 2007. Exercise builds brain health: key roles of growth factor cascades and inflammation. Trends Neurosci. 30 (9), 464–472. Dubouchaud, H., Butterfield, G.E., Wolfel, E.E., Bergman, B.C., Brooks, G.A., 2000. Endurance training, expression, and physiology of LDH, MCT1, and MCT4 in human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 278 (4), E571–E579. Erickson, K.I., Kramer, A.F., 2009. Aerobic exercise effects on cognitive and neural plasticity in older adults. Br. J. Sports Med. 43 (1), 22–24. Etnier, J.L., Nowell, P.M., Landers, D.M., Sibley, B.A., 2006. A meta-regression to examine the relationship between aerobic fitness and cognitive performance. Brain Res. Rev. 52 (1), 119–130. Fabre, C., Chamari, K., Mucci, P., Masse-Biron, J., Prefaut, C., 2002. Improvement of cognitive function by mental and/or individualized aerobic training in healthy elderly subjects. Int. J. Sports Med. 23 (6), 415–421. Foldvari, M., Clark, M., Laviolette, L.C., Bernstein, M.A., Kaliton, D., Castaneda, C., Pu, C.T., Hausdorff, J.M., Fielding, R.A., Singh, M.A., 2000. Association of muscle power with functional status in community-dwelling elderly women. J. Gerontol. A Biol. Sci. Med. Sci. 55 (4), M192–M199. Forte, R., Boreham, C.A., Leite, J.C., De Vito, G., Brennan, L., Gibney, E.R., Pesce, C.,

4.4. Limitations Though our results support the potential benefits of HVCRT on measures of cognition in an elderly population, several limitations must be addressed. Firstly, the current study did not include blinded pretest and post-testing assessments. We acknowledge that this increased the possibility for tester bias; however, measures were taken to ensure that assessors who conducted pre-testing on a specific participant did not complete post-testing on that participant. A second limitation was the failure to incorporate an intention-to-treat model; therefore, our results did not adequately reflect participants' attrition. Another consideration that must be addressed is the unequal distribution of male and females in each group. Since the majority of the participants were female, generalizability to the population as a whole may be an issue. By conducting an ANCOVA, we have nullified baseline values that may have resulted from sex differences. In addition, given that there were a large number of outcome variables, and the chosen statistical approach had the potential to increase Type 1 error probability, previous research (Vickers and Altman, 2001) has stated that accounting for baseline measures is the choice method for analyzing randomized trials with baseline and follow-up measurements. Finally, formal monitoring of the control groups exercise or dietary habits was not incorporated into the study design. 5. Conclusion Our findings indicate that the implementation of a 12-week HVCRT regimen provided cognitive improvements, potentially through an increase in AP. This information adds to the literature by presenting resistance training as a specific method of interval training that enhances cognition. Furthermore, implications from this study suggest that resistance training may counteract age-related cognitive impairment and help individuals to maintain their independence later in life. Although the precise mechanisms underlying this association remain unclear, 86

Experimental Gerontology 112 (2018) 76–87

N. Cherup et al.

correlates and modifiers. Neurosci. Biobehav. Rev. 30 (6), 730–748. Reid, K.F., Fielding, R.A., 2012. Skeletal muscle power: a critical determinant of physical functioning in older adults. Exerc. Sport Sci. Rev. 40 (1), 4–12. Reid, K.F., Pasha, E., Doros, G., Clark, D.J., Patten, C., Phillips, E.M., Frontera, W.R., Fielding, R.A., 2014. Longitudinal decline of lower extremity muscle power in healthy and mobility-limited older adults: influence of muscle mass, strength, composition, neuromuscular activation and single fiber contractile properties. Eur. J. Appl. Physiol. 114 (1), 29–39. Rhea, M.R., Oliverson, J.R., Marshall, G., Peterson, M.D., Kenn, J.G., Ayllón, F.N., 2008. Noncompatibility of power and endurance training among college baseball players. J. Strength Cond. Res. 22 (1), 230–234. Robinson, J.M., Stone, M.H., Johnson, R.L., Penland, C.M., Warren, B.J., Lewis, R.D., 1995. Effects of different weight training exercise/rest intervals on strength, power, and high intensity exercise endurance. J. Strength Cond. Res. 9 (4), 216–221. Royall, D.R., Palmer, R., Chiodo, L.K., Polk, M.J., 2004. Declining executive control in normal aging predicts change in functional status: the freedom house study. J. Am. Geriatr. Soc. 52 (3), 346–352. Saez de Asteasu, M.L.S., Martínez-Velilla, N., Zambom-Ferraresi, F., Casas-Herrero, Á., Izquierdo, M., 2017. Role of physical exercise on cognitive function in healthy older adults: a systematic review of randomized clinical trials. Ageing Res. Rev. 37, 117–134. Singh-Manoux, A., Kivimaki, M., Glymour, M.M., Elbaz, A., Berr, C., Ebmeier, K.P., Ferrie, J.E., Dugravot, A., 2012. Timing of onset of cognitive decline: results from Whitehall II prospective cohort study. BMJ 344 (7622), 1–8. Smiley-Oyen, A.L., Lowry, K.A., Francois, S.J., Kohut, M.L., Ekkekakis, P., 2008. Exercise, fitness, and neurocognitive function in older adults: the “selective improvement” and “cardiovascular fitness” hypotheses. Ann. Behav. Med. 36 (3), 280–291. Smith, P.J., Blumenthal, J.A., Hoffman, B.M., Cooper, H., Strauman, T.A., Welsh-Bohmer, K., Browndyke, J.N., Sherwood, A., 2010. Aerobic exercise and neurocognitive performance: a meta-analytic review of randomized controlled trials. Psychosom. Med. 72 (3), 239–252. Snowden, M., Steinman, L., Mochan, K., Grodstein, F., Prohaska, T.R., Thurman, D.J., Brown, D.R., Laditka, J.N., Soares, J., Zweiback, D.J., Anderson, L.A., 2011. Effect of exercise on cognitive performance in community-dwelling older adults: review of intervention trials and recommendations for public health practice and research. J. Am. Geriatr. Soc. 59 (4), 704–716. Strassnig, M.T., Signorile, J.F., Potiaumpai, M., Romero, M.A., Gonzalez, C., Czaja, S., Harvey, P.D., 2015. High velocity circuit resistance training improves cognition, psychiatric symptoms and neuromuscular performance in overweight outpatients with severe mental illness. Psychiatry Res. 229 (1–2), 295–301. Sweet, T.W., Foster, C., Mcguigan, M.R., Brice, G., 2004. Quantitation of resistance training using the session rating of perceived exertion method. J. Strength Cond. Res. 18 (4), 796–802. Tabata, I., Atomi, Y., Kanehisa, H., Miyashita, M., 1990. Effect of high-intensity endurance training on isokinetic muscle power. Eur. J. Appl. Physiol. Occup. Physiol. 60 (4), 254–258. Tan, B., 1999. Manipulating resistance training program variables to optimize maximum strength in men: a review. J. Strength Cond. Res. 13 (3), 289–304. Tivadar, B.K., 2017. Physical activity improves cognition: possible explanations. Biogerontology 18 (4), 477–483. Tjonna, A.E., Lee, S.J., Rognmo, O., Stolen, T.O., Bye, A., Haram, P.M., Loennechen, J.P., Al-Share, Q.Y., Skogvoll, E., Slordahl, S.A., Kemi, O.J., Najjar, S.M., Wisloff, U., 2008. Aerobic interval versus continuous moderate exercise as a treatment for the metabolic syndrome: a pilot study. Ciculation 118 (4), 346–354. Tsai, C.L., Wang, C.H., Pan, C.Y., Chen, F.C., 2015. The effects of long-term resistance exercise on the relationship between neurocognitive performance and GH, IGF-1, and homocysteine levels in the elderly. Front. Behav. Neurosci. 9 (23), 1–12. Vickers, A.J., Altman, D.G., 2001. Analysing controlled trials with baseline and follow up measurements. BMJ 323 (7321), 1123–1124. Voelcker-Rehage, C., Godde, B., Staudinger, M., 2010. Physical and motor fitness are both realted to cognition in old age. Cogn. Neurosci. 31, 167–176. Voss, M.W., Prakash, R.S., Erickson, K.I., Basak, C., Chaddock, L., Kim, J.S., Alves, A., Heo, S., Szabo, A.N., White, S.M., Wojcicki, T.R., Mailey, E.M., Gothe, N., Olson, E.A., McAuley, E., Kramer, A.F., 2010. Plasticity of brain networks in a randomized intervention trial of exercise training in older adults. Front. Aging Neurosci. 2 (32), 1–17. Weintraub, S., Dikmen, S.S., Heaton, R.K., Tulsky, D.S., Zelazo, P.D., Bauer, P.J., ... Gershon, R.C., 2013. Cognition assessment using the NIH Toolbox. Neurology 80, S54–S64. Weintraub, S., Dikmen, S.S., Heaton, R.K., Tulsky, D.S., Zelazo, P.D., Slotkin, J., ... Gershon, R., 2014. The cognition battery of the NIH Toolbox for assessment of neurological and behavioral function: validation in an adult sample. J. Int. Neuropsychol. Soc. 20 (6), 567–578. Weuve, J., Kang, J.H., Manson, J.E., Breteler, M.B., Ware, J.H., Grodstein, F., 2004. Physical activity, including walking, and cognitive function in older women. JAMA 292 (12), 1454–1461. Wilson, J.M., Marin, P.J., Rhea, M.R., Wilson, S.M., Loenneke, J.P., Anderson, J.C., 2012. Concurrent training: a meta-analysis examining interference of aerobic and resistance exercises. J. Strength Cond. Res. 26 (8), 2293–2307. Young, J., Angevaren, M., Rusted, J., Tabet, N., 2015. Cochrane Database Syst. Rev. 2015 (4), CD005381. https://doi.org/10.1002/14651858.CD005381.pub4.

2013. Enhancing cognitive functioning in the elderly: multicomponent vs resistance training. Clin. Interv. Aging 8, 19–27. Gershon, R., Wagster, M., Hendrie, H., Fox, N., Cook, K., Nowinski, C., 2013. NIH toolbox for the assessment of neurological and behavioral function. Neurology 80 (Suppl. 3), S2–S6. Gligoroska, J.P., Manchevska, S., 2012. The effect of physical activity on cognition—physiological mechanisms. Mater. Sociomed. 24 (3), 198–202. Gothe, N.P., Fanning, J., Awick, E., Chung, D., Wójcicki, T.R., Olson, E.A., Mullen, S.P., Voss, M., Erickson, K.I., Kramer, A.F., McAuley, E., 2014. Executive function processes predict mobility outcomes in older adults. J. Am. Geriatr. Soc. 62 (2), 285–290. Gray, M., Paulson, S., 2014. Developing a measure of muscular power during a functional task for older adults. BMC Geriatr. 14, 145. Gregory, M.A., Gill, D.P., Petrella, R.J., 2013. Brain health and exercise in older adults. Curr. Sports Med. Rep. 12 (4), 256–271. Guiney, H., Machado, L., 2013. Benefits of regular aerobic exercise for executive functioning in healthy populations. Psychon. Bull. Rev. 20 (1), 73–86. Haennel, R., Quinney, A., Kappagoda, T., 1991. Effects of hydraulic circuit training following coronary artery bypass surgery. Med. Sci. Sports Exerc. 23 (2), 158-165. Hafstad, A.D., Boardman, N.T., Lund, J., Hagve, M., Khalid, A.M., Wisloff, U., Larson, T.S., Aasum, E., 2011. High intensity interval training alters substrate utilization and reduces oxygen consumption in the heart. J. Appl. Physiol. 111, 1235–1241. Helgerud, J., Hoydal, K., Wang, E., Karlsen, T., Berg, P., Bjerkaas, M., Simonsen, T., Helgesen, C., Hjorth, N., Bach, R., Hoff, J., 2007. Aerobic high-intensity intervals improve VO2max more than moderate training. Med. Sci. Sports Exerc. 39 (4), 665–671. Hillman, C.H., Erickson, K.I., Kramer, A.F., 2008. Be smart, exercise your heart: exercise effects on brain and cognition. Nat. Rev. Neurosci. 9 (1), 58–65. Izquierdo, M., Cadore, E.L., 2014. Muscle power training in the institutionalized frail: a new approach to counteracting functional declines and very late-life disability. Curr. Med. Res. Opin. 30 (7), 1385–1390. Janssen, E., 2015. Validation of the Tendo Weightlifting Analyzer as a Method to Assess Muscular Power (Unpublished Thesis). University of Arkansas, Fayetteville. Johnson, J.K., Lui, L., Yaffe, K., 2007. Executive function, more than global cognition, predicts functional decline and mortality in elderly women. J. Gerontol. A Biol. Sci. Med. Sci. 62 (10), 1134–1141. Kelly, M.E., Loughrey, D., Lawlor, B.A., Robertson, I.H., Walsh, C., Brennan, S., 2014. The impact of exercise on the cognitive functioning of healthy older adults: a systematic review and meta-analysis. Ageing Res. Rev. 16, 12–31. Kemi, O.J., Haram, P.M., Loennechen, J.P., Osnes, J., Skomedal, T., Wisloff, U., Ellingsen, O., 2005. Moderate vs. high exercise intensity: differential effects on aerobic fitness, cardiomyocyte contractility, and endothelial function. Cardiovasc. Res. 67 (2005), 161–172. Kraemer, W.J., Patton, J.F., Gordon, S.E., Harman, E.A., Deschenes, M.R., Reynolds, K.A.T.Y., Newton, R.U., Triplett, N.T., Dziados, J.E., 1995. Compatibility of highintensity strength and endurance training on hormonal and skeletal muscle adaptations. J. Appl. Physiol. 78 (3), 976–989. Kramer, A.F., Hahn, S., Cohen, N.J., Banich, M.T., McAuley, E., Harrison, C.R., Chason, J., Vakil, E., Bardell, L., Boileau, R.A., Colcombe, A., 1999. Ageing, fitness and neurocognitive function. Nature 400 (6743), 418–419. Kramer, A.F., Colcombe, S.J., McAuley, E., Scalf, P.E., Erickson, K.I., 2005. Fitness, aging and neurocognitive function. Neurobiol. Aging 26 (1), 124–127. Leyva, A., Balachandran, A., Britton, J.C., Eltoukhy, M., Kuenze, C., Myers, N.D., Signorile, J.F., 2017. The development and examination of a new walking executive function test for people over 50 years of age. Physiol. Behav. 171, 100–109. Liu-Ambrose, T., Nagamatsu, L.S., Graf, P., Beattie, B.L., Ashe, M.C., Handy, T.C., 2010. Resistance training and executive functions: a 12-month randomized controlled trial. Arch. Intern. Med. 170 (2), 170–178. Lojovich, J.M., 2010. The relationship between aerobic exercise and cognition: is movement medicinal? J. Head Trauma Rehabil. 25 (3), 184–192. McKinnon, N.B., Montero-Odasso, M., Doherty, T.J., 2015. Motor unit loss is accompanied by decreased peak muscle power in the lower limb of older adults. Exp. Gerontol. 70, 111–118. Miszko, T.A., Cress, M.E., Slade, J.M., Covey, C.J., Agrawal, S.K., Doerr, C.E., 2003. Effect of strength and power training on physical function in community-dwelling older adults. J. Gerontol. A Biol. Sci. Med. Sci. 58 (2), M171–M175. North, T.C., McCullagh, P., Tran, Z.V., 1990. Effect of exercise on depression. Exerc. Sport Sci. Rev. 18 (1), 379-416. Nouchi, R., Taki, Y., Takeuchi, H., Sekiguchi, A., Hashizume, H., Nozawa, T., Nouchi, H., Kawashima, R., 2014. Four weeks of combination exercise training improved executive functions, episodic memory, and processing speed in healthy elderly people: evidence from a randomized controlled trial. Age 36 (2), 787–799. Oltmanns, J., Godde, B., Winneke, H., Richter, G., Niemann, C., Voelcker-Rehage, C., Schömann, K., Staudinger, M., 2017. Don't lose your brain at work — the role of recurrent novelty at work in cognitive and brain aging. Front. Psychol. 8 (117), 1–16. Orr, R., De Vos, N.J., Singh, N.A., Ross, D.A., Stavrinos, T.M., Fiatarone-Singh, M.A., 2006. Power training improves balance in healthy older adults. J. Gerontol. A Biol. Sci. Med. Sci. 61 (1), 78–85. Potiaumpai, M., Gandia, K., Rautray, A., Prendergast, T., Signorile, J.F., 2016. Optimal loads for power differ by exercise in older adults. J. Strength Cond. Res. 30 (10), 2703–2712. Raz, N., Rodrigue, K.M., 2006. Differential aging of the brain: patterns, cognitive

87