Cardiorespiratory Fitness and Cognitive Function are Positively Related Among Participants with Mild and Subjective Cognitive Impairment

Abstract

Background:

By 2030, about 74 million people will be diagnosed with dementia, and many more will experience subjective (SCI) or mild cognitive impairment (MCI). As physical inactivity has been identified to be a strong modifiable risk factor for dementia, exercise and physical activity (PA) may be important parameters to predict the progression from MCI to dementia, but might also represent disease trajectory modifying strategies for SCI and MCI.

Objective:

A better understanding of the relationship between activity, fitness, and cognitive function across the spectrum of MCI and SCI would provide an insight into the potential utility of PA and fitness as early markers, and treatment targets to prevent cognitive decline.

Methods:

121 participants were stratified into three groups, late MCI (LMCI), early MCI (EMCI), and SCI based on the Montreal Cognitive Assessment (MoCA). Cognitive function assessments also included the Trail Making Test A+B, and a verbal fluency test. PA levels were evaluated with an interviewer-administered questionnaire (LAPAQ) and an activity monitor. An incremental exercise test was performed to estimate cardiorespiratory fitness and to determine exercise capacity relative to population normative data.

Results:

ANCOVA revealed that LMCI subjects had the lowest PA levels (LAPAQ, p = 0.018; activity monitor, p = 0.041), and the lowest exercise capacity in relation to normative values (p = 0.041). Moreover, a modest correlation between MoCA and cardiorespiratory fitness (r = 0.25; p < 0.05) was found.

Conclusion:

These findings suggest that during the earliest stages of cognitive impairment PA and exercise capacity might present a marker for the risk of further cognitive decline. This finding warrants further investigation using longitudinal cohort studies.

Keywords

Cardiorespiratory fitness mild cognitive impairment Montreal Cognitive Assessment physical activity subjective cognitive impairment

INTRODUCTION

Worldwide, over 46 million people live with dementia, and this is expected to rise to about 74 million by 2030 [1]. The prevalence of mild cognitive impairment (MCI), which is a risk state for dementia [2], is about 4% in community-dwelling older adults aged between 50–80 years [3]. As such, cognitive decline, dementia, and Alzheimer’s disease (AD), add to the significant economic impact of an aging population [1], and are identified as global health and healthcare priorities [4]. There is growing recognition of the importance of the early detection of cognitive decline, such as subjective cognitive impairment (SCI), which is associated with an increased risk to progress to MCI and dementia [5, 6]. There has also been a shift in research efforts to identify possibly reversible risk factors that are associated with, and present during, the stages of SCI [7] and MCI [2].

Various modifiable risk factors (e.g., physical inactivity, mid-life hypertension) have been identified as potentially contributing to the development of AD [8]. Of these, physical inactivity has been identified as the strongest independent risk factor, accounting for up to 12.7% of the risk of AD [8, 9]. Furthermore, an active lifestyle and the adoption of physical activity (PA) during mid-life (35 to 64 years) and even after the age of 65 years maintain or improve cognitive function in ageing and reduce the risk of AD [10 –14]. A longitudinal study reported that participants with MCI are more likely to develop dementia after 7 years if they reported being previously inactive [15]. This indicates that differences in daily PA levels among participants with MCI might be an important parameter to predict further cognitive decline.

To date there is limited data available on the daily PA levels of participants with MCI and SCI. While cognitive impairment leads to less activity on average compared with healthy older adults, this finding does not take into account the broad range of activity levels, which ranges from no PA up to relatively high levels of PA among individuals with SCI [16] and MCI [17]. Previous studies have often been limited by the use of indirect methods of PA assessment [11 , 15–17]. Furthermore, the relationship between PA and cognitive function in participants with MCI or SCI is not clear. Some studies reported no correlations between PA and cognitive function [12, 16], whereas another study reported a weak relationship between daily leisure-time PA, and cognitive function [18]. A better understanding of the relationship between activity and cognition, particularly among those in the earliest stages of cognitive decline, is important to establish if PA might be a potential risk factor and treatment target.

Exercise training and increased levels of PA are associated with an improved cardiorespiratory fitness. Cardiorespiratory fitness may be beneficial for maintaining cognitive function in those at an increased risk for AD [19, 20]. In subjects with AD, a higher cardiorespiratory fitness might improve some aspects of cognitive function (e.g., sustained attention, visual memory) [21, 22]. Recently, it has been reported that maximal exercise capacity during mid-life (mean age 59 y), which provides an estimate of cardiorespiratory physical fitness [23, 24], is associated with cognitive function in later life (mean age 77 y) [25]. To date the influence of SCI and MCI on exercise capacity and the relationship with cognitive function has not been determined.

A better understanding of the relationship between PA, cardiorespiratory fitness, and cognitive function during the early stages of cognitive decline would provide an insight into the potential utility of PA and fitness as early markers, and treatment targets. Therefore, the primary aim of the present study was to compare PA levels, exercise capacity and cardiorespiratory fitness between older adults across the spectrum of SCI and MCI, and to assess the strength of the relationship between these physical characteristics and cognitive function. It is hypothesized that lower levels of PA and of exercise capacity are reflected by lower cognitive performance levels across the spectrum of SCI and MCI.

MATERIALS AND METHODS

Participants

All participants were recruited through the NeuroExercise Project [26], a multi-centered randomized controlled trial of exercise therapy in participants with MCI across three European countries. For the purpose of the present sub-study, the participants were recruited in Germany at the German Sport University (GSU). The study was conducted in accordance with the declaration of Helsinki (1975) and approved by the research ethics committee of the GSU. Participants were recruited through newspaper advertisements and editorials. All participants provided their informed consent to the study procedures.

Participants were initially interviewed via telephone and were required to meet the following eligibility criteria, which has been published elsewhere [26]. For the purpose of this study, persons with a MOCA score >25, who reported memory impairments but did not meet the clinical criteria for MCI used in the NeuroExercise Project, were included in this substudy as participants with SCI. 121 participants met the inclusion criteria and were stratified into three different groups depending on their MoCA scores. Group 1 included those with late mild cognitive impairment (LMCI), group 2 included participants with early MCI (EMCI), and group 3 included participants with SCI (Table 1).

Table 1

Characteristics of the total sample and subgroups

	Group	LMCI (MoCA 19–21)	EMCI (MoCA (22–25)	SCI (MoCA >25)
Sample Size^#	121	21	64	36
Age (years)^*	72.6±6.3	75.0±5.1	73.6±5.97	69.3^**±6.39
Sex	64 f	6 f	35 f	23 f
Education (years)⁺	2.7±0.58	2.7±0.57	2.7±0.65	2.7±0.45
MoCA^*** (mean scores)	24.0±2.4	20.2^***±0.8	23.7^***±1.0	27.0^***±0.9
MoCA-MIS^*** (mean scores)	7.79±3.9	4.4^***±2.8	7.4^***±3.6	10.5^***±3.3
calculated MMSE^** (mean scores)*	28.15^***±1.2	26.24^***±0.83	28.11^***±0.62	29.33^***±0.48

Data are mean±SD. ⁺Education level was assessed using categorical levels (1 = less than 10 years of education; 2 = between 10–13 years of education, 3 = more than 13 years of education). MoCA = Montreal Cognitive Assessment; calculated MMSE = Mini-Mental State Examination, scores were calculated based on the individual MoCA score; f = female; ^* = significant difference between the groups; ^** = significantly younger than the other groups (p < 0.05); ^*** = significant differences between all of the groups (p < 0.001); LMCI = late mild cognitive impairment; EMCI = early mild cognitive impairment; SCI = subjective cognitive impairment; ^# = data for the activity monitor and the incremental exercise test could not be generated for all participants (see missing data).

Study overview

The participants attended two appointments for the assessment of cognitive function, exercise capacity, and daily activity levels. During their first visit the participants underwent a neuropsychological test battery. This consisted of five different tests: the Montreal Cognitive Assessment (MoCA) [27], the Trail Making Test (TMT) A and B [28], and two tests for verbal fluency [29]. The Longitudinal Aging Study Amsterdam Physical Activity Questionnaire (LAPAQ) [30] was used to assess daily PA levels. At the end of their first visit the participants were instructed to wear a watch with an activity monitor (Polar M400^©, USA) for the assessment of PA over seven consecutive days. During the second visit, the participants performed an incremental exercise test on a cycle ergometer (Ergoline er900^©, Bitz, Germany) for the determination of maximal exercise capacity and the estimation of cardiorespiratory fitness (VO₂peak).

Cognitive function assessment

The MoCA is a 12-item test with scores from 0 to 30, where higher scores indicate better cognitive function. MoCA scores from 17 to 25 or from 19 to 25 have previously been used to delineate MCI [27, 31], although test sensitivity has been shown to be improved using lower cut-off scores of 22-23 [32 –34]. To enable comparisons between participants across the spectrum of SCI and MCI, we adopted the lower cut-off score and stratified the sample into three groups (see Participants): 1) LMCI (late MCI, score 19–21) [34]; 2) EMCI (early MCI, score 22–25) [28]; 3) SCI (SCI, score >25).

Besides MoCA total score, the MoCA memory index score (MoCA-MIS) was calculated to evaluate verbal episodic memory, which is a sensitive factor to discriminate between early and late MCI [35, 36]. The MoCA-MIS is calculated by adding the number of words in free delayed recall, category-cued recall, and multiple choice-cued recall. The number of words is multiplied by 3, 2 and 1, respectively, with a total MoCA-MIS between 0–15 [37]. Furthermore, we used the equi-percentile equating method with log-linear smoothing to calculate Mini-Mental State Examination (MMSE) scores for our participants based on the individual MoCA scores [31, 38]. The MMSE is a common screening tool used by many studies with SCI and MCI. Although the MMSE lacks sensitivity and specificity to detect MCI in comparison to the MoCA [27, 31], it is still widely used to assess cognitive function and will therefore allow further comparisons to other studies in the field of interest.

The TMT A and B are validated neuropsychological tests, which assess speed of processing and executive function [28]. In Part A, the participants were required to connect numbers in ascending order from 1 to 25. In Part B, the participants were instructed to connect numbers and letters in alternating and ascending order (e.g., 1 – A – 2 – B – 3 – C, etc.). A mistake made by the participants was immediately pointed out by the test administrator and was corrected before proceeding. Both tests were completed as fast as possible and the time taken to complete the tests was measured.

The verbal fluency test is a short test of verbal functioning, which typically consists of two tasks [29]. These tasks are letter fluency [39], and category fluency [40]. For both tasks participants were given 1 minute to produce as many unique words as possible either starting with a given letter (letter fluency) or unique words within a semantic category (category fluency). The participant’s score was the number of unique correct words in each task. All cognitive function assessments were administered in-person by a trained researcher and supervised by an experienced neuropsychologist.

Physical activity assessment

Self-reported PA was assessed using the LAPAQ, which is a valid and reliable interview-administered questionnaire, and captures PA across six categories (walking outdoors, bicycling, gardening, light household activities, heavy household activities, and sport and exercise activities) over the preceding 14 days [30]. Mean daily activity scores, and mean time spent in sport and exercise activities were calculated by summing the reported activities in minutes and dividing those by the number of days.

Daily PA levels were also objectively assessed using the activity watch. The GPS watch is equipped with a triaxial accelerometer, which uses wrist movements to calculate distance walked per day (in km) as an objective measure of PA [41, 42]. The activity monitor was worn on the non-dominant arm for seven consecutive days to sample behavior on both week and weekend days [41]. The participants were asked to wear the activity monitor 24 h a day. Only days with ≥10 h of captured data, and only data-sets with ≥7 days of valid data were included for analysis [41]. Distance walked per day (in km) has been shown to be validly measured by the activity monitor [43] and demonstrates a proxy for steps per day [42].

Exercise capacity and estimated cardiorespiratory fitness

An incremental exercise test was performed on an electronically braked cycle ergometer (Ergoline, type er900^©, Bitz, Germany) under the supervision of a cardiologist. The progressive exercise test commenced with 3 min of unloaded cycling followed by step increments of 25W every 2 min in accordance with clinical guidelines [44]. 12-lead electrocardiography (ECG) and heart rate (HR) were monitored throughout the test. Blood lactate levels from earlobe capillary samples, and rating of perceived exertion (RPE) [45] were assessed during the final 30 s of each stage. The participants continued to exercise until their volitional maximum and to the point where they were unable to maintain the required pedal cadence with increasing exercise intensity.

Exercise capacity was defined as the peak power (W) reached at final stage and was used to estimate cardiorespiratory fitness using the following equation (VO₂peak = (exercise capacity (W)/weight (kg)) * 10.8 + 3.5 + 3.5) [46]. To determine exercise capacity relative to population normative data, we calculated reference scores for exercise capacity using the following equation from a cross-sectional epidemiologic survey of the local population in Germany (Males: reference exercise capacity (W) = –103.512–1.5576 * Age + 2.2114 * Height –0.1198 * Weight. Females: reference exercise capacity (W) = –80.628–0.7698 * Age + 1.4038 * Height + 0.2873 * Weight) [47, 48]. Exercise capacity was subtracted from the reference score. This parameter is referred to as relative exercise capacity.

Missing data

121 participants completed the neuropsychological test battery and the LAPAQ during their first visit to the GSU. Valid PA monitor data sets were collected for 84 of the 121 participants, with many participants unfortunately failing to wear the activity watch for the required daily periods. The incremental exercise test was completed sufficiently by 86 of the 121 participants. 12 participants were not given clearance by the cardiologist to undertake the test due to uncontrolled hypertension or irregularities in their resting ECG; for 7 participants, the test was terminated prior to their maximum because of adverse blood pressure, ECG or symptomatic responses; and in 16 cases the participants did not meet the criteria for maximum effort. There were no differences in baseline characteristics and cognitive function between those with complete and incomplete data sets within each of the study subgroups, and therefore all data have been included for analysis.

Statistical analysis

Data were analyzed using IBM SPSS Statistics 23^©. Variables pertaining to cognitive function (TMT A + B, letter and category fluency), the PA assessment (LAPAQ + activity monitor), exercise capacity and estimated cardiorespiratory fitness (exercise capacity, estimated VO₂peak, RPE, heart rate, blood lactate levels, and relative exercise capacity) were compared between the three groups by one-way univariate ANCOVA analysis with age and sex as covariates. Post-hoc pairwise comparisons between groups were performed using Fisher’s least significant difference (LSD). Additionally, Spearman correlation was used to determine the strength of relationships between estimated VO₂peak, relative exercise capacity and the MoCA, as well as its memory index score (MoCA-MIS). Furthermore, Pearson’s correlation was used to assess correlations between the PA assessment, VO₂peak and relative exercise capacity. Data are presented as means±standard deviations, and p values <0.05 were regarded as statistically significant.

RESULTS

Cognitive function

Group differences were found for category fluency (F_(2,119) = 10.355; p < 0.001), TMT A (F_(2,120) = 11.690; p < 0.001), and TMT B (F_(2,120) = 14.988; p < 0.001), but not for letter fluency (F_(2,120) = 2.955; p = 0.56). Post-hoc pairwise comparisons revealed lower cognitive function for category fluency in LMCI than EMCI and SCI. Additionally, post-hoc comparisons showed that EMCI had lower cognitive function than SCI in category fluency (Fig. 1). Significant differences were also found for the TMT A and TMT B test where LMCI needed significantly longer to complete both Trail Making tests than the other two groups. Furthermore, EMCI needed significantly more time to complete the TMT A than SCI (Fig. 1).

Fig.1

Results of the cognitive function assessment; (^*= p < 0.05, ^**= p < 0.01, ^***= p < 0.001), LMCI = late mild cognitive impairment, EMCI = early mild cognitive impairment, SCI = subjective cognitive impairment. A) Verbal fluency tests (category + letter fluency). B) Results of the Trail Making Test (TMT) A + B (total means were divided by 10 for a clear depiction in this figure).

Physical activity assessment

Differences between the groups were found for mean daily activity scores (LAPAQ, F_(2,120) = 4.169; p = 0.018), and average distance walked per day (activity monitor, F_(2,83) = 3.878; p = 0.025) but not for mean time spent in sports and exercise activities (F_(2,120) = 2.885; p = 0.60, subcategory of the LAPAQ). Post-hoc pairwise comparisons revealed that LMCI had lower mean daily activity scores (LAPAQ) than EMCI (p = 0.006) and SCI (p = 0.036). Furthermore, LMCI had a lower average distance walked per day (4.4±2.4 km) in comparison to EMCI (6.4±2.7 km, p = 0.007), but not SCI (5.9±2.5 km p = 0.089). Means and standard deviations are presented in Table 2.

Table 2

Data of the physical activity and physical fitness assessment

Group	LMCI	EMCI	SCI
Exercise capacity (W)	94.74±37.8	104.02±36.6	114.58±44.5
Estimated VO₂peak (ml kg-1 min-1)	20.2±5.6	22.6±4.8	23±5.1
Rating of perceived exertion	16.5±2.4	16.7±2.5	17.7±2.3
Heart rate (bpm)	123.4±26.7	126.5±24.8	134.7±21.1
Blood lactate (mmol)	4±1.9	4.7±5.9	4.5±2.0
Relative exercise capacity (W)	–46.38^*, #±39.16	–24.18±25.20	–20.9±21.02
LAPAQ (min)	163.57^*, #±135.95	270.09±157.09	239.78±153.10
LAPAQ – Sports &Exercises (min)	12.79±21.24	24.27±29.78	14.42±13.98
Activity monitor	4.4^*±2.4	6.4±2.7	5.9±2.5
Average distance walked per day (km)

The data presents peak data of the incremental exercise test, mean daily activity levels (LAPAQ), mean daily time spent in sports and exercises, and mean distance walked per day; ^* = significant difference to EMCI (p < 0.05); ^# = significant difference to SCI (p < 0.05); LAPAQ = Longitudinal Aging Study Amsterdam Physical Activity Questionnaire, LMCI = late mild cognitive impairment, EMCI = early mild cognitive impairment, SCI = subjective cognitive impairment; km = kilometer.

Exercise capacity and estimated cardiorespiratory fitness

No differences were found for exercise capacity (F_(2,85) = 0.192; p = 0.825), VO₂peak (F_(2,85) = 1.009; p = 0.369), RPE (F_(2,85) = 1.477; p = 0.236), HR (F_(2,85) = 0.117; p = 0.890), and blood lactate level (F_(2,84) = 0.631; p = 0.535). Relative exercise capacity differed significantly between the groups (F_(2,85) = 3.321; p = 0.041), with post-hoc test showing LMCI to have a lower relative exercise capacity than EMCI and SCI. The detailed results (means and standard deviations) can be seen in Table 2.

Correlations

Spearman correlation revealed a significant correlation between the MoCA scores and estimated VO₂peak (r = 0.245, p = 0.022), as well as between the MoCA and relative exercise (r = 0.228, p = 0.035) capacity. Indicating that a higher cardiorespiratory fitness and better exercise capacity were reflected by a better cognitive performance (Fig. 2). Furthermore, a positive correlation was found between estimated VO₂peak and average distance walked per day (r = 0.266, p = 0.027), but not between mean daily activity levels (LAPAQ) and VO₂peak (p = 0.600) or between VO₂peak and mean time spent in sports and exercise activities (p = 0.545, subcategory of the LAPAQ). Relative exercise capacity had a modest positive correlation with average distance walked per day (r = 0.321; p = 0.008), but not with the outcomes of the self-reported activity questionnaire (p = 0.384) or its subcategory (p = 0.651).

Fig.2

A) Correlation between estimated VO₂peak and cognitive function (Montreal Cognitive Assessment (MoCA) score; B) Correlation between relative exercise capacity and cognitive function (MoCA score).

DISCUSSION

We assessed PA levels and exercise capacity across a group of participants with MCI and SCI. The main finding of this study was that a greater cognitive impairment (LMCI) was reflected by significantly lower PA levels and a lower exercise capacity across the spectrum of SCI and MCI. Furthermore, we observed a modest correlation between cognitive function and estimated VO₂peak, as well as relative exercise capacity, but not between self-reported daily activity levels and cognitive function.

In this study, we compared three groups of participants across the spectrum of SCI and MCI. There is some inconsistency in the reported MoCA cut-off scores for MCI [27, 34], and on this basis, we applied novel criteria to stratify participants to SCI, EMCI, or LMCI. We estimated MMSE scores, which have been previously used to describe participants with EMCI and LMCI [48], using every participant’s individual MoCA score. The estimated MMSE scores were quite similar to the ones published by Lee et al. [49]. Furthermore, the three groups performed significantly different in the Trail Making Tests, the category fluency tests and their MoCA memory scores, which are another sensitive factor to discriminate between early and late MCI [36, 37]. As expected, LMCI demonstrated the worst cognitive function followed by EMCI and then SCI on all tests of the neuropsychological test battery. This follows the definition that participants with EMCI show milder degrees of cognitive impairment than participants with LMCI [50]. Interestingly, our verbal fluency tasks showed varying results, with significant differences among the groups in the category, but not the letter fluency task. Our results extend previous findings, which showed that even between participants with AD and healthy controls often only category fluency differs significantly [51]. We suggest, that differences across the spectrum of SCI and MCI can rather be determined by semantically demanding tasks (category fluency) than phonemic tasks (letter fluency) [52].

Our findings showed that the results of the cognitive tests were mirrored by the results of the PA assessment. LMCI had the lowest activity levels (LAPAQ and activity monitor) compared to the other two groups. This extends previous findings that activity levels are lower in people with MCI compared with healthy adults, and lower again in people with AD [17]. Gagliardi et al. reported a wide range of PA levels among individuals with MCI [17], and our findings show that this range of activity levels is related to cognitive function. It has previously been shown that low levels of engagement in PA are associated with the development of AD among people with MCI over a seven year follow up period [15]. Our findings raise the possibility that physical inactivity might also be a contributing risk factor for the progression from SCI and EMCI to LMCI.

Previous studies of PA among people with MCI and healthy older adults have usually relied on self-reported activity recall questionnaires [15 , 53, 54]. These subjective methods are open to reporting bias, and older adults aged 65–84 years tend to overestimate PA time in comparison to objectively measured accelerometer data [55]. Aadahl et al. [56] pointed out that the so-called social desirability response bias might be responsible for this, because of the known benefits of PA and exercise in the population. Therefore, we suggest using objective measurement devices to assess PA. The use of the activity monitor was a strength of the current study, although it is acknowledged that use of an activity monitor also has the potential to positively influence activity behaviors [57].

In line with their lower activity levels, we observed that the LMCI group had a lower relative exercise capacity than the EMCI and SCI groups and tended to have a lower estimated cardiorespiratory fitness. The use of relative exercise capacity presented an opportunity to report differences in exercise capacity between the groups relative to normative values, and thereby take into account each participant’s age, sex, height, and weight, which influence exercise capacity and cardiorespiratory fitness [47 , 59]. Indeed, small differences in age and sex distribution between the groups may explain why we did not find significant differences in estimated VO₂peak between the groups. Nonetheless, it is noteworthy that none of the groups in our study met the normative exercise capacity levels for healthy older adults of their age [46, 47]. This is also consistent with the low VO₂peak values that ranged from 20.2 ml kg-1 min-1 (LMCI) to 23.0 ml/min (SCI), which are lower than population reference values [60, 61]. Exercise capacity and cardiorespiratory fitness are strongly influenced by exercise and PA, and indeed we found a modest correlation between the average distance walked per day (activity monitor) and estimated VO₂peak (r = 0.266, p = 0.027), as well as between the distance walked per day and relative exercise capacity (r = 0.321; p = 0.008). That this relationship was not present for the self-reported daily PA levels reinforces the importance of objective PA assessment.

An important finding was the modest correlation between cognitive function (MoCA) and estimated cardiorespiratory fitness, as well as between cognitive function (MoCA) and the relative exercise capacity. This adds to previous evidence that the large differential in cognitive function between healthy adults and those with MCI is associated with fitness [19 , 62, 63]. Our findings show that this relationship is also present among a cohort ranging from SCI to MCI. Observational evidence in humans associates higher levels of cardiorespiratory fitness and PA with greater brain volume, reduced brain atrophy, and the reduced risk and slower progression of dementia [11 , 64–68]. An increase in cardiorespiratory fitness has also been associated with an attenuation of the effects of cerebral amyloid on cognition [19, 63]. Additionally, PA and exercise are known to improve neuroplasticity by the expression of brain derived factor [69, 70] and insulin-like growth factor [71], and increase hippocampal volume [72]. Combined with this evidence, our cross-sectional observations support the notion that cardiorespiratory fitness may be a sensitive marker of cognitive decline, and the progression from SCI to a more severe presentation of MCI. Unlike previous findings, which reported improved spatial memory after aerobic exercise [72], we did not find a correlation between the memory index of the MoCA and cardiorespiratory fitness. We suggest that spatial memory might be correlated to aerobic exercise, but that this might not be the case for verbal episodic memory (MoCA-MIS).

It is tempting to speculate that higher levels of cardiorespiratory fitness, and other factors such as improved neural connectivity [73], might underpin the association between cognitive function and PA observed in the present study. In contrast, we cannot exclude the possibility that the progression of cognitive decline leads to less engagement in PA, e.g., due to a loss of independence and/or overprotection by the individuals’ families and carer [74]. Therefore, we cannot exclude a certain threshold effect due to different stages of cognitive impairment (LMCI, EMCI, SCI). Further longitudinal studies are needed to confirm these relationships, and to investigate whether prescribed changes in PA among those with SCI and MCI have positive effects on cognitive function and/or the progression of impairment.

Study limitations

The present study has some limitations. Firstly, the assessment period for the LAPAQ and the PA monitoring were not matched, and therefore comparisons of the two measures is somewhat compromised. However, all participants reported that they undertook their normal activities during the week where PA was assessed using the monitor. Secondly, there was a small but significant difference in the age of group participants (Table 1) where those with SCI were younger than the other two groups. While this may have biased the results, as age affects cognitive performance and PA [75], we calculated an ANCOVA to statistically control for the effect of age. Still, this cannot completely neglect a significant influence of age. Thirdly, cardiorespiratory fitness was estimated based on the measure of exercise capacity. This was done so as to avoid the need for the collection of expired gasses during exercise, which can sometimes cause anxiety among older participants with cognitive impairment. This estimation assumes a fixed level of oxygen economy for all participants, and has been reported to overestimate VO₂peak in older subjects [76, 77]. Fourthly, we assessed EMCI and LMCI with the use of the MoCA and estimated MMSE scores. In our opinion, the MoCA is the better tool to detect early cognitive impairments, because its ceiling effect is less and its sensitivity and specificity higher in the detection of mild cognitive impairment [27 , 38].

Conclusion

In conclusion, our results showed that within the spectrum of early cognitive decline, the most severely impaired participants were less engaged in PA and had a lower relative exercise capacity. Furthermore, we found a positive relationship between cardiorespiratory fitness and cognitive function across the stages of mild and subjective cognitive impairment. These findings may suggest that PA and especially cardiorespiratory fitness might be sensitive to the changes in cognitive function associated with subjective and mild cognitive impairment. However, it is also possible that people with MCI might select to be less active and are therefore less fit. This cannot be answered by our study. Further research that aims to determine the effect of changes in PA and cardiorespiratory fitness among those with SCI and MCI on cognitive function and/or the progression of impairment are warranted.

Footnotes

ACKNOWLEDGMENTS

This study was supported by The EU Joint Programme – Neurodegenerative Disease Research (JPND) and the Federal Ministry of Education and Research of Germany (BMBF; grant number: BMBF 01ED1510A). Some aspects of the work were supported by the Universities-Australia and German Academic Exchange Service (DAAD) cooperation scheme. Furthermore, Tim Stuckenschneider is supported by a DAAD Scholarship.

The authors would like to thank all the participants who have been involved and dedicated their precious time to this study. We would like to thank three unknown reviewers for their valuable feedback on a previous version of the manuscript.

Authors’ disclosures available online ().

References

Prince

, Wimo

, Guerchet

, Ali

, Wu

, Prina

(2015) The Global Impact of Dementia: An analysis of prevalence incidence, cost and trends, Alzheimer’s Disease International (ADI), London.

Petersen

, Roberts

, Knopman

, Boeve

, Geda

, Ivnik

, Smith

, Jack

Jr (2009) Mild cognitive impairment: Ten years later. Arch Neurol 66, 1447–1455.

Dlugaj

, Weimar

, Wege

, Verde

, Gerwig

, Dragano

, Moebus

, Jockel

, Erbel

, Siegrist

, Heinz Nixdorf Recall Study Investigative G (2010) Prevalence of mild cognitive impairment and its subtypes in the Heinz Nixdorf Recall study cohort. Dement Geriatr Cogn Disord 30, 362–373.

Kirton

(2013) A summit of significant success: Prospects for theG8 leaders at Lough Erne. G8 Research Group, http://www.g8.utoronto.ca/evaluations/2013lougherne/kirton-prospects-2013.html

Donovan

, Amariglio

, Zoller

, Rudel

, Gomez-Isla

, Blacker

, Hyman

, Locascio

, Johnson

, Sperling

, Marshall

, Rentz

(2014) Subjective cognitive concerns and neuropsychiatric predictors of progression to the early clinical stages of Alzheimer disease. Am J Geriatr Psychiatry 22, 1642–1651.

Fernandez-Blazquez

, Avila-Villanueva

, Maestu

, Medina

(2016) Specific features of subjective cognitive decline predict faster conversion to mild cognitive impairment. J Alzheimers Dis 52, 271–281.

Stewart

(2012) Subjective cognitive impairment. Curr Opin Psychiatry 25, 445–450.

Norton

, Matthews

, Barnes

, Yaffe

, Brayne

(2014) Potential for primary prevention of Alzheimer’s disease: An analysis of population-based data. Lancet Neurol 13, 788–794.

Barnes

, Yaffe

(2011) The projected effect of risk factor reduction on Alzheimer’s disease prevalence. Lancet Neurol 10, 819–828.

10.

Ngandu

, Lehtisalo

, Solomon

, Levalahti

, Ahtiluoto

, Antikainen

, Backman

, Hanninen

, Jula

, Laatikainen

, Lindstrom

, Mangialasche

, Paajanen

, Pajala

, Peltonen

, Rauramaa

, Stigsdotter-Neely

, Strandberg

, Tuomilehto

, Soininen

, Kivipelto

(2015) A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): A randomised controlled trial. Lancet 385, 2255–2263.

11.

Tolppanen

, Solomon

, Kulmala

, Kareholt

, Ngandu

, Rusanen

, Laatikainen

, Soininen

, Kivipelto

(2015) Leisure-time physical activity from mid- to late life, body mass index, and risk of dementia. Alzheimers Dement 11, 434–443.e6.

12.

Ahlskog

, Geda

, Graff-Radford

, Petersen

(2011) Physical exercise as a preventive or disease-modifying treatment of dementia and brain aging. Mayo Clin Proc 86, 876–884.

13.

Bowen

(2012) A prospective examination of the relationship between physical activity and dementia risk in later life. Am J Health Promot 26, 333–340.

14.

Matura

, Fleckenstein

, Deichmann

, Engeroff

, Fuzeki

, Hattingen

, Hellweg

, Lienerth

, Pilatus

, Schwarz

, Tesky

, Vogt

, Banzer

, Pantel

(2017) Effects of aerobic exercise on brain metabolism and grey matter volume in older adults: Results of the randomised controlled SMART trial. Transl Psychiatry 7, e1172.

15.

Bidzan

, Bidzan

, Pachalska

(2016) The effects of intellectual, physical, and social activity on further prognosis in mild cognitive impairment. Med Sci Monit 22, 2551–2560.

16.

Miyawaki

, Bouldin

, Kumar

, McGuire

(2017) Associations between physical activity and cognitive functioning among middle-aged and older adults. J Nutr Health Aging 21, 637–647.

17.

Gagliardi

, Papa

, Postacchini

, Giuli

(2016) Association between cognitive status and physical activity: Study profbaselile on ine survey of the My Mind Project. Int J Environ Res Public Health 13, E585.

18.

de Souto Barreto

, Andrieu

, Rolland

, Vellas

, DSA MAPT Study Group (2018) Physical activity domains and cognitive function over three years in older adults with subjective memory complaints: Secondary analysis from the MAPT trial. J Sci Med Sport 21, 52–57.

19.

Schultz

, Boots

, Darst

, Zetterberg

, Blennow

, Edwards

, Koscik

, Carlsson

, Gallagher

, Bendlin

, Asthana

, Sager

, Hogan

, Hermann

, Cook

, Johnson

, Engelman

, Okonkwo

(2017) Cardiorespiratory fitness alters the influence of a polygenic risk score on biomarkers of AD. Neurology 88, 1650–1658.

20.

Mavros

, Gates

, Wilson

, Jain

, Meiklejohn

, Brodaty

, Wen

, Singh

, Baune

, Suo

, Baker

, Foroughi

, Wang

, Sachdev

, Valenzuela

, Fiatarone Singh

(2017) Mediation of cognitive function improvements by strength gains after resistance training in older adults with mild cognitive impairment: Outcomes of the study of mental and resistance training. J Am Geriatr Soc 65, 550–559.

21.

Morris

, Vidoni

, Johnson

, Van Sciver

, Mahnken

, Honea

, Wilkins

, Brooks

, Billinger

, Swerdlow

, Burns

(2017) Aerobic exercise for Alzheimer’s disease: A randomized controlled pilot trial. PLoS One 12, e0170547.

22.

Maliszewska-Cyna

, Xhima

, Aubert

(2016) A comparative study evaluating the impact of physical exercise on disease progression in a mouse model of Alzheimer’s disease. J Alzheimers Dis 53, 243–257.

23.

Myers

, Kaykha

, George

, Abella

, Zaheer

, Lear

, Yamazaki

, Froelicher

(2004) Fitness versus physical activity patterns in predicting mortality in men. Am J Med 117, 912–918.

24.

Pescatello

, American College of Sports Medicine (2014) ACSM’s guidelines for exercise testing and prescription, Wolters Kluwer/Lippincott Williams & Wilkins Health, Philadelphia.

25.

Muller

, Chan

, Myers

(2017) Association between exercise capacity and late onset of dementia, Alzheimer disease, and cognitive impairment. Mayo Clin Proc 92, 211–217.

26.

Devenney

, Sanders

, Lawlor

, Olde Rikkert

MGM

, Schneider

, NeuroExercise Study Group (2017) The effects of an extensive exercise programme on the progression of Mild Cognitive Impairment (MCI): Study protocol for a randomised controlled trial. BMC Geriatr 17, 75.

27.

Nasreddine

, Phillips

, Bedirian

, Charbonneau

, Whitehead

, Collin

, Cummings

, Chertkow

(2005) The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. J Am Geriatr Soc 53, 695–699.

28.

Tombaugh

(2004) Trail Making Test A and B: Normative data stratified by age and education. Arch Clin Neuropsychol 19, 203–214.

29.

Lezak

, Howieson

, Bigler

, Tranel

(2012) Neuropsychological Assessment, Oxford University Press, New York, NY.

30.

Stel

, Smit

, Pluijm

, Visser

, Deeg

, Lips

(2004) Comparison of the LASA Physical Activity Questionnaire with a 7-day diary and pedometer. J Clin Epidemiol 57, 252–258.

31.

Trzepacz

, Hochstetler

, Wang

, Walker

, Saykin

, Alzheimer’s Disease Neuroimaging Initiative (2015) Relationship between the Montreal Cognitive Assessment and Mini-mental State Examination for assessment of mild cognitive impairment in older adults. BMC Geriatr 15, 107.

32.

Luis

, Keegan

, Mullan

(2009) Cross validation of the Montreal Cognitive Assessment in community dwelling older adults residing in the Southeastern US. Int J Geriatr Psychiatry 24, 197–201.

33.

Lee

, Dong Woo

, Cho

, Na

, Hong Jin

, Kim

, You Ra

, Youn

, Kwon

, Lee

, Cho

(2008) Brief screening for mild cognitive impairment in elderly outpatient clinic: Validation of the Korean version of the Montreal Cognitive Assessment. J Geriatr Psychiatry Neurol 21, 104–110.

34.

Freitas

, Simoes

, Alves

, Santana

(2013) Montreal cognitive assessment: Validation study for mild cognitive impairment and Alzheimer disease. Alzheimer Dis Assoc Disord 27, 37–43.

35.

Aggarwal

, Wilson

, Beck

, Bienias

, Bennett

(2005) Mild cognitive impairment in different functional domains and incident Alzheimer’s disease. J Neurol Neurosurg Psychiatry 76, 1479–1484.

36.

Clement

, Belleville

, Mellah

(2010) Functional neuroanatomy of the encoding and retrieval processes of verbal episodic memory in MCI. Cortex 46, 1005–1015.

37.

Julayanont

, Brousseau

, Chertkow

, Phillips

, Nasreddine

(2014) Montreal Cognitive Assessment Memory Index Score (MoCA-MIS) as a predictor of conversion from mild cognitive impairment to Alzheimer’s disease. J Am Geriatr Soc 62, 679–684.

38.

Roalf

, Moberg

, Xie

, Wolk

, Moelter

, Arnold

(2013) Comparative accuracies of two common screening instruments for classification of Alzheimer’s disease, mild cognitive impairment, and healthy aging. Alzheimers Dement 9, 529–537.

39.

Thurstone

(1973) Primary Mental Abilities. In the Measurement of Intelligence Springer, Netherlands.

40.

Benton

, Hamsher

, Sivan

(1989) Multilingual Aphasia Examination, AJA Associates, Inc., Iowa City, IA.

41.

Matthews

, Hagstromer

, Pober

, Bowles

(2012) Best practices for using physical activity monitors in population-based research. Med Sci Sports Exerc 44, S68–76.

42.

Bassett

Jr , Toth

, LaMunion

, Crouter

(2017) Step counting: A review of measurement considerations and health-related applications. Sports Med 47, 1303–1315.

43.

Pobiruchin

, Suleder

, Zowalla

, Wiesner

(2017) Accuracy and adoption of wearable technology used by active citizens: A marathon event field study. JMIR Mhealth Uhealth 5, e24.

44.

Fletcher

, Balady

, Amsterdam

, Chaitman

, Eckel

, Fleg

, Froelicher

, Leon

, Pina

, Rodney

, Simons-Morton

, Williams

, Bazzarre

(2001) Exercise standards for testing and training: A statement for healthcare professionals from the American Heart Association. Circulation 104, 1694–1740.

45.

Borg

(1998) Borg’s perceived exertion and pain scales Human Kinetics, Champaign.

46.

Glass

, Gregory

(2007) ACSM’s Metabolic Calculations Handbook, Lippincott Williams & Wilkins, Baltimore.

47.

Koch

, Schaper

, Ittermann

, Spielhagen

, Dorr

, Volzke

, Opitz

, Ewert

, Glaser

(2009) Reference values for cardiopulmonary exercise testing in healthy volunteers: The SHIP study. Eur Respir J 33, 389–397.

48.

Glaser

, Ittermann

, Schaper

, Obst

, Dorr

, Spielhagen

, Felix

, Volzke

, Bollmann

, Opitz

, Warnke

, Koch

, Ewert

(2013) [The Study of Health in Pomerania (SHIP) reference values for cardiopulmonary exercise testing]. Pneumologie 67, 58–63.

49.

Lee

, Ryoo

, Park

, Jeong

, Alzheimer’s Disease Neuroimaging Initiative (2017) Morphological and microstructural changes of the hippocampus in early MCI: A study utilizing the Alzheimer’s Disease Neuroimaging Initiative Database. J Clin Neurol 13, 144–154.

50.

Aisen

, Petersen

, Donohue

, Gamst

, Raman

, Thomas

, Walter

, Trojanowski

, Shaw

, Beckett

, Jack

Jr , Jagust

, Toga

, Saykin

, Morris

, Green

, Weiner

, Alzheimer’s Disease Neuroimaging Initiative (2010) Clinical core of the Alzheimer’s Disease Neuroimaging Initiative: Progress and plans. Alzheimers Dement 6, 239–246.

51.

Laws

, Duncan

, Gale

(2010) ‘Normal’ semantic-phonemic fluency discrepancy in Alzheimer’s disease? A meta-analytic study. Cortex 46, 595–601.

52.

Shao

, Janse

, Visser

, Meyer

(2014) What do verbal fluency tasks measure? Predictors of verbal fluency performance in older adults. Front Psychol 5, 772.

53.

Kim

, Im

, Kwon

, Seo

, Ye

, Cho

, Noh

, Lee

, Kim

, Park

, Kim

, Hwang

, Kang

, Jeong

, Na

(2016) Higher physical activity is associated with increased attentional network connectivity in the healthy elderly. Front Aging Neurosci 8, 198.

54.

Lam

, Ong

, Dikot

, Sofiatin

, Wang

, Zhao

, Li

, Dominguez

, Natividad

, Yusoff

, Fu

, Senanarong

, Fung

, Lai

(2015) Intellectual and physical activities, but not social activities, are associated with better global cognition: A multi-site evaluation of the cognition and lifestyle activity study for seniors in Asia (CLASSA). Age Ageing 44, 835–840.

55.

Dyrstad

, Hansen

, Holme

, Anderssen

(2014) Comparison of self-reported versus accelerometer-measured physical activity. Med Sci Sports Exerc 46, 99–106.

56.

Aadahl

, Jorgensen

(2003) Validation of a new self-report instrument for measuring physical activity. Med Sci Sports Exerc 35, 1196–1202.

57.

Bravata

, Smith-Spangler

, Sundaram

, Gienger

, Lin

, Lewis

, Stave

, Olkin

, Sirard

(2007) Using pedometers to increase physical activity and improve health: A systematic review. JAMA 298, 2296–2304.

58.

Kostka

, Drygas

, Jegier

, Zaniewicz

(2009) Aerobic and anaerobic power in relation to age and physical activity in 354 men aged 20-88 years. Int J Sports Med 30, 225–230.

59.

Ross

, Blair

, Arena

, Church

, Despres

, Franklin

, Haskell

, Kaminsky

, Levine

, Lavie

, Myers

, Niebauer

, Sallis

, Sawada

, Sui

, Wisloff

, American Heart Association Physical Activity Committee of the Council on Lifestyle and Cardiometabolic Health, Council on Clinical Cardiology, Council on Epidemiology and Prevention, Council on Cardiovascular and Stroke Nursing, Council on Functional Genomics and Translational Biology, Stroke Council (2016) Importance of assessing cardiorespiratory fitness in clinical practice: A case for fitness as a clinical vital sign: A scientific statement from the American Heart Association. Circulation 134, e653–e699.

60.

Loe

, Rognmo

, Saltin

, Wisloff

(2013) Aerobic capacity reference data in 3816 healthy men and women 20-90 years. PLoS One 8, e64319.

61.

Edvardsen

, Hansen

, Holme

, Dyrstad

, Anderssen

(2013) Reference values for cardiorespiratory response and fitness on the treadmill in a 20- to 85-year-old population. Chest 144, 241–248.

62.

Barnes

, Yaffe

, Satariano

, Tager

(2003) A longitudinal study of cardiorespiratory fitness and cognitive function in healthy older adults. J Am Geriatr Soc 51, 459–465.

63.

Schultz

, Boots

, Almeida

, Oh

, Einerson

, Korcarz

, Edwards

, Koscik

, Dowling

, Gallagher

, Bendlin

, Christian

, Zetterberg

, Blennow

, Carlsson

, Asthana

, Hermann

, Sager

, Johnson

, Stein

, Okonkwo

(2015) Cardiorespiratory fitness attenuates the influence of amyloid on cognition. J Int Neuropsychol Soc 21, 841–850.

64.

Erickson

, Raji

, Lopez

, Becker

, Rosano

, Newman

, Gach

, Thompson

, Ho

, Kuller

(2010) Physical activity predicts gray matter volume in late adulthood: The Cardiovascular Health Study. Neurology 75, 1415–1422.

65.

Jedrziewski

, Ewbank

, Wang

, Trojanowski

(2010) Exercise and cognition: Results from the National Long Term Care Survey. Alzheimers Dement 6, 448–455.

66.

Larson

, Wang

, Bowen

, McCormick

, Teri

, Crane

, Kukull

(2006) Exercise is associated with reduced risk for incident dementia among persons 65 years of age and older. Ann Intern Med 144, 73–81.

67.

Vidoni

, Honea

, Billinger

, Swerdlow

, Burns

(2012) Cardiorespiratory fitness is associated with atrophy in Alzheimer’s and aging over 2 years. Neurobiol Aging 33, 1624–1632.

68.

Weuve

, Kang

, Manson

, Breteler

, Ware

, Grodstein

(2004) Physical activity, including walking, and cognitive function in older women. JAMA 292, 1454–1461.

69.

Huang

, Larsen

, Ried-Larsen

, Moller

, Andersen

(2014) The effects of physical activity and exercise on brain-derived neurotrophic factor in healthy humans: A review. Scand J Med Sci Sports 24, 1–10.

70.

Dinoff

, Herrmann

, Swardfager

, Liu

, Sherman

, Chan

, Lanctot

(2016) The effect of exercise training on resting concentrations of peripheral brain-derived neurotrophic factor (BDNF): A meta-analysis. PLoS One 11, e0163037.

71.

Rojas Vega

, Knicker

, Hollmann

, Bloch

, Struder

(2010) Effect of resistance exercise on serum levels of growth factors in humans. Horm Metab Res 42, 982–986.

72.

Erickson

, Voss

, Prakash

, Basak

, Szabo

, Chaddock

, Kim

, Heo

, Alves

, White

, Wojcicki

, Mailey

, Vieira

, Martin

, Pence

, Woods

, McAuley

, Kramer

(2011) Exercise training increases size of hippocampus and improves memory. Proc Natl Acad Sci U S A 108, 3017–3022.

73.

Gordon

, Rykhlevskaia

, Brumback

, Lee

, Elavsky

, Konopack

, McAuley

, Kramer

, Colcombe

, Gratton

, Fabiani

(2008) Neuroanatomical correlates of aging, cardiopulmonary fitness level, and education. Psychophysiology 45, 825–838.

74.

Sabia

, Dugravot

, Dartigues

, Abell

, Elbaz

, Kivimaki

, Singh-Manoux

(2017) Physical activity, cognitive decline, and risk of dementia: 28 year follow-up of Whitehall II cohort study. BMJ 357, j2709.

75.

Rosso

, Taylor

, Tabb

, Michael

(2013) Mobility, disability, and social engagement in older adults. J Aging Health 25, 617–637.

76.

Arena

, Myers

, Williams

, Gulati

, Kligfield

, Balady

, Collins

, Fletcher

, American Heart Association Committee on Exercise, Rehabilitation, and Prevention of the Council on Clinical Cardiology, American Heart Association Council on Cardiovascular Nursing (2007) Assessment of functional capacity in clinical and research settings: A scientific statement from the American Heart Association Committee on Exercise, Rehabilitation, and Prevention of the Council on Clinical Cardiology and the Council on Cardiovascular Nursing. Circulation 116, 329–343.

77.

de Souza

ESCG

, Franklin

, Forman

, Araujo

(2016) Influence of age in estimating maximal oxygen uptake. J Geriatr Cardiol 13, 126–131.