Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Longitudinal associations between energy utilization and brain volumes in cognitively normal middle aged and older adults

Abstract

Peak energy capacity of the whole person is associated with neurodegeneration. However, change in ability to utilize energy manifests as combination of declining peak energy capacity and rising energetic costs of mobility in mid-to-late life. We examined longitudinal associations between change in energy utilization and brain volumes. Cognitively normal participants from the Baltimore Longitudinal Study of Aging (N = 703, age = 70.4 ± 12.1 years, 54.1% women, 30% black) had concurrent data on brain volumes and energy utilization (defined as ratio of energetic cost of walking to peak energy capacity (“cost-to-capacity ratio”) at ≥ 1 visit between 2008 and 2018. We performed linear mixed-effect models, adjusting for demographics, medical history and walking engagement. Average baseline cost-to-capacity ratio was 0.55 ± 0.16, with average annual increase of 0.04 ± 0.13 over 3.9 follow-up years. A 10% higher baseline cost-to-capacity ratio was associated with 2.00 cm3 (SE = 0.44) larger baseline ventricular volume (p < 0.001), and 0.10 cm3 (SE = 0.03) greater annual increase in ventricular volume (p = 0.004) after adjustment. Longitudinal change in cost-to-capacity ratio was not associated with brain volumes. These findings highlight, among cognitive-normal adults, poorer baseline energy utilization is associated with subsequent ventricular enlargement, an indirect measure of central brain atrophy. Future studies should explore whether early detection of worsening energy utilization may act as a marker of underlying brain atrophy.

Introduction

The brain accounts for approximately 20% of total daily energy expenditure in humans1, yet little is known about the association between efficiency of energy utilization and brain health. To date, research has primarily focused on the construct of peak energy capacity (e.g., peak/max VO2) on cognitive performance and brain health2,3,4,5,6,7. However, beginning in mid-life, changes in the ability to utilize energy manifest as a combination of declining peak capacity and rising energetic costs of mobility8. These additional energetic costs are important, particularly as they expand to consume a larger portion of peak energy availability, leaving less energy for daily activities and leading to higher fatigability9,10. Many underlying mechanisms contribute to these rising energetic costs, including age- and disease-related alterations in metabolism, biomechanics, and neuromotor control11,12, which may further reduce energy availability. Over time, this “compression” of energy reserves pushes the cost of mobility beyond 50% of peak energy capacity, often approaching 100% of peak energy capacity in old age12,13,14. Previous work by our group has shown that compromised energy utilization is associated with slow and declining gait speed, greater fatigability, and faster decline in memory with aging12,13,14, but its association with concurrent brain structure and changes in brain structure has not been explored.

Previous neuroimaging studies among mid-to-late life populations, relying largely on cross-sectional design, suggest some associations between peak energy capacity and brain volumes but findings are inconsistent2,3,4,5. Specifically, a meta-analysis of 11 cross-sectional studies found a positive association between VO2max and white matter volume5; yet another study of middle-to-older aged adults found no association with white matter volume, but a significant positive association between VO2max and gray matter volume2. Importantly, the associations between peak energy capacity and brain volumes may differ by cognitive status; a study of older adults aged ≥ 60 years found VO2peak, an approximation of VO2max, was associated with whole brain and white matter volumes in participants with early Alzheimer’s disease, but not in nondemented participants3. Moreover, the cross-sectional nature of these works limits inferences in temporality.

Although limited longitudinal work suggests peak energy capacity is associated with brain volumes in selected middle and medial regions, including the parahippocampal gyrus6,7. Focusing on peak energy capacity alone may not sufficiently inform the effects of daily energy availability on brain structure, particularly among those whose energy demands for mobility exceed 50% of the peak energy capacity15,16. Accordingly, we focused on energy utilization, which simultaneously examines the age-associated upward trend in energetic costs for mobility in relation to the downward trend in peak energy expenditure14, and examined the magnitude of the cross-sectional and longitudinal associations between energy utilization and brain volumes in cognitively normal participants from the Baltimore Longitudinal Study of Aging (BLSA). We hypothesized that those with poorer energy utilization at baseline, or greater declines in energy utilization, would exhibit greater brain atrophy over time.

Methods

Study participants

The BLSA is a study of human aging established in 1958 and conducted by the National Institute on Aging Intramural Research Program. The BLSA consists of a continuously enrolled cohort of community dwelling volunteers who pass comprehensive health and functional screening evaluations and are free of major chronic conditions, except controlled hypertension, and cognitive and functional impairment at the time of enrollment. Once enrolled, participants are followed for life or until a diagnosis of dementia and undergo extensive testing every 1–4 years depending on age (< 60 every 4 years, 60–79 every 2 years, ≥ 80 every year)8.

A sub-sample of the BLSA of around 1000 participants has been prospectively assessed using magnetic resonance imaging (MRI) to measure global and regional brain structure. The current study sample consisted of all participants ≥ 50 years old with brain imaging, energy assessments, and body composition measures (N = 756). For each participant, the first visit with all assessments served as baseline. Visits were excluded if participants had missing data on any global or lobar brain regions (n = 18), body composition (n = 25), education (n = 2), or Center for Epidemiologic Studies Depression Scale (CES-D) score (n = 9). Participants were also excluded from the analysis after the onset of mild cognitive impairment (MCI) or Alzheimer’s disease, and all included participants reported no history of Parkinson’s disease. Comparison of baseline characteristics between those included and excluded did not reveal significant differences (Supplementary Table 1). The final sample included 703 participants aged 50–94 years assessed between 2008 and 2018. The study protocol was approved by the Institutional Review Board of the Intramural Research Program of the National Institutes of Health, and all methods were performed in accordance with the relevant guidelines and regulations. All participants provided written informed consent at each visit.

Energy utilization (cost-to-capacity ratio)

Average peak walking capacity VO2 (ml/kg/min) was assessed during the 400-m segment of the long-distance corridor walk; a rapid-paced endurance walking test and a validated proxy of cardiorespiratory fitness in older adults17. The test was performed on a 20-m course in an uncarpeted corridor marked by cones at both ends, with the participant wearing a portable metabolic analyzer, the Cosmed K4b2 (Cosmed, Rome, Italy). Participants were instructed to walk “as fast as possible, at a pace you can sustain for 400 m”. Standardized encouragement was given with each lap along with the number of laps remaining. Split times for each lap and total time to walk 400 m were recorded. To calculate average peak walking energy expenditure per kilogram of body weight (peak VO2 ml/kg/min), readings from the first 1.5 min of the test were discarded to allow the participant to adjust to the peak workload and the remaining readings were averaged to arrive at single measure of the average energy expended (ml/kg/min) during 400 m of peak sustained walking11,14.

The energetic cost of slow walking (ml/kg/min) was assessed via indirect calorimetry (Medical Graphics Corp, St Paul, MN) during 5 min of treadmill walking at 0.67 m/s (1.5 mph), 0% grade. This constant speed and duration were used for all participants, providing a standardized measure by which to gauge changes in energy expenditure (e.g., walking efficiency) during a low-demand task that minimizes exclusion of lower functioning participants11,14. To calculate the average volume of oxygen consumed per kilogram of body weight (ml/kg/min) during the walking task, energy expenditure readings from the first 2 min of testing were discarded to allow the participant to adjust to the workload. The final 3 min were averaged to derive a single measure of the average VO2 (ml/kg/min) consumed, or the average energetic cost of a slow standardized walking task11,14.

A ratio of the energetic cost of walking to peak walking energy expenditure (the “cost-to-capacity ratio”)11,14 was calculated to define the percentage of peak walking capacity needed for mobility:

$$\frac{{{\text{Energetic}}\,{\text{cost}}\,{\text{of}}\,{\text{slow}}\,{\text{walking }}\left( {{\text{ml}}/{\text{kg}}/{\text{min}}} \right)}}{{{\text{Peak}}\,{\text{walking}}\,{\text{energy}}\,{\text{expenditure}}\left( {{\text{ml}}/{\text{kg}}/{\text{min}}} \right)}}.$$

The cost-to-capacity ratio is bounded at 0 and 1, with a higher ratio indicating less efficient energy utilization (i.e., poorer energy utilization).

MRI acquisition

Scanning was performed on a 3 T Philips Achieva at the National Institute on Aging (NIA) Clinical Research Unit. T1-weighted magnetization-prepared rapid gradient echo (MPRAGE) scans were acquired (repetition time = 6.8 ms, echo time = 3.1 ms, flip angle = 8°, image matrix = 256 × 256, 170 slices, pixel size = 1 × 1 mm, slice thickness = 1.0 mm, field-of-view = 240 mm2). For regional labeling and volumetric analysis, a standardized procedure was applied to T1-weighted volumetric scans to align images in parallel with the anterior–posterior commissure place, remove extracranial tissue, and define regional labels for gray matter, white matter, and cerebrospinal fluid (CSF)18,19. The intracranial volume (ICV) was estimated for the baseline scan of each individual and used as a covariate in statistical analyses.

Regions of interest (ROIs) included total brain, gray matter, white matter, ventricles, frontal lobe, frontal gray matter, frontal white matter, temporal lobe, temporal gray matter, temporal white matter, parietal lobe, parietal gray matter, parietal white matter, occipital lobe, occipital gray matter, occipital white matter, and hippocampus. The volumes of ROIs (cm3) measured at each visit were treated as dependent variables in separate models.

Covariates

Demographic and lifestyle characteristics, including baseline age, sex, race (white, non-white), years of education, smoking status (never/former smoker, current smoker), and alcohol consumption (> 7 vs. ≤ 7 drinks per week) were obtained via an interviewer-administered questionnaire. During the interview, participants were also asked to report whether they were ever told by a doctor or other health professional that they had any of the following conditions: diagnosed cardiovascular diseases including heart attack, angina, myocardial infarction, congestive heart failure, peripheral arterial disease, and vascular-related procedures (including coronary artery bypass grafting or angioplasty); stroke; hypertension or high blood pressure; high cholesterol; chronic obstructive pulmonary disease; diabetes mellitus; cancer; or osteoarthritis. A morbidity index was created using a binary variable, defined as having 0–1 or 2 or more conditions mentioned above. Self-reported engagement in volitional walking was also recorded in the question “did you do brisk walking in the past 2 weeks” (yes/no).

Weight and height were assessed in light clothing according to standard protocols. Total body dual-energy X-ray absorptiometry (DEXA) was performed using a Prodigy Scanner (GE, Madison, WI) and analyzed with version 10.51.006 software to obtain fat mass and lean mass. Depressive symptoms were measured by CES-D20. Apolipoprotein E (APOE) e4 carrier status (≥ 1 vs. 0 e4 alleles) was determined from DNA samples21. All covariates were treated as time-fixed at baseline.

Diagnostic status (normal, MCI/dementia) was determined by standard procedures detailed previously22. Participants meeting pre-specified criteria were reviewed at a research diagnostic case conference. Although participants were excluded from the analyses after the onset of mild cognitive impairment (MCI) or Alzheimer’s disease, they were still followed-up and we were able to determine their latest diagnostic status.

Statistical analysis

Baseline participant characteristics were compared by the baseline cost-to-capacity ratio (≤ 0.50 vs > 0.50) using t-test for continuous variables and χ2 tests for categorical variables and were reported as mean ± standard deviation (SD) or frequencies (percentages). Alpha was set to 0.01 for ROIs analyses after correction of multiple comparison using false discovery rate (FDR). All analyses were performed using Stata version 16 (Statacorp, College Station, TX).

To quantify trajectories of changes in brain volumes in the overall sample, linear mixed-effect models with restricted maximum likelihood estimation were used. This approach can combine cross-sectional and longitudinal data, as it borrows information and improves the precision for cross-sectional associations23. Time was modeled as a continuous variable scaled to years since baseline. The fixed effects of interest consisted of baseline cost-to-capacity ratio, change in cost-to-capacity ratio from baseline, time, all covariates, and two-way interactions between baseline cost-to-capacity ratio and change in cost-to-capacity ratio with time. Change in cost-to-capacity ratio was calculated as cost-to-capacity ratio measured at the current visit minus the baseline cost-to-capacity ratio, and it was modeled as a time-varying predictor. Unstructured variances and covariance for random incepts and slopes were used. Test for multi-collinearity did not suggest collinearity among predictors (all VIF < 3). We also further explored the independent associations between energetic cost of slow walking and peak walking energy expenditure, after adjusting for each other, with brain volumes using the same linear mixed-effect model approach.

To address potential confounding, Model 1 was adjusted for ICV, the latest diagnostic status (subsequently impaired vs. remain normal), baseline age, sex, race, and years of education. Model 2 added height, fat mass, and lean mass to the covariates from Model 1, and Model 3 further adjusted for morbidity index, CES-D scores, APOE e4 allele status and walking engagement. Several sensitivity analyses were also conducted to further test the robustness of our results: (1) restricting the main analyses to a sub-population who were never diagnosed with MCI or dementia over extended follow-up; (2) adding interaction terms to assess the potential for effect modification by MCI/dementia diagnosis or APOE status (cost-to-capacity × subsequent normal/MCI/dementia status and cost-to-capacity × APOE status), and (3) stratifying participants based on their baseline energy utilization at a cost-to-capacity ratio of ≤ 0.50 vs > 0.5014.

Results

Participant characteristics

Of the overall sample with a mean age of 70.4 ± 12.2 years, over half (54.1%) were women, and around two thirds (70.0%) were of white race. The percentages of current smokers (1.7%) and heavy drinkers (16.9%) were low (Table 1). More than half reported a diagnosis of high cholesterol (59.9%), osteoarthritis (54.6%) and/or hypertension (52.2%), and the majority of participants (70.0%) had more than two morbidities. The average baseline cost-to-capacity ratio was 0.55 ± 0.16, meaning on average participants used approximately 55% of their energy capacity to walk at a slow pace (Table 1). On average, those with baseline cost-to-capacity ratio > 0.5 (n = 402, mean = 0.65 ± 0.12) tended to be older and less physical active. In addition, those with the higher baseline cost-to-capacity ratio had greater fat mass, lower lean mass, more adverse health conditions (e.g., hypertension, osteoarthritis, CVD, stroke, and more than two morbidities), higher depressive symptoms, and a higher proportion of subsequent MCI/dementia diagnosis. The average time since baseline for the overall sample (N = 703) was 2.2 ± 2.4 years, and among those with > 1 visits (n = 399) the average follow-up time was 3.9 ± 2.0 years (Table 1).

Table 1 Baseline characteristics of 703 BLSA participants by baseline cost-to-capacity ratio (≤ 0.5 vs. > 0.5), defined as the energy cost of slow walking to peak walking energy expenditure.

Energy utilization-related brain volumetric change

In adjusted models (Model 3), a 10% higher baseline cost-to-capacity ratio was significantly associated with 2.0 cm3 (SE = 0.44, p < 0.001) larger ventricular volume at baseline (Table 2, Column A), and a 0.10 cm3 (SE = 0.03, p = 0.004) greater annual increase in ventricular volume (Table 2, Column B). Participants with poorer baseline energy utilization appeared to have a steeper annual change in ventricular volume (Fig. 1), but there were no significant longitudinal associations (Table 2, Column C, D).

Table 2 Longitudinal mixed models of the associations between energy utilization and brain volumes among 703 BLSA participantsa.
Figure 1
figure 1

Adjusted baseline and annual change of ventricular volume at years since baseline (index). Solid black lines represent the estimated average if baseline cost-to-capacity ratio = 0.2 (good energy utilization), dashed black lines represent the estimated average volume if baseline cost-to-capacity ratio = 0.5 (moderate energy utilization), and dotted black lines represent the estimated average volume if baseline cost-to-capacity ratio = 0.8 (poor energy utilization). To better see individual trajectories, gray dots and gray lines at the background represent crude individual data points in this study. All estimated averages were fitted using linear mixed effect models after adjustmenta. a Model (model 3) adjusted for ICV, the latest diagnostic status, baseline age, sex, race, years of education, height, fat mass, lean mass, number of multi-morbidities, CES-D, APOE e4 status and brisk walking.

When examining the independent associations of energetic cost of slow walking and peak walking energy expenditure with brain volumes, in the fully adjusted models: (i) a 1 ml/kg/min higher baseline energetic cost of slow walking was associated with a 0.09 cm3 (SE = 0.03, p = 0.004) greater annual increase in ventricular volume (Supplementary Table 2, Column B), and (ii) a 1 ml/kg/min higher baseline peak walking energy expenditure was associated with a 0.67 cm3 (SE = 0.17, p < 0.001) smaller ventricular volume at baseline (Supplementary Table 3, Column A).

Our results largely remained robust with sensitivity analyses. When restricted to participants who never developed MCI or dementia over extended follow-up, a higher baseline cost-to-capacity ratio remained associated with a larger baseline ventricular volume (p < 0.001) and an annual increase in ventricular volume (p = 0.005) (Supplementary Table 4). Interaction terms with cost-to-capacity × subsequent normal/MCI/dementia status and cost-to-capacity × APOE status were not significant at the FDR-adjusted level (p < 0.01) and removed from the final models.

Further, we found some discrepancies when stratifying by baseline cost-to-capacity ratio: among those with baseline cost-to-capacity ratio ≤ 0.5 (good/modest energy utilization), a higher baseline cost-to-capacity ratio was only associated with an annual increase in ventricular volume (β = 0.16, SE = 0.08, p = 0.045) at the non-FDR adjusted level, but not the main baseline effect of brain volumes. Whereas, among those with a baseline cost-to-capacity ratio > 0.5 (poor energy utilization), a higher baseline cost-to-capacity ratio was only associated with a larger baseline ventricular volume (β = 2.34, SE = 0.91, p = 0.010) but not with annual changes in any brain volumes (Supplementary Table 5).

Discussion

This prospective longitudinal study among middle-to-older-aged adults found that poorer energy utilization was associated with larger ventricular volume and steeper increase in ventricular volume over a mean 3.9 years of follow-up. However, no associations were found with other brain regions, and longitudinal change in energy utilization over time was not associated with brain structure. Notably, those with relatively efficient energy utilization at baseline, as indicated by a cost-to-capacity ratio ≤ 0.5, had a trending association between baseline cost-to-capacity ratio and subsequent annual increase in ventricular volume, whereas those with poor baseline energy utilization did not, hinting at a possible threshold effect. Collectively, these findings suggest that adverse changes in energy utilization may contribute to increases in ventricular volume early in the pathological process when energy utilization is still relatively efficient, warranting future follow-up in younger populations.

Ventricular enlargement provides an indirect measure of atrophy of central brain structures24. Although ventricular enlargement is seen with normal aging, it is also evident in a variety of neurodegenerative diseases, including Alzheimer’s disease25. In the current study, we surmise that the association between energy utilization and brain volumes limited to ventricular enlargement at baseline may be explained as follows: higher cardiorespiratory fitness has been associated with smaller ventricles7,26, and higher cardiorespiratory fitness also contributes to better energy utilization. Thus, it is plausible that energy utilization, defined by both cost and capacity, is related to ventricular enlargement. However, other brain-related diseases, such as Lewy body disease and vascular dementia, where gait dysfunction is common, may also be potential underlying contributors to the association between energy utilization and ventricular enlargement27. The majority of our impairments in our sample were Alzheimer’s disease and only a small number were dementia for other reasons (n = 19). As a result, we are unable to further disentangle the association regarding dementia subtypes.

Contrary to our hypothesis, changes in energy utilization were not related to annual changes in brain volumes. This may be due to the relatively older mean baseline age of the current sample (70 years old), or a potential threshold effect of energy utilization on brain atrophy. In our sample, change in cost-to-capacity ratio (0.04 ± 0.13) was minimal over the follow-up years, which could restrict our statistical power to detect significant associations. More importantly, the average cost-to-capacity ratio at baseline was 0.55 in this population, indicating that 55% of peak walking energy was used for ambulation. Previous work suggests that the cost-to-capacity ratio begins to increase around age 50, with an exponential increase each decade thereafter8. Thus, it is possible that once individuals exceed a cost-to-capacity ratio threshold of 40–50%, the effect of energy utilization on brain atrophy has already occurred, and subsequent changes in energy utilization have no further effect on brain volumes. We tested such possibility in the stratified analysis based on baseline energy utilization at a cost-to-capacity ratio of ≤ 0.50 vs > 0.50. We found that when baseline energy utilization was good/moderate (≤ 0.5), it was marginally associated with subsequent annual changes in ventricular volume; yet when baseline energy regulation was poor (> 0.50), there was no association with annual changes in brain volumes. Collectively, these results support our conjecture, but we cannot rule out the possibility of survival bias; those with less efficient energy utilization were less likely to return for follow-up visits, possibly attenuating longitudinal effects. Future studies with a longer follow-up time and more younger populations are warranted to disentangle these associations.

When examining the independent components of the cost ratio, we found that baseline energetic cost of slow walking was associated with an annual increase in ventricular enlargement, while baseline peak walking energy expenditure was associated with smaller ventricular volume at baseline only. These findings illustrate the differential utility of these energetic measures in understanding changes in brain structure with aging. Previous work from the BLSA examining the bidirectional association of time to walk 400 m, a validated measure of cardiorespiratory fitness17, with brain volumes also showed that fitness was not associated with future changes in ventricular volume, but smaller ventricular volume predicted future higher fitness7. The current study adds to this body of literature by illustrating that energetic cost of walking can be informative beyond measures of fitness, showing stronger associations with future brain changes, and adding to the growing body of literature linking the brain and aspects of mobility. Indeed, another recent study from the BLSA further showed that energetic cost of usual-paced walking was associated with accelerated annual increase in ventricular volume and decline in hippocampal volume28. Future research is needed to fully understand the association between the energy needed for walking at various speeds and brain volumes, especially among those less healthy than BLSA participants who may be more susceptible to dementia-related pathology.

To our knowledge, this is the first longitudinal study examining energy utilization—with a novel measure of energy cost and energy capacity—and brain volumes. However, the present study is not without limitations. First, the BLSA participants are highly educated, mostly Caucasian, and relatively healthy, thus findings may not generalize to other populations of older adults. Second, about half (49.6%) of the sample only contributed to the baseline data, and mid-life BLSA participants contribute less longitudinal data due to a longer interval between follow-up visits. Finally, the BLSA is observational, and although its prospective design allows for assessment of temporality between energy utilization and subsequent changes in brain volumes, we cannot rule out unmeasured confounding and reverse causality—that brain changes in early life may affect energy utilization in mid-to-late life.

In conclusion, less efficient energy utilization at baseline is associated with subsequent annual ventricle enlargement among middle-to-older-aged adults. Future longitudinal studies are needed to better characterize the value of energy utilization, which may be a potential early indicator of susceptibility to Alzheimer’s disease, in understanding changes in brain structure in younger and middle-aged adults.

Data availability

Because of the sensitive nature of the data collected for this study, requests to access the data set from qualified researchers trained in human subject confidentiality protocols may be sent to the Intramural Research Program of the National Institute on Aging at https://blsa.nih.gov.

References

  1. Clark, D. & Sokoloff, L. (Lippincott, 1999).

  2. Gordon, B. A. et al. Neuroanatomical correlates of aging, cardiopulmonary fitness level, and education. Psychophysiology 45, 825–838. https://doi.org/10.1111/j.1469-8986.2008.00676.x (2008).

    Article  PubMed  PubMed Central  Google Scholar 

  3. Burns, J. M. et al. Cardiorespiratory fitness and brain atrophy in early Alzheimer disease. Neurology 71, 210–216. https://doi.org/10.1212/01.wnl.0000317094.86209.cb (2008).

    Article  PubMed  Google Scholar 

  4. Erickson, K. I., Leckie, R. L. & Weinstein, A. M. Physical activity, fitness, and gray matter volume. Neurobiol. Aging 35(Suppl 2), S20-28. https://doi.org/10.1016/j.neurobiolaging.2014.03.034 (2014).

    Article  PubMed  Google Scholar 

  5. Sexton, C. E. et al. A systematic review of MRI studies examining the relationship between physical fitness and activity and the white matter of the ageing brain. Neuroimage 131, 81–90. https://doi.org/10.1016/j.neuroimage.2015.09.071 (2016).

    Article  PubMed  Google Scholar 

  6. Tian, Q., Studenski, S. A., Resnick, S. M., Davatzikos, C. & Ferrucci, L. Midlife and late-life cardiorespiratory fitness and brain volume changes in late adulthood: Results from the Baltimore Longitudinal Study of Aging. J. Gerontol. A Biol. Sci. Med. Sci. 71, 124–130. https://doi.org/10.1093/gerona/glv041 (2016).

    Article  PubMed  Google Scholar 

  7. Tian, Q. et al. A prospective study of focal brain atrophy, mobility and fitness. J. Intern. Med. 286, 88–100. https://doi.org/10.1111/joim.12894 (2019).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  8. Kuo, P. L. et al. A roadmap to build a phenotypic metric of ageing: Insights from the Baltimore Longitudinal Study of Aging. J. Intern. Med. 287(4), 373–394. https://doi.org/10.1111/joim.13024 (2020).

    Article  PubMed  PubMed Central  Google Scholar 

  9. Schrack, J. A. et al. Longitudinal association between energy regulation and fatigability in mid-to-late life. J. Gerontol. Ser. A 75, e74–e80. https://doi.org/10.1093/gerona/glaa011 (2020).

    CAS  Article  Google Scholar 

  10. Liu, F. et al. Association between walking energetics and fragmented physical activity in mid-to-late life. J. Gerontol. Ser. A Biol. Sci. Med. Sci. https://doi.org/10.1093/gerona/glab127 (2021).

    Article  Google Scholar 

  11. Schrack, J. A., Simonsick, E. M. & Ferrucci, L. The energetic pathway to mobility loss: An emerging new framework for longitudinal studies on aging. J. Am. Geriatr. Soc. 58(Suppl 2), S329-336. https://doi.org/10.1111/j.1532-5415.2010.02913.x (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  12. Schrack, J. A., Zipunnikov, V., Simonsick, E. M., Studenski, S. & Ferrucci, L. Rising energetic cost of walking predicts gait speed decline with aging. J. Gerontol. Ser. A Biol. Sci. Med. Sci. 71, 947–953. https://doi.org/10.1093/gerona/glw002 (2016).

    Article  Google Scholar 

  13. Schrack, J. A., Simonsick, E. M., Chaves, P. H. & Ferrucci, L. The role of energetic cost in the age-related slowing of gait speed. J. Am. Geriatr. Soc. 60, 1811–1816. https://doi.org/10.1111/j.1532-5415.2012.04153.x (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  14. Schrack, J. A., Simonsick, E. M. & Ferrucci, L. The relationship of the energetic cost of slow walking and peak energy expenditure to gait speed in mid-to-late life. Am. J. Phys. Med. Rehabil. 92, 28–35. https://doi.org/10.1097/PHM.0b013e3182644165 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  15. Waters, R. L. & Mulroy, S. The energy expenditure of normal and pathologic gait. Gait Posture 9, 207–231. https://doi.org/10.1016/s0966-6362(99)00009-0 (1999).

    CAS  Article  PubMed  Google Scholar 

  16. Larish, D. D., Martin, P. E. & Mungiole, M. Characteristic patterns of gait in the healthy old. Ann. N. Y. Acad. Sci. 515, 18–32. https://doi.org/10.1111/j.1749-6632.1988.tb32960.x (1988).

    ADS  CAS  Article  PubMed  Google Scholar 

  17. Simonsick, E. M., Fan, E. & Fleg, J. L. Estimating cardiorespiratory fitness in well-functioning older adults: Treadmill validation of the long distance corridor walk. J. Am. Geriatr. Soc. 54, 127–132. https://doi.org/10.1111/j.1532-5415.2005.00530.x (2006).

    Article  PubMed  Google Scholar 

  18. Doshi, J. et al. MUSE: MUlti-atlas region segmentation utilizing ensembles of registration algorithms and parameters, and locally optimal atlas selection. Neuroimage 127, 186–195. https://doi.org/10.1016/j.neuroimage.2015.11.073 (2016).

    Article  PubMed  Google Scholar 

  19. Davatzikos, C., Genc, A., Xu, D. & Resnick, S. M. Voxel-based morphometry using the RAVENS maps: Methods and validation using simulated longitudinal atrophy. Neuroimage 14, 1361–1369. https://doi.org/10.1006/nimg.2001.0937 (2001).

    CAS  Article  PubMed  Google Scholar 

  20. Radloff, L. S. The CES-D scale: A self-report depression scale for research in the general population. Appl. Psychol. Meas. 1, 385–401. https://doi.org/10.1177/014662167700100306 (1977).

    Article  Google Scholar 

  21. Thambisetty, M., Beason-Held, L., An, Y., Kraut, M. A. & Resnick, S. M. APOE e4 genotype and longitudinal changes in cerebral blood flow in normal aging. Arch. Neurol. 67, 93–98. https://doi.org/10.1001/archneurol.2009.913 (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  22. Resnick, S. M., Pham, D. L., Kraut, M. A., Zonderman, A. B. & Davatzikos, C. Longitudinal magnetic resonance imaging studies of older adults: A shrinking brain. J. Neurosci. 23, 3295–3301. https://doi.org/10.1523/JNEUROSCI.23-08-03295.2003 (2003).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  23. Yang, J., Zaitlen, N. A., Goddard, M. E., Visscher, P. M. & Price, A. L. Advantages and pitfalls in the application of mixed-model association methods. Nat. Genet. 46, 100–106. https://doi.org/10.1038/ng.2876 (2014).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  24. Zahr, N. M. et al. A mechanism of rapidly reversible cerebral ventricular enlargement independent of tissue atrophy. Neuropsychopharmacology 38, 1121–1129. https://doi.org/10.1038/npp.2013.11 (2013).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  25. Fox, N. C. & Schott, J. M. Imaging cerebral atrophy: Normal ageing to Alzheimer’s disease. Lancet 363, 392–394. https://doi.org/10.1016/s0140-6736(04)15441-x (2004).

    Article  PubMed  Google Scholar 

  26. Raichlen, D. A., Klimentidis, Y. C., Bharadwaj, P. K. & Alexander, G. E. Differential associations of engagement in physical activity and estimated cardiorespiratory fitness with brain volume in middle-aged to older adults. Brain Imaging Behav. https://doi.org/10.1007/s11682-019-00148-x (2019).

    Article  Google Scholar 

  27. De Cock, A.-M. et al. Comprehensive quantitative spatiotemporal gait analysis identifies gait characteristics for early dementia subtyping in community dwelling older adults. Front. Neurol. 10, 313. https://doi.org/10.3389/fneur.2019.00313 (2019).

    Article  PubMed  PubMed Central  Google Scholar 

  28. Dougherty, R. J., Ramachandran, J., Liu, F. et al. Association of walking energetics with amyloid beta status: Findings from the Baltimore Longitudinal Study of Aging. Alzheimers Dement (Amst). 13(1), e12228. https://doi.org/10.1002/dad2.12228 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Funding

This research was supported by the Intramural Research Program of the National Institute on Aging of the National Institutes of Health and by U01AG057545 from the National Institute of Aging. Data used in the analyses were obtained from the Baltimore Longitudinal Study of Aging, a study performed by the National Institute on Aging Intramural Research Program.

Author information

Authors and Affiliations

Authors

Contributions

Authors Y.Q., E.M.S., L.F., and J.A.S. had full access to all of the data for the study and take responsibility for the integrity of the data and accuracy of the data analyses. All authors: interpretation of data, critical revision of the manuscript for important intellectual content. All authors read and approved the submitted manuscript.

Corresponding author

Correspondence to Yujia Qiao.

Ethics declarations

Competing interests

A.S. received an honorarium as a consultant for Merck and received honoraria from Springer Nature Switzerland AG for guest editing special issues of Current Sleep Medicine Reports. Other co-authors have no conflict of interest to report.

Additional information

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Qiao, Y., Wanigatunga, A.A., An, Y. et al. Longitudinal associations between energy utilization and brain volumes in cognitively normal middle aged and older adults. Sci Rep 12, 6472 (2022). https://doi.org/10.1038/s41598-022-10421-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41598-022-10421-7

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing