Cardiorespiratory fitness and brain atrophy in early Alzheimer disease

Sample and recruitment.

Subjects without dementia (Clinical Dementia Rating [CDR] 0, n = 64) and subjects with early-stage AD (CDR 0.5 and 1, n = 57) aged 60 and over were enrolled in the University of Kansas Brain Aging Project. The CDR 0.5 (n = 45) and CDR 1 (n = 12) participants together represent early-stage AD. Study exclusions include neurologic disease other than AD, diabetes mellitus (defined as a clinical diagnosis and use of an antidiabetic agent), history of ischemic heart disease (acute coronary artery event, angina), schizophrenia, clinically significant depressive symptoms, abnormalities in B12, RPR, or thyroid function, use of antipsychotic and investigational medications, and significant visual or auditory impairment, systemic illness, or orthopedic issues that could impair completion of the study.

Clinical assessment.

The clinical assessment included a semi-structured interview with the participant and with a collateral source knowledgeable about the participant. Diagnostic criteria for AD require the gradual onset and progression of impairment in memory and in at least one other cognitive and functional domain.¹⁴ The presence or absence of dementia, and its severity if present, was determined using the CDR.¹⁵ These methods have a diagnostic accuracy for AD of 93%,¹⁶ are sensitive to detecting the earliest stages of AD by focusing on intraindividual change rather than comparison with group norms,¹⁷ and are accurate in identifying the subset of individuals meeting criteria for MCI who have early-stage AD.¹⁸ Only CDR 0, 0.5, and 1 subjects were enrolled into this study. The CDR sum of boxes was used as a measure of dementia severity.

Medications, past medical history, education, and demographic information were collected from the collateral source by the nurse clinician. A standard physical and neurologic examination was performed to assess abnormalities in visual fields, cranial nerves, motor strength, sensation, reflexes, plantar responses, coordination, praxis, and gait.

Neuropsychological assessment.

A trained psychometrician administered a psychometric battery including standard measures of memory (WMS-R Logical Memory I and II,¹⁹ Free and Cued Selective Reminding Task²⁰), language (Boston Naming Test–15 item²¹), working memory (WMS III Digit Span Forwards and Backwards,¹⁹ WAIS Letter-number sequencing²²), executive function (Trailmaking A and B,²³ Verbal Fluency [animals, fruits and vegetables], and Stroop Color-Word Test²⁴), and visuospatial ability (WAIS Block Design²²). The Mini-Mental State Examination (MMSE)²⁵ was administered as a measure of global cognition. Additionally, all cognitive performance scores were converted to z-scores (with higher scores representing better performance) based on the mean and SD of all subjects without dementia and subjects with early-stage AD. The mean of each participant’s performance z-scores was determined to create an index of global cognitive performance.

Cardiorespiratory fitness.

VO₂^peak, the standard measure of cardiorespiratory fitness, was measured during a symptom-limited graded treadmill test, as previously described.²⁶ Subjects were attached to a 12-lead electrocardiograph to monitor cardiac stability and wore a non-rebreathing facemask. Subjects began walking at a pace of 1.7 miles per hour at 0% grade. Every 2 minutes the speed or grade was increased according to a protocol designed for a geriatric population.²⁷ Blood pressure and the participant’s self-assessment of exertion using the 15-point Borg Rating of Perceived Exertion scale²⁸ were acquired during the last 30 seconds of each 2-minute stage. Heart rates were recorded every 15 seconds. Expired air was measured for oxygen and carbon dioxide at 15-second intervals using a Parvomedics system. Gas calibrations were performed on the metabolic cart before each test according to the specifications of the manufacturer.

VO₂^peak was considered the highest observed value during the test.²⁹ Participant effort during the treadmill test was assessed with measures of the respiratory exchange ratio (value ≥ 1.0 considered near-maximal effort³⁰) and 17 individuals were excluded for not attaining an RER ≥ 1.0 (6 subjects without dementia vs 11 subjects with early-stage AD, p = 0.17). Subjects excluded for an RER < 1.0 were predominantly women (14 women and three men) and tended to be older than those achieving an RER ≥ 1.0 (77.5 [7.1] vs 73.5 [6.5] years, p = 0.04). There was no difference in global cognition, MMSE, and dementia severity in those excluded vs those achieving an RER ≥ 1.0.

The primary fitness measure was VO₂^peak normalized by lean mass determined by DEXA scanning (Lunar Prodigy, GE Medical Systems, Madison, WI). This measure was chosen as our primary fitness measure for several reasons. Normalizing VO₂^peak for lean mass is more physiologic than indexing for body weight²⁹ given that 95% of oxygen consumption during exercise occurs in exercising muscles (i.e., lean mass). Additionally, VO₂^peak normalized by lean mass minimizes gender differences in VO₂^peak. For instance, gender differences were not evident in our overall cohort (n = 121) for VO₂^peak normalized by lean mass (37.1 [6.8] in men vs 36.1 [5.1] mL/kg/min in women, p = 0.37) while robust gender differences are present when normalizing VO₂^peak for total body weight (24.4 [5.2] in men vs 19.9 [3.6] mL/kg/min in women, p < 0.0001).

Physical activity and frailty assessments.

The participant’s level of habitual physical activity was estimated using the Physical Activity Scale in the Elderly (PASE). The PASE is a reliable and valid measure of physical activity and physical function developed specifically for older individuals.³¹ The PASE assesses an individual’s level of physical activity within the last 7 days as an estimate of habitual physical activity. We modified the PASE by administering it to the subject’s study partner for both subjects without dementia and subjects with AD. The Physical Performance Test is a short battery of timed physical tasks serving as a composite measure of physical performance and used to assess physical frailty. The modified Physical Performance Test³² includes 1) writing a sentence, 2) simulated eating, 3) lifting a book and placing it on a shelf above shoulder height, 4) putting on and removing a jacket, 5) picking up a penny from the floor, 5) turning 360 degrees and 7) walking 50 feet, 8) chair rises, and 9) a progressive Romberg test.

Neuroimaging.

Structural MRI was obtained using a Siemens 3.0 Tesla Allegra MRI scanner at the Hoglund Brain Imaging Center. High-resolution T1-weighted anatomic images were acquired to provide detailed gross anatomy with high gray-white matter contrast (magnetization-prepared rapid gradient echo; 1 × 1 × 1 mm voxels; repetition time = 2,500, echo time = 4.38, inversion time = 1,100, field of view 256 × 256 with 18% oversample, 1 mm slice thickness, flip angle = 8 degrees). Normalized whole brain volume was computed for each image session using a validated set of imaging tools from FMRIB Software Library (FSL; www.fmrib.ox.ac.uk/fsl). The images were preprocessed and skull-stripped using Brain Extraction Tool. The skull stripped images were then segmented into white matter, gray matter, and CSF using FMRIB’s Automated Segmentation Tool by registering them to a Montreal Neuroimaging Institute avg152 template. Image processing was conducted utilizing the Laboratory of Neuroimaging Pipeline (University of California Los Angeles, www.pipeline.loni.ucla.edu). Normalized volumes for white matter, gray matter, and whole brain (sum of white and gray matter) were calculated by dividing each by the total intracranial volume (the sum of white, gray, and CSF volumes) and expressed as the percent of total intracranial volume. Normalized brain volumes minimize gender differences and produce an estimate of brain atrophy. As expected, normalized whole brain volumes were not related to gender (77.1% [3.3] in women vs 76.9% [3.4] in men, p = 0.76), lending validity to the normalization procedure.

Statistical analyses.

Analyses were conducted using SPSS 15.0. Continuous variables were summarized by means and standard deviations while categorical variables were summarized by frequency and percent. Student t test was used to compare continuous demographic and imaging variables in early AD and no-dementia groups. A χ² test was used to compare categorical variables. Pearson correlation coefficients were calculated to assess simple relationships between variables. Linear regression was used to assess the relationships between variables while controlling for covariates. A univariate general linear model was used to examine for the presence of group × VO₂^peak in predicting brain volume. To account for the apparent differential relationship between fitness and brain volume across groups (group × VO₂^peak interaction, see Results), we conducted all analyses within diagnostic groups (early AD and no-dementia subjects separately) as opposed to the entire cohort combined. To minimize the confounding influence of age-related changes in brain volume we controlled all analyses for age.

Descriptives.

Table 1 summarizes the characteristics of the no-dementia and AD groups. The mean age of the cohort (n = 121) was 73.5 (6.5) years with no difference between no-dementia and early-stage AD groups. Individuals in the early AD group had mild global cognitive dysfunction with a mean MMSE score of 26.2. Whole brain atrophy was apparent in the early AD group, with whole brain differences primarily related to group differences in gray matter volume, consistent with studies suggesting gray matter is more vulnerable to AD than white matter.^33,34 Whole brain volume was associated with global cognitive performance (r = 0.53, p < 0.001) and MMSE performance (r = 0.33, p < 0.001).

Cardiorespiratory fitness.

Modest group differences in cardiorespiratory fitness were observed between AD (34.7 [5.0] mL/kg/min) and no-dementia groups (38.1 [6.3] mL/kg/min, p = 0.002). Group means of peak heart rate and RER were consistent with maximal effort (i.e., 90% of predicted maximum heart rate [220 − age] and RER ≥ 1.1 for both groups). The duration of exercise testing to achieve VO₂^peak was lower in the early AD group than in participants without dementia.

The role of dementia severity on treadmill testing was examined by assessing the correlation between dementia severity (CDR box score) and measures of effort and VO₂^peak in the early AD group. Dementia severity was not related to VO₂^peak (r = −0.09, p = 0.49), RER (r = −0.01, p = 0.94), peak heart rate (r = 0.14, p = 0.30), exercise duration (r = −0.16, p = 0.25), and rating of perceived exertion (r = −0.12, p = 0.37).

Fitness and brain structure.

Simple correlations were first examined between VO₂^peak, brain atrophy, and cognition (table 2). Given the presence of a group × VO₂^peak interaction in predicting whole brain volume (p = 0.004), all analyses were performed within diagnostic groups (i.e., no dementia and early AD) rather than the overall group combined. VO₂^peak was associated with whole brain volume in the early AD group (r = 0.54, p < 0.001; figure 1) but not in controls without dementia (r = 0.18, p = 0.17, figure 2). In early AD, fitness was associated with both white matter volume (r = 0.39, p = 0.003) and gray matter volume (r = 0.36, p = 0.007) with higher fitness levels associated with increased brain volume.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Relationship between whole brain volume and cardiorespiratory fitness in participants with early Alzheimer disease

Increased fitness levels are associated with increased normalized whole brain volume (less brain atrophy). Filled circles indicate participants with very mild dementia (Clinical Dementia Rating 0.5) and open circles indicated participants with mild dementia (Clinical Dementia Rating 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 Relationship between whole brain volume and cardiorespiratory fitness in participants with no dementia

Next, these relationships were assessed when controlling for the effect of age (table 1) and other covariates. In early AD, controlling for age did not attenuate the relationship of fitness with whole brain volume (beta = 0.35, p = 0.02) and white matter volume (beta = 0.35, p = 0.04). Additionally, controlling for age and individual covariates of gender, dementia severity (CDR sum of boxes), physical activity level, and a measure of physical frailty (physical performance test) did not alter the relationship of fitness with whole brain volume and white matter volume. An overall model including all covariates did not alter the relationship between fitness and whole brain volume (beta = 0.30, p = 0.03) and partially attenuated the relationship of fitness and white matter volume (beta = 0.27, p = 0.096). After controlling for age, there was no relationship between fitness and gray matter volume in the early AD participants. In participants without dementia, there was no relationship between fitness and whole brain, white matter, and gray matter volumes after controlling for age.

Cognitive performance.

Simple correlations between VO₂^peak and overall cognitive performance were assessed first. In early AD, VO₂^peak was not associated with global cognitive performance or performance on the MMSE. On specific cognitive tasks, VO₂^peak was associated with performance on delayed memory (logical memory II, r = 0.30, p = 0.02) and digit symbol (r = 0.37, p = 0.05) in early AD although these associations were not present when controlling for age. In controls without dementia, VO₂^peak was associated with global cognitive performance (r = 0.30, p = 0.02) with increasing fitness associated with higher global cognitive performance. On specific cognitive tasks, VO₂^peak in participants with no dementia was most strongly associated with timed tasks of Trailmaking A (r = −0.32, p = 0.01) and Trailmaking B (r = −0.25, p = 0.005). Additionally, fitness was associated with performance on the Stroop interference task (r = 0.30, p = 0.02), a measure reflecting executive function, and performance on delayed memory (logical memory II; r = 0.26, p = 0.04). These associations, however, were primarily driven by age and were not present when controlling for the effect of age.