Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Effects of Lutein and Zeaxanthin Supplementation on Brain Morphology in Older Adults: A Randomized, Controlled Trial

The Effects of Lutein and Zeaxanthin Supplementation on Brain Morphology in Older Adults: A... Hindawi Journal of Aging Research Volume 2019, Article ID 3709402, 11 pages https://doi.org/10.1155/2019/3709402 Research Article The Effects of Lutein and Zeaxanthin Supplementation on Brain Morphology in Older Adults: A Randomized, Controlled Trial 1 1 1 Catherine M. Mewborn, Cutter A. Lindbergh, B. Randy Hammond, 1,2 1 Lisa M. Renzi-Hammond, and L. Stephen Miller Department of Psychology, University of Georgia, Athens, GA 30605, USA Institute of Gerontology, Department of Health Promotions and Behavior, College of Public Health, University of Georgia, Athens, GA 30605, USA Correspondence should be addressed to L. Stephen Miller; lsmiller@uga.edu Received 27 July 2019; Accepted 29 October 2019; Published 1 December 2019 Academic Editor: Carmela R. Balistreri Copyright © 2019 Catherine M. Mewborn et al. )is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. A growing literature emphasizes the importance of lifestyle factors such as nutrition in successful aging. )e current study examined if one year of supplementation with lutein (L) and zeaxanthin (Z), two nutrients with known antioxidative properties and cognitive benefits, impacted structural brain outcomes in older adults using a double-blind, randomized, placebo-controlled trial design. Community-dwelling older adults (20 males and 27 females) aged 65–87 years (M � 71.8 years, SD � 6.04 years) were randomized into supplement (N � 33) and placebo groups (N � 14) using simple randomization. )e supplement group received 10 mg L + 2 mg Z daily for 12 months while the placebo group received a visually identical, inert placebo. L and Z were measured via retinal concentrations (macular pigment optical density or MPOD). Structural brain outcomes, focusing on global and frontal- temporal lobe regions, were acquired using both T1-weighted and DTI MRI sequences. We hypothesized that the supplement group would increase, maintain, or show attenuated loss in hypothesized regions-of-interest (ROIs) while the placebo group would show age-related declines in brain structural integrity over the course of the trial. While results showed age-related declines for frontal and temporal gray and white matter volumes, as well as fornix white matter microstructure across both groups, only minimal differences were found between the supplement and placebo groups. However, exploratory analyses showed that individuals who responded better to supplementation (i.e., showed greater increases in MPOD) showed less decline in global and prefrontal gray matter volume than supplement “nonresponders.” While results suggest that one year of L and Z supplementation may have limited effects on structural brain outcomes overall, there may be a subsample of individuals for whom supplementation of L and Z provides greater benefits. ClinicalTrials.gov number, NCT02023645. structure and cognitive functioning [4]. To combat the 1. Introduction negative effects of oxidation, researchers have studied nu- Aging is associated with many changes, both cognitive and trients such as vitamins, flavonoids, and carotenoids for their neural, that contribute to negative outcomes such as de- potential in preventing and treating age-related cognitive creased functional independence, significant personal and and neural decline. Intake of these nutrients, along with societal economic burden, and psychological distress for healthy fatty acids and adherence to a balanced healthy diet, both aging individuals and their caregivers [1, 2]. One of the has been associated with positive neural effects, including more common theories of biological aging is the Free preserved gray and white matter volume, white matter Radical/Oxidative Stress )eory of Aging [3], which states microstructure, and lower risk of cerebral infarcts, even after that oxidative stress causes damage to DNA and proteins. In controlling for demographics and vascular risk factors [5–7]. turn, oxidation leads to neural inflammation, neurotoxicity, Lutein (L) and zeaxanthin (Z) are two nutrients in the reduced cerebral perfusion, and disruption of neural xanthophyll carotenoid family that have been suggested to 2 Journal of Aging Research benefit cognition and neural outcomes in older adults. 2. Materials and Methods Compared to other carotenoids, L and Z are the dominant 2.1. Participants. Community-dwelling older adults were carotenoids in the central nervous system (CNS) in both recruited for participation in a year-long randomized, early- and late-life, where they account for 66–77% of the double-blind, placebo-controlled trial evaluating the impact total carotenoid concentration in human brain tissue [8, 9]. of lutein (L) and zeaxanthin (Z) supplementation on vision, Although the cognitive effects of L and Z have been well cognitive functioning, and neural integrity. Recruitment established and there is a growing literature on the direct methods included newspaper advertisements, flyers, and neural effects, particularly regarding neural functioning electronic media (e.g., listservs). Exclusion criteria included and neural efficiency, much of what is known about the macular degeneration, corrected visual acuity worse than relation between L and Z and the brain has been de- 20 : 40, xanthophyll carotenoid supplementation within the termined through postmortem studies (e.g., [8–10]). Re- six-month period prior to enrollment (with the exception of cent randomized control trials (RCTs) have demonstrated multivitamins that contained less than 1 mg L + Z/day), that the effects of L and Z supplementation can be mea- gastric conditions known to impair absorption of nutritional sured at a neural level using functional neuroimaging supplements (e.g., gastric bypass or gastric ulcer), left- technology [11]. However, there remains limited literature handedness, traumatic brain injury, previous history of on the structural brain effects of L and Z, and, to our stroke, dementia, Parkinson’s disease or other neurological knowledge, the only published study that examined the condition known to impair cognitive function, and MRI effect of L and Z on brain structure in vivo was cross- incompatibility (e.g., cardiac pacemaker). sectional [12]. )us, the aim of the current study was to Sixty participants (23 males and 37 females), aged 65–92 extend previous literature on the relation between L and Z years (M � 72.3 years, SD � 6.77 years), met the inclusion and brain structure in older adults by using a randomized, criteria and were randomized into either the active supple- double-blind, placebo-controlled trial design to evaluate ment group (N � 43) or the placebo group (N � 17). Of the 60 the impact of L and Z supplementation on several metrics randomized participants, 47 participants (20 males and 27 of brain structure. females) aged 65–87 years (M � 71.8 years, SD � 6.04 years) Specifically, we examined if one year of supplemen- completed the study, with the final sample size of 33 par- tation of L and Z impacted brain volume in older adults ticipants in the supplement group and 14 participants in the using T1-weighted sequences and white matter micro- placebo group. A visual depiction of the study screening, structure using diffusion tensor imaging (DTI) MRI se- randomization, intervention, and attrition process can be quences. We examined global measures of brain volume found in Figure 1, consistent with CONSORT guidelines [17]. (i.e., global gray and white matter volume and white matter hypointensity volume) as well as specific regions-of-in- terest (ROIs) in the frontal and temporal lobes (i.e., prefrontal cortex, orbitofrontal cortex, anterior cingulate 2.2. Procedure. Eligible participants were randomly assigned to groups using a 2 :1 active supplement to placebo group cortex, medial temporal cortex, and hippocampus), as gray and white matter volume declines are typically seen first in ratio. Simple randomization was conducted by the study anterior regions of the brain in aging individuals (e.g., coordinator, who was not involved in data collection. A [13, 14]). We also examined global white matter micro- master list of participant randomization was kept confi- structure and integrity of several anterior white matter dential by the study coordinator. All study personnel, in- tracts (i.e., genu of the corpus callosum, fornix, and an- cluding the staff who performed the assessments, were terior cingulum) that are particularly vulnerable to age- blinded to participant randomization throughout the course related decline (e.g., [15, 16]). of the trial. Blinding was broken only after all data collection We hypothesized that L and Z supplementation would was complete and when necessary for statistical analysis of intervention effects. positively relate to brain structure such that the L and Z supplement group would increase, maintain, or show at- Both the active supplement and placebo were provided by DSM Nutritional Products (Besel, Switzerland). )e tenuated loss of their brain volume and white matter microstructure over the course of the trial while the active supplement contained 10 mg L and 2 mg Z. )e placebo group would show age-related declines in brain placebo was visually identical to the active supplement, and volume and white matter microstructure (i.e., lower both the supplement and placebo were contained in iden- fractional anisotropy (FA), higher mean diffusivity (MD), tical, opaque, sealed bottles with labels that were visually and higher radial diffusivity (RD)); axial diffusivity (AD) identical except for the randomization code on the label. was also examined as an exploratory measure of white )us, participants were also blinded to intervention con- matter microstructure but was not associated with di- dition. Participants were instructed to take one tablet from rectional hypotheses. Additionally, we hypothesized that L the bottle daily with a meal for 12 months. Participants completed several preintervention visits to and Z supplementation would negatively relate to white matter hypointensity volume such that the supplement collect visual, cognitive, and neuroimaging measures. Par- ticipants also completed follow-up visits at 4 months and 8 group was expected to maintain or attenuate increases of global white matter hypointensity volume, while the months for ongoing data collection. Compliance to the placebo group was expected to show age-related increases intervention was monitored through twice monthly tele- in these measures. phone calls and pill counts from bottles returned by the Journal of Aging Research 3 Assessed for eligibility (n = 82) Enrollment Excluded (n = 22) (i) Not meeting inclusion criteria (n = 22) (ii) Declined to participate (n = 0) Randomized (n = 60) 1:2 Placebo: intervention Allocation Allocated to intervention (n = 43) Allocated to placebo (n = 17) (i) Received allocated intervention (n = 42) (i) Received allocated placebo (n = 17) (ii) Did not receive allocated intervention (ii) Did not receive allocated placebo (n = 0) (withdrew due to death in the family and associated stress (n = 1) Follow-up Lost to follow-up (completed at least one baseline measure but failed to show up to testing appointments) (n = 8) Lost to follow-up (completed at least one baseline measure but failed to show up to Discontinued intervention (withdrawn from the testing appointments) (n = 3) study by study personnel due to noncompliance on four or more bimonthly compliance phone calls, or failure to maintain inclusion criteria) (n = 1) Analysis Analyzed (n = 14) Analyzed (n = 33) Excluded from analysis (no baseline MRI data Excluded from analysis (n = 0) due to claustrophobia) (n = 1) Figure 1: CONSORT –ow diagram. participants during follow-up visits. Participation could be Scale [19] to con‡rm eligibility. ˆe CDR is a semistructured discontinued by study personnel if individuals reported interview conducted with both participants and collateral noncompliance on four or more of the telephone check-ins; informants. ˆe interviewer rates an individual’s abilities in however, no participants were withdrawn from the study due six cognitive and functional domains; scores from each of to noncompliance. Postintervention data were collected at 12 these domains are then combined to create a global rating of dementia severity ranging from 0 (no dementia) to 3 (severe months and followed the same acquisition procedure as the preintervention data collection. Of note, although the larger dementia). A global score of 0.5 is often used as a proxy RCT included a more extensive battery of outcome measures, measure for mild cognitive impairment (MCI). ˆus, to the current project focused only on retinal L and Z data, ensure the cognitive health of the sample, individuals who together with structural neuroimaging data collected at pre- received a global rating of 0 or 0.5 were eligible for the study. and postintervention visits. Results from other outcomes can be found in Lindbergh et al. and Hammond et al. [11, 18]. 2.2.2. Macular Pigment Optical Density (MPOD). Retinal concentrations of L and Z were measured as macular pig- 2.2.1. Clinical Dementia Rating Scale (CDR). Dementia ment optical density (MPOD) and assessed using custom- severity was assessed using the Clinical Dementia Rating ized heterochromatic –icker photometry (cHFP). ˆis 4 Journal of Aging Research method of data acquisition has been well validated as an in vivo normalization, and atlas registration, with processing of each measure of macular pigment density and has been fully de- time point initiated from a within-subjects template that represents mean subject anatomy across time points [23]. scribed elsewhere (e.g., [20, 21]). Briefly, participants viewed a disc that was composed of two wavelengths of light (460 Following image processing, global gray matter, global white nanometer (nm) shortwave “blue” light and 570 nanometer matter, and global white matter hypointensity volume (mm ) (nm) midwave “green” light); the two wavelengths were pre- were extracted. )e Desikan-Killiany atlas [24] was used to sented in square-wave, counter-phase orientation, which extract region-of-interest (ROI) volumes (see Figure 2), and caused the disc to appear to “flicker.” )e task was customized all volumes were corrected for intracranial volume (ICV) to individual participants based on their critical flicker fusion prior to statistical analysis according to the formula: nor- frequency (CFF) values, which were measured in the same malized volume � raw volume – b (ICV × mean ICV), where b session. Participants then turned a knob to adjust the intensity is the slope of the regression of an ROI volume on ICV. When of the 460 nm light until it appeared to match the luminance of appropriate, right and left hemisphere values were summed to the 570 nm light, causing the “flickering” to cease. )is pro- create a single value for each ROI. cedure was conducted in both the foveal and parafoveal regions of the retina. MPOD was calculated as the log of the intensity of 2.4.2. White Matter Microstructure. Diffusion weighted im- 460 nm light required to match the 570 nm light in the fovea ages (DWIs) were preprocessed using the Oxford Centre’s (where macular pigment is the densest) compared to the log of Functional MRI of the Brain (FMRIB) Diffusion Toolbox (FTD) the intensity needed in the parafovea (where macular pigment [25]. Preprocessing followed a standard pipeline, including head is absent). MPOD data collection followed the same procedure motion and eddy current correction, brain extraction, correc- at both pre- and postintervention visits. tion of distortion via fieldmap processing, and estimation of diffusion tensors for each voxel. Following preprocessing, Tract- Based Spatial Statistics (TBSS) [26] was used to optimize reg- 2.3. Neuroimaging Acquisition. All images were acquired istration and create the mean diffusion images, which were using a General Electric Signa HDx 3T MRI scanner (GE; thinned to create mean diffusion skeletons that represent the Waukesha, WI, USA). A high-resolution 3D T1-weighted centers of all tracts common to the group of participants. )e fast spoiled gradient echo (FSPGR) sequence was used to Johns Hopkins University (JHU) ICBM-DTI-81 White Matter collect structural scans (TE � 3.2 ms; TR � 7.5 ms; flip Atlas [27] was used to create binary masks for each ROI, which angle � 20 ; 154 axial slices; slice thickness � 1.2 mm; were then multiplied with the diffusion skeletons to create FOV � 256 × 256 mm in a 256 × 256 matrix). )ese images skeletonized masks for each ROI (see Figure 3). We examined provided coverage from the top of the head to the brainstem, four standard diffusivity values for each ROI: fractional an- with a total acquisition time of 6 minutes and 20 seconds. isotropy (FA), mean diffusivity (MD), radial diffusivity (RD), Diffusion weighted imaging (DWI) scans were acquired and axial diffusivity (AD). Average diffusivity values were axially using a single-shot diffusion-weighted spin echo-EPI extracted from each skeletonized ROI and used in statistical sequence. Slices covered from the top of the head to the analysis. When appropriate, right and left hemisphere values brainstem and were acquired aligned to the anterior com- were summed to create a single mean value for each ROI. missure-posterior commissure line. Scan parameters in- cluded: TE � 3.2 ms, TR � 15900 ms, 90 flip angle, 60 interleaved slices, slice gap � 0 mm, 2 mm isotropic voxels, 2.5. Statistical Analysis. Baseline MPOD was used as a co- acquisition matrix � 128 ×128, FOV � 256 × 256 mm, par- variate in all analyses to ensure that intervention effects were allel acceleration factor � 2, b-value: 1000, and 30 optimized not due to variations in baseline L and Z concentrations gradient directions with three b0 images. Total scan time for alone. Four individuals in our supplement group had pos- the DWI acquisition was 9 minutes and 38 seconds. sible MCI, as assessed by the CDR. We conducted analyses Additionally, two pairs of magnitude and phase images both with and without these participants and found no were acquired for fieldmap-based unwarping of DWIs significant effect based on the removal of these outliers. (TE1 � 5.0 ms and TE2 � 7.2 ms, TR � 700 ms, 60 slices, slice )us, in order to improve statistical power, we conducted gap � 0 mm, 2 mm isotropic voxels, acquisition matrix � final analyses with the whole sample, controlling for both age 128 ×128, and FOV � 256 × 256 mm). Acquisition for each and baseline CDR scores, as age and level of cognitive pair of images took approximately 2 minutes and 20 seconds. impairment are strong predictors of brain structure in older All neuroimaging acquisition followed the same procedure adult populations (e.g., [13, 15]). at both pre- and postintervention visits. Changes in structural brain outcomes over time as a function of intervention condition were determined using analysis of covariance (ANCOVA). Global volumes (i.e., 2.4. Neuroimaging Processing global gray matter, white matter, and white matter hypo- 2.4.1. Brain Volume. T1-weighted structural images were intensity volumes) and global diffusivity measures (i.e., global processed and segmented using FreeSurfer (v 6.0) (http:// FA, RD, MD, and AD) were entered as dependent variables surfer.nmr.mgh.harvard.edu; [22]). Due to the longitudinal into a series of two-way mixed ANCOVAs with intervention design of the study, the FreeSurfer longitudinal processing group (active supplement vs. placebo) and timepoint (pre- vs. postintervention) as the independent variables and baseline stream was utilized, which includes motion correction, skull stripping, automated transformation to Talairach space, age, CDR scores, and MPOD as the covariates. Journal of Aging Research 5 Figure 2: Volumetric regions-of-interest (ROIs). )e figure depicts the masks used for volumetric ROI analyses from left to right: prefrontal cortex, orbitofrontal cortex, anterior cingulate cortex, medial temporal cortex, and hippocampus. Masks are superimposed on a repre- sentative T -weighted image from a participant in our sample. (a) (b) Figure 3: White matter microstructure regions-of-interest (ROIs). )e figure depicts the masks used for white matter microstructure ROI analyses for the genu (red), fornix (blue), and anterior cingulum (yellow) in the sagittal view (a) and axial view (b). Masks are superimposed on a single-subject diffusion-weighted template in MNI space provided by Johns Hopkins University (JHU) in FMRIB’s Software Library (FSL). )e mean skeleton for the sample is overlaid on the diffusion-weighted image in bright green. Region-of-interest (ROI) outcomes were similarly entered significantly at preintervention on age, education level, or together in groups of two-way mixed MANCOVAs with in- baseline MPOD concentrations. Chi-square tests also con- tervention group (active supplement vs. placebo) and time- firmed that the two groups did not differ significantly at point (pre- vs. postintervention) as the independent variables preintervention in terms of sex or level of cognitive im- and baseline age, CDR score, and MPOD as the covariates. )e pairment (CDR score). Similarly, there were no significant first group of analyses included ICV-corrected gray matter differences between participants who completed the study volume for orbitofrontal cortex, prefrontal cortex, anterior (N � 47) and participants who attrited (N � 13) with respect cingulate cortex, medial temporal cortex, and the hippocampus to age, sex, education, baseline MPOD, and cognitive im- in the conglomerate dependent variable. )e second group pairment (CDR score). included ICV-corrected subcortical white matter volume of the orbitofrontal, prefrontal, anterior cingulate, and medial tem- 3.1. Brain Volume. Pre- and postintervention values for both poral cortices in the conglomerate dependent variable. )e final group included white matter diffusivity values for the genu of global and regional brain volumes can be found in Table 2. When controlling for baseline age, MPOD, and cognitive the corpus callosum, fornix, and anterior cingulum in the conglomerate dependent variable. Analyses for each diffusivity impairment (CDR score), there were no significant main effects of group for any of the measures of global brain parameter (i.e., FA, MD, RD, and AD) were performed sep- arately. If the MANCOVAs reached significance, planned volume. Additionally, there were no significant main effects follow-up two-way mixed ANCOVA analyses were conducted of time, although both groups showed nonsignificant age- related changes in brain volumes (i.e., decreased global gray to determine changes in specific ROIs as a function of in- tervention condition and timepoint, controlling for baseline and white matter volume and white matter hypointensity volume). )ere were no significant group ∗ time interactions age, CDR score, and MPOD values. for any measures of global brain volume. A two-way mixed MANCOVA showed a significant 3. Results and Discussion main effect of time for gray matter ROIs (i.e., gray matter Demographic characteristics of the sample can be found in volume of the prefrontal cortex, orbitofrontal cortex, an- Table 1. Independent-samples t-tests confirmed that the terior cingulate cortex, medial temporal lobe, and hippo- supplement group and placebo group did not differ campus) (F � 3.02, p � 0.022), but no significant main effect 6 Journal of Aging Research Table 1: Preintervention characteristics. % or M (SD) Supplement (N � 33) Placebo (N � 14) Overall sample (N � 47) Age (years) 72.4 (6.27) 70.4 (5.43) 71.8 (6.04) Sex (% female) 51.5% 71.4 % 57.4% Race (% Caucasian) 100% 100% 100% Education (years) 16.6 (3.31) 16.7 (3.02) 16.6 (3.19) Cognitive impairment (%) No impairment (CDR � 0) 87.9% 100% 91.5% Mild impairment (CDR � 0.5) 12.1% 0% 8.5% Note: CDR � clinical dementia rating scale. Table 2: Brain volume. M (SD) in mm Supplement (N � 33) Placebo (N � 14) Overall sample (N � 47) Pre Post Pre Post Pre Post Gray matter Global 589258 (41075) 583090 (43892) 598888 (26886) 595608 (27435) 592127 (37387) 586819 (39829) Regions-of-interest (ROIs) Prefrontal 102435 (10156) 101378 (10354) 105206 (7147) 104517 (6733) 103260 (9372) 102313 (9460) Orbitofrontal 23839 (2388) 23809 (2208) 23430 (1645) 23590 (1719) 23717 (2184) 23744 (2058) Anterior cingulate 7989 (1319) 7910 (1447) 7887 (1055) 7864 (1102) 7958 (1236) 7896 (1341) Medial temporal 27361 (2589) 27234 (2558) 27063 (2342) 27234 (2658) 27212 (2496) 27234 (2559) Hippocampus 8681 (989) 8407 (1050) 9144 (743) 8962 (886) 8819 (939) 8572 (1027) White matter Global 437384 (74298) 432185 (81221) 427438 (40081) 423846 (39899) 434422 (65691) 429701 (71090) Regions-of-interest (ROIs) Prefrontal 84364 (8774) 83736 (10028) 83119 (7414) 82317 (7317) 83994 (8332) 83313 (9247) Orbitofrontal 18494 (2083) 18421 (2345) 18141 (1484) 18083 (1531) 18389 (1915) 18313 (9247) Anterior cingulate 9228 (902) 9160 (951) 9336 (696) 9294 (699) 9260 (840) 9200 (878) Medial temporal 17092 (2499) 16955 (2736) 16803 (2039) 16961 (2015) 17006 (2353) 16957 (2521) Lateral ventricle 33250 (15696) 35066 (16500) 28855 (11638) 30229 (12088) 31941 (14622) 33625 (15352) White matter hypointensities 7759 (14588) 8090 (16083) 5543 (4446) 5625 (4690) 7099 (12437) 7356 (13692) Note: All volumes are corrected for intracranial volume (ICV) and have been rounded to the nearest mm . indicates a significant change from pre- to postintervention, p< 0.05, controlling for age and Clinical Dementia Rating (CDR) score. 3.2. White Matter Microstructure. Pre- and postintervention of group or group∗ time interaction was observed. When conducting planned follow-up ANCOVAs, only changes values for both global and regional white matter micro- structure can be found in Table 3. When controlling for over time in medial temporal lobe volume, regardless of group status, were individually significant (F � 6.72, baseline age, MPOD, and cognitive impairment (CDR p � 0.013). )ere were no significant changes over time, score), results showed a significant main effect of time for between group differences, or group∗ time interactions for global FA (F � 5.31, p � 0.004), but no significant main effect prefrontal cortex, orbitofrontal cortex, anterior cingulate of group or group∗ time interaction. Contrary to hypoth- cortex, or hippocampal volume. eses, both groups showed a significant increase in global FA. Similarly, a two-way mixed MANCOVA showed a )ere were no significant main effects of time for global RD, significant main effect of time for white matter ROIs MD, or AD, and no significant group differences or (i.e., subcortical white matter volume of the prefrontal group∗ time interactions were observed. Two-way mixed MANCOVAs for ROIs (i.e., genu of the cortex, orbitofrontal cortex, anterior cingulate cortex, and medial temporal lobe) (F � 4.00, p � 0.016), but no corpus callosum, fornix, and anterior cingulum) showed a significant main effect of group or group∗ time in- significant main effect of time for FA (F � 5.31, p � 0.004), teraction. However, follow-up ANCOVAs showed that but no significant main effects of group or group∗ time changes over time in the anterior cingulate cortex, re- interactions. Follow-up ANCOVAs showed that changes gardless of group status, were individually significant over time in the anterior cingulum (F � 8.60, p � 0.005) and (F � 6.19, p � 0.017). No other significant main effects of fornix (F � 8.29, p � 0.006), regardless of group status, were time, main effects of group, or group∗ time interactions individually significant. No other significant main effects of were observed for subcortical prefrontal cortex, orbi- time, main effects of group, or group∗ time interactions tofrontal cortex, or medial temporal lobe white matter were observed for genu FA and RD, MD, or AD values in the volume. ROIs. Journal of Aging Research 7 Table 3: White matter microstructure. M (SD) Supplement (N � 33) Placebo (N � 14) Overall sample (N � 47) Pre Post Pre Post Pre Post Fractional anisotropy (FA) 0.55076 0.55257 0.55384 0.56195 0.55167 0.55537 Global (0.02855) (0.03287) (0.01966) (0.02032) (0.02604) (0.02978) Regions-of-interest (ROIs) 0.61228 0.63634 0.63892 0.62361 Genu 0.61821 (.04982) 0.62021 (0.04750) (0.05182) (0.03231) (0.02889) (0.04573) 0.62050 0.62390 0.63488 0.64467 0.62478 0.63008 Anterior cingulum (0.05117) (0.05587) (0.03001) (0.03492) (0.04606) (0.05107) 0.37296 0.37186 0.37936 0.37860 0.37487 0.37387 Fornix (0.09130) (0.10075) (0.08800) (0.10308) (0.08942) (0.10037) Radial diffusivity (RD) 0.00053 0.00051 0.00050 0.00053 Global 0.00053(0.00005) 0.00052 (0.00006) (0.00061) (0.00003) (0.00003) (0.00005) Regions-of-interest (ROIs) 0.00050 0.00052 0.00047 0.00047 0.00049 Genu 0.00050 (0.00009) (0.00010) (0.00010) (0.00049) (0.00042) (0.00009) 0.00053 0.00052 0.00049 0.00048 0.00052 Anterior cingulum 0.00051 (0.00009) (0.00009) (0.00010) (0.00006) (0.00006) (0.00009) 0.00156 0.00158 0.00154 0.00156 0.00155 Fornix 0.00157 (0.00043) (0.00040) (0.00043) (0.00047) (0.00045) (0.00041) Mean diffusivity (MD) 0.00081 0.00081 0.00079 0.00078 0.00081 Global 0.00080 (0.00005) (0.00005) (0.00005) (0.00003) (0.00003) (0.00042) Regions-of-interest (ROIs) 0.00086 0.00087 0.00083 0.00083 0.00085 Genu 0.00086 (0.00008) (0.00009) (0.00009) (0.00004) (0.00003) (0.00008) 0.00090 0.00090 0.00087 0.00085 0.00089 Anterior cingulum 0.00088 (0.00007) (0.00008) (0.00008) (0.00005) (0.00004) (0.00007) 0.00191 0.00193 0.00189 0.00191 0.00190 Fornix 0.00193 (0.00038) (0.00036) (0.00038) (0.00044) (0.00039) (0.00038) Axial diffusivity (AD) 0.00138 0.00138 0.00135 0.00134 0.00137 Global 0.00137 (0.00005) (0.00004) (0.00005) (0.00004) (0.00003) (0.00004) Regions-of-interest (ROIs) 0.00156 0.00158 0.00155 0.00155 0.00156 Genu 0.00157 (0.00007) (0.00010) (0.00008) (0.00004) (0.00003) (0.00008) 0.00164 0.00166 0.00161 0.00160 0.00163 Anterior cingulum 0.00164 (0.00006) (0.00007) (0.00006) (0.00005) (0.00003) (0.00006) 0.00261 0.00264 0.00260 0.00261 0.00260 Fornix 0.00263 (0.00030) (0.00029) (0.00301) (0.00038) (0.00030) (0.00032) Note: indicates a significant change from pre- to postintervention, p � 0.05, controlling for age and Clinical Dementia Rating (CDR) score. 3.3. Intervention Response. To confirm intervention effec- individuals in the supplement group appearing to fail to tiveness, paired-samples t-tests were conducted in both the respond to intervention (i.e., showing stable or decreased MPOD concentrations). To explore this heterogeneity, in- supplement group and placebo group to assess statistically significant changes in MPOD concentrations over the course dividuals were classified into two categories: (1) those who of the intervention. Analyses confirmed that the supplement showed an increase in MPOD of 0.10 + log units from pre- to group showed a significant increase in MPOD (t � 2.27, postintervention were classified as “responders” and (2) all p � 0.030) over the course of the trial, while the placebo others who showed a stable, decreased, or nonsignificant group did not show any significant changes in MPOD increase (<0.10 log units) in MPOD from pre- to post- (t � 0.788, p � 0.445). However, there was heterogeneity in intervention were classified as “nonresponders” (see Table 4). both groups, with some individuals in the placebo group )en, exploratory analyses were undertaken to determine if showing increased MPOD concentrations and some there were any brain changes corresponding to increased L 8 Journal of Aging Research Table 4: Treatment response–change in MPOD concentrations. and Z concentrations within the supplement group (“in- tervention responders”). ˆe same analyses were repeated in M (SD) the placebo group to determine if increases in MPOD, in the Preintervention Postintervention absence of supplementation, were associated with brain Supplement group structural changes. As with the primary analyses above, all Responder (N  15) 0.4547 (0.1992) 0.6947 (0.2329) exploratory analyses were controlled for baseline age, MPOD, Nonresponder (N  18) 0.5717 (0.1658) 0.5111 (0.1705) and CDR scores. Placebo group 0.4414 (0.0373) 0.4843 (0.0536) Independent-samples t-tests con‡rmed that there was no ∗ Responder (N  7) 0.3871 (0.1602) 0.5943 (0.1605) signi‡cant dižerence between treatment “responders” and Nonresponder (N  7) 0.4957 (0.9888) 0.3743 (0.1820) “nonresponders” in either group with respect to baseline age, Total sample 0.4955 (0.0259) 0.5617 (0.0317) education, and MPOD concentrations. Additionally, chi- Responder (N  22) 0.4332 (0.1866) 0.6627 (0.2141) Nonresponder (N  25) 0.5504 (0.1521) 0.4728 (0.1811) square tests con‡rmed that responders and nonresponders were not dižerent with regard to sex and cognitive im- Note: indicates a signi‡cant change from pre- to postintervention of p < 0.05. pairment (CDR scores). In the supplement group, those who were classi‡ed as intervention responders (N  15) showed signi‡cantly less decline in total gray matter volume (∆R  0.144, F  5.37, 6.2 p  0.028) (see Figure 4) and prefrontal cortex gray matter 6.1 volume (∆R  0.125, F  4.26, p  0.048) (see Figure 5) than those classi‡ed as nonresponders (N  18). However, in the 5.9 placebo group, there were no signi‡cant dižerences in any brain volume measures between individuals who showed 5.8 increased MPOD (N  7) versus those who showed stable or 5.7 decreased MPOD (N  7). ˆere were also no signi‡cant 5.6 dižerences in global or regional white matter microstructure 5.5 measures between those classi‡ed as responders versus Pre Post nonresponders in either the supplement or placebo groups. Responder Nonresponder 4. Conclusions Figure 4: Changes in total gray matter volume. ˆe ‡gure shows 3 5 ˆe current study tested whether one year of supplemen- changes in total gray matter volume (mm ×10 ) between sup- plement group “responders” and “nonresponders.” tation with lutein (L) and zeaxanthin (Z), two xanthophyll carotenoids with known antioxidative properties and cog- nitive bene‡ts, could prevent or slow age-related structural brain changes in a sample of community-dwelling older 1.08 adults. Using a randomized, double-blind, placebo-con- 1.06 trolled trial design, we hypothesized that older adults re- ceiving supplementation with L and Z would increase, 1.04 maintain, or show attenuated loss of brain volume and white 1.02 matter microstructure in areas that are vulnerable to age- related decline (i.e., frontal and temporal gray and white 1 matter regions) relative to older adults receiving a placebo. 0.98 Results showed expected age-related declines for frontal 0.96 and medial-temporal gray and white matter, particularly Pre Post medial temporal lobe gray matter and subcortical white Responder matter of the anterior cingulate cortex, across both groups Nonresponder over the course of the trial. However, L and Z supplemen- tation did not appear to in–uence this loss. No signi‡cant Figure 5: Changes in prefrontal cortex volume. ˆe ‡gure shows 3 5 group dižerences or changes over time were observed in changes in total gray matter volume (mm ×10 ) between sup- global brain volume outcomes (i.e., global gray matter, global plement group “responders” and “nonresponders.” white matter, and white matter hypointensity volume). Ad- ditionally, no interactions between group and time were found for any of the brain volume measures. Average percent and placebo groups. Results did show an average increase in change for both global volumes and frontal-temporal volumes global FA, across both groups. Signi‡cant changes were also ranged from less than 1% to 3%, consistent with previous seen in anterior white matter tracts, including increased FA estimates of the average annual rate of gray and white matter in the anterior cingulum and an expected, age-related de- volume decline in older adults [14, 28]. cline in fornix FA. While it is encouraging to see im- We also did not ‡nd signi‡cant dižerences in white provements in global and anterior cingulum white matter matter microstructure outcomes between the supplement microstructure, it is important to note that these changes Journal of Aging Research 9 were seen in both the supplement and placebo groups and, population at baseline, prior to supplementation. In this thus, cannot be attributed to the intervention per se. Average respect, it is encouraging that a small dietary change percent change for both global and frontal-temporal white provided benefit to even a subsample of individuals in our matter microstructure ranged from less than 1% to around study who were already well nourished, educated, and 2%, again consistent with estimates of annual changes in affluent. Finally, as with any nutritional intervention, there diffusion metrics in healthy older adult populations [29, 30]. was no true control condition; our entire sample has Although results confirmed that the intervention ma- presumably been exposed to both L and Z throughout their nipulation was effective (i.e., the supplement group showed a lifetimes, and even those participants in the placebo significant increase in retinal L and Z concentrations, while condition continued to consume a normal diet for the the placebo group did not), there was individual variability duration of the trial, which likely contained some amount in both groups. )is individual variability is not uncommon of L and Z which are naturally occurring in many foods. It for nutritional intervention studies. )ere are many factors is possible that the biggest effects of L and Z on the brain which may affect how one processes and absorbs nutritional could be driven by deficiency. While the current study found that one year of supplements. For example, some studies have suggested that cholesterol levels and cosupplementation with other nu- supplementation of L and Z had limited effects on the trients can impact the absorption and bioavailability of L and brain structural integrity of community-dwelling older Z [31, 32]. Additionally, the very nature of being in a nu- adults, there is a growing body of literature to suggest that tritional study may cause some participants to choose these nutrients are important for other aspects of cog- healthier and more nutritious foods, whether consciously or nitive and brain health. Other analyses from the current unconsciously. )us, exploratory analyses were conducted RCT have shown that supplementation with L and Z to determine whether individuals who responded better to benefited cognition, particularly complex attention, the intervention, or who otherwise increased their retinal L cognitive flexibility [18], and verbal learning [11], as well and Z concentration, showed better neural outcomes than as neural functioning in dorsolateral prefrontal cortex those who showed no change or decreased retinal L and Z. and anterior cingulate, areas that are vulnerable to age- related decline [11]. )e importance of L and Z for While there were no differences in terms of white matter microstructure outcomes, results showed that in the sup- cognitive and neural outcomes has been replicated cross- plement group, intervention “responders” (i.e., increased sectionally in other samples and by other researchers retinal L and Z) had significantly less decline in global and [9, 37–39] and is beginning to be confirmed longitudi- prefrontal gray matter volume than intervention “non- nally as well [40, 41]. responders” (i.e., stable or decreased retinal L and Z), even )e strength of this study lies in the novel approach after controlling for factors such as baseline age, MPOD, and used to investigate in vivo structural brain outcomes of a cognitive impairment. )ere were no concurrent differences year-long, double-blind, placebo-controlled trial of lutein observed in the placebo group. )ese results suggest that L (L) and zeaxanthin (Z) supplementation in older adults. and Z may slow age-related gray matter decline for a subset Exploratory analyses suggested that there was a small group of individuals who appeared to reap greater benefit from the of individuals who reaped greater benefit from L and Z supplementation regimen. supplementation and who showed less decline in global and As with any study, the present trial had limitations. For prefrontal gray matter volume than individuals who did not example, a similar prospective study on the effects of Vi- appear to benefit from the intervention. To that end, future tamin B on brain volume found significant results but studies could explore factors which may impact the efficacy followed the older subjects for five years as opposed to only of L and Z supplementation, such as vascular health, one [33]. )us, longer time periods may be necessary to cosupplementation with other nutrients, and other factors allow enough sensitivity to discriminate how dietary known to interact with nutrient bioavailability and ab- components influence morphological loss. Additionally, we sorption. Replication of this study in a larger and more did not have access to biomarkers, such as blood vitamin diverse sample may also determine whether there are some levels or lipid metabolites, in order to analyse the effect that individuals who could benefit even more from L and Z these individual differences might have on intervention supplementation, such as individuals who are more cog- nitively impaired or have had less access to nutritious diets efficacy. Our sample size was small (e.g., the B study [33] tested 107 older adults) and suffered from considerable across their lifetime. Factors such as supplement dosage and particularly a longer intervention duration should also attrition (approximately 22%), which may have limited our power to detect certain group differences and intervention be considered when determining intervention efficacy. effects. Additionally, our sample was homogenous in terms Finally, future studies will need to determine the long-term of background, consisting of 100% Caucasian and pre- effects of L and Z supplementation on outcomes such as the dominantly cognitively healthy and highly educated in- development of dementia and accumulation of brain pa- dividuals. Research has shown that more educated people thologies. While L and Z supplementation over one year also tend to have healthier diets [34], and our data support appears to have limited effects on structural brain outcomes this trend. Baseline mean MPOD values for our sample in older adults, it is likely not harmful [42] and appears to were slightly higher than other published data for older provide benefits for other aspects of cognitive and brain health that could improve or extend quality of life for older adults [35, 36], suggesting that our sample may have consumed a more nutritious diet than the general adults. 10 Journal of Aging Research [9] E. J. Johnson, R. Vishwanathan, M. A. Johnson et al., “Re- Data Availability lationship between serum and brain carotenoids, α-tocoph- erol, and retinol concentrations and cognitive performance in )e data used to support the findings of this study are the oldest old from the Georgia centenarian study,” Journal of available from the corresponding author upon request. Aging Research, vol. 2013, Article ID 951786, 13 pages, 2013. [10] T. den Heijer, L. J. Launer, J. C. de Groot et al., “Serum Conflicts of Interest carotenoids and cerebral white matter lesions: the Rotterdam scan study,” Journal of the American Geriatrics Society, vol. 49, During a portion of data collection time, LMRH was no. 5, pp. 642–646, 2001. employed by Abbott Nutrition while holding a joint ap- [11] C. A. Lindbergh, L. M. Renzi-Hammond, B. R. Hammond pointment at the University of Georgia. No other conflicts of et al., “Lutein and zeaxanthin influence brain function in older adults: a randomized controlled trial,” Journal of the In- interest exist for the study authors, including CMM, CAL, ternational Neuropsychological Society, vol. 23, no. 1, BRH, and LSM. All statistical analyses were completed in- pp. 77–90, 2018. dependent of the supporting agencies. [12] C. M. Mewborn, D. P. Terry, L. M. Renzi-Hammond, B. R. Hammond, and L. S. Miller, “Relation of retinal and Acknowledgments serum lutein and zeaxanthin to white matter integrity in older adults: a diffusion tensor imaging study,” Archives of Clinical )is work was supported by Abbott Nutrition, Columbus, Neuropsychology, vol. 33, no. 7, pp. 861–874, 2017. OH, USA (research grant to LMRH, BRH, and LSM); )e [13] O. Abe, H. Yamasue, S. Aoki et al., “Aging in the CNS: University of Georgia’s Bio-Imaging Research Center (ad- comparison of gray/white matter volume and diffusion tensor data,” Neurobiology of Aging, vol. 29, no. 1, pp. 102–116, 2008. ministrative support to LSM); and DSM Nutritional Prod- [14] D. H. Salat, J. A. Kaye, and J. S. Janowsky, “Prefrontal gray and ucts (Switzerland) who provided the supplements and white matter volumes in healthy aging and Alzheimer dis- placebos for the randomized-controlled trial from which the ease,” Archives of Neurology, vol. 56, no. 3, pp. 338–344, 1999. current data were derived. A version of this work was th [15] I. J. Bennett and D. J. Madden, “Disconnected aging: cerebral presented at the 47 Annual Meeting of the International white matter integrity and age-related differences in cogni- Neuropsychological Society. Published Abstract: Journal tion,” Neuroscience, vol. 276, no. 6, pp. 187–205, 2014. of the International Neuropsychological Society, 25(s1). doi: [16] D. Head, R. L. Buckner, J. S. Shimony et al., “Differential 10.1017/S1355617719000663. vulnerability of anterior white matter in nondemented aging with minimal acceleration in dementia of the Alzheimer type: evidence from diffusion tensor imaging,” Cerebral Cortex, References vol. 14, no. 4, pp. 410–423, 2004. [17] K. F. Schulz, D. G. Altman, and D. Moher, “CONSORT 2010 [1] M. Couture, N. Larivi`ere, and R. Lefrançois, “Psychological statement: updated guidelines for reporting parallel group distress in older adults with low functional independence: a randomized trials,” Annals of Internal Medicine, vol. 152, multidimensional perspective,” Archives of Gerontology and no. 11, pp. 726–732, 2010. Geriatrics, vol. 41, no. 1, pp. 101–111, 2005. [18] B. R. Hammond, L. S. Miller, M. O. Bello, C. A. Lindbergh, [2] A. N. Puente, D. P. Terry, C. C. Faraco, C. L. Brown, and C. Mewborn, and L. M. Renzi-Hammond, “Effects of lutein/ L. S. Miller, “Functional impairment in mild cognitive im- zeaxanthin supplementation on the cognitive function of pairment evidenced using performance-based measurement,” community dwelling older adults: a randomized, double- Journal of Geriatric Psychiatry and Neurology, vol. 27, no. 4, masked, placebo-controlled trial,” Frontiers in Aging Neuro- pp. 253–258, 2014. science, vol. 9, 2017. [3] D. Harman, “Aging: a theory based on free radical and ra- [19] J. C. Morris, “)e Clinical dementia rating (CDR): current diation chemistry,” Journal of Gerontology, vol. 11, pp. 289– version and scoring rules,” Neurology, vol. 43, no. 11, p. 2412, 300, 1956. [4] M. Abdollahi, M. Y. Moridani, O. I. Aruoma, and [20] B. R. Wooten and B. R. Hammond, “Spectral absorbance and S. Mostafalou, “Oxidative stress in aging,” Oxidative Medicine spatial distribution of macular pigment using heterochro- and Cellular Longevity, vol. 2014, Article ID 876834, 2 pages, matic flicker photometry,” Optometry and Vision Science, vol. 82, no. 5, pp. 378–386, 2005. [5] Y. Gu, A. M. Brickman, Y. Stern et al., “Mediterranean diet [21] B. R. Wooten, B. R. Hammond, R. I. Land, and and brain structure in a multiethnic elderly cohort,” Neu- D. M. Snodderly, “A practical method for measuring macular rology, vol. 85, no. 20, pp. 1744–1751, 2015. pigment optical density,” Investigative Ophthalmology and [6] C. A. Raji, K. I. Erickson, O. L. Lopez et al., “Regular fish consumption and age-related brain gray matter loss,” Visual Science, vol. 40, no. 11, pp. 2481–2489, 1999. [22] B. Fischl, D. H. Salat, E. Busa et al., “Whole brain segmen- American Journal of Preventive Medicine, vol. 47, no. 4, pp. 444–451, 2014. tation,” Neuron, vol. 33, no. 3, pp. 341–355, 2002. [23] M. Reuter, N. J. Schmansky, H. D. Rosas, and B. Fischl, [7] N. Scarmeas, J. A. Luchsinger, Y. Stern et al., “Mediterranean diet and magnetic resonance imaging-assessed cerebrovas- “Within-subject template estimation for unbiased longitu- dinal image analysis,” NeuroImage, vol. 61, no. 4, pp. 1402– cular disease,” Annals of Neurology, vol. 69, no. 2, pp. 257–268, 2011. 1418, 2012. [24] R. S. Desikan, F. Segonne, ´ B. Fischl et al., “An automated [8] N. E. Craft, T. B. Haitema, K. M. Garnett, K. A. Fitch, and C. K. Dorey, “Carotenoid, tocopherol, and retinol concen- labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest,” NeuroImage, trations in elderly human brain,” 6e Journal of Nutrition, Health and Aging, vol. 8, no. 3, pp. 156–162, 2004. vol. 31, no. 3, pp. 968–980, 2006. Journal of Aging Research 11 [25] T. E. J. Behrens, M. W. Woolrich, M. Jenkinson et al., elderly population: results of the EVA study,” 6e Journals of Gerontology Series A: Biological Sciences and Medical Sciences, “Characterization and propagation of uncertainty in diffu- sion-weighted MR imaging,” Magnetic Resonance in Medi- vol. 62, no. 3, pp. 308–316, 2007. [41] E. J. Johnson, H.-Y. Chung, S. M. Caldarella, and cine, vol. 50, no. 5, pp. 1077–1088, 2003. [26] S. M. Smith, M. Jenkinson, H. Johansen-Berg et al., “Tract- D. M. Snodderly, “)e influence of supplemental lutein and docosahexaenoic acid on serum, lipoproteins, and macular based spatial statistics: voxelwise analysis of multi-subject pigmentation,” 6e American Journal of Clinical Nutrition, diffusion data,” NeuroImage, vol. 31, no. 4, pp. 1487–1505, vol. 87, no. 5, pp. 1521–1529, 2008. [42] )e Age-Related Eye Disease Study 2 (AREDS2) Research [27] S. Mori, K. Oishi, H. Jiang et al., “Stereotaxic white matter Group, “Lutein + zeaxanthin and omega-3 fatty acids for age- atlas based on diffusion tensor imaging in an ICBM template,” related macular degeneration: the Age-Related Eye Disease NeuroImage, vol. 40, no. 2, pp. 570–582, 2008. Study 2 (AREDS2) randomized clinical trial,” JAMA, vol. 309, [28] T. L. Jernigan, S. L. Archibald, C. Fennema-Notestine et al., no. 19, pp. 2005–2015, 2013. “Effects of age on tissues and regions of the cerebrum and cerebellum,” Neurobiology of Aging, vol. 22, no. 4, pp. 581–594, [29] T. R. Barrick, R. A. Charlton, C. A. Clark, and H. S. Markus, “White matter structural decline in normal ageing: a pro- spective longitudinal study using tract-based spatial statis- tics,” NeuroImage, vol. 51, no. 2, pp. 565–577, 2010. [30] S. J. Teipel, T. Meindl, M. Wagner et al., “Longitudinal changes in fiber tract integrity in healthy aging and mild cognitive impairment: a DTI follow-up study,” Journal of Alzheimer’s Disease, vol. 22, no. 2, pp. 507–522, 2010. [31] Y. Sato, R. Suzuki, M. Kobayashi et al., “Involvement of cholesterol membrane transporter Niemann-Pick C1-Like 1 in the intestinal absorption of lutein,” Journal of Pharmacy and Pharmaceutical Sciences, vol. 15, no. 2, pp. 256–264, 2012. [32] S. A. Tanumihardjo, J. Li, and M. P. Dosti, “Lutein absorption is facilitated with cosupplementation of ascorbic acid in young adults,” Journal of the American Dietetic Association, vol. 105, no. 1, pp. 114–118, 2005. [33] A. Vogiatzoglou, H. Refsum, C. Johnston et al., “Vitamin B12 status and rate of brain volume loss in community-dwelling elderly,” Neurology, vol. 71, no. 11, pp. 826–832, 2008. [34] T. N. Akbaraly, A. Singh-Manoux, M. G. Marmot, and E. J. Brunner, “Education attenuates the association between dietary patterns and cognition,” Dementia and Geriatric Cognitive Disorders, vol. 27, no. 2, pp. 147–154, 2009. [35] L. M. Renzi, M. J. Dengler, A. Puente, L. S. Miller, and B. R. Hammond, “Relationships between macular pigment optical density and cognitive function in unimpaired and mildly cognitively impaired older adults,” Neurobiology of Aging, vol. 35, no. 7, pp. 1695–1699, 2014. [36] L. M. Renzi, B. J. Hammond, M. Dengler, and R. Roberts, “)e relation between serum lipids and lutein and zeaxanthin in the serum and retina: results from cross-sectional, case-control and case study designs,” Lipids in Health and Disease, vol. 11, no. 33, 2012. [37] S. Ajana, D. Weber, C. Helmer et al., “Plasma concentrations of lutein and zeaxanthin, macular pigment optical density, and their associations with cognitive performances among older adults,” Investigative Opthalmology and Visual Science, vol. 59, no. 5, pp. 1828–1835, 2018. [38] J. Feeney, C. Finucane, G. M. Savva et al., “Low macular pigment optical density is associated with lower cognitive performance in a large, population-based sample of older adults,” Neurobiology of Aging, vol. 34, no. 11, pp. 2449–2456, [39] R. Vishwanathan, A. Iannaccone, T. M. Scott et al., “Macular pigment optical density is related to cognitive function in older people,” Age and Ageing, vol. 43, no. 2, pp. 271–275, [40] N. T. Akbaraly, H. Faure, V. Gourlet, A. Favier, and C. Berr, “Plasma carotenoid levels and cognitive performance in an MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 International Journal of Journal of Immunology Research Endocrinology Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com BioMed PPAR Research Research International Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2013 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Neurology Research and Treatment Cellular Longevity Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Aging Research Hindawi Publishing Corporation

The Effects of Lutein and Zeaxanthin Supplementation on Brain Morphology in Older Adults: A Randomized, Controlled Trial

Loading next page...
 
/lp/hindawi-publishing-corporation/the-effects-of-lutein-and-zeaxanthin-supplementation-on-brain-ozG1V8xaA5
Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2019 Catherine M. Mewborn et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
2090-2204
eISSN
2090-2212
DOI
10.1155/2019/3709402
Publisher site
See Article on Publisher Site

Abstract

Hindawi Journal of Aging Research Volume 2019, Article ID 3709402, 11 pages https://doi.org/10.1155/2019/3709402 Research Article The Effects of Lutein and Zeaxanthin Supplementation on Brain Morphology in Older Adults: A Randomized, Controlled Trial 1 1 1 Catherine M. Mewborn, Cutter A. Lindbergh, B. Randy Hammond, 1,2 1 Lisa M. Renzi-Hammond, and L. Stephen Miller Department of Psychology, University of Georgia, Athens, GA 30605, USA Institute of Gerontology, Department of Health Promotions and Behavior, College of Public Health, University of Georgia, Athens, GA 30605, USA Correspondence should be addressed to L. Stephen Miller; lsmiller@uga.edu Received 27 July 2019; Accepted 29 October 2019; Published 1 December 2019 Academic Editor: Carmela R. Balistreri Copyright © 2019 Catherine M. Mewborn et al. )is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. A growing literature emphasizes the importance of lifestyle factors such as nutrition in successful aging. )e current study examined if one year of supplementation with lutein (L) and zeaxanthin (Z), two nutrients with known antioxidative properties and cognitive benefits, impacted structural brain outcomes in older adults using a double-blind, randomized, placebo-controlled trial design. Community-dwelling older adults (20 males and 27 females) aged 65–87 years (M � 71.8 years, SD � 6.04 years) were randomized into supplement (N � 33) and placebo groups (N � 14) using simple randomization. )e supplement group received 10 mg L + 2 mg Z daily for 12 months while the placebo group received a visually identical, inert placebo. L and Z were measured via retinal concentrations (macular pigment optical density or MPOD). Structural brain outcomes, focusing on global and frontal- temporal lobe regions, were acquired using both T1-weighted and DTI MRI sequences. We hypothesized that the supplement group would increase, maintain, or show attenuated loss in hypothesized regions-of-interest (ROIs) while the placebo group would show age-related declines in brain structural integrity over the course of the trial. While results showed age-related declines for frontal and temporal gray and white matter volumes, as well as fornix white matter microstructure across both groups, only minimal differences were found between the supplement and placebo groups. However, exploratory analyses showed that individuals who responded better to supplementation (i.e., showed greater increases in MPOD) showed less decline in global and prefrontal gray matter volume than supplement “nonresponders.” While results suggest that one year of L and Z supplementation may have limited effects on structural brain outcomes overall, there may be a subsample of individuals for whom supplementation of L and Z provides greater benefits. ClinicalTrials.gov number, NCT02023645. structure and cognitive functioning [4]. To combat the 1. Introduction negative effects of oxidation, researchers have studied nu- Aging is associated with many changes, both cognitive and trients such as vitamins, flavonoids, and carotenoids for their neural, that contribute to negative outcomes such as de- potential in preventing and treating age-related cognitive creased functional independence, significant personal and and neural decline. Intake of these nutrients, along with societal economic burden, and psychological distress for healthy fatty acids and adherence to a balanced healthy diet, both aging individuals and their caregivers [1, 2]. One of the has been associated with positive neural effects, including more common theories of biological aging is the Free preserved gray and white matter volume, white matter Radical/Oxidative Stress )eory of Aging [3], which states microstructure, and lower risk of cerebral infarcts, even after that oxidative stress causes damage to DNA and proteins. In controlling for demographics and vascular risk factors [5–7]. turn, oxidation leads to neural inflammation, neurotoxicity, Lutein (L) and zeaxanthin (Z) are two nutrients in the reduced cerebral perfusion, and disruption of neural xanthophyll carotenoid family that have been suggested to 2 Journal of Aging Research benefit cognition and neural outcomes in older adults. 2. Materials and Methods Compared to other carotenoids, L and Z are the dominant 2.1. Participants. Community-dwelling older adults were carotenoids in the central nervous system (CNS) in both recruited for participation in a year-long randomized, early- and late-life, where they account for 66–77% of the double-blind, placebo-controlled trial evaluating the impact total carotenoid concentration in human brain tissue [8, 9]. of lutein (L) and zeaxanthin (Z) supplementation on vision, Although the cognitive effects of L and Z have been well cognitive functioning, and neural integrity. Recruitment established and there is a growing literature on the direct methods included newspaper advertisements, flyers, and neural effects, particularly regarding neural functioning electronic media (e.g., listservs). Exclusion criteria included and neural efficiency, much of what is known about the macular degeneration, corrected visual acuity worse than relation between L and Z and the brain has been de- 20 : 40, xanthophyll carotenoid supplementation within the termined through postmortem studies (e.g., [8–10]). Re- six-month period prior to enrollment (with the exception of cent randomized control trials (RCTs) have demonstrated multivitamins that contained less than 1 mg L + Z/day), that the effects of L and Z supplementation can be mea- gastric conditions known to impair absorption of nutritional sured at a neural level using functional neuroimaging supplements (e.g., gastric bypass or gastric ulcer), left- technology [11]. However, there remains limited literature handedness, traumatic brain injury, previous history of on the structural brain effects of L and Z, and, to our stroke, dementia, Parkinson’s disease or other neurological knowledge, the only published study that examined the condition known to impair cognitive function, and MRI effect of L and Z on brain structure in vivo was cross- incompatibility (e.g., cardiac pacemaker). sectional [12]. )us, the aim of the current study was to Sixty participants (23 males and 37 females), aged 65–92 extend previous literature on the relation between L and Z years (M � 72.3 years, SD � 6.77 years), met the inclusion and brain structure in older adults by using a randomized, criteria and were randomized into either the active supple- double-blind, placebo-controlled trial design to evaluate ment group (N � 43) or the placebo group (N � 17). Of the 60 the impact of L and Z supplementation on several metrics randomized participants, 47 participants (20 males and 27 of brain structure. females) aged 65–87 years (M � 71.8 years, SD � 6.04 years) Specifically, we examined if one year of supplemen- completed the study, with the final sample size of 33 par- tation of L and Z impacted brain volume in older adults ticipants in the supplement group and 14 participants in the using T1-weighted sequences and white matter micro- placebo group. A visual depiction of the study screening, structure using diffusion tensor imaging (DTI) MRI se- randomization, intervention, and attrition process can be quences. We examined global measures of brain volume found in Figure 1, consistent with CONSORT guidelines [17]. (i.e., global gray and white matter volume and white matter hypointensity volume) as well as specific regions-of-in- terest (ROIs) in the frontal and temporal lobes (i.e., prefrontal cortex, orbitofrontal cortex, anterior cingulate 2.2. Procedure. Eligible participants were randomly assigned to groups using a 2 :1 active supplement to placebo group cortex, medial temporal cortex, and hippocampus), as gray and white matter volume declines are typically seen first in ratio. Simple randomization was conducted by the study anterior regions of the brain in aging individuals (e.g., coordinator, who was not involved in data collection. A [13, 14]). We also examined global white matter micro- master list of participant randomization was kept confi- structure and integrity of several anterior white matter dential by the study coordinator. All study personnel, in- tracts (i.e., genu of the corpus callosum, fornix, and an- cluding the staff who performed the assessments, were terior cingulum) that are particularly vulnerable to age- blinded to participant randomization throughout the course related decline (e.g., [15, 16]). of the trial. Blinding was broken only after all data collection We hypothesized that L and Z supplementation would was complete and when necessary for statistical analysis of intervention effects. positively relate to brain structure such that the L and Z supplement group would increase, maintain, or show at- Both the active supplement and placebo were provided by DSM Nutritional Products (Besel, Switzerland). )e tenuated loss of their brain volume and white matter microstructure over the course of the trial while the active supplement contained 10 mg L and 2 mg Z. )e placebo group would show age-related declines in brain placebo was visually identical to the active supplement, and volume and white matter microstructure (i.e., lower both the supplement and placebo were contained in iden- fractional anisotropy (FA), higher mean diffusivity (MD), tical, opaque, sealed bottles with labels that were visually and higher radial diffusivity (RD)); axial diffusivity (AD) identical except for the randomization code on the label. was also examined as an exploratory measure of white )us, participants were also blinded to intervention con- matter microstructure but was not associated with di- dition. Participants were instructed to take one tablet from rectional hypotheses. Additionally, we hypothesized that L the bottle daily with a meal for 12 months. Participants completed several preintervention visits to and Z supplementation would negatively relate to white matter hypointensity volume such that the supplement collect visual, cognitive, and neuroimaging measures. Par- ticipants also completed follow-up visits at 4 months and 8 group was expected to maintain or attenuate increases of global white matter hypointensity volume, while the months for ongoing data collection. Compliance to the placebo group was expected to show age-related increases intervention was monitored through twice monthly tele- in these measures. phone calls and pill counts from bottles returned by the Journal of Aging Research 3 Assessed for eligibility (n = 82) Enrollment Excluded (n = 22) (i) Not meeting inclusion criteria (n = 22) (ii) Declined to participate (n = 0) Randomized (n = 60) 1:2 Placebo: intervention Allocation Allocated to intervention (n = 43) Allocated to placebo (n = 17) (i) Received allocated intervention (n = 42) (i) Received allocated placebo (n = 17) (ii) Did not receive allocated intervention (ii) Did not receive allocated placebo (n = 0) (withdrew due to death in the family and associated stress (n = 1) Follow-up Lost to follow-up (completed at least one baseline measure but failed to show up to testing appointments) (n = 8) Lost to follow-up (completed at least one baseline measure but failed to show up to Discontinued intervention (withdrawn from the testing appointments) (n = 3) study by study personnel due to noncompliance on four or more bimonthly compliance phone calls, or failure to maintain inclusion criteria) (n = 1) Analysis Analyzed (n = 14) Analyzed (n = 33) Excluded from analysis (no baseline MRI data Excluded from analysis (n = 0) due to claustrophobia) (n = 1) Figure 1: CONSORT –ow diagram. participants during follow-up visits. Participation could be Scale [19] to con‡rm eligibility. ˆe CDR is a semistructured discontinued by study personnel if individuals reported interview conducted with both participants and collateral noncompliance on four or more of the telephone check-ins; informants. ˆe interviewer rates an individual’s abilities in however, no participants were withdrawn from the study due six cognitive and functional domains; scores from each of to noncompliance. Postintervention data were collected at 12 these domains are then combined to create a global rating of dementia severity ranging from 0 (no dementia) to 3 (severe months and followed the same acquisition procedure as the preintervention data collection. Of note, although the larger dementia). A global score of 0.5 is often used as a proxy RCT included a more extensive battery of outcome measures, measure for mild cognitive impairment (MCI). ˆus, to the current project focused only on retinal L and Z data, ensure the cognitive health of the sample, individuals who together with structural neuroimaging data collected at pre- received a global rating of 0 or 0.5 were eligible for the study. and postintervention visits. Results from other outcomes can be found in Lindbergh et al. and Hammond et al. [11, 18]. 2.2.2. Macular Pigment Optical Density (MPOD). Retinal concentrations of L and Z were measured as macular pig- 2.2.1. Clinical Dementia Rating Scale (CDR). Dementia ment optical density (MPOD) and assessed using custom- severity was assessed using the Clinical Dementia Rating ized heterochromatic –icker photometry (cHFP). ˆis 4 Journal of Aging Research method of data acquisition has been well validated as an in vivo normalization, and atlas registration, with processing of each measure of macular pigment density and has been fully de- time point initiated from a within-subjects template that represents mean subject anatomy across time points [23]. scribed elsewhere (e.g., [20, 21]). Briefly, participants viewed a disc that was composed of two wavelengths of light (460 Following image processing, global gray matter, global white nanometer (nm) shortwave “blue” light and 570 nanometer matter, and global white matter hypointensity volume (mm ) (nm) midwave “green” light); the two wavelengths were pre- were extracted. )e Desikan-Killiany atlas [24] was used to sented in square-wave, counter-phase orientation, which extract region-of-interest (ROI) volumes (see Figure 2), and caused the disc to appear to “flicker.” )e task was customized all volumes were corrected for intracranial volume (ICV) to individual participants based on their critical flicker fusion prior to statistical analysis according to the formula: nor- frequency (CFF) values, which were measured in the same malized volume � raw volume – b (ICV × mean ICV), where b session. Participants then turned a knob to adjust the intensity is the slope of the regression of an ROI volume on ICV. When of the 460 nm light until it appeared to match the luminance of appropriate, right and left hemisphere values were summed to the 570 nm light, causing the “flickering” to cease. )is pro- create a single value for each ROI. cedure was conducted in both the foveal and parafoveal regions of the retina. MPOD was calculated as the log of the intensity of 2.4.2. White Matter Microstructure. Diffusion weighted im- 460 nm light required to match the 570 nm light in the fovea ages (DWIs) were preprocessed using the Oxford Centre’s (where macular pigment is the densest) compared to the log of Functional MRI of the Brain (FMRIB) Diffusion Toolbox (FTD) the intensity needed in the parafovea (where macular pigment [25]. Preprocessing followed a standard pipeline, including head is absent). MPOD data collection followed the same procedure motion and eddy current correction, brain extraction, correc- at both pre- and postintervention visits. tion of distortion via fieldmap processing, and estimation of diffusion tensors for each voxel. Following preprocessing, Tract- Based Spatial Statistics (TBSS) [26] was used to optimize reg- 2.3. Neuroimaging Acquisition. All images were acquired istration and create the mean diffusion images, which were using a General Electric Signa HDx 3T MRI scanner (GE; thinned to create mean diffusion skeletons that represent the Waukesha, WI, USA). A high-resolution 3D T1-weighted centers of all tracts common to the group of participants. )e fast spoiled gradient echo (FSPGR) sequence was used to Johns Hopkins University (JHU) ICBM-DTI-81 White Matter collect structural scans (TE � 3.2 ms; TR � 7.5 ms; flip Atlas [27] was used to create binary masks for each ROI, which angle � 20 ; 154 axial slices; slice thickness � 1.2 mm; were then multiplied with the diffusion skeletons to create FOV � 256 × 256 mm in a 256 × 256 matrix). )ese images skeletonized masks for each ROI (see Figure 3). We examined provided coverage from the top of the head to the brainstem, four standard diffusivity values for each ROI: fractional an- with a total acquisition time of 6 minutes and 20 seconds. isotropy (FA), mean diffusivity (MD), radial diffusivity (RD), Diffusion weighted imaging (DWI) scans were acquired and axial diffusivity (AD). Average diffusivity values were axially using a single-shot diffusion-weighted spin echo-EPI extracted from each skeletonized ROI and used in statistical sequence. Slices covered from the top of the head to the analysis. When appropriate, right and left hemisphere values brainstem and were acquired aligned to the anterior com- were summed to create a single mean value for each ROI. missure-posterior commissure line. Scan parameters in- cluded: TE � 3.2 ms, TR � 15900 ms, 90 flip angle, 60 interleaved slices, slice gap � 0 mm, 2 mm isotropic voxels, 2.5. Statistical Analysis. Baseline MPOD was used as a co- acquisition matrix � 128 ×128, FOV � 256 × 256 mm, par- variate in all analyses to ensure that intervention effects were allel acceleration factor � 2, b-value: 1000, and 30 optimized not due to variations in baseline L and Z concentrations gradient directions with three b0 images. Total scan time for alone. Four individuals in our supplement group had pos- the DWI acquisition was 9 minutes and 38 seconds. sible MCI, as assessed by the CDR. We conducted analyses Additionally, two pairs of magnitude and phase images both with and without these participants and found no were acquired for fieldmap-based unwarping of DWIs significant effect based on the removal of these outliers. (TE1 � 5.0 ms and TE2 � 7.2 ms, TR � 700 ms, 60 slices, slice )us, in order to improve statistical power, we conducted gap � 0 mm, 2 mm isotropic voxels, acquisition matrix � final analyses with the whole sample, controlling for both age 128 ×128, and FOV � 256 × 256 mm). Acquisition for each and baseline CDR scores, as age and level of cognitive pair of images took approximately 2 minutes and 20 seconds. impairment are strong predictors of brain structure in older All neuroimaging acquisition followed the same procedure adult populations (e.g., [13, 15]). at both pre- and postintervention visits. Changes in structural brain outcomes over time as a function of intervention condition were determined using analysis of covariance (ANCOVA). Global volumes (i.e., 2.4. Neuroimaging Processing global gray matter, white matter, and white matter hypo- 2.4.1. Brain Volume. T1-weighted structural images were intensity volumes) and global diffusivity measures (i.e., global processed and segmented using FreeSurfer (v 6.0) (http:// FA, RD, MD, and AD) were entered as dependent variables surfer.nmr.mgh.harvard.edu; [22]). Due to the longitudinal into a series of two-way mixed ANCOVAs with intervention design of the study, the FreeSurfer longitudinal processing group (active supplement vs. placebo) and timepoint (pre- vs. postintervention) as the independent variables and baseline stream was utilized, which includes motion correction, skull stripping, automated transformation to Talairach space, age, CDR scores, and MPOD as the covariates. Journal of Aging Research 5 Figure 2: Volumetric regions-of-interest (ROIs). )e figure depicts the masks used for volumetric ROI analyses from left to right: prefrontal cortex, orbitofrontal cortex, anterior cingulate cortex, medial temporal cortex, and hippocampus. Masks are superimposed on a repre- sentative T -weighted image from a participant in our sample. (a) (b) Figure 3: White matter microstructure regions-of-interest (ROIs). )e figure depicts the masks used for white matter microstructure ROI analyses for the genu (red), fornix (blue), and anterior cingulum (yellow) in the sagittal view (a) and axial view (b). Masks are superimposed on a single-subject diffusion-weighted template in MNI space provided by Johns Hopkins University (JHU) in FMRIB’s Software Library (FSL). )e mean skeleton for the sample is overlaid on the diffusion-weighted image in bright green. Region-of-interest (ROI) outcomes were similarly entered significantly at preintervention on age, education level, or together in groups of two-way mixed MANCOVAs with in- baseline MPOD concentrations. Chi-square tests also con- tervention group (active supplement vs. placebo) and time- firmed that the two groups did not differ significantly at point (pre- vs. postintervention) as the independent variables preintervention in terms of sex or level of cognitive im- and baseline age, CDR score, and MPOD as the covariates. )e pairment (CDR score). Similarly, there were no significant first group of analyses included ICV-corrected gray matter differences between participants who completed the study volume for orbitofrontal cortex, prefrontal cortex, anterior (N � 47) and participants who attrited (N � 13) with respect cingulate cortex, medial temporal cortex, and the hippocampus to age, sex, education, baseline MPOD, and cognitive im- in the conglomerate dependent variable. )e second group pairment (CDR score). included ICV-corrected subcortical white matter volume of the orbitofrontal, prefrontal, anterior cingulate, and medial tem- 3.1. Brain Volume. Pre- and postintervention values for both poral cortices in the conglomerate dependent variable. )e final group included white matter diffusivity values for the genu of global and regional brain volumes can be found in Table 2. When controlling for baseline age, MPOD, and cognitive the corpus callosum, fornix, and anterior cingulum in the conglomerate dependent variable. Analyses for each diffusivity impairment (CDR score), there were no significant main effects of group for any of the measures of global brain parameter (i.e., FA, MD, RD, and AD) were performed sep- arately. If the MANCOVAs reached significance, planned volume. Additionally, there were no significant main effects follow-up two-way mixed ANCOVA analyses were conducted of time, although both groups showed nonsignificant age- related changes in brain volumes (i.e., decreased global gray to determine changes in specific ROIs as a function of in- tervention condition and timepoint, controlling for baseline and white matter volume and white matter hypointensity volume). )ere were no significant group ∗ time interactions age, CDR score, and MPOD values. for any measures of global brain volume. A two-way mixed MANCOVA showed a significant 3. Results and Discussion main effect of time for gray matter ROIs (i.e., gray matter Demographic characteristics of the sample can be found in volume of the prefrontal cortex, orbitofrontal cortex, an- Table 1. Independent-samples t-tests confirmed that the terior cingulate cortex, medial temporal lobe, and hippo- supplement group and placebo group did not differ campus) (F � 3.02, p � 0.022), but no significant main effect 6 Journal of Aging Research Table 1: Preintervention characteristics. % or M (SD) Supplement (N � 33) Placebo (N � 14) Overall sample (N � 47) Age (years) 72.4 (6.27) 70.4 (5.43) 71.8 (6.04) Sex (% female) 51.5% 71.4 % 57.4% Race (% Caucasian) 100% 100% 100% Education (years) 16.6 (3.31) 16.7 (3.02) 16.6 (3.19) Cognitive impairment (%) No impairment (CDR � 0) 87.9% 100% 91.5% Mild impairment (CDR � 0.5) 12.1% 0% 8.5% Note: CDR � clinical dementia rating scale. Table 2: Brain volume. M (SD) in mm Supplement (N � 33) Placebo (N � 14) Overall sample (N � 47) Pre Post Pre Post Pre Post Gray matter Global 589258 (41075) 583090 (43892) 598888 (26886) 595608 (27435) 592127 (37387) 586819 (39829) Regions-of-interest (ROIs) Prefrontal 102435 (10156) 101378 (10354) 105206 (7147) 104517 (6733) 103260 (9372) 102313 (9460) Orbitofrontal 23839 (2388) 23809 (2208) 23430 (1645) 23590 (1719) 23717 (2184) 23744 (2058) Anterior cingulate 7989 (1319) 7910 (1447) 7887 (1055) 7864 (1102) 7958 (1236) 7896 (1341) Medial temporal 27361 (2589) 27234 (2558) 27063 (2342) 27234 (2658) 27212 (2496) 27234 (2559) Hippocampus 8681 (989) 8407 (1050) 9144 (743) 8962 (886) 8819 (939) 8572 (1027) White matter Global 437384 (74298) 432185 (81221) 427438 (40081) 423846 (39899) 434422 (65691) 429701 (71090) Regions-of-interest (ROIs) Prefrontal 84364 (8774) 83736 (10028) 83119 (7414) 82317 (7317) 83994 (8332) 83313 (9247) Orbitofrontal 18494 (2083) 18421 (2345) 18141 (1484) 18083 (1531) 18389 (1915) 18313 (9247) Anterior cingulate 9228 (902) 9160 (951) 9336 (696) 9294 (699) 9260 (840) 9200 (878) Medial temporal 17092 (2499) 16955 (2736) 16803 (2039) 16961 (2015) 17006 (2353) 16957 (2521) Lateral ventricle 33250 (15696) 35066 (16500) 28855 (11638) 30229 (12088) 31941 (14622) 33625 (15352) White matter hypointensities 7759 (14588) 8090 (16083) 5543 (4446) 5625 (4690) 7099 (12437) 7356 (13692) Note: All volumes are corrected for intracranial volume (ICV) and have been rounded to the nearest mm . indicates a significant change from pre- to postintervention, p< 0.05, controlling for age and Clinical Dementia Rating (CDR) score. 3.2. White Matter Microstructure. Pre- and postintervention of group or group∗ time interaction was observed. When conducting planned follow-up ANCOVAs, only changes values for both global and regional white matter micro- structure can be found in Table 3. When controlling for over time in medial temporal lobe volume, regardless of group status, were individually significant (F � 6.72, baseline age, MPOD, and cognitive impairment (CDR p � 0.013). )ere were no significant changes over time, score), results showed a significant main effect of time for between group differences, or group∗ time interactions for global FA (F � 5.31, p � 0.004), but no significant main effect prefrontal cortex, orbitofrontal cortex, anterior cingulate of group or group∗ time interaction. Contrary to hypoth- cortex, or hippocampal volume. eses, both groups showed a significant increase in global FA. Similarly, a two-way mixed MANCOVA showed a )ere were no significant main effects of time for global RD, significant main effect of time for white matter ROIs MD, or AD, and no significant group differences or (i.e., subcortical white matter volume of the prefrontal group∗ time interactions were observed. Two-way mixed MANCOVAs for ROIs (i.e., genu of the cortex, orbitofrontal cortex, anterior cingulate cortex, and medial temporal lobe) (F � 4.00, p � 0.016), but no corpus callosum, fornix, and anterior cingulum) showed a significant main effect of group or group∗ time in- significant main effect of time for FA (F � 5.31, p � 0.004), teraction. However, follow-up ANCOVAs showed that but no significant main effects of group or group∗ time changes over time in the anterior cingulate cortex, re- interactions. Follow-up ANCOVAs showed that changes gardless of group status, were individually significant over time in the anterior cingulum (F � 8.60, p � 0.005) and (F � 6.19, p � 0.017). No other significant main effects of fornix (F � 8.29, p � 0.006), regardless of group status, were time, main effects of group, or group∗ time interactions individually significant. No other significant main effects of were observed for subcortical prefrontal cortex, orbi- time, main effects of group, or group∗ time interactions tofrontal cortex, or medial temporal lobe white matter were observed for genu FA and RD, MD, or AD values in the volume. ROIs. Journal of Aging Research 7 Table 3: White matter microstructure. M (SD) Supplement (N � 33) Placebo (N � 14) Overall sample (N � 47) Pre Post Pre Post Pre Post Fractional anisotropy (FA) 0.55076 0.55257 0.55384 0.56195 0.55167 0.55537 Global (0.02855) (0.03287) (0.01966) (0.02032) (0.02604) (0.02978) Regions-of-interest (ROIs) 0.61228 0.63634 0.63892 0.62361 Genu 0.61821 (.04982) 0.62021 (0.04750) (0.05182) (0.03231) (0.02889) (0.04573) 0.62050 0.62390 0.63488 0.64467 0.62478 0.63008 Anterior cingulum (0.05117) (0.05587) (0.03001) (0.03492) (0.04606) (0.05107) 0.37296 0.37186 0.37936 0.37860 0.37487 0.37387 Fornix (0.09130) (0.10075) (0.08800) (0.10308) (0.08942) (0.10037) Radial diffusivity (RD) 0.00053 0.00051 0.00050 0.00053 Global 0.00053(0.00005) 0.00052 (0.00006) (0.00061) (0.00003) (0.00003) (0.00005) Regions-of-interest (ROIs) 0.00050 0.00052 0.00047 0.00047 0.00049 Genu 0.00050 (0.00009) (0.00010) (0.00010) (0.00049) (0.00042) (0.00009) 0.00053 0.00052 0.00049 0.00048 0.00052 Anterior cingulum 0.00051 (0.00009) (0.00009) (0.00010) (0.00006) (0.00006) (0.00009) 0.00156 0.00158 0.00154 0.00156 0.00155 Fornix 0.00157 (0.00043) (0.00040) (0.00043) (0.00047) (0.00045) (0.00041) Mean diffusivity (MD) 0.00081 0.00081 0.00079 0.00078 0.00081 Global 0.00080 (0.00005) (0.00005) (0.00005) (0.00003) (0.00003) (0.00042) Regions-of-interest (ROIs) 0.00086 0.00087 0.00083 0.00083 0.00085 Genu 0.00086 (0.00008) (0.00009) (0.00009) (0.00004) (0.00003) (0.00008) 0.00090 0.00090 0.00087 0.00085 0.00089 Anterior cingulum 0.00088 (0.00007) (0.00008) (0.00008) (0.00005) (0.00004) (0.00007) 0.00191 0.00193 0.00189 0.00191 0.00190 Fornix 0.00193 (0.00038) (0.00036) (0.00038) (0.00044) (0.00039) (0.00038) Axial diffusivity (AD) 0.00138 0.00138 0.00135 0.00134 0.00137 Global 0.00137 (0.00005) (0.00004) (0.00005) (0.00004) (0.00003) (0.00004) Regions-of-interest (ROIs) 0.00156 0.00158 0.00155 0.00155 0.00156 Genu 0.00157 (0.00007) (0.00010) (0.00008) (0.00004) (0.00003) (0.00008) 0.00164 0.00166 0.00161 0.00160 0.00163 Anterior cingulum 0.00164 (0.00006) (0.00007) (0.00006) (0.00005) (0.00003) (0.00006) 0.00261 0.00264 0.00260 0.00261 0.00260 Fornix 0.00263 (0.00030) (0.00029) (0.00301) (0.00038) (0.00030) (0.00032) Note: indicates a significant change from pre- to postintervention, p � 0.05, controlling for age and Clinical Dementia Rating (CDR) score. 3.3. Intervention Response. To confirm intervention effec- individuals in the supplement group appearing to fail to tiveness, paired-samples t-tests were conducted in both the respond to intervention (i.e., showing stable or decreased MPOD concentrations). To explore this heterogeneity, in- supplement group and placebo group to assess statistically significant changes in MPOD concentrations over the course dividuals were classified into two categories: (1) those who of the intervention. Analyses confirmed that the supplement showed an increase in MPOD of 0.10 + log units from pre- to group showed a significant increase in MPOD (t � 2.27, postintervention were classified as “responders” and (2) all p � 0.030) over the course of the trial, while the placebo others who showed a stable, decreased, or nonsignificant group did not show any significant changes in MPOD increase (<0.10 log units) in MPOD from pre- to post- (t � 0.788, p � 0.445). However, there was heterogeneity in intervention were classified as “nonresponders” (see Table 4). both groups, with some individuals in the placebo group )en, exploratory analyses were undertaken to determine if showing increased MPOD concentrations and some there were any brain changes corresponding to increased L 8 Journal of Aging Research Table 4: Treatment response–change in MPOD concentrations. and Z concentrations within the supplement group (“in- tervention responders”). ˆe same analyses were repeated in M (SD) the placebo group to determine if increases in MPOD, in the Preintervention Postintervention absence of supplementation, were associated with brain Supplement group structural changes. As with the primary analyses above, all Responder (N  15) 0.4547 (0.1992) 0.6947 (0.2329) exploratory analyses were controlled for baseline age, MPOD, Nonresponder (N  18) 0.5717 (0.1658) 0.5111 (0.1705) and CDR scores. Placebo group 0.4414 (0.0373) 0.4843 (0.0536) Independent-samples t-tests con‡rmed that there was no ∗ Responder (N  7) 0.3871 (0.1602) 0.5943 (0.1605) signi‡cant dižerence between treatment “responders” and Nonresponder (N  7) 0.4957 (0.9888) 0.3743 (0.1820) “nonresponders” in either group with respect to baseline age, Total sample 0.4955 (0.0259) 0.5617 (0.0317) education, and MPOD concentrations. Additionally, chi- Responder (N  22) 0.4332 (0.1866) 0.6627 (0.2141) Nonresponder (N  25) 0.5504 (0.1521) 0.4728 (0.1811) square tests con‡rmed that responders and nonresponders were not dižerent with regard to sex and cognitive im- Note: indicates a signi‡cant change from pre- to postintervention of p < 0.05. pairment (CDR scores). In the supplement group, those who were classi‡ed as intervention responders (N  15) showed signi‡cantly less decline in total gray matter volume (∆R  0.144, F  5.37, 6.2 p  0.028) (see Figure 4) and prefrontal cortex gray matter 6.1 volume (∆R  0.125, F  4.26, p  0.048) (see Figure 5) than those classi‡ed as nonresponders (N  18). However, in the 5.9 placebo group, there were no signi‡cant dižerences in any brain volume measures between individuals who showed 5.8 increased MPOD (N  7) versus those who showed stable or 5.7 decreased MPOD (N  7). ˆere were also no signi‡cant 5.6 dižerences in global or regional white matter microstructure 5.5 measures between those classi‡ed as responders versus Pre Post nonresponders in either the supplement or placebo groups. Responder Nonresponder 4. Conclusions Figure 4: Changes in total gray matter volume. ˆe ‡gure shows 3 5 ˆe current study tested whether one year of supplemen- changes in total gray matter volume (mm ×10 ) between sup- plement group “responders” and “nonresponders.” tation with lutein (L) and zeaxanthin (Z), two xanthophyll carotenoids with known antioxidative properties and cog- nitive bene‡ts, could prevent or slow age-related structural brain changes in a sample of community-dwelling older 1.08 adults. Using a randomized, double-blind, placebo-con- 1.06 trolled trial design, we hypothesized that older adults re- ceiving supplementation with L and Z would increase, 1.04 maintain, or show attenuated loss of brain volume and white 1.02 matter microstructure in areas that are vulnerable to age- related decline (i.e., frontal and temporal gray and white 1 matter regions) relative to older adults receiving a placebo. 0.98 Results showed expected age-related declines for frontal 0.96 and medial-temporal gray and white matter, particularly Pre Post medial temporal lobe gray matter and subcortical white Responder matter of the anterior cingulate cortex, across both groups Nonresponder over the course of the trial. However, L and Z supplemen- tation did not appear to in–uence this loss. No signi‡cant Figure 5: Changes in prefrontal cortex volume. ˆe ‡gure shows 3 5 group dižerences or changes over time were observed in changes in total gray matter volume (mm ×10 ) between sup- global brain volume outcomes (i.e., global gray matter, global plement group “responders” and “nonresponders.” white matter, and white matter hypointensity volume). Ad- ditionally, no interactions between group and time were found for any of the brain volume measures. Average percent and placebo groups. Results did show an average increase in change for both global volumes and frontal-temporal volumes global FA, across both groups. Signi‡cant changes were also ranged from less than 1% to 3%, consistent with previous seen in anterior white matter tracts, including increased FA estimates of the average annual rate of gray and white matter in the anterior cingulum and an expected, age-related de- volume decline in older adults [14, 28]. cline in fornix FA. While it is encouraging to see im- We also did not ‡nd signi‡cant dižerences in white provements in global and anterior cingulum white matter matter microstructure outcomes between the supplement microstructure, it is important to note that these changes Journal of Aging Research 9 were seen in both the supplement and placebo groups and, population at baseline, prior to supplementation. In this thus, cannot be attributed to the intervention per se. Average respect, it is encouraging that a small dietary change percent change for both global and frontal-temporal white provided benefit to even a subsample of individuals in our matter microstructure ranged from less than 1% to around study who were already well nourished, educated, and 2%, again consistent with estimates of annual changes in affluent. Finally, as with any nutritional intervention, there diffusion metrics in healthy older adult populations [29, 30]. was no true control condition; our entire sample has Although results confirmed that the intervention ma- presumably been exposed to both L and Z throughout their nipulation was effective (i.e., the supplement group showed a lifetimes, and even those participants in the placebo significant increase in retinal L and Z concentrations, while condition continued to consume a normal diet for the the placebo group did not), there was individual variability duration of the trial, which likely contained some amount in both groups. )is individual variability is not uncommon of L and Z which are naturally occurring in many foods. It for nutritional intervention studies. )ere are many factors is possible that the biggest effects of L and Z on the brain which may affect how one processes and absorbs nutritional could be driven by deficiency. While the current study found that one year of supplements. For example, some studies have suggested that cholesterol levels and cosupplementation with other nu- supplementation of L and Z had limited effects on the trients can impact the absorption and bioavailability of L and brain structural integrity of community-dwelling older Z [31, 32]. Additionally, the very nature of being in a nu- adults, there is a growing body of literature to suggest that tritional study may cause some participants to choose these nutrients are important for other aspects of cog- healthier and more nutritious foods, whether consciously or nitive and brain health. Other analyses from the current unconsciously. )us, exploratory analyses were conducted RCT have shown that supplementation with L and Z to determine whether individuals who responded better to benefited cognition, particularly complex attention, the intervention, or who otherwise increased their retinal L cognitive flexibility [18], and verbal learning [11], as well and Z concentration, showed better neural outcomes than as neural functioning in dorsolateral prefrontal cortex those who showed no change or decreased retinal L and Z. and anterior cingulate, areas that are vulnerable to age- related decline [11]. )e importance of L and Z for While there were no differences in terms of white matter microstructure outcomes, results showed that in the sup- cognitive and neural outcomes has been replicated cross- plement group, intervention “responders” (i.e., increased sectionally in other samples and by other researchers retinal L and Z) had significantly less decline in global and [9, 37–39] and is beginning to be confirmed longitudi- prefrontal gray matter volume than intervention “non- nally as well [40, 41]. responders” (i.e., stable or decreased retinal L and Z), even )e strength of this study lies in the novel approach after controlling for factors such as baseline age, MPOD, and used to investigate in vivo structural brain outcomes of a cognitive impairment. )ere were no concurrent differences year-long, double-blind, placebo-controlled trial of lutein observed in the placebo group. )ese results suggest that L (L) and zeaxanthin (Z) supplementation in older adults. and Z may slow age-related gray matter decline for a subset Exploratory analyses suggested that there was a small group of individuals who appeared to reap greater benefit from the of individuals who reaped greater benefit from L and Z supplementation regimen. supplementation and who showed less decline in global and As with any study, the present trial had limitations. For prefrontal gray matter volume than individuals who did not example, a similar prospective study on the effects of Vi- appear to benefit from the intervention. To that end, future tamin B on brain volume found significant results but studies could explore factors which may impact the efficacy followed the older subjects for five years as opposed to only of L and Z supplementation, such as vascular health, one [33]. )us, longer time periods may be necessary to cosupplementation with other nutrients, and other factors allow enough sensitivity to discriminate how dietary known to interact with nutrient bioavailability and ab- components influence morphological loss. Additionally, we sorption. Replication of this study in a larger and more did not have access to biomarkers, such as blood vitamin diverse sample may also determine whether there are some levels or lipid metabolites, in order to analyse the effect that individuals who could benefit even more from L and Z these individual differences might have on intervention supplementation, such as individuals who are more cog- nitively impaired or have had less access to nutritious diets efficacy. Our sample size was small (e.g., the B study [33] tested 107 older adults) and suffered from considerable across their lifetime. Factors such as supplement dosage and particularly a longer intervention duration should also attrition (approximately 22%), which may have limited our power to detect certain group differences and intervention be considered when determining intervention efficacy. effects. Additionally, our sample was homogenous in terms Finally, future studies will need to determine the long-term of background, consisting of 100% Caucasian and pre- effects of L and Z supplementation on outcomes such as the dominantly cognitively healthy and highly educated in- development of dementia and accumulation of brain pa- dividuals. Research has shown that more educated people thologies. While L and Z supplementation over one year also tend to have healthier diets [34], and our data support appears to have limited effects on structural brain outcomes this trend. Baseline mean MPOD values for our sample in older adults, it is likely not harmful [42] and appears to were slightly higher than other published data for older provide benefits for other aspects of cognitive and brain health that could improve or extend quality of life for older adults [35, 36], suggesting that our sample may have consumed a more nutritious diet than the general adults. 10 Journal of Aging Research [9] E. J. Johnson, R. Vishwanathan, M. A. Johnson et al., “Re- Data Availability lationship between serum and brain carotenoids, α-tocoph- erol, and retinol concentrations and cognitive performance in )e data used to support the findings of this study are the oldest old from the Georgia centenarian study,” Journal of available from the corresponding author upon request. Aging Research, vol. 2013, Article ID 951786, 13 pages, 2013. [10] T. den Heijer, L. J. Launer, J. C. de Groot et al., “Serum Conflicts of Interest carotenoids and cerebral white matter lesions: the Rotterdam scan study,” Journal of the American Geriatrics Society, vol. 49, During a portion of data collection time, LMRH was no. 5, pp. 642–646, 2001. employed by Abbott Nutrition while holding a joint ap- [11] C. A. Lindbergh, L. M. Renzi-Hammond, B. R. Hammond pointment at the University of Georgia. No other conflicts of et al., “Lutein and zeaxanthin influence brain function in older adults: a randomized controlled trial,” Journal of the In- interest exist for the study authors, including CMM, CAL, ternational Neuropsychological Society, vol. 23, no. 1, BRH, and LSM. All statistical analyses were completed in- pp. 77–90, 2018. dependent of the supporting agencies. [12] C. M. Mewborn, D. P. Terry, L. M. Renzi-Hammond, B. R. Hammond, and L. S. Miller, “Relation of retinal and Acknowledgments serum lutein and zeaxanthin to white matter integrity in older adults: a diffusion tensor imaging study,” Archives of Clinical )is work was supported by Abbott Nutrition, Columbus, Neuropsychology, vol. 33, no. 7, pp. 861–874, 2017. OH, USA (research grant to LMRH, BRH, and LSM); )e [13] O. Abe, H. Yamasue, S. Aoki et al., “Aging in the CNS: University of Georgia’s Bio-Imaging Research Center (ad- comparison of gray/white matter volume and diffusion tensor data,” Neurobiology of Aging, vol. 29, no. 1, pp. 102–116, 2008. ministrative support to LSM); and DSM Nutritional Prod- [14] D. H. Salat, J. A. Kaye, and J. S. Janowsky, “Prefrontal gray and ucts (Switzerland) who provided the supplements and white matter volumes in healthy aging and Alzheimer dis- placebos for the randomized-controlled trial from which the ease,” Archives of Neurology, vol. 56, no. 3, pp. 338–344, 1999. current data were derived. A version of this work was th [15] I. J. Bennett and D. J. Madden, “Disconnected aging: cerebral presented at the 47 Annual Meeting of the International white matter integrity and age-related differences in cogni- Neuropsychological Society. Published Abstract: Journal tion,” Neuroscience, vol. 276, no. 6, pp. 187–205, 2014. of the International Neuropsychological Society, 25(s1). doi: [16] D. Head, R. L. Buckner, J. S. Shimony et al., “Differential 10.1017/S1355617719000663. vulnerability of anterior white matter in nondemented aging with minimal acceleration in dementia of the Alzheimer type: evidence from diffusion tensor imaging,” Cerebral Cortex, References vol. 14, no. 4, pp. 410–423, 2004. [17] K. F. Schulz, D. G. Altman, and D. Moher, “CONSORT 2010 [1] M. Couture, N. Larivi`ere, and R. Lefrançois, “Psychological statement: updated guidelines for reporting parallel group distress in older adults with low functional independence: a randomized trials,” Annals of Internal Medicine, vol. 152, multidimensional perspective,” Archives of Gerontology and no. 11, pp. 726–732, 2010. Geriatrics, vol. 41, no. 1, pp. 101–111, 2005. [18] B. R. Hammond, L. S. Miller, M. O. Bello, C. A. Lindbergh, [2] A. N. Puente, D. P. Terry, C. C. Faraco, C. L. Brown, and C. Mewborn, and L. M. Renzi-Hammond, “Effects of lutein/ L. S. Miller, “Functional impairment in mild cognitive im- zeaxanthin supplementation on the cognitive function of pairment evidenced using performance-based measurement,” community dwelling older adults: a randomized, double- Journal of Geriatric Psychiatry and Neurology, vol. 27, no. 4, masked, placebo-controlled trial,” Frontiers in Aging Neuro- pp. 253–258, 2014. science, vol. 9, 2017. [3] D. Harman, “Aging: a theory based on free radical and ra- [19] J. C. Morris, “)e Clinical dementia rating (CDR): current diation chemistry,” Journal of Gerontology, vol. 11, pp. 289– version and scoring rules,” Neurology, vol. 43, no. 11, p. 2412, 300, 1956. [4] M. Abdollahi, M. Y. Moridani, O. I. Aruoma, and [20] B. R. Wooten and B. R. Hammond, “Spectral absorbance and S. Mostafalou, “Oxidative stress in aging,” Oxidative Medicine spatial distribution of macular pigment using heterochro- and Cellular Longevity, vol. 2014, Article ID 876834, 2 pages, matic flicker photometry,” Optometry and Vision Science, vol. 82, no. 5, pp. 378–386, 2005. [5] Y. Gu, A. M. Brickman, Y. Stern et al., “Mediterranean diet [21] B. R. Wooten, B. R. Hammond, R. I. Land, and and brain structure in a multiethnic elderly cohort,” Neu- D. M. Snodderly, “A practical method for measuring macular rology, vol. 85, no. 20, pp. 1744–1751, 2015. pigment optical density,” Investigative Ophthalmology and [6] C. A. Raji, K. I. Erickson, O. L. Lopez et al., “Regular fish consumption and age-related brain gray matter loss,” Visual Science, vol. 40, no. 11, pp. 2481–2489, 1999. [22] B. Fischl, D. H. Salat, E. Busa et al., “Whole brain segmen- American Journal of Preventive Medicine, vol. 47, no. 4, pp. 444–451, 2014. tation,” Neuron, vol. 33, no. 3, pp. 341–355, 2002. [23] M. Reuter, N. J. Schmansky, H. D. Rosas, and B. Fischl, [7] N. Scarmeas, J. A. Luchsinger, Y. Stern et al., “Mediterranean diet and magnetic resonance imaging-assessed cerebrovas- “Within-subject template estimation for unbiased longitu- dinal image analysis,” NeuroImage, vol. 61, no. 4, pp. 1402– cular disease,” Annals of Neurology, vol. 69, no. 2, pp. 257–268, 2011. 1418, 2012. [24] R. S. Desikan, F. Segonne, ´ B. Fischl et al., “An automated [8] N. E. Craft, T. B. Haitema, K. M. Garnett, K. A. Fitch, and C. K. Dorey, “Carotenoid, tocopherol, and retinol concen- labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest,” NeuroImage, trations in elderly human brain,” 6e Journal of Nutrition, Health and Aging, vol. 8, no. 3, pp. 156–162, 2004. vol. 31, no. 3, pp. 968–980, 2006. Journal of Aging Research 11 [25] T. E. J. Behrens, M. W. Woolrich, M. Jenkinson et al., elderly population: results of the EVA study,” 6e Journals of Gerontology Series A: Biological Sciences and Medical Sciences, “Characterization and propagation of uncertainty in diffu- sion-weighted MR imaging,” Magnetic Resonance in Medi- vol. 62, no. 3, pp. 308–316, 2007. [41] E. J. Johnson, H.-Y. Chung, S. M. Caldarella, and cine, vol. 50, no. 5, pp. 1077–1088, 2003. [26] S. M. Smith, M. Jenkinson, H. Johansen-Berg et al., “Tract- D. M. Snodderly, “)e influence of supplemental lutein and docosahexaenoic acid on serum, lipoproteins, and macular based spatial statistics: voxelwise analysis of multi-subject pigmentation,” 6e American Journal of Clinical Nutrition, diffusion data,” NeuroImage, vol. 31, no. 4, pp. 1487–1505, vol. 87, no. 5, pp. 1521–1529, 2008. [42] )e Age-Related Eye Disease Study 2 (AREDS2) Research [27] S. Mori, K. Oishi, H. Jiang et al., “Stereotaxic white matter Group, “Lutein + zeaxanthin and omega-3 fatty acids for age- atlas based on diffusion tensor imaging in an ICBM template,” related macular degeneration: the Age-Related Eye Disease NeuroImage, vol. 40, no. 2, pp. 570–582, 2008. Study 2 (AREDS2) randomized clinical trial,” JAMA, vol. 309, [28] T. L. Jernigan, S. L. Archibald, C. Fennema-Notestine et al., no. 19, pp. 2005–2015, 2013. “Effects of age on tissues and regions of the cerebrum and cerebellum,” Neurobiology of Aging, vol. 22, no. 4, pp. 581–594, [29] T. R. Barrick, R. A. Charlton, C. A. Clark, and H. S. Markus, “White matter structural decline in normal ageing: a pro- spective longitudinal study using tract-based spatial statis- tics,” NeuroImage, vol. 51, no. 2, pp. 565–577, 2010. [30] S. J. Teipel, T. Meindl, M. Wagner et al., “Longitudinal changes in fiber tract integrity in healthy aging and mild cognitive impairment: a DTI follow-up study,” Journal of Alzheimer’s Disease, vol. 22, no. 2, pp. 507–522, 2010. [31] Y. Sato, R. Suzuki, M. Kobayashi et al., “Involvement of cholesterol membrane transporter Niemann-Pick C1-Like 1 in the intestinal absorption of lutein,” Journal of Pharmacy and Pharmaceutical Sciences, vol. 15, no. 2, pp. 256–264, 2012. [32] S. A. Tanumihardjo, J. Li, and M. P. Dosti, “Lutein absorption is facilitated with cosupplementation of ascorbic acid in young adults,” Journal of the American Dietetic Association, vol. 105, no. 1, pp. 114–118, 2005. [33] A. Vogiatzoglou, H. Refsum, C. Johnston et al., “Vitamin B12 status and rate of brain volume loss in community-dwelling elderly,” Neurology, vol. 71, no. 11, pp. 826–832, 2008. [34] T. N. Akbaraly, A. Singh-Manoux, M. G. Marmot, and E. J. Brunner, “Education attenuates the association between dietary patterns and cognition,” Dementia and Geriatric Cognitive Disorders, vol. 27, no. 2, pp. 147–154, 2009. [35] L. M. Renzi, M. J. Dengler, A. Puente, L. S. Miller, and B. R. Hammond, “Relationships between macular pigment optical density and cognitive function in unimpaired and mildly cognitively impaired older adults,” Neurobiology of Aging, vol. 35, no. 7, pp. 1695–1699, 2014. [36] L. M. Renzi, B. J. Hammond, M. Dengler, and R. Roberts, “)e relation between serum lipids and lutein and zeaxanthin in the serum and retina: results from cross-sectional, case-control and case study designs,” Lipids in Health and Disease, vol. 11, no. 33, 2012. [37] S. Ajana, D. Weber, C. Helmer et al., “Plasma concentrations of lutein and zeaxanthin, macular pigment optical density, and their associations with cognitive performances among older adults,” Investigative Opthalmology and Visual Science, vol. 59, no. 5, pp. 1828–1835, 2018. [38] J. Feeney, C. Finucane, G. M. Savva et al., “Low macular pigment optical density is associated with lower cognitive performance in a large, population-based sample of older adults,” Neurobiology of Aging, vol. 34, no. 11, pp. 2449–2456, [39] R. Vishwanathan, A. Iannaccone, T. M. Scott et al., “Macular pigment optical density is related to cognitive function in older people,” Age and Ageing, vol. 43, no. 2, pp. 271–275, [40] N. T. Akbaraly, H. Faure, V. Gourlet, A. Favier, and C. Berr, “Plasma carotenoid levels and cognitive performance in an MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 International Journal of Journal of Immunology Research Endocrinology Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com BioMed PPAR Research Research International Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2013 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Neurology Research and Treatment Cellular Longevity Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Journal of Aging ResearchHindawi Publishing Corporation

Published: Dec 1, 2019

References