Neurobiology of Aging 47 (2016) 63e70 Contents lists available at ScienceDirect Neurobiology of Aging journal homepage: www.elsevier.com/locate/neuaging Obesity associated with increased brain age from midlife Lisa Ronan a, *, Aaron F. Alexander-Bloch b, Konrad Wagstyl a, Sadaf Farooqi c, Carol Brayne d, Lorraine K. Tyler e, Cam-CANe, Paul C. Fletcher a a Brain Mapping Unit, Department of Psychiatry, University of Cambridge, UK Yale School of Medicine, Yale University, USA c Department of Clinical Biochemistry, Institute of Metabolic Sciences, Cambridge, UK d Institute of Public Health, University of Cambridge, Cambridge, UK e MRC Cognition and Brain Sciences Unit, Cambridge Center for Ageing and Neuroscience (Cam-CAN), Cambridge, UK b a r t i c l e i n f o a b s t r a c t Article history: Received 15 October 2015 Received in revised form 14 July 2016 Accepted 15 July 2016 Available online 27 July 2016 Common mechanisms in aging and obesity are hypothesized to increase susceptibility to neurodegeneration, however, direct evidence in support of this hypothesis is lacking. We therefore performed a cross-sectional analysis of magnetic resonance image-based brain structure on a population-based cohort of healthy adults. Study participants were originally part of the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) and included 527 individuals aged 20e87 years. Cortical reconstruction techniques were used to generate measures of whole-brain cerebral white-matter volume, cortical thickness, and surface area. Results indicated that cerebral white-matter volume in overweight and obese individuals was associated with a greater degree of atrophy, with maximal effects in middle-age corresponding to an estimated increase of brain age of 10 years. There were no similar body mass index-related changes in cortical parameters. This study suggests that at a population level, obesity may increase the risk of neurodegeneration. ! 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Keywords: Obesity White-matter volume Structural MRI Population-based 1. Introduction The link between obesity and adverse health outcomes such as diabetes, cancer, and cardiovascular disease is well established and poses a major challenge to current and future health care provision. Moreover, it is increasingly recognized that obesity may act to accelerate or advance the onset of age-related changes such as neurodegeneration, either directly or through associated comorbidities (Doherty, 2011). These associations, taken together with the increased rate of obesity in elderly populations (Flegal et al., 2012) render it critical to understand the full impact of obesity on brain health, in particular as evidence suggests that adverse outcomes may be mitigated through intervention (Gunstad et al., 2011). A number of strands of evidence have related biological processes associated with obesity to changes found in normal aging. For example, as with normal aging, obesity increases oxidative stress (Furukawa et al., 2004) and promotes inflammation through * Corresponding author at: Brain Mapping Unit, Department of Psychiatry, Downing Site, Downing Street, Cambridge CB2 3EB, UK. Tel.: 01223 764421; fax: 01223 764760. E-mail address: lr344@cam.ac.uk (L. Ronan). the production of proinflammatory cytokines produced in adipose tissue (Arnoldussen et al., 2014; Chung et al., 2009). In turn, cytokines and proinflammatory markers such as interleukin 6 and tumor necrosis factor-a have been linked to cognitive decline (Chung et al., 2009; Griffin, 2006; Wilson et al., 2002) and have been shown to be upregulated in regions undergoing neurodegeneration (Wilson et al., 2002). Inflammatory biomarkers have been associated with increased brain atrophy, a common marker of aging (Jefferson et al., 2007), as have other endophenotypes such as shortened telomere length (Wikgren et al., 2014). Conversely, a considerable body of evidence exists suggesting that caloric restriction may be neuroprotective, leading to a delay or slowing of aging (Colman et al., 2014, 2009; Masoro, 2005; Sohal and Weindruch, 1996), a reduction in age-related apoptosis (Someya et al., 2007), and age-related production of proinflammatory cytokines (Kalani et al., 2006; Spaulding et al., 1997). In short, the growing body of literature that relates common markers of aging to those observed in obesity supports the hypothesis that obesity may accelerate or advance the onset of brain aging. However, direct studies in support of this link are lacking. For example, although many studies have reported a link between increased body mass index (BMI) and declining cognitive function 0197-4580/! 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). http://dx.doi.org/10.1016/j.neurobiolaging.2016.07.010 64 L. Ronan et al. / Neurobiology of Aging 47 (2016) 63e70 (Cournot et al., 2006; Debette et al., 2011), as well as increased risk of dementia and Alzheimer’s Disease (Gustafson et al., 2004; Whitmer et al., 2005; Xu et al., 2011), other studies contradict these findings (Qizilbash et al., 2015), and indeed, it has even been suggested that lower, rather than higher, body mass may be predictive of the onset of AD in the years immediately preceding the onset of clinical symptoms (Fielding et al., 2013; Knopman et al., 2007). The literature on brain structural changes too is complex. Although many studies report a negative correlation between BMI and gray matter volume (GMV) (increased BMI linked to lower GMV) (Brooks et al., 2013; Debette et al., 2014; Gunstad et al., 2008; Hassenstab et al., 2012; Veit et al., 2014), other reports are contradictory (Haltia et al., 2007; Pannacciulli et al., 2007; Sharkey et al., 2015). More significantly, despite a considerable number of often highly powered studies across the adult lifespan (Taki et al., 2008), there is a conspicuous lack of either global findings related to obesity or evidence of an aging interaction (for a review, see Willette and Kapogiannis, 2015). Thus, although current neuroimaging evidence certainly suggests altered brain structure is associated with obesity, it fails to support the hypothesis that obesity influences age-related atrophy of the brain. There are a number for reasons for why this might be. Different tissue types in the brain age at different rates (Walhovd et al., 2005), perhaps limiting the sensitivity of cross-sectional studies over limited age-periods. Moreover, there is a complex and somewhat compensatory interaction between the change in cortical thickness and surface area (Storsve et al., 2014), that may confound analysis by morphometric methods such as voxel-based morphometry commonly employed in structural studies of obesity. In addition, voxel-based morphometry methods are designed to obviate global changes in favor of regional analyses. If obesity, like aging affects the brain globally, it may be the case that a significant global interaction may be obfuscated. Analysis of white matter too may be confounded. Although some studies suggest obesity and inflammation are both associated with smaller fractional anisotropy in diffusion tensor imaging (Stanek et al., 2011; Verstynen et al., 2013), it is also the case that additional factors related to obesity and aging such as blood pressure are positively associated with fractional anisotropy (Verstynen et al., 2013), raising the possibility that competing effects may hamper identification of an age-by-BMI interaction. The alternative to these propositions is that obesity may increase the rate of aging of brain tissue but that these effects are subtle and within the scope of normal aging parameters. In this cross-sectional population-based study, we assessed the impact of obesity on brain structure across the adult lifespan using global parameters of volume, cortical thickness, and surface area. The goal of our study was to establish the overall effect of obesity on gray (i.e., cortical thickness and surface area) and white matter; to determine whether obesity affected tissue types differentially; and crucially to investigate whether obesity was associated with an increase in brain age, evaluated with reference to lean controls. We were particularly interested in whether changes associated with obesity (i.e., deviations from lean age-matched controls) might occur during a particular vulnerable period. 2. Materials and methods 2.1. Subjects A total of 527 subjects with an age range of 20e87 years were included in this study. Participants were cognitively healthy adults recruited from the local community over a period of 5 years as part of an ongoing project to investigate the effects of aging on memory and cognition at the Cambridge Centre for Aging and Neuroscience (Shafto et al., 2014). Ethical approval for the Cam-CAN study was obtained from the Cambridgeshire 2 (now East of EnglandeCambridge Central) Research Ethics Committee. Of these, 54 subjects were excluded on the basis of being underweight (BMI < 18.5kgm!1), under the age of 20, or for reasons of poor MR image quality (see below). Subject demographics are detailed in Table 1. The mean age was 54 years (range 20e87), and mean BMI was 26 kg/m2 (18.5e45.5). The final cohort included 246 (51%) lean controls (BMI between 18.5e25 kgm!2), 150 overweight subjects (31%; BMI 25e30 kgm!2), and 77 obese subjects (BMI >30 kgm!2). There was a significant positive correlation between age and BMI (r ¼ 0.24, p < 0.001). Various health and lifestyle factors were recorded including self-reported history of diagnosis of diabetes, stroke, cancer, myocardial infarction, high blood pressure, and high cholesterol. A self-report questionnaire was used to calculated total estimated physical activity per week (measures as kJ/d/Kg). Education level was binarized to those with or without degree-level qualifications. Gross household income was also included, defined as those above and below a threshold income of £30,000. There were no incidences of Parkinson’s disease or multiple sclerosis. Cognitive Table 1 Demographic information Variables Lean Overweight Obese BMI (kg/m2) (mean) No. of subjects (%) Sociodemographic variables Age (years) Female/male University degree or higher Household income (above median) Health behaviors Current smoking (%) Physical activity (kJ/d/Kg) Health measures Systolic blood pressure (BP) (mm Hg) Diastolic BP (mm Hg) Disease diagnosis Myocardial infarction Cancer Diabetes Stroke High cholesterol High BP 18.5e24.99 22.7 # 1.7 246 (51) 25e29.99 27.1 # 1.6 150 (31) $30 33.5 # 3.8 77 (18) 48 # 16 122/124 180 149 57 # 17 66/84 89 84 61 # 16 49/28 33 38
Purchase answer to see full attachment