Review Article

Structural Brain Changes Associated with Overweight and Obesity

Table 2

Characteristics and summary results of neuroimaging studies on obesity.

Author, countryStudy populationAssessmentMain findings
N; age: mean (SD) years; BMI: mean (SD) (kg/m2)

Herrmann et al. [72],GermanyMeta-analysis of 10 studies [73-82]MRI studiesObese status: ↓ GMV: B IFG (including insula), L MFG, L preFC, L MTG, L PCG, L cerebellum, BA 34 (including L amygdala and L lenticular nucleus)
Voxel-wise meta-analysis: GMV
BMI

Taki et al. [83], JapanN = 1,428 Men = 690, A: 44.5 (16.1) Women = 738, A: 46.4 (14.1)MRI: VBM: GMV, GMrMen: ↑ BMI, ↓ GMV: B uncus, B cerebellum (anterior lobe), L FG, R SPL, R preCG, R IFG, B precuneus, B SFG, R midbrain
BMI↑ BMI, ↑ GMV: B IFG, B cerebellum (posterior lobe), R SFG, R STG, R ITG, L MTG, B Th (pulvinar), R CG, B Cn (heads), L preCG
No significant correlations in women

García-García et al. [86], CanadaMeta-analysis of 21 studies (including 73, 76-80, 82-83, 85, 88, 115) N = 5,882MRI studies: VBM analyses: GMVObesity-related variables are associated with ↓ GMV mainly in mPFC, B cerebellum, L TPole
Obesity-related variables (BMI, WC, WHR, BFD)Less robust results: preCG, IPC

Yokum et al. [87], USAN = 83 (women), A: 18.4 (2.8) BMI range: 17.3–38.9MRI: VBM:Total GMV: obese < OW, lean
WMV and GMVOverall WMV: obese < OW
BMI↑ BMI, ↑ GMV: MOG
↑ BMI, ↑ WMV: MTG, FG, PHipG, RO, DS
One-year follow-up: ↑ BMI, ↓ GMV: SFG, MF

Papageorgiou et al. [89], GreeceN = 268, A: 46.3 (16.1) Obese = 52 OW = 96 Normal weight = 120MRI: DTI: brain fractional anisotropy↑ BMI was related with ↓ ATR, PTR, IFOF, ILF, SLF, CC, UF, IC, CST, cingulum
WM
BMI

Repple et al. [90], GermanyN = 1,433 Münster Neuroimaging Cohort = 369 Human Connectome Project = 1064MRI: DTI: brain fractional anisotropy, WM↑ BMI globally, ↓ WM integrity (replication in an independent sample)
BMI

Kullman et al. [91], Germany/UKN = 48MRI: DTI, voxel-based quantification analysis↑ BMI, ↓ predominantly in the SLF, ATR, IC, and CC
Obese = 12, A: 26.9 (4.5), BMI: 33.2 (3.2)WM
OW = 12, A: 26.1 (2), BMI: 28.1 (1.4)BMI
Lean = 24, A: 26.7 (3.7), BMI: 22.4 (1.6)

Kim et al. [92], South KoreaN = 54 Obese = 27, A: 23.15 (3.2), BMI: 30.12 (3.8); lean = 27, A: 26.1 (5.78), BMI: 21.76 (1.7)MRI: vertex-based and shape analysis↑ BMI, ↓ B Cn
Subcortical structures volume↑ BMI, ↑ B Th, B Pu, B GP, B hip
BMI

Raji et al. [93], USAN = 94MRI: tensor-based morphometryGlobally: ↑ BMI, ↓ GMV and WMV
Obese = 14, A: 76.9 (2.8), BMI: 32.9 (2.5)BMI, FPI, T2DM↑ BMI, ↓ V: OFC, hip, Pu, GP, Th
OW = 51, A: 77.2 (2.6), BMI: 27.5 (1.4)↑ FPI, ↓ V: splenium CC, OFC, hip
Lean = 29, A:77.5 (4), BMI: 22.5 (1.9)↑ T2DM, ↓ V: genu and splenium CC, ACC, mTL, OL, Cn, Pu, GP

Dekkers et al. [94], NetherlandsN = 12,087 (from UK Biobank study)MRI: multiparametric brain mapping; DTIIn women: ↑ TBF, ↑ GMV, WMV; ↓ GP
TBFIn men: ↑ TBF, ↓ GMV, Th, Cn, Pu, GP, hip, Nac

Le et al. [96], USA/TaiwanN = 105 (58 women), A: 31.4 (12.6)MRI: VBM and rsFC analyses Hypothalamic GMV↑ BMI, ↑ hypothalamic GMV, and ↑ connectivity with the insula, Th, GP, and cerebellum but ↓ with hypothalamic connectivity with the SPL
BMI: 25.8 (4.7)BMI

Ou et al. [97], USAN = 24MRI: VBM and TBSS analyses: GMV, WMVObese: ↓ GMV: R MTG, B Th, L SPL, L preCG, postCG, L cerebellum
Obese = 12, A: 9.1 (0.9), BMI: 24.4 (3.4)BMIObese: ↑ WMV: B posterior IFOF, B posterior SFOF, B SCR
Lean = 12, A: 9.8 (0.7), BMI: 15.8 (1)

Perlaki et al. [98], Hungary/Netherlands/ Germany/SwedenN = 51, (32 female), A: 13.8 (1.9)MRI: VBM: GMV, GMMVBM analysis controlled for age, sex, and ICV; no significant associations between GMM and BMI z-score
ROIsROI: ↑ BMI z-score, ↑ B amygdala, ↑ Nac
BMI z-score

Sharkey et al. [100], CanadaN = 378 (from NIH Pediatric MRI Data Repository); age range: 4–18MRI: cortical thicknessNo significant correlations were found between cortical thickness and BMI z-score using FDR multiple comparison correction
ROIs
BMI z-score

Yokum and Stice [101], USA [72]N = 162, (81 female), A: 15.3 (1.1)MRI: VBM: GMV, WMVBF gain < BF loss: Pu
N = 60 completed a second scan at two- or three-year follow-up.BMI: BF gain, BF stability, BF lossBF gain > BF loss: dorsal/subgenual ACC
BF gain > BF stability: subgenual ACC

Marqués-Iturria et al. [103], SpainN = 39MRI: ICV, TGM, CGM, sub-CGM, WMV, mean surface, mean thickness↑ BMI, ↓ L CGM: SFG, MFG (caudal), ACC (rostral and caudal), preCG
Obesity = 19, A: 33.7 (5.7), BMI: 36.1 (5.9)BMI↑ BMI, ↓ R CGM: mOFC, SFG, MFG (rostral), FPole, OFC
Lean = 18, A: 32.3 (5.9), BMI: 22.54 (1.94)↑ BMI, ↓ V: L vDC, R vDC, brainstem

Ottino-González et al. [104], Spain/CanadaN = 63MRIOverweight: ↑ AL index, ↓ cortical thickness Lean: ↑ AL index, ↑ cortical thickness in pars triangularis, L medial SFG, supramarginal gyrus, IPC, precuneus
OW-obese = 34, A: 31.8 (6.1), BMI: 31.4 (5)Cortical thickness
Lean = 29, A: 30.07 (6.2), BMI: 22.35 (1.82)BMI, WC
AL index (15 biomarkers)

Shan et al. [105], ChinaN = 76MRI: GMV↑ Creatinine, ↑ GMV ↑ HbA1c, LDL, triglyceride, ↓ GMV
Obese = 37, A: 27.8 (6.9), BMI: 40 (6.5)BMI, DBP, creatinine; FBS, HbA1c; HDL, LDL; SBP, SUA, triglyceride
Lean = 39, A: 26.7 (6.8), BMI: 21.8 (1.8)Obese < lean
R SPG, SOG, AG, cerebellum

Beyer et al. [106], GermanyN = 748, A: 68.4 (4.8)MRI: VBM: GMV, ICVMetabolic profile (↑ BMI, WHR, HbA1c, leptin, CRP, ↓ adiponectin) associated with ↓ GMV
BMI: 27.7 (4.1)BMI, WHR, metabolic profile (adiponectin, CRP, HbA1c, HDL, IL-6, SBP, TC)Th, B insular Cx, L amygdala-hip, TPole, cerebellum

Kotkowski et al. [107], USAN = 208MRI: VBM: GMVMetabolic syndrome was associated with ↓ GMV
MetS = 104, A: 37.3 (13.2), BMI: 34.5 (5.5)Cerebellum, OFC, R insula, cuneus, B Cn (body), R TL, R amygdala, BA34.
Controls = 104, 37.3 (13.2), BMI: 24.3 (3.8)MetS

Climie et al. [108], AustraliaN = 560MRI: VBM: GMV, hip volumeT2DM is associated with ↓ GMV, ↓ hip volume, ↑ WHR, BMI, ↓ physical activity
T2DM = 258, A: 67 (7)WHR, BMI, physical activity (steps/day), and blood biochemistryT2DM-GMV association is mediated by WHR
Non-T2DM = 302, A: 72 (7)

Nouwen et al. [109], UKN = 53MRI: VBM, TBSS: GMV, WMV, T2DM, obesity GMV: control > T2DM: B Cn, B PuControl > obese: R hip, R amygdala-hip, L amygdala, L Pu
T2DM = 14, A: 16.1 (1.5)Control > T2DM + obese: R Pu/Cn, L Cn, L Pu, L Th/Cn, L hip/amygdala
Obesity = 20, A: 14.9 (2)WM: T2DM < controls: L CST, mCC, L fornix, L ThR, L RLIC, L IFOF, R ACR, CC (genu), L uncinate, cingulum

Bernardes et al. [110], Brazil/ NetherlandsN = 75MRI: CT, surface area, GMVCT: obese + T2DM < controls, obese (normoglycemic) < controls: R insula
Obese + T2DM = 28, A: 60.4 (5)T2DM, adiposity, glucose, and insulinSurface area: obese + T2DM < controls: L lateral occipital
Obese (normoglycemic) = 16, A: 58 (8.4)Surface area, GMV: Th, obese + T2DM < obese (normoglycemic): L paracentral
Control = 31, A: 57 (7.1)

Rofey et al. [111], USAN = 15MRI: GMVGMV, WM integrity: Cn, Th
T2DM = 5, A: 18 (1.4)BMI, T2DMT2DM < obese < lean: Cn
Obese (without T2DM) = 5, A: 15.4 (2.2)T2DM < obese; T2DM < lean: Th
Lean = 5, A: 14.5 (2.5)

Redel et al. [112], USAN = 40MRI: VBM: GMV, voxel-wise GM, WMVGlobal GMV: obese + T2DM < control
T2DM = 20, A: 16.7 (2.6), BMI: 37.8 (4.7)BMI, T2DMRegional GMV: Obese + T2DM < control: R mTL, L fusiform, L IPG, R cerebellum, B ITG, R cingulum, R MOG, R IOG, L Cn, L angular, R lingual
Control = 20, A: 16.7 (2), BMI: 24.5 (5.2)Obese + T2DM > control: B Pu, R ITG, R Th, B paracentral

Yoon et al. [113], South KoreaN = 150MRI: GMV, CTGlobal mean CT: T2DM + OW and T2DM < control
T2DM + OW = 50, A: 49 (7.4), BMI: 28.1 (2)T2DM, BMIRegional CT: T2DM + OW and T2DM < control: B FG, L ITG, B STG, B SFG, L MFG, B IFC, L insula, B lingual Cx, R OFC
T2DM = 50, A: 49.3 (8.1), BMI: 22.8 (2)WM: T2DM + OW and T2DM < control: Fornix, CC, CST, IFOF, ATR
Control = 50, A: 49 (7.8), BMI: 22.7 (1.8)Duration-related alteration in CT: T2DM + OW > T2DM

Caunca et al. [114], USA/JapanN = 1,289 (from the Northern Manhattan MRI Substudy), A: 64 (8)MRI: ICV, mean CT, WMHV↑ BMI and ↑ WC z-score, ↓ mean CT
Obese < normal: CT
BMI z-score, WHR z-score, WC z-score, adiponectinNo association between obesity and cerebral small vessel disease (WMHV)

Hayakawa et al. [116], JapanN = 792MRI: voxel-wise analysesMen: ↑ WC, BMI, ↓ GMV: B ITG, R MTG, L fusiform, L Th, L red nucleus, L mFG, R SFG, R IFG, R preCG, R postCG, B cerebellum
Men = 523, A: 55.3 (9.7)GMV; GMr, ICVWomen: ↑ WC, BMI, ↓ GMV: B Th, R red nucleus, L preCG, L IFG, L mFG, R hip
BMI: 24.7 (3.1), WC (cm): 88.5 (8.1)BMI, WCRegions were more widespread in men than in women
Women = 269, A: 55.2 (9.9)
BMI: 22 (3.3), WC (cm): 81.2 (9.8)

Hamer and Batty [117], UKN = 9,652 (from UK Biobank study) Obese = 1,805, A: 55 (7.3)MRI: GMV↑ BMI, WHR, total fat mass, ↓ GMV
OW = 4,167, A: 56 (7.5)BMI, WHR, total FM, FM index (total FM (kg)/height (m))Combination of overall obesity (BMI) and central obesity (WHR) was associated with the lowest GMV
Lean = 3,680, A: 54.7 (7.5)↑ BMI, ↓ Pu, Pa, Nac
↑ WHR, ↓ Th, Cn, Pa
↑ Fat index, ↓ Pu, Pa, Nac

Croll et al. [118], NetherlandsN = 3,648 (from Rotterdam Study) Cross-sectional study = 3,648, A: 65.9 (11.1)MRI: GMV, WMV↑ BMI and FM index were cross-sectionally associated with ↓ WMV
Longitudinal study = 1,844, A: 60.9 (9.9)BMI, DXA: FM index, FFM index↑ FFM index was associated with ↑ GMV

Note. Brain Anatomy. ACC: anterior cingulate cortex; ACR: anterior corona radiata; AG: angular gyrus; ATR: anterior thalamic radiation; B: bilateral; Cn: caudate nucleus; CC: corpus callosum; CG: cingulate gyrus; CST: corticospinal tract; DS: dorsal striatum; FG: fusiform gyrus; FPole: frontal pole; Hip: hippocampus; IC: internal capsule; IFC: inferior frontal gyrus; IFOF: inferior fronto-occipital fasciculus; ILF: inferior longitudinal fasciculus; IOG: inferior occipital gyrus; IPG: inferior parietal gyrus; ITG: inferior temporal gyrus; L: left; mCC: medial corpus callosum; mTL: medial temporal lobe; mOFC: medial orbitofrontal cortex; mFG: medial frontal gyrus; MFG: middle frontal gyrus; mPFC: medial prefrontal cortex; MTG: middle temporal gyrus; MOG: middle occipital gyrus; Nac: nucleus accumbens; OFC: orbitofrontal cortex; OL: occipital lobe; Pa: pallidum; PHipG: parahippocampal gyrus; PTR: posterior thalamic radiation; postCG: postcentral gryrus; pre CG: precentral gyrus; Pu: putamen; R: right; RLIC: retrolenticular internal capsule; RO: Rolandic operculum; SCR: superior corona radiata; SFG: superior frontal gyrus; SFOF: superior fronto-occipital fasciculus; SLF: superior longitudinal fasciculus; SPL: superior parietal lobe; SPG: superior parietal gyrus; SOG: superior occipital gyrus; STG: superior temporal gyrus; Th: thalamus; TL: temporal lobe; TPole: temporal pole; UF: uncinate fasciculus; vDC: ventral diencephalon. Neuroimaging. CGM: cortical gray matter; CT: cortical thickness; DTI: diffusion tensor imaging; FDR: false discovery rate; GMM: gray matter mass; GMr: gray matter ratio (percentage of GMV in the intracranial volume); GMV: gray matter volume; ICV: intracranial volume; MRI: magnetic resonance imaging; rsFC: resting-state functional connectivity; ROI: region of interest; sub-CGM: subcortical gray matter; VBM: voxel-based morphometry; V: volume; TBSS: track-based spatial statistic; TGM; total gray matter. Obesity Assessment. BMI: body mass index; BMI z-score: standardized BMI; BF: body fat; BFD: body fat distribution; DBP: diastolic blood pressure; DXA: dual-energy X-ray absorptiometry; FFM: fat-free mass; FBS: fasting blood sugar (fasting glucose); FM: fat mass; FPI: fasting plasma insulin levels; HbA1c: glycated hemoglobin; HDL: high-density lipoprotein; LDL: low-density lipoprotein; MetS: metabolic syndrome; OW: overweight; SBP: systolic blood pressure; SUA: serum uric acid; TBF: total body fat; TC: total cholesterol; WC: waist circumference; WHR: waist-to-hip ratio; WMV: white matter volume.