Cover Page The handle http://hdl.handle.net/1887/45008

The handle http://hdl.handle.net/1887/45008 holds various files of this Leiden University dissertation

Author: Sala, Michiel

Title: MR and CT evaluation of cardiovascular risk in metabolic syndrome

Issue Date: 2016-12-14

Michiel L. Sala Jeroen van der Grond Renée de Mutsert Diana van Heemst P. Eline Slagboom Lucia J.M. Kroft Albert de Roos

AJR Am J Roentgenol, 2016 May 21;206(5):1087-92 CHAP TER 4

markers of incipient brain injury in middle-

aged to elderly persons with overweight

ABSTRACT

Objective: Obesity has been related to structural brain abnormalities. Increasing evidence sug- gests that accumulation of fat in the liver is an important determinant of cardiometabolic compli- cations of obesity. We aim to investigate the association between computed tomography (CT) assessed liver to spleen (L/S) attenuation ratio as a measure of liver steatosis and magnetic resonance imaging (MRI) assessed brain tissue integrity in middle-aged to elderly persons.

Materials and Methods: CT and MRI were performed in 213 participants (99 men; mean age 65 ± 7 years) from the Leiden Longevity Study. Brain tissue integrity was assessed by magnetiza- tion transfer imaging. Linear regression analysis was adjusted for age, sex, vascular risk factors, and total body fat estimated from bioelectrical impedance analysis.

Results: 79 participants had normal weight (body mass index [BMI] 18.5 – 24.9 kg/m

) and 134 were overweight (BMI ≥ 25 kg/m

). Significant interaction was found between L/S ratio and BMI (p = 0.001). In the overweight group, liver fat was associated with reduced brain tissue integrity in both grey (standardized β = 0.22, 95% confidence interval [CI] 0.07 to 0.36) and white matter (standardized β = 0.31, 95% CI 0.15 to 0.45). These associations were not found in the normal weight group (grey matter: standardized β = -0.08, 95% CI -0.33 to 0.16; white matter: standardized β = -0.09, 95% CI -0.36 to 0.14).

Conclusion: Our data indicate that liver fat assessed by CT relates to MRI markers of incipient

brain injury in middle-aged to elderly overweight persons.

Chap ter 4

INTRODUCTION

Obesity, particularly in midlife, is associated with an increased risk of dementia in later life (1).

Magnetic resonance imaging (MRI) enables in vivo characterization of cerebral changes, and previous imaging studies have shown that obesity relates to overt brain abnormalities (2,3).

Adiposity is frequently assessed by using anthropometric measures, the most common be- ing body mass index (BMI). Although BMI is an easily obtainable clinical measure of obesi- ty, it does not distinguish between body fat and body lean mass, and furthermore does not account for the distribution of fat between different compartments. It has been proposed that specific fat depots may predispose certain individuals to developing obesity related illnesses (4).

Increasing evidence suggests that accumulation of fat in the liver is an important determinant of the cardiometabolic complications of obesity (5). For example, it has been shown that fatty liver is associated with dyslipidemia and dysglycemia, also after adjusting for the amount of abdominal visceral fat (6). Fatty liver is frequently associated with obesity, especially in type 2 diabetes (7), although multiple other factors including significant alcohol consumption, use of steatogenic medication, and hereditary disorders may also cause liver fat accumulation (8).

In the obese state, overexposure of the liver to nonesterified fatty acids, that result from in- creased lipolysis, and to increased amounts of inflammatory cytokines from the blood pool, may lead to liver steatosis. Accordingly, fatty liver contributes to increased cardiovascular risk through the systemic release of several inflammatory, hemostatic, and oxidative-stress medi- ators, and by contributing to atherogenic dyslipedimia and insulin resistance (9). It has been shown that circulating levels of several proatherogenic factors are highest in nonalcoholic steatohepatitis, intermediate in simple steatosis, and lowest in control subjects without ste- atosis (9). These findings indicate that there is a graded relationship between the severity of histologic changes and the systemic abnormalities observed in patients with liver steatosis.

Accumulation of fat in the liver is closely related to metabolic syndrome (10), a cluster of vascular and metabolic risk factors which in turn has been associated with brain damage and cognitive decline (11,12). We hypothesized that liver fat may constitute increased cardiovascular risk and thus may aggravate subclinical brain disease in obesity. Accordingly, the purpose of our study was to investigate the association between liver fat assessed by computed tomography and brain tissue integrity assessed by magnetization transfer imaging (MTI) in normal weight and overweight middle-aged to elderly individuals.

MATERIALS AND METHODS Study design and study population

Our institutional review board approved the study and all participants gave informed consent.

For the current study, offspring of long-lived siblings and their partners were included from the

Leiden Longevity Study, which has been described previously (13). Long-lived siblings had been

selected by the following inclusion criteria: (1) having at least one other living sibling who fulfilled the age criteria and who was willing to participate, (2) men had to be aged 89 years or older and women had to be aged 91 years or older and (3) the sib pairs had to have the same parents For the present study, offspring of these long-lived siblings were asked to participate. It has been shown that offspring of long-lived siblings have a 30% lower mortality rate compared to the general population (13). Their partners, who share the same geographical and socio-eco- nomic background, were also enrolled (13). Although there were no selection criteria on health or demographic characteristics, the selection results in a relatively healthy study population.

Results from analyses with visceral adipose tissue (VAT) and MTI have been reported previously (14). For this study, liver fat and total body fat data were newly and independently analyzed in association with brain MTI, these results were not reported previously. Exclusion criteria were diabetes mellitus, transient ischemic attack or stroke, and BMI < 18.5 kg/m

. Information on smoking status was obtained from the study subjects. Blood pressure measurements were per- formed between November 2006 and May 2008. Four blood pressure measurements were taken from the arm while study participants were seated, and the average of these measurements were used for further analyses. In total, 213 individuals comprising 119 offspring and 94 partners were included. All participants underwent MRI of the brain and abdominal CT on the same day.

Imaging was performed between September 2009 and December 2010. Within this timeframe, blood samples were obtained for the determination of non-fasting serum glucose levels (13).

Assessment of total body fat

Total body fat (TBF) was assessed using direct segmental multi-frequency bioelectrical imped- ance analysis, procedure has been described in more detail elsewhere . During this session, study participants were weighted and their body height was measured for calculating BMI. (15) Assessment of abdominal and liver fat

CT acquisition

All examinations were performed on a 320 multidetector-row CT scanner (Aquilion ONE, Toshi- ba Medical Systems, Otawara, Japan). For abdominal fat measurement, a single-slice 8.0 mm acquisition was planned at the level of the 5th lumbar vertebra. Scan parameters were: tube volt- age 120 kV, tube current 310 mA, rotation time: 0.5 sec. Imaging was performed during breath hold after expiration. For liver fat measurement a single-slice 8.0 mm acquisition was planned at the T12/L1intervertebral disc. Scan parameters were: tube voltage: 120 kV, tube current: 155 mAs, rotation time: 0.5 sec.

CT analysis

Analysis of VAT has been described previously (14). Briefly, measurements were performed with

dedicated fat measurement software available at the CT scanner console. An automatic tool

Chap ter 4 was used for recognizing fat. Predefined thresholds for adipose tissue were set from -150 to -30

Hounsfield units (16). By manual outlining of the abdominal muscular wall, subcutaneous adi- pose tissue was separated from intraabdominal adipose tissue surrounding the internal abdom- inal organs, which includes intraperitoneal and retroperitoneal adipose tissue. Intraabdominal adipose tissue is referred to as VAT. Accordingly, VAT areas (cm

) were calculated automatically.

To quantify liver fat, the attenuation was measured in Hounsfield units (HU) by placing a region of interest (ROI) peripheral in the right liver lobe (figure 1). The ROI was made as large as possi- ble (at least 1 cm

), avoiding the hepatic vessels or any focal lesions. Accordingly, to normalize measures, HU were measured in the spleen. One cross-sectional slice has been shown to ad- equately capture the majority of variance in liver fat content, and for a single versus three ROI measures in the liver, the intraclass coefficient has been shown 0.99 (17). To measure liver fat content, liver-spleen ratio (L/S) was calculated, where L was the hepatic attenuation in HU and S is the splenic attenuation in HU. Lower attenuation values correspond to lower tissue density, and indicated greater fat content. Low L/S ratios indicated more liver steatosis. Fatty liver disease was defined as L/S ratio < 1 (18).

Figure 1. Unenhanced Computed Tomography (CT) scan of the upper abdomen, axial view. Left:

61-year-old female with normal liver; liver attenuation was 62 Hounsfield units (HU), spleen was 54 HU, liver to spleen ratio was 1.1. Right: 62-year-old male with fatty liver; liver attenuation was 24 HU, spleen attenuation was 76 HU, liver to spleen ratio was 0.3.

Assessment of brain tissue integrity MRI acquisition

Brain imaging was performed on a whole body MRI system operating at 3 Tesla field strength

(Philips Medical Systems, Best, The Netherlands). 3D T1-weighted images were acquired with

repetition time (TR) = 9.7 ms, echo time (TE) = 4.6 ms, flip angle (FA) = 8°, and 224 x 177 x 168

mm field of view (FOV), resulting in a nominal voxel size of 1.17 x 1.17 x 1.4 mm. Magnetization

transfer imaging was performed with TR = 100 ms, TE = 11 ms, FA = 9°, FOV = 224 x 180 x 144

mm, matrix size 224 x 169, 20 slices, slice thickness 7 mm.

MRI analysis

FSL (Functional MRI of the Brain Software Library) software package was used for analysis (19).

To create individual brain masks for cortical grey and white matter, 3D T1-weighted images were skull-stripped (20) and subsequently segmented (21). In magnetization transfer imaging, MTR is the most commonly used measure, reflecting the efficiency of magnetization exchange between protons in tissue compared with surrounding water (22). Individual MTR maps were calculated voxel by voxel following the equation MTR = (M0-M1)/M0. Low MTR is usually considered indicative of damage to myelin and other cellular structures. The frequency distribution of all MTR values were displayed as histograms. The peak height of the histogram indicates the number of voxels that show the most common MTR value per region of interest (23). In healthy individuals, the histogram is characterized by a single sharp peak, indicating that normal brains are relatively homogeneous in terms of MT imaging characteristics. Lower MTR peak height may be indica- tive of brain tissue decline (23). It has been shown previously that MTR histogram peak height is a sensitive marker of changes in brain tissue integrity (24-26). The peak height of the MTR histogram was divided by the number of voxels of brain tissue to normalize head size and brain atrophy (23). MT-MRI measures did not exceed - 3 or + 3 standard deviations.

Statistical analysis

Variables are expressed as mean ± standard deviation or percentage. Differences in baseline characteristics and measures of adiposity and brain tissue integrity between normal weight (BMI 18.5 – 24.9 kg/m

) and overweight (BMI ≥ 25 kg/m

) participants were assessed using stu- dent’s t-test and Pearson chi-square test. The correlation between L/S ratio, TBF, and VAT was tested by Pearson correlation analysis. Linear regression analysis was performed to investigate the association between L/S ratio and grey and white matter mean MTR and MTR peak height, with adjustment for age, sex, descent (offspring of long-lived siblings or partner), systolic blood pressure, and smoking (model 1). The presence of interaction between L/S ratio and BMI was tested by including an interaction term to this model. The model was then adjusted for TBF (model 2) and TBF and VAT (model 3). Standardized β-coefficients, 95% confidence intervals (CI), and P-values are reported. P-values < 0.05 were considered significant. Statistical analysis was per- formed with SPSS version 22.

RESULTS

Baseline characteristics of the study group are presented in table 1. Twelve patients with diabetes,

5 patients with clinically manifest cerebrovascular disease, and 2 patients with BMI < 18.5 kg/m

were excluded. In total, 213 participants were included: 79 had normal weight (BMI

18.5 – 24.9 kg/m

) and 134 were overweight (BMI ≥ 25 kg/m

) of whom 28 were obese

(BMI ≥ 30 kg/m

). Mean age was 65 ± 7 years and mean BMI was 26.4 ± 3.7 kg/m

.

Chap ter 4 Table 1. Baseline characteristics

Whole cohort Normal weight (BMI 18.5 – 24.9)

Overweight

(BMI ≥ 25) P-Value

Participants (n) 213 79 134

Age (years) 65 ± 7 65 ± 7 66 ± 3 0.186

Male (%) 99 (46%) 31 (39%) 68 (51%) 0.104

Offspring of long-lived (%) 119 (56%) 44 (56%) 75 (56%) 0.969

Body mass index (kg/m

) 26.4 ± 3.7 22.9 ± 1.6 28.4 ± 3.1 <0.001

Nonfasting glucose (mmol/L) 5.8 ± 1.0 5.9 ± 1.2 5.8 ± 1.0 0.404

Systolic blood pressure (mm Hg) 141 ± 19 137 ± 20 143 ± 19 0.050

Current smoking (%) 28 (13%) 13 (16%) 15 (11%) 0.272

Fatty liver (L/S ratio < 1) (%) 45 (21%) 8 (10%) 37 (28%) 0.003

Values are represented as means ± SD or percentage. L/S ratio: liver to spleen CT attenuation ratio, low LS ratio values represent increasing liver steatosis.

Prevalence of fatty liver disease was higher in the overweight group com- pared to the normal weight group (10% versus 28%, p = 0.003).

Overweight participants had a higher body fat percentage (p < 0.001), more VAT (p <

0.001), lower liver attenuation (p < 0.001) and L/S ratio (p < 0.001) compared to nor- mal weight participants (Table 2). Spleen attenuation was similar between the two groups (p = 0.915). MTR peak height was lower in overweight compared to normal weight par- ticipants (grey matter: p = 0.003; white matter: p = 0.002). Mean MTR values were sim- ilar between the two groups (grey matter: p = 0.926; white matter: p = 0.692) (Table 2).

Table 3 shows the results from Pearson correlation analysis between L/S ra- tio, TBF, and VAT. L/S ratio was correlated to VAT (r = -0.44, p < 0.001) but not TBF (r = -0.10, p = 0.135). VAT was correlated to TBF (r = 0.31, p < 0.001).

Table 4 shows the association between L/S ratio and MTR peak height. Liver fat was asso- ciated with reduced brain tissue integrity in both grey ( β = 0.21, 95% CI: 0.09 to 0.34) and white matter ( β = 0.21, 95% CI: 0.09 to 0.34) after adjusting for age, sex, descent (off- spring of long-lived siblings or partner), systolic blood pressure, and smoking. Associa- tions remained for grey ( β = 0.14, 95% CI: 0.02 to 0.27) and white matter (β = 0.15, 95%

CI: 0.03 to 0.29) after additional adjustment for TBF. After additionally adjusting for the amount of VAT, the association between liver fat and brain tissue integrity attenuated (grey matter, β = 0.09, 95% CI: -0.04 to 0.22; white matter, β = -0.08, 95% CI: -0.05 to 0.22).

There was a significant interaction between L/S ratio and BMI (p < 0.001 for mod-

el with grey matter MTR peak height as outcome variable; p = 0.001 for model with

Table 2. Measures of adiposity and brain tissue integrity

Whole cohort Normal weight (BMI 18.5 – 24.9)

Overweight

(BMI ≥ 25) P-Value

Total body fat (%) 30 ± 8 25 ± 7 33 ± 8 <0.001

VAT (cm

) 126 ± 57 92 ± 53 146 ± 49 <0.001

Liver attenuation (HU) 61 ± 10 64 ± 9 59 ± 10 <0.001

Spleen attenuation (HU) 55 ± 4 55 ± 3 55 ± 4 0.915

L/S ratio 1.1 ± 0.2 1.18 ± 0.2 1.09 ± 0.2 0.001

MTR peak height (pixel count x 10

)

Grey matter 76 ± 11 79 ± 11 75 ± 11 0.003

White matter 118 ± 22 125 ± 22 115 ± 22 0.002

Mean MTR (pixel count x 10

)

Grey matter 335 ± 9 335 ± 10 335 ± 9 0.926

White matter 394 ± 9 394 ± 12 394 ± 8 0.692

Values are represented as means ± SD or percentage. Total body fat was estimated from bioelectrical impedance analysis. Visceral adipose tissue (VAT) area in cm

and liver and spleen attenuation were assessed by computed tomography. L/S ratio: liver to spleen CT attenuation ratio, low LS ratio values represent increasing liver steatosis. Brain tissue integrity was assessed by magnetization transfer ratio (MTR). HU: Hounsfield units.

Table 3. Results from Pearson correlation analysis between L/S ratio, total body fat, and VAT

L/S ratio – TBF -0.44**

L/S ratio – VAT -0.10

VAT - TBF 0.31**

Pearson correlation coefficients are shown. L/S ratio: liver to spleen CT attenuation ratio, low LS ratio values represent increasing liver steatosis. TBF: total body fat in percentage. VAT: visceral adipose tissue area in cm

. * P-value < 0.05, ** P-value < 0.01

Table 4. Association between L/S ratio and MTR peak height

L/S Grey matter White matter

β 95% CI β 95% CI

Crude 0.26 0.13 to 0.40 0.26 0.13 to 0.40

1. Adjusted 0.21 0.09 to 0.34 0.21 0.09 to 0.34

2. Adjusted + TBF 0.14 0.02 to 0.27 0.15 0.03 to 0.29

3. Adjusted + TBF, VAT 0.09 -0.04 to 0.22 0.08 -0.05 to 0.22

Standardized beta coefficients with corresponding 95% confidence interval (CI) are shown. Model 1 adjusted for age, sex,

descent (offspring of long-lived siblings or partner), systolic blood pressure, and smoking. Model 2 included model 1 and adjusted

Chap ter 4 white matter MTR peak height as outcome variable), indicating that the relationship be-

tween L/S ratio and brain tissue integrity is significantly different across BMI groups. There- fore, analyses were repeated separately in normal weight and overweight individuals.

Results of linear regression analyses per standard deviation of L/S ratio in normal weight and over- weight participants are shown in Figure 2. In the overweight group, liver fat was associated with reduced brain tissue integrity in both grey ( β = 0.29, 95% CI: 0.14 to 0.44) and white matter (β = 0.27, 95% CI: 0.11 to 0.41) after adjusting for age, sex, descent (offspring of long-lived siblings or partner), systolic blood pressure, and smoking. These associations remained when TBF was includ- ed in the model (grey matter, β =0.22, 95% CI: 0.07 to 0.36; white matter, β = 0.31, 95% CI: 0.15 to 0.45). After additionally adjusting for the amount of VAT, liver fat remained associated to grey ( β

= 0.17, 95% CI: 0.01 to 0.32) and white matter MTR peak height ( β = 0.16, 95% CI: 0.00 to 0.31).

In these overweight participants, one standard deviation decrease in L/S ratio was associated with an overall decrease in MTR peak height of 2% in the grey matter and 5% in the white matter.

Liver fat was not associated with MTR peak height in the normal weight group (grey mat- ter: β = -0.08, 95% CI -0.33 to 0.16; white matter: β = -0.09, 95% CI -0.36 to 0.14).

There was no association between liver fat and grey or white matter mean MTR (data not shown).

DISCUSSION

The main finding of our study was that liver fat is associated with incipient brain injury in over- weight middle-aged to elderly individuals without clinically manifest cerebrovascular disease.

Decreasing L/S ratio was associated with lower MTR peak height in both the grey and white matter. This was independent of age, sex, descent (offspring of long-lived siblings or partner), systolic blood pressure, smoking, and total body fat. However, after additionally adjusting for the amount of abdominal VAT, associations between liver fat and brain tissue integrity attenu- ated. Nevertheless, after repeating our analyses separately in normal weight and overweight participants, we found that decreasing L/S ratio was associated with grey and white matter MTR peak height in overweight participants only, not in normal weight participants. This finding was independent of total body fat and VAT, indicating that the observed relation between liver fat and brain tissue integrity in overweight individuals cannot be accounted for by overall adiposity nor differences in visceral abdominal fat deposition.

Previous studies have shown that overall adiposity relates to brain abnormalities assessed by MRI

(2,3,27). To the best of our knowledge, no previous study has tested the hypothesis that liver fat

relates to brain injury after taking total body fat and visceral fat into account. One recent study

used MTI to assess brain damage in obesity. This study showed that increasing total body fat was

associated with higher mean MTR values in both grey and white matter in healthy adolescents

(12 – 18 years old) (28).

Figure 2. Association between L/S ratio and grey (Fig. 2A) and white matter (Fig. 2B) MTR peak height in normal weight and overweight participants. Standardized beta coefficients with corresponding 95%

confidence interval are shown. Although 95% confidence intervals were relatively wide in the normal weight group (Fig. 2A and B, upper panels) compared to the overweight group (Fig 2. A and B, lower panels), point estimates were negative and close to zero, indicating no relation between L/S ratio and brain tissue integrity in normal weight participants. In contrast, in overweight individuals, L/S ratio was associated with grey (Fig 2A, lower panel) and white matter (Fig. 2B, lower panel) MTR peak height, even after adjusting for total body fat (TBF) and visceral adipose tissue (VAT). Adjusted: adjusted for age, sex, descent (offspring of long-lived siblings or partner), systolic blood pressure, and smoking. L/S ratio:

liver to spleen CT attenuation ratio, low L/S ratio values represent increasing liver steatosis. BMI: body mass index in kg/m

.

The authors suggested that the positive association between adiposity and mean MTR values may be due to higher myelination of white matter fibers and intracortical axons in those with high- er adiposity (28). We observed that increasing amounts of liver fat were associated with lower MTR peak height, which may indicate brain tissue decline (23). Consistent with our observed direction of the relation between adiposity and MTR, another recent study showed that increasing BMI was associated with lower MTR peak height values (12). In addition, one study showed that increasing amounts of VAT, assessed by CT, related to lower MTR peak height in middle-aged to elderly individuals (14). Furthermore, previous diffusion tensor imaging studies have shown reduced white matter integrity, potentially due to axonal and myelin degeneration, in association with greater adiposity (27,29,30).

It has been shown that increasing BMI as a measure of obesity is associated with brain tissue

decline assessed by MTI (12). However, BMI does not account for the distribution of fat between

different compartments and does not necessarily represent reduced health or pathologic state

such as fatty liver (4). We therefore used liver fat as imaging biomarker that may better discrim-

inate between health and disease. Indeed, decreasing L/S attenuation ratio was associated

with decreased MTR peak height in overweight individuals. This finding was independent of total

body fat estimated from bioelectrical impedance analysis, indicating that our observed relation

Chap ter 4 between liver fat and brain tissue integrity cannot be accounted for by overall adiposity or body

weight.

While increasing evidence suggests that accumulation of fat in the liver is an important determi- nant of cardiometabolic complications in obesity (5), VAT is another fat depot that has been asso- ciated with increased cardiovascular risk (4). It has also been shown that abdominal VAT relates to structural brain abnormalities (14,31). Liver fat is strongly correlated to VAT, we therefore ad- justed our analyses for the amount of VAT. Then, the association between liver fat and brain tissue integrity attenuated when liver fat and VAT were considered simultaneously, suggesting that liver fat exerted no additional effect above and beyond VAT. Surprisingly, stratifying by BMI revealed that liver fat was associated with reduced brain tissue integrity in overweight participants, even after adjusting for VAT.

In line with our findings, one recent study showed that liver fat was associated with metabol- ic syndrome in overweight, but not in normal weight people (32). A possible explanation for this finding is that metabolic risk is greatest in those with the most extensive fatty liver infiltration (6,32). Indeed, our study comprised relatively healthy persons in which prevalence of severe liver steatosis (e.g. L/S ratio < 0.8 (33)) was relatively low. Because the amount of liver fat is correlated with adiposity, it is conceivable that leaner individuals do not have enough liver fat for developing adverse metabolic effects. It has been shown that there is a graded relationship between the severity of histologic changes and the systemic abnormalities observed in patients with fatty liver (9).

One putative mechanism underlying the association between liver fat and brain injury is chronic inflammation. The fatty liver is an amplifier of systemic inflammation through increased transcrip- tion of several proinflammatory genes (9), while inflammation in turn has been related to structur- al brain abnormalities in obesity (11,31).

MTI has been used to probe the integrity of macromolecular proteins and phospholipids in the

brain. Post mortem imaging and histopathology studies in patients with multiple sclerosis have

shown that in white matter lower MTR is associated with axonal loss and myelin compromise

(34,35). In grey matter, reduction in dendritic density and neuronal size or number and damage

to cell membranes may colzlectively or independently lead to lower MTR (36). The biologi-

cal mechanism underlying our observed reduced MTR peak height in overweight participants

remains to be elucidated, but it may be speculated that low-grade inflammation, insulin resis-

tance, or expression of adipose tissue derived hormones including adiponectin and leptin may

be involved in the association between liver fat and MTR peak height representing brain tissue

integrity (37-41).

There were several limitations in our study. First, sample size was relatively small, in particular in the normal weight group. Although 95% confidence intervals of the linear regression analyses were relatively wide in the normal weight group compared to the overweight group, point es- timates were negative and close to zero. It is therefore unlikely that our negative finding on the association between liver fat and brain tissue integrity in this group is due to lack of statistical power. Second, CT offers a semi-quantitative method for the assessment of liver fat with high reproducibility (42). Although ultrasound has been accepted as an initial screening for fatty liver without radiation exposure, this modality is highly operator dependent and does not provide reproducible quantitative information (42). Third, our findings support the idea that imaging bio- markers may help in advanced risk stratification. However, performing CT imaging for the sole purpose of liver fat measurement as to establish cardiometabolic risk is currently not indicated.

On the other hand, detection of liver steatosis at unenhanced CT performed for other reasons may warrant reporting and may trigger an indicated workup for metabolic syndrome. Finally, due to the observational nature of our study we cannot infer causality. Longitudinal studies should further elucidate associations between liver fat, brain injury, and cognitive function by using larg- er sample sizes through multicentre collaborations.

In conclusion, liver fat was associated with incipient brain injury in overweight middle-aged to el- derly individuals without clinically manifest cerebrovascular disease. Reduced brain tissue integri- ty consequent to normal aging processes may be exacerbated by mechanisms related to obesity.

Awareness of the underlying mechanism between fat accumulation in specific depots and brain

damage may offer more focused individual treatment than considering overall adiposity alone.

Chap ter 4

REFERENCE LIST

1. Whitmer RA, Gunderson EP, Barrett-Connor E, Quesenberry CP, Jr., Yaffe K: Obesity in middle age and future risk of dementia: a 27 year longitudinal population based study. BMJ 330:1360, 2005 2. Jagust W, Harvey D, Mungas D, Haan M: Central obesity and the aging brain. Arch Neurol 62:1545-

1548, 2005

3. Gustafson D, Lissner L, Bengtsson C, Bjorkelund C, Skoog I: A 24-year follow-up of body mass index and cerebral atrophy. Neurology 63:1876-1881, 2004

4. Tchernof A, Despres JP: Pathophysiology of human visceral obesity: an update. Physiol Rev 93:359- 404, 2013

5. Stefan N, Kantartzis K, Machann J, Schick F, Thamer C, Rittig K, Balletshofer B, Machicao F, Fritsche A, Haring HU: Identification and characterization of metabolically benign obesity in humans. Arch Intern Med 168:1609-1616, 2008

6. Speliotes EK, Massaro JM, Hoffmann U, Vasan RS, Meigs JB, Sahani DV, Hirschhorn JN, O’Donnell CJ, Fox CS: Fatty liver is associated with dyslipidemia and dysglycemia independent of visceral fat:

the Framingham Heart Study. Hepatology 51:1979-1987, 2010

7. Tolman KG, Fonseca V, Dalpiaz A, Tan MH: Spectrum of liver disease in type 2 diabetes and man- agement of patients with diabetes and liver disease. Diabetes Care 30:734-743, 2007

8. Chalasani N, Younossi Z, Lavine JE, Diehl AM, Brunt EM, Cusi K, Charlton M, Sanyal AJ: The di- agnosis and management of non-alcoholic fatty liver disease: practice Guideline by the American Association for the Study of Liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association. Hepatology 55:2005-2023, 2012

9. Targher G, Day CP, Bonora E: Risk of cardiovascular disease in patients with nonalcoholic fatty liver disease. N Engl J Med 363:1341-1350, 2010

10. Marchesini G, Bugianesi E, Forlani G, Cerrelli F, Lenzi M, Manini R, Natale S, Vanni E, Villanova N, Melchionda N, Rizzetto M: Nonalcoholic fatty liver, steatohepatitis, and the metabolic syndrome.

Hepatology 37:917-923, 2003

11. Yates KF, Sweat V, Yau PL, Turchiano MM, Convit A: Impact of metabolic syndrome on cognition and brain: a selected review of the literature. Arterioscler Thromb Vasc Biol 32:2060-2067, 2012 12. Sala M, de RA, van den Berg A, Altmann-Schneider I, Slagboom PE, Westendorp RG, van Buchem

MA, de Craen AJ, van der Grond J: Microstructural brain tissue damage in metabolic syndrome.

Diabetes Care 37:493-500, 2014

13. Schoenmaker M, de Craen AJ, de Meijer PH, Beekman M, Blauw GJ, Slagboom PE, Westendorp RG: Evidence of genetic enrichment for exceptional survival using a family approach: the Leiden Lon- gevity Study. Eur J Hum Genet 14:79-84, 2006

14. Widya RL, Kroft LJ, Altmann-Schneider I, van den Berg-Huysmans AA, van der Bijl N, de RA, Lamb

HJ, van Buchem MA, Slagboom PE, van HD, van der Grond J: Visceral adipose tissue is associated

with microstructural brain tissue damage. Obesity (Silver Spring) 23:1092-1096, 2015

15. Ling CH, de Craen AJ, Slagboom PE, Gunn DA, Stokkel MP, Westendorp RG, Maier AB: Accuracy of direct segmental multi-frequency bioimpedance analysis in the assessment of total body and segmen- tal body composition in middle-aged adult population. Clin Nutr 30:610-615, 2011

16. Yoshizumi T, Nakamura T, Yamane M, Islam AH, Menju M, Yamasaki K, Arai T, Kotani K, Funahashi T, Yamashita S, Matsuzawa Y: Abdominal fat: standardized technique for measurement at CT. Radiol- ogy 211:283-286, 1999

17. Speliotes EK, Massaro JM, Hoffmann U, Foster MC, Sahani DV, Hirschhorn JN, O’Donnell CJ, Fox CS: Liver fat is reproducibly measured using computed tomography in the Framingham Heart Study. J Gastroenterol Hepatol 23:894-899, 2008

18. Foster T, Budoff MJ, Saab S, Ahmadi N, Gordon C, Guerci AD: Atorvastatin and antioxidants for the treatment of nonalcoholic fatty liver disease: the St Francis Heart Study randomized clinical trial. Am J Gastroenterol 106:71-77, 2011

19. Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TE, Johansen-Berg H, Bannister PR, De LM, Drobnjak I, Flitney DE, Niazy RK, Saunders J, Vickers J, Zhang Y, De SN, Brady JM, Matthews PM: Advances in functional and structural MR image analysis and implementation as FSL. Neuroim- age 23 Suppl 1:S208-S219, 2004

20. Smith SM, Zhang Y, Jenkinson M, Chen J, Matthews PM, Federico A, De SN: Accurate, robust, and automated longitudinal and cross-sectional brain change analysis. Neuroimage 17:479-489, 2002 21. Zhang Y, Brady M, Smith S: Segmentation of brain MR images through a hidden Markov random field

model and the expectation-maximization algorithm. IEEE Trans Med Imaging 20:45-57, 2001 22. Seiler S, Cavalieri M, Schmidt R: Vascular cognitive impairment - an ill-defined concept with the need

to define its vascular component. J Neurol Sci 322:11-16, 2012

23. van Buchem MA, Grossman RI, Armstrong C, Polansky M, Miki Y, Heyning FH, Boncoeur-Martel MP, Wei L, Udupa JK, Grossman M, Kolson DL, McGowan JC: Correlation of volumetric magnetization transfer imaging with clinical data in MS. Neurology 50:1609-1617, 1998

24. Benedetti B, Charil A, Rovaris M, Judica E, Valsasina P, Sormani MP, Filippi M: Influence of aging on brain gray and white matter changes assessed by conventional, MT, and DT MRI. Neurology 66:535-539, 2006

25. Ropele S, Enzinger C, Sollinger M, Langkammer C, Wallner-Blazek M, Schmidt R, Fazekas F: The impact of sex and vascular risk factors on brain tissue changes with aging: magnetization transfer imaging results of the Austrian stroke prevention study. AJNR Am J Neuroradiol 31:1297-1301, 2010 26. van den Bogaard SJ, Dumas EM, Milles J, Reilmann R, Stout JC, Craufurd D, van Buchem MA, van der

Grond J, Roos RA: Magnetization transfer imaging in premanifest and manifest Huntington disease.

AJNR Am J Neuroradiol 33:884-889, 2012

27. Stanek KM, Grieve SM, Brickman AM, Korgaonkar MS, Paul RH, Cohen RA, Gunstad JJ: Obesity

is associated with reduced white matter integrity in otherwise healthy adults. Obesity (Silver Spring)

19:500-504, 2011

Chap ter 4 28. Schwartz DH, Dickie E, Pangelinan MM, Leonard G, Perron M, Pike GB, Richer L, Veillette S, Pauso-

va Z, Paus T: Adiposity is associated with structural properties of the adolescent brain. Neuroimage 103C:192-201, 2014

29. Mueller K, Anwander A, Moller HE, Horstmann A, Lepsien J, Busse F, Mohammadi S, Schroeter ML, Stumvoll M, Villringer A, Pleger B: Sex-dependent influences of obesity on cerebral white matter investigated by diffusion-tensor imaging. PLoS One 6:e18544, 2011

30. Verstynen TD, Weinstein AM, Schneider WW, Jakicic JM, Rofey DL, Erickson KI: Increased body mass index is associated with a global and distributed decrease in white matter microstructural integrity.

Psychosom Med 74:682-690, 2012

31. Debette S, Beiser A, Hoffmann U, Decarli C, O’Donnell CJ, Massaro JM, Au R, Himali JJ, Wolf PA, Fox CS, Seshadri S: Visceral fat is associated with lower brain volume in healthy middle-aged adults.

Ann Neurol 68:136-144, 2010

32. Kim LJ, Nalls MA, Eiriksdottir G, Sigurdsson S, Launer LJ, Koster A, Chaves PH, Jonsdottir B, Garcia M, Gudnason V, Harris TB: Associations of visceral and liver fat with the metabolic syndrome across the spectrum of obesity: the AGES-Reykjavik study. Obesity (Silver Spring) 19:1265-1271, 2011 33. Park SH, Kim PN, Kim KW, Lee SW, Yoon SE, Park SW, Ha HK, Lee MG, Hwang S, Lee SG, Yu ES,

Cho EY: Macrovesicular hepatic steatosis in living liver donors: use of CT for quantitative and qualita- tive assessment. Radiology 239:105-112, 2006

34. Schmierer K, Scaravilli F, Altmann DR, Barker GJ, Miller DH: Magnetization transfer ratio and myelin in postmortem multiple sclerosis brain. Ann Neurol 56:407-415, 2004

35. van Waesberghe JH, Kamphorst W, De Groot CJ, van Walderveen MA, Castelijns JA, Ravid R, Ni- jeholt GJ, van d, V, Polman CH, Thompson AJ, Barkhof F: Axonal loss in multiple sclerosis lesions:

magnetic resonance imaging insights into substrates of disability. Ann Neurol 46:747-754, 1999 36. Bagary MS, Symms MR, Barker GJ, Mutsatsa SH, Joyce EM, Ron MA: Gray and white matter brain

abnormalities in first-episode schizophrenia inferred from magnetization transfer imaging. Arch Gen Psychiatry 60:779-788, 2003

37. Haslam DW, James WP: Obesity. Lancet 366:1197-1209, 2005

38. Pou KM, Massaro JM, Hoffmann U, Vasan RS, Maurovich-Horvat P, Larson MG, Keaney JF, Jr., Meigs JB, Lipinska I, Kathiresan S, Murabito JM, O’Donnell CJ, Benjamin EJ, Fox CS: Visceral and subcutaneous adipose tissue volumes are cross-sectionally related to markers of inflammation and oxidative stress: the Framingham Heart Study. Circulation 116:1234-1241, 2007

39. Schmidt R, Schmidt H, Curb JD, Masaki K, White LR, Launer LJ: Early inflammation and dementia: a 25-year follow-up of the Honolulu-Asia Aging Study. Ann Neurol 52:168-174, 2002

40. Tan ZS, Beiser AS, Fox CS, Au R, Himali JJ, Debette S, Decarli C, Vasan RS, Wolf PA, Seshadri S: As-

sociation of metabolic dysregulation with volumetric brain magnetic resonance imaging and cognitive

markers of subclinical brain aging in middle-aged adults: the Framingham Offspring Study. Diabetes

Care 34:1766-1770, 2011

41. Jefferson AL, Massaro JM, Wolf PA, Seshadri S, Au R, Vasan RS, Larson MG, Meigs JB, Keaney JF, Jr., Lipinska I, Kathiresan S, Benjamin EJ, Decarli C: Inflammatory biomarkers are associated with total brain volume: the Framingham Heart Study. Neurology 68:1032-1038, 2007

42. Schwenzer NF, Springer F, Schraml C, Stefan N, Machann J, Schick F: Non-invasive assessment

and quantification of liver steatosis by ultrasound, computed tomography and magnetic resonance. J

Hepatol 51:433-445, 2009