Association of body weight and serum uric acid with Alzheimer's disease biomarkers and cognitive impairment

Abstract

Background

Given the limited effective treatments for Alzheimer's disease (AD), obesity and serum uric acid (SUA) levels which are considered modifiable risk factors for dementia are of interest. However, research indicates conflicting results.

Objective

We aimed to further investigate the association of body weight (BW) and SUA with AD biomarkers and cognitive impairment.

Methods

Clinical data were collected from 139 adults (mean age 66.9 years) with chronic cognitive impairment. Cerebrospinal fluid (CSF) biomarkers and PET imaging were used to assess amyloid-β (A) and Tau (T) tangles load, classifying participants into AT profiles based on the results. The association of BW and SUA with AT profiles was evaluated using multivariable logistic regression, and their relationship with cognitive function (Mini-Mental State Examination (MMSE) scores) were analyzed using multivariable linear regression.

Results

Lower BW levels significantly influenced the presence of Aβ positive state (A+) (p = 0.007), while SUA levels did not (p = 0.263). Higher dementia proportion (p = 0.021), lighter BW (p = 0.019), and lower mean arterial pressure (MAP) levels (p = 0.025) were associated with AD pathological progress (A-T-→A+T-→A+T+), but SUA was not observed statistically significant. Among all participants regardless of Aβ state, high education levels (p < 0.001), high BW (p = 0.010), and high SUA (p=0.036) were associated with high MMSE scores, and high serum creatinine (p = 0.003) was associated with low MMSE scores.

Conclusions

Lower BW may accelerate AD pathology and cause cognitive impairment, while SUA is not linked to AD pathological progression but protects cognitive function.

Keywords

Alzheimer's disease pathology biomarkers body weight cognitive function serum uric acid

Introduction

Alzheimer's disease (AD) is the most prevalent neurodegenerative cause of dementia,¹ posing significant challenges to global healthcare and social security systems. Given the limited availability of efficacious treatments for AD, it is imperative to underscore strategies for preventing its onset and progression. Therefore, there is great interest in modifiable risk factors.

Although several risk factors contribute to dementia,² the precise impact of obesity remains uncertain. Obesity or overweight in midlife has been found to be associated with an increased risk of subsequent dementia,^3–5 whereas among older adults, they may be linked to a decreased risk of dementia.^4,5 A retrospective cohort study of 2 million patients from UK Clinical Practice Research Datalink showed that obesity was associated with lower risk of developing dementia both in midlife and late-life individuals.⁶ Nevertheless, there is also evidence that increased body weight (BW) or abdominal obesity are associated with increased dementia incidence in older adults.⁷

Uric acid (UA) is considered another modifiable factor for dementia. Studies have indicated that UA may combat neurodegenerative diseases by powerfully attenuating oxidative stress and free radicals.^8,9 On the contrary, certain studies propose that UA may heighten the risk of AD or cognitive dysfunction by influencing amyloidosis or amplifying amyloid-β effects.^10,11 One plausible explanation is that the biological characteristics of serum UA (SUA) may vary depending on its chemical micro-environment and concentration in biological fluid in diverse populations.^11,12 The oxidation reaction results in the formation of UA free radicals, which can enhance the pro-oxidative effect and disrupt the delicate balance between antioxidation and pro-oxidation.¹³ The pro-oxidative properties of UA may elucidate its detrimental association with neurobiological alterations in the progression of AD. On the other hand, hyperuricemia has been linked to various vascular risk factors and diseases that could predispose individuals to dementia.¹⁴ Population-based longitudinal studies have demonstrated that notwithstanding the associated increased risk of vascular disease, higher levels of UA exert protective effect in the development of dementia,¹⁵ irrespective of whether it is AD or vascular dementia (VaD).¹⁶ Conversely, another large prospective cohort study of elderly people found high UA levels were associated with increased risk of dementia, especially vascular or mixed dementia.¹⁷ The association between UA and AD or cognitive impairment diseases remains inconclusive, with conflicting results.

Elevated SUA levels have been observed to coexist with obesity, as evidenced by epidemiological studies demonstrating positive associations between SUA and various adiposity markers such as body mass index (BMI) and body fat,¹⁸ suggesting a potential interaction between UA and BW that could confound their association with dementia. One prospective population-based cohort study demonstrated that higher BMI and SUA levels were independently associated with a reduced risk of developing dementia, including AD and VaD.¹⁹

Many previous studies that relied solely on clinical diagnostic criteria for AD without incorporating biomarkers may have included patients with other types of dementia, compromising the study's validity. With advancements in AD biomarkers, several studies have investigated the association between BW and AD biomarkers,^20–22 as well as UA and AD biomarkers^10,23 respectively. A consensus on the impact of BW and SUA together on AD biomarkers and cognitive impairment remains elusive.

In this study, we aimed to investigate the association between BW and SUA levels with amyloid deposition/tau tangle (AT) biomarkers according to the NIA-AA research framework,²⁴ as well as their relationship with cognitive abilities, in a hospital-based sample population involving adult individuals potentially diagnosed with AD or other forms of dementia.

Methods

Ethics approval

The study protocol has been approved by the Medical Ethical Review Board of the Third Affiliated Hospital of Sun Yat-sen University (ID: II2024-077-02). Written informed consent was obtained from each participant before the study procedure.

Participants

This cross-sectional analysis was conducted in the neurology department of Third Affiliated Hospital of Sun Yat-sen University. Participants aged 18 years or older were recruited from both inpatient and outpatient departments who exhibited the main clinical manifestations of “chronic hidden cognitive decline” between November 2018 and March 2023.

The cognitive function, BW, SUA levels, and other biochemical indicators of the subjects may exhibit excessive short-term fluctuations due to acute cerebral disorders or systemic diseases. These fluctuations can significantly impact the study results as confounding factors. Therefore, participants were excluded for the following reasons: suffering from rapidly progressive dementia such as CNS infection, autoimmune encephalitis, metabolic encephalopathy etc.; acute cerebrovascular disease (infarction or bleeding); multiple sclerosis; myelitis; nonorganic psychiatric disorders such as schizophrenia; severe systematic disorders including active stage tumors, severe liver insufficiency or renal insufficiency and severe infection etc.; or having major deficiency in clinical data. According to inclusion and exclusion criteria, 139 participants were included in the study.

Clinical data collection and cognitive assessment

Clinical data of participants were collected retrospectively from the electronic medical record database of Third Affiliated Hospital of Sun Yat-sen University, including age, gender, educational level, BW, past medical history (hypertension, diabetes etc.), smoking habit, blood pressure (BP), APOE genotyping, blood sample testing (glycosylated hemoglobin, lipids, UA, creatinine, aminotransferase, etc.), CSF sample testing and brain imaging [magnetic resonance angiography (MRA), computed tomography angiography (CTA) and positron emission tomography (PET)], etc.

Deep white matter lesions on magnetic resonance imaging were graded according to modified Fazekas scale²⁵ into grade 0, absent; grade 1, punctate; grade 2, beginning confluent; and grade 3, large confluent. Cerebral arteriosclerosis was defined as vascular wall stiffness, uneven thickness, tortuous running or local blood flow signal weakened, distal blood flow display poor with or without luminal stenosis on 3.0T MRA or multilayer spiral CTA by experienced radiologist. The APOE ε4 carrier status was determined by having at least one APOE ε4 allele. Cognitive performances were assessed with the Mini-Mental State Examination (MMSE) and dementia status was defined as scores ≤24 for middle school education and above, scores ≤20 for primary school education and scores ≤17 for illiteracy.²⁶

CSF biomarker testing and cut-off definitions

CSF biomarker data were available for 119 participants. The levels of CSF amyloid-β 40 (Aβ₄₀), amyloid-β 42 (Aβ₄₂), phosphorylated-tau181 (P-tau181), and total-tau (T-tau) were measured using enzyme-linked immunosorbent assay (ELISA) kits (EUROIMMUN) and absorbance microplate reader (TECAN Sunrise™). Aβ pathology positivity (A+) was defined as CSF Aβ_42/40 ≤ 0.1, and tau pathology positivity (T+) was defined as CSF P-tau181>61 pg/ml. The reference range was set according to the test kit laboratory performance verification, data accumulation and references.^27–29

¹⁸F-florbetapir PET and ¹⁸F-flortaucipir PET images collection and processing

¹⁸F-florbetapir PET (¹⁸F-AV45 PET) images were acquired in 24 participants, and ¹⁸F-flortaucipir PET (¹⁸F-AV1451 PET) were in 15 participants. ¹⁸F-AV45 and ¹⁸F-AV1451 PET were used to estimate cerebral Aβ load and tau tangles load, respectively. Aβ and tau tangles positive or negative (A+/A-and T+/T-) were estimated with semi-quantitative analysis. The regional standardized uptake value ratios (SUVRs) from both PET images were calculated.

AT profile classification definitions

Participants were divided into A- and A+ groups based on their CSF biomarkers and PET test results, and further classified into four AT biomarker profiles²⁴: A-T-, A+T-, A+T+, and A-T+. Only three AT profiles (A-T-, A+T-, and A+T+) were analyzed in this study due to the low number of individuals in the A-T+ profile which also standing for non-AD pathological change. A-T-→A+T-→A+T+ conforms to the current mainstream view that the process of gradual deposition and development of AD pathology.

Majority AT profiles were grouped based on CSF biomarkers, and PET was used when participants lacked CSF biomarkers. In cases of inconsistent results between CSF and PET data, PET data were used for the AT classification in our analyses because CSF analysis data are transient, and PET date represent a cumulative effect, more stable and more widely used in clinic.^30,31

Statistical analyses

Continuous variables were reported as mean ± standard deviation (SD) for normally distributed data, or median (P25, P75) for skewed distributions, determined using the Shapiro–Wilk test. Categorical variables were reported as counts (%). Student's t-test, Pearson's Chi-square test, Mann-Whitney U test, Analysis of Variance (ANOVA or Welch test, with Tukey's post hoc analysis) and Kruskal Wallis test were used to assess univariate differences among groups as appropriate.

After naturally grouping the subject population, an initial exploratory analysis is conducted to identify any imbalanced features between groups. Variables exhibiting a significance level of p ≤ 0.05 (indicating potential group imbalances) are selected as candidate confounding factors in subsequent univariate analysis. Moreover, certain variables such as age and gender, which are theoretically expected to act as confounders, are also taken into consideration. The variables of interest (e.g., BW, SUA) along with potential confounders are incorporated into a multivariate regression model for further analysis. The multivariate analysis included a test for multicollinearity, wherein tolerance values and variance inflation factor (VIF) were presented.

Multivariable logistic regression analysis were used to evaluate the association of weight and SUA with Aβ status, and ordinal logistic regression analysis were used to evaluate associations of clinical characteristics with AD pathology development (A-T-→A+T-→A+T+). Data were reported as adjusted ORs and 95% CIs. As exploratory post hoc analyses, comparisons of association between BW, SUA, and Aβ status (A+ and A-) across subgroups of interest (age, gender, and cognitive capacity) were performed using multivariable logistic regression analysis and data of p for interaction were presented.

The factors associated with cognitive function (MMSE scores) were analyzed using univariable linear regression analysis and further assessed using multivariable linear regression analyses across all participants and within pre-defined subgroups based on Aβ status (A+ and A-). Data were reported as adjusted Unstandardized Coefficient β and 95% CIs, and data of p for interaction were presented. A two-tailed p ≤ 0.05 was considered significant in all tests. Statistical analyses were performed using the SPSS version 23.0 software.

Results

Participants description

In the study, 139 participants aged 39 to 92 years were included, with 77.7% of them being 60 years or older. The participants’ characteristics are detailed in Table 1.

Table 1.

Demographic, clinical characteristics and biomarker data of the entire participants (n = 139).

Characteristics	Statistical description	N (%) with available data
Age (y)	66.9 ± 10.46	139 (100%)
Female, n (%)	69 (49.6%)	139 (100%)
Education (y)	9 (5, 13)	125 (89.9%)
MMSE scores	18.5 (12, 25)	124 (89.2%)
Dementia, n (%)	84 (67.7%)	124 (89.2%)
Weight (Kg)	59.2 ± 10.65	132 (95%)
APOE ε4 carriers, n (%)	36 (32.1%)	112 (80.6%)
Hypertension, n (%)	54 (38.8%)	139 (100%)
Diabetes, n (%)	28 (20.1%)	139 (100%)
Smoking, n (%)	25 (18%)	139 (100%)
Mean arterial pressure (mmHg)	95 (88.67, 103.67)	139 (100%)
Serum
HbA1c (%)	5.7 (5.4, 6.1)	137 (98.6%)
Total cholesterol (mmol/L)	4.79 (4.12, 5.56)	139 (100%)
LDL-cholesterol (mmol/L)	2.94 (2.49, 3.55)	139 (100%)
Uric acid (μmol/L)	347.71 ± 86.47	139 (100%)
Creatinine (μmol/L)	65 (55, 77)	138 (99.3%)
AD core biomarkers
CSF Aβ₄₂ (pg/mL)	571.45 (337.14, 976.44)	119 (85.6%)
CSF Aβ_42/40 ratio	0.08 (0.05, 0.13)	119 (85.6%)
CSF P-tau181(pg/mL)	48.09 (24.47, 109.24)	119 (85.6%)
CSF T-tau (pg/mL)	410.83 (192.53, 590.73)	119 (85.6%)
¹⁸F-AV45 PET positive, n (%)	15 (62.5%)	24 (17.3%)
¹⁸F-AV1451 PET positive, n (%)	4 (26.7%)	15 (10.8%)
Brain Imaging (MRA/CTA)
Fezakas scores	2 (2, 4)	138 (99.3%)
Cerebral arteriosclerosis, n (%)	78 (70.3%)	111 (79.9%)

Data are presented as mean ± SD, M (P₂₅, P₇₅) or n (%).

MMSE: Mini-Mental State Examination; APOE ε4: Apolipoprotein Ε4; HbA1c: glycated hemoglobin; LDL-cholesterol: low density lipoprotein cholesterol; CSF: cerebrospinal fluid; Aβ: amyloid-beta; P-tau181: phosphorylated-tau181; T-tau: total-tau; ¹⁸F-AV45: ¹⁸F-florbetapir; PET: positron emission tomography; ¹⁸F-AV1451: ¹⁸F-flortaucipir; MRA: magnetic resonance angiography; CTA: computed tomography angiography.

Differences in BW and SUA levels between A+ and A-

Table 2 displays the clinical characteristics of participants divided into two groups based on whether Aβ is positive (A+) or negative (A-). Significant differences were found in the dementia proportion, ratio of APOE ε4 carriers, weight, and SUA levels. Participants in group A+ showed a higher proportion of dementia (χ²= 5.483, P = 0.019) and APOE ε4 carriers (χ²= 5.472, p = 0.019) compared to group A-. Group A+ also had lighter weight (Mean Difference −6.8, 95%CI −10.6 to −3.01, t = −3.549, p = 0.001) and lower levels of SUA (Median Difference −38, 95%CI −69 to −5, z = −2.301, p = 0.021) (Figure 1(a) and (b)). Other variables did not show significant differences between the two groups.

Figure 1.

Student's t-test of the levels of BW levels (a) and Mann-Whitney U test of the levels of SUA (b) grouped by Aβ state (A- versus A+); adjusted odds ratios (OR) and 95% confidence intervals for the association of BW and SUA with Aβ state (c).

Table 2.

Differences of characteristics between groups classification by Aβ state.

Characteristic	A- (n = 46)	A+ (n = 84)	p
Age (y)	66.28 ± 10.1	66.42 ± 10.38	0.943
Female, n (%)	18 (39.1%)	45 (53.6%)	0.115
Dementia, n (%)	18 (51.4%)	59 (73.8%)	0.019
APOE ε4 carriers, n (%)	7 (17.9%)	26 (40.4%)	0.019
Weight (Kg)	64.1 ± 10.49	57.30 ± 10.02	0.001
Mean arterial pressure (mmHg)	95.5 (88.5, 107.33)	95 (89, 102.58)	0.644
Serum
HbA1c (%)	5.75 (5.4, 6.2)	5.6 (5.4, 6)	0.369
Total cholesterol (mmol/L)	4.59 (4.06, 5.39)	4.78 (4.12, 5.63)	0.548
LDL-cholesterol (mmol/L)	2.85 (2.43, 3.62)	3.00 (2.43, 3.6)	0.659
Uric acid (μmol/L)	371.5 (330.75, 411.25)	333.5 (272.5, 399)	0.021
Creatinine (μmol/L)	68.85 (59, 78.75)	63 (53, 77)	0.107
Brain Imaging (MRA/CTA)
Fezakas scores	3 (2, 4.25)	2 (2, 4)	0.266
Cerebral arteriosclerosis, n (%)	23 (69.7%)	48 (69.6%)	0.989

Data are presented as mean ± SD, M (P₂₅, P₇₅) or n (%).

Associations of BW and SUA with Aβ state (A+ versus A-)

After adjusting for potential confounders, a multivariate logistic regression analysis (Table 3, Figure 1(c)) revealed that lower weight level significantly influenced the presence of Aβ positive state (A+) (adjusted odds ratio = 0.919, 95%CI 0.865 to 0.977, p = 0.007), while SUA level did not (adjusted odds ratio = 0.996, 95%CI 0.990 to 1.003, p = 0.263). The test for multicollinearity showed no collinearity issues among these variables (tolerance values ≥0.760, VIF ≤ 1.462).

Table 3.

Logistic regression analysis to evaluate the association of weight and SUA with Aβ state.

Characteristic	Adjusted OR	95% CI	p
Age (y)	1.036	0.984–1.090	0.177
Female	1.110	0.353–3.488	0.858
Dementia	1.222	0.730–7.364	0.154
APOE ε4 carriers	2.318	0.427–3.500	0.709
Weight (Kg)	0.919	0.865–0.977	0.007
Uric acid (μmol/L)	0.996	0.990–1.003	0.263

Stratified association between BW, SUA and Aβ state (A+ versus A-) by age, gender, and cognitive capacity

The exploratory post hoc subgroup analyses by age, gender, and cognitive capacity were presented in Table 4. The relationship between BW and Aβ state was consistent across age subgroups, while the association of SUA with Aβ state remained stable across all three specified subgroups.

Table 4.

Stratified association between BW, SUA and Aβ state (A- versus A+) by age, gender and cognitive capacity.

BW (kg)
Variables	N (%)	Mean ± SD /M (P25, P75)		Adjusted OR (95% CI)	p	p_interaction
Variables	N (%)	A-	A+	Adjusted OR (95% CI)	p	p_interaction
Age, y						0.985
≤60	35 (27%)	66 (49.75, 71)	56 (50.75, 60.63)	0.997 (0.795–1.249)	0.977
>60	95 (73%)	64.48 ± 10.35	56.98 ± 9.53	0.915 (0.853–0.981)	0.012
Female						0.042
No	67 (52%)	67.5 (63.5, 73.13)	60 (56, 69)	0.833 (0.735–0.945)	0.004
Yes	63 (48%)	58.21 ± 11.62	53.08 ± 9.51	0.988 (0.923–1.057)	0.725
Dementia						0.009
No	38 (29%)	69.35 ± 7.73	59.48 ± 10.24	0.814 (0.816, 1.000)	0.050
Yes	77 (59%)	58.4 ± 10.05	56.42 ± 9.87	0.978 (0.908, 1.054)	0.560
All patients	130 (100%)	64.10 ± 10.49	57.30 ± 10.02	0.919 (0.865–0.977)	0.007

SUA (umol/L)
Variables	N (%)	Mean ± SD /M (P25, P75)		Adjusted OR (95% CI)	p	p_interaction
Variables	N (%)	A-	A+	Adjusted OR (95% CI)	p	p_interaction
Age, y						0.258
≤60	35 (27%)	389.9 ± 62.99	327.14 ± 92.99	0.973 (0.929–1.018)	0.230
>60	95 (73%)	366.5 (303, 417.75)	342.85 (281, 395.25)	0.998 (0.991–1.005)	0.553
Female						0.080
No	67 (52%)	370 (340.5, 421)	380 (299, 433)	1.002 (0.993–1.011)	0.622
Yes	63 (48%)	365.13 ± 77.65	311.43 ± 79.78	0.992 (0.983–1.000)	0.060
Dementia						0.937
No	38 (29%)	376 (339, 417)	357 (286, 415)	0.997 (0.983, 1.010)	0.635
Yes	77 (59%)	364.5 ± 68.88	332.55 ± 85.83	0.994 (0.984, 1.003)	0.196
All patients	130 (100%)	371.5 (330.75, 411.25)	333.5 (272.5, 399)	0.996 (0.990–1.003)	0.263

In the male and non-dementia subgroup, a significantly lower weight level was found to influence the presence of Aβ positive state (A+) (adjusted odds ratio = 0.833, 95%CI 0.735 to 0.945, p = 0.004; adjusted odds ratio = 0.814, 95%CI 0.816 to 1.000, p = 0.050), whereas no significant association was observed in the female (adjusted odds ratio = 0.988, 95%CI 0.923 to 1.057, p = 0.725) (P for interaction = 0.042) and dementia subgroup (adjusted odds ratio = 0.978, 95%CI 0.908 to 1.054, p = 0.560) (p for interaction = 0.009).

Differences in BW and SUA levels between groups divided by AT profiles

Statistically significant differences were found in dementia proportion (χ²= 15.856, p < 0.001), APOE ε4 carriers’ ratio (χ²= 9.07, p = 0.011), weight (F = 8.088, p = 0.001) and mean arterial pressure (MAP) levels (F = 4.942, p = 0.010) between AT profiles (Table 5). Lower weights were found in group A+T+ (mean 56.3, SD 8.47, p < 0.001) and A+T- (mean 58.79, SD 11.5, p = 0.021), compared with group A-T- (mean 65.09, SD 9.98) and reduced level of MAP was found in group A+T+ (mean 93.44, SD 8.79, p = 0.014), compared to A+T- (mean 99.93, SD 11.11) (Figure 2(a) and (b)). No significant differences were observed in SUA and other variables across AT profiles (Table 5, Figure 2(c)).

Figure 2.

Tukey's post hoc analysis of the levels of BW levels (a), the levels of MAP (b) and the levels of SUA (c) grouped by AT state (A-T- versus A+T- versus A+T+); adjusted odds ratios (OR) and 95% confidence intervals for the association of BW, MAP, and SUA with AT state (d).

Table 5.

Differences of characteristics between groups divided by AT profiles.

Characteristic	A-T- (n = 40)	A+T- (n = 38)	A+T+ (n = 45)	p
Age (y)	67.43 ± 9.35	67.71 ± 11.26	65.33 ± 9.68	0.496
Female, n (%)	14 (35%)	20 (52.6%)	24 (53.3%)	0.172
Dementia, n (%)	12 (41.4%)	22 (61.1%)	37 (86%)	<0.001
APOE ε4 carriers, n (%)	7 (21.2%)	7 (25%)	19 (52.8%)	0.011
Weight (Kg)	65.09 ± 9.98	58.79 ± 11.5	56.3 ± 8.47	0.001
Mean arterial pressure (mmHg)	98.17 ± 11.12	99.93 ± 11.11	93.44 ± 8.79	0.010
Serum
HbA1c (%)	5.85 (5.5, 6.5)	5.6 (5.4, 6)	5.7 (5.4, 6.05)	0.244
Total cholesterol (mmol/L)	4.59 (4.02, 5.4)	4.67 (4.09, 5.72)	4.79 (3.99, 5.58)	0.829
LDL-cholesterol (mmol/L)	2.89 (2.46, 3.65)	2.97 (2.42, 3.35)	2.99 (2.56, 3.65)	0.874
Uric acid (μmol/L)	366.5 (307.5, 417.75)	360 (270.5, 428)	317 (281, 393.5)	0.074
Creatinine (μmol/L)	68.85 (59, 82.85)	64.6 (50.5, 78)	62 (53, 76.5)	0.340
Brain Imaging (MRA/CTA)
Fezakas scores	3 (2, 5)	3 (2, 4)	2 (2, 3)	0.100
Cerebral arteriosclerosis, n (%)	21 (75%)	25 (78.1%)	23 (63.9%)	0.389

Data are presented as mean ± SD, M (P₂₅, P₇₅) or n (%).

Association of BW and SUA with AT profiles

An ordinal multivariate logistic regression analysis model showed that higher dementia proportion (adjusted odds ratio = 3.188 95%CI 1.194 to 8.510, p = 0.021), lighter weight (adjusted odds ratio = 0.939, 95%CI 0.891 to 0.990, p = 0.019) and lower MAP (adjusted odds ratio = 0.954, 95%CI 0.915 to 0.994, p = 0.025) were associated with AD pathological progress (A-T-→A+T-→A+T+) (Table 6, Figure 2(d)). SUA and other variables were not observed statistically significant. The test for multicollinearity showed no collinearity issues among these variables (tolerance values ≥0.719, VIF ≤ 1.469).

Table 6.

Ordinal logistic regression analysis to evaluate the association of weight and SUA with AT profiles (A-T-→A+T-→A+T+).

Characteristic	Adjusted OR	95% CI	p
Age (y)	0.985	0.941–1.032	0.524
Female	1.214	0.438–3.367	0.709
Dementia	3.188	1.194–8.510	0.021
APOE ε4 carriers	1.979	0.767–5.102	0.158
Weight (Kg)	0.939	0.891–0.990	0.019
Mean arterial pressure (mmHg)	0.954	0.915–0.994	0.025
Uric acid (μmol/L)	0.999	0.993–1.005	0.263

Association of BW and SUA with AT profiles

Univariable linear regression analysis revealed that female sex (Unstandardized Coefficient β = −3.435, 95% CI: −6.299 to −0.572, p = 0.019) and positive Aβ state (A+) (Unstandardized Coefficient β = −4.702, 95% CI: −7.869 to −1.534, p = 0.004) were associated with low MMSE score. High education level (Unstandardized Coefficient β = 0.799, 95% CI: 0.547 to 1.052, p < 0.001), high weight (Unstandardized Coefficient β = 0.293, 95% CI: 0.162 to 0.424, p < 0.001) (Figure 3(a)), and high SUA (Unstandardized Coefficient β = 0.025, 95% CI: 0.008 to 0.041, p = 0.004) (Figure 3(b)) were associated with high MMSE scores, while other factors did not show statistical significance. The association between each variable and MMSE scores was re-evaluated through univariable linear regression analysis within pre-defined subgroups based on Aβ status (A+ and A-), respectively.

Figure 3.

Univariable linear associations between SUA, BW and MMSE scores in total participants (a, b), adjusted unstandardized coefficient β and 95% confidence intervals for the association of education, BW, SUA, and creatinine with MMSE scores in total participants (c). b: unstandardized coefficient β; 95% CI: 95% confidence intervals.

The association of weight and SUA with MMSE scores in participants analyzed by multivariable linear regression are shown in Table 7. In model 1 including variables of age, sex, Aβ state, education years, weight, SUA and creatinine, high education level (p < 0.001), high weight (p = 0.010), and high SUA (p = 0.036) were associated with high MMSE scores; however, high serum creatinine (p = 0.003) was associated with low MMSE score, in all participants regardless of Aβ state (Figure 3(c)). Model 2 and model 3 showed the multivariable linear regression analysis to evaluate the association of weight and SUA with MMSE scores among participants stratified into negative Aβ state (A-) and positive Aβ state (A+). In exploratory analyses, the relationship between BW, SUA, and MMSE scores were consistent across pre-defined Aβ state subgroups (A- and A+) (Table 8). The tests for multicollinearity showed no collinearity issues among these variables in model 1(tolerance values ≥0.518, VIF ≤ 1.930), model 2 (tolerance values ≥0.615, VIF ≤ 1.625) and model 3 (tolerance values ≥0.501, VIF ≤ 1.995).

Table 7.

Multivariable linear regression analysis to evaluate the association of weight and SUA with MMSE scores.

Characteristic	Adjusted UnstandardizedCoefficient β	95% CI	p
Model 1
Age (y)	−0.024	(−0.146, 0.098)	0.694
Female	−1.951	(−5.197, 1.295)	0.236
Positive Aβ state (A+)	−0.979	(−3.813, 1.855)	0.495
Education (y)	0.677	(0.425, 0.929)	<0.001
Weight (Kg)	0.189	(0.047, 0.332)	0.010
Uric acid (μmol/L)	0.019	(0.001, 0.037)	0.036
Creatinine (μmol/L)	−0.127	(−0.210, −0.044)	0.003
Model 2
Age (y)	−0.022	(−0.217, 0.172)	0.817
Female	−1.559	(−6.927, 3.809)	0.556
Education (y)	0.542	(0.066, 1.018)	0.027
Weight (Kg)	0.255	(0.000, 0.511)	0.050
Uric acid (μmol/L)	−0.004	(−0.032, 0.024)	0.764
Total cholesterol (mmol/L)	−1.604	(−3.509, 0.301)	0.095
Model 3
Age (y)	−0.009	(−0.176, 0.158)	0.913
Female	−1.022	(−5.191, 3.148)	0.626
Education (y)	0.722	(0.408, 1.037)	<0.001
Weight (Kg)	0.134	(−0.047, 0.315)	0.145
HbA1c (%)	−0.853	(−2.432, 0.726)	0.285
Uric acid (μmol/L)	0.026	(0.008, 0.041)	0.018
Creatinine (μmol/L)	−0.131	(−0.288, −0.034)	0.009

Model 1 Multivariable linear regression analysis to evaluate the association of weight and SUA with MMSE scores in all participants regardless of Aβ state.

Model 2 Multivariable linear regression analysis to evaluate the association of weight and SUA with MMSE scores in participants of negative Aβ state (A-).

Model 3 Multivariable linear regression analysis to evaluate the association of weight and SUA with MMSE scores in participants of positive Aβ state (A+).

Table 8.

Stratified association between body weight, SUA and MMSE scores by Aβ state.

BW (kg)
Variables	N (%)	Adjusted UnstandardizedCoefficient β	95% CI	p	p_interaction
Aβ state					0.172
A-	35 (28%)	0.255	(0.000, 0.511)	0.050
A+	80 (65%)	0.134	(−0.047, 0.315)	0.145
All patients	124 (100%)	0.189	(0.047, 0.332)	0.010

SUA
Variables	N (%)	Adjusted UnstandardizedCoefficient β	95% CI	p	p_interaction
Aβ state					0.391
A-	35 (28%)	−0.004	(−0.032, 0.024)	0.764
A+	80 (65%)	0.026	(0.008, 0.041)	0.018
All patients	124 (100%)	0.019	(0.001, 0.037)	0.036

Discussion

In the present study, we found that lower BW was associated with aggregated amyloid pathology. Furthermore, individuals with A+T+ biomarker profile exhibited lower BW than those with A+T- and A-T- profiles. A progressive decline in BW was along with AD pathological deteriorated progress (A-T-→A+T-→A+T+). Aβ positive group (A+) exhibited lower SUA level compared to the Aβ negative group (A-), with a gradual downtrend observed from A-T- to A+T- and further to A+T+. However, upon adjusting for confounding factors including BW, no significant association was found between SUA levels and aggregated amyloid pathology or the progression of AD pathology.

Weight reduction is common among individuals diagnosed with AD,^32,33 and it has been associated with the progression and severity of the disease.³⁴ A longitudinal clinical-pathological study of 298 brain autopsies conducted had demonstrated BMI was negatively associated with the extent of AD pathology in elderly individuals with and without dementia, but was not related to the presence of cerebral infarctions or Lewy body disease.³⁵ Subsequent a series of cross-section and longitudinal studies had exhibited in vivo lower BMI was associated with markers of increased AD burden (i.e., CSF Aβ, tau, or PiB PET image) among elderly healthy/no dementia, mild cognitive impairment (MCI) and AD groups.^{20–22,36–38} The results of our current study corroborate previous findings, and we introduce a novel observation that, in vivo, through the ordered multi-classification regression, BW gradually decreases with the pathological progress of AT profile, which further confirms the postmortem pathological studies of AD (which is related to the progression and severity of the disease),³⁵ to our knowledge the relevant data are scarce in the literature to date.

Intriguingly, we noticed a concurrent decline in MAP, which was independently correlated with the progression of AD pathology. Hypertension occurring during midlife is a risk factor for cognitive decline and dementia in older adults.³⁹ Longitudinal evidence also indicates that individuals who develop or have hypertension in early adulthood (median age of 28.9 years) were associated with whole brain volume and hippocampus volume reduction (AD and related dementias pathology) in later life (median age of 74.8 years).⁴⁰ However, in elderly individuals with frailty, higher BP is correlated with improved cognitive function.⁴¹ In a post-hoc analysis of Systolic Blood Pressure Intervention Trial, the prevalence of “probable dementia” and MCI was higher in hypertensive patients with frailty, and intensive BP reduction is likely to strengthen this association.⁴² Our data consisted mostly of the elderly, with an average age of 66.9 ± 10.46 years old, which were in line with the above findings. The following mechanisms may account for our findings: Firstly, the reduction in MAP, along with weight loss, serves as a marker of “frailty” in the elderly population and is independently associated with the pathological progression of AD. Secondly, atherosclerosis is prevalent in the elderly, and the decline in MAP against this backdrop in elderly persons with frailty may indicate an exacerbated reduction in cerebral perfusion, further contributing to the advancement of AD pathology.^43,44

In post-hoc subgroup analysis, we found that lower BW was associated with aggregated amyloid pathology in the male and non-dementia population, but not in females or individuals with dementia. This discrepancy may be attributed to the higher proportion of muscle (non-fat) in men compared to women, who tend to have a higher percentage of body fat. Estrogen exerts a neuroprotective effect,⁴⁵ and the presence of adipose tissue enables endogenous estrogen supply to postmenopausal women, thereby partially attenuating the impact of low body weight on Aβ deposition. Moreover, muscle loss is considered a characteristic feature of frailty.⁴¹ Low BW primarily reflects muscle loss and weakness in males, which elucidates the more pronounced association between low BW and Aβ deposition. These findings further substantiate the link between muscle loss, weakness, and the progression of AD pathology. The increase in Aβ deposition is known to be frequently along with a decline in cognitive function.²⁴ Our subgroup analysis suggests that low BW has a stronger association with Aβ deposition in the non-dementia population compared to the dementia population, indicating that weight loss may play a promoting role during the early development of AD-related dementia, specifically during the period of Aβ deposition.

The literature on the relationship between UA and AD has been conflicting. Due to its potent antioxidant properties, UA has been demonstrated to exert a neuroprotective role in neurodegenerative diseases, such as Parkinson's disease.⁴⁶ Huang et al. reported that SUA levels exhibit a biphasic pattern during the entire AD development process (CU-MCI-dementia), increasing initially and then decreasing, reaching a peak at the MCI stage, and they speculated that the potential neuroprotective effects of SUA predominantly impact Aβ₄₂ and the subsequent pathological cascade.²³ And in a population-based prospective cohort study, BMI and SUA levels were independently associated with a reduced risk of dementia (including AD and VaD).¹⁹ However, Li et al. identified SUA as a potential risk factor for AD, demonstrating a strong correlation with CSF AD biomarkers in the preclinical stage of AD which suggested that elevated SUA levels may lead to disruptions in brain Aβ metabolism, thereby increasing the risk of AD.¹⁰ It is worth noting that the study by Huang et al.²³ considered the progression of AD biomarkers and cognitive impairment as outcome factor together, and the study by Li et al.¹⁰ used the clinical diagnostic criteria for AD that are not considered biomarkers. These approaches could potentially obscure the independent impacts of UA, BW, and other factors on AD biomarkers and cognitive impairment to some extent. Our findings revealed a downward trend in UA levels concurrent with both Aβ deposition and AD pathologic progression; however, the regression was not statistically significant after adjusting for confounding factors (including weight and dementia). This suggests that the decline in UA levels may be a concurrent phenomenon in the context of weight loss and cognitive decline in the elderly population, and UA does not appear to be independently associated with AD pathology. The findings of our research indicate that UA cannot be regarded as having either a protective or exacerbating influence on the pathological progression of AD, that contradicts some prior studies. The potential explanation for this contradiction lies in the potential confounding effect of other metabolic factors or the impact of chronic inflammation.⁴⁷ Previous research employing a bidirectional Mendelian randomization approach have suggested elevated SUA is an outcome rather than a causal factor of increased adiposity.¹⁸ And as well known, inadequate protein and fat intake can result in weight loss, which is associated with a decrease in UA production. Alternatively, chronic inflammation may exist, leading to alterations in UA levels indirectly. It is important to acknowledge that UA should not be regarded in isolation as a sole influencing factor.

The pathological progression of AD is not invariably accompanied by cognitive decline among older adults. Approximately 30 to 40 percent of cognitively unimpaired individuals exhibit AD neuropathological changes at autopsy^48,49 or in vivo Aβ biomarker abnormalities.⁵⁰ AD-derived cognitive impairment is only one type of cognitive-related disease. Therefore, in the current study, we also analyzed the associations between BW, UA levels with cognitive function in diverse populations, including the general population, individuals in the Aβ+ state, and those in the Aβ- state.

We discovered a positive correlation between BW, UA and cognition, as well as a negative correlation between serum creatinine and cognition, within the general population, irrespective of Aβ status. These findings were in line with previous research.^19,51 The relationship between BW, SUA, and MMSE scores remained consistent across pre-defined subgroups based on Aβ state in exploratory analyses. The findings suggest that reductions in BW and SUA are linked to cognitive decline, regardless of whether the decline is caused by AD or non-AD sources. This implies that BW and SUA may independently affect cognitive function, regardless of the presence of Aβ deposition.

The potential explanation for this phenomenon is that, given Aβ- and Aβ+ represent a grade cut-off, an Aβ- state does not necessarily imply the absence of an Aβ+ in the future. This is because the Aβ deposit is a continuous accumulation process.⁵² Low-weight individuals are accompanied by the worsening of Aβ deposition (from A- to A+) and further increased risk of cognitive impairment due to possible higher risk of weakness and difficulty in instrumental activities of daily life.⁵³ The increase of individual cognitive impairment is bound to affect the intake of calories and nutrients, resulting in further weight loss. Although SUA has no significant correlation with the pathological development of Aβ deposition, its correlation with cognitive function retention may be due to its powerful anti-inflammatory and antioxidant functions, which protect cognitive function based on the protective effect on nerve cells themselves, rather than by reducing the deposition of Aβ. Moreover, multiple studies have demonstrated a positive association between UA and cognitive abilities. Due to the structural similarity with brain-active substances like caffeine and theobromine, UA is postulated to potentially stimulate the positive response in the cerebral cortex. Consequently, as early as in 1955, Orowan proposed a hypothesis suggesting that the evolutionary inactivation of the uricase gene increases UA levels, leading to the evolution of intelligence in ancient primates.⁵⁴ The existence of intriguing studies has been demonstrated the incidence of gout appears to be higher in populations with high intelligence compared to the general population.⁵⁵

The present study may contribute to understand the inconsistencies of previous research concerning the associations between weight, UA, and dementia. Furthermore, it elucidated the relationships between weight and UA with AD biomarkers and cognitive function, respectively. Since measuring BW is easy to obtain in clinical settings, combined with other factors such as SUA, BP, and kidney function, may offer a valuable information to the physicians to rapidly predict Aβ deposition status and cognitive function at initial diagnosis, thereby facilitating the differentiation of dementia subtypes, in patients with cognitive complaints. In clinical practice, ensuring appropriate and sufficient nutrition and calorie intake, as well as avoiding low body weight, is crucial for the prevention of AD pathology in middle-aged and elderly individuals at risk of dementia. Furthermore, maintaining optimal UA concentration levels and avoiding deficiencies can significantly contribute to the preservation of cognitive function.

This study has several limitations. Firstly, this cross-sectional study confined findings to the correlation between BW, UA and AD biomarkers, cognitive functions, which might not provide a comprehensive causal inference between them. Second, these data were collected from a distinct subset of individuals seeking medical intervention at our hospital, which may affect the study's generalizability, and the modest sample size may reduce the efficiency of the assessment and augmented the likelihood of unstable results. Data collection was conducted during outpatient or initial admission visits, with most patients being assessed prior to receiving systematic medical intervention treatment. However, a limited number of individuals may have received medical interventions for cognitive impairment before assessment, such as cognition-enhancing medications (e.g., cholinesterase inhibitors and memantine) that can potentially cause gastrointestinal reactions, lifestyle modifications including increased physical exercise, or dietary adjustments involving the reduction of high-calorie diets. These medical interventions may have exerted an influence on the patient's weight and UA levels, potentially introducing slight confounding factors to the study outcomes. Thirdly, caution should be exercised when interpreting the results of subgroup analyses, considering their potential deviation from randomness, limited sample size, inadequate statistical power, or increased risk of false positives. These findings should be used solely for reference purposes. Finally, we discussed BW in adults, rather than BMI, including the effect of height, which in some ways also reflects differences in genetics, early nutritional status, socioeconomic conditions,⁵⁶ and other factors, impacting the association with AD. Weight is not a specific indicator of body composition, and it does not adequately represent obesity. Studies have shown that the ratio of non-fat tissue to height begins to decline after the age of 45, particularly in women.⁵⁷ Sarcopenia and frailty in old age have been shown to be positively associated with the progression of AD and dementia.^41,58 Considering these aforementioned limitations, it is still necessary to conduct large-scale and prospective studies to further explore the correlation between various body composition measures (e.g., body fat percentage, waist circumference, waist-to-hip ratio, subcutaneous fat thickness, muscle content, etc.), UA, and AD biomarkers or cognitive disorders in adults and derive the exact causal relations between them.

In summary, in individuals with cognitive complaints during midlife and late-life stages, a lower BW was found to be associated with Aβ deposition, while the progressive decline in both BW and MAP independently correlated with the progression of AD pathology. However, SUA has not been demonstrated such significance. More intriguing finding is that, both lower BW and SUA levels were found to be associated with cognitive decline within the population, regardless of the presence or absence of Aβ deposition. Our study suggest that weight loss may promote the accumulation of Aβ and result in cognitive impairment, whereas SUA may play a significant role in protecting cognition regardless of whether it is related to AD or non-AD cognitive impairment.

Footnotes

Acknowledgments

The authors have no acknowledgments to report.

ORCID iDs

Qing Tian

Qing Dong

Zhumin Su

Yingying Liu

Lili Ma

Huimin Dong

Yiru Xu

Zhan Ma

Xiaohong Chen

Xiaomeng Ma

Author contributions

Qing Tian (Data curation; Formal analysis; Investigation; Methodology; Project administration; Writing – original draft; Writing – review & editing); Qing Dong (Data curation; Formal analysis; Methodology; Project administration; Writing – original draft; Writing – review & editing); Zhumin Su (Software); Yingying Liu (Methodology; Resources; Software; Validation); Lili Ma (Investigation; Resources; Writing – review & editing); Huimin Dong (Investigation; Methodology; Software; Writing – review & editing); Yiru Xu (Funding acquisition; Writing – review & editing); Zhan Ma (Investigation; Project administration; Resources); Xiaohong Chen (Conceptualization; Data curation; Resources; Supervision; Writing – review & editing); Xiaomeng Ma (Conceptualization; Data curation; Investigation; Resources; Supervision; Writing – review & editing)

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Social Science Foundation of China (22BYY078 and 21&ZD294).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

The authors confirm that the data supporting the findings of this study are available within the manuscript.

References

Alzheimer’s Association. 2023 Alzheimer's disease facts and figures. Alzheimers Dement 2023; 19: 1598–1695.

Livingston

Huntley

Sommerlad

, et al. Dementia prevention, intervention, and care: 2020 report of the lancet commission. Lancet 2020; 396: 413–446.

Whitmer

Gustafson

Barrett-Connor

, et al. Central obesity and increased risk of dementia more than three decades later. Neurology 2008; 71: 1057–1064.

Pedditzi

Peters

Beckett

. The risk of overweight/obesity in mid-life and late life for the development of dementia: a systematic review and meta-analysis of longitudinal studies. Age Ageing 2016; 45: 14–21.

Liu

Ang

, et al. BMI decline patterns and relation to dementia risk across four decades of follow-up in the Framingham study. Alzheimers Dement 2023; 19: 2520–2527.

Qizilbash

Gregson

Johnson

, et al. BMI and risk of dementia in two million people over two decades: a retrospective cohort study. Lancet Diabetes Endocrinol 2015; 3: 431–436.

Ajnakina

Steptoe

, et al. Higher risk of dementia in English older individuals who are overweight or obese. Int J Epidemiol 2020; 49: 1353–1365.

Ames

Cathcart

Schwiers

, et al. Uric acid provides an antioxidant defense in humans against oxidant- and radical-caused aging and cancer: a hypothesis. Proc Natl Acad Sci U S A 1981; 78: 6858–6862.

Squadrito

Cueto

Splenser

, et al. Reaction of uric acid with peroxynitrite and implications for the mechanism of neuroprotection by uric acid. Arch Biochem Biophys 2000; 376: 333–337.

10.

, et al. Serum uric acid may aggravate Alzheimer's disease risk by affecting amyloidosis in cognitively intact older adults: the CABLE study. J Alzheimers Dis 2021; 81: 389–401.

11.

Desideri

Gentile

Antonosante

, et al. Uric acid amplifies Abeta amyloid effects involved in the cognitive dysfunction/dementia: evidences from an experimental model in vitro. J Cell Physiol 2017; 232: 1069–1078.

12.

Desideri

Castaldo

Lombardi

, et al.

Is it time to revise the normal range of serum uric acid levels?

Eur Rev Med Pharmacol Sci 2014; 18: 1295–1306.

13.

Wen

Arakawa

Tamai

. Uric acid in health and disease: from physiological functions to pathogenic mechanisms. Pharmacol Ther 2024; 256: 108615.

14.

Vannorsdall

Jinnah

Gordon

, et al. Cerebral ischemia mediates the effect of serum uric acid on cognitive function. Stroke 2008; 39: 3418–3420.

15.

Euser

Hofman

Westendorp

, et al. Serum uric acid and cognitive function and dementia. Brain 2009; 132: 377–382.

16.

Scheepers

Jacobsson

Kern

, et al. Urate and risk of Alzheimer's disease and vascular dementia: a population-based study. Alzheimers Dement 2019; 15: 754–763.

17.

Latourte

Soumare

Bardin

, et al. Uric acid and incident dementia over 12 years of follow-up: a population-based cohort study. Ann Rheum Dis 2018; 77: 328–335.

18.

Lyngdoh

Vuistiner

Marques-Vidal

, et al. Serum uric acid and adiposity: deciphering causality using a bidirectional Mendelian randomization approach. PLoS One 2012; 7: e39321.

19.

Cao

Yang

, et al. Associations of BMI and serum urate with developing dementia: a prospective cohort study. J Clin Endocrinol Metab 2020; 105: dgaa638.

20.

Vidoni

Townley

Honea

, et al. Alzheimer disease biomarkers are associated with body mass index. Neurology 2011; 77: 1913–1920.

21.

Hsu

Mormino

Schultz

, et al. Lower late-life body-mass index is associated with higher cortical amyloid burden in clinically normal elderly. J Alzheimers Dis 2016; 53: 1097–1105.

22.

Rabin

Shirzadi

Swardfager

, et al. Amyloid-beta burden predicts prospective decline in body mass index in clinically normal adults. Neurobiol Aging 2020; 93: 124–130.

23.

Huang

Wang

Fan

, et al. The association of serum uric acid with cognitive impairment and ATN biomarkers. Front Aging Neurosci 2022; 14: 943380.

24.

Jack

Bennett

Blennow

, et al. NIA-AA Research framework: toward a biological definition of Alzheimer's disease. Alzheimers Dement 2018; 14: 535–562.

25.

Fazekas

Chawluk

Alavi

, et al. MR Signal abnormalities at 1.5T in Alzheimer's dementia and normal aging. AJR Am J Roentgenol 1987; 149: 351–356.

26.

Folstein

McHugh

. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189–198.

27.

Shaw

Vanderstichele

Knapik-Czajka

, et al. Cerebrospinal fluid biomarker signature in Alzheimer's disease neuroimaging initiative subjects. Ann Neurol 2009; 65: 403–413.

28.

Mattsson

Andreasson

Persson

, et al. The Alzheimer's association external quality control program for cerebrospinal fluid biomarkers. Alzheimers Dement 2011; 7: 386–395.

29.

Hansson

Batrla

Brix

, et al. The Alzheimer's association international guidelines for handling of cerebrospinal fluid for routine clinical measurements of amyloid beta and tau. Alzheimers Dement 2021; 17: 1575–1582.

30.

Blennow

Hampel

. CSF Markers for incipient Alzheimer's disease. Lancet Neurol 2003; 2: 605–613.

31.

Landau

Joshi

, et al. Comparing positron emission tomography imaging and cerebrospinal fluid measurements of beta-amyloid. Ann Neurol 2013; 74: 826–836.

32.

White

Pieper

Schmader

, et al. Weight change in Alzheimer's disease. J Am Geriatr Soc 1996; 44: 265–272.

33.

Cronin-Stubbs

Beckett

Scherr

, et al. Weight loss in people with Alzheimer's disease: a prospective population based analysis. BMJ 1997; 314: 178–179.

34.

White

Pieper

Schmader

. The association of weight change in Alzheimer's disease with severity of disease and mortality: a longitudinal analysis. J Am Geriatr Soc 1998; 46: 1223–1227.

35.

Buchman

Schneider

Wilson

, et al. Body mass index in older persons is associated with Alzheimer disease pathology. Neurology 2006; 67: 1949–1954.

36.

Ewers

Schmitz

Hansson

, et al. Body mass index is associated with biological CSF markers of core brain pathology of Alzheimer's disease. Neurobiol Aging 2012; 33: 1599–1608.

37.

Sun

Tan

, et al. Weight loss is a preclinical signal of cerebral amyloid deposition and could predict cognitive impairment in elderly adults. J Alzheimers Dis 2020; 77: 449–456.

38.

Grau-Rivera

Navalpotro-Gomez

Sanchez-Benavides

, et al. Association of weight change with cerebrospinal fluid biomarkers and amyloid positron emission tomography in preclinical Alzheimer's disease. Alzheimers Res Ther 2021; 13: 46.

39.

Ninomiya

Ohara

Hirakawa

, et al. Midlife and late-life blood pressure and dementia in Japanese elderly: the Hisayama study. Hypertension 2011; 58: 22–28.

40.

George

Maillard

Gilsanz

, et al. Association of early adulthood hypertension and blood pressure change with late-life neuroimaging biomarkers. JAMA Netw Open 2023; 6: e236431.

41.

Rogers

Steptoe

Cadar

. Frailty is an independent predictor of incident dementia: evidence from the English longitudinal study of ageing. Sci Rep 2017; 7: 15746.

42.

Wang

Pei

, et al. Association between the frailty and cognitive impairment among patients with hypertension-A post hoc analysis of the SPRINT trial. J Am Heart Assoc 2023; 12: e28736.

43.

Wolters

Zonneveld

Hofman

, et al. Cerebral perfusion and the risk of dementia: a population-based study. Circulation 2017; 136: 719–728.

44.

Wanleenuwat

Iwanowski

Kozubski

. Alzheimer's dementia: pathogenesis and impact of cardiovascular risk factors on cognitive decline. Postgrad Med 2019; 131: 415–422.

45.

Rahman

Jackson

Hristov

, et al. Sex and gender driven modifiers of Alzheimer's: the role for estrogenic control across age, race, medical, and lifestyle risks. Front Aging Neurosci 2019; 11: 315.

46.

Paganoni

Schwarzschild

. Urate as a marker of risk and progression of neurodegenerative disease. Neurotherapeutics 2017; 14: 148–153.

47.

Dalbeth

Phipps-Green

House

, et al. Body mass index modulates the relationship of sugar-sweetened beverage intake with serum urate concentrations and gout. Arthritis Res Ther 2015; 17: 263.

48.

Knopman

Parisi

Salviati

, et al. Neuropathology of cognitively normal elderly. J Neuropathol Exp Neurol 2003; 62: 1087–1095.

49.

Bennett

Schneider

Arvanitakis

, et al. Neuropathology of older persons without cognitive impairment from two community-based studies. Neurology 2006; 66: 1837–1844.

50.

Jansen

Ossenkoppele

Knol

, et al. Prevalence of cerebral amyloid pathology in persons without dementia: a meta-analysis. JAMA 2015; 313: 1924–1938.

51.

Garcia-Ptacek

Trevisan

, et al. Kidney function, kidney function decline, and the risk of dementia in older adults: a registry-based study. Neurology 2021; 96: e2956–e2965.

52.

Villemagne

Pike

Chetelat

, et al. Longitudinal assessment of Abeta and cognition in aging and Alzheimer disease. Ann Neurol 2011; 69: 181–192.

53.

Ren

, et al. Associations of body mass index, waist circumference and waist-to-height ratio with cognitive impairment among Chinese older adults: based on the CLHLS. J Affect Disord 2021; 295: 463–470.

54.

Orowan

. The origin of man. Nature 1955; 175: 683–684.

55.

Sofaer

Emery

. Genes for super-intelligence? J Med Genet 1981; 18: 410–413.

56.

Tyrrell

Jones

Beaumont

, et al. Height, body mass index, and socioeconomic status: Mendelian randomisation study in UK Biobank. BMJ 2016; 352: i582.

57.

Barlett

Puhl

Hodgson

, et al. Fat-free mass in relation to stature: ratios of fat-free mass to height in children, adults, and elderly subjects. Am J Clin Nutr 1991; 53: 1112–1116.

58.

Beeri

Leugrans

Delbono

, et al. Sarcopenia is associated with incident Alzheimer's dementia, mild cognitive impairment, and cognitive decline. J Am Geriatr Soc 2021; 69: 1826–1835.

Association of body weight and serum uric acid with Alzheimer's disease biomarkers and cognitive impairment

Abstract

Background

Objective

Methods

Results

Conclusions

Keywords

Introduction

Methods

Ethics approval

Participants

Clinical data collection and cognitive assessment

CSF biomarker testing and cut-off definitions

18F-florbetapir PET and 18F-flortaucipir PET images collection and processing

AT profile classification definitions

Statistical analyses

Results

Participants description

Differences in BW and SUA levels between A+ and A-

Associations of BW and SUA with Aβ state (A+ versus A-)

Stratified association between BW, SUA and Aβ state (A+ versus A-) by age, gender, and cognitive capacity

Differences in BW and SUA levels between groups divided by AT profiles

Association of BW and SUA with AT profiles

Association of BW and SUA with AT profiles

Discussion

Footnotes

Acknowledgments

ORCID iDs

Author contributions

Funding

Declaration of conflicting interests

Data availability

References

¹⁸F-florbetapir PET and ¹⁸F-flortaucipir PET images collection and processing