Association Between Non-alcoholic Fatty Liver Disease and the Risk of Dementia

A Nationwide Cohort Study

Gi-Ae Kim; Chi Hyuk Oh; Jung Wook Kim; Su Jin Jeong; In-Hwan Oh; Jin San Lee; Key-Chung Park; Jae-Jun Shim

Liver International. 2022;42(5):1027-1036.

Results

Baseline Characteristics of the Study Subjects

The study population comprised 4 031 948 subjects who met the inclusion criteria (Figure 1). The baseline characteristics of the study subjects are presented in Table 1 and Table S2. In our study population, 31.3%, 50.2% and 18.4% of the subjects were in the NAFLD, intermediate and non-NAFLD groups respectively (Table 1). NAFLD subjects were older (mean age, 53.7 vs. 53.6, p < .001) and more likely to be diabetic (10.4% vs. 1.1%, p < .001) or hypertensive (20.6% vs. 6.7%, p < .001), to be disabled (6.2% vs. 5.6%, p < .001), to be residing in a rural area (56.1% vs. 53.4%, p < .001), and to have higher serum ALT (28 vs. 16 U/L, p < .001) and GGT (32 vs. 19 U/L, p < .001) than non-NAFLD subjects.

Incidence Rate of Dementia Among Groups

During the observation period of 38 302 553 person-years (median 9.5 years), a total of 414 851 subjects developed dementia (annual incidence, 10.83/1000 PY), and 160 680 subjects died (annual incidence, 4.20/1000 PY).

The annual incidence of dementia was higher in the NAFLD group (11.55/1000 PY) than in the intermediate (10.71/1000 PY) and non-NAFLD groups (9.95/1000 PY; p < .001; Table 2).

Association Between NAFLD and Development of Dementia

The risk of dementia among the NAFLD and intermediate groups was calculated using the non-NAFLD group as a reference (Table 2). The non-adjusted HR for dementia in the NAFLD group (HR, 1.16; 95% CI, 1.15–1.17; p < .001) was significantly higher than that in the intermediate (HR, 1.08; 95% CI, 1.07–1.08; p < .001) and non-NAFLD groups. When clinical variables were adjusted in model 2, the NAFLD group had a significantly higher risk of developing dementia (HR, 1.05; 95% CI, 1.04–1.06; p < .001). When the competing risk of death was considered in model 3, the NAFLD group was associated with the highest risk of dementia (HR, 1.08; 95% CI, 1.07–1.09; p < .001; Table 2).

When stratified by sex, a total of 245 782 females (13.7%; annual incidence, 14.53/1000 PY) and 169 069 males (7.5%; annual incidence, 7.90/1000 PY) developed dementia, while 39 473 females (2.3%) and 121 207 males (5.7%) died. NAFLD male subjects showed a risk of dementia that was comparable to that of non-NAFLD male subjects in model 2 (HR, 0.99; 95% CI, 0.97–1.00; p = .09; Table 3). When the competing risk of death was considered in model 3, NAFLD male subjects showed a higher risk of developing dementia than non-NAFLD male subjects (HR, 1.02; 95% CI, 1.01–1.04; p = .001). As for NAFLD female subjects, they showed a higher risk of developing dementia than non-NAFLD female subjects in both model 2 (HR, 1.15; 95% CI, 1.14–1.17; p < .001) and model 3 (HR, 1.16; 95% CI, 1.15–1.18; p < .001; Table 3).

When the analysis was performed without excluding stroke at baseline, consistent results were observed. NAFLD was associated with an increased risk of dementia in model 1 (HR, 1.24; 95% CI, 1.23–1.25; p < .001), model 2 (HR, 1.03; 95% CI, 1.02–1.04; p < .001) and model 3 (HR, 1.14; 95% CI, 1.12–1.15; p < .001; Table S3).

Propensity Score-matched Analysis

Propensity score-matching analysis generated 610 596 pairs of the NAFLD and non-NAFLD groups, and the characteristics of the pairs were balanced with a standardized difference of <10% for variables (Table S4). In the propensity score-matched cohort, the NAFLD group showed a higher risk of dementia than the non-NAFLD group (HR, 1.09; 95% CI, 1.08–1.10; p < .001).

Subgroup Analysis

To further investigate the association of NAFLD with dementia risk, the NAFLD group was subdivided into NAFLD group 1 and NAFLD group 2 (Table S5). NAFLD group 1, whose HSI was greater than 36 at all health check-ups, was associated with the highest risk of dementia in model 1 (HR, 1.29; 95% CI, 1.27–1.30; p < .001), model 2 (HR, 1.06; 95% CI, 1.04–1.07; p < .001) and model 3 (HR, 1.08; 95% CI, 1.07–1.10; p < .001; Table 2). When stratified by sex, NAFLD group 1 had the highest risk of dementia both in males (HR, 1.04; 95% CI, 1.02–1.06; p < .001) and females (HR, 1.17; 95% CI, 1.15–1.18; p < .001; Table 3) in model 3.

Association Between NAFLD and Dementia Considering BMI

Dementia risk was analysed based on the BMI of the subjects. When subjects were categorized as non-obese (BMI <25 kg/m²) and obese (BMI ≥25 kg/m²), non-obese subjects with NAFLD showed a greater association between NAFLD and dementia (HR, 1.09; 95% CI, 1.07–1.10; p < .001) than obese subjects with NAFLD (HR 0.98; 95% CI, 0.88–1.09; p = .69; Table S6).

1 2 3 4

Next Section

Table 1. Baseline characteristics of the patients
	Non-NAFLD	NAFLD	Intermediate	p
Number of patients	742 934	1 263 376	2 025 638
Age, mean ± SD, years	53.6 ± 8.4	53.7 ± 7.8	53.9 ± 7.9	<.001
Male sex, n (%)	436 441 (58.7%)	729 204 (57.7%)	1 074 368 (53.0%)	<.001
SBP, mmHg, median (IQR)	120 (110, 130)	130 (119, 138)	121 (111, 132)	<.001
DBP, mmHg, median (IQR)	75 (70, 80)	80 (72, 87)	80 (70, 83)	<.001
BMI, kg/m²				<.001
<18.5 kg/m², n (%)	64 599 (8.7%)	822 (0.1%)	6269 (0.3%)
18.5–22.9 kg/m², n (%)	614 588 (82.7%)	86 624 (6.9%)	724 530 (35.8%)
23.0–24.9 kg/m², n (%)	60 761 (8.2%)	255 284 (20.2%)	818 182 (40.4%)
≥25 kg/m², n (%)	2986 (0.4%)	920 646 (72.9%)	476 657 (23.5%)
Diabetes mellitus, n (%)	8300 (1.1%)	131 299 (10.4%)	72 957 (3.6%)	<.001
Hypertension, n (%)	50 051 (6.7%)	260 701 (20.6%)	255 509 (12.6%)	<.001
Disabled person, n (%)	41 561 (5.6%)	78 662 (6.2%)	108 027 (5.3%)	<.001
Rural area^a, n (%)	396 448 (53.4%)	709 382 (56.1%)	1 083 619 (53.5%)	<.001
Income^b, n (%)				<.001
Quintile 1, lowest	113 411 (15.4%)	185 810 (14.9%)	304 659 (15.2%)
Quintile 2	95 352 (13.0%)	142 077 (11.4%)	241 455 (12.1%)
Quintile 3	122 212 (16.6%)	196 677 (15.8%)	310 308 (15.5%)
Quintile 4	152 742 (20.8%)	277 871 (22.3%)	421 120 (21.1%)
Quintile 5, highest	251 876 (34.2%)	444 630 (35.7%)	722 903 (36.1%)
Laboratory parameters
AST, IU/ml, median (IQR)	23 (19, 27)	24 (20, 30)	23 (19, 27)	<.001
ALT, IU/ml, median (IQR)	16 (13, 20)	28 (21, 39)	20 (15, 26)	<.001
GGT, IU/ml, median (IQR)	19 (14, 29)	32 (21, 54)	23 (15, 36)	<.001
TC, mg/dl, median (IQR)	188 (167, 211)	202 (179, 228)	196 (174, 221)	<.001
FBG, mg/dl, median (IQR)	90 (82, 98)	97 (87, 110)	92 (84, 102)	<.001

Abbreviations: AST, aspartate aminotransferase; ALT, alanine aminotransferase; BMI, body mass index; DBP, diastolic blood pressure; FBG, fasting blood glucose; GGT, gamma-glutamyltransferase; IQR, inter-quartile range; NAFLD, non-alcoholic fatty liver disease; SBP, systolic blood pressure; SD, standard deviation; TC, total cholesterol.
^aRural area is defined as the area not included in the urban area (Seoul, Incheon or Gyeonggi-do).
^bIncome ranked by insurance premium. Quintile 1 includes subjects who pay the lowest insurance premium, while Quintile 5 includes subjects who pay the highest insurance premium.

Table 2. Association between NAFLD and development of dementia
	No. of patients	PYs	Events	Incidence rate per 1000 PY	Model 1^a		Model 2^b		Model 3^c
	No. of patients	PYs	Events	Incidence rate per 1000 PY	HR (95% CI)	p	HR (95% CI)	p	HR (95% CI)	p
Classification 1^d
Non-NAFLD	742 934	7 034 537	69 982	9.95	1.00		1.00		1.00
NAFLD	1 263 376	11 988 399	138 424	11.55	1.16 (1.15–1.17)	<.001	1.05 (1.04–1.06)	<.001	1.08 (1.07–1.09)	<.001
Intermediate	2 025 638	19 279 616	206 445	10.71	1.08 (1.07–1.08)	<.001	1.01 (1.00–1.02)	.05	1.03 (1.02–1.04)	.001
Classification 2^e
Non-NAFLD	742 934	7 034 537	69 982	9.95	1.00		1.00		1.00
NAFLD
Group 1	501 854	4 742 137	60 534	12.77	1.29 (1.27–1.30)	<.001	1.06 (1.04–1.07)	<.001	1.08 (1.07–1.10)	<.001
Group 2	761 522	7 246 262	77 890	10.75	1.08 (1.07–1.09)	<.001	1.05 (1.04–1.06)	<.001	1.08 (1.06–1.09)	<.001
Intermediate	2 025 638	19 279 616	206 445	10.71	1.08 (1.07–1.08)	<.001	1.01 (1.00–1.02)	.048	1.03 (1.02–1.04)	<.001

Note: NAFLD subjects were further categorized into two groups: NAFLD group 1 (HSI >36 at all health check-ups) and NAFLD group 2 (HSI >36 at one or more, but not all, health check-ups).
Abbreviations: CI, confidence interval; HR, hazard ratio; NAFLD, non-alcoholic fatty liver disease; PY person-years.
^aModel 1: non-adjusted.
^bModel 2: adjusted for age, sex, income, disability, residence area, hypertension, total cholesterol, fasting blood glucose, systolic blood pressure and diastolic blood pressure.
^cModel 3: considers competing risk of death in addition to the results of model 2.
^dClassification 1: The subjects were categorized into three groups: non-NAFLD, NAFLD and intermediate.
^eClassification 2: The subjects were categorized into four groups: non-NAFLD, NAFLD group 1 (HSI >36 at all check-ups), NAFLD group 2 (HSI >36 at one or more, but not all check-ups) and intermediate.

Table 3. Association between NAFLD and development of dementia by sex
	Male						Female
	Model 1^a		Model 2^b		Model 3^c		Model 1^a		Model 2^b		Model 3^c
	HR (95% CI)	p	HR (95% CI)	p	HR (95% CI)	p	HR (95% CI)	p	HR (95% CI)	p	HR (95% CI)	p
Classification 1^d
Non-NAFLD	1.00		1.00		1.00		1.00		1.00		1.00
NAFLD	0.71 (0.70–0.72)	<.001	0.99 (0.97–1.00)	.09	1.02 (1.01–1.04)	.001	1.77 (1.74–1.79)	<.001	1.15 (1.14–1.17)	<.001	1.16 (1.15–1.18)	<.001
Intermediate	0.82 (0.81–0.83)	<.001	0.96 (0.94–0.97)	<.001	0.98 (0.97–0.99)	.008	1.33 (1.32–1.35)	<.001	1.09 (1.07–1.10)	<.001	1.09 (1.08–1.11)	<.001
Classification 2^e
Non-NAFLD	1.00		1.00		1.00		1.00		1.00		1.00
NAFLD
Group 1	0.70 (0.68–0.71)	<.001	1.00 (0.98–1.01)	.61	1.04 (1.02–1.06)	<.001	1.92 (1.90–1.95)	<.001	1.16 (1.14–1.18)	<.001	1.17 (1.15–1.18)	<.001
Group 2	0.71 (0.70–0.72)	<.001	0.98 (0.97–1.00)	.04	1.02 (1.00–1.03)	.02	1.64 (1.62–1.67)	<.001	1.15 (1.14–1.17)	<.001	1.16 (1.14–1.18)	<.001
Intermediate	0.82 (0.81–0.83)	<.001	0.96 (0.94–0.97)	<.001	0.98 (0.97–0.99)	.008	1.33 (1.32–1.35)	<.001	1.09 (1.07–1.10)	<.001	1.09 (1.08–1.11)	<.001

Note: NAFLD subjects were further categorized into two groups: NAFLD group 1 (HSI >36 at all health check-ups) and NAFLD group 2 (HSI >36 at one or more, but not all, health check-ups).
Abbreviations: CI, confidence interval; HR, hazard ratio; NAFLD, non-alcoholic fatty liver disease.
^aModel 1: non-adjusted.
^bModel 2: adjusted for age, income, disability, residence area, hypertension, total cholesterol, fasting blood glucose, systolic blood pressure and diastolic blood pressure.
^cModel 3: considers competing risk of death in addition to the results of model 2.
^dClassification 1: The subjects were categorized into three groups: non-NAFLD, NAFLD and intermediate.
^eClassification 2: The subjects were categorized into four groups: non-NAFLD, NAFLD group 1 (HSI >36 at all check-ups), NAFLD group 2 (HSI >36 at one or more, but not all check-ups) and intermediate.

processing....

Association Between Non-alcoholic Fatty Liver Disease and the Risk of Dementia

A Nationwide Cohort Study

Results

Baseline Characteristics of the Study Subjects

Incidence Rate of Dementia Among Groups

Association Between NAFLD and Development of Dementia

Propensity Score-matched Analysis

Subgroup Analysis

Association Between NAFLD and Dementia Considering BMI

Tables

Tables

Tables

References

Authors and Disclosures

Authors and Disclosures

Sidebar

Sidebar

Lay Summary

Email This

Feedback