Skip to main content

Table 1 Extraction table of included studies assessing the association between NAFLD and cognitiona

From: Is there an association between non-alcoholic fatty liver disease and cognitive function? A systematic review

Author, Year Country Study design; Length Sample size NAFLD mean age, sex (n females, n males) Risk factors/co-morbidities Cognitive tests NAFLD diagnosis Findingsb
Weinstein et al. (2018) [27] United States Cross-sectional, population-based sampling survey; NHANES Survey data from 2011 to 2014 1102 participants
239 NAFLD participants
Mean NAFLD age: 68.6 years NAFLD -
BMI 35.0 kg/m2
Obesity: 92.5%
Insulin resistance: 62.1%
HTN: 76.1%
High cholesterol: 82.2%
MetS: 58.5%
Control –
BMI 25.00 kg/m2
Obesity: 19.4%
Insulin resistance: 17.5%
HTN: 61.2%
High cholesterol: 74.2%
MetS: 11.1%
(1) The Word Learning subset - used to assess immediate and delayed learning ability
(2) The Animal Fluency Test - categorical verbal fluency
(3) The digit symbol substitution - involves processing speed, sustained and working memory
Fatty liver index score ≥ 60 Spanish. (1) The Word Learning subset – Immediate verbal memory: NAFLD only (SD: 20.5 ± 0.4) vs the control group (SD: 20.0 ± 0.3)
Delayed verbal memory: NAFLD only group (SD: 6.6 ± 0.2) vs the control group (6.4 ± 0.1)
(2) AFT –
Verbal fluency: NAFLD only (18.6 ± 0.3) vs control group (18.3 ± 0.4)
(3) DSST –
Processing speed, sustained attention, working memory: NAFLD (55.9 ± 1.05) vs control group (53.6 ± 1.2)
Celikbilek et al. (2018) [22] Turkey Cross-sectional study 143 participants
70 NAFLD participants
Mean NAFLD age: 46.9 years, F (NAFLD): 41, M (NAFLD): 29 NAFLD –
T2DM: 30%
HTN: 20%
Hyperlipidaemia: 7.1%
MetS: 48.6%
Control –
T2DM: 4.1%
HTN: 2.7%
Hyperlipidaemia: 1.4%
MetS: 34.3%
(1) Montreal Cognitive Assessment Turkish version. (a) visuospatial abilities - clock-drawing task and three-dimensional cube-copying task, (b) Memory - delayed recall (c) Executive functioning - Trail Making B task, a phonemic fluency task, and a two-item verbal abstraction task, (d) Attention - a serial subtraction task and digits forward and backward tasks, (e) Language - three-item confrontation naming task, repetition of two syntactically complex sentences (f) orientation - orientation to time and place is also evaluated Abdominal ultrasonography (1) MoCA-TR - scores significantly lower in NAFLD group in comparison to control group (P < 0.001)
(a) Visuospatial abilities: P < 0.05
(b) Memory: P = 0.16
(c) Executive functioning: P < 0.05
(d) Attention: P = 0.56
(e) Language: P = 0.89
(f) Orientation: P = 0.29
Takahashi et al. (2017) [28] Japan Cross-sectional study 39 participants
24 NAFLD participants
Mean NAFLD age: 54 years NAFLD –
T2DM: 41.6%
Dyslipidaemia: 58.3%
HTN: 29.2%
Control –
Not reported
(1) A verbal fluency task (VFT) Asia-Pacific Working Party guidelines for the assessment and management of NAFLD.
Ultrasonography (in the absence of other causes of chronic liver disease e.g., hepatitis C antibody negative, hepatitis B surface antigen negative, and alcohol consumption < 20 g/day)
(1) VFT –
Number of words during VFT was significantly higher in controls compared to the NAFLD: p < 0.032
Number of words during VFT NAFLD: 12
Number of words during VFT controls: 14
Filipovic et al. (2018) [21] Serbia Case-control study 76 participants
40 NAFLD participants
Mean NAFLD age: 47.9 years,
M (NAFLD): 22,
F (NAFLD): 18
NAFLD –
T2DM: 30%
HTN: 80%
Obese: 50%
MetS: 85%
Control –
T2DM: 25%
HTN: 44%
Obese: 11%
MetS: 22%
(1) Montreal Cognitive Assessment test Serbian version.
(a) alternating connections, (b) vasoconstrictive abilities - draw a cube and a clock in 11:10 position of clock hands, (c) memory - numbers repeated in the same and reverse order, (d) attention - tap whenever you hear a letter A, serial subtraction of 7, starting with a hundred (e) sentence repeating and verbal fluency.
Abdominal ultrasonography, Sonographic evaluation (US) of hepatic steatosis (1) MOCA-SR –
The cognitive status was lower in NAFLD compared to control:
OR 0.096, 95% CI 0.032–0.289, and p < 0.001
Elliott et al. (2013) [23] UK (Europe) Cohort, Prospective follow-up over 3 years 431 participants
224 NAFLD participant
Mean NAFLD age: 59 years, F (NAFLD): 101, M (NAFLD): 123 Not reported (1) Cognitive Failures Questionnaire (CFQ) - measures memory, attention, concentration, forgetfulness, word finding abilities and confusion. Liver Serum Biochemistry - ALT, ALP, ALB, Bilirubin. (1) CFQ –
Cognitive symptoms CFQ associated with worse functional ability in NAFLD compared to controls: r = 0.4, p < 0.0001
Weinstein et al. (2019) [29] United States Population-based, multi-generational study (The Framingham Heart Study) Cross-sectional study 1287 participants
378 NAFLD participants
Mean NAFLD age: 61.1 years, F (NAFLD): 157, M (NAFLD): 221 NAFLD –
BMI: 31.0
T2DM: 20.5%
HTN (stage 1): 59.3%
Controls –
BMI: 26.8
T2DM: 6.9%
HTN (stage 1): 38.2%
(1) The Wechsler Memory Scale - verbal and visual memory
(2) Time to complete trail-making B minute time to complete trail-making A test (TrB-TrA) - executive function
(3) The similarities test (SIM) - abstract reasoning skills
(4) Hooper visual organization test (HVOT) - visual perception.
Multi-detector CT with 8-slice, A calibration phantom (Image Analysis, Lexington, KY) with a water equivalent compound - three areas from the liver and one from an external phantom were measured (NAFLD was defined as having a liver/phantom ratio ≤ 0.33). (1) The Wechsler Memory Scale -
p = 0.610
(2) TrB-TrA – p = 0.418
(3) SIM – p = 0.746
(4) HVOT – p = 0.528
No significant association of NAFLD with cognitive measures identified.
Participants with NAFLD with advanced fibrosis was associated with poorer performance on the TrB-TrA (P = 0.028) and SIM test (P = 0.009)
Tarter et al. (1984) [30] United States Case-control study 40 participants
30 NAFLD participants
Mean NAFLD age: 40.9 years Not reported (1) Peabody picture intelligence test – verbal intelligence.
(2) Raven’s progressive matrices - nonverbal intelligence.
(3) Digit span forward.
(4) Digit span backward.
(5) Weschler Memory Scale): (a) Logical memory (b) Figural memory, (c) Paired associates, (d) Suprasan.
(6) Perceptual-Motor: (a) Finger tapping, (b) Purdue Pegboard, (c) Stary tracing, (d) Symbol digit modalities.
(7) Spatial: (a) Block design, (b) Tactual performance, (c) Trail making.
(8) Language: (a) Fluency, (b) Confrontation naming, (c) Responsive naming
(9) The Token Test -comprehensive capacity
Hepatic diagnosis - clinical, biochemical, serological confirmed by liver biopsy (1) Peabody picture intelligence test – no difference
(2) Raven’s progressive matrices – no difference
(3) Digit span forward – no difference
(4) Digit span backward – no difference
(5) Weschler Memory Scale) – no difference
(6) Perceptual-Motor –
(a) Finger tapping: no difference
(b) Purdue Pegboard: p < 0.01
(c) Stary tracing: no difference
(d) Symbol digit modalities: p < 0.01
(7) Spatial –
(a) Block design: p < 0.06
(b) Tactual performance: p < 0.05
(c) Trail making: p < 0.05
(8) Language –
(a) Fluency: no difference
(b) Confrontation naming: no difference
(c) Responsive naming: no difference
(9) The Token Test – no difference
Tarter et al. (1987) [31] United States Case-control study 46 participants
23 NAFLD participants
Mean NAFLD age: 37.2 years, M (NAFLD): 11, F (NAFLD): 12 Not reported (1) Digit Span Test - immediate memory
(2) Supraspan Learning Test - the number of trials required accurately to repeat a string of digits that is one digit longer than the digits forward score. (3) Benton Visual Retention Test (BVRT) - short term visual
(4) Rey-Osterreith Complex Figure Test - incidental learning ability
(5) Brown-Peterson Test - measures the rate of decay of information from short-term memory.
Serological tests, clinical history and physical examinations, and was confirmed by a percutaneous liver biopsy. (1) Digit Span Test – no difference
(2) Supraspan Learning Test – p < 0.05
(3) BVRT – p < 0.0001
(4) Rey-Osterreith Complex Figure Test – no difference
(5) Brown-Peterson Test – no difference
Seo et al. (2016) [32] United States Cross-sectional study (NHANES III - survey) 4472 participants
874 NAFLD participant
Mean NAFLD age: 40.9 years NAFLD –
T2DM: 13.6%
HTN: 36.3%
Hypercholesterolaemia: 30.4%
Stroke: 1.5%
Controls –
T2DM: 5.2%
HTN: 17.5%
Hypercholesterolestelemia: 22.2%
Stroke: 0.5%
(1) The Simple Reaction Time Test (SRTT) - measures response time, visual-motor speed
(2) The Symbol Digit Substitution Test (SDST) - assesses visual attention and coding ability
(3) The Serial Digit Learning Test (SDLT) - test learning, recall, and concentration.
Gallbladder examination by ultrasound with a Toshiba SSA-90A machine with a 3.75 and 5.0 MHz transducer. Liver enzymes (ALT, AST) were assayed with Hitachi 737 automated multichannel chemistry analyser. NAFLD was defined as moderate/severe steatosis as deter- mined by ultrasound. (1) SRTT – B = 7.827, 95% CI 0.975 to 14.679
(2) SDST – B = 0.110, 95% CI 0.004 to 0.216
(3) SDLT – B = 0.880, 95% CI 0.317 to 1.443
Participants with NAFLD had lower performances on SRTT, SDST and SDLT
Felipo et al. (2012) [24] Spain (Europe) Case control study 179 participants
29 NAFLD participant
Mean NAFLD age: 45 years, F (NAFLD): 24, M (NAFLD): 5 Not reported (1) The Digit Symbol test (DST) – mental speed
(2) The Number Connection Test A (NCT-A) – mental speed
(3) The Number Connection Test B (NCT-B) – mental speed
(4) The Serial Dotting Test – mental speed
(5) The line tracing test (LTT) – visuospatial
Liver biopsy obtained during histological study. (1) DST – no difference in NAFLD, cirrhosis or NASH
(2) NCT-A – no difference in NAFLD, impaired in patients with NASH and cirrhosis (P < 0.001)
(3) NCT-B – no difference in NAFLD, impaired in NASH and cirrhosis (P < 0.001)
(4) The Serial Dotting Test – no difference in NAFLD, impaired in cirrhosis (P < 0.001) and NASH (P < 0.01)
(5) LTT – no difference in NAFLD, impaired in NASH and cirrhosis (P < 0.001)
Tuttolomondo et al. (2018) [33] Italy (Europe) Case control study 163 participants
80 NAFLD participants
Mean NAFLD age: 53.7 years NAFLD –
BMI: 28.7
HTN: 35.0%
T2DM: 38.8%
Dyslipidaemia: 18.8%
Previous CVD: 3.75%
Controls –
BMI: 25.3
HTN: 15.7%
T2DM: 8.4%
Dyslipidaemia: 16.9%
Previous CVD: 0
(1) Mini-mental state examination (MMSE) Increased serum levels of ALT for at least 6 months and alcohol consumption of < 20 g/day in the previous year. Liver ultrasound finding of steatosis and a liver stiffness value > 6 kPa. Liver biopsy for NAFLD. Hepatic steatosis was defined by detection of Bright Liver Echo pattern. (1) MMSE
NAFLD had a lower mean MMSE score when compared to controls (26.9 ± 1.6 vs. 28.0 ± 1.36; p = 0.005)
  1. aAbbreviations: AFT animal fluency test, ALT alanine aminotransferase, ALP alkaline phosphatase, AST aspartate aminotransferase, BMI body mass index, BVRT Benton visual retention test, CFQ Cognitive Failures Questionnaire, DSST The digit symbol substitution test, DST digit symbol test, HVOT hooper visual organisation test, HTN hypertension, LTT line tracing test, MoCA-TR Montreal cognitive assessment Turkish version, MoCA-SR Montreal cognitive assessment Serbian version, MCI mild cognitive impairment, MetS metabolic syndrome, NAFLD non-alcoholic fatty liver disease, NASH non-alcoholic steatosis, NCTA number connection test A, NCTB number connection test B, SIM similarities test, SRRT simple reaction time test, SDST symbol digit substitution test, SDT serial dotting test, SDLT serial digit learning test, VFT verbal fluency task
  2. bSignificant effect (p < 0.05); no effect (p > 0.05); CI confidence interval, OR odds ratio, RR relative risk