DOI: https://doi.org/10.21203/rs.3.rs-799868/v1
Background: There are several reports on the prevalence of metabolic dysfunction associated fatty liver disease (MAFLD) in the general population. But the fibrosis burden of MAFLD in general population is largely unknown. We aimed to evaluate the prevalence of significant and advanced fibrosis associated with MAFLD in a health check-up cohort.
Materials and methods: Participants who underwent magnetic resonance elastography (MRE) at nationwide thirteen health check-up centers were included in this cross-sectional study. Fatty liver was evaluated using ultrasonography. Significant (≥F2) and advanced (≥F3) hepatic fibrosis were defined by MRE thresholds of 3.0 kPa (range: 2.99–3.65 kPa) and 3.6 kPa (range: 3.4–3.9 kPa), respectively. The sex- and age-standardized prevalence of MAFLD and hepatic fibrosis were estimated.
Results: Total 6,775 health check-up adults included analysis. The sex- and age-standardized prevalence of MAFLD was 33.9%. Prevalence of MAFLD was higher in males than in females, and increased with age. The prevalence of diabetes in MAFLD was 13.3%, and 73.6% of subjects with diabetes had MAFLD. The sex- and age-standardized prevalence of significant (≥F2) and advanced hepatic fibrosis (≥F3) was 9.7% (range: 3.0–9.8%) and 3.0% (range: 2.6–4.6%), respectively, in MAFLD subjects. The prevalence of advanced hepatic fibrosis in diabetic with MAFLD 9.5% (range: 7.5–12.7%).
Conclusion: The sex- and age-standardized prevalence of advanced fibrosis was 3.0% (range: 2.6–4.6%) in subjects with MAFLD, respectively.
An international group of hepatologists have proposed a new definition for metabolic dysfunction associated with fatty liver disease (MAFLD), which is based on a set of positive diagnostic criteria associated with metabolic dysfunction regardless of other hepatic diseases or alcohol consumption.[1] Although there are many controversies regarding the new definition of MAFLD, it is clear that MAFLD include more subjects with hepatic fibrosis than conventional nonalcoholic fatty liver disease NAFLD.
There are seven studies on the epidemiology of MAFLD in community based cohorts.[2–8] Prevalence of MAFLD is reported to be 25.9–39.1%.[2–8] However, data on the prevalence of significant and advanced hepatic fibrosis of MAFLD in the general population are veiled. It is important to diagnose the fibrosis stage, because it is the most powerful prognostic factor to predict long-term hard outcomes.[9] Unfortunately, most studies estimated the hepatic fibrosis burden using a formula to screen for hepatic fibrosis such as the Fibrosis-4 (FIB-4) index or NAFLD fibrosis score (NFS) in subjects with MAFLD.[10–13] And the prevalence of significant fibrosis was 15%[11] and that of advanced fibrosis was 8.8–26.6%, respectively.[12, 13] But FIB-4 and NFS is widely used as rule out strategy for advanced hepatic fibrosis instead of confirmative test, because of low positive predictive value. Moreover, the validation of FIB-4 and NFS was still under-evaluated in new definition of MAFLD. Only two studies have examined the prevalence of hepatic fibrosis in MAFLD using transient elastography (TE) in the general population; these studies were performed in a limited scale (500 ~ 1,700 selected subjects).[2, 6]
To the best of our knowledge, no study has evaluated the burden of hepatic fibrosis in subjects with MAFLD using magnetic resonance elastography (MRE) in a health check-up cohort. Therefore, we aimed to evaluate the prevalence of MAFLD and the prevalence of significant and advanced fibrosis of MAFLD in a nationwide health check-up cohort using MRE.
We conducted a nationwide, multicenter, retrospective, cross-sectional, observational study of a health check-up health check-up cohort.
The Industrial Safety and Health Act requires health check-ups for all adults over the age of 40 and all employees in Korea. Thirteen health promotion centers belonging to the Korea Association of Health Promotion were included. These centers perform about 10% of health check-ups provided by the Korean National Health Insurance Corporation (NHIC) in Korea [14]. The Korean NHIC covers the entire population of Korea and provides medical health check-ups. The MRE test was performed voluntarily among those undergoing health screening. This study was approved by the Institutional Review Board (IRB No. HY-2021-04-001-001).
All employees or over the age of 40 or Koreans have been requested regular health check-up according to Industrial Safety and Health Act. MRE test is not included as basic check-up program. The MRE test is performed voluntary and at your own expense. NHIC serves abdominal sonography and AFP twice a year in patients with chronic viral hepatitis or cirrhosis as different program free of charge. So, known chronic liver disease patients rarely choose the MRE test at their own expense for a health check-up.
Participants aged 18 years or older who underwent a health check-up at the Health Promotion Center between January 2017 and May 2020 in South Korea were included. The exclusion criteria were as follows: inappropriate information about alcohol intake and laboratory variables, missed ultrasonography data for evaluation of fatty liver, and technical issues during MRE that led to examination failure.
All MRE examinations were performed with a 1.5-T imaging system (GE Healthcare, Waukesha, Wisconsin, USA; Gyroscan Intera, Philips Healthcare, Best, Netherlands). Two-dimensional MRE was performed using previously described techniques [15]. The acquired images were processed automatically without manual intervention at the scanner to produce quantitative maps of liver stiffness, known as elastograms. Elastograms with 95% confidence maps were used to obtain the liver stiffness values. The liver stiffness values were measured by the placement of four regions of interest (ROIs) covering the largest liver, excluding artifacts, large vessels, gallbladder, and fissures, by attending radiologists. Median values measured for the four ROIs were used.
According to previous studies,[16–20] significant fibrosis (≥ F2) was defined as fibrosis with a threshold of 3.0 kPa (range of thresholds: 2.99 kPa to 3.65 kPa) and advanced fibrosis was defined as that with a threshold of 3.6 kPa (range of thresholds: 3.4 kPa to 3.9 kPa).
Clinical and biochemical measurements were collected during health examinations. The variables included alcohol consumption, blood pressure, waist circumference, and body mass index (BMI). Significant alcohol intake was defined as an intake of more than 30 g/day for males and 20 g/day for females. Abnormal waist circumference was defined as ≥ 90 cm in males and ≥ 80 cm in females. We also assessed concomitant liver disease, including hepatitis B or C virus infection, defined as positivity for serological markers of hepatitis B or C, and alcoholic liver disease, defined as chronic parenchymal liver disease or liver cirrhosis with significant alcohol consumption. Patients fasted overnight before blood samples were collected for the following tests: full blood count, aspartate aminotransferase (AST), alanine aminotransferase (ALT), r-glutamyl transpeptidase, albumin, total cholesterol, high-density lipoprotein cholesterol, triglyceride, glucose, and hemoglobin A1c. The presence of obesity, impaired fasting glucose, diabetes, hypertension, dyslipidemia, and metabolic syndrome were defined according to Korean guidelines. The diagnosis of hepatic steatosis was assessed using ultrasonography results, regardless of severity.
For each subject, we investigated four surrogate blood indices of liver fibrosis: the AST to platelet ratio (APRI), AST/ALT ratio (AAR), NFS, and FIB-4. APRI, AAR, NFS, and FIB-4 indices were calculated as previously described [21–23].
NAFLD was defined as the presence of ultrasonographic fatty liver without significant alcohol consumption or concomitant liver disease [24]. MAFLD was defined as the presence of hepatic steatosis with one or more of the following [25]: (1) an overweight or obese condition (BMI ≥ 23 kg/m2; MAFLD group I); (2) lean/normal weight with at least two metabolic abnormalities as described previously (MAFLD group II); (3) presence of type 2 diabetes (MAFLD group III). Although the homeostasis model assessment-insulin resistance score and the plasma high-sensitivity C-reactive protein level indicate metabolic risk abnormalities, they were not available in our dataset.
Our health check-up cohort showed male dominance, although this was a result of the demographics of the general population due to the nature of health check-ups. Therefore, we adjusted the prevalence according to the 2018 sex and age data from the Korean Statistical Information Service. The ages of the population were divided into the following six age groups: 20–29, 30–39, 40–49, 50–59, 60–69, and ≥ 70 years. The sex- and age-standardized prevalence of MAFLD and NAFLD were estimated. Additionally, the prevalence of significant and advanced fibrosis in these patients were also calculated.
Baseline characteristics were reported as frequency and percent, or as mean and standard deviation. Continuous variables were compared using the Student’s t-test, and categorical variables were compared using the chi-square test. Subjects were further categorized into the following mutually exclusive groups: neither MAFLD nor NAFLD, MAFLD-only (compatible with only the MAFLD definition), NAFLD-only (compatible with only the NAFLD definition), and both MAFLD and NAFLD (compatible with both MAFLD and NAFLD definitions). Analysis of variance was used to compare the characteristics of multiple groups. Statistical significance was set at a two-sided P-value of < 0.05. All analyses were conducted using the SPSS 25 software (SPSS Inc., Chicago, IL, USA).
A total of 8,545 participants consecutively underwent MRE as part of a health check-up between January 2017 and May 2020. Among them, we included 6,775 subjects who also underwent ultrasonography for analysis (Fig. 1). The mean age was 46.8 years, and males were predominant (80.6%). The mean BMI was 24.8 kg/m2, and the proportion of subjects with abnormal waist circumference was 35.2%. The proportion of subjects with impaired fasting glucose, hypertension, and dyslipidemia was 32.2%, 28.0%, and 18.9%, respectively. The proportion of subjects with overall alcohol intake and significant alcohol intake was 54.6% and 17.4%, respectively (Table 1). The overall sex- and age-standardized prevalence of significant fibrosis was 7.3% (range, 2.4–7.6%) and that of advanced fibrosis was 2.5% (range, 1.8–3.4%) in whole participants.
Total (n = 6,775) |
Non-MAFLD (n = 3,754) |
MAFLD (n = 3,021) |
P-value |
|
---|---|---|---|---|
Clinical parameters |
||||
Male, n (%) |
5,460 (80.6) |
2,725 (72.6) |
2,735 (90.5) |
< 0.001 |
Age, years |
46.8 ± 10.3 |
46.3 ± 10.7 |
47.4 ± 9.8 |
< 0.001 |
Significant alcohol consumption, (%) |
1,182 (17.4) |
663 (17.7) |
519 (17.2) |
0.604 |
Waist circumference, cm |
85.4 ± 9.1 |
81.1 ± 8.1 |
90.7 ± 7.3 |
< 0.001 |
Abnormal waist circumference, (%) |
2,386 (35.2) |
699 (18.6) |
1,687 (55.8) |
< 0.001 |
BMI, kg/m2 |
24.8 ± 3.2 |
23.3 ± 2.7 |
26.7 ± 2.9 |
< 0.001 |
Fat mass, kg |
18.6 ± 5.8 |
16.1 ± 4.7 |
21.6 ± 5.6 |
< 0.001 |
Lean mass, kg |
49.2 ± 8.9 |
46.7 ± 8.7 |
52.3 ± 8.2 |
< 0.001 |
Alcohol consumption |
94.5 ± 166.9 |
93.2 ± 163.8 |
96.1 ± 170.6 |
0.474 |
Systolic blood pressure, mmHg |
116.4 ± 13.5 |
114.0 ± 13.2 |
119.4 ± 13.2 |
< 0.001 |
Diastolic blood pressure, mmHg |
74.8 ± 9.6 |
73.2 ± 9.2 |
76.8 ± 9.7 |
< 0.001 |
Laboratory findings |
||||
Platelets, ⅹ109/L |
246.3 ± 54.1 |
244.4 ± 55.5 |
248.6 ± 52.3 |
0.001 |
AST, U/L |
30.5 ± 19.8 |
27.3 ± 16.3 |
34.5 ± 22.9 |
< 0.001 |
ALT, U/L |
32.8 ± 33.4 |
25.0 ± 25.9 |
42.5 ± 38.8 |
< 0.001 |
GGT, U/L |
55.7 ± 84.9 |
43.5 ± 66.1 |
70.9 ± 101.5 |
< 0.001 |
Albumin, g/dl |
4.50 ± 0.27 |
4.48 ± 0.28 |
4.53 ± 0.26 |
< 0.001 |
Cholesterol, mg/dl |
199.9 ± 38.1 |
196.2 ± 35.6 |
204.7 ± 40.5 |
< 0.001 |
Triglyceride, mg/dl |
142.7 ± 113.6 |
108.6 ± 72.4 |
185.1 ± 138.5 |
< 0.001 |
Glucose, mg/dl |
98.8 ± 20.9 |
94.4 ± 15.8 |
104.3 ± 24.8 |
< 0.001 |
HbA1c, % |
5.8 ± 0.8 |
5.6 ± 0.6 |
6.0 ± 0.9 |
< 0.001 |
Comorbid disease |
||||
Metabolic syndrome, n (%) |
1,512 (22.4) |
317 (8.5) |
1,195 (39.7) |
< 0.001 |
Obesity, n (%) |
3,081 (45.5) |
932 (24.8) |
2,149 (71.1) |
< 0.001 |
Diabetes, n (%) |
545 (8.0) |
144 (3.8) |
401 (13.3) |
< 0.001 |
Impaired fasting glucose, n (%) |
2,182 (32.2) |
833 (22.2) |
1,349 (44.7) |
< 0.001 |
Hypertension, n (%) |
1,895 (28.0) |
754 (20.1) |
1,141 (37.8) |
< 0.001 |
Dyslipidemia, n (%) |
1,280 (18.9) |
454 (12.1) |
826 (27.3) |
< 0.001 |
Hepatic fibrosis index |
||||
FIB-4 |
1.14 ± 0.70 |
1.17 ± 0.70 |
1.11 ± 0.63 |
0.002 |
FIB-4 (Intermediate or high risk), (%) |
1,701 (25.1) |
993 (26.5) |
708 (23.4) |
0.004 |
NFS |
-2.32 ± 1.19 |
-2.42 ± 1.19 |
-2.20 ± 1.18 |
< 0.001 |
AAR |
1.11 ± 0.43 |
1.25 ± 0.41 |
0.94 ± 0.38 |
< 0.001 |
APRI |
0.39 ± 0.32 |
0.35 ± 0.30 |
0.43 ± 0.35 |
< 0.001 |
MR elastography |
||||
MRE, kPa |
2.34 ± 0.57 |
2.30 ± 0.57 |
2.40 ± 0.56 |
< 0.001 |
≥F2 (threshold: 3.0 kPa) (%) |
554 (8.2) |
254 (6.8) |
300 (9.9) |
< 0.001 |
≥F3 (threshold: 3.6 kPa) (%) |
151 (2.2) |
79 (2.1) |
72 (2.4) |
0.440 |
Note. Data are presented with mean ± standard deviation, number of subjects (percentage). *Age- and sex-standardized prevalence. | ||||
Abbreviations: BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; FIB-4, fibrosis-4 index; GGT, gamma glutamyl transferase; NFS, NAFLD fibrosis score; AAR, AST/ALT ratio; APRI, AST to platelet ratio; MR, magnetic resonance; NAFLD, non- alcoholic fatty liver disease; MAFLD, metabolic dysfunction associated fatty liver disease |
In a health check-up cohort (n = 6,775), 47.5% (n = 3,215) of the subjects had fatty liver on ultrasonography (Fig. 1). The sex- and age-standardized prevalence of MAFLD and NAFLD was 33.9% and 28.7%, respectively. The sex- and age-standardized prevalence of MAFLD in males was significantly higher than that in females (45.7% vs. 22.2%, P < 0.001). Prevalence of MAFLD increased according to age, but the prevalence of MAFLD showed a slight decrease during the 70s in both males and females (Supplement Fig. 1). Among subjects with MAFLD, 95.2% (n = 2,875) were categorized into group I (obese MAFLD), 4.3% (n = 129) into group II (lean MAFLD), and 13.3% (n = 402) into group III (diabetes). Most of the group III subjects overlapped with those of group I (91.5%, 368/402) (Supplementary Fig. 2 and Table 2).
MAFLD (n = 3,021) |
MAFLD group I (n = 2,875) |
MAFLD group II (n = 129) |
MAFLD group III (n = 402) |
|
---|---|---|---|---|
Clinical parameters |
||||
Male, n (%) |
2,735 (90.5) |
2,615 (91.0) |
106 (82.2) |
365 (90.8) |
Age, years |
47.4 ± 9.8 |
47.2 ± 9.7 |
52.7 ± 8.7 |
52.4 ± 9.2 |
Waist, cm |
90.7 ± 7.3 |
91.2 ± 7.1 |
81.7 ± 4.7 |
92.0 ± 8.3 |
BMI, kg/m2 |
26.7 ± 2.9 |
26.9 ± 2.7 |
22.0 ± 0.8 |
26.9 ± 3.2 |
Fat mass, kg |
21.6 ± 5.6 |
21.9 ± 5.4 |
14.7 ± 2.5 |
21.9 ± 6.3 |
Lean mass, kg |
52.3 ± 8.2 |
52.7 ± 8.0 |
44.1 ± 7.7 |
51.8 ± 8.3 |
Alcohol consumption |
96.1 ± 170.6 |
96.7 ± 171.1 |
86.6 ± 169.1 |
104.2 ± 193.5 |
Systolic blood pressure, mmHg |
119.4 ± 13.2 |
119.4 ± 13.3 |
119.9 ± 12.6 |
123.4 ± 14.6 |
Diastolic blood pressure, mmHg |
76.8 ± 9.7 |
76.8 ± 9.7 |
78.2 ± 9.5 |
79.1 ± 10.5 |
Laboratory findings |
||||
Platelet, ⅹ109/L |
248.6 ± 52.3 |
248.6 ± 52.2 |
251.6 ± 53.9 |
241.8 ± 53.1 |
AST, U/L |
34.5 ± 22.9 |
34.5 ± 22.7 |
34.4 ± 25.7 |
41.3 ± 29.2 |
ALT, U/L |
42.5 ± 38.8 |
42.9 ± 39.5 |
33.3 ± 17.7 |
45.4 ± 32.7 |
GGT, U/L |
70.9 ± 101.5 |
70.1 ± 99.1 |
86.5 ± 144.9 |
102.7 ± 172.1 |
Albumin, g/dl |
4.53 ± 0.26 |
4.53 ± 0.26 |
4.53 ± 0.26 |
4.56 ± 0.27 |
Cholesterol, mg/dl |
204.7 ± 40.5 |
204.5 ± 39.9 |
209.7 ± 49.8 |
189.8 ± 50.2 |
Triglyceride, mg/dl |
185.1 ± 138.5 |
184.1 ± 138.2 |
212.8 ± 147.4 |
217.2 ± 201.6 |
Glucose, mg/dl |
104.3 ± 24.8 |
103.9 ± 24.1 |
104.0 ± 13.9 |
149.5 ± 40.4 |
HbA1c, % |
6.0 ± 0.9 |
6.0 ± 0.9 |
5.9 ± 0.6 |
7.6 ± 1.4 |
Comorbid disease |
||||
Metabolic syndrome, n (%) |
1,195 (39.7) |
1,149 (40.1) |
46 (35.7) |
280 (70.4) |
Obesity, n (%) |
2,149 (71.1) |
2,149 (74.7) |
0 |
286 (71.1) |
Diabetes, n (%) |
401 (13.3) |
368 (12.8) |
16 (12.4) |
402 (100) |
Impaired fasting glucose, n (%) |
1,349 (44.7) |
1,255 (43.7) |
78 (60.5) |
383 (95.3) |
Hypertension, n (%) |
1,141 (37.8) |
1,084 (37.7) |
57 (44.2) |
236 (58.7) |
Dyslipidemia, n (%) |
826 (27.3) |
770 (26.8) |
51 (39.5) |
138 (34.3) |
Hepatic fibrosis index |
||||
FIB-4 |
1.11 ± 0.63 |
1.10 ± 0.62 |
1.36 ± 0.86 |
1.45 ± 0.96 |
FIb-4 (Intermediate or high risk), (%) |
708 (23.4) |
652 (22.7) |
48 (37.2) |
157 (39.1) |
NFS |
-2.20 ± 1.18 |
-2.21 ± 1.18 |
-2.13 ± 1.25 |
-1.23 ± 1.05 |
AAR |
0.94 ± 0.38 |
0.93 ± 0.37 |
1.10 ± 0.43 |
1.02 ± 0.52 |
APRI |
0.43 ± 0.35 |
0.43 ± 0.34 |
0.43 ± 0.43 |
0.54 ± 0.45 |
MR elastography |
||||
MRE (kPa) |
2.40 ± 0.56 |
2.40 ± 0.56 |
2.26 ± 0.58 |
2.68 ± 0.83 |
≥F2 (threshold: 3.0 kPa), (%) |
300 (9.9) |
285 (9.9) |
10 (7.8) |
86 (21.4) |
≥F3 (threshold: 3.6 kPa), (%) |
72 (2.4) |
65 (2.3) |
4 (3.1) |
38 (9.5) |
Note. Data are presented with mean ± standard deviation, number of subjects (percentage). | ||||
Abbreviations: BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; FIB-4, fibrosis-4 index; GGT, gamma glutamyl transferase; NFS, NAFLD fibrosis score; AAR, AST/ALT ratio; APRI, AST to platelet ratio; MR, magnetic resonance; NAFLD, non- alcoholic fatty liver disease; MAFLD, metabolic dysfunction associated fatty liver disease |
Among subjects with MAFLD, 22.7% (686/3,021) had concomitant other liver disease (significant alcohol consumption (71.9%), chronic viral hepatitis (24.3%), and both (3.8%)). The prevalence of obesity (BMI ≥ 25 kg/m2) in MAFLD was 71.1%, and 79.0% of obese subjects had MAFLD. The prevalence of diabetes in MAFLD was 13.3%, and 73.6% of subjects with diabetes had MAFLD. The prevalence of metabolic syndrome, hypertension, and dyslipidemia was 39.7%, 37.8%, and 27.3%, respectively, in the MAFLD group. The prevalence of MAFLD in metabolic syndrome, hypertension, and dyslipidemia was 79.0%, 60.2%, and 64.5%, respectively (Fig. 2). The prevalence of significant and advanced fibrosis was 20.9% and 9.5% in patients with diabetes (Fig. 3).
The prevalence of advanced fibrosis also increased with age (Fig. 4). The sex- and age-standardized prevalence of significant fibrosis was 9.7% (range, 3.0–9.8%) and that of advanced fibrosis was 3.0% (range, 2.6–4.6%) in subjects with MAFLD (Table 2). Meanwhile, the sex- and age-standardized prevalence of significant fibrosis was 9.4% (range, 3.1–9.4%) and that of advanced fibrosis was 3.3% (range, 2.8–4.5%) in subjects with NAFLD. Advanced fibrosis was significantly higher in females than in males (3.8% vs. 2.2%, P = 0.036).
The prevalence of significant hepatic fibrosis in MAFLD group I (overweight or obese MAFLD) was 9.9% (range, 2.1–10.1%). The prevalence of significant hepatic fibrosis in MAFLD group II (lean MAFLD) and group III (diabetic MAFLD) was 7.8% (range, 2.3–7.8%) and 21.4% (range, 9.2–21.4%), respectively. The prevalence of advanced hepatic fibrosis in MAFLD group I was 2.3% (range: 1.5–3.8%), 3.1% (range: 2.3–4.7%) in MAFLD group II, and 9.5% (range: 7.5–12.7%) in MAFLD group III.
Of the 6,775 subjects who underwent MRE, those with overlapping MAFLD and NAFLD comprised 34.5% (n = 2,335). The MAFLD-only and NAFLD-only groups comprised 10.1% (n = 686) and 2.2% (n = 148), respectively (Table 3 and Fig. 5). Subjects who were compatible with only the MAFLD definition included a higher proportion of men, and exhibited higher blood pressure, triglyceride levels, fasting blood glucose levels, and rate of comorbid disease than those who were compatible with only the NAFLD definition (P < 0.001 in all cases) (Table 3). The MRE value of the MAFLD-only group was significantly higher than that of the NAFLD-only group (2.44 ± 0.62 kPa vs. 2.27 ± 0.51 kPa, P = 0.002).
None |
NAFLD-only |
MAFLD-only |
Both NAFLD and MAFLD |
P-value |
|
---|---|---|---|---|---|
N |
3,606 (53.2) |
148 (2.2) |
686 (10.1) |
2,335 (34.5) |
|
Clinical parameters |
|||||
Male, n (%) |
2,609 (72.4) |
116 (78.4) |
622 (90.7) |
2,113 (90.5) |
< 0.001 |
Age, years |
46.3 ± 10.7 |
46.6 ± 10.3 |
47.0 ± 9.5 |
47.5 ± 9.8 |
< 0.001 |
Waist, cm |
81.1 ± 8.2 |
78.9 ± 5.1 |
90.6 ± 7.4 |
90.8 ± 7.3 |
< 0.001 |
BMI, kg/m2 |
23.4 ± 2.7 |
21.7 ± 1.1 |
26.6 ± 2.9 |
26.7 ± 2.9 |
< 0.001 |
Fat mass, kg |
16.2 ± 4.8 |
13.6 ± 2.8 |
21.6 ± 5.7 |
21.5 ± 5.5 |
< 0.001 |
Lean mass, kg |
46.7 ± 8.8 |
45.3 ± 6.6 |
52.4 ± 8.0 |
52.2 ± 8.3 |
< 0.001 |
Alcohol consumption |
95.5 ± 166.3 |
37.4 ± 57.1 |
304.7 ± 248.7 |
34.8 ± 54.5 |
< 0.001 |
Systolic blood pressure, mmHg |
113.9 ± 13.3 |
113.0 ± 10.3 |
117.9 ± 13.4 |
119.8 ± 13.2 |
< 0.001 |
Diastolic blood pressure, mmHg |
73.2 ± 9.2 |
72.1 ± 7.0 |
75.8 ± 9.9 |
77.1 ± 9.6 |
< 0.001 |
Laboratory findings |
|||||
Platelet, ⅹ109/L |
244.1 ± 55.5 |
252.6 ± 54.7 |
240.4 ± 54.3 |
251.0 ± 51.5 |
< 0.001 |
AST, U/L |
27.2 ± 15.9 |
29.2 ± 22.8 |
34.4 ± 25.4 |
34.5 ± 22.1 |
< 0.001 |
ALT, U/L |
24.8 ± 25.9 |
28.9 ± 26.2 |
42.8 ± 46.6 |
42.4 ± 36.1 |
< 0.001 |
GGT, U/L |
42.9 ± 60.9 |
57.9 ± 142.5 |
62.4 ± 63.9 |
73.4 ± 110.0 |
< 0.001 |
Albumin, g/dl |
4.48 ± 0.28 |
4.51 ± 0.30 |
4.52 ± 0.27 |
4.54 ± 0.26 |
< 0.001 |
Cholesterol, mg/dl |
196.1 ± 35.6 |
198.1 ± 33.6 |
203.9 ± 40.8 |
204.9 ± 40.4 |
< 0.001 |
Triglyceride, mg/dl |
108.7 ± 72.9 |
105.7 ± 55.0 |
180.3 ± 160.9 |
186.5 ± 131.2 |
< 0.001 |
Glucose, mg/dl |
94.5 ± 15.9 |
92.1 ± 9.0 |
103.4 ± 24.4 |
104.6 ± 24.9 |
< 0.001 |
HbA1c, % |
5.62 ± 0.62 |
5.59 ± 0.39 |
5.96 ± 0.93 |
6.00 ± 0.89 |
< 0.001 |
Comorbid disease |
|||||
Metabolic syndrome, n (%) |
317 (8.8) |
0 |
250 (36.5) |
945 (40.6) |
< 0.001 |
Obesity, n (%) |
932 (25.8) |
0 |
481 (70.1) |
1,668 (71.4) |
< 0.001 |
Diabetes, n (%) |
143 (4.0) |
1 (0.7) |
81 (11.8) |
320 (13.7) |
< 0.001 |
Impaired fasting glucose, n (%) |
814 (22.6) |
19 (12.8) |
290 (42.3) |
1,059 (45.4) |
< 0.001 |
Hypertension, n (%) |
743 (20.6) |
11 (7.4) |
236 (34.4) |
905 (38.8) |
< 0.001 |
Dyslipidemia, n (%) |
448 (12.4) |
6 (4.1) |
191 (27.8) |
635 (27.2) |
< 0.001 |
Hepatic fibrosis index |
|||||
FIB-4 |
1.17 ± 0.70 |
1.13 ± 0.73 |
1.13 ± 0.56 |
1.11 ± 0.65 |
0.014 |
NFS |
-2.39 ± 1.19 |
-2.89 ± 1.19 |
-2.13 ± 1.18 |
-2.22 ± 1.19 |
< 0.001 |
AAR |
1.26 ± 0.41 |
1.17 ± 0.37 |
0.94 ± 0.32 |
0.94 ± 0.39 |
< 0.001 |
APRI |
0.35 ± 0.28 |
0.39 ± 0.58 |
0.45 ± 0.36 |
0.43 ± 0.34 |
< 0.001 |
MR elastography |
|||||
MRE (kPa) |
2.29 ± 0.57 |
2.27 ± 0.51 |
2.44 ± 0.62 |
2.39 ± 0.54 |
< 0.001 |
≥F2 (threshold: 3.0 kPa) (%) |
245 (6.8) |
9 (6.1) |
90 (13.1) |
320 (9.0) |
< 0.001 |
≥F3 (threshold: 3.6 kPa) (%) |
75 (2.1) |
4 (2.7) |
18 (2.6) |
54 (2.3) |
0.479 |
Note. Data are presented with mean ± standard deviation, number of subjects (percentage). | |||||
Abbreviations: BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; FIB-4, fibrosis-4 index; GGT, gamma glutamyl transferase; NFS, NAFLD fibrosis score; AAR, AST/ALT ratio; APRI, AST to platelet ratio; MR, magnetic resonance; NAFLD, non alcoholic fatty liver disease; MAFLD, metabolic dysfunction associated fatty liver disease. |
We evaluated the prevalence of significant and advanced fibrosis associated with MAFLD in a health check-up cohort. The sex- and age-standardized prevalence of MAFLD was 33.9% in a large health check-up cohort in South Korea. It was higher in males than in females, and increased with age. The prevalence of significant hepatic fibrosis (≥ F2) was 9.7% (range: 3.0–9.8%), and that of advanced hepatic fibrosis (≥ F3) was 3.0% (range: 2.6–4.6%). In particular, significant and advanced fibrosis was higher in the diabetes MAFLD group, 21.4% and 9.5%, respectively.
Our study is based on a large cohort of MAFLD patients from thirteen health check-up health centers nationwide. All thirteen centers used the same vendor for MRE. Taking into account that liver biopsy cannot estimate the prevalence of hepatic fibrosis in the community population; this study is the largest study to investigate the hepatic fibrosis burden in a real-world setting. We also tried to evaluate the prevalence of fibrosis by standardizing sex and age and by presenting the prevalence with ranges according to various MRE threshold ranges. Significant fibrosis (≥ F2) was defined as fibrosis with a threshold of 3.0 kPa (range of thresholds: 2.99 kPa to 3.65 kPa) and advanced fibrosis was defined as that with a threshold of 3.6 kPa (range of thresholds: 3.4 kPa to 3.9 kPa) according to previous studies.[16–20]
There are two studies that investigated the burden of hepatic fibrosis in MAFLD subjects in the health check-up cohort and the HHANES cohort using fibroscan.[2, 6] The prevalence of advanced fibrosis in MAFLD was 7.4% with a cutoff of 9.7 kPa in the U.S. cohort [6] and 3.0% with a cutoff of 10.0 kPa in the Hong Kong cohort [2], whereas it was 3.0% (range, 2.6–4.6%) in the present cohort. Our data thus showed a similar prevalence of advanced hepatic fibrosis to that observed in Hong Kong, but the prevalence was lower than that in the United States. This difference is thought to be due to differences in race, prevalence of diabetes, BMI, and diagnostic methods.[26] Most important point is prevalence of diabetes of each cohort. The prevalence of diabetes in Hong Kong and the United States was found to be 5.8% and 13%, respectively. The prevalence of diabetes in the present study was 8.0%. Second, the measurement tools and cut-offs also differed. MRE has so far shown the best diagnostic performance in discriminating significant or advanced hepatic fibrosis and estimating hepatic fibrosis burden.[27] The summary estimate of the area under the curve of TE for diagnosis of significant fibrosis and advanced fibrosis were 0.83 and 0.85, respectively, whereas those of MRE were 0.91 and 0.92, respectively, in a meta-analysis.[27] To the best of our knowledge, this is the first and largest nationwide health check-up MRE cohort. It is noteworthy that 7.3% and 2.5% of the participants (sex- and age-adjusted prevalence) were respectively accompanied by significant and advanced fibrosis according to the reference method of MRE within a large health check-up health check-up cohort.
In this study, the thresholds to diagnose significant hepatic fibrosis and advanced hepatic fibrosis in MRE were set at 3.0 kPa and 3.6 kPa, respectively. The thresholds of 3.0 kPa and 3.6 kPa are the most widely used thresholds for predicting significant hepatic fibrosis and advanced hepatic fibrosis, respectively, with previous studies reporting areas under the curve of 0.92 and 0.93, respectively.[28–30] However, the thresholds of MRE for diagnosing significant hepatic fibrosis and advanced hepatic fibrosis vary widely across studies.[16–20] For this reason, we presented the prevalence of hepatic fibrosis in MAFLD subjects as a range of representative values.
Our study has several limitations. First, the subjects of our cohort did not have the same sex and age distributions as the general population. In particular, the proportion of females was significantly lower than that of males due to differences in gender employment. In order to correct for age and gender imbalance, the sex- and age-standardized prevalence of MAFLD, NAFLD, and hepatic fibrosis were estimated. Second, there is a high likelihood of selection bias based on the possibility that subjects who are concerned about their liver health are likelier to undergo MRE. However, the prevalence of hypertension, diabetes, and significant alcohol intake in our cohort were comparable to that of the general population. Third, although MRE is the most accurate diagnostic tool, except for liver biopsy, the threshold for either significant or advanced fibrosis has not been standardized. To compensate for this limitation, we used a range of MRE values to analyze the prevalence.
In conclusion, the prevalence of MAFLD was 33.9%, which was higher in males and increased with age. The prevalence of significant and advanced fibrosis was 9.7% and 3.0%, respectively, in MAFLD, and it was remarkably higher in subjects with diabetic MAFLD in the nationwide health check-up cohort.
Conflict of interest statement: All authors declare that they have no conflict of interest.
Financial support statement: This study was supported by National Research Foundation of Korea 2020R1A2C2009227 (Dae Won Jun).
Ethics approval: This study was approved by the Institutional Review Board (IRB No. HY-2021-04-001-001).
Availability of data and material: All the data were obtained upon the request to the corresponding authors.
Author contributions: Concept and design: Dae Won Jun. Collection and management of data: Dae Won Jun, Eun-Hee Nah. Interpretation of data: Dae Won Jun, Mimi Kim, Writing of the manuscript: Mimi Kim, Eileen L. Yoon. Supervision and critical review: Eileen L. Yoon, Seon Cho, Chul-min Lee, Bo Kyeong Kang. The authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Acknowledgment: We express our sincere thanks to Soorack Ryu, statistician who provided useful information about the prevalence and standardized prevalence.