Sage Journals: Discover world-class research

Abstract

Background:

Cardiovascular disease (CVD) has emerged as the most significant complication and leading cause of death among metabolic-associated fatty liver disease (MAFLD).

Objectives:

This study aims to investigate the CVD risk among MAFLD subgroups.

Design:

Data of participants from June 2017 to January 2023 in the Physical Examination Center of the Third Hospital of Hebei Medical University were collected. MAFLD were divided into four subgroups: metabolic healthy lean/normal weight MAFLD (MHL), metabolic healthy overweight/obese MAFLD (MHO), metabolic dysfunctional lean/normal weight MAFLD (MDL), and metabolic dysfunctional overweight/obese MAFLD (MDO).

Methods:

The risk assessment for atherosclerotic CVD was performed based on the flowchart for primary prevention risk assessment in Chinese adults.

Results:

The proportions of MHL, MHO, MDL, and MDO were 0.77% (n = 185), 10.05% (n = 2406), 1.29% (n = 310), and 16.86% (n = 4038), respectively. After adjustment for gender, age, smoking history, drinking history, and significant liver fibrosis, the subgroup of MAFLD was still an independent risk factor for high adverse cardiovascular events (HACE). Compared with the MHL, the MDL had the highest risk, followed by MDO (all p < 0.05), and there was no significant difference between MHO and MHL. We performed regression analysis according to age (65 years) and gender (male or female), respectively, and the results were similar to those of the total population.

Conclusion:

MAFLD is associated with a higher risk of CVD, especially in MDO. Classification of MAFLD based on body mass index and metabolic status helps in risk stratification, which will mitigate or prevent the development of CVD.

Plain language summary

The relationship between MAFLD and CVD in a health check-up Chinese population

In parallel with the escalation of the global obesity epidemic, metabolic-associated fatty liver disease (MAFLD) currently has become the most prevalent chronic liver disease globally. Among the array of complications of MAFLD, cardiovascular diseases (CVD) notably stands out as the leading cause of mortality in individuals with MAFLD. Therefore, this study aimed to investigate the risk of cardiovascular events in different MAFLD subgroups. Our findings implied potential variations in HACE risks across distinct subtypes of MAFLD. Furthermore, a significant association was observed between HACE and the presence of lean status and/or metabolic dysregulation, underlining the importance of classifying MAFLD subgroups based on BMI and metabolic status to better assess CVD risk in individuals with MAFLD.

Keywords

atherosclerotic cardiovascular diseases lean/normal weight liver fibrosis metabolic-associated fatty liver disease metabolic dysfunction

Introduction

In parallel with the escalation of the global obesity epidemic, metabolic-associated fatty liver disease (MAFLD), currently, has become the most prevalent chronic liver disease globally, affecting approximately 1 billion people.^1,2 As a high-incidence area of MAFLD, the prevalence of steatosis was 44.39% based on the database of the largest health check-up chain in China.³ Cardiovascular disease (CVD) is a leading cause of mortality in individuals with MAFLD, attributed to the predictive capacity of MAFLD for adverse CVD outcomes, independent of traditional risk factors,⁴ coupled with the overlap between MAFLD and conventional CVD risk factors. CVD and MAFLD link via insulin resistance (driving hepatic steatosis/atherogenic dyslipidemia), systemic inflammation (TNF-α, IL-6), oxidative stress, altered lipoproteins (sdLDL↑, HDL↓), ectopic fat, genetic variants (PNPLA3), and gut dysbiosis.⁵

MAFLD often coexists with metabolic disorders such as diabetes and dysfunctional lipid metabolism. Research indicated that a substantial proportion of diabetic patients, approximately 70% in regions including the United States and Hong Kong, had hepatic steatosis.^6,7 The progression of CVD is exacerbated by the interference of MAFLD with abnormal glucose and lipid metabolism. It has been found that the risk of CVD in MAFLD combined with diabetes is 2.2 times higher than that in patients with diabetes only.⁸ A meta-analysis inclusive of 38 observational studies indicated that patients with moderate to severe hepatic steatosis detected by liver ultrasound had significantly higher rates of clinical and subclinical atherosclerotic cardiovascular disease (ASCVD) than those with mild hepatic steatosis.⁹ This underscores the potential role of MAFLD as a significant risk factor for CVD.

A substantial proportion of individuals diagnosed with MAFLD, categorized as lean/normal weight MAFLD, exhibited a normal body mass index (BMI). The proportion of lean/normal weight MAFLD varied significantly worldwide, ranging from 5% to 26% across different research groups, constituting 15%–50% of MAFLD cases.¹⁰ Despite having a lower BMI, lean/normal weight MAFLD is not a benign phenotype. Increasing research demonstrated that lean/normal weight MAFLD usually presented with more severe metabolic dysregulation and was strongly associated with increased CVD risk. The CVD risk in lean/normal weight MAFLD was notably higher than in healthy controls,¹¹ and, in certain studies, even surpassed that of overweight/obese MAFLD.^12,13

Although low BMI and metabolic dysregulation are independently associated with increased CVD risk, the combined influence of these factors on CVD risk has not been fully determined. Therefore, this study focuses on ASCVD, the most prevalent CVD, to investigate the risk of high adverse cardiovascular events (HACE) in different MAFLD subgroups characterized by different BMI and metabolic states. The study aims to clarify the relationship between MAFLD subgroups and HACE, providing a theoretical foundation for mitigating the burden of MAFLD-related CVD complications.

Materials and methods

Study population

Individuals who underwent routine health checks at the Physical Examination Center of The Third Hospital of Hebei Medical University, provided informed consent between June 2017 and January 2023, were prospectively selected in this study, with baseline data collected for longitudinal risk assessment of ASCVD. The study excluded individuals who were: (1) aged <20 or > 90 years; (2) pregnancy or breastfeeding; (3) inadequate data for MAFLD and HACE classification (approximately 1.0%).

The Ethics Committee of The Third Hospital of Hebei Medical University granted ethical approval for the study (Number HB/KY-06-06/2.0), which adhered to the tenets of the Declaration of Helsinki. Informed consent was obtained from all participants, and patient data were identified.

Definition and classification of patients with MAFLD

The diagnosis of MAFLD adhered to the Asian Pacific Association for the Study of the Liver clinical practice guidelines for the diagnosis and management of MAFLD.¹⁴ The diagnosis of MAFLD necessitates that participants fulfill specific criteria, which include: (a) the hepatic steatosis detected by liver ultrasound and iLivTouch; (b) the presence of at least one of the following three criteria: (1) being overweight/obese; (2) type 2 diabetes mellitus (T2DM); (3) clinical evidence of metabolic dysfunction. Metabolic dysfunction was identified if participants exhibited at least two of the following metabolic risk factors: (1) waist circumference ≥90 cm in men and ≥80 cm in women; (2) systolic blood pressure (SBP) ≥130 mmHg or diastolic blood pressure (DBP) ≥85 mmHg or in the treatment with antihypertensive medications; (3) plasma triglyceride (TG) ≥1.70 mmol/L and/or high-density lipoprotein cholesterol (HDL-C) <1.0 mmol/L for men and <1.3 mmol/L for women or in the treatment with lipid-lowering medications; (4) fasting plasma glucose (FPG) levels between 5.6 and 6.9 mmol/L or glycosylated hemoglobin (HbA1c) between 5.7% and 6.4%.

Participants possessing fewer than two metabolic risk factors were classified as metabolic health, whereas those with more than two metabolic risk factors were classified as metabolic dysfunction.¹⁰ Furthermore, participants with BMI of <23 kg/m² were defined as lean/normal weight and ≥23 kg/m² as overweight/obese. Based on their BMI and metabolic status, the participants were divided into four subgroups: metabolic healthy lean/normal weight/normal weight MAFLD (MHL), metabolic healthy overweight/obese MAFLD (MHO), metabolic dysfunctional lean/normal weight MAFLD (MDL), and metabolic dysfunctional overweight/obese MAFLD (MDO).^10,15

Noninvasive scoring systems play important roles in the assessment of MAFLD and CVD risk⁵: Significant liver fibrosis (SLF) was defined as the cut-off values of Fibrosis-4 Index (FIB-4) ≥1.3 or non-alcoholic fatty liver disease fibrosis score (NFS) ≥−1.455.^16,17

Clinical and laboratory parameters

Clinical data were collected, including age, gender, height, weight, waist circumference, SBP, DBP, past medical history, family history, smoking history, and alcohol consumption. BMI = weight (kg)/height (m)² (kg/m²). Peripheral blood tests, including white blood cell count (WBC), red blood cell count (RBC), platelet count (PLT), and hemoglobin (HGB), were conducted using the Beckman Coulter LH 750 Hematology Analyzer. Biochemical analyses of serum albumin (ALB), alanine aminotransferase (ALT), aspartate aminotransferase, total bilirubin, gamma-glutamyl transpeptidase, total cholesterol, TG, HDL-C, low-density lipoprotein cholesterol (LDL-C), FPG, serum urea, serum creatinine (CREA), and serum uric acid (UA) were performed using the Olympus AU2700 Automatic Biochemistry Analyzer.

Ultrasound imaging and ultrasound attenuation parameters

Participants were instructed to fast for 8–12 h prior to the tests. Hepatic ultrasonography and fat ultrasound attenuation measurements were uniformly conducted by experienced ultrasonographists using the Philips HD15 Color Ultrasound Diagnostic System (Hamburg, Germany) and the iLivTouch system (Wuxi Hisky Medical Technologies Co., Ltd, Wuxi, China). Features such as diffuse hyperechogenicity of the liver parenchyma, blurring of the intrahepatic vessels, and posterior attenuation were used to confirm fatty liver. The threshold for diagnosis was established at an ultrasound attenuation parameter of ≥244 dB/m.

ASCVD risk assessment

Based on the flowchart of relevant guidelines for primary prevention risk assessment in Chinese adults,¹⁸ the participants were divided into extreme risk of ASCVD, high-risk of ASCVD, high grade of the 10-year ASCVD risk, high grade of the risk of ASCVD for the rest of life, low grade of the risk of ASCVD for the rest of life, and low grade of the 10-year ASCVD risk. The HACE is defined as a composite event by incorporating the extreme risk of ASCVD, high-risk of ASCVD, high grade of the 10-year ASCVD risk, and high grade of the risk of ASCVD for the rest of life.

Statistical analysis

The statistical analysis was performed using SPSS 26.0, (IBM, Armonk, NY, USA), with statistical significance set at p < 0.05. Continuous variables were described as median (interquartile range), and non-normal distribution was confirmed by the Shapiro–Wilk test (p ⩽ 0.05). For two-group comparisons, the Mann–Whitney U test was used. For four-group comparisons, a Kruskal–Wallis test was first performed; if significant overall differences were found (p < 0.05), pairwise comparisons were conducted using Dunn’s test with Bonferroni correction (significance level α = 0.05/6). Binary categorical variables were presented as frequencies and percentages (n, %). The continuity-corrected Chi-square test was applied for two-group comparisons. For four-group comparisons, Pearson’s Chi-square test was first used to assess overall differences; if significant (p < 0.05), six pairwise comparisons were performed with Bonferroni-corrected significance level (α = 0.05/6). Binary logistic regression identified risk factors for HACE, including Odds ratios (ORs) with 95% confidence intervals (95% CIs).

Results

Population characteristics and clinical data of MAFLD subgroups

A total of 23,945 participants were recruited for our cohort. The observed proportion of MAFLD was 28.98% (6939/23,945). The proportions of MHL (n = 185), MHO (n = 2046), MDL (n = 310), and MDO (n = 4038) subgroups were 0.77%, 10.05%, 1.29%, and 16.86%, respectively (Figure 1).

Figure 1.

Flowchart of the study population enrollment.

The distribution of MAFLD prevalence across different age groups stratified by sex revealed an increasing trend of MHL and MDL groups with age, peaking at 50–59 years, followed by a subsequent decline. The peak prevalence of MHO and MDO group achieved at 30–39 years. In the most age groups of MHL and MDL, the prevalence rate was higher in female patients than in male patients, while an opposite result was obtained in MHO and MDO groups. However, in the group aged ≥70 years old for MHO and ≥60 years old for MDO, female patients outnumbered male patients (Figure 2). When stratified by metabolic status, the MHL and MDL groups had a higher median age (MHL: 47.00 vs MHO: 39.00 years old, p < 0.001; MDL: 57.00 vs MDO: 48.00 years old, p < 0.001), and had a lower proportion of males (MHL: 38.92% vs MHO: 67.58, p < 0.001; MDL: 37.10% vs MDO: 67.76%, p < 0.001) correspondingly compared with the MHO and MDO groups. When classified by BMI, the MDL and MDO groups had a higher median age than MHL and MHO groups, and there were no significant gender differences observed (Table 1).

Figure 2.

Age trend of MAFLD proportion stratified by gender in subgroups.

Table 1.

Comparison of characteristics between MAFLD subgroups.

Characteristic	Total MAFLD	MHL	MHO	MDL	MDO	p Value
Among total, n (%)	6939 (28.98%)	185 (0.77)	2406 (10.05)	310 (1.29)	4038 (16.86)	–
Among lean/normal weight MAFLD (%)	–	1.89	–	3.16	–	–
Among overweight/obese MAFLD (%)	–	–	17.02	–	28.56	–
Among MAFLD (%)	–	2.67	34.67	4.47	58.19	–
Age (year)	45.00 (35.00–56.00)	47.00 (35.00–56.00)	39.00 (33.00–49.00)^a	57.00 (47.75–65.00)^ab	48.00 (37.00–58.00)^bc	<0.001
Males	4549 (65.56)	72 (38.92)	1626 (67.58)^a	115 (37.10)^b	2736 (67.76)^ac	<0.001
BMI (kg/m²)	26.64 (24.80–28.70)	22.15 (21.48–22.66)	26.49 (24.86–28.37)^a	22.37 (21.55–22.78)^b	27.09 (25.39–29.07)^abc	<0.001
SBP (mmHg)	129.00 (119.00–139.00)	121.00 (112.00–127.00)	121.00 (114.00–129.00)	135.00 (126.00–144.00)^ab	134.00 (125.00–143.00)^ab	<0.001
DBP (mmHg)	80.00 (72.00–87.00)	72.50 (66.00–80.00)	75.00 (69.00–81.00)	81.00 (74.00–87.00)^ab	83.00 (75.00–89.00)^ab	<0.001
UAP (dB/m)	288.25 (264.00–312.20)	242.45 (222.45–274.19)	284.15 (262.93–301.53)^a	266.30 (242.25–277.88)^b	294.90 (269.85–318.25)^abc	<0.001
LSM (kPa)	5.60 (4.70–6.70)	5.35 (4.55–9.77)	5.60 (4.70–6.50)	5.25 (4.80–6.18)	5.60 (4.70–6.80)	0.587
Waist circumference (cm)	93.00 (87.00–99.00)	81.00 (76.84–84.00)	89.00 (84.00–95.25)^a	86.00 (81.00–91.00)^b	94.00 (90.00–100.00)^abc	<0.001
Drinking history (n, %)	1344 (19.37)	29 (15.68)	860 (35.74)^a	70 (22.58)^b	1535 (38.01)^ac	<0.001
Smoking history (n, %)	2494 (35.94)	25 (13.51)	455 (18.91)	57 (18.39)	807 (19.99)	0.135
WBC (×10⁹/L)	6.29 (5.40–7.27)	5.88 (4.79–6.82)	6.15 (5.28–7.11)^a	6.23 (5.21–7.23)	6.41 (5.50–7.38)^abc	<0.001
RBC (×10⁹/L)	4.86 (4.53–5.18)	4.65 (4.36–4.92)	4.88 (4.54–5.17)^a	4.59 (4.35–4.86)^b	4.87 (4.56–5.22)^ac	<0.001
PLT (×10⁹/L)	234.00 (197.60–273.70)	231.40 (199.80–271.10)	236.80 (200.40–275.70)	234.20 (191.40–281.35)	231.95 (196.25–271.78)^b	0.032
HGB (g/L)	149.70 (139.00–158.60)	140.40 (132.90–151.85)	149.80 (138.65–158.05)^a	140.60 (132.40–150.30)^b	150.65 (140.20–159.50)^abc	<0.001
ALB (g/L)	47.49 (45.69–49.33)	47.64 (45.76–49.43)	47.31 (45.63–49.15)	47.24 (45.23–49.30)	47.64 (45.78–49.40)^b	0.004
ALT (U/L)	27.00 (18.00–41.00)	22.00 (16.00–39.00)	25.00 (17.00–37.00)	21.00 (16.00–32.00)^b	28.00 (19.00–43.25)^abc	<0.001
AST (U/L)	21.00 (17.00–26.00)	20.00 (16.00–26.00)	20.00 (17.00–24.00)	20.00 (17.00–24.25)	21.00 (18.00–27.00)^bc	<0.001
TBIL (µmol/L)	14.96 (11.62–19.14)	13.89 (10.98–18.09)	15.50 (12.08–19.85)^a	14.36 (11.38–18.38)^b	14.70 (11.46–18.79)^b	<0.001
GGT (U/L)	32.00 (22.00–48.00)	28.00 (20.00–63.75)	29.00 (20.00–42.00)	26.00 (19.00–38.00)	34.00 (24.00–49.00)^bc	0.024
TC (mmol/L)	5.10 (4.50–5.76)	5.11 (4.37–5.95)	5.02 (4.48–5.60)	5.19 (4.48–5.98)^b	5.16 (4.54–5.83)^b	<0.001
TG (mmol/L)	1.78 (1.30–2.49)	1.70 (1.19–2.29)	1.38 (1.08–1.69)^a	2.02 (1.52–2.77)^ab	2.06 (1.57–2.83)^ab	<0.001
LDL-C (mmol/L)	3.09 (2.65–3.57)	3.15 (2.54–3.75)	2.99 (2.59–3.42)	3.11 (2.61–3.77)^b	3.14 (2.70–3.66)^b	<0.001
HDL-C (mmol/L)	1.23 (1.09–1.39)	1.22 (1.10–1.41)	1.28 (1.15–1.43)^a	1.26 (1.08–1.45)	1.19 (1.06–1.35)^bc	<0.001
FPG (mmol/L)	5.55 (5.17–6.14)	5.64 (5.18–5.85)	5.25 (4.98–5.49)^a	6.20 (5.41–7.49)^ab	5.81 (5.37–6.52)^abc	<0.001
UREA (mmol/L)	4.82 (4.09–5.66)	4.72 (3.92–5.64)	4.77 (4.03–5.57)	4.87 (4.09–5.80)	4.87 (4.14–5.71)^b	0.001
CREA (µmol/L)	74.16 (63.76–83.79)	67.55 (59.59–77.91)	75.97 (64.39–84.74)^a	67.15 (60.88–76.68)^b	73.99 (64.09–83.72)^abc	<0.001
UA (µmol/L)	385.00 (323.00–451.00)	333.50 (284.25–402.75)	385.00 (326.50–451.00)^a	357.00 (291.75–400.75)^b	390.00 (327.00–459.00)^ac	<0.001
FIB-4 ≥1.3	1157 (16.84)	44 (24.58)	248 (10.33)^a	93 (31.74)^b	772 (19.32)^bc	<0.001
NFS ≥1.455	1583 (22.89)	24 (13.48)	261 (10.85)	104 (35.25)^ab	119 (29.57)^ab	<0.001

Values were shown as n (%) or medians with IQRs.

Versus MHL, p < 0.05; ^bversus MHO, p < 0.05; ^cversus MDL, p < 0.05.

ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; CREA, serum creatinine; DBP, diastolic blood pressure; FIB-4, fibrosis-4 index score; FPG, fasting plasma glucose; GGT, gamma-glutamyl transpeptidase; HDL-C, high-density lipoprotein cholesterol; HGB, hemoglobin; IQR, interquartile ranges; LDL-C, low-density lipoprotein cholesterol; LSM, liver stiffness measurement; MAFLD, metabolic-associated fatty liver disease; MDL, metabolic dysfunctional lean/normal weight MAFLD; MDO, metabolic dysfunctional overweight/obese MAFLD; MHL, metabolic healthy lean/normal weight MAFLD; MHO, metabolic healthy overweight/obese MAFLD; NFS, non-alcoholic fatty liver disease fibrosis score; PLT, platelet count; RBC, red blood cell count; SBP, systolic blood pressure; TBIL, total bilirubin; TC, total cholesterol; TG, triglyceride; UA, serum uric acid; UAP, ultrasonic attenuation parameters; UREA, serum urea; WBC, white blood cell count.

When pairwise comparing the groups under the same metabolic conditions, the MHO and MDO groups showed higher levels of obesity indicators (BMI (26.49, 27.90 vs 22.15, 22.37) and waist circumference (89.00, 94.00 vs 81.00, 86.00 cm)), alcohol consumption ratio (35.74%, 38.01% vs 15.68%, 22.58%), CREA (75.97, 73.99 vs 67.55, 67.15 µmol/L), and UA (385.00, 390.00 vs 333.50, 357.00 µmol/L) levels compared to MHL and MDL groups, while the proportion of T2DM (2.95%, 23.01% vs 10.81%, 45.81%), higher TG (0.54%, 15.53% vs 24.31%, 40.92%), levels of FPG (5.25, 5.81 vs 5.64, 6.20 mmol/L), and proportions of abnormally high levels of LDL-C (5.27%, 11.59% vs 11.35%, 15.21%) were lower in MHO and MDO groups. When pairwise comparing the groups under the same BMI range, the MDL and MDO groups exhibited higher proportion of T2DM, hypertension (HTN) compared with MHL and MHO groups, respectively (62.26%, 60.25% vs 12.97%, 15.42%), all p < 0.05 (Table 1 and Figure 3).

Figure 3.

Proportion of metabolic conditions in MAFLD subgroups.

HACE risk assessment in different MAFLD subgroups and liver fibrosis severity

The number (proportion) of HACE in MHL, MHO, MDL, and MDO subgroups was 61 (32.97%), 634 (26.35%), 211 (68.06%), and 2163 (53.57%), respectively. The HACE proportions in MDL and MDO groups were higher than those in MHL and MHO groups, respectively (p < 0.001). The HACE proportions in MHL and MDL groups were correspondingly higher than that in MHO and MDO groups (p < 0.05; Figure 4).

Figure 4.

ASCVD risk stratification in MAFLD subgroups.

Compared the HACE proportions between different subgroups with SLF defined by FIB-4, the patients of MDL and MDO had a higher proportion of HACE than MHL and MHO patients, respectively (81.72%, 80.96% vs 34.09%, 60.48%; all p < 0.001). The result was similar when SLF was defined by NFS (90.38%, 81.57% vs 54.17%, 60.54%; all p < 0.001). Among different subgroups with non-SLF (NSLF), the MDL group had a higher HACE proportion as indicated by FIB-4 (61.00% vs 31.85%, 22.43%, 46.93%; all p < 0.001) or NFS (55.50% vs 29.22%, 21.19%, 41.81%; all p < 0.001; Supplemental Figure 1).

Furthermore, we compared the HACE proportions between patients with NSLF and SLF in the four subgroups. When SLF was assessed by NFS, the proportions of HACE in SLF patients were significantly higher than NSLF patients within MHL and MDL subgroups (p < 0.001). The HACE proportions between NSLF and SLF patients in MHO and MDO were similar. When the SLF was assessed by FIB-4, there was no significant difference of HACE proportions between SLF and NSLF patients in all the four subgroups.

Correlation analysis between MAFLD subgroups and HACE

To mitigate the influence of multicollinearity, metabolic factors and BMI were not included in this analysis. Univariate regression analysis showed that MAFLD subgroups, age, smoking history, drinking history, SLF were associated with the occurrence of HACE in MAFLD (Supplemental Table 1). When further adjusted for gender, age, smoking history, drinking history, and SLF, we found that MAFLD subgroups were an independent risk factor for HACE. The MDL group exhibited the highest risk of SLF (OR: 2.758, 95% CI: 1.710–4.449 SLF assessed by FIB-4; OR: 2.684, 95% CI: 1.663–4.332 SLF assessed by NFS), followed by the MDO group (OR: 2.399, 95% CI: 1.638–3.514 SLF assessed by FIB-4; OR: 2.247, 95% CI: 1.534–3.292 SLF assessed by NFS). No significant risk difference was observed between the MHO and MHL groups (Figure 5).

Figure 5.

The risk factors for HACE in MAFLD subgroups. Model 1: adjusted for age, smoking history, drinking history, and SLF (assessed by FIB-4). Model 2: adjusted for the same factors as Model 1, and SLF was assessed by NFS.

For patients <65 years old, the univariate regression analysis showed that MDL and MDO subgroups had higher risk of HACE than MHL subgroup (MDL: OR = 3.545, 95% CI: 2.284–5.503, p < 0.001; MDO: OR = 2.341, 95% CI: 1.642–3.338, p < 0.001). Gender, age, smoking history, drinking history, and SLF were identified as risk factors of HACE. After adjusted for age, gender, smoking history, drinking history, and SLF (assessed by FIB-4 or NFS), the multivariate regression analysis showed the similar result. For patients ≥65 years old, the univariate regression analysis also showed that MDL and MDO subgroups had higher risk of HACE than MHL group (MDL: OR = 11.833, 95% CI: 2.724–51.411, p = 0.001; MDO: OR = 3.721, 95% CI: 1.451–9.543, p = 0.006), with age and SLF assessed by NFS identified as risk factors. The multivariate regression analysis showed the similar results after adjusted for age and SLF (Supplemental Table 2).

Next, we analyzed the effect of gender difference among the MAFLD subgroups. For males, the univariate regression analysis showed that MDL and MDO subgroups carried a higher risk of HACE in comparison to MHL subgroup (MDL: OR = 3.789, 95% CI: 2.007–7.154, p < 0.001; MDO: OR = 2.421, 95% CI: 1.464–4.002, p = 0.001). Factors such as age, smoking history, drinking history, and SLF were risk factors of HACE. Upon adjusting for age, smoking history, drinking history, and SLF, the multivariate regression analysis showed that MDO subgroup presented a higher risk but the MHO and MDL subgroups did not exhibit statistical differences compared with MHL subgroup. In the case of females, the MHO group had a lower risk of HACE, while the MDL and MDO subgroups had higher risks than MHL group (MHO: OR = 0.623, 95% CI: 0.402–0.964, p = 0.033; MDL: OR = 4.656, 95% CI: 2.795–7.754, p = 0.001; MDO: OR = 3.721, 95% CI: 1.451–9.543, p = 0.006). Age and SLF assessed by FIB-4 were deemed as risk factors of HACE. Adjusted for age and SLF, the multivariate regression analysis showed that MDL and MDO subgroups had higher risks but the MHO subgroup had no statistical differences from MHL subgroup (Supplemental Table 3).

Discussion

Based on cross-sectional data from urban populations in northern China, this study examined the risk factors associated with the occurrence of high risk of HACE in MAFLD. Notably, the MDL group exhibited the highest proportion of HACE, followed by the MDO, MHL, and MHO groups. These findings implied potential variations in HACE risks across distinct subtypes of MAFLD. Furthermore, a significant association was observed between HACE and the presence of lean/normal weight status and/or metabolic dysregulation, underlining the importance of classifying MAFLD subgroups based on BMI and metabolic status to better assess CVD risk in individuals with MAFLD.

MAFLD is a complex multi-system metabolic disorder, primarily affecting the liver and intertwining with various diseases in multiple organs. Among these complications, CVD emerges as the most significant and the leading cause of death in patients with MAFLD.¹⁴ During the progress of MAFLD, the interplay of insulin resistance, abnormal fat deposition, and oxidative stress can trigger low-grade chronic inflammation, endothelial dysfunction, and a hypercoagulable state, which may contribute to the acceleration of atherosclerosis and CVD. This could elucidate the significantly increased risk of incidence and proportion of non-fatal and fatal CVDs observed in MAFLD compared to healthy controls.¹⁹

Currently, there are some internationally recognized ASCVD risk assessment models, such as Framingham model, SCORE model, QRISK score, and PCE model. However, the aforementioned models established based on European and American data may not be suitable for Chinese due to differences in the epidemiological characteristics of CVD and risk factors between the Chinese and Western populations.^20–22 To enhance the accuracy of the CVD risk assessment in Chinese population, we chose the Chinese 10-year ASCVD incidence risk assessment scheme, which draws from extensive longitudinal data from Chinese CVD cohorts.¹⁸

Remarkably, our research conducted pairwise comparisons among the four subgroups. The MHL and MDL groups had higher mean age, heightened rates of morbidities in female, higher proportions of metabolic comorbidities and SLF, along with abnormal levels of metabolic markers. Currently, the distribution characteristics of gender and age in lean/normal weight MAFLD have shown inconsistencies in different studies, with the male-to-female ratio fluctuating between 0.42 and 1.35.²³ Some research found that patients with lean/normal weight MAFLD were usually older, and more likely to be female compared with those classified as overweight/obese MAFLD.^24–26 These results align closely with our findings.

Within the same BMI group, the MDL and MDO groups had a higher mean age and exhibited a higher risk of HACE compared to the MHL and MHO groups. This may be related to the positive correlation between metabolic dysregulation and age. The age-related increase in metabolic dysregulation, characterized by declining levels of sex hormones and gradual deterioration of physiological functions, leads to a decreased capacity for metabolic substances. Additionally, chronic diseases like T2DM, HTN, and dyslipidemia have an increased incidence with advancing age.

SLF is an independent risk factor for both liver-related and non-liver-related mortality in MAFLD, a fact confirmed by studies on patients diagnosed with SLF through either liver biopsy or non-invasive serological diagnostic models.^27–32 In our study, when SLF was assessed by NFS, there was an increased proportion of HACE of SLF patients compared to NSLF patients in each subgroup. This may stem from the progression of MAFLD-related liver fibrosis, a consequence of long-term exposure to cardiovascular metabolic risk factors. These risk factors are closely associated with systemic insulin resistance, mild inflammation, and increased oxidative stress, all of which can exacerbate hepatocyte damage through various pathways, triggering the activation of hepatic stellate cells and fibrosis. Therefore, the burden of liver fibrosis can reflect the pathogenesis and severity of CVD to some degree.^33,34 Therefore, we suggest that the noninvasive scoring systems (such as FIB-4) be used as a supplementary indicator for risk stratification, independent of traditional factors.

In addition, metabolic abnormalities, particularly the disturbances in glucose metabolism, are classical and crucial risk factors for ASCVD. The glucose disturbances have a bidirectional relationship with MAFLD, promoting the occurrence and progression of MAFLD through various complex mechanisms. In our study, the proportion of T2DM and FPG levels was higher in the metabolic dysfunctional groups compared to the metabolic healthy groups. Disturbances in glucose metabolism can lead to endothelial dysfunction and oxidative stress, resulting in inflammation and thrombosis, promoting the risk of atherosclerosis and ASCVD.⁵ Additionally, a study found that insulin resistance was an independent risk factor for liver fibrosis in MAFLD, and the insulin resistance increased the risk of ASCVD by accelerating the development of liver fibrosis.³⁵ A nationwide population-based cohort study from Korea found that patients combined with T2DM and MAFLD had the highest cardiovascular-specific mortality rate.¹³ Another study including middle-aged individuals without a history of CVD found the similar result after a 10-year follow-up.³²

Notably, even in the groups within the same metabolic conditions, lean/normal weight MAFLD showed a higher proportion of metabolic comorbidities and metabolic indicator levels than overweight/obese MAFLD, thus challenging the prevailing notion that lean/normal weight MAFLD had a less severe metabolic profile than overweight/obese MAFLD.³⁶ A large prospective cohort study conducted by Nabi et al.³⁷ found that lean/normal weight MAFLD had a higher prevalence of late fibrosis and elevated ALT at baseline, and presented with more aggressive liver disease progression and a higher risk of death compared with overweight/obese MAFLD. The research from the NHANES (National Health and Nutrition Examination Survey) cohort³⁸ and the French CONSTANCES cohort¹³ showed similar results. In our research, we found the similar trend of liver fibrosis but the opposite trend of ALT levels. Currently, the relationship between ALT and lean/normal weight or overweight/obese MAFLD was not clear. A research inclusive of 2247 Chinese MAFLD showed that obese MAFLD had higher ALT level than lean/normal weight MAFLD (29 vs 20 U/L).³⁹ The differences in baseline ALT levels may be due to the heterogeneity of study populations.

Additionally, the phenotype of lean/normal weight MAFLD may associate with a higher risk of ASCVD. Multiple studies have confirmed that lean/normal weight MAFLD is an independent risk factor for CVD.⁵ It is often associated with low body weight and sarcopenia. The decreased skeletal muscle mass can lead to impaired glucose metabolism, worsened insulin resistance, and decreased exercise tolerance, promoting weight gain and exacerbating insulin resistance. In contrast, the large adipose tissue in overweight/obese MAFLD may provide protection against CVD and reverse disease progression.⁴⁰ Furthermore, BMI may not accurately reflect differences in fat distribution and function. Increased visceral fat is positively correlated with hyperinsulinemia and insulin resistance, while subcutaneous fat in the lower limbs may have a protective effect against lipotoxicity, and reduce the risk of ASCVD.⁴¹ Finally, clinicians often underestimated the severity of lean/normal weight MAFLD because of lean/normal weight physiques of patients, which can result in missing the optimal intervention window.

There are some limitations of this study. Firstly, the investigation of medical history was based on self-reports from participants during health examinations, potentially veil the true association between MAFLD and ASCVD. Secondly, this study was based on health examination for urban populations dwelling in North China, which required a caution to extrapolate our findings to other populations. Future researches are recommended to investigate more diverse populations to strengthen the generalizability of findings. Thirdly, reliance on non-invasive fibrosis scores rather than direct histological evaluations may not provide the most accurate diagnosis of hepatic fibrosis. Fourthly, it lacks long-term follow-up data on cardiovascular events among participants, making it impossible to directly assess the incidence of CVD and prognosis in different MAFLD subgroups over the long term. Fifthly, as a cohort study, while it is possible to observe the association between MAFLD subgroups and CVD risk, it cannot directly prove causality. Further intervention studies or randomized controlled trials are needed to verify the causality of these associations.

Conclusion

In summary, stratified management based on distinct metabolic conditions and BMI is crucial for improving prognosis and impeding the disease progression. It is recommended to incorporate the MAFLD subgroup analysis into the CVD risk assessment process, emphasizing intensified intervention for individuals with metabolic dysfunction (especially the lean/normal weight type).

Supplemental Material

sj-docx-1-taj-10.1177_20406223251378867 – Supplemental material for The relationship between MAFLD and CVD in a health check-up Chinese population: a prospective cohort study

Supplemental material, sj-docx-1-taj-10.1177_20406223251378867 for The relationship between MAFLD and CVD in a health check-up Chinese population: a prospective cohort study by Yao Dou, Jiawei Cui, Qi Gu, Xiwei Yuan, Mengmeng Hou, Wenjing Ni, Chen Dong, Chudi Chang, Jinhua Shao, Qiuling Wang, Jie Li, Liang Qiao and Yuemin Nan in Therapeutic Advances in Chronic Disease

Footnotes

Acknowledgements

All authors would like to thank all the staff for their great support and conscientious work during this study.

Declarations

ORCID iDs

Yao Dou

Jiawei Cui

Xiwei Yuan

Mengmeng Hou

Chen Dong

Chudi Chang

Jie Li

Liang Qiao

Yuemin Nan

Supplemental material

Supplemental material for this article is available online.

References

Younossi

Golabi

Paik

, et al. The global epidemiology of nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH): a systematic review. Hepatology 2023; 77(4): 1335–1347.

Stefan

Yki-Järvinen

Neuschwander-Tetri

BA.

Metabolic dysfunction-associated steatotic liver disease: heterogeneous pathomechanisms and effectiveness of metabolism-based treatment. Lancet Diabetes Endocrinol 2025; 13(2): 134–148.

Man

Deng

, et al. Prevalence of liver steatosis and fibrosis in the general population and various high-risk populations: a nationwide study with 5.7 million adults in China. Gastroenterology 2023; 165(4): 1025–1040.

Meyersohn

Mayrhofer

Corey

, et al. Association of hepatic steatosis with major adverse cardiovascular events, independent of coronary artery disease. Clin Gastroenterol Hepatol 2021; 19: 1480–1488.e14.

Zheng

Sechi

Navarese

, et al. Metabolic dysfunction-associated steatotic liver disease and cardiovascular risk: a comprehensive review. Cardiovasc Diabetol 2024; 23(1): 346.

Ciardullo

Monti

Perseghin

High prevalence of advanced liver fibrosis assessed by transient elastography among U.S. adults with type 2 diabetes. Diabetes Care 2021; 44: 519–525.

Kwok

Choi

Wong

, et al. Screening diabetic patients for non-alcoholic fatty liver disease with controlled attenuation parameter and liver stiffness measurements: a prospective cohort study. Gut 2016; 65: 1359–1368.

Zhou

, et al. Synergistic increase in cardiovascular risk in diabetes mellitus with nonalcoholic fatty liver disease: a meta-analysis. Eur J Gastroenterol Hepatol 2018; 30: 631–636.

Toh

JZK

Pan

Tay

PWL

, et al. A meta-analysis on the global prevalence, risk factors and screening of coronary heart disease in nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol 2022; 20(11): 2462–2473.e10.

10.

Eslam

El-Serag

Francque

, et al. Metabolic (dysfunction)-associated fatty liver disease in individuals of normal weight. Nat Rev Gastroenterol Hepatol 2022; 19: 638–651.

11.

Lee

Kim

, et al. Metabolic dysfunction-associated fatty liver disease and incident cardiovascular disease risk: a nationwide cohort study. Clin Gastroenterol Hepatol 2021; 19: 2138–2147.e10.

12.

Kim

Han

Lee

, et al. Cardiovascular risk is elevated in lean subjects with nonalcoholic fatty liver disease. Gut Liver 2022; 16: 290–299.

13.

Chung

Yoo

, et al. Lean or diabetic subtypes predict increased all-cause and disease-specific mortality in metabolic-associated fatty liver disease. BMC Med 2023; 21: 4.

14.

Eslam

Sarin

Wong

VW-S

, et al. The Asian Pacific Association for the study of the liver clinical practice guidelines for the diagnosis and management of metabolic associated fatty liver disease. Hepatol Int 2020; 14: 889–919.

15.

WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004; 363: 157–163.

16.

Sun

Cui

, et al. Comparison of FIB-4 index, NAFLD fibrosis score and BARD score for prediction of advanced fibrosis in adult patients with non-alcoholic fatty liver disease: a meta-analysis study. Hepatol Res 2016; 46: 862–870.

17.

Castera

Friedrich-Rust

Loomba

Noninvasive assessment of liver disease in patients with nonalcoholic fatty liver disease. Gastroenterology 2019; 156: 1264–1281.

18.

Chinese Society of Cardiology of Chinese Medical Association; Cardiovascular Disease Prevention and Rehabilitation Committee of Chinese Association of Rehabilitation Medicine; Cardiovascular Disease Committee of Chinese Association of Gerontology and Geriatrics, et al. Chinese guideline on the primary prevention of cardiovascular diseases. Chin J Cardiol 2020; 48: 1000–1038.

19.

Mantovani

Csermely

Petracca

, et al. Non-alcoholic fatty liver disease and risk of fatal and non-fatal cardiovascular events: an updated systematic review and meta-analysis. Lancet Gastroenterol Hepatol 2021; 6: 903–913.

20.

Wilson

D’Agostino

Levy

, et al. Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97: 1837–1847.

21.

Liu

Hong

D’Agostino

Sr , et al. Predictive value for the Chinese population of the Framingham CHD risk assessment tool compared with the Chinese Multi-Provincial Cohort Study. JAMA 2004; 291: 2591–2599.

22.

Liu

, et al. Estimation of 10-year risk of fatal and nonfatal ischemic cardiovascular diseases in Chinese adults. Circulation 2006; 114: 2217–2225.

23.

Cheng

Kao

Wang

CC.

The metabolic profiles and body composition of lean metabolic associated fatty liver disease. Hepatol Int 2021; 15: 405–412.

24.

Guo

, et al. Prevalence and risk factors of metabolic associated fatty liver disease in Xinxiang, China. Int J Environ Res Public Health 2020; 17: 1818.

25.

Ahmed

Gidener

Mara

, et al. Natural history of nonalcoholic fatty liver disease with normal body mass index: a population-based study. Clin Gastroenterol Hepatol 2022; 20: 1374–1381.e6.

26.

Fracanzani

Petta

Lombardi

, et al. Liver and cardiovascular damage in patients with lean nonalcoholic fatty liver disease, and association with visceral obesity. Clin Gastroenterol Hepatol 2017; 15: 1604–1611.e1.

27.

Han

Lee

Kim

, et al. Nonalcoholic fatty liver disease and sarcopenia are independently associated with cardiovascular risk. Am J Gastroenterol 2020; 115: 584–595.

28.

Tamaki

Kurosaki

Takahashi

, et al. Liver fibrosis and fatty liver as independent risk factors for cardiovascular disease. J Gastroenterol Hepatol 2021; 36: 2960–2966.

29.

Taylor

Bayliss

, et al. Association between fibrosis stage and outcomes of patients with nonalcoholic fatty liver disease: a systematic review and meta-analysis. Gastroenterology 2020; 158: 1611–1625.e12.

30.

Angulo

Kleiner

Dam-Larsen

, et al. Liver fibrosis, but no other histologic features, is associated with long-term outcomes of patients with nonalcoholic fatty liver disease. Gastroenterology 2015; 149: 389–397.e10.

31.

Dulai

Singh

Patel

, et al. Increased risk of mortality by fibrosis stage in nonalcoholic fatty liver disease: systematic review and meta-analysis. Hepatology 2017; 65: 1557–1565.

32.

Lee

Lim

Kim

, et al. Long-term cardiovascular outcomes differ across metabolic dysfunction-associated fatty liver disease subtypes among middle-aged population. Hepatol Int 2022; 16: 1308–1317.

33.

Zhou

X-D

Targher

Byrne

, et al. An international multidisciplinary consensus statement on MAFLD and the risk of CVD. Hepatol Int 2023; 17: 773–791.

34.

Targher

Byrne

Tilg

MASLD: a systemic metabolic disorder with cardiovascular and malignant complications. Gut 2024; 73(4): 691–702.

35.

Sohn

Kwon

Chang

, et al. Liver fibrosis in Asians with metabolic dysfunction-associated fatty liver disease. Clin Gastroenterol Hepatol 2022; 20: e1135–e1148.

36.

Chen

Esmaili

Rogers

, et al. Lean NAFLD: a distinct entity shaped by differential metabolic adaptation. Hepatology 2020; 71: 1213–1227.

37.

Nabi

Lapidus

Boursier

, et al. Lean individuals with NAFLD have more severe liver disease and poorer clinical outcomes (NASH-CO Study). Hepatology 2023; 78: 272–283.

38.

Dao

Nguyen

Ito

, et al. Prevalence, characteristics, and mortality outcomes of obese and nonobese MAFLD in the United States. Hepatol Int 2023; 17: 225–236.

39.

Guan

Zhen

, et al. Metabolic (dysfunction)-associated fatty liver disease in Chinese patients with type 2 diabetes from a subcenter of the national metabolic management center. J Diabetes Res 2022; 2022: 8429847.

40.

Joo

Kim

Interaction between sarcopenia and nonalcoholic fatty liver disease. Clin Mol Hepatol 2023; 29: S68–S78.

41.

Han

Lee

, et al. Anatomic fat depots and cardiovascular risk: a focus on the leg fat using nationwide surveys (KNHANES 2008–2011). Cardiovasc Diabetol 2017; 16: 54.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.06 MB