Obesity is an important determinant of severity in newly defined metabolic dysfunction-associated fatty liver disease

Ji Hye Huh , Kwng Joon Kim , , Seung Up Kim , Bong-Soo Ch , Byung-Wn Lee ,

a Division of Endocrinology and Metabolism, Department of Internal Medicine, Hallym University Sacred Heart Hospital, Anyang 14068, Korea

b Division of Geriatrics, Department of Internal Medicine, Yonsei University College of Medicine, Seoul 03722, Korea

c Division of Endocrinology and Metabolism, Department of Internal Medicine, Yonsei University College of Medicine, Seoul 03722, Korea

d Institute of Gastroenterology, Department of Internal Medicine, Yonsei University College of Medicine, Seoul 03722, Korea

Keywords:Metabolic dysfunction-associated fatty liver disease Metabolically-healthy obesity Liver fibrosis

ABSTRACT

Introduction

As per the current national and international guidelines, histological or imaging evidence of hepatic steatosis, after excluding secondary causes and excessive alcohol intake, is indicative of nonalcoholic fatty liver disease (NAFLD) [1–3] . However, there has been a serious criticism concerning the nomenclature of NAFLD regarding not only the prominent role that alcohol plays in the definition of NAFLD, but also the negative impacts of the nomenclature, which leads to the trivialization and stigmatization of the disease, as well as the impact on the consideration of this disease in health policies [4] .

Accordingly, after a consensus of international experts, a proposition has been made to change the name of the disease from NAFLD to metabolic dysfunction-associated fatty liver disease(MAFLD) [4] . The suggested diagnostic criteria for MAFLD include the presence of hepatic stenosis and one of the following three factors: overweight/obesity, type 2 diabetes mellitus(T2DM), and metabolic abnormalities such as hyperglycemia, dyslipidemia, and high blood pressure. According to these criteria,both obesity and metabolic abnormalities are the main factors contributing to MAFLD diagnosis. These criteria are well-defined and clear, given that they consider underlying metabolic abnormalities and recognize that MAFLD commonly coexists with other conditions [5] . Obesity and metabolic abnormalities are prevalent worldwide [6] , leading to an increased incidence of chronic metabolic disorders, including T2DM, NAFLD, and cardiovascular diseases [7–9] . Given that both obesity and metabolic abnormalities are risk factors modifiable via interventions, determining their clinical impact on severe fatty liver disease will help prevent the progression of MAFLD. We previously reported that obesity is more closely related to hepatic steatosis than to abnormalities in the general population [10] . However, there have been few studies that used the new diagnostic criteria of MAFLD until now. Moreover,the impact of obesity and metabolic dysfunction on MAFLD severity has not been fully elucidated.

Fig. 1. Study participant screening process and study groups. MAFLD: metabolic dysfunction-associated fatty liver disease; BMI: body mass index; CAP: controlled attenuation parameter.

To better understand the relationship between metabolic health/obesity status and MAFLD severity, we investigated the effects of metabolic abnormalities and obesity on hepatic steatosis and fibrosis using the controlled attenuation parameter (CAP) and liver stiffness measurement (LSM) values determined via transient elastography (TE) in patients with MAFLD [11] .

Methods

Patients

Individuals who underwent TE during their checkup at a center for health promotion and personalized medicine were included.The details are mentioned in a previous report [10] . Initially, 2521 individuals participated in the study ( Fig. 1 ). Participants were excluded if they had unreliable LSM values (n= 15) or incomplete information (n= 78) or did not meet the diagnostic criteria of MAFLD [CAP score＜238 dB/m (n= 1131); CAP score ≥238 dB/m with BMI＜23 kg/m2and no metabolic risk abnormalities and T2DM (n= 134)]. Thus, after excluding the abovementioned patients, 1163 participants aged ≥20 years who met diagnostic criteria of MAFLD were included in this study. Of the study participants, 60 (5.2%) had positive serologic markers for hepatitis B, and 10 (0.9%) had positive serologic markers for hepatitis C. Participants were categorized into four groups according to the presence or absence of metabolic abnormalities and their obesity status. All participants completed a questionnaire about their medical and/or surgical history, current alcohol consumption status, and smoking status.

Definitions

MAFLD was characterized by the presence of hepatic steatosis,as determined by TE, with one or more of the following conditions: (1) overweight/obesity [body mass index (BMI) ≥23 kg/m2,as per the Asia-Pacific criteria]; (2) T2DM; or (3) at least two of the metabolic abnormalities described in Table S1. Obesity was defined according to the Asia-Pacific BMI criteria (non-obese, BMI＜25 kg/m2; and obese, BMI ≥25 kg/m2), which are defined by the World Health Organization Western Pacific Region [12] . Metabolic abnormality was defined according to the National Cholesterol Education Program – Adult Treatment Panel III guidelines [13] . The waist circumference criterion was not used because of its collinearity with BMI. Participants who satisfied at least two of the following criteria were defined as having metabolic abnormalities,whereas those who did not were defined as not having metabolic abnormalities: (1) a systolic blood pressure ≥130 mmHg and/or a diastolic blood pressure ≥85 mmHg or ongoing anti-hypertensive treatment; (2) triglyceride level ≥150 mg/dL; (3) fasting glucose level ≥100 mg/dL or ongoing anti-diabetes treatment; and (4)high-density lipoprotein (HDL) cholesterol level＜40 mg/dL in males and＜50 mg/dL in females.

Based on these criteria, study participants were categorized into four groups: Group 1, comprising those who were non-obese and did not have metabolic risk abnormalities (BMI＜25 kg/m2and＜2 metabolic risk abnormalities); Group 2, comprising those who were non-obese and had metabolic risk abnormalities (BMI＜25 kg/m2and ≥2 metabolic risk abnormalities); Group 3, comprising those who were obese and did not have metabolic risk abnormalities (BMI ≥25 kg/m2and＜2 metabolic risk abnormalities); or Group 4, comprising those who were obese and had metabolic risk abnormalities (BMI ≥25 kg/m2and ≥2 metabolic risk abnormalities) [14] .

Measurements and key variables

Height and weight were measured while participants wore light clothing and were barefoot. BMI was calculated by dividing weight(kg) by height squared (m2). Blood pressure was obtained by averaging the results of three blood pressure recordings in a sitting position, with each recording measured after at least 5 min of rest. Blood samples were collected from each participant after overnight fasting. Fasting plasma glucose levels were measured using the glucose oxidase method. The plasma lipid profile and liver enzymes and uric acid levels were assayed using the Hitachi 7600 Auto Analyzer (Hitachi Instruments Service, Tokyo, Japan). Lowdensity lipoprotein (LDL) cholesterol levels were calculated using the Friedewald equation: LDL cholesterol (mg/dL) = total cholesterol (mg/dL) - HDL cholesterol (mg/dL) - triglyceride (mg/dL)/5.Glycated hemoglobin (HbA1c) levels were measured using highperformance liquid chromatography using VariantTMII Turbo (Bio-Rad Laboratories, Hercules, CA, USA).

Measurement of LSM and CAP using TE

The methods of the measurement of CAP and LSM measurements via TE [FibroScan 502 machine (Echosens, Paris, France)]have been described previously [ 11 , 15 ]. TE was performed by an experienced technician who has performed＞10 0 0 0 examinations. The technician was blinded to clinical subject data. Results were expressed in kPa for LSM and in dB/m for CAP. The ratio of the interquartile range (IQR) of LSM to the median LSM value was measured as an indicator of variability. Only cases with at least 10 valid acquisitions, a success rate of at least 60%, and an IQR-tomedian ratio of＜0.3 were considered reliable and used for statistical analysis [15] . We first used the M probe and if M probe failed,we employed XL probe to measure LSM. The CAP value was considered valid only when the LSM was reliable for the same signal at the same volume of liver parenchyma. In this study, a cutoff CAP value of 238 dB/m was used to define hepatic steatosis (TE-defined hepatic steatosis) based on a previous South Korean study [16] .CAP was scored according to steatosis grade using the following cutoffs: 238-259 dB/m for mild steatosis (S1), 260-292 dB/m for moderate steatosis (S2), and＞292 dB/m for severe steatosis(S3) [17] . Significant liver fibrosis was characterized by an LSM value of 7.0 kPa based on previous studies [ 18 , 19 ].

Statistical analysis

Continuous variables were expressed as mean ± standard deviation. Categorical variables were expressed as proportions (%),and values were compared using the Chi-squared test. Differences in demographic and biochemical characteristics between the four study groups, which were categorized by their metabolic and obesity status, were analyzed by one-way analysis of variance using Scheffe’s method for the post-hoc analysis. A multiple logistic regression analysis was performed to determine the odds ratios(ORs) of severe hepatic steatosis and significant liver fibrosis as assessed by TE in each group, with Group 1 as a reference group.Statistical analysis was performed using SPSS version 20.0 (IBM,Chicago, IL, USA). APvalue of＜0.05 was considered statistically significant.

Results

Baseline characteristics

The differences in clinical and biochemical characteristics between the study groups are presented in Table 1 . Overall, 295(25.4%) patients were categorized in the non-obese without metabolic risk abnormalities (Group 1), 244 (21.0%) in the nonobese with metabolic risk abnormalities (Group 2), 210 (18.1%) in the obese without metabolic risk abnormalities (Group 3), and 414(40%) in the obese with metabolic risk abnormalities (Group 4).Alanine aminotransferase (ALT) and gamma-glutamyl transferase(GGT) levels were significantly higher in obese participants with metabolic risk abnormality (Group 4 vs. the remaining groups).Prevalence of elevated ALT levels was significantly higher in patients with obesity than in those without obesity. The CAP scores were 260.4 ± 21.7, 268.5 ± 26.1, 275.0 ± 26.7, and 288.6 ± 34.5 dB/m in Groups 1, 2, 3 and 4, respectively (P＜0.001). The LSM values (kPa) were 3.8 ± 1.0, 4.0 ± 1.1, 4.6 ± 2.9, and 5.1 ± 2.8 kPa in Groups 1, 2, 3 and 4, respectively (P＜0.001). These findings show that the severity of hepatic steatosis and fibrosis was higher in the obese groups than that in the non-obese groups.

Distribution of CAP score and prevalence of hepatic steatosis with elevated liver enzyme levels across the study groups

Fig. 2 A presents the distribution of the CAP score among the four study groups. The proportion of participants with a lower grade of hepatic steatosis (S1 grade) was lower in Group 3 (obese without metabolic risk abnormalities) than in Group 2 (nonobese with metabolic risk abnormalities) (35.2% vs. 44.3%,P＜0.001). However, the proportions of participants with severe hepatic steatosis (S3 grade) were 10.2%, 14.3%, 22.9%, and 37.4% in Groups 1, 2, 3, and 4, respectively (Ptrend＜0.001). The proportion of participants with severe hepatic steatosis (S3 grade) was higher in Group 3 (obese without metabolic risk abnormalities) than that in Group 2 (non-obese with metabolic risk abnormalities) (22.9%vs. 14.3%,P= 0.01). Prevalences of hepatic steatosis with elevated ALT levels were 40.7%, 44.7%, 47.1%, and 63.3% in Groups 1, 2, 3,and 4, respectively (Ptrend＜0.001) ( Fig. 2 B). Obese participants with metabolic risk abnormalities (Group 4) showed the highest prevalence of severe hepatic steatosis (S3 grade) and elevated liver enzyme levels.

Association of metabolic abnormality and obesity status with severe hepatic steatosis and liver fibrosis

Table 2 shows the risks of severe hepatic steatosis and liver fibrosis as assessed by TE according to the presence of metabolic abnormalities and obesity. The unadjusted ORs [95% confidence interval (CI)] for severe hepatic steatosis were 1.4 8 (0.88-2.4 9), 2.62(1.59-4.30), and 5.29 (3.45-8.10) in Groups 2, 3, and 4, respectively(Ptrend＜0.001), with Group 1 as the reference group. After adjustments for age, sex, current smoking, alcohol intake, uric acid levels, ferritin levels, and diabetes, the adjusted ORs (95% CI) for severe hepatic steatosis were 1.07 (0.61-1.88), 2.43 (1.44-4.08), and 4.07 (2.56-6.48) in Groups 2, 3, and 4, respectively (Ptrend＜0.001).The fully adjusted ORs for significant hepatic fibrosis were 1.31(0.30-5.74), 4.70 (1.24-17.82), and 6.43 (1.88-22.02) in Groups 2, 3,and 4, respectively (Ptrend＜0.001) ( Table 2 ). There was no significant increase in the OR of both severe hepatic steatosis and significant liver fibrosis in the non-obese groups, even in the presence of metabolic abnormalities, whereas there was a significant increase in the OR in the obese groups. The OR of severe hepatic steatosis and significant liver fibrosis was the highest in Group 4.

Discussion

This study focused on the association of obesity and metabolic status with MAFLD severity. MAFLD was assessed by TE, which is a validated and non-invasive tool to assess significant liver fibrosis and hepatic steatosis severity. This study had two main findings.First, obesity was more significantly associated with severe hepatic steatosis and liver fibrosis than with other metabolic abnormalities. Second, the risk of severe hepatic steatosis and liver fibrosis in patients with MAFLD was immensely aggravated in the presence of both obesity and metabolic abnormalities. To the best of our knowledge, this study is the first to report the relationship ofmetabolic health and obesity with the severity of MAFLD, particularly liver fibrosis, as assessed by TE.

Table 1 Characteristics of patients with in metabolic dysfunction-associated fatty liver disease according to the presence of obesity and metabolic risk abnormalities.

Fig. 2. Distribution of controlled attenuation parameter (CAP) score grade (S1-3) ( A ) and prevalence of hepatic steatosis with elevated alanine aminotransferase levels ( B )in patients having metabolic dysfunction-associated fatty liver disease (MAFLD) with or without metabolic abnormalities and obesity. CAP grade: 238-259 dB/m for mild steatosis (S1), 260-292 dB/m for moderate steatosis (S2), and ＞ 292 dB/m for severe steatosis (S3). Elevated ALT levels were defined as levels ＞ 33 U/L for males and ＞ 25 U/L for females.

Table 2 Associations between metabolic abnormalities/obesity and transient liver elastography findings.

A new definition of MAFLD has been proposed by an international panel of experts, implying profound conceptual changes regarding the metabolism-related etiology and disease heterogeneity of MAFLD. Unlike the exclusive diagnosis of NAFLD, MAFLD is diagnosed based on etiology. The criterion is based on the co-existence of hepatic steatosis, which is determined by histological evidence,imaging, blood biomarkers, and scores, and the presence of one of the following metabolism-related criteria: overweight/obesity,T2DM, and metabolic abnormalities in the lean population [20] .This well-defined diagnostic approach would not only enable clinicians to optimally manage liver disease-related comorbidities, but also lower the risk of misdiagnosis and missed diagnosis among patients with hepatic steatosis. However, few studies have investigated the impact of different metabolism-related etiologies on MAFLD severity. Considering that an advanced stage of fatty liver disease (especially liver fibrosis) is a prognostic indicator of the development of liver-related or non-liver-related morbidity and mortality, interventions to prevent and/or delay fatty liver disease progression are necessary [3] . However, there are no proven effective medications that can delay or improve fatty liver disease in clinical practice. It is universally acknowledged that diet and lifestyle interventions are important parts of fatty liver disease therapy [21] .With the help of lifestyle interventions and even the use of several medications, including anti-obesity drugs, anti-hypertensive drugs,and anti-dyslipidemic drugs, body weight control and metabolic abnormalities may be improved. Therefore, we attempted to determine a dominant risk factor for severe fatty liver disease among modifiable risk factors, including obesity and metabolic abnormalities, in patients with MAFLD.

In this study, we demonstrated that severe hepatic steatosis and significant liver fibrosis were closely associated with obesity, independent of metabolic abnormalities. We also found that the risk of severe hepatic steatosis and significant liver fibrosis was remarkably higher in obese participants with metabolic abnormalities than in the participants in the remaining groups. However,non-obese participants, irrespective of presence of metabolic risk abnormalities, did not demonstrate a higher risk of severe hepatic steatosis and significant liver fibrosis. Moreover, in a multivariable regression analysis, the absolute values of the regression coeffi-cients (β) for CAP and LSM were the largest in obesity (β= 0.272 in CAP, 0.197 in LSM) among various metabolic parameters (Tables S2 and S3). Although T2D was also closely associated with CAP and LSM, the degree of association was relatively lower than the association between obesity and CAP and LSM value (β= 0.090 in CAP, 0.127 in LSM). These findings suggest that a healthy metabolic profile does not protect or attenuate the impact of obesity on fatty liver disease severity. Furthermore, we assume that, while determining a higher risk of liver fibrosis in patients with MAFLD, it may be important to particularly consider obesity status. Based on the findings of this study, obesity status may serve as a guide for clinicians when preparing surveillance and therapeutic strategies for severe hepatic steatosis and liver fibrosis in patients with MAFLD.

Using the definition of NAFLD, the association between obesity/metabolic health and fatty liver disease severity has been widely demonstrated in the literature. In line with our findings, recent studies have shown that BMI is positively associated with the worsening of non-invasive fibrosis markers, regardless of metabolic health status in patients with NAFLD [ 22 , 23 ]. However, these studies used noninvasive fibrosis markers and tools that were not appropriately validated for assessing liver fibrosis. Furthermore,these studies did not assess patients with MAFLD. In fact, a recent population-based study found that approximately 15% of patients with NAFLD who failed to meet the diagnostic criteria of MAFLD exhibited significantly lower non-invasive fibrosis scores than those who met the MAFLD criteria [24] . After analyzing all patients, including those without hepatic fibrosis, who underwent check-ups, we found that individuals with MAFLD had higher levels of CAP score and LSM than those without hepatic steatosis or those with fatty liver disease without metabolic association (Table S4),indicating that MAFLD itself is strongly related to MAFLD severity. Despite conceptual differences between NAFLD and MAFLD, to date, no study has investigated the association between metabolic health and obesity status and fatty liver disease severity in patients with MAFLD. Therefore, this study focused on the association between obesity, metabolic abnormalities and liver fibrosis,as assessed by TE, in patients with MAFLD. Although both obesity and metabolic health are important factors for MAFLD, we found that these factors differentially affected MAFLD severity, showing stronger effect of obesity than metabolic abnormalities on MAFLD severity. Our findings support the growing evidence on the heterogeneity of MAFLD regarding clinical presentation, disease course,and etiology.

This study has several limitations. First, the cross-sectional study design precludes solid conclusions regarding the causal relationship of obesity and metabolic health status with severe hepatic steatosis and significant liver fibrosis. Therefore, further studies with large samples and a well-balanced spectrum are needed to validate our data in a longitudinal manner. Second, regarding CAP and LSM, the exact cutoffs to optimize the sensitivity and specificity for predicting hepatic steatosis and significant liver fibrosis are not well delineated. Third, using the CAP score but not the conventional methods of liver biopsy or magnetic resonance imaging, to assess the severity of hepatic steatosis also may be a limitation. Moreover, as the LSM values assessed using TE would be influenced by high BMI, high levels of ALT, and cholestasis, assessing liver fibrosis using TE can be inaccurate in patients with extreme obesity or in those with liver enzyme abnormalities. However, these factors may not greatly influence the overall results of this study because most of our study participants did not have extreme obesity and liver enzyme abnormalities. Fourth, we were unable to investigate the association between abdominal obesity and MAFLD severity because of the lack of data on waist circumference.Furthermore, as this study did not collect data on the index of insulin resistance, we were unable to investigate the impact of insulin resistance on MAFLD severity. Finally, we were unable to collect data on other possible confounders, such as diet, the presence of other liver diseases, medications, and genetic predisposition that may influence fatty liver disease severity. However, given that the variables were obtained from health examination data, the individuals included in this study were relatively healthy; we believe that the impact of other possible confounders on the results may not be considerable. Despite these limitations, it is the first study to demonstrate an association between obesity and the risk of severe hepatic steatosis and liver fibrosis as assessed by TE in patients with MAFLD.

In conclusion, the findings of this study suggest that a healthy metabolic profile does not protect against severe hepatic steatosis or fibrosis in obese patients with MAFLD. In fact, it appears that obesity negatively impacts MAFLD ranging from hepatic steatosis to fibrosis. Thus, it is important to consider obesity status when predicting and designing strategies to treat and prevent hepatic steatosis and liver fibrosis in the general population.

Acknowledgments

None.

CRediT authorship contribution statement

Ji Hye Huh: Conceptualization, Formal analysis, Writing – original draft. Kwang Joon Kim: Data curation, Formal analysis, Writing – review & editing. Seung Up Kim: Resources, Writing – review& editing. Bong-Soo Cha: Project administration, Writing – review& editing. Byung-Wan Lee: Supervision, Validation, Writing – review & editing.

Funding

This research was supported by a grant from the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2020R1F1A1076198).

Ethical approval

This study was approved by the Institutional Review Board of the Severance Hospital (4-2020-1421) and was conducted in accordance with the ethical standards oftheDeclarationofHelsinki.

Competing interest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi: 10.1016/j.hbpd.2022.03.009 .