In this study of 9719 fatty liver patients, 79.7% was “both FLD”, which fulfills both diagnostic criteria of NAFLD and MAFLD. The new population (MAFLD only group; 12%) has more people than the missed population (NAFLD only group; 7.1%) after the change of diagnostic criteria from NAFLD to MAFLD. Compared with NAFLD patients, MAFLD patients had higher frequency of male gender, DM, and hypertension. Furthermore, several metabolic components including BMI, uric acid, glucose and lipid profiles were higher in MAFLD patients than NAFLD patients even a high degree of overlap between the two groups of patients. In addition, the new population had higher frequency of metabolic diseases; poor metabolic profiles; increased severity of hepatic steatosis and liver fibrosis; and risk of atherosclerosis than the missed population. Therefore, the new diagnostic criteria of MAFLD not only include more patients, but also high-risk patients of metabolic, liver and cardiovascular (CV) diseases, suggesting MAFLD may be a better nomenclature than NAFLD in clinical practice.
Fatty liver patients with alcohol or other known etiologies of chronic liver disease were excluded from the diagnosis of NAFLD. In contrast, the diagnosis of MAFLD includes fatty liver plus metabolic risk and permits to have other concomitant liver diseases.18 Two large population-based studies from Korea and China showed the added population had more people than the missed population after the change of disease name from NAFLD to MAFLD.19,20 However, another population-based study from NHANES-III database, a representative sample of United State (US) general population had inconsistent findings. MAFLD patients were diagnosed in 31.24% participants, while NAFLD in 33.23% amongst the overall population.21 Our large, population-based study from Taiwan showed the diagnostic criteria of MAFLD could include more patients than NAFLD.
The diagnostic criteria of MAFLD have two essential factors: fatty liver and metabolic dysfunction. The difference of diagnostic criteria between MAFLD and NAFLD is metabolic dysfunction and any of other concomitant liver diseases. A recent study from China showed MAFLD patients had greater proportions of DM and dyslipidemia, higher BMI, WC, blood glucose, and lipid levels than NAFLD patients.19 Our study also found higher frequency of metabolic diseases and abnormal metabolic factors including BMI, glucose, lipid, and uric acid metabolism in MAFLD patients, suggesting the definition of MAFLD can diagnose fatty liver patients with higher degree of metabolic dysfunction than NAFLD.
The top three causes of death in NAFLD patients were CV disease, liver-related diseases, and non-liver cancers. Therefore, whether the new diagnostic criteria of MAFLD have the advantage of picking out high-risk patients of liver or CV diseases needs research to confirm. As we know, the severity of liver fibrosis was associated with overall mortality and the risk of HCC development.22 The severity of atherosclerosis correlates with the risk of CVD.23 A Taiwan study of 166 cases with pathologic diagnosis of hepatic steatosis or cryptogenic cirrhosis showed that “MAFLD alone group” patients had higher NAFLD activity score and percentage of advanced fibrosis than those with “NAFLD alone group”.24 In another study of 765 Japanese fatty liver patients using shear wave elastography, they found liver stiffness was higher in MAFLD patients than those of NAFLD. In addition, the diagnostic criteria of MAFLD had higher sensitivity for detecting significant liver fibrosis than the previous NAFLD criteria.25 In our study, MAFLD patients had higher serum AST, ALT, GGT, fatty liver index and NFS than NAFLD patients. Furthermore, the “MAFLD only group’ had higher serum AST, ALT, GGT, fatty liver index, NFS, and FIB-4 score than “NAFLD only group”. Taking together, the new diagnostic criteria of MAFLD can help identifying those with a high degree of disease activity including hepatic steatosis, liver inflammatory and fibrosis than previous NAFLD criteria.
Regarding CVD risk, a nationwide health-screening database from Korea cohort followed by a median of 10.1 years revealed NAFLD and MAFLD patients had higher risk of CV events. In addition, when the “neither FLD” was used as reference, “both FLD” had highest hazard ratios for CV events, followed by MAFLD only and NAFLD only groups, suggesting the added patients (MAFLD only) had higher CV risk than the missed population (NAFLD only) after the change of diagnostic criteria from NAFLD to MAFLD.26 Another study of 2985 subjects followed for 7 years found that although NAFLD and MAFLD had similar metabolic traits at baseline and similar outcome of CV events, “MAFLD only” patients had higher risk of adverse outcomes than “NAFLD only” patients.28 Using the data of carotid duplex ultrasound, our study found MAFLD patients had higher IMTs than those with NAFLD. In addition, the new population after the change of disease name from NAFLD to MAFLD had higher IMT and percentage of carotid plaques than the missed population. Our data consistently confirmed the new terminology of MAFLD better identifies high-risk patients of CVD.
The missed population after the change of diagnostic criteria from NAFLD to MAFLD includes fatty liver patients who fulfilled the definition of NAFLD, but no metabolic risk (MR) (“NAFLD only group”; non-MR NAFLD). These subjects presented with fatty liver in imaging or histology without metabolic dysfunction, such as overweight/obese, DM, and metabolic dysregulation and without other etiologies of chronic liver disease. Since these patients had no metabolic diseases and other concomitant liver diseases, the risk of liver and CVD is expected to be minimal. Some previous studies confirmed the subjects with non-MR NAFLD without urgent diagnostic and therapeutic intervention needs due to a potentially favorable disease.25 However, a recent brief report using NHANES III database noted “non-MR NAFLD only” patients with severe fatty liver in ultrasound might have significant liver injury and fibrosis and need more attention in clinical practice.28 Since the definition of non-MR NAFLD in this study excluded only excessive alcohol consumption and metabolic risk, the results might be confounded by viral hepatitis or other causes of liver disease. In another study of 1217 cases with liver biopsy, there was no significant difference in the degree of liver inflammation and fibrosis among MAFLD, NAFLD, and “non-MR NAFLD” groups histologically. They suggested MAFLD criteria might overlook a subset of patients with steatohepatitis and significant fibrosis. However, since the majority of the cases were infected with HBV (93.26%), the confounding effect cannot be excluded.29 Our study revealed that the missed population had lower risk of liver fibrosis and CVD compared with the new population after the change of disease name and diagnostic criteria from NAFLD to MAFLD.
Our study had some strengths. First, this is a large, population-based study from Taiwan bio-bank with simultaneous assessment of liver and CVD risk. Second, this was the first study to compare the risk of CVD between NAFLD and MAFLD patient using the data of carotid duplex ultrasound. Third, since the database have detailed description about the type, amount, frequency and duration of alcohol drinking, the amount of alcohol consumed in a day can be accurately calculated to make the correct judge about the positive or negative diagnosis of NAFLD. However, some limitations should also be addressed. First, fatty liver was determined by ultrasound without histology in our study. However, liver biopsy is not well suited in population-based study. Furthermore, ultrasound-based studies remain poor diagnostic accuracy of fatty liver when hepatic steatosis is < 30%, for which ultrasound tends to underestimate the true prevalence of NAFLD and MAFLD. Second, this was a cross-sectional study and unable to demonstrate the causal relationship. Third, since the data of high-sensitivity C reactive protein and insulin resistance were not available in our study, the prevalence of MAFLD might be underestimated.
In summary, MAFLD is an “inclusive” diagnosis with the mention of underlying causes, which makes explanation and education more easily between physicians and patients. In addition, this large-scale, population-based study confirmed the change of nomenclature and diagnostic criteria from NAFLD to MAFLD could identify more patients at risk of metabolic, liver and CV complications for early intervention, suggesting MAFLD may be a better nomenclature than NAFLD in clinical practice. However, since MAFLD patients could have other concomitant liver diseases such as alcohol, viral hepatitis, or autoimmune liver diseases etc., the natural history and clinical outcomes of those patients need further investigations.30