The literature search found 1469 MAFLD papers published in 434 journals (Fig. 1a, S1). The number of publications showed a steady annual increase with 533 solely in 2023 at the time of data collection (Fig. 1a). Based on the analysis of publication type, the steady increase in article number over time suggests ongoing research interest in MAFLD (Fig. 1b). The categories of journals publishing a high number of papers were not limited to gastroenterology and hepatology but included other disciplines such as endocrinology, metabolism, and general medicine and science (Fig. 1c).

MAFLD literature and the publication journals. a The annual number of MAFLD papers from 2020 and 2023. b The annual number of MAFLD papers according to the three publication types. c Top journals publishing a high number of MAFLD papers

Research topics in MAFLD were assessed using author keywords. In 1195 papers with author keywords (Fig. S1), the most common were MAFLD and other terms for fatty liver disease (Fig. 2a), which is consistent with the scope of the literature search. These were followed by terms related to metabolic dysfunction, including obesity, metabolic syndrome, (type 2) diabetes, insulin resistance, and CVD (Fig. 2a). Another category of frequent keywords included liver pathology and complications from MAFLD, such as steatosis, inflammation, fibrosis, cirrhosis, and hepatocellular carcinoma (HCC) (Fig. 2a). COVID-19 was listed as a common keyword, suggesting the influence of the pandemic on the field (Fig. 2a).

Author keywords in MAFLD and NAFLD literature. a Frequent author keywords in the MAFLD literature. b Frequent author keywords in the NAFLD literature during 2016 and 2019. c Co-occurrence network of 67 common author keywords in the MAFLD literature with 5 keyword groups. AMPK, adenosine monophosphate-activated protein kinase; COVID-19, coronavirus disease 2019; HCC, hepatocellular carcinoma; MAFLD, metabolic (dysfunction)-associated fatty liver disease; NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis; NHANES, National Health and Nutrition Examination Survey; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2

To compare the research topics between NAFLD (2016–2019) and MAFLD (2020 onwards), common keywords were analyzed for 7568 NAFLD papers published during that period (Fig. 2b, S2). Apart from the terms for fatty liver disease, the most frequent keywords related to metabolic dysfunction, as in MAFLD (Fig. 2b). This finding confirms that metabolic dysfunction was recognized as an underlying association of NAFLD prior to the proposal of MAFLD. In contrast, a major difference between NAFLD and MAFLD papers was that CVD has become one of the common keywords in the MAFLD literature (Fig. 2b). This finding implies increased awareness of cardiometabolic outcomes in MAFLD, as compared to NAFLD.

To examine for associations between keywords, a co-occurrence network of 67 common keywords was constructed (Fig. 2c). Community detection analysis of the network identified 5 keyword groups (A to E) according to the number of included terms (Fig. 2c). The largest, group A contained terms for fatty liver disease, including MAFLD. Other common terms in this group were associated with epidemiological metrics such as prevalence, mortality, and risk factor, as well as diagnostic tools for fatty liver disease, including fatty liver index and transient elastography [1]. Moreover, this group comprised CVD and CKD, two major complications of MAFLD [1, 4, 5]. Thus, group A was the central community in the network with a focus on clinical epidemiology. Group B was enriched in features of metabolic dysfunction, including obesity, (type 2) diabetes, insulin resistance, and metabolic syndrome. Group C included histological findings of MAFLD such as steatosis, steatohepatitis, inflammation, and fibrosis [29]. This group contained multiple terms for molecular mechanisms of MAFLD, including autophagy, lipotoxicity, mitochondria, and oxidative stress [29,30,31,32]. Thus, group C related to the pathophysiology of MAFLD. Group D comprised COVID-19 and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), as well as hepatic complications such as cirrhosis, liver cancer, and HCC. The smallest group E contained gut microbiota and gut-liver axis, which play important roles in the pathogenesis of MAFLD [20, 31, 32]. Altogether, the major focus of MAFLD was clinical epidemiology and its relationship with metabolic dysfunction, whereas other domains of research included disease mechanisms, hepatic complications, and associations with COVID-19.

Citation analysis was performed to gain insights into the research findings from 1418 MAFLD papers with citation records (Fig. S1). To this end, highly cited publications were identified based on citation counts in the databases (Fig. 3a, S3) and those from the MAFLD literature included in this study (Fig. 3b) [1,2,3, 6,7,8,9, 11, 16, 33,34,35,36,37,38,39,40,41,42]. The two most cited publications were the initial statements proposing MAFLD in 2020 with more than 1500 overall citations [2, 3]. The highly cited publications mostly pertained to discussion of the definition [1, 6,7,8,9, 11] and validation of the criteria [33, 35, 37,38,39, 41, 42]. These studies show that individuals diagnosed as MAFLD or NAFLD are concordant but not fully equivalent [33, 35, 37,38,39, 41, 42]. Moreover, the risks of hepatic and extrahepatic complications, including liver fibrosis [35], CVD [37, 42], CKD [39], and all-cause mortality [38], were greater in MAFLD than in NAFLD. Several publications with high citation counts reported the association between MAFLD and COVID-19 (Fig. 3a) [16, 34, 40].

Citation analysis of the MAFLD literature. a Highly cited MAFLD papers recorded in the Scopus database. b Top publications with high citations from the MAFLD literature. c Annual citation counts of representative papers from MAFLD literature. d. Annual counts of representative keywords in cited MAFLD literature. Note that the vertical axes in c and d are log-transformed to facilitate visualization. COVID-19, coronavirus disease 2019; HCC, hepatocellular carcinoma; MAFLD, metabolic (dysfunction)-associated fatty liver disease

In representative highly cited publications related to the redefinition [1,2,3] and original studies investigating the clinical outcomes of MAFLD [33,34,35, 37,38,39], annual citation counts received from MAFLD papers showed overall increases during the search period (Fig. 3c). In addition, frequent keywords in cited MAFLD papers showed a similar annual increasing trend (Fig. 3d). The exceptions were papers and keywords related to COVID-19, which showed a peak in 2021 with a subsequent decrease in counts (Fig. 3c, d).

To assess the relationships between publications, we constructed a co-citation network of 722 MAFLD papers (Fig. S1). Community detection analysis in this network divided 713 papers into 5 major publication groups (A to E) (Fig. 4a). Subsequently, 10 key publications in each group were identified based on high strength of the nodes to infer the central research findings (Fig. 5a–e).

Co-citation network of MAFLD literature. a Co-citation network of 713 MAFLD papers in 5 major publication groups. The node size corresponds to node strength. The annotated numbers in the nodes indicate the reference number of key publications in Fig. 5. b The annual number of papers in each publication group of the co-citation network. The number in each group label indicates the total number of manuscripts in the publication group

Key publications in the co-citation network of MAFLD literature. a–e 10 Key publications in each publication group determined by high node strength in the co-citation network of the MAFLD literature. Each reference number of the publication corresponds to the annotated number for the node in the co-citation network of Fig. 4a

Group A was mainly associated with the redefinition of fatty liver disease (Fig. 5a) and comprised publications with high citations (Fig. 3a, b) [1,2,3, 8, 9, 11, 33, 35, 39, 41]. This group included the two initial proposals for MAFLD [2, 3], which were followed by endorsements or arguments in this context [1, 8, 9, 11]. In addition, several early clinical studies demonstrated better performance of the MAFLD criteria for detecting the risk of complications, such as hepatic fibrosis [33] and CKD [39] than NAFLD.

The main topic in Group B was clinical outcomes of MAFLD (Fig. 5b) [37, 38, 42,43,44,45,46,47,48,49]. Key publications were clinical studies demonstrating that MAFLD is more strongly associated with higher risks of CVD, CKD, and all-cause mortality than NAFLD [37, 38, 42,43,44,45]. Moreover, the importance of concomitant liver diseases was suggested by findings that the clinical course of MAFLD is affected by chronic hepatitis virus infection and excessive alcohol consumption [46, 49]. As an aside, a few publications demonstrated a greater association with liver fibrosis of MAFLD than NAFLD [47, 48].

Group C covered other epidemiological features of MAFLD, including prevalence, risk factors, and non-invasive fibrosis assessment (Fig. 5c) [13, 14, 50,51,52,53,54,