1. Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) has emerged as a growing global epidemic, with a prevalence estimated between 30–38% [1,2]. The burden is particularly high in South America and the Middle East, where MASLD represents a significant public health concern [3]. In parallel, the clinical relevance of MASLD in Turkey has become increasingly evident, given the rising rates of associated cardiometabolic comorbidities in the country.

Despite its public health implications, the prevalence and characteristics of MASLD in Turkey remain poorly defined, largely due to inconsistencies in study design, methodology, and diagnostic criteria across existing research. Reported prevalence rates in ostensibly healthy populations vary widely, from 10.6% to 65.6%, although many of these studies are limited by suboptimal diagnostic modalities [4,5], retrospective designs [6–8], or small sample sizes [9,10]. A retrospective study with a larger cohort (n = 113.239) estimated a prevalence of 48.3% in the general population [11]. Given that MASLD is strongly associated with cardiometabolic risk factors, higher prevalence rates would be expected in at-risk populations with metabolic disorders. However, only a limited number of studies have investigated this in Turkish at-risk groups. Among patients with diabetes, retrospective reports of MASLD defined by ultrasound has reported prevalence rates ranging from 69.9% to 71.9% [12,13]. Small-scale studies among subjects with diabetes [14] and obesity reported a high prevalence of 94.3% and 88.7% [15], respectively, though MASLD risk factors remain uninvestigated in both. Other key metabolic drivers of MASLD—such as insulin resistance, abdominal obesity, and metabolic syndrome (MetS)—remain unexplored in Turkish populations. According to current MASLD diagnostic criteria, individuals with abdominal obesity, dyslipidemia, impaired glucose metabolism, or elevated blood pressure are considered at risk for MASLD. The independent contribution of these metabolic factors to disease severity in at-risk Turkish cohorts remains to be elucidated.

Therefore, we aimed to investigate the prevalence and risk factors of MASLD and significant fibrosis in at-risk populations to provide a more comprehensive understanding of MASLD burden and its metabolic correlates in this specific population.

2. Materials and methods

We conducted a cross-sectional study involving adult participants who visited Istanbul University-Cerrahpasa Hospital obesity division of the endocrinology clinic and an affiliated secondary care internal medicine outpatient center in Istanbul for regular follow-up or initial assessment. The study was originally designed and registered as a single-center study to be conducted at the outpatient center. Subsequently, the second center employing the same inclusion and exclusion criteria was incorporated to decrease selection bias and improve the representativeness of the sample, particularly with respect to socioeconomic and health status. Participant enrollment occurred at both sites between February 2022 and December 2024. The study received ethical approval from the relevant institutional review boards (Istanbul University-Cerrahpasa Clinical Research Ethics Committee, Ref. 2023/134; Istanbul Health Sciences University Clinical Research Ethics Committee, Ref. 2021/224) and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants. The study was prospectively registered with ClinicalTrials.gov (ID: NCT05194553).

3. Results

3.1. Patient characteristics

A total of 1,094 individuals were recruited for this study, of whom 1,039 were included in the final analysis (Fig 1). The mean age of the study population was 51.7 ± 13.1 years, and 49.4% (n = 513) were male. The mean body mass index (BMI) was 30.54 ± 5.85 kg/m². Obesity and central obesity were prevalent in 48.5% (n = 504) and 75.1% (n = 780) of the subjects, respectively. Detailed characteristics of the study population are presented in Table 1. The demographic profile of the study cohort was broadly representative of the general Turkish population, with the exception of educational attainment, which was higher among study participants (S2 Table). Differences in socioeconomic and health status of the participants in two research sites are provided in S3 Table.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of patients with and without MASLD based on transient elastography.

https://doi.org/10.1371/journal.pone.0341214.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Study flowchart.

https://doi.org/10.1371/journal.pone.0341214.g001

3.2. Prevalence of MASLD and associated factors

The overall prevalence of MASLD, as determined by CAP, was 57.5% (95% CI: 54.4%–60.4%), with severe steatosis (S3) observed in 35.9% (95% CI: 33.0%–38.9%) of cases (Fig 2). When stratified by sex, males (59.8%, 95% CI: 55.5%–64.0%) had higher MASLD prevalence compared with females (55.1%95% CI: 50.9%–59.3%). MASLD was most prevalent among individuals with obesity and MetS, and least common in those without MetS. The largest disparity in MASLD prevalence was also observed between participants with and without MetS (Fig 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Stages of MASLD and fibrosis in the study population.

https://doi.org/10.1371/journal.pone.0341214.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Prevalence of MASLD and significant fibrosis according to presence of obesity, metabolic syndrome, diabetes and hypertension.

Abbreviations: DM, diabetes mellitus; DysL, dyslipidemia, HT, hypertension; MetS, metabolic syndrome.

https://doi.org/10.1371/journal.pone.0341214.g003

Subjects with MASLD were older, had lower educational attainment, and exhibited higher BMI and heart rate values. In terms of laboratory findings, MASLD cases showed significantly elevated levels of HbA1c, fasting blood glucose, ALT, AST, GGT, triglycerides, and HOMA-IR, along with significantly reduced HDL levels. Prevalence of obesity, MetS and all its components were higher in the MASLD group (Table 1). According to a multivariable logistic regression, independent factors associated with higher odds of MASLD were male sex, the presence of obesity, insulin resistance, MetS and central obesity, as well as the research site (Table 2, S4 Table). All necessary assumptions were met for the analysis in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Multivariable logistic regression analyses of factors associated with the presence of MASLD based on transient elastography.

https://doi.org/10.1371/journal.pone.0341214.t002

3.3. Prevalence of fibrosis and associated factors

The overall prevalence of significant liver fibrosis, as determined by LSM, was 7.6% (95% CI: 6.1%–9.4%), with advanced fibrosis (>F3) observed in 2.6% (95% CI: 1.8%–3.8%) of cases (Fig 2). Among individuals with MASLD, the prevalence of significant fibrosis increased to 11.1% (95% CI: 8.8%–13.8%). When stratified by sex, females (8.2%, 95% CI: 6.1%–10.8%) had slightly higher fibrosis prevalence compared with males (7.0%, 95% CI: 5.1%–9.6%). Fibrosis was most prevalent among individuals with DM and obesity, and least common among those without obesity. The largest disparity in fibrosis prevalence was also observed between obese and non-obese participants (Fig 3).

Subjects with significant fibrosis had lower income, lower educational attainment and exhibited higher BMI and heart rate values. In terms of laboratory findings, fibrosis cases showed significantly elevated levels of HbA1c, fasting blood glucose, ALT, AST, GGT, LDL, triglycerides, and HOMA-IR, along with significantly reduced HDL levels. Prevalence of obesity, MetS and all its components were higher in the fibrosis group, whereas noninvasive indices of fibrosis were significantly lower (Table 3). On multivariable logistic regression, independent factors associated with higher odds of significant fibrosis were the presence of obesity, diabetes, MetS and insulin resistance, while hypertension was associated with lower odds of significant fibrosis. Compared to the very low-income group, the high-income group has a 75% reduction in the odds of significant fibrosis, when adjusting for research site, sex, age, marital status, obesity, diabetes, insulin resistance, hypertension, and metabolic syndrome (Table 4, S5 Table). All necessary assumptions were met for the analysis in Table 4.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Characteristics of patients with and without advanced fibrosis based on transient elastography.

https://doi.org/10.1371/journal.pone.0341214.t003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Multivariable logistic regression analyses of factors associated with the presence of significant fibrosis based on transient elastography.

https://doi.org/10.1371/journal.pone.0341214.t004

4. Discussion

This multicenter cross-sectional study conducted among individuals at risk for MASLD in Turkey demonstrated that more than half of those with at least one metabolic risk factor were affected by MASLD. In contrast, the prevalence of liver fibrosis in this population was considerably lower. To the best of our knowledge, this is the first study to date utilizing VCTE to assess MASLD and fibrosis in metabolically at-risk Turkish populations.

Our findings indicate a lower MASLD prevalence compared to some earlier VCTE-based screening studies in at-risk populations, which have reported prevalence rates ranging from 61.3% to 83.6% [25–30]. This discrepancy may, in part, be attributed to differences in inclusion criteria across studies. While only a few smaller studies included individuals with at least one metabolic risk factor—similar to our approach [31–33], the majority of investigations targeting at-risk MASLD populations have focused specifically on individuals with diabetes and/or obesity—conditions already known to be strongly associated with higher MASLD prevalence. Indeed, meta-analyses have reported MASLD rates of 69.9% in overweight individuals, 75.3% in obese individuals and 65.3% in those with diabetes [34,35]. A recent meta-analysis focusing on studies from the Middle East reported a MASLD prevalence of 68.7% in the region and 70.0% in Turkey among patients with type 2 diabetes [3]. In contrast, MASLD appears to be less common in individuals with hypertension or dyslipidemia, with lower reported rates among these subgroups [36]. Consistent with this pattern, our study also found substantially higher MASLD prevalence in obese and diabetic participants compared to those with hypertension or dyslipidemia. These findings align with previous sporadic reports on MASLD prevalence among Turkish patients with diabetes [12–14] and obesity [15]. Although earlier Turkish reports, despite their methodological limitations, yielded MASLD prevalence estimates consistent with international data, our study provides a more robust confirmation of them, strengthening the evidence for MASLD burden in Turkish at-risk populations.

In our study, male sex was associated with an increased odds of MASLD, in line with previous evidence [11,32,37–41]. Among clinical predictors, central obesity emerged as the strongest determinant of MASLD, followed by general obesity. Although obesity defined by BMI has consistently been associated with MASLD [4,7,11,14,42], central obesity had not been previously investigated in Turkey. Worldwide evidence suggests that fat distribution, rather than overall fat mass, may play a more critical role in the pathogenesis of MASLD [43,44]. While it remains unclear whether the effect of central obesity is independent of general obesity, our findings address this gap and suggest that central adiposity may exert a comparable—if not greater—impact on MASLD risk in Turkish populations. This is similar to the findings from Asian populations, where both central and general obesity have been shown to similarly elevate MASLD risk [45]. Furthermore, dysfunctional adipose tissue in central obesity also causes insulin resistance – a key pathophysiological mechanism responsible for the development and progression of MASLD. Insulin resistance promotes adipose tissue lipolysis, leading to elevated circulating free fatty acids and their subsequent deposition in the liver, resulting in hepatic steatosis [46]. In line with this, we showed that the presence of insulin resistance, but not diabetes, is an independent predictor of MASLD in Turkey, which corroborates previous studies linking insulin resistance markers to increased MASLD risk [47,48]. The lack of a significant association between diabetes and MASLD in our study contrasts with many population-based reports but likely reflects the specific characteristics of our at-risk cohort. Because all participants fulfilled at least one metabolic criterion for MASLD diagnosis, the background prevalence of metabolic disturbances was already high, diminishing the discriminatory effect of diabetes on steatosis presence. Moreover, most individuals with diabetes had relatively well-controlled glycemia (mean HbA1c = 6.6 ± 1.5) and were under active treatment, which may have limited hepatic fat accumulation. The presence of MetS, on the other hand, showed the second strongest association with MASLD in our cohort, corresponding to a 2.5-fold increase in the odds of disease presence. In comparison, a cohort study involving 11,647 individuals reported an 11.5-fold increase for the same association [49]. However, their reliance on ultrasound for MASLD diagnosis may have compromised the accuracy of the findings. Notably, studies employing VCTE as the diagnostic modality have reported results more consistent with ours [18,31,32]. Finally, subjects enrolled at the tertiary center showed considerably higher odds of MASLD than those from the outpatient clinic, which may be explained by several factors. These include referral bias, as tertiary centers often manage more complex cases; and a higher burden of underlying comorbidities that we did not adjust for. Additionally, patients at tertiary centers may exhibit distinct health-seeking behaviors, potentially reflecting more advanced or symptomatic disease. Taken together, these findings reiterate the need for further research to better characterize MASLD risk factors in at-risk populations in various clinical settings.

To date, only a limited number of studies have investigated the prevalence of liver fibrosis in at-risk populations. Three nationwide studies and a large cross-sectional study consistently reported similar prevalence rates, ranging between 13–14% in individuals with metabolic syndrome, 11–14% in those with obesity, %12 in central obesity and 14–21% in patients with diabetes [50–53]. Two relatively small studies that included individuals with at least one metabolic risk factor—similar to our inclusion criteria—reported fibrosis prevalences of 9.4% and 16.8%, respectively [31,32]. In line with these findings, our results demonstrated that fibrosis was most prevalent among participants with obesity, diabetes, and MetS, with subgroup-specific rates comparable to those reported in previous studies. However, earlier studies conducted in Turkish cohorts have yielded different prevalence estimates. For example, fibrosis was reported in 6.6% and 9.8% of patients presenting to tertiary centers for dyspepsia [5,54]. Among adults with diabetes, prevalence ranged from 9.5% to 16.9% [12–14], while a study in apparently healthy individuals reported a rate of 15.8% [7]. These discrepancies likely reflect differences in study populations, settings, and possibly methodological approaches, suggesting that previous cohorts may not be representative of the broader at-risk population in Turkey. By focusing on a well-defined at-risk population using standardized assessment methods, our study helps address this gap and provides prevalence estimates that are consistent with international figures.

In our study, obesity emerged as the strongest predictor of significant liver fibrosis, consistent with previous reports demonstrating a similar association between obesity and fibrosis risk [50,55–57]. This relationship may be interpreted in several ways: obesity may directly contribute to fibrogenesis, or alternatively, it may exert its effect indirectly by promoting metabolic disturbances such as impaired glucose metabolism, which are known contributors to liver fibrosis. Supporting this, our findings -as well as those of other studies [52,58–60]- highlight insulin resistance and diabetes as strong predictors of clinically significant fibrosis. These observations underscore the critical role of these potentially modifiable risk factors in fibrosis progression and emphasize the importance of early intervention targeting insulin resistance and/or diabetes. Adjusted for its individual components, MetS itself was independently associated with fibrosis in our cohort. Very few studies have examined the role of MetS in hepatic fibrosis. One study reported more than a three-fold increase in fibrosis risk among individuals with MetS compared to those without [53], whereas a biopsy-proven MASLD cohort from Turkey did not identify MetS as a significant predictor of fibrosis [61]. These conflicting findings highlight the need for larger, prospective studies to clarify the independent contribution of MetS to hepatic fibrosis risk. We also observed an unexpected inverse association between hypertension and significant fibrosis. While this finding aligns with results from prior work reporting an attenuated risk among hypertensive individuals [62], it should be interpreted cautiously, as unmeasured confounding, treatment effects, or selection bias may partially explain this association. Finally, we observed a significantly lower odds of fibrosis among individuals in the highest income group. A similar trend was reported in a nationwide study involving 5.7 million participants, where fibrosis prevalence gradually declined with increasing GDP per capita, although the associations did not reach statistical significance [50]. This may reflect the complex and context-dependent relationship between socioeconomic status and liver health, which can be influenced by various mediating factors such as healthcare access, lifestyle habits, and comorbidities. Further research is warranted to explore these pathways and determine whether income-related disparities contribute to fibrosis risk in different populations.

The strengths of the present study include a relatively large sample, and the use of a reliable, imaging-based, non-invasive technique to assess hepatic steatosis and fibrosis, offering advantages over surrogate blood-based markers. Although MASLD has been widely studied in different regions, our study fills a major knowledge gap by providing the first multicenter, VCTE-based assessment of steatosis and fibrosis prevalence among metabolically at-risk adults in Turkey—a population with one of the world’s highest rates of obesity and type 2 diabetes. The standardized design across centers and exclusion of secondary liver disease sources strengthen the reliability and comparability of our findings. Importantly, the identification of fibrosis-related metabolic phenotypes within this high-risk cohort highlights specific subgroups who may benefit from early MASLD screening in metabolic or endocrine clinics. These results carry direct clinical implications for prioritizing non-invasive liver assessment in cardiometabolic care pathways and for informing national strategies aimed at early detection and prevention of advanced liver disease. Moreover, although this was not a nationwide study, the socioeconomic characteristics of the study population were moderately representative of the broader population, thereby enhancing the generalizability of our findings. Nonetheless, as the study was partially conducted at a tertiary care center and included individuals with established risk factors, the metabolic burden among participants was likely higher than that of the general population. Therefore, the findings may not be generalizable to the ostensibly healthy Turkish population. Our findings should be interpreted also in light of several limitations. Although we identified important associations, the observational nature of the study precludes conclusions regarding causality. Additionally, we did not use the gold standard of liver biopsy to assess fibrosis. However, performing liver biopsies in all participants of a cohort would have been impractical, and VCTE remains a validated and widely accepted non-invasive tool for fibrosis assessment. Although the inclusion of both an outpatient clinic and a tertiary care center enhanced the diversity and size of our sample, it may have introduced selection bias. Patients attending tertiary centers are typically referred for more complex or advanced health conditions and may differ systematically from those seen in primary or outpatient settings. As such, the higher prevalence of MASLD observed among participants from the tertiary center may reflect an overrepresentation of individuals with more severe metabolic risk profiles. Also, study site was included as a fixed effect to account for systematic differences between centers; however, this approach does not explicitly model within-site correlation, which should be considered when interpreting the results. Finally, as the study was conducted in a metropolitan setting, our findings may not fully reflect the epidemiological patterns of MASLD in rural populations in Turkey.

To conclude, this study demonstrated that the prevalence of MASLD in at-risk populations in Turkey is considerably high, while MASLD-related signficant fibrosis remains less common. Obesity, insulin resistance, and metabolic syndrome emerged as shared risk factors for both conditions. Our findings support targeted screening and monitoring of these high-risk groups to identify individuals at greater risk for advanced MASLD. Nonetheless, further research is needed to evaluate the long-term benefits and cost-effectiveness of such an approach.

Supporting information