Abstract
Background
Metabolic dysfunction-associated steatotic liver disease (MASLD) is highly prevalent among patients with type 2 diabetes mellitus (T2DM). Non-invasive tools (NITs) such as the FIB-4 index, are widely used for risk stratification to identify patients at higher risk of progressive liver disease. This study aimed to assess the utility of the FIB-4 index and identify stronger predictors of MASLD in a T2DM cohort.
Methods
A retrospective cross-sectional study was conducted on 186 patients with T2DM. Hepatic steatosis was diagnosed via abdominal ultrasonography, categorizing patients into MASLD (n=100) and non-MASLD (n=86) groups. The FIB-4 index was calculated, and group comparisons were performed. A novel simplified clinical scoring system was constructed to identify independent predictors of ultrasound-confirmed steatosis.
Results
Patients with MASLD exhibited significantly higher body mass index (BMI) (P < 0.001), alanine aminotransferase (ALT) (P = 0.005), Aspartate aminotransferase (AST) (P =0.05) and triglycerides (P = 0.034) compared to the non-MASLD group. The multivariate logistic regression analysis established the weight of each clinical marker. The FIB-4 index failed to significantly differentiate between the two groups (P = 0.272).
Conclusion
The FIB-4 index demonstrated limited diagnostic utility for MASLD in this T2DM population. Conversely, a newly derived score, incorporating readily available parameters like BMI, ALT, AST, and triglycerides, offers a highly specific, practical alternative for non-invasive MASLD screening in resource-limited primary care settings.
Keywords: MASLD, type 2 diabetes mellitus, fib-4 index, hepatic steatosis
Introduction
Metabolic dysfunction-associated steatotic liver disease (MASLD) is defined as the presence of hepatic steatosis in conjunction with at least one cardiometabolic risk factor (CMRF) and the absence of other discernible causes.1,2
Adult Criteria of CMRF are shown in Figure 1 adapted from the Delphi consensus statement.1
Figure 1.
Cardio-metabolic risk factor (CMRF).
MASLD may be present in up to 30% to 75% of patients with type 2 diabetes mellitus, and the overall prevalence of MASLD is expected to continue to increase.2,3
The prevalence of nonalcoholic fatty liver disease (NAFLD: the older terminology equivalent to the current MASLD) in Middle East and North Africa (MENA) is an estimated 36.53%, (28.63–45.22%).4 A cross-sectional study conducted in Jordan, using ultrasonographic criteria to diagnose NAFLD, estimated that the prevalence of NAFLD among participants with diabetes was 80.4%.5
Ultrasound can support a diagnosis of MASLD when it shows characteristic signs of fat accumulation in the liver — such as increased liver echogenicity (“bright” liver), reduced visibility of the portal vein walls and diaphragm, and attenuation of the ultrasound beam — which reflect hepatic steatosis.6–9 Conventional ultrasound is reliable in detecting moderate-to-severe steatosis (when ≳ 20–30% of hepatocytes have fat), but has lower sensitivity for mild steatosis — so a “normal” ultrasound does not rule out early-stage MASLD.
Given the rising global burden of MASLD, the use of non-invasive tools (NITs) has become crucial for identifying patients at greatest risk of fibrosis progression and liver-related complications. These tools—including serum-based scores such as FIB-4 index and NAFLD Fibrosis Score, as well as imaging modalities like vibration-controlled transient elastography (FibroScan) and shear-wave elastography—allow reliable assessment of liver fibrosis without the risks, costs, or limited accessibility of biopsy. FIB-4 index is the best validated scoring system to rule out advanced fibrosis and cirrhosis. When the FIB-4 index is >1.3, it cannot rule out advanced fibrosis, and a second fibrosis assessment should then be performed. Incorporating NITs into routine care enables early detection, better risk stratification, and timely referral for specialist management, ultimately supporting more effective and efficient care for the growing MASLD population.9–11
This retrospective study aimed to determine the stronger risk factors for MASLD in our cohort. We aimed to check FIB-4 index to further define our patients with higher risk of progressive liver disease. The results could inform clinical practice by promoting the use of non-invasive, cost-effective screening tools, ultimately enhancing early detection and management of MASLD in diabetic patients.
Methods
Study Design and Population
This study is a cross-sectional retrospective study conducted at King Abdullah University Hospital (KAUH), Irbid, Jordan. We analyzed the medical records of 391 patients who visited King Abdullah University Hospital between April 2018 and June 2024, who had abdominal ultrasonography (US) and had a diagnosis of diabetes. 186 patients remained eligible after applying exclusion criteria mentioned below.
Inclusion Criteria
Patients aged 18 years or older with a documented diagnosis of Type 2 Diabetes Mellitus (T2DM) who underwent abdominal ultrasonography during the study period. The diagnosis of T2DM was verified by comprehensive clinical records, active use of glucose-lowering medications (eg, Metformin, sulfonylurea, etc), and laboratory data. To account for incomplete medical records and ensure patients with well-controlled T2DM were not omitted, no minimum cross-sectional HbA1c threshold was enforced for patients with an established diagnosis; for patients lacking a formal coded diagnosis but displaying clinical indicators, an HbA1c greater than or equal to 6.5% was required for inclusion.
Exclusion Criteria
Patients with the following conditions were excluded from the study: Viral hepatitis, Autoimmune hepatitis, Primary biliary cholangitis, Alcoholic hepatitis or congestive hepatopathy. History of chemotherapy, hematologic malignancies (such as leukemia, lymphoma, or multiple myeloma), hypopituitarism, and abnormal thyroid function.
Data Collection and Procedures
The data collection time point was defined as the date of abdominal ultrasonography (US). Liver function tests (LFTs) and other relevant laboratory data were collected from a time point within one year of the abdominal US. Abdominal US reports were reviewed to assess for features of hepatic steatosis. Ultrasound can support a diagnosis of MASLD when it shows characteristic signs of fat accumulation in the liver.
The FIB-4 index was calculated using the following formula:
 |
Relevant demographic and clinical variables were also extracted, including age, gender, BMI calculated as body mass (kg) divided by the square of height (m2), and HbA1c. Laboratory parameters included lipid profile, platelet count, aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), gamma-glutamyl transpeptidase (GGT), bilirubin, albumin, and creatinine.
A low FIB-4 index was defined as 1.3 and high as 2.67.
Ethical Considerations
The study protocol was reviewed and ethically approved by the Institutional Review Board (IRB) at King Abdullah University Hospital (Approval No.782-2023). All steps performed in the study were in accordance with the 1964 Helsinki Declaration and its later amendments. For this type of study, formal consent is not required. IRB waived the requirement for informed consent due to the retrospective nature of the study and the use of de-identified patient data. To ensure confidentiality, all patient identifiers (such as names and national ID numbers) were removed from the dataset. Data were stored on a password-protected server, and access was restricted to the primary investigators.
Statistical Analysis
Continuous variables were evaluated for normality using the Shapiro–Wilk test. Normally distributed data are expressed as mean ± standard deviation (SD), whereas non-normally distributed data are presented as median and interquartile range (IQR). To evaluate differences between two independent groups, the independent samples t-test was applied for normally distributed variables, and the Mann–Whitney U-test was utilized for non-normally distributed variables. Categorical variables are summarized as frequencies and percentages. Associations between categorical variables were evaluated using the Chi-square test or Fisher’s exact test when any expected cell frequency was less than five. Comparisons across the three FIB-4 categories were performed using one-way analysis of variance (ANOVA) or the Kruskal–Wallis test, as appropriate.
To identify predictors associated with hepatic steatosis, a multivariate logistic regression analysis was conducted. The dependent variable was the ultrasound (US) result, coded as 1 (positive) and 0 (negative). Independent variables included body mass index (BMI), aspartate aminotransferase (AST), alanine aminotransferase (ALT), and triglyceride levels. Prior to analysis, case-wise deletion was implemented for individuals presenting with missing data across the target variables, yielding a definitive evaluation cohort of 116 patients (66 MASLD cases and 50 controls). Model fit was evaluated using deviance, the Akaike Information Criterion (AIC), and McFadden’s pseudo R2. Multicollinearity among independent variables was assessed using the variance inflation factor (VIF) and tolerance values. Model discrimination was evaluated via receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC), applying a classification cut-off value of 0.618.
To establish a functional clinical utility score, a linear predictor equation was derived directly from the mathematical regression coefficients (β). The mathematical formulation of the Diabetic Steatosis Score is defined as follows:
 |
The discriminating capability of the generated score was determined via Receiver Operating Characteristic (ROC) curve analysis. The optimal clinical cut-off point was mathematically established by maximizing the Youden Index (J = Sensitivity + Specificity − 1). Standard diagnostic performance parameters, including Sensitivity, Specificity, Positive Predictive Value (PPV), and Negative Predictive Value (NPV), were calculated at this optimal threshold.
Internal Validation and Calibration
To evaluate the internal validity, minimize potential over-fitting optimization, and calculate reliable variance estimates, an internal validation process utilizing bootstrapping with 1000 resamples was performed. Empirical 95% Confidence Intervals (CI) were computed using the 2.5th and 97.5th percentiles from the bootstrap distributions for the Area Under the Curve (AUC), Sensitivity, and Specificity. Model calibration was formally assessed using the Hosmer–Lemeshow goodness-of-fit test (g = 10) along with a visual calibration curve comparing predicted probabilities against observed MASLD fractions. All statistical analyses were conducted using Python 3.10 with the scikit-learn, statsmodels, and matplotlib library packages. A two-tailed p < 0.05 was considered statistically significant.
Results
The key differences between patients with ultrasound-confirmed MASLD (N=100) and non-MASLD (N=86) are primarily metabolic and biochemical.
Statistical analysis identified several factors that were significantly higher in the MASLD group (Table 1):
Body Mass Index (BMI): Patients with MASLD had a significantly higher median BMI of 30 kg/m2 compared to 28.1 kg/m2 in the non-MASLD group (p < 0.001).
Liver Enzymes: Both AST and ALT levels were significantly higher in the MASLD group. The median ALT was 23.3 U/L for MASLD patients vs 16.1 U/L for those without (p = 0.005).
Triglycerides: Median triglyceride levels were significantly elevated in the MASLD group (1.71 mmol/L) compared to the non-MASLD group (1.58 mmol/L; p = 0.034).
Table 1.
Baseline Characteristics of the Study Population (MASLD vs Non-MASLD)
| Variable | MASLD (n = 100) | Non-MASLD (n = 86) | P value |
|---|---|---|---|
| Demographics | |||
| Age (years), median (IQR) | 57.5 (42.0–80.0) | 54.5 (40.0–85.0) | 0.512 |
| Male gender, n (%) | 46 (51.7%) | 43 (48.3%) | 0.586 |
| Anthropometrics | |||
| BMI (kg/m2), median (IQR) | 30.0 (22.0–78.5) | 28.1 (16.3–47.9) | <0.001 |
| Glycemic Status | |||
| HbA1c (%), median (IQR) | 6.88 (5.70–14.6) | 6.63 (5.70–12.7) | 0.749 |
| Liver Function Tests | |||
| AST (U/L), median (IQR) | 20.6 (8.0–72.0) | 17.5 (7.8–80.4) | 0.05 |
| ALT (U/L), median (IQR) | 23.3 (5.6–102.0) | 16.1 (1.5–157.0) | 0.005 |
| ALP (U/L), median (IQR) | 83.5 (7.0–234.0) | 103.7 (41.0–440.0) | 0.12 |
| GGT (U/L), median (IQR) | 36.3 (8.0–625.0) | 30.0 (7.0–493.0) | 0.438 |
| Bilirubin (µmol/L), median (IQR) | 7.86 (2.32–73.4) | 7.95 (1.92–88.8) | 0.852 |
| Synthetic & Renal Function | |||
| Albumin (g/L), median (IQR) | 44.0 (29.0–51.0) | 43.4 (19.2–51.4) | 0.098 |
| Platelets (103/mm3), median (IQR) | 280 (76.0–574.0) | 263 (43.1–510.0) | 0.245 |
| Creatinine (µmol/L), median (IQR) | 69.0 (33.4–1647.0) | 69.0 (38.5–1216.0) | 0.369 |
| Lipid Profile | |||
| HDL (mmol/L), mean ± SD | 1.11 ± 0.298 | 1.17 ± 0.361 | 0.28 |
| LDL (mmol/L), median (IQR) | 1.85 (0.580–61.24) | 1.64 (0.620–6.65) | 0.057 |
| Triglycerides (mmol/L), median (IQR) | 1.71 (0.58–61.24) | 1.58 (0.62–6.01) | 0.034 |
| Cholesterol (mmol/L), median (IQR) | 5.0 (2.17–8.84) | 4.82 (2.22–10.5) | 0.733 |
Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; GGT, gamma-glutamyl transferase; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MASLD, Metabolic Dysfunction-Associated Steatotic Liver Disease.
Patients with a FIB-4 index > 1.3 (combining the Intermediate and High Risk categories) exhibited several distinct clinical and biochemical characteristics compared to the Low Risk group (Table 2).
Table 2.
Association Between FIB-4 Index Categories and Metabolic Syndrome Components
| Component | Low Risk (<1.3) (N=135) |
Intermediate Risk (1.3–2.67) (N=39) |
High Risk (>2.67) (N=12) |
P value |
|---|---|---|---|---|
| Demographics | ||||
| Age (years), median (IQR) | 53 (40–78) | 66 (43–85) | 72 (42–80) | <0.001 |
| Male gender, n (%) | 57 (64.0%) | 23 (25.8%) | 9 (10.1%) | 0.027 |
| BMI (kg/m2), median (IQR) | 29 (22.0–78.5) | 29.4 (16.3–47.9) | 29.4 (24.0–32.0) | 0.848 |
| HbA1c (%), median (IQR) | 6.7 (5.7–12.7) | 6.4 (5.79–14.6) | 7.55 (5.9–12.7) | 0.477 |
| Liver Function Tests | ||||
| AST (U/L), median (IQR) | 17.5 (7.8–59.4) | 22.4 (8.0–73.5) | 43.5 (24.0–80.4) | <0.001 |
| ALT (U/L), median (IQR) | 20.2 (5.6–63.0) | 17.1 (5.7–102.0) | 24.3 (1.5–156.8) | 0.406 |
| ALP (U/L), median (IQR) | 83.5 (7.0–275.0) | 83.5 (49.0–379.0) | 124 (54.0–440.0) | 0.035 |
| GGT (U/L), median (IQR) | 30 (7.0–625.0) | 35 (8.0–363.0) | 102.4 (40.4–493.0) | <0.001 |
| Bilirubin (µmol/L), median (IQR) | 7 (1.92–73.4) | 10 (3.0–39.3) | 18.96 (3.7–88.8) | <0.001 |
| Albumin (g/L), median (IQR) | 44.3 (25.6–51.4) | 43 (27.7–48.0) | 36.3 (19.2–47.0) | <0.001 |
| Platelets (103/mm3), median (IQR) | 293 (124–574) | 222 (43.1–332) | 137 (56.0–341) | <0.001 |
| Creatinine (µmol/L), median (IQR) | 66 (33.4–1216) | 81.5 (36.5–1647) | 87.5 (50.0–781) | 0.01 |
| Lipid Profile | ||||
| HDL (mmol/L), mean ± SD | 1.149 ± 0.287 | 1.2 ± 0.401 | 0.738 ± 0.323 | 0.005 |
| LDL (mmol/L), median (IQR) | 2.97 (0.44–7.31) | 2.32 (1.39–5.24) | 1.96 (0.89–7.0) | 0.069 |
| Triglycerides (mmol/L), median (IQR) | 1.86 (0.58–6.65) | 1.44 (0.67–61.2) | 1.71 (1.45–4.15) | 0.059 |
| Cholesterol (mmol/L), median (IQR) | 5.02 (2.17–10.5) | 4.29 (2.82–7.45) | 3.65 (2.77–8.85) | 0.003 |
Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; FIB-4 index, fibrosis-4 index; GGT, gamma-glutamyl transferase; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
Demographic Profile
Older Age: Patients with higher FIB-4 index are significantly older. The median age is 66 years for the Intermediate group and 72 years for the High Risk group, compared to 53 years in the Low Risk group (p < 0.001).
Gender: There is a higher proportion of males in the elevated FIB-4 index categories (25.8% Intermediate, 10.1% High Risk) compared to females (16.5% Intermediate, 3.1% High Risk) (p = 0.027).
BMI and HbA1c: Interestingly, there were no significant differences in median BMI or HbA1c levels across the different FIB-4 index risk categories.
Liver Function and Biochemical Markers
Patients with FIB-4 index > 1.3 showed significantly altered liver and systemic markers:
Elevated Enzymes: There are statistically significant increases in median AST, ALP, and GGT as the risk category increases.
Lower Platelet Count: A key component of the FIB-4 index formula, platelets are significantly lower in high-risk patients.
Liver Function: Median Albumin levels are significantly lower in the high-risk group, while Bilirubin is significantly higher.
Renal Function: Creatinine levels are significantly higher in patients with elevated FIB-4 index compared to the low-risk group.
Lipid Profile
Total cholesterol is significantly lower in the high-risk group compared to the low-risk group.
Mean HDL also drops significantly in the high-risk category.
FIB-4 Index in MASLD Patients
Interestingly, the FIB-4 index did not show a statistically significant difference between the two groups (MASLD vs non-MASLD group). The majority of MASLD patients (78%) fell into the Low Risk (<1.3) category (Table 3).
Table 3.
FIB-4 Index Scores in MASLD Patients
| Variable | MASLD (n = 100) | Non-MASLD (n = 86) | P value |
|---|---|---|---|
| FIB-4 Index | |||
| FIB-4 index median (IQR) | 0.928 (0.249–7.04) | 1.093 (0.188–5.88) | 0.272 |
| FIB-4 index Categories, n (%) | 0.098 | ||
| Low Risk (<1.3) | 78 (57.8%) | 57 (42.2%) | |
| Intermediate Risk (1.3–2.67) | 15 (38.5%) | 24 (61.5%) | |
| High Risk (≥2.67) | 7 (58.3%) | 5 (41.7%) |
Abbreviations: FIB-4 Index, fibrosis-4 index; MASLD, Metabolic Dysfunction-Associated Steatotic Liver Disease.
Furthermore, when stratified into established risk categories, the distribution of patients across low (<1.3), intermediate (1.3–2.67), and high-risk (≥2.67) groups did not differ significantly between the two cohorts, indicating suboptimal diagnostic performance of FIB-4 index for identifying MASLD in this study population (Figure 2).
Figure 2.
The distribution of patients across low (<1.3), intermediate (1.3–2.67), and high-risk FIB-4 score (≥2.67) groups.
A binomial logistic regression analysis was performed to identify predictors associated with ultrasound (US) found steatosis in diabetics (Supplementary Methodology and Supplementary Results).
An increase in BMI was associated with higher odds of a positive US finding (OR = 1.18, p = 0.003). Similarly, ALT levels showed a significant positive association with US results (OR = 1.07, p = 0.012). AST and triglyceride levels were not statistically significant predictors (Table 4).
Table 4.
Predictors Associated with Ultrasound (US) Found Steatosis
| Predictor | Estimate (β) | Standard Error | Odds Ratio (95% CI) | P value |
|---|---|---|---|---|
| BMI (kg/m2) | 0.167 | 0.057 | 1.18 (1.06–1.32) | 0.003 |
| AST (U/L) | −0.052 | 0.033 | 0.95 (0.89–1.01) | 0.107 |
| ALT (U/L) | 0.064 | 0.025 | 1.07 (1.01–1.12) | 0.012 |
| Triglycerides (mmol/L) | 0.207 | 0.183 | 1.23 (0.86–1.76) | 0.258 |
| Intercept | −5.495 | 1.819 | 0.004 (0.0001–0.14) | 0.003 |
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index.
The model showed good discriminative ability with an Area Under the Receiver Operating Characteristic curve (AUC) of 0.712, sensitivity of 59.0%, and specificity of 83.1% (Supplementary Table 1 and Supplementary Figure 1).
Model Formulation
On the other hand, the multivariate logistic regression analysis established the weight of each clinical marker. The resulting finalized linear predictive index is defined as:
 |
The optimal discrimination threshold for this score was determined to be ≥ 5.942. Patients scoring ≥ 5.942 are classified as high risk (Predictive of MASLD), whereas individuals scoring < 5.942 are classified as low risk (non-Predictive of MASLD).
Diagnostic Performance and Internal Validation
The diagnostic utility metrics of the model at the optimal Youden cut-off are summarized in Table 5. The model achieved a baseline AUC of 0.756. Internal validation via 1000 bootstrap resamples confirmed the stability of the index, yielding robust 95% confidence bounds across all primary metrics (Figure 3).
Table 5.
Diagnostic Performance Metrics the Diabetic Steatosis Score Model
| Diagnostic Parameter | Point Estimate | 95% Confidence Interval (CI)* |
|---|---|---|
| Area Under Curve (AUC) | 0.756 | 0.677–0.854 |
| Sensitivity | 62.12% | 49.23–89.29% |
| Specificity | 82.00% | 54.35–96.36% |
| Positive Predictive Value (PPV) | 82.00% | Not Applicable |
| Negative Predictive Value (NPV) | 62.12% | Not Applicable |
| Youden Index (J) | 0.4412 | Not Applicable |
Note: *Derived from 1000 bootstrap iterations.
Figure 3.
Receiver Operating characteristic (ROC) curve of the multi-marker logistic regression score for predicting MASLD. The orange solid line shows the discriminative path across varying thresholds, yielding an Area under the curve (AUC) of 0.756 (95% CI: 0.677–0.854). The red dot identifies the optimal diagnostic threshold calculated via the Youden index (J = 0.4412), corresponding to a raw score cut-off of 5.942 (sensitivity: 62.12%, Specificity: 82.00%). The diagonal dashed line represents a random classifier (AUC = 0.50).
Model Calibration
The Hosmer–Lemeshow test demonstrated excellent statistical calibration matching between predicted risks and clinical outcomes (χ2 = 10.20, degrees of freedom = 8, p = 0.251). The non-significant p-value (p > 0.05) signifies that the model’s predictions do not significantly deviate from real-world observations (Figure 4).
Figure 4.
Calibration plot outlining predicted vs observed probabilities of MASLD. The dark blue line maps the mean predicted probability from the logistic regression model against the actual fraction of positive cases grouped into uniform clinical bins. The gray dashed line represents a perfectly calibrated model where predicted probabilities perfectly mirror true disease frequencies. The proximity of the model’s track to the ideal reference line matches the non-significant Hosmer–Lemeshow test result (p = 0.251).
Discussion
Extensive longitudinal research indicates that even among individuals without baseline metabolic dysfunction, an elevated BMI serves as an independent risk factor for both the onset of MASLD and the progression of hepatic fibrosis over a 2- to 8-year period.12,13 In our cohort as well, patients with MASLD had a significantly higher median BMI of 30 kg/m2 compared to 28.1 kg/m2 in the non-MASLD group (p < 0.001).
The heavy reliance of the FIB-4 index on age can introduce confounding in a diabetic cohort, potentially misclassifying older diabetic patients without advanced fibrosis into high-risk categories, while underestimating risk in younger patients with aggressive metabolic disease.14,15
Additionally, ethnic and genetic variations must be considered; the anthropometric characteristics, visceral adiposity distribution, and prevalence of genetic polymorphisms (such as PNPLA3) in the Jordanian and broader Middle Eastern population may alter lipid metabolism and hepatic enzyme expression in ways not accounted for by indices validated primarily in Western cohorts.16–19
Patients with a FIB-4 index > 1.3 in our study are generally older, predominantly male, and show biochemical evidence of more advanced liver stress or dysfunction (higher enzymes, lower platelets/albumin) and poorer lipid profiles, despite having similar BMI and blood sugar control to those in the low-risk category. This observation may be reasonably explained by the progressive nature of this disease. Why males? It seems like mathematics skews higher for males as ALT is physiologically significantly higher for men than for women.20,21 As well, platelets counts are different between males and females.22,23 That was documented in several studies, and it points to the need to have that in mind when using FIB-4 index.24
In our cohort, ultrasound findings did not help predict the FIB-4 index. This discordance aligns with established pathophysiology, as conventional B-mode ultrasound and the FIB-4 index evaluate distinct, non-linear domains of the disease spectrum. Ultrasound is highly sensitive for detecting changes in tissue echogenicity caused by hepatic lipid accumulation (steatosis) but is inherently limited in its ability to detect tissue stiffness or scarring.25–28
In our analysis, an increase in BMI was independently associated with higher odds of a positive ultrasound finding for steatosis, as were elevated ALT levels. These results are highly concordant with established literature. Large-scale population studies, such as the Rotterdam study, have consistently identified total adiposity and hepatocellular enzyme elevation as primary drivers of sonographically detectable liver fat.29,30 Furthermore, this relationship holds true even within populations that already possess widespread metabolic dysfunction. Studies evaluating T2DM cohorts have similarly demonstrated that BMI and ALT remain robust, independent predictors of steatosis, overcoming the baseline confounding of diabetes.31,32 The predictive strength of these two variables justifies their inclusion in historical screening algorithms33–39 and forms the physiological basis for our proposed simplified clinical score.
To predict MASLD, we compared a primary two-parameter model (BMI and ALT; Table 4) with a secondary four-parameter model (BMI, ALT, AST, and triglycerides; Table 1). As shown in Table 5 and the Supplementary Table 1, the second (four-parameter) model yielded slightly better predictive characteristics. However, the performance gap was narrow, suggesting that the simpler, two-parameter model is a highly viable and accessible alternative for clinical risk stratification.
Conclusion
In conclusion, our findings demonstrate that the widely used FIB-4 index possesses limited diagnostic utility for identifying MASLD in this cohort of Jordanian patients with Type 2 Diabetes Mellitus. The index failed to adequately differentiate between patients with and without ultrasound-confirmed hepatic steatosis, highlighting the clinical limitations of applying scoring systems originally derived from viral hepatitis cohorts to populations with complex metabolic phenotypes. Conversely, our newly derived, simplified clinical scoring system—incorporating readily available parameters such as BMI, ALT, AST, and triglycerides—demonstrated superior discriminatory ability (AUC 0.703). These results underscore the critical need to transition toward population-specific, metabolically focused non-invasive screening strategies. Future prospective studies, ideally incorporating advanced modalities such as transient elastography or liver biopsy, are warranted to externally validate this proposed scoring system and refine clinical care pathways for at-risk diabetic populations in the Middle East.
Study Limitations
A retrospective cross-sectional cohort of 186 patients is relatively modest. We acknowledge the need for prospective, multi-center validation.
The data is derived from a Jordanian cohort, the generalizability to other populations with different baseline anthropometric and metabolic characteristics is questionable given the ethnic and genetic variables at play.
The reliance on ultrasound-confirmed hepatic steatosis, while highly practical and standard for non-invasive screening, lacks the sensitivity of a liver biopsy or magnetic resonance elastography (MRE) to definitively stage early fibrosis. That may have contributed to overlapping clinical profiles between the MASLD and non-MASLD groups.
While our novel clinical prediction tool shows high performance in this regional cohort, prospective validation across geographically and ethnically diverse cohorts is essential to demonstrate its broader generalizability.
Funding Statement
No funding was received.
Declaration of Generative AI and AI-Assisted Technologies in the Writing Process
During the preparation of this paper, the authors used Google Gemini 1.5 Pro. This tool was used to refine grammar, and improve overall readability. The authors take full responsibility for the integrity, accuracy, and overall content of the whole paper.
Data Sharing Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Author Contributions
Nesreen A Saadeh: conceptualization, methodology, writing-review and editing. Khaled A Obeidat: formal analysis, writing-original draft, review and editing. Nadeem Wesam Banihani: data curation, formal analysis, and writing-original draft. Deaa Darawsheh: data curation, formal analysis, and writing-original draft. Muayad A Alrawashdeh: data curation, formal analysis, and writing-original draft. Abdelrahman Khaled Ahmad: data curation, formal analysis, and writing-original draft.Rawan M Alnatour: data curation, and writing-original draft. Batool Tariq Mayyas: data curation, and writing-original draft. Shaima Ahmad Almomani: data curation, and writing-original draft. Nasr Alrabadi: supervision, validation, visualization, writing-review and editing. All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
The authors declare that they have no conflict of interest.
References
- 1.Rinella ME, Lazarus JV, Ratziu V, et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature. Ann Hepatol. 2024;29(1):101133. doi: 10.1016/j.aohep.2023.101133 [DOI] [PubMed] [Google Scholar]
- 2.Michalopoulou E, Thymis J, Lampsas S, et al. The triad of risk: linking MASLD, cardiovascular disease and type 2 diabetes; from pathophysiology to treatment. J Clin Med. 2025;14(2):428. doi: 10.3390/jcm14020428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sandireddy R, Sakthivel S, Gupta P, Behari J, Tripathi M, Singh BK. Systemic impacts of metabolic dysfunction-associated steatotic liver disease (MASLD) and metabolic dysfunction-associated steatohepatitis (MASH) on heart, muscle, and kidney-related diseases. Front Cell Dev Biol. 2024;12:1433857. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Younossi ZM, Golabi P, Paik JM, Henry A, Van Dongen C, Henry L. The global epidemiology of nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH): a systematic review. Hepatology. 2023;77(4):1335–12. doi: 10.1097/HEP.0000000000000004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Almahmoud MH, Al Khawaja NM, Alkinani A, Khader Y, Ajlouni KM. Prevalence of fatty liver disease and its associated factors among Jordanian patients with type 2 diabetes mellitus: a cross-sectional study. Ann Med Surg. 2021;68:102677. doi: 10.1016/j.amsu.2021.102677 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hernaez R, Lazo M, Bonekamp S, et al. Diagnostic accuracy and reliability of ultrasonography for the detection of fatty liver: a meta-analysis. Hepatology. 2011;54(3):1082–1090. doi: 10.1002/hep.24452 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.European Association for the Study of the Liver. EASL clinical practice Guidelines on non-invasive tests for evaluation of liver disease severity and prognosis - 2021 update. J Hepatol. 2021;75(3):659–689. doi: 10.1016/j.jhep.2021.05.025 [DOI] [PubMed] [Google Scholar]
- 8.Cleveland E, Gawrieh S. Magnetic resonance imaging-proton density fat fraction for quantification of hepatic steatosis. Clin Liver Dis. 2020;15(1):6–11. doi: 10.1002/cld.874 [DOI] [Google Scholar]
- 9.Chalasani N, Porter L, Bhala N, et al. AASLD practice guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology. 2023;77(5):1797–1835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Castera L, Friedrich-Rust M, Loomba R. Noninvasive assessment of liver disease in patients with nonalcoholic fatty liver disease. Gastroenterology. 2019;156(5):1264–1281. doi: 10.1053/j.gastro.2018.12.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.McPherson S, Hardy T, Henderson E, et al. Evidence of underdiagnosis and undertreatment of nonalcoholic fatty liver disease in the real world: the value of fibrosis scoring tools. Gut. 2017;66(5):865–871. doi: 10.1136/gutjnl-2016-312239 [DOI] [Google Scholar]
- 12.Xu Y, Wang Y, Yao X, et al. Prevalence, incidence, and recovery of metabolic dysfunction-associated steatotic liver disease and associations with weight loss and lipid reduction in a Chinese community-based cohort. J Epidemiol. 2025;35(4):195–205. doi: 10.2188/jea.JE20240224 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Koliaki C, Dalamaga M, Kakounis K, Liatis S. Metabolically healthy obesity and metabolic dysfunction-associated steatotic liver disease (MASLD): navigating the controversies in disease development and progression. Curr Obes Rep. 2025;14(1):46. doi: 10.1007/s13679-025-00637-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.McPherson S, Hardy T, Dufour JF, et al. Age as a confounding factor for the accurate noninvasive diagnosis of advanced NAFLD fibrosis. Am J Gastroenterol. 2017;112(5):740–751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ishiba H, Sumida Y, Tanaka S, et al. The FIB-4 index overestimates liver fibrosis in elderly patients with nonalcoholic fatty liver disease. Semin Liver Dis. 2018;38(4):379–389.30357775 [Google Scholar]
- 16.Chemaitelly H, Abu-Raddad LJ, Al-Kandari W, et al. Epidemiology of non-alcoholic fatty liver disease (NAFLD) in the Middle East and North Africa: systematic review and meta-analysis. Therap Adv Gastroenterol. 2019;12:1756284819845367. [Google Scholar]
- 17.Younossi ZM, Koenig AB, Abdelatif D, et al. Global epidemiology of nonalcoholic fatty liver disease -meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology. 2016;64(1):73–84. doi: 10.1002/hep.28431 [DOI] [PubMed] [Google Scholar]
- 18.Sookoian S, Pirola CJ. Meta-analysis of the influence of I148M variant of patatin-like phospholipase domain containing 3 gene (PNPLA3) on the susceptibility and histological severity of nonalcoholic fatty liver disease. Hepatology. 2011;53(6):1883–1894. doi: 10.1002/hep.24283 [DOI] [PubMed] [Google Scholar]
- 19.Eslam M, Valenti L, Romeo S. Genetics and epigenetics of NAFLD and NASH: clinical impact. J Hepatol. 2018;68(2):268–279. doi: 10.1016/j.jhep.2017.09.003 [DOI] [PubMed] [Google Scholar]
- 20.Prati D, Taioli E, Zanella A, et al. Updated definitions of healthy ranges for serum alanine aminotransferase levels. Ann Intern Med. 2002;137(1):1–10. [DOI] [PubMed] [Google Scholar]
- 21.Dong MH, Bettencourt R, Brenner DA, et al. Serum levels of alanine aminotransferase decrease with age in longitudinal analysis. Clin Gastroenterol Hepatol. 2012;10(3):285–290.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Biino G, Santimone I, Minelli C, et al. Age- and sex-related variations in platelet count in Italy: a proposal of reference ranges based on 40987 subjects’ data. PLoS One. 2013;8(1):e54289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Troussard X, Vol S, Cornet E, et al. Full blood count normal reference values for adults in France. J Clin Pathol. 2014;67(4):341–344. [DOI] [PubMed] [Google Scholar]
- 24.Lonardo A, Nascimbeni F, Ballestri S, et al. Sex differences in nonalcoholic fatty liver disease: state of the art and identification of research gaps. Hepatology. 2019;70(4):1457–1469. doi: 10.1002/hep.30626 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Castera L, Friedrich-Rust M, Loomba R. Noninvasive assessment of liver disease in patients with nonalcoholic fatty liver disease. Gastroenterology. 2019;156(4):1264–1281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Khov N, Sharma A, Riley TR. Bedside ultrasound in the diagnosis of nonalcoholic fatty liver disease. World J Gastroenterol. 2014;20(22):6821–6825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Amer J, Alnees M, Salameh M, et al. The diagnostic utility of FIB-4 as a non-invasive tool for liver fibrosis scoring among NAFLD patients: a retrospective cross-sectional study. Eur Rev Med Pharmacol Sci. 2024;28(8):3104–3111. doi: 10.26355/eurrev_202404_36026 [DOI] [PubMed] [Google Scholar]
- 28.Shaji N, Singhai A, Sarawagi R, Pakhare AP, Mishra VN, Joshi R. Assessment of liver fibrosis using non-invasive screening tools in individuals with diabetes mellitus and metabolic syndrome. Cureus. 2022;14(2):e22682. doi: 10.7759/cureus.22682 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Browning JD, Szczepaniak LS, Dobbins R, et al. Prevalence of hepatic steatosis in an urban population in the United States: impact of ethnicity. Hepatology. 2004;40(6):1387–1395. doi: 10.1002/hep.20466 [DOI] [PubMed] [Google Scholar]
- 30.Koehler EM, Schouten JN, Hansen BE, et al. Prevalence and risk factors of non-alcoholic fatty liver disease in the elderly: results from the Rotterdam study. J Hepatol. 2012;57(6):1305–1311. [DOI] [PubMed] [Google Scholar]
- 31.Leite NC, Salles GF, Araujo AL, Villela-Nogueira CA, Cardoso CR. Prevalence and associated factors of non-alcoholic fatty liver disease in patients with type-2 diabetes mellitus. Liver Int. 2009;29(1):113–119. [DOI] [PubMed] [Google Scholar]
- 32.Lu H, Zeng L, Liang B, Shu X, Xie D. High prevalence of coronary heart disease in type 2 diabetic patients with non-alcoholic fatty liver disease. Arch Med Res. 2009;40(7):571–575. doi: 10.1016/j.arcmed.2009.07.009 [DOI] [PubMed] [Google Scholar]
- 33.Bedogni G, Bellentani S, Miglioli L, et al. The fatty liver index: a simple and accurate predictor of hepatic steatosis in the general population. BMC Gastroenterol. 2006;6:33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Shah AG, Lydecker A, Murray K, Tetri BN, Contos MJ, Sanyal AJ. NASH clinical research network. Comparison of noninvasive markers of fibrosis in patients with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol. 2009;7(9):1104–1112. doi: 10.1016/j.cgh.2009.05.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Assy N, Kaita K, Mymin D, Levy C, Rosser B, Minuk G. Fatty infiltration of liver in hyperlipidemic patients. Dig Dis Sci. 2000;45(10):1929–1934. doi: 10.1023/a:1005661516165 [DOI] [PubMed] [Google Scholar]
- 36.Bazick J, Donithan M, Neuschwander-Tetri BA, et al. Clinical model for NASH and advanced fibrosis in Adult patients with diabetes and NAFLD: guidelines for referral in NAFLD. Diabetes Care. 2015;38(7):1347–1355. doi: 10.2337/dc14-1239 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Wong VW, Adams LA, de Lédinghen V, Wong GL, Sookoian S. Noninvasive biomarkers in NAFLD and NASH-current progress and future promise. Nat Rev Gastroenterol Hepatol. 2018;15(8):461–478. doi: 10.1038/s41575-018-0014-9 [DOI] [PubMed] [Google Scholar]
- 38.Angulo P, Hui JM, Marchesini G, et al. The NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology. 2007;45(4):846–854. doi: 10.1002/hep.21496 [DOI] [PubMed] [Google Scholar]
- 39.Adams LA, Angulo P. Role of liver biopsy and noninvasive tests in NAFLD. Clin Liver Dis. 2007;11(1):25–35. doi: 10.1016/j.cld.2007.02.003 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.