To investigate the value of ultrasound fat fraction (UDFF) in evaluating the severity of disease and glucose lipid metabolism in young and middle-aged patients with metabolism-related fatty liver disease (MAFLD).

A retrospective analysis was carried out on the clinical data of 152 young and middle-aged MAFLD patients admitted to the Second People's Hospital of Neijiang City from March 2022 to March 2025. All patients were examined for UDFF and magnetic resonance proton density fat fraction (MRI-PDFF). According to the pathological grade of liver steatosis, all patients were divided into two groups: 102 cases in the mild fatty liver disease group (PDFF S1) and 50 cases in the moderate-to-severe fatty liver disease group (PDFF S2、S3); the relevant indicators related to glycolipid metabolism [fasting blood glucose (FBG), glycated hemoglobin (HbA1c), insulin resistance index (HOMA-IR), total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C)] of the patients were collected; and the patients with MAFLD were grouped based on the different degrees of glycolipid metabolism, they were also grouped based on whether there was obvious liver fibrosis, among them, patients with obvious liver fibrosis were classified as group A and patients without/with minor liver fibrosis were classified as group B, and the UDFF differences between different groups were compared. The receiver operating characteristic (ROC) curve was plotted, and the area under the curve (AUC) was calculated to evaluate the diagnostic value of UDFF for the disease classification and glycolipid metabolism of MAFLD patients.

The levels of FBG, HOMA-IR, TC, TG and LDL-c in the mild group were higher than those in the moderate to severe group, while HbA1c, HDL-c and UDFF were lower than those in the moderate to severe group, with statistically significant differences (P<0.05). The UDFF in group A was (7.66±1.20)%, which was lower than that in group B [(12.23±3.49)%], with a statistically significant difference (t=11.400, P<0.001). The results of Logistic regression analysis showed that UDFF was positively correlated with the severity of MAFLD (OR=7.503, 95% CI: 3.427-16.456, P<0.001), the ROC curve showed that the AUC of UDFF in evaluating the severity of the disease in patients with MAFLD was 0.856 (95% CI: 0.825-0.886), the optimal cut-off value was 18.5%, the sensitivity was 86%, and the specificity was 82%. The results of Logistic regression analysis showed that UDFF was significantly positively correlated with abnormal glycolipid metabolism (P<0.001), for every 1% increase in UDFF, the risk of abnormal glycolipid metabolism in patients with MAFLD increased by 132% (OR=2.325, 95% CI: 1.852-2.912, P<0.001), the ROC curve showed that the AUC of UDFF in evaluating the degree of glycolipid metabolism in patients with MAFLD was 0.873, the optimal cut-off value was 9.5%, the sensitivity was 80.3%, and the specificity was 76.5%.

By quantifying liver fat content, UDFF has a good diagnostic efficacy for MAFLD patients with different severity of disease, which helps to evaluate the degree of abnormal glucose and lipid metabolism and liver fibrosis, and provides an important reference for clinical diagnosis and treatment.