Introduction
Nonalcoholic fatty liver disease (NAFLD) has emerged as a global health challenge, particularly among individuals with metabolic disorders like Type 2 Diabetes. The degree of liver fibrosis is the most significant prognostic factor for these patients, determining their long-term clinical outcomes. While liver biopsy remains the "gold standard" for staging fibrosis, its invasive nature and potential for sampling error have led to the widespread adoption of non-invasive alternatives. FibroScan, which measures liver stiffness (LSM) via transient elastography, is a leading tool in clinical practice. However, its accuracy can be influenced by extra-hepatic factors. This study investigates how abnormal glucose metabolism affects LSM readings in NAFLD patients.
Methods
This cross-sectional study included 91 patients with NAFLD and abnormal glucose metabolism (defined as fasting glucose ≥ 5.6 mmol/L or 2 h glucose ≥ 7.8 mmol/L) who underwent liver biopsy at Zhongshan Hospital, Fudan University. LSM was performed using the FibroScan M probe. The diagnostic performance of LSM was evaluated using receiver operator characteristic (ROC) curves, with biopsy results as the reference. Multivariate linear regression and subgroup analyses were conducted to identify metabolic indicators independently associated with LSM values after adjusting for histological findings like inflammation and ballooning.
Results
FibroScan demonstrated strong diagnostic performance, particularly for advanced stages of fibrosis. The area under the ROC curve (AUROC) for detecting cirrhosis (Stage 4) was 0.902, while the AUROC for Stage ≥3 was 0.837. The optimal cut-off values for stages ≥1, ≥2, ≥3, and 4 were 6.3 kPa, 7.6 kPa, 8.3 kPa, and 13.8 kPa, respectively.
A key finding was the impact of metabolic "noise" on these measurements. Multivariate analysis revealed that hemoglobin A1c (HbA1c) and aspartate aminotransferase (AST) were independently associated with LSM values, even after adjusting for histological fibrosis stage. Subgroup analysis showed that patients with poor glycemic control (HbA1c ≥ 7%) or obesity (BMI ≥ 30 kg/m²) exhibited higher LSM values regardless of actual scarring. For instance, the cut-off for significant fibrosis (Stage ≥2) shifted from 7.7 kPa in patients with well-controlled sugar (HbA1c < 7%) to 8.3 kPa in those with poor control (HbA1c ≥ 7%).
Discussion
The study confirms that while FibroScan is a valuable tool for evaluating liver fibrosis in NAFLD patients with glucose metabolism issues, its readings are not purely reflective of physical scarring. The independent association of HbA1c and AST with LSM suggests that metabolic state and active liver inflammation can artificially inflate stiffness readings. This "metabolic skew" is particularly relevant when diagnosing early-to-moderate fibrosis, where the margin for error is smaller. The increased distance ultrasound waves must travel in patients with higher BMI also likely contributes to signal attenuation and higher readings.
Conclusion
FibroScan is highly effective for ruling out advanced fibrosis and cirrhosis in NAFLD patients with abnormal glucose metabolism. However, clinicians must interpret LSM values within the context of the patient's metabolic profile. Future clinical guidelines may benefit from stratified cut-off values that account for HbA1c and BMI to improve diagnostic precision and facilitate personalized patient management.
Summary updated