Burden of non-alcoholic fatty liver disease and usefulness of non-invasive tests to identify advanced liver disease in patients with type 2 diabetes mellitus: interimanalysis of an Italian prospective multicentre study

Background and Aims: Patients with type 2 diabetes mellitus (T2DM) are at increased risk of advanced liver fibrosis and should be entered into risk stratification pathways for appropriate referral to hepatologists. An optimal model of care should allow the identification of the largest proportion of high-risk patients without overwhelming hepatology clinical services. In a cohort of patients with T2DM consecutively referred by diabetes Units, we aimed to investigate the efficacy of different models of care based on Fibrosis-4 (FIB-4) and/or vibration controlled transient elastography (VCTE) for the identification of patients needing referral to hepatology clinic. Method: From April 2021 to October 2023, a total of 706 consecutive T2DM patients at their first referral to 6 different diabetology clinics were prospectively enrolled. All patients underwent liver stiffness measurement (LSM) by VCTE and liver steatosis assessment by controlled attenuation parameter (Fibroscan, Echosens, France). A LSM cut-off value of 8.0 kPa was used to rule out advanced liver fibrosis. Patients were stratified according to FIB-4 values <1.30, 1.30–2.67, and >2.67. Results: Patients’ median age was 59 (IQR 53–65) years and 411 (58.2%) of them were males; 370 (52.4%) patients were obese (BMI ≥ 30 kg/m2) and 446 (64.3%) had arterial hypertension. Median LSM was 5.3 (IQR 4.3–6.5) kPa; 111 (14.6%) of patients had LSM ≥ 8.0 kPa. Median FIB-4 value was 1.05 (IQR 0.80–1.40); 490 (69.4%) patients had FIB-4 < 1.30, 197 (27.9%) had FIB-4 between 1.30 and 2.67, while 19 (2.7%) patients showed FIB-4 values ≥ 2.67. In our cohort of T2DM patients, a 2-tier screening for advanced liver fibrosis by FIB-4 followed by VCTE would have led to 50 (7.1%) patients referred to the hepatologist, despite a false negative rate of 8.6% (n = 61). A single tier screening based on FIB-4 ≥1.3 would have resulted in an overflow of patients (n = 216; 30.6%) to hepatology clinic, while a FIB-4 >2.67 would have caused a disproportionate rate of false negative patients (n = 103; 14.6%). Finally, a first line screening based on VCTE would have resulted in a referral rate to hepatologists of 14.6% (n = 111) with a low probability of missing patients at risk of advanced liver fibrosis. However, first-line screening with VCTE could be unfeasible in resource-limited settings. Conclusion: In diabetology clinics, the most appropriate screening approach for the identification of high-risk patients that require hepatologist’s referral should be determined according to local health-care resources.