Liver stiffness measurement independently predicts mortality risk in type 2 diabetes

medwireNews: Liver stiffness measurement (LSM) is an independent risk factor for overall mortality in people with type 2 diabetes, even over just 2 years of follow-up, conclude US researchers in a data-linkage analysis.

“These findings support incorporating noninvasive liver stiffness assessment into routine [type 2 diabetes] management to improve risk stratification and identify patients at elevated risk of adverse outcomes,” write Fernando Bril (University of Alabama at Birmingham) and colleagues in JAMA Network Open.

“Future prospective studies and health system-level interventions are needed to validate these findings and explore how best to implement fibrosis screening in clinical practice.”

As “the risk of clinically significant fibrosis and cirrhosis is substantially increased” in people with type 2 diabetes, the American Diabetes Association recommends screening with the Fibrosis (FIB)-4 index, followed by confirmatory testing with the LSM or enhanced liver fibrosis test, the authors explain.

The FIB-4 index has, however, been found to have limited sensitivity, “prompting interest in LSM as a primary screening strategy,” they say.

The team studied adults with complete information on vibration-controlled transient elastography (VCTE) and controlled attenuation parameter (CAP), and no history of chronic liver disease other than metabolic dysfunction-associated steatotic liver disease (MASLD).

MASLD was defined as a CAP of 274 dB/m or above with at least one cardiometabolic risk factor and a weekly alcohol consumption below 140 g for women and 210 g for men. Stage 2 and 3 fibrosis were defined as an LSM of 8.2–9.6 kPa and 9.7–19.9 kPa on VCTE, respectively, while 20.0 kPa or more was classed as cirrhosis. Stage 2 fibrosis was considered clinically significant and stage 3 or 4 fibrosis advanced disease. A FIB-4 index less than 1.30 was considered low risk.

Baseline data, including demographics and laboratory tests, were obtained from the National Health and Nutrition Examination Survey 2017–2018, and linked to the National Center for Health Statistics and National Death Index up to December 2019.

In all, 4102 patients were included in the study. The average age was 47 years, 51% were women, and 63% identified as non-Hispanic White. The mean BMI was 29.5 kg/m².

Over an average follow-up of 24 months, 59 (1.4%) patients died. Compared with those who survived, they were significantly older (62 vs 47 years), more likely to be non-Hispanic White (85 vs 63%), and more likely to have diabetes (36 vs 14%) with a higher glycated hemoglobin level (HbA1c; 6.2 vs 5.6%; 7.3 vs 6.3 mmol/L).

Patients with both MASLD and diabetes had a significantly increased overall mortality risk compared with those with neither condition, at an adjusted hazard ratio (aHR) of 2.77 after accounting for age, sex, BMI, and liver fibrosis using LSM as a continuous variable. By contrast, diabetes alone and MASLD alone were not associated with an increased mortality risk.

The coexistence of advanced liver fibrosis and diabetes was also significantly associated with overall mortality, at an aHR of 6.41, but again the presence of either condition alone was not linked to an increased mortality risk.

Further investigation of the impact of liver disease among people with diabetes indicated that LSM, as assessed on VCTE, was independently associated with overall mortality, at an aHR of 1.06 for each 1-kPa increment in the LSM, whereas the FIB-4 index was not independently associated with overall mortality.

And the presence of advanced fibrosis versus no clinically significant fibrosis was also significantly associated with overall mortality in patients with diabetes (aHR=4.24), as was cirrhosis (aHR=11.13). The mortality risk was again significantly increased in patients with uncontrolled diabetes (HbA1c ≤8 vs >8%; ≤10.2 vs >10.2 mmol/L) in combination with LSM above 8 kPa (aHR=11.71), compared with controlled diabetes and low LSM, but it was not significantly increased in patients with uncontrolled diabetes and low LSM or those with controlled diabetes and high LSM.

In an accompanying commentary, Emad Qayed (Emory University School of Medicine, Atlanta, Georgia, USA) says the study “provides a strong argument to reconsider current strategies for detecting liver fibrosis in patients with diabetes,” highlighting “that reliance on FIB-4 as the initial screening tool may result in a substantial proportion of patients with high-risk diabetes being missed” and therefore excluded from “emerging pharmacologic therapies for MASLD.”

He adds, however: “At present, universal implementation of LSM is impractical, given the size of the population with diabetes, limited access to transient elastography, and logistical, cost, and infrastructure constraints,” underlining the need for “more accurate, low-cost biomarkers that can approximate the prognostic performance of liver stiffness measurement and guide appropriate risk stratification and treatment.”

medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2026 Springer Healthcare Ltd, part of Springer Nature

JAMA Netw Open 2026; 9: e260762
JAMA Netw Open 2026; 9: e260674