From the AGA Journals

Liver stiffness predicts hepatic events in NAFLD

View on the News

Study highlights potential for noninvasive LSMs

As the prevalence of nonalcoholic fatty liver disease (NAFLD) continues to rise, risk-stratifying those who will develop liver-related complications remains a major challenge. Although progression of liver fibrosis is a key risk factor for developing liver-related complications, the clinical application of noninvasive fibrosis markers for prognostication has been largely unexplored in NAFLD.

Dr. George Cholankeril, Baylor College of Medicine, Houston

Dr. George Cholankeril

This study by Dr. Petta and colleagues highlights the potential for liver stiffness measurements (LSMs) as a noninvasive method. Increased LSM that was suggestive of clinically significant portal hypertension (kPa >21) had a nearly fourfold risk of hepatic decompensation. Furthermore, a longitudinal increase in LSM by greater than 20% was associated with a greater than 50% increased risk for hepatic decompensation, hepatocellular carcinoma, and death.

Transient elastography is a widely available and accurate tool for the noninvasive assessment of liver fibrosis for NAFLD in routine clinical practice. Routine serial measurements of LSM with transient elastography during clinic visits can provide clinicians with important information in the management of NAFLD, which can aid in treatment decisions, response to therapy, and monitoring of disease progression.

Further research is needed to validate these findings and to evaluate how longitudinal changes in LSM and other noninvasive fibrosis markers can prognosticate outcomes in NAFLD.

George Cholankeril MD, MS, is an assistant professor in the section of gastroenterology & hepatology of the department of medicine and in the division of abdominal transplantation of the department of surgery at Baylor College of Medicine in Houston. He reported having no conflicts of interest.



Among patients with nonalcoholic fatty liver disease (NAFLD) and compensated advanced chronic liver disease, liver stiffness measurements (LSMs) are associated with risks of hepatic events, according to a retrospective analysis of more than 1,000 patients.

“[N]oninvasive markers that can predict liver disease severity and outcomes in patients with NAFLD and advanced fibrosis are a major unmet need,” wrote lead author Salvatore Petta, MD, of the University of Palermo, Italy, and colleagues. Their report is in Clinical Gastroenterology and Hepatology. “Data about the accuracy of LSM in the prediction of events in NAFLD, and especially in patients with NAFLD and F3-F4 fibrosis, are scarce.”

To address this knowledge gap, the investigators retrospectively analyzed data from 1,039 consecutive patients with NAFLD who had baseline LSMs of more than 10 kPa and/or histologically diagnosed F3-F4 fibrosis. Patients were prospectively recruited at 10 centers in 6 countries, then followed for a median of 35 months, ranging from 19 to 63 months.

All patients had their liver stiffness measured with an M or XL probe at baseline. In addition, approximately half of the patients (n = 533) had a follow-up measurement using the same method, generating a subgroup with changes in liver stiffness. “Improved” liver stiffness was defined as a decrease in LSM greater than 20% from baseline, “impaired” liver stiffness was defined as an increase in LSM greater than 20% from baseline, and “stable” liver stiffness was defined as a change falling between 20% lower and 20% higher than baseline.

At baseline, mean LSM was 17.6 kPa. Cox regression analysis revealed that baseline LSM was independently associated with HCC (hazard ratio, 1.03; 95% confidence interval, 1.00-1.04; P = .003), liver decompensation (HR, 1.03; 95% CI, 1.02-1.04; P < .001), and liver-related death (HR, 1.02; 95% CI, 1.00-1.03; P = .005), but not extrahepatic events.

According to the investigators, the association between LSM at baseline and risk of liver decompensation was maintained after adjustment for the severity of liver disease and for surrogate markers of portal hypertension, they noted. Furthermore, patients with a baseline LSM of at least 21 kPa – which indicates high risk of clinically significant portal hypertension (CSPH) – were at greater risk of liver decompensation than were those with an LSM less than 21 kPa (HR, 3.71; 95% CI, 1.89-6.78; P = .04).

In the subgroup with follow-up measurements, approximately half of the patients had an improved LSM (53.3%), while 27.2% had a stable LSM, and 19.5% had an impaired LSM, a pattern that was significantly associated with diabetes at baseline (P = .01).

“These data agree with the available literature identifying diabetes as a risk factor for liver disease progression and liver-related complications,” the investigators wrote.

Cox regression showed that, among those with follow-up LSM, changes in LSM were independently associated with HCC (HR, 1.72; 95% CI, 1.01-3.02; P = .04), liver decompensation (HR, 1.56; 95% CI, 1.05-2.51; P = . 04), liver-related mortality (HR, 1.96; 95% CI, 1.10-3.38; P = .02), and mortality of any cause (HR, 1.73; 95% CI, 1.11-2.69; P = .01).

These risks could be further stratified by level of change in liver stiffness, with greater impairment predicting greater risk: The crude rate of liver decompensation was 14.4% among those with impaired LSM, compared with 6.2% among those with stable LSM and 3.8% among those with LSM improvement. That said, the categories of changes in LSM were not predictive of decompensation among patients with high risk of CSPH at baseline; however, they remained predictive among those with low risk of CSPH at baseline.

“[T]his study … showed that an integrated assessment of baseline LSM or [changes in LSM] can help in stratifying the risk of development of liver-related complications and of both hepatic and overall mortality,” the investigators concluded. “These data, if further validated, could help personalize prognosis and follow-up in NAFLD with [compensated advanced chronic liver disease].”

The investigators disclosed relationships with AbbVie, Novo Nordisk, Gilead, and others.

Next Article: