Kenny Walter is an editor with HCPLive. Prior to joining MJH Life Sciences in 2019, he worked as a digital reporter covering nanotechnology, life sciences, material science and more with R&D Magazine. He graduated with a degree in journalism from Temple University in 2008 and began his career as a local reporter for a chain of weekly newspapers based on the Jersey shore. When not working, he enjoys going to the beach and enjoying the shore in the summer and watching North Carolina Tar Heel basketball in the winter.
Currently, guidelines call for liver biopsies to diagnose NASH disease in patients.
Theodore C. Friedman, MD, PhD
There is a need for non-invasive methods to diagnosing nonalcoholic steatohepatitis (NASH) without a liver biopsy.
A team, led by Theodore C. Friedman, MD, PhD, Charles R. Drew University of Medicine and Science, compared 3 non-invasive methods for identifying NASH using data from NHANES III (1988-1994) to determine variables associated with published formulas to identify the disease in data planned to be presented at ENDO 2020, the Endocrine Society’s annual meeting.
NASH is considered a serious liver disease that is marked by hepatic steatosis, cell damage, and inflammation, while increasing the risk of developing cirrhosis and hepatic cancer. Currently, NASH is diagnosed with a liver biopsy, which is both expensive and risky to the patient.
Investigators have attempted to find a non-invasive alternative diagnosis method for identifying patients with NASH, but have yet to develop a technique that can be considered the industry standard and replace liver biopsies.
The investigators used ultrasound data to identify patients with moderate-to-severe hepatic steatosis. Furthermore, they identified the NASH population among those with hepatic steatosis using either the HAIR score, the NASH liver fat score, or the Gholam score.
The investigators developed the HAIR score in a sample of obese patients based on hypertension, insulin resistance, and alanine transaminase (ALT) levels. This method had an AUROC of 0.9, a sensitivity of 0.8, and a specificity of 0.89.
They developed the NASH liver fat score in a Finnish population undergoing gastric bypass and validated this score in an Italian population of liver biopsy patients. The NASH liver fat score incorporates metabolic syndrome, type 2 diabetes, serum insulin, AST, and ALT.
In the Finnish and Italian populations, it had AUROCs of 0.73 and 0.74, sensitivities of 59.5 and 92.9, and specificities of 79.7 and 32.7.
The investigators created the Gholam score in a sample of obese patients using aspartate aminotransferase and type 2 diabetes diagnosis. This method had an AUROC of 0.82, a sensitivity of 0.76, and a specificity of 0.66.
They performed multinomial logistic regression to compare each NASH population to the normal population (those with no or only mild hepatic steatosis).
Overall, the investigators identified 1236 patients having NASH using at least 1 of the 3 methods, with 18% identified by all 3 models and 20% identified by 2 methods. All 3 methods identified significant risk factors for NASH (P <0.05) as being overweight or obese, having elevated AST or ALT levels, and having elevated C-peptide, serum glucose, or serum triglyceride levels.
The investigators also found using the HAIR and Gholam methods Mexican-American participants had a higher risk factor. However, this was not found using the NASH liver fat score.
Being a former alcohol drinker and not meeting guidelines for physical activity were also significant risk factors when using the NASH liver fat score.
“Considerable care must be taken in interpreting risk factors, because the results differ depending which method is used,” the authors wrote. “This could have implications in clinical practice as well, where patients and their risk factors may be misidentified if formulas are used and not for liver biopsy.”
Friedman explained that the non-conclusive nature of the evidence show that before data is needed for NASH diagnosis.
“The 3 non-invasive methods we investigated agreed on a NASH diagnosis for only about one-fifth of the participants in the database,” Friedman said in a statement. “These results imply that better methods are needed to predict NASH.”