Reuters Health Information: Non-invasive tests for NAFLD are unreliable in South Asians
Non-invasive tests for NAFLD are unreliable in South Asians
Last Updated: 2017-12-27
By Lorraine L. Janeczko
NEW YORK (Reuters Health) - Blood-test-based non-invasive tests for non-alcoholic fatty liver disease (NAFLD) are unreliable alternatives to liver biopsy in South Asians, new research suggests.
Non-invasive liver tests (NILTs) include those based on biomarkers or on clinical and laboratory data such as NAFLD fibrosis score (NFS); Fibrosis-4 (Fib-4); BMI, AST/ALT ratio, diabetes (BARD); aspartate transaminase (AST)-to-platelet ratio index (APRI); and the AST/alanine transaminase (ALT) ratio, researchers note in Frontline Gastroenterology, online November 16.
"Liver stiffness, measured by transient elastography (TE), acoustic radiation force impulse or MRI, can be a surrogate marker of fibrosis, but requires specialist equipment and/or skilled personnel to conduct the tests," write Dr. Wing-Kin Syn of the Medical University of South Carolina in Charleston and colleagues.
"Providers should consider ethnicity when recommending assessment and treatment algorithms for" NAFLD, Dr. Syn told Reuters Health by email. "Current guidance advocates the use of noninvasive tests to stratify patients into risk groups. However, most of the work to develop and validate the scores has been done in largely Caucasian populations. The applicability of these NILTs to different patient groups - including those of different ethnicities - has yet to be determined."
Dr. Syn and colleagues conducted a retrospective cross-sectional study of 175 patients with histologically confirmed NAFLD who were being treated at one specialist liver center. Their ethnicities were South Asian (Indian, Pakistani, Bangladeshi, Sri Lankan and Nepalese; 90 patients), white (79), black (2), and East Asian and other (4).
South Asians were almost a decade younger than whites (44 vs. 52 years), and fewer of them were obese (60% with BMI >27.5 kg/m2 vs. 75% with BMI >30 kg/m2). There were no significant differences in the grade or stage of liver injury, however.
NILTs were less sensitive in the South Asian patients (range, 0.09-0.52) compared with the whites (range, 0.50 - 0.75), but specificities were similar. The relative risk of correct diagnosis in white patients compared with South Asians was 1.85 (95% confidence interval, 1.35 to 2.56).
In binary logistic regression analyses, ethnicity, platelet count, and serum ALT predicted accuracy. In both whites and South Asians, transient elastography was highly accurate.
If the new "data can be confirmed, they should prompt efforts to develop adjustment of the blood-based test scores based on ethnicity," Dr. Raymond T. Chung, director of hepatology at Massachusetts General Hospital in Boston, told Reuters Health by email. "They suggest that one size (or score) does not fit all."
Dr. Chung, who was not involved in the study, added that he was somewhat surprised by the results because the original studies developing these indices were performed across multiple ethnic groups.
"However," he noted, "it is quite possible that South Asians were not prominently represented in these studies."
Dr. Syn cautioned, "Because the sensitivity of these commonly used tests is lower in patients of South Asian ethnicity, it suggests that a significant number of South Asian patients may be inappropriately reassured that they do not have advanced disease. As with any clinical test, there is a danger that, if a test is applied or interpreted inappropriately, wrong decisions can be made and harm may ultimately come to a patient."
Frontline Gastroenterol 2017.