This article is more than 5 years old. Information may no longer be current.

Lower PDFF cut-off values increased sensitivity of steatosis diagnosis

Results of a recently published study showed that the currently implemented cut-off value of 5% for diagnosing liver steatosis with in vivo proton magnetic resonance spectroscopy to calculate proton density fat fraction is too high.

“While it is important to understand the upper limit of normal for hepatic lipid content by imaging, it is unknown whether mild steatosis is associated with adverse clinical outcome,” the researchers wrote. “Given that PDFF is more accessible than liver biopsy, the lower we make the threshold for diagnosis of pathological lipid content, the more likely we are to count benign physiological cases of steatosis as disease.”

The researchers aimed to assess the reliability and accuracy of in vivo proton magnetic resonance spectroscopy proton density fat fraction (¹H-MRS PDFF) in diagnosing hepatic steatosis using histopathological grading as a reference standard and compare with lower cut-off values.

The researchers enrolled 94 patients with elevated liver enzymes who required a biopsy for diagnosis in the study between 2007 and 2014. Mean patient age was 53.3 years and 55% were men. Mean fasting glucose was 109 mg/dL and mean BMI was 26.3 (63% of patients had a BMI of 25 kg/m² or greater).

Twenty-five patients with ¹H-MRS PDFF of 5% or greater had steatosis according to the histopathologist (100% specificity). Twenty-two of 69 patients with ¹H-MRS PDFF of less than 5% were also diagnosed with steatosis by the histopathologist (53% sensitivity).

Thirty-seven patients were diagnosed with non-alcoholic fatty liver disease. Reducing the ¹H-MRS PDFF standard cut-off value of 5% to 3% correctly identified NAFLD in 31 out of 37 patients (100% specificity and 79% sensitivity) and reduction to 2% correctly identified NAFLD in 35 out of 37 patients (94% specificity and 87% sensitivity).

The researchers concluded that if the cut-off value of 5% had been used to define hepatic steatosis with ¹H-MRS PDFF, 12 out of 37 patients with elevated liver enzymes due to NAFLD would have been incorrectly considered to not have steatosis. Additionally, 11 of those 12 patients who had PDFF values of less than 5% had stage 1 steatosis and four patients had stage 3 or 4 advanced fibrosis. “Thus, our results strongly indicate that NAFLD patients with PDFF [of less than 5%] constitutes a clinically important group in terms of diagnosis and prognosis.” – by Talitha Bennett

Disclosure: The researchers report no relevant financial disclosures.