The rising incidence of obesity and metabolic syndrome has led to a dramatic increase in non-alcoholic fatty liver disease (NAFLD). NALFD is characterized initially by isolated hepatic steatosis which in some patients progresses to steatohepatitis (NASH) and eventually cirrhosis. Reductions in steatosis, fibrosis, and inflammation have been observed histologically as a result of weight loss surgery (WLS), indicating treatment response (1). Non-invasive biomarkers that track treatment response are desirable, specifically steatosis using quantitative water-fat imaging (2) as well as fibrosis and inflammation with MR elastography (3,4). The purpose of this study was to measure liver stiffness reduction after WLS to determine which histological features were predictive of a larger decrease in liver stiffness.

Patients undergoing WLS (vertical gastrectomy or Roux-en-Y gastric bypass) were recruited at the University of California – San Diego and the University of Wisconsin – Madison for IRB-approved MRI studies at 2 times points: 1-2 days prior to surgery and approximately 6 months after surgery. All imaging was performed at either 1.5T or 3T systems (Signa HDxt and MR750, GE Healthcare, Waukesha. WI).

At both time points, 2D GRE MRE was performed (1.5T/3T): TR=50ms, TE=22/19ms, in-plane resolution range= 1.9x3.1mm – 3.8x5.0mm, slice thickness=5-10cm, shear wave frequency=60Hz. In addition, chemical shift encoded fat quantification was performed (1.5T/3T): 6 echoes, TR=13.4/8.6ms, ΔTE=2.0/1.0ms, flip angle=5°/3°, resolution=1.7x2.8x8 /1.7x3.4x8mm.

Intraoperative biopsies were assessed by two experienced pathologists for steatosis, inflammation and fibrosis using the NASH CRN histologic scoring system (5). For each histological feature, the staging score was reduced to either a feature present or feature absent classification. Patients with stage 0 fibrosis, grade 0 steatosis and grade 0 or 1 NASH CRN were classified as feature absent. Similarly, patients with fibrosis stage ≥ 1, steatosis grade ≥ 1, and NASH CRN grade ≥ 2 were classified as feature present.

Wilcoxon rank sum tests were performed to test for statistical difference in stiffness changes between histologically defined subgroups: fibrosis, inflammation, steatosis, suspicious or definite NASH. Spearman’s correlations were used to assess relations between changes in stiffness with changes in proton density fat fraction (PDFF) and body mass index (BMI).

22 patients were successfully recruited (17 females, 5 males, age = 46.3±10.1 years, BMI at surgery= 41.8±6.1 kg/m², BMI at 6 months after surgery=33.5±5.4 kg/m²) There was a significant overall reduction in liver stiffness (Figure 1) 6 months after WLS (2.95±1.03 kPa, 2.58±0.72 kPa, p-value = 0.04 ). The reduction in stiffness for patients with fibrosis was significantly greater than for patients without fibrosis (Figure 2). No statistical difference was observed in patients with inflammation, steatosis, or suspicious or definite NASH (Table 1). Additionally, no significant correlations between changes in stiffness with changes in PDFF (r=-0.06, p-value=0.7779) or BMI (r=0.29, p-value=0.1892) were found. Repeated measurements of liver stiffness immediately prior to and 6 months after WLS showed significant reductions in liver stiffness, especially in patients with baseline fibrosis. These findings suggest that weight loss and WLS led to improvement in hepatic fibrosis, although future research with end-of-treatment biopsies are needed for confirmation. If validated by such research, our results support the use of MRE as a non-invasive biomarker to monitor hepatic treatment response after weight loss surgery.