Intestinal failure-associated steatosis and fibroblast growth factor 21 plasma levels among adult chronic intestinal failure patients.

BACKGROUND & AIMS: Adult patients with chronic intestinal failure (CIF) may develop intestinal failure-associated steatosis. Asymptomatic steatosis can lead to steatohepatitis and its downstream complications. Monitoring steatosis in daily practice in adult CIF patients is hampered by limited, reliable, accessible, non-invasive methods to measure liver fat content (LFC). Fibroblast growth factor 21 (FGF21) is a hormone that is mainly produced by hepatocytes, and higher plasma levels are associated with the presence and the degree of liver steatosis in several clinical conditions. Furthermore, FGF21 analogues have been shown to reduce fatty liver. FGF21 has previously been suggested as a biomarker for liver steatosis. The aim of this study was to assess the diagnostic performance of FGF21 plasma levels to detect steatosis and steatosis severity in adult CIF patients.

METHODS

FGF21 plasma levels were quantified using enzyme-linked immunosorbent assay (ELISA) in 48 adult CIF patients who had been receiving home parenteral nutrition (HPN) or intravenous fluids for ≥3 months for ≥2 times per week. Liver fat content (LFC, %) was assessed with proton magnetic resonance spectroscopy (1H-MRS). Patient characteristics of patients with steatosis (LFC >5.5 %) and without steatosis (LFC ≤5.5 %) were compared using the Mann-Whitney U test or Fisher's exact test. The diagnostic value of FGF21 levels to diagnose the presence of steatosis (LFC >5.5 %) was performed by determining the area under the receiver operating characteristics curve (AUC), and the optimal cut-off value was determined. Furthermore, Spearman's rho correlation coefficient was calculated to evaluate the association between FGF21 levels and LFC.

RESULTS

FGF21 plasma levels were measured in 48 patients (median age of 56 years, 71 % female) with a median duration of HPN use of 57 months. Steatosis was diagnosed in 8/48 (17 %) patients, with a median LFC of 8.4 % (range 5.7-39.9 %). CIF patients with steatosis had higher median FGF21 plasma levels (658 pg/mL) than patients without steatosis (299 pg/mL). The area under the curve (AUC) of FGF21 to predict steatosis (LFC >5.5 %) was 0.80 [95 % CI 0.63, 0.96]. With the optimal FGF21 cut-off point at 453 pg/mL, the sensitivity as well as the specificity was 75 %. The calculated Spearman rho correlation found a significant positive correlation (ρ = 0.65, p < 0.001) between FGF21 plasma levels and LFC (%).

CONCLUSION

Adult CIF patients with steatosis had higher FGF21 plasma levels than CIF patients without steatosis. FGF21 is a good predictor for diagnosing steatosis and has a good correlation with LFC. FGF21 should be considered as a biomarker for steatosis in adult patients with CIF.