The diagnostic value of controlled attenuation parameter (CAP) and magnetic resonance imaging derived proton density fat fraction (MRI-PDFF) for the evaluation of intestinal failure-associated liver steatosis among adult chronic intestinal failure patients.

BACKGROUND & AIMS: Chronic intestinal failure (CIF) can cause intestinal failure-associated liver disease (IFALD), with hepatic steatosis being a predominant feature in adult CIF patients. Asymptomatic steatosis may progress to steatohepatitis and its downstream complications. While proton magnetic resonance spectroscopy (1H-MRS) is considered one of the most accurate non-invasive techniques for quantifying liver steatosis, its high cost and limited availability hinder widespread clinical use. Therefore, more accessible and reliable alternatives are needed for routine clinical assessment of liver fat in CIF patients in daily practice. Controlled Attenuation Parameter (CAP) is a non-invasive, ultrasound-based method, obtained via transient elastography, to quantify hepatic steatosis. This study aims to evaluate the diagnostic accuracy of controlled attenuation parameter (CAP) and magnetic resonance imaging-derived proton density fat fraction (MRI-PDFF) for assessing liver steatosis in CIF patients, using 1H-MRS as the reference standard.

METHODS

Sixty adult CIF patients, receiving home parenteral nutrition or intravenous fluids for at least three months, were enrolled. Liver fat content (LFC) was measured using 1H-MRS. The diagnostic performance of CAP and MRI-PDFF in detecting liver steatosis (defined as LFC >5.5 %) was evaluated through receiver operating characteristic (ROC) curve analysis, and optimal cut-off values were determined. Spearman's rho correlation was used to assess associations between LFC as measured by 1H-MRS and the two non-invasive modalities.

RESULTS

Valid liver fat measurements were obtained in 60 patients with 1H-MRS, 53 with MRI-PDFF and 34 with CAP. According to 1H-MRS, liver steatosis was present in 17 % of patients. The Spearman correlation coefficients between 1H-MRS and CAP and between 1H-MRS and MRI-PDFF were 0.70 and 0.82 respectively. The area under the curve for CAP and MRI-PDFF in detecting steatosis (LFC >5.5 %) were 0.95 (p < 0.001, CI 0.88-1.00) and 1.00 (p < 0.0005, CI 1.00-1.00) respectively.

CONCLUSION

Both MRI-PDFF and CAP demonstrate high diagnostic accuracy in detecting liver steatosis (LFC >5.5 %) in adult CIF patients. Additionally, CAP shows a strong correlation, while MRI-PDFF exhibits an excellent correlation with severity of 1H-MRS-derived LFC. Both non-invasive techniques, may serve a valuable role in routine assessment of steatosis and steatosis severity in this population, potentially improving early diagnosis, monitoring and understanding of IFALD pathogenesis, ultimately improving patient outcomes.