Influence of different frequencies and insertion depths on the diagnostic accuracy of liver elastography by acoustic radiation force impulse imaging (ARFI)

Abstract

Background: Acoustic Radiation Force Impulse Imaging (ARFI) is an innovative elastography for staging of liver fibrosis. We evaluated the diagnostic accuracy of different probes to perform ARFI at different insertion depths.

Methods: In a prospective study, 89 chronic HCV infected patients underwent ARFI elastography using both available probes (c-ARFI: C4-1-MHz; l-ARFI: L9-4 MHz) in comparison to Fibroscan(®). Variability of ARFI elastography at different insertion depths was systematically evaluated in 39 patients (44%). According to Fibroscan(®) elastography, 32 patients (36%) presented with liver cirrhosis, 23 patients (26%) had significant fibrosis and 34 patients (38%) had no significant fibrosis.

Results: Mean propagation velocity with c-ARFI was 1.70±0.67m/s and 1.91±0.87m/s with l-ARFI. Results of both probes were correlated to each other (p<0.001; r=0.70) and to Fibroscan(®) (p<0.001, r=0.82 and 0.84, respectively). In patients with significant fibrosis or with cirrhosis, mean values by l-ARFI were significantly higher than by c-ARFI (p<0.001). For detection of liver cirrhosis, AUROC was 0.97 for c-ARFI (cut-off level 1.72m/s) and 0.90 for l-ARFI (cut-off 2.04m/s). Correlation coefficients of c-ARFI with Fibroscan(®) were highest at an insertion depth of 5-6cm (r=0.882 and 0.864, respectively, p<0.001) and at 3-4cm for l-ARFI (r=0.850 and 0.838, respectively, p<0.001).

Conclusions: ARFI elastography with the linear and with the convex probes showed comparable validity and accuracy in the estimation of liver stiffness. The linear probe gave higher ARFI values. The most accurate insertion depth was 5-6cm for c-ARFI and 3-4cm for l-ARFI indicating that measurements should not be performed close to the liver capsule.