Feasibility of liver shear wave elastography with different transducers

Abstract

The usefulness of Shear Wave elastography (SWE) in the diagnosis of liver fibrosis has been investigated by several groups. Admitted values for fibrosis staging were often obtained using a dedicated abdominal probe (6 MHz). This study investigates the impact of the transducer frequency on Shear-wave elastography (SWE) measurements. Quantitative liver SWE measurements were obtained for 62 patients with various body morphologies and liver textures using a SuperSonic Imagine machine. For each patient, measurements were performed with two transducers (SC6-1 and SL10-6) with the same scanning preset. The mean (E) and standard deviation (SD) of the elasticity values in circular regions of interest obtained with both probes were compared. The quality of the exam was also computed as the percentage of non-filling (PNF), the temporal variability (TV) and the spatial variability (SV) of the elasticity map. SWE measurements failed in 4 patients. Correlation of elasticity estimates between the two probes was poor (r=0.35, p<10−2) for the whole population but correlated for normally weighted patients and thin skin-to-liver-surface thickness (SLT) (r=0.80 and 0.78 respectively, p<10−5). Elasticity estimates did not agree: E values obtained with the 10 MHz probe were about 20% higher (p<10−2) than that obtained using the 6 MHz probe. In phantom study, elasticity estimates were the same for all probes. In all cases, quality indexes as PNF and TV were better with the higher frequency probe (mean PNF 2.8% for SL10-2 vs 11.1% for SC6-1, p<10−3 and mean TV of 0.5 for SL10-2 vs 0.9 for SC6-1, p<10−5). Mean SV did not differ between the two probes (6.07 vs 7.26, NS). The SL10-2 probe could be an alternative to the SC6-1 when the elasticity map filling is poor, keeping in mind for staging purposes that the measured elasticity depends on the frequency.