Ultraschall Med 2016; 37 - SL19_6
DOI: 10.1055/s-0036-1587809

Inter/intra-observer reproduciibility of a 2D- Shear Wave Elastography (Logiq E9 system from GE) technique and the impact of ultrasound experience in achieving reliable data

TV Moga 1, C Pienar 1, AM Stepan 1, A Popescu 1, R Sirli 1, M Danila 1, I Sporea 1
  • 1University of Medicine and Pharmacy "Victor Babes", Timisoara, Romania

The aim of this paper was to evaluate the reproducibility of a new SWE technique -2D SWE.GE and the impact of ultrasound experience in acquiring reliable measurements, since no recommendation are available.

Liver stiffness measurements were made in 60 healthy subjects by 2D-SWE.GE using Logiq E9. Three examiners, (one with both ultrasonographic and elastographic experience of more than one year – E1; one with only ultrasonographic experience for more than one year – E2; and one without any experience – E3) performed 10 valid measurements for each subject and a median and interquartile range (IQR) were calculated (m/s). The inter-observer reproducibility was assessed by calculating the interclass correlation coefficients (ICCs) for median and IQR. For the intra-observer reproducibility, we calculated, for each examiner, the medians for the first five and the last five measurements, respectively. We then calculated, for each examiner, the ICCs for the two medians.

The final study group included 58 subjects: 56.9% women, with a mean age of 32.69 ± 13.37 years and mean BMI 22.68 ± 3.96 Kg/m2. 92.7% were healthy volunteers and 8.3% had compensated liver cirrhosis. The ICCs for median values of valid measurements were excellent both between E1 and E2, and E1 and E3 (0.872, 95% CI: 0.784 – 0.924 vs. 0.842, 95% CI: 0.735 – 0.907), p < 0.0001. The ICCs for IQR were only fair between E1 and E2 (0.411, 95%CI: 0.033 – 0.645, p = 0.018) and between E1 and E3 (0.443, 95% CI: 0.055 – 0.672, p = 0.004). The agreement between measurements was excellent for each examiner, however the ICCs were higher for the more elastographic experienced examiner: 0.956, (95% CI: 0.925 – 0.974) vs. 0.953 (95% CI: 0.920 – 0.972) vs. 0.984 (95% CI: 0. 973 – 0.991).

The good ICCs for the median values show that 2D-SWE.GE is a reproducible method for liver stiffness measurements; however experience plays a role in obtaining reliable results.