Ultraschall Med 2016; 37 - SL19_3
DOI: 10.1055/s-0036-1587806

The performance of 2D SWE.GE compared to transient elastography for the evaluation of liver stiffness

I Sporea 1, F Bende 1, R Sirli 1, A Popescu 1, M Danila 1, R Mare 1, AM Stepan 1, R Lupusoru 1
  • 1University of Medicine and Pharmacy "Victor Babes", Gastroenterology, Timisoara, Romania

Aim: To evaluate the performance of 2D-SWE.GE in noninvasive fibrosis assessment as compared to a validated method-Transient Elastography (TE).

Material and Method: Our study included 255 consecutive subjects with or without chronic hepatophaties (only compensated liver disease evaluated for decision regarding treatment), in which liver stiffness (LS) was evaluated in the same session by means of 2 elastographic methods: TE (M or XL probes) and 2D-SWE.GE (LOGIQ E9, General Electrics Healthcare). Reliable LS measurements were defined as follows: for TE -the median value of 10 measurements with a success rate of ≥60% and an interquartile range< 30% and for 2D-SWE.GE – the median value of 10 measurements acquired in a homogenous area and an interquartile range (IQR)< 30%.

Results: Reliable LS measurements were obtained in 241/255 (94.5%) subjects by 2D-SWE.GE, and in 240/255 (94.1%) by TE (p = 0.997). The final analysis was performed on 229 subjects with valid measurements by both methods. Based on TE cut-off values (F1 – 6, F2 – 7.2, F3 – 9.6 and F4 – 14.5 kPa), we divided our cohort into 3 groups: F< 2: 62/229 (27%); F2-F3: 44/229 (19.3%); F = 4: 123/229 (53.7%). We found a strong correlation between the LS values obtained by the 2 methods: r = 0.833, p < 0.0001. The mean values obtained by 2D-SWE.GE considering TE cut-off values as reference were: F0-F1: 5.88 ± 1.39; F2-F3: 9.86 ± 2.21; F4: 13.57 ± 2.76 (p < 0.001). The best cut-off values for 2D-SWE.GE were: F> 2 – 6.7 kPa (AUROC = 0.975, Sensitivity = 96.97%, Specificity = 87.1) and for F> 4: 10.7 kPa (AUROC = 0.911, Sensitivity = 81.68%, Specificity = 84.37%).

Conclusions: 2D-SWE.GE and TE (using M and XL probes) had good feasibility (94.5% and 94.1%), with no statistical differences between them (p = 0.997). There was a strong correlation between the two methods, with LS values significantly increasing with the severity of fibrosis. The best 2D-SWE.GE cut-off values for predicting F> 2 and F> 4 was 6.7 kPa, and 10.7 kPa.