Comparison between two 2D-SWE techniques using transient elastography as a reference method for liver stiffness assessment

Objectives Ultrasound-based liver elastography techniques are non-invasive methods used for the assessment of liver stiffness (LS). In addition to Transient Elastography (TE), new methods were developed. Aim to compare the performance of 2D-SWE technique implemented on two different ultrasound probes from different vendors for the assessment of liver stiffness measurements (LSM) using transient elastography (TE) as reference method. Materials A prospective study was conducted in which LSM were performed in 201 consecutive patients with or without chronic liver disease, evaluated in the same session by 2D-SWE and TE implemented on the following systems: Siemens ACUSON Sequoia (5C-1 convex transducer and Deep Abdominal Transducer-DAX), Aixplorer Mach 30 (C2-1X convex transducer) and FibroScan Compact M 530 (M and XL probes). Reliable measurements were defined as the median value of 10 measurements and an IQR/M<0.3. For significant fibrosis a cut-off value for TE of 7 kPa was used, for advanced fibrosis 9.5 kPa and for liver cirrhosis 12 kPa [1] Results From 201 patients, 198 patients had reliable measurements in all techniques and were included in the final analysis, mean age 54.8±13.3 years, mean BMI28.8 ± 5.0, 58% (116/198)men. 58.5% were without or with mild fibrosis, 14.1% had significant fibrosis, 6.2% had advanced fibrosis and 21.2% had liver cirrhosis. For significant fibrosis the performance was slightly better for 2D-SWE.SSI (AUROC=0.89, p<0.0001, >7.3 kPa, Se=85.1%, Sp=87.9%) followed by 2D-SWE.5C1 (AUROC=0.79, p<0.0001, >6.9 kPa, Se=33.7%, Sp=96.7%) and 2D-SWE.DAX (AUROC=0.78, p<0.0001, >6.3 kPa, Se= 36.4%, Sp=96.7%), p=0.01. For advanced fibrosis the best performance was slightly better by 2D-SWE.SSI (AUROC=0.92, p<0.0001, >8.8 kPa, Se=92.5%, Sp=91.9%), and by 2D-SWE.DAX (AUROC=0.86, p<0.0001, >7.6 kPa, Se= 38.8%, Sp=99.3%), followed by 2D-SWE.5C1 (AUROC=0.84, p<0.0001, >8.6 kPa, Se=38.8%, Sp=96.5%), p=0.02. For liver cirrhosis the performances were similar: 2D-SWE.SSI (AUROC=0.91, p<0.0001, >10.3 kPa, Se=92.8%, Sp=90.3%), followed by 2D-SWE.DAX (AUROC=0.90, p<0.0001, >10 kPa, Se= 23.8%, Sp=98.7%) and 2D-SWE.5C1 (AUROC=0.84, p<0.0001, >9.9 kPa, Se=33.3%, Sp=96.7%), p=0.10.The cut off values for predicting different stages of fibrosis ranged from 6.3-7.3 kPa for F2, 7.6-8.8 kPa for F3 and 9.9-10.3 for F4. Conclusions The performance of the evaluated 2D SWE techniques for liver fibrosis assessment was similar.