Variability of Liver Shear Wave Measurements Using a New Ultrasound Elastographic Technique.

ABSTRACT

OBJECTIVES: A new 2-dimensional (2D) shear wave elastographic (SWE) device has been developed for the noninvasive assessment of liver fibrosis. Guidelines on measurement acquisition parameters are not yet well established for this technique. Our study aimed to assess 2D SWE measurement variability and to determine the number of measurements required per patient to reliably assess liver stiffness. METHODS: Two-dimensional SWE was assessed in 55 patients with mixed-etiology chronic liver disease on an Aplio 500 ultrasound system (Toshiba Medical Systems Corporation, Tochigi, Japan). Ten measurements were obtained per patient by an operator blinded to all preceding readings. Results were analyzed with clinical information obtained from medical records. RESULTS: The median interquartile range/median ratio for 2D SWE was 0.131 (quartiles 1-3, 0.089-0.174). Five readings provided an approximation within 0.11 m/s, or 4.2% of the median velocity of 10 measurements. Factors associated with increased measurement variability included body mass index (ρ = 0.388; P = .01), increased skin-to-liver capsule distance (ρ = 0.426; P = .002), and measurements taken within 1.5 cm of the liver capsule (P < .001). Measurements with heterogeneous shear wave profiles (indicated by a region of interest [ROI] SD/speed ratio > 0.15) showed greater deviation from the set's median velocity than those with an ROI SD/speed ratio of 0.15 or lower (0.42 versus 0.22 m/s; P = .001). CONCLUSIONS: Two-dimensional SWE showed low overall measurement variability, with a minimum of 5 readings providing equivalent precision to the existing method using 10 samples. Obesity, increasing abdominal wall thickness, subcapsular measurements and an ROI SD/speed ratio of greater than 0.15 were all associated with increased measurement variability. The ROI SD/speed ratio warrants further evaluation as a quality assessment metric, to allow objective operator assessment of individual 2D SWE measurement reliability in real time.