Purpose: To assess the reproducibility of ultrasound shear wave elastography (SWE) measurements of the liver made in a clinical environment.
Methods: 40 patients undergoing clinically-indicated abdominal ultrasonography were recruited under an IRB-approved protocol. Each patient had liver SWE measurements performed using a GE LE9 scanner by 2 sonographers randomly selected from a pool of 15 trained in the technique, in a blinded fashion. SWE screening categories were defined using Youngâ€™s modulus cut-off values of 8.3KPa and 11.9KPa. SWE acquisition success rate, reproducibility of screening category, and reproducibility of modulus measurements were all assessed.
Results: 74% of the 80 SWE acquisitions resulted in acceptable modulus measurements with respect to measurement consistency and image artifacts. 23 of 40 patients had 2 acceptable measurements and contributed to the reproducibility assessments. Measurements for these patients are shown in Figure 1. The number of these measurements falling in the <8.3KPa, intermediate, and >11.9KPa categories were 26, 12, and 8, respectively. SWE screening category agreed for 19 of the 23 patients (83%). In two of the patients, modulus values straddled the cut-off values but were within 5% of their mean. In one of these patients, the modulus difference was assessed as an outlier (see Figure 2). The remaining data were used to estimate the standard deviation (SD) of Youngâ€™s modulus measurements as 0.96KPa.
Conclusion: The SWE acquisition success rate was 74% (but has been improved via refinement of the measurement protocol and preset). Screening category agreement of 83% was deemed acceptable for effective clinical use. The Youngâ€™s modulus SD of 0.96KPa measured here under blinded conditions is likely a worst case estimate for performing follow-up measurements since, in clinical practice, the initial exam would be used for guidance when performing a follow-up exam. Further work to assess and reduce the SWE measurement SD is planned.
Funding Support, Disclosures, and Conflict of Interest: This work was funded in part by a grant from General Electric Medical Systems. Dr Chen and Mayo have been granted a patent related to this technology, that is licensed to General Electric Medical Systems.
Not Applicable / None Entered.