Diane M Dickson1, Henrietta O Fawole1,2, Gordon J Hendry1, Stephanie L Smith1. 1. Glasgow Caledonian University, School of Health and Life Sciences, Glasgow, Scotland. 2. Physiotherapy Department, University of Benin, School of Basic Medical Sciences, Benin City, Nigeria.
Abstract
OBJECTIVES: To evaluate intermachine variation and compare intraoperator and interoperator agreement and repeatability characteristics of 2 ultrasound (US) systems for measurements of quadriceps and patellar tendons by strain elastography (SE). METHODS: Forty tendons from 20 healthy participants were investigated by operators with different experience (operator 1, 12 years of US experience and >50 SE examinations; operator 2, no US experience and 1 day of SE training). Repeated measures were performed on GE Healthcare (Waukesha, WI) and Esaote (Genoa, Italy) US systems. The percentage of agreement, Cohen κ, intraclass correlation coefficient, and correlation tests assessed agreement, repeatability, and associations of SE measures. A paired t test and Wilcoxon signed rank test assessed differences in SE measures. RESULTS: The study participants included 5 male and 15 female volunteers (mean [range] age, 29.3 [21-39] years). Better agreement and repeatability characteristics were observed for the patellar compared to the quadriceps tendon and the color score (CS) method over the elasticity ratio (ER). Intraoperator agreement was better for the experienced operator. Intraoperator repeatability was achieved in 55% of ER (intraclass correlation coefficient, 0.40-0.91; P < .05) and 77% to 85% (κ = -0.25-1) of CS measures. Interoperator repeatability was achieved in 35% (t/z, -2.93-7.94; P < .001-.048) of all ER measures. No significant differences in proximal (z, -0.13- -0.78) and distal patellar (z, -1.52-2.26; P > .5) patellar ER measures were observed. Seventy-four percent to 75% mean agreement (κ = 0-0.5) for CS measures comparable across both US systems was observed. Intermachine ER associations were poor (r = -0.39-0.13; P > .05), whereas greater than 70% agreement (κ = -0.87-0.53) for the CS was achieved. CONCLUSIONS: The reproducibility of knee tendon SE measurements is influenced by the operator experience, US system, and tendon site.
OBJECTIVES: To evaluate intermachine variation and compare intraoperator and interoperator agreement and repeatability characteristics of 2 ultrasound (US) systems for measurements of quadriceps and patellar tendons by strain elastography (SE). METHODS: Forty tendons from 20 healthy participants were investigated by operators with different experience (operator 1, 12 years of US experience and >50 SE examinations; operator 2, no US experience and 1 day of SE training). Repeated measures were performed on GE Healthcare (Waukesha, WI) and Esaote (Genoa, Italy) US systems. The percentage of agreement, Cohen κ, intraclass correlation coefficient, and correlation tests assessed agreement, repeatability, and associations of SE measures. A paired t test and Wilcoxon signed rank test assessed differences in SE measures. RESULTS: The study participants included 5 male and 15 female volunteers (mean [range] age, 29.3 [21-39] years). Better agreement and repeatability characteristics were observed for the patellar compared to the quadriceps tendon and the color score (CS) method over the elasticity ratio (ER). Intraoperator agreement was better for the experienced operator. Intraoperator repeatability was achieved in 55% of ER (intraclass correlation coefficient, 0.40-0.91; P < .05) and 77% to 85% (κ = -0.25-1) of CS measures. Interoperator repeatability was achieved in 35% (t/z, -2.93-7.94; P < .001-.048) of all ER measures. No significant differences in proximal (z, -0.13- -0.78) and distal patellar (z, -1.52-2.26; P > .5) patellar ER measures were observed. Seventy-four percent to 75% mean agreement (κ = 0-0.5) for CS measures comparable across both US systems was observed. Intermachine ER associations were poor (r = -0.39-0.13; P > .05), whereas greater than 70% agreement (κ = -0.87-0.53) for the CS was achieved. CONCLUSIONS: The reproducibility of knee tendon SE measurements is influenced by the operator experience, US system, and tendon site.