The purpose of this study is to check the intermuscular fluid collection in a phantom using ultrasound after comparison with the original injected volume of gel and to evaluate interobserver reliability.

One orthopedic surgeon injected ultrasonographic gel into the muscle of slaughtered cow meat using a syringe and needle, and recorded the injected volume. The injected volume of gel ranged from 1 ml to 5 ml, and total number of trials was 30 times. The volume was calculated using ultrasound by two experienced radiologists. The ultrasonographic measured volme was assessed and interobserver variability was measured.

The mean ultrasonographic measured volume was 71.35% of the original injected volume of gel. Thus, the measured volume using ultrasound showed a tendency to underestimate the injected volume into the muscle. For evaluation of interobserver reliability, interclass correlation using SPSS and consistency using Bland and Altman plot were calculated. Interclass correlation was 0.95, and Bland and Altman plot showed that the differences of mean volume measured by two observers were within ±1.96 standard deviation over 95%. Therefore, it showed good interoberver reliability.

Ultrasonographic measurement of the intramuscular fluid collection in a phantom showed mean 71.35% of the original injected volume of gel, and it tended to be underestimated in proportion to the increased volume of intermuscular gel. This study showed good interobserver reliability.