Diagnostic Value of Real-time Sonoelastography in Congenital Muscular Torticollis

Abstract

The purpose of this study was to evaluate the possible use of real-time sonoelastography in infants with congenital muscular torticollis for predicting treatment outcomes. The study included 20 infants with a sternocleidomastoid muscle thickness of greater than 10 mm, a sonoelastographic score of 4, and involvement of the entire length of the muscle (group 1) and 30 infants with a sternocleidomastoid muscle thickness of less than 10 mm, a sonoelastographic score of 3, and involvement of only part of the muscle (group 2). A physiatrist performed B-mode sonography and sonoelastography together, measured the thickness of the sternocleidomastoid muscle, and calculated the cross-sectional area of the involved muscle in both groups. On color scale sonoelastography, the sonoelastographic score of the sternocleidomastoid muscle was graded from 1 (purple to green: soft) to 4 (red: stiff), and the color histogram of the muscle was subsequently analyzed. The thickness and cross-sectional area of the sternocleidomastoid muscles in group 1 were significantly greater than those in group 2 (P = .001). On the color histograms, the median red pixel values in group 1 were significantly greater than those in group 2 (P = .001). In group 1, the mass in the affected muscle completely disappeared in 16 infants (80%), and a residual mass was detected in 4 (20%) on B-mode sonography at the final outcome. However, in group 2, the mass in the affected sternocleidomastoid muscle completely disappeared in all of the infants. These findings suggest that real-time sonoelastography, although an ancillary technique to conventional sonography, may predict treatment outcomes of congenital muscular torticollis.