Wei Chen, Lawrence H. Le, Edmond H. M. Lou


October 2016, Volume 25, Issue 10, pp 3265 - 3273 Original Article Read Full Article 10.1007/s00586-016-4492-6

First Online: 07 March 2016

Purpose

This study aimed to investigate the intra- and inter-observer reliability of the axial vertebral rotation (AVR) measurements of adolescent idiopathic scoliosis (AIS) using the center of lamina (COL) method on ultrasound transverse images.

Methods

Three cadaver vertebrae were scanned with 42 AVR configurations by both ultrasound and radiograph. In this in vitro study, four observers measured the AVR using the COL method on ultrasound transverse images and three observers measured the AVR using the Stokes’ method on radiographs. In the in vivo study, 13 AIS subjects were recruited. Eighteen spinal curvatures were identified and 48 vertebrae were selected for the AVR measurements. Two observers performed the AVR measurements on both the ultrasound images and radiographs. All measurements were performed twice with 1 week interval apart to reduce memory bias. The intraclass correlation coefficient (ICC), mean absolute differences (MAD), and standard deviation (SD) were used to analyze the intra- and inter-observer reliability of the AVR measurements. The Bland–Altman plot was used to analyze the 95 % limit of the differences between the two methods.

Results

The proposed COL method had high intra- and inter-observer reliability on both the in vitro and in vivo studies (ICCs > 0.91, MADs < 1.4°) and agreed well with the experimental setup (ICCs > 0.96, MADs < 2.3°). The COL method showed good agreement with the Stokes’ method for the in vitro study (ICC 0.84–0.85, MAD 4.5°–5.0°), while poor agreement for the in vivo study (ICC 0.49–0.54, MAD 2.7°–3.5°).

Conclusions

The pilot study indicated the proposed COL method was a simple and reliable method to evaluate the AVR on ultrasound images. Standardization of the posture during ultrasound scan and taking radiograph is important.


Read Full Article