Please use this identifier to cite or link to this item:
Title: Three-dimensional assessments of adolescent idiopathic scoliosis using three-dimensional ultrasound
Authors: Wang, Qian
Degree: Ph.D.
Issue Date: 2018
Abstract: Adolescent idiopathic scoliosis (AIS) is a three-dimensional spinal deformity characterized by lateral curvature and vertebral rotation of spine. It occurs in approximately 3% of adolescents with unknown reasons. Nowadays, the radiographic assessment of scoliotic spine continues to be the most widely used method in a scoliosis clinic. In routine clinical practice, radiographic assessments are performed throughout the course of treatment of the patients with AIS. However, the frequency of radiation exposure in monitoring scoliosis concerns many adolescents and their parents in light of evidence that cumulative radiation exposure could increase cancer risk. In addition, radiographic assessment of scoliotic spine is limited in the coronal and sagittal planes, which represent a simplification of the true 3-dimensional (3-D) spinal deformity involved in scoliosis. Thus, attempts to reduce or eliminate radiation exposure in adolescents and visualization of 3-D characteristics of scoliotic spine have led researchers to develop new imaging technologies, such as stereo-radiography (EOS), ultrasound imaging, and magnetic resonance imaging (MRI). Currently, ultrasound has gained considerable attention in the assessment of scoliosis. Ultrasound imaging is a non-radiation and cost-effective method, which is accessible in the majority of medical institutes. The posterior structure of vertebrae could be displayed by ultrasound imaging in the transverse plane. The development of the 3-D ultrasound system can enable the 3-D reconstruction of vertebral images and facilitate the measurement of scoliotic spine in various anatomical planes that could not be accomplished previously. A series of the related research have been conducted in Canada, Hong Kong, Japan, Australia, Netherlands and other places. Spinous processes, laminae and transverse processes can be visualized and used as landmarks to measure the lateral curvature and vertebral rotation in the coronal and transverse planes of the ultrasound images. The center of laminae (COL) method has been proposed to measure the spinal curvature and vertebral rotation in the coronal and transverse planes of the 3-D ultrasound images. The reliability and validity of this proposed method have been demonstrated. However, the evidence is limited to the experiment phantom studies. Thus, the objective of this study was to explore the possibility of using the proposed 3-D ultrasound methods to assess the coronal curvature, vertebral rotation, kyphotic and lordotic angles in the subjects with AIS under the clinical setting, and to evaluate its reliability and validity with the concurrent MRI methods. Due to the gravitational effect, spinal orientation between standing and supine positions may change their corresponding lateral curvatures and vertebral rotations. Thus, the second purpose of this study was to investigate the gravitational effect on the coronal curvature and vertebral rotation between standing and supine postures using 3-D radiation-free ultrasound assessments in the patients with AIS. Sixteen female AIS subjects were recruited from the Prince of Wales Hospital, Hong Kong. The ultrasound examinations were performed using a 3-D ultrasound unit with a SonixGPS system. A purpose-design couch with central slot was used for supine ultrasound scanning. Ultrasound scanning was performed continuously along the coronal plane from C7 to S1, with the subjects in the standing and supine positions, respectively. Two observers performed the ultrasound scanning and 3 times of angle measurements for each parameter after 3-D image reconstructions. In the coronal plane, the spinal curvatures were measured using the center of laminae (COL) method. In the transverse plane, the apical vertebral rotations were also assessed with the center of laminae (COL) method. In the sagittal plane, the kyphosis and lordosis angles were estimated by the spinous process angle (SPA) method.
To compare with the ultrasound measurements, magnetic resonance imaging (MRI) examination was conducted within the same morning. A 3.0T MR scanner (Achieva, Philips Medical Systems, Netherlands) and a spine array coil were used. The spinal curvature was measured with the Cobb method in the coronal plane. The apical vertebral rotation was calculated using the Aaro-Dahlborn method in the transverse plane. The kyphotic and lordotic angles were assessed by the Cobb method in the sagittal plane. The raters 1 and 2 had 5- year and 2-year experience of using ultrasound to measure the scoliotic spine respectively. Prior to the study, each rater was required to practice ultrasound scanning at the supine position and measurements for more than 10 subjects. During this study, the 3-D ultrasound and MRI images were randomly assigned without specific order for measurements. The two raters were blinded to the subjects' clinical information and they performed the 3-D ultrasound and MRI measurements independently in 3 trials each with one week interval. The intra-class correlation coefficient (ICC, [2, k]) with 95% confidence intervals (CI) was used to evaluate the intra- and inter-observer reliabilities of the 3-D ultrasound and MRI assessments. In addition, the mean absolute deviation (MAD), standard deviation (SD) and standard error of measurement (SEM) were used to assess the intra- and inter-observer measurement variability of these two methods. In order to determine the validity of 3-D ultrasound assessments, the comparison of means, the Bland-Altman method and the Pearson correlation analysis were applied between the 3-D ultrasound and MRI measurements in the patients with AIS. The results suggested that the 3-D ultrasound presented high intra- and inter-rater reliability when measuring the coronal curvature (COL method), the apical vertebral rotation (COL method), the kyphosis and lordosis (SPA method) in the patients with AIS. In addition, the validity of 3-D ultrasound measurements has been verified, including the spinal curvature angle in the coronal plane (COL method); the vertebral rotation in the transverses plane (COL method); the kyphotic other than lordotic angles in the sagittal plane (SPA method). Besides, the difference between the supine and standing positions was 1.9°~11.7° and 0.0°~5.9° for the coronal curvature and the vertebral rotation respectively. Multi-linear regression revealed that the possible relevant factors were the coronal curvature, the vertebral rotation and the variation in the selected upper-end vertebra to these changes. Furthermore, a high correlation between the supine and standing postures was demonstrated. The radiation-free 3-D ultrasound presented to be a reliable & valid method for measuring the spinal curvature in the coronal plane and the vertebral rotation in the transverse plane in the patients with AIS under the clinical setting. The reliable assessments of kyphotic and lordotic angles in the sagittal plane have been obtained using the 3-D ultrasound method (SPA), however, its validity has yet been proved in this study. The possibility of using 3-D ultrasound to measure the lateral curvature and the vertebral rotation of AIS at the supine and standing positions were verified, the difference and correlation between these two positions have been demonstrated. Further studies on the 3-D changes of AIS using the radiation-free ultrasound are deserved in order to optimize the 3-D ultrasound scanning and measuring procedures, and to further validate the 3-D ultrasound measurements in a larger clinical trial. With these efforts, 3-D ultrasound will become a potential option used as an alternative to radiography for screening and routine assessment of scoliosis.
Subjects: Hong Kong Polytechnic University -- Dissertations
Musculoskeletal system -- Ultrasonic imaging
Pages: 216 pages : color illustrations
Appears in Collections:Thesis

Show full item record

Page views

Citations as of May 22, 2022

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.