The complete protocol of this cross-sectional study (registration no. ChiCTR1900021700) was approved by the academic board of the China Institute of Sports Science (no. 201518-21). All methods were performed in accordance with the guidelines and regulations of the institution. Before the study, written informed consent was obtained from all study participants and their parents. We excluded all students who refused to participate in the study. During the analysis, all participants were deidentified.
From September 2014 to September 2016, the middle schools in the Xicheng District of Beijing were divided into demonstration, ordinary, and general schools. Three schools at each level were selected by random sampling. At each school, three classes (junior 2, senior 1, and senior 2) were selected by simple random sampling. A cluster survey of the selected classes was conducted to test spine morphology and function.
Before the test, all students and parents were given questionnaires (the main contents of the questionnaires included questions about whether the student had a definite diagnosis of spinal diseases, whether there had been back or lumbar pain during the past month, whether there had been an acute back injury during the past month, whether there had been sports-related injuries to other body parts during the past month, and others).
During the pre-experiment, which included 34 boys and 28 girls, the mean thoracic kyphosis angle (TKA) values were 37.3° and 35.4° for boys and girls, respectively. The standard deviation was 10.2 for all participants. With a type I error rate of 0.05 and power of 80%, the study used the following formulas to compute the sample size:
As a result, each group required at least 450 participants because of the 10% rejection rate. Our design included at least 500 male and female participants.
Participants completed screening questionnaires and were examined by physical therapists. The exclusion criteria were as follows: history of spine fracture, spine-related surgery, shoulder joint motion injury, pelvis-related injury, or definitive diagnosis of spine-related diseases (such as cauda equina syndrome, lumbar disc herniation, spinal stenosis, congenital scoliosis, idiopathic scoliosis); spine pain during the past month; acute spine-related injuries during the past month; limb fracture, joint trauma, and other phenomena during the past month; structural kyphosis with obvious structural abnormalities; positive Adams flexion test result (to exclude possible scoliosis); and anomalous kyphosis with thoracic deformity.
Test methods to determine spinal morphology and function
The gold standard of spinal morphology testing is X-ray irradiation, but the radiation produced is contrary to the ethical requirements. Therefore, in this study, a non-invasive device called the Spinal Mouse, which has good reliability and validity, was used to test the spine morphology and function [18-20]. Several previous studies have used the Spinal Mouse to measure the spinal alignment and function of different types of participants [9, 21-23].
Participants were asked to remove their shoes and jackets to expose the entire back from the seventh cervical vertebra (C7) to the third sacral vertebra (S3). We located the iliospinale posterius, at the same level of the S2 spinous process; then, we located the S3 down to S2 (approximately 2 cm below) and around the anal crease [19, 24]. Surface localization of all spinous processes from C7 to S3 was marked with a fluorescent pen. The test protocol included the standing sagittal plane, maximum flexion, maximum extension, and Matthiass test. All body position tests determined all the spinous process markers on the body surface; then, tests were performed to assess the spine based on those markers. The specific testing methods for each body position were as follows: upright position, study participants were instructed to stand upright in a casual position with feet shoulder-width apart bearing equal weight, arms by their sides, and looking straight ahead; forward bending position, study participants were instructed to stand with their feet shoulder-width apart and to bend the torso forward while keeping the legs straight and allowing the arms to fall naturally; extended position, study participants were instructed to stand with their feet shoulder-width apart bearing equal weight, arms by their sides or supported by the hips, looking straight ahead, with the jaw close to the chest, and to stretch the torso backwards as far as possible; and the Matthiass test. During Matthiass test, study participants were instructed to stand with their feet shoulder-width apart, looking straight ahead. When asked, the participants pushed their arms forward horizontally while holding a certain amount of weight (5% of participant’s body weight) until their arms created a 90° angle with the shoulders. Spinal curvature and pelvic tilt were measured. Participants were instructed to retain this posture for 30 seconds; then, the second spinal curvature and pelvic tilt were measured. The Matthiass test is a useful method for detecting spinal posture changes related to strength levels [9, 25, 26].
The test indexes included the TKA, lumbar lordosis angle (LLA), sacral/hip angle (SA), and incline angle (INA) of the sagittal plane while the participants were in the standing and sitting positions. The TKA is the Cobb angle from T1 to T12, and the LLA is the Cobb angle from L1 to S1. The SA reflects pelvic positioning (the angle between the surface contour line and the vertical line of the sacrum) and pelvic mobility as a result of the limited movement of the sacroiliac (SI) joint. The incline line is the line between T1 and S1. The angle between the incline line and the vertical line is called the INA. The range of motion (ROM) of the thoracic spine and the ROM of the lumbar spine in the frontal plane were recorded. The ROM values of the sacrum, thoracic spine, and lumbar spine from the fully flexed position to the fully extended position in the sagittal plane, which can reflect the overall mobility of the spine, were also recorded. The Matthiass test was used to measure the changes in the TKA, LLA, SA, and INA upon loading.
Normal ranges for the TKA and LLA
The normal range of the TKA is 20°–40° [27, 28]. However, few studies have reported the normal range of the LLA for adolescents. Jean et al. reported that the mean normal LLA was 48.0°±11.7° ; therefore, in this study, the normal range for the LLA was based on that result (mean value ± 2 standard deviations).
We created groups based on sex and grade. Because of the differences in sexes and grades, there was great diversity in the lifestyles of the Chinese students involved in this study. Using SPSS 19.0 statistical software, an independent sample t test was performed using height, weight, and age to test whether there were significant differences among the groups. An independent samples t-test was used for all characteristics of those with the same sex and age. The likelihood ratio chi-square test was used to analyse whether there were differences in the incidence of abnormal thoracic kyphosis and lumbar lordosis among the different groups. A simple correlation analysis was performed for age and spinal morphology and function. The difference was statistically significant (P<0.05).
SAS 9.4 was used for data cleaning and analysis. The CANCORR process was used to analyse four indexes (sacral INA, TKA, lumbar lordosis angle, INA) reflecting spinal morphology and eight indexes (sacral activity, thoracic activity, lumbar activity, INA activity, sacral inclination load, thoracic kyphosis load, and lumbar lordosis load) reflecting spinal function.