Prevalence of adolescent idiopathic scoliosis in Shijiazhuang, Hebei, China: a cross-sectional study

Scoliosis refers to the lateral curvature of the human spine in the coronal plane, which is generally accompanied by vertebral rotation and alteration of physiological curvature in the sagittal plane, and even rotation and tilting of ribs and pelvis, as well as abnormalities of the surrounding soft tissues. As the etiologies of approximately 90% of scoliosis are unclear, the disease is also termed as idiopathic scoliosis (IS). Based on the age of the patients, IS is further classified into infantile (0–2 years), juvenile (3–9 years), and adolescent (≥ 10 years) IS (AIS), of which AIS is the most common type [1]. AIS has reportedly become a major issue, following myopia and mental health, severely affecting the healthy growth of adolescents in China [2]. The school screening of AIS was first performed in the USA in 1960s [3] and was performed in various regions in China in 1990s, based on which the prevalence was reported. However, the screening could not be performed in most regions or performed in a small scale and limited patients due to the lack of support from relevant governments. The findings of screenings varied due to the differences in the regions, age groups, sample sizes, methods, and awareness of this disease in different regions. AIS screening has been included since 2018 in the routine health service in XXX, XX, China, thanks to the support from the XX provincial government, the XX Provincial Department of Education, the XX Provincial Health Commission, and the XX Provincial Sports Bureau. Since the outbreak of coronavirus disease 2019, online learning increased while physical activities reduced substantially in primary and secondary school students in mainland China, which further increased the spinal burden on children that was capable of inducing scoliosis [4]. This study aimed to screen this condition so as to explore the prevalence and features of AIS in XXX, XX, and provide timely interventions and accurate guidance to children with mild scoliosis.

The findings of this study showed that the AIS prevalence was 2.5% in XXX, XX, which was higher than the average AIS prevalence (1.23%) reported by a previous survey in China [6]. One possible cause of the difference in the prevalence could be that previous surveys did not consider reexamination rate and imaging rate of positive cases, and thus, relatively high report bias was found. In this study, the inverse probability weight was calculated based on the propensity score modeling of the general characteristics of children to adjust the prevalence. The IS prevalence varied substantially in countries other than China, ranging from 0.62 to 3.26% [7‐12] (4.3% in Goiania, Brazil; 2.3% in Turkey; 3.1% in semiarid region of Brazil; 3.26% in Korea; 0.62% in Isfahan, Iran; 2.93% in Indonesia), which could be associated with the differences in regions, age of enrollment, methods of screening, and diagnostic criteria in the studies. The male-to-female prevalence ratio was 1:1.85, and the prevalence was higher in female than in male children, which was in agreement with the findings of previous studies [6]. The higher prevalence in female children could be associated with the early development in these children compared with their male counterparts, while their physical activities were less than male children, and their muscle strength was weak and incapable of supporting the trunk. Moreover, the higher prevalence in female children could also be associated with their dance training. Previous studies reported that the incidence of back and tendon injuries was relatively high in dancers with a younger age, and overstretching, especially back stretching, was not recommended in children under 10 years of age during dance training [13]. The IS prevalence in professional dancers was as high as 41.6% [14].

The findings of this study showed that the IS prevalence increased with age, which was in agreement with the results of a previous study that the IS prevalence was 0.1%, 0.3%, and 1.2% in children aged 6–8 years, 9–11 years, and 12–14 years, respectively [15]. The AIS prevalence was 1.19%, 1.04%, and 2.02% in children aged 7, 8, and 9 years, respectively, in this study. However, several other studies reported no IS in children aged 6–9 years and thus recommended no IS screening for children aged ≤ 10 years [16]. As the AIS prevalence in younger children was relatively low, the necessity of screening in this group still needed to be further evaluated. According to previous studies, the age of menarche in girls in China showed a long-term declining tendency, and is 11.8 ± 1 years currently [17]. The peri-menarche period is the peak stage of growth of girls, which is also the high-incidence stage of AIS. Therefore, IS screening should be performed in primary school as well, otherwise the best conservative treatment timing could be missed even if IS is screened.

The findings of this study showed that the percentage of mild IS (Cobb angle, 10°–19°) was 79.5% in XXX, XX, which was higher than the percentage reported in a western Istanbul study (42.8%) [18]. The findings of this study indicated that the AIS in XXX was mainly mild, while the essence of screening was to help early intervention on mild IS, prevent IS progression, and reduce the rate of surgery. The findings of this study also showed that the Cobb angle in children with AIS tended to worsen with age, which could be associated with the education mode in China. For instance, the pressure for learning was substantially higher in senior students than in junior students, and long-time sitting could also induce imbalance of trunk muscles, therefore increasing the risk of IS [19].

AIS could be further classified according to the number of spinal curves. In this study, most children with AIS were with single curve (75.4%), of which 46.4% were lateral thoracolumbar curve, consistent with the results reported by a Korean study (single curve accounted for 87.69%, and lateral thoracolumbar curve accounted for 53.6%) [10]. These findings indicated that the thoracolumbar segment of adolescents had poor stability and thus had a higher risk of IS.

Adam’s FBT is still the extensively acknowledged effective method for IS screening. Previous studies demonstrated that ATR correlated with the Cobb angle of IS (r = 0.7) [20], and ATR ≥ 5° could be used as the criterion for referral. However, this criterion had the risk of over-referral, and the false positive rate was 35.7–61.1% [7, 10, 21] (35.7% in Goiania, Brazil; 53.6% in Korea; and 61.1% in Chongming Island, China). However, when using ATR ≥ 7° as the criterion for referral, the false positive rate decreased, but still approximately 25% of positive cases had no IS, while the false negative rate increased [22]. The difference in the false positive rate could be associated with the age of patients receiving the screening. Previous studies demonstrated that ATR was not associated with the Cobb angle in children < 9 years, and the association decreased gradually with the increase in the curvature. Therefore, the best cutoff value of ATR for referral is still not decided [23]. Both ATR ≥ 5° and trunk asymmetry abnormality were included as criteria to identify more children suspected with AIS in this screening, which could be a cause of the substantial higher positive rate in the school screening in this study (9.16%) than the average prevalence of 4.4% reported in previous studies [6]. Second screening by superior doctors was performed for positive children identified during the school screening to reduce the unnecessary imaging examinations due to high false positive rate; the final diagnosis rate was 41.0% and the false positive rate was 59.0%.

This study had several limitations. First, the awareness of students and parents on this disease was insufficient, and thus, the rate of second screening and imaging examination was relatively low, which was 12.3% and 41.0%, respectively. The school–family–hospital communications were to be strengthened to gradually improve the recognition and attention of students, teachers, and parents on AIS. Second, some positive children might also have visited to other hospitals for reexamination. In the future, communications and cooperation with other hospitals in XXX should be strengthened, and clinical studies on large data from multiple platforms should be conducted for better and effective outcomes. The AIS prevalence was estimated based on the reexamination rate, imaging examination rate, and weights of children in different age groups and sexes, in order to exclude the interference of aforementioned factors on the data analysis of AIS in children in XXX, XX. Thirdly, a detailed examination of leg length discrepancy was not conducted for lateral bends less than 20, so it was not clear whether the LLD and the curve really have apical torsion, and need to be further clarified in the later stage.

The findings of this study showed that the AIS prevalence in primary and secondary school students in XXX, XX, was relatively high, indicating the need to pay more attention to AIS to achieve early screening, prevention, and treatment. Moreover, a number of images showed that AIS is often accompanied by straightening of cervical curvature, straightening of lumbar curvature, collapse of thoracic spine, and thoracic kyphosis. Previous studies showed that spinal abnormality in the sagittal plane was a risk factor for IS [24]. Therefore, we need to enhance the overall screening and data analysis of spines in adolescents in the future for an in-depth epidemiological investigation to improve prevention and treatment strategies in patients with IS.