Scoliosis is defined as a lateral curvature of the spine greater than 10 degrees[1]. It most commonly starts between aged 10 years and skeletal maturity[2] (adolescent idiopathic socliosis, AIS), and has an incidence ranging between 1-3% depending on populations studied[3, 4], although this incidence does appear to be increasing[5]. There is an equal gender prevalence for minor scoliotic curves (10 degrees), but far more girls have curves greater than 30 degrees compared with boys[6]. For unknown reasons, 80-90 percent of thoracic and thoracolumbar curves are to the right[7] suggested to be linked with handedness[8], but 70% of single lumbar curves are to the left[7]. Again, for reasons that are not clearly understood, some scoliotic curves progress: estimates vary from 3.5%[9] to 15%[10], while others remain static.

Scoliosis is not always a benign structural abnormality, although the mortality rate for individuals with AIS is comparable to that of the general population[11]. There is a direct correlation between the magnitude of the thoracic scoliotic curve and pulmonary function[12]; back pain is a complaint in 80% of scoliotic patients in long-term follow-up studies[13]; and the psychological and social effects of scoliosis are real but variable and hard to quantify. Nevertheless, studies that have looked into this area consistently find mild social dysfunction such as an unwillingness to wear snug clothing or bathing suits[12], but also high rates of work disability[14,15] and reduced rate of marriage[15], although no studies that we could find assess the psychological and social effects in children and young adults.

Current knowledge about the causes of the initiation or induction of the scoliotic curve is scarce. Some work has been carried out on the determinants of the initiation or induction of the scoliotic curve, but very little using a longitudinal prospective study design. Some work has also been done on the determinants of curve progression, but contradictory results have been found for the association with bone density, anthropometry, and hypermobility, probably because of small sample sizes. No coherent picture on the genetic determinants of either induction or curve progression has been found, probably because scoliosis is due to a complex interplay between environmental and genetic influences. This also suggests that scoliosis may be a heterogeneous group: more accurate phenotyping according not only to curve structure, but associated clinical features, may therefore allow more accurate classification into subgroups with different prognosis and different genetic determinants.

Therefore the aims of this project are

1. to quantify the incidence of scoliosis in the UK population from aged 7 to 17 years, and to describe the percentage progression and regression (i.e. natural history), using a large population based birth cohort.

2. to assess the association between the standard Adams forward bending test with Cobb angle measured by DXA analysis in children

3. to assess the impact of scoliosis induction and progression in children (in terms of time off school, pain, sleep, physical activity, fractures)

4. to accurately phenotype and therefore classify into relevant subgroups, children with scoliosis in terms of curve structure and associated clinical features such as hypermobility, bone density, muscle function, family history

5. to prospectively investigate the risk factors for scoliotic curve induction or initiation, focussing on early life events, bone and muscle quality and growth

6. to prospectively investigate the risk factors for scoliotic curve progression or regression, focussing on growth, anthropometry, puberty, bone density and hypermobility

7. to investigate the genetic determinants of scoliosis using a large population based birth cohort to carry out genome wide association studies, with replication and further study to be carried out in a secondary care disease cohort

Methods

The Adams forward bending test was performed at research clinics when the offspring were aged 7, 9, 10, 11 and 13 by trained research assistants. We are requesting funding for this measure to be repeated at aged 17. The results of the Adams forward bending test will be used as the main outcome of interest, with children being categorised into having or not having scoliosis on the basis of an ATI greater or less than 7 degrees. Analyses will be repeated using the ATI as a continuous outcome to assess trends. In addition, total body DXA scans have been performed with full spinal DXA results. Cobb angles (traditional method of assessing scoliosis on plain X-rays) will be measured on these scans.

Exposures of interest

(1) demographics and early life events, (2) growth, (3) bone and muscle parameters, (4) genetics

Concept: Specific measure: Person: Source: Time point:

demographics gender child Q birth

ethnicity child Q birth

SES mother Q preg, birth, aged 4

early life event gestational age child Q birth

birth weight and length child assessment birth

growth anthropometry child assessment 7, 9,10,11,13,15,17

puberty child Q 9, 11, 13, 15

R/L handedness child Q

arachnodactyly child assessment 12

bone and muscle parameters total body DXA child assessment 9,12,15

grip strength child assessment 13

hypermobility child assessment 13,17

genetics GWAS data child biological

Impact of scoliosis

(1) chronic pain, (2) sleep, (3) physical activity, (4) time off school, (5) psychological, (6) lung function (7) fractures

Concept: Specific measure: Person: Source: Time point:

chronic pain chronic pain child Q 17

sleep disrupted/poor sleep child Q 13,14,16

physical activity reported child Q 9

accelerometer child assessment 12, 17

time off school time off school child Q 14,16

psychological DAWBA child Q 7,9,10,11,13,15

CIS-R child Q 7,9,10,11,13,15

lung function methacholine challenge child assessment 8

spirometry child assessment 9

NO, salbutamol challenge child assessment 15

fractures fractures child Q 12, 15

[1] Kane W (1997) Clin Orthop 126:43-46

[2] Dobbs MB et al (1999) Orthop Clin N Am 30:331-341

[3] Wang Y et al (1996) Chinese J Epi 17:160-162

[4] Grivas et al (2002) Studies Health Tech Info 91:76-80

[5] Wong HK et al (2005) Spine 30:1188-1196

[6] Roach JW (1999) Orthop Clin N Am 30:353-365

[7] Rinsky LA et al (1988) West J Med 148:182-191

[8] Burwell RG (2003) Develop Neurorehab 6:137-170

[9] Robitaille YVON et al (1984) Int J Epi 13:319-323

[10] Dickson RA et al (1980) BMJ 281:165-167

[11] Weinstein SL et al (1983) J Bone Joint Surg Am 65:447

[12] Aaro S et al (1984) Spine 9:220

[13] Weinstein SL et al (1981) J Bone Joint Surg Am 63:702

[14] Fowles JV et al (1978) Clin Orthop 134:212-217

[15] Nachemson A (1968) Acta orthop Scand 39:466-476