Introduction

Adolescent idiopathic scoliosis (AIS) is the most common paediatric spinal deformity, affecting about 3% of school-aged children worldwide. This disorder occurs in otherwise healthy children who bear no obvious deficiencies in the components of the spinal column itself. The cause of AIS is poorly understood, as is implied by the name. Lesions of the bony composition of the vertebrae, the vertebral endplates, the paraspinous muscles, or the neurological system each have been proposed to explain disease pathogenesis. Progress has been hampered by the absence of an obvious AIS animal model. Consequently we have used genetic studies in human populations to identify factors underlying AIS susceptibility.

The complex inheritance and population frequency of AIS suggest that many genetic factors are involved in this disease. To search comprehensively for such factors we previously undertook the first genome-wide association study (GWAS) of AIS susceptibility in a cohort of 419 families in Texas, USA. We found that chromosome 3 SNPs in the proximity of the CHL1 gene yielded strongest results, which we replicated in additional cohorts (rs10510181 OR 1·49, 95% CI 1·29–173, p=2·58×10–8). CHL1 is of interest because it encodes an axon guidance protein and is functionally related to the ROBO3 gene that causes hereditary gaze palsy with progressive scoliosis (HGPPS), a rare disease marked by severe scoliosis. Here we expanded the study to 702 Texas families.

Background and methods: Adolescent idiopathic scoliosis (AIS) is the most common spinal deformity in children, with a prevalence of 1–2%. The disease generally displays complex inheritance. Various family studies have produced many first reports of AIS susceptibility regions, but confirmation of these is lacking. In the present study we investigated extension of our own data, and reproducibility of other published results, by testing linkage in a new collection of fifty-four AIS families. Altogether fifteen candidate regions were evaluated in a two-stage design.

Results: Strongest results were obtained for linkage to microsatellite loci within a candidate region of proximal 8q previously identified by chromosomal breakpoint mapping. Although positive lod scores were obtained for other regions, none exhibited significance less than or equal to P = .05. Lod scores remained stable after analysis of an independent panel of SNP loci in the 8q candidate region and were strengthened with inclusion of additional affected family members (multipoint NPL = 3.02, P = 0.001). Two SNPs near the peak of linkage produced evidence of association to AIS susceptibility. Both SNPs are found within plausible candidate genes for AIS susceptibility.

Conclusion: These results support linkage of the 8q11-8q13 region to AIS susceptibility. Bashiardes et al. previously described a chromosomal break in the 8q11 region that disrupted the gamma-1- syntrophin (SNTG1) gene and segregated with AIS in an extended kindred. In that study, possible rare splice site mutations were identified an additional affected family and one sporadic case. The peak of linkage and association detected in this study appears to be distinct from the SNTG1 gene. This suggests the possibility that more than one gene in the region may contribute to disease. A more detailed analysis of the region encompassing this linkage peak, and the SNTG1 gene, is warranted in larger family collections.

Of 193 children with Perthes' disease at the Texas Scottish Rite Hospital for Cripppled Children, 24 were found to have only minimal changes. The cases were grouped relative to the area of the femoral head involved, with 10 involving the anterior portion, seven the posteromedial, three the lateral, and four the central portion. Almost all hips lost some height as measured by the epiphysial index but all had good results by the Mose criteria irrespective of treatment. The anterior lesion is the same as that described by Catterall as Group 1. The three additional groups have a similar benign natural history but distinct radiographic features. The need for early recognition of these patterns is emphasised if unnecessary treatment is to be avoided. Possible correlations of these lesions with the segmental blood supply of the femoral head are proposed and an hypothesis relating the Catterall classifications to the blood supply is put forward.

Experience in the management of fourteen children with melorheostosis has been reviewed. The principal and presenting clinical features were unilateral soft-tissue contractures associated with inequality of limb length. In contrast to the disease in adults, pain occurred infrequently and was never intense. The average interval between the discovery of the clinical features and the correct diagnosis was six years. The distinctive radiographic feature in the child was an endosteal pattern of hyperostosis marked by streakiness of the long bones and spotting of the small. This differs from the usual subperiosteal or extracortical pattern of hyperostosis seen in adults. The surgical treatment of the contractures proved difficult and recurrence of the deformity was the rule. Distal ischaemia occurred when the chronically contracted and flexed joint was rapidly extended.