RESUMO
Fractures are common in childhood with incidence maximal during puberty, around the time of peak height velocity. The relationships between single nucleotide polymorphisms (SNPs) in COL1A1 and COL1A2, bone mass acquisition, and childhood fractures are unclear. We recruited 394 children and adolescents aged 4 to 16 years into a noninterventional case control study. All had suffered an episode of trauma leading to hospital presentation; 205 had sustained a fracture. We determined the frequency of COL1A1 Sp1 and COL1A2 PvuII SNPs. Lumbar spine dual-energy X-ray absorptiometry (DXA) measurements were compared between fracture and control groups according to genotype. Subgroup analyses were performed according to sex, pubertal status, and site of injury. We found that the COL1A2 'PP' genotype approximately halved the odds of fracture in the study group as a whole (OR=0.45 [95% CI=0.24-0.82], p=0.01). In particular, possession of the same genotype by subjects who had not yet progressed beyond midpuberty was associated with reduced odds of fracture (OR=0.38 [95% CI=0.19-0.79], p=0.01) and significantly increased lumbar spine bone mineral content (p=0.03) and areal bone mineral density (p=0.007). The COL1A1 Sp1 binding site 's' allele was associated with a trebling of the odds of fracture in prepubertal children (OR=3.1 [95% CI=1.43-6.61], p=0.004), but there was no association with any DXA measures. This is the first paediatric study to our knowledge that shows an association of the COL1A2 PvuII restriction site 'PP' genotype with a reduced risk of fracture and of the COL1A1 Sp1 binding site 's' allele with an increased risk. The association of these variants with fracture risk is greatest during periods of predominantly appendicular bone growth.
