RESUMEN
PURPOSE: A long debate exists about the connection between anorexia nervosa (AN) and scoliosis due to conflicting evidence. No study so far has evaluated the prevalence of scoliosis in patients with AN. The aim of the study is to evaluate the prevalence of idiopathic scoliosis in patients with AN. METHODS: Design: cross-sectional study. STUDY GROUP: convenience sample of all patients matching the inclusion criteria. CONTROL GROUP: female participants coming from an epidemiological screening for scoliosis. INCLUSION CRITERIA: patients had a diagnosis of AN during adolescence according to the DSM-IV-TR criteria. We applied a two-level screening using a Bunnell scoliometer and a radiograph. We calculated the odds ratio compared with participants coming from a school screening. RESULTS: Seventy-seven females with AN were compared to 816 females screened for scoliosis. The prevalence of scoliosis in the AN group was 16.9% (OR 5.77, 95% CI 3.12-10.67) with respect to the control group. If we consider as positive only those who received a scoliosis diagnosis during adolescence, the OR would be 3.15 (95% CI 1.55-6.42). DISCUSSION: This is the first study performed on patients with AN showing a sixfold greater odds of presenting with scoliosis. A cause-effect relationship cannot be determined due to the design.