Reducción quirúrgica de la cadera congénita displásica luxada con y sin ligamentopexia. Comparación de tres métodos de inmovilización postoperatoria

Abstract

Objective. To assess efficacy and security among 3 surgical procedures for congenital hip dislocation. Materials and methods. This is a prospective trial about a historical cohort of 193 children, aged 10 to 35 months, 205 hips classified into 3 groups: A) 97 hips with simple reduction and 6 months of casting in 2 positions (3 months human position and 3 months Lorenz position) B) 53 hips with attachment of ligamentum teres to the acetabulum and 3 months casting in the Lorenz position, and C) 55 hips with ligament attachment and 3 months casting in human position. Evaluation was through the accumulated rate of redislocation and necrosis, as well as by severity degree of necrosis. Hypothesis were matched by Chi2 test and treatments were peered compared for relative risk with confidence intervals of 95%. Tolerance for alfa-error was fixed in 5%. Since the size of samples was not calculated at the begining, statistical power was assessed post hoc for each contrast. Results. Rate of redislocation was: A) 72%, B) 7.2% and C) 1.8% (p=0.006). Peering comparison of treatments, technique A had a significant protective effect in relationship to B, but not to C, which had a lower rate than A. However, technique B had a relative risk for redislocation 10 times higher than C (p=0.003) and 3 times higher than A (p=0.03). Accumulated rates for necrosis were for A) 33.0%, B) 30.2% and C) 20.0% (p=0.22). Peer treatments comparison for global rate of necrosis, C was near to be significantly lower than A (p=0.08) but not than B (p=0.22). When comparing distribution of severity degrees of necrosis, A had a slight advantage over B and C, since A had the highest rate of degree-I severity of necrosis and the lowest rate of degrees II-IV (A vs C, (p=0.27). Conclusion. Treatment C had a better security and efficacy 16 in a relatively low redislocation rate, while A had a tendency to a higher efficacy in decreasing severity of necrosis. Finally, most peer comparison of treatments for frequency and severity of necrosis lack of statistical power, with a top risk of beta-error. This will improve by increasing the universe of group C