Conventional versus locked plates for hip dislocation in cerebral palsy: a matched cohort analysis.

PURPOSE: Bone quality of non-ambulatory patients with cerebral palsy (CP) is a matter of concern for proximal femoral varus derotational osteotomies (VDRO). Locking plates (LCP) have been designed to compensate this biological downfall. Little data exist comparing the LCP with the conventional femoral blade plate. METHODS: We retrospectively studied 32 patients submitted to VDRO (40 hips), operated with blade plates or LCP. Groups were matched, and the minimal follow-up was 36 months. Clinical (age at surgery, sex, GMFCS class, CP patterns) and radiological characteristics (neck shaft angle [NSA], acetabular index [AI], Reimers migration index [MP] and time until bone healing), as well as postoperative complications and the cost of treatment, were evaluated. RESULTS: Preoperative clinical characteristics and radiographic measurements were comparable, except for a higher AI in the BP group (p < 0.01). Mean follow-up was longer in the LCP group (57.35 vs 34.6 months). Mean NSA, AI and MP had comparable correction with surgery (p < 0.01). At final follow-up, dislocation recurrence speed was higher in BP group although not statistically significant (0.56% vs 0.35%/month; p = 0.29). The complication rate was similar in both groups (p > 0.05). Finally, the cost of the treatment was 62% higher in the LCP group (p = 0.01). CONCLUSION: Our cohorts showed LCP or BP equivalence clinically and radiographically in mid-term follow-up, with the former increasing the cost of treatment by a mean of 62%. This may raise a question on the real necessity of locked implants for these operations. LEVEL OF EVIDENCE: Level III-Retrospective comparative study.