Implant-related fractures in children: A 15-year review

BACKGROUND: To our knowledge, there are no comprehensive clinical studies of implant-related fractures in children. Our goal was to identify the incidence, skeletal location, and associated diagnoses of implant-related fractures. METHODS: We reviewed our institutional database to identify cases of implant insertion (7584 cases) in patients less than 18 years old from January 1, 1995 through December 31, 2009. We calculated the overall incidence of these fractures and stratified the incidence by skeletal location and preoperative diagnoses. Fisher exact test was used to ascertain differences in fracture incidence. Risk ratios were calculated when appropriate. Significance was set at P<0.05. RESULTS: There were 25 cases of implant-related fractures: 22 in the femur, 2 in the tibia, and 1 in the radius. The overall incidence of implant-related fracture was 0.33%; the incidence by skeletal location was: femur, 0.89%; tibia, 0.1%; and radius, 0.14%. Associated diagnoses were cerebral palsy (9 cases), hip dysplasia (3 cases), spina bifida (2 cases), and avascular necrosis (1 case); 10 cases were associated with "other diagnoses," which included various skeletal syndromes (5 cases) and traumatic fractures (5 cases). The incidences of implant-related fractures by diagnoses were: cerebral palsy, 1.1%; hip dysplasia, 1.1%; spina bifida, 1.3%; and avascular necrosis, 0.35%. The incidence of implant-related fracture in the "other diagnoses" group was 0.16%, and the incidence of fracture in otherwise healthy patients was 0.084%. The femur was 15.2 times more likely to fracture than other bones (P<0.001). Diagnoses of cerebral palsy, hip dysplasia, spina bifida, and avascular necrosis were 6.1 times more likely to be associated with implant-related fractures than the "other diagnoses" (P<0.001). The mean time to fracture in the study was 2.8 years. The overall implant removal rate at our institution was 24.3%, and it varied significantly by patient diagnosis (P<0.01). CONCLUSIONS: Skeletal location and preoperative diagnosis should be factors of consideration in a surgeon's decision about removing implants to prevent implant-related fractures. LEVEL OF EVIDENCE: Prognostic Level III.