BACKGROUND - The purpose of this study was to compare hemiepiphysiodesis implants for late-onset tibia vara and to evaluate patient characteristics that may predict surgical failure.
METHODS - This is a retrospective review of late-onset tibia vara patients treated with temporary hemiepiphysiodesis from 1998 to 2012. Mechanical axis deviation (MAD), mechanical axis angle, mechanical lateral distal femoral angle, and medial proximal tibial angle were measured on standing bone length radiographs. Surgical failure was defined as residual deformity requiring osteotomy, revision surgery, or MAD exceeding 40 mm at the time of final follow-up. Implant failure was recorded. Costs included implants and disposables required for construct placement. Staple constructs included 2 or 3 staples. Plate constructs included the plate, screws, guide wires, and drill bits.
RESULTS - A total of 25 patients with 38 temporary lateral proximal tibia hemiepiphysiodeses met the inclusion criteria. The average body mass index (BMI) was 39.1 kg/m with an average follow-up of 3.0 years (minimum 1 y). Surgical failure occurred in 57.9% of patients. Greater BMI (P=0.05) and more severe deformity (MAD, mechanical axis angle, and medial proximal tibial angle; P<0.01) predicted higher rates of surgical failure. Younger age predicted higher rates of implant failure (P<0.01). There were no differences in surgical or implant failure between staple and plate systems. Hospital costs of plate constructs ($781 to $1244) were 1.5 to 3.5 times greater than the staple constructs ($332 to $498).
CONCLUSIONS - Greater BMI, more severe deformity, and younger age were predictive of surgical or implant failure. There was no difference in success between implant types, whereas the cost of plate constructs was 1.5 to 3.5 times greater than staples. The rate of surgical failure was high (58%) and consideration should be given to reserving hemiepiphysiodesis for patients with lower BMI and less severe deformity. In our population, if hemiepiphysiodesis was not offered to patients with BMI>35 or MAD>80 mm varus, the surgical failure rate would diminish to 28%. The failure rate outside these parameters would be 88%.
LEVEL OF EVIDENCE - Level II—Prognostic.