Static proximal fixation of intramedullary nails used to treat inter-trochanteric femur fractures does not increase the risk of failure. Radiographic outcomes from a randomized controlled trial.

Abstract

To compare the outcomes of intramedullary fixation of intertrochanteric femur fractures treated with a single lag screw (Gamma3) and a dual integrated screw design (Intertan), including outcomes depending on the mode of proximal lag screw fixation (static or dynamic). Design. A pragmatic, single-blinded RCT with a three-arm parallel group design. A multicentre PRCT, with a Level 1 academic trauma centre and a second linked smaller level 2 hospital. Patients aged over 60 undergoing intramedullary screw fixation of a standard obliquity intertrochanteric femur fracture (AO/OTA 31A1 or A2) were randomized into three groups: single lag screw (dynamically locked); dual integrated lag screw (dynamically locked) and dual integrated lag screw (statically locked). The primary outcome measure was radiographic failure of the device by 6 months, judged by any one of cut-out requiring re-operation, a change in tip-apex distance of more than 10mm, or breakage of the metal. Pairwise comparisons were performed between the 3 study groups. Secondary outcomes included all cause re-operation rates, and degree of secondary collapse. 477 patients were randomised. 27 patients were excluded after randomisation and 95 died prior to 6 months. 226 had full follow up to the primary outcome point of 6 months: (80 Gamma, mean age 83 (range 60 -101), 60% female; 72 Intertan dynamic, mean age 80 (range 60 - 101), 63% female; 74 Intertan static, mean age 82 (range 61 - 97), 72% female). A further 129 had clinical follow-up but no radiographs. No difference was seen in radiographic failure by 6 months between the Gamma nail (single lag screw) and the Intertan Dynamic (dual lag screw) groups (11.3% vs 9.7%, p=0.74); Initial tip-apex distance remained statistically the most significant independent predictor of failure (Mean TAD of 15.7mm in non-failure group, 23mm in failure group, p<0.001 ). The Intertan group with a statically locked proximal lag screw had a lower (non-statistically significant) radiological failure rate (1.4%) than either dynamically locked group (Gamma 11.3%, Intertan dynamic 9.7%, p=0.05). Re-operation rates were similar for all groups (Intertan static 2%, Gamma 3.3%, Intertan dynamic 5.3%, p=0.42). In patients over 60 undergoing intramedullary fixation of standard obliquity intertrochanteric fractures, the failure rate was not higher when using the Intertan nail in the proximally locked mode, when compared with either the Intertan nail or Gamma nail used in the dynamic proximal locking mode. Therapeutic Level I.