Unstable intertrochanteric fractures may be treated by several types of implants, most frequently by dynamic sliding hip screw or some form of intramedullary implant. Intramedullary implants began to be used in cases with an expectation of further improvement of osteosynthesis stability. A need to determine the advantages of single implants for selected types of fractures in randomized trials was defined. In addition to biomechanical principles, bone quality is considered, together with increasing possibilities in recent years of further improving density measurements, especially qCT with respect to local specificity. A series of 86 patients (24 men, 62 women, average age 77,6 years) was operated on from September 6, 2005 to June 30, 2009 for unstable intertrochanteric fracture (31 A2.1, A2.2, A2.3), either by DHS of PFN osteosynthesis after randomization. A CT examination of both hip joints in a predefined manner was performed before surgery. Using special software the relative density of the central spherical part of the femoral head 2 and 3 centimetres in diameter was determined. After fracture healing, the dynamization of the neck screw of both implants and the reduction of vertical distance between the tip of the neck screw and subchondral bone of the femoral head were determined. In addition to evaluation of osteosynthesis stability and osteosyntheis failure, clinical parameters such as surgical time, blood loss and length of hospital stay were compared between the two groups of patients. Survival of patients was evaluated with respect to April 21, 2010. In the patient series, 4 failures of DHS osteosynthesis (cut out) and 2 failures of PFN osteosynthesis (cut out) were noted. Sliding of the DHS was on average 11,9 mm, and was significantly higher in comparison to dynamization of the PFN neck screw, which was 6,9 mm (p=0,005). When comparing the vertical distance between the tip of the neck screw and subchondral bone of the femoral head immediately after surgery and after fracture healing the average reduction of the vertical distance was 1,6 mm in DHS osteosynthesis and 0,8 mm in PFN osteosynthesis. The difference was statistically significant (p=0,025). PFN seems to provide a more stable fixation, based on the measurements. The number of failed DHS osteosyntheses is higher in comparison to the number of failed PFN osteosyntheses but the difference is not statistically significant. The influence of femoral head density on osteosynthesis failure could not be determined due to a low number of failed osteosyntheses in both patient groups. At the same time, after statistical analysis, influence of the relative femoral head density on vertical distance reduction between the screw tip and femoral head subchondral bone in healed fractures was not proven. Statistically, average length of surgical time, length of hospital stay, mean blood loss and survival did not differ significantly between the two patient groups.

Aims: In spite of approved methods of osteosynthesis of proximal femoral fractures using modern implants stabilisation still may fail especially in unstable osteoporotic fractures which is a cause of revision surgeries and unsatisfactory functional results. The goal of our study was to determine predisposing factors of failure of either DHS or PFN osteosynthesis with respect to the degree of osteoporosis. At the same time we evaluated clinical results one year after surgery and evaluated occurrence of further osteoporotic fractures.

Methods: Within the framework of a research plan (2005–2009) patients with low energy fractures of trochanteric area with qCT proven osteoporosis have been randomised. Unstable intertrochanteric fractures were operated by either DHS or PFN osteosynthesis after adequate reduction. During surgery one bone sample was taken from the femoral head prior to insertion of head screw located at the tip of the screw and the second sample was taken from iliac crest. Samples from the femoral head were examined by histomorphometry. Relationship between histomorphometry and migration of osteosynthetic material was evaluated. After surgery patients were examined in osteology department including DEXA and received appropriate treatment of osteoporosis. Orthopaedic follow up was performed 6 weeks, 3, 6 and 12 months after surgery when patients were evaluated by Harris hip score. Results were evaluated statistically.

Results: From September 1. 2005 to August 31. 2006 55 patients with unstable intertrochanteric fractures had been randomised. DHS was used in 26 patients and PFN in 29 patients. The average age of the patients was 75,6 years. Only patients who were able to sign informed consent were elegible for randomisation. The average qCT T-score was −3,2 and the qCT Z-score was −1,1. In addition to osteoporosis osteomalacia was proven histologically in one patient. Secondary osteoporosis was proven in 15 per cent of all patients. 49 patients were examined 1 year after operation. Failure of osteosynthesis was observed in four cases (7,3 per cent, 2x DHS and 1x PFN cut out phenomenon, 1 case of PFN head screw migration). Migrating PFN screw was removed. There were no other revision surgeries. The average qCT T-score in patients with failure of osteosynthesis was −4,3, Z-score −2,1. The average HHS one year after surgery was 67,3.

Conclusions: In patients with proven osteoporosis in spite of correct surgical technique risk of osteosynthesis failure is increased. Optimized surgical techniques and implants may still improve surgical results in patients with severe osteoporosis (qCT T-score lower than −4).