Introduction Bone strength plays an important role in implant anchorage. Bone mineral density (BMD) is used as surrogate parameter to quantify bone strength and to predict implant anchorage. BMD can be measured by means of quantitative computer tomography (QCT) or dual energy X-ray absorptiometry (DXA). These noninvasive methods for BMD measurement are not available pre-or intra-operatively. Instead, the surgeon could determine bone strength by direct mechanical measurement. We have evaluated mechanical torque measurement for (A) its capability to quantify local bone strength and (B) its predictive value towards load at implant cut-out. Materials and methods Our experimental study was performed using sixteen paired human cadaver proximal femurs. BMD was determined for all specimens by QCT. The torque to breakaway of the cancellous bone structure (peak torque) was measured by means of a mechanical probe at the exact position of subsequent DHS ® placement. The Wxation strength of the DHS ® achieved was assessed by cyclic loading in a stepwise protocol beginning with 1,500 N increasing 500 N every 5,000 cycles until 4,000 N. Results A highly signiWcant correlation of peak torque with BMD (QCT) was found (r = 0.902, r 2 = 0.814, P < 0.001). Peak torque correlated highly signiWcant with the load at implant cut-out (r = 0.795, P < 0.001). All specimens with a measured peak torque below 6.79 Nm failed at the Wrst load level of 1,500 N. The specimens with a peak torque above 8.63 Nm survived until the last load level of 4,000 N. Conclusion Mechanical peak torque measurement is able to quantify bone strength. In an experimental setup, peak torque identiWes those specimens that are likely to fail at low load. In clinical routine, implant migration and cut-out depend on several parameters, which are diYcult to control, such as fracture type, fracture reduction achieved, and implant position. The predictive value of peak torque towards cut-out in a clinical set-up therefore has to be carefully validated.

Osteoporotic changes start in cancellous bone due to the underlying pathophysiology. Consequently, the metaphyses are at a higher risk of "osteoporotic" fracture than the diaphysis. Furthermore, implant purchase to fix these fractures is also affected by the poor bone quality. In general, researchers and developers have worked on three different approaches to address the problem of fractures to osteoporotic bone: adapted anchoring techniques, improved load distribution as well as transfer with angular stable screws, and augmentation techniques using bone substitutes. A patent-based review was performed to evaluate which ideas were utilized to improve fixation in osteoporotic, metaphyseal bone, especially in the proximal femur, and to analyze whether the concept had entered clinical use. Anchoring devices that are either extramedullary or intramedullary have a long clinical history. However, demanding surgical techniques and complications, especially in poor quality bone, are justification that such implants and their corresponding surgical techniques need to be improved upon. Expanding elements have been evaluated in the laboratory. The results are promising and the potential of this approach has yet to be fully exploited in the clinics. Internal fixators with angular stable screws open the door for many new anchorage ideas and have great potential for further optimization of load distribution and transfer. Augmentation techniques may improve anchorage in osteoporotic bone. However, the properties of bone substitute materials will need to be modified and improved upon in order to meet the demanding requirements. If we summarise the development process and the clinical use of implants to date, we have to clearly state that more factors than simply biomechanical advantage will determine the clinical success of a new fixation principle or a new implant. Instead, fracture treatment of patients with osteoporosis really needs an interdisciplinary approach!

The importance of osteoporosis in proximal humerus fractures is well recognized. However, the local distribution of bone quality in the humeral head may also have a significant effect because it remains unclear in what quality of bone screws of standard implants purchase. The goal of this study was to investigate whether the failure of proximal humerus locked plating can be predicted by the DensiProbe (ARI, Davos, Switzerland). A 2-part fracture with metaphyseal impaction was simulated in 12 fresh-frozen human cadaveric humeri. Using the DensiProbe, local bone quality was determined in the humeral head in the course of 6 proximal screws of a standard locking plate (Philos; Synthes GmbH, Solothurn, Switzerland). Cyclic mechanical testing with increasing axial loading until failure was performed. Bone mineral density (BMD) significantly correlated with cycles until failure. Head migration significantly increased between 1000 and 2000 loading cycles and significantly correlated with BMD after 3000 cycles. DensiProbe peak torque in all screw positions and their respective mean torque correlated significantly with the BMD values. In 3 positions, the peak torque significantly correlated with cycles to failure; here BMD significantly influenced mechanical stability. The validity of the DensiProbe was proven by the correlation between its peak torque measurements and BMD. The correlation between the peak torque and cycles to failure revealed the potential of the DensiProbe to predict the failure of locked plating in vitro. This method provides information about local bone quality, potentially making it suitable for intraoperative use by allowing the surgeon to take measures to improve stability.

The data presented clearly indicate that the DensiProbe instrumentation and measurement principle are eligible for routine intraoperative use by trained surgeons. Interpretation of possible correlations between BMD values measured by means of DEXA and the Peak Torque values assessed by DensiProbe has to be considered very carefully, because BMD and Peak Torque analyse bone at a different scale. Only within the framework of a multicenter study it will be possible to include a sufficient number of patients for calculation of the methods' predictive value towards implant failure and to verify acceptance of the method by the surgeons.

No abstract