Abstract
Background
Hip fractures affect 1.6 million people globally per annum, associated with significant morbidity and mortality. A large proportion are extracapsular neck of femur fractures, treated with the dynamic hip screw (DHS). Mechanical failure due to cut-out is seen in up to 7% of DHS implants. The most important predictor of cut-out is the tip-apex distance (TAD), a numerical value of the lag screw”s position in the femoral head. This distance is determined by the psychomotor skills of the surgeon guided by fluoroscopic imaging in theatre. With the current state of surgical training, it is harder for junior trainees to gain exposure to these operations, resulting in reduced practice. Additionally, methods of simulation using workshop bones do not utilise the imaging component due to the associated radiation risks. We present a digital fluoroscopy software, FluoroSim, a realistic, affordable, and accessible fluoroscopic simulation tool that can be used with workshop bones to simulate the first step of the DHS procedure. Additionally, we present the first round of accuracy tests with this software.
Methods
The software was developed at the Royal National Orthopaedic Hospital, London, England. Two orthogonally placed cameras were used to track two coloured markers attached to a DHS guide-wire. Affine transformation matrices were used in both the anterior-posterior (AP) and cross table lateral (CTL) planes to match three points from the camera image of the workshop bone to three points on a pre-loaded hip radiograph. The two centre points of each marker were identified with image processing algorithms and utilised to digitally produce a line representing the guide-wire on the two radiographs. To test the accuracy of the system, the software generated 3D guide-wire apex distance (GAD) (from the tip of the guide-wire to a marker at the centre of calibration) was compared to the same distance measured with a digital calliper (MGAD). In addition, the same accuracy value was determined in a simulation scenario, from 406 attempts by 67 medical students.
Results
The median absolute inaccuracy of FluoroSim with 270 measurements was 3.35mm (IQR = [1.15mm, 6.53mm]). The absolute inaccuracy showed a graded increase the further away the tip of the guide-wire was from the centre of calibration; MGAD ≤10mm, median absolute inaccuracy = 1.53mm; MGAD 10mm<x≤20mm, median absolute inaccuracy = 4.97mm; MGAD >20mm, median absolute inaccuracy = 7.23mm. Comparison between all three groups reached significance (p < 0.001). In a simulation scenario with medical students, FluoroSim had a significantly greater median absolute inaccuracy of 4.79mm compared to the testing scenario (p < 0.001).
Conclusion
FluoroSim is a safe and inexpensive digital imaging adjunct to workshop bones simulation. To our knowledge this technology has not been explored in the context of DHS simulation, and has the potential to be extended to other orthopaedic procedures.
