All experiments were performed in accordance with relevant guidelines and regulations of the Ethics Commission for the Federal State of Salzburg (EK No: 1198/2021).
Tests on anatomical specimens:
A commercially available AO hand fixator (DePuy Synthes, Raynham, MA, U.S.) was used for the EFF [13]. Pins were selected in 3.5 and 4.0 mm diameter. The clamping jaws and carbon rods were available in standard dimensions.
A CT scan was performed of 3 cadaver skulls in the following settings: non-fractured, after setting the fractures and after applying the EFF and fracture reduction (Computed tomography: Somatom Emotion 6, Siemens Healthineers, Germany).
The determination of the maximum pull-out force was carried out by a calibrated tensile force gauge (Sauter FK 500, Fmax = 500N, Germany). The anatomical specimen was fixed tightly to the table with tension straps. The device was attached to the respective pin with an angle-stable hook system and pulled in orthograde direction. The pull-out forces were measured for the palatal pins, the pins in the zygomatic bone, and the supraorbital pins. As a reference, a mandibular fixator was also attached and the pull out forces for each pin separately measured, since this technique is widely established [14].
CMFs were set with a chisel on 13 anatomical specimens in the sense of a combination of a Le Fort 1, 2 and 3 fracture with a palatal split and a Naso-Orbitoethmoid fracture as this combination of fractures is common in LIC [15, 16]. Primarily, the hard palate was split longitudinally like a Le Fort 1 fracture. Then the zygomatic bone was fractured on both sides like a Le Fort 2 fracture. Finally, a fracture of the orbita was placed laterally on both sides and of the nasale bone in the sense of a Le Fort 3 fracture.
After setting the fractures, the EFF was applied. The reduction was performed bottom-up. Primarily, the palatal pins were inserted, and a crossbar was attached to close the palatal split. If a rotational instability was still present, or the placement of lateral palatal pins was unsatisfactory, a 2.0 mm Kirschner wire was inserted behind the cuspid of the canines or 1st premolar teeth, bridging the split to improve the rotational instability of the construct. Subsequently, two pins were inserted into the zygomatic bones and connected to the first rod. Finally, the supraorbital pins were placed on both sides. The final step was the biomechanical evaluation of the pull-out forces for the individual pins.
Application of the EFF step by step
The self-drilling pins were inserted after performing a stitch incision with a scalpel, blade size 11.
1. Reduction and fixation of the palatal split
Primarily, the reduction of the palatal split was performed. For this purpose, two medial palatal pins were inserted into the hard palate, directly below the nostrils on both sides. They were drilled until the tip could be palpated with the finger under the oral mucosa. The pins were connected to a carbon rod and, after the palatal split had been reduced with the finger under sight and control, were turned tight (Fig. 1A and B). If the pins in the lateral hard palate showed decreased clinical stability, a K-wire osteosynthesis was inserted as an alternative in combination with the pins in the medial hard palate to increase rotational stability (Fig. 1b).
2. Reduction and fixation of the Le Fort 1 and 2 fracture components
If the zygomatic bone was not affected by a comminuted fracture, one pin was inserted into it on both sides and the pins were connected to the transverse carbon rod from step one. The reduction was achieved by temporary occlusion. In this position, the clamping jaws were turned tight. The reduction was checked clinically.
3. Reduction and fixation of the Le Fort 3 fracture component
As a last step, the EFF was completed by inserting the supraorbital pins in the lateral third of the upper orbital margin. After inserting the pins, they were connected to one another with a rod and to the carbon rod of the zygomatic pins. After reduction of the Le Fort 3 fracture component, the clamping jaws were tightened (Fig. 1C and D). If the zygomatic pins could not be inserted because of a comminuted fracture of the bone, the reduction was achieved by temporary occlusion at this step.
Analysis of the optimal insertion angle for the longest intraosseous drilling possibility:
To give a detailed impression of the bony conditions for the pin insertion areas of the EFF with optimal insertion angles and the corresponding possible intraosseous drilling depth, the following data of 100 CTs of the skull of healthy subjects were digitally determined:
-
Supraorbital area
-
Zygomatic bone
-
Hard palate medial
-
Hard palate lateral
The measured values of the angles and possible intraosseous drilling depth are shown in Fig. 2.
Digital measurement of CT scans for analysis of optimal pin insertion angles
To define the optimal insertion angles of the supraorbital pins, digital analysis of a clinical dataset of 100 healthy CT scans of the skull was performed (Age: 33.4 ± 9.7 years). A central line was drawn in axial slice and insertion site for supraorbital pins was determined in lateral third of the os frontalis pars orbitalis. Subsequently, the possible intraosseous drilling depth was measured at different angles (-45°, -40°, -30°, -20°, -10°, 0°, 10°, 20°, 30°, 40°, 45°). The percentile analysis is used to determine the optimal surgical angle. In a second step, we examined the distribution of the optimal and worst surgical angles using a histogram. The goal is to determine the optimal and unfavourable window for the insertion angle seen over all patients.
A skull was anatomically dissected postoperatively and after removal of the EFF by an anatomist to identify structures at risk and document any damage.
Statistical methodology
To show the differences in the pull-out force among the different pin insertion areas, an ANOVA analysis with Kruskal-Wallis and a Dunn's Multiple Comparison Test was carried out. To determine the optimal insertion angle, we use simple statistical methods, such as percentile analysis and histograms.