Parietal bone surrogates for graft lift training

Objective: Fast bone machining procedures are surgical standard procedures. To avoid damage, appropriate forces and speeds have to be applied. Therefore extensive training is necessary. One training possibility is through model simulators, which consist of a patient phantom with realistic haptic feedback. Studies have shown that simulators with haptic feedback have higher surgical skill-transfer to novel surgeons than visual training only. One surgical procedure, where fast machinery procedures are used is the lift of outer table grafts from the parietal bone. Taking grafts from the skull vault is an accepted standard practice and popular in facial-skeletal surgery to treat traumatic, reconstructive or cosmetic deformities. The goals of this study were, first the development of parietal bone surrogates which are validated for drilling, milling, sawing training and second, the development of a simulator prototype for parietal bone grafting. Methods: Two artificial skulls made of polyurethane and selected additives were molded. Due to a three-step molding process, the skull surrogates represent the three bony layers of the skull: a cortical inner and outer table sandwiching the cancellous diploe. Two parietal bones from female donors (67 and 83 years) were used as a reference. According to the surgical procedure of the graft lift, the maximum insertion forces during drilling and milling of the outer table and during sawing of the diploe were analyzed. The surgical tips were driven into the bones with a constant speed of 1 mm/s (drilling, milling) and 0.5 mm/s (sawing) respectively for given insertion depths (10 and 5mm). The speed of the hand drive was set to 40.000 rotations per minute. Insertion forces were measured with an axial force sensor. Additionally all specimens were scanned with a CT to gather outer table and diploe thickness values. Results: Statistical analysis was performed using the software SPSS Statistics 22. The examined data showed normal distribution (Shapiro-Wilk-test) and homogenous variances (Levene-test). Thus unpaired t-tests were used to detect differences between human and artificial skulls. For all tests, a p-value of 0.05 or less was considered significant. No significant differences of measured forces between human and artificial specimens could be observed. Significant differences were found for the diploe thickness of both bone surrogates and for both cortical tables’ thickness values of the first artificial skull. Conclusion: The measured forces of the surgical tips during bone insertion revealed comparable haptics for both artificial skull surrogates in comparison to human specimens. In contrast the thickness measurements of the bony layers indicated that the anatomical structures of the first artificial skull significantly differ from the human references. The results further identified the second artificial skull as a suitable surrogate mimicking the properties of human parietal bone during bone grafting. As a result a training simulator for harvesting cranial bone grafts was built with this artificial bone surrogate.