Taking bone grafts from the skull is an accepted standard practice in facial-skeletal surgery to treat traumatic or reconstructive deformities , . The main advantages of skull grafts are a barely visible scar under the hairline, pain reduction of the donor site, a short withdrawal period and rapid revascularization. One method to gather bone material from the parietal skull is the lift of outer table transplants. Therefore a transplant is contoured and the edges are flattened with a pear-shaped milling head. This enables the cutting of the diploe with an oscillating saw followed by transplant lifting and cutting of the remaining diploe with the aid of an osteotome. The inner table thus remains intact . However, the lift of cranial grafts also harbors risks such as bleeding, numbness, subdural hematomas, intracranial injury or even the death of the patient . To minimize these risks, appropriate forces and speeds of the hand drives have to be applied during the surgical procedure , . Therefore, for novice surgeons, extensive training is necessary . Traditionally, this training occurs in the operating room by watching a surgery and making first hands-on experience under supervision of an experienced surgeon. Learning is also done on human and animal specimen, live animals or simulators . Here, model simulators or patient phantoms provide realistic haptic feedback  and represent a good training possibility and simulators with haptic feedback showed enhanced surgical-skill transfer rates . Commercially available models are very often based on polyurethane (PU) foams as defined in the regulatory ASTM F1839 . Several studies show that these PU models match in mechanical properties with human bones , . However investigations dealing with orthopedic applications on these PU based models show controverse results , , . The goal of this investigation was the development of parietal bone surrogates which are validated for drilling, milling and sawing. Therefore, characteristic parameters of these machinal processes were recorded. Additionally microtomography (μCT) images were evaluated to manufacture realistic cortical and cancellous layers with realistic thicknesses values.
2 Material and methods
2.1 Human specimen
Two human parietal bones (both from female donors, one left, 67 years and one right, 83 years respectively), which were taken from two routine craniectomies at BGU Murnau hospital, were completely released from soft tissues, rinsed with water and sterilized due to autoclaving after their removal. Afterwards the parietal calottes were cut into rectangular pieces. Adjacent skull proportions, like the frontal or temporal bone, were omitted (see Figure 1). Care was taken that the cut samples were as flat as possible with a size of approximately 2 × 8 cm, respectively. The harvested parietal bone specimens were numbered and scanned with a μCT (see Section 2.4). For storage reasons, the parietal bone pieces were frozen at −37° and were completely defrosted before the measurements (24 h in saline solution at room temperature/4°C).
2.2 Artificial specimen
The custom made artificial skull caps described in this study were made of a two-component PU resin as base material. Further, varying amounts of additives (water, PU color, fillers and surfactants) were added in order to create a cortical shell with cancellous core and to enable fluoroscopy visualization. Two different combinations were molded in a two step approach resulting in a harder outer layer mimicking the cortical shell and a softer, porous structure imitating the inner cancellous diploic structure (see Figure 1). One surrogate (AS1) was molded with the same materials for cortical and cancellous layers. Another bone model (AS2) was manufactured with varying mixtures. For the machinery measurements, four parietal specimens for each composition were cut from the skull caps.
2.3 Measurement setup
For the measurement of forces during drilling (Fdrill), milling (Fmill) and sawing (Fsaw) a custom made test rig was used. A surgical drive (Implantmed SI-923,W&H Dental, Bürmoos, Austria) was placed vertically in a rig made of profiles. The following surgical tips were used: a 2 mm drillhead (Stryker Corp., Kalamazoo, MI, USA), an engraving mill head (Variodent, Neuss, Germany) and a 10 mm sawblade (W&H Dental, Bürmoos, Austria). The specimens were clamped on to a six degree of freedom (DOF) sensor (resolution 1/16N, nano25, ATI Industrial Automation, Apex, USA) with a custom made adapter and further fixed with screws. The force sensor was further adapted to a linear actuator (DNCE-63-400-BS-”10”P-Q, Festo AG & Co. KG, Esslingen, Germany) carrying the specimen to the surgical tip. The surgical tip was fixed within the surgical drive and the specimen was placed above the surgical tip with the linear actuator. The specimen was placed in a manner that the surgical tip was not touching the specimen before the start of a measurement. Force measurements started simultaneously with the movement of the actuator and hence, the specimen and the surgical hand piece. The speed of the surgical handpieces was uniformly set to the maximum of 40.000 rotations per minute. The tools were penetrated into the bones using a constant speed of 1 mm/s (drilling, milling), 0.5 mm/s (sawing) for given insertion depths of 10 mm (drilling, reaming), 5 mm (sawing), respectively. After the planned measurement depth was reached, the measurement was automatically stopped and the specimen was pulled away from the surgical tip. For drilling and milling the specimens were positioned flat to enable a perpendicular insertion of the tool tip into the bone specimen. For sawing measurements, the bone specimens were tilted 90° to allow only the cutting of the diploic bone. According to the surgical procedure, the maximum insertion forces during drilling and milling of the outer table and during sawing of the diploe were analyzed. Five measurements for each specimen and procedure were performed.
All human specimen were scanned with a μCT (μCT80, Scanco Medical, Brüttisellen, Switzerland, see Figure 2). For all images, the resolution preferences of 70 kVp, 114 μA and a 200 ms integration time were used resulting in a slice thickness of 0.09 mm. The cut artificial specimen were tilted for 90 degrees and were photographed. The datasets and photos were analyzed using ImageJ (1.49 V, National Institutes of Health, Bethesda, MD, USA, ). Prior to the evaluation the datasets were edited according to Larsson and colleagues . The images were filtered with a 3D median filter and a manual color threshold was set. Tabula externa and diploe thickness values were measured with the caliper tool in ImageJ. For each sample, thickness measurements were performed on five locations evenly distributed on the bone specimen. Furthermore, the same five locations were measured for four different μCT slices resulting in 20 measurements for each bone specimen.
2.5 Statistical analysis
Statistical analysis was performed using the software SPSS (SPSS Statistics 22, IBM, Armonk, NY, USA). The examined data were tested for normal distribution (Shapiro-Wilk test) and homogenous variances (Levene test). For normal distributed data unpaired t-tests were used for the testing of differences between groups. Non-normal distributed data were tested with Whitney-U-test. For all tests, a p-value of 0.05 or less was considered significant.
The examined data showed normal distribution and homogenous variances, thus unpaired t-tests were used. The maximum insertion forces in axial direction were analyzed. All measurement results are summarized in Figure 3. Significant differences between the tested groups were not observed.
The examined data showed normal distribution and homogenous variances. The measurement results of the evaluation of the μCT images are summarized in Table 1. Unpaired t-tests revealed significant differences between the thickness values of both cortical tables of AS1 (p-values <0.001* and 0.004*) and the diploe of both artificial parietal bones (p-values 0.001* and 0.002*).
Results of thickness measurements (n = 20) of parietal skulls (Human) and two customized artificial skulls (AS1, AS2; mean ± standard deviation (p-value for comparison to human specimen).
|Specimen||Externa (mm)||Diploe (mm)||Interna (mm)|
|Human||1.44 ± 0.32||2.40 ± 0.31||0.79 ± 0.21|
|AS1||0.89 ± 0.09|
|6.73 ± 0.12|
|1.93 ± 0.99|
|AS2||1.71 ± 0.39|
|5.65 ± 0.32|
|0.63 ± 0.10|
Commercially available bones do not deliver realistic haptic feedback during orthopedic interventions , , . In addition, a surgical simulator for the training of parietal graft lifts is currently not available and therefore it is apparent that sufficient artificial bones would be beneficial for surgical training. According to the surgical procedure of a parietal graft lift, the maximum insertion forces during drilling and milling of the outer cortical table and the average forces during sawing of the cancellous diploic bone were analyzed. The measured forces during drilling, milling and sawing of the bone surrogates did not show significant differences to those gathered from human bones. The spreads of all measured data of the artificial skulls was as large as those from human bones. The ingredients of the custom made artificial skulls were all weighed in with an acceptable range of ± 5%. Thus, variances in compositions as well as cortical coating thicknesses arise. But, these fluctuations are desireable leading to varying training scenarios for the surgeons. Significant differences were recognized for the diploic thickness of the artificial skulls. This is due to the molding process and the extension of the open cell PU foam in the mold. The parietal skull caps were already tested by two cranio-maxillo-facial surgeons. Both surgeons certify good haptical feedback for both bone surrogates, however, they report a thicker diploic layer with insufficient porosity compared to human specimens.
In conclusion, the results identified the AS2 as a suitable surrogate mimicking the properties of human parietal bone during outer table grafting. As a result a training simulator for harvesting cranial bone grafts was built with this artificial bone surrogate.
The authors would like to thank W&H Dentalwerk Bürmoos GmbH for their generous provision of machinal surgery tool required for this study. M.Hollensteiner thanks TÜV Austria for financial support.
Research funding: The Research Group for Surgical Simulators Linz (ReSSL) acknowledges the financial support by the Austrian Research Promotion Agency (FFG) within the program line Cooperation & Innovation (COIN) and project number 845436. Conflict of interest: Authors state no conflict of interest. Material and Methods: Informed consent: Informed consent has been obtained from all individuals included in this study. Ethical approval: The research related to human use complies with all the relevant national regulations, institutional policies and was performed in accordance with the tenets of the Helsinki Declaration, and has been approved by the authors’ institutional review board or equivalent committee.
Bruno B. Cranial bone harvest grafting. AORN J. 1994;59:242–51.
Jackson R, Smith I, Mixter J. Nasal bone grafting using split skull grafts. Ann Plast Surg. 1983;11:533–40.
Strong E, Moulthrop T. Calvarial bone graft harvest: a new technique. Otolaryngol Head Neck Surg. 2000;123:547–52.
Díaz I, Gil J, Louredo M. Bone drilling methodology and tool based on position measurements. Comput Methods Programs Biomed. 2013;112:284–92.
Okada D, de Sousa A, Huertas R, Suzuki F. Surgical simulator for temporal bone dissection training. Braz J Otorhinolaryngol. 2010;76:575–8.
Lin Y, Wang X, Wu F, Chen X, Wang C, Shen G. Development and validation of a surgical training simulator with haptic feedback for learning bone-sawing skill. J Biomed Inform. 2014;48:122–9.
Kneebone R. Simulation in surgical training: educational issues and practical implications. Med Educ. 2003;37:267–77.
Ström P, Hedman L, Saerna L, Kjellin A, Wredmark T, Fellaender-Tsai L. Early exposure to haptic feedback enhances performance in surgical simulator training: a prospective randomized crossover study in surgical residents. Surg Endosc. 2006;20:1303–8.
ASTMF1839-08(2012), Standard Specification for Rigid Polyurethane Foam for Use as a Standard Material for Testing Orthopaedic Devices and Instruments. American Society for Materials and Testing Std. 2012;08.
Heiner D. Structural properties of fourth-generation composite femurs and tibias. J Biomech. 2008;41:3282–4.
Johnson A, Keller T. Mechanical properties of open-cell foam synthetic thoracic vertebrae. J Mater Sci Mater Med. 2008;19:1317–23.
Aziz M, Nicayenzi B, Crookshank M, Bougherara H, Schemitsch E, Zdero R. Biomechanical measurements of cortical screw purchase in five types of human and artificial humeri. J Mech Behav Biomed Mater. 2014;30:159–167.
Hollensteiner M, Fürst D, Schrempf A. Artificial vertebrae for a novel simulator in minimally invasive spine surgery. Biomed Tech. 2013;58:4409–11.
Tsuji M, Crookshank M, Olsen M, Schemitsch E, Zdero R. The biomechanical effect of artificial and human bone density on stopping and stripping torque during screw insertion. J Mech Behav Biomed Mater. 2013;22:146–56.
Doube M, Kłosowski MM, Arganda-Carreras I, Cordeliéres FP, Dougherty RP, Jackson JS, et al. BoneJ: Free and extensible bone image analysis in ImageJ. Bone. 2010;47:1076–9.
Larsson E, Brun F, Tromba G, Cataldi P, Uvdal K, Accardo A. Quantification of structural differences in the human calvarium diploe by means of X-ray computed microtomography image analysis: a case study. IFMBE Proceedings. 2011;37:599–602.