Computer-Assisted Maxillofacial Surgery

Mandibular fractures are very common in maxillofacial trauma surgery. While previous studies have focused on possible risk factors related to post-operative complications, none have tried to identify pre-existing conditions that may increase the risk of mandibular fractures. We hypothesized, through clinical observation, that anatomical conditions involving poor dental contacts, such as malocclusions, may increase the risk of mandibular fractures. This work was subdivided into two parts. In the first part, Digital Imaging and Communications in Medicine (DICOM) data of four healthy patients characterized by different dentoskeletal occlusions (class I, class II, class III, and anterior open bite) have been used to develop four finite element models (FEMs) that accurately reproduce human bone structure. A vertical and lateral impact have been simulated at increasing speed on each model, analyzing the force distribution within the mandibular bone. Both vertical and lateral impact showed higher level of stress at the impact point and in the condylar area in models characterized by malocclusion. Specifically, the class III and the open bite models, at the same speed of impact, had higher values for a longer period, reaching critical stress levels that are correlated with mandibular fracture, while normal occlusion seems to be a protective condition. In the second part of this study, the engineering results were validated through the comparison with a sample of patients previously treated for mandibular fracture. Data from 223 mandibular fractures, due to low-energy injuries, were retrospectively collected to evaluate a possible correlation between pre-existing malocclusion and fracture patterns, considering grade of displacement, numbers of foci, and associated CFI score. Patients were classified, according to their occlusion, into Class I, Class II, Class III, and anterior open bite or poor occlusal contact (POC). Class I patients showed lower frequencies of fracture than class II, III, and open bite or POC patients. Class I was associated with displaced fractures in 16.1% of cases, class II in 47.1%, class III in 48.8% and open bite/POC in 65.2% of cases (p-value < 0.0001). In class I patients we observed a single non-displaced fracture in 51.6% of cases, compared to 12.9% of Class II, 19.5% of Class III and 22.7% of the open bite/POC group. Our analysis shows that class I appears to better dissipate forces applied on the mandible in low-energy injuries. A higher number of dental contacts showed a lower rate of multifocal and displaced fractures, mitigating the effect of direct forces onto the bone. The correlation between clinical data and virtual simulation on FEM models seems to point out that virtual simulation successfully predicts fracture patterns and risk of association with different type of occlusion. Better knowledge of biomechanics and force dissipation on the human body may lead to the development of more effective safety devices, and help select patients to plan medical, orthodontic/dental, and/or surgical intervention to prevent injuries. Full article

This randomized prospective clinical study aims to analyze the differences between the computer-assisted planned implant position and the clinically realized implant position using dynamic navigation. In the randomized prospective clinical study, 30 patients were recruited, of whom 27 could receive an implant (BLT, Straumann Institut AG, Basel, Switzerland) using a dynamic computer-assisted approach. Patients with at least six teeth in their jaws to be implanted were included in the study. Digital planning was performed using cone beam tomography imaging, and the visualization of the actual situation was carried out using an intraoral scan. Two different workflows with differently prepared reference markers were performed with 15 patients per group. The actual clinically achieved implant position was recorded with scan bodies fixed to the implants and an intraoral scan. The deviations between the planned and realized implant positions were recorded using evaluation software. The clinical examinations revealed no significant differences between procedures A and B in the mesiodistal, buccolingual and apicocoronal directions. For the mean angular deviation, group B showed a significantly more accurate value of 2.7° (95% CI 1.6–3.9°) than group A, with a value of 6.3° (95% CI 4.0–8.7°). The mean 3D deviation at the implant shoulder was 2.35 mm for workflow A (95% CI 1.92–2.78 mm) and 1.62 mm for workflow B (95% CI 1.2–2.05 mm). Workflow B also showed significantly higher accuracy in this respect. Similar values were determined at the implant apex. The clinical examination shows that sufficiently accurate implant placement is possible with the dynamic navigation system used here. The use of different workflows sometimes resulted in significantly different accuracy results. The data of the present study are comparable with the published findings of other static and dynamic navigation procedures. Full article