Jeffrey A. Fialkov

In vivo bone engineering in a rabbit femur.

The repair of bone defects in reconstructive surgery has significant limitations. Donor site morbidity, limited supply of autograft, and risks and complications associated with allografting and synthetic bone substitutes are among the most significant. In an effort to address these problems, the search for an ideal bone replacement has led to the development of a new method of poly(lactide-co-glycolide) (PLGA) foam processing, enabling the production of a biodegradable scaffold with similar porosity to human trabecular bone. In this study, these scaffolds were evaluated for bone repair in vivo in a femoral critical-sized segmental defect in New Zealand White (NZW) rabbits. Three groups of nine animals were investigated. In the first group, the critical-sized defects were empty. Scaffolds alone were implanted in the second group, whereas autologous bone marrow cell-loaded scaffolds were implanted in the third group. Animals ambulated freely for 8 weeks after surgery, and bone formation throughout the defects was serially assessed radiographically and quantified using a bone formation index (BFI) measure. Postmortem radiography and histology were also undertaken to examine bone formation. There was a significant effect of applying this technology to the amount of bone formed in the defects as determined by the BFI (F = 3.41, P < 0.05). The mean BFI for the cell-loaded scaffolds was greater than for the control group at all measured time points (2-, 4-, 6-, and 8-week radiographs). This difference was significant for the 2- and 8-week radiographs (P < 0.05). Qualitative histological assessment confirmed these findings. We concluded from these findings that these PLGA scaffolds loaded with marrow-derived progenitor cells yield significant bone formation in a critical-sized rabbit femoral defect. This technology comprising a novel scaffold design and autologous cells may provide an alternative to current strategies for reconstruction of bony defects.

Internal fixation of mandibular angle fractures: a meta-analysis.

Background: The degree of rigidity of internal fixation required for the treatment of mandibular angle fractures has long been at the center of debate in the literature. A statistical comparison between rigid fixation and monocortical fixation has been difficult because of multiple terms, definitions, and technical variations. The purpose of this study was to use the meta-analysis tool to combine information from multiple studies and to compare complication rates for different fixation methods. Methods: An English language literature search was conducted for articles on mandibular angle fractures. Information was collected on four variables of interest: compression/noncompression technique, monocortical/bicortical screws, number of plates, and location of plates. Five outcome rates were analyzed: infection, reoperation, hardware removal, malunion, and nonunion. Meta-analyses were run using Comprehensive Meta Analysis, version 2.2.03. Results: Twenty-four studies with relevant data on the variables and outcomes of interest met the inclusion criteria. Significantly higher rates of infection, reoperation, and hardware removal were found for compression compared with noncompression, two plates compared with one plate, and for plates located on both the inferior and superior borders as compared with superior or inferior only. There were also significantly higher infection rates for bicortical screws compared with monocortical screws and higher malunion rates for compression compared with noncompression plating techniques. Conclusion: The results of this meta-analysis found lower complication rates with the use of noncompression, monocortical, and single-plate fixation, supporting the trend toward a single, superiorly placed, monocortical miniplate for fixation of mandibular angle fractures.

Toward characterization of craniofacial biomechanics.

Surgical reconstruction of craniofacial deformities has advanced significantly in recent years. However, unlike orthopedic surgery of the appendicular skeleton, the biomechanical characterization of the human craniofacial skeleton (CFS) has yet to be elucidated. Attempts to simplify facial skeletal structure into straightforward mechanical device analogies have been insufficient in delineating craniofacial biomechanics. Advanced computational engineering analysis methods offer the potential to accurately and completely define the internal mechanical environment of the CFS. This study developed a finite element (FE) model in the I-deas 10 FEM software package of a preserved cadaveric human CFS and compared the predictions of this model against in vitro strain measurement of simulated occlusal loading forces from a single masseter muscle. The FE model applied shell element modeling to capture the behavior of the thin cortical bone that may play an important role in stabilizing the facial structures against functional loads. In vitro testing included strain measurements at 12 locations for a total of 16 independent channels with less than 150 N of tensile force applied through the masseter muscle into the zygomatic arch origin at 4 different orientations, with 3 trials of 500 recorded data points for each loading orientation. Linear regression analysis yielded a moderate prediction (r2 = 0.57) between the model and experimentally measured strains. Exclusion of strain comparisons in regions that required greater modeling assumptions greatly improved the correlation (r2 = 0.70). Future validation studies will benefit from improved placement of strain gauges as guided by FE model predicted strain patterns.

Postoperative infections in craniofacial reconstructive procedures.

The rate of, and possible risk factors for, postoperative craniofacial infection is unclear. To investigate this problem, we reviewed 349 cases of craniofacial skeletal procedures performed from 1996 to 1999 at our institution. Infection rate was determined and correlated with the use of implants, operative site, and cause of deformity. The inclusion criteria consisted of all procedures requiring autologous or prosthetic implantation in craniofacial skeletal sites, as well as all procedures involving bone or cartilage resection, osteotomies, debridement, reduction and/or fixation. Procedures that did not involve bone or cartilage surgery were excluded. The criteria for diagnosis of infection included clinical confirmation and one or more of 1) intravenous or oral antibiotic treatment outside of the prophylactic surgical regimen; 2) surgical intervention for drainage, irrigation, and or debridement; and 3) microbiological confirmation. Among the 280 surgical cases that fit the inclusion criteria and had complete records, there were 23 cases of postoperative infection (8.2%). The most common site for postoperative infection was the mandible (infection rate = 16.7%). Multiple logistic regression analysis revealed gunshot wound to be the most significant predictor of postoperative infection. Additionally, porous polyethylene implantation through a transoral route was correlated with a significant risk of postoperative infection.

A stereotactic system for guiding complex craniofacial reconstruction

A stereotactic system has been designed to address the problem of achieving symmetry in complex and extensive craniofacial defects. Preliminary testing suggests that such a system, which allows for the intraoperative application of preoperative CT planning, will be useful in guiding the reconstruction of congenital or acquired bony time, is being used to investigate the correlation of intraoperative globe position following enophthalmos correction with long-term outcome, particularly as it relates to the size and location of the orbital defect, and the timing of the procedure.

A New System for Severity Scoring of Facial Fractures: Development and Validation

Facial fractures are often the result of high-velocity trauma, causing skeletal disruption affecting multiple anatomic sites to varying degrees. Although several widely accepted classification systems exist, these are mostly region-specific and differ in the classification criteria used, making it impossible to uniformly and comprehensively document facial fracture patterns. Furthermore, a widely accepted system that is able to provide a final summary measure of fracture severity does not exist, making it difficult to investigate the epidemiologic data surrounding facial fracture severity. In this study, a comprehensive method for panfacial fracture documentation and severity measurement is proposed and validated through a retrospective analysis of 63 patients operated on for acute facial fracture. The severity scale was validated through statistical analysis of correlation with surrogate markers of severity (operating room procedure time and number of implants). Spearman correlation coefficients were calculated, and a statistically significant correlation was found between severity score and both number of implants and operating room procedure time (R = 0.92790 and R = 0.68157, respectively). Intraclass correlation coefficients were calculated to assess intrarater and interrater reliabilities of the severity scale and were found to be high (0.97 and 0.99, respectively). This severity scale provides a valuable, validated research tool for the investigation of facial fracture severity across patient populations, allowing for systematic evaluation of facial fracture outcomes, cost-benefit analysis, and objective analysis of the effect of specific interventions.

Distraction osteogenesis of radiation-induced orbitozygomatic hypoplasia.

In the last decade, the application of distraction osteogenesis to the craniofacial skeleton has grown to include not only deformities of the mandible, but of the midface, palate, dentoalveolar region, and calvarium. A major advantage of distraction osteogenesis lies in the simultaneous soft tissue histogenesis that accompanies the bony distraction process, allowing for potentially lower relapse rates and improved cosmesis. Although this may seem appropriately suited to irradiation-induced deformities of both hard and soft tissues, there is little in the literature as to the efficacy of this technique in patients who have received radiotherapy. To introduce an effective application of this technology, and highlight some advantages and disadvantages of its application in the irradiated craniofacial skeleton, we present a case of distraction osteogenesis of the orbitozygomatic complex in a patient with radiation induced orbitozygomatic hypoplasia.

Characterization of the bending strength of craniofacial sutures

The complex, thin and irregular bones of the human craniofacial skeleton (CFS) are connected together through bony articulations and connective tissues. These articulations are known as sutures and are commonly divided into two groups, facial and cranial sutures, based on their location in the CFS. CFS sutures can exhibit highly variable degrees of interdigitation and complexity and are believed to play a role in accommodating the mechanical demands of the skull. This study aimed to evaluate the mechanical behavior of CFS bone samples with and without sutures and to determine the effect of sutural interdigitations on mechanical strength. Sagittal, coronal, frontozygomatic and zygomaticotemporal sutures along with adjacent bone samples not containing sutures were excised from six fresh-frozen cadaveric heads. The interdigitation of the sutures was quantified through μCT based analysis. Three-point bending to failure was performed on a total of 29 samples. The bending strength of bone samples without sutures demonstrated a non-significant increase of 14% as compared to samples containing sutures (P=0.2). The bending strength of bones containing sutures was positively correlated to the sutural interdigitation index (R=0.701, P=0.002). The higher interdigitation indices found in human cranial vs. facial sutures may be present to resist bending loads as a functional requirement in protecting the brain.

Measuring pulsatile forces on the human cranium.

The cyclic stresses in the cranium caused by pulsation of the brain play an important role in the design of materials for cranioplasty, as well as craniofacial development. However, these stresses have never been quantified. In this study, the force in the epidural space against the cranium was measured intraoperatively in 10 patients using a miniature force probe. Heart and ventilatory rates computed from the force tracing correlated closely with the corresponding measured values in the patients, confirming that the forces measured were indeed a result of brain pulsation. The mean outward systolic normal and tangential stresses were 54.2 kilo-Pascals (kPa) and 345.4 kPa, respectively. The systolic shear stress was 199.8 kPa. Through mechanotransduction, these stresses play a role in cranial development. The calculated yield stress of a cranioplasty repair was 0.4 MPa, which is within one order of magnitude of the known strength of common calcium-phosphate cements. This indicates a possible relation of these pulsatile forces and occult failure of calcium-phosphate cement cranioplasties through material fatigue.

High resolution bone material property assignment yields robust subject specific finite element models of complex thin bone structures

Accurate finite element (FE) modeling of complex skeletal anatomy requires high resolution in both meshing and the heterogeneous mapping of material properties onto the generated mesh. This study introduces Node-based elastic Modulus Assignment with Partial-volume correction (NMAP) as a new approach for FE material property assignment to thin bone structures. The NMAP approach incorporates point spread function based deblurring of CT images, partial-volume correction of CT image voxel intensities and anisotropic interpolation and mapping of CT intensity assignment to FE mesh nodes. The NMAP procedure combined with a derived craniomaxillo-facial skeleton (CMFS) specific density-isotropic elastic modulus relationship was applied to produce specimen-specific FE models of 6 cadaveric heads. The NMAP procedure successfully generated models of the complex thin bone structures with surface elastic moduli reflective of cortical bone material properties. The specimen-specific CMFS FE models were able to accurately predict experimental strains measured under in vitro temporalis and masseter muscle loading (r=0.93, slope=1.01, n=5). The strength of this correlation represents a robust validation for CMFS FE modeling that can be used to better understand load transfer in this complex musculoskeletal system. The developed methodology offers a systematic process-flow able to address the complexity of the CMFS that can be further applied to create high-fidelity models of any musculoskeletal anatomy.