Virtual surgical planning, surgical navigation, intraoperative imaging, and customizable implants have emerged as reliable tools in implant surgery and orthognathic surgery and are gaining traction in the trauma setting.
In the trauma patient, difficulty exists in restoring the 3-dimensional facial skeleton for several reasons.
The technological options discussed in this article can be used either independent of one another or in concert, depending on the case, surgeon preference, and hospital resources. The approaches can be implemented for both the acute trauma patient or for secondary repair.
Over the past decade, the utilization and implementation of various technologies into the trauma workflow have increased substantially in oral and maxillofacial surgery.
Virtual surgical planning, surgical navigation (SN), intraoperative imaging, and customizable implants have emerged as reliable tools in implant surgery and orthognathic surgery and are gaining traction in the trauma setting as well. Adoption of these modalities has been driven by a desire from surgeons to improve accuracy, predictability, and precision while reducing time and cost, therefore improving patient outcomes. Correction of orbital volume, re-creation of facial symmetry, and establishment of premorbid occlusion continue to elude many practitioners who manage the complex trauma patient.
In the trauma patient, difficulty exists in restoring the 3-dimensional (3-D) facial skeleton for several reasons. Poor visualization from the overlying soft tissue envelope, difficulty assessing deep skeletal junctions, and inaccurate plate bending/internal fixation all can yield suboptimal results. Intraoperative errors often can have a cumulative result, yielding a patient with facial asymmetry, diplopia, malocclusion, occlusal cants, and the like.
The technological options discussed in this article can be used either independent of one another or in concert, depending on the case, surgeon preference, and hospital resources. The approaches can be implemented both for the acute trauma patient or for secondary repair.
Three-dimensional computer-assisted (virtual) surgical planning
Three-dimensional computer-assisted surgical planning–computer-aided design/computer-assisted manufacturing (CAD/CAM) technology in oral and maxillofacial surgery has become a reliable resource in all aspects of the specialty. Proprietary CAD/CAM software exists for many applications of oral and maxillofacial surgery, which can be applied to the trauma patient. An aspect that may need consideration for the trauma patient is the scanning protocol of the emergency or trauma services. Trauma patients may have a maxillofacial computed tomography (CT) scan from the emergency department; however, they may require a second CT suitable for Digital Imaging and Communications in Medicine (DICOM) formatting and CAD/CAM utilization. When desiring a DICOM data set, either a medical grade CT scan using 1.25mm cuts or cone-beam CT (CBCT) should be obtained.
The CAD/CAM software allows a surgeon to import the DICOM data set and generate a 3-D virtual representation of the patient’s craniomaxillofacial skeleton. A simulated surgical repair can be carried out using mirroring, segmentation, reduction, or virtual osteotomies ( Fig. 1 ). , The virtual reconstruction/data set then can be used in several ways.
The simplest application is to allow the surgeon to manipulate the data set via segmentation, mirroring, reflection, or insertion to establish a treatment plan. The next application is the fabrication of a stereolithic model of the virtual reconstruction. In addition, patient-specific implants (PSIs) can be manufactured from the virtual session. Finally, the surgical plan can be uploaded to a SN instrument for real-time intraoperative feedback, discussed later in the article.
Preoperative CAD/CAM virtual surgery is the most straightforward application of this technology and does not add significant down time or increase the length of hospital stay, in general. Once the digital workflow is established, there are few limitations to its access. A conversation with the emergency department/trauma team about admission CT scanning protocols avoids the need for a second scan because of poor quality. Surgical manipulation and planning in the trauma patient can take place once the DICOM data set has been uploaded to the system. There are no additional equipment needs and it does not rely on a capitol purchase by the hospital; therefore, it can be widely adopted in most clinical settings. A DICOM formatted disc is submitted to a third-party vendor, and, within hours, the images can be made available for virtual surgery. The surgeon performs the surgery in the virtual setting and then applies the plan in vivo.
A second application of CAD/CAM technology in the trauma patient is the fabrication of stereolithic models as a template for plate contouring. On completion of the virtual surgery, a stereolithic model or models can be printed or milled, representing the virtual reconstruction. The availability of reconstructed models allows the surgeon to contour, or prebend, plates to the desired, new positions ( Fig. 2 ). The plates then are sterilized preoperatively and internally fixated to the patient. This approach is intended to improve accuracy and reduce operating room time.
A third application of CAD/CAM technology is the manufacture and printing of customized plates or PSIs. After a virtual modeling/reconstruction session has been completed, hardware can be fabricated specifically to a patient’s needs. The access to customized plating solutions allows the practitioner to achieve precision and accuracy in the most complex cases. Indications include comminution, continuity defects, severe displacement, loss of landmarks, absence of dentition, and malunion cases ( Fig. 3 ).
Utilization of virtual surgical planning is intended to improve precision; however, in inexperienced hands, this may not always be the case. Mistakes can be introduced into the workflow. Prebending to stereolithic models or custom fabrication of plates to ideal skeletal conditions without the consideration of soft tissue limitations can yield plates that are not practical or achievable once applied to the patient ( Fig. 4 ). Additionally, stereolithic models do not reproduce bone integrity, density, or quality and plating may not have an acceptable recipient site for screw placement once applied to a live patient. Lastly, without a bite registration, stereolithic models and/or custom plate fabrication cannot be relied on for establishing occlusion. Generally speaking, if dentate segments are involved, a dental impression or dental scan should be obtained to assist in the virtual fracture reduction process.
In some trauma patients, dental impressions may not be possible and malocclusions may occur because of this. As pictured ( Fig. 5 ), a reconstruction plate was adapted to the stereolithic model with precision but without obtaining dental impressions. The resultant anterior open bite was an unfortunate result despite a seemingly excellent adaptation and virtual reduction of the mandible fracture.
There are advantages and disadvantages to CAD/CAM of PSIs. Goals of custom plating include decreased operating room time, improved accuracy, and reduced perioperative complexity. Disadvantages include increased direct cost of hardware, turnaround time for plate manufacturing in the trauma setting, and unforeseen soft tissue barriers, which can prevent placement of the customized hardware.
The manufacture of custom plates remains proprietary among the major hardware vendors, and they approach the customization process uniquely. Plates are manufactured either by a 3-D CAD/CAM printing or milling. Milling describes a process whereby material is cut away (milled) after it has been poured into a mold, to create customized solutions ( Fig. 6 ). Customized plates can eliminate plate fracture from bending fatigue and offer a high degree of variability, including cribs for bone grafting, inferior border retention hooks for perioperative stability, and complex orbital floor plates (see Fig. 3 ). In addition to custom titanium options, customization of polyether ketone (PEEK) implants and porous polyethylene implants can be fabricated for soft tissue augmentation ( Fig. 7 ).
Intraoperative navigation/surgical navigation
Intraoperative navigation/image-guided navigation has emerged as a tool available to the oral and maxillofacial surgeon by allowing real-time visualization of bony landmarks via comparison to preoperative CT scans as well as the planned images that were reconstructed in the virtual environment. In addition, it allows the user to view anatomic structures that are not routinely visualized during surgery, such as the orbital apex.
SN relies on 3 primary components, similar to Global Positioning Systems, which are used daily. The components are a localizer (the satellite), a surgical probe (car), and a CT (the map). The CT/DICOM is uploaded into the navigation software/workstation and then used as a road map for surgery, with the probe functioning as the location device.
Navigation technology is based on the synchronization of the intraoperative position of the instruments with the imaging of a patient’s anatomy obtained previously via CT or magnetic resonance imaging. The synchronization is realized through image registration, which is akin to mapping between a coordinate system of the preoperative CT images and that of the real-world/real-time patient during surgery. Registration of fiducial markers on or in the patient must occur, and this can be accomplished with either invasive or noninvasive techniques. A noninvasive Light Emitting Diode (LED) surgical mask can be applied to the patient’s facial skin ( Fig. 8 A), with this approach having some obvious limitations in maxillofacial trauma. For one, the mask can become distorted or displaced by saliva, sweat, surgical preparation, and/or the surgeon. In addition, a phenomenon known as structural image drift can be encountered in the trauma patient. Structural image drift describes topographic soft tissue changes secondary to edema. Images acquired preoperatively may not represent a patient’s soft tissue morphology at the time of surgery. Displacement of the registration device or structural image drift introduces errors and should be avoided. An alternative noninvasive indexing option is known as surface tracing and relies on numerous points with the probe to calibrate the device.
Invasive registration options include a skull post or screws placed in strategic areas, such as the maxilla or infraorbital rims ( Fig. 8 B). Although these approaches are less likely to be displaced during surgery, the position of the invasive registration devices should be monitored, in particular in cases of extended length.
Lastly, special considerations should be taken for mandibular surgery due to its lack of bony attachment to the viscerocranium. In a trauma setting, the mobility of the mandible can be circumvented by placing the patient in maxillomandibular fixation for both the CT scan and the live surgery.
After the fiducial registration, the orientation of any tracked instrument can be displayed on the screen, demonstrating its real-time position in relation to the preoperative images and surgical plan ( Fig. 9 ).
Preoperative surgical planning (CAD/CAM) can be combined with SN to execute a surgical plan from a virtual environment. Virtual data are imported into a navigation system, which then can be used to provide guidance for accurate reduction, placement of hardware, and identification of crucial structures. For example, a virtual reconstruction of a zygoma fracture can be performed by mirroring the uninjured side (see Fig. 9 ). The mirrored image then is overlaid on the affected side using an alternative color. The reconstruction plan is uploaded to the navigational system and used as a virtual surgical template for navigation. The surgery is performed, and the navigation system is used to verify the final position, which was established preoperatively.
When utilizing navigation without a CAD/CAM plan, this technology is used most commonly for retrieval of foreign bodies and identification of hard-to-see structures in the deep spaces of the head and neck. For example, SN can be useful for treatment of orbital floor fractures without a preoperative planning session. After patient indexing, the navigation probe can be used to indirectly identify the depth of the posterior ledge from which an orbital floor plate rests. The depth of the orbital apex can be measured with precision and a plate can be fashioned to an exact length based on the data obtained from the navigation. Plate placement presumably then is more accurate, while also avoiding the optic canal and its contents. This approach can be used for any fracture whereby the final position can be used in comparison with the reference/uninjured side.
A newer approach that has shown promise is known as statistical shape modeling, which can be considered an alternative to mirroring. This approach is based on diffeomorphic deformations and principle component analysis to extract the main modes of variation. The registration on the defect serves as a tool to find the model most suitable to match the anatomic structure of the unaffected parts of the target skull. The goal of statistical shape modeling is to automate the process and obviate preoperative CAD/CAM modeling sessions, which can be time consuming. Although early in development, it has been shown to be a reliable and accurate alternative to mirroring and may represent a more simplistic pathway in the digital workflow.
Advantages of SN are that it provides real-time feedback to the surgeon, in particular from deep anatomic regions that cannot be visualized directly; and, for surgeons who are fluent with their available system, it saves time, is more precise (reported accuracy to ±2 mm), reduces radiation exposure, and can provide less invasive approaches. , Disadvantages include a steep learning curve, hospital cost/availability, introduced error from unstable bony or soft tissue references, and difficulty referencing the mandible. ,
As technology advances, wider availability of integrated navigation systems with intraoperative CT scanning and/or CBCT scanning leads to better use of the technology by improved perioperative feedback. Integrated systems allow for perioperative reassessment and are based on images obtained during surgery as opposed to the virtually predicted positions.
Intraoperative three-dimensional imaging
Two modalities that are commonly available for intraoperative imaging are the intraoperative CT scanner ( Fig. 10 ) and the 3-D C-arm (CBCT). The foundation for using intraoperative imaging is to improve outcomes by eliminating the need for postoperative revision. The technology can be indispensable for evaluating final reconstructive positions and hardware position and for confirming procedure completion. Fracture patterns that cannot be visualized directly, such as the posterior aspect of the orbital floor, are excellent candidates for intraoperative imaging. This technology can assist in surgical decision making, including CT-directed revisions that otherwise would go unobserved.
Intraoperative computed tomography scanning
Since its inception in 1972, the utilization of CT scanning has been universally adopted in the practice of oral and maxillofacial surgery. It is known to have significantly better resolution, sensitivity, and specificity compared with plain films, in particular in the setting of midface injuries. With advances in CT technology has come the availability of intraoperative/portable CT scanners. Although available since the 1990s, wide accessibility to intraoperative scanning was not immediate. Even at the time of this writing, portable CT scanners remain costly ($200,000–$800,000) and hospitals may not want to indulge in the expense. Initial studies on intraoperative CT utility in maxillofacial surgery were focused primarily on orbital floor reconstruction, which remains a primary indication for its use even today. With improved access, however, expanded applications of intraoperative CT scanning include panfacial fractures, comminuted mandible fractures, displaced zygomaticomaxillary complex fractures, naso-orbitoethmoid fractures, displaced condyle fractures, and Le Fort II/III fractures. A good rule of thumb for its use is to scan the patient only if the resultant image will lead to an alteration in medical decision making. In 2018, Cuddy and colleagues demonstrated a CT-directed revision rate of 28%. Advantages of intraoperative scanning include improved accuracy of fracture reduction and hardware placement and reduced postoperative revisions. Disadvantages to intraoperative CT scanning include cost, hospital availability, added time, radiation exposure, and small size of the O-arm. Likely due to the obesity epidemic in the United States, some patients have been encountered who do not fit in the O-arm. A patient’s size limitation generally is limited to the mandible; however, obesity has been a soft contraindication encountered on several occasions.
The 3-D C-arm, also known as CBCT, first was introduced in the 1990s. It has been a widely adopted technology in neurosurgery and orthopedics and it may be more accessible than intraoperative CT scanners in many institutions. The 3-D C-arm is an open C-shaped ring, which rotates 180° around the patient in 20 seconds to 60 seconds, with less radiation dose than conventional CT. , With the dosing exposure significantly less than CT scanning, repeat scanning can be performed with less total radiation exposure than 1 single CT scan. One drawback of the C-arm is the image resolution of soft tissue. The C-arm exhibits excellent resolution for bony structures; however, soft tissue resolution is inferior to that of the CT scan ( Fig. 11 ). For a majority of maxillofacial procedures, this aspect is without consequence; however, the user should be made aware of this drawback. Patient positioning with the C-arm is not limited by patient size, because it is an open ring and should be considered in a larger patient, if available.
As with intraoperative CT scanning, hardware positioning (orbital floor plate) and skeletal landmarks (zygomaticofrontal suture) can be verified prior to the termination of the procedure and allows for changes should C-arm findings indicate the need. Either modality reduces the likelihood of surgical revisions, may reduce operating room time, and arguably improves outcomes.
Both options provide high-quality, 3-D images of the craniomaxillofacial skeleton that can be obtained under general anesthesia with the patient on the operating room table. Although not universally available at all institutions, 1 of the 2 is likely available, in particular if the hospital employs neurosurgeons. It is important to meet with the organization’s facilities management to best understand what an institution has to offer, which includes SN.