Bony contouring in head and neck microvascular reconstruction

Head and neck bony microvascular reconstruction

The simplest strategy for restoring bony contour involves “pre-bending” an IF reconstructive plate to the native maxillofacial skeleton prior to tumor resection. In these situations, once the bony segment intended for extirpation is exposed, an IF implant is templated against the native bone surface prior to osteotomies and resection. However, when tumors involve the plating surface of the bone or contour is already compromised by fractures or erosion, this disruption can preclude the practice. For example, exophytic tumors of the buccal surface of the mandible can disrupt accurate contouring of a reconstructive plate along the outer cortex of the bone and spoil this contouring option. This predicament has fostered innovative approaches to bony fixation.

We present here three strategies that can be employed to restore bony contour when using composite free tissue transfer and provide suggested applications for each.

External fixation

First reported in 1981, Bergman et al.^[9] showcased a series of 20 patients with oral and oropharyngeal cancers treated with an intraoperatively applied external fixator system for 12-14 weeks post-operatively. Notably, these patients were left “swinging” without bony replacement of the resected segments. The use of the system in this setting was aimed at preventing soft tissue contracture and the resulting deviation of the native mandible toward the resected side. In their series, they proved the feasibility of these strategies for oncologic segmental mandibulectomies reporting improvements in trismus, oral care, and no delay in receiving adjuvant chemoradiotherapies.

The next evolution of the technique came in 2001 with a Kirschner wire based system^[10]. Louis et al.^[10] utilized this EF system for providing intraoperative stabilization of the mandible to facilitate en-bloc mandibular tumor extirpation. This was followed by application of a titanium IF reconstruction plate to bridge the resected segment. No bony replacement is described in their technique, leaving the reconstruction plate covered by soft tissue alone.

By the early 2000’s, composite microvascular free-tissue transfer was reported citing improved bony union, decreased infectious complications and resorption when compared to free or allograft bony reconstructons^[11-13]. The adjunct of temporary intraoperative EF and composite free flap reconstruction was introduced in a series of six patients successfully treated in 2002^[14]. Their technique involved placing an EF device prior to tumor extirpation under wide mandibular exposure, and then removing the external bracketing during the bony osteotomies. The removal of the external frame allowed for increased working space during the oncologic resection. Once the mandibular segment was removed, the external frame was reapplied to the bone posts to restore the bony conformation. The vascularized bone grafts were then inset with titanium mini-plates and thereafter the device was explanted entirely. All six patients had complete bony union with no evidence of cross-bite deformities or condylar head dislocation and satisfactory external projection and contour^[14].

Apart from oncologic defects, EF has utility in traumatic injuries of the maxillofacial skeleton, congenital deformities, edentulous patients, and pathologic fractures with osteoradionecrosis. In a mixed series of 10 patients treated with a modern titanium EF system, Cornelius et al.^[15] provided further support of the technique. Their population included eight patients with sequelae from treatment of their head and neck cancers including pathologic fractures, orocutaneous fistulas and locoregional tumor recurrences. They utilized EF as a stabilizing measure for a range of 40-170 days prior to converting to either IF hardware alone or vascularized bone transfer. Their experience shows EF as a means of allowing interval soft tissue wound healing to occur prior to definitive management with tissue transfer and IF to avoid hardware complications or malunion.

EF has also been shown to be a useful tool when tumor resection involves or necessitates removal of the mandibular condyle with or without portions of the adjacent maxilla. In these situations, intraoperative stabilization can be accomplished by capitalizing on the bony stability of the uninvolved zygoma or remaining maxilla as an anchoring point for an EF post. We do not routinely remove the EF hardware during the tumor resection and osteotomies, but instead fashion the conformation of the external frame in a way that facilitates adequate working space for the ablation [Figures 1 and 2].

Figure 1. External fixation of a single segment mandibular reconstruction.

Figure 2. Anterior composite resection with external fixation intraoperative stabilization prior to scapula composite free flap reconstruction.

Though useful, EF is not without potential complications. When used in a temporary intraoperative manner, the primary concern is loosening of shifting of the apparatus disrupting the integrity of the alignment. In this situation, realignment through maxillomandibular fixation with traditional arch bars or interdental wiring is advocated in dentulous patients^[3]. Pin placement through soft tissue can also result in facial nerve injury at a reported rate of 7%^[3]. With intraoperative use only, the concern for pin-site infections or soft-tissue complications is minimal given the wide direct bony exposure typically required for tumor extirpation.

When considering the cost, the use of EF intraoperative can be compared to other modalities such as VSP. Since the construct is only used intraoperatively, the only implant charge is typically for the Schanz screws or posts that directly drill into the native bone at our institution. For example, in a case such as Figure 2 with an anterior composite resection construct, only the three Schanz screws pictured are considered implanted and charged for as the remainder of the construct is re-sterilized. This equates to approximately $1200-2400 USD in hardware costs and can avoid the need for VSP and fabricated hardware costs.

Virtual surgical modeling and planning

Imaging-guided surgical navigation, modeling and preoperative planning have been in development for craniofacial surgery since the 1990’s, and with recent advances have gained widespread adoption in head and neck reconstruction^[4,5,16-18]. The most common variant of this technology is often referred to as a VSP in which computed tomography studies are used to guide both resection margins and free flap osteotomies to recreate the bony defect using vascularized bone by free tissue transfer [Figure 3A-C]. Culie et al.^[5] described their experience with VSP in 18 patients against a historic internal cohort showing decreased operative and ischemia time with use of the technology. This correlated with similar contemporary published series at the advent of VSP in mandibular reconstruction touting benefits of improved accuracy of the bony contour, improved aesthetics, and increased rates of bony union^[19-22]. Supporting conclusions from these series cited the benefit of allowing for bony contouring in-situ at the donor site under perfused conditions prior to harvest of the osteocutaneous composite free flap. This lent to decreased ischemia time for the flaps while also allowing improved workflows when a two-team approach was utilized for ablative and reconstructive portions of the operations.

Figure 3. Anterior mandible ossifying fibroma with 3D reconstructed imaging for VSP (A). Ossifying fibroma specimen after anterior composite resection (B). Ex vivo bony contour of a fibula composite free flap reconstruction with VSP after closing osteotomies made with pre-fabricated cutting guides (C). Reconstruction with the three segment fibula composite flap in vivo with a patient specific implant reconstruction plate (D). VSP: Virtual surgical planning.

However, the cost-effectiveness of routine VSP use in head and neck reconstruction must be factored in a modern medical climate that prioritizes value. An institutional analysis by Fatima et al.^[23] estimated an average added cost of $7099 per person with use of VSP over conventional strategies and described an increased rate of flap loss and post-operative infectious complications. Even with controls for decreased operative time and free-flap ischemia time, the added cost of VSP did not add objective value in their analysis^[23]. Similar studies have supported VSP as an economically viable technology in head and neck reconstruction on the basis of saved operative time alone^[24-26].

There are situations in bony head and neck reconstruction that VSP offers a seemingly substantial advantage to other techniques. Guiding indications for the use of VSP have been suggested, which include: defects requiring multiple free flaps, defects involving multiple sites or subsites of the maxillofacial skeleton, the need for multiple osteotomies, presence of osteoradionecrosis or irradiated fields, or a history of a high-velocity ballistic injury with severe bony comminution^[5,27]. In our review, no studies have focused specifically on the outcomes and cost-effectiveness of VSP in the setting of these specific indications compared to less complex reconstructions.

Rapid prototyping for head and neck reconstruction

Used as either an adjunct or substitute for VSP systems, RP or three-dimensional printing has proved valuable in the realm of head and neck reconstruction. RP refers to the process of generating a physical model from digital data taken from a computed tomography or magnetic resonance image, commonly referred to as “3D printing”. There are various methodologies for creating these medical prototypes which can be either inert or biologic depending on the anticipated application. The technique for rendering these models also varies both by equipment and process. Modeling can be either additive or subtractive in nature. In additive processes, a model is built de novo by sequential on-laying of the desired material to gradually create a final prototype. Conversely, subtractive processes utilize removal of material from an existing block to arrive at a final model^[28]. Both processes are utilized in modern head and neck reconstructive applications depending on the usage of the material.

RP can be utilized at multiple levels of the head and neck reconstruction process. It can be utilized to build either patient-specific implants such as inert titanium plating for maxillary or mandibular reconstruction or for biologic implants such as tissue scaffolding^[6-8]. RP can also be employed to build tailored cutting guides based on image-guided modeling to improve the osteotomy accuracy in bony contouring. Finally, RP can be used to create patient-specific idealized craniofacial models to allow for ex vivo reconstruction plate contouring and bony segment templating in the absence of VSP cutting guides^[29-31] [Figure 4A-C]. Mirror image digital processing can be incorporated to create idealized contours from the non-diseased side of the maxillofacial skeleton to avoid contour disruptions from an exophytic or destructive tumor.

Figure 4. Templating utilizing a rapid-prototype model intraoperatively to achieve restored mandibular contour (A). Intraoperative photo showing the applied mandibular hardware after an oral composite resection of the left mandibular body (B). Reconstruction utilizing a subscapular system composite flap to achieve bony contour restoration (C).

An increasing number of institutions have moved to internal RP production to mitigate the cost and time required in formal VSP^[32]. The 3D printer equipment costs have been estimated as minimal as $1000 USD with material costs for printing these models reaching $1 USD and 4-5 h of printing time required per application^[32]. This has expanded the usage of RP in many medical centers and allowed for increasing research and educational applications in head and neck reconstruction. Table 1 gives a generalized comparison of the available modalities for bony contouring of head and neck defect reconstruction.