Mandibular Reconstruction With Axially Vascularised Bone Substitutes

Background:

Mandibular reconstruction is necessary following tumour resections, infections or trauma resulting in severe defects of mandibular arch continuity and sacrifice of teeth. Basic reconstruction involves the use of non-vascularized bone grafts together with restoration of lost teeth by means of dental implants and implant-supported prostheses. Smaller bony defects (<6 cm) are commonly treated with nonvascularized corticocancellous grafts harvested from the anterior or posterior iliac crest. (Goh et al. 2008) For reconstructing larger and recurrent defects, currently the vascularized free flaps are regarded as the "gold standard". The use of these flaps, however, presents several major inconveniences. Harvesting of autologous tissue may result in a significant donor site morbidity, the extent of which may vary, according to the donor site and possibly according to the intervention technique. The problems include bleeding, pain, infections, donor site fractures and prolonged hospital stay. (Hartman et al. 2002; Rogers et al. 2003) The field of Regenerative Medicine promises new alternatives to surgical reconstruction through harnessing the regenerative capacity of the human body to repair itself. In the last few decades the rapid expansion of knowledge about the biological basis of wound healing and the role of cells, signals, and biological scaffolds has drawn the attention from ''tissue reconstruction'' to ''tissue regeneration''. New strategies started to emerge aiming at mimicking the normal healing process in regenerating lost or damaged tissues. The term "tissue engineering" was officially coined at a National Science Foundation workshop in 1988 to mean "the application of principles and methods of engineering and life sciences toward fundamental understanding of structure-function relationships in normal and pathological mammalian tissues and the development of biological substitutes to restore, maintain or improve tissue function".

Tissue engineering and Regenerative medicine depend on the presence of a biomaterial promoting cell growth and proliferation. In order to populate this biomaterial (scaffold) with new tissue, the body must effectively interact with this biomaterial. This necessitates the establishment of an early and reliable angiogenic response leading to the development of an adequate blood supply for the restoration of structure and function. (Hodde 2002) The three main components required for regeneration are the cells, scaffolds, and induction molecules. When growing tissues in vitro (Tissue Engineering), all three components should exist, however, when referring to Regenerative Medicine, any of these can be provided to the body in an attempt to optimize its capacity for regenerating its own tissues. Adding cells or growth factors to the biomaterials can reinforce tissue regeneration (Pellegrini et al. 2009), but the vascularization, and thus the integration, of theses biomaterials is still considered the determining issue in the success of any critical size defect regeneration. (Novosel et al. 2011) Applying principles of regenerative medicine in the field of cranio-maxillofacial reconstruction has now become a daily practice. The wide spectrum of applications ranges from simple addition of bioactive bone fillers to much more sophisticated techniques for bone replacement and reconstruction. The indications included reconstruction after minor developmental defects, trauma, infections, benign cysts or tumours but seldom following malignant tumour excision. (Clokie and Sandor 2008; Schuckert et al. 2009; Trautvetter et al. 2011) Warnke et al (Warnke et al. 2004), who used a completely different technique than those used in the previous case reports, reported the only case of regeneration after cancer ablation. The main technical difference was related to vascularization of the regenerated tissue. While all the reported cases for mandibular regeneration used the conventional extrinsic vascularization strategy, where the constructs were left to acquire a parasitic blood supply form the recipient site of implantation, Warnke et al (Warnke et al. 2004) used an axial vascularization strategy through a prelamination procedure in the Latissimus dorsi (LD) muscle followed by free tissue transfer of the regenerated mandible. Although this technique avoided bony donor site morbidities, the need to harvest the LD muscle represented a major drawback of this prelamination technique. This single case report highlighted the need for an efficiently vascularized construct if the regenerative therapy is to be applied for larger and recurrent defects.

Axial vascularization of scaffolds aims at providing the construct with blood supply through a defined and dedicated vascular axis. In this context, the blood supply of the construct is not randomly acquired from the implantation site, and thus the implantation in an area of low vascularization potential, as in irradiated or fibrosed surgical sites could be possible (Kneser et al. 2006). The two major techniques for axial vascularization are prelamination and prefabrication.

Prefabrication of a tissue construct is simply done by implanting an arterio-venous fistula or loop (AVL) or a vascular pedicle underneath or within the construct. This results in spontaneous sprouting of vessels from the loop or pedicle and subsequent revascularization of the whole tissue construct (Erol and Spira 1979; Guo and Pribaz 2009). Prelamination is another technique introduced by Pribaz and Fine (Pribaz and Fine 1994) in 1994 where the implantation of a construct into a vascularized territory (flap) is performed to create a customized vascularized unit. The end result of both techniques is an axially vascularized unit that depends for its nourishment upon a defined vascular axis.

Two more important terms to mention in this context are ''intrinsic'' and ''extrinsic'' vascularization modes. The extrinsic vascularization of a construct denotes acquiring its blood supply from the periphery towards the centre, while the intrinsic vascularization mode denotes that the core region of the construct is being vascularized first (Lokmic and Mitchell 2008). Accordingly, prefabrication is considered an intrinsic axial vascularization strategy. The construct in prelamination, however, is extrinsically vascularized within an intrinsically vascularized territory (Eweida et al. 2012).

As the reconstruction of challenging or irradiated bone defects requires an axially vascularized tissue bulk, applying the prelamination strategy will always end up in a remarkable donor site morbidity where the whole vascularized territory (mostly a muscle flap) should be transferred to the recipient site (Mesimaki et al. 2009; Warnke et al. 2004). The prefabrication technique, however, when applied to a tissue construct, entails only the transfer of this construct with its pedicle, thus diminishing the donor site morbidity to the minimal. Moreover, the prefabrication technique could be applied at the recipient site as a primary reconstruction technique, abolishing the donor site morbidity completely (Eweida et al. 2014; Horch et al. 2014).

One of the most extensively investigated techniques to induce axial vascularization in the tissue constructs is the arterio-venous loop or fistula (Arkudas et al. 2013; Horch et al. 2012) and its superiority over the vascular bundle in terms of vascular density and tissue regeneration potential was clearly demonstrated (Tanaka et al. 2003).

Preclinical data:

The first documented idea for axial vascularization using the AV loop was described by Erol and Spira (Erol and Spira 1979) in 1979 in a rat model. Morrison et al further developed the model and inserted the loop into isolation chambers (Hofer et al. 2003). They successfully demonstrated the induction of vascularization in polymer and gel matrices (Cassell et al. 2001). Since 2006, the design and characterization of the isolation chambers and the inset of the AV loop were further developed by the work of Horch et al (Kneser et al. 2006) where the engineering of vascularized transplantable bone was first successfully demonstrated by this research group. After successful evaluation of axial vascularisation in different bone substitutes, the concept was translated from the rat model to a large animal model (sheep) (Beier et al. 2011; Boos et al. 2012). Based on that model, axial vascularization was induced in constructs made of processed bovine cancellous bone substitutes and ß-TCP/HA (Tricalcium-phosphate/ hydroxyapatite) scaffolds. The sheep AV loop showed similar vascularization patterns like that of the rat but the optimum vascularization density was reached after a longer duration (8-12 weeks versus 4 weeks respectively) (Boos et al. 2011). Induction of new bone formation in the axially vascularized sheep isolation chamber was then demonstrated through implantation of MSC (Mesenchymal Stem Cells) in combination with osteogenic growth factor BMP based on a clinically applicable ß-TCP/HA bone substitute (Boos et al. 2013).

In 2011 the investigators introduced the AV Loop model for the first time for mandibular reconstruction in goats (Eweida et al. 2012; Eweida et al. 2011). The investigators could demonstrate successful regeneration of critical size marginal mandibular defects through axial vascularization of ßTCP/HA scaffolds charged with BMP (Bone morphogenic protein). The technique was discussed previously in details (Eweida et al. 2014; Eweida et al. 2012; Eweida et al. 2011). Briefly, a critical size (3 x 2 cm) marginal defect was created at the angle of the goat mandible. An equivalent sized scaffold made of ßTCP/HA was grooved to accommodate the AV Loop created by direct anastomosis of locally available vessels under the operative microscope. The scaffold was then mounted on a titanium plate and fixed to the mandible. Through our comparative studies between the AV Loop and the non-AV loop constructs, the axially vascularized constructs showed significantly more central vascularization and markedly enhanced central bone formation. The biomechanical characteristics were remarkably enhanced as well. The safety and efficacy of the model at a preclinical level was successfully demonstrated along a 6 months follow-up period (Eweida et al. 2014).

Clinical data to date:

Only two previous studies represented case reports of craniofacial bone regeneration using axially vascularized bone substitutes (non-randomised vascularization). The first study was published in 2004 by Warnke et al (Warnke et al. 2004) and further evaluated in 2006 (Warnke et al. 2006). The authors used a prelamination technique in the LD muscle of bovine mineral blocks (BioOss) to reconstruct of a large mandibular defect. The defect was reconstructed 8 years following subtotal mandibulectomy and irradiation. The construct was charged with BMP and autogenous bone marrow aspirate from the iliac crest. Although this prelamination strategy for mandibular regeneration has shown some promising initial results, the long-term results were not free of complications.

The second report was published in 2009 by Mesimaki et al (Mesimaki et al. 2009) who reported a similar prelamination technique for vascularizing a bone substitute made of ß-TCP (beta Tricalcium phosphate) in the rectus abdominis muscle. The vascularized construct was used to reconstruct a complex maxillary defect following hemi-maxillectomy due to a large keratocyst. An extensive review of literature (up to 06.2019) has shown that the prefabrication using the AV loop was never used for bone regeneration within the craniofacial region.

The introduced technique of axial vascularization of bone substitutes, however, was successfully demonstrated in humans. A case report was presented by Horch et al who have demonstrated successful bone regeneration and in situ axial vascularization using the AV loop model in two patients with bony defects in the radius and the tibia (Horch et al. 2014). Horch et al have demonstrated a safe and successful technique, encouraging results, and a complication-frees follow up period of 72 months.

Aim:

Applying and assessment of the technique of axial vascularization of a bone substitute using the arteriovenous loop (AVL) to reconstruct a mandibular defect

Patients and Methods:

Ten patients will be included in this prospective study.

Informed consent:

The patients will be fully oriented through a documented written informed consent in Arabic language. The informed written consent will include the following points:

• The procedure is related to research (Application of this technique in the craniofacial region)
• The purpose of this research.
• Alternative methods for mandibular reconstruction.
• The technical details of the procedure.
• The advantages and disadvantages of the technique.
• The risks of the procedure.
• The possible peri- and postoperative side effects.
• The management plan for the possible side effects.
• A statement that participation is voluntary, refusal to participate will involve no penalty or loss of benefits to which the patient is otherwise entitled, and the patient may discontinue participation at any time without penalty or loss of benefits, to which the subject is otherwise entitled
• An explanation of whom to contact for answers to questions about the procedure and patient's rights, and whom to contact in the event of research-related injury to the patient.

Preoperative assessment:

The patient will be subject to thorough history taking and full medical examinations.

Investigations will include:

• Routine laboratory work-up (Coagulation profile, fasting blood sugar, serum creatinine, blood urea, Complete Blood Count)
• Panoramic x-ray view of the mandible
• Computerized tomographic angiography (CTA) of the head and neck region.
• Three-dimensional printing of the mandible and the defect to facilitate preoperative orientation, plate and mesh pre-bending

Surgical procedures:

The procedure will be done under general anaesthesia in a supine position. Through a submandibular skin incision, the mandible will be exposed. The defect will be refreshed through bone curettage. The preformed reconstruction plate will be fixed to the mandible and traversing the defect. A preformed U-shaped titanium mesh will be mounted to the mandible and fixed with screws. The ipsilateral facial artery and vein (or other available vascular axes in case of lack of the facial vascular axis) will be anastomosed using the operative microscope via a reversed vein graft to create an AV loop harvested from the forearm. The AV loop will be laid within the defect. The titanium mesh will be filled with a mixture of:

1. Silicated Hydroxylapatite granules (NanoBone® vial 1.2 ml, ARTOSS, Rostock, Germany).
2. Autogenous bone marrow aspirate from the iliac crest.
3. Bone morphogenic protein 2 (BMP 2- InductOs®, Medtronic BioPharma B.V., Tolochenaz, Switzerland) The submandibular wound closed in layers.

Follow up and Evaluation:

Close monitoring of the patient for the vital signs, patency of the AV loop (via hand Doppler) and possible perioperative complications. The patient will be discharged on the 3rd day postoperative. The patient will be followed up regularly every week for wound healing, patency of the AVL and possible postoperative complications. Serial X-ray (panoramic view) will be done monthly to monitor the new bone formation. CTA will be performed after 6 months.

According to the clinical and radiological findings further dental rehabilitation will be planned after complete healing of the defect (6-9 months postoperative). Bone biopsies from dental rehabilitation (bone drilling for implants) will be studied for bone quality and vascularization using the standard H\&E staining after decalcification.

Qualitative and quantitative data from Panoramic X-rays, CTA, and histological analysis will be evaluated for:

1. Bone regeneration and density.
2. Vascularization of bone substitutes. Results will be tabulated and compared with literature.

Risk analysis:

After successful demonstration of the safety and efficacy of the ''AV loop vascularization technique'' at the preclinical and clinical levels, we believe it is time to start applying the technique for mandibular regeneration. This new application, however, will not be without challenges. A major concern would be the hemodynamic changes resulting from the AV fistula. Although a similar model was never reported in literature, reports of the use of a distal arteriovenous fistula provide evidence that a controlled fistula placed in a peripheral artery is well-tolerated, providing that the size of the fistula is less than 1 cm and the fistula accepts less than 20 percent of the cardiac output (Blaisdell et al. 1966; Eweida et al. 2013; Horch et al. 2014; Tukiainen et al. 2006). Regarding bone density and mineralization distal to an AV fistula, contradictory results are shown in literature ranging from an increase of bone growth (Vanderhoef et al. 1963) to decreased mineralization and augmented osteopenia (Muxi et al. 2009). These studies focused on the hemodynamics and its impact on bone growth distal to the fistula. The long-term consequences of an intraosseous fistula, however, were not adequately studied in literature.

An early throbbing sensation or hum would be expected due to the initial high flow. The long-term studies at the preclinical level, however, have shown that the vascular bed matures with time leading to dumping down of the direct arteriovenous jet (Polykandriotis et al. 2007; Polykandriotis et al. 2009).