Reconstruction of Dural Defects of the Endonasal Skull Base

Endonasal, endoscopic approaches to the cranial base have undergone significant technique refinement over the past decade. Repair of the resultant defects remains perhaps the most important concern with these approaches; however, recent advances suggest that with careful attention to the closure, these procedures can be done with acceptable rates of morbidity. In this review, the authors discuss known techniques for the repair of endonasal defects, and provide some insight based on their experience.

The last decade has seen a marked increase in the ability to remove cranial base lesions with minimal disturbance to the unaffected brain or cranial bone. Improved endoscopic and surgical technology have made intracranial surgery partly or entirely within tight spaces, such as the nasal cavity, obviating a craniotomy or brain retraction to approach lesions of the anterior cranial fossa, sella, clivus, and even many off-midline structures.

Although the possibility of performing complex intracranial procedures without a skin incision or a craniotomy is appealing for many, the limited space makes a neat dural incision and watertight dural closure essentially impossible. While the risk of postoperative cerebrospinal fluid (CSF) leak is manageable in purely intrasellar surgery, because larger lesions are addressed via the endonasal approaches and a greater area of the cranial base is disrupted to expose these lesions, reconstruction of these defects becomes more important, and more difficult.

Several techniques have been used to attempt to repair these defects. Admittedly, no current technique is perfect in all situations; however, existing repair methods represent significant improvements, and are an essential part of the repertoire of anyone contemplating endonasal endoscopic approaches to the cranial base.

Basic principles of repair

Perhaps the sole factor working in favor of successful repair of large cranial base defects is the well-vascularized and highly proliferative nasal mucosa surrounding the defect. If allowed to, mucosa overgrows even large defects, providing a lasting barrier against CSF egress via secondary intention. Perhaps the most important tool for repairing these defects is the recognition during the exposure that preservation of as much mucosa as possible with its native blood supply, prevention of unnecessary barriers to mucosal contact with the closure construct, such as unnecessary bony spurs, and minimization of unnecessary mucosa trauma or excessive cauterization probably all improve the likelihood of a permanent mucosa seal being formed.

Another critical feature of successful repair is layered closure. Regardless of closure method, if continued CSF egress is permitted by the method, even if the flow rate is modest, it can maintain and eventually epithelialize a cranionasal fistula, and prevent a permanent mucosal seal from forming at the edges of the defect. Simply put, each additional layer of closure provides one more barrier around which the CSF must go to enter the nose. Although there is a limit, in general, the more layers of different closure materials, the better.

Finally, once the layered closure is in place, it is important to assure that it can resist reasonable stresses without moving. Various different methods for achieving this have been described including the use of bioabsorbable plates to buttress the intradural portion of the repair, the temporary implantation and inflation of the balloon of a Foley catheter, and securing thicker portions of the repair to surrounding tissue using the U-Clip (Medtronic, Minneapolis, MN, USA). Regardless of the method used, the repair must be given some supporting structure to provide a head start for healing, because extrusion of the repair construct almost guarantees failure.

Repair materials

Nonvascularized Autografts

Autografts are the classic endonasal repair substrates, and can be used to serve several functions in the repair of these defects. Although these are technically simple to harvest and place, they are devascularized tissue, and generally are slowly resorbed and replaced by viable tissue, as opposed to forming their own healthy tissue.

Fat is the most commonly used autograft, and can easily be harvested from the lower abdominal quadrant or the lateral thigh. Fat grafts have come into common use because they are easily malleable and shapeable, and thus not only serve as a barrier to CSF egress but also can prevent the prolapse of neural tissue, like the optic apparatus, through the bony defect left after the approach. Although fat grafts are compressible, there is a limit, and overpacking the defect with fat can cause mass effect and neurologic deficit. The authors have found a variant of the bath-plug technique to be useful in making the fat graft into a firm source of support without the risk of compression-critical structures. In this technique, a slightly oversized piece of fat is wrapped in Surgicel, which is encircled with a 4-0 suture with the ends of the suture threaded through the holes of an absorbable buttressing plate. After introducing the fat into the cavity and tucking the ends of the plate under the edges of the bony defect, the suture is pulled tight against the buttressed plate to compact the fat behind the plate. If sized correctly this fat can form a good seal, and obliterate the potential site of neural prolapsed with minimal risk of overpacking.

Fascia lata can be harvested from the lateral thigh and can be taken at the same time as fat. As a biologic covering, it might be better incorporated into the eventual healed tissue than synthetic grafts. There is a minor risk of injury to lateral femoral cutaneous nerve, so the benefits of fascia should be tempered against this, especially if fat is not going to be used in the repair.

Vascularized Mucosal Flaps

Although harvesting these flaps can be complex and time consuming, their intact blood supply, their large potential surface area, and their thick tissue mass makes them an important recent development in endonasal surgery at the skull base. It is important to plan these flaps into the exposure if they are to be part of the closure, as not only do most endonasal approaches involve procedures that potentially could injure the blood supply of these flaps, namely the sphenopalatine artery and its branches, but also the flap needs to be safeguarded in some region of the nose that will not be in the primary working axis of the surgery, such as the choana or the maxillary sinus. The ideal location to store the flap depends on the approach, and this should be planned ahead.

Nasoseptal flap

This is a workhorse vascularized mucosal flap, owing to the relative simplicity of elevating this flap with its blood supply, its large potential surface area, and the simplicity by which it can be mobilized to cover defects in nearly every region of the skull base. The flap is based on the posterior nasal branch of the sphenopalatine artery, which is identified by locating the natural osteum of the sphenoid sinus and the choana, between which the artery runs. Using monopolar cautery, the mucosa is incised on the lower edge of the sphenoid osteum and the upper edge of the choana from lateral to medial onto the nasoseptal mucosa. The mucosa incisions spread rostrally and caudally to widen the flap on the nasal septum, but care should be taken not to bring the upper incision too high on the septal mucosa until anterior to the middle turbinate, for fear of injury to the olfactory epithelium. The incisions continue from posterior to anterior, widening to the maximum width possible just anterior to the middle turbinate, stopping at the mucosa-epithelium junction. The flap is then elevated off the nasal septum with a Cottle or other periosteal dissector ( Fig. 1 ) and stored in a safe place until needed.

Inferior turbinate flap

The inferior turbinate flap represents a back-up flap when the nasoseptal flap is not an option, usually when previous transphenoidal surgery has either cut or otherwise devascularized the nasoseptal mucosa. This flap is based on the posterolateral nasal artery, a branch of the sphenopalatine artery. Given its early take off after the sphenopalatine foramen, it has a short vascular pedicle in the nasal vault, and thus it is typically necessary to perform a complete uncinectomy, anterior ethmoidectomy, and a wide maxillary antrostomy, to identify and skeletonize the sphenopalatine artery through the posterior maxillary sinus wall following it medially into the nasal vault. The posterolateral nasal artery is then followed onto the inferior turbinate, and a mucosal flap is elevated from the medial surface of the turbinate up to the anterior head of the turbinate.

Tunneled periosteal flaps

In cases when both the nasoseptal and inferior turbinate flaps are not available, it is possible to harvest pericranium through a linear hemicoronal scalp incision. After performing a wide maxillary antrostomy, the sphenopalatine artery is followed out into the pterygopalatine fossa, the vidian nerve is sacrificed, and the pterygopalatine ganglion is mobilized so that the medial pterygoid plate can be drilled to provide room for the flap. The posterolateral maxillary wall is removed in part to communicate the sinus with the infratemporal fossa. After a pericranial graft is harvested based on the anterior branch of the superficial temporal artery, a lateral canthotomy incision is used to dissect the temporalis muscle down the pterygomaxillary fissure and the lateral maxillotomy made internally. After soft tissue dilation of the tract with tracheostomy dilators passed over guidewires, the pericranium is tunneled into the nasal cavity through the maxillary tunnel created endoscopically, and rotated into position.

Alternatively, to close anterior skull base defects, for example after a transcribriform approach, a pericranial flap can be tunneled through the frontal sinus through a small trephination in the nasion. This technique obviously is used following a wide bilateral frontal sinusotomy (ie, Draf 3 procedure).

Rescue flaps

In the rare case when the nasoseptal, inferior turbinate, and tunneled transpterygoid flaps are not available, other more technically challenging flaps have been described. A somewhat smaller flap from the middle turbinate based on the posterior sphenopalatine branches can be raised with some difficulty. In rare circumstances, the palatal mucosa can be harvested and a flap elevated based on the descending palatine arteries.

Nonautogenous Grafts and RStructural Material

Various materials for providing a barrier to CSF egress are commercially available, most notably bovine pericardium and dural matrix. Few data support the use of one material over another, what is important is that these layers provide a smooth, flat surface for mucosa to grow over, and that they completely cover the defect. The authors have found the use of a single row of a bioabsorbable plate (Stryker) to be a useful buttress to keep the material from leaving the defect under normal physiologic stress.

Biologic Adhesives

A wide variety of adhesive materials exist, with different durations needed for absorption. Although no specific data exist supporting their use, it seems wise to provide a more watertight seal by applying one of these adhesives to the edges of the repair. The authors have been successful with BioGlue (CryoLife, Kennesaw, GA, USA), particularly with repair of small sellar defects, owing to its thick consistency and long duration before it is biologically degraded. No specific data support the use of one material over another at present.

Only gold members can continue reading. Log In or Register to continue