HEAD AND NECK
CHAPTER 31 RECONSTRUCTION OF THE SCALP, CALVARIUM, AND FOREHEAD
J. GUILHERME CHRISTIANO, NICHOLAS BASTIDAS, AND HOWARD N. LANGSTEIN
Scalp and forehead defects result from trauma, burns, oncologic resection, infection, radionecrosis, and congenital abnormalities. Reconstruction is dictated primarily by the size and depth of the defect and is accomplished by the simplest means possible following the reconstructive ladder. Nevertheless, complicating features of the soft tissues and underlying bone may require a more complex approach.
The forehead and scalp share five distinct anatomic layers: Skin, subcutaneous Connective tissue, galea Aponeurotica or muscle, Loose areolar tissue, and Pericranium (SCALP) (Figure 31.1). The first three layers are bound together by numerous vertical septae between the skin and the galea aponeurotica, forming a single unit that glides along the loose areolar tissue over the pericranium.
The skin of the scalp is the thickest in the body (between 3 and 8 mm) and has numerous sebaceous glands. Immediately beneath the skin lies a layer of dense connective and fatty tissue, which contains a rich net of arteries, veins, lymphatics, and sensory nerves, along with the hair follicles of the scalp. The underlying galea aponeurotica is part of a broad fibromuscular layer that covers the upper cranium from the forehead to the occiput and serves as the central tendinous confluence of the occipitalis muscle posteriorly and the frontalis muscle anteriorly. The occipitalis and frontalis muscles are thin, quadrilateral muscles, each consisting of two bellies joined in the midline by extensions of the galea. The occipitalis muscle arises from the lateral two thirds of the superior nuchal line of the occipital bone and from the mastoid part of the temporal bone. The frontalis muscle has no bony attachments. Its medial fibers are continuous with those of the procerus muscles, while its lateral fibers blend with those of the corrugator and the orbicularis oculi. The frontalis muscle joins the galea aponeurotica in the upper forehead. The galea aponeurotica is contiguous with the temporoparietal fascia (also known as superficial temporal fascia) and with the subcutaneous musculoaponeurotic system (SMAS) of the face (see Chapter 47).
FIGURE 31.1. Layers of the forehead and scalp.
Deep to the galea lies the loose areolar layer, a relatively avascular plane also known as the subaponeurotic layer, subgaleal fascia, or innominate fascia. It enables the layers above it (skin, subcutaneous connective tissue, and galea) to slide as a unit over the cranium. As such, this layer is easily dissected and is often the plane of cleavage in avulsion or scalping injuries.
The pericranium is the periosteum of the calvarium. Laterally, at the superior temporal line, it is continuous with the deep temporal fascia (temporalis muscle fascia). More inferiorly, the deep temporal fascia divides into two layers, deep and superficial, which envelop the superficial temporal fat pad and insert into the superficial and deep aspects of the zygomatic arch, respectively (Figure 31.2).1
The scalp and forehead are supplied by five paired arteries that form rich interconnections within the subcutaneous layer (Figure 31.3). From anterior to posterior, these are the supratrochlear and supraorbital arteries (internal carotid), and the superficial temporal, posterior auricular, and occipital arteries (external carotid).
The main blood supply to the anterior scalp and forehead derives from the supratrochlear and supraorbital arteries, which arise from the ophthalmic artery (first branch of the internal carotid) and enter the forehead vertically at the level of the supraorbital rim. These vessels become superficial above the brow, piercing the frontalis muscle to reach the superficial layer, where they anastomose with anterior branches of the superficial temporal arteries.
As terminal branches of the external carotid arteries, the superficial temporal arteries are the largest of the scalp vessels and supply blood to the temporal and central scalp. They course through the superficial lobes of the parotid glands and ascend in front of the auricles, traveling with the auriculotemporal nerves. Above the zygomatic arch, the superficial temporal arteries lie within the superficial temporal fascia and divide into anterior and posterior branches. These branches anastomose liberally with anterior and posterior blood supplies of the scalp. The anterior branch usually crosses the most anterior temporal branch of the facial nerve just above the lateral brow, an important anatomic landmark for finding either the nerve or the vessel.
FIGURE 31.2. Anatomic relationships in the temporal region.
FIGURE 31.3. Blood supply to the scalp and forehead.
The occipital arteries provide blood supply to the posterior scalp above the nuchal line. They run from the external carotid arteries along the vertebral muscles and join the scalp through the cranial attachments of the trapezius muscle or in the space between the trapezius and sternocleidomastoid muscles. These arteries are usually found within 2 cm of the midline at the nuchal line and usually divide into medial and lateral branches above this location. The blood supply of the posterior scalp caudal to the nuchal line derives from perforating branches of the trapezius and splenius capitis muscles.
The region of the mastoid is supplied by the posterior auricular arteries, the smallest of the main vessels of the scalp. They are sufficient for some local flaps in this vicinity but not robust enough to support the entire scalp.
The scalp is drained by veins that accompany the named arteries. Venous blood also flows through the diploe of the cranium to the dural sinuses via emissary veins. These emissary veins do not have valves and can be a portal for intracranial spread of infections in the scalp, especially those in the loose areolar layer.2 The supratrochlear and supraorbital veins join to form the angular vein on each side. They also communicate with the ophthalmic veins. The superficial temporal veins descend in front of the auricle and enter the parotid glands to join the maxillary veins and form the retromandibular veins. The posterior auricular veins join the posterior division of the retromandibular veins to form the external jugular veins. The occipital veins pierce the cranial attachments of the trapezius muscle and join the deep cervical and vertebral veins in the venous plexus of the posterior triangle on each side.
The scalp lymphatics are mainly located in the subdermal and subcutaneous levels. There are no lymph nodes in the scalp region and hence no barriers to lymphatic flow. Lymph from the scalp drains freely toward the parotid, pre- and postauricular nodes, upper cervical nodes, and occipital nodes.
The muscles of the forehead are innervated by the frontal (also known as temporal) branches of the facial nerve (cranial nerve VII). As many as five separate branches may course in the loose areolar plane below the SMAS, cross the midportion of the zygomatic arch, and reach the undersurface of the frontalis muscle (Figure 31.2). The occipitalis muscle is innervated by the posterior auricular branches of the facial nerve. The temporalis muscle is supplied by motor branches from the third division of the trigeminal nerve (cranial nerve V).
The sensory nerve supply to the anterior scalp and forehead derives from the ophthalmic division of the trigeminal nerves. The supratrochlear and supraorbital nerves arise from this branch and leave the skull through the supraorbital foramina or grooves at the supraorbital rim. The temporal scalp is supplied by the maxillary division of the trigeminal nerve (zygomaticotemporal nerve) and the preauricular scalp by the mandibular division (auriculotemporal nerves). The postauricular scalp is supplied by dorsal rami of the cervical spinal nerves (greater occipital nerve and third occipital nerve).
Several variables need to be taken into account when devising the best therapeutic approach for a scalp defect. While size and depth are the most obvious, other features may prove to be just as important in each individual case. As always, the plastic surgeon should employ the procedure at the lowest level of the reconstructive ladder that suits both the defect and the patient.
Key Planning Points
1. The Defect
Size and shape along with depth (see below) are the main determinants of the amount of tissue needed for closure. Categorizing scalp defects by their size can facilitate an algorithmic approach to reconstruction. However, defects come in different shapes, and a 2 × 10 cm defect may represent a completely different reconstructive challenge compared with a 4 × 5 cm wound, even though both span 20 cm2. For this reason, while recognizing that most scalp wounds in the real world are not perfectly round, the authors choose to categorize the size of the defects based on diameter, as depicted below.
Location. The parietal regions allow the most advancement of scalp tissue. Therefore, parietal defects are more amenable to scalp undermining and primary closure, while defects in other areas may take advantage of full advancement of parietal scalp as a flap. Reconstruction of the temporal scalp may benefit from the extra soft tissue padding provided by the underlying temporalis muscle. The occipital region provides very limited scalp mobility. Finally, for best cosmetic results, defects in hair-bearing areas should be covered using hair-bearing scalp, unless hair transplantation is planned.
Depth. Lacerations involving the galea require specific closure of this layer. When skin grafting over the calvarium, an intact pericranium is needed for reliable and durable take. In defects involving the calvaria, all exposed devitalized bone needs to be debrided before coverage. Exposed dura frequently requires some form of calvarial reconstruction for brain protection, but small areas may be resurfaced with well-vascularized scalp or other soft tissue.
2. The Surrounding Tissues
Quality. Thickness, elasticity, and vascular supply to the tissues surrounding the defect can be compromised by several factors. In addition, scars from previous operations are frequently seen in the scalp in patients referred for oncological reconstruction, further limiting the usefulness of local tissues. Burns and previous radiation therapy also diminish the use of local flaps. Infections should be controlled before the creation of flaps or insertion of tissue expanders.
For oncological defects, clear margins of malignancy are verified, before reconstruction, especially in repairs involving large areas of undermining, as these may be then need to be removed in an oncological re-excision. Local wound care or temporizing reconstruction with dermal regeneration templates, allografts, or xenografts is considered when clear margins cannot be confirmed at the time of tumor excision.
The hair line can be distorted by excessive tissue undermining and recruitment. It should also be addressed when designing rotational flaps or planning scalp tissue expansion. Defects involving the hair line may require reconstruction with hair-bearing and non–hair-bearing tissues.
3. The Patient
Before considering complex procedures or multi-stage scalp reconstruction, it is paramount to assess the patient’s overall health, level of function, compliance, and personal preference. Patients with significant comorbidities may not be candidates for prolonged surgical procedures. Patients’ compliance and ability to keep pressure off the surgical site are required after pressure ulcer reconstruction. Individuals undergoing tissue expansion need to cope with a lengthy process, some level of pain, and significant disfiguration. Some may prefer a single-stage surgical treatment instead.
Oncological patients require special considerations. Attention is given to preoperative chemotherapy agents that could compromise wound healing. Nutritional state is checked. Patients may present for reconstruction before radiation and/or chemotherapy and require a reliable short-term reconstructive approach that will allow them to continue cancer treatment in a timely fashion. Therefore, tissue expansion is usually not an initial option. In severe cases, patient’s life expectancy may influence the choice of reconstruction.
Primary Closure. Primary closure is possible for scalp defects up to 3 cm in diameter. The scalp has a rich vascular supply, which allows for some degree of tension. On the other hand, the galea is relatively inelastic and prevents tissue recruitment. Undermining the surrounding tissues at the subgaleal plane yields some mobilization, which may be sufficient for wound coaptation. Open wounds are irrigated and devitalized or fibrotic edges are sharply debrided before closure.
Secondary Closure. Although this is an option for all soft tissue wounds, wound healing by secondary intention should be discouraged in the scalp, unless other reconstructive techniques are not available to the surgeon. Complete wound healing can be a lengthy process, depending on its size and depth. Exposed calvarial bone devoid of pericranium is prone to osteomyelitis and stable spontaneous coverage is unlikely. Finally, the hairless scar that results is cosmetically problematic.
Skin Grafts. Skin allografts and xenografts can be used for temporary closure of scalp defects of any size. One to two weeks may pass before the graft is lost by rejection. During this time, for instance, conclusive pathological information can be obtained on the resected specimen (benign versus malignant tumor, margin status, need for radiation therapy, etc.) before a final reconstructive plan is implemented. In other occasions, the surgeon may not be certain of the ability of a scalp wound to accept a skin graft and heal, most commonly because of lingering infection, injury from radiation therapy, tissue ischemia, or malnutrition. In such cases, a trial period with xeno- or allografts can provide the wound with time to improve and at the same time be informative to the surgeon, with no expense of the patients’ donor sites. Immediate or early loss of the xeno/allograft indicates that the wound bed is not healthy, while good take of the xeno/allograft prompts the surgeon to proceed with the definitive reconstruction.
Split-thickness autografts are used in scalp reconstruction for coverage of the primary defect, the secondary defect (donor area of a local flap), or a flap devoid of skin (pericranial, muscular, omental, etc.) and require a healthy vascularized bed for success. Alopecia and some degree of color mismatch are expected. When placed over the calvarium, an intact pericranium is preferred (Figure 31.4). Skin grafts over pericranium, however, have low tolerance to shearing forces and are prone to wound breakdown. Burring down the outer cortical layer of the skull to induce the formation of granulation tissue prior to grafting is also discouraged for the same reason. When available, pericranial or temporoparietal fascial flaps can be used to recreate a thicker vascularized wound bed and decrease the wound’s depth. For best cosmetic results, skin autografts to the scalp are not meshed.
Dermal Regeneration Templates. Integra (Integra LifeSciences, Plainsboro, NJ) is a synthetic bilaminate composed of a collagen lattice covered with a thin silastic sheet, which renders it impermeable to water. Vascularization of the deeper layer usually occurs in a few weeks, at which point the silastic sheet is removed and a thin split-thickness skin autograft is applied. Advantages of the use of Integra in scalp reconstruction include the simplicity of initial wound management, the potential as a wound-temporizing technique, the increase in thickness of the wound bed allowing for more stable coverage with a subsequent autograft, and the decrease in donor site morbidity because a thinner skin graft is needed. Disadvantages may include the availability and cost of the product, the relatively long period required with a pressure dressing, and alopecia.
Local Flaps. Local scalp flaps can be designed as partial or full thickness. Partial-thickness flaps, such as pericranial and galeal flaps, are most useful to create a vascularized wound bed when full-thickness scalp flaps (FTSFs) are not available or desired. Unlike coverage with dermal regeneration templates, which require time for vascular ingrowth, partial-thickness scalp flaps can be covered with a skin graft immediately. The pericranial flap includes the periosteum of the skull along with the overlying loose areolar tissue (Figure 31.5). The flap should be based on one of the named scalp vessels or originate from their general vicinity. Perfusion across the midline, often meager, can be enhanced by including the galea with the flap. This galeal–pericranial flap is a much more robust parcel of tissue that can be dissected off the neighboring scalp, placed over the bare skull, and covered with a skin graft.3,4 This dissection is tedious and bloody and may result in some alopecia at the donor site.
FIGURE 31.4. Scoring of the galea of a scalp flap.
FIGURE 31.5. Closure of a cerebrospinal fluid leak following craniotomy for tumor ablation.
In general, coverage of scalp defects with FTSFs provides superior functional and aesthetic results compared with reconstruction with grafts, especially when postoperative radiation therapy is anticipated. An FTSF is an axial flap, based on the scalp’s major supplying vessels. Although the rich subcutaneous anastomotic connections would allow the entire scalp to live on a single major vessel, FTSFs should incorporate as many vessels as possible without compromising the mobilization needed. A long, wide flap is preferable to a small flap. Long, curvilinear incisions (Figure 31.6) facilitate flap advancement and allow primary closure of the secondary defect (donor area). A large flap is also advantageous in the event of peripheral wound breakdown, being more amenable to re-advancement than a small flap. The spherical shape of the calvaria is to be considered, as two-dimensional measurements of a scalp defect (from a photograph, for instance) underestimate the size of flap needed for coverage. For this reason, measurements of the defect are performed in vivo and after calvarial reconstruction. Scalp flaps are raised below the galea, making use of the relatively avascular plane at the loose areolar layer. When incising hair-bearing scalp, an attempt is made to bevel the cut parallel to the hair shaft direction to avoid follicular damage and incisional alopecia. Scalp flaps using in radiated scalp tissue are less robust and require skin grafting the pericranium of the donor site or surgical delay.
Most defects with a diameter of 3 to 6 cm are amenable to closure with FTSFs. Donor sites are usually closed primarily. If necessary, once the scalp incisions are made, a few maneuvers can be used to facilitate mobilization of the flap and decrease wound closure tension: (1) wide undermining of the surrounding tissues, (2) galeal scoring, and (3) back cutting of the flap, in that order. Frequently, significant undermining is required to close the donor site. Galeal scoring is performed perpendicular to the main advancement axis of the scalp flap. Incisions are made 1 to 2 cm apart in the galea, trying to avoid injury to the vessels lying just superficial to it. One can expect 1 to 2 mm of release for each galeal incision score line, sometimes adding up to 1 cm or more (Figure 31.7). Finally, sometimes a substantial back-cut is required to provide the necessary flap mobilization.3,5 Two flaps can also be used, skin grafting the donor site, if necessary (Figure 31.8).
For defects ranging from 6 to 9 cm in diameter, one large scalp flap based on a major pedicle can be used, but the secondary defect will likely require skin grafting (Figure 31.9). For this reason, it is important to preserve the pericranium in the donor defect. A “bucket-handle” flap, based on both superficial temporal vessels, can be used to resurface anterior scalp defects, and the requisite skin graft can be placed in the vertex, which can be camouflaged by hair. Alternatively, multiple scalp flaps are used, frequently in the shape of a pinwheel. Another variant of using multiple scalp flaps involves elevation of the remaining scalp, with one posterior flap based on the occipital or posterior auricular vessels and two anterior flaps based on the superficial temporal vessels. The posterior flap is based on the side opposite to the defect and recruits excess tissue found in the nape of the neck region. These flaps are widely undermined and then interdigitated, covering the defect but often leaving an anterior secondary defect that requires grafting when the primary defect is posterior. Conversely, when the defect is anterior, one of the anterior flaps is used to cover the primary defect. The other flaps are pieced together to close any secondary defects. Skin grafts are used as needed.6
Whichever flap design is chosen, the main reconstructive principles are as follows: (1) mobilize as much scalp tissue as available to cover the primary defect (flap + wide undermining) and to minimize the size of the secondary defect (wide undermining) and (2) plan the location of the secondary defect so as to maximize cosmetic and functional results. Patients tolerate surprisingly well a favorably located area of alopecia (ideally occipital) that can be camouflaged with their own hair.
Tissue Expansion. Tissue expansion represents an invaluable asset in scalp reconstruction, allowing the replacement of like with like (Chapter 10). In an optimal situation, time is available for scalp expansion. Most commonly, however, patients present to the surgeon with an open wound already present or with one that is covered in a temporary or suboptimal fashion, limiting the practicality of scalp expansion.
A major advantage of scalp expansion is the potential to use expanded hair-bearing tissue, therefore, avoiding alopecia or treating existing alopecia. In addition, when given enough time, scalp tissue can be expanded to a significant degree. As much as 50% of the scalp can be reconstructed by expanding the remaining scalp. Unfortunately, the use of tissue expanders has many limitations. The wound needs to be free of infection. The tissues to be expanded should be healthy and well vascularized. Expanding previously radiated tissue is not advisable. Alternatively, adjacent non-radiated areas can be expanded and subsequently used to cover radiated wounds and/or replace radiated tissues. The patient needs to be well informed and preoperatively screened in regard to social support, medical compliance, status of the underlying disease and its treatment course, and the will to endure the lengthy expansion period, the consequent physical deformity, and the multi-stage reconstruction. Expander complication rates may be as high as 25% and include infection, exposure, extrusion, and device failure.
Scalp tissue expanders are placed in the subgaleal plane. In general, one or more expanders are inserted through remote incisions, oriented radially to the defect. It is critical to anticipate where the future flap incisions will be made prior to expander insertion. Overexpansion by up to 50% of the estimated amount is necessary, because the flaps do not yield the theoretical tissue gain. Significant back-cuts are usually required to advance the expanded tissue and are well tolerated due to the vascular delay effect from expansion. The periprosthetic capsule is incised to allow maximal flap movement (Figure 31.10). See Chapter 10 for further information on tissue expansion.
FIGURE 31.6. Skin graft reconstruction of anterior scalp and forehead. A. Anterior scalp and forehead defect following re-resection of persistent low-grade sarcoma. B. Re-resection defect with inner table burred down. C. Full-thickness skin graft placed. D. Full take of graft. E. Hairstyle change covering graft.
Regional Flaps. Some posterior scalp defects can be reconstructed with regional pedicled flaps such as the trapezius musculocutaneous flap and the latissimus dorsi musculocutaneous flap.7 Preliminary delay of these flaps should be considered to enhance vascularity and maximize flap success.
Free Tissue Transfer. Scalp defects greater than 9 cm in diameter are usually not amenable to closure with pedicled flaps and require free tissue transfer. Although tissue expansion might still be applicable, free tissue transfer offers a one-step solution for resurfacing large scalp defects with good results, especially in patients with preexisting alopecia.3,8 The superficial temporal vessels are frequently available as recipient vessels, although the vein is occasionally inadequate or absent, in which case interpositional vein grafting to the neck may be necessary. They are easily found through a preauricular incision with elevation of the skin flap to the location of the palpable pulse of the artery. The depth of these vessels decreases as the incision advances superiorly, such that they are quite superficial in the temporal region and lie within the parenchyma of the parotid gland more inferiorly. Occasionally, the occipital vessels can be used as recipients.
FIGURE 31.7. Large, wide-based full-thickness scalp rotation flap is raised to cover exposed skull in occiput. Multiple galeal scorings are evident, running parallel to the axial blood supply.
Available options include musculocutaneous and fasciocutaneous flaps, as well as muscle and omental flaps in combination with non-meshed split-thickness skin grafts. Muscle flaps show significant atrophy with time. The latissimus dorsi and the rectus abdominis muscle flaps are traditional favorites for coverage of large defects of the scalp. They have consistent vascular pedicles and a large muscle mass (Figure 31.11). A large omental flap can also be used, but it often becomes thin over time and may not be suitable for long-term durable coverage. The serratus anterior muscle free flap has a long pedicle that can reach the neck vessels without the need for vein grafts and is a good choice for smaller defects. Fasciocutaneous flaps, such as the anterolateral thigh, have recently grown in popularity. Primary closure of the thigh donor site is usually not possible when the harvested skin paddle is wider than 8 cm, requiring a skin graft from the contralateral thigh.
A few unique features of the forehead should be considered when planning reconstruction. The forehead comprises the upper part of the face. It forms a single, relatively large, and conspicuous facial aesthetic unit, whose limits may depend on the presence and location of the hairline. The eyebrows are adjacent to its lower border. Their position and function are intrinsically related to the soft tissues of the forehead.
Wide undermining and primary closure of defects is facilitated by the substantial soft tissue laxity usually found in the forehead. Defects up to 2.5 cm may be closed primarily. Shape and orientation of the wound are very important, as the main vector of tissue mobilization should be horizontal, to avoid disruption of the eyebrows. Significant wound tension is well tolerated and subsides with time. Different from the scalp, the forehead heals secondarily very well and surprisingly good results can be expected, as observed in forehead flap donor sites in older patients (Figure 31.12).
When contemplating the use of flaps or skin grafting for forehead reconstruction, the surgeon adheres to aesthetic unit principles. Reconstruction of wounds comprising more than 50% of the forehead often yields a superior aesthetic result by replacing the entire unit, even if it means excising healthy tissue. Alternatively, the forehead is amenable to tissue expansion. Incisions placed around the aesthetic unit (at the hairline and just above the brows) are the least conspicuous. Older patients may provide the surgeon with plenty of skin rhytids to help camouflage scars in other locations. It is preferable to lower the anterior hairline rather than leave mid-forehead incisions.
Free tissue transfer may be required for coverage of total or near total forehead defects, especially in the presence of exposed bone without pericranium or after local radiation therapy.
Cranioplasty or other forms of calvarial reconstruction are considered to be among the oldest recorded surgical procedures in history, dating back to 3000 BC.9 The Incas filled the trephination defects created (to allow evil humors/spirits to egress from the body) with gold and silver. Callus formation noted in ancient skulls indicates that some patients survived. Today, the plastic surgeon is often consulted for calvarial defects resulting from vascular accident, trauma, infection, and post-oncologic resection. The current indications for cranioplasty include the following: restoration of the aesthetic contour of the calvarium, protection of the underlying brain, and to provide treatment for “syndrome of the trephined” (characterized by dizziness and fatigue after craniectomy and thought to be related to intracranial transmission of atmospheric pressures altering cerebral circulation). Prior to reconstruction, the surgeon must take into account patient stability and systemic complicating factors, and must determine if adequate soft tissue exists for coverage.
FIGURE 31.8. A. Exposed cranium in a vertex defect following re-excision for recurrent skin cancer. Opposing scalp flaps are designed. B. Scalp flaps are rotated, advanced, and inset. Split-thickness graft was needed to cover one of the secondary defects.
Loss of bone flaps from infection after craniotomy has a reported incidence of 2%. Retrospective studies have suggested delaying calvarial reconstruction for at least 90 days after the infection to reduce the potential for recurrence.10 Other authors advocate waiting up to 1 year. A recent meta-analysis suggested comparable infection rates for early reconstruction (<90 days) and late reconstructions, with a mean re-infection rate of 7.9%.11
FIGURE 31.9. A. Large anterior scalp defect with exposed cranium following resection of neglected skin cancer. B. Posteriorly based scalp flap rotated and advanced, leaving large secondary defect, which required skin grafting. C. Inset of large rotational scalp flap demonstrating dog-ear at base, which was not removed at this setting, as it frequently resolves without surgery, and resection may limit vascularity of the flap.
The ideal calvarial replacement after craniotomy is the original “bone flap.” At the time of craniotomy, bone flaps may be preserved for delayed replacement by banking the bone flap either subcutaneously (typically in the abdomen) or in a deep freezer (recommended temperatures ranging from −30°C to −80°C).12 After resolution of cerebral edema or infection, the bone flap is then replaced. The bone flap then functions as a conduit for “creeping substitution,” where it serves as a scaffold for ingrowth of new bone from the edges of the defect. Proponents of subcutaneous banking of the bone flap suggest a lower rate of resorption and subsequent secondary surgery in comparison to the deep freezing method.13 Subcutaneous placement, however, may require fracturing of the bone flap and adds the morbidity of an abdominal wound and scar.
FIGURE 31.10. Tissue expansion and scalp flap reconstruction for unstable skin graft on posterior skull. A. Unstable skin graft following dermatofibrosarcoma protuberans excision and radiation therapy. B. Placement of large crescent tissue expander at anticipated donor region. C. Rotation of expanded scalp allowed coverage of exposed skull. D. Postoperative result showing stable wound but persistent alopecia from irradiation.
More often, the original bone flap is not available when the plastic surgeon is consulted, and the options for reconstruction then include autologous and alloplastic methods of calvarial reconstruction:
Autogenous Bone Grafting
Autologous reconstruction is considered the gold standard particularly in the setting of bone flap loss after infection. Bone grafts replace “like with like,” are thought to have a lower incidence of infection, and allow calvarial growth in the pediatric population. Donor sites typically include the calvarium, rib, and iliac crest, all of which may be split to increase the surface area and reduce donor site defects.
Split calvarial bone is the optimal autologous donor graft, reportedly offering less resorption (thought to be related to a high density of cortical bone and its intra-membranous embryologic origin). Some studies, however, still suggest a 25% rate of resorption at 5 years, which may lead to the necessity for secondary cranioplasty.10 Since the donor site is located in the same operative field as the defect, additional scarring is avoided. Cranial bone is typically harvested from the parietal bone, splitting the bone using a sagittal saw and/or an osteotome in the diploic space. The remaining posterior table is thought to maintain 67% of its breaking strength after splitting, increasing the risk of skull fracture in this area.10 In children less than 4 years of age, the bone may be too thin to split even if a full-thickness bone flap is removed and splitting is attempted on the back table. Harvesting full-thickness grafts and splitting them ex vivo may help facilitate the process in older children or when larger defects are grafted. Using trans-illumination techniques, thicker bone segments with a well-defined diploic space can be identified and split using a sagittal saw or scalpel. The posterior table is placed on the graft site and the anterior table is returned to the donor. Additionally, full-thickness defects in neonates may heal spontaneously, likely an effect of the growing brain (dura) on the induction of osteogenesis.
Particulate bone graft harvested from the inner or outer cortex may also be used to repair critical size defects, with some authors reporting up to a 97.8% success.14 These grafts are easily harvested using a Hudson brace and flat craniotomy bit and can be stabilized using fibrin glue. A disadvantage is the irregular, “bumpy” contour, making these grafts more useful for filling donor site defects in less visible areas of the calvarium. Split rib and iliac crest grafts are considered secondary options, given the necessity of an additional scar, donor site pain, and the potential for donor site complications (such as pneumothorax and painful neuroma, respectively). Rarely, free vascularized bone may be transferred along with soft tissue coverage for calvarial defects, such as a free latissimus/serratus with rib, though this is typically in the setting of large radiation defects.15
FIGURE 31.11. Free latissimus dorsi flap reconstruction of scalp defect. A. Large scalp defect following angiosarcoma resection, radiation therapy, and unsuccessful skin graft placement. B. Latissimus dorsi free flap placed on defect with anastomoses to superficial temporal vessels. C. Postoperative result with satisfactory contour.
Alloplastic materials offer a potentially unlimited off-the-shelf resource without donor site morbidity. The ideal implant material would be biologically inert, osteoconductive, and biomechanically compatible. Numerous materials have been described historically, with currently titanium, polymethylmethacrylate (PMMA), and hydroxyapatite (HA) being most frequently used today (Chapter 7).
Titanium mesh offers a corrosive resistant, biocompatible, strong material that produces minimal encapsulation. Titanium implants can be obtained pre-formed (CAD/CAM) or can be contoured in the operating room from a straight piece of mesh. A review of the literature reports infection rates from 0% to 4.5%.16 Critics of titanium mesh cite the conduction of hot and cold temperature as uncomfortable for the patient postoperatively and also the difficulty of obtaining high-quality imaging post-cranioplasty due to artifact scatter.
PMMA is the most widely used alloplastic material in cranioplasty. PMMA does not integrate, which helps facilitate removal in revision surgery when compared with other alloplastic implants.17Similar to titanium mesh, it can be pre-formed using CAD/CAM to help achieve a more accurate reconstruction. One study reported a 9.6% infection rate in a series of 31 patients, requiring removal of the implant and secondary cranioplasty.18 When the reagents are mixed and sculpted intraoperatively, there is a risk of thermal damage to the underlying brain from an exothermic reaction, with temperatures potentially reaching as high as 100°C. Actively cooling the surrounding tissues with water may decrease thermal conduction during the maturation process.
FIGURE 31.12. Reconstructive choices for forehead defects with local flaps. Solid lines, “H” flap advancement; dashed lines, and rotation advancement.
HA is a calcium phosphate salt putty that is easily moldable, but may be prone to fracture and/or fragmentation. HA is the only alloplastic material to offer the potential for osteoconduction and integration with the native bone.19 In large defects, it is often combined with titanium mesh since it cannot tolerate axial loading stresses alone. High rates of complications have been reported in the setting of large-scale defects (>25 cm2) or areas of previous irradiation. An infection rate as high as 40% can be seen particularly if the material communicates with the frontal sinus.19
Overall complication rates are relatively high in cranioplasty procedures, with a typical range of 20% to 30%.20 Contour deformities and infection being the most common, followed by exposure, hematoma, and seroma. Higher rates of complications are reported in the post-tumor ablation group, likely a result of adjunctive chemotherapy and radiation. Contour irregularities from autologous bone graft are reported in high frequencies due to partial resorption of the graft over time.
Some authors advocate the following tips to decrease potential complications10: (1) obliterate dead space, if any, between the cranial reconstruction and underlying dura, (2) obliterate sinuses (bone, pericranial flap), and (3) eliminate gaps if bone graft is used.
Staphylococcus aureus is the most common organism isolated in cranioplasty infections, which may occur up to 6 or 7 months after surgery.21 One study demonstrated a 5% overall incidence of infection of 5%, which rose to 14% after a documented previous infection. Exposure of the graft or implant to a sinus significantly increases the incidence of postoperative infection, illustrating the importance of sinus ablation in these defects. Treatment of infections should include broad-spectrum antibiotics, cultures, operative exploration, and, most importantly, debridement. Reports of salvage have been described combining debridement with continuous irrigation/drainage systems.22
Three-dimensional computed tomography imaging has been instrumental in performing accurate reconstruction of calvarial defects based on mirror imaging of the unaffected side. Acrylic models can now be affordably casted for use in preoperative planning and conforming of alloplastic material, reducing intraoperative time. Some companies offer custom-designed alloplastic implants without the need for constructing an acrylic model. Medical models may also allow for the creation of templates for patching together autologous bone grafts.23
While calvarial reconstruction using modeling has been found to be accurate within 2% of the normal, the surgeon must be aware of the soft tissue limitations that exist from skin contracture/debridement from the previous surgeries preventing a fully symmetric anatomic reconstruction.
1. Zide BM, Jelks GW. Surgical Anatomy of the Orbit. New York, NY: Raven Press; 1985:13-20.
2. McCarthy JG, Zide BM. The spectrum of calvarial bone grafting: introduction of the vascularized calvarial bone flap. Plast Reconstr Surg. 1984;74(1): 10-18.
3. Hoffman JF. Management of scalp defects. Otolaryngol Clin North Am. 2001;34(3):571-582.
4. Horowitz JH, Persing JA, Nichter LS, Morgan RF, Edgerton MT. Galeal-pericranial flaps in head and neck reconstruction. Anatomy and application. Am J Surg. 1984;148(4):489-497.
5. Worthen EF. Transposition and rotation scalp flaps and rotation forehead flap. In: Strauch B, Vasconez LO, Hall-Findlay, EJ, eds. Grabb’s Encyclopedia of Flaps. Boston, MA: Little Brown and Company; 1990:5-10.
6. Ortico chea M. New three-flap reconstruction technique. Br J Plast Surg. 1971;24(2):184-188.
7. Mustoe TA, Corral CJ. Soft tissue reconstructive choices for craniofacial reconstruction. Clin Plast Surg. 1995;22(3):543-554.
8. Hussussian CJ, Reece GP, W. Microsurgical scalp reconstruction in the patient with cancer. Plast Reconstr Surg. 2002;109(6):1828-1834.
9. Marbacher S, Andres RH, Fathi AR, Fandino J. Primary reconstruction of open depressed skull fractures with titanium mesh. J Craniofac Surg. 2008;19(2):490-495.
10. Zins JE, Langevin CJ, Nasir S. Controversies in skull reconstruction. J Craniofac Surg. 2010;21(6):1755-1760.
11. Yadla S, Campbell PG, Chitale R, Maltenfort MG, Jabbour P, Sharan AD. Effect of early surgery, material, and method of flap preservation on cranioplasty infections: a systematic review. Neurosurgery. 2011;68: 1124-1129.
12. Grossman N, Shemesh-Jan HS, Merkin V, Gideon M, Cohen A. Deep-freeze preservation of cranial bones for future cranioplasty: nine years of experience in Soroka University Medical Center. Cell Tissue Bank. 2007;8(3): 243-246.
13. Movassaghi K, Ver Halen J, Ganchi P, Amin-Hanjani S, Mesa J, Yaremchuk MJ. Cranioplasty with subcutaneously preserved autologous bone grafts. Plast Reconstr Surg. 2006;117(1):202-206.
14. Rogers GF,Greene AK, Mulliken JB, Proctor MR, Ridgway EB. Exchange cranioplasty using autologous calvarial particulate bone graft effectively repairs large cranial defects. Plast Reconstr Surg. 2011;127(4): 1631-1642.
15. Afifi AM, Gordon CR, Pryor LS, Sweeney W, Papay FA, Zins JE. Calcium phosphate cements in skull reconstruction: a meta-analysis. Plast Reconstr Surg. 2010;126(4):1300-1309.
16. Cabraja M, Klein M, Lehmann TN. Long-term results following titanium cranioplasty of large skull defects. Neurosurg Focus. 2009;26(6):E10.
17. Marchac D, Greensmith A. Long-term experience with methylmethacrylate cranioplasty in craniofacial surgery. J Plast Reconstr Aesthet Surg. 2008;61(7):744-752; discussion 753.
18. Goh, RC, Chang CN, Lin CL, Lo LJ. Customised fabricated implants after previous failed cranioplasty. J Plast Reconstr Aesthet Surg. 2010;63(9): 1479-1484.
19. Cheng YK, Weng HH, Yang JT, Lee MH, Wang TC, Chang CN. Factors affecting graft infection after cranioplasty. J Clin Neurosci. 2008;15(10):1115-1119.
20. Moreira-Gonzalez A, Jackson IT, Miyawaki T, Barakat K, DiNick V. Clinical outcome in cranioplasty: critical review in long-term follow-up. J Craniofac Surg. 2003;14(2):144-153.
21. Kumar AR, Bradley JP, Harshbarger R, et al. Warfare-related craniectomy defect reconstruction: early success using custom alloplast implants. Plast Reconstr Surg. 2011;127(3):1279-1287.
22. Baumeister S, Peek A, Friedman A, Levin LS, Marcus JR. Management of postneurosurgical bone flap loss caused by infection. Plast Reconstr Surg. 2008;122(6):195e-208e.
23. Molnar JA, DeFranzo AJ, Marks MW. Single-stage approach to skin grafting the exposed skull. Plast Reconstr Surg. 2000;105(1):174-177.