Skull reconstruction is most successful when the patient’s own bone is used, but alloplastic materials can work as well
Deformities of the bones of the skull are usually the result of trauma or tumor resection. Patients present with complaints of asymmetry or gross deformity, as well as concerns about large areas of the brain being unprotected.
In recent years, reconstructive options have grown as newer techniques have emerged and newer materials have been developed. The overriding goal, however, is to recontour the cranium to appear more natural while providing protection for the underlying brain and meninges.
All bones arise from mesenchyma. The bones of the skull develop by a process called intramembranous ossification, in which mesenchymal models of the bones appear during the embryo-nic period and proceed to ossify in the fetal period.
This differs from the mechanism for the long bones of the extremities, which develop by endochondral ossification. Here, cartilage models of bones develop from the mesenchyma and are subsequently replaced by bone. In either situation, the histologic appearance and properties of the bone are similar.
The bones of the skull constitute seven eighths of the head in the neonate, but only two thirds in the adult; the rest of the head is composed of the facial skeleton. The bones are separated by sutures at the edges and fontanelles at the corners. The sutures allow the bones to overlap and contort the head during delivery and are sites of growth during the first years of life.
Growth of the skull is greatest during the first 2 years of life. Its size increases more slowly until early adulthood. Most of the sutures fuse when the individual is between 30 and 40 years of age.
Skull deformation in infancy may be the result of positioning, in which all the sutures are patent, yet continuous pressure on one aspect of the head produces flattening due to the malleability of the developing bones. Treatment modalities include changes in position and molding helmets.
Less commonly, the deformation may be the result of premature closure of one or more of the sutures. This is termed craniosynostosis and produces characteristic changes in the shape of the head, depending on which sutures are involved. The deformation does not improve with time and requires operative release of the fused suture and reconstruction of the skull (Figure 1, page 26).
At birth, the bones of the skull are unilaminar, possessing no space between the cortices. In the adult, there are generally three layers of bone. Sandwiched between internal and external layers of cortical bone is a middle layer of porous, cancellous bone. Both the outer and inner layers are covered by periosteum. As people age, the bones become lighter and thinner and the blood cells and fat of the diploic space are replaced by a gelatinous material.
The skull lies beneath several layers of scalp. Beneath the skin lies a layer of subcutaneous fat; a layer of epicranial aponeurosis that covers the bones and serves as a site of attachment for the muscles of the forehead, occiput, and temporal sides; a loose areolar tissue plane; and a layer of periosteum that is easily stripped from the bones (unless it is continuous with the intervening sutures).
The scalp is immensely vascular and, as a result, resists infection very well. It does not provide much elasticity and, therefore, it lacks the ability to stretch and cover large defects. Some movement occurs within the loose areolar plane. Evaluation of any patient with a defect in the skull must include the overlying scalp, because closure of the scalp following cranioplasty is imperative.
For many tumors of the central nervous system, neurosurgeons need to remove large areas of bone to gain access to the lesion. Removed bone is wrapped in moist saline gauze during the procedure and replaced at the end of the case as a graft.
If all goes well, the graft will heal to the surrounding bone and continue to provide viable protection to the underlying brain and meninges. Sometimes, the bone does not survive or becomes infected and must be removed. In such cases, later reconstruction, once the infection has resolved, is important for protection.
Bone may also be “banked” in subcutaneous tissue (such as the abdomen) after craniotomy if the surgeon thinks that immediate replacement would be prone to infection or failure. Later replacement may be attempted when the circumstances are more favorable. This technique has provided reasonable cranial reconstruction with acceptable rates of infection and of additional revision.1
Delayed replacement is especially useful after surgical decompression following trauma, as well as after removal of an intracranial tumor. The technique usually provides a graft with the precise dimensions needed to fit the defect and acts as a scaffold for remodeling. Disadvantages include discomfort at the banking site, infection, and the potential for resorption of the graft prior to use.
Options for skull reconstruction in the absence of banked bone are numerous. They include autogenous tissue in the form of bone harvested from the surrounding calvarium or more remote sites, or from alloplastic material.
The choice of the substance to use depends on a variety of factors. Patient age is important, because the very young will continue to grow and may outgrow any alloplastic material used. Older patients may have poor bone stock and may require alternative material with which to restore the missing bone.
The use of the patient’s own bone is a well-documented means of calvarial reconstruction. Donor sites for bone appropriate to reconstruct the calvarium include the neighboring bones of the skull, the ribs, and the iliac crest. These sites share the bone strength and size necessary in many reconstructive cases.
Adjacent skull bone is often the most reasonable option, because a separate donor wound is not usually required. Its use is only possible in adult patients with adequate bone stock and diploic space.
For smaller defects, the desired placement size is outlined on a remote area of the skull in an attempt to avoid areas adjacent to vital structures, such as the midline sagittal sinus or the dominant motor cortex beneath the parietal bone (on the left side in right-handed individuals). A burr is used to retract the outlined piece through the diploic space down to the level of the inner table.
The surgeon will know which level has been reached by the amount of bleeding from the edges. The outer and inner cortices bleed relatively little, compared with the middle diploic space. Once a circumferential outline has been created, a slope is added around the edge to permit an osteotome to slide beneath the cells of the diploic space and outer table and over the inner table to remove the donor bone graft carefully.
To minimize violation of the intracranial space, the osteotome should pass parallel to the inner table. Violation of the inner table often requires wider exposure around the site of entry and confirmation that there is no epidural bleeding or dural injury.
Because larger bone grafts tend to fracture during elevation, a full-thickness harvest may be preferable in such cases. With the help of a neurosurgical colleague, an adequate piece of full-thickness calvarium is removed and split on a back table using either hand osteotomes or a reciprocating saw.
The two pieces may then be replaced as individual grafts—one for the donor defect and the other for the recipient site. Using this technique, inadvertent injuries to the underlying dura and brain are minimized, because the structures are exposed and any bleeding or violation of the dura can be immediately recognized and repaired.
The other sites of donor bone grafts include the ribs and the iliac crest. The ribs are numerous and may be harvested with minimal adverse sequelae (Figure 2). In general, large numbers of contiguous ribs should not be used so that the strength and integrity of the rib cage are preserved. Alternate ribs are a better option when a large amount of bone is required.
Each length of rib can provide two grafts, because the rib may be split without the need for replacement at the donor site. Rib bone is not as strong as calvarial bone, but it provides adequate strength as a reconstructive option. The outer table of the iliac crest may similarly be used as a donor site.
Autogenous bone is well tolerated and is associated with a relatively low rate of infection. It is a preferable reconstructive option in patients with prior infection or the potential for infection, to avoid placement of a foreign body.
In younger patients, autogenous bone grows with the host and, once healed, has the ability to resist infection. Shortcomings of any autogenous reconstruction include possible morbidity at the donor site, the limited quantity available, and the potential for later resorption of the graft.
Distraction osteogenesis is a newer reconstructive option that may have a role in the cranial vault. The technique involves separating one portion of bone from another and attaching a device across the gap that slowly creates space between the edges. Like a fracture, the space is surrounded by a healing callus that provides a milieu of growth factors and cells necessary for healing.
As the space between the edges is increased or distracted, the size of the callus stretches and new bone is laid down in the gap. With distraction osteogenesis, no donor site is required and the potential for adverse sequelae is diminished. The technique is useful for large defects that would otherwise heal with a fibrous union. It has been used in the maxillofacial skeleton in adults—in both healthy bone and irradiated tissue2—and in younger patients (including neonates) for a variety of congenital anomalies.3 Its use in the bones of the skull is still evolving.4
In situations in which autogenous tissue is lacking or unsuitable, alloplastic compounds may be used. Numerous alloplastic materials have been used for skull reconstruction. Historically, the inert metals, notably titanium, have been used successfully to reconstruct the cranial vault.5 Titanium, although expensive and difficult to contour in the operating room (OR), is the most biologically inert of the metals and is radiolucent.
The simplest reconstruction is wire mesh affixed to the periphery of the defect with screws (Figure 3, page 26). Ceramics have also been used in cranioplasty because of their strength and biocompatibility, but, unlike metals, they are prone to shattering when force is applied.6
Poly(methyl methacrylate) (PMMA) was introduced in the 1940s and is now one of the more common materials used for alloplastic calvarial-vault reconstruction.7 Through its use over the past several decades, PMMA has proven itself a stable, durable material in cranioplasty.
PMMA may be used in either of two ways. First, it may be placed within a cranial defect and cured intraoperatively. The reaction is exothermic, so continuous cooling of the surface is important to minimize injury to the surrounding tissues.
Alternatively, an implant may be fabricated preoperatively with the use of computerized systems to enhance the precision of the graft and minimize time spent in the OR. A preoperatively fabricated implant is useful in larger defects for which intraoperative contouring may be time consuming.
The manufacturing process entails obtaining preoperative, thin-slice computed tomography (CT) data reformatted to create a 3D image and, more importantly, a 3D model of the surrounding skull, with the implant fitting into the defect precisely (Figure 4). The model is created by polymerizing liquid resin by means of a laser, using the CT data.8 The completed model can be used to construct an implant fabricated from hydroxyapatite, titanium, or PMMA.
Preoperative preparation requires only that the implant be sterilized the day before surgery and affixed into place. Preoperative fabrication of the implant also avoids the potentially harmful exothermic reaction in the OR and minimizes operative time. As the implant is set into place, it may be modified using standard burrs to improve its fit in the calvarial defect (Figure 4).
Preoperative fabrication of an implant does carry some tolerable shortcomings. A precise CT scan, following a specific protocol of the implant manufacturer, is required preoperatively. Manufacturing an implant takes time, so custom implants cannot be acquired in acute settings. In addition, intraoperative modification of the implant may still be required.
Recently, scientists have sought to create off-the-shelf substitutes that more closely simulate the histologic characteristics of native bone. Ideally, a more porous material would serve as a more natural template or conduit for new growth from the bony margins. In time, the material would be completely incorporated by the neighboring tissue, thus becoming more resistant to infection and having less chance of late failure.
Hydroxyapatite is a relatively new osteoconductive material used in cranioplasty. It is one of the natural mineral constituents of human bone, and it can be processed from sea coral for use in cranioplasty. It is prepared as porous blocks, granules, and cement.
The porous blocks may be fabricated to fit a given defect, but they are brittle and may be difficult to use. Similarly, the granules may be prone to migration out of the defect after placement.9 Hydroxy-apatite cement, however, molds easily, cures quickly, and has been shown to produce favorable results in skull reconstruction10–12 (Figure 5).
Hydroxyapatite may be preferable to PMMA in calvarial reconstruction because it has been demonstrated to produce partial bone replacement and vascular ingrowth.13 Complications associated with hydroxyapatite include infection, exposure, and extrusion; and complication rates have been reported to be as high as 33%.14
The alloplastic materials used in calvarial-vault reconstruction, including PMMA, hydroxyapatite, and titanium, are all valuable, but do carry higher risks of infection, deformity as a result of continued growth, and increased expense. In the future, the ideal material for calvarial-vault reconstruction will be relatively inexpensive and will mimic native bone so that neighboring tissue grows into the implant, eventually replacing it.
Stem-cell research is exploring ways of applying pluripotent cells to a critical-sized defect and allowing the body’s host growth factors to influence the stem cells to form osteoblasts, which would then lay down healthy new bone. This technology is still evolving, but the options available meanwhile are safe, with small complication rates, if the proper technique is chosen for the patient.
Peter J. Taub, MD, FACS, FAAP, is a full-time faculty member of the Division of Plastic and Reconstructive Surgery at the Mount Sinai Medical Center in New York City . He is board certified in both general and plastic and reconstructive surgery, and he is fellowship trained in craniofacial surgery. He is also an assistant professor of surgery and pediatrics in the School of Medicine and serves as the associate program director for the plastic and reconstructive surgery residency training program. He can be reached at (212) 241-1968 or [email protected]
Philip J. Torina, MD,has completed his 3-year prerequisite training in general surgery and is currently a resident in the Division of Plastic and Reconstructive Surgery at the Mount Sinai Medical Center.
1. 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;117:202–206.
2. Taub PJ, Bradley JP, Kawamoto HK Jr. Closure of an oronasal fistula in an irradiated palate by distraction osteogenesis. J Craniofac Surg. 2001;12: 495–499.
3. Taub PJ, Koch RM, Merer D, Geldzhaler G. Mandibular distraction in a neonate with muscular dystrophy. Ann Plast Surg. 2005;55: 519–523.
4. Cho BC, Hwang SK, Uhm KI. Distraction osteogenesis of the cranial vault for the treatment of craniofacial synostosis. J Craniofac Surg. 2004;15: 135–144.
5. Blake GB, MacFarlane MR, Hinton JW. Titanium in reconstructive surgery of the skull and face. Br J Plast Surg. 1990;43:528–535.
6. Kobayashi S, Hara H, Okudera H, Takemae T, Sugita K. Usefulness of ceramic implants in neurosurgery. Neurosurgery. 1987;21:751–755.
7. Kleinschmidt O. Plexiglas zur Deckung von Schädellücken. Chirurg. 1941;13:273.
8. Taub PJ, Rudkin GH, Clearihue WJ, Miller TA. Prefabricated alloplastic implants for cranial defects. Plast Reconstr Surg. 2003;111:1233– 1240.
9. Waite PD, Morawetz RB, Zeiger HE, Pincock JL. Reconstruction of cranial defects with porous hydroxylapatite blocks. Neurosurgery. 1989;25:214.
10. Eppley BL. Hydroxyapatite cranioplasty: 2 clinical experiences with a new quick-setting material. J Craniofac Surg. 2003;14:209.
11. Stelnicki EJ, Hoffman WY, Ousterhout DK. A method for repairing zygomatic arch fractures using a hydroxyapatite cement paste. J Craniofac Surg. 1997;8: 236.
12. Burstein FD, Williams JK, Hudgins Roger, et al. Hydroxyapatite cement in craniofacial reconstruction: Experience in 150 patients. Plast Reconstr Surg. 2006;118:484–489.
13. Costantino PD, Friedman CD, Jones K, Chow LC, Sisson GA. Experimental hydroxyapatite cement cranioplasty. Plast Reconstr Surg. 1992;90: 174.
14. 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;14:144–152.