Plastic surgery is a discipline that predominantly involves surgical problems of the integument, particularly that of the face. Plastic surgical correction of deformities of congenital, post-traumatic, or postablative origin is usually intended not only to optimize function, but also to restore form. Most procedures involve the rearrangement of local tissues or the transfer of more distant ones, such as grafts and flaps. The plastic surgeon seeks to optimize conditions for the wound healing process, with attention to technique and detail.

Wound healing is a non-specific response to injury. It involves the biological processes of inflammation, collagen metabolism, and contraction in an overlapping integrated continuum. Wound healing is temporally divided into three phases: an inflammatory phase, a fibroblastic phase, and a remodelling phase. The nature and condition of the tissues as well as the mechanism of wound closure determines the relative duration of these phases and the end result of the healing process. With this background, wound healing is classically divided into three types: primary intention, delayed primary intention, and secondary intention healing.

The coaptation of sharply incised wound edges within hours of injury is referred to as primary repair and usually results in healing by primary intention. Because little or no tissue is lost and the injury is confined to the wound edges, the healing phases are relatively well defined. In this situation, the inflammatory phase which begins at the time of injury largely subsides by 5 days. The fibroblastic phase, in which collagen synthesis by fibroblasts is a dominant process, overlaps the initial phase by 2 to 3 days and dominates from days 5 to 14. Wound strength increases rapidly with collagen content. The final phase of maturation is a protracted one. During this phase, which begins at about day 14 and lasts for up to 6 months, collagen formation is balanced by collagen degradation. The end result of the wound healing process is the bridging of the soft tissue gap with collagen and its resurfacing with epithelium.

In delayed primary intention healing the wound edges are brought together several days after wounding. This delay allows host inflammatory and immune processes to control wound contamination. Delayed primary intention healing is appropriate when initial wound cleansing and debridement is unlikely to be adequate to avoid infection if the wound were closed primarily. Delayed primary closure is effected by placing sutures between the wound edges at the initial debridement, but leaving them untied until day 5. Delayed primary closure does not delay the development of wound strength; however more scar tissue develops between the native wound edges, and soft tissue appearance and function may be compromised.

Wounds heavily contaminated by bacteria or with significant necrotic tissue may be left unsutured and allowed to heal by secondary intention. These wounds become gradually filled with capillaries and fibrous tissue (‘granulation tissue’) and the inflammatory and fibroblastic periods of wound healing are markedly prolonged. These wounds are eventually closed, to a varying extent, by the processes of contraction and epithelialization. Wound contraction is believed to be controlled by the myofibroblast, a fibroblast with smooth muscle components. Although secondary intention healing may be the optimal means to achieve wound closure, in certain instances, the end result is an increased amount of scar and a thin epithelial covering.

When tissue loss prevents direct closure and secondary intention healing is not possible or would result in significant compromise in function or appearance, skin grafts or flaps are required to close the wound.

Facial soft tissue injuries are usually self evident. Those resulting from blunt trauma should always first be considered a sign of underlying significant injury to the facial skeleton or brain. Appropriate physical examination and diagnostic imaging should be undertaken.

The repair of facial soft tissue injuries should restore both pre-injury function and appearance. To this end, it is especially important to minimize the amount of scar tissue, to restore anatomical landmarks, and to avoid contour irregularities. Wound closure is ideally performed as soon as possible after injury, thereby minimizing distortion secondary to oedema, decreasing the effects of initial contamination, and decreasing the opportunity for secondary contamination. Lacerations accompanying facial fractures should be closed after fracture fixation if performed acutely. However, if fracture therapy is to be delayed, the lacerations should be closed acutely to avoid wound and possible fracture site contamination. Clean soft tissue injuries that have been cleansed well initially can be closed up to 24 h after injury. This interval can be extended if the wound margins are excised prior to closure.

Wound cleansing and irrigation is intended to remove foreign bodies and decrease bacteria. Normal saline solution, which is readily available and causes no tissue damage, is best used for irrigation. Because small particulate matter can only be removed by mechanical disruption and not by volume dilution, irrigation is best performed with a fluid jet delivering 7 lb of pressure per square inch. This goal can be accomplished by injecting sterile saline from a hand-held 35 ml syringe through an 18 gauge needle and for larger wounds by using a jet lavage system. A bulb syringe does not generate adequate pressure to debride the wound.

Debridement is more conservative in the face than in other areas of the body: severely damaged skin often survives because of the rich facial blood supply, and anatomical landmarks are virtually impossible to reconstruct secondarily. When tissue is in relative excess, such as in the cheek or forehead, ragged lacerations may be excised and a ‘new surgical wound’ created.

Antibiotics are appropriate when a surgically clean wound cannot be obtained, when lacerations communicate with the mouth or sinuses, and when systemic or local factors known to diminish host resistance are present. The need for tetanus immunization depends on the nature of the wound and the patient's previous immunization history.

For orientation, ‘cardinal’ sutures are placed to realign anatomical landmarks, such as lid margins, eyebrows, vermilion borders, or significant wrinkles. To avoid contour deformities, wounds are closed by anatomical layers and dead spaces are obliterated. To this end, muscle, subcutaneous tissue, and the deep dermis are approximated with 4 - 0 or 5 - 0 polyglycolic acid sutures. Skin edges are then precisely aligned with slight eversion using interrupted sutures of 6 - 0 nylon or prolene. Loupe magnification is often helpful.

Facial sutures are removed between 3 and 5 days after surgery to avoid leaving suture marks. The wound may be reinforced at this time with skin tapes. The presence of suture marks is largely determined by the tightness with which sutures are tied and by the length of time they are left in.

Because the wound healing process is long, scar revision is seldom performed sooner than 6 months to 1 year after injury. Hypertrophic scars and keloids result from imbalances during the proliferative and the maturation phases of wound healing. Hypertrophic scars are raised and thickened, but remain within the borders of the original scar. They undergo some maturation and improvement with time. Hypertrophic scars often respond to pressure which causes realignment and remodelling of the collagen bundles within the scar. If they are the result of a wound that healed by secondary intention or of a laceration across a flexion crease, they may improve by excision and appropriate reclosure.

Keloids grow beyond the limits of the original scar. Intralesional injection of triamcinolone acetonide (10 mg/ml) directly into the lesion may decrease their bulk. Injections can be given every 3 to 4 weeks for up to 6 months.

Lacerations in the cheek area should always alert the surgeon to the possibility of injury to the parotid gland, Stensen's duct, or branches of the facial nerve. Prior to administering an anaesthetic, the function of the facial nerve should always be tested. Lacerations of the parotid gland need not be repaired, but are managed by layered wound closure and drainage of the area. Persistent parotid fistula is exceedingly uncommon. Management of duct lacerations are determined by their location. Lacerations within the substance of the gland are best managed by ligation, allowing reflex atrophy. Lacerations to the portion overlying the masseter muscle may be repaired over a polyethylene splint. Those distal to the masseter are managed by ligating the distal cut and then directing the proximal end into the oral cavity.

Buccal branches of the facial nerve run parallel to the course of Stensen's duct and are usually involved when the parotid duct has been lacerated. As with injuries of other facial nerve branches, the nerve should be repaired at the time of presentation using microsurgical techniques. If emergency repair cannot be performed, the nerve ends should be tagged at presentation, since the distal divided ends lose their ability to respond after 48 h, making later localization difficult.

The plastic used to laminate windscreens distributes impact forces and retains the resulting broken windshield pieces. Windscreen impact causes a mosaic of tiny lacerations and avulsion flaps that mimic the windshield fracture pattern. Effective management requires removal of all pieces and meticulous approximation of all lacerations. If allowed to heal without repair, a cobblestone appearance which is impossible to correct later will result.

Foreign bodies embedded in soft tissue wounds must be removed at the time of presentation. In most cases this can be accomplished by vigorous scrubbing with a surgical scrub brush, but dermabrasion or excision may be required. Delayed removal of traumatic tattooing is difficult, if not impossible.

Bite injuries to the face can be disfiguring because of the crush avulsion mechanisms involved. They are treated as dirty facial lacerations with vigorous irrigation and debridement. If possible, the wound is surgically recreated by removing 1 to 2 mm of the wound margins both at the skin surface and in the depths of the wound. If this would result in significant disfigurement, only clearly devitalized tissues are removed. These wounds are then closed in standard fashion. Amputated parts are cleansed, debrided, and replaced as composite grafts. If the amputated part has vessels of a sufficient calibre, microvascular replantation should be considered.

Prophylactic antibiotics are administered. Because Pasteurella is the most frequent pathogen in dog and cat bites, penicillin or its equivalent should be administered to those who are not penicillin allergic. Both Streptococcus and Staphylococcus organisms are prevalent in human bites: cephalosporin or erythromycin or penicillin and a penicillinase-resistant penicillin may be administered. Guidelines for tetanus and rabies prophylaxis should be consulted.

Facial lacerations seldom cause tissue loss. Placement of ‘cardinal’ sutures to restore landmarks will usually show that little, if any, tissue is missing. If tissue cannot be reapproximated at the time of immediate repair, flap reconstruction is rarely indicated. If the defect is significant, emergency skin graft should be considered to allow definitive reconstruction to be performed under ideal circumstances.

Patients frequently request the excision of benign lesions. Occasionally these lesions masquerade as malignant lesions and excision is required to clarify the diagnosis. If there is any question about the diagnosis, biopsy examination is necessary. These lesions are usually excised as an elliptical or lenticular shape which allows a straight line closure. The patient must realize that removal of the benign lesion will cause a scar. The orientation of the ellipse is dictated by wrinkle lines, lines of least tension on the face, contour line, and facial feature outlines (Fig. 6) 2802. If a reconstructive technique more elaborate than a primary closure is required, it may be best to leave the benign lesion alone.

Basal cell carcinoma is a skin malignancy derived from the basal epithelial cells. These slow growing lesions rarely metastasize. They grow either as flat diffuse lesions, morpheaform cancers, or as raised, well circumscribed nodules. Morpheaform lesions are best excised with intraoperative control of margins. The nodular lesions have a clear border which allows them to be excised with a reasonable expectation of removal by gross inspection. Subsequent pathological inspection of the margins of all lesions is necessary to be sure that removal is complete. Positive margins after excision do not invariably mean the lesion will recur, but repeat excision is usually recommended.

Squamous cell skin malignancies, originating from the differentiated epithelial cells, are more serious than basal cell carcinoma since metastasis is more common. Like basal cell carcinomata, squamous cell cancers are caused in part by sun exposure, and, therefore, will be found on chronically exposed skin surfaces. The face, especially the lower lip, and the back of the neck and hands are often involved. Because the histological margins may be difficult to determine by gross inspection and because invasion into the subcutaneous tissue is possible, resection should be performed with a wider margin than is required for basal cell cancers. Intraoperative examination of the resection margins by frozen section pathology is advised.

Because of the increased risk of metastasis, the draining lymph nodes should be palpated for enlargement. Clinically enlarged nodes suggest the need for a standard lymph node dissection in addition to the primary resection.

The most serious of the skin cancers, melanoma requires rapid resection and lymph node dissection depending on the depth of invasion of the primary lesion. Melanoma is derived from the melanocytes interspersed among the basal cell layer of the epithelium. The lesions occur in a superficially spreading or a nodular invasive pattern. The superficial lesion may be the predecessor of the nodular pattern.

Melanomas, like other skin cancers, occur on sun-exposed surfaces. Melanoma may also occur in mucosa, the retina, the palmar and plantar skin, and subungual areas. The prognosis for metastasis is dependent on the depth of the primary invasion. There have been two classification systems proposed: Clark's and Breslow's (Fig. 7) 2803. Clark's classification divides the primary lesion into five levels: Level I is confined to the epithelium; Level II passes through the basement membrane; Level III passes to the level of the junction of papillary and reticular dermis; Level IV passes into the reticular dermis; and Level V invades the subcutaneous tissue. Breslow's system measures the depth of penetration of the melanoma from the epithelial surface. Metastasis is rare for lesions less than 1 mm in depth. From 1.5 or 1.67 to 3.6 mm, there is a significant and increasing risk of local nodal metastasis, and therapeutic or prophylactic lymph node dissection is recommended. Deeper lesions have a high incidence of metastasis, and local nodal dissections are often futile.

Melanoma is excised with a wider margin than squamous cell cancers. A margin of 1.5 cm is recommended on the face near vital features. A margin of 3 cm is recommended elsewhere. Although many adjunctive measures have been tried, including irradiation, chemotherapy, and immunotherapy, none has been successful for melanoma. Early detection and complete initial surgical resection are the keys in caring for patients with melanoma.

Split-thickness skin grafts are thin sections of skin consisting of varying amounts of dermis and its overlying epidermis which are totally detached from the body and transplanted to a recipient site (Fig. 8) 2804. The skin graft is, therefore, totally dependent on the vascularity of the recipient site for its survival. Any tissue bed with an exposed capillary circulation will potentially support a skin graft. The skin graft must be in close contact with the recipient bed so that a plasmatic circulation (serum inbibition) can support the graft until revascularization (inosculation) of the graft can occur. These processes are called the ‘taking’ of the graft. The immobilization and intimate contact of the graft with the bed are accomplished by the use of dressings. In the extremities, a circumferential dressing is used; in the face and other areas, tie-over stent dressings are usually employed (Fig. 9) 2805. The ‘take’ of the graft is also determined by the type and amount of bacterial contamination. Pseudomonas and Streptococcus are particularly detrimental to graft ‘take’. Tissues containing less than 10&sup5; organisms per gram of tissue by quantitative culture usually allow a graft to ‘take’.

Thin grafts are more likely to take than are thick grafts because there is less tissue to be supported by imbibition and there are a larger number of blood vessel openings on their undersurface—a reflection of the increasing branching vascular pattern with ascent from the deep dermis to the surface. Split thickness grafts can be harvested with a drum dermatome, a power dermatome, or a free-hand knife. Split-thickness grafts are often cut to form a mesh to allow better conformability to the wound, to cover a larger surface area, and to allow drainage of any exudate. The open areas of meshed grafts are allowed to close by re-epithelialization and contraction (Fig. 10) 2806.

Full-thickness skin grafts are harvested to contain the entire thickness of the papillary and reticular dermis. Full thickness grafts with their greater metabolic demands require optimal conditions for graft ‘take’. They are usually used to resurface surgically created wounds, and provide reconstruction of the best quality, contour, colour match, durability, and with lack of wound contraction. Unlike split-thickness graft donor sites, which heal by re-epithelialization, full-thickness graft donor sites require surgical closure. No special instruments are required to harvest full-thickness grafts. Useful donor sites include the postauricular area, the upper eyelid, the groin, and the flank.

In addition to skin and subcutaneous tissue, composite grafts include another tissue element, most commonly cartilage or hair follicles. Because of the volume of tissue transferred, composite grafts need a well-vascularized bed with optimal conditions for graft ‘take’. A common composite graft is the free edge of the ear used to reconstruct a full-thickness defect of the alar border. Since the graft must survive to its core by imbibition and diffusion until inosculation is completed, the dimensions of this graft are limited. The graft should be designed so that no part is greater than 5 mm from any part of the nutrient bed.

Occasionally, there is an insufficient supply of skin graft donor sites, such as in a patient with a large burn, and a skin graft substitute is required for temporary or permanent wound coverage. An open wound is a portal for bacterial invasion and water vapour and protein loss: a skin graft substitute must fulfil the role of a barrier. Some skin graft substitutes have been designed to attempt not only to create a barrier, but also to produce a ‘skin’ with the supple biomechanical properties of normal skin.

Cadaver skin, which can be harvested as a split-thickness skin graft is the best temporary skin substitute. The graft will adhere to the wound and become vascularized. However, the barrier provided is only temporary: the graft will be rejected in the first 2 to 3 weeks after placement, depending on the immunological capability of the host. Cadaver allograft skin can be stored in a frozen state to be made available for wound coverage when needed. Such frozen allografts are particularly useful for the immediate closure of large excised burn wounds, but have also been used as biological dressings on venous stasis ulcers.

The barrier function of skin can also be temporarily simulated by an alloplastic sheet. Silicone sheets with or without small pores may be placed on the wound and encouraged to adhere to the wound by materials on their undersurface. A model for this type of synthetic skin is Biobrane, a porous silastic sheet with a bottom layer of fine nylon coils. Such silastic sheets, being foreign bodies, are prone to infection. The sheets can only be left in place for 3 to 6 weeks, after which they must be replaced with a permanent skin.

Various types of collagen sponge have also been used as temporary skin substitutes. Skin is a bilaminar structure of epidermis and dermis. The barrier function of skin resides primarily in the epidermis; the dermis contributes to the elasticity and durability of skin. In an effort to produce a dermal substitute that might approximate the biomechanical behaviour of a full-thickness skin graft, collagen sponges covalently linked to glycosaminoglycans have been developed. The outer surface of these dermal substitutes is covered temporarily with Silastic. After the collagen sponge has been incorporated into the wound (3 - 6 weeks) the outer Silastic layer is removed and is replaced with a thin split-thickness skin graft. Whether these materials will lead to development of a permanent dermal substitute is unknown; their use is still in an experimental phase.

Human epithelial cells can be grown in tissue culture to produce coherent sheets of cells that can be transferred as a pure epithelial autograft or allograft (Fig. 12) 2808. From a small biopsy specimen of 4 cm², up to 2 m² can be grown in 3 weeks. The rapid growth of epithelial cells in tissue culture allows such grafts to be used for the closure of large burn wounds. The grafts can be placed on the excised burn wound and will adhere and produce a permanent skin substitute. Long-term histological examination of cultured epithelial grafts as a permanent skin show the differentiation of a multilaminar epithelium with the development of a basement membrane, anchoring fibrils, and rete ridges and pegs. The differentiation of the subjacent collagen in response to the epithelial grafts may indicate that a dermis does not need to be provided by a collagen sponge.

Flaps are transferable blocks of tissue that maintain their viability based on their own circulation. They are, therefore, used when the recipient bed does not have a sufficient capillary circulation or when the bulk, colour, or contour of this specific flap tissue is desired. The flap closure of a surgically created wound allows healing by primary intention with all of its advantages.

Flap design is based on the knowledge of the vascular anatomy of the skin and on geometric principles. The skin receives its blood supply from the dermal - subdermal plexus of vessels. In most areas of the body, these vessels are supplied by vertically oriented perforating branches from large vessels running in the underlying fascia or muscle. The muscle or fascia, in turn, is supplied by one or more vascular pedicles (Fig. 13) 2809. In a few areas of the body, the dermal - subdermal plexus is supplied by vessels that course directly in the subcutaneous tissues.

Flaps can be classified on the basis of their blood supply, the type of tissue transferred, their proximity to the wound, and their geometric design.

Flaps designed without regard to the underlying vascular anatomy are termed random. These flaps have a blood supply based on the subdermal plexus of vessels. Except for the richly vascularized areas of the face and scalp, their length to width ratio should not exceed one to one. Flaps of greater length to width ratio are elevated in stages to alter their pattern of nutritional blood flow—a process termed ‘delaying the flap’.

Axial flaps are supplied by large vascular pedicles that run in the subcutaneous tissue. As long as the vascular pedicle is included in the flap they can be designed without regard to the length to width ratio.

The groin flap is an axial flap that is particularly useful for hand reconstruction. This flap is supplied by the superficial circumflex iliac artery, a branch of the common femoral or profunda femoris arteries. This artery runs laterally parallel to the inguinal ligament and will support the skin overlying this area to the midlateral line. The deltopectoral flap based on branches of the internal mammary artery and the forehead flap based on the supratrochlear vessels are other useful axial cutaneous flaps.

Most cutaneous areas of the body receive a blood supply from the underlying muscle. Knowledge of the vascular anatomy of muscles allows one to transfer flap of the muscles alone covered by a skin graft or of the overlying skin and subcutaneous tissue as well. Because the pattern of blood supply to muscles determines its usefulness as a potential flap, the vascular anatomy of the muscles of the body has been classified in terms of the number of pedicles, their size, and their point of entry (Fig. 15) 2814. This classification allows the surgeon to predict the muscle's reliability, arc of rotation, or potential for free tissue transfer. The reliable blood supply and the bulk which allows them to obliterate dead space accounts for the prominent role of muscle and myocutaneous flaps in reconstruction.

An intercommunicating system of vessels derived from perforators from the underlying muscle runs in the muscular fascia, and in certain areas of the body these fascial vessels will provide a sufficient blood supply to the overlying skin. Either the fascia alone or the fascia with its overlying skin may be transferred.

Flaps derived from tissues immediately next to the wound are termed local flaps. Obviously, a local flap is an easier tool to choose for a particular reconstruction because no new incision is added to another part of the body. However, reconstructive choices with a local flap may be limited because sufficient skin with the appropriate blood supply may not be available. Local flap reconstruction of small facial defects is common; the design of such flaps is discussed below.

When local tissue is insufficient to cover an adjacent defect, a flap must be transferred from a remote location. Traditionally, this was done in one of three ways: the flap donor site and defect recipient site were brought near to one another, as in the case of the cross leg flap; the recipient site was brought near to the recipient site, as when a hand defect is covered by a groin flap; or the flap was brought to the recipient site attached temporarily to the blood supply provided by an intermediate carrier site, such as the wrist. All of these techniques require prolonged, sometimes awkward positioning of the patient, and several operations to be successful.

The blood supply of any flap that is derived from an artery at least 1 mm in diameter may be completely detached from the donor site and transferred to a distant recipient site with microvascular anastomosis to restore its blood supply. This process of flap transfer is called free tissue transfer; the flap is called a free flap (Fig. 16) 2815. The flap may consist of any type of tissue reliably supplied by an artery and vein. Free flaps allow the reconstruction of defects where no local tissue is available for transfer and where use of a distant flap might require more than one operation. The variety of types of tissue which can be transferred, the obscurity of the donor site, and the ingenuity of the reconstruction frequently provide a result whose excellence justifies the length and risk of the operation.

In addition to categorization by blood supply design, local flaps can be categorized by their geometric design, that is, by the manner through which areas of relative skin excess are transferred to areas of dearth. The design and movement of these flaps from the donor to the recipient site is dependent on a balance of blood supply and tissue tension. Excess tension will decrease blood supply and lead to flap necrosis.

Advancement flaps depend on the laxity of their constituent skin to provide excess tissue when separated from its underlying structures. Advancement flaps are most useful for small defects when skin is in excess and possesses inherent good blood supply, such as in the ageing face.

V - Y advancement flap is a modification of the advancement principle. A small amount of length along the vertical axis is gained from movement of tissue from the horizontal axis. The V-shaped flap is advanced and its donor site is closed to make the vertical limb of the Y.

When tissues are rotated into defects they are called rotation flaps; when moved laterally into a defect, they are called transposition flaps. Many flaps combine in varying degree the principles of rotation and transposition. One of the two motions may predominate in a flap depending on the size of the primary defect and the laxity of the adjacent tissues. The secondary defect may be closed by primary suture or with a skin graft.

In principle, rotation flaps are designed so that the defect and flap lie in a semicircle. Pure rotation flaps depend on tissue laxity and a redistribution of tension to close the secondary defect. If the primary defect is sufficiently large, a back-cut is made to allow the flap also to be transposed. The secondary defect is usually best managed with a skin graft to avoid excess tension.

Transposition flaps have lateral motion as their predominant movement, even though the flaps do pivot on an axis, a rotational movement. Transposition flaps are designed as rectangles adjacent to the defect and are moved laterally to fill it. They are usually designed to allow primary closure of the secondary defect, but may require skin grafts (Fig. 19) 2818. A back-cut may be necessary to reduce tension at the base of the flap and allow its rotation into the defect. However, the back cut reduces the circulation to the flap.

A Z-plasty is two opposing transposition flaps that exchange places and thereby alter the direction and length of scars. The name is derived from the Z shape of the outline of the two flaps. Z-plasties are useful because after transposition there is a gain in length along the central axis of the original Z and a change in its direction by 90°. The increase in length is useful in the treatment of contractures and the change in direction is often useful in the revision of hypertrophic scars (Fig. 20) 2819.

The variables involved in constructing the Z are orientation, limb length, and angle size. When lengthening contractures, the vertical limb is placed in the axis of the contracture. To lengthen a specific scar, the central limb is drawn over the scar and the two limbs are drawn of the same length and at an angle of 60° to the central member. To change the direction of a scar, the proposed resultant central member is drawn across the scar and of the same length. The two limbs are then drawn from the ends of the new central member to the ends of the old central member, with some variation of angle allowed. For a given limb length, increasing the angle increases the amount of length gained. Clinical experience shows that angles of 60° are most effective. Greater angles produce flaps that are difficult to transpose, while smaller angles are less effective in gaining length and are progressively less well vascularized. For a given angle, the length of the limb determines the amount of lengthening. This length is determined by the amount of tissue available on either side.

The requirements of the reconstruction dictate the types of tissue which will be included in the flap. Most often the flap contains skin and its underlying subcutaneous tissue. This tissue may be supported by its random or axial blood supply or may be carried by the subjacent fascia or muscle. These last structures may contribute to the reconstruction with their bulk or function. Occasionally, muscle will be transferred alone to provide bulk to fill a large cavity or to be a functional muscle unit. In special circumstances, the flap may be designed to contain bone, nerves, intestine, or omentum. The number of different tissues within a flap is limited only by the requirements of the reconstruction and the ingenuity of the surgeon.

Defects in the eyelids resulting either from trauma or tumour resection should be reconstructed immediately to provide protective cover for the globe.

Loss of up to 25 per cent of the lid may be reconstructed by primary closure, which must be carefully layered to maintain lid function. The conjunctiva is closed with fine, buried, absorbable sutures. The tarsal plate is approximated separately with more slowly absorbed sutures. The orbicularis oculi is closed with fine, absorbable sutures. The grey line must be precisely lined up to avoid notching. Three 6 - 0 silk sutures at the inner and outer edges of the grey line and through the Meibomian glands in the centre anatomically align the edges. The skin is closed with 6 - 0 nylon sutures (Fig. 21) 2820. When the defect is just greater than 25 per cent of the lower lid, a lateral canthotomy may aid the primary closure. When up to half of the lower lid is lost, the tissue lateral to the eye must be elevated as a rotation flap. A curved incision rising above the lateral canthus allows the temple skin to be rotated medially. The excess conjunctiva in the lateral fornix will provide a lining for the flap after the septum orbitale and inferior lateral canthus are divided.

For defects affecting more than 50 per cent of the lower lid, the lateral incision may be continued down to the preauricular area to rotate a large cheek flap. In these cases, there will be insufficient conjunctiva for lining, and a free graft of septal cartilage with mucoperichondrium on one side provides lid support and a conjunctival replacement. Alternatively, an advancement flap from the upper lid including part of the tarsus and the conjunctiva can be brought to the lower lid, particularly for narrow horizontal defects. The outer skin surface is replaced with a full-thickness skin graft or with a pennant flap from the upper lid.

Upper lid defects of up to 25 per cent may be closed similarly to lower lid defects. For larger defects, tissue is borrowed from the lower lid in the form of a cross lid flap. The lower lid lacrimal punctum should be preserved, but the borrowed portion of lower lid can be replaced by the techniques mentioned above depending on the size of the cross-lid flap.

Reconstruction of the nose is one of the oldest plastic surgical procedures. Nasal amputations, a not uncommon form of punishment, were reconstructed in India in approximately 500 bc by Susruta. Nasal reconstruction was adopted by Western European medicine in sixteenth century Italy where a flap of skin from the upper arm as well as the ‘Indian’ forehead flap was used.

Restoration of a functional and aesthetic nose needs a careful analysis of the missing parts. The cantilevered part of the nose is a sandwich of lining, support, and cover, all of which must be considered in the reconstruction. The contours, features, and margins of the face separate the face into distinct areas which are generally best reconstructed as a whole.

Full thickness loss of the dorsal skin of the nose due to tumour resection or trauma is best restored by a local flap first and a full-thickness or composite skin graft later. The nasal skin may be thick and flaps may not transpose easily. A defect of the tip of the nose may be restored by elevating a flap of the entire nasal dorsum including a V cut into the glabellar skin. The flap is transposed to fill the defect and the glabellar V is closed a Y of a straight line. A graft from the concha of the ear can be used to replace any missing cartilage.

A side wall defect in the nose is best replaced by a nasolabial flap. The flap is based on the angular arterial blood supply and may be designed with a superior or inferior base. Nasal reconstructions will usually be based on the superior supply and can be cut with an island paddle of skin to tailor the flap inset and the donor site closure.

Subtotal or total nose reconstruction is best effected by transposition of a forehead flap. This procedure is one of the oldest in plastic surgery. A flap based on the supratrochlear vessels may be turned 180° to cover the nasal structures.

Reconstruction of all or part of the ear, due to either a congenital or a post-traumatic defect, requires a cartilage framework and a thin flap skin cover. The best source of cartilage graft is the confluence of the lower costal cartilages. Depending on the size of cartilage required, the cartilage may be harvested from the same or opposite side. A slightly exaggerated framework is carved from the rib to simulate the helical and antihelical folds. The key to the operation is the provision of a thin skin flap with enough redundancy to reveal the architecture of the underlying cartilage graft.

Small breasts may be enlarged surgically by the placement of an implant behind the mammary tissue or behind the subjacent pectoral muscle. The indications for the procedure are the cosmetic interests of the patient or marked asymmetry due to dystrophy of one breast.

The operation may be performed either under general anaesthesia or with local anaesthesia and intravenous sedation. The route of access is through an incision in the areola, just above the inframammary crease, or within the axilla. The dissection is carried down to a plane under the pectoralis major and above the pectoralis minor muscles where a pocket quite a bit larger than the proposed implant is made. The pocket used to be made above the pectoralis major muscle, but the incidence of scar capsule contraction seems to be higher in that position. Implants for breast augmentation or reconstruction are usually Silastic outer envelopes filled with either a gel of the same material or with saline.

The most frequent complication of the procedure is scar capsule contraction around the implant, causing the configuration of the breast to be rounder and firmer than desired. There may also be decreased nipple sensation, haematoma formation within the pocket, a hypertrophied scar at the incision, or asymmetry of implant position.

Overly large breasts may cause neck and back pain, grooving of the shoulders from the brassiere straps, and intertrigo. A breast reduction is performed either to alleviate these symptoms or to improve the appearance of the breasts. There are no anthropomorphic criteria for breast hypertrophy, but it is generally accepted that a breast reduction will remove in excess of 300 g per breast. There is a limit to how large a reduction may be effected without producing a flattened or square appearance of the breast. General anaesthesia is required for the procedure. Because blood loss is moderate, the patient will usually have a unit of autologous blood stored before surgery.

The pattern for breast reduction is best drawn on the patient in a standing position before surgery. Many patterns and operative designs for breast reduction have been devised. The design described here will be based on the Wise template for breast shape and on a combination of a bipedicle-augmented inferior pedicle for blood supply to the transposed nipple.

The central pedicle is de-epithelialized so that it can be buried in a subcutaneous position. The nipple, with a reduced diameter, is left as an island on the pedicle. Medial and lateral segments of breast tissue and fat are removed. A suction lipectomy may be performed to reduce axillary fat. The flaps of breast created are tailored to leave symmetrical amounts of tissue on both sides. The incisions are closed in several layers over suction drains with the nipple folded on to the centre of the ‘keyhole’.

Possible complications include haematoma, decreased nipple sensation, hypertrophied scars, asymmetry, and nipple loss due to ischaemia. Patient satisfaction with the change in breast shape and weight is usually high.

A breast reconstruction is not necessary after mastectomy since adequate external prostheses are available. However, the convenience and appearance of the reconstructed breast is superior. Recent advances in implant technology and operations for breastreconstruction promise reconstructions which more accurately simulate the opposite breast.

The operation may be performed immediately after the mastectomy under the same anaesthesia, or at any time in the future. The immediate procedure has the advantage of saving a later operation and perhaps in shoring up the psyche of the woman. The delayed operation may give the woman more time to decide which type of reconstruction she desires, if any, as well as preventing complications of the reconstruction interfering with required postoperative radiation or chemotherapy.

The simplest breast reconstruction is an implant placed in a submuscular pocket, similar to breast augmentation. In contrast to breast augmentation, the pocket created is completely beneath muscle (pectoralis major, serratus anterior, and rectus abdominis) to protect the implant from overlying flap necrosis and to minimize scar capsule contraction. This type of reconstruction suffers from its inability to simulate anything but the smallest and least ptotic breast.

The ellipse of skin resected in the mastectomy may be replaced by a transposed latissimus dorsi island myocutaneous flap with an implant placed under the latissimus dorsi and pectoralis major muscles. A larger and more ptotic breast may be created, but at the expense of a back scar. There is little functional deficit from the transposition of the muscle. The latissimus dorsi flap is particularly useful when the breast has been irradiated preoperatively.

The resected skin can also be ‘replaced’ by skin expansion before implant placement. This procedure has the advantage of providing additional skin without requiring a new donor site. At the initial operation a Silastic envelope is inserted with a subcutaneous filling portal. After surgery, saline is injected through this portal at weekly or biweekly intervals. Much like a pregnant abdomen, the overlying skin is stretched to a larger surface area. The breast is often expanded to a volume greater than that ultimately desired to allow for skin hysteresis. Some 3 to 6 months after expander placement, a permanent implant is placed in the expanded pocket with some attempt to create an inframammary crease. The breast so reconstructed is larger than that made by an implant alone, but despite some hint of ptosis, will not match an opposite pendulous breast.

If the opposite breast is large or ptotic, final symmetry is enhanced by performing a mastopexy or breast reduction on that side. This operation is performed at the time of breast reconstruction, or at a later date to maximize symmetry.

The breast is most naturally reconstructed with a reasonable match to even a large or ptotic opposite breast by a transverse island rectus abdominis myocutaneous flap. The operation is suited to the patient who does not want a permanent implant or who would not want any surgery on the opposite large or ptotic breast. The operation has the further advantage of providing a ‘tummy tuck’ at the time of breast reconstruction. However, the scale, duration, and extent of the procedure should not be underestimated.

A transverse island of skin from the lower abdomen is elevated based on one or two of the rectus abdominis muscles. The upper abdominal skin is elevated, and the flap is passed through this tunnel to the reconstruction site. Only slightly more than one half of the lower abdominal skin receives sufficient blood supply from the ipsilateral rectus abdominis muscle. The flap is turned into the site of the new breast to recreate an axillary fold as well as the pendulous form of the breast (Fig. 25) 2824.

If the patient is obese, even the ipsilateral skin may not entirely survive on one rectus abdominis muscle. In that case, two muscles may be transposed together to support the flap. However, the double muscle transfer weakens the abdominal wall. Alternatively, the flap can be transferred as a free flap based on the inferior epigastric vessels anastomosed to branches of the axillary vessels. The flap may also be transferred on one muscle with a microvascular anastomosis of the inferior epigastric vessels.

Transverse island rectus abdominis myocutaneous flap breast reconstruction is an extensive operation. The blood loss from the abdominal dissection is moderate, and an autologous unit of blood is often required. The anterior rectus abdominis fascia superior to the accurate line is partially or completely resected, which results in some laxity of the abdominal wall; this defect can be supported with a patch of synthetic mesh. Although it was initially thought that ventral hernias would not be seen, observation over time is revealing an incidence of 5 to 10 per cent.

The breast tissue may be removed with preservation of the entire breast skin and nipple in the treatment of benign breast disease. Incapacitating mastodynia is the primary indication for this procedure. The indications have been extended to removal of the breasts with premalignant disease. Although one would expect a decrease in the incidence of breast cancer in the breasts so operated, that decrease has not yet been documented without question. In addition, the presence of an implant may hamper the detection of breast cancer in the small vestige of breast tissue left after surgery.

The operation is performed in a similar manner to breast reduction. The nipple is maintained primarily by a superior pedicle and the breast is degloved through an inframammary crease incision. A total submusclular pocket is made to accept an implant.

Hypertrophy of the male breast during adolescence usually resolves spontaneously in several years. Endocrinological evaluation is unwarranted. Persistent breast enlargement in later years may be reduced by a combination of resection and suction lipectomy. Unilateral breast enlargement in older men should be biopsied because of the rare occurrence of breast cancer.

The operation can be performed under general anaesthesia or local anaesthesia and sedation. If there is true glandular hypertrophy, an areolar incision is made around the inferior portio of the areola. A portion of the glandular (firm and white) tissue is resected. Care should be taken not to resect the entire mass since the nipple without some support will indent. For the fatty breast and for the perimeter of the breast with true hypertrophy, suction lipectomy is performed to reduce the surrounding fat and to ‘feather’ the perimeter of the excision into the surrounding area. Some surgeons believe that the entire procedure can be performed with suction alone. However, since true glandular tissue resists even the scalpel, suction alone is not likely to be able to remove it. All tissue removed should be sent for pathological examination.

Large abdominal wall defects resulting from trauma, infection, extirpative surgery for cancer, resection of fistulae, and late sequelae of radiation therapy can be difficult to close. Fistulae may be present to bowel or bladder, and may need concomitant repair in combined procedures. The local tissues are often infected and friable, requiring extensive preoperative debridement. These local tissues will be inadequate for primary repair, leaving exposed viscera or a skin closure without fascial support resulting in herniation.

The ideal abdominal wall reconstruction should restore the normal layers of the wall and contain vascularized fascia, subcutaneous tissue, and skin. These are preferably supplied by a single reconstructive unit in a one-stage procedure. Knowledge of the vascular anatomy of the abdominal wall allows the design of safe elective incisions and of appropriate flaps for abdominal wall reconstruction.

Many methods are available. Zone I comprises the upper abdomen and epigastrium and is defined as the area from xiphisternal junction superiorly to the umbilicus inferiorly and to the midaxillary line laterally. Although this area is central on the body surface, it is one of the more difficult areas for which to provide flap coverage, particularly when the superior epigastric arteries have been sacrificed in the defect. Myocutaneous flaps are most easily transposed from the centre outward or from the extremities toward the centre. There are few good flaps in the centre of the body that can be transposed locally. Flaps useful in repairing Zone I defects include the rectus abdominis muscle flap, the extended deep inferior epigastric flap, the external oblique muscle flap, the lateral intercostal flap, the latissimus dorsi muscle flap, and the greater omentum.

Zone II comprises the mid-abdomen, which includes the periumbilical area from 2 cm above the umbilicus to 2 cm above the pubis and laterally to include both flanks. One of the most reliable and versatile flaps for the reconstruction of Zone II defects is the tensor fascia lata flap, which can supply fascial support as well as skin coverage. Other possible flap donor sites include the rectus abdominis flap, the rectus femoris flap, the extended deep inferior epigastric flap, the groin flap, the posterior thigh flap, and the external oblique flap.

Zone III consists of the suprapubic and groin areas. Defects in this zone present the combined reconstructive problems of potential herniation, contaminated pubic bone, and exposed femoral vessels. The flaps available for defects in Zone III include the tensor fascia lata flap, the groin flap, the gracilis muscle flap, the sartorius muscle flap, the rectus abdominis flap, and the vastus lateralis muscle flap.

The rectus abdominis muscle extends from the lower costal margin and the cartilages of the sixth, seventh, and eighth ribs to the pubic tubercle on the pubic crest. The muscles are bilaterally symmetrical around the midline. The dominant blood supply is from the deep inferior and superior epigastric systems, and each can usually support the entire muscle. The muscle can therefore be reflected superiorly onto the chest wall for reconstruction as well as inferiorly into the perineum and groin. Its motor and sensory nerves derive from multiple intercostal nerves. If the superior epigastric artery is intact, the rectus abdominis flap is the best choice for repair of epigastric defects. Based on the superior pedicle, with a point of rotation just below the xiphisternum, the flap will cover the lateral abdomen, upper flank, lower chest, and sternum. It may be especially helpful in filling defects about the xiphisternum, where pectoralis major flaps will not easily reach. The rectus abdominis flap can be used with its overlying skin for chest wall reconstruction, taking skin from the lower abdomen and transferring it to the chest wall. The rectus abdominis muscle primarily provides soft tissue coverage and is a weak support for abdominal wall integrity, even when the anterior sheath is included. It can be used in conjunction with a supporting Prolene or Marlex mesh if necessary.

The posterior rectus sheath above the arcuate line (semilunar line of Douglas) should be preserved to facilitate donor site closure and avoid production of a hernia. Below the arcuate line, if the anterior sheath is taken with the muscle flap, the defect should be supported by synthetic mesh to avoid hernia formation. Donor sites can usually be closed primarily.

The tensor fascia lata muscle originates from the anterior superior iliac spine and from the greater trochanter of the femur. The fascia lata is an extension of this small muscle and inserts into the lateral aspect of the knee, where it functions as a knee stabilizer. If one draws a line from the greater trochanter to the midlateral aspect of the knee, it should bisect the centre of the tensor fascia lata muscle as well as the flap. The anterior and posterior limits of the tensor fascia lata flap are the rectus femoris muscle and the biceps femoris muscle, respectively. The dominant blood supply is from the lateral circumflex vessels which enter the muscle approximately 10 cm below the anterior superior iliac spine and the inguinal ligament. The motor nerve of the tensor fascia lata muscle branches from the gluteal nerve, and the sensory nerve to the overlying skin derives from the lateral cutaneous nerve of the thigh and sensory branches of T12.

Wangensteen described the use of this flap as a pedicled island flap for replacement of the abdominal wall. The muscle and its fascial extension can be elevated based on the vascular pedicle with a variable size skin paddle either with a proximal subcutaneous base or as a true skin island. The tensor fascia lata flap will reach to just above the umbilicus and is the flap of choice for most defects below this level. The cephalad reach of the flap can be extended to Zone I by dissecting the lateral circumflex vascular pedicle back to the femoral vessels, ligating the side branches to the rectus femorisand vastus lateralis muscles, and tunnelling the entire island flap under the rectus femoris muscle to exit over the femoral vessels. There is little or no functional loss associated with harvest of the tensor fascia lata muscle. The donor site can be closed directly when the skin paddle is less than 8 cm in width.

In addition to its primary role in the closure of Zone II defects, the tensor fascia lata flap is also a good flap for reconstruction of pubic defects and for coverage of ischial pressure sores.

The gracilis muscle originates from the pubic tubercle on the inferior ramus of the pubis and inserts on to the medial knee. Its main function is a flexor of the knee as well as an accessory adductor of the thigh. It has a proximal blood supply from the profundus femoris artery which enters the muscle approximately 10 cm below the inguinal ligament. Occasionally, the dominant vessels will enter as low as 16 to 17 cm below the inguinal ligament, which limits the arc of rotation of this flap, making it useless for abdominal wall reconstruction. Doppler evaluation of the level of vascular ingress into the muscle is advised before operation. There is a distal minor vascular pedicle, which can be divided without vascular compromise to the distal muscle. The motor nerve to the gracilis muscle comes from the femoral nerve, and the sensory nerve to the skin paddle arises from the medial cutaneous nerve of the thigh.

The blood supply to the subcutaneous tissue and skin overlying this muscle comes from vessels wrapping around this muscle in the intermuscular septa and not from direct muscular perforators. Because of this septal blood supply, the skin paddle of this flap, particularly in the obese patient, is not as reliable as that of other myocutaneous flaps.

The gracilis muscle is generally used for primary defects of the perineum, vagina, scrotum, and penis. After pelvic exenteration, paired gracilis myocutaneous flaps can be joined in the midline to close the pelvic floor and to recreate a vaginal sheath. This flap may also be used to cover irradiated tissue and exposed vessels in the ipsilateral groin.

Microvascular surgery and microneurosurgery developed as a combination of the microsurgical techniques of ophthalmology and the macrovascular techniques derived from the work of Alexis Carrel. In plastic surgical reconstruction the microscope is used primarily for limb replantation, nerve surgery, and free tissue transfer.

Microvascular surgery, the anastomosis of 1 mm vessels, began in the laboratory in 1960. Early demonstrations that small vessels could be successfully anastomosed led very quickly to clinical revascularization and replantation. The first arm replantation was performed by Malt; the first clinically successful revascularization of a thumb was performed by Kleinert in 1963; and the first true replantation of a thumb was performed by Komatsu and Tamai in 1965. In the next decade large series of successful limb replantations were reported from around the world.

The instruments of microvascular surgery include the double-head binocular operating microscope, the microsurgical instruments of ophthalmology, and very fine nylon sutures with swaged-on needles. The development of these sutures actually enabled microvascular surgery to be initiated.

Vascular anastomosis is begun by placing the vessel ends approximate to each other in a non-traumatic clamp. Initial interrupted nylon sutures are placed 120° on the circumference of the vessels. After an additional one or two more sutures are placed in the ‘front wall’, the clamp and vessels are turned over. Since the ‘front wall’ sutures were placed relatively close together, the flaccid ‘back wall’ can be sutured without danger of suturing the lumen. Veins, because of the lower intraluminal pressure, require fewer sutures than do arteries.

Replantation of amputations has grown in the past 30 years from a laboratory experimental model to a common technique in trauma surgery. Microvascular surgery and replantation have developed together, the one stimulated by the other.

All patients who have an amputated part should be considered candidates for an attempt at replantation. Only the poor condition of the patient that would discourage a prolonged operation or the anticipation that the replanted part will function less well than a prosthesis should prevent a replantation attempt.

In general, young, healthy patients with a clean (guillotine) amputation have the best result. The older the patient, the less good function is likely to be. Major limb amputations at the upper arm, forearm, or hand should be replanted if possible because of the devastating functional loss. Smaller amputated parts frequently provide the best reconstruction for the injury when replanted. The thumb and multiple fingers should be replaced if possible. Even smaller parts of single fingers may be replanted in some patients.

The amputated part should be cooled from the time it is retrieved by the medical team. The cooling is usually effected by wrapping the part in moist gauze and placing it in a plastic bag immersed in iced saline solution. Cold ischaemia preserves a distal part for up to 24 h or more; proximal amputations that contain a large bulk of muscle can only be preserved for 6 to 8 h.

The bone is usually shortened proximally and distally to allow easy approximation of all structures. The sequence of repair is carried from deeper to more superficial structures and from larger to more microscopic structures. The bones are fixed with interosseous wires, Kirschner wires, or fixation plates. The extensor and flexor tendons are repaired. The vascular repairs may begin with the artery or vein first, although initial venous repair will minimize blood loss when the arterial repair is unclamped. The nerve repairs are performed next, followed by the soft tissue closure.

The ideal reconstruction for an amputated thumb is replantation. Even if the replanted thumb does not move well, a sensate thumb post provides good function. Very distal amputations of the thumb are replanted since thumb length is important. As the amputation approaches the eponychium, it may be difficult to find suitable veins for anastomosis. In this case the nail bed or the soft tissue can be allowed to bleed to decompress the part (Fig. 28) 2830,2831.

Amputation of multiple fingers at all levels except the most distal deserve a replantation attempt. From inspection of the parts it may be difficult to determine which will survive or which will be the most functional. Single fingers amputated distal to the proximal interphalangeal joint provide excellent function after replantation. Fingers amputated through ‘no man's land’ because of the devastating combined injury to all structures as well as the flexor tendons may be expected to require a prolonged course of physical therapy to regain motion. Single fingers replanted in ‘no man's land’ may be bypassed in future hand activity, but multiple replanted fingers in the well motivated patient usually function better than alternatives.

Amputations through the wrist or distal forearm produce some of the best functional restorations of all replantations. The forearm bones are fixed with plates. All tendons and nerves are repaired. The structures are large, and two arteries and several veins should be repaired if possible. Amputations within the carpal bones may require row carpectomy for fixation.

Success in revascularizing the amputated part depends to a large degree on the mechanism of injury. Clean, guillotine amputations should have an anastomotic patency rate in excess of 90 per cent. A crushing component to the mechanism widens the zone of injury, increasing the risk of vascular thrombosis. Avulsing injuries will produce a zone of injury particularly within the vessels which may make revascularization impossible, even with long vein grafts. The overall rate of viability should be in the 80 per cent range if the replantation team is attempting difficult as well as easy replantations.

Functional results are dependent on the mechanism of injury and the age and motivation of the patient. Young patients will regain more motion than old patients.

Following the success of replantation, several investigators determined that microsurgery could be used to transfer flaps from one area of the body to another. Such flaps were initially shown to be possible by Goldwyn and Krizek in the laboratory. In 1973, Daniel performed the first successful free flap in a patient by transferring the groin flap based on the superficial circumflex iliac vessels.

Axial arterial flaps and myocutaneous flaps are usually chosen for free tissue transfer because their arteries and veins are of sufficient calibre and provide a blood supply to a sufficiently large area to function as a reconstructive transfer.

The groin flap was the most commonly chosen flap in the early era of free tissue transfer. Maxwell reported on the use of the latissimus dorsi muscle, either alone or with an island skin paddle, and emphasized its large size, malleability, and long vascular leash as desirable characteristics. The latissimus dorsi flap remains as one of the most reliable of free tissue transfers.

Over the ensuing years, multiple flaps for microsurgical transfer have been devised by ingenious surgeons confronted by difficult reconstructive problems. The flaps whose vascular supplies had been researched during the myocutaneous flap development era were sequentially employed as free flaps. In addition, several new flaps, such as the great toe for thumb reconstruction, were devised in direct response to the microsurgical potential.

The operation of free tissue transfer requires the preparation of the flap in one area of the body and the preparation of the recipient site in another. The procedure is frequently performed by two surgical teams working simultaneously.

The donor flap is dissected out fully with the vessels skeletonized, but not ligated, so that it remains perfused until the recipient vessels are ready for anastomosis. The surgical crew working on the recipient site prepare the wound to be reconstructed. This may involve the excision of an ulcer, a tumour, or osteomyelitic bone. Suitably nearby vessels of the appropriate calibre and blood flow for the flap vessel size and perfusion requirements are then dissected out. When both operative sites are ready, the microscope is positioned, the flap is divided, and the anastomosis is performed.

The microsurgical anastomosis of free flaps is more often done in an end-to-side fashion than in the end-to-end orientation commonly employed in replantation surgery. Since the recipient, or inflow vessels, may be important or essential to maintaining the distal portion of the limb being reconstructed, the end-to-side anastomosis will allow flow to the flap and the limb. In some cases, it may be possible that the flap will ‘steal’ arterial flow from the distal limb through its end-to-side anastomosis. On the other hand, however, it has been documented in operations on limbs which are chronically ischaemic from arterial insufficiency that this flow may, in fact, revascularize the distal limb through neovascular connections to the bed in which the flap is inset.

High velocity injuries to the lower extremity frequently result in open fractures that are further complicated by soft tissue loss and perhaps by distal devascularization. Aggressive early debridement and internal fracture fixation necessary for limb salvage can only be performed when soft tissue coverage can be provided in the acute or subacute phase. The microsurgical techniques initiated by replantation and free tissue transfer have revolutionized lower limb salvage.

The central tenets of this surgery include the fixation of the bone fractures with preservation of limb length, maintenance or restoration of normal distal blood flow, and closure of the wound by ample, well-vascularized soft tissue. This surgery should only be performed if sensibility can be expected in the foot and if the anticipated limb function will be better than would be expected for an amputation and prosthesis. Lower limb function with a prosthesis is quite good, particularly for a below-knee amputation. Thus the anticipated function of the reconstructed part must be very good to be worth the multiple surgical procedures and prolonged rehabilitation usually required.

Chronic osteomyelitis of the lower limb is best treated by complete excision of the affected bone followed by coverage with a well-vascularized flap. For osteomyelitis in sites as distal as the midtibia there are adequate local muscle flaps which can be transposed into position. However, for distal osteomyelitis close to the malleolus flaps for soft tissue coverage are not available, and free tissue transfer is required.

The essential part of the surgical therapy is the debridement of infected and necrotic bone. Initial surgical debridement is performed before antibiotic administration to be sure that the offen- ding organisms are identified by culture. The complete excision of all involved bone is assured by repeated surgical procedures (sometimes three are required) with examination of the bone surfaces in the wound between procedures for the development of granulation tissue. Occasionally, the extent of the infected bone necessitates the removal of the complete circumference or a segment of the bone. In this case an external fixator must be applied to hold the bones in alignment and the limb at proper length. At the least the limb is maintained in a protective cast or splint and in a non-weightbearing state to avoid pathological fracture. Once the bone is covered by granulation tissues, the wound may undergo final debridement and free muscle flap closure. Usually this operation is preceded by an arteriogram to check that the recipient vessels are adequate to support the flap.

Postoperative care is similar to that of the replantation patient. Anticoagulant agents may be given intravenously. The patient is maintained at bedrest with the affected limb elevated. Suction drains are kept in the donor and recipient wounds until drainage is minimal or until the flap is deemed to be well adherent to the bone in the wound bed. The myocutaneous or muscle flap can be examined directly to be sure that anastomotic thrombosis has not occurred. In addition, the subcutaneous or vessel surface temperature can be followed for quantitative evidence of thrombosis. If thrombosis is suspected, the patient should be returned as soon as possible to the operating room for direct examination of the anastomosis and any repairs that are necessary. Free flaps have an average initial viability rate of about 90 per cent; of the 10 per cent which undergo early thrombosis perhaps half can be salvaged.

Trauma or cancer resection may result in large defects of skull and scalp. Dural resection, if required, is usually performed in concert with a neurosurgeon. In a non-infected surgical field, the bone defect may be restored at the same time as wound closure. If the wound is contaminated or if there is concern about potential flap necrosis, the bone defect may be reconstructed at a later date.

Local transposition flaps of the entire thickness of the scalp are based on the rich vascular supply from the anastomotic network of the superficial temporal and occipital vessels. Large flaps may be transposed safely to cover the defect with the donor site closed with a split-thickness skin graft. Because the galea is the least elastic layer of the scalp, the flap can be stretched to a larger surface or greater reach by scoring the underlying galea. Care must be taken in the course of the scoring to avoid injury to the vessels which run immediately above the galea.

Skin expansion is particularly useful for scalp closure. The skull provides a firm base on which to stretch the overlying skin, and of course, the flap transfers hair-bearing tissue without leaving a glabrous donor site (Fig. 31) 2840,2841.

An adjacent, but not local, flap which may be used for scalp reconstruction is the trapezius myocutaneous flap. Based on the circumflex cervical vessels and the trapezius muscle, a paddle of skin can be brought from the midback to the cheek, orbit, or scalp. This flap is particularly suitable for occipital reconstruction. The donor site can be closed primarily or skin grafted.

As with other sites at the body's periphery where local flaps may not be sufficient for large defects, the scalp is frequently the site offree tissue transfer reconstruction. Free flaps that provide the broad, flat characteristics required for the scalp include the latissimus dorsi muscle, the scapular flap, and the omentum covered by a split-thickness skin graft. The superficial temporal vessels are frequently small and spastic, and vein grafts may be necessary to derive an inflow from the external carotid.

Resection of cancer of the floor of the mouth may leave a defect affecting portions of the tongue, the floor of the mouth, the mandible, and the neck lymph nodes. An ideal flap for reconstructing this wound is the pectoralis major myocutaneous flap. An island of skin can be carried on the pectoralis major muscle with its blood supply from the thoracoacromial vessels. The flap and muscle are turned back on themselves through the dissected neck to provide coverage of the exposed carotid artery. The skin paddle is inset into the mucosal and tongue defect.

A mandibular defect may be reconstructed at the time of resection and primary flap closure, or at a later time. If not reconstructed, a mandibular defect will allow deviation of the mentum to the defect side and will prevent the fitting of functional dentures. However, it is important to emphasize that not every defect requires reconstruction. The bulk of the pectoralis flap will minimize deviation, and a bone graft will not necessarily allow mastication.

Mandibular defects can be simply reconstructed with a plate and screws. For lateral, short segment losses, the plate functions well. Unfortunately, plates in the chin or in an irradiated area may become exposed, and removal is then required.

For short segments of bone loss (less than 6 cm), a non-vascularized bone graft from the rib or iliac crest will heal. Defects greater than 6 cm long require a vascularized bone graft. A section of vascularized rib may be carried with the pectoralis major myocutaneous flap. Large defects are, however, best reconstructed by microvascular transfer of free bone flaps. The best free bone grafts are the iliac crest and the fibula. The iliac crest will replace up to 15 cm of absent mandible; the fibula is needed for longer reconstructions. The iliac crest has a configuration which is suitable to the particular confirmation of the mandible and the ramus and angle. The fibula provides a long piece of bone with a thick cross-section suitable for dental implants.

While not all maxillectomy defects require reconstruction, the hollowness of the cheek, the absence of the malar prominence, and the incontinent palate can be corrected with an osteocutaneous free flap from the scapula. The circumflex scapular vessel penetrates the triangular space to supply the lateral border of the scapula and the overlying skin. After osteotomy, the bone spans the infraorbital defect, the de-epithelialized flap fills the cheek, and the skin on the tail of the flap closes the palate. If there is also a defect of the external cheek skin, there is an effective need for three skin flaps to cover the external cheek, the soft and hard palate, and the lateral nasal wall. These types of defects can be replaced by designing a large free flap with multiple separate petals of skin separated by de-epithelialized bands. In this fashion one flap is divided into several flaps to be inset in each of the separate areas required. The latissimus dorsi myocutaneous flap or the scapular free flap is particularly suitable for this type of reconstruction. The scapular flap has the added advantage that a section of bone from the lateral border of the scapula can be carried with the vascular bundle and used to reconstruct the zygoma or maxilla. The maxillary defect can also be reconstructed using a radial forearm flap and a separate graft of vascularized fibular bone anastomosed to the radial forearm flap. This technique is particularly useful for bilateral maxillary defects.

Soft tissue deformities of the face, particularly of a hemifacial nature, may be congenital, acquired, or postsurgical. These can be corrected by the microvascular free tissue transfer of soft tissues from other parts of the body. The first free tissue transfers attempted for treatment of hemifacial microsomia were free omental grafts. Free tissue transfer flaps of omentum based on a single vascular pedicle were transferred to the face in a subcutaneous position with multiple tongues of omental fat carefully placed in the abnormal contours of the face and revascularized by microvascular and venous anastomosis to the superficial temporal system. Although the initial restoration of facial contour was excellent, the long-term result was one of gravitational drop of the omental fat to the lower face.

A better flap for subcutaneous augmentation is the de-epithelialized scapular free flap. This flap is based on the circumflex scapular branch of the subscapular artery. A large flap of upper back skin and fat can be harvested to restore contour to the forehead, cheek, and jawline. This flap is de-epithelialized so that it can be completely buried under the normal facial skin. The flap can be cut to fit the necessary contours. After suturing in place, the flap is less susceptible to the gravitational change seen with free omental transfers. The flap is frequently so full that a subsequent procedure is needed to decrease the prominence of the augmented soft tissues.

Cervical oesophageal reconstruction is ideally effected with a free transfer of a segment of jejunum. Although the colon or the stomach can be advanced superiorly to the cervical oesophagus, the vascularity of these tissues at the level of the neck is decreased. The ideal tissue to replace the cervical oesophagus is the jejunum. A loop of jejunum up to 16 or 18 cm in length can be harvested based on a single proximal artery and vein supplying the arcade to the entire segment. The segment is transferred to the cervical oesophagus and placed in an antegrade direction for normal peristalsis. The artery and vein are repaired using microvascular techniques to restore blood supply to the intestine. The proximal and distal anastomoses are performed in a standard fashion. Intestinal continuity is returned by a standard intestinal anastomosis. Because the cervical anastomoses are performed to well-vascularized intestine, the risk of fistula or stricture is low. Oral alimentation can be resumed in 7 to 10 days.

Total parotidectomy generally entails truncal resection of the facial nerve. The ideal reconstruction is immediate sural nerve grafting. Multiple cables of sural nerve are grafted in an interfascicular fashion from the trunk to all branches.

The management of major fractures of the facial skeleton has undergone a dramatic revolution in the last 15 years. Several new diagnostic and therapeutic modalities have evolved, including CT scans, which make possible the precise definition of fracture patterns, extended open reduction techniques, rigid fixation systems, and acute bone grafting. These techniques make restoration of preinjury anatomy and function realistic goals, even following the most severe injuries.

The mechanism of injury is important in estimating not only the extent of injury to the facial skeleton, but also the likelihood of injury to the central nervous system and other major organ systems.

Shortly after injury and before the onset of soft-tissue swelling, architectural deformity with facial asymmetry may be obvious. This abnormality is often best noted by looking from above. Numbness in the distribution of a branch of the trigeminal nerve suggests fracture of the bone through which the branch passes. Hence, numbness in the infraorbital nerve distribution suggests zygomatic complex fracture; supraorbital nerve numbness, frontal bone fracture; and mental nerve distribution numbness, mandibular fracture.

Fractures of the maxilla or the mandible distort the preinjury relationship of the maxillary and mandibular dentition and therefore should initiate a search for fractures in one or both areas.

Palpable crepitation and false motion should be searched for by gently palpating all contours of the facial skeleton. Midface mobility is elicited by stabilizing the frontal area with one hand and gently grasping and rocking the maxillary alveolus with the other. Motion indicates fracture at one of the LeFort maxillary fracture levels.

Radiological examination defines the location and extent of fracture and aids in determining the need for and the method of therapy. In the acute setting, plain films are limited by the need for precise patient positioning and by bone outline overlap, making interpretation difficult. The Water's—mento-occipital position, the submental vertex, and oblique views of the mandible are all helpful. The single most revealing plain film is the reversed Water's view which requires little patient positioning and can be obtained in almost all patients with portable equipment (Fig. 38) 2856. Thin-section CT scan has become the gold standard for facial skeletal imaging (Fig. 39) 2857. It provides precise information about fracture patterns and displacement. In addition, the status of the orbital and intracranial soft tissues is also made clear. Three-dimensional CT scans generated from summation of averages of two-dimensional slices lose detail, but provide excellent perspective (Fig. 40) 2858. Patients suffering high energy trauma should have the entire cervical spine including C1-C2 and C6-C7 visualized. If not visualized, a cervical injury should be assumed and the head immobilized with a cervical collar.

Clinical experience has shown that superior results are achieved when fracture reduction and fixation is performed in the acute phase. Treatment delayed after 10 days becomes limited by contraction of the contused soft tissue envelope over the malaligned skeletal infrastructure, bone resorption, and bone healing. Emergency therapy is ideal when experienced teams and safe monitoring are available. Haemorrhage, significant risk to vision, ventilatory insufficiency, and elevated intracranial pressure are contraindications to immediate therapy.

Closed reduction techniques may be suitable for uncomplicated nasal, zygomatic, and mandibular fractures. Fractures with displacement and comminution are usually best managed with extended open reduction and fixation techniques.

Surgical incisions used to approach the facial skeleton are borrowed from aesthetic surgery. They include the bicoronal, lower blepharoplasty and intraoral gingivobuccal sulcus incisions. Mandibular fractures anterior to the mandibular angle are usually explored intraorally. The angle, ramus, and complicated fractures usually require extraoral incisions (Fig. 41) 2859. Through these incisions, subperiosteal dissection is performed to allow exposure of all fracture segments and to attain an anatomical reduction.

Visibility of scars from elective incisions is considerably influenced by their placement. Incisions should be placed within or parallel to skin lines with minimal tension that perpendicular to the overlying musculature. In the face they are obvious at ‘wrinkle lines’ of the muscles of facial expression. Placement of incisions at junctions of unlike tissues such as the brow and forehead skin, or hairline and forehead skin make the resultant scar less conspicuous.

Wire osteosynthesis is being gradually replaced by the use of miniaturized plates and screws borrowed from orthopaedic surgery. Screws that can gain a purchase on both cortices and are of sufficient scale to counteract the forces of the muscles of mastication are used in the mandible. Stability increases when compression osteosynthesis by the special design of the plate face is employed. In the upper face, small scale ‘mini-plates’ that purchase a single cortex and provide three-dimensional stability are employed.

When bone is lost or extensively comminuted, the preinjury contour is restored by replacing the severely damaged or missing bone with bone grafts placed at the time of acute reconstruction. The outer table of the skull, ribs, and iliac crest are preferred donor sites. Acute bone grafting is never performed in the mandible, where it is accompanied by high incidence of infection.

In addition to providing three-dimensional stability of the reconstructed skeleton, plate and screw fixation allows the release of maxillomandibular fixation immediately after operation, in most cases thereby decreasing the need for tracheostomy, improving oral hygiene, and allowing better postoperative nutrition as well as decreasing hospital stay. Disadvantages of rigid fixation include the expense of the equipment, a steep learning curve, and the unforgiving nature of the fixation and technique.

The nasal bones are the most commonly fractured bones of the facial skeleton. When fractures result from high energy forces, they are frequently associated with fractures of contiguous areas including the maxilla, orbital, frontal, and cranial base. Plain radiographs are of little help in diagnosis. CT scans aid in diagnosis of injury to adjacent structures in high energy injuries.

Nasal fractures are grouped into two general types, depending on the direction of impact forces. Lateral impact forces may fracture both nasal bones and the cartilaginous septum while deviating them to the opposite side. Frontal impact forces cause telescoping of the nasal septum and splaying of the nasal bones. This results in nasal shortening, dorsal saddling, and upper nasal widening. If impact forces are severe enough, fractures extend beyond the nasal bones into the medial orbital area and even into the cranial base (Fig. 42) 2860. Ideally, nasal fractures are treated immediately after injury, before swelling makes adequacy of reduction difficult to determine. This is usually undertaken with local anaesthesia, using xylocaine with adrenaline local infiltration and cocaine packs to anaesthetize the nasal mucosa topically. Septal haematoma should always be drained acutely to avoid problems from pressure necrosis and/or secondary infection.

In laterally deviated fractures, a Kelly clamp or a knife handle is used to centralize the nasal bones. A greenstick fracture on the opposite side must be completed to allow its repositioning. Frontal impact injuries which result in collapse often require Asch forceps placed on either side of the septum to allow elevation of the nasal bones into their proper position. Fractures that extend into the nasoethmoid region may need bone graft support.

After reduction, a protective dorsonasal splint maintains bony position. Nasal packing of non-adherent antibiotic-coated gauze is used to maintain the reduction of the septum and to limit haematoma formation. In comminuted injuries it prevents nasal bone collapse by providing counter pressure against the external splint.

High energy blunt trauma to the central face may result in fractures beyond the limits of the nasal skeleton into the medial orbits, frontal, and ethmoid sinuses. When that portion of the medial orbital rim which bears the medial canthal tendon is fractured with the potential for displacement, acute treatment is required to avoid telecanthus. The characteristic deformities of the fractures when not adequately treated are telecanthus, nasal shortening, and loss of nasal support. Fractures of this region are often accompanied with injuries to adjacent structures.

Proper management requires open reduction and internal fixation, taking care to reposition the medial canthal ligament. Reduction is usually accomplished with transnasal canthopexy. Acute bone graft of the nasal dorsum is usually required to restore nasal height. No attempts are made to assure the patency of the lacrimal system. Proper reduction and internal fixation reduces the need for lacrimal surgery to less than 20 per cent of patients with this injury.

Mistakenly labelled a ‘tripod’, the zygoma has articulations with five adjacent structures: the frontal bone, the arch, the greater wing of the sphenoid (lateral orbit), the medial maxilla (orbital rim), and the maxillary alveolus (lateral buttress). Characteristically, frontal impact causes malar depression and arch bowing; lateral impacts cause arch depression. Fractures usually extend into the internal orbit, and if severe, disrupt the floor and medial wall enough to increase orbital volume, resulting in enophthalmos and ocular globe dystopia. Entrapment of the extraocular muscles in the fracture site may cause diplopia. Infraorbital nerve anaesthesia results from impingement on the infraorbital nerve. Several radiographic views are necessary to evaluate the zygoma. The Water's view shows displacement at the infraorbital rim and fluid in the maxillary sinus. The Caldwell view is necessary to show displacement at the zygomaticofrontal suture, while only the submental vertex view will show posterior displacement of the malar complex. CT scans define the entire area including the status of the internal orbit and its structures.

Isolated zygomatic arch fractures usually have a classic W-shaped fracture pattern. They can be treated by closed reduction using the Gillies approach. This is a small scalp incision made within the temporal hairbearing skin and carried to the deep temporal fascia. A long instrument is passed beneath this fascia beneath the zygomatic arch to allow its elevation. Closed reduction is usually stable.

Closed reduction of the zygomatic body fractures can be performed in selected patients without comminution or instability of their fractures. Closed reduction can be successful in patients who have only infraorbital rim and zygomatic maxillary displacement. Closed reduction should be performed within 72 h after injury and patients should be monitored closely for subsequent displacement.

Open reduction and internal fixation is performed through a subciliary skin muscle flap, lateral rim of upper lid incision, or intraoral gingivobuccal sulcus incision, as required. Alignment is confirmed at the zygomaticofrontal suture, the infraorbital rim and the lateral buttress arch prior to intraosseous wire or, preferably, plate and screw fixation at one or two of these points. Three point fixation, and preferably plate and screw fixation, resists the downward pull of the masseter muscle. In the past, antral packing or K-wire fixation has been used to prevent displacement.

Orbital floor fractures inevitably accompany zygomatic complex fractures. Less frequently they occur as isolated ‘blow out’ fractures. Surgery is indicated for persistent entrapment as determined by force duction test, enophthalmos, or vertical dystopia. CT scanning is helpful in assessment. Defects greater than 1.5 cm result in enophthalmos if not treated. Treatment is intended to restore the preinjury anatomy of the internal orbit. After any fat which may have prolapsed into the maxillary antrum is retrieved, the defect is defined and reconstructed with an alloplastic implant or autogenous bone or cartilage. Defects involving more than the floor are best managed with autogenous bone. Ophthalmic consultation should be obtained before and after surgery.

Little can be added to the classic work of LeFort who described the areas of weakness through which midface fractures tended to pass (Fig. 44) 2862. The use of CT scans preoperatively and of extended open reduction techniques in treatment have shown that fractures rarely occur in ‘pure’ Lefort patterns. Rather they tend to occur in combinations. The more severely damaged side tends to have more comminution in a higher level of fracture.

The classic deformity of untreated Lefort fractures is midface elongation and retrusion— ‘dish face deformity’. Malocclusion and midface mobility are the clinical findings. Treatment consists of first restoring the preinjury occlusion by placing the jaws in maxillomandibular fixation. This restores the proper preinjury facial projection. Height is restored by realigning the four maxillary buttresses, using the least injured one as a guide. Osteosynthesis is accomplished with intraosseous wires or, better, with miniplates which are placed at each buttress (Fig. 45) 2863,2864. Bone defects greater than 0.5 cm are filled with bone grafts. All devitalized sinus mucosa is debrided. The maxillary antra are not drained.

When plate and screw fixation is used, maxillomandibular fixation is removed and occlusion examined. If malocclusion occurs, the fixation has distracted the mandibular condyles out of its proper position. Because plate and screw fixation is non-yielding, the plates must be removed and the fixation process repeated. When rigid fixation is used for fixation, maxillomandibular fixation can usually be removed after surgery. Use of extended open reduction and intraosseous wiring alone or maxillomandibular fixation and suspension wires requires jaw immobilization for 4 to 6 weeks. The use of extended open reduction and rigid fixation techniques has replaced the use of suspension wires and external fixation systems in many centres.

Mandibular fractures tend to occur in structurally weak areas of the mandible, including the subcondylar area, the angle, and the cuspid region. In the dentate mandible the subcondylar area is most frequently fractured, while in the edentulous mandible fractures occur most commonly in the body and angle area. Fracture displacement is determined by the direction of the blow, the pattern of the fracture, and the direction of pull of the strong muscles of mastication. The unique shape of the mandible results in a high incidence of bilateral fractures. These often occur in relatively standard combinations: symphysis and bilateral subcondylar, parasymphysis and contralateral subcondylar, angle and contralateral body.

Signs of mandibular fracture include malocclusion, local pain, swelling, and crepitation on palpation. Standard radiographs confirm and delineate the fractures. Few fractures escape detection by panorex or CT examination.

Restoration or a functional reapproximation of the preinjury occlusion is the goal of therapy. Maxillomandibular fixation is the standard treatment for the majority of mandibular fractures. Arch bars are ligated to the teeth. Elastic traction will overcome the muscle pull to correct small degrees of displacement. Closed be successful with fractures whose muscle force is opposed by the direction of the fracture line, thereby preventing displacement. In unfavourable fractures, the fracture lines do not oppose the muscle pull and displacement tends to occur when maxillomandibular fixation is used alone. Open reduction and internal fixation is, therefore, necessary. In the edentulous mandible dentures or dental splints may be used to restore the maxillary and mandibular occlusion and provide a means for fixation. These appliances are fixated to the mandible by circum-mandibular wiring and to the maxilla by fixing to the alveolar process or with suspension wiring. Maxillomandibular fixation is maintained for 4 to 6 weeks, with the exception of subcondylar fractures. In these cases, excursion must be initiated at 3 to 4 weeks to prevent fibrous ankylosis of the temporomandibular joint which is usually damaged in subcondylar injuries. Subcondylar fractures are rarely open in adults and virtually never so in children.

Open reduction is indicated when fracture displacement cannot be corrected and stabilized by maxillomandibular fixation alone. Intraosseous wires should be positioned at right angles to the fracture line for best stability. Stabilization is provided postoperative by splinting with fixation for 4 to 6 weeks. Plate and screw stabilization has increased the indications for open reduction with internal fixation since it obviates the need for maxillomandibular fixation in the postoperative period for many patients (Fig. 46) 2865. It also has the potential for better results with complex injuries. Despite the increased exposure required, there is no increase of infection when used correctly. Plate and screw fixation should only be used when the surgeon is well versed in the principles and techniques of their application.

When teeth are involved in the fracture line, they are removed if they pose a risk for infection or necrosis. All injuries with teeth in the fracture line and all open fractures should be treated with prophylactic antibiotics.

Cosmetic surgery is frequently considered as a series of surgical procedures which are somewhat apart from reconstructive plastic surgery. Indeed, there are surgical procedures, such as will be discussed below, which are performed only for the improvement in appearance from an acceptable ‘normal’ to a more ‘coveted normal’. Reconstructive procedures are likely to have a dominant functional goal. However, all reconstructive procedures have in addition a strong ‘cosmetic’ motivation for their performance and strive to produce a pleasing appearance. By the same token, cosmetic procedures require the same expertise of plastic surgical techniques as reconstructive procedures, and it may be quite difficult to say whether any one procedure is cosmetic or reconstructive. Is a breast reduction or restoration of facial motion after facial nerve paralysis to be considered cosmetic or reconstructive? For purposes of explanation only the following surgical procedures are categorized as cosmetic. However, all plastic surgical procedures have both cosmetic and reconstructive aspects and the best plastic surgeons will be accomplished in both.

The external appearance of the nose may be altered in patients with congenital, developmental, or traumatically acquired abnormalities. Rhinoplasty is usually performed after the nasal skeleton has matured in adolescence. The indications for performing rhinoplasty are the desire by the patient to change a perceived contour irregularity which is seen by the surgeon as to be worth the risks involved. The surgeon must be ready to probe the patient's inner feelings regarding the appearance of the nose, which may have strong psychosocial aspects.

The procedure is usually performed on an outpatient basis, using local anaesthesia with intravenous sedation. The goal of the surgery is to alter the underlying bone and cartilage skeleton and allow the skin to contract around this new form (Fig. 47) 2866. Surgery is performed through the nostrils with intranasal incisions as much as possible.

The midline dorsal prominence is reduced using a rasp or osteotome for the bone and a scalpel for the cartilage. The medial borders of the upper lateral cartilages may require resection as they also produce some dorsal prominence. The upper border of the lower lateral cartilages is resected either with the initial intranasal incision or under direct vision with the cartilages everted through another incision just under the nostril rim. The nasal bones, which have been separated from each other by the initial bone rasping, are now cut at their junction with the maxilla so that they may be inclined toward each other to narrow the nose and close the dorsal ‘roof’. The aesthetic goal of rhinoplasty is the production of a balanced nose that does not appear to have had surgery (Fig. 48) 2867.

Post-traumatic nasal deformity is often associated with septal deviation that may interfere with nasal breathing and alter the timbre of the voice. This septal deformity can be corrected at the same time as a rhinoplasty. Prior to the osteotomy of the nasal bones, the mucoperichondrium is elevated from both sides of the septum through an incision in the membranous septum. The deviated portion of the quadrilateral cartilage is resected or sculpted to correct its curvature. In addition, the surgeon must correct the usual concomitant deformity in the perpendicular plate of the ethmoid and the vomer to assure that this posterior portion of the septum is not left obstructing the airway. Occasionally, the inferior turbinates will need to be removed from a prominent position in the airway by outfracture or resection. Although rhinitis sicca can result from an airway made too patent, the most common complication of septal surgery is failure to clear the airway of all obstructions.

Excess facial skin and fat, particularly in the lower face, can be removed or reduced by a face lift procedure. Two symmetrical incisions are made from the scalp to the preauricular skin to the postauricular sulcus and into the posterior scalp to allow the elevation of bilateral cheek flaps which are advanced posteriorly to resect an effective ellipse of facial and scalp skin and tighten the facial tissues along a vector parallel to the jawline. The cheek flaps are elevated in a subcutaneous plane across the cheeks, jawline, and down the neck to the level of the hyoid. Frequently the platysma muscle and the submuscular aponeurotic system are elevated as a separate layer in the face lift and transposed superiorly and posteriorly to tighten the deeper tissues. A submental incision may be made specifically to excise submental fat and to excise or suture the medial borders of the platysma muscle. The supraplatysmal fat within the neck may be removed by direct excision or by suction lipectomy. The platysma flap and the skin flaps are then pulled posteriorly and superiorly under tension and sutured. The major tightening of the face lift occurs from the level of the oral commissure to the hyoid. There is little change to the nasolabial folds. The major complications of face lifting include haematoma with loss of pre- or postauricular skin and injury to the facial nerve branches.

Upper and lower eyelid excess skin and periorbital fat can be corrected by a blepharoplasty or a blepharoplasty combined with a brow lift. The lower lid may have excess or herniated periorbital fat, even in a young person. Such a configuration of the lid can be a familial characteristic that worsens with age. Young people will occasionally have a lower lid blepharoplasty to remove this excess tissue, which will not reaccumulate until quite late in life. The lower lid blepharoplasty is performed through a subciliary incision carried out laterally in one of the laterally radiating wrinkles. The incision can also be created through the conjunctival fornix. Through the subciliary incision the skin and orbicularis oculi muscle are elevated as a single flap. The dissection is carried down to the orbital septum which is entered in the medial, middle, and lateral fat pockets. These fat pockets are removed back to a level of the infraorbital rim. The skin and muscle are then tailored to remove excess and the wound is closed.

The upper lid blepharoplasty usually removes a larger ellipse of skin from the supratarsal fold superiorly. A section of orbicularis oculi muscle is also removed to enhance fixation of the supratarsal fold to the levator aponeurosis. The orbital septum is then entered and the medial and lateral pockets of periorbital fat are excised back to the level of the supraorbital rim (Fig. 49) 2868.

The major complications of blepharoplasty relate to exposure of the conjunctiva through excessive tissue release or scar contraction. Rarely, a haematoma in the deep periorbital fat may lead to temporary or permanent blindness.

The effects of upper lid blepharoplasty can be significantly improved by a brow lift, which also elevates the forehead tissues. Muscle excision is often performed to minimize furrows and wrinkles on the brow. The brow lift is performed through a coronal incision, a pre-hairline incision, or through two separate elliptical incisions just above the brow; the coronal incision is the most common. The forehead subcutaneous tissue is elevated superficial to the pericranium with careful preservation of the supraorbital nerves. The corrugator muscles are resected medially just under the supratrochlear vessels and nerves. The brow is then elevated to improve the contours of the lateral upper lid and the wound closed. The major potential complication to this procedure comes from injury to the supraorbital or supratrochlear nerves and decreased forehead sensation or injury of the facial nerve branches to the frontalis muscle and an inability to elevate the forehead and eyebrow on that side.

Facial contour abnormalities can be improved by various types of alloplastic facial implants. The most common implant used is placed over the chin to correct microgenia. The implant is either of silastic or proplast. The alloplastic implant is placed through an incision either in the buccal sulcus or the submental area. Care must be taken to preserve the mental nerve sensation to the lower lip. The incision is closed in multiple layers to insure healing without infection. Although the implant is placed through the oral cavity, infection is rare. The main complication of this procedure is decreased lip sensation and long-term bone erosion. The cosmetic result of the surgery is generally excellent. When combined with rhinoplasty, a major improvement in the balance of the profile can be achieved (Fig. 50) 2869.

Prominent malar eminences can be created by the placement of malar implants. These implants can either be positioned in a lateral, middle, or medial position to change the contours of the cheek bones. The implants may be of a tear drop shape or specifically contoured to mimic the infraorbital rim and zygomatic arch. The implants are placed through an intraoral incision passing in a supraperiosteal position lateral to the infraorbital nerve and superior to the zygomaticus major muscle on to the malar prominence of the zygomatic bone. The implants can also be placed through a standard subciliary blepharoplasty incision. The potential complications of this procedure include damage to the infraorbital nerve and decreased upper lip sensation or asymmetry of the implants.

Indented scars on the skin, wrinkles, and other surface irregularities can be somewhat improved by dermabrasion or chemical peel. Dermabrasion is a method of abrading the outer layer of the skin, including the epidermis into the mid-dermis. After the dermabrasion the open wound is allowed to re-epithelialize spontaneously. In the course of this re-epithelialization there is an increase in the thickness of the collagen bundles within the deep dermal tissue. Dermabrasion is particularly suited for acne scars. The effect of dermabrasion is to bevel the sharp edges of the acne scar so that a less prominent shadow is cast. The scar is not truly removed, but is altered in a way that it may be less visible. The actual abrasion is usually done with a mechanical wheel with a wire brush or a carborundum drum. The procedure may be performed under local anaesthesia or general anaesthesia. For maximal effect it is important to carry the abrasion to the deep dermis. Of course, epithelial remnants must be left within the hair follicles or sweat glands to allow for epithelialization. After surgery the dermabraded area is bandaged and allowed to epithelialize under this bandage. The epithelium will regrow in 4 to 7 days. The area of abrasion remains quite erythematous for several months. Eventually, however, the abraded area is usually paler than the surrounding skin. The potential complications of dermabrasion include hypertrophic scars, milia, and hyperpigmentation. The dermabrasion can be repeated at a later time with some additional slight improvement in the scars.

Chemical peel is more suited to the correction of fine facial wrinkles or hyperpigmented areas. In chemical peel a 50 per cent solution of phenol or a weaker solution of trichloroacetic acid is applied to the area and then covered with an occlusive dressing. The caustic material penetrates the skin to a variable depth causing epithelial necrosis and reorientation of the collagen structure of the dermis on healing. The occlusive dressing is left for 48 h and then removed with the necrotic outer layer of epidermis. The underlying wound is then dressed and allowed to epithelialize spontaneously. The epithelium usually regenerates in 4 to 10 days. The wound remains erythematous for 3 to 6 months. After healing, the depth and prominence of normal facial wrinkles is significantly decreased.

A biproduct of the chemical peel is hypopigmentation of the skin. This effect can be used therapeutically to remove dark facial pigmentation such as cloasma. The chemical peel solution is applied to the pigmented area and dressed in the same way. The adjacent areas of normal facial skin are not necessarily treated. The pigmented area will frequently be bleached significantly.

The complications of chemical peel include delayed healing, prolonged erythema, hypertrophic scarring, and distortion of facial features.

Excess subcutaneous tissue can be removed from several areas of the body to improve aesthetic contours. The subcutaneous fat may be removed by suctioning under the skin or if there is also excess skin, by direct excision of fat and skin.

The most common procedure involving direct excision of fat and skin is an abdominoplasty. Excess lower abdominal fat and skin, particularly common in postpartum women, can be removed from the level of the umbilicus down to the pubis and closed primarily. The upper abdominal skin is undermined just superficial to the anterior abdominal fascia so that a large upper abdominal advancement flap is created which is sutured at the level of the inguinal creases and the upper pubis. The umbilicus is left in its original position, but is brought out through a new opening in the centre of the flap. When the abdominal flap is elevated and after the lower abdominal ellipse of skin and fat has been resected, any diastasis recti can be plicated in the midline to further tighten the waistline. This is a major surgical procedure with major blood loss. There is a risk of skin loss in the lower abdomen and appreciable scarring.

Other areas of the body that can be contoured by excision of skin and fat include the medial upper arms, the lower midback around the circumference of the hips, and the posterior lower buttocks and thighs. These procedures are less common and do produce significant scarring.

When there is excess subcutaneous tissue without lax skin, or in a patient whose skin may be expected to contract after fat removal, a solid tube cannula can be introduced into the subcutaneous tissue and when placed on a suction close to a vacuum, the subcutaneous fat can be specifically removed without depriving the overlying skin of a blood supply. Generally this suctioning technique is carried out by passing the cannula through the subcutaneous tissue several times in slightly different directions so that the fat is removed evenly. The level of fat suctioning is at the deep premuscular fascial level so that surface irregularities are minimized. The total amount of fat removed must be carefully monitored as the blood loss can be impressive within the material removed by suction. There is much oedema within the tissues after surgery. The areas most frequently treated by suction lipectomy include the lateral and posterior thighs and buttocks, the medial knee, the lower abdomen, and the submental tissue. The potential complications of this procedure include surface irregularities, decreased skin sensation, and major fascial infection.

Adams WM. Internal wiring of fixation of facial fractures. Surgery, 1942; 12: 523.

Aghababian RV, Conte JE Jr. Mammalian bite wounds. Ann Emerg Med, 1980; 9: 19.

Amaratunga NA. The effect of teeth in the line of mandibular fracture or healing. J Oral Maxillofac Surg, 1987; 45: 312.

American College of Surgeons Committee on Trauma. Guide to Prophylaxis Against Tetanus in Wound Management. Chicago: American College of Surgeons, 1983.

Angle EH. Classification of malocclusion. Dent Cosmos, 1899; 41: 240.

Aronourty JA, Freeman BS, Spira M. Long-term stability of teflon orbital implants. Plast Reconstr Surg, 1986; 78: 166.

Branemark PI, Erkholm R. Tissue injury caused by wound disinfectants. J Bone Joint Surg, 1967; 49A: 48.

Brentano L, Craven DL. Wound closure: a method for the quantification of bacteria in burn wounds. Appl Microbiol, 1967; 15: 670.

Cannon B, Cope O. Rate of epithelial regeneration. Ann Surg, 1943; 117: 85.

Cannon B, et al. The use of open jump flaps in lower extremity repairs. Plast Reconstr Surg, 1947; 2: 336.

Converse JM, et al. Plasmatic circulation in skin grafts. Transplant Bull, 1957; 4: 154.

Converse JM, et al. Inosculation of vessels of skin graft and host bed: a forfeitures encounter. Br J Plast Surg, 1975; 28: 274.

Cormack GC, Lamberty BGH. A classification of fasciocutaneous flaps according to their pattern of vascularization. Br J Plast Surg, 1984; 37: 80.

Crikelair GF. Skin suture marks. Am J Surg, 1958; 96: 631.

Daniel RK, Taylor GI. Distant transfer of an island flap by microvascular anastomosis: a clinical technique. Plast Reconstr Surg, 1973; 52: 111.

Dingman RO, Natvig P. Surgery of Facial Fractures. Philadelphia: WB Saunders, 1964.

Edlich RF, et al. Management of soft tissue injury. Clin Plast Surg, 1977; 4: 191.

Gruss JS, MacKinnon SE, Kassl E, Cooper PW. The role of primary bone grafting in complex craniomaxillofacial trauma. Plast Reconstr Surg, 1985; 75: 17.

Jackson IT, Somers PC, Kjan JG. The use of Champy miniplates for osteosynthesis in craniofacial deformities and trauma. Plast Reconstr Surg, 1986; 77: 729.

Knight JS, North JF. The classification of malar fractures: an analysis of displacement as a guide for treatment. Br J Plast Surg, 1961; 13: 325.

Manson PN, Crawler WA, Yaremchuk MJ, Rochman G, Hoopes JE, French JH. Midface fractures: advantages of immediate extended open reduction and bone grafting. Plast Reconstr Surg, 1985; 76: 1.

Manson PN, Su CT, Hoopes JE. Structural pillars of the facial skeleton. Plast Reconstr Surg, 1980; 66: 54.

Manson PN. Facial injuries. In McCarthy JG, ed. Plastic Surgery. Philadelphia: W.B. Saunders Co. 1990.

McGregor IA, Morgan G. Axial and random pattern flaps. Br J Plast Surg, 1973; 26: 202.

Michelet FX, Deymes J, Dersus B. Osteosynthesis with miniaturized screwed plates in maxillofacial surgery. J Maxillofac Surg, 1973; 1: 79.

Schilli W. Compression osteosynthesis. J Oral Surg, 1977; 35: 802.

Schmoker R, Von-Allmen G, Tschopp HM. Applications of functionally stable fixation in maxillofacial surgery according to ASIF principles. J Oral Maxillofac Surg, 1982; 40: 457.

Spiessl B, ed. New Concepts in Maxillofacial Bone Surgery. New York: Springer-Verlag, 1976.

Tanner JC Jr., Vandeput J, Olley JF. The mesh skin graft. Plast Reconstr Surg, 1964; 34: 287.