If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Correspondence to Dr Rajiv Y. Chandawarkar MD, Plastic and Reconstructive Surgery, MD Anderson Cancer Center, Box 443, 1515, Holcombe Blvd, Houston, TX 77030, USA
Affiliations
Division of Plastic Surgery, Akron General Medical Center, Akron, OH, USA
Reconstruction of partial-thickness defects of the cheek can be challenging. In addition to maintaining function, the repair must restore contour, minimise donor-site deformity and not distort the eyelid, mouth or jaw line. Since the demands for repair differ according to the site, so do the reconstructive options. To aid in selecting the best method, we classify the cheek into five subunits based on vascular anatomy and relevant landmarks. Based on our experience with 160 patients over a 12 year period, we present an algorithm that helps to select the best reconstructive option for each site. This algorithm is easy to follow and conforms to anatomical principles.
The cheek is the widest skin expanse of the face and is an important cosmetic and functional unit. Centrally, it provides a platform for crucial midface structures, while supporting the lips, nose and lower eyelids. Peripherally, it frames the face. Skin quality and cheek volume are important in facial appearance, as illustrated by the process of ageing.
The aim of cheek reconstruction is to restore appearance with like tissue, while minimising donor-site morbidity. Though large cheek defects may require regional or distant flaps, most partial-thickness wounds can be repaired with local tissue. The cheek provides an ample source of tissue for use in the repair of a wide variety of defects. When properly selected, these local flaps can achieve good contour and colour match without distorting critical landmarks.
Our study examines reconstructive options for partial defects in various areas of the cheek. The central tenet is the division of the cheek into anatomical subunits similar to those described for the nose by Burget and Menick.
Based on our results, we propose an algorithm for the repair of partial-thickness cheek defects.
1. Materials and methods
We studied 160 patients who underwent reconstruction of the cheek between 1990 and 2002. Demographic data, as well as details of the nature, location, histology and size of their primary lesions, were noted. All patients underwent a wide excision of the lesion followed by immediate reconstruction. All defects were partial thickness, and averaged 2.8 cm in diameter. Patients with full-thickness defects (through-and-through into the oral cavity) were excluded from this study. The minimum follow-up was 3 months and the maximum was 5 years.
Reconstructive decisions were largely based upon the location of the lesion and the nature and availability of the local tissues. Fundamental considerations were using like tissue for like, minimising distortion at the donor site and in the reconstructed area, and, in males, restoring the pattern of facial hair. None of the patients in this study had facial scars that influenced the choice of reconstruction.
1.1 Anatomical classification of defects
The cheek was anatomically subdivided into five zones, as shown in Figure 1. Zones were created following two basic principles: anatomical contour and function, and neurovascular distribution (Fig. 1(A)).
Figure 1Subunits of the cheek. (A) The neurovascular anatomy was used as a guideline for division of the cheek into subunits. (B and C) Zones 1–5 seen in (B) frontal and (C) lateral views. Horizontal (solid) and vertical (dashed) lines mark the boundaries. Note that zone 5 is a central area spanning a portion of each of the other four zones.
Key landmarks for this zoning were simply identifiable anatomical structures, from which we defined three horizontal and three vertical lines (Fig. 1(B) and (C)). The horizontal boundary lines were the zygomatic arch, the horizontal plane from the nasal sill and the lower border of the mandible. The vertical lines were the lateral margin of the nose and the nasolabial crease, the lateral canthal line (a vertical line from the lateral canthus), and the preauricular crease.
1.1.1 Zone 1
The boundaries are the lower eyelid crease superiorly, the lateral canthal line (vertical line from the lateral canthus) laterally, a horizontal line from the nasal sill inferiorly, and the lateral margin of the nose and the nasolabial crease medially.
1.1.2 Zone 2
The boundaries are a horizontal line from the nasal sill superiorly, the lateral canthal line (vertical line from the lateral canthus) laterally, the lower jaw line inferiorly, and the vertical line from the angle of the mouth medially.
1.1.3 Zone 3
The boundaries are a horizontal line from the nasal sill inferiorly, the lateral canthal line (vertical line from the lateral canthus) medially, the preauricular crease laterally, and a horizontal line from the lateral canthus superiorly.
1.1.4 Zone 4
The boundaries are a horizontal line from the nasal sill superiorly, the lateral canthal line (vertical line from the lateral canthus) medially, the lower jaw line inferiorly, and the preauricular crease and the posterior margin of the ascending mandibular ramus laterally.
1.1.5 Zone 5
This is a centrally located area encompassing a diameter of 2 cm (Fig. 1(B) and (C)).
1.2 Reconstructive methods
Local flaps were used to reconstruct defects in each of these zones. The choice of flap and its design were dictated entirely by the location of the defect.
Zone 1 defects were reconstructed using a V–Y advancement flap, bringing skin and subcutaneous tissue from zones 2 and 5 (Fig. 2) . This flap was based on the angular arterial supply from the facial artery.
Figure 2Reconstruction in zone 1. Lesions in (A) the upper portion and (B) the lower portion of the zone. Both were reconstructed using V–Y advancement flaps. Postoperative results are shown 8 months after excision.
Zone 2 defects required a local transposition flap from the neck, just below the jaw line. The flap base was located medially or laterally, depending upon the defect size and the hair distribution in males (Fig. 3) .
Figure 3Reconstruction in zone 2 using cervical flaps.
Zone 3 defects were reconstructed using a local transposition flap based medially. Again, the hair distribution was a critical element in the choice of its orientation.
Zone 4 defects required a medially based local transposition flap (Fig. 4) . This design allowed better skin mobility and caused the least distortion of the skin.
Figure 4Reconstruction in zone 4. Note that the flap is based anteriorly as it offers better mobility.
In both zone 3 and zone 4, as they are essentially peripheral units, vertical scars were accepted as a compromise since they were less visible frontally and, in males, could be hidden under a beard.
Zone 5 defects were reconstructed using flaps based on the angle of the jaw (Fig. 5) .
Figure 5Reconstruction in zone 5. Preoperative appearance, and postoperative view 6 months after excision.
All patients were treated in outpatient surgery and allowed home the same day. There were no major complications. Minor wound breakdown at the incision occurred in two patients, both of whom were being treated with steroids for other medical conditions. There were no partial or complete flap losses, and all the local flaps healed satisfactorily. There was no haematoma or seroma formation and no donor-site morbidity.
Cosmetic results were excellent, and all patients were pleased with the outcome. The match of skin colour and texture was excellent. Symmetry was strongly maintained, although subtle contour distortion was evident in five patients after reconstruction in zone 5; their defects were approximately 3 cm in diameter. None of the patients had distortion of either the oral commissure or the lower eyelid. There was no evidence of hypertrophic scarring or contracture in the region.
Using a pre-set algorithm for facial reconstruction does not challenge the dogma of customising each repair. On the contrary, it provides a useful framework for preoperative planning and the design of the reconstructive process. In our algorithm, each zone is based on neurovascular patterns and provides a template for rapid selection of reconstructive options. This is in contrast to other studies describing subunits of the cheek.
points out that cheek defects are reconstructed using tissue from adjacent areas and not by interpolating flaps as in nasal defects. Also, it is not necessary to excise adjacent normal tissue within a subunit of the cheek as is commonly done in the nose.
Important to the success of any reconstructive algorithm is the preoperative planning. The lesion and its excision margins are carefully marked, and the cheek zone is noted. We commonly employ square or rhomboid excisions since these patterns maximise ‘tissue economy’ and avoid the need to discard large dog-ears when using facial advancement flaps (Figure 3, Figure 4, Figure 5). The excision pattern can be rotated clockwise to accommodate the best flap for each zone or to provide an appropriate axis for a V–Y flap. Needless to say, clear margins must be confirmed prior to definitive repair.
Cosmetic results after using this reconstructive algorithm can be excellent. One can provide good colour and texture match while restoring contour. Lip and eyelid deformities are avoided in zones 1, 2 and 4 by transferring tissue into the defect rather than closing by extensive undermining. Mobilising the base of these rotation flaps allows redistribution of tissue and avoids any long-term dog-ears.
V–Y flaps were used almost exclusively in Zone 1. As shown by Pontes et al,
one can use this flap for extensive defects not only in the cheek but also elsewhere in the head and neck. We have used this flap for large combined defects involving the medial cheek, lower eyelid and lateral nose. In our study we did not modify these flaps by using a bilateral pedicle as described by Pontes et al.
However, adding these features can extend the reach of the flap. The design of the axis of the V–Y flap was key in avoiding conspicuous scarring. We recommend that the axis of the flap be curved to allow for an oblique rather than a vertical scar. Compared with the Z-plasty reconstruction for defects in zones 1 and 2,
vertical lines in the central face are distracting and clearly visible on frontal view. These flaps have a good blood supply, lack long-term lymphoedema and generally result in an excellent donor scar, which is frequently hidden in the nasolabial fold.
In both zone 3 and zone 4 the defect determines the position of the base of the flap. Rotation flaps in zone 3 can be based medially or laterally, as both allow for mobile skin.
We have routinely preferred laterally based flaps, as they avoid scars in the anterior neck and preserve essential blood supply from the lateral aspect. Similar flaps in zone 4 are medially based, as the ear tethers the skin in the preauricular area and distortion can result.
These local flaps must be large and generous in their volume. As described by Menick,
large flaps share the donor burden, incorporate a larger vascular supply and permit later re-elevation and advancement if required. In addition, they should be thick enough to include an adequate vascular supply.
None of our patients required flap re-elevation but, given that 30–50% of patients with basal or squamous cell cancer will develop a second tumour, this consideration is important.
As in most patients undergoing reconstructive surgery, risk factors such as age, smoking and associated co-morbidities dictate options. Moreover, since random skin flaps are inherently prone to tip necrosis in high-risk patients, we recommend that flaps be elevated with deeper subcutaneous fat and include the superficial muscular aponeurotic system where applicable. These flaps are particularly useful in restoring volume to the cheek. Care must be taken to avoid injury to deeper structures, especially the facial nerve in the midface.
Although none of our patients required further surgery to modify contour or revise scars, this option remains available.
In conclusion, partial-thickness defects of the cheek place specific demands on the surgeon. Restoration of contour using local tissue with minimal donor-site deformity is the goal. An algorithm for preoperative planning provides a useful framework for effective reconstruction.
Division of Plastic Surgery, Summa Health System, Akron General Medical Center, Crystal Clinic and North Eastern Ohio Universities College of Medicine, 400 Wabash Avenue, Akron, OH 44307, USA
00026-2&id=gr1.jpg){kind=link}
00026-2&id=gr2.jpg){kind=link}
00026-2&id=gr3.jpg){kind=link}
00026-2&id=gr4.jpg){kind=link}
00026-2&id=gr5.jpg){kind=link}