Figure 1.
Image not available

Facial defect reconstruction paradigm: Nodes 1 through 5, factors to consider when deciding to use healing by second intention are shown. Nodes 6 through 11, factors to consider when deciding on the use of primary closure or a skin graft are shown (described in detail in "Compartment No. 1"). RSTL indicates relaxed-skin-tension line; NSAID, nonsteroidal anti-inflammatory drug; and DM, diabetes mellitus.

Figure 2.
Image not available

Eight flap types to use during facial defect reconstruction. The shaded areas marked by broken lines indicate underscored skin; black areas, discarded skin; unbroken lines, the line of incision; 2-way arrows, the vector of force; and D, diameter. A, The note flap: For reconstructing a circular defect, an underminable arc of about 90° is required for a distance equal to about 2 diameters of the defect. With optimal orientation, at least 50% of the scar can be oriented in relaxed skin tension lines (RSTLs), and, for all 8 flap types, the vectors of closing tension are as shown. B, The rhomboid flap requires a 90° arc and 1.5 diameters. The scar can be at least 60% in the RSTLs. C, The bilobed flap requires a 90° arc and 2.5 diameters. The resulting scar can be at least 65% in the RSTLs. D, The O-to-Z flap requires 2 opposing 90° arcs and 1.5 diameters. The resulting scar can be at least 80% in RSTLs. E, The subcutaneously pedicled flap requires 90° and 2 to 3 diameters. The scar can be at least 65% in the RSTLs. F, The triple rhomboid flap requires 360° and 1.5 diameters. The scar can be at least 50% in the RSTLs. G, The V-to-Y flap requires a 90° arc and 1.5 diameters. The resulting scar can be at least 65% in the RSTLs. H, The A-to-T flap requires a 180° arc and at least 2 diameters. The long limb of the resulting scar is ideally camouflaged in an anatomic boundary.

Figure 3.
Image not available

A skin defect and the type of flap reconstruction used: (A) the defect; (B) an analysis of the defect, as detailed in "Compartment No. 2"; (C) possible orientations of the rhomboid flap; and (D) the result of the reconstruction. See the legend to Figure 2 for an explanation of the diagrams.

Figure 4.
Image not available

A circular defect of the left temple: (A) the defect, (B) an analysis of the defect, and (C) the result of the reconstruction. See the legend to Figure 2 for an explanation of the diagram.

Figure 5.
Image not available

The defects used for the field test.

Table 1. 
Image not available
Nondistortable Structures and Facial Anatomic Units That Often Provide Skin for Reconstruction
Table 2. 
Image not available
Comparative Ranking or Scoring of Possible Reconstructive Options in the Field Test*
Table 3. 
Image not available
Individual Pretest and Posttest Scores of All Residents*
Table 4. 
Image not available
Average Scores for Each Postgraduate Year (PGY)*
Parker  SLTong  TBolden  SWingo  PA Cancer statistics, 1996. CA Cancer J Clin. 1996;465- 27Article
Young  VLBartell  TH Burns of the face. Thomas  JRHolt  GReds.Facial Scars: Incision, Revision, and Camouflage St Louis, Mo CV Mosby Co1989;229- 270
Original Article
January 1998

Teaching Paradigm for Decision Making in Facial Skin Defect Reconstructions

Author Affiliations

From the Department of Otolaryngology, University of Texas Medical Branch, Galveston.


Copyright 1998 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.1998

Arch Otolaryngol Head Neck Surg. 1998;124(1):60-66. doi:10.1001/archotol.124.1.60

Objective  To present a decision paradigm for facial defect reconstruction, and test the ability of this paradigm to improve resident performance.

Design  A decision paradigm for reconstruction of facial skin defects is proposed and explained, with patient examples. The paradigm's usefulness is then tested with residents.

Setting  Otolaryngology residency training program at a tertiary hospital.

Study Participants  Otolaryngology residents.

Interventions  Twelve residents took a pretest wherein they were presented with drawings of skin defects and asked to choose the "best" (most aesthetically pleasing) type of reconstruction from a closed set. This paradigm was presented to these residents, and their posttest consisted of choosing again with the same defects and closed set of choices.

Main Outcome Measures  Cosmetic outcomes of reconstructive decisions on the pretest and posttest were rated on a scale of 0 to 5 (with 0 indicating poor; 5, excellent).

Results  There was a significant improvement in reconstructive choices between the pretest and posttest (P<.001, Student t test).

Conclusion  This paradigm can be easily modified to accommodate different surgical approaches preferred by individual surgeons and is thus useful in almost any reconstructive teaching situation.

FACIAL BASAL and squamous cell carcinomas are common, with more than 800000 new cases occurring each year.1 The facial skin defects resulting from the excision of these cancers require reconstructive choices that range from healing by second intention through the use of skin grafts to the use of local or regional flaps. Surgeons experienced in reconstructive surgery make decisions based on an "instinctive" understanding of the size and location of the defect and individual patients' skin type and healing potential. The result is almost always good to excellent function and cosmesis. Teaching this type of decision making to novice reconstructive surgeons requires an examination of each facet of the decision. Because seldom is only 1 reconstructive method acceptable, teaching this decision making process can be difficult. We describe our decision paradigm for the reconstruction of facial skin defects and the field testing of this paradigm. Reconstructive surgeons have their own preferences, but this decision paradigm can be modified to accommodate different surgical approaches and is thus useful for teaching in almost any reconstructive practice.


Three otolaryngologists, 2 experienced reconstructive surgeons and the other holding an advanced degree in information and computer science, collaborated on this project. A sample of actual cases of facial skin defect reconstruction, representing a wide variety of anatomic locations and patient characteristics, was chosen from cases performed in the past 3 years. For each case, a decision flowchart was developed that diagrammed the actual decision making process used. These decision making diagrams were then merged to construct a composite decision making paradigm for use with most defect locations of the head and neck.

The decision-guidance system consists of 2 compartments. Compartment No. 1 includes a flowchart that is used to decide whether healing by second intention, a skin graft, primary closure, or the use of a local skin flap is the best choice for a patient. Compartment No. 2 details the process of choosing a specific local flap for the reconstruction of simple defects.

Each defect is first described by the anatomic (or aesthetic) unit or units it occupies.2 Those involving more than 1 unit are considered as 2 or more separate defects. Any defect in the nose, lip, ala, or eyelid area; those without available adjacent skin; or defects requiring bone or cartilage for reconstruction are complex defects and are not discussed in detail in this article.


Compartment No. 1 begins with the patient's general health (Figure 1). There are many relative contraindications to further surgical treatment. These include unstable angina and anticoagulation therapy that cannot be interrupted. Patients with the most severe of these rarely seek surgical reconstruction. Second-intention healing is often a wise choice for the few such patients who do present for reconstruction.

The patient's treatment desires are then considered. If a patient unequivocally does not desire any further surgical intervention, an immediate decision is made for healing by second intention (node 2 of Figure 1). Relative contraindications for healing by second intention are defects of the nasal ala, eyelid, and lip because contracture in these areas can cause notching and possibly alar notching, ectropion, or oral incompetence. The use of a skin graft or flap is preferred in these areas unless the patient understands and accepts the likelihood of poor cosmetic and functional results (nodes 3 and 4 of Figure 1). The other firm indication for second-intention healing in our practice is a skin-only defect of less than 1 cm2 in the medial canthal area. Other facial areas where healing by second intention may produce a reasonable appearance are the temple, forehead, and periauricular areas. Second-intention healing in convex areas such as the lower nose and eyelids generally produces poorer results.

All defects not undergoing second-intention healing are considered for suitability for primary closure (node 6 of Figure 1). These defects should have a satisfactory shape, orientation, and laxity to allow such closure. A defect that can be made to be long and narrow with approximately a 3:1 length-to-width ratio (to avoid standing cones) and where the scar can be oriented in or near the relaxed-skin-tension line (RSTL) is suitable for primary closure (nodes 6, 7, and 8 of Figure 1). A defect that is long and narrow has its orientation already determined. If the defect is circular, a 3:1 ellipse can be created that is oriented to place the scar in an RSTL. Skin laxity varies with age, and the younger the patient, the more undermining is required for a low-tension closure. The closure of an ellipse generally requires undermining the equivalent of the width of the defect on either side at the center, tapering to a total of 1 width at each end of the ellipse. Neither the undermining nor the creation of the ellipse can distort a nondistortable structure (the distortion of which yields functionally [eyelid ectropion] or cosmetically [eyebrow, hairline] unacceptable results [Table 1]). If a long and narrow defect is oriented perpendicular to the RSTL or if creating a 3:1 ellipse oriented along the RSTL or 1-width-equivalent of undermining along the long axis would impinge on a nondistortable structure, this defect is probably not suitable for primary closure.

Other factors can affect the likelihood of significant perioperative problems (nodes 9 and 10 of Figure 1). Heavy smokers and patients with insulin-dependent diabetes mellitus are at greater risk for small-vessel disease, and patients currently taking nonsteroidal anti-inflammatory drugs have a greater risk of hematoma formation, leading to flap necrosis. A full-thickness skin graft may offer a better chance of a good result than a local or regional flap in these patients. Skin grafts are also offered to patients who prefer the quickest possible surgery over the best cosmesis (node 11 of Figure 1). In some anatomic areas such as the nasal sidewall (subunit of the upper nose), skin grafting may offer the best cosmesis, but the remaining patients are candidates for local flaps.


We use the following decision framework for the consideration of relevant factors in choosing the best local flap. Complex flaps involving bone or cartilage replacement and reconstruction of the lip, nose, ala, or eyelids are not considered. In this compartment, the surgeon first describes the defect in detail, including size, shape, condition, and location of surrounding structures. The characteristics are then matched to flaps being considered, and the best flap for reconstructing the defect is then chosen.

For this study, we chose 8 flap types that are most commonly used in our practice: the note, rhomboid, bilobed, O-to-Z, V-to-Y, A-to-T, triple rhomboid, and subcutaneously pedicled flaps. Five characteristics were described for each: the arc of skin required, the diameter of skin required, normal skin discarded, scar orientation with respect to RSTLs, and the vector of maximal tension after closure (Figure 2, A through H). Other flap types can be similarly described.

The amount and location of skin required for the use of a particular flap is described by the arc and the number of diameters usually needed for elevating the flap and undermining for closure (eg, the note flap requires a 90° arc, the radius of which is 2 times the diameter of the defect). These flap diagrams also show the amount of normal skin that is discarded in transferring this flap and the shape of the scar resulting after flap transfer. When the scar is optimally oriented with respect to RSTLs, some of the scar will parallel (or nearly so) the RSTLs, and some of the scar will be perpendicular (or nearly so). An ellipse, eg, when oriented in the RSTL, will be 100% parallel. The use of the note flap results in a scar that is about 50% oriented with the RSTLs and about 50% against. In Figure 2, the numbers following the word "scar" describe this characteristic, as "100/0" for the ellipse (or 0/100 for an ellipse oriented perpendicular to RSTLs) or as 50/50 for the note flap. The final flap characteristic noted is the vector of the tension of flap closure, indicated by the vector-of-force arrows in Figure 2. It is undesirable to orient a flap so that the primary tension of closure pulls on (is perpendicular to) a nondistortable structure such as the lip or the eyelid.

A description of the defect begins by noting the size, shape, average diameter of the defect, and the involved anatomic units. The adjacent nondistortable features, good facial lender units (from which skin can often be borrowed with minimal distortion) and nonlender facial units (from which skin can rarely be borrowed with a good result) are identified (Table 1). Then the arc and the number of diameters of skin potentially available for reconstruction are diagrammed and described.

The amount and location of skin available for the reconstruction is compared with the amount and location of skin required for each of the flaps being considered. Those that are obviously incompatible (eg, the use of an O-to-Z flap when the skin available is limited to a 90° arc) are discarded from consideration. Any flap that would place the vector-of-closure tension perpendicular to a nondistortable feature is similarly discarded from consideration.

The final step of a successful flap reconstruction is to choose from among the remaining flaps the one that will yield the least noticeable scars. The least visible scars are those hidden in an anatomic boundary, such as the hairline or the nasolabial crease. Scars that parallel RSTLs are also minimally noticeable, especially in older patients. The 2 subcutaneously pedicled flaps (V-to-Y and tunneled) have slightly more tenuous blood supply than the other 6 flaps and are used only if they offer substantial advantages over these other flaps. If necessary, several sketches can be made in which each of the remaining flaps is tried, thus leading to the choice of the best possible flap for the reconstruction of that defect.

Figure 3 shows a skin defect (Figure 3, A), with its finished working diagram (Figure 3, B). It is a roughly rectangular defect, 14×22 mm, with an average diameter of 18 mm. It occupies the lateral lower lip and chin, with the adjacent lower lip being a nondistortable area. The chin and cheek are possible lender units. There is an arc of about 225° available for reconstruction that covers the lower lateral cheek, jaw, and submental and chin areas. The diameters potentially available for reconstruction are about 1 diameter medially, 1 to 2 diameters inferiorly, and 3 to 4 diameters laterally. For a flap-to-defect match, an inadequate arc is available for the O-to-Z or triple rhomboid flaps. There is no good boundary for long-limb placement for an A-to-T flap. Note, rhomboid, and bilobed flaps are all possibilities. The use of the technically more complex V-to-Y and subcutaneously pedicled flaps provides no advantage. Because the shape of the defect was closest to a rhomboid, this reconstructive option was chosen. The 4 possible rhomboid flap designs are shown in Figure 3, C. The vector-of-closure tension for options B and D potentially would pull the lower lip downward. Option C was chosen over A because the flap is developed laterally on the cheek, rather than being visible in the midline of the chin, and because we generally choose an inferiorly based flap over a superiorly based one when possible to ensure optimal venous and lymphoid drainage. The final result of this reconstruction is shown in Figure 3, D.

The second example involves a 38×42-mm roughly circular defect of the left temple (Figure 4, A). Adjacent nondistortable landmarks include the hairline, beard line, eyebrow, and eyelids. The patient intended to wear a full beard for the rest of his life. Possible lender units are the forehead and cheek (Figure 4, B). The available skin has an arc of about 180°, with the central portion unavailable, and approximately 2 diameters on the forehead and 2 diameters on the cheek. Thus, the arc is inadequate for a triple rhomboid flap. The note, rhomboid, and bilobed flaps are all possible choices. The V-to-Y and subcutaneously pedicled flaps are not necessary. Because skin is available on opposite sides of the defect, the O-to-Z and A-to-T flaps are good choices. There is a boundary in which to hide the long limb of the A-to-T flap, so this is chosen. Figure 4, C, shows the final reconstructive result.


The usefulness of this composite decision paradigm was field-tested on 12 resident physicians. The pretest consisted of each field tester taking 10 minutes to decide the 3 most suitable reconstructive options for 4 diagrammed facial defects (medial canthus, cheek, temple, and forehead) (Figure 5). A 15-minute talk was delivered explaining the facial reconstructive paradigm. The decision tree was illustrated with 2 example defects dissimilar to the defects under consideration. During the posttesting period, each resident was retested with the same defect diagrams and the same allotted time (10 minutes). Diagrams of compartment No. 1, a summary of compartment No. 2, RSTLs, facial subunits, and appropriate equipment (compass, protractor, ruler) were supplied.

The cosmetic outcomes of the reconstructive decisions on the pretest and posttest facial defects were rated on a scale of 0 to 5 (with 0 indicating poor; 5, excellent) (Table 2). The tests were scored in accordance with this rating, with the best possible score being 42. The residents were also asked to rate the paradigm as not helpful, helpful, or very helpful at the end of the field testing. A paired t test was used to assess the data for statistical significance.

Three resident physicians from each of the 4 otolaryngology training years (n=12) were tested. The individual pretest and posttest scores are shown in Table 3. The average scores for each postgraduate training year are shown in Table 4. Reconstructive choices between pretest and posttest scores were significantly improved (P<.001). All the residents rated the reconstructive facial paradigm as helpful.


The rapid and consistent improvement of the resident physicians' reconstructive decisions indicates the usefulness of this paradigm for teaching novice surgeons. It is also of use to experienced surgeons because it can be adapted to reflect their practice preferences.


This paradigm provides a technique for teaching the complex decision making required for consistent and successful functional and aesthetic reconstruction of facial skin defects. It can be adapted to reflect individual surgeons' practice preferences and is easily followed by novice surgeons. A standardized framework such as this for decision making enhances the training of neophyte reconstructive surgeons by providing them with a clear understanding of basic considerations in choosing a flap. Using this understanding as a base, beginning surgeons can devote energy to developing the technical facility and artistic nuances that distinguish skilled reconstructive surgeons.

Back to top
Article Information

Accepted for publication October 8, 1997.

This paper was presented in part at the Annual Meeting of the American Academy of Facial Plastic and Reconstructive Surgery in New Orleans, La, September 16, 1995.

Corresponding author: Karen H. Calhoun, MD, Department of Otolaryngology, University of Texas Medical Branch, Galveston, TX 77551 (e-mail:

Parker  SLTong  TBolden  SWingo  PA Cancer statistics, 1996. CA Cancer J Clin. 1996;465- 27Article
Young  VLBartell  TH Burns of the face. Thomas  JRHolt  GReds.Facial Scars: Incision, Revision, and Camouflage St Louis, Mo CV Mosby Co1989;229- 270