Palatal fistula is a difficult complication after cleft palate repair. The repair of a palatal fistula can be challenging, particularly in wide and recurrent fistulas (
Figure 1). Large defects after cleft palate repair produce various symptoms, including regurgitation of fluid into the nasal cavity, hearing loss, and velopharyngeal insufficiency. In these cases, the palatal tissue around the fistula can be quite scarred and in short supply. A variety of reconstructive options are commonly employed, using local flaps of muscle and mucosa or tongue tissue or using distant flaps [
1,
2,
3]. The combination of buccal mucosal flaps and buccinator muscle as an axial myomucosal flap based on the facial artery has been described by Pribaz et al. [
2].
This flap consists of mucosa, submucosa, part of the buccinator and orbicularis muscles, and the facial artery with its venous plexus. This is known as the facial artery musculomucosal (FAMM) flap [
2]. The FAMM flap can be designed as an anteriorly based (for anterior fistula repair) or posteriorly based (for posterior fistula repair) flap. The facial artery is a branch of the external carotid artery and enters the face at the lower border of the mandible at the anterior border of the masseter muscle. It then passes superiorly and anteriorly following a tortuous course. It lies deep to the zygomatic muscles and superficial to the buccinator and the levator anguli oris muscles.
Abundant branches pass into the skin and mucosa through these facial muscles but some branches go directly into the skin (
Figure 2) [
4,
5]. The facial artery most commonly bifurcates at the mouth to become the lateral nasal artery and superior labial artery. The facial vein takes a separate course from the artery in a deeper plane and it is not included in local flaps (
Figure 3). In palatal fistulas, the FAMM flap is used for oral mucosa reconstruction. However, to achieve two-layer repair of the fistula, the nasal lining is repaired by folding the flap on itself, using pharyngeal flaps, local turndown flaps from oral mucosa, or split-thickness skin grafts. For large fistulas, free flaps can be used.
Cadaveric studies of the facial artery have suggested an area of skin at the nasolabial fold level that may be harvested with muscle and mucosa based on several cutaneous perforators as a facial artery composite flap [
6,
7,
8,
9]. There are other uses of composite flaps, including free flaps as the nasolabial composite flap with buccal mucosa described by Sasaki et al. for eyelid reconstruction [
7]. This flap provides an additional skin component for eyelid lining reconstruction; however, it requires microvascular technique.
To our knowledge, there are no reported clinical series evaluating outcomes of composite pedicle flaps (including skin component) based on the facial artery in palatal fistula repair. In addition, all of these flaps were described as arterialized flaps based on the facial artery; more accurately, these flaps are all based on the distal and major branch of the facial artery: the lateral nasal artery. The facial artery is only included at the distal end for anteriorly based flaps or the proximal end for posteriorly based flaps. This artery ends at the level of the alar base and becomes a cutaneous perforator, the inferior alar branch and superior alar branch (
Figure 2) [
6].
We present here the nasal artery musculomucosal cutaneous (NAMMC) flap, which can be based anteriorly or posteriorly, as a reliable surgical option in the management of large palatal fistulas. This technique provides a pedicle composite flap for large palatal fistula reconstruction in a single surgical procedure (
Figure 4 and
Figure 5). The skin flap component reduces the size of myomucosal component and avoids the use of pharyngeal flaps for nasal mucosa reconstruction.
The Outreach Surgical Center Program Lima is supported by ReSurge International and Smile Train Foundation from United States and receives patients from all the country to be operated in three different hospitals (as reference centers) in Peru.
Patients and Methods
Clinical Cases
This study was performed by the Outreach Surgical Center Program based in Lima, Peru. Twelve patients (seven male and five female) with palatal fistula after primary cleft palate repair were operated using the NAMMC-based anterior or posterior flap between March 2007 and December 2010 in three different hospitals in Peru. All of these patients had a large palatal fistula, defined as a defect wider than 2 cm (
Table 1). We have an established recall system for patient evaluation after surgery at 1, 6, and 12 months postoperatively. Average follow-up was 17 months (range, 6 to 28 months).
Surgical Technique
The surgery is performed under general anesthesia with oral endotracheal intubation. The course of the facial artery is mapped from the lower border of the mandible to the nasal alar base using a Doppler and physical exam (palpation;
Figure 6). A Dingman mouth gag is used to facilitate exposure. The fistula margins are infiltrated with lidocaine plus adrenaline and dissected 2 to 3 mm around the fistula as a turndown flap from oral mucosa. This procedure is not adequate to close the nasal layer in larger fistulas. In the design of the NAMMC flap, the width of the palatal oral mucosa defect is used as the width of the mucosa component of the flap, and the width of the palatal nasal mucosa defect left after dissection of the turndown flaps is used to define the size of the skin component of the flap.
After identifying the location of the artery and determining the size of the flap, we mark the oral mucosa using methylene blue. Incisions are made following the markings using a number 15 blade. The first incision is made through the mucosa and muscle tissue following the anterior border of the flap. The upper labial artery is easy to identify at the level of the oral commissure; after this, the facial artery can be located following this labial branch. In older patients, palpation of the artery pulse can be used with the same purpose. The superior labial artery must be ligated to release the flap. The facial artery should not be dissected to avoid any injury, and some subcutaneous tissue should remain over the artery. A branch of the buccal nerve (from the facial nerve) is usually observed during this dissection and can be carefully retracted, avoiding nerve injury. Then, the flap can be raised, making the posterior border incision and preserving the facial artery.
Posteriorly Based Flap
The length of the skin component of the flap is the distance between the most anterior and posterior point of the nasal mucosa defect. The length of the muscle mucosa component in the posteriorly based flap is the same as the distance from the most anterior point of the oral mucosa defect to the pivot point. The pivot point in this case is located behind the maxillary arch (retromolar trigone). The lateral nasal artery is identified and divided at the distal border of the flap and distal to the cutaneous perforator.
A main perforator is located a few millimeters before the distal division of the lateral nasal artery (below the alar base) and can be used in posteriorly based flap (
Figure 2). This perforator should not be dissected to avoid its injury, and the subcutaneous tissue is dissected around it in connection with the skin island and the musculomucosal component of the flap. The skin island is located over the nasolabial fold and below the alar base.
Anteriorly Based Flap
The length of the muscle mucosa component in the anteriorly based flap is the same as the distance from the most distal point of the oral mucosa defect to the pivot point. The superior labial artery must be ligated to release this type of flap. The pivot point for the anteriorly based flap is located at the level of the upper border of the alveolar cleft. The facial artery is identified and divided at the distal border of the flap and distal to the cutaneous perforator. Several small perforators are located at the distal portion of the flap, and the subcutaneous tissue is dissected around them in connection with the skin island and the musculomucosal component of the flap. The skin island is located over the nasolabial fold and above the oral commissure. Finally, in both anteriorly and posteriorly based flaps, the flap should be designed anterior and inferior to the level of the parotid duct to avoid any injury. The island skin component is located at the distal portion of the flap in both types.
The skin component is incised and dissected in connection with the muscle mucosa component including subcutaneous tissue around the lateral nasal artery skin perforators. It is only necessary to include one perforator. The composite flap is then dissected in the subcutaneous level, maintaining the artery and ramifications within the pedicle until the pivot point (
Figure 7). Once completely raised, the flap is inset and the donor defect closed primarily with absorbable sutures. If the flap is too bulky, division and insetting of the flap in a second session is sometimes required (
Figure 8). The flap is an axial pattern flap designed along and including the length of the facial artery, based either inferiorly with orthograde flow or superiorly with retrograde flow. The postoperative regimen includes a soft diet, and the patient is usually discharged from the hospital in 24 to 48 hours.
Results
Clinical Cases
The postoperative results were evaluated subjectively at each patient examination to assess the results. Demographic data and reconstruction details are listed in
Table 2. Eighty-three percent of the patients (10) returned for postoperative evaluation at 1 month and only 3 (25%) were evaluated at 6 months postoperatively. Eight anteriorly and four posteriorly based NAMMC flaps in 12 patients are presented. The mean age at the time of operation was 5.3 years (range, 2 to 8 years). None of the patients had undergone previous attempts at closure of palatal fistulae. NAMMC flap sizes varied from 6 to 4.5 cm in length and 2 to 2.5 cm in width. The skin portion of the flap varied from 2 to 3 cm in length and 1 to 1.5 cm in width. The donor site could be closed primarily without tension and with good cosmetic results. There were no complaints of skin donor site scarring. Complications are presented in
Table 3. We have seen short-term affection of the facial animation in 41.6% of the patients during the first months. Only one patient required a secondary procedure; all other flaps achieved resolution of the fistula. There was one case of partial necrosis: a 20% flap loss was encountered, due to venous congestion. In this case, the wound healed secondarily with no fistula recurrence. The partial flap failure rate is 8.3% (1/12).
The only other complication was a wound dehiscence in a patient with a bilateral wide anterior oronasal fistula. The anterior defect was closed using local flaps. We have no seen facial nerve injury in these cases.
The donor sites were closed primarily, and there were no postoperative complications. There was minimal deformation of the donor site (
Figure 17 and
Figure 18). Three representative cases follow.
Case 11
An 8-year-old boy with a history of bilateral cleft lip and palate presented with an anterior fistula measuring 33 × 19 mm after palatal surgery (
Figure 9). The fistula was closed with a left anteriorly based NAMMC flap (
Figure 10).
Case 9
A 7-year-old boy with a history of bilateral cleft lip and palate presented with an anterior fistula measuring 40 × 28 mm after palatal surgery (
Figure 11). The fistula was closed with a left anteriorly based NAMMC flap (
Figure 12,
Figure 13 and
Figure 14).
Case 4
An 8-year-old boy with a history of isolated cleft palate presented with a large fistula measuring 35 × 19 mm after palatal surgery (
Figure 15). The fistula was closed with a left posteriorly based NAMMC flap (
Figure 16 and
Figure 19).
Discussion
Numerous flaps based on the facial artery are well known and have widespread use. These flaps are supplied by perforators from the facial artery. Their venous perfusion is achieved by the venous vasa vasorum accompanying the facial artery. Therefore, this flap does not require the facial vein. However, we have found in many studies [
4,
6,
8] that the facial artery usually bifurcates at the oral commissure and becomes two major branches: the superior labial artery and the lateral nasal artery (70.8%).
We consider the angular artery usually to be a thin branch rather than a main continuation of the facial artery, as Nakajima et al. [
8] described. Under this pattern, the lateral nasal artery is the main facial artery distal branch. In the flap described here (NAMMC flap), the main blood supply comes from the lateral nasal artery, a terminal branch of facial artery. In a previous study [
6], we found that this artery runs toward the alar base and provides a skin perforator a few millimeters before the inferior alar branch. In this cadaveric study, we found multiple facial artery perforators that could be identified and dissected to their origin from the facial artery (
Figure 20).
In ~1 to 10% of cases, the facial artery is missing in the nasolabial area, terminating in the vicinity of the angle of the mouth [
5]. However, the facial artery was present in the nasolabial area in all cases in our study [
6]. Therefore, before initiating this procedure, the position and suitability of the blood vessels should be confirmed by Doppler. Another important consideration is that the facial artery has a very tortuous course in this region; therefore, care must be taken to include the arterial axis in the flap. Closure of large palate fistulae is a challenging problem, and a variety of regional and distant flaps have been described.
Flaps including the facial artery and all layers from the skin to the mucosa have been used as pedicled flaps for reconstruction of the lower lip, nose, and lower eyelid but not for palatal fistula reconstruction [
7,
9,
10]. The FAMM flap has some advantages over other regional flaps: it includes good blood supply, it requires only a single stage, and it involves less risk of flap dehiscence. However, in larger palate fistula repair, this flap is limited for nasal layer reconstruction, and additional locoregional flaps (extended turndown flap from oral mucosa, pharyngeal flap) or skin grafts are usually necessary.
The nasal layer repair reduces the recurrence of palatal fistulas. The FAMMC flap (named by us as NAMMC flap) provides a pedicle composite flap for large palatal fistula reconstruction, reduces the size of myomucosal component, reduces the surgical dissection over the maxilla, and avoids the use of pharyngeal flaps for nasal mucosa reconstruction. The NAMMC flap as a single-stage operation is a better option than a two-stage flap or microvascular reconstruction. It is associated with minimal donor site morbidity (
Figure 14,
Figure 17 and
Figure 18). Observed complications were related mainly to the postoperative pain and scar disorders. We believe this composite flap is a good surgical choice for recurrent and wide hard palate fistulae and any other difficult secondary palate repair.
Conclusions
A modification of a common technique (FAMM flap) is presented here. The FAMM cutaneous flap is a reliable and safe procedure and can be used as an alternative for the closure of difficult palatal fistulas. Because the main blood supply comes from the lateral nasal artery, this flap should be named as “nasal artery musculomucosal cutaneous flap”. The aesthetically pleasing donor site offers a versatile flap because of the presence of perforators with a large arc of rotation.