Pharyngeal Flap and the Internal Carotid in Velocardiofacial Syndrome

ORIGINAL ARTICLE Pharyngeal Flap and the Internal Carotid in Velocardiofacial Syndrome

Sherard A. Tatum III, MD; JaKwei Chang, MD; Natalie Havkin, MS; Robert J. Shprintzen, PhD

nternal carotid artery anomalies have been documented as a common clinical feature in velocardiofacial syndrome. There has been some controversy over the need for preoperative imaging procedures, such as magnetic resonance angiography, when planning pharyngeal surgery for correcting velopharyngeal insufficiency. The purpose of this article is to Idescribe 20 patients with velocardiofacial syndrome who received comprehensive evaluation and underwent pharyngeal flap surgery within a 2-year period and to report the technique used for dissecting the flap and the surgical outcomes. Anomalies of the major neck vessels were present in all cases, but 5 of these 20 cases had particularly severe anomalies of the internal carotid arteries that placed the vessels directly deep within the donor site for the pharyngeal flap. Surgery was car- ried out successfully in all 20 cases using a modified approach after radiographic imaging was performed to locate the arteries. In the 5 cases with severe malpositioning of the internal carotid arteries, it was clear that the vessels could have been injured had their location not been identified and the surgical approach modified to avoid them. Arch Facial Plast Surg. 2002;4:73-80

Velocardiofacial syndrome (VCFS) is the standard angiography to demonstrate ec- most common multiple anomaly syn- topic and medial placement of the inter- drome associated with cleft palate, con- nal carotid arteries in 3 cases selected spe- stituting 8% of patients with cleft palate,1 cifically because of previous observations including overt, submucous, and occult from nasopharyngoscopy that showed submucous cleft palate. Although the fre- prominent arterial pulsations in the pos- quency of VCFS among individuals with terior pharyngeal wall during workup for cleft lip is not known, cleft lip does occur pharyngeal flap surgery. The abnormal as a finding in the syndrome at least oc- placement of the arteries was considered casionally.2 It has been reported that ap- a contraindication to pharyngeal flap sur- proximately 5% of all patients at large in- gery in these cases.3 In a more compre- terdisciplinary cleft palate–craniofacial hensive study using magnetic resonance centers have VCFS.1,2 Because velopha- angiography (MRA), Mitnick et al5 as- ryngeal insufficiency (VPI) is such a com- sessed 19 consecutive patients with VCFS mon disorder in the spectrum of anoma- referred for pharyngeal flap surgery. The lies in VCFS, it is likely that many of the MRA results were correlated to findings patients will present for surgical manage- from nasopharyngoscopic examinations ment of hypernasal speech. for observations of visible pulsations in the Anomalies of the internal carotid ar- pharyngeal walls. It was found that obser- teries in VCFS were initially reported in vations of pulsations did not predict me- 1987.3,4 MacKenzie-Stepner et al3 used dial deviation of the internal carotid arteries, and medially deviated arteries did From the Division of Facial Plastic and Reconstructive Surgery, Departments of not always result in visible pulsations. Otolaryngology and Pediatrics (Dr Tatum); the Division of Neuroradiology, 5 Department of Radiology (Dr Chang); the Communication Disorder Unit (Ms Havkin); Mitnick et al concluded that some type and the Center for the Diagnosis, Treatment, and Study of Velo-Cardio-Facial of vascular imaging procedure was nec- Syndrome, Department of Otolaryngology and Communication Science essary before undertaking pharyngeal flap (Dr Shprintzen), State University of New York Upstate Medical University, Syracuse. surgery because the placement of the

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 4, APR-JUNE 2002 WWW.ARCHFACIAL.COM 73

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/28/2021 SUBJECTS AND METHODS Velopharyngeal insufficiency was assessed from clinical speech and language evaluation, multiview videofluoroscopy, and nasopharyngoscopy using the International SUBJECTS Working Group rating scale (Figure 1).8 The ratings of the components of velopharyngeal closure are listed in Table The study sample comprised 20 consecutively referred pa- 2 along with the degree of perceived hypernasality. tients with VCFS. All cases were confirmed by FISH (fluo- rescent in situ hybridization) to have a 22q11 deletion, and FLUOROSCOPIC AND ENDOSCOPIC ASSESSMENTS all patients were examined by the fourth author (R.J.S.) to confirm the clinical diagnosis. There were 11 male sub- Preoperatively, velar motion varied among the sample but jects and 9 female subjects, ranging in age from 4 to 17 years was rated above 0.5 for only 1 case (Table 2).8 A rating of (Table 1). This sample represents all cases of VCFS re- 0.5 indicates velar motion of half of the distance from the ferred for surgical management of VPI from the Center for rest position to the posterior pharyngeal wall. In 14 cases, the Diagnosis, Treatment, and Study of Velo-Cardio- velar motion was under 0.5. Poor or absent lateral pharyn- Facial Syndrome of the State University of New York Up- geal wall motion (a rating of 0.2 or lower) was found in all state Medical University, Syracuse, within a 2-year period cases except 1. In 2 cases, there was asymmetric lateral pha- (1998-1999). These 20 cases represented 25% of all cases ryngeal wall motion, with the right lateral pharyngeal wall of VCFS referred to the VCFS center within this 2-year pe- showing motion rated at 0.2 and the left lateral wall show- riod. The other cases were not referred for surgery for a ing no motion (0.0) in both cases. A rating of 0.2 indicates variety of reasons, including age (too young), no evidence motion less than half of the distance to the pharyngeal mid- of VPI (about 10% of the sample), refusal of additional sur- line. In all other cases, the lateral pharyngeal walls were gery because of previous failures elsewhere, or successful rated at 0.0 or 0.1 bilaterally. In such cases, very wide sub- treatment elsewhere prior to referral. obstructing pharyngeal flaps are recommended. All patients received a comprehensive evaluation that included MRA or computed tomography (CT) scanning, MAGNETIC RESONANCE ANGIOGRAPHY video nasopharyngoscopy, and multiview videofluoroscopy. Five patients had previously undergone other sur- Nineteen patients had an MRA prior to pharyngeal flap surgical procedures for VPI (Table 1). Four of the patients had gery. One patient required contrast-enhanced CT scan- previous failed sphincter pharyngoplasties, and 1 had a failed ning because of the presence of a pacemaker. In most cases, Furlow palate repair as a secondary procedure. In all 5 cases, MRA was performed within a week of surgery, but in a few hypernasality was not corrected by the surgery. cases MRA was done several months prior to admission. Of the total sample, 1 patient had an overt cleft of the The MRA protocol includes scanning of the entire head and secondary palate, 10 had obvious submucous clefts includ- neck and the upper chest to the aortic arch using 7-mm- ing bifid uvula, and 8 had occult submucous clefts (Table thick abutting slices. We also scan the spine because of the 1). One patient had no evidence of a cleft, and 2 patients frequency of tethered cord and other spinal anomalies in had asymmetric VPI related to pharyngeal hypotonia on the VCFS. The brain is also assessed from the magnetic reso- left (Table 2). The frequency of congenital heart anoma- nance imaging scans. The MRA is formatted in coronal, lies is also listed in Table 1. All patients had grossly nor- transverse, and sagittal views, and 3-dimensional recon- mal expressive language at the time of surgery. structions of the vessels are done as well. The common carotid, internal carotid, external carotid, and vertebral ar- ASSESSMENT PROCEDURES teries are all isolated in relation to their position within the pharyngeal soft tissues. One patient had a pacemaker, which All patients were evaluated by the interdisciplinary team necessitated the substitution of contrast-enhanced 3-di- at the Center for the Diagnosis, Treatment, and Study of mensional CT angiography instead of MRA. Velo-Cardio-Facial Syndrome. Evaluation procedures included the following: ASSESSMENT OF TONSILS 1. speech and language evaluation including a group rat- Previous reports have shown that tonsillectomy prior to pha- ing of nasal resonance on a 5-point scale (hyponasal, ryngeal flap surgery is an important component in the avoid- normal, mild, moderate, and severe hypernasality) ance of obstructive sleep apnea following surgery.9,10 Ton- 2. genetic/dysmorphologic evaluation sils were assessed using both videofluoroscopy and 3. cytogenetic/molecular genetic evaluation including FISH nasopharyngoscopy (Figure 2 and Figure 3). It has been 4. flexible fiber optic nasopharyngoscopic evaluation our experience that tonsillar hypertrophy is not always well 5. multiview videofluoroscopic evaluation in at least fron- recognized on oral examination. In cases of VCFS, the phar- tal and lateral views ynx (including both the oropharynx and nasopharynx) is 6. facial plastic surgery typically deep secondary to platybasia11 and a short, defi- 7. immunologic evaluation cient palate.12 When tonsils are assessed perorally using the 8. audiologic evaluation familiar scale (0 to 4+), the rating is based on the medial pro- 9. magnetic resonance angiography of the neck vessels jection of the tonsils. When the pharynx is deep, as in VCFS, and magnetic resonance imaging of the brain and spine it may be that the path of least resistance for tonsillar growth or CT scanning is posterior, posteroinferior, or posterosuperior. Previous re- 10. otolaryngologic evaluation ports have documented that tonsils can grow behind the pal- 11. a variety of evaluations from other disciplines, as needed, ate and faucial pillars.13,14 When this posterior growth oc- including endocrinology, neurology, cardiology, ne- curs, the tonsils can be seen on endoscopic examination phrology, and hematology. (Figure 2) and on fluoroscopic assessment when barium

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 4, APR-JUNE 2002 WWW.ARCHFACIAL.COM 74

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/28/2021 contrast is used (Figure 3). In our sample, when tonsils are 2-0 Chromic sutures on a tapered needle are passed seen posteriorly in the pharyngeal airway, they are always through the oral mucosa of the soft palate near the junction removed 6 or more weeks before pharyngeal flap surgery of the hard and soft palate into the pocket, brought out to avoid respiratory complications. If not removed, the en- through the pocket, and passed through the inferior edge larged tonsils would occlude both the pharynx as well as of the flap. They are then passed back through the soft pal- the lateral ports beneath the pharyngeal flap, which has been ate pocket and back out through the oral mucosal layer, ap- associated with the development of obstructive sleep approximately 5 mm away from the initial entry of the suture. nea.10 In addition, the presence of tonsils intruding into the Five sutures are equally placed along the soft palate and in- pharynx would reduce the ability to create a flap of ad- ferior edge of the flap from one corner to the next. They are equate width intraoperatively. not tied until all sutures are passed. These sutures are all tied, pulling the flap up into the pocket of the soft palate. SURGICAL TECHNIQUE Attention is then turned to the posterior pharyngeal wall defect. The muscle and mucosa on the inferior aspect of the The pharyngeal flap is intended to be as short a musculo- defect are elevated. This posterior pharyngeal wall elevation mucosal flap as possible; leaving a small donor site, which extends down into the hypopharyngeal region, freeing up the results in less throat discomfort and less circumferential nar- musculomucosal layer and allowing it to be advanced up- rowing of the pharynx after surgery.10,15 After induction of ward into the base of the flap where it is sutured to the pre- general endotracheal anesthesia, standard prepping and drap- vertebral fascia and muscle with 2-0 chromic sutures (stents ing is performed. Clindamycin, 10 mg/kg (maximum, 900 are generally not placed). At the termination of the proce- mg), and dexamethasone, 0.5 mg/kg (maximum, 12 mg), dure, neither the flap nor the posterior pharyngeal wall de- are administered intravenously.16 The Dingman mouth gag fect is visible through the mouth without superior retrac- is introduced and suspended with towel rolls on the chest tion of the soft palate. The nasal cavity, nasopharynx, and rather than the Mayo stand to reduce tongue ischemia. The oropharynx are all copiously irrigated, and all clots are re- gag is let down for 5 minutes every 30 minutes for ischemia moved. Oxymetazoline drops (0.05%) are instilled into the reduction as well.17 The soft palate and posterior pharyn- nasal cavity, and the patient is extubated in the operating room. geal wall are infiltrated with 1% lidocaine with 1:100000 epi- The patient is observed for several minutes for airway diffi- nephrine in preparation for the superiorly based flap. The culties before leaving the operating room. palate is typically not split but retracted superiorly to pro- vide adequate exposure for the surgery. POSTOPERATIVE CARE The desired flap width, having been predetermined by endoscopy and/or fluoroscopy, is then marked with a scal- The protocol at our institution following pharyngeal flap is pel on the posterior pharyngeal wall. The width selected to keep patients in the pediatric intermediate care unit for at for this group of patients was typically 75% to 100% of the least 1 postoperative day under apnea, heart rate, and pulse total posterior wall width. The length of the flap is deter- oximetry monitoring. Patients are sent to a regular pediatric mined by measuring the distance from the midsection of room once it is confirmed that they have not had any res- the soft palate to the posterior pharyngeal wall, while re- piratory complications or obstructive apnea. Their postop- tracting the free edge of the soft palate slightly toward the erative diet is advanced from clear liquids on the first post- posterior pharyngeal wall. The typical length of the flap is operative day to full liquids and a soft diet on the second 1.5 to 2 cm, extending no lower than the midlevel of the and third postoperative days, respectively. They continue to oropharynx. Once this length is determined, the trans- receive intravenous clindamycin until peroral intake is ad- verse portion of the incision is marked with a scalpel. equate. Dexamethasone is also given for 36 to 48 hours (0.25 The relationship of carotid pulsations is then noted rela- mg/kg every 8 hours). Oxymetazoline (0.05%) drops are tive to the mucosal markings and the imaging studies. The changed to isotonic sodium chloride solution after 24 to 48 incisions are carefully extended down through the constric- hours. Patients are discharged with a prescription for amoxi- tor musculature with a scalpel. Hemostasis is obtained with cillin-clavulanic acid and acetaminophen-codeine typically topical 1:1000 epinephrine and electric cautery. The dissec- 2 to 3 days after surgery. An outpatient follow-up visit is gen- tion is extended through the visceral fascia, leaving the alar erally scheduled for 7 to 10 days after surgery. fascia intact. The superiorly based myomucosal flap is then POSTOPERATIVE EVALUATION elevated in an inferior to superior direction in the retropha- ryngeal space, superficial to the danger space or preverte- Patients are asked to return to the center approximately 6 bral space. The internal carotid artery, if underlying the dis- months after surgery. Of this group of patients, 2 lived within section, is retracted laterally by direct pressure. The artery 20 miles (32 km) of the hospital, 2 within 120 miles (192 km), remains covered by the alar fascia. 4 within 250 miles (400 km), 4 within 500 miles (800 km), Blunt dissection with a Kitner dissector is performed as 6 within 2000 miles (3200 km), 1 had to travel from the West much as possible to elevate the flap off of the alar fascia. The Coast, and 1 was from the United Kingdom. Therefore, follow- flap is elevated at least to the level of the atlas if not higher. A up, though preferred at 6 months, 1 year, and 2 years, was not transverse incision is then made on the nasal surface of the always according to this schedule because of the distances in- soft palate beginning at the base of the uvula approximately volved.Allpatientshadclinicalspeechevaluationatthesesame 5 mm superior to the free edge on either side of the uvula and intervals, and we also evaluated all out-of-town patients by extending out laterally to the most lateral aspect of the soft videotape on semimonthly intervals. Nearly all patients were palate. The incision remains straight, not following the cur- in speech therapy under our prescription, and the speech pa- vature of the free edge of the soft palate own along the pala- thologists treating in local communities were instructed to topharyngeal fold. Angled scissors are used to create a pocket send videotapes with specific speech samples and spontane- through this incision into the substance of the soft palate ex- ous speech. Ratings of nasal resonance were made in the same tending close to the junction of the hard and soft palate. manner as the preoperative protocol.

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 4, APR-JUNE 2002 WWW.ARCHFACIAL.COM 75

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/28/2021 Table 1. Twenty Subjects With Velocardiofacial Syndrome According to Age, Cleft Type, History of Previous Surgery for Velopharyngeal Insufficiency, and the Presence of Congenital Heart Anomalies*

Case No. Age, y Cleft Type Previous Surgery Heart Anomalies 1 5 SMCP None . . . 2 4 SMCP None VSD, pulmonic stenosis 3 17 SMCP Sphincter pharyngoplasty . . . 4 15 . . . Sphincter pharyngoplasty . . . 5 9 SMCP None . . . 6 4 OSMCP None TOF 7 5 OSMCP None TOF 8 8 OSMCP Furlow palatoplasty . . . 9 6 OSMCP None VSD 10 4 OSMCP None . . . 11 6 OSMCP Sphincter pharyngoplasty . . . 12 5 SMCP None IAA, type B 13 5 OSMCP None TOF 14 4 CP None ASD 15 4 SMCP None . . . 16 4 SMCP None . . . 17 4 SMCP None . . . 18 5 SMCP Sphincter pharyngoplasty VSD 19 5 OSCMP None IAA, type B, VSD 20 5 SMCP None . . .

*SMCP indicates submucous cleft palate; VSD, ventriculoseptal defect; OSMCP, occult submucous cleft palate; TOF, tetralogy of Fallot; IAA, interrupted aortic arch; CP, cleft plate; ASD, atrial septal defect; and ellipses, not present.

Table 2. Perceptual Rating of Nasality Related to Rating of Velar Motion, Lateral Pharyngeal Wall Motion, and Posterior Pharyngeal Wall Motion Prior to Pharyngeal Flap Surgery*

Case No. Hypernasality Velar Motion Right Lateral Wall Motion Left Lateral Wall Motion Posterior Wall Motion 1 Severe 0.3 0.0 0.0 0.0 2 Severe 0.5 0.2 0.2 0.0 3 Severe 0.2 0.0 0.0 0.0 4 Moderate 0.2 0.2 0.0 0.0 5 Severe 0.5 0.0 0.0 0.0 6 Severe 0.2 0.0 0.0 0.0 7 Severe 0.4 0.0 0.0 0.0 8 Severe 0.3 0.0 0.0 0.0 9 Severe 0.2 0.0 0.0 0.0 10 Severe 0.5 0.1 0.1 0.0 11 Severe 0.0 0.0 0.0 0.0 12 Severe 0.0 0.0 0.0 0.0 13 Severe 0.2 0.1 0.1 0.0 14 Severe 0.5 0.0 0.0 0.0 15 Severe 0.2 0.1 0.1 0.0 16 Moderate 0.7 0.3 0.3 0.0 17 Severe 0.3 0.0 0.0 0.0 18 Severe 0.5 0.2 0.0 0.0 19 Severe 0.0 0.0 0.0 0.0 20 Severe 0.3 0.0 0.0 0.0

*Ratings are according to the standards set by the International Working Group (Golding-Kushner et al,8 1990).

arteries in several of their cases was directly within the of MRA studies. In a discussion of the Witt et al6 article, donor site of a pharyngeal flap. Shprintzen7 pointed out design flaws and lack of scientific Witt et al,6 using a questionnaire and anecdotal re- evidence. Shprintzen7 reinforced the scientific evidence from ports to determine if there had been any fatalities in pa- the original prospective research of Mitnick et al,5 point- tients with VCFS during pharyngeal flap surgery, re- ing out hard scientific data that supported the necessity of ported the absence of data to support the notion that internal preoperative vascular imaging in patients with VCFS. carotid anomalies warranted preoperative MRA in pa- The purpose of this article is to describe 20 con- tients with VCFS. Based on a sample of 30 surgeons (se- secutive patients with VCFS who had pharyngeal flap sur- lection criteria were not reported), they indicated that the gery within a 2-year period. The surgical technique, modi- absence of reported deaths or bleeding complications was fications, outcomes, and the intraoperative status of the sufficient evidence to recommend against the added costs internal carotid arteries are also reported.

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 4, APR-JUNE 2002 WWW.ARCHFACIAL.COM 76

LPW LPW v p

C D v p Anterior t

PPW

Figure 1. Assessment of velopharyngeal closure in a patient using both multiview videofluoroscopy in lateral (A), frontal (B), and base (C) projections and flexible fiberoptic nasopharyngoscopy (D). In A and D, a indicates adenoid; p, posterior pharyngeal wall; and v, velum. The arrows in B mark the lateral pharyngeal walls (LPW). In C, the posterior pharyngeal Figure 3. Hypertrophic tonsils (t) seen in lateral view videofluoroscopy wall (PPW) is marked by the white arrow. behind the velum (v) and positioned posteriorly in the oropharynx.

A B p

t Lateral t Pharyngeal Wall v v Flap

Figure 2. Hypertrophic tonsils (t) as seen in the oropharynx during nasopharyngoscopy in 2 patients (A and B) with velocardiofacial syndrome. The tonsils are interposed between the velum (v) and posterior pharyngeal wall (p).

RESULTS

SPEECH RESULTS Lateral Port Figure 4. Wide pharyngeal flap as seen endoscopically 1 year following Hypernasal resonance and abnormal nasal air escape dur- surgery. The lateral ports are patent, but small. ing speech was successfully eliminated in 18 of 20 cases. Hyponasality is a typical short-term finding, usually per- CAROTID ARTERY PLACEMENT sisting for 6 to 12 months. With growth, hyponasality has decreased, and resonance balance has normalized in The radiographic studies performed prior to surgery all cases that are more than 1 year postoperative. Post- proved to be highly predictive of potential problems re- operative endoscopic assessment of velopharyngeal clo- lated to the abnormal placement of the internal carotid sure has shown very wide pharyngeal flaps in all 18 cases arteries. Five cases had severe medial displacement of the that had a successful outcome (Figure 4). In the 2 cases arteries potentially placing them directly within the do- with some residual nasal air escape and hypernasality dur- nor site for the flap (Figure 5). Table 3 shows the place- ing speech, 1 lateral port was noted to be wider than the ment of the arteries as noted during surgery in relation other, resulting in a unilateral VPI. Of interest, in both to endoscopic and MRA or CT assessments. In 10 cases, of these cases with unilateral VPI, a prominent pulsa- the level of maximum medial deviation was at or above tion of the internal carotid artery was seen in the wider the base of C1. Although prominent pulsations were seen port. Although there was improvement in the speech in in the posterior pharyngeal wall in 13 cases, surgical ex- these 2 cases, there was residual VPI and hypernasality, posure of the artery beneath the alar fascia occurred in and they are considered to be failures.18 Of the last 10 5. In the cases with the most significant medial devia- cases undergoing pharyngeal flap, there has been a 100% tion of the arteries, the placement was also close to the resolution of abnormal hypernasality and VPI (Table 2). mucosal undersurface of the posterior pharyngeal wall.

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 4, APR-JUNE 2002 WWW.ARCHFACIAL.COM 77

Pharynx

Internal Carotid

Figure 5. Medially deviated internal carotid arteries in a coronal magnetic resonance angiography section (A) that are located close to the mucosal surface as seen in an axial view (B). These images are from case 15 in Table 3.

Table 3. Carotid Pulsations as Seen on Endoscopy in Relation to Magnetic Resonance Angiography or Computed Tomography Evaluation and Findings at Surgery*

Pulsation on Radiographic Evaluation Endoscopic Intraoperative Carotid Case No. Evaluation 25% to Midline 50% to Midline At or Near Midline Level of Medial Deviation Visualization in Flap Donor Size 1 Yes Yes Mid C1 Yes 2 No Yes Mid C2 No 3 No Yes Base of C1 No 4 No Yes Base of C1 No 5 No Yes TopofC2 No 6 Yes Yes Mid C2 No 7 Yes Yes Base of C1 Yes 8 No Yes Base of C2 No 9 Yes Yes Top of C2 No 10 Yes Yes Base of C1 No 11 No Yes Base of C1 No 12 Yes Yes Mid C1 Yes 13 Yes Yes Base of C1 Yes 14 No Yes Mid C2 No 15 Yes Yes Base of C1 Yes 16 Yes Yes Base of C1 No 17 Yes Yes Base of C2 No 18 Yes Yes Base of C2 No 19 Yes Yes Mid C2 No 20 Yes Yes Top of C2 No

*Carotid artery placement is reported radiographically according to its location in relation to the pharyngeal midline and the vertical plane of the maximum displacement in relation to the cervical vertebrae.

This abnormal position of the internal carotid arteries ing complications from the flap donor site. One patient placed the arteries directly within the operative field for developed a nosebleed unrelated to the surgery that re- the donor site of the pharyngeal flap. quired a return to the operating room to rule out the possibility that the bleeding was from the flap or soft palate. POSTOPERATIVE COURSE The total length of hospital stay was 3 days for this pa- AND COMPLICATIONS tient. Several patients had loud snoring in the immedi- ate postoperative period without associated apnea or oxy- The average length of postoperative stay was 2.7 days, gen desaturations based on the monitoring protocol. The ranging from 2 days to 6 days. Ten patients were dis- snoring tended to reduce markedly by the 14th postop- charged after 2 postoperative nights, 9 after 3 postoperative day. Obstructive sleep apnea has not been ob- erative nights, and 1 after 6 nights. There were no bleed- served in any of the 20 cases during hospitalization nor

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 4, APR-JUNE 2002 WWW.ARCHFACIAL.COM 78

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/28/2021 following discharge. There have been no bleeding com- surgery or discontinued the operation when the promi- plications and no transfusions. There were no other post- nent vessel pulsations were observed intraoperatively. Al- operative complications. though all patients assessed to be at risk intraoperatively had visible pulsations preoperatively, not all patients COMMENT with preoperatively visible pulsations were assessed to be at risk. This finding may be related to alteration in ca- Velocardiofacial syndrome is the most common syn- rotid position from the typical sitting position for en- drome of clefting, constituting a high percentage of the doscopy to the supine neck–extended position for im- disorders treated in cleft palate centers. Speech disor- aging or surgery. ders, VPI, and hypernasality in particular, are more preva- The success rate with the specific procedure de- lent among patients with VCFS than in other patients with scribed in this report is acceptable, especially when one clefts. It is likely that there is a disproportionately in- considers that the 2 failures occurred in the earliest of creased frequency of cases of VCFS among patients un- the surgical procedures performed. As is often true with dergoing surgical correction of VPI compared with the any modification of a procedure, there is a learning curve, overall frequency of patients with VCFS in the cleft popu- and once the goals and techniques of surgery are relation. Because patients with VCFS have been docu- fined, implementation becomes easier over time. In the mented to have significant pharyngeal hypotonia (sup- final 12 patients in the series, there were no failures, and ported by our findings of poor lateral pharyngeal wall the goal for elimination of hypernasality was achieved motion in the present study), it is also likely that a rela- in all cases. Because almost all of these patients demon- tively high percentage of failures to resolve VPI are in pa- strated poor or absent motion in 1 or both lateral pha- tients with VCFS. ryngeal walls (19 of 20 were bilaterally hypotonic), the Abnormal placement of the internal carotid arter- goal in nearly all cases was to place a subobstructing flap. ies has been well documented in VCFS, leading to the In 1 of the 2 cases in which VPI and hypernasality were recommendation for imaging of the cervical vessels prior not completely eliminated, it was noted that there was a to reconstructive pharyngeal surgery.5 However, in a study prominent pulsatile vessel (the internal carotid artery) critical of that recommendation, Witt et al6 protested the directly in the lateral port that persistently impinged on financial cost to the medical care system that would be the lateral edge of the flap. It is unclear if the vessel caused induced by the addition of MRA to the preoperative evalu- that port to be stented open because of the constant pres- ation. In a response critical of the manner in which Witt sure of the artery against the healing flap, or if the prob- and colleagues reached their conclusion, Shprintzen7 sug- lem was some other unanticipated result of healing. How- gested that the risk of encountering abnormally placed ever, in the later procedures in the series, the problem internal carotid arteries during surgery far outweighed was not encountered. In both of the suboptimal cases, any potential cost issues. there was a unilateral insufficiency: 1 lateral port clos- Based on our findings in the present study, it is clear ing completely and 1 lateral port closing only partially that preoperative MRA or CT angiography is essential in with speech. In both of these cases, even though hyper- patients with VCFS who are to undergo pharyngeal flap nasality was markedly diminished, it was not com- surgery. In this series, 5 (25%) of the 20 cases had arter- pletely eliminated and therefore could not be catego- ies that were within the donor site dissection. The proce- rized as a completely successful outcome.17 It should be dure for pharyngeal flap described in this article is spe- noted that 5 of these 20 patients had already experi- cifically designed to raise a very short musculomucosal flap enced surgical failure with other procedures. leaving the alar fascia down, thus exposing the smallest There were essentially no significant postoperative possible segment of the carotids to potential injury. Even complications in this series, including no evidence of ob- so, without specific knowledge of the placement of the ar- structive sleep apnea either short or long term. Al- teries in these patients, the possibility for injury of the in- though respiration had initially been altered in many cases ternal carotids was high in these 5 cases. As reflected in from predominantly nasal to oral, in most cases there was Table 3, in those 5 cases with the most medial placement a gradual increase in the ability to exchange air nasally of the arteries, the vessels were located very close to the after 6 months. However, in no cases was the continued mucosa of the posterior pharyngeal wall. The danger of use of predominantly oral respiration accompanied by this ectopic placement of the arteries is compounded by exercise intolerance or decreased vitality. The absence an abnormally thin pharyngeal muscle wall in patients with of postoperative complications in this sample is related VCFS. Therefore, surgical dissection in these cases must to 2 factors that have been previously reported in the lit- be done very carefully. Other pharyngeal flap procedures erature: the removal of tonsils and the confinement of require that very long and very wide flaps be raised, some- the flap donor site to a short area almost entirely within times extending to the hypopharynx. In such proce- the upper oropharynx and nasopharynx.9,10 This short dures, there is no doubt that without specific knowledge flap donor site, closed vertically, prevents narrowing of of the placement of the internal carotid arteries, they would the airway beneath the flap so that even if nasal respira- have been at risk to be injured or severed during surgery tion is partially compromised, oral respiration is unim- if performed in the 5 cases with arteries that were within peded in the pharynx. the donor site dissection in our series. The elimination of VPI was followed by intensive Our series may have an ascertainment bias with re- articulation therapy that was checked frequently by speech spect to these arteries because several patients were re- evaluation for local patients and videotape for those com- ferred only after other surgeons had recommended against ing from a distance. Speech therapy to eliminate abnor-

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 4, APR-JUNE 2002 WWW.ARCHFACIAL.COM 79

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/28/2021 mal articulatory compensations using specific tech- H. Abnormal carotid arteries in the velocardiofacial syndrome: a report of three niques designed to eliminate glottal stop substitutions has cases. Plast Reconstr Surg. 1987;80:347-351. 4. D’Antonio LL, Marsh JL. Abnormal carotid arteries in the velocardiofacial syn- been rapidly successful in nearly all cases and resulted drome. Plast Reconstr Surg. 1987;80:471-472. in a complete normalization of speech intelligibility, ar- 5. Mitnick RJ, Bello JA, Golding-Kushner KJ, Argamaso RV, Shprintzen RJ. The ticulation, and resonance.19 use of magnetic resonance angiography prior to pharyngeal flap surgery in patients with velo-cardio-facial syndrome. Plast Reconstr Surg. 1996;97:908- 919. CONCLUSIONS 6. Witt PD, Miller DC, Marsh JL, Muntz HR, Grames LM. Limited value of preoperative cervical vascular imaging in patients with velocardiofacial syndrome. Plast The specific type of pharyngeal flap surgery described in Reconstr Surg. 1998;101:1184-1195. the present study has been highly successful in eliminat- 7. Shprintzen RJ. Discussion: limited value of preoperative cervical vascular imaging VPI in a patient population who have been consid- ing in patients with velocardiofacial syndrome. Plast Reconstr Surg. 1998;101: ered to be at surgical risk and who have had a signifi- 1196-1199. 8. Golding-Kushner KJ, Argamaso RV, Cotton RT, et al. Standardization for the re- cant failure rate. Preoperative imaging of the major neck porting of nasopharyngoscopy and multiview videofluoroscopy: a report from arteries is recommended for all patients with VCFS who an International Working Group. Cleft Palate J. 1990;27:337-347. are to undergo pharyngeal flap surgery. 9. Shprintzen RJ. Pharyngeal flap surgery and the pediatric upper airway. Int An- esthesiol Clin. 1988;26:79-88. Accepted for publication April 10, 2001. 10. Shprintzen RJ, Singer L, Sidoti EJ, Argamaso RV. Pharyngeal flap surgery: postoperative complications. Int Anesthesiol Clin. 1992;30:115-124. This study was supported in part by funds from the Chil- 11. Arvystas M, Shprintzen RJ. Craniofacial morphology in the velocardiofacial syn- dren’s Miracle Network Telethon and by grants and dona- drome. J Craniofac Gen Dev Biol. 1984;4:39-45. tions to the Joseph and Annette Cooper Fund for Research 12. Golding-Kushner KJ. Craniofacial Morphology and Velopharyngeal Physiology in Velo-Cardio-Facial Syndrome at State University of New in Four Syndromes of Clefting [dissertation]. New York: Graduate School and York Upstate Medical University and by grant University Center, City University of New York; 1991. 13. Shprintzen RJ, Sher AE, Croft CB. Hypernasal speech caused by hypertrophic 5PO1HD34980-03 from the National Institutes of Health, tonsils. Int J Pediatr Otorhinolaryngol. 1987;14:45-56. Bethesda, Md (Dr Shprintzen). 14. MacKenzie-Stepner K, Witzel MA, Stringer DA, Laskin R. Velopharyngeal insuf- Corresponding author: Sherard A. Tatum III, MD, Di- ficiency due to hypertrophic tonsils: a report of two cases. Int J Pediatr Otorhi- vision of Facial Plastic and Reconstructive Surgery, Depart- nolaryngol. 1987;14:57-63. ments of Otolaryngology and Pediatrics, State University 15. Argamaso RV. The pharyngeal flap in cleft lip and palate. In: Kernihan DA, Ro- senstein SW, eds. Cleft Lip and Palate: A System of Management. Baltimore, Md: of New York Upstate Medical University, 750 E Adams St, Williams & Wilkins; 1990:263-269. Syracuse, NY 13210 (e-mail: [email protected]). 16. Senders CW, Emery BE, Sykes JM, Brodie HA. A prospective, double-blind, ran- domized study of the effects of perioperative steroids on palatoplasty patients. Arch Otolaryngol Head Neck Surg. 1996;122:267-270. REFERENCES 17. Senders CS, Eisele JH. Lingual pressure induced by mouthgags. Int J Pediatr Otorhinolaryngol. 1995;33:53-60. 1. Shprintzen RJ, Siegel-Sadewitz VL, Amato J, Goldberg RB. Retrospective diag- 18. Shprintzen RJ. Conceptual framework for pharyngeal flap surgery. In: Bardach noses of previously missed syndromic disorders among 1,000 patients with cleft J, Morris H, eds. Management of Unilateral Cleft Lip and Cleft Palate. Philadel- lip, cleft palate, or both. Birth Defects Orig Artic Ser. 1985;21:85-92. phia, Pa: WB Saunders; 1990:806-809. 2. Lipson AH, Yuille D, Angel M, Thompson PG, Vandervoord JG, Beckenham EJ. 19. Golding-Kushner KJ. Treatment of articulation and resonance disorders associ- Velocardiofacial (Shprintzen) syndrome: an important syndrome for the dys- ated with cleft palate and VPI. In: Shprintzen RJ, Bardach J. Cleft Palate Speech morphologist to recognize. J Med Genet. 1991;28:596-604. Management: A Multidisciplinary Approach. St Louis, Mo: Mosby–Year Book Inc; 3. MacKenzie-Stepner K, Witzel MA, Stringer DA, Lindsay WK, Munro IR, Hughes 1995:327-351.

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 4, APR-JUNE 2002 WWW.ARCHFACIAL.COM 80