Degrees of hypernasality in the control group (A) and the velocardiofacial syndrome group (B) after pharyngoplasty. Dark blue indicates always; light blue, sometimes; and light grey, never. X indicates insufficient data.
Closure patterns in the control group (A) and the velocardiofacial syndrome group (B) after pharyngoplasty. Dark blue indicates never; dark grey, sometimes; light blue, often; and light grey, always. X indicates insufficient data.
Widdershoven JCC, Stubenitsky BM, Breugem CC, MinkvanderMolen AB. Outcome of Velopharyngoplasty in Patients With Velocardiofacial Syndrome. Arch Otolaryngol Head Neck Surg. 2008;134(11):1159-1164. doi:10.1001/archotol.134.11.1159
Copyright 2008 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2008
To compare the outcomes of surgical correction of velopharyngeal insufficiency (VPI) in patients with velocardiofacial syndrome (VCFS) and a non-VCFS group.
Twenty-five patients with VCFS (16 girls and 9 boys) underwent palatal lengthening for VPI between 1986 and 2001. The mean age at surgery was 6.4 years. Revision was defined as the need for secondary sphincter pharyngoplasty as determined by speech investigation, nasal endoscopy, and acoustic nasometry. A comparison was made to a control group made up of a randomized group of patients without VCFS who underwent palatal lengthening for VPI (32 patients: 10 girls and 22 boys).
Wilhelmina Children's Hospital, a tertiary referral center in Utrecht, the Netheralands.
A total of 57 patients who underwent palatal lengthening for VPI, 25 with VCFS and 32 without VCFS.
Primary surgery consisted of a palatal lengthening technique. If revision was needed, a sphincter pharyngoplasty was carried out.
Main Outcome Measures
Pharyngeal function was assessed using perceptual speech investigation, nasal endoscopy, and acoustic nasometry.
In the VCFS group, 16% of the patients required surgical revision (4 of 25). These patients were slightly older at the time of primary surgery than those who did not require surgical revision (mean age, 6 vs 5.5 years). In the control group, no patients required revision. Preoperative speech analysis showed a more pronounced VPI in the VCFS group than in the control group. Outcomes of endoscopy and speech hypernasality improved significantly more in the control group than in the VCFS group. Improvement in the results of acoustic nasometry did not differ significantly between the 2 groups.
Treatment of VPI using palatal lengthening in children with VCFS is both safe and effective. The discrepancy in improvement between the speech analysis and the nasal endoscopy results within the VCFS group indicates that mechanical improvement does not necessarily correspond to an improvement in speech and emphasizes the complexity of speech disorders found in VCFS.
Hypernasality is a feature of velopharyngeal insufficiency (VPI), the inability to completely close the velopharyngeal port during speech production. Velopharyngeal insufficiency can result from a variety of causes, the most common being an overt or submucous cleft of the secondary palate.1 In patients with a cleft palate, velopharyngeal closure may be impaired by insufficient tissue, insufficient muscular activity, or scar tissue.
If hypernasality persists after primary repair of the cleft palate, a pharyngoplasty can be performed to normalize resonance during speech production. In our hospital, we routinely use a palatal lengthening procedure for patients with VPI.2,3 Palatal lengthening is achieved by pushback with a pharyngeal flap using only mucosal flaps instead of full-thickness mucoperiosteal flaps for the oral lining of the defect.3 When the periosteum and the palatine arteries are preserved, vascularization of the hard palate is maintained and bone is not exposed, thus avoiding potential detrimental scar formation overlying the hard palate, which may affect normal outgrowth of the maxilla. The level of improvement after surgical intervention varies among patients. One of the explanations for this variation is the existence of a comorbid syndromal abnormality such as velocardiofacial syndrome (VCFS).
Velocardiofacial syndrome has also been called DiGeorge syndrome, Shprintzen syndrome, and conotruncal anomaly face syndrome. Velocardiofacial syndrome is an autosomal dominant disorder caused by the 22q11.2 deletion and is also often referred to as 22q11.2 deletion syndrome. The incidence is approximately 1 per 4000 live births.4 It has a highly variable phenotype and manifests with multiple anomalies such as heart malformations, typical facial characteristics, developmental difficulties, a cleft palate, VPI, and a characteristic pattern of hypernasal speech.5- 8 Pharyngeal muscle hypotonia is present in approximately 75% of patients with VCFS, contributing to hypernasality.9
In the present report, our experiences in functional outcomes after palatal lengthening velopharyngoplasty in patients with VCFS are described and compared with those in a group of patients without VCFS who were similarly treated. In addition, the clinical features and preoperative and postoperative speech analyses of patients with and without VCFS are reviewed.
Records of the patients who underwent palatal lengthening for VPI in the 1986-2001 period in our hospital were reviewed. The VCFS group consisted of 25 consecutively treated patients, 15 girls and 10 boys, with a mean age of 6.4 years at the time of surgery. The mean follow-up time was 5 years. Comparisons were made with 32 nonsyndromic cases of palatal lengthening surgery to correct hypernasality (14 submucous cleft palates, 5 congenital short vela, 6 cleft palates, 6 unilateral complete clefts, and 1 bilateral complete cleft). The control group was made up of 10 girls and 22 boys with a mean age at the time of surgery of 7.4 years. All 57 patients had undergone fluorescence in situ hybridization testing to either confirm or rule out the presence of the 22q11.2 deletion that causes VCFS.
All patients underwent preoperative and postoperative testing of velopharyngeal function by a phoniatrician and a speech pathologist. Screening included a perceptual speech investigation, nasal endoscopy, and acoustic nasometry.
Perceptual speech investigation was performed using Eurocleft sentences,10 which are designed so that every sentence contains 2 explosive consonants in the same position in 5 languages, including Dutch. Hypernasality during speech was categorized as “always,” “sometimes,” and “never.”
Nasal endoscopy was performed using a Pentax 2.3-mm flexible endoscope (Pentax Corp, Tokyo, Japan) and documented on videotape. Patients were asked to repeat high- and low-pressure oral- and nasal-loaded speech. Overall velopharyngeal function, including velar elevation, lateral wall movement, and closure pattern, was rated as “always,” “most of the time,” “sometimes,” and “never.”
Acoustic nasometry data were obtained using “Dr. Speech for Windows”11 in combination with the NasalView12 instrument. Nasalance scores of nasal and oral sounds were registered both preoperatively and postoperatively.
The results of all investigations were statistically analyzed. For the acoustic nasometry, a paired samples t test and an independent sample t test were used. The results of nasoendoscopy, the hypernasality speech tests, and the velum function tests were analyzed with a Wilcoxon signed rank test and a Mann-Whitney test. For the entire statistic analysis, we used complete result couples. Statistical significance was defined as P < .05.
In the VCFS group, 4 of the 25 patients required surgical revision (16%). These patients were slightly older at the time of initial surgery: 5.5 vs 6.0 years. Average time interval to revision was 6.7 years (range, 3-8 years). This delay had a variety of causes, including loss to follow-up (1 patient), reluctance of parents for reoperation (1 patient), and surgeon doubt regarding the efficacy of reoperation in patients with VCFS and poor palatal mobility (2 patients). Revision was performed using a sphincter pharyngoplasty.13,14 This technique yielded a clear anatomic improvement in 3 of the 4 patients (full closure achieved during nasoendoscopy) and an audible improvement in speech; however, speech normalized in only 1 case. The one patient whose speech did not improve after revision showed minimal to absent activity of the velum and pharyngeal wall. In the control group, none of the patients required revision. No cases of sleep apnea were observed in either group.
In the VCFS group, 24 of 25 patients were preoperatively determined to always manifest speech hypernasality (96%). The remaining patient was rated as sometimes (Figure 1). Postoperatively, 15 of 25 patients with VCFS did not show a decrease in hypernasality (58%). The remaining 10 showed an audible improvement (42%), but none of them achieved normal speech after the operation.
In the control group, 29 of 32 patients always displayed hypernasal speech preoperatively (91%). The remaining 3 patients sometimes showed hypernasal speech (9%). Twenty-five patients showed improvement postoperatively (78%), and 16 experienced completely normalized speech (50%). Multivariate analysis confirmed the significant improvement in hypernasality after the primary operation in both of the study groups (P = .001 for the VCFS group; P < .001 for the control group) and between the study groups (P ≤ .001).
Not all patients cooperated sufficiently to provide complete nasal endoscopy data preoperatively and postoperatively (Figure 2). For statistical analysis, only the complete data couples were used. As a result, the VCFS group was reduced to 16 patients, and the control group to 25 patients. Multivariate analysis demonstrated improvement of nasopharyngeal closure after surgery in both groups (VCFS group, P = .002; control group, P < .001) and between the VCFS and non-VCFS groups (P < .01).
In the VCFS group, 10 patients were tested preoperatively and postoperatively with acoustic nasometry. In the control group, 28 patients were tested. There was significant improvement in both groups (standard text, P < .01 and denasal text, P < .001 for both groups) (Table). An independent samples t test comparing the degree of improvement between the VCFS and non-VCFS groups demonstrated no significant difference.
Speech and language problems are among the most common characteristics of VCFS.15 The treatment of VPI in patients with VCFS is particularly challenging. Children with VCFS have several intrinsic anatomic and physiologic characteristics that may influence velopharyngeal closure. In addition to the cleft palate, several factors may influence VPI: pharyngeal hypotonia, platybasia (resulting in an increased distance from the palate to the posterior pharyngeal wall), unilateral vocal fold paralysis, and adenoid hypoplasia. With the palatal lengthening procedure used in our hospital, speech evaluation and analysis showed a satisfactory result in 84% of patients with VCFS. After revision, all but 1 patient showed further improvement in speech. All patients in the control group demonstrated improvement in speech, and none required revision.
Golding-Kushner15 suggests that communication skills in children with VCFS may be syndrome specific. Nearly 98% of patients with VCFS have developmental delay.15 When compared with the patients in our control group, the patients with VCFS had trouble cooperating with and completing the different preoperative and postoperative tests, reducing the available complete data scores. This seems to be inherent in the psychosocial development and personality characteristics of patients with VCFS.
This discrepancy in revision rate in our study may be explained by the fact that the VCFS group had more pronounced VPI preoperatively. The results of speech evaluations and nasal endoscopy demonstrated significant discrepancies between VCFS and controls. Both before and after surgery, patients without VCFS performed better. Postoperatively, the control group showed better velopharyngeal closure, decreased hypernasality, and improved velum function compared with the VCFS group. Our nasometry numbers indicate that the degree of improvement did not differ significantly between groups, but because the severity of the hypernasality was more pronounced in the VCFS group, the improvement was often not enough to fully alleviate VPI in the VCFS group. Therefore, a mechanical improvement in the VCFS group may not result in normalization of speech.
Functional brain imaging has provided insight into and illustrated the complexity of the abnormal neurologic mechanisms underlying the mathematic and speech problems seen in VCFS.16 In VCFS, many other factors are presumed to play a role in the outcome after surgery, including abnormal levator muscle anatomy, abnormal dimensions of the oropharynx, abnormal brain development, and psychosocial disorders.9,17- 19 Reoperation rate was low, although most patients with VCFS did not achieve normal speech by the time of postoperative measurement. Nonetheless, the final outcome is sufficient for children to attend school and have good social contacts. We monitor both of these parameters closely, and school and/or social problems due to incomprehensible speech are an indication to perform additional surgery in children with VCFS in our unit.
In our experience, the beneficial effects of speech-correcting surgery in the VCFS population are best evaluated 1 year after velopharyngoplasty. Patients with VCFS generally need such a time period to adjust and correct their speech pattern and techniques. In our experience, this adjustment to the new anatomic situation after surgery is markedly slower than that experienced by patients without VCFS. At present, we can only speculate about the cause of this slower adjustment. The lower average intelligence of the VCFS population compared with a normal population is a possible factor of importance. One of the limitations of our study was that patients were not matched for intelligence quotient. However, other studies do not mention the slower adjustment in the VCFS population.20 In the future, we intend to report on the long-term effects of palatal lengthening on speech in patients with VCFS. Good prospective randomized studies will be required to get the final answer.
An important benefit of the palatal lengthening technique as used in our hospital is the fact that a sphincter pharyngoplasty can still be carried out if the initial result is unsatisfactory, an especially important consideration in cases of asymmetry, as reported in patients with VCFS,21 when a unilateral or bilateral sphincter pharyngoplasty could further improve speech. In 4 of the 25 patients in our VCFS group, reoperation was required (16%). A 16% revision rate compares favorably with the results of Losken et al,20 who compared the results of sphincter pharyngoplasty for the treatment of VPI in patients with VCFS and a control group without VCFS, similar to our report. They found, as we did, that patients with VCFS had significantly less favorable results and a relatively high revision rate of 22%. However, the data collection in their study was different from that in ours, precluding a detailed comparison between outcomes.
In 2004, Mehendale et al21 also reported on the management of VPI in patients with VCFS. Their study lacks the comparison with a control group. They advocate a staged approach, repairing and retropositioning the muscle first in cases of anterior levators or submucous cleft palate. If this procedure yields insufficient effect, a Hynes pharyngoplasty is carried out. It is evident from their study that normal speech cannot always be achieved in patients with VCFS, even after a staged procedure.
Our study has several shortcomings. It is a retrospective study and includes several limitations such as possible loss of patients for follow-up, selection bias, and our inability to adequately control for additional variables. The intent of our analysis was to evaluate our surgery method and investigate whether this is a safe and reliable method to improve speech problems in VCFS. Formulating a hard conclusion from our results is further hampered by the fact that outcome of VPI surgery generally is not measured and reported in a uniform way, which makes comparisons between studies very difficult if not impossible. The relatively scarce literature reporting outcomes of VPI surgery in the VCFS population shares the same limitation. Based on the available data, we conclude that a palatal lengthening procedure to treat VPI in patients with VCFS is as successful as any other technique reported. The procedure is certainly safe, and it has the advantage that a sphincter pharyngoplasty (Orticochea or Hynes technique) is still possible. In this aspect, our approach resembles that of others like Mehendale et al21 in the treatment of patients VCFS: it proceeds incrementally and does not close off additional surgical options, should the need arise, to achieve an optimal result in the treatment of VPI.
Correspondence: Josine C. C. Widdershoven, MD, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, the Netherlands (firstname.lastname@example.org).
Submitted for Publication: September 5, 2007; final revision received January 23, 2008; accepted February 20, 2008.
Author Contributions: Drs Widdershoven and MinkvanderMolen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Widdershoven, Stubenitsky, and MinkvanderMolen. Analysis and interpretation of data: Widdershoven, Stubenitsky, Breugem, and MinkvanderMolen. Drafting of the manuscript: Widdershoven and Stubenitsky. Critical revision of the manuscript for important intellectual content: Widdershoven, Breugem, and MinkvanderMolen. Statistical analysis: Widdershoven. Study supervision: Breugem and MinkvanderMolen.
Financial Disclosure: None reported.