ORIGINAL ARTICLE Otolaryngologic Manifestations of the 22q11.2 Deletion Syndrome Orville Dyce, MD; Donna McDonald-McGinn, MS; Richard E. Kirschner, MD; Elaine Zackai, MD; Kathleen Young, BSE; Ian N. Jacobs, MD Background: The 22q11.2 chromosome deletion syndrome occurs at a frequency of 1 in 4000 live births. Fluorescent in situ hybridization is a reliable means of testing for this genetic abnormality. Objective: To describe the otolaryngologic manifestations of the 22q11.2 deletion syndrome to improve recognition and management of these disorders. Patients and Design: A retrospective medical record review of 102 patients with chromosome 22q 11.2 deletions confirmed by fluorescent in situ hybridization. Setting: A multidisciplinary 22q11.2 deletion clinic at an academic children s hospital. Outcome Measure: All otolaryngologic problems were recorded, including facial dysmorphic features, velopharyngeal insufficiency, speech and airway abnormalities, feeding difficulties, gastroesophageal reflux, hearing loss, otitis media, sinus problems, and vascular anomalies. Additionally, available objective test results were recorded, including those from audiograms, imaging studies, endoscopies, speech evaluations, and vascular studies. Results: Dysmorphic facial features were found in most patients. Velopharyngeal incompetence was noted in 76 patients, while overt submucosal clefts were found in 11 patients. Most patients had speech and language delays. In addition, 53 patients had chronic or recurrent otitis media, and 28 had recurrent sinorhinitis. Furthermore, feeding problems were found in 48 patients, while vascular anomalies of the head and neck were found in 16 patients. Conclusion: Otolaryngologic abnormalities are relatively common and important to recognize with the 22q11.2 deletion syndrome. Arch Otolaryngol Head Neck Surg. 2002;128:1408-1412 From the Division of Pediatric Otolaryngology (Drs Dyce and Jacobs) and Genetics (Mss McDonald-McGinn and Young and Drs Kirschner and Zackai), The Children s Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia. IN 1978, Shprintzen et al1 first described the velocardiofacial syndrome when they reported on a series of patients with a particular constellation of findings including conotruncal cardiac anomalies, dysmorphic facial features, and palatal dysfunction. Along with his colleagues, Shprintzen further elucidated the characteristics and diagnostic findings of this syndrome. 1-5 In 1982, at The Children s Hospital of Philadelphia (CHOP), Philadelphia, Pa, a patient with DiGeorge syndrome was found to have a deletion on the long arm of chromosome 22 using fluorescent in situ hybridization (FISH) techniques. 6 In further studies, a submicroscopic deletion of the long arm of chromosome 22 (22q11.2) has been identified in most patients with DiGeorge, velocardiofacial, and conotruncal anomaly face syndromes. 7,8 Some patients with Opitz G/BBB and Cayler cardiofacial syndromes also have the deletion. 9,10 The 22q11.2 chromosome deletion syndrome occurs at a frequency of 1 in 4000 live births. 11 In addition to the characteristic pattern of conotruncal cardiac anomalies, a significant number of common otolaryngologic problems are found in these syndromes. These include velopharyngeal insufficiency, cleft palate, characteristic facial dysmorphisms, otitis media, sinorhinitis, hearing loss, speech and language difficulties, feeding problems, gastroesophageal reflux, as well as congenital airway and vascular abnormalities. At CHOP, a large multidisciplinary clinic is dedicated to the care of children with the 22q11.2 deletion syndrome. Patients are evaluated by members of the genetics, plastic surgery, immunology, endocrine, speech pathology, pediatrics, and otolaryngology divisions of the institution. The main objective of the present study was to describe the otolaryngologic 1408
manifestations in a randomly selected group of 102 patients seen over an 8-year period. METHODS We performed a retrospective medical record review of the outpatient records from the Genetics Department of CHOP. In total, 102 patients were randomly selected using the first 102 patients from an alphabetical listing. All patients had a 22q11.2 deletion confirmed by FISH analysis. Each patient was evaluated by the multidisciplinary 22q11.2 clinic, which is coordinated by the Genetics Division. Patients are all first seen by genetics specialists and then sent to the various specialists at CHOP including otolaryngology, plastics, endocrinology, cardiology, general pediatrics, immunology, and audiology. The decision to send patients to the various specialists was based on patient needs, available time, and specialist availability. Many patients were evaluated by a pediatric otolaryngologist in the outpatient clinic of the Otolaryngology Division. These patients underwent a complete otolaryngologic history review and physical examination including pneumatic otoscopy and/or microtympanostomy. In addition, many, but not all, patients underwent audiological testing that included sound field audiometry as a screening tool for infants and toddlers. Patients who had recent hearing evaluations outside of CHOP did not undergo a second testing unless the findings were abnormal. When sound-field testing findings were abnormal or indeterminate, children underwent brainstem evoked-response testing. Older children underwent pure-tone audiometry (voluntary or conditioned play audiometry). From the medical record review, otological history was recorded including the presence or absence and duration of chronic otitis media with effusion or recurrent acute otitis media, as well as the dates of tympanostomy tube insertion, tympanomastoid ear surgery, and sequelae. When hoarseness was present and laryngeal abnormalities suspected, patients underwent flexible fiberoptic laryngoscopy with topical anesthesia. Patients were evaluated for velopharyngeal insufficiency (VPI) by a pediatric plastic surgeon and speech pathologist (both experienced in diagnosing VPI), who physically examined the palates of all patients. The palate was inspected for the presence of a bony cleft as well as notching, dimpling, and a bifid uvula. All postlingual patients underwent perceptual speech examination by a speech pathologist experienced in evaluating speech as the initial screening for VPI. Prelingual patients were assessed for hypernasality by fluid reflux through the nose during feeding. When notable VPI was suspected from the speech evaluation findings and pharyngeal flap surgery was being considered, VPI was confirmed by nasopharyngoscopy in cooperative children or by videofluoroscopy in uncooperative children. Incomplete closure of the velum was considered confirmatory evidence of VPI. A palpable bony defect in the hard palate was considered a submucosal cleft, while an occult cleft was defined as a VPI, diagnosed by endoscopy or fluoroscopy, without a bony defect. Patients being considered for pharyngeal flap surgery underwent magnetic resonance angiography or computed tomography to determine the position of the carotid arteries. We also recorded available data in several categories. Signs of facial dysmorphisms were recorded including eye, auricular, and nasal abnormalities. The results of standard palatal evaluations were reviewed including any functional or anatomic evaluation for velopalatal insufficiency. Next, general otolaryngologic problems were recorded including hearing loss, otitis media and sinorhinitis. All available audiometric data were included. In addition, abnormal airway and vascular findings were tabulated. RESULTS OVERALL DEMOGRAPHICS Records of an otolaryngologic evaluation were found for 59 patients. The referral source was recorded on the medical chart for 90 of the 102 patients: 52 patients were referred to the multidisciplinary clinic by the genetics service; 22 by the cardiology service; 8 by the plastic surgery service; 2 by the speech service, 2 by the otolaryngology service, 2 by the rheumatology service, 1 by the immunology service, and 1 by the endocrinology service. FACIAL DYSMORPHISMS Dysmorphic facial features were extremely common in the 102 patients. Nasal abnormalities, with bulbous nasal tip being the most common, are listed below. Nasal Abnormality Bulbous nasal tip 61 Thickened nasal bridge 34 Narrow nares 31 Prominent nasal root 28 Nasal dimpling 10 Auricular abnormalities, with overfolded helix being the most common, are listed below. External Auricular Abnormality Overfolded helix 45 Protuberant ears 23 Thickened helix 18 Cupped ears 16 Low-set ears 16 Squared-off helix 15 Posteriorly rotated ears 13 Microtia 12 Other facial features are listed below. Other Dysmorphic Finding Facial asymmetry 28 Small mouth 21 Retrognathia 21 Down-slanting corners of the mouth 14 Hooded eyelids 26 Epicanthal folds 18 Hypertelorism 13 Up-slanting eyes 3 Some the common features of a child with the 22q11.2 deletion syndrome can be seen in Figure 1. Common auricular anomalies are shown in Figure 2. SPEECH AND PALATAL ABNORMALITIES More than 70% of the patients had a history of significant speech problems or language delays. The most common speech abnormality was velopharyngeal insufficiency (hypernasal speech), which was diagnosed by perceptual voice analysis in 76 patients. Velopharyngeal insufficiency was confirmed by nasopharyngoscopy in 17 patients and by videofluoroscopy in 17 patients. In addition, an overt cleft palate was seen in 11 patients, a submucosal cleft in 14 patients, and an occult cleft of the palate in 6 additional patients. Uvular abnormalities, ranging from a grossly bi- 1409
a mixed pattern, and another 4 had an indeterminate pattern at the time of testing. AIRWAY, VASCULAR, AND SWALLOWING ABNORMALITIES Congenital airway problems, while less common, occurred relatively frequently when considered as a group in this patient population. Nineteen patients who had signs or symptoms of airway obstruction were evaluated by airway fluoroscopy, flexible fiberoptic laryngoscopy, direct laryngoscopy, or bronchoscopy. Fourteen patients were found to have laryngotracheal abnormalities (listed below), with 3 subjects having multiple airway problems. Figure 1. A child with the 22q11.2 deletion syndrome. Laryngotracheal Abnormality Subglottic level narrowing 4 True vocal cord paralysis 2 Glottic level narrowing 2 Tracheomalacia 1 Tracheoesophageal fistula 1 Vocal cord nodules 1 Laryngeal cleft 1 High tracheal bifurcation or abnormal 2 tracheal curvature Evaluation of the vascular anatomy was completed with magnetic resonance imaging, and arterial or venous phasing was performed on 11 patients. Two studies demonstrated medialization of the carotid arteries. Overall, vascular abnormalities were relatively uncommon in this group. Persistent feeding and swallowing difficulties were reported in 41 patients and gastroesophageal reflux in 61 patients. COMMENT CLASSIFICATION AND DIAGNOSIS OF THE 22q11.2 SYNDROME Figure 2. Common auricular anomalies. fid uvula to minor dimpling, were seen in 23 patients in our cohort. GENERAL OTOLARYNGOLOGIC PROBLEMS A history of chronic or recurrent otitis media, defined as effusions lasting longer than 30 days or more than 3 episodes of acute otitis per year, was recorded in 53 patients. A history of myringotomy and tympanostomy tube placement was found for 24 patients. Eight patients had multiple sets of tubes. The mean age was 34.7 months at the time of the last tube placement. Most patients improved after insertion of tympanostomy tubes, although some continued to develop infections. Seventyone patients underwent audiometry. Hearing loss, as diagnosed by sound-field audiometry, pure-tone audiometry, or brainstem evoked-response testing, was found in 42 patients. Conductive hearing loss was the most common abnormality and seen in 32 patients. Two patients had evidence of sensorineural loss, 4 patients exhibited The use of FISH has improved our ability to make the diagnosis of the 22q11.2 deletion. 7 This technique has allowed the recognition of a shared common chromosomal defect among several distinct syndromes. We found considerable phenotypic overlap among patients listed as having velocardiofacial and DiGeorge syndromes. We were unable to separate the syndromes based on otolaryngologic features alone. Furthermore, a lack of reliable correlation between the velocardiofacial and DiGeorge syndromes and distinct loci on chromosome 22 made any attempt at subclassification of syndromes impossible. 12-14 Thus, we attempted to identify the more common features of all the 22q11.2 deletion as a single syndrome rather than several distinct syndromes. In some children the 22q11.2 deletion syndrome may be difficult to recognize and diagnose. This is particularly true if the typical features are subtle or not present at all. For the otolaryngologist, this is of some concern because routine interventions such as an adenoidectomy or tonsillectomy may be of considerable detriment to these patients. Nonetheless, by recognizing the aforementioned findings it is possible to develop a strong clinical suspicion warranting further investigation with a FISH screen. 1410
The facial features can be highly variable and are frequently mild. Clinicians have reported on many of the features discussed herein, including characteristic facial dysmorphisms of the 22q11.2 syndromes as well as common otolaryngologic findings. 3-5,15-19 In addition, many of the eye abnormalities have been described elsewhere. 18 While subtle and not the primary focus of the otolaryngologist, external eye findings in conjunction with other facial dysmorphisms can contribute to the recognizability of these patients (Figure 1). SPEECH AND COMMUNICATION PROBLEMS Speech and language difficulties were features also common to patients with the 22q11.2 deletion. A formal speech evaluation was performed on most patients in our cohort. Hypernasality, articulation errors, and problems with intelligibility are likely to be among the reasons requiring otolaryngologic evaluation. Most patients underwent speech therapy. This is in line with the findings of others. 3 The speech and language difficulties may be further confounded by neurodevelopmental problems frequently found in association with the 22q11.2 syndrome. 20,21 Overall, presumptive evidence of velopharyngeal insufficiency was present in three quarters of the study population and was not limited to those patients with cleft palate abnormalities. Some patients manifested nasopharyngeal reflux, while others had varying degrees of hypernasality. Velopharyngeal wall dysmotility in the presence or absence of an overt submucous cleft palate are often cited as causative factors of VPI. 15 Often patients may have velopharyngeal disproportion with foreshortened soft palates. In fact, Zori and colleagues 22 reviewed a population of patients with isolated VPI without a clear cause and found positive FISH results in 7 of 23 patients tested. While this group represented a select group of patients followed by a specialty clinic, these findings lend credence to the need for careful evaluation of patients with functional pharyngeal abnormalities without clearly identifiable anatomic abnormalities. Considering the above, it is not surprising that feeding problems would also be present in a considerable number of patients with the 22q11.2 deletion. These have been well described in prior reports. 3,23 Poor suck reflexes early in life in addition to nasopharyngeal regurgitation were frequent findings. However, not all feeding difficulties are directly attributable to upper pharyngeal dysfunction. Cricopharyngeal dysfunction, esophageal spasm, and frequent emesis were not uncommon findings in our series. Directly attributable to these feeding difficulties was the frequent diagnosis of failure to thrive. Many of these patients eventually required placement of nasogastric or gastrostomy feeding tubes. Further confounding the language and communication problems were the frequent findings of chronic otitis media and hearing loss. Typically, a mild to moderate conductive loss was present. This was found in close to 40% of the patients in our cohort. The causes of the hearing loss and chronic otitis media were likely multifactorial, with immune deficiency, palatal dysfunction, Eustachian tube dysfunction, and chronic middle ear effusions 15,16 all contributing. A smaller subset of patients had anomalies of the middle ear, which appeared to lead to their hearing loss. The percentage of patients with both 22q11.2 deletion and hearing loss, which widely varies in the literature, has been reported as high 75%. 24 The basis for the wide discrepancy is unclear and may simply reflect referral patterns for genetic testing. VASCULAR AND AIRWAY ABNORMALITIES While cardiac anomalies are fairly common, head and neck vascular anomalies occur less frequently. Of particular concern to us was the finding of medial displacement of the carotid arteries, which placed them at increased risk of injury during pharyngeal surgery. We incidentally discovered this anomaly in 2 patients, one of whom was noted to have prominent pharyngeal pulsations during a routine tonsillectomy. Because magnetic resonance angiography was not routinely performed on all patients in our cohort, the true prevalence of these findings in our group is unknown. The frequency of this anomaly has been investigated by others and may be much more common than reported here. 3 If gone unrecognized, such medial displacement could represent a potentially life-threatening anomaly during pharyngeal surgery. Therefore, magnetic resonance imaging should be performed on the children with 22q11.2 deletion who are under consideration for any type of pharyngeal surgery such as adenotonsillectomy or pharyngeal flap. Airway evaluations were occasionally warranted in patients with 22q11.2 deletion because of stridor, apneic episodes, cough, or other evidence of respiratory distress. Primary congenital anomalies of the larynx, trachea, or bronchi were uncommon in our series. The more prevalent findings of subglottic stenosis, vocal cord paralysis, or laryngeal stenosis were each seen in less than 5% of our patients. Airway symptoms were not infrequent sequelae of vascular aberrances including vascular rings, aberrant innominate vessels, and tortuous aortic arches with tracheobronchial compression. Based on these findings, we believe that airway abnormalities were nonspecific, contributing little to the diagnosis of the syndrome, while the aforementioned vascular anomalies warrant further investigation. While the group of evaluated patients came from a diverse geographic location, there may be certain selection bias inherent in such a select group, such as referral filter or diagnostic assessment bias. 25 The families of patients with more severe findings may seek out the evaluations at a multispecialty clinic. Nevertheless, 100 patients were referred for nonotolaryngologic reasons. Thus, the large diverse group of patients still demonstrates the general otolaryngological characteristics of the syndrome. In addition, diagnostic suspicion bias may have played some role, especially with respect to the screening evaluation of VPI because patients with 22q11.2 deletion may have been expected to have speech problems. However, when VPI was found on screening, further objective testing was performed to confirm these findings. The objective structural and functional palatal abnormalities, diagnosed on physical examination, nasopharyngoscopy, or videofluoroscopy, are less likely to be influenced by expectation. 1411
While individual findings may be nonspecific, certain constellations of symptoms should heighten one s suspicion (eg, clefting in association with a history of cardiac anomalies, velopharyngeal insufficiency without clearly identifiable anatomic causes, or individual syndromic facial features with evidence of head and neck vascular anomalies). When these problems are found in concert, it may be worthwhile to refer these patients for further genetic testing and counseling. CONCLUSIONS Many patients with 22q11.2 deletion syndrome should be recognizable to the general otolaryngologist. Head and neck abnormalities are common in patients with the 22q11.2 deletion syndrome. The typical facial features, speech problems, cardiac defects, and palatal abnormalities may serve as clues and aid in the diagnosis. As otolaryngologists, our ability to recognize these patients and the mere avoidance of a potentially harmful intervention, such as tonsillectomy in the patient with the medially displaced carotid artery or adenoidectomy in a patient with velopharyngeal insufficiency, would be beneficial in the absence of active intervention. Furthermore, referring these patients to the appropriate services could greatly aid their early development. Accepted for publication April 1, 2002. This study was presented at the Society for Ear, Nose, and Throat Diseases in Children (SENTAC), Williamsburg, Va, December 5, 1999. Corresponding author: Ian N. Jacobs, MD, Division of Pediatric Otolaryngology, Richard D. Wood Center, First Floor, 34th Street and Civic Center Boulevard, Philadelphia, PA 19104 (e-mail: jacobsi@e-mail.chop.edu). REFERENCES 1. Shprintzen RJ, Golberg RB, Lewin ML, et al. A new syndrome involving cleft palate, cardiac anomalies, typical facies, and learning disabilities: velo-cardiofacial syndrome. Cleft Palate J. 1978;15:56-120. 2. Motzkin B, Marion R, Goldberg R, Shprintzen R, Saenger P. Variable phenotypes in velo-cardio-facial syndrome with chromosomal deletion. J Pediatr. 1993; 123:406-410. 3. Shprintzen RJ. Velo-Cardio-Facial Syndrome Educational Foundation, Inc. VCFS Clinical Database Project. Last updated February 14, 2000. Available at: http:// www.vcfsef.org. Accessed March 5, 2000. 4. Mitnick RJ, Bello JA, Golding-Kushner KJ, Argamaso RV, Shprintzen RJ. The use of magnetic resonance angiography prior to pharyngeal flap surgery in patients with velo-cardio-facial syndrome. Plast Reconstr Surg. 1996;97:908-919. 5. Goldberg R, Motzkin B, Marion R, Scambler PJ, Shprintzen RJ. Velo-cardiofacial syndrome: a review of 120 patients. Am J Med Genet. 1993;45:313-319. 6. Kelley RI, Zackai EH, Emanuel BS, Kistenmacher M, Greenberg F, Punnett HH. The association of the DiGeorge anomalad with partial monosomy of chromosome 22. J Pediatr. 1982;101:197-200. 7. Driscoll DA, Budarf ML, Emanuel BS. A genetic etiology for DiGeorge syndrome: Consistent deletions and microdeletions of 22q11. Am J Hum Genet. 1992; 50:924-933. 8. Burn J, Takao A, Wilson D, et al. Conotruncal anomaly face syndrome is associated with a deletion within chromosome 22. J Med Genet. 1993;30:822-824. 9. Gianotti A, Diglio MC, Marino B, Mingarelli R, Dallapiccola B. Cayler cardiofacial syndrome and del 22q11: part of the CATCH22 phenotype. Am J Med Genet. 1994; 53:303-304. 10. McDonald-McGinn DM, Driscoll DA, Bason L, et al. Autosomal dominant Opitz G/BBB syndrome due to a 22q11.2 deletion. Am J Med Genet. 1995;59:103-112. 11. Devriendt K, Fryns J, Mortier G, VanThienen M, Keymolen K. The annual incidence of DiGeorge/velo-cardio-facial syndrome. J Med Genet. 1998;35:789-790. 12. McQuade L, Christodoulou J, Budarf M, et al. Patient with a 22q11.2 deletion with no overlap of the minimal DiGeorge syndrome critical region (MDGCR). Am J Med Genet. 1999;86:27-33. 13. Stevens CA, Carey, JC, Shigeoka AO. DiGeorge anomaly and velo-cardio-facial syndrome. Pediatrics. 1990;85:526-530. 14. Scambler PJ, Kelly D, Lindsay E, et al. Velo-cardio-facial syndrome associated with chromosome 22 deletions encompassing the DiGeorge locus. Lancet. 1992; 339:1138-1139. 15. Vantrappen G, Rommel N, Cremers CW, Devriendt K, Frijns JP. The velo-cardiofacial syndrome: the otorhinolaryngeal manifestations and implications. Int J Pediatr Otorhinolaryngol. 1998;45:133-141. 16. Vantrappen G, Devriendt K, Swillen A, et al. Presenting symptoms and clinical features in 130 patients with the velo-cardio-facial syndrome: the Leuven experience. Genet Couns. 1999;10:3-9. 17. McDonald-McGinn DM, LaRossa D, Goldmuntz E, et al. The 22q11.2 deletion: Screening, diagnostic workup, and outcome of results report on 181 patients. Genet Test. 1997;1:99-108. 18. McDonald-McGinn DM, Kirschner R, et al. The Philadelphia story: the 22q11.2 deletion report on 250 patients. Genet Couns. 1999;10:11-24. 19. Gripp KW, McDonald-McGinn DM, et al. Nasal dimple as part of the 22q11.2 deletion syndrome. Am J Med Genet. 1997;69:90-92. 20. Gerdes M, Solot C, Wang Paul, et al. Cognitive and behavior profile of preschool children with chromosome 22q11.2 deletion. Am J Med Genet. 1999,85:127-133. 21. Moss E, Batshaw M, Solot C, et al. Psychoeducational profile of the 22q11.2 microdeletion: a complex pattern. J Pediatr. 1999;134:193-198. 22. Zori RT, Boyar FZ, Williams WN, et al. Prevalence of 22q11 deletion in patients with VPI. Am J Med Genet. 1998;77:8-11. 23. Rommel N, Vantrappen G, Swillen A, Devriendt K, Feenstra L, Fryns JP. Retrospective analysis of feeding and speech disorders in 50 patients with velo-cardiofacial syndrome. Genet Couns. 1999;10:71-78. 24. Thomas JA, Graham JM. Chromosome 22q11 deletion syndrome: an update and review for the primary pediatrician. Clin Pediatr (Phila). 1997;36:2253-2266. 25. Sackett DL. Bias in research. J Chronic Dis. 1979;32:51-63. 1412