Laryngeal Web in Pediatric Population: Evaluation and Management PDF

Original Research Otolaryngology– Head and Neck Surgery Lar...

Original Research Otolaryngology– Head and Neck Surgery Laryngeal Web in the Pediatric 1–7 Ó American Academy of Otolaryngology–Head and Neck Population: Evaluation and Management Surgery Foundation 2019 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599819893985 http://otojournal.org Claire M. Lawlor, MD1,2, Natasha D. Dombrowski1,2, Roger C. Nuss, MD1,2, Reza Rahbar, DMD, MD1,2, and Sukgi S. Choi, MD, MBA1,2 No sponsorships or competing interests have been disclosed for this article. Keywords laryngeal web, glottic web, laryngeal stenosis, 22q11.2 dele- Abstract tion syndrome Objective. To discuss the presentation, evaluation, and man- Received June 18, 2019; accepted November 20, 2019. agement of pediatric laryngeal web. Study Design. Retrospective case series. L aryngeal webs result in stenosis of the airway at the Setting. Single tertiary care center. level of the glottis, although subglottic extension is Subjects. All patients with laryngeal web at Boston Children’s common. Laryngeal web was first described by Hospital in the past 22 years. Fleischmann in 1882 in an autopsy of an infant.1 Sometimes called ‘‘glottic webs,’’ they can be described as congenital Methods. No exclusion criteria. Charts mined for age at pre- or acquired. Congenital laryngeal webs account for 5% of sentation, presenting symptoms, degree/location of web, congenital anomalies of the larynx and result from failure of associated syndromes, number/type of surgical procedures, the larynx to recanalize during embryogenesis.2-4 They can and postoperative outcomes. occur in isolation or be associated with syndromes, such as Results. Thirty-seven patients were included (13 male, 24 22q11.2 deletion syndrome.5-7 Laryngeal webs are described female). Average age at diagnosis was 3.7 years (0-19.5 as acquired if they are the result of endolaryngeal surgery, years). Mean follow-up was 4.4 years (range, 0-16.4 years). trauma, caustic ingestion, infection, or prolonged intubation.5 There were 26 congenital webs (70.2%) and 11 acquired In 1985, Cohen8 published a review of laryngeal webs webs (29.8%). Presenting symptoms were vocal (29 patients, and proposed a grading system for anterior webs that is now 78.4%) and respiratory (22 patients, 60%). Underlying syn- of common use. Type 1 laryngeal webs are thin, are mem- dromes or synchronous airway lesions included the follow- branous, and involve less than 35% of the glottis. Type 2 is ing: premature (n = 5), congenital heart disease (n = 18), a web with 35% to 50% glottic involvement (Figure 1). A subglottic stenosis (n = 5), 22q11.2 deletion syndrome (n = type 3 web encompasses 50% to 75% of the glottis and often 10), and recurrent respiratory papillomatosis (n = 4). There extends to the anterior cricoid cartilage in the subglottis were 20 type 1 webs, 6 type 2 webs, 8 type 3 webs, and 3 (Figure 2). Type 4 is a laryngeal web involving up to 99% type 4 webs; 10 had subglottic extension of the laryngeal of the glottis. Complete laryngeal stenosis, if congenital, is web. Twelve patients were managed conservatively with referred to as congenital high airway obstruction syndrome, observation. Eighty-four interventions were performed: 18 or CHAOS.9 Presentation can range from an incidental find- open and 66 endoscopic (sharp division, 32; dilation, 33; ing to mild hoarseness, weak cry, stridor, dyspnea, aphonia, mitomycin C, 14; laser, 5; keel, 6; triamcinolone injection, 8; stent, 15; removal of granulation tissue, 5). Tracheotomy was required in 11 patients, and 5 patients were decannu- 1 Department of Otolaryngology, Children’s National Medical Center, lated. Voice improved in 12 patients, with respiratory symp- Washington, DC, USA 2 toms in 12 patients. Web recurred in 17 patients. One Department of Otolaryngology & Communication Enhancement, Boston patient died due to airway complications. Children’s Hospital, Boston, Massachusetts, USA Presented at the American Laryngological Association’s 2019 Annual Conclusions. Pediatric laryngeal web is an uncommon but Meeting at COSM; May 1, 2019; Austin, Texas. challenging lesion. Patients need to be evaluated for comor- bid syndromes and synchronous airway lesions. Management Corresponding Author: Sukgi S. Choi, MD, MBA, Department of Otolaryngology, Boston Children’s includes open and endoscopic procedures. Procedures Hospital, 300 Longwood Ave BCH3129, Boston, MA 02115, USA. should be tailored to the child’s presentation. Email: [email protected] 2 Otolaryngology–Head and Neck Surgery This study describes a single institutional experience with the diagnosis and management of pediatric laryngeal webs over a 22-year period. Methods The Institutional Review Board (IRB) at Boston Children’s Hospital (BCH) approved this retrospective review of all patients aged 0 to 18 years who were diagnosed with laryn- geal web at our institution between July 1, 1997, and February 28, 2019. Patients were identified using data cap- ture tools that queried medical records based on diagnosis code and/or keywords in the chart. Patients were excluded if they did not have a confirmed laryngeal web diagnosis, were treated at an outside hospital, or the records were incomplete. All laryngeal web diagnoses were confirmed via direct laryngoscopy and bronchoscopy. Complete medical records were obtained and reviewed for all patients in the study. Data collected included demo- Figure 1. Type 2 (35%-50%) thin membranous laryngeal web. graphic information, age at diagnosis, presenting symptoms, medical comorbidities, type of laryngeal web, treatment, recurrence, outcome, and follow-up. Particular attention was paid to the type of surgical procedure performed and the number of procedures necessary for complete treatment of recurrent laryngeal webs. Procedures were categorized as endoscopic, open, or both. Methods of laryngeal web repair included laser lysis, sharp lysis, laryngofissure with lysis and/or graft placement, airway stent placement, laryngeal keel placement, mitomycin C application, steroid injection, balloon dilation, and removal of granulation tissue. Results Thirty-seven patients met criteria for inclusion (Table 1). The patients were considered by etiology (congenital vs acquired) and degree of web (Cohen types 1 and 2 vs types 3 and 4). The etiologies of acquired laryngeal webs were prior laryngotracheal reconstruction, recurrent respiratory papillomatosis (RRP), and repeated or prolonged intuba- tions. Presenting symptoms were classified as vocal or respiratory. Vocal complaints were more common (29/37 patients, 78.4%) and included weak cry, hoarseness, high Figure 2. Type 3 (50%-75%) thick web with subglottic extension. pitch, dysphonia, and aphonia. Respiratory complaints (22/ 37, 60%) included stridor, respiratory distress, recurrent croup, and need for tracheotomy. Many of the patients had and severe respiratory distress at birth with potential death synchronous airway lesions, syndromes, or medical comor- due to airway obstruction.2-5 bidities, including prematurity, congenital heart disease, Cohen type 1 and 2 webs can be observed if asympto- subglottic stenosis independent of the laryngeal web, matic. When surgery is necessary, they can often be man- 22q11.2 deletion, and recurrent respiratory papillomatosis. aged endoscopically. The web may be divided with cold The incidence of Cohen type web classification examined steel or a CO2 laser (Figure 3). A laryngeal keel may be by etiology can be found in Table 2. placed to keep the opposing mucosal surfaces of the anterior The 37 patients in our cohort underwent a total of 84 vocal folds apart during healing (Figure 4).1,10-12 Type 3 procedures with intervention during the period of our and 4 webs often require a tracheotomy prior to definitive review (Table 3) with 20 different surgeons. The number of repair.13 Repair can be performed endoscopically, but often procedures per patient ranged from 0 (observation) to 21, an open laryngotracheal reconstruction with placement of with an average of 2.3 procedures/patient. Observation graft is necessary to resolve the stenosis.14 alone was possible in 12 patients; 25 patients required Lawlor et al 3 Figure 3. Laryngeal web before and after endoscopic sharp division. Figure 4. Laryngeal web with keel placement and after keel removal. operative intervention. Endoscopic procedures were per- patients who underwent open procedures and after 16 of 18 formed in 24 of 25 patients who underwent procedures, open procedures (89%). Of the 11 patients who required tra- with 66 endoscopic procedures performed in total. Open cheotomy placement, 5 were decannulated during the study procedures were performed in 9 of 25 patients, with 18 period. operative procedures performed in total. Only 1 patient underwent an open procedure without a prior endoscopic Discussion attempt. Tracheotomy placement was required in 11 To our knowledge, we present the largest review of pedia- patients. tric patients with both congenital and acquired laryngeal The outcomes of our cohort can be found in Table 4. webs in the literature to date. The incidence of congenital Many patients presented with both respiratory and voice webs in our series was significantly higher than that of symptoms. Of the 22 patients who presented with respira- acquired webs, with a ratio of 2.4:1. There was a 2.3:1 tory symptoms, 12 noted improvement after undergoing pro- female to male predominance in congenital laryngeal web cedures. A total of 29 patients presented with voice patients. This was lost in the acquired webs, where the complaints; 12 noted improvement after interventions. female to male ratio was only 1.2:1. Laryngeal webs classi- Recurrence of the web occurred in 11 of the 25 patients cally present with respiratory and voice symptoms, includ- who underwent procedures (44%). Improvement in web, but ing a soft or hoarse cry, aphonia, stridor, and severe not complete resolution, occurred in 6 patients who under- respiratory distress.2-4 This was consistent in our population, went procedures. Recurrence of web occurred to some with a slight predominance of vocal symptoms over respira- degree in 16 of 24 patients who underwent endoscopic pro- tory (78% vs 59%). Comorbid conditions included congeni- cedures and following 50 of 66 (76%) endoscopic proce- tal heart disease, prematurity, vocal fold paralysis, RRP, and dures. Recurrence of web occurred to some degree in 6 of 8 separate subglottic stenosis (SGS). In our congenital web 4 Otolaryngology–Head and Neck Surgery Table 1. Laryngeal Web Patient Demographics. Characteristic Total Congenital Acquired Types 1-2 Types 3-4 Number 37 26 11 26 11 Age, mean (range) 3.7 (0-19.5) 1.5 (0-5.7) 9.2 (0.2-19.5) 4.6 (0-19.5) 1.9 (0-6.2) Sex, No. Male 13 8 5 8 5 Female 24 18 6 18 6 Length of follow-up, mean (range), y 4.4 (0-16.4) 4.0 (0-16.4) 5.2 (0-12.1) 4.1 (0-12.9) 5.0 (0-16.4) Presenting symptoms, No. Respiratory 22 15 7 11 11 Voice 29 22 7 21 8 Comorbid conditions, No. 22q11.2 10 10 0 7 3 RRP 4 0 4 4 0 CHD 18 14 4 12 6 Prematurity 5 3 2 2 3 Vocal fold paralysis Unilateral 1 1 1 Bilateral 2 2 2 SGS 5 4 1 2 3 Etiology of acquired web, No. Repeated/prolonged intubations, No. 5 RRP procedures 4 LTR 2 Abbreviations: CHD, congenital heart disease; LTR, laryngotracheal reconstruction; RRP, recurrent respiratory papillomatosis; SGS, subglottic stenosis. Table 2. Cohen Web Classification by Congenital or Acquired Lesion. Type, No. Characteristic 1 (\35%) 2 (35%-50%) 3 (50%-75%) 4 (75%-99%) Subglottic Extension, No. Total 20 6 8 3 10 Congenital 13 3 8 2 6 Acquired 7 3 0 1 4 population, 10 patients had 22q11.2 deletion syndrome other glottic injuries are known complications of the surgical (38%). There were no other associated genetic syndromes in management of RRP, whether using sharp or laser tech- our population. niques.17,18 Laryngotracheal reconstructions for a preexisting Although relatively uncommon, congenital webs have SGS were the cause of laryngeal web in 2 patients (2/11, been an area of active investigation, particularly the link 18%). Although not always possible, care must be taken to between laryngeal webs and 22q11.2 deletion syndrome and avoid violating the anterior commissure of the larynx as this contemporary surgical management.6,7,15 Acquired laryngeal can lead to development of laryngeal web and poor vocal webs have also been studied, although not as commonly in outcomes.19 the pediatric population. Etiologies described in the litera- Our 37 patients underwent a total of 84 procedures with ture include intubation trauma, external laryngeal trauma, intervention. The average number of procedures for conge- caustic ingestion, endolaryngeal surgery, and laryngeal nital web was 1.96 procedures/patient (range, 0-11), com- tumors.5,16 In our series, the most common etiology of pared to 3.18 procedures/patient for acquired webs (range, acquired laryngeal web was intubation trauma and/or pro- 0-21). The decisions to perform the procedure endoscopi- longed intubation (5/11, 45%). The second highest risk cally vs open, as well as which techniques to employ, were factor for acquired laryngeal web was endolaryngeal surgery surgeon dependent. The 86 procedures were performed by for the treatment of RRP (4/11, 36%). Laryngeal web and 20 different surgeons over the 22-year period. This may Lawlor et al 5 Table 3. Procedures Performed for Laryngeal Web. Characteristic Total No. Congenital, No. Acquired, No. Types 1-2, No. Types 3-4, No. Total procedures 84 50 34 24 60 Observation 12 8 4 12 0 Open 18 12 6 3 15 Endoscopic 66 38 28 21 45 Sharp 32 24 8 16 16 CO2 laser 5 4 1 1 4 Stent ETT 13 9 4 12 1 Other 2 2 0 2 0 Keel 6 4 2 2 4 Mitomycin C 14 8 6 5 9 Dilation 33 15 18 1 32 Steroid injection 8 4 4 2 6 Granulation removal 5 2 3 2 3 Tracheotomy 11 7 4 3 8 Abbreviation: ETT, endotracheal tube. Table 4. Laryngeal Web Surgical Outcomes. Characteristic Total, No. (%) Congenital, No. (%) Acquired, No. (%) Types 1-2, No. (%) Types 3-4, No. (%) Respiratory improvement 12/22 (55) 8/15 (53) 4/7 (57) 5/11 (45) 7/11 (64) Voice improvement 12/29 (41) 10/22 (45) 2/7 (29) 8/21 (38) 4/8 (50) Trach decannulation 5/11 (45) 3/7 (43) 2/4 (50) 1/3 (33) 4/8 (50) Resolved web 8/25 (32) 7/18 (39) 1/7 (14) 5/14 (36) 3/11 (27) Improved web 6/25 (24) 4/18 (22) 2/7 (29) 3/14 (21) 3/11 (27) Recurrent web 11/25 (44) 7/18 (39) 4/7 (57) 6/14 (43) 5/11 (45) Deceased 1/37 (3) 1/26 (4) 0/11 (0) 0/26 (0) 1/11 (9) have contributed to the diversity of surgical approaches. In separation of the anterior vocal folds during the initial heal- our series, open cases were reserved for type 3/4 laryngeal ing process. webs with significant aberrant soft tissue (rather than a We stratified our patient series by Cohen type web clas- membranous web) or webs that had failed endoscopic man- sification. Type 1 and 2 webs were significantly more agement. Observation alone was possible in 8 congenital common than the more severe type 3 and 4 webs (70% and patients (31%) and 4 acquired patients (36%), because 30%, respectively). Unsurprisingly, the more severe webs symptoms were mild and not of significant concern to the were more difficult to manage surgically. The average patient or family. Open procedures were performed in 24% number of procedures for type 1 and 2 webs was 0.96 pro- of cases on congenital webs and 17% of cases on acquired cedures/patient (range, 0-5), compared to 5.55 procedures/ webs. In our population, tracheotomy was required in 30% patient for acquired webs (range, 1-21). Observation was of congenital cases and 36% of acquired webs. The laryn- possible in 12 of the type 1 and 2 patients (46%) but was geal webs were surgically resolved in 39% of congenital not possible in the type 3 and 4 patients due to the severity webs and 14% of acquired webs. The webs persisted but of symptoms. Open procedures were performed in 15% of were improved after surgery in 22% of congenital cases and cases on type 1 and 2 webs and 25% of cases on type 3 and 29% of acquired cases. Recurrence occurred in 44% of con- 4 webs. In our population, tracheotomy was required in genital cases and 39% of acquired cases. This suggests that 12% of type 1/2 cases and 73% of type 3/4 webs. The laryn- both congenital and acquired laryngeal webs present chal- geal webs were surgically resolved in 36% of type 1/2 webs lenging surgical dilemmas. The opposing raw mucosal sur- and 27% of type 3/4 webs. The webs persisted but were faces on the anterior vocal folds increase the risk of web improved after surgery in 21% of type 1/2 cases and 27% of recurrence. Keel placement or laryngeal stenting, often with type 3/4 cases. Recurrence occurred in 43% of type 1/2 an endotracheal tube, is often required to maintain webs and 45% of type 3/4 webs. In this series, it is clear 6 Otolaryngology–Head and Neck Surgery that stratifying patients by severity of laryngeal web, rather for comorbid syndromes and airway lesions. Management than congenital or acquired etiology, is a greater predictor of techniques include open and endoscopic procedures, with a clinical complexity, including possibility of observation, wide array of techniques discussed in the literature. number of surgical procedures needed to treat, and need for Laryngeal webs classically present with respiratory and voice tracheotomy. Unfortunately, neither stratifying patients by web symptoms. Improvement in respiratory and voice symptoms etiology nor severity was predictive of surgical outcomes, can remain challenging in complex cases. Further understand- including symptomatic improvement or risk of recurrence. ing of the etiology of acquired cases and optimization of sur- Unfortunately, recurrence of web was a common compli- gical techniques should lead to improved outcomes. cation in our cohort and occurred frequently after both endoscopic (76%) and open (89%) procedures. The slightly Author Contributions higher rate of recurrence after open procedures may be con- Claire M. Lawlor, acquisition and analysis of data, drafting of the founded by the greater severity and complexity of the webs work, final approval of version to be published, agreement to be that required an open approach. A greater number of accountable for all aspects of work; Natasha D. Dombrowski, patients in our series saw improvements in their respiratory acquisition and analysis of data, drafting of the work, final symptoms than their vocal symptoms (55% vs 41%). approval of version to be published, agreement to be accountable Symptomatic improvement was based on parental reporting for all aspects of work; Roger C. Nuss, design of work, critical and, less often, formal voice evaluation; this was also sur- revision, final approval of version to be published, agreement to be accountable for all aspects of work; Reza Rahbar, design of work, geon dependent. Decannulation was achieved in 45% of critical revision, final approval of version to be published, agreement patients who required tracheotomy. Historically, respiratory to be accountable for all aspects of work; Sukgi S. Choi, substantial and swallowing were the primary outcome measures in the contribution to conception, critical revision, final approval of version treatment of airway pathology. While improvement in to be published, agreement to be accountable for all aspects of work. airway symptoms is paramount, there is increasing recogni- tion of the handicap of pediatric dysphonia. Persistent dys- Disclosures phonia can have both a social and an academic impact on pediatric patients.20 While vocal outcomes have been stud- Competing interests: None. ied in patients undergoing laryngotracheal reconstruction, Sponsorships: None. there are no data on vocal outcomes after management of Funding source: None. laryngeal webs. This is an area of potential future study. This study is limited by the retrospective nature of the References chart review. Chart review often does not reveal the sur- 1. Alkan U, Nachalon Y, Vaisbuch Y, Katz O, Hamzany Y, geon’s preoperative thought process, including why one Stern Y. Treating paediatric anterior glottic web: single-centre intervention was selected over another in each individual experience of 20 patients with comparison among techniques. case. In addition, although this represents the largest review Clin Otolaryngol. 2017;42:893-897. of pediatric congenital and acquired laryngeal webs, this is 2. Holinger PH, Brown WT. Congenital webs, cysts, laryngoceles still a relatively small case series of only 37 patients over a and other anomalies of the larynx. Ann Otol Rhinol Laryngol. 22-year review. While single-surgeon case series have their 1967;76:744. own limitations, this represents the results of 20 different 3. Benjamin B. Chevalier Jackson Lecture. Congenital laryngeal surgeons, including fellows. This underscores that the man- webs. Ann Otol Rhinol Laryngol. 1983;92(4, pt 1):317-326. agement of laryngeal webs is universally difficult. 4. Nicollas R, Triglia JM. The anterior laryngeal webs. Otolaryngol In this series, we tried a wide variety of endoscopic and Clin North Am. 2008;41:877-888. open management techniques in our approach to laryngeal 5. Stephenson KA, Wyatt ME. Glottic stenosis. Semin Pediatr webs. When coupled with the high recurrence rate, this sug- Surg. 2016;25(3):132-137. gests that there is not one ideal surgical technique for the 6. Miyamoto RC, Cotton RT, Rope AF, et al. Association of management of laryngeal webs, highlighting an area for anterior glottic webs with velocardiofacial syndrome (chromo- continued investigation and innovation within our field. some 22q11.2 deletion). Otolaryngol Head Neck Surg. 2004; Specifically, future studies should focus on optimizing the 130:415-417. surgical approach based on the severity and etiology of the 7. Ebert B, Sidman J, Morrell N, Roby BB. Congenital and iatro- web to aid surgeons in preoperative planning. Furthermore, genic laryngeal and vocal abnormalities in patients with 22q11. with greater understanding of the causes of acquired laryn- 2 deletion. Int J Pediatr Otorhinolaryngol. 2018;109:17-20. geal webs, increased consideration can be given to avoiding 8. Cohen SR. Congenital glottic webs in children: a retrospective this frustrating complication. review of 51 patients. Ann Otol Rhinol Laryngol Suppl. 1985; 121:2-16. Conclusions 9. Lim FY, Crombleholme TM, Hedrick HL, et al. Congenital Pediatric laryngeal web is an uncommon but challenging high airway obstruction syndrome: natural history and man- laryngeal lesion. It is important to evaluate these patients agement. J Pediatr Surg. 2003;38:940-945. Lawlor et al 7 10. Cao Y, Sun G. Surgical treatment modalities for iatrogenic ante- 16. Cotton RT, Tewfik TL. Laryngeal stenosis following carbon rior glottic stenosis. Ann Otol Rhinol Laryngol. 2018;127:946-952. dioxide laser in subglottic hemangioma. Ann Otol Rhinol 11. Milczuk HA, Smith JD, Everts EC. Congenital laryngeal Laryngol. 1985;94:494-497. webs: surgical management and clinical embryology. Int J 17. Holland BW, Koufman JA, Postma GN, McGuirt WF Jr. Pediatr Otorhinolaryngol. 2000;52:1-9. Laryngopharyngeal reflux and laryngeal web formation in 12. Yoo MJ, Roy S, Smith LP. Endoscopic management of conge- patients with pediatric recurrent respiratory papillomas. nital anterior glottic stenosis. Int J Pediatr Otorhinolaryngol. Laryngoscope. 2002;112:1926-1929. 2015;79:2056-2058. 18. Perkins JA, Inglis AF Jr, Richardson MA. Iatrogenic airway 13. de Trey LA, Lambercy K, Monnier P, Sandu K. Management stenosis with recurrent respiratory papillomatosis. Arch of severe congenital laryngeal webs: a 12 year review. Int J Otolaryngol Head Neck Surg. 1998;124:281-287. Pediatr Otorhinolaryngol. 2016;86:82-86. 19. Pullens B, Hakkesteegt M, Hoeve H, Timmerman M, Joosten 14. Wyatt ME, Hartley BE. Laryngotracheal reconstruction in con- K. Voice outcome and voice-related quality of life after sur- genital laryngeal webs and atresias. Otolaryngol Head Neck gery for pediatric laryngotracheal stenosis. Laryngoscope. Surg. 2005;132:232-238. 2017;127:1707-1711. 15. Jones JW, Tracy M, Perryman M, Arganbright JM. Airway 20. Hseu A, Ayele N, Dombrowski N, et al. Prevalence of voice anomalies in patients with 22q11.2 deletion syndrome: a 5- disturbances in pediatric airway patients: are we missing any- year review. Ann Otol Rhinol Laryngol. 2018;127:384-389. thing? Int J Pediatr Otorhinolaryngol. 2019;118:59-61.