- Email: [email protected]
Late-onset laryngomalacia: A cause of pediatric obstructive sleep apnea
International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Late-onset laryngomalacia: A cause of pediatric obstructive sleep apnea§ Sally M. Revell *, William D. Clark The University of Texas Health Science Center at San Antonio, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 5 July 2010 Received in revised form 2 November 2010 Accepted 3 November 2010 Available online 27 November 2010
Objective: To describe the presentation, diagnosis, and treatment of late-onset laryngomalacia in children with obstructive sleep apnea syndrome (OSAS). Design: Retrospective study. Setting: Tertiary care children’s hospital. Patients: Seventy-seven children were identified who had OSAS diagnosed by polysomnography and underwent airway endoscopy to evaluate for laryngomalacia between July 2006 and December 2008. Children with significant neurologic disease or craniofacial malformations were excluded. Seven children under 3 years of age had laryngomalacia and OSAS (Group A), 19 children 3–18 years of age had laryngomalacia and OSAS (Group B), and 51 children 3–18 years of age had OSAS but not laryngomalacia (Group C). Main outcome measures: Comparison of pre-operative findings, intra-operative findings, interventions, and outcomes between the 3 groups. Results: Group A was consistent with previous reports of congenital laryngomalacia with respect to presentation, diagnosis, and treatment. Groups B and C had similar pre-operative findings, including a high incidence of adenotonsillar hypertrophy, and the only significant difference was the intra-operative finding of laryngomalacia in Group B. Treatments were individualized to include supraglottoplasty, adenoidectomy, tonsillectomy, adenotonsillectomy, or a combination of the above. Of the 52 patients who returned in follow-up, 44 noted improvement, but this was rarely confirmed by polysomnogram. Conclusions: Late-onset laryngomalacia may act alone or in concert with additional dynamic or fixed lesions to cause pediatric OSAS. Although there is no specific pre-operative indicator to diagnose lateonset laryngomalacia, it can be readily identified intra-operatively and effectively treated with supraglottoplasty, with or without concurrent adenotonsillectomy. ß 2010 Elsevier Ireland Ltd. All rights reserved.
Keywords: Laryngomalacia Obstructive sleep apnea Adenotonsillectomy Supraglottoplasty
1. Introduction Obstructive sleep apnea syndrome (OSAS) is a syndrome of partial or complete upper airway obstruction occurring during sleep and causing a disruption of normal ventilation and sleep patterns. While this may occur in patients of any age, the classic description of pediatric OSAS is in a child 2–8 years old with symptoms of loud snoring, witnessed apneas, frequent nighttime arousals, and chronic mouth breathing [1]. Ten percent of children have some form of sleep disordered breathing (including primary snoring) and 1–3% of children have obstructive sleep apnea [2]. These numbers are important because pediatric OSAS is associated with neurocognitive, cardiovascular, and metabolic sequelae that
§ Poster Presentation at the American Society of Pediatric Otolaryngology Spring Meeting Las Vegas, NV, April 30, 2010. * Corresponding author at: 8300 Floyd Curl Drive, MSC 7777, San Antonio, TX 78229, United States. Tel.: +1 210 450 0709; fax: +1 210 562 9374. E-mail address: [email protected] (S.M. Revell).
0165-5876/$ – see front matter ß 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2010.11.007
can have lasting consequences throughout the child’s life. School performance suffers due to behavioral problems, developmental delay, and inattention, while overall health is damaged by systemic inflammation and metabolic derangements, all of which contribute to a poor quality of life for the child. Major risk factors for OSAS in children include adenotonsillar hypertrophy, obesity, neuromuscular disease, and craniofacial anomalies [3–5]. Pediatric OSAS is often attributed to adenotonsillar hypertrophy, yet research suggests the cause is often multifactorial in nature, with decreased neuromuscular tone during REM sleep also playing a large role [1,6,7]. Despite a recognition that the cause of obstruction is multifactorial, adenotonsillectomy continues to be the most common intervention, supported by studies citing improvements in behavior [8] and quality of life [9] and overall cure rates of 75–100% after adenotonsillectomy [10]. As larger studies with better levels of evidence emerge, however, it is now clear that adenotonsillectomy is not a cure-all for pediatric OSAS. Tauman et al. [11] studied 110 patients and found that while all showed improvement in sleep parameters after adenotonsillectomy, only 25% had complete cure of OSAS (defined as an apnea
232
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
hypopnea index (AHI) < 1). Mitchell [12] reported on 79 nonobese, otherwise healthy children with OSAS, all of whom underwent pre- and post-operative polysomnograms. Sixteen percent of children had persistent disease when defined by an AHI greater than or equal to 5 and 27% of children had persistent disease after adenotonsillectomy when defined by a respiratory distress index (RDI) of greater than or equal to 1. The largest and most recent study by Bhattacharjee et al. [13] retrospectively evaluated the efficacy of adenotonsillectomy in 548 children with OSAS. Although 90.1% of patients showed a reduction in AHI, only 27.2% had complete normalization of their breathing patterns during sleep after adenotonsillectomy. Patients more likely to have persistent disease were greater than 7 years old, had a higher body mass index (BMI), had a higher pre-operative AHI, and had asthma. Looking specifically at obese patients, Mitchell and Kelly [14] found that 76% of obese patients had persistent disease after adenotonsillectomy. The results of these studies support the notion of OSAS as being multi-factorial in nature and raise the question of what additional upper airway pathology, whether fixed or dynamic, is contributing to OSAS in children. Congenital laryngomalacia is a well-described dynamic lesion of the larynx, causing collapse of the supraglottic structures during inspiration. It is the most common congenital laryngeal abnormality, presenting with inspiratory stridor during the first 2 weeks of life, worsening over the next 6–8 months, and resolving by 2 years of age [15]. In this patient population, only 10% of affected patients require intervention due to complications such as failure to thrive, obstructive sleep apnea sydrome, resting dyspnea, hypoxia or hypercapnea, pulmonary hypertension, or cor pulmonale [16]. More recently, variations of this typical presentation have been reported, including state-dependent and late-onset laryngomalacia. State-dependent laryngomalacia was described by Smith et al. [17] after identifying 5 patients between the ages of 3 and 4 years old who were asymptomatic while awake, but experienced stridor and airway collapse during sleep. This was thought to be secondary to decreased neuromuscular tone in the sleeping child, causing it to be missed on an office exam including flexible fiberoptic laryngoscopy. Late-onset laryngomalacia was described by Richter et al. [18] in patients older than 2 years of age with sub-categories of feeding-disordered, sleep-disordered, and exercise-induced laryngomalacia. Seven school-aged children presented with sleep disordered breathing, five of whom had already undergone adenotonsillectomy without resolution of symptoms. They were all identified as having sleep-disordered laryngomalacia and improved after treatment with supraglottoplasty. As more data emerges about pediatric obstructive sleep apnea syndrome, more questions arise regarding the pathophysiology of this disease and the appropriate work-up, intervention, and follow-up of these patients. Airway endoscopy has not historically been a part of the initial work-up for pediatric OSAS, so it is possible that late-onset laryngomalacia is present in a number of these patients but is not being recognized. Using the results of routine airway endoscopy in children with OSAS, this study aims to increase awareness of late-onset laryngomalacia as a cause of pediatric OSAS and to compare the presentation, diagnosis, and management of these patients to that of younger children with congenital laryngomalacia and to age-matched children with obstructive sleep apnea but not laryngomalacia. 2. Materials and methods With approval of the institutional review board, a retrospective study was conducted on patients treated in the pediatric otolaryngology clinic at Christus Santa Rosa Children’s Hospital (San Antonio, TX) for laryngomalacia and/or obstructive sleep apnea between July 2006 and December 2008. Only patients with
obstructive sleep apnea diagnosed by polysomnogram and airway endoscopy performed by the senior author to confirm or deny the diagnosis of laryngomalacia were included. Patients with significant craniofacial malformations or neurologic disorders were excluded. Data was collected from inpatient and outpatient electronic and paper charts to include history and physical exam findings, polysomnogram results, operative reports, operative photos and videos, and post-operative course. Seventy-seven patients met inclusion criteria for the study. For the purposes of this study, a cut-off of 3 years of age was used to separate congenital from late-onset laryngomalacia, allowing for division of the patients into three distinct groups. Group A (congenital laryngomalacia) contained patients ages 0–3 years old with obstructive sleep apnea and laryngomalacia, Group B (lateonset laryngomalacia) contained patients ages 3–18 years old with obstructive sleep apnea and laryngomalacia, and Group C (classic pediatric OSAS) contained patients ages 3–18 years old with obstructive sleep apnea but without laryngomalacia. Pre-operative data points included sleep symptoms, medical comorbidities, weight, presence of stridor and/or retractions, and tonsil and adenoid size. Patient weights were plotted on CDC growth curves and standard formulas were used to calculate the growth percentile for each patient. For patients under 24 months of age, weight to length percentiles were used to determine if the patient was underweight (<5th percentile), normal weight (5th to 95th percentile), or overweight (>95th percentile). Patients over 24 months of age were assessed based on body mass index percentiles and divided into underweight (<5th percentile), normal weight (5th to 85th percentile), overweight (85th to 95th percentile), and obese (>95th percentile) categories [19]. Tonsil size was graded from 1 to 4 with size 1 tonsils hidden within the pillars, size 2 tonsils extending to the edge of the pillars, size 3 tonsils extending beyond the pillars, and size 4 tonsils meeting in the midline. Adenoids were simply described as obstructing or nonobstructing. Mild laryngomalacia was diagnosed if the supraglottic collapse was intermittent or led to incomplete obstruction of the laryngeal inlet while severe laryngomalacia was diagnosed if the collapse was persistent with complete obstruction. All patients underwent pre-operative full-night polysomnogram, most of which were conducted by a pediatric pulmonologist at the same children’s hospital. Respiratory events were scored based on American Academy of Sleep Medicine guidelines [20] and disease was categorized as mild (AHI 1–5), moderate (AHI 5–10), or severe (AHI > 10) for each patient. Direct laryngoscopy and tracheobroncoscopy were performed with each patient spontaneously ventilating under intravenous anesthesia. A propofol drip was started at 200–250 mcg/kg/min and titrated to achieve deep sedation. Direct laryngoscopy was then performed with a Parson’s laryngoscope, a weight-appropriate dose of 2% lidocaine was applied to the vocal cords, and then tracheobronchoscopy was performed with a Hopkin’s rod telescope. During this period, observations were made regarding adenotonsillar size, cobblestoning of the tracheal mucosa, and presence of any other airway pathology. Once this had been completed, the laryngoscope was kept in the vallecula while the anesthesiologist turned off the propofol, allowing for evaluation of dynamic laryngeal pathology as the patient began to awaken from anesthesia and produced vigorous inspiratory efforts. During this period, the aryepiglottic folds were observed for collapse with inspiration allowing for diagnosis of mild or severe laryngomalacia or no laryngomalacia. Indications for supraglottoplasty in a child with polysomnogram-proven OSAS included severe laryngomalacia with or without adenotonsillar hypertrophy or mild laryngomalacia
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
without adenotonsillar hypertrophy. When indicated, this was performed using laryngeal micro-scissors to excise portions of the aryepiglottic folds deemed to be redundant, with or without the associated cuneiform cartilage. Indications for adenotonsillectomy included a diagnosis of OSAS with adenotonsillar hypertrophy, with or without associated laryngomalacia. If necessary, tonsillectomy was performed with needle-tip electrocautery and adenoidectomy was performed with a microdebrider. If both laryngomalacia and adenotonsillar hypertrophy were diagnosed, treatment was individualized, as will be further addressed in the discussion section. Any children with continued symptoms and positive polysomnography after the completion of surgical interventions were referred back to the sleep center for medical management.
233
Table 1 Comorbidities.
Overweight/obesity Asthma Allergic rhinitis Attention deficit hyperactivity disorder Reflux disease Prematurity Diabetes mellitus Congential heart disease History of respiratory syncitial virus Speech delay Recurrent pneumonia Cystic fibrosis Hypertension Failure to thrive
Group A
Group B
Group C
1
12 10 6 6 2
25 18 10 4 4 4 4 2 2 1 1 1 1 1
1 1
3. Results
[()TD$FIG]
After excluding patients with significant neurologic disease and craniofacial anomalies, the most common comorbidities were overweight/obesity, asthma, allergic rhinitis, attention deficit hyperactivity disorder, and gastroesophageal reflux disease (Table 1). Demographic information was not consistently recorded and is therefore not available to report. There were no surgical complications noted for any of the study patients. The findings of each group are described below and represented in Fig. 1.
3.1. Group A Seven patients under 3 years of age were identified as having both laryngomalacia and obstructive sleep apnea syndrome. The patients ranged in age from 3 to 10 months with a mean age of 6 months. All seven patients were male. The most common presenting symptoms in this group were stridor (5), witnessed apnea (5), snoring (3), noisy breathing (2), feeding difficulties (2),
Fig. 1. Treatment group comparisons for pre-operative data. Data are presented as: (a) percentage of patients with each presenting symptoms, (b) percentage of patients in each weight category, and (c) percentage of patients with mild, moderate, and severe OSA by polysomnogram.
[()TD$FIG]
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
234
Fig. 2. Congenital laryngomalacia. Photos demonstrate (a) shortened aryepiglottic folds and (b) redundant mucosa of the aryepiglottic folds.
Additional operative findings included cobblestoning of the tracheal mucosa in 4 patients and an antrochoanal polyp in one patient. Four patients had severe laryngomalacia with varying degrees of adenotonsillar hypertrophy. They underwent supraglottoplasty alone. One patient reported some subjective improvement, but had no objective studies performed. Another had continued snoring and behavior problems but a pre-operative AHI of 2.5 dropped to 0.3 and no additional interventions were required. A third patient had subjective improvement and resolution of snoring, but went on to have a tonsillectomy and adenoidectomy 9 months later for recurrent tonsillitis. The last patient had no improvement in symptoms, and a repeat sleep study was performed. It showed continued severe obstructive sleep apnea, so the patient underwent adenotonsillectomy, after which snoring resolved. Eight patients had obstructing adenotonsillar tissue with varying degrees of laryngomalacia. One patient in this group had severe adenotonsillar hypertrophy and laryngomalacia, but only adenotonsillectomy was performed initially due to lack of consent for the supraglottoplasty. He did not improve until he underwent supraglottoplasty as well, as will be further described in the discussion section. The remaining seven patients in this group had adenotonsillar hypertrophy as the predominant obstructing pathology with mild laryngomalacia, so they underwent tonsillectomy or adenotonsillectomy without supraglottoplasty. Five had subjective improvement with 2 of these demonstrating complete resolution by sleep study (AHI 15.4–0.1 and 4.5–0.1). One had continued somnolence, enuresis, and
mouth breathing (2), and cyanosis (1). One patient was underweight, 5 patients were of normal weight, and 1 patient was overweight. Physical exam findings included stridor (4) and retractions (1) and no patients were noted to have adenotonsillar hypertrophy. OSA severity was mild in 2 patients, moderate in 2 patients, and severe in 3 patients. Intra-operative findings were significant for laryngomalacia alone in 6 patients and for laryngomalacia in addition to tracheomalacia and bronchomalacia in 1 patient. Laryngomalacia in these patients was characterized by shortened aryepiglottic folds with redundant overlying mucosa that was pulled into the airway resulting in airway obstruction (Fig. 2). This was most commonly noted during emergence from anesthesia and was accompanied by stridor and substernal retractions. Four of the seven patients underwent supraglottoplasty and three returned for follow-up. One had subjective improvement but no post-operative polysomnogram and another had subjective improvement and polysomnogram revealed improvement in apnea hypopnea index from 13.1 to 2.3. The last had subjective improvement but residual symptoms prompted re-evaluation. Repeat endoscopy showed no collapse and polysomnography revealed a decrease in AHI from 6.1 to 1.0. 3.2. Group B Nineteen patients between 3 and 18 years of age were identified with a diagnosis of laryngomalacia and obstructive sleep apnea syndrome. The age range was 3.8–11.1 years with a mean age of 7.3 years. Eight patients were male and 11 were female. The most common presenting symptoms were snoring (17), apnea (15), daytime somnolence (12), difficulty awakening (10), mouth breathing (9), behavior problems (8), gasping/choking during sleep (7), enuresis (6), difficulty feeding (2), recurrent tonsillitis (2), noisy breathing (1), stridor (1) and cyanosis (1). One patient was underweight, 6 patients were of normal weight, no patients were overweight, and 12 patients were obese. Tonsil and adenoid size were recorded in 18 and 12 patients, respectively. Tonsillar hypertrophy (size 2 or greater) was noted in 89% of patients (11% size 1, 28% size 2, 39% size 3, and 22% size 4). Adenoids were nonobstructing in 33% of patients and obstructing in 67%. None of these patients had stridor or retractions noted on office exam. Polysomnogram severity was mild in 10 patients, moderate in 5 patients, and severe in 4 patients. All of the patients in this group, by definition, were also found to have laryngomalacia. The predominant feature in these patients was the redundant mucosa of the aryepiglottic folds that was pulled into the airway, causing obstruction during forceful inspiration (Fig. 3). Shortened aryepiglottic folds were a less common feature in this population.
[()TD$FIG]
Fig. 3. Late-onset laryngomalacia. Photo demonstrates the typical finding of redundant mucosa of the aryepiglottic folds.
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
behavior problems, but repeat endoscopy was normal and polysomnogram showed an improvement in AHI from 6 to 1.3. The last patient in this group was lost to follow-up. Three patients had significant adenotonsillar hypertrophy and laryngomalacia. They underwent supraglottoplasty plus tonsillectomy or adenotonsillectomy. All three reported significant improvement with resolution of symptoms, but no post-operative polysomnograms were performed. Another patient also had a fixed obstruction in the form of an antrochoanal polyp. He had been treated with adenotonsillectomy for sleep-disordered breathing at another facility and was then referred when symptoms persisted. He was diagnosed with severe OSAS by polysomnogram (AHI 39.3) and scheduled for direct laryngoscopy and tracheobronchoscopy. He had mild laryngomalacia in addition to a large, obstructing antrochoanal polyp. He underwent endoscopic removal of the polyp with subjective improvement and a post-operative polysomnogram revealed a decrease in AHI from 39.3 pre-operatively to 2.2 post-operatively. No intervention was performed for the mild laryngomalacia. Three patients did not undergo any surgical intervention. Two were offered supraglottoplasty for laryngomalacia, but the families declined, and the other had mild, intermittent laryngomalacia, no obstructing adenotonsillar tissue, and mild OSAS, so no intervention was recommended. None of these patients returned for follow-up. 3.3. Group C Fifty-one patients between 3 and 18 years of age were identified with obstructive sleep apnea but without laryngomalacia. Their ages ranged from 3.0 to 16.6 years with a mean age of 7.2 years. There were 26 male and 25 female patients. The main presenting symptoms were snoring (45), apnea (33), difficulty wakening (29), daytime somnolence (22), mouth breathing (21), behavior problems (10), recurrent tonsillitis (9), enuresis (9), gasping/ choking during sleep (6), poor weight gain (4), noisy breathing (4), difficulty feeding (2), stridor (2), cyanosis (1). Three patients were underweight, 23 patients were of normal weight, 7 patients were overweight, and 18 patients were obese. Tonsil and adenoid size were recorded in 46 and 35 patients, respectively. Tonsillar hypertrophy (size 2 or greater) was noted in 91% of patients (9% size 1 tonsils, 33% size 2, 39% size 3, and 20% size 4). Adenoids were non-obstructing in 23% of patients and obstructing in 77%. No patients were noted to have stridor or retractions during office exam. OSAS severity was available for 49 patients with disease being mild in 23 patients, moderate in 12 patients, and severe in 14 patients. None of these patients had any evidence of laryngomalacia on airway endoscopy. Additional pathology included 15 patients with cobblestoning of the tracheal mucosa and 2 patients with nodules on the true vocal folds. Thirty-eight patients underwent adenotonsillectomy and 28 of them returned for follow-up. Twenty-one reported subjective improvement and underwent no further testing. Four patients reported subjective improvement which was confirmed by postoperative polysomnography (AHI 9.1–0.6, AHI 19.4–2.2, AHI 18.2– 1.3, AHI 53.2–0.2). Three additional patients only followed up in the sleep clinic and subjective data is not available. Of these, one patient continued to have mild OSAS (AHI 3–1.6), another dropped from severe to mild OSAS (AHI 16.4–3.5), and the last continued to have severe OSAS (AHI 18.5–13.7). These patients continued to follow-up with the sleep clinic and were treated with medical management. Five patients underwent tonsillectomy alone for isolated tonsillar hypertrophy. Two patients returned in follow-up, both of whom had subjective improvement. One of these had a repeat polysomnogram which confirmed a decrease in AHI from 12.6 to
235
1.4, but the other had no post-operative studies performed. Three did not return for follow-up. Three patients underwent adenoidectomy alone. One reported subjective improvement initially, but then apnea returned. A repeat polysomnogram showed improvement (AHI 11.7–3.6), but continued disease prompted the initiation medical management by the sleep medicine clinic. The second patient reported subjective improvement but had no additional studies performed, and the third did not return for follow-up. Five patients did not show significant adenotonsillar hypertrophy or any other surgically correctable site of obstruction, so no surgical interventions were performed. These patients were all referred back to the sleep clinic for medical management. 4. Discussion While the otolaryngology literature has a large volume of data on laryngomalacia and pediatric obstructive sleep apnea syndrome, with both being recognized as common diseases of childhood, the interaction between these two diseases has been less well studied. The association between the laryngomalacia and OSAS has been best described in the infant population. In these patients, laryngomalacia, with its tell-tale stridor, is often diagnosed first, with subsequent polysomnography being used to demonstrate the resulting obstructive sleep apnea and to prompt the intervention of supraglottoplasty for treatment. Zafereo et al. [21], O’Connor et al. [22] and Pereira Valera et al. [15] have reported statistically significant improvements in polysomnography parameters, to include respiratory distress index [15,21,22] and lowest oxygen saturation levels, [21,22] in patients who underwent supraglottoplasty as a treatment for moderate to severe laryngomalacia. These studies and their results can be correlated with Group A in this study where the most common symptoms were stridor (71%) and apnea (71%), most patients were of normal weight, no patients had adenotonsillar hypertrophy, and supraglottoplasty was an effective treatment. In contrast to infants with congenital laryngomalacia and obstructive sleep apnea, older children are first identified by their symptoms of sleep disordered breathing that lead to adenotonsillectomy with or without pre-operative polysomnogram. While this is an accepted practice, there has been recent evidence that adenotonsillectomy is not as successful at curing pediatric OSAS as was once believed. This is likely due to the multi-factorial nature of the obstruction, some of which is dynamic and some of which is fixed in nature. It is for this reason that every child with symptoms of sleep disordered breathing at our institution undergoes intraoperative direct laryngoscopy and tracheobronchoscopy prior to adenotonsillectomy. This allows for the separation of Groups B and C as each patient in Group C has been evaluated for laryngomalacia and found not to have it. This is an important distinction from previous studies, as many patients treated elsewhere will only undergo endoscopy after failing empiric treatment with adenotonsillectomy. On initial presentation, the patients in both Groups B and C fall under the more classic description of pediatric obstructive sleep apnea syndrome. As it is classically defined, pediatric OSAS has a peak incidence between 2 and 8 years of age, occurs equally in males and females, is characterized by snoring, mouth breathing, and restless sleep, and is associated with normal body weight and adenotonsillar hypertrophy. As the patients move toward adolescence, patients are more commonly male and obese, but continue to have the same symptom complex of snoring, mouth breathing, and restless sleep [23,24]. Both Groups B and C in this study revealed a similar incidence among males and females and had a mean ages of 7.3 and 7.2 years old, respectively. The most common presenting symptoms for Groups B and C were also in accordance
236
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
with previous studies. They included snoring (89%, 88%), apnea (79%, 65%), daytime somnolence (63%, 43%), difficulty awakening (53%, 57%), and mouth breathing (47%, 41%). There were differences between Groups B and C in less commonly reported symptoms, such as behavior problems, enuresis, and gasping/ choking during sleep, but little can be deduced from this, as not all patients were specifically asked about every symptom. The distribution of OSAS severity was also similar between the groups, with Group B having 53% mild, 26% moderate, and 21% severe disease, while Group C had 47% mild, 24% moderate, and 29% severe disease. The distribution of adenotonsillar hypertrophy was also similar, with 89% of Group B and 91% of Group C having tonsillar hypertrophy (tonsil size 2 or greater) and 67% of Group B and 77% of Group C having adenoid hypertrophy (obstructing). While obesity is recognized as a risk factor for pediatric OSAS, this is not part of the ‘‘classic’’ presentation of pediatric OSAS. In this study, the percentage of overweight and obese patients is high in both Group B (63%) and Group C (49%), which is not surprising given the three-fold increase in obesity in school-aged children and adolescents in the United States since 1980 [25]. Dayyat et al. [26] reported that obese patients are likely to have an ‘‘adult-like’’ presentation and disease course, which is supported by the presence of symptoms like daytime somnolence and difficulty wakening in this study population. The numbers in this study are too small to show a difference in presentation and disease course between the normal weight and obese children, but Mitchell and Kelly [14] and Bhattacharjee et al. [13] have demonstrated that increasing BMI is associated with a higher incidence of persistent disease than is seen in normal weight children. Larger studies will be needed to evaluate the relationship between obesity and lateonset laryngomalacia. This study makes a distinction between groups based on the findings on airway endoscopy. The patients who were found to have laryngomalacia were placed into Group B and diagnosed with late-onset laryngomalacia. Late-onset laryngomalacia, as described by Richter et al. [18] has a different symptom complex and disease course than congenital laryngomalacia. He described a group of 7 school-aged children with symptoms of sleep disturbance who were found to have type I laryngomalacia (redundant mucosa prolapsing over the arytenoids into the airway) and were treated successfully with supraglottoplasty. He further classified these patients as having ‘‘sleep-disordered laryngomalacia.’’ All 19 patients in Group B of this study would fall into this category, as they presented with typical pediatric OSAS symptoms, were found to have laryngomalacia on airway endoscopy, and were often treated successfully with supraglottoplasty. The main difference between this group and Richter’s group is the weight. He reported a mean weight percentile of 37% [18] while mean body mass index percentile in Group B of this study was 79%. This could be due to a regional variation in overweight and obese children, or may simply reflect small case numbers in each study. In the patients with late-onset laryngomalacia (Group B), the findings on airway endoscopy were more similar to the patients with congenital laryngomalacia (Group A) while the patient presentation was much more similar to that of the children and adolescents with pediatric OSAS without laryngomalacia (Group C). Another interesting finding is the very natural break between the ages of the patients in Groups A and B. The initial study design used a cut-off of 3 years of age to separate congenital laryngomalacia (typically in patients younger than 2 years of age) from late-onset laryngomalacia to ensure that no patients with congenital laryngomalacia would be grouped with the lateonset patients. When the data was analyzed, however, the oldest patient with congenital laryngomalacia and OSAS was 10 months old while the youngest patient with late-onset laryngomalacia and
OSAS was 3 years 9 months old. This further supports the concept that late-onset laryngomalacia is a distinct clinical entity from congenital laryngomalacia. There is no clear explanation for the cause of this atypical laryngomalacia, especially in patients without underlying neuromuscular disorders. Richter et al. [18] suggested that gastro-esophageal reflux disease (GERD), decreased laryngeal tone, and supraglottic edema may all contribute and exacerbate one another in a cyclical fashion. This is supported by research in the infant population showing that no matter the etiology of laryngomalacia, the negative pressure generated to overcome the obstruction worsens airway edema, subsequently worsening the supraglottic airway obstruction of laryngomalacia [27]. While not directly studied, it could be hypothesized that negative pressure due to any upper airway obstruction (including adenotonsillar hypertrophy) could lead to reflux and subsequent airway edema that may be severe enough to lead to the supraglottic collapse of laryngomalacia. Although only one patient in Group B had been formally diagnosed with GERD, findings of cobblestoning of the tracheal mucosa in 21% of these patients supports a role for reflux disease in the pathophysiology of lateonset laryngomalacia. It is also possible, given the large number of obese patients in our study, that obesity itself may play a role in the pathophysiology, especially given the relationship between GERD and obesity. In this study, clinically significant late-onset laryngomalacia was more common in obese children, with 6 of the 8 patients (75%) who required supraglottoplasty being obese. However, given the small numbers of patients involved, additional studies will be needed to further elucidate this point. While previous studies only sought unusual pathology after a patient had failed adenotonsillectomy, our study is unique in that the only patient who had undergone previous adenotonsillectomy was the patient with the antrochoanal polyp. In all other cases, an intra-operative decision was made as to the relative contributions of adenotonsillar hypertrophy (fixed obstruction) and laryngomalacia (dynamic obstruction) to each patient’s obstructive pathology. Of the 15 patients in Group B who returned for follow-up, 3 patients improved with supraglottoplasty alone, 6 patients improved after adenotonsillectomy (or isolated tonsillectomy or adenoidectomy), and 3 improved after concurrent supraglottoplasty and tonsillectomy or adenotonsillectomy. Two patients required additional procedures prior to improving. One was a patient who was found to have severe adenotonsillar hypertrophy (tonsil size 4 and obstructing adenoids) and severe laryngomalacia on direct laryngoscopy. He was not initially consented for supraglottoplasty, and his parents could not be located anywhere in the hospital or by telephone to give this consent. He therefore underwent adenotonsillectomy alone, and was admitted to the hospital overnight for observation. He continued to have airway obstruction through the night, and the parents agreed to have him return to the operating room on post-operative day one for supraglottoplasty. The next night, he was noted to have significant improvement in his breathing and the family noted continued subjective improvement with more energy and improved behavior at home. Another patient was initially treated with supraglottoplasty alone, only later to require adenotonsillectomy for continued symptoms. After the second intervention, the patient was clinically improved. These two cases further confirm the multifactorial nature of pediatric OSAS. Based on our experience, it is often difficult to determine the relative contributions of each site of obstruction, and the added morbidity of performing supraglottoplasty and adenotonsillectomy in the same setting must be weighed against the possibility of needing a second surgery in the future. Additional studies in the future will hopefully shed more light on this topic and provide more definitive recommendations. Pediatric obstructive sleep apnea syndrome can be seen in children of all ages and may present in different ways. Previous
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
studies have shown that supraglottoplasty is effective in the treatment of infants [15,21,22] and school-aged children [18] with laryngomalacia and OSAS, as was also seen in our patient population. Our study furthers the description of late-onset laryngomalacia, as our larger numbers (19) and identification of patients prior to empiric adenotonsillectomy allowed for improved assessment of the true presenting features of these patients and how they compare to the presentation of patients without laryngomalacia. It also illuminates the dilemma involved in deciding which sites of obstruction to address surgically in a multi-factorial disease process. While there were no surgical complications in any of these patients, the morbidity of a postoperative hemorrhage after tonsillectomy or supraglottic stenosis after supraglottoplasty cannot be overlooked and must be considered in the decision-making process. It would be reasonable to argue that the negative pressure generated by a child attempting to breathe with obstructing tonsils may induce supraglottic collapse and that by removing the pharyngeal obstruction, there may be spontaneous resolution of the laryngomalacia. However, as is shown with one of the patients in this study, although the adenotonsillar hypertrophy may have induced the tissue redundancy and supraglottic collapse, merely removing the tonsils cannot reverse the tissue changes, and symptomatic laryngomalacia may continue until it is addressed surgically. The results of the recent study by Bhattacharjee et al. [13] suggest that patients greater than seven years old, with increasing BMI, with increasing AHI, and with asthma are at higher risk for persistent disease. This may help surgeons in the future decide to be more aggressive surgically in these patients, choosing to perform airway endoscopy in patients at high risk for persistent disease and address both laryngomalacia and adenotonsillar hypertrophy when identified. The main limitation of this study is its retrospective nature. There was no standardized way to collect data in the clinic or in the operating room, such that not all findings were reported in a consistent manner. Significant gaps in data, especially in the postoperative setting, make it impossible to make statistical evaluations of data and limit the study to being purely descriptive in nature. Demographic data is largely missing as well, although a large number of patients treated in this clinic are of Hispanic heritage and are funded by Medicaid. This may also limit the application of findings to a larger, more diverse population. It is possible that some of these patients had laryngomalacia as infants that was never diagnosed, or was not reported in a way that could be accessed by a retrospective study, and so the diagnosis would be a re-emergence or recurrence of larygomalacia in an older child rather than a true late-onset variety. There was also no documentation on sleep study reports regarding the presence or absence of stridor during the overnight study to help determine if this may help identify these patients pre-operatively. An argument may be made that evaluating the patient under general anesthesia with a rigid laryngoscope alters the anatomy and physiology of the larynx. Propofol has been widely used in sleep endoscopy and the adult literature suggests that while it alters normal sleep patterns, failing to induce REM sleep, it does not induce snoring in patients without OSAS and does not significantly alter the AHI [28]. It is possible that the methods used in this study actually underestimate the finding of laryngomalacia in these children, as the rigid endoscope may stent the airway open and the anesthetic agent does not replicate the decreased neuromuscular tone of REM sleep. Future studies in pediatric OSAS should continue to consider dynamic laryngeal pathology as a cause of or contributing factor to OSAS. As more multi-institutional prospective studies are planned, care should be taken to account for this pathology. Sleep center technicians should be trained to differentiate stridor from snoring, and this should be a routine finding noted on the polysomnography report. Because the current standard of care for pediatric OSAS is to
237
perform adenotonsillectomy, a future prospective study could perform routine sedated fiberoptic laryngoscopy prior to all adenotonsillectomies and document the presence or absence of supraglottic collapse. These patients could then be followed postoperatively to determine which patients failed to have normalization of breathing patterns and correlate these findings to the preoperative fiberoptic exam findings. This could help future providers determine the need for routine endoscopy in all patients prior to adenotonsillectomy, to a select high-risk group of patients prior to adenotonsillectomy, or only after a patient has failed to improve after adenotonsillectomy. 5. Conclusion Late-onset laryngomalacia may be a cause of or contributing factor to airway obstruction in children with obstructive sleep apnea syndrome. When seen in older children, the presentation of laryngomalacia is more similar to that of patients with OSAS without laryngomalacia than to younger patients with congenital laryngomalacia. While no specific diagnostic criteria were identified by this study, late-onset laryngomalacia is readily diagnosed by direct laryngoscopy and can be treated safely and effectively with supraglottoplasty. The identification of this disease entity in otherwise healthy children suggests that the diagnosis should be considered in children with OSAS, especially if they have failed initial management with adenotonsillectomy. Acknowledgements The authors would like to thank Dr. Karen Hentschel-Franks and her staff at the Children’s Sleep Center at Christus Santa Rosa Hospital for their help in evaluating and treating patients with obstructive sleep apnea. References [1] R.M. Ray, C.M. Bower, Pediatric obstructive sleep apnea: the year in review, Curr. Opin. Otolaryngol. Head Neck Surg. 13 (2005) 360–365. [2] N.J. Ali, D.J. Pitson, J.R. Stradling, Snoring, sleep disturbance, and behaviour in 4–5 year olds, Arch. Dis. Child. 68 (1993) 360–366. [3] American Academy of Pediatrics, Clinical practice guideline: diagnosis and management of childhood obstructive sleep apnea syndrome, Pediatrics 109 (2002) 704–712. [4] C.E. Ievers-Landis, S. Redline, Pediatric sleep apnea: implications of the epidemic of childhood overweight, Am. J. Respir. Crit. Care Med. 175 (5) (2007) 436–441. [5] M. Benninger, D. Walner, Obstructive sleep-disordered breathing in children, Clin. Cornerstone 9 (Suppl. 1) (2007) S6–S12. [6] V. Kirk, A. Kahn, R.T. Brouillette, Diagnostic approach to obstructive sleep apnea in children, Sleep Med. Rev. 2 (1998) 255–269. [7] R. Arens, C.L. Marcus, Pathophysiology of upper airway obstruction: a developmental perspective, Sleep 27 (5) (2004) 997–1019. [8] N.A. Goldstein, M. Fatima, T.F. Campbell, R.M. Rosenfeld, Child behavior and quality of life before and after tonsillectomy and adenoidectomy, Arch. Otolaryngol. Head Neck Surg. 128 (7) (2002) 770–775. [9] L.M. De Serres, C. Derkay, K. Sie, et al., Impact of adenotonsillectomy on quality of life in children with obstructive sleep disorders, Arch. Otolaryngol. Head Neck Surg. 128 (5) (2002) 489–496. [10] M.S. Schechter for the Section on Pediatric Pulmonology, Subcommittee on Obstructive Sleep Apnea Syndrome, Technical Report: Diagnosis and management of childhood obstructive sleep apnea syndrome, Pediatrics 109 (4) (2002) e69. [11] R. Tauman, T.E. Gulliver, J. Krishna, et al., Persistence of obstructive sleep apnea syndrome in children after adenotonsillectomy, J. Pediatr. 149 (6) (2006) 803–808. [12] R.B. Mitchell, Adenotonsillectomy for obstructive sleep apnea in children: outcome evaluated by pre- and post-operative polysomnography, Laryngoscope 117 (9) (2007) 1844–1854. [13] R. Bhattacharjee, L. Kheirandish-Gozal, K. Spruyt, et al., Adenotonsillectomy outcomes in treatment of obstructive sleep apnea in children: a multicenter retrospective study, Am. J. Respir. Crit. Care Med. 182 (5) (2010) 676–683. [14] R.B. Mitchell, J. Kelly, Outcome of adenotonsillectomy for obstructive sleep apnea in obese and normal-weight children, Otolaryngol. Head Neck Surg. 137 (1) (2007) 43–48. [15] F.C. Pereira Valera, E. Tamashiro, M.M. de Araujo, H.H. Sander, D.S. Kupper, Evaluation of the efficacy of supraglottoplasty in obstructive sleep apnea syndrome associated with severe laryngomalacia, Arch. Otolaryngol. Head Neck Surg. 132 (2006) 489–493.
238
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
[16] D.R. Olney, J.H. Greinwald Jr., R.J. Smith, N.M. Bauman, Laryngomalacia and its treatment, Laryngoscope 109 (1999) 1770–1775. [17] J.L. Smith II, D.M. Sweeney, B. Smallman, A. Mortelliti, State-dependent laryngomalacia in sleeping children, Ann. Otol. Rhinol. Laryngol. 114 (2) (2005) 111–114. [18] G.T. Richter, M.J. Rutter, A. deAlarcon, L.J. Orvidas, D.M. Thompson, Late-onset laryngomalacia: a variant of disease, Arch. Otolaryngol. Head Neck Surg. 134 (1) (2008) 75–80. [19] R.J. Kuczmarski, C.L. Ogden, L.M. Grummer-Strawn, et al., CDC growth charts: United States, Adv. Data 314 (2000) 1–27. [20] C. Iber, S. Ancoli-Israel, A. Chesson, S.F. Quan for the American Academy of Sleep Medicine, The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Termnology and Technical Specifications, 1st ed., American Academy of Sleep Medicine, Westchester, IL, 2007. [21] M.E. Zafereo, J. Taylor, K.D. Pereira, Supraglottoplasty for laryngomalacia with obstructive sleep apnea, Laryngoscope 118 (10) (2008) 1873–1877. [22] T.E. O’Connor, P. Bumbak, S. Vijayasekaran, Objective assessment of supraglottoplasty outcomes using polysomnography, Int. J. Pediatr. Otolaryngol. 73 (9) (2009) 1211–1216.
[23] T.F. Hoban, R.D. Chervin, Sleep-related breathing disorders of childhood: description and clinical picture, diagnosis, and treatment approaches, Sleep Med. Clin. 2 (2007) 445–462. [24] T.F. Hoban, Sleep disorders in children, Ann. N.Y. Acad. Sci. 1184 (2010) 1–14. [25] C.L. Ogden, K.M. Flegal, M.D. Carroll, C.L. Johnson, Prevalence and trends in overweight among US children and adolescents, 1999–2000, JAMA 288 (14) (2002) 1728–1732. [26] E. Dayyat, L. Kheirandish-Gozal, O. Sans Capdevila, M.M. Maarafeya, D. Gozal, Obstructive sleep apnea in children: relative contributions of body mass index and adenotonsillar hypertrophy, Chest 136 (1) (2009) 137–144. [27] B.L. Matthews, J.P. Little, W.F. Mcguirt Jr., J.A. Koufman, Reflux in infants with laryngomalacia: results of 24-hour double pH probe monitoring, Otolaryngol. Head Neck Surg. 120 (6) (1999) 860–864. [28] F.A.W. Rabelo, A. Braga, D.S. Kuppper, et al., Propofol-induced sleep: polysomnographic evaluation of patients with obstructive sleep apnea and controls, Otolaryngol. Head Neck Surg. 142 (2) (2010) 218–224.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Late-onset laryngomalacia: A cause of pediatric obstructive sleep apnea§ Sally M. Revell *, William D. Clark The University of Texas Health Science Center at San Antonio, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 5 July 2010 Received in revised form 2 November 2010 Accepted 3 November 2010 Available online 27 November 2010
Objective: To describe the presentation, diagnosis, and treatment of late-onset laryngomalacia in children with obstructive sleep apnea syndrome (OSAS). Design: Retrospective study. Setting: Tertiary care children’s hospital. Patients: Seventy-seven children were identified who had OSAS diagnosed by polysomnography and underwent airway endoscopy to evaluate for laryngomalacia between July 2006 and December 2008. Children with significant neurologic disease or craniofacial malformations were excluded. Seven children under 3 years of age had laryngomalacia and OSAS (Group A), 19 children 3–18 years of age had laryngomalacia and OSAS (Group B), and 51 children 3–18 years of age had OSAS but not laryngomalacia (Group C). Main outcome measures: Comparison of pre-operative findings, intra-operative findings, interventions, and outcomes between the 3 groups. Results: Group A was consistent with previous reports of congenital laryngomalacia with respect to presentation, diagnosis, and treatment. Groups B and C had similar pre-operative findings, including a high incidence of adenotonsillar hypertrophy, and the only significant difference was the intra-operative finding of laryngomalacia in Group B. Treatments were individualized to include supraglottoplasty, adenoidectomy, tonsillectomy, adenotonsillectomy, or a combination of the above. Of the 52 patients who returned in follow-up, 44 noted improvement, but this was rarely confirmed by polysomnogram. Conclusions: Late-onset laryngomalacia may act alone or in concert with additional dynamic or fixed lesions to cause pediatric OSAS. Although there is no specific pre-operative indicator to diagnose lateonset laryngomalacia, it can be readily identified intra-operatively and effectively treated with supraglottoplasty, with or without concurrent adenotonsillectomy. ß 2010 Elsevier Ireland Ltd. All rights reserved.
Keywords: Laryngomalacia Obstructive sleep apnea Adenotonsillectomy Supraglottoplasty
1. Introduction Obstructive sleep apnea syndrome (OSAS) is a syndrome of partial or complete upper airway obstruction occurring during sleep and causing a disruption of normal ventilation and sleep patterns. While this may occur in patients of any age, the classic description of pediatric OSAS is in a child 2–8 years old with symptoms of loud snoring, witnessed apneas, frequent nighttime arousals, and chronic mouth breathing [1]. Ten percent of children have some form of sleep disordered breathing (including primary snoring) and 1–3% of children have obstructive sleep apnea [2]. These numbers are important because pediatric OSAS is associated with neurocognitive, cardiovascular, and metabolic sequelae that
§ Poster Presentation at the American Society of Pediatric Otolaryngology Spring Meeting Las Vegas, NV, April 30, 2010. * Corresponding author at: 8300 Floyd Curl Drive, MSC 7777, San Antonio, TX 78229, United States. Tel.: +1 210 450 0709; fax: +1 210 562 9374. E-mail address: [email protected] (S.M. Revell).
0165-5876/$ – see front matter ß 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2010.11.007
can have lasting consequences throughout the child’s life. School performance suffers due to behavioral problems, developmental delay, and inattention, while overall health is damaged by systemic inflammation and metabolic derangements, all of which contribute to a poor quality of life for the child. Major risk factors for OSAS in children include adenotonsillar hypertrophy, obesity, neuromuscular disease, and craniofacial anomalies [3–5]. Pediatric OSAS is often attributed to adenotonsillar hypertrophy, yet research suggests the cause is often multifactorial in nature, with decreased neuromuscular tone during REM sleep also playing a large role [1,6,7]. Despite a recognition that the cause of obstruction is multifactorial, adenotonsillectomy continues to be the most common intervention, supported by studies citing improvements in behavior [8] and quality of life [9] and overall cure rates of 75–100% after adenotonsillectomy [10]. As larger studies with better levels of evidence emerge, however, it is now clear that adenotonsillectomy is not a cure-all for pediatric OSAS. Tauman et al. [11] studied 110 patients and found that while all showed improvement in sleep parameters after adenotonsillectomy, only 25% had complete cure of OSAS (defined as an apnea
232
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
hypopnea index (AHI) < 1). Mitchell [12] reported on 79 nonobese, otherwise healthy children with OSAS, all of whom underwent pre- and post-operative polysomnograms. Sixteen percent of children had persistent disease when defined by an AHI greater than or equal to 5 and 27% of children had persistent disease after adenotonsillectomy when defined by a respiratory distress index (RDI) of greater than or equal to 1. The largest and most recent study by Bhattacharjee et al. [13] retrospectively evaluated the efficacy of adenotonsillectomy in 548 children with OSAS. Although 90.1% of patients showed a reduction in AHI, only 27.2% had complete normalization of their breathing patterns during sleep after adenotonsillectomy. Patients more likely to have persistent disease were greater than 7 years old, had a higher body mass index (BMI), had a higher pre-operative AHI, and had asthma. Looking specifically at obese patients, Mitchell and Kelly [14] found that 76% of obese patients had persistent disease after adenotonsillectomy. The results of these studies support the notion of OSAS as being multi-factorial in nature and raise the question of what additional upper airway pathology, whether fixed or dynamic, is contributing to OSAS in children. Congenital laryngomalacia is a well-described dynamic lesion of the larynx, causing collapse of the supraglottic structures during inspiration. It is the most common congenital laryngeal abnormality, presenting with inspiratory stridor during the first 2 weeks of life, worsening over the next 6–8 months, and resolving by 2 years of age [15]. In this patient population, only 10% of affected patients require intervention due to complications such as failure to thrive, obstructive sleep apnea sydrome, resting dyspnea, hypoxia or hypercapnea, pulmonary hypertension, or cor pulmonale [16]. More recently, variations of this typical presentation have been reported, including state-dependent and late-onset laryngomalacia. State-dependent laryngomalacia was described by Smith et al. [17] after identifying 5 patients between the ages of 3 and 4 years old who were asymptomatic while awake, but experienced stridor and airway collapse during sleep. This was thought to be secondary to decreased neuromuscular tone in the sleeping child, causing it to be missed on an office exam including flexible fiberoptic laryngoscopy. Late-onset laryngomalacia was described by Richter et al. [18] in patients older than 2 years of age with sub-categories of feeding-disordered, sleep-disordered, and exercise-induced laryngomalacia. Seven school-aged children presented with sleep disordered breathing, five of whom had already undergone adenotonsillectomy without resolution of symptoms. They were all identified as having sleep-disordered laryngomalacia and improved after treatment with supraglottoplasty. As more data emerges about pediatric obstructive sleep apnea syndrome, more questions arise regarding the pathophysiology of this disease and the appropriate work-up, intervention, and follow-up of these patients. Airway endoscopy has not historically been a part of the initial work-up for pediatric OSAS, so it is possible that late-onset laryngomalacia is present in a number of these patients but is not being recognized. Using the results of routine airway endoscopy in children with OSAS, this study aims to increase awareness of late-onset laryngomalacia as a cause of pediatric OSAS and to compare the presentation, diagnosis, and management of these patients to that of younger children with congenital laryngomalacia and to age-matched children with obstructive sleep apnea but not laryngomalacia. 2. Materials and methods With approval of the institutional review board, a retrospective study was conducted on patients treated in the pediatric otolaryngology clinic at Christus Santa Rosa Children’s Hospital (San Antonio, TX) for laryngomalacia and/or obstructive sleep apnea between July 2006 and December 2008. Only patients with
obstructive sleep apnea diagnosed by polysomnogram and airway endoscopy performed by the senior author to confirm or deny the diagnosis of laryngomalacia were included. Patients with significant craniofacial malformations or neurologic disorders were excluded. Data was collected from inpatient and outpatient electronic and paper charts to include history and physical exam findings, polysomnogram results, operative reports, operative photos and videos, and post-operative course. Seventy-seven patients met inclusion criteria for the study. For the purposes of this study, a cut-off of 3 years of age was used to separate congenital from late-onset laryngomalacia, allowing for division of the patients into three distinct groups. Group A (congenital laryngomalacia) contained patients ages 0–3 years old with obstructive sleep apnea and laryngomalacia, Group B (lateonset laryngomalacia) contained patients ages 3–18 years old with obstructive sleep apnea and laryngomalacia, and Group C (classic pediatric OSAS) contained patients ages 3–18 years old with obstructive sleep apnea but without laryngomalacia. Pre-operative data points included sleep symptoms, medical comorbidities, weight, presence of stridor and/or retractions, and tonsil and adenoid size. Patient weights were plotted on CDC growth curves and standard formulas were used to calculate the growth percentile for each patient. For patients under 24 months of age, weight to length percentiles were used to determine if the patient was underweight (<5th percentile), normal weight (5th to 95th percentile), or overweight (>95th percentile). Patients over 24 months of age were assessed based on body mass index percentiles and divided into underweight (<5th percentile), normal weight (5th to 85th percentile), overweight (85th to 95th percentile), and obese (>95th percentile) categories [19]. Tonsil size was graded from 1 to 4 with size 1 tonsils hidden within the pillars, size 2 tonsils extending to the edge of the pillars, size 3 tonsils extending beyond the pillars, and size 4 tonsils meeting in the midline. Adenoids were simply described as obstructing or nonobstructing. Mild laryngomalacia was diagnosed if the supraglottic collapse was intermittent or led to incomplete obstruction of the laryngeal inlet while severe laryngomalacia was diagnosed if the collapse was persistent with complete obstruction. All patients underwent pre-operative full-night polysomnogram, most of which were conducted by a pediatric pulmonologist at the same children’s hospital. Respiratory events were scored based on American Academy of Sleep Medicine guidelines [20] and disease was categorized as mild (AHI 1–5), moderate (AHI 5–10), or severe (AHI > 10) for each patient. Direct laryngoscopy and tracheobroncoscopy were performed with each patient spontaneously ventilating under intravenous anesthesia. A propofol drip was started at 200–250 mcg/kg/min and titrated to achieve deep sedation. Direct laryngoscopy was then performed with a Parson’s laryngoscope, a weight-appropriate dose of 2% lidocaine was applied to the vocal cords, and then tracheobronchoscopy was performed with a Hopkin’s rod telescope. During this period, observations were made regarding adenotonsillar size, cobblestoning of the tracheal mucosa, and presence of any other airway pathology. Once this had been completed, the laryngoscope was kept in the vallecula while the anesthesiologist turned off the propofol, allowing for evaluation of dynamic laryngeal pathology as the patient began to awaken from anesthesia and produced vigorous inspiratory efforts. During this period, the aryepiglottic folds were observed for collapse with inspiration allowing for diagnosis of mild or severe laryngomalacia or no laryngomalacia. Indications for supraglottoplasty in a child with polysomnogram-proven OSAS included severe laryngomalacia with or without adenotonsillar hypertrophy or mild laryngomalacia
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
without adenotonsillar hypertrophy. When indicated, this was performed using laryngeal micro-scissors to excise portions of the aryepiglottic folds deemed to be redundant, with or without the associated cuneiform cartilage. Indications for adenotonsillectomy included a diagnosis of OSAS with adenotonsillar hypertrophy, with or without associated laryngomalacia. If necessary, tonsillectomy was performed with needle-tip electrocautery and adenoidectomy was performed with a microdebrider. If both laryngomalacia and adenotonsillar hypertrophy were diagnosed, treatment was individualized, as will be further addressed in the discussion section. Any children with continued symptoms and positive polysomnography after the completion of surgical interventions were referred back to the sleep center for medical management.
233
Table 1 Comorbidities.
Overweight/obesity Asthma Allergic rhinitis Attention deficit hyperactivity disorder Reflux disease Prematurity Diabetes mellitus Congential heart disease History of respiratory syncitial virus Speech delay Recurrent pneumonia Cystic fibrosis Hypertension Failure to thrive
Group A
Group B
Group C
1
12 10 6 6 2
25 18 10 4 4 4 4 2 2 1 1 1 1 1
1 1
3. Results
[()TD$FIG]
After excluding patients with significant neurologic disease and craniofacial anomalies, the most common comorbidities were overweight/obesity, asthma, allergic rhinitis, attention deficit hyperactivity disorder, and gastroesophageal reflux disease (Table 1). Demographic information was not consistently recorded and is therefore not available to report. There were no surgical complications noted for any of the study patients. The findings of each group are described below and represented in Fig. 1.
3.1. Group A Seven patients under 3 years of age were identified as having both laryngomalacia and obstructive sleep apnea syndrome. The patients ranged in age from 3 to 10 months with a mean age of 6 months. All seven patients were male. The most common presenting symptoms in this group were stridor (5), witnessed apnea (5), snoring (3), noisy breathing (2), feeding difficulties (2),
Fig. 1. Treatment group comparisons for pre-operative data. Data are presented as: (a) percentage of patients with each presenting symptoms, (b) percentage of patients in each weight category, and (c) percentage of patients with mild, moderate, and severe OSA by polysomnogram.
[()TD$FIG]
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
234
Fig. 2. Congenital laryngomalacia. Photos demonstrate (a) shortened aryepiglottic folds and (b) redundant mucosa of the aryepiglottic folds.
Additional operative findings included cobblestoning of the tracheal mucosa in 4 patients and an antrochoanal polyp in one patient. Four patients had severe laryngomalacia with varying degrees of adenotonsillar hypertrophy. They underwent supraglottoplasty alone. One patient reported some subjective improvement, but had no objective studies performed. Another had continued snoring and behavior problems but a pre-operative AHI of 2.5 dropped to 0.3 and no additional interventions were required. A third patient had subjective improvement and resolution of snoring, but went on to have a tonsillectomy and adenoidectomy 9 months later for recurrent tonsillitis. The last patient had no improvement in symptoms, and a repeat sleep study was performed. It showed continued severe obstructive sleep apnea, so the patient underwent adenotonsillectomy, after which snoring resolved. Eight patients had obstructing adenotonsillar tissue with varying degrees of laryngomalacia. One patient in this group had severe adenotonsillar hypertrophy and laryngomalacia, but only adenotonsillectomy was performed initially due to lack of consent for the supraglottoplasty. He did not improve until he underwent supraglottoplasty as well, as will be further described in the discussion section. The remaining seven patients in this group had adenotonsillar hypertrophy as the predominant obstructing pathology with mild laryngomalacia, so they underwent tonsillectomy or adenotonsillectomy without supraglottoplasty. Five had subjective improvement with 2 of these demonstrating complete resolution by sleep study (AHI 15.4–0.1 and 4.5–0.1). One had continued somnolence, enuresis, and
mouth breathing (2), and cyanosis (1). One patient was underweight, 5 patients were of normal weight, and 1 patient was overweight. Physical exam findings included stridor (4) and retractions (1) and no patients were noted to have adenotonsillar hypertrophy. OSA severity was mild in 2 patients, moderate in 2 patients, and severe in 3 patients. Intra-operative findings were significant for laryngomalacia alone in 6 patients and for laryngomalacia in addition to tracheomalacia and bronchomalacia in 1 patient. Laryngomalacia in these patients was characterized by shortened aryepiglottic folds with redundant overlying mucosa that was pulled into the airway resulting in airway obstruction (Fig. 2). This was most commonly noted during emergence from anesthesia and was accompanied by stridor and substernal retractions. Four of the seven patients underwent supraglottoplasty and three returned for follow-up. One had subjective improvement but no post-operative polysomnogram and another had subjective improvement and polysomnogram revealed improvement in apnea hypopnea index from 13.1 to 2.3. The last had subjective improvement but residual symptoms prompted re-evaluation. Repeat endoscopy showed no collapse and polysomnography revealed a decrease in AHI from 6.1 to 1.0. 3.2. Group B Nineteen patients between 3 and 18 years of age were identified with a diagnosis of laryngomalacia and obstructive sleep apnea syndrome. The age range was 3.8–11.1 years with a mean age of 7.3 years. Eight patients were male and 11 were female. The most common presenting symptoms were snoring (17), apnea (15), daytime somnolence (12), difficulty awakening (10), mouth breathing (9), behavior problems (8), gasping/choking during sleep (7), enuresis (6), difficulty feeding (2), recurrent tonsillitis (2), noisy breathing (1), stridor (1) and cyanosis (1). One patient was underweight, 6 patients were of normal weight, no patients were overweight, and 12 patients were obese. Tonsil and adenoid size were recorded in 18 and 12 patients, respectively. Tonsillar hypertrophy (size 2 or greater) was noted in 89% of patients (11% size 1, 28% size 2, 39% size 3, and 22% size 4). Adenoids were nonobstructing in 33% of patients and obstructing in 67%. None of these patients had stridor or retractions noted on office exam. Polysomnogram severity was mild in 10 patients, moderate in 5 patients, and severe in 4 patients. All of the patients in this group, by definition, were also found to have laryngomalacia. The predominant feature in these patients was the redundant mucosa of the aryepiglottic folds that was pulled into the airway, causing obstruction during forceful inspiration (Fig. 3). Shortened aryepiglottic folds were a less common feature in this population.
[()TD$FIG]
Fig. 3. Late-onset laryngomalacia. Photo demonstrates the typical finding of redundant mucosa of the aryepiglottic folds.
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
behavior problems, but repeat endoscopy was normal and polysomnogram showed an improvement in AHI from 6 to 1.3. The last patient in this group was lost to follow-up. Three patients had significant adenotonsillar hypertrophy and laryngomalacia. They underwent supraglottoplasty plus tonsillectomy or adenotonsillectomy. All three reported significant improvement with resolution of symptoms, but no post-operative polysomnograms were performed. Another patient also had a fixed obstruction in the form of an antrochoanal polyp. He had been treated with adenotonsillectomy for sleep-disordered breathing at another facility and was then referred when symptoms persisted. He was diagnosed with severe OSAS by polysomnogram (AHI 39.3) and scheduled for direct laryngoscopy and tracheobronchoscopy. He had mild laryngomalacia in addition to a large, obstructing antrochoanal polyp. He underwent endoscopic removal of the polyp with subjective improvement and a post-operative polysomnogram revealed a decrease in AHI from 39.3 pre-operatively to 2.2 post-operatively. No intervention was performed for the mild laryngomalacia. Three patients did not undergo any surgical intervention. Two were offered supraglottoplasty for laryngomalacia, but the families declined, and the other had mild, intermittent laryngomalacia, no obstructing adenotonsillar tissue, and mild OSAS, so no intervention was recommended. None of these patients returned for follow-up. 3.3. Group C Fifty-one patients between 3 and 18 years of age were identified with obstructive sleep apnea but without laryngomalacia. Their ages ranged from 3.0 to 16.6 years with a mean age of 7.2 years. There were 26 male and 25 female patients. The main presenting symptoms were snoring (45), apnea (33), difficulty wakening (29), daytime somnolence (22), mouth breathing (21), behavior problems (10), recurrent tonsillitis (9), enuresis (9), gasping/ choking during sleep (6), poor weight gain (4), noisy breathing (4), difficulty feeding (2), stridor (2), cyanosis (1). Three patients were underweight, 23 patients were of normal weight, 7 patients were overweight, and 18 patients were obese. Tonsil and adenoid size were recorded in 46 and 35 patients, respectively. Tonsillar hypertrophy (size 2 or greater) was noted in 91% of patients (9% size 1 tonsils, 33% size 2, 39% size 3, and 20% size 4). Adenoids were non-obstructing in 23% of patients and obstructing in 77%. No patients were noted to have stridor or retractions during office exam. OSAS severity was available for 49 patients with disease being mild in 23 patients, moderate in 12 patients, and severe in 14 patients. None of these patients had any evidence of laryngomalacia on airway endoscopy. Additional pathology included 15 patients with cobblestoning of the tracheal mucosa and 2 patients with nodules on the true vocal folds. Thirty-eight patients underwent adenotonsillectomy and 28 of them returned for follow-up. Twenty-one reported subjective improvement and underwent no further testing. Four patients reported subjective improvement which was confirmed by postoperative polysomnography (AHI 9.1–0.6, AHI 19.4–2.2, AHI 18.2– 1.3, AHI 53.2–0.2). Three additional patients only followed up in the sleep clinic and subjective data is not available. Of these, one patient continued to have mild OSAS (AHI 3–1.6), another dropped from severe to mild OSAS (AHI 16.4–3.5), and the last continued to have severe OSAS (AHI 18.5–13.7). These patients continued to follow-up with the sleep clinic and were treated with medical management. Five patients underwent tonsillectomy alone for isolated tonsillar hypertrophy. Two patients returned in follow-up, both of whom had subjective improvement. One of these had a repeat polysomnogram which confirmed a decrease in AHI from 12.6 to
235
1.4, but the other had no post-operative studies performed. Three did not return for follow-up. Three patients underwent adenoidectomy alone. One reported subjective improvement initially, but then apnea returned. A repeat polysomnogram showed improvement (AHI 11.7–3.6), but continued disease prompted the initiation medical management by the sleep medicine clinic. The second patient reported subjective improvement but had no additional studies performed, and the third did not return for follow-up. Five patients did not show significant adenotonsillar hypertrophy or any other surgically correctable site of obstruction, so no surgical interventions were performed. These patients were all referred back to the sleep clinic for medical management. 4. Discussion While the otolaryngology literature has a large volume of data on laryngomalacia and pediatric obstructive sleep apnea syndrome, with both being recognized as common diseases of childhood, the interaction between these two diseases has been less well studied. The association between the laryngomalacia and OSAS has been best described in the infant population. In these patients, laryngomalacia, with its tell-tale stridor, is often diagnosed first, with subsequent polysomnography being used to demonstrate the resulting obstructive sleep apnea and to prompt the intervention of supraglottoplasty for treatment. Zafereo et al. [21], O’Connor et al. [22] and Pereira Valera et al. [15] have reported statistically significant improvements in polysomnography parameters, to include respiratory distress index [15,21,22] and lowest oxygen saturation levels, [21,22] in patients who underwent supraglottoplasty as a treatment for moderate to severe laryngomalacia. These studies and their results can be correlated with Group A in this study where the most common symptoms were stridor (71%) and apnea (71%), most patients were of normal weight, no patients had adenotonsillar hypertrophy, and supraglottoplasty was an effective treatment. In contrast to infants with congenital laryngomalacia and obstructive sleep apnea, older children are first identified by their symptoms of sleep disordered breathing that lead to adenotonsillectomy with or without pre-operative polysomnogram. While this is an accepted practice, there has been recent evidence that adenotonsillectomy is not as successful at curing pediatric OSAS as was once believed. This is likely due to the multi-factorial nature of the obstruction, some of which is dynamic and some of which is fixed in nature. It is for this reason that every child with symptoms of sleep disordered breathing at our institution undergoes intraoperative direct laryngoscopy and tracheobronchoscopy prior to adenotonsillectomy. This allows for the separation of Groups B and C as each patient in Group C has been evaluated for laryngomalacia and found not to have it. This is an important distinction from previous studies, as many patients treated elsewhere will only undergo endoscopy after failing empiric treatment with adenotonsillectomy. On initial presentation, the patients in both Groups B and C fall under the more classic description of pediatric obstructive sleep apnea syndrome. As it is classically defined, pediatric OSAS has a peak incidence between 2 and 8 years of age, occurs equally in males and females, is characterized by snoring, mouth breathing, and restless sleep, and is associated with normal body weight and adenotonsillar hypertrophy. As the patients move toward adolescence, patients are more commonly male and obese, but continue to have the same symptom complex of snoring, mouth breathing, and restless sleep [23,24]. Both Groups B and C in this study revealed a similar incidence among males and females and had a mean ages of 7.3 and 7.2 years old, respectively. The most common presenting symptoms for Groups B and C were also in accordance
236
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
with previous studies. They included snoring (89%, 88%), apnea (79%, 65%), daytime somnolence (63%, 43%), difficulty awakening (53%, 57%), and mouth breathing (47%, 41%). There were differences between Groups B and C in less commonly reported symptoms, such as behavior problems, enuresis, and gasping/ choking during sleep, but little can be deduced from this, as not all patients were specifically asked about every symptom. The distribution of OSAS severity was also similar between the groups, with Group B having 53% mild, 26% moderate, and 21% severe disease, while Group C had 47% mild, 24% moderate, and 29% severe disease. The distribution of adenotonsillar hypertrophy was also similar, with 89% of Group B and 91% of Group C having tonsillar hypertrophy (tonsil size 2 or greater) and 67% of Group B and 77% of Group C having adenoid hypertrophy (obstructing). While obesity is recognized as a risk factor for pediatric OSAS, this is not part of the ‘‘classic’’ presentation of pediatric OSAS. In this study, the percentage of overweight and obese patients is high in both Group B (63%) and Group C (49%), which is not surprising given the three-fold increase in obesity in school-aged children and adolescents in the United States since 1980 [25]. Dayyat et al. [26] reported that obese patients are likely to have an ‘‘adult-like’’ presentation and disease course, which is supported by the presence of symptoms like daytime somnolence and difficulty wakening in this study population. The numbers in this study are too small to show a difference in presentation and disease course between the normal weight and obese children, but Mitchell and Kelly [14] and Bhattacharjee et al. [13] have demonstrated that increasing BMI is associated with a higher incidence of persistent disease than is seen in normal weight children. Larger studies will be needed to evaluate the relationship between obesity and lateonset laryngomalacia. This study makes a distinction between groups based on the findings on airway endoscopy. The patients who were found to have laryngomalacia were placed into Group B and diagnosed with late-onset laryngomalacia. Late-onset laryngomalacia, as described by Richter et al. [18] has a different symptom complex and disease course than congenital laryngomalacia. He described a group of 7 school-aged children with symptoms of sleep disturbance who were found to have type I laryngomalacia (redundant mucosa prolapsing over the arytenoids into the airway) and were treated successfully with supraglottoplasty. He further classified these patients as having ‘‘sleep-disordered laryngomalacia.’’ All 19 patients in Group B of this study would fall into this category, as they presented with typical pediatric OSAS symptoms, were found to have laryngomalacia on airway endoscopy, and were often treated successfully with supraglottoplasty. The main difference between this group and Richter’s group is the weight. He reported a mean weight percentile of 37% [18] while mean body mass index percentile in Group B of this study was 79%. This could be due to a regional variation in overweight and obese children, or may simply reflect small case numbers in each study. In the patients with late-onset laryngomalacia (Group B), the findings on airway endoscopy were more similar to the patients with congenital laryngomalacia (Group A) while the patient presentation was much more similar to that of the children and adolescents with pediatric OSAS without laryngomalacia (Group C). Another interesting finding is the very natural break between the ages of the patients in Groups A and B. The initial study design used a cut-off of 3 years of age to separate congenital laryngomalacia (typically in patients younger than 2 years of age) from late-onset laryngomalacia to ensure that no patients with congenital laryngomalacia would be grouped with the lateonset patients. When the data was analyzed, however, the oldest patient with congenital laryngomalacia and OSAS was 10 months old while the youngest patient with late-onset laryngomalacia and
OSAS was 3 years 9 months old. This further supports the concept that late-onset laryngomalacia is a distinct clinical entity from congenital laryngomalacia. There is no clear explanation for the cause of this atypical laryngomalacia, especially in patients without underlying neuromuscular disorders. Richter et al. [18] suggested that gastro-esophageal reflux disease (GERD), decreased laryngeal tone, and supraglottic edema may all contribute and exacerbate one another in a cyclical fashion. This is supported by research in the infant population showing that no matter the etiology of laryngomalacia, the negative pressure generated to overcome the obstruction worsens airway edema, subsequently worsening the supraglottic airway obstruction of laryngomalacia [27]. While not directly studied, it could be hypothesized that negative pressure due to any upper airway obstruction (including adenotonsillar hypertrophy) could lead to reflux and subsequent airway edema that may be severe enough to lead to the supraglottic collapse of laryngomalacia. Although only one patient in Group B had been formally diagnosed with GERD, findings of cobblestoning of the tracheal mucosa in 21% of these patients supports a role for reflux disease in the pathophysiology of lateonset laryngomalacia. It is also possible, given the large number of obese patients in our study, that obesity itself may play a role in the pathophysiology, especially given the relationship between GERD and obesity. In this study, clinically significant late-onset laryngomalacia was more common in obese children, with 6 of the 8 patients (75%) who required supraglottoplasty being obese. However, given the small numbers of patients involved, additional studies will be needed to further elucidate this point. While previous studies only sought unusual pathology after a patient had failed adenotonsillectomy, our study is unique in that the only patient who had undergone previous adenotonsillectomy was the patient with the antrochoanal polyp. In all other cases, an intra-operative decision was made as to the relative contributions of adenotonsillar hypertrophy (fixed obstruction) and laryngomalacia (dynamic obstruction) to each patient’s obstructive pathology. Of the 15 patients in Group B who returned for follow-up, 3 patients improved with supraglottoplasty alone, 6 patients improved after adenotonsillectomy (or isolated tonsillectomy or adenoidectomy), and 3 improved after concurrent supraglottoplasty and tonsillectomy or adenotonsillectomy. Two patients required additional procedures prior to improving. One was a patient who was found to have severe adenotonsillar hypertrophy (tonsil size 4 and obstructing adenoids) and severe laryngomalacia on direct laryngoscopy. He was not initially consented for supraglottoplasty, and his parents could not be located anywhere in the hospital or by telephone to give this consent. He therefore underwent adenotonsillectomy alone, and was admitted to the hospital overnight for observation. He continued to have airway obstruction through the night, and the parents agreed to have him return to the operating room on post-operative day one for supraglottoplasty. The next night, he was noted to have significant improvement in his breathing and the family noted continued subjective improvement with more energy and improved behavior at home. Another patient was initially treated with supraglottoplasty alone, only later to require adenotonsillectomy for continued symptoms. After the second intervention, the patient was clinically improved. These two cases further confirm the multifactorial nature of pediatric OSAS. Based on our experience, it is often difficult to determine the relative contributions of each site of obstruction, and the added morbidity of performing supraglottoplasty and adenotonsillectomy in the same setting must be weighed against the possibility of needing a second surgery in the future. Additional studies in the future will hopefully shed more light on this topic and provide more definitive recommendations. Pediatric obstructive sleep apnea syndrome can be seen in children of all ages and may present in different ways. Previous
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
studies have shown that supraglottoplasty is effective in the treatment of infants [15,21,22] and school-aged children [18] with laryngomalacia and OSAS, as was also seen in our patient population. Our study furthers the description of late-onset laryngomalacia, as our larger numbers (19) and identification of patients prior to empiric adenotonsillectomy allowed for improved assessment of the true presenting features of these patients and how they compare to the presentation of patients without laryngomalacia. It also illuminates the dilemma involved in deciding which sites of obstruction to address surgically in a multi-factorial disease process. While there were no surgical complications in any of these patients, the morbidity of a postoperative hemorrhage after tonsillectomy or supraglottic stenosis after supraglottoplasty cannot be overlooked and must be considered in the decision-making process. It would be reasonable to argue that the negative pressure generated by a child attempting to breathe with obstructing tonsils may induce supraglottic collapse and that by removing the pharyngeal obstruction, there may be spontaneous resolution of the laryngomalacia. However, as is shown with one of the patients in this study, although the adenotonsillar hypertrophy may have induced the tissue redundancy and supraglottic collapse, merely removing the tonsils cannot reverse the tissue changes, and symptomatic laryngomalacia may continue until it is addressed surgically. The results of the recent study by Bhattacharjee et al. [13] suggest that patients greater than seven years old, with increasing BMI, with increasing AHI, and with asthma are at higher risk for persistent disease. This may help surgeons in the future decide to be more aggressive surgically in these patients, choosing to perform airway endoscopy in patients at high risk for persistent disease and address both laryngomalacia and adenotonsillar hypertrophy when identified. The main limitation of this study is its retrospective nature. There was no standardized way to collect data in the clinic or in the operating room, such that not all findings were reported in a consistent manner. Significant gaps in data, especially in the postoperative setting, make it impossible to make statistical evaluations of data and limit the study to being purely descriptive in nature. Demographic data is largely missing as well, although a large number of patients treated in this clinic are of Hispanic heritage and are funded by Medicaid. This may also limit the application of findings to a larger, more diverse population. It is possible that some of these patients had laryngomalacia as infants that was never diagnosed, or was not reported in a way that could be accessed by a retrospective study, and so the diagnosis would be a re-emergence or recurrence of larygomalacia in an older child rather than a true late-onset variety. There was also no documentation on sleep study reports regarding the presence or absence of stridor during the overnight study to help determine if this may help identify these patients pre-operatively. An argument may be made that evaluating the patient under general anesthesia with a rigid laryngoscope alters the anatomy and physiology of the larynx. Propofol has been widely used in sleep endoscopy and the adult literature suggests that while it alters normal sleep patterns, failing to induce REM sleep, it does not induce snoring in patients without OSAS and does not significantly alter the AHI [28]. It is possible that the methods used in this study actually underestimate the finding of laryngomalacia in these children, as the rigid endoscope may stent the airway open and the anesthetic agent does not replicate the decreased neuromuscular tone of REM sleep. Future studies in pediatric OSAS should continue to consider dynamic laryngeal pathology as a cause of or contributing factor to OSAS. As more multi-institutional prospective studies are planned, care should be taken to account for this pathology. Sleep center technicians should be trained to differentiate stridor from snoring, and this should be a routine finding noted on the polysomnography report. Because the current standard of care for pediatric OSAS is to
237
perform adenotonsillectomy, a future prospective study could perform routine sedated fiberoptic laryngoscopy prior to all adenotonsillectomies and document the presence or absence of supraglottic collapse. These patients could then be followed postoperatively to determine which patients failed to have normalization of breathing patterns and correlate these findings to the preoperative fiberoptic exam findings. This could help future providers determine the need for routine endoscopy in all patients prior to adenotonsillectomy, to a select high-risk group of patients prior to adenotonsillectomy, or only after a patient has failed to improve after adenotonsillectomy. 5. Conclusion Late-onset laryngomalacia may be a cause of or contributing factor to airway obstruction in children with obstructive sleep apnea syndrome. When seen in older children, the presentation of laryngomalacia is more similar to that of patients with OSAS without laryngomalacia than to younger patients with congenital laryngomalacia. While no specific diagnostic criteria were identified by this study, late-onset laryngomalacia is readily diagnosed by direct laryngoscopy and can be treated safely and effectively with supraglottoplasty. The identification of this disease entity in otherwise healthy children suggests that the diagnosis should be considered in children with OSAS, especially if they have failed initial management with adenotonsillectomy. Acknowledgements The authors would like to thank Dr. Karen Hentschel-Franks and her staff at the Children’s Sleep Center at Christus Santa Rosa Hospital for their help in evaluating and treating patients with obstructive sleep apnea. References [1] R.M. Ray, C.M. Bower, Pediatric obstructive sleep apnea: the year in review, Curr. Opin. Otolaryngol. Head Neck Surg. 13 (2005) 360–365. [2] N.J. Ali, D.J. Pitson, J.R. Stradling, Snoring, sleep disturbance, and behaviour in 4–5 year olds, Arch. Dis. Child. 68 (1993) 360–366. [3] American Academy of Pediatrics, Clinical practice guideline: diagnosis and management of childhood obstructive sleep apnea syndrome, Pediatrics 109 (2002) 704–712. [4] C.E. Ievers-Landis, S. Redline, Pediatric sleep apnea: implications of the epidemic of childhood overweight, Am. J. Respir. Crit. Care Med. 175 (5) (2007) 436–441. [5] M. Benninger, D. Walner, Obstructive sleep-disordered breathing in children, Clin. Cornerstone 9 (Suppl. 1) (2007) S6–S12. [6] V. Kirk, A. Kahn, R.T. Brouillette, Diagnostic approach to obstructive sleep apnea in children, Sleep Med. Rev. 2 (1998) 255–269. [7] R. Arens, C.L. Marcus, Pathophysiology of upper airway obstruction: a developmental perspective, Sleep 27 (5) (2004) 997–1019. [8] N.A. Goldstein, M. Fatima, T.F. Campbell, R.M. Rosenfeld, Child behavior and quality of life before and after tonsillectomy and adenoidectomy, Arch. Otolaryngol. Head Neck Surg. 128 (7) (2002) 770–775. [9] L.M. De Serres, C. Derkay, K. Sie, et al., Impact of adenotonsillectomy on quality of life in children with obstructive sleep disorders, Arch. Otolaryngol. Head Neck Surg. 128 (5) (2002) 489–496. [10] M.S. Schechter for the Section on Pediatric Pulmonology, Subcommittee on Obstructive Sleep Apnea Syndrome, Technical Report: Diagnosis and management of childhood obstructive sleep apnea syndrome, Pediatrics 109 (4) (2002) e69. [11] R. Tauman, T.E. Gulliver, J. Krishna, et al., Persistence of obstructive sleep apnea syndrome in children after adenotonsillectomy, J. Pediatr. 149 (6) (2006) 803–808. [12] R.B. Mitchell, Adenotonsillectomy for obstructive sleep apnea in children: outcome evaluated by pre- and post-operative polysomnography, Laryngoscope 117 (9) (2007) 1844–1854. [13] R. Bhattacharjee, L. Kheirandish-Gozal, K. Spruyt, et al., Adenotonsillectomy outcomes in treatment of obstructive sleep apnea in children: a multicenter retrospective study, Am. J. Respir. Crit. Care Med. 182 (5) (2010) 676–683. [14] R.B. Mitchell, J. Kelly, Outcome of adenotonsillectomy for obstructive sleep apnea in obese and normal-weight children, Otolaryngol. Head Neck Surg. 137 (1) (2007) 43–48. [15] F.C. Pereira Valera, E. Tamashiro, M.M. de Araujo, H.H. Sander, D.S. Kupper, Evaluation of the efficacy of supraglottoplasty in obstructive sleep apnea syndrome associated with severe laryngomalacia, Arch. Otolaryngol. Head Neck Surg. 132 (2006) 489–493.
238
S.M. Revell, W.D. Clark / International Journal of Pediatric Otorhinolaryngology 75 (2011) 231–238
[16] D.R. Olney, J.H. Greinwald Jr., R.J. Smith, N.M. Bauman, Laryngomalacia and its treatment, Laryngoscope 109 (1999) 1770–1775. [17] J.L. Smith II, D.M. Sweeney, B. Smallman, A. Mortelliti, State-dependent laryngomalacia in sleeping children, Ann. Otol. Rhinol. Laryngol. 114 (2) (2005) 111–114. [18] G.T. Richter, M.J. Rutter, A. deAlarcon, L.J. Orvidas, D.M. Thompson, Late-onset laryngomalacia: a variant of disease, Arch. Otolaryngol. Head Neck Surg. 134 (1) (2008) 75–80. [19] R.J. Kuczmarski, C.L. Ogden, L.M. Grummer-Strawn, et al., CDC growth charts: United States, Adv. Data 314 (2000) 1–27. [20] C. Iber, S. Ancoli-Israel, A. Chesson, S.F. Quan for the American Academy of Sleep Medicine, The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Termnology and Technical Specifications, 1st ed., American Academy of Sleep Medicine, Westchester, IL, 2007. [21] M.E. Zafereo, J. Taylor, K.D. Pereira, Supraglottoplasty for laryngomalacia with obstructive sleep apnea, Laryngoscope 118 (10) (2008) 1873–1877. [22] T.E. O’Connor, P. Bumbak, S. Vijayasekaran, Objective assessment of supraglottoplasty outcomes using polysomnography, Int. J. Pediatr. Otolaryngol. 73 (9) (2009) 1211–1216.
[23] T.F. Hoban, R.D. Chervin, Sleep-related breathing disorders of childhood: description and clinical picture, diagnosis, and treatment approaches, Sleep Med. Clin. 2 (2007) 445–462. [24] T.F. Hoban, Sleep disorders in children, Ann. N.Y. Acad. Sci. 1184 (2010) 1–14. [25] C.L. Ogden, K.M. Flegal, M.D. Carroll, C.L. Johnson, Prevalence and trends in overweight among US children and adolescents, 1999–2000, JAMA 288 (14) (2002) 1728–1732. [26] E. Dayyat, L. Kheirandish-Gozal, O. Sans Capdevila, M.M. Maarafeya, D. Gozal, Obstructive sleep apnea in children: relative contributions of body mass index and adenotonsillar hypertrophy, Chest 136 (1) (2009) 137–144. [27] B.L. Matthews, J.P. Little, W.F. Mcguirt Jr., J.A. Koufman, Reflux in infants with laryngomalacia: results of 24-hour double pH probe monitoring, Otolaryngol. Head Neck Surg. 120 (6) (1999) 860–864. [28] F.A.W. Rabelo, A. Braga, D.S. Kuppper, et al., Propofol-induced sleep: polysomnographic evaluation of patients with obstructive sleep apnea and controls, Otolaryngol. Head Neck Surg. 142 (2) (2010) 218–224.