The American Academy of Pediatrics (AAP) first recommended universal screening for snoring in 2002 (28). Although a PSG remains the most commonly recommended method for diagnosing SDB, a comprehensive “sleep history” is a vital clinical tool for evaluating the presence of SDB in children. This history, more often provided by caregivers, should include information about nightly hours of sleep, sleep hygiene, sleep latency, abnormal sleep behaviors, the presence and character of snoring, daytime behavior (including hypersomnia), behavioral disorders, and school performance. It is also important to understand napping behavior of children with suspected SDB.

Parents and caregivers routinely report witnessing nightly snoring, apneas, choking or gasping, increased work of breathing, hyperextension of the neck, restless sleep and night sweats, nocturnal enuresis, and parasomnias (Table 139.2). Commonly witnessed parasomnias include confusional arousals, sleep walking, and night/sleep terrors; however, the presence of these symptoms does not necessarily correlate with the presence of SDB in children. Caregivers may also report daytime symptoms including frequent mouth-breathing, chronic nasal obstruction, poor school performance, hyperactive behavior or attention deficit disorder, aggression, and less frequently, daytime somnolence (Table 139.2). Although hypersomnolence is present in some children with OSA, it is more commonly seen in adolescent and adult patients. For children with hypertrophic tonsils, caregivers may also report eating habits including avoidance of bulky foods such as meat.

While sleep questionnaires can be used in the office as a screening tool to identify children most at risk of having OSA and to acquire information about common symptoms associated with SDB and OSA (29, 30, 31), their use as a single tool to diagnose OSA has not been supported, thus they should only be used as an aid when pursuing a thorough sleep history. There are limitations to the accuracy of these data in differentiating SDB from OSA. In fact, parental reporting has been shown to be an inaccurate method of diagnosing OSA in children (30), and a recent metaanalysis of 12 articles that examined the accuracy of clinical assessment found that discrimination between primary snoring and OSA could not be precisely determined by parental reporting alone (32).

Narrowing of the upper airway is the primary reason that children develop SDB and OSA. This may be the result of multiple factors including anatomic abnormalities, and thus, a complete head and neck exam is recommended. Identification of the anatomic location of the obstruction is necessary to tailor the appropriate treatment, especially in those children with multilevel obstruction.

A systematic approach to the physical examination is necessary so that no relevant findings are missed (Table 139.3). The exam should begin with evaluation of the general appearance of the patient, including height, weight, blood pressure, general craniofacial appearance, voice, and the presence or absence of mouth breathing. Voice abnormalities can be useful in identification of the location of
pathology. For example, a hyponasal voice can suggest enlarged adenoids, while a muffled voice can result from tonsillar hypertrophy. Care should be taken to perform a thorough cardiovascular exam including auscultation for any murmurs. Systemic findings including failure to thrive, cor pulmonale, and pectus excavatum should be considered (33). Special consideration is given to obesity and body habitus, especially in the adolescent patient. Finally, genetic and syndromal dysmorphisms should also be noted.

At the start of the exam, an overall assessment of head and neck shape and proportions should be performed. For example, midface hypoplasia is seen in Down syndrome and Crouzon syndrome, whereas mandibular hypoplasia is often seen in children with Treacher Collins or Pierre Robin sequence. Extensive cervical adipose tissue and increased neck circumference may be seen in children with obesity. Examination of the nasal cavity should include details regarding the mucosa, septum, inferior turbinates, nasal valve, and presence of polyps or masses. The nasal mucosa may be erythematous, suggesting chronic inflammation, or dusky, as is seen with allergic rhinitis. Anterior rhinoscopy can assist in identification of septal deviation and abnormalities of the inferior turbinates such as turbinate hypertrophy, contributing to nasal obstruction. Often, chronic rhinorrhea is seen in children with chronic sinusitis or adenoid hypertrophy. Functional nasal valve collapse can be seen in children during deep inspiration through the nose and can contribute to obstructed flow through the nasal passages. The presence of nasal polyps is rare except in children with cystic fibrosis. Patency of the nasal cavities, information often useful in the newborn, can be determined by passing an 8 French catheter through the nose and into the oropharynx. Finally, nasal endoscopy can be used in children to evaluate multiple anomalies of the nasal cavity, such as choanal atresia or stenosis, pyriform aperture stenosis, posteriorly positioned polyps or masses, or adenoid hypertrophy.

Examination of the oral cavity and oropharynx is one of the most useful portions of the head and neck exam in a child with suspected SDB and OSA. The oropharynx, and specifically the retropalatal region, is one of the most narrow portions of the airway in normal children (34). Initial exam should assess for size and position of the mandible and dentition, including malocclusion. Tongue size and position, especially macroglossia and glossoptosis, may have a significant contribution to SDB if present. Evaluation of the palate should include inspection for overt or submucosal clefting of the hard and soft palate, a high arched or narrow palate, and palatal masses, as well as presence of a bifid uvula. The tonsils are graded based on a four-point scale as described by Brodsky et al. (35) (Fig. 139.1): 0 for surgically absent tonsils, 1 for tonsils that lie within the anterior and posterior pillars, 2 for tonsils seen just beyond the pillars, 3 for tonsils that reach beyond 50% towards the midline,
and 4 for tonsils that “kiss,” or approximate, in the midline. An assessment using the modified Mallampati score (36, 37, 38) can also be performed by evaluating the view of the uvula and posterior pharyngeal wall while asking the child to open his/her mouth completely while the tongue is in a resting position. Scoring is grade I when the entire uvula is visible, grade II for a partial view of the uvula, grade III when none of the uvula is visible but some of the soft palate is visible, and grade IV for a view of the hard palate only. Flexible laryngoscopy is recommended in all children under 1 year of age with OSA and when further evaluation is needed in older children. Flexible laryngoscopy allows direct visualization of the base of tongue, pharyngeal walls, vallecula, epiglottis, hypopharynx, vocal folds, and potentially, a portion of the subglottis. This procedure is useful in the evaluation of possible laryngomalacia, lingual tonsil hypertrophy, and persistent OSA despite previous tonsils and adenoids (T&A), and is also useful to document vocal cord mobility and to rule out pharyngeal/hypopharyngeal masses.

Lateral neck films can be used in children that do not tolerate nasopharyngoscopy or flexible laryngoscopy. This film is most commonly used to identify adenoidal size but can also identify structural abnormalities within the nasal cavity, as well as lingual tonsil hypertrophy (39). Cine magnetic resonance imaging (MRI) has been described as a tool for assessment of the child with persistent sleep apnea despite surgical treatment with an adenotonsillectomy (40). In recent years, both pediatricians and pediatric otolaryngologists have deferred the use of airway fluoroscopy and computed tomography (CT) unless absolutely necessary, given the long-term risks of early radiation exposure. In those patients with a higher likelihood of associated cardiovascular complications from OSA, such as children with underlying congenital cardiac anomalies, electrocardiograms and echocardiograms should be considered in the workup of OSA.

Polysomnography (PSG) is currently considered the traditional, and preferred, method in the evaluation and diagnosis of OSA in children (28). PSG includes the recovering of a number of different simultaneous recordings during sleep as noted in Table 139.4 (41). The results of this test provide objective measurements regarding the severity of airway obstruction in children during sleep. The apnea-hypopnea index (AHI) is obtained by averaging the number of apneas and hypopneas per hour of sleep, and this number is used to categorize the severity of OSA. Obstructive events/apneas in children are defined by continued respiratory effort with an absolute termination of airflow through the nose and mouth. A reduction in airflow of ≥30% to 50% along with either an arousal or an oxygen desaturation ≥3% from the baseline is considered a hypopnea (42). In addition, the definition of flow limitation of apnea in children only requires an event to last for two or more consecutive breathing cycles, versus the 10-second criteria necessary for adults (42). Similar to adult evaluation, multiple series have demonstrated that the test-retest reliability, comparing multiple nights of PSG in the same child, is consistent and that PSG serves as a valid test for OSA in children (43, 44, 45, 46).

The categories for OSA in adults are well-established; however, a standard definition of SDB disease severity does not exist for pediatric OSA, thus study interpretation may differ in their criteria used for mild, moderate, or severe OSA. One series that evaluated nonsnoring children defined a normal PSG as ≤1 obstructive apnea per hour, ≤0.9 central apneas per hour, an oxygen saturation nadir of 89% or higher, and an end tidal CO₂ greater than 45 mm Hg for 10% of the total sleep time (47). Current research definitions for OSA commonly define mild OSA when obstructive AHI is between 1 and less than 5 events per hour, moderate OSA if the AHI is 5 to less than 10, and severe OSA when the AHI is greater than 10.

While the use of PSG to diagnose OSA is widely accepted for adults, some controversy exists regarding the use of PSG in the diagnosis of pediatric OSA. It is not universally accepted that all otherwise healthy children with suspicion of OSA should undergo PSG. Issues include the fact that there are not enough sleep centers to test children in a reasonable time frame and that many sleep centers are not designed to accommodate children (i.e., with extra beds for caregivers, CO₂ monitoring, and staff specifically trained to work with children). If the PSG for a child is done in an “adult” sleep center, it is important to confirm that pediatric definitions are used in the scoring of the study. The extra cost and time to obtain a sleep study is questioned if adenotonsillectomy is planned, regardless of the PSG result.

Two recent guidelines were published in 2011 to address some of these issues. The first, by the American Academy of
Sleep Medicine (AASM), recommends that PSG is indicated when OSA is suggested by clinical assessment, and all children being considered for adenotonsillectomy for OSA should undergo PSG (48). These guidelines also recommend that PSG be used prior to decannulation in children that required tracheostomy placement, be used postoperatively in children with symptoms of persistent OSA after adenotonsillectomy as well as those with moderate to severe preoperative OSA, and those with high risk for persistent disease such as those with obesity or craniofacial abnormalities (48).

Alternatively, practice guidelines released by the American Academy of Otolaryngology—Head and Neck Surgery (AAO-HNS) recommend that all children with obesity, Down syndrome, craniofacial abnormalities, neuromuscular disorders, sickle cell disease, or mucopolysaccharidoses be referred for PSG, since these children are at the highest risk of having OSA (49). In addition, PSGs are recommended in order to determine which otherwise healthy children should be admitted after adenotonsillectomy, specifically those with severe OSA and thus high risk for respiratory distress after surgery (50). They also recommend PSGs when the need for surgery is uncertain or if the size of the tonsils found on physical exam does not correlate with the reported symptoms of SDB (49, 50).