Pediatric Obstructive Sleep Apnea

Etiology

Sleep apnea is divided into separate categories based on two causes; central or obstructive.[2] Central sleep apnea is due to central nervous system pathology. It is a neurologically mediated respiratory drive impairment with no associated respiratory effort.[3] 

Obstructive sleep apnea, 95% of diagnosed sleep apnea, is due to complete collapse of the upper airway or partial collapse, resulting in arousal from sleep or 3% or more oxygen desaturation. Anything that can decrease airway diameter or integrity can contribute to OSA, including anatomic, genetic, or neuromuscular issues.[2] The upper airway can have an increased risk of abnormal collapse due to both intrinsic and extrinsic factors. The intrinsic factors are based on the critical pressure in the airway that is needed to maintain patency. The extrinsic factors are fat deposits, hypertrophy of tissues, and craniofacial features that stray from normal anatomy that contribute to increased incidence of collapse.[1]

Epidemiology

The incidence of pediatric OSA peaks between 2 to 8 years of age due to the increased growth of tonsils and adenoids relative to the size of the upper airway in this age group. Risk factors for early-onset OSA include prematurity, Down syndrome, African American race, and daycare attendance. The severity can be increased in those with obesity, tobacco exposure, and reduced family income. Boys are at an increased risk after puberty, but the prepubertal risk is equal among boys and girls.[3]

Pathophysiology

Diagnosis of sleep apnea is made by measuring the apneic events during sleep. The diagnostic criteria are detailed in the evaluation section. The four main features that contribute to OSA are obesity, lymphoid hyperplasia, craniofacial abnormalities, and neuromuscular dysfunction. All of these abnormalities provide increased extrinsic factors that can lead to upper airway compression or collapsibility. Obesity causes fat deposits to surround the upper airway and increase extrinsic pressure that causes collapse. The presence of lymphoid hyperplasia causes tonsillar and adenoid obstruction of the airway, and the increased relaxation during sleep allows these to become significantly more problematic. Neuromuscular dysfunction can be seen in both central and obstructive causes of sleep apnea. This is most often seen in cases like Down syndrome, where there is hypotonia, which contributes to increased susceptibility of the airway to collapse. Additionally, patients with craniofacial abnormalities, including those with Crouzon, Pierre-Robin, or Apert syndromes, as well as those with cleft lip or palate, are at an increased risk. The presence of alteration of the normal airway anatomy as well as features such as micrognathia, micro or macroglossia, and midface hypoplasia all contribute to decreased posterior oropharynx space and the increased incidence of pediatric OSA.[3]

History and Physical

It is important to evaluate sleep quantity and quality during the evaluation of the pediatric patient. Studies have shown that sleep screening differs across practices. Although there is no standard screening tool that is widely used specifically for pediatric OSA, guidelines suggest that practitioners should be asking about sleep quantity and quality as well as screening for snoring at every well-child visit. Additionally, if there is concern about sleep quality, questions should also be posed to inquire about frequent night awakenings, unusual sleep positioning, and significant disruption of bed coverings as signs of increased nighttime movements.[3] This is especially important because parents may not always discuss concerns about their child’s sleep without first being prompted or explicitly asked about these issues.[4]

Children can present in ways that often differ from the classic adult OSA presentation, most notably resulting in behavioral changes. Parents often endorse a history of snoring, mouth breathing, witnessed apneas, frequent nighttime awakenings, and secondary nocturnal enuresis. Children with OSA have disrupted sleep, which can lead to behavioral issues, including hyperactivity, irritability, or even aggression. These issues are often what prompts parents to bring children to the doctor for evaluation.[2]

On physical exam, patients may appear tired or fatigued or may exhibit a hyperactive state. An exam may reveal “allergic shiners,” swollen nasal mucosa, micrognathia, macroglossia, high arched palate, adenoidal facies, or hypertrophied tonsils. Additionally, hyponasal speech and nasal congestion can often be appreciated. A growing risk factor for pediatric OSA is childhood obesity. All children should be screened with height, weight, and BMI evaluation at every visit. For every kg/m^2 increase in BMI above the 50th percentile, there is a 12% increase in the risk for OSA.[2]

Craniofacial abnormalities, certain syndromes, and events in infancy can also carry an increased risk of OSA. Those with neuromuscular disorders, Down Syndrome, and craniofacial disorders should be screened repeatedly for OSA. Additionally, children with a history of prematurity are at increased risk for OSA even without the presence of other risk factors.[5] Studies have found that as high as 9.4% of school-aged children who were born prematurely were also eventually diagnosed with OSA. This patient group may not always exhibit the typical signs of pediatric OSA, and there should be a low threshold for sleep study referral.[5]

Evaluation

When the history and physical are concerning for pediatric OSA, the gold standard for diagnosis is nocturnal polysomnography (PSG). However, evaluation with a PSG can be expensive, time-consuming, and resource-limited. Overnight oximetry at home can be used as a tool to provide more information, but it does not replace a PSG in making a diagnosis of OSA.[3] If there is a concern for an underlying cardiopulmonary process, a chest x-ray and EKG are also warranted as additional evaluation. If appropriate, the cardiac workup may also include an echocardiogram, which is especially important prior to surgical management for children with severe OSA. There may also be increased utility in measuring certain inflammatory biomarkers, specifically kallikrein-1, uromodulin, urocortin-3, and orosomucoid-1, which can be elevated in children with OSA. It is also appropriate to obtain a CBC, iron studies, and TSH looking for other primary causes of sleep disruption.[2] Imaging studies may aid in assessing underlying anatomic abnormalities that can contribute to OSA, but are only useful in conjunction with a PSG to make a definitive diagnosis.[6] 

During PSG, certain specific parameters of sleep quality and quantity are measured. This is done by using sensors to measure brain activity, heart rate, nasal and oral air movement, blood oxygen levels, limb movements, eye movements, and the presence of snoring. The data collected allows for calculations to include sleep onset latency, sleep efficiency, and time in each stage of sleep. Important for the diagnosis of OSA, a PSG will measure the apnea/hypopnea index (AHI), which is the average of the apneic and hypopneic episodes per hour of sleep. An AHI score of 1 to 4.9 events/hour is mild OSA, 5 to 9.9 events/hour is moderate, and more than 9 events/hour is severe.[3] An AHI of 1 or greater is abnormal in children up to 13 years old, although there is some controversy over the clinical significance of an AHI 1 to 1.9 events/hour. [1] From 13 to 17 years of age, either pediatric or adult criteria can be used depending on what is most appropriate based on the clinical picture and patient development. Because children have higher airway integrity and less risk of collapse like adults, their PSG is usually dominated by hypopneic events rather than frank apnea. Interestingly, in the diagnosis of pediatric OSA, hypercarbia may also be present in the pediatric population that is absent in the adult OSA population. This is caused by a state of prolonged hypoventilation and is less common in adults.[3]

Treatment / Management

Many different treatment options are depending on the severity of the OSA. These include both surgical and non-surgical interventions. A therapeutic trial of leukotriene inhibitors (montelukast) may be appropriate for pediatric patients who are diagnosed with mild to moderate OSA.[7] Montelukast has been shown to decrease adenotonsillar size significantly after three months of treatment, leading to a decrease in AHI in appropriate patients.[8](A1)

Systemic glucocorticoids are not effective in the treatment of pediatric OSA. However, intranasal steroid treatment for six weeks has been shown to improve the AHI and can also be considered a treatment option for patients with mild to moderate OSA.[2] When intranasal steroids are used in combination with leukotriene inhibitors, the majority of patients will have a clinically significant decrease in AHI.[9] 

If there is adenotonsillar hypertrophy, the most effective treatment is adenotonsillectomy (A&T). This is recommended for most patients with an AHI more than 9 events/hour and those with mild or moderate disease with significant symptoms. A partial tonsillectomy is also an option that decreases both postoperative complications and recovery time, but it has been shown that tonsillar regrowth rates are between 7.2% to 16.6%.[9] Other surgical methods, including lingual tonsillectomy and uvulopalatopharyngoplasty, have limited data and have been shown to be non-superior.[9] One of the more conservative treatments, adenotonsillotomy, has not been shown to be superior to A&T in the treatment of pediatric OSA. There is an increased risk of recurrence of OSA, and the possible need for repeat surgery with adenotonsillectomy should be taken into consideration.[10] 

For patients with mild to moderate OSA, "watchful waiting" for up to six months can be appropriate. This is usually done to attempt to correct underlying problems, like obesity or allergic rhinitis, that could improve OSA and AHI without surgical intervention. However, in pediatric patients who are non-obese and non-syndromic, A&T is superior in the improvement of the AHI.[11]

Continuous positive airway pressure (CPAP) is another potential treatment option. While this is the first-line treatment in adults, there are limitations to the use of PAP in children. PAP should be considered during the perioperative phase before A&T for severe OSA if the child is not a good surgical candidate or has persistent moderate to severe OSA despite surgery. [2] Compliance can be difficult for children, and the use of the same mask over time has the potential to change the facial structure.

Although oral appliances (OAs) have been shown to decrease AHI in adults, there have not been enough studies on pediatric populations to know their efficacies or which patients might benefit from these.[9] Changing OAs to accommodate a child's growth would also require frequent refittings. Rapid maxillary expansion, however, is a potential option for prepubescent patients and may be especially useful for those with high arched palates and non-obese patients with residual OSA after A&T.[12](B3)

Myofunctional therapy is a new area of study in the treatment of both pediatric and adult OSA. This consists of retraining the muscles of the oral cavity and oropharyngeal structures, as well as proper tongue positioning. There is limited data about efficacy in children, with only small studies on pediatric patients with OSA. However, this could be an area of future research and could serve as an adjunct in treating OSA.[13](A1)

Any patient with a reversible risk factor for OSA, especially obesity, should be counseled on reversing the issue. Weight loss can improve OSA and can be considered adjunctive therapy in older children.[9] Parents should also be advised to limit children's exposure to second-hand smoke and tobacco use in the house.

Differential Diagnosis

• Allergic rhinitis - Not all snoring is OSA, which should be discussed with the parents at the time of the initial evaluation, and nasal congestion should be treated.
• Attention deficit hyperactivity disorder (ADHD) - Pediatric OSA can be mistaken for behavioral problems because it frequently manifests as hyperactivity and difficulty with focus and attention in children.
• Developmental delay - OSA can lead to learning difficulties due to compromised focus and attention and sometimes be mistaken for a developmental delay.
• Gastroesophageal reflux - When reflux occurs at night, it can cause a brief pause in breathing that may be mistaken for OSA. The two are not, however, mutually exclusive and reflux can increase the risk of OSA by worsening adenotonsillar hypertrophy.
• Nocturnal enuresis - Primary nocturnal enuresis is rarely associated with OSA, but secondary nocturnal enuresis should prompt screening for other symptoms of OSA. 
• Morning headaches - These can be a consequence of carbon dioxide retention overnight, but the patient should also be screened for more concerning qualities of the headaches (waking them from sleep at night, worse when lying supine, etc.) that may require imaging or further evaluation. A typical headache from OSA is dull and generally goes away on its own shortly after getting out of bed without using medication or caffeine. 
• Parasomnias - OSA can increase the occurrence of sleepwalking, night terrors, and confusional arousals but may also be unrelated. 
• Narcolepsy - OSA usually does not present as excessive daytime sleepiness in younger children. Patients frequently falling asleep during the day after five years of age should prompt screening questions for narcolepsy.
• The only definitive way to distinguish between OSA and any of these is polysomnography.

Prognosis

Patient populations that are at a higher risk of having OSA have an increased risk of neurocognitive disability later in life compared to their peers if OSA is left untreated.[5] If identified and managed promptly, patients will not suffer long-term consequences or complications of pediatric OSA.[2] 

After A&T, it is appropriate to wait up to 6 months for the inflammation to resolve and for normal sleep patterns to return. Patients should be evaluated at that time for improvements in sleep function and quality; however, it is not usually necessary to do an additional PSG after surgery. The success of the intervention can be based on the evaluation of improvement in initial behavior issues or cognitive function as well as reduced concern for ongoing observed sleep disturbances. If a PSG was performed after surgery, this would provide concrete data on the reduction in AHI. In a study that compared the success of A&T with watchful waiting, patients who had undergone A&T were found to have a significant improvement in OSA symptoms, AHI, and behavior. However, there was not a difference between the two groups in regard to executive function or attention.[14]

The majority of children experience improvement after surgery. There has been data to support that A&T results in better outcomes as compared to watchful waiting and that PSG findings were normalized in 79% of children who underwent A&T as compared to 46% with watchful waiting.[14] In children with mild OSA (AHI >1), symptoms persist 19% to 73% of the time, but with severe OSA (AHI >5), symptoms including snoring and disrupted sleep persist only in 13-29% of patients.[2] Most children will not require long-term monitoring unless there is an ongoing concern for persisting symptoms. Unfortunately, A&T is generally less effective in patients with Down Syndrome, craniofacial abnormalities, or the combination of the two, and these children will require post-operative monitoring and at least yearly repeat PSG due to the high rate of persistent or recurrent OSA.[6]

Complications

If left untreated, pediatric OSA can have serious morbidities and long-term complications. Sustained hypoxia can increase pulmonary vasoconstriction and lead to pulmonary hypertension and right heart failure at an early age. Cognitive dysfunction, impaired learning, and poor school performance are associated with undiagnosed and untreated pediatric OSA. Additionally, increased work of breathing can be associated with failure to thrive seen in younger populations.[2]

It should be recognized, though, that surgical intervention is not benign and should be carefully considered like any surgical intervention. If a patient is treated with A&T, there are certain surgical risks and complications that are associated with the procedure. Unfortunately, tonsillectomy can have adverse events, although these are rare in children. The most common are localized pain, decreased oral intake, and dehydration in the acute post-surgical timeframe. More concerning complications of A&T are bleeding, secondary infection, respiratory compromise, velopharyngeal insufficiency, and subglottic stenosis. As with any surgery that requires airway management, there is always a risk of bronchospasm or laryngospasm. The increased degree of severity of OSA is predictive of a higher rate of post-surgical complications, and those with more severe OSA may require longer monitoring in the PACU or inpatient ward after surgery.[2]