Childhood Sleep Apnea Workup

Compared with the adult literature, the available normative data for sleep and cardiorespiratory parameters are rather sparse in the pediatric literature, such that most pediatric sleep laboratories use individually established reference ranges rather than referring to an authoritative text. Nevertheless, the general consensus criteria for a normal finding on sleep study are presented below and have been derived from the published literature on this subject and the authors' experience.

Reference-range parameters for sleep gas exchange and gas exchange in children are as follows (see also the image below):

• Sleep latency - More than 10 minutes

• Total sleep time (TST) - More than 5.5 hours

• Percentage of rapid eye movement (REM) sleep - More than 15% of TST

• Percentage of stage 3-4 non-REM sleep - More than 25% of TST

• Respiratory arousal index (number per hour of TST) - Less than 5

• Periodic leg movements (number per hour of TST) - Less than 1

• Apnea index (number per hour of TST) - Less than 1

• Hypopnea index (nasal/esophageal pressure catheter; number per hour of TST) - Less than 3

• Nadir oxygen saturation - More than 92%

• Mean oxygen saturation - More than 95%

• Desaturation index (>4% for 5 s; number per hour of TST) - Less than 5

• Highest carbon dioxide level - 52 mm Hg

• Carbon dioxide level of more than 45 mm Hg - Less than 20% of TST

  Normal parameters for sleep gas exchange and gas exchange in children.

  View Media Gallery

The adult criteria usually used around the world for the diagnosis of obstructive sleep apnea do not apply to children. In fact, the finding of 10-15 obstructive apneic events per hour of sleep, which represents mild obstructive sleep apnea in an adult patient in whom treatment may not even be contemplated, represents a sleep-related respiratory disturbance corresponding to a severely affected child definitely in need of therapeutic intervention. Thus, an apnea hypopnea index (AHI) of more than 5 events per hour clearly represents an indication for treatment in children. An AHI of fewer than 3 events per hour does not require any intervention, and, in children with an AHI of more than 3 but fewer than 5 events per hour, the benefit of treatment remains to be determined.

Other diagnostic studies may be warranted to evaluate for complications of obstructive sleep apnea or to better assess the contribution of an underlying condition. In patients with severe obstructive sleep apnea, electrocardiography and echocardiography are particularly important to assess for pulmonary hypertension and cor pulmonale.

Currently, the only available tool for definitive diagnosis of obstructive sleep apnea is an overnight polysomnographic evaluation in the sleep laboratory. An overnight polysomnographic study usually includes multiple channels that aim to monitor sleep state, as well as cardiac and respiratory parameters (see the images below).

Nasopharyngoscopy or direct laryngoscopy and bronchoscopy may be required to determine anatomy prior to contemplated otolaryngologic surgery.

Compressed overnight polysomnography tracing of a 6-year-old boy who snores, showing multiple events of obstructive apnea (green-shaded areas) associated with oxyhemoglobin desaturation (yellow-shaded areas) and EEG arousals (red-shaded areas).

View Media Gallery

Parameters monitored during an overnight pediatric sleep study.

View Media Gallery

Polysomnography remains the criterion standard for establishing the diagnosis of obstructive sleep apnea (OSA) in infants, children, and adults. Ideally, polysomnography should be performed overnight and during the patient's usual bedtime.

Polysomnography provides the following measures:

• Sleep state (≥2 EEG leads)

• Electrooculogram (right and left)

• Submental electromyelogram (EMG)

• Airflow at nose and mouth (thermistor, capnography, or mask and pneumotachygraph)

• Chest and abdominal wall motion (impedance or inductance plethysmography)

• Electrocardiogram (preferably with R-R interval derivation technology)

• Pulse oximetry (including a pulse waveform channel)

• End-tidal carbon dioxide (sidestream or mainstream infrared sensor)

• Video camera monitor with sound montage (analog or digital)

• Transcutaneous oxygen and carbon dioxide tensions (in infants and children < 8 y)

Multiple physiologic parameters are monitored during polysomnography, although the specific montage may vary slightly between sleep laboratories. Generally, electrooculography, chin and leg surface electromyography (EMG), and at least 2 EEG channels are included to confirm sleep and assess sleep architecture. Breathing is assessed using nasal/oral airflow sensors, pulse oximetry, and end-tidal (ET) CO₂ measurements monitoring and by placing piezo crystal belts across the chest and abdomen to detect respiratory efforts. At least one ECG channel is necessary to determine heart rate and rhythm. Occasionally, other channels are incorporated into the study as needed. These might include additional EEG leads to better detect seizure activity, esophageal pH measurements, or transcutaneous carbon dioxide monitoring.

Polysomnographic normal standards differ between children and adults. In the pediatric age range, abnormalities include oxygen desaturation under 92%, more than one obstructive apnea per hour, and elevations of ET CO₂ measurements of more than 50 mm Hg for more than 9% of sleep time or a peak level of greater than 53 mm Hg.

See the related polysomnographic image below.

Example of an obstructive apnea and an obstructive hypopnea recorded during polysomnography.

View Media Gallery

Polysomnography is necessary to document obstructive sleep apnea and gauge its severity. A history of snoring alone is not adequate for making a diagnosis of obstructive sleep apnea or for determining its seriousness.

Some children with obstructive sleep apnea have primarily obstructive hypoventilation in which repetitive partial obstructions occur with some degree of relative oxygen desaturation and hypercapnia. Because of this, pediatric polysomnographic testing should include some means of determining CO₂ levels, such as end-tidal (ET) CO₂ monitoring or transcutaneous CO₂ monitoring.

PSG, continuously monitored by appropriately trained technical personnel, may be difficult to arrange due to relative unavailability, with long waiting periods between referral and testing times. For these reasons, attempts have recently been made to evaluate the role of outpatient overnight studies to provide more accessible and practical approaches to the diagnosis of pediatric obstructive sleep apnea. However, these outpatient studies are not well validated yet or covered by third party payers and, thus, remain largely available only as research tools.

Individuals with obstructive sleep apnea syndrome have pathologic degrees of obstructive apnea, obstructive hypopnea, or both. Severity is quantified using a polysomnographic-derived index known as the apnea hypopnea index (AHI). The AHI is the total number of apneas and hypopneas that occur divided by the total duration of sleep in hours. An AHI of 1 or less is considered to be normal by pediatric standards. An AHI of 1-5 is very mildly increased, 5-10 is mildly increased, 10-20 is moderately increased, and greater than 20 is severely abnormal.

Obstructive hypopnea (OH) in children is a sleep-related breathing disorder that is considered a variation of obstructive sleep apnea. Children with OH may have an AHI in the normal range, but they have episodic periods of hypercapnia, as identified on the basis of end-tidal (ET) CO₂ monitors. Peak ET CO₂ measurements of greater than 53 mm Hg are considered abnormal. The percentage of sleep time spent with ET CO₂ measurements greater than 50 mm Hg should not be more than 9%.

Most physicians who treat children with sleep apnea generally recommend specific interventions when the AHI is greater than 5 or respiratory events are associated with oxygen desaturations of less than 85%. When the AHI falls to between 1 and 5, other clinical factors must be taken into account to determine whether to pursue adenotonsillectomy or other therapy.

Daytime nap studies are specific, but not sensitive, in detecting sleep apnea. This is because obstructive events are more likely to occur during rapid eye movement (REM) sleep than during other sleep stages, and very little (if any) REM sleep occurs during daytime naps in noninfants. Therefore, children with symptoms of obstructive sleep apnea who have normal nap study findings must undergo nocturnal polysomnography to exclude the diagnosis. Sleep studies should be performed without sedation.

Unattended home overnight oximetry has been proposed as a screening study. However, it may miss the child with significant obstructive sleep apnea who does not have marked episodes of oxygen desaturation.

Overnight pulse oximetry by itself is not adequate for establishing the diagnosis or excluding obstructive sleep apnea in children because it provides no information concerning sleep staging/sleep fragmentation or carbon dioxide.

The results of initial studies indicate that, although home audio tape recordings appear relatively insensitive, oximetry trend analysis with or without additional measures may provide a useful alternative in establishing the definitive cases that require intervention. However, despite high specificity, home oximetry has low sensitivity, and children with negative findings on studies still require complete nocturnal polysomnography.

Neck radiography for soft tissue detail help define upper airway anatomy and adenoid size and exclude the possibility of rare nasal pharyngeal neoplasms.

Assessment of tonsillar size usually does not require any type of imaging; however, lateral neck radiographs can be used to determine adenoid size. Although MRI can provide very detailed images of soft tissues and bony structures underlying the nasopharynx, such images are not usually required, except in cases of suspected aberrant anatomy.

Cine MRI during sleep may be helpful in identifying specific sites of airway obstruction in the complicated patient being evaluated for surgical interventions. This technique is currently only available at a handful of specialized tertiary care facilities.

Thyroid function studies are useful to exclude hypothyroidism, which is associated with tongue enlargement, weight gain, and obstructive sleep apnea.

These studies are not necessary in all children with suspected sleep apnea. However if very severe long-standing obstruction is suspected, an ECG and echocardiography are helpful in assessing ventricular thickness and function and to check for evidence of pulmonary hypertension.

If the clinical history suggests the possibility of narcolepsy, the MSLT should be ordered in conjunction with overnight polysomnography.

A history of severe snoring, headaches, neck pain, urinary frequency, or swallowing problems raises the suspicion of Chiari malformation. Chiari malformations may occur in otherwise normal children and in association with congenital myelomeningocele. If brainstem dysfunction is suspected, MRI is necessary. Cranial CT imaging is not adequate to assess for brainstem and upper cervical cord lesions.

Additional studies include the following:

• CBC count may be useful because chronic hypoxia related to recurrent airway obstruction may lead to polycythemia

• Cephalometric radiography and 3-dimensional CT reconstruction imaging are rarely, if ever, necessary in the pediatric age group

• Multichannel studies lack reliable assessment of sleep disruption

The following sleep study parameters are under investigation:

• Nasal pressure–flow measurements

• Esophageal manometry

• Continuous noninvasive blood pressure monitoring

• Autonomic nervous system tone using finger tonometry