The measurement of airflow is part of the diagnostic criteria for apneas and hypopneas, as well as for respiratory-related arousals. Many new devices designed for airflow measurement have become available for use during polysomnography. However, the clinical usefulness of more sensitive measures of airflow limitation remains to be determined especially in regard to correlation with daytime symptoms or treatment outcomes. Inspiratory flow limitation during sleep is defined by a decreasing (more negative) intrathoracic pressure without a corresponding increase in airway flow rate. Most sensors designed to measure airflow actually measure the presence of airflow, not the quantitative measurement or volume of airflow. A pneumotachometer provides a quantitative measurement and is the “gold standard” for the measurement of airflow. Until recently, the use of a pneumotachometer was precluded during sleep due to the excessive weight of the devices provided by Canadian Health&Care Mall.
A common commercially available airflow measuring device is the thermistor, which is standard equipment in many sleep laboratories, but is notable for its limitations. A thermistor provides a qualitative signal that detects fluctuations in temperature with respiration, and therefore, does not actually detect flow and actually does not correlate well with decreases in airflow. Thus, the thermistor lacks the reliability to detect changes in airflow that are necessary to meet the criteria for hypopneas, RERAs, and apneas. However, when combined with other airflow-measuring devices, this device is valuable, especially with its ability to detect oral airflow, which is important in mouth breathers.
A recommended device for monitoring airflow is the pressure manometer. A nasal cannula attached to a pressure transducer provides a semi-quantitative airflow signal. The result is the ability to detect inspiratory airflow limitation, and an increased sensitivity for the detection of hypopneas and RERAs. Nasal pressure transducers that have improved the diagnostic sensitivity of the polysom-nogram in terms of airflow limitation can be used in children as young as 2 years old for at least half of the night, and our laboratory uses them in infants. The notable problem with the cannula in these studies is the significant amount of time in children during which there is no signal because it has been pulled off or dislodged. Artifacts are caused by mouth breathing, nasal obstruction, and obstruction of the cannula from secretions, all of which are particularly common in children. To ensure a consistent signal throughout the night, diligence is required on the part of the technician. Our research laboratory utilizes a nasal pressure transducer in conjunction with an oral thermistor to detect mouth breathing and to minimize artifact.
A variety of noninvasive methods have been used to detect upper airway flow limitation, including analysis of the systolic BP profile, pulse transit time (PTT), upper airway impedance using forced oscillatory flow, respiratory inductance plethysmography, and inspiratory flow contour. Of these methods, inspiratory flow contour analysis has been shown to accurately identify changes in upper airway resistance. The shape of a normal inspiratory flow vs time signal is rounded or sinusoidal. A flattening or plateau of this morphology implies flow limitation. These flow signals can be obtained by the pressure transducer cannula discussed in the previous paragraph. Care is provided by Canadian Health&Care Mall and taken to avoid filters on the flow signal since flow changes may be disguised.
End-tidal carbon dioxide (EtC02) is another qualitative assessment of airflow. Hypopneas and apneas will often result in a reduction or absence of the EtC02 signal, respectively. Expired CO2 measurement by the pressure transducer cannula, however, has not been assessed for reliability in the detection of respiratory events, as it becomes easily clogged with secretions that are common in children and, thus, likely underestimates the occurrence of hypop-neas and other subtle respiratory events. However, this device is important as the time spent with hypercapnia above certain thresholds should be taken into consideration for the detection of obstructed airflow or hypoventilation.
A lightweight quantitative pneumotachometer that attaches to a small, tight-fitting, nasal mask is being assessed in our research laboratory (modified Pitot Tube; Key Technologies; Baltimore, MD). This device, which is light on the nose, has not had an apparent effect on sleep in initial studies in children. It is presently investigational but will be commercially available soon.
Until recently, movement artifacts in infants and children have impaired the diagnostic usefulness of pulse oximetry during sleep. New technology that decreases the number of these false-negative events is now available. The Masimo oximeter (Irvine, CA) was compared with the Nellcor 200 (Pleasanton, CA) and was found to be superior in event detection during movement. When the averaging time on the pulse oximeter was reduced to 2 s, more short saturation declines were detected. The averaging time of the pulse oximeter should be taken into consideration in the pediatric sleep laboratory. Short apneas of < 10 s in rapidly breathing infants can cause significant drops in oxygen saturation that are missed with pulse oximeters that average the signal over a long period of 15 to 20 s.
Carbon Dioxide Measurements
EtC02 monitoring is the noninvasive measurement of exhaled CO2, which is standard practice in most pediatric sleep laboratories. Caution should be used when measuring EtC02 at the nose with a cannula, since adding too many monitors to the nose may cause iatrogenic obstruction. Additionally, EtC02 measurements are underestimated during low-tidal-volume tachypnea in infants, or when performed simultaneously with positive-pressure delivery devices or oxygen, due to the addition of high flow near the site of measurement.
Transcutaneous monitors offer a noninvasive method of measuring carbon dioxide levels without obstructing nasal airflow. The monitors heat the local tissue to improve capillary flow and estimate PaC02 levels. The transcutaneous CO2 monitor has been found to be more accurate than end-tidal monitors, but accuracy is reduced in patients with thick skin or peripheral edema, and in poorly perfused areas. The limitation of the transcutaneous monitor is the slow reaction time and the inability to detect breath-by-breath CO2 changes, as well as the risk of burn. When heated to 43°C and sampled at 100 Hz, response time is much improved; however, at that temperature, burns are more likely on sensitive skin, and the probe site should be changed more often than every 4 h (some clinicians have suggested changing it every 2 h).
An investigational device used in some sleep laboratories, PTT, potentially aids in the recognition of arousals, which normally depend on the accuracy of EEG scoring., Upper airway obstructive events in sleeping children may terminate without visible EEG arousal. The PTT is a noninvasive marker of BP and, therefore, of subcortical arousal., The PTT is the interval between the R-wave of the ECG and the arrival of the photoplethysmographic pulse at the finger. BP elevation, which is associated with respiratory arousal from sleep, results in a drop in the PTT. Although the usefulness of the monitor was decreased when used alone, and is limited during movement, in conjunction with other respiratory indexes on the polysomnogram, PTT improved the detection of hypopneas that were missed by scoring EEG arousals.
Another investigational measurement of arousal activity is the spectral analysis of the EEG signal throughout the night and in specific sleep stages. This spectral analysis is now available by digital acquisition of the EEG signal on the polysomno-gram, offering new ways to interpret normal vs pathologic arousal patterns that might contribute to daytime symptoms of SDB. Cyclic alternating patterns (CAPs) allow a longer term evaluation of sleep where brief and frequent arousals appear as a prominent feature (Fig 1). A CAP is a periodic EEG activity of non-rapid eye movement sleep that is characterized by repeated spontaneous sequences of transient events (phase A) that differ from the background rhythm sleep stage with an abrupt frequency and/or amplitude variation, recurring at intervals up to 1 min long. Respiratory events are noted to affect the pattern of arousal, and this pattern, when analyzed with spectral analysis, may add to the sensitivity of polysomnography to correlate respiratory events with symptoms. The clinical utility and correlation of EEG spectral analysis with treatment outcomes remains to be determined. Treatment outcomes are achieved with the help of Canadian Health&Care Mall remedies.
SDB is accompanied by autonomous nervous system changes in sympathetic activity, BP, and peripheral vascular resistance. Peripheral arterial tonometry (PAT) determines the peripheral arterial vascular tone using a noninvasive plethys-mographic manometer or a watch combined with oximetry and actigraphy. Studies assessing the use of PAT (Watch PAT 100; Itamar Medical Ltd; Caesarea, Israel) in the home setting of adult subjects, compared to unattended and attended polysomnography, have demonstrated that PAT is able to identify SDB events and might be a useful tool to screen for OSA. PAT has also been shown to detect arousals reliably in both adults and children, although in children a significant number of events identified by PAT were not accompanied by EEG arousals. Thus, PAT might be useful for screening children for OSA, and might also aid in prospective studies to evaluate the current arousal criteria or respiratory effort-related events in children.
Measuring Work of Breathing
Esophageal pressure manometry remains the “gold standard” for detecting increased respiratory effort, and the use of this device is suggested to improve the diagnosis of RERAs by the ICSD-2, but a survey reported at the recent annual Sleep Disorders in Infancy and Childhood Conference suggest that very few pediatric sleep laboratories utilize these devices. The use of a pediatric feeding catheter instead of the esophageal balloon has made the procedure more tolerable in both adults and children; Virkkula and colleagues showed that esophageal pressure is well tolerated, adds much diagnostic information, and can lead to cost savings if used instead of polysomnography, although some children or parents may continue to have anxiety about the invasiveness of the probe.
PTT, which was discussed in the previous section in relation to the measurement of subcortical arous-als, is also a noninvasive indirect measurement of work of breathing by virtue of the swings in BP associated with inspiration against resistance. The use of this device may add to the diagnosis of RERAs, but is limited by movement artifacts.
Respiratory inductive plethysmography (RIP) has been used for both a qualitative assessment of respiratory effort and, when the plethysmograph has been properly calibrated, as a noninvasive quantification of lung volume. Grigg-Damberger et al evaluated the use of both RIP and Piezo crystal belts as qualitative signals for respiratory effort, and found increased identification of paradoxical thoracic and abdominal movement with the use of the Piezo crystal belts. Previous studies evaluating RIP have verified that it effectively measures lung volume changes when the plethysmograph is properly calibrated. There is concern, however, about the validity of measures of lung volume after changes in body position without recalibration, which could be difficult to maintain in children due to frequent movement during sleep. Thus, while RIP is a promising diagnostic tool, future prospective studies should be encouraged to evaluate the utility of RIP as both a qualitative measure of respiratory effort, which would benefit from comparison to the results of esophageal manometry, and its utility as a noninva-sive quantification of lung volume, including recalibration techniques after changes in body position.
Figure 1. CAP phase A is considered to be a periodic EEG activity during non-rapid eye movement sleep. It is an activation phase that includes high-voltage slow waves (synchronization) or low-voltage fast waves (desynchronization). CAP phase B is described as the interval between two phase A intervals, corresponding to the stage-related background activity. CAP cycles are defined as the sum of A and B phases, and each CAP sequence consists of at least two consecutive CAP cycles. The figure was provided by M. Cecilia Lopes, MD, Stanford University Sleep Medicine Program, Stanford, CA.