APNEA MONITORING

Most of the principles of apnea monitoring have already been discussed. In practice, the appropriateness of a particular apnea monitor depends on the situation in which it is to be used. For example, a head canopy monitor of ventilation^[227] may be appropriate in an ICU but entirely inappropriate for use at home in a young child. Apnea monitors are usually based on one of three general principles—detection of gas flow, chest wall movement, and gas exchange (i.e., monitors of PaCO₂ or SaO₂ ). Guyatt and colleagues^[238] described a method in which a pressure transducer was connected to a nasal O₂ cannula. Periodic fluctuations of about 1 cm H₂ O due to cyclical respiratory flow were observed during nose breathing. Respiratory rates could easily be obtained with this technique, even when O₂ was flowing through the canula. Direct monitoring of ventilation can be obtained easily in a patient whose trachea is intubated. Use of a rigid, airtight canopy that encases the subject's head with a neck seal has been described by Sorkin and coworkers.^[227] The canopy is continuously flushed with fresh gas. The subject's respiration produces a net flow in and out of the canopy,

1468

Another method of airflow detection was described by Werthammer and associates.^[239] An acoustic monitor encapsulated in silicone rubber was taped 0.25 cm inside a nostril. Eight premature infants were continuously monitored for 1 to 2 hours; 26 episodes of apnea lasting 15 seconds or longer were detected and confirmed by direct observation. An impedance monitor detected only seven of these episodes. Hok and colleagues^[240] described an acoustic method, which was demonstrated to detect hypoventilation and apnea with greater sensitivity than pulse oximetry. An extremely unobtrusive method is to use a tiny, rapid-response hygrometer taped close to a nostril. This type of sensor can monitor respiratory rates at least up to 60 breaths/min as effectively as capnometry in infants or adults.^[241]

Chest wall movement may be detected in several ways. One method is inductive plethysmography,^[220] in which the abdomen and thorax are each encircled by a coil, and respiratory movements are detected as changes in the self-inductance. A more commonly used approach is transthoracic impedance, a technique that is commonly implemented in commercially available electrocardiographic monitors. In this method, a small alternating current (typically 100 microamps) at about 100 kHz is passed through a pair of electrocardiographic leads, allowing transthoracic electrical impedance to be continuously measured by a change in the induced 100-kHz voltage. Low-frequency changes in respiratory impedance can easily be demodulated from the signal.^{[95] [239]} Electromyography of the respiratory muscles can also been used to monitor respiration, ^[206] although it is more difficult because of contamination of the electromyographic signal with a much higher voltage electrocardiographic potential. Photoplethysmography is the monitoring of changes induced by respiration in the light absorption of skin and blood vessels. Photoplethysmographic (PPG) sensors can detect cyclic changes in peripheral blood flow caused by breathing. PPG sensors, consisting of an infrared source and a photocell to measure back-scattered light, were applied near a forearm vein and could detect respiratory movements in patients after general anesthesia.^[242] The main disadvantage of these monitors of patient movement is that they may fail to detect obstructive apnea, in which ventilatory effort or movement can occur without gas flow.

Because apnea may not result in any physiologic abnormality, a measure of its possible adverse effects (i.e., hypoxia or hypercapnia) may be preferable. Continuous measurement of end-tidal CO₂ , although easy in an intubated patient, is more difficult in a patient whose trachea is not intubated because of discomfort from the catheter placement and clogging with mucus and saliva. This problem frequently accompanies direct placement of a cannula into the nasopharynx. Some improvements may be obtained if a catheter is placed inside a nasal airway. A more satisfactory solution is to use specially designed nasal cannulas in which one prong is used to sample exhaled gas while the other is used to deliver O₂ .

Another approach is to use O₂ saturation. George and coworkers^[243] compared arterial desaturation with manually detected events using a variety of monitors, including chin electromyography, airflow measurement, and respiratory movement measured by inductance plethysmography. Apnea was defined as cessation of respiration for more than 10 seconds. Nine overnight records from six patients with sleep apnea were analyzed and compared with SaO₂ measurement using an ear oximeter. Decreases in SaO₂ of more than 3% from baseline were considered significant. Using this criterion, only 1.32% of 4008 apneic episodes were undetected. Catley and associates^[79] used ear oximetry and respiratory inductive plethysmography to monitor patients postoperatively, and they demonstrated that morphine analgesia, compared with regional analgesia with local anesthetic, was associated with a high incidence of hypoxemia and associated obstructive apnea, central apnea, paradoxic breathing, and slow respiratory rate.

Given the proven reliability of pulse oximetry, it seems that this method, perhaps in association with capnography, may provide an excellent method of apnea monitoring. Principles of apnea monitoring have been reviewed elsewhere in detail.^[244]