When the infant was clinically stable, usually within 2 h of ICU admission, prospective respiratory monitoring was commenced during quiet sleep and continued throughout the illness until complete clinical resolution of both the illness and CWD. Throughout the progression of the illness, the tcPos was maintained greater than 55 mm Hg by judicious manipulation of the Flo, necessitating at times up to an Flo of 0.65. Racemic epinephrine was routinely administered to these infants, but all recordings were performed at least 1 h after a treatment with this inhaled medication.
The displacement of the chest wall and abdomen was monitored using a respiratory inductance plethysmograph (RIP) (Respitrace, Ambulatory Monitoring, Ardesley, NY) with the ventilatory move ments measured each 8 h until resolution of the illness. The inductance bands for the RIP (Respibands) were fixed on the chest at the nipple line and midabdomen as previously described for infants by Duffty et al,u with the position of the inductance bands not altered between measurements. Volume calibration for the RIP was not performed but the signals are recorded as an inductance (volts). The volume of each breath (Vt) was obtained from airflow (V), sensed at the mouth by a mask and pneumotachograph (Hewlett-Packard 21609B) and the V was subsequently integrated to Vt (Hewlett-Packard 8815A). All volume measurements were performed during quiet (non-REM) sleep as determined by direct observation and by electro-oculogram, recorded by a high-gain biologic signal amplifier (Hewlett-Packard 8816A). These epochs of quiet sleep constituted less than 15 percent of total time monitored, and were frequently of short duration, often lasting only 10 to 15 min. Simultaneously the transcutaneous oxygen and carbon dioxide (tcPoa and tcPcoJ tensions were measured (Kontron Scientific Ltd, Mississauga, Ontario), and heart rate was recorded.
Analysis of the pattern of breathing was performed during epochs of quiet sleep when the tePco was stable for 5 min. Twenty consecutive breaths were analyzed for the displacement of the chest wall (Irc) and abdominal (Iabd) compartments, and for ventilatory parameters, using the measured Vt (obtained from V) respiratory frequency to obtain the instantaneous minute ventilation (Ve) and the mean inspiratory flow (Vt/Ti). All results were reported as a mean and standard deviation, with the ventilatory volumes normalized by body weight to allow comparison between the infants.
The pattern of movement of each compartment was analyzed. The relationship of the motion of the two compartments was derived with the outward movement (inspiration) arbitrarily defined as a positive signal to derive the Lissajous figures of relative motion of the two compartments after the method of Allen et al. When CWD was present, the inductance of the ribcage (Irc) was derived for the maximal displacement such that a negative Irc was possible (Fig 1) and expressed as the measured phase angle (0m). Furthermore, a calculated phase angle (0c) was also obtained, after the scalar method of Agostini and Mognoni, by using the difference in time to peak displacement for the ribcage and abdomen, divided by the total respiratory cycle time, and expressed as degrees. Statistical analysis was performed using regression techniques, and the two methods of deriving the phase angles were compared by the method of Bland and Altman with statistical significance assumed at the 0.05 level.
Figure 1. Recordings obtained by respiratory inductance plethysmography (RIP) and pneumotachograph during an episode of laryngotracheobronchitis in one infant. The left panel demonstrates synchrony of movement between the chest wall (Irc) and abdomen (Iabd), as well as the tidal volume (Vt), with the Lissajous loop of the relative motion of the Irc and Iabd. The center panel demonstrates moderate chest wall distortion with the initial negative deflection of the Irc, and the change in orientation of the Lissajous loop. The right panel illustrates severe distortion of the chest wall throughout inspiration, with a negative deflection in the Irc, and the loop demonstrates the large phase angle between the Iabd and Irc.