Comparison of BiPAP Nasal Ventilation and Ventilation via Iron Lung in Severe Stable COPD: Statistics
The signals of RC and AB were replayed to form “Lissajous figures” (Fig 1) on software (Lotus version 2). A “phase angle” 0 was calculated as an index of TAA. 0 can be measured from a Lissajous figure, in which AB and RC motion are plotted on the x and у axis, respectively, with time as an implicit variable (Fig 1A). Figure 1B shows an example of the Lissajous figure from one breath of one of our patients. The width of the “loop” is an index of asynchrony and is quantified by dividing the width of the figure at mid-RC excursion (m) by the width of the figure at the extremes of AB excursion (s). When 0<9O degree, sin 0 = m/s; for 9O<0<18O, 0 = 180 — ц., where sin ц = m/s.
The integrated diaphragmatic EMG (iEMGdi) was calculated during inspiration using software developed by Belman et al.” The EMG of sternocleidomastoid muscle (iEMGst) was directly rectified and integrated without kicking out the QRS activity because the cardiac electrical activity was negligible on the neck.
After the recording devices were carefully set up, the patient rested in the supine position for 10 min. The data were recorded on both personal computer and strip chart recorder, using those obtained in the last minute as a baseline. Afterward, the patients were randomized to either positive nasal ventilation via BiPAP (Respironics, Monroeville, Pa) or iron lung (Emerson, Cambridge, Mass) ventilation for 40 min. During ventilatory support, the data were recorded for 1 min every 10 min. Pimax and Pt:max were measured again immediately after the ventilatory support. After resting for 10 min, the patient was shifted to the other ventilatory support for another 40 min. The data w’ere also recorded for 1 min every 10 min. During BiPAP ventilation, spontaneous (S) mode w as used with 1PAP around 10 cm H20 and EPAP at 2 cm H20 as the patient could tolerate. The pressure level of the iron lung was set around 20 cm H20 without causing any discomfort. The rate of the iron lung was set around 18 cycle/min, which can be adjusted according to the comfort of patients.
Subsequently, the patient rested for another 10 min to obtain the last baseline data that were recorded again in the last minute. Pimax and PEmax were measured again. Throughout the study, the signals of EMG, ECG, and respiratory movements were displayed on the monitor of the personal computer.
One-way analysis of variance for repeated measures w as used to compare the values during spontaneous breathing, BiPAP, and iron lung. Students t test was used to compare the AiEMGst, pulmonary function, and ABG between the patients with FEV, below’ 0.55 L and those with FEV, al>ove 0.55 L. Simple linear regression was used to evaluate the correlation l>etween FEV, and AiEMGst. Differences were considered statistically significant when p<0.05. All results are reported as mean ± SD.