Comparison of BiPAP Nasal Ventilation and Ventilation via Iron Lung in Severe Stable COPD: Study Limitations
In this study, we used surface electrodes to record EMGdi. Doubt could be raised that the EMG might be contaminated by the activities of intercostal or abdominal muscles; nevertheless, this kind of contamination can be negligible in both normal subjects and COPD patients. Actually, good correlation between surface and esophageal EMG has been described by several studies.
To minimize the discomfort of the patients and the consequent influence on the study, we did not record esophageal or transdiaphragmatic pressures (Pdi). However, the excellent correlation found between pressure time integral (PTI) and iEMGdi (r = 0.88 by Belman et al and r = 0.801 by Rochester et al) strongly supported the use of iEMGdi to reflect the diaphragmatic activity in our study. In addition, the study of Rochester and coworkers also showed similar correlation between peak Pdi and iEMGdi (r = 0.818).
Thoracoabdominal asynchrony has long been regarded as a sign of respiratory muscle fatigue, for example, as seen during weaning from mechanical ventilation.2 Although this concept has been challenged by Tobin et al and Pourriat et al, it still reflects a condition in which the load on the inspiratory muscles is on the increase. Since one of the goals of the ventilators is to unload the respiratory muscles, it is reasonable to use TAA as a parameter in assessing the effects of mechanical ventilation. Allen and coworkers have successfully used “phase angle” derived from the time lag between AB and RC motion to reflect TAA in pediatric patients with airflow obstruction. They found that substantial TAA existed and improvement in TAA was closely linked to the improvement in lung mechanics. A similiar study done by Sivan and coworkers also evidenced a strong association between the degree of stridor and the degree of TAA in children. In adult patients with stable COPD, phase angle and inspiratory asynchrony index were also used to reflect TAA.
We could not find statistically significant improvements in iEMGst, iEMGdi, phase angle, Vt, Vt/Ti, RR, PR, etC02, Sa02, Pimax, and PEmax after the use of BiPAP This result agreed with that of Strumpfet al but contradicted other investigations. The differences in the disease severity may be the underlying reason. Since our study excluded patients with orthopnea because of the use of iron lung in supine position, subjects had to be those with less severe COPD. The patients in the study of Carrey and coworkers did have much higher PaC02 level than our patients had (58 vs 45 mm Hg). Some other studies done in patients with acute exacerbation of COPD using nasal or face mask showed a tendency toward less intubation during hospitalization. This also supports the assumption that disease severity could influence the effects of mask ventilation. Further support comes from our finding that the patients with lower FEV, had more obvious reduction in iEMGst. The correlation was statistically significant. The data in Table 2 showed that PaC02 was significantly higher (p = 0.0256) and FVC was also much lower (p = 0.0199) in patients with FEV, below 0.55 L than those with FEV, above 0.55 L. All three parameters, ie, FEV,, FVC, and PaC02, are crucial in reflecting ventilatory function. This suggests that the effect of BiPAP correlates with ventilatory function impairment in patients with severe and stable COPD.