Monitoring of nonlinear respiratory elastance using a multiple linear regression analysis

Eur Respir J. 2001 Jun;17(6):1158-66. doi: 10.1183/09031936.01.00017801.

Abstract

The elastic pressure/volume (P/V) curve obtained by the multiple linear regression (MLR) technique using a new model, was compared with the quasi-static P/V points obtained by the rapid airway occlusion technique. Seven infants were studied during mechanical ventilation using a pressure controlled mode. The resistive pressure was subtracted from airway opening pressure, thus determining the elastance related pressure, which was then plotted against the volume to make an MLR-elastance curve. Quasi-static P/V curves of the rapid occlusion technique were constructed by plotting the different inspiratory and expiratory volumes against the corresponding values of the quasi-static airway pressure. The calculated MLR-elastance curves closely fit the experimental quasi-static P/V points obtained by the occlusion technique. There were, however, some discrepancies due to the viscoelastic behaviour of the respiratory system. Although slightly altered by these discrepancies, the multiple linear regression-elastance curves did fit the observed quasi-static pressure/volume characteristics for use in clinical practice. The multiple linear regression technique may prove to be clinically useful by continuous monitoring of respiratory system mechanics during mechanical ventilation.

MeSH terms

  • Airway Resistance / physiology*
  • Elasticity
  • Female
  • Humans
  • Infant, Newborn
  • Linear Models*
  • Lung Compliance / physiology*
  • Lung Volume Measurements*
  • Male
  • Nonlinear Dynamics*
  • Respiration, Artificial / statistics & numerical data*
  • Respiratory Distress Syndrome, Newborn / physiopathology*
  • Respiratory Distress Syndrome, Newborn / therapy