@article {Peslin871, author = {R Peslin and JF da Silva and F Chabot and C Duvivier}, title = {Respiratory mechanics studied by multiple linear regression in unsedated ventilated patients}, volume = {5}, number = {7}, pages = {871--878}, year = {1992}, doi = {10.1183/09031936.93.05070871}, publisher = {European Respiratory Society}, abstract = {Respiratory mechanics during artificial ventilation are commonly studied with methods which require a specific respiratory pattern. An alternative is to analyse the relationship between tracheal pressure (P) and flow (V{\textquoteright}) by multiple linear regression (MLR) using a suitable model. The value of this approach was evaluated in 12 unsedated patients, mechanically-ventilated for acute respiratory failure, and most with a history of chronic obstructive or restrictive respiratory disease. After correction for the non-linear resistance of the endotracheal tube, the data were analysed with the linear first order model: P = P0 + E.V + R.V{\textquoteright} where E and R are total respiratory elastance and resistance, and P0 is the static recoil pressure at end-expiration. After exclusion of the cycles which clearly exhibited muscular activity, a good fit was observed in 25 out of 36 records (relative root-mean-square error less than 10\%); the values of E and R were reproducible within cycles, and consistent with the patient{\textquoteright}s condition and the ventilatory mode. The intrinsic positive end-expiratory pressure (PEEPi), as derived from P0 and the applied PEEP, averaged 1.1 +/- 1.0 hPa. Using more sophisticated models, allowing for mechanical non-homogeneity or non-linearity of R or E, rarely improved the fit and often provided unrealistic data. In several subjects the discrepancy between the data and the first order model was consistent with expiratory flow limitation, which may severely impair the analysis. We conclude that, except in the case of expiratory flow limitation, the method is useful for routine clinical use and better implemented with the simple linear model.}, issn = {0903-1936}, URL = {https://erj.ersjournals.com/content/5/7/871}, eprint = {https://erj.ersjournals.com/content/5/7/871.full.pdf}, journal = {European Respiratory Journal} }