Comparison of static and dynamic intrinsic positive end-expiratory pressure using the Campbell diagram

Intrinsic positive end-expiratory pressure (PEEPi) due to dynamic hyperinflation has been measured as a plateau airway opening pressure during airway occlusion (PEEPi,stat). PEEPi has also been dynamically determined as a fall in esophageal pressure (Pes) before the inspiratory flow starts (PEEPi,dyn). The aims of the current study were to systematically compare PEEPi,stat and PEEPi,dyn and to explain the underlying mechanisms of their difference. The study was performed in healthy subjects with dynamic hyperinflation induced by expiration through a Starling resistor. The Campbell diagram was constructed for each subject by determining the static pressure-volume curves of the lung (Pst,[l]) and chest wall (Pst,[w]). For a given end-expiratory volume, PEEPi,stat was measured on the Campbell diagram as the pressure difference between Pst(w) and -Pst(l). PEEPi,dyn was measured as mentioned above. The effects of respiratory muscle recruitment on PEEPi,dyn were estimated by the Pes values when Pes started to fall relative to Pst(w). We found that: (1) there was a great variability of the PEEPi,dyn/PEEPi,stat ratio among and within subjects; (2) expiratory muscle recruitment was evident on most occasions; (3) persistent inspiratory muscle activity during expiration was present in some subjects; (4) the Pes values at the start of inspiratory flow were frequently on the left of -Pst(l), which contributed to the difference between PEEPi,stat and PEEPi,dyn and implied a greater dynamic than static elastance presumably due to viscoelastic properties; (5) chest wall distortions characterized by inflation of the abdomen with deflation of the rib cage during the initial inspiratory efforts were observed in three subjects. In conclusion, interpretation of PEEPi,dyn needs to be cautious because both expiratory and tonic inspiratory muscle activities that lead to significant over- or underestimation of PEEPi by PEEPi,dyn, respectively, are associated with acute dynamic hyperinflation. In addition, the effects of viscoelastic properties and chest wall distortions on PEEPi,dyn need to be further investigated.