End-inspiratory pause prolongation in acute respiratory distress syndrome patients: effects on gas exchange and mechanics.

Aguirre-Bermeo, Hernan; Morán, Indalecio; Bottiroli, Maurizio; Italiano, Stefano; Parrilla, Francisco José; Plazolles, Eugenia; Roche-Campo, Ferran; Mancebo, Jordi

BACKGROUND End-inspiratory pause (EIP) prolongation decreases dead space-to-tidal volume ratio (Vd/Vt) and PaCO2. We do not know the physiological benefits of this approach to improve respiratory system mechanics in acute respiratory distress syndrome (ARDS) patients when mild hypercapnia is of no concern. METHODS The investigation was conducted in an intensive care unit of a university hospital, and 13 ARDS patients were included. The study was designed in three phases. First phase, baseline measurements were taken. Second phase, the EIP was prolonged until one of the following was achieved: (1) EIP of 0.7 s; (2) intrinsic positive end-expiratory pressure ≥1 cmH2O; or (3) inspiratory-expiratory ratio 1:1. Third phase, the Vt was decreased (30 mL every 30 min) until PaCO2 equal to baseline was reached. FiO2, PEEP, airflow and respiratory rate were kept constant. RESULTS EIP was prolonged from 0.12 ± 0.04 to 0.7 s in all patients. This decreased the Vd/Vt and PaCO2 (0.70 ± 0.07 to 0.64 ± 0.08, p < 0.001 and 54 ± 9 to 50 ± 8 mmHg, p = 0.001, respectively). In the third phase, the decrease in Vt (from 6.3 ± 0.8 to 5.6 ± 0.8 mL/Kg PBW, p < 0.001) allowed to decrease plateau pressure and driving pressure (24 ± 3 to 22 ± 3 cmH2O, p < 0.001 and 13.4 ± 3.6 to 10.9 ± 3.1 cmH2O, p < 0.001, respectively) and increased respiratory system compliance from 29 ± 9 to 32 ± 11 mL/cmH2O (p = 0.001). PaO2 did not significantly change. CONCLUSIONS Prolonging EIP allowed a significant decrease in Vt without changes in PaCO2 in passively ventilated ARDS patients. This produced a significant decrease in plateau pressure and driving pressure and significantly increased respiratory system compliance, which suggests less overdistension and less dynamic strain.