ECMO criteria for influenza A (H1N1)-associated ARDS: role of transpulmonary pressure.

Purpose: To assess whether partitioning the elastance of the respiratory system (E (RS)) between lung (E (L)) and chest wall (E (CW)) elastance in order to target values of end-inspiratory transpulmonary pressure (PPLAT(L)) close to its upper physiological limit (25 cmH(2)O) may optimize oxygenation allowing conventional treatment in patients with influenza A (H1N1)-associated ARDS referred for extracorporeal membrane oxygenation (ECMO). METHODS: Prospective data collection of patients with influenza A (H1N1)-associated ARDS referred for ECMO (October 2009-January 2010). Esophageal pressure was used to (a) partition respiratory mechanics between lung and chest wall, (b) titrate positive end-expiratory pressure (PEEP) to target the upper physiological limit of PPLAT(L) (25 cmH(2)O). RESULTS: Fourteen patients were referred for ECMO. In seven patients PPLAT(L) was 27.2 ± 1.2 cmH(2)O; all these patients underwent ECMO. In the other seven patients, PPLAT(L) was 16.6 ± 2.9 cmH(2)O. Raising PEEP (from 17.9 ± 1.2 to 22.3 ± 1.4 cmH(2)O, P = 0.0001) to approach the upper physiological limit of transpulmonary pressure (PPLAT(L) = 25.3 ± 1.7 cm H(2)O) improved oxygenation index (from 37.4 ± 3.7 to 16.5 ± 1.4, P = 0.0001) allowing patients to be treated with conventional ventilation. CONCLUSIONS: Abnormalities of chest wall mechanics may be present in some patients with influenza A (H1N1)-associated ARDS. These abnormalities may not be inferred from measurements of end-inspiratory plateau pressure of the respiratory system (PPLAT(RS)). In these patients, titrating PEEP to PPLAT(RS) may overestimate the incidence of hypoxemia refractory to conventional ventilation leading to inappropriate use of ECMO.