Compliance of the respiratory system as a predictor for successful extubation in very-low-birthweight infants recovering from respiratory distress syndrome

Objective. To develop additional criteria to predict for successful extubat ion of very-low-birth-weight infants recovering from respiratory distress s yndrome. Design. Prospective study. Setting. Neonatal intensive care unit at a university teaching hospital. Interventions. Infants ready for extubation according to clinical, ventilat ory and blood gas criteria were studied. Before extubation, tidal volume (V t) minute ventilation, respiratory rate/Vt and mean inspiratory flow were m easured during two different ventilatory settings: (i) during intermittent mandatory ventilation (IMV); and (ii) while breathing spontaneously with en dotracheal continuous positive airway pressure (CPAP). Tidal volume was obt ained through electronically integrated flow measured by a hot-wire anemome ter. Total respiratory compliance (Crs) was determined during IMV and was d erived from the formula Vt/PIP-PEEP, where the difference between peak insp iratory pressure (PIP) and positive end-expiratory pressure (PEEP) represen ted the ventilator inflation pressure. Measurements and main results. Each of 49 infants was studied once before e xtubation. 33 infants (67%) were successfully extubated and 16 (32.6%) requ ired reintubation. Infants in the success and failure groups were matched f or gestation postconceptional age, study weight and methylxanthine therapy at the time of study. Successful extubation was associated with a higher me an absolute Crs value (ml/cm H2O) specific Crs value (standardised for body length; ml/cm H2O/cm) compared with infants in whom extubation failed (0.6 7 v. 0.46; P = 0.01 and 0.018 v. 0.014; P = 0.03, respectively). Analysis o f ROC curves detected thresholds for Crs (05 ml/cm H2O) and Vt (7 ml) for p redicting successful extubation. An absolute Crs value 0.5 ml/cm H2O or mor e improved the likelihood of successful extubation when compared with clini cal/ventilator and blood gas criteria. The likelihood of successful extubat ion was 81% if the Crs value was greater than or equal to 0.5 ml/cm H2O. A tidal volume of 7 mi or more was less sensitive in contributing to successf ul extubation (sensitivity 69%). The major causes for extubation failure in cluded atelectasis (diffuse and/or localised) and the presence of a patent ductus arteriosus. Conclusions. In addition to following very precise venti latory criteria for extubation, we found that bedside measurement of total respiratory system compliance added to the likelihood of extubation success in infants recovering from respiratory distress syndrome. Prospective stud ies are needed to validate the findings of this study.