Prediction of prolonged ventilator dependence in children by respiratory function measurements

Complications of ventilatory support are more common if this assistance is prolonged. Our aim was to determine if results of respiratory function measurement on the first day of ventilation identified children who would develop prolonged ventilatory dependence (> or = 4 days) and whether such results were a more accurate predictor than readily available clinical data. Thirty three children, median age 2 years (range 0.1-13.6), who were supported by a constant flow ventilator and hence had measurements of compliance of the respiratory system (CRS) and resistance of the respiratory system (RRS) on the first day of ventilatory support, were retrospectively identified. Those who needed prolonged ventilatory support had a lower CRS on day one (p < 0.01) and required at any time during their ventilatory career both a higher maximum inspired oxygen concentration (p < 0.01) and peak inspiratory pressure (PIP) (p < 0.01). Logistic regression analysis demonstrated that only a low CRS and high maximum PIP were significantly correlated with prolonged ventilator dependence. A low CRS (<0.4 (ml/cmH2O) kg(-1)) and a high maximum PIP (>27 cmH2O) had similar sensitivities (83%) and specificities (71% and 67% respectively) in predicting prolonged ventilatory dependence. The CRS results, unlike the maximum PIP results, however, were always available on the first day of ventilatory support. We therefore conclude that respiratory function measurements have a role in identifying children who would benefit from strategies to prevent prolonged ventilator dependence.