Effects of inertance on respiratory mechanics measurements in mechanically ventilated children

OBJECTIVE: The use of the first-order linear single compartment model when studying respiratory mechanics classically neglects inertance (Irs). We hypothesized that Irs would affect compliance (Crs) and resistance (Rrs) estimates in mechanically ventilated young children. DESIGN: Prospective study; single-center evaluation. SETTING: University-affiliated tertiary pediatric intensive care unit. PATIENTS: Forty-four patients with and without respiratory disease. INTERVENTIONS: Patients were studied during volume-controlled constant inspiratory flow ventilation. MEASUREMENTS AND MAIN RESULTS: Pressure (PaO) and flow (V') were analyzed according to two different models: a one-compartment first-order linear model according to PaO = (1/Crs) x V + Rrs x V' and a one-compartment second-order linear model according to PaO = (1/Crs) x V + Rrs x V' + Irs x V''. Irs was higher in children with vs. those without respiratory disease (median 0.00224 cm H2O/L/sec2, Q1-Q3 0.00180-0.00321 vs. median 0.00133 cm H2O/L/sec2, Q1-Q3 0.00072-0.00210; p < .001)). A positive correlation between Irs and the difference of Crs estimates between the first- and the second-order model was found in both groups (r = .84, p < .05 and r = .67, p < .05). Rrs estimates were similar in both groups. CONCLUSIONS: This study showed that the linear single-compartment model may not adequately estimate the respiratory mechanical properties in mechanically ventilated children, particularly in the presence of respiratory disease. Including an Irs term significantly diminished Crs estimates. A one-compartment second-order linear model might be a useful clinical tool in more adequately measuring respiratory mechanics and optimizing ventilatory settings in children with respiratory disease.