Are tidal volume measurements in neonatal pressure-controlled ventilation accurate?
Lily C Chow,Andre Vanderhal,Jorge Raber,Augusto Sola
DOI: https://doi.org/10.1002/ppul.10150
Abstract:Bedside pulmonary mechanics monitors (PMM) have become useful in ventilatory management in neonates. These monitors are used more frequently due to recent improvements in data-processing capabilities. PMM devices are often part of the ventilator or are separate units. The accuracy and reliability of these systems have not been carefully evaluated. We compared a single ventilatory parameter, tidal volume (V(t)), as measured by several systems. We looked at two freestanding PMMs: the Ventrak Respiratory Monitoring System (Novametrix, Wallingford, CT) and the Bicore CP-100 Neonatal Pulmonary Monitor (Allied Health Care Products, Riverside, CA), and three ventilators with built-in PMM: the VIP Bird Ventilator (Bird Products Corp., Palm Springs, CA), Siemens Servo 300A (Siemens-Elema AB, Solna, Sweden), and Drager Babylog 8000 (Drager, Inc., Chantilly, VA). A calibrated syringe (Hans Rudolph, Inc., Kansas City, MO) was used to deliver tidal volumes of 4, 10, and 20 mL to each ventilator system coupled with a freestanding PMM. After achieving steady state, six consecutive V(t) readings were taken simultaneously from the freestanding PMM and each ventilator. In a second portion of the bench study, we used pressure-control ventilation and measured exhaled tidal volume (V(te)) while ventilating a Bear Test Lung with the same three ventilators. We adjusted peak inspiratory pressure (PIP) under controlled conditions to achieve the three different targeted tidal volumes on the paired freestanding PMM. Again, six V(te) measurements were recorded for each tidal volume. Means and standard deviations were calculated.The percentage difference in measurement of V(t) delivered by calibrated syringe varied greatly, with the greatest discrepancy seen in the smallest tidal volumes, by up to 28%. In pressure control mode, V(te) as measured by the Siemens was significantly overestimated by 20-95%, with the biggest discrepancy at the smallest V(te), particularly when paired with the Bicore PMM. V(te), as measured by the VIP Bird and Drager paired with the Ventrak PMM, had a tendency to underestimate V(t) by up to 25% at the smallest V(te). However, when paired with the Bicore PMM, these same two ventilators read over target by up to 18%. Under controlled laboratory conditions, we demonstrated that true delivered V(te), as measured by the three ventilators and two freestanding PMM, differed markedly. In general, decreasing dynamic compliance of the tubing was not associated with greater inaccuracy in V(te) measurements.