Abstract
BACKGROUND: Mechanical ventilation of the neonate requires ventilators than can deliver precise and accurate tidal volume (V(T)) and PEEP to avoid lung injury. Due to small neonatal V(T) and the disproportionate effect of endotracheal tube leak in these patients, accomplishing precise and accurate V(T) delivery is difficult. Whereas neonatal ICU ventilators are validated in this population, thorough studies testing the performance of anesthesia ventilators in delivering small V(T) in neonates are lacking. METHODS: Three anesthesia ventilators, Dräger Apollo, GE Avance, and Getinge Flow-i; and 2 ICU ventilators, Medtronic PB980 and Nihon Kohden NKV-550, were tested under volume control mode at V(T) of 5, 20, 40, and 60 mL. Three combinations of lung compliance and airway resistance were tested using a Servo ASL 5000 lung simulator. RESULTS: In a scenario without leak, the measured V(T) was greater than the set V(T) by > 10% in the Apollo (21.0% [18.8-26.0]); measured V(T) was less than the set V(T) by > 10% in the Flow-i (-19% [-20.8 to -18.7]). The Avance, PB980, and NKV-550 presented a volume error < 10% (-9.50% [-10.8 to -4.4], -5.8% [-11.8 to -3.5], and 5.4% [-4.5 to 18.9], respectively). Considering all combinations of set V(T), leaks, and respiratory mechanics, none of the anesthesia ventilators were able to deliver a median measured V(T) within a 10% error. The bias between measured V(T) and set V(T) varied widely among ventilators (from 4.27 mL to -10.59 mL). Additionally, in the Apollo ventilator, PEEP was underdelivered with the largest leak value. CONCLUSIONS: Our results suggest that in comparison with the 2 neonatal ICU ventilators tested, the anesthesia ventilators did not greatly differ in terms of V(T) delivery in the presence of a gas leak.