We hypothesized that during partial liquid ventilation (PLV) with perflubron in spontaneously breathing lung injured animals, a fully-synchronized mode of ventilation, assist-control (A/C), would reduce subject effort when compared to IMV and SIMV. Ten newborn piglets(1.19±0.40 kg) with saline lavage-induced lung injury (PaO2<100 torr at FiO2 1.0) were randomized to sequential 30 minute periods of PLV during IMV→SIMV→A/C, or A/C→SIMV→IMV. Piglets were ventilated using time-cycled, pressure limited, volume targeted (15 cc/kg) ventilation(Dräger Babylog®). Respiratory rate (RR) and minute ventilation (Ve) were determined as 1 minute moving averages every 15 seconds; tidal volume(Vt), mean airway presure (MAP), and an esophageal pressure-time index(PE·RR) to estimate subject, not mechanical, work of breathing were determined for all breaths. PE·RR was defined as the area below baseline of the esophageal pressure-time curve × RR, and was recorded using a computer-assisted lung mechanics analyzer (VenTrak®). Blood gases were recorded every 30 seconds using an in-line continuous blood gas analyzer(Paratrend 7®); a/A was calculated. Vt variation was assessed using the coefficient of variation (V; SD/mean × 100). Data analysis used paired t-tests with Bonferroni correction. Wilcoxon rank-sum test was used for nonparametric data. Results: Subject work, estimated by PE·RR, was significantly lower with A/C vs SIMV during PLV (A/C vs IMV, p=0.06). pH was higher and a/A was significantly improved using A/C. Ve and MAP increased during A/C. RR was significantly less in AC vs SIMV, and trended lower in A/C vs IMV (p=0.07). Vt was always more consistent during A/C.Conclusion: In spontaneously breathing animals, fully-synchronized PLV using A/C, which required the least patient effort, increased pH, Ve, and a/A at the lowest RR and least variable Vt. These data suggest substantial physiologic benefit from A/C during PLV in nonparalyzed subjects. Perflubron provided by Alliance Pharmaceutical Corp./Hoechst Marion Roussel. VenTrak provided by Novametrix Medical Systems. Babylog provided by Dräger Inc. Table

Table 1 Table legend: *p<0.05 A/C vs IMV, SIMV; **p<0.05 A/C vs SIMV. Data displayed as mean values±standard error.