To study the effect of ventilation strategy on perfluorochemical (PFC) elimination profile (evaporative loss profile; EL), 6 ml/kg of perflubron were instilled into anesthetized normal rabbits. The strategy was to maintain minute ventilation (V˙e, in ml/min) in three groups:V˙e L (low-range V˙e, 208 ± 2), V˙e M (midrangeV˙e, 250 ± 9), andV˙e H (high-range V˙e, 293 ± 1) over 4 h. In three other groups, respiratory rate (RR, breaths/min) was controlled at 20, 30, or 50 with a constant V˙e and adjusted tidal volume. PFC content in the expired gas was measured, and EL was calculated. There was a significant V˙e- and time-dependent effect on EL. Initially, percent PFC saturation and loss rate decreased in the V˙e H >V˙e M >V˙e L groups, but by 3 h the lower percent PFC saturation resulted in a loss rate such thatV˙e H <V˙e M <V˙e L at 4 h. For the groups at constant V˙e, there was a significant time effect on EL but no RR effect. In conclusion, EL profile is dependent on V˙e with little effect of the RR-tidal volume combination. Thus measurement of EL andV˙e should be considered for the replacement dosing schemes during partial liquid ventilation.
- partial liquid ventilation
- minute ventilation
partial liquid ventilation (PLV) with perfluorochemical (PFC) has been shown effective in treating a variety of respiratory diseases and is currently in Phase II/III clinical trials with perflubron. Numerous investigators have reported improved gas exchange, lung mechanics, and survival, as well as a reduction in the degree of lung injury and inflammatory infiltrates in animals and humans on PLV compared with gas ventilation (2, 4, 12, 20, 24, 25, 33, 38, 39, 42). A stable PFC volume in the lungs is desirable to optimize the beneficial effects of PLV (29). However, the best way to maintain the PFC liquid level in the lungs is still under investigation. Evaporation of PFC is related to a number of factors that make PFC replacement dosing unclear. Previous investigations have shown that the elimination of PFC from the lungs is associated with the air-PFC liquid interface and influenced by the PFC physical properties, the dosing volume and frequency of dosing, body position, lung volume (possibly in combination with positive end expiratory pressure), and the effect of lung injury in combination with extracorporeal membrane oxygenation (16, 17, 29, 37, 41, 43). However, there is no available information on ventilation strategy and its effects on elimination of PFC during PLV.
With regard to ventilation strategy, we hypothesized that the PFC elimination profile (EL; percent PFC saturation and PFC loss rate) is dependent on minute ventilation (V˙e) and the respiratory rate (RR)-tidal volume (Vt) combination. Therefore, the present study was designed to test the effects of V˙e and RR-Vt combination on the PFC elimination from the respiratory system. In addition, a secondary endpoint of the study was to evaluate the effect of PFC elimination on pulmonary gas exchange and function.
All animals were managed according to the NIH regulationsGuide for the Care and Use of Laboratory Animals. In addition, all procedures were approved by the Institutional Animal Care Use Committee of Temple University.
Thirty-six New Zealand White juvenile rabbits (weight 2.2 ± 0.1 kg) with normal lungs were anesthetized with an intramuscular injection of a mixture of ketamine (23 mg/kg), acepromazine (0.58 mg/kg), and xylazine (0.8 mg/kg). This particular animal model was chosen on the basis of our previous experience with the model, the suitability of the model as it applies to neonatal and pediatric medicine, and the history of this preparation for PLV studies (1,16, 17, 21, 37-41). The skin and soft tissues were anesthetized with 0.5% lidocaine HCl (4 mg/kg) and instrumented with tracheostomy placement of a 3-mm-ID endotracheal tube (Hi-Lo Jet Tube, Mallincrodkt, Saint Louis, MO). Catheters were placed in a jugular vein and carotid artery. Conventional gas ventilation with a time-cycled, pressure-limited ventilator (BP-200, Bear Medical Systems, Riverside, CA) was established to maintain physiological cardiopulmonary condition. Subsequent anesthesia was maintained with intravenous injection of the above anesthesia drug (one-tenth of initial anesthesia dose) hourly. Muscle relaxation was induced by intravenous administration of pancuronium bromide (0.1 mg/kg) and maintained with continuous infusion (0.15 mg · kg−1 · h−1) throughout the experiments. Maintenance fluid was provided by a continuous infusion of 5% dextrose at a rate of 5 mg · kg−1 · h−1. Arterial blood pressure was monitored by attaching the arterial catheter to a standard pressure transducer, connected to a neonatal monitor (Athena/Neonatal 9040, S & W Medico Teknik, Albertslund, Denmark). Electrocardiogram electrodes and a rectal temperature probe were inserted for monitoring. The animal's rectal temperature was maintained within 37–38°C.
Nineteen animals were randomly assigned to one of three groups to study the effect of V˙e on the PFC elimination profile. Constant V˙e was maintained in each group throughout the experiment: low-range V˙e target (V˙e L = 205–215), midrangeV˙e target (V˙e M = 245–255), and high-range V˙e target (V˙e H = 285–395 ml · kg−1 · min−1), which were determined by our pilot study to keep the arterial partial pressure of CO2 (PaCO2) in a physiological range. The V˙e was maintained constant in each group by monitoring Vt every 15–30 min using pneumotachography (PEDS-LAB, MAS, Hatfield, PA) and appropriately adjusting ventilator peak inflation pressure. In addition, several healthy rabbits ventilated without PFC were also studied, as described by Tutuncu et al. (40), to confirm cardiopulmonary stability of the animal preparation over time.
Ventilation strategy protocol.
Seventeen animals were randomly assigned to one of three additional groups to study the effect of varying RR and Vt on PFC elimination from the respiratory system. RR (breaths/min) was controlled at RR20 = 20, RR30 = 30, and RR50 = 50, and Vt was corrected by adjusting peak inflation pressure and positive end-expiratory pressure to maintain V˙e (V˙e = 275–290 ml · min−1 · kg−1) and mean airway pressure (7–8 cmH2O) in a constant range. Mean airway pressure was maintained constant in an attempt to regulate end-expiratory lung volume, which is associated with the air-PFC liquid interface.
In all study groups, other ventilation profiles were maintained constant: inspiratory time 0.3 s, positive end-expiratory pressure 5–6 cmH2O, inspired O2 fraction = 1.0. Baseline expired gas and arterial blood samples were obtained followed by sequential measurements every 60 min. Pulmonary mechanics (PEDS-LAB, MAS, Hatfield, PA) were measured every 60 min as previously described (3). Animal position (supine), airway temperature (35°C), and humidity (100%) were kept constant throughout the experiment. Perflubron (6 ml/kg; LiquiVent, Alliance Pharmaceutical, San Diego, CA) was instilled into the lungs via the side port of the endotracheal tube for 5–10 min. During instillation, the animal was repositioned to optimize PFC distribution by using four equal increments of the total dose, with Trendelenberg, reverse Trendelenberg, and left and right lateral decubitus positions.
Expired gas samples were manually collected over 30 s (using a stopcock on a side port of the endotracheal tube into a 50-ml rebreathing bag) at 15 min after PFC instillation and then every 30 min (16, 17, 21, 29, 41). In each sample of expired gas, the percent PFC saturation and expired CO2 tension (NOVA blood gas instrument) at 37°C were measured by using a modification of a previously described thermal detector device (29). The percent PFC saturation values were corrected for the CO2 content and water vapor in the expired gas samples as previously described (29).
All of the experiments were continued for 4 h. Arterial partial pressure of O2 (PaO2) was maintained >100 mmHg and pH was maintained in a physiological range. At the end of the experiments, the animals were euthanized with 15% potassium chloride.
One-way analysis of variance (ANOVA) was used to compare basic physiological data as a function of time. Two-way ANOVA and post hoc analysis with Student-Newman-Keuls correction for multiple comparisons were performed to compare EL and other parameters as a function of time and groups. One-way ANOVA and post hoc analysis with Student-Newman-Keuls correction were performed to compare the values of EL and other parameters at each time point. Significance was accepted at the P < 0.05 level.
As shown in Table 1 (mean ± SE), body weight did not differ significantly across groups.V˙e was maintained constant in three significantly (P < 0.01) different ranges (V˙e H = 293 ± 1 > MHM = 250 ± 9 >V˙e L = 208 ± 2 ml · kg−1 · min−1). Figure1 shows the percent PFC saturations and the loss rate of PFC over time for each group. As shown, the relationship is significantly (P < 0.05) dependent onV˙e and time. After the initial instillation, the expired gas in all groups was saturated with approximately the same amount of PFC (50% saturation). The percent PFC saturation of expired gas declined over time for all groups; however, the expired gas remained relatively saturated with PFC for a longer time at lower levels of V˙e. Initially, the loss rate of the groups with higher V˙e was greater (V˙e H >V˙e M >V˙e L) because of theV˙e, but, by 3 h, the lower percent PFC saturation resulted in a loss rate such thatV˙e H <V˙e M <V˙e L at 4 h.
The relationship of residual PFC in the lung and the PFC loss as a percentage of initial dose over time for each group is shown in Fig.2. There were lower amounts of residual PFC in the lungs and a higher loss (percentage of initial PFC dose) in the groups with higher V˙e during the first 3 h. Table 2 (means ± SE) demonstrates how the redosing interval is dependent on the V˙e such that replacement doses are needed sooner when the V˙eis higher. For example, it takes 20 min longer for theV˙e L group to lose 20% of the initial dose compared with the V˙e H group.
Table 3 illustrates the cardiopulmonary profiles of animals in the V˙e groups. As a function of time and group, there were significant changes in PaO2 and Cdyn over time (P < 0.05). However, there are significant changes in Cdyn% as a function of time only and in PaCO2 as a function of group only (P< 0.05).
Ventilation Strategy Study
As shown in Table 1, the body weight and V˙e did not differ significantly among groups. By design, different breathing rates were set for each group, and thus Vt was adjusted to maintain a similar V˙e = 283 ± 4 SE ml · kg−1 · min−1. Figure3 demonstrates that percent PFC saturation and PFC loss rate were each significantly (P < 0.05) dependent on time, but there was no effect of RR. As shown, Fig.4 demonstrates that the PFC loss and the residual amount of PFC in the lung are each significantly (P< 0.01) correlated with time, but there was no significant difference with respect to RR.
Table 4 illustrates the cardiopulmonary profiles in the ventilation strategy study. As a function of time and group, there is a significant change in Cdyn over time (P < 0.05). However, there are significant changes in Cdyn% and PaO2 as a function of time only and in PaCO2 as a function of group only (P < 0.05).
Figure 5 illustrates the relationship between residual PFC and both PaO2 and Cdyn% in all study groups (V˙e and ventilation strategy groups). Although parametric in time, these data represent time points from 15 min to 4 h after PFC instillation. As shown in Fig. 5(top), there was a significant (P < 0.001) positive correlation between the amount of residual PFC and the PaO2 (r = 0.55). Likewise, as shown in Fig. 5 (bottom), there was a significant (P < 0.001) positive correlation between the amount of residual PFC and Cdyn% (r = 0.87). These data indicate a physiological correlation between oxygenation, lung mechanics, and residual perflubron in the lungs; thus, as residual perflubron is diminished as a result of evaporative loss, there is a deterioration in both oxygenation and lung compliance.
The data presented herein demonstrate that whenV˙e is increased during PLV, the evaporative loss of PFC is increased. This information has significant implications with respect to subsequent redosing during the clinical application of PLV. Higher evaporative loss from the lungs in subjects treated with higherV˙e indicates it is necessary to replace PFC sooner than those with lower V˙e. As noted herein, to maintain 80% of the initial dose in the lungs, theV˙e H group would require dosing 20 min more frequently than the V˙e L group. In addition, as noted in Fig. 5, there are graded physiological consequences in oxygenation and pulmonary compliance associated with evaporative loss and no replacement dosing. Furthermore, during PLV with the same V˙e, it was shown that all components of EL are dependent on time after instillation and not ventilation strategy (high RR vs. low RR). Thus ventilation strategy has little impact on PFC redosing schedule during PLV.
The physiochemical properties of PFC liquid have been studied by previous investigators (7, 18, 22, 27, 44). The vapor pressure of breathable PFC liquids ranges from 0.2–400 mmHg at a temperature of 37°C, so the rate of evaporative loss from the respiratory system is vapor pressure dependent (29). Also, PFC volatilization from the lungs is the major route of elimination from the body, because PFC is not metabolized in the body (31,43). In addition to vapor pressure, several other factors influence the elimination of PFC from the lungs, including the physicochemical properties of PFC and physiological factors of animals or humans (43).
As previously reported, it appears that maintaining the amount of PFC liquid in damaged lungs prevents time-dependent changes in gas exchange and pulmonary function during PLV (1, 6, 8, 9, 11-14,23-25, 38, 39, 42, 44). Therefore, it is valuable to have a continuous assessment of the amount of PFC in the lung to maintain the physiological effectiveness of PLV. Several radiological studies have also confirmed the evaporation and redistribution processes in animals and humans after administration of various amounts of PFC into the lungs (8, 10, 17, 34, 36, 45). On the basis of these findings, investigators focused on the design of methods for monitoring the elimination of PFC from the lungs (15, 21, 29). Using the thermal detector method, Miller et al. (17) reported that the elimination rate after multiple doses of PFC would be significantly higher than that after a single dose. Likewise, Weis et al. (41) reported that the elimination of PFC was dependent on the initial dose and the time after instillation. Preliminary data indicate that repositioning during PLV modulates PFC elimination from the lungs and that position (supine vs. prone) during PLV results in a different loss rate (5, 17, 29). To prevent PFC elimination during PLV, several ventilator circuit designs for PFC dose maintenance are currently under investigation (19,28).
In the present study, the primary focus was on V˙e and ventilation strategy and subsequent effects on perflubron elimination during 4 h of PLV. In addition, the physiological consequences associated with perflubron evaporation were correlated withV˙e and ventilation strategy. Numerous animal and clinical studies have reported the need to redose perflubron during short-term and long-term PLV; however, none of these studies was able to correlate the need for redosing with the actual evaporative loss of perflubron (9, 11-13, 23, 26, 42). The results presented herein show that, with regard to ventilator management,V˙e is a major factor that influences the elimination of perflubron from the lungs. Thus, whenever V˙e is adjusted during PLV as a result of ventilator management and clinical care, it is necessary to consider the influence of V˙echanges on perflubron elimination and to adjust the redosing schedule appropriately. As noted by Weis et al. (41), the need for redosing is potentiated by initial dose requirements; that is, lower initial doses (6 ml/kg) require more frequent dosing than larger initial doses (18 ml/kg). It is also noteworthy that in the present study it was found that, as long as V˙e was maintained constant, alterations in RR from 20 to 50 breaths/min had little effect on PFC loss profile.
The issue of an “optimum initial PFC dose” is still under investigation, although doses from 25 to 100% of FRC have been studied with varied success (6, 14, 26, 32, 42). In the present experiment, we initially administered 6 ml/kg perflubron into the lungs, ∼1/3 the normal FRC volume of a juvenile rabbit (17,41). We based this initial dose on previous studies (17,41) as well as on the low dose protocol for the current phase III, adult clinical trial for PLV (personal communication, Dr. Mark Wedel, Alliance Pharmaceutical). In addition, it has been reported that this volume of perflubron is sufficient to coat the lung because of a conformational change from a droplet to a film during inflation (35).
Partial liquid ventilation with PFC liquid has been reported to be effective in treating experimental respiratory failure (12,44). Success has been associated with the high solubility of respiratory gases in the PFC liquid that support gas exchange. Furthermore, several investigators also noticed that lung compliance could improve due to the recruitment of alveoli and reducing the surface tension in damaged lungs (23, 35, 38, 39, 41, 42). In contrast to the injured lung, numerous studies have shown that residual PFC liquid in a healthy lung does not improve pulmonary mechanics or gas exchange. In addition, it has been shown that the gas exchange response to residual PFC liquid in the lungs is inversely related to lung condition (30). Furthermore, it has been shown that during prolonged PLV (without replacement dosing), there is a deterioration of lung mechanics, gas exchange, and histology, which suggested that atelectasis does occur during return to conventional mechanical ventilation. These findings have also been shown in the short-term studies presented herein. It is noteworthy that time-dependent deterioration in oxygenation and compliance were not observed in our pilot studies of rabbits with healthy lungs without PLV. In addition, to the degree that these studies were in agreement with that of Tutuncu et al. (40), time-dependent deterioration in cardiopulmonary function can be excluded, with reason, as a possible covariant in the relationship between oxygenation and compliance with residual PFC volume.
The mechanism for alterations in gas exchange and lung mechanics can be explained as illustrated in Fig. 6. As shown, when the lungs are initially instilled with PFC (depending on the initial dose), most of the alveoli are recruited. Thus the lung is divided into three compartments: 1) gas filled with a PFC film lining (nondependent); 2) partially gas filled with a PFC film lining and partially PFC liquid filled (midlung region); and3) PFC liquid filled (dependent). As PFC liquid evaporates and the animals are still paralyzed, the alveoli with PFC are more likely to remain expanded. Nondependent and midlung regions demonstrate the highest degree of clearing, whereas the dependent regions remain relatively liquid filled (17). Thus derecruitment of alveoli secondary to PFC evaporation is probably associated with the observed deterioration in oxygenation and compliance (Fig. 5). Although the present study is not supported with histopathological evidence, previously we have shown that nonuniform PFC distribution and histopathological correlates during PLV of the respiratory distress syndrome lung (44). Therefore, to prevent unnecessary compromise in oxygenation and lung mechanics, as shown herein, accurate replacement of evaporated PFC is crucial to maintain optimum treatment with PLV, especially when the V˙e is changed.
In conclusion, for the presented juvenile animal model, it was found that the evaporative loss profile is dependent on V˙ewith little effect of the RR-Vt combination. Thus assessment of PFC evaporative loss and V˙e should be considered for the replacement dosing schemes during PLV. Finally, the present study was conducted in healthy juvenile animals. These data suggest that further studies of injured, larger, and human lungs in a clinical setting are warranted to evaluate the effect of ventilation strategy on PFC elimination.
We gratefully acknowledge the expert technical assistance of Rita Garbarino and Rachel Laudadio.
This work was performed at Temple University School of Medicine and was supported in part by Alliance Pharmaceutical.
Address for reprint requests and other correspondence: M.-J. Jeng, Dept. of Pediatrics, Children's Medical Center, Veterans General Hospital-Taipei, No. 210, Section 2, Shih-Pei Road, Taipei 11217, Taiwan, ROC (E-mail:or ).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society