Main

ARDS in neonates and infants is characterized by hypoxemia and decreased lung compliance due to intrapulmonary shunting and a disturbance of surfactant synthesis and function(1). Although natural surfactant is very effective for improving oxygenation and lung compliance in preterm infants, it has limited success in infants and children with ARDS(2, 3).

In recent years liquid ventilation with PFC has gained increasing interest to improve pulmonary gas exchange in experimental models of ARF(4–6). PFCs are organic hydrogen molecules in which carbon-bound hydrogen atoms are replaced by fluorine atoms. They are biologically inert, non-biotransformable, and immiscible in both aqueous and lipid media. PFC liquids are absorbed minimally by the respiratory epithelium and are eliminated by evaporation through the lungs. Oxygenated PFC liquids have the ability to lower surface tension in surfactant-deficient lungs and are able to dissolve large amounts of respiratory gases. However, total liquid ventilation requires a special apparatus to deliver and remove tidal volumes of liquid, and to oxygenate and remove CO2 from the liquid extracorporeally. In 1991 Fuhrman et al.(7) reported the use of a combined technique of liquid ventilation and tidal gas volume ventilation in 13 piglets with normal lungs. The lungs were filled with a volume of 30 mL kg-1 PFC liquid according to the normal residual capacity followed by tidal gas ventilation using conventional ventilator settings. This technique is much simpler than total liquid ventilation, does not require a special device, and is well tolerated by the animals.

As ARDS in neonates and infants is often combined with PH, the use of a selective pulmonary vasodilator might be of clinical benefit. Recently, inhaled NO has been described as an effective selective pulmonary vasodilator in different experimental models of PH(8–10). It improved oxygenation and reduced elevated pulmonary artery pressure in neonates, infants, and adults with PH or ARF combined with PH(11–15). Inhaled NO improves a ventilation/perfusion mismatch by selectively vasodilating pulmonary vessels of ventilated areas.

The aim of this study was to investigate the effects of PLV combined with inhaled NO on gas exchange, lung mechanics, and pulmonary hemodynamics in an animal model of ARF with PH.

METHODS

The protocol was approved by the Institutional Animal Research Committee and the care of the animals was in accordance with guidelines for ethical animal research.

In 12 piglets of either sex, weighing 12.1 ± 0.6 kg, premedicated with azaperone (8 mg kg-1) and atropine (0.02 mg kg-1), anesthesia was induced with ketamine (10 mg kg-1) and thereafter maintained by a continuous infusion of fentanyl (0.15 μg kg-1 min-1), pentobarbital sodium (4 mg kg-1 h-1), and pancuronium (0.3 mg kg-1 h-1). All animals were placed in a supine position. After placing a 5.5-mm inside diameter Hi-Lo cuffed tube(Hi-Lo Jet Tube, National Catheter Corp., Malinckrodt, Glen Falls, NY) via tracheostomy, controlled mechanical ventilation was established using a volume-controlled, time-cycled ventilator (Evita 2, Dräger Company, Lübeck, FRG). Tidal volume was set at 10-12 mL kg-1, respiratory rate at 24 min-1, inspiratory/expiratory ratio at 1:2, PEEP at 0.4 kPa, and Fio2 at 0.21. Initially a 4-Fr double-lumen catheter (Duocath, Peter von Berg Medizintechnik, Kirchseeon, FRG) was inserted into the right subclavian vein for nutrition and anesthesia. After induction of anesthesia, a Ringer solution was infused, initially, at a rate of 20 mL/kg in 30 min followed by a continuous infusion of 5 mL kg-1 h-1. A 5.5-Fr thermodilution O2 saturation fiberoptic pulmonary artery catheter(Edwards Swan-Ganz Oximetry TD catheter, Edwards Critical Care Division, Irvine, CA) was placed into the pulmonary artery by peripheral cutdown of the right external jugular vein. A 4-Fr O2 saturation catheter (Edslab double lumen O2 Sat II catheter, Edwards Critical Care Division) was inserted via the right common carotid artery and placed into the thoracic aorta for continuous arterial oximetry measurement. A short 16-gauge catheter(Abbocath, Abbott Ireland LTD, Sligo, Republic of Ireland) was placed into the femoral artery for continuous pressure recording. A second 4-Fr double-lumen catheter (Duocath, Peter von Berg Medizintechnik, Kirchseeon, FRG) was placed via the left external jugular vein.

Hemodynamic monitoring. ECG, Pra, Ppa, Ppc′, and Pa pressures were recorded continuously on a multichannel recorder (SMU 612 monitor, PPG Hellige Corp., Freiburg, FRG) using 0.9% saline-filled transducers (Monitoring Kit NM-081-D, Peter von Berg Medizintechnik, Kirchseeon, FRG). All pressures were referred to a zero level taken at the middle of the ventral to dorsal thoracic distance. Q′ was measured at end-expiration by thermodilution technique as the mean of three determinations after injection of 5 mL of 0.9% saline at 0°C. SVR and PVR were calculated using the following equations (SVR, [(Pa - Pra)·60]/Q′; PVR,[(Ppa - Ppc)·60]/Q′). Two oxygen saturation monitors (Baxter Sat2 Oximeter) were used for continuous arterial and mixed venous oximetry measurements.

Respiratory monitoring. Arterial and mixed venous blood samples were taken for measurements of Hb concentration, oxygen saturation, Po2, Pco2, and pH using an automatic blood gas system (AVL 995, AVL Corp., Graz, Austria). In addition the arterial oxygen saturation and mixed venous oxygen saturations and arterial methemoglobin fraction(CO-Oxylite AVL 912, AVL Corp., Graz, Austria) were analyzed. End-tidal CO2 concentrations and the fraction of inspired (Fio2) and expired (Feo2) oxygen were measured with a multigas analyzer (PPG Hellige Corp., Freiburg, FRG). Airway pressures, tidal volume, Cdyn, and expiratory Raw were measured with a Bicore CP-100 monitor (Bicore Monitoring Systems, Irvine, CA) using the neonatal Varflex flow transducer connected directly to the endotracheal tube and the air-filled SmartCath esophageal balloon catheter placed into the distal third of the esophagus. Intrapulmonary shunt fraction was calculated using the equation Qs/Qt: (Cc′o2 - Cao2)/(Cc′o2 - Cvo2), Cdyn was normalized to body weight (Cdyn: mL kPa-1 kg-1).

PLV. PLV was started after the induction of ARF by repeated lung lavages using the Rimar 101 PFC liquid (Rimar 101®, Miteni Corp., Milan, Italy). The PFC liquid was oxygenated at an Fio2 of 1.0 and warmed to 38°C. Thereafter the oxygenated PFC liquid (30 mL kg-1) was instilled into the trachea over 5 min via the distal port of the endotracheal tube. PLV was continued for 4 h without changing the ventilator settings. Evaporative losses of the liquid were compensated by adding 5 mL kg-1 h-1 of the oxygenated perfluorocarbon liquid. Rimar 101 has a specific gravity of 1.77 g mL-1 at 25°C, a surface tension of 1.47 dyne kPa-1, vapor pressure of 8.5 kPa at 37°C, an oxygen solubility of 52 mL/100 mL and CO2 solubility of 160 mL/100 mL at 37°C and 1 atm of pressure.

Administration of NO. NO was obtained as 600 ppm in nitrogen(Pulmomix forte, Messer-Griesheim, Austria) and applied using the Pulmonox system. The Pulmonox is a microprocessor-controlled system that allows NO delivery at concentrations from 1 to 40 ppm and continuous inspiratory measurement of NO/NO2 using the chemiluminescence method. A flow-box incorporated into the inspiratory limb of the ventilatory circuit transfers the inspiratory flow rate of the ventilator to the Pulmonox system which adapts the NO dosage on a continuous basis. NO was administered into the inspiratory limb of the ventilatory circuit close to the endotracheal tube using an airway sampling adapter (Engström Sampling adapter, Gambro Engström, Bromma, Sweden). The system was calibrated before each treatment/measurement with special calibration gases.

Protocol. After instrumentation the animals were allowed to rest for 30 min before control measurements were performed. Before starting the lung lavage procedure, the Fio2 of the ventilator was increased to 1.0 and the PEEP levels to 0.6 kPa. ARF (arterial oxygen saturation < 85%) was induced by 4-10 lung lavages using 0.9% saline (20-30 mL kg-1). After the last lung lavage the animals were randomly assigned either for PLV or CMV. After 2 h of PLV or CMV the stable endoperoxide analog of thromboxane(U46619, Cascade Biochem Ltd., Bershire, UK) was infused at a rate of 0.4-0.8μg kg-1 min-1 to increase the ratio of mean pulmonary artery to mean arterial pressure (Ppa/Pa) to 0.5. Thereafter the animals were treated with different doses of inhaled NO (1, 10, and 20 ppm) applied in random order. Measurements were taken after 15 min of each NO concentration or after a stable period of 5 min. Five minutes after termination of NO application, a last measurement was taken under hypoxic conditions.

All animals tolerated the experimental protocol. After the end of the trial the animals were killed with an overdose of potassium chloride.

Data analysis. All statistical analyses on recorded data were performed using the Statview 4.5 software for Macintosh. Values are given as mean ± SEM. All intragroup comparisons were made by analysis of variance for repeated measures with Bonferroni/Dunn correction. For comparison between groups significant differences were tested by analysis of variance at each point with a Bonferroni/Dunn correction. A p value of <0.05 was considered significant.

RESULTS

Five to ten lung lavages significantly decreased Cdyn from 0.11 ± 0.005 to 0.05 ± 0.004 mL kPa-1 kg-1 (p < 0.01) and Pao2/Fio2 from 61.1 ± 1.5 to 7.7 ± 0.5 kPa (p < 0.01). Simultaneously Raw increased from 2.8± 0.13 to 4.9 ± 0.2 kPa l-1 s-1 (p < 0.01).

Gas exchange. Blood gas, acid base, and pulmonary mechanical variables are summarized in Table 1. As shown in Fig. 1, Pao2/Fio2 increased significantly from 8.2 ± 0.4 kPa to 34.8 ± 5.1 kPa (p < 0.01) during PLV, whereas there was only a slight change in oxygenation in the control group. Simultaneously, Q′s/Q′t decreased significantly from 52 ± 4% to 25 ± 3.7% (p < 0.01) during PLV and remained unchanged during CMV (Fig. 2). Paco2 values were not significantly different between the PLV and control groups.

Table 1 Blood gas, acid base, and pulmonary mechanical variables in 12 piglets with ARF and PH during combined treatment of PLV with inhaled NO or CMV combined with inhaled NO (mean ± SEM)
Full size table
Figure 1
figure 1

Pao2/Fio2 ratio (mean ± SEM) during combined treatment of PLV with inhaled NO and CMV with inhaled NO in acute respiratory failure with pulmonary hypertension. BL, baseline;L, lung lavage; asterisks indicate differences within a group(baseline vs other values) *p < 0.05, **p< 0.01; open circles indicate differences between the PLV and CMV groups at corresponding points: ○p < 0.05, ○○ p < 0.01.

Full size image
Figure 2
figure 2

Q′s/Q′t (mean ± SEM) during combined treatment of PLV with inhaled NO and CMV with inhaled NO in ARF with PH. BL, baseline; L, lung lavage; asterisks indicate differences within a group (baseline vs other values)*p < 0.05, **p < 0.01; open circles indicate differences between the PLV and CMV groups at corresponding points:○p < 0.05, ○○ p < 0.01.

Full size image

During the infusion of the endoperoxane analog of thromboxane the Pao2/FiO2 ratio decreased rapidly from 34.8 ± 5.1 kPa to 14.1 ± 0.4 kPa (p < 0.01) in the PLV group and from 9.5± 0.4 kPa to 6.9 ± 0.2 kPa (p < 0.05) in the control group. Inhaled NO rapidly increased Pao2/Fio2 to 45.7± 5.3 kPa during PLV and to 25.9 ± 4.7 kPa during CMV. In addition, Q′s/Q′t decreased significantly from 37± 5% to 16 ± 2.6% (p < 0.01) in the PLV group and from 56 ± 1% to 28 ± 6% (p < 0.01) in the control group. After termination of inhaled NO Pao2/Fio2 ratio decreased rapidly from 45.7 ± 5.3 to 17.3 ± 2 kPa (p < 0.01) in the PLV group and from 25.9 ± 4.7 to 7.4 ± 0.5 kPa(p < 0.01) in the control group. After the induction of ALI, oxygenation was significantly better in the PLV group than in the control group (Fig. 1). When inhaled NO was applied to both ventilatory strategies better oxygenation was also observed in the PLV group than in the control group.

Lung mechanics. There was no significant difference in airway pressures during PLV or CMV throughout the study period. Whereas Cdyn remained between 0.05 ± 0.004 and 0.63 ± 0.01 mL kPa-1 kg-1 during CMV, Cdyn significantly increased from 0.05 ± 0.004 to 0.091± 0.01 mL kPa-1 kg-1 (p < 0.01) during PLV(Table 1). In addition, Raw decreased significantly from 4.7 ± 0.5 to 3.3 ± 0.3 kPa l-1 s-1(p < 0.01) during PLV and remained unchanged at 5 kPa l-1 s-1 during CMV. There was no change of Cdyn or Raw during inhalation of NO in both groups.

Hemodynamics. As shown in Table 2 there was no significant difference in hemodynamic parameters between the PLV and control groups. The infusion of the endoperoxane analog of thromboxane increased mean Ppa > 7.0 kPa in both groups. Inhaled NO rapidly reduced mean Ppa from 7.8 ± 0.26 kPa to 4.2 ± 0.39 kPa (p < 0.01) during PLV and from 7.4 ± 0.26 kPa to 5.1 ± 0.13 kPa(p < 0.01) during CMV. Fig. 3 depicts that there was no significant difference in the reduction in mean Ppa between inhalation of 1, 10, and 20 ppm of NO in both groups. As shown in Table 2, inhaled NO significantly reduced PVR in both groups. On the contrary the continuous infusion of the endoperoxane analogue of thromboxane significantly increased SVR associated with a significant fall in cardiac output in both groups.

Table 2 Hemodynamic data in 12 piglets with ARF and PH during combined treatment of PLV with inhaled NO or conventional mechanical ventilation combined with inhaled NO (mean ± SEM)
Full size table
Figure 3
figure 3

Mean Ppa (mean ± SEM) during combined treatment of PLV with inhaled NO and CMV with inhaled NO in acute respiratory failure with pulmonary hypertension. BL, baseline; L, lung lavage; asterisks indicate differences within a group (baseline vs other values): *p < 0.05, **p < 0.01.

Full size image

All animals survived the 4-h trial. In all animals methemoglobin levels were <1% throughout the experiment.

DISCUSSION

The main features of ARDS are severe hypoxemia, due to ventilation-perfusion mismatch and intraalveolar shunting, and decreased lung compliance with a high alveolar surface tension, due to a disturbance of surfactant synthesis and function. Aggressive ventilatory management is often necessary to maintain adequate blood gases. Based on recent experimental work, it is now evident that overinflation with static transpulmonary pressures higher than 2.9 kPa may aggravate, or even cause, alveolar epithelial and capillary injury(16). A fatal vicious cycle leading to irreversible lung damage and death might occur.

In 1991 Fuhrman et al.(7) introduced the technique of PFC-associated gas exchange or PLV. This new technique combined liquid ventilation at functional residual capacity levels and tidal gas volume ventilation using a conventional ventilator. The authors showed that this technique of liquid ventilation was uniformly well tolerated in 13 piglets with normal lungs.

In 1993 Tütüncü et al.(17) published their experience with intratracheal PFC administration combined with mechanical ventilation in 12 New Zealand rabbits with lung lavage-induced ARDS. The authors observed a rapid improvement in oxygenation (increase of Pao2 from 10.0 ± 2.0 to 55.9 ± 3.6 kPa), an increase in respiratory compliance from 0.19 ± 0.01 to 0.26 ± 0.01 mL kPa-1, and a fall in airway resistance from 4.2 ± 0.3 to 3.2± 0.4 kPa L-1 s-1. The authors concluded that PLV proved to be a successful supportive technique to improve gas exchange at low inflation pressures in ARF.

In the same year Leach et al.(18) published their experience with PLV in premature lambs with RDS. They observed a rapid improvement in oxygenation within 5 min in the PLV group, whereas the animals with CMV developed progressive hypoxemia, hypercarbia, and acidosis. In addition, the authors found a 3-fold increase in Cdyn and no change in Raw during PLV.

In a recent publication Leach et al.(19) confirmed their experience with improved lung mechanics and gas exchange during PLV in premature lambs with RDS. In addition, these authors tested the combined use of PLV with an exogenous surfactant in five premature lambs with RDS. There was a similar but slightly delayed response on Cdyn and gas exchange during PLV combined with surfactant application. The authors concluded that PLV is compatible with exogenous surfactant and that PLV was superior to the used surfactant in this study.

Nesti et al.(20) found a significant increase in static lung compliance due to an increased effective tidal volume and significant lower static end-inspiratory pressures during volume controlled PLV in a gastric aspiration model of ARDS. Recently, Houmes et al.(21) reported that the use of PFC liquid did not result in deleterious effects on hemodynamic parameters in large pigs with ALI. Wolfson et al.(22) showed that vasoactive substances can be delivered to the lungs during PFC liquid ventilation. PFC liquid improved mechanical properties of the lung decreasing surface-active forces and enhanced a uniform distribution of biochemical agents across the lung. This application route of vasodilators had a greater effect on the pulmonary than on the systemic circulation, thereby avoiding systemic hypotension.

In a recent abstract Leach et al.(23) published the effects of inhaled NO during PLV in six lambs with normal lungs and PH induced by the endoperoxane analog of thromboxane (U46619). During PLV and CMV pulmonary artery pressure, pulmonary vascular resistance, cardiac output, and gas exchange were similar. Inhaled NO significantly decreased pulmonary artery pressure and pulmonary vascular resistance during both PLV and CMV. The onset of activity of NO was within 1 min. There was no significant difference between 10 and 80 ppm NO during PLV and CMV. The authors speculated that alveolar recruitment with PLV might enhance the efficacy of inhaled NO when parenchymal disease and PH coexist.

In 1995 Wilcox et al.(24) published in a prospective, nonrandomized, controlled study their initial experience with PLV combined with inhaled NO in fetal lambs with congenital diaphragmatic hernia. PLV significantly improved gas exchange and lung compliance. Inhalation of 80 ppm of NO significantly improved oxygenation and decreased pulmonary artery pressure during PLV. The authors concluded that PLV may be beneficial in the treatment of congenital diaphragmatic hernia and that inhaled NO can be delivered during PLV.

The current study evaluates the efficacy of PLV combined with inhaled NO in an animal model of severe lung injury associated with pulmonary hypertension. First, this study confirms recent findings that PLV substantially improves oxygenation and lung mechanics without any deleterious hemodynamic side effects. Like Tütüncü we noticed a significant decrease in Raw during PLV. In contrast, Nesti et al. measured a higher Paw and Raw during PLV in a gastric aspiration model of ARDS(20). The authors speculated that the increased Raw might be caused by the high density and viscosity of the PFC liquid. In our lavage-induced model of ALI, repeated lung lavages significantly increased Raw. Whereas Raw remained high in the control group, Raw significantly decreased during PLV. This difference in Raw during PLV might be explained on two different experimental models of ALI. On the other hand, we used PEEP levels of 0.6 kPa during PLV. This PEEP level might prevent movement of PFC liquid to the central airways during expiration, which might be responsible for transient increases in Raw. In addition, this study demonstrates that inhaled NO can be delivered easily and safely during PLV, resulting in improved pulmonary hemodynamics and oxygenation when ARF is associated with PH. In this situation PFC liquid ventilation improves lung mechanics and oxygenation, and on the other hand inhaled NO improves pulmonary circulation, resulting in a further increase in oxygenation. However, exact data on NO gas solubility and diffusion capacity in PFC liquid are missing. NO dose-response testing showed that the inhalation of low doses of NO was very effective in improving pulmonary hemodynamics and oxygenation both during PLV and CMV.

CONCLUSION

We confirmed the findings of the literature that PLV significantly improves oxygenation and lung mechanics in severe ALI. In addition, we showed that the combined treatment of PLV with low doses of inhaled NO is feasible and improves pulmonary hemodynamics, resulting in improved oxygenation when ARF is associated with severe PH. However, further prospective, randomized, and controlled experimental as well as clinical studies are necessary to evaluate the impact of this treatment strategy on the course of severe ARF.