Effect of Inhaled Prostacyclin in Combination with Almitrine on Ventilation–Perfusion Distributions in Experimental Lung Injury

The experimental protocol was approved by the appropriate governmental institution (Bezirksregierung, Koeln, Germany), and the study was performed according to the Helsinki convention for the use and care of animals.

In 24 female pigs weighing 30 ± 3 kg (mean ± SD), anesthesia was induced with 5 mg/kg thiopental and maintained with continuous infusion of 5–10 mg · kg⁻¹· h⁻¹thiopental and 8–12 μg · kg⁻¹· h⁻¹fentanyl. Muscle relaxation was achieved with 0.2–0.4 mg · kg⁻¹· h⁻¹pancuronium. Animals were positioned supine, intubated with a 8.0–9.0-mm ID endotracheal tube (Mallinckroth, Athlone, Ireland), and submitted to volume-controlled mechanical ventilation (Servo 300 A Ventilator; Siemens Elema, Lund, Sweden) with an Fio²of 1.0, a respiratory rate of 20 breaths/min, a tidal volume of 10 ml/kg, an inspiratory/expiratory time ratio of 1:2, and a positive end-expiratory pressure of 5 cm H²O. The ventilator setting remained unchanged throughout the study protocol.

A 16-gauge arterial catheter (Vygon, Ecouen, France) and a 8.5-French venous sheath (Arrow Deutschland GmbH, Erding, Germany) were inserted percutaneously into femoral vessels. A right heart catheter (model AH-05050-7.5 F, Arrow Deutschland GmbH) was positioned in a pulmonary artery during transduced pressure guidance.

The blood temperature, determined by means of the pulmonary artery catheter, was maintained at 36.7 ± 0.9°C during the experiment using an infrared warming lamp and a warming pad. A continuous infusion of 4–5 ml · kg⁻¹· h⁻¹of a balanced electrolyte solution was administered for adequate hydration.

All hemodynamic measurements were taken in the supine position with zero reference level at the midchest. Central venous pressure, mean arterial pressure, mean pulmonary artery pressure (MPAP), and pulmonary capillary wedge pressure were transduced (pvb, Medizintechnik, Kirchseeon, Germany) and recorded (AS/3 Compact, Datex-Ohmeda, Achim, Germany). Cardiac output was measured using standard thermodilution techniques and expressed as the mean of three measurements at end expiration of different respiratory cycles. Heart rate was traced by the blood pressure curve.

Blood samples were collected simultaneously in duplicate, and analysis of arterial and mixed venous blood gases (Po², Pco²), hemoglobin, Met-hemoglobin(MetHb), CO-hemoglobin (COHb), and oxygen saturation (Hbo²) was performed immediately. Blood gases were determined using standard blood gas electrodes (ABL 510, Radiometer, Copenhagen, Denmark). The parameters hemoglobin, MetHb, COHb, Hbao², and Hbv̄o²were measured via  species-specific spectroscopy (OSM 3, Radiometer). Venous admixture (^VA/^T) was calculated using the secondary parameters arterial (Cao²), mixed venous (Cv̄o²), and arterial capillary oxygen content (Cco²):

P^alvo²(alveolar oxygen partial pressure, mmHg) = P^Baro− P^H2O− P^alvco², assuming Fio²= 1.0, P^Baro(barometric pressure) = 760 mmHg, P^H2O(water vapor pressure) = 47 mmHg, and P^alvco²(alveolar carbon dioxide partial pressure, mmHg) = Paco²: 0.8.

The data are presented as the mean of each measurement taken in duplicate.

a/ distributions were analyzed using the multiple inert gas elimination technique (MIGET). Briefly, 45 min before the first blood sampling, an isotonic saline solution equilibrated with six inert gases (sulfur hexaflouride, ethane, cyclopropane, halothane, ether, and acetone) was infused into a peripheral vein at a constant rate of 4 ml/min. Samples of arterial and mixed venous blood and mixed expired gas were collected simultaneously at each study point during several respiratory cycles and analyzed immediately by gas chromatography. The expiratory tubing and the mixing box for the expired gas samples were heated above body temperature to avoid a loss of the more soluble gases in condensed vapor. All samples were taken in duplicate. For each inert gas retention (the ratio of the gas concentration in arterial to that in mixed venous blood) and excretion (the ratio of the gas concentration in expired gas to that in mixed venous blood) were calculated. a/ distributions were estimated as previously described by Wagner et al . 13The duplicate samples were processed separately, resulting in two a/ distributions for each condition investigated in this study. The presented data are the mean values of a/ distributions taken in duplicate.

Shunt (^S/^T) was defined as the fraction of pulmonary blood flow (^T) perfusing unventilated alveoli (a/ = 0). Low a/ regions were defined as those with a/ ratios between 0.005 and 0.1, normal a/ regions as those with a/ ratios between 0.1 and 10, and high a/ regions as those with a/ ratios between 10 and 100. Data for a/ distributions are presented as the fraction of pulmonary blood flow perfusing each lung region and expressed as ^low, ^normal, and ^high. Dead space ventilation (^D/^T) was defined as the fraction of gas entering unperfused lung units (a/ > 100). The position of the distributions was also described by the mean a/ ratio for perfusion and ventilation (mean, meana) and their dispersion by the log SD of both perfusion (log SD) and ventilation (log SDa). These parameters of dispersion do not take into account either shunt or dead space.

The residual sum of squares was the result of testing the compatibility of the inert gas data to the derived a/ distribution by the least squares method. An indication of acceptable quality of the a/ distributions is a residual sum of squares of 5.348 or less in half of the experimental runs (50th percentile). 14In the present study, 57% of residual sum of squares was less than 5.348.

After animal preparation, a baseline measurement of all values was performed. ALI was induced by surfactant depletion caused by repeated lung lavage with saline, as previously described and evaluated by Lachmann et al . 15Each lavage was performed with 40 ml/kg saline prewarmed to a temperature of 37.0°C. During the lavages the animals were disconnected from the respirator (Servo 300 A Ventilator, Siemens Elema) for less than 1 min. Values for ALI were collected after the Pao²remained persistently below 100 mmHg for 1 h without any additional lavage. Subsequently, animals were randomized to investigate four different conditions for 30 min: (1) 25 ng · kg⁻¹· min⁻¹aerosolized prostacyclin alone (n = 6); (2) 1 μg · kg⁻¹· min⁻¹almitrine alone (n = 6); (3) combination of 25 ng · kg⁻¹· min⁻¹aerosolized prostacyclin and 1 μg · kg⁻¹· min⁻¹almitrine (n = 6); or (4) controls with no specific treatment (n = 6).

As previously described, prostacyclin (Wellcome, London, United Kingdom) was administered via  continuous ultrasonic nebulization of a prostacyclin solution into the inspiratory limb of the respiratory circuit. 16The rate of the nebulized solution was calculated before the study by measuring the consumption of the fluid in the nebulizer chamber (Model 6302 595 E 400 E, Siemens Elema) over the time. The prostacyclin solution was prepared by dissolving 500 μg prostacyclin in 50 ml glycine diluent. Before the administration, the solution was individually diluted according to the nebulization rate and the weight of each animal.

Almitrine (Euthérapie, Neuilly-sur-Seine, France) was infused continuously as a solution of 15 mg almitrine bismesylate dissolved in 5 ml malonic acid and diluted in 45 ml saline solution immediately before use.

To exclude a possible effect of the diluent, during the experimental conditions a corresponding amount of glycine buffer alone was nebulized in controls and animals receiving almitrine alone, and a corresponding dose of malonic acid was administered in controls and animals receiving prostacyclin alone.

After 30 min, all hemodynamic and gas exchange parameters as well as MIGET data were determined. At the end of the study, all animals were killed with an intravenous application of potassium chloride.

All values are expressed as mean ± SD. Statistical analyses were performed using the SigmaStat for Windows 5.0 software package (Jandel, San Rafael, CA). Each parameter was analyzed by two-way analysis of variance for repeated measures to compare ALI values with those after the intervention within each group and to compare ALI values and intervention values between the intervention groups. These statistical analyses were followed by the Student-Newman-Keuls test for all pairwise comparison when analysis of variance revealed significant results. P  values < 0.05 were considered significant.

All animals survived the entire study period. Examination of all animals by a veterinary surgeon before the study confirmed the absence of any sign of infection or pulmonary disease. No differences in baseline parameters were observed between the animals. Total hemoglobin, Met-hemoglobin, and CO-hemoglobin remained unchanged throughout the study. A mean of 10 ± 2 lavages had to be performed to obtain a stable ALI with a decrease of Pao²from 522 ± 20 to 62 ± 15 mmHg. In controls, no differences in hemodynamics, gas exchange, and MIGET parameters were determined within 30 min after initiation of ALI (tables 1–3).

Hemodynamic parameters are summarized in table 1. Systemic hemodynamics (heart rate, mean arterial pressure, central venous pressure, pulmonary capillary wedge pressure, cardiac output) remained unchanged throughout the entire study period, whereas MPAP revealed significant changes as a result of experimental procedures: inhaled prostacyclin and combined prostacyclin–almitrine induced a decrease in MPAP when compared with ALI and almitrine (P < 0.05). In contrast, almitrine increased MPAP when compared with ALI, prostacyclin, and combined prostacyclin–almitrine (P < 0.05).

Gas exchange parameters are summarized in table 2. None of the experimental interventions had any influence on Paco². Inhalation of prostacyclin, infusion of almitrine, as well as the combination of prostacyclin and almitrine caused an increase in Pao²and a decrease in ˙Q^VA/˙Q^Twhen compared with ALI (P < 0.05). No differences in Pao²and ˙Q^VA/˙Q^Twere revealed between combined prostacyclin–almitrine and prostacyclin or almitrine alone.

During the entire study period, the MIGET analysis showed no lung regions with a high ˙Va/˙Q ratio. A summary of the other MIGET data is presented in table 3. Analysis of ˙Va/˙Q distributions revealed no changes in ˙Q^lowand ˙V^D/˙V^Tas a result of either intervention. A decrease in ˙Q^S/˙Q^Tand an increase in ˙Q^normalwere observed with prostacyclin alone, almitrine alone, and combined prostacyclin–almitrine (P < 0.05). No differences in ˙Q^S/˙Q^Tand ˙Q^lowwere revealed between combined prostacyclin–almitrine and prostacyclin or almitrine alone. For all conditions, mean ˙Q and mean ˙V^Aas well as log SD˙Q and log SD˙V^Aremained unchanged.

The distribution of ventilation and perfusion after 30 min without treatment, inhalation of prostacyclin alone, administration of almitrine alone, and combined application of prostacyclin–almitrine is demonstrated for one animal from each group in figure 1.

The aim of this study was to determine the effects of the combined application of inhaled prostacyclin and intravenous almitrine on the ˙Va/˙Q distributions in experimental lung injury. Our major finding was that, although combined prostacyclin–almitrine improved gas exchange as a result of redistribution of pulmonary blood flow from nonventilated regions toward ventilated lung areas when compared with ALI, no additional improvement of the ˙Va/˙Q distributions was revealed when compared with inhaled prostacyclin or intravenous almitrine alone.

In ARDS, pulmonary gas exchange is impaired because of a mismatching of ventilation and perfusion in the lung. 2Experimental ALI in this study was also characterized by disturbed ˙Va/˙Q distributions and revealed no differences in hemodynamics, gas exchange, and MIGET data when compared between the groups. Furthermore, control animals remained stable with regard to all measured parameters during the study period of 30 min (fig. 1). These findings indicate a stable and reproducible model of lung injury with a decreased gas exchange caused by impaired ˙Va/˙Q distributions similar to those in ARDS. Nevertheless, this experimental model may reflect clinical situations in which the disturbances in gas exchange are mainly caused by atelectasis. However, although the model reflects primarily atelectasis, the histologic findings correspond to clinical ARDS. 15 

Aerosolized prostacyclin has been shown to improve pulmonary gas exchange in experimental ALI and ARDS. 4,5,17Accordingly, inhaled prostacyclin alone improved gas exchange because of a decrease of ˙Q^S/˙Q^Tand an increase of ˙Q^normalin this study (fig. 1). The simultaneous decrease of MPAP is in accordance with the suggested mechanism of a selective pulmonary vasodilation.

Almitrine bismesylate is a peripheral chemoreceptor stimulant 18that has been reported to improve oxygenation in ALI. 6In contrast to prostacyclin, the mechanism by which almitrine acts is not well understood, but it is suggested that almitrine may cause a pulmonary vasoconstriction preferentially in unventilated lung areas by affecting pulmonary vessels directly. 19,20This mechanism of almitrine may divert blood flow from nonventilated to ventilated lung regions and therefore reduce ˙Q^S/˙Q^T, as was shown in the report by Reyes et al.  6in patients with ARDS receiving almitrine. In the present study, almitrine also reduced ˙Q^S/˙Q^Tfor the benefit of perfusion of better-ventilated lung regions (fig. 1). The fact that almitrine produced an increase in MPAP is consistent with the hypothesis of pulmonary vasoconstriction caused by almitrine.

Our group has shown that the combination of inhaled nitric oxide with almitrine has an additional effect on the improvement of ˙Va/˙Q distribution in the surfactant washout model of experimental ALI. 12In contrast, combined prostacyclin–almitrine in this study failed to reveal additional effects (fig. 1).

Usually, the administration of aerosolized prostacyclin is associated with the problem that alveolar concentrations cannot be measured. Therefore, doses of inhaled prostacyclin demonstrating selective vasodilation in the literature may not reflect the effective portion of the drug in the lung. However, in nonclinical and clinical trials, doses used for selective vasodilation ranges from 1 to 50 ng · kg⁻¹· min⁻¹prostacyclin. 4,5,15,21,22In this study, with 25 ng · kg⁻¹· min⁻¹prostacyclin, the mean of those doses was chosen to avoid vasodilation in lung areas with decreased ventilation but to provide the proposed effect of selective vasodilation. MIGET data suggested the administered dose to be effective in this way, and stable mean arterial pressure values exclude a possible spillover effect on systemic circulation.

In recent trials reporting an effect of almitrine in experimental and clinical lung injury, almitrine was administered in different doses ranging from 2 to 16 μg · kg⁻¹· min⁻¹, demonstrating an improvement of pulmonary gas exchange with high-dose as well as low-dose almitrine infusion. 6,9,23–25To minimize the expected increase in MPAP, we previously performed a dose–response trial that revealed a dosage of 1 μg · kg⁻¹· min⁻¹almitrine to be effective with regard to an improvement of gas exchange. 26Furthermore, 1 μg · kg⁻¹· min⁻¹almitrine has been shown to be effective alone and in combination with inhaled nitric oxide in the same setting of experimental ALI. 12According to these data, we used this dosage in the present study. Consequently, the effects of almitrine alone on ˙Va/˙Q distribution were comparable to those observed previously.

The combination of both drugs did not reveal any improvement of the effects observed with each drug alone. Therefore, we conclude that combined inhaled prostacyclin and intravenous almitrine are not suitable for the treatment of ˙Va/˙Q mismatching in this model of experimental ALI.