Main

Surfactant replacement therapy is an effective treatment for neonatal RDS by reducing neonatal mortality and morbidity of pulmonary complications(1). Surfactant instillation has also been reported to have effects on both systemic and pulmonary hemodynamics, and a decrease in pulmonary and systemic arterial pressures has been observed in some clinical trials(28). In animal studies, a transient decrease in pulmonary arterial pressure was observed after natural surfactant instillation in preterm lambs with RDS(9). Similarly, Clyman et al.(10) reported that surfactant instillation significantly prevented a progressive increase in the ratio of pulmonary vascular resistance/systemic vascular resistance in preterm lambs with RDS, implying a pulmonary arterial vasodilatation induced by surfactant. Recently, we studied systemic and pulmonary hemodynamics in surfactant depleted newborn piglets after endotracheal instillation of porcine surfactant(11, 12). Both MABP and systemic vascular resistance decreased significantly after surfactant instillation. CO, however, remained stable. Our data suggest that porcine surfactant instillation induces a peripheral vasodilatation. However, no reductions in pulmonary arterial pressure and pulmonary vascular resistance were found after instillation of surfactant(12).

NO is synthesized from the amino acid precursor L-arginine in the presence of NO synthase(13, 14) and has been identified as the major endothelium-derived relaxing factor(15). It is well known that NO plays an important role in regulation of vascular tone in both the pulmonary and systemic circulations(1618). NO participates in the vasodilatation evoked by many mediators such as: bradykinin, substance P, vasoactive intestinal peptide, and halothane(1922). The aims of this study were to test the hypothesis that NO might be involved in the surfactant-induced vasodilatation in surfactant-deficient newborn piglets, and to further study any effects of surfactant instillation on the pulmonary hemodynamics in newborn piglets with surfactant deficiency.

METHODS

Approval. The experimental protocol was approved by the hospital's ethics committee for animal studies.

Surgical procedures. Seventeen 2-6-d-old piglets were delivered from a local farm on the day of the experiment. The animals were anesthetized with an intraperitoneal injection of pentobarbital sodium (15 mg/kg) and an intramuscular injection of pethidine (2.5 mg/kg), followed by pentobarbital sodium 20 mg/kg i.v. and pethidine 2.5 mg/kg i.v. A continuous pentobarbital sodium infusion (6 mg/kg/h) was administered throughout the experiment. Supplemental doses of pentobarbital sodium were given only as necessary during the surgical procedure. Lidocaine 1% was used for the surgical incision. A peripheral i.v. infusion containing 0.7% NaCl and 1.25% glucose was given at a rate of 10 mL/kg/h. Rectal temperature was kept between 38 and 39 °C with a heating pad and a radiant heating lamp. Heart rate was recorded via skin electrodes.

A tracheotomy was performed, and a 3.5-mm cuffed endotracheal tube was inserted. The animals were ventilated with a Bourns BP 200 pressure-regulated infant ventilator (Bourns Medical Systems, Inc., Riverside, CA). Peak inspiratory pressure and frequency were adjusted to keep Paco2 between 4.0 and 6.0 kPa. Fio2 was set between 0.21 and 0.30 to maintain a Pao2 between 9.0 and 13.0 kPa.

Polyethylene catheters (Portex PE-50; inside diameter 0.58 mm) were positioned in the left femoral artery for sampling arterial blood gases and for recording MABP, and in the inferior caval vein through the left femoral vein for recording CVP and for administration of NO inhibitor. Temperature-corrected blood gases were analyzed with an AVL 945 automatic blood gas system (Schaffhausen, Switzerland). A sternotomy and cardiac catheterization were performed as previously described(12, 23). Briefly, after sternotomy an 8-mm ultrasonic transit-time flow probe (CardioMed flow probe, MediStim, Oslo, Norway) was fitted snugly around the proximal part of the common pulmonary artery for measurement of CO. The flow probe diameter was chosen not to restrict the diameter of the pulmonary artery, and the positioning was checked before each experiment. Catheters were inserted into the pulmonary artery via the anterior wall of the right ventricle for recording pulmonary arterial pressure, and into the left atrium for recording left atrial pressure. All catheters were filled with heparinized saline (4 U/mL), and connected to pressure transducers (Senso-Nor 840; Senso-Nor, Horten, Norway). Mean blood pressures obtained by electronic averaging were obtained from a computer via a signal amplifier (Gould Transducer or Universal, Gould, Cleveland, OH). CO was corrected by each animal's weight to give the cardiac index. Systemic vascular resistance was calculated as (MABP - CVP)/cardiac index, pulmonary vascular resistance was calculated as (pulmonary arterial pressure - left atrial pressure)/cardiac index. The a/A ratio was calculated as Pao2/[(Fio2 × 0.95) - Paco2]. All baseline variables were measured at 40 min after finishing the surgical procedure.

Surfactant deficiency model. Repeated lung lavages were performed through the endotracheal tube with 20 mL/kg 0.9% NaCl heated to 38°C(12, 24). Positive end-expiratory pressure was set at 2 cm H2O. Peak inspiratory pressure was gradually increased to 25 cm H2O. The I:E ratio was 1:2. Fio2 after lavage was 1.0. Ventilator frequency (20-30 min-1) was adjusted to maintain normocapnia(Paco2 was kept between 4 and 6 kPa). The lung lavage was repeated until Pao2 was <10 kPa. Mean ± SD number of lavages in the L-NAME and the saline groups were 13 ± 5 and 13 ± 8, respectively. There was some improvement in the blood gases after completion of lung lavage. Fio2 was thereby reduced to avoid Pao2 outside the predefined range between 9.0 and 13.0 kPa.

One animal in the L-NAME group was given 4% albumin 10 mL/kg because MABP after lung lavage was less than 55 mm Hg. Three animals were excluded before randomization owing to lethal hemorrhage, pneumothorax, and pericarditis. Finally, 14 animals entered the study after lung lavage.

Experimental design. Two randomized groups of piglets were studied. After completion of lung lavage, a period of 60 min of stabilization was allowed before i.v. injection. The L-NAME group (n = 7, weight = 1.9 ± 0.2 kg, age = 4.3 ± 0.8 d) received an i.v. injection of 3 mg/kg L-NAME (Sigma Chemical Co., St. Louis, MO), a NO synthase inhibitor; the saline group (n = 7, weight = 2.0 ± 0.2 kg, age = 4.3± 0.8 d) received the same volume of 0.9% NaCl i.v. The i.v. injection of saline or L-NAME was completed in 5 min. Forty-five minutes after the i.v. injection, both groups received an identical dose of porcine surfactant (200 mg/kg) by endotracheal instillation in two bolus doses (100 mg/kg for each dose) through a feeding catheter inserted into the lumen of the endotracheal tube. The endotracheal tube was temporarily disconnected from the ventilator. The time disconnected from the ventilator to perform each instillation did not exceed 10 s. After the first dose the animals were manually ventilated for 4-6 respiratory cycles, and then reconnected to the ventilator. One minute after the first instillation, the second dose of surfactant was instilled in the same way as the first dose. After the instillation procedure, Fio2 was increased for 2 min to avoid transient hypoxemia. After the endotracheal instillation, the animals were studied for 30 min, and all measurements were taken at 2.5, 5, 10, 20, and 30 min after the instillation. Fio2 and peak inspiratory pressure were adjusted to maintain normoxia and normocapnia, whereas the I:E ratio, frequency, and positive end-expiratory pressure were not changed after instillation.

Data analysis. Results are given as mean ± SD. The Wilcoxon signed rank test was used to determine the effects of repeated lung lavages and injection of L-NAME (or saline). The stability of the model before surfactant instillation was determined by comparing values at 30 and 45 min after injection of either L-NAME or saline. Our previous studies revealed that the maximum effect on hemodynamics was seen 2-5 min after surfactant instillation(11, 12). Therefore, we chose to perform statistical evaluation only at 5 min after surfactant instillation to evaluate the acute hemodynamic effect, and at 30 min for evaluation of delayed effects of surfactant instillation. Friedman's test was used for repeated measures within each group to determine the effects of surfactant instillation starting at 30 min after L-NAME. If Friedman's test showed a significant time difference, nonparametric Dunnet's test was used as a post hoc test. Differences between the two groups were evaluated by comparing Δ values within each group for the effect of surfactant. Δ value = the value before surfactant instillation minus the value at 5 or 30 min after surfactant instillation. The Mann-Whitney U test was used for comparisons between the groups. Two-tailed p values <0.05 were considered statistically significant.

RESULTS

Effects of the lung lavage. There were no significant differences in any of the hemodynamic variables and arterial blood gases between the groups before and after repeated lung lavages(Table 1 and Figs. 14). Pulmonary arterial pressure, pulmonary vascular resistance, and heart rate increased significantly, and the a/A ratio decreased significantly in both groups after lung lavage, whereas MABP and systemic vascular resistance were unaffected(Table 1 and Figs. 1, 2, and 4). Fio2 and peak inspiratory pressure were significantly increased to maintain normoxia and normocapnia (Table 1).

Table 1 Arterial blood gases, acid-base balance, ventilator settings, and heart rate
Figure 1
figure 1

MABP and pulmonary arterial pressure (PAP) in the two groups. The data are expressed as mean ± SD. B, baseline (before lung lavage); L, after lung lavages (before injection L-NAME or saline); Surf., surfactant instillation.*p < 0.05 compared with baseline; #p < 0.01, comparing before and after L-NAME or saline i.v.; §p < 0.01, comparing before and after surfactant instillation.

Figure 4
figure 4

a/A ratio in the two groups. The data are expressed as mean ± SD. B, baseline (before lung lavage); L, after lung lavage (before L-NAME or saline injection); Surf., surfactant instillation. *p < 0.05, comparing before and after surfactant instillation.

Figure 2
figure 2

Systemic vascular resistance (SVR) and pulmonary vascular resistance (PVR) in the two groups. The data are expressed as mean ± SD. B, baseline (before lung lavage);L, after lung lavage (before L-NAME or saline injection);Surf., surfactant instillation. *p < 0.05 compared with baseline; #p < 0.01, comparing before and after L-NAME or saline i.v.; §p < 0.05, comparing before and after surfactant instillation.

Effects of injection of L-NAME or saline. Thirty minutes after injection of L-NAME, MABP increased significantly from 63 ± 11 to 84± 8 mm Hg (p < 0.01), and pulmonary arterial pressure increased significantly from 24 ± 2 to 41 ± 5 mm Hg (p< 0.01) (Fig. 1). Simultaneously, systemic vascular resistance increased significantly from 0.30 ± 0.11 to 0.54 ± 0.16 mm Hg/mL/min/kg (p < 0.01), and pulmonary vascular resistance increased significantly from 0.10 ± 0.04 to 0.25 ± 0.12 mm Hg/mL/min/kg (p < 0.01) (Fig. 2). Cardiac index decreased significantly from 241 ± 135 to 178 ± 106 mL/min/kg (p < 0.05) 30 min after L-NAME injection (Fig. 3). In contrast, the injection of 0.9% NaCl did not modify any variables in the saline group (Figs. 13). No significant differences in any variable between 30 and 45 min after injection were found in either the L-NAME or the saline groups. Heart rate in both groups increased significantly after i.v. injection of either L-NAME or saline (p < 0.01) (Table 1).

Figure 3
figure 3

Cardiac index (CI) in the two groups. The data are expressed as mean ± SD. B, baseline (before lung lavage); L, after lung lavage (before L-NAME or saline injection);Surf., surfactant instillation. *p < 0.05, comparing before and after L-NAME or saline i.v.

Effects of surfactant instillation on systemic hemodynamics. In the L-NAME group, MABP and systemic vascular resistance were not modified by the instillation of porcine surfactant (Figs. 1 and 2). In contrast, in the saline group a significant decrease in MABP (from 66 ± 10 to 53 ± 9 mm Hg, p < 0.01) was observed at 5 min after surfactant instillation (Fig. 1). Moreover, the differences in ΔMABP 5 min after surfactant instillation between the groups were significant (p < 0.01). Systemic vascular resistance decreased significantly (from 0.40 ± 0.13 to 0.33 ± 0.12 mm Hg/mL/min/kg, p < 0.01) 5 min after surfactant instillation in the saline group (Fig. 2), and the differences inΔ-systemic vascular resistance 5 min after surfactant instillation between the groups were also significant (p < 0.01). No significant changes in cardiac index, CVP, and heart rate were seen in any group after surfactant instillation.

Effects of surfactant instillation on pulmonary hemodynamics. A significant decrease in pulmonary arterial pressure (from 29 ± 6 to 23± 6 mm Hg, p < 0.05) was seen only in the saline group 5 min after surfactant instillation, whereas pulmonary arterial pressure in the L-NAME group was not modified (Fig. 1). The differences inΔ-pulmonary arterial pressure between the groups 5 min after the instillation were significant (p < 0.05). The decrease in pulmonary vascular resistance 5 min after surfactant instillation in the saline group was nearly significant (p = 0.06), and the differences in Δ-pulmonary vascular resistance between the groups were significant(p < 0.05). Pulmonary vascular resistance in the L-NAME group, and left atrial pressure in both groups remained unchanged after surfactant instillation.

Effects of L-NAME and surfactant instillation on pulmonary gas exchange. There were no significant differences in both arterial blood gases and base excess before and after L-NAME (or saline) injection, and before and after surfactant instillation. After surfactant instillation, however, the decrease in Fio2 (p < 0.01) and increase in a/A ratio (p < 0.05) were significant in the saline group, but not in the L-NAME group (Fig. 4 and Table 1). Moreover, the Δa/A ratio in the saline group 30 min after surfactant instillation was significantly larger than that in the L-NAME group(p < 0.05). The decrease in peak inspiratory pressure was not significant after surfactant instillation in any group (Table 1).

DISCUSSION

In this study we found that endotracheal instillation of porcine surfactant in surfactant depleted newborn piglets induced a transient systemic vasodilatation and a transient decrease in pulmonary arterial pressure, and these hemodynamic changes were completely abolished by pretreatment with the NO synthase inhibitor L-NAME.

The present model differs from the clinical situation of RDS in many aspects(12). However, in vivo lung lavage mimics one important aspect of RDS, surfactant deficiency. Lachmann et al.(24) first used in vivo lung lavage as an experimental model of RDS in guinea pigs. Suguihara et al.(25) also reported that repeated lung lavages in newborn piglets significantly decreased oxygenation and lung function to the same degree as in newborn infants with RDS. Similarly, after the repeated lung lavages the newborn piglets in this study required a high peak inspiratory pressure (23-25 cm H2O) and Fio2 (0.7-0.8), suggesting that a surfactant deficiency model was successfully established.

L-NAME is a competitive inhibitor of NO synthase(26). The dose of L-NAME, 3 mg/kg, was determined on the basis of pilot studies and previous studies by other investigators(27). Gibson et al.(27) as well as our own pilot studies showed that larger doses of L-NAME caused a rapid and marked decline in CO, leading to death of the animal before the end of the experiment. Although we did not measure any NO-related variables (for instance, plasma nitrite), our study confirms earlier findings by demonstrating that 3 mg/kg L-NAME caused a significant increase in MABP, systemic vascular resistance, pulmonary arterial pressure, and pulmonary vascular resistance, and a significant decrease in CO(27, 28). Obviously, the activity of NO synthase was inhibited by this dose of L-NAME.

This study also confirms our previous findings that porcine surfactant instillation induces significant decreases in MABP and systemic vascular resistance in newborn piglets(11, 12). The time point of the lowest MABP in this study was 5 min after surfactant instillation, which was also similar to our previous study (3.5 min after surfactant instillation). Both MABP and systemic vascular resistance were maintained after surfactant instillation in the animals pretreated with L-NAME, and the changes in MABP and systemic vascular resistance after surfactant instillation between the animals with and without L-NAME pretreatment were significant, suggesting that the decreases in MABP and systemic vascular resistance were completely prevented by L-NAME. Surfactant instillation might therefore induce NO synthase.

Pulmonary arterial pressure transiently decreased from 29 to 23 mm Hg after surfactant instillation in the animals not pretreated with NO synthase inhibitor, which was comparable to data reported by Jobe et al.(9). They used a natural surfactant to treat preterm lambs with RDS, and pulmonary arterial pressure transiently decreased from 32 to 25 mm Hg(9). Compared with the effect of synthetic surfactant on pulmonary arterial pressure, the effect of the porcine surfactant was more short lasting. Kääpä et al.(6, 7) used Doppler ultrasound techniques to assess the changes in blood pressure, and found that the significant decrease in systolic pulmonary arterial pressure was maintained until 45 min or one hour after surfactant treatment. A similar result was reported by Hamdan and Shaw(8). The reasons for the difference may be associated with the different species, the methods for assessing pulmonary arterial pressure, or different surfactant preparations.

Although the decrease in pulmonary vascular resistance after surfactant instillation was not statistically significant (p = 0.06) in the saline group possibly due to relatively high standard deviations and a small number of animals, the decrease in pulmonary arterial pressure was statistically significant. Thus, we speculate that porcine surfactant instillation may also induce a transient pulmonary vasodilatation in newborn piglets with surfactant deficiency.

In contrast to the present data, we previously found that pulmonary arterial pressure did not decrease after surfactant instillation in newborn piglets(12). This discrepancy could at least partly be due to a higher pulmonary arterial pressure before surfactant instillation in the present study (29 mm Hg) compared with the previous study (25 mm Hg). In the present study pulmonary arterial pressure after surfactant instillation decreased to the same level as pulmonary arterial pressure before surfactant instillation in our previous study(12), implying that pulmonary arterial pressure before surfactant instillation in our earlier study might not have been high enough to be affected by the surfactant treatment. Another difference between the protocols is that, in the present study, we increased Fio2 for 2 min after the instillation maneuver to avoid possible pulmonary hypertension caused by hypoxemia. In addition, there was also a longer observation time before surfactant instillation due to the L-NAME or saline injection in the present study.

Pulmonary blood flow did not change after surfactant instillation. This is in agreement with Halliday et al.'s(29) finding. Our previous study showed that surfactant instillation did not cause any significant shunting across the ductus arteriosus in surfactant-depleted newborn piglets(11). Supporting this, Belik and Light(30) found that the ductus arteriosus was functionally closed in newborn piglets. Additionally, we have previously discussed the possibilities of shunting across the foramen ovale, which we considered was unlikely to play any significant role(12).

In the present study, we found that pulmonary arterial pressure did not decrease after surfactant instillation in animals pretreated with L-NAME, and the differences in Δ-pulmonary vascular resistance between the animal pretreated with and without L-NAME were also significant. These findings suggest that the decrease in pulmonary arterial pressure and the possible pulmonary vasodilatation after surfactant instillation are associated with activation of NO synthase.

In the saline group Fio2 requirement decreased significantly, and a/A ratio increased significantly after surfactant instillation, whereas no effect was seen in the L-NAME group. The reason for this non- or poor response to surfactant instillation in the L-NAME group is unclear. A larger intrapulmonary shunting associated with pulmonary hypertension in the L-NAME group may be one of the mechanisms. Supporting this, inhalation of NO significantly reduced pulmonary arterial pressure and intrapulmonary shunting with improvement of oxygenation in adult RDS(31). Furthermore, Karamanoukian et al.(32) found that a combination of inhaled NO with surfactant instillation in lambs with congenital diaphragmatic hernia produced a significantly better oxygenation compared with surfactant instillation alone.

In summary, NO plays an important role in maintenance of vascular tone in systemic and pulmonary circulations in surfactant deficient newborn piglets. Endotracheal instillation of porcine surfactant in newborn piglets with surfactant deficiency induces a transient systemic vasodilatation and a decrease in pulmonary arterial pressure, and these effects are at least partly associated with activation of NO synthase. This mechanism may be responsible for the decrease in blood pressure after surfactant instillation in infants with RDS.