INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: neonatal intensive care; phosphodiesterase inhibitors; sildenafil; persistent fetal circulation syndrome; meconium aspiration syndrome

Persistent pulmonary hypertension of the newborn (PPHN) secondary to meconium aspiration syndrome remains an important cause of morbidity and mortality in term neonates (1). Severe meconium aspiration syndrome produces acute parenchymal lung disease with a rapid onset of pulmonary hypertension with intracardiac and intrapulmonary right-to-left shunting, and together these give rise to a vicious cycle of worsening oxygenation.

Current management of PPHN includes therapies that are targeted at lowering the pulmonary vascular resistance including inhaled nitric oxide, which produces its specific pulmonary vasodilator actions through augmentation of cyclic-guanosine 5'-monophosphate (cGMP) concentrations in the pulmonary artery smooth muscle. This activates cGMP-dependent protein kinase, which increases the opening of calcium-sensitive potassium channels, in turn causing membrane hyperpolarization, and inhibits calcium influx through the L-type calcium channel. This leads to relaxation of the vascular smooth muscle cell (2). cGMP exerts these potent vasodilatory effects in many vascular beds including the lungs, where it is subsequently inactivated by phosphodiesterase-5 (PDE-5). Therefore an alternative, or adjunctive approach to the management of pulmonary hypertension may be to increase endogenous pulmonary cGMP concentrations using specific PDE-5 inhibitors.

Previous studies using the PDE-5 inhibitors dipyridamole (3) and zaprinast (4) have demonstrated the potent pulmonary vasodilator properties of this class of agents. More recently, the potential role of another PDE-5 inhibitor, sildenafil, has been introduced. One of the theoretical advantages of this agent may be its greater selectivity for the pulmonary vascular bed. Oral sildenafil was recently shown to prevent hypoxic pulmonary vasoconstriction without affecting systemic hemodynamics in healthy adults breathing 11% oxygen (5). The efficacy of nebulized sildenafil has recently been demonstrated in lambs with pulmonary hypertension induced by a thromboxane A₂ analog (6).

A recent study published in this journal demonstrated the pulmonary vasodilator properties of enteral sildenafil in a healthy adult model, but raised concerns regarding its adverse effects on oxygenation and intrapulmonary shunt fraction (7). Based on our current understanding of the molecular basis of its endothelial actions, further investigation of the pulmonary vascular effects of sildenafil in neonatal pulmonary hypertension is clearly warranted. The route of administration of the drug can be a key factor determining its usefulness in the critically ill neonate, and with an enteral formulation, its bioavailability, which clearly depends on absorption from the gastro-intestinal tract, may be unpredictable in these patients. Furthermore, these infants may also be at risk of gut ischemia secondary to hypoxia. Aerosolized drug therapy may be difficult in the neonate with severe meconium aspiration not only because of the likely need for continuous treatment given the short action of sildenafil when delivered by this route, but also its limited distribution in the presence of meconium and inflammatory exudates. The intravenous formulation therefore would seem potentially more appealing in these patients.

The purpose of this study is to investigate the effects of intravenous sildenafil on pulmonary hemodynamics and oxygenation, and to compare these effects with inhaled nitric oxide, in a porcine model of neonatal pulmonary hypertension secondary to meconium aspiration.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The study conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No 85-23, revised 1996). Eighteen Danish Landrace piglets weighing 4.9 ± 0.4 kg were premedicated with an intramuscular injection of ketamine (2 mg/kg) and midazolam (1 mg/kg), followed by an intravenous bolus of ketamine (1 mg/kg) administered through the ear vein. A stable level of anesthesia was maintained throughout the experiment with an intravenous infusion of fentanyl (30 µg/kg/h) and midazolam (2 mg/kg/h).

Animals were intubated with a cuffed endotracheal tube and ventilated using volume-controlled ventilation with a Servo ventilator 900D (Siemens, Solna, Sweden); the cuff was inflated throughout the study period. Initial ventilator settings were preset to deliver a minute volume of 300 ml/kg, rate of 30 breaths/min, positive end-expiratory pressure of 4-5 cm H₂O, and inspired oxygen fraction of 0.25.

Valved introducers (William Cook, Denmark) were advanced through a cut-down incision from the right internal carotid artery to the brachiocephalic artery for pressure measurement and blood sampling, and from the right external jugular vein to the superior caval vein for measurement of systemic venous pressure. A 5F thermodilution catheter (Baxter Healthcare Corp., Irvine, CA) for measurement of pulmonary artery pressure and cardiac output was passed under fluoroscopic guidance through a sheath placed in the left external jugular vein to the origin of the left pulmonary artery. Systemic blood pressure, pulmonary arterial pressure and central venous pressure were continuously monitored (Sirecust, Siemens, Sweden), and cardiac output was measured in triplicate by thermodilution.

After completing the preparation, hemodynamics were allowed to stabilize for 15 minutes and then arterial blood gas analysis was performed. Ventilation was adjusted to maintain a PO₂ of 70-100 mm Hg, and PCO₂ of 35-45 mm Hg. After 30 minutes, hemodynamic and ventilatory variables were recorded, cardiac output was measured and blood gas analysis was performed; this set of measurements was taken as baseline (0 minutes).

Following baseline measurements, each animal received a deep intratracheal instillation (3 ml/kg) of a 20% solution of pooled human meconium. First pass human meconium was obtained and stored at 20° C. Before commencing the studies, meconium was thawed and diluted with 0.9% saline to make a 20% solution; the resulting mixture was filtered to remove large particulate matter.

Animals were assigned to one of three groups: in six animals (control group), no further therapeutic interventions other than ventilatory adjustments were made. Six animals received inhaled nitric oxide (20 ppm) beginning 120 minutes after meconium instillation, and the remaining six animals (sildenafil group) received an intravenous infusion of sildenafil beginning 120 minutes after meconium aspiration (2 mg/kg over 2 hours; Pfizer Ltd., Sandwich, England). In all animals, for the first hour after meconium instillation, a blood gas measurement was made at 30 minutes and ventilation was adjusted accordingly. Hemodynamic and blood gas measurements were subsequently recorded at 60, 120, 180, and 240 minutes, and ventilation was adjusted as necessary. At the end of the study period (240 minutes), each animal received a 10-ml injection of radio-opaque contrast into the right atrium under fluoroscopic control to confirm the absence of an intracardiac shunt. Following this, the animals were killed with a lethal injection of pentobarbital.

Statistics

Measures in the three groups at each time point were compared with a one-way analysis of variance (ANOVA). Within group responses to interventions were evaluated with a repeated measures ANOVA, using Bonferroni's correction for multiple comparisons. Results are expressed as mean ± standard error of the mean; a p value of less than 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary and Systemic Hemodynamics

Baseline pulmonary artery pressure, pulmonary capillary wedge pressure, systemic arterial pressure, central venous pressure, and cardiac output were similar in the three groups (Table 1, Figure 1). There was no difference between pulmonary vascular resistances; systemic vascular resistance was slightly lower at baseline in the sildenafil group but this did not reach statistical significance (p = 0.08).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Hemodynamic changes during the study in control (), nitric oxide (), and sildenafil groups (). In all groups, meconium aspiration was induced after baseline readings (after 0 minutes). Intervention with sildenafil or inhaled nitric oxide was commenced after 120 minutes measurements. Error bars represent standard error of the mean (SEM). (A) Changes in mean pulmonary artery (PA) pressure (mm Hg) in at baseline (0 minutes) then hourly during the study period. (B) Changes in pulmonary vascular resistance (PVR; U/kg) in all groups. (C) Changes in the pulmonary-to-systemic vascular resistance ratio (PVR/SVR) in all groups.

Instillation of meconium increased the pulmonary artery pressure significantly in all three groups at 120 minutes (Table 2). There was no difference between the groups in any of the hemodynamic parameters at 120 minutes. Pulmonary capillary wedge pressure was unchanged, and cardiac output was slightly higher in the sildenafil group at 120 minutes, but this difference was not significant (p = 0.15). Pulmonary vascular resistance increased following meconium instillation by 71 ± 37% in the control group, 75 ± 24% in the nitric oxide group, and 72 ± 35% in the sildenafil group (p < 0.05 for all groups). Systemic arterial pressure and central venous pressure were unchanged after meconium instillation, and thus systemic vascular resistance did not change. Pulmonary-to-systemic vascular resistance ratio increased significantly in all groups between 0 and 120 minutes (p < 0.05).

In the control group, there were no further changes in pulmonary and systemic hemodynamics between 120 and 240 minutes. Inhaled nitric oxide reduced pulmonary artery pressure by 21 ± 10% (p < 0.05), and did not significantly affect pulmonary capillary wedge pressure or cardiac output. Thus, pulmonary vascular resistance fell by 36 ± 16% (p < 0.05). Aortic pressure and central venous pressure were unchanged, as was systemic vascular resistance. Pulmonary-to-systemic vascular resistance ratio fell significantly following inhaled nitric oxide (p < 0.05). Intravenous sildenafil reduced the pulmonary artery pressure by 29 ± 7% (p < 0.05), and did not change the pulmonary capillary wedge pressure. The cardiac output increased significantly between 120 and 180 minutes, and this increase was subsequently maintained until 240 minutes. Pulmonary vascular resistance fell by 48 ± 8% with intravenous sildenafil, again with a rapid fall between 120 and 180 minutes, and subsequent maintenance of this level (p < 0.05). Sildenafil lowered the mean aortic pressure by 5 ± 11%, but this was not significant. Systemic vascular resistance was unchanged by intravenous sildenafil, and the pulmonary-to-systemic vascular resistance ratio fell to baseline values at 240 minutes. There was no significant difference in the change in pulmonary artery pressure or pulmonary vascular resistance between intravenous sildenafil and inhaled nitric oxide.

Oxygenation Index

Meconium aspiration produced marked increases in oxygenation index in all animals (p < 0.05 for all groups), but there was no difference in oxygenation index between the three groups of animals at individual time points from 0 to 120 minutes. In control animals the oxygenation index remained at a plateau from 120 minutes until the end of the study period without any further significant changes. Inhaled nitric oxide and intravenous sildenafil had no effect on oxygenation index during the study period.

No animal was found to have an intracardiac right-to-left shunt on contrast injection.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Intravenous sildenafil is an effective and selective pulmonary vasodilator, which completely reversed the changes in pulmonary vascular resistance in this model of meconium aspiration syndrome. The fall in pulmonary vascular resistance was achieved by a combination of a reduction in pulmonary artery pressure without a change in wedge pressure, and an increase in cardiac output. The pulmonary hemodynamic changes produced by sildenafil in this model were comparable to those seen with inhaled nitric oxide therapy. Neither agent had any significant effect on oxygenation.

Aspiration of meconium into the airways (8) causes acute parenchymal (9) and vascular pulmonary pathology, with intrapulmonary and intracardiac right-to-left shunting, and a vicious cycle of hypoxemia and escalating ventilatory requirements, which can further exacerbate the pulmonary hypertension and lung damage. Current therapy for severe meconium aspiration and other causes of PPHN is aimed at controlling and reversing pulmonary hypertension using protective ventilatory strategies and pulmonary vasodilators (10); and in the most severe cases providing a period of extracorporeal life support with lung rest (11).

This study used an established in vivo model of meconium aspiration described by Davey (12), which has since been applied in animals ranging from less than 1 kg to 25 kg (13, 14). All of the piglets in our study had acute, sustained pulmonary hypertension induced by meconium aspiration, with an elevation in oxygenation index, and thus provided us with an ideal model in which to investigate a potential pulmonary vasodilator in PPHN secondary to meconium aspiration syndrome. The timing of our therapeutic intervention (sildenafil or nitric oxide administration) was based on the observation in a pilot group of animals receiving meconium only (not shown here) that pulmonary vascular resistance increased acutely for the first 2 hours, and thereafter reached a plateau without any further significant changes until the end of the study period. We therefore commenced intravenous sildenafil 2 hours after meconium aspiration.

The pharmacokinetics of intravenous sildenafil are currently undetermined, and the dose used in this study was based on limited anecdotal experience, essentially extrapolated from adult regimes using enteral doses of 1-2 mg/kg per dose in the pediatric population, and a shorter study using an intravenous dose of 1 mg/kg in children undergoing cardiac catheterization (15). Our data would suggest that the maximal pulmonary vasodilator effect of sildenafil had been achieved by the end of the first hour of infusion, and that continuation of the infusion did not further reduce pulmonary vascular resistance; nor, however did it produce any adverse effects. An intravenous dose of 2 mg/kg appeared to be well tolerated in this model, and sustained the desired pulmonary vasodilator effect over the 2-hour period without producing any unwanted systemic side effects. However, we cannot yet speculate on the duration of action of this dose of intravenous sildenafil based on our observations and further pharmacokinetics studies are necessary to determine the optimal dosing regime.

We were encouraged by the lack of systemic vasodilatation which intravenous sildenafil produced. One of the major limitations to intravenous vasodilator therapy in patients with pulmonary hypertension has previously been concern regarding the relative lack of selectivity for the pulmonary vasculature. Mean aortic pressure in this study at 4 hours was lower in the sildenafil than in control animals; the latter group had become became relatively hypertensive by the end of the study. However, within the sildenafil group there was no significant change in aortic pressure or systemic vascular resistance after the drug was given. The ratio of pulmonary-to-systemic vascular resistance, a sensitive marker of selectivity of a vasodilator for one or other vasculature, increased significantly in all animals as the pulmonary vascular resistance increased following meconium aspiration. The ratio fell to baseline values after sildenafil administration, thus confirming selectivity for the pulmonary vascular bed. Sildenafil therefore appears to have an advantage over other intravenous agents such as prostacyclin (16), and even less selective PDE-inhibitors such as dipyridamole (3) and zaprinast (4), whose use may be associated with unwanted systemic vasodilatation.

We, like others (6), observed an improvement in cardiac output, which accompanied the fall in pulmonary artery pressure with sildenafil administration. The mechanism for the increase in cardiac output is unclear; this could be secondary to a reduction in right ventricular afterload, or to a myocardial effect of an increase in cGMP. cGMP has been previously shown in isolated preparations to inhibit myocardial cyclic-AMP breakdown by PDE-3, and thus could theoretically augment cardiac performance (17).

Interestingly the improvement in cardiac output during sildenafil administration was not accompanied by a deterioration in oxygenation This observation contrasts in part with data published in this journal by Kleinsasser's group: they demonstrated an increase in cardiac output coupled with a reduction in oxygenation and an increase in intrapulmonary shunt in older, healthy pigs receiving medium and high (> 2 mg/kg) doses of enteral sildenafil (6). Furthermore, a detrimental effect of sildenafil on intrapulmonary shunt might be expected to be more exaggerated in the presence of acute lung injury. We would emphasize that intrapulmonary shunt was not measured in the current study, and the extent of acute lung injury induced by meconium aspiration was not demonstrated by pathologic examination. Our observations of stable indices of ventilation during sildenafil infusion should therefore not be over-interpreted in the absence of an objective assessment of the degree of lung injury, and without measurements of pulmonary ventilation-perfusion distribution. These limitations of the current study would form an important part of future investigations.

Inhaled nitric oxide now has an established role in the acute treatment of PPHN, and leads to vasodilatation in ventilated areas of the lungs, and improvement in oxygenation, and reduces the need for extracorporeal life support (11). However, the beneficial effect of nitric oxide on arterial oxygenation may be limited in neonates with hypoxemic respiratory failure with severe parenchymal lung disease, and without extrapulmonary right-to-left shunting (1). Furthermore, inhaled nitric oxide therapy can be associated with the problems of toxicity, `tolerance' which can occur even after a relatively short treatment period, and rebound pulmonary hypertension after its discontinuation (18). By enhancing intracellular and circulating cGMP concentrations, oral sildenafil has been reported to ameliorate the effects of nitric oxide withdrawal in patients with congenital heart disease who had demonstrated rebound pulmonary hypertension on its initial discontinuation (19). In this journal, Dukarm and coworkers reported a significant reduction in pulmonary vascular resistance following administration of an intravenous E4021 (a PDE-5 inhibitor) to neonatal lambs with pulmonary hypertension induced by in uteroductal ligation (20). They found no further enhancement of the pulmonary hemodynamic effects of this PDE-5 inhibitor after the addition of inhaled nitric oxide, but did observe an improvement in oxygenation at a dose as low as 0.5 ppm. Adjunctive therapy of nitric oxide with sildenafil may therefore offer maximal pulmonary vasodilatation, while allowing nitric oxide dose reduction and reducing the risk of rebound pulmonary hypertension. It would be of interest to examine this interaction in future studies.

Conclusions

Intravenous sildenafil is a selective pulmonary vasodilator, which produces a profound reduction in pulmonary vascular resistance without adversely affecting systemic hemodynamics or oxygenation. In this study, sildenafil was at least as effective a pulmonary vasodilator as inhaled nitric oxide, and neither agent had any discernible effect on oxygenation in this model.

Our data would support suggestions that the PDE-5 inhibitor sildenafil may play an important role in the treatment of neonatal PPHN. The advantage of intermittent or continuous parenteral dosing in this population is clear. The intravenous formulation may represent the ideal route of administration in the unstable neonate with pulmonary hypertension, who may have substantial ventilatory requirements, questionable gastro-intestinal absorption, and who may be at increased risk of gut ischemia.