Toxicity of Prolonged High Dose Inhaled PGE₁ in Ventilated Neonatal Pigs

Abstract

Objective

To study the toxicity of inhaled PGE₁ (IPGE₁) in healthy ventilated piglets.

Methods

Mechanically ventilated anesthetized piglets received either high dose IPGE₁ (IPGE₁ group) or nebulized saline (control group) continuously for 24 hours. Cardio-respiratory parameters, complete blood counts and serum electrolytes were monitored. Lung histology was evaluated by a masked pathologist for the severity (minimal, moderate, and severe) and extent (focal, multifocal, and diffuse) of histologic injury.

Results

Ten neonatal pigs were instrumented. Four received nebulized saline and six received high dose IPGE₁. There was no evidence of adverse cardio-respiratory effects, bronchial irritation or hypernatremia related to IPGE₁. Diffuse/multifocal alveolar edema and focal polymorphonuclear infiltration was observed in both the control and IPGE₁ groups suggesting that alveolar alterations may be secondary to effects of mechanical ventilation. The most distinct histomorphological abnormalities observed in the IPGE₁ animals were focal ulceration, flattening of the bronchial epithelium and loss of cilia of moderate to severe degree in the trachea and bronchi.

Conclusion

In healthy piglets, inhalation of high dose IPGE₁ was not associated with adverse cardiorespiratory effects, bronchial irritation, or hypernatremia and produced minimal signs of pulmonary toxicity even after 24 hours. Prolonged inhalation of high dose PGE₁ therefore appears safe in newborn piglets.

1. Introduction

Neonatal hypoxemic respiratory failure (NHRF), often referred to as persistent pulmonary hypertension of the newborn (PPHN) because of failure of the elevated pulmonary vascular resistance to decrease postnatally, is associated with a variety of neonatal diseases. Inhaled nitric oxide (INO), the only selective pulmonary vasodilator approved for the treatment of NHRF, is ineffective in 30–46% of infants and requires specialized delivery systems making the treatment expensive and limiting availability (1). Intravenous PGE₁ (ivPGE₁) and PGI₂, potent vasodilators used empirically in the treatment of NHRF, are associated with systemic hypotension and worsening of oxygenation due to increased venous admixture (2–6). This has led investigators to explore the delivery of PGE₁ and PGI₂ directly to the lungs as an inhalation, thus minimizing systemic effects and achieving selective pulmonary vasodilation (7–21). Compared to PGI₂, PGE₁ has a shorter half-life, lower pH (6.3 versus 10.5), bronchodilator action, anti-proliferative and anti-inflammatory effects on the alveolar, interstitial and vascular spaces of the lung (8, 22–25). We have described the emitted dose, stability and aerosol particle size distribution of inhaled PGE₁ in a neonatal ventilator circuit and demonstrated effective pulmonary delivery using magnetic resonance imaging in a ventilated piglet model (26, 27). We have also reported the feasibility, safety and effective delivery of inhaled PGE₁ (IPGE₁) when administered for a maximum duration of three hours in a phase I–II study in term/near-term neonates with NHRF (28, 29). Treatment of PPHN associated with NHRF with IPGE₁ would likely require inhalation for a longer period of time. Although significant data exist regarding the safety of long term ivPGE₁ in neonates with heart disease, the possible toxic effects of IPGE₁, especially local pulmonary toxicity, following prolonged inhalation have not been investigated. In addition, there are concerns that IPGE₁ may be a bronchial irritant and that continuous aerosolization of medications dissolved in normal saline for prolonged periods may result in hypernatremia in the neonatal subject. The objective of this study is to evaluate the safety of 24-hour inhalation of PGE₁ in a neonatal animal model undergoing assisted ventilation with high fractional inspired O₂ concentration (FiO₂) that closely mimics the intended clinical use.

2. Materials and Methods

Subjects

The study was performed on ten healthy domestic piglets (1–9 days old) from a specific pathogen free (SPF) litter after approval by the Institutional Animal Care and Use Committee. All animals received care in compliance with the NIH guidelines.

Animal Preparation

Animals were fasted 1–3 hours prior on the day of the experiment. They were intubated oro-tracheally after intramuscular pre-medication with midazolam (1 mg/kg) and ketamine (33 mg/kg). A tracheotomy was performed if oro-tracheal intubation was unsuccessful. The cephalic vein was catheterized percutaneously for vascular access. The carotid artery was cannulated through a neck incision for continuous blood pressure monitoring and arterial blood sampling. The external jugular vein was also cannulated if percutaneous intravenous access was unsuccessful. Anesthesia was maintained by intravenous infusion of ketamine (20 mg/kg/hour), midazolam (2 mg/kg/hr) and fentanyl (30 μg/kg/hr) titrated to the level of anesthesia as assessed by monitoring heart rate, blood pressure, and response to painful stimuli. Maintenance fluids consisting of dextrose-saline were administered at a rate of 5–10 ml/kg/hour and the rate modified based on the animals’ hemodynamic status. Body temperature was maintained at 38–39° C. Blood pressure was monitored continuously and recorded every 2 hours and as needed. Additional saline was infused, when necessary, to maintain baseline systemic blood pressure. Prophylactic antibiotics were given at 12 hour intervals.

Mechanical Ventilation

Time cycled, pressure limited assisted ventilation (Sechrist) was initiated at a rate of 15 bpm, FiO₂ of 1.0, peak inspiratory pressure of 18 cm H₂O, and positive end expiratory pressure of 4 cm H₂O. During the study, ventilator parameters were adjusted to maintain normocapnia. Oxygen saturations were monitored continuously and arterial blood gas analyses performed every 6 hours and as needed. FiO₂ was maintained at 1.0 throughout the experiment.

Drug Preparation

Animals were randomly assigned to receive IPGE₁ (IPGE₁ group) or nebulized saline (control group) continuously for 24 hours. PGE₁, supplied as 500 μg of Alprostadil in 1 ml of ethanol (Gensia Sicor Pharmaceuticals, Irvine, California), was diluted in normal saline to yield a dose of 1200 ng/kg/min when infused into the nebulizer chamber at a rate of 4 ml/hr.

Administration of Continuous Aerosol

The low flow MiniHeart jet nebulizer (Westmed Inc., Lakewood, Colorado) was used to generate continuous aerosols as described previously (28). The nebulizer was placed in the inspiratory limb of the ventilator circuit ~20 inches from the andotracheal tube (ETT). The oxygen flow through the jet nebulizer was set at 2 LPM. During nebulization, the ventilator flow was adapted according to the additional flow of the nebulizer to maintain alveolar ventilation. At the start of aerosol therapy, the nebulizer chamber was primed with 2 ml of the study medication followed by continuous delivery into the nebulizer chamber at a rate of 4 ml/hour.

Blood Sampling

Blood for arterial blood gas analyses was obtained 20 minutes after initiation of mechanical ventilation and every 6 hours thereafter (ABL 5, radiometer, Copenhagen, Denmark). Additional blood gases were obtained if indicated. Blood samples were obtained at baseline and at the end of the experiment for evaluation of complete blood counts (CBC), and serum electrolytes.

Histomorphological Analysis of the Lungs

At the end of the experimental protocol, the piglets were euthanized by intravenous overdose of pentobarbital (90 mg/kg). At autopsy, the lungs and heart were removed en bloc and put in formalin. Representative sections were obtained from all pulmonary lobes (12 sections per animal) and trachea (five sections per animal), processed routinely for paraffin embedding, and stained with hematoxylin/eosin. Lung histomorphology was assessed by a single pathologist masked to group assignment using a scoring system adapted from those described in literature (30–33). Lung injury was characterized on the basis of both severity of injury and extent of injury. Severity of injury was graded as no, minimal, moderate or severe. Extent of injury was classified based on the percentage of sections per animal with abnormal histological features as none (0%), focal (<25%), multifocal (25–50%), and diffuse (>50%). Trachea, bronchi and alveoli were evaluated separately for evidence of toxicity. The trachea and bronchi were evaluated for epithelial flattening or desquamation, loss of cilia, ulceration, hemorrhage and polymorphonuclear infiltration. The alveoli were scored for the presence of hyaline membranes, edema, alveolar wall thickness, and polymorphonuclear infiltration.

Statistical Analyses

Descriptive statistics were used to summarize sample characteristics. Group differences for continuous variables were assessed using the independent samples t-test. Paired samples t-test was used to compare subjects at baseline and at the end of the experimental protocol. Heart rates and systolic blood pressures were plotted against time to evaluate trends during the experiment. The General Linear Model (GLM) was used to assess the differences in heart rate and systolic blood pressure between the PGE₁ and control groups, considering effects from the repeated measurement of these variables (34). Cross-tabulation was used to evaluate severity (minimal or moderate to severe) and extent (focal, multifocal and diffuse) of lung injury. Two-tailed significance level was set at 0.05. Statistical analyses were performed using the SPSS^® statistical package, version 15.0.1 (SPSS Inc., Chicago, IL, USA) and SAS/STAT^® software, Version [9.1.3] (SAS Institute Inc., Cary, NC, USA).

3. Results

Ten piglets underwent the experimental protocol. Of these, six were randomized to the PGE₁ group and four to the control group.

Baseline Characteristics

The mean birth weight of the piglets was 2.0 kg (range 1.4 to 2.5 kg) and the mean age was 4 days (range 1 to 9 days). Majority of the piglets were male (90%).

Hemodynamic Parameters

The mean heart rate and systolic blood pressure were comparable in animals in the IPGE₁ and control groups (Figs 1 and 2). Although the heart rate was higher and the systolic blood pressure lower at the start of the experiment (probably as a result of the recent instrumentation); no significant time trends in heart rate or blood pressure were observed during the aerosol administration in either group. None of the piglets in either group required vasopressors.

Figure 1. Changes in Heart Rate during Aerosol Administration by Study Group.

Error bars represent ± 1 SD

Figure 2. Changes in Systolic Blood Pressure During Aerosol Administration by Study Group.

Error bars represent ± 1 SD

Respiratory Support and Arterial Blood Gases

The baseline ventilator support was comparable in the two groups (Table 1). In both groups of animals, PaO₂ decreased over time (p<0.01). However, there was no difference between the two groups over the course of the study in the need for ventilator support or blood gas parameters. There was no evidence of airway irritation as manifested by coughing or wheezing associated with significant changes in breathing pattern, heart rate, blood pressure, anesthetic requirement, ventilator peak pressures, ventilator rate or PaCO₂ during PGE₁ or saline inhalation in the anesthetized mechanically ventilated subjects of this study.

Table 1. Respiratory parameters at baseline and at end of study.

	Baseline		24 Hours
	Control (n=4)	PGE1 (n=6)	Control (n=4)	PGE1 (n=6)
Ventilator settings
Rate (bpm)	15±1	16±1	19±2	21±5
Peak pressure (cm H2O)	18±1	19±2	17±1	20±3
Blood gas analyses
pH	7.37±0.14	7.33±0.14	7.29±0.1	7.31±0.1
PaCO2 (mmHg)	34±15	36±13	37±5	36±9
PaO2 (mmHg)	427±46	451±66	238±82	248±136

Values represent mean±SD

Hematological and Biochemical indices

Paired blood samples at baseline and end of study were available for total leukocyte count, hematocrit, platelet count, and serum electrolytes (Table 2). Hematological and biochemical indices were comparable between the two groups at baseline and at the end of the study. Hypernatremia was not documented in any animal in either group.

Table 2. Hematological and Biochemical indices at baseline and at end of study.

	Baseline		End of Study
	Control (n=4)	PGE1 (n=6)	Control (n=4)	PGE1 (n=6)
TLC (1000/mm3)	8.2±4.4	13.0±6.8	12.9±3.5	16.0±5.5
Hematocrit (%)	29.6±5.2	34.6±7.7	32.4±5.1	31.0±5.5
Platelets (1000/mm3)	217±28	406±152	215±64	400±163
Sodium (mMol/L)	133.4±0.5	138.5±6	131.0±7.0	132.0±6.9
Potassium (mMol/L)	4.0±0.2	3.8±0.6	5.2±0.8	4.3±1.3

Values represent mean±SD

Light Microscopy

Direct pulmonary toxicity of the inhaled medication was evaluated by histologic examination of five tracheal sections and 12 parenchymal sections from each animal.

There was no evidence of severe toxic injury in the form of hyaline membranes, diffuse/severe neutrophilic infiltration or ulceration in the lungs and trachea in either the control or IPGE₁ groups. Majority of the animals in the control and IPGE₁ groups had normal appearing lungs (Figure 3) or showed the presence of focal injury of minimal severity (Figure 4). Two animals in the PGE₁ group had evidence of moderate-severe focal tracheal or bronchial injury in the form of flattening of epithelium, loss of cilia, ulceration and polymorphonuclear infiltration. Moderate to severe multifocal/diffuse alveolar edema and focal polymorphonuclear infiltration was observed in one animal in the control group and two animals in the IPGE₁ group.

Figure 3.

Normal bronchus and adjacent alveoli in a piglet who received IPGE₁ for 24 hours

Figure 4.

Few neutrophils and red blood cells in the bronchus (arrow) and minimal alveolar edema (arrowhead) in a piglet who received IPGE₁ for 24 hours

In addition to the tracheal, bronchial and alveolar changes, atelectasis was evident in 2 piglets (one each in the control and IPGE₁ groups). Evidence of bile aspiration was present in three animals in the PGE₁ group and two animals in the control group.

4. Discussion

We have described for the first time a 24 hour model of anesthetized ventilated neonatal piglets to study the safety of continuous aerosol therapy. We used healthy piglets to ensure that changes observed during long-term inhalation were not confounded by underlying disease or by the experimental model of pulmonary hypertension.. Although the subjects of this study were healthy piglets, FiO₂ of 1.0 was used to simulate the treatment of neonates with NHRF who typically are ventilated with 100% oxygen. This would allow the evaluation of pulmonary toxicity related to IPGE₁ in the presence of high FiO₂ while undergoing positive pressure ventilation.

The neonatal pig was a satisfactory model for this study as the lung of the neonatal piglet demonstrates several similarities to the lung of the human newborn, and the functional changes occurring in the pulmonary circulation during the first two weeks of life follow a similar time course to those in the human neonate (35).

Existing animal studies for testing drug safety in the treatment of respiratory failure are limited by the fact that most often spontaneously breathing, relatively mature animals have been used, and if undergoing mechanical ventilation, the duration of exposure to the inhaled agent has been relatively short (8 hours) (31). Histomorphological changes in tracheal, bronchial and alveolar epithelial tissues after 8-h inhalation of PGI₂ or normal saline in 14 healthy ventilated lambs have been previously reported (28.5 to 48.5 kg in weight). There was no difference in light microscopy findings following inhalation of PGI₂ or normal saline. Histological abnormalities were seen in 57% of tracheal sections and included focal flattening of the epithelium, loss of cilia, slight inflammatory cell infiltration. Alveolar changes were seen in 12% of sections and included thickening of alveolar septal space and focal inflammatory cell infiltration. Similarly, mild acute sterile tracheitis was reported following inhalation of PGI₂ in four intubated large white landrace piglets (11 to 21 kg) (22). This was attributed to the alkaline glycine diluent used to prepare the PGI₂ solution (pH 10.5). However, PGE₁ solution prepared in normal saline has a lower pH (6.5) and therefore is less likely to be associated with pulmonary toxicity. Pulmonary toxicity following inhalation of PGE₁ has not previously been described.

We have previously reported the effective delivery of IPGE₁ at doses ranging from 50 to 300 ng/kg/min in a phase I–II study in term/near-term neonates with NHRF as assessed by plasma PGE₁ levels (29). In an in vitro study, we have demonstrated that the emitted dose of IPGE₁ following jet nebulization in a neonatal ventilator circuit was 32–40% (26). In the present study, we used a high dose of IPGE₁, 1200 ng/kg/min, corresponding to a total dose of ~3,500 μg of PGE₁ administered over 24 h. Compared to doses known to reduce pulmonary hypertension in patients (8–300 ng/kg/min), this represents a high dose (7–9, 20, 28, 36). A dose of 1200 ng/kg/min was chosen as this represents a dose four times the maximal dose that has been reported in humans and when given over a 24-hour period represents the cumulative dose that would be delivered over several days in the clinically used doses. Moreover, this dose is more likely to reveal evidence of toxicity especially when given continuously over 24 hours as the extent of adverse effects on the hemodynamic parameters and lung histomorphology are directly related to the dose delivered.

We chose to administer high dose IPGE₁ continuously for 24 hours in the anesthetized ventilated piglets as this would allow sufficient time for the manifestation of adverse effects on hemodynamic parameters and pulmonary histomorphology. Hubbard et al reported that initial histomorphological changes in the respiratory tract appeared as early as 15 to 60 min after exposure depending on the severity of toxic exposure (37). Changes included necrosis and sloughing of respiratory epithelium, loss of cilia, and surface erosions. Inflammatory response, manifested by the formation of pseudomembranes, was observed two hours after the onset of toxic inhalation and was associated with simultaneous appearance of neutrophils in the lamina propria, epithelium, and lumen of trachea and bronchi. The acute inflammatory cell response was maximal by 24 hours. Increased mucus production and metaplastic changes were evident by 12 hours. The extent of injury was related to dose of toxic inhalation. Proceeding from these observations, appearance or worsening of pathological changes after 24-h inhalation of high doses of PGE₁ is unlikely.

There was no evidence of adverse effects of high dose IPGE₁ when administered continuously for 24 hours on cardiorespiratory, hematological and biochemical parameters. The absence of significant changes in arterial pressure indicates that significant systemic vasodilation did not occur during prolonged high dose IPGE₁. This is in accordance with the first pass pulmonary metabolism and lack of significant active metabolites of PGE₁. Although, PGE₁ is known to have significant bronchodilator action, there is a concern that transient airway irritation may result in coughing, wheezing and occasionally bronchoconstriction (23, 38–40). In the current report, no evidence of coughing or wheezing associated with significant changes in breathing pattern, heart rate, blood pressure, anesthetic requirement, ventilator peak pressures, ventilator rate or PaCO₂ were recorded during PGE₁ inhalation, indicating that there was no significant bronchospasm as a result of airway irritation in the anesthetized subjects of this report. There is a concern that continuous administration of IPGE₁ dissolved in normal saline at a rate of 4 ml/hr may result in hypernatremia in newborn infants. In the current report, normal saline was the vehicle for the administration of aerosolized PGE₁ or placebo for both groups of animals. Hypernatremia was not observed in any animal either in the control or the IPGE₁ groups even after 24 hours of continuous aerosol delivery.

The decline in oxygenation during the 24 hour period in animals in both groups probably reflects oxygen toxicity. Prolonged exposure to hyperoxia in newborns has been reported to be toxic resulting in destruction of the alveolar-capillary barrier leading to pulmonary edema, impaired gas exchange, pulmonary hypertension and eventually death (41–43). During the first 24 to 72 hr after exposure to 100% oxygen, most animal species do not demonstrate significant light microscopic changes although biochemical changes including increased oxygen consumption of cells and increased production of oxygen free radicals has been described. The earliest morphologic changes seen in the lung in response to hyperoxic stress involve subtle changes in endothelial cell ultra-structure seen 48 hours after hyperoxic exposure. Structural remodeling of pulmonary arteries is seen after 7 days of exposure to hyperoxia.

Pathological changes on lung histology were minimal or focal even after 24 hours of high dose IPGE₁ in conjunction with positive pressure ventilation with FiO₂ of 1.0. There was no evidence of severe toxic injury in the form of hyaline membranes, diffuse/severe neutrophilic infiltration or ulceration in the lungs and trachea in either the control or IPGE₁ groups. Moderate to severe diffuse/mutifocal alveolar edema was observed in both the control and IPGE₁ groups suggesting that the alveolar alterations may be secondary to effects of mechanical ventilation with high FiO₂ (30, 44–46). The commonest abnormality observed in the IPGE₁ group was the presence of moderate-severe focal tracheal or bronchial epithelial flattening, loss of cilia and ulceration in two animals. The most significant lesions after toxic inhalation have been reported to affect the trachea with the most severe changes being observed in the tissue adjacent to the tip of the ETT (31, 37). This is because, following aerosol delivery, a large fraction of the aerosol accumulates in the conducting airways, particularly in the trachea and ETT (47). Bile aspiration, evident in five animals undergoing the experimental protocol, could also be responsible for some of the histological abnormalities observed. These findings suggest that high dose IPGE₁ is associated with minimal signs of pulmonary toxicity even after 24 hours of therapy. The possibility of severe pulmonary toxicity occurring after 24 hours is unlikely.

The most significant strength of our study is that we have successfully established a neonatal animal model undergoing assisted ventilation and anesthesia to study safety of continuous IPGE₁ therapy over 24 hours thus closely simulating intended clinical use. Although significant data exist regarding the safety of long term ivPGE₁ in neonates with heart disease, the possible toxic effects of IPGE₁, especially local pulmonary toxicity, following prolonged inhalation have not been previously investigated. We have demonstrated that the continuous delivery of high dose IPGE₁ over 24-hours in a ventilated newborn animal model is relatively safe without significant adverse effects on hemodynamic, respiratory, hematological, and biochemical parameters and pulmonary histology.

Despite the important findings described in this study, there are potential deficiencies. We have evaluated the direct pulmonary toxicity of IPGE₁ following continuous administration for only 24 hours in mechanically ventilated anesthetized neonatal pigs. It is likely that clinical use may require aerosol administration for a longer period of time. We have tried to compensate for this by administering a high dose of IPGE₁ (1200 ng/kg/min, four times the maximal dose reported in humans) representing the cumulative dose that would be delivered over several days in the clinically used doses. This dose is more likely to reveal evidence of toxicity as the extent of adverse effects on the hemodynamic parameters and lung histomorphology are directly related to the dose delivered. Additionally, based on pulmonary toxicology studies following toxic inhalation, appearance or worsening of pathological changes after 24-h inhalation of high doses of PGE₁ is unlikely (37).

5. Conclusion

In summary, in healthy piglets, 24-h inhalation of high dose PGE₁ was not associated with adverse cardiorespiratory effects, bronchial irritation, or hypernatremia and produced minimal signs of pulmonary toxicity. Prolonged inhalation of high dose PGE₁ therefore appears safe in newborn piglets and might be applied without serious harm to human patients. Randomized controlled clinical trials are needed to document efficacy and safety in humans and to establish the role of IPGE₁ as an additional selective pulmonary vasodilator in the treatment of PPHN associated with NHRF.

ABBREVIATIONS

ETT

endotracheal tube

NHRF

neonatal hypoxemic respiratory failure

PPHN

persistent pulmonary hypertension of the newborn

PGE₁

prostaglandin E₁

IPGE₁

inhaled PGE₁

ivPGE₁

intravenous PGE₁

INO

inhaled nitric oxide

FiO₂

fractional inspired oxygen concentration

NICU

neonatal intensive care unit