Noninvasive inhaled nitric oxide for persistent pulmonary hypertension of the newborn: A single center experience

1 Introduction

Persistent pulmonary hypertension of the newborn (PPHN) is a clinical condition caused by a disruption of the postnatal transition from fetal to neonatal circulation. The consequence of this maladaptation is a sustained pulmonary vascular resistance (PVR) rather than the reduction in PVR that normally occurs at birth. High PVR causes extrapulmonary right-to-left shunting of blood across the ductus arteriosus and/or foramen ovale which leads to hypoxemia, cyanosis, and poor perfusion [1].

PPHN is estimated to occur at the rate of two per 1,000 live births among full-term and post-term infants. Treatment requires prompt diagnosis and expert management to lower pulmonary pressures. Traditionally, the treatment arsenal has included intubation with invasive mechanical ventilation, high frequency oscillatory ventilation, inhaled nitric oxide (iNO), and extracorporeal membrane oxygenation (ECMO).

Nitric oxide is a potent, selective pulmonary vasodilating agent that has been proven to decrease PVR. In December 1999, the Food and Drug Administration recognized and approved the use of iNO for the treatment of near-term and term neonates with hypoxic respiratory failure with evidence of pulmonary hypertension. Research indicates that iNO decreases the need for ECMO without an increase in comorbidities. Based on the analysis by Golombek and Young, iNO has been shown to effectively improve oxygenation as evidenced by oxygenation index, a marker of respiratory and oxygen support [2].

In more recent years, neonatal databases, such as the Vermont Oxford Network (VON), have reported an increase in noninvasive ventilation in clinical practice neonatology. This is due to the concern for volutrauma and increasing the inflammatory cascade. The application of nasal continuous positive airway pressure (NCPAP) and high flow nasal cannula (HFNC) has increased, while conventional mechanical ventilation (CMV) has decreased [3].

Oxygen (O₂) toxicity has also been implicated as a potential source of mortality and morbidity. As early as 1997, studies with immature lambs proved that even a short exposure to mechanical ventilation and excessive O₂ exposure could be harmful [4]. A potential lung protective strategy for infants with PPHN may be to administer iNO in a noninvasive manner by utilizing NCPAP or nasal cannula (NC) with a lower fraction of inspired oxygen (FiO₂) exposure. Early noninvasive delivery of iNO may potentially decrease O₂ toxicity and volutrauma by reducing ventilation to perfusion (V/Q) mismatch, improving oxygenation, and reducing hyperoxia exposure, and the need for invasive ventilation.

Little research is available on noninvasive (NIV) iNO for the treatment of PPHN, although NIV methods have been successfully reported in literature as part of the weaning process [5]. The aim of this retrospective chart review is to describe the outcomes of infants diagnosed with PPHN in the early newborn period whose iNO therapeutic regimen was purely via NIV delivery methods.

2 Methods

3 Results

Of the 46 eligible infants in the database, 24 met inclusion criteria. The gestational age range was 35 0/7 to 41 0/7 weeks at birth, with a mean of 38 4/7 weeks. Birth weights averaged 3410 g, with a range of 2322 g to 5108 g. Seven babies had low APGAR scores (0–3) recorded at one minute, and two babies scored low at five minutes of life. Underlying pathology was as follows: eight (33%) meconium aspiration syndrome (MAS), 15 (63%) respiratory distress syndrome (RDS), and one (4%) pneumonia (Table 1).

Curosurf^® (Poractant alfa) was administered to eight babies (33%). Mean duration of iNO administration was 60 hours (Table 2). Three babies (12%) were intubated: one supported with conventional mechanical ventilation (CMV), two managed with endotracheal (ETT) continuous positive airway pressure (CPAP). Twenty-one babies (88%) had a strictly NIV course of iNO administration (Table 3).

Initially, all 24 babies were dependent upon supplemental O₂ with a mean O₂ exposure period of approximately six days (Table 4). All blood culture results were negative. The survival rate for the 24 patients was 100%.

4 Discussion

The practice of iNO utilization for PPHN at WPH was updated in early 2009 based on published information, such as that provided by the Neonatal Inhaled Nitric Oxide Study Group (NINOS) and previous iNO phase II studies [6, 7]. The approach at WPH changed to emphasize avoiding invasive ventilation, but maintaining optimal V/Q match, avoiding hyperventilation and alkalosis agents, and avoiding hyperoxemia and hyperoxia exposure. The primary practice became noninvasive treatment with iNO for babies greater than 34 weeks gestational age with an increased FiO₂ requirement and echocardiographic evidence of PPHN, rather than invasive ventilation.

Several studies are available on iNO administration for PPHN for intubated babies on ventilatory support, but few studies describe NIV iNO delivery methods as the primary treatment. A 2009 pilot trial (study population of four) found that iNO via an oxygen hood was feasible [8]. Significant drawbacks to the hood are the cumbersome equipment and interference with parental kangaroo care.

A single case report by Nair, Orie, and Lakshminrusimha describes the successful treatment of a full-term infant with a frontline NIV iNO approach similar to ours [9]. This retrospective study expands on their work to provide an evaluation of a larger population.

The 24 study babies had initial echocardiograms within the first day of life in the neonatal unit. All babies were receiving NIV O2 at ≥0.50 FiO₂ and eight babies received Curosurf^® prior to the first echocardiogram and diagnosis of PPHN. The surfactant therapy was delivered via ETT with the intent to rapidly extubate after administration. All study subjects were diagnosed with pulmonary hypertension as evidenced by right ventricular hypertension, interventricular septal configuration, and right-to-left bidirectional shunting at the ductus arteriosus and/or foramen ovale by a pediatric cardiologist. In addition, structural or congenital anomalies were ruled out.

Underlying pathology was defined by chest radiographs and descriptions of clinical presentation within the physician’s progress notes. Criteria for MAS were delivery history, i.e. meconium stained amniotic fluid and visualization of meconium in the oropharynx and radiographic evidence of bilateral pulmonary infiltrates. RDS was determined by clinical presentation of progressive hypoxemia, cyanosis, increased work of breathing, and retractions while the patient was on a high level of NIV O₂ support (≥0.50 FiO₂). Chest x-ray findings for the RDS group varied: ten babies had low lung volumes and five babies had adequate bilateral aeration with no sign of infiltrates. A chest x-ray report of haziness in addition to the clinical picture led to a differential diagnosis of pneumonia in one case (Table 1).

Neonatologists at WPH administer iNO on the recommendations of iNO Therapeutics, which were determined by the pivotal NINOS and CINRGI trials [10]. The infants were effectively treated with iNO at a maximum level of 20 ppm. The nitric oxide was delivered via the INOvent^® system which features continuous inline integrated monitoring of delivered nitric oxide and a comprehensive alarm system to ensure delivery accuracy [10]. The methemoglobin (MetHgb) levels were monitored before, during, and after iNO administration; MetHgb levels were 1.0% for all samples analyzed.

Babies were weaned according to the WPH NICU guideline as follows: wean iNO level from 20 ppm to 10 ppm when FiO₂ is 0.40 to 0.50 to maintain oxygen saturation (SaO₂) ≥0.95, wean iNO level to 5 ppm when FiO₂ is at 0.30 to maintain SaO₂ ≥0.95, then decrease iNO level in increments by 1 ppm to discontinuation as SaO₂ is maintained ≥0.95.

According to animal and human neonatal literature, invasive ventilation increases the inflammatory cascade within hours of initiation. This inflammatory process has been demonstrated in the lungs, but additionally affects other distant organs, such as the brain and kidneys [11]. Animal and human research affirms that high FiO₂ exposure attenuates the effects of iNO by various means, including an increase in phosphodiesterase enzyme activity and reactive oxygen species. In lamb PPHN models, FiO₂ exposure of 0.50 was sufficient to attenuate iNO response [12]. An important finding was that neonates had a low incidence of deterioration to invasive ventilation, as well as a low rate of escalating FiO₂ need once NIV iNO therapy was initiated. Two babies remained intubated for respiratory support with ETT CPAP after receiving surfactant, one baby for 12 hours and the other for four days. Only one baby had significant decline with hypercarbia that required intubation and CMV for a 48 hour period.

The rapid improvement of our neonates resulted in a short exposure to supplemental O₂ and a relatively short length of stay. Eighty-eight percent of the babies demonstrated a narrowing of the pre- and postductal SaO₂ gradient and tolerated progressive weaning from supplemental O₂ within the first four hours of administration of NIV iNO. Babies weaned from iNO with little need for continued O₂ therapy: 25% required <0.30 FiO₂ via nasal cannula with a flow rate less than 1 L/min, 75% on room air. Repeat echocardiograms were not routinely performed during and after iNO, but those babies who had repeat testing were found to have resolved PPHN in every case. Noninvasive ventilation also meant less exposure to negative oral stimuli (i.e. intubations, ETT securement, etc.) which seemed to enhance oral feeding progress, another factor that can impact length of stay [13]. All babies were nipple feeding (either at breast or bottle) at least a portion of the total ordered volume by day six of life.

Current neonatal studies suggest the requirement of supplemental O₂ therapy is a risk factor for prolonged hospital stay. Furthermore, resources and cost differences exist between babies who need supplemental O₂ and those babies who do not [14]. Only one baby required supplemental O₂ upon discharge.

The length of stay (LOS) was a key indicator of the benefits of this NIV approach (Table 5). The range for LOS was five days to 52 days with a mean of two weeks. The Dixon’s outlier test found that the 52 day stay was an outlier with a P value <0.001. The baby with the 52 day hospital stay weaned from NIV iNO in two days but oral pathology delayed discharge. Early or late-onset sepsis can also slow clinical progress, but neither was identified within the study group.

This retrospective review has some limitations. We could not clearly demonstrate that improvement was from NIV iNO alone. Babies were on supplemental O₂ therapy and some received surfactant prior to iNO administration. The lack of a control group for direct comparison makes the study difficult to draw significant conclusions. It is possible these babies would have remained extubated and discharged home in a similar time period without iNO exposure. The definition of PPHN was largely based on echocardiography, which may be another flaw. It is possible the degree of PPHN was overestimated. In addition, the neonates were not initially intubated, which may denote a less severe clinical acuity.

5 Conclusion

Noninvasive ventilation is a growing trend in neonatal respiratory support that may potentially decrease the inflammatory response associated with invasive ventilation. At WPH, NIV iNO was a well-tolerated, successful frontline treatment for infants with PPHN. The additional advantage of minimizing supplemental FiO₂ exposure may maximize the benefits of iNO therapy.

This research provides the basis for future prospective study of this intervention to support NIV iNO as a primary therapy to treat PPHN in near-term and term neonates with an intact respiratory drive. Further study should continue, specifically a randomized controlled clinical trial, which could provide the necessary evidence on clinical outcomes as well as cost effectiveness to guide best clinical practice.

Financial disclosure statement

D.P. Smith has no competing financial interests. J.A. Perez has received speaker honoraria from Ikaria, Inc., but did not receive any compensation for this study.

Other disclosure statement

Orlando Health, Inc. received a grant from Ikaria, Inc. used solely to support this research study.