Nebulized Furosemide in the Treatment of Bronchopulmonary Dysplasia in Preterm Infants

Abstract

Bronchopulmonary dysplasia (BPD) is a chronic pulmonary disease commonly seen in preterm infants who require supplemental oxygen and/or assisted mechanical ventilation. BPD, a major cause of morbidity and mortality among premature infants, occurs in 5,000 to 10,000 premature infants in the United States each year. Despite numerous medical advances, no single intervention will prevent or treat BPD; hence, premature infants have an increased risk for developing significant sequelae that affect both cognitive and motor function. This article provides a brief overview of BPD and reviews the available literature regarding the safe and effective use of nebulized furosemide in the treatment of this disorder.

INTRODUCTION

Advances in neonatal intensive care have led to remarkable improvements in the survival rates of premature infants over the past 20 years. Data from the National Institute of Child Health and Human Development (NICHD) have documented survival rates of 21% at 22 weeks gestational age, 30% at 23 weeks gestational age, 50% at 24 weeks gestational age, and 75% at 25 weeks gestational age during the period 1995–1996.¹ Over the next 5 years, survival rates for infants at 23 weeks gestational age increased from 30% to 60%, almost a 4% improvement in survival for each additional day in utero.²

Among premature infants, there has not been a substantial reduction in morbidity in low-, very-low-, or extremely low-birth-weight infants.^1–3 Bronchopulmonary dysplasia (BPD), a major cause of morbidity and mortality among premature infants, occurs in 5,000 to 10,000 premature babies in the United States each year.⁴ Infants weighing <700 grams have an 85% risk of developing BPD compared to only a 5% risk in infants weighing >1,500 grams at birth.⁵

Because of the gaps in knowledge about the prevention and management of BPD, premature infants are at increased risk for developing cerebral palsy, microcephaly, and neurodevelopmental delay affecting both cognitive and motor functions.^6–8 Despite numerous medical advances, such as antenatal steroids, surfactant replacement, nitric oxide administration, and gentler ventilation strategies, no single intervention has been shown to prevent or has been used to treat BPD. This article provides a brief overview of BPD and reviews the data available for the safety and efficacy of nebulized furosemide in the treatment of BPD.

DEFINITION AND INCIDENCE

Since 1967, the definition of BPD has been modified several times. Northway et al.⁹ first described the disorder in premature infants who developed respiratory distress syndrome and were placed on prolonged mechanical ventilation with high pressures and FiO₂.⁵ The definition was later refined to include abnormal findings on chest x-ray and the need for supplemental oxygen at 28 days of life to maintain arterial oxygen concentrations greater than 50 mm Hg.^10,11 Although the additional need for supplemental oxygen at 36 weeks postmenstrual age (PMA) was shown to be a better predictor of long-term respiratory outcomes,¹² changes in the clinical presentation of BPD precluded the application of this definition to the majority of patients.

In 2001, the NICHD developed new diagnostic criteria describing the severity of BPD based on an infant's supplemental oxygen requirements for ≥28 days at 36 weeks PMA (Table 1).^10,13 Specific radiographic findings were not included in the consensus definition. Although this definition has been widely accepted, it has been criticized because of the limitations of supplemental oxygen as a diagnostic criteria for BPD.¹⁴ In addition, radiographic changes have shown to be most predictive of long-term respiratory outcomes.^15,16

Table 1.

Diagnostic Criteria*

The use of a variety of diagnostic criteria for BPD and the evolution of “old” to “new” definitions of BPD has led to considerable variability in the reported incidence of the disease, ranging from 0% to 44%.^1,17 The differences in the use of supplemental oxygen and positive pressure and the level of neonatal care provided within the institution can also impact the incidence of BPD. Likewise, the incidence may differ according to birth weight, with the likelihood of BPD increasing as birth weight decreases. Approximately 600,000 very-low birth weight (VLBW) infants (∼1.5% of all newborns) are born in the United States each year.¹⁸ Although 85% of VLBW infants survive, 30% of these infants develop BPD.^13,19 Infants weighing <1,250 grams and infants born prior to 30 weeks gestation constitute 97% of the infants with BPD.^13,17

PATHOPHYSIOLOGY AND ETIOLOGY

There are five stages of development in the normal human lung: 1) embryonal, 2) pseudoglandular, 3) canalicular, 4) saccular, and 5) alveolar.²⁰ The embryonal stage occurs during the first 2 months of gestation, during which formation of the right and left mainstem bronchi occurs. At 8 to 16 weeks of gestation, the second stage (i.e., pseudoglandular) occurs. During this time, budding and branching of the airways continues, and formation of the small airways including the terminal bronchioles are complete. From 16 to 26 weeks of gestation (i.e., canalicular stage), the lung periphery and air–blood barrier are formed to allow for gas exchange. In the saccular stage, 26 to 36 weeks of gestation, expansion of air spaces occurs, and surfactant is detected. During the final stage, which occurs between 36 weeks of gestation and 2 years of age, the alveoli form from terminal endings of the alveolar sacculi. Over time, the alveoli increase in diameter.

Following premature birth, the final stages of normal lung development are interrupted, causing the infant to have reduced capacity for gas exchange due to decreased lung volumes and capillary surface areas. Impaired gas exchange leads to hypoxemia and prolonged oxygen requirements.^20,21

There are several proposed etiologies of BPD.^5,11,18,22 Premature infants, especially those <26 weeks gestational age, develop BPD due to lung immaturity. Surfactant deficiency causes respiratory distress syndrome (RDS), which also contributes to the development of BPD. Prolonged exposure to high oxygen concentrations and free oxygen radicals can cause tissue damage and impaired gas exchange. Data suggest that oxygen saturation concentrations of >95% and PaO₂ of >70 mm Hg are correlated with worse respiratory outcomes.¹¹ BPD may also be attributed to volutrauma caused by excess tidal volume or barotrauma resulting from positive pressure ventilation. Compliant alveoli can hyperinflate, whereas those with increased surface tension collapse. The combined effects of oxygen toxicity and barotrauma/volutrauma on immature lungs can worsen lung damage.^5,18

Inflammation is another cause of BPD and is activated by oxygen toxicity, barotrauma/volutrauma, or other injury (e.g., infection).^5,22 The inflammatory process enhances lung injury and causes capillary leakage. Finally, nutrients deficient in vitamin A or trace elements (e.g., copper, zinc, selenium) can also play a role in the development of BPD as these serve an important role in antioxidant function, protect against infection, and help in lung repair.^5,22

Although BPD has been associated with lung inflammation and fibrosis due to oxygen toxicity and/or barotrauma, advances in neonatal care have improved survival rates among infants born at earlier gestational ages. As a result, the classic or old BPD seen in both preterm and term infants has been replaced by a milder form of BPD, which is frequently seen in extremely low-birth-weight infants. The changes seen in the new BPD are primarily due to an arrest in lung development rather than mechanical injury (Table 2).^23,24

Table 2.

“Old” versus “New” BPD Criteria*

MEDICAL MANAGEMENT

Management of infants with BPD includes minimizing the duration of mechanical ventilation and the exposure to high concentrations of oxygen while maintaining adequate gas exchange. Long-term management often requires a combination of fluid restriction, drug therapy (i.e., bronchodilators, pulmonary vasodilators, diuretics, steroids), and nutritional intervention including supplementation with vitamin A.⁶

Although inhaled bronchodilators, such as beta-agonists or anticholinergic agents, are used to decrease airway resistance in patients with episodes of acute exacerbation of airway obstruction; they do not improve long-term outcomes.⁶ Likewise, pulmonary vasodilators (e.g., nitric oxide) may reduce pulmonary vascular resistance and inflammation in infants with severe pulmonary hypertension, but there is no evidence that these agents improve long-term outcomes associated with BPD.⁶ Corticosteroids decrease inflammation in the injured lung, enhance production of surfactant and antioxidant enzymes, decrease pulmonary and bronchial edema and fibrosis, and improve vitamin A concentrations.⁶ Despite these positive effects, the prolonged use of steroids is associated with poor neurological outcome; therefore, they are recommended for use after the first 2 weeks of life or in severe cases of pulmonary damage.^6,25

Vitamin A administered intramuscularly three times a week for 4 weeks has been shown to promote lung growth and repair as well as to decrease the incidence of sepsis, chronic lung disease (CLD), and death.^6,26 However, the need for repeated intramuscular injections to VLBW infants with insufficient muscle mass limits use of vitamin A in neonatal intensive care units in the United States. In addition to vitamin A supplementation, aggressive caloric intake helps reduce infection rates, decrease oxygen toxicity secondary to decreased diminished muscle strength and prolonged mechanical ventilation, and promotes alveoli development.⁶

Diuretics are commonly used in the management of BPD. They reduce interstitial lung water, lower pulmonary vascular resistance, improve gas exchange, and reduce oxygen requirements.^5,6,27 When pulmonary edema persists despite fluid restriction, loop diuretics are frequently incorporated in the management plan, as these agents produce rapid improvements in lung function.^5,6,27 However, chronic intravenous or oral administration of furosemide is associated with significant side effects including electrolyte imbalances, metabolic alkalosis, hearing loss, and nephrocalcinosis. Furthermore, routine or sustained use of systemic loop diuretics in infants with CLD cannot be recommended based on the lack of long-term outcome data. Because of these concerns, studies have looked at alternate-day dosing of furosemide as well as administering furosemide via nebulization.^28–33

MECHANISM OF ACTION OF NEBULIZED FUROSEMIDE

Nebulized furosemide was shown to inhibit bronchoconstrictive effects of exercise and antigen challenge in adult patients with asthma. It has also been shown to protect against early and late allergen-induced asthmatic reactions.^34,35 The efficacy of nebulized furosemide is attributed to its local effect on the lung, rather than diuresis associated with its renal effect. This improvement in pulmonary function in adult patients with asthma has led to its use in preterm infants with BPD.

The exact mechanism by which nebulized furosemide improves lung function is unclear. Proposed mechanisms include 1) inhibition of the Na⁺/K⁺/Cl⁻ cotransport system in airway epithelium, thus modifying airway osmolarity^36,37; 2) reduction in intracellular Na and Ca, thereby inducing airway smooth muscle relaxation^36,37; 3) inhibition of the release of LTC4, histamine, and neutrophil chemotactic factors by the inflammatory cells^38,39; and 4) increase in the airway epithelium-derived prostaglandin E₂ and pulmonary endothelium-derived prostaglandin I₂, which has been shown to be vasoprotective.⁴⁰

STUDIES

Efficacy

Five studies have assessed the efficacy of nebulized furosemide in a total of 57 premature infants with BPD (Table 2).^29–33 Enrollment size has ranged from as few as 8 to as many as 19 infants. Doses of 0.1,²⁹ 0.25,²⁹ 0.5,²⁹ 1,^29–32 and 2 mg/kg³³ have been evaluated using a variety of study designs ranging from open-label to randomized, double-blind, placebo-controlled.

In an open-label dose comparison study, Rastogi and colleagues²⁹ evaluated whether nebulized furosemide given to infants with BPD would improve respiratory function without causing systemic complications that have been described with long-term systemic furosemide therapy. Infants older than 14 days postnatal age (PNA) with a history of RDS and persistent respiratory failure that required mechanical ventilation, with dependence on oxygen and radiographic evidence of pulmonary parenchymal disease, were included. Patients who had received medications other than vitamins for at least 72 hours prior to enrollment and those with evidence of congenital malformations (including heart defect) or infection were excluded.

Eight premature infants with a mean gestational age of 27.4 ± 1.6 weeks (mean weight, 814.5 ± 296.5 grams) were evaluated. Mean PNA and weight at the time of study were 33.13 ± 13 days and 1,013 ± 331 grams, respectively. Each patient received 0.1, 0.25, 0.5, and 1 mg/kg nebulized furosemide on 4 consecutive days at 24-hour intervals. The furosemide was diluted with 0.9% sodium chloride to a final volume of 2 mL. Dose and order of the dose administered to each patient was determined in a randomized manner using a table of random numbers. Total pulmonary compliance, pulmonary resistance, and tidal volume values were measured at baseline, 0.5, 1, 2, and 4 hours after furosemide treatment. Serum and urine calcium, sodium, potassium, chloride, and creatinine concentrations were measured daily. Body weight, fluid intake, and urine output were recorded daily and 24 hours after the final dose of furosemide.

There was no improvement in pulmonary mechanics observed with the 0.1, 0.25, or 0.5 mg/kg dose; however, the 1 mg/kg dose improved lung compliance (51% at 2 hours, p<0.001), decreased pulmonary resistance (28% at 1 hour, p<0.05), and increased tidal volume (43% at 1 hour, p<0.01). Within 30 minutes of nebulization, 7 patients had an increase in pulmonary compliance that persisted for up to 4 hours in 6 patients. Seven patients had lower pulmonary resistance at 4 hours than at baseline. Tidal volume increased within 30 minutes and remained above baseline at 4 hours in 6 of 8 patients. No increases in urine output (UOP), calcium excretion, fractional excretion of sodium, and creatinine were noted with any dose, and no side effects were attributed to nebulized furosemide.

In a randomized, crossover study, Prabhu et al.³⁰ compared the effectiveness of single doses of nebulized and intravenous furosemide. All infants were mechanically ventilated, had evolving CLD, and had been oxygenated for more than 14 days since birth. Infants did not have any major congenital anomalies and were not given other diuretics or steroids. Although bronchodilators were withheld 4 hours prior to, during, and 2 hours after the study treatment, theophylline therapy was continued.

Nineteen preterm infants with a mean gestational age of 26.2 ± 2.1 weeks and a mean weight of 800 ± 170 grams were evaluated. Mean PNA and weight at the time of enrollment were 27 ± 11 days and 910 ± 180 grams, respectively. Infants were randomized to receive 1 mg/kg intravenous or nebulized furosemide 24 hours apart on 2 separate days. Furosemide was diluted with 0.9% sodium chloride to a final volume of 2 mL and administered via nebulization at a flow rate of 2 L/min over 5 to 10 minutes, using the Miniheart nebulizer (Westmed Inc, Tucson, AZ).

Pulmonary compliance, pulmonary resistance, and tidal volume were measured 10 minutes before and 30, 60, and 120 minutes after treatment. Urine was collected for 6 hours before and after treatment. There was a 34% increase in pulmonary compliance (p=0.002), a 31% increase in tidal volume (p=0.001), and no change in pulmonary resistance 2 hours after nebulized treatment. No pulmonary changes were noted with intravenous furosemide. Conversely, nebulized furosemide had no effect on urine output. The authors concluded that a single dose of nebulized furosemide improves pulmonary function in premature infants with evolving CLD, without producing any change in fluid and electrolyte balance.

In a randomized, double-blind, placebo-controlled study, Kugelman et al.³¹ evaluated the safety and efficacy of a single dose of nebulized furosemide in mechanically ventilated infants with severe BPD. Infants were ≤32 weeks gestational age and had birth weights between 500 and 2,500 grams. The infants had persistent respiratory failure that required mechanical ventilation and had radiographic changes suggestive of BPD; however, their conditions were stable, and they had no evidence of cardiac disease or sepsis.

Nine preterm infants with a mean gestational age of 29 ± 1 weeks (mean weight, 1.1 ± 0.1 kg) and a mean PNA of 47 ± 6 days (mean weight, 1.8 ± 0.2 kg) were evaluated. Infants were randomized to receive 1 mg/kg nebulized furosemide (diluted in 2 mL of 0.9% sodium chloride) or placebo (2 mL of 0.9% sodium chloride) and were followed by crossover to alternate therapy. Nebulized furosemide and placebo were delivered through the inspiratory limb of the ventilator by using a neonatal nebulizer placed 10 to 12 cm from the endotracheal tube. The side flow rate was 5 to 6 L/min over 5 to 10 minutes. All bronchodilators and diuretics were discontinued 6 hours prior to study.

Pulmonary compliance, pulmonary resistance, and tidal volume were measured at baseline and at 1 and 2 hours after treatment. Serum electrolytes, blood urea nitrogen (BUN), serum creatinine (SCr), and total calcium concentrations were measured on the day of the study. Neither pulmonary compliance, pulmonary resistance, nor tidal volume were improved at 1 and 2 hours posttreatment with nebulized furosemide or placebo. Gas exchange, as assessed by pO₂ and pCO₂, and UOP, blood pressure (BP), heart rate (HR), and respiratory rate (RR) did not change significantly. The authors concluded that while 1 mg/kg nebulized furosemide was safe, it did not improve pulmonary mechanics in ventilated infants with severe BPD.

Prabhu et al.³² compared the duration and effects of 1 mg/kg to those of 2 mg/kg nebulized furosemide on pulmonary mechanics. Infants were ventilated and oxygenated for 14 days since birth and had no major congenital anomalies. Although bronchodilator treatments were held for 4 hours prior to and during study treatment, theophylline therapy was continued. No infants received other diuretics or steroids.

Thirteen preterm infants with a mean gestational age of 25.4 ± 1.5 weeks (mean weight, 713 ± 132 grams) and a mean PNA of 24 ± 9.5 days (mean weight, 862 ± 258 grams) were evaluated. Infants were randomized to receive nebulized furosemide, 1 mg/kg or 2 mg/kg, followed by crossover to alternate therapy 24 hours apart. Furosemide was diluted with 0.9% sodium chloride to a final volume of 2 mL and administered at a flow rate of 2 L/min over 5 to 10 minutes, using the Miniheart nebulizer.

Pulmonary compliance, pulmonary resistance, and tidal volume were measured 10 minutes before and at 2, 4, and 6 hours after nebulization. UOP and electrolytes were measured for 6 hours before and after nebulization. Nebulized furosemide dosages of 1 mg/kg and 2 mg/kg resulted in significant improvement in tidal volume and lung compliance (p=0.005 and p=0.003, respectively) values compared to those at baseline; however, there was no significant difference in the magnitude of response between doses. Although the maximum effects seen from both doses occurred at 4 hours, the duration of effect lasted up to 6 hours. Regardless of dose, nebulized furosemide had no effect on pulmonary resistance. Finally, neither of the two treatments was associated with increased UOP or electrolyte losses. The authors concluded that increasing the dose from 1 mg/kg to 2 mg/kg showed no further improvement in pulmonary function. Based on the duration of effect, the authors recommended that nebulized furosemide be administered every 6 hours or longer.

Ohki and colleagues³³ evaluated whether nebulized furosemide improved pulmonary function without excessive diuresis by using a double-blind, placebo-controlled, crossover study design. Infants with radiographic evidence of CLD who required mechanical ventilation for more than 14 days and weighed <1,500 grams were evaluated. Diuretics were held for at least 48 hours prior to study.

Eight preterm infants with a mean gestational age of 26.5 ± 1.9 weeks (mean weight, 798 ± 225 grams) and a mean PNA of 32 ± 15 days (mean weight, 807 ± 236 grams) were evaluated. Infants were randomized to receive a nebulized furosemide dose of 2 mg/kg (diluted in 3 mL of 0.9% sodium chloride) or placebo (3 mL of 0.9% sodium chloride) via an ultrasonic nebulizer (Devilbiss 500; Devilbiss Inc, Somerset, NJ). Patients were crossed over to receive the alternate therapy 48 hours apart.

Pulmonary compliance, pulmonary resistance, and tidal volume were measured at baseline and at 1 and 2 hours after nebulization. Urine output was followed for 24 hours, and serum electrolytes were measured before and 48 hours after each inhalation. Nebulized furosemide significantly increased pulmonary compliance and tidal volume values at both time intervals compared to those at baseline. There was no significant change in pulmonary resistance in either group. There were no differences in UOP and serum sodium and chloride concentrations between the two groups. No side effects were reported. The authors concluded that nebulized furosemide improves pulmonary function within 1 hour of treatment in infants with CLD and the effects can be seen up to 2 hours postnebulization.

Toxicity

Nebulized furosemide at 0.1, 0.25, 0.5, 1, and 2 mg/kg/dose had no effect on UOP or electrolyte losses.^29–33 In addition, doses up to 1 mg/kg were not associated with changes in creatinine clearance or renal calcifications.²⁹ Finally, no systemic effects (i.e., heart rate, respiratory rate, blood pressure, and body temperature) were noted following administration of nebulized furosemide.³¹

STUDY LIMITATIONS AND FUTURE RESEARCH

All of the studies suffered from small sample size,^29–33 with the largest study group consisting of only 19 preterm infants.³⁰ All 5 studies reported that either not all ^29,31 or none^30,32,33 of the patients reached 36 weeks postmenstrual age; hence, they did not meet the new NICHD diagnostic criteria for BPD. Study designs varied among the literature reviewed (Table 3). Only 3 studies used a control group,^30,31,33 2 of which used placebo^31,33 and 1 of which used intravenous furosemide.³⁰ Three of the studies randomized patients,^30–32 and 4 studies used a crossover design.^30–33

Table 3.

Study designs varied among the literature reviewed

All of the studies evaluated a single-dose effect and did not support the long-term use of nebulized furosemide. Although 2 studies involved dose comparisons and evaluated either 2³² or 4 different doses,²⁹ the washout period between doses may have been insufficient. Because the half-life of loop diuretics is longer in premature infants, ≥24 hours, a prolonged washout period of >48 hours should have been used in order to eliminate additive effects from multiple doses.^41,42

In addition, the duration of observation following a nebulized dose varied from 2 to 6 hours,^29–33 with 3 studies using only a 2-hour period.^30,31,33 A short observation period and failure to use repeated doses may not provide a real assessment of the potential for negative effects on urine output, electrolytes, or pulmonary mechanics.

Several of the studies allowed the use of bronchodilators and diuretics prior to the start of study,^30–32 and 2 studies allowed infants to continue theophylline therapy during the study period.^30,32 Because these treatments produce bronchodilation and have a weak diuretic effect, use of these agents may have influenced the study results. Only 1 study prohibited the use of diuretics for at least 48 hours prior to enrollment in the study.³³

The only study to report no effect from 1 mg/kg nebulized furosemide on pulmonary mechanics noted that the lack of response might be attributed to technical problems in drug delivery.³¹ To achieve a high local concentration of furosemide in the lung, the drug was delivered through the inspiratory limb of the ventilator. Study discussion noted that deposition of the drug in the ventilator tubing has been reported to result in poor and variable delivery of medication to the lungs. Variations in nebulizer efficacy (i.e., ultrasonic vs. jet nebulizer) could have altered medication deposition in the lungs, anywhere from 1% to 15%.³¹

In view of the lack of data regarding the use of nebulized furosemide, future randomized, double-blinded, placebo-controlled trials should be designed using well-defined inclusion/exclusion criteria, an increased sample size, multiple doses, and longer observational periods. Studies should focus on determining optimal delivery, dosage, frequency of inhalations, duration, and safety and efficacy of chronic administration. Such studies should also evaluate whether nebulized furosemide in combination with surfactants, corticosteroids, methylxanthines, and/or systemic furosemide will improve pulmonary compliance in patients with BPD. Finally, studies should focus on evaluating whether nebulized furosemide has beneficial effects on ventilator support, length of stay, survival, and long-term outcomes.

CONCLUSION

Despite numerous medical advances, no single intervention will prevent or treat BPD; hence, premature infants have an increased risk for developing significant sequelae that affect both cognitive and motor functions. The use of diuretics provides short-term improvement in lung mechanics, but they do not decrease the need for ventilatory support, length of hospital stay, or survival.⁴³ Nebulized furosemide doses of 1 or 2 mg/kg may be effective in improving pulmonary mechanics in preterm infants with BPD, but recommendations on frequency and duration of therapy are limited due to the single-dose studies available in this patient population.

Abbreviations

BP

blood pressure

BPD

bronchopulmonary dysplasia

CLD

chronic lung disease

HR

heart rate

GA

gestational age

NICHD

National Institute of Child Health and Human Development

PMA

postmenstrual age

PNA

postnatal age

NICU

Neonatal Intensive Care Unit

RDS

respiratory distress syndrome

RR

respiratory rate

UOP

urine output

VLBW

very-low birth weight