Clinical Considerations for the Pharmacologic Management of Patent Ductus Arteriosus with Cyclooxygenase Inhibitors in Premature Infants

INTRODUCTION

In term infants, the ductus arteriosus (DA) typically closes in the first day of life, but in preterm infants, closure is often either delayed or does not occur at all.3 Persistence of the DA is inversely related to gestational age. Spontaneous patent ductus arteriosus (PDA) closure occurs in up to 40% of prematurely born infants, but up to 70% of those born weighing less than 1,000 g require some form of intervention due to significant hemodynamic symptomatology.4 The persistence of the PDA puts the infant at higher risk for undesirable pulmonary, hemodynamic, gastrointestinal, and renal effects including intraventricular hemorrhage (IVH), pulmonary edema, congestive heart failure, cerebral vascular accidents, necrotizing enterocolitis (NEC), poor weight gain, bronchopulmonary dysplasia (BPD), and death.25

Available treatment options include conservative medical management, pharmacologic therapy, and surgical ligation. There is a high failure rate with conservative medical management, which involves watchful waiting, fluid restriction, and possibly mechanical ventilation, especially in very low birth weight infants.4 Surgical ligation in infants of less than 28 weeks gestation requires a thoracotomy and is associated with significant morbidities including pneumothorax, chylothorax, infection, laryngeal nerve paralysis, respiratory compromise, blood pressure fluctuation, intracranial hemorrhage, bronchopulmonary dysplasia, retinopathy of prematurity, neurosensory impairment, and even death.46–8 Pharmacologic therapy involves the use of cyclooxygenase (COX) inhibitors, which have been shown to be safe and effective in 70%–80% of treated infants.12

Currently, there are 2 equally effective intravenous (IV) COX inhibitors available: IV indomethacin (Indocin; Ovation Pharmaceuticals, Deerfield, IL) and IV ibuprofen lysine (NeoProfen; Ovation Pharmaceuticals, Deerfield, IL).1 Oral ibuprofen has been shown to be effective in closing PDAs in 3 studies; however, these were small, nonrandomized studies, and definite conclusions of efficacy and safety cannot be made from these trials.9–11 Prior to 2006, when IV ibuprofen lysine was approved by the US Food and Drug Administration (FDA), IV indomethacin had been the standard of pharmacologic care for closure of PDA; however, adverse effects related to vasoconstriction of renal, mesenteric and cerebral blood flow caused by IV indomethacin left clinicians concerned about its use. Recent research has shown IV ibuprofen lysine to be associated with less of these adverse vasoconstrictive effects, making it a desirable alternative to indomethacin.

This review examines the clinical considerations surrounding the approved IV COX inhibitors, indomethacin and ibuprofen lysine. In particular, this article focuses on the safety and efficacy of these COX inhibitors with emphasis on the important differences demonstrated by clinical trials, including a comparison of these agents' effects on blood flow to the cerebral, renal and mesenteric arteries, and on renal function. Further, key clinical issues discussed in this paper include comparative effects on pulmonary outcomes, protein binding, platelet aggregation, gastrointestinal bleeding, and mortality.

CLINICAL CONSIDERATIONS IN THE MEDICAL MANAGEMENT OF PDA

Premature infants with respiratory distress and echocardiographically confirmed PDA warrant treatment with COX inhibitors to prevent the hemodynamic instability and undesirable sequelae associated with persistent PDA. IV ibuprofen lysine and IV indomethacin are chemically different and inhibit COX-1 and COX-2 isoforms to a different degree. By inhibiting COX-1 and COX-2 isoforms, prostaglandins, which support the patency of the DA, are inhibited.812 Indomethacin has stronger COX-1 inhibition, which produces more severe gastrointestinal, cerebral, and renal side effects than COX-2 inhibition.12 IV ibuprofen lysine inhibits both COX-1 and COX-2 enzymes but has less COX-1 inhibition than indomethacin, which results in less severe vasoconstrictive side effects on these vital organs.13–17 IV ibuprofen lysine and indomethacin have been shown to be equally efficacious in the closure of PDA in preterm infants12818–25 and significantly more effective than placebo (P < .000126 and P = .0018,2728 respectively).

COMPARATIVE VASOCONSTRICTIVE EFFECTS

Clinical trials have shown that adverse vasoconstrictive effects to renal, cerebral and mesenteric vessels are less pronounced with the use of IV ibuprofen lysine than with IV indomethacin. Ibuprofen closes the DA in animals without affecting basal cerebral blood flow (CBF) or renal or intestinal hemodynamics during positive pressure ventilation.29 Supporting this evidence, studies of premature infants have shown that IV ibuprofen lysine did not significantly reduce CBF.3031 In contrast, indomethacin induced significant reductions in CBF velocity in clinical trials.3233 Comparative drug studies of IV ibuprofen versus IV indomethacin showed that indomethacin, but not ibuprofen, induced significant reductions in CBF (Figure 1).34–36

View Figure

• Download as Powerpoint
• Download Figure

In addition, trials using indomethacin in preterm infants showed significant reduction in mesenteric 3738 and renal39 blood flow velocity measured by Doppler ultrasound.17 Using indomethacin during positive pressure ventilation compromised regional hemodynamic blood flow to the kidneys (decreased 37%–57%), to the ileum (decreased 58%–74%), and to the colon (decreased 53%–71%) in a study by Malcolm.13 A drug comparison trial of indomethacin vs. ibuprofen by Pezzati et al. confirmed that mesenteric and renal perfusion were more compromised with the use of indomethacin.17

Renal Effects

The greatest and most significant differences observed in clinical trials in patient tolerance of the 2 available COX inhibitors were in renal effects. When comparing renal and mesenteric blood flow using IV ibuprofen vs. IV indomethacin, Pezzati et al.17 found significantly less decrease in renal blood flow (P < .0001) and mesenteric blood flow (P < .0001) with IV ibuprofen. The differential effects upon renal blood flow with IV indomethacin have been observed clinically. At the recommended dose of indomethacin, oliguria is a frequent side effect. In clinical settings, this often requires interruption of the treatment regimen. To counter this side effect, furosemide and/or dopamine are often used; however, use of furosemide increases circulating prostaglandin levels and may counteract treatment effectiveness.51

Decrease in urine output as a side effect of therapy was significantly less with IV ibuprofen lysine compared to indomethacin in a pooled analysis of 4 studies (P < .00001).121841 Infants treated with IV ibuprofen lysine had higher creatinine clearance, lower serum creatinine values (P = .03),18 and lower blood urea nitrogen (BUN) values than infants treated with indomethacin.19 Oliguria was significantly higher in the indomethacin group (P = .017).18 Pezzati et al.17 reported that urine output (P < .05) and serum creatinine (P < .02) were significantly affected by indomethacin; urine output took 7 days to return to pretreatment levels. Infants treated with indomethacin had a significant decrease in urine output on all 3 days of therapy (P < .05), while infants treated with IV ibuprofen lysine had a decrease only on day 1.17 Patel et al.35 found that infants developed oliguria and elevated serum creatinine after the first dose of indomethacin, while none of the ibuprofen infants in his study experienced adverse renal events.

Van Overmeire et al.,1 in a comparison trial of IV ibuprofen lysine and IV indomethacin for closure of PDA (n = 148), showed significantly less oliguria with ibuprofen lysine (P = .001 over all treatment days) (Figure 2) and significantly lower serum creatinine concentrations in the indomethacin group (P = .04 overall) (Figure 3). A meta-analysis of 5 IV ibuprofen lysine vs. indomethacin trials (n = 443) confirmed these findings, revealing a smaller increase in serum creatinine (P < .001) and a significantly smaller decrease in urine output (P < .001) with IV ibuprofen lysine.25 In multi-center trials of IV ibuprofen lysine, there was a small decrease in urinary output on days 2–6 of life in the ibuprofen groups with a compensatory increase on day 9; however, there were no statistically significant differences in total renal-related events between placebo and IV ibuprofen lysine.20 These data underscore that IV ibuprofen lysine does not cause a reduction in renal blood flow velocity11730 as described under vasoconstrictive effects and is therefore associated with significantly less impairment of renal function in these tiny babies.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

PULMONARY EFFECTS

Due to right-to-left cardiac shunting with a PDA, pulmonary function is compromised. The lower the gestational age of a prematurely born infant, the greater the occurrence of PDA, respiratory distress syndrome, and the greater the risk of resulting chronic lung disease (CLD) or BPD. CLD in preterm infants has a multi-factorial etiology. Genetics, oxygen exposure, mechanical ventilation, degree of immaturity, infection, and PDA are all compounding factors in the development of CLD. Perinatal steroids and tocolytics, coupled with postnatal surfactant and an increased use of noninvasive modes of respiratory support, have lowered the incidence and severity of CLD but have not eliminated it.

The definition of BPD or CLD differs among authors and clinicians. The definition of BPD/CLD as “the need for supplemental oxygen at 28 days of life” has been modified by clinicians who emphasize that an infant born at 24–25 weeks gestation is still 11–12 weeks premature at 28 days of life. Shennan et al.63 found that the definition “the need for supplemental oxygen at 36 weeks postmenstrual age (PMA)” is a more accurate marker of adverse pulmonary outcome.

EFFECTS ON HEMOREGULATION

Both indomethacin and ibuprofen are COX inhibitors that block the conversion of arachidonic acid to prostaglandins. Their side effects include inhibition of platelet aggregation. In adult studies, IV ibuprofen lysine has a minimal effect on platelet aggregation that can cause a modest increase in bleeding time not exceeding the upper limit of normal, and these effects are short-lived and dissipate between 6 and 24 hours after dosing.67 Indomethacin's effect on platelet aggregation persists longer, into the second day.68 The pharmacodynamics and pharmacokinetics differ between adult and neonatal groups. Due to renal function immaturity, premature infants have substantially slower elimination and prolonged half-life of these medications. Drug clearance increases and half-life decreases with advancing postnatal age, but in infants the half-life is generally more than 10 times greater than in adults.6669 Concerns regarding IVH due to the inhibition of platelet aggregation have led investigators to include IVH as an endpoint in clinical trials of both IV ibuprofen lysine and indomethacin. The relative risk for developing IVH is similar for both indomethacin- and ibuprofen-treated infants, and there may be a risk of increasing preexisting hemorrhages with the use of COX inhibitors. However, comparison trails of the 2 drugs showed no significant differences in the occurrence of initial IVH or the extension of IVH.12181941 Even so, if a preexisting IVH is present, many practitioners will wait 24 hours before initiating treatment with COX inhibitors to prevent the risk of exacerbation of the IVH.

EFFECTS ON PROTEIN BINDING

COX inhibitors are highly protein-bound (95%–98%) drugs and can potentially displace bilirubin from its binding site on albumin, increasing free bilirubin levels. Hyperbilirubinemia is a common finding in premature infants, who are at increased risk for kernicterus. Unbound bilirubin that crosses the blood brain barrier is the best indicator of risk for kernicterus, but risk thresholds for adverse neurologic outcomes currently do not exist.70 The degree of protein binding with COX inhibitors is concentration-dependent. At a bilirubin-to-albumin ratio of greater than or equal to 1 and a bilirubin concentration of greater than or equal to 10 mg/dL, potential displacement of serum bilirubin could occur with different serum concentrations of ibuprofen.71 Studies in premature infants indicate that bilirubin can be displaced at an ibuprofen concentration of 100 mg/L, but not at 50 mg/L or below.7273 In the intended population, with the approved dose and infusion time of IV ibuprofen lysine, plasma levels of ibuprofen rarely exceed 50 mg/L.66

In clinical trials of IV ibuprofen lysine vs. placebo or no treatment, there were no clinically important differences in elevated bilirubin levels. In a pooled analysis of major studies, neonatal hyperbilirubinemia was reported in a lower percentage of infants receiving IV ibuprofen lysine than placebo or no treatment. A lower percentage of infants in the indomethacin comparator group experienced hyperbilirubinemia compared with infants receiving IV ibuprofen lysine, placebo or no treatment.1202666 In 2 comparison studies by Van Overmeire, the incidence rates for neonatal hyperbilirubinemia were similar for infants treated with IV ibuprofen lysine or indomethacin1 and infants treated with IV ibuprofen lysine or placebo.26 A clinical trial of 15 preterm newborns tested for unbound bilirubin before and after IV ibuprofen administration at the recommended dosage showed no significant change in free or unbound bilirubin before or after ibuprofen infusion (1.07 μg/dL vs. 1.0 μg/dL).74 These observations suggest that, at the doses recommended, IV ibuprofen lysine does not produce significant bilirubin-drug-albumin interactions.21

OTHER EFFECTS

Comparative clinical trials and meta-analyses of these studies have demonstrated that the 2 available COX inhibitors, IV ibuprofen lysine and IV indomethacin, are both safe and equally effective for PDA closure in preterm infants. Of the other multiple variables studied, significant differences were seen in maintenance of cerebral, mesenteric and renal blood flow and in less impairment of urine output and less elevation of creatinine levels favoring IV ibuprofen lysine.

No statistically significant differences in other adverse outcomes were found in any of the individual comparison trials or placebo-controlled trials or in meta-analyses.23 The incidence of PDA reopening or the need for surgical PDA ligation were not statistically different in meta-analyses of IV ibuprofen lysine vs. placebo trials 27 or in drug comparison trials.25 Neonatal mortality at less than 28 days of age,23242750 at 36 weeks PMA,27 or during the infants' initial hospital stay 24–2650 showed no statistically significant differences between groups. Days on mechanical ventilation were similar in placebo-controlled 2850 and comparison trials.25 Gastrointestinal bleeding, defined as either occult blood in the stool, gastric bleeding, or localized bowel perforations, showed no statistically significant difference between trial groups using either COX inhibitor25 or in trials comparing IV ibuprofen lysine with placebo.27 Occurrence of retinopathy of prematurity50 or sepsis24 was not significantly different in those trials reporting these variables or in 2 meta-analyses.2527 Time to reach full enteral feeding and length of hospital stay were not statistically different.50 Outcomes after initial hospital discharge, including neurologic outcomes, have not been reported and require further study.

CONCLUSIONS

Persistent PDA in premature infants is associated with serious complications and adverse outcomes. Close medical monitoring and informed decision-making are essential in assuring the best treatment plan for optimal patient outcomes. Conservative medical treatment fails in two-thirds of infants delivered before 28 weeks gestation with birth weights less than 1,750 g.28 Surgical ligation is associated with significant and serious short- and long-term morbidities.46–8 When conservative measures fail to close the PDA, pharmacologic treatment with a COX inhibitor, IV Ibuprofen lysine or IV indomethacin, is recommended. Both agents are equally effective, closing the DA in 70%–80% of patients. Among the ibuprofen agents, only ibuprofen lysine is approved for use in the US.

Serious concerns with the use of these agents regarding renal compromise, PPHN, NEC, gastric perforation, IVH, CLD, hyperbilirubinemia, and coagulopathy have prompted multiple clinical trials. These studies have shown IV ibuprofen lysine to be more favorable in renal side effects, with significantly less rise in creatinine levels and less oliguria. PPHN has been seen only with ibuprofen THAM, which is not available in the US. There is no increase in the incidence of gastrointestinal complications with either agent. Prophylactic indomethacin has shown some protective properties against severe IVH, but concerns regarding treating children whose DAs might spontaneously close (60%) have prevented this from becoming widespread practice. Comparison treatment trials show no differences in the incidence of initial IVH or extension of existing IVH between the 2 drugs in treatment trials. There is no increase in the incidence of CLD at 36 weeks PMA with either drug, nor is there an increase in significant hyperbilirubinemia or bleeding dyscrasia. For treatment of PDA, IV ibuprofen lysine would be the treatment of choice due to significantly less renal side effects.