|
|
| |||||||||||||||||||||||||||||||||||||||||||||
Background Indomethacin is the conventional treatment for patent ductus arteriosus in preterm infants. However, its use is associated with various side effects. In a prospective study, we compared ibuprofen and indomethacin with regard to efficacy and safety for the early treatment of patent ductus arteriosus in preterm infants.
Methods We studied 148 infants (gestational age, 24 to 32 weeks) who had the respiratory distress syndrome and an echocardiographically confirmed patent ductus arteriosus. The infants were randomly assigned at five neonatal intensive care centers to receive three intravenous doses of either indomethacin (0.2 mg per kilogram of body weight, given at 12-hour intervals) or ibuprofen (a first dose of 10 mg per kilogram, followed at 24-hour intervals by two doses of 5 mg per kilogram each), starting on the third day of life. The rate of ductal closure, the need for additional treatment, side effects, complications, and the infants' clinical course were recorded.
Results The rate of ductal closure was similar with the two treatments: ductal closure occurred in 49 of 74 infants given indomethacin (66 percent), and in 52 of 74 given ibuprofen (70 percent) (relative risk, 0.94; 95 percent confidence interval, 0.76 to 1.17; P=0.41). The numbers of infants who needed a second pharmacologic treatment or surgical ductal ligation did not differ significantly between the two groups. Oliguria occurred in 5 infants treated with ibuprofen and in 14 treated with indomethacin (P=0.03). There were no significant differences with respect to other side effects or complications.
Conclusions Ibuprofen therapy on the third day of life is as efficacious as indomethacin for the treatment of patent ductus arteriosus in preterm infants with the respiratory distress syndrome and is significantly less likely to induce oliguria. (N Engl J Med 2000;343: 674-81.)
Ibuprofen has been shown to close the ductus in animals11 without affecting basal cerebral blood flow and intestinal or renal hemodynamics during positive-pressure ventilation,12 and it has been shown to have different effects on regional circulation from those of indomethacin.13,14,15 Furthermore, ibuprofen enhanced cerebral blood-flow autoregulation and had some neuroprotective effects in animals subjected to oxidative stress.16,17 More recently, ibuprofen has been shown to be effective for the closure of patent ductus arteriosus in premature infants,18,19,20 without reducing mesenteric, renal,21 or cerebral22 blood flow. These trials, however, included only a small number of infants. Therefore, we conducted a prospective, multicenter trial involving a larger number of patients. The primary objective was to compare the efficacy of ibuprofen with that of indomethacin in inducing closure of patent ductus arteriosus in premature infants with the respiratory distress syndrome. The comparison of the side effects in the two groups was a secondary objective.
Methods
Patients
Five tertiary neonatal intensive care centers in Belgium participated in the trial (the University Hospital in Antwerp, the University Hospital in Ghent, Queen Paola Children's Hospital in Antwerp, Sint Jan Ziekenhuis in Bruges, and Clinique Saint Vincent in Rocourt). The study was approved by the medical ethics committee of each center. Neonates were enrolled after written informed consent was obtained from their parents. The criteria for enrollment were a gestational age of 32 weeks or less; an age of 2 to 4 days; echocardiographic evidence of patent ductus arteriosus; and the respiratory distress syndrome necessitating respiratory support (all received mechanical ventilation starting at birth). Exclusion criteria were major congenital anomalies; life-threatening infection or hydrops fetalis; recent intraventricular hemorrhage (within the previous 24 hours); urine output below 1 ml per kilogram of body weight per hour during the preceding 8 hours; a serum creatinine concentration of 1.6 mg per deciliter (140 µmol per liter) or higher; a serum urea nitrogen concentration greater than 40 mg per deciliter (14 mmol per liter); a platelet count of 60,000 per cubic millimeter or less; a tendency to bleed, as revealed by hematuria, blood in the endotracheal aspirate, gastric aspirate, or stools, and oozing from puncture sites; and hyperbilirubinemia necessitating exchange transfusion.
Study Design
The infants at each unit were randomly assigned to a treatment group by means of cards in sealed opaque envelopes. Each infant received three doses of either indomethacin (Indocid I.V., Merck, West Point, Pa.; 0.2 mg per kilogram at 12-hour intervals) or ibuprofen (an initial dose of 10 mg per kilogram, followed by two doses of 5 mg per kilogram each, after 24 and 48 hours). The medications were infused continuously over a period of 15 minutes. The doses and intervals for ibuprofen were the same as in our earlier trial20 and were in accordance with recommendations for use in infants and neonates that were based on preliminary pharmacodynamic data.23,24 Ibuprofen was prepared for intravenous use from a commercially available product intended for parenteral administration (Imbun I.M., Merckle, Blaubeuren, Germany), delivered in vials containing 400 mg of dry, sterile powder of ibuprofen as a lysine salt, corresponding to 234 mg of ibuprofen. The contents of one vial were aseptically reconstituted with 23.4 ml of water for injection, yielding a sterile and pyrogen-free solution of 10 mg of ibuprofen per milliliter. This solution was aseptically dispersed in pyrogen-free ampules containing 2 ml each. Chromatographic analysis showed no appreciable degradation of the product after three months when it was stored at –20°C. All vials were prepared in a single pharmacy and distributed to the participating centers.
When the ductus arteriosus was still patent after the randomly assigned treatment in a patient in either group who was receiving mechanical ventilation, indomethacin (three doses of 0.2 mg per kilogram at 12-hour intervals) was given as a nonrandomized rescue treatment. If this therapy also failed to promote ductal closure and the patient continued to receive mechanical ventilation, or if there was a contraindication to the second pharmacologic treatment, surgical ligation of the ductus was performed.
Echocardiography
In all infants, color Doppler echocardiography (Sonos 1500 imaging system, Hewlett–Packard, Andover, Mass., or Ultramark 4 Plus, Advanced Technology Laboratories, Bothell, Wash., with a 7.5-MHz or 10-MHz transducer) was performed by physicians who were unaware of the infants' treatment assignments. The purpose was to evaluate patency of the ductus arteriosus and shunting at the time of inclusion and after the last dose of the study drug was given. The internal ductal diameter, the maximal shunt velocity, the ratio of the diameter of the left atrium to that of the aortic root (left-atrium-to-aortic-root ratio), and the degree of shunting were recorded as described earlier.25 Shunting was graded as moderate if a disturbed diastolic flow was easily detectable in the main pulmonary artery and there was a diastolic backflow in the aorta immediately beneath the ductus arteriosus and a forward flow above the ductal insertion; it was graded as severe if a diastolic backflow was easily detectable in the pulmonary trunk and in the aorta and if dilatation of the left atrium was present, with a left-atrium-to-aortic-root ratio above 1.6. The second evaluation was performed no more than 24 hours after the last dose of the assigned treatment. A third echocardiographic evaluation was performed to evaluate the effect of a second nonrandomized rescue treatment or whenever there was suspicion on clinical grounds that the ductus had reopened after closure.
Concomitant Treatment
Fluid intake was guided by the body weight, serum sodium concentration, and serum osmolarity. For infants with a gestational age of 26 weeks or less, the daily fluid intake began at 80 ml per kilogram and was increased by 20 ml per kilogram each day to a maximum of 130 ml per kilogram per day. For infants with a gestational age of 27 or 28 weeks, the fluid intake began at 70 ml per kilogram, and for those with a gestational age of 29 weeks or more it began at 60 ml per kilogram, with increases of 10 ml per kilogram each day to a maximum of 120 ml per kilogram per day by the end of the first week. The use of furosemide was restricted during the first week. When there was hypotension refractory to fluid-replacement therapy, a dopamine infusion was started. For treatment of the respiratory distress syndrome, infants received respiratory support (conventional or primary high-frequency oscillatory ventilation), oxygen supplements, and early rescue therapy (given from 2 to 18 hours after birth; mean, 6 hours) with surfactant (Survanta, Abbott Laboratories, Brussels, Belgium, or Alvofact, Boehringer Ingelheim Pharma, Biberach an der Riss, Germany) in doses of 100 mg per kilogram. Prophylactic antibiotics were administered from admission to the neonatal intensive care unit and stopped after three to four days if the results of the bacterial cultures (of blood, tracheal aspirate, and urine) remained negative.
Clinical Course and Outcome
Biologic data, demographic information, and clinical outcomes, including death, were prospectively recorded on data sheets especially designed for this study.
The dose of antenatal indomethacin or glucocorticoid treatment and the interval between the administration of the last antenatal dose and delivery were noted. Gestational age, birth weight, daily fluid intake, urine production, and body weight were also recorded. Urine was collected in adhesive urine bags. Oliguria was defined as a urine output of 1 ml per kilogram per hour or less during a 24-hour collection period. A tendency to bleed was defined according to the criteria used for exclusion from the study. During the first week, urine was screened daily for microscopic hematuria with use of a color-reagent test strip (Multistix, Bayer, Tarrytown, N.Y.). When the test strip indicated more than 5 to 20 red cells per microliter of urine, or when microscopy revealed more than 5 red cells per microliter, microscopic hematuria was diagnosed.
Cranial ultrasound scans were performed during the first three to four days of life, at the end of the first week, and weekly thereafter. All five centers used similar techniques. Infants were assessed for intraventricular hemorrhage (grade 1 to 4) and for periventricular leukomalacia (grade 1 to 3), which were graded according to standard classification systems (higher grades indicated greater severity).26,27,28 The serum creatinine concentration, serum sodium concentration, hematocrit, and platelet count were recorded during the first week.
The outcome in surviving patients was evaluated on the basis of clinical symptoms, the need for respiratory support (days during which mechanical ventilation and supplemental oxygen were needed), and the time required for the infant to regain his or her birth weight and to be ready for full enteral feeding. Bronchopulmonary dysplasia was defined by the need for supplemental oxygen after 28 days of life, in association with typical radiographic findings. Necrotizing enterocolitis was diagnosed when the clinical signs and radiographic findings generally accepted as characteristic of this condition were present.29
Statistical Analysis
We calculated that a study group of 140 neonates would be necessary for the study to be able to detect a difference of at least 20 percentage points in the closure rate between the ibuprofen and indomethacin groups, assuming a closure rate of 80 percent with indomethacin, with a P value of 0.05 and a power of 80 percent. The t-test, Mann–Whitney U test, and chi-square test or Fisher's exact test were used to compare continuous normally distributed data, nonparametric continuous data, and categorical data, respectively. All reported P values are two-tailed.
To assess the occurrence of side effects in the groups, repeated-measures analysis of variance was performed, with time as the within-subject variable and treatment allocation as the between-subject variable. To evaluate the difference in the rates of death, we performed Kaplan–Meier analysis with log-rank testing. Relative risks were used to indicate the efficacy of treatment and odds ratios to evaluate the probability of ductal closure. Multiple logistic regression was performed to assess the influence of predictive factors on the rate of closure of the ductus and the occurrence of oliguria; the factors we analyzed were sex, birth weight, gestational age, use or nonuse of antenatal glucocorticoid or indomethacin therapy, assignment to indomethacin or ibuprofen, the degree of increase in the serum creatinine concentration before treatment for patent ductus arteriosus, fluid balance and respiratory status before treatment, ductal diameter, maximal velocity of the left-to-right ductal shunt, severity of shunting before treatment, and use or nonuse of high-frequency oscillatory ventilation. We used Statistica '99 software (StatSoft, Tulsa, Okla.) and SPSS software (version 9.0, SPSS, Chicago).
Results
Efficacy of Treatment
We studied 148 infants, 74 of whom were assigned to each group. There was no significant difference between the two groups in base-line clinical and echocardiographic characteristics (Table 1). The rate of closure of the patent ductus arteriosus was similar in the indomethacin and ibuprofen groups, as was the need for rescue treatment (Table 2). No reopening of the ductus after closure was observed. The ductus was surgically ligated because of the presence of contraindications to a second pharmacologic treatment in four infants in the indomethacin group and in three infants in the ibuprofen group. The final numbers of infants who underwent ductal ligation for any reason were also similar in the two groups (Table 2). If there was only minor ductal shunting after therapy in a patient who did not require respiratory support, no further treatment of the ductus was attempted; in all such cases the ductus closed spontaneously before discharge.
|
|
26 weeks) but there was no difference in effect between the drugs within each category. The antenatal use of indomethacin as a tocolytic agent and a lower ductal shunt velocity on the third day of life (the latter reflecting a higher pulmonary arterial pressure) were independently associated with treatment failure.
|
Survival at one month was similar in the two groups (P=0.76) (Figure 1). The most common causes of death were refractory hypoxemia and overwhelming sepsis. There was no significant difference between the treatment groups in the number of infants with severe neonatal complications (Table 4). The rate of survival to discharge, the rate of bronchopulmonary dysplasia, the proportion needing respiratory support, the time required to regain the birth weight, the time until full enteral feeding was possible, and the rate of microscopic hematuria were similar (Table 4). No difference in the likelihood of a tendency to bleed was noted. No localized irritation, redness, or extravasation was noted during any of the infusions of the study medication.
|
|
Oliguria developed in 14 infants in the indomethacin group and in 5 infants in the ibuprofen group during the three days beginning with the start of treatment (P=0.03). Urine output was significantly lower from day 3 to day 7 in the indomethacin group than in the ibuprofen group (P<0.001), whereas the base-line values were similar (Figure 2). The increase in the serum creatinine concentration from day 4 to day 8 in the indomethacin group was significantly greater than that in the ibuprofen group (P=0.04) (Figure 3). The rate of administration of furosemide and the pattern of changes in daily weight were similar in the two groups (data not shown). Fifteen infants assigned to indomethacin and 14 assigned to ibuprofen received an infusion of dopamine during the first week of life, at similar doses.
|
|
|
On the basis of our findings, it appears that ibuprofen is as effective as indomethacin in promoting ductal closure in premature infants. The rate of closure in the group assigned to indomethacin was similar to rates previously reported. In 15 of 21 infants with birth weights of less than 1750 g (71 percent)30 and in 87 of 113 infants with similar gestational ages and birth weights (77 percent),5 a patent ductus arteriosus disappeared after indomethacin treatment at the age of two to seven days. More recently, a ductus arteriosus was closed after prophylactic treatment with indomethacin in 22 of 31 preterm infants (71 percent).31 Although we observed differences in the overall rate of ductal closure among the centers participating in our study, the efficacy of the two drugs remained similar in each of the centers.
Ibuprofen has been shown to constrict the ductus arteriosus effectively in lambs.11 Earlier, smaller studies suggested that ibuprofen might be effective in the prevention and early treatment of patent ductus arteriosus in human neonates.16,17,18 When ibuprofen was administered within three hours after birth, it reduced the subsequent incidence of patent ductus arteriosus in a preliminary study.19 In another study, it reduced the incidence of patent ductus arteriosus without altering cerebral blood flow.18 In 20 premature infants, it seemed to be as effective as indomethacin in closing the ductus arteriosus on the third day of life.20
Since in our earlier trial we observed that urine production was less affected by ibuprofen than by indomethacin,20 we investigated renal function in more detail in the present study. Oliguria (defined as urine output of less than 1 ml per kilogram per hour) developed in significantly fewer infants, and serum creatinine values did not increase, after ibuprofen treatment. Our observations underscore recent findings that ibuprofen does not cause a reduction in renal blood-flow velocity.21 Of the two cyclooxygenase isoenzymes that are known (COX-1 and COX-2), COX-1 seems to be primarily involved in basal physiologic processes in the kidney.32 Although both isoforms are inhibited by ibuprofen and indomethacin, indomethacin is more potent against COX-1.33 Although it is uncertain whether these findings also apply to preterm infants with patent ductus arteriosus, a difference in the renal effects of the two drugs remains a possibility. In addition, some experimental evidence suggests that there may be mechanisms unrelated to the inhibition of prostaglandins that partly explain the differences in the effects of indomethacin and ibuprofen on regional circulations.15,34
It is remarkable that necrotizing enterocolitis developed in a significantly higher proportion of the infants with oliguria than of those without oliguria (21 percent vs. 5 percent, P=0.02), probably reflecting an associated impairment of mesenteric perfusion. Moreover, twice as many infants had necrotizing enterocolitis in the indomethacin group as in the ibuprofen group, although the difference was not statistically significant. In an experimental model of bowel ischemia in rats, animals treated with ibuprofen had a significantly lower incidence of intestinal necrosis, and it was suggested that ibuprofen may have a cytoprotective role in animals at risk for bowel ischemia.14 In piglets, ibuprofen did not affect intestinal and renal hemodynamics during positive-pressure ventilation with high mean airway pressures12 and had no significant effects on gastrointestinal vascular resistance.15 In preterm infants, ibuprofen did not significantly reduce mesenteric blood-flow velocity.21 In our patients, however, this lack of effect was not reflected in earlier full enteral feedings or in the rate of weight gain. Feeding guidelines were, however, not strictly specified in the protocol.
A limitation of our trial is the relatively small number of patients, which limited the power of the study to detect significant differences in other clinical effects that we observed — notably, those related to outcomes such as necrotizing enterocolitis, isolated bowel perforation, intraventricular hemorrhage, and periventricular leukomalacia. For the same reason, small differences in the efficacy of the drugs according to gestational age may not have become apparent.
In summary, our data indicate that ibuprofen is as effective as indomethacin in promoting ductal closure on the third day of life in premature infants. However, ibuprofen is associated with significantly less impairment of renal function. No significant differences with regard to other side effects were observed. A lower gestational age (26 weeks), antenatal indomethacin use, receipt of high-frequency oscillatory ventilation, and an elevated pulmonary-artery pressure increased the risk of treatment failure.
We are indebted to the parents of the infants for their willingness to allow their children to participate in this study, to the staff and nurses of the participating centers for their cooperation, and to P. Bruyneel for preparing the study medication.
Source Information
From the Department of Pediatrics, Division of Neonatology, University Hospital Antwerp (B.V.O., K.D.); the Department of Pediatrics, Division of Neonatology, Ghent University Hospital (K.S.); the Neonatal Intensive Care Unit, Sint Jan Ziekenhuis, Bruges (D.L.); the Neonatal Intensive Care Unit, Queen Paola Children's Hospital, Antwerp (H.V.B.); the Department of Epidemiology and Community Medicine, University of Antwerp (J.W.); and the Neonatal Intensive Care Unit, Clinique Saint Vincent, Rocourt (J.-P.L.) — all in Belgium.
Address reprint requests to Dr. Van Overmeire at the Department of Pediatrics, Division of Neonatology, University Hospital Antwerp, Wilrijkstraat 10, B-2650 Edegem, Belgium, or at bart.van. overmeire{at}uza.uia.ac.be.
References
| |||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
