Early versus late treatment of PDA with indomethacin

Abstract

We wish to express concerns about the conclusions drawn by Van Overmeire et al1Van Overmeire B Van de Broek H Van Laer P Weyler J Vanhaesebrouck P Early versus late indomethacin treatment for patent ductus arteriosus in premature infants with respiratory distress syndrome.J Pediatr. 2001; 138: 205-211Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar on timing of indomethacin treatment for closure of patent ductus arteriosus (PDA). Although sample size for this early versus late indomethacin trial (127 mechanically ventilated preterm infants) was calculated specifically to determine whether delaying indomethacin treatment affects the rate of PDA closure, the authors also concluded that timing of treatment does not affect the incidence of bronchopulmonary dysplasia (BPD) (supplemental O2 >28 days). We disagree that this conclusion is necessarily valid or generalizable. With the use of the ~40% BPD rate in their early treatment group and their statistical criteria, we calculated that 324 study infants would be needed to detect a difference of at least 20% in BPD between these groups (α =.05, β =.2).2Hermansen M Sample size.in: Biostatistics: some basic concepts. Caduceus Medical Publishers, New York1990: 185-189Google Scholar We also suspect the treatment groups may be different. The use of high-frequency oscillatory ventilation (HFOV) before randomization was significantly higher in the early (21/64) than in the late (11/63) treatment group. This greater prerandomization use of HFOV suggests these infants had worse respiratory disease. Although the authors inferred that the similar mean airway pressures (MAP) between the groups indicated similar respiratory illness, MAP measured in conventional ventilation and HFOV cannot be meaningfully compared, and MAP is not a valid surrogate for illness severity scoring. Van Overmeire et al1Van Overmeire B Van de Broek H Van Laer P Weyler J Vanhaesebrouck P Early versus late indomethacin treatment for patent ductus arteriosus in premature infants with respiratory distress syndrome.J Pediatr. 2001; 138: 205-211Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar also concluded that late treatment was associated with significantly fewer adverse renal effects. In the early treatment group, not unexpectedly, mean urine output was lower and mean serum creatinine concentration was higher. Although statistically significant, it is not evident that these transitory deviations are clinically significant or adversely affected outcomes. Finally, the authors drew attention to other extrapulmonary adverse events in the early treatment group. A bowel perforation occurred after early indomethacin administration and, in 4 infants, intraventricular hemorrhage extended during treatment. Randomized clinical trials have not supported these associations.3Ojala R Ikonen S Tammela O Perinatal indomethacin treatment and neonatal complications in preterm infants.Eur J Pediatr. 2000; 159: 153-155Crossref PubMed Scopus (35) Google Scholar, 4Fowlie PW Intravenous indomethacin for preventing mortality and morbidity in very low birth weight infants.Cochrane Database Syst Rev. 2000; 2 (CD 000174)PubMed Google Scholar Consequently, we are surprised that anecdotal observations featured prominently in the report discussion, and they were used to support the conclusion: “We advocate an expectant attitude with regard to postnatal indomethacin treatment.” Hemodynamically significant PDA in sick preterm infants can lead to respiratory, intestinal, and neurodevelopmental morbidities. It follows that a delay in treatment needs careful justification. The expectant attitude suggested by Van Overmeire et al1Van Overmeire B Van de Broek H Van Laer P Weyler J Vanhaesebrouck P Early versus late indomethacin treatment for patent ductus arteriosus in premature infants with respiratory distress syndrome.J Pediatr. 2001; 138: 205-211Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar is based on insufficiently powerful post hoc analysis, transitory suppression of renal perfusion, and anecdote. Clinical practice should be guided by more rigorous standards. We are not persuaded that benefits of delayed indomethacin treatment outweigh the costs.