Article Text

Download PDFPDF

Should a prolonged or short course of indomethacin be used in preterm infants to treat patent ductus arteriosus?
  1. Sachin Shah
  1. Hospital for Sick Children and University of Toronto; sshahdoc{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

A 27 week gestation infant is diagnosed to have a significant patent ductus arteriosus (PDA) on echocardiography on day 2. The infant is ventilator dependent. You decide to treat with indomethacin. The resident suggests that it would be better if prolonged indomethacin therapy could be administered over 5–7 days to ensure that the PDA remains closed. What evidence did she have?

Structured clinical question

In preterm infants with patent ductus arteriosus [patient], is prolonged course indomethacin [intervention] better than short course conventional therapy [comparator] in preventing recurrences and decreasing the need for surgical ligation [outcomes]?

Search strategy and outcome

Medline (1966–Dec 2002): “indomethacin” AND “patent ductus arteriosus” AND “infant, newborn”; LIMIT to “english language” and “human”—375 references. Other databases searched were Cochrane Controlled Trials Register (Issue 4, 2002), EMBASE (1980–Dec 2002), CINAHL (1982–Dec 2002), abstracts published in Pediatric Research (1990–2002). Five relevant trials were identified. See table 2.

Table 2

Prolonged versus short course indomethacin in treatment of patent ductus arteriosus in preterm infants

Clinical bottom line

  • The evidence is inconclusive as to which is the superior regimen.

  • The prolonged course regime may have a role in those with borderline impairment of renal function.


The five RCTs comparing prolonged versus short course indomethacin differ in the dosage regimes (for both prolonged and short course groups), diagnosis of PDA, and onset of treatment. Also there is a wide variation in the gestational age and birth weight.

In all but one study (Rennie et al), PDA was diagnosed by echocardiography. In two studies (Lee et al; Rhodes et al), PDA was detected on echocardiographic screening at predetermined intervals, while in two studies (Tammela et al; Hammerman et al), a clinically symptomatic PDA was confirmed on echocardiography.

There is significant “in between study heterogeneity” as far as outcomes like failure of PDA closure, need for surgical ligation, recurrence of PDA, and mortality rates are concerned.

In the majority of studies, the incidence of renal side effects was less in the prolonged indomethacin group compared to the short course group. There were no significant differences in the other co-morbidities.

In theory, short term indomethacin therapy suppresses dilator prostanoids and facilitates ductal constriction. This transient suppression may not allow sufficient time for anatomic ductal closure in many infants. This may be especially true for the extremely low birth weight infant whose PDA is more sensitive to prostaglandins and whose PDA does not constrict as tightly as the more mature infant. Also the ductus tends to retain its sensitivity for a longer duration, making it more susceptible to reopening.

This theoretical advantage has not translated into clinical practice in all the studies. The evidence is inconclusive as to which is the superior regimen. Further trials are needed to determine the optimum regimen and dose of indomethacin, especially in extremely low birth weight infants (<1000 g). A four arm trial where infants are randomised and allocated at birth, either to prolonged therapy or to short course therapy, may be the most efficient trial design. Within the prolonged and short course treatment arms, the infants would then be randomised to receive indomethacin on echocardiographic determination of PDA within the first seven days or when the PDA becomes clinically symptomatic. The outcomes assessed should be mortality and need for surgical ligation.



  • Bob Phillips