Prophylactic intravenous indomethacin for preventing mortality and morbidity in preterm infants.

Patent ductus arteriosus (PDA) and intraventricular haemorrhage (IVH) are both associated with increased mortality and morbidity in preterm infants. Indomethacin has been used successfully to treat symptomatic PDA and may also prevent or limit IVH in the neonatal period. There are however potential unwanted side effects of indomethacin, in particular a potential for reduced organ perfusion that might outweigh any clinical benefits. The prophylactic use of indomethacin, where infants who may not have gone on to develop a symptomatic PDA or IVH would be exposed to indomethacin, warrants particular scrutiny. This review examines the effectiveness of prophylactic intravenous indomethacin in reducing the mortality and morbidity associated with PDA and IVH in preterm infants. An initial literature search was conducted in three computerised databases: MEDLINE; EMBASE; and the Oxford Database of Perinatal Trials in October 1994. The search was updated in February 1997 and October 2001. Strict selection criteria were applied to clinical trials: the population had to be newborn preterm infants (less than 37 completed weeks gestation); the intervention had to be prophylactic intravenous indomethacin; the trial had to be randomised and controlled; and at least one of several prespecified outcomes had to be reported in the results. The methodological quality of each study was assessed using explicit criteria. Data on relevant outcome measures were extracted and, where appropriate, the results of individual trials were combined using meta-analysis techniques to provide a pooled estimate of effect. Nineteen eligible trials randomising 2872 infants were identified. There is no evidence of difference in mortality at latest follow-up between infants receiving prophylactic indomethacin and controls, pooled relative risk (RR) = 0.96 [95% CI 0.81 to 1.12]. There is no evidence to suggest prophylactic indomethacin is associated with any reduction in long-term neurosensory impairment, ie no significant difference in rates of cognitive delay, cerebral palsy, blindness or deafness. The incidence of symptomatic patent ductus arteriosus is significantly reduced in treated infants, pooled RR = 0.44 [0.38 to 0.50] but there is no evidence that treatment affects respiratory outcomes. Prophylactic indomethacin reduces the need for surgical PDA ligation [RR = 0.51 (0.37,0.71)]. Prophylactic indomethacin significantly reduces the incidence of Grade 3 and 4 intraventricular haemorrhage, pooled RR = 0.66 [0.53 to 0.82]. There is no evidence of difference in rates of necrotising enterocolitis, excessive clinical bleeding or sepsis. Increased incidence of oliguria is seen with prophylactic indomethacin [RR = 1.90 (1.45,2.47] but this is not associated with major renal impairment. Prophylactic treatment with indomethacin has a number of immediate benefits, in particular a reduction in symptomatic patent ductus arteriosus, the need for duct ligation and severe intraventricular haemorrhage. There is no evidence to suggest either benefit or harm in longer term outcomes including neurodevelopment. Depending on clinical circumstances and personal preferences, there may be a role for prophylactic indomethacin in some infants on some neonatal units.