Sodium bicarbonate infusion during resuscitation of infants at birth.

For many years, intravenous sodium bicarbonate has been used to reverse acidosis during newborn resuscitation. However, controversy surrounds its use. Most of the evidence has been derived from studies in animals, adult humans, or in uncontrolled, descriptive experiments. Despite the lack of evidence from the human neonatal population and concerns about its safety, some international resuscitation guidelines still recommend the use of sodium bicarbonate in resuscitation of the newborn. To determine whether an intravenous infusion of sodium bicarbonate, compared to placebo or no treatment, reduces mortality and morbidity (in particular regarding neurodevelopmental outcome) in infants receiving resuscitation in the delivery room at birth. We used the standard search strategy of the Cochrane Neonatal Review Group. Searches were conducted of the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 3, 2005), MEDLINE (1966 - September 2005), EMBASE (1980 - September 2005) and CINAHL (1982 - September 2005) and Pediatric Research (1987 - September 2005). Unpublished trials were sought by handsearching the conference proceedings of American Pediatric Society/Society for Pediatric Research (1990 - 2005) and European Society for Paediatric Research (1993 - 2005). Randomised or quasi-randomised controlled trials of newborn infants receiving sodium bicarbonate infusion during any resuscitation in the delivery room at birth. Two review authors independently assessed trial quality and extracted data. Study authors were contacted for additional information. We found one randomised controlled trial that fulfilled the eligibility criteria (Lokesh 2004) that compared treating asphyxiated newborn infants (infants continuing to need positive pressure ventilation at 5 minutes after birth) with sodium bicarbonate infusion (N = 27) versus 5% dextrose (N = 28). They found no evidence of an effect on mortality prior to discharge [Relative risk 1.04 (95% confidence interval 0.49 to 2.21)], abnormal neurological examination at discharge [Relative risk 0.86 (95% confidence interval 0.30 to 2.50)] or a composite outcome of death or abnormal neurological examination at discharge [Relative risk 0.97 (95% confidence interval 0.59 to 1.60)]. There was no statistically significant difference in the incidence of encephalopathy [Relative risk 1.30 (95% confidence interval 0.88 to 1.92)], intraventricular haemorrhage [Relative risk 1.04 (95% confidence interval 0.23 to 4.70)] and neonatal seizures [Relative risk 1.19 (95% confidence interval 0.50 to 2.82)]. No long term neurodevelopmental outcomes were assessed. There is insufficient evidence from randomised controlled trials to determine whether the infusion of sodium bicarbonate reduces mortality and morbidity in infants receiving resuscitation in the delivery room at birth.