Does fluid balance affect neurodevelopmental outcomes in hypoxic-ischaemic encephalopathy?

Abstract

Hypoxic-ischaemic encephalopathy (HIE) is a major cause of morbidity and mortality in neonates worldwide.1 Neonates with HIE are treated with therapeutic hypothermia (TH) and other supportive measures. Restrictive fluid regimen, with attention to fluid balance, is advocated in the treatment of HIE to prevent exacerbation of cerebral oedema.2, 3 However, there is no evidence to support this practice or to provide insight into the effect of fluid balance on outcomes of neonates with HIE.4, 5 This study by Ottolini et al. aimed to determine any association between positive fluid balance and death or moderate to severe brain injury on post-rewarming magnetic resonance imaging (MRI) in neonates with HIE treated with TH. All neonates were initially administered fluids at a standard rate of 60 ml/kg/day. Infusion rates were adjusted subsequently as per clinical discretion to maintain target sodium and glucose levels. Retrospective analysis of data showed positive fluid balance during the first 24 h of TH conferred 3.4 times higher odds of death or moderate to severe brain injury in neonates. The odds of this primary adverse outcome increased to 5.8 times in neonates with positive fluid balance at 48–72 h of TH. Ottolini et al. propose acute kidney injury and syndrome of inappropriate secretion of anti-diuretic hormone (SIADH), as mechanisms underlying the higher incidence of free water retention and lower plasma sodium levels observed in neonates who died or had moderate-to-severe brain injury in this study.6-8 Accordingly, whilst it could be hypothesised that positive fluid balance itself contributes to the adverse neurodevelopmental outcome, the degree of positive fluid balance may concurrently indicate the severity of hypoxic injury in neonates with HIE. Therefore positive fluid balance may be used as an early prognostic marker of severity of brain injury. Previous randomised controlled trials (RCT) found fluid restriction precipitated hypoglycaemia and hyponatremia, potentially confounding the observed adverse outcomes.4, 7 Anticipation and correction of these electrolyte disturbances gives this study's results with greater validity. However, the lack of further details of fluid management makes it difficult to determine exact fluid volumes administered throughout the study period. Consequently, it is unclear whether positive fluid balance, as a result of hypoxic injury, was exacerbated by the fluid strategy undertaken in the study. Apart from a median of 9 days of life, and an interquartile range of 6–10 days of life, definitive details of timing of the post-rewarming MRI assessment of brain injury have not been provided. Furthermore, twenty neonatal deaths were recorded prior to MRI, and deaths beyond this point were not reported. A multicentre RCT comparing a standard versus restrictive fluid regimen with long-term neurodevelopmental surveillance may address the controversy around the ideal fluid strategy in HIE. Comparing levels of biomarkers that indicate the severity of HIE and SIADH in both groups may also provide information on prognostication. Despite the limitations discussed, the study illustrates a significant association between positive fluid balance and adverse outcomes, and further research is warranted on fluid therapy in HIE. None declared. URL to the full review on the EBNEO website.