Cold 'cuts baby brain damage'

"Babies who are starved of oxygen at birth have a much lower risk of brain damage if they are given mild hypothermia,” reported BBC News. The website said research in more than 300 babies found that full-term babies that were affected by oxygen deprivation when born were 57% more likely to survive without brain damage if they were cooled.

This is a solid study that is part of a growing body of research on this treatment. This study brings closer the decision on whether this treatment should become standard practice. However, the evidence is not yet conclusive, and some issues about how the treatment is delivered and the extent of any long-term benefit still need to be resolved.

Where did the story come from?

The research was carried out by Dr Denis Azzopardi from the Division of Clinical Sciences and Medical Research Council, Clinical Sciences Centre, based at Imperial College London and colleagues from other UK institutions. It was supported by grants from the UK Medical Research Council and the Department of Health. It was published in the peer-reviewed medical journal The New England Journal of Medicine .

What kind of scientific study was this?

This research investigated whether cooling (hypothermic therapy) of newborn babies with brain symptoms (e.g. lethargy) after being starved of oxygen (asphyxial encephalopathy) reduces results such as death or severe disability at 18 months of age.

The study is called the Total Body Hypothermia for Neonatal Encephalopathy Trial (TOBY). This was a randomised-controlled trial in infants who were less than six hours old and had been born within four weeks of their expected due date with a condition known as perinatal asphyxial encephalopathy. This condition is partly caused by a lack of oxygen to the brain during birth and is responsible for high rates of disability and death worldwide.

The researchers randomised 325 newborn infants with the condition to receive either intensive care alone (162 subjects) or intensive care with cooling to 33.5°C for 72 hours (163 subjects). Death and severe disability at 18 months of age was the main result (outcome) they were looking closely at. They also looked for 12 other neurologic (nervous system) outcomes and 14 adverse outcomes, including bleeding inside the head, low blood pressure, kidney failure and clotting abnormalities.

The diagnosis of brain asphyxia was made by recognised criteria such as an Apgar score of five 10 minutes after birth, or a continued need for resuscitation, blood test abnormalities or seizures. The cooling was performed with parental consent and consisted of treatment in incubators with the power turned off. A target temperature of 33 to 34°C was maintained by placing the infant on a fluid-filled cooling blanket. The blanket’s temperature was regulated by a manually adjusted thermostat. This trial did not use head cooling.

The data were analysed in the groups to which patients had been assigned regardless of whether they later moved groups. For example, a baby who began treatment with intensive care only, but then was later treated with cooling was analysed as if they had remained in the control group.

What were the results of the study?

Of the 325 infants, 163 underwent intensive care with cooling, and 162 underwent intensive care alone.

In the cooled group, 42 infants died and 32 survived but with severe neurodevelopmental disability, whereas in the non-cooled group, 44 infants died and 42 had severe disability. This difference was not statistically significant (RR 0.86, 95% confidence interval 0.68 to 1.07).

Taking into account two infants who were not followed to 18 months, there were 71 infants in the cooled group who survived without neurological abnormality and 21 with multiple developmental disabilities (some of which were severe). This compared to 45 infants in the non-cooled group without any neurological abnormality and 33 with multiple neurodevelopmental disabilities. This means that infants in the cooled group did have a statistically significant increase in survival without neurologic abnormality. In the study group 44% survived without abnormality compared to 28% in the control group (RR 1.57, 95% confidence interval 1.16 to 2.12).

Other noted improvements from cooling were reduced risks of cerebral palsy and improved scores on the scales of mental, psychomotor and overall infant development scales and the Gross Motor Function Classification System.

Improvements in other neurologic outcomes in the cooled group were not significant. Adverse events were mostly minor and not associated with cooling.

What interpretations did the researchers draw from these results?

The researchers conclude that moderate hypothermia for 72 hours in infants who had perinatal asphyxia did not significantly reduce the combined rate of death or severe disability but resulted in improved neurologic outcomes in survivors.

What does the NHS Knowledge Service make of this study?

Other studies have investigated similar (but not identical) cooling techniques and the authors discuss these in context. They state that a 2007 review found eight randomised-controlled trials that included 638 term infants. It concluded that cooling of newborn babies who may have suffered from a lack of oxygen at birth reduces death or disability, without increasing disability in survivors. However, they note that the review, done two years ago, was based on less than half of all infants currently known to be randomised into eligible trials of cooling.

It is important to note other issues mentioned by these authors:

• Minor heart and respiratory problems were common, but serious adverse events were rare and not associated with cooling.
• Some adverse events occurred in both groups. For example, mild-to-moderate bleeding inside the head (intracranial haemorrhage) was frequently seen on MRI in both groups but not on ultrasound, and sinus thrombosis occurred very infrequently in both groups at similar rates.
• The researchers call for longer-term studies as they say that to date, no trial has reported neurologic outcomes after 18 months and the less severe impairments are not reliably assessable at this age. Assessment later in childhood (for example, at age six or seven) is necessary for an accurate, comprehensive evaluation of cognitive function, behaviour and learning, fine motor development, attention, and psychosocial health.

Systematic reviews of this topic may need to be updated so that they include this trial. As many patients are needed to detect small differences in these important outcomes, systematic reviews are an essential technique to combine the data from smaller studies. This is a large study, and so could contribute a lot of patient data to such a review.

Systematic reviews that assess different types of cooling (for example, assessing whether head cooling is better than body cooling) could also be helpful in the debate about whether this treatment should become standard practice.