Cold Treatment Protects Against Infant Disability And Death From Oxygen Loss

The study appears in the October 13, 2005 New England Journal of Medicine.

"The experimental cooling of newborns to prevent death andinjury from oxygen deprivation during birth is extremely promising,"said NICHD Director Duane Alexander, M.D. "Yet it would be premature toimplement the study results under any but the most carefully controlledand monitored circumstances. The potential for serious harm exists ifthe conditions followed in this protocol are not carried out preciselyas they were during the study, by personnel skilled in their use."

The study was led by Seetha Shankaran, M.D., of the Division ofNeonatal-Perinatal Medicine, at Wayne State University School ofMedicine in Detroit, one of the participating NICHD Neonatal ResearchNetwork study sites.

Hypoxic ischemic encephalopathy (HIE) occurs when an infant'sbrain fails to receive sufficient oxygen or sufficient blood beforebirth. HIE may occur hours before birth, or, in some cases, duringlabor and delivery. The condition may result from a variety of causes.These include compression of the placenta, tearing of the placenta fromthe uterine wall before birth, compression of the umbilical cord, andrupture of the uterus. Dr. Shankaran explained that HIE is estimated tooccur from 0.5 to 1 times per every thousand births.

The study authors wrote that 10 percent of infants with moderate HIEdie, as do 60 percent of infants with severe HIE. "Many, if not all,"survivors of severe HIE experience major disability, they added.

Previous studies, conducted in laboratory animals, suggestedthat cooling the brain from 2 to 5 degrees Celsius after HIE couldreduce the chances for the death and disability that often result fromHIE, the authors wrote.

To conduct the study, researchers enrolled infants from the 15centers making up the NICHD neonatal network. All the infants hadexperienced oxygen deprivation during the birth process. A total of 208infants took part in the study. They were assigned at random to 1 of 2groups, with 102 infants undergoing the experimental cooling(hypothermia) treatment and 106 receiving standard care. Standard carefor HIE may involve placing the infant on a ventilator to assistbreathing monitoring blood pressure, and providing fluidsintravenously, and other newborn intensive care supportive therapies.

The infants were cooled by placing them on a soft plasticblanket through which water circulates. The blanket's temperature isregulated by computer. For the study, the blankets were set at 5degrees Celsius (41 degrees Fahrenheit). The infant's temperatures werelowered to 33.5 degrees Celsius (92.3 degrees Fahrenheit), as measuredby a temperature probe placed in an infant's esophagus. The infants inthe hypothermia group were enrolled within the first 6 hours of birth,and remained on the cooled blanket for 72 hours. After 72 hours hadpassed, they were gradually warmed to a normal body temperature.

Infants in both the hypothermia group and the control groupreceived standard newborn intensive care including monitoring of vitalsigns and were watched carefully for signs of organ dysfunction.

When the infants were examined at 18 to 22 months of age, 44percent of those in the hypothermia group developed a moderate tosevere disability or had died, as compared to 62 percent in the controlgroup.

Dr. Shankaran explained that when the study's three principaloutcomes--death, moderate disability, and severe disability--wereconsidered as one unit, the difference between the two groups ofinfants was statistically significant. However, when these threeadverse outcomes were analyzed as separate categories, the differencebetween the two groups of infants for any individual outcome was notstatistically significant.

Dr. Shankaran explained that it was not possible to recruit enoughinfants to arrive at statistically significant measures for thedifferences in the various outcomes between the two groups. Because HIEoccurs infrequently, it took 3 years to enroll enough infants toconduct the current study from the 15 participating NICHD NeonatalResearch Network sites.

In terms of the actual number of infants affected, fewerinfants in the hypothermia group died or experienced moderate or severedisability than was experienced by infants in the control group. Forexample, 24 infants in the hypothermia group died, as compared to 38 inthe control group. Similarly, 15 infants in the hypothermia groupexperienced disabling cerebral palsy, compared to 19 infants in thecontrol group. Blindness occurred in 5 infants in the hypothermia groupand in 9 infants in the control group. Infants in the hypothermia groupalso averaged higher on measures of infant mental and physicaldevelopment than did infants in the control group.

"A concern with any therapy that reduces mortality amonginfants at high risk of death and disability is the possibility of anincrease in the number of infants who survive with disabilities," thestudy authors wrote. "In our study there was no evidence of increasedrates of moderate or severe disability at 18 to 22 months of age amonginfants treated with hypothermia."

Side effects of the treatment consisted of a temporaryhardening and drying of the skin where the skin came in contact withthe cooling blanket, Dr. Shankaran said.

"Physicians need to exercise extreme caution in putting thestudy's results into practice,"said Rose Higgins, M.D., programscientist for the NICHD Neonatal Research Network and an author of thestudy. "Most newborn intensive care units don't have the resources orexperienced personnel to duplicate the carefully controlled conditionsof the study."

Dr. Higgins added that comparatively minor fluctuations in aninfant's body temperature--perhaps by as little as a few degrees--couldresult in serious harm if not closely monitored by trained personnel.

During the 72 hours of the hypothermia treatment, personneltrained in life support and use of the cooling blanket monitored allinfants continuously. Fluctuations in the infant's temperature werecompensated for immediately by adjustments to the cooling blanket.

Moreover, only full-term infants took part in the study, Dr.Higgins said. It is not known whether preterm infants with HIE wouldbenefit or be harmed from hypothermia treatment.

Dr. Higgins said that the NICHD is currently advising theAmerican Academy of Pediatrics to develop practice recommendations fortreating infants with HIE. Moreover, three ongoing studies ofhypothermia treatment are expected to provide additional information onthe most effective ways to carry out the treatment.

Dr. Higgins added that the NICHD Neonatal Research Networkwill also follow both groups of children until they reach the ages of 6or 7, to compare the incidence of health problems or learningdifficulties.

The NICHD is part of the National Institutes of Health (NIH), thebiomedical research arm of the federal government. NIH is an agency ofthe U.S. Department of Health and Human Services. The NICHD sponsorsresearch on development, before and after birth; maternal, child, andfamily health; reproductive biology and population issues; and medicalrehabilitation.