Objective: To assess the safety of selective head cooling in birth-asphyxiated term newborn infants while maintaining the rectal temperature at 35.0 degrees C or 34.5 degrees C.
Methods: Twenty-six term infants with Apgar <or=6 at 5 minutes or cord/first arterial pH <7.1, plus evidence of encephalopathy, were studied. After parental consent had been obtained, 13 infants received selective head cooling with the rectal temperature maintained at 35.0 degrees C in 6 infants and at 34.5 degrees C in 7 infants. The remaining 13 infants were normothermic. Cooling was achieved by circulating water at 10 degrees C through a cap placed around the head. Rectal, fontanelle, and nasopharyngeal temperatures were monitored.
Results: One cooled infant died 2 days after rewarming, and 3 control infants died. Seizures occurred in 9 (69%)of 13 cooled infants and 5 (38%) of 13 control infants. Respiratory support within the first 72 hours of life was required in 10 of 13 infants in both the cooled and control groups. Three cooled infants and 1 control infant received nitric oxide for persistent pulmonary hypertension. During the same interval, 6 of the cooled infants and 4 of the control infants had episodes in which their blood pressure fell to <40 mm Hg; in 2 infants in each group, the lowest blood pressure was below 35 mm Hg. No requirement for volume expansion or increased inotropic support was seen in any infant during stepwise rewarming. All of the cooled infants demonstrated a fall in heart rate during cooling, but the rate was <80/min in only 2 cases and no infant had a rate <70/min. No infant demonstrated an abnormal rhythm or was clinically compromised by the change in heart rate. One infant cooled to a rectal temperature of 34.5 degrees C had a prolonged QT interval of 570 ms associated with a heart rate of 85/min on electrocardiogram aged 34 hours. This returned to normal after rewarming. Platelet counts below 150 x 10(9)/L, hypoglycemia below 2.6 mmol/L, and highest creatinine were not statistically different between cooled and control infants. Positive precooling blood cultures were found in 1 cooled and 1 control infant. The mean cap water input temperature used during cooling was 10 +/- 1 degrees C. During active cooling, the mean difference between rectal and nasopharyngeal temperature was 1.4 degrees C in the infants who were not receiving respiratory support, but this gradient could not be measured in those who were receiving respiratory support that involved delivery of warmed gases to the nasopharynx.
Conclusions: This study suggests that selective head cooling combined with mild systemic hypothermia of 34.4 degrees C or 35.0 degrees C is a stable, well-tolerated method of reducing cerebral temperature in term newborn infants after perinatal asphyxia.