Brain Hypothermia and QT Interval

To the Editor. When any new therapy is proposed, it is important to monitor for adverse events. Mild systemic hypothermia to 35.5°C by head cooling in term infants with hypoxic ischemic encephalopathy has recently been reported to be safe, and a multinational study with the power to evaluate efficacy is planned.1 We have continued pilot studies with lower systemic temperatures, aiming to keep the nasopharyngeal temperature below 34°C, which was a critical temperature for neuronal rescue after carotid artery occlusion in our studies in fetal sheep.2 Hypothermia is associated with recognized electrocardiography changes including sinus bradycardia, prolonged PR interval, prolonged QT interval and Osborn or J waves, and in some cases cardiac arrhythmias, all of which are reversible with rewarming.3We report an infant with markedly prolonged QT interval during cooling after hypoxic ischemia. This infant girl was born to a 35-year-old G2 P1 epileptic mother who refused her Tegretol medication during pregnancy. In early labor at 41 weeks, 4 days gestation, she had a grand mal seizure; delivery was by emergency cesarean section delayed for 10 minutes by difficulties in maternal intubation. The baby weighed 3050 g with Apgars of 4 and 6 at 1 and 5 minutes, respectively, and required intubation and ventilation for 10 minutes. The arterial umbilical cord blood gas pH was 6.78, Pco2 was 22.4 kPa, base excess was −15. At 3 hours of age she remained extremely hypotonic and flaccid, without suck or response to painful stimuli. Because of this evidence of moderate encephalopathy, parental consent was obtained for head cooling, which was started at 3.5 hours of life. Initial rectal temperature was 36.7°C and the heart rate was 134 bpm. After 1 hour of cooling the rectal temperature was 34.1°C and the heart rate was 96. This slow heart rate persisted and at 34 hours of age, when the rectal temperature was 34.3°C, an electrocardiogram showed a rate of 85 bpm with a markedly prolonged QT duration of 570 milliseconds (ms) in lead V5. (The normal QT interval range at 85 bpm is 290 to 380 ms, and the normal QT interval for this age is 215 to 345 ms.4 The venous serum sodium was 130 mM, potassium was 4.3 mM, and calcium was 2.46 mM. No cardiac arrhythmia was detected and cooling continued for the full 3 days. Brief seizures were treated with phenobarbitone and a pneumomediastium was treated with a 24-hour period of continuous positive airway pressure (with inspiratory oxygen fraction of 1.0). On rewarming the infant, the heart rate increased from 87 to 139 bpm, over 2 hours. The day after rewarming an electrocardiogram with a heart rate of 125 bpm showed a QT duration of 300 ms, within the normal range for age and heart rate.4These data suggest that in future studies of hypothermic therapy for neuronal protection, if there is a sustained decrease in heart rate below 90, then serial electrocardiograms should be performed. If prolongation of the QT interval is seen, then close monitoring for ventricular arrhythmias is mandatory; ventricular tachycardia associated with prolonged QT interval would be an indication to reduce or stop hypothermia. Prolonged QT in the absence of ventricular arrhythmias may be safe; other therapies that lengthen the QT interval, such as macrolide antibiotics, should be avoided.