Presented at the Neonatal Society 2004 Summer Meeting.

Iwata O, DeVita E, Thornton J, O’Brien F, Iwata S, Shanmugalingam S, Peebles D, Scaravilli F, Cady E, Ordidge R, Wyatt JS, Robertson NJ

Departments of Paediatrics and Child Health, Medical Physics and Bioengineering, University College London, London WC1E 6JJ, UK

Background: Increasing evidence suggests that mild to moderate hypothermia is neuroprotective when commenced shortly after a hypoxic-ischaemic (HI) insult (Sirimanne et al., 1996). The extent of the neuroprotection, however, may be dependent on the delay, duration and depth of hypothermia and on the severity of the HI injury. A precise definition of the patient group who will respond to neuroprotective intervention is urgently required.

Objective: To assess the relationship between the severity of the HI insult and the efficacy of hypothermic neuroprotection assessed histologically in an experimental model.

Design/Methods: 19 piglets were anaesthetised within 24 hours of birth using isofluorane and morphine, mechanically ventilated and randomised to 3 groups; (i) normothermic (n-HI, n=6); (ii) core temperature 35ºC (35-HI n=6); and (iii) core temperature 33ºC (33-HI n=7). Animals were then subjected to a transient HI insult (bilateral carotid occlusion and FiO2 12-16% for ~1 hr). Phosphorus magnetic resonance spectroscopy gave a measure of the duration and magnitude of acute depletion of nucleotide triphosphate (NTP) relative to the exchangeable high energy phosphate pool (EPP): Animals were subgrouped into: (i) moderate insult (n-HI-m, 35-HI-m, 33-HI-m); and (ii) severe insult (n-HI-s, 35-HI-s, 33-HI-s). Animals were maintained at target temperature for 24hr commencing 2hr after the end of the insult and then rendered normothermic. At 48hr the animal was sacrificed and the brain perfusion fixed. Hematoxylin and eosin stained slices were assessed in 10 regions in the cortex and 6 regions in the deep grey matter. Percentages of viable and necrotic neurons in each region were compared in subgroups.

Results: Consolidating all grades of HI insult, compared to the n-HI group, 33-HI and 35-HI animals both had less dead neurons; only the 33-HI group had more viable neurons (Fig 1, all p< 0.05).

In the normothermic group histological scores were similar in moderate and severe insults (Fig 2). Hypothermic intervention improved histological scores in the moderate insult groups (35-HI-m, 33-HI-m) but not in the severe insult groups (Fig 2).

Conclusion: These data suggest that HI insult severity affects the efficacy of subsequent hypothermic intervention; systemic hypothermia of 35ºC and 33ºC were neuroprotective only after moderate HI insults.

References
1. Sirimanne ES, Blumberg RM, Bossano D et al., The effect of prolonged modification of cerebral temperature on outcome after hypoxic-ischaemic brain injury in the infant rat. Pediatr Res 1996; 39:591-7