How long is sufficient for optimal neuroprotection with cerebral cooling after ischemia in fetal sheep?

Abstract

The optimal duration of mild "therapeutic" hypothermia for neonates with hypoxic-ischemic encephalopathy is surprisingly unclear. This study assessed the relative efficacy of cooling for 48 h versus 72 h. Fetal sheep (0.85 gestation) received sham ischemia (n = 9) or 30 min global cerebral ischemia followed by normothermia (n = 8) or delayed hypothermia from 3 h to 48 h (n = 8) or 72 h (n = 8). Ischemia was associated with profound loss of electroencephalogram (EEG) power, neurons in the cortex and hippocampus, and oligodendrocytes and myelin basic protein expression in the white matter, with increased Iba-1-positive microglia and proliferation. Hypothermia for 48 h was associated with improved outcomes compared to normothermia, but a progressive deterioration of EEG power after rewarming compared to 72 h of hypothermia, with impaired neuronal survival and myelin basic protein, and more microglia in the white matter and cortex. These findings show that head cooling for 48 h is partially neuroprotective, but is inferior to cooling for 72 h after cerebral ischemia in fetal sheep. The close association between rewarming at 48 h, subsequent deterioration in EEG power and increased cortical inflammation strongly suggests that deleterious inflammation can be reactivated by premature rewarming.

Keywords: Hypoxia-ischemia; encephalopathy; fetus; hypothermia; neuroprotection.

Figure 1.

Change in extradural temperature, esophageal temperature, and carotid artery blood flow before, during, and after 30 min of global cerebral ischemia (time 0) in near-term fetal sheep, showing the ischemia-normothermia (n = 8), ischemia-48 h hypothermia (n = 8), and ischemia-72 h hypothermia (n = 8) groups. Extradural temperature was significantly reduced in both hypothermia groups during treatment (p < 0.05) and returned to baseline when cooling ended at 48 or 72 h. A significant reduction in esophageal temperature was seen with the onset of hypothermia (p < 0.05). CaBF increased in the ischemia-normothermia group between 6 and 36 h after ischemia compared to either hypothermia group (p < 0.05). Data are mean ± SEM, *p < 0.05 ischemia-72 h hypothermia vs. ischemia-normothermia, ^#p < 0.05 ischemia-48 h hypothermia vs. ischemia-normothermia.

Figure 2.

Changes in EEG activity, spectral edge frequency, and impedance before, during, and after 30 min of global cerebral ischemia (time 0) in the ischemia-normothermia (n = 8), ischemia-48 h hypothermia (n = 8), and ischemia-72 h hypothermia (n = 8) groups. EEG activity was suppressed in all groups during/immediately after ischemia, followed by a transient increase during the seizure period (8–48 h). EEG activity in the ischemia-normothermia group was reduced for the remainder of the experiment, whilst both hypothermia groups showed a significant recovery from 24 h onward (p < 0.05). The end of hypothermia was associated with a rapid reduction in EEG power in the ischemia-48 h hypothermia group, which remained significantly lower than the ischemia-72 h hypothermia group (p < 0.05). Spectral edge was suppressed in all groups during ischemia and remained suppressed in the ischemia-normothermia group but increased significantly in both hypothermia groups between 8 and 12 h and from 84 h onward (p < 0.05). Ischemia was associated with an increase in impedance, which resolved after the end of ischemia. In the ischemia-normothermia group, impedance increased between 24 and 72 h, followed by a significant reduction from 120 h. A reduction in impedance was seen in the ischemia-48 h hypothermia group from 144 h onward compared to the ischemia-72 h hypothermia group (p < 0.05). Data are mean ± SEM, *p < 0.05 ischemia-72 h hypothermia vs. ischemia-normothermia group, ^#p < 0.05 ischemia-72 h vs. ischemia-48 h hypothermia group, ^ap < 0.05 ischemia-48 h hypothermia vs. ischemia-normothermia group.

Figure 3.

Neuronal survival 7 days after 30 min of global cerebral ischemia in the sham control (n = 9), ischemia-normothermia (n = 8), ischemia-48 h hypothermia (n = 8), and ischemia-72 h hypothermia (n = 8) groups. (a) Ischemia was associated with a significant reduction in neuronal survival in the cortex, CA1, CA3, CA4, and dentate gyrus of the hippocampus compared to sham control. Neuronal survival was significantly increased in the ischemia-72 h hypothermia group in all regions except the CA1 but only in the cortex and dentate gyrus in the ischemia-48 h hypothermia group. Neuronal survival was significantly greater in the cortex and CA4 in the ischemia-72 h hypothermia group compared to the ischemia-48 h hypothermia group. *p < 0.05 vs. sham control, ^#p < 0.05 vs. ischemia-normothermia, ^ap < 0.05 vs. ischemia-48 h hypothermia. (b) Photomicrograph showing NeuN-positive staining in the cortex (A, F, K, P), CA1 (B, G, L, Q), CA3 (C, H, M, R), CA4 (D, I, N, S), and dentate gyrus (E, J, O, T) in the sham control (A–E), ischemia-normothermia (F–J), ischemia-48 h hypothermia (K–O), and the ischemia-72 h hypothermia (P–T) groups. Scale bar 200 µm.

Figure 4.

Oligodendrocyte survival in the intragyral white matter of the first and second parasagittal gyri and the periventricular white matter in the sham control (n = 9), ischemia-normothermia (n = 8), ischemia-48 h hypothermia (n = 8), and ischemia-72 h hypothermia (n = 8) groups. (a) Ischemia was associated with a significant reduction in oligodendrocyte survival in all white matter regions with a partial improvement in both hypothermia groups in all regions, which remained significantly lower than sham control. Data are mean ± SEM, *p < 0.05 vs. sham control, ^#p < 0.05 vs. ischemia-normothermia. (b) Photomicrograph of Olig-2 positive labelling in the intragyral white matter of the first parasagittal gyrus (A, D, G, J), the second parasagittal gyrus (B, E, H, K) and the periventricular white matter (C, F, I, L) in the sham control (A–C), ischemia-normothermia (D–F), ischemia-48 h (G–I), and ischemia-72 h (J–L) groups. Scale bar 200 µm.

Figure 5.

Microglial number in the intragyral white matter of the first and second parasagittal gyri, the periventricular white matter and parasagittal cortex in the sham control (n = 9), ischemia-normothermia (n = 8), ischemia-48 h hypothermia (n = 8), and ischemia-72 h hypothermia (n = 8) groups. (a) Ischemia was associated with a significant increase in microglial number in all regions, which was attenuated in both hypothermia groups but remained significantly elevated in the intragyral white matter of the first parasagittal gyrus in the ischemia-72 h hypothermia group and in all regions in the ischemia-48 h hypothermia group. Microglial number was significantly higher in the ischemia-48 h hypothermia group compared to the ischemia-72 h hypothermia group all white matter regions. (b) Photomicrograph of Iba1-positive microglial labelling in the intragyral white matter of the first parasagittal gyrus (A, D, G, J), the second parasagittal gyrus (B, E, H, K), and the periventricular white matter (C, F, I, L) in the sham control (A–C), ischemia-normothermia (D–F), ischemia-48 h (G–I), and ischemia-72 h (J–L) groups. Scale bar 200 µm. (c) Ischemia was associated with a significant increase in microglial number in the parasagittal cortex compared to sham control, which was attenuated in the ischemia-72 h hypothermia group but not the ischemia-48 h hypothermia group. (d) Photomicrograph of Iba1-positive microglial labelling in the parasagittal cortex in the sham control (A), ischemia-normothermia (B), ischemia-48 h hypothermia (C), and ischemia-72 h hypothermia (D) groups. Scale bar 100 µm. Data are mean ± SEM, *p < 0.05 vs. sham control, ^#p < 0.05 vs. ischemia-normothermia, ^ap < 0.05 vs. ischemia-48 h hypothermia.

Figure 6.

Myelin basic protein (MBP) area fraction in the intragyral white matter of the first and second parasagittal gyri and the periventricular white matter in the sham control (n = 9), ischemia-normothermia (n = 8), ischemia-48 h hypothermia (n = 8), and ischemia-72 h hypothermia (n = 8) groups. (a) MBP area fraction was significantly reduced in all regions in the ischemia-normothermia group compared to sham control. Loss of MBP was partially attenuated in the ischemia-48 h hypothermia group in the first parasagittal gyrus only, whereas in the ischemia-72 h hypothermia group, MBP loss was reduced in all regions. Data are mean ± SEM, *p < 0.05 vs. sham control, ^#p < 0.05 vs. ischemia-normothermia, ^ap < 0.05 vs. ischemia-48 h hypothermia. (b) Photomicrograph showing MBP staining in the intragyral white matter of the first parasagittal gyrus (A, D, G, J), the second parasagittal gyrus (B, E, H, K), and the periventricular white matter (C, F, I, L) in the sham control (A–C), ischemia-normothermia (D–F), ischemia-48 h (G–I), and ischemia-72 h (J–L) groups. Scale bar 200 µm.