Hypoxic-ischemic encephalopathy due to insufficient oxygen delivery to brain tissue is a leading cause of death or severe morbidity in neonates. The early recognition of the most severely affected individuals remains a clinical challenge. We hypothesized that hypoxic-ischemic injury can be detected using PET radiotracers for hypoxia ([¹⁸F]EF5), glucose metabolism ([¹⁸F]FDG), and inflammation ([¹⁸F]F-DPA).

Methods

A preclinical model of neonatal hypoxic-ischemic brain injury was made in 9-d-old rat pups by permanent ligation of the left common carotid artery followed by hypoxia (8% oxygen and 92% nitrogen) for 120 min. In vivo PET imaging was performed immediately after injury induction or at different timepoints up to 21 d later. After imaging, ex vivo brain autoradiography was performed. Brain sections were stained with cresyl violet to evaluate the extent of the brain injury and to correlate it with [¹⁸F]FDG uptake.

Results

PET imaging revealed that all three of the radiotracers tested had significant uptake in the injured brain hemisphere. Ex vivo autoradiography revealed high [¹⁸F]EF5 uptake in the hypoxic hemisphere immediately after the injury (P < 0.0001), decreasing to baseline even 1 d postinjury. [¹⁸F]FDG uptake was highest in the injured hemisphere on the day of injury (P < 0.0001), whereas [¹⁸F]F-DPA uptake was evident after 4 d (P = 0.029), peaking 7 d postinjury (P < 0.0001), and remained significant 21 d after the injury. Targeted evaluation demonstrated that [¹⁸F]FDG uptake measured by in vivo imaging 1 d postinjury correlated positively with the brain volume loss detected 21 d later (r = 0.72, P = 0.028).

Conclusion

Neonatal hypoxic-ischemic brain injury can be detected using PET imaging. Different types of radiotracers illustrate distinct phases of hypoxic brain damage. PET may be a new useful technique, worthy of being explored for clinical use, to predict and evaluate the course of the injury.