Cognitive deficits are a devastating neurological outcome seen in survivors of cardiac arrest. We previously reported water electrolysis derived 67% hydrogen gas inhalation has some beneficial effects on short-term outcomes in a rat model of global brain hypoxia-ischemia induced by asphyxia cardiac arrest. In the present study, we further investigated its protective effects in long-term spatial learning memory function using the same animal model. Water electrolysis derived 67% hydrogen gas was either administered 1 hour prior to cardiac arrest for 1 hour and at 1-hour post-resuscitation for 1 hour (pre- & post-treatment) or at 1-hour post-resuscitation for 2 hours (post-treatment). T-maze and Morris water maze were used for hippocampal memory function evaluation at 7 and 14 days post-resuscitation, respectively. Neuronal degeneration within hippocampal Cornu Ammonis 1 (CA1) regions was examined by Fluoro-Jade staining ex vivo. Hippocampal deficits were detected at 7 and 18 days post-resuscitation, with increased neuronal degeneration within hippocampal CA1 regions. Both hydrogen gas treatment regimens significantly improved spatial learning function and attenuated neuronal degeneration within hippocampal CA1 regions at 18 days post-resuscitation. Our findings suggest that water electrolysis derived 67% hydrogen gas may be an effective therapeutic approach for improving cognitive outcomes associated with global brain hypoxia-ischemia following cardiac arrest. The study was approved by the Animal Health and Safety Committees of Loma Linda University, USA (approval number: IACUC #8170006) on March 2, 2017.