Male C57BL/6 mice (8–10-week old, weigh: 20–25 g) were purchased from Japan SLC (Tokyo, Japan). The mice were housed at 23 ± 2 °C under a 12-h light–dark cycle with free access to standard food and water. All experiments were performed between 09:00 and 17:00 under normal room light and temperature (23 ± 2 °C) conditions. A total of 44 mice were used in this study.
Hypoxic–ischemic brain injury
Hypoxic–ischemic brain injury was induced via a combination of permanent left common carotid artery occlusion and exposure to a low-oxygen environment, as previously described [9, 10]. Briefly, mice were placed in a dorsal position, and a middle neck incision was made under isoflurane anesthesia. The left common carotid artery was isolated from the vagus nerve, and then ligated and cut. One hour after the common carotid artery occlusion, mice were exposed to a low-oxygen environment (8% O2 balanced with nitrogen) for either 40 min or 25 min. In the 40-min injury model, the amiodarone groups received a single-bolus intraperitoneal injection of amiodarone (Sanofi K.K., Tokyo, Japan, 50 mg/kg) either immediately (0 min) or 10 min after the induction of hypoxic–ischemic brain injury (n = 10 and 6 respectively), while the control group received only normal saline immediately after the induction hypoxic–ischemic brain injury (n = 10). In the 25-min injury model, the amiodarone group received 50 mg/kg amiodarone immediately after the induction of hypoxic–ischemic brain injury, while the control group received only normal saline. Two mice were excluded due to the death before 24 h after induction (one per each group); eight mice were included in each group. The amiodarone dose was chosen based on the body surface area  and previously published data . The rectal temperature was monitored and maintained at 37 ± 0.5 °C using a heating pad. Heart rates and non-invasive blood pressures were monitored during the hypoxic–ischemic brain injury.
In the 40-min hypoxic–ischemic injury model, overall neurological function and survival rates over 7 days were evaluated. The neurological deficit score was used to evaluate global neurological deficit as follows: 0—no deficit; 1—torso flexion; 2—spontaneous circling; 3—leaning or longitudinal circling; 4—no spontaneous movement; 5—death . In the 25-min injury model, the mice were euthanized 24 h after the hypoxia–ischemic insult, followed by infarct volume analysis or water/sodium content analysis.
Measurement of infarct volume
Twenty-four hours after the 25-min hypoxic–ischemic injury, mice were deeply anesthetized with 5% isoflurane and euthanized by cervical dislocation. Brains were removed and coronal slices with a thickness of 1 mm were prepared. Brain slices were immersed in 2% 2,3,5-triphenyltetrazolium chloride (TTC; Sigma Aldrich, St. Louis, MO) solution, and incubated at 37 °C for 15 min. The area of infarction was traced and measured using image analysis software (ImageJ; National Institutes of Health, Bethesda, MD). The infarct area was calculated as follows to correct for edema: [1 − (total ipsilateral hemisphere − infarct region)/total contralateral hemisphere] × 100% . Total infarct volume was calculated as the sum of all infarct areas multiplied by section thickness (n = 8 each).
Evaluation of brain water content
Twenty-four hours after the 25-min hypoxic–ischemic injury, mice were deeply anesthetized with 5% isoflurane and euthanized by cervical dislocation. The whole brain was harvested and the olfactory bulbs and cerebellum were removed. The wet-weight of the brain was measured using a digital scale, and the brain was then freeze-dried for 72 h. The difference between the wet-weight and dry-weight was defined as brain water content (n = 8 each).
Inductively coupled plasma analysis
The freeze-dried brain was subjected to mortar pulverization, and immersed in 13 M nitric acid on a hot plate for chemical decomposition. The solution was then diluted with distilled water into pH 3–4, and subjected to inductively coupled plasma analysis (SPS3520, Hitachi High-Technologies, Tokyo, Japan). Sodium concentration of the sample solutions and the total brain sodium content was determined via the method of standard addition. Values were expressed in mEq/kg dry weight (n = 8 each).
Three-needle probe electrocardiogram monitoring was performed during general anesthesia and the surgical procedures (Powerlab, Bioamp, and LabChart 8, AD Instruments, NSW, Australia). The arterial blood pressure was non-invasively measured using a tail-cuff method (Softron, Tokyo, Japan). In the 25-min hypoxic–ischemic injury model, these hemodynamic parameters were recorded immediately before induction of hypoxic–ischemic injury (baseline) and immediately after exposure to the hypoxic environment (n = 8 each).
Statistical analysis was performed using Prism 6 software (GraphPad Software, San Diego, CA). The Log-rank test was used to analyze the survival rates; a two-tailed t test was used for the infarct volume, water content, and sodium content analysis; and the Kruskal–Wallis test followed by Dunn’s post hoc test was used to analyze neurological deficit scores. Values are presented as mean ± standard deviation in the infarct volume, water content, and sodium content analysis; values are presented as median and range in the neurological function analysis. The sample size of 8 mice per group was sufficient to provide 80% power with an α level of 0.05 to detect a mean difference of 10% in infarct volume. P < 0.05 was considered statistically significant.