This was an experimental randomized posttest-only control group design.

Research activities were carried out at the Research and Development Centre for Stem Cells, Universitas Airlangga in July, and August 2021. The adaptation phase was carried out for one week, while the treatment and symptom observation phase lasted 48 hours.

Experimental animals used were all healthy, six-month-old, male Wistar rats, a strain of Rattus norvegicus. Rats were divided into a control group and a treatment group. Animals in the control group underwent no experimental interventions. In the treatment group, the right common carotid artery (CCA) was ligated. The rats were then placed for 45 minutes in a hypoxic atmosphere chamber, with an ambient gas composition of 8% O₂ and 92% N₂11, 12, 13. Following hypoxia, reperfusion was performed by opening the CCA ligation after 60 minutes. After 48 hours, the rats were sacrificed. NSE levels were measured in blood serum by ELISA, using NSE reagent E-EL0R0058 (Elabscience Biotechnology Inc., China). Plates were read using a SPECTROstar® Nano plate reader (BMG Labtech, Germany).

We established a cerebral ischemia model according to the Vannucci method. Wistar rats were anaesthetized with xylazine 10%, injected intramuscularly. An incision was made in the midventral cervical area in the middle of the neck, on the upper edge of the sternum. This allowed access to the sternocleidomastoid muscle and the sternohyoid muscle. The sternocleidomastoid muscle deep inside the sternohyoid muscle was pulled slowly until the sternohyoid muscle appeared. The common carotid artery (CCA) is normally enwrapped in fibrous connective tissue, which contains vagus nerves. The CCA was carefully separated from the attached tissue using ophthalmic forceps. The unilateral CCA was then bound twice with two stitches with 4-0 silk, next to each other, and then the surgical wound was closed with a suture. Reperfusion was performed by opening of the CCA ligation for two hours.

Clinical evidence of brain ischemia was assessed with a neurological examination, using a six-point Longa score scale as follows: 0: no neurological deficit; 1: left forefoot fails to fully extend, indicating a mild focal neurologic deficit; 2: turns left, meaning a moderate focal neurologic deficit; 3: falls to the left, indicating a severe focal neurologic deficit; 4: does not run spontaneously and exhibits a lower level of consciousness; 5: death due to brain ischemia. If the animal's score was 0 or 5, it was excluded from the study14.

All animal protocols were approved by the Committee of Animal Care and Use, Faculty of Veterinary Medicine, Airlangga University No: 2.KE.019.02.2018.

Normality of the control and treatment group data distributions was tested using the Shapiro–Wilk test, followed by a t-test for independent sample analysis. The Statistical Package for the Social Sciences (SPSS) 21.0 software program was used for this analysis.

Table 1

Bodyweight (BW)	Group
	Control	HIE
BW baseline (g)
Mean	203.4	206.4
Standard Deviation	4.307	5.476
BW 48 hours post Ligation (g)
Mean	199.0	204.3
Standard Deviation	4.629	5.599

Table 2

	HIE Group
No Rats	1	2	3	4	5	6	7	8
Neurology Score after 24 hours	1	2	2	2	1	1	1	1

Table 3

Group	N	Mean	SD	95%CI	Mean Difference	P
Control	8	0.946	0.093	0.300-0.715	0.185	0.004*
HIE	8	1.132	0.118	0.300-0.709

*Significant P < 0.05 with T-test for independent sample

Twenty-two healthy male Wistar rats that had not been used in previous studies were selected for the experiment. The characteristics of these rats are shown in Table 1.

We evaluated animals phenotypically for evidence of brain ischemia by performing a neurological examination on each animal, using the six-point Longa score scale (see methods). We found that all HIE group rats experienced hemiparesis at different levels, as shown in Table 2.

The normality of the NSE variable data was tested using the Shapiro–Wilks test. NSE data in each group showed a normal distribution. Levene's test was used to test whether our two groups had homogeneous variances. We found that the NSE levels were significantly higher in the treatment group compared with the control group (Table 3).

It is worth noting that the mortality of rats during the experiment reached 27% (3 rats) each in the control group and treatment group. The exact cause of death is unknown, but a severe neurological disorder is suspected. A total of 16 rats completed the study.

Enolase, or 2-phospho-D-glycerate hydrolase, is an essential metabolic enzyme and is found in the cytoplasm. Enolase is one of the enzymes of the glycolytic pathway that converts glucose into pyruvate. Brain enolase consists only of alpha and gamma subunits. Gamma enolase is categorized as neuron-specific enolase (NSE) because of the distinction of its neurons15.

Previous studies have shown that NSE is detectable at high levels in CSF after brain injury. One outstanding question not addressed by those studies is the level of NSE in the blood serum in brain injury. This study set out to answer that question by measuring NSE level from blood serum of HIE rats model. We used CCA ligation to induce hypoxic ischemia and measured NSE levels after reperfusion. We found that NSE levels were higher in the serum of the study group compared to controls. NSE is considered a precise marker for neuronal injury but not glial injury8, 16. It is also a good marker for brain ischemic animal models, including local infarct and global ischemia16, 17. Cerebrospinal fluid (CSF) NSE levels are correlated with ischemia duration and the size of cerebral infarct. We interpreted our serum results after ischemia to mean that the level of neuron death was higher in our experimental animals than in controls.

One previous study compared the NSE levels in mouse CSF before and until seven days after ligation. The NSE level was measured using a radioimmunoassay kit. In those experiments, consolidation of the damaged area on the frontal lobe was evaluated to measure the infarct volume. There was a correlation between CSF NSE levels on the third day postligation and the infarct volume18.

NSE has been used as a more global marker of ischemia in humans, e.g., after cardiac arrest, status epilepticus, or drowning16, 18. A second study reported that focal ischemia treatment efficacy can be evaluated using NSE as a marker. They investigated the correlation between NSE and infarct volume in test animals by ligating the carotid artery. Ten minutes after ligation, animals were given intraperitoneal nicardipine 1.2 mg/kg and then again at 8, 16 and 24 hours after ligation. The mice that were given nicardipine had a 19% smaller infarct volume than the ligated group that was not treated. When NSE levels were measured 24 hours after ligation, they found that NSE was three times higher in ligated animals than in a control group that was not ligated. Interestingly, animals that were ligated and also treated with nicardipine had a lower level of NSE than those that were ligated and untreated, with the difference being up to 50% lower at 24 hours, 42% at 48 hours and 59% at 72 hours after ligation. These data showed that NSE levels may be correlated with different outcomes from therapeutic intervention for stroke19, 20.

Earlier studies have reported that NSE can be measured not only from CSF but also from blood10. In one such study, serum NSE levels were measured in mice with status epilepticus (SE) induced by lithium pilocarpine, and in healthy mice at the ages of 1, 2, 3 and 4 weeks. Brain damage was defined using a scale of 0 for no damage and up to 5 for more than 50% cellular loss. The 1- and 2-week-old mice had higher baseline serum NSE (s-NSE) than the other mice. After induction of SE, the 1-week-old mice did not show an increase in s-NSE, and there was no histological evidence of damage. In the older mice, NSE increased between 18.9 ± 0.8 ng/ml in 2-week-old mice (vs 11.5 ± 0.5 for the control group) and 35.8 ± 2.1 ng/ml in 3-week-old mice (vs 12.1 ± 0.8 for the control group). After SE, the s-NSE of the mice also increased from 5.4 ± 0.4 for the control group to 30.4 ± 1.3. They found that serum NSE after HIE injury correlated with histological changes and suggested that further studies to validate the use of NSE as a peripheral readout for brain injury may be warranted21.

In the current study, we were unable to replicate their study parameters and assess the presence of neuronal damage in the brain, since visualization of damage to the brain could was constrained by the limitations of equipment at our location.

Induced HIE in a rat model led to detectable increase in serum levels of NSE. Our study suggests that serum NSE may be a promising biomarker of brain damage. However, due to the limitations of our study (sample size, brain damage histological correlation with serum findings), further exploration will be needed, in varied neuropathological conditions and with larger sample sizes.