Detection of elevated levels of taurine and creatine in the brain of hypoxic-ischemic newborn rat using a magnetic resonance spin echo full intensity acquired localization sequence

Introduction

Neonatal hypoxic-ischemic encephalopathy (HIE) is defined as hypoxic-ischemic brain damage (HIBD) in perinatal newborns due to asphyxia, and is one of the common neurological disorders in newborns (1,2). HIE has the potential to result in neonatal mortality as a consequence of intrapartum asphyxia. Furthermore, those who manage to survive severe cases of HIE often develop varying degrees of central nervous system abnormalities, including cerebral palsy, epilepsy, and cognitive impairment (3).

Currently, mild hypothermia therapy (MHT) could substantially reduce the mortality rate of children with HIE, as well as the incidence of adverse neurodevelopmental outcomes and cerebral palsy among survivors (4). However, the treatment time window of MHT is less than 6 hours, and the diagnostic criteria for HIE are primarily based on clinical manifestations and medical history, with no objective quantitative criteria. It is necessary to investigate objective indicators that can diagnose HIE at an early stage for active treatment within the time window as an important reference for clinical diagnosis, treatment, and prognosis, which will greatly improve the survival rate of children and reduce the disability rate.

The technique of ¹H-magnetic resonance spectroscopy (¹H-MRS) is widely used in medical research to noninvasively assess the metabolism of various substances within tissues, including the brain, skeletal muscle, and prostate. Currently, the most common technique to perform ¹H-MRS acquisitions is the use of the point resolved selective spectroscopy (PRESS) sequence. It has a higher signal-to-noise ratio (SNR) in comparison to stimulated echo acquisition mode (STEAM) sequences. Additionally, the PRESS sequence demonstrates a reduced susceptibility to motion artifacts (5). Nevertheless, the longer duration of the echo time (TE) in the PRESS sequence is not favorable to effectively detecting metabolites with short T2 time, such as inositol (Ins), taurine (Tau), glutamine + glutamate (Glu + Gln), and so on.

Spin echo full intensity acquired localization (SPECIAL) sequence, developed by Mlynárik et al., contains both the high SNR of the PRESS sequence and the short TE advantage of the STEAM sequence. It is based on spin echo (SE) sequences combined with one-dimensional (1D) image-selected in vivo spectroscopy (ISIS) sequences (6,7), and can be used to detect more kinds of metabolites at the same time with improved SNR and resolution, reduced specific absorption rate (SAR), and increased bandwidth. Mekle et al. (7) effectively quantified metabolites in the human brain utilizing the SPECIAL sequence on 3- and 7-T, obtaining the spectra with a higher SNR and resolution, which enhanced the precision of quantitative analysis. Cudalbu et al. (8) evaluated the feasibility of quantifying macromolecules in the nervous system utilizing the SPECIAL sequence. Heo et al. (9) and Saleh et al. (10) conducted quantitative analysis and reproducibility studies on specific macromolecular metabolites in the brain utilizing SPECIAL sequences. However, as we know, there are relatively few studies that have investigated the metabolic changes in the brains of HIE rats using SPECIAL sequences (11,12).

The aim of this study was (I) to investigate the feasibility of brain magnetic resonance (MR) spectra in newborn rats with HIE using the SPECIAL sequences. (II) To assess and compare the parameters of spectra quality, including SNR, resolution, and fitting error of metabolites [standard deviation (SD)], of the SPECIAL sequence and the PRESS sequence in the same volume of interest (VOI). (III) Furthermore, to explore and compare the abnormal concentration changes in the brain area corresponding to the ligation side acquired from SPECIAL and PRESS spectra. The relationship among metabolites in the brain of newborn HIE rats by the two kinds of spectra was also detected and compared. This investigation provides significant implications for the development of reliable biomarkers that may help in the diagnosis and treatment of HIE in early stage. We present this article in accordance with the ARRIVE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0074/rc).

Methods

Subjects

All animal experiments were performed under a project license (No. sumc2015-078) granted by the Animal Experimentation Ethics Committee of Shantou University Medical College, in compliance with institutional guidelines for the care and use of animals. A protocol was prepared before the study without registration. Six male Sprague-Dawley neonatal rats, each aged 7 days with their weights ranged from 10 to 13 g, were used in this study. The rats were obtained from the Animal Experimentation Ethics Committee of Shantou University Medical College and were confirmed to be in a clean condition. The lighting conditions in the animal room undergo a cyclical pattern, transitioning between periods of darkness and light every 12 hours. The relative humidity within the room remains constant at a level of 55%, while the temperature is maintained at approximately 22 degrees Celsius.

HIE animal model production

The classic Rice-Vannucci model was used in this study (13-15), which involves the permanent closure of the carotid artery followed by exposure to hypoxia as an approach to induce brain injury. The newborn rats were administered anesthesia by being placed in an anesthetic box and exposed to isoflurane gas. After the anesthesia was successfully induced, the rats were taken out of the box, and the left carotid artery was ligated. The model preparation process was efficiently conducted within a time limit of five minutes in order to reduce the potential pain caused by trauma. Subsequently, the neonatal rats were introduced inside a sealed glass box wherein a gas mixture consisting of 8% oxygen and 92% nitrogen, maintained at a flow rate ranging from 1.5 to 2.5 L/min, was administered. Following a duration of 2 hours, the normal oxygen supply was restored.

Image acquisition

This study used an Agilent small animal MR imaging (MRI) system (7-T/160/AS, Agilent Technologies, Santa Clara, CA, USA) for data acquisition. Newborn rats were anesthetized by introducing 2.2% inhaled gas anesthetic isoflurane (R510-22, Reward, Shenzhen, China) under the condition of oxygen circulation through the anesthesia mask. The anesthetized newborn rats in the supine position in the sample feeder. Rats were anesthetized by 1.5% isoflurane in oxygen gas, and its’ physiological state were monitored by respiratory waveforms during the whole experimental procedure under a MR-compatible small animal monitoring and gating system (Model 1030, SAII, Stony Brook, NY, USA). Brain T2-weighted imaging was performed using the fast SE (FSE) sequence with a two-channel orthogonal volume coil. The scan parameters of the FSE sequence were as follows: repetition time (TR), 3,000 ms; effective TE, 10 ms; matrix, 192×128; field of view (FOV), 35 mm × 30 mm; and slice thickness, 2 mm.

A temporomandibular surface coil (Mercer Medical, Rochester, MN, USA) was used to perform PRESS and SPECIAL MRS. Two VOIs were placed in the left and right hippocampi and thalamus, respectively (Figure 1A,1B). Three-dimensional (3D) shim and manual shim are used for keeping the full width at half maximum (FWHM) of water less than 10 Hz. The MR spectra of two sequences were acquired in the same VOI. The position of the water peak was automatically aligned, and the VAPOR method was used to suppress the water.

Figure 1 T2WI and pathological findings of HIE rats. The location of VOI of MR spectra in rat brain in axial image (A) and sagittal image (B). The green boxes indicate VOI placement. (C,D) Two typical hippocampal HE staining of a rat model of HIE. (C) Left hippocampus; (D) right hippocampus corresponding to the lighten side. The neuronal cells on the ligation side are wrinkled, the neural fiber network is obviously widened, and the space of glial cells and micro-vessels is obviously widened. HE, hematoxylin and eosin; HIE, hypoxic-ischemic encephalopathy; MR, magnetic resonance; T2WI, T2-weighted imaging; VOI, volume of interest.

The PRESS sequence was conducted with the following parameters: a TR of 3,000 ms, TE of 13.33 ms, number of repeated acquisitions of 256, empty scans of 2, a VOI of 3 mm × 3 mm × 4 mm, and a total acquisition time of 12 minutes and 53 seconds. Following the scanning process, the spectra without and with water suppression were acquired, respectively.

The parameters of the SPECIAL sequence are as following: TR of 3,000 ms, TE of 5 ms, mixing time (TM) of 10 ms, 2 dummy scans, spectral width of 4,006 Hz, outer volume suppression, and a VOI of 3 mm × 3 mm × 4 mm. The total count of repetitions in the study was 256, while the duration of the scanning process amounted to 12 minutes and 53 seconds. The MR spectra of the areas of interest with and without water suppression were also acquired.

Spectra post-processing

The MR spectra were processed, and metabolite concentrations were quantified using LCModel (version 6.3, LCModel Inc., Oakville, ON, Canada). The basis set of a 7-T-PRESS-TE13 was selected for the PRESS sequence. A simulated 7-T-PRESS-TE0 acquired from Dr. Provencher was used for the fitting of the SPECIAL spectra. Water scaling and eddy current correction (ECC) were also used in LCModel processing (16,17).

The FWHM and the SNR of SPECIAL and PRESS spectra were provided by LCModel software. The concentrations and SD of metabolites, include N-acetylaspartate + N-acetylaspartylglutamate (NAA + NAAG), Glu and Glu + Gln, creatine + phosphocreatine (Cr + PCr), glycerophosphocholine + phosphocholine (GPC + PCh), lactate + macromolecule methyl at 1.4 ppm + lipid methyl at 1.3 ppm (Lac + MM14 + Liq13), Ins, and Tau, were calculated by LCModel software. The spectra with SNR greater than 5 and the metabolites with fitting error (SD) greater than 30% were selected for analysis to ensure the reliability and accuracy of the fitting results.

Pathological examination

The animals were sacrificed immediately after the completion of MR scanning. Following heart perfusion, the brain was extracted and subsequently immersed in a 4% formaldehyde solution for a duration of 48 hours, and then the olfactory brain and cerebellum were removed. Taking the posterior fontanel as the baseline, five sections with a thickness of 2 mm were obtained, extending from the foot side to the head side, corresponding to the MRI scanning layer as far as possible. HE staining was carried out to determine the presence of neuronal injury in the cortex and hippocampus. Histopathological evaluation was performed on hematoxylin and eosin (HE)-stained coronal brain sections using a semi-quantitative grading system adapted from established criteria for hypoxic-ischemic brain injury. The severity of neuronal damage was assessed according to the extent of neuronal necrosis, karyopyknosis, cytoplasmic vacuolization, and parenchymal disruption.

Statistical analysis

The data was evaluated by the SPSS 19.0 statistical analysis software. Firstly, the Shapiro-Wilk test was used to detect whether the normal distribution of data acquired by the SPECIAL or PRESS sequence on the left or right side of the brain region is satisfied. Square root transform was used if the data did not satisfy the normal distribution. If satisfy the normal distribution, the data were expressed using mean ± SD, and independent sample t-tests were used to detect the differences in FWHM, SNR, and the fitting SD in metabolites from the two kinds of spectra in different brain areas. Otherwise, the data were shown as median (25^th percentile, 75^th percentile), and the Mann-Whitney U test was used to compare between-group differences. Subsequently, paired t-tests were used to assess the differences in metabolite concentration between the bilateral brain regions acquired from two kinds of MR spectra, respectively. At last, the correlation among the concentration of metabolites from SPECIAL and PRESS spectra was analyzed separately using the Pearson test. P<0.05 is considered as statistically significant.

Results

Differences of metabolite concentrations in bilateral hippocampus and thalamus acquired from SPECIAL and PRESS sequence

HE staining confirmed the presence of neuronal damage in this side compared to the contralateral brain area (Figure 1C,1D). SPECIAL spectra with higher resolution were acquired in hippocampal and thalamic regions corresponding to the ligation side compared with PRESS spectra (Figure 2A-2D). Elevated concentrations of Tau and Cr + PCr were found in hippocampal and thalamic regions corresponding to the ligation side based on SPECIAL spectra (P<0.05). Meanwhile, increased Cr + PCr and Lac + MM14 + Liq13, as well as decreased NAA + NAAG, were observed by PRESS spectra (P<0.05) (Table 1).

Figure 2 Typical MR SPECIAL and PRESS spectra of bilateral hippocampal and thalamus in a HIE rat brain. The PRESS spectra in left hippocampus (A) and in right hippocampus corresponding to the lighten side (B). The SPECIAL in left hippocampus (C) and in right hippocampus corresponding to the lighten side (D) with higher spectral resolution than PRESS spectra (B). HIE, hypoxic-ischemic encephalopathy; MR, magnetic resonance; PRESS, point resolved selective spectroscopy; SPECIAL, spin echo full intensity acquired localization.

Differences of FWHM, SNR, and SD of metabolites from SPECIAL and PRESS sequence

Table 2 shows that the SPECIAL spectrum had a lower FWHM in the brain region corresponding to the ligation side than the PRESS spectra (0.03±0.00 Hz in SPECIAL vs. 0.06±0.01 Hz in PRESS, P<0.05). There are no statistic differences of SNR from two kinds of spectroscopy, but the fitting error of Glu and Glu + Gln in the brain region corresponding to the ligation side and the fitting error of NAA + NAAG in the contralateral side by SPECIAL spectra are higher than those by PRESS spectra (P<0.05).

Relationship among metabolite concentrations

More correlations between metabolites were found in SPECIAL spectra. The levels of Cr + PCr were significantly positively correlated with Tau (r=0.7, P<0.01), Glu + Gln (r=0.75, P<0.01), GPC + PCh (r=0.63, P<0.01), and Lac + MM14 + Liq13 (r=0.54, P=0.03), while no correlation of the above metabolites was found in PRESS spectra (Figure 3).

Figure 3 Correlations between metabolites in SPECIAL spectra and PRESS spectra. Tau and Cr + PCr in SPECIAL (A), Glu + Gln and Cr + PCr in SPECIAL (B), GPC + PCh and Cr + PCr in SPECIAL (C) and Lac + MM14 + Liq13 and Cr + PCr in SPECIAL (D). Tau and Cr + PCr in PRESS (E) and Glu + Gln and Cr + PCr in PRESS (F). Cr + PCr, creatine + phosphocreatine; Glu + Gln, glutamine + glutamate; GPC + PCh, glycerophosphocholine + phosphocholine; Lac + MM14 + Liq13, lactate + macromolecule methyl at 1.4 ppm + lipid methyl at 1.3 ppm; PRESS, point resolved selective spectroscopy; SPECIAL, spin echo full intensity acquired localization; Tau, taurine.

A total of six neonatal male rats were included in this study, and histopathological brain injury was assessed via HE staining. Representative photomicrographs revealed that all animals exhibited moderate to severe brain injury, characterized by disrupted neuronal arrangement, karyopyknosis, cytoplasmic vacuolization, and tissue rarefaction.

Discussion

This study used PRESS and SPECIAL sequences to acquire hippocampus spectra from HIE rats, followed by quantitative analysis of metabolite concentrations and spectral quality factors. The study identified distinct spectra of fetal rat brain samples, referred to as SPECIAL spectra, which showed superior line resolution compared to the PRESS spectra. Furthermore, the SPECIAL spectra had greater levels of Tau, Cr + PCr, and Lac + MM14 + Liq13, with statistically significant differences. The SNR of the PRESS sequence exhibited a higher value, allowing the detection of reduced levels of NAA + NAAG, as well as increased levels of Cr + PCr. These findings offer potential for the early assessment of HIE-related metabolic changes.

Tau, a protein that is present in high quantities in the developing brain, has inhibitory effects on the release of norepinephrine and acetylcholine, which are excitatory neurotransmitters. Additionally, tau shows antioxidant properties. It is postulated that the compensatory upregulation of Tau may also be associated with the heightened neuronal activity observed in the impacted region of the brain. Li et al. reported that in newborns with HIE, elevated plasma levels of Tau are linked to death or severe brain damage as well as faulty cerebral autoregulation (18). McGowan et al. also found that tau was the most reliable indicator of unfavorable neurological outcomes, especially when evaluated during or following rewarming (19). This study observed an increase in Tau levels within the SPECIAL sequences, which is consistent with previous studies. Moreover, it is more direct evidence in detecting Tau changes in lesion brain region than in the blood index.

The current study also observed that both SPECIAL and PRESS sequences exhibited increased concentrations of Cr + PCr on the injured side. Creatine is a promising neuroprotective substance. The compound Cr + PCr is closely associated with the process of energy metabolism and has the capacity to supply energy to nerve cells. An elevated concentration of Cr + PCr indicates elevated energy metabolism on the afflicted side, suggesting its potential use as a biological indicator for monitoring subcooling treatment. Tran et al. assessed the preclinical animal research on creatine’s role in prenatal neuroprotection, and reported that higher increases in creatine/phosphocreatine in HIE animals compared to adolescent-adult equivalent studies (20). The results of this study are consistent with the above conclusions.

The PRESS sequences used in this experiment revealed a decline in NAA, consistent with findings from other research. NAA is a unique metabolite found in neuronal cells. It serves as a marker within neurons, indicating their functional status and density. A decline in NAA levels indicates potential shortages or problems in neural synthesis (5).

Lac is indicative of an elevation in anaerobic metabolic activity occurring inside the tissue. Normal brain tissue is incapable of detecting it. In instances of cellular hypoxia, there is a notable impairment in mitochondrial metabolism, leading to a reduction in adenosine triphosphate (ATP) synthesis. Consequently, the cell’s energy production is primarily sustained by the anaerobic glycolysis of glucose, which serves to maintain metabolic processes. This metabolic shift ultimately culminates in the buildup of Lac (5).

In this experimental study, after the process of modelling, prominent Lac peaks were observed at 1.3 ppm on the spectra of the two sequences in the brains of four HIE mice. This finding suggests a significant buildup of lactic acid inside the examined subjects. The sensitivity of SPECIAL spectra to variations in Lac is greater compared to that of PRESS spectra. Nevertheless, due to the overlapping of the Lac peak with Lip13 and MM14 at TE 15 ms, the LCModel is unable to effectively distinguish between them. Therefore, in this study, the combination of Lac + MM14 + Liq13 is used to reflect changes in Lac. However, the presence of a Lac peak was not seen in two cases. A possible reason is because previous research has indicated that the closure of the unilateral common carotid artery mostly results in chronic harm, possibly due to the compensating effect caused by other blood arteries, such as the vertebral artery. The specific region affected by this injury is located at the confluence of grey and white matter.

According to the reference, Glu is classified as an excitatory amino acid. The present investigation observed that the levels of Glu and Glu + Gln in the two sequences exhibited a changing trend, but no statistically significant difference was seen. Furthermore, the changing trends of the two sequences were found to be inconsistent. The previous results may be attributed to the specific time point chosen for spectra scanning. Several studies have indicated that there is an initial rise in Glu concentration during hypoxia (21), which is then followed by a decline starting at the 3-hour timepoint. However, it is necessary to validate the accuracy and reliability of Glu detection by MR sequence and subsequent data post-processing. The findings of this study indicate that the reduced fitting error of Glu and Glu + Gln with PRESS was found compared to SPECIAL; however, a decline levels of Glu and Glu + Gln in PRESS spectra is inconsistent with previous report (21). The correlations between glutamate and other metabolites with SPECIAL spectra were found suggesting the data from SPECIAL spectra is more reliable.

This study focused on the application of MRS in HIE assessment. A fundamental limitation of our approach is that MRS detects small metabolites (e.g., NAA, Lac) at millimolar concentrations but cannot access protein-based biomarkers (e.g., NSE, VEGF, CK, SOD) lacking specific MRS signals, making them invisible to this technique even with high-field optimization.

However, MRI encompasses multiple additional sequences that hold significant value for HIE diagnosis, including diffusion-weighted imaging (DWI) (22), diffusion tensor imaging (DTI), arterial spin labeling (ASL), and susceptibility-weighted imaging (SWI). In subsequent investigations, we will integrate these complementary sequences to establish a more comprehensive imaging framework for HIE, thereby refining diagnostic precision and enhancing the reliability of prognostic evaluation.

This study has the following shortcomings: (I) the animals used in this study are neonatal rats aged 7 days with a small number of samples, so it is necessary to further expand the sample size to obtain more accurate results. (II) Neonatal Sprague-Dawley rats are collected under anesthesia during MR scanning. Some studies suggest that the use of anesthetic drugs may affect cerebral blood flow and metabolite concentration (23). (III) The fitting base-set of a PRESS sequence with TE of 0 ms was used in this study for the fitting and quantification of SPECIAL spectra with LCModel software. The fitting accuracy may be further improved by using a base set of SPECIAL spectra with TE of 5 ms.

Conclusions

This study validated the feasibility of using the SPECIAL sequence to detect brain metabolic changes in neonatal HIE models. It was found that the SPECIAL sequence has high spectral line resolution and is more sensitive to changes in Lac, Cr + PCr, and Tau. It can be used as an early brain metabolic indicator for HIE and has potential clinical value in guiding the diagnosis and treatment of HIE in clinical practice.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the ARRIVE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0074/rc

Data Sharing Statement: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0074/dss

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0074/prf

Funding: This research was funded by the Natural Science Foundation Project of Guangdong Province, China (No. 2017A030313718).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0074/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All animal experiments were performed under a project license (No. sumc2015-078) granted by the Animal Experimentation Ethics Committee of Shantou University Medical College, in compliance with institutional guidelines for the care and use of animals.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. She HQ, Sun YF, Chen L, et al. Current analysis of hypoxic-ischemic encephalopathy research issues and future treatment modalities. Front Neurosci 2023;17:1136500. [Crossref] [PubMed]
2. Hoiland RL, Rikhraj KJK, Thiara S, et al. Neurologic Prognostication After Cardiac Arrest Using Brain Biomarkers: A Systematic Review and Meta-analysis. JAMA Neurol 2022;79:390-8. [Crossref] [PubMed]
3. Lambing H, Gano D, Li Y, et al. Using Neonatal Magnetic Resonance Imaging to Predict Gross Motor Disability at Four Years in Term-Born Children With Neonatal Encephalopathy. Pediatr Neurol 2023;144:50-5. [Crossref] [PubMed]
4. Ji X, Zhou Y, Gao Q, et al. Functional reconstruction of the basal ganglia neural circuit by human striatal neurons in hypoxic-ischaemic injured brain. Brain 2023;146:612-28. [Crossref] [PubMed]
5. Lee JW, Sreepada LP, Bevers MB, et al. Magnetic Resonance Spectroscopy of Hypoxic-Ischemic Encephalopathy After Cardiac Arrest. Neurology 2022;98:e1226-37. [Crossref] [PubMed]
6. Mlynárik V, Gambarota G, Frenkel H, et al. Localized short-echo-time proton MR spectroscopy with full signal-intensity acquisition. Magn Reson Med 2006;56:965-70. [Crossref] [PubMed]
7. Mekle R, Mlynárik V, Gambarota G, et al. MR spectroscopy of the human brain with enhanced signal intensity at ultrashort echo times on a clinical platform at 3T and 7T. Magn Reson Med 2009;61:1279-85. [Crossref] [PubMed]
8. Cudalbu C, Mlynarik V, Xin L, et al. Quantification of in vivo short echo-time proton magnetic resonance spectra at 14.1 T using two different approaches of modelling the macromolecule spectrum. Meas Sci Technol 2009;20:104034.
9. Heo H, Kim S, Lee HH, et al. On the Utility of Short Echo Time (TE) Single Voxel 1H-MRS in Non-Invasive Detection of 2-Hydroxyglutarate (2HG); Challenges and Potential Improvement Illustrated with Animal Models Using MRUI and LCModel. PLoS One 2016;11:e0147794. [Crossref] [PubMed]
10. Saleh MG, Near J, Alhamud A, et al. Reproducibility of macromolecule suppressed GABA measurement using motion and shim navigated MEGA-SPECIAL with LCModel, jMRUI and GANNET. MAGMA 2016;29:863-74. [Crossref] [PubMed]
11. van de Looij Y, Chatagner A, Quairiaux C, et al. Multi-modal assessment of long-term erythropoietin treatment after neonatal hypoxic-ischemic injury in rat brain. PLoS One 2014;9:e95643. [Crossref] [PubMed]
12. Sanches EF, Van de Looij Y, Toulotte A, et al. Brain Metabolism Alterations Induced by Pregnancy Swimming Decreases Neurological Impairments Following Neonatal Hypoxia-Ischemia in Very Immature Rats. Front Neurol 2018;9:480. [Crossref] [PubMed]
13. Vannucci RC. Cerebral carbohydrate and energy metabolism in perinatal hypoxic-ischemic brain damage. Brain Pathol 1992;2:229-34. [Crossref] [PubMed]
14. Shi Z, Luo K, Jani S, et al. Mimicking partial to total placental insufficiency in a rabbit model of cerebral palsy. J Neurosci Res 2022;100:2138-53. [Crossref] [PubMed]
15. Shi Z, Vasquez-Vivar J, Luo K, et al. Ascending Lipopolysaccharide-Induced Intrauterine Inflammation in Near-Term Rabbits Leading to Newborn Neurobehavioral Deficits. Dev Neurosci 2018;40:534-46. [Crossref] [PubMed]
16. Provencher SW. Automatic quantitation of localized in vivo 1H spectra with LCModel. NMR Biomed 2001;14:260-4. [Crossref] [PubMed]
17. Pfeuffer J, Tkác I, Provencher SW, et al. Toward an in vivo neurochemical profile: quantification of 18 metabolites in short-echo-time (1)H NMR spectra of the rat brain. J Magn Reson 1999;141:104-20. [Crossref] [PubMed]
18. Li R, Lee JK, Govindan RB, et al. Plasma Biomarkers of Evolving Encephalopathy and Brain Injury in Neonates with Hypoxic-Ischemic Encephalopathy. J Pediatr 2023;252:146-153.e2. [Crossref] [PubMed]
19. McGowan MM, O'Kane AC, Vezina G, et al. Serial plasma biomarkers of brain injury in infants with neonatal encephalopathy treated with therapeutic hypothermia. Pediatr Res 2021;90:1228-34. [Crossref] [PubMed]
20. Tran NT, Kelly SB, Snow RJ, et al. Assessing Creatine Supplementation for Neuroprotection against Perinatal Hypoxic-Ischaemic Encephalopathy: A Systematic Review of Perinatal and Adult Pre-Clinical Studies. Cells 2021;10:2902. [Crossref] [PubMed]
21. Suzuki T, Sato Y, Kushida Y, et al. Intravenously delivered multilineage-differentiating stress enduring cells dampen excessive glutamate metabolism and microglial activation in experimental perinatal hypoxic ischemic encephalopathy. J Cereb Blood Flow Metab 2021;41:1707-20. [Crossref] [PubMed]
22. Ambwani G, Shi Z, Luo K, et al. Distinguishing Laterality in Brain Injury in Rabbit Fetal Magnetic Resonance Imaging Using Novel Volume Rendering Techniques. Dev Neurosci 2025;47:55-67. [Crossref] [PubMed]
23. Du F, Zhang Y, Iltis I, et al. In vivo proton MRS to quantify anesthetic effects of pentobarbital on cerebral metabolism and brain activity in rat. Magn Reson Med 2009;62:1385-93. [Crossref] [PubMed]