Changes in Brain Electrical Activity after Transient Middle Cerebral Artery Occlusion in Rats

Round 1

Reviewer 1 Report

This is a well written paper on the use of EEG in experimental stroke research in the rat model. Introduction and methods are well done, results are well described, discussion is adequate. The paper shows that EEG data are appropriate to provide additional insight in the process of functional recovery in the MCA-occlusion in rats. It remains to be clarified whether this additional information is related to clinical recovery and responsiveness to medication.

Author Response

It should be noted that in this paper, we focused on the use of ECoG for experiments in rodents. The observed changes in the BEA are specific to rats (and most likely, to other rodents), but may not be relevant for humans. This may be due to the fact that rats and humans may have different ranges of EEG rhythms; for example, it was shown that rodent θ (6-10 Hz) is an overlapping but overall faster frequency range than human θ (4-7 Hz) [Watrous, 2013]. In addition, electrode characteristics (e.g., material, size), reference electrode location, recording device parameters, and other methodological factors play an essential role in the amplitude-spectral characteristics of ECoG in rats. Despite the fact that attempts are being made to standardize EEG studies in rodents [Maheshwari, 2020], the obtained results often do not correlate not only with data from humans, but also among rodent studies performed by different laboratories. Nevertheless, if, in the future, the method we have described could indicate the ability of a potential neuroprotective agent to ameliorate pathological BEA changes in experimental stroke, as shown earlier for TBI [Brain Sciences, 2021], it would possibly predict positive effects of said agent in human patients as well.

Reviewer 2 Report

 

The authors used EEG to monitor the brain activity changes after the transient middle cerebral artery occlusion and concluded that IS can be represented by the dynamics Theta rhythm. EEG is a low cost and portable technique and can be used for studying brain functionality. This study provides an innovative approach for assessing brain status following ischemic stroke. A few comments may need to be addressed.

Comments:

1) Please add scale bars in Figure 1.

2) Figures 4, 5, and 6 used rhythm indices, mean amplitude, mean power of different waves, respectively, for analysis. However, the heatmaps look similar. Any redundancy in the analysis? Can the authors briefly state what the advantages for each metric are?

3) It is hard to visualize the variations in the heatmaps. Is it possible to provide a few representative examples of bar plots with mean and standard deviations? For instance, FP1 Theta of IS 30 vs. that of control.

 

Author Response

1) Scale bars have been added

2) Undoubtedly, all these indicators are interconnected, as we wrote in the discussion section: «This similarity is quite understandable, since all three parameters are closely interrelated, and an increase in δ rhythm index seems unlikely in the absence of at least some increase in its mean amplitude». Nevertheless, in our previous work [Brain Sciences, 2021], mean rhythm amplitudes were shown to re-spresent the general pattern of ongoing processes in the brain during traumatic brain injury (a decrease in the mean amplitudes of θ-, α-, and β-rhythms, regardless of the lead). For the rhythm indices, distinct changes were observed in the injured hemi-sphere (leads FP1, C3, and O1), while in the healthy hemisphere they changes were mild, and it could be concluded that they almost do not occur, if at all, in the absence of any pathological changes. Therefore, in this study, we focused on the changes in the three parameters mentioned above, however unfortunately, unlike the case with TBI, we did not observe different heatmap patterns here.

3) We added all results as data tables in supplementary files (S1, S2).

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The paper presents a study on  EEG alteretion after ischemia induce by  closure of the middle cerebral artery in rat models. Discussion can be improved, disclosure of the methods used is weak.

Transient cerebral ischemia->  this term is stressed in the paper. However, transient closure of MCA lasts 30 and 45 minutes. Can we be sure that the ischemia is transient (as the transient ischemic attack in men, which includes no MRI lesions)? I am not an expert on animal models, but I find it very likely that transient closure of this duration might cause ischemic lesions. Thus i suppose that this term, even used in the title, might be wrong.
 
Spectral analysis-> description of neurophysiological signal analysis methods is not sufficient. Did u use the welch method, what window, and the sampling rate? Was the signal preprocessed for artifact removal? Which filters did you use? There are many premises that glorify EEG in animal models, but then you do not make your methods explicit.
 
Statistical analysis-> really short and non-sufficient.
 

Figures 2,3and 4, the authors do not specify the units of measure. Which unit of measure was used  power spectrum/relative to power, or else ?? Peanuts?

Figure 5 how was statistic analysis done? Did the authors use multiple sample corrections? 

 

"In addition, in patients with anterior cerebral artery territory infarcts, the Brain symmetry index values (representing the differences in mean signal spectral powers in the 1-25 Hz band between the two hemispheres) are highly correlated with the National Institutes of Health stroke scale (NIHSS) scores [29]. Another study [30] showed that acute delta change index (aDCI) values have a stronger correlation with NIHSS scores 30 days after IS than do infarct volumes seen on diffusion-weighted MRI in the early phase of stroke"

I would suggest expanding this part of the discussion by adding considerations on the use of new ways to parameterize the spectrum since correlation is not only between delta and NIHSS but also between the power-law shape of the spectrum and nihss and is the pendence of the PSD slope that seems to be responsible for the significant differences in delta band in stroke patients. Here are some relevant papers.

EEG in rodents and spectral parameterization: "Rodent EEG: Expanding the Spectrum of Analysis."

The exponent in humans and NIHSS"EEG spectral exponent as a synthetic index for the longitudinal assessment of stroke recovery"

importance of 1/f dynamics: "Inferring synaptic excitation/inhibition balance from field potentials"

Minor points

"The main difficulty that they pose to
researchers are the need for surgical implantation of EEG/ECoG electrodes, which usually increases the time and labor intensity of the study, and also limits the number of experimental animals. In addition, the obtained neurophysiological data are complex and quite underexplored, and may therefore require high levels of skill and expertise in order to be interpreted correctly" --> I would tune down this section. It seems a bit self-congratulatory.
 

 

Reviewer 2 Report

This manuscript was aimed to detail brain electrographic signals following brain ischemia in a rat model. The idea/theoretic consideration behind this study is excellent, but experiments in characterization of brain ischemia in their model appeared to be weak or not thorough. As MCAO is a well described model, detailed assessments of the rat model of MCAO (brain injury) may be essential for charactering associated changes in EEG signals.

 

Specific
Line numbers may be provided to facilitate review.

Page 4, Electrocorticogram acquisition and analysis

It is mentioned that “Corticogram fragments of up to 5 min long which corresponded to awake, resting state with no locomotion, exploratory and/or grooming behavior were analyzed….”. More information may be provided for the “5 mim data segment”. For example, did rats reach a stable sleep or immobility for a certain length of time before the “5 min”? In which time periods of day (morning or afternoon) the 5-min segment was chosen?

It is mentioned that “Cortical electrical activity was recorded at a 0.5-35 Hz bandwidth and a 500 Hz sampling rate. However, frequency signals were analyzed in bandwidth up to 20-35 Hz. Can input frequency band of 0.5-35 Hz allow analysis of rhythmic signals of up to 35 Hz?

It is mentioned that “amplitude, spectral, and cross-correlation analyses were carried out, and group means were compared for each parameter.” However, no numeric data was presented in Results about these analyses.

Results

Numbers of ischemic rats examined, including those experienced successful or non-successful ischemic episodes, should be clearly indicated.

How did the authors verify/quantify ischemic brain injury in individual rats? When examined 7 days post ischemia, infarction should be clearly recognizable in targeted hemisphere. However, no data from brain histology was provided in this manuscript. Monitoring acute changes of brain ischemia, such as changes in blood flow and/or EEG signals during and shortly after MCAO, were not mentioned. Therefore, it is unclear whether individual rats experienced successful an ischemic episode.