An Early Detection of Acute Compartment Syndrome in Fastened Zip-Tie Rat Model Using Dynamic Phosphorous-31 Magnetic Resonance Spectroscopy at 9.4 Tesla

Acute compartment syndrome (ACS) leads to a series of health problems, and even In vivo phosphorus-31 magnetic resonance spectroscopy ( 31 P-MRS) provides a unique non-invasive method to assess skeletal muscle metabolisms such as inorganic phosphate (Pi), phosphocreatine (PCr), and adenosine triphosphate (ATP). The study aims to assess the ability of dynamic 31 P-MRS in the early detection of muscular damage in ACS. Materials & Methods The study induced the fastened zip-tie model of ACS on normotensive Sprague-Dawley rats (n = 6). The spectra were acquired in Bruker 9.4-Tesla preclinical scanner using 1 H/ 31 P surface coil. 31 P-MRS spectra and blood samples were obtained at time 0 (pre-ischemic phase) and every 15 minutes during the compression (120 minutes) and the reperfusion phase (90 minutes). 31 P-MRS spectra ndings were compared with plasma creatine phosphokinase (CPK). of ACS. This technique is a promising non-invasive method to detect the early ischemic muscular damage in ACS.


Introduction
Acute compartment syndrome (ACS) of the extremities can damage the muscles, nerves ,and vasculature [1,2]. ACS occurs more frequently in disaster-prone countries [3]. The time to decompressive fasciotomy is crucial; if not treated within 6-8 hours, the limb will lose its viability, limb salvage, and the patient will be even death [4,5]. Thus, monitoring the disease progression that involves both ischemic and recovery is essential to evaluate skeletal muscle status and differentiate reversible and irreversible tissue damage of skeletal muscles. However, detecting the early ischemic muscular damage in ACS remains a di cult task. While there are several different diagnostic methods, including pressure levels, biomarkers, and imaging techniques, the conventional diagnosis techniques provide insu cient evidence to diagnose a patient's condition is most cases [6]. It is currently in the stage of research and development. For early diagnosis and early treatment of ACS, it is necessary to evaluate muscle bioenergetics and metabolism in vivo non-invasively [6][7][8].
One key parameter used in evaluating ACS signs is skeletal muscles' energy metabolism [9]. Magnetic resonance spectroscopy (MRS) provides a unique technique that allows for assessing tissue metabolic properties. Phosphorous-31 MRS ( 31 P-MRS) offers the ability to directly quantify essential metabolites such as inorganic phosphate (Pi), phosphocreatine (PCr), and adenosine triphosphate (ATP). More speci cally, 31 P-MRS can dynamically monitor high-energy phosphate depletion and re-synthesis rates during ischemic to reperfusion state. In addition, it can measure the intracellular pH and mitochondrial oxidative capacity. [9,10] These are valuable indices used to differentiate between normal and pathological states.
While most ACS cases have been scanned at clinical Magnetic Resonance Imaging (MRI) machine, we used a pre-clinical ultra-high eld (9.4 T) to gain a higher signal-to-noise ratio (SNR), shorten data acquisition time, improve spectra quality, and reduce metabolite overlapping. Therefore, the quanti cation of energy metabolism is more accurate. [11][12][13] The use of serum biomarkers in ACS has several limitations that have not shown speci city in detecting ACS's early symptoms [7,14,15]. Creatinine phosphokinase (CPK), known as CK (creatinine kinase), is an enzyme present in various tissues and cell types such as skeletal muscle, myocardium, and brain. The elevated CPK level frequently occurs in heart attack, skeletal muscle injury, and exhausting exercise.
According to the previous study, CPK concentrations greater than 4,000 U/L are associated ACS development in humans [16]. It has been suggested that measurement of CPK could be a potential biomarker to indicate ACS [17].
This study aims to examine whether dynamic 31 P-MRS will provide early detection of skeletal muscle bioenergetic changes in ACS. In the present experiment, we acquired 31 P-MRS spectra, Hydrogen Magnetic Resonance Imaging's image ( 1 H-MRI), and blood samples from the fastened zip-tie rat model of ACS. Comparison of 31 P-MRS with blood test will offer essential information about muscle pathophysiology from pre-ischemic to ischemic and reperfusion phase.

Animals
Male normotensive Sprague-Dawley rats (n = 6; 9 to 12 weeks old; 276-499g) obtained from Nihon SLC (Japan SLC, Inc. Shizuoka, Japan) were housed in individual cages in a controlled room (temperature: 24-(CE-2, CLEA Japan, Inc., Tokyo, Japan) and water. The research protocol was approved by the Institutional Animal Care and Use Committee of The Jikei University School of Medicine (Protocol number: 2019-043C1). All experimental procedures were conducted under the Fundamental Guidelines for Proper Conduct of Animal Experiments and Related Activities in Academic Research Institutions issued by the Japanese Ministry of Education, Culture, Sports, Science and Technology [18].
The fastened zip-tie rat model of ACS procedure A vascular surgeon performed the fastened zip-tie rat model of ACS. Anesthesia was induced using 3% iso urane kept within the laboratory "Small Animal Anesthesia System" (SBN-487, Shinano, Tokyo, Japan) and titrated to maintain an acceptable standard during the experiment. This iso urane concentration is commonly used for MR procedures without impacting the results obtained. Respiratory was monitored in real-time using a Magnetic Resonance (MR)-compatible Small Animal Monitoring & Getting System (SA Instruments, Inc.,10, NY, USA). First, the rats were placed on a heating pad in a supine position and shifted into a lateral position. Then, two plastic zip-ties (ELPA, Osaka, Japan; length: 150mm, width: 3.6mm) were held with three xed positions on the rat's inguinal region (3 xed points were sewn directly around the inguinal region avoiding vascular damages) ( g.1). After the pre-ischemic scanning, the ischemic was induced by fastening two zip ties. Pre-ischemic and ischemic states were monitored continuously from time zero to 120 minutes. After releasing the zip-ties, the reperfusion state was followed for 90 minutes.

Experimental setup
After preparation for the fastened zip-tie rat model of ACS, the rats were placed on a supine position and immobilized on the MR cradle with the hindlimb in the full-extended position. The MR cradle is connected to a small animal ventilator (SA Instruments, Inc., 10, NY, USA). MR coils were positioned on the hindlimb's lateral side and rmly secured with tape at pre-ischemic state. Ischemic was then induced by fastening the zip tie. After 120 minutes of ischemic, perfusion was resumed immediately by cutting the zip tie. 31 P-MRS spectra were acquired at rest, during muscle was compressed (120 minutes) and reperfusion (90 minutes). During the MR procedure, the concentration of pH and CPK were analyzed by blood samples obtained from the tail artery. The rat's respiration rate was continuously monitored (SA Instruments, Stony Brook, NY, USA) and controlled 80 to 120/minutes throughout the MR scanning. Body temperature was maintained at about 37 degrees by heating bath circulator equipment (CW-05G, lab perform shimming, and the knee was centered over the surface coil. The leg was secured to the 1 H/ 31 P dual coil covered by masking tape to avoid potential motion during the experiment. Localizer 1 H-images were acquired to detect the leg's position within the sensitive area of the 1 H/ 31 P dual coil. Subsequently, the wobble adjustment was performed that allows the radiofrequency (RF) coil (measuring its absorption spectra) for manual tuning/ matching. We selected two coils (elements 1 H and 31 P, respectively) with an isotropic voxel of 48 x 48 x 48 mm 3 in the surface coil's sensitive eld. The wobble curve was tuned and matched until the dip reaches the center, and its minimum is close to the zero line. After that, the B0 map, used to measure a eld map of the object used in a study to calculate shims, was set up. The shim was calculated based on a previously measured B0-eld map to optimize the eld homogeneity within the Shim volume. As a result, a proton linewidth of 120-140 Hz was obtained. When the line width at half height of the proton signal was about 0.5 ppm for one free induction decay (FID), the magnetic eld homogeneity was accepted. Then, the spectrometer was turned to 31 P nuclei. The entire procedure took about ve minutes on average. 31 P spectra were acquired, followed by T2-weighted images, leading to a total acquisition time of 10 minutes at pre-ischemic. The interleaved 31 P-MRS, T2-weighted images, and blood samples were acquired on the hindlimb of rats from pre-ischemic to ischemic and followed by reperfusion phase. The 31 P-MRS dynamic protocol consisted of 15 min for pre-ischemic, 120 min for ischemic, and 90 min for recovery phase, respectively, to quantify phosphate metabolite changes. Each acquisition of 31  Field of view = 35x50mm 2 ; Slice thickness = 1.00 mm, Slices = 18; Scanning time = 3minutes 12 seconds). The experimental procedure work ow is illustrated in g.3.
Blood sampling procedure Blood samples were collected for ex vivo analysis. The 22-gauge catheter for invasive blood sampling collection was inserted into the ventral tail artery. After sampling arterial blood at the pre-ischemic state, the heparinized physiological saline was ushed into the catheter. The catheter was kept until the end of the experiment. Each amount of blood sample (0.25mL) was obtained and lled into the i-STAT CG4 + cartridge (Abbott product, USA) and FUJI DRI-CHEM slide (Fuji lm Medical, Tokyo, Japan) blood test kit.
Blood samples were obtained at pre-ischemic, ischemic (60 minutes & 120 minutes), and after pressure removal (5 minutes, 60 minutes & 90 minutes). It was assessed at each time point for measurement of in ammatory biomarker CPK. After the needle was withdrawn, we applied gentle pressure with a cotton ball to the puncture site until the bleeding stopped.

Data Processing, Statistical Analysis
Data processing The MRS data were rst processed with TopSpin 4.0.7 software (Bruker Biospin Corp., Billerica MA). The resulting datasets were tted in the time domain using AMARES (Advanced Method for Accurate, Robust, and E cient Spectral Fitting) algorithm implemented in jMRUI software. The NMRSCOPE tool (jMRUI software package) created a basic set of ve metabolites spectra that include Pi, PCr, and ATP. PCr and Pi peaks were tted to Lorentzian line shapes, whereas µ-ATP, α-ATP, and β-ATP signals were tted to Gaussian line shapes. [19] PCr was used as an internal reference for calculating the absolute concentration of Pi and ATP [20,21]. The PCr/(Pi+PCr) ratio, a marker of energy state's level, was calculated from Pi and PCr areas [10]. The intracellular pH was calculated from the chemical shift of Pi relative to PCr utilizing the following equation [22] pH = 6.75 + log (δ-3.27/ (5.69-δ), Where δ is the chemical shift of the Pi peak in parts per million relatives to PCr.
T2-weighted images were analyzed using ImageJ software (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA) [23]. In order to quantitively analyze the signal intensity changes, the signal intensity was measured in two compartments that correspond to the tibialis anterior and the gastrocnemius muscles (locations were indicated in g.8). The signal intensities of these two regions of interest (ROI) were calculated as the mean values of all pixels within the ROI. The tibia bone's signal intensity was then used as the reference value to normalize the signal intensity. The signal intensity through each different phase from the pre-ischemic phase to the recovery phase was quanti ed.
Blood samples analysis: Blood samples were analyzed using an automatic biochemical analyzer (FUJI DRI-CHEM 3500v, Fuji lm Medical, Tokyo, Japan) and handheld blood analyzer i-STAT 1(Abbott product, USA). Results were obtained after 2 minutes. The analysis included pH and CPK [16,24]. We set all the results over 2000 U/L to 2000 U/L.

Statistical analysis
Results are presented as mean ± SD. A repeated-measures nonparametric Friedman test was used to compare 31 P-MRS and CPK between time points. Potential relationships between T2-weighted and CPK were assessed using Pearson's correlation analysis or the nonparametric Spearman rank-order correlation. Statistical signi cance was accepted at P <0.05. Statistical analyses performed using GraphPad Prism (version 8.0.2, GraphPad Software, Inc.)

Results
The model successfully induced ACS on rat's hindlimb compartments by using the fastened zip-tie.
technique. 31 P-MRS detected signi cant changes in muscles at all the time points ( g.5). Energy metabolism, signal intensities, and pH at rest: the metabolite concentration such as PCr, Pi, and ATP were observed as a baseline value. Typical T2-weighted fast spin-echo images were shown in g.8.
Energy metabolism, signal intensities, and pH during muscle compression and reperfusion: After muscle compression, PCr/(Pi+PCr) ratio immediately decreased compare to pre-ischemic level, which was 0.858 ± 0.100 (mean ± SD, n = 6) ( g.6, tab.1). After 60 and 120 minutes of ischemic state, PCr/(Pi+PCr) ratio was 0.182 ± 0.062 and 0.036 ± 0.012, respectively. Friedman test revealed a signi cant difference in PCr/(Pi+PCr) ratio among pre-ischemic, 60 minutes, and 120 minutes of ischemic state (P <0.05, g.6). Intracellular pH signi cantly changed due to time (P <0.05, g.7, tab.2). It decreased gradually from 7.092 ± 0.013 to 6.430 ± 0.121 after 120 minutes of ischemic. At thirty minutes of ischemic state, PME (phosphomonoester) peak was detected, which was not seen in the pre-ischemic phase, in the region of about from 4.7 ppm to 5.0 ppm to the left of the Pi peak ( g.5). SI of TA and GA muscles gradually increased in the T2-weighted images.
Once the zip-ties were released, 31 P-MRS spectra instantly started to recover to the pre-ischemic level. Nevertheless, PCr/(Pi+PCr) ratio did not ultimately return to normal values. At 90 minutes of reperfusion state, the corrected ratios of PCr/(Pi+PCr) was 0.676 ± 0.199, of those in the pre-ischemic state (tab.1). The intracellular pH started to re-increase during the reperfusion time (tab.2). It did not fully recover to the baseline level at 90 minutes of reperfusion state (6.533 ± 0.492). PME peak gradually decreased and disappeared during reperfusion state ( g.5) After release, muscles' SI became inhomogeneous, especially the SI of TA and GA muscles wherein they are remarkably higher than those of other areas ( g.8). At 90 minutes of reperfusion state, the high SI of TA and GA muscles were found to be preserved.

Discussion
In this study, the goal was accepted as 31 P-MRS provided a reliable, sensitive measure of muscle metabolites changes in this fastened zip-tie rat model of ACS. This study is the rst nding in the world expected to lead to limb salvage with ACS patients. These data are consistent with previous data from the arterial occlusion model [25]. The 31 P-MRS spectra showed a predictable change to a constant level when plastic zip ties were fastened, suggesting a loss of perfusion to the tissues. 31 P-MRS spectra data con rms that the fastened zip-tie rat model of ACS signi cantly reduced the blood ow into the compartment's tissues. It is signi cant to note that 31 P-MRS spectra changes occur immediately after decreased perfusion demonstrating real-time responsiveness. These data showed that from 60 to 120 minutes of ischemic state, the ratio of PCr/(Pi + PCr) was rmly lower than the baseline value. In contrast to this, the CPK concentration slightly changed compared to the ischemic state concentration [7]. It could be suggested that 31 P-MRS values detect more rapidly and sensitively bio-energetics' change status in muscles than CPK. 31 P-MRS is able to detect early muscle damage signs of ACS non-invasively. This observation is essential to differentiate between invasive measures such as intra-compartmental pressures and non-invasive measures such as 31 P-MRS. Additionally, during the ischemic phase and recovery phase, we also found the PME peak was considered to be delivered mainly from some kinds of phospholipids. It is an additional sign that helps in detecting early muscle damage of ACS. [26] The edema of hindlimb's compartment depending on the location in the muscles were determined by 1 H-MR images. The SI gradually increased during the ischemic phase and overshoots immediately after release the muscle compression as shown in T2-weighted images (Fig. 7). During the recovery process, the elevated SI of T2-weighted images showed that the relationship with the water content increased in the extracellular space phenomenon [27]. After the release of muscle compression, it is understood that which is thought to be caused by increased vascular permeability in skeletal muscle induced by oxygenderived free radicals, apparent edema around the skeletal muscle occurs. These ndings suggest that alterations in skeletal muscle high-energy phosphate metabolism occur early in the pathophysiology of ACS.
In most cases of the reperfusion state, the CPK concentration showed over 2000 U/L. The blood analyzer equipment could not display the speci c value of CPK concentration, which ere exceeded the upper limit of measurement (2000 U/L). However, it was considered that detailed numerical measurement was unnecessary because it was su ciently high compared to the pre-ischemic level. The CPK concentrations signi cantly increased over pre-ischemic at 90 minutes of reperfusion state. However, it was not signi cantly increased and unstable over the pre-ischemic level at 60 minutes after zip-ties fastened. The current data revealed the difference in the timing of CPK and 31 P-MRS metabolites concentration during the ischemic state. It proved that the increase in CPK concentration observed was not immediately, and so, CPK is not suitable to detect the early ischemic muscular damage in ACS. However, CPK can be used as a useful biomarker to indicate the extent of muscle damage in the sub-acute phase. [28] Moreover, CPK has been used as a primary biomarker in recognizing trauma for patients with ACS. [16] In addition, a correlation between T2-weighted images signal intensity (Tibialis anterior & Gastrocnemius muscles regions) and CPK has been found. Once the zip-ties were released, the phenomenon of blood reperfusion injury is more natural to occur with the tissue swelling, uid in ow, and compartment compression. This result found demonstrated the muscle damage was consistent for edema symptoms.
The fastened zip-tie rat model is a simple model to research crush injuries as encountered seen in a time of accidents or disasters. It can be useful to evaluate the systematic effects of releasing pressure from trapped or crush extremities. This model showed an advanced method of pressure creation compared to the previous study. [28][29][30] In our study, there was a difference between the control pH of 31 P-MRS and the blood test. Especially at the ischemic state, ischemic tissue pH (6.5) from 31 P-MRS spectra was lower than arterial pH (7.4) from blood tests. The value of pH decreased more extensively in ischemic tissue than in circulation blood. Anaerobic glycolysis lowered the pH in the ischemic area. However, homeostasis controlled systematic pH value tightly, and the value was almost unchanged (Normal mammals blood pH = 7.40 ± 0.02). According to the previous study, there was a difference between the arterial blood pH and the super cial pH value. [31] The present study has several limitations. Firstly, we were only able to use the surface coil to examine the rats' legs. The surface coil sensitivity area was limited; therefore, the SNR was not as good as we expected, even though it provided precise 31 P-MRS results. As a result, we only measured the signal intensity of the two types of muscles. However, it is enough to prove the evidence of edema appeared in the ACS model. [25] Secondly, the requirement for several groups with multiple animals was outside the experimental range of this study. Certainly, comparing 31 P-MRS and blood samples in ACS from multiple groups with longer ischemic time and treating the injury hindlimb with medicine is the next step for this line of research.
There is also tremendous research in the eld of regenerative medicine, with possible application in emergency medicine. It has been reported that the clinical e cacy of mesenchymal stem cell therapy for ischemic diseases [32]. Therefore, mesenchymal stem cell therapy has the potential for ACS. By advancing this research, we hope that the survival of patients rescued from disasters will be improved and that the deterioration of quality of life due to unnecessary amputation of the lower limbs will be prevented.

Conclusion
Our study showed the possibility of a non-invasive, reliable, and sensitive assessment of muscle   The experimental protocol. White arrows indicate time of collecting data for 31P-MRS and T2-weighted images. Red arrows designate time of arterial blood sampling for creatine phosphokinase and pH concentration.31P-MRS spectra were acquired at rest, during muscle was compressed (for 120 minutes) and reperfusion (for 90 minutes). During the MR procedure, the concentration of pH and CPK were analyzed by blood samples obtained from the tail artery.  Especially, at ischemic state, phosphate monoester (PME) peak appeared at 30 minutes and gradually increased until 120 minutes of ischemic state. At the beginning of reperfusion state, PME peak started to decrease and completely disappeared at 185 minutes of reperfusion state.  Comparison of quantitative intracellular pH and arterial pH among pre, ischemic and reperfusion. ns (not signi cant); * (0.01< P <0.05), ** (0.001< P <0.005) and *** (P <0.001) based on Friedman test. Figure 8