Metabolic Changes During and After Global Prolonged Reversible Cerebral Ischemia in Pigs. Is it Possible to Change Brain Metabolism to Maintain the Brain Alive?

Anile. Metabolic Changes During and After Global Prolonged Reversible Cerebral Ischemia in Pigs. Is it Possible to Change Brain Metabolism to Maintain the Brain Alive?. Abstract Background: Cerebral ischemic disorders can be divided into global and focal. In condition of simultaneous and similar drop of cerebral blood flow global ischemia occurs while whenever there is a partial brain involvement it is defined as focal ischemia. Lots of paper deal with focal ischemia and the relative changes in metabolism or blood supply while few reports deal with global ischemia secondary to the difficulties in creating a representative model. Objective: This paper aim to analyze changes in brain metabolism during ischemia in order to achieve a better cognition of the time limit for reversible global ischemia. Methods: Sixteen guinea pigs underwent global reversible cerebral ischemia. Four of them for 30 minutes and other four for one hour. During the entire procedure pH, pO2 and pCO2 values were recorded while brain metabolites were collected through microdialysis. Secondarily, other 8 pigs were pretreated with L-NAME and monitored as the previous eight to analyze changes in metabolism. Results: In animals not pretreated with L-NAME anaerobic glycolysis takes place with glucose consumption, increase in pyruvate and lactate levels and pH decrease. L-NAME pretreatment modify those data transforming the anaerobic glycolysis in a partial aerobic one in which glucose decrease is followed by pyruvate increase with stability of lactate and pH levels. Conclusion: Brain metabolism seems to remain active even after cortical activity stops. Such a finding becomes crucial in defining scientific criteria for brain death and in the research for neuroprotective measures against brain ischemia. Brain microdialysis represents a very accurate monitoring system for neurological Intensive Care Unit (ICU) patients. Pharmacological pretreatment can modify brain metabolism during tamponade demonstrating the persistence of cellular activity even in conditions of supposedly brain death like 30 and 60 minutes of cerebral ischemia.

a regulator of the mechanisms itself. Hypoxia/ischemia interferes with mitochondrial oxidation of pyruvate leading to cytosolic lactate formation. Such a shift leads to ATP generation only through glycolysis causing a progressive pH decrease. Acidosis development results then in cellular ion homeostasis progressive breakdown.
This cascade is on the other hand reversible as in cases of short time ischemia ion gradients gradually normalize with a recovery in metabolism after about 15 minutes. However, due to the inhibition of Pyruvate Dehydrogenate (PDH) complex by free radicals, lactate/pyruvate ratio is still significantly higher than normal.
This finding is related to the lack of pyruvate metabolism in the mitochondria. Then, during reperfusion, cell death results from a progressive inactivation and structural damage [1]. The loss in ATP modifies intracellular homeostasis blocking ATP-dependent pumps in favor of calcium dependent ones with an increase in calcium levels that activates endothelial Nitric Oxide Synthase (eNOS) and neuronal Nitric Oxide Synthase (nNOS). Nitric Oxide (NO) levels then rise. competing with O 2 during oxidative phosphorylation and subsequently decreasing even more cellular ATP formation [2].

Nonetheless, the timing of the aforementioned metabolic
cascades is yet to be clearly defined but, as shown in animal models, reversible global ischemia can lead either to deterioration of the neurological status either to recovery [3,4]. Furthermore, there are no clear evidence throughout the literature on how long global cerebral ischemia has to last to produce irreversible neuronal damages in humans. Many reports instead suggest how specific neuronal populations vary substantially in ischemic tolerance [5,6] and how this tolerance can be modified by chemical and physical agents [3,7]. The still open questions mentioned before represent one of the major issues in clinical practice. In fact, guessing the severity of brain damage and the correct timing for treatment is vital in the choice whether to treat or not and in the choice for the proper neuroprotective measures. In order to clarify those questions in our paper an evaluation of brain metabolism during tamponade was carried on in basal conditions and after Nitro-Larginine methyl ester (L-NAME) administration. L-NAME was chosen for its interference with NO synthesis being such a molecule strongly involved in brain metabolism during ischemia.

Material and Methods
The experimental protocol followed the Guidelines for the Care and Use of Laboratory Animals of the Ethical Committee of our Institution or the Helsinki Declaration of 1975 (and as revised in 1983) and the Gunma University Graduate School of Medicine experimental protocol [8]. Eight adult guinea pigs weighting 25-30 g each underwent global reversible ischemia that was induced using a Fogarty balloon inflated intracranially. In four cases brain tamponade lasted for 30 minutes while in other four it lasted 60 minutes. Blood specimen were collected from the Superior Sagittal Sinus (SSS), the right femoral artery and the right femoral vein in order to measure pH, pO 2 and pCO 2 . Cerebral microdialysis was performed to analyze brain metabolites (glucose, pyruvate, lactate, glycerol). A second experiment was performed on other eight animals that were submitted to brain tamponade similarly to the previous ones adding L-NAME pretreatment.

Surgical Preparation
Specimens were placed in sphinx position on the operating fixed flow rate of 2 mcg/min [9]).
In four animals (A30, B30, C30, D30) brain tamponade lasted 30 minutes obtaining samples at the following time intervals: T0=baseline; T1=start brain tamponade; T2=start brain reperfusion after 30 minutes of brain tamponade; T3= first 30 min of brain reperfusion; T4=60 min of brain reperfusion; T5=90 min of brain reperfusion; T6=120 min of brain reperfusion and T7=150 min of brain reperfusion. Repeated measurements were taken to have a proper evaluation of the metabolic changes alongside the whole experiment in order not to miss unexpected values modifications.
L-NAME is an antagonist of the eNOS and nNOS [10] which reduce NO levels promoting a switch in energetic metabolism to creatinine-phosphate/creatinine system (CrP/Cr) [11,12]. L-NAME was administered intravenously at the dosage of 1 mg/Kg [13,14].
No measurement of L-NAME metabolites was undertaken as this was not the scope of the paper. at the end of the 30 minutes (PaCO 2 =82,9 mmHg and PaO2=44,5 mmHg) tending to normalize at reperfusion (Table 1).  These values showed the same trend during tamponade while differing from the one measured without L-NAME pretreatment in their trend (PaCO 2 =30,6 and PaO 2 =63,9 mmHg) ( Table 3)

Statistical Analysis
Secondary to the difficulties in creating an animal model for prolonged global cerebral ischemia only 16 guinea pigs were collected divided, as mentioned before in 4 groups of 4 samples.
For this reason, having a statistically significant analysis was not possible being de facto a limit of our work that had to remain observational and inferring on the results. We are well aware of such a limit thus suggesting further studies to invigorate our results. Nonetheless, mean and standard deviation for samples collected from the SSS were calculated and reported graphically in Figure 1 and Tables 5 & 6. Those were the chosen value studied being the most representative for post tamponade metabolism.

Discussion
Brain metabolism is a complex mechanism in which glycolysis and oxidative phosphorylation are the main protagonist in ATP generation and thus energetic supply. During prolonged ischemia such mechanism are impaired and there is a shift towards different anaerobic pathways that only partially can maintain ATP formation.
In lasting ischemia, a progressive cellular breakdown, due to ATP dependent pump blockage, lead to brain death. Most of the paper are centered on focal ischemia secondary to the simplicity to evaluate the condition as well as to the frequency of the condition.
On the other hand, prolonged global ischemia is hard to evaluate, and it is hard to find clinical conditions in which proper evaluation are feasible. For this very reason, in this paper we have used animal specimens in which brain tamponade was achieved through ICP increase. Such a choice was made not hinder our data with changes in metabolism secondary to the tamponade etiology. Furthermore, as our work intend to evaluate metabolite no clinical or neurological evaluation has been carried on. Our experiment analyzed different metabolic processes occurring during different length of brain tamponade, namely 30 and 60 minutes of null CBF.
Our results reflect the shift from aerobic metabolism to anaerobic one with consequent production of acidic metabolite.
Such a modification is visible from the progressive increase of the PaCO 2 SSS in contraposition with the stability of the arterial PaCO 2 Art that tend to remain stable. It is the manifestation of the progressive glycolytic metabolism and consumption of the stored glucose. In the same way, PaO 2 SSS decrease while PaO 2 Art tend to be 15% higher than baseline in the attempt to correct the lack of brain oxygen. Sign of the severity of the condition is the absence of electrical EEG recovery at the end of brain tamponade. This data seems to suggest that a longer global brain ischemia has an important effect on how cerebral cells try to preserve function during reperfusion.
Nonetheless, even though brain cells cease their electrical activity they seem to be still metabolically "alive". These data are of the utmost importance in defining irreversible cerebral damage and more in general brain death. Nowadays, the internationally accepted medical criteria for detecting brain death, both for the American "whole brain" [15] and the British "brain stem death" doctrine [16], are based on the finding of two isoelectric lines in two EEGs at least 24 hours apart. As previously suggested the electrical measurement only address part of the problem. Even though a metabolically active brain is not the same as a functioning brain we suggest that adding metabolite analysis to our brain monitoring can be useful in discriminating the window between brain ischemia and the start of electrical inactivity.
In our opinion, the use of microdialysis [17,18] can provide a more accurate evaluation during monitoring for intensive care unit patients in comparison with other techniques (ICP monitoring [19], brain oxygenation [17,18,20], transcranial Doppler ultrasound [17,18] or scalp EEG [21]). In fact, despite its minimally invasiveness, it gives an "instantaneous picture" of brain metabolism. An even more interesting data supporting the persistence of active brain metabolism during brain tamponade is given from the results of the L-NAME pretreatment experiment. NO and its metabolites are involved in several detrimental pathways, such as apoptotic signals, inflammatory mediators and activation of proteinase enzymes. Also, combination of NO and superoxide anion leads to the formation of a very toxic compound of piroxantrone, which yields to the protein nitro tyrosination and cell death. Finally, proximitized can distribute its poisonous effects into neighboring cells and tissues [14]. In addiction NO increase seems to disrupt the blood brain barrier creating even more damage. For this very reason, L-NAME was chosen to better understand if in ischemic brain NO reduction can affect brain metabolism during tamponade.
It seems that after L-NAME pretreatment there is a switch to a "partial aerobic glycolysis" as previously reported by Regli [2,10].
This process seems testified by the increase in pyruvate that is used in the mitochondria to ensure energetic metabolism.
Furthermore, the aforementioned interference with NO and ONOO-radical formation lead to the maintenance of creatinine phosphatase activity adding, as a support in cellular metabolism, the effect of the creatinine-phosphate/creatinine system (CrP/Cr).
This system has the double role of ATP producer and mitochondrial support in dealing with ischemic damage, thus preserving oxidative phosphorylation [11,12]. Such findings seem to suggest that an organ capable of changing its metabolism can't be "dead" as previously thought. In addition, EEG evaluation in L-NAME pretreated animals follows what seen in metabolism. In fact, while in basal condition after 60 minutes of brain tamponed there is no recover in electrical activity, in treated one's partial recover can be witnessed. Though interesting such data are of difficult interpretation being taken from animal samples but remain an intriguing one. Such findings are obviously far from clinically useful but may open a discussion on whenever considering the brain dead. Furthermore, a suggestion in implementing ICU monitoring with microdialysis is warranted as a way to better monitor patients as well as a way to collect more human data to achieve a better comprehension of this complex problematic.

Conclusion
Metabolic changes during prolonged global brain ischemia are yet to be clearly defined. Despite cortical activity disappearance, brain metabolism seems to remain active especially in the first 30 minutes of ischemia. Such finding might change the way we define brain death and, in the future, might increase the possibility for neuroprotection. In this contest, brain microdialysis may represent a much more accurate monitoring system for neurological intensive care unit patients than the one of common use. Furthermore, pharmacological pretreatment might, with further studies, modify brain metabolism during tamponade increasing the gap between brain lesions and brain death. Obviously, there is still lot to investigate and many aspects of brain metabolism are still prone to be addressed by further studies, but our data might represent a starting point in a better definition of brain death and irreversible brain damage.

Declarations of Interest
The authors declare that they have no conflict of interest.