Skip to main content
NCBI home page
Science Advances logoLink to Science Advances
. 2023 Dec 13;9(50):eadj2417. doi: 10.1126/sciadv.adj2417

Memantine inhibits cortical spreading depolarization and improves neurovascular function following repetitive traumatic brain injury

Mark A MacLean 1, Jamil H Muradov 2, Ryan Greene 2, Gerben Van Hameren 2, David B Clarke 1, Jens P Dreier 3, David O Okonkwo 4, Alon Friedman 1,2,5,*
PMCID: PMC10848720  PMID: 38091390

Abstract

Cortical spreading depolarization (CSD) is a promising target for neuroprotective therapy in traumatic brain injury (TBI). We explored the effect of NMDA receptor antagonism on electrically triggered CSDs in healthy and brain-injured animals. Rats received either one moderate or four daily repetitive mild closed head impacts (rmTBI). Ninety-three animals underwent craniectomy with electrocorticographic (ECoG) and local blood flow monitoring. In brain-injured animals, ketamine or memantine inhibited CSDs in 44 to 88% and 50 to 67% of cases, respectively. Near-DC/AC-ECoG amplitude was reduced by 44 to 75% and 52 to 67%, and duration by 39 to 87% and 61 to 78%, respectively. Daily memantine significantly reduced spreading depression and oligemia following CSD. Animals (N = 31) were randomized to either memantine (10 mg/kg) or saline with daily neurobehavioral testing. Memantine-treated animals had higher neurological scores. We demonstrate that memantine improved neurovascular function following CSD in sham and brain-injured animals. Memantine also prevented neurological decline in a blinded, preclinical randomized rmTBI trial.

Memantine inhibited cortical spreading depolarization and prevented neurobehavioral decline following traumatic brain injury.

INTRODUCTION

Decades of research have been directed toward discovery of interventions that attenuate the effect of trauma on the nervous system. Despite this, translation of laboratory-based, mechanism-directed traumatic brain injury (TBI) research to the clinical setting remains a challenge (1). To date, no pharmacological agent has received approval for targeting the pathophysiology underlying TBI-related acute and long-term sequelae. Expert working groups have proposed strategies to facilitate the translation of new therapies (1, 2). Select examples include (i) identifying predictive and pharmacodynamic biomarkers of therapeutic response, (ii) confirming treatment engagement of pathomechanistic targets, and (iii) rigorously conducting randomized, mechanism-directed preclinical trials (RCTs) with blinded treatment examiners and predetermined outcomes measures.

Cortical spreading depolarization (CSD) represents a promising mechanistic biomarker for monitoring individual patients and targeted neuroprotective therapy. CSD has a pathological role in subarachnoid hemorrhage, ischemic stroke, and TBI (3–10). Animal and human studies indicate that CSDs occur in 50 to 90% of severe cases following cerebral injuries and are associated with lesion development and poor outcomes (6, 9, 11–13). Animal studies suggest that CSDs also occur following mild TBI (5, 14, 15), although corresponding human data are lacking. Given that the estimated incidence of mild TBI exceeds 300,000 cases per year among American high school football athletes (16), and 17% among military veterans (17), this area of research should be a focus of future translational efforts.

CSD involves an abrupt, sustained depolarization of neurons and glia cells within the cerebral cortex, which precludes firing of action potentials and synaptic transmission (18). The near-complete breakdown of transmembrane ion gradients during CSD is associated with neuronal swelling. In electrically active tissue, CSD is observed as a negative direct current (DC) shift associated with a wave of electrocorticographic (ECoG) silence, termed spreading depression of high-frequency brain activity. In rats, swine, cats, and humans, CSD induces a microvascular response with predominant hyperperfusion in naïve tissue, which is variably followed by mild oligemia (normal neurovascular response) (3, 19–21). In contrast, the inverse hemodynamic response to CSD in rat and human cortex with disturbed neurovascular coupling is characterized by severe initial vasoconstriction and ischemia (3, 22); this perfusion deficit spreads together with the depolarization wave (i.e., spreading ischemia) and may be followed by hyperemia instead of oligemia. The full continuum from normal hemodynamic response to increasingly inverse hemodynamic response has been observed in patients with cerebral injuries, including TBI (13, 23–25).

The excitatory neurotransmitter glutamate is excessively released following TBI and CSDs (26). The N-methyl-d-aspartate glutamate receptor (NMDAR) antagonist ketamine has been shown to suppress CSDs in animals (7, 27–36) and humans (9, 37–41). Few studies have examined CSDs and ketamine in the context of mild TBI (5, 14, 15). This likely owes to the limited clinical applicability of ketamine in sporting athletes, given the known dissociative effects and interaction with numerous receptors aside from NMDAR (42, 43).

To date, there remains a limited understanding of mechanisms underlying development of brain complications following mild and moderate TBI (modTBI) and a lack of interventions to prevent these complications. As such, here, we sought to systematically explore the role of CSDs using established rodent models of TBI. We also sought to identify a clinically applicable CSD-directed therapy that may be used across the spectrum of TBI severity.

RESULTS

Effect of NMDAR antagonism on CSD following repeated cortical stimulation in sham animals

We first explored the effect of NMDAR antagonism on CSD parameters following electrical cortical stimulation in vivo (Fig. 1). We investigated topical and intraperitoneal NMDAR drug delivery routes. Figure 2A highlights characteristic changes in measured neurovascular parameters in response to stimulation in naïve tissue including CSD occurrence (i.e., percentage of electrical stimuli resulting in CSD), apparent duration (“duration”), apparent amplitude (“amplitude”), and CSD-associated change in cerebral blood flow (CBF) measured via laser Doppler flowmetry.

Fig. 1. Experimental study design.

Fig. 1.

Open in a new tab

(A) Closed head injury models of TBI. Nine-week-old, male Sprague-Dawley rats received either mild (500 g mass, 85 cm height) or moderate (450 g mass, 1.02 m height) head impacts via a modified Marmarou weight drop. rmTBI involved one mild TBI per day for up to four consecutive days. A subgroup of rmTBI animals received memantine (10 mg/kg, intraperitoneally) after the first and before the second, third, and fourth impacts (rmTBI-DM). Sham animals received brief isoflurane anesthetic without TBI. (B) Microsurgical model to investigate effects of NMDAR antagonism on in vivo ECoG and vascular parameters following CSDs. Sham and brain-injured animals underwent cranial window surgery 24 hours following final impact or brief isoflurane anesthetic (sham). Experiments were stratified by impact severity, drug, and route of delivery. Vehicle (2.5 cm3/kg normal saline, intraperitoneally), ketamine (100 μM, topically applied to the exposed cortex), ketamine (25 mg/kg, intraperitoneally), and memantine (10 mg/kg, intraperitoneally) were tested in vivo. Treatment group and sample sizes are described in Results and table S1. (C) Blinded, preclinical randomized trial of NMDAR antagonism for the treatment of rmTBI. Following a single-arm pilot trial of rmTBI, in which animals (n = 13) received memantine (10 mg/kg, intraperitoneally), 31 animals were blindly randomized to memantine (10 mg/kg, intraperitoneally) or saline (2.5 cm3/kg, intraperitoneally). The primary outcome was the proportion of animals that scored ≥6 on the NSS composite (of 12 points).

Fig. 2. NMDAR antagonism inhibits CSD and reduces CSD amplitude and duration following repeated electrical stimulation of the cortex in sham animals.

Fig. 2.

Open in a new tab

(A) Exemplary vascular and ECoG changes during CSD triggered via electrical stimulation of the cortex. Laser Doppler flowmetry was used to record changes in CBF following stimulation. A characteristic negative shift in near-DC/AC-ECoG was observed in cases where CSD occurrence was not inhibited. CSD-induced depression of brain activity is demonstrated via AC-ECoG and integral of AC-ECoG power tracings. (B to D) In each experiment, an initial cortical stimulus was delivered before study drug delivery, serving as an internal control, and demonstrating the ability to elicit a CSD. Heat maps demonstrate the effect of intraperitoneal (IP) delivery of vehicle saline (Veh; 2.5 mg/kg; n = 7), ketamine (Ket; 25 mg/kg; n = 6), or memantine (Mem; 10 mg/kg; n = 9) on CSD-related parameters including occurrence (i.e., percentage of stimuli resulting in CSD), duration, and amplitude (both in relative units) using near-DC/AC ECoG recordings. Repeated cortical stimuli were delivered 45, 60, 75, and 90 min following drug delivery. Topical (top) ketamine (n = 5) was applied to the cortex as 100 μM, with aCSF solvent dilutant. Detailed data pertaining and statistical significance values may be found in table S1.

In vehicle-treated sham controls (n = 7), repeated cortical stimulation resulted in CSDs in 92% of cases, with no change in the duration or amplitude between subsequent stimuli in the same animal (P = 0.60; table S1 and Fig. 2, B to D). Ketamine (intraperitoneal) (n = 6) and memantine (intraperitoneal) (n = 9) significantly reduced the occurrence of CSDs (21%, P = 0.0012 and 27%, P < 0.001, respectively) (Fig. 2B), as well as the mean relative change in CSD duration (14% and 29%, both P < 0.001) (Fig. 2C) and amplitude (30% and 40%; both P < 0.001) (Fig. 2D) following repeated cortical stimuli. Ketamine (topical) (n = 5) delivery resulted in a trend toward reduced occurrence of CSDs (55%; P = 0.08), whereas significantly reduced mean relative changes in CSD amplitude (54%, P < 0.001) and duration (58%, P = 0.01) (table S1) were observed. In cases that NMDAR antagonism did not block CSD initiation entirely, a normalizing effect (i.e., reduced depression) on alternating current (AC) brain activity (AC-ECoG power integral) was observed.

In sham controls, there was no difference in CSD occurrence between ketamine (intraperitoneal) and memantine (intraperitoneal) (P > 0.5) or between memantine (intraperitoneal) and ketamine (topical) (P > 0.05). Furthermore, there was no difference in CSD amplitude or duration comparing ketamine (intraperitoneal) and memantine (intraperitoneal) (duration, P > 0.05; amplitude, P > 0.5) or comparing ketamine (topical) and memantine (intraperitoneal) (amplitude and duration, both P > 0.5), respectively.

Effect of NMDAR antagonism on CSD following repeated cortical stimulation in TBI animals

The inhibitory effect of NMDAR antagonism on CSD parameters was explored in animals subjected to modTBI and repetitive mild TBI (rmTBI) (5). Using a model of rmTBI (5), we also tested the effect of daily memantine delivery (rmTBI-DM).

In TBI animals receiving vehicle treatment intra-operatively, CSDs were elicited following most of the electrical cortical stimuli (modTBI: 88%; rmTBI: 71%; rmTBI-DM: 90%; all P > 0.05; six animals per group; table S2 and Fig. 3A). In vehicle-treated, brain-injured animals, mean relative CSD amplitude (modTBI: 68%, n = 5, P = 0.004; rmTBI: 59%, n = 6, P < 0.001; rmTBI-DM: 73%, n = 4, P = 0.007) and duration (59%, n = 6, rmTBI; P < 0.001) were reduced with repeated stimuli (table S2 and Fig. 3, B and C).

Fig. 3. NMDAR antagonism inhibits CSD occurrence and reduces CSD amplitude and duration following repeated electrical stimulation of the cortex in animals following modTBI or rmTBI.

Fig. 3.

Open in a new tab

A subset of animals received memantine once per day for four consecutive days while receiving rmTBI (i.e., rmTBI-DM). (A to C) Heat maps demonstrate the effect of intraperitoneal (IP) delivery of vehicle saline (Veh; 2.5 mg/kg; modTBI, n = 5 to 6; rmTBI, n = 6; rmTBI-DM, n = 4 to 6), ketamine (Ket; 25 mg/kg; modTBI, n = 5 to 6), or memantine (Mem; 10 mg/kg; modTBI, n = 5 to 6; rmTBI, n = 5; rmTBI-DM, n = 11) on CSD-related parameters, including occurrence (i.e., percentage of stimuli resulting in CSD), duration, and amplitude (both in relative units) as measured using near-DC/AC-ECoG recordings. The topical (top) (ketamine; 100 μM; modTBI; n = 4) drug delivery route was also tested. (D) Memantine (10 mg/kg, intraperitoneally) reduced the rate of occurrence of CSDs before and after intraoperative drug delivery in each respective TBI model. P > 0.05; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Detailed data pertaining and statistical significance values may be found in table S2.

The occurrence of CSD following delivery of ketamine was 56% and 12%, when delivered by topical (n = 4) and intraperitoneal (n = 5) routes, respectively (Fig. 3A). The occurrence of CSD following delivery of intraperitoneal memantine was also reduced (modTBI: 50%, n = 6, P = 0.04; rmTBI: 40%, n = 5, P = 0.024; rmTBI-DM: 33%, n = 11, P < 0.001; table S2 and Fig. 3, A and D). Both ketamine and memantine reduced the CSD amplitude and duration in TBI animals. Following ketamine delivered by topical and intraperitoneal, the CSD duration was 61% and 13% compared to baseline, respectively (n = 4 and n = 6, modTBI; P = 0.016 and P < 0.001, respectively; table S2 and Fig. 3B). Following memantine, the CSD duration was 22% (n = 5, modTBI), 39% (n = 5, rmTBI), and 31% (n = 11, rmTBI-DM) compared to baseline (all P < 0.001; Fig. 3B). Following ketamine (topical or intraperitoneal) and memantine (intraperitoneal), the CSD amplitude was 56% (modTBI: n = 4, P = 0.004), 25% (modTBI, n = 5, P < 0.001), and 33 to 48% (modTBI: n = 6; rmTBI: n = 5; rmTBI-DM: n = 11; all P < 0.001) (table S2 and Fig. 3C).

Following modTBI, there was no difference in CSD occurrence between ketamine (intraperitoneal) and memantine (intraperitoneal) (P > 0.05) or between memantine (intraperitoneal) and ketamine (topical) (both P > 0.50). Furthermore, there was no difference in CSD amplitude or duration comparing ketamine (intraperitoneal) and memantine (intraperitoneal) (duration, P > 0.5; amplitude, P > 0.05). However, memantine (intraperitoneal) reduced the duration and amplitude significantly compared to ketamine (topical) (both P = 0.008).

We found that vehicle-treated brain-injured animals had significantly reduced CSD duration (P = 0.007) and amplitude (P = 0.01) compared to sham animals (tables S3 and S4). Following NMDAR antagonist treatment, these differences were no longer observed (all P > 0.5).

A single intra-operative dose of memantine (intraperitoneal) affected all three parameters associated with CSD-induced spreading depression of high-frequency (integral of AC-ECoG power) brain activity [area under the curve (AUC); the nadir of AC-ECoG power and rate of decline of brain activity after stimulation] following rmTBI (Fig. 4, A to D). The power at the nadir was increased following memantine delivery (pre-memantine, stimulation 1; n = 16; 0.22 ± 0.21 versus post-memantine, stimuli 2 to 5; n = 28; 0.51 ± 0.28; P = 0.001). The mean AUC was also increased, indicating normalization of brain activity following memantine administration (pre-memantine, stimulation 1; n = 12; 179.33 ± 61.61 versus post-memantine, stimulations 2 to 5; n = 28; 293.87 ± 112.96; P = 0.002). A statistically significant reduction in the rate of decline in brain activity following stimulation was observed following memantine (pre-memantine, stimulation 1; n = 16; slope −0.0043 ± 0.0017 versus post-memantine, stimulation 4; n = 7; slope −0.0026 ± 0.0011; P = 0.026 and stimulation 5; n = 7; slope −0.0021 ± 0.00080; P = 0.004).

Fig. 4. Memantine mitigates brain activity depression during CSD in rmTBI-DM animals.

Fig. 4.

Open in a new tab

(A) Normalized AC-ECoG power integral tracings demonstrate the effect of intraperitoneal injection of memantine (10 mg/kg) on brain activity during the 300-s period following each respective cortical stimulation (Stim) before (i.e., stimulations 1 to 2) and after (i.e., stimulations 3 to 5) drug delivery. (B) The nadir (n = 16) was identified as the lowest value of the power integral recording during the 5-min period after stimulation. (C) The rate of change (n = 16) was calculated as the slope of power integral decay from the normalized baseline to its nadir. (D) Area under the curve (AUC) (n = 12) was measured for each stimulation between 150 and 300 s and expressed as a percentage of AUC during a 150-s period preceding the stimulation (baseline). Means and SDs are displayed on bar graphs. P > 0.05; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001.

Effect of memantine on CSD-induced spreading depression of high-frequency brain activity following TBI

We tested the effect of daily memantine treatment on electrically triggered, CSD-induced spreading depression of high-frequency (AC-ECoG integral of power) brain activity in animals subjected to rmTBI (i.e., rmTBI-DM). At the start of each experiment, before commencing intraoperative drug testing, we found no significant difference in power spectral density across animal models (n = 5 in each group: sham, modTBI, rmTBI, rmTBI-DM) (P = 0.39) or brain partial pressure of oxygen between sham (n = 7) and modTBI (n = 11) animals (P = 0.33). A single head impact was delivered per day for four consecutive days. Memantine was delivered following the first head impact and before the second, third, and fourth impacts. Craniotomy and stimulation of the frontal cortex was performed 24 hours following the last head impact, without any intra-operative drug treatment. Quantitative analysis of the AC-ECoG power integral tracings revealed that rats receiving daily memantine [n = 16; rmTBI-DM; 179.33 ± 61.61 relative units (RUs)] had reduced CSD-induced spreading depression (normalization) compared with vehicle-treated animals that received either modTBI (n = 20; 100.09 ± 56.7 RUs; P = 0.001) or rmTBI (n = 9; 106.73 ± 37.04 RUs; P = 0.032) (Fig. 5, A to D). AUC for rmTBI-DM animals was also comparable and trended toward being increased (i.e., further normalized) compared to vehicle-treated sham animals (129.24 ± 55.21 RUs; n = 32; P = 0.066).

Fig. 5. Daily memantine use has a lasting effect and reduces CSD-induced spreading depression of brain activity following rmTBI.

Fig. 5.

Open in a new tab

(A) Normalized AC-ECoG power integral tracings demonstrate the degree of brain activity depression during the 300-s period following an initial electrical cortical stimulus (stim), before drug delivery, for each respective TBI model (modTBI, rmTBI, and rmTBI-DM). (B) The nadir was identified as the lowest value of the power integral recording during a 5-min period after stimulation. (C) Rate of change represents the slope of power integral decay from the normalized baseline to the respective nadir. (D) Power integral AUCs were measured for each animal group between 150 and 300 s and are expressed as a percentage of the AUC during a 150-s period preceding the stimulation (normalized baseline). Means and SDs are displayed on bar graphs. N values are displayed on the figure. P > 0.05; *P ≤ 0.05; **P ≤ 0.01.

Local cortical blood flow and pial arterial vasoconstriction response to CSD following TBI

The effect of daily memantine treatment on the local hemodynamic response to electrically triggered CSD was measured using laser Doppler flowmetry (Fig. 6, A to C). All animals had a normal hemodynamic response to CSD, characterized by a predominant hyperemic phase followed by oligemia. There was no difference in AUC, peak maximum, or slope of the hyperemic response to CSD between sham and TBI animals. In contrast, AUC for the oligemic phase was more pronounced in animals after modTBI (105.1 ± 26.7 RUs, n = 16, P = 0.005) or rmTBI (104.7 ± 23.1 RUs, n = 10, P = 0.01) compared to vehicle-treated sham animals (137.1 ± 30.4 RUs, n = 16). Animals that received daily memantine treatment with rmTBI had a reduced oligemic phase of CSD (122.9 ± 29.0 RUs, n = 15), which was comparable to sham animals (P = 0.22). Pial arterial diameter measurements revealed that post-CSD–induced vasoconstriction (rmTBI-DM versus rmTBI and modTBI) was smaller in memantine compared to vehicle-treated animals (Fig. 6, D and F).

Fig. 6. Daily memantine use improves local cortical blood flow and reduces the degree of pial arterial vasoconstriction following CSD in rmTBI.

Fig. 6.

Open in a new tab

(A) Laser Doppler flowmetry recordings depict hyperemic and oligemic phases of the CBF response to CSD. AUC measurements for the hyperemic (B) and oligemic (C) CBF phases were measured between 30 to 120 s and 150 to 300 s after stimulation, respectively. AUCs for the hyperemic and oligemic phases are expressed as a percentage of the AUC during a 150-s period preceding the stimulation (normalized baseline). (D) Cortical pial artery diameter was measured using direct intravital microscopy. AUCs for vasodilation (E) and vasoconstriction (F) were calculated as an integral of the time-dependent changes in 30 to 120 s and 150 to 300 s after stimulation, respectively, and are represented as a percentage of the AUC during a 150-s period preceding the stimulation (normalized baseline). Means and SDs are displayed on bar graphs. N values are depicted on the figure. P > 0.05; **P ≤ 0.01; ***P ≤ 0.001.

Pilot trial of memantine for the treatment of rmTBI

We conducted a single-arm pilot trial of rmTBI, in which rodents (n = 13) received memantine (10 mg/kg, intraperitoneally). The primary outcome was the proportion of animals that scored ≥6 on the neurological severity scale (NSS) composite (see Materials and Methods for details). Memantine was delivered immediately following the first and 1 hour before the second, third, and fourth mild closed head impacts (Fig. 7A). Ninety-two percent of memantine-treated rodents scored >6 on neurobehavioral testing (5).

Fig. 7. Memantine improves neurological outcomes following rmTBI.

Fig. 7.

Open in a new tab

(A) Pilot trial: mean NSS scores are depicted for animals that received memantine (10 mg/kg, intraperitoneally ) once daily for four consecutive days with rmTBI (i.e., rmTBI-DM; n = 13). (B and C) Preclinical RCT: A blinded RCT of rmTBI-DM (n = 15) versus daily saline (2.5 mg/kg, intraperitoneally; n = 16) was conducted, and individual NSS (B) and mean aggregate scores (C) are demonstrated. (D) NSS subcomponent scores are shown for the aggregate treatment arms (rmTBI, left panel; rmTBI-DM, right panel). (E) Individual NSS values were compared for animals enrolled in the pilot study, the RCT, as well as an additional historical cohort of animals that underwent rmTBI. P > 0.05; **P ≤ 0.01; ***P ≤ 0.001. Means are presented with SEs.

A randomized, blinded, pre-clinical trial of memantine for rmTBI

We subsequently conducted a randomized, blinded, preclinical trial of daily memantine (intraperitoneal) versus vehicle (intraperitoneal) (for power calculation: see Materials and Methods). Mean neurological scores were higher in the memantine group at the second, third, fourth, and fifth assessments [final: memantine 9.27, 95% confidence interval (CI) 7.56 to 10.97 versus vehicle 5.56, 95% CI 3.93 to 7.19, P = 0.002]. Memantine prevented neurological decline following rmTBI compared to vehicle (memantine baseline versus final assessment, P = 0.14; vehicle baseline versus final assessment, P < 0.001; Fig. 7, B to D). Regarding the primary outcome, the proportion of rats that scored ≥6 on the NSS was higher in the memantine compared with the vehicle group (fifth assessment: 14 of 15, 93.3% versus 9 of 16 animals, 56.3%, P = 0.023; table S3). Additional secondary outcome data are described in the supplementary materials.

Post hoc analysis: Matched pilot, randomized preclinical trial, and historical control data

Memantine-treated (n = 28) animals from both pilot and randomized preclinical experiments were matched to a historical cohort of vehicle-treated animals (n = 36) (Fig. 7E) (5). Findings were similar to those obtained in the blinded study; mean NSS scores were higher in the memantine group at the second, third, fourth, and final assessments (final: memantine 8.79, 95% CI 7.86 to 9.71 versus vehicle 5.72, 95% CI 4.44 to 7.00, P = 0.002). Furthermore, the proportion of animals that scored ≥6 on the NSS was higher in the memantine group at the fourth and fifth assessment times (final: n = 26 of 28, 92.8% versus n = 15 of 36, 41.7%, P < 0.001; table S4).

Association between NSS scores and CSD-induced spreading depression and oligemia severity following CSD in rmTBI

A post hoc analysis was performed for a subgroup of animals that received rmTBI and daily vehicle treatment, and that subsequently underwent craniotomy for intra-operative recording of cortical activity and local hemodynamic response measurement following electrically triggered CSD. CSD-induced spreading depression and local hemodynamic response were compared between animals with NSS scores ≥6 versus <6. Notably, animals with NSS scores <6 had a significantly more profound CSD-induced reduction of cortical activity (n = 9, 181.48 ± 21.3 versus n = 6, 117.5 ± 12.5; P = 0.045) and oligemia (n = 17, 123.6 ± 27.1 versus n = 8, 100.6 ± 22.8; P = 0.045) compared to animals that scored ≥6 on NSS testing.

DISCUSSION

Overall, our findings demonstrate that the NMDAR antagonist memantine (intraperitoneal) was comparably effective to ketamine (intraperitoneal) at reducing the occurrence, duration, and amplitude of electrically induced CSD in sham and brain-injured animals. Memantine also alleviated CSD-induced spreading depression and post-CSD oligemia. Considering these findings, we conducted the first preclinical RCT of memantine using a previously validated rodent model of rmTBI (5), with predetermined outcome measures and blinded examiners (1). Regarding the primary outcome, memantine prevented neurological decline compared to vehicle treatment.

Ketamine and CSDs in animals and humans

Ketamine is a noncompetitive NMDAR antagonist known to inhibit CSD occurrence and to reduce CSD propagation velocity and negative DC shift duration (28, 34–36). We similarly found that intraperitoneal ketamine treatment inhibited CSD occurrence in vivo in sham and modTBI animals. Memantine (intraperitoneal) was comparably effective to ketamine (intraperitoneal) at reducing CSD occurrence, duration, and amplitude. In cases that ketamine did not prevent CSD induction, the amplitude and duration were reduced compared to nontreated controls. These results are in accordance with a therapeutic dose-dependent benefit observed for CSD parameters in healthy gyrencephalic swine cortex (30). Recently, exploratory trials have shown that ketamine suppresses a fraction of CSDs in humans with cerebral injuries (37, 38, 40, 41). Carlson et al. (38) recently conducted the first prospective, randomized cross-over trial of ketamine versus other sedative agents on the occurrence of CSD in patients with severe cerebral injuries and found a dose-dependent suppressive effect.

Few studies have examined modalities of ketamine delivery aside from intraperitoneal injection. We found that topical application of ketamine perfused over the cortex in healthy and TBI animals resulted in a reduction in CSD amplitude and duration. Topical drug application may have clinical potential, obviating side effects associated with systemic use, particularly in elderly patients. The presence of CSDs following evacuation of chronic subdural hematoma (cSDH) has been associated with worse clinical outcomes (44), and application of topical ketamine via irrigation fluid to the subdural hematoma cavity following evacuation may represent an area for future study.

Effect of memantine on CSD and cortical blood flow following rmTBI

Given that CSDs are common in animal models of mild TBI, there remains an outstanding need to identify an alternative clinically applicable CSD-directed therapy with a tolerable side effect profile that may be used across the spectrum of TBI severity (5). Ketamine may induce dissociative, cognitive, and autonomic alterations (42). Such effects may hinder broad clinical application, particularly for milder cases on the TBI spectrum. Given this, we did not explore ketamine delivery in the context of rmTBI, nor its effect on the oligemic phase of CSD, given the potential limited clinical applicability. Here, we chose to investigate the U.S. Food and Drug Administration–approved NMDAR antagonist memantine, given its placebo-like side-effect profile in humans (45) and similar affinity for the phencyclidine binding site of the NMDAR complex (46).

In our previously reported cohort of rmTBI animals (5), approximately 50% of rats experienced a decline in NSS scores to <6 (of 12) with consecutive impacts (i.e., bimodal distribution with a trough value of 6 of 12). Since CSD was also recorded immediately after injury in ~50% of rats (5), we sought to investigate the effect of daily memantine treatment on NSS scores with rmTBI. First, we assessed the impact of memantine injection on neurobehavioral (NSS) scores in healthy rats. The absence of deleterious effects on NSS scores observed in healthy animals is in keeping with prior literature (47–49). In our pilot trial, 92% of memantine-treated rodents scored >6 on neurobehavioral testing, compared with 50% in our historical cohort (5). In a blinded, randomized preclinical of memantine versus vehicle for rmTBI, we found that daily memantine treatment improved NSS scores following rmTBI.

Preclinical studies of memantine for the treatment of TBI and/or CSDs are limited, but our results are consistent with two prior preclinical studies that demonstrated that memantine suppresses CSDs (49, 50). Human studies of memantine for the treatment of cerebral injuries are sparse. Reinhart et al. (50) used memantine (30 mg delivered enterally) twice daily to reduce CSD duration, as well as ECoG suppression following CSD, in a single patient with cSDH. Dosing was extrapolated from a trial by Mokhtari et al. (51); however, dosing and safety studies were not conducted before that trial, despite recommendations for such (1, 2).

Here, we found that brain-injured animals had more pronounced oligemia in response to CSD compared to controls. These findings are consistent with prior animal and human studies (10), particularly in the setting of multiple CSDs (52). Prior report of low-dose ketamine infusion in healthy gyrencephalic swine did not find reduction of post-CSD oligemia (32). However, given that memantine has been reported to improve CBF in Alzheimer’s patients, we also investigated the drug effect on hemodynamic response following CSD (53).

We found that post-CSD–induced vasoconstriction was reduced in memantine compared to vehicle-treated TBI animals. These findings may contribute to the beneficial effects of memantine in our study. Mechanistically, memantine may have a direct effect on the hemodynamic response to CSD via the astrocyte-endothelial cell vasodilatory axis, in which cortical astrocytes drive changes in arterial diameter, which is at least partially dependent on endothelial NMDAR activation and nitric oxide production (54). Prior studies support a direct glutamate effect on cortical vasculature, including blood-brain barrier (BBB) dysfunction and nitric oxide production (55, 56). A recent study found that memantine facilitated BBB repair in a rat model of valproic acid–induced autism (57). An alternative hypothesis involves reducing CSD-associated neuronal depolarization and the metabolic burden of the propagation event, preserving subsequent neurovascular coupling via downstream signaling products (e.g., nitric oxide and arachidonic acid) to match vasomotor effects with demand (30). The effect of ketamine on the oligemic phase of CSD following rmTBI was not explored, given the lack of clinical applicability of ketamine among mild and moderate TBI populations. When delivered intra-operatively, ketamine blocked electrically induced CSD in most instances (88%); therefore, the oligemic phase could not be measured.

Neuroprotective role of memantine: Potential application toward CSDs and TBI

To test the applicability of memantine for TBI treatment, we treated rats with memantine and tested recovery after rmTBI. Daily memantine treatment improved NSS scores following rmTBI. A prior study reported that memantine reduced neuronal death in CA2 and CA3 of the hippocampus in a rodent model of modTBI (47) and rmTBI (58). A separate preclinical study showed that a single dose of memantine 1 hour after the final (fourth) impact attenuated acute β-amyloid precursor protein expression and accumulation of phosphorylated tau in an rmTBI model (59). One month following injury, memantine had reduced microglial activation, rescued NMDAR subunit loss, and partially restored long-term potentiation. In the clinical setting, memantine was shown to improve neurological function in patients with thromboembolic stroke (60).

Our blinded, preclinical randomized trial showed that animals receiving memantine had better neurobehavioral scores compared to those receiving normal saline treatment following rmTBI. We therefore suggest that memantine may become a candidate for future clinical trials in humans. Regarding dosing, we used 10 mg/kg, as this was the lowest dose in rats that consistently blocked CSD induction. In rats, this dose yields brain extracellular fluid and serum concentrations of approximately 0.4 to 0.5 μM (61). In human frontal cortex, the Ki (inhibition constant) value of memantine at the phencyclidine binding site of the NMDAR is 0.5 μM (62); similar serum and cerebrospinal fluid (CSF) levels are observed following daily administration of 30 mg via oral route (63). This dose is comparable to the dose used by Reinhart et al. (50). While the dose used here appears comparable to clinically relevant doses in humans for treatment of Alzheimer’s disease (64), further studies are warranted to elucidate the dose required to suppress CSD in humans. The optimal dosing frequency should also be considered, given that the half-life of memantine in rats (<4 hours) and humans (>60 hours) markedly differs (65).

Overcoming failed trials of NMDAR antagonism for TBI: Leveraging timing of drug delivery and improved CSD detection

Many human trials of NMDAR antagonism for the treatment of acquired brain injuries have failed (66) or were terminated early for drug-related morbidity and mortality (2, 67). Notably, most human therapeutic trials in TBI have involved drug delivery hours following injury, long after excitatory neurotransmitter levels have risen and pathological neuronal damage and cell death pathways have been triggered (2). An alternative explanation may relate to the lack of biomarker-driven strategies to confirm treatment engagement. For example, many prior trials did not stratify based on the presence or absence of CSDs, despite the incidence of CSDs being reportedly greater than 50% following mild and moderate TBI in animals (5). In patients with severe TBI, CSDs are measured in 60% of cases (6) and, in other entities such as aneurysmal subarachnoid hemorrhage, in up to 90% of cases (13). The incidence of CSD following mild TBI in humans is unknown, likely given the lack of noninvasive approaches for detection.

The timing of therapeutic delivery following TBI may influence clinical outcomes. Masse et al. (68) recently used a rodent model of mild TBI and found that the timing of NMDAR antagonism affected extracellular, excitatory amino acid concentrations. Only drug delivery before trauma resulted in normalized glutamate levels and righting times comparable to sham animals. Given that neurological decline occurs following multiple repetitive mild head impacts rather than a single impact (5), there may be clinical value in selectively prescribing prophylactic treatment after the diagnosis of mild TBI is established. This was the rationale behind timing of drug delivery in our preclinical RCT; drug delivery was performed after the first and before the second, third, and fourth closed head impact.

Limitations

Our study has several limitations. First, we tested a single alternative NMDAR antagonist to ketamine. This approach was chosen to allow for a detailed comparative investigation of memantine’s effect on CSD parameters to both systemic and topical ketamine in healthy and TBI animals. Second, we chose a unique timing for memantine and vehicle delivery, with pre-impact dosing following second to fourth impacts. Alternative dosing strategies, such as those previously investigated (59), may not have similar outcomes. Third, instead of true DC-AC recordings, we used near-DC-AC recordings. The latter represents a surrogate signal for true DC current and was used to systematically quantify signal characteristics across experiments. Near-DC-AC recordings allow only approximate conclusions about whether the amplitude and duration of the negative shifts of the slow potential of SDs are pharmacologically altered (69). This limitation was partially mitigated by including quantitative analyses of the integral of ECoG power (i.e., CSD-induced spreading depression), as per expert guideline recommendations (69). Finally, we acknowledge that the beneficial neurobehavioral effect observed in our trial of memantine for rmTBI could be independent of an effect on CSD. However, we have recently shown that spontaneously occurring CSD following TBI and electrically triggered CSD in sham animals are both associated with decline in neurobehavioral function (70).

Future work

The optimal agent, timing, and route of delivery of CSD-directed therapies remain unclear. There remains an outstanding need for clinically applicable CSD-directed therapies that may be used across the TBI severity spectrum. This includes therapies applicable for use in sporting athletes and military service people, given the high incidence of mild TBI among both populations (16, 71). However, whether CSDs are common in humans with mild TBI has not been demonstrated to date. Therefore, future work may focus on developing noninvasive methods and determine the occurrence of CSDs following mild-to-moderate TBI in humans, given the high incidence in animals (5). Eventually, the efficacy of memantine may be examined through clinical trials in humans, particularly emphasizing CSD as a biomarker of therapeutic response and treatment engagement.

It is possible that the neurological benefit observed with memantine in our preclinical trial relates to CSD inhibition or reduction in post-CSD oligemia (which was observed to last >24 hours following the last dose of memantine) in a subset of animals. However, this should be confirmed with direct intracranial monitoring, for example, using implanted epidural electrodes and real-time local blood flow monitoring, as has been recently demonstrated (5). Additionally, there remains a need for development of noninvasive means by which CSDs can be detected in nonoperative patients. Alternative biomarkers should be considered. For example, BBB dysfunction has been associated with CSDs (5, 72), mild TBI in rodents (5, 73) and humans (5, 74–77), and persistent neural dysfunction (78, 79). Established noninvasive dynamic contrast-enhanced magnetic resonance imaging techniques have been reported for the detection of BBB dysfunction in humans (74–76).

Here, we compared the in vivo effects of NMDAR antagonism on CSD-related electrophysiological and vascular parameters across validated preclinical models of TBI, varied by impact severity, drug, and route of drug delivery. Memantine (intraperitoneal) was as effective as ketamine (intraperitoneal) at reducing the proportion, amplitude, and duration of electrically triggered CSDs. Memantine also had a lasting protective effect against spreading depression and oligemia following CSD. In the first randomized, blinded, preclinical RCT of memantine for rmTBI, memantine prevented neurological decline compared to vehicle treatment. This work may set the stage for clinical trials of memantine in humans with acquired brain injuries.

MATERIALS AND METHODS

Study design

All procedures were approved by the Institutional Animal Care and Use Committee and were performed in accordance with Public Health Service Policy on Humane Care and Use of Laboratory Animals. Nine-week-old, adolescent wild-type male Sprague-Dawley rats (Charles River, Montreal, Quebec, Canada) were double-housed in standard cages and exposed to a reversed life cycle (12:12 hours). Animals had access to food, water, and shelter ad libitum. Experiments were conducted during the dark (active) phase of the light cycle. Experimental reporting adheres to the ARRIVE guidelines (80). Euthanization was performed if animals deteriorated during experimental testing (i.e., sustained paralysis; 15% loss of total body weight).

A modified Marmarou weight-drop model was used to induce either a single mild (500 g mass, 85 cm height; mild TBI) or moderate (450 g mass, 1.02 m height; modTBI) head impact (5, 70, 81, 82). Impacts were delivered anterior to the lambda suture line and posterior to bregma. Impacts were delivered in the midline via alignment with the animal’s ears as an anatomical reference. Following mild TBI, behavioral scores were transiently reduced 10 min following impact and return to baseline within 24 hours. Mortality was rare, and both brain magnetic resonance imaging and gross post-mortem pathological assessment revealed no intracranial bleeding or structural injury (5). Twenty minutes following modTBI, behavioral scores were reduced and followed a normal distribution (70). Forty-eight hours following modTBI, behavioral scores remained reduced, and the distribution was bimodal. Mortality was 2.5%. Pathological assessment revealed bruises on the rat’s skull and epidural hematomas, without evidence for subdural or cortical intracerebral bleeding. Subarachnoid hemorrhage was observed, most often adjacent the skull base or brainstem. rmTBI involved one mild TBI per day for up to four consecutive days. A subgroup of rmTBI animals received intraperitoneal memantine (10 mg/kg) after the first and before the second, third, and fourth impacts (rmTBI-DM). Sham animals received brief anesthetic (3 ± 1 min) without TBI; handling was otherwise comparable to animals that received TBI. Impacts were delivered as previously described (76). Pressure-sensitive films were previously used in a subset of rmTBI experiments to ensure that the delivered pressure was equivalent for each experiment (5).

We conducted a two-part study to achieve the following primary research objectives: (i) to use an established rodent microsurgical model to investigate the effects of NMDAR antagonism on in vivo ECoG and neurovascular parameters following CSDs in both sham and brain-injured animals, stratified by impact severity, drug, and route of drug delivery, and (ii) conduct a blinded, preclinical RCT of NMDAR antagonism for the treatment of rmTBI.

In vivo animal preparations (cranial microsurgery)

We explored the effect of NMDAR antagonism on CSD in animals that received either modTBI, rmTBI, rmTBI-DM, or sham impacts. Normal saline (2.5 mg/kg, intraperitoneally) was tested as a control vehicle. Twenty-four hours following final cranial impact (or brief isoflurane anesthetic in sham controls), animals underwent cranial window surgery as previously reported (5, 83). In brief, surgical procedures were performed under deep isoflurane anesthesia (3% for induction and 1.5% for maintenance; 2 liters/min O2), monitored using the toe-pinch reflex. The animal was subsequently placed in a stereotaxic cranial frame. Body temperature was monitored continuously and maintained at 37.5 ± 0.5°C via heating pad and rectal temperature probe. Heart rate and oxygen saturations were continuously monitored using an animal oximeter (ML325, ADInstruments, Colorado Springs, CO, USA) and a PowerLab data acquisition device (PL3508, ADInstruments). A parietal cranial window was exposed. The cortex was gently superfused with artificial CSF (aCSF) (5). Regional cortical blood flow was measured using laser Doppler flowmetry (Oxford Optronix, OxyFlo, 2000). Intravital cortical microscopy was performed (Axio Zoom V16, Zeiss GmbH and CMOS camera, PCO Edge 5.5 model, PCO-Tech Canada). Recording electrodes were constructed using Teflon-insulated silver wire (280 mm diameter, A-M Systems Inc.), and differential ECoG recordings were done using a multi-channel Octal Bioamplifier (ML138, ADInstruments, Sydney, Australia). Recording electrodes were placed on the cortex, 2 mm and 6 mm posterior to bregma, and 2.5 mm from midline. A ground electrode was placed in the subcutaneous tissue of the animal’s neck. Near-DC/AC-ECoG recordings were acquired (sampling rate: 0.4 kHz) and were high (0.02 Hz), low (100 Hz), and notch (60 Hz) filtered. CBF recordings were low pass filtered (0.25 Hz). In a subgroup of animals (n = 7 for sham controls and n = 11 for modTBI), Unisense OX-10 probe was inserted 1 mm into the cortex to measure brain tissue oxygen partial pressure.

CSDs were electrically triggered by cortical stimulation (20 V, 20 Hz, 2 s, and 5 to 20 ms pulse width) through stainless steel electrodes (2.5 cm length, 1.5 mm diameter) via a separate frontal burrhole in the ipsilateral hemisphere. At the start of each experiment, a single cortical stimulus was applied to ensure that CSD could be elicited before any drug delivery; the first electrical stimulus resulted in a CSD in all experiments. CSDs were recorded when stimulations resulted in an abrupt near-DC/AC ECoG negativity shift, cortical activity depression, hemodynamic response, and change in intrinsic optical signaling of the cortex. To ensure peak systemic drug levels, four additional electrical cortical stimuli were delivered every 15 min, beginning either 30 min following saline or ketamine (27, 84), and 45 min following memantine delivery (65). Topical solutions were added to aCSF applied directly to the exposed cerebral cortex. Ketamine was dosed as previously described (100 μM topical and 25 mg/kg, intraperitoneally) (27, 84). We used the lowest dose of intraperitoneal memantine (10 mg/kg) at which CSDs were consistently blocked (47, 49). Memantine and ketamine were purchased from Sigma-Aldrich Ltd. (St. Louis, USA) or CDMV (Quebec, Canada). Ketamine contained a racemic mixture of its enantiomers [dl-2-(o-chlorophenyl)-2-(methylamino) cyclohexanone hydrochloride].

Neurobehavioral assessments

Neurobehavioral assessments were scored using the NSS as previously described (5, 74). Assessments were conducted 1 hour before each impact, as well as 24 hours following the final impact. The multidimensional test battery was conducted in awake, unanesthetized animals. Assessments were performed using beam walk, inverted wire mesh, and open-field tests (5). Each subtest elicits a spectrum of neurobehavioral deficits, spanning several domains: orienting response, locomotion and exploration, gait and balance, neuromotor function, habituation, arousal, and response to environmental stimuli. Neurobehavioral performance on each subtest was graded by a trained observer, according to a five-point scale (0 to 4), based on standardized metrics capturing graded behavioral impairment per respective subtest. A composite score of 12 indicated unimpaired performance in all three subtests. Lower scores correspond to increasing post-injury neurobehavioral impairment. All neurobehavioral assessments were filmed (Canon R700). Convulsive movements (presence, duration) and latency to locomotion were recorded after impact, respectively.

Neurobehavioral assessments and offline image analysis were performed by trained observers for both pilot and randomized trials. For the preclinical trial, observers were blinded to the treatment assignment. Time to locomotion was defined as purposeful movement of all four paws. Preclinical RCT results were analyzed after all data were collected.

Before commencing our pilot trial of memantine treatment for rmTBI, we sought to assess the impact of daily memantine injection on neurobehavioral (NSS) scores. Five animals received a daily injection of memantine (10 mg/kg, intraperitoneally). Animals underwent daily NSS neurobehavioral testing for 5 days following injection. Scores remained similar to baseline at all time points (all P > 0.05).

Blinded, randomized preclinical trial of memantine versus vehicle saline for rmTBI

Thirty-one animals were blindly randomized to memantine (10 mg/kg) or saline (2.5 cm3/kg); of the 36 animals ordered, 2 animals did not survive transport and delivery, and 3 did not pass our standardized pre-experiment neurobehavioral apparatus initiation and training. Randomization was performed using GraphPad (Dotmatics, San Diego, CA) software to minimize selection bias and reduce systematic differences in animal characteristics allocated to treatment and control groups (85).

Planned post hoc neurobehavioral score analysis

Neurological scores from rodents that received rmTBI and daily memantine treatment that were included in our pilot trial and preclinical trial were matched against rodents that received rmTBI and daily vehicle saline treatment that were included in our preclinical randomized trial, as well as those from our recently published rmTBI vehicle saline–treated cohort study (5). This additional planned post hoc statistical comparison was conducted to improve the power of our analysis, as well as reduce any potential bias in animal handling and animal shipments via the addition of animals from experiments conducted by different researchers, at different times, under the same conditions.

ECoG and cortical blood flow data processing and analysis

ECoG recordings were measured and analyzed using LabChart software (version 8), as recommended and described by the Co-Operative Studies on Brain Injury Depolarizations (COSBID) guidelines (69). To account for low CSD event incidence numbers where NMDAR antagonism prevented the elicitation of CSDs, analyses were carried out by comparing the average of grouped post-drug (stimulations 2 to 5) to pre-drug (stimulation 1) CSD parameters for all in vivo experiments. Further postprocessing efforts are described in the Supplementary Materials. Using the normalized integral of ECoG power tracings, we measured the following parameters: (i) AUC, (ii) the nadir (lowest recorded value) for ECoG power, and (iii) the rate of decline (slope) in brain activity (from the normalized prestimulation baseline to the nadir following stimulation). AUC analyses were performed using time-locked ECoG recordings 50 to 350 s after electrical cortical stimulation, as no CSD was observed before 50 s after stimulation for any experiment. Integrals of ECoG power analyses are provided in an effort to ease visual assessment of changes in high-frequency brain activity; they are derived from the squared spontaneous high-frequency activity, quantifying local brain activity over time by computing time integrals over a sliding window according to a time decay function (5-s decay constant) for smoothed curves (69). A similar technique was applied to measure AUC for the hyperemic (after stimulation: 30 to 120 s) and oligemia (after stimulation: 150 to 300 s) CBF phases using time-locked laser Doppler flowmetry recordings. Time-dependent changes in vascular diameter were measured using an in-house MatLab code.

Power spectral density analysis was performed via MatLab using 120-s-long DC-ECoG data collected at the baseline from each animal and sliding sampling windows of 10 data points (25 ms). The average power spectral density was determined (n = 5 animals per group) and compared across animal models (sham, modTBI, rmTBI, and rmTBI-DM).

Power analysis and statistics

Statistical analysis was performed using SPSS version 27 (IBM Corp.). Continuous variables are described with means and SEs, while categorical variables are described as frequencies and percentages. For the blinded randomized trial of memantine versus vehicle saline, the sample size calculation was done using prior neurological scoring results from our previously published results using the same established rodent model of rmTBI (5), in conjunction with the current pilot trial. The primary outcome was the proportion of animals with neurobehavioral scores ≥6 of 12. This cutoff was selected based on results from our previously reported cohort of rmTBI animals (5), in which approximately 50% of rats experienced a decline in NSS scores to <6 (of 12) with consecutive impacts (i.e., bimodal distribution with a trough value of 6 of 12); CSD was also recorded immediately after injury in ~50% of rats (5, 86). Assuming 1:1 allocation, to determine the difference between two independent proportions, 18 animals were required per group, or 36 total, to detect significance with 80% power and α of 0.05.

A one-way analysis of variance (ANOVA) was performed for continuous variables, with Bonferroni post hoc testing as applicable. In the event continuous data were nonnormal, the Mann-Whitney U test was used. Tests of proportion were compared using the chi-square or Fisher’s exact test. Significance was obtained at P < 0.05, except in the event multiple comparisons were performed, then the Bonferroni correction was used. Figures were prepared using GraphPad Prism version 8.0 (La Jolla, CA, USA).

Acknowledgments

We thank K. Murphy and J. Kukurin for their technical expertise. We thank E. Arora, E. Ruzicka, and A. Amiree for their assistance with data processing and blind video scoring. We thank R. A. Ghazleh for methodological and technical insight regarding surgical procedures.

Funding: This work was supported by European Research Area Network Neuron Award (CIHR Award no. NDD 168164) (A.F.), Deutsche Forschungsgemeinschaft (DFG) (German Research Council) (DFG DR 323/10-2) (J.P.D.), BMBF Bundesministerium fuer Bildung und Forschung (Era-Net Neuron EBio2, with funds from BMBF 01EW2004) (J.P.D.), Neurosurgery Research Education Foundation and Academy of Neurological Surgeons Research Fellowship Grant (M.A.M. and A.F.), and Canadian Institute of Health Research Project Grant (grant number 488048) (M.A.M., G.V.H., D.B.C., and A.F.).

Author contributions: Conceptualization: M.A.M., J.H.M., G.V.H., R.G., D.B.C., J.P.D., D.O.O., and A.F. Methodology: M.A.M., J.H.M., G.V.H., R.G., and A.F. Investigation: M.A.M., J.H.M., G.V.H., R.G., and A.F. Funding acquisition: M.A.M., J.H.M., G.V.H., D.B.C., J.P.D., and A.F. Project administration: M.A.M., J.H.M., G.V.H., and A.F. Supervision: M.A.M., G.V.H., D.B.C., J.P.D., and D.O.O. Writing—original draft: M.A.M., J.H.M., G.V.H., R.G., and A.F. Writing—review and editing: M.A.M., J.H.M., R.G., G.V.H., D.B.C., J.P.D., D.O.O., and A.F.

Competing interests: The authors declare that they have no competing interests.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

This PDF file includes:

Tables S1 to S6

Secondary Pre-Clinical Trial Outcomes and Data Post-Processing and Measurement of CSD-Related Parameters

Click here for additional data file. (348.8KB, pdf)

REFERENCES AND NOTES

  • 1.R. Diaz-Arrastia, P. M. Kochanek, P. Bergold, K. Kenney, C. E. Marx, C. J. B. Grimes, L. T. C. Y. Loh, L. T. C. G. E. Adam, D. Oskvig, K. C. Curley, W. Salzer, Pharmacotherapy of traumatic brain injury: State of the science and the road forward: Report of the Department of Defense neurotrauma pharmacology workgroup. J. Neurotrauma 31, 135–158 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.R. K. Narayan, M. E. Michel, B. Ansell, A. Baethmann, A. Biegon, M. B. Bracken, M. R. Bullock, S. C. Choi, G. L. Clifton, C. F. Contant, W. M. Coplin, W. D. Dietrich, J. Ghajar, S. M. Grady, R. G. Grossman, E. D. Hall, W. Heetderks, D. A. Hovda, J. Jallo, R. L. Katz, N. Knoller, P. M. Kochanek, A. I. Maas, J. Majde, D. W. Marion, A. Marmarou, L. F. Marshall, T. K. McIntosh, E. Miller, N. Mohberg, J. P. Muizelaar, L. H. Pitts, P. Quinn, G. Riesenfeld, C. S. Robertson, K. I. Strauss, G. Teasdale, N. Temkin, R. Tuma, C. Wade, M. D. Walker, M. Weinrich, J. Whyte, J. Wilberger, A. B. Young, L. Yurkewicz, Clinical trials in head injury. J. Neurotrauma 19, 503–557 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.J. P. Dreier, The role of spreading depression, spreading depolarization and spreading ischemia in neurological disease. Nat. Med. 17, 439–447 (2011). [DOI] [PubMed] [Google Scholar]
  • 4.M. Lauritzen, J. P. Dreier, M. Fabricius, J. A. Hartings, R. Graf, A. J. Strong, Clinical relevance of cortical spreading depression in neurological disorders: Migraine, malignant stroke, subarachnoid and intracranial hemorrhage, and traumatic brain injury. J. Cereb. Blood Flow Metab. 31, 17–35 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.E. Parker, R. Aboghazleh, G. Mumby, R. Veksler, J. Ofer, J. Newton, R. Smith, L. Kamintsky, C. M. A. Jones, E. O’Keeffe, E. Kelly, K. Doelle, I. Roach, L. T. Yang, P. Moradi, J. M. Lin, A. J. Gleason, C. Atkinson, C. Bowen, K. D. Brewer, C. P. Doherty, M. Campbell, D. B. Clarke, G. van Hameren, D. Kaufer, A. Friedman, Concussion susceptibility is mediated by spreading depolarization-induced neurovascular dysfunction. Brain 145, 2049–2063 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.J. A. Hartings, N. Andaluz, M. R. Bullock, J. M. Hinzman, B. Mathern, C. Pahl, A. Puccio, L. A. Shutter, A. J. Strong, A. Vagal, J. A. Wilson, J. P. Dreier, L. B. Ngwenya, B. Foreman, L. Pahren, H. Lingsma, D. O. Okonkwo, Prognostic value of spreading depolarizations in patients with severe traumatic brain injury. JAMA Neurol. 77, 489–499 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.K. Schoknecht, M. Kikhia, C. L. Lemale, A. Liotta, S. Lublinsky, S. Mueller, P. Boehm-Sturm, A. Friedman, J. P. Dreier, The role of spreading depolarizations and electrographic seizures in early injury progression of the rat photothrombosis stroke model. J. Cereb. Blood Flow Metab. 41, 413–430 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.E. G. Lapilover, K. Lippmann, S. Salar, A. Maslarova, J. P. Dreier, U. Heinemann, A. Friedman, Peri-infarct blood-brain barrier dysfunction facilitates induction of spreading depolarization associated with epileptiform discharges. Neurobiol. Dis. 48, 495–506 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.J. P. Dreier, J. Woitzik, M. Fabricius, R. Bhatia, S. Major, C. Drenckhahn, T.-N. Lehmann, A. Sarrafzadeh, L. Willumsen, J. A. Hartings, O. W. Sakowitz, J. H. Seemann, A. Thieme, M. Lauritzen, A. J. Strong, Delayed ischaemic neurological deficits after subarachnoid haemorrhage are associated with clusters of spreading depolarizations. Brain 129, 3224–3237 (2006). [DOI] [PubMed] [Google Scholar]
  • 10.H. Nakamura, A. J. Strong, C. Dohmen, O. W. Sakowitz, S. Vollmar, M. Sué, L. Kracht, P. Hashemi, R. Bhatia, T. Yoshimine, J. P. Dreier, A. K. Dunn, R. Graf, Spreading depolarizations cycle around and enlarge focal ischaemic brain lesions. Brain 133, 1994–2006 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.J. A. Hartings, M. R. Bullock, D. O. Okonkwo, L. S. Murray, G. D. Murray, M. Fabricius, A. I. Maas, J. Woitzik, O. Sakowitz, B. Mathern, B. Roozenbeek, H. Lingsma, J. P. Dreier, A. M. Puccio, L. A. Shutter, C. Pahl, A. J. Strong, Spreading depolarisations and outcome after traumatic brain injury: A prospective observational study. Lancet Neurol. 10, 1058–1064 (2011). [DOI] [PubMed] [Google Scholar]
  • 12.J. A. Hartings, A. J. Strong, M. Fabricius, A. Manning, R. Bhatia, J. P. Dreier, A. T. Mazzeo, F. C. Tortella, M. R. Bullock, Spreading depolarizations and late secondary insults after traumatic brain injury. J. Neurotrauma 26, 1857–1866 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.J. P. Dreier, M. K. L. Winkler, S. Major, V. Horst, S. Lublinsky, V. Kola, C. L. Lemale, E.-J. Kang, A. Maslarova, I. Salur, J. Lückl, J. Platz, D. Jorks, A. I. Oliveira-Ferreira, K. Schoknecht, C. Reiffurth, D. Milakara, D. Wiesenthal, N. Hecht, N. F. Dengler, A. Liotta, S. Wolf, C. M. Kowoll, A. P. Schulte, E. Santos, E. Güresir, A. W. Unterberg, A. Sarrafzadeh, O. W. Sakowitz, H. Vatter, M. Reiner, G. Brinker, C. Dohmen, I. Shelef, G. Bohner, M. Scheel, P. Vajkoczy, J. A. Hartings, A. Friedman, P. Martus, J. Woitzik, Spreading depolarizations in ischaemia after subarachnoid haemorrhage, a diagnostic phase III study. Brain 145, 1264–1284 (2022). [DOI] [PubMed] [Google Scholar]
  • 14.J. Bouley, D. Y. Chung, C. Ayata, R. H. Brown, N. Henninger, Cortical spreading depression denotes concussion injury. J. Neurotrauma 36, 1008–1017 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.J. M. Pacheco, A. Hines-Lanham, C. Stratton, C. J. Mehos, K. E. McCurdy, N. J. Pinkowski, H. Zhang, C. W. Shuttleworth, R. A. Morton, Spreading depolarizations occur in mild traumatic brain injuries and are associated with postinjury behavior. eNeuro 6, ENEURO.0070-19 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Z. Y. Kerr, G. B. Wilkerson, S. V. Caswell, D. W. Currie, L. A. Pierpoint, E. B. Wasserman, S. B. Knowles, T. P. Dompier, R. D. Comstock, S. W. Marshall, The first decade of web-based sports injury surveillance: Descriptive epidemiology of injuries in united states high school football (2005-2006 Through 2013-2014) and national collegiate athletic association football (2004-2005 Through 2013-2014). J. Athl. Train. 53, 738–751 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.J. E. Wilk, R. K. Herrell, G. H. Wynn, L. A. Riviere, C. W. Hoge, Mild traumatic brain injury (concussion), posttraumatic stress disorder, and depression in U.S. soldiers involved in combat deployments. Psychosom. Med. 74, 249–257 (2012). [DOI] [PubMed] [Google Scholar]
  • 18.J. P. Dreier, C. L. Lemale, V. Kola, A. Friedman, K. Schoknecht, Spreading depolarization is not an epiphenomenon but the principal mechanism of the cytotoxic edema in various gray matter structures of the brain during stroke. Neuropharmacology 134, 189–207 (2018). [DOI] [PubMed] [Google Scholar]
  • 19.M. Wahl, M. Lauritzen, L. Schilling, Change of cerebrovascular reactivity after cortical spreading depression in cats and rats. Brain Res. 411, 72–80 (1987). [DOI] [PubMed] [Google Scholar]
  • 20.N. Hadjikhani, M. Sanchez del Rio, O. Wu, D. Schwartz, D. Bakker, B. Fischl, K. K. Kwong, F. M. Cutrer, B. R. Rosen, R. B. H. Tootell, A. G. Sorensen, M. A. Moskowitz, Mechanisms of migraine aura revealed by functional MRI in human visual cortex. Proc. Natl. Acad. Sci. U.S.A. 98, 4687–4692 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.M. Lauritzen, Pathophysiology of the migraine aura. The spreading depression theory. Brain 117, 199–210 (1994). [DOI] [PubMed] [Google Scholar]
  • 22.J. P. Dreier, K. Körner, N. Ebert, A. Görner, I. Rubin, T. Back, U. Lindauer, T. Wolf, A. Villringer, K. M. Einhäupl, M. Lauritzen, U. Dirnagl, Nitric oxide scavenging by hemoglobin or nitric oxide synthase inhibition by N-nitro-L-arginine induces cortical spreading ischemia when K+ is increased in the subarachnoid space. J. Cereb. Blood Flow Metab. 18, 978–990 (1998). [DOI] [PubMed] [Google Scholar]
  • 23.J. P. Dreier, S. Major, A. Manning, J. Woitzik, C. Drenckhahn, J. Steinbrink, C. Tolias, A. I. Oliveira-Ferreira, M. Fabricius, J. A. Hartings, P. Vajkoczy, M. Lauritzen, U. Dirnagl, G. Bohner, A. J. Strong; COSBID study group , Cortical spreading ischaemia is a novel process involved in ischaemic damage in patients with aneurysmal subarachnoid haemorrhage. Brain 132, 1866–1881 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.J. M. Hinzman, N. Andaluz, L. A. Shutter, D. O. Okonkwo, C. Pahl, A. J. Strong, J. P. Dreier, J. A. Hartings, Inverse neurovascular coupling to cortical spreading depolarizations in severe brain trauma. Brain 137, 2960–2972 (2014). [DOI] [PubMed] [Google Scholar]
  • 25.J. Woitzik, N. Hecht, A. Pinczolits, N. Sandow, S. Major, M. K. L. Winkler, S. Weber-Carstens, C. Dohmen, R. Graf, A. J. Strong, J. P. Dreier, P. Vajkoczy; COSBID study group , Propagation of cortical spreading depolarization in the human cortex after malignant stroke. Neurology 80, 1095–1102 (2013). [DOI] [PubMed] [Google Scholar]
  • 26.L. Medina-Ceja, K. Pardo-Peña, A. Morales-Villagrán, J. Ortega-Ibarra, S. López-Pérez, Increase in the extracellular glutamate level during seizures and electrical stimulation determined using a high temporal resolution technique. BMC Neurosci. 16, 11 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.A. I. Srienc, P.-P. Chiang, A. J. Schmitt, E. A. Newman, Cortical spreading depolarizations induced by surgical field blood in a mouse model of neurosurgery. J. Neurosurg. 132, 1820–1828 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.T. Amemori, J. Bures, Ketamine blockade of spreading depression: Rapid development of tolerance. Brain Res. 519, 351–354 (1990). [DOI] [PubMed] [Google Scholar]
  • 29.K. M. Reinhart, C. W. Shuttleworth, Ketamine reduces deleterious consequences of spreading depolarizations. Exp. Neurol. 305, 121–128 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.R. Sánchez-Porras, E. Santos, M. Schöll, C. Stock, Z. Zheng, P. Schiebel, B. Orakcioglu, A. W. Unterberg, O. W. Sakowitz, The effect of ketamine on optical and electrical characteristics of spreading depolarizations in gyrencephalic swine cortex. Neuropharmacology 84, 52–61 (2014). [DOI] [PubMed] [Google Scholar]
  • 31.R. Sánchez-Porras, M. Kentar, R. Zerelles, M. Geyer, C. Trenado, J. A. Hartings, J. Woitzik, J. P. Dreier, E. Santos, Eighteen-hour inhibitory effect of s-ketamine on potassium- and ischemia-induced spreading depolarizations in the gyrencephalic swine brain. Neuropharmacology 216, 109176 (2022). [DOI] [PubMed] [Google Scholar]
  • 32.R. Sánchez-Porras, E. Santos, M. Schöll, K. Kunzmann, C. Stock, H. Silos, A. W. Unterberg, O. W. Sakowitz, Ketamine modulation of the haemodynamic response to spreading depolarization in the gyrencephalic swine brain. J. Cereb. Blood Flow Metab. 37, 1720–1734 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Z. Zheng, M. Schoell, R. Sanchez-Porras, C. Diehl, A. Unterberg, O. W. Sakowitz, Spreading depolarization during the acute stage of experimental subarachnoid hemorrhage in mice. Acta Neurochir. Suppl. 127, 97–103 (2020). [DOI] [PubMed] [Google Scholar]
  • 34.R. Marrannes, R. Willems, E. De Prins, A. Wauquier, Evidence for a role of the N-methyl-D-aspartate (NMDA) receptor in cortical spreading depression in the rat. Brain Res. 457, 226–240 (1988). [DOI] [PubMed] [Google Scholar]
  • 35.J. Hernándéz-Cáceres, R. Macias-González, G. Brozek, J. Bures, Systemic ketamine blocks cortical spreading depression but does not delay the onset of terminal anoxic depolarization in rats. Brain Res. 437, 360–364 (1987). [DOI] [PubMed] [Google Scholar]
  • 36.N. A. Gorelova, V. I. Koroleva, T. Amemori, V. Pavlík, J. Burĕs, Ketamine blockade of cortical spreading depression in rats. Electroencephalogr. Clin. Neurophysiol. 66, 440–447 (1987). [DOI] [PubMed] [Google Scholar]
  • 37.O. W. Sakowitz, K. L. Kiening, K. L. Krajewski, A. S. Sarrafzadeh, M. Fabricius, A. J. Strong, A. W. Unterberg, J. P. Dreier, Preliminary evidence that ketamine inhibits spreading depolarizations in acute human brain injury. Stroke 40, e519–e522 (2009). [DOI] [PubMed] [Google Scholar]
  • 38.A. P. Carlson, M. Abbas, R. L. Alunday, F. Qeadan, C. W. Shuttleworth, Spreading depolarization in acute brain injury inhibited by ketamine: A prospective, randomized, multiple crossover trial. J. Neurosurg. , 1–7 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.E. Santos, A. Olivares-Rivera, S. Major, R. Sánchez-Porras, L. Uhlmann, K. Kunzmann, R. Zerelles, M. Kentar, V. Kola, A. H. Aguilera, M. G. Herrera, C. L. Lemale, J. Woitzik, J. A. Hartings, O. W. Sakowitz, A. W. Unterberg, J. P. Dreier, Lasting s-ketamine block of spreading depolarizations in subarachnoid hemorrhage: A retrospective cohort study. Crit. Care 23, 427 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.D. N. Hertle, J. P. Dreier, J. Woitzik, J. A. Hartings, R. Bullock, D. O. Okonkwo, L. A. Shutter, S. Vidgeon, A. J. Strong, C. Kowoll, C. Dohmen, J. Diedler, R. Veltkamp, T. Bruckner, A. W. Unterberg, O. W. Sakowitz, Effect of analgesics and sedatives on the occurrence of spreading depolarizations accompanying acute brain injury. Brain 135, 2390–2398 (2012). [DOI] [PubMed] [Google Scholar]
  • 41.A. J. Schiefecker, R. Beer, B. Pfausler, P. Lackner, G. Broessner, I. Unterberger, F. Sohm, M. Mulino, C. Thome, C. Humpel, E. Schmutzhard, R. Helbok, Clusters of cortical spreading depolarizations in a patient with intracerebral hemorrhage: A multimodal neuromonitoring study. Neurocrit. Care 22, 293–298 (2015). [DOI] [PubMed] [Google Scholar]
  • 42.P. Zanos, R. Moaddel, P. J. Morris, L. M. Riggs, J. N. Highland, P. Georgiou, E. F. R. Pereira, E. X. Albuquerque, C. J. Thomas, C. A. Zarate, T. D. Gould, J. M. Witkin, Ketamine and ketamine metabolite pharmacology: Insights into therapeutic mechanisms. Pharmacol. Rev. 70, 621–660 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.R. D. Andrew, E. Farkas, J. A. Hartings, K. C. Brennan, O. Herreras, M. Müller, S. A. Kirov, C. Ayata, N. Ollen-Bittle, C. Reiffurth, O. Revah, R. M. Robertson, K. D. Dawson-Scully, G. Ullah, J. P. Dreier, Questioning glutamate excitotoxicity in acute brain damage: The importance of spreading depolarization. Neurocrit. Care 37, 11–30 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.L. M. Mohammad, M. Abbas, C. W. Shuttleworth, R. Ahmadian, A. Bhat, D. A. Hill, A. P. Carlson, Spreading depolarization may represent a novel mechanism for delayed fluctuating neurological deficit after chronic subdural hematoma evacuation. J. Neurosurg. 134, 1294–1302 (2020). [DOI] [PubMed] [Google Scholar]
  • 45.B. Winblad, R. W. Jones, Y. Wirth, A. Stöffler, H. J. Möbius, Memantine in moderate to severe Alzheimer’s disease: A meta-analysis of randomised clinical trials. Dement. Geriatr. Cogn. Disord. 24, 20–27 (2007). [DOI] [PubMed] [Google Scholar]
  • 46.J. Kornhuber, M. Weller, K. Schoppmeyer, P. Riederer, Amantadine and memantine are NMDA receptor antagonists with neuroprotective properties. J. Neural Transm. Suppl. 43, 91–104 (1994). [PubMed] [Google Scholar]
  • 47.V. L. Rao, A. Dogan, K. G. Todd, K. K. Bowen, R. J. Dempsey, Neuroprotection by memantine, a non-competitive NMDA receptor antagonist after traumatic brain injury in rats. Brain Res. 911, 96–100 (2001). [DOI] [PubMed] [Google Scholar]
  • 48.A. Dogan, M. A. Eras, V. L. Rao, R. J. Dempsey, Protective effects of memantine against ischemia-reperfusion injury in spontaneously hypertensive rats. Acta Neurochir. 141, 1107–1113 (1999). [DOI] [PubMed] [Google Scholar]
  • 49.M. Peeters, M. J. Gunthorpe, P. J. L. M. Strijbos, P. Goldsmith, N. Upton, M. F. James, Effects of pan- and subtype-selective N-methyl-D-aspartate receptor antagonists on cortical spreading depression in the rat: Therapeutic potential for migraine. J. Pharmacol. Exp. Ther. 321, 564–572 (2007). [DOI] [PubMed] [Google Scholar]
  • 50.K. M. Reinhart, A. Humphrey, K. C. Brennan, A. P. Carlson, C. W. Shuttleworth, Memantine improves recovery after spreading depolarization in brain slices and can be considered for future clinical trials. Neurocrit. Care 35, 135–145 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.M. Mokhtari, H. Nayeb-Aghaei, M. Kouchek, M. M. Miri, R. Goharani, A. Amoozandeh, S. Akhavan Salamat, M. Sistanizad, Effect of memantine on serum levels of neuron-specific enolase and on the Glasgow coma scale in patients with moderate traumatic brain injury. J. Clin. Pharmacol. 58, 42–47 (2018). [DOI] [PubMed] [Google Scholar]
  • 52.H. T. Zhao, M. C. Tuohy, D. Chow, M. G. Kozberg, S. H. Kim, M. A. Shaik, E. M. C. Hillman, Neurovascular dynamics of repeated cortical spreading depolarizations after acute brain injury. Cell Rep. 37, 109794 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.T. Araki, R. Wake, T. Miyaoka, K. Kawakami, M. Nagahama, M. Furuya, E. Limoa, K. Liaury, S. Hashioka, K. Murotani, J. Horiguchi, The effects of combine treatment of memantine and donepezil on Alzheimer’s disease patients and its relationship with cerebral blood flow in the prefrontal area. Int. J. Geriatr. Psychiatry 29, 881–889 (2014). [DOI] [PubMed] [Google Scholar]
  • 54.L. Lu, A. D. Hogan-Cann, A. K. Globa, P. Lu, J. I. Nagy, S. X. Bamji, C. M. Anderson, Astrocytes drive cortical vasodilatory signaling by activating endothelial NMDA receptors. J. Cereb. Blood Flow Metab. 39, 481–496 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.A. D. Hogan-Cann, P. Lu, C. M. Anderson, Endothelial NMDA receptors mediate activity-dependent brain hemodynamic responses in mice. Proc. Natl. Acad. Sci. U.S.A. 116, 10229–10231 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.U. Vazana, R. Veksler, G. S. Pell, O. Prager, M. Fassler, Y. Chassidim, Y. Roth, H. Shahar, A. Zangen, R. Raccah, E. Onesti, M. Ceccanti, C. Colonnese, A. Santoro, M. Salvati, A. D’Elia, V. Nucciarelli, M. Inghilleri, A. Friedman, Glutamate-mediated blood-brain barrier opening: Implications for neuroprotection and drug delivery. J. Neurosci. 36, 7727–7739 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.H. Kumar, B. Sharma, Memantine ameliorates autistic behavior, biochemistry & blood brain barrier impairments in rats. Brain Res. Bull. 124, 27–39 (2016). [DOI] [PubMed] [Google Scholar]
  • 58.G. B. Effgen, B. Morrison, Memantine reduced cell death, astrogliosis, and functional deficits in an in vitro model of repetitive mild traumatic brain injury. J. Neurotrauma 34, 934–942 (2017). [DOI] [PubMed] [Google Scholar]
  • 59.Z. Mei, J. Qiu, S. Alcon, J. Hashim, A. Rotenberg, Y. Sun, W. P. Meehan, R. Mannix, Memantine improves outcomes after repetitive traumatic brain injury. Behav. Brain Res. 340, 195–204 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.H. Kafi, J. Salamzadeh, N. Beladimoghadam, M. Sistanizad, M. Kouchek, Study of the neuroprotective effects of memantine in patients with mild to moderate ischemic stroke. Iran. J. Pharm. Res. 13, 591–598 (2014). [PMC free article] [PubMed] [Google Scholar]
  • 61.M. B. Hesselink, B. G. De Boer, D. D. Breimer, W. Danysz, Brain penetration and in vivo recovery of NMDA receptor antagonists amantadine and memantine: A quantitative microdialysis study. Pharm. Res. 16, 637–642 (1999). [DOI] [PubMed] [Google Scholar]
  • 62.J. Kornhuber, J. Bormann, W. Retz, M. Hübers, P. Riederer, Memantine displaces [3H]MK-801 at therapeutic concentrations in postmortem human frontal cortex. Eur. J. Pharmacol. 166, 589–590 (1989). [DOI] [PubMed] [Google Scholar]
  • 63.J. Kornhuber, G. Quack, Cerebrospinal fluid and serum concentrations of the N-methyl-D-aspartate (NMDA) receptor antagonist memantine in man. Neurosci. Lett. 195, 137–139 (1995). [DOI] [PubMed] [Google Scholar]
  • 64.S. J. Thomas, G. T. Grossberg, Memantine: A review of studies into its safety and efficacy in treating Alzheimer’s disease and other dementias. Clin. Interv. Aging 4, 367–377 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.M. G. Beconi, D. Howland, L. Park, K. Lyons, J. Giuliano, C. Dominguez, I. Munoz-Sanjuan, R. Pacifici, Pharmacokinetics of memantine in rats and mice. PLOS Curr. 3, RRN1291 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.C. Willis, S. Lybrand, N. Bellamy, Excitatory amino acid inhibitors for traumatic brain injury. Cochrane Database Syst. Rev. 2003, CD003986 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.K. Beauchamp, H. Mutlak, W. R. Smith, E. Shohami, P. F. Stahel, Pharmacology of traumatic brain injury: Where is the “golden bullet”? Mol. Med. 14, 731–740 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.I. Masse, L. Moquin, C. Bouchard, A. Gratton, L. De Beaumont, Efficacy of prophylactic versus therapeutic administration of the NMDA receptor antagonist MK-801 on the acute neurochemical response to a concussion in a rat model combining force and rotation. J. Neurosurg. 136, 1650–1659 (2022). [DOI] [PubMed] [Google Scholar]
  • 69.J. P. Dreier, M. Fabricius, C. Ayata, O. W. Sakowitz, C. W. Shuttleworth, C. Dohmen, R. Graf, P. Vajkoczy, R. Helbok, M. Suzuki, A. J. Schiefecker, S. Major, M. K. L. Winkler, E.-J. Kang, D. Milakara, A. I. Oliveira-Ferreira, C. Reiffurth, G. S. Revankar, K. Sugimoto, N. F. Dengler, N. Hecht, B. Foreman, B. Feyen, D. Kondziella, C. K. Friberg, H. Piilgaard, E. S. Rosenthal, M. B. Westover, A. Maslarova, E. Santos, D. Hertle, R. Sánchez-Porras, S. L. Jewell, B. Balança, J. Platz, J. M. Hinzman, J. Lückl, K. Schoknecht, M. Schöll, C. Drenckhahn, D. Feuerstein, N. Eriksen, V. Horst, J. S. Bretz, P. Jahnke, M. Scheel, G. Bohner, E. Rostrup, B. Pakkenberg, U. Heinemann, J. Claassen, A. P. Carlson, C. M. Kowoll, S. Lublinsky, Y. Chassidim, I. Shelef, A. Friedman, G. Brinker, M. Reiner, S. A. Kirov, R. D. Andrew, E. Farkas, E. Güresir, H. Vatter, L. S. Chung, K. C. Brennan, T. Lieutaud, S. Marinesco, A. I. R. Maas, J. Sahuquillo, M. A. Dahlem, F. Richter, O. Herreras, M. G. Boutelle, D. O. Okonkwo, M. R. Bullock, O. W. Witte, P. Martus, A. M. J. M. van den Maagdenberg, M. D. Ferrari, R. M. Dijkhuizen, L. A. Shutter, N. Andaluz, A. P. Schulte, B. M. Vicar, T. Watanabe, J. Woitzik, M. Lauritzen, A. J. Strong, J. A. Hartings, Recording, analysis, and interpretation of spreading depolarizations in neurointensive care: Review and recommendations of the COSBID research group. J. Cereb. Blood Flow Metab. 37, 1595–1625 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.G. van Hameren, J. Muradov, A. Minarik, R. Aboghazleh, S. Orr, S. Cort, K. Andrews, C. McKenna, N. T. Pham, M. A. MacLean, A. Friedman, Mitochondrial dysfunction underlies impaired neurovascular coupling following traumatic brain injury. Neurobiol. Dis. 186, 106269 (2023). [DOI] [PubMed] [Google Scholar]
  • 71.L. M. Oberman, S. Exley, N. S. Philip, S. H. Siddiqi, M. M. Adamson, D. L. Brody, Use of repetitive transcranial magnetic stimulation in the treatment of neuropsychiatric and neurocognitive symptoms associated with concussion in military populations. J. Head Trauma Rehabil. 35, 388–400 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Y. Gursoy-Ozdemir, J. Qiu, N. Matsuoka, H. Bolay, D. Bermpohl, H. Jin, X. Wang, G. A. Rosenberg, E. H. Lo, M. A. Moskowitz, Cortical spreading depression activates and upregulates MMP-9. J. Clin. Invest. 113, 1447–1455 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73.C. H. Acosta, G. A. Clemons, C. T. Citadin, W. C. Carr, M. S. B. Udo, V. Tesic, H. W. Sanicola, A. H. Freelin, J. B. Toms, J. D. Jordan, B. Guthikonda, K. M. Rodgers, C. Y.-C. Wu, R. H.-C. Lee, H. W. Lin, PRMT7 can prevent neurovascular uncoupling, blood-brain barrier permeability, and mitochondrial dysfunction in repetitive and mild traumatic brain injury. Exp. Neurol. 366, 114445 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74.C. A. Tagge, A. M. Fisher, O. V. Minaeva, A. Gaudreau-Balderrama, J. A. Moncaster, X.-L. Zhang, M. W. Wojnarowicz, N. Casey, H. Lu, O. N. Kokiko-Cochran, S. Saman, M. Ericsson, K. D. Onos, R. Veksler, V. V. Senatorov, A. K. Jr, X. Z. Zhou, O. Miry, L. R. Vose, K. R. Gopaul, C. Upreti, C. J. Nowinski, R. C. Cantu, V. E. Alvarez, A. M. Hildebrandt, E. S. Franz, J. Konrad, J. A. Hamilton, N. Hua, Y. Tripodis, A. T. Anderson, G. R. Howell, D. Kaufer, G. F. Hall, K. P. Lu, R. M. Ransohoff, R. O. Cleveland, N. W. Kowall, T. D. Stein, B. T. Lamb, B. R. Huber, W. C. Moss, A. Friedman, P. K. Stanton, A. C. M. Kee, L. E. Goldstein, Concussion, microvascular injury, and early tauopathy in young athletes after impact head injury and an impact concussion mouse model. Brain 141, 422–458 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75.I. Weissberg, R. Veksler, L. Kamintsky, R. Saar-Ashkenazy, D. Z. Milikovsky, I. Shelef, A. Friedman, Imaging blood-brain barrier dysfunction in football players. JAMA Neurol. 71, 1453–1455 (2014). [DOI] [PubMed] [Google Scholar]
  • 76.R. Veksler, U. Vazana, Y. Serlin, O. Prager, J. Ofer, N. Shemen, A. M. Fisher, O. Minaeva, N. Hua, R. Saar-Ashkenazy, I. Benou, T. Riklin-Raviv, E. Parker, G. Mumby, L. Kamintsky, S. Beyea, C. V. Bowen, I. Shelef, E. O’Keeffe, M. Campbell, D. Kaufer, L. E. Goldstein, A. Friedman, Slow blood-to-brain transport underlies enduring barrier dysfunction in American football players. Brain 143, 1826–1842 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.E. O’Keeffe, E. Kelly, Y. Liu, C. Giordano, E. Wallace, M. Hynes, S. Tiernan, A. Meagher, C. Greene, S. Hughes, T. Burke, J. Kealy, N. Doyle, A. Hay, M. Farrell, G. A. Grant, A. Friedman, R. Veksler, M. G. Molloy, J. F. Meaney, N. Pender, D. Camarillo, C. P. Doherty, M. Campbell, Dynamic blood-brain barrier regulation in mild traumatic brain injury. J. Neurotrauma 37, 347–356 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 78.D. Shlosberg, M. Benifla, D. Kaufer, A. Friedman, Blood-brain barrier breakdown as a therapeutic target in traumatic brain injury. Nat. Rev. Neurol. 6, 393–403 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 79.B. Obermeier, R. Daneman, R. M. Ransohoff, Development, maintenance and disruption of the blood-brain barrier. Nat. Med. 19, 1584–1596 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 80.N. P. du Sert, V. Hurst, A. Ahluwalia, S. Alam, M. T. Avey, M. Baker, W. J. Browne, A. Clark, I. C. Cuthill, U. Dirnagl, M. Emerson, P. Garner, S. T. Holgate, D. W. Howells, N. A. Karp, S. E. Lazic, K. Lidster, C. J. MacCallum, M. Macleod, E. J. Pearl, O. H. Petersen, F. Rawle, P. Reynolds, K. Rooney, E. S. Sena, S. D. Silberberg, T. Steckler, H. Würbel, I. Boutron, The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. PLOS Biol. 18, e3000410 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81.A. Marmarou, M. A. Foda, W. van den Brink, J. Campbell, H. Kita, K. Demetriadou, A new model of diffuse brain injury in rats. Part I: Pathophysiology and biomechanics. J. Neurosurg. 80, 291–300 (1994). [DOI] [PubMed] [Google Scholar]
  • 82.M. J. Kane, M. Angoa-Pérez, D. I. Briggs, D. C. Viano, C. W. Kreipke, D. M. Kuhn, A mouse model of human repetitive mild traumatic brain injury. J. Neurosci. Methods 203, 41–49 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 83.O. Prager, Y. Chassidim, C. Klein, H. Levi, I. Shelef, A. Friedman, Dynamic in vivo imaging of cerebral blood flow and blood-brain barrier permeability. Neuroimage 49, 337–344 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 84.X. Li, P. Martinez-Lozano Sinues, R. Dallmann, L. Bregy, M. Hollmén, S. Proulx, S. A. Brown, M. Detmar, M. Kohler, R. Zenobi, Drug pharmacokinetics determined by real-time analysis of mouse breath. Angew. Chem. Int. Ed. Engl. 54, 7815–7818 (2015). [DOI] [PubMed] [Google Scholar]
  • 85.Dotmatics, GraphPad: Randomisation tool; https://www.graphpad.com/quickcalcs/randomize1/.
  • 86.R. Aboghazleh, E. Parker, L. T. Yang, D. Kaufer, J. P. Dreier, A. Friedman, G. van Hameren, Brainstem and cortical spreading depolarization in a closed head injury rat model. Int. J. Mol. Sci. 22, 11642 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Tables S1 to S6

Secondary Pre-Clinical Trial Outcomes and Data Post-Processing and Measurement of CSD-Related Parameters

Click here for additional data file. (348.8KB, pdf)

Articles from Science Advances are provided here courtesy of American Association for the Advancement of Science

RESOURCES