A systematic review of neuroprotective strategies after cardiac arrest: from bench to bedside (Part I – Protection via specific pathways)

N-Methyl-D-aspartate (NMDA) receptor antagonist

Excessive activation of NMDA receptors under conditions of energy substrate depletion results in glutamate excitotoxicity [11]. This pathology has been demonstrated in animal models of global brain ischemia [12, 13] and in human patients recovering from CA [14]. Consequently, much effort has gone into targeting glutamate receptor subtypes in an attempt to limit ischemic brain damage.

However, in the setting of global brain ischemia, pre or post CA infusion with the NMDA receptor antagonist MK-801 exacerbated post-resuscitation neurological deficits in the dog model [15]. Similarly post-CA intravenous treatment with the NMDA receptor antagonist GPI 3000 was associated with poor survival and neurologic function along with increased neuronal death in the neocortex and hippocampus in dogs subjected to CA and CPR [16]. An interaction between ischemia and competitive NMDA receptor antagonism may be the cause of deleterious outcomes [15]. Indeed, pharmacological attempts to use NMDA receptor blockers for stroke have had very limited clinical success because these compounds produce additional adverse side effects such as profound psychotomimetic behavioral changes as well as intrinsic neurotoxicity at proposed neuroprotective concentrations [17–19].

Lamotrigine

An alternative approach to ameliorate glutamate excitotoxicity can be to inhibit the presynaptic release of glutamate by using the phenyltriazine seizure drug lamotrigine. Lamotrigine has been previously shown to improve survival and neurologic function as well as hippocampal neurohistopathology up to 21 days in a gerbil model of bilateral cerebral artery occlusion [20]. Similar neuroprotection was consistently found in a rat model of CA [21]. Lamotrigine administered 15 minutes after 8.5 minutes of CA significantly improved the number of viable cells compared to non-treated CA rats at 3 weeks after CA. The mechanism of action of lamotrigine is purported to be inhibition of pre-synaptic glutamate release by blockade of voltage dependent sodium channels [22]. Additionally, decreased sodium influx can prevent intracellular calcium overload, further favoring outcome [23].

Xenon

The inert gas xenon has both anesthetic properties and a pharmaceutical profile of low-affinity use-dependent NMDA receptor antagonism at nonanesthetic concentrations [24]. It may avoid or reduce adverse side effects and potential neurotoxicity associated with prototypical NMDA receptor antagonists [19, 25]. A number of pre-clinical studies have shown potential utility for xenon as a neuroprotectant when administered alone or in combination with TH [26–29]. In a pig model of CA and CPR, combined mild hypothermia (33 degrees C for 16 hours) with xenon (70% xenon and 30% O2 for 1 hour) showed significantly improved neurologic deficit scores over 5 days post-ROSC relative to untreated controls [30]. While neuronal viability was similar to mild TH, combined mild TH + xenon treatment was associated with less astrogliosis and microgliosis, suggesting synergistic protection from the combination.

Other xenon studies have been conducted without combined TH. Pigs underwent 8 minutes of ventricular fibrillation (VF) arrest followed by 5 minutes of CPR [31]; one hour after ROSC the pigs were randomized to 70% xenon in 30% O₂ for 1 hour or 70% xenon in 30% O₂ for 5 hours or 70% nitrogen in 30% O₂ (control). Improvement in neurologic function was transient as benefit was only seen in the first 3 post-ischemia days with no significant difference observed on post-ischemia day 4. At necropsy on day 5, there were significantly reduced numbers of necrotic neurons in pigs ventilated with xenon [31].

A similar study evaluated pigs receiving 70% xenon/30% O₂ vs. 69% N₂/1% isoflurane/30% O₂ vs. 70% N₂/30% O₂ for one hour starting 10 minutes after ROSC; findings revealed no differences with respect to neurologic deficit scores over 4 days or neurohistopathological analysis at day 5 [26].

However, a separate xenon study without TH revealed that the time to initiate xenon treatment appears to be critical [32]. Late rather than early treatment with xenon after CA resulted in neurologic benefit. Xenon 70%/O₂ 30% administered for one or five hours starting one hour after ROSC was associated with improved neurologic function compared to the control group receiving 70% N₂/30% O₂[32].

As of this writing, a search of the ClinicalTrials.gov site for xenon and cardiac arrest reveals that there are two human clinical trials recruiting subjects (NCT00879892; NCT01262729). These two clinical trials may provide a better picture about the translation potential of xenon use following CA.

Argon

Argon is a non-anesthetic gas that is relatively cost-effective and available compared to xenon. Intriguing data have shown neuroprotective properties of argon in animal models of global brain ischemia secondary to CA. Rats underwent 7 minutes of CA followed by 3 minutes of CPR and one hour after CPR were then randomized to ventilation with 70% argon/30% O₂ vs. 70% N₂/30% O₂ for one hour [33]. Argon treated rats had significantly improved neurologic deficit scores for 7 days and demonstrated statistically better neurohistopathological outcome in the neocortex and the CA3/4 region of the hippocampus. In a separate pig model of CA (8 minutes) and CPR (5 minutes), four hour ventilation with 70% argon/30% O₂ after ROSC resulted in significantly improved neurologic recovery and less histopathology in brain tissue [34].

In general, the mechanisms of neuroprotection by argon are poorly understood. The mechanism of neuroprotection afforded by argon may involve activation of anti-apoptotic signaling through increasing BcL-xL or Bcl-2 levels, thus promoting cell survival [35]. Argon may also affect gamma aminobutyric acid type A receptors, although further data is required to clarify whether this is the mechanism for cytoprotection [27].

Ischemic post-conditioning (IPC)

Although resuscitation requires reperfusion of ischemic tissue with oxygenated blood to restore aerobic metabolism and organ function, reperfusion concomitantly activates multiple pathogenic mechanisms and results in what is collectively known as reperfusion injury [23]. At the center of reperfusion injury are mitochondria, playing a critical role as effectors and targets of injury. Mitochondrial calcium overload can worsen cell damage by compromising mitochondrial capability to sustain oxidative phosphorylation [36] and by promoting the release of pro-apoptotic factors [37]. Although interrupting chest compressions for an extended period during CPR may lead to poor outcome [38], IPC can limit reperfusion injury, thus exerting anti-apoptotic effects through mitochondrial protection [23, 39–42]. In pigs subjected to 15 minutes of VF arrest, IPC was given during the first 3 minutes of CPR as 4 cycles of 20 seconds of chest compressions followed by a 20 second pause and compared to a control group of CPR without IPC. Both groups received mild TH for 12 hours after CPR. Left ventricular ejection fraction (LVEF) and neurological functional recovery were significantly better in the CPR + IPC group at 2 days compared to regular CPR. The same investigators further compared the effects of standard CPR, CPR + IPC, CPR + cardioprotective vasodilator therapy (CVT), or CPR + IPC + CVT [43]. CVT (IV sodium nitroprusside 2 mg and adenosine 24 mg) was administered during the first minute of CPR. CPR + IPC showed significantly better 48-hour survival, LVEF, neurological assessment and histological scores compared to the other groups.

Caspase-3 inhibitor

Caspase-3 activation plays a central role in apoptotic pathways following brain injury [44]. zDEVD-FMK directly interrupts the apoptotic pathway by inhibiting caspase-3 in focal brain ischemia [45]. However, the same benefits were not observed in the setting of global ischemia secondary to CA. In a rat model of CA and CPR, a 7-day intracerebroventricular (ICV) infusion of the caspase-3 inhibitor zDEVD-FMK after ROSC did not improve neurologic deficit or neurohistopathology at 1, 3 and 7 days [46]. The discrepancy may be attributed to either whole body ischemia/reperfusion syndrome and/or a systemic inflammatory response syndrome, which increases the pathological complexity of brain injury in CA. Additionally, the lack of protective effects may result from high zDEVD-FMK clearance inside the ICV system and/or an impaired uptake into the brain parenchyma [46].

Buffering metabolic acidosis

CA prompts a shift to anaerobic metabolism leading to rapid development of intense and sustained intracellular acidosis systemically and locally in the brain. This subsequently triggers a battery of pathological processes resulting in cytosolic calcium overload. In the “low flow” or “no flow” state followed by reperfusion, both CO₂ elimination by ventilation and metabolic correction by HCO₃ buffering may be necessary to optimize pH recovery [47]. In this context, buffering metabolic acidosis using sodium bicarbonate (NaHCO₃) and carbicarb have been evaluated as neuroprotective strategies, but the efficacy remains controversial.

After dogs underwent VF arrest for either 5 minutes (short) or 15 minutes (long), NaHCO₃ was administered as 1 mmol/kg initially with additional doses as necessary to correct base deficit to -5 mEq/l [48]. Dogs treated with NaHCO₃ demonstrated equivalent rates of ROSC and 24 hour survival with short arrest but significantly improved ROSC and survival in the long arrest group when compared to controls. Acidosis was also significantly less in the prolonged arrest group treated with NaHCO₃, while coronary and systemic perfusion pressures were significantly better. Neurologic deficit scores were improved in both groups treated with NaHCO₃, but histopathologic staining was not different.

Carbicarb is an alkalinizing agent given to combat acidosis inherent to CA. While NaHCO₃ produces increased CO₂ and a paradoxical cerebral acidosis, carbicarb is a mixture of HCO₃ and sodium carbonate that does not produce paradoxical cerebral acidosis. In a rat model of CA (8 minutes) and CPR, low dose (3 ml/kg) carbicarb post-ROSC treatment had positive outcomes for 7-day survival, neurologic deficits and hippocampal cell death compared to controls [49]. The protection was secondary to attenuation of brain pH decreases as well as an increase in post-resuscitation mean arterial pressure (MAP). A high dose (6 ml/kg) group completely neutralized pH but had negative outcomes when compared to the control group with increased neuronal cell death, increased neurologic deficit and decreased MAP [49].

In CA patients, retrospective studies showed some benefits of NaHCO₃[50, 51], but a prospective randomized double blind clinical trial did not [52]. In one of the retrospective human studies, low versus high dose NaHCO₃ administration during resuscitation had no impact on immediate ROSC but low dose NaHCO₃ favored long-term outcomes including survival and neurologic outcome [51]. The authors found that administering more than 1 mEq/kg NaHCO₃ during CPR had a negative impact on long-term survival and neurologic outcome. In another retrospective study associated with the Brain Resuscitation Clinical Trial III [50], patients were included in the study if they suffered out of hospital CA and advanced cardiovascular life support (ACLS) was initiated within 30 minutes by emergency medical services. NaHCO₃ administration was optional in the study and was preferentially more frequently administered at some institutions than others. Patients treated in the “high NaHCO₃” user sites had significantly better rates of ROSC and neurologic outcomes. One potential confounder in this study is that the patients who received the higher NaHCO₃ doses also had a significantly shorter time to initiation of ACLS (1.7 minutes). One prospective randomized double-blinded clinical trial in 875 pre-hospital CA patients showed empirical early administration of NaHCO₃ (1 mEq/kg) had no effect on overall outcomes in brief (<5 minutes) and moderate (5-15 minutes) CA. However, there was a trend toward improved outcome in prolonged (>15 minutes) CA [52].

As of this writing, a search of the ClinicalTrials.gov site for bicarbonate AND cardiac arrest reveals 1 human trial that is not yet recruiting subjects (NCT01377337). This study proposed to investigate the survival at CPR termination impact resulting from IV administration of 1 mEq/kg NaHCO₃ after the first IV dose of epinephrine and up to 2 additional doses administered at 5 to 10 minute intervals during CPR.

The discouraging results of attempts to apply this approach from bench to bedside may be due to opposing mechanisms of cellular action. High sodium may help by eliminating H + via the antiporter, but intracellular Na + accumulation may lead to decreased Ca2+ elimination via the Na + -Ca2+ exchanger and Ca2+ accumulation [47]. Ca2+ overload is associated with deleterious cellular consequences that may compromise anti-acidosis protective effects. In addition, acidosis depends on the patient’s medical status before arrest, along with the duration and efficacy of CPR. Treatment may be improved by titration guided by venous or arterial pH measurement. Arrest times longer than 15 minutes will most likely not benefit from bicarbonate administration [47]. Therefore, NaHCO₃ was recommended for use during CA resuscitation in 1974 but was removed from the CPR algorithm in 1986 secondary to the potential to increase acidosis. The 1992 and 2000 AHA guidelines de-emphasized the use of NaHCO₃.

Fluoxetine

Fluoxetine is a selective serotonin reuptake inhibitor and has been shown to protect neurons through anti-inflammatory effects in focal brain ischemia [54]. In mouse models of CA and CPR, high dose (10 mg/kg) but not low dose (5 mg/kg) fluoxetine was associated with decreased histological damage in the caudate putamen as well as decreased sensorimotor deficits at 3 days when administered 30 minutes after ROSC [55]. No difference in the hippocampus was observed with either dose.

Matrix metalloproteinase-9 (MMP-9) inhibitor

MMP-9 activation plays an important role in blood–brain barrier (BBB) disruption, resulting in brain edema and inflammation following brain ischemia [53]. SB-3CT is a specific inhibitor of MMP-9. In a rat model, CA was induced by occlusion of the airway. CPR was started 1 minute after CA onset. Compared to control CA rats, intraperitoneal (IP) injections of SB-3CT at 5 minutes after ROSC was associated with significantly reduced brain tissue expression of MMP-9 protein and messenger ribonucleic acid, brain water content, Evans Blue content and cytokine levels at 3, 9, 24 and 48 hours [56].