A systematic review of neuroprotective strategies after cardiac arrest: from bench to bedside (part II-comprehensive protection)

A literature search was conducted of articles indexed in Medline and published between 1980 and October 2013 using combination of keywords including “brain injury”, “cardiac arrest”, “neuroprotection”, “cerebral protection”, “cardiopulmonary resuscitation”, “global ischemia”, “global cerebral ischemia”, and “global brain ischemia” (Table 1). Bibliographies of relevant articles were cross-referenced for pertinent articles. Articles were selected for review if postulated mechanisms of neuroprotection and some measure of neurologic outcomes were included. Only neuroprotective strategies tested in animal models relevant to global brain ischemia associated with CA were reviewed. Case reports, pediatric studies and articles not written in English were excluded. Studies of therapies administered before CA were not included as our goal was to investigate potential therapies to improve neurologic outcome that can be employed in the clinical setting (during or after CPR). Similarly, the many studies related to neuroprotection from anesthetic agents were not included, as administration of anesthetic agents during or immediately after CPR may be impractical or associated with undesirable hemodynamic effects. Due to the pathophysiologic differences between focal and global cerebral ischemia, the extensive literature regarding neuroprotective strategies in focal cerebral ischemia is acknowledged but not included in this review. In this part, a total of 17 preclinical studies were reviewed. We focus on several strategies that demonstrate comprehensive neuroprotective mechanisms evaluated in the setting of whole body and global brain ischemia associated with CA. Due to the limited numbers of publications in this area, we included pre-clinical studies using animal models of global brain injury involving bilateral carotid occlusion along with systemic hypotension.

Adenosine was proposed in the late 1980’s to be an endogenous neuroprotective molecule with multiple functions [6]. Effects of adenosine and its receptor agonists include reducing the release of excitotoxic neurotransmitters, vasorelaxation, anti-inflammatory effects, reduction of metabolism, scavenging free radicals and moderate reduction of brain temperature [6, 7]. Only one pre-clinical study evaluated its treatment effects in the setting of global brain injury associated with CA [8]. In a rat model of CA and CPR, post-ischemic administration of adenosine (7.2 mg/kg) was associated with a transient increase in brain total adenylates, hypothermia, increased post-ischemia brain blood flow and reduced brain edema and delayed CA1 neuronal loss in the hippocampus [8].

In the brain, adenosine acts as a neurotransmitter through activation of four specific G-protein-coupled adenosine (A) receptors: A1, A2A, A2B and A3 receptors [9]. The A1 receptor mediates neuroprotection, mostly by blockade of Ca²⁺ influx, which results in inhibition of glutamate release and reduction of glutamate excitatory effects at a postsynaptic level [10]. Unfortunately, selective activation of A1 receptors as a therapeutic approach for brain injury was hampered by major cardiovascular side effects such as bradycardia and hypotension [6, 9]. Given that selective activation of A2A and A3 receptors mediated glutamate release and/or microglia activation and cell death, respectively, development of selective antagonists targeting A2A and A3 may offer a possible treatment for brain disorders via anti-excitoxicity, anti inflammation and anti-apoptosis mechanisms [10, 11]. Therapeutic applications of discrete adenosine ligands need to be further explored in the setting of global brain ischemia following CA.

Under focal ischemic conditions in the brain, administration of several types of exogenous growth factor/hormones has been shown to be neuroprotective. Limited animal studies evaluated their effects in global brain ischemia secondary to CA. However, the neuroprotective effects were not as robust as that found in cerebral ischemia that does not concurrently involve complete systemic circulatory stasis. Furthermore, a short half-life and a low rate of transport through the blood brain barrier (BBB) could challenge clinical applications of growth factor/hormones for brain protection.

Hydrogen sulfide (H₂S) was originally known as a toxic gas and environmental pollutant. However, there has been growing interest in the potential role of H₂S as an endogenous signaling molecule: a third gasotransmitter [35, 36]. By opening ATP-sensitive potassium (K_ATP) channels, endogenous H₂S acts as a vasorelaxant [35‐37]. A number of preclinical studies suggested exogenous low concentration H₂S protects cells from oxidative stress by increasing levels of glutathione (a cellular major and potent antioxidant) and/or directly scavenging oxygen species [38]. It also exerts anti-apoptotic and anti-inflammation effects [37, 39]. In addition, H₂S facilitates the induction of hippocampal long-term potentiation by enhancing the activity of NMDA receptors [40]. In animal studies of global ischemia brain injury, H₂S or H₂S donors provided mixed results. Although H₂S can protect against inflammation, it may also be involved in intracellular calcium overload and worsen secondary neuronal injury [35, 36, 39]. These data suggested that H₂S may have beneficial and/or detrimental effects depending on what pathological state and/or what phase in a certain pathological state it is administered.

In mouse model of CA and CPR, administration (IV) of sodium sulfide (Na₂S, a hydrogen sulfide donor) 0.55 mg/kg at CPR initiation significantly improved survival and neurological deficits at 24 hours [41]. It also significantly decreased apoptosis in the hippocampus, and prevented myocardial dysfunction [41]. Delaying Na₂S administration to 10 minutes after CPR was also investigated but did not confer the same advantages as when administered at CPR initiation. No formal long-term analysis was performed in this study, but pilot studies show a significant long-term survival advantage in treatment group mice. A similar study investigated the mechanism by which by sodium hydrogen sulfide (NaHS; another hydrogen sulfide donor) conferred benefit [42]. Study methods were similar to Na₂S studies, but data was collected over 10 days and abnormal water diffusion in the brain was investigated using hyperintense diffusion-weighted imaging. Mice treated with NaHS had significantly improved 10-day survival, neurological outcome, and reduced brain water diffusion. Results differed in a mouse study utilizing a very similar protocol [43]. At CPR initiation, Na₂S was administered (0.5 mg/kg IV bolus) and continued as an infusion (1 mg/kg/hr) for 6 hours. The only significant difference was better neurological function scoring at 3 days in mice treated with Na₂S. No other differences were seen between groups with respect to neurological function or neurohistopathology.

In contrast, researchers using a pig model of prolonged CA and CPR reported detrimental effects. Na₂S (1 mg/kg or 0.3 mg/kg) injection and infusion given at 1 min after starting CPR did not improve initial resuscitation success. It further compromised early postresustication hemodynamics by reducing cardiac output, heart rate and pulmonary arterial pressure [44]. Such deleterious outcome was more significant in the high dose Na₂S group. H₂S mediated cerebral ischemia damage has been attributed to increased excitotoxicity through activating NMDA receptors in experimental focal ischemia [45] or due to its vasodilatation effect [44]. These conflicting results may be due to differences in drug administration time between the rodent model (at CPR initiation) and porcine model (1 minute after starting CPR). It suggests that the temporal window of opportunity for improving the outcome of CA/CPR is relatively narrow and may require delivering Na2S immediately after reperfusion. This feature echoes a previous study of ischemic post-conditioning that found the delay of onset of post-conditioning by only 1 minute aborts protection in rabbit heart [46].

HBO at low pressure (within 3 ATA) has been shown to ameliorate brain injury in a variety of animal models including focal cerebral ischemia, neonatal hypoxia–ischemia and subarachnoid hemorrhage [47]. Possible neurotoxicity was only found when HBO was applied at high pressure (>3ATA) and long duration [47]. Adverse effects include increased spontaneous respiratory rate that can lead to arterial hypocapnia with reduced cerebral blood flow, increased lipid peroxidation and seizures [47]. The multiple mechanisms of HBO include enhancing tissue oxygenation/metabolism, reduction of oxidative stress, inflammation modulation, and inhibition of apoptosis as well as enhancing neurogenesis [47, 48]. It suggests low pressure of HBO may provide a promising neuroprotective strategy for CA patients favoring neurological function recovery. Neuroprotective effects of HBO therapy have been reported in animal models of global brain ischemia induced by clamping of the ascending aorta and cavae [49], infusing Elliott’s B solution into the subarachnoid space [50] and 4-vessel occlusion [51].

Positive results were consistently observed in two studies using a global brain ischemia model involving whole body ischemia. In a dog model of CA, animals were randomized to control or HBO after CA and resuscitation [52]. HBO was administered at 2.7 ATA for 60 minutes starting 1 hour after ROSC. The functional status of the dogs was significantly better and neuronal damage was significantly less in the HBO group at 1 day after ROSC. Cerebral oxygen extraction ratio was also significantly less in the HBO group. In a rat model of global brain ischemia with hypotension, HBO (3 ATA) applied at 1 h after ischemia for total of 2 h significantly reduced neuronal death in the CA1 region of the hippocampus and the cortex [53].

Additional preclinical studies using clinically relevant CA models are needed to establish the optimal dose and time for treatment as well as the safety of applying HBO treatment following ROSC, particularly when intensive care is required.

Molecular hydrogen has previously been proposed as a novel selective hydroxyl radical and peroxynitrite scavenger in the treatment of focal brain ischemia [54]. Emerging evidence has demonstrated its protective a role in a variety of diseases through anti-oxidant, anti-inflammation and anti-apoptotic effects [55, 56]. There is no risk of combustion at concentrations less than 4% H₂[56]. In two animal models of global brain ischemia, hydrogen gas or hydrogen rich saline benefitted overall outcome after CA, including neurological function [57, 58]. The protective effects have been shown to be comparable to therapeutic hypothermia [57].

Rats underwent 5 minutes of CA followed by CPR and were randomized to sham, control (ventilation with 98% O₂ + 2% N₂), therapeutic hypothermia (TH), hydrogen gas (H₂, ventilation with 98% O₂ + 2% H₂), or TH + H₂ groups [57]. Twenty-four and seventy-two hour survival was significantly improved in rats treated with H₂, H₂ + TH, or TH compared to controls. At 24 and 48 hours, the H₂, H₂ + TH, and TH groups were significantly better than the control group, and the H₂ + TH group was significantly better than the TH and H₂ groups alone. Systemic inflammatory response measured at 2 hours post ROSC showed significantly reduced levels of IL-6 in H₂ and H₂ + TH groups compared to control or TH groups. Other significant effects seen with the H₂ gas included maintaining low left ventricular end diastolic pressure during the first 2 hours after ROSC when compared to controls, decreased lung edema at 24 hours post-CA (levels same as sham rats), and attenuation of cardiomyocyte degeneration.

In rabbits subjected to CA followed by CPR, intraperitoneal injection of H₂ rich saline or placebo was done at CPR initiation [58]. Animals were assigned to sham (no CA); CA; CA + low dose H₂ (10 ml/kg; CA + H₂ group 1); and CA + high dose H₂ rich saline (20 ml/kg; CA + H₂ group 2). Intraperitoneal H₂ rich saline (low and high dose) significantly improved 72-hour survival, reduced neuronal injury, and reduced plasma oxidative markers. Neurologic functional recovery was significantly improved only with 20 ml/kg H₂ compared to controls at 72 hours.

H₂ is cost-effective and has low toxicity with little drug-drug interaction. These features could make H₂ administration an ideal neuroprotection approach that could be implemented during CPR and/or after ROSC.