A small numbers of patients (10-15%) treated with cardiopulmonary resuscitation (CPR) are discharged from hospitals with a favorable neurologic outcome. However, a higher incidence of chest injuries (30-70%), mainly rib and sternum fractures, are observed among the survivors. It's no surprise that 6 months after cardiac arrest (CA) 50-70% of the patients who have survived continue to have pain and stress-related problems. Based on the need for the pain/stress treatment in these patients and several experimental evidences demonstrating neuroprotective features of anesthetics it is logical to presume that application of anesthesia during CPR may be indicated. In rodents exposed to hypoxic gas (5% 02, 95% N2) for 70 min, all seven animals died at the end of the experiments in the naloxone pre-treated group while only one out of seven rats died in the morphine pre-treated group, and five of seven rats died in the control group. In human volunteers, intravenously administered 60 mg of morphine did not alter cerebral blood flow and cerebral vascular resistance but markedly depressed cerebral oxygen uptake. Interestingly, in critical patients, morphine even in low doses is effective in relieving dyspnea by altering central perception and decreasing anxiety. In rats, morphine demonstrates dose- depending reduction of cerebral glucose utilization in limbic and forebrain regions. Thus, the main points of application for morphine in treatment of CA may be a reduction of oxygen/glucose consumption. Pre-treatment of zebrafish with ketamine protects against cardiac arrest-induced brain injury by inhibiting Ca2+ wave propagation and consequently it improves survival rate. Inhibition of NMDA receptors by ketamine reduces neuronal apoptosis and attenuates the systemic inflammatory response to tissue injury. The sympathomimetic effects of ketamine may facilitates recovery of systemic blood pressure during CPR. Retrospective investigations demonstrate that patients who are treated with opioids before or during CA have a statistically significantly higher survival rate and much better neurological outcome compared to untreated patients. Experimental studies have a limitation as all animals are treated with anesthesia and therefore survival rate varies between 50-90%. Thus, prospective research is urgently needed to investigate the influence of morphine or/and ketamine on survival and neurological outcome in patients with CA.

Almost 35 years ago, Dr Peter Safar wrote that "cerebral recovery from more than 5 min of cardiac arrest is hampered by complex secondary derangements of multiple organ systems after reperfusion". Actually, these 5 "golden" minutes determine the ability of cerebral neurones to regain ordinary function after anoxia. The ordinary function of cerebral neurones is conduction of electrical impulses across their length from the post-synaptic membrane of dendrites to the presynaptic membrane of an axon. The process is based on exchange of Ca2+, Na+ and K+ between the extra- and intracellular space of cerebral neurones, and therefore a lot of energy in the form of adenosine triphosphate (ATP) is needed to remove Ca2+ and Na+ from the intracellular space of these cells. Cardiac arrest (CA) initiates a switch to glycolytic metabolism with very low production of ATP and the increased levels of lactate and H+. Both acidosis and the lack of ATP inhibit the ions pumps, which are responsible for handling excessive intracellular accumulation of Ca2+ and Na+. Preclinical studies demonstrate that acute hypoxia results in an uncontrolled release of glutamate with consequent stimulation of the N-methyl-D-aspartate (NMDA) receptors causing also an excessive Ca2+ influx. Meanwhile, the ATP reservoir in neurones can be completely depleted after 5 minutes of no-flow state. In case of oxygen supplying restoration, two molecules of ATP are initially required to split glucose and restart the cellular respiration. Thus, the presence or absence of these two molecules of ATP in neurones actually determine restoring of both oxidative phosphorylation and the ordinary function of the neurones. Finally, prolonged intracellular Ca2+ overload results in increased mitochondrial permeability causing following release of cytochrome C from mitochondria, and consequent cleavage and activation of caspase-3. Caspase-3 is an essential protease, which is involved in the early stage of apoptosis and it is generally accepted as a hallmark of irreversible cell death. Interestingly, in an experimental study with rats, the activation of caspase-3 was observed in a significant number of neurones of the cerebellum and neocortex only after 9 hours following asphyxial cardiac arrest.

Today, only therapeutic hypothermia has been shown to have a beneficial impact on the ion pump dysfunction, and thereby reduce neurotoxicity. Interestingly, in hibernators, hypothermia is also believed to protect against hypoxic brain damage. Meanwhile, if naloxone, a non selective opioid receptor antagonist, is injected during the maintenance phase of hibernation, arousal is quickly achieved and the protective effects are vanished. Delta opioid peptides, previously discovered to induce hibernation have also been shown to protect rats from hypoxic brain damage. Based on the ability of opioids to reduce the level of cyclic adenosine monophosphate (cAMP), and consequently to block Na+ channels, it would be logical to propose that opioids might prevent the disturbance of ionic homeostasis during acute hypoxia. Indeed, preclinical studies demonstrate that opioids can preserve cellular integrity status during acute hypoxia in many organs and tissues including: intestine, skeletal muscle, myocardium and brain. Moreover, Morphine has been shown to significantly increase the survival of mice and rats in acute hypoxia conditions. In the experimental model with rats exposed to hypoxic gas (5% oxygen, 95% N2) for 70 min, all seven rats in the naloxone pre-treated group died at the end of the experiments while only one out of seven rats died in the Morphine (5 mg/kg) pretreated group, and five from the seven rats died in the control group. In the experiments where the rats were exposed to 8 min anoxia, pre-treatment with Morphine (5mg/kg), or Ketamine (40 mg/kg), resulted in higher survival in both groups as compared to the control group (data not yet published). No publications looking at the survival rate in animals with treatment by Morphine before cardiac arrest have been published yet. Meanwhile, two recent retrospective studies demonstrated that patients who were treated with opioids before or during cardiac arrest had a statistically significantly higher survival rate and much better neurological outcome compared to untreated patients. Recently published data shows that the Na+ influx initiating the action potential in neurones consumes one third the ATP of synaptic potentials associated with Ca2+ influx. In theory, Ketamine that inhibits the synaptic potentials by NMDA receptor blockade, might save much more ATP in the neurones compared to Morphine, which inhibits only Na+ influx and accordingly the action potentials. Certainly, after restoration of blood flow, the residual saved ATP may contribute to restoration of both neuronal oxidative phosphorylation and ionic exchange. Newly published experimental data demonstrates that pre-treatment of zebrafish with Ketamine protects against cardiac arrest induced brain injury by inhibiting Ca2+ wave propagation, which consequently improves survival rate. Contrary to the results seen in these studies, two NMDA antagonist, MK-801 and GPI-3000 in high doses did not improve survival rate and brain outcome after cardiac arrest and resuscitation in a dog model. These studies did not suggest any mechanisms of the negative results, but they contributed to a lack of interest for testing NMDA blockade in CA for years. More recently, a study of the effects of using the noncompetitive NMDA antagonist Ifenprodil demonstrated a significant reduction of brain oedema following asphyxial cardiac arrest in rats. In this study, i.v. injection of Ifenprodil also resulted in much more stable hemodynamic status after CA as compared with salt treated animals. Another experimental study of different anaesthesia regimes in a rodent cardiac arrest model also demonstrated much better hemodynamic status in the early post resuscitation period in the rats treated with Ketamine and Medetomidine as compared to anaesthesia with Sevoflurane and Fentanyl. All anaesthetics, with their ability to antagonise glutamate mediated excitotoxicity and inflammation might be logical candidates for neuroprotective treatment during cardiac arrest. However, the ability of anaesthetics to produce vasodilatation with a significant reduction of blood perfusing pressure can be the main argument against the idea to test their effects during cardio pulmonary resuscitation (CPR) in human. However, due to their minimal influences on hemodynamic status in therapeutic doses, Ketamine as well as Morphine can be considered as the safe candidates during neuroprotective treatment trials in CPR patients. Another argument for possible application of Morphine or Ketamine during CPR could be as an analgesic. Vigorous thoracic compression with possible trauma of the ribs may lead to severe pain and stress reactions in patients surviving CPR.

The rationale for analysing plasma levels of S-100B protein and NSE in this trial will be their different distribution within the white (S100B protein) and grey (NSE) matter of the brain, and the fact that both of them are extensively involved in the pathogenesis of anoxial brain damage. S100 B protein is an intracellular calcium-binding dimer that has a molecular weight of 21 kDa and two hours of half life. Thanks to the low molecular weight, S100 B protein easily cross the blood-brain barrier and rapidly end up in the systemic circulation. NSE is a neuronal isoform of the glycolytic enzyme enolase that has a molecular weight of 78 kDa and a twenty four hours half life. Further, NSE is extensively involved in glucose metabolism in the neurones and can be detected only in neuronal and neuroendocrine tissues. Due to this organ specificity, concentration of NSE in blood is often elevated as a result of relative rapid and massive neuronal destruction. In clinical practice, elevated serum NSE levels, above 30 ng/ml, correlate well with a poor outcome in coma, particularly when caused by an hypoxic insult. Thus, these two markers of early neuronal damage is a good fit for testing the neuroprotective features of Morphine or Ketamine application during CPR. A retrospective evaluation of patients after cardiac arrest in the University hospital of Northern Norway demonstrated significantly higher 1, 2, 3 and 28 days survival rate and reduced duration of CPR in the patients additionally treated with opioids compared to ordinary resuscitation. Two years later another retrospective, observational cohort study from Pittsburgh, USA reported that despite poor baseline prognostic factors, survival after recreational drug overdose-related cardiac arrests (CA) was no worse than after non overdose-related arrest, and among survivors the majority had a good neurological outcome. Interestingly, the same research group reported in one previous retrospective study a higher rate of survival to hospital discharge (19% vs. 12%, p = 0.014) in the overdose CA group compared to the non-overdoses one. However, patients in these overdose cases were significantly younger (45 vs. 65, p < 0.001), but less likely to be witnessed by a bystander (29% vs. 41%, p < 0.005). Suspected overdose cases had a higher overall chest compression fraction (0.69 vs. 0.67, p = 0.018) and higher probability of adrenaline, sodium bicarbonate, and atropine administration (p < 0.001). Application of Naloxone in these overdose cases might have had an influence on survival as well. One previously published clinical case described full neurological recovery in a young man who overdosed on opioids and who regained sinus rhythm many minutes after resuscitation had been abandoned. Based on all of the above, it can be hypothesised that treatment with Morphine or Ketamine might have a beneficial impact on the conservation of ATP in the brain, and thereby the treatment might increase the ability of cerebral neurones to survive and regain ordinary function after CPR.

Inclusion Criteria:

- All adult patients with in-hospital cardiac arrest

Exclusion Criteria:

• an age of less than 18 years
• drugs poisoning or the administration of opioids or Ketamine 24 hours before the cardiac arrest
• terminal phase of oncological or other chronic diseases
• poor communication and physical capabilities due to psychiatric or neurological diseases
• dementia or Alzheimers
• extremely reduced weight or physical ability and activity
• known history of chronic use of opioids/Ketamine

Acute bag for cardiopulmonary resuscitation (CPR) will have a trial pack containing 3 blinded ampules that will be labelled with only randomisation number. Anaesthesiologists and nurses performing CPR and research treatment will not know the content of the ampules as well as a meaning of the randomisation number. Neurologist, who will not be aware of treatment assignments, will assess neurological outcomes in all survived patients. Statistician, who will not be aware of treatment assignments, will assess statistical data in all included patients. An independent Data Monitoring Committee will assess the progress, safety data (death, long-term hospitalization, and disability) and, critical efficacy endpoints of the clinical study (survival at 1, 2, 3 and 28 days in patients after in-hospital CA) in all included patients.