Our analysis reveals significant discrepancies between preclinical study designs and clinical practice, underscoring the urgent need for more clinically relevant and multimodal approaches to enhance the translational potential of experimental findings. Specifically, our study shows that most of preclinical studies on CA were not designed based on a clinically derived approach, thus limiting the translatability of results.

The need for neuroprotective treatments

There is an urgent need to identify novel neuroprotective treatments after CA. Over the past few decades, many treatments have been tested in the preclinical setting. Remarkably, the majority of them showed promising results in animal studies but failed to replicate the same success in large clinical trials. One reason for this translational failure is the discrepancy between the methods used in preclinical research and those applied in clinical practice.

In clinical neuroprognostication—the prediction of neurological recovery—it is recommended to use a multimodal approach as no single test has sufficient specificity to avoid false positives[2]. Embracing a multimodal methodology in animal research, could increase the translatability of findings, mirroring clinical practice, where combining multiple assessment tools improves the prediction of neurological outcomes.

Post-resuscitation care and neuroprognostication

The post-resuscitation care in years has gained increasing importance. In 2010 guidelines, the post-resuscitation care was a paragraph incorporated in Advance Life Support section [15]. In 2015, the ERC and ESICM published guidelines specifically focused on the post-resuscitation care section, emphasizing the importance of high-quality post-resuscitation care and highlighting how this is a crucial factor in the chain of survival. Moreover, in 2015, it has been introduced for the first time the concept of the multimodal approach for neuroprognostication [16].

Two-thirds of deaths occurring in OHCA patients are due to PCABI, with the majority resulting from the active withdrawal of life-sustaining treatment (WLST) based on neuroprognostication. Indeed, it was important to minimize the risk of falsely poor prediction [16]. ERC guidelines 2015, suggested for the first time a multimodal neuroprognostication strategy, so a combination of distinct parameters that can increase the sensitivity to predict a poor outcome in patients. Specifically, this prognostication strategy algorithm was composed by one or both of no pupillary and corneal reflexes and bilaterally absent N20 SSEP wave [16]. In addition, they recommended that in none of those signs were present, it can be evaluated also a group of less accurate predictors: the presence of early status myoclonus, high values of NSE at 48–72 h after Return of Spontaneous Circulation (ROSC), an unreactive malignant EEG pattern and the presence of anoxic injury evaluated by brain CT or brain MRI scans [16]. ERC and ESICM proposed this algorithm based on evidence that none of these predictors singularly predicted poor outcome with 100% of accuracy, but combined together, then, they can increase safety and sensitivity to avoid falsely pessimistic prediction.

In the latest 2021 guidelines the concept of neuroprognostication has been revisited and improved: since 2015 there has been many studies regarding prognostication, which validated and confirmed the reliability of the algorithm presented in the last guidelines of 2015. Therefore, based on these data, they simplified the two-stage strategy algorithm so that poor outcome is considered likely when two or more listed predictors are present: no pupillary or corneal reflex at ≥ 72 h, bilaterally absent N20 SSEP wave at ≥ 24 h, NSE > 60 μg/L at 48 h and/or 72 h, presence of status myoclonus ≤ 72 h and a diffuse and extensive anoxic brain injury on CT or MRI scans [2].

The role of histological analysis

Histological analysis is an essential component of preclinical research, providing critical insights into the biological mechanisms underlying injury evolution following CA. Although the findings from histological studies are not always translatable to clinical settings, they are vital for understanding the pathophysiological changes that occur after brain injury. Histological and immunohistochemical methods remain essential tools for evaluating the severity of PCABI. Techniques such as hematoxylin and eosin (H&E), Nissl and Fluoro-Jade staining allow researchers to examine neuronal death and neurodegeneration. [17]. Immunohistochemistry is particularly valuable for investigating the inflammatory response following CA [17], focusing on key cellular populations like microglia and astrocytes, which mediate the brain's immune response and hold a pivot role in contributing to long term outcome. [1, 5]. These glial cells are highly active in the injured brain and play a pivotal role as immune cells that mediate inflammation. Their behavior and activation states, as revealed through immunohistochemistry, provide important clues about the extent and nature of the brain's response to injury after CA [1, 5]. Susceptibility to ischemia reperfusion injury due to CA, varies significantly depending on the neuronal subtype and region, with area more susceptible than others: the neocortex, the hippocampus, basal ganglia, cerebellum and thalamus [5].

However, it is essential to recognize that histological analysis cannot be obtained at the bedside and not always directly correlate with functional outcomes, which can limit its applicability in clinical contexts. Therefore, integrating histological analysis with other assessments, such as neuroimaging, biomarkers, and functional outcome measures, is crucial for a more comprehensive understanding of brain injury and recovery in conjunction with behavioral assessments.

Multimodal approaches in preclinical studies

Inspired by the multimodal approach used in clinical practice for neuroprognostication, we propose applying a similar strategy in preclinical research to improve the assessment of brain injury and functional recovery after CA.

Functional outcome

Given the poor prognosis of cardiopulmonary resuscitation (CPR) with regard to both survival and neurological outcome, functional evaluation constitute one of the primary measures of outcome [18].

In preclinical research on CA, neurological deficit tests play a critical role in assessing the extent of brain injury and functional recovery in animal models. These tests can be broadly categorized into two groups: those evaluating general neurological behavior and those focused on cognitive and behavioral assessments.

General neurological behavior

General neurological behavior tests provide a basic overview of the animal’s neurological status post-CA. These tests are essential for identifying the immediate and overt effects of brain injury. Specifically, distinct aspects of animal’s behavior can be assessed using clinical deficit scores with specific grids for each species.

They are based on the clinical evaluation of consciousness, behavior, breathing, reflexes, coordination, motor and sensory activity and seizure. The consciousness and general behavior, is asses by observing any changes in movement or responsiveness of the animal. The brainstem performance, is the evaluation of reflexes controlled by brainstem, i.e., pupillary, responses and breathing. The coordination, is assessed by testing the motor coordination through balance observations. Furthermore, the corneal reflex is evaluated by blink response to corneal stimulation and it indicates the cranial nerve function. The motor and sensory activity is the assessment of animal’s voluntary movements and response to sensory stimuli. Finally, is often present the study of seizure activity, so monitoring the occurrence of seizure or convulsions post-CA.

These analyses provide insight into the severity of neurological damage and recovery following CA. However, they primarily focus on broad neurological outcomes and may not capture the full spectrum of cognitive or behavioral impairments.

Cognitive and behavioral assessments

The second group offers a more specific evaluation of specific brain functions. These tests explore aspects like memory, anxiety, exploratory behavior, and learning, critical in understanding the deeper impact of CA on brain function. The key tests in this category include:

Open field test: used to assess anxiety levels and exploratory behavior by measuring the animal's movement and interaction with a novel environment [8, 19,20,21,22];

Novel object recognition test: evaluates recognition memory, based on the animal's ability to differentiate between familiar and new objects [20, 23];

Morris water maze: a well-established test for assessing spatial learning and memory, where animals must navigate to a hidden platform in a pool of water [19, 21, 24,25,26,27,28,29];

Tape removal test: evaluating sensorimotor function by placing adhesive tape on the forepaw and recording the time it takes for the animal to detect and remove it, reflecting sensory and motor coordination [8, 30, 31];

Y-maze: Used to assess working memory by analyzing spontaneous alternation behavior as the animal explores different arms of the maze [32, 33];

Motor activity test: assesses general locomotor activity by tracking the animal’s movement, measuring distinct parameters like distance traveled, speed, time spent moving versus resting and spontaneous motor activity during light and dark phase of day [8, 9, 19, 21, 22, 34].

These cognitive and behavioral tests are particularly valuable for examining long-term brain function, providing a more detailed understanding of how CA impacts learning, memory, and emotional regulation.

Biomarkers

Multi-organ failure and neuronal damage can be measured in the serum or plasma after CA as biomarkers. Major advantages of blood biomarkers are that they are easy to obtain and offer a quantitative and easily interpreted measure of the extent of brain injury [1].

Neuron-specific enolase

NSE is a neuronal glycolytic enzyme that is abundant in the neurons of brain gray matter and involved in axonal transport [35]. In healthy individuals, serum levels of NSE remain low. In contrast, upon damage to neuronal tissue, such as anoxic brain injury, NSE serum concentration increases and consequently acts as a biomarker for brain damage. European Resuscitation Council (ERC) guidelines 2021 indicates that increasing values between 24 and 48 h or 72 h in combination with high values at 48 and 72 h indicate a poor prognosis [2].

S100 calcium‐binding protein B

S100B is abundant in glial cells expressed in astrocytes surrounding the blood vessels in the brain and in the Schwann cells of the peripheral nervous system, where they stimulate cellular processes such as proliferation, differentiation, and regulation of intracellular Ca2+ homeostasis. At least 80–90% of the total S100B pool is found within the brain, the remainder being located in other non-neuronal tissues. S100B is released from damaged astrocytes into the bloodstream after ischemic–reperfusion injury that occur after CA. Is considered an early biomarker, as the level usually peaks at 24 h and elevated levels are associated with poor outcome [35].

Neurofilament light

Neurofilament light chain (NfL) is a subunit of neurofilaments, which are cylindrical proteins exclusively located in the neuronal cytoplasm, predominantly within large, myelinated axons within the cerebral white matter. Their function is largely unknown but hypothesized to be essential for radial growth and enabling rapid nerve conduction [35]. Under physiological conditions, low levels of NfL are constantly released from axons, probably in an age-dependent manner, however, in response to CNS axonal damage because of pathological problem, such as an ischemic insult as occurs during CA, the release of NfL sharply increases. High levels in CA patients are index of poor outcomes [35].

Lactate

Lactate, a product of pyruvate reduction during glycolysis, has been suggested to be an indicator of multi-organ failure hypoxia resulting from reduced cardiac output and in CA patients elevated arterial blood lactate levels are associated with poor neurologic outcome [36]. Lactatemia during the first hours after CA is associated with hypoperfusion after the cessation of blood flow and the inflammatory reaction due to ischemia–reperfusion injury [36]. Hyperlactatemia few h after ROSC may indicate complication in patients, such as poor neurological function. Hence, lactate levels in CA are a critical marker for assessing the severity of ischemic–reperfusion injury, as well as predicting neurological outcomes.

Electroencephalography

Electroencephalography (EEG) is a highly sensitive tool for detecting the severity of PCABI since assesses cortical synaptic activity [5]. Moreover, EEG is able to evaluate the occurrence of seizures as well as the appearance of spikes/sharp waves and epileptiform discharges. These, together with malignant EEG patterns (persistent iso-electricity, low voltage activity, or low burst-suppression patterns), are used to prognosticate outcome after CA. ERC guidelines 2021 suggest that highly malignant EEG after 24 h is an indicator of poor neurological outcome [2].

Imaging

Brain CT is extensively used shortly after CA to rule out neurological causes of arrest, especially an intracranial hemorrhage that would contraindicate percutaneous coronary interventions. However, CT also allows assessing the severity of PCABI by detecting brain oedema [1].

The use of MRI for prognostication is more recent, but has rapidly gained interest during the last decade. One of the main advantages of MRI is the ability to assess the anatomical distribution of diffusion restrictions [37]. MRI allows the detection of cytotoxic edema, which occurs within hours after CA. Restricted diffusion by cytotoxic edema can be quantified by the Apparent Diffusion Coefficient (ADC) value of each voxel. Low ADC values, thus restricted water diffusion (DWI lesions) are associated with poor outcome after CA [2]. Moreover, an ADC values < 650X10−6 mm2/s in > 10% of the brain at 7 days after CA is highly specific for poor outcome.

DTI is an extension of DWI that allows the evaluation of microstructural integrity of brain white matter by directional assessment of water diffusion. Although DTI is not a criterion in the strategy algorithm for neuroprognostication, it was found that changes in DTI parameters can predict poor outcome in CA patients with 85% sensitivity [37].

Bridging the gap between preclinical and clinical studies

Our review highlights several gaps between preclinical models and clinical practices. Preclinical studies often rely heavily on brain histopathology as a primary endpoint, neglecting correlations with functional outcomes or biomarkers. Furthermore, animal models typically involve young, healthy subjects, which do not adequately reflect the comorbidities present in clinical patients.

Significant correlations are seen between imaging, neurological deficits, and biomarkers, reflecting the severity of brain injury in CA patients, suggesting similar correlations should be explored in preclinical models. Indeed, in our analysis we did not find studies that explored possible correlations between outcomes. Interestingly, our review highlights a significant gap in the literature, as very few studies have explored potential correlations between histology, neuroimaging, electrophysiology, blood biomarkers, and neurobehavioral outcomes. Notably, there are only two studies addressing these correlations that demonstrated strong, positive correlations between apparent diffusion coefficient (ADC) values and the neurological deficit score, showing that the severity of brain cytotoxic edema is closely associated with worsened neurological function in the early phase after CA [8, 9]. In addition, one of these studies found a strong correlation between diffusion tensor imaging (DTI) metrics and the brain injury biomarker NfL [8]. These findings provide compelling evidence that such correlation can be identified underscoring a critical parallel between preclinical models and the clinical scenarios, emphasizing the translational relevance of incorporating multimodal assessments in preclinical research.

To improve translational value, preclinical models should integrate multimodal assessments and better mimic clinical conditions providing a more comprehensive assessment of the brain during CA.