Proteomic Profiling to Reveal Novel Prognostic Markers for Neurological Outcome Following Resuscitation

Background: Cardiac arrest is a life-threatening event. Intensivists are challenged with an increasing number of patients with uncertain neurological outcome following cardiopulmonary resuscitation (CPR). The prognostic value of current biomarkers for neurophysiologic long-term outcome is limited. Therefore, identification of novel plasma markers with higher predictive value for neurophysiological recovery is critical for patient management after CPR.

Hypothesis: We hypothesize that specific brain-derived tissue leakage proteins can be identified to reveal novel, more reliable prognostic biomarkers for good neurological outcome.

Methods: This translational study (n=100) is a combination of a prospective basic science study intended to reduce the number of potential plasma biomarker candidates by proteomic shotgun analyses in brain tissue autopsy samples and plasma samples from resuscitated patients (n=10) and a prospective clinical validation study in a large study population (n=90) by high-throughput analyses. Samples will be analyzed by proteomic shotgun analyses using the Q-Exactive quadrupole-orbitrap mass spectrometer (MS). MS/MS data will be interpreted by the MaxQuant and Perseus Software. In order to identify brain-derived proteins within plasma, the plasma proteome of 10 resuscitated patients will be compared to the proteomic profile of brain tissue. This will reduce the number of potential plasma biomarker candidates associated with neurologic outcome. The prospective validation in plasma samples will be performed by a targeted proteomics approach using selected reaction monitoring (SRM) on a triple quadrupole ion MS. Neurological outcome will be assessed by the five-point scale (death, persistent vegetative state, severe disability, moderate disability, and good recovery) according to the cerebral performance categories (CPC). A CPC sore of <3 is considered a good neurological outcome. Selection of proteomic markers and signature estimation will be performed by L1 regularized logistic regression, where the tuning parameter will be optimized by cross-validated model performance. The signature's ability to discriminate patients with good and poor outcome will be described by ROC analysis.

Clinical perspective: An accurate predictor of neurological outcome following CPR is of utmost clinical importance. However, previous studies focused on a very limited array of biomarkers. Therefore, a structured proteomic analysis approach might identify the best marker out of all proteins liberated during cellular damage.

Inclusion criteria

• Nontraumatic, normothermic cardiac arrest due to cardiac disorders, respiratory failures, or hemodynamic or metabolic factors.
• A Glasgow Coma Scale of 3, none of the patients will be conscious at the time of hospital admission.
• No previous cardiac arrest, as well as known or coexisting neurological disorders or neoplasms of the central nervous system.
• No history of psychiatric illness, no alcohol or drug dependency, and no psychotropic medication.
• Initiation of mild therapeutic hypothermia

Exclusion criteria

• hydrocephalus and shunt artifact
• severe movement artifacts
• intracerebral hemorrhage
• old large ischemic lesion