What is the approach to neuroprognostication in adults post cardiac arrest?

Neuroprognostication Post Cardiac Arrest

Neuroprognostication in comatose adults after cardiac arrest must employ a multimodal approach with assessment delayed until at least 72 hours after return of spontaneous circulation (ROSC) or return to normothermia, as no single test has sufficient specificity to eliminate false positives. 1

Critical Timing Principles

• Wait a minimum of 72 hours after ROSC (or 72 hours after rewarming if targeted temperature management [TTM] was used) before making definitive prognostic assessments 1, 2
• No clinical signs, electrophysiologic studies, biomarkers, or imaging can reliably predict death or poor neurologic outcome within the first 24 hours after cardiac arrest 1, 2
• Ensure complete clearance of sedation effects and absence of confounders (hypotension, hypothermia, neuromuscular blockade, seizures, hypoxemia) before prognostication 2, 3

Multimodal Assessment Components

Clinical Examination (≥72 hours post-ROSC)

Pupillary Light Reflex:

• Bilateral absence of pupillary light reflex at ≥72 hours is the most reliable clinical predictor 1, 3
• In patients treated with TTM: false-positive rate (FPR) 0% (95% CI 0-3%) - Class I recommendation 1
• In patients not treated with TTM: FPR 0% (95% CI 0-8%) - Class IIa recommendation 1

Corneal Reflex:

• Bilateral absence of corneal reflex at ≥72 hours predicts poor outcome with FPR 0% (95% CI 0-9%) 1, 2

Motor Response:

• Absent motor movements or extensor posturing alone should NOT be used for predicting poor outcome due to high FPR (10-15%) 1
• Glasgow Coma Scale motor score ≤2 at 72 hours has FPR of 5% (95% CI 2-9%) 2

Myoclonus:

• Presence of myoclonus or status myoclonus within 7 days after ROSC should be used in combination with other tests 1
• Always record EEG when myoclonic jerks are present to detect associated epileptiform activity 1

Electrophysiological Testing

Somatosensory Evoked Potentials (SSEP):

• Bilateral absence of N20 cortical response at 24-72 hours after cardiac arrest is a reliable predictor with FPR 1% (95% CI 0-3%) 1, 2
• SSEPs are less affected by sedation or temperature manipulation compared to EEG and clinical examination 1
• Must be performed with appropriate technical expertise to avoid electrical interference 1

Electroencephalography (EEG):

• Absence of EEG background reactivity alone should NOT be used to predict poor outcome 1
• Presence of seizure activity on EEG should be used in combination with other indices 1
• Burst suppression on EEG (after sedation effects have cleared) should be used in combination with other indices 1
• In normothermic patients without confounders, highly malignant patterns (generalized suppression <20 μV, burst-suppression with epileptic activity, diffuse periodic complexes on flat background) have FPR 3% (95% CI 0.9-11%) 2
• EEG findings within 24 hours after ROSC are NOT reliable predictors 2

Serum Biomarkers

Neuron-Specific Enolase (NSE):

• Use NSE within 72 hours after ROSC in combination with other tests 1
• Thresholds ranging from 33-120 μg/L predict poor outcome with specificity 75-100% and sensitivity 7.8-83.6% 1
• No consensus exists on a specific threshold value 1

Other Biomarkers:

• S-100B protein is NOT recommended for prognostication due to wide variability in thresholds 1
• Glial fibrillary acidic protein, serum tau protein, and neurofilament light chain are NOT recommended due to insufficient evidence 1

Neuroimaging

Brain CT:

• Marked reduction of gray-white ratio (GWR) on CT obtained within 2 hours may be reasonable to predict poor outcome in patients not treated with TTM 1
• Use GWR on brain CT for predicting neurological outcome, though no specific threshold for 100% specificity can be recommended 1

Brain MRI:

• Use diffusion-weighted brain MRI for predicting neurological outcome 1
• Extensive cortical and subcortical lesions on MRI are associated with poor outcome 2
• Optimal timing is 2-6 days post-ROSC 2

Algorithmic Approach to Prognostication

Step 1: Timing Verification

• Confirm ≥72 hours have elapsed since ROSC (or return to normothermia if TTM used) 1, 2, 3

Step 2: Exclude Confounders

• Verify absence of sedation effects, neuromuscular blockade, hypotension, hypothermia, seizures, hypoxemia 2, 3

Step 3: Clinical Examination

• Assess pupillary light reflex bilaterally 1, 3
• Assess corneal reflex bilaterally 1
• Document motor response and presence of myoclonus 1

Step 4: Electrophysiological Studies

• Obtain bilateral SSEP testing for N20 response 1
• Perform EEG to assess background reactivity, seizure activity, and burst suppression patterns 1

Step 5: Biomarker Assessment

• Measure NSE within 72 hours (interpret in context of other findings) 1

Step 6: Neuroimaging

• Obtain brain CT or MRI (preferably MRI at 2-6 days) 1, 2

Step 7: Integration

• Never base decisions on a single finding 1, 2
• Poor prognosis requires convergence of multiple poor prognostic indicators 1, 3
• If prognosis remains indeterminate, extend observation period for awakening if consistent with goals of care 3

Critical Pitfalls to Avoid

• Making prognostic decisions before 72 hours is the most common error, as premature assessment leads to inaccurate predictions 1, 2, 3
• Relying on a single test violates the fundamental principle that no single test has sufficient specificity 1, 2
• Ignoring confounding effects of sedation can falsely suggest poor prognosis when recovery is still possible 2, 3
• Using motor examination alone has unacceptably high false-positive rates (10-15%) and should never guide withdrawal decisions 1
• Self-fulfilling prophecy occurs when prognostic testing influences withdrawal of life-sustaining therapy, which then causes the predicted poor outcome 4, 5
• Premature withdrawal of life-sustaining therapy based on perceived poor prognosis is the most common mechanism of death after cardiac arrest 6, 5

Special Considerations

Patients on ECMO:

• Classical prognostic tools maintain acceptable performance in patients undergoing ECMO after cardiac arrest 7
• The combination of at least two poor outcome criteria has 100% specificity in ECMO patients 7
• NSE levels are higher in ECMO patients but other prognostic items remain similar 7

Impact of TTM:

• The potential impact of TTM on prognostication remains to be fully determined 1
• Timing should be adjusted to 72 hours after return to normothermia in TTM-treated patients 1, 2