Post-Cardiac-Arrest Targeted Temperature Management and Brain Death Prognostication
A 62-year-old woman with hypertension and type 2 diabetes has a witnessed out-of-hospital cardiac arrest (ventricular fibrillation), bystander CPR, ROSC after 22 minutes. She is comatose, intubated, on 0.1 mcg/kg/min norepinephrine, target temperature management at 33 C initiated. Day 3 after rewarming to 36.5 C: still unresponsive, GCS 3, pupils 3 mm and reactive, corneals present, no cough with suction, no spontaneous breathing on a brief apnea test attempt (the team aborted because SpO2 fell to 82%). Sedation was stopped 36 h ago; she received fentanyl and propofol during TTM. Labs: lactate 1.8, creatinine 1.9, AST 320, INR 1.3, platelets 88,000. CT head: diffuse loss of gray-white differentiation, no herniation. Family asks "is she brain dead?" and whether to continue.
1. Is this patient brain dead? Walk through the determination and identify what is missing or confounded.
Read the worked answer
No - she does not meet brain death criteria on these data, and the exam wants you to say why.
Brain death (death by neurologic criteria) requires: (1) a known cause sufficient to explain irreversible coma; (2) exclusion of confounders (sedation, neuromuscular blockade, hypothermia, severe metabolic/endocrine derangement, hypotension); (3) coma with no motor response to pain; (4) absent brainstem reflexes (pupillary, corneal, oculocephalic, oculovestibular, gag, cough); and (5) apnea.
Here: pupils are reactive and corneals are present - that alone precludes the diagnosis. The apnea test was aborted, so the apnea criterion is not established (and an aborted test due to desaturation is a failed, not positive, test). Confounders remain: she was on propofol and fentanyl (fentanyl can be prolonged with hepatic/renal dysfunction - AST 320, creatinine 1.9, platelets 88k suggesting possible DIC/ischemic hepatitis), and she is hypotensive on norepinephrine (MAP may be inadequate for a valid exam). Hypothermia has resolved, which is good, but the metabolic picture (renal and hepatic dysfunction) and possible residual sedation must be excluded. Also, INR 1.3 and platelets 88k - coagulopathy matters for any ancillary testing and for interpreting catastrophic injury.
What to do: ensure normothermia, MAP ≥65–70 (wean pressors if possible), confirm no neuromuscular blockade (train-of-four), allow adequate drug washout (consider fentanyl effect - maybe switch to a short-acting agent or hold), correct electrolytes, then repeat a full clinical exam. If the exam remains confounded or apnea cannot be safely tested, use ancillary testing (e.g., CT angiography showing no intracranial filling, or EEG showing electrocerebral silence - noting EEG can be confounded by sedation). A single exam is not enough; a second examiner at a later time is standard.
2. The family asks about prognosis and whether to continue. How do you counsel them, and what multimodal prognostication do you order?
Read the worked answer
Counseling: Be honest that it is too early to declare death by neurologic criteria and that the current exam is confounded, but also that the picture is grave: 22 minutes of arrest, diffuse loss of gray-white differentiation on CT, and persistent coma at 72 h off sedation are poor prognostic features. Avoid definitive statements ("she will never wake up") - prognostication after cardiac arrest is probabilistic; some patients with poor early features have meaningful recovery, though the probability here is low. Emphasize that decisions will be revisited as data accrue, and that the goal is to give the family time and information to make a values-based decision. Ask about her prior wishes/advance directive.
Multimodal prognostication (per ERC/ESICM and AHA guidance):
- Clinical: serial exam off sedation (GCS motor, brainstem reflexes, myoclonus).
- EEG: look for malignant patterns (burst suppression with epileptiform discharges, generalized periodic discharges, low-voltage) - note sedation confound.
- SSEP: bilateral absent N20 responses are a strong predictor of poor outcome (do when off sedation, normothermic).
- Biomarkers: neuron-specific enolase at ≥48–72 h (trend; interpret with hemolysis caveat), GFAP/NfL where available.
- Neuroimaging: CT (already shows diffuse edema - a poor sign); MRI DWI later if she survives.
- Multimodal combination - no single test is sufficient; require concordant poor-prognosis features before recommending withdrawal, and integrate the family's goals. If the exam evolves to brain death criteria, follow the institutional protocol with a second exam and ancillary testing as needed.