18.1 Coma and Other States of Altered Consciousness (03.A)

Coma and Other States of Altered Consciousness

Quick Answer: Coma means unarousable unresponsiveness with eyes closed and no sleep–wake cycle. Vegetative state / unresponsive wakefulness syndrome (UWS) has eyes open and sleep–wake without awareness. Minimally conscious state (MCS) has inconsistent but reproducible awareness. Locked-in syndrome is aware ventral pontine injury — not coma. Akinetic mutism is a frontal/cingulate paucity of movement with wakefulness. True coma localizes to both hemispheres (or diencephalon) or to the brainstem ARAS. Check glucose with the ABCs.

A silent, motionless patient in the neuro ICU is not automatically “in coma.” On the ABIM Neurocritical Care examination, stems mix a locked-in basilar occlusion, a vegetative survivor after anoxia, an MCS patient who inconsistently follows a command, and a hypoglycemic patient who looks brainstem-dead until the glucometer reads 28 mg/dL. Independent OpenExamPrep teaching here covers coma and other states of altered consciousness as listed among Neurocritical care diseases in the ABPN Content Specifications. This guide is not an ABIM or ABPN product.

Why the label matters

The word coma is a localization claim and a communication claim. Families hear “coma” as unconsciousness. If the patient can hear you, track vertically, and spell with eye blinks, you have just described locked-in syndrome as if the person were gone. Prognostication, rehabilitation, and whether you talk to the patient at the bedside all change with the correct syndrome name. The Glasgow Coma Scale (GCS) and FOUR score grade responsiveness; they do not by themselves name the syndrome. A GCS of 5T can be bihemispheric anoxia, a pontine hemorrhage, or residual neuromuscular blockade. Report subscores, pupils, tracking, and whether sleep–wake is present.

Named states you must not collapse into one word

Coma, in the Plum and Posner sense still used at the bedside, is unarousable unresponsiveness. The patient does not open eyes to voice or pain in a sustained way, does not have a sleep–wake cycle, and shows no purposeful behavior. Duration is hours to (by convention) about 2–4 weeks; after that, if the brainstem supports wakefulness without awareness, the label usually moves to vegetative/UWS or MCS rather than “chronic coma.”

Vegetative state and unresponsive wakefulness syndrome describe the same clinical picture: wakefulness without awareness. Eyes open, sleep–wake cycling returns, breathing may be spontaneous, and there may be startle, grimace, or roving eyes, but there is no reproducible evidence that the person understands or intends. Do not call this coma. Do not call a few days of eyes-closed unresponsiveness after cardiac arrest “vegetative” either — that diagnosis is not a day-1 label.

Minimally conscious state, defined in the Aspen and later Giacino work, requires inconsistent but reproducible signs of awareness. MCS− is often visual pursuit, localization to noxious stimulation, or contingent affective behavior. MCS+ includes command following, intelligible words, or yes/no (even if unreliable). One lucky twitch is not MCS. You need a behavior you can elicit again. Emergence from MCS is scored when the patient has functional interactive communication or functional use of two different objects.

Locked-in syndrome is the highest-stakes trap on this list. The lesion is the ventral pons: corticospinal and corticobulbar fibers are destroyed; the tegmentum, including the ascending reticular activating system (ARAS) and much of the horizontal gaze machinery, may be relatively spared, and vertical gaze and eyelid opening (midbrain) often remain. The patient is awake and aware, cannot move limbs or speak, and communicates with vertical eye movements or blinks. EEG is typically near-normal for the awake state. Do not call locked-in “coma.” Do not give a FOUR eye score of 0 without checking whether the eyes open and track vertically. Cover the “coma” label in the note and you will miss both the basilar occlusion and the person in the bed.

Akinetic mutism looks like a quiet, awake patient who does not speak or move much. It is a disorder of initiation, classically from bilateral anterior cerebral artery infarction, cingulate injury, or paramedian thalamic lesions. Eyes may be open and may track. Tone is not the flaccid silence of medullary failure. It is not locked-in (the face and limbs can often be shown to move if sufficiently stimulated) and it is not coma.

Localization: two hemispheres or the brainstem core

Consciousness requires a functioning ARAS in the paramedian tegmentum of the midbrain and rostral pons, plus at least one working cerebral hemisphere (in practice, bihemispheric or thalamic integrity). That anatomy yields two coma localizations:

1. Bihemispheric (or bilateral diencephalic) coma. Diffuse anoxia, hypoglycemia, hepatic encephalopathy, meningitis/encephalitis, bilateral subdural hematomas, and bi-thalamic infarcts shut down both hemispheres. Brainstem reflexes are often preserved early: pupils may be small and reactive, corneal and oculocephalic reflexes present, and breathing regular or Cheyne–Stokes. Motor findings may be symmetric (withdrawal, or nothing). A unilateral cortical stroke does not cause coma unless there is herniation with brainstem compression, massive edema with diencephalic failure, seizure/post-ictal state, or a second lesion. If the stem gives a dense left hemiparesis, a blown right pupil, and unresponsiveness, you are no longer in “left MCA coma” — you are in uncal herniation.

2. Brainstem coma. A paramedian midbrain or pontine tegmental lesion knocks out ARAS. Look for asymmetric or lost cranial-nerve signs: midposition or pinpoint pupils, lost corneal reflexes, lost oculocephalics, skew, internuclear ophthalmoplegia, ataxic or apneustic breathing, or hyperthermia from pontine injury. Ventral pontine injury without tegmental destruction is locked-in, not brainstem coma — the pupils and vertical gaze save the diagnosis.

Psychogenic unresponsiveness and catatonia can mimic coma. The eyes often close actively, vestibulo-ocular testing may show nystagmus (which requires a working cortex), and laboratory and imaging are unrevealing. Do not lead with that diagnosis in a febrile, acidotic, or post-arrest patient.

Confounders that fake a dead brainstem — residual sedation, hypothermia, neuromuscular blockade, and profound metabolic failure — are taught with the ICU examination. Subtract them before you name a syndrome or discuss death by neurologic criteria (a later chapter).

First-hour workup: treat the reversible while you image

Stabilization is not a slogan. Airway, breathing, circulation come first because hypoxia and hypotension make every later scan worse. In parallel, check a point-of-care glucose. Hypoglycemia is common, immediately reversible, and will produce a “brainstem-dead” exam until the sugar is up. Give dextrose when the glucose is low or unobtainable. Give thiamine when malnutrition or alcohol use is plausible — before or with glucose — so you do not precipitate Wernicke encephalopathy. Naloxone belongs in the opioid-looking patient (pinpoint pupils, hypoventilation), not as a reflex for every GCS of 6.

Then obtain a noncontrast head CT unless the cause is already obvious and immediately treatable (documented hypoglycemia that is already correcting with a now-arousing patient). CT finds hemorrhage, hydrocephalus, large infarct with edema, and mass effect. If the story is sudden collapse with pinpoint pupils or locked-in features, add vascular imaging for basilar occlusion. Send CBC, chemistry, liver enzymes, ammonia when indicated, ABG, blood and urine toxicology, cultures if febrile, and consider lumbar puncture after imaging when meningitis or encephalitis is in play. EEG is next when nonconvulsive status epilepticus is possible (subtle twitching, unexplained coma after a seizure, or a flat-looking patient with a history of epilepsy). MRI is for the unclear case after CT, not a substitute for glucose and an airway.

Worked stems

A 62-year-old is found down. GCS E1 V1 M2, pupils 2 mm and reactive, glucose 32 mg/dL. The next action is dextrose (and thiamine if risk is present), not an immediate brain-death discussion and not a first-line MRI.

A 55-year-old with atrial fibrillation is mute and quadriplegic. Eyes are open and look up and down to command. This is locked-in syndrome from ventral pontine ischemia. Call it that. Image the basilar artery. Speak to the patient.

A 40-year-old opens eyes, has sleep–wake cycles three weeks after anoxia, and never follows a command despite off sedation. That is UWS/vegetative, not coma and not MCS until you document a reproducible aware behavior.

Exam traps

Calling locked-in “coma.” Diagnosing vegetative state on hospital day 1. Treating a unilateral hemisphere infarct as an adequate explanation for coma without herniation or a second lesion. Skipping glucose. Using GCS total without pupils or tracking. Independent practice items at /practice/abim-neurocritical-care drill these distinctions; they are a study bank, not the computer-based examination administered by ABPN.