Does subvocalization occur when a person is comatose?

August 20, 2004 12:52 PM

Does subvocalization occur when a person is comatose?

I mean, I take it that subvocalization occurs in sleep, since people can actually talk out loud during sleep, but does that happen when people are in comas? I suppose it probably also depends on the type of coma...

posted by badstone to Health & Fitness (1 answer total)

I asked a neurologist friend about this. This is his reply:

The question you sent me is a very interesting one. The short answer is "no," but let's see if we can't make it a little more interesting.

Basically the comatose state is defined by the subject's level of responsiveness to her environment. For the sake of the quick bedside assessment, I divide alertness (or responsiveness, or arousal, or level of consciousness/awareness - all the same concept as far as I am concerned) into five broad areas:

Hypervigilant: someone is agitated, looking around a lot, paying attention to all features of the surroundings. Often seen in mania or delirium.

Alert: The normal waking state. Someone who is asleep but can be awakened and remains awake thereafter is normal.

Lethargic: Less alert than normal, but rousable to a normal level of alertness. Will descend into lethargy again if left unstimulated for a time.

Obtunded: Less alert than normal, and cannot be roused to a normal level of alertness, but still shows some level of responsiveness to stimuli.

Coma: Responsiveness to stimuli is absent or consists wholly of stereotyped reflexes.

Neurologists will argue about the precise definitions and the borders of these definitions, but few would disagree with the above.

The Glasgow Coma Scale was developed as a quick scale to assess the level of consciousness of a presumably acutely AMS (altered mental status) patient. The verbal part of the scale scores up to 5 points. You get 1 point for showing up with no verbal output. 2 points is incoherent mumbling, 3 points is aphasic babbling (words used, but no meaning conveyed), 4 points is confusion or disorientation, and 5 points is normal speech.

In practice the incoherent mumbling, when obtained, is usually evoked by a strong painful stimulus, such as a 'sternal rub' (medical term for a noogie) or nailbed crush. A patient that will not do this is comatose almost by default. The 3 point response is usually due to a structural lesion (stroke, tumor, hemorrhage) in the left cerebral cortex, and the 4 point response is most frequently caused by delirium.

Patients who are not responsive never produce speech. Even in the long-term evolution of structural coma to the 'persistent vegetative' or 'minimally aware' state, I have not encountered patients who vocalized words or recognizable phonemes. If a patient so much as cried out to a painful stimulus, or spontaneously, I would immediately discard a diagnosis of coma, and I would strongly question a diagnosis of persistent vegetative or minimally aware state.

There was some recent work at Cornell in which comatose patients were subjected to deep brain stimulation with an electrode implanted in the thalamus. These people had 'structural midbrain coma', where the coma results from damage to the alertness centers in the midbrain that activate the cerebral hemispheres. The theory was that in the absence of hemispheral damage, the electrode could possibly replace the effect of the alertness center, known as the 'reticular activating system.'

Some of these patients spoke whole words during stimulation, in one case after several years of coma. The experiments were eventually suspended owing to concerns about the ethics of proxy consent for such procedures.

Rather interesting implications, don't you think?

David Filippi, MD
posted by caitlinb at 6:27 PM on August 21, 2004