Researchers who did the scans in Belgium say it justifies giving pain relief to all patients in this "minimally conscious state" (MCS).
"These findings might be objective evidence of a potential pain perception capacity in patients with MCS, which supports the idea that these patients need painkilling treatment," write Steven Laureys and his colleagues at the Coma Science Group of the Cyclotron Research Centre at the University of Liege in The Lancet Neurology.
But they found much less evidence that "brain-dead" coma patients in a so called persistent vegetative state (PVS) react to pain.
Laureys and his colleagues discovered after taking brain scans of patients and healthy volunteers as they received mild electric shocks.
They performed the procedure on five MCS patients, 15 healthy controls and 15 "brain-dead" patients.
Patterns of brain activity were the same MCS patients and the healthy controls, who rated the pain they received as "highly unpleasant to painful". Blood flow increased to parts of the brain that form the so-called "pain matrix", incorporating the thalamus and various parts of the cortex activated when we feel pain.
"I think it definitely means they feel pain because they activate the whole pain matrix," Laureys told New Scientist. "But what they feel is still an open question, whether they feel it the same way we do," he says.
Conversely the activity was much reduced in the PVS patients. "The difference between patients with MCS and PVS was very striking," says Laureys.
Stuart Derbyshire, who studies the connections between neural activity and pain at the University of Birmingham in England, cautions that the brain activation patterns don't necessarily mean the patients actually experienced pain. "We cannot read subjectivity from activation patterns any more than we can read it from changes in breathing, heart rate or blood pressure," he says.
"Patients undergoing procedures with sedation might also activate cortical regions during noxious intervention, but one might hesitate before calling that pain," says Derbyshire.
He points out that patients under general anaesthetic regularly grimace, wince and flinch, and these are probably indications of reflex signals from the much more primitive brain stem that can't reach the level of conscious experience. "Having said that, this study provides good evidence that whatever is happening in MCS patients during noxious stimulation is clearly more than a brain stem response," says Derbyshire.
Reacting to environment
John Whyte of the Moss Rehabilitation Research Institute in Elkins Park, Pennsylvania, comments in the same issue of The Lancet Neurology that the study "supports the conclusion that patients in MCS have the essential neural systems required to experience pain subjectively and that patients in VS might not".
But the difficulty for nurses and doctors, he says, is identifying patients who are MCS. "They're patients who react with their environment in a way that can't be purely reflexive," says Whyte. Examples might include patients following a moving object with their eyes, or somehow responding to a spoken command.
In the absence of a definitive test, Whyte says that doctors should assume that patients might sometimes be conscious even though they can't show it, so pain relief should be given as a precautionary measure, even to PVS patients. He says that most doctors in the US do this anyway, but the Belgian results provide further justification for it. "To not give it is to assume that patients are unconscious all the time, and I don't think the data fits with that."
Whyte says that coma patients who recover seldom complain of painful events or treatment during their comatose state, but this shouldn't be taken as evidence they never felt pain, because their memories are so impaired during this time that they wouldn't remember anyway even if they did feel pain. "They often don't report having felt pain, but it doesn't mean they didn't," he says.
Derbyshire says that an experiment that could resolve some of the issues would be to take scans while patients undergo procedures under general anaesthetic. These would show whether the same "pain matrix" regions of the brain light up even thought they are spared experience of pain by the anaesthetic.
Laureys says that it will be difficult for doctors to decide how much analgesia to give patients with MCS, because too much would sedate the patients, "decreasing the chance of seeing signs of consciousness and recovery".