Dr Ed Walker, who helped treat victims of Hillsborough, reflects on one of modern medicine's most difficult dilemmas

Is PVS (persistent vegetative state) a fate worse than death? Some think so. Resuscitation often salvages only a living shell of the former person. So should we be more realistic - perhaps by not using all our technology to its full extent, just because we can?

All doctors have seen victims of serious brain damage who are written off as having no meaningful life left but then astound everyone by waking up. A recent research paper in the journal Science documents the case of a 25-year-old woman injured in a road accident in 2005. The authors claim that she can hear and respond to questions. The validity of this one-off experiment has been questioned. But if it is replicated elsewhere, the results could have profound implications for the way cases of PVS are managed in future.

In March 1993, Tony Bland became the 96th victim of the Hillsborough stadium disaster, four years after the event. He was in PVS and died after a lengthy legal debate, which led to doctors being allowed to stop feeding him. Other well-documented cases, such as that of Terri Schiavo in the US, illustrate the torment that results from PVS.

There are two ways we could cut the number of severely brain-damaged patients. One is obvious but costly: new public health and safety measures could reduce the number of preventable brain injuries. But there is a simpler option, which would save NHS resources: improving the level and availability of pre-hospital care. Not through ambulance-response targets - unless we can fund a service that can reach each victim within three minutes. Defibrillation can treat 85 per cent of cardiac arrests, and with dramatic results, if done quickly. The safest city on the planet in which to have your heart stop is Seattle, which has thousands of defibrillators in public places and thousands of members of the public who know how to use them. Were we to follow this example, our intensive care units might be spared a lot of hopeless cases, and many families in the future saved from making truly impossible decisions.

Have you ever been unfortunate enough to see someone being resuscitated, or having cardiopulmonary resuscitation (CPR)? You will notice something strange: the focus is on the victim's chest. Air will be blown into it to make up for the lack of breathing, and someone may be thumping or compressing it to mimic the action of the heart. The chest, when it comes to resuscitation, is where the action is. This is strange, as the organ we are really trying to save lies further north.

The late Peter Safar, regarded by many as the father of modern resuscitation techniques, coined the term "CPCR" (cardiopulmonary-cerebral resuscitation) to correct the anomalyand to remind everyone that the organ they were trying to reanimate was in fact the brain. But it never stuck: CPR it is, and has been for a long time.

With the technology now available, we are better than ever at restarting hearts and keeping them going: technology can take over lung function and ensure excellent oxygenation and clearance of waste gases; failing kidneys can be bypassed and substituted with machines; infusions of drugs maintain blood pressure and ward off infection; flotation beds can prevent pressure sores. All of this to keep the "person", the brain, alive and well. But this clever medical trickery often goes to waste because it is far, far too late by the time it is applied.

The brain is a truly remarkable organ, infinitely and subtly complex. But this intricacy comes at a price - one of extreme vulnerability. Deprive the organ of oxygen for more than a few minutes and you risk anoxic brain injury. Strike its bony protective layer too hard and you will easily do irreparable damage to its fragile structure. The end result varies enormously, from almost instant death to minor personality changes discernible only to someone who has known the person for many years. Somewhere in this spectrum of brain-damage symptoms lies the condition known as PVS, a term coined in the 1970s to describe a person who appears awake but seems to have no awareness of, or interaction with, their surroundings.

Here is a typical scenario that will be playing out now somewhere not too far from you. A 78-year-old man has collapsed in the street. By the time someone thinks to check for a pulse, it is already 90 seconds since his heart stopped pumping. By the time that person begins chest compressions, three minutes have ticked by. Eight minutes after this (it's his lucky day in one respect - he collapsed near an ambulance station), paramedics arrive. With a mixture of IV adrenalin and electric shocks from a defibrillator, a heart rhythm is established, one that generates a feeble but definite pulse. So the man arrives at hospital with a heartbeat, and someone, somewhere gets to tick the box that says "life saved". But no one realistically expects him ever to leave hospital alive.

In one other respect, this man is fortunate, too. Given his age, he will be unlikely ever to reach the status of "vegetative" (used not in a pejorative way, but in the dictionary definition of "an organic body capable of growth and development but devoid of sensation and thought"). Nature, probably in the guise of an overwhelming infection, will take its course within a few days.

But the families of many young people every year, whose brains have been starved of oxygen but whose bodies are still strong, can face weeks, months or years of anguish with a loved one who is not quite dead, but neither very much alive.

PVS: the facts

Patients in PVS are awake, but are thought to be unconscious and unaware of their surroundings.

They may respond to pain or noise, sleep, wake up, smile, cryor moan.

Most vegetative states last a few weeks. After 30 days, patients are said to be in PVS. After a year, chances of recovery are very low.

About 1,500 patients in the UK have been in a vegetative state for more than three months, but the condition is rare and often misdiagnosed.

One 1996 study of 40 patients in PVS concluded that 43 per cent had been misdiagnosed and were in fact able to communicate.