Admissions: A Life in Brain Surgery



For a surgeon to help operate on patients he did not know, whom he would scarcely ever see again, for whom he carried no practical responsibility – if there were problems Dev would have to deal with them – had always been anathema to me. And yet I had already discovered, to my surprise, that my lack of human contact with the Nepali patients both before and after surgery had not reduced my anxiety when I was operating. It did not seem to matter after all. Operating in Kathmandu I was in the same state of tense concentration as I was in London and it seemed that I cared just as much for the patients, even though my concern for them had now become entirely abstract and impersonal. I used to feel critical of surgeons if they were remote and detached from their patients but now, very late in my career, I was forced to recognize that some of this had perhaps been vanity on my part, and simply yet another attempt to feel superior to other surgeons.

Surgeons describe operating on patients with whom they have no personal or emotional contact as being veterinary. There was a veterinary surgery near the old hospital in Wimbledon and one of the vets there – Clare Rusbridge – specialized in veterinary neurological disorders. Devoted pet owners can take out insurance for their pets which includes the cost of MRI brain and spinal scans. Clare would bring to our weekly X-ray meetings fascinating scans of cats and dogs with neurological disorders. We would look at the scans at the end of the meeting and called it Pets’ Corner. They provided a bizarre contrast in anatomy to the images of human brains and spines with which we were so familiar. Cavalier King Charles spaniels, we learnt, often suffer from the brain abnormality known as a Chiari malformation, which humans also get. Labradors can develop malignant meningiomas. The spaniels’ problem is the result of selective breeding aimed to produce the small round head which wins points at dog shows. The malformation leads to spinal cord damage, and the poor creatures suffer from intractable pain and scratch themselves incessantly.

I operated with Clare on a couple of occasions, though she was unable to find an owner of a King Charles spaniel who was willing to let us operate on their pet. We did, however, once operate on a badger, which had been found confused and wandering on Epsom Downs and had been rescued and brain-scanned by an animal charity. The brain scan suggested that she might have hydrocephalus, although, to be honest, not much is known about badger brains. She was a beautiful creature and once she had been anaesthetized, I held her on my lap for a few minutes, stroking her grey and white fur, before Clare removed most of it with a pair of clippers in preparation for the surgery. I tried to carry out an operation for the possible hydrocephalus. I already had an article published entitled ‘Brain Surgery in Ukraine’ and I hoped I would be able to add to my CV ‘Brain Surgery in Badgers’, but the operation was not a success and the poor creature died. Or rather, she was ‘euthanased’.

‘At least our patients don’t have to suffer,’ one of Clare’s colleagues, who had watched the operation, commented afterwards. ‘Unlike yours.’

The first case I had done with Dev – two days before the operation on the child – had been an awake craniotomy for a tumour. This was the first time that such surgery had been carried out in Nepal. I had brought the equipment for cortical brain stimulation from London in my suitcase. Many years ago I had been the first surgeon in Britain to use the technique of awake craniotomy for treating a particular type of brain tumour known as a low-grade glioma. It was unorthodox at the time, but is now standard practice in most neurosurgical departments. It is, in fact, a very simple way of operating which allows you to remove safely more of a tumour in the brain than with the patient asleep under a general anaesthetic. The problem is that the ‘tumour’ is in fact part of the brain which has tumour growing in it – brain and tumour are muddled up together. The abnormal area, especially at its edges, looks almost identical to normal brain and only by having the patient awake, so that you can see what is happening to them as you remove the tumour, can you tell if you are straying into normal brain and running the risk of causing serious damage. Patients tolerate the procedure much better than you might expect, once they understand why it has been recommended.

The brain cannot feel pain: pain is a sensation created within the brain in response to electrochemical signals sent to it from the nerve endings in the body. When I see patients with chronic pain, I try to explain to them that all pain ‘is in the mind’ – that if I pinch my little finger, it is an illusion that the pain is in my finger. It is not ‘in’ my finger but really in my brain – an electrochemical pattern in my brain, in a map that my brain has made of my body. I try to explain this in the hope that the patient will understand that a psychological approach to pain might be just as effective as a ‘physical’ treatment. Thought and feeling, and pain, are all physical processes going on within our brains. There is no reason why pain caused by injury to the body to which the brain is connected should be any more painful, or any more ‘real’, than pain generated by the brain itself without an external stimulus from the body. The phantom limb pain of an amputated arm or leg can be excruciating. But most patients with chronic pain problems or conditions like chronic fatigue syndrome find this hard to accept. They feel that their symptoms are being dismissed – as they often are – if it is suggested that there is a psychological component to their problem and that a psychological approach might help. The dualism of seeing mind and matter as separate entities is deeply ingrained in us, as is the belief in an immaterial soul which will somehow outlive our bodies and brains. My ‘I’, my conscious self, writing these words, does not feel like electrochemistry, but that is what it is.

So, for an awake craniotomy, only the scalp needs to be anaesthetized and the rest of the operation is painless, although patients find having their skull drilled into very noisy – the skull acts like a sounding board. I therefore usually do this part of the operation under a brief general anaesthetic. The patient is then woken up, but unlike normal operations, where you wake up in a bed back on a hospital ward, with an awake craniotomy you wake up in the operating theatre, in the middle of the operation. There are various ways of conducting the ‘awake’ part of the operation. All involve using an electrode to stimulate the patient’s brain, which tells you where, in functional terms, you are on its surface. You will be able to produce limb movement or interfere with the ability to talk as the electrode momentarily stimulates or stuns the relevant part of the brain. It is a little like pulling the strings of a puppet. You also need to ask the patient to perform simple tasks or name and identify pictures if the tumour is near the speech centres of the brain. Some surgeons rely on speech therapists or physiotherapists to talk to the patient and assess them as the operation proceeds. I always relied on my anaesthetists, in particular Judith Dinsmore, whose highly skilled and reassuring manner never failed to keep the patients calm and cooperative.

I operate with a transparent screen between myself and the patient. Judith would sit facing the patient, talking to them and assessing the relevant functions – their ability to talk fluently, or to read, or to move the limbs on the opposite side of their body to the tumour (for obscure evolutionary reasons, each half of the brain controls the opposite side of the body). I would be standing behind and above the patient’s opened head and exposed brain, and watch and listen to Judith through the transparent screen as she put the patient through their paces. When she started to look anxious, I knew it was time to stop. If the patient had been under a general anaesthetic for all of the operation, I would have had to stop much earlier and would have removed less of the tumour. There would have been no way of knowing whether I was still removing tumour or normal, functioning brain. Obviously, more subtle social or intellectual functions cannot be tested, but this is not usually a problem. It would seem that low-grade gliomas have to be very extensive indeed – and effectively inoperable – before the patient’s personality is at risk.

I operate with a microscope which has a camera connected to a video monitor. The operation is mainly done with a simple sucker or an ultrasonic aspirator (which is a sucker with an ultrasonic tip that emulsifies what you are operating on). All you can see, as you look into the patient’s brain with the microscope, is the brain’s white matter, which is like a smooth, thick jelly. It is usually – but not always – slightly darker than normal because of the presence of tumour within it. It took me some time to learn to operate like this, with the patient awake. I am always a little anxious when operating and, at first, having the patient awake made this worse, especially as I had to affect a complete calm and confidence for their sake that I did not inwardly feel.

‘Do you want to see your brain?’ I will usually ask the patient. Some say yes and some say no. If they say yes, I go on to say: ‘You are now one of the few people in the history of the human race who have seen their own brain!’ And the patients gaze in awe at their brains on the monitor. I have even had the left visual cortex – the part of the brain responsible for seeing things on the right-hand side – looking at itself. You feel there should be some philosophical equivalent of acoustic feedback when this happens, a metaphysical explosion, but there is nothing, although one patient, having looked at his speech cortex, as I brushed it with a sucker and told him that was what was talking to me, commented: ‘It’s crazy.’

Towards the end of the first ever awake craniotomy in Nepal, the patient’s leg had suddenly become paralysed.

‘It’s probably temporary,’ I assured Dev. ‘It can happen when operating in the supplementary motor area, which was where the tumour was.’

I nevertheless awoke next morning feeling miserable. But Dev came to find me as I sat with a cup of coffee in the garden of his home – I had only stayed two nights in the hotel to which I had first been taken, and was now living in the guest house at the end of Dev’s garden. He told me that the juniors had rung to say that the patient had started to move his leg.

‘I knew you were upset, though you said nothing,’ he said.

The morning was instantly transformed.

‘Were there any admissions overnight?’ I asked.

‘Couple of head injuries,’ Dev replied.

I would often be rung at night when I was on call in London, although unlike Dev I was not on call every night. The telephone would ring and I would be dragged out of sleep, often with the strange illusion that I had chosen to wake up before the phone started ringing. These emergency cases were usually cerebral haemorrhages – bleeding into the brain caused by head injuries or a weakened blood vessel. I had to decide whether the patient should be operated on or not. Sometimes it was obvious that they would die if they did not undergo surgery and that they would make a good recovery after surgery. Sometimes it was obvious that they did not need surgery and would survive without it, and sometimes it was obvious that they would die whatever we did. But often it was not clear whether to operate and, if you did, whether they would make a good recovery. If the haemorrhage had been a big one, the patient was going to be left disabled, however well the operation went, as the brain – being so intricate and delicate – has much less capacity for repair and recovery than other parts of the body. The question then was whether the disability might be so severe – the patient left a ‘vegetable’, as the saying goes – that it might be kinder to let them die.

You can rarely predict with absolute certainty from a brain scan what sort of recovery the patient might make, but if we operate on everybody (as some surgeons do), without any regard to the probable outcome, we will create terrible suffering for some of the patients, and even more for their families. It is estimated that there are 7,000 people in the UK in a ‘persistent vegetative or minimally conscious state’. They are hidden from view in long-term institutions or cared for at home, twenty-four hours a day, by their families. There is a great underworld of suffering away from which most of us turn our faces. It is so much easier to operate on every patient and not think about the possible consequences. Does one good result justify all the suffering caused by many bad results? And who am I to decide the difference between a good result and a bad result? We are told that we must not act like gods, but sometimes we must, if we believe that the doctor’s role is to reduce suffering and not just to save life at any cost.

‘Twenty-six-year-old. Collapsed last night while in the shower. Looks like a spontaneous ICH. Probably an underlying AVM – there’s some calcification. Blown a big hole in the left basal ganglia, a bit into the midbrain too. GCS four according to the paramedics. Blown left pupil but came down with mannitol and ventilation. Lots of shift on the CT. Basal cisterns just visible. Now tubed and ventilated.’

‘Hang on a mo, I’ll have a look at the scan,’ I said. I pulled my laptop off the shelf by my bed and, balancing it on my knees, I spent a few minutes connecting to the hospital X-ray system over the internet. I looked at the scan.

‘He’s not going to do well, is he?’

‘No,’ my registrar replied.

‘Have you spoken to the family?’

‘Not yet. He isn’t married. There’s a brother, who’s coming in. Should be here soon.’

‘What’s the time?’


‘Well, we can probably wait until the brother’s here.’

Translated, the story – or ‘history’, as doctors call it – was of a young man who had suffered an intracerebral haemorrhage (ICH) from an AVM, an arterio-venous malformation. This is a kind of rare birthmark, a tangle of weak, abnormal blood vessels that often burst and bleed into the brain. The haemorrhage was into the left side of his brain, and also into part of the brain called the midbrain, which is important in keeping us conscious. It looked unlikely to me, from the scan, that if we operated he would get back to any kind of independent life. It is rarely possible to be certain, but I doubted if he would ever regain consciousness, let alone walk or talk again. His GCS was four, which meant he was in a deep coma. The scan showed a critical build-up of pressure in his head (‘lots of shift on the CT’, as my registrar put it). The fact that the pupil of his left eye was ‘blown’ – enlarged and no longer reacting to light – was a warning sign that without surgery he would probably die within the next few hours. The pupil had become smaller with a drug called mannitol, which temporarily reduces intracranial pressure, so we had a little time to decide what to do.

I couldn’t get back to sleep and went into the hospital an hour later. The sun was rising over south London, a long line of bright orange seen through the hospital windows. The corridors were quiet and empty as it was so early in the morning, but the ITU was very busy and full of noise. The twelve beds were all occupied and the nursing shift was about to change over, so there were many staff milling around the nurses’ station. There was a forest of drip-stands for intravenous fluids and syringe pumps, and flashing monitors standing guard beside each bed, the constant sound of the monitors bleeping and the softer, sighing sound of the ventilators doing the patients’ breathing for them. The nurses were all talking, handing over their patients to each other. The unconscious patients lay immobile, covered by white sheets, with tubes in their mouths connected to ventilators, IV lines in the veins of their arms, nasogastric tubes in their noses and catheters in their bladders. Some had drainage tubes and pressure-monitoring cables coming out of their heads.

My patient was in the far corner and there was a young man sitting at the bedside. I went up to him.

‘Are you his brother?’


‘I’m Henry Marsh, the consultant responsible for Rob. Can we go and talk?’

We shook hands, and left Rob’s bed to go to a small room used for interviews, for breaking bad news. I signalled to one of the nurses to join us. My registrar appeared, slightly out of breath.

‘I didn’t know you were going to come in so early,’ he said.

I gestured to the patient’s brother to sit down and sat opposite him.

‘We need to have a very difficult conversation,’ I said.

‘Is it bad?’ the brother asked, but he would have known already from my tone of voice that it was.

‘He’s suffered a major bleed into his brain.’

‘The doctor here,’ he said, pointing to the registrar, ‘said you had to operate.’

‘Well,’ I replied, ‘I’m afraid it’s a bit more complicated than that.’

I went on to explain that if we operated and he survived, there was very little chance of his getting back to an independent life.

‘You know him better than I do,’ I said. ‘Would he want to be disabled, in a wheelchair?’

‘He loved the outdoor life, went sailing… had his own boat.’

‘Are you close to him?’

‘Yes. Our parents died when we were kids. We were best mates.’


‘Not at the moment. Broke up recently.’ He sat with his hands between his knees, looking at the floor.

We sat in silence for several minutes. It is very important not to try to fill these sad silences with talking too much. I find it very difficult, but have got a little better at it over the years.

‘No chance?’ his brother asked me after a while, looking up at me, into my eyes.

‘I doubt it,’ I replied. ‘But to be honest you can never be entirely certain.’

There was another long silence.

‘He’d hate to be disabled. He told me that once. He’d rather be dead.’

I said nothing.

‘Rob was my best friend.’

‘I think it’s the right decision,’ I said slowly, even though neither of us had explicitly stated what we had decided. ‘If he was a member of my family, that’s what I would want. I’ve seen so many people with terrible brain damage. It’s not a good life.’

So the decision was made and we did not operate. Rob died later that day – at least, he became brain-dead, the ventilator was switched off and his organs were used for transplantation. I suppose it was just possible that I might have been wrong and he might have got back to some semblance of an independent life, or perhaps his brother was wrong, and Rob would have come to terms with being disabled, or simply have had no insight into it and led a happy, minimally conscious existence, no longer the person that he once was. Perhaps, perhaps – but doctors deal with probabilities, not certainties. Sometimes, if you are to make the right decision, you have to accept that you might be wrong. You may lose one patient with a good outcome but save a far greater number – and their families – from great suffering. It’s a difficult truth that even now I find hard to accept. When I received phone calls at night about cases like this, if I told the surgeon on call in the hospital to operate, I would roll over and get back to sleep. If I told him not to operate, and that it was better to let the patient die, I would lie awake until it was time to go to work.

The operation on the six-year-old child in Nepal was only two days after the awake craniotomy and not especially difficult. I had done many similar cases before, but rarely with tumours of this size. It involved separating the two halves of the brain – the cerebral hemispheres – using what is known as a transcallosal approach. It was an operation I had always taken a particular interest in because it was the one that had saved my son’s life many years earlier when, at the age of only three months, he had undergone surgery for a brain tumour. But his tumour had been only a fraction of the size of this one.

Dev and I had already decided that my principal role in his hospital would be to help him train his juniors so that they could learn how to do more than just the opening and closing and the emergency work at night. Within the first few days of coming to Nepal I knew, with the blind confidence of a lover, that as long as I could still usefully work, I would want to spend as much time as possible in the country.

‘You can’t go on running your hospital single-handed like this for ever,’ I told him, ‘and you need to think about the succession. You’re not that much younger than me. What do you want to leave behind?’

‘I know,’ he said. ‘It’s been worrying me a lot of late.’

‘I’d like to help, if I can be useful,’ I went on. ‘But you must promise to tell me as soon as I stop being useful.’

‘I agree,’ he replied.

You always get more nervous operating on children than on adults because of the terrible anxiety of the parents waiting outside the operating theatre. I had trained in paediatric neurosurgery and for many years had done most of the paediatric surgery in the hospital in Wimbledon. When that was closed and we were moved to the huge teaching hospital three miles away, I was unhappy about the way the paediatric neurosurgical ward was arranged. It was a very long distance from the neurosurgical theatres and my office, and it would no longer be possible for me to visit the children’s ward several times a day as I had done in the past. It had been my way of coping with the parents’ anxiety, which I understood all too well from the time when my own son had undergone surgery for his brain tumour and before I had become a neurosurgeon myself. To my shame and slight dismay, I found that I did not miss the paediatric work. In fact it was a relief to stop doing it.

‘Can I have the microscope please?’ I asked. The microscope – brand-new and as good as anything I have in London – was pushed into place.

‘Has it been counterbalanced properly?’ I asked. The second operation I had been involved with after arriving in Nepal had almost ended in disaster. The microscope, with an optical head weighing at least thirty kilos, had not been properly counterbalanced. The registrar had accidentally pressed the button that releases the perfectly floating optical head, and instead of floating it had crashed down onto my hands, almost forcing the instruments I was holding into the patient’s brain.

With Dev’s agreement we introduced a checklist to be completed before each operation, making sure (hopefully) that this would not happen again. I found it ironic that despite my well-known hatred of paperwork and checklists in my own hospital, I was now trying to introduce them in the hospital in Nepal.

‘Yes, sir,’ said Pankash, the registrar assistant, in answer to my question about counterbalancing the microscope. The registrars are very polite and respectful. If they do not know the answer to a question, they find it quite impossible to admit this. Rather than say no, they will stand speechless. The silence can last for many minutes and makes teaching very difficult. I had quickly resigned myself to never knowing what they really thought about what I told them or the questions I asked them.

I positioned the microscope and cautiously pressed the release button. The optical head remained steady and I settled down in the operating chair, which had also been pushed up to the table.

‘What is the first rule of microsurgery?’ I asked Pankash.

‘To be comfortable, sir,’ he replied. I had told him this the day before.

‘Look,’ I said. Pankash was peering down the microscope’s side-arm so he could see what I was doing. Dev was watching on the monitor. I put the retractor against the inner side of the right hemisphere and gently pulled it a few millimetres to the right, away from the thick midline membrane called the falx cerebri that separates the two hemispheres. As I looked down the binocular microscope, it was as though I was descending a ravine or negotiating a narrow crevice, with the shiny, silvery-grey surface of the falx to the left and the pale surface of the brain, etched with thousands of fine blood vessels, glittering in the microscope’s brilliant light, to the right. Even after thirty years I still find using an operating microscope profoundly exhilarating – the feeling of beauty and mystery and exploration has never left me. After years of practice, the perfectly balanced instrument is like an extension of your own body and you feel – until things go wrong – equipped with superhuman powers.

‘If we are lucky we will quickly drop right down onto the corpus callosum – there it is!’

The white corpus callosum came into view at the floor of the chasm, like a white beach between two cliffs. Running along it, like two rivers, were the anterior cerebral arteries, one on either side, bright red, pulsing gently with the heartbeat, which you must not damage under any circumstances. The corpus callosum contains countless millions of nerve fibres joining the two halves of the brain. If all of the corpus callosum is divided – as is occasionally done for severe epilepsy – patients develop what is called the ‘split-brain phenomenon’. Outwardly they appear normal enough, but if placed in an experimental situation where the two cerebral hemispheres each see a different image, the two halves of the brain can end up disagreeing about what they are seeing – in particular, what something is called and what it is used for: knowledge of names is in the left hemisphere, and knowledge of how to use things is in the right. The self has been split. Actual conflicts between the two halves of the brain are rare, but it is said that one patient, on losing his temper with his wife, attacked her with his right hand while his left hand tried to restrain it.

Who has not felt contradictory impulses within themselves? The more you learn about the brain, about our true selves, the stranger it becomes. It is almost as if we have many competing and cooperating selves within our brains, and yet somehow they all resonate together to produce a coherent individual capable of thought and action. There was a famous experiment many years ago by the American neuroscientist Benjamin Libet – confirmed many times since – that showed that the conscious decision to move the hand is preceded by electrical activity in the hand area of the brain. Nobody has yet provided a satisfactory account of what this really means. It is as though the deciding self is no different from the sailor in the storm who is forced to steer his boat in the storm’s direction, but then claims to have chosen the direction himself. And yet to claim, as some do, that the conscious self is an illusion, that it is no more than running before the wind, or a consoling fairy story, somehow seems as implausible as maintaining that pain is an illusion, and not ‘really’ painful.

I was only going to make a small hole in the child’s corpus callosum. This would bring me straight to the centre of her brain where the tumour was growing. A small callosotomy, as it is called, does not seem to produce any obvious problems for the patient after the operation. Besides, there were other, much greater risks to the child from what I was going to do, just as there were from not removing the tumour in the first place. And yet I was not at all certain that it was worth it: the child was almost certainly doomed whatever we did. Although the tumour was benign it was simply too large to remove completely without causing awful damage, both to the fornices and to an area near them called the hypothalamus. The hypothalamus controls vital functions such as thirst, appetite and growth. Children with hypothalamic damage typically become morbidly obese dwarves.

It was easy enough to find the tumour. It was at least four or five centimetres in diameter and of a soft, grey consistency which ‘sucked easily’, as neurosurgeons say. It was difficult to see if the all-important fornices still existed or whether they had been obliterated by the tumour. I managed to preserve a thin thread of white matter on one side that might, or might not, have been all that was left of them.

‘Have a look,’ I said to Dev, gesturing to the side-arm on the microscope. ‘Do we know what her memory was like before the operation? It may already have been very poor, so there is less to be lost.’

‘No,’ he replied, ‘but it’s not going to make much difference to what we do, is it?’

‘No, I suppose not,’ I said, a little reluctant to accept his brutal realism. He scrubbed up and, once he had taken over from me, I went to have a cup of coffee.

Dev’s office has a broad window looking out onto the space in front of the hospital, one floor down. In the distance you can see the green foothills to the north, which, whenever I look at them, invariably make me think of the celestial, snow-covered high Himalayas that lie hidden beyond them and which I long to see.

I drank my coffee and went back to the operating theatre. I scrubbed up and joined Dev and, after removing more of the tumour, we found that we had passed all the way through the child’s brain onto the base of the skull on the right side. The tumour was so large that it had, effectively, split the lower part of the front of her brain in half.

‘I think it’s time to stop,’ I said to Dev. ‘We’ve taken out at least one fornix and one half of the hypothalamus – any more damage and she’ll be completely wrecked.’

‘I agree,’ said Dev. ‘She can have radiotherapy for what’s left.’

‘It’s hard to know how much of the tumour is still there,’ I said ‘Maybe twenty per cent.’

There is no pleasure or glory in this kind of operating, I thought to myself as I tidied up the bleeding in the large cavity we had left in the girl’s brain by removing the tumour. There’s tumour left behind, she’s almost certainly going to be left very damaged, and all we have done is slow down her dying.