Blood and Guts
Page 22
Rebuilding the face of William Spreckley during the First World War. The second picture shows the arrowshaped piece of cartilage implanted under the skin of Spreckley's forehead. Looking at the final picture, it is hard to believe that Spreckley was once without a nose.
Drawing of an early 19th-century operation for 'restoring' the nose. Although crude, these procedures were often successful.
Four members of the Guinea Pig Club in their ward at East Grinstead in the 1940s. They are all sporting tube pedicles showing the intermediate stages of facial reconstruction.
An 1850 interpretation of the passage of the tamping iron through Phineas Gage's skull.
A mask made of Phineas Gage's face (probably when he was still alive) pictured next to the railroad worker's skull. The partially healed hole can be seen clearly on the top.
Another day at the office for lobotomist Walter Freeman as he performs his procedure in front of a fascinated audience. This picture dates from 1949, before lobotomies became completely discredited.
Harvey Cushing, the brilliant, god-like surgeon who was adored by his patients.
Harvey Cushing operating on his 2000th brain tumour in 1931. During his impressive career, Cushing would save hundreds of lives and transform brain surgery.
A modern operation for the treatment of Parkinson's disease. The fearsome looking clamp is to hold the patient's head in place so that surgeons can place implants precisely in the affected area of the brain.
CHAPTER 5
SURGERY
OF THE SOUL
THE MAN WHO SHOULD HAVE DIED
Vermont, 13 September 1848
* * *
The navvies said that Phineas Gage was the best foreman they'd ever had. The twenty-five-year-old was fair and honest, a good worker and a fine leader. He was employed on the Rutland & Burlington Railroad. The promoters of the railroad hoped it would soon make them rich: winding through the wooded hills of New England, it would link Vermont to the cities of the east coast, bringing trade and creating new markets for the state's agricultural and mineral wealth. It was a fine enterprise indeed, and one well suited to an industrious and practical worker such as Gage.
Gage's work gang had been toiling since early morning near the town of Cavendish. They were building the roadbed – clearing and levelling the land in preparation for the rails. The plans called for a deep cutting, which had to be blasted through the granite hillside. When it was finished, rock would dwarf the trains as they rounded the sharp bend in the track. It would be a proud moment, thought Gage, when the first steam engine – wheels pounding, head lamp blazing – rolled along the track into town. For that to happen there was still much work to do. Gage would make sure it was done, on time and to the highest standards.
As foreman, Gage was highly skilled in the use of explosives, but it was a tricky and dangerous job. First his men would drill a hole in the rock – using a manual drill on solid granite was not an easy task. Gage made sure that the hole was carefully positioned so that the natural fractures in the rock could be used to maximize the effects of the explosion. Next he lowered a measured amount of gunpowder into the shaft and inserted a fuse. He tamped the powder gently with his tamping iron before adding a layer of sand. The sand helped confine the explosion to a small space, focusing the charge into the rock rather than back up through the hole, which was simply a waste of good gunpowder. Finally, Gage tamped the sand good and hard, lit the fuse and stood well back. It was a job he did every day.
Gage was so practised with explosives that he even had his own custom-made tamping iron. The three foot seven inch-long iron bar was an inch and a quarter in diameter. The bar was round, flat at one end – the end he used to pack the explosives and sand – and tapered to a point at the other. It was more than a crowbar; it was styled almost like a javelin. A fine iron bar for an iron-willed man. An unfortunate description given what was about to occur.
It was half past four. They were nearing the end of another hard day and Gage could not wait to get back to the inn where he was staying. Most of his men were looking forward to an evening drink, but Gage rarely touched alcohol himself. They were loading lumps of rock on to a flat car as Gage prepared to blast another section from the hillside.
He lowers the string of the fuse into the hole and pours in the gunpowder. He begins to tap the powder gently with his iron. Distracted by the work going on behind him, he leans forward over the hole. Perhaps he forgets that the sand hasn't yet been poured, or perhaps he slips. But when he tamps the iron again, it goes in too hard and catches on the granite. It ignites a spark.
The gunpowder explodes.
The iron rod shoots out of the hole like a bullet from a gun. It goes straight through Gage's cheek, passes through the floor of his left eye into the front of his brain and tears out of the top of his head. His skull is splayed apart as the iron continues its journey upwards, eventually returning to earth some eighty feet away, smeared with blood and bits of brain. Some of Gage's brain is later found splattered across the rocks where the rod landed.*
* The men who found the iron reported that it was 'covered with blood and brains'. They washed it in a nearby brook, but it still had a 'greasy' appearance and was 'greasy to the touch'.
Gage was knocked on to his back by the force of the explosion. His men ran across to find him twitching on the ground. A few moments later he spoke. Then, to everyone's astonishment, he got up and started to walk towards the road. He was helped on to an ox cart and driven the three-quarters of a mile into the centre of town. When he arrived at the tavern of Joseph Adams, where he was lodging, he walked with only a little assistance and sat in a chair on the veranda. He chatted with some of the men who gathered around and answered questions about his injury. Gage had rarely missed a day's work in his life and said he was keen to get back to the railroad.
When Dr Edward Williams arrived at around five o'clock he could not believe what he seeing. It made no sense – how could this man possibly be alive? Gage remained perfectly lucid, insisting that the bar did indeed pass right through his head. One of the labourers corroborated the story: 'Sure it was so, sir, for the bar is lying in the road below, all blood and brains.'
Despite the burn marks on Gage's cheek, the copious amounts of blood dribbling down the poor man's face and the fragments of bone sticking from his head, Williams was still unable to accept what had happened. It wasn't until Gage started vomiting a large quantity of blood and, as Williams noted, 'about half a teacupful of the brain, which fell upon the floor' that the doctor finally came round to the idea that Gage had survived the firing of an inch and a quarter-wide tamping iron through his head.
Williams was completely flummoxed by the case, and seemed reluctant to administer any treatment. So an hour later, when Dr John Harlow arrived, Gage was still sitting on the veranda answering questions and recounting his dramatic tale; also occasionally vomiting blood, bone and lumps of brain that had dropped through the hole from the top of his head into his mouth. Harlow was impressed with how Gage 'bore his sufferings with the most heroic firmness'. Despite becoming increasingly exhausted from the massive loss of blood, Gage recognized the doctor at once and needed little assistance to make his way up the stairs to his room.
Harlow was much more practical although, unsurprisingly, somewhat taken aback by the mess. 'His person and the bed on which he was laid were literally one gore of blood,' he recalled. However, this didn't stop the doctor passing his fingers completely through the hole. 'I passed in the index finger its whole length, without the least resistance, in the direction of the wound in the cheek, which received the other finger in like manner,' he later reported.
Together the doctors cleaned and dressed Gage's wounds. They shaved his scalp and removed a few bits of bone and a stray piece of brain that was 'hung by a pedicle', as well as bandaging the burns on his hands and arms. Harlow pressed the jigsaw of bones on the top of Gage's skull back into position as best he could and left the man propped up in bed,
where his bandages gradually became saturated with blood. A couple of the men volunteered to watch over him.
When Harlow returned at seven the next morning, Gage was still conscious. He had even managed to snatch some sleep during the night. Harlow didn't expect him to live for much longer, and the undertaker was called so that Gage could be measured for his coffin. It seemed the prudent thing to do. As the undertaker took his measurements, Gage's mother arrived to say her last goodbyes.
By 15 September Gage's condition had indeed deteriorated. He was passing in and out of consciousness, he was delirious and incoherent. On the 16th Harlow replaced the dressings but described 'a fetid sero-purulent discharge, with particles of brain intermingled'. That couldn't be good.
Harlow continued to visit his patient every day, and by the 22nd it seemed that the stubborn (and iron-willed) Gage was finally ready to die. He was hardly sleeping at all; he threw his arms and legs about as if he was trying to get out of bed. His body was hot, his wounds fetid. He even told the doctor, 'I shall not live long so.'
One month later Gage was walking up and down the stairs, even into the street. His wounds were healing rapidly and he was eating well. His bowels were described as 'regular' and he had even stopped vomiting globules of brain. By the end of November all the pain had subsided and Gage told the doctor that he was 'feeling better in every respect'. He could walk, talk and eat. There was only one problem: Gage was no longer Gage.
He described it as a 'queer feeling'. Others said the man had completely changed. The accident had radically transformed his personality. The railroad foreman who had once been described as sober, patient and industrious was now vulgar, impatient and impulsive. Gage was rude, they said, and could suddenly break forth into vile profanity. When he reapplied for his position as foreman his employers said the change in his mind was so marked that they refused to take him on. He was described as childlike in his attitude, but 'with the animal passions of a strong man'.
Gage's accident went beyond mere medical curiosity. When the iron bar tore away part of his brain it revealed the inner workings of the mind. It demonstrated that the brain is not some homogeneous grey pudding, but is made up of different parts doing different things. This is a concept known as localization, and would become vitally important for our understanding of the brain and for the first tentative advances in brain surgery.
Most of our personality, our sense of 'self', is contained behind the forehead, in the frontal lobes of the brain. These were the parts that were blasted away by Gage's tamping iron – the parts sprayed on to the rock or those that he later vomited across the floor. The frontal lobes are where we think and plan things. When the rod ripped through Gage's brain it tore away his personality and made him more impulsive. A century later surgeons would employ smaller rods to do much the same thing.
Gage never did return to the railroad. With his tamping iron as his constant companion, he travelled across New England. He eventually ended up in New York, where for a while it is said he became a sideshow in the famous Barnum's American Museum. For a few cents, punters could see a living man with a hole in his head. Although anyone expecting to see something truly gruesome would have been sorely disappointed. They could see (and perhaps if they were lucky, touch) the tamping iron, but the hole was now healed and there was little to show for Gage's trauma. Instead visitors could listen to Gage as he used another skull to regale his dramatic story.
In December 1848 Harlow's account of the case was printed in the Boston Medical and Surgical Journal. It was greeted with scepticism by the medical establishment, most of whom believed Gage's survival to be completely impossible. Surely Harlow must be mistaken? What would a rural doctor like Harlow know about the anatomy of the brain anyway? However, by 1849 Gage's case had attracted the attention of the new professor of surgery at Harvard Medical School, Henry Jacob Bigelow, who compiled a detailed account of the accident and paraded Gage (and his tamping iron) in front of surgical colleagues, suggesting that this was 'the most remarkable history of injury to the brain which has been recorded'. Thanks to Bigelow, Gage's accident would become a medical sensation and one of the most curious incidents in the whole history of surgery.
In the first days after the explosion it had been reported in the local paper as merely a 'Horrible Accident'. Workers died all the time on the railroad; it was hardly big news. But now, as more and more newspapers heard about the case, Gage's fame spread. He could have made a comfortable living on the medical freak show circuit – travelling around the USA from circus to surgical symposium (they often amounted to the same thing). He would be the nineteenth-century equivalent of a daytime chat show guest.
However, Gage's new impulsive nature took him in a different direction. The new Gage discovered that he enjoyed working with animals and went to work at a livery stable. For a while he cared for horses and drove a stagecoach in Chile. But with his health failing, he returned to the United States in 1859, finding employment on a farm in California. Then, in 1860, the accident that should have killed him finally did.
In February 1860, while ploughing a field, he suffered an epileptic fit. During these final few months of his life he started to suffer more and more fits and convulsions. Doctors did the only thing they knew how – which was to bleed him – but the treatment seemed to have little effect. Phineas Gage finally died in May 1860, twelve years after a three foot seven inch-long iron rod passed through his brain.
Although Harlow's treatment of Gage was exemplary, it is one thing to piece back together a fractured skull or even care for major head injuries such as those sustained by Gage, but it is quite another to open up the head and poke around in the brain – to have a crack at brain surgery. As far as most surgeons were concerned, any attempt to go further than repairing a head injury was to be avoided at all costs. Anaesthetics, advances in anatomy and, later, Joseph Lister's antiseptic operating techniques (see Chapter 1) might have transformed nineteenth-century medicine, but the brain was still a mystery, locked away in the sealed casket of the skull. Few surgeons were prepared to open this casket, and those who did usually came quickly to regret it. The only exception to the unwritten 'no operating on the brain' rule was the ancient practice of trepanning.
Trepanning is arguably the world's oldest surgical practice – although amputation is likely to run it a close second. It involves drilling a hole, from half an inch to two inches across, into the skull. The patient would have had their hair and skin scraped away before the prehistoric equivalent of a surgeon started to bore into their head with a sharpened stone or, later, a crude metal drill.
The practice of trepanning was in widespread use from around 10,000 BC, before the invention of reading or writing. The incredible thing is that archaeologists have found skulls with holes drilled in them all over the world. The evidence suggests that trepanning was being practised by many different peoples in completely different locations. These were communities that were segregated by geography. They had no possible way of contacting each other or, indeed, any knowledge of the others' existence. This implies that either many groups developed trepanning separately, or that the practice was passed down from our earliest human ancestors.
The big question is why? Why on earth would you want to drill a hole in someone's head? There could be any number of reasons, which historians can only guess at. In some civilizations, those who were trepanned also show evidence of head injuries, suggesting that trepanning was used as a treatment. It might also have been used to cure headaches, epilepsy or insanity. Perhaps it allowed demons to escape. There is speculation that it might even have given the recipient magical powers – a window, perhaps, to the gods.*
* You might think you need trepanning like a hole in the head, but it is still practised today. Surgeons have to use drills to access the brain, but there are also alternative therapy groups that recommend trepanning for all sorts of mental health conditions. In 2000, for instance, a British woman decided to do her own DIY brain surgery. Twen
ty-nine-year-old Heather Perry from Gloucester injected herself with a local anaesthetic before drilling a one-inch hole in her own skull. Unfortunately, the drill went in too far, damaging a membrane and requiring emergency medical help. Despite the mishap, she told reporters she had no regrets about the procedure.
Back in the nineteenth century there were three major challenges facing budding brain surgeons: the risk of infection, the danger of haemorrhage and the fact that they had very little idea what each bit of the brain did. The issue of infection applied to any major surgery, but infection from operations was gradually being defeated as Lister's reforms were adopted. As for controlling blood loss, this was still a significant problem. The brain has more than four hundred miles of blood vessels and consumes nearly two pints of blood every minute. The scalp, brain and bone are all extremely bloody, so cutting into the skin, skull and membranes surrounding the brain means there is a good chance of a patient bleeding to death on the operating table.
As for understanding how the brain works and the functions of its different areas, these are things that scientists are still grappling with today. Phineas Gage gave surgeons an insight into the role of the frontal lobes, but any sort of accurate map of the mind was still a long way off. All this left surgeons powerless to help people with brain injuries or tumours. Or to find a surgical cure for insanity. Even so, despite limited knowledge and unrefined techniques, some surgeons were still prepared to have a go.
AN OPERATION ON THE BRAIN
Hospital for Epilepsy and Paralysis, London, 1884
* * *