8
Curious Influences
What is the oldest profession in the world? The stock answer is, as we all know, prostitution. However, a very long time before any prostitute ever walked the streets, and well before any female even existed, going right back to the genesis of the world in the Old Testament, we know that the Lord God ‘… took one of the man’s ribs and then closed up the place with flesh’ (Genesis 2:21–24). This, of course, was the famous rib from which Adam’s companion, Eve, was supposed to have been made. Be that as it may, to us cardiothoracic surgeons, a rib resection is a rib resection, and it is a stock-standard operation in the surgical repertoire of any of us. This means that the world’s oldest profession is, in fact, cardiothoracic surgery, and its first practitioner was God himself.
I used to enjoy telling this somewhat lame joke to anyone who would listen at medical dinners, until I was challenged by a particularly irritating anaesthetist who gleefully pointed out that the full text of that excerpt from Genesis is as follows (my italics):
So the Lord God caused the man to fall into a deep sleep; and while he was sleeping, he took one of the man’s ribs and then closed up the place with flesh. Then the Lord God made a woman from the rib he had taken out of the man, and he brought her to the man.
In other words, the world’s oldest profession was in fact not cardiothoracic surgery, but cardiothoracic anaesthesia.
There is little love lost between surgeons and anaesthetists. They work in close proximity to one another. They are separated only by a surgical drape. On one side of the drape is the anaesthetist, with the requisite paraphernalia of respiratory machines, monitoring equipment, and powerful drugs and the means to administer them. On the other side is the sterile field, the open wound, the blood, the sharp instruments, and the surgeon. This drape has been christened by some witty anaesthetist with the provocative name of the blood–brain barrier.* So to anaesthetists, they are the brain (intelligent, reasonable, thoughtful) and we surgeons are the blood (gory, impetuous, sanguine).
[* There is a real blood–brain barrier: it is the intricate cellular and molecular complex that protects the brain by stopping certain drugs and other bad substances from crossing into the brain from the bloodstream, and has nothing to do with surgical drapes.]
In the eyes of many surgeons, anaesthetists are lazy, crossword-obsessed tea and coffee addicts who would go to any length to avoid doing some work and who are constantly watching the clock, their only interest in the life-or-death operation taking place before their very eyes being summarised thus: ‘When will this bloody procedure finish so I can go home?’ Predictably, the view from the anaesthetic side of the blood–brain barrier is that surgeons are a bunch of bloodthirsty, arrogant, aggressive, overconfident cowboys with massive egos and small brains. These are, of course, somewhat extreme caricatures, but, like all caricatures, there is more than just a grain of truth in them.
Anaesthetists can be viciously eloquent in mocking surgeons. In the second edition of the serious educational textbook Core Topics in Cardiac Anaesthesia by Jonathan Mackay and Joseph Arrowsmith, there is an appendix that contains a table of the phonetic alphabet. Alongside this, another table contains what the editors (both anaesthetists) describe as the ‘surgical alphabet’. The phonetic alphabet of course goes like this:
Alpha, Bravo, Charlie, Delta, Echo, Foxtrot … and so on.
The surgical alphabet, according to McKay and Arrowsmith, goes like this:
Accuse, Blame, Criticise, Deny … and so forth.
Another famous and oft-repeated anaesthetic quip is that there are only two types of surgeons: the bastards and the slow bastards. A third is the very well-known ‘definition’ of anaesthesia:
Anaesthesia is the half-awake
watching the half-asleep
being half-murdered
by the half-witted.
There are many other examples of anaesthetic scurrilous attacks on surgeons, but it most certainly is not one-way traffic. If we surgeons have not yet quite reached the same exalted level of vitriolic invective in abusing our beloved anaesthetic colleagues, it certainly has not been for lack of trying:
You can easily identify the professions of those who work in an operating room by the spillage on their shoes: On the shoes of a cardiac surgeon, there is blood. On the shoes of a urologist, there is urine. On the shoes of a liver surgeon, there is bile, and on the shoes of a colorectal surgeon, there is, of course, shit. How do you recognise an anaesthetist? Easy: on their shoes, there’s coffee.
One saying that is a favourite plaint by cardiac surgeons is the following:
Heart surgery is a team effort
(until the patient dies, then it’s just the surgeon’s fault)
This speaks volumes, if only to highlight how exasperated some surgeons are with the inescapable fact that they, and they alone, tend to be the ones held responsible for poor results, when the care is actually delivered by an entire team. Most people do not know that, in a standard cardiac operation such as CABG, there are, on average, around nine people working within the operating theatre, and they are as follows:
On the surgical side:
The above members of staff are scrubbed, gowned, and gloved. The others, who are not scrubbed (and who are stationed beyond the blood–brain barrier), are:
In total, then, with nine people working as a team to deliver an effective and safe operation, why is it that, if the patient dies, it is purely the surgeon’s fault?
A few years ago, at the major annual national meeting of cardiothoracic surgeons in the UK, there was an important presentation dealing with the outcomes of surgery. Numbers of operations were presented, and so were their outcomes, and the distribution of such activity among all the heart surgery hospitals in the country. The hall was packed with surgeons from all over the nation as these results were presented, but this happens every year at this particular meeting. The novelty in this particular year was the announcement that these outcomes were now to be made public and presented alongside surgeons’ names. Needless to say, not all the surgeons in the auditorium were especially thrilled with this idea. One particularly aggravated heart surgeon stood up and addressed the conference, saying that, in his opinion, when it came to death from heart surgery, the heart surgeon is merely an ‘innocent bystander’, or, put in other words, the death is everybody else’s fault.
Many laughed at this somewhat polarised view, but the reality is that we were somewhere between two extremes. At one end, there was an increasingly prevalent culture that blamed the surgeon squarely and entirely for the outcome, and, at the other end, a minority represented by our surgeon who felt that a bad outcome was not at all his fault, but everybody else’s. The truth, as always, probably lies somewhere in between. On listening to this exchange, a number of thoughts coursed through my mind. The first, I’m afraid, was a rather vicious one: I wondered whether that very same surgeon who claimed that he was merely an ‘innocent bystander’ when a patient dies would also rush to claim that he was merely an ‘innocent bystander’ when a brilliant and difficult operation goes particularly well, and the patient survives against the odds. Being of a somewhat uncharitable disposition, and knowing surgeons as well as I do, I thought ‘Probably not’.
My second thought was a little more productive. We all know that the surgeon makes a difference to the outcome, but do other members of the team? Perhaps we should investigate this to find out. Which member of the team shall we investigate first? The answer, of course, came by itself, screaming loudly from all the rooftops through a megaphone: the anaesthetist! Who else?
Briefly, we looked at more than 18,000 patients operated on at Papworth Hospital up to the year 2012, with 21 senior surgeons and 29 senior anaesthetists in charge of conducting the operations. We examined their predicted mortality and their actual mortality, and tried to see if there was any evidence of significant variation associated with who the surgeon was, and also if there was any evidence of significant variation associated with who the anaesthetist was (Papachristophi 2014). What we found was extraordinary: who the surgeon was actually did make a difference. Who the anaesthetist was made not an iota of difference: there was no appreciable difference in outcomes that could be linked to the anaesthetist. In fact, we have just extended and further analysed the data on surgeons and anaesthetists, and actually measured their impact on survival. If you have a heart operation at Papworth Hospital, whether you survive or succumb to the operation will depend on the following factors: 97 per cent of the outcome is determined by EuroSCORE, just under 3 per cent is determined by which surgeon operates on you, and a measly 0.01 per cent depends on which anaesthetist puts you to sleep. To put it another way, by far the most important factor is you, your operation, and your risk factors. Which surgeon you choose has a minuscule effect, and it doesn’t matter a jot which anaesthetist you have, so just go for the one with a friendly smile.
This study, published in the Journal of Cardiothoracic and Vascular Anesthesia in 2014, was important for two reasons. The first was that it confirmed that our anaesthetic department, which usually follows harmonious and standardised protocols, and administers anaesthetic care according to fairly rigid, pre-agreed guidelines, was safe no matter who the anaesthetist was. Other anaesthetic departments in other institutions do not necessarily do this, and various anaesthetists in other hospitals are free to select the methods of anaesthesia that suit them, their habits, and their prejudices. To my mind, this study provides a strong stimulus for similar studies to be conducted in such hospitals. If their findings are the same as ours, and there is no variation, then there is no problem, but, if they find that one anaesthetist has better outcomes than another, then there may be a strong case for studying the technique used by that anaesthetist to see if others could learn by adopting it and thus improve their results. My anaesthetist colleague Andrew Klein has just completed a similar study of a number of UK hospitals, and found that there is indeed variation in anaesthetic outcomes in some of them, and, in one of these hospitals, the effect of the anaesthetists has almost the same magnitude as the effect of the surgeon at Papworth. This study is not yet published, but, when it is, it will have important implications for the anaesthetic protocols (or lack of them) in some of these institutions.
The second reason was more personal. For the first time ever, I was able to give lectures to prominent anaesthetists in many parts of the world to show that we were unable to find that the anaesthetist had a discernible influence on outcomes, and conclude perfectly legitimately that, according to our study, we had proved that there is no such thing as a good anaesthetist.
You may well think that this is a little contrived, slightly petty, or frankly pathetic, but at least it scores one point for the surgeons in the endless and unequal war of abuse between these two great but disparate professions.
What about other external influences on the outcome of an operation? What can our quality-measurement tools reveal about those?
Arthur Hailey was a British-Canadian novelist who wrote a string of best-selling novels that almost always have an industrial or commercial backdrop. Among his most famous titles are Hotel, which became a successful and long-running television series; Airport, which was made into a Hollywood blockbuster; Wheels, which took in the American motor industry in Detroit; and The Final Diagnosis, which is set in a hospital pathology department, and happens to be my favourite. Hailey spent a very long time in the environments in which he set his novels, and his work is characterised by meticulous research and insider knowledge of the background to his stories. This meant that the casual reader effortlessly learned a massive amount about the intricacies of airports, hotels, and pathology departments while being entertained with multiple parallel storylines. As each strand seemed always to end a chapter on a cliffhanger, and the next chapter took up the cliffhanger from a previous strand, this had the effect of making the book truly difficult to put down despite the lamentations of the critics about the lack of literary style.
In Wheels, published in 1971, Hailey writes that a well-known fact in the motor trade is that one should avoid buying a car built on a Monday or a Friday. The reasons he gives for this are obvious and intuitive: on a Monday, the workers are dragging their feet back into the factory, and, on a Friday, they can’t wait to get away, so that on both of these days their workmanship may be expected to suffer. I am not aware if any independent research has ever confirmed or refuted this allegation, but, as I have an obsession with the subject of the quality of surgical work, I could not help wondering whether surgeons may also be prone to such vagaries. I was not specifically interested in the Monday/Friday business, as surgeons, by and large, enjoy their work, and many of their long-suffering families would testify to the fact that it is difficult to get them to drag their feet away from the darned hospital rather than the other way around. What I wondered about, however, was the impact of a prolonged break, such as a holiday, or a prolonged period at work, such as on the day immediately before a holiday, on surgical performance.
It is conceivable that surgical skills need to be exercised continuously to be maintained, and that a holiday may cause these skills to become a little rusty, so having an adverse effect on performance on the first day back. It is also conceivable that the day immediately before a holiday is the last day of a prolonged work period without a break, and a surgeon may be tired or even feel burnt out, badly in need of a refreshing change and a rest, and perform sub-optimally as a result of these feelings. When you think about it, both scenarios are equally and unpleasantly plausible ones. With risk-adjusted outcomes readily available to us, it was not too difficult to find the answer (White 2007).
We studied 7,873 patients who had heart operations at Papworth Hospital over a four-year period, and we began by dividing these patients into three groups. The first group consisted of those patients who had their operations done on the very last day before a surgeon went on holiday. The second included those patients who had their operation on the very first day of the surgeon returning to work after a holiday. The third group (the majority) was the control group: everybody else. We adjusted all of these for risk, just in case one of the groups contained higher-risk patients than others, and we looked to see if mortality in the groups operated just before a holiday or just after a holiday differed from the rest. The results of this study were quite a revelation.
The first thing we found was that the patients were remarkably similar: our surgeons had clearly not made any special efforts to operate on (or to avoid operating on) particular types of patient if they were about to go on holiday, or if they had just come back: they took all patients as they came.
The second thing we found was that mortality for the entire group was only 4 per cent. For the period of the study at the beginning of the last decade, this was a good result overall. The mortality for the control group was similar to the total, but mortality in patients operated just before a holiday was more than double the mortality on the first day after a holiday. This did not quite reach statistical significance,* but there was a strong and persuasive trend.
[* The p-value was 0.053. You need a p-value of less than 0.05 for the difference to be considered statistically significant.]
The study confirmed irrefutably that holidays do not cause surgical skills to become rusty. On the contrary, there seemed to be a patient-protective effect of operating on the first day after an absence, and a patient-damaging effect of operating on the last day before hopping on the plane.
Why is that? There can be many reasons for this, but I have a theory relating to personality types that I will explain later on. In the meantime, surgeons who are planning a holiday would be well advised to stay at their desk on the last working day before the holiday, where they are likely to do less damage.
Another question that our new tools can address is whether surgeons should stop operating if a patient dies on the table. When a patient dies under the care of a physician rather than a surgeon, it is often the patient who gets the blame, and the stock phrase used is: ‘I am sorry, but he or she “failed” to respond to the treatment.’ The underlying assumption behind this platitude is that it was somehow the poor patient’s fault. Of course, the right treatment will sometimes truly fail, and patients will sometimes fail to respond, or the wrong treatment is given, or the right treatment is given but badly, or there is no right treatment available. Any of these can and do cause treatment failure and death, but when a physician’s patient dies while undergoing treatment, no automatic assumption is made that the physician prescribing the treatment is responsible.
This is somewhat different in surgery. When a patient is operated as an emergency with a view to salvage life, the surgeon may be treated in a similarly charitable and forgiving manner, and thus the patient may also be seen to have ‘failed to respond to treatment’, perhaps because the injury or the condition were too severe. Yet most surgical operations are not carried out as an emergency: they are carefully planned and performed to relieve symptoms or improve outlook in patients who spontaneously and willingly walk into hospital on their own two feet. If, a few days later, the patient is carried out of hospital in a wooden box after an operation has been performed, it is hard for the casual observer to avoid, consciously or subconsciously, linking such an outcome with the surgeon who did it.
If it is difficult for a third party to separate the outcome from the operation, and the operation from the operator, then it is doubly difficult, if not impossible, for the surgeons themselves to make such a separation in their own minds. I have worked as a surgeon surrounded by many other fellow surgeons for the past 25 years, and have never known a surgeon to take the loss of a patient lightly. That said, the reactions of surgeons to such events can be very interesting and very, very different, and this is often a reflection of their different personalities.
One of my finest trainees was an intelligent, softly spoken, and kind young Irishman called Andrew Drain. He tragically died of a virulent form of leukaemia before completing his specialist training. He had an unusually mature early interest in the psychology of the surgical mentality, and he and I once tried to categorise the reaction experienced by surgeons when they ‘lose’ a patient.
The closest we got to understanding these reactions was to recognise that the surgeons who experience them could be classified into one of two broad stereotypes. The first is a self-flagellator: this surgeon would go back over the case, reviewing every single detail, asking ‘What if?’ at every juncture, and self-blaming for any aspect of the patient management where care may have been anything but optimal. After an intense period of self-scrutiny, sadness, and bitterness, this surgeon draws a line under the events, and returns to work the following day, a little chastened and perhaps a little more careful. The second type would also go back over the case, reviewing every single detail, asking ‘What if?’ at every juncture, until the discovery and identification of a minor error or omission by somebody else. This type of surgeon is then absolutely convinced that the death is somebody else’s fault, becomes visibly and palpably cheerful, and returns to work as if nothing had happened. I am not exaggerating: I still remember a senior surgeon who rushed into hospital on a Sunday night after his patient unexpectedly died. This was not in order to speak to bereaved relatives or provide moral support for his junior staff, but specifically to pore laboriously over the case notes of the recently deceased patient in the nurses’ office. Twenty minutes later, he identified a minor electrolyte abnormality that was not, in his opinion, adequately dealt with by the hapless resident medical officer on call two nights previously. He looked up from the notes, turned to me, and said, ‘Aha: he killed my patient’, then smiled, got up, and went home to his gin and tonic, satisfied that his amour-propre was intact. This was not a cavalier surgeon, nor was he one who did not care for his patients. He was a hard-working, dedicated, and selfless man who devoted his life to surgery, but his behaviour here was driven by the need to restore self-confidence before facing the next operating day.
In a way, this is not all that surprising. Regardless of the coping mechanism employed, a surgeon must ensure that he or she is going to be able to function almost immediately after the tragedy. After all, other patients are awaiting treatment the very next day, and the day after that, and operations must be performed by competent and confident individuals who are capable of making correct decisions and making them quickly. The operating theatre is no room for indulging in excessive introspection and wallowing in the paralysis of all-pervasive self-doubt.
If all of the above is true, then it must apply with even greater poignancy when a death occurs not a day or two after surgery, but on the operating table itself. It is here that separating operator and operation from outcome is nigh impossible. And it is here that one would expect to find the most profound impact of the outcome of the last case on the outcome of the next one.
The issue first came to the fore in abdominal surgery, the branch of surgery that deals with the contents of the belly, such as the gall bladder and the intestines. Some abdominal surgery is carried out in desperate emergencies, such as when patients are brought in with the rupture of an internal organ and dangerous infections of the body cavity. These patients may arrive in the emergency room in a desperate state, with raging infection and failure of multiple organ systems, and it is not surprising if some of these patients die despite the best efforts of the surgeons. Most abdominal surgery, however, is routine, carefully planned, and carried out electively to remove a diseased gall bladder, fix a hernia, and so on. It is true that many patients who present for such surgery are elderly, and some may have heart and lung problems purely as a result of their age. Despite the worsening risk profile of these patients, advances in anaesthesia, monitoring, and the general overall care of the surgical patient have resulted in elective abdominal surgery becoming a very safe treatment indeed. All of this means that, in such elective surgery, the death of a patient on the operating table — an intra-operative death — is now a vanishingly rare event.
Rare events do sometimes happen, though, and they occasionally happen in rapid succession. In Scotland, one particular surgeon (and his hapless patients) had an almost incredible misfortune: having experienced the unexpected with the death of one patient on the operating table, he continued with the day’s scheduled operating list only for another patient to suffer the same tragic fate. For two patients to die consecutively on an operating table was virtually unheard of, and, not surprisingly, the media took notice, and there was a clamorous outcry. One focus of the reaction to this bit of medical news was that many sections of the media found it incomprehensible that surgeons ‘carried on regardless’ with the day’s work after such a calamity.
In the UK, a death during surgery must be reported to the appropriate authorities. The coroner’s courts deals with such deaths in England, Wales, and Northern Ireland, while the procurators fiscal and sheriff courts deal with them in Scotland. These cases were duly reported to the local procurator fiscal, and were heard some time later in court. During the hearing in the Falkirk Sheriff Courthouse on 26 January 1999, some expert witnesses advised the court of their educated opinion that, after an intra-operative death, surgeons should cease operating that day. Sheriff Albert Sheehan listened to this advice, and recommended that the Scottish Royal Colleges and the Scottish Intercollegiate Guidelines Network should consider whether guidelines or advice were needed for surgeons to follow if they experience a death on the table. Should they carry on regardless? Should they take some time off?
At that stage, there was an utter lack of consensus on the matter. Opinions were divided, and appeared to be deeply held, but none of them was backed by the slightest shred of evidence. A survey of Welsh orthopaedic surgeons looked for some information about working after an intra-operative death. Among the survey participants, only one of 16 surgeons who had such a death actually cancelled further operations that day, but eight of the surgeons who experienced the death of a patient during surgery felt that some time without operating would have been advisable. Some cardiac surgeons (who, by the nature of their work, are more familiar with such events than their orthopaedic colleagues) said that they behaved differently after a death on the table, and many anaesthetists stated that if surgeons and their performance are affected by such an event, then their close colleagues, the anaesthetists, are equally affected, if not even more so. Questions were also asked about the rest of the operating team.
The events in Falkirk raised many questions that the medical profession had not previously addressed. Indeed, it was a question that we doctors did not know was there. Should we stop operating if a patient dies on the table?
There are many ways of addressing the issue, and many ancillary questions can be asked. A few of these are listed below:
In cardiac surgery, a death on the operating table is still a rare event, but it certainly occurs far more often than in abdominal or orthopaedic surgery, or indeed in any other kind of surgical specialty. There are, of course, many reasons for this.
The most obvious reason is that the heart and the great vessels that pour into it or out of it are of course both big and full of blood. A tear or rupture, whether caused by disease, injury, or a surgeon’s knife, can result in catastrophic bleeding. Your heart pumps your entire blood volume around your body once a minute, which means that a large hole in a heart chamber or a big blood vessel, while the heart is still beating, will result in ‘bleeding out’ in around one minute.
The second reason is to do with the heart itself. When an orthopaedic surgeon sets a broken bone, that bone, together with the limb within which it is situated, is rested in plaster until healing occurs. When a piece of bowel or stomach is taken out, the alimentary system is given a period of rest, with ‘NIL BY MOUTH’ instructions until the gut recovers. Almost every organ that is operated on is given a chance to rest and recover before it is expected to take up its duties once more. Not so the heart. The sick heart, made temporarily even sicker by being operated on, is simply expected to get on with it, and go back to work as soon as the operation is finished. This is mandatory: a patient whose heart does not work at all at the end of an operation is almost always a dead patient. (I say almost always, because we now have artificial hearts that can buy some time in the hope of a quick recovery of the real thing, but they are imperfect machines, risky to use, fraught with complications of their own, and ridiculously expensive.)
Sometimes, bleeding out and a non-functioning heart can happen together in the same patient. For such a patient, death is a double certainty.
Most heart surgeons will experience a death on the table at least once. One such heart surgeon, Stephen Large, is a colleague and friend of mine. He wrote an account of the death on the operating table of a young man, aged only 17, who died after a moderately high-risk operation (see Appendix B). In this account, he describes the events leading to the death, and the impact that these events have had on him many years later, which he calls a ‘true haunting’. My own experience of death on the table is sharpest in relation to its immediate impact, and it is a feeling of catastrophic and absolute loneliness. When a patient dies on the table, everybody in the operating theatre is still physically present, but, to my mind, they have all simply disappeared. All that is left is the patient who has been failed, and the surgeon who has failed him or her, with the harsh glare of the operating-room light focused unforgivingly on us both.
As such a tragic event is a rare but regular feature of heart surgery, I felt that we heart surgeons were the ideally placed specialty to look into its impact and address the questions listed above. In particular, I wanted to find out if having a death on the table affects the performance of the surgeon immediately afterwards. I discussed the idea with Tony Goldstone and Chris Callaghan, who at the time were two promising surgical trainees in our department, and they were very keen on the concept. Together with Jon Mackay, an anaesthetist, and Susan Charman, a statistician, we designed a study to look into this question.
The first thing we set out to do was to ask cardiac surgeons and anaesthetists what their actual practices were. We sent questionnaires to all the senior cardiac surgeons and anaesthetists in the country. We first asked them if they had experienced a death on the table. We then asked them whether they had stopped work for a period afterwards. We also asked them, regardless of what they actually did after such a death, if they thought they should stop work for a period after such an event. We also asked them if they thought it would be helpful to be given some guidance as to what they should do after such an event. Finally, we gave them some free space to express their views on the subject.
The second thing we did was to seek some evidence to confirm or refute whether continuing to operate after a death on the table had an adverse effect on the outcomes of the patients who had their operations immediately afterwards (in the following 48 hours). We took all the patients who, within 48 hours after a death on the table, were operated by the same surgeon who had experienced that death, and tried to see how their outcomes compared with patients operated at other, more ‘normal’ times.
The results of both the survey of surgeons and anaesthetists and of the outcome comparison were published in the British Medical Journal (Goldstone 2004). Some of our findings were predictable, but others were a complete surprise.
The survey was a very successful one, in that of just under 500 senior surgeons and anaesthetists who were approached, 76 per cent responded. This is an exceptionally large proportion for an anonymous and optional postal survey, and indicates that the subject was one about which people felt strongly enough to want their opinion heard.
The majority of surgeons (86 per cent) and anaesthetists (95 per cent) had encountered at least one death on the table. Just over half of the surgeons (53 per cent), but only around one in five (22 per cent) of the anaesthetists, actually stopped work immediately afterwards. Both surgeons and anaesthetists cited fatigue, emotion, medico-legal concerns, and the advice of colleagues as reasons to stop. Most of these doctors felt that guidelines would be helpful to them.
We had given the survey participants the opportunity for freehand comment so that they could express their own thoughts on the subject, and one salient topic came up repeatedly. Most of those who said that they wanted guidelines also commented that, in their view, it is important that such guidelines make a crucial distinction between the types of death on the table. To paraphrase the feelings expressed by the majority: if the death on the table was to some extent ‘expected’, in other words a high-risk or emergency death, then it doesn’t matter too much, but, if the death on the table was totally unexpected, such as in low-risk elective surgery, then stopping and taking stock is probably a good idea.
This is what the professionals thought. Now for the evidence: the study aimed to find if those patients operated in the immediate aftermath of a death on the table fared differently from other patients. We identified a total of 81 deaths on the table at Papworth Hospital during the five-year study period. At first glance, that may sound like a lot, but bear in mind that Papworth performs about 2,000 operations per year, so that would be a lot less than 1 per cent chance of dying on the table during the period of the study.
In the 48 hours immediately after a death on the table, 233 patients were operated by the very same surgeons who had just experienced the death. These were the patients we were interested in: were their outcomes affected? When we compared them to virtually identical patients in every respect who were not operated immediately after a death, we found that there was no difference in survival rates. We did, however, find that these 233 patients had more complications and spent longer in hospital than their peers. Clearly, therefore, a death on the table does affect performance, but not enough to endanger the life of the next patient.
So far, all of the above, if not exactly predictable, can be described as not all that surprising. The real surprise comes next.
Our survey of the profession had indicated that a death that was to some extent ‘expected’ (high risk or emergency) should not matter so much, but that a death that was a bolt from the blue (low risk, elective) should be followed with caution and introspection. With the survey finding in mind, we subdivided the 233 patients operated immediately after a death into two groups: those preceded by a high-risk or emergency death (in other words, a somewhat ‘expected’ death) and those preceded by a low-risk or elective death (in other words, an unexpected, bolt-from-the-blue death). Here the result was astounding: patients operated on after a death that was somewhat to be ‘expected’ fared worse, and had a higher mortality than those operated on after an unexpected, low-risk death. This was exactly the opposite of what the professionals thought.
We do not know why this is so, though I suspect that once again it has something to do with the personality of surgeons, which I explore in the next chapter. What we do know, however, is that a death on the table does indeed have an impact on subsequent performance. At the very least, there are longer hospital stays due to complications, even if overall mortality is not affected. And those hoped-for guidelines? We are, of course, still waiting for them.