Risky business. 20.06.20
Welcome to the Plague Pit – issue number 33.
In the middle of the COVID-19 pandemic, perhaps it’s odd to talk of ‘disasters coming in threes’. But this week has felt a bit like that.
First – I went to a safety meeting about a young patient who had an unexpected catastrophic outcome after routine minor surgery. I was involved in the patient’s care.
Then, yesterday, a colleague told me about his own unsuccessful attempts to resuscitate a hospital patient in cardiac arrest a few years ago – ‘I just happened to be in the corridor outside’. The patient was having a procedure with a low but well-recognised risk of death. Resuscitation efforts included the most sophisticated possible techniques and went on ten times longer than usual. The hospital lawyers have just got involved.
I’m still waiting for number three…….
(Not) everyone’s a winner
Surgeons and anaesthetists don’t expect every one of their patients to make an immediate, full recovery after their operation. Minor complications are very common. What happened to my patient is, mercifully, extremely rare. Overall death rates from surgery are higher than you might think, though. In a large European study of 46,000 patients having non-cardiac surgery, 4% of patients died within 60 days of their surgery. (1)
This study is a good example of an observational study. The authors did not set out to compare one treatment with another, as in a randomised controlled trial. Instead, they recorded the outcome from surgery (dead/alive) in a large number of patients. For each patient, they also recorded data that they thought might be linked to risk of postoperative death.
It’s a prospective cohort study, which is good. That means the authors decided in advance what data they wanted, which patients to get it from and when, and how they would analyse it. Only then did they set about collecting it from patients. In the alternative – a retrospective cohort study –investigators look at stuff they already have in the cupboard and see what theory it might support (said one cynic). Bias is a major problem.
The authors of the Lancet study used multivariate regression (MVR) to establish which factors were statistically associated with a higher risk of death after surgery. Major surgery and emergency operations are bad. If you’re old, alcoholic or have cancer that has already spread – that’s bad too. And if you live in Ireland, Latvia, Poland or Romania, you might think about having your surgery elsewhere…..
.
MVR fascinates me. It involves thinking about multidimensional states, i.e. more than three at a time. Absolutely terrifying if you are not a mathematician, and I’m not. I had to learn about MVR for my doctoral research – I wanted to establish whether the dose of propofol that a patient needs to stop them talking is related to their age, ethnicity, weight, sex, blood pressure, heart rate etc. [The answer is ‘yes’ to all]. (2)
Back then, I really enjoyed ‘Multivariate statistics : a practical approach’ by Bernhard Flury and Hans Riedwyl (London, Chapman and Hall, 1988). I was desperate and just picked it at random off the library shelf. It’s short on maths formulae. Hoorah. Instead uses a great practical example to explain the concepts – namely genuine and counterfeit dollar bills. It still sticks in my mind ten years later. Anxious not to recommend a dud to you, though, I checked out the reviews today.
The first one I came across was a real stinker. The second, too (I assume) – one star, in Japanese. But I’m pleased to say that it got the thumbs up in 1988 from ‘Choice : the publication of the Association of College and Research Libraries’ – ‘…..a very readable book on multivariate statistics’. Not a phrase you hear very often……..
Public health warnings
When things go wrong during surgery, patients or relatives are understandably upset – especially if they feel they were not warned of the risks beforehand.
Five years ago, a major decision by the Supreme Court – Montgomery vs Lanarkshire [2015] – completely changed the legal obligation doctors have in respect of informed consent for surgery. The basic principle remains, in that patient and surgeon both sign the same form before the operation. This states the procedure they have agreed – and the risks and potential complications they have discussed. (Occasionally patients needing surgery aren’t in a condition to give informed consent, and then a different process applies.)
Before Montgomery, doctors explained risks to patients on the basis of Sidaway v Board of Governors of the Bethlem Royal Hospital [1985]. Essentially, the principle was that the doctor’s decision regarding what information to give to a patient during informed consent should be one that would be supported by a responsible body of clinicians.
In 2015, that changed. As one writer has put it, a patient should now be told ‘whatever they want to know, not what the doctor thinks they should be told’. On the face of it, that sounds entirely reasonable. But it’s not as straightforward as it sounds. How can a patient know that they want to know something, if they don’t know what it is to start with?
The ruling in Montgomery gives some guidance on this. Doctors should explain a risk of treatment if “a reasonable person in the patient’s position would be likely to attach significance to the risk, or the doctor is or should reasonably be aware that the particular patient would be likely to attach significance to it.”
Death during surgery is a risk most people would regard as significant. And for anaesthetists working in UK hospitals today, that’s awkward. Many patients are now admitted only an hour or two before surgery, even for major operations. They’ll have been assessed by a specialist nurse beforehand but they won’t meet their anaesthetist until the last moment.
Personally, I find it hard to tell an anxious teenager I’ve just met – scheduled for minor surgery – that they might die in the next few hours.
.
Before Montgomery, things were different. Back then, I suppose I always assumed that every patient (1) knew there was a risk (2) knew it was a small one if they were healthy (3) would rather I didn’t go on about it too much. All rather patriarchal. Even then, I had a sneaking suspicion that these assumptions were more for my own comfort and convenience than the patient’s. And I knew that patients can, and still do, struggle with concepts of risk
As a trainee, I recall a tricky preoperative conversation with a sick elderly patient and his relatives. We talked about how the operation was the only effective treatment and about the things that might (very easily) go wrong. Over half an hour we covered strokes, heart attacks, postoperative pneumonia, intensive care admission – and death. The patient’s daughter seemed indignant with each new complication I brought up. At the end, she asked forcefully: ’Yes – but you wouldn’t be doing anything if it was dangerous, would you?’
Thankfully, Montgomery has prompted anaesthetists to consider different ways of informing patients about the risks of anaesthesia. The Royal College of Anaesthetists has an excellent collection of leaflets about everything from feeling sick to waking up blind.
They also have this nice poster.
.
Going home, done my time
I’m delighted to say I’ve just started working back at my old hospital. I wrote a bit in issue 7 about my time away during the pandemic – anaesthetising NHS cancer patients in a private facility. Back in the NHS this week, everything looks as reassuringly tired and shabby as before. Staff and decor alike.
But it’s home.
One utter joy has been working with my regular surgeons again. Mr Y was as unkempt as I am – his hair ‘too complicated’ to get a trim from his wife. During the pandemic, his young family have been self- isolating in the country while he is alone in London, working. ‘I got a bit fitter and read lots of philosophy – mostly the Stoics.’
As a hand surgeon, he spends a lot of time repairing injuries. Together on Fridays, he and I used to tackle an endless of list of:
(a) ‘Fight clubbers’. Fractured fifth metacarpal, right hand. “So I lost it – and punched a wall, Doc.”
(b) DIY incompetents. Mostly ‘Stanley warriors’. Only occasionally, thank goodness, the angle grinder amputees.
(c) Hungover gourmets. Avocado preparation injures from the professional classes. A more prosaic array of cleaver wounds from the takeaway chefs.
(d) Dodgy dog owners. Staffordshire terriers especially. ‘Had him since he was a pup. Can’t understand it. Never hurt a fly’. (Said while bleeding profusely from partly-detached fingers).
Credit: Wellcome Collection. Attribution 4.0 International (CC BY 4.0)
.
Surgeon Mr Y was telling me that, while we were apart during lockdown, he has seen some changes. Every now and then, before COVID-19, we used to put together a patient with multiple severed tendons, nerves, and arteries, deep to a transverse wrist wound. The serious self-harmer.
Since March, it’s been one a week. In fact, that’s what we spent yesterday morning doing – for more than two hours. It’s a graphic illustration that mental health is as big an issue in the pandemic as COVID lung.
The Good Soldier
[1] Pearse RM et al. Mortality after surgery in Europe: a 7-day cohort study. Lancet 2012; 380: 1059–65. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(12)61148-9/fulltext
[2] Morley AP et al. The effect of pre‐operative anxiety on induction of anaesthesia with propofol. Anaesthesia 2008;63(5):467-473. https://onlinelibrary.wiley.com/doi/full/10.1111/j.1365-2044.2007.05402.x