Transcript
MIC CAVAZZINI: Welcome to Pomegranate Health. A podcast about the culture of medicine. I’m Mic Cavazzini, for the Royal Australasian College of Physicians.
COMMENTATOR: Wonderful bold summer colours on this idyllic Copenhagen evening.
MIC CAVAZZINI: It’s the 12th June 2021, a clear summer evening at Copenhagen’s international football stadium. Denmark is hosting Finland in the opening match of their European Cup campaign.
COMMENTATOR: Each of these two second in their qualifying groups. Denmark unbeaten over their eight group games.
MIC CAVAZZINI: The tournament had been put off for the COVID-19 pandemic, so it’s a big deal to see 14,000 fans packed into a stadium for the spectacle. And Finland is keen for an upset in their first ever appearance at this level. For much of the first half, however, the play is all going Denmark’s way.
They make several shots on goal and behind a couple of these is a chisel-jawed player wearing number 10 on his red and white jersey. Christan Eriksen is widely regarded as one of the best midfielders of his generation, a creative playmaker and goalscorer. At 29 years of age, he’s been awarded Danish Football Player of the Year a record five times.
COMMENTATOR: Eriksen more and more starting to pull the strings in midfield here. Switching it for Delaney. That’s a good ball in!
MIC CAVAZZINI: 43 minutes into today’s match, the Finns are desperately holding off the Danish attack. Deep in the opponent’s half, Christian Eriksen stumbles to receive a throw in and then crashes to the ground after the ball bounces from his shins.
COMMENTATOR: Eriksen trying to get in and around the box. But down goes Christian Eriksen.
MIC CAVAZZINI: Play continues for a few seconds before the referee recognises that Eriksen isn’t picking himself up.
COMMENTATOR: He does seem to have glazed over. And I have to say this looks very unpleasant.
MIC CAVAZZINI: The ref blows the whistle and waves urgently to the sideline, but at first the players aren’t sure if this the usual sort of injury or time stalling. Danish defender, Simon Kjær, runs the length of the pitch, puts Eriksen on his side and make sure the airway is clear. Other players gather round and as the team doctor arrives they make a wall, not to protect the goal this time but their fallen teammate’s privacy.
COMMENTATOR: This is truly awful, Tony. So many of the players are in tears. Football at the moment certainly is furthest from our thoughts.
MIC CAVAZZINI: The doctors feel urgently for a carotid pulse and a minute after the collapse they’re joined by a paramedic team with apparatus. After another 45 seconds one of the team doctors straddles the stricken player, driving his arms down into Eriksen’s sternum as others set up a defibrillator.
COMMENTATOR: Copenhagen has suddenly got a horrible hush.
MIC CAVAZZINI: At a press conference after the match, the doctor will say, “He was gone… but we got him back after one defib.” In the moment, Eriksen’s tearful partner is seen climbing from the stands onto the pitch before he is stretchered off the field awake and waving feebly to the stunned but cheering crowd.
COMMENTATOR: Well, those in the stadium can barely believe what we’ve just witnessed.
MIC CAVAZZINI: Three days later Eriksen posted a photo from his hospital bed and thanked the public for their support. Although the cause of his ventricular fibrillation wasn’t immediately identified, he was fitted with an implantable cardioverter-defibrillator device as a precaution. There were doubts that Eriksen would ever again play professional football and the ICD automatically ruled him out from continuing his contract with his club team in the Italian Serie A league. But 8 months on from that fateful day Eriksen returned to the field for Premier League Club Brentford. And a month after that he scored for Denmark just two minutes after coming on against the Netherlands.
As confronting as Christian Eriksen’s collapse was, it did raise public awareness of sudden cardiac arrest and bystander CPR, and that’s a good thing. But as we’ll hear today people have vastly overoptimistic expectations about the benefit of CPR. Even medics who know better, tend to administer chest compressions to a wide range of patients in hospital who are deteriorating terminally.
In today’s podcast we discuss the systems and the culture that contribute to this over-intervention. And also the semantic confusion that results in some people not receiving other resuscitative measures because Do Not Attempt CPR orders have been misunderstood. My guests were cardiologist Dr Tammy Pegg, intensivist Dr Alex Psirides, palliative care specialist Dr Kate Grundy and special guest Dr Kathryn Mannix, joining us from Northumberland in the UK.
KATRHYN MANNIX: I'm Kathryn Mannix. I'm a retired palliative care physician in the UK and I work to promote better public understanding of ordinary dying. I do that by writing and broadcasting and podcasts largely.
MIC CAVAZZINI: Fabulous and, as you put it in a TED talk where I found this anecdote. “When Christian Eriksen’s heart failed, his legs were still running. That is a cardiac arrest that can be resuscitated from”. Can you unpack that a bit further?
KATRHYN MANNIX: Okay, so here is a young man at the top of his game. He is running. He is breathing. His organs are filled with oxygen-enriched blood being circulated around healthy organs by a healthy heart. When his heart suddenly stops everything else is in full working order. So, if the reason that his heart has stopped is reversible and if there's an adequate bridge to protect those organs, then this man can, and we know did, make a complete recovery even returning to playing international football again.
And that is completely different from the situation we see when somebody is becoming progressively less well as their organs are working less well, as their blood pressure is dropping, as circulation of blood that may no longer be fully oxygenated through their organs is starting to change the way everything works. And there's a progressive system failure, often with medical staff working valiantly against it in a hospital setting. At the end of that system failure, the body no longer working adequately allows things to stop. And the last thing that stops is the heart, it's completely the opposite situation to that young footballer on his football pitch.
MIC CAVAZZINI: And in that out of hospital setting, we know that the survival rates from cardiac arrest, are 5 to 10 percent. For example, there’s a massive European registry study from 2020 that looked at over 25,100 cases of out of hospital arrest where CPR was initiated. A fifth of rhythms were shockable and spontaneous circulation actually returned in a third of patients. Most of those made it to hospital, but only a quarter made it out again. So that’s where you get this overall survival to discharge rate of 8 percent for people who were given any form of chest compressions. Alex Psirides, lots of variables here but can you explain how important CPR is in that 7-12 minute time window to defibrillation?
ALEX PSIRIDES: Sure. Kia ora, my name is Alex. I'm an intensive care doctor in Wellington in New Zealand and I've probably always been more interested in the dying than the living in terms of the things we do in intensive care possibly to try and not prolong that process any more than we have to. In terms of defibrillation, we have a lot of data because we're pretty good at collecting data on cardiac arrest. It's a relatively binary event, your heart stopped or it didn't, although there is some complexity within that. And these tell us essentially within two minutes, there's a general assumption that if you have defibrillation within two minutes of a shockable rhythm, which is an important point to make, then you will have almost certainly a good, if not perfect neurological outcome.
But there's so many other factors that are included in that, whether your arrest was witnessed or unwitnessed. Obviously, in Eriksen's case, it was witnessed by a very large audience around the world. We know that bystander CPR is a good prognostic sign for a good neurological outcome, but that's only a bridge when done well. You know, if you've got a firie, you've got an ambulance officer jumping on your chest, you'll get much better cardiac output. Obviously, that also causes rib fractures, sternal fractures, direct cardiac injury, valve compression, and liver damage. So, the better your cardiac output from CPR, the more chance you have, perversely, of having significant other major injuries, not inconsistent with major trauma you see when people have been hit by cars or fall out of trees.
MIC CAVAZZINI: Just to add to these comments from Dr Psirides. We know that chest compressions deliver 20 to 40 percent of cardiac output depending on how well they’re done. So just how much does bystander CPR help patients as a bridge to defibrillation? You’ll be hearing a lot about the Swedish registry data. In one analysis of these from 2015 researchers excluded non-witnessed cardiac arrests to eliminate those with unknown time delay. From the remaining 30,000 witnessed events, about half underwent bystander CPR before the arrival of paramedics. The 30-day survival rate for these patients was 10.5 percent compared to just 4 percent for those who didn’t receive CPR until the paramedics arrived an average of 11 minutes after arrest.
Another study published only in February this year examined delay to CPR in even more detail using data from 195,000 cardiac arrests witnessed in the USA over a decade. The researchers found, unsurprisingly, that those receiving bystander CPR in the first minute after arrest had the best survival-to-discharge rates and neurological outcomes. Both measures were about 85 percent better than in the paramedic care group only, which for some reason is much lower than the 260 percent difference reported in the previous paper. If bystander CPR was initiated in the eighth minute after arrest, outcomes were only about 25 percent better than the group that didn’t receive it. And when initiated ten minutes after arrest, both survival and neurological outcomes were actually worse. So, as Dr Psirides has already mentioned, CPR can be a valuable bridge to defibrillation, but it’s also a very crude insult.
In fact, the authors of the initial work on CPR in the sixties, James Jude and colleagues at Johns Hopkins, stated that “Not all dying patients should have CPR attempted. Some evaluation should be made before proceeding. The cardiac arrest should be sudden and unexpected. The patient should not be in the terminal stages of a malignant or other chronic disease and there should be some possibility of return to a functional existence”. But as we’ve already foreshadowed, today’s podcast is about the overuse of CPR in deteriorating, hospitalised patients. Here’s Dr Tammy Pegg to outline the scope of this problem. She’s a consultant cardiologist at Nelson Marlborough hospital on the South Island of Aotearoa New Zealand.
TAMMY PEGG: I mean, the origins of CPR really has a very narrow scope. It was developed for essentially iatrogenic complications during the operating theatre because formative general anaesthesia was so dangerous. It was also used a lot in the cardiac catheter lab, which I'm aware of as a cardiologist. And you've got to remember back in the 1950s when these techniques were developed, to take a coronary angiogram, they used to actually arrest the heart with adenosine and take a plain x-ray film. So, in those formative days, CPR was a vital bridge to the reversal of these causes, either with time or with medication/ defibrillation.
The modern hospital environment is a very different place and these iatrogenic complications are actually incredibly rare. So, most of the time when CPR is actually deployed, it's deployed in situations of progressive deterioration, not sudden unexpected cardiac arrest. And if you go out of the coronary care unit or the operating environment and you're on a general ward, only one in ten cardiac arrests—and I'm using inverted commas there because they're probably not actually a cardiac arrest—only one in ten of those are actually with a shockable rhythm.
MIC CAVAZZINI: Thank you. Yeah, that's a good overview. And you and Alex coauthored a study published 2023 in the Internal Medicine Journal to quantify the understanding of the technique. About 500 respondents, mostly from the NZ medical community, were presented with quite detailed clinical vignettes of different patients and asked what response would be appropriate in the event of cardiac arrest. For example, one vignette was about a 74-year-old with new-onset heart failure and syncope and severe aortic stenosis shown on echocardiography with mildly reduced size and function of the left ventricle. The respondents could tick various options from a list, and for this example, 45 percent of respondents did propose CPR, despite it being, as you put it “not effective as a bridge to aortic valve surgery in aortic stenosis.” In all there were five very different clinical vignettes. Alex, I noticed the bias in favour of CPR was found in all but one of the scenarios. Can you unpack a little bit what these results reveal about the clinical reasoning going on in those scenarios.
ALEX PSIRIDES: What we found was interesting because we found that CPR would still be offered by a significant number of people to cases where the literature suggests there's less than a 5 percent survival chance. One of those in particular, one of the scenarios was a patient with severe lung disease who had a hospital-acquired pneumonia, a fractured neck of femur who'd been deemed not for ventilation, but despite all those things, 12 percent of respondents said they would attempt CPR in this case.
I guess the key we'd come down to is CPR as a bridge to what? And if you don't have somewhere to go then why are you causing harm to someone to get them to somewhere that you can't get them out of? And it seems that people didn't really think it through. There is a famous bridge in Whanganui in New Zealand that was built for a settlement that subsequently never happened. And you can visit it and there's nothing on the other side—it’s called the Bridge to Nowhere and it's the most powerful descriptor I can think of with regard to CPR.
TAMMY PEGG: You know, many conditions, particularly in my world, cannot be effectively treated by CPR. We spoke about aortic stenosis, I'm also aware of a person who arrested, as Kathryn said, probably didn't arrest, probably just died during a TAVI valve procedure where the clinical team undertook chest compressions and rescue breaths, but the valve was destroyed and each compression and release was just pushing blood backwards and forwards like a wave through a destroyed valve, not producing any effective circulation. So, these patients shouldn’t be offered cardiopulmonary resuscitation, it’s an ineffective treatment.
MIC CAVAZZINI: And there’s a study that followed over 22,600 cardiac arrests in 144 British hospitals—a study from 2014. Seventeen percent of arrests were shockable, so that’s ventricular fibrillation or pulseless ventricular tachycardia. Survival to hospital discharge associated with these rhythms were 49 percent. Tammy, we sort of touched on this earlier. If there was a defibrillator at every bedside, would chest compressions still be a thing in this hospital setting?
TAMMY PEGG: Well, certainly in the cardiac catheter lab, we hardly ever use them. I can't think of a time when you've managed to do early defibrillation that subsequent chest compressions have improved the cardiac circulation. So I think, you know, already without a bedside defibrillator available, we are achieving time to defibrillation in hospital of around a minute already, and that's from large US data. They trialled having a bedside defibrillator in Ottawa actually, and they found that it improved survival, nurse led AED discharges improved survival, but only in monitored wards. And that's because in non-monitored wards, in our general medical and surgical wards, most cardiac arrest is not shockable. So having a bedside defibrillator will not impact survival at all.
MIC CAVAZZINI: And going on with some of the data from this study, 83 were of arrests were non-shockable so flatlins or Pulseless Electrical Activity. But survival rate for this cohort wasn’t zero, it was 11 percent. So, Alex, if defibrillation isn’t indicated in those patients, is CPR potentially useful as a bridge to some other life-saving intervention?
ALEX PSIRIDES: I think there’s huge heterogeneity here with regard to—you can start with shockable versus non-shockable rhythms. But in your non-shockable rhythm group, you have a whole lot of causes. Traditionally, there's the 4 H's and the 4 T's. There may be obstructive causes, there may be PEs, there might be myocardial thrombi that will be dislodged by physical activity. So, some of those patients will be included in that. And there's a lot more causes I would argue of non-shockable rhythms than there are shockable. It's also worth pointing out, obviously, that ultimately all arrhythmias become asystole eventually. So, you may just find your shockable rhythm, if you don't shock it, becomes asystole because that's what happens when your heart stops. So, your sampling error may be, when did you pick it up and if it's asystole it's because they had their VFRS 20 minutes ago and no one noticed.
MIC CAVAZZINI: Thank you. So, overall survival to discharge in this British audit was 18% for all rhythms, averaging those two cohorts we already described. In a smaller Australian study it was 26%. But there’s a lot of data missing on what kind of shape these patients are in and how much longer they survive. Tammy, in your recent review for the New Zealand Medical Journal you summarised some neurological findings reported from US survivors like this; “despite [80%] of US in-hospital cardiac arrest survivors attaining CPC category 1–2, 40% have life-changing disability after discharge and only half are discharged home. This harm is not readily available in the statistics.” So, can you expand on this data gap? What else do we need to know to make sense of effectiveness on this intervention?
TAMMY PEGG: Well, this is from the Get With The Registry data from the US, which is a national cardiac arrest database. And basically, if you go into the details of it you find that only half of patients are discharged home, half are discharged to hospice or other nursing care facilities. But of those who are discharged home, there's still quite a high burden of morbidity. So, there's quite significant mental health issues. There's quite significant burden of chronic pain. And you know, Swedish registry data also suggests that there's a continued decay in survival after discharge. So, it's not like you're discharged home and you're completely fine. Almost a half are dead within sort of 12 to 24 months after discharge. And so, I think the problem is, is we assume that survival to discharge, life at any cost, is acceptable, but we need to put this in the context of what patients are actually living like further on down the line to understand whether those outcomes are genuinely acceptable in a patient-centred manner.
MIC CAVAZZINI: We know that laypeople vastly overinflate the likelihood of survival from CPR. Who can blame them when shows like Grey’s Anatomy, depict the survival rate at over 70 percent. That series is still running after 20 years and is just one of many popular tv dramas about critical care. But clinicians shouldn’t be prey to such optimism bias. In order to quantitatively gauge their understanding about survival following cardiac arrest, Drs Tammy Pegg and Alex Psirides surveyed 500 healthcare practitioners, prompting them with several different presenting rhythms or comorbidities. The results, published in the Internal Medicine Journal of 2023, actually show that clinicians often underestimate recovery from resuscitation. So, if they know the intervention is not likely to help, what could explain the overuse of CPR for in hospital cardiac arrest? Dr Psirides believes that a major problem is the crudeness of the algorithms that medics are trained to use.
ALEX PSIRIDES: Algorithms treat in hospital and out of hospital cardiac arrest as the same thing. There's no algorithm I've seen that distinguishes between them. And the key point Kathryn raised at the start, which is if you're outside a hospital and you're walking around and you collapse in the supermarket, that's told me that you're not bed-bound, you're independent and you're mobile. If you're in hospital, bed-bound with chronic disease for which your arrest is the end point, you have the same algorithm as applied as if you collapsed in the supermarket.
We spent a bit of time a while ago looking at the teaching that in New Zealand the New Zealand Resuscitation Council provides around not doing CPR and it's essentially a single paragraph in a 200 page document that talks about not doing this because it may be futile as opposed to 199 pages of telling you how to do it. So, there's a huge algorithm bias because resuscitationists want to resuscitate people, they don't want to talk about not resuscitating people. And when it's taught it's taught as this is the default you will receive this. Based on data largely extrapolated from out-of-hospital patients to in hospital patients. So, there's an algorithm bias essentially because it treats them as the same event and they are absolutely not because there's significant differences for all the reasons we've talked about in the patients that suffer arrests outside versus inside hospitals.
MIC CAVAZZINI: Kathryn we’ll bring you back in here. So, your mention of the algorithms as just defaulting to CPR might lead clinicians to practice defensive medicine; “Well, if I don't follow the algorithms, even if I know this patient can't be saved, if I don't follow the algorithm I'll get into trouble.” Do you think that plays into it, Kathryn?
KATRHYN MANNIX: So, I think it's part of a more complicated emotional state in the people who respond to a person's heart stopping in hospital. And it's interesting that we've indiscriminately used the word arrest throughout this conversation, even though heart stopping moments aren't all, I think, cardiac arrests. So, it's partly worry about getting into trouble, but the getting into trouble worry comes from the presumption for CPR if somebody's heart stops in hospital. And I think we need to go back and actually examine that a little bit. Why is there no declaration of “escalate comfort care, don't proceed to CPR” statement for somebody who's got a foreseeable medical emergency that might result in them dying.
Because until we call this state of gradual diminution of health to the point that our person's heart stops beating, dying, we can't differentiate cardiac arrest where the appropriate response is full resuscitative measures that may or may not include bridging chest compressions and rescue breaths. And ordinary dying, where we're anticipating a person's heart is going to stop. And we ought to share our anticipation of that with the person if they want to know, and with the family around them. And that is a conversation about, “I think your dad is so sick that he could be dying. And we're doing everything we possibly can to stop him from dying, or we've done everything he's given us permission to do to stop him from dying. But it's possible that in the next 24, 48 hours, your dad's body won't be strong enough to withstand what's going on and his heart will eventually stop beating.”
And I think we need to take this further. I think we need to be able to say, “when that happens, we're not going to mistake it for one of those TV drama, cardiac arrest events. You know, we're not going to damage his dying by jumping on his chest and sticking tubes down his throat. We want the people who want to be alongside him to be there. We want him to be as comfortable as he can be. We want you all to be aware that this is the current state of play. And whilst we'll continue to do everything we can to offer the best care that's available, when his heart stops, that will be a non-restartable event.”
MIC CAVAZZINI: I'm going to ask you to explore those experiences you've had in your career of those conversations later. But it occurs to me now that this ties into the defensive medicine, that once the understanding among the public and among the clinicians changes, that for all the bells and whistles and lights, camera action, you're not going to save these patients, then the defensive element might be brought down. “If everyone understands that this is a hopeless situation, I can't be accused of not doing enough”. I'm actually doing more by sending them down a more caring pathway.
KATRHYN MANNIX: I think that's true. And I also think that when we're thinking about outcomes, one of the outcomes ought to be that at our morbidity and mortality review meetings, we need to ask, “Did this person pitch up in hospital with a foreseeable medical emergency because of the long-term condition people already knew they had? And was there or was there not an anticipatory care plan about what should be done and what the ceilings of intervention would be? And if there was not an anticipatory care plan, then that should be a ‘never event’. Somebody somewhere didn't have a conversation that led us, who didn't know this person so well, to have to make on-the-hoof decisions that were not the best match for what was going to be good whole person care for this person. And we can't hold the emergency department responsible for the geriatricians, the oncologists, the respiratory physicians, et cetera, et cetera, not having had the appropriate conversations in a timely manner in the first place.
MIC CAVAZZINI: Alex, another idea that you alluded in one of your presentations to is that of a conveyer belt. Sort of a systems problem that everyone just takes on and escalates care without really thinking about the long game. Is that right?
ALEX PSIRIDES: But that's what we're trained to do. That's what paramedics are trained to do, that's what EDs are trained to do, that's what ward doctors are trained to do, and that's what intensivists are trained to do. So, we do what we're trained to do. We're not trained not to do that. And we're never trained not to ask, should we be doing that? The problem with the situation is, course, that everyone blames the person before them. The intensivist goes, why did ED intubate this patient? ED goes, well, why did the paramedic bring them in? And the paramedic has to because they got called by someone who was desperate, who didn't know what else to do. So, we're all weighting the system poorly because the conveyor belt of critical care carries you to an intensive care unit where two weeks later we have that heartsink conversation of family members, at least not infrequently, saying those awful words “No one wanted any of this.” And I hate it when that happens. And you know that when families are telling you that they're doing it with a sense of gratefulness about what you've done but actually they just got brave enough to tell you now after X days of intensive care that this isn't something they wanted.
There's a quote that has been attributed to me unfairly that came from a palliative care physician in Queensland called Dr Will Cairns, which is, “Two weeks in ICU can save you one hour of difficult conversation”. That increasingly is true. It's a difficult conversation, possibly because it's delegated to the most junior doctor in the hospital, not by the senior experts present here. And arguably, my argument has always been it's every one's business to have that conversation. And there's a variety of tools available; the Serious Illness Conversation Guide that’s come from the US and Tammy has been implementing across New Zealand has been very helpful, hopefully, in training people to know how to have these conversations, which essentially is taking many, many, many years of palliative care wisdom and trying to distil it down for non-palliative care people.
MIC CAVAZZINI: Going back to the algorithms. We do know something about survival from different causes of cardiac arrest. Tammy has already mentioned that for highly reversible causes like general anaesthesia, intoxication and hypothermia, survival rates after resus are around 50 percent. For cardiac patients one-year survival for cardiac patients is 39 percent while it’s closer to 11 percent for non-cardiac patients. Meanwhile for, cardiac arrest after sepsis and ICU care survival is very low. I've come across some literature machine learning algorithms that pull in hundreds of different variables that might be able to predict which patients will benefit and which ones won't much better. Could we imagine this selection of patients being put into practice in the hospital environment or there just isn't really time to process all that? Let’s bring Kate Grundy in now.
KATE GRUNDY: Hi, I'm Kate Grundy. I'm a palliative care physician based in Christchurch Hospital. I've always worked in a consult liaison hospital service and therefore have kind of a long experience of looking after seriously unwell people, many of whom actually do end up dying in the acute environment. I don't think any algorithm is going to take the place of having honest, open conversations with patients who—humans are not machines, and they're not very predictable as far as how things are going to go and we tend to—I mean, Alex alluded to this, you know, doctors and nurses are part our people within society and we have our own fears and our biases and we also are often just as fearful of dying and of facing dying as our patients are. So, we tend to fall back onto the comfortable territory of treatment options that we have available and CPR is one of those treatment options.
So, we're very comfortable in talking about ways in which we can try and avoid dying, without actually mentioning the word of course, but actually talking about the treatments that are available without recognising overtly and opening up the opportunity for discussion with patients about what would happen if those treatments didn't work. And what are your—you know, “it's really important for us to know that even though we're trying really hard to get you through this pneumonia or get you through this latest recurrence of your cancer, that if we're not successful, how can we still look after you well”. And my experience, and I know Kathryn’s experience is exactly the same, is when you have those types of conversations, people don't want CPR. They want not to be given up on. They want you to involve them in the decisions. They want you to give them time to think through and to become accustomed to an idea that things are changing and to recognise what's happening in their bodies that they are becoming weaker, frailer, that they're not responding as well as they did.
I have these conversations regularly and I barely ever have a, “why are you still wanting CPR conversation?” It's a very rare patient that actually still wants CPR if nothing else is working. So, I think we focus so much on this one thing that we're not going to do and miss out on the opportunities of all the wonderful things that we can still offer people as their health deteriorates. And that's what we do in palliative care, but that's where the Serious Illness Conversation Guide comes, because it gives us some really lovely, open, patient-facing, patient-tested language to use that isn't scary for people.
MIC CAVAZZINI: I think, in your papers you cited some studies suggesting that only a third of patients with advanced cancer had discussions about resuscitation at end of life with their treating clinicians, but those papers were about 20 years old. Maybe I'll put this one to Kathryn, has anything changed in that 20 years? Other physicians won't have had the same training that you and Kate have had. Are general physicians and intensivists more prepared to have those conversations today?
KATHRYN MANNIX: So, we need to talk about that, don't we? Because actually, it's too late by the time the palliative care team is having this conversation. This is an upstream conversation as part of the progressive illness dialogue with the cardiologist, respiratory physician, who knows this person, and possibly also their general practitioner. So, we've got a drive, across high income countries, for better advanced care planning, it's called different things in different places. And we know that it's not happening in more than between 10 and 25 percent of people, so if anything we're getting worse. That might be because we stopped counting only cancer and we've started to realize that these conversations are much more applicable to many more diagnoses and at the same time as that, treatments for cancer have opened up and become very much more effective. And so, there is an increase in reluctance again of having those conversations.
MIC CAVAZZINI: That's interesting. And from the review that Tammy and Alex and Kate, who's also an author, there's a quote which I found interesting. “Patient selections are often confused and paradoxical…. For example, respondents to a treatment preference survey indicated that 74% wished to receive chest compressions, whereas only 61% selected defibrillation and 42% ventilation…. Simply asking the patient, “if they would like their ‘heart restarted’….undersells the complexity of treatment and does not represent shared decision-making”. Tammy, you make the point that these conversations, you know, if you are trying to dissuade someone from the trauma of CPR, you can’t just scare the living daylights out of them, or coerce them to what you want. Can you point to the shared goals of care process? I don't think we have that in Australia.
TAMMY PEGG: Yeah, so in New Zealand we moved away from a binary DNA CPR process to a graded process where you can have curative or restorative treatment inclusive of CPR, remembering that CPR is just chest compressions and rescue breaths in Australia and New Zealand. You can have curative or restorative treatment exclusive of CPR and those would be things for conditions where the circulation is not able to be supported by these simple measures. Or you could have treatment that's aimed at easing symptoms, still might end up aiming to get the person home, but these treatments are less burdensome. And then you've got level D, which is care of the dying patient. And that offers the clinician a sort of a wider range of options to consider.
We need to come back and move away from informed decision making where we just give a patient information about what treatments are an option and start to understand what's important to a patient, what are their values, goals and preferences and make sure we match available treatments to best deliver those goals. And if we do that, then we don't end up getting in this sort of toing and froing situation where we explain a treatment and we use manipulative language or scare tactics to dissuade somebody from selecting it.
MIC CAVAZZIN: Alex in your rockstar appearance at the SMACC conference; you noted how the squeamishness around talking about death, all the euphemisms lead to further confusion and distress. Can you describe for us the experience you had right after you moved to New Zealand. And found that perhaps the kiwis are less buttoned up than the English.
ALEX PSIRIDES: Am I allowed to swear? Are you going to bleep this out? I mean, I shall use the language that was used. So, I guess probably for some context, I moved to Wellington in 2002 and had come from large London teaching hospitals where no one really said what they thought, everyone said what the politics required them to do. And relatively early on, went into a meeting with a family in the intensive care unit that was led by one of my consultant colleagues. At that time, I was a relatively junior ICU trainee.
And he began a very well conducted family meeting talking about how things were going badly. And the family who were there didn't really understand what he was saying. So, he tried another way to do this gently and he used the D word, as you always should in intensive care; “things were likely to end in death” and it was clear that the family also didn't really understand what was being said to them. So, he just stopped and said, look, “They're fucked. They're not going to wake up. Their brain is fucked and there's nothing we can do.”
And I think my jaw audibly dropped at that point having come from a hospital in London where this would never have happened. And I was deeply unsure about what was going to happen next. And this family stood up, shook his hand, said, “Thank you for telling us something in a way we can understand it”, went back to the bed space and we extubated them and let them die. And it was a glorious example to me of the value of clear communication and a shared mental model and that was what got the message across the line and resulted in stopping what was essentially prolonging this poor person's death.
Everywhere else I'd worked, including my year in Melbourne, in a very large quarternary ICU, seems to favour quantity of life over quality. And the reason I came back and stayed in New Zealand is that there is clearly a favour of quality over quantity, which means that quantity isn't sought at all costs. And I find that refreshingly pragmatic as an intensivist who has the ability to cause innumerable amounts of harm to people by dragging them through things to get them in a state they never wanted to be in. I wouldn't practice intensive care anywhere else in the world because of that.
MIC CAVAZZINI: So far we’ve described how “cardiac arrest” is a catch all outcome for some vastly different pathophysiologies. Maybe some of the confusion about who should be resuscitated could be avoided if the term were used with the specificity that was originally intended. But even when advanced conversations about end of life treatments do take place there exists another semantic trap. Let’s say the patient notes are marked DNR for Do Not Resuscitate or NFR, Not For Resuscitation. Does that instruct you to withhold any form of resuscitation or just cardiopulmonary resus?
Or if the more specific DNA-CPR is used, Do Not Attempt CPR, that doesn’t always clear things up because of a semantic creep in the use of that term. Many people take CPR to mean all and any interventions for resuscitation, something that was demonstrated in that IMJ paper coauthored by Dr Psirides and Dr Pegg, along with Niamh Berry-Kilgour and Dr Jono Paulin.
The 500 survey respondents were probed with this tick-a-box question, “Cardiopulmonary resuscitation for in-hospital cardiac arrest includes; Chest compressions, Rescue breaths, Defibrillation, Adrenaline and/or Atropine for bradycardia”. 97 percent of responsdents included defibrillation as part of the CPR package and 57 percent ticked all of the above treatments. To the more direct question “Defibrillation is part of cardiopulmonary resuscitation,” 91 percent of respondents agreed or strongly agreed.
That’s to say, CPR is often used as a shorthand for any form of resuscitation rather than just one possible example in the toolbox. The consequence of this, identified by Tammy Pegg and coauthors, is that DNACPR orders then risk “being perceived as an unofficial stop sign to other evidence-based interventions.” I asked Dr Pegg for an example from her experience.
TAMMY PEGG: The case that highlighted this for me was a nurse brought a case to my attention about two years ago of a person who was in an aged residential care facility and she'd been asked to do a case review. And they had choked on an egg sandwich and because there was DNA CPR nobody sought to remove the egg sandwich from their mouth. Which was just a terrible devastating way to die, irrespective of whether they needed CPR or not. It showed, you know, a lack of kindness, but also a real problem around what cardiopulmonary resuscitation is and what it's trying to achieve. And it's ironic Alex, isn't it, that this CPR has such a narrow scope, such a very specific set of indications, and yet it's broadened out and now when we say people are not for this very specific treatment for a very set of circumstances, their overall care is embarrassed through a whole range of implicit biases.
ALEX PSIRIDES: And there doesn't really seem to be concordance with the international societies themselves that publish guidelines on what CPR means. Is it chest compressions plus rescue breaths? Is it chest compressions plus rescue breaths plus adrenaline? Is it that plus defibrillation? Does that mean fluids? Does that mean antibiotics? So, there is a justifiable concern that if we say someone is not for resuscitation, then that also excludes them from things that may be beneficial, and in some cases may exclude them, for example, from receiving analgesia at end of life to lessen their symptoms. And that is a real concern, it's described in the literature as this halo effect of “Not For” extending to things that actually may be beneficial.
It's also a failure to recognize that people who may not be for CPR may actually be for a whole lot of things and may leave hospital alive. And in many cases may leave hospital alive, go home back to productive lives that they enjoy. So, it is this odd decision that we're asking people to make because by default in hospital it's the only treatment we provide to everyone generally unless you opt out. Every other treatment in hospital is opt-in. CPR for reasons I fail to understand is an opt-out that we ask patients to make and clinicians as you say to take a tick box approach to well, “we'll do a bit of that and a bit of this but we won't do some of that”.
MIC CAVAZZINI: And there’s a paper cited in your review suggesting that in more than half of advanced care directives there's an incongruence between what the patient and clinician have understood.
KATE GRUNDY: That is the reason why the shared goal of care model is so much better than the binary Not for CPR or for CPR. But it's contingent on us learning and being prepared to have the conversation. Because in my experience, what it does is it opens up opportunities for people to understand more about what their future health is going to look like, rather than the magical thinking that everything is always going to be fixed. So, I see it not as something that should only be done in those people who we have high risk, high expectations that they're going to arrest or going to die or going to deteriorate, but actually make it more a routine part of our conversation so that we can—people can understand what choices they genuinely have and if we understood more about what people wanted and what their goals and wishes and priorities are then we can tailor our treatment to them in a much better way.
MIC CAVAZZINI: That's excellent. I like all these sorts of practical systems recommendations. From your study, the trend, the misunderstandings in the terms CPR was unrelated to the respondent’s seniority or their particular professional roles. Maybe we can infer from this that the devolution in the semantics has been established for quite some time now. So much so that in some of the literature from the UK, including the National Cardiac Arrest Audit, they actually say, “CPR is defined as the receipt of chest compressions and/or defibrillation.” But as you write in your paper, “CPR is just a subset of resuscitation and needs recapturing as only chest compressions and rescue breaths intended solely to maintain organ perfusion.” Alex, why insist on reclaiming this, admittedly, rather vague term rather than coming up with a more precise one. Or perhaps even going back to the original CCCM for “closed chest cardiac massage” which is very specific?
ALEX PSIRIDES: I guess the simple answer is that we should all be talking about the same thing. Here we are talking about a technique that all of us at some point in our career may need to apply and we can't agree on what it is. That just seems like basic definition failure to me. And if we can't agree within ourselves, how the hell can we expect patients to understand it? So, there needs to be to clarity, there needs to be consistency and we need to be removing confusion around what resuscitation actually is.
I'd go out on a limb and possibly suggest that close chest CPR was way ahead of its time that hadn't thought through the ethical implications. Prior to 1960 when closed chest massage was described, if you wished to do cardiac massage, you took a scalpel, you made an incision literally into the side of the chest through a thoracotomy, and you did internal cardiac massage. Closed chest CPR made it, dare I say, too easy, that meant that therefore it was open to everyone, as the paper says, with a pair of hands. What they possibly hadn't thought through was the cognition behind whether they should be doing that or not.
MIC CAVAZZINI: Almost at the end, and I think we'll let Kathryn close just as she opened. So yeah, this conversation isn't about whether you'd like to be resuscitated is much bigger than that. It's not just something to be considered on your way into hospital or critical care, but it's about philosophy and culture and community. And Kathryn, you've written how for most of human history, most people died at home right up to the 1950s. What's changed in the period since then and what's the situation we find ourselves in now?
KATHRYN MANNIX: So what’s changed in high-income countries—and we need to remember that this is a this is a high-income privileged conversation in the first place—is that there have been such technical advances in medicine in that where once we sat around the bed of a person who was dying very often of an effective illness, we now can take that person to hospital and there are better anaesthetics that allow longer surgery, there are new agents to combat cancers, there are innovations that support failing organs or even replace them with a machine or transplant somebody else's organ in, there are so many more possibilities over the last 80 or so years in medicine that the way people respond to being sick enough to die, has been completely transformed.
And so that familiarity with watching friends and relatives dying decades earlier than people die now was something that people were very familiar with. My grandmother at the age of 23 had looked after many, many dying people at home in the early 1920s. I qualified from medicine at age 23 having been told and taught nothing at all about the process of dying. So, we're trying to talk to people whose only view of dying now is the vacuum of personal experience that's been filled by Hollywood deaths, soap opera deaths, terrifying newspaper articles. And there isn't an understanding that just like giving birth, dying is a process, it's got phases and stages. They get confounded sometimes by the addition of medical treatments, but if they don't, then human dying, like any other animal dying, is a recognisable process.
It usually involves a period of deep unconsciousness during which there will be reflex breathing cycles that the public have not previously seen. So, they misinterpret agonal breathing as sighing, as groaning, as distress. They misinterpret breathing through a film of saliva that you're not clearing from the back of your throat because you're so deeply unconscious that you're not triggering the glossopharyngeal reflex anymore as choking or drowning and they become traumatized by the experience of watching somebody actually deeply unconscious undergoing ordinary dying.
And we don't in fact talk to families about the fact that the heart of a person who has taken their last breath and they haven't breathed for a minute or more, almost certainly their heart is still doing something in there. But we take the time of death as the last breath, not the last beat of their heart. We've uncoupled it. Medicine has gained so many things but it's lost its own understanding and familiarity with ordinary dying. It's lost the midwifing of dying where we explain to people what to expect in advance so that the companions around the bed don't awfulise it, don't misinterpret it and take a better understanding of ordinary dying into their bereavement with them. And that is a really important service that we need to reclaim. Not because we're medical, but because we're human.
MIC CAVAZZINI: That was Dr Kathryn Mannix ending this episode of Pomegranate Health. At her own website you can find links to her bestselling books, With the End in Mind and Listen. You also heard from Dr Tammy Pegg, Dr Alex Psirides and Dr Kate Grundy who were all incredibly generous with their time and expertise. The views expressed are their own and don’t necessarily represent those of the Royal Australasian College of Physicians or the institutions they’re employed by. These podcasts really are just meant to be a conversation starter, not the final word. But you can find links to the all the research we discussed in the show notes at our website, racp.edu.au/podcast then click on episode 131.
At that web page you’ll also find thankyous to the RACP members who provided feedback on early drafts of this story. I hope you liked the final product- please share it around or vent your frustration by sending an email to podcast@racp.edu.au. One helpful new feature to the recordings is that they now have embedded chapter headings, that most podcast players will allow you to skip to. The website has also been tidied up to make it easier to navigate to subscription links and older episodes. This podcast was produced on the lands of the Gadigal clans of the Yura nation. I pay respect to their healers past and present. I’m Mic Cavazzini, thank you for listening.