Bioethics: Comfort care vs. intensive care for extremely premature infants

[wiggin]

I recently ran across this study in NEJM talking about outcomes for extremely premature infants. Some background: generally in the US hospitals will use intensive interventions to resuscitate and treat premature infants born at around 24 or 25 weeks gestation. Infants younger than 22 weeks of gestation almost all die. In between, however, are a few weeks of extremely premature infants whose prognosis is not very good. A common approach (for about 3/4 of these babies) is to use 'comfort care', which amounts to drying off the infant, wrapping them up, and putting them in their parents' arms until they die (incidentally, I find this one of the most depressing images I've ever thought about). For the remaining 1/4 of babies in this 22-23 week range, intensive care is attempted, with survival rates of around 25% (and survival rates without substantial medical issues of around 15%). This works out to around 5% and 3% of babies born at this age.

The interesting thing about the article I cited above is that it appears that these numbers mask substantial differences between hospitals - some hospitals routinely intervene aggressively, with higher subsequent survival rates (albeit at substantial cost and, admittedly, still a majority of infants dying), while other hospitals almost exclusively choose 'comfort care'. There are also substantial between-country variations - I have heard some physician colleagues of mine speculate that part of the US' surprisingly high infant mortality rates (for a developed country) is due to a more aggressive approach to treating extremely premature infants - generally, they're counted as stillborn (and thus not an infant death) if comfort care is chosen over intensive care.

There are all sorts of interesting angles to look at this, but I'm interested in the ethical one. Say you are either a physician or a parent put in this situation. You do not have time to adequately evaluate the infant's chances before a decision must be made about provision of care. For that matter, there's precious little good data out there that would actually quantify a child's chances anyways - visual assessments are routinely wrong, and there is little else to go on until they start responding to treatment. To make matters worse, gestational age is generally an inexact science based on iffy ultrasound measurements, so a borderline 22 week infant might actually be a more robust 24 week infant. Lastly, there is the obvious risk of substantial disabilities and later health problems due to prematurity (currently, almost half of babies who survive from this early age). If an infant stabilizes but is likely to have absolutely debilitating ailments for the rest of his or her life, parents and physicians must decide whether to discontinue life-sustaining care.


So, a question for all of you: In the absence of a currently well-articulated bioethical rubric for these situations, how do you think parents and physicians should behave? Given no data on likely outcomes and no recommendations from expert bodies on the issue (current recommendations are for a 'consultative process' which more or less means that they're leaving it up to the doctors and parents), how do you articulate an ethical position? Doctors are in a bind - torn between torturing dying infants unnecessarily (at significant emotional costs to parents and doctors, and substantial monetary costs to society) - with a ~10% chance that they won't die but will end up horribly disabled - and withholding care from babies who have a 15% chance (currently) of growing up more or less normal. Parents are rarely in a position to make informed decisions at this juncture, and doctors just don't know what to do. So how do you decide?

This is a very hot topic among neonatologists - I have a friend in the field who recently left clinical practice because she felt doctors were being too aggressive in treatment options. Others feel that the majority of hospitals that use almost exclusively comfort care for these borderline cases are making wildly unethical decisions.

I don't really know how to address this - what do you think?

[RandBlade]

Its a very difficult dilemma that first off I'm glad never to have been in and wouldn't want to be in. So caveat that nothing I'm thinking below implies I think others who disagree are wrong or unethical, and ask me on a different day and I might give a different answer. Its a very tough question that I feel sorry for any parents who find themselves in this situation.

Getting off the fence though, as I think right now my personal view is that if for example a disabled adult patient got cancer and there was a treatment that offered a 25% chance of survival, would it be taken or would they be given comfort care? I'd think a treatment would be attempted. So why not take the chance to help a premature baby survive? We're not talking about the equivalent of a brain-dead or inoperable patient.

[EDIT: Or perhaps a more appropriate analogy is if there was a car crash and the occupants might likely not survive (and even if they do might be left disabled) would we try and save the patients? Again, I'd think so.]

[Timbuk2]

A complex issue.

I'd like to see money poured into research into care for those born very prematurely. An artificial surrogate womb of sorts seems to be needed, for continued development of the child through 'normal' prenatal means.

Oh were I a rich philanthropist.

[Aimless]

It's a little funny that you should ask this question now. Just a couple of days ago, an intense debate arose in Sweden concerning the great regional variability in the care of extremely preterm infants, specifically wrt whether or not hospitals will even try to save them. I'm not sure if I brought up this matter (focusing on NICUs in the US) in one of the discussions on healthcare we had a few years ago.

This question, like many others in the field, is fraught with problems that arise from insufficient knowledge. I haven't updated myself on this topic recently so my views should be taken with a pinch of salt, but my impression is that we know more today than we did only a few years ago. A colleague of mine has a baby who was born GW 23 and a more distant acquaintance has a baby who was born GW 22 so I've gotten some informal updates via them and those of my acquaintances who work in the field. My hospital and my dad's hospital are the two centres who are the absolute best in Sweden at treating extremely preterm infants and are also the only ones who offer intensive care to babies born at GW 22 by default (barring obvious problems). Studies based on the Swedish registry of extremely pre-term infants show one thing that is probably true of every single country in the world, the US in particular (as is demonstrated by the NEJM article and others before it): there is a great deal of regional variability when it comes to outcomes and that's not only due to the problem of self-fulfilling prophecies (ie. you don't initiate intensive care in time because you believe it will be futile, thus worsening the prognosis considerably).

The consensus here and in many other places is that, if you're going to treat extremely preterm infants you need to at least have a tertiary (or quaternary ) care center that gets a high volume of patients in this category and is manned by an experienced and knowledgable team with well-trained experts in every aspect of care--whether medical or nursing or whatever--starting with the first seconds or with the transport to the hospital. Overall one-year survival for kids born w 22 in Sweden was ca 10% a few years ago, but considerably higher in the two hospitals that specialise in their care and offer that care right from the start (even better today). Modifiable/preventable complications such as malnutrition (or inappropriate nutrition) and sepsis were significant contributors to mortality and morbidity so we should be able to expect better outcomes even with current methods. Anecdotally, those extremely preterm infants who survive past the first year have "long-term" outcomes similar to those born a couple of weeks later when it comes to disability, although obv rates of moderate disability are fairly high. The Swedish data at least bear this out and I expect the US to be similar despite our many socioeconomic and cultural differences.

Sometimes we can say with a reasonable degree of certainty that an infant will definitely die in the very near future. But, on the whole, we have very limited prognostic tools at our disposal in this group of patients. My personal belief is that, when we can't tell those who will die no matter what we do from those who may survive if we do everything right from the get-go, our decisions should at first be biased in favour of giving the best possible care at least long enough to get a better idea of the likely prognosis. It is, in my view, the most just approach when there is ambiguity. This is esp. true for younger patients. We often discriminate unfairly against the elderly, but the truth is that, even under ideal circumstances, a very old person with a severe (esp. acute) illness will likely die in the very near future, so intensive care other than palliation may only give them a shot at a little more life, spent in a hospital, at significant cost in terms of suffering. With the young, we're talking about getting a shot at having a long and full life, even if it may be with some disability. The risk of disability itself is not something that is sufficient to resolve the ethical problems, in my view, although I understand its relevance. Currently there is some ethical leeway in that there is a cultural acceptance for not trying to save every extremely preterm infant, partly due to extremely poor survival coupled with extremely high rates of painful neonatal morbidity. As that changes, and as our culture adapts, I think the ethical aspects of this problem may become a little simpler--at least when it comes to the first year--but I think that may instead make things even harder for the parents of these infants. As if it wasn't hard enough already

[wiggin]

Its a very difficult dilemma that first off I'm glad never to have been in and wouldn't want to be in. So caveat that nothing I'm thinking below implies I think others who disagree are wrong or unethical, and ask me on a different day and I might give a different answer. Its a very tough question that I feel sorry for any parents who find themselves in this situation.

Getting off the fence though, as I think right now my personal view is that if for example a disabled adult patient got cancer and there was a treatment that offered a 25% chance of survival, would it be taken or would they be given comfort care? I'd think a treatment would be attempted. So why not take the chance to help a premature baby survive? We're not talking about the equivalent of a brain-dead or inoperable patient.

[EDIT: Or perhaps a more appropriate analogy is if there was a car crash and the occupants might likely not survive (and even if they do might be left disabled) would we try and save the patients? Again, I'd think so.]

I'm not sure cancer is a fair comparison. There, the patient has a say, and I think their wishes are generally paramount, within reason. There is also a lot more time to evaluate the prognosis of that particular patient and have a detailed discussion. In the case of an extremely preterm infant, we're talking about making a decision very rapidly - often within seconds - when there is little to no prognostic indicators (and precious little data)... and the baby can't tell you how they value the potential tradeoffs. Even for a car crash victim we have some more detailed understanding of their prognosis from a vast literature on types of trauma and the likelihood of death of severe disability. And to an extent, physicians are sometimes uncomfortable extending truly heroic lifesaving measures to trauma patients who have such a bad prognosis - I had some friends who rotated through Shock Trauma, and even though they are by definition inclined to serious efforts to save the lives of their patients, there is a point when even they question the value of aggressive intervention. There's some interesting discussion in bioethical literature about triage decisions that also relates to this.

I do appreciate your thoughts, though. In general I'm inclined towards attempting a medical intervention, but the distress to the baby and likely death/disability is a pretty high cost to consider given the relatively meager chance of reward. I come from a cultural tradition that discounts the cost of future disability in favor of lifesaving measures in general (though certainly there is a value placed on quality of life), so I suspect I would try to intervene and would argue against withdrawing care unless it became obvious the infant was dying. That's with my infant, without any direct experience with the clinical reality. But my patient (if I were a neonatologist)? What if the parents stress they don't want extraordinary lifesaving measures carried out? How do I weigh their opinion (and define 'extraordinary') against my clinical experience and personal inclination?

I just don't see an ethical framework that addresses this issue. Many ethical decisions can be framed by the data available, which greatly eases the challenge. Here, stripped of most of the data and expertise, we're left with a raw ethical dilemma - how do we balance suffering (short and long term) vs. a chance at life, especially when the patient has no say? Does the infant have agency once they are born, even if it's at the edge of viability? Is their life or suffering valued differently?

A complex issue.

I'd like to see money poured into research into care for those born very prematurely. An artificial surrogate womb of sorts seems to be needed, for continued development of the child through 'normal' prenatal means.

Oh were I a rich philanthropist.

I don't disagree with you, Tim, but that avoids the question. Yes, technology (in the broadest sense of the word, including improvements in data-driven clinical practice) will reduce but not eliminate this problem. But that's not what I care about - I'm interested in what you would do right now, as a neonatologist. How do you work through the conflicting values and responsibilities to arrive at an ethical and just course of action?

To an extent, technology doesn't solve the problem, just shifts it. Just a few decades ago, a 26 or 27 week infant would be commonly left to die due to a combination of insufficient knowledge that these babies were in principle viable, and insufficient technology to care for them. Nowadays, that would be a wildly unethical decision (far all but exceptional cases) in any rich world hospital. This is partly due to cultural changes and a lot due to technological and clinical changes. If we were to perfect techniques to allow, say, a 22 week infant to survive, the problem would get shifted to a new grey area - 21 or even 20 week infants, who today aren't even considered potentially viable.

I'm at a loss on how to articulate a clear decision-making process.

[Khendraja'aro]

There's also the question of long-term health of such infants - as far as I know, they can suffer from quite a big range of maladies.

[RandBlade]

I think part of the problem in this is that the parents are not remotely dispassionate or rational in this instance. The decision on whether to extend extraordinary attempts to save the life needs to be made swiftly and not slowly deliberated over. The mother especially may not be in a physical (let alone mental) state to get involved in such a critical decision. The father likely might not be in a position of mental strength either and I imagine most parents then would look to the medical experts.

One possibility if you want parents involved is to think about the equivalent of a living will during prenatal appointments - but I don't think that's appropriate.

[Timbuk2]

I don't disagree with you, Tim, but that avoids the question. Yes, technology (in the broadest sense of the word, including improvements in data-driven clinical practice) will reduce but not eliminate this problem.

Well, my rather sci-fi proposition of an artificial womb is easily extended so that a foetus could be carried at any point from conception to term without a need for a mother's womb at all. So removing the issue altogether, not just shifting it.

But yes, that is still me avoiding the question

But that's not what I care about - I'm interested in what you would do right now, as a neonatologist. How do you work through the conflicting values and responsibilities to arrive at an ethical and just course of action?

To an extent, technology doesn't solve the problem, just shifts it. Just a few decades ago, a 26 or 27 week infant would be commonly left to die due to a combination of insufficient knowledge that these babies were in principle viable, and insufficient technology to care for them. Nowadays, that would be a wildly unethical decision (far all but exceptional cases) in any rich world hospital. This is partly due to cultural changes and a lot due to technological and clinical changes. If we were to perfect techniques to allow, say, a 22 week infant to survive, the problem would get shifted to a new grey area - 21 or even 20 week infants, who today aren't even considered potentially viable.

I'm at a loss on how to articulate a clear decision-making process.

Right now?

I suppose I would be in favour of information. As much information as possible, to be provided to the parents by neonatal doctors.

Every premature birth is different, every baby is different. A team undertakes a comprehensive and detailed assessment of the level of development of the premature baby, and provides as detailed a prognosis as possible on;
a. the chances of life beyond the immediate short term, and
b. the quality of life if the chances to a. are good.

This will allow the parents to make an informed decision as to which step they take with their child.

If a. is looking good but b. isn't, then some parents will choose your 'comfort care' option as they cannot bare the thought of the pain involved in raising a child with developmental issues. To some parents the very thought of not raising the child, whatever the developmental challenges may be, is anathema.

Further, even if a. is not looking good, some parents may still wish to avoid the 'comfort care' option and have everything possible done to fight for its life.

So your asking the question of how do you arrive at an ethical and just course of action? Well, any course of action that the parents choose is 'ethical' or 'just', in the context of this discussion. The burden is on them. The responsibility is on them. All the doctors should do is provide cold probabilities followed up with non-obligatory recommendations. The course of action is up to the parents.

I cannot see it as unethical for a parent to opt for comfort care if either a. or b. don't look good. Nor do i see it as unethical for a parent to opt to raise the child if either a. or b. don't look good.

~

NB. and yes I know I'm not including certain contributory factors here, such as the financial implications of any chosen course of action, but I'm trying to keep things simple to stick to the high-level question

[RandBlade]

It's not 100% sci-fi, we have had for over a century now incubators for premature babies. While not a womb it is designed to replicate and create the conditions to nurture the babies until they're strong enough and the technology involved is ever-improving as wiggin suggested. However it doesn't remove the dilemmas involved.

I like the way Aimless phrased it in that "when we can't tell those who will die no matter what we do from those who may survive if we do everything right from the get-go, our decisions should at first be biased in favour of giving the best possible care at least long enough to get a better idea of the likely prognosis." There is no perfect answer but I'd rather err on the side of trying too hard than not trying enough, and as we try harder our abilities will improve whether it be technological, experience or just best practice.

PS while Amanda was pregnant with Chloe, she was tracking the pregnancies of a number of our friends on Facebook who were also pregnant at the same time (a lot of overlap for that with our generation over nine months!) One of her friends from school in South Africa gave birth prematurely to a very underweight baby, though thankfully not as premature as we're discussing in this thread. The baby stayed in hospital in an incubator for I believe nearly two months but now is thankfully happily home and healthy.

[GGT]

...

I just don't see an ethical framework that addresses this issue. Many ethical decisions can be framed by the data available, which greatly eases the challenge. Here, stripped of most of the data and expertise, we're left with a raw ethical dilemma - how do we balance suffering (short and long term) vs. a chance at life, especially when the patient has no say? Does the infant have agency once they are born, even if it's at the edge of viability? Is their life or suffering valued differently?

That "ethical framework" has been part of the equation since medicine entered the modern era of science. The only thing that's changed is what the "general public" expects from its medical experts. And now they not only expect that 70 yr old people can live longer lives....but that babies borne at 20 weeks gestation should live. No matter what.


edit: if you want to discuss bioethics, and palliative care vs life-saving care, then why focus on premature babies.....when millions of adults have a greater need?

[Aimless]

You'll note that expectations have changed as medicine has progressed