Pediatric heart transplantation with high-risk donors may confer similar outcomes to low-risk donors
A recent study published in the Annals of Thoracic Surgery found that cardiac allografts from high-risk donors implanted in children resulted in similar posttransplant survival as those from low-risk donors.
David L.S. Morales, MD, director of congenital heart surgery at the Heart Institute and Clark-Helmsworth Chair of Cardiovascular Surgery at Cincinnati Children’s and professor of surgery and pediatrics at University of Cincinnati College of Medicine, and colleagues aimed to determine whether the use of high-risk donors would confer similar outcomes to low-risk donors. Through this, the issues associated with pediatric heart transplantation could potentially decrease such as long waitlist times and high waitlist mortality; up to one-quarter of children die awaiting a donor heart.
Healio spoke with Morales about what was learned through the study, the potential benefits of using perceived high-risk donors and how cardiologists should approach this area of heart transplantation.
Question: What do the findings add to the knowledge base?
Answer: The number of pediatric transplants that we do fall short of the number of patients that are actually on the pediatric transplant list, so kids wait and often die waiting. In fact, pediatric heart transplant has the highest waitlist mortality of all solid organ transplantations. Why is that? Everyone says there’s a shortage of donors, but there are a lot of donors that aren’t used. As we know in all of medicine, even if papers come out and literature proves things, it takes years to decades for that knowledge to be put into day-to-day clinical practice. Many donors that are considered to be “high risk” maybe aren’t high risk and can be used successfully.
We demonstrated that a lot of the perceived donor risk do not affect outcomes and that the recipients’ characteristics have a much greater effect on outcomes. There were two other published studies that looked at what makes a donor high risk. What we did is we said, “OK, why don’t we make sure that all of the recipients are the same, have the same characteristics, and look at how those donors did in those patients.” What we demonstrated is whether they got a high-risk donor or a low-risk donor, they did the same, and that it was a recipient-driven process more than a donor process. We’re not saying that the donor heart doesn’t make a difference, but many of the things that we think are risk factors aren’t necessarily risk factors. There’s literature to demonstrate that. It’s just that we don’t put that into practice.
Q: Do you think it’s because doctors are trying to be cautious about what donors are used?
A: This is probably one of over 50 papers that I’ve written with Farhan Zafar, MD, who’s also on this paper and my colleague, looking at how organs are distributed in transplantation. We recently received a multimillion-dollar RO1 grant from the government to investigate how to potentially improve thoracic organs distribution in pediatrics.
The goal is not to get to transplant; the goal is for the child to be alive and healthy for a long life. Right now, our system is built in getting patients to transplant. The sickest patients are the highest on the list, get organs the quickest and usually get the best organs. However, we and others have done studies showing that this doesn’t make sense. Whether they get a good organ or a less than ideal organ, their survival is the same, but if I took that good organ and gave it to a recipient who isn’t as sick, then they would do much better than if they had gotten a less ideal organ.
If the goal is to get the most years posttransplant for the community, then what we want are best matches. We want the best recipient and donor pairing to maximize survival. If we just focus on getting someone to transplant, when we have data to show that their survival will not be good posttransplant, it doesn’t help neither the patient nor the community.
Hopefully with the use of machine learning (artificial intelligence), we have enough data to start creating algorithms that will predict the best matches. Right now, there are no agreed upon standards across institutions when patients get listed for transplantation, what donor organs are accepted or even the acceptable weight range of donors for a particular size of recipient. How we manage our recipient transplant list can and is often completely different than a children’s hospital in a neighboring city or state. It’s all based on what the provider’s and the program’s experiences and beliefs are. We have enough data now to create data-driven algorithms so we can standardize approaches and predict best matches and maximize outcomes.
The decisions around transplantation shouldn’t be emotionally driven. Often caregivers get attached to a child who then gets put on a lifesaving machine such as [extracorporeal membrane oxygenation] and who we know is probably not going to do well even after transplantation regardless of the organ they get, yet they get the best organ out there. However, if we passed on this ideal organ for a less ideal organ, we know that it would not affect their outcome, but it would greatly improve the outcome of a “less critical” recipient. However, you cannot really ask the physicians, the nurses and the people who are at the bedside to make those kinds of decisions. When we are at the bedside, we have an obligation to that patient, but from a policy standpoint, if a data-driven algorithm determined best matches, we would not have to make these decisions and it could potentially increase overall survival for the community, tremendously.
Q: What are some of the precautions that need to be taken when high-risk donors are considered for heart transplantation in children?
A: You have to make sure that you have all of the information and that you feel comfortable about moving forward. For example, it used to be thought that if the donor had a cardiac arrest and CPR, that the organ was not usable or “high risk,” which makes sense. However, there are studies demonstrating that transplantations using these donor organs have equivalent outcomes to donors that did not have CPR.
It is important to be abreast of the literature, understand it and try to be more data-driven rather than just practice how you are used to practicing. The issue is that even the largest pediatric transplant programs do anywhere from 15 to 30 transplants at most, so no one is doing hundreds of pediatric transplants. Most programs do five to 10 transplants. I, like others, find it difficult to take all the different risk factors into account and change practice when with these low numbers, it is hard to see if your changes have a positive or negative effect on your programs’ patients. This is why having some national standards and/or a matching algorithm that is data-driven makes sense.
Another important point is patients on the transplant list are listed with a weight range of donors that they are willing to accept. Let’s say I have a 10 kg patient. I might list them from 8 kg to 20 kg, but that’s just based on nothing. Not nothing, but just based on my experience, how my program has done it and how I was trained. However, this could be very different at another program. We are “guesstimating” that these donors will have hearts that fit in our recipient.
We’re starting to use virtual transplantation — and this work is also part of that government grant — that can say if you’re this size, we can tell you by these calculations which hearts would fit and which won’t fit. We’ve shown that you could increase the amount of offers you get significantly, so it’s something else that needs to be taken out of programs’ colloquial habits, and be driven by data. Programs would just input their patient’s total cardiac volume, then an algorithm can figure out what hearts are going to fit and thus the donor range for their patient’s listing.
Q: What needs to be done to remove the stigma of “high-risk donors” so that children and other patients receive the hearts that they need to survive?
A: Change is difficult. It’s easy for me to say follow the data because I do not have a patient in front of me, but when I do, I’m just as much at fault as anyone else. When I’m getting an offer, I start thinking about patients in the past, the present patient and their family and what my experiences are. It’s hard for me just to trust the data, but that’s what we should do.
A significant change in the system will have to be driven by the government or medical societies. These policies can’t be done at the individual physician level because you’re taking care of a patient who needs the transplant and you know his parents and the rest of your team want him to get a transplant. It would be hard to say, “You know what? I know that organ would be good for you, but it would be better for someone else.”
This somehow has to be automated so that those decisions are taken out of our hands and we follow a data-driven methodology. We’re getting to a point where we have enough data, computer power and machine learning techniques to create an algorithm that only gets better and better over time. As the algorithm gets used, it will learn from its experiences which will be nationwide and only get better at predicting and making best matches. Change and improvement will be fast. Our present system is not much different than it was a decade ago, and individual programs do not change/learn rapidly. Therefore, if we are really going to make a significant improvement in pediatric thoracic transplantation, we need to change out, not modify the present system.
Q: What are some calls to action that you have for cardiologists who perform heart transplants?
A: Pediatric transplants caregivers and the transplant population is such a wonderful community and has done so many great things like the Pediatric Heart Network to improve the outcomes of our patients. Everyone in the field is trying to do the best for our patients. However, we need to take the next step, which is a scary step, and go to an organized, data-driven system with algorithms and national standards. This is a big ask because caregivers will have to put aside our personal biases and feelings, as well as give up meaningful control of the system, which is difficult for professionals to do. One has to remember, it may not be the best system in the beginning, but is the best method to continue to learn and improve.
This will be required if we are to move toward a more data-driven algorithm of how to distribute organs. Also, we have to always remember getting to transplant is not the goal. Often, when a child is sick, people say, “If we can just get to transplant, they’ll do fine.” We already know that if they have certain characteristics and they get a transplant, their survival is going to be decreased by 25% or 30%. We should do what we can to get them in the best shape possible before transplant so when they get it, they can enjoy the longest and most fulfilling life they can with their new organ. There are ways to do that through the use of ventricular assist devices or total artificial hearts.
Given the field’s push for transparency, these changes may be driven by parents and others. I’m sure most families don’t realize the diversity of how people get listed, what organs get accepted and that there are no standards and thus, their transplant journey can be very different depending on what hospital you go to. There are examples in transplantation where standardization has occurred; for example, renal transplantation. Since standardization of their system, they have shown increases in organ utilization and posttransplant outcomes. – by Darlene Dobkowski
For more information:
David L.S. Morales, MD, can be reached at email@example.com; Twitter: @dlsm_md.
Disclosures: Morales reports he received support from the NIH through an RO1 grant. Please see the study for all other authors’ relevant financial disclosures.