Pediheart Podcast #327: Long Term Outcomes Of The Ross Procedure In The Newborn And Infant

Introduction to PD Heart Podcast and Episode Overview

[00:00] Welcome to PD Heart Pediatric Cardiology today. My name is Dr. Robert Pass.

[00:20] I'm the host of this podcast. I am Professor of Pediatrics at the Icahn School of Medicine at Mount Sinai where I'm the Chief of Pediatric Cardiology. Thank you very much for joining me for this 327th episode of the podcast. I hope everybody enjoyed last week's episode on the concept of SVT in the newborn or in infancy and whether or not it's

[00:40] be treated. We spoke with Dr. Anthony Pampa of Washington University in St. Louis. For those of you interested in the topic of general cardiology or electrophysiology, I'd recommend you take a listen to last week's important episode, 326. As I say most weeks if you'd like to get in touch with me, my email is easy to remember. It's PDHeart at

Background on the Ross Procedure in Neonates and Infants

[01:00] Gmail dot com. This week we move on to the world of cardiovascular surgery. The title of the work we'll be reviewing is Long-Term Outcomes Following the RAS Procedure in Neonates and Infants, a Multi-Institutional Analysis. The first author of this work is Jason Greenberg and the senior author of this work is David Winlaw and this work comes to us

[01:20] from Cincinnati Children's Hospital and multiple centers throughout the world, as well as the so-called Ross Collaborative. When we're done reviewing this paper, the senior author Dr. David Winlaw has kindly agreed to speak with us about it. Therefore, let's get straight onto this article and then a conversation about the Ross procedure in newborns and young infants. The authors began

[01:40] In this week's work with some comments about the difficulty of managing infants with critical aortic stenosis with annular hypoplasia, and how there are so few options for valve replacement in this age range, with the ROSS procedure, whereby the aortic root and valve are replaced with a pulmonary valve autograph, being the most effective. However, prior reports

[02:00] Reports have suggested substantial morbidity and mortality after the ROS, particularly in newborns or infants who are undergoing this procedure, where some reports have suggested operative mortality as high as 30%. The authors explain that these somewhat suboptimal report results have to use their words, quote, led to restraint in the use of the ROS.

[02:20] And the authors wondered if outcomes were really as bad in the present era, and also wonder about the fate of the neo aortic root, as there is a belief that smaller infants have the potential for tissue remodeling potentially resulting in better long term autograph function. With this as a background the authors write and I quote, In this, the first report from a

[02:40] Multinitiational and Multinational ROS Collaborative. The authors report contemporary early and long-term operative outcomes for neonates and infants after the ROS procedure, emphasizing anatomic and functional changes in the autograph and freedom from surgical reintervention. The primary study objective was to describe long-term

Study Overview: Multi-Institutional Analysis of Long-Term Outcomes

[03:00] meaning more than five year autographed or neaeortic durability and function in newborns defined as less than or equal to 30 days, or infants described as less than 365 days at the time of surgery. The autographed durability and function was assessed by freedom from surgical reintervention on the autographed or LV-outflow tract, as well as freedom from

[03:20] moderate to severe neaeortic valve regurgitation and growth or dilation of the valvular annulus and the autograph over time. The authors described the secondary outcomes of this work to be in-hospital and long-term mortality rates. The work included all infants and newborns who had a ROS at six participating centers from January of 1998

[03:40] in December of 2016, and they were found retrospectively. And this date in 2016 was chosen so that there was a full minimum five years of follow-up for every patient in this work. Serial echocardiograms were performed and reviewed, and the authors explained that they had follow-up in 83% of the hospital's survival.

[04:00] fibers with serial echoes. The authors reviewed the data from the echoes and what normative data they used for comparison as well as what Z-score data they used. Many different demographics were recorded as well as the presence of genetic syndromes like the George or Goldenhar syndrome to just name two. And onto the results. There were 133 infants

[04:20] in this group with 23% or 30 neonates. The diagnosis of these patients were isolated aortic stenosis in 19% or 14%. Shone's complex in 19 patients or 14%. Aortic stenosis with other features not with Shone's complex in 95 patients or 71% including

[04:40] obstruction in 55, LV hypoplasia 9, and mitral valve disease with moderate or greater stenosis or regurgitation in 31. The median age at Ross was 96 days with a median weight of 4.4 kilos. Amongst the neonates, the median age was 9 days with a median weight of 3.3 kilos.

[05:00] and a quarter of all the patients were ventilator-dependent at the time of the ROS procedure. Importantly, many of the patients had undergone surgical interventions before the ROS, including aortic arch repair in 53 of 74 patients who had arch obstruction and 3 of 50 with mitral disease had mitral valve repair or replacement before the ROS.

[05:20] Also, and perhaps not surprisingly, 64% had had a balloon dilation of the aortic valve and 32 or 38% of those who had a balloon performed pre-ROS had moderate or severe aortic regurgitation at the time of the ROS and this was the rationale for the operation. The in-hospital mortality seen in this group

[05:40] was 13 of the 133 presenting patients, or 10%, and was 4 of 30, or 13%, amongst the neonates. All patients who survived the ROSTA discharge had 30-day survival after discharge, and the median follow-up duration after discharge in this work was 10.8 years. And what

Patient Demographics and Surgical Details

[06:00] What happened to them after discharge? Well, post-discharge mortality occurred in 10 of 120 discharged patients, or 8%, at a median of 298 days following the ROS. One of these 10 was a neonate. The authors state that late death was due to cardiac-related pathology in 7 of 10 or 7

[06:20] with LV failure seen in four and three mortalities with no obvious explanation recorded. Six of ten who died later had pre-ROS obstruction of the aortic arch and five of ten had mitral disease. In all, the all-cause mortality in this entire group was 17% both early and late combined.

[06:40] And what happened to the autograph over time? Well, there was significant dilation over time at the annulus, aortic root, and sinus tubular junction. Dilation seems to have begun shortly after the ROS and peaked at the 2 to 3 year mark, reaching 4 to 5 standard deviations from normal. After this, the Z-squared

[07:00] scores fell and normalized, with normalization of the autograph size occurring at roughly 6 to 7 years for the aortic valve annulus, 8 to 9 years for the root, and 12 to 13 years for the sinotubular junction. Importantly, neo-aortic valve function was good in follow-up, with 98.5% freedom from

[07:20] significant Neo-AI at five years post-ROS, and 86% at 10 years and beyond. Figure 2 on page 1726 demonstrates nicely the changes in aortic dimension Z-scores with time following the ROS that I've just explained. And what percentage of those who survived ultimately needed some

[07:40] intervention on either the autographed or the alveofil tract? Well, the answer was just five patients of the 120 hospital survivors, or 4% needed such, with a median time to intervention of 10.3 years, and patients had such things as an aortic root replacement with a homograft, subaortic stenosis resection with a cono,

[08:00] aortic valve repair in 1, and aortic valve replacement 13 years later in 2. In their discussion, the authors state and I quote, The present multi-institution, multinational study of neonates and infants undergoing the ROSS procedure, the first report of the ROSS collaborative represents the largest cohort to date and includes the longest median follow-up duration

[08:20] of any yet published. Long-term survival after the ROS and freedom from autographed LV outflow track reintervention were excellent in the cohort, with acceptable early postoperative survival and excellent preservation of neo aortic function over time. Echo analysis confirmed that neo aortic dilatation does occur in infants after the ROS.

[08:40] but that the autograph subsequently returns to near baseline Z-scores in the long term. Such findings have significant clinical implications and may explain the outstanding freedom from autograph replacement seen in the cohort. The current study serves as a benchmark of contemporary outcomes of the ROSS procedure in the high-risk cohort of infants with aortic

Post-Surgical Outcomes: Mortality, Complications, and Survival Rates

[09:00] valvular disease, while providing novel and useful information for clinicians caring for the high-risk patient population of neonates and infants with critical aortic stenosis and annular hypoplasia. The authors then briefly review the history of the ROS, stating that it was first reported by Dr. Donald ROS back in 1967 and they

[09:20] They review how a large recent study in the German RAS Registry showed outstanding post-RAS survival in adults up to 25 years that was similar to the general population. They review the oft-mentioned, most important possible drawback of the RAS, specifically the jeopardizing of function of two valves instead of one, the pulmonary and the aorta.

[09:40] potentially taking a disorder in which there is a single valve problem, the aortic valve, and creating a scenario of two-valve disease. The authors then review the fact that in adults and older kids it is known that the autograph has a tendency to dilate with time in a progressive and irreversible manner, and how re-operation in the second decade after the ROS is not on

[10:00] common. In younger children, however, the fate of the autograph is less well established, and the authors suggest that there may be some sort of enhanced ability to heal and remodel tissue that may protect against autograph dilation and degeneration that is more commonly seen in older patients who undergo this operation. They point to a recent work

[10:20] showing similarly that the aortic dimension study do enlarge after repair, but they contrast that work to this present one, suggesting that if one follows or watches these patients long enough, the dilatation seen initially normalizes with longer follow-up after the ROS. They re-emphasize the very low rate, 5 of 120,

[10:40] of reinterventions on the autographed in this study, and suggest that this may be related to the ultimate stabilization of the autographed in this patient group, and they are hardened by only 15% of moderate to severe AR at 10 years out. The authors emphasize that if a valve can be palliated either by repair or balloon dilation in the young patient,

[11:00] that this should be the first step prior to consideration of a ROS, and they suggest that it only be used in patients in whom the first line approaches fail or where there is pessimism regarding the likelihood of succeeding. They then review the question of whether it would be better to palliate and leave a patient with a sub-optimal aortic valve function with the goal of later ROS,

[11:20] and review how that would potentially be more deleterious in regards to LV long-term function, but also that earlier ROS does have a significant mortality associated with it and so it is quite a balancing act. In regards to limitations, the authors point to the fact that because the goal was to see how patients were doing more than at least five years after the ROS, by definition,

[11:40] in the patients in this work reflect a historic cohort, and newer surgical approaches, including the increased use of Ross-Kono rather than separating them, and the use of ECMO more readily after surgery, may mean that the results reported herein may actually underestimate the excellence of outcomes, and that things may actually be superior today.

Analysis of Neo-Aortic Root Changes and Long-Term Function

[12:00] than we're seeing in this cohort. They speak also about how we sometimes use things like beta blockers and are more aggressive with our blood pressure goals these days to reduce autographed enlargement and how this has not been studied, but may yield even more superior long-term results. And so they suggest that the all-in 17% mortality rate

[12:20] that's seen in this work may not reflect present 2025 day outcomes. They also point to things not studied in this work, specifically things like the RV to PA conduit stability with time and the long-term LV function to name just two. And so they conclude, in this the largest and most complete analysis of neonate and

[12:40] outfit Ross outcomes to date. Long-term autograph durability, function, and survival extending several decades after surgery were excellent. Although early and mid-term autograph dilatation occurred after the Ross, there is a ventral return to baseline Z score. The study results suggest that the Ross may be the ideal course of

[13:00] for neonates and infants with aortic valve disease and certain left-sided pathologies, rather than postponing valve replacement in this patient population. Notably, complex operative candidates did not fare worse than more straightforward cases, suggesting that differences in technical performance may explain some of the early outcomes and that

[13:20] lessons from high-performing centers can inform practice. Well, this is a most interesting report. On the one hand, it's remarkable that 90% of patients undergoing such a complex and challenging operation survive to leave the hospital, and the authors suggest that these numbers are potentially even better today. However, a 17% all-in-

[13:40] mortality is not tiny, and I wondered as I read this why the STS considers this to be a STAT2 operation given these relatively dangerous sounding numbers. I think it's fascinating that the authors have demonstrated normalization of the autograph in regards to size, and I think that this is one of the most important contributions of this work.

[14:00] As I think most of us think that a dilated neoriotic route is basically a guarantee for this repair. And these authors have shown us that this is not the case and in fact argue that maybe something about the young patient may account for this. There's a lot to unpack about this and so in the interest of time I think we'll proceed with our interview with the work's senior author, Dr.

[14:20] David Winlaw. David Winlaw is the Division Head of Pediatric Cardiovascular Surgery at Lurie Children's Hospital in Chicago. He is also the Willis Potts Professor of Surgery at Northwestern University. Dr. Winlaw received his medical degree from the University of New South Wales Faculty of Medicine in Sydney, Australia and performed fellowships in pediatric

[14:40] CT surgery at both Birmingham Children's Hospital in the UK as well as the Children's Hospital at West Med in Sydney, Australia. It is a delight and honor to have him join us this week. Welcome Dr. Winlaw to P.D. Hart. I'm here now with Dr. David Winlaw of Northwestern University. Dr. Winlaw, thank you very much for joining us this week on the podcast. It's a pleasure

Interview with Dr. David Winlaw: Insights on Ross Procedure in Infants

[15:00] to be here. Thank you very much for your interest in our work. Thank you. Very happy to have you and thank you. You know, Dr. Winlaw, although this was not really the emphasis of your work, I have to say that as I read the work, I wondered to myself why the STAT categories from STS list the ROSS procedure as a STAT2 procedure despite mortality

[15:20] rates in your work that were about 10% and total of 17% at long-term follow-up, which actually, from what I have read, is superior to many prior works on this topic. And I understand that your work basically establishes that the ROS is an excellent option in the right hands for the right small

[15:40] And I wondered if you agreed with categorizing this particular procedure as a STAT2 STS procedure. Well, it's interesting. It's not really a matter whether I agree with it or not because the STAT categories are based on empiric observations. So registry data, the success of previous operations,

[16:00] operations in both the STS and the European registry determine what the STAT score is. And actually in the most recent revision, the 2021 version, a straightforward ROS operation is a STAT1 operation. So expected mortality of 1% or less.

[16:20] So most of the procedures that we've described in infants and neonates are ROS operations with a cono incision in the interventricular septum to accommodate the pulmonary valve within a small annulus, and when you add a cono procedure

[16:40] to the ROS, it becomes a STAT4 operation. So with expected mortality of around about 10%, which is kind of what we reported in this multi-center experience over many years. And so I think the STAT categories are accurately reflecting the

[17:00] mortality. But there is, of course, you would understand there is a big difference between an elective ROS operation and a 15-year-old that is critically neonate. And much of the mortality in the ROS cono group is actually driven not by the ROS or the cono. It's

[17:20] driven by the coexisting mitral valve disease, which is such a common association in the Shones type patients. Yeah, yeah, very interesting points, all of them. Thank you, Dr. Winlow. That was very clear. You know, in your work, neo aortic roots appear dilated initially, but then very fascinatingly, the Z-scores

[17:40] normalized in the group over time. And this has not previously been demonstrated in other studies of RAS patients, and I think that's one of the many important contributions of your work. I'm wondering if you could speak about what you believe are the factors that may account for the RAS neo aortic roots eventually in this particular patient.

Current Surgical Approaches and Future Improvements in the Ross Procedure

[18:00] patient group not being as dilated as in other patient groups. And I'm wondering if you think that there are surgical factors that may account for infant-ross aortas sort of normalizing with time, or do you think that there are other factors? And I also wondered if you thought maybe the fact that the aortic root normalizes might be an

[18:20] you meant to think about doing a RAS earlier in a baby who has a problem with their aortic valve? Yeah, great thoughts. There are all the questions that we were asking in construction of this study and I think mostly it's about the adaptation of the neo-aortic route and that this adaptation is possible

[18:40] early in life, but cannot occur later in life. All of our patients were neonates and infants. The threshold after which this adaptation doesn't occur is not really known because we don't have detailed data like this on patients that are beyond the infant.

[19:00] to age having a RAS procedure. But we know, for example, an adolescent patient is at pretty high risk of neo aortic dilatation and development of AR over time. And so I think it's like so many things, there is a window of opportunity as a neonate and an infant. The coronary circulation

[19:20] shows similar opportunities. You can restore circulation, for example, in an alcarpa. Ventricular function will improve over time. So I think that ability to adapt, we have as a young baby, but not as an older person. I think actually there are not

[19:40] technical reasons why this is better. In fact most of the Ross roots that are done in babies are done as unsupported roots because you need to allow some growth, quite a lot of growth over time and so we don't do many of the adjunctive techniques to support

[20:00] the route that we would do in an adolescent, for example. But I think there is reason to think about the Ross-Conor more often. And I think this paper addresses one of the great concerns that we would just get dilatation, which will be progressive and problematic.

[20:20] I see. Well, you certainly showed that that in fact was not the case, so that was very interesting indeed. You know, Dr. Winla, the work you wrote was written with a historical cohort in which a Ross procedure was at least five years ago, intentionally so that you could have long-term follow-up. But I'm wondering

[20:40] You know, the average patient in your study had been followed over 10 years. Are there any novel approaches to the ROS in this patient group in terms of technical aspects or management that you think might actually result in better outcomes, meaning that if you like were to study the present day baby ROS

[21:00] operation patients? Do you think they're going to be better in terms of what you've observed and why do you think that might be? You know, Robert, it was fascinating reading all the old operation reports to see the problems that surgeons of a different era had encountered and actually many of those problems are often

[21:20] still encountered and the solutions we have are not particularly different. I would say that the improvement in outcomes is mostly related to all the little iterative improvements that all members of the team have made to their work. You know, surgeons are much more familiar operating in the neonatal and infant period.

[21:40] now than there were a generation ago. Perfusion is much better. I think the ability to support the patient postoperatively with ECMO is another important thing. We have a much lower threshold to do now and many of the patients

Management of RV-PA Conduits and Long-Term Considerations

[22:00] have benefited from a short period of ECMO to tie them through the period that their cardiac output may have necessarily been diminished postoperatively. So I think this is an important tool. We're doing it more often. Of course, we don't like to do it. We do it when necessary.

[22:20] Dr. Winlaw, you're a highly experienced surgeon and as such, in this very young patient group, how do you decide between a RAS operation and someone with bad aortic valve in this age range and repair? And I'm thinking particularly in the newborns, you know, these valves look very abnormal. We sometimes describe them

[22:40] is looking like a blob and yet sometimes they can be repaired as a reasonable palliative approach. And how do you make that decision? Is this a decision typically that makes before you make before you even start the operation or is it more of shall we say a game time decision in the operating room when you're inspecting the valve? Do you think that where

[23:00] underutilizing this approach of the Ross and maybe harming patients by not offering it for less severe forms of aortic valve disease? Yeah, a blob, a cauliflower, there's lots of descriptions of a very unsatisfactory aortic valve and I think where you can do a repair and get a reasonable

[23:20] duration of free of importance to nurses or regurgitation or repair is what you should do and I think deferring the rest is a sensible approach when possible but in in most people's practice neonatal aortic stenosis is treated by balloon valor

[23:40] alveoplasty and the outcomes of that will either be good or terrible and it's the terrible ones that end up having a neonatal Rosscono and I think that's a very appropriate pathway. I think there are certainly surgeons who will repair more

[24:00] valves than some. But I think we probably have the indication correct here. I would not advocate a Roscono over a successful balloon valve ileplasty, for example. I don't think that's the comparison. But

[24:20] When you have a patient with moderate AI who's not doing well after a balloon or there's residual stenosis or it's just a terrible unicuspid valve, then Roscono is a good approach and we know that we can do that with relative safety and efficacy. Yeah, of course, if we had Dr. Hruska on today, he would be telling us

[24:40] that it was criminal to use a bull inhalation and everybody should have a surgical valvuloplasty. It's a debate for another day. Yes, he was a very adamant believer in that. Well, for those in the audience, it's right in the middle of Dr. Winlaw's morning and so I don't want to take up much more of his time, so I'm going to finish up with this final question.

[25:00] And you know, although this was not the focus of this work, I couldn't help but ask you, since you I'm sure know this data very well, if you could comment to the audience on the outcomes of the RV to PA conduit, which was, I'm guessing is going to be in another paper, but I'm wondering how often did they need reintervention and how

[25:20] soon on average do patients need to have that intervened upon in one way or the other? Yeah, it's one of the downsides of the approach. But in our mind, maintaining left ventricular function and satisfactory growth is the fundamental determinant of long-term outcomes. So we're happy to look

[25:40] after the left at the expense of the right to some extent. Most of these babies will receive a homograft as part of their rv to PA connection. They will need that replaced sometime between five and seven years of age. Probably that would last another eight to 10 years.

Conclusion and Final Thoughts

[26:00] So another replacement in adolescence would be expected and thereafter I think the approaches are partly interventional and partly surgical. So yes this is a problem. We don't have an IVDPA conjure that grows with the patient yet. We don't have ones that are particularly durable but I think

[26:20] think it's an acceptable compromise at this point. Yeah, absolutely. Well, Dr. Winlaw, I can't thank you enough for taking time out of your busy schedule to speak with us. I also want to congratulate you and all the many authors from six different major centers throughout the world who were kind enough to participate with you and share their data for us all too.

[26:40] learn a lot about the Ross and this young patient population. So congratulations and thank you so much for joining us this week on the podcast. Yeah, thank you very much. Great pleasure. As Dr. Winlaw was an uncommonly clear speaker, I think there's little for me to add. I think he really fleshed out a number of important points and takeaways from this work and I remain so impressed that the Ross Collaborative was able to

[27:00] collects so many unusual cases of babies needing a RAS procedure at such a young age, and the length of follow-up is also notably impressive. So often that operation feels as if it is a bailout operation, and we're clinically just pleased that we get out of the early years of life alive and with reasonable functioning aortic valve and LV. However, these

[27:20] data clearly show that even in the long term, outcomes are quite reasonable. And I'm sure that this collaborative group will continue to report on this group even over longer periods of time so that we can better anticipate the future in this complex group of patients. I'd once again like to thank Dr. Winlaw for taking time from his very busy schedule to speak with us this week on the podcast. To conclude

[27:40] his 327th episode of Pedyard Pediatric Cardiology today, we hear the American baritone John Brancy, who's originally from the state of Pennsylvania, but grew up in New Jersey and studied at Juilliard. His career is slowly taking off in venues throughout the United States, Canada, and Europe. Today we hear him sing the achingly sad and beautiful

[28:00] Aria, Yavas Lublu from Tchaikovsky's Picdam, in which Prinzioletsky sings of his love of Lisa, singing Yavas Lublu, Lublubismerna, which translates, I love you, love you beyond all measure. Thank you for joining me for this 327th episode and thanks once again to

[28:20] I hope all have a good week ahead.