#435 - On With VON - Transfusion Thresholds

Hello friends 👋

The transfusion threshold consensus is here — but practice hasn't fully caught up. In the second episode of On with VON, Ben and Daphna sit down with Dr. Roger Soll and Dr. Ravi Patel to extend the conversation from the Vermont Oxford Network Grand Rounds on evidence to practice for transfusion thresholds.

The core finding across trials is consistent: lower thresholds for both packed red blood cells and platelets appear safe. The guidelines are freely available in JAMA Network Open and actionable — 11, 10, 9 grams per deciliter across the first three weeks for infants on respiratory support. So why hasn't practice shifted uniformly?

The group works through the populations the trials didn't fully capture — hypoxic-ischemic encephalopathy, the most premature infants, and babies in the first week of life when intraventricular hemorrhage risk peaks. On NEC and feeding during transfusion: the data may surprise you. On transfusion volume and infusion duration: an underappreciated variable, particularly for platelets.

The episode closes with practical guidance on implementing transfusion guidelines at the unit level — who needs to be in the room, how informatics tools can support decision-making, and why understanding protocol deviations matters as much as the guidelines themselves.

Link to episode on youtube: https://youtu.be/8xFvVlcDl80

----

Short Bios

Dr. Roger Soll: Dr. Roger Soll is the H. Wallace Professor of Neonatology at the Larner College of Medicine and is President of the Vermont Oxford Network and Director of Network Clinical Trials and Follow-up. Dr. Soll is an authority in evidence-based medicine and randomized clinical trials. He is the coordinating editor of Cochrane Neonatal, part of the Cochrane Collaboration, and author or co-author of the Cochrane Reviews of surfactant therapy. He is the author of numerous peer reviewed articles and book chapters on the subject of surfactant replacement therapy and evidence-based medicine. A native of New York City, Dr. Soll graduated from Cornell University with a degree in Genetics and History of Science in 1975. He received his M.D. degree from the University of Health Sciences/Chicago Medical School in 1978. He returned to New York City to complete his residency training in Pediatrics at Bellevue Hospital/New York University Medical Center in 1981. After 2 years with the Public Health Service, Dr. Soll returned to academic training. He completed the post graduate fellowship in Neonatal-Perinatal Medicine at the University of Vermont in 1983 and has remained in Vermont ever since.

Dr. Ravi Patel: Dr. Ravi Mangal Patel is an Associate Professor of Pediatrics and Director of Neonatal Clinical Research at Emory University and Children's Healthcare of Atlanta. He is passionate about evidence-based medicine. His research interests include necrotizing enterocolitis, neonatal transfusion, caffeine therapy and perinatal epidemiology. He is an executive committee member of the American Academy of Pediatrics, Section on Neonatal-Perinatal Medicine. He is also a principal investigator in the NICHD Neonatal Research Network, chair of the International Society for Evidence-Based Neonatology (EBNEO) and president of the Southern Society for Pediatric Research. Dr. Patel received his MD from the Medical College of Georgia and an MSc from Emory University.

----

The transcript of today's episode can be found below 👇

Dr. Ben Courchia (00:00.801) Hello everybody. Welcome back to the Incubator Podcast. We are back today for the second episode of our series, On with Vaughn, where we extend the conversation from the Vermont Oxford Network (VON) Grand Rounds. Many people in the studio today. Daphna, how are you?

Dr. Daphna Barbeau (00:19.810) I'm doing well. Our first On with Vaughn was very well received — I'm not surprised. The VON team is doing a wonderful job. Their webinars are excellent. We're just trying to add to the conversation, go deeper, and make sure we get to questions from the community that we weren't able to address in the full hour.

Dr. Ben Courchia (00:43.751) Absolutely. To remind everybody: questions in the chat are addressed live during the VON Grand Rounds session. The ones we can't get to are the questions we'll dive into here on the podcast. We are in the studio this week with Dr. Roger Soll. Roger, welcome back to the podcast.

Dr. Roger Soll (01:11.270) Thanks Ben, thanks Daphna.

Ben Courchia (01:13.019) And the speaker of the latest VON Grand Rounds Evidence to Practice, Dr. Ravi Patel. Ravi, welcome back to the show.

Dr. Ravi Patel (01:21.582) It's great to see you both, and thrilled to be here with Roger.

Dr. Ben Courchia (01:25.239) The topic of the Grand Rounds we're reviewing today is evidence to practice: transfusion thresholds. The Grand Rounds took place on February 4th, 2026. The first comment I want to make — and get your take on, Roger — is that as you review the evidence for transfusion thresholds, both for packed red blood cells (PRBCs) and for platelets, the different trials seem to be in agreement: transfusing at lower thresholds does not appear to be harmful for our patient population. Is that a fair assessment?

Dr. Roger Soll (02:04.851) That is. And let me take two seconds to thank you for having us — VON is really pleased to have this opportunity to work with the Incubator. We're also glad to extend the conversation because we always shut things down before questions are answered.

When we address thresholds, there's always a fundamental challenge: one person's high threshold is another person's low. As you sift through the literature, what was a high threshold in one study is the low threshold of another. There's a relative paucity of large, well-designed studies, and the thresholds used are very discrete. That said, we can say something specific — both for PRBC transfusions and for platelets — about what we know as we proceed toward lower thresholds. In both cases, and it was an unusual choice to combine them in one discussion because they have very different issues, we were able to look at the primary studies and make conclusions mostly about the relative safety of going to lower transfusion thresholds.

Dr. Ben Courchia (03:30.241) What's interesting is that, as you say, as robust studies have been published, we continue to see variability in practice across both Europe and the US. Despite the evidence, consensus has been elusive. Ravi, how much variability exists, and why do you think it persists — whether for packed red blood cells or platelets?

Dr. Ravi Patel (04:09.836) That's exactly the right question, and the nice thing is we can quantify it. One study we've done used something called the National Heart, Lung, and Blood Institute REDS-4P program, which has been around for several decades and allows us to measure practice by looking at what the platelet count or hemoglobin was right before a transfusion was given — essentially a surrogate for the threshold being used.

What you see is a large range of variation for red cells that spans the thresholds from the low and high arms of the trials. This was before the most recent Cochrane review and some of the more recent evidence, but it does suggest that the variability in practice was actually tested in the trials — and within that variable range, lower thresholds appear to be safe.

Platelet practice is even more variable. We've also looked at how practice has shifted over time. For red cell transfusions, clinicians have moved toward lower pre-transfusion hemoglobin levels — suggesting some adoption of the restrictive evidence. For platelets in extremely low birth weight infants, practice appears largely unchanged between 2019 and 2023.

Dr. Ben Courchia (06:12.311) That goes back to whether we agree as a community on the rationale for what we do. Why do we transfuse packed red blood cells? Why do we transfuse platelets? The divergence may stem from the fact that with platelets, we're worried about bleeding events — what counts as significant, how to define it — whereas with packed red blood cells, the oxygen-delivery rationale is more intuitive. Roger, I'll let you go first on that, then Ravi.

Dr. Roger Soll (06:55.551) There's a fundamental human nature aspect to this discussion. In critical care, one of the most important lessons we learn is how to accept values outside the normal range — how willing we are to tolerate that before we intervene. And our babies are always outside the normal range.

Dr. Daphna Barbeau (07:13.624) Always.

Dr. Roger Soll (07:17.847) We've spent decades pushing them back to normal — whether it's CO2, sodium, whatever. There's always this teleological pull to restore what we think is natural. And the more we test that impulse, the more we find that babies are resilient to these aberrations, and that our interventions carry harm — whether it's the ventilator, fluid management, or in this case, transfusion thresholds.

When I was a student, we transfused to replace what we took for testing. We transfused arbitrarily, even in children with normal hematocrit. The thresholds chosen in early trials like the PINT trial reflected what we were comfortable with physiologically, not an evidence-based minimum. Time and again, the lesson is the same: babies are resilient, and our interventions have harms. I'll turn it to Ravi for specifics.

Dr. Ravi Patel (08:33.866) I'd build on that by asking: what are you trying to achieve when you give a red cell transfusion? If you're just trying to increase hemoglobin, you can achieve that. But we want more — we want to improve oxygenation and outcomes: survival, neurodevelopmental integrity. When you look at those outcomes, they haven't been reliably achieved with red cell transfusion. Survival without neurodevelopmental impairment has not been shown to improve.

Even oxygenation — which is the physiologic rationale — shows quite variable responses, possibly related not just to when we transfuse, but to what is actually being transfused. This may explain why we don't always see improvement in oxygenation by near-infrared spectroscopy following a transfusion. The starting point should be: what are you trying to achieve? And the best available evidence should guide that decision.

Dr. Daphna Barbeau (09:55.572) That's very helpful. Some listeners may be thinking: we've changed thresholds before, why does the degree of variability matter? And on the other side — we don't know how helpful transfusions are, but we know there are potential harms. Can we talk about those? Ravi?

Dr. Ravi Patel (10:31.256) I'll touch on that. It's important to consider not just our preferences as clinicians, but what parents and patients think. In work we did as part of a guideline panel that included parent representatives from the European Foundation for Immunities of Care, parents were fairly clear: if you can avoid a transfusion and there is no clear benefit, they prefer not to give one unless truly necessary.

We see this clinically. Parents care deeply about transfusions. They feel invasive, even when we've obtained consent. If we can safely avoid one, that in itself is an important aim. The key word is safely.

The concern that drove some of the transfusion threshold trials was whether tolerating anemia — keeping hemoglobin low — might compromise cognitive outcomes. That concern did not pan out in the trials. Two large trials following PINT supported the safety of lower thresholds. There is, however, an important caveat highlighted in the Cochrane review: we understand the effects within the thresholds that have been studied. We don't know what happens below those thresholds. If you become more restrictive than what was studied, there is biologic rationale for concern, including as it relates to outcomes like necrotizing enterocolitis (NEC).

Dr. Roger Soll (12:41.917) On the point about parental concerns — at the bedside, families receive input from many directions. When apnea increases or growth stalls, they hear from everyone that maybe their baby needs a transfusion. The idea of setting safe, evidence-based standards does something important: it gives families clarity and a communication framework. They want to understand what will trigger a transfusion.

Dr. Daphna Barbeau (13:34.735) We underestimate how much a transfusion feels like an invasive procedure to families. Even in our own unit, with our best efforts, a baby can sometimes receive a transfusion without the parent being immediately aware — especially babies who have had multiple transfusions. It still registers as a stressor. And that's why a consensus statement matters so much. Roger, how many consensus statements do we actually have in neonatology?

Dr. Roger Soll (14:31.499) Honestly, I don't know the precise number, Daphna. Consensus statements are fascinating to produce — they're difficult and they can still end up highly variable depending on interpretation of evidence and the specific populations being considered. I would guess that systems with social medicine — Canada, Australia, the UK — generate more consensus statements, because there's a push to set standards in the context of resource allocation. The US tends to be more resistant to that.

Dr. Ravi Patel (16:07.086) Guidelines are critically important. We want to practice evidence-based medicine, and we want patients to receive the best available care. But how do you make it effective? You have excellent Cochrane reviews that synthesize the evidence. You translate that into recommendations based on effect sizes, values and preferences of families and clinicians, and the strength of the evidence. That's how the evidence actually reaches patients — otherwise it lives in PubMed and never reaches the bedside.

For centers without current transfusion guidelines, or those that haven't revisited them recently, this is a great moment to refresh them based on the most recent evidence.

Dr. Daphna Barbeau (17:18.078) Before we get into the nuances, can you walk us through some of the actual thresholds so that listeners have a reference point?

Dr. Ravi Patel (17:39.790) The guidelines I'll reference were developed in partnership with a number of organizations, led by colleagues in Europe from the Neonatal Transfusion Network — Moky Deshmukh and Simon Stanworth. They took the evidence from the clinical trials and focused on the outcomes most important to patients and clinicians.

Consistent with the Cochrane review, there is no clear evidence of benefit with higher thresholds. These guidelines make a conditional recommendation for lower thresholds. The approach to simplify it into practice: use a single number rather than a range, and make it easily applicable — including in clinical decision support tools.

For a baby on respiratory support, positive airway pressure for example: transfuse below 11 grams per deciliter in the first week, 10 grams per deciliter in the second week, 9 grams per deciliter in the third week and thereafter. Eleven, ten, nine — first, second, third week. Those are easy mental triggers for rounds. These guidelines are freely available, published in JAMA Network Open, and a good starting point for local guideline development or revision.

Dr. Ben Courchia (19:31.093) We reviewed these PRBC transfusion guidelines on episode 220 of the podcast for anyone who wants to go deeper. Now — many questions from the Grand Rounds chat pointed to the collision between evidence and bedside reality. The trials were conducted in specific populations with specific inclusion and exclusion criteria. How well do those findings apply to babies with encephalopathy, critically ill infants, or infants with pulmonary hypertension? Roger, you were nodding — go ahead.

Dr. Roger Soll (20:34.715) This is always a limitation of trials, even those with broad inclusion criteria. The PRBC trials were in preterm infants — not term infants with hypoxic-ischemic encephalopathy (HIE), not critically ill term infants, not post-operative term infants. The platelet studies appropriately excluded infants with active bleeding — these were relatively asymptomatic thrombocytopenic infants. So both sets of trials have real limits when applied to populations outside the studied cohort.

This can create pushback against guidelines at the unit level — the "what about this one-off case" concern, the feeling that guidelines tie clinical hands. I think that's a misinterpretation of what guidelines are intended to do. As Ravi noted, the same caution about extrapolating to lower thresholds should apply to extrapolating to populations not studied.

Dr. Ravi Patel (22:16.526) One important point Dr. Soll made: understand the population studied. If the intervention was prophylactic platelet transfusion in patients without bleeding, we cannot simply apply that threshold to patients who are bleeding. Once a patient is hemorrhaging and needs hemostasis, management may need to be different.

For red cell transfusions, the recommendations are conditional — meaning individual circumstances may warrant deviation. The trials themselves had built-in exceptions: patients in shock or on 100% oxygen, for example, were managed differently. Guidelines are appropriate for most patients, but for the specific patient in front of you with unique circumstances, thoughtful deviation is appropriate. I would be concerned if clinicians followed guidelines 100% of the time without clinical reasoning.

Dr. Roger Soll (23:55.436) At the unit level, it can be very useful to look at protocol deviations from the guidelines. As in complex respiratory care guidelines and weaning protocols, asking "why didn't you act the way the guideline said?" can be illuminating. It helps identify gaps in the guideline or areas where staff need better communication about intent.

Dr. Ben Courchia (24:24.523) When you look at specific trials — say babies born between 400 and 1,000 grams — the distribution isn't always even. You may have over-representation of more mature infants. How does that affect interpretation and bedside application?

Dr. Roger Soll (25:09.355) That's a very fair question and one we face constantly as we attempt to support infants at 22 and 23 weeks gestation. This applies to essentially every intervention. There are never enough patients in this unique developmental and physiologic state to know with confidence that findings from broader trials are fully applicable. One could argue these populations deserve dedicated study.

Dr. Ravi Patel (25:50.318) Two points. First: does the evidence apply to the baby in front of me? There are nuances worth noting. In the PlaNeT-2 trial — one of the largest platelet trials — enrollment required a cranial ultrasound within six hours so that bleeding status was known before the assigned threshold was applied. Babies with bleeding could be transfused differently. We don't always have that in US practice, where screening ultrasound approaches may differ from Europe. That is an important contextual difference.

Second: when a trial includes babies under 1,000 grams, the mean gestational age often skews slightly higher due to birth numbers. The average treatment effect from a trial is still your best estimate for that population. It is technically very difficult to establish that the effect differs meaningfully between the smallest and the slightly larger babies within that group. Most of the trials don't identify different effects within subgroups of extremely low birth weight infants. Our trials are run precisely because we want trustworthy estimates, and although they have limitations, the average treatment effect remains the most reliable guide we have — more reliable than individual clinical judgment, which can be subject to systematic biases.

Dr. Ravi Patel (28:11.950) There's this belief that we can be precise in identifying exactly how to manage each individual patient. That is very difficult to do. Our experience as neonatologists is unique and valuable, but we should be careful about assuming we know better than the evidence.

Dr. Roger Soll (28:33.099) That's borne out in individual patient data meta-analyses. When we see no effect overall and then try to look at the smallest babies specifically, the general treatment effect tends to hold.

Dr. Daphna Barbeau (29:02.166) I want to address some of the questions about subgroups before we move on. There were quite a few. Many people have a sense that babies with platelet counts between 25,000 and 60,000 tend not to be doing particularly well. The patient you described, Ravi, was at an extreme. But most of our extremely low birth weight (ELBW) infants with moderate thrombocytopenia would be considered quite sick. How do we help clinicians decide which babies fall outside the guidelines?

Dr. Ravi Patel (30:01.806) There are really two groups when thinking about severe thrombocytopenia. First, babies who develop early thrombocytopenia, often in the setting of placental insufficiency and growth restriction — they're not always very ill, but they have a low platelet count. Second, babies who develop later thrombocytopenia from sepsis or NEC — these can be quite ill.

From the data, including additional analyses from the platelet trials looking at babies at higher versus lower risk of bleeding, the best current approach appears to be consistency regardless of perceived illness severity — unless there is active bleeding. Once a baby is bleeding, management needs to change. There has been interesting work from the same group who ran the trial looking at individual prediction of outcomes using a paper published in JAMA, but I think that remains more intellectually interesting than practically actionable at the bedside right now.

Dr. Daphna Barbeau (31:11.782) Another subgroup that came up repeatedly: babies in the first week of life, when intraventricular hemorrhage (IVH) risk is highest. There seems to be a widely held belief that more red blood cells and more platelets reduce IVH risk. But there is actually some data suggesting the opposite — that increased transfusions may increase IVH risk. Can you speak to that?

Dr. Ravi Patel (31:58.808) The IVH and pulmonary hemorrhage concern is exactly why we think carefully about transfusions in that first week. The PlaNeT-2 trial had an important finding: using a higher platelet threshold — keeping the platelet count higher — did not reduce the risk of bleeding, and may potentially have increased it.

One caveat: about 39% of enrolled babies had already received a platelet transfusion before enrollment, which adds some uncertainty about how the first week specifically was managed. But if the rationale for transfusing is to prevent bleeding in a baby with no evidence of active bleeding and no current IVH, that indication is not supported by the available evidence.

Dr. Ben Courchia (33:05.065) A related question coming from the chat: feeding during PRBC transfusion and the risk of transfusion-related gut injury. Listeners are asking whether being NPO (nil per os, or nothing by mouth) makes a real difference and whether there are any recommendations for holding feeds before, during, or after transfusion. Ravi, you've thought about this extensively.

Dr. Ravi Patel (33:37.134) It's an interesting phenomenon in neonatology — we tend to look at what happened recently when a baby develops a complication. There's a natural human tendency to link the two most proximate events. But if we step back and ask: does transfusion cause NEC, the best experimental approach is to give more versus fewer transfusions — which is exactly what the threshold trials did. And the Cochrane review shows no difference in NEC rates between high and low transfusion threshold groups. So at the thresholds that have been studied, we can say there is no difference.

How did that concern translate into withholding feeds? I'm not entirely sure. When we looked at the actual incidence of NEC after red cell transfusion in our 2016 JAMA study, it was approximately 0.5% of babies. As a clinician you might perceive it as far more common, but the data suggests it's quite rare — confirmed by another group as well.

If you're considering withholding feeds because of NEC concern, you have to ask: how many babies would you withhold feeds from to prevent one case — assuming there's even an effect? At 0.5%, the number needed to harm is very large. There are ongoing large trials trying to address the feeding question specifically, but today there is no good data supporting routine feed withholding around transfusions.

Dr. Roger Soll (35:35.283) Based on that back-of-the-envelope calculation, you're suggesting you'd withhold feeds from roughly 200 babies to prevent one case of NEC.

Dr. Ravi Patel (35:45.198) At least that. The point stands.

Dr. Daphna Barbeau (35:54.704) I think part of the confusion is that there does seem to be a link between very low pre-transfusion hemoglobin — severe anemia — and NEC risk. That may cloud our memory of which babies developed NEC after transfusion. Maybe it wasn't the transfusion itself; it was the anemia. Can you speak to that data?

Dr. Roger Soll (36:30.709) Perhaps that's precisely why Ravi cautions against going below the studied lower thresholds. But looking at the trials: the high-threshold group — more frequently transfused — did not have more NEC, and the low-threshold group — allowed to go lower — also did not have more NEC. At the current thresholds, it doesn't appear to be a concern at a population level.

Dr. Ravi Patel (36:55.146) The tricky part of the literature is that you treat anemia with a transfusion, but you also prevent anemia with a transfusion — and disentangling those two is difficult. There is mechanistic plausibility that anemia causes intestinal inflammation through various pathways in the gut, possibly at levels lower than those clinically used, and elegant animal studies support that. But you can't practice based on biologic plausibility alone. You have to return to what the Cochrane review tells us: using the lower thresholds tested in the trials, you are not going to see more NEC.

Dr. Daphna Barbeau (37:56.086) As we wrap up — the guidelines note that for red blood cells, both hemoglobin and hematocrit measurements should be used. Can you speak to that? And is there data on which measurement source — venous, arterial, or capillary — is most accurate?

Dr. Ravi Patel (38:36.494) From a transfusion medicine perspective, hemoglobin tends to be preferred — it's what the adult transfusion medicine community uses. But hematocrit is widely used in neonatology and I understand why. Some trials have differentiated capillary versus non-capillary specimens and accounted for potential threshold differences. That adds real complexity to clinical application.

My pragmatic view: measure what you're able to measure in a reasonable way and apply the guidelines, understanding there will be some analytic variability. You'll sometimes see two specimens drawn in close succession give slightly different values — that variability is real. I would not want clinicians to feel they need arterial lines or invasive sticks solely to more precisely measure hemoglobin. Apply the guidelines using your available measurement and use clinical judgment alongside.

Dr. Ben Courchia (39:37.331) One final question from the chat: volume of transfusion. If we've established thresholds, what about how much we transfuse? At the time of recording, 15 mL/kg is standard for PRBCs and approximately 10 mL/kg for platelets. Is there reason to think more carefully about volume and infusion duration?

Dr. Ravi Patel (40:23.406) The two largest threshold trials used different volumes: the TOP trial used 15 mL/kg, and the ETTNO trial used 20 mL/kg. That may partly explain why the number of transfusions differed slightly between trials, since larger doses extend the interval before the next transfusion threshold is reached.

We need to be careful about large volume transfusions given rapidly — there is a risk of transfusion-associated circulatory overload, and we have concerns about the hemodynamic effects of rapid volume shifts in tiny infants.

This is particularly relevant for platelets. If you think about the equivalent dose in adults, we tend to give substantially more platelets to babies on a weight-adjusted basis. The PlaNeT-2 trial used 15 mL/kg — slightly higher than what US surveys suggest is typical practice — and didn't specify infusion duration, so it could have been given as a bolus. That raises an important question: was the observed signal from the higher threshold, the greater platelet volume, or the rate of infusion? Slowing down the infusion and using a lower dose might yield different results, and that remains an area of uncertainty.

In US practice, most centers use approximately 10 mL/kg typically over one to two hours — a slower infusion to minimize hemodynamic effects.

Dr. Roger Soll (42:08.201) Well summarized.

Dr. Ben Courchia (42:10.519) The last question: local implementation. This conversation makes clear that transfusion thresholds remain complex — there are many variables, and ongoing research, including the NeoPLAT trial for platelet transfusion in the US, will continue to shape the field. For centers that want to develop or revise guidelines, what is your practical recommendation?

Dr. Ravi Patel (43:18.542) You have to wear two hats — the researcher, who lives in uncertainty and wants more evidence, and the clinician, who has to make decisions during rounds and cannot function in constant uncertainty.

My recommendation: if a center is trying to establish practice, others have already done the work of assembling expert panels and synthesizing the evidence. Use that. The Cochrane reviews and the published guidelines I described are out there — don't try to redo that work from scratch if you don't have to. Then gauge the values and preferences of your clinicians locally, address specific concerns, and bring in outside perspectives where there's persistent worry.

Transfusion decisions are particularly amenable to informatics tools — electronic health record alerts and decision support can make thresholds easily applicable in daily practice. These are frequent, repetitive decisions, and if you can embed the guideline into the EHR workflow, it becomes measurable. When deviations occur, you can ask why — and as some qualitative research out of Utah has shown, the reasons are often rational: a clinician knew a baby was trending toward the threshold and anticipated the transfusion. Those are thoughtful clinical decisions, not guideline failures.

Dr. Roger Soll (45:17.001) You also need the right people in the room. This will not work without expertise from pathology and the blood bank, bedside nurses, and parent representation. And you need to push people as far as the evidence supports.

Years ago when the PINT trial first came out, our routine practice was well above even the high threshold. In those early discussions, I could only convince people to adopt the high threshold — which was itself a radical change. Now, surveys show clinicians are no longer choosing the high level, but they're not fully at the low level either. Change is iterative. Get the right people in the room, address resource issues, and center the values of both staff and families.

Dr. Ben Courchia (46:19.319) A great note to conclude on. Thank you, Dr. Ravi Patel and Dr. Roger Soll, for joining us on this second episode of On with Vaughn. I invite the audience to register for the next VON Grand Rounds webinar taking place May 6th from 3 to 4 PM Eastern: Evidence to Practice — Erythropoiesis Stimulating Agents. You can register at vtoxford.org. Roger, Ravi — it was a pleasure.

Dr. Daphna Barbeau (46:49.938) A hot topic. We'll be ready.

Dr. Roger Soll (47:03.243) Thank you so much.

Dr. Ravi Patel (47:04.619) Really enjoyed it. Thanks, everyone.

Dr. Ben Courchia (47:06.455) Thank you.