Improving Outcomes in Pediatric Orthopedic Surgery with Dr. Brett Lullo

With advanced surgical training and a background in computer science, Dr. Brett Lullo brings a unique set of skills to Lurie Children’s pediatric orthopedic programs and is dedicated to improving outcomes for children with both common and challenging spinal conditions. In this episode, Dr. Lullo describes many of the conditions he treats and explains his passion for bringing new technologies, data science and machine learning into surgical pediatric orthopedic care.

Show Notes

• Dr. Lullo was drawn to pediatric orthopedics because he was compelled to work with children with significant disabilities and offer care that can improve the quality of life for young patients.  
• For children with profound physical disabilities, a surgery may help them sit upright, engage with family, reduce pain, and regain daily function.
• He is also dedicated to working with typically-developing children who are diagnosed with scoliosis, including idiopathic, congenital, syndromic, and neuromuscular scoliosis.  He is especially interested in treating early-onset scoliosis (before age 10), where ongoing growth and lung development make treatment decisions more complex and often require long-term, multi-step care planning.
• He says his team at Lurie Children's is highly coordinated and supports families from their first clinic visit through the operating room and recovery. They objectively try to measure what families value most in treatment decisions, so surgical plans can better match patient goals and improve outcomes.
• Another goal of Dr. Lullo’s and his team is to reduce repeated surgeries. Using his background in computer science, Dr. Lullo is currently conducting research using multiple machine learning models to predict which early-onset scoliosis patients are at risk for unplanned return to the operating room.

Transcript

[00:00:00] Erin Spain, MS:
This is Precision Perspectives on Children's Surgery from Ann & Robert H. Lurie Children's Hospital of Chicago. I'm your host, Erin Spain. On this podcast, we introduce you to surgeons at one of the country's most renowned children's hospitals to find out how they are transforming pediatric medicine. Children with spinal issues such as severe scoliosis often require treatment through surgery. Lurie Children's Division of Orthopedic Surgery and Sports Medicine is one of the top pediatric orthopedic programs in the country and works with the hospital's scoliosis program to provide specialized surgical care for growing children and teenagers with this condition. Dr. Brett Lullo, a pediatric orthopedic surgeon at Lure Children's has expertise in caring for children with spinal deformities, including early onset congenital neuromuscular and adolescent idiopathic scoliosis, as well as fractures, growth plate injuries and limb deformities. What makes his perspective unique is a combination of advanced surgical training and a background in computer science, which drives his passion for bringing new technologies, data science and machine learning into pediatric orthopedic care. We welcome him to the show today to talk about his approach to care and how he hopes to improve outcomes for children with challenging spine conditions. Welcome to the show,

[00:01:29] Dr. Brett Lullo:
Thank you for having me.

[00:01:30] Erin Spain, MS:
Well, It's wonderful to have you here, and your background is so interesting. So tell me how you use your knowledge in computer science to enhance pediatric spine care.

[00:01:40] Dr. Brett Lullo:
I think from the outside looking in, it does look like an interesting path to medical school, but for me it always felt pretty traditional. I knew I wanted to go to medical school even from the time I was in high school. And I had a great opportunity in high school actually here in the Chicagoland area to do a lot of computer science courses. And that really sparked my interest in that field. And in college I was able to major in computer science and I had a number of great opportunities to do internships in the medical field and really utilize the computer science tools in pretty innovative ways that have. Honestly inspired a lot of the work that I'm still doing today.

[00:02:17] Erin Spain, MS:
You've witnessed some severe disabilities in children with spinal deformities. Can you talk about that and share some of those limitations with me that these patients experience and how that impacts their lives?

[00:02:28] Dr. Brett Lullo:
So it's working with children with those disabilities that really got me into medicine in the first place. My mom was a special ed teacher and that really fueled my passion for medical school. And I've gotten into my orthopedic training and specifically my pediatric orthopedic training. It's working with that population and helping children overcome those limitations that has really fueled my desire clinically and on the research end of things. Some of the main disabilities that the children I treat have are, as simple as even being able to sit up straight in their wheelchair and interact with the world around them. So when we're deciding to do surgery and planning for surgery. We always keep that at the forefront. If we can help a child to even sit up and better interact with his family, with their peers, that is such a quality of life improvement for these children. On the other spectrum of things, I also treat children with back pain and being able to, relieve that back pain or straighten the spine and let people get back to the functioning that they're used to really goes a long way for our typically developing population as well.

[00:03:23] Erin Spain, MS:
Let's talk specifically about scoliosis. So you treat early onset congenital neuromuscular and adolescent scoliosis. For families listening. How do these forms of scoliosis differ and why does early diagnosis matter so much?

[00:03:39] Dr. Brett Lullo:
Scoliosis is any curve to the back over 10 degrees, and that can occur at any age. Most commonly we see this in teenage patients, but it can. Occur even at birth. So in my practice, I get to treat children as early as a few months old, all the way through their teens and into their twenties. There's a few different types of scoliosis. The most common one we treat is idiopathic scoliosis. So it just occurs in healthy, typically developing children. But there are other forms of scoliosis as well, so related to neuromuscular issues. Syndromes scoliosis that's related to congenital issues of the spine that they're born with. And these can occur anytime throughout your development. If it happens before age 10, we refer to that as early onset scoliosis. And those are the patients that really give us a lot of the challenges and force us to, really work together to deliver really high quality care, to see them not only through one surgery, but multiple surgeries throughout their lifetime.

[00:04:33] Erin Spain, MS:
Is that because they're growing so rapidly at that age and the spine keeps changing as you're doing these surgeries?

[00:04:40] Dr. Brett Lullo:
Yeah. As opposed to a teenager, an adult who's pretty much done growing these kids with early onset scoliosis have so much growth. Potential left. And it's not even just a height thing, our spinal height also determines our pulmonary function. So we're ensuring that these children have maintained growth throughout their early childhood years to really support their pulmonary function so they can do all the things that they want to do, once they're a teenager and an adult. So every treatment we do, we're keeping growth at the forefront to really allow them to have a high quality of life.

[00:05:10] Erin Spain, MS:
You really try to keep the patients and their families at the center of what you do. Tell me how you approach conversations with families who may be anxious about a new diagnosis or a potential surgery, especially for scoliosis.

[00:05:24] Dr. Brett Lullo:
Yeah, I think that's one of the most valuable things that we can do as, clinicians and surgeons, is to, one, establish a relationship with a patient and the family because, patients and families come to you at different levels of, acceptance and willingness to understand and move forward with treating their diagnoses. You can have a family who has been told from a few different surgeons that they need surgery and they're coming to you and they just want to get this done. You can also treat families that are hearing about this diagnosis for the first time and can't truly wrap their head around about what exactly the treatment's going to entail. So you have to meet the families where they're at and sometimes that takes. Telling them, come back in three months and we could discuss this a little bit further. Sometimes, surgery may be the gold standard answer, that family needs a little bit more time with bracing or other non-operative therapies. So we really get to tailor the treatments that we do to the needs of each patient and their family.

[00:06:15] Erin Spain, MS:
What do these surgeries that you perform on patients with scoliosis look like you mentioned the timing already? That can be more difficult when they're younger, but why is timing so important with these surgeries?

[00:06:26] Dr. Brett Lullo:
Yeah, so in scoliosis surgery, there's two main types of surgery. The most common one we talk about is a posterior spinal fusion, so that's where we make an incision in the back of the spine, and then we put screws into each of the vertebrae of the spine. In the evolved part of the scoliosis, we connect those screws with rods and we can use those rods to help straighten out the spine and give 'em the chance to heal or fuse. In younger kids with early onset scoliosis, as we talked about before, we can't fuse the spine because we want them to continue to grow. So the mainstay of treatment for these very young kids is what we call growing rods. So they're rods that we can put to temper, rarely stabilize the spine, but allow a child to continue to grow. And in the past, that required multiple surgeries, but a lot of newer technologies allow us to lengthen these rods in our clinic with some pretty fancy, interesting technology and that saves these patients and families, a lot of surgeries, a lot of trips to the hospital, and overall improves quality of life.

[00:07:20] Erin Spain, MS:
Tell me about those in clinic appointments where you're using lengthening rods? Is that what you said?

[00:07:25] Dr. Brett Lullo:
Yeah, so we refer to 'em as magnetically controlled growing rods. So the rods that we implant have an induction motor inside of them, we can use an external, remote control we call it, but it basically creates a magnetic field and spins the motors inside these rods to help them to lengthen. And children tolerate this very well in clinic with a little bit of iPad anesthesia as we like to call it.

[00:07:46] Erin Spain, MS:
iPad Sure comes in handy for a lot of things, doesn't it?

[00:07:49] Dr. Brett Lullo:
Very much so.

[00:07:50] Erin Spain, MS:
Teamwork is something that you say is so important in the clinic and in the operating room. Talk to me about what the team looks like for a child with scoliosis at Lurie Children's and why that collaborative model is so important.

[00:08:02] Dr. Brett Lullo:
It really is the most important thing. And I tell families that as we're talking about surgery, that it's not just me taking care of your patient. It really is an entire team. And I think here at Lurie Children's, we're so lucky to have such a rockstar team involved in each child's care that starts in the clinic. I have a wonderful nurse coordinator, Nelly, who helps communicate with these families and set them up for success leading up to the surgery. We have a team in our office, my administrative assistant, Delia helps reach out to these families, clear insurance hurdles that I know are way heavily on these families. And then, on the day of the surgery we had about a dozen people in the operating room with me. So anesthesiologists, scrub techs, circulating nurses, the residents and fellows that we work with. We have perfusion specialists, neuromonitoring specialists, we have imaging and implant specialists. So all of these people need to come together at the right time in the right place to really ensure high quality care for each patient who's undergoing spinal surgery.

[00:09:00] Erin Spain, MS:
Something that a lot of parents want to know right away is, how long are these procedures going to last and what is recovery going to look like? How many days in the hospital recovery at home? Can you talk to that a little bit?

[00:09:10] Dr. Brett Lullo:
Yeah, so for our typical idiopathic scoliosis patient who's having a posterior spinal fusion, that's about a five hour procedure. And they come in first thing in the morning, it usually ends sometime in the middle of the afternoon and patients spend about two or three days in our hospital. And that, by the work of many team members has come down. It used to be that patients were in the hospital for a week plus and now many of my patients go home two days after surgery. I see patients back in three weeks and make sure they're doing well. We get some x-rays at six weeks and, by six weeks I let patients get back to light activities and most. Patients by that point are really interested in getting back. We do protect people for about four to six months from higher impact activity as the bone around the spine and around the implants we put in is still healing. But, I tell patients it's my goal to get people back to whatever they wanna be doing, whether that's swimming, gymnastics, volleyball, football, wrestling, I want them to get back to everything that they loved doing before they had to worry about this spine surgery.

[00:10:06] Erin Spain, MS:
You are one of several pediatric orthopedic surgeons at Lurie Children's who have experience doing these complicated spine surgeries. Can you talk about that and why that might give families some confidence going to see you in clinic, that, this is something that your team has a lot of expertise in.

[00:10:24] Dr. Brett Lullo:
I'm the newest spine surgeon here at Lurie Children's Hospital having joined three years ago but I joined a team of my senior partners who have been really driving spinal care in Chicago for the past three or four decades. And they've done it at such a high level. And I think that, reputation for great spine care was really a product of everything that they've been doing over these past 30 or 40 years. And it's really nice, in the clinic, to be able to calm families by telling them, this is a big, life event for them. This is a very complex thing that they are going through. But, for us, we do so many of these, we have such great volume that some of these surgeries are truly just another day for us. And because of that, we can objectively provide the best care possible for these patients.

[00:11:04] Erin Spain, MS:
Let's shift a little bit and talk about your research. As you mentioned earlier, you have this background as a computer scientist. You're able to really infuse that into your work at Lurie Children's. Your recent study in spine deformity used 10 different machine learning algorithms to predict which early onset scoliosis patients might require an unplanned return to the operating room. Tell me a little bit about this study. What inspired you to apply machine learning to this question?

[00:11:32] Dr. Brett Lullo:
When I was in college, I was one of 36 computer science majors at Princeton in my graduating class. And that's changed and just as we've seen computer science and artificial intelligence machine learning become so prevalent in our society. The number one major at Princeton now with over 250 students per year is computer science. People are realizing that they can utilize this in whatever field that they want to go into. But when I did it in 2011, it was a pretty unique idea and I think people were confused that I was going to medical school but wanted to do computer science. While I was still in medical school, I took some online courses through Stanford to actually learn about the science of machine learning and artificial intelligence. And with that, I started doing work on different research projects throughout residency trying to integrate machine learning to predict outcomes for patients. And, this was my first pediatric spine project that I tackled and I started doing it while I was still in fellowship with the Children's Hospital of Philadelphia team and I was able to assess a lot of different patient radiographic and surgical factors that go into the surgeries that we do and predict which patients were gonna have an unplanned return to the operating room, so have to return to the operating room before we expected that they might. Whether it's due to an infection or a hardware issue, and by looking at a subset of these, patient radiographic and surgical factors, with pretty good accuracy, we can use machine learning to predict, which patients are at higher risk and which patients, we less expect to have issues throughout their treatment course.

[00:12:57] Erin Spain, MS:
This is something that. Been virtually impossible to do with humans, right? Going through all this data, this is something where machine learning can really make a difference.

[00:13:08] Dr. Brett Lullo:
Yeah, it's one of those things where, you know, we as orthopedic surgeons can look at an x-ray and pull out. Factors from that image that we see, with pattern recognition and not necessarily having to think about it systematically, but just by recognizing things that we see, machine learning and artificial intelligence do the same thing. They can pull out these kinds of hidden relationships that are not obvious to see from someone just looking at it by themselves or even, the traditional statistical methods that we used previously. So we can pull out some of these hidden relationships from the data that we weren't able to previously, and use that to help better inform the care that we're delivering.

[00:13:40] Erin Spain, MS:
Any surprises to you in this study? Any of these predictive factors? That was surprising to you?

[00:13:46] Dr. Brett Lullo:
Honestly the answer is no. It really just informed the trends that we've been seeing over the past few years in early onset scoliosis care, which is that if we wait until children are a little bit older. Gain a little bit more weight and height and we use methods that require less surgery. These patients are gonna have less complications overall. So this really confirmed a lot of the things that we see and the trends that we're seeing. If at all possible, we try to delay these surgeries nowadays, and if we can avoid using growing rods altogether and get patients to a size and age where they could just have a definitive spinal fusion, that's also a great way to improve quality of life outcomes and avoid complications.

[00:14:23] Erin Spain, MS:
Why was there a rush to surgery in some cases in the past versus this waiting until the children were a little bigger?

[00:14:31] Dr. Brett Lullo:
Yeah, so even in some of these young children, as early as three, four years old, you can have curves that grow to 70, 80, 90, a hundred plus degrees. So really it is about. Trying to prevent these curves from getting too large because at some point they truly become unmanageable. Growing rods were a wonderful opportunity for us to intervene a little bit earlier and decrease the size of these curves so we can better treat them later in life. But I think, just as we try to hone our operative skills, we've also been able to hone our non-operative skills over the past decade or so, and use simple things like bracing and casting to avoid surgeries and improve outcomes.

[00:15:05] Erin Spain, MS:
So, are you able to use this predictive model now at Lure Children's or is this something that we're going to be seeing in the future be a part of care?

[00:15:13] Dr. Brett Lullo:
So I've since expanded upon the project that I did initially and has since been published in the Journal of Pediatric Orthopedics, and now I'm looking at an international multi-center cohort of these same patients. We're looking specifically at growing rod surgeries and patients from all over the world, and we're creating models that can be generalized to a larger population. And the goal is to set up a web-based calculator that anyone has access to. I've actually just recently won a grant from the Pediatric Spine Foundation to help build an API to help build some more of these tools that everyone can have access to. And the idea is that, one day this is gonna be. Built right into the systems that we already use. So for us, that's our electronic medical record. So when I see a patient in the clinic and we get x-rays, hopefully these tools can alert me right away without having to think too much or do too much extra work, at some of the complications that these children might have and other factors that may not be easy to calculate on our own.

[00:16:06] Erin Spain, MS:
Why is it important to you that this is an international project and that this is open access for anybody someday to be able to use these tools?

[00:16:14] Dr. Brett Lullo:
The first part of that is that not every child's the same, even within early onset scoliosis, every patient I treat is different. They come from different backgrounds, they have different etiologies as we talked about previously. They have different size curves and they have families that are willing or not willing to do different procedures. So we have to factor all of that in, especially when we're creating these machine learning and artificial intelligence models. Because if you don't factor that into the input, then the. The outputs that we have are gonna be biased. And, I really want these tools to be accessible to surgeons all around the world because, we have a wonderful infrastructure here at Lurie Children's, the average doctor, other places in the world or even other places in our country don't have access to that same infrastructure. So I wanna be able to package up these tools that we're creating so anyone has access to 'em. Any patient across the world can benefit from them.

[00:16:59] Erin Spain, MS:
Looking ahead, how might machine learning data science and other emerging technologies support better outcomes for children with complex spine conditions?

[00:17:07] Dr. Brett Lullo:
I was recently asked to be a faculty member at the recent Scoliosis Research Society meeting, and there was an entire morning dedicated to just that question is how are artificial intelligence and machine learning going to affect the care we provide to our patients in, spine deformity surgery and we had the wonderful opportunity to hear people who are using in all facets of care. Here at Lurie Children's we're starting to use some artificial intelligence scribes in our clinics. And, that helps us to stay more focused on the relationships that we're having with the patients, but also not lose out on the details that pop up in the clinic. We're using machine learning techniques to start planned surgery. So, automated measuring of scoliosis angles automated planning of some of the surgical corrections that we have to do. These are tools that are, I would say in their infancy right now, but are really being developed and I think one day are gonna help us to deliver better quality care for each patient and individualized to each patient.

[00:17:59] Erin Spain, MS:
Lurie Children's is a leader in pediatric orthopedic surgery and something that we've talked about quite a bit is that families and patient preference is very important part of this, when you're thinking about which surgery you're going to do or when to do surgery, and you've recently worked on a project that looks specifically at the preference of families and patients, can you tell me about that?

[00:18:21] Dr. Brett Lullo:
In that study that you referenced we specifically tried to objectively assess what is important to families because so much of what we do in orthopedics, there is not one right answer. And as we discussed, every patient is different and every family brings with them their own histories and desires for what their child needs and what their family is ultimately gonna benefit from. So, we use some advanced, actually marketing technologies where we partnered with a company that usually focuses on marketing preferences in the general population, and we were able to create a 12 question survey where we ask specific questions about the surgeries we do and the outcomes that we expect. And in a really objective, detailed manner, we are able to assess the objective preferences of, family of young children that may have to decide between having a growing rod surgery or a definitive posterior spinal fusion. And, the outcome of that is that we can better do the surgeries and the treatments that we're offering. To each specific patient and each specific family. Because, if we decide that one treatment is correct, but you know it, the family doesn't feel a hundred percent on board with it, then the outcomes are gonna be less than ideal. So if we can match our desired treatment to the family's preferences and expectations, that patient is gonna do better in the long run.

[00:19:33] Erin Spain, MS:
What message would you like to leave listeners with today about the kind of experience they can expect when they come to see you at Lurie Children's?

[00:19:41] Dr. Brett Lullo:
Because of the amazing resources that we have here at Lure Children's, both in the research field, but also clinically, families truly can expect to get the best care and care that is specifically tailored to their child.

[00:19:52] Erin Spain, MS:
Thank you so much Dr. Brett Lullo for coming on the show, talking about your incredible research and the procedures offered here at Lurie Children's. Appreciate your time today.

[00:20:02] Dr. Brett Lullo:
Yeah. Thank you for having me. This was wonderful.

[00:20:04] Erin Spain, MS:
For more information, including how to make a referral or an appointment, visit luriechildrens.org.