Selected Podcast
Clinical Characteristics of Fractures in Pediatric Patients Exposed to PPIs
Dr. Nathan Fleishman discusses the clinical characteristics of fractures in pediatric patients exposed to PPIs.
Featured Speaker:
Learn more about Nathan Fleishman, MD
Nathan Fleishman, MD
Nathan Fleishman, MD is a Pediatric Gastroenterology Fellow at Children's Mercy Kansas City.Learn more about Nathan Fleishman, MD
Transcription:
Clinical Characteristics of Fractures in Pediatric Patients Exposed to PPIs
Dr. Michael Smith (Host): There are increasing concerns regarding proton pump inhibitors and risk of fractures in adults, but few studies have actually evaluated this risk among pediatric patients. This is Transformational Pediatrics, the podcast from Children’s Mercy. I'm Dr. Mike. Let’s talk with Dr. Nathan Fleishman. He’s a pediatric gastroenterology fellow at Children’s Mercy Kansas City. Dr. Fleishman, what are the known concerns regarding proton pump inhibitors and in general the risk of fractures?
Nathan Fleishman, MD (Guest): Sure. So in adult populations, people have been concerned about the side effects related to PPIs and especially fractures really for some time. For the better part of 15 years this has been investigated. In pediatrics, we prescribe the medications a lot as well. So it’s kind of a trickle down effect. People are wondering is there risk of fractures in pediatric populations. Well in adults, this all kind of stemmed from more post-menopausal women and the risk of hip fractures. The initial study that I'm aware of by Vestergaard in 2006 looked at this risk in adults and specifically PPI exposure within a year and did find that there was some association with increased risk of fractures. Several studies have been done since then including med analyses have kind of confused this. In terms of children, there's really been limited literature. So only in the past kind of three or four years have there been studies kind of looking at this. Of those studies, most of them have either looked at older children or younger children, not all children. They have some conflicting results.
Host: I know this is not maybe well known or I know a lot of people are looking into this, but what exactly is it about the PPI that is associated with fractures? Is it simply just lost of mineral? Are people becoming low in magnesium? What do you think is going on with the PPIs?
Dr. Fleishman: That is a great question. The short answer is nobody knows. Like you said, there could be a mineral component. Some people postulated more of a hormonal component or a direct effect on cells that are involved in bone turnover. So nobody knows the exact mechanism. Additionally you'll see some differences in adults and pediatrics in terms of maybe how they handle medications, how they metabolize them. So that plays into them to. The other question is is it just an effect from the underlying disorder that you're treating with the medication? So all these are good questions, but I'm not aware that anyone knows this answer yet although there are many theories out there.
Host: Yeah, I know. I just kind of wanted to hear what you thought. So now going back to the pediatric patient, obviously you mentioned there was a trickle down effect, right, as more and more kids are being prescribed PPIs. That is a good question, right. Is there that same fracture risk? So tell us about what you actually studied and what was the design of what you did?
Dr. Fleishman: Yeah. So our study what our main goal was because there's so limited literature out there is basically to see does this relationship even exist in pediatrics. We didn’t set out to determine cause or anything like that. We basically wanted to look out and say hey, if a child is exposed to a proton pump inhibitors do they have an increased risk of fracture? Do they seem to have more fractures? So our study was a little bit unique in that we used a database called the FIZZ database made up of more than 50 children’s hospitals across the United States. So we have access to pooled data which gives us a lot of numbers and encounters in various settings. So emergency departments, in patient settings, surgical settings, and observation settings and we track children. So we look to see does this child have exposure to PPI based on kind of what they were given in that hospital encounter. Then we track them over a two year period of time to see if they had an incident of fracture being coded. So we went back and looked to see the patient’s that were exposed to PPI, what percentage of them had fractures. Then we did a control group by taking children who matched up pretty closely in terms of demographic fractures and location of their encounter and we tracked them over two years and looked to see if they weren’t exposed to PPI on the day, what was their fracture risk?
Host: Yeah. So very interesting right. So here you're simply asking—The question is is this same association with PPIs that we see in adults with fractures, is that same association happening in kids? So straight forward and nice. I like the setup. What were the results?
Dr. Fleishman: So we tracked over 32,000 counters where PPI was documented over this two year period of time. We tracked the same number of what we call non-exposed patients—patients that didn’t receive PPIs—during that period of time. What we found was interestingly that patients that had documented exposure to PPIs were in fact for higher risk for having a fracture in that period of time. It as an odds ratio of 1.2:1 which basically in kind of layman’s terms means that our cohort documented that PPI exposure was documented with a 1.2 times greater likelihood of having a fracture during that two year period of time.
Host: Wow. Were you surprised by those results?
Dr. Fleishman: You know, honestly I was. When I was setting up this project, I wouldn’t be surprised if we didn’t see any relationship at all and there's no increased, which I thought would still be useful because a lot of parents want to know what are the risks taking this medication. If I could reassure them from a fracture standpoint there doesn’t appear to be an increased risk then that would be great as well. So I wasn’t afraid of getting a negative result, but yeah I was a little surprised.
Host: So yeah. I'm sure you documented this. What types of fractures were most common among the PPI group?
Dr. Fleishman: So in both groups—the PPI and non-exposed group—we still see upper extremity fractures predominate. As you can imagine, most children—if you look at population data—tend to have arm fractures, wrist fractures, that type of thing during childhood. So that was consistent with what we found in both groups. If you broke both groups down and looked at other locations of fractures, we also found that the PPI exposed group was more likely to have lower extremity fractures, spinal fractures, rib fractures as well which I didn’t anticipate that. So for whatever reason, there does appear to be a propensity to have fractures in other locations if your exposed to a PPI based on our data.
Host: Yeah. That’s very interesting right. So there aren’t a lot of studies in the pediatric population with PPIs, but I think there are some. How was yours different from maybe some of those other studies?
Dr. Fleishman: Yeah. So really I think to date there’s three to four other studies that I know about. I think what separated ours is one is the age range. Other studies tend to either focus on older children or younger children but not necessarily both. So we tried to include all because we felt like that was useful data to see how that effects all groups if you combine them. The other thing we did was the location of fracture. A more recent study has looked at that which got published around the same time as ours, but prior to that nobody had really talked about if there is an increased risk of fracture, where are the fractures? Are they the same we’d expect or are they more kind of pathologic fractures? So I think those two separate us out a bit. Then the numbers. I mean having access to the FIZZ database and the power of getting over 32,000 encounters is a bit unique.
Host: That’s pretty good. That’s pretty good. All studies have limitations, right. So what were some of the limitations of your study?
Dr. Fleishman: Yeah. So definitely limitations. Big data’s kind of overarching, like looking at the big picture. You don’t get some of the granular data like more duration effects and things like that. The biggest weakness I see in it and hopefully this will get better over time as we use medical records a little more efficiently and they become more accurate is that this isn’t a closed system. For instance, a child who’s captured in the FIZZ database, yes they may be at a large children’s hospital there. They can go outside of this little bubble of what we’re able to capture. They can go to their pediatrician if they have a finger fracture. They can go directly to the orthopedic doctor if they have an arm fracture. So we’re only capturing this group in these settings. Additionally with medications, as you know, a lot of these are prescribed outside of these particular settings or you can go buy a proton pump inhibitor and histamine blocker over the counter. So I wouldn’t be able to necessarily capture those. Hopefully, those would, if anything, underestimate our exposure but probably effect both groups equally we would assume.
Host: Yeah. So you mentioned one of the limitations is duration of exposure to the PPI, right? You didn’t really dig down that deep. Is that something you're interested in now? Now that you've shown this association to increased fracture versus that control, is that the next step for you? Dosing of PPI, how long they should be on PPI. Is that kind of where you want to go?
Dr. Fleishman: I think so. Whether I’ll be able to do it in the same manner remains to be seen or whether you have to use a different database or more granular database. I think those are all legitimate questions. Because if we’re implying that a medication has a certain side effect well my next question is I think it’s the medication they need. If they need it, what can I do to decrease this risk? That becomes, exactly as you mentioned, a dose or a duration question.
Host: Yeah, excellent. So in summary, what would you like pediatricians, GI specialists, community doctors, what would you like for them to know about PPIs in kids?
Dr. Fleishman: So I think the first think I would want to emphasize is that our data shouldn’t be interpreted necessarily as PPIs causing a fracture. That’s not what we set out to do. There seems to be an association and why that is, as we discussed before, I don’t know. I don’t think anyone knows at this point. Secondly I would think overall these medications appear to be safe as a whole. There's an increased risk of fracture, but if you look at the overall numbers it’s a small increased risk, but nonetheless it’s there. I think they're effective medications and they can be very helpful to a lot of patients. We should still be inclined to use them if we need it, but for all providers and patients alike we should always be considering whether a patient still needs to be on a medication and should never leave a medication on their list without truly thinking over time do they still need this? Yes or no. If they don’t, how can I safely get that medication off of their list and still make sure that they're being treated appropriately. I think it also goes to show that like I said there is an increased risk, but I think parents should be reassured that this is a minimal increase. Therefore from a fracture standpoint, if your doctor thinks you need this I think it’s still a safe decision.
Host: Excellent summary. That’s Dr. Nathan Fleishman. He’s a pediatric gastroenterology fellow at Children’s Mercy Kansas City. Thanks for checking out this episode of Transformational Pediatrics. Please visit childrensmercy.org to get connected with Dr. Fleishman or another provider. If you found this podcast helpful, please share it on your social channels and be sure to check out the entire podcast library for topics of interest to you. Be sure to check back soon for the next podcast. I'm Dr. Mike. Thanks for listening.
Clinical Characteristics of Fractures in Pediatric Patients Exposed to PPIs
Dr. Michael Smith (Host): There are increasing concerns regarding proton pump inhibitors and risk of fractures in adults, but few studies have actually evaluated this risk among pediatric patients. This is Transformational Pediatrics, the podcast from Children’s Mercy. I'm Dr. Mike. Let’s talk with Dr. Nathan Fleishman. He’s a pediatric gastroenterology fellow at Children’s Mercy Kansas City. Dr. Fleishman, what are the known concerns regarding proton pump inhibitors and in general the risk of fractures?
Nathan Fleishman, MD (Guest): Sure. So in adult populations, people have been concerned about the side effects related to PPIs and especially fractures really for some time. For the better part of 15 years this has been investigated. In pediatrics, we prescribe the medications a lot as well. So it’s kind of a trickle down effect. People are wondering is there risk of fractures in pediatric populations. Well in adults, this all kind of stemmed from more post-menopausal women and the risk of hip fractures. The initial study that I'm aware of by Vestergaard in 2006 looked at this risk in adults and specifically PPI exposure within a year and did find that there was some association with increased risk of fractures. Several studies have been done since then including med analyses have kind of confused this. In terms of children, there's really been limited literature. So only in the past kind of three or four years have there been studies kind of looking at this. Of those studies, most of them have either looked at older children or younger children, not all children. They have some conflicting results.
Host: I know this is not maybe well known or I know a lot of people are looking into this, but what exactly is it about the PPI that is associated with fractures? Is it simply just lost of mineral? Are people becoming low in magnesium? What do you think is going on with the PPIs?
Dr. Fleishman: That is a great question. The short answer is nobody knows. Like you said, there could be a mineral component. Some people postulated more of a hormonal component or a direct effect on cells that are involved in bone turnover. So nobody knows the exact mechanism. Additionally you'll see some differences in adults and pediatrics in terms of maybe how they handle medications, how they metabolize them. So that plays into them to. The other question is is it just an effect from the underlying disorder that you're treating with the medication? So all these are good questions, but I'm not aware that anyone knows this answer yet although there are many theories out there.
Host: Yeah, I know. I just kind of wanted to hear what you thought. So now going back to the pediatric patient, obviously you mentioned there was a trickle down effect, right, as more and more kids are being prescribed PPIs. That is a good question, right. Is there that same fracture risk? So tell us about what you actually studied and what was the design of what you did?
Dr. Fleishman: Yeah. So our study what our main goal was because there's so limited literature out there is basically to see does this relationship even exist in pediatrics. We didn’t set out to determine cause or anything like that. We basically wanted to look out and say hey, if a child is exposed to a proton pump inhibitors do they have an increased risk of fracture? Do they seem to have more fractures? So our study was a little bit unique in that we used a database called the FIZZ database made up of more than 50 children’s hospitals across the United States. So we have access to pooled data which gives us a lot of numbers and encounters in various settings. So emergency departments, in patient settings, surgical settings, and observation settings and we track children. So we look to see does this child have exposure to PPI based on kind of what they were given in that hospital encounter. Then we track them over a two year period of time to see if they had an incident of fracture being coded. So we went back and looked to see the patient’s that were exposed to PPI, what percentage of them had fractures. Then we did a control group by taking children who matched up pretty closely in terms of demographic fractures and location of their encounter and we tracked them over two years and looked to see if they weren’t exposed to PPI on the day, what was their fracture risk?
Host: Yeah. So very interesting right. So here you're simply asking—The question is is this same association with PPIs that we see in adults with fractures, is that same association happening in kids? So straight forward and nice. I like the setup. What were the results?
Dr. Fleishman: So we tracked over 32,000 counters where PPI was documented over this two year period of time. We tracked the same number of what we call non-exposed patients—patients that didn’t receive PPIs—during that period of time. What we found was interestingly that patients that had documented exposure to PPIs were in fact for higher risk for having a fracture in that period of time. It as an odds ratio of 1.2:1 which basically in kind of layman’s terms means that our cohort documented that PPI exposure was documented with a 1.2 times greater likelihood of having a fracture during that two year period of time.
Host: Wow. Were you surprised by those results?
Dr. Fleishman: You know, honestly I was. When I was setting up this project, I wouldn’t be surprised if we didn’t see any relationship at all and there's no increased, which I thought would still be useful because a lot of parents want to know what are the risks taking this medication. If I could reassure them from a fracture standpoint there doesn’t appear to be an increased risk then that would be great as well. So I wasn’t afraid of getting a negative result, but yeah I was a little surprised.
Host: So yeah. I'm sure you documented this. What types of fractures were most common among the PPI group?
Dr. Fleishman: So in both groups—the PPI and non-exposed group—we still see upper extremity fractures predominate. As you can imagine, most children—if you look at population data—tend to have arm fractures, wrist fractures, that type of thing during childhood. So that was consistent with what we found in both groups. If you broke both groups down and looked at other locations of fractures, we also found that the PPI exposed group was more likely to have lower extremity fractures, spinal fractures, rib fractures as well which I didn’t anticipate that. So for whatever reason, there does appear to be a propensity to have fractures in other locations if your exposed to a PPI based on our data.
Host: Yeah. That’s very interesting right. So there aren’t a lot of studies in the pediatric population with PPIs, but I think there are some. How was yours different from maybe some of those other studies?
Dr. Fleishman: Yeah. So really I think to date there’s three to four other studies that I know about. I think what separated ours is one is the age range. Other studies tend to either focus on older children or younger children but not necessarily both. So we tried to include all because we felt like that was useful data to see how that effects all groups if you combine them. The other thing we did was the location of fracture. A more recent study has looked at that which got published around the same time as ours, but prior to that nobody had really talked about if there is an increased risk of fracture, where are the fractures? Are they the same we’d expect or are they more kind of pathologic fractures? So I think those two separate us out a bit. Then the numbers. I mean having access to the FIZZ database and the power of getting over 32,000 encounters is a bit unique.
Host: That’s pretty good. That’s pretty good. All studies have limitations, right. So what were some of the limitations of your study?
Dr. Fleishman: Yeah. So definitely limitations. Big data’s kind of overarching, like looking at the big picture. You don’t get some of the granular data like more duration effects and things like that. The biggest weakness I see in it and hopefully this will get better over time as we use medical records a little more efficiently and they become more accurate is that this isn’t a closed system. For instance, a child who’s captured in the FIZZ database, yes they may be at a large children’s hospital there. They can go outside of this little bubble of what we’re able to capture. They can go to their pediatrician if they have a finger fracture. They can go directly to the orthopedic doctor if they have an arm fracture. So we’re only capturing this group in these settings. Additionally with medications, as you know, a lot of these are prescribed outside of these particular settings or you can go buy a proton pump inhibitor and histamine blocker over the counter. So I wouldn’t be able to necessarily capture those. Hopefully, those would, if anything, underestimate our exposure but probably effect both groups equally we would assume.
Host: Yeah. So you mentioned one of the limitations is duration of exposure to the PPI, right? You didn’t really dig down that deep. Is that something you're interested in now? Now that you've shown this association to increased fracture versus that control, is that the next step for you? Dosing of PPI, how long they should be on PPI. Is that kind of where you want to go?
Dr. Fleishman: I think so. Whether I’ll be able to do it in the same manner remains to be seen or whether you have to use a different database or more granular database. I think those are all legitimate questions. Because if we’re implying that a medication has a certain side effect well my next question is I think it’s the medication they need. If they need it, what can I do to decrease this risk? That becomes, exactly as you mentioned, a dose or a duration question.
Host: Yeah, excellent. So in summary, what would you like pediatricians, GI specialists, community doctors, what would you like for them to know about PPIs in kids?
Dr. Fleishman: So I think the first think I would want to emphasize is that our data shouldn’t be interpreted necessarily as PPIs causing a fracture. That’s not what we set out to do. There seems to be an association and why that is, as we discussed before, I don’t know. I don’t think anyone knows at this point. Secondly I would think overall these medications appear to be safe as a whole. There's an increased risk of fracture, but if you look at the overall numbers it’s a small increased risk, but nonetheless it’s there. I think they're effective medications and they can be very helpful to a lot of patients. We should still be inclined to use them if we need it, but for all providers and patients alike we should always be considering whether a patient still needs to be on a medication and should never leave a medication on their list without truly thinking over time do they still need this? Yes or no. If they don’t, how can I safely get that medication off of their list and still make sure that they're being treated appropriately. I think it also goes to show that like I said there is an increased risk, but I think parents should be reassured that this is a minimal increase. Therefore from a fracture standpoint, if your doctor thinks you need this I think it’s still a safe decision.
Host: Excellent summary. That’s Dr. Nathan Fleishman. He’s a pediatric gastroenterology fellow at Children’s Mercy Kansas City. Thanks for checking out this episode of Transformational Pediatrics. Please visit childrensmercy.org to get connected with Dr. Fleishman or another provider. If you found this podcast helpful, please share it on your social channels and be sure to check out the entire podcast library for topics of interest to you. Be sure to check back soon for the next podcast. I'm Dr. Mike. Thanks for listening.