Sky to emergency medicine. Meet Greet, Treat man, Street him. Today's date is 05/31/2024, and I'm your skeptical host, Dennis Ren. The title of today's podcast is finding Ne or finding pneumonia in feb infants. Our guest skeptic is doctor Christina Lin, who is a pediatric emergency medicine attending at Children's National Hospital an assistant professor of pediatrics and emergency medicine at the George Washington University's School of Medicine and health sciences in Washington Dc.
She also serves as the associate program director for the pediatric emergency medicine fellowship. Doctor Lin, welcome to S pe. Thank you. Excited to be here. So I'm super excited to have you on because as I mentioned, you are the App ap of the Pe fellowship, and you also happen to be my Ap as well. So this is a very special episode for me. Such a privilege Dentist. And we go way back. I don't know if you remember the first interaction we ever had when I was a pediatric in intern.
In the emergency department on my first ever rotation. I think I worked with you, and I said, I think I wanna do this I totally do. And you've been ever so supportive throughout the course of my career. I think we met up a couple... Times where you looked at my application, and we just, like, hung out. So I I just have to say very publicly in this forum. I truly truly appreciate everything that you did and helped me along in my journey. Dennis, it's a privilege you're at the best.
Look how far you come. Well, it's great to have you also join me in this podcasting adventure, and I understand that you brought us a case. Sure did. Alright. So you have a 6 week old boy who's brought in by his family to the emergency department for a fever. He's found to have a temperature of 38.4 degrees celsius at home on the night that they're bringing him in. His parents tell you that he has been congested for the past few days and his cough seems to have worsened.
They've been using a bulb suction device at home to help him breathe better, and he's still to breast milk formula. Of course, there's a school aid something at home with coffin congestion as well, which she has successfully passed on to the rest of the family. On your physical exam, you note that he has clear nasal secretions, normal oxygen saturation and looks quite well. His parents ask you, He's so little. Do you think he has pneumonia? His sister and pneumonia in the past
and got a chest x ray? Does he need a chest x rays as well? We have covered topics about feb infants and pediatric pneumonia multiple times on the S, and that included the step by step rule, the Clinical prediction rule, the Aap guidelines, the safer trial, the Scout cap trial, it only makes sense really that today we get to combine both of those topics into 1 episode and talk about pneumonia in feb infants less than 60 days.
Pneumonia is a tough diagnosis in this very young population based on just clinical exam alone? Don't you think, Dennis? Oh, yeah. I totally agree. And, you are much wiser than I am and you trained me. So how in the world do you to Determine when and if you get a chest x rate in this population. I have to say that this is tough. I would say in general, if the child has a fever, plus my a cough and doesn't have classic exam findings consistent with
bro. So do I mean w, pops, squeak, that sort of washing machine sound. Also, if they focal lung findings that don't clear with suction or other interventions. I also consider what their overall work of breathing looks like. And if they're epoxy. But overall, I think this is really challenging because as we all know, there's a lot of crossover with bro. And, ultimately, there's no evidence based guidance as to who gets a chest x ray and
who does not. And so, unfortunately, I think this leads to a lot of practice variation which we see in our daily care. And there's also limitations when it comes to interpreting chest x ray findings, but but I think I'm getting ahead of myself here. We'll revisit this in the talk nerdy section? I just got too excited. Christina, what is the clinical question we're asking today?
What are the factors we consider such as demographics, clinical information, laboratory findings that are associated with radio graphic pneumonia, in several infants. And what's the reference? The study was published by Todd Flo at all, entitled radio graphic pneumonia and young federal incidents presenting to the emergency department. Secondary analysis of a prospective cohort study. And it was published in the emergency Medicine Journal in 20 23.
Okay. Let's break it down by the Peacock questions, what was the population that they included in the study? So this is federal infants less than or equal to 60 days with a rec temperature greater than or equal to 38 degrees Celsius with a chest x rate performed.
And they excluded infants who appeared critically ill who are already getting antibiotics who were premature defined as less than 37 weeks gestation had significant comorbidities, ind dwelling devices or focal bacterial infections like cell. What was the intervention? So they were looking at the evaluation of radio pneumonia classified into definite pneumonia, possible pneumonia and known pneumonia. And their comparison? They have no comparison.
Okay. Let's talk about the outcomes you had mentioned a couple different factors they were looking at. So their primary outcome, they were looking at demographic clinical and laboratory factors associated with radio graphic pneumonia. And then those clinical factors that included things like their vital signs, their oxygen saturation, the respiratory exam findings, and they also use the yale observation scale.
Laboratory factors included white blood cell count and see your absolute count, platelets, pro, urine in testing and viral testing, which was a Pcr bio. And lastly, what kind of trial was this? So this was a secondary analysis of data from a previous p card study, which have been conducted from June 20 16 to April 20 19. I feel like this group just keeps looking at the data from that study. It it is truly like a gift that keeps on giving it seems. I agree. We do have a very special guest on
this episode with us. We are joined by doctor. Todd Flo, who is the associate division Head of academic affairs and research in the division of pediatric emergency medicine at Lu Children's Hospital of Chicago, And associate professor of pediatrics at Northwestern University Fe School of Medicine. He just so happens to be the... Doctor Todd Flo, who is the lead author on the paper we're discussing today. Todd Welcome to S p. It's so
great to be here. Thanks so much for having me to talk about 1 of my favorite topics in pediatric emergency medicine. Doctor. Flo, do you mind giving us your conclusions from the paper? Radio photographic pneumonia were uncommon in feb infants. Viral detection was common. Pneumonia was associated with respiratory distress but few other factors.
Although Anc c and levels were elevated in infants with definite pneumonia, further work is necessary to evaluate the role of blood biomarkers in infant pneumonia. Okay. It's time for our quality checklist. First question, did the study address a clearly focused issue. I would say yes. Do the authors use an appropriate method to answer their question? Yes. They use a secondary analysis of a pro? Prospective cohort study, which is appropriate for identifying clinical features and associations.
Was the cohort recruited in an acceptable way. Yes. The cohort was recruited from 18 Eds in the Peak car network and included federal infants evaluated for serious bacterial infection. Was the exposure accurately measured to minimize bias? I'm not sure if it was. What about the outcome was the outcome accurately measured to minimize bias? Yes. Outcomes were measured using consistent criteria for classifying pneumonia... Are based on prior studies and collecting clinical and laboratory data.
Have the authors identified all important con compounding factors. I'm not sure. I think this is hard to do. Do you think the follow up of subjects was complete enough? I would say yes. How precise are the results? So this is tough and we see this a lot with this very small patient cohort because... It's such a small sample. So I think we're kind of unsure about the precision? But do you believe the results? I do. I believe them. Can the results be applied to the local population?
Again, I'm unsure, again, because was a small cohort? So I think you have to evaluate based on where you practice. Do the results of this study fit in with other available evidence. Yes. The findings are consistent with other studies on pneumonia in young feb infants. And let's talk about the conflicts of interest now. Were there any conflicts of interest cleared and who funded the study? Yes. Doctor Ram reported personal fees from San pastor, Merck and Pfizer.
And grants from Jensen and the Bill and Melinda Gates Foundation, unrelated to this study, there were no other conflicts that were reported. And the study was funded by the Unique Kennedy Sc National Institute of Child health and human Development. The National Institute of Allergy and infectious diseases the National Heart Lung and Blood Institute and the Health Resources and services administration. Okay. It sounds like no glaring conflicts of interest here with this study. No.
Let's talk about the results now. They enrolled a 2612 infants of whom 22 percent had chest x rays performed. And the median age was 38 days a little bit more than half for male, 59 percent in 19, and that's only 3 percent had a definite pneumonia and 34 or 6 percent had a possible pneumonia. So really not that much definite or possible pneumonia here. Christina, what was the key result? Radio pneumonia in federal infants was uncommon and associated with respiratory distress.
For some reason, I actually find that kind of reassuring. Right? Respiratory distress Like, oh, I can actually trust my clinical exam this time. Right. But how do you define respiratory distress? That's the tough part. Oh, now you're asking the hard questions. We talked about all these factors that they were looking at you tell us a bit about the demographic factors? What do they find?
Yeah. They found no statistically significant differences biz based on age, sex, race or ethnicity among the groups that had no possible and definite pneumonia. And when it came to clinical fact, the infants who had definite or possible pneumonia were more likely to be to kip, have worked of breathing, respiratory distress, required mission or have influenza or Rs
v detected. Yeah. And there were no statistical differences based on the yale observations scale, but a higher proportion of infants with definite pneumonia scored higher on the yale observation scale, which intuitively makes sense. Yeah. What about those lab factors we talked about? Yeah. The median white blood cell count, a and c and pro we're higher in infants with possible or definite pneumonia.
It's important to know that 57 percent of the infants or 324 infants had pro and results actually reported. There were a higher proportion of infants with definite or possible pneumonia who had influenza or Rs v detected. And back to was rare and no infant who had the possible or definite pneumonia ended up having back to. Doctor Flo and doctor Lin. Are you ready for my favorite section? Are you ready to talk nerdy? Always ready to talk Nerdy, doctor Ren.
Alright, Todd. You are on the hot seat for this 1, but I will let Christina lead us off. Our first point is about selection bias. Alright. So in the initial Peak concurrent study, chest x rays were performed at the Ed provider's discretion. Unfortunately, researchers were unable to capture the Ed provider's clinical reasoning behind performing the chest x. Was it part of an institutional pathway for working up a feb infant? Was it because the infant was exhibiting signs of respiratory distress?
In the infants who did not have Chest x performed, what proportion would also have potentially had evidence of radio graphic pneumonia. I'm not advocating for radiating all babies, but how do you think this selection bias may have impacted your results? Yeah. So this is a legitimate concern and truly a limitation of our paper. We really did not know the reasons why Chester rays
were ordered. They could have been ordered because the infant had respiratory findings, institutional pathways could be in place or It could just be that provider's preference. That said, we also view this limitation as a bit of a strength because it also reflected clinical practice in the real world. We did not look at how many children who were initially not radio graft, returned or had radio later that made had have demonstrated pneumonia.
But, you know, I'd venture a guess that this would be few to none. Our results are pretty consistent with some of the smaller studies that are out there in the literature, and and I think even with possible selection bias at play, our results are in the ball park for pneumonia incidents in young federal infants, So we don't think that our results are that far off the mark as a result of any potential selection bias.
Moving on to our second question for you, it's this definite possible and no pneumonia. We talked earlier in the show about the limitations in the use of chest x rays in detecting pneumonia, and we can't tell based on chest x rays whether a pneumonia is back or viral. And I think we've all gotten those radiology reads along the lines of or developing pneumonia correlate clinically. And there's even poor inter rate reliability among radio.
Your study characterize these chest x rays into 3 categories. No possible and definite pneumonia. What was the process for evaluating the chest expert by these study investigators and how many investigators reviewed these x rays? How was the inter rate reliability amongst the study investigators and amongst the study investigators and the attending radio epidemiologists who perform the initial review. Did you all agree?
Of course, we agreed. No. This is this is this is the bane of every pneumonia researchers existence is the interpretation of chest radio. It is not a great reference standard for all the reasons that you cited earlier. That said, There is evidence to suggest that the negative predictive value of chest is very high in the ballpark of 98 percent or greater. So what we can say that is if a chest x rate was read as definitively normal by the radio.
It is highly unlikely, though not impossible that the child will return with pneumonia. In this study, we did not have the ability to review the images directly. So we could not have multiple radio or multiple emergency medicine physicians, all review the films and compare inter integrator reliability. We relied solely on the local radio report. So we classified the reports using a standardized process that we've used in prior studies. And that is the reason actually why we
included a possible group. Just given the limitations of reviewing, radio reports and the limitations of of, reading chest rated graphs in general. Now I performed the review on all non definitively normal reports using this standardized scheme and then classified them. Right. So just to clarify, you saw the the text based reports, you you didn't see the images themselves. That is correct. Moving on to our third question for you, this is 1 that I'm were curious about, and that is viral testing.
So in this study cohort, a large proportion of the children tested positive for viruses, and a higher proportion of children with possible or definite pneumonia is tested positive for influenza or Rs. We know that detection of certain viruses such as influenza, Rs or even Sars C 2 may decrease the risk of invasive back atrial infections in this young age group. At the same time, we know that viral Pcr testing is limited and positive results can linger after active infection.
Positive testing does not eliminate the possibility of bacterial c infection. So with that being said, how are you using respiratory viral testing in this population if at all. Yes. So there's huge variability in the use of viral testing for Feb young infants by provider by institution. I think what the data shows us is that viral testing generally has little clinical utility other than possibly influenza testing.
And the reason for that is that os tam is generally indicated in young infants older than 2 weeks of age with influenza, due to their increased risk of severe disease and complications. So if the infant is presenting during influenza season, I typically will obtain an isolated influenza Pcr tests that could guide my treatment decisions based on the result.
There are studies in young feb infants with Covid that have suggested that the rate of serious bacterial infection is less in Sars C 2 positive children, but not ins substantial. So the knowledge of Covid status is actually less helpful in this age group because it does not eliminate the possibility of a serious bacterial infection.
Now large scale multiplex viral testing, you know, those panels with 22 pathogens really has very little utility, and generally has not been shown to alter outcomes in older children. Though fewer studies are done in this young feb infant population. That said, I do think that the prevalence of detected viral illness in our population suggests that more work is necessary. To understand the role of viral illness in feb respiratory disease
in the youngest infants. It's it's interesting because there is some work going on that is looking at the severity of disease of specific viral strains, and it may be that the future of viral testing for these young infants with respiratory disease is actually strain specific as a predictor of severe outcomes. So more to come on that in the scientific literature in the years to come.
Todd, I'm I'm curious. 1 of the limitations of viral Pcr testing, which you're not a fan of, and I think that makes us fast friends. Is that when you when you detect a virus. Right? It doesn't tell you whether or not the virus is active. Right? You're you're detecting, like, we've we got different
parts of this virus. So can you comment on the utility or the limitation just simply based on that, you can test positive for 567 all of the viruses, but it's just how many of those are actually active contributing to the disease process. Right. So I think with regard to this population in particular, we're talking about infants less than 6 days, which means that they didn't have a whole lot of time on this earth to be able to have these viruses make their way into
the note. So unlike an older child who may have rhino virus, you know, that they had from their common cold 8 weeks ago still lingering in their nose. It's a little less likely that that's going to happen with a 60 day old or 30 day old infant. Now that being said, I think they're you know, there's data that we can extrapolate from older children about colonization versus infection.
And that data would suggest that if you're seeing Rs v for instance, in the nas or the nose, of a child with a disease that is consistent with Rs clinically. It is highly likely that that that is not some residual Rs v, but that the Rs v is actually playing an active role in infection. It's a little bit different with some other pathogens like rhino virus.
And rhino coronaviruses are just so common that it is difficult to say that a child with rhino coronavirus detected in their nose or their nas. Is truly having coronavirus as the active cause of their infection. That being said, I'm not sure that it matters all that much because I think in this age group, we're not going to use the viral test as the sole determinant of figuring out whether this child should should not receive antibiotics with concerns for serious bacterial
infection. So The fact that you have rhino virus in your nose is ultimately not going to change the clinical management of that child, whether that rhino virus is active or whether that is residual. Fantastic. Thank you for that answer, And that's actually perfect segue as we talked about feb infants into our next question about inflammatory markers. Previous low risk feb infant prediction tool risk strat based on the results of the uri analysis, an Anc and c and pro.
And that pro cutoff off was 0.5 nano per Ml. Your study also looked at white blood cells, Anc c and pro. And the pro in this study even in the groups of possible or definite pneumonia range from 0.1 all the way up to 1.5. So how do you interpret an elevated pro level that's greater than that 0.5 and in pneumonia on chest x ray? Is that reassuring to you that you've now found the source of infection? I think that if you're seeing a high pro in the setting of fever and pneumonia on
chest x right? I think it's likely that you found the source of the infection, and I would treat accordingly. I would still, however, follow the urine in the blood cultures as well. At this point, I would also kicked by the Feb infant rule as published. Since pneumonia was too uncommon to draw definitive conclusions and and was not included in the original rule der.
Now it's interesting about pro on the lower end is that there are studies in older children with pneumonia that suggests that a pro concentration of less than 0.25 nano per milliliter is likely indicative of a low risk of a typical bacterial infection. So I think that if the pro is very low and the other pecan markers are negative, it's highly unlikely that you are dealing with a typical bacterial pneumonia.
What's interesting about these studies in older children with pneumonia is that when pro levels were higher, they were really not sensitive or specific.
So where pro seems to be most useful as actually on the lower end that less than 0.25 concentration to be able to feel reassured that it is likely not a typical bacterial infection, whereas higher levels are non specific, and therefore, given the vulnerability of this age group, I would always air on the side of treating as these would fall out of the low risk criteria. Let's say you're working at a hospital that doesn't have access to a pro.
Do the results of your white count or Anc c help guide your to vision to perform a chest x. We know there's a lot of overlap in the ranges provided amongst the groups with no possible and definite pneumonia. We've been using white blood cell count for a long time. And I think the data is now becoming more clear that white blood cell count and Anc and c are less useful in determining the ideology of pneumonia across all age groups.
I do not think that white blood cell counter or Anc c should be used to guide chest x use at this time given what we know.
We know that white blood cell count in Anc are known to not determine the et ideology of pneumonia, and nor are they particularly predictive of the presence of a radio graphic pneumonia, and you point out very astute that there is huge overlap in the ranges in our study across groups, which further emphasizes this point about the challenges of using white blood cell count or Anc and in isolation. Christina, you wanna round this out with our last question for doctor Flo? Yes.
So where are we going? What do you thinks next And where do you think we're gonna go with a diagnosis of pneumonia and federal infants and children? Oh my gosh. There's so much to do, and that's why I find this topic so citing, it's a careers worth of study. So I'll name just a few things just to keep it short because I could go on forever about this. I think that we have not yet fully evaluated the role of ultrasound in the diagnosis of pneumonia in the youngest
infants. So there is more and more work in older children that ultrasound can be useful at the point of care to be able to detect pneumonia at least in a similar fashion to chest radio graph. This has not really been looked at in these smallest infants where you've got a small chest and what probe you're gonna use and what technique you're gonna use So I think there's some role for looking at the role of ultrasound in the pneumonia diagnosis in these kids.
There are also investigators that are currently evaluating systems biology approaches, so things like Pro, meta metabolism, transcript, to be able to actually look at differences in gene and protein expression in infants and children with bacterial versus viral infection. Which may help to guide our diagnostic approach. We've seen this work in older kids with pneumonia, but mainly in kids with sepsis.
Where we're discovering panels of new biomarkers that come out through host gene expression in response to a pathogen. That are more accurate than any of our current tests. So I think that is a super exciting field of study for the future in terms of using these systems biology approaches to improve our idiomatic discrimination in kids with pneumonia. Todd, you mentioned a lot of O, and I would consider myself an on this particular topic. But are these tests all serum testing blood
testing. We're gonna have to poke these kids and and draw blood for them. So what's fascinating is you can actually look at these markers of gene and approach expression in several different sample type. There are people that have looked at rna gene expression. So that's transcript in the nas. We have looked at meta
in the urine. So looking at what meta are being in the urine, there are folks that are looking at pro, so looking at protein expression in blood in urine, to be able to differentiate patients with and without pneumonia with bacterial versus viral pneumonia. So it really is looking across multiple sample types. I think that blood is really
the most commonly looked at. But I think given the prevalence of viruses in the nose, you know, folks are really exploring the nas as a possible source for this as well. And we shouldn't forget in all of this, the microbiome. So there are folks that are looking at the nasal microbiome and comparing microbial flo, in kids who have pneumonia don't have pneumonia,
have bacterial disease have viral disease. To be able to look at if we can discriminate kids with and without these diff pneumonia, with and without viral infection, bacterial infection, just looking at the microbiota in the nas. It's super cool stuff. It's very star trek. Your your passionate and nerd truly come across very well, and I I appreciate that. That is not that is not a dig That is, like, truly, yeah. Nerds.
Well Well, we are talking nerdy. So I told you at the beginning that I was super excited to join in this, so I appreciate that opportunity to expo on that a bit. That's perfect. Doctor Lin, can you comment on the author's conclusion compared to the S gem conclusion? We agreed with the author's conclusions. And give us the S gym bottom line. Consider obtaining a Test x, rate in feb infants with clinical signs of respiratory distress.
And can you resolve the case from the beginning that little baby with the sniff in this school age siblings, feel like we've all been there. What do we do?
So I would say you let the family know that their potential harms and benefits of ordering a Chest x ready for their child, there's little evidence to guide decision making around ordering a chest x. Given that the child appears well, and does not appear to be in respiratory distress, you engage and share decision making an opt to defer the chest x at this time, which is the added benefit of not exposing
the child radiation. However, you will plan to obtain urine and blood testing based on the peak low risk feb in scent prediction tool. Fantastic. Yeah. I've always feel so bad, and it's always so tempting to be like, yeah. You know, you're you're tiny, but You might have a cold, But when when you add that fever, you just... I think you just got an air on the side of caution. Totally. Doctor Flo, how do you apply the findings of this study to your clinical practice?
I think that clinically, this work demonstrates that... We should not uniformly include Chest x rays in the work up of the Feb young infant. It should not be part of standard institutional path. Ways. It should not be part of national guidelines as you must do a chest graph in every infant that has a fever. It just the rates of pneumonia too uncommon. But in those where there is respiratory distress, signs of lower respiratory tract disease, it it really should be considered.
And in addition, we just talked about laboratory findings and they certainly were not definitive in our study. I would consider a chest x those infants whose... Anc and c or pro are really through the roof. Just thinking that this may be another source of infection, I would not consider it for those whose Anc c's or are sort of borderline as we just discussed the limitations of these biomarkers. Was there a feature of respiratory distress? You know, we... I that was cited in the
article quite a bit. That was uniquely different from what we see in patients with... Bro, because personally in my practice, this is this is the crux. Right? And I asked because we have quality measures in our institutions specifically to decrease chest x use in infants with bro who are actively demonstrating respiratory distress. You know, as you mentioned, there's a good deal of overlap between the clinical
presentation of pneumonia and bro and infants. And our measure of respiratory distress in this study was really a composite measure that included the present of retract grunt, head bobbing and to kip. So it really was a composite measure. Now, anecdotally, 1 of the key differentiator is
the 3. So bro almost always starts with, upper respiratory symptoms that then migrate down to the lower tract if you are seeing a child that has no upper respiratory symptoms and only has lower track disease that may anecdotally lead you more to think about an pneumonia than bro. On au consultation as well, generally, with bro, you're gonna hear w and Wrong ky we tends to be negatively associated with pneumonia. So if you're hearing a distinct we is...
It's going to make it less likely that it is pneumonia. Understanding that oftentimes in these infants with small chest, it's very difficult to differentiate sometimes w from Wrong guy from Ra or crackle. And so it's a real challenge that I think our clinical exam is not going to be able to uncover. But if you are hearing, you know, mono w that is not associated with anything else it's less likely to be a pneumonia.
So some of these findings are things that you can use to help differentiate these 2 similar conditions with the caveat that Much of this is anecdotal and extrapolation of the literature rather than truly evidence based. It is comforting to me to hear that
even next... It, like yourself has some trouble distinguishing because I feel like, most of the time, when you lay your stethoscope the scope at least my stethoscope, the scope on 1 of these 80 bitty babies for I'm often performing a cardiac respiratory and abdominal exam at the same time. That is absolutely true. And that's... That is a known limitation of why this work is so difficult because the au sculpt exam is really really...
Not reliable in this young age group. So let's bring it back to the patient case that we presented earlier on that 6 week old. What are you telling the parents? So I think that you can confidently tell parents that pneumonia is uncommon in feb infants, particularly if they're not demonstrating signs or symptoms with respiratory distress.
That said, given the limitations of the clinical history and the physical exam in discriminating those with and without pneumonia as we've discussed throughout this episode, I think it's reasonable to engage in a shared decision making process with the parents to determine the best course of action that's specific to their situation. Well, doctor Flo, doctor Lin, thank you both so much for joining us on S Beads.
We learned a lot through this discussion. Thanks so much for having us and remember to be skeptical of anything you learn. Even if you heard it on this skeptics guide to emergency medicine, Talk to everyone next time.