This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.
Todd A. Florin, MD, MSCE: Hello. I am Dr Todd Florin, and I'm an associate professor at Northwestern University Feinberg School of Medicine and the director of research for the Division of Emergency Medicine at Ann and Robert H. Lurie Children's Hospital of Chicago. Welcome to the Medscape InDiscussion podcast series on pediatric pneumonia. Today, we'll discuss the latest treatment strategies for pediatric pneumonia, including three recent trials that should make us all reconsider those long-standing dogmatic approaches to the treatment of pneumonia in children with mild disease. First, let me introduce my guest, Dr Derek Williams. Dr Williams is an associate professor and chief of the Division of Pediatric Hospital Medicine at Vanderbilt Children's Hospital. Dr Williams is also a clinical and health services researcher. His research program centers on improving care delivery and outcomes for children with pneumonia and other acute respiratory illnesses. Welcome to InDiscussion, Derek. How are you?
Derek J. Williams, MD, MPH: Great, Todd. It's awesome to be here. Thank you so much.
Florin: You and I have been working in this field for a long time. We've been collaborating together for many years. And there have been many changes in medicine. Like I ask all of my guests for this series, what do you think are the most exciting changes in pediatric healthcare you've seen over the course of your career?
Williams: What I always come back to when I'm talking about pneumonia and things that have been most impactful in my career is the introduction of pneumococcal conjugate vaccines because of the dramatic impact they have had on pneumococcal disease, seeing reductions in disease overall, and seeing changes in penicillin resistance as a result. The other thing that has been really exciting is all of the new molecular diagnostics — our ability to identify different pneumonia pathogens and different ways to do this. It also gives us a lot of work to do as we think about how to incorporate these things into our clinical practice.
Florin: I resonate with both of those, clearly. You and I talk about molecular diagnostics and how we can improve the approach to pneumonia all the time. I will put a plug in for one of our other InDiscussion pneumonia podcasts on vaccines that we did with Dr Rick Malley from Boston Children's Hospital. I encourage folks to listen to that episode. In terms of molecular diagnostics, this is a key innovation in medicine we have not seen in the diagnostic space prior to this with regard to its speed, sensitivity, and comprehensive nature. This discussion of molecular diagnostics leads me to the first point I'd love to discuss with you, which are the data on etiology of pneumonia in children. You brought up the pneumococcal vaccine, Derek, and how that changed the etiology of pneumonia over time. Pneumonia used to be largely a bacterial disease. But the epidemiology and etiology have shifted over time. Can you talk about some more contemporary data on pneumonia etiology and how they might impact our treatment approach?
Williams: Sure. I will highlight two studies — one on the etiology of pneumonia and community-acquired pneumonia that I was a part of, EPIC, and then an analogous study, PERCH. EPIC was based in the United States. PERCH was an international study. Both studies were focused on understanding pneumonia etiology in the era of pneumococcal conjugate vaccines. Both studies used comprehensive diagnostics for viruses and bacteria, including a lot of these new molecular diagnostics — specifically, polymerase chain reaction (PCR) — for a number of different respiratory viruses sampled from the upper airway. These studies also looked at whole blood by PCR to detect pneumococcus, historically the most common cause of bacterial pneumonia. What both of these studies showed us is that viruses are incredibly common in children who are diagnosed with pneumonia in both inpatient and outpatient settings. Respiratory syncytial virus is one of the most common, and human rhinovirus is also very common. Place an asterisk around rhinovirus because we also know that this pathogen can be detected in a lot of children who are asymptomatic, which brings up some of the challenges with these new molecular diagnostics. Also, importantly, in the EPIC study, very few children had pneumococcus detected — less than 5% — similar to the PERCH study. This is a big shift in our understanding of pneumonia as we are seeing a lot more viruses and less and less pneumococcus.
Florin: That finding was surprising to me and probably surprising to a lot of our listeners as well — there was a really low prevalence of pneumococcal detection in both of those studies. Can you hypothesize as to how much of it is that pneumococcus is truly becoming rarer vs our diagnostic limitations to be able to detect a localized pneumococcal infection in the lungs that may not end up in the bloodstream and certainly may not be detected by other methods. Is it a prevalence issue? Is it a detection issue? Or, is it a little bit of both?
Williams: I think it has to be both. If you look at other data on culture-confirmed disease caused by pneumococcus — and this is all types of invasive pneumococcal disease — we clearly see dramatic reductions in invasive pneumococcal disease following the introduction of these conjugate vaccines. Clearly, vaccines are having an impact on clinically relevant disease due to Streptococcus pneumoniae. In other studies that look at the prevalence of pneumonia overall, it has not declined to the same degree, which tells us that as these etiology studies are showing us, there are a lot of different pathogens playing a role in pneumonia. Even though we see less pneumococcus, overall prevalence has not changed that much, telling us that these other pathogens are playing an important role. The other really big issue here is our ability to identify typical bacteria like pneumococcus. We can't use the same methods we use to identify viruses from the upper airway just because of issues with colonization. Our ability to detect these bacterial pathogens is less sensitive. We don't have the same ability to detect these pathogens. I also think we are undercounting some of the disease caused by pneumococcus. I can't tell you exactly what that number is. I do think that in a lot of these children, based on the epidemiology data and also my clinical experience over the course of my career, classic lobar pneumonia is much less common. This tells me that there are certainly a lot of different pathogens at play here. Viruses are playing an important role. Pneumococcus is still around.
Florin: Yes. That will segue into our treatment discussion. The one pathogen I would like to make a brief mention about is Mycoplasma pneumoniae and the so-called atypical pneumonia. You hadn't brought that up previously. What did the EPIC study show, and what are your thoughts on the prevalence and importance of mycoplasma in children with pneumonia?
Williams: We do have molecular diagnostics using PCR from the upper airway to test for mycoplasma, which is thought to be less concern for colonization as we see with other typical bacteria. In the EPIC study, mycoplasma was actually the most common bacteria identified in over 10% of children — still much less than viruses that were detected but certainly playing a role. The interesting thing, though, about mycoplasma is that it does not seem to be a pathogen that is always around. It seems to come in waves in these epidemic years where we have lots of disease, and then other times we have much less disease. We saw that within the EPIC study as well, with most of the detections being concentrated in a certain period of time.
Florin: Now that we've talked a bit about the important etiologies, let's dive right into the treatment question and the treatment guidelines. In 2011, the Pediatric Infectious Diseases Society and the Infectious Diseases Society of America published guidelines on the management of pediatric community-acquired pneumonia. Can you elucidate for our listeners what these guidelines say about how to best manage inpatients and outpatients with community-acquired pneumonia with respect to treatment?
Williams: Absolutely. It's easiest to dichotomize this into outpatient and inpatient. That's how the guidelines are organized. Recommendations are largely similar, however, and really hinge on the fact that we know that pneumococcus is historically the most common bacterial pathogen in pediatric pneumonia. As I mentioned, the data around penicillin resistance in the setting of pneumococcal conjugate vaccine resistance have declined significantly. As a result, the primary recommendations for treatment really focus on narrow-spectrum aminopenicillins. In the outpatient setting, we're talking about amoxicillin. In the inpatient setting, we're largely talking about ampicillin, and that really contrasts with the use of third-generation cephalosporins, which were more common in the decade preceding publication of the guidelines when penicillin resistance was quite a bit higher. This did, I think, prompt a shift in how we think about treatment — focusing on more narrow-spectrum therapy in the outpatient setting. Also, in the outpatient setting, it is important to recognize the importance of viruses, especially in young children. There was a recommendation to consider withholding antibiotics in those young children, especially those who present like they have a viral illness with bilateral disease and mild illness because we know viruses play such an important role. There are less data and there is less evidence around the use of macrolides for mycoplasma in both the inpatient and outpatient setting. In the outpatient setting because these children are less ill, the recommendation probably shifts more toward withholding therapy unless you have an identified pathogen, such as mycoplasma, that you want to consider treating. On the inpatient side, again, for the straightforward noncomplicated pneumonia, we really emphasize narrow-spectrum therapy. On the inpatient side, we do certainly get into more issues with complicated pneumonia, children, and more severe disease. You have to start thinking a little bit about other pathogens such as Staphylococcus aureus and things of that nature. Those are much less common, though. Really, the emphasis is on use of narrow-spectrum therapy in most cases. The caveat is presuming that most of these children we're seeing in most settings have been vaccinated against pneumococcus and other common pathogens.
Florin: That was a great summary, and I think the move to narrow-spectrum aminopenicillins, amoxicillin, and penicillin has been successful. Studies have demonstrated that there has not been an increase of complications of pneumonia or worsening of disease as a result of using narrow-spectrum antibiotics compared to broader spectrum cephalosporins and the like. However, amoxicillin allergies or labels of amoxicillin allergy are pretty common, and clinicians are often faced with a conundrum of how they maintain the philosophy of narrow-spectrum antibiotics. What are the alternatives if a child has an amoxicillin allergy or a labeled amoxicillin allergy?
Williams: The first thing I think about is, as you alluded to — is this a true allergy? Because we know that about 10% of children will have a penicillin allergy label. Of those who have an allergy label, about 90% or more, when they are challenged, do not in fact have a documented penicillin allergy or do not have a penicillin allergy response, and it is likely safe for them to use penicillin. There are several studies by several groups that are trying to implement protocols around delabeling that include risk stratification. If you can identify the low-risk patients for true penicillin allergy, you can challenge them right there in the emergency department or hospital with a single dose of amoxicillin, for example. Hopefully you can transition them to use that therapy. if you're concerned about severe penicillin allergy, yes, there are other antibiotic options. Clindamycin is the one that is commonly used. Our local guideline recommends that when a patient has a severe penicillin allergy, consider something like clindamycin.
Florin: A hot topic now is duration. Clearly, the duration of antibiotic therapy is a major stewardship target. A lot of durations are simply based off of the days in the week. We have 7 days in a week, and you get 7 days of antibiotics. Seven to 10 days has been our dogma for four decades for how long to treat these infections. The doctor always told us to finish the whole bottle of bubble gum medicine. But new data are actually questioning this. There have been three major well-designed trials in three different countries, one of which you were intimately involved with, that suggest maybe we need to be rethinking this dogma. Can you talk a little bit about antibiotic duration and what these trials have shown us?
Williams: Yes, absolutely. The three studies we are going to talk about are the CAP-IT trial, the SAFER trial, and the SCOUT-CAP trial. The last trial was the one I was involved in. These were all three studies that were randomized trials. They all had slightly different designs, as you said, and were in different countries. CAP-IT was in the United Kingdom, SAFER was in Canada, and SCOUT-CAP was in the United States. They were all similarly designed, comparing short courses of therapy of 3-5 days vs more standard longer courses of therapy of 7-10 days.
Florin: How was the SCOUT-CAP trial performed, and what were the major findings?
Williams: The SCOUT-CAP trial enrolled 380 children. These were all outpatients who were less than 6 years of age. They were coming in to either their primary care doctor, which was most of the patients, or an urgent care or an emergency department. There was some pragmatic design gear in that we were really focused on clinician-diagnosed pneumonia. We did not approach and enroll subjects until they had already been initiated on therapy for pneumonia. Nearly all of these children were receiving amoxicillin. We approached them around days three to five of therapy. At that time, if they had shown some clinical improvement, they were enrolled and randomized to a short-course or a standard-course therapy strategy. For the short-course therapy, they got five more days of matching placebo. In the standard-course therapy, they received an additional 5 days of whatever drug they were started on. Again, most always that was amoxicillin.
Florin: I do think it's an important caveat for our listeners to understand that these were all kids who showed some clinical improvement before day five. These are not the kids who are showing persistent symptoms or progressive symptoms. These are the kids who are getting better. In most of these viral illnesses that we talked about to begin with, we see some improvement before day five. This is going to be relevant for the vast majority of outpatient children with mild pneumonia. What did you find? What were the results?
Williams: I'll get into the results. The other thing I want to highlight is that this was designed as a superiority study because we wanted to show 5 days was not only as effective as 10 days, but we thought maybe 5 days could be considered better if we consider a child's full spectrum of outcomes. We were not just looking at whether it was yes or no to treatment failure, but we looked at treatment failure which we called inadequate clinical response. Did they come back to see their primary care provider for worsening pneumonia? Did they have to get a new antibiotic? Did they go to the hospital? We also looked at persistent symptoms, such as prolonged cough or fever, and then we looked at antibiotic adverse effects, such as gastrointestinal side effects, rash, etc. We organized all of that into an ordinal outcome that had eight different levels. So not only were we looking at that inadequate clinical response but also if they had persistent symptoms. Plus, did they have antibiotic adverse effects? If you stop there and just compare the clinical outcomes I mentioned, there were no differences in the group of those who received 5 days of antibiotics vs the group who was randomized to 10 days. However, we also accounted for antibiotic duration because we have an ordinal outcome we can compare. We did pairwise comparisons for all of the patients. If they had the same rank on that ordinal outcome, we then used antibiotic duration to break those ties, essentially saying that short-course antibiotics are better because you achieve the same outcome with shorter duration. That was an a priori hypothesis, and that's actually what we showed — when we considered antibiotic duration, we saw similar clinical outcomes, but we achieved this with shorter durations in the short-course arm. We concluded that short-course therapy was actually superior to the standard-course therapy.
Florin: One really innovative and fascinating analysis that you all did in the SCOUT-CAP trial was this resistome analysis. Can you talk a little bit about what you found in terms of antibiotic resistance genes between the two groups of therapy, 5 days vs 10 days?
Williams: Yes, absolutely. To add a little bit more objective data to the claim that short-course therapy was superior, we hypothesized that if we looked at antibiotic resistance genes and the upper airway of children in both groups, we might see fewer antibiotic resistance genes in the short-course therapy because they're receiving antibiotics for less time. And that is actually what we found. We looked at antibiotic resistance genes about 2 weeks after stopping therapy, and for total resistance genes as well as beta-lactam resistance genes, we saw fewer resistance genes in the short-course arm compared to the long-course arm. Those were significant differences, which again provide some objective evidence that short course is, in fact, doing something in longer courses of therapy or having some sort of impact on our microbiome.
Florin: I really love that finding, and I love that you all went after that because I think a lot of clinicians say, "What's the difference? Five fewer days of antibiotics doesn't make a difference." Clearly, we are seeing some objective differences. I also really love that secondary analysis you all performed. What was the dose of amoxicillin? Was this a high dose of 80-90 mg/kg/day, or was this a standard dose of 40-50 mg/kg/day?
Williams: The focus was on the higher end of dosing. There was a range because it was a pragmatic approach using whatever the clinicians had prescribed. The emphasis was on high dose because that is what has been recommended in the guidelines.
Florin: That leads to another trial. You mentioned the CAP-IT trial that was conducted in the UK, which was a factorial design, meaning that it was able to compare a standard dose vs high dose of antibiotics in addition to 3 days vs 7 days. It was a noninferiority design, not a superiority design. The noninferiority margin was 8%, and they enrolled 824 children 6 months and older who had clinical diagnosis of pneumonia across 29 hospitals in the UK. Their primary outcome was a clinically indicated retreatment for respiratory infections within a month after enrollment. What did the CAP-IT trial show and how does it augment the results of the SCOUT-CAP trial?
Williams: The findings overall were very similar when they looked at comparing low- vs high-dose antibiotics. Overall, the outcome of antibiotic retreatment, which was the primary outcome, occurred in about 12% of patients, so that is slightly higher than what we saw in SCOUT-CAP. Obviously, we're talking about slightly different outcomes. Importantly, there were no differences in either of the main comparisons groups, the low vs high dose and the 3 days vs 7 days. About 12% of children in both groups had that antibiotic retreatment outcome. Another important analysis they did is to include both emergency department patients as well as inpatients, so about a quarter of the children were in the inpatient setting. They did have some additional analyses looking at children with more severe disease, which I think is important because that's certainly something we did not see and did not include in SCOUT-CAP.
Florin: That was the one trial that looked at inpatients. Now, granted, just to caution our listeners — it's really not sufficiently powered to look at this outcome. When you look at results in the subset of patients who had severe pneumonia, there was about a 4%-5% higher rate of kids needing retreatment who received standard dose therapy vs high-dose therapy. Similarly, there was about a 2% higher need for retreatment in kids who received 3 days vs 7 days of therapy. The trial was inadequately powered for that outcome. I would say to our listeners that caution is still warranted for hospitalized kids with pneumonia. You're a hospitalist. What is your current take on the treatment of inpatients in terms of duration of therapy?
Williams: We can certainly extrapolate some of these data. There is probably a subset of children in the inpatient setting who are only mildly ill from the perspective of someone who takes care of hospitalized children, meaning that they maybe require a little bit of oxygen. Or, maybe they're dehydrated and need some intravenous fluids, but they're in and out of the hospital within a day or two. I personally feel, and this is more anecdotal, that in my practice we are moving more toward using shorter durations of therapy in those children especially if they tend to appear to have a viral illness, whether or not they have a virus detected, have bilateral findings, or their chest x-ray is equivocal. Certainly, if you're talking about more severe disease, the kid who is in the hospital for 4 or 5 days, is in the intensive care unit, or has complications, we are very far outside the realm of 5 days of therapy in those populations. At the same time, we should acknowledge that children in the intensive care unit with pneumonia very often will have only a virus identified. If we're treating those with more complicated and more severe disease with antibiotics for presumed bacterial pneumonia, we're usually treating for a longer duration until we have better data.
Florin: Great. We're looking forward to seeing what the data might show us in the future about these shorter courses of therapy in hospitalized kids. I want to touch on the third trial, the SAFER trial out of Canada, because it was a well-done trial. It adds support to these two trials we've already discussed. What did the SAFER trial show us?
Williams: The SAFER trial had very similar conclusions to SCOUT-CAP and CAP-IT with no differences in either of the groups. Their primary outcome of initial improvement was about 85% in both groups. A caveat there, though, was that they included things such as if the patient had a fever spike. They included a secondary analysis where they were more stringent in that outcome of treatment failure. When they did that, more than 90% of children in both groups had adequate clinical response. Overall, all three of these trials have had very similar results, confirming that short courses of therapy are effective in the outpatient setting for children with uncomplicated pneumonia.
Florin: The theme of this discussion that you and I have been having today is about revisiting the approach, both in terms of the spectrum and the duration. Can you summarize what you feel to be the largest opportunities to minimize antibiotic overuse in children with pneumonia?
Williams: For me, the Holy Grail here is how can we can limit antibiotic use in children who don't have bacterial disease? I know that's a really hard thing to answer, especially when you have a patient who's breathing fast sitting in front of you and you're concerned about bacterial pneumonia. Where we can make the most impact is not starting antibiotics at all in children who do not have bacterial pneumonia. Again, I know that's a hard thing to answer. There's a lot of exciting stuff happening in the diagnostic space that might help us get there. Short of that, we should focus on using narrow-spectrum therapy as recommended by the guidelines, using amoxicillin in the outpatient setting, using ampicillin in the inpatient setting, avoiding broad spectrum therapy unless it's absolutely necessary, and reducing therapy duration. Pneumonia is extremely common. Even that can have a significant impact on overall antibiotic utilization. I think we can really move the needle.
Florin: That's a great summary, and it's a great place for us to wrap up our discussion. Thank you all for tuning in. Please take a moment to download the Medscape app to listen and subscribe to this podcast series on pediatric pneumonia. This is Dr Todd Florin for the Medscape InDiscussion podcast.
Resources
Potential Impact of Conjugate Pneumococcal Vaccines on Pediatric Pneumococcal Diseases
Forty Years of Molecular Diagnostics for Infectious Diseases
Community-Acquired Pneumonia Requiring Hospitalization Among U.S. Children
Detection of Streptococcus pneumoniae in Whole Blood by PCR
Mycoplasma pneumoniae Infections: Pathogenesis and Vaccine Development
Follow Medscape on Facebook, Twitter, Instagram, and YouTube
Medscape © 2023 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: The Latest Strategies and Clinical Trials Challenging Long-standing Treatment Approaches for Pediatric Pneumonia - Medscape - Jul 12, 2023.
Comments