
Community-acquired pneumonia (CAP) is one of the most common infections in children and a top-five cause of pediatric hospitalizations in the U.S. Yet, diagnosing pneumonia at the bedside remains surprisingly difficult. While IDSA/PIDS guidelines advise diagnosing CAP primarily on clinical features without routine chest radiographs, clinicians are left without a concrete set of objective diagnostic criteria—and often find themselves disagreeing on what they actually hear through their stethoscopes.
In this episode, Pediatric Emergency Medicine Physician and Researcher Dr. Todd Florin joins us to unpack the evidence behind clinical predictors of pediatric pneumonia, discuss his JAMA Network Open study on the surprisingly low inter-rater reliability of lung auscultation, and share how to balance chest radiography against antibiotic stewardship.
1. Diagnosing Pneumonia: Clinical Features vs. Imaging
- No Single Slam-Dunk Sign: Large systematic reviews and meta-analyses show that no single clinical sign or symptom reliably predicts radiographic pneumonia. Individual findings like fever, cough, and classic auscultatory changes perform almost like a coin flip on their own.
- Strongest Predictors: Work of breathing and hypoxemia carry the strongest individual association with radiographic pneumonia.
- Combinations Matter: Reliable bedside diagnosis depends on clustering multiple findings—such as prolonged duration of fever, hypoxemia, tachypnea, older age, and focal exam findings—to stratify whether a patient is at low, intermediate, or high risk.
2. The Auscultation Controversy: Do We Actually Agree?
In a JAMA Network Open sub-analysis of 252 children (derived from the ongoing multi-center PEDCAPS cohort), Dr.Florin’s team had two independent clinicians examine the exact same child within 30 minutes to test how often they agreed on physical exam findings:
- Poor Agreement on Classic Findings: Signs traditionally taught as “pathognomonic” for pneumonia showed poor inter-rater agreement even before adjusting for statistical chance:
- Decreased breath sounds: Raw agreement was only 46%.
- Fine crackles / rales: Raw agreement was only 55%.
- What Clinicians Actually Agree On: The only physical exam findings with reasonable, reliable agreement were retractions and wheezing.
- Why We Disagree: Small pediatric chest sizes allow transmitted upper airway sounds to mimic lower tract pathology, stethoscopes have acoustic limitations, and trainees often lack a standardized reference point for labeling lung sounds.
- Clinical Takeaway: Treat an exam finding like any single lab biomarker—never interpret it in isolation. When auscultatory findings are ambiguous, lean heavily on the broader clinical picture (oxygen saturation, work of breathing, and fever duration).
Looking Ahead: The PEDCAPS Study
Dr. Florin and the PECARN network are currently completing external validation for the PEDCAPS (Pediatric Community-Acquired Pneumonia Severity) score—a prospective study aimed at helping clinicians accurately predict which children with pneumonia will progress to moderate or severe disease. Because early clinical gestalt frequently misses children who later deteriorate, knowing which physical exam findings are actually reliable is the critical foundation for building a trustworthy bedside tool.
How do you diagnose pneumonia in kids? Do you routinely get chest x-rays? Perform your own POCUS? We’d love to her hear how you practice and whether this episode changes what you do. Share your thoughts with us on social media @empulsepodcast or at ucdavisem.com
Host:
Dr. Julia Magaña, Professor of Pediatric Emergency Medicine at UC Davis
Guests:
Dr. Todd Florin, Professor of Pediatrics (Emergency Medicine) at Northwestern University Feinberg School of Medicine; Emergency Medicine Associate Division Head for Academic Affairs & Research at Lurie Children’s Hospital; Scientific Director of Research Development at Stanley Manne Children’s Research Institute
Resources:
***************************************************************************************
Thank you to the UC Davis Department of Emergency Medicine for supporting this podcast and to Orlando Magaña at OM Productions for audio production services.
Weitere Episoden von „EM Pulse Podcast™“



Verpasse keine Episode von “EM Pulse Podcast™” und abonniere ihn in der kostenlosen GetPodcast App.








