Pediatric Pneumonia, Diagnostic Uncertainty, and Communication

By Philip R. Fischer, MD, DTM&H

SYNOPSIS: Community-acquired pneumonia is a common cause of hospitalization among children in resourced countries and is the leading cause of childhood death in lower-income countries. However, the diagnosis of pneumonia is fraught with uncertainty, and management practices vary widely. Clinician behaviors and attributes that engender trust in the face of diagnostic uncertainty have been documented.

SOURCE: Lehrer BJ, Mutamba G, Thure KA, et al. Optimal pediatric outpatient antibiotic prescribing. JAMA Netw Open 2024;7:e2437409.

Antibiotics are prescribed more often for pediatric patients in the United States than are any other sorts of medications. In fact, there are approximately 60 antibiotic prescriptions for every 100 children each year. In the state of Tennessee, the rate of antibiotic prescriptions for pediatric outpatients is 50% greater than the national average. The majority of these prescriptions are unnecessary and/or inappropriate. Of course, inappropriate antibiotic use is associated with antimicrobial resistance, adverse clinical reactions, and health-altering changes in the microbiome.

In the face of antibiotic overuse, there are limited data about the actual prescribing of antimicrobial therapy that involves the most appropriate antibiotic for the most appropriate duration. Thus, investigators in Tennessee used state-wide data to evaluate the appropriateness of outpatient antibiotic prescriptions for children.

Lehrer and colleagues used large (involving 90% of medical care in Tennessee) databases to link diagnostic codes with the name, dose, frequency, and duration of prescribed antibiotics. Outpatient encounters were included in the study when patients were younger than 20 years of age and received either an oral antibiotic or outpatient intramuscular ceftriaxone or penicillin during the 365 days of 2022. Encounters were excluded from study if related to travel medicine, pregnancy, or an immunocompromised state. Based on critically acclaimed, published guidelines, diagnoses were classified as Tier 1 (essentially always require antibiotic treatment), Tier 2 (sometimes require antibiotic treatment), and Tier 3 (essentially never require antibiotic treatment). Diagnosis-specific antibiotic treatment of Tier 1 and Tier 2 conditions was assessed for optimal choice and duration of the antibiotic. Tier 3 antibiotic prescriptions were automatically considered non-optimal in terms of both choice and duration.

During 488,818 encounters with pediatric patients (near-equal numbers of boys and girls, mean age 8 years), a total of 506,633 antibiotic prescriptions were provided, with 4% for Tier 1 diagnoses, 57% for Tier 2 diagnosis, and 39% for Tier 3 diagnoses (for which antibiotic treatment was not indicated). The duration of antimicrobial therapy was for five or more days in 99% of prescriptions.

The most common diagnoses prompting antibiotic prescriptions were otitis media, pharyngitis, and sinusitis. The most commonly used antibiotics were amoxicillin, cefdinir, and amoxicillin-clavulanate. The investigators determined that the specific antibiotic was appropriate for the diagnosed condition in just 39% of cases, 51% provided treatment for an optimal duration, and 31% of prescriptions were for both an appropriate antibiotic and an optimal duration.

Only 1.2% of antibiotic prescriptions were for pneumonia; 61% of those involved an appropriate first-line antibiotic choice. Nearly 90% of antibiotic prescriptions for pneumonia were for five or more days, in line with previous guidelines — even though five days is now considered adequate for most patients.

The authors highlighted the “substantial opportunity” for clinician education about antimicrobial stewardship in Tennessee. They also noted the need to disseminate knowledge that five days of antibiotic treatment is adequate for community-acquired pneumonia.

Commentary

Lehrer and colleagues accepted the given diagnoses as accurate, even though we know that not all clinicians agree on clinical diagnoses of otitis media, sinusitis, and pneumonia. While accepting that these conditions might be over-diagnosed as caused by bacterial etiologies, the majority of antibiotic-treated children in Tennessee received the wrong antibiotic and/or too much antibiotic. There is clearly a need for improved antimicrobial stewardship by clinicians managing sick children.

In the United States, there have been clear guidelines for the management of children with non-severe otitis media for more than a decade; antibiotics are not usually required.¹ Similar guidelines suggest that symptomatic sinusitis should be either significantly worsening or persisting for more than 10 days before antibiotics are indicated. It is unlikely that those strict criteria were followed in the huge number of patients with “sinusitis” in the Tennessee study.²

Since 2011, amoxicillin has been the first-line treatment for children with uncomplicated pneumonia; even then, it was considered that short (less than 10 days) treatment may be effective.³ In Canada, coupling new treatment guidelines with prospective audits and feedback about stewardship was effective in improving appropriate antibiotic use for children with pneumonia.⁴

So, why are such common infections treated in ways that are inconsistent with evidence-based guidelines? Sometimes the answer is related to clinicians having significant risk-aversion. They do not want to risk a poor outcome (or patient and family dissatisfaction) if a particular patient happens to be the rare individual who seemed well at the time of the clinical encounter but goes on to develop serious consequences. Sometimes, especially in cases of respiratory symptoms raising the question of pneumonia, over-treatment is the result of legitimate diagnostic uncertainty.

Truly, as noted elsewhere in this issue of Infectious Disease Alert for adults with pneumonia, the diagnosis of pediatric pneumonia is subject to diagnostic uncertainty and treatment variation. In children with cough and fever, excessive tachypnea is a key finding to suggest bacterial disease. However, fever alone can raise a respiratory rate, and viral infections also can cause tachypnea. In children, as noted in adults, interpretation of chest imaging studies is imprecise.

The etiologic epidemiology of community-acquired pneumonia changed with the addition of routine pediatric vaccinations against Haemophilus influenzae (1980s) and Streptococcus pneumoniae (2000s) during the past four decades.⁵ Now, 80% to 90% or more of children presenting with radiologically confirmed community-acquired pneumonia have viruses, rather than bacteria, causing their illness.⁵ Prior to the COVID-19 pandemic, viruses (especially respiratory syncytial virus in children younger than 5 years of age) were the most common cause of lower respiratory tract infection in hospitalized children in the United States, and Mycoplasma pneumoniae was the most common bacterial cause (especially in children older than 5 years of age).⁵

Physical exam findings, laboratory testing, and radiological imaging are incompletely reliable to differentiate viral from bacterial lower respiratory tract infection in children; testing beyond the physical exam is not usually recommended for children well enough to be treated as outpatients.⁵ Etiologic testing from lung aspirates could be used if needed in hospitalized children but is more invasive than necessary for less ill pediatric patients.⁵ Aspirates expectorated or sampled through the oropharynx often are contaminated by bacteria and viruses that are not etiologically responsible for the child’s symptoms.⁵ Thus, treatment decisions must be based on a thoughtful clinical diagnosis based on the child’s presentation while considering likely etiologies in similarly aged children.⁵

Children with non-severe lower respiratory tract infection likely have viral disease and may be offered supportive care. Amoxicillin (to cover pneumococcus, especially in young or unvaccinated children) or a macrolide, such as azithromycin (to cover mycoplasmal infection, especially in older children), may be considered, but watchful waiting without antibiotic therapy while observing the clinical course also is very reasonable.⁵ In fact, there is an active clinical trial testing whether placebo is as good as azithromycin for children with confirmed mycoplasmal pneumonia.⁶ More seriously ill children with tachypnea and chest retractions who require hospitalization still could have either bacterial (especially if with focal lung findings) or viral (especially if with diffuse lung findings and/or wheeze) disease but would more likely receive an antibiotic.

It has been said that acute diagnosis in children “is a complex, iterative, and nonlinear process, often occurring over time.”⁷ Faced with challenges in accurately determining the etiology of lower respiratory tract infection in children, how do clinicians handle diagnostic uncertainty? When seeing sick children, pediatric clinicians in a tertiary care setting consider “red flags” and “gut feelings” as they seek to avoid evolving “cannot miss” diagnoses.⁷ They might take an intentional pause and/or seek a colleague’s “fresh eyes” before finalizing a diagnosis, especially with a very sick child.⁷ Clinicians tend to communicate diagnostic uncertainty clearly among colleagues; however, they are less forthright with patients and families, sometimes with fear of losing trust.⁷

Faced with diagnostic uncertainty, what do patients and families want? Interestingly, when patients are asked what they consider ideal physician behaviors, they rarely mention knowledge-related behaviors. Rather, they prefer that physicians display confident, empathetic, humane, personal, forthright, respectful, and thorough behaviors.⁸ Technical skills and knowledge still have value, but patients prioritize interpersonal and relational skills in assessing ideal physicians.⁸

A recent Canadian study assessed trust of physicians by parents of children with complex medical problems.⁹ Fostering collaboration between the family and the clinician and maintaining a flow of communication between the clinician and the family were key attributes that engendered trust.⁹ Presumably these same characteristics would be appreciated by parents of children whose presentations generate diagnostic uncertainty. Communication should be open as clinical courses evolve.

Patients and families also have identified words and phrases that should “never” be used during episodes of serious and uncertain illness.¹⁰ One such phrase is “everything will be fine.”¹⁰ Patients and families do not desire superficial (and unfounded) reassurance; they prefer realistic and humane support.¹⁰ Whether faced with diagnostic uncertainty or serious illness, patients prefer to hear “I’m here to support you throughout this process.”¹⁰ Seeing the diagnosis of pediatric pneumonia as a longitudinal process, we can give children and their families the help they need, usually without antibiotics.

References

1. Lieberthal AS, Carroll AE, Chonmaitree T, et al. The diagnosis and management of acute otitis media. Pediatrics 2013;131:e964-e999.
2. Wald ER, Applegate KE, Bordley C, et al. Clinical practice guideline for the diagnosis and management of acute bacterial sinusitis in children aged 1 to 18 years. Pediatrics 2013;132:e262-e280.
3. Bradley JS, Byington CL, Shah SS. The management of community-acquired pneumonia in infants and children older than 3 months of age: Clinical practice guidelines by the Pediatric Infectious Diseases Society and the Infectious Diseases Society of America. Clin Infect Dis 2011;53:e25-e76.
4. Le Saux NMA, Bowes J, Viel-Thériault I, et al. Combined influence of practice guidelines and prospective audit and feedback stewardship on antimicrobial treatment of community-acquired pneumonia and empyema in children: 2012 to 2016. Paediatr Child Health 2020;26:234-241.
5. Meyer Sauteur PM. Childhood community-acquired pneumonia. Eur J Pediatr 2024;183(3):1129-1136.
6. Meyer Sauteur PM, Seiler M, Tilen R, et al. A randomized controlled non-inferiority trial of placebo versus macrolide antibiotics for Mycoplasma pneumoniae infection in children with community-acquired pneumonia: Trial protocol for the MYTHIC Study. Trials 2024;25:655.
7. Patel SJ, Ipsaro A, Brady PW. Conversations on diagnostic uncertainty and its management among pediatric acute care physicians. Hosp Pediatr 2022:e2021006076.
8. Bendapudi NM, Berry LL, Frey KA, et al. Patients’ perspectives on ideal physician behaviors. Mayo Clin Proc 2006;81:338-344.
9. Dewan T, Whiteley A, MacKay LJ, et al. Trust of inpatient physicians among parents of children with medical complexity: A qualitative study. Front Pediatr 2024;12:1443869.
10. Lee Adawi Awdish R, Grafton G, Berry LL. Never-words: What not to say to patients with serious illness. Mayo Clin Proc 2024;99:1553-1557.

Philip R. Fischer, MD, DTM&H, is Professor of Pediatrics, Department of Pediatric and Adolescent Medicine, Mayo Clinic, Rochester, MN.