Stewardship in the pediatrician’s office: Antibiotic use in community-acquired pneumonia

Ulka Kothari

Asif Noor

Community-acquired pneumonia, or CAP, is an inflammation of the lungs caused by a microbe. It is a lower respiratory tract infection acquired in the community as opposed to hospital-acquired or nosocomial pneumonia.

Cases of pneumonia are seen in the pediatric office throughout the year, with increases in prevalence during the fall and winter months. Pediatric office visit rates for pneumonia ranged from 16.9 to 22.4 cases per 1,000 children between 1994 and 2007. Children aged younger than 5 years accounted for most of these cases. Over the past 2 decades, a decline in childhood hospitalizations due to pneumonia coincided with universal pneumococcal vaccination.

Viruses and bacteria are responsible for CAP in infants and children. The true prevalence of the etiologic agent is difficult to determine because of the lack of dependable diagnostic tests and the difficulty in differentiating true pathogens from colonization. However, the age of the child, clinical characteristics and chest X-ray appearance may help differentiate viral from bacterial causes. In addition, the advent of rapid viral testing, both antigen and respiratory multiplex PCR, can assist further in making this important distinction. In this column, we discuss the clinical features, viral diagnostics and chest X-ray findings that aid in distinguishing viral from bacterial pneumonia and the choice of empiric antibiotic treatment.

1. Confronting the conundrum of appropriate antibiotic prescription

Microbiology

The microbiology of CAP varies by age. Viruses account for most cases in young children aged younger than 2 years. Mycoplasma is the most common cause of CAP in school-aged children, aged older than 5 years. Overall, viruses make up for 66% of cases in children, bacteria make up 8%, and a mix of viruses and bacteria comprise 7% of cases among children aged younger than 18 years. When categorized by age, viruses accounted for 80% cases among children aged younger than 2 years and 49% of cases among those aged older than 2 years.

Respiratory syncytial virus is the most common cause of CAP in children. It is responsible for 28% of all cases, with the highest incidence in children aged younger than 2 years. Other viruses such as influenza and human metapneumovirus are also significantly associated with childhood pneumonia. The role of parainfluenza virus, rhinovirus and nonseasonal coronaviruses is difficult to attribute to CAP because these viruses can also be detected in well-appearing children, and because of issues with prolonged shedding. During the current COVID-19 pandemic, SARS-CoV-2 is an important cause of pneumonia, particularly in adolescents with high BMI and underlying medical problems.

Regarding bacterial pathogens, Streptococcus pneumoniae is the most common cause of CAP in all age groups. In the post-Prevnar 13 (Pfizer) era, it is more common among older children and those with underlying medical problems. An atypical bacterial pathogen, Mycoplasma pneumoniae, is implicated in CAP cases among school-aged children and adolescents.

Clinical presentation

Classic symptoms of pneumonia include acute onset of fever, cough and difficulty breathing. However, there exists variability based on the pathogen and host characteristics. Infants might only have poor feeding and irritability. Children aged younger than 5 years might have upper respiratory symptoms of rhinorrhea and low-grade fever, followed by signs and symptoms of lower respiratory infection. Older children and adolescents may complain of pleuritic chest pain.

The clinical signs and symptoms do not distinguish between viral and bacterial causes. Nonetheless, a gradual progression from upper respiratory tract signs and symptoms to lower respiratory tract infection in an otherwise well-appearing child is consistent with a viral infection. On the other hand, an abrupt increase in fevers, difficulty breathing and toxic appearance suggests bacterial pneumonia.

Chest X-ray

Chest X-rays should be reserved for children who require hospitalization, or to look for complications such as empyema. In the office setting, if diagnosis in uncertain, a chest X-ray may be obtained. The presence of bilateral diffuse infiltrates suggest viral pneumonia. This pattern can occasionally be seen with Mycoplasma pneumonia in older children. Lobar or segmental opacification or large pleural effusion indicates a bacterial infection.

Diagnostic testing

In the office setting, the availability of either a rapid antigen test (influenza, SARS-CoV-2) or a PCR panel (respiratory multiplex for common respiratory viruses as well as SARS-CoV-2 PCR) can assist in confirmation of a viral infection. This information can decrease the need for further testing and help avoid antibiotic prescriptions. Routine use of complete blood count, blood cultures and inflammatory markers are not specific and thereby not recommended.

Choice of antibiotic

In the office setting, the choice of antibiotic is empiric.

• Most infants aged older than 6 months and children aged younger than 5 years have a viral pneumonia. If the child is well appearing, well hydrated, with the presence of upper respiratory tract signs such as rhinorrhea, antibiotics are not indicated. A rapid viral antigen or PCR test can assist with the decision to hold empiric antibiotic treatment.

• Children with confirmed influenza can be treated with neuraminidase inhibitors such as oseltamivir.

• If a child is suspected to have bacterial pneumonia based on a focal physical exam with findings of pneumonia (ie, crackles and/or decreased air entry, chest X-ray showing presence of lobar consolidation), high-dose amoxicillin 90 mg/kg per day every 12 hours is the first-line agent. Oral cephalosporins are inferior to amoxicillin, given their low bioavailability and susceptibility compared with amoxicillin.

• Children with non-type I hypersensitivity can be treated with a second- or third-generation cephalosporin, such as cefdinir. Children with type I hypersensitivity can be treated with oral clindamycin (first-line therapy) or oral levofloxacin (second-line therapy).

• Children aged older than 5 years who have clinical signs and symptoms suggestive of atypical pneumonia, such as bilateral crackles but are otherwise well appearing can be treated with an oral macrolide, azithromycin being the first-line antibiotic of choice.

• The duration of treatment for typical pneumonia is 5 days for otherwise healthy children and 7 days for children with underlying medical problems or those with slow initial response in cases of typical CAP. Duration of treatment for atypical pneumonia is 5 days.

2. Practice guidelines for outpatient settings

See the Infectious Diseases Society of America guidelines for recommendations on the management of CAP in infants and children.

3. Operationalizing an antimicrobial stewardship program (ASP) in your office through quality improvement and electronic health records

Antibiotic prescribing for common pediatric infections varies among practices, and in at least one study, patient-specific factors could not explain this variability. A deeper dive into understanding and addressing provider barriers would help design intervention strategies. Personalized audit and feedback on primary care providers' antibiotic prescribing practices for CAP positively impacted clinicians' selection of guideline-recommended antibiotics. ASP quality improvement efforts use education coupled with a documented and agreed-upon care process model, electronic decision support, data tracking and feedback to frontline providers. The CDC's core elements of outpatient antibiotic stewardship succinctly summarize these interventions and can be found here.

References:

Bradley JS, et al. Clin Infect Dis. 2011;doi. 10.1093/cid/cir531.

Diaz MCG, et al. Clin Pediatros (Phila). 2020;doi:10.1177/0009922820928054.

Gerber JS, et al. J Pediatric Infect Dis Soc. 2015;doi:10.1093/jpids/piu086.

Griffin MR, et al. MMWR Morb Mortal Wkly Rep. 2014;63:995-998.

Jain S, et al. N Engl J Med. 2015;doi:10.1056/NEJMoa1405870.

Kaplan SL, et al. Abstract 77. IDWeek; Oct. 8-12; Philadelphia.

Kronman MP, et al. Pediatrics. 2011;doi:10.1542/peds.2010-2008.