Identifying Children Likely to Benefit From Antibiotics for Acute Sinusitis: A Randomized Clinical Trial

Abstract

Importance: The large overlap between symptoms of acute sinusitis and viral upper respiratory tract infection suggests that certain subgroups of children being diagnosed with acute sinusitis, and subsequently treated with antibiotics, derive little benefit from antibiotic use.

Objective: To assess if antibiotic therapy could be appropriately withheld in prespecified subgroups.

Design, setting, and participants: Randomized clinical trial including 515 children aged 2 to 11 years diagnosed with acute sinusitis based on clinical criteria. The trial was conducted between February 2016 and April 2022 at primary care offices affiliated with 6 US institutions and was designed to evaluate whether symptom burden differed in subgroups defined by nasopharyngeal Streptococcus pneumoniae, Haemophilus influenzae, or Moraxella catarrhalis on bacterial culture and by the presence of colored nasal discharge.

Interventions: Oral amoxicillin (90 mg/kg/d) and clavulanate (6.4 mg/kg/d) (n = 254) or placebo (n = 256) for 10 days.

Main outcomes and measures: The primary outcome was symptom burden based on daily symptom scores on a validated scale (range, 0-40) during the 10 days after diagnosis. Secondary outcomes included treatment failure, adverse events including clinically significant diarrhea, and resource use by families.

Results: Most of the 510 included children were aged 2 to 5 years (64%), male (54%), White (52%), and not Hispanic (89%). The mean symptom scores were significantly lower in children in the amoxicillin and clavulanate group (9.04 [95% CI, 8.71 to 9.37]) compared with those in the placebo group (10.60 [95% CI, 10.27 to 10.93]) (between-group difference, -1.69 [95% CI, -2.07 to -1.31]). The length of time to symptom resolution was significantly lower for children in the antibiotic group (7.0 days) than in the placebo group (9.0 days) (P = .003). Children without nasopharyngeal pathogens detected did not benefit from antibiotic treatment as much as those with pathogens detected; the between-group difference in mean symptom scores was -0.88 (95% CI, -1.63 to -0.12) in those without pathogens detected compared with -1.95 (95% CI, -2.40 to -1.51) in those with pathogens detected. Efficacy did not differ significantly according to whether colored nasal discharge was present (the between-group difference was -1.62 [95% CI, -2.09 to -1.16] for colored nasal discharge vs -1.70 [95% CI, -2.38 to -1.03] for clear nasal discharge; P = .52 for the interaction between treatment group and the presence of colored nasal discharge).

Conclusions: In children with acute sinusitis, antibiotic treatment had minimal benefit for those without nasopharyngeal bacterial pathogens on presentation, and its effects did not depend on the color of nasal discharge. Testing for specific bacteria on presentation may represent a strategy to reduce antibiotic use in this condition.

Trial registration: ClinicalTrials.gov Identifier: NCT02554383.

Figure 1.. Enrollment, Randomization, and Follow-Up of Children in a Trial of Antibiotics for Acute Sinusitis

^aDefined by the clinical practice guideline from the American Academy of Pediatrics, which defines acute sinusitis as persistent nasal symptoms, cough, or both for 11 to 30 days, or worsening nasal symptoms, worsening daytime cough, or new onset fever in children on days 6 to 10 of illness who appear to be recovering from a viral upper respiratory infection. ^bChildren had a mean Pediatric Rhinosinusitis Symptom Scale score of 22 among those with parents who declined to participate and a score of 24 among those with parents who agreed to participate. In addition, 52% of children had colored nasal discharge (parents declined to participate) vs 67% (parents agreed to participate). Neither the age nor the sex of children whose parents declined to participate differed from that of children whose parents agreed to participate. ^cThere were 391 children who had another acute infection requiring systemic antibiotics (eg, acute otitis media, streptococcal pharyngitis, or pneumonia). ^dThe Pediatric Rhinosinusitis Symptom Scale is a validated scale designed to follow symptoms of children with acute sinusitis. Parents were asked how their child had been doing over the past 24 hours and were required to rate each of 8 symptoms (stuffy nose, runny nose, daytime cough, tiredness, irritability, trouble breathing through the nose, nighttime cough, and trouble sleeping) as none (score of 0), almost none (score of 1), a little (score of 2), some (score of 3), a lot (score of 4), and an extreme amount (score of 5). The 8 ratings were summed to obtain an overall Pediatric Rhinosinusitis Symptom Scale score. Scores ranged from 0 to 40; higher scores indicate greater severity. ^eStratified according to the presence or absence of colored (yellow or green) nasal discharge. At each site of enrollment, within each stratum, children were randomized 1:1 in block sizes of 4. Allocation was concealed.

Figure 2.. Symptom Burden Assessed by Mean Score on the PRSS During the First 10 Days of Follow-Up

A, The mean scores on the PRSS over the course of the first 10 days of follow-up differed significantly between the 2 groups. Study day 1 is the day of enrollment. The PRSS is a validated scale designed to follow symptoms of children with acute sinusitis. Parents were asked how their child had been doing over the last 24 hours by rating each of 8 symptoms (stuffy nose, runny nose, daytime cough, tiredness, irritability, trouble breathing through the nose, nighttime cough, and trouble sleeping) as none (score of 0), almost none (score of 1), a little (score of 2), some (score of 3), a lot (score of 4), and an extreme amount (score of 5). The 8 ratings were summed to obtain an overall PRSS score. Scores ranged from 0 to 40; higher scores indicate greater severity. The median number of daily electronic symptom diary entries was 10 in both treatment groups. B, The between-group difference in symptom scores depends on whether at least 1 pathogen (Streptococcus pneumoniae, Haemophilus influenzae, or Moraxella catarrhalis) was present in the bacterial culture of the nasopharynx at the time of diagnosis (P = .02 for the interaction between nasopharyngeal colonization and treatment). C, The between-group difference in symptom scores was not related to color of nasal discharge at the time of diagnosis (P = .52 for the interaction between nasal discharge and treatment).