乌鸦传媒

Antibiotic treatment provided "minimal" benefit for children with acute sinusitis and was accompanied by more cases of clinically significant diarrhea, researchers reported.

In a randomized trial of kids ages 2 to 11 years, mean symptom scores during the first 10 days were significantly lower in those who received amoxicillin and clavulanate compared with a placebo group (9.04 vs 10.60 on the Pediatric Rhinosinusitis Symptom Score).

That difference translated to a significant between-group difference of -1.69 (95% CI -2.07 to -1.31) on the 40-point scale, according to researchers led by Nader Shaikh, MD, MPH, of UPMC Children's Hospital of Pittsburgh, and colleagues.

In addition, the length of time to symptom resolution was significantly lower for children in the antibiotic group (7 days) versus the placebo group (9 days; P=0.003), they wrote in .

Although small, the differences were probably noticeable to the children and to their parents, Shaikh and team said. "A between-group difference of 2 points in the mean symptom scores would be seen when comparing a hypothetical child with 'an extreme amount' of irritability from day 2-11 versus a child with 'some' irritability," they wrote.

Of note, antibiotics provided more benefit to those with nasopharyngeal pathogens detected compared with those without pathogens detected, with between-group differences in mean symptom scores of -1.95 (95% CI -2.40 to -1.51) and -0.88 (95% CI -1.63 to -0.10), respectively.

Efficacy did not significantly differ whether colored nasal discharge was present or not (between-group difference -1.62 for colored nasal discharge vs -1.70 for clear nasal discharge; P=0.52 for the interaction between treatment group and the presence of colored nasal discharge).

As for secondary outcomes, children in the antibiotic group were significantly less likely to experience treatment failure, develop acute otitis media, or receive additional systemic antibiotics compared with those in the placebo group.

The study provides some key context for therapeutic decision making, but it isn't practice-changing, wrote Aaron M. Milstone, MD, MHS, of Johns Hopkins University School of Medicine in Baltimore, and colleagues in an , noting that because nasopharyngeal swabs were cultured for all of the patients, the results might not be applicable to most therapeutic settings.

Although the study "offers interesting nuances related to 3 of the 4 moments of antibiotic decision making, it does not seem to be a game changer in the understanding of acute bacterial sinusitis," they added.

"Overall, the current study suggests that antibiotics may provide mild relief in symptom burden in children with acute bacterial sinusitis, especially in those with common respiratory bacteria," they concluded. "However, given the limitations of nasopharyngeal bacterial pathogen testing, widespread adoption of a test-based treatment approach is not there yet."

Universal, or at least more frequent, testing would greatly improve antibiotic stewardship, Shaikh and team said. If antibiotic use was limited to those with culture-proven bacterial infections, it would have reduced use by 28%.

In an exploratory analysis, Shaikh and team determined that the highest clinical benefit with antibiotics was seen in patients infected with Streptococcus pneumoniae or Haemophilus influenzae. There was no significant interaction between treatment and the presence of Moraxella catarrhalis.

"If antibiotic use was limited to children with H. influenzae or S. pneumoniae in their nasopharynx at the time of diagnosis, antibiotic use would decrease by 53%," they pointed out.

In a second accompanying editorial published in , Sophie G. Shay, MD, of the Medical College of Wisconsin in Milwaukee, and Jennifer J. Shin, MD, SM, of Brigham and Women's Hospital and Harvard Medical School in Boston, noted that "[d]evelopment of a ... practical point-of-care test option for acute bacterial rhinosinusitis could invoke a paradigm shift for ambulatory care professionals assessing patients whose presentation raises uncertainty in the diagnosis of viral or bacterial acute rhinosinusitis."

"Point-of-care multiplex polymerase chain reaction testing with same-day results could potentially be leveraged to reduce antibiotic prescriptions," they wrote. "While this approach requires nasopharyngeal swabs, which have traditionally been viewed as invasive by patients and families, as the authors note, in the setting and needs of the COVID-19 pandemic, such swabs have become more frequently accepted. In this setting, swabbing amid acute rhinosinusitis could be a more feasible proposition."

Study Details

All of the children included in the study had to meet the American Academy of Pediatrics' diagnostic criteria for acute sinusitis: persistent illness, which the study defined as symptoms for 11-13 days without improvement; worsening or new onset of nasal discharge, daytime cough, or fever after initial improvement; or severe onset with concurrent fever and purulent (yellow or green) nasal discharge for at least 3 consecutive days.

Of the 510 children included, 64% were ages 2 to 5 years, 54% were boys, and 52% were white.

The children were divided into two groups based on the presence or absence of purulent nasal discharge (67% had colored nasal discharge present). They were randomized to a 10-day course of oral amoxicillin (90 mg/kg per day) and clavulanate (6.4 mg/kg per day) or matching placebo.

At enrollment, nasopharyngeal specimens for culture to identify S. pneumoniae (26-30% of patients), H. influenzae (29-31%), and M. catarrhalis (44-53%) were obtained.

The primary outcome was measured by the 40-point Pediatric Rhinosinusitis Symptom Score during the first 10 days after diagnosis.

There were no serious adverse events associated with treatment. Children who received antibiotics were no more likely to have a treatment-resistant pathogen at follow-up than those who received placebo (13% vs 12%). However, more than twice as many developed clinically significant diarrhea (11.4% vs 4.7%; P=0.005). Rash was also twice as common, but the difference was not statistically significant.