Macrolides may not reduce hospitalization length in paediatric pneumonia
The addition of a macrolide* to empirical β-lactam** therapy did not reduce length of hospitalization in children with pneumonia, a recent study found, raising questions on the routine use of macrolides in this setting.
Using data from the multicentre, prospective EPIC*** Study, researchers analysed 1,418 children (median age, 27 months) hospitalized at three children’s hospitals in Tennessee and Utah with radiographically-confirmed community-acquired pneumonia between January 2010 and June 2012 who during the first 2 days of hospitalization, received either β-lactam monotherapy (n=1,019) or β-lactam plus macrolide combination therapy (n=399). Azithromycin was the predominant macrolide used (98.2 percent). Atypical bacteria were present in 8.8 percent of the children, primarily Mycoplasma pneumoniae (95.2 percent), while a virus was present in 73.7 percent.
Length of hospitalization was comparable between children who received β-lactam monotherapy and β-lactam plus macrolide combination therapy (median, 55 vs 59 hours, adjusted hazard ratio [HR], 0.87, 95 percent confidence interval [CI], 0.74–1.01). Results were similar in the propensity score-matched cohort which comprised 554 children (median, 53 vs 62 hours, HR, 0.88, 95 percent CI, 0.74–1.03). [JAMA Pediatr 2017;doi:10.1001/jamapediatrics.2017.3225]
The addition of macrolides did not improve hospitalization length in patients who should have derived the most benefit (children aged >5 years, those with presence of atypical bacteria, those in intensive care, and those with acute wheezing, of whom 55.2, 60.8, 29.3, and 25.7 percent, respectively, received the combination treatment; HR, 1.08, 1.07, 0.86, and 0.93, respectively, for combination therapy vs monotherapy).
The addition of macrolides also did not affect admittance to intensive care after the first day of hospitalization, with a similar number of children in the combination and monotherapy groups admitted to intensive care (38 percent vs 35 percent, adjusted odds ratio [adjOR], 1.04, 95 percent CI, 0.41–2.64).
Of the 873 children in the unmatched cohort who were followed up 3–10 weeks after hospital discharge, 769 had recovered from the initial illness, 86.9 and 91.1 percent of those who received monotherapy and combination therapy, respectively (adjOR, 1.11).
“In this study, almost 30 percent of children received macrolide combination therapy, even though atypical pathogens were detected in less than 9 percent ... our study did not demonstrate benefits of empirical macrolide therapy in those with atypical bacteria,” said the researchers.
“Thus, clinicians must weigh the theoretical individual benefits of empirical macrolide therapy against the risk of adverse drug effects and the societal risks associated with antimicrobial resistance,” they said.
The high incidence of viral infection suggests that antibiotics may not be necessary for a majority of children who are hospitalized for pneumonia, said Dr Michael J. Smith from Duke University, Durham, North Carolina, US, in an editorial. [JAMA Pediatr 2017;doi:10.1001/jamapediatrics.2017.3828]
“[I]t would seem that a significant benefit of azithromycin would have been observed in the EPIC trial if it existed. This is not to say macrolides should be avoided completely. There are clearly cases in which they are needed. But we should approach routine azithromycin use with a healthy scepticism,” he said.
The researchers acknowledged that the observational study design, baseline differences between the two groups, and limiting the study to antibiotic therapy received in the first 2 days of hospitalization may have affected the results, and called for large, multicentre studies to confirm these findings, particularly among children who should benefit the most from macrolide therapy.