Lack of Effectiveness for Antibiotics

Context Acute otitis media (AOM) is the most common condition for which antibiotics are prescribed for US children; however, wide variation exists in diagnosis and treatment.

Objectives To perform a systematic review on AOM diagnosis, treatment, and the association of heptavalent pneumococcal conjugate vaccine (PCV7) use with AOM microbiology.

Data Sources PubMed, Cochrane Databases, and Web of Science, searched to identify articles published from January 1999 through July 2010.

Study Selection Diagnostic studies with a criterion standard, observational studies and randomized controlled trials comparing AOM microbiology with and without PCV7, and randomized controlled trials assessing antibiotic treatment.

Data Extraction Independent article review and study quality assessment by 2 investigators with consensus resolution of discrepancies.

Results Of 8945 citations screened, 135 were included. Meta-analysis was performed for comparisons with 3 or more trials. Few studies examined diagnosis; otoscopic findings of tympanic membrane bulging (positive likelihood ratio, 51 [95% confidence interval {CI}, 36-73]) and redness (positive likelihood ratio, 8.4 [95% CI, 7-11]) were associated with accurate diagnosis. In the few available studies, prevalence of Streptococcus pneumoniae decreased (eg, 33%-48% vs 23%-31% of AOM isolates), while that of Haemophilus influenzae increased (41%-43% vs 56%-57%) pre- vs post-PCV7. Short-term clinical success was higher for immediate use of ampicillin or amoxicillin vs placebo (73% vs 60%; pooled rate difference, 12% [95% CI, 5%-18%]; number needed to treat, 9 [95% CI, 6-20]), while increasing the rate of rash or diarrhea by 3% to 5%. Two of 4 studies showed greater clinical success for immediate vs delayed antibiotics (95% vs 80%; rate difference, 15% [95% CI, 6%-24%] and 86% vs 70%; rate difference, 16% [95% CI, 6%-26%]). Data are absent on long-term effects on antimicrobial resistance. Meta-analyses in general showed no significant differences in antibiotic comparative effectiveness.

Conclusions Otoscopic findings are critical to accurate AOM diagnosis. AOM microbiology has changed with use of PCV7. Antibiotics are modestly more effective than no treatment but cause adverse effects in 4% to 10% of children. Most antibiotics have comparable clinical success.

From press release:

Using antibiotics to treat newly diagnosed acute ear infections among children is modestly more effective than no treatment, but comes with a risk of side effects, according to a new study designed to help advise efforts to rewrite treatment guidelines for the common illness.

Researchers found no evidence that name-brand antibiotics work any better in general than generic antibiotics and that careful examination of the eardrum by a clinician for signs of infection is critical for accurate diagnosis of acute ear infections. The study is published in the Nov. 17 edition of the Journal of the American Medical Association.

"Our findings reinforce the existing knowledge that the best antibiotic treatment for common childhood ear infections may be no antibiotic treatment at all," said lead author Dr. Tumaini R. Coker, a pediatrician at Mattel Children's Hospital UCLA and a researcher at the RAND Corporation, a nonprofit research organization.

"Prescribing antibiotics early may help cure ear infections a little bit faster, but also raises the risk that children will suffer antibiotic-related side effects such as a rash or diarrhea," Coker said. "Parents and their children may value these different outcomes differently."

Researchers from the Southern California Evidence-Based Practice Center conducted the study by examining previously published research about the diagnosis and treatment of acute ear infections among children. The review was requested by the American Academy of Pediatrics as part of its effort to update practice guidelines for treating acute ear infections among young children.

The study found that a major limitation of efforts to improve the treatment of acute ear infections is that there is no one definitive test for making the diagnosis. The use of an otoscope to look into the ear for signs of infection is a key component of any diagnosis and improving clinicians' ability to conduct this exam may help improve the accuracy of diagnosis, researchers said.

Acute ear infections are the most common childhood illness in the United States where antibiotics are routinely prescribed. A 2006 study estimated that it costs about $350 to treat each episode of the illness, with yearly national costs totaling $2.8 billion.

The new study reviews published data from 1999 to 2010 about the treatment of acute ear infections among children. This systematic review of the literature found that prescribing antibiotics early for acute ear infections had a modest benefit. Of 100 average-risk children with acute ear infections, about 80 would be expected to get better within about three days without antibiotics. If all were treated with antibiotics immediately at diagnosis, evidence suggests an additional 12 would improve, but three to 10 children would develop a rash and five to 10 would develop diarrhea, according to the study.

"Early prescribing of antibiotics offers a slight benefit and poses a slight risk," said the study's senior author, Dr. Glenn Takata, a pediatrician and researcher at Children's Hospital Los Angeles. "Clinicians will have to take these modest benefits and risks into consideration when deciding whether to treat with antibiotics or not."

Researchers also found no evidence that using newer, name-brand antibiotics to treat uncomplicated acute ear infections in normal-risk children offers an advantage over generic antibiotics, namely the commonly-used amoxicillin. Despite these findings, the higher-priced antibiotics are often prescribed for uncomplicated acute ear infections. Savings could be substantial if physicians used amoxicillin as the first line of defense against uncomplicated acute ear infections, according to researchers.

Funding for the study was provided by the U.S. Agency for Healthcare Research and Quality, which provides ongoing support to the Southern California Evidence-Based Practice Center. The center is based at RAND and the VA Greater Los Angeles Healthcare System.

Other authors of the study are Linda S. Chan and Dr. Mary Ann Limbos of the Keck School of Medicine of USC, Sydne J. Newberry and Marika J. Suttorp of RAND, and Dr. Paul. G. Shekelle of RAND and the VA Greater Los Angeles Healthcare System.