AOM and AOMT are different entities.

Some controversy exists as to whether acute otitis media (AOM) and acute otitis media through tympanostomy tubes (AOMT) are different entities. The issue is clinically relevant because if there is a difference in their natural history, then there might be a difference in the way each is treated. I believe without question that they are indeed different.

Microbiology

Microbiologically, the organisms associated with AOM are Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis, and some miscellaneous respiratory bacterial pathogens. (1) In AOMT, the most common organisms are H influenzae, S pneumoniae, Staphylococcus aureus, Pseudomonas aeruginosa, fungi, and miscellaneous respiratory, skin, oral, and even bowel flora. (1) So while AOM is associated with the usual respiratory pathogens, AOMT is linked to those as well as to organisms that are associated with water and skin.

Natural history

AOM. Rosenfeld and Kay performed a meta-analysis of 63 articles to determine the natural history of untreated AOM and otitis media with effusion (OME). (2) They found that symptoms of AOM resolved in 61% of children within 24 hours of onset and in 80% within 3 days. Moreover, there was no statistically significant difference in the rates of suppurative complications between treated and untreated children.

In another meta-analysis, Takata et al analyzed 80 articles and reported that only 19% of untreated children were still clinically symptomatic 7 days after the onset of AOM and that few of them experienced any suppurative complications. (3) They also found that treatment with amoxicillin reduced the 7-day cure rate by only 12%. Finally, they found no evidence to support the choice of one antibiotic regimen over any another.

AOMT. By far the best study of the natural history of AOMT was published in 2003 by Ruohola et al in Finland. (4) They conducted a randomized, double-blind, placebo-controlled trial of amoxicillin/clavulanate in a volunteer sample of 66 children, aged 6 months to 6 years (mean: 2 yr 1 mo), who experienced episodes of AOMT that lasted less than 48 hours. Patients received either active treatment with 45 mg/kg/day of amoxicillin/clavulanate (n = 34) or matching placebo (n = 32) for 7 days; all patients also underwent daily suction of middle ear fluid through the tube. At study's end, otorrhea had resolved in 28 of the 34 treated children (82.4%) and in 13 of 32 controls (40.6%)--a statistically significant difference. Also, the duration of otorrhea was significantly shorter in the active-treatment group (3 vs. 8 days).

A comparison of the results of these three studies suggests that antibiotic treatment does not confer much of a benefit in AOM (at least as it was diagnosed in these studies), but it does have a significant impact on AOMT. This supports the idea that the two are different diseases.

Definitions

We would perhaps be able to more easily differentiate AOM from AOMT if we had better working definitions of the two entities.

AOM. One reason that defining AOM has been problematic is the vagaries of AOM studies. For example, many of the children who were enrolled in AOM studies in the past did not actually have AOM. We now know that many AOM trials that were conducted in the 1980s included patients with OME. In these studies, we were often treating a syndrome rather than a specific disease with a specific cause. These infectious or noninfectious entities had a variety of causes. I would say that at most, only 75 to 80% of these patients actually had AOM; a more likely figure is 60%. Granted, it is difficult to distinguish between AOM and OME; the pediatric otolaryngologists at the University of Pittsburgh went to great lengths to train their personnel how to differentiate ears with middle ear fluid from those without fluid, but my estimate is that they were less successful at differentiating infected fluid (AOM) from noninfected fluid.

Another complicating factor is that it was considered unethical to give placebo to children with AOM if they had severe otalgia and a bulging tympanic membrane. Therefore, many of these children were excluded from trials, and the remaining study population included patients with only mild or moderate disease. This practice of excluding serious cases increased the likelihood that the diagnosis in these children was equivocal and that they might not have had AOM at all.

A third problem is that the doses of amoxicillin typically used in the earlier studies were assumed to be representative of effective or even optimal antibiotic therapy. But studies at that time involved the use of low-dose amoxicillin--typically 20 to 30 mg/kg/day in 3 divided doses. That dose, as we now know from double-tap eradication studies, is likely to fail to eradicate 25 to 40% of even the beta-lactamase-negative strains, and it appears to be no better than placebo against beta-lactamase-producing strains of H influenzae and M catarrhalis. (5,6)

Finally, most comparative antibiotic trials in AOM are designed to show only noninferiority rather than superiority. With such a goal, investigators can almost guarantee a desired result by enrolling the right number of patients with the right set of clinical characteristics. And in those few cases where the desired results are not achieved, the results can be selectively cherry-picked or completely buried. Therefore, the data that go into the mix that makes up "evidence-based medicine" are skewed because the evidence is incomplete and may even be misleading.

AOMT. AOMT can be viewed as what I might call "quasi-external otitis" in a nonintact eardrum. The reason I call it "quasi-external otitis" is that most AOMT patients--even those with P aeruginosa infection--do not have particularly tender or otherwise symptomatic ear canals, at least not as symptomatic as the ear canals in patients with external otitis. While AOMT patients have some pathogens typically seen in external otitis (e.g., P aeruginosa and S aureus), they also have some pathogens that are not typical in external otitis, and these atypical pathogens may be reaching the middle ear.

Therapeutic options

In AOM, the tympanic membrane is intact, and ototopical agents placed in the ear canal do not reach the middle ear. Therefore, systemic therapy (or surgical drainage) is required to treat the middle ear infection. Because of the rarity of pseudomonads and staphylococci in AOM, we do not need to treat for those organisms and the range of potential organisms is more narrow. While the use of systemic steroids with antimicrobials in AOM would likely hasten recovery, such use would also expose children to the systemic problems associated with steroids.

In AOMT we can administer topical antibiotics and steroids directly to the site of infection without significant systemic exposure. The high concentration of antibiotics in ototopicals--even the amount that makes it through the tube and into the middle ear and mastoid system--is almost always higher than the minimum inhibitory concentration of what our microbiology laboratories would characterize as "resistant strains" (a definition that is based on antibiotic levels that can be achieved by systemic therapy).

In conclusion, some cases of AOMT are similar to AOM, but there are substantial differences in the range of causes, in the apparent spontaneous resolution rate, and in the options for therapy. The astute clinician will not be misled by the power and verbiage of meta-analyses and what is often erroneously called "evidence-based medicine." Rather, one should be able to place those data within the context of the limitations of study designs, the importance of tangential (e.g., basic science and pharmacokinetic/pharmacodynamic) data, and one's own clinical experience. The herd may be wrong. And in the case of the natural history of true AOM, we now know that for 20 years, the herd was indeed wrong.

Michael D. Poole, MD, PhD

References

(1.) Roland PS, Parry DA, Stroman DW. Microbiology of acute otitis media with tympanostomy tubes. Otolaryngol Head Neck Surg 2005;133:585-95.

(2.) Rosenfeld RM, Kay D. Natural history of untreated otitis media. Laryngoscope 2003; 113:1645-57.

(3.) Takata GS, Chan LS, Shekelle P, et al. Evidence assessment of management of acute otitis media: 1. The role of antibiotics in treatment of uncomplicated acute otitis media. Pediatrics 2001; 108:239-47.

(4.) Ruohola A, Heikkinen T, Meurman O, et al. Antibiotic treatment of acute otorrbea through tympanostomy tube: Randomized double-blind placebo-controlled study with daily follow-up. Pediatrics 2003;111:1061-7.

(5.) Dagan R, Leibovitch E, Fliss DM, Leiberman A. Treatment failures in otitis media--What can we learn? Ear Nose Throat J 1998;77(6 suppl): 16-19; discussion 20-1.

(6.) Dagan R. McCracken GH Jr. Flaws in design and conduct of clinical trials in acute otitis media. Pediatr Infect Dis J 2002;21:894-902.