Otitis media: the viral alternative

June 2002

---Philip A. Brunell, MD

There appears to be increasing frustration with the treatment of acute otitis media (AOM) with antibacterial drugs. Perhaps then it is appropriate to explore the role of respiratory viruses in AOM.

Viral agents have been suspected in cases of AOM in which no bacterial pathogen has been recovered. Clinical and virologic data supporting this association of viruses with AOM have been accumulating over the years. The association of otitis media with bronchiolitis is well recognized. As respiratory syncytial virus is the most common cause of bronchiolitis, it must somehow be associated with ear infection. Laboratory evidence linking this and a variety of other respiratory viruses with AOM has been accruing (New Engl J Med. 1999;340:260; Pediatrics. 2002;109:826).

In addition, it has been recognized for more than two decades that the changes in middle ear pressure observed prior to the onset of bacterial otitis, negative pressure followed by effusion, are what might be expected if viral infections preceded AOM (J Pediatr. 1979;63:435). Thus, even when viruses themselves were not the cause of otitis, by changing middle ear dynamics they predisposed to bacterial otitis media. Perhaps, then it is time to reconsider our half century long antibacterial approach to AOM.

Isolation techniques

Despite the clinical evidence linking viruses to AOM, establishing their role by isolation of viruses from middle ear fluid generally has been unrewarding. However, using techniques that did not require the isolation of viruses clearly demonstrate the association of viruses. In a series of studies of children with AOM by the Galveston group by a combination of serologic and of antigen testing of middle ear fluids and nasal wash specimens evidence for viral involvement was found in 41%. RSV was the most common virus identified in children 2 months to 7 years of age. Pneumococcus was isolated from 25%, Haemophilus influenzae from 23% and Moraxella from 15%. If you have been paying attention, you will have noticed that this adds up to more than 100%. Multiple bacterial or viral agents or concomitant bacterial and viral agents were found in many patients (New Engl J Med. 1999;340:260).

Although it appeared that the isolation of a viral agent with a bacterium tended to make these more resistant to antibacterial therapy, this was not borne out by later studies (New Engl J Med. 1999;340:260).

In trying to define the role of viruses in otitis media changes in middle ear pressure were studied in a group of school children (Pediatrics. 2002;109:826). It was found that two-thirds had an abnormal tympanogram within two weeks following the onset of a URI at least on one day following the onset of illness most commonly during the first three days after onset.

Using PCR to test nasopharyngeal swabs the researchers found evidence of viral infection in almost two-thirds of the patients. Unfortunately, there is no report of otoscopic examination of the ears in this study. It would have been useful to be able to confirm the statement of the authors of the study that “retraction of the tympanic membrane attributable to negative middle ear pressure may produce redness of the eardrum, in the absence of microorganisms (presumably bacterial organisms).” It would have been wonderful to have photos of a red tympanic membrane that was proven to be caused by a virus and not by bacteria. None of the patients in this study developed bacterial otitis media.

Role of current approach

These data support the long held thesis that preceding viral infections, by a variety of mechanism, may result in negative pressure in the middle ear granting pharyngeal bacteria access. By the time we treat bacterial otitis, there is a collection of pus and bacterial in a closed space, the middle ear. Most ID specialists will tell you this is a situation, which will not readily respond to antimicrobials. Thus, it is not surprising that the current approach is less than optimal. In addition, viruses themselves may be causing AOM. In one study, viruses were found in almost 40% of ears in which no bacteria were recovered (New Engl J Med. 1999;340:260). Thus, it may be difficult to demonstrate the effectiveness of antibacterial drugs because some of the cases of AOM are due to viruses which would not be expected to respond.

Does our current approach to the treatment of AOM intervene too late? Should we be trying to prevent the preceding viral infections or treat their consequences in ways, which would diminish the likelihood of bacterial AOM? Oseltamivir has been shown to decrease the risk of AOM diagnosed by tympanometry by 44% if the drug was started within 48 hours following the onset of influenza (Pediatr Infect Dis J. 2001;20:127). The use of oseltamivir probably prevented some inappropriate antibiotic use. Whether the office visits for AOM prevented would have been greater than the visits for treatment of influenza with oseltamavir is uncertain. Although influenza may be a significant cause of otitis during the flu season, it does not appear to be a major player in the overall scheme of things. Only 2.5% of total cases of AOM (New Engl J Med. 1999;340:260) and 4.5% of total URIs appeared to be due to influenza virus (Pediatrics. 2002;109:826).

New approaches

There is some evidence that prevention of influenza infections by immunization with live intranasal (New Engl J Med. 1998;338:1405) or killed influenza vaccine (Arch Ped Adol Med. 1995;149:1113) will decrease the risk of otitis during the flu season. However, more recent reports (see article on page 5) and others have failed to confirm this finding (J Infect Dis. 2000;182:1218). There are other compelling reasons to give this vaccine to children. Developing vaccines against other viruses involved in AOM is an attractive option. An RSV vaccine is a major priority. Unfortunately there is, at most recent count, more than a hundred different serotypes or rhinoviruses so it is unlikely that we will have a rhinovirus vaccine. In addition to coronaviruses and paramyxoviruses, there are undoubtedly other viruses, e.g. the recently described parapneumoviruses (Med J Aust. 2002;176:188) which may be involved in the etiology of otitis.

Another alternative would be to try to use agents that interfere with the adverse effect of these viruses. Diphenydramine, a drying agent, sudaphedrine, a sympathomimetic, and steroids, an anti-inflammatory agent, all have been ineffective when given as an adjuvant to antimicrobials in treatment of otitis. Agents that reduce inflammation during viral infections, which precede AOM, might be worth investigating.

If rhinoviruses are a major cause of otitis during the spring and fall would it not be preferable to evaluate antiviral rather than using antibacterial prophylaxis during these seasons in “frequent cryers.” There are at least three classes of drugs that are active against this group of viruses and one, pleconaril, already has been studied in children. Oseltamivir prophylaxis might be worthwhile evaluating in these children during the influenza season. There is still no antiviral drug in sight of clinical trials for RSV.

There now is extensive evidence to implicate viruses as etiologic agents in otitis. They explain why some cases of otitis will get better without antibacterial therapy, and why typanocentesis does not always yield a bacterial agent. If a significant proportion of otitis is caused by viruses and these are not affected by antibacterial drugs, it is no wonder that it is hard to show a more substantial effect of these drugs or of pneumococcal vaccine on AOM. It should give us added reason to withhold antibacterial drugs. Finally, a combination of approaches including immunization against bacterial and viral agents, the use of antiviral and of anti-inflammatory agents might be an alternative approach to the management of AOM.

For more information:

• Winther B, Hayden FG, Arruda E, et al. Viral respiratory infection in schoolchildren: effects on middle ear pressure. Pediatrics. 2002;109(5):826-832.
• Nissen MD, Siebert DJ, Mackay IM, et al. Evidence of human metapneumovirus in Australian children. Med J Aust. 2002;176(4):188.
• Whitley RJ, Hayden FG, Reisinger KS, et al. Oral oseltamivir treatment of influenza in children. Pediatr Infect Dis J. 2001;20(2):127133.
• Hurwitz ES, Haber M, Chang A, et al. Studies of the 1996-1997 inactivated influenza vaccine among children attending day care: immunologic response, protection against infection, and clinical effectiveness. J Infect Dis. 2000;182(4):1218-1221.
• Heikkinen T, Thint M, Chonmaitree T. Prevalence of various respiratory viruses in the middle ear during acute otitis media. New Engl J Med. 1999;340(4):312-4.
• Belshe RB, Mendelman PM, Treanor J, et al. The efficacy of live, attenuated, cold-adapted, trivalent, intranasal influenzavirus vaccine in children. New Engl J Med. 1998;338(20): 1405-1412.