Summary of the findings of a joint consensus panel on the use of potentially ototoxic ototopicals

Ear, Nose & Throat Journal, Sept, 2004 by Peter S. Roland

In 2003, Dr. Roland was appointed chairman of the Consensus Panel on Role of Potentially Ototoxic Antibiotics for Topical Middle Ear Use by the president of the American Academy of Otolaryngology--Head and Neck Surgery (AAO--HNS), John S. Johnson, MD. Other members of the committee were Peter C. Weber, MD, and Fred Owens, MD, appointed by the leadership of the AAO--HNS; Rick Friedman, MD, PhD, and Spiros Manolidis, MD, appointed by the American Neurotology Society; Leonard Rybak, MD, PhD, and Greg Matz, MD, appointed by the American Otological Society; and Janet Casey, MD, appointed by the American Academy of Pediatrics. Maureen Hannley, PhD, served as a scientific advisor, and Michael G. Stewart, MD, MPH, served as an advisor in evidence-based medicine. What follows is a brief summary of their findings.

The number of claims lodged against physicians for allegedly causing iatrogenic ototoxic injury appears to be increasing. While British and Canadian medical organizations had developed recommendations for limiting the use of potentially ototoxic ototopical antibiotics, no systematic, evidence-based review of this issue had been performed in the United States until our consensus panel was formed and was charged with this responsibility. Our findings were published in March 2004 (see "The panel's recommendations"). (1) Our task was to answer two questions:

* Should potentially ototoxic antibiotics be used to treat ear disease in patients who have an open tympanic or mastoid cavity?

* If potentially ototoxic antibiotics are prescribed, should limits or conditions be placed on their use?

Methods

The panel conducted separate reviews of four areas of research: animal studies, (2) ototoxicity in humans, (3) therapeutic efficacy, (4) and antibiotic resistance. (5)

Animal studies. The data on animal ototoxicity are consistent. It is well established that aminoglycosides and some other agents are reliably and severely ototoxic in animals. But even though animal data are important, extrapolation to humans must be done with caution because of the differences in the round window membranes of animals and humans.

Ototoxicity in humans. Some evidence exists of both cochlear and vestibular toxicity in humans. The evidence for cochlear toxicity is sparse, but researchers at the University of Toronto have reported that aminoglycoside drops can cause significant vestibular injury.

Therapeutic efficacy. There is no evidence that aminoglycoside eardrops are more effective than fluoroquinolone eardrops. Therefore, there is no justification for the use of a potentially ototoxic agent over a nontoxic agent in terms of efficacy.

Antibiotic resistance. We found no evidence that the use of ototopical agents induces or selects for resistant organisms.

Conclusions

In addition to its four formal recommendations, the panel arrived at several other conclusions of note:

* For both clinical and medicolegal reasons, physicians should not use an aminoglycoside-containing drop without very good cause. These agents are associated with the onset of hearing loss and vestibular injury.

* Neomycin-containing drops can cause topical sensitization that can be difficult to distinguish from persistent infection.

* The addition of a steroid to an antibiotic drop can reduce the formation of granulation tissue, speed the resolution of pain, and hasten the eradication of infection in some cases.

The panel's recommendations

Pursuant to its review of evidence-based research, the panel made four recommendations. (1)

Recommendation 1. When possible, topical antibiotic preparations that are free of potential ototoxicity are preferable to those that do have the potential for otologic injury in patients with an open middle ear or mastoid.

Keep in mind, however, that this recommendation is not intended to unjustifiably discourage the use of potentially ototoxic agents when clinical circumstances warrant and when the judgment of the treating physician is that their use is appropriate in a particular case. Such circumstances might include, but are certainly not limited to, previous treatment failure with quinolone drops, culture results that suggest that a quinolone drop would not be appropriate, and the unavailability of potentially nontoxic preparations.

Recommendation 2. When a potentially ototoxic antibiotic is chosen, it should be used only in infected ears and it should be discontinued shortly after the infection has resolved.

Recommendation 3. When a potentially ototoxic antibiotic drop is prescribed for a patient with an open middle ear or mastoid, the patient or parent should be warned of the risk of ototoxicity.

The patient or parent should be specifically instructed to call the physician or return to the office if the patient develops (1) dizziness or vertigo, (2) hearing loss or a worsening of heating if such an impairment was already present, or (3) tinnitus. The treating physician should consider the possibility of ototoxic injury when evaluating these new or exacerbated symptoms.