Printer Friendly

Aminoglycoside-containing eardrops: avoiding ototoxicity and choosing safer alternatives.

I have no problem using aminoglycoside-containing topical antibiotic drops in the presence of an intact tympanic membrane. However, a major problem could arise when a patient has a defect in a tympanic membrane--for example, a ventilation tube or a perforation--that might allow these drops to be absorbed into the inner ear via the round window membrane. Under these circumstances, I prefer to use a topical quinolone, which in my practice has been every bit as effective.

If there is no alternative but to use an aminoglycoside, then I do so carefully; I prescribe it for the shortest duration necessary and certainly for no longer than 5 or 7 days without reviewing the patient clinically to see if further treatment is required. Implicit is the fact that patients on topical aminoglycoside drops must be warned about their adverse effects and be instructed to stop the drops immediately if any dizziness, imbalance, hearing loss, or tinnitus occurs during therapy. Our experience with inadvertent topical gentamicin vestibulotoxicity (which is described in detail below) revealed that the earliest toxicity occurred on day 7 of treatment. In our study of intentional gentamicin toxicity in patients with incapacitating unilateral Meniere's disease, the earliest we recognized a deafferentation was approximately day 9. The longer these drops are used, the more likely toxicity will occur.

In general, ototoxic drugs should not be used in a patient whose middle ear is dry. They should be used only in infected ears. One exception might be for a patient with a perforated tympanic membrane who allowed some water into the ears by accident; even then, they should be used for only a short duration.

Clinical studies of topical gentamicin ototoxicity

At the University of Toronto, we discovered that the topical aminoglycoside gentamicin was ototoxic to the vestibular portion of the ear back in the 1980s. At that time, we were treating patients who had incapacitating unilateral Meniere's disease with high-concentration intratympanic gentamicin. We would inject approximately 24 mg/ml of gentamicin three times a day for 3 days through a piece of butterfly catheter tubing that had been inserted into the eardrum. This regimen was very successful in controlling vertigo in these patients. We often found, however, that approximately 2 days alter the completion of treatment, patients would suddenly become vertiginous.

Delayed ototoxicity. In an effort to explain this delayed reaction, we surmised that the aminoglycoside would have had to pool in the area around the round window niche before it was absorbed through the round window membrane. It would then travel through the cochlea and eventually into the labyrinth, where it did its damage. Additionally, many of these patients would experience a high-frequency hearing loss, which was not surprising considering the proximity of the basal turn of the cochlea to the round window membrane.

To demonstrate topical toxicity, we performed caloric testing on 24 patients before and after they had undergone gentamicin ablation. We believed that this would be an objective method of determining what occurred in the inner ear as measured by changes in the caloric excitability difference. At study's end, we found that 18 of these patients (75%) experienced a statistically significant deafferentation of vestibular function (figure). (1)


In the meantime, we saw some patients in consultation who had discharging ears that bad been treated with a low-concentration topical antibiotic/steroid combination--specifically, 3 mg/ml of gentamicin and 1 mg/ml of betamethasone. Three of these patients had complaints of ataxia, imbalance, and oscillopsia. Their history revealed that all three had taken their gentamicin/betamethasone drops for 2 to 3 weeks, even after their ears had become dry. This prolonged exposure to the drug led to bilateral vestibulotoxicity. Because we did not have any pretreatment audiometric or vestibular data on these early patients, we could not prove what we believed. Nevertheless, we were certain that a commercially available topical gentamicin preparation--with or without a steroid--became ototoxic after an average of 2 weeks' use. Since then, we have identified incontrovertible evidence of topical gentamicin toxicity in approximately 30 more patients.

"Therapeutic ototoxicity." In light of our findings with inadvertent gentamicin ototoxicity, we decided to study the feasibility of inducing intentional vestibulotoxicity wit the same low-concentration gentamicin/betamethasone preparation in patients with incapacitating Meniere's disease. In an ongoing study, we have thus far treated 37 patient We insert a ventilation tube into the ear and tell patients to instill 3 drops four times a day until they begin to experience dizziness, and then to continue the drops for 2 more days. After the 2 days, they were instructed to stop taking the medication. Electronystagmography and audiometry were performed before and after each patient's treatment course.

We found that most patients experience ototoxicity 12 days into their treatment regimen. The onset of vertigo in these patient occurred more rapidly than it did in the patients with discharging ears who experienced inadvertent ototoxicity as they had become dizzy after an average of 14 or 15 days of treatment. The reason for the 2- or 3-day difference was that the otorrhea in the latter group had prevented the drops from reaching the middle ear space right away. Once the otorrhea subsided, the drops were in a situation where they could be absorbed into the inner ear.

Of the 37 patients, approximately 75% have exhibited a significant change in caloric excitability difference. Also, approximately 50% of patients experienced a worsening of their hearing, which indicates that patients with Meniere's disease might be more sensitive to the effects of gentamicin. Approximately 40% of our patients developed permanent perforations in their eardrums once their tubes had been extruded, which might possibly be explained as a result of the steroid component. Last year, we published an article reporting the results of intentional ablation in 20 of these patients. (2)

Genetic predisposition

Finally, in what is something of a disconcerting development, we have learned that some Asian and Middle Eastern patient groups appear to be exquisitely sensitive to the aminoglycoside kanamycin because they have a mitochondrial RNA mutation. People with this genetic mutation are at risk of completely losing their hearing and possibly their vestibular function after one or two doses. As people continue to move around the world in greater numbers, it's possible that you might see such a patient in your office.


Dr. Roland: Does anyone else feel comfortable using potentially ototoxic antibiotic drops in ears that have an open tympanic membrane? A consensus panel of the American Academy of Otolaryngology--Head and Neck Surgery (AAO--HNS) is attempting to address this issue, but it hasn't been resolved.

Dr. Schapowal: As far as bacterial infections are concerned, I do not see any reason to use an ototoxic topical antibiotic as a first-line treatment. Not only do these agents pose safety problems, but they also are simply less effective than quinolones. The only argument one could make for ototoxic agents is that they are inexpensive, and this would be a valid consideration in underdeveloped countries.

In patients with a perforated tympanic membrane, ototoxic effects can occur after only a single dose, regardless of how small the perforation is. So if you do use an ototoxic drop in an open eardrum, you should inform the patient of the risk and you should perform audiometry before and after treatment. If patients begin to notice a loss of hearing while on therapy, they should report it immediately and should undergo audiometry as soon as possible. If a patient cannot undergo audiometry within 24 hours, he or she should stop taking the medication until a visit can be scheduled.

Before ciprofloxacin/hydrocortisone otic solution became available, I would use ciprofloxacin eyedrops to treat middle ear infections in patients who had a perforated tympanic membrane, as well as in patients who had external otitis. Ciprofloxacin/hydrocortisone drops seldom fail, and they are especially effective against problematic bacteria such as Pseudomonas ae ruginosa. Therefore, I use ototoxic eardrops only as a second-line therapy and only in patients who have an intact tympanic membrane.

Also, antimycotic topical treatment is required for the 4% of cases of external otitis that are caused by fungi. In these patients, I have achieved good results with natamycin.

Dr. Haynes: I see no reason to use an aminoglycoside-containing drop when we have a safe alternative. I don't have an alternative to gentian violet or acetic acid, and I occasionally have to use them. But I won't use an aminoglycoside-containing eardrop, because the quinolone drops are available.

Prof. Hawke: One thing you do not want to do is continue to put aminoglycoside drops into a normal middle ear. If you do, there is a strong likelihood that ototoxicity will develop sometime after 7 days.

Dr. Roland: When you prescribe any drug for otorrhea, do you instruct the patient to use it for a specific number of days, or until the drainage stops, or for a specified number of days after the resolution of drainage? Many of my partners instruct patients to take their drops until the drainage stops and then for 1 additional day.

Dr. Dohar: We tell all otorrhea patients, regardless of which drug they are taking, to take their drops for 7 days and then return for a follow-up visit. We don't rely on patients to subjectively assess otorrhea, because their evaluations don't correlate well with otoscopic documentation. Also, with ofloxacin drops, we have found that 3 or 5 days of treatment is inadequate; in fact, it's no better than placebo. A minimum of 7 days is necessary. I don't know if the duration would be different with another antibiotic or with the addition of a steroid.

Dr. Rutka: I would stop the drops shortly after the drainage resolves. I would not continue them for any longer than 48 hours.

Dr. Roland: Does anyone support the use of ototoxic antibiotics just because they're inexpensive?

Dr. Rutka: In the ideal world, of course, cost would not be an issue, and safety and efficacy would reign supreme. Nevertheless, if you are considering treating a patient with a potentially ototoxic drug for longer than 1 week, 1 would strongly suggest that you reconsider and choose an alternative. Remember that if a patient does develop ototoxicity, the money you saved by prescribing a less-expensive ototoxic drug will be outweighed by the adverse effects on the patient and the consequences of possible medicolegal action.

Dr. Manning: It's not that I support their use as much as I have no choice but to use them. In most urban centers in the United States, approximately 50% of the patients in a children's hospital are on Medicaid, and Medicaid doesn't cover the more expensive, patented drugs if there is a generic available. Therefore, the only available options might be neomycin drops--which we never use because of their ototoxicity and low pH, and because they sting the skin--and a generic antibiotic such as gentamicin. When forced to use an ototoxic drop, we tell the patient not to take it for more than 5 to 7 days and not to take it once the otorrhea resolves. It's quite a dilemma when the only treatment available is one that we might be sued for prescribing.

Dr. Dohar: We have a similar situation at the Children' s Hospital of Pittsburgh in that we see perhaps 95% of the indigent population in our area. We have asked third-party payers to put a quinolone on their formulary, but they will not do it. So for these patients, we have no choice but to use an ototoxic drug.

Dr. Roland: Like Dr. Schapowal, we at the University of Texas Southwestern have been able to bypass formulary restrictions by prescribing an ophthalmic quinolone, and these eyedrops are on most formularies. But that's just skirting the real issue. Our profession must continue to emphasize to managed care organizations that the use of ototoxic drugs is not good medicine.

Dr. Haynes: Would a position statement from the AAO--HNS improve our access to nonototoxic eardrops?

Dr. Roland: Such a statement would put pressure on insurance companies, but the extent to which they would fight or succumb to that pressure would depend on internal factors beyond just medical safety. Insurance companies don't have to do anything. They can simply say, "Sorry, we don't agree with you." But the fact remains that we face a dilemma in that our patients must choose between generics, which are less expensive but more dangerous, and quinolones, which are more expensive but safer.

Dr. Poole: Our responsibility is not to make economic decisions for our patients. Our responsibility is to inform them of the risks and benefits of the different therapeutic options. If they want to spend $8 rather than $40, that's their decision. All we can do is give them the facts.

Dr. Roland: Which brings us to informed consent. A failure to obtain informed consent is frequently the pivotal point in malpractice lawsuits. What do you tell your patients when you prescribe an ototoxic drug because you know they can't afford anything else?

Dr. Manning: Every treatment decision we make involves a risk and a benefit. That's the nature of medicine. When you're treating a patient for a draining ear and you know he doesn't have much money, the risk of doing nothing is greater than the risk of ototoxicity. But we must explain the risks as best we can and we must do whatever it takes to obtain informed consent.

Dr. Rutka: It would be good practice to inform patients of the risks of aminoglycoside-containing drops, but we should acknowledge that the likelihood of ototoxicity is extremely low during a limited treatment course. As Dr. Manning suggested, untreated disease also carries risks for sensorineural hearing loss as well as vestibular loss.

Dr. Dohar: We have a standard printed form that we ask patients to read and sign, and we keep it in their chart. The form says that there is a risk that the drug will cause permanent injury to either the hearing organ or the balance organ, although the risk is very small. We define small as less than 5%. We also put a limit on the duration of therapy, even though, as Dr. Schapowal and Dr. Rutka have pointed out, an ototoxic drug can cause hearing loss after only a single dose.

Dr. Haynes: How do you feel about the issue of informed consent for eardrops? I have seen articles in the Canadian literature recommending that patients should be informed of the potential risks of aminoglycoside eardrops. This certainly is new ground.

Dr. Roland: I don't believe the consensus panel will make any recommendation as to whether consent should be written or oral. All prescribers are already required be law to inform patients of the risks and complications of any treatment, although we haven't been very conscientious about doing this for oral medications. On the other hand, I know that some rheumatologists are very careful when they prescribe methotrexate or cyclophosphamide for autoimmune inner ear disease, and they obtain consent after a fairly detailed discussion.

Dr. Dohar: If the AAO--HNS does formally establish guidelines for informed consent, I believe the American Academy of Family Practice and the American Academy of Pediatrics will follow suit.

Dr. Roland: Should we warn patients to call or return to the office if certain symptoms arise? I am aware of two lawsuits that have been filed because a patient complained of vertigo while taking drops, but the otolaryngologist didn't discontinue therapy because he didn't make the connection.

Prof. Hawke: There are two similar lawsuits in Canada in which a patient complained of dizziness, but the otolaryngologist said that it could not be caused by the drops, and he told the patient to continue them. We have a significant physician education problem, especially when you consider that in the United States, most eardrops are prescribed by physicians other than otolaryngologists.

Dr. Rutka: Other symptoms of concern are imbalance, worsening hearing loss, and tinnitus. If they occur, the patient should stop the drops immediately and be reassessed at the earliest opportunity.

Dr. Roland: When you do prescribe an ototoxic antibiotic, how do you monitor the patient? Do you perform audiometry every week or two?

Dr. Coates: That would be ideal, but in the poorer areas of Australia, it's difficult to achieve. We are forced to prescribe ototoxic drugs in charity cases, but we can't know whether these patients are taking them correctly. Furthermore, patients out in the rural areas don't always return for scheduled follow-up visits.

Dr. Rutka: I'm not sure that we actually can monitor patients. We can warn them and advise them, but we can't really watch over them. It is impractical to perform electronystagmography on every patient during therapy. And it would be nice if we could obtain a pretreatment audiogram, or at least some audiometry to use for comparison purposes, but this, too, is very difficult. Our lack of pretreatment baseline and follow-up data is one of the reasons it is difficult to know whether it is the drops, the disease process, or some other factors that might account for any cochleovestibular loss.

Some places have built-in safeguards. In England, physicians cannot prescribe an aminoglycoside drop for longer than 7 days, and they cannot renew the prescription without seeing the patient again. For many years in Canada, we did not pay too much attention to ototoxicity. In all fairness, it doesn't occur very often, but when it does, it can be serious. About 5 years ago, our national health agency also put a 7-day limit on aminoglycoside drops. But last year, the agency basically banned the use of these drops in patients with perforated eardrums. Such use is now indefensible in a court of law.


(1.) Rutka J, Topical aminoglycosides? No. The case against using these agents in chronic ear disease. Ear Nose Throat J 2003: 82(Suppl 1):9-12.

(2.) Kaplan DM, Hehar SS, Bance ML, Rutka JA. Intentional ablation of vestibular function using commercially available topical gentamicin-betamethasone eardrops in patients with Meniere's disease: Further evidence for topical eardrop ototoxicity. Laryngoscope 2002; 112:689-95.
COPYRIGHT 2003 Medquest Communications, LLC
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rutka, John
Publication:Ear, Nose and Throat Journal
Geographic Code:1USA
Date:Aug 1, 2003
Previous Article:Value and safety of steroids in treating ear disease.
Next Article:Topical treatment of chronic suppurative otitis media in Aboriginal children.

Related Articles
Use of a new device, the [MicroWick.sup.TM], to deliver medication to the inner ear.
Characteristics of systemic and topical agents implicated in toxicity of the middle and inner ear. (Review).
Topical aminoglycosides? No. The case against using these agents in chronic ear disease. (Debate).
Topical aminoglycosides? Yes. The case for using these agents in chronic ear disease. (Debate).
Other topical agents are safer and just as effective. (Rebuttal).
Benefits of aminoglycoside drops still outweigh the risks. (Rebuttal).
Value and safety of steroids in treating ear disease.
The formation and management of middle ear granulation tissue in chronic ear disease.
Topical antibiotics: strategies for avoiding ototoxicity.
How serious a problem is topical aminoglycoside ototoxicity?

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters