Potential complications associated with steroid use in the middle and inner ear.

Glucocorticoids have been used to treat disorders such as sudden sensorineural hearing loss, autoimmune hearing loss, Meniere's disease, and hearing loss secondary to trauma, viral insult, and idiopathic causes. When used to treat disorders of the inner ear, steroids have traditionally been administered systemically. However, it has been postulated that the steroid levels achieved in the inner ear with systemic administration are not consistent because of the presence of the blood-labyrinth barrier. (1) This can lead to potentially suboptimal therapeutic concentrations and subsequent treatment failure.

Another concern with the systemic administration of steroids is the side effects associated with this route. Possible complications include fluid and electrolyte imbalance, muscle weakness and a loss of muscle mass, tendon rupture, osteoporosis, vertebral fractures, aseptic necrosis of the humeral and femoral heads, peptic ulcer, impaired wound healing, a cushingoid state, adrenocortical insufficiency, cataracts, and decreased carbohydrate tolerance.

In view of these concerns with systemic administration, other delivery routes have been investigated. Among the alternatives are intratympanic injection and topical administration into the middle ear cavity via a tympanostomy tube. But these routes have their own possible complications, including delayed healing of the tympanic membrane, vertigo, heating loss, and morphologic changes in the round window membrane. (2-5) In this article, the author reviews the advantages and disadvantages of topical steroid use in the middle and inner ear.

Literature search

The author searched the MEDLINE database for evidence-based articles that would help answer the clinical question, Are there potential risks associated with the use of steroids administered directly to the middle and inner ear spaces? The search terms used were "glucocorticoids and inner ear toxicity," "glucocorticoids and middle ear toxicity," "ototoxicity and ototopical steroids," "intratympanic steroids-middle ear toxicity," and "intratympanic steroids-inner ear toxicity."

The articles that were retrieved were reviewed in detail to ascertain whether they answered the question. Most of those that did were reports of animal studies, although a few small series in humans were applicable, as well.

Extrapolating animal data to humans

Can data obtained from animal studies be extrapolated to humans? For practical reasons, the bulk of information available on the potential toxicity of steroid administration to the middle and inner ears via transtympanic administration has been derived from animal studies; sampling of inner ear fluids in humans poses a risk of permanent hearing loss.

There are several anatomic differences between animal models and human models. For example, guinea pigs have a patent cochlear aqueduct, and this can lead to CSF contamination during perilymph sampling. Also, the human round window membrane is thicker than that of lower-order animals (70 [micro]m for humans vs. 10 to 14 [micro]m for rodents), and this difference may affect drug transport into the middle ear. Finally, because the human inner ear is considerably larger than that of animals used in most studies, it is difficult to make generalizations about drug concentrations within inner ear fluids. (6)

Roland et al identified several anatomic as well as nonanatomic differences between animals and humans and described some of the ramifications of these differences. (7) For example, the human round window membrane is recessed in a triangular fossa and therefore is less accessible than the round window membrane in animals. As a result, substances placed in the middle ear of humans may remain in contact with the round window for a shorter period of time, thus potentially lessening their effect at this site. In an investigation of inner ear perfusion, Silverstein et al reported that nearly 30% of the human ears in their study had obstructing mucosal folds that obscured the round window membrane. (8)

Differences between humans and animals in eustachian tube function may exist as well, although much investigation in this area remains to be done. Again, if the anatomy of the human eustachian tube promotes the rapid movement of substances from the middle ear, potential contact time with the round window membrane would be reduced.

With regard to the nonanatomic differences, histologic changes in the round window have been seen in animals with otitis media. Specifically, these changes include increased vascularity, infiltration of immunoreactive cells and, eventually, thickening of the membrane, which may result in decreased permeability of the membrane over time.

Pending the accumulation of more data, we can only conclude for now that the extent to which animal data can be extrapolated to represent a human model is unknown.

Physiology: How steroids exert their effect

Corticosteroids exert their effect by strongly suppressing the immune system-mediated inflammatory response. This suppression occurs as a result of the interaction between the drug and the intracellular glucocorticoid receptors, which have been shown to exist in both cochlear and vestibular tissues of humans and rats. With respect to the location of steroid receptors, Rarey and Curtis found that the highest concentrations of receptors were in the cochlear labyrinth, specifically the spiral ligament, followed by the organ of Corti; the lowest concentration was in the stria vascularis. (9) In the vestibular labyrinth, the highest concentrations of receptors were in the crista ampullaris and utricular macula, and the lowest levels were in the macula of the saccule.

Before a glucocorticoid (or any other substance) can reach the inner ear and interact with the receptors there, it must pass through the round window. The transport of substances through this membrane is affected by its thickness, by the size and charge of various particles, by the presence of inflammation or injury to the membrane, and by the presence of facilitating agents.

Once the glucocorticoid enters the inner ear, we do not know much about how it travels in the inner ear fluid, the amount of time it spends in the inner ear, and its eventual anatomic distribution. Nor do we know if there are any differences between the various glucocorticoids with respect to distribution, concentration, and efficacy in the middle ear.

Pharmacokinetics in the middle and inner ear

The pharmacokinetic properties of the glucocorticoids must be taken into consideration because they can have an impact on the safety, efficacy, and feasibility of steroid delivery to the inner ear.

In 1999, Parnes et al established cochlear fluid pharmacokinetic profiles for hydrocortisone, methylprednisolone, and dexamethasone in the guinea pig following oral, intravenous, and intratympanic administration. (10) All three of these steroids successfully penetrated the blood-labyrinthine barrier and round window membrane and entered the inner ear fluid. The authors compared the concentrations of these drugs administered via these routes with simultaneous pharmacokinetic profiles of drug in blood and CSF. They found that administration via the intratympanic route resulted in significantly greater inner ear levels of all three drugs at every sampling time when compared with administration via the systemic route. Perilymph concentrations of all three steroids peaked within the first hour, then rapidly declined. Similarly, peak endolymph levels were reached in 1 to 2 hours and they, too, rapidly declined. Of the three steroids, methylprednisolone had the highest concentrations in both perilymph and endolymph, and levels remained measurable for the longest period of time.

In the same report, Parnes et al then described how they used these pharmacokinetic data in a clinical study of 37 patients with sensorineural hearing loss secondary to various inner ear disorders. (10) Of this group, 20 patients were treated with intratympanic dexamethasone and 17 with intratympanic methylprednisolone. The best overall responses were seen in 2 patients with Cogan's syndrome and 1 patient with clinical autoimmune inner ear disease. No case of treatment-induced hearing loss was seen, as auditory thresholds remained stable even in patients who did not respond to therapy. Some patients reported temporary vertigo at the time of injection, but no persistent dysfunction was seen. No permanent tympanic membrane perforations were found.

In 2000, Chandrasekhar et al reported the findings of a guinea pig study in which they compared perilymph concentrations of dexamethasone following systemic and intratympanic administration. (11) Within 1 hour of administration, levels of the intratympanically delivered dexamethasone were significantly higher than levels of the intravenously administered drug. The difference was believed to be attributable in part to the low molecular weight and the high lipid solubility of dexamethasone, which would enhance its transport across the round window membrane. Intratympanic delivery did not result in any significant systemic absorption. Potential ototoxicity was not investigated in this study.

One year later, Chandrasekhar published the results of a clinical study of 10 patients with sudden sensorineural hearing loss who were treated with intratympanic dexamethasone. (12) These findings correlated with the findings of his group's earlier study (11) of guinea pigs--that is, perilymph concentrations of steroid were significantly higher with intratympanic administration than with intravenous delivery. No case of otitis media, persistent tympanic perforation, or other complication of intratympanic injection was seen.

In an attempt to further define the inner ear pharmacokinetics of intratympanically injected dexamethasone, Hargunani et al explored the conversion of the pro-drug dexamethasone 21-phosphate to its active form in the inner ear. (13) Detection of either the prodrug or its active form revealed that dexamethasone was present in the inner ear at 15 minutes and completely gone after 24 hours. The active form was detected as early as the prodrug, suggesting that the conversion occurred rapidly. This study provided confirmation that dexamethasone given by intratympanic injection delivers active drug to the inner ear structures, but it also showed that the therapeutic window is probably open for less than 24 hours after administration.

Through a series of simulated application scenarios, Plontke and Salt demonstrated that controlling middle ear drug clearance is of critical importance when administering the drugs to the inner ear. (14) Their simulations revealed that the longer a drug is applied to the round window membrane, the higher the drug level is within the inner ear. Also, drug levels in the inner ear are significantly higher when the rate of middle ear clearance is low.

Essentially, clearance can be defined as the removal of a substance from the cochlear fluid. When a drug is applied to the round window, the interplay between diffusion and clearance determines its distribution in the inner ear. Substances that are quickly cleared from the cochlear fluid will rapidly reach a steady state in which diffusion and clearance are equal. As a result, a drug will never reach the apical portion of the cochlea in appreciable amounts. Substances with faster diffusion rates will travel farther along the cochlea. (15)

In summary, the highest intracochlear drug levels will be obtained with continuous delivery, and the lowest with a brief application.

Studies of potential complications

Results of the treatment of sudden sensorineural hearing loss with intratympanic glucocorticoids have been encouraging. Likewise, the success of the ototopical steroid treatment of tympanostomy tube otorrhea has been well described--and well received by the medical community. Even so, the concern for potential side effects associated with these therapies has been raised.

In a study in which 2% hydrocortisone was instilled through a tympanic membrane perforation into the round window niche of rats for 5 consecutive days, Spandow et al demonstrated that auditory brainstem responses were permanently impaired in the frequency region beyond 8 kHz. (3) Morphologic investigation detected no damage to the inner ear. The tympanic membrane perforations through which the steroid was instilled remained unhealed during the first 3 weeks. At 2 months, 4 of 7 animals had persistent perforations.

To look at the effects of antibiotics and steroids on the middle ear mucosa, Park and Yeo inoculated the right ears of 27 rats with Streptococcus pneumoniae; the left ears served as controls. (4) The animals were divided into three groups: 9 rats received intramuscular penicillin G, 9 rats received a combination of penicillin G and dexamethasone, and the remaining 9 rats were not treated. Three animals in each group were sacrificed on day 4 after challenge, 3 more were sacrificed on day 7, and the remaining 3 in each group on day 14. Examination of their tympanic membranes and middle ear mucosa demonstrated that the antibiotic/steroid combination was more effective than the antibiotic alone in reducing or preventing the persistence of mucosal changes. In fact, the inflammatory responses in the middle ear mucosa in the antibiotic-treated group were similar (although not as severe) to the responses in the infected controls. By day 14, all of the tympanic membrane perforations in the infected control group had healed. In the antibiotic-treated group, the tympanocentesis site was not found in any animal on day 7. Healing of the tympanocentesis site was slowest in the antibiotic/steroid group, but no animal exhibited any persistent tympanic membrane perforation by day 14.

The aforementioned study by Spandow et al revealed that hearing loss occurred in rats treated with hydrocortisone applied to the middle ear. (3) On the other hand, Takeuchi and Anniko found that dexamethasone reduced the severity of hearing loss caused by Pseudomonas aeruginosa exotoxin A. (16) Nordang et al conducted a controlled experiment to determine whether different glucocorticoids have the same effect on the round window membrane. (5) They instilled either hydrocortisone or dexamethasone into the middle ears of 20 rats. Light microscopy and transmission electron microscopy revealed an accumulation of inflammatory cells close to the round window membranes of those rats that had been exposed to hydrocortisone but not in the round window membranes of rats that had been exposed to dexamethasone. Furthermore, the hydrocortisone-exposed membranes were thicker than the dexamethasone-exposed and control membranes. The effect of hydrocortisone on the round window membrane epithelium seen by Nordang et al (5) corresponded with the findings of Spandow et al. (3)

Another potential complication associated with intratympanic steroid injection is the worsening of residual hearing. Thus far, studies have been done on sensorineural hearing loss secondary to Meniere's disease (in which spontaneous fluctuations in hearing are seen, making evaluation of changes in hearing extremely difficult), sudden sensorineural hearing loss, and autoimmune inner ear disease. Hearing loss associated with the latter two conditions is also difficult to evaluate because of the natural progression of the disease. In the series by Parnes et al, no patient exhibited worsening of hearing, and auditory thresholds remained stable in all unsuccessful treatment cases. (10) Similarly, in the study of humans by Chandrasekhar, no patient experienced a rapid reduction in hearing after treatment (although audiometric data were not provided in this report). (12)

Clearly, the is sue of a further reduction in hearing thresholds secondary to intratympanic steroid use has not been definitely settled by the studies done thus far. Therefore, we should give consideration to several other factors as potential etiologies for hearing loss. These possible factors include the natural progression of the disease, the action of carrier molecules in the steroid preparation, the presence of concomitant infection (bacterial or viral), the effect of trauma to the round window resulting in a fistula, and the possible side effects of an anesthetic agent.

Conclusion

In summary, it appears that steroids in the inner ear achieve peak concentration rapidly (<1 hr) and are eliminated quickly (<24 hr). Thus, when these agents are administered for a short time, high intracochlear drug levels will not be achieved. It would appear, then, that a short-term course of steroid would be an appropriate ototopical regimen for the treatment of tympanostomy tube otorrhea.

Finally, although some studies have shown that hydrocortisone may be associated with changes in round window membrane morphology, this does not seem to be the case with dexamethasone. As for the potential for tympanic membrane perforations associated with the use of steroids, it appears that myringotomy healing is transiently impaired with the use of dexamethasone.

References

(1.) Seidman MD, Van De Water TR. Pharmacologic manipulation of the labyrinth with novel and traditional agents delivered to the inner ear. Ear Nose Throat J 2003;82(4):276-80, 282-3,287-8 passim.

(2.) Doyle KJ, Bauch C, Battista R, et al. Intratympanic steroid treatment: A review. Otol Neurotol 2004;25(6): 1034-9.

(3.) Spandow O, Anniko M, Hellstrom S. Hydrocortisone applied into the round window niche causes electrophysiological dysfunction of the inner ear. ORL J Otorhinolaryngol Relat Spec 1989;51 (2): 94-102.

(4.) Park SN, Yeo SW. Effects of antibiotics and steroid on middle ear mucosa in rats with experimental acute otitis media. Acta Otolaryngol 2001;121(7):808-12.

(5.) Nordang L, Linder B, Anniko M. Morphologic changes in round window membrane after topical hydrocortisone and dexamethasone treatment. Otol Neurotol 2003;24(2):339-43.

(6.) Banerjee A, Parues LS. The biology of intratympanic drug administration and pharmacodynamics of round window drug absorption. Otolaryngol Clin North Am 2004;37(5): 1035-51.

(7.) Roland PS, Rybak L, Hannley M, et al.Animal ototoxicity of topical antibiotics and the relevance to clinical treatment of human subjects. Otolaryngol Head Neck Surg 2004;130(3 Suppl):S57-78.

(8.) Silverstein H, Rowan PT, Olds MJ, Rosenberg SI. Inner ear perfusion and the role of round window patency. Am J Oto1 1997; 18(5): 586-9.

(9.) Rarey KE, Curtis LM. Receptors for glucocorticoids in the human inner ear. Otolaryngol Head Neck Surg 1996;115(1):38-41.

(10.) Parnes LS, Sun AH, Freeman DJ. Corticosteroid pharmacokinetics in the inner ear fluids: An animal study followed by clinical application. Laryngoscope 1999; 109(7 Pt 2): 1-17.

(11.) Chandrasekhar SS, Rubinstein RY, Kwartler JA, et al. Dexamethasone pharmacokinetics in the inner ear: Comparison of route of administration and use of facilitating agents. Otolaryngol Head Neck Surg 2000;122(4):521-8.

(12.) Chandrasekhar SS. Intratympanic dexamethasone for sudden sensorineural hearing loss: Clinical and laboratory evaluation. Otol Neuroto1 2001 ;22(1): 18-23.

(13.) Hargunani CA, Kempton JB, DeGagne JM, Trune DR. Intratympanic injection of dexamethasone: Time course of inner ear distribution and conversion to its active form. Otol Neurotol 2006;27(4): 564-9.

(14.) Plontke SK, Salt AN. Simulation of application strategies for local drug delivery to the inner ear. ORL J Otorhinolaryngol Relat Spec 2006;68(6):386-92.

(15.) Salt AN, Plontke SK. Local inner-ear drug delivery and pharmacokinetics. Drug Discov Today 2005; 10(19): 1299-1306.

(16.) Takeuchi N, Anniko M. Dexamethasone modifies the effect of Pseudomonas aeruginosa exotoxin A on hearing. Acta Otolaryngol 2000;120(3):363-8.

Ann L. Edmunds, MD, PharmD