Aminoglycoside-induced hearing deficits – a review of cochlear ototoxicity

Abstract

This article aims to inform the reader about current knowledge of the effects of aminoglycosides on hearing. The hearing loss caused by aminoglycosides is permanent and can negatively affect the individual’s quality of life. Although treatment with aminoglycosides cannot be avoided in some instances, timely intervention can minimise the impact of the hearing loss on the individual’s everyday life. The discussion includes the incidence of aminoglycoside-induced hearing loss and predisposing factors. In addition, monitoring and management of cochlear ototoxicity is described.

Keywords:

• amikacin
• aminoglycosides
• cochleotoxicity
• hearing loss
• kanamycin
• ototoxicity

Introduction

Aminoglycosides are often used as part of treatment of life threatening illnesses such as septicaemia and multidrug-resistant tuberculosis (MDR-TB). Research has firmly established that aminoglycosides cause permanent hearing loss in humans.¹ In a country like South Africa, where drug-resistant tuberculosis prevalence and incidence is among the highest in the world,² health professionals’ knowledge about the impact of a permanent hearing loss due to aminoglycosides is central to appropriate management of affected individuals.

With respect to the inner ear, aminoglycosides may injure either the cochlea or the vestibular system.^3,4 The aminoglycosides enter the inner ear fluids via the bloodstream and result in intracellular biochemical and morphological changes of the cochlear outer hair cells (OHCs).³ Aminoglycosides bind with iron, forming an oxidative compound that contributes to the formation of free radicals, which are involved in tissue damage in the body due to oxidative activities with proteins and other targets.⁵

Initial ototoxic effects of aminoglycosides occur in the OHCs and supporting cells at the basal end of the cochlea, where high frequencies are transduced.^3,6,7 With prolonged use of aminoglycosides damage progresses towards the apical portion of the cochlea responsible for the processing of low frequencies.^3,8 Cochlear damage due to aminoglycosides is predominantly irreversible since it has not been firmly established that the cochlear hair cells in humans are capable of regeneration.⁹

Incidence

Wide variability exists in the reported incidence of aminoglycoside cochleotoxicity. The incidence of cochlear damage due to aminoglycosides varies from 7–90%, as can be seen in Table 1. According to the American Speech-Language-Hearing Association (ASHA)¹⁰ the actual frequency of cochleotoxicity associated with aminoglycosides is unclear due to the inconsistent reporting of results. Incidence rates appear variable and controversial due to differences in study design and methodologies.¹¹ Studies have used different criteria to define cochleotoxicity as well as various means to monitor hearing.^4,10 The variability can also be ascribed to a lack of standardised guidelines for the monitoring of cochleotoxicity, as well as the lack of universal criteria for ototoxic changes. In addition, it is known that cochleotoxicity starts in the high frequencies, thus monitoring techniques examining these high frequencies are more sensitive to detecting cochlear damage than standard frequency methods (≤8 kHz).

Table 1: Incidence of cochleotoxicity

Author	Year	Drug	Method	Participants	Incidence	Comments
Al-Malky et al^24	2011	Amikacin, tobramycin	Prospective high frequency pure tone audiometry	39	21%	Treatment for ±14 days
Black et al^25	1976	Amikacin	Prospective, pure tone audiometry	44	24%	Significant loss was defined as a reduction of ≥ 15 dB; 84% of participants who developed hearing loss were asymptomatic
De Jager et al^26	2002	Amikacin, kanamycin, streptomycin	Pure tone air conduction	110	18%	Treatment for > 14 days. Hearing loss was defined as a loss of 15 dB at two or more frequencies, or a minimum of 20 dB hearing loss of at least one frequency between 0.25 and 8.0 kHz
Duggal et al^27	2007	Amikacin, kanamycin, capreomycin	Pure tone air and bone conduction	64	25%	Treatment for 18–24 months
Fausti et al^28	1993	Aminoglycoside not specified	High frequency audiometry	94 ears	69%	Treatment for > 4 days
Fee et al^29	1980	Gentamicin, tobramycin	Prospective audiometry	113 participants, 138 courses	Gentamicin 16.4%	Hearing loss progressed after discontinued drug
Harris et al^19	2012	Kanamycin, streptomycin, capreomycin	Prospective, pure tone audiometry	153 participants with MDR-TB	57%	Monitored monthly for 3 months
Hotz et al^30	1994	Amikacin	Transient evoked otoacoustic emissions	10 ears	90%	Treatment for more than 12 days
Lerner et al^21	1979	Gentamicin, amikacin	Prospective, pure tone audiometry	54 participants for each drug	Gentamicin 7%; amikacin 9%	Cochlear toxicity defined as 15 dB decrease at two frequencies
Moore et al^31	1984	Amikacin	Prospective, pure tone air conduction	135	22.3%
Ramma et al^18	2012	Kanamycin, amikacin	Prospective, pure tone air and bone conduction	53	47%	Treatment between 1–18 months
Sturdy et al^32	2011	Amikacin, capreomycin, streptomycin	Retrospective, pure tone audiometry	50	28%	Treatment for > 2 weeks

Different aminoglycosides vary in terms of their level of ototoxicity, with for example, kanamycin being more cochleotoxic than streptomycin.⁴ Other reasons for the variety of incidence rates could be the differences in dosage, courses of administration and individual susceptibility to aminoglycoside cochleotoxicity.¹²

Disability and quality of life

Hearing loss can affect an individual’s quality of life, due to the negative impact on one’s ability to communicate.¹³ This impaired communication can affect the person in all spheres of life, including socialising, employment, self-esteem and general well-being. In the USA, a survey found that 36% of individuals with hearing difficulty were unemployed in comparison to an unemployment rate of 25% for individuals with no disability.¹⁴ Another study found that, regardless of the aetiology and type of hearing loss, hearing-impaired individuals are likely to suffer from anxiety, lack of self-confidence, depression and/or social isolation.¹⁵ Similarly, Cohen et al.¹⁶ found that hearing impairment impacted negatively on social function, emotional state and communication. Therefore, even in the case of individuals where permanent hearing loss due to aminoglycoside exposure cannot be avoided, timely detection of cochlear damage is important to enhance quality of life with early intervention.

The impact of aminoglycoside-induced ototoxicity is pertinent in the local context, as the South African MDR-TB guidelines include aminoglycosides as part of the treatment regimen.¹⁷ These negative effects of permanent hearing loss can have dire consequences on economic and social well-being in a country like South Africa, which has one of the highest rates of MDR-TB in the world.² Local data suggest that the occurrence of hearing loss due to aminoglycoside treatment ranges from 24–57%,^18–20 which highlights the magnitude of ototoxicity in our context.

Factors influencing occurrence

Medication variables

Streptomycin, an aminoglycoside generally used to treat tuberculosis,²¹ has been found to be predominantly vestibulotoxic, and has a fairly minor effect on hearing.^4,22,23 Gentamicin has a similar effect on the auditory system as streptomycin.³³ To the contrary, kanamycin is primarily and profoundly cochleotoxic, as it destroys outer hair cells in the cochlea.^23,34 Kumana and Yuen³⁵ suggested that its toxicity is such that kanamycin should be regarded as obsolete. Amikacin is a derivative of kanamycin and structurally similar.³⁶ In South Africa streptomycin, kanamycin and amikacin form part of the drug regimen administered to MDR-TB patients.¹⁷ In addition, capreomycin, a polypeptide antibiotic, is also used, and has a similar effect on the auditory system as its aminoglycoside counterparts.

Variables such as drug dosage, once vs. multiple daily administration, and peak serum levels are poorly correlated with the symptoms of cochleotoxicity.^9,37−39

Patient variables

Risk factors for cochleotoxicity include previous treatment with an ototoxic agent, renal impairment, pre-existing hearing loss and vestibular dysfunction. For a comprehensive list, see Table 2. It is important that the medical practitioner questions the patient about these risk factors. However, one cannot solely rely on patient report, as an individual will only notice a hearing loss when significant cochlear damage has already occurred.²⁸

Table 2: Intrinsic and extrinsic risk factors for cochleotoxicity

Intrinsic risk factors	Extrinsic risk factors
Previous treatment with ototoxic medication	Concomitant treatment with other ototoxic or nephrotoxic drugs
Prolonged exposure	Dosage in relation to weight
Renal impairment	Health practitioners’ knowledge of cochleotoxicity
Genetic susceptibility	Attitude of staff when symptoms are reported
Poor nutrition	Availability of diagnostic/monitoring services
Advanced age
Excessive noise exposure
Pre-existing hearing loss and vestibular dysfunction
HIV/AIDS

Healthcare practitioner variables

It is important that the healthcare practitioner has adequate knowledge about ototoxicity in order to manage and refer the patient appropriately. There is a need for the family general practitioner, nurses, specialists such as otolaryngologists or physicians involved in direct patient care, and the audiologist in the team to focus on the risk of ototoxicity in each individual patient. Clinical risk–benefit evaluations should be undertaken⁴⁰ to balance the need for intervention against the risk of permanent, irreversible hearing loss and the ramifications of living with such a disability. The use of available clinical guidelines should encourage clinicians to evaluate their practice with aminoglycosides.⁴¹ Although clinical guidelines can promote the use of evidence to shape practice at a clinical level, adoption of such guidelines can be inconsistent for reasons which include conflicting opinion and the complexity of the documents.⁴² With particular reference to ototoxicity, there has been a call for ototoxicity grading systems to be easy for clinicians to apply and understand and be standardised.⁴³

Knowledge can be disseminated through literature, clinical guidelines and professional society recommendations and protocols. Application of such knowledge in the clinical setting is, however, rather under-investigated. While not focused on ototoxicity specifically, a small study of ICU nurses in Brazil with an average of 12 years’ experience showed interesting findings. ICU nurse were asked to identify pairs of drugs which were likely to cause either drug interactions or adverse effects. Gaps in knowledge were identified with very few nurses being aware of the danger of combining gentamicin and furosemide, which would potentiate ototoxicity.⁴⁴

It could be proposed that to optimise the care of the patient undergoing treatment with ototoxic drugs, a team of healthcare practitioners could include the audiologist to give information regarding risk factors and pre-existing hearing loss and to inform, monitor and support the patient during and after treatment, the medical and nursing personnel involved with prescribing and administering the treatment, and the pharmacologist with knowledge of drug effects and interactions. All team members could give input into case management meetings to effectively guide appropriate intervention and management.

In addition to knowledge, the healthcare practitioner’s attitude can determine whether the patient seeks help when noticing symptoms of hearing loss. A study into attitudes of nurses in South Africa caring for patients with tuberculosis found that nurses could be authoritarian, scolding and frustrated.⁴⁵ These behaviours can possibly deter patients from reporting symptoms. However, more recent unpublished data (Nokhara and Rogers, personal communication) suggested that even if patients encountered negative attitudes from nursing staff overseeing their treatment, MDR-TB patients would not be deterred from having their hearing tested when such facilities were available. For variables related to the healthcare practitioner, see the extrinsic factors in Table 2.

Diagnosis/monitoring

Due to all the variables that influence an individual’s predisposition to developing an ototoxic hearing loss, the only way to ensure that ototoxic damage is not missed is to provide hearing testing for all individuals who are treated with aminoglycosides. Due to the negative impact of permanent hearing loss, it is advantageous to identify patients at risk for cochleotoxicity and establish an effective screening and monitoring programme that will either detect the onset of chronic cochleotoxicity early, i.e. before it affects the patient’s ability to communicate.⁴⁶⁻⁴⁹ Vasquez and Mattucci⁴⁹ recommended a flexible approach when monitoring patients for the effects of cochleotoxicity; and suggested that even if bedside testing is the only method available to monitor patients, this is preferable to no monitoring at all; a view supported by Dhanireddy et al.⁵⁰

The case history is a vital part of the diagnostic process. It would seem obvious to ask patients about the onset or aggravation of symptoms such as hearing loss, tinnitus, nausea or disequilibrium.⁵¹ During the case history-taking it is important to carefully scrutinise all previous drug exposure, as well as other predisposing or risk factors (see Table 2 for a list).

While the case history is crucial, in a multi-lingual setting such as South Africa, as in many developing countries, there may not be words in indigenous languages to describe symptoms of cochleotoxicity adequately. In addition, the grasp of these languages by staff may not be sufficient for subtle nuances to register their significance.

Even if patients report their symptoms, these complaints are often too late; i.e. when patients notice a hearing problem, permanent damage to the auditory system has already occurred.²⁸ Thus careful monitoring of hearing and cochlear status is important for purposes of hearing conservation during and after aminoglycoside treatment.¹⁰ The only way to ensure that ototoxic damage is not missed is to provide testing of cochlear function for all individuals who are treated with aminoglycosides.

In order to monitor cochleotoxicity, other possible causes of decreased auditory function, for example middle ear problems, need to be excluded.⁹ Thus the clinician must ensure that the hearing loss is sensorineural in nature. A test battery approach is essential, because sensitivity and specificity of the various audiometric tests are not well established.⁵¹

The available tests that can be included in the diagnostic and/or monitoring test battery include case history, conventional pure tone audiometry (cPTA), high frequency pure tone audiometry (hfPTA), otoacoustic emissions (OAEs), and the auditory brainstem response (ABR). Please see Table 3 for a summary of these tests’ functions and limitations.

Table 3: Battery of tests for diagnosing/monitoring cochleotoxicity

Test	Aim of evaluation	Result if cochleotoxic	Notes
Case history	Establish complaints of hearing loss, tinnitus, dysequilibrium, nausea	Positive for presence of symptoms	Subjective hearing loss noted only once there is marked hearing loss, so poor method. Tinnitus should always prompt audiological investigation
Conventional pure tone audiometry from 250–8 000 Hz (air and bone conduction)	Quantifies presence and severity of hearing loss across the middle frequency range of human hearing	Progressive, initially high frequency sensorineural hearing loss, usually symmetrical in both ears	Will not detect hearing loss until there has been significant high frequency hearing loss, so a comparatively ‘late’ finding
			Can be done at bedside but easily contaminated by ambient noise and patient attending to the task
High frequency pure tone audiometry (including frequencies up to 20 000 Hz)	Quantifies presence and severity of hearing loss across the high frequency range of human hearing	Progressive, ultra-high frequency sensorineural hearing loss, usually symmetrical in both ears, will progress to middle frequency range and then be detected with conventional pure tone audiometry	Patient acts as own control. Detects hearing loss sooner than cPTA
			Requires more specialised equipment
			Again subject to patient’s ability to attend to task
Otoacoustic emissions	Measures outer hair cell function	Abnormal, i.e. either reduced or absent	Requires only passive cooperation from patient, suitable for moribund patients and can be done at bedside
			Reasonably resistant to ambient noise. Suitable for screening and diagnosis
			Screening/monitoring can be performed by audiological assistant
Auditory brainstem response	Measures integrity of VIII^th cranial nerve. Also used for threshold estimation but an indirect measure	Abnormal	Requires only passive cooperation from patient, suitable for moribund patients and can be done at bedside
			However, sensitive to both ambient noise and electrical interference
			Requires skilled interpretation by audiologist

If compiling a test battery for cochleotoxicity monitoring, apart from the relative sensitivity and specificity of the tests, other factors such as the costs of procedures and equipment, the ease of use, applicability in the clinical environment, and the state of the patient should be considered. For example, individuals receiving ototoxic medication are frequently seriously ill; thus extensive testing might be too stressful.⁵² It is also important that the test(s) detect cochlear damage early, before the damage affects an individual’s speech perception.⁵³

When conducting pure tone testing (cPTA and hfPTA), the audiologist relies on the behavioural responses of the patient, which may not be accurate if the patient is too ill to participate.⁵² Although hfPTA for hearing loss due to ototoxicity detects damage earlier than cPTA,^23,28,54 the equipment for hfPTA is more expensive than standard audiometers and transducers, which is challenging in a country like South Africa where healthcare resources are limited. Thus a reliable objective measure that requires minimal participation of the individual being tested is preferred in a population where active participation cannot always be ensured.

The otoacoustic emission (OAE) test constitutes the only non-invasive means of objective cochlear investigation to date.⁵⁵ Evoked OAEs are sounds that originate due to activity of the outer hair cells (OHCs) in the cochlea in response to sound stimuli.⁵⁶ The presence of evoked OAEs is associated with normal, or near normal, mechanically active functioning of the OHCs,^57,58 and can only be measured reliably with normal middle ear functioning.⁸ As OHCs are the cochlear components most susceptible to injury from ototoxic drugs such as aminoglycosides,⁵⁵ evoked OAE tests seem to be appropriate tools in the monitoring of ototoxic effects on the cochlea.

One advantage of using OAEs to investigate cochlear functioning is that it is objective ,and thus does not rely on the testee’s cooperation or judgement. Another advantage is the fact that measurement of cochlear status is obtained in a non-invasive manner. It is also a relatively straightforward, quick (i.e. a few seconds per ear) and reliable test of cochlear function, especially with individuals who are ill, when compared to pure tone audiometry.⁵⁶ The cost of OAE equipment needed for cochleotoxicity monitoring is comparable to that required for conventional pure tone audiometry.

However, due to the fact that OAEs are absent with a moderate hearing loss, cochleotoxicity monitoring with OAEs can only be conducted with individuals who have, at the worst, a minimal hearing loss at baseline. Currently, there are no accepted protocols or criteria for ototoxic change using objective measures like OAEs.⁹

The auditory brainstem response (ABR) is an electrophysiological test that measures the integrity of the auditory system from the auditory nerve to the brainstem pathways. Due to the length of testing and general lack of frequency specificity; it is not a test that has been implemented in ototoxicity monitoring today. However, current research into high frequency tone-burst stimuli holds promise, but has not implemented clinically yet.⁵¹ In addition, this test needs a skilled audiologist to conduct and interpret the test.

Treatment

Spontaneous recovery

Due to the permanent nature of cochleotoxicity caused by aminoglycosides, it is highly unlikely that a person’s cochlear function and hearing will recover after treatment.⁵³ In fact, cochlear damage due to aminoglycoside treatment is known to progress/worsen up to 6 months after cessation of treatment due to the fact that aminoglycosides remain detectable in the hair cells up to 4 months after administration.⁵⁹

Pharmacology

Numerous studies on otoprotective agents such as d-methionine, antioxidants, free radical scavengers and cell death inhibitors show great promise.⁶⁰⁻⁶² The future use of otoprotectants may minimise or prevent ototoxic hearing loss.⁹ However, currently there is no commercially available medication that will prevent cochleotoxicity, or reverse its effects yet.

Cutting edge research is being conducted worldwide in an attempt to rebuild, replace or generate cochlear hair cells via gene and stem cell therapy.^63,64 Some success has been achieved in animals, but research in humans is not imminent.

Management of hearing loss

Since cochlear damage due to aminoglycoside treatment is permanent, it is key to minimise the patient’s chances of developing a debilitating hearing loss. Prevention is key and can be achieved by not prescribing ototoxic medication. However, this is often not possible, especially when a person’s life is at stake. Secondary prevention can take the form of advising the patient of the known risk factors, for example noise exposure. ¹⁹

Early detection of cochleotoxicity is crucial for timely management. This management includes counselling of the patient and family, communication strategies, audiological rehabilitation including amplification or cochlear implantation, if indicated. Being able to counsel the patient and family early on will enable them to adjust to this life-changing disability. It is important to not underestimate the depth of support that is often needed to accept a hearing loss, which necessitates intensive counselling. Thus, it is most beneficial for the patient with an ototoxic hearing loss if the management is done via a multidisciplinary team approach (see Figure 1 for a flowchart of healthcare professionals involved).

Figure 1: Idealised scheme of professionals and services required for the management of a patient with cochleotoxicity

Conclusion

Authors in developed countries have called for the use of aminoglycosides to be banned due to their adverse effects.⁶⁵ Due to the epidemic of MDR-TB in South Africa, it might not be possible to remove aminoglycosides from the treatment regimen in the near future. However, it is important that awareness be created about the devastating effect of aminoglycosides on an individual’s hearing, and thus their ability to communicate effectively. In doing so, appropriate measures can be put into place to manage aminoglycoside-induced cochleotoxicity effectively, in order to maximize the affected individual’s quality of life.

Additional information

Funding

Funding. This work was supported by University of Cape Town.