Cystic fibrosis (CF) is a life-threatening autosomal recessive genetic disorder characterized by poor cellular ion regulation and water balance leading to poor mucous transport from the lungs and other organs. CF-related illnesses are often caused by bacterial lung infections, such as Pseudomonas aeruginosa, or methicillin-resistant Staphylococcus aureus (MRSA). These infections often require hospitalizations for intravenous treatment with aminoglycosides (typically tobramycin or amikacin) that can result in sensorineural hearing loss. Dosing with aminoglycosides often occurs in conjunction with other antibiotics, including vancomycin (a glycopeptide) or azithromycin (a macrolide), which may also have concomitant ototoxic effects. Those, however, are less well-established (Wood. J Antimicrob Chemother 1996;37:209-222http://jac.oxfordjournals.org/content/37/2/209). Life expectancy for CF patients has increased from the mid-20s to the late 30s over the last decade (Cystic Fibrosis Foundation Patient Registry Annual Data Report. 2013). Identifying patient characteristics that potentiate drug-induced ototoxicity will help reduce the risk of developing hearing loss, leading to a better and longer quality of life. Yet, the incidence of and the risk factors that can potentiate aminoglycoside-induced hearing loss in the CF population remain poorly defined.
Margaret Kenna and her colleagues at Boston Children's Hospital set out to describe the relationship between ototoxic medications and hearing loss in a large cohort of subjects with CF. They explored whether age, number of hospitalizations, and cumulative (recent) aminoglycoside exposure are associated with an increased risk of hearing loss. The data in this paper was obtained from 177 subjects (mean age=18yr, SD=±10 yr) referred for audiometry with a prevalence of (mostly bilateral) hearing loss of about 20 percent (>20 dB in the 0.25-8 kHz range). Only 61 subjects received in-hospital intravenous antibiotic treatments in the year prior to audiometric testing. Older subjects or those with a higher number of hospitalizations (and hence increased antibiotic dosing) in the year prior to audiometry had a significantly higher risk of hearing loss. This greater risk associated with an increasing number of hospitalizations echoes the findings of a study by Al-Malky and colleagues, where pediatric CF subjects who had more than 10 hospitalizations were at greater risk of high-frequency hearing loss (Al Malky. J Cyst Fibros 2015;14:248-254http://www.cysticfibrosisjournal.com/article/S1569-1993(14)00170-2/abstract). The risk associated with individual antibiotics, however, was not significantly associated with hearing loss, although amikacin did trend close to significance despite the small numbers in their sample. The authors cautioned their narrow discovery of antibiotic dosing as a limitation of their study, and imply that correlating lifetime dosing with audiometry may reveal which dosing paradigms (and/or their frequency) increase the risk of hearing loss. They had previously reported, as did a group in London, that increased dosing may be associated with increased risk of hearing loss in pediatric subjects (Al Malky. J Cyst Fibros 2015;14:248-254http://www.cysticfibrosisjournal.com/article/S1569-1993(14)00170-2/abstract; Cheng. Otolaryngol Head Neck Surg 2009;141:86-90http://www.sciencedirect.com/science/article/pii/S019459980900254X). Both groups recommended that all patients with cystic fibrosis receiving intravenous antibiotics, particularly aminoglycosides, are routinely screened for hearing loss.
Risk Factors for Hearing Loss in Patients with Cystic Fibrosis
Tarshish, Y, Huang, L, et al
J Am Acad Audiol
This study also examined a second class of antibiotics, macrolides (e.g., azithromycin), that have been infrequently reported to potentiate the risk of hearing loss when given in conjunction with aminoglycosides. Again, as the authors noted, dosing was restricted to the year prior to audiometry for this study. Examination of dosing history will become increasingly important as subjects with CF live longer and are treated with a variety of drugs known to (or suspected to be) ototoxic. This examination will become increasingly complex as dosing paradigms will often be concomitant or sequential between different classes of drugs over their lifetime. Another drug class that will need to be studied in the CF population is the glycopeptide antibiotics (e.g., vancomycin) that are preferred by some institutions, and yet have also been infrequently associated with ototoxicity. Other frequently administered drugs include loop diuretics, sulfonamides and polymyxins;their effect on hearing outcomes, alone or with other antibiotics, in the population with CF will also be of importance.
The authors described the prevalence of hearing loss in their self-selected sample as likely an over-estimate of that reflected in the CF population as a whole. They found that children and teenagers have hearing loss rates of 10-11 percent compared with just five percent in adolescents in national studies. Other data derived from national studies show hearing loss rates in children up to 9 years old are less than one percent. Thus, hearing loss in young subjects with CF in this study were up to 1 order of magnitude greater than in national databases. The authors stressed that high sound level exposure can also potentiate drug-induced ototoxicity, particularly in younger subjects listening to personal stereo devices that may contribute to this increased risk. Furthermore, older subjects with CF had a greater prevalence of hearing loss than the national average for each decade of life. There was also a possible association with gender, as males with CF were at greater risk of hearing loss in this study, and this almost-significant trend will need validation in a larger study. Differences in the reported prevalence of hearing loss rates are likely due to various methodologies, dose calculating methods, treatment regimens and other study related factors.
The authors also noted that Al-Malky and colleagues pointed to lower pulmonary function data (predicted FEV1%) as a significant risk factor for earlier onset of hearing loss. This is a crucial data point, as a low FEV1 percentage is also indicative of pulmonary infections and suggests that these subjects may be more likely to receive intravenous aminoglycoside treatments. Such infections trigger systemic inflammatory responses that were recently demonstrated in preclinical models to exacerbate the ototoxicity of aminoglycosides (Koo. Sci Transl Med 2015; 7:298ra118http://stm.sciencemag.org/content/7/298/298ra118). Another risk factor is a mitochondrial mutation that pre-disposes individuals for aminoglycoside-induced ototoxicity; however, this mutation has a very low prevalence in the CF population that is typically of northern European origin. Nonetheless, there may be unidentified genomic associations with drug-induced hearing loss, although specific polymorphisms have yet to be identified in the CF population.
As pharmacological treatments for cystic fibrosis continue to improve the lifespan of individuals with CF, it becomes more urgent to preserve their quality of life, and particularly their auditory function, so they may remain better integrated within their families and mainstream society. More importantly, new pharmacological candidates are being developed as otoprotectants against drug-induced hearing loss, typically in healthy preclinical models. It will be crucial that these new otoprotective drug candidates are effective during aminoglycoside treatment in the setting of cystic fibrosis, and particularly in individuals with infection and/or inflammation, and not only in healthy subjects.