Osteoporosis, Bisphosphonate Use, and Risk of Moderate or Worse Hearing Loss in Women

Sharon G. Curhan MD, ScM; Konstantina Stankovic MD, PhD; Christopher Halpin PhD; Molin Wang PhD; Roland D. Eavey MD, SM; Julie M. Paik MD, ScD; Gary C. Curhan MD, ScD


J Am Geriatr Soc. 2021;69(11):3103-3113. 

In This Article


In this large nationwide longitudinal study of nearly 144,000 women with up to 34 years of follow-up, we found that osteoporosis or LBD was independently associated with higher risk of incident moderate or worse hearing loss. The magnitude of the elevated risk was similar among women who did and did not use bisphosphonates. We also found that risk of hearing loss was higher among women with a history of VF, but not HF. To the best of our knowledge, this is the first large longitudinal study to evaluate the relations of bone density, bisphosphonate use, fractures and risk of hearing loss.

It is estimated that ~25% of U.S. women aged 65 and older have osteoporosis,[26] characterized by reduced bone density and qualitative changes in bone microarchitecture that lead to skeletal fragility and higher risk of fractures.[27] Osteoporosis results in an estimated 1.5 million fractures per year in the United States, most of which occur in postmenopausal women.[28]

Osteoporosis has been suggested as a potential risk factor for hearing loss, but the underlying mechanisms are unclear. Bone mass at peripheral sites is correlated with bone mass at central sites, such as hip and spine, with correlation coefficients between 0.6 and 0.7.[29] Plausibly, systemic bone demineralization could involve the temporal bone, the otic capsule, and the middle ear ossicles.[30–32] Evidence suggests pathways involved in bone homeostasis may influence cochlear sensorineural integrity; demineralization of the otic capsule and the cochlea have been associated with neuronal degeneration and sensorineural hearing loss.[33] Hearing loss has been demonstrated in other pathologic bone disorders, such as otosclerosis and Paget's disease, with pathologic changes in bone density and bone quality that overlap with those associated with osteoporosis.[34,35] In individuals with Paget's disease, cochlear demineralization was associated with both sensorineural and conductive hearing loss.[35] Among individuals with otosclerosis, high-resolution computer tomography demonstrated lower bone density at specific loci on the cochlear capsule.[36] Notably, in both Paget's disease and otosclerosis, abnormal bone remodeling in the otic capsule juxtaposed to the cochlea's lateral wall may lead to alterations in ion and fluid homeostasis in the perilymphatic space of the cochlea.[37] Potentially, imbalances in bone formation and resorption in osteoporosis may lead to alterations in ionic metabolism that result in hearing loss. Furthermore, a simulation study indicated that osteoporosis of the ossicles could increase high frequency displacement of the umbo and stapes footplate, suggesting ossicular osteoporosis could contribute to hearing loss.[38]

Most previous studies of bone density and hearing loss have been cross-sectional, relatively small, and findings have been inconsistent. While some studies found poorer hearing among individuals with lower bone density,[11] others have not.[10,39] A cross-sectional study among ~2000 participants (1082 women) in The Health, Aging and Body Composition Study found lower total hip BMD was associated with higher odds of hearing loss among black men (n = 310; odds ratio [OR]: 1.36 [1.05, 1.75]), but not white men (n = 660; OR: 0.86 [0.72, 1.03]); no associations between bone density and hearing loss in women were observed.[40] One cross-sectional study in the Korea Health and Nutrition Examination Survey found no association between femoral neck or lumbar spine bone density measurements and hearing loss,[39] but another found higher odds of hearing loss among those with lower femoral neck BMD and no association with femoral shaft or lumbar spine BMD.[41]

Increased bone resorption coupled with decreased bone formation causes reductions in bone mass quantity and bone quality leading to compromised bone strength and increased risk of fractures. As a complement to our analyses of osteoporosis, we also examined whether VF or HF, the most common osteoporosis-related fractures, was associated with risk of hearing loss. Our findings of up to a 40% higher risk among women with VF, but not HF, were intriguing and merit further study. Despite shared risk factors between these fracture types, substantial heterogeneity in risk factors exists.[7] The discordant findings between these skeletal sites may reflect differences in composition and metabolism of bones in the spine and hip and could provide insight into the pathophysiological changes in the ear that may lead to hearing loss. The temporal bone differs from both the hip and vertebrae in that it is a flat bone with a layered structure with a cancellous bone layer sandwiched between two layers of dense cortical bone. Differential influences of mechanical, hormonal and other factors on osteoporosis-related changes at different bone sites have been demonstrated.[42,43]

Our study did not observe that systemic bisphosphonates influenced the risk of hearing loss among women with osteoporosis. Bisphosphonates are the primary therapy for managing osteoporosis or other skeletal conditions characterized by increased osteoclast-mediated bone resorption. Some research suggests a beneficial influence of bisphosphonates on slowing or stabilizing progression of hearing loss among patients with otosclerosis, but the findings were not conclusive. A double-blind clinical trial found a nonsignificant trend toward stabilization or improvement in air conduction thresholds at 1 and 4 kHz after treatment with etidronate.[14] A small study of patients with otosclerosis and progressive hearing loss before treatment was suggestive of stabilization of hearing thresholds and word recognition scores after zoledronate treatment, but there was no control group for comparison.[44] A promising study in noise-exposed mice found zoledronate reversed noise-induced damage to primary afferent auditory synapses and subsequent recovery of peripheral auditory function. Specifically, regeneration of inner hair cell synapses and recovery of Auditory Brainstem Response wave I amplitudes were observed.[13] Studies in animals and humans have shown survival of spiral ganglion neuronal cell bodies for months or years after noise-induced peripheral synaptic and neuronal injury, suggesting a therapeutic window for synaptic recovery.[45] Possibly, a potential influence of bisphosphonates on the relation of osteoporosis and hearing loss in humans may depend on the type, dose and timing of bisphosphonate administration. Additional studies that evaluate these factors could be informative.

In our cross-sectional analysis of osteoporosis/LBD and audiometric hearing thresholds, the mean thresholds among women with osteoporosis/LBD who used bisphosphonates were significantly worse than those among women without osteoporosis/LBD. We did not observe a significant association among women who did not use bisphosphates. Although the cross-sectional findings suggest poorer audiometric thresholds among those who used bisphosphonates, the longitudinal findings demonstrate no significant difference in the risk of hearing loss among those with osteoporosis who did and did not use bisphosphonates (the 95% CIs overlap). Possibly, the women who were treated with bisphosphonates had more substantially reduced bone density, indicating that more severe osteoporosis was associated with poorer audiometric thresholds. The influence of bisphosphonate use on the bony and membranous labyrinth as well as middle ear merits further investigation.

Study strengths include the large sample size, longitudinal design, and use of well-validated covariate information. Our findings among two large and distinct cohorts of women were consistent. We also complemented our findings with those from a subcohort of women with longitudinal audiometric assessments. Limitations of our study include hearing assessment based on self-report. Hearing decline is often subtle, thus there is imprecision in date of onset. Pure-tone audiometry is the standard approach for hearing evaluation, but does not capture many aspects of hearing experience in real-world environments.[46] Hearing assessments based on self-report have been found to be reasonably reliable.[19] We a priori chose to examine moderate or worse hearing loss to minimize potential misclassification. The sensitivity of a single question to detect moderate or severe hearing loss among women of similar age to our population was high (95% and 100%, respectively).[20] The potential low specificity of a single question could mean the magnitudes of the associations may be even larger. This is an observational study, thus residual confounding could have influenced the results; nevertheless, the analyses were carefully adjusted for potentially confounding variables, many of which were demonstrated to be well-reported in this cohort. The study population included predominantly white female healthcare professionals, which enhanced validity of the health information collected and reduced variability in educational achievement and socioeconomic status, but research in non-white women and additional populations is warranted.