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

Disclosures

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

In This Article

Results

A total of 60,821 NHS and 83,078 NHS II women were included in the analyses (total n = 143,899). The baseline characteristics of participants in the NHS (1982) and NHS II (1995) according to osteoporosis/LBD are presented in Table 1. In both cohorts, women with osteoporosis or LBD were more likely to be older, menopausal, and to use hormone therapy, but there were no other large differences. Among women who used bisphosphonates, the mean (SD) duration of use was 5.8 (3.2) years in NHS and 3.4 (1.9) years in NHS II.

After 1,509,711 person-years of follow-up in NHS and 1,468,175 person-years of follow-up in NHS II, there were 12,460 (NHS) and 4565 (NHS II) incident cases of moderate or worse hearing loss reported. In multivariable-adjusted analyses, the risk of moderate or worse hearing loss was higher among women with osteoporosis or LBD in both cohorts (Table 2). In NHS, compared with women without osteoporosis, the multivariable-adjusted relative risk (MVRR, 95% CI) of moderate or worse hearing loss was 1.14 (1.09, 1.19) among women with osteoporosis; in NHS II, the MVRR was 1.30 (1.21, 1.40) among women with osteoporosis/LBD. The magnitude of the elevated risk was similar among women who did and did not use bisphosphonates (Table 3). In NHS, the MVRR among women with osteoporosis who did not use bisphosphonates was 1.13 (1.07, 1.20); the MVRR among those who did use bisphosphonates was 1.14(1.06, 1.22). In NHS II, the MVRR among women who did not use bisphosphonates was 1.28 (1.17, 1.40) and among women who used bisphosphonates was 1.40 (1.15, 1.70). In additional analyses further adjusting for furosemide use and intakes of individual nutrients, results were not appreciably changed (data not shown).

A history of VF was also associated with higher risk of hearing loss (Table 4). In NHS, compared with women without VF, the MVRR among women with confirmed VF was 1.31 (1.16, 1.49). In NHS II, compared with women without VF, the MVRR among women with self-reported VF was 1.39 (1.13, 1.71). There were only six cases among women with confirmed VF, and the MVRR was 1.62 (0.72, 3.61). We did not observe a significant association between history of HF and risk of hearing loss; the MVRRs were 1.00 (0.86, 1.16) in NHS and 1.15 (0.75, 1.74) in NHS II.

In additional analyses among the subcohort of NHS II participants in the CHEARS AAA, we examined the association of osteoporosis/LBD and individual audiometric hearing thresholds (Figure 1 and Table S2). Compared with women without osteoporosis/LBD, the mean hearing thresholds were significantly higher (worse) among women with osteoporosis/LBD who used bisphosphonates, particularly at audiometric frequencies of 2 kHz and above. For example, mean worse ear hearing thresholds were 2.19 dB HL higher at 2 kHz, 2.6 dB HL higher at 6 kHz, and 3.66 dB HL higher at 8 kHz than among women without osteoporosis/LBD. There was a suggestion that mean thresholds were also higher at 3 and 4 kHz, but the differences were not statistically significant. The mean thresholds among women with osteoporosis/LBD who did not use bisphosphonates were not significantly different from women who did not have osteoporosis.

Figure 1.

Multivariable-adjusted mean differencesa in audiometric thresholdsb (dB HL) among women in the NHS II (n = 3749) with and without prevalent osteoporosis/LBD and bisphosphonate use (cross-sectional analysis). aMultivariable-adjusted differences between the mean threshold values, measured in decibels hearing level (dB HL), among women with osteoporosis/LBD compared with women without osteoporosis/LBD (referent group). Mean differences are adjusted for age, race/ethnicity, body mass index, waist circumference, physical activity, DASH diet score, alcohol intake, thiazide use, history of hypertension, history of diabetes mellitus, type II, smoking, ibuprofen use, acetaminophen use, menopausal status, and oral hormone use. bPure-tone audiometric thresholds indicate hearing sensitivity for calibrated pure tones. Audiometric hearing thresholds were measured at individual frequencies (0.5–8 kilohertz, kHz) in each ear. *p < 0.03, indicating statistically significant differences between the mean threshold values among women with osteoporosis/LBD and BP use compared with women without osteoporosis/LBD (referent group). Bisphos, bisphosphonate; Bisphos—No, women with osteoporosis/LBD who did not use bisphosphonates; Bisphos—Yes, women with osteoporosis/LBD who used bisphosphonates; dB HL, decibels hearing level; kHz, kilohertz; LBD, low bone density; No OP/LBD, women without osteoporosis or low bone density; OP, osteoporosis

processing....