Dental mercury amalgam fillings associated with a deterioration of high-frequency auditory acuity

hearingl-lossHearing loss is by far the most common disorder of the senses, affecting more than 36 million people in the US alone. Mercury has been shown to affect the auditory system at a wide range of levels, from the cochlea to the cortex.  In this study, we compared the number and surface area of different types of dental fillings with auditory thresholds in the range of .25 to 16 kHz. having more amalgam fillings was associated with a deterioration of high-frequency auditory acuity (8 kHz and above), independent of socio-economic factors. Thus, these results suggest a detrimental, dose-dependent effect of amalgams on hearing. There is also a likely duration-dependent effect.


In this study we investigated the effects of amalgam dental fillings on auditory thresholds. Participants (n39) were non-smoking women age 40 to 45. Regression and correlation analyses were performed between auditory thresholds, measured from 0.25 to 16 kHz, and the number/surface area of dental fillings, using the ASHA criteria for ototoxic change as a reference for comparison.

No significant correlation (p0.05) was found between composite (non-amalgam) filling or drilling data and auditory thresholds. However, there was a significant positive linear correlation between amalgam filling data and auditory thresholds at 8, 11.2, 12.5, 14, and 16 kHz. 

The strongest association (r0.587, n39, pB.001, r2 0.345) was at 14 kHz, where each additional amalgam filling was associated with a 2.4 dB decline in hearing threshold (95% confidence interval [CI], 1.33.5 dB).

The results suggest an association between more amalgam fillings and poorer thresholds at higher frequencies, which could contribute to presbyacusis in developed countries. This provides further argument for the use of amalgams to be phased out where suitable alternatives exist. 


hearingloss-chartsAmalgam fillings and auditory thresholds An increase in the number of amalgams (range 0 to 16, mean 7.1) was associated with poorer thresholds (expressed in dB HL) at 8, 11.2, 12.5, 14, and 16 kHz (Table 1, Figure 1). The greatest correlation was at 14 kHz (n39, pB0.001, r20 35, F19.5), where each amalgam was associated with a 2.4 dB decline in hearing sensitivity (95% CI 1.33.5 dB). Using ASHA’s (1994) criteria of change (]20 dB at a single frequency), an ototoxic effect was observed for women aged 4045 with nine or more amalgam fillings.

When ASHA’s (1994) alternative criteria of ]20 dB at one frequency and/or ]10 dB at two adjacent frequencies was used, an ototoxic effect was observed with six or more amalgam fillings. Very similar results were obtained when considering the amalgam scores (taking into account the area of the fillings), but these results are not presented here.

In addition to the numbers and scores of the fillings at the time of testing, ”adjusted” values were also calculated, taking into account estimated exposure differences due to a reported substitution of filling type. Using the adjusted calculation, the effect upon thresholds was more pronounced. The worsening in threshold per amalgam at 8, 11.2, 12.5, 14, and 16 kHz for adjusted dental numbers (recorded numbers in parentheses) was 0.97 (0.85), 1.31 (1.14), 1.98 (1.75), 2.75 (2.39), and 1.43 (1.0) dB, respectively. Regardless of these findings, the statistical analysis in this study concentrated mostly on the actual number of dental fillings and score values recorded, because adjusted values were derived using a common-sense but unproven calculation method.

Composite fillings/other dental data and auditory thresholds Three tests of correlation were used: Kendall’s tau_b, Spearman’s rho, and Pearson’s. There was no significant correlation (p0.05) between the number (range 09, M2.4) or score (range 043, M10.1) of composite fillings and auditory thresholds at any frequency. Participants had a mixture of dental filling types, and those with many amalgams generally had fewer composites (rs0.32, n39, p0.04, r20.10).

There was no significant correlation (p0.05) between drilling episodes (range 425, M12.0) and auditory thresholds. Porcelain- and gold-filling data did not demonstrate a normal distribution and were not specifically compared with auditory thresholds; however, these data are included within the drilling episode data. Figure 2 shows the 14 kHz thresholds versus the number of amalgam fillings (upper panel), composite fillings (middle panel), and the number of drilling episodes (lower panel). The number of drilling episodes includes placement/ removal of amalgam, composite, porcelain, and gold fillings.


In this study, we compared the number and surface area of different types of dental fillings with auditory thresholds in therange of .25 to 16 kHz. For female participants taken from the local population, having more amalgam fillings was associated with a deterioration of high-frequency auditory acuity (8 kHz and above), independent of socio- conomic factors. Thus, these results suggest a detrimental, dose-dependent effect of amalgams on hearing. There is also a likely duration-dependent effect.

For the study participants, age and questionnaire data were not confounding factors. Also, there was no correlation between total drilling episodes and thresholds or number of composite fillings and thresholds, which implies that the observed effect was not due to another related event such as noise damage caused by the drilling process. 

It may be suggested that blood-mercury levels should have been recorded in order to determine the relationship between blood-mercury levels, amalgam numbers, and auditory thresholds; however, it was decided not to take this approach as blood mercury levels have been shown to be a poor indicator of the levels that can accumulate in other parts of the body (Goerling et al., 1992).

”Presbyacusis” is the term generally used to describe the process by which hearing acuity reduces with age, starting in the high frequencies and gradually progressing towards lower frequencies (and becoming more clinically significant). There is considerable variability between individuals in the degree and rate of this hearing loss, which is presumed to be due to a variety of individual and environmental factors. Tambs et al. (2003) found that 31 – 64% of the variance in age-related hearing loss was linked to noise exposure, ear infections, head injury, and gender, and the remaining variance was postulated to be due, at least in part, to environmental toxins. 

Regression data from this study indicate that women aged 40 – 45 with six or more amalgam fillings exhibit ototoxicity according to ASHA criteria (ASHA, 1994). With increasing age and, therefore, ongoing exposure, the percentage of participants exhibiting ototoxicity, as well as the degree of hearing loss, will gradually increase. Also, because ototoxicity progresses from high to low frequencies, it is probable that the detrimental effect will gradually become more apparent in speech frequencies, ultimately producing a clinically significant impairment. This may contribute to the hearing differences in older people between isolated and developed populations and the marked variations seen within developed populations. Indeed, it seems possible that the ototoxicity of amalgam fillings may be a component of what is currently labelled ”presbyacusis” in developed countries.

The mechanisms of mercury ototoxicity were not the subject of the present study, but it is relevant to consider these here. 

Mercury has been shown to affect the auditory system at a wide range of levels, from the cochlea to the cortex, and there would appear to be scope for further audiological investigation into the effects of amalgam fillings. In particular, the use of Auditory Brainstem Responses (ABRs) could yield useful information. Counter (2003) noted a relationship between ABR measures and exposure to mercury vapour in gold-mining activities. Murata et al. (2004) reported increased latencies of ABR peaks in those exposed to long-term methyl mercury and suggested that the measures indicated both permanent and reversible components. 

Thus, further research is needed in this area. For the age group investigated in the present study, a high-frequency tone burst or a high-frequency band-limited click stimulus is likely to be more sensitive than a conventional ABR click stimulus because audiometric thresholds appear to be more readily affected at 8 kHz and above. 

As far as we are aware, this is the first study that has shown a dose-dependent effect of amalgam fillings on health in the general population. In studies on children (Derouen et al., 2006; Bellinger et al., 2006), no adverse neurobehavioural, cognitive, or renal effect from exposure to amalgam was found. However, the children will inevitably have lower accumulated mercury levels due to the lower number of amalgam fillings and the shorter duration of exposure.

In addition, auditory thresholds were not considered in these studies. 

Based on the findings of this study, the use of alternative dental-filling materials, which do not show any evidence of ototoxicity, would be prudent; however, there is a dilemma regarding amalgam fillings already in place. Using a standard hearing test, Siblerud and Kienholz (1997) demonstrated improved audiometric thresholds six months after amalgam removal in a group of women with multiple sclerosis. However, although this is of considerable interest, the risk of complications due to removal of the amalgam may exceed the risk of long-term side effects (Berlin, 2003). Additional drilling can cause problems with existing teeth and causes a transient increase in blood mercury (Clarkson et al., 2003; Haikel et al., 1990), which may trigger a reaction in sensitive individuals, such as those with auto-immune conditions.

Replacement of amalgam fillings (by another type) at the end of their useful life may be a sensible solution in the majority of cases. Furthermore, the use of certain precautionary techniques (such as high-volume air extraction) can minimize mercury vapour exposure to the patient, dentist, and dental staff during drilling. 

In terms of public health and financial costs to health services (e.g. hearing aids), further research is clearly needed. The use of other dental filling materials, notably composites, has increased in recent years in many countries. However, the data from the current study provide an additional argument against the continued use of dental amalgam where suitable alternatives exist. 

[iframe width=”560″ height=”315″ src=”” frameborder=”0″ allowfullscreen]

About The Author

Leave a Reply