Health effects of removing amalgam fillings in people with subjective health complaints: systematic review
Systematic review
|Published
This systematic review summarizes research on the health effects of removing amalgam fillings (either extracting the tooth or replacing the filling) compared to no removal in individuals with subjective health complaints they attribute to amalgam fillings.
Key message
Some individuals with amalgam fillings report health issues that they believe may be attributed to their amalgam fillings.
We conducted a systematic review of the health effects of removing compared to not removing amalgam fillings in individuals who believe their health complaints are related to amalgam fillings. We performed a systematic literature search for controlled studies and included all types of health outcomes. We assessed the risk of bias in the included studies and used the GRADE method (Grading of Recommendations, Assessment, Development, and Evaluations) to assess the certainty in the evidence base.
Two small studies met the inclusion criteria:
A Norwegian controlled before-and-after study from 2011 with 40 participants and a German randomized controlled trial from 2008 with 90 participants investigated whether the removal of amalgam fillings affected subjective health complaints. Both studies had significant methodological limitations, including a small number of participants, incomplete reporting, and a lack of blinding of participants, personnel, and outcome assessors. Therefore, we have very low confidence in the results from these studies.
This systematic review shows that there is a lack of solid research evidence on the effect of amalgam removal on health outcomes in people with subjective health complaints that they attribute to their amalgam fillings. Based on this, we conclude that it is highly uncertain whether removing amalgam has any effect on patients' self-reported health complaints.
Summary
Introduction
Dental amalgam has been the most commonly used filling material for the repair of teeth affected by cavities/tooth decay for over 100 years. Due to environmental concerns, mercury-containing products, including amalgam, were banned in Norway from 2008. However, the material also has many positive properties: it is durable, long-lasting, affordable, and technically simple to produce. The most common alloy used in Norway consisted of mercury mixed with an alloy where the main components were silver and tin. Most people with amalgam fillings experience no health complaints. However, some individuals report health problems they believe can be attributed to their amalgam fillings, such as fatigue, dizziness, pain, and psychological symptoms. One measure to alleviate these health complaints has been to remove the amalgam fillings and replace them with composite resin or other filling materials.
Objective
This systematic review summarizes research on the health effects of removing amalgam fillings (either extracting the tooth or replacing the filling) compared to no removal in individuals with subjective health complaints they attribute to amalgam fillings.
Method
We searched for controlled studies in databases and other resources. Two researchers independently assessed the studies against the inclusion criteria, evaluated the risk of bias in the included studies, extracted data, and assessed the certainty in the evidence using the GRADE approach.
Results
We included one Norwegian controlled before-and-after study, published in 2011, and one German randomized controlled trial, published in 2008. Both studies included individuals with subjective health complaints that they attributed to amalgam fillings, without other medical or psychological findings that could explain the symptoms.
In the Norwegian study, there were 20 individuals in the intervention group who had their amalgam fillings removed and replaced with other materials. The control group, also consisting of 20 individuals, received no treatment. The authors additionally randomized 10 individuals to serve as replacements in case of dropout from the intervention group. The follow-up period was a total of 3 years, with measurements taken at baseline, after 3 months, 1 year, and 3 years.
Main findings of the Norwegian study at the 3-year follow-up and our assessment of the certainty in the results (GRADE):
Outcome after 3 years* | Effect* | Absolute effect | # participants (# studies) | Certainty in the estimate (GRADE) | |
Amalgam removal Mean difference (95% CI) | No amalgam removal Mean difference (95% CI) | ||||
Intra-oral index (numerical rating scale 0-60, lower is better) | 3.5 (0.4 til 6.6) | -0.6 (-6.4 to 5.2) | 6.9 index score (1.3 to 12.4) | 40 | ⨁◯◯◯ VERY LOW a |
Extra-oral index (numerical rating scale 0-50, lower is better) | 2.0 (-2.2 to 6.2) | -0.6 (-4.8 to 3.6)
| 4.3 index score (-1.6 to 10.3) | 40 | ⨁◯◯◯ VERY LOW a |
General health index (numerical rating scale 0-120, lower is better) | 10.1 (5.0 to 15.2) | -2.3 (-13.1 to 8.5) | 14.2 index score (2.4 to 26.0) | 40 | ⨁◯◯◯ VERY LOW a |
CI: Confidence interval. The absolute effect is calculated using an analysis of covariance (ANCOVA) and shows changes in index scores in the intervention group and the control group, adjusted for gender, age, and health complaints from baseline. *Difference between Questionnaire 1 (baseline) and 3-year follow-up. Positive values indicate a reduction in health complaints, while negative values indicate increased complaints; **Results presented from an intention-to-treat analysis. | |||||
Justification for the GRADE assessments: a Downgraded due to challenges with outcome measurements, differences in the methods of outcome measurement between the groups, relatively high dropout rates, and a small number of participants. | |||||
Explanation: Intra-oral index: intraoral burning sensation, intraoral pain ⁄ tenderness, taste disturbances, intraoral stiffness ⁄ paresthesia, dry mouth and increased saliva secretion ⁄ mucus. Extra-oral index: extraoral burning sensation, extraoral pain ⁄ tenderness, extraoral stiffness ⁄ paresthesia, extraoral skin problems and pain from the jaw joints. General health complaints: musculoskeletal complaints, gastrointestinal complaints, cardiovascular complaints, skin complaints, eye and vision complaints, ear and hearing complaints, nose and throat complaints, fatigue, dizziness, headaches, memory problems, concentration difficulties and anxiety and depression. | |||||
- It is uncertain whether the removal of amalgam fillings affects subjective health complaints compared to not removing amalgam fillings.
In the German study, 30 individuals were randomized to have their amalgam fillings removed (Group A), 29 individuals to have their amalgam fillings removed and receive a biological detoxification intervention (Group B), and 31 individuals to participate in a lifestyle program without the removal of amalgam fillings (Group C, control). The detoxification program consisted of a series of vitamins administered over 12 weeks, starting 4 weeks prior to the amalgam removal. The control group participated in a health-related lifestyle program consisting of 14 consultations with two-hour group sessions. The follow-up period was 1.5 years, with measurements taken at baseline, as well as after 1, 2, 6, 12, and 18 months. The main outcome, from which the results were reported, was at the 12-month follow-up.
Main findings of the German study at the 12-month follow-up, along with our assessment of certainty in the evidence (GRADE):
Outcome after one year* | Effect | p-value of adjusted ANOVA | # participants (studies) | Certainty in the estimate | |
A: Amalgam removal, B: Amalgam removal + detox Mean difference (SD) | Lifestyle programme without the removal of amalgam fillings (Control) Mean difference (SD) | ||||
Main outcome: Change in weighted sum score of the three main complaints (0-9, lower values are better) | A –3.5 (SD 2.2) B –3.5 (SD 2.2) | -2.5 (2.4) | 0.091* | 90
| ⨁◯◯◯ VERY LOW a,b |
Health complaints: Total symptom score (0-150, lower is better) | A -21.8 (17.6) B -24.0 (16.5)
| -16.3 (12.2) | 0.230 | 90 | ⨁◯◯◯ VERY LOW a,b |
Number of complaints (0-50, lower is better) | A -9.7 (8.4) B -10.5 (9.2)
| -2.5 (2.4) | 0.091 | 90
| ⨁◯◯◯ VERY LOW a,b
|
Physical health: SF-36 (0-100, higher is better) | A 3.9 (8.5) B 2.4 (11.5)
| 1.0 (6.9) | 0.546 | 71 | ⨁◯◯◯ VERY LOW a,b
|
Mental health: SCL-90-R Global severity index (90 items, 0-4 each item, lower is better) | A –5.5 (7.4) B –4.9 (8.9) | -6.9 (5.4) | 0.633 | 74 | ⨁◯◯◯ VERY LOW a,b
|
Internal locus of control, KGG (Stanine values) | A –0.2 (9.3) B 2.3 (7.8) | 4.1 (7.2) | 0.197 | 74 | ⨁◯◯◯ VERY LOW a,b
|
ANOVA: analysis of variance (ANOVA); KKG: The Health Locus of Control Scale; SCL-90-R: Symptom Checklist SCL-90-R; SD: standard deviation; SF-36: The 36-Item Short Form Health Survey. Absolute effect is calculated as an ANOVA adjusted for baseline values and based on the intention-to-treat population. * P-value for ANCOVA (analysis of covariance) between the groups, adjusted for baseline values. See Appendix 6 for effect estimates calculated in different ways, none of the analytical methods yielded statistically significant differences. For stanine values, scores of 4-6 are average, scores lower than or equal to 3 are below average, while scores of 7 or higher are above average. | |||||
Justification for the GRADE assessments: a Downgrade two levels due to challenges with lack of blinding of patients, therapists, and outcome assessors combined with subjective outcome measures, relatively high dropout, and lack of reporting of outcome measures in the protocol. b Downgraded one level due to the small number of participants. | |||||
- It is uncertain whether the removal of amalgam (removal alone or removal plus biological detoxification) affects subjective health complaints compared to a lifestyle programme without amalgam removal.
Discussion
We included one Norwegian controlled before-and-after study and one German randomized controlled trial that examined the health effects of removing amalgam fillings. The small number of participants and challenges related to outcome measurements resulted in a high risk of systematic biases in the studies, weakening our confidence in the evidence base. Therefore, there is considerable uncertainty about whether amalgam removal affects health outcomes in individuals with subjective health complaints they attribute to amalgam fillings. This is consistent with findings from a previous systematic review. Conducting controlled studies that can provide reliable or stronger evidence of the health effects of removing amalgam in individuals reporting health complaints they attribute to amalgam fillings is challenging, as blinding is not possible, and the outcome measures consist of subjective health complaints.
Conclusion
This systematic review summarizes the existing research on the health effects of removing compared to not removing amalgam in individuals with health complaints they attribute to amalgam. Our confidence in the evidence is very low, and it is therefore not possible to draw definitive conclusions about the effect.