Treatment of health complaints attributed to amalgam.

Abstract

The aim of the present study was to compare the reduction of subjective complaints by 3 treatment strategies in 90 "amalgam patients" whose complaints could not be explained by a medical or psychological disorder. The individuals were randomly assigned either to removal of dental amalgam only (removal group), or removal in combination with a "biological detoxification" therapy with high doses of vitamins and trace elements (removal-plus group), or participation in a health promotion program without removal of dental amalgam (no-removal group). Between baseline and month 12, the sum score of main complaints decreased by 3.5 (SD=2.2) points on average in the removal group as well as in the removal-plus group, and by 2.5 (SD=2.4) points in the no-removal group (p=0.152). Both removal groups showed a significant decrease in steady-state levels of inorganic mercury compared with the no-removal group. Thus, all 3 interventions were associated with clinically relevant improvements.

Full text

Page 1

349
RESEARCH REPORTS
Clinical
D. Melchart',2*, S. Vogt3, W. Köhler',
A. Streng', W. Weidenhammer',
L. Kremers3, R. Hickel3, N. Felgenhauer4,
T. Zilker4, E. Wühr5, and S. Halbach6
1Centre for Complementary Medicine Research, Internal
Medicine II, Technische Universität München, Munich,
Germany;
Department of Internal Medicine, University Hospital Zürich,
Switzerland;
Periodontology and Pediatric Dentistry, University of
Munich, Germany; 4Department of Clinical Toxicology,
Internal Medicine II, Technische Universität München,
Munich, Germany; 5International Association for Holistic
Dental Medicine, Mannheim, Germany; and 6Institute of
Toxicology, GSF Research Centre for Environment and
Health, Munich, Germany; *corresponding author, Kaiserstr.
9, D-80801 Munich, Germany, dieter.melchart@lrz.tum.de
2Division of Complementary Medicine,
3Department of Restorative Dentistry,
J Dent Res 87(4):349-353, 2008
ABSTRACT
The aim of the present study was to compare the
reduction of subjective complaints by 3 treatment
strategies in 90 "amalgam patients" whose
complaints could not be explained by a medical or
psychological disorder. The individuals were
randomly assigned either to removal of dental
amalgam only (removal group), or removal in
combination with a "biological detoxification"
therapy with high doses of vitamins and trace
elements (removal-plus group), or participation in
a health promotion program without removal of
dental amalgam (no-removal group). Between
baseline and month 12, the sum score of main
complaints decreased by 3.5 (SD = 2.2) points on
average in the removal group as well as in the
removal-plus group, and by 2.5 (SD = 2.4) points
in the no-removal group (p = 0.152). Both
removal groups showed a significant decrease in
steady-state levels of inorganic mercury compared
with the no-removal group. Thus, all 3
interventions were associated with clinically
relevant improvements.
KEY WORDS: Dental amalgam, mercury,
subjective health complaints, controlled trial.
Received July 26, 2006; Last revision October 17, 2007;
Accepted December 14, 2007
A supplemental appendix to this article is published
electronically only at
http://jdr.iadrjournals.org/cgi/content/full/87/4/349/DC1
Treatm ent of Health Com plaints
Attributed to Amalgam
INTRODUCTION
ental amalgam is still widely used as a filling material in the treatment of
dental caries, but safety concerns relating to its mercury content have been a
topic of discussion for many years (Hörstedt-Bindslev et al., 1991; LSRO,
2004; Bates, 2006). This and the known toxicity of methylmercury for the
developing brain have recently initiated two randomized longitudinal studies on
the neuropsychological effects of amalgam in children (Bellinger et al.,
2006; DeRouen et al., 2006). Some amalgam-bearers with chronic,
subjective, non-specific health problems suspect that their complaints are
caused by mercury released from their amalgam fillings. Typically, these so-
called "amalgam patients" report a large number and variety of symptoms
(Yontchev et al., 1986; Melchart et al., 1998). Some of the reported
symptoms correspond well with the major toxic effects on the nervous and
immune systems that are considered to be associated with chronic subtoxic
exposure to mercury (Molin, 1990; Eneström and Hultman, 1995).
In 1995, approximately 1500 persons filed a law suit against the
Degussa company (former main manufacturer of amalgam in Germany),
claiming health injury by dental amalgam. The ensuing settlement generated
funds which were allocated by an independent research funding
organization for investigation of controversial questions such as the
detrimental potential of amalgam, diagnosis of injury by amalgam, and
treatment of persons who relate their health problems to amalgam. As part
of the research program, we conducted a controlled trial to investigate the
effectiveness of 3 treatment strategies for "amalgam patients" in reducing
subjective health complaints that could not be explained by other medical or
psychological disorders. Thus, the underlying null hypothesis assumed no
differences among the treatment regimens.
MATERIALS & METHODS
Design
The study was randomized and controlled, comparing removal of dental
amalgam (removal group), removal of dental amalgam combined with a
'biological detoxification' therapy (removal-plus group), and participation in a
health promotion program without removal of amalgam (no-removal group).
Participants were not blinded to treatment. Randomization was done by
telephone according to a random list generated in advance, and stratified
according to the number of amalgam restoration surfaces. The protocol was
approved by two university ethics committees. All study participants provided
written informed consent.
Participants
Inclusion criteria were: persons with dental amalgam restorations who suspected
that their health complaints were caused by dental amalgam; having reported at
least 10 symptoms (including at least 3 of strong intensity); and age 20 to 50 yrs.
Exclusion criteria were: persons with bridges, crowns, or gold inlays;
persons having undergone unsuccessful endodontic treatment; having relevant
D

Page 2

2 Melchart et al. J Dent Res 87(4) 2008

by professionals from the Centre for
Complementary Medicine Research.
Outcomes
Initially, extensive toxicological,
psychiatric, and dental screenings
were performed.
treatment, compliance with the
medication regimen in the removal-
plus group and participation in the
health promotion sessions in the no-
removal group, and the occurrence
of serious adverse events were
documented by the physician at each
visit. Total and inorganic mercury
levels were determined in plasma
and erythrocytes by cold-vapor
atomic absorption. Total mercury
was also determined by urinalysis
(Halbach et al., 1998, 2003).
Concomitant
At baseline, participants were
given a pre-defined symptom list
with 50 items (scoring from 0 = not
present to 3 = strong intensity)
and additionally were asked to rank
their three main complaints (rating
scale from 0 = not present to 9 =
extreme), resulting in a weighted
sum score. This procedure was
repeated for the
mos after the start of treatment.
The number of complaints as
well as a total symptom score were
organic, allergic, or mental disorders; inability to understand the
study; alcohol or drug abuse; pregnancy or lactation; participation
in any clinical research study in the preceding 3 months.
Treatments
Removal of amalgam and replacement by other restoration
materials were performed by the University Unit of Dentistry,
including provisions for protecting both participant and physician
during treatment. Amalgam was removed by quadrant, with at least
1 wk between visits; underlying restorations and carious dentin
were removed completely. Calcium hydroxide was used as a liner
in cases of very deep restorations (caries profunda) before being
restored with ceramic or gold inlays, or a composite restoration.
In the removal-plus group, participants were additionally treated
with high doses of vitamins and trace elements (following the
recommendations of the International Association for Holistic Dental
Medicine), intended to support the excretion of mercury from the
body. Biological detoxification therapy lasted for 12 wks, beginning
4 wks before amalgam removal, and consisted of the daily intake of
tablets containing: vitamin B6 (100 mg), vitamin C (1 g), vitamin E
(300 mg), calcium (500 mg), selenium (200 µg as sodium selenite),
zinc orotate (2 x 40 mg, i.e., 2 x 6.3 mg zinc), and a garlic preparation
(1 x 100-3 00 mg). Participants were not allowed to take vitamin C
and selenium at the same time of day, to avoid neutralization.
Those in the no-removal group participated in a health
promotion program, aimed at developing health-related lifestyle
management skills suitable for individuals' everyday life
(Wunderlich and Melchart, 2002). The program consisted of 14
two-hour group sessions (up to 12 participants) and was
conductedinitially selected complaints at visits 1, 2, 6, 12, and 18
also determined. At baseline and 6, 12, and 18 mos after the start of
treatment, participants completed the SF-36 to assess health-related
quality of life (Ware et al., 1993), the Symptom Checklist SCL-90-
R, providing a general index of symptom severity as a measure for
overall psychological distress (Derogatis, 1992), and the KKG
questionnaire to estimate the participant's healthrelated locus of
control (Lohaus and Schmitt, 1989). At baseline, participants
completed two additional psychometric instruments: the SAM
questionnaire, assessing dispositional self-consciousness (Filipp
and Freudenberg, 1989); and FPI-R, an inventory assessing basic
personality traits (Fahrenberg et al., 1994).
The main outcome measure was the difference in the main
complaints sum score between baseline and 12 mos. Pre-defined
secondary outcomes included: total symptom score after 18 mos,
quality of life, psychic symptoms and signs, and mercury levels in
blood and urine after 12 and 18 mos.
Statistics
Since there were no data to estimate the expected effect sizes for
each treatment group, mean reductions for the main outcome
measure were assumed as: 2.5 in the removal group, 3.0 in the
removal-plus group, and 1.0 in the no-removal group, with a
common standard deviation of 2.5 (a = 5%, two-sided). A
minimum sample size of 29 participants per group met these
conditions, showing an 80% power to reject the null hypothesis
(Elashoff, 2000).
All randomized participants with baseline data on symptom
score and dental status were defined as the intention-to-treat (ITT)
population, while protocol violators (see Fig. 1) until month 12

Figure 1. Trial flow chart (ITT = intention to treat, MOM = main outcome measure, PP = per protocol).

Page 3

J Dent Res 87(4) 2008 The German Amalgam Trial (GAT) 3
349
were excluded from the per-protocol (PP) population.
Statistical testing of the main outcome measure was
performed on the ITT population, with missing data replaced with
baseline values (thus setting the differences compared with
baseline to zero) by analysis of variance. In case of rejection of the
null hypothesis, posteriori pair-wise comparisons were planned. In
addition, sensitivity analyses were performed for the main outcome
measure, with all available data or replacement of missing data by
the 'last value carried forward' method. Exploratory analyses
(without adjustment for multiple testing) were done for pre-defined
secondary outcome measures. All data were analyzed descriptively
(mean values with standard deviations, 95% confidence intervals),
and percentages were provided. In case of baseline differences (p <
0.1), analyses of covariance with the baseline values as covariates
were carried out. Additional sensitivity analyses were performed
excluding participants who had expressed a preference for a
specific treatment at baseline, since there was some concern about
whether one of the study physicians had followed the correct
randomization procedure (see Fig. 1).
Participants
Persons were included in the study between April, 1998, and July,
2002. Most participants were recruited through reports in local
newspapers. Approximately 1200 persons expressed interest in
participating in the study (Fig. 1), 164 entered at baseline, and 91
were randomized. One individual randomized to the removal group
dropped out before treatment without complete baseline data on
dental status (90 people in the ITT population). Two people in the
removal group, two in the removal-plus group, and eight in the no-
removal group withdrew or were lost to follow-up at month 12 (p
= 0.041, x2 test).
RESULTS
Disposition of Participants
As their most important complaint, 64% of the participants
reported either skin disease, headache, mental complaint (e.g.,
nervousness, sleeplessness), general tiredness/weakness, or an
infection/low resistance to infections. Complaints like allergies,
sensory disturbances, and urological, gastrointestinal, or
cardiovascular symptoms were reported less frequently.
At baseline, groups were comparable for most variables.
No statistically significant differences between the treatment
groups could be found (APPENDIX Table).
The mean total numbers of treatment sessions were 10.3

Figure 2. Course of mean weighted main complaints sum score (means
and 95% CI; ITT with missing values replaced by baseline values) for all
three groups.
(SD = 6.1) in the removal group, 13.4 (SD = 7.4) in the
removal-plus group, and 9.9 (SD = 2.9) in the no-removal
group (p = 0.058). In both removal groups, an average of 4.1
(SD = 1.4) and 4.4 (SD = 1.3) sessions was needed purely for
amalgam removal; the remaining sessions were preparations
due to the final restoration therapy. In the no-removal group,
19% of the participants participated in fewer than 8 sessions,
and 31% attended at least 12 sessions, but only two people
complied with the full program.
Primary and Secondary Outcomes
Between baseline and month 12, the 'main complaints' sum
score decreased by an average of 3.5 (SD = 2.2) points each in
the removal and removal-plus groups, and by 2.5 (SD = 2.4) in
the no-removal group (p = 0.152; ITT population; see Fig. 2).
The sensitivity analyses confirmed the main result (Table 1).
Table 1. Main Outcome Measurement: Mean Differences (standard deviations) between Baseline and Month 12 and Sensitivity Analyses, Unadjusted
(ANOVA), and Adjusted for Baseline Values (ANCOVA)
Group A
Removal of
Dental Amalgam
n
Mean (SD1)
Group B
Removal Plus
Biological Detoxification
n Mean (SD1)
Group C
Health Promotion Program
without Removal
n
Mean (SD1)

p2 p3
Missing values replaced by baseline value
Missing values replaced by last value carried forward
No replacement of missing values
Per protocol population (no missing values)

1 Standard deviation. 2 p-value for ANOVA among groups. 3 p-value for ANCOVA among groups adjusted for baseline values.
30
30
28
25
-3.5 (2.2)
-3.5 (2.2)
-3.8 (2.1)
-3.6 (2.1)
29
29
26
21
-3.5 (2.2)
-3.6 (2.0)
-3.9 (2.0)
-3.9 (1.8)
31
31
23
21
-2.5 (2.4)
-2.8 (2.3)
-3.4 (2.1)
-3.3 (2.2)
0.152
0.286
0.683
0.626
0.091
0.194
0.557
0.597

Page 4

4

Melchart et al. J Dent Res 87(4) 2008
Table 2. Mean Differences (standard deviations) between Baseline and Month 12 in Secondary Outcomes for the
Three Treatment Groups


Group A
Removal of
Dental Amalgam
n Mean (SD1)

Group B
Removal Plus
Biological Detoxification
n Mean (SD1)

Group C
Health Promotion Program
without Removal
n Mean (SD1)

p2
Participant questionnaires
Total symptom score (0-150)
No. of complaints (0-50)
No. of strong complaints (0-50)
SF-36 physical health3,4
SF-36 mental health3,4
SCL-90-R Global Severity Index4
KKG internal locus of control5
KKG external locus of control5
KKG fatalistic externality5

27
27
27
26
26
26
26
26 1.1
26
- 21.8 ( 17.6)
- 9.7 ( 8.4)
- 4.5 ( 3.2)
3.9 ( 8.5)
3.5 ( 8.3)
- 5.5 ( 7.4)
- 0.2 ( 9.3)
( 11.8)
0.8 ( 12.1)
25 - 24.0 ( 16.5)
25 - 10.5 ( 9.2)
25 - 4.7 ( 3.0)
24 2.4 ( 11.5)
24 4.4 ( 10.6)
26 - 4.9 ( 8.9)
26 2.3 ( 7.8)
26 0.2 ( 10.4)
26 - 0.4 ( 9.2)
23
23
23
21
21
22
22 4.1
22
22 -
- 16.3 ( 12.2)
- 7.6 ( 7.7)
- 3.6 ( 2.5)
1.0 ( 6.9)
5.0 ( 9.1)
- 6.9 ( 5.4)
( 7.2)
0.7 ( 8.9)
1.8 ( 11.0)
0.230
0.487
0.437
0.546
0.858
0.633
0.197
0.947
0.711
Mercury concentrations
Total in blood plasma (ng/mL)
Inorganic in blood plasma (ng/mL) 26
Total in erythrocytes (ng/mL)
Inorganic in erythrocytes (ng/mL)
In urine (ng/mL)
In urine excretion (ng/8 hrs)

26 - 0.43 ( 0.39)
- 0.44 ( 0.38)
0.06 ( 1.77)
- 0.41 ( 0.38)
26 - 1.15 ( 1.42)
25
-4 8 9 .4 (4 7 0 .0 )
26 - 0.46 ( 0.63)
26 - 0.45 ( 0.57)
26 - 0.42 ( 1.69)
26 - 0.43 ( 0.49)
26 - 1.86 ( 2.71)
26 -718.5 (1004.3)
23
21
23
22
23
23 -1 6 2 .3 (3 7 3 .6 )
- 0.16 ( 0.27)
- 0.12 ( 0.17)
- 0.73 ( 1.57)
- 0.12 ( 0.19)
- 0.48 ( 1.43)
0.049
0.013
0.259
0.012
0.054
0.022
26
26

amalgam
without
detoxification therapy)
was associated with a
marked reduction in
the participants' sub -
jective complaints and
mental stress. How -
ever, a health pro -
motion program with -
out amalgam removal
was similarly effective.
The im provements
observed in all groups
were clinically relevant
and persisted through -
out the follow-up
period of 18 mos.
This study is the
first comparative ran -
domized controlled trial
of treatment strategies
for adults with amal -
gam restor ations, and
benefited from the
cooperation of experts
from medical, dental,
and toxicological
departments. Strengths
of the study were
extensive screening
procedures, strict ex -
clusion criteria, and
(with
biological
or
1
2
3
4
5
Standard deviation.
p-value for ANOVA among groups.
Positive differences indicate improvement.
tvalues.
Stanine values.
The groups did not differ significantly with regard to mean
differences in secondary outcomes between examinations at
baseline and 12 mos later (Table 2). Descriptive analyses
indicated persistent improvements up to month 18.
With the exception of total mercury in erythrocytes,
mercury concentrations of all measured blood and urine
parameters were significantly lower after amalgam removal as
compared with concentrations in the no-removal group (Table
2). In all three groups, the concentrations of inorganic mercury
in erythrocytes and plasma, of total mercury in plasma, and
of the urine parameters hardly changed between months 12 and
18.
Safety and Tolerability
Three serious adverse events (in-patient surgical treatments
considered unrelated to study condition and intervention) were
documented; all three persons continued the trial.
During the study, 43 participants (12/14/17, respectively,
for the three groups) reported 73 complaints as 'new'. Overall,
16% of the complaints were gastrointestinal symptoms, 15%
arthralgia/back pain, 11% dental problems due to the
replacement of restorations (only in the removal groups), 10%
skin diseases, 7% infections, 7% sensory disturbances, and
34% other complaints. Four women (all in the removal-plus
group) became pregnant.
DISCUSSION
The focus of this controlled trial was to investigate treatment
options for so-called "amalgam patients". Removal of dental
high-quality treatment schedules for amalgam removal following
generally accepted guidelines. The validity of our results is
supported by the consistency of findings from various variables.
The study participants cannot be regarded as representative
of all persons with amalgam. Further limitations are that the
definition of "amalgam patients" and the measurements of
improvements are based on subjective criteria and may be
underpowered. Recruitment for the trial turned out to be
difficult and required several years of effort, mainly because
individuals with other dental materials in combination with
amalgam were excluded. Many individuals insisted on
amalgam removal and therefore refused randomization. Since
blinding was not possible, the relevant effects observed in the
removal groups may also be due to the expectations of the
person, the natural course of the complaints, or placebo effects.
Individuals randomized to the no-removal group showed a high
drop-out rate. Due to the unequal drop-out, they tended to
benefit less than those in both removal groups in the main
analysis (missing values replaced by baseline values), but the
results of the sensitivity analyses suggest improvements similar
to those reported in the other groups.
The strong effects of the health promotion program on the
subjective complaints of "amalgam patients" were unexpected,
especially since we observed only a weak relationship between
numbers of treatment sessions and symptom relief. A possible
explanation may be that by adopting a health-promoting lifestyle,
including good nutrition, exercise, and relaxation techniques, the
individuals' general health improved, e.g., by strengthening the

Page 5

J Dent Res 87(4) 2008 The German Amalgam Trial (GAT) 5
immune system and reducing 'amalgam anxiety'. Another reason
may be that participants acquired coping strategies for their
complaints (Gottwald et al., 2002). Placebo effects in the no-
removal group cannot be ruled out as well (Grandjean et al., 1997).
The mercury measurements revealed that, although
removal plus biological detoxification therapy tended to result
in slightly lower mercury values than amalgam removal alone,
participants from both removal groups showed similar
improvements in their subjective outcome measurements.
Furthermore, a low mercury level was not a precondition for
subjective improvement. The strong effects on health
observed in the no-removal group are unlikely to be
explained by the slight decrease in mercury levels (possibly
caused by the participants' avoiding additional mercury uptake by
adopting a healthier diet, thereby also ingesting more vitamins
and trace elements, similar to what would occur in biological
detoxification therapy). As a side-effect, the measurement of
total mercury in the erythrocytes showed no significant
difference in the three groups at months 12 and 18, while the
drop in inorganic mercury in both removal groups was
significant. Since the value for total mercury is the sum of
inorganic plus organic mercury, a latent increase in organic
mercury cannot be excluded for the post-removal data.
A recent review on the health effects of dental amalgam
(LSRO, 2004) concluded that, apart from allergic sensitivity,
there is insufficient evidence that various non-specific
complaints attributed to dental amalgam are actually caused by
mercury release from restorations. The review also suggested
that "amalgam patients" should be screened for underlying
dental, physical, and psychiatric conditions to exclude affective
symptoms independent of mercury exposure. The participants
in our study met these preconditions exactly.
Removal of dental amalgam and other metal alloys supported
by anti-oxidant therapy resulted in improved quality of life in
"amalgam patients" in a large retrospective study (Lindh et al.,
2002). Several, mainly observational, trials have reported the
improvement of various complaints after amalgam removal
(Nerdrum et al., 2004; Lygre et al., 2005; Tillberg et al., 2005),
while recent randomized trials showed no specific health effects
of amalgam restorations in children (Bellinger et al., 2006;
DeRouen et al., 2006). In our trial, amalgam removal was
associated with a marked reduction in the participants' subjective
health complaints. However, similar improvements were
observed after a health promotion program without amalgam
removal, while mercury levels deviated only slightly from
baseline. In conclusion, although the reasons for amalgam-related
complaints are still unclear, our results suggest that amalgam
removal is not the only treatment option, since all treatments were
associated with clinically relevant improvements.
ACKNOWLEDGMENTS
The study was funded by the Stifterverband für die Deutsche
Wissenschaft. The assistance of Bärbel Maier and Brigitte
Wahl is gratefully acknowledged. Trial registration:
http://www.controlled-trials.com/isrctn/trial/|/0/51459238.html,
ISRCTN51459238.
REFERENCES
Bates MN (2006). Mercury amalgam dental fillings: an epidemiologic
assessment. Int J Hyg Environ Health 209:309-3 16.
Bellinger DC, Trachtenberg F, Barregard L, Tavares M, Cernichiari E,
Daniel D, et al. (2006). Neuropsychological and renal effects of dental
amalgam in children: a randomized clinical trial. J Am Med Assoc
295:1775-1783.
Derogatis LR (1992). SCL-90-R. Administration, scoring and procedures
manual II for the r(evised) version. Towson, MD, USA: Clinical
Psychometric Research, Inc.
DeRouen TA, Martin MD, Leroux BG, Townes BD, Woods JS, Leitão J, et
al. (2006). Neurobehavioral effects of dental amalgam in children: a
randomized clinical trial. J Am Med Assoc 295:1784-1792.
Elashoff JD (2000). nQuery Advisor. Version 4.0 user's guide. Los Angeles,
CA.
Eneström S, Hultman P (1995). Does amalgam affect the immune system?
A controversial issue. Int Arch Allergy Immunol 106:180-203.
Fahrenberg J, Hampel R, Selg H (1994). Das Freiburger
Persönlichkeitsinventar FPI. Revidierte Fassung FPI-R und teilweise
geänderte Fassung FPI-A1 (Freiburg Personality Inventory). Göttingen,
Germany: Hogrefe.
Filipp S-H, Freudenberg E (1989). Der Fragebogen zur Erfassung
dispositionaler Selbstaufmerksamkeit SAM (A scale for assessing
dispositional self-consciousness). Göttingen, Germany: Hogrefe.
Gottwald B, Kupfer J, Traenckner I, Ganss C, Gieler U (2002).
Psychological, allergic, and toxicological aspects of patients with
amalgam-related complaints. Psychother Psychosom 7 1:223-232.
Grandjean P, Guldager B, Larsen IB, Jorgensen PJ, Holmstrup P (1997).
Placebo response in environmental disease. Chelation therapy of
patients with symptoms attributed to amalgam fillings. J Occup
Environ Med 39:707-7 14.
Halbach S, Kremers L, Willruth H, Mehl A, Welzl G, Wack FX, et al.
(1998). Systemic transfer of mercury from amalgam fillings before and
after cessation of emission. Environ Res 77:115-123.
Halbach S, Vogt S, Köhler W, Felgenhauer N, Kremers L, Melchart D, et al.
(2003). Relationships between dose and concentrations of total and
inorganic mercury in blood and urine of amalgam patients.
Environmental Sci 10:71-8 1.
Hörsted-Bindslev P, Magos L, Holmstrup P, Arenholt-Bindslev D (1991).
Dental amalgam—a health hazard? Copenhagen: Munksgaard.
Lindh U, Hudecek R, Danersund A, Eriksson S, Lindvall A (2002).
Removal of dental amalgam and other metal alloys supported by
antioxidant therapy alleviates symptoms and improves quality of life in
patients with amalgam-associated ill health. Neuro Endocrinol Lett
23:459-482.
Lohaus A, Schmitt GM (1989). Fragebogen zur Erhebung von
Kontrollüberzeugungen zu Krankheit und Gesundheit KKG (The
Health Locus of Control Scale). Göttingen, Germany: Hogrefe.
LSRO Life Science Research Office (2004). Report on health effects of
mercury from dental amalgam. Final report. www.Lsro.org/amalgam/
frames_amalgam_report.html (accessed: 26 December 2007).
Lygre GB, Gjerdet NR, Björkman L (2005). A follow-up study of patients
with subjective symptoms related to dental materials. Community Dent
Oral Epidemiol 33:227-234.
Melchart D, Wühr E, Weidenhammer W, Kremers L (1998). A multicenter
survey of amalgam fillings and subjective complaints in non-selected
patients in the dental practice. Eur J Oral Sci 106:770-777.
Molin M (1990). Mercury release from dental amalgam in man. Influences
on selenium, glutathione peroxidase and some other blood and urine
components. Swed Dental J 14(Suppl 71):1-122.
Nerdrum P, Malt UF, Høglend P, Oppedal B, Gundersen R, Holte M, et al.
(2004). A 7-year prospective quasi-experimental study of the effects of
removing dental amalgam in 76 self-referred patients compared with
146 controls. J Psychosom Res 57:103-111.
Tillberg A, Berglund A, Marell L, Bergdahl J, Eriksson N, Linden G, et al.
(2005). Changes in health over time in patients with symptoms
allegedly caused by their dental restorative materials. Community Dent
Oral Epidemiol 33:427-437.
Ware JE, Snow KK, Kosinski M, Gandek B (1993). SF-36 Health Survey
manual and interpretation guide. Boston: New England Medical Center,
The Health Institute.
Wunderlich S, Melchart D (2002). Strukturiertes Gesundheitstraining in der
klinischen Praxis. In: Naturheilverfahren. Leitfaden für die ärztliche
Aus-, Fort- und Weiterbildung. Melchart D, Brenke R, Dobos G,
Gaisbauer M, Saller R, editors. Stuttgart: Schattauer, pp. 74-88.
Yontchev E, Hedegard B, Carlsson GE (1986). Reported symptoms,
diseases and medication of patients with orofacial discomfort
complaints. Int J Oral Maxillofac Surg 15:687-695.