| Sign In to gain access to subscriptions and/or personal tools. |
Dental Amalgam and Antibiotic- and/or Mercury-resistant Bacteria
1 1 Box 357234, Departments of, Environmental and Occupational Health Sciences, and Correspondence: * corresponding author, marilynr{at}u.washington.edu
Mercury emitted from dental amalgam may select for increased numbers of antibiotic- or mercury-resistant commensal bacteria in patients and increase their risk for bacterial diseases that are resistant to common therapies. We hypothesized that the presence of dental amalgams would increase the level of mercury-, tetracycline-, ampicillin-, erythromycin-, or chloramphenicol-resistant oral and urinary bacteria as compared with levels in children receiving composite fillings. Samples were collected at baseline, 3–6 months after the initial dental treatment, and annually for 7 years of follow-up. There were no statistically significant differences between treatment groups in the numbers of bacteria growing on antibiotic- or mercury-supplemented plates. This study provided no evidence that amalgam fillings on posterior teeth influenced the level of antibiotic- or mercury-resistant oral or urinary bacteria as detected by culture.
Key Words: mercury • amalgam • antibiotic-resistant • mercury-resistant
Dental amalgam has been used for 150 years, is relatively low-cost, durable, and easy to use. The majority of human mercury (Hg) exposure has been reported to be from dietary sources and from dental amalgams (50% mercury), which emit mercury vapor (Dye et al., 2005). Two long-term dental amalgam safety studies (Bellinger et al., 2006; DeRouen et al., 2006) concluded that children receiving amalgam did not, on average, have differences in neurobehavioral assessments or nerve conduction velocity compared with children treated with composite materials.
Antibiotic and mercury resistance (Hgr) genes are often associated on the same conjugative elements in bacteria. Therefore, it might be possible that mercury released from amalgam may select for increased numbers of antibiotic-resistant commensal bacteria, which could pass these traits to pathogenic bacteria and increase the risk of bacterial diseases resistant to common therapies. Such infections are more difficult and costly to treat and have increased morbidity and mortality (Pike et al., 2002; Mindlin et al., 2005; Baker-Austin et al., 2006). One study on 6 monkeys suggested that amalgam did increase the numbers of Hgr Enterobacteriaceae, Streptococcus, and Enterococcus species (Wireman et al., 1997). Another study found that 71% of the children without amalgam restorations carried Hgr ( The aim of this study was to verify whether the presence of dental amalgam increased the levels of mercury- and antibiotic-resistant bacteria from baseline levels as compared with levels in children receiving composite fillings, which would be an unintentional consequence of dental treatment, and to examine the linkage between acquisition of antibiotic and Hg resistance in bacteria.
Study Design The participants were a subset of children who participated in a randomized controlled trial designed to assess the safety of amalgam (DeRouen et al., 2006). All children attended the Casa Pia schools in Lisbon, Portugal, and met the following inclusion criteria: (1)  1 caries lesion in a permanent tooth, (2) no prior amalgam fillings, (3) urinary mercury levels < 10 µg/L, (4) blood lead levels < 15 µg/L, (5) IQ score of 67, and (6) no interfering health conditions, including existing psychological, behavioral, neurologic, immunosuppressive, or renal disorders. The study protocol was approved by the institutional review boards at the University of Washington and University of Lisbon, and required written informed consent from parents or guardians and signed assent from the children prior to their participation. The sample size of 150 children was selected based on the requirement of 80% power to detect a half-log10 change in bacterial levels between two groups, assuming a standard deviation of 1 on the log-scale and 15% drop-out.
The Amalgam group received caries treatment with dental amalgam for posterior restorations, whereas the Control group received a resin composite material. In both groups, smaller and anterior restorations were treated with other materials selected from a list of materials normally used in the United States and Portugal. At each visit, the children received a dental cleaning by a trained professional and oral hygiene instructions for daily oral care. If further restorations were needed in the follow-up visits, these were done with the same treatments as those used initially (Tables 1
Sample Collection Samples were collected at baseline, after dental treatment (3–6 mos later), and at 7 annual follow-up visits. Oral samples were collected by means of the BBLTMCultureSwabTM Transport System (Becton Dickinson, Sparks, MD, USA). The first specimen was collected from the buccal mucosa (lower right quadrant) and the second sample from the right buccal gingiva. A trained dental hygienist swabbed the site from the midline to the back of the oral cavity and back again for 20 sec. Midstream urine was collected, and a 10-mL quantity was transferred to a sterile conical tube. Urine samples were centrifuged at 3000 rpm for 15 min and decanted; a 1-mL quantity of urine plus urine sediment was packaged on ice for shipping. Oral and urine samples were shipped to the University of Washington laboratory on the same day as they were collected, and typically arrived 50 hrs after they were shipped.
Laboratory Analysis The urine samples were serially diluted (1:10) in PRAS buffer, and a 100-µL quantity from each dilution was plated onto MacConkey media (MAC) without the crystal violet (Difco Lab.) and MAC supplemented with 25 3g/mL ampicillin, chloramphenicol, or tetracycline. At year 1 follow-up, samples were also plated on MAC supplemented with 200 3M HgCl2 made fresh weekly. Plates were incubated at 37°C aerobically for 24 to 48 hrs and counted to give numbers of bacteria/mL of urine. The percentage of children with Hgr bacteria was determined with the number of positive samples for growth on the unsupplemented MAC as the denominator. Urinary mercury concentrations were previously determined (DeRouen et al., 2006). Previously published work verified the correlation between phenotypic Hgr and antibiotic resistance and carriage of Hg and/or antibiotic resistance genes of randomly selected isolates (Luna et al., 2002; Ojo et al., 2004).
Statistical Analysis We compared the numbers of antibiotic-resistant or Hgr bacteria in each group using linear regression models applied to log10-transformed bacteria counts with the same covariates as listed above and the log-count for the unsupplemented plate as an additional covariate. Log-transformed counts were used to stabilize the variance and provide approximately normally distributed dependent variables. The GEE method was used to account for correlations as described above. For analyses of oral samples, the average of bacteria counts for gingival and buccal mucosal samples was used. In all treatment group comparisons, participants were retained in their assigned groups, and all data available on randomized children were included in the analyses, in accordance with the intent-to-treat principle. We performed additional analyses to test for associations between antibiotic and Hg resistance. These analyses used GEE and logistic regression models that controlled study year (as a categorical variable), gender, and race. Similar analyses tested the associations between the presence of Hgr or Apr bacteria in the urine samples and the presence of Hgr or Apr bacteria in oral samples.
Characteristics of Participants The demographics were similar at baseline and at follow-up year 7 in the two groups (Table 1  ). Children who were sampled at year 7 follow-up did not differ in baseline characteristics from those not followed up.
Dental Treatment and Mercury Exposure
Oral Bacterial Levels
The numbers of bacteria on all media increased after dental treatment and then decreased to approximately baseline levels over the subsequent few years. The highest numbers of bacteria were on tetracycline-supplemented media and were from 10- to 1000-fold less on the mercury media (Table 3 ). No significant associations between the level of urinary Hg and the presence of Hgr oral bacteria were found within either treatment group. However, children with Hgr bacteria were more likely to have Apr bacteria than those without Hgr bacteria (odds ratio 3.43, 95% CI 2.61– 4.51, P < 0.01).
Urinary Bacterial Levels
Logistic regression analysis showed that the presence of Hgr bacteria in the urine sample was associated with the presence of antibiotic-resistant bacteria, with odds ratios of 3.67 (95% CI 2.46–4.49, P < 0.01) for ampicillin-, 4.55 (95% CI 3.08–6.73, P < 0.01) for chloramphenicol-, and 1.82 (95% CI 1.29–2.56, P < 0.01) for tetracycline-supplemented media. Similar odds ratios were found when Tcr rather than Hgr bacteria were used, and were similar in the two treatment groups. There was no clear association between the level of urinary Hg and the presence of Hgr bacteria (data not shown), and no correlation between the presence of antibiotic- or Hg-resistant bacteria in the oral and urine samples collected at the same time period in either treatment group.
Mercury is toxic to bacteria, and, as a result, bacteria have acquired genes that confer Hgr (Foster, 1987; Barkay et al., 2003). These Hgr genes may pre-date man, since they were identified in Gram-negative Pseudomonas and Acinetobacter spp. isolated from permafrost samples dated at 15,000–40,000 years old (Mackert and Berglund, 1997), and certainly pre-date the antibiotic era (Essa et al., 2003). These studies suggest that the presence of Hgr genes in bacteria is ancient and evolved initially due to environmental mercury rather than in response to human activities. Hgr genes are thought to differ from most bacterial antibiotic-resistant genes, which have clearly appeared in response to human use of these agents over the last 50 years (Hughes and Datta, 1983). In the current study, the numbers of oral and urinary bacteria able to grow on 3 different antibiotic- and Hg-supplemented media were determined over 7 yrs. We found that, as expected, the children in the Amalgam group had more Hg exposure and had higher urinary Hg levels compared with those in the Composite group. No statistically significant differences between treatment groups were found in either the percentage of children with growth on Hg media or the numbers of Hgr bacteria for either the oral or urine samples. Similarly, we found no statistically significant differences between the groups in the percentage of children with growth on antibiotic media or antibiotic-resistant bacteria for either the oral or urine samples. We also found no correlation between carriage of antibiotic- and/or Hg-resistant oral and urinary bacteria at the same time period. Analysis of the data suggests that amalgam dental fillings had little influence on the numbers of children carrying oral or urinary antibiotic-/Hg-resistant bacteria. Similar results have been reported when a single time-point was used for oral samples from 41 children with dental amalgam fillings and 42 children without (Pike et al., 2002). However, other than age, no other demographics of the children were available in this previous study. The Casa Pia schools offered fish/seafood meals 2.4 times per wk, and there was the possibility of confounding by dietary mercury exposure. To rule out this possibility, we tested fish samples from the schools and found them to have low levels of methylmercury (13.9–23.6 ng/g), which were unlikely to influence oral and urinary bacteria. In addition, blood samples collected from the children at 1 year follow-up had lower Hg levels than baseline blood samples, while urinary Hg levels increased in the Amalgam group during this time period (Evens et al., 2001). For both treatment groups, the numbers of oral bacteria that grew on both supplemented and unsupplemented media increased at the second visit (after therapy), and then stabilized near baseline levels. Whether a comparable increase existed in the urine samples is not clear, because the length of time the samples were stored before shipment changed after the collection of the baseline samples. Previously, the merA genes from individual Hgr Gram-negative bacteria, from both treatment groups, were analyzed, and all carried the Gram-positive merA genes and were able to transfer these genes to Gram-positive Enterococcus faecalis. In addition, 78% of the isolates carried both Gram-positive and Gram-negative merA genes and were able to transfer the merA genes to both Gram-positive E. faecalis and Gram-negative E. coli recipients (Ojo et al., 2004). Only 2% of the Hgr Gram-positive bacteria were antibiotic-susceptible. The odds of a child carrying either oral Apr or urinary Apr or Cmr bacteria were higher when the sample contained Hgr bacteria. This study provides no evidence that treatment of children with amalgam dental fillings for posterior restorations influences the level of antibiotic-/mercury-resistant oral or urinary bacteria.
This study was supported by grant U01 DE-1189 and contract N01 DE-72623 from the National Institute of Dental and Craniofacial Research of the National Institutes of Health, Bethesda, MD, USA. Received for publication July 13, 2007. Revision received January 2, 2008. Accepted for publication February 4, 2008.
Journal of Dental Research, Vol. 87, No. 5,
475-479 (2008) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
