Skip to main content
NCBI home page
Journal of Dental Research logoLink to Journal of Dental Research
. 2012 Sep;91(9):853–858. doi: 10.1177/0022034512455031

White-spot Lesions and Gingivitis Microbiotas in Orthodontic Patients

ACR Tanner 1,3,*, AL Sonis 4, P Lif Holgerson 6, JR Starr 2,5,7, Y Nunez 8, CA Kressirer 1,3, BJ Paster 1,3, I Johansson 6
PMCID: PMC3420397  PMID: 22837552

Abstract

White-spot lesions (WSL) associated with orthodontic appliances are a cosmetic problem and increase risk for cavities. We characterized the microbiota of WSL, accounting for confounding due to gingivitis. Participants were 60 children with fixed appliances, aged between 10 and 19 yrs, half with WSL. Plaque samples were assayed by a 16S rRNA-based microarray (HOMIM) and by PCR. Mean gingival index was positively associated with WSL (p = 0.018). Taxa associated with WSL by microarray included Granulicatella elegans (p = 0.01), Veillonellaceae sp. HOT 155 (p < 0.01), and Bifidobacterium Cluster 1 (p = 0.11), and by qPCR, Streptococcus mutans (p = 0.008) and Scardovia wiggsiae (p = 0.04) Taxa associated with gingivitis by microarray included: Gemella sanguinis (p = 0.002), Actinomyces sp. HOT 448 (p = 0.003), Prevotella cluster IV (p = 0.021), and Streptococcus sp. HOT 071/070 (p = 0.023); and levels of S. mutans (p = 0.02) and Bifidobacteriaceae (p = 0.012) by qPCR. Species’ associations with WSL were minimally changed with adjustment for gingivitis level. Partial least-squares discriminant analysis yielded good discrimination between children with and those without WSL. Granulicatella, Veillonellaceae and Bifidobacteriaceae, in addition to S. mutans and S. wiggsiae, were associated with the presence of WSL in adolescents undergoing orthodontic treatment. Many taxa showed a stronger association with gingivitis than with WSL.

Keywords: orthodontic, adolescents, white-spot lesions, microbial ecology, Scardovia wiggsiae, HOMIM

Introduction

Orthodontic treatment with fixed appliances increases the risk of development of decalcified white-spot lesions (WSL; Lovrov et al., 2007; van der Veen et al., 2010), which can occur within 6 mos of appliance placement (Tufekci et al., 2011) (Fig. 1). While the majority of WSL re-mineralize when appliances are removed, WSL can progress to cavitation (van der Veen et al., 2010) and are a cosmetic problem (Maxfield et al., 2012). WSL result from increased plaque accumulation due to inadequate oral hygiene around orthodontic appliances (Chapman et al., 2010), which also leads to gingivitis (Naranjo et al., 2006; Rego et al., 2010).

Figure 1.

Figure 1.

Open in a new tab

White-spot lesions at gingival margins of lower left lateral incisors between fixed orthodontic bracket and gingival margin. Marked gingival inflammation, edema, and redness can be seen at the gingival margins of central and lateral lower incisors.

Evaluation of caries-associated bacteria in orthodontic patients has focused principally on Streptococcus mutans and lactobacilli (Boyar et al., 1989; Ahn et al., 2007). Plaque and WSL development under orthodontic bands observed by a multispecies microarray was associated with a diverse microbiota (Torlakovic et al., 2012), including Scardovia wiggsiae, a newly recognized species (Downes et al., 2011) associated with severe early childhood caries (ECC; Tanner et al., 2011a,b). Prevotella, Capnocytophaga, Selenomonas, and Fusobacterium species have also been detected in childhood caries (Aas et al., 2008; Gross et al., 2010; Tanner et al., 2011b). These latter taxa are more frequently associated with gingivitis, and their role in dental caries is unclear.

The goal of this study was to evaluate the microbiota of WSL in orthodontic patients as a model of initial caries development in children. We assessed the relationship between WSL and gingivitis and whether they share microbial risk factors.

Materials & Methods

Clinical Methods

Children aged between 10 and 19 yrs old, with stainless steel brackets bonded with non-fluoride-releasing adhesive (Dentsply GAC, Bohemia, NY, USA), were recruited from the dental department of Children’s Hospital, Boston. Children were medically healthy and had not used antibiotics within the preceding 3 mos. The parent or guardian provided informed consent, and the children with orthodontia agreed to participate. Participants resided in water-fluoridated areas (1 ppm) and reported using fluoridated toothpastes. They were sequentially approached until 30 children with WSL and 30 without WSL had been recruited. The Institutional Review Boards of Children’s Hospital, Harvard University, and The Forsyth Institute approved the study design, protocol, and informed consent. The clinical phase of the study was conducted between July 2009 and January 2010.

The numbers of decayed, missing, and filled teeth (DMFT) when appliances were placed were recorded. At a 6- to 12-month follow-up appointment, bracketed teeth were cleaned and examined for WSL adjacent to the bonded brackets by direct visualization with 2X magnification (dental loops), and from intra-oral photographs. Gingival and plaque indices were measured with a 1 to 4 score by quadrant. Each child completed a short survey. WSL were usually sampled from buccal anterior tooth surfaces, and from a matched site in children without WSL. One plaque sample was taken from each child by means of sterile toothpicks, and DNA from samples was purified with MasterPure kits (Epicentre Biotechnologies, Madison, WI, USA; Kanasi et al., 2010b).

Microbiological Methods

Microarray Analysis

Samples were analyzed by microarray to approximately 300 bacterial taxa with the HOMIM assay (http://mim.forsyth.org) as described previously (Tanner et al., 2011a; Torlakovic et al., 2012). Probes to 100 taxa included in the current analyses were selected based on reactivity to at least one, but not to all samples.

Bacterial-specific PCR Analysis

Quantitative-PCR was performed to detect S. mutans (Psoter et al., 2011) and Bifidobacteriaceae (Matsuki et al., 2004). We developed qPCR for S. wiggsiae using primers designed using Unipro UGENE: forward, 5′-TGCGTGAAGCCCAGGACG TA -3′, and reverse, 5′-TGTGTGGTGTGGTGAGTGGACTTT AT-3′. The qPCR standard curve reaction mixture, total volume 5 µ0L, consisted of Roche SYBR Green master mix 2X (2.5 µL), 20 µM of each primer (0.4 µL), PCR-grade water (0.1 µL), and S. wiggsiae FO 424 genomic DNA from 5 ng/µL to 5 fg/µL (2 µL). The qPCR conditions were an initial denaturation of 95oC for 15 min, followed by 45 cycles at 95oC for 15 sec, 67oC for 30 sec, and 72oC for 40 sec, and generated a 766-bp amplicon. S. wiggsiae was 10-fold serially diluted, plated, and cultured anaerobically, and colonies were counted after 5 days. Aliquots of dilutions were assayed by qPCR for DNA quantitation.

We log10-transformed the resulting DNA levels after adding 0.001 to Bifidobacterium and S. wiggsiae counts and 0.01 to the S. mutans counts. To derive qualitative qPCR results, we denoted the presence of taxa as S. mutans > 0.55. S. wiggsiae > 0.55, and Bifidobacterium > 7.8.

Personnel performing microbiology assays were not aware of the clinical status of samples.

Statistical Analyses

WSL and Bacteria

We used logistic regression to estimate associations between the presence of WSL and the presence of individual bacterial taxa (microarray or PCR) or levels of specific taxa (qPCR). If the statistical analyses involved any group with < 5 children, we used exact logistic regression. For all odds ratio (OR) estimates, we calculated 95% confidence intervals and p-values. Because the study was exploratory and potentially underpowered, we did not adjust these for multiple comparisons and do not interpret them as dichotomous significance tests (Thomas et al., 1985; Rothman, 1990). For hypothesis generation, we report taxa that exhibited strong or moderately strong associations, with p < 0.2. To explore the possibility of confounding by bacterial associations with gingivitis, we refit all models including each child’s mean gingivitis score as a covariate. We repeated the analysis after excluding individuals with S. mutans DNA< 5 (n = 5 and 8 for those with and without WSL, respectively).

Gingivitis and Bacteria

We fit logistic or exact logistic regression models to estimate the associations between gingivitis level (high, ≥ 2; or low, < 2) and the presence of each taxon. Secondarily, we refit the models adjusting for the presence or absence of WSL.

The statistical analyses were performed with Stata software version 10.1 (StataCorp. 2007, Stata Statistical Software: Release 10. StataCorp LP, College Station, TX, USA).

Partial Least-squares Discriminant Analysis (OPLS-DA)

Multivariate partial least-squares discriminant analysis (OPLS-DA) was performed (SIMCA P+, version 12.0, Umetrics AB, Umeå, Sweden) as described previously (Kanasi et al., 2010b; Lif Holgerson et al., 2011). Partial least-squares (PLS), which defines the maximum separation between class members (here WSL), is suitable for data where the number of observations is smaller than the number of variables, and where the X variables co-vary. Dichotomous HOMIM and qPCR signals, plaque, gingivitis, and drinks between and with meals built the X-block, and the presence of WSL the Y-block (outcome). The criteria for including a HOMIM variable into the x-block were that the detection prevalence differed by ≥ 14% or that the odds ratio to have WSL if having a species was > 2 or < 0.2, or that it differed between groups at p < 0.05. All variables were autoscaled to unit variance, and cross-validated prediction of Y was calculated.

Results

The study population was comprised of 28% Hispanic, 34% White, 17% Asian, and 25% Black children, and most were born in the USA. Cases were, on average, over a year older than controls, with higher proportions of girls and Hispanic children (Appendix Table 1). Children with WSL had a higher mean gingival index, but did not differ in DMFT or plaque index, compared with children without WSL. Most children reported flossing. There were minimal differences in the diet (Appendix Table 1). Children reported averages of 1.6 mealtime and 1.8 between-meal beverages, and 1.3 nighttime snacks. In comparisons of mealtime with between-meal beverages, frequencies were, respectively, 20% and 7% milk, 30% and 37% juice, and 7% and 12% diet sodas, with no differences in water or sugar-containing sodas.

HOMIM Microarray Data

White-spot Lesions

Granulicatella elegans (OR 12.0, p = 0.01) and Veillonellaceae species HOT 155 (OR 4.6, p = 0.01) were detected more frequently in samples from WSL than in those without WSL (Fig. 2A, Appendix Table 2). Other taxa exhibiting moderately strong associations with the presence of WSL (p < 0.2) included Bifidobacterium Cluster I (OR = 5.8), Selenomonas sputigena (OR 2.3), Prevotella Cluster IV (OR 2.3), Streptococcus sp. HOT 071/070 (OR 2.5), Prevotella melaninogenica (OR 2.3), and S. wiggsiae (OR 5.7). Cardiobacterium hominis was associated with controls (OR = 0.3, p = 0.02). The magnitude of ORs and p values attenuated only slightly after adjustment for gingivitis levels or exclusion of children with S. mutans DNA > 5 (data not shown).

Figure 2.

Figure 2.

Open in a new tab

Microbiota from HOMIM microarray analysis. (A) Microbiota associated with white-spot lesions (WSL) ordered according to detection in children with or without WSL. Granulicatella elegans and Veillonellaceae species were detected more frequently in samples from WSL than in those without WSL, whereas Cardiobacterium hominis was detected more frequently from children without WSL. The exact p values, odds ratios, and 95% confidence intervals are in Appendix Table 2. (B) Microbiota associated with gingivitis ordered according to detection in children with or without WSL, as in (A). In these children, more taxa were associated with higher gingivitis than with WSL, including Actinomyces sp. HOT 448 and Gemella sanguinis. The exact p values, odds ratios, and 95% confidence intervals are in Appendix Table 3.

The Lactobacillus cluster I (Lactobacillus casei, Lactobacillus paracasei, Lactobacillus rhamnosus) was detected in a low proportion of children, whereas Actinomyces cluster I (Actinomyces meyeri, Actinomyces odontolyticus, Actinomyces oricola, Actinomyces naeslundii II) and Actinomyces gerensceriae were detected in over 65% children, but their detection and other Actinomyces did not differ between children with and those without WSL (Appendix Table 2). Actinomyces sp. HOT 448 was detected in 30% of children with WSL compared with 17% of those without WSL (OR 2.1). S. mutans and Streptococcus sobrinus were detected infrequently and were not WSL-associated in the microarray data. In children with low levels of S. mutans, as assessed by qPCR, S. wiggsiae by qPCR and P. melaninogenica and Veillonellaceae species (HOT 155) from HOMIM were associated with WSL with or without adjustment for gingivitis.

Gingivitis

Gemella sanguinis (OR 8.4, p = 0.002), Actinomyces sp. HOT 448 (OR 9.8, p = 0.003), Prevotella Cluster IV (OR 3.5, p = 0.021), Streptococcus sp. HOT 071/070 (OR 5.3, p = 0.023), Streptococcus parasanguinis (OR 3.2, p = 0.035), Leptotrichia sp. HOT 417/462 (OR 9.2, p = 0.041), and Gemella haemolysans (OR 3.0, p = 0.042) were associated with gingivitis (Fig. 2B, Appendix Table 3). Other taxa exhibiting moderate associations with more gingivitis (p < 0.2) included: Dialister invisus (OR 2.9), Megasphera micronuciformis (OR 2.8), Veillonellaceae sp. HOT 155 (OR 2.7), Selenomonas sputigena (OR 2.6), Selenomonas infelix (OR 2.8), Actinomyces Cluster I (OR 2.6), Prevotella pallens (OR 4.5), S. wiggsiae (OR 6.1), Leptotrichia hofstadii (OR 6.1), and Slackia exigua (OR 2.3). Taxa in reduced gingivitis included Fusobacterium periodonticum (OR 0.04) and Neisseria elongata (OR 0.04). The magnitude of ORs and p values changed only slightly after adjustment for the presence of WSL. This adjustment strengthened the magnitude of ORs more often than it attenuated them.

PCR Data

qPCR detection of S. wiggsiae ranged from > 102 to 107fg DNA (see Appendix Fig. for amplification and standard quantification plots). A 0.1-pg quantity of DNA was equivalent to 37.5 CFU. By qPCR, S. mutans and S. wiggsiae were associated with WSL in species levels, and with S. mutans in detection frequency (Fig. 3, Appendix Table 2). S. mutans and Bifidobacterium were associated with higher gingivitis in levels and detection frequency (Fig. 3, Appendix Table 3). S. wiggsiae was associated with higher gingivitis only in detection frequency (Fig. 3, Appendix Tables 2 and 3).

Figure 3.

Figure 3.

Open in a new tab

S. mutans, S. wiggsiae, and Bifidobacterium detected by qPCR. (A) Mean S. mutans, S. wiggsiae, and Bifidobacterium levels comparing children with and without white-spot lesions (WSL). Species levels are in pg of DNA/1 µL sample, total sample size 100 µL. Mean levels of S. mutans and S. wiggsiae were higher in children with WSL than in those without WSL. The odds ratios and 95% confidence intervals, and adjustment for gingivitis, are in Appendix Table 2. (B) Mean S. mutans, S. wiggsiae, and Bifidobacterium levels (by qPCR) comparing children with and without elevated gingivitis. Species levels are in pg of DNA/1 µL sample, total sample size 100 µL. S. mutans and Bifidobacterium were detected at higher levels in increased gingivitis. The odds ratios and 95% confidence intervals and adjustment for WSLs are in Appendix Table 3. (C) Detection frequencies of S. mutans, S. wiggsiae, and Bifidobacterium in children with and without white-spot lesions (WSL). S. mutans and S. wiggsiae, at a modest level, were detected more frequently in samples from children with WSL than in those without WSL at a comparable threshold of detection (0.55 pg DNA). Bifidobacterium were not associated with WSL at a detection threshold >7.8 ng DNA, reflecting the higher levels at which Bifidobacterium were detected in samples (Fig. 3A). The odds ratios and 95% confidence intervals and adjustment for gingivitis are in Appendix Table 2. (D) Detection frequencies of S. mutans, S. wiggsiae, and Bifidobacterium in children with and without elevated gingivitis.S. mutans, S. wiggsiae, and Bifidobacterium were detected more frequently in children with higher levels of gingivitis. The odds ratios and 95% confidence intervals and adjustment for WSL are in Appendix Table 3.

PLS-DA Modeling

PLS-DA modeling identified a significant component that explained 40.7% and predicted 23.9% (R2 = 0.407 and Q2 = 0.239), although some overlap was observed between WSL and non-WSL children (Fig. 4A). Cardiobacterium hominis detection and drinking milk between meals were significantly associated with controls. Having WSL was significantly associated with the presence of Granulicatella elegans, Veillonellaceae sp. HOT 155, S. wiggsiae (qPCR), S. mutans (qPCR), S. sputigena and Prevotella Cluster IV, sugar drinks with meals, and gingival and plaque scores (Fig. 4B).

Figure 4.

Figure 4.

Open in a new tab

Partial least-squares (PLS) modeling plots. (A) PLS score plot of children with and without WSL. WSL children fell mainly toward the upper right, and non-WSL children mainly toward the lower left, with some overlap in the middle. The PLS model used WSL as the dependent variable, and the microarray, PCR, dietary, clinical, and demographic information as the independent matrix. Cross-validation was done by a systematic prediction of one 7th of the data by the remaining 6/7th of the data. R2- and Q2-values give the capacity of the X-block to explain (R2) and predict (Q2) the outcome. (B) PLS column loading plot of children with and without WSL. Bars show mean PLS correlation coefficients with measurement error (and error bars representing 95% confidence interval) for children with and without WSL. Variables with highest correlation coefficients in each group are displayed. The variables with the strongest influence in the model were C. hominis and drinking milk between meals in non-WSL children, and G. elegans, Veillonellaceae [GI] sp. HOT 155, drinking sugar-containing drinks with meals, S. wiggsiae (qPCR), S. mutans (qPCR), and higher levels of gingivitis in the WSL children.

Discussion

In this pilot study, children with WSL associated with fixed orthodontic appliances showed few differences in demographics and diet compared with control children without WSL. Microbiology findings confirmed the diversity of the microbiota of dental plaque and the relationship of S. mutans with fixed orthodontia previously observed in longitudinal studies of eleven (Boyar et al., 1989) and eight (Torlakovic et al., 2012) children. Our novel findings included the association of Scardovia wiggsiae with WSL in the presence and absence of S. mutans, and after adjustment for the presence of gingivitis, extending the association of caries with S. wiggsiae from severe ECC (Tanner et al., 2011b).

The Human Oral Microbial Identification Microarray (HOMIM) facilitates detection of multiple taxa in plaque samples, but we observed few microbial differences between WSL- and non-WSL-associated plaques. Granulicatella elegans was associated with severe ECC by clonal (Kanasi et al., 2010a), but not cultural or microarray, analysis of severe ECC (Tanner et al., 2011a). G. elegans, Veillonellaceae sp. HOT 155, and Bifidobacterium Cluster I were not associated with developing WSL according to HOMIM (Torlakovic et al., 2012), suggesting that association with WSL of these taxa requires further study. The increases in lactobacilli and actinomyces by culture (Boyar et al., 1989; Kupietzky et al., 2005), but not in the current study, may reflect differences between the microbiological assays. Slackia exigua was previously associated with severe ECC (Tanner et al., 2011a), but in the current study, S. exigua was associated with gingivitis, consistent with its detection in periodontitis (Abiko et al., 2010).

The microbiota of gingivitis with fixed orthodontic appliances has been described (Naranjo et al., 2006; Rego et al., 2010), but not with adjustment for WSL. We observed a strong association between gingivitis and detection of WSL, as we had observed with severe ECC (Tanner et al., 2011b). Gemella sanguinis, Actinomyces sp. HOT 448, Prevotella cluster IV, Streptococcus parasanguinis, Leptotrichia sp. HOT 417/462, and Gemella haemolysans showed a stronger association with gingivitis than with WSL, suggesting that the primary disease association is with gingivitis. Other species we detected in association with gingivitis, including Prevotella, Selenomonas, Actinomyces, Dialister, Slackia, and Parvimonas species, have been reported in association with childhood caries (Aas et al., 2008; Gross et al., 2010), associations that could reflect unmeasured confounding by gingivitis.

The multivariate PLS-DA modeling correctly identified most, but not all, children with WSL. The overlap may reflect similarities in the plaque and caries histories of children with and those without WSL. It may also reflect that WSL represents an early and reversible stage of caries, with only a subset of sites likely to progress if left untreated. This modeling approach is of particular value when there are more variables than study participants, and species that might be interdependent based on shared environmental requirements. Validity of the resulting model for classifying study participants will need testing in other samples.

Although the sample size was not unusually small, it could not take into account multiple hypothesis testing. Correction for multiple comparisons would have reduced statistical power to a low level; none of the findings met the ‘false discovery rate’ threshold (Benjamini and Hochberg, 1995). Applying significance thresholds is one way of evaluating results but is not always useful in non-randomized studies (Rothman, 1990, 1998). The magnitude of the differences and their confidence intervals can still be interpreted in a preliminary fashion.

A strength of this study was the use of a microarray to detect multiple species and taxon-specific qPCR to improve sensitivity and quantitation of high-interest species. Including analyses for both WSL and gingivitis clarified clinical associations for some species. Multivariate modeling allowed for visualization of the differences between clinical groups. Limitations of this study include the cross-sectional design, so we could not determine whether the microbiota changed with or could predict development of WSL. Further, the use of a defined probe set may not have adequately detected key species in WSL ecology.

We conclude that there is a complex microbiota associated with WSL that can include S. wiggsiae and possibly G. elegans, Veillonellaceae, and Bifidobacteriaceae, in addition to S. mutans. Of clinical significance, procedures to prevent the development of WSL should include testing more bacterial taxa than mutans streptococci and Lactobacillus species, and consider the impact of treatment regimens on the diverse microbiota.

Acknowledgments

We thank Winston Kuo and Alex Trachtenberg for assistance with qPCR assays, and Ralph Kent and Natalia Chalmers in project development.

Footnotes

This work was conducted with support from USPHS grants DE-015847, DE-021796, and DE-007327 from the NIDCR (NIH), Throne-Holst’s Foundation, the Bingham Trust, and Harvard Catalyst UL1 RR 025758, RC1 DE020549.

The authors declare no potential conflicts of interest with respect to the authorship and/or publication of this article.

A supplemental appendix to this article is published electronically only at http://jdr.sagepub.com/supplemental.

References

  1. Aas JA, Griffen AL, Dardis SR, Lee AM, Olsen I, Dewhirst FE, et al. (2008). Bacteria of dental caries in primary and permanent teeth in children and young adults. J Clin Microbiol 46:1407-1417 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Abiko Y, Sato T, Mayanagi G, Takahashi N. (2010). Profiling of subgingival plaque biofilm microflora from periodontally healthy subjects and from subjects with periodontitis using quantitative real-time PCR. J Periodontal Res 45:389-395 [DOI] [PubMed] [Google Scholar]
  3. Ahn SJ, Lim BS, Lee SJ. (2007). Prevalence of cariogenic streptococci on incisor brackets detected by polymerase chain reaction. Am J Orthod Dentofacial Orthop 131:736-741 [DOI] [PubMed] [Google Scholar]
  4. Benjamini Y, Hochberg Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 57:289-300 [Google Scholar]
  5. Boyar RM, Thylstrup A, Holmen L, Bowden GH. (1989). The microflora associated with the development of initial enamel decalcification below orthodontic bands in vivo in children living in a fluoridated-water area. J Dent Res 68:1734-1738 [DOI] [PubMed] [Google Scholar]
  6. Chapman JA, Roberts WE, Eckert GJ, Kula KS, Gonzalez-Cabezas C. (2010). Risk factors for incidence and severity of white spot lesions during treatment with fixed orthodontic appliances. Am J Orthod Dentofacial Orthop 138:188-194 [DOI] [PubMed] [Google Scholar]
  7. Downes J, Mantzourani M, Beighton D, Hooper S, Wilson MJ, Nicholson A, et al. (2011). Scardovia wiggsiae sp. nov., isolated from the human oral cavity and clinical material, and emended descriptions of the genus Scardovia and Scardovia inopinata. Int J Syst Evol Microbiol 61(Pt 1):25-29 [DOI] [PubMed] [Google Scholar]
  8. Gross EL, Leys EJ, Gasparovich SR, Firestone ND, Schwartzbaum JA, Janies DA, et al. (2010). Bacterial 16S sequence analysis of severe caries in young permanent teeth. J Clin Microbiol 48:4121-4128 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Kanasi E, Dewhirst FE, Chalmers NI, Kent R, Jr, Moore A, Hughes CV, et al. (2010a). Clonal analysis of the microbiota of severe early childhood caries. Caries Res 44:485-497 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kanasi E, Johansson I, Lu SC, Kressin NR, Nunn ME, Kent R, Jr, et al. (2010b). Microbial risk markers for childhood caries in pediatricians’ offices. J Dent Res 89:378-383 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kupietzky A, Majumdar AK, Shey Z, Binder R, Matheson PB. (2005). Colony forming unit levels of salivary Lactobacilli and Streptococcus mutans in orthodontic patients. J Clin Pediatr Dent 30:51-53 [PubMed] [Google Scholar]
  12. Lif Holgerson P, Harnevik L, Hernell O, Tanner AC, Johansson I. (2011). Mode of birth delivery affects oral microbiota in infants. J Dent Res 90:1183-1188 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Lovrov S, Hertrich K, Hirschfelder U. (2007). Enamel demineralization during fixed orthodontic treatment – incidence and correlation to various oral-hygiene parameters. J Orofac Orthop 68:353-363 [DOI] [PubMed] [Google Scholar]
  14. Matsuki T, Watanabe K, Fujimoto J, Kado Y, Takada T, Matsumoto K, et al. (2004). Quantitative PCR with 16S rRNA-gene-targeted species- specific primers for analysis of human intestinal bifidobacteria. Appl Environ Microbiol 70:167-173 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Maxfield BJ, Hamdan AM, Tufekci E, Shroff B, Best AM, Lindauer SJ. (2012). Development of white spot lesions during orthodontic treatment: perceptions of patients, parents, orthodontists, and general dentists. Am J Orthod Dentofacial Orthop 141:337-344 [DOI] [PubMed] [Google Scholar]
  16. Naranjo AA, Trivino ML, Jaramillo A, Betancourth M, Botero JE. (2006). Changes in the subgingival microbiota and periodontal parameters before and 3 months after bracket placement. Am J Orthod Dentofacial Orthop 130:275.e17-e22 [DOI] [PubMed] [Google Scholar]
  17. Psoter WJ, Ge Y, Russell SL, Chen Z, Katz RV, Jean-Charles G, et al. (2011). PCR detection of Streptococcus mutans and Aggregatibacter actinomycetemcomitans in dental plaque samples from Haitian adolescents. Clin Oral Investig 15:461-469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Rego RO, Oliveira CA, dos Santos-Pinto A, Jordan SF, Zambon JJ, Cirelli JA, et al. (2010). Clinical and microbiological studies of children and adolescents receiving orthodontic treatment. Am J Dent 23:317-323 [PubMed] [Google Scholar]
  19. Rothman KJ. (1990). No adjustments are needed for multiple comparisons. Epidemiology 1:43-46 [PubMed] [Google Scholar]
  20. Rothman KJ. (1998). Writing for epidemiology. Epidemiology 9:333-337 [DOI] [PubMed] [Google Scholar]
  21. Tanner AC, Kent RL, Jr, Lif Holgerson P, Hughes CV, Loo CY, Kanasi E, et al. (2011a). Microbiota of severe early childhood caries before and after therapy. J Dent Res 90:1298-1305 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Tanner AC, Mathney JM, Kent RL, Chalmers NI, Hughes CV, Loo CY, et al. (2011b). Cultivable anaerobic microbiota of severe early childhood caries. J Clin Microbiol 49:1464-1474 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Thomas DC, Siemiatycki J, Dewar R, Robins J, Goldberg M, Armstrong BG. (1985). The problem of multiple inference in studies designed to generate hypotheses. Am J Epidemiol 122:1080-1095 [DOI] [PubMed] [Google Scholar]
  24. Torlakovic L, Klepac-Ceraj V, Ogaard B, Cotton SL, Paster BJ, Olsen I. (2012). Microbial community succession on developing lesions on human enamel. J Oral Microbiol 4. 10.3402/jom.v4i0.16125 10.3402/jom.v4i0.16125 Epub 2012 Mar 14. (in press) [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Tufekci E, Dixon JS, Gunsolley JC, Lindauer SJ. (2011). Prevalence of white spot lesions during orthodontic treatment with fixed appliances. Angle Orthod 81:206-210 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. van der Veen MH, Attin R, Schwestka-Polly R, Wiechmann D. (2010). Caries outcomes after orthodontic treatment with fixed appliances: do lingual brackets make a difference? Eur J Oral Sci 118:298-303 [DOI] [PubMed] [Google Scholar]

Articles from Journal of Dental Research are provided here courtesy of International and American Associations for Dental Research

RESOURCES