Orthodontic treatment interferes with oral hygiene and promotes plaque retention, which leads to gingival inflammation and enamel demineralization. Although removable clear aligners (CAs) are designed to improve oral hygiene compared with fixed appliances (FAs), comprehensive studies comparing their respective effects on the oral microbiome are limited. This longitudinal study investigated the microbial changes during orthodontic treatment with FA and CA in correlation with clinical parameters.
Clinical parameters and supragingival plaque were collected from 12 study participants for the FA or CA treatment groups at baseline and at least twice at the 1, 3, 6, and 12-month follow-up appointments. The plaque was also harvested from the aligner tray for the CA group. Microbiome composition was determined via 16S rRNA gene sequencing, compared between groups, and correlated with clinical parameters.
Plaque (PI) and gingival indexes (GI) increased significantly in the FA but not the CA group. Beta but not alpha diversities of the microbial communities were distinct between the 2 treatment groups, even though genus-level differences were not significant except for Leptotrichia . The CA tray harbors a unique plaque community. Elevated PI and GI in the FA group correlated with a higher abundance of disease-related genera.
Orthodontic treatments trigger appliance-related plaque community shifts from baseline, and the CA tray environment attracts distinct microbial communities. In comparison with FA, the use of CA resulted in better oral health index outcomes, which is reflected by the corresponding PI and GI-associated oral microbial communities.
Assessment of 2 different frequently employed treatment types: Fixed vs removable appliances.
At the clinical level, removable appliances resulted in better oral health indexes.
Microbiomes of patients using fixed or removable appliances are distinct with similar composition.
Clear aligner tray harbors a unique and less diverse microbial community.
Plaque and gingival indexes correlated with disease-associated microbes in fixed appliances.
Dental plaque is a complex, multilayered, tooth-attached biofilm composed of many microbial species. The accumulation of plaque on tooth surfaces can lead to inflammation of the gingival soft tissues and enamel demineralization, eventually resulting in periodontal diseases and dental caries. To date, plaque removal by oral hygiene practices is the most effective measure to avert biofilm maturation and achieve disease prevention.
Conventional orthodontic treatments, especially fixed appliances (FAs), are commonly employed to address malocclusion, interfere with routine oral hygiene habits, and promote higher bacterial retention. , The resulting increase in plaque accumulation can lead to gingival inflammation and white spot lesions. About 2 decades ago, clear aligners (CAs) were introduced as an alternative to FAs that, among other benefits, allowed to alleviate the challenge of maintaining oral hygiene for patients receiving orthodontic treatment. Several studies, which followed the patients for up to 6 months, reported that in comparison with FAs, treatment with CAs resulted in improved periodontal health, , , and reduction in plaque accumulation. , In contrast, a long-term study carried out over an 18 months time period found that no significant differences exist for the periodontal parameters between various treatment modalities. Clinical outcomes are closely interrelated with the associated microbial communities. Therefore, investigating the changes in microbiome composition after introducing orthodontic appliances is critical for understanding their impact on oral health.
Limited information is available on the perturbations introduced to the tooth-associated plaque environment by orthodontic treatments and the accompanying elevated plaque accumulation. Although some studies reported an increase in the prevalence of Porphyromonas gingivalis , Tannerella forsythia, and Fusobacterium nucleatum after using FAs, , others demonstrated a decrease in these periopathogens. A systematic review and meta-analysis found that the disturbance of the microbial community is transient and reverts to pretreatment levels after the appliances are removed. Most of these studies assessed the changes in a few selected pathogens by quantitative methods, , , which does not provide insight into changes at the community level. Recent microbiome studies investigating the effect of CAs treatment on the oral microbial community revealed a nonpathologic shift in community composition or reported nonsignificant changes in salivary microbial biodiversity. With the immense popularity of CAs over FAs, it is important to examine the oral microbiome of the 2 different treatments to better understand the clinical and microbial implications. However, only a selected few studies have focused on comparing the microbiomes between the 2 different orthodontic treatments (CAs and FAs), leaving a significant gap in our understanding of the microbial shift and the associated oral health consequences of different orthodontic appliances.
An earlier pilot study by our laboratory, including 4 patients in each group, reported that although an overall microbial shift was observed from the baseline, no significant differences were observed between the orthodontic treatments. However, the CA tray biofilm, which was analyzed for the first time, revealed a unique microbial community that was notably less diverse than the corresponding tooth plaque communities. A recent study revealed that orthodontic treatments lead to a dysbiotic community; however, no marked improvement was observed over the abundance of periopathogens by CAs compared with FAs. Apart from the studies above, detailed comparisons of the oral communities between the 2 treatment modalities are largely lacking.
Thus, the present study aimed to provide a comprehensive, longitudinal view of the microbiome for orthodontic treatment with FAs compared with CAs. In addition, the composition of the plaque that accumulates on the CA trays was also investigated. These microbial data are further examined in correlation with the corresponding clinical parameters (plaque [PI] and gingival indexes [GI]) to assess the implication of these 2 different treatment modalities on oral health.
Material and methods
This longitudinal study was approved by the University of California Los Angeles (UCLA) Institutional Review Board under IRB no. 16-001258. Subjects were screened and recruited by the investigators at the UCLA School of Dentistry Orthodontics Clinic. Twenty-four participants who were due to start treatment with FAs or CAs were recruited and grouped on the basis of the treatment modality they received (ie, FA and CA). Informed consent was obtained from each subject or, for minors, their legal guardians. The inclusion criteria required participants to be in good general health. Participants with active dental caries, periodontal disease, chronic systemic diseases, and other medical conditions or who used antibiotics within 30 days of their treatment were excluded from the study. The study visits were planned at the start of orthodontic treatment (T0), with follow-up appointments scheduled at 1 (T1), 3 (T2), 6 (T3), and 12 months (T4). Because of the large variability in the patient recall intervals and missed appointments, samples were collected from subjects at T0 and at least 2 more time points of the remaining 4 visits (T1, T2, T3, and T4).
At each visit, patients were rinsed with TRACE Disclosing Solution (Young Dental, LLC, Earth City, Miss) to visualize the extent of plaque accumulation on their teeth. Once the plaque could be visualized, plaque levels were scored for both anterior and posterior teeth with the Turesky Modified Quigley Hein Plaque Index (TQHPI).
The GI score for both anterior and posterior teeth was calculated at each time point according to the protocol developed by Loe et al, with a score ranging from 0 to 3. This GI score is based on 2 characteristic signs of gingival inflammation: swelling and bleeding. The GI does not consider quantitative changes of the periodontium (eg, pocket depths or clinical attachment loss) but focuses on qualitative changes in the periodontal soft tissue.
Supragingival plaque samples were collected from the anterior and posterior teeth at T0, T1, T2, T3, and T4. In addition, the tray plaque was collected from the inner surface of the most recently worn CAs for the CA group, along with the anterior and posterior plaque. All plaque samples were collected in 15% glycerol and stored at −80°C until further use.
DNA was extracted from the supragingival plaque using MasterPure DNA extraction and Purification Kit (Epicentre, Madison, Wis) according to the manufacturer’s instructions, with some minor modifications. DNA quality and quantity were measured using a nanodrop (Thermo Scientific, Waltham, Mass). The 16S rRNA gene was amplified according to the HOMINGS protocol ( homings.forsyth.org ) using V1-V3 primers (forward: 5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGAGAGTTTGATCMTGGCTCAG-3′ and reverse: 5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGATTACCGCGGCTGCTGG-3′) instead of the V3-V4 primers of the original protocol. Briefly, the V1-V3 region of the 16S rRNA gene was amplified using gene-specific primers with overhang adapters (Illumina, Inc, San Diego, Calif). Each 25 μL polymerase chain reaction (PCR) reaction mixture included 25ng DNA, 0.4 μM each of the forward and reverse primers and 1× of the HF Taq polymerase Master mix (NEB, Ipswich, Mass). The PCR cycles consisted of an initial denaturation step of 98°C for 3 minutes followed by 35 cycles of 98°C for 30 seconds, 55°C for 30 seconds, 72°C for 30 seconds with a final extension of 72°C for 5 minutes. Three PCR reactions were carried out for each sample and pooled together to reduce bias before cleaning with AMPure XP beads (A63881; Beckman Coulter, Irving, Tex). The amplicons were indexed using the Nextera XT Index kit (Illumina) and purified with AMPure XP beads (A63881; Beckman Coulter). After quantification with the DNA KAPPA kit (ROCHE Diagnostics, Indianapolis, Ind), equal amounts of each sample were pooled into a single library. The quality and quantity of the library were checked at the Technology Center of Genetics and Bioinformatics Core at UCLA before Miseq (2 × 300 base pairs) paired-end sequencing on the Illumina platform.
After demultiplexing and trimming the barcodes, low-quality sequences containing bases with Phred quality values <20 and sequences with >3% uncertain basepairs were removed. The 16S rRNA sequences were clustered into operational taxonomic units at a 98% similarity level using QIIME1 and taxonomically assigned by comparison to the Human Oral Microbial Database. Alpha diversity (Shannon index), beta diversity (weighted UniFrac), and principal coordinate analyses were calculated using QIIME1.
Before performing the statistical analysis, the normality of the TQHPI and GI data was assessed using Shapiro-Wilk analysis, and appropriate further analyses were selected for each index. The power of the study was calculated with the G∗Power statistical analysis program (version 188.8.131.52; Franz Faul, Christian-Albrechts-Universitat, Kiel, Germany) using the number of plaque samples. The overall calculated power for the TQHPI data was 0.892, whereas, for the GI, the power was 0.973. Statistical significance was determined using a 2-tailed t test for the TQHPI scores and the Mann-Whitney U-test for the GI scores at a level of P ≤0.05 . The random missing data that resulted from patients missing appointments were addressed by using regression imputation, according to Kang. No significant difference was found between the measured clinical data and the data sets completed with regression imputation. All statistical analyses were performed using GraphPad Prism (version 9.1.0; GraphPad Software, Inc, San Diego, Calif).
In this longitudinal study, a total of 24 subjects participated, with their age ranging from 8-56 years. Based on the type of orthodontic treatment they received, the participants were stratified into the following 2 groups: FA (n = 12) and CA (n = 12). Although the average age for the CA group (29 ± 12 years) was higher than for the FA group (22 ± 13 years), the difference was not statistically significant. Both groups had equal gender distribution with 4 males and 8 female participants.
PI measurements revealed that in the FA group, the TQHPI score steadily increased from T0 to T4 in both the anterior (T0, 1.80 ± 0.82; T4, 2.96 ± 0.78) and the posterior teeth (T0, 1.33 ± 0.68; T4, 2.36 ± 0.42) ( Fig 1 , A ). The plaque accumulation was significantly higher in the FA group in the anterior and posterior teeth at T3 and T4 than T0. In contrast, no significant changes in PI were observed in the CA group. Between groups (FA and CA), the plaque levels were significantly different at the T3 and T4 time points in the anterior teeth and at T4 in the posterior teeth ( Fig 1 , A ).
Similar to the TQHPI scores, the GI scores followed an upward trend from T0 to different time points in the FA group in the anterior and posterior regions ( Fig 1 , B ). Significant differences were observed between T0 and T3 in the anterior area and T0 vs T3 and T0 vs T4 in the posterior region. In the CA group, no significant differences in GI were observed, and the gingival score was similar to the baseline throughout the study period. However, when the same time points were compared between the 2 treatment modalities (FA and CA group), the GI showed significant differences between T3 in the anterior location and at the T3 and T4 time points in the posterior region.
Furthermore, linear regression analysis of PI and GI in the different treatment modalities revealed a significant correlation between the PI and GI in the FA group ( Fig 2 , A ). However, in the CA group, no significant correlation between these clinical parameters was observed ( Fig 2 , B ).
Next, the community profiles of FA and CA groups were determined via 16S rRNA gene amplicon sequencing. A total of 2,435,143 reads were obtained from 165 plaque samples, with an average of 14,758 sequences per sample. Microbial composition analysis revealed that the anterior and posterior plaque community profiles were similar in FA ( Fig 3 , A ) and CA ( Fig 3 , B ) groups. The following 10 major genera constituted 75% of the communities: Streptococcus , Actinomyces , Corynebacterium , Capnocytophyga , Leptotrichia , Neisseria , Rothia , Fusobacterium , Prevotella , and Veillonella .
Although the microbial communities overall were very similar between the subjects treated with either 1 of the 2 different orthodontic appliances, some genera exhibited differential relative abundance. Only the genus Leptotrichia differed significantly between the 2 treatment modalities ( Fig 3 , C ). Specifically, in the anterior plaque, a higher abundance of Leptotrichia was observed for the FA group after 3 months (T2) compared with the CA group. Similarly, in the posterior plaque, a significant increase in the levels of Leptotrichia was observed for T1, T2, and T3 time points in the FA group. Within the FA group, an increased abundance of Leptotrichia was observed for the posterior plaque at later time points, with a significant increase specifically at T2 compared with T0 and T1 time points.
Although the anterior and posterior plaque exhibited similar community profiles in the CA group, a distinct profile was observed for the tray plaque ( Fig 3 , B ). Comparing the relative abundance of genera in anterior, posterior, and tray samples revealed that the levels of Actinomyces , Corynebacterium , and Selenomons decreased significantly in the tray samples. In contrast, for Streptococcus and Granulicatella , a significant increase was observed at various time points ( Figure 4 ). Specifically, a significant decrease in the relative abundance of Actinomyces was observed at T2 and T3 time points in the tray plaque compared with anterior and posterior teeth plaque ( Fig 4 , A ). Similarly, Corynebacterium and Selenomonas showed significantly decreased relative abundance for the tray samples at T3 ( Figs 4 , B and C ). In addition, a significant decrease in the relative levels of Corynebacterium between anterior teeth and tray plaque was observed at T2 ( Fig 4 , B ).
Interestingly, Streptococcus and Granulicatella exhibited a steady increase in the tray microbial community samples from T1 to T4 ( Figs 4 , D and E ). However, a significant increase was only detected between tray samples and plaque (anterior and posterior) at T3 for Streptococcus and T2 and T3 for Granulicatella . For Streptococcus , only the levels between anterior plaque and tray samples were significant at T2.
Alpha diversity was calculated using the Shannon index for anterior and posterior teeth plaque samples at the various collection time points. The community diversity in both the anterior and posterior teeth was comparable in the FA group, and no significant differences were observed ( Fig 5 , A ). Similarly, in the CA group, the anterior and posterior teeth did not exhibit any significant differences in microbial diversity ( Fig 5 , B ). Although the tray (T) samples at T4 showed decreased diversity compared with anterior and posterior teeth at T4, the observed differences were not significant. Analysis of the alpha diversity in correlation to their respective PI and GI in the FA group exhibited increased diversity in correlation to elevated PI ( Fig 5 , C ) and GI ( Fig 5 , D ); however, these differences were not statistically significant. The CA group exhibited a similar diversity index for both PI and GI.
Community diversity was further investigated by beta diversity analysis using weighted UniFrac distance measures. Analysis of the baseline samples confirmed no difference at the community level between the subjects before starting treatment with the different appliances (FA or CA) ( Fig 6 , A ). However, a significant ( P <0.001) albeit not dramatic shift in the plaque communities in the sample collected during treatment was observed on the basis of the type of appliance used ( Fig 6 , B ). Although the communities identified in plaque from anterior and posterior teeth were similar to each other in both the FA group ( Fig 6 , C ) and the CA group ( Fig 6 , D ), a distinct community for tray samples plaque samples was indicated by their significant clustering ( P <0.001). Furthermore, the communities exhibited significant differences within the FA group according to GI ( P <0.001) and PI ( P <0.005) ( Figs 6 , E and F ). However, no such pattern was observed for the GI and PI in the CA group ( Figs 6 , G and H ).
Further analysis of the microbial taxa revealed differential relative abundances for several genera associated with PI and GI. Specifically, a higher relative abundance of the genera Campylobacter, Veillonella, and Prevotella were observed with increasing PI in both treatment groups ( Fig 7 , A ). Although relative levels of Selonomonas and Leptotrichia increased with higher PI, a linear relationship was apparent only in the FA group. In contrast, the relative abundance of Rothia , Lautropia , and Haemophilus followed an inverse relationship with PI in both the groups exhibiting a decrease with higher PI.
Interestingly, for GI, Selonomonas, Leptotrichia, Veillonella, Prevotella, and Saccharibacteria exhibited an increased relative abundance with higher GI, whereas the levels of Capnocytophaga , Haemophilus , Rothia , Cardiobacterium , and Kingella decreased with increasing GI ( Fig 7 , B ). Again, this trend was only observed for the FA group. The relative abundance of all the genera mentioned above in the CA group remained unchanged. In addition, in both groups, the proportions of Lautropia and Pseudopropionibacterium decreased with increasing GI.