The effect of orthodontic appliances on the Oral Candida colonisation: a systematic review

Objectives: To evaluate the influence of Fixed (FOA) and Removable Orthodontic Appliances (ROA) on oral Candida colonisation. Methods: A search for articles published in the English language until September 2021, was carried out using Pubmed, Scopus and Web of Knowledge databases and by applying the search terms “orthodontic” OR “orthodontics” OR “fixed appliance” OR “removable appliance” OR “bracket” OR “removable aligner” AND “Candida” OR “Candidiasis” OR “Candidosis” to identify all potentially relevant human studies. After the removal of duplicate articles and data extraction according to the PICOS scheme, the methodological quality of the included papers was assessed by applying the Swedish Council on Technology Assessment in Health Care Criteria for Grading Assessed Studies (SBU). Results: The initial search identified 533 articles, 157 of which were selected by title and abstract. After full-text reading, sixteen articles were selected. The evidence quality for all the studies was moderate. Conclusions: ROA induced a temporary increase of Candida counts from the early stage of treatment but which returned to the pre-treatment level after ROA removal. Contrasting results were reported for FOA treatment which promoted the oral colonisation of non- albicans species, although the most prevalent species was Candida albicans in both groups. This review should be interpreted with caution because of the number, quality, and heterogeneity of the included studies.


Background
Candida species is a commensal yeast that colonises the oropharyngeal region of more than 60% of healthy subjects without resulting in clinical symptoms of infection. 1 Oral candidiasis is a common opportunistic infection of the oral cavity caused by an overgrowth of the Candida species, the most common being Candida albicans. 2 The ability of Candida to become a pathogenic microorganism is determined by risk factors, including systemic diseases (diabetes or infection) and local factors (orthodontic appliances, removable dentures and poor oral hygiene). 2 The onset of candidiasis represents a serious clinical problem, especially in immune-compromised patients, 3 because the infection can spread via the vascular route or upper gastrointestinal tract and lead to severe systemic infection. 2 Due to an increase in the use of corticosteroid and immuno-suppressive therapies as well as the improved survival of certain diseases (such as AIDS), an increasing number of immune-compromised patients present for orthodontic treatment, along with their healthy peers. 3 Moreover, the number of children diagnosed with cancer is increasing and the greater efficacy of children's oncological treatment has globally increased the number of cancer survivors. 4 Therefore, the number of oncological children or adolescents seeking orthodontic treatment is also increasing 3 and attention must be paid to the possible complications that orthodontic treatment could induce in immunocompromised patients 5 as it has been shown that orthodontic appliances (fixed and removable) could promote changes in the oral microbiota. 6 A Fixed Orthodontic Appliance (FOA) is the most common treatment method used in contemporary orthodontics. 7 Its complex design based on the fixed placement of brackets and bands can affect oral hygiene practices, thereby promoting the accumulation of dental plaque and altering saliva properties and microbial counts. 8 A Removable Orthodontic Appliance (ROA) is another common device used to move or retain teeth during or after orthodontic treatment. 9 Because these acrylic appliances cover a large area of mucosa for extended periods, the prolonged wear of an ROA reduces salivary flow and pH levels, protects the microbiome from the natural flow of saliva and the mechanical removal effects of the oral musculature. 9 Published literature has shown that these variables would possibly lead to pathogenic Candida colonisation, especially if there are favourable conditions and a reduction in immune function. 3 However, the effect of orthodontic treatment by fixed and/or removable appliances on Candida colonisation has not been assessed in an evidence-based manner. The only existing systematic review of treatment-induced Candida changes assessed few databases and found a limited number of studies. 3 Therefore, the evaluation of the effects of orthodontic treatment on oral Candida status is helpful for clinicians to decide the most appropriate and individualised treatment based on the patient's clinical conditions, especially in susceptible patients who might have a high risk of local or systemic complications. 10 Therefore, the aim of this review was to evaluate if orthodontic appliances induce changes in Candida colonisation in order to answer the following questions: 1. Does an orthodontic appliance affect the number and the composition of Candida colonies in the oral cavity? 2. Are there any differences in Candida populations related to FOA and ROA?

Protocol
This systematic review was performed according to the PRISMA statement. 11

Eligibility criteria
According to Participants-Intervention-Comparison-Outcome-Study design schema (PICOS), the inclusion and exclusion criteria are summarised in Table I.

Information sources and literature search
The search for articles was carried out using four electronic databases (Pubmed, Scopus, Web of Knowledge, CENTRAL), and included publications in the English language from inception up to September 2021. Human studies which featured the keywords "orthodontic" OR "orthodontics" OR "fixed appliance" OR "removable appliance" OR "bracket" OR "removable aligner" AND "Candida" OR "Candidiasis" OR "Candidosis", were identified. In addition, the reference and citation lists of the included trials and relevant reviews were manually searched.

Study selection
All titles identified from the literature were screened and selected by two independent authors (A.C.; E.L.M.). Duplicate studies were eliminated. The abstracts were examined and full texts were obtained if additional data were needed to fulfil the eligibility criteria. Conflicts were resolved by discussion with a third author (L.L.M.).

Data collection
The characteristics of the included studies (study design, patients, age, orthodontic appliance, sample site, timing, analysis method, outcome, additional measures, quality of the study) were independently extracted by two authors (A.C.; E.L.M.). For further clarification, missing or unclear information was directly requested of the respective authors.

Methodological quality assessment
The methodological quality of the included studies was assessed according to the "Swedish Council on ORTHODONTIC APPLIANCES AND CANDIDA COLONISATION Technology Assessment in Health Care Criteria for Grading Assessed Studies" (SBU) method. 12 Articles were ranked into three levels (A, B, C) of evidence (Table II) and, based on the score assigned to each study, the review level of available evidence was further scored into four grades (1,2,3,4) (Table III). Quantitative and qualitative analysis of Candida colonies, from intra-oral mucosal sites, saliva or supra/sub-gingival plaque.

No clear mention of the analysis or time-point
All available time-points will be included and categorized into pre-treatment, short-term (< 3 months) treatment, mid-term (3-6 months) treatment and long-term (< 6 months) treatment, post-treatment Study design Randomized clinical trials or non-randomized, prospective or retrospective, cohort studies Note: Tx, treatment; ROA, removable orthodontic appliance; FOA, fixed orthodontic appliance.

Data synthesis
Due to the lack of homogeneity in the study setting (study design, sample site, sample collection time and methods), only a systematic review could be conducted rather than a meta-analysis.

Study selection
The initial search identified 533 articles from Pubmed, Scopus and Web of Knowledge. After eliminating duplicates and ineligible studies by title and abstract, a total of 157 full texts were screened. Finally, a total of sixteen papers were identified according to the eligibility criteria.
The flow chart of the selection of eligible studies for this review is summarised in Figure 1. One study with level "A" and at least two studies with level "B" 3 Limited At least two studies with level "B" 4 Inconclusive Fewer than two studies with level "B"

Assessment of methodological quality
According to the SBU tool, the quality of evidence for nine studies was moderate (grade B) and for seven studies was low (grade C). As a result, the level of evidence for the conclusions of this review was considered limited (level 3).

Study characteristics
The characteristics of the studies are presented in Table IV. Of the 16 included studies, all were prospective in nature and included four reports which described the changes in Candida in patients treated using a ROA 13,17,20,26 and nine studies which described those treated with FOA. 8,[14][15][16][21][22][23][24]27 Three studies analysed and compared treated and non-treated patients, two of which involved a ROA 17,18 and one used FOA 25 . An additional untreated group served as a control. Only one study compared the changes in candida between ROA and FOA therapies. 19

Results of individual studies
The results are summarised in Figure 1.

Short-term (<3 months) Candida changes
Two studies 20,26 described the short-term changes occurring during ROA treatment; eight studies 8,14,15,[21][22][23][24]27 analysed the effects related to FOA therapy. From baseline to one month of ROA therapy, a significant increase (p < 0.001) in the number of Candida albicans counts was observed in saliva 20 and on the oral mucosa. 26 During the early stages of FOA, Hägg et al. 14 found a significant Candida increase on the dorsum of the tongue (p < 0.001), but not in saliva and plaque samples. Arslan et al. 15 reported an increase in the number of colony-forming units (CFU) was statistically significant (p < 0.001) both in saliva and on tooth surfaces. This was confirmed by the salivary results (p < 0.001) of Arab et al. 8 and by a plaque analysis (p < 0.05) conducted by Shukla et al. 22 Zheng et al. 21 also reported a significant increase of Candida counts (p < 0.001) in gargled samples, finding a higher (p < 0.05) percentage of Candida carriers after 2 months of FOA, compared to pre-treatment. Different results were reported by Lee et al. 16 and Grzegocka et al., 23 who determined a non-significant increase of Candida, after analysing saliva samples. In addition, Soler et al. 24 found no significant differences at the vestibular level, while Kouvelis et al. 27 reported that Candida was not identified in any sample before and after 4 weeks of FOA.

Mid-term (3-6 months) Candida changes
One study 20 reported the mid-term effects of ROA treatment and found a significant increase (p < 0.001) in the Candida counts in saliva after 3 months. Seven studies 8,14,16,[21][22][23]27 analysed the mid-term effects after FOA placement. Hägg et al. 14 found a significant increase in candida on the dorsum of the tongue, but not in saliva and plaque samples. However, Lee et al. 16 reported a significant increase in the presence of candida in saliva. Zheng et al. 21 also showed that the presence of Candida was significantly higher after 3 months of FOA treatment compared to baseline, finding a significant increase (p < 0.05) of Candida counts in a gargled sample. The increase was confirmed by Arab et al. 8 and Shukla et al. 22 who analysed saliva and dental plaque, respectively. Grzegocka et al. 23 showed a non-significant upward trend of yeast numbers in saliva after 12 weeks of FOA treatment. Only one study 25 analysed the differences in candida between FOA and untreated patients suggesting that, 3-6 months after FOA placement, Candida was a frequently isolated species (12%) in orthodontic patients compared to a control group, and that, in the mid-term, the frequency of Candida significantly increased in FOA patients, compared to untreated cases.

Long-term (> = 6 months) Candida changes
One study 20 investigated the long-term Candida changes in patients treated with ROA, and observed a significant increase of Candida counts (p < 0.001) in saliva after 6 months. Two studies 17,18 compared Candida counts between ROA and control groups, and found conflicting results. Mahmoudababi et al. 17 reported the prevalence of the salivary colonisation of Candida spp. was significantly higher (p < 0.001) in ROA patients, compared to untreated subjects. Gonçalves et al. 18 observed no statistically significant differences in saliva yeast counts between the ROA and a control group. A further study 19 compared the differences between two orthodontic groups, one treated with ROA and one with FOA, through an analysis of the salivary samples of 80 subjects (40 for each group). A statistical significance (p < 0.001) was found in an increased colonisation of Candida in patients treated using FOA, compared to those treated with a ROA. 19 Four studies 15,16,21,24 analysed the alteration in candida counts in patients treated with FOA. In comparing pre-treatment and longterm values, Arslan et al. 15 found a significant increase (p < 0.001) of Candida in saliva and tooth samples, although the increase was not significant during the 6 to 12 month period. Lee et al. 16 also observed significant differences (p < 0.005) in the presence of oral Candida in the saliva of FOA patients, at longterm follow-up. Alternative results were reported by Zheng et al. 21 in which, after 6 months, the candida levels were comparable with those prior to treatment. In addition, Sanz-Orrio-Soler et al. 24 observed no statistical difference in the frequency of Candida over the long-term.

Candida changes after orthodontic appliance removal
Two studies evaluated the differences in Candida counts before and after ROA 13 and FOA 24 treatment. Arendorf et al. 13 observed a significant decrease in candida (p < 0.001) to baseline levels after ROA removal, although a transient significant increase (p < 0.001) occurred during therapy, especially on the posterior (63.6%) and anterior palate (60.6%). Sanz-Orrio-Soler et al. 23 , reported no statistically significant increase in Candida colonisation during FOA treatment. The slight increase in Candida levels from pre-treatment (T0 = 3.2%) to post-treatment (T4 = 4.8%) was not significant. Moreover, no significant differences in the presence of Candida were found between the two different analysed fixed appliances (metal or ceramic brackets). 24

Candida species changes during orthodontic treatment
Eleven studies described the changes in the frequency of the different candida strains during orthodontic treatment using a ROA [17][18][19]26 and FOA. [14][15][16]21,[23][24][25] Mahmoudabi et al. 17  (9.1%). Six yeast species (C. parapsilosis, famata, sake, glabrata, dubliniensis, S. cerevisiae, P.etchellsii) were isolated only in the ROA group. Gonçalves et al. 18 also found a higher incidence of non-albicans Candida in the ROA group (55.2%) compared to a control group (42.9%), such as C. Iusitaniae (10.3%/4.8%), C. krusei (10.3%/0), C. Tropicalis (13.3%/9.5%), C. parapsilosis (6.9%/4.8%). Rodríguez-Rentería et al. 26 noted that, after 4 weeks of ROA treatment, C. albicans and C. glabrata were the most prevalent species. Hägg et al. 14 found that the predominant Candida species isolated during the first stages of FOA treatment was C. albicans (83-87%), while C. parapsilosis, C. tropicalis and C. guillermondii were less common. Arslan et al. 15 found that the 58.5% (42 of the 72 patients) of an initial FOA group were Candida carriers and the most common species identified was Candida albicans (73.8%), followed by C.tropicalis, C.krusei and C. kefyr (7.14%) and by C. parapsilosis (4.76%). No long-term qualitative evaluation was carried out. Lee et al. 16 observed that C. albicans was the most isolated species, while the non-albicans species identified were: C. tropicalis (4 isolates), C. parapsilosis (2 isolates), S. cerevisiae (2 isolates), C. globosa (1 isolate). Zheng et al. 21 evaluated the Candida strains in a long-term followup, and determined that the presence of C. albicans was 85.7% of that at T0, which subsequently further decreased during treatment in favour of an increase in other strains, specifically C. parapsilosis, C. krusei and C. tropicalis. Grzegocka et al. 23 identified that 58.8% of subjects were Candida-carriers (two were colonised after bracket placement), with a predominant colonisation of C. albicans (91.1%), followed by C. tropicalis (4.5%) and C. guillermondii (4.5%). Soler et al. 24 reported that the most isolated candida strains in FOA patients were C. albicans, while C. glabrata and C. krusei were each found in one patient out of 124, respectively. Pellisari et al. 25 observed that, in patients treated with FOA, the isolated fungal strains were C. albicans and C. krusei, compared to untreated subjects. Khanpayeh et al. 19 noted a higher frequency of salivary Candida carriers (p = 0.0001) and a higher colonisation of non-albicans Candida species (p = 0.001) in a FOA group compared to a ROA sample (p = 0.0001). The negative saliva culture was 22.5% in ROA patients but only 5% in FOA patients. The most frequent species in the ROA group was C. albicans (62.5%), while in the FOA group, the frequency of C. albicans was lower (45%).

Discussion
Candida yeasts are able to form a biofilm on abiotic surfaces, such as the brackets of FOA or the acrylic surfaces of ROA, leading to an increased oral Candida presence to produce pathogenic oral mycoses, especially in immunodeficient patients. 23 A recent review 28 revealed a strong relationship between orthodontic treatment and the oral colonisation of Candida species. density on the dorsum of the tongue was found when an imprint culture was used, although the overall prevalence rates of candida obtained using oral saliva and pooled plaque techniques did not demonstrate a change. In contrast, Lee et al. 16 found a significant increase in Candida counts in oral saliva after 5 months of FOA. In a mid-term follow-up, a statistical increase in Candida counts was found in patients with FOA in gargled samples 21 and dental plaque, 22,25 although Grzegocka et al. 23 reported a non-significant upward trend. Limited studies reported the long-term effect of FOA on Candida density in contrast with two studies which reported a significant increase 15,16 and two other studies reporting non-significant changes. 15,16,21,24 Only one study by Soler et al. 24 investigated the candida effects after FOA removal, finding that FOA (both metal and ceramic appliances) did not influence the presence of Candida albicans. Khanpayeh et al. 19 compared the salivary sample of 80 orthodontic subjects, treated with ROA and FOA by dividing subjects into two groups matched by gender and age. A higher frequency of Candida colonisation was found in the FOA group, compared to the ROA group.

Candida species and orthodontic treatment: summary of evidence
It is accepted that the most common aetiological contributor of oral candidiasis is Candida albicans, which causes 45-75% of the total incidence of candidiasis, whereas C. tropicalis, C. glabrata, C. parapsilosis, C. krusei account for about 7% of all cases. 21 The assessment of candida colonies in orthodontic patients compared to untreated controls showed that Candida albicans was the most prevalent species isolated in both groups, although the method of sampling and analysis differed. 17,18,25 The analysis of the Candida species showed that the most prevalent species in orthodontic patients was C. albicans, while other yeast species were less common during ROA 26 and FOA [14][15][16]21,23,24 treatment. Varying levels of Candida strains were reported among the analysed studies, likely due to individual variability and to the different collection methods.
Furthermore, differences in the oral yeasts of patients with or without orthodontic appliances have demonstrated a higher Candida diversity in the orthodontic group. The wearing of orthodontic appliances significantly increased the tendency for colonisation by multiple yeast species, especially nonalbicans species (as C. parapsilosis, C. famata, C. sake, C. glabrata). [14][15][16][17][18][19]21,[23][24][25][26] A higher colonisation of non-albicans Candida species was also seen in FOA subjects compared with ROA subjects (P = 0.001). 19 The most frequent species in the ROA group was C. albicans (62.5%), while in the FOA group, the frequency of C. albicans was lower (45%) and the frequencies of other species (C. tropicalis, C. parapsilosis, C. Krusei, C. Kefyr) was higher than in the ROA group. Therefore, FOA seemed to promote an increase in the presence of salivary Candida, particularly non-albicans Candida species compared to ROA patients. 19 All of these yeast species have a great ability to form biofilms 23,28 in patients undergoing orthodontic therapy, 6 mainly in FOA cases. 19,28 Candida strains aggregate or adhere more easily to orthodontic fixed appliances. 16 The increase in Candida species other than C. albicans in FOA patients, may be due to the different environmental conditions of non-albicans Candida strains. After FOA placement, the pH of plaque, the strains and number of micro-organisms in the oral cavity are altered, 8,14,22,23,27 which allows non-albicans strains to proliferate and adhere more easily to FOA. 16 Moreover, the increased risk of Candida colonisation in orthodontic patients could be attributed to a varying degree of gingival inflammation and mucosal damage that is often seen during orthodontic therapy, and which could have decreased the local defense mechanisms. 20,21 Recent literature 28 reported that other local factors, such as mucosal barriers, contributed to the formation of Candida colonies. The first line of defense against the Candida species is an intact mucosa, 28 and therefore, there will be an increased risk of infection if there are oral lesions due to local trauma associated with orthodontic appliances. 3,24 It is important to consider that the presence of oral appliances does not appear to increase the clinical signs of candida in individuals who are healthy carriers. 31 However, Goncalves et al. 18 and Zheng et al. 21 , respectively, reported that the presence of microtrauma of the oral mucosa in orthodontic patients, did not produce candidiasis in the studied patients, despite Candida colonisation.
This situation may be explained by the opportunistic pathogenic character of these micro-organisms, that may cause infection in cases of immuno-suppression. Therefore, clinicians should be cautious when providing orthodontic treatment in immuno-compromised ORTHODONTIC APPLIANCES AND CANDIDA COLONISATION children because of an increased risk of candida infection. This is especially valid during FOA treatment because traumatic mucositis often occurs to the oral mucosa due to FOA irritation throughout treatment. 15 Additional host-dependent variables, such as sialometric variations, 8,13 immuno-deficiency, a diet rich in sugar and deficient oral hygiene, 14,23 should also be considered as contributors to the formation of a Candida spp. biofilm. 28 The oral prevention, correct hygiene habits and a greater awareness of children under orthodontic treatment and their parents, not only guarantees the success of treatment, but can also decrease the risk of systemic and/or local diseases, especially in immunocompromised patients. 14,19,[21][22][23][24]26 Limitations Considering the clinical heterogeneity of the reviewed studies, as well as the differences between the sample sites, the analytical methods and, in the quantitative assessment (the number composition was expressed as counts of CFU or as a percentage of frequency), the present review reflects only the changing trend in the colonisation of oral Candida during orthodontic treatment. Further high-quality randomised clinical trials are needed to increase the quality of evidence regarding the changes in the candida population during orthodontic treatment.