Longitudinal changes in gingival crevicular fluid after placement of fixed orthodontic appliances


Bacterial plaque is an etiologic factor in the development of gingival inflammation and periodontitis. The presence of orthodontic bands and brackets influences plaque growth and maturation. The purposes of this research were to monitor microbiologic and periodontal changes after placement of orthodontic attachments over a 1-year period and to link these changes to alterations in cytokine concentrations in the gingival crevicular fluid (GCF).


This longitudinal split-mouth trial included 24 patients. Supragingival and subgingival plaque composition, probing depth, bleeding on probing, and GCF flow and composition were assessed at baseline (Tb) and after 1 year (T52). A statistical comparison was made over time and between the banded and bonded sites. Prognostic factors for the clinical reaction at T52 in the GCF at Tb were determined.


Between Tb and T52, the pathogenicity of the plaque and all periodontal parameters increased significantly, but intersite differences were not seen, except for bleeding on probing. The cytokine concentrations in the GCF did not differ significantly between the sites or between Tb and T52. The interleukin-6 concentration in the GCF at Tb was a significant predictive value for the GCF flow at T52 ( P <0.05). The same relationship was found between the interleukin-8 concentration at Tb and the increase in probing depth at T52 ( P <0.05).


Interleukin-6 and interleukin-8 concentrations before orthodontic treatment were shown to be significant predictive factors for some potential inflammatory parameters during treatment.

It is well established that bacterial plaque is the primary etiologic factor in the development of gingival inflammation and periodontitis. The quantity and the quality of plaque are known to be influenced by many factors, including surface characteristics. Especially surface roughness and high surface free energies were found to be positively correlated with plaque growth and maturation. Gingival inflammation is known to further increase this. In addition to the total amount of bacteria, the ratio between the aerobic and anaerobic bacteria is also an important marker for plaque pathogenicity.

The placement of orthodontic bands and brackets influences plaque growth and maturation because of the above-mentioned factors. Significant differences in biofilm formation on bonded teeth compared with control teeth were reported.

Most studies reporting on gingival changes after bracket placement suggested only temporary reversible periodontal changes. Another study, however, reported significant attachment loss during orthodontic treatment.

Another way of studying the changes during orthodontic therapy is by analysis of the composition of the gingival crevicular fluid (GCF). GCF reflects the immune and inflammatory reactions from host-parasite interactions and biomechanical stresses. It is a noninvasive method, and, until now, many substances involved in the inflammatory process and produced by the periodontal ligament cells in sufficient amounts to diffuse into the GCF have already been studied. Giannopoulou et al compared the GCF composition of orthodontic patients with that of nonorthodontic controls. However, the wide intersubject differences in GCF composition made reasonable comparisons difficult and warrant prospective intrasubject studies. Among many inflammatory and immune mediators identified in GCF, cytokines have attracted particular attention.

Interleukin-6 (IL-6) is known to be a major modulator of inflammation in chronic local inflammatory reactions. It regulates the immune cell recruitment in the transition from the acute (recruitment of neutrophils) to the chronic (recruitment of monocytes) form of inflammation. Interleukin-8 (IL-8), produced by various cells (polymorphonuclear leukocytes, monocytes, macrophages, and fibroblasts), plays a key role in the accumulation of leukocytes at the sites of inflammation.

Our aims in this study were to monitor the microbiologic and clinical periodontal changes after placement of orthodontic attachments over a 1-year period and to link these changes to alterations in cytokine concentrations in the GCF. The null hypothesis was that the cytokine concentrations before orthodontic treatment have no predictive value for the clinical periodontal reaction during treatment. The dependant variables were change in probing depth, number of sites bleeding on probing, and GCF flow.

Material and methods

Twenty-four subjects (10 boys, 14 girls) aged 14.6 ± 1.1 years (mean ± SD) referred to the postgraduate clinic of the Department of Orthodontics of the Catholic University of Leuven in Belgium were included in the study. The subjects and their parents were given a written explanation of the background of the study, its objectives, and their involvement. After screening for suitability and after good comprehension of the protocol, the parents all gave their written informed consent. This study was approved by the ethical committee of the same university. The patients were selected if they fulfilled the following inclusion criteria: no smoking, no orthodontic treatment with extractions, no extensive dental restorations or adhesive fixed partial dentures, a sulcus bleeding index of less than 0.3, no periodontal disease, and no use of antibiotics during or up to 4 months before the study. The patients were asked whether they were right- or left-handed; in the right-handed patients, the right quadrants might be brushed more thoroughly, leading to a healthier gingival condition. All patients were right-handed. Fourteen subjects (6 boys, 8 girls) of the 24 subjects were treated with headgear (headgear group) and received bands on the maxillary first molars for 18 weeks before bonding the brackets to the remaining maxillary teeth. In the headgear group, it was possible to make intrasubject comparisons between the bonded and banded sites. The other 10 subjects were treated with brackets only (nonheadgear group).

The study had a longitudinal prospective design. During the study period, the subjects were periodontally analyzed 2 or 3 times ( Table I ). The first time, at minus 18 weeks (T–18), was to record the status praesens of the periodontium, to sample the subgingival and supragingival plaque, and to place the molar bands. At the second visit, after 18 weeks (T0), the measurements and samples were repeated, and brackets were bonded on the remaining maxillary teeth (headgear group). For the nonheadgear group, this was the first visit. At T0, the initial orthodontic archwire was also placed (0.014-in nickel-titanium alloy). T–18 was considered the baseline for the banded sites, and T0 was the baseline for the bonded sites. The orthodontic therapy was performed by using Generus full edgewise brackets with an 0.018-in slot (GAC International, Bohemia, NY). The extraction therapies were excluded; after 1 year (T52) of orthodontic treatment, all subjects were in the treatment phase with 0.016 × 0.022-in stainless steel wires. At T52, the final measurements were made. The 1.4 and 1.6 were sampled; for the headgear groups, the first molar was a banded site, and the first premolar was a bonded site. For the nonheadgear group, both teeth were bonded. Standardized oral hygiene instructions with an orthodontic toothbrush (Oral-B, Kirkland, Quebec, Canada) with the Bass technique and a single tufted brush (Oral-B) were instructed. Interdental cleaning was taught with extra-fine interdental wood sticks (Oral-B). The youngsters were able to use these wood sticks also after placement of the bands, brackets, and orthodontic wires. The patients were told to always brush their teeth for 3 minutes. The hygiene protocol was taught with a model; then the subjects’ brushing was analyzed and improved by a clinician (J.v.G.) to achieve good comprehension. At each visit, the teeth were stained with erythrosine disclosing solution (4% erythrosine in water solution) to show the patients how to remove the remaining plaque.

Table I
Flow chart of the study with interventions per treatment and group
Intervention Tb
T–18 T0 T52
Headgear group Nonheadgear group Headgear group Nonheadgear group Headgear group Nonheadgear group
Molar-band placement x
Bracket placement x x
Crevicular fluid sampling x x x x x
Probing depth measuring x x x x x
Bleeding on probing measuring x x x x x
Supragingival microbial sampling x x x x x
Subgingival microbial sampling x x x x x
Oral-hygiene instruction x x x x x
Clinical photos x x x x x

Tb , Baseline; T–18 , baseline for the headgear group; T0 , baseline for the nonheadgear group; T52 , end of the study after 52 weeks of treatment.

For band placement at T–18, only the patients from the headgear group received orthodontic bands on their maxillary first molars ( Table I ). The teeth were pumiced with a rubber cup, the orthodontic bands were fitted, and the correct size was selected. The gingival band margins were trimmed to be placed supragingivally. After disinfecting the bands with alcohol and drying them, Transbond Plus glass ionomer cement (Multi-Cure Ionomer Orthodontic Band Cement, 3M Unitek, Monrovia, Calif) was mixed according to the manufacturer’s instructions. The bands were placed, and any excess cement was removed from the occlusal and cervical margins of the bands and the teeth. All band selections and cementations were performed by the same clinician (J.v.G.). The cement was light cured (QHL75 halogen curing light, Dentsply, Addlestone, Surrey, United Kingdom) for 30 seconds from the occlusal side. The preformed headgear was adjusted, and the patients were instructed to wear it for 14 hours a day.

At T0, all patients received brackets in the maxilla ( Table I ). For the nonheadgear group, these were the first orthodontic attachments in the mouth. The headgear group received brackets on all remaining teeth in the maxilla. The teeth were pumiced with a rubber cup, and the quadrant to bond was isolated with cotton rolls and saliva suction. A 1-step adhesive (Transbond Plus Self Etching Primer, 3M Unitek) was applied with a microbrush, and the excess was blown away with dry air in the incisal-occlusal direction to prevent contact with the gingiva. The composite bonding material (Transbond Plus color change adhesive, 3M Unitek) was applied to the bracket base, the bracket was pressed firmly onto the enamel surface, and any excess adhesive was removed. Then the composite was light cured (Dentsply QHL75 halogen curing light, Dentsply) for 30 seconds from the occlusal and gingival directions. After placement of the brackets, an initial nickel-titanium orthodontic wire was placed and ligated to the brackets with elastic ligatures.

After isolating the teeth from the saliva with cotton rolls and gently drying them to prevent contamination, the supragingival plaque was carefully removed with sterile curettes without traumatizing the gingiva, because this would increase the production of crevicular fluid. The supragingival plaque was transferred into flip-capped vials containing 2.0 mL of prereduced transport medium to be processed within 2 hours.

Each sample was homogenized by vortexing for 30 seconds and coded. The coding was not revealed until all analyses were completed, leading to blinded microbiologic analyses.

The subgingival plaque was sampled after collecting the GCF without traumatizing the crevice, because this would increase the GCF flow. To sample the subgingival plaque per tooth, 6 sterile medium paper points (RoekoA, Roeko, Langenau, Germany) were inserted into the sulcus (3 mesially, 3 distally) and kept in place for at least 10 seconds. The subgingival plaque samples were transferred into flip-capped vials containing 2.0 mL of prereduced transport medium to be processed within 2 hours. The mesiobuccal and distobuccal sites of the maxillary right first premolar and first molar were sampled. For the headgear group, the first molar was a banded site, and the first premolar wasa bonded site; the samples were separately analyzed. For the nonheadgear group, both teeth were bonded, and the samples were pooled.

To minimize the effect of tooth movement on the composition of the crevicular fluid, we did not perform any orthodontic activations during the last 2 months before sampling, and intermaxillary elastics were not worn. After removing all supragingival plaque as described, the GCF was sampled. The absence of plaque is important because dental plaque has also been shown to have a marked effect on the recorded volume of crevicular fluid in the strip.

Periopaper absorbent strips (593525, Ora Flow, Amityville, NY) were placed into the sulcus until light resistance was felt. After keeping the strip in place for 30 seconds, the absorbed volume was measured with the Periotron 6000 device (Ora Flow), which was calibrated before each measurement according to the standard curve obtained with bovine serum. Strips with blood contamination were discarded. The volume measurements were performed within 5 seconds after removal of the strip from the crevice to minimize evaporation. Per site, 3 strips were used. After measuring the collected volumes, the Periopaper strips were placed in coded sterile screw-capped vials and stored at –70°C.

The mesiobuccal and distobuccal sites of the maxillary right first premolar and first molar were sampled. In the headgear group, the first molar was a banded site, and the first premolar was a bonded site; the samples were analyzed separately. For the nonheadgear group, both teeth are bonded, and the strips were pooled.

After collecting all GCF samples, the fluid was extracted from the strips by adding 50 μL of phosphate-buffered saline solution containing 0.05% Tween 20 (Fluka, Buchs, Switzerland) and storing the vials at 4°C for 12 hours. After centrifuging the vials, the fluid was collected, and again 50 μL of phosphate-buffered saline solution containing 0.05% Tween 20 was added, and the whole procedure was repeated. The fluid extracted from the strips was pooled and stored at –70°C until analysis.

A Bio-Plex human cytokine assay for simultaneous quantification of interleukin-2 (IL-2), interleukin-4 (IL-4), interleukin-6 (IL-6), interleukin-8 (IL-8), interleukin-10 (IL-10), granulocyte monocyte colony stimulating factor (GM-CSF), interferon- γ (IFN-γ), tumor necrosis factor-α (TNF-α), monocyte chemotactic protein-1 (MCP-1), and interferon-inducible protein-10 (IP-10) was run according to the procedure recommended by the manufacturer (Bio-Rad Laboratories, Hercules, Calif). For each cytokine, a standard curve (1/4 dilutions) ranging from 0.5 pg per milliliter (IL-4), 2 pg per milliliter (IL-2, IL-6, IL-8/CXCL8, IL-10, GM-CSF, IFN- γ, MCP-1/CCL2, and IP-10/CXCL10) or 8 pg per millileter (TNF- α), to 32,000 pg per millileter was prepared. The filter plate was prewet with Bio-Plex assay buffer (Bio-Rad Laboratories), and 50 μL of beads coupled to specific monoclonal antibodies against the analyzed cytokines was added per well. Subsequently, the standards and samples (12.5 μL/sample) were prepared by using the Bio-Plex serum diluent kit (Bio-Rad Laboratories), and 50 μL per tube was transferred to the test plate, followed by incubation (30 minutes) at room temperature on a shaker (300 rpm). After 3 washes with Bio-Plex wash buffer (Bio-Rad Laboratories), detection antibody was added, and the plate was again incubated (30 minutes, room temperature, 300 rpm) to allow interaction with the beads. After removal of any excess detection antibody (3 washes with Bio-Plex wash buffer), a final incubation with streptavidin phycoerythrin was performed (10 minutes, room temperature, 300 rpm). The beads were washed 3 times and resuspended in 125 μL of Bio-Plex assay buffer before analysis in the Bio-Plex suspension array system. The data were analyzed by using the Bio-Plex Manager software (version 4.1) with 5-parameter logistic regression curves.

At all visits, digital color photographs were taken to ascertain the status praesens of the periodontium and the dental plaque accumulation.

The probing depths were measured at the proximal buccal sides of the teeth with a Merrit B probe (Hu-Friedy, Chicago, Ill) and rounded off to the nearest 0.5 mm. The bleeding-on-probing tendency for each site per tooth was also registered, 20 seconds after probing the depth of the pocket (absence, 0; presence, 1). These periodontal parameters were scored at all visits, and the examiner (J.v.G.) was blinded to the previous scores.

All samples were transferred to the laboratory and processed within 2 hours. Serial 10-fold dilutions were prepared in the prereduced transport medium. Dilutions of 10:2 to 10:4 were plated in duplicate with a spiral platter (Spiral Systems, Cincinnati, Ohio) onto nonselective blood agar plates (Blood Agar Base II, Oxoid, Basingstoke, United Kingdom), supplemented with hamine (5 μg/mL), menadione (1 μg/mL), and 5% sterile horse blood.

After 7 days of anaerobic incubation (80% N 2 , 10% CO 2 , and 10% H 2 ) in an anaerobic chamber and 3 days of aerobic incubation in an aerobic incubator at 37°C, the total numbers of anaerobic and aerobic colony-forming units (CFU) were counted. From these data, the CFU ratio (CFU aerobe:CFU anaerobe) was also calculated. The number of specific dark pigmented colonies (black pigmented bacteria) on a nonselective anaerobic plate, containing approximately 100 colonies, was counted. From the black-pigmented bacteria in the plaque samples, every third colony was subcultured on a blood agar plate. After 48 hours of anaerobic incubation, the pure cultures were identified by means of a series of biochemical tests (including N-acetyl-β-D-glucosaminidase, α-glucosidase, α-galactosidase, α-fucosidase, esculine, indole, and trypsin activity) to differentiate Porphyromonas gingivalis and P intermedia from other pigmented porphyromonas and prevotella species.

Statistical analysis

To compare the groups, a linear mixed model was fit to the data, with time and treatment as the fixed factors. The data at Tb were compared with those at T52 (for the banded sites, Tb was T–18; for the bonded sites, Tb was T0). The repeated measurement aspect of the data was modeled by taking the patient factor as the random variable. Residual analysis indicated that CFU counts, crevicular fluid, and cytokine concentrations needed a log transformation before analysis to achieve normal distribution. Multiple comparisons between treatments and time groups were corrected for simultaneous hypotheses according to the general linear hypothesis model, resulting in global confidence of 95% for all comparisons per variable. For variables with values below quantification limits, a 1000-fold Monte Carlo simulation was built up, replacing values below the quantification limit with a uniform, randomly distributed value between 0 and the quantification limit. Only P values for which 95% of the simulations were below 0.05 were considered significant.

To find relationships between the cytokine concentrations at baseline and the severity of the inflammatory reaction (ie, worsening of the periodontal parameters) after 1 year of treatment, a stepwise regression was performed. The periodontal parameters (bleeding on probing, probing depth, GCF flow) after 1 year of orthodontic treatment were used as response and cytokine concentrations (IL-6, IL-8, IFN-γ, IP-10) at the start as explanatory variables. Values were entered in the model in the forward step and omitted from the model in the backward step by using a threshold P value of 0.2. The variable modeling of the difference in the models was by default used as the first variable in the model. Only variables having a P value below 0.05 were considered significant. A correlation coefficient would be difficult to interpret for the relationship between the cytokine concentrations at baseline and the severity of inflammatory reaction after 1 year of treatment because the data were not bivariately normally distributed and not independent. Instead, we used the P values of the regression to reflect the strength of the relationship.


At Tb, there were no differences in microbial composition of the dental plaque between the bonded and the banded sites, and no differences in CFU ratios (aerobe:anaerobe) between the sexes at Tb were seen ( Table II ). The supragingival and subgingival CFU ratios (aerobe:anaerobe) decreased significantly over time for both the banded and the bonded sites ( P <0.05). The decrease in the ratio was not significantly different for the banded compared with the bonded sites. The presence of P gingivalis increased over time for both banded and bonded sites, but no intersite or intersex differences were seen.

Apr 11, 2017 | Posted by in Orthodontics | Comments Off on Longitudinal changes in gingival crevicular fluid after placement of fixed orthodontic appliances
Premium Wordpress Themes by UFO Themes