INTRODUCTION

After orthodontic treatment, the enamel surface should be returned to its original condition as close as possible without any composite remnants and without damaging the tooth surface.¹ So far, no consensus exists as to the most efficient technique to remove composite remnants after bracket debonding.² Despite many different approaches used in daily practice, the tungsten bur remains the most preferred tool to remove composite remnants.^3,4 Tungsten carbide burs are effective in removing composite remnants but result in a rough enamel surface. Therefore, enamel polishing is required to avoid bacterial adhesion.⁵ This is recommended to be performed with Sof-Lex discs (Sof-Lex,3M/ESPE, St. Paul, Minnesota, USA).² Although the ideal procedure for removing composite has been the subject of several investigations, less effort has gone into establishing a technique that facilitates the removal of composite remnants.^6–8

The fluorescence properties of luminescent chemicals and tooth structures are different under a wavelength of 405 nm. Therefore, fluorescence can be a good tool to use as a noninvasive method for detecting composites.^9–12 Recently, this method has been used to assess the effectiveness of composite remnant removal after orthodontic debonding.^6–8 However, the volumetric assessment of composite remnants and enamel loss after debonding as well as the time needed for removal under fluorescent light has not been investigated. The aims of this in vitro study were to evaluate the amount of composite remnants, tooth substance loss, and the time needed for the removal procedure using a fluorescence-aided identification technique (FIT) compared with a conventional light source (CLS).

MATERIALS AND METHODS

One hundred and twenty permanent bovine incisors were extracted, cleaned, and stored in 0.5% chloramine-T solution at room temperature until further processing. To the naked eye, these teeth had an intact surface without any staining, demineralization, caries, enamel cracks, or fractures. Twelve upper dental arches were produced as follows: Ten teeth, ranging from tooth 15 to 25, were mounted on a wax plate with interdental contacts mimicking a maxillary dental arch, interlocked with hot-setting glue, and embedded in hot polymer (ProBase, Ivoclar Vivadent AG, Schaan, Liechtenstein). Teeth were selected and assembled according to the size of the buccal surface to produce a harmonious model. Size of the buccal surface had to be at least double the size of the corresponding bracket to provide enough unaltered surface for subsequent superimposition of the optical scans. The models were randomized into two groups: FIT group (FIT; n = 6) and conventional light source group (CLS; n = 6). A digital 3D surface scan (CEREC Omnicam, Software SW 4.5,1 Dentsply Sirona, York PA, USA) was performed from the labial, covering the area from tooth 15 to 25 on every model from both groups.

After cleaning with rubber cups using nonfluoridated and oil-free pumice and water for 10 seconds, each tooth was bonded with a conventional bracket (Victory Series, 3M, St. Paul, Minn, USA) by etching the middle third of the crown with 35% phosphoric acid (Opal Etch, Ultradent, South Jordan, Utah, USA) for 60 seconds using a silicon pattern with holes of 5 mm diameter. Then, the acid was rinsed with water for 60 seconds, and the tooth air-dried for 10 seconds. Afterward, the primer and sealant (Opal Seal, Ultradent) was applied and light-cured for 5 seconds (Bluephase 20i, Ivoclar Vivadent AG, Schaan, Lichtenstein). Opal Bond (Ultradent) was applied on the brackets and, once firmly placed and the excess removed, light-cured for 20 seconds mesial and distal of each bracket as suggested by the manufacturer. The prepared models were stored in water for approximately one day until the debonding procedure. To mimic the clinical debonding situation, the models were mounted in a dental mannequin (Frasaco GmbH, Tettnang, Germany), which was permanently fixed to a dental chair (Teneo, Dentsply Sirona, York, Pa, USA).

Two blinded right-handed operators were recruited for the debonding procedure: an experienced orthodontist (A) and an undergraduate student in the fifth year (B). The Ishihara test was performed to exclude color blindness and color weakness. They were instructed to remove the brackets with a bracket-removing plier (678-220L, Hu-Friedy, Chicago, Ill, USA) and all of the composite remnants with a six-blade tungsten carbide bur (H23RA, Gebr. Brasseler GmbH, Lemgo, Germany) mounted in a low-speed contra-angle handpiece (KaVo Master Series, Biberach, Germany) by applying water cooling first and air cooling second. Afterward, the enamel surface was polished with multistep Sof-Lex discs (coarse, medium, fine, super fine; Sof-Lex, 3M) using air cooling. For each model, a new tungsten carbide bur and Sof-Lex discs were used. The operators were instructed to clean the enamel surface until no visible composite remnants could be detected. Magnification was not allowed and, for detection, the operators used dental mirrors, probes and the multifunctional syringe only.

Each operator prepared three of the six models under illumination of the operating lamp (LEDview, Sirona, Bensheim, Germany) of the dental chair (CLS group), and the other three models (FIT group) by using a prototype fluorescence inducing (λ = 405 ± 7 nm) headlamp (Karl Storz GmbH & Co. KG, Tuttlingen, Germany). A postoperative optical three-dimensional scan was performed of each model (Figure 1). A digital stopwatch was used to record the time (seconds) from start of composite remnant removal to the end of polishing.

View Figure

• Download as Powerpoint
• Download Figure

The best-fit method was used to superimpose the pre- and postoperative scans through dedicated superimposition software (OraCheck, Version 2.13.8676, Cyfex AG, Zurich, Switzerland). The unaltered part of the incisal, buccal, and oral surfaces were used as reference points.^12,13 Each tooth surface was superimposed separately and the differences in the distances between any point of the pre-and postoperative scans were identified by color-coding (Figure 2) to the nearest 0.01 mm. All surfaces were analyzed independently by two examiners once. The differences between these two measurements were used to determine the interexaminer variation. The following parameters were measured:

• The number of teeth with composite remnants

• Composite remnant height: the largest perpendicular distance between the tooth and the remnant surface in μm with the “cursor-distance” tool

• Composite remnant volume: the volume of remaining composite in mm³ with the “volume analysis” tool

• Defect depth: the largest perpendicular distance between the tooth and the defect surface in μm with the “cursor-distance” tool

• Defect volume: the volume of the enamel defect in mm³ with the “volume analysis” tool

• The time required for composite remnant removal

View Figure

• Download as Powerpoint
• Download Figure

RESULTS

For each continuous variable, the interexaminer variation was smaller than the standard deviation of either the first or second sets of measurements, meaning that the systematic error was small enough to have no clinical relevance (Table 1). The chi-square statistics revealed no significant difference between Operator A and B regarding the percentage of teeth with composite remnants. Both operators left significantly less teeth with composite remnants with FIT compared with CLS (Figure 3).

Table 1 Interexaminer Variation: Results from the Double Measurements

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Height and volume of the composite remnants were significantly higher in the CLS group than in the FIT group, regardless of the operator (Table 2). Both operators produced significantly larger defect depths in the CLS group than in the FIT group (Table 2). The defect volume was significantly smaller in the FIT group than in the CLS group for operator A, but not for operator B (Table 2). Both operators were significantly faster when the removal was performed with FIT (Figure 4). Operator A was significantly faster and left more composite volume than B by using CLS (Figure 4, Table 2). By the FIT method, operator A generated thinner and smaller defects than operator B (Table 2).

Table 2 Results for Operator A and B for Both Methods (CLS or FIT): Surface Changes (μm) and Volume Changes (mm³), Standard Deviation, Minimum and Maximum and P Values^a

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

The present study showed that the FIT was superior to the conventional light technique in amount of composite removed, avoiding enamel defect generation, and time required for clean-up procedures. To simulate a clinical setting and the related operational limitations due to patient's head and teeth position, full arch models were mounted in a dental mannequin. The bovine teeth used have been shown to be an acceptable substitute for human teeth for bonding and polishing tests.^14,15 The collection of human teeth with the same quality regarding the age and storage time without any tooth decay is often difficult.¹⁶ The large and flat tooth surfaces of the bovine incisor teeth provided enough unaltered surface, which was used as a reference for the 3D superimposition. The bovine teeth of this study were in an identical development stage at the time of collection. By means of a standardized manner of cleaning, disinfection, and storage time, the comparability could be guaranteed.

Intraoral scanning was chosen, since it is a cost-effective, quick, and easily applicable measuring tool for clinical investigations with a high reliability.^12,17 The direct 3D digitalization reduced the risk of imprecision due to the elimination of intermediate steps required by the scanning of plaster casts.¹⁸ Since scanning precision improves from full arch to quadrant and to single tooth,^19–21 each tooth was superimposed separately.

The number of teeth with composite remnants was relatively high in this study (CLS: A: 66% and B: 73%) compared with the study by Ryf et al. (27%).¹ Polishing systems with good composite polishing properties may foster more composite remnants due to the lustrous surface that can be achieved with them.¹ Therefore, the good polishing properties of the Sof-Lex discs²² may have resulted in more teeth showing composite remnants. In this study, the height of composite remnants ranged from 0 to 57 μm and the composite remnant volumes from 0 to 0.15 mm³. The study of Ryf et al.¹ reported an average height of 229.2 μm, which was considerably more, and a mean composite volume of 0.22 mm³. The polishing with the Sof-Lex discs may have left a thin, but larger, area with composite remnants on the teeth, leading to a comparable result for composite volume. Further discrepancies may not only be explained by the different experimental conditions, but also by the different methods of measurement and calculation.

The phosphoric acid etching technique leads to bonding infiltration between 10–20 μm in the enamel of human teeth²³ and 8.7 μm in bovine teeth.⁶ Due to infiltration, complete removal of the composite remnants is never possible without damaging the enamel surface.^2,23 By conventional light, the primer and sealant agent may still remain on the teeth even though, clinically, the surface might look clean.²⁴ The resin infiltrated enamel could be responsible for color changes of the enamel in the long term.²⁵ Due to the FIT method, this infiltrated enamel layer becomes visible, which will then be removed during the clean-up process. This can lead to an increased enamel loss of up to 60 μm with a tungsten carbide bur.⁶

Enamel scratches should be avoided as much as possible²⁶ since they enhance bacterial adhesion⁵ and cannot be eliminated by polishing.^27,28 These enamel alterations of bracket debonding and adhesive removal affect the optical properties of the enamel surface.²⁵ The FIT method could reduce the depth of enamel defects.

Improper handling of the tungsten carbide bur at the line angle and cervical areas can lead to visible grooves²⁹ and the pressure against the enamel is operator-dependent.³⁰ In this study, operator B produced flatter, but a similar amount of enamel defects with both methods compared to the more experienced orthodontist who, to a significant degree, produced flatter and smaller enamel defects using FIT. This difference may have been due to an improper use of the instruments by the less experienced operator.

Examination of the dry enamel by conventional light is crucial for detecting surface irregularities³¹ and, therefore, helps in the detection of composite remnants.²⁴ The previously reported high sensitivity of the FIT method¹² allowed the detection of even small composite remnants in hard to find areas during clean-up, for example in grooves or pits.^28,32 Therefore, the FIT allows the use of rotating instruments by water cooling, thus limiting a potential temperature increase that may damage the pulp with temperature increases of the pulp observed during dry clean-up.³³

Any time reduction is essential for the clinician¹ and might also increase patient comfort. In the present study, the mean time required for one tooth clean-up by CLS (A: 123 seconds, B: 146 seconds) was slightly increased compared with the studies of Krell³⁴ a (113 seconds) and Vidor³⁵ (79 seconds) using the same method. The FIT method reduced the time significantly to 78 seconds (A) and 82 seconds (B). The increased speed of the FIT method during clean-up, in addition to the high detection sensitivity and specificity, is another advantage of the technique.¹²

CONCLUSIONS

• The FIT method improves detection and removal of composite remnants.

• The clean-up is faster with FIT than with CLS.

• The damage to the enamel is operator-dependent even with the FIT method.