Performance of a Conventional Sealant and a Flowable Composite on Minimally Invasive Prepared Fissures
Three different fissure preparation procedures were tested and compared to the non-invasive approach using a conventional unfilled sealant and a flowable composite. Eighty permanent molars were selected and divided into 4 groups of 20 teeth each. All the teeth were split into 2 halves, and the exposed fissures were photographed under a microscope (35×) before and after being prepared using the following methods: (I) Er:YAG laser (KEY Laser, KaVo) 600 mJ pulse energy, 6 Hz; (II) diamond bur; (III) Er: YAG laser (KEY Laser, KaVo) 200 mJ pulse energy, 4 Hz; (IV) Control group: Powder jet cleaner (Prophyflex, KaVo, Germany). The pre-and post-images were superimposed in order to evaluate the amount of hard tissue removed. Ten teeth in each group were then acid etched and sealed with an unfilled sealant (Delton opaque, Dentsply), while the remaining 10 teeth were acid etched, primed and bonded (Prime & Bond NT, Dentsply) and sealed with a flowable composite (X-flow, DeTrey, Dentsply). Material penetration and microleakage were evaluated after thermocycling (5000 cycles) and staining with methylene blue 5%. ANOVA and Mann-Whitney tests were applied for statistical analysis. The laser 600 mJ and bur eliminated the greatest amount of hard tissue. The control teeth presented the least microleakage when sealed with Delton or X-flow. A correlation between material penetration and microleakage could not be statistically confirmed. Mechanical preparation prior to fissure sealing did not enhance the final performance of the sealant.SUMMARY
INTRODUCTION
Fissure sealants are considered an effective measure for occlusal caries prevention, especially when combined with regular fluoride application, proper dietary counseling and implementation of adequate oral hygiene (Kidd & Nunn, 2003). This treatment is, however, considered risky by many researchers and clinicians who are reluctant to seal incipient (undetected) caries (Weerheijm & others, 1992; Primosch & Barr, 2001). It is well known that accurate occlusal caries detection is a difficult task, and many decayed fissures go unnoticed or falsely classified as caries-free (Pitts, 2004). When such fissures are sealed, the caries below becomes masked, hindering the subsequent evaluation of the lesion (Deery & others, 1995). On the other hand, some authors propagate minimal invasive dentistry and claim that sealing caries provides the opportunity to arrest the decay process (Simonsen, 2002). The acid etching and sealing processes (Kramer, Zelante & Simionato, 1993) inhibit penetration of the bacterial nutrient supply. Viable microorganisms present in the fissure are thus reduced significantly, and caries progress arrestment can be expected (Handelman, Washburn & Wopperer, 1976; Going & others, 1978; Mertz-Fairhurst & others, 1995). However, if the sealant get partially lost, it will no longer be efficient at preventing caries, since the uncovered areas are unprotected from acid attack (Deery & others, 1997; Weintraub, 2001; Simonsen, 2002; Mejàre & others, 2003). Full sealant retention rate ranges between 31% and 87% (Weintraub, 2001), and the expected sealant loss is between 5% and 10% per year (Feigal, 1998). Thus, the invasive approach of widening the fissures prior to sealing (the so-called-enameloplasty) is recommended by many authors and acknowledged by many clinicians (Primosch & Barr, 2001). This procedure not only allows fissure inspection (Pereira, Verdonschot & Huysmans, 2001; Primosch & Barr, 2001), but it also increases sealant retention (Shapira & Eidelman, 1986; Geiger, Gulayev & Weiss, 2000). Moreover, many authors agree with the idea that the invasive approach reduces the risk of microleakage (Salama & Al-Hammad, 2002), increases sealant penetration (Geiger & others, 2000; Salama & Al-Hammad, 2002) and eliminates organic materials and the prismless layer (Gwinnett, 1973; Burrow, Burrow & Makinson, 2001). No conclusive results were obtained in terms of which material should be chosen to seal fissures, thus, the decision was left to the clinicians (Waggoner & Siegal, 1996). It was hypothesized that sealants on non-prepared fissures behave differently from fissures that undergo enameloplasty. A different performance can also be expected when an unfilled sealant or a more viscous flowable composite is applied. Therefore, the aim of this study was to analyze the amount of hard tissues eliminated by different invasive procedures on artificially decayed fissures and evaluate the influence of each technique in the final performance of the sealant. Further, the behavior of a conventional unfilled sealer and a flowable composite was investigated.
METHODS AND MATERIALS
Selection and Preparation of the Samples
Eighty human molars without cavitation were chosen from a pool of extracted teeth that had been stored in a Chloramine 1% solution after extraction. After plaque and soft tissue were removed from the tooth surface, the occlusal surfaces were screened for caries at different sites of the fissure system. For that purpose, a laser fluorescence device was used (DIAGNOdent, KaVo, Biberach, Germany). Only sound teeth with fluorescence values equal to or smaller than 4 (indicative for sound surfaces) were used (Lussi & others, 1999). The samples were then divided into 4 groups of 20 teeth each. Groups I, II and III were assigned as experimental groups and underwent an invasive treatment, while Group IV was considered the control group. In each group, lower (n 10) and upper molars (n 10) were equally distributed. The occlusal aspect of each tooth was photographed. In order to recreate incipient caries, an artificial fissure lesion was created. To do so, the teeth were fully coated, using melted utility wax, leaving approximately 2-mm around the fissures uncovered. The samples were then submitted to a computer assisted automatic caries machine for 4 weeks. This process consisted of 4 steps:
-
From a reservoir containing 2 l of a demineralization solution (pH=4.6) (ten Cate, Buijs & Damen, 1995), 400 ml were sucked into a container into which the waxed teeth were placed so as to cover them completely. The teeth remained in contact with the demineralization solution for 1 hour. During this time, about 50% of the solution was renewed with fresh solution from the demineralization reservoir after 30 minutes.
-
After 1 hour, the demineralization solution was sucked from the container and the teeth were rinsed twice with fresh deionized water (total 800 ml).
-
The container containing the teeth was then filled with 400 ml of a remineralization solution (pH=7.0) from a reservoir (2 l) (ten Cate & others, 1995). During a total of 6 hours, about 50% of the solution was renewed with fresh solution from the remineralization reservoir every 30 minutes.
-
The samples were then rinsed twice with fresh deionized water and the process was restarted.
The solutions were renewed each week. The pH values of the remineralization and demineralization solutions remained constant during the entire procedure.
After being retrieved from the caries machine, the teeth had their wax coatings stripped off and their roots embedded in a self-curing resin block (Technovit 4071, Heraeus Kulzer GmbH & Co, Wehrheim, Germany) in a rectangular mold, so as to leave the crowns exposed. The samples in the 3 experimental groups were then bisected bucco-lingually with a low-speed diamond saw 0.3-mm thick (Isomet, Buehler, Lake Bluff, IL, USA). Each cut was performed along a mesio-distal-running-fissure, thus resulting in an axial cut. One half of each tooth was selected and exposed, so as to observe the fissure longitudinally (as shown in Figure 1b). Two small marks were drilled with a diamond bur at high speed (FG 200S-Intensiv, Grancia, Switzerland) on both sides of the main fissure, as shown in Figure 1, in order to aid the following superimposition. The surface was then digitally photographed under a light microscope at 35× magnification with a camera (Leica M420, Leica, Heerbrugg, Switzerland) linked to a computer. Both halves were then reassembled tightly with a light-curing resin from the buccal and lingual aspect of the crown, so that no apparent break in the surface was noticed (Figure 1c) (Celiberti, Francescut & Lussi; 2006).



Citation: Operative Dentistry 31, 5; 10.2341/05-91
Mechanical Preparation
The teeth were prepared according to the following procedures:
Group 1 (n 20): The samples had their occlusal fissures treated by an Er:YAG Laser (KEY Laser 1243 (KaVo, Biberach, Germany), emission wavelength 2.94 µm, pulse energy 600 mJ and frequency of 6 Hz. Fissure preparation took place under water spray cooling (7ml/minute) with handpiece #2060 provided by the manufacturer for hard tissues preparation. The approximate distance to the fissure was 10-mm in a non-contact mode.
Group 2 (n 20): The occlusal fissures were opened with a specially designed conical drill for minimal preparation and fissure opening (#889M.314.007; Komet, Gebr Brasseler GmbH& Co, Lemgo, Germany). The drill was used in a high-speed hand-piece under water cooling.
Group 3 (n 20): The samples had their occlusal fissures treated by an Er:YAG Laser KEY Laser 1243 (KaVo, Biberach, Germany), emission wavelength 2.94 µm, pulse energy 200 mJ and frequency 4 Hz. Fissure preparation occurred under water spray cooling (7ml/minute), with handpiece #2060 in a non-contact modus at an approximate distance of 10 mm.
Group 4 (n 20): Control group. No separation of the crowns or opening of the fissures was carried out. The fissures were cleaned using a powder jet cleaner (PROPHYflex II, KaVo, Biberach, Germany) with sodium hydrogen carbonate (NaHCO3) particles for approximately 5 seconds (3–5 mm distance) followed by rinsing and drying.
These procedures were carried out by the same operator with the aid of a dental magnifying glass (2.5×) (Sandy Grendel Design, Dr A Grendelmeier & Co, Aarburg, Switzerland). After preparation, the samples of the 3 groups were separated, and a new picture of the longitudinal cut was taken with the light microscope at the same magnification (35×) as performed previously to the preparation (Figure 1d). Only the selected tooth halves that were photographed were used for sealing. The other halves were discarded.
Superimposition of Pre- and Post-preparation Images
The outlines of the fissures before and after preparation were digitally delineated, and both pictures were superimposed with the aid of a design program (Adobe Photoshop 6.0) (see Figure 1e). The resulting area obtained from superimposing pre- to post-preparation images was measured with the help of an image processing program (IM500, Leica M420, Leica, Heerbrugg, Switzerland) and expressed in µm2.
Sealing of the Samples
The samples of all groups were then allotted to 2 subgroups of 10 teeth each for either a sealant (Subgroup A) or flowable composite application (Subgroup B). In both cases, the occlusal surfaces were etched with phosphoric acid gel 35% (Ultra Etch, Ultradent Products Inc, South Jordan, UT, USA) for 30 seconds for ground enamel (Groups 1, 2 and 3) or 60 seconds for unground enamel (Group 4). The etchant was gently stirred on the occlusal surfaces by means of the etchant applicator. Care was taken to avoid contact of the applicator with the enamel surfaces. The teeth were then rinsed with water/air spray for 15 to 20 seconds and dried with oil and water-free compressed air for 5 seconds until the surface appeared chalky.
Teeth in Subgroup A had their fissures sealed with an unfilled sealant (Delton, opaque, light cured, Dentsply DeTrey, Konstanz, Germany). This material was directly applied to the etched and dried enamel with a round-ended applicator BR 06/08 (A Deppeler SA, Rolle, Switzerland). Care was taken to not overfill the fissures. In order to enhance penetration ability, the sealant was left undisturbed for 20 seconds before poly-merization in order to allow it to flow into the fissures system and etched enamel (Chosack & Eidelman, 1988). The sealant was then light cured for 40 seconds with a halogen curing light unit on pulse mode (Astralis 10, Ivoclar Vivadent, Liechtenstein).
Teeth in Subgroup B had their fissures sealed with a flowable composite as follows: Prime & Bond NT (Dentsply, DeTrey, Konstanz, Germany) was applied in ample amounts as recommended by the manufacturer and left undisturbed for 20 seconds. After removing with an air syringe, the material was light cured for 20 seconds. Finally, the flowable composite (X-flow, Dentsply DeTrey) was applied and cured for 40 seconds. Prime & Bond NT and X-flow were light cured with the same polymerization unit at the same pulse mode that was used for subgroup A. The sealant material (Delton) and the bonding system (Prime & Bond NT) had previously been labeled with 0.1% fluorescent dye rhodamin-B-isothiocyanate in a dark chamber in order to show a better contrast between the material and fissure walls (Pioch, 1996; D'Souza & others, 1999). All sealing procedures were performed under ambient relative humidity (45 ± 2%) at a temperature of 25 ± 1°C. Finally, all samples were stored in individual containers at 37°C for 1 week in artificial saliva (Eisenburger & others, 2001).
Thermal Cycling and Microleakage Procedure
Once all the samples had been sealed, they were submitted to a thermocycling process in water for 5000 cycles between 5 ± 2°C and 55 ± 2°C, dwelling time 30 seconds. The surfaces were then coated with a thin layer of nail varnish and a second layer of melted utility wax. Finally, another layer of nail varnish was applied. The sealant and approximately 2-mm of enamel around it were left uncovered. The teeth were then immersed in 5% methylene blue for 24 hours to allow for dye penetration into possible gaps in the tooth-sealant interface. The coatings were stripped off, and the teeth rinsed thoroughly with tap water and embedded in a self-curing resin (Technovit 4071, Heraeus Kulzer GmbH & Co, Wehrheim, Germany) to prevent chipping of the material. Using a low-speed saw (Isomet), the samples were sectioned bucco-lingually with parallel cuts 1-mm wide. As a result, 3 to 6 sections per sample were obtained, yielding 2 to 5 surfaces for analysis.
Microleakage and Sealant Penetration Evaluation
Each cut was observed under a light microscope at 35× magnification (Leica M420, Leica, Heerbrugg, Switzerland). In each sample, the microleakage proportion was analyzed and expressed as the length of dye penetration (µm) divided by the length of the sealant-tooth interface (µm).
The penetration ability of the material was evaluated as the unfilled area of the fissure (µm2). The same procedure was followed for the unfilled areas. In all cases, the average values per tooth were used for further analysis.
Data Analysis
QQ plots were used to assess the distribution. Since the results were not normally distributed, non-parametric ANOVA and Mann-Whitney tests were applied (SPSS 11.0.3, SPSS Chicago, IL, USA). When multiple comparisons were carried out, the Bonferroni-Holm adjustment procedure was performed (Holm, 1979). The level of significance was set at p<0.01 for all tests.
RESULTS
From a total of 80 teeth, 3 were lost when prepared for dye penetration evaluation. Thus, 77 samples were finally analyzed (9 samples in Group 1 and 10 in Groups 2, 3 and 4 in the Delton group, and 10 samples in Groups 1 and 2 and 9 in Groups 3 and 4 in the X-flow group).
The amount of enamel removed by opening the fissures with the laser device at 600 mJ was not statistically significantly different compared to the amount eliminated with the bur (p>0.01). On the other hand, a significantly smaller amount of enamel was ablated by the laser device at 200 mJ (p<0.001) compared to the bur (Table 1). Considering material penetration, the control group showed a tendency towards a larger unfilled area irrespective of the material used when compared to the other 3 experimental groups. However, no statistical differences were registered (Table 2). The smallest microleakage rates in the Delton group were obtained with the bur, followed by the control group and the laser 600 mJ (p>0.01). On the other hand, with the flowable composite, the control group showed the least microleakage rate compared to any of the other 3 preparation techniques (p<0.01), while the teeth lasered at 600 mJ showed the greatest microleakage overall (p<0.01). The control teeth sealed with X-flow showed similar microleakage values than the control teeth sealed with Delton (p>0.01). Fissures prepared with the bur showed significantly less microleakage values when sealed with Delton than when filled with X-flow (p<0.01). The same findings were observed in the laser group at 600 mJ (p<0.01).


DISCUSSION
Some clinicians and investigators show a certain amount of reluctance with respect to sealing incipient caries, while other authors encourage sealing soft, wet, infected dentin caries (Kidd, 2004) and incipient decayed fissures (Handelman & others, 1976; Going & others, 1978; Kramer & others, 1993; Simonsen, 2002). These authors assume that, by doing so, a detrimental environment for the residual microorganisms is created and biofilm activity and caries progress will be arrested. Decayed enamel and dentin surfaces will be preserved and remineralization can occur (Mertz-Fairhurst & others, 1995; Kidd, 2004).
In this study, 3 minimally invasive fissure preparation methods prior to sealing were compared to a non-invasive approach.
No statistical differences were observed in the material penetration of any of the 4 techniques. Many studies state that enameloplasty allows for higher penetration of the sealant material (Geiger & others, 2000; Salama & Al-Hammad, 2002). The methodology applied in those studies for the evaluation of sealant penetration (percentage of sealant penetration, binomial answer: yes or no) does not allow for a justified comparison with the results of this study. Moreover, labeling the white sealant and the transparent bond with Rhodamine, as performed in this study, enhances the distinction of both materials from the enamel background, allowing for better evaluation of material penetration.
The unfilled sealant selected for this study (Delton) was chosen because of its popularity in the clinic and its good record in previous studies (Hatibovic-Kofman, Butler & Sadek, 2001; Blackwood & others, 2002; Duangthip & Lussi, 2003a). The flowable composite was combined with Prime&Bond NT, as suggested by the manufacturer. No statistical differences were detected in the penetration ability in either of the 2 materials. This confirms the results of previous studies that also found no differences in penetration between sealant materials of a different viscosity (Courson & others, 2003; Duangthip & Lussi, 2003a). It can also be speculated that use of the hydrophilic bonding agent Prime & Bond NT, in combination with the flowable composite, may have increased the penetration ability of the latter (Güngör & others, 2003). According to many authors, better penetration results in better retention (Shapira & Eidelman, 1986; García-Godoy & de Araujo, 1994), though not necessarily in less microleakage. In this study, all control teeth presented the poorest penetration ability. However, this did not increase microleakage. Indeed, the least microleakage rates were detected in both control groups. Thus, no relationship could be confirmed between penetration level and microleakage. These findings confirm previous results (Courson & others, 2003; Duangthip & Lussi, 2003a,b).
It was shown that more overfilling resulted in more shrinkage (Geiger & others, 2000). Thus, overfilling can lead to an increased detachment from the enamel surface and, subsequently, to greater microleakage. In this study, care was taken to avoid overfilling. This was, however, more difficult to achieve with the more viscous flowable composite X-flow. From a total of 106 surfaces, 53 (50%) were overfilled with flowable composite, while only 3 surfaces from the 106 (2.8%) sealed with Delton showed overfilling. In the 3 experimental groups, considerably larger fissures resulted from preparation. Far more material was then required to fill in such areas. Since a greater amount of composite might increase shrinkage, this leads to a rise in contraction forces and an increase in dye penetration (Becker, Schriever & Heidemann, 1994). This can explain the significantly greater microleakage observed in the 3 invasive methods when sealed with flowable composite compared to the control teeth. Similarly, significantly greater microleakage rates were detected with X-flow than with Delton when applying the more invasive preparation methods (laser 600 mJ and diamond bur). On the other hand, when Delton was applied, no statistically significant differences were detected in dye penetration between fissures widened with the bur, those treated with the laser at 600 mJ and the control group. The same findings were shown in previous studies (Blackwood & others, 2002; Salama & Al-Hammad, 2002). Even at energies as high as 1000 mJ, the laser obtained results that were comparable to those of bur-prepared fissures (Lupi Pegurier & others, 2003). Unexpectedly, in the Delton group, fissures treated with the laser at 200 mJ showed greater microleakage rates than those prepared with the bur. This contradicts the results of Borsatto and others (2004), who found no statistical differences between laser-prepared and/or conventionally prepared bur and non-prepared fissures.
It is important to note that, in the control group, even if not opened up, the fissures did undergo an air polishing treatment with powder jet cleaner in order to eliminate possible debris and calculus. This procedure may have influenced the final results, thus affecting comparisons with the control groups of other studies whose surfaces were cleaned only with pumice. Furthermore, it is important to bear in mind that the possible detrimental effect of loads was not taken into account in this study. According to Zervou and others (2000), less microleakage can be expected when load is applied on fissures that undergo enameloplasty. This point has to be further investigated.
The technique used to compare the areas post-and pre-preparation allowed for a perfect superimposition of both images, thus providing very confident results. By splitting the teeth with a diamond saw, a 300-µm gap was created between both halves of the teeth. The parts could, however, be very well reassembled, avoiding deviations from the fissure track during preparation. The enamel surfaces treated with the laser showed a pattern of crater-shaped pits similar to micro-explosions, corresponding to the laser pulses. This resulted in a rough, flaky appearance of the surface. Even when preliminary tests were carried out to familiarize the operator with the laser device, controlled preparation was very difficult to achieve due to the non-contact modus and the discontinuous emission mode. As a result, the pits or micro-explosions were not always uniformly distributed on the laser-treated fissure. In some cases, they were present on one fissure wall, with the other wall remaining intact (see Figure 2a), depending on the incidence of the laser beam. At higher pulses of energy and higher pulse frequencies, wider and deeper craters could be observed (see Figures 2a and 2b). Moreover, deep and narrow fissures were likely to be hit only at their entrance. In no case did the laser beam reach the bottom of these fissures (see Figure 2b). Scanning electron microscope images confirm these findings (Armengol & others, 1999; Manhart & others, 2004).



Citation: Operative Dentistry 31, 5; 10.2341/05-91
CONCLUSIONS
-
No inverse relationship could be found between material penetration and microleakage.
-
The diamond bur and laser at 600 mJ eliminated the greatest amount of enamel.
-
Laser-treated fissures at 600 mJ or those prepared with the bur showed statistically significantly greater microleakage rates when sealed with X-flow than when sealed with Delton.
-
Native fissures or fissures widened with the diamond bur presented the least microleakage rates, especially when sealed with Delton.

Diagram of the experimental procedure step-by-step (a specimen prepared with the bur is presented):
a. Bucco-oral cut through the lesion center.
b. Light-microscope images (35×) before preparation. Note the marks with the drill on both sides of the fissure to facilitate posterior superimposition.
c. Reassembly of both tooth halves with resin composite
d. Light-microscope image (35×) after preparation.
e. Superimposition of light-microscope pre- and post-preparation images

Light microscope images (35×) of 2 laser-treated fissures:
a. Preparation at 600 mJ, 6 Hz.
b. Preparation at 200 mJ, 4 Hz.
Notice that, irrespective of the parameters applied, the laser beam hits only the right wall of the fissure, leaving the left wall almost intact in several cases (a). Narrow fissures were usually ground only at their entrance, since the laser beam could not penetrate deeper (b).
Contributor Notes