Abstract
Sealing caries fissures is considered an appropriate treatment option for arresting the caries process. However, little information is available regarding the sealing of occlusal cavitated dentin lesions. The hypothesis tested in this in vitro study was that no difference in microleakage and sealant penetration depth exists between cavitated and sound sealed fissures when a resin is used. Eighty molars, each with an occlusal cavitated dentin lesion, were treated according to 5 experimental protocols and compared with a control group of sealed sound molars. In the experimental groups, fissure sealants were placed with and without an adhesive, and in various ways. All teeth were sectioned, and microleakage and sealant penetration into the fissure were evaluated. Sealed caries fissures showed significantly more microleakage and insufficient sealant penetration depth than sound fissures. Neither the use of an adhesive nor its intermediate curing influenced the microleakage score and the penetration ability of sealants.
INTRODUCTION
The prevalence of dental caries in Western societies is still dominated by the occurrence of caries lesions at occlusal surfaces of, predominantly, permanent molars (Marthaler et al., 1996; Eriksen, 1998). Fissure sealants can be applied to prevent occlusal caries (Mejàre et al., 2003; Ahovuo-Saloranta et al., 2004). Several factors influence adequate placement of fissure sealants, such as operator skill, sealant material, fissure type, quality of enamel, and phase of eruption (Simonsen, 2002). The application of an adhesive can improve sealant quality when sound fissures are sealed (Symons et al., 1996; Feigal et al., 2000), but other researchers have raised doubts about this (Boksman et al., 1993; Pinar et al., 2005). Contemporary protocols for the treatment of dental caries support the use of sealants, not only as a preventive treatment for sound fissures, but also to arrest progression of dental caries by sealing over active occlusal caries lesions (Welbury et al., 2004; Feigal and Donly, 2006). Whether caries fissures can be sealed as effectively as sound fissures has not been sufficiently investigated. Some clinical studies have shown that a fissure sealant applied in a caries fissure has resulted in reduction of viable micro-organisms, eventually leading to lesion arrest (Mertz-Fairhurst et al., 1986; Handelman et al., 1987). However, the observation time in these studies was limited, and the caries status of the sealed teeth was not documented precisely. Caries fissures are often surrounded by demineralized enamel and dentin with reduced adhesive properties (Xie et al., 1996; Say et al., 2005). Moreover, cavitated fissures may contain biofilm (Fejerskov and Kidd, 2003), which is impossible to remove because of difficult access, thus preventing a sealant from adapting properly in the fissure. No in vitro studies are available on the sealing of cavitated occlusal dentin lesions, and it is not known whether an adhesive can enhance sealant penetration ability in cavitated caries fissures. The objective of this in vitro study was to compare microleakage and penetration depths of sealants applied in caries and sound fissures, using different adhesives and curing methods. The hypothesis tested was that there is no difference in microleakage scores and sealant penetration depths of sealants placed in caries and sound fissures, by different adhesives and curing methods.
MATERIALS & METHODS
Tooth Selection and Sealant Application
From a batch of third molars stored in tap water immediately after extraction, 80 were selected on the basis of the presence of a small occlusal cavitated dentin lesion. These molars were collected with the donors’ informed consent. Institutional Review Board approval was secured for the use of human teeth.
Cavitated lesions to which a dental probe had access (1 mm) were included in the study. Bitewing radiographs were taken to exclude molars without occlusal radiolucency (D1 and D2 lesions) or with a radiolucency indicating a D4 lesion close to the pulp (Fejerskov and Kidd, 2003). Sixteen molars with sound non-discolored fissures were also selected. The carious molars were randomly divided into 5 experimental groups, and the sound molars formed the control group. Teeth were cleansed with pumice and treated by one operator according to 6 different protocols (Table 1).
All occlusal surfaces were etched for 15 sec with 37% phosphoric acid, rinsed thoroughly with water for 10 sec, and gently air-dried until the surface appeared ’frosted’. In groups 1 and 6 (control), a fissure sealant (Teethmate F-1, Kuraray, Osaka, Japan) was applied. A dental probe was used to remove entrapped air from the sealant. The sealant was cured for 20 sec, by means of an LE-Demetron I curing device (Kerr, Danbury, CT, USA) with an output of 800 mW/mm2. In groups 2–5, 2 three-step adhesives were applied to the etched surface before sealant placement, according to 1 of 4 protocols. In groups 2 and 3, OptiBond FL (Kerr) was used. The primer was scrubbed vigorously for 15 sec over the occlusal surface and gently air-dried for a few sec, care being taken to avoid desiccating dentin. The adhesive was then applied and cured for 30 sec (group 2), or left uncured (group 3). In groups 4 and 5, Clearfil SA primer/Photo Bond (Kuraray) was applied and gently air-dried. In group 4, the adhesive was spread by a gentle stream of air and cured for 10 sec, whereas in group 5 the adhesive was left uncured before sealant application. The sealant (Teethmate F-1, Kuraray) was subsequently applied in groups 2–5 as in groups 1 and 6.
Teeth were thermocycled (500 cycles; 5–55°C; dwell time, 30 sec), covered with nail varnish (except on the occlusal surfaces), and immersed in 1% methylene blue for 24 hrs. Thereafter, teeth were embedded in acrylic resin (Struers, Copenhagen, Denmark) and sectioned in a buccolingual direction through the mesial, central, and distal areas of the fissure by means of sawing apparatus (Leica, Heidelberg, Germany: blade thickness, 300 3m). This resulted in 4 molar fragments and 6 section sides per molar.
Evaluation
Using a microscope with a 20x magnification (Forty; American Optical, New York, NY, USA), two independent observers scored microleakage per section side on a three-point rating scale (Score 0, no microleakage visible; Score 1, microleakage in up to half of the fissure; Score 2, microleakage in more than half of the fissure). In case of disagreement, a third independent observer was consulted, who made the final decision. Each molar received 1 microleakage score calculated from the 6 section sides according to a worst-case scenario, so the highest microleakage score of all section sides of the same tooth determined the final score. To determine intra-examination reliability, we randomly selected and re-examined 10 molars (10%) after 1 wk.
Furthermore, section sides were digitally photographed, by means of a microscope with a 40x magnification (Leica). The presence of an unfilled area below the sealant material (incomplete sealant penetration) was determined on the basis of the digitized photographs of the section sides. Moreover, each section side was evaluated on the presence of a caries lesion in dentin, leading to the score "caries present" or "caries absent". Since both section sides adjacent to 1 sawing line were closely related to each other, a final score for the presence or absence of an unfilled area below the sealant material was formulated for those combined sections. Accordingly, when 1 section side was diagnosed as carious, it was decided that the final score for the combined sections would be considered "caries present". Finally, the micro-morphological fissure type was scored either as wide (including U or V fissure type) or as narrow (including Y1 or Y2 fissure type), because several studies have shown an influence of fissure type on sealant penetration (Duangthip and Lussi, 2003). To determine intra-examiner agreement, we randomly selected and re-examined 46 section sides (7%).
Statistical Analysis
Data were analysed with SPSS 12 (SPSS Inc., Chicago, IL, USA). Chi-square tests with Fisher’s Exact tests were conducted, and risk ratios (RR) and 95% confidence intervals were calculated, to test whether the independent variables (caries status, use and type of adhesive, intermediate light curing of adhesive) influenced the performance of sealant application. Microleakage scores per molar and sealant penetration per combined section were analyzed. The influence of adhesive application and intermediate curing was calculated from the ’carious combined’ sections only. We used the chi-square test to see whether the division of micro-morphological fissure type (wide vs. narrow) among the experimental groups was balanced.
For microleakage, sealant penetration measurements, and presence of a dentin lesion, Kappa scores were calculated from those molars that had been evaluated twice.
RESULTS
Examples of incomplete sealant penetration and microleakage are shown in the Fig. Statistical analysis revealed that the molars in control group 6 showed significantly less microleakage than did carious molars (group 1) (chi-square test: p = 0.004). No significant differences could be observed (p = 0.80) among the different experimental groups (groups 1–5) (Table 2).
Results after comparison of the combined sections in groups 1 and 6, where no adhesive was used, showed that significantly better sealant penetration was achieved when a sound fissure was present (chi-square test: p < 0.0001) (RR = 2.70). Thus, the chances of having incomplete sealant penetration when the fissure is carious are 2.7 times (170%) higher than when the fissure is sound (Table 3). The use and type of adhesive had no significant influence on sealant penetration in caries fissures (chi-square test: p = 0.92). Moreover, separate curing of the adhesive before sealant placement did not influence sealant penetration in caries fissures (chi-square test: p = 1.00) (Table 3).
Analysis of the distribution of micro-morphological fissure types revealed an independent distribution of fissure types among the 5 experimental groups (chi-square test: p = 0.53).
For microleakage measurements, Kappa statistics showed a perfect intra-examination reliability (κ = 1). The intra-examiner reliability for sealant penetration ability scores showed a κ-value of 0.83, and for the presence of a dentin lesion, a κ-value of 0.79 was calculated.
DISCUSSION
In contemporary cariology, there is a paradigm shift from the concept that carious tissues should be removed before a tooth with a caries lesion is restored, toward the view that a caries lesion can be sealed and, thus, be arrested (Fejerskov and Kidd, 2003). Consequently, excavation of carious tissues can be limited, or even omitted, as long as the seal is properly placed (Kidd, 2004). The present study deals with the question as to whether a cavitated occlusal caries lesion can be sealed adequately. The hypothesis tested in this in vitro study was that no difference would be found in microleakage scores and sealant penetration depths of sealants placed in caries and sound fissures if different bonding systems and curing methods were used. This hypothesis was rejected, because analysis of the data revealed that, in caries fissures, significantly more microleakage was present, and penetration of the sealant in the caries fissure was often incomplete, with unfilled space left under the sealant.
Clinical studies have shown that sealants placed in caries fissures can arrest the caries process, but either their evaluation periods were short or the carious status of the fissures was unclear (Mertz-Fairhurst et al., 1986; Handelman et al., 1987). However, in a retrospective study (Weerheijm et al., 1992), cariogenic micro-organisms were found in 50% of the sealed teeth, and the dentin was soft and moist, which indicated the presence of active caries. In the present study, molars with occlusal cavitated dentin lesions surrounded by demineralized enamel were used. Therefore, inadequate adhesion to demineralized enamel could be one reason why microleakage scores were higher in the ’carious combined’ sections than in the ’sound combined’ sections. Several in vitro studies have been carried out to investigate the influence of resin penetration into white-spot enamel lesions on smooth enamel surfaces, some showing good infiltration of resin into these lesions (Gray and Shellis, 2002; Bjarnason et al., 2003). A recent study, however, showed that infiltration depended on the type of adhesive being used (Mueller et al., 2006). In the current study, demineralized enamel was not visible on smooth surfaces, but was located in the fissures, which could explain the different outcome. This argument is supported by recently published research (Celiberti and Lussi, 2007) reporting higher microleakage scores in sealants placed over natural enamel caries lesions, especially when the outline was situated on carious enamel. In any event, application of an adhesive did not improve microleakage and sealant penetration depth. It is probably more likely that the irregular shapes of cavitated fissures might have hampered sealant penetration into the caries fissure, as found in this study, even when a probe was used to prevent porosities inside the sealant material. The suboptimal sealant adaptation found in the caries fissures might also have been caused by the presence of biofilm, left in the deeper parts, which are difficult to access (Fejerskov and Kidd, 2003). These findings indicate that cavitated caries fissures cannot be sealed as adequately as sound fissures. However, the clinical implication of these findings is unclear, owing to the in vitro design of the study.
Molars in the experimental groups were selected if a small occlusal cavitated dentin lesion was present. In the current study, the sealed molars were cut 3 times with a saw, resulting in 4 fragments and 6 section sides for scoring. Some sections showed no sealant material in the fissures. These sections were excluded from the analysis, which explained the lower number of combined section sides than expected. To establish a relationship between the presence of an unfilled area below the sealant and the presence of a dentin lesion, we divided the molars into combined carious and sound section sides. The score per combined section side was used in the analysis, on the assumption that no statistical dependence existed between combined section sides within one molar. This assumption was justified by the presence of differences in morphological fissure type and caries status within one molar, and by the fact that the combined sides of the fragments were located at a distance from each other.
We used a three-point rating scale to score microleakage, instead of the more sophisticated methods, such as measuring the length of dye penetration, as used in other in vitro studies (Duangthip and Lussi, 2003; Celiberti and Lussi, 2007). This was because the sealed cavitated dentin lesions in the present study had many irregular borders, making the current technology standard unusable. Second, in most sections, microleakage was seen in small amounts, either at the bottoms of the fissures or at their edges. To underline the impact of differences in microleakage at these two sides, we chose a three-point rating scale.
The use of an adhesive system before sealant placement was not shown to reduce microleakage or enhance sealant penetration for the carious treatment groups. Also, no differences among groups could be observed regarding intermediate curing of the adhesive. Application of an adhesive and intermediate curing of the adhesive had been expected to result in better adaptation of the sealant. That no better results were achieved may have been due to the caries status of the fissures and to cavitation with irregular borders that might have prevented proper application of an adhesive.
Within the restrictions of this in vitro study, it can be concluded that sound fissures can be sealed with resin more adequately than cavitated caries fissures. Neither the use of an adhesive nor intermediate curing of the adhesive improved adaptation of the sealant material.
Treatment Groups and Treatment Procedures
Microleakage Scores per Molar for Each Group
Results for Sealant Penetration per Combined Section Related to Caries Status, Application of Adhesive, and Curing Technique
Examples of incomplete sealant penetration and microleakage score.
Footnotes
Notes
Acknowledgements
The authors thank W.E. Schulte for technical support of the present study. This study was supported by the School of Dentistry, Radboud University Nijmegen Medical Centre, The Netherlands.