Search

To evaluate the in vitro marginal integrity of CAD-CAM composite inlays with the proximal box margin relocated with different direct restorative materials before and after thermomechanical aging.Standardized 4 mm-wide Class II cavities were prepared in 40 sound human upper molars. The mesial cervical margin was located in enamel (control) and 1 mm occlusal to the cemento-enamel junction (CEJ), while the distal cervical margin was located in dentin (experimental) and 2 mm apical to the CEJ. In the experimental groups (n=10), the distal margins were relocated up to the CEJ level by using a highly viscous conventional glass ionomer cement (Equia Forte), low-viscosity bulk-fill composite (SureFil SDR Flow), highly viscous nanofilled composite (Filtek Supreme XT), or a highly viscous bulk-fill composite (Filtek Bulk Fill). The inlays were milled from CAD-CAM nanohybrid composite blocks (Brava Block) and luted with dual-cure resin cement (Allcem). The specimens were submitted to thermocycling (5-55°C, 7,800 cycles) and fatigue loading (50 N, 1 Hz, 240,000 cycles). Epoxy resin replicas of the margins were obtained before and after aging and evaluated through scanning electron microscopy. The percentage of continuous margin in relation to individual assessable margin was calculated as the marginal integrity. The data were statistically analyzed using Wilcoxon and Kruskal-Wallis tests (α= .05).The marginal integrity before and after aging was not significantly different (P0.05). Similarly, no significant differences were observed between margins relocated with different restorative materials in comparison to enamel margins (P0.05).Cervical margin relocation is a valid protocol to simplify intra-oral scanning and luting of CAD-CAM composite inlays with deep proximal margins. The marginal integrity at deep margins with relocation material was comparable to that of enamel margins.

To determine the prevalence of non-carious cervical lesions (NCCLs) in maxillary premolars of different torques and simulated cervical stress profiles of the premolars under coincident loadings using finite element analysis (FEA).The CBCT scans of 616 maxillary premolars from 154 subjects were retrospectively evaluated. The premolars were ascribed into low torque group (LTG)-10.9°, medium torque group (MTG) -10.9° to -3.9°, and high torque group (HTG)-3.9°, when the torque was referring to the occlusion plane. The prevalence of NCCLs in each group was evaluated. Then finite element models of a maxillary first premolar, its adjacent teeth and alveolar bone were established. The models were prepared with ANSYS software generating the premolars presenting different torques. The mastication scenario for the premolars in maximum intercuspation position was simulated.The prevalence of NCCLs was 15.7% in LTG, 7.9% in MTG and 5.5% in HTG. The prevalence of LTG was significantly higher than that of MTG (P0.05) and HTG (P0.01). As for FEA, the stresses at the buccal necks of the premolars basically increased with decrease of the torque. The tensile stress peaks were in the cemento-enamel junction in most premolars of the LTG, while in the middle of the crowns in premolars of MTG and HTG.Low torque with excessive lingual inclination is a risk factor for NCCLs of maxillary premolars, and excessive tensile stress concentration in buccal necks during mastication may be responsible for that.

This systematic review on abfraction studied whether stress is a mechanism in the formation of non-carious cervical lesions (NCCLs).A literature search was performed on three electronic databases (PubMed, ISI Web of Science, and EMBASE) using the keyword "abfraction" in publications published in English. The inclusion criteria were clinical and laboratory studies that investigated the role of abfraction in NCCLs. The title and abstract of the identified publications were screened by two investigators independently. Reviews, case reports, and irrelevant papers were excluded. Full text of the remaining publications were retrieved. A manual search was performed on the bibliographies of the selected publications to identify additional relevant publications for review.A total of 372 publications were identified, and 165 duplicated publications and 166 irrelevant publications were excluded. From the bibliographies of the remaining 41 publications, 28 relevant publications were found. Therefore, 69 publications (31 clinical studies and 38 laboratory studies) were included in this review and the majority (56/69, 81%) found an association between occlusal stress and NCCLs. Although no clinical study demonstrated that NCCL was caused by stress alone, 23 studies reported that stress or occlusal factors were associated with NCCLs. Of the 38 laboratory studies, 24 that used finite element analysis found that stress was concentrated at the cervical region of the tooth. Nine laboratory studies suggested that stress was a mechanism for NCCLs, whereas five studies reported the opposite. In conclusion, current literature supported an association between occlusal stress and NCCLs.This systematic review of abfraction found the majority of studies reported an association between occlusal stress and non-carious cervical lesions.

To evaluate the occlusive potential of four toothpastes by atomic force microscopy (AFM) before and after bleaching and quantify the hydrogen peroxide (HP) diffusion into the pulp chamber after application of desensitizing toothpastes in teeth with cervical lesions.In 52 human extracted premolars, 2-mm deep artificial cervical lesions (ACL) were prepared and rinsed with EDTA for 10 seconds. Then teeth were adapted in a brushing machine and brushed with one of the following toothpastes [Regular toothpaste with no occlusive compounds Colgate Cavity Protection (CP), Oral-B Pro Health (OB), Colgate ProRelief (PR) and Sensodyne Rapid Relief (RR)] under constant loading (250 g; 4.5 cycles/seconds; 3 minutes). In 13 teeth (control group), no artificial cervical lesion was prepared. After that, the teeth were bleached with 35% HP with three 15-minute applications. The HP diffusion was measured spectrophotometrically as a stable red product based on HP reaction with 4-aminoanthipyrine and phenol in presence of peroxidase, at a wavelength of 510 nm and the dentin surfaces of ACL were evaluated before and after bleaching by AFM. Data was statistically analyzed by one-way ANOVA and Tukey's test (alpha = 0.05).In the AFM images, some modifications of the dentin surface were observed after application of OB and RR. However, only for RR the formation of a surface deposit was produced, which occluded the majority of the dentin tubules. Also, only for RR, this deposit was not modified/removed by bleaching. Despite this, all groups with ACL showed higher HP penetration than sound teeth, regardless of the toothpaste used (P0.001).

To evaluate the effects of lasers (Nd:YAG and Er:YAG) and of topical desensitizing agents on dentin tubule occlusion by measuring real-time dentin fluid flow (DFF).32 molars were prepared with V-shape cavity at the cervical area, acid-etched, water rinsed, blotted dry, and treated with (1) Nd:YAG laser; (2) Er:YAG laser; (3) SuperSeal, a desensitizing agent; (4) ClinproXT, a resin-modified glass-ionomer (RMGI) varnish (n = 8 each). A real-time fluid flow measuring instrument (nano-Flow) was used to measure the DFF throughout the procedures. The DFF rates before and after the treatment were compared. Moreover, the surface topography of dentin tubules after each desensitizing method was examined using SEM.DFF varied among the groups. The DFF rate was significantly reduced after laser irradiation/application of the desensitizing agents (P0.05). ClinproXT showed the greatest reduction of DFF rate (71.9%), followed by the SuperSeal (34.8%) and laser groups (P0.05). However, there was no significant difference between the Nd:YAG (24.1%) and Er:YAG (20.6%) groups (P0.05). In SEM images, narrowed dentin tubules were observed in both lased groups and SuperSeal group. In the ClinproXT group, the occluded dentin tubules by the RMGI covering were observed.

An equivalence randomized controlled trial within the subject was organized to evaluate the clinical long-term success of a new 2-step etchrinse adhesive and a new nano-filled ormocer.50 subjects, 21 males and 29 females aged between 21 and 65, were randomized to receive 150 restorations, 100 with the new restorative material, 50 with the composite as control, placed in non-carious cervical lesions with the same bonding system. The main outcome measure was the cause of failure at 8 years. Randomization was number table-generated, with allocation concealment by opaque sequentially numbered sealed and stapled envelopes. Subjects, examiner, and analyst were blinded to group assignment. Two interim analyses were performed. Data were analyzed by ANOVA and Cox test (P0.05).After 8 years, 40 subjects and 120 teeth were included in the analysis of the primary outcome. There were eight failures in the experimental group and four failures in the control group. The cumulative loss rate was 7% for both restorative materials, with the annual failure lower than 1%, without any statistically significant difference. There were two key elements of failure: the presence of sclerotic dentin and the relationship between lesion and gingival margin.

To evaluate the esthetic outcome of four different approaches to treat gingival recession, associated with non-carious cervical lesion (combined defects) and the possible roles of patient-related factors in this esthetic outcome.78 combined defects were previously treated by: coronally advanced flap (CAF), CAF plus cervical restoration using resin-modified glass-ionomer material (CAF+R), connective tissue graft (CTG) and CTG+R. After a follow-up of 2 years, esthetic evaluations were performed using a modification of the Root Coverage Esthetic Score (MRES) and Qualitative Cosmetic Evaluation (QCE). Additionally, regression analyses were performed to evaluate the influence of patient-related factors in the final esthetic outcome.The MRES showed that CAF and CTG had statistically significantly better results, when compared to the other groups (P0.05). Similarly, the QCE showed that CAF and CTG, along with CAF+R presented better results, and CTG+R showed the poorest esthetic outcome. Regression analyses showed that the overall gingival inflammation (full mouth bleeding index--FMBI) was negatively associated with CTG MRES score (P = 0.04 and R = -0.48). This means that the greater the FMBI during the study period, the lower the final esthetic score.

This 3-year prospective clinical study evaluated the clinical performance of a one-step self-etching adhesive (Futurabond NR) in combination with a nanohybrid composite (Grandio) for the treatment of different Class V cavities.122 restorations were placed in 42 patients (mean age of 54 +/- 13.2 years) evaluated according to modified Ryge-criteria at baseline, 6 months, 1, 2 and 3 years. The lesions comprised 91 Class V non-caries cervical lesions (NCCL) and 31 Class V cavities due to caries or restoration replacement. While carious lesions as well as restoration replacement required preparation of dentin, it was not roughened in case of NCCL. Macro-mechanical retention with undercuts was not used. The statistical analysis was carried out based on Bonferroni adjusted McNemar test (global alpha = 0.05) including the criteria marginal adaptation, color match, surface texture, anatomical form, retention and clinical acceptance and pain. The occurrence of secondary caries, preoperative and postoperative sensitivity was also examined.After 3 years of clinical service the restorations showed a significant deterioration of all studied parameters. After 3 years, 65% of the restorations were rated as excellent or acceptable in terms of clinical acceptance, 9% appeared tolerable and 26% were rated as not acceptable which mainly resulted from restoration losses. Within the observed timeframe the retention rate decreased to 75%, which means that 30 out of the 122 restorations were partially or completely lost. There was a difference in the 3-year retention rate of NCCL (71%) and the caries/restoration replacement group (87%) but it was not statistically significant.

To evaluate the putative role of occlusal loading in the pathogenesis of non-carious cervical lesions (NCCLs) in subjects who exhibited mixed excursive guidance [i.e., immediate canine guidance on one side and group function (GF) on the other].20 subjects with Angle Class 1 occlusion and having from 1 to 5 NCCLs on separate teeth were selected. Only subjects who displayed mixed excursive guidance were recruited so that they could serve as their own controls. Non-carious cervical lesions were recorded on casts mounted in semi-adjustable articulators.On the GF sides, 22.5% of all teeth that contacted in working excursions exhibited NCCLs; only 2.1% of the teeth on the canine guided sides exhibited NCCLs, which were found exclusively in canines. Although a case for the multifactorial etiology of NCCLs remains strong, our data, albeit limited, seems to support the dominant role of occlusion in lesion formation.

To compare the desensitizing effects of a glutaraldehyde-based desensitizing system (Gluma) and an Er:YAG laser treatment on cervically exposed hypersensitive dentin.A total of 22 subjects (mean age: 39 +/- 13.7 years; 15 females, 7 males) suffering from cervical dentin hypersensitivity was included in a prospective, split-mouth clinical study. The teeth were treated on one side of the mouth with the glutaraldehyde-based desensitizing system and on the other side with the Er:YAG laser. Sensitivity perception was recorded before treatment (baseline), during and immediately after treatment, after 1 week, 1 month, 3 months and 6 months. The subjects were asked to rate the sensitivity experienced during air stimulation by placing a mark on a visual analogue scale (VAS).Both techniques showed an effective reduction of cervical dentin hypersensitivity. The subjects experienced equal improvements compared to their status before and 6 months after treatment with both methods (P0.001).

To present an updated knowledge on the restorative options and their performance in the particular environment of non-carious cervical lesions (NCCL).Evidence-based support for a proper restoration of NCCL was searched. For this purpose, literature over the last 10 years available in the MEDLINE database was reviewed. Previous systematic reviews, meta-analysis and clinical trials as randomized clinical trials with a testing period of minimum 1 year were selected.Until recently, glass-ionomer-based materials were considered the treatment of choice in most of the NCCL, although consistent improvements are still required. Today, based on the excellent esthetic properties and the good clinical performance, there is a general indication to use composite in NCCL. Clinical behavior was found to be highly product-dependent. The choice of an adhesive with tested efficiency is decisive for the retention and marginal adaptation, whereas differences in flexibility were not found to be clinically important. A periodontal-esthetic approach was suggested by combining the surgical root coverage with the correction of the emergence profile by composite.

Due to an increased prevalence of non-carious cervical lesions (NCCL), a clinical strategy for this lesion type should be considered. Previous reviews focused mainly on etiology and prevalence. In Part 1 of this paper, an evidence-based support for a preventive strategy of NCCL was elaborated.Literature over the last 10 years available in the MEDLINE database was reviewed in order to find clinical evidence for a preventive approach to NCCL. Recommendations were based primarily on systematic reviews, clinical evaluations and a monograph.The etiology of NCCL is currently considered to be rather multifactorial, as clinical investigations found multiple factors associated with this type of lesions and due to the lack of evidence to support exclusively one or another factor. Based on the hypothesis of multifactorial origin, a preventive protocol has been established. No clinical research exists with respect to the prevention of NCCL and long-term clinical evaluations of the proposed preventive measures are needed.

This 24-month randomized paired tooth clinical study evaluated the performance of All Bond 3 used in the simplified (2-step) and full (3-step) versions.33 patients, with at least two similar sized non-carious cervical lesions participated in this study. A total of 66 restorations were placed, half using the 2-step All Bond 3 (AB3-2) and the other half using 3-step All Bond 3 (AB3-3). The restorations were placed incrementally using the composite resin Aelite. The restorations were evaluated at baseline and after 6, 12 and 24 months following the modified USPHS criteria. Statistical differences between the adhesive were tested using with McNemar's test and clinical performance over time for each material with the Fisher's exact test (alpha = 0.05).After 24 months, six AB3-2 and four AB3-3 were rated as bravo for marginal discoloration but did not differ from each other significantly (P0.05). The retention rates at 24 months of AB3-2 and AB3-3 were 90.9% and 97.0%, respectively (P0.05).

To quantify the amount of peroxide penetration from the pulp chamber to the external surface of teeth during the walking bleaching technique.Seventy-two bovine lateral incisors were randomly divided over five experimental groups and one control (n = 12 per group): (1) 35% hydrogen peroxide (HP); (2) 35% carbamide peroxide (CP); (3) sodium perborate (SP); (4) (HP+SP); (5) (CP+SP) and (6) Control (CG), deionized water. All groups were treated according to the walking bleach technique. After 7 days at 37 degrees C in an acetate buffer solution, 100 microl violet leukocrystal coloring and 50 microl peroxidase was added, producing a blue stain that could be measured in a spectrophotometer and then converted into peroxide microg/ml.G5 exhibited the greatest penetration, while G2 and G3 produced the lowest values. All bleaching agents penetrated from the pulp chamber to the external root surface. There was a direct correlation between the presence of oxidative agents and penetration potential. Sodium perborate in distilled water was less oxidative and appeared to be the least aggressive bleaching agent.

To evaluate the clinical effectiveness of two different one-step self-etching adhesives.Two single-step self-etching adhesive systems, Clearfil Tri-S Bond and G-Bond, were evaluated. As a control, a two-step self-etching adhesive system, Clearfil SE Bond, was used. The teeth to be restored were randomly assigned. The resin composite used to restore the teeth was Clearfil AP-X. The three adhesive systems were evaluated by Modified USPHS at baseline, 3 months, 6 months, and 12 months. The evaluation consisted of retention rate, color match, marginal discoloration (interfacial staining), marginal adaptation (integrity), wear, post-operative sensitivity, caries recurrence, and other failures. Changes over time and across groups were evaluated statistically using generalized estimating equations.During the 12-month study period, no statistical differences were observed among the three groups (P0.05) in retention rate, color match, marginal discoloration (interfacial staining), marginal adaptation (integrity). No wear, post-operative sensitivity, caries recurrence, or other failures were detected in any groups. The two one-step self-etching adhesives tested showed good clinical performance at the end of 12 months.

To evaluate the plaque removal efficacy and safety of an advanced rotation-oscillation power toothbrush relative to a newly-introduced sonic toothbrush.This study used a randomized, examiner-blind, two-treatment, four-period, four-sequence crossover design. Subjects received both toothbrushes (Oral-B Triumph and Sonicare FlexCare) and a standard dentifrice from the study site and used each toothbrush at home during an acclimation phase prior to their plaque measurement visits. After abstaining from all oral hygiene for 24 hours, subjects returned to the study site and were assessed with the Rustogi Modified Navy Plaque Index. They then brushed for 2 minutes with their first randomly-assigned toothbrush and post-brushing plaque scores were recorded. This procedure was followed for three additional study visits, with subjects using their normal at-home toothbrush and dentifrice for the 2- to 5-day washout periods between visits. Subjects always abstained from all oral hygiene for 24 hours prior to their visits.45 subjects completed the study. Both brushes were found to be safe and both significantly reduced plaque after a single brushing. Oral-B Triumph was statistically significantly (P0.0001) more effective in plaque removal than Sonicare FlexCare for whole mouth plaque scores, gingival marginal plaque scores and interproximal plaque scores. Compared to Sonicare FlexCare, the adjusted mean plaque reduction scores for Oral-B Triumph were 21%, 23% and 22% greater for whole mouth, marginal and interproximal areas, respectively.

To evaluate the influence of a desensitizing agent (VivaSens) on efficacy of a paint-on bleaching agent (VivaStyle Paint On Plus).Bleaching was performed for 7 days with VivaStyle Paint On Plus. The varnish was applied twice a day for 10 minutes each. 80 subjects were included in the study and randomly distributed in two groups (n = 40) according to exposed cervical dentin and perceived hypersensitivities. Group A used VivaStyle without VivaSens while Group B used VivaStyle Paint On Plus after a single application of VivaSens. Tooth color was assessed on facial surfaces of first upper incisors with Vita shade guide at baseline and 10 days after bleaching therapy. Sensitivity, with intensity graded from 0 (no sensitivity) to 10 (high sensitivity), was assessed chair-side using a blow of air at baseline, at the end of therapy (7 days) and 10 days after bleaching therapy. Statistical evaluation was performed with non parametric ANOVA.Thirteen subjects dropped out of the study; six due to gingival burning sensation (A: 3; B: 3) related to the bleaching regimen and seven due to lack of compliance. Directly after completion of bleaching therapy, tooth color had changed significantly compared to baseline in both treatment groups without difference among the groups. Color changes (Delta) according to Vita shade guide were as follows (mean +/- standard deviation): Group A: Delta 2.7 +/- 1.0, Group B: Delta 2.8 +/- 0.9. After bleaching (7 days) the intensity of tooth hypersensitivity (mean +/- standard deviation) was increased significantly compared to baseline in both groups (P0.05): Group A: 1.58 +/- 1.91 (baseline: 0.4 +/- 0.5); Group B: 1.3 +/- 1.8 (baseline: 0.5 +/- 0.7). The number of subjects reporting tooth hypersensitivity increased in Group A by 5 (n = 13) and in Group B by 1 (n = 10) subject. Although degree of hypersensitivities and number of subjects with hypersensitivities were lower in Group B, there was no significant difference between the groups.

To evaluate the retention and marginal staining of restorations placed in non-carious cervical lesions using the all-in-one self-etching system, One-Up Bond F and Palfique Estelite resin composite.Fifty-one non-undercut non-carious cervical lesions received restorations with One-Up Bond F and Palfique Estelite in 15 subjects (mean age 57.7 years) by one operator. Teeth were cleaned with pumice and water, restored following the manufacturer's instructions and evaluated at 6 months, 1, 2, 4 and 5 years for retention and presence of marginal staining.At 1 year, 42 restorations were evaluated, and all were intact with three restorations showing slight marginal staining. At 2 years, 37 restorations were evaluated from 11 subjects, and all were intact. Marginal staining was observed around nine restorations. At 4 years, 27 restorations and at 5 years, 24 restorations could be evaluated. Marginal staining showed a significant increase at the enamel and cervical margins over the 5 years of the study, with two restorations showing severe, nine mild and 13 no staining. Although a number of restorations could not be evaluated, the retention rate of 92% at 5 years showed One-Up Bond F/Palfique Estelite resin composite for the restoration of non-carious cervical lesions produced an excellent outcome. Marginal staining was limited to a small number of subjects and is unlikely to be a problem if margins are refurbished at recall visits.

To evaluate the effect of various bleaching agents on the cemento-enamel junction (CEJ) of human teeth by scanning electron microscopy (SEM) analysis.30 intact teeth were selected and longitudinally sectioned, yielding 60 specimens. Thirty specimens served as controls; the other 30 were divided into six groups with five specimens each (n = 5) and bleached according to six protocols (Group 1: External bleaching with 10% carbamide peroxide; Group 2: External bleaching with 35% hydrogen peroxide; Group 3: External bleaching with 35% hydrogen peroxide; Group 4: Internal/external bleaching with 35% hydrogen peroxide; Group 5: Internal/external bleaching with 35% hydrogen peroxide; and Group 6: Intracoronal bleaching with a paste of sodium perborate mixed with 9% hydrogen peroxide). After treatment the specimens were prepared and examined in a scanning electron microscope.The bleaching agents used in this study caused morphological changes in the CEJ and increased dentin exposure.

To evaluate in vitro the gingival microleakage of Class II direct and indirect composite restorations with cervical margins 0.5 mm apical to the cemento-enamel junction (CEJ).Mesial-occusal (MO) preparations of similar size were made in 10 homologous pairs of caries-free extracted human third molars. One specimen for each pair was prepared for a direct composite restoration and the other for an indirect composite restoration. Direct preparations were restored per manufacturer's instructions. Indirect preparations were impressed, and composite restorations were fabricated and cemented with a dual-cure cement. A total-etch technique was used for all restorations. Restorations were finished, polished, stored for 1 week in distilled water at 37 degrees C, thermocycled (5 degrees-55 degrees C x 1000), sealed with fingernail polish (leaving a 1.5 mm open periphery adjacent to the gingival margin), and placed in 0.5% basic fuchsin dye for 24 hours. Teeth were sectioned longitudinally (mesio-distally) within the restoration in two cuts and the four resulting surfaces (two inner cut surfaces, two outer cut surfaces) were evaluated for dye penetration with a x10 stereoscope using a scale of 0 (no penetration) to 4 (dye penetration involving more than half the axial wall).All 20 specimens had at least one score of three (dye penetration involving less than half of the axial wall) or four. Statistical analysis (Wilcoxon paired-sample test) disclosed a significant decrease in the indirect composite microleakage scores for the two outer cuts (P = 0.006, P = 0.002). No significant differences in microleakage scores were found between materials for the inner cut surfaces of the specimens. Overall, the results of die penetration showed no statistical difference between Class II direct and indirect composite restorations for microleakage.

To evaluate the clinical performance of two self-etching and one total-etch adhesives in Class V non-carious cervical lesions (NCCL).Two self-etching primers, Clearfil SE Bond and Hybrid Bond, and one total etch adhesive, Admira Bond, were placed in 195 NCCL. Restorations were evaluated at baseline, 1 and 2 years using the USPHS criteria.No restoration was lost after 1 and 2 years for all materials. There was no significant difference between the baseline and 2-year results for Admira Bond and Clearfil SE Bond restorations. In contrast, Hybrid Bond restorations showed significant deterioration in marginal adaptation and cavosurface marginal discoloration after 2 years. Also, there was no significant difference between Admira Bond and Clearfil SE Bond at each recall period.

To evaluate the effectiveness of two agents for treating cervical dentin sensitivity associated with gingival recession or noncarious cervical lesions.44 patients with at least mild sensitivity affecting cervical dentin were enrolled in a longitudinal randomized clinical trial. A resin-based desensitizer or an experimental glass-ionomer was assigned to treat at most two teeth from each side of the mouth. Sensitivity was assessed by tactile and cold tests, measured with a Visual Analogue Scale at baseline, after treatment, and at 1 week, 1, 3, 6, and 12 months after treatment. Other noteworthy clinical observations were recorded.Both treatments effectively reduced dentin sensitivity (mixed linear model analysis). Sensitivity score for the glass-ionomer was significantly lower than for the resin-based desensitizer after treatment and at all follow-up periods (P0.0001). Some overhanging margins were observed in the glass-ionomer group, which could accumulate plaque and cause gingivitis. Despite material loss from some teeth treated with the glass-ionomer, the follow-up sensitivity scores were still lower than baseline scores.

To investigate the desensitizing ability of a one-bottle bonding agent and a glutaraldehyde-based HEMA formulation on sensitive tooth cervical areas for a period up to 9 months.The sample consisted of 40 patients with cervical hypersensitivity. Three sensitive teeth per patient were treated; one received One-Step (one-bottle bonding agent), the other Gluma Desensitizer (glutaraldehyde-based agent) and the third distilled water (control group). The hypersensitivity level was determined before, immediately after the desensitizing session, at 8 weeks, and 9 months post-treatment. Measurements of sensitivity were determined by the patient's response to tactile and air-blast stimuli. A verbal rating scale was used and scored as follows: 0, no discomfort; 1, discomfort but no severe pain; 2, severe pain during simulation; 3, severe pain after simulation. The results were subjected to statistical analysis by Kruskal-Wallis test (a=0.05).Both treatment procedures resulted in reduction of hypersensitivity to both stimuli, for up to 9 months. No significant differences were recorded between One-Step and Gluma Desensitizer at immediate and 8-week examinations, whereas Gluma Desensitizer produced lower hypersensitivity than One-Step at the 9-month assessment. In general, a lower level of reduction was found for the 9-month interval compared to the 8-week hypersensitivity score for both agents tested. A placebo effect was observed with water treatment, ranging from 4.7 to 27.5% reduction of hypersensitivity.

To compare the morphology of experimentally induced cervical toothbrush abrasion lesions to teeth demonstrating non-carious cervical lesions in vivo.Eighteen premolars extracted for orthodontic reasons were each subjected to 80 hours (1.4 million strokes) of horizontal brushing in a custom fabricated toothbrushing machine. Toothpaste slurry was applied continuously and specimens were subjected to 300 gms of toothbrushing force. Denture base resin was used to simulate gingival recession of 1 mm and a sulcus width of 0.1 mm. Cervical abrasions were analyzed by optical and scanning electron microscopy and classified as either V-shape/wedged vs. U-shape/rounded lesions. When lesions exhibited both shapes, they were classified as mixed. Shapes of experimentally induced lesions then were compared to naturally occurring ones found on extracted premolars. RESULTSts: Experimentally induced toothbrush abrasion duplicated the classical clinical shapes. Half (9/18) of the experimentally induced toothbrush abrasions exhibited wedged lesions, 28% (5/18) showed a mixed wedged/rounded lesion, and 22% (4/18) showed rounded lesions. Serial photography showed progression of the morphology of the lesions. Toothbrush abrasion apparently begins apical to the cemento-enamel junction, progresses to dentin, and then undermines enamel with loss of the original cemento-enamel junction. Wedged lesions may appear with the apex oriented coronal or apical, or may be symmetrical.

To investigate the durability of three simplified systems in Class V non-carious abrasion/erosion lesions.144 non-carious cervical dentin lesions were restored either with Clearfil Liner Bond 2, a 2-step self-etching primer (n = 46), One Coat Bond, a one bottle total-etch system applied with one coat (n = 46), or Prompt-L-Pop, a 1-step self-etching primer ("all-in-one") (n = 52), in 90 individuals. Ninety-eight of the lesions showed sclerotic dentin and 46 were non-sclerotic. Sixty-one were slightly roughened with a diamond bur before conditioning. The restorations were evaluated every 6 months during a 2-year period with slightly modified USPHS criteria.All except three restorations were evaluated over 2 years. The cumulative loss rates for Clearfil Liner Bond 2, One Coat Bond and Prompt-L-Pop were at 6 months: 4.3, 2.2 and 3.9%, at 18 months: 4.3, 10.9 and 15.4% and at 24 months: 8.7, 13.0 and 21.2%, respectively. The cumulative loss rates of the materials in sclerotic lesions (15.7%) versus non-sclerotic lesions (14.0%) were not significanty different. Restorations placed with a diamond bur-roughened lesions showed a loss rate of 14.5%, while for the non-roughened lesions the frequency was 14.8%.

To evaluate the clinical effectiveness of a resin-modified glass-ionomer adhesive FujiBond LC (GC) in non-carious Class V cervical lesions.28 patients and 52 abrasion-erosion lesions were restored using FujiBond LC and a microfilled resin composite in a cavity design with the adjacent enamel margins beveled. Clinical parameters as retention, marginal integrity, clinical microleakage, caries recurrence, esthetics, gingival response, tooth vitality and post-operative sensitivity were evaluated at 6 months, 1 and 2 years.A 100% retention rate was recorded after 2 years of clinical service. However, only 15% of the restorations showed a perfect marginal adaptation. All marginal defects were small, either located at the incisal or at the cervical margin. The percentage of small marginal defects was obviously higher at the incisal enamel margin than at the cervical dentin margin. At the 2-year recall, half of the restorations exhibited a superficial, localized discoloration at the restoration margin, however none of the restorations showed deep, generalized discoloration. Remaining criteria of clinical effectiveness were rated as excellent.

To investigate the effect of a low-viscosity resin-based composite (LVC) liner on the microleakage of cervical resin-based composites bonded with new adhesive systems.180 wedge-shaped cervical cavities on bovine incisors were treated with AQ Bond (all-in-one), ABF (experimental two-step self-etching primer containing antibacterial monomer) or Clearfil SE Bond (two-step self-etching primer) systems according to the manufacturers' instructions. Half of the treated cavities in each adhesive system were lined with an LVC (Protect Liner F) and filled with a hybrid resin-based composite (Clearfil AP-X). Ten specimens from each group were immersed in 0.5% basic fuchsin solution immediately after finishing the restorations. The specimens to be subjected to thermocycling (5-60 degrees C, 15 seconds dwell time, 5,000 cycles) or to flexural load cycling (approximately 0.5 mm labio-lingual displacement at the incisal edge, 10,000 cycles, 1 Hz) prior to immersion in the dye solution were also prepared. The teeth were sectioned through the center of the restorations and the degrees of dye penetration were scored. The data were analyzed with the Mann-Whitney U-test or the Kruskal-Wallis test (P0.05).The use of the LVC liner significantly improved marginal sealing of AQ Bond. However, the LVC liner could not prevent the deterioration in the marginal integrity caused by thermocycling or flexural load cycling. In the case of ABF, the use of the LVC liner showed a trend toward less microleakage. For Clearfil SE Bond, the LVC liner prevented the deterioration in the marginal seal caused by thermocycling.

To evaluate the effect of the cavity configuration factor (CF) on the marginal microleakage of cervical restorations with four aesthetic restorative materials.Conventional cavities, 2.9 mm in diameter and 1.5 mm deep, with CF=2.7 and "saucer"-shaped ones with CF=2 were created in 60 extracted premolars. The following groups were established: G1: Z100/Single Bond, G2: Freedom/Stae, G3: Vitremer/Primer and G4: Durafill/Durafill Bond, following each manufacturer's directions. Thermocycling of the specimens was performed in an aqueous solution of 2% buffered methylene blue, with the temperature varying between 5 to 55 degrees C, for a total of cycles of 60 minutes per day, for 7 days. The specimens were then sectioned and evaluated by two observers using photographs acquired from a stereomicroscope. The values were subjected to Kruskal-Wallis analysis and the Dunn and Wilcoxon test.The averages of the microleakage scores observed in the conventional and "saucer"-shaped cavities were respectively: G1: 0.66/0.46; G2: 0.92/0.69; G3: 1.8/1.86; G4: 3.54/2.3.

International audience; Purpose: To test how the number of sections affects the maximum depth of tracer penetration.Materials and methods: This study was simultaneously performed in three different centers. C-shaped Class V cavities were made on 60 human third molars. The cavities were located across the cementum-enamel junction. One restorative system was used: Scotchbond Multi-Purpose and Z100. After restoration, the samples were thermally cycled 3000 times, between 5 and 55 degrees C with a dwell time immersion of 10 seconds. Silver nitrate was used by two centers and basic fuchsin was used by the third. The teeth were longitudinally sectioned with a diamond saw: 5 sections per tooth provided 10 surfaces for evaluation. The penetration of the tracer was recorded on a scale from 0 to 3. The deepest leakage per restoration was identified for comparison with lesser measured values elsewhere in the tooth. The Spearman test was applied to evaluate the relationship between the reference and data from one, two and three sections (i.e. two, four and six measurements). The Kruskal-Wallis test was applied to compare the three centers.Results: Whatever the study center, the Spearman correlation coefficient (r(s)) increased as a function of the number of sections (S) up to three: Center 1 (1S, 0.47; 2S, 0.68; 3S, 1.0), Center 2 (1S, 0.60; 2S, 0.99; 3S, 0.99), Center 3 (S1, 0.40; 2S, 0.73; 3S, 1.0). No statistically significant difference was found between the three study centers

To evaluate the micro-shear bond strength of current adhesive systems to cervical and mid-coronal enamel.Two commercially available resin adhesives were investigated; a self-etching primer system (Clearfil SE Bond) and a one-bottle adhesive system (Single Bond) intended for use with the total-etch wet-bonding technique were employed. Two regions of enamel, cervical and mid-coronal regions, were chosen from the buccal surface of extracted molars and were then bonded with each adhesive system and submitted to the micro-shear bond test. In addition, the conditioned enamel surfaces without any bond and the enamel-resin adhesive interfaces were studied morphologically using scanning electron microscopy (SEM).No significant differences were found between the adhesive systems used (P0.05). However, for both bonding systems, cervical enamel showed significantly lower bonding than mid-coronal enamel (P0.05). The relatively lower bond strengths obtained from cervical enamel probably resulted from its aprismatic structure, which appears to be more resistant to dissolving in acids than prismatic mid-coronal enamel.

To evaluate microleakage at enamel and dentin margins of two composite resins, using bulk and incremental placement techniques, "rebonding", and facial and lingual curing methods.One hundred standardized Class V cavity preparations were made on the facial surface of extracted human premolars and were randomly assigned to 10 groups. Single Bond was used as the dentin/enamel adhesive. A heavily filled composite resin, Z250, and a microfill, Silux Plus, were inserted and polymerized using five different techniques: (1) incremental placement and facial curing; (2) incremental placement, facial curing and rebonding; (3) bulk placement and facial curing; (4) bulk placement, facial curing and rebonding; (5) incremental placement and lingual and facial curing. After the restorations were finished and polished, the margins of those in the rebonded groups were etched, rinsed, and dried. The adhesive resin, Single Bond, was applied at the composite resin-tooth interface and light-cured. All the specimens were thermocycled, stained with 1% methylene blue, sectioned, and evaluated for leakage (0-4 scale) by two examiners.Almost no leakage occurred at enamel margins. At the cementum margins, differences in microleakage related to restorative material or technique were not statistically significant. However, leakage at the cementum margins was significantly greater than at the enamel margins for both composite resin materials.

To clinically evaluate the 3-year performance of a one-bottle dentin adhesive in the restoration of non-carious cervical lesions (NCCLs) with resin composite.105 NCCLs were restored with either AElite Flow, Bisco Glaze (both flowable resin-based composites) or Silux (a microfilled resin-based composite), using One-Step dentin adhesive. The cavities were cleaned with pumice and water slurry, the dentin lightly roughened with a fine diamond bur, and the restorations placed according to the manufacturer's instructions. Patients were recalled at 6 months, 1 year, 2 years and 3 years, and restoration integrity and marginal discoloration assessed.The overall retention rates were: 6 months, 100%; 1 year, 95%; 2 years, 87%; 3 years, 75%. Marginal discoloration was negligible, and there was no significant difference among the materials, although there was a trend for it to be greater around the Bisco Glaze restorations.

To assess the marginal seal of Class II resin-based composite restorations cured by a high intensity halogen light or a plasma arc light.Class II cavities were prepared in extracted molar teeth with cervical margins located apical to the cemento-enamel junction. Cavities were restored with a posterior resin-based composite (Tetric Ceram) using either a two-increment or a bulk placement and cure restorative technique after application of bonding agent (Excite). Test parameters included light unit type/curing mode, application of flowable composite and restorative technique. After thermocycling, the teeth were immersed in basic fuchsin dye. Sectioned restorations were examined under a stereomicroscope (x 25) and extent of microleakage (mm) was recorded using an image analysis system. Data was analyzed by non-parametric statistical tests.No leakage was observed at the occlusal enamel margin of a nyrestoration. Median gingival leakage values for the different test groups ranged from 0.02-2.51 mm. Light unit type had a highly significant effect on leakage (P=0.0002). The highest leakage scores were recorded with the plasma arc light used in standard curing mode and the lowest with the halogen light in turbo-boost mode.

To evaluate microleakage at the dentin margins of Class II resin-based composite restorations using four simplified adhesive systems: EtchPrime 3.0, PrimeBond NT, Clearfil Liner Bond 2V and Clearfil SE Bond, upon thermal and mechanical stresses.Forty slot preparations (5mm x 3mm x 1.5mm) were cut in sound human molars with gingival margins below the cement/enamel junction. All adhesive systems were applied strictly according to manufacturers' instructions. Cavities were restored using Clearfil AP-X composite resin. After finishing and polishing, teeth were submitted to thermal (x700, 5-55 degrees C, 60-second dwell time) and consecutively mechanical cycling (100,000 cycle; 8 Kgf load). Specimens were coated with nail polish, immersed in silver nitrate for 2 hours and sectioned longitudinally with a diamond disc. The extent of leakage was ranked using a 0-4 scale.Statistical analysis was performed using Kruskal-Wallis test. Minimal leakage occurred with Clearfil SE Bond. There was a statistically significant difference (P0.01) between this system and Clearfil Liner Bond 2V, although the latter also showed low values of dye penetration. No statistically significant difference was found between EtchPrime 3.0 and PrimeBond NT. Both showed statistically greater leakage than the two other systems (P0.01).

To evaluate dye penetration in adhesively bonded cervical amalgam restorations.The specimens were randomly divided into four groups and the adhesives PrimeBond 2.1, PrimeBond 2.1 Dual Cure, Scotchbond Multi Purpose Plus and Amalgambond Plus were tested. After being restored and polished, the teeth were thermocycled 1,000 times at temperatures varying between 5 degrees C +/- 2 degrees C and 55 degrees C +/- 2 degrees C. Statistical analysis was carried out using the Kruskal-Wallis test and multiple comparisons.Scotchbond Multi-Purpose Plus demonstrated lower levels of microleakage in enamel and dentin. The dual adhesives were more effective than the "one bottle" adhesive.

To evaluate the interface with tooth structure, mode of failure and stability of Fuji Bond LC (FBLC) resin-modified glass-ionomer bonding system.The study was in three main parts. The first part with interfacial characteristics of FBLC. The micropermeability of the FBLC/dentin interface of restored cavities in extracted teeth, after fluorescent dye was introduced into the pulp chamber, was assessed by confocal microscopy. In other teeth, confocal microscopy was also used to examine trhe failure of FBLC bonded to flat dentin surfaces, on shear loading. In the second part, cervical cavities restored with FBLC (with and without light-curing) and a resin-based composite were examined for stability of the interface over the first 15 minutes. In the last part of the study, a new fluorescent dye was mixed with FBLC for restoration of cavities and subsequent examination using 2-photon imaging techniques.FBLC adapted well to tooth structure, had an interface with dentin that was permeable to fluid and displayed the absorption layer. Shear loading at the dentin interface resulted in complex failure with areas of cohesive failure where the absorption layer was present. Light-curing of FBLC seemed to produce a stable restorative while "co-curing" resulted in an unstable situation, with the appearance of an emmisable exudate from the FBLC. Advanced fluorescent imaging of the material itself seemed to suggest that there was a movement of aluminum ions in a manner similar to other glass-ionomer cements. The presence of HEMA in the matrix of the cement associated with the absorption layer was also demonstrated.

To evaluate the effectiveness of a dentin bonding agent containing HEMA compared to a resin emulsion in the reduction of cervical dentin hypersensitivity (CDH).The study was a randomized-within-subject, double blind design. Twenty patients who were suffering from CDH from at least two teeth (canines and/or premolars) were enrolled. The subjective perceptions of pain in response to tactile, air blasts and cold ethyl chloride stimuli were evaluated using a visual analog scale (VAS) at baseline and 10 minutes, 1 week and 4 weeks after the treatments. Scotchbond 1 (a dentin bonding agent similar to Single Bond in the USA) and MS Coat (a resin emulsion), both applied according to the manufacturers' directions, were selected for the study.Multiple regression ANOVA for repeated measures showed that both treatments significantly reduced the CDS with respect the baseline values after just 10 minutes. There were no significant differences between the ability of the two treatments to reduce CDH for each of the three evaluation stimuli or for subjective evaluation.

To evaluate the effectiveness of a "one-bottle" dentin adhesive for reducing or eliminating hypersensitivity in exposed cervical dentin.Eight patients with hypersensitive teeth were enrolled in this clinical trial. The hypersensitive teeth included a total of 22 premolars and anterior teeth with exposed cervical dentin that was sensitive to cold or other stimuli. A single layer of PrimeBond 2.1, an acetone-based "one-bottle" adhesive, was applied to the cervical dentin and was light-cured. Sensitivity to compressed air and cold stimuli was determined immediately before and after treatment, and at periods ranging to 24 wks after treatment. Patient responses to the stimuli were timed, and were also evaluated by the use of a visual analogue scale.Application of a single coat of PrimeBond 2.1 significantly reduced cervical dentin hypersensitivity to both compressed air and cold stimuli immediately after treatment. The reduction in hypersensitivity remained significant at 24 wks after treatment.

To investigate the microleakage of four hydrophilic adhesive systems: one "multiple-bottles" (Scotchbond Multi-Purpose Plus); two "one-bottle" (Single Bond, Stae); and one self-etching (EtchPrime 3.0).120 bovine incisor teeth were divided into four groups (n = 30) and Class V cavities were prepared at the cemento-enamel junction. The cavities were restored with the adhesive systems and with Z100 composite. The teeth were thermocycled 1,000 times between 5+/-2 degrees C and 55+/-2 degrees C with a dwell time of 1 min, and then placed in a 2% methylene blue dye (pH 7.0) for 4 hrs, washed and sectioned vertically through the center of the restorations. The qualitative evaluation was made by three examiners who distributed pre-established scores (0-4) for each tooth using a stereomicroscope at x30 magnification.In enamel margins little microleakage was observed and the Kruskal-Wallis analysis did not show differences. In dentin margins the Kruskal-Wallis and multiple comparison analyses were applied: microleakage was significantly greater with Stae (median 3) and Scotchbond MP Plus (median 4). Single Bond (median 1) and EtchPrime 3.0 (median 2) showed the best results in dentin margins, and the statistical analysis did not demonstrate differences in microleakage among these groups.

To compare the ability of three packable resin-based composite (RBC) systems (ALERT, SureFil, and Solitaire), placed and light-cured in either incremental layers or in bulk, to seal the gingival margin of Class II preparations when the gingival margin is placed apical to the cementoenamel junction.Packable RBCs were used to restore 60 extracted human premolars with Class II preparations using an incremental-cure technique (each layer no greater than 2 mm in thickness when light-cured) or a bulk-cure technique (n=10). Manufacturers' directions were followed; the cure technique was the only variable. After the restorations were completed, the specimens were thermocycled, stained, sectioned, and viewed under a light microscope for microleakage at the gingival margin.All specimens showed leakage at the gingival margin. No statistically significant difference was found between the bulk-cure and incremental-cure techniques for each of the RBC systems evaluated. ALERT had significantly less leakage than SureFil and Solitaire.

To evaluate in vitro the dentin morphology in root canals in terms of tubule orientation, density and increase in surface area after etching.Thirty anterior teeth were divided in 3 groups at random: the samples of Group 1 were used to study tubular morphology in SEM. Groups 2 and 3 samples were etched with 32% phosphoric acid. The teeth in Group 2 were examined by SEM without further treatment. The samples in Group 3 were treated with a bonding system and fiber posts were luted into the canal. These teeth were then processed for evaluation of hybrid layer formation and resin tags in dentin tubules. The observations were made according to location in the root dentin, tubule density was estimated, and the increase in area available for bonding after etching was calculated.The observations revealed variability in tubule density and orientation within different areas of any one sample. Statistically significant differences in the density of tubules were found depending on location. The dentin surface area available for bonding increased by 202% after etching in the cervical third, 156% in the middle third, and 113% in the apical third of the root dentin. Group 3 samples showed that the thickness of the hybrid layer depended on the density of tubules. In the sectors with a low density of tubules, the hybrid layer was significantly thinner than in areas with a higher density of tubules. The increase in dentin surface area might be responsible for the enhanced bond strength after acid etching, but not all areas exhibited equal responses to etching.

To evaluate dye penetration of proximal vertical slot resin-based composite (RBC) restorations as a function of cavity size and restorative material employed (RBC, ceramic inserts with low-viscosity resin, and RBC with low-viscosity resin liner). Microleakage was measured using rank scoring and linear measurement of dye penetration and ANOVA and non-parametric tests were used to evaluate the statistical significance of the results.Ninety erupted crack- and caries-free mandibular third molars, extracted for orthodontic reasons, were used. Proximal vertical slots were prepared on the mesial surfaces of the teeth using diamond instruments at high speed under distilled water coolant. Teeth were placed randomly into nine groups. Three sizes of diamond burs of an ultrasonic cavity preparation system were used under low energy in order to standardize cavity size and shape, each tip size being used in 30 different teeth. The gingival margins were prepared 1 mm above the cemento-enamel junction. Specimens were etched, prepared with bonding agent, restored using three different materials, and immersed in a 50% AgNO3 solution for 8 hrs and sectioned to evaluate the dye penetration.Statistical analysis revealed no statistically significant differences among the nine groups.

To evaluate the 5-yr clinical performance of a resin-modified glass-ionomer cement and a polyacid-modified resin composite in restoring non-carious cervical lesions.Non-carious cervical lesions in 46 incisors, canines, and premolars were restored either with Fuji II LC (n=18) or with Dyract (n=28) in 16 healthy patients. The lesions were restored without cavity preparation strictly according to the manufacturer's instructions. The restorations were clinically evaluated single blind after 5 yrs using modified USPHS criteria.The percentage of Alfa ratings were as follows (Dyract/Fuji II LC): color match 81.3%/28.6%, surface texture 93.8%/21.4%, anatomic form 75.0%/28.6%, marginal integrity (enamel) 62.5%/42.9%, marginal integrity (dentin) 68.8%/28.6%, marginal discoloration (enamel) 56.3%/42.9%, marginal discoloration (dentin) 68.8%/21.5%. Five-yr data revealed a significant difference between the clinical ratings of Dyract and Fuji II LC for all criteria except marginal integrity and marginal discoloration in enamel. A total of five Dyract restorations and four Fuji II LC restorations failed within the study period.A considerably high and almost similar overall failure rate was found for both restorative materials in restoring non-carious cervical lesions. However, retained Dyract restorations presented superior clinical performance compared to Fuji II LC restorations.

To evaluate the effect of cavity preparation device (i.e. carbide bur, diamond bur, air abrasive, Sonicsys, and Er:YAG laser) on microleakage of Class V resin-based composite (RBC) restorations, the tooth restoration interface and dentin ultrastructure.Eight groups (n=18) of 9 human molars each were prepared on the buccal and lingual surfaces for Class V RBC restorations using five preparation devices: Group 1: carbide bur #8 round; Group 2: diamond bur #801; Group 3: KCP 1000 air abrasion unit; Group 4: Sonicsys Approx.; Group 5: Er:YAG laser; Group 6: carbide bur plus air abrasive; Group 7: carbide bur plus laser, and Group 8: laser without etching. Circular cavity preparations, 3 mm diameter and 1.5 mm deep, were cut at the CEJ on the buccal and lingual surfaces of each tooth. A bevel was placed on the enamel margin. All cavities were restored using Single Bond adhesive system and Silux Plus according to manufacturer's directions, with the exception that no etchant was used with Group 8. After thermocycling, specimens were tested for microleakage using silver nitrate methodology. Dye penetration data was collected and subjected to non-parametric statistical analysis. SEM analysis of the dentin-RBC interface, as well as the effect of each cavity preparation device on the surface of dentin was performed.Microleakage did not occur in enamel for any method of cavity preparation when the enamel was etched prior to adhesive application. Kruskal-Wallis One-way ANOVA revealed differences in dentin microleakage among the tested groups at alpha=0.05. Mann-Whitney Test for Paired Comparisons at alpha=0.01 demonstrated significantly greater dentin microleakage in Groups 5 and 7 compared to all groups except Group 3 (air abrasive). Although Group 8 had the least microleakage in dentin, this was not significantly different from Groups 1, 2, 4 and 6. SEM observation revealed hybridization at the dentin-resin interface for all groups except for Group 8 (laser prepared, non-etched). Surface analysis showed differences related to the operating mode of each preparation device.The method of cavity preparation did not affect microleakage in etched enamel. When using a laser for cavity preparation, enamel should be etched although not etching dentin may help to improve marginal seal.