ot te of ty Se oo wvanbzarion “o @ ciassionower “AND. COMGSITE RESIN : GIOMER: (een. * STEGHNGISEY Giomer International Meeting / Japan Proceedings SHOEYING. a. haGiomer International Meeting / Japan Proceedings Suday, Ist July, 2001 International Conference Room Makuhari Messe Chiba, Japan Presidents /Editors Hideo Onose, D.D.S.,D.D.Sc. Professor and Chair. Department of Operative Dentistry School of Dentistry, Nihon University (Japan) Ivar A.Mjér, B.D.S.,M.S.D., M.S., Dr. Odont. Professor Department of Operative Dentistry Colleve of Dentistry, University of Florida (U.S.A.)Greetings : Katsuya Ohta (President of SHOFU INC.) ...... Opening Address Special Lecture : Nairn H.F. Wilson ‘The University of Manchester (U.K.) Hideo Onose, President (Nihon University) “Practitioners” Reasons for the Replacement of Restorations” Session 1 : Characteristics of GIOMER Products + Ultrastructure of Reactmer Bond, a Glass lonomer-based Adhesive Franklin R. Tay University of Hong Kong (China) .. + Analysis of Resin-Dentin Interface with Laser Raman Microscopy Masashi Miyazaki Nihon University (Japan) .. + A Laboratory Evaluation of the Water and Fluoride Equilibrium Characteristics of Reactmer John F. McCabe University of Neweastle (UK) + Biocompatibility of Restorations with GIOMER Junji Tagami Tokyo Medical and Dental University (Japan) + Discussion: Session 1... Session 2 : The Behavior and Effect of Fluoride * Diffusion of Fluoride lon from GIOMER Products into Dentin Toshio Teranaka Kanagawa Dental College (Japan) + Effect of Fluoride-Releasing Adhesives to Dentin -Lnhibition of secondary caries and remineralization of decalcified dentin- Masahiro Yoshiyama Okayama University (Japan). + Fluoride Recharge Effect of Various Dental Materials: Fluoride Release and Inhibition of Caries Progression Katsushi Okuyama Hokkaido University Japan)3: The Clinical Performance of REACTMER + FE-SEM Observation on the Interfacial Ultrastructure of The Shofu Reactmer Restorative System and Its Clinical Application Toshimoto Yamada Toranomon Hospital (Japan) ........csccceseceeeereenereteeeeenenne Ob * Clinical Evaluation of Reactmer Restoration Takashi Matsuo The University of Tokushima (Japan) .........2..... REA periayeiiare eR Aca eee 88 * Reactmer Restorative Material: Its Clinical Reliability in Downsizing (MILART) Treatment Ko Hinoura Hinoura Dental Office (Japan) ......sseseecsssesessrseseermerseeseuneeesrenecsesseestesesescerneesrneeereene OD eADISCUSSION= SESSIOTS hae. Oimeme ienin atime Aum ERE SINE SDN Oriente MIE TIN ocenrynane nant 94 4: The Clinical Performance of BEAUTIFIL * Clinical Performance of Class I and Class V Giomer Restorations (6 month evaluation) Michelle C. Sunico University of the Philippines Manila (Philippines) . . 101 * Two-centre Evaluation of Fluorobond-bonded Class I and Class II Restorations of Beautifil: Two-year findings Nairn H.F. Wilson University of Manchester (U.K.) ...2..2-.:c::.:cesescsecsececeteneeeeescnsseeeneenenenes 108 » Eighteen Month Clinical Evaluation of Beautifil and Silux Plus Restorative Materials and a Non-restored Cervical Erosion Lesion Bruce A; Matis Indiana, University (U:S:A9) scan sieeiecnieneonacdawaccaemsmees 172 SS PDISCUSSIOM? SESSION, A eaacscyaraeescot eat eced ae snuses cers sens esec Uae sb wot Gea svesra Stare sceaneve nounced cavevenwecars 117 Discussion: Hideo Onose, president ......0...:.cescseceresetersisensseneesenenens oneeesenenseseseneeaessaenenanenene 125 ddress: Ivar A. Mjér, PresidentGood morning, my name is Ohta, president of SHOFU. On the opening of the international research meeting, 1 would wish to say a few words. This is the second meeting to the first Giomer conference held in Washington in April last year. Around 100 doctors have gathered here for this conference from all over Japan and outside of Japan. I should like to express my deepest appreciation for all of you. Many of you here, | believe, attended the 4-day IADR meeting held here in Makuhari. In such a tight schedule, I am very happy you are attending this meeting today. Especially, I wish to express my deepest appreciation for approximately 20 clinical doctors, who are in leading positions in the daily clinical practice in Japan, to be with us during your precious holiday. In holding this conference, we owe a great deal to support and cooperation from Professor Tagami of Tokyo Medical and Dental University and many other doctors who have been doing research on Giomer. Especially, 1 would like to extend my gratitude to Professor Onose of Nihon University and Professor Mjér of University of Florida, who have kindly accepted the presidency of this meeting. Furthermore, I would like to express my greatest gratitude to Professor Wilson of The University of Manchester, who has so kindly accepted to give a special lecture at this conference. It was in 1985 that we launched the light-cure composite into the Japanese market, which was first developed in Japan. At the same time, we have been supplying various unique glass-ionomer products. As the result of these activities, we strongly recognized the importance of developing such material that has the properties of glass-ionomer and composite, which lead to the development of PRG technology. At first, we applied PRG technology for fluoride source to Fluoro Bond. But recently, using this technology, we have completed the development of new restorative materials named Beautifil and Reactmer. Then we named the “dental resinous products with glass-ionomer property”, which are characterized by PRG technology, as “Giomer”. Among the Giomer’s basic properties, we, SHOFU, believe the feature, which is to strengthen the tooth structure, is going to be an extremely essential requirement for restorative materials from now on. Through many reports of long-term research regarding Giomer, which will be delivered today, and the deliberations and discussions, SHOFU should like to find the way to further contribute to the advancement of the dental care worldwide. On behalf of the sponsor, I wish to express my deepest hope that this conference will he meaninetil academically for all af vou who are here todav Thank vou very much.Opening Address Dr. Hideo Onose: President Ladies and gentlemen, this is my great honor to declare the opening of the second Giomer International Meeting, which is being held for the second time in Japan. And I would like to welcome many distinguished researchers and clinicians who have come here from all over the world to join this meeting, Trecognize that many of you here today also attended the first meeting at Washington DC last year. Since then, PRG technology has been significantly improved, as you know: Looking at today’s program, | understand that many interesting reports on PRG technology will be discussed during this meeting. I’m sure these reports will bring many interesting, constructive, and enlightening comments. Our program for the only one-day meeting is very intensive, 50, | would like to ask you to make your remarks brief and concise to enable us to cover as many comments as possible. During the breaks between sessions and at the social gathering after the meeting, you will have chances to get to know each other. We hope your exchange of information will greatly contribute to a better understanding of PRG technology. Finally, let me once again extend our warm welcome to all of you, and we hope that deliberation during this meeting will be fruitful. Thank you. wSpecial Lecture PRACTITIONERS' REASONS FOR THE REPLACEMENT OF RESTORATIONS Nairn H.F.Wilson, Ph.D.,M.Sc., B.D.S., F.D.S., D.R.D., FE.G.D-P. (UK.). Professor / Department of Restorative Dentistry Tumer Dental School University Dental Hospital of Manchester (U.K.)Special Lecture Practitioners’ reasons for replacing restorations. Nairn H. F. Wilson, Ph.D., M.Sc.. B.D... ED.S., D.R.D., FE.G.D.P (UK). Dental Institute, King’s College London, formerly University Dental Hospital of Manchester (UK.) ABSTRACT Despite many important improvements in recent years in materials, equipment and operative techniques, most restorations placed in the “real world” environment of general dental practice continue to have a relatively limited lifespan. The principal reason for the replacement of restorations continues to be secondary caries, as diagnosed clinically. The reasons for these circumstances are many and varied, including: lack of criteria and systems to reliably diagnose significant deterioration in the clinical acceptability of restorations, notably the onset of secondary caries; outdated teaching; a fuilure to develop, validate and promote the repair and refurbishment of restorations as an alternative to replacement, and a range of related factors which perpetuate traditional approaches to the provision of routine dental care, Central to modernizing existing approaches to the replacement of restorations — activity which accounts for >60% of all restorative work undertaken, are new knowledge and understanding of secondary caries and ways in which to predict, prevent and, where necessary, counter the deterioration of restorations in clinical service, Such developments should be linked to a more preventatively orientated approach to routine dental care, whereby clinicians are rewarded more for preserving rather than repeatedly replacing tooth tissues with restorations of ever-increasing size and complexity. Ongoin: investigations on reasons for the replacement of restorations provide valuable information on trends in primary dental care, highlighting priorities for future research. INTRODUCTIO! Over a period of 50 years, numerous surveys have been conducted on reasons for the placement and replacement of restorations.! Although subject to certain limitations", the data from such surveys extend existing knowledge and understanding of patterns of provision of dental care, including tends in the use of restorative ‘materials. The importance of such new knowledge and understanding may be considered to be unquestionable given that the replacement of existing restorations comprises c. 60% of all restorative work undertaken” — treatment which, as a consquence, accounts for a very large proportion of the global spend on oral healthcare provision. There are many, varied interrelated reasons for the replacement of restorations - decision-making in clinical practice not being an exact science. In addition, new evidence as to the nature of secondary caries'® — the principal reason for the replacement of restorations, has added substantial new justification for considering investigations of practitioners reasons for the replacement of restorations and related issues to be an important esearch priority This contribution to the existing literature on reasons for the replacement of restorations brings together an overview of the findings of surveys on the reasons for the replacement of restorations, considers the impact of new thinking in the ficld, and highlights the need for renewed research efforts to better understand andimprove decision-making in relation to existing restorations. Such contributions to the existing literature are considered timely given the growing interest in setting aside traditional approaches to, in particular, operative dentistry, in favour of a more patient-centred, preservative approach to restorative dental care. Findings of surveys As highlighted by Deligeorgi er al,' in the most recent overview of reasons for the placement and replacement of restorations, secondary caries, as diagnosed clinically, has been consistently found to be the principal reason for the replacement of restorations, both in anterior and posterior teeth and irrespective of the restorative material forming the restoration. Other common reasons for the replacement of restorations include marginal deterioration, bulk fracture, poor anatomic form, fracture of some element of the remaining tooth structure and, for tooth-coloured restorations, marginal and bulk discoloration. In many surveys, notably the most recent studies, the number of restorations replaced as a consequence of secondary caries has nearly matched the number of restorations replaced for all other reasons‘. This finding is cause for concern given the continuing uncertainty and lack of consistency in the clinical diagnosis of secondary caries.'! Furthermore, this finding places all the more importance on the consequences of the conclusion of Mjér and Toffenetti!” that secondary caries, as most commonly found in association with the cervical margins of restorations, is in effect a new primary lesion of caries located adjacent to a restoration rather than a lesion of caries comprising an outer element and a wall section joining at the cavosurface margin of the affected restoration.” While reports of <60% of restorations having been replaced as a consequence of secondary caries are typical of cross-sectional studies in general dental practice, secondary caries in controlled clinical trials is typically found to account for the failure of only about 3% of restorations per annum after three or more years in clinical service. There are various explanations for this dramatic difference in the findings of the two types of study; however, it has been suggested that the principal reason relates to different criteria being applied by practitioners and clinical investigators". The adoption and application of common criteria could be anticipated to have a profound effect on both reasons for the failure of restorations and restoration longevity, Criteria for secondary caries Traditional teaching reinforced by long-held views on, for example, microleakage has encouraged practitioners to consider restorations with clinically detectable marginal gaps, with or without associated staining or discoloration, to have failed, or be failing as a consequence of secondary caries. Contrary to the thinking behind such an approach, a recent study by Ozer" found that the size of the gap between a restoration and tooth has no influence on the initiation of secondary caries, unless the gap size exceeds 250 um, and only then if the gap is in an inaccessible position in which oral hygiene procedures have little, if any, effect. The decisive factor in the initiation of secondary caries, according to Ozer", is plaque accumulation in and around the caries susceptible gap. Such criteria may be considered to be of particular importance in helping to distinguish between arrested and active, progressive secondary caries. As concluded by Mjér and Toffenetti'’, diagnostic radiography may be found to be of limited value, if there is any shadowing effect caused by existing restorations. However, in radiographic studies involving sound and defective restorations, for example, the investigations of Hewlett et a/.!, the incidence of secondary caries may be found to be three-fold higher for defective restorations than for sound restorations.Confounding factors Notwithstanding the widely recognised limitations in the diagnosis of secondary caries in everyday clinical practice, other factors may encourage practitioners to consider restorations with less than ideal characteri ies as having failed, or to be failing. Increasing risks of litigation stemming from allegations of a failure to diagnose and treat disease, funding systems which encourage a “drill and fill” approach to patient care, and the expectations of patients based on previous dental treatment may, together with other interrelated factors associated with the demands of present day clinical practice, adversely influence decisions to replace restorations. Under such circumstances, practitioners may be found to be more inclined to replace rather than accept certain restorations as clinically acceptable, and if asked for a reason for making such decisions, it is suggested that secondary caries is the most likely reason to be given. ‘The way forward To limit the extent to which restorations may continue to be unnecessa replaced in the belief that they are failing as a consequence of secondary caries, it is suggested that it is necessary to: review existing teaching in relation to secondary caries; find ways to promote the refurbishment and repair rather than the replacement of restorations with less than ideal characteristics,'" and replace the “drill and fill” approach to restorative care with a preservative “predict and prevent” philosophy to long term patient management". The difficulties of realising such goals, let alone the difficulties of establishing new criteria for the failure of restorations, are not to be underestimated; however, success in such endeavours would, itis suggested, be of immense benefits to existing and future generations of patients. In the meantime, further developments in restorative materials and operative techniques may enhance the performance of restorations in clinical service, and thereby positively influence restoration longevity. At one and the same time certain changes may be observed in the distribution of reasons for the failure of restorations. CONCLUSION The most common reason for the replacement of restorations is secondary caries as diagnosed clinically, albeit that the scientific basis for such a diagnosis is meagre in all but those situations in which validated criteria for secondary caries are satisfied. Further studies on practitioners’ reasons for the replacement of restorations are important in developing and encouraging the adoption of preventative, minimal invasive approaches and techniques aimed at extending the longevity of restorations in clinical service. REFERENCES + ‘Deligeorgi, V, Mj ; ILA. and Wilson, N.H.B. An overview of reasons for the placement and replacement i of restorations. Primary Dental Care 2001 + *Swift EJ, Bader, JD and Shugars DA. Glass-ionomer cement restorations and secondary caries (letter), 82. + ‘Forsten L. Glass-ionomer cement restorations and secondary caries: a preliminary report (letter). Quintessence International 1996; 27: 581-5 Quintessence International 1996; 27: p.791 : + ‘Mount GT. Glass-ionomer cement restorations (letter). Quintessence International 1996; 27: 582-583. tem. British Dental Journal + ‘Nuttall N. Financial implications of the 1985 GDS patient charging sy 1984; 159: 375-376. + ‘Quist V; Thylstrup A and Mjér IA. Restorative treatment pattem and longevity of amalgam restorations. 10 in Denmark. Acta Odontologica Scandinavia 1986; 44: 343-350, 7Qvist V, Thylstrup A and Mjér IA. Restorative treatment pattern and longevity of resin restorations in Denmark. Acta Odontologica Scandinavia 1986; 44: 351-259. ‘Maryniuck GA and Caplan SH. Longevity of restorations: survey results of dentists” estimates and attitudes. Journal of the American Dental Association 1986; 112: 39-45. "Klausner L and Charbeneau G. Amalgam restorations: a cross-sectional survey of placement and replacement. Journal of the Michigan Dental Association 1985; 67: 249-252. "Mijér, LA. and Toffenetti, FE. Secondary caries: A literature review with case reports. Quintessence nternational 2000; 31: 165-179. "Bader JD and Shugars DA. Understanding dentists’ restorative treatment decisions. Journal of Public Health Dentistry 1992; 52: 102-110. "Ozer L. The relationship between gap size, microbial accumulation and the structural features of natural caries in extracted teeth with Class IT amalgam restorations (thesis). University of Copenhagen, 1997. “Hewlett ER, Atchinson KA, White SC and Flack V. Radiographic secondary caries prevalence in teeth with clinically defective restorations. Journal of Dental Research 1993; 72: 1604-1608. “Wilson, NHF, Setcos, JC and Brunton P. Replacement or repair of dental restorations. In: Advances in Operative Dentistry. Vol 2: Challenges of the future. Quintessence, London (in press). 'Anusavice KJ. Treatment regimens in preventive and restorative dentistry. Journal of the American Dental Association 1995; 126: 727-740.Session | Characteristics of GIOMER Products Chairperson John F. McCabe, B.Sc., Ph.D., D.Sc. Professor of Dental Material Science Department of Restorative Dentistry ‘The Dental School University of Newcastle (U.K.)Ultrastructure of Reactmer Bond, a glass ionomer-based adhesive Franklin R. Tay, B.D.Sc.(Hons), Ph.D. Franklin R. Tay! ,Hidehiko Sano”, Junji Tagami*, Masanori Hashimoto?, Keith M. Moulding‘, Cynthia Yiu!, David H. Pashley? Conservative Dentistry, Faculty of Dentistry, The University of Hong Kong (China) 2 Department of Operative Dentistry, Hokkaido University School of Dentistry (Japan) * Department of Operative Dentistry, Tokyo Medical and Dental University (Japan) + Materials Characterisation and Preparation Facility, The Hong Kong University of Science andTechnology (China) © Department of Oral Biology and Maxillofacial Pathology, School of Dentistry, Medical College of Georgia, Augusta (USA) Abstract Objective Reactmer Bond (Shofu Inc., Kyoto, Japan), is a glass ionomer (GI) based, tri-curable, all-in-one, filled adhesive. Both fluoroaluminosilicate glass (FASG) and fully pre-reacted glass (F-PRG) are used as fillers. This study examined the ultrastructure and elemental composition of resin-dentine interfaces that were treated with this adhesive. Materials and methods Dentine disks prepared from human third molars were abraded with either 600- or 60- grit SiC paper to create smear layers of different thickness. They were bonded using Reactmer Bond. Cryo- fractured dentine surfaces devoid of smear layers were also bonded by chemical-activation and GI reaction without additional light-activation, or allowing the Gl reaction to proceed for 1 min before the adhesive was applied and light activated. Undemineralised and demineralised sections were processed for TEM examination and STEM/EDX analysis. Resulty. Resin-dentine interface from specimens with smear layers consisted of a mineral-dense surface layer that resided on top of a partially demineralised dentine. The partially demineralised zone was considerably thicker in the 600-grit than the 60-grit specimens, In smear layer-free specimens that were cured by chemical- activation/GI modes only, the surface layer concurred with the partially demineralised zone, and appeared as an electron-dense layer over the undemineralised intact dentine, Smear layer-free specimens that were cured by the light-activation of the partially neutralised adhesive contained incomplete amorphous surface layers only. Apatt from colloidal silica, FASG fillers were the predominant filler type within the resin matrices, Peripheral hydrogel layers that contained electron-dense “seeds” were found around the FASG fillers. F-PRG fillers were only sparsely observed. In specimens that were laboratory demineralised with formic acid, phase separation of the unstained resin matrices into electron-dense and electron-lucent domains occurred. Artefactual dendritic deposits were found within the electron-dense domains. Significance The presence of a surface interaction layer on top of a partially demineralised zone along the resin-dentine interface suggests that either a Gl-type reaction of, precipitation of insoluble carboxylate salts around remnant apatite crystallites may occur when this single-step adhesive interacts with dentine. Appearance ofartefactual dendritic deposits suggests that continuous ion movement is possible within the hydrophilic portion of the resin matrix in this fluoride-releasing adhesiveIntroduction Contemporary self-etching adhesive systems are important additions to the clinician’s bonding armamentarium. As smear plugs are not removed prior to the application of these adhesives, the potential for, post-operative sensitivity that is caused by incomplete resin infiltration of patent dentinal tubules can be substantially reduced. Moreover, as water is an essential component of these systems to dissociate protons to enable demineralisation of hard dental tissues, the technique sensitivity associated with variable hydration of a demineralised collagen matrix is also eliminated, They are also user-friendly in terms of the reduced number of steps required for bonding. Self-ctching adhesives may be classified into two-step self-etching primers, and single-step all-in-one adhesives. Self-etching primers contain hydrophilic monomers, as well as increased concentration of organic acids or acidic adhesion promoting monomers to etch and prime the tooth substrates.' However, they require a second layer of strength-conveying bonding resin with more hydrophobic resin components in order to couple the resin composites to the primed enamel and dentine. To maintain adequate shelf-lives, the recently introduced all-in-one adhesives are also supplied as two-component assemblies, They are mixed together immediately before use and the mixture of hydrophilic and hydrophobic resin components is then applied to the tooth substrate in a single-step application. Examples of these all-in-one adhesives include Prompt L-Pop (ESPE, Seefeld, Germany),’ One-Up Bond F (Tokuyama Corp., Tokyo, Japan)‘, Reactmer Bond (Shofu Inc., Kyoto, Japan), Etch&Prime 3.0 (Degussa AG, Hanau, Germany), AQ Bond (Sun Medical, Kyoto, Japan) and Xeno CF Bond (Sankin, Tokyo, Japan). One-Up Bond F and Reactmer Bond may further be classified as glass ionomer-based adhesives,® as they contain ion-leachable, basic, fluoroaluminosilicate glass (FASG) in the adhesive formulations. Similar to polyelectrolytes in glass ionomer (GI) cements, resin monomers with carboxylic groups in these adhesives, such as methacryloxyundecane dicarboxylic acid (MAC-10) and 4-acryloxyethyltrimellitic acid (4-AET) are proton donors. They can react with the ion-leachable glass (proton acceptors) in the presence of water to initiate a Gl-type reaction. Like polyalkenoic acids that can bond chemically to hydroxyapatite’ or collagen’, 4- AET has also been shown to bond chemically to both dentine apatite and collagen.’ This provides the potential of chemical bonding via an ion-exchange process between the glass particles and partially demineralised tooth substrates.!0"! In addition to conventional FASG fillers, Reactmer Bond utilises a novel filler material known as fully pre-reacted glass polyalkenoate fillers (F-PRG) to enhance the sustained fluoride-releasing and recharging potential of the material.’ These fillers are formed by the complete reaction of FASG glass with polyalkenoic acids in the presence of water to form a wet siliceous hydrogel. Upon freeze-drying, the desiccated “xerogel” is further milled and silane-treated to form F-PRG fillers of a specific size range.! It was proposed that of the use of F-PRG fillers promotes rapid fluoride release through a ligand exchange within the pre-reacted hydrogel? In order to be an effective self-etching adhesive, Reactmer Bond should demonstrate the potential to etch through thick smear layers and demineralise the underlying intact dentine for micromechanical retention." The manufacturer claims that apart from the purported acid-base GI-type reaction, both redox chemical initiators and photoinitiators are included in this adhesive to make it tricurable, but this hypothesis needs testing. In order to be truly tricurable, the adhesive should not only be able to polymerise in the absence of light, but should exhibit evidence of a surface interaction layer along the adhesive-dentine interface that is similar to those found in conventional GI cements.'' Furthermore, the polymerised resin matrix should also demonstrate evidence of ion diffusion in order to justify the claim that it is a fluoride-releasing adhesive.'? This study investigated the 14interfacial ultrastructure and elemental composition of sound dentine bonded with this single-step adhesive, using transmission electron microscopy (TEM) and scanning transmission electron microscopy/energy dispersive x-ray microanalysis (STEM/EDX). The null hypotheses tested were: 1) there is no difference in the ultrastructure of resin-dentine interfaces when the adhesive is applied to dentine abraded with silicone carbide (SiC) paper of different roughness, and 2) there is no difference in the ultrastructure of resin-dentin interfaces when the adhesive is used in the light-activation mode or chemical activation/GI mode. Materials and Methods Tooth preparation ‘There were four experimental groups, two containing smear layers of different thickness, and two {ayer-free groups that were cured by either the light-activation mode or the chemical-cured/GIC mode. Sixteen noncarious, human third molars, stored in 0.5% chloramine T at 4°C, were used within one month following, extraction, For the smear layer groups, the occlusal enamel was first removed using a slow-speed saw with a diamond-impregnated disk (Isomet, Buehler Ltd., Lake Bluff, IL, USA) under water lubrication, A dentine disk of about 1.0 — 1.5 mm thick was prepared from each tooth us ng the same saw. In Group F-600, the bonding surfaces of four dentine disks were abraded with 600-grit SiC paper under running water, to create a relatively smooth surface with a thin smear layer between 0.4-0.7 jm thick,'* In Group F-60, the same number of disks were ground with 60-grit SiC paper to create rough bonding surfaces with coarse, thick smear layers between 2.9-4.1 um thick."* The use of rougher SiC papers such as 180- or 60-grit for abrading dentine was found to be more representative of dentine that were prepared clinically with the use of medium to coarse diamond burs." Each of the smear layer-free groups consisted of two teeth that had circumferential grooves placed around the teeth through enamel into peripheral dentine. They were submerged in liquid nitrogen for 10 min, and fractured while frozen with a hammer and scalpel blade. These specimens had no smear layer on their bonding surfaces. We realized that the use of dentine bonding surfaces devoid of smear layers was far removed fiom what was practiced clinically. The rationale for bonding to these surfaces was to eliminate the possibility ‘of mistaking a resin-infiltrated smear layer for the surface interaction layer that may be formed between dentine and the GI component of the adhesive. In Group CC, the adhesive was applied to the fractured dentine and allowed to cure completely in the absence of light. It was anticipated that both the self-etching effect and that produced by Gl interaction with dentine would be maximised and readily recognised in the absence of a pre- existing smear layer, In Group LC, the adhesive was mixed and left in the dark for the acid-base reaction to proceed for | min before applying to dentine. It was then polymerised in the light-activation mode. The rationale for using this protocol was to minimise the self-etching effect of the adhesive with the use of this “neutralised” adhesive, so that any observed interaction between the adhesive and dentine would represent that contributed by the GI phase of the adhesive. 15Bonding procedures The general composition and batch number of the all-in-one adhesive are listed in Table 1. Equal droplets of the Reactmer Bond bonding agent A (white liquid) and the bonding agent B (amber liquid) were dispensed into a mixing well. They were thoroughly mixed for 5 s until a homogenous liquid ruintnie was obtained. For groups S-600 and S-60, the all-in-one adhesive mixture was then applied to each bonding surface and left undisturbed for 20 s. Without rinsing, the adhesive was gently air-dried to evaporate the solvent and spread the mixture evenly. Each bonded surface was light-cured for 20 s using a halogen light- Table 1 Composition of the all-in-one adhesive used in the study Bottle Composition Lot number Procedures Reactmer Bond A F-PRG PASG (Shofu Inc,, Initiators Kyoto, Japan } Water Acetone 040002 a; b (208); c; d (20s); ¢ Reactmer Bond RB 4-AET 4-AETA (Shofu Inc., HEMA Kyoto, Japan ) UDMA Photo-initiator Procedures - a: mix all-in-one adhesive; b: apply adhesive mixture without agitation: ¢: gently air-dry; d: light-cure (suggested); e: place restoration, light-cure and polish. (Abbreviations: F-PRG: full-reaction type pre-reacted glass ionomer fi FASG: fluoroaluminosilicate glass; 4-AET: 4-acryloxyethyltrimellitic acid; 4-AETA: 4-acryloxyethyltrimellitate anhydride: HEMA: 2-hydroxylethyl = ; : ; ; : -thacrylate; UDMA: urethane dimethacrylate). curing unit (Variable Intensity Polymeriser, Bisco, 9 "*1y!#*e: UPMA: urethane dimethacrylate) Schamburg, IL, USA), with the curing intensity set at 500 mWem?®. For group CC, the adhesive-coated teeth were left in the dark for 15 min to enable the adhesive to chemical cure in the absence of light, and for any GI reaction to proceed to completion. For group LC, after the adhesive components were mixed, they were left in the dark for | min. The slightly viscous mixture was then applied to the fractured tooth surfaces, left undisturbed for 20 s and then light-cured for 20 s. After bonding, dentine disks from each group were restored using a thin layer of light-cured lining composite (Protect Liner F, Kuraray Co. Ltd., Osaka, Japan). TEM preparation Specimens were aged in 100% relative humidity for seven days. Two strips, each approximately 2 mm wide, were sectioned longitudinally across the bonded interface of each specimen. Half of the strips in each group were examined undemineralised. The other half were completely demineralised in an aqueous solution of 8 M formic acid that was buffered with sodium formate to a pH of 2.5. After demineralisation, the specimens were fixed in Karnovsky’s fixative (2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.3) for 1 h and rinsed thoroughly with 0.1 M sodium cacodylate buffer. They were dehydrated in an ascending ethanol series (30% to 100%), immersed in propylene oxide as a transition fluid and embedded in epoxy resin (TAAB 812 resin, TAAB Laboratories, Aldermaston, UK). 90 nm thick undemineralised sections were prepared with an ultramicrotome (Reichert Ultracut S, Leica, Wein, Austria) and collected on single slot, carbon- and formvar-coated copper grids (Electron Microscopy Sciences, Fort Washington, PA, USA). 70 nm thick demineralised sections were similarly prepared and retrieved with 150-mesh copper grids (TAAB laboratories, Aldermaston, UK). Both undemineralised and demineralised sections were examined unstained. They were examined with a transmission electron microscope (Philips EM2088, Eindhoven, The Netherlands) operating at 80-100 kV, Images were recorded with photographic films and also in form of digitised images. The latter were done using the charge couple device (CCD) camera (Bioscan, Model 792, Gatan Inc., Pleasanton, CA, USA) attached to the microscope. Five digitised images taken from undemineralised sections of each group were imported into an image analysis software (NIH Image 1.60, Scion Corp., Frederick, MD, USA). The thickness of the surface interaction layer and the partially demineralised zone above the intact, undemineralised dentine were measured from two positions in each micrograph. 16STEM/EDX analysis After TEM examination, undemineralised, unstained sections were further coated with carbon and examined with a scanning transmission electron microscope (STEM; Philips CM20, Eindhoven, The Netherlands) operating at 200 kV and equipped with an EDX system (ISIS Pentafet, Oxford Scientific Instruments Corporation, High Wycombe, UK). Area mappings were performed for elements present along the resin-dentine interfaces. Spot analyses were also done on the FASG fillers as well as other sites of interest. Results In undemineralised sections, the resin-dentine interface from specimens with smear layers consisted of an electron-dense, calcium and phosphorus rich surface layer that resided on top of a relatively electron-lucent zone of calcium and phosphorus sparse, partially demineralised dentine (Figure 1). Remnant apatite crystallites ‘were also found within the electron-dense surface layer. The partially demineralised zone, which contained sparsely distributed apatite remnants, was considerably thicker in Group S-600 (Figure 1) than in Group S-60 (Figure 2), In smear layer-free specimens that were cured by chemical-activation/GT modes only (Group CC), the surface layer concurred with the partially demineralised zone, and together they form a thick electron- dense layer over the undemincralised intact dentine (Figure 3), Smear layer-free specimens that were cured by the light-activation of the partially “neutralised” adhesive (Group LC) contained incomplete amorphous surface layers only (Figure 4), The mean thickness of the electron-dense surface layer and the relatively electron-lucent pattially demineralised zone in the four groups are summarised in Table 2. Apart from colloidal sili: The glass cores of these FASG ZASG fillers were the predominant filler type within the resin matrices. llers contained silicon, aluminium, calcium and lanthanum as the major clements. Peripheral hydrogel layers that contained electron-dense “seeds” were found around the FASG fillers, F-PRG fillers were only sparsely observed and did not contain electron-dense “seeds” (Figure 5). In specimens that were laboratory demineralised with formic acid, phase-separation of the unstained resin matrices into electron-dense and electron-lucent domains was observed (Figure 6). Artefactual dendritic deposits were found within the electron-dense domains, Fig. 1A 4 ow magnifivation TEM avistograph of a6 unstained, Codeminraned pein flowing he pplication of Bondo. Sc papers sound dene eoap S40) The sl-ching eft fe aide ein monomer FAET ean evil ye pesence fa alae classic part) ‘ened ne(8) ob pat iicunderminersivedntetbalat ins (Cy Poon ha mri, «move electron-dense ‘Sac er) soul be ested etwcen the asi (M4) Sth pally demoed Tye (8) As hese aye has {oe Genieralises before ns deine reshe he sect tease src lager camo be eased bp the demineraistion oss don, Th eid aes resin tM) een Tics an oraine PASG filets 5 wel a cro lusts, deminciss pein zhong the aplication of React ond to ori SiC pape bad sound detineazop 5-0 Sin othe S-600 group, a clestioadens arses per) canbe sum on to ft mach tines partly deminer ‘eons (1). L: unkominralisinterbai enine: Meade res mati, Bar= 180m,Discussion For effective micromechanical retention, a self-etching primer must be able to penetrate beyond the smear layer into the underlying intact dentine. It-has been shown that hybrid layers created by self. -etching primers that are as thin as 0.5 lm can achieve good initial tensile bond strength in sound dentine.!? The all-in- one adhesive investigated in this study produced partially demineralised zones of variable thicknesses in intact dentine when dentine surfaces were abraded with different grits of SiC paper (Figures 1A and 2). In the 600 grit group (S-600), a partially dennnerabeed zone with a mean thickness of 0.7 1m was consistently observed. However, in the S-60 group, the mean thickness of this zone was only 0.1 tm thick and was absent in some specimens (Table 2). This leads to a rejection of the first null hypothesis. Our results were similar to those of Koibuchi et al.,'° in that the bond strength of this adhesive to dentine will be compromised when thick smear layers are encountered. In this study, the adhesive mixture was left on the dentine surface without agitation, according to the manufacturer’s instructions. We found better bonding is achieved by agitating the adhesive during application, that brings fresh adhesive to replace the buffered layer of adhesive that is in contact with dentine. Our microtensile bond strength results showed that when bonded to thick smear layers, this all-in-one adhesive achieved 25,8+3.6 MPa when left undisturbed on dentine, according to the manufacturer’s instructions, and 38.6+6.5 MPa with continuous agitation. The difference may be caused by deeper and more consistent demineralisation of the underlying intact dentine in the presence of thick smear layers, Fig. 1B mf, STEM/EDX area mapping of the resin-dentine interface from an unstained, undemineralised section similar to the one shown FSnne Lanthanurn in Figure 1A. The two layers that are shown in the bright field i ne image can be discerned in the elemental mapping results. The surface layer (X) is rich in calcium and phosphorus while aluminium, silicon, lanthanum and fluorine are only detected minimally. There is also less calcium and phosphorus in the underlying partially demineralised zone (B). Aluminium Preece Silicon The presence of a mineral-dense, surface layer between the adhesive and the partially demineralised zone was difficult to interpret in these two groups. As the adhesive was applied on smear layer-covered dentine, one could not exclude the possibility that such a layer represented smear layer remnants that were trapped along the resin-dentine interfaces. Even so, it was unusual that the mineral density in the interface was the reverse of what was normally seen when self-etching primers were applied to smear layer-covered dentine.” As the smear layer is loosely arranged,"’ and has to be demineralised before intact dentine is reached, there should be an increasing gradient of mineral density from the dentine surface down to the demineralisation front even if smear layer remnants are present STEM/EDX analysis also confirmed that a high concentration of calcium and phosphorus was present in this surface layer (Figure 1B). Thus, the increased mineral density in the surface layer could not be solely explained by the retention of smear layer remnants. This was further supported by the observation of a similar layer when the adhesive was bonded to cryo-fractured dentine that were devoid of a smear layer. 18Be One possible interpretation of the “Thickness ofthe electron-dense surface layer and relatively electron surface layer observed in this study is that it Ineent, partially demineraised zone from undemineralised sections represents the ultrastructural manifestation of the | Graap Sorta ner ae interaction between the GI component of the san oss0.25 oresos adhesive and dentine. Wesenberg et al. reported sa sas0.44 ounce a similar surface layer of increased calcium and sen F e que phosphorus contents above a partially 1 ousen puna demineralised zone in when a GI cement was: FERS denincsalsed ze was decor in vo specimens Ge farout often measures) ‘Bat jer cosmid he sare poston ad appeared 6a eestton- dens ayer above iia undneraind dertine “Thess ler was cotinine specimens Ge tmorofeamesuencns) Ts pall demineaiss zone ss dentin usps {ig it tt ten mesures) bonded to dentine." Ina conventional GI cement, ions released through the interaction of FASG fillers and polyalkenoic acid can interact with calcium and phosphate ions extracted from the ‘dentine substrate to form an ion-exchange interphase.'! This interphase is morphologically manifested as an “intermediate layer" ora surface interaction layer along the Gl cement-dentine interface. Using acid conditioners ‘with different degrees of aggressiveness, we were able to observe the coexistence of demineralised dentine ith the surface i “conditioner employed, there was a downward migration of the interaction layer into the zone of demineralised eraction layers in a conventional Gl cement." Depending on the aggressiveness of the “dentine, with the possibility that the interaction layer corresponded with the depth of deminerlisation.” A “similar phenomenon was observed in the present study. When Reactmer Bond was used in the chemical-cure/ iI mode (Group CC), the electron-dense layer coincided with the zone of partially demineralised dentine ‘(Figures 3A and B). Fig. 3A, ‘A low magnificition TEM micrograph of an unstained aerated apecinen flowing the application of Reaemer ad facut dene Le. po aearlayen) The adesve ws Jett in he dark for 19 min to permit she adhesive 2 auto Dulpmerisandary il races to poceed to cumpstion Groep (CC). A 500 am ick elecees-deas layer DX is present long ‘hese of he ueninerliodisrobular dentine (U) No ‘Nigh magiiton micrograph fe interface in his group sowed ha he mrs layer) conc with he positon af be rly deminer 2ne Thi so continously unas ssdle-shaped apatite ste (arowbeod) tha te parialy ewinaried Ge th cennl Heda ling tie ‘Thee crystallites are surmeaded by gry, moephous deposits (coer). Me adneve ria matin. Bar 10 a. Pra dence zoe can be at he magic ‘This may be explained by the longer period available for the Gi reaction to proceed before the adhesive schemically-cured via free radical polymerisation. During this period, the acidic resin monomers could also ‘bane partially demineralised the intact dentine, The surface layer/partially demineralised zone in this smear layer-free group was similar in thickness to the partially demineralised zone found in Group S-600. This is sbably because only thin smear layers were present when dentine were abraded with 600-grit SiC papers.!* the thickness of the surface layer were different when this Gl-based adhesive was used in either the light- 19activation mode or chemical-activation/GI mode, we have to reject the second null hypothesis. This situation is comparable to what was reported for resin modified-G] cements, in which the acid-base reaction was extensively reduced by photopolymerisation during their early setting stages.” It is also interesting to see that a partially demineralised zone was almost completely absent when the mixed adhesive was left in the dark for | min before it as applied to cryo-fractured dentine (Group LC; Figure 4A). In a conventional GI cement, tartaric and polyalkenoic acids are completely neutralised within one hour of mixing when not placed on dentine.*! Presumably, the acidic adhesive mixture in Group LC was partially neutralised after | min, to a pH that was not aggressive enough to etch into intact dentine. Similar to resin modified G1 cement, it is possible that continuous formation of the GI phase may occur on maturation of the material.” This probably accounts for the observation of incomplete surface layers along the adhesive-dentine interface in Group LC (Figure 4B). Another possible explanation of the increased mineral density in the surface layer may be the precipitation of insoluble calcium salts of 4-AET around the partially dissolved apatite crystallites. SEM-X-ray micro- analyzer examination of the bonded interface of a self-etching primer containing the acidic resin monomer 4- AET indicated that much calcium remained at the interface.” In an aqueous medium, the two ionized carboxylic groups of the 4-AET molecule can react with the Ca*' cation of dentine apatite to form insoluble calcium carboxylate salts. It has been shown in a Raman spectroscopic study that formation of similar insoluble calcium carboxylate salts between 4-methacryloxyethyltrimellitic acid (4-MET) and enamel powder was possible.** However, this reaction occurred very slowly in a 4-META/methylmethacrylate-tri-7-butyl borane (MMA-TBB) adhesive, probably as a result of the poor ionization of 4-MET in the hydrophobic environment of MMA. On the contrary, formation of these carboxylate salts may be faster in the water-containing all-in-one adhesive investigated, due to better ionization of the 4-AET. This is similar to the formation of insoluble calcium polyalkenoate salts within the hydrogel matrices of water-based, conventional GI cements.*' This, however, does not imply that these carboxylate salts are chemically chelated to dentine apatites. It must be stressed that in the absence of chemical analysis, it is impossible to elucidate the bonding mechanism of an adhesive based on ultrastructural examination alone. Based on the morphological manifestation of a siliceous hydrogel layer Fig. 4A Fig. 4B eG - “= dentine surface in this A low magnification TEM micrograph of an unstained, A high magnification micrograph of the 20 undemineralised specimen following the application of partially “neutralised” Reactmer Bond (ic., adhesive mixture left in the dark for | min for GI reaction to oceur) to fractured dentine and then light-activaied after 20 s (Group LC), At this magnification, neither a surface layer nor a partially demineralised zone can be seen on the surface of the undemineralised intertubular dentine (U). Within the resin matrix (M), some of the FASG fillers (G)} are surrounded by thin hydrogel layers (arrowhead) that contain electron-dense "seeds". A region similar to the one labelled with the asterisk is shown at higher magnification in Figure 4B. P: peritubular dentine. Bar = 500 om. group. The surface of the undemineralised dentine (U) is incompletely covered with an amorphous surface layer (X) that contains electron-dense, needle-shaped apatite crystallites (arrowhead), Within the adhesive resin matrix (M), hydrogel layers are formed around the FASG glass fillers and contain electron- dense "seeds" (arrowhead). Bar = 200 nm_around the periphery of the PASG fillers, it is indisputable that a Gl-type acid-base reaction occurs between the acidic resin monomers and the glass fillers.2! However. the existence of either a Gl-type ion exchange interphase, or the formation of insoluble 4-AET calcium saits along the adhesive-dentine interface will remain speculative until further confirmation is established using definitive chemical analytical technique such as X- ray photoelectron spectroscopy’ or micro-Raman spectroscopy.* Fig. SB Secale hase hand STEMIEDX enatgy spoctrum ofthe FASG sass core ‘The elemental composition shows that i isan alkaline acth metal MuoroaTuminosilicate glass. Part of the alkaline earth metal is repluced by lanthanum, The TEM miccograph taken from an unstained. udéminralised section from the S-60 group, showing the various types of fillets within polymerised resin nati. The eeston-dense core (G) of an FASG filer is ‘ually surrounded by a siliceous hydrogel layer (pointer) ‘Ua contains elcten-danse "seeds. Smale FASG fillers ietusion of tantuanum in the glass particles results ina radinpague adhesive $ Copper is from the single stot copper gi. nay have completely reacted withthe acc monomers to for elestron-lucene hydrogels that contain eleetron= ange "seads (see Figure 3B). Unlike FASG filles, F PRG fillers are only sparsely present. An F-PRG iller (0) isgrey throughout ihe ene ills an doesnot contain electron-dense “Seeds” Agulomerated mieofiler clusters (arrow) are also observed within the relatively homogencus, eesin marx (M). Bar~ $00 nm. Another consistent observation in this study was the presence of electron-dense “seeds” within the hhycirogel layers of lanthanum-containing FASG particles (EDX - Figure SB) but not the caleium-containing P- PRG fillers (EDX not shown). These “seeds” are not demineralisation or staining artefacts, as they were found in both unstained, undemineralised (Figure 4B) and demineralised (Figure 6C) sections. Similar spherical “seeds” were also identified in ChemFlex (Denstply, DeTrey. Konstanz, Germany), a high strength GI cement that contains strontium alumino-fluorosilicate glass particles.” These electron-dense “seeds” represent segregated regions of fluoride-rich phases within a more continuous glass phase in certain reactive glass compositions.” They were resistant to acid dissolution even after complete laboratory dissolution of the glass cores (Figure 6C), Their occurrence may be regarded as phenomenological as their function is unknown. 21‘A low magnification TEM micrograph taken from an unstained, demineralised specimen in the 5-600 group. The unstained hybrid layer is not apparent, although a thin surface intermediate layer (arrowhead) can be discerned. After laboratory formic acid demineralisation, there is phase separation of the resin matrix into an clectron-lucent phase (L) and an electron-dense ($) phase. FASG fillers are located predominantly in the electron- ai a Higher magnification of the unstained resin matrix showing the phase separation depicted in Figure 6A The electron-lucent phase (L) of the resin matrix contains only microfiller clusters (arrowhead). By contrast, F-PRG fillers (open arrow) are found almost exclusively within the electron-dense phase of the resin matrix. Bar = 2 um, Very high magnification of the electron-dense phase of the unstained resin matrix, showing the dispersion of artefactual dendritic deposits (arrowhead) within the resin matrix. Electron- dense FASG fillers are completely dissolved after formic acid demineralisation, leaving behind electron-lucent cores that contain remnant electron- dense, spherical "seeds" (arrow). dm: laboratory demineralised dentine, Bar = 200 nm. dense phase of the resin matrix. dm: laboratory demineralised intertubular dentine. Bar = 3 pm. The polymerised resin matrix in demineralised sections was modified by a heavy deposition of electron- dense, crystalline dendritic outgrowths (Figures 6A and 6C). These artefactual deposits probably represented insoluble complexes that were precipitated within parts of the resin matrix as the FASG fillers were dissolved by the formic acid. Similar to citrate ions, formate ions may also possess the ability to complex metal cations that result in hastening the degradation of ion-leachable glass fillers in resin modified GI cements and compomers.”* This suggested that formic acid should not be used in the future for investigation of adhesive interfaces containing ion-leachable glass fillers. However, the detection of these artefacts indirectly demonstrates that water and ion diffusion and formation of reaction products is possible within the resin matrix of this all-in- one adhesive. This, unexpectedly, also permits the identification of phase separation within the resin matrix. The phase that contained these artefactual outgrowths as well as FASG fillers probably contained hydrophilic resin components that facilitated such ion movements. Similar ion exchange within the hydrophilic phase of the resin matrix of this adhesive probably accounts for its fluoride-releasing and in vitro cariostatic potential.” Conversely, silanisation of the colloidal silica probably rendered them hydrophobic, so that they were retained within the hydrophobic phase of the resin matrix (Figure 6B). It may be tentatively concluded that amalgamation of adhesive resin and GI technology in an all-in-one adhesive represents a novel approach in dentine bonding that awaits further clinical testing. At the moment, it is not known whether there is real clinical benefit derived from GI-based dentine adhesives when compared to either the use of conventional purely resin-based, multi-step dentine adhesives or resin modified GI cements.*° Acknowledgements The authors thank Dr. Kunio Ikemura, Shofu, Inc., for his expert advice on the chemistry aspects of Reactmer Bond. The adhesive used in this study is generously supplied by Shofu, Inc, Kyoto, Japan. This study was supported, in part, by CRC grant 1020 3278 24993 08004 323 01 from The University of Hong Kong, Hong Kong SAR, China, by Grant in Aid for Scientific Research #11470401 from the Ministry of Education, Science and Culture, Japan, and by grant DE 06427 from the National Institute of Dental and Craniofacial Research, USA. The authors are grateful to Michelle Barnes for secretarial support. References 1. Watanabe I, Nikaido T, Nakabayashi N. Effect of adhesion promoting monomers on adhesion to ground 22dentin, Journal of the Japanese Society of Dental Materials and Devices 1990; 9:888-893. 2, Hayakawa T, Kikutake K, Nemoto K. Influence of self-etching primer treatment on the adhesion of resin composite to polished dentin and enamel, Dental Materials 1998; 14: 99-105. 3. Frey O, Hansen M, Dede K. Adhesive attachment of dental filling materials. European Patent number EP0937448; 1999. 4, Kazama H, Oguri M. Visible ray polymerization initiator and visible ray polymerizable composition. United States Patent number 5,744,511; 1998. 5. Roberts TA, Miyai K, Ikemura K, Fuchigami K, Kitamura T. Fluoride ion sustained release preformed glass ionomer filler and dental compositions containing the same. United States Patent number 5,883,153; 1999. 6. Inoue S, Van Meerbeek B, Vargas M, Yoshida Y, Lambrechts P, Vanherle G. Adhesion mechanism of self= etching adhesives. In: Tagami J, Toledano M, Prati C, editors. Proceedings of Conference on: Advanced Adhesive Dentistry. Third International Kuraray Symposium. December 3-4, 1999. Granada, Spain. Cirimido, Italy: Grafiche Erredue, 2000: 131-148. 7. Yoshida Y, Van Meerbeek B, Nakayama Y, Snauwaert J, Hellemans L, Lambrechts P, Vanherle G, Wakasa K. Evidence of chemical bonding at biomaterial-hard tissue interfaces. Journal of Dental Research 2000; 79: 709-714. $, Nezu T, Winnik FM. Interaction of water-soluble collagen with poly (acrylic acid). Biomaterials 2000; 21 415-419. 9, tkemura K, Shoda K, Endo T. An ATR-FTIR spectroscopic study on the chemical interaction of a new monomer bearing carboxylic moiety with dentin apatite and collagen. Journal of the Adhesion Society of Japan 1999; 35: 9-20. 10. Geiger SB, Weiner S. Fluoridated carbonatoapatite in the intermediate layer between glass ionomer and dentin. Dental Materials 1993; 9: 33-36. 11, Sennou HE, Lebugle AA, Grégoire GL. X-ray photoelectron spectroscopy study of the dentin-glass ionomer cement interface. Dental Materials 1999; 15: 229-237. 12, Ikemura K, Kouro Y, Endo T. A new fluoride-releasing dental adhesive and its bonding durability to teeth under long-term water-immersion. Journal of the Adhesion Society of Japan 1998: 85-97 13, Tay FR, Sano H, Carvalho R, Pashley EL, Pashley DH. An ultrastructural study of the influence of acidity of self-etching primers and smear layers thickness on bonding to intact dentin, Journal of Adhesive Demtistry 2000; 2: 83-98. 14. Ngo H, Mount GJ, Peters MC. A study of glass-ionomer cement and its interface with enamel and dentin using a low-temperature, high-resolution scanning electron microscopic technique. Quintessence International 1997; 28:63-69. 15. Mair LH. The silver sorption layer in dental composites: three year results. Dental Materials 1999; 15: 408-412. 16. Koibuchi H, Yasuda N, Nakabayashi N. Bonding to dentin with a self-etching primer: the effect of smear layers, Dental Materials 2001; 17: 122-126. ’ 17. Pashley DH, Tao L, Boyd L, King GE, Homer JA. Scanning electron microscopy of the substructure of smear layers in human dentine. Archives of Oral Biology 1988: 33:265-270, 18. Wesenberg G, Hals E. The in vitro effect of a glass ionomer cement on dentine and enamel walls. An electron probe and microradiographic study. Journal of Oral Rehabilitation 1980; 7: 35-42.20: 2 = 22, 23. 24. 2 aA 2 2 “I 28. 22) 30 24 Dp . Tay FR, Smales RJ, Ngo H, Wei SHY, Pashley DH. Effect of different conditioning protocols on adhesion of a high strength GIC to dentin. Journal of Adhesive Dentistry 2001 (in press). Kakaboura A, Eliades G, Palaghias G, An FTIR study on the setting mechanism of resin-modified glass ionomer restoratives. Dental Materials 1996; 12: 173-178. . Young AM, Sherpa A, Pearson G, Schottlander B, Waters DN. Use of Raman spectroscopy in the characterisation of the acid-base reaction in glass-ionomer cements, Biomaterials 2000; 21: 1971- 1979. Bourke AM, Walls AW, McCabe JF. Light-activated glass polyalkenoate (ionomer) cements: the setting reaction. Journal of Dentistry 1992; 20: 115-120. Ikemura K, Kouro ¥, Endo T. Effect of 4-Aryloxyethyltrimellitic acid in a self-etching primer on bonding to ground dentin. Dental Materials Journal 1996; 15:132-143. Leung Y, Morris MD. Characterization of the chemical interactions between 4-MET and enamel by Raman spectroscopy. Dental Materials 1995; 11: 191-195. . Hatton PV, Brook JM. Characterisation of the ultrastructure of glass-ionomer (poly-alkenoate) cement. British Dental Journal 1992; 173: 275-277. Spencer P, Wang Y, Walker MP, Wieliczka DM, Swafford JR. Interfacial chemistry of the dentin/adhesive bond. Journal of Dental Research 2000; 79: 1458-1463. . Barry TI, Clinton DJ, Wilson AD. The structure of a glass-ionomer cement and its relationship to the setting process. Journal of Dental Research 1979; 58: 1072-1079. Marks LA, Verbeeck RM, De Maeyer EA. Martens LC. Effect of a neutral citrate solution on the fluoride release of resin-modified glass ionomer and polyacid-modified composite resin cements. Biomaterials 2000; 21: 2011-2016. Okuda M, Pereira PNR, Nikaido T, Yoshimitsu K, Tagami J. Evaluation of caries inhibition effect with Confocal laser scanning microscope. Journal of Dental Research 2000; 79 (Special Issue): 147 (abstract 26). . Haller B, Trojanski A. Effect of multi-step dentin bonding systems and resin-modified glass ionomer cement liner on marginal quality of dentin-bonded resin composite Class II restorations. Clinical Oral Investigations 1998; 2: 130-136.Analysis of Resin-Dentin Interface with Laser Raman Microscopy ‘Masashi Miyazaki, D.D.S., Ph.D. ashi Miyazaki, Hideo Onose Department of Operative Dentistry, Nihon University School of Dentistry (Japan) Int roduction Good marginal adaptation to the entire tooth surface is believed to reduce marginal discoloration and ipa irritation related to microleakage. Such a bonding systems, which bonds to tooth sufficiently strong 10 ‘vithstand the internal and external stresses, have been desired for long time. Since the introduction of acid w technique for bonding restorative to enamel surface (Buonocore, 1965), many attempts have been mnade to develop restorative materials both for enamel and dentin, The development of promising bonding “sjstem is dependent on understanding of the bonding mechanism and characterization of the bonding interface “betyeen resin composites and tooth substrates. Comparing tothe clinical success of enamel bonding, development ‘of predictable and clinically durable dentin bonding has been a more challenged goal to achieve. More factors related to the difficulty in dentin bonding than enamel bonding, including higher organic component in dentin, fluid pressure from the dentinal tubules, and the presence of smear layer (Miyazaki et al., 1996; Van ecrbeek et al., 1998; Perdigio et al., 1999). The smear layer prevents infiltration of monomers into dentin, and even if this layer is removed with acid etching followed by bonding agent application, fluid flow from pulp iuht interfere stable bonding (Pashley, 1991). Bonding systems can be categorized by how they treat the smear layer to obtain micromechanical ation by infiltrating resin into the demineralized dentin and exposed collagen fibers. One-bottle adhesive system removes the smear layer and smear plugs with acidic conditioners followed by water rinsing. Complete al ofthe smear layer, demineralization of the top 3~5 4m of the dentin surface with phosphoric acid, and ration of resin creates a hybrid layer (Nakabayashi et al., 1982; Van Meerbeek ct al., 1992; Inokoshi et al, 3), Self-etching primer system uses acidic functional monomers mixed with hydrophilic monomers, such as HEMA. These primers are designed to dissolve the mineral component of the smear layer, and alter the underlying superficial dentin (Hasegawa et al., 1994: Watanabe et al., 1994; Ikemura et al., 1996). Penetration Recently one-application bonding tems, which combine functions of the self- iple, Reactmer Bond (Fig. 1), have been eloped. Theoretically, the acidic adhesive ssolves the smear layer, incorporates it into Ithough the bond strength is not as high as FAG ORT two-step bonding systems, the advantages of ‘Adper Plompt L-Pop. je-application bonding procedure with a p Development of dentin boning systems has been foeusod on te eli of bond ‘potential for fluoride-release are noteworthy : pa strength as wel 05 the reduction of elinical tes 25for clinical use (Ikemura et al., 1998; Kitasako et al., 2000; Miyazaki et al., 2001). To understand the mechanism of dentin bonding through creation of a hybrid layer (Nakabayashi et al., 1982; Van Meerbeek et al., 1992; Inokoshi et al, 1993), we have to examine the relationship between the depth , of demineralization and the extent of monomer penetration. Excessive etching of the dentin produces weak bonding because the base of the demineralized dentin does not completely impregnate the resin adhesives. It has been suggested that poor infiltration of adhesive resin into the demineralized dentin leaves nano-spaces in the hybrid layer (Sano et al, 1994; Sano etal, 1999; Spencer et al., 1999; Miyazaki et al., 2002). Such a region of unprotected collagen fiber may be susceptible to degradation from oral fluids and bacterial enzymes. The adhesive systems demineralize the dentin surface to a certain depth followed by the resin infiltration in an attempt to hybridize the exposed collagen fibers and partially demineralized underlying dentin. It is a major concern whether resin infiltration takes place to the entire depth of demineralized dentin. Numerous morphological and chemical studies have been done in an effort to understand and obtain a clearer picture of the dentin-resin interface. Because of the very small thickness of the hybrid layer, particularly with self-etching primer systems (Harada et al, 1998; Prati et al., 1998), methods to analyze this layer must have a very high resolution similar to that of electron microscopy. Raman microspectroscopy is a useful analytical technique for studying the composition and structure of bonding of a sample (Tsuda & Arends, 1997). Since the acquired spectra are attributed to molecules rather than to single elements, and the laser beam can be focused to a very small-sized spot, a high spatial resolution of the sample surface can be achieved. Dehydration of the sample is not required and the measurements can be done under room conditions. Penetration of resin monomers into decalcified dentin has been reported with use of Raman microscopy (Suzuki et al, 1991; Ozaki et al., 1992; Van Meerbeck et al., 1993: Wieliczka et al., 1996; Wieliczka et al., 1997; Lemor et al., 1999; Spencer et al., 2000; Miyazaki et al., 2002). However, differences in the extent of resin penetration among adhesive systems have not been fully characterized. The purpose of this study is to analyze the degree of demineralization of bovine dentin and the extent of resin monomer penetration into demineralized dentin with the use of adhesive resin and a laser Raman microscopy. Materials and Methods 1) Materials Tested Two self-etching primer systems Mac Bond II/ Palfique Estelite (Tokuyama Dental., Tokyo, Japan) and Imperva Fluoro Bond/ Lite-Fil II A (Shofu Inc., Kyoto, Japan), a one-bottle adhesive system Single Bond/ Filtck Z250 (3M ESPE Sr. Paul, MN, USA), and a one-application bonding system Reactmer Bond/ Reactmer (Shofu Inc., Kyoto, Japan), were employed (Table 1). The adhesive systems were used according to the manufacturers’ instructions. A curing unit (Optilux 401, Demetron/ Kerr, Danbry, CT, USA) was connected to a variable transformer in order to adjust the light intensity to 600 mW/cm? as measured with a dental radiometer (Model 100, Demetron/Kerr).‘Table 1: Bonding systems used in this study (Cade System Primen/Btehant ‘Adhesive Procedure Restorative (Manufacturer) (Main components) (Main components) MB Mac Bond I Primer Bonding Agent loses priming Palfique Estelite (Tokxyama Dental.) (4-AET,-AETA,HEMA —(Mac-10, HEMA, Losec eure "water, ethanol) Bis-GMA, initiator) cata FB Imperva FB Primer FB Bond 10s priming Lite-Fil LA Fluoro Bond (4-AET, 4-AETA,HEMA (4-AET, HEMA, 10s cure (Shoiu Inc.) catalyst, water, ethanol) UDMA, filler, initiator) Single Bond Etchant Single Bond ISsce etch Filtek 2250 (GM ESPE) (35% phosphoric acid) BisGMA,HEMA, ——_Lsee rinse Vitremer copolymer, Blot dry ethanol, water, initiator) 10sec cure RB Reactmer Reactmer Bond 2 Reactmer (Shofu Ine.) - (LAET,4-ARTAUDMA, 2 HEMA.PRG filler, F aluminosilicate glass, aacctone, wate, initiator) 2) Sample Preparation Mandibular incisors, extracted from 2-3 year old cattle and stored frozen for up to 2 weeks, were used asa substitute for human teeth. After removing the roots with a low-speed saw (Isomet, Buheler Ltd., Lake Bluff, IL, USA), the pulps were removed, and final finish was accomplished by grinding with wet 600-grit silicon carbide paper until sufficient area of dentin was exposed. After ultrasonic cleaning with distilled water for | minute to remove the excess debris, these surfaces were washed and dried with a three-way syringe. The tooth surfaces were treated and adhesives were applied according to each manufacturer's recommendations. A Teflon mold, 2.0 mm high and 4.0 mm in diameter, was used to form and hold the restorative materials to the tooth surface. According to each manufacturer's instruction, resin composite was condensed into the mold and cured for 40 seconds with light irradiation, Afer 24 hours’ storage in 37°C water, these specimens were embedded |, Tokyo, Japan) and stored at 37°C for 12 hours. After setting, in selfcuring epoxy resin (Epon 812, Nisshin E1 the epoxy casts were sectioned parallel to dentinal tubules and the sectioned surfaces of the cut halves were polished to high gloss using silicon carbide papers of 600, 1200 and 2000-grit size, successively. Then the surface was mirror polished with a special soft cloth with diamond paste to a grit size of Iyum (Buchler Ltd.).A special care was taken not to aggressive-polish the surface to avoid making a gap between the dentin and adhesive. At least three specimens from different bovine teeth were prepared for each adhesive system. Care was taken to avoid aggressive polish that might change the specimen height, and no destructive effect wi found around the dentin-resin interface where the laser beam was focused on. 3) Laser Raman Microscopy System Raman spectra were obtained with a computer Fig. 2 controlled laser Raman microscope equipped with a monochromator and a back-thinned MPP type CCD camera (System 2000, Renishaw, UK). A He-Ne laser (GLG-5900, NEC Co., Tokyo, Japan) tuned to a wavelength of 632.8 nm with an output level of 75 mW was used as an excitation source (Fig. 2). The focus of the laser beam in conjunetion with the CCD (70% quantum yield) provided a spatial resolution of 0.6~0.8 1m. Instrument calibration was Laser Raman microscopy system used in his study a7determined before data acquisition by comparison of spectra from pure $i (520 em”). The spectral resolution of the laser Raman microscopy was determined by use of a silicon solid plate ion-coated with thin Au. The plate was put on a precision X-¥ stage moved across from Au-coated to Au- uncoated region of the plate surface in steps of 0.2Nm. The spectra of Si (520 emr') were obtained at positions corresponding to each measuring point and the relative intensities of the spectra were calculated. The obtained spectra of Si are shown in Fig. 3. From the result of curve fitting data of this spectrum, the obtained half-width was 1.6-1.9 em’, hence the spectral resolution was determined under | cm’. The variations of the Raman intensities of partially Au coated silicon plate as a function of measuring points are shown in Fig. 4. From the acquired data, differential rate was calculated and the spatial resolution of the laser Raman microscopy was determined as 0.6-0.8 um. This device was designed as a confocal microscopy, the sampling depth of the Raman scattering from the | um x | Um area was calculated as 2 um. Fig. 3 Fig. 4 28000 s20en! 24000 20000 go] Acquired dave @ = Dilerennal re 30 trom acquired data 16000 20 Wes 1008. sooo sun 0 os Lo 1s 20 a3 Measuring point (4m) x00) oo 300, 00) 700 Wavenumber (cm) Raman spectra obtained from Si plate. Obtained Raman intensities of Si as a function of each measuring point 4) Raman Spectrum Acquisition The sample was placed on a precision X-Y stage with a spatial resolution of 0.1 um. An optical microscope (BH2-UMA, Olympus, Tokyo, Japan) allowed for visual identification of the position, from which the signal was obtained. The laser beam was focused on the sample surface with a 100x microscope objective taking care to avoid resin tags, which would disturb the measurement. Fig. 5 The sample was moved perpendicularly to the interface in steps of | Adhesive 0.2 um, and the spectra were obtained at each position across the dentin-adhesive interface with an integration time of 120 seconds for each measurement (Fig. 5). Raman spectra of unaltered dentin and cured adhesive resins were also recorded as references for each obtained spectra. The acquired spectra in the region of interest were analyzed using a curve-fitting program with the Raman microscope software, and the relative amounts of hydroxyapatite, adhesive resin, — schema of the spectrum obtained at each position of and organic substance in the dentin-resin interface were calculated, the deatin-adhesive interface. Raman spectra were ' 2 i i obtained with the laser beam at a spatial resolution of The measurements were done three times with different specimens 9 6.9.8 um. The sample was moved relative to the for each bonding system, and the reproducibility of this technique — '2ye" spot position in steps of 0.2 uum, and the spectra were obtained across the dentin-adhesive interface. was confirmed. 28Results 1) Representative Raman Spectrum The Raman spectra of bovine dentin recorded in the region of 500~1800 om" are shown in Fig. 6. The intense peak at 960 cm is associated with the P-O stretching vibration in the mineral apatite component of dentin (O'Shea et al., 1974). Weak bands at 1242, 1450 and 1669 cm are attributed to the organic components of dentin, such as collagen matrix (Frushour & Koenig, 1975; Xu et al., 1997) The Raman spectra of cured adhesives are shown in Fig. 7. Several intense peaks are observed; from these 640 cm, which is assigned to the aromatic ring group of Bis-GMA (Wieliczka et al, 1996), and 1450 om" which is assigned to the C-H alkyl group [10, 18], were selected as measures of the resin adhesive. Because the question arises whether Bis-GMA, whose molecular weight (MW=512) is significantly larger than HEMA (MW=130), can be representative of all the monomers in the hybrid layer, the C-H alkyl group was selected as a measure of all of the organic components in the hybrid layer. Fig. 6 Fig.7 ‘Obiained Raman spectra of bovine dentin in the region of | Odtainad Raman specta of cured bonding agents nthe region of 2) Change in Intensities of Selected Raman Bands The intensities of the selected Raman bands scanned across the resin-dentin interface of the Mac Bond Iladhesive system are shown in Fig. 8. Scans were started from the dentin side to the adhesive side. A gradual decrease in intensity of the P-O peak from the dentin to Fig. 8 adhesive side was observed. On the other hand, the peal of ‘Mac Bond I the Bis-GMA and the peak of the organic substrate increased across the dentin-resin interface. Since the C-H alkyl groups also appeared in the collagen spectrum, the changes in relative intensities of this peak were considered from the dentin to the resin, From these curves, mineral removal seemed to start at position 2.6 pum and end at 4.6 um, and the resin penetration corresponded to this same. erento Bs esin-dentin interface of Mac 2.0 pm wide region. Bond I adThe intensities of the selected Raman bands Fig. 9 scanned across the resin-dentin interface of lmperva Fluoro _ Imperva Fluoro Bond Bond are shown in Fig. 9. Since the base resin monomer of z = u * ofe oP*se %8 © |e Apatite the Fluoro Bond is UDMA, no aromatic ring was observed. = ee Seat ‘ =| The same general tendencies found with the Raman : oe ae agheshve specimens made with Mac Bond II were observed. The : et fe — mineral removal seemed to start at 2.2m and end at 4.2um “iain “|teeetee,, and CH penetration corresponded to this depth. ae a eee Perpendicular position (jm) Representative line scans across the resin-dentin interface of Imperva Fluoro Bond adhesive system. The intensities of the selected Raman bands scanned across the resin-dentin interface of Single Bond are shown in Fig. 10. A gradual change in Raman intensity of the peaks of mineral and organic components was observed. From the observation of the intensity of the P-O peak, the mineral loss seemed to start at 4.0 um and reach its plateau at 7.4 um. On the other hand, a gradual increase of the C-H band was observed from 4.0 to 11.2 im, and penetration of Bis-GMA began at 5.6 um to a plateau at 11. 2 Um. The gradual changes in the Raman intensities indicate the region from 7.4 to 11. 2 um, where very weak hydroxyapatite concentration and increased resin substrate were detected, might correspond to the zone of decalcified exposed collagen fibers. An area of decalcified dentin without aromatic ring penetration was observed and the width of this zone was about 1.6 uum. Though the exposed collagen sublayer was partially impregnated by Bis-GMA, this resin failed to penetrate into the deeper demineralized dentin. The intensities of the selected Raman bands scanned across the resin-dentin interface of Reactmer Bond are shown in Fig. 11. The same general tendencies found with the Raman specimens made with self- etching primer system were observed. The mineral removal seemed to start at 2.4 um and end at 4.2 um and CH penetration corresponded to this depth. Fig. 10 Fig. 11 Single Bond Reactmer Bond 100 anata Fa |e - ae z 230 2 = Eats 5 % ° os. || * |e Aromatic ring = 40 © Organic = °, oo ye 6 Organle s substance| 5 | = ieee £ Dentin £ commegentht Adhesive Bs Ge oS 2 2 ue 2 0 2 4a es Z a ve seecnpet ee a SF 8 Ele esos om : UST E eae tite 0.005 1,0 152.0 2.5 3.0 35 4.0 4.5 5.0 5.5 6.0 6.5 0123 4 5 6 7 8 8 10 11 12 13 14 Perpendicular position (urn) Perpendicular pasition (jm) Representative line scans across the resin-dentin interface of Single Representative line scans across the resin-dentin interface of Bond adhesive system Reactmer Bond adhesive system Discussion Micromechanical interlocking through the creation of a hybrid layer (Nakabayashi et al., 1982: Van Meerbeek et al., 1992; Inokoshi et al, 1993) is currently accepted as one of the major mechanisms of adhesive resin bonding to dentin. The adhesive systems demineralize the dentin surface to a certain depth followed by the resin infiltration in an attempt to hybridize the exposed collagen fibers and partially demineralized dentin. Tt is a major concern whether resin infiltration takes place to the entire depth of demineralized dentin. 30The representative Raman bands scanned across the resin-dentin interface showed transitional changes in the hydroxyapatite and resin component, Gradual changes in intensity of Raman peaks were also reported by others (Van Meerbeek et al., 1993; Wieliezka et al., 1997; Lemor et al., 1999), indicating that the minerals removed by the acid were gradually replaced by resin. The line scans of the dentin-resin interface of the self etching primer systems and the single application bonding system suggested that the depth of dentin demineralization was 1.8~2.0 jim and resin penetration was detected with the entire depth of demineralized entin, In the case of these bonging systems, penetration of the primer / adhesive into the dentin substrate occurs simultaneously with the demineralization of the mineral component in the dentin. There is little possibility ‘hata region of demineralized dentin that is not infiltrated by the resin monomers exists with these systems Phosphoric acid etching with Single Bond produced a transitional change in the dentin from complete removal of hydroxyapatite to partial demineralization. The layer, which was demineralized with phosphoric acid, seemed to consist of two different phases, phase I and II. The surface layer of the dentin was totally decalcified, leaving exposed collagen fibers, and this layer may correspond to the phase I hybrid layer. Underneath the exposed collagen fibers (phase I hybrid layer), there exists gradually demineralized dentin continuing into the intact dentin. This partially demineralized dentin may correspond to the phase II hybrid layer. Though the alkyl groups of the organic components (including HEMA) penetrated into the entire depth of the “phase IT” hybrid layer, the Bis-GMA could not penetrate into the deepest area with Single Bond. HEMA is included in ‘most dentin adhesive systems to wet the dentin substrate and to facilitate subsequent penetration of adhesive resins that ean polymerize with other monomers in order to create hybrid layer (Harashima & Hirasawa, 1990). From the results of the present study, it is speculated that HEMA diffuses more readily into the partially ‘demineralized dentin zone than Bis-GMA component, and fills the spaces among the collagen fibers as reported by previous study (Eick et al., 1996) For the self-etching primer systems, simultaneous etching and priming facilitates penetration of the adhesive resin monomer into etched dentin (Hasegawa et al., 1994; Watanabe et al., 1994; Ikemura ct al., 1996). The penetration of the acidic monomers into tooth surface creates a hybrid layer for dentin with the application of bonding agent. The same phenomenon might be observed with the single application bonding system. In order to create a hybrid layer, the resin monomers have to penetrate into the demineralized dentin surface (Fig. 12). The thickness of the hybrid layer depends on the bonding systems used, and the variations of the thickness were observed among specimens with the same bonding system and different dentin substrates {Uno & Finger, 1996; Vagas et al., 1997; Prati et al., 1999). Though different hybrid layer thicknesses were ‘observed, no correlation with the bonding strength to the dentin was found. The hybrid layer and the presence of resin tags may not be the only mechanism influencing Fig. 12 dentin bond strengths. ‘The micromechanical entrapment of resin in the demineralized dentin substrate is of importance to create a hybrid layer. The dentin-resin interface has been ‘smgle Bora investigated by use of micromorphological techniques and ‘the importance of resin penetration through the entire depth of decalcified dentin has been emphasized. After the infiltration of resin monomers into the decaleified dentin, Reena subsequent polymerization of monomers is required t0 sex pictures oreross-sectional view of treated dentin, 31create a stable bond. If the polymerization of these monomers is not complete, hydrophilic monomers or small oligomers might be extracted or hydrolyzed by the presence of nano-leakage (Sano et al, 1994; Sano et al, 1999). The quality of the hybrid layer and uniformity of resin penetration are important factors to understand \ the contributions of hybrid layer for dentin bonding mechanism. Conclusions t Laser Raman microscopy used in this study was a useful tool for study of the variations in chemical composition across the dentin-resin interface. The hybrid layer represents a gradual decrease in the relative amount of adhesive from the resin side to dentin side. The diffusion of some resin monomers into demineralized dentin might produce a gradient resin penetration with the highest resin concentration at the surface of the adhesive, lower concentration in the middle of the hybrid layer, and little concentration of resin at the deepest part of the demineralized zone. Further research is needed to determine the relationship between the presence of this poorly resin saturated demineralized dentin and the subsequent hybrid layer degradation. Acknowledgements This work was supported, in part, by grant in aid (C)(2) 14571826 from the Japan Society for the Promotion of Science. References * Buonocore MG (1965). A simple method of increasing the adhesion of acrylic filling materials to enamel surfaces. J Dent Res 34: 849-853. * Eick JD, Miller RG, Robinson SJ, Bowles CQ, Gutshall OL, Chappelow CC (1996). Quantitative analysis of the dentin adhesive interface by Auger spectroscopy. J Dent Res 75: 1027-1033. * Frushour BG, Koenig JL (1975). Raman scattering of collagen, gelatin and elastin. Biopolymers 14: 379- 391. * Harada N, Inokoshi 8, Tagami J (1998), Changes in microtopography across polished resin-dentin interfaces. Am J Dent 11: 137-142. + Hasegawa T, Manabe A, Itoh K, Wakumoto 5 (1989). Investigation of self-etching dentin primers. Dent Mater 5: 408-410. + Harashima I, Hirasawa T (1990). Adsorption of 2-hydroxyethyl methacrylate on dentin from aqueous solution. Dent Mater J 9: 36-46. * Ikemura K, Kouro Y, Endo T. Effect of 4-acryloxyethyltrimellitic acid in a self-etching primer on bonding to ground dentine. Dent Mater J 1996; 15: 132-143. + Ikemura K, Nishino M, Tomita S, Endo T (1998). A SIMS study on fluoride ion uptake into dentin substrate derived from a new pre-reacted glass-polyalkenoate filler in a dental adhesive. J Adhesion Soe Japan 34; 334-344, « Inokoshi S, Hosoda H, Harnirattisai C and Shimada Y (1993). Interfacial structure between dentin and seven dentin-bonding systems revealed using argon ion beam etching. Oper Dent 18; 8-16. * Kitasako T, Nakajima M, Pereira PN, Okuda M, Sonoda H, Otsuki M, Tagami J (2000). Monkey pulpal response and microtensile bond strength beneath a one-application resin bonding system in vivo J Dent 28 193-198. 32+ Lemor RM, Kruger MB, Wieliczka DM, Swafford JR, Spencer P (1999). Spectroscopic and morphologic ‘characterization of the dentin’ adhesive interface. J Biomed Optics 4: 22-27 Miyazaki M, Oshida Y, Xirouchaki L (1996), Dentin bonding system. Part I: Literature review. Bio-Med Mater Eng 6: 15-31 Miyazaki M, Iwasaki K, Onose H and Moore BK (2001). Enamel and dentin bond strengths of newly developed one-step bonding systems. Aim J Dent 14: 361-366. Miyazaki M, Onose H, Moore BK (2002). Analysis of the dentin-resin interface by use of laser Raman spectroscopy. Dent Mater 18: 576-580. Nakabayashi M, Kojima K, Masuhara E (1982). The promotion of adhesion by the infiltration of monomers into tooth substrates, J Bio-med Mater Res 16: 265-273. (O'Shea DC, Bartlett ML, Young RA (1974). Compositional analysis of apatites with laser-Raman spectroscopy (OH, F, Cl) apatites. Are Oral Biol 19: 995-1006. (Ozaki M, Suzuki M, Itoh K, Wakumoto S, Hisamitsu H (1992). Laser-Raman spectroscopic study of the adhesive interface; Analysis between 4-META/MMA-TBB resin and bovine or human dentin, Dent Mater 11: 70-76. Pashley DH (1991). Dentin bonding: overview of the substrate with aspect to adhesive material. J Esthet Dent 3: 46-50. Perdigdo J, Lopes M (1999), Dentin bonding - questions for the new millennium. J Adhes Dent 1: 191- 209. Prati C, Chersoni S, Mongiorgi R, Pashley DH (1998). Resin-infiltrated dentin layer formation of new bonding systems. Oper Dent 23: 185-194. Prati C, Chersoni $, Mongiorgi R, Montanari G, Pashley DH (1999). Thickness and morphology of resin- infiltrated dentin layer in young, old, and sclerotic dentin. Qper Dent 24: 66-72. Sano H, Takatsu T, Ciucei B, Horner JA, Matthews WG, Pashley DH (1994), Nanoleakage: leakage wi the hybrid layer. Oper Dent 20: 18-25. Sano H, Yoshikawa T, Pereira PNR, Kanemura N, Morigami M, Tagami J, Pashley DH (1999). Long-term durability of dentin bonds made with a self-etching primer, in vivo. J Dent Res 78: 906-911 Spencer P, Swafford JR (1999). Unprotected protein at the dentin-adhesive interface. Quintessence Int 30: 501-307. Spencer P, Wang Y, Walker MP, Wieliczka DM, Swafford IR (2000). Interfacial chemistry of the dentin’ adhesive bond. J Dent Res 79: 1458-1463. + Suzuki M, Kao H, Wakumoto $ (1991). Vibrational analysis by Raman spectroscopy of the interface between dental adhesive resin and dentin. J Dent Res 70: 1092-1097. Tsuda H, Arends J (1997). Raman spectroscopy in dental research: a short review of recent studies. Adv Dent Res 11: 539-547. Uno S, Finger W1 (1996). Effects of acidic conditioners on dentine demineralization and dimension of hybrid layers. J Dent 24: 211-216. ‘Van Meerbeek B, Inokoshi , Braem M, Lambrechts P, Vanherle {3 (1992). Morphological aspects of the resin-dentin interdiffusion zone with different adhesive systems. J Dent Res 71: 1530-1540. ‘Van Meerbeck B, Mohrbacher H, Celis JP, Roos JP, Brae M, Lambrechts P, Vanherle G (1993). Chemical characterization of the resin-dentin interface by micro-Raman spectroscopy. J Dent Res 72: 1423-1428, 33Van Meerbeek B, Perdigiio J, Lambrechts P, Vanherle G (1998). The clinical performance of adhesives. J Dent 26: 1-20. Vargas MA, Cobb DS, Denehy GE (1997), Interfacial micromorphology and shear bond strength of single-bottle primer/adhesives. Dent Mater 13: 316-324. Watanabe I, Nakabayashi N, Pashley DH (1994). Bonding to ground dentin by a phenyl-p self-etching primer. J Dent Res 73: 1212- 1220. Wieliczka DM, Spencer P, Kruger MB (1996). Raman mapping of the dentin/adhesive interface, Appl Spectroscopy 50: 1500-1504. Wieliczka DM, Kruger MB, Spencer P (1997). Raman imaging of dental adhesion diffusion. Appl Spectroscopy 51: 1593-1596. Xu J, Stangel I, Butler [S, Gilson DFR (1997). An FT-Raman spectroscopic investigation of dentin collagen surfaces modified by 2-hydroxyethylmethacrylate. J Dent Res 76: 596-601. 34A laboratory evaluation of the water and fluoride equilib: characteristics of Reactmer im John F. McCabe, B.Sc. Ph.D., D.Sc. Dental School, University of Newcastle (UK) Introduction Composites and Glass ionomers have been used as dental fil ig materials for well over 30 years. The ‘advantages and disadvantages of the two groups of products are well established and understood [1]. Whereas these materials have certain advantageous characteristics, they also have limitations and shortcomings, which fi it their clinical applications. For example, composites are relatively strong and tough, but show no inherent adhesive characteristics and do not have the ability to release fluoride. On the other hand, glass ionomers have an inherent adhesive characteristic and release fluoride ions, but are relatively weak and brittle and their properties modified lass jonomers’. have evolved as a means of attempting to combine the most favourable characteristies of both areadversely affected by moisture contamination. Materials, which can loosely be described as ‘resi Parent groups of materials whilst overcoming the shortcomings. As a result, there now exists an almost continuous scale linking the parent groups with products ranging from cement-like, acid-base setting through to resin Inaitix type setting by polymerisation and almost every possible combination in between, Generally, the properties of hese ‘hybrid’ type materials depend on the extent to which they resemble the parent materials from which they are derived. Some products undergo a limited acid-base setting combined with polymerisation, which becomes the dominant setting mode. Other products undergo initial setting by polymerisation and are then able toundergo a limited acid-base reaction as the material absorbs water and the surface of the ion-leachable glass ‘neaks down and liberates ions which react with acid groups present within the resin [2,3]. Within the range of ‘ass ionomer - resin hybrids, Shofu Inc. of Japan have developed a new category of restorative resins (Giomer) based on the use of pre-reacted glass-ionomer fillers. These products are unique in that they allow both the acid-base, glass ionomer reaction and resin polymerisation to proceed to an extent corresponding with the reactions of the parent groups; albeit the avid base reaction occurs before blending the glass with resin to form ‘acomposite-like structure. These products are produced by pre-reacting an ion leachable glass with a poly-acid inorderto create a surface layer of glass from which fluoride ions can be liberated and which can also act as a “sink* for fluoride absorption from and re-release into the environment. The Newcastle group has previously ‘worked with Shofu on the development and evaluation of the first Giomer, Beautifil@. This product was shown ‘have a superior stability in water when compared with a compomer [4]. Fluoride release was demonstrated for Beautifil, although this was at a relatively low level, Since the surface reacted zone on Reactmer is greater than that of Beautifil, the ability for fluoride absorption and re-release should be of a greater magnitude. It was the purpose of the present study to measure fluoride absorption and release for Reactmer and to monitor the ‘water absorption characteristics as it was expected that the pre-reacted zones on the ion-leachable glass particles ‘may act as channels for effective water transport. ‘ Materials The Giomer, Reactmer was compared with two compomer materials (Dyract AP, Dentsply Inc, F-2000, 3M Co) and one fluoride releasing composite (Xeno CF, Sankin). The A3 shade of each material was used for 35all tests. Methods Water absorption, Solubility and Swelling Procedure (Specimen preparation) Specimens were prepared in polypropylene moulds 1.0 mm thick with a 15 mm. dia. hole. The mould was positioned on top of a sheet of PMMA covered with a square of polyester. The test material was packed to slight excess into the hole using a plastic spatula. Another square of polyester , backed by a strip of PMMA was placed on top of the mould and excess material was expressed using hand pressure. The material was irradiated (Visilux I Light Curing Unit) through the polyester film for 60 s from each side. Any flash was removed using a scalpel and the specimen was carefully removed from the mould. Procedure (Test method) Testing was performed in one of four ways: a) over 7 days according to ISO 4049 (desiccated specimens) b) over 7 days according to ISO 4049 except that specimens were not desiccated before starting the test c) over 6 months with prior desiccation d) over 6 months without prior desiccation The above procedures were adopted since it became clear that some materials contain ‘inherent’ water. After the specimens had been removed from the mould, they were either stored dry in the laboratory for 24h or transferred to a desiccator in an oven at 37°C. After 24 hrs, these latter specimens were placed ina desiccator at 23°C for one hour. This cycle was repeated until a constant mass was obtained. All specimens were weighed to an accuracy of + 0.2mg. The specimens were immersed in distilled water and maintained at 37+1°C. At regular intervals, the specimens were removed, rinsed with water, blotted until free from visible moisture, and waved in the air for 15 s. They were weighed | min after removal from the water. After the final weighing, the specimens were reconditioned to constant mass in the a desiccator. The diameter and thickness of each specimen were measured at the centre and at four equally spaced points on the circumference and the volume in cubic millimetres was calculated. Procedure (Swelling) Flat - bottomed holes (5mm dia. and 4mm deep) were cut in Perspex (cast PMMA) and used as moulds. The test material was packed into the mould and cured (60 s with Visilux I). The surface was lapped flush with the surface of Perspex (800 grit paper) and a baseline profile was recorded across two perpendicular diameters of the specimen. The specimens were stored in water at 37°C and re-profiled after 24 h , 1 week, 1 month and 6 months.Fluoride Release and Up-take ‘Specimen Preparation Discs of each material (1.0 mm thick by 10 mm diameter) were prepared in polyethylene moulds betveen PMMA plates covered with Melinex film, The materials were cured using a 60 s exposure to a Visilux (3M Co) curing unit on each side and stored dry for 24 h before beginning the test procedure, The surfaces of the specimens were grounded with P800 grit abrasive paper to remove the resin-rich layer. Water immersion and fluovide re-charge regime Specimens for fluoride determination were weighed and measured (dia and thickness) and then stored indistilled/de-ionised water (5 ml) in individual plastic containers. At the time for fluoride determination, the specimen was removed from the container, rinsed and the storage water decanted for analysis. The specimen ‘vas returned to the container, re-charged with a fresh 5 ml aliquot of water, and soaking continued. After days 8, 16, 23 and 30, each specimen was soaked for Ih in separate 5 ml aliquots of 250 ppm sodium fluoride Solution and then re-immersed in water as before. The water was changed daily (except weekends) and determinations of fluoride release were made at regular daily (first 32 days) and then weekly intervals for a total of 180 days. After 180 days each specimen was soaked for I h in separate 5 ml aliquots of 250 ppm sodium luorde solution and then re-immersed in water as before. Regular daily measurements of fluoride release were fo determine how much fluoride had been absorbed and then re-released by the specimens. oride Determinations Fluoride concentration in the water used for specimen immersion was determined by two methods: lective Electrode Method 3 ml of aqueous solution was used for analysis, to which 0.3 ml of TISAB III solution was added (ie 10 ic stirrer. A reading of the electrode meter was made at 10 min after first immersing the electrode. lation of fluoride concentration was made through reference to daily calibrations which were performed standard solution of sodium fluoride (Russell) diluted to give solutions of 10 ppm to 0.0001 ppm. In Ion Chromatography 1 ml of aqueous solution was injected onto a sample loop of 250 UL. The flow rate was 0.5 ml/min and. ‘made by interpolation using peak height.Results and Discussion Water absorption, Solubility and Swelling Table | shows water absorption and solubility determined at 7 days according to ISO 4049. Table 2 shows water absorption and solubility for the same materials where there had been no desiccation of the samples prior to beginning the test (Note: this is a departure from the ISO 4049 methodology). Table 1: Water absorption and solubility, ,1g mm at 7 days (starting with desiccated samples as in ISO 4049) n=5 Material Water Absorption Solubility mean s.d. mean s.d Reactiner 82.88 4.20 1.82 2.19 DyactAP 28.69 0.83 2.16 0.38 F 2000 36.19 147 7.04 0.43 Xeno CF 18.85 1.54 1.36 0.28 Table 2 : Water absorption and solubility, ug mm® at 7 days (starting with un-desiccated samples) Material Water Absorption Solubility mean s.d. mean s.d Reactmer 71.53 448 4.80 Bias DyactAP 26.02 1.36 TAS 0.77 F 2000 40.29 0.87 15.92 0.55 Xeno CF 18.72 0.54 6.93 0.29 Figures 1 and 2 show long term water absorption and desiccation characteristics for all materials starting either with or without desiccation. The standard deviations are not shown on these plots to maintain clarity. However, the s.d. was very small in all cases. desiccation starts 168 days water absorption ug mm’> 100 150 200 OReacimer 4 F2000 ODyract AP y Xeno CF time days Long term water absorption and desiccation (starting with desiccated specimens as in ISO 4049) Fig. 2 140 120 100 80 water absorption ug mir? -20 9 50 100 200 OReactmer « F2000 time days ODyract AP y Xeno CF Long term water absorption (starting with un-desiccated specimens) up to 112 days. The water absorption of Reactmer is significantly greater than other materials irrespective of starting with desiccated or un-desiccated samples. The values of solubility increase for all materials when starting with un-desiccated samples indicating that all materials contain small quantities of loosely bound constituents or moisture within the original formulation. 38Long term water absorption confirms the data of the 7 day ISO test. Reactmer absorbs significantly mote water than other materials over both the long term and short term. Data for un-desiccated specimens show «similar trend to those for desiccated specimens . Fig. 3 Fig. 4 A fon | wears] ae = Profle of Reamer at baseline, 1, week, | month and 3 months Profile traces ler 3 months water storage Figure 3 shows profile traces for Reactmer re ded at various stages during water storage, Figure 4 shows profile traces for all four test materials after 3 months water storage. Table 3 show’ average expansion (swelling) values for each material over a 3 month period. Table 3: Swelling from a PMMA eavity(im) over 6 months Material Swelling T day Tweek [TL month 3 months | 6 months Reaotmer 1.93.80) | 10.84.27) | 33.26(648) | 5608(8.60) | 63.15(8.87) DyactAP 0902.19) | 384.98) | 2.425.91) 2.38(2.16) (0.38(1.82) LF 2000 2.20038) | 2.250066) | 1.13.27) | -039(0.73)_|-0.21(1.09) | Xeno CF =1.680.83) | 227247 | 0.1047) | 0.05222) | 0.412.053) _| Reactmer gives a sustained expansion over 3 months. F 2000 and Xeno CF appear to give an initial small contraction (or erosion) followed by a very small expansion over 3 months. At 3 months, the overall expansion/contraction for Dyract AP, F 2000 and Xeno CF is not significantly different from zero (p>0.05), Reactmer gives a significantly greater expansion than the other materials from 1 week to 3 months (p<0.05), Comparison of swelling data with water absorption data suggests that swelling of Reactmer is caused by water absorption. aoThe interaction of dental restorative materials with water is a key factor in controlling material integrity and durability. The consequences of water absorption depend upon the amount of water absorbed, the effect on dimensional stability and the effect on material structure (eg. hydrolysis of resin-filler interfacial coupling for composites). It is possible for a material to absorb large quantities of water without any adverse effect on dimensions or properties (eg. acting like a sponge). Alternatively, water absorption may generate internal stress and outward pressure, which may be damaging to the surrounding tissues. It has been shown previously that water absorption of dental composites can generate a significant pressure against the cavity walls [5]. Swelling of resin-modified glass ionomers due to water absorption has been reported [2]. In a similar experiment that carried out here, two materials protruded by 15 -50 um after 6 months in water. For those products, the high water absorption and swelling values were explained by the presence of HEMA. For Giomer materials, the high water absorption may be related to the pre-reacted glass ionomer zones becoming saturated rapidly with water. It is important to gain understanding of the process through which this occurs. If the pre-reacted zones are acting like mini sponges. it may be possible to accommodate the water absorption without generating too much outward pressure. On the other hand, if the water is absorbed through an osmotic effect, an outward pressure would be generated and this would limit the clinical applications of the material. Fluoride Release and Up-take Figure 5 shows a series of calibration plots for the specific ion electrode. Figure 6 shows a calibration plot for the ion chromatogram. Fig. 6 100 Fig. 5 200 ISE Calibration calibration for ion chromatogram 0.01 ppm & 0.1 ppm er cles 0.001 ppm 5 0.0001 ppm gs 5 g [= Z 100 5 E S 8 nag 50. 1ppm F =10 mol/L, oO 2 = a7) & 0 3 8 7 6 5 “4 3 e Concentration mol/Lx10 C000) 0.0001 = 0.01 0.01 04 1 10 100 theoretical concentration ppm Examples of calibration plots for Specific lon Electrode A calibration graph for the ion chromatogram. Produced from five separate determinations made on serial dilutions of fluoride standard solutions (theoretical values) plotted against calculated values produced through the instrument software. Bars indicate standard deviations. The fluoride release value was consistently higher when measured using the specific ion electrode than when using ion chromatography. It was impossible to make sensible measurements of fluoride concentrations much below 0.1 ppm with the electrode as the calibration graph was curved in this region and measured mV values correspond to two possible fluoride concentrations. The electrode method would not therefore be suitable for monitoring fluoride release beyond 1 month for the materials evaluated in this study. For this reason, we believe that ion chromatography is the preferred method for long-term fluoride release studies. Over the whole period of testing, F2000 gave the consistently higher values of 24h fluoride release. Reactmer and Dyract AP were similar and Xeno CF gave the lowest values of fluoride release. 40Figure 7 shows fluoride release against time for all four materials. Fig.7 te ect Pc tea 2000 ers roa or on 1 Sp vce ae Beas os Es Bun a Dac AP uc rtease re eno OF Fuori eeass ES 2 cetera i 1 Sp See pon limcammngemy =) Bae © br choncipeoy Ete tas Days Fuoride Release (24h values) over 40 days with re-harging at 8,1 6,23 and 30 days Table 4 shows fluoride release values (compared to ‘background’ levels) during the 24h period after re- charging with 250 ppm fluoride solution. This data shows that all materials are able to absorb and re-release a certain amount of fluoride. Reactmer and F2000 are able to absorb and release significantly more fluoride than Dyract AP and Xeno CF over the first 30 days. Over this time period, the amount of fluoride, which can be ahsorbed and then released, appears to increase with time. It appears that as the material loses its “natural” Sluoride, it creates the opportunity for more absorption of fluoride from the environment. The results at 180 days show that the performance of Xeno CF seems to have been maintained at a relatively low and constant level of fluoride absorption and release. The value for Dyract AP has increased to the levels measured for F2000. Fluoride absorption and release for Reactmer appears to have been reduced slightly. It would be interesting to speculate how the pattern of results may change if fluoride soaking had been maintained at ‘regular intervals between 30 days and 180 days. Table 4: 24 h fluoride release after fluoride recharge 11gmm* x 10° 3 by chromatography (calculated as release value less previous 24h value). This corresponds to peak heights in figure 7. Material 24 h fluoride release AfterSd | After 16d After 33d | After30d | After I80d | Reaetmer 1307) | 13704) | 13.606) | 1500.3) | 10.7005) DyactAP 5.60.8) 5.10.9) 6207) | 81a) | 13203) | 2000 10.5(0.3) 630.7) 1242.1) 17.6(1.6) 13.804) | Xeno CF 3.704) 44a | 4907) | 620.2) 460.0) | 4lSee The concentration of fluoride ion required in order for a therapeutic effect to be manifested is not known. However, it is known that all materials, which inherently release fluoride from within their structure, do so most noticeably within the first few days and that the rate of release into water declines rapidly as a function of time[6,7] . Furthermore, materials having a ‘compomet-type’ structure are known to release significantly less fluoride than conventional glass ionomers or resin-modified glass ionomers[7]. Hence, the ability of a material to absorb fluoride from its environment and then re-release it may be more important in the long term than the inherent ability to release fluoride. Furthermore, the mechanism of absorption followed by release implies a more stable structure for the host material compared with sustained inherent release over a long time. The latter implies that a material must be yielding its fluoride by a sustained process of degradation. Conclusions Water absorption and swelling over 6 months is significantly greater for Reactmer than for any other material tested. This is potentially an area of concern. Much will depend upon the mechanism of absorption and any associated build up of pressure. Ton chromatography is the most appropriate method for monitoring long-term fluoride release. Reactmer, F2000 and Dyract AP have similar patterns of fluoride release. These materials are able to absorb and release significant amounts of fluoride from their environment. There is some evidence that fluoride absorption and re-release may be clinically more important than ‘inherent’ release, particularly in the long term. References . McCabe JF and Walls AWG (1998) Applied Dental Materials, Blackwell Science, UK 2. McCabe JF (1998) Resin-modified glass ionomers. Biomaterials 19 521-527 3. Sidhu SK and Watson TF (1995) Resin-modified glass ionomer materials. 4m J Dent 8 59-67 4. McCabe JF Carrick TE and Rusby § (2000) Properties of a composite containing pre-reacted glass a ionomer filler. Dent Res Special Issue Abstract 5. Momoi Y and McCabe JF (1994) Hygroscopic expansion of resin based composites during 6 months water storage. Br Dent J 176 91 -96 6. Momai Y and McCabe JF (1993) Fluoride release from light-activated glass ionomer restorative cements. Dent Mater 9 151-154 7. Shaw AJ, Carrick TE and McCabe JF (1998) Fluoride release from glass-ionomer and compomer restorative materials: 6-month data. J Dentistry 26 355-359Biocompatibility of Restorations with Giomer Junji Tagami, D.DS.. D.DSe Junji Tagami, Masayuki Otsuki, Hidekazu Sonoda, Pisol Senawong, Yasutaka Sasafuchi Yuichi Kitasako, Makoto Arakawa, Kanitha Nirasli"” Cariology and Operative Dentistry, Tokyo Medical and Dental University (Japan) ‘Department of Operative Dentistry, Faculty of Dentistry, Mahidol University (Thailand) Introduction Restorative materials categorized as Giomer consists of adhesive resins and resin based filling materials With the pro-reacted glass ionomer particles, which are applied as the fillers, The Giomer technology was developed by Shofu company and the Giomet due to the fluoride released from the fillers. The present paper consists of 3 studies. that is, pulpal response to expected to exhibit unti-cariogenic effect around the restorations ‘the Giomer restoration (Sonoda et al., 2002), in vivo secondary caries inhibition by Giomer, and also caries inhibition around the Giomer restoration after recharging of fluoride in vitro. 1, Pulpal response of Giomer restoration Methods and Materials Pulpal response of Giomer restorations were investigated using Japanese monkeys (Macaca Fuscata) aecording to the guidelines of the Tokyo Medical and Dental University Animal Research Center. Deep class V ‘cavities were prepared on the buccal surfaces of 48 intact teeth with a high-speed tapered diamond bur (#170) under water spray coolant. The 48 experimental teeth were divided into two groups (n=8 for each time period). In the first group (Reactmer group), Reactmer Bond A and B were mixed, and applied to the cavity for 20s using a miero-brush to reduce the formation of blisters, and then light-cured for 20s. Then the cavity was restored with Reaetmer Paste and light-cured for 30s, In the second group (Beautifil group), the cavity was treated with a mixture of the FB Primer A and B for 10s. Next, excess primer was removed with gentle air drying, and then the low viscosity FB Bond was placed and light-cured for 10s, and the cavity was restored with Beautifil and light ‘ured for 40s. After 3, 30 and 90 days, the experimental animals were sacrificed with an overdose of thiopental sovlum (Ravonal, Tanabe Pharmaceutical). Procedures for the specimens preparation were the same as previous studies (Sonoda et al., 2001, Taylor, 1966). The pulpal response and the presence of bacteria in the cavity were evaluated and measured using a Tight microscope. The thickness of the remaining dentin underneath cach cavity was measured following the pourse ofthe der “section of the tooth. Odontoblastic change, inflammatory cell infiltration, and irritation dentin formation were al tubules. The thinnest tissue section of each tooth sample was selected as the representative “scored in four degrees of none (~), slight (£), moderate (+) and severg (+), These criteria are the same as in the “previous study (Sasafuchi et al., 1999). These specimens where bacterial penetration was detected along the -cwity walls and fixed inadequately were omitted. The results of odontoblastic change, inflammatory cell “infiltration, and irritation dentin formation were statistically analyzed using the Mann-Whitney U Test at the 95% level of confidence (n=! 8 for each time period). Parametric one-way analysis of variance (ANOVA) and BFisher’s protected least significant difference (PLSD) test at the 99% level of confidence (n=8 for each time period) were used for analyzing the results of the remaining dentin thickness in each group. Results The remaining dentin thickness (RDT) and the graded scores for observed odontoblastic change, inflammatory cell infiltration, and irritation dentin formation are summarized in Table 1. Severe inflammatory reactions of the pulp such as necrosis and abscess formation were not observed. Pulpal changes were located adjacent to the cavity. Odontoblastie Change: The disappearance of primary odontoblasts of the remaining dentin underlying the cavity preparation, as well as aspiration of cell nuclei into these same dentinal tubules were observed in both the Reactmer group and the Beautifil group. There were no statistically significant differences in odontoblastic changes between the two groups for the same time intervals. Inflammatory Cell Infiltration: In both groups, no specimen was observed to have more than moderate inflammatory cell infiltration over the experimental periods, and there were no statistically significant differences in the incidence of inflammatory cell infiltration between the two groups for the same time intervals. Irritation Dentin Formation: There were no statistically significant differences in irritation dentin formation within any of the two groups at 30 and 90 days. Bacterial Penetration: There was no bacterial presence except for one case, and the specimen where bacterial penetration was detected along the cavity wall was omitted. Table 1:Results histopathological findings Time Intervals 3 days 30 days 90 days Experimental Groups Reactmer Beautifil Reactmer Beautifil Reactmer Beautifil # of Specimens 8 8 8 8 8 8 Odontoblastic none a 6 7 8 8 8 Change slight 1 1 1 0 0 0 moderate 0 1 0 0 0 0 severe 0 0 0 0 0 0 Inflamation none 6 i 7 8 8 8 Cell Infiltration — slight 2 1 I 0 0 0 moderate 0 0 0 0 0 0 severe 0 0 0 0 0 0 Irritation Dentin none 8 8 ie 7 4 4 Formation slight 0 0 1 1 3 3 moderate 0 0 0 0 0 I severe 0 0 0 0 0 0 Remaining mean 0.66 0.65 0.18 O41 0.38 0.43 Dentin Thickness minimum 0.25 0.10 0.05 0.15 0.15 0.05 (mm) maximum 1.10 1.30 0.44 1.10 0.60 1.20 Discussion Only 3 of 24 pulps showed a slight inflammatory cell infiltration in the Reactmer group. Our Reactmer data were comparable with those of Kitasako et al. (2000) who investigated an experimental one-application resin bonding system, TOF-1, showing a slight pulpal response in only two of 30 cases. Our data demonstrated that the Reactmer system was biologically compatible with vital pulps. 44Bacterial-staining data indicated that the Reactmer system provided an almost complete seal against bucterial microleakage through all time intervals. The optimum sealing capacity of Reactmer was its biological capacity to prevent bacterial microleakage. Many studies (Inokoshi et al., 1982; Harnirattisai & Hosoda, 1991; Fujitani et al., 1992) have demonstrated that the presence of bacteria within the dentinal tubules and the dental pulp was the major factor that led to pulpal irritation, and a significant relationship between bacterial penetration and pulpal response was confirmed. Sasafuuchi et al. (1999) detected some bacteria with the same bacterial stain technique as well inthe present study when evaluating the correlation between bacterial penetration due to microleakage and pulpal response, Kitasako et al. (2000) showed that TOF-1 did not permit bacterial growth from the tooth surface, In this study, there was only a slight pulpal response, since the Reaetmer system provided an excellent biological seal. This acceptable pulpal response was dependent upon the exclusion of bacterial penetration or the lack of toxicity of Reactmer, Imperva Fluorobond, which isa self-etching priming system, requires two steps (etching/priming and bonding) during the bonding procedure. Reactmer Bond is a self-etching/priming/bonding resin system, and requires only one step during the bonding procedure. Sano et al. (1998) pointed out that reducing the steps of the bonding procedure was an important factor in obtaining a more reliable, strong bond between the resin and, dentin, Many adhesive systems have been reported to be technique-sensitive (Fundingsland et al., 1996), however, this one-step bonding procedure was easy to use. The Reactmer system was expected to reduce the pulpal imitation because of its fewer steps and no rinsing. The mean dentin bond strength of Reactmer was less than that of mperva Fluorobond (Ikemura et al, 2000, Koda 2000) however, there was no pulp inflammation in both groups at 90 days. The adhesive system of Reacimer might show high bond strength over the long-term on the enamel and dentin in cavities, Consequently, Reacimer may be able to provide a long-term hermetic seal, preventing bacterial microleakage. However, Tarim e1al. (1997) reported that high bond strengths of laboratory studies should not be equated with long-term in vivo resistance to bacterial microleakage, and high bond strength should not be misconstrued with long-term clinical success. Further studies are needed to characterize in vivo and laboratory dentin bond strengths of these materials, and to evaluate the long-term marginal seal. Procedures of acid etching, rinsing, and air drying has been pointed out to cause the initial pulpal Invitation (Fujitan et al., 1992). Kitasako et al. (2000) reported that the adhi applied thick in their study to obtain stress-absorbing effect, and the bonding material was not air thinned after ive layer, which was intentionally application because adhesive layer thickness could not be controlled in vivo. In this study, irritation dentin formation at 90 days showed only slight presence in 3 of 8 cases for Reactmer. Since the Reactmer system ‘needed no rinsing and Reactmer Bond was not air dispersed after application, there was little initial chemical and mechanical irritation, That might be the reason why irritation dentin formation at 90 days was shown to be slight. Reactmer Bond and Imperva Fluorobond are fluoride-containing restorative materials which have been developed with the attempt to prevent recurrent caries. Okuda et al. (2000) reported that the thickness of the inhibition zone around Reactmer in bovine root dentin was comparable to that of a resin-modified glass polyalkenoate cement, Fuji If LC improved. In this study, there was only a slight pulpal response at all time periods in both groups. From a clinical standpoint, this slight reaction might be affected by fluoride release from the restorative materials (Six et al., 2000) over the long term, but further study is needed the confirming this. 45Conclusions The fluoride-releasing all-in-one resin bonding system that was tested provided an almost complete c seal against bacterial microleakage through all time intervals. The pulpal response of the Reactmer group was little different from that of the control group. Only 3 of 24 pulps showed a slight inflammatory cell infiltration in the Reactmer group; therefore the Reactmer system is believed to be biologically compatible with vital pulps. There was little initial chemical and mechanical irritation from Reactmer, providing a reason why irritation dentin formation at 90 days was only slight. 2. In vivo secondary caries inhibition by Reactmer Bond Methods and Materials The monkey teeth were used for this in vivo investigation. The guidelines and the fundamental procedures are the same as mentioned above. The root dentin of the monkey teeth was exposed, and cavities were prepared. Then the cavities were restored using Reactmer Bond and Reactmer Paste according to the procedures mentioned above. After 7, 30 and 90 days, monkeys were sacrificed and the subjected teeth were removed from the jaws (n=4 for each group). As a control group, cavities were prepared on extracted bovine root dentin and restored with either Reactmer Bond/Reactmer Paste, or Clearfil Liner Bond 2V and Clearfil AP-X (Kuraray Medical), which are non fluoride releasing adhesive and resin composite, according to the previously established method (Okuda et al. 2000). Procedures of the specimen preparation are illustrated in Figure 1. Fig. | METHOD (demineralizing) Wax ' i Nail varnish Exposed cavity margin ee Ge esssnesstunutesnonnmennrevnnn Sectioning “" and polishing Demineralizing stored for 3.5 days ae CLSM examination Restoration Procedures for the specimen preparation to observe the inhibition zone and outer lesion under CLSM established by Okuda et al. (2001) Specimens were immersed in a buffered demineralizing solution for 3.5 days. Longitudinal half sections were then cut through the center of the restoration and polished for analysis under a confocal laser scanning microscope (CLSM). Data were analyzed by one-way ANOVA, Fisher’s PLSD Test, and Scheffe’s Test at the 95% level of confidence. Results The depth of the outer lesions for the monkey and bovine roots were approximately 180 Lm, in which there were not significant differences among all groups (p>0.05). Regarding to the widths of the inhibition zones adjacent to Reactmer, the results were (mean and SD micrometer): 46‘days =26 (6) 30 days Tdays (bovine) = 21 (4) They were not significantly different (p>0.05). Inhibition zones were not observed adjacent to LB2V/AP-X 32(11) 90 days = 38 (18) restorations. Discussion From the results of depths of outer lesions, it was suggested that there were no differences between ‘monkey and bovine roots in inhibiting demineralization with the acidic solution. Inhibition zones produced by Reactmer were not significantly different among the groups of 7, 30 and 90) days. As expected, non fluoride-releasing material did not produce the inhibition zone. Therefore the formation ofthe inhibition zone was material independent, however, not time independent. The greatest amount of fluoride release from Reactmer Bond oceurred during the first 7days rather than during the later period. Small amount of fluoride is believed to be released form Reactmer Bond even until 90 days after the filling, However, on the cavity walls, the increased accumulation of fluoride did enhance the inhibitory effect on demineralization and taused a significant increase in the inhibition areas during the first 7 days. Conclusion It was concluded that Reactmer Bond/Reactmer Paste appeated to inhibit demineralization of monkey ‘oo! dentin adjacent to the restoration in vivo. ition zone around Giomer restoration in vitro 4. Effect of recharging on caries in Methods and Materials Class V cavities were prepared in human molars and the occlusal margin was placed in enamel, and the ‘cervical margin was placed in dentin, A conventional glass ionomer cement, FujilX (GC), a Giomer, Reactmer Bond/Reactmer Paste (Shofu), and a compomer, Dyract AP with Prime and Bond (Dentsply), and Clearfil SE Bond/Clearfil AP-X (Kuraray Medical) were used for the restorations. After the finishing and polishing, the ‘specimens were stored in tap water and divided into 3 groups for observation at 7 days, 90 days storage and ‘fler recharging at 90 days storage in water by application of 2.2% NaF gel for 5 minutes, Then the artificial lesion was formed with the method developed by Okuda et al. (2002). Also, the depth of outer lesion and the “thickness of the inhibition zone were evaluated under the confocal laser-scanning microscope. Results The results are shown in Table 2. Fuji 1X demonstrated wider caries inhibition zone than those created ‘vith other materials at both enamel and dentin margins. Only Reactmer restorations recharged after 90 days demonstrated the significant increase in thickness of the inhibition zone at enamel and dentin margins compared “to he groups at 7 day and 90 days storage without recharging. In the groups of Clearfil AP-X, inhibition zone was not created at all. Regarding to the depth of outer lesion, no statistically significant difference was found among all groups at 7 days and 90 day's storage. However, rechargeq groups tended to show the smaller depths, of outer lesion. 47Table 2:The means and Standard Deviations of Outer Lesion and Inhibition Zone (micrometer). Substrate Materials Inhibition zone Outer lesion - 7 days 90 days Recharge at 90 days 7 days 90 days Recharge at 90 days Enamel Fuji IX 19.74" 16.53" 22.448 179.32° 135.70° —-143.59° (12.57) (24.46) (23.04) | 8.55) (45.11) (65.47) Clearfil AP-X 0° 0° 0° 236.79 215.988 195,008 (0) (0) (0) (16.37) | (40.40) (38.53) Reactmer 0° L 0° |_ 5.00 209,078 |.233,71" 173.54 (0) (0) (5.39) L (42.94) | (27.16) (39.77) Dyract AP ot of 2.904 187.88'5 217.624 173.86! (0) (0) (7.32) (41.92) (46.49) (27.41) Dentin Fuji IX 32.60 30.28F 4435k _i7679P _17031P 151.26? (8.51) (19.04) (11.03) | (24.96) | (1.76) — (41.36) Clearfil AP-X 0! o o! L 194.269 | 195.519 166.364 (0) (0) (0) (24.87) | (24.34) | (39.66) Reactmer 6.76" 3.68" 15.59 185,31°° | 222.65"! 160.378 (6.56) (5.37) (9.34) (46.93) | (38.35) (39.51) Dyract AP 0.24" 1st" 362° Luog.4st Lise37' 149.58 (0.75) (2.48) (4.74) (34.59) (30.66) (31.21) The data at the same alphabet shows no statistical difference. Discussion This study demonstrated the efficacy of recharging to increase the thickness of inhibition zone around the restorations using Reactmer. The pre-reacted glass ionomer filler particle is considered to be very effective to incorporate the fluoride by recharging. However, conventional glass ionomer cement did not show the recharging effect to increase the thickness of inhibition zone. Though the recharging of fluoride was reported to increase the fluoride releasing (Yip and Smales 1999, Takahasgji et al., 1993, Hatibovic-Hofiman et al., 1997), it was not effective to increase the thickness of the inhibition zone. Presumably, the amount of fluoride recharged was not enough to change the thickness of the inhibition zone. Regarding to Dyract AP, the amount that could be incorporated by recharging was not enough to influence the inhibition zone formation. Each material has its own mechanism in releasing and recharging fluoride. These differences may have influenced the results in this study. Conclusion Reactmer restoration demonstrated the increase in the thickness of inhibition zone after fluoride recharging by the application of 2.2% NaF gel after 90 days storage in water. Acknowledgment The authors would like to thank Dr. Michael F Burrow for his helpful advice in preparing the manuscript. 48References + Fujitani M, Inokoshi $ & Hosoda H (1992) Effect of acid etching on the dental pulp in adhesive composite restorations International Dental Journal 42(1) 3-11 + Fundingsland JW, Aasen SM, Ario PD & Glasspoole EA (1996) Investigation into technique variability in bonding studies Journal of Dental Research (Abstract 868) 75. + Harnirattisai C & Hosoda H (1991) Pulpal responses to various dentin bonding systems in dentin cavities Dental Materials Journal W(2) 149-164. + Hatibovie-Kofinan S, Koch G & Bkstrand (1997) Glass ionomer materials as a rechargeable fluoride- release system Int J Pediaur Dent 7 65-73 + Ikemura K, Kouro Y, Kimoto K. & Shinno K (2000) Development of a new fluoride-releasing one-step bonding resin Journal of Dental Research (Abstract 1694) 79 355. + Inokoshi S, Iwaku M & Fusayama T (1982) Pulpal response to a new adhesive restorative resin Journal of Dental Research 61(8) 1014-1019. + Kitasako Y, Nakajima M, Pereira PNR, Okuda M, Sonoda H, Otsuki M & Tagami J (2000) Monkey pulpal response and microtensile bond strength beneath a one-application resin bonding system in vivo Journal of Demtistry 28(3) 193-198. + Koda K (2000) A study on light-cured resins; chronological change of dentin bond strength of one-step systems with storage time (in Japanese) The Japanese Journal of Conservative Dentistry 43(3) 684- 694 + Okuda M, Pereira PNR, Nikaido T & Tagami J (2001) Evaluation of in vitro secondary caries using confocal laser scanning microscope and X-ray analytical microscope American Journal of Dentistry (accepted). + Sano H, Kanemura N, Burrow MF, Inai N, Yamada T & Tagami J (1998) Effect of operator variability on dentine adhesion: student vs, dentist Dental Materials Journal 17(1) 51-58. + Sesafluchi ¥, Otsuki M, Inokoshi $ & Tagami J (1999) The effects on pulp tissue of microleakage in resin composite restorations Journal of Medical and Dental Sciences 46(4) 155-164. + Six N, Lasforgues J & Goldberg M (2000) In vivo study of the pulp reaction to Fuji IX, a glass ionomer cement Journal of Dentistry 28(6) 413-422. + Sonoda H, Sasafuchi Y, Kitasako Y, Arakawa M, Otsuki M & Tagami J(2002) Pulpal response to a fluoride-releasing All-in-one resin bonding system Operative Dentistry 27(3) 271-277 + Sonoda H, Inokoshi S, Otsuki M & Tagami J (2001) Pulp tissue reaction to four dental cements Operative dentistry 26(2) 201-207. + Takahashi K, Emilson CG & Birkhed D (1993) Fluoride release in vitro from various glass ionomer cements and resin composites after exposure to NaF solutions Dental Materials 9(6) 350-354. + Tatim B, Hafez AA, Suzuki SH, Suzuki S & Cox CF (1997) Biocompatibility of compomer restorative systems on nonexposed dental pulps of primate teeth Operative Dentistry 22(4) 149-158, + Taylor RD (1966) Modification of the Brown and Brenn gram stain for the differential staining of Gram- positive and Gram-negative bacteria in tissue sections American Journal of Clinical Pathology 46 472-474. + Yip HK & Smales RJ (1999) Fluoride release and uptake by aged resin-modified glass ionomers and a polyacid-modified resin composite Ini Dent 49 217-225, 49(DISCUSSION: Session 1) Chairperson: Dr McCabe Dr McCabe: We’ve covered a lot of ground during this session. We’ve introduced the composition of the materials, the method and mechanism of adhesion. We’ve covered biocompatibility. And we started to cover fluoride-releasing characteristics although this will be covered again to some extent in our next session. And we covered the interaction of the materials with water. 3 So now I would like to ask whether any member of our audience would like to make any comment or ask a question at this stage. And do I have Dr Miyai in the audience? Perhaps first of all, make a comment on the water absorption characteristics of these materials. Dr Miyai: I’m Kozo Miyai, researcher of R&D Department of Shofu. Thanks very much for very interesting and new studies of 4 presenters. As one of the researchers in charge of the development of Reactmer, I’d like to give some comments on presentation of Dr McCabe. According to what I’ve been informed by clinicians who had been clinically using Reactmer for more than two years, so far, any poor marginal adaptation or dislodgement due to swelling of Reactmer restorations has not been reported. And also, as Dr McCabe showed in his data regarding the flexural strength, swelling did not cause much degradation of the quality of Reactmer although the swelling rate was high. PRG filler contained in Reactmer is dried glass-ionomer phase, which shows the characteristics of usual glass-ionomer when it swells by water absorption. That is our idea. Also, we think that this idea is supported by our one-year of flexural strength test. Within | month or 3 months period, the strength went down a little bit, but then after that, it stabilised there. Also, Reactmer system was applied to cervical restoration, and we conducted micro-leakage test with thermocycle. We had a bonding test after 2 years of soaking in water, and found only a slight decrease in the shear bond strength. From these 2 experiments, we also think that water absorption and swelling do not affect the bonding performance of Reactmer. However, regarding the internal stress, which can be caused by swelling, we don’t have enough data. We do wish to continue the basic research of the water absorption and swelling of F-PRG. Dr McCabe: I’m reassured to some extent by the comment that you made. | would be concerned if there is a plan to use similar products as luting cement, and there would be some need to take a very serious look at the material. And of course I await the outcome of our own measurement of internal pressure generation during the absorption and swelling of these materials. Any further comments or questions? Attendee: We had a similar study in our laboratory using Reactmer Bond to try to close artificial gaps created in cavities that were prepared in other firm glass. Glass is, as you know, very vital material. For most categories of materials that we investigated, they did not close gaps completely. But with Reactmer Bond, there was a very interesting phenomenon because 70% of the glass cavities cracked during the second week of our experiment, and we had to stop the part of our experiment for this material. So this is just my contribution to our experience. Thank you. 50Attendee: Tilike to ask for a comment, Dr McCabe. You were using 250 ppm fluoride sol While professor Tagami used 2.2 percent fluoride, which is 22,000 ppm. So would that actually raise the rate of the fluoride content enough? And how long have you soaked it in the solution? How long would it take for mn to recharge your material, Resetmer to intake enough fMuoride? Dr McCabe: Ofcourse, the environment in which the material was placed, vary dramatically from almost zero fluoride to pethaps one part per million in fluoridated water supply to thousands of parts per million in fluoridated toothpaste supply. And in actual fact, there is an equilibrium position somewhere along this chain, and we've done a little bit of work on this. If you take the fluoride content in water to below something like a hundred parts per ion, then there is little chance of taking up fluoride. But at 250 ppm or certainly upper level that you would have in toothpaste, the rate of absorption of fluoride does increase quite markedly. Now we chose 250 ppm, “which is relatively low level compared with mouth rinses or toothpastes, but we extended the time to I hour to compensate for that, Certainly there would be a concentration effect on the diffusion of fluoride in and out of ‘the material. Attendee: Yeah, I guess. So, you would like to look at the area under the curve. DrMcCabe: Yes, exactly. DeMjtir: How would the pellicle — organic coating that we have on all restorations and all teeth and all surfaces? How ‘would that be affected? Dr McCabe: [think although we haven't studied about it, it’s fairly obvious that it can have an affect in one or two ways. For example, if the patient was brushing, one would hope that most of the pellicle will be removed. DrMj It would not be removed by regular tooth brushing. That's for sure. DrMeCabe: ‘Also, one would hope that fluoride would be concentrated within the pellicle itself, and that would give a mechanism thorough which the fluoride could be absorbed into the material. Drlmazato: . Thanks very much for the nice presentation of all the four speakers. 1’ like to ask one very basic question. T think ifs a very important point to clarify the possible advantage of the material when companies or we are going to justify the usage of different terminology. Now, we know the name of Giomer, but for example, its recharging effect of fluoride is quite similar to Compomers. And although Giomer's biocompatibility is ok, itis 51not so different from Compomers. So, could I have some additional comments about the clear possible advantages of technology of PRG or Giomer? I ask any of four speakers or any person from Shofu. Dr McCabe: That’s a very good question. Perhaps someone can actually clarify whether this material is a composite-type system or glass ionomer-type system in terms of standardisation. Perhaps, Dr Tagami can make a comment on ' this. Dr Tagami: Thank you very much for the question. From the viewpoint of biocompatibility of this kind of material, Reactmer was very excellent. | think the excellency is because of the bonding procedure, that is, sclf-etching primer system. Especially Reactmer Bond is just one-step system and also we are not required to apply the phosphoric acid etching. According to the previous studies, the phosphoric acid etching produces considerable odontoblastic change due to etching, rinsing and sometimes drying and so on. So that’s why we could obtain the very good results at the initial 3-day evaluation in case of the odontoblastic change and the intra-mouth reaction. Also, at the same time, this material releases fluoride, so we can expect very good bonding, good sealing and anti-caries effect of fluoride-release. Dr Imazato: 1 understand that well. I think this material is very good, but I'd like to know more clearly about it, covering the whole concept of this material. Dr Nishino (Shofu): Thank you, Dr Imazato. I would like to make one point. At the last IADR meeting in Washington, Dr McCabe presented a comparison between Beautifil, one of the Giomer products, and Compomer. It showed the difference between them very clearly. Material strength of Reactmer and Beautifil, the Giomer products, were very stable against long-term storage in the water, while the Compomer showed somewhat large decrease in the strength. This difference in the long-term storage can be attributed to the fact that Giomer products consist of completely reacted glass —ionomer phase, while for Compomer, glass-ionomer reaction proceeds after light curing in clinical service. That means, in oral condition, Compomer starts ionic acid-base reaction, and that is why the material deteriorated. That is Compomer. But Giomer was stable even after six months” storage according to Dr McCabe’s data. Thank you. Dr McCabe: So what we are saying here is that the special characteristic of Giomer is that the fluoride release occurs without any degradation of the material, whereas there is some evidence that in case of Compomer, fluoride release is only achieved if the material begins to undergo some degradation. But your question really, I think, is, “can we justify a special name for this group of materials?” and, “are we confusing dental practitioners by having different names for every group of materials?” Is that your point, Dr Imazato? Dr Imazato: ‘Yes, that’s right. I hope I can see many good clinical results of Giomer later on by several other speakers, and at 52he end, eventually, | would see the benefit of this material. But for clinicians, | think it’s a very important point o differentiate the material from other materials — why you use this material, why this is very good, what the ‘everybody is now looking forward to a short break. We'll reconvene the next session, I think, in 10 minutes, 53The Behavior and Effect of Fluoride airperson junji Tagami, D.D.S.,D.D.Sc. Professor Department of Operative Dentistry Faculty of Dentistry ‘Medical and Dental University (Tapan)Diffusion of Fluoride Lon from Giomer Products into Dentin Toshio Teranaka, D.DS.. Ph.D. Toshio Teranaka, Shusaku Okada, Yoshiharu Mukai, Koji Hanaoka Department of Operative Dentistry & Endodontics, Kanagawa Dental College (Japan) Introduction Prevention of secondary and recurrent caries is one of the major purposes of operative dentistry. It is widely accepted that the gradual release of fluoride ion from glass ionomer cement prevents teeth from demineralization and promotes remineralization"™. ten Cate et al® reported even hypermineralization of incipient caries-like dentinal lesion can occur in adjacent contact with glass ionomer cement, Previous our study"! has demonstrated that diffusion of fluoride ion from fluoride-releasing dental materials into dentin is influenced by pretreatment of dentin, concentration and the type of fluoride in materials, and presence of dentinal fluid. Recently new fluoride-releasing filling materials and the bonding agent, “Giomer”, which shows gla: jonomer cement-like fluoride releasing and recharging abilities, became available on the market by using pre- reacted glass ionomer filler in the material, This study consisted of two experiments, The first object was to determine the fluoride diffusion depth from Giomer products into dentin. The second was to evaluate acid resistibility of dentin adjacent to filling products of Giomer. Materials and Methods Experiment The materials used in the experiment I are shown in Table 1, Seven combinations of adhesive systems and filling materials and one glass ionomer cement vere conducted; Lite-Fil Il with Imperva Bond (LF/ Paces harene pee 1), Lite-Fil Il with Imperva Fluorobond (LF/F), Lite Lite-FHTT Imperva Fluorabond Ler Fil Il with Reactmer Bond (LF/R), Beautifil with gesagt Renae eedsaaltoataee Imperva Fluorobond (BF/F), Beautifil with Reactmer Beautifil Reactmer Bond BRR Bond (BF/R), Reaetmer with Imperva Bond (RE/I). _Resctmey ere roe fee Reactmer with Reactmer Bond (RE/R). The Glasionomer-F GF combination of Lite-Fil II and Imperva Bond, which Table) Mattias sed n experiment donot contain fluoride, was used as a negative control, For the positive control, Glasionomer-F (GIF), conventional glass ionomer cement, was used. Sx7mm of 120 dentin specimens were obtained from freshly extracted bovine incisor coronal dentin. Each material was applied onto the polished dentin surface according to the manufacture’s instruction (Figure 1), There were five specimens for each experimental group. 57Fig. 1 Adhesion ; (Manufacturer’s Instruction) Section LP —» 9 —_.<3g —. 1a 5 Analysis 5x7mm Storage 1,7 and 28days 37°C, DW Coronal Dentin Bovine 'F Diffusion Experimental design of experiment I. Depth The specimens were stored in deionized water for 1, 7 and 28 days at 37°C. After each storage period, the specimens were cut perpendicularly to the bonded surface and polished. Elemental analysis was conducted by electron probe micro-analyzer (EPMA). A line profile of fluoride was shown on a fluoride map to measure the fluoride diffusion depth and intensity. The EPMA analysis conditions are shown in Table 2. The +Accelerating Voliage : 20kV EPMA-8705EIH System (Shimadzu, Kyoto JPN) concentration of element was expressed in a 16 “Sample Current : 15nA pseudo colors mapping image according to the x- *Mapping Area : 40 X40um (512X512 Point) tay strength. Calcium, Silicon and Aluminum were —_Element__Spectral Diffruction Crystal Soa Veonve: Beat Ca LiF (Lithium Fluoride) Kars 3.3604 also analyzed to determine the dentin-resin border. Si PET (Pentaerythritol) Kay» 71268 : . z . Al RAP (Rubidium Acid Phtharate) Kays 8.3994, Fluoride diffusion depth was measured from line ea aa (ene profile of fluoride. Measuring method of fluoride tasie2 EPMA analysis condition diffusion depth are also shown in Figure 1. Statistical *LSA has 55 A lattice width and is highly imporved for Fluorine (Fand Oxgen (Q), analysis was carried out using ANOVA and Student test. Experiment IT The effect of Giomer on the dentin demineralization was evaluated in the experiment II. Materials used and experimental design in the experiment IT are shown in Figure 2. 38 Fig. 2 SESS ASSES Bovine Excision Dentin Embedding Specimen root thin section Experimental design of experiment II.Bovine root dentin segments were cut from just below the cervical area, Segments were cut longitudinally into 300 tum thick specimens, each to be subjected to the embedding with unfilled resin. Acrylie molds were prepared having a circle cavity with 4mm in diameter and 2mm in depth, and a slit. Lite-Fil II (negative control), Beautifil, Reactmer and Glasionomer-F (positive control) were filled, and trimmed single sections were inserted into the slit. These specimens were mounted on the inside of lids of plastic tubes. 20ml of 80% ethyl cellulose ze] was poured in the bottomless test tube, and 20 ml of 0.1 M lactic acid buffer at pH 4.8 was poured on the sel. The specimens were subjected to a subsequent demineralization for 7 days at 37°C. After the demineralization period, the specimens were microradiographed. Results Experiment I Figure 3 shows the typical fluotide-mapping image of the negative control LF/I at 28 days. No fluoride diffusion was found into dentin from the material Fig.3 Fluoride map of Lite Fil and Imperva Bond Figure 4 demonstrates fluoride map of RE/R at 1, 7 and 28 days. The material in red indicates high fluoride content. It is obvious that a high concentration of fluoride was found in the bordering dentin, and the diffusion depth gradually increased with time Fig, 4 Fluoride map of Reacimer and and Reactmer Bond aeThe Figure 5 shows the positive control Glasionomer-F. Diffused fluoride was found significantly deeper than with Reactmer and increased from | day to 28 days. Fig. 5 Fluoride map of Glasiomer-F The mean fluoride diffusion depths in the groups are shown in Figure 6. Except for the negative control, the combination of LF/F showed the minimum fluoride diffusion within the groups studied for the period of 28 days’ storage (p<0.005), though Imperva Fluorobond is a fluoride-containing bonding agent. This result showed that the amount of fluoride content in Imperva Fluorobond might not be sufficient for significant fluoride diffusion into dentin during the experimental period. 1 7 sg. Days Among the Giomer products, the combination of Mea fluoride diffusion depth of each experimental group RE/R demonstrated the significantly deepest diffusion and BF/R was the second. The significant difference between these two was thought to be the fluoride status within the materials. There were no significant differences among the three; RE/L, LF/R and BE/F (p>0.05). BF/F indicated significantly deeper diffusion than that of LF/ F (p<0.005), in spite of using an identical adhesive system. This difference is attributed to the presence of fluoride in the filling material. It is thought that fluoride ion in fluoride containing resin composite can pass the bonding and hybrid layer by fluoride concentration gradient. Therefore, the amount of fluoride in the filling materials is crucial for fluoride uptake into the dentin from dental materials. 60Experiment IT Fig mictoradiograghic images of lesions developed in dentin adjacent to Lite-Fil IT, Glasionomer-F, Beautifil and Reactmer. The preservation of surface layer in the Lite-Fil © 7 presents typical transverse Fig.7 Il was very slight. Comparing Giomer (Beautifil and Lite FATT Glasionomer-? [| Resetmer Glasionomer-F against acid attack was outstanding. sant Beauift Reactmer) image with the Lite-Fil II, a striking difference inthe presence of a thick and dense remineralized surface layer was obvious. Reactmer showed a thick but less radio- dense surface compared with Beatifil. The protection with Actually, the demineralization was hardly detectable on the ypiss! transverse misromubograghic images Glasionomer-F lesion. Conclusions 1. Fluoride in Giomer passed through the bonding and hybrid layer. 2, Fluoride diffusion depth into dentin showed gradual increases with the passage of time. 3. The combination of Reactmer and Reactmer Bond demonstrated the deepest diffusion, while Lite-Fil I and Imperva Fluorobond showed the minimum 4, Giomer inhibited dentin demineralization effectively. 5, These results are encouraging for the clinical use of Giomer. References \. Arends J, Ruben JL, Christoffersen J, Jongebloed WL and Zuidgeest TG. (1990). Remineralization of human dentine in vitro. Caries Res 24:432-435, 2. LiJ, Nakagaki H, Tsuboi S, Kato S, Huang S, Mukai M, Robinson C and Strong M. (1994). Fluoride profiles in different surfaces of human permanent molar enamels from a naturally fluoridated and a non-fluotidated area. Arch Ora! Biol 39:727-31 3. Skartveit L, Wefel JS and Ekstrand J. (1991). Effect of fluoride amalgams on artificial recurrent enamel and root caries. Seand J Dent Res 99:287-94, 4.ten Cate JM and van Duinen RNB. (1995). Hypermineralization of Dentinal Lesions Adjacent to Glass- ionomer Cement Restorations. J Dent Res 74:1266-1271 $. Okada $, Hanaoka K and Teranaka T. (2000). Diffusion of Fluorine from Fluoride Releasing Dental Materials into Dentin. Bull Kanagawa Dent Col 28-30-31. 6. Roberts TA, Miyai K, Ikemura K, Fuchigami K and Kitamura K. (1998). US Par. 8:892, 766 7 Ikemura K. and Endo T. (1999). An ATR-FTIR spectroscopic study on the interaction of adhesive to dentin. J Dent Res (Special Issue) Abst #954, 78 8 Ikemura K, Kouro Y, Kimoto K and Shino k. (2000). Development of a new Fluor epbonding resin. J Dent Res 79 (Special Issue) Abst #1694, 355 le-releasing one-st 61Effect of Fluoride-releasing Adhesives on Dentin -Inhibition of secondary caries and remineralization of decalcified dentin- Masahiro Yoshiyama, D.DS., Ph.D. Masahiro Yoshiyama, Toshiyuki Itota, Satoshi Nkabo, Junichi Doi, Yasuhiro Torii + Department of Operative Dentistry, Okayama University (Japan) Abstract Fluoride-releasing materials can be expected to inhibit secondary caries and enhance remineralization of decalcified dentin underneath restorations. The aim of this study was to evaluate the inhibitory effect on secondary caries and remineralization of decalcified dentin by fluoride-releasing adhesives. Two commercial fluoride- releasing adhesives, Reactmer Bond (RB, Shofu) and One-up Bond F (OB, Tokuyama Dental), and a commercial adhesive without fluoride release, Mac-Bond II (MB, Tokuyama Dental), were used in this study. For examination of the inhibitory effect, class V cavities were prepared on extracted human premolars and restored with a restorative material following application of the each adhesive. The restored teeth were incubated in bacterial medium for 14 days after storage for 14 days at 37°C, 100% humidity. For examination of remineralization of decalcified dentin, decalcified dentin was prepared by using a bacterial caries induction system at the cavity wall. The cavities were then restored with resin composite after application of each adhesive and incubated for 4 weeks at 37°C, 100% humidity, Microradiographs in the inhibitory test showed an acid-resistant layer adjacent to the restoration in the caries-like lesions. The acid-resistant layers in the RB and OB groups with fluoride release were thicker than that in the MB group. Microradiographs in remineralization tests showed that the radiopacities of the decalcified dentin layers in the RB and OB groups were significantly higher than that in the MB group, which does not release fluoride. These results indicated that fluoride-releasing adhesives are effective in inhibiting wall lesions and in enhancing the remineralization of decalcified dentin. Introduction It is well known that the major causes of secondary caries are poor adhesion to the tooth substance or remnant decalcified dentin underneath the restoration (Kidd et al., 1992). In the former case, numerous adhesive systems promoting good adhesion to dentin have been developed and the bond strength to the dentin is improved (Yokota et al., 1996). However, it is doubtful that adhesion between the dentin and the restoration can be maintained for long term in the severe oral environment. In the latter case, the complete removal of decalcified dentin is difficult in general because complete removal may result in dental pulp exposure. Therefore, some decalcified dentin is frequently left in the cavity, with the associated danger of occurrence of the secondary caries even when good adhesives are used. It has been reported that fluoride ions are effective for secondary caries inhibition and can enhance remineralization of decalcified dentin in vitro (Nagamine et al., 1997; Ten Cate et al., 1995). Recently, many fluoride-releasing adhesives have been developed. Giomer, a new fluoride-releasing restorative system based on a pre-reacted glassionomer (PRG) technology, has been developed by Shofu Inc., and is recommended as a restorative system for cervical defects and root caries. These adhesive systems can be expected to prevent secondary caries around restorations due to their fluoride release (Itota et al., 2001). As fluoride-releasing adhesives directly contact the cavity wall, fluoride ions released from them easily penetrate and diffuse into the 62cavity wall dentin. Therefore, the fluoride-releasing adhesives may be beneficial for the inhibition and prevention of secondary caries. The aim of this study was to evaluate the inhibitory effect of secondary caries and the ‘temineralization of decalcified dentin by fluoride-releas ig adhesives, Materials and Methods The materials used in this study are ‘Table 1: Materials used in the study shown in Table 1. Two commercial fluoride- “Sisseratatcode, Manufactured Compesition releasing adhesives, Reactmer Bond (RB, Shoful, Kyoto, Japan) and One-up Bond F (OB, Tokuyama Dental, Tokyo, Japan), and a aaa ‘Renctmor Bond (RB, Shofs) AI, -AITA, PAS One-up Bond F (OB, Tékuyama) 40-19, yhoepats monomer, AS et Mac‘Bond 11 (MB, Tokuyama) MAC-10, phosphate monomer, commercial adhesive without fluoride release, on nriny Mac-Bond 11 (MB, Tokuyama Dental), Were Resctmer Paste (RP, Shafi) UDMA, filer, FRO «i used. Reactmer Paste (RP, Shofu), Lite-Fil I A. ‘Lite flITA (LF, Shofu) UDMA, filer ete (LE Shofu), Estelite (EL, Tokuyama Dental) and Eztelite: (EL, Tokuyama) Pix GMA, filler ete. Clearfill AP-X (AP, Kuraray Medical, Osaka, __ClearfllAPX (AP, Kuraray) Bie Gala flere . PRG = preted grasionomer FASG ~ Nur aluminosilicate gas Japan) was used as the restorative material in S this study. Measurement of fluoride release ‘We measured the fluoride release from specimens coated with adhesives and specimens made of ahesives only. For the former, specimens are coated with adhesive except for the top surface (inner diameter of ‘6mm and depth of 2.5 mm) were prepared. For the latter, specimens made with adhesive only (inner diameter 3.5 mm and depth 2.0 mm) were prepared. After storage at 37°C in 100% humidity for 24 h, these specimens ‘vere individually immersed in 5 mL distilled water and stored at 37°C. The amount of Muoride release from the specimens was measured every week for 10 weeks Inhibitory effect of secondary caries Extracted upper premolars without caries and other defects were used in this study. Class V cavities ‘were prepared at the cement-enamel junction on the buccal and palatal surfaces of each tooth. Cavities were ‘teated with the adhesives according to the respective manufacturers’ instructions and restored with restorative materials followed by light irradiation for 40s. The restorations were polished with a white carborundum point afer storage at 37°C in 100% humidity for 24 h. The specimens were mounted individually on plastic tubes and stored for 14 days at 37°C in 100% relative humidity. Hanks’ balanced salt solution was poured into cach tube. fier storage, each tooth surface was coated with nail varnish except fora peripheral zone around the restoration. ‘They were incubated in brain-heart infusion broth containing 1% sucrose inoculated with Sireplococeus mutans at 37°C for [4 days. After incubation, the specimens were embedded in epoxy resin, cut through the restorations 4nd axial sections approximately 80 jum in thickness were prepared. Contact microradiographs were taken with «soft xray source and the artificial secondary carious lesions at the gingival margin in each root were observed ‘vith a microscope. For evaluation of the inhibitory effect of artificial secondary caries, the thickness of the acid-resistant layer in the root was measured at a depth of 250 pm under the surface of the restoration, and the depth of the outer lesion in the root was measured at a distance of 100 jim from the restoration margin using a microscope attached to a micrometer. 63Remineralization of decalcified dentin Class V cavities were prepared at the cement-enamel junction on the buccal and palatal surfaces of each tooth, and the teeth were mounted individually on plastic tubes. Decalcified dentin was then prepared by a caries inducing system at the cavity wall. Cavities were treated with adhesive systems according to the respective manufacturers’ instructions and restored with a resin composite, AP-X, followed by light irradiation for 40 s. The specimens were stored for 4 weeks at 37°C in 100% relative humidity after polishing of the restoration with a white carborundum point. After storage, the specimens were removed from the tube and embedded in epoxy resin. The specimens were cut through the restorations and axial sections approximately 80 Lum in thickness were prepared. Contact microradiographs were taken and the decalcified dentin underneath the restorations was observed with a microscope. The specimens without restoration and storage for 4 weeks were used as a control. For the evaluation of remineralization of decalcified dentin, the density of decalcified dentin underneath each restoration was analyzed with a computer program. The density in a round field (diameter = depth of the decalcified dentin layer) of the decalcified dentin was measured. Sound dentin underneath the decalcified dentin layer was also analyzed using the same specimen. The ratio of the radiopacity of decalcified dentin/ sound dentin was used as the degree of remineralization of decalcified dentin in this study. Results The amounts of fluoride released from each group during 10 weeks are shown in Fig. 1. Among the specimens coated with adhesives, the RB/RP and RB/LF groups showed higher amounts of fluoride release than the OB/EL group at all test periods. No fluoride release was detected in the MB/EL group. However, among the specimens made by adhesive only, the amount of fluoride release from the OB group was larger than that of the RB group. In the MB group, no fluoride release was detected at any time. Fig..1 Adhesive / Restorative NT oive —H RB/RP aE RBILE hk OBJEL —&- MBIEL, amount of fluoride released (ppm) " amount of fluoride released (ppm) _ he Ad A dh A A A A A A 2 4 6 8 10 0 2 4 6 8 10 immersion time (week) immersion time (week) Cumulative amount of fluoride release from the specimensFig. 2 shows typical microradiographs of the lesions in roots. In all specimens, there were outer lesions onthe exposed tooth surface but no wall lesions were observed in the root. Moreover, acid-resistant layers were clearly observed in the lesions adjacent to the restorations of RB/RP, RB/LF and OB/EL groups, but not clear in the MB/EL group. Fig. 2 ‘Typical microradiogeaphs ofthe inhibitory effect study The mean thickness of the acid-resistant layers and the mean depth of outer lesions are shown in Table 2, The layers in the RB/RP, RB/LF and MB/EL groups were significantly thicker than in the MB/EL group. ‘There was no significant difference among the RB/RP, RB/LF and OB/ L groups. The outer lesions in the RB: RP group were significantly the shallowest of all groups. Table 2: The thickness of acid-resistance layers and the depth oF outer lesions There was no significant difference among RB/LF,OB/— ~aiiaic Tike EL and MBVEL groups in the depth of the outer lesion. Fig. 3 shows typical microradiographs of i decalcified dentin in each group. In the RB/AP group and ae aul OB/AP group, the radiopacity of the layer was higher than oe era sores that in the MB/AP or control groups. The means of the er set sisei0 ratio of radiopacity in the RB/AP and OB/AP groups were significantly higher than those in the MB/AP or control groups (Fig. 4). There were no significant differences between the RB/AP and OB/AP groups, and MB/AP and control groups. Fig, 3 Fig. 4 | s | 4 2 06 : 3 I 4 2 | 3Discussion and Conclusion In this study, fluoride-releasing adhesives inhibited the formation of wall lesions. It was reported that poor adhesion allows marginal leakage to create a wall lesion (Gilmour et al., 1993; Grieve et al., 1980). Therefore, the absence of wall lesions shows that the fluoride-releasing adhesive systems produced good adhesion to the cavity wall dentin. Moreover, thicker acid-resistant layers were recognized adjacent to the restorations at the cavity wall dentin in the RB/RP. RB/LF, and OB/EL groups. Hsu ef al. (1998) reported that an inhibition area along the cavity wall was obser in fluoride-releasing restorative materials. This layer was formed by the fluoride ions released from the RB or OB, and resisted acid-attack during artificial caries formation. Therefore, a fluoride-releasing adhesive is desirable for inhibition of wall lesions because of the reinforcement of the wall dentin by fluoride release. On the other hand, outer lesions were formed on the exposed tooth surface around the restoration in all specimens. The formation of outer lesions is caused by S.mutans in the medium (Zambon & Kasprzak, 1995). Inhibition of outer lesions was recognized in the RB/RP group only. The inhibition of outer lesions is caused by the uptake of fluoride to the root surface (Dijkman & Arend, 1992). 4 high concentration of fluoride around the restoration is thus required for the inhibition of outer lesions. The RB/RP group was restored with RP, a fluoride- releasing restorative material, which was exposed directly to the medium and probably supplied a high concentration of fluoride ion at the root surface. In the experiment of remineralization of decalcified dentin, the RB/AP and OB/AP groups with fluoride telease exhibited higher radiopacity than the MB/AP or control groups. The high radiopacities of microradiographs on the decalcified dentin layers suggest remineralization of the decalcified dentin by fluoride ions released from the adhesive resin. This suggestion may be supported by the previous finding that fluoride ions enhanced remineralization of carious dentin (ten Cate & van Duinen, 1995). In addition, a radio-opaque layer was observed in the decalcified dentin layer in RB/AP group. This radio-opaque layer may be produced by the deposition of inorganic components as aluminum or strontium released from the RB containing glassionomer filler. The deposition of these components would occur at the decalcified dentin layer in the RB/ AP group. Therefore, this radio-opaque layer may contribute to inhibition of the progress of secondary caries underneath the restoration. In conclusion, it was found that the use of fluoride-releasing adhesive systems was effective in inhibiting wall lesions and enhancing the remineralization of decalcified dentin. References * Dijkkman GEHM, Arends J. Secondary caries in situ around fluoride-releasing light-curing composites: A quantitative model investigation on four materials with a fluoride content between 0 and 26 vol%. Caries Research 1992; 26: 351-357. * Gilmour ASM, Edmunds DH, Newcombe RG, Clark MF. An in vitro study into the effect of a bacterial artificial caries system on the enamel adjacent to composite and amal gam restorations, Caries Research 1993; 27: 169-175. * Grieve AR, Jones JC. An in vitro study of marginal leakage associated with composite restorations using an acidified agar technique. Journal of Oral Rehabilitation 1980; 7: 215-223. * Hsu CYS, Donly KJ, Drake DR, Wefel JS. Effects of aged fluoride-containing restorative materials on recurrent root caries. Journal of Dental Research 1998; 77: 418-425. + Itota T, Nakabo S, Iwai Y, Konishi N, Nagamine M, Torii Y, Yoshiyama M. Effect of adhesives on the 66\ i 2 Fluoride recharge effect of various dental materials: Fluoride release an inhibition of caries progression Katsushi Okuyama, Ph.D., D.D.S. Katsushi Okuyama"), Yasuhiro Murata’), Pereira NR Patricia”, Hisanori Komatsu”, Hidehiko Sano! P Division of Cariology and Operative Dentistry, Department of Oral Health Sciences, Hokkaido University Graduate School of Dental Medicine (Japan) 2 Department of Operative Dentistry, School of Dentistry, The University of North Carolina (USA) Abstract Objectives: The purpose of this study was: (1) to evaluate the fluoride release and recharge from Giomer materials (Reactmer Paste and Beautifil) and (2) to investigate the effect of Giomer materials on inhibition of caries progression. Material and Method: The amount of daily fluoride release into distilled water from two Giomers, two glass- ionomers (Fuji 2 and Fuji? LC) and resin composite (AP-X) were measured up to 3 weeks. After 3 weeks, the specimens were immersed into 1000 ppm NaF solution for 5 min. Half of the specimens from each material were immersed into fluoride solution once, and the other half were done this procedure for 2 weeks consecutively. For the inhibition of caries progression, two bonding agents, containing fluoride (Reactmer Bond: RB) and non-fluoride containing (non-F)(Clearfil Mega Bond: MB), and two resin composites containing fluoride (Reactmer Paste: RP) and non-F (Clearfil AP-X: AP) were used. Each combination of a bonding agent and a resin composite, RB/RP (RR), RB/AP (RA), MB/RP (MR) and MB/AP (MA), was filled in a cavity on root dentin of extracted human molars. The pH cycling, the model of alternating demineralization (D: pH4.5) and remineralization (R: pH7.0), was done on a daily schedule of three cycles of [hr-D and 4hrs-R and followed by 9hrs-R storage. Half of specimens were immersed into 0.05% NaF solution for Imin once a day as an F mouth rinse simulation. After the pH cycling for 6 weeks, specimens were observed using contact microradiography, Results: Giomer materials and glass-ionomers released some amount of fluoride. High amount of fluoride released from two Giomer materials was measured, immediate after single fluoride immersion. Afier consecutive fluoride immersion, higher amount of fluoride release from the Giomer material was maintained compared to fluoride release before repeated fluoride recharging. The microradiographs showed the remineralized layers when fluoride containing materials or fluoride immersion were used. Conclusions: The Giomer materials were found to be effective for artificial caries prevention. Moreover, when combining with fluoride application, Giomer materials showed more effective caries prevention due to fluoride recharge.Introduction The cariostatic effects of fluoride have been Table 1: List of Mater used measurement of fluoride rel ‘well established (Ogaard B et al., 1986; Ogaard B, Trade name Material type Manufacturer Code 1990; Ten Cate JM, 1990; Inaba D, 1992; Heilman. Reacimer Paste Fluoride containing SHOFU. Kyoto, pe comgesie Goren apon IR et al., 1997: Teranaka T et al., 1987). For = . Senshi Forse ontainng SHOPS BT prevention of dental caries, various fluoride comooste (Gomer) lapan eae a Fujtonomer eco, F2 applications are available: water fluoridation. topical aiona coneuaatad ¢ Te fluoride application to tooth surface, fluoride mouth Fullonamer —pesinmeaneg ge EET, «LC rinse and dentifrice (Araki H et al., 1986; Thylstrup eA, ee Actal., 1994), Many reports have proved that dental Clea aP-x —_—Resincomgante “uray onaka, AP caries is decreased by fluoride applications (Stamm, IWetal., 1990; Locker WG et al., 1989; Hunt RJ et al., 1989; Burt BA et al., 1986; Weerheijm KL etal., 1997), Based on the clinical observations that the teeth treated with silicate cement showed relatively low rate of secondary caries in the area adjacent to the cement (Lind et al., 1964), some materials containing fluoride \were emphasized as a great benefit for prevention of dental caries, Although silicate cement released some amount of fluoride, the cement disappeared in the market because of its low mechanical properties and no adhesion to tooth structure. Since the introduction of glass-ionomer cement, the use of dental materials containing fluoride is receiving increased emphasis again. Glass-ionomer cement is a popular fluoride containing material. This cement has some advantages, such as adhesive properties, fluoride release and uptake (Mount GJ, 1994; 1999), However, it has some disadvantages, such as poor mechanical properties, poor bond strength to enamel and denti initial sensitivity to moisture and desiccation and aesthetic properties (Mount GJ, 1994: 1999). Recently, new fluoride containing materials are available, Resin-modified glass-ionomer cement was developed to over come the disadvantages of conventional glass-ionomer cements (Saito S et al., 1999; MeCabe JF, 1998; Sidhu SK et al., 1995). Other fluoride containing materials were poly-acid modified resin composite (Saito S et al., 1999; Meyer JM et al., 1998) and fluoride containing resin composite (Eichmiller FC et al., 1998). These fluoride containing materials ate reported to release fluoride to some extent (Vieira AR et al., 1999; Aoyagi et al., 1997; Grobler SR et al. 1998; Carvalho AS et al., 1999; Milller U et al., 2000), Fluoride recharge is the uptake of fluoride into a material and release from the material again (Eichmiller FC et al, 1998: Gao W et al., 2000). In clinical situation, the recharge of the restoration occurs during topical fluoride application (Kupietzky A et al., 1994) and brushing with fluoride dentifrice. The potential of recharge is very important to maintain and enhance the fluoride release from material. Glass-ionomer cement releases a large amount of fluoride after recharge due to its high potential of recharge (Takahashi K et al., 1993; Creanor SL et al, 1994; 1995; Diaz-Arnold AM et al., 1995; Gao W et al., 2000; Rothwell M et al., 1998). On the other hand, recharge of resin composite is poor due to its low potential of recharge (Takahashi K et al., 1993). For this reason, materials which are superior in fluoride release and recharge, have been in demand Recently, PRG technology was developed by Shofu company. Using PRG technology, the stable glass- jonomer phase is made by pre-reacted glass filler and poly acid. This technology is divided into two types. One is F-PRG; whole glass was reacted. The other is S-PRG; surface of glass filler was reacted. PRG filler has induces fluoride release and recharge, this action is a superior characteristic of PRG (Tay FR et al., 2001). In this paper, two objectives were investigated. One is the fluoride release and recharge from materials. 69The other is the effect of various materials on inhibition of caries progression. Materials and methods ; (1) Fluoride release from various dental restorative materials Materials Five materials were used in this study (Table |). Two Giomers; Reactmer Paste (Shofu, Kyoto, Japan) and Beautifil (Shofu, Kyoto, Japan), two Glassionomner cements; Fuji ionomer Type II (GC, Tokyo, Japan) and Fuji lonomer type I] LC (GC, Tokyo, Japan) and a resin composite; Clearfil AP-X (Kuraray, Osaka, Japan) were examined. Methods Cylindrical test specimens were made by placing each material into a plastic mold, with inside dimensions of 9mm diameter and 3mm height. All materials were prepared according to the manufactures’ instructions. After setting, the specimens were neither varnished nor finishing, but were transferred to humidity condition at 37°C. After one hour, the specimens were put individually into plastic beakers with 8 ml distilled deionized water and incubated at 37°C without shaking. After one day, the specimens were removed from the beakers, washed with 2 ml distilled water and then stored in 8 ml of fresh distilled water. This procedure was done up to 3 weeks. Three weeks after, samples were washed in 2ml fresh distilled water and then exposed to 1000ppmF (0.211%) NaF solution for 5 min at room temperature. Following the fluoride exposure, these samples were washed in non-fluoride containing water and then were placed in 8 ml of distilled water. All samples were divided into two groups. For one group, this procedure was done only once, for the other group, this procedure was repeatedly done for 2 weeks. Sample solution was taken 3 ml from each storage solution and used for measuring the fluoride concentration. These were mixed with 0.3ml of TISAB III buffer solution (Orion Research, Inc.) and analyzed for fluoride using an ion-specific electrode (96-O9BN, Orion Research, Inc.) connected to an expandable ion- analyzer (pH/ISE meter model 710A, Orion Research, Inc.). The amount of fluoride release from the various materials was expressed as wg/mm’. (2) Inhibition of caries progression using fluoride releasing materials Materials Two adhesives, Reactmer Bond (fluoride-releasing: Shofu, Kyoto, Japan) and Clearfil Mega Bond (non-fluoride releasing: Kuraray, Osaka, Japan), and two composites, Reactmer Paste (fluoride releasing: Shofu, Kyoto, Japan) and Clearfil AP-X (non-fluoride resin composite: Kuraray, Osaka, Japan), were used in this study (Table 2). Using these materials, four combined restorative procedures were used; Reactmer Bond and Reactmer Paste (RR), Reactmer Bond and Clearfil AP-X (RA), Clearfil Mega Bond and Reactmer Paste (MR), Clearfil Mega Bond and Clearfil AP-X (MA) (Table 3). ‘Table 2: List of materials used for investigation of caries inhibition Table 3: Combination of materials for filling Combination ‘Giga Reactmer Bond (SHOFU: Fluoride releasing eer + Bonding agent J Reactmer Bond + Reactmer Paste RR Clearfil MegaBond (Kuraray) Non-fluoride Reactmer Bond + Clearfil AP-X Be Reactmer Paste (SHOFU) Fluoride releasing Filling material Clearfil MegaBond + Reactmer Paste MR Clearfil AP-X (Kuraray) Non-fluoride Clearfil MegaBond + Clearfil AP-X MA 70Methods These combined materials were applied into prepared cavity on root dentin (width: 2mm, length: 4mm, depth: Imm). Soon after that, the pH cycling model was carried out. The pH eycling was carried out on a schedule of three cycles of 1-hour demineralization and 4-hours remineralization, and followed by 9 hours remineralization storage. After the third demineralization period, half of the samples were immersed in 0.05% sodium fluoride solution for 1 min, The demineralizing and remineralizing solutions were used as follows: The demineralizing solution contained 3.0mM CaCl,, 1.8mM KH;PO., 2% carboxymethylcellulose sodium and 0.2M Lactic acid, This solution was adjusted with IM NaOH to pH of 4.5. The remineralizing solution contained 3.0mM CaCh, 1.8mM KH,PO,, 2% carboxymethylcellulose sodium and 0.02M HEPES (2- (4-(2-hydroxyethyl)- L-piperazinyl] ethanesulfonic acid). The remineralizing solution was adjusted with 1M NaOH to pH of 7.0. The caries progression test that employed pH cycling was performed for 6 weeks. Following the test, samples were Aichydrated, embedded, and sectioned into 150420 yun thick specimens. All sections were subjected to contact microradiography (CMR) observation. Results (1) Fluoride release Figure 1 shows the amount of fluoride release and recharge from materials with single fluoride immersion. The fluoride release patterns of different fluoride containing materials were similar, but the amount. of fluoride release was different among each material, The fluoride release was highest at the First day. Dramatic decrease of fluoride release was found alter 2 or 3 days. After 14 days, the fluoride release reached a plateau. On the 21st day, the specimens were immersed in fluoride solution for Smin. Then, the fluoride release was increased on the next day of fluoride application, but two days later fluoride release was decreased again, The amount of fluoride release from Reactmer and Beautifil is shown in Figure 2. Both materials showed the highest amount of fluoride release after fluoride recharge (22nd day). The amount of fluoride release of 42nd and $6th day was higher than before fluoride immersion. Fig. 1 Fig. 2 ee i ese : is i aaa i 7 won mo « bmn eae es ee eee ay) ate 21 days (42y) and 35 days 6-9) The amount of fluoride release from various dental materials that were exposed to repeated fluoride application is shown in Figure 3. Before fluoride application, the fluoride release pattern was similar to that showed in Figure J, The repeated fluoride recharge tended to inerese the amount of fluoride released from two {lass-ionomer cements, On the other hand, for Reactmer and Beautifil, the amounts of fluoride release increased only for the first few days. However, there after, the amount of fluoride release was constant, Figure 4 shows the amount of fluoride released from Reactmer and Beautifil with repeated fluoride recharge. Both materials, after the initiation of repeated recharge, the amount of fluoride release was higher than before fluoride immersion However, there was no significant difference in the amount of fluoride release between I-and 14 days after repeated recharge. 7Bigs d Fig.4, z ——RE -*8T < E of, -s-F2 LC Ss S { AP os 3a S | = ae 325 ans i = aol | be a 0a) a Start repeated F application 1000ppmF: NaF solution 3 ¢ 3 Smin, immersion 2 1 i é 8, 107 0.88 2 S (on) tory 3 = = ae & = a Q 7 Bers Ftd Ftd Bere Fad add 0 7 14 at m8 m6 Mean(sn) RP BT Inpersion period Faye) Values connected by herizontal lines are not significantly different (a>0.05) {n=5) Daily amount of fluoride released from materials with 2 weeks Daily amount of fluoride release from RE and BT with 2 weeks y Me continuously fluoride application continuously fluoride application at before fluoride immersion (1-day), first immersion day (F-Id: 22-day) and L4th day (F-14d: 35-day) (2) Inhibition of caries progression Figure 5 shows the image of the CMR on RR and RA without fluoride application, respectively. A radiopaque layer adjacent to cavity wall is shown in both materials. However, the radiopaque layer on the outer lesion is only indicated in the materials (Fig. 5-a). Figure 6 shows the image of the CMR on MR and MA without fluoride application, respectively. With both combinations, the radiopaque layer was not shown from adjacent area to cavity wall. The images of the CMR on RR and RA with fluoride application are shown in Figure 7. There were radiopaque layers on the outer lesion and adjacent area to cavity wall on both materials. The images of the CMR on MR and MA with fluoride application are shown in Figure 8. Comparing with the same material without fluoride application (Figure 6), the radiopaque layers on the outer lesion were enhanced. However the radiopaque layer adjacent to cavity wall was not created. Fig. 5 (a) Fig. 6 (a) Fig. 6 (b) Microradiograph of specimens afler pH-cycling without fluoride Microradiograph of specimens after pH-cycling without fluoride application (a: RR, b: RA), R: Restorative material, D: Dentin, Arrow a: application (a: MR, b: MA), R: Restorative material, D: Dentin, Arrow radiopaque layer adjacent to cavity wall. Arrow b: the radiopaque layer b: the radiopaque layer on the outer lesion on the outer lesion Fig. 7 (a) Fig. 8 (a) Fig. 7 (b) Fig. 8 (b) Microradiograph of specimens after pH-cycling with fluoride application Microradiograph of specimens afier pl I-cycling with fluoride application (a: RR, b; RA), R: Restorative material, D: Dentin, Arrow a: radiopaque (a: MR, b: MA), R: Restorative material, D: Dentin, Arrow b: the layer adjacent to cavity wall Arrow b: the radiopaque layer on the outer radiopaque layer on the outer lesion lesion 72Table 4 indicates whether there was a radiopaque layer on the lesion or not on each of the combined restoratives. A circle indicates that a radiopaque layer was exhibited on ‘Table 4: Radiopaque layers on lesions the lesion. X indicates that a radiopaque layer was not within the lesion, aa ‘When using a fluoride application or Reactmer Paste, a radiopaque layer ee |e ‘was found on the outer lesion, When the Reactmer Bond was applied to ha, =|» cavity, a radiopaque layer was shown the area where adjacent to cavity a ae wall. wR ol» wa ol. (Rasen eras pet x Tl wots Discussion (1) Fluoride release Many reports are available for the fluoride release from dental materials (Creanor SL et al. 1994, 1995; Takahashi et al., 1993; Vieira AR et al., 1999; Aoyagi et al., 1997; Grobler SR et al., 1998; Carvalho AS etal., 1999; Diaz-Arnold AM et al., 1995; Gao W et al., 2000, Miiller U et al., 2000). Recently, many fluoride- containing resin composites are marketed so that the amount of fluoride release from these material were measured (Vieira AR et al., 1999; Aoyagi etal., 1997; Grobler SR et al., 1998; Carvalho AS et al., 1999; Miller Uctal., 2000}. However, most of them showed that the amount of fluoride release from glass-ionomer cement was greater than that of from resin composite (Vieira AR et al.. 1999; Aoyagi et al., 1997; Grobler SR et al., 1998; Carvalho AS et al., 1999:). The results of this study were in agreement with previous studies. However, it siill unclear whether or not amount of fluoride release from Reactmer or Beautifil was larger than that of from other fluoride-containing resin composite. ‘When fluoride immersion was performed, higher amounts of fluoride release from Reactmer or Beautifil ‘were shown than those before fluoride immersion. This result indicates that the Giomer material has the fluoride recharge effect due to the PRG-filler in the materials. Repeated fluoride recharge maintained higher amounts of fluoride release from Reactmer or Beautifil compared to fluoride release before repeated fluoride recharging, It is found from the result that long-term fluoride release could be improved by repeated fluoride recharging. It is well known that glass-ionomer cement takes up and re-releases fluoride (Takahashi et al., 1993; Creanor SL etal, 1994, 1995; Diaz-Arnold AM et al., 1995; Gao W et al., 2000). Eichmiller FC et al, (1998) explained the mechanism of this recharge effect was that extrinsic fluoride was deposited back into the ionomer, re-supplying the release from the ionomer into the surrounding environment. On the other hand, Gao W et al, (2000) indicated that the mechanism of fluoride recharge was still unclear. Future research will have to show the mechanism of HMuoride recharge. Glass-ionomer cement was called as a reservoir of fluoride (Mount Gl, 1999). It was thought that glass-ionomer cement serve fluoride to oral environment. The Giomer materials also have a potential of fluoride reservoi (2) Inhibition of caries progression ‘When glass-ionomer cement was filled in a dentin cavity, the radiopaque layer was appeared on the outer lesion and in area adjacent to cavity wall (Nagamine M et al., 1997; Tam LE et al., 1997; Ten Cate JM et al, 1995; Okuyama et al, 2001: Torii ¥ et al., 2001). This layer wagcalled as remineralized layer (Ten Cate JM etal., 1995; Okuyama K et al., 2001) or acid-resistant layer (Nagamine M et al., 1997; Tam LE et al., 1997; Torii Y ct al., 2001). In other word, this layer indicated an effective layer for inhibition of caries progression. ‘This layer might related with pH eycling (immersion in de- and remineralized solution) and fluoride (Okuyama Ketal., 2001). Fluoride is an important factor to create this layer. Glass-ionomer cements were suggested to Breduce or prevent secondary caries, because the cements include fluoride and create the radiopaque layer around materials. Other fluoride containing materials (resin-modified glass-ionomer cement, poly-acid modified resin composite (compomer)) were also believed to have a possibility for caries prevention because of same reasons (Dunne SM et al., 1996; Nagamine M et al., 1997; Tam LE et al., 1997; Creanor SL et al., 1998: Pereira PNR et al., 1998; Torii ¥ et al., 2001). From the results of this study, some conclusions about the inhibition of caries can be drawn. Regarding the outer lesion, when fluoride application was done, all specimens had the effect of the inhibition of artificial caries, Moreover, fluoride containing filling material also seemed to be effective to caries inhibition without fluoride application. Regarding the area adjacent to cavity wall, when only using fluoride containing bonding agent, this material made effective to caries inhibition, whether fluoride application used or not. Since fluoride from the restorative material might not have sufficiently penetrated into the cavity walls directly, there was observed no radiopaque layer adjacent to cavity wall (Okuyama K et al., 2001). Presumably some amount of fluoride from restorative material could not go through Clearfil Mega Bond. The fluoride recharging effect was unclear due to fluoride application, since fluoride application was more effective than fluoride recharge from materials. Figure 9 shows that the supposition of fluoride shifting around tooth structure and materials (Okuyama K etal., 2001). Arrows indicate the shifting of fluoride. Fluoride moves material to the tooth structure directly or through a surrounding oral fluid (i.¢., saliva). A radiopaque layer adjacent to cavity wall may be developed by directly applying fluoride from material. The radiopaque layer on the outer lesion may be developed by fluoride application and released fluoride from material. Fig. 9 Fluoride application Fluoride ———_» | Material Tooth structure! From “Caries preventive restoration" forum in Sapporo “99: 69-80, 2001 Schematic drawing of shifling fluoride around fluoride containing-materials. Arrows indicate shifting fluoride, Conclusions There are three findings from our research. 1) In Giomer materials, there was the fluoride recharge effect. 2) When repeating the fluoride application, high amounts of fluoride recharge remained. 3) Giomer materials were effective to artificial caries inhibition. According to these findings, the Giomer materials seemed to be effective materials for caries prevention. Moreover, when combining with fluoride application, Giomer materials appeared more effective to caries prevention due to fluoride recharge. 74(DISCUSSION: Session 2) Chairperson: Dr Tagami DrTagami Thank you very much, Dr Okuyama for a very interesting presentation. Now, we are running about 20 minutes behind the schedule, but I think we can take sometime for discussion, At the beginning of the discussion, 1d like to introduce you our special guest today, Professor David Pashley from Medical College of Georgia in the United States. Many of you in the audience and the speakers have collaborated with him, and used to be students of Dr Pashley, and I'm very happy as one of his former students. I’m very happy to introduce good news that he was awarded Wilmer Souder Award at the [ADR meeting this year. I'd like to ask Dr Pashley to give some comments. Dr Pashley: ‘Thank you. This morning has been very, very interesting. Particularly the last three presentations were fascinating because they were clearly showing, that this restorative material can release fluoride rapidly - perhaps too rapidly compared with mine - and that this material can recharge fluoride. [ have a number of thoughts and questions. For instance, over the first week of the research, what was the percentage of the total fluoride being released? Secondly, when you recharged the material with fluoride, did it go above the percentage, which was recorded during the first week, or above one hundred percent? Thirdly, what was the type of the fluoride? Does the recharge matter if you use neutral fluoride in stead of acidic fluoride? Could you trap the fluoride ion with calcium? There are several interesting things to discuss. Well, there has to be a rational, ideal pair between the type of adhesive to be used and the type of fuoride-releasing material to be put on top of it. For instance, if you use Reacimer paste as restorative material, but with Fluaro Bond as adhesive, you would have very little fluoride going through the bonding agent, thus have very little fluoride get into the dentin. So, if your goal is to get Auoride into the dentin, then you need a very special kind of adhesive, and that is what Reactmer Bond is. It's a very interesting bond. I don’t know the detail of it, but it clearly permits bi-directional fluxes of fluoride. So you obviously have to have permeable adhesive if you want the fluoride in the restorative to get into the dentin. So you have to have a special kind of adhesive, and it perhaps needs to be made very hydrophilic, and there may be some certain disadvantages because of the hydrophilicity ~ water sorption being one. However, after you apply the bond and put the restorative material on the tooth, what would happen? For in order to trap the fluoride in the restorative, the sealing must be tight, so that the fluoride won't go out so fast. On the other hand, torecharge fluoride, bonding agent must be permeable. Inoticed that in many of the microradiographs, there were a lot of cases of dentin shrinkage. Did you notice that? Which is indicating that the material got a high water content in the water storage. If they have soaked those specimens in ethylene glycol before they stored them in the water, you know, they may have got slightly different results - perhaps somewhat better results. But it an interesting development that must be evaluated clinically. You have a good deal of in vitro information, and you have some in vivo information, but the final test should be done in the mouth - in all kinds of mouths, and in all kinds of patients. And Giomer products may ‘umn out to be very useful for people with low salivary flows in geriatrics, and perhaps for paediatric patients. And it remains to be seen exactly how Giomer products should be used, and how we can manipulate them to maximise all the properties of the material for various treatments that we want to do with it. Anyway, I want to congratulate Shofu for their beautifull R&D effort in developing this fascinating material that obviously will get 75better over time. Thank you. Dr Tagami: OK. Now I think it is available to have a few questions or comments before lunch. Yes, please. Attendee: : we : Thanks for the interesting results. As Dr Pashley was telling us, I was so interested in the results of fluoride diffusion through the bonding agent. So can anybody explain it to me? Dr Teranaka, or anybody? Please. Dr Teranaka: I'm not sure of the detail, but the polymers, consisting of the cured bonding agent are porous enough for the fluoride ion to diffuse. As the water goes down from higher place to lower place, the higher concentration of fluoride in the material moves to the lower concentration part through the cured bonding agent, too. What shall I say, an equilibrium of fluoride concentration will be achieved. Dr Tagami: Thank you. Yes, please. Dr Pashley: Thank you very much. In my opinion - it’s just a guess - fluoride is moving through water-filled pores, or water- filled channels. Dr Tay and I have been doing some very interesting works, in which we simply seal dentin with resin - just an adhesive layer, which is maybe 50 microns thick. And then, we tried to push water through the layer, and the layer sealed pretty good. But, when we put hypotonic calcium chloride solution on top of that adhesive, water was pulled right thorough the dentin and right through the adhesive, and I could not believe that. Therefore, we did it over and over with many different materials. But we have not done it yet with Reactmer. However, I suspect that we'll be able to move water through Reactmer relatively easily. So if water can go through Reactmer, fluoride can go through Reactmer, also. And that’s good when you want to increase the amount of fluoride being taking up by dentin. Dr Tay and I have been working on a lot of things - for instance, if you use silver nitrate, you can trace these channels for fluoride shift. And we've intended to trace these channels with radioactive molecules of different sizes. But the adhesive layers are not glassy, solid impermeable teflon membranes. They have some spaces in them. Sometimes when you think that you are evaporating all solvent, you are really not. And that may be important that there is some spaces left there. It’s just a guess, but an educated guess. Dr Tagami: Thank you very much. Dr Fujitani, please. Dr Fujitani: Thank you very much for your wonderful, informative presentations. I’d like to ask three of the presenters. Actually, Reactmer showed higher rate of fluoride releasing, and high inhibition efficacy against demineralisation of dentin. But the conventional glass ionomer cement showed much higher value of fluoride releasing - so my question is, how much amount of fluoride releasing is necessary for caries inhibition? This is a very basic 76