RESEARCH dental materials Clinical performance of a condensable metal-reinforced glass ionomer cement in primary molars

N. Krämer,1 R. Frankenberger,2

Objective Aim of the present study was to evaluate the clinical resistance, moisture susceptibility, and inferior flexural strength suitability of the condensable metal-reinforced glass ionomer characteristics lead to the development of several modifications in cement Hi-Dense in classes I and II cavities of primary molars. the field of conventional GICs.12–14 Methods Seventeen children received a total of fifty four Hi- Metal reinforced glass ionomer cements (e.g. Ketac Silver, Espe) Dense fillings (nineteen class I and thirty five class II). The contain silver-tin alloy fibers or flakes being sintered to the glass- restorations were clinically assessed at baseline, after one and particles for a better incorporation of the metal particles into the after two years of clinical service according to modified USPHS cement matrix.15,16 In both cases the metal particles should codes and criteria. The restorations were replicated in each recall improve the mechanical properties of the cements and increase and representative samples were qualitatively analysed under a their wear resistance.17-19 SEM. Due to the amalgam-like optical performance, metal-reinforced Results Over the observation period of two years, five materials showed an enormous commercial success and were conse- restorations failed due to total retention loss, two fillings needed quently used as posterior materials for the primary dentition as replacement because of persisting hypersensitivity, one filling well.2 A clinical study of Kilpatrick,20 however, demonstrated that was lost because of an unsuccessful endodontic treatment, and Ketac Silver was inferior to the conventional non-metal-reinforced four restorations remained intact until natural exfoliation (Two glass ionomer cement Ketac Fil in Class II cavities. Clinical reports year survival rate: 92% for Class I and 66% for Class II). The SEM dealing with non-resin-modified glass ionomers in Class II restora- analysis of surfaces and marginal areas exhibited an inferior tions of primary molars show unfavourable results in general (Table adhesive performance primarily in proximal areas, whereas a 1).21–23 negative step formation due to wear was frequently observed in Further development of glass ionomer cements focussed on a occlusal parts. higher powder-to-liquid ratio, a lower water content, and smaller Conclusions The results clearly indicate that the condensable, glass particles leading to high viscosity glass ionomer cements (non- metal-reinforced GIC Hi-Dense reveals no enhanced performance metal-reinforced: Fuji IX, Ketac Molar; metal-reinforced: Hi Dense) and lifetime expectancy for class II restorations in primary molars which should be packable like amalgam and reveal enhanced flex- when compared to other non-resin-modified GICs. ural strength characteristics.13,18,23 Various in vitro studies demon- strated that both wear resistance and flexural strength have been he efforts to replace amalgam as a restorative material for pri- enhanced for the viscous metal-reinforced material Hi-Dense com- Tmary molars increased during the last decade.1–3 Especially in pared to Ketac Silver.2,14,18 However, this observation is only Germany the need for alternative materials arose after the Federal reported for initial flexural strength. After cyclic fatigue of the mate- Institute for Drugs and Medical Devices (so-called BfArM, Govern- rials Hi-Dense and Ketac Silver no differences were found ment Health Office) recommended to absolutely restrict the use of anymore.14 amalgam in the primary dentition.4 Currently several materials can Apart from material properties alone, success in children´s be considered as alternatives to amalgam in the primary dentition: restorative therapy primarily depends on the compliance of the conventional glass-ionomer cements (GIC), cermet cements, metal young patients.15,16,23 Consecutively, moisture control frequently is reinforced GIC, resin-modified GIC, resin composites, and poly- only possible for a limited period of time. Hickel and Voss showed acid-modified composites (PMC, compomers).5–7 that Ketac Silver and amalgam restorations performed clinically Glass ionomer cements, e.g. Ketac Fil, Chemfil (Espe, Seefeld, similar in younger children with lower compliance.1 Germany), Fuji II (GC, Tokio, Japan), are classical water-based The fluoride release of conventional glass ionomer cements is cements with an acid/base setting reaction of aluminumsilicate estimated higher compared with light-curing glass ionomer glass and a polyalkenoic acid.8–10 Almost twenty years ago, origi- cements and compomers.24,25 Especially in young children the fluo- nally developed as succeeding material for silcate cements, the fluo- ride release may not only help to prevent secondary caries around ride release and the ion-exchange adhesion with both tooth the filling but may also reduce the risk of caries at the surface of the structures have been reported to be the main advantages of this par- adjacent tooth.25 ticular group of materials.11 However, the evidently low wear The successful use of viscous glass ionomer cements in the ART- technique (Atraumatic Restorative Treatment) made them a 1Associate Professor, 2Associate Professor, Polyclinic for Operative Dentistry and promising alternative to amalgam in the primary dentition of very Periodontology, University of Erlangen-Nuremberg, Glueckstrasse 11, D-91054 young or uncooperative children.26,27 However, except for the ART ErlangenGermany studies no results of clinical long-term trials with viscous glass *Correspondence to: N. Kramer ionomers are published in the literature of the field.8,15 E-mail: [email protected] Therefore the aim of the present study was to evaluate the clinical REFEREED PAPER Received 22.09.99; Accepted 03.04.00 performance of the metal-reinforced, viscous glass ionomer cement © British Dental Journal 2001; 189: 317–321 Hi-Dense in Class I and Class II cavities in primary teeth.

BRITISH DENTAL JOURNAL, VOLUME 190, NO. 6, MARCH 24 2001 317 RESEARCH dental materials

Table 1 Results of comparable studies with glass ionomers in Class II Table 2 Modified USPHS codes and criteria cavities Description USPHS criteria Author Period Material Survival rate Clinically perfect ‘alpha’ (A) present study 2 years Hi-Dense 66% Few defects, correction impossible without ‘bravo’ (B) Kilpatrick 1995 2.5 years Ketac-Fil 77% damage of tooth or restoration. Ketac Silver 59% No negative effects expected Hickel and Voss 1990 3 years Ketac Silver 58% Holst 1996 3 years Ketac Silver 46% Severe defects, prophylactic removal for ‘charlie’ (C) Espelid 1999 3 years Ketac Silver 74% prevention of severe failures Immediate replacement necessary ‘delta’ (D)

Method that had dropped out, natural exfoliation, and other reasons for All patients were treated in the Polyclinic for Operative Dentistry replacement of the restoration (e.g. new proximal lesion indepen- and Periodontology, University of Erlangen-Nuremberg, by one dent of evaluated filling) were estimated as censored at the last recall clinician (associate professor) having experience with restorative appointment and did not cause step formations in the Kaplan- procedures in pediatric dentistry. The patients agreed to a recall Meier curves. programme of two years consisting of baseline, one year and two Representative samples of failed restorations were investigated years recalls. using scanning electron microscopy (SEM). Epoxy replicas (Epoxy- Seventeen children were included in the clinical trial. A total of Die, Ivoclar, Schaan, Liechtenstein) were sputtered with gold (Balz- fifty four restorations was placed by one operator. The mean age of ers SCD 40, Balzers, Liechtenstein) and viewed under SEM (Leitz the children was 6.9 years with a range of 3–11 years. Nineteen fill- ISI 50, Akashi, Tokyo, Japan) at up to X200 magnification. ings were placed in Class I cavities and 35 fillings in Class II cavities. None of the patients dropped out of the study. Results The preparations of the cavites were performed with slight The results of the clinical evaluation at the latest examination are undercuts without bevelling of the margins using 80 µm diamond summarized in Table 3. A total of five Hi-Dense restorations failed burs (Komet, Lemgo, Germany), and finished using 25 µm finish- due to total retention loss. Two fillings had to be replaced due to per- ing diamonds. The minimum depth was 1–1.5 mm with the sisting hypersensitivity, both within the same patient. One filling occluso-axial angles being rounded. The dentine was not pre- was lost because of an unsuccessful endodontic treatment. Four fill- treated, the cavity was rinsed with air-water spray and thoroughly ings remained intact until natural exfoliation having occured dur- dried. Dentine close to the pulp was then covered with a calcium ing the two-year observation period. hydroxide cement (Calxyl, OCO-Praeparate GmBH, Dirmstein, Figure 1 shows the cumulative survival curves (Kaplan–Meier) Germany). for Class I and Class II Hi-Dense restorations. After two years, the In Class II cavities a matrix band was placed and interproximally survival rate for Class I fillings was 92% (standard error 7.4%) and wedged. The material was mixed for 15s in a CapMix mixing for Class II restorations a survival rate of 66% (standard error 10%) machine (Espe, Seefeld, Germany) and immediately applied into was computed. the cavity. After modelling of the anatomical surface, the restoration Over the whole clinically observed period of 2.5 years, however, the was coated with a thin layer of Hi-Glaze (Shofu) being light cured Kaplan–Meier algorithm displays a mean survival time for Class I fill- for 20 s using an Elipar (Espe) light curing unit. The matrix band ings of 2.4 years with a 95% confidence interval between 2.2 and 2.7 was removed after 3 min. Ten minutes after placement material years. For Class II restorations the mean survival time was 2.4 years as excess was removed with a finishing diamond bur (Intensiv). The well with a 95% confidence interval between 1.9 and 2.9 years. surface of the restoration was finally finished with a silicone polisher Regarding the success rate, the test statistic for equality of the (Shofu). survival distribution resulted in no statistical differences between At initial recall (baseline), after one, and two years, the restora- Class I and Class II restorations within the observation time (log tions were assessed according to modified United States Public Health Service (USPHS) codes and criteria28 (Table 2) by one inves- tigator using mirrors, probes, and intraoral photographs. Statistical analysis was computed using SPSS for Windows 8.0. 100 The descriptive statistics and survival analysis were calculated using the algorithm according to Kaplan-Meier and the log rank test, sta- 80 tistical unit was one restoration. Criteria for failure of a restoration are bulk fracture, retention loss, and secondary caries. The fillings 60

Table 3 Results of the clincial evaluation 40

Investigation 1 year 2 years Survival rate (%) (assessed cases) H1(n=42) H2(n=22) 20 Date of investgation (S.D.) 1.10 years (0.25) 2.05 years (0.29) Surface A: 0 B: 100 A: 5 B: 95 anatomic form A: 47 B: 50 C: 3 A: 10 B: 90 C: 0 marginal integrity A: 42 B: 53 C: 5 D: 0 A: 20 B: 60 C: 10 D: 0 0 1 2 3 integrity tooth A: 94 B: 6 C: 0 D: 0 A: 72 B: 14 C: 14 D: 0 Years integrity filling A: 36 B: 49 C: 5 D: 10 A: 29 B: 61 C: 10 D: 0 complaints A: 100 B: 0 C: 0 D: 0 A: 95 B: 0 C: 0 D: 5 Fig. 1 Kaplan-Meier algorithm for Class I and Class II restorations. Red line, class 1 (n=19); yellow dot, censored (n=17) orange line, class II (n=23); % A, alfa; B, bravo; C, charlie; D,delta red dot, class II censored (n=23)

318 BRITISH DENTAL JOURNAL, VOLUME 190, NO. 6, MARCH 24 2001 RESEARCH dental materials

Fig. 2 Characteristic Hi-Dense restoration a, Clinically evident crack formation b, Restoration of Figure 2a under the SEM c, occlusal: negative step formation due to wear d, proximal: marginal disintegration and loss of adhesion e, Bulk fracture after 15 months.

c

a

d

b

e

rank test, p=0.0614). shows the replica of the same filling viewed under a SEM. Marginal After 24 months, wear within contact free areas was generally SEM analysis revealed a distinct adhesive failure between tooth sub- acceptable. No restoration exhibited dentine exposure due to con- strates and restorative material above all in marginal areas of proxi- tact free wear. In the area of occlusal contacts, however, surface mal boxes (Fig. 2c). In contrast to the proximal box, the marginal fatigue and crack formation were obvious. integrity of occlusal parts remained intact over the observation Figure 2 displays a representative sample of a Class II Hi-Dense period of two years of clinical service, although distinct wear phe- restoration. Fig. 2a shows a clinically visible surface crack. Fig. 2b nomena are evident (Fig. 2d). The clinical view of Fig. 2e shows the

BRITISH DENTAL JOURNAL, VOLUME 190, NO. 6, MARCH 24 2001 319 RESEARCH dental materials failed restoration due to bulk fracture and consequently retention the matrix technique. Reports in the literature are inconsistent con- loss of the proximal part of the filling. cerning the conditioning of cavities prior to the application of glass ionomers. While laboratory tests commonly result in advantages for Discussion a separate conditioning step,30 clinical reports do not always sup- Compared with other clinical studies in the field of restorative pedi- port this hypothesis.31 In the course of the present in vivo study it atric dentistry, the restorations of the present study were placed in has to be considered that the preparations have been performed an early stage of the deciduous dentition at the age of 3–11 years with undercuts providing macromechanical retention for the expecting a quite long period of clinical service until exfoliation. restorations and not relying on adhesion of the materials to the This simulates realistic clinical conditions in pediatric dentistry. tooth hard tissues. It is obvious that the highest in vitro bond The relatively low age of the patients furthermore guarantees the strengths of glass ionomers may be not sufficient to allow non- presence of limited childrens‘ compliance during the placement of macromechanical retentive preparations in stress-bearing areas.12 most restorations.1 To enable an easy removal of the matrix band after the polymeriza- The aim of the present study was to evaluate whether the more tion of the varnish, the matrix band could be coated with varnish as recently developed viscous, metal-reinforced GIC Hi-Dense would well. We did not coat the matrix band, neither with varnish nor with be able to fulfil the rather high expectations regarding improved oily substances like vaseline because varnish may copolymerize with material properties such as wear resistance and flexural strength.18 the covering glaze and oily substances may affect the setting of the It is obvious that the present study provides no control group rep- material itself. Due to the undermining preparations, the careful resenting another material. Therefore previous studies providing removal of the matrix band obviously did not interfere with the similar observation periods in the literature of pediatric dentistry clinical results observed. dealing with the older cermet cement material Ketac Silver should Isolating the results of the 24 months recall, the survival rate of serve as comparison due to the similar chemical characteristics Class I restorations (92%) seems to be superior compared with (Table 1).1,12,22,28 Class II fillings (66%). Over the complete observation period dis- Ketac Silver and Hi-Dense imply different principles of incorpo- played in the Kaplan–Meier curves, however, the lower number of rating metal particles into a glass-ionomer cement matrix.17,18 cases with occlusal cavities combined with occuring natural exfolia- Consecutively, Hi-Dense is characterized by a higher viscosity than tions complicates the statistical evaluation, finally resulting in a p- Ketac Silver being achieved by a higher powder-to-liquid ratio and a value of slightly more than 0.05 when the log rank test was smaller particle size in the case of Hi-Dense. In contrast to the com- performed. Nevertheless, the data of the present study demonstrate positional differences, the flexural fatigue characteristics of Ketac that the tested material is suitable as occlusal restorative material for Silver and Hi Dense are not markedly different.14 This laboratory the primary dentition. fact is clearly reflected in the clinical results of both materials. The Altogether the more recent material Hi-Dense does not seem to survival rate of Hi-Dense compares well with the results of previous represent an overall amalgam substitute for class II restorations in studies with the metal-reinforced glass-ionomer cement Ketac Sil- deciduous molars, neither fracture resistance nor adhesive perfor- ver (Table 1).12,23,29 mance have been promising. A comparison of failed restorations revealed that Ketac Silver Without any doubt, resin composites, polyacid-modified com- more frequently suffered bulk fractures than did Hi-Dense.1 How- posites (compomers) and amalgam remain the most promising ever, Hi-Dense primarily failed due to total loss of the restoration. materials for the primary dentition.6,7 However, viscous GICs profit This observation can be interpreted as the effect of the different vis- from their easy handling characteristics which remain an important cosities of both materials, therefore Ketac Silver, due to its lower vis- advantage in less cooperative children. cosity, may adapt better to the cavity walls than Hi Dense.12 The reported persisting hypersensitivity with Hi-Dense restorations Conclusion support this interpretation assuming that the complaints were • Compared to conventional glass ionomers, the tested material caused by an improper seal of the margin. exhibited no improved clinical performance and lifetime Bulk fractures have been primarily observed with restorations expectancy for Class II restorations in deciduous teeth. exhibiting surface cracks. This high incidence is distinctly sup- • Primarily in uncooperative children this group of restorative ported by SEM analysis and may be attributed to the inferior adhe- materials exhibits advantageous handling properties. sive potential and to the low flexural strength of the material, 1 Hickel R, Voss A. A comparison of glass cermet cement and amalgam because the fracture risk consecutively increases after debonding. restorations in primary molars. J Dent Child 1990; 57: 184-188. The ion-exchanging adhesion to tooth substrates without the 2 Krämer N, Pelka M, Kautetzky P, Sindel J, Petschelt A. Wear resistance of rather complicated use of a dentin bonding agent is one of the argu- compomers and viscous glass ionomer cements (Abstract in English). ments in favour of glass-ionomer cements.8,20 Compared to the Dtsch Zahnärztl Z 1996; 52: 186-189. 3 Frankenberger R, Sindel J, Krämer N. Viscous glass ionomer cements - adhesion of resin-bonded composites, however, adhesion to enamel an alternative to amalgam in the primary dentition? Quintessence Int (10 MPa) and dentin (5 MPa, data of manufacturer) is significantly 1997; 28: 667-676. inferior. On the other hand, with glass ionomers there is no need for 4 Bundesgesundheitsamt: Amalgame - Nebenwirkungen und Bewertung compensating shrinkage stresses during setting as occurs with resin der Toxizität (Amalgams: Side effects and toxicity). Zahnärztl Mitt 1992; composites.6 82/19: 36- 5 Qvist V, Laurberg L, Poulsen A, Teglers P T. Longevity and cariostatic An important point in discussing the present results are the effects of everyday conventional glass-ionomer and amalgam restorations apparently different viscosities of the test material Hi-Dense and the in primary teeth: three-year results. J Dent Res 1997; 76: 1387-1396. material Ketac Silver. It is an obvious assumption that the higher 6 Attin T, Opatowski A, Meyer C, Zingg-Meyer B, Hellwig E. Clinical eval- viscosity of Hi-Dense in comparison to Ketac Silver possibly uation of a hybrid composite and a polyacid-modified composite resin in Class-II restorations in deciduous molars. Clin Oral Investig 1998; 2: affected the internal cavity wall adaptation. SEM findings primarily 115-119. in proximal areas tend to confirm this hypothesis because in these 7 Marks L A, Weerheijm K L, van Amerongen W E, Groen H J, Martens L particular sections of the margins loss of retention was probably vis- C. Dyract versus Tytin Class II restorations in primary molars: 36 ible (Fig. 2c). These observations of marginal irregularities like months evaluation.Caries Res 1999; 33: 387-392. washouts were clinically always followed by complete loss of the 8 Mount G J. Glass-ionomers: A review of their current status. Oper Dent 1999; 124: 115-124. proximal parts of the restorations. Important points for the discus- 9 Naasan M A, Watson T F. Conventional glass ionomers as posterior sion of these observations are the pre-treatment of the cavities and restorations. Am J Dent 11, 36 (1998).

320 BRITISH DENTAL JOURNAL, VOLUME 190, NO. 6, MARCH 24 2001 RESEARCH dental materials

10 Walls A W G, Murray J J, McCabe J F. The use of glass polyalkenoate 21 Forsten L, Karjalainen S. Glass ionomers in proximal cavities of primary (ionomer) cements in the deciduous dentition. Br Dent J 1988; 165: 13-17. molars. Scand J Dent Res 1990; 98:70-73. 11 Wilson, A D, Crisp S, Prosser H J, Lewis B G, Merson S A. 22 Östlund J, Moller K, Koch G. Amalgam, composite resin and glass Aluminosilicate glasses for polyelectrolyte cements. Int Eng Chem ionomer cement in Class II restorations in primary molars - a three year Product R & D 1980; 19: 263–270. clinical evaluation. Swed Dent J 1992; 16: 81-86. 12 Espelid I, Tveit A B, Tornes K H, Alvheim H. Clinical behaviour of glass 23 Holst A. A 3-year clinical evaluation of Ketac Silver restorations in pri- ionomer restorations in primary teeth. J Dent 1999; 27: 437-442. mary molars. Swed Dent J 1996; 20: 209-214. 13 Guggenberger R, May R, Stefan K-P. New trends in glass-ionomer chem- 24 Forsten L. Fluoride release and uptake by glass-ionomers and related istry. Biomaterials 1998; 19: 479-483. materials and its clinical effect. Biomaterials 1998; 19: 503-508. 14 Frankenberger R, Krämer N, Graf A, Petschelt A. Cyclic fatigue of glass 25 Pereira P N, Inokoshi S, Tagami J. In vitro secondary caries inhibition ionomers and compomers (Abstract in English). Dtsch Zahnärztl Z 1999; around fluoride releasing materials. J Dent 1998; 26: 505-510. 54: 269-271. 26 Mallow P, Durward M, Klaipo M. Comparison of two glass ionomer 15 Nicholson J W, Croll T P. Glass-ionomer cements in restorative dentistry. cements using the ART technique. J Dent Res 1995; 74: 405. Quintessence Int 1997; 28: 705-714. 27 Frencken J E, Holmgren C J. How effective is ART in the management of 16 Mount G J. Clinical performance of glass-ionomers. Biomaterials 1998; dental caries? Community Dent Oral Epidemiol 1999; 27: 423-430. 19: 573-579. 28 Ryge G, Cvar J F. Criteria for the clinical evaluation of dental restorative 17 Walls A W G, Adamson J, McCabe J F, Murray J J. The properties of a materials. San Francisco: United States Dental Health Center, US glass polyalkeonate (ionomer) cement incorporating sintered metallic Government Printing Office, 1971, publication no. 7902244. particles. Dent Mater 1987; 3: 113-116. 29 Kilpatrik N M. Durability of restorations in primary molars. J Dent 18 Williams J A, Billington R W, Pearson G J. The comparative strengths of 1993; 21: 67-73. commercial glass ionomer cements with and without metal additions. Br 30 Peutzfeldt A. Compomers and glass ionomers: bond strength to dentin Dent J 1992; 172: 279-282. and mechanical properties. Am J Dent 1996; 9: 259-263. 19 McKinney I E, Antonucci J M, Rupp N W. Wear and microhardness of a 31 Van Dijken J W. Four-year evaluation of the effect of 10% polyacrylic silver-sintered glass-ionomer cement. J Dent Res 1988; 67: 831-835. acid or water rinsing pretreatment on retention of glass polyalkenoate 20 Kilpatrick N M, Murray J J, McCabe J F. The use of a reinforced cement. Eur J Oral Sci 1996; 104:64-66. glass-ionomer cermet for the restoration of primary molars: a clinical trial. Br Dent J 1995; 179: 175-179.

BRITISH DENTAL JOURNAL, VOLUME 190, NO. 6, MARCH 24 2001 321