Quality and durability of marginal adaptation in bonded composite restorations F. Lulz I. Krejci F. Barbakow Department of Preventive Dentistry, Periodontology and Cariology Z0rich University Dental Institute Plattenstrasse 11 CH-8028 Z0rich Switzerland Received March 13, 1990 Accepted December 25, 1990 Dent Mater 7:107-113, April, 1991 Abslract- Excellent marginal adaptation extends the longevity of restorations. Unfortunately, polymerization shrinkage of composite restorations adversely affects this quality requirement. The residual stress within the cured resin compromises the material's properties, causes marginal openings, and flexes cavity walls. In this study, the wall-towall contraction in MOD cavities was measured for different placement techniques. In addition, the restoration margins were quantitated before and after thermo-cycling and mechanical stressing. Factors which enhanced adaptation also optimized marginal quality and reduced the amount of residual stress. The latter was expressed by intercuspal narrowing after the restoration was completed. Both quality and stress resistance of the marginal adaptation were inversely correlated to the intercuspal narrowing caused by the polymerization contraction of bonded and excellently adapted resin restorations. The most effective factors which optimized marginal quality included: guidance of the shrinkage vectors; reducing the ratio of bonded to free, unbonded restoration surfaces; and minimizing the mass of in situ-cured composite. The latter principle was followed best in the adhesive inlay technique. In medium-sized adhesive MOD composite inlays, the volume loss induced by the polymerization contraction of the composite cement was nondestructively compensated for by an inward flexing of each cavity wall of approximately 10 ~m.

sthetics and longevity of composite restorations are influenced by the quality and durability of the marginal adaptation. The factors governing the quality of marginal adaptation of bonded composite restorations are listed in Table 1. Among these factors, polymerization contraction ranging from 1.7 to 5.7% (Bausch et al., 1982; Goldman, 1983; Feilzer et al., 1988; Hay and Shortall, 1988; Walls et al., 1988) is the most adverse property of the c u r r e n t l y available resin-based restorative materials (Bausch et al., 1982; Goldman, 1983; Feilzer et al., 1989). The durability of marginal adaptation is negatively influenced by three factors: (1) Residual internal stresses, generated by the polymerization shrinkage, challenge the adhesive bond to the cavity walls and margins unless they are relieved by structural changes within the resinbased restoration or the adjacent enamel and dentin; (2) chemical degradation debonds the tooth-restoration interface; and (3) the differing physical properties of the dental hard tissues and the bonded materials have the potential to become destructive d u r i n g mechanical and t h e r m a l stressing. Quality and stress resistance of marginal adaptation have previously been analyzed and optimized in vitro by an SEM-based q u a n t i t a t i v e method of assessing marginal micromorphology as well as by a dye penetration method (Lutz et al., 1985). However, these evaluation techniques failed to explain the stress resistance or longevity of the excellent marginal adaptation of composite inlays and composite restorations placed by the three-sited light-curing technique (Krejci et al., 1987; Ffillemann and Lutz, 1988). Therefore, the purpose of this in v i t r o s t u d y was to i n v e s t i g a t e whether the magnitude of the unre-

E

lieved residual internal stress in bonded resin restorations was directly correlated to the durability of the marginal adaptation. This correlation was evaluated in two stages: first, by analyzing the tooth deformation induced by polymerization shrinkage, and second, by evaluating the changes in marginal quality produced by mechanical and thermal stresses. MATERIALS AND METHODS

Operative techniques and restoration placement.- Thirty-six caries-free extracted lower human molars were selected which measured 11.8 _+ 0.4 mm mesio-distaUy and 10.6 -- 0.5 ram bucco-lingually. They were divided into six groups, and the six operative techniques used in this study, with their pertinent adaptation-enhancing factors, are listed in Table 2. In 30 teeth, randomly assigned to groups 1 to 5, box-shaped mesio-occluso-distal cavities were cut by diamond preparation and finishing burs. The cavity walls were prepared exactly parallel to each other and the 90° cavosurface angles were in enamel (Reller et al., 1989). The cavities measured approximately 4.0 mm from vertical line angle to line angle, and they were 3.0 mm wide occlusally. The occlusal boxes were 3.0 mm deep, measured from the deepest point of the secondary fissures, while the cervical proximal margins were placed exactly 1 mm coronal to the cemento-enamel junction. Cavities were accepted only if their sizes did not deviate by more than 0.05 ram. Teeth with enamel fractures detected under the light microscope at 50 × magnification were discarded. The cavities in group 6 were identical in size and quality to those in the other groups, but their cavity walls diverged 4°. After the measuring device was mounted and calibrated, the restorations were placed strictly acDental Materials/April 1991 107

TABLE

1

A SUMMARY OF THE DIFFERENTFACTORS (CATEGORIZEDAS MATERIAL PROPERTIES, CAVITY PREPARATION, AND OPERATIVETECHNIQUE) WHICH INFLUENCETHE QUALITY OF MARGINAL ADAPTATION OF RESIN-BASED RESTORATIONS Material Properties Cavity Preparation Composite Resin Cavity Size Polymerization Shrinkage Cavity Design Resin Composition Cavosurface Configuration Filler Content Finishing Technique - Pre-gelation Flow Incorporated Air Yield Strength of Cavity Walls** - Curing Mode & Velocity - Curing Degree Yield Strength Coefficient of Thermal Expansion Water Sorption Bonding Agent Wetting Properties Polymerization Shrinkage - Pre-gelation Flow Yield Strength / Elasticity * Ratio of bonded to free, unbonded, restoration surfaces. ** Invalid variables.

cording to the clinical protocols at a standardized room temperature of 22°C and 55 _+ 5% humidity. The nondestructive finishing was delayed for 40 min until the measurement of the tooth deformation was completed (Lutz et al., 1983). The teeth were then stored in water for 14 days before they were exposed to the mechanical and thermal stresses.

Measurement of tooth deformation. - The measuring device was individually mounted on each tooth by metal pins placed buccally and lingually in the measuring points. These points were defined as the intersection of the equator line and the bucco-lingual median line. A miniaturized infrared light barrier (Figs. 1 and 2) was attached to the lingual pin and con-

TABLE

Operative Technique In situ-cured Composite Mass Built-up Base - Inlay Technique C-Factor* - Adhesion to Base Material - Adhesion to Dentin - Adhesion to Enamel Bonding Agent Application Insertion Technique Curing Mode & Velocity Curing Technique Finishing Time & Technique

Curing Intensity (Incomplete Cure)**

nected to an xy-recorder. The temperature and light output were kept constant by an additional control sensor. A slice of photographic film, measuring 4 x 10 mm and impermeable to infrared radiation, was attached to the buccal pin and inserted in the slot of the infrared light barrier. The zero position was adjusted to the center _ 20 txm of the mea-

2

APPLIED OPERATIVETECHNIQUESAND PERTINENT FACTORS (SEE TABLE 1) ENHANCING MARGINAL ADAPTATION Group

OperativeConcepts Factors Enhancing Marginal Adaptation Enamel Etch Technique; Non Functional Bonding Agent*; Chemically Cured Adhesion to Enamel, C-Factor, Bonding Agent Application, Composite**; Bulk Placement; Finishing Diamond Burs & Flexible Discs Curing Mode, Finishing Technique*** EET; NFBA'; Light Cured Composite"; BP, Curing from an Occlusal Direction, AtE, CF, BAA, FT*** 60 seconds' Irradiation; FDB&FD EET; NFBA'; LCC"; Two-step Incremental Technique: First Increments -- ProxAtE, CF, BAA, Insertion Technique, FT*** imal Boxes to the Bottom of the Occlusal Box, Second Increments = Unfilled Parts of the Proximal Boxes and Occlusal Box, Curing from an Occiusal Direction, 60 seconds' Irradiation per Increment; FDB&FD EET; NFBA'; LCC"; Three-Sited Incremental Technique: First Increments = AtE, CF, BAA, optimized IT, Curing Technique, FT*** Apical Third of the Proximal Boxes, Curing from a Gingivoproximal Direction via Light Reflecting Wedges, Second Increments = Buccal Two-thirds of the Proximal Boxes, Curing from a Buccal Direction through the Buccal Wall, Third Increments = Lingual Third of the Proximal Boxes, Curing from a Lingual Direction through the Lingual Wall, Fourth Increment = Occlusal Box, Curing from an Occlusal Direction; 60 Seconds' Irradiation per Increment; FDB&FD EET; NFBA'; LOG"; Glass-ionomer Cement: Unetched Built-up Base; TSIT; In situ-cured Composite Mass, AtE, CF, BAA, optimized FDB&FD IT, CT, FT*** EET; NFBA'"; Light & Chemically Cured Resin-based CementA, Partial Curing Minimized ISCCM, AtE, OF, BAA, CM, IT, CT, FF*** from a Gingivoproximal Direction via Light-reflecting Wedges; Extra-orally Heatcured Composite InlayAA, FDB&FD * Silar, Bonding Agent, 3M, St Paul, MN, USA ' Silux, Bonding Agent, 3M ^ Duo-Bond, Colt~ne ** Silar, 3M " Silux, 3M ^^ DI 500, Colt~,ne *** Composhape Set, Intensiv, Lugano, CH; Soflex-Discs, 3M '" DI 500, Bonding Agent, Colt~ne, CH

108 LUTZ et al./MARGINAL ADAPTATION OF COMPOSITES

xy-Recorder

TABLE3 INCREASE OF THE INTERCUSPALDISTANCE [p,m] IN THE PREPAREDTEETH OF GROUPS 1 TO 6 UNDER A SPREADINGFORCEOF 2 N

Group 1 (n=6) 2 3 4 5 6 1-6 (n = 36)

Mean 0.717 0.617 0.700 0.633 0.617 0.583 0.644

SD 0.183 0.264 0.253 0.250 0.075 0.330 0.190

suring range. The measuring plane was placed 3 mm above the equator line of the tooth to facilitate the placement of the restorations. When the cavity walls were deformed in the bucco-lingual direction, the film was slightly displaced in the slot, and the intensity of the infrared light changed. This change was detected by the photodiode-receiver. The conversion of the photocurrent into voltage was linear in a range of _+ 150 ~m with a 1% potential error. The initial yield strength and the restoring capacity of each tooth were assessed by a standardized application of a 2-N spreading force for 60 s in the middle of the occlusal box. The intercuspal distance was graphed continuously during r e s t o r a t i o n p l a c e m e n t and for 40 rain a f t e r placement. After the restorations were finished, a second measure-

Maximum 1.0 1.0 1.0 1.1 0.7 0.8 0.7

urrent to °Jtageon

Minimum 0.5 0.3 0.4 0.4 0.5 0.4 0.3

Receiver

AtD I

ment was made before and after exposure to an occlusal load of 90 N.

I

=-

Control Sensor

T_ransmittir

Assessment of marginal adaptation.Three replicas were made of the occlusal and proximal surfaces of the restorations: first, immediately after the finishing procedure was completed; second, after the restorations were stressed with an occlusal load of 90 N; and third, after the mechanical and thermo-cycle stressing in a chewing machine. This simulated 120,000 masticatory cycles with a maximum force of 49 N and 300 full temperature cycles between 5o-55°. 5°C (Krejci et al., 1990). At these three junctures, the marginal quality of each group was quantitated by assessment of the percentage of 'excellent margin' and 'marginal opening' in the SEM at 200 x magnification after the replicas had been

~ Cusp Reference Current

\ I Fig. 2. Schematic description of the miniaturized infrared light barrier.

sputtered with gold (Lutz et al., 1985). The data for marginal adaptation and tooth deformation were statistically analyzed by the T-method. The

CONTRACTION OF CAVITY WALLS M a D C o m p o s i t e Restoration Bulk C h e m i c a l l y C u r e d 0

5

10

I

I

I

15

20

Registration Time [rain] 25 30

-10 O O c a t~

-20te

£ o c

-30.

Fig. 1. Miniaturized infrared barrier in situ attached to a test molar.

[pm] F/gs. 3 to 8. Examples of the contraction curves during polymerizationfor each experimental group.

Dental Materials/April 1991 109

CONTRACTION OF CAVITYWALLS

MOD Composite Restoratlon Bulk Light Cured Registration Time [min] 0

5

10

15

25

3O

I

I

I

I

I

i

I*

O-

Irradiation Exposure

-I0" ® 0

-20"

! -30"

[tim]' Fig. 4. CONTRACTION

OF CAVITY

WALLS

M O D C o m p o s i t e Restoration 2 - S t e p Light C u r e d RegistrationTime [min] 0

5

10

15

20

25

I

I

I

I

i

I

3O

OIrradiation Exposure

-10 " e o :~::~::::: a

~

-20 -

-30 -

[pm], Fig. 5.

t test was used for paired comparisons between the marginal qualities before and after exposure to mechanical and thermo-cycle stress within each group.

RESULTS Table 3 shows the increased intercuspal distance in the prepared teeth of all groups induced by a spreading force of 2 N. Figs. 3 to 8 show examples of contraction curves for each

experimental group, characterizing the interaction between the intercuspal distance and the rigid contraction induced by the different operative techniques. Table 4 lists the quantitated results of the marginal adaptation of the restorations before and after application of the masticatory forces. The data quantitate the induced loss of quality and the narrowing of the intercuspal distance caused by the rigid contraction. The

110 LUTZ et al./MARGINAL ADAPTATION OF COMPOSITES

application of a single central occlusal load of 90 N on the freshly finished restorations did not cause any detectable dimensional changes.

DISCUSSION The teeth selected and the cavity preparations were adequately standardized. No differences in either yield strength or elastic after-effect were recorded when the prepared teeth were subjected to a spreading force of 2 N applied for 60 s (Table 3). Furthermore, in molars with MOD cavities prepared as described above, the relation between tooth deformation and spreading forces up to 40 N was linear. The hysteresis one min after this stress-exposure was in the order of 1 ~m. Accuracy and linearity of the measuring technique were also sufficiently well-standardized. As expected, the inlay cavities deformed the least. Nonetheless, the readings represent only relative values. On the one hand, they were approximately 20% too high, because the measuring plane was positioned 3 mm above the cavity. On the other hand, they were too low because of two reasons, i.e., the working temperature of 22°C and the timing. Cuspal flexures days after restoration p l a c e m e n t - w h i c h supposedly indicate structural changes within the t o o t h - h a v e already been reported (Causton et al., 1985). This is explained by the fact that ambient temperatures of 37°C increase the polymerization shrinkage by 20-34% (Bausch et al., 1982). Thirty min after the start of cure, Silar and Silux attained only 87% and 71%, respectively, of their 24-hour shrinkage values (Feilzer et al., 1988). In this study, the readings were stopped 40 min after the start of cure, because all the contraction curves leveled out within this time interval. The contraction curves mirrored the operative techniques used, with chemical cure causing a smooth continuous decrease in intercuspal distance. In contrast, light cure was characterized by a sudden decrease followed by a rebound at the end of the irradiation time. This phenomenon is explained by an interference between the curing light and the light barrier. The initial marginal adaptation in groups 1 and 2 was so poor that it

exerted no bending forces on the cavity walls. Consequently, the loss of marginal adaptation under stressing was minimal, because residual stresses were already partially relieved by the gap formation at the tooth/restoration interface. The percentage of 'excellent margin' at different sites along the tooth-restoration interface in group 1 before stressing confirmed the vector theory in polymerization shrinkage. The marginal qualities were uniform at the different sites in group 1, because, in chemically cured composites, the vectors uniformly run to the center of mass. However, in lightcuring composites, the shrinkage vectors point toward the light source, thus inducing material displacement away from the cervical margins. Therefore, the marginal quality in group 2 was worst along the cervical line angle (Table 4). The incremental technique (group 3) produced significantly better marginal qualities while simultaneously inducing wall deformations of roughly 2 × 20 ~m. The crippling loss of marginal qualities during stressing mirrors the residual stress within both the composite restoration and the deformed cavity walls. The results in groups 4, 5, and 6 confirm the potential of the marginal-adaptation-enhancing factors listed in Table 2. Their effect is twofold: When applied singularly, they improve the initial marginal adaptation; and in sophisticated combinations, they also decrease the internal stress in perfectly adapted restorations. One can assume that the intercuspal narrowing reflects the magnitude of the residual stress induced by the rigid contraction. Thus, it can be hypothesized that the residual stress and the loss in marginal restoration quality under stress are well-correlated provided that sufficient excellent margin is established initially (groups 3-6). The results are in accordance with previously published reports (Jensen and Chan, 1985; McCullok and Smith, 1986; Smith and Caughman, 1988). Unfortunately, direct comparisons cannot be made because of differences between the present and earlier studies regarding measuring technique, cavity preparation, and operative procedures. Operative techniques requiring less than 20 ~m

likely to cause structural changes within the tooth. Most probably, the stress is relieved at the enamel/dentin junction (Causton et al., 1985). In thin resin-based cement layers, the C-factor is extremely large (Feilzer et al., 1987, 1989). The rigid contraction therefore becomes univectorial, runs perpendicular to the bonded interface, and equals the polymerization shrinkage in lightcured systems (Itoh et al., 1986; Feilzer et al., 1987, 1989; Kato et al., 1988). Consequently, in inlays bond-

cuspal flexures to compensate for the rigid contraction provide stress-resistant marginal adaptation. In molars with medium-sized MOD cavities, the cavity walls either yield without causing stress or undergo plastic deformation up to 10 ~m (Causton et al., 1985; Jensen and Chan, 1985). The restoring capacity is almost 100% when contracting forces are applied for only a short period of time. Therefore, bonded resin restorations which permanently flex the adhering cavity walls are

CONTRACTION OF CAVITY WALLS MOD Composite Restoration 3-Sited

Light C u r e d

0

5

10

15

20

25

I

I

I

I

I

I

RegistrationTime [min] 30

0-

.:.. :.: ::.::..: ...

Irradiation Exposure +:+:.:.:,:

.:,::+:,:.

-10 O O ¢O

-20 2

-30

[~m]

Fio. 6. CONTRACTION OF CAVITY WALLS MOD Composite

Restoration

GIC-Base, 3-Sited Light C u r e d Registration Time [min] 0

5

I

I

10 |

ii!iii+i 15

20

I

I

":::2:::::: ............ :::::::::::::

::::::::::::: ::2:::::::: ::::::::::::; ::+:.:.:.: ...+....... :.:.:.:,:.:. :::::::::::::

-10 0 O C

............, :::::::::::::

~

-20

2@

...... 2:::::::::

-30 •

::::::::::::

.:.:.:+:.:.:

::::::::::::: :::::::::::::

::::::::::::: ::::::2:::: ?~!i!i!!!ii!!

2::::::::::

:::::::::::::

...........

++i+i++ii++++i++ii+i+++

30 z

b

~:iiiiii~iiil

i~i~i~iiiiiiiii!iiiiiiiiii! +:,:.:.:+:+:.:.:,:.: iiiiiiiiiiili IrradiationExposure :::::::::::: ::::::::::::

iiiiiiiiiiii: .:.:,:.:.:+: :1:1:1:131: :::::::::::::

..,.,........

:::::::::::::

25 i

::::::::::::: ::::::::::::: :2::::::::: ............

:::::::::::::

::2:::::::: iii~!i!~!!!i!

iiiiiii!iiiii ...........+ ::::::::::::;

............. iiiiii!iiiiii

[.m],

i+++++i+++++++ +i++ +++++

+++++,

i i li i i l

iiiilili!

iiiiiiiiiiill

Fig. 7.

Dental Materials~April 1991

111

TABLE4 MARGINAL QUALITYOF RESTORATIONSIN GROUPS 1 TO 6 BEFOREAND AFTER STRESSINGWITH THE INDUCED QUALITY LOSS OF 'EXCELLENT MARGIN' [%] FOR THE CERVICAL, LATERAL, AND OCCLUSALMARGINS AND THE TOTALTOOTH-RESTORATIONINTERFACE;THE DECREASEOF THE INTERCUSPALDISTANCE [~m] IN THE RESTORATIONSOF GROUPS 1 TO 6 40 MIN AFTER START OF RESTORATIONPLACEMENT Marginal Quality: Percentage of 'Excellent Margin' @, n = 6) Cervical Lateral Occlusal Total Before After Loss Before After Loss Before After Loss Before After Groups Stressing Stressing Stressing Stressing 1 40.5 20.6 19.9 34.3 25.3 09.0 38.5 31.8 06.7 37.8 25.9 2 27.1 16.3 10.4 48.4 38.8 09.6 56.2 40.6 15.6 43.9 31.9 3 66.0 28.4 37.6 63.7 30.2 33.5 71.5 39.4 32.1 67.1 32.7 4 79.9 72.9 07.0 84.6 71.7 12.9 82.6 69.8 12.8 82.4 71.5 5 87.7 85.5 02.2 89.3 81.1 08.2 88.8 83.7 05.1 86.6 83.5 6* 90.2 88.4 03.8 91.9 88.1 03.8 92.4 88.5 03.9 91.5 87.7 95% Confidence interval, percentageof 'excellent margin' before exposureto masticatory forces = 09.7. 95% Confidence interval, percentageof 'excellent margin' after exposure to masticatory forces = 10.2. *Mean cement layer thickness -- 65 p,m.

ed to enamel and dentin, the intercuspal narrowing follows the formula

the adhesive surface in enamel was about 40 mm 2 per wall. Consequently, at the composite enamel interface the tensile stress was in the order of 1 MPa. This is somewhat less than the low cohesive strength of etched enamel cut parallel to the long axis of the prisms. The 90° external line angle in enamel is therefore an acceptable cavosurface configuration for adhesive inlays or optimally placed composite restorations (groups 5 and 6). This fact is supported by the absence of marginal enamel fractures after stress exposure (Krejci et al., 1987; Ftillemann and Lutz, 1988). Reducing the in situ-cured composite mass was the most effective of the adaptation-enhancing factors.

Ain = {2 × Cth × Vol%} : 100 with Ai, representing intercuspal narrowing, Cth the thickness of the cement layer, and Vol% the polymerization shrinkage of the resinbased composite. As can be computed, cement layers up to 200 ~m wide can be tolerated if it is assumed that the polymerization shrinkage of the composite cement is 5 Vol%. As seen in the test for linearity between intercuspal n a r r o w i n g and flexing force, a 20-~m intercuspal narrowing corresponds to a force of approximately 40 N. In the test cavities, CONTRACTION

OF CAVITY

WALLS

MOD Composite Restoration Adhesive Inlay O

5

IO

15

20

25

I

I

I

I

I

I

30 i

~i~i~iiiiiiiiiii

b

IrradiationExposure -10

~ ! ! i i ! ~!~:!!!:i~:!:i~:!!!!!: ! ~:!:

,,-,,

o

-20 -

i i i iiiiiiiiiii i i i i

2e C

-30

i!iiiiiiiiil!ii}ii!iiii

[~m] Fig. 8.

112

11.9" 12.0 34.4*** 10.9" 03.1 03.8

IntercuspalNarrowing in #m @, n = 8) 14.6 20.1 37.6 30.6 21.9 11.6

This concept is readily and easily obtained in the inlay technique. In order for stress resistance of the excellent marginal adaptation to be secured, the cement layer must be < 200 ~m wide. Because of esthetic reasons and for fear of excessive wear, particularly caused by toothbrush bristles with diameters as small as 160-180 ~m, cement layers < 100 ~m wide are considered to be sufficiently protected by the enamel and inlay margins and therefore to be adequate. Apart from the excellent and stress-resistant marginal quality, composite inlays also have superior material properties. This is because the unrestricted curing at higher temperatures results in a less stress-loaded and optimally polymerized inlay. CONCLUSIONS

Registration Time [mini

O O c O

Loss

LUTZ et aL/MARGINAL ADAPTATION OF COMPOSITES

(1) In resin restorations with excellent marginal adaptation, stress resistance and durability of the interfacial bond are inversely proportional to the residual internal stress induced by the rigid contraction of the in situ-polymerized resin composite. (2) Although m a n y factors enhance marginal adaptation, reducing the amount of in situ-cured composite mass and reducing the ratio of bonded to free, unbonded, restoration surfaces are the key factors that help to reduce the residual stress in excellently adapted bonded resin restorations. REFERENCES BAUSCH, J.R.; DELANGE, K.;

DAVIDSON, C.L.; PETERS,A.; and DEGEE,A.J.

(1982): Clinical Significance of Polymerization Shrinkage of Composite Resins, J Prosthet Dent 48: 59-67. CAUSTON, B.E.; MILLER, B.; and SEFTON, J. (1985): The Deformation of Cusps by Bonded Posterior Composite Restorations: An In Vitro Study, Br Dent J 159: 397-400. FEILZER, A.J.; DEGEE, A.J.; and DAVmSON, C.L. (1987): Setting Stress in Composite Resin in Relation to Configuration of the Restoration, J Dent Res 66: 1636-1639. FEILZER, A.J.; DEGEE, A.J.; and DAVIDSON, C.L. (1988): Curing Contraction of Composites and Glass-Ionomer Cements, J Prosthet Dent 59: 297-300. FEILZER, A.J.; DEGEE, A.J.; and DAVIDSON, C.L. (1989): Increased Wall-to-Wall Curing Contraction in Thin Bonded Resin Layers, J Dent Res 68: 48-50. F~LLEMANN,J. and LUTZ,F. (1988): Direktes Kompositinlay, Schweiz Monatsschr Zahnmed 98: 759-764. GOLDMAN, M. (1983): Polymerization Shrinkage of Resin-based Restorative Materials, Aust Dent J 28: 156-161. HAY, J.N. and SHORTALL,A.C. (1988): Polymerization Contraction and Re-

action Kinetics of Three Chemically Activated Restorative Resins, J Dent 16: 172-176. ITOH, K.; YANAGAWA,T.; and WAKUMOTO,S. (1986): Effect of Composition and Curing Type of Composite on Adaptation to Dentin Cavity Walls, Dent Mater J 5: 260-262. JENSEN, M.E. and CHAN, D.C.N. (1985): Polymerization Shrinkage and Microleakage. In: Posterior Composite Resin Dental Restorative Materials. G. Vanherle and D.C. Smith, Eds., St. Paul: 3M, pp. 243-262. KATO, H.; ITOH, K.; and WAKUMOTO,S. (1988): The Bonding Efficiency of Chemically and Visible Light Cured Composite Systems, Dent Mat J 7: 1318. KREJCI, I.; REICH, F.; ALBERTONI,M.; and LUTZ, F. (1990): In vitro Testverfahren zur Evaluation dentaler Restaurationssysteme. 1. Computergesteuerter Kausimulator, Schweiz Monatsschr Zahnmed 100: 953-960. KREJCI, I.; SPARR, D.; and LUTZ, F. (1987): A Three-Sited Light Curing Technique for Conventional Class II Composite Resin Restorations, Quint Int 18: 125-131. LUTZ, F.; IMFELD, T.; BARBAKOW,F.;

and ISELIN,W. (1985): Optimizing the Marginal Adaptation of M0D Composite Restorations. In: Posterior Composite Resin Dental Restorative Materials. G. Vanherle and D.C. Smith, Eds., St. Paul: 3M, pp. 405419. Lvwz, F.; SETCOS, J.C.; and PHILLIPS, R.W. (1983): New Finishing Instruments for Composite Resins, J A m Dent Assoc 107: 575-580. MCCULLOK, A.J. and SMITH, B.G.N. (1986): In Vitro Studies of Cuspal Movement Produced by Adhesive Restorative Materials, Br Dent J 161: 405409. RELLER, U.; GEIGER, F.; and LUTZ, F. (1989): Quantitative Investigation of Different Finishing Methods in Conventional Cavity Preparations, Quint Int 20: 453-460. SMITH, C.D. and CAUGHMAN, W.F. (1988): The Effects of Composite Polymerization Shrinkage on Intercuspal Distance, J Dent Res 67: 221, Abstr. No. 864. WALLS, A.W.G.; MCCABE, J.F.; and MURRAY, J.J. (1988): The Polymerization Contraction of Visible-Light Activated Composite Resins, J Dent 16: 177-181.

Dental Materials~April 1991 113