Posterior composite Class II restorations: in vitro comparison o! preparation designs and restoration techniques K.J. Donly 1. T.W. Wild 1 M.E. Jensen2 1Department of Pediatric Dentistry University of Texas Dental Branch 6516 John Freeman Avenue Houston, Texas 77030 2Dows Institute for Dental Research University of Iowa College of Dentistry Iowa City, Iowa 52242 Received May 8, 1989 Accepted March 1, 1990 *Corresponding author Dent Mater 6:88-93, April, 1990

Abstract-The aim of this study was to evaluate conservative preparation designs for therestoration of Class II lesions with posterior resin composite. Fourteen primary and 14 permanent molars were obtained. Consenvative modified MO and DO preparations were placed in half the teeth; conventional MO and DO preparations were placed in the remaining teeth. Randomly, a glass-ionomer liner was placed over the exposed dentin in one preparation of each tooth; a calcium hydroxide liner was placed in the remaining preparations. Posterior resin composite was placed in all teeth, and the teeth were loaded with a 17-kg force. Teeth were thermocycled, stored in 37°C water, then immersed in 50% silver nitrate solution and placed in developer. The teeth were sectioned and photographed. Microleakage was calculated according to the depth of dye penetration, on a 6-degree scale. Results demonstrated the conservative modified restorations and conventional restorations, when glass-ionomer liner was used, to have less marginal microleakage, in both primary and permanent teeth, than their calcium hydroxide counterparts.

esin composite has progressed to being one of the most widely utilized contemporary restorative materials, over the past 20 years. Resin composite has become accepted due to its excellent esthetics, relatively low thermal conductivity, preselwation of tooth st~ctta'e in cavity preparation, and continued advancement in the stability of compositional properties of the material. More recently, resin composites have been recommended for use as posterior restorations. Several studies report success with Class II posterior composite restorations, as well as Class I posterior restorations, in prhnary teeth (Tonn and Ryge, 1985; Roberts et al., 1985; Oldenburg et al., 1985, 1987; Nelson et al., 1980). When the placement of a Class II posterior composite restoration is under consideration, the knowledge of an acceptable cavity preparation design is necessary. The specific cavity design for resin composite in primary molars is presently controversial. Nelson et al. (1980) reports success using a conventional Class II amalgam preparation. Oldenburg et al. (1985) completed a study comparing three different preparation de-

R

OCCLUSALBEVEL (0.5- 1.0ram) ~

~

signs for posterior Class II resin composite restorations. The first design was a conventional Class I] preparation as suggested by Nelson. The second design was a conventional Class II preparation with a 1mm, 45-degree cavosurface bevel. The thh'd design was a modified preparation in which enamel was removed only for access to decay, the cavosurface margin also being beveled. Findings demonstrated a failure rate of 21.4% for the modified/ bevel preparation design, 8.3% failure rate for the conventional preparation design, and 1.3% failure for the conventional/bevel preparation design after two years. Four-year results demonstrated failure rates of 34%, 15%, and 8.3% for the same restorations, respectively (Oldenburg et al., 1987). These findings were in agreement with a similar study performed by Paquette et al. (1983). Therefore, a conventional Class II preparation with a cavosurface bevel appeared to be the most widely accepted preparation for primary molars. Obviously, a reduction in the loss of tooth structure, during cavity preparation, would be pertinent to

EXTERNALBEVEL ONPROXIMALCAVOSURFACE MARGIN

OCCLUSAL BEVEL

-

ROUNDED AXIAL-PULPAL LINEANGLES

Fig. 1. Schematic diagram illustrating the conservative modified preparation.

88 DONLY et al./CONSERVATIVE CLASS H RESIN COMPOSITE RESTORATIONS

BEVEL

the concepts of operative dentistry. Therefore, possible reasons for the failure of the modified/bevel preparation design should be discussed. Paquette et al. (1983) stated that the failures may be associated with insufficient retention and resistance form in cavity preparation. This may indeed be true, since the conventional preparation design has resistance because of the converging walls. Another possible cause for failure suggested that the enamel thickness in primary teeth may not provide sufficient surface area for a resin layer thick enough to withstand the displacing forces of occlusion. This seems to be a very realistic problem. The studies of Paquette et a/. (1983) and Oldenburg et aL (1985) involved the placement of a calcium hydroxide base over all exposed dentin; therefore, all retention was obtained from the acid-etch enamel bond. Once the enamel-composite bond was fractured, worn, or polished away, the restoration would literally fall out due to the lack of further retention. Perhaps a dentin bonding agent would give added retention, prevent marginal leakage, and improve the success of the modified/bevel preparation design. Another point which must be considered is the possibility of hydration of the calcium hydroxide base. Water sorption of resin composite or inadequate marginal integrity may BUCCAL-LINGUAL INCREMENTAL POLYMERIZATION

FIRSTINCREMENT PLACEDAND PHOTO-CURED

expose the calcium hydroxide to moisture, leaving an unstable lining which could be compressed under occlusal forces. This, in turn, creates a condition which, under occlusal forces, may cause marginal breakdown more frequently than ff a dentin bonding liner had been utilized. Polymerization shrinkage could also have an effect on the success of the restoration. Previous studies have not placed the resin composite in increments (Oldenburg et at, 1985; Paquette e$ al., 1983). Incremental placement has been shown to minimize shrinkage stresses (Donly and Jensen, 1986; Hansen, 1986; Eick and Welch, 1986), eliminating the potential effect of polymerization shrinkage to cause inadequate marginal adaptation. A r e c e n t s t u d y by McConnell et al. (1986) demonstrated that composite polymerization shrinkage pulled a calcium hydroxide base away from the dentinal wall, leaving a gap. The purpose of this study was to evaluate, through in vitro comparison, different preparation designs and restorative techniques using posterior resin composite.

sio-occlusal and disto-occlusal surfaces. The preparation outline was cut with a #330 carbide bur, placing the occlusal margin extension 25% the tooth mesiodistal length, and the depth of 1.5 mm. A 0.5-mm, 45-degree bevel was placed on all enamel margins (Fig. 1). In each tooth, one preparation had a calcium hydroxide base (Dycal, L.D. Caulk Co., Milford, DE) placed over all exposed dentin; the other preparation had a glass-ionomer liner (Ketac Bond, ESPE-Premier Sales Corp., Norristown, PA) placed on all exposed dentinal surfaces. Before the placement of the glass-ionomer liner, a 40% polyacrylic acid (Durelon, ESPEPremier Sales Corp., Norristown, PA) was applied to the dentin for 10 s and thoroughly rinsed. Each tooth was acid-etched (Etching Gel, 3M Dental Products, St. Paul, MN) for 60 s, then thoroughly rinsed for 30 s. The glass-ionomer liner was etched for 10 s. An unfilled resin (Scotchbond, 3M Dental Products., St. Paul, MN) was placed and polymerized (P30, 3M Dental Products, St. Paul, MN), followed by restoration with posterior resin composite (Visilux, Visible Light Curing Unit, 3M Den-

MATERIALSAND METHODS Fourteen extracted or exfoliated primary molars and 14 permanent molars, with intact clinical crowns, were obtained. The teeth were stored in 0.1% thymol (Mallinckrodt, Inc., St. Louis, MO) solution until the experimental procedure was initiated, thereby preventing dehydration. The teeth were mounted in 2.5-cm retention tubes, being held fast by acrylic (Fastray, Harry J. Bosworth, Skokie, IL). Fourteen teeth, seven primary and seven permanent, had conservative modified preparations cut in the me-

IPRESSUREROD RESTORATIVE MATERIAL

LINGUAL

BUCCAL

MOUNTINGMEDIA FORTOOTH

~

Fig. 4. Schematic diagram of a mounted tooth with a pressure rod in place for application of an axial loading force.

OCCLUSAL BEVEL (0.5 - 1.0 mm)

~

L..~ .~.-~

..... 7

/.-

I/ ~ . . . : . i i ! l / ~ ~

SECONDINCREMENT PLACEDAND PHOTO-CURED

Fig. 2. Schematic diagram illustrating the buccolingual increment polymerizatlull.

1

;

OCCLUSAL BEVEL EXTERNALBEVEL ON PROXIMAL-

AVOSO.,ACE

II

//

ROONOEO

AXlAL-PULPAL/ LINE ANGLES(

ROUNDE~

AXIAL-PULPAL LINE ANGLES

//

"~

/7

I PROXI MA EXTERNA BEVEL

II

~UNDEI

GIBEVEL NGIVAL

INTERNAL LINE ANGLES

Fig. 3. Schematic diagram illustrating the conventional preparation.

Dental Materials/April 1990 89

FREQUENCY DISTRIBUTION OF M I C R O L E A K A G E WITHIN P R I M A R Y TEETH TREATED C O N S E R V A T I V E L Y 50"

LEGEND

4.5"

~JJ

Ca(OH)2

40"

3.5.

30"

2.5

20

I

15

~O

5

O 4 -

-

o

I DEGREE

2 3 OF PENETRATION

4

Fig. 5. Bar graph illustrating die penetration in primary leeth with conservative restorations.

FREQUENCY DISTRIBUTION OF M I C R O L E A K A G E WITHIN P R I M A R Y TEETH TREATED C O N V E N T I O N A L L Y z.O

LEGEND ~J~

Co(OH)2

3.5

30

25

2.0

~5

10

.5

O

i

0

1 DEGREE

OF

2 ,3 PENETRATION

4.

Fig. 6. Bar graph illustrating die penetration in primary teeth with convenlional restorations.

tal Products, St. Paul, MN). The preparations with the calcium hydroxide liner had the resin composite placed and polymerized in one complete unit. The preparations that were lined with glass ionomer had the resin composite placed in buccolingual increments (Fig. 2). Fourteen teeth, seven primary and seven permanent, had two conven-

tional Class II preparations placed in the mesio-occlusal and disto-occlusal surfaces. The cavity preparations had the cervical margin placed 1 mm above the CEJ with the axial wall extended 1.5 ram. The occlusal extension was placed approximately 40% the mesiodistal length at a 2mm pulpal depth. A 0.5-mm, 45-degree bevel was placed on all enamel

90 DONLY et al./CONSERVATIVE CLASS H R E S I N COMPOSITE RESTORATIONS

margins (Fig. 3). One preparation, randomly, had a calcium hydroxide base placed over all exposed dentin, while the second preparation received a glass-ionomer liner over all exposed dentin. Posterior resin composite was placed as one complete unit in the preparation with a calcium hydroxide base and in buccolingual increments in the preparation with glass-ionomer liner. All restorations were exposed to 17-kg axial loading periods of five s, alternating with load-free periods of the same length, over a total time period of two rain. The load was applied to the tooth by a sphere, attached to the upper member of the testing instrument, brought into contact with the buccal cusp, marginal ridge, and lingual cusp (Fig. 4). The use of this loading weight was determined by the fact that 17 kg has been cited as the average tooth load during mastication (Anderson, 1956; Fields et al., 1986). All teeth were then thermocycled from 10°C to 50°C, 40 cycles/day, for 30 days. The dwell times were two min during thermocycling, after which the teeth were stored in 37°C water when not being thermocycled. Thirty days was chosen because P30, the resin composite used in this study, has approached its maximum water sorption over this time period (Oysaed and Ruyter, 1986). Following this water storage period, the teeth were again axially loaded with 17 kg. After the teeth were loaded, fingernail polish was placed on all exposed tooth surfaces and mounting rings to within 1 mm of all restoration margins. The samples were immersed in 50% aqueous silver nitrate solution in darkness for four hours, then immersed in rapid photodeveloping solution under bright light for two hours. The specimens were sectioned, under water, mesiodistally in a plane perpendicular to the long axis of the tooth. Examinations of these sections (two sections p e r restoration; therefore, four sections per tooth) were made by two independent investigators, using a light optical stereomicroscope (Olympus PM10-AK Semi-Automatic Photomicrographic Camera System, Olympus Corp., Lake Success, NY) at 4× magnification. Photomicro-

graphs were exposed for evaluation of silver nitrate penetration of the restoration margins. The two sections for each restoration were ranked, then averaged for analysis. The penetration of the dye was scored according to standardized categories reported by Fuks and Shey (1983). Six degrees of marginal leakage are distinguished as follows: Degree 0: No dye penetration. Degree 1: Dye penetration along the occlusal or gingival wall of the restoration, adjacent to enamel only. Degree 2: Dye penetration along the entire length of the occlusal or gingival wall of the restoration, but not along the pulpal wall. Degree 3: Dye penetration along the entire length of the occlusal or gingival wall of the restoration, including the pulpal wall. Degree 4: Dye penetration along the restoration and diffusion of dye into dentin from the pulpal wall. Degree 5: Dye penetration along the restoration, and diffusion of dye through dentin to the pulp chamber.

RESULTS Results demonstrating the marginal microleakage are presented in Figs. 5-8. Conservative modified restorations and conventional restorations, using glass-ionomer liner, had less marginal microleakage, in both primary and permanent teeth, than theft" calcium hydroxide counterparts. The chi-square test indicated statistically that there was significantly less marginal microleakage in Class II conservative and conventional resin composite restorations when a glassionomer liner was used, compared with a calcium hydroxide liner (p