- Email: [email protected]
Margin analysis of posterior composites in vivo
Margin analysis of posterior composites in vivo J.F. Roulet B. Salchow M. Wald
Free University of Berlin Department of Operative Dentistry and Endodontics North FOhrer Strasse 15 D-IO00 Berlin 65, Germany Received January 30, 1990 Accepted August 13, 1990 Dent Mater 7:44-49, January, 1991
Abstract-One of the major shortcomings of posterior composites is their tendency to shrink during polymerization. Thirty-three class II cavities were filled with Occlusin (OCC) (12), Coltene EX-D2 (CEX) (10), and Dispersalloy (AM) (11). The composites were placed in beveled, enamel-etched cavities by means of an incremental technique. One week and seven mo after placement, replicas were obtained and quantitatively analyzed in the SEM. The composites showed higher percentages of "excellent margin" (EM) (OCC, 70.6%; CEX-D2, 66.6%) than AM (49.2%) at baseline (p < 0.01) and after seven mo (p < 0.01). During the experimental period, the percentage of EM decreased by 10% for AM and OCC and by 14% for CEX-D2 (p < 0.01). With the composites, an increase of "submargination" (SM) was found: OCC, 8.8%, and CEX-D2, 12.7%. An increase (11.2%) in margin fractures occurred for amalgam. Only OCC showed a low but significant increase (6%) in "marginal openings" (MO). This study confirms the inferior micromorphology of AM at the margins. The high amount of MO with the resin composites (16% after seven mo) indicates that, despite the complicated application technique, leakage of the restorations will occur. The margin analysis showed SM (OCC, 18%; CEXD2, 21%) after seven mo, indicating severe wear. Glass II amalgam restorations should not be replaced with the composites used in this study.
ecause of increased esthetic deimands and increasing environmental concerns r e g a r d i n g mercury, the use of amalgam substitutes is an important subject. In order for Class II restorations to function, these restorations must fulfill many requirements (Lutz and Roulet, 1982): (1) Wear resistance must be as good as or better than that of amalgam. (2) The restorations must have an impermeable seal and ideal marginal adaptation which must not deteriorate in clinical service. (3) The substitute must have a radiopacity at least greater than that of enamel. (4) The longevity of the restorations must be predictable. (5) The restorations must be dimensionally stable under occlusal forces during use. (6) The restorative material must require technically simple working procedures. (7) The material must be able to be finished quickly, by exact techniques, without tooth destruction. (8) Restorations must be esthetically pleasing in terms of color and translucency. The marginal integrity is the most important of all these points, because the absence of a tight seal allows micro-organisms to penetrate the restoration interface (BriinnstrSm and Nyborg, 1971, 1973; BriinnstrSm and Nordenvall, 1978), with subsequent pulpal disease (BrfinnstrSm and Nordenvall, 1978; BriinnstrSm and Vojinovic, 1976). Composite restorations must have a perfect marginal morphology, Le., no visible defects at the margins, because any gap formation means that the restoration will leak. In contrast, amalgam restorations, even with morphologically poor margins, do not leak, because the marginal gaps fill with corrosion products over time (Grossman et al., 1986; Passi et at, 1983).
B
44 ROULET et a£/in vivo MARGINS OF POSTERIOR COMPOSITES
There are many techniques for assessing the quality of restoration m a r g i n s (Roulet, 1987b, 1989). Functional tests for assessing restoration leakage are destructive and therefore not applicable for longitudinal studies (Roulet, 1989). Therefore, morphological methods are the best at the present time. Quantitative margin analysis in the SEM is a valuable tool for the clinical assessm e n t of composite r e s t o r a t i o n s (Roulet et al., 1989). The data for composites used to restore anterior teeth are excellent. When the adhesive technique is used, it is possible for high percentages of "excellent margins" to be achieved (Noack, 1988a, b). However, its use for class II cavities has shown poorer results (Roulet, 1987a; Resch and Roulet, 1986). In vivo studies of posterior composites when only clinical parameters were used (USPHS criteria) have given satisfying results (Wilson et al., 1986; Letzel et aL, 1989). However, these methods are not sensitive enough to give reliable data on the marginal behavior of posterior restorations. The objective of this study was to use the quantitative margin analysis to test the in vivo marginal behavior of two posterior composites and an amalgam. MATERIALS AND METHODS
Thirty-one patients were selected for the study. They all needed posterior restorations and were willing to participate in the ~xperiment. Only patients with a complete periodontally sound dentition and without any parafunctions were included in the study. Ages ranged between 18 and 35 years. All dentitions were completely reconstructed before patients received single experimental restorations from among the materials listed in Table 1. The location and distribution of the restorations are shown in Table 2. At baseline, 12 Occlusin (OCC), 10 Coltene Experimental D2 (CEX),
and 11 amalgam (AM) restorations were placed. The cavities for the composite restorations were cut with pear-shaped diamond burs in a high-speed handpiece and were prepared to fulfill the requirements of the adhesive technique, i.e., as much enamel as possible was conserved. Additionally, the enamel margins were beveled one mm with a fine-grained diamond bur (Composhape, Intensiv SA, CH-6962 Viganello-Lugano, Switzerland) at a maximum speed of 40,000 rpm. Excavation of the caries lesion and all subsequent steps were performed with a rubber dam in place. The enamel margins were etched for 60 s with 40% phosphoric acid, rinsed for 60 s, and dried carefully with a blast of warm air. A thin layer of bonding resin was applied, after which transparent matrix bands were placed with a Tofflemire matrix bandholder. So that good cervical adaptation would be ensured, the matrix bands were carefully wedged. The dental unit light was turned off and the operative area illuminated with yellow light only, so that the composites would be prevented from polymerizing during application. The composite m a t e r i a l was applied in increments with a Centrix syringe and cured for 60 s per increment. Proximal boxes were filled with three and the occlusal portions with two increments. After placement of the composite, the rubber dam was removed and the restoration finished by use of Soflex discs (3M, St. Paul, MN 55144-1000, USA) and Composhape diamonds. The occlusal contacts were established and carefully checked with a thin occlusal indicator foil. The amalgam restorations were also placed by use of rubber dam and conventional hand condensation. Resin composite and amalgam r e s t o r a t i o n s were polished from seven to 10 days later with Soflex discs and Composhape diamonds. In addition, the amalgam restorations were polished with silicone rubber polishers. Right after being polished and after seven mo in service, replicas were obtained by the following technique: After placement of a rubber dam, the teeth were carefully cleaned with a
TABLE 1
COMPOSITE MATERIALSUSED IN THE STUDY OCC (%w) 85 15 49.5 49.5
Fillers (total)
Resin (total) UDMA TEGDMA BisGMA Ethox.BisGMA Initiators
CEX (%w) 79 21 30 35 34 1
-
1
TABLE 2
DISTRIBUTION OF THE RESTORATIONS Tooth Number (ISO)
16
17
CEX
3
0
OCC
0
AM
1
1
26 2 1
27 0 1
36 0 3
37 1 2
46 3 1
47 1 1
0
0
0
2
2
4
2
soft-bristle brush and toothpaste, dental floss, and sodium hypochloride for removal of all organic films and plaque. After restorations were carefully rinsed with water spray and dried with a blast of warm air, a condensation silicone impression material (Silasoft, Detax, Karl Huber & Co KG, D-7500 Karlsruhe, Germany) was applied with a syringe. Special care was taken that the proximal areas were filled, after which a specially designed impression tray (Fig. 1) was applied. After the material had cured, the mouth of the impression tray was opened so that the impression material would be torn at the thinnest part of the proximal area. The mouth of the impression tray was then closed again, so that the original impression form was restored. With this method, it was possible for impressions showing the complete perimeters of the MOD restorations to be obtained. The impressions were then boxed, cast with an epoxy resin (Stycast, Emerson and Cummings, B-2431 Esterlo-Oevel, Belgium), and prepared for SEM investigation. The quantitative margin analysis in the SEM (Roulet et al., 1989) was performed by use of the following criteria: MQ 1, excellent margin MQ 2, submargination MQ 3, overhang MQ 4, marginal opening MQ 5, restoration fracture MQ 6, enamel fracture MQ 7, marginal opening and submargination
MQ 8, marginal opening and overhang MQ 9, marginal opening and restoration fracture MQ 10, marginal opening and enamel fracture MQ 11, submargination and restoration fracture MQ 12, submargination and enamel fracture MQ 13, overhang and restoration fracture MQ 14, overhang and enamel fracture. The occlusal and the proximal portions of the restorations were assessed separately, image by image, at a magnification of 200x (Roulet et al., 1989). The statistical analysis was performed with the SPSS program, with
i¸ i
i
il i
Fig. 1. Single-tooth impression tray in the closed position. When the lock (top) is removed, the "jaws" can be hinged to tear the impression at the thinnest part (proximal area).
Dental Materials~January
1991
45
TABLE3 PERCENTMARGIN QUALITIESOF CLASS I] RESTORATIONS AT BASELINEAND AFTERSEVENMONTHS OCC EM SM OH MO RF
t=0 70.6 8.6 17.7 9.4 3.1
CEX t=7 60.5 17.5 14.4 15.7 5.5
t=0 66.6 8.5 21.4 10.5 2.7
AM t=7 52.6 21.2 20.4 15.8 4.9
t=0 48.5 9.4 35.7 12.4 9.0
t=7 39.8 13.2 37.6 15.1 19.9
Occlusal and proximal data are pooled. EM = excellent margin, SM = submargination, OH= overhang, MO = marginal opening, and RF = restoration fracture. p < 0.01; underlined = not significant.
TABLE4 PERCENTMARGIN QUALITIESOF THE OCCLUSALPORTIONSOF CLASS II RESTORATIONSAT BASELINEAND AFTERSEVENMONTHS OCC EM SM OH MO RF
t=0 77.0 10.6 9.1 4.7 3.9
CEX t=7 64.5 22.5 6.4 12.4 6.8
t=0 74.9 8.9 13.4 6.1 3.1
AM t=7 56.3 24.4 13.8 14,5 6.6
t=0 51.5 8.6 31.9 11.3 13.1
t=7 35.4 18.4 33.7 18.0 32.8
EM = excellent margin, SM = submargination, OH = overhang, MO = marginal opening, and RF= restoration fracture. p < 0.01; underlined = not significant.
a nonparametric analysis of variance. The entire restoration was analyzed, but separate calculations were also done for the occlusal and proximal areas. The original margin qualities were converted to more conclusive categories according to the following scheme: Excellent mar- = Q 1 gin (EM) Submargination = MQ 2 + (SM) MQ 7 + MQll + MQ 12 Overhang (OH) = MQ 3 + MQ8 + MQ 13 + MQ 14 Marginal open- = MQ 4 + ing (M0) MQ 7 + MQ8 + MQ9 + MQ 10 Restoration = MQ 5 + fracture (RF) MQ 9 + MQll + MQ 13 Enamel fracture = MQ 6 + (EF) MQ 10 + MQ 12 + MQ 14
RESULTS
The means of the converted margin qualities are given for the entire restorations in Table 3, for the occlusal portions in Table 4, and for the proximal portions of the restorations in Table 5. Note that in all cases the percent distribution of "excellent margins" decreased significantly (p<0.01) during use. After seven mo of use, the resin composite restorations showed significantly more "submargination" (p < 0.01). Occlusin also showed a significant increase of "marginal openings" up to 16% (p < 0.01). It is also obvious that the restorations were worn more occlusally, since all materials showed a significant increase of "submargination" on the occlusal surface (p < 0.01) (Table 4). On the proximal surfaces, only Occlusin showed statistically significant increased "submargination" values. After seven mo, the "restoration margin fractures" of amalgam had increased significantly to the high value of 33% (p < 0.01) (Table 5). The material comparisons are shown in Fig. 2 and Table 6. It is
46 ROULET et at/in vivo MARGINS OF POSTERIOR COMPOSITES
obvious that both composites behaved similarly in this clinical test. There were significant differences compared with amalgam, which showed fewer "excellent margins", fewer "submarginations", more "overhangs", and more "restoration margin fractures" after seven mo (p < 0.01). DISCUSSION This study can be classified as a controlled clinical study (Letzel, 1982). If well-defined assessment criteria are used, small sample sizes are sufficient (Roulet et al., 1989; Henisch, 1989). However, due to difficulties in finding participants, the sample group in the present study is not randomized. The materials were applied with the utmost care, regardless of the time needed. Composite restorations were placed only if the cavities w e r e completely s u r r o u n d e d by enamel, so that the enamel-etching technique could be used most advantageously. The cavity form was optimized so t h a t the n e g a t i v e properties of the composite (e.g., polymerization shrinkage) would be minimized. With the incremental technique (horizontal increments), good margin qualities were expected in the proximal boxes as well. The composites were finished and polished with the Composhape diamonds (Intensiv) and/or Sofiex discs (3M) exclusively. Both procedures are the least destructive to resin composite margins (Krejci and Lutz, 1984; Noack, 1988a). These treatment criteria may be more ideal than what is true for the average clinical case. Therefore, the above results are likely to favor the composite restorations compared with what can be expected to be typical clinical behavior. Experience has shown that it is much more difficult to place composite than amalgam fillings. Tight proximal contact and proper occlusal morphology were the most difficult steps in the placement of the composites. Therefore, we agree with other studies (Leidal, 1985; Leidal et at, 1985) that the placement of posterior composites is much more timeconsuming than the placement of amalgam restorations. Quantitative margin analysis in the
SEM offers many advantages (Roulet, 1989; Roulet et al., 1989) because it is non-destructive and gives a high degree of detail reproduction. The technique is also highly reliable (Roulet et al., 1989; Henisch, 1989), and early changes can be detected, which explains the short experimental time of seven mo. The problem of this study was to produce replicas which also showed the cervical margins on the proximal surfaces. The technique that used the variable impression trays, combined with a condensation silicone impression material with a low tear strength, worked well as long as the proximal areas were clean and located supragingivally. If the margins were located subgingivally, production of proper replicas was difficult. Fourteen margin qualities were determined so that assessment would be facilitated. With these criteria, the evaluator was not forced to weight the quality decision. Since clinical combinations of different criteria often occur, it seems reasonable for combinations to be defined as single categories. A screening of the raw data has shown that it is advantageous, for the characterization of the changes, for the 14 margin qualities to be reduced to six categories. Thus, it was possible for the behavior of the materials to be shown more clearly. Since we expected different clinical wear rates in the occlusal and the proximal areas, both areas were assessed separately. Due to the high specificity of the quantitative margin analysis in the SEM, the results are comparable only with those of other studies that used the same technique. It was obvious that large restorations had a much lower percentage of "excellent margins" than the best results obtained with small class III cavities in vitro (Noack, 1988a) and in vivo (Noack, 1988b), where 88% "excellent margins" were reported at baseline and 82% "excellent margins" were still present after six mo. The baseline results of the present study (71% "excellent margins" with Occlusin and 67% "excellent margins" with Colt~ne EX-D2) are slightly lower than those reported in an in vitro study by Resch and Roulet (1986), where values from 78 - 85% "excellent mar-
TABLE 5 PERCENTMARGIN QUALITIES OF THE PROXIMAL PORTIONS OF CLASS II RESTORATIONSAT BASELINE AND AFTER SEVENMONTHS OCC EM SM OH MO RF
t=O 64 2 6.7 26.3 14.0 2.3
CEX
t=7 56.4 12.5 21.6 19.1 3.4
t=O 58.3 88.0 29.4 14.8 2.4
AM
t=7 48.9 17.9 27.0 17.0 3.2
t=O 45.6 10.2 39.5 13.5 4.9
t=7 44.2 8.1 41.5 12.2 7.0
EM = excellent margin, SM = submargination, OH = overhang, MO = marginal opening, and RF = restoration fracture. p > 0.01; underlined = not significant.
gins" were found for hybrid posterior composites. The difference can be easily explained by the greater difficulties encountered in the in vivo situation. This difference (approximately 10%) was also observed for a n t e r i o r composite r e s t o r a t i o n s (Noack, 1986). After seven mo in vivo, the number of "excellent margins" from all three materials decreased by 12%. Resch and Roulet (1986) also reported a drop of 12% with "excellent margins" after 2000 thermal cycles from 5°C to 55°C. Therefore, the results of this study support the use of thermocycling as a screening test, because it can be concluded that thermocycling is a good predictor of the clinical margin behavior of resin composite restorations. In the present study, relatively high amounts of "marginal openings" were found (Tables 3-5). The value of approximately 15% after seven mo in vivo is a clear indication that leakage occurred. With the resin composites, the percentage of marginal openings was higher in the proximal area (Table 5). These data are in contrast to data found with adhesive inlay systems. Roulet (1987a) reported fewer than 10% marginal openings after seven mo in vivo for experimental microfilled composite inlays, and Herder (1988) reported fewer than 1% marginal openings for cast glass-ceramic inlays. In the present study, significant increases of "submargination" were seen. After seven mo, the portions showing submarginated restorations increased by 9% for Occlusin and 13% for Colt~ne EX-D2. This was considerably better than the values re-
p o r t e d by Roulet (1987a) with experimental composites (43%-65% "negative ledges"). Despite the considerable improvement, these values c l e a r l y indicate t h a t the composites used are subjected to severe wear in the contact-free zone. Since it is known that wear is much more severe in the occlusal contact zone (Lutz et al., 1984), considerable wear of these materials occurs. These observations must be confirmed with measured wear data. In contrast to the composites, the decrease in the number of excellent margins of the amalgam restorations was due to the significant increase in margin fractures. Such fractures were found especially in the occlusal area, where an increase of 20% was seen. This confirms the known fact that amalgam restorations are prone to margin f r a c t u r e s ( J S r g e n s e n and W a k u m o t o , 1968; L e t z e l and Vrijhoef, 1983). Comparing the margin quality of composite restorations and amalgams is problematic, because the margin quality of both restoration types has a different clinical impact. As mentioned in the opening paragraphs, marginal openings of composite restorations are the reason for multiple problems (Br~innstrSm and Nyborg, 1971, 1973; Br~innstr5m and Vojinovic, 1976; Br~innstrSm and Nordenvall, 1978). Amalgam restorations tend to fill their marginal gaps with corrosion products (Grossman et al., 1986; Passi et aI., 1983), which seems to prevent further damage. Both composites tested showed poor margins after seven mo in vivo. Occlusin was rated slightly better after seven mo, because it had significantly more "excellent margins" "Dental Materials~January 1991 47
100
90
I-10CC,
80
i
70
!
m c
t=o
t=o
~,t=o
[ ] OCC, tffi7
CEX,t=7
L I AM, tffi7
60
%
50
n.s.= not significant
40 30
OCClGEI(
*.-l~$-a
CIEX/AM OCClAIM
i-$~ ~---@-----J
L $.2 [email protected] L...$...--=
L-I~I--~ L-ns .-~ L---ns----J
EM
Lns--J ~@--J L---.ns---J
t--lII--J L..O..j L-..@-...~
SM
Lf14--4 L@.., ~..@ ,
L--f1--J L na--J L---ns----J
OH
L--ltS--J ~ne--J L--nS----J
MO
LIIS--J LnS..J L--nS....J
=--.'41.2 [email protected] [email protected]
RF
Percent margin quality of class II restorations
Fig. 2. Percent margin qualities of class II restorationsat baseline and after seven mo. Material comparisons,Occlusaland proximal data are pooled. EM = excellent margin, SM = submarglnaUon,OH = overhang, MO = marginal opening, and RF = restorationfracture.
than Colt~ne EX-D2. However, both materials showed the same amount of "marginal openings" (16%). CONCLUSIONS
Despite very careful application of the posterior composites, by use of the adhesive technique and a horizontal incremental technique, it was not possible for good margin qualities to be obtained with the composites t e s t e d . P r e s e n c e of 16% "marginal openings" indicates restoration leakage and poor function. The rather large number of submarginated composite restorations (23% occlusally) after seven mo in vivo suggests that wear occurred. Therefore, we conclude that class II amalgam r e s t o r a t i o n s should not be replaced on a routine basis with the composites used in this study. REFERENCES
BR~NNSTROM, M. and N O R D E N V A L L , K.J. (1978): Bacterial Penetration, Pulpal Reaction and the Inner Surface
of Concise Enamel Bond. Composite Fillings in Etched and Unetched Cavities, J Dent Res 57:3-10. BRANNSTR(~M,M. and NYBORG,H. (1971): The Presence of Bacteria in Cavities Filled with Silicate Cement and Composite Resin Materials, Swed Dent J 64:149-155. BRkm~STROM,M. and NYBORG,H. (1973): Cavity Treatment with a Microbicidal Fluoride Solution: Growth of Bacteria and Effect on the Pulp, J Prosthet Dent 30:303--310. BR~STROM, M. and VoJmovIc, O. (1976): Response of the Dental Pulp to Invasion of Bacteria around Three Filling Materials, J Dent Child 43:1521. GROSSM.~_~, E.S.; WITCOMB, M.J.; and JODMrdN, A. (1986): Elements in Marginal Seals at Amalgam-Tooth Interfaces, J Dent Res 65:998--1000. HENISCH, G. (1989): E i n computergestiAtztes System zur Erfassung und Verwaltung yon Messdaten im Rahmen der quantitativen Randanalyse am Rasterelektronenmikroskop. The-
sis, Free University, Berlin. HERDER, S. (1988): In-vivo-Untersu-
48 ROULET et a£/in vivo MARGINS OF POSTERIOR COMPOSITES
chung der marginalen Adaptation adh ~ i v befestigter Glaskeramikinlays.
Thesis, Free University, Berlin. JORGE~SEN, K.D. and W~U~OTO, S. (1968): Occlusal Amalgam Fillings: Marginal Defects and Secondary Caries, Odontol Tidskr 76:43--54. KREJCI, I. and LUTZ, F. (1984): Kompositf'tillungen-das 1 real 1 des Ausarbeitens, Schweiz Mschr Zahnheilk 94:1015--1028. LEIDAL, T.I. (1985): Discussion Paper: Accomplishments and Expectations with Posterior Composite Resins. In: Composite Resin Restorative Materials, G. Vanherle and D.C. Smith, Eds., Minneapolis, MN: Minnesota Mining & Mfg. Co. LEIDAL, T.I.; SOLEM,H.; and RYKKE,M. (1985): A Clinical and Scanning Electron Microscopic Study of a New Restorative Material for Use in Posterior Teeth, Acta Odontol Scand 43:1--8. LETZEL, H. (1982): Anforderungen an die klinische Prtifung yon Kunststoff-Fiillungsmaterialien. In: Ffillungswerkstoffe auf Kunststoffbasis, M.F. Mfiller-Kalben, Ed. Leverkusen, Germany: Bayer Dental.
LETZEL, H.; VAN'T HOF, M.A.; VRIJHOEF, M.M~A.; MARSHALL,G.W.; and MARSHALL,S.J. (1989): Failure, Survival, and Reasons for Replacement of Amalgam Restorations. In: Quality Evaluation of Dental Restorations, K.J. Anusavice, Ed., Chicago: Quintessence, pp. 83-94. LETZEL, H. and VRIJHOEF, M.M.A. (1983): Survival Rate of Dental Amalgam Restorations, J Dent Res 61:269, Abstr. No. 820. L~TZ, F.; PHILLIPS, R.W.; ROULET,J.F.; and SETCOS, J.C. (1984): Potential Posterior Composites-An in vivo and in vitro Comparison of Wear, J Dent Res 63:914-920. LUTZ, F. and ROULET,J.F. (1982): Composite Resin Systems: Curing Techniques and Clinical Performance. In: Ffillungswerkstoffe auf Kunststoffbasis, M.F. Mfiller-Kalben, Ed., Leverkusen: Bayer Dental, pp. 21-29. NOACK,M. (1986): In vitro- und in vivoVergleich eines experimentellen Hybridkomposits mit einem inhomogehen Mikrofi~llerkomposit bei Klasse III-Kavitgten. Thesis, Free University, Berlin. NOACK, M. (1988a): Quantitativ Filllungsrandanalyse von Frontzahnkom-
TABLE6 STATISTICALDIFFERENCESOF MATERIALS Before TC OCC/CEX CEX/AM OCC/AM
After TC
EM
MO
SM
OH
RF
EM
MO
SM
OH
RF
NS * *
NS NS NS
NS NS NS
NS * *
NS NS NS
* * *
NS NS NS
NS * NS
NS * *
NS * *
positen im Rasterelektronenmikroskop nach thermischer Wechselbadbelastung, Dtsch Zahn(trztl Z 43:295-299. NOACK, M. (1988b): Randschluss yon Frontzahnfiillungen mit Hybrid- und Mikroffillerkompositen nach einer Tragezeit von 6 Monaten, Dtsch Zahndrztl Z 43:919-922. PASSI, P.; MUSAJO, F.; and FURLAN,C. (1983): La chiusura marginale delle otturazioni in amalgama: microanalisi dei detriti inorganici che concorrono a formare il sigillo (nota II), Dent Cadmos 51:37-48. RESCH, H. and ROULET,J.F. (1986): In vitro Marginal Integrity of MOD Fillings with Hybrid and Microfilled Composites, J Dent Res 65: 826, Abstr. No. 902. ROULET, J.F. (1987a): Degradation of Dental Polymers. Basel: Karger.
ROULET, J.F. (1987b): A Materials Scientist's View: Assessment of Wear and Marginal Integrity, Quint Int 18:543552. ROULET, J.F. (1989): Margin Quality: Criteria and Techniques for Assessment. In: Quality Evaluation of Dental Restorations, K.J. Anusavice, Ed., Chicago: Quintessence, pp. 223-241. ROULET, J.F.; REICH, T.; BLUNCK, U.; and NOACK, M. (1989): Quantitative Margin Analysis in the Scanning Electron Microscope, Scanning Microsc 3:147-159. WILSON, N.H.F.; SMITH, G.A.; and WILSON, M.A. (1986): A Clinical Trial of a Visible Light Cured Posterior Composite Restorative Material: Three-year Results, Quint Int 17:643652.
Dental Materials~January 1991 4g
Free University of Berlin Department of Operative Dentistry and Endodontics North FOhrer Strasse 15 D-IO00 Berlin 65, Germany Received January 30, 1990 Accepted August 13, 1990 Dent Mater 7:44-49, January, 1991
Abstract-One of the major shortcomings of posterior composites is their tendency to shrink during polymerization. Thirty-three class II cavities were filled with Occlusin (OCC) (12), Coltene EX-D2 (CEX) (10), and Dispersalloy (AM) (11). The composites were placed in beveled, enamel-etched cavities by means of an incremental technique. One week and seven mo after placement, replicas were obtained and quantitatively analyzed in the SEM. The composites showed higher percentages of "excellent margin" (EM) (OCC, 70.6%; CEX-D2, 66.6%) than AM (49.2%) at baseline (p < 0.01) and after seven mo (p < 0.01). During the experimental period, the percentage of EM decreased by 10% for AM and OCC and by 14% for CEX-D2 (p < 0.01). With the composites, an increase of "submargination" (SM) was found: OCC, 8.8%, and CEX-D2, 12.7%. An increase (11.2%) in margin fractures occurred for amalgam. Only OCC showed a low but significant increase (6%) in "marginal openings" (MO). This study confirms the inferior micromorphology of AM at the margins. The high amount of MO with the resin composites (16% after seven mo) indicates that, despite the complicated application technique, leakage of the restorations will occur. The margin analysis showed SM (OCC, 18%; CEXD2, 21%) after seven mo, indicating severe wear. Glass II amalgam restorations should not be replaced with the composites used in this study.
ecause of increased esthetic deimands and increasing environmental concerns r e g a r d i n g mercury, the use of amalgam substitutes is an important subject. In order for Class II restorations to function, these restorations must fulfill many requirements (Lutz and Roulet, 1982): (1) Wear resistance must be as good as or better than that of amalgam. (2) The restorations must have an impermeable seal and ideal marginal adaptation which must not deteriorate in clinical service. (3) The substitute must have a radiopacity at least greater than that of enamel. (4) The longevity of the restorations must be predictable. (5) The restorations must be dimensionally stable under occlusal forces during use. (6) The restorative material must require technically simple working procedures. (7) The material must be able to be finished quickly, by exact techniques, without tooth destruction. (8) Restorations must be esthetically pleasing in terms of color and translucency. The marginal integrity is the most important of all these points, because the absence of a tight seal allows micro-organisms to penetrate the restoration interface (BriinnstrSm and Nyborg, 1971, 1973; BriinnstrSm and Nordenvall, 1978), with subsequent pulpal disease (BrfinnstrSm and Nordenvall, 1978; BriinnstrSm and Vojinovic, 1976). Composite restorations must have a perfect marginal morphology, Le., no visible defects at the margins, because any gap formation means that the restoration will leak. In contrast, amalgam restorations, even with morphologically poor margins, do not leak, because the marginal gaps fill with corrosion products over time (Grossman et al., 1986; Passi et at, 1983).
B
44 ROULET et a£/in vivo MARGINS OF POSTERIOR COMPOSITES
There are many techniques for assessing the quality of restoration m a r g i n s (Roulet, 1987b, 1989). Functional tests for assessing restoration leakage are destructive and therefore not applicable for longitudinal studies (Roulet, 1989). Therefore, morphological methods are the best at the present time. Quantitative margin analysis in the SEM is a valuable tool for the clinical assessm e n t of composite r e s t o r a t i o n s (Roulet et al., 1989). The data for composites used to restore anterior teeth are excellent. When the adhesive technique is used, it is possible for high percentages of "excellent margins" to be achieved (Noack, 1988a, b). However, its use for class II cavities has shown poorer results (Roulet, 1987a; Resch and Roulet, 1986). In vivo studies of posterior composites when only clinical parameters were used (USPHS criteria) have given satisfying results (Wilson et al., 1986; Letzel et aL, 1989). However, these methods are not sensitive enough to give reliable data on the marginal behavior of posterior restorations. The objective of this study was to use the quantitative margin analysis to test the in vivo marginal behavior of two posterior composites and an amalgam. MATERIALS AND METHODS
Thirty-one patients were selected for the study. They all needed posterior restorations and were willing to participate in the ~xperiment. Only patients with a complete periodontally sound dentition and without any parafunctions were included in the study. Ages ranged between 18 and 35 years. All dentitions were completely reconstructed before patients received single experimental restorations from among the materials listed in Table 1. The location and distribution of the restorations are shown in Table 2. At baseline, 12 Occlusin (OCC), 10 Coltene Experimental D2 (CEX),
and 11 amalgam (AM) restorations were placed. The cavities for the composite restorations were cut with pear-shaped diamond burs in a high-speed handpiece and were prepared to fulfill the requirements of the adhesive technique, i.e., as much enamel as possible was conserved. Additionally, the enamel margins were beveled one mm with a fine-grained diamond bur (Composhape, Intensiv SA, CH-6962 Viganello-Lugano, Switzerland) at a maximum speed of 40,000 rpm. Excavation of the caries lesion and all subsequent steps were performed with a rubber dam in place. The enamel margins were etched for 60 s with 40% phosphoric acid, rinsed for 60 s, and dried carefully with a blast of warm air. A thin layer of bonding resin was applied, after which transparent matrix bands were placed with a Tofflemire matrix bandholder. So that good cervical adaptation would be ensured, the matrix bands were carefully wedged. The dental unit light was turned off and the operative area illuminated with yellow light only, so that the composites would be prevented from polymerizing during application. The composite m a t e r i a l was applied in increments with a Centrix syringe and cured for 60 s per increment. Proximal boxes were filled with three and the occlusal portions with two increments. After placement of the composite, the rubber dam was removed and the restoration finished by use of Soflex discs (3M, St. Paul, MN 55144-1000, USA) and Composhape diamonds. The occlusal contacts were established and carefully checked with a thin occlusal indicator foil. The amalgam restorations were also placed by use of rubber dam and conventional hand condensation. Resin composite and amalgam r e s t o r a t i o n s were polished from seven to 10 days later with Soflex discs and Composhape diamonds. In addition, the amalgam restorations were polished with silicone rubber polishers. Right after being polished and after seven mo in service, replicas were obtained by the following technique: After placement of a rubber dam, the teeth were carefully cleaned with a
TABLE 1
COMPOSITE MATERIALSUSED IN THE STUDY OCC (%w) 85 15 49.5 49.5
Fillers (total)
Resin (total) UDMA TEGDMA BisGMA Ethox.BisGMA Initiators
CEX (%w) 79 21 30 35 34 1
-
1
TABLE 2
DISTRIBUTION OF THE RESTORATIONS Tooth Number (ISO)
16
17
CEX
3
0
OCC
0
AM
1
1
26 2 1
27 0 1
36 0 3
37 1 2
46 3 1
47 1 1
0
0
0
2
2
4
2
soft-bristle brush and toothpaste, dental floss, and sodium hypochloride for removal of all organic films and plaque. After restorations were carefully rinsed with water spray and dried with a blast of warm air, a condensation silicone impression material (Silasoft, Detax, Karl Huber & Co KG, D-7500 Karlsruhe, Germany) was applied with a syringe. Special care was taken that the proximal areas were filled, after which a specially designed impression tray (Fig. 1) was applied. After the material had cured, the mouth of the impression tray was opened so that the impression material would be torn at the thinnest part of the proximal area. The mouth of the impression tray was then closed again, so that the original impression form was restored. With this method, it was possible for impressions showing the complete perimeters of the MOD restorations to be obtained. The impressions were then boxed, cast with an epoxy resin (Stycast, Emerson and Cummings, B-2431 Esterlo-Oevel, Belgium), and prepared for SEM investigation. The quantitative margin analysis in the SEM (Roulet et al., 1989) was performed by use of the following criteria: MQ 1, excellent margin MQ 2, submargination MQ 3, overhang MQ 4, marginal opening MQ 5, restoration fracture MQ 6, enamel fracture MQ 7, marginal opening and submargination
MQ 8, marginal opening and overhang MQ 9, marginal opening and restoration fracture MQ 10, marginal opening and enamel fracture MQ 11, submargination and restoration fracture MQ 12, submargination and enamel fracture MQ 13, overhang and restoration fracture MQ 14, overhang and enamel fracture. The occlusal and the proximal portions of the restorations were assessed separately, image by image, at a magnification of 200x (Roulet et al., 1989). The statistical analysis was performed with the SPSS program, with
i¸ i
i
il i
Fig. 1. Single-tooth impression tray in the closed position. When the lock (top) is removed, the "jaws" can be hinged to tear the impression at the thinnest part (proximal area).
Dental Materials~January
1991
45
TABLE3 PERCENTMARGIN QUALITIESOF CLASS I] RESTORATIONS AT BASELINEAND AFTERSEVENMONTHS OCC EM SM OH MO RF
t=0 70.6 8.6 17.7 9.4 3.1
CEX t=7 60.5 17.5 14.4 15.7 5.5
t=0 66.6 8.5 21.4 10.5 2.7
AM t=7 52.6 21.2 20.4 15.8 4.9
t=0 48.5 9.4 35.7 12.4 9.0
t=7 39.8 13.2 37.6 15.1 19.9
Occlusal and proximal data are pooled. EM = excellent margin, SM = submargination, OH= overhang, MO = marginal opening, and RF = restoration fracture. p < 0.01; underlined = not significant.
TABLE4 PERCENTMARGIN QUALITIESOF THE OCCLUSALPORTIONSOF CLASS II RESTORATIONSAT BASELINEAND AFTERSEVENMONTHS OCC EM SM OH MO RF
t=0 77.0 10.6 9.1 4.7 3.9
CEX t=7 64.5 22.5 6.4 12.4 6.8
t=0 74.9 8.9 13.4 6.1 3.1
AM t=7 56.3 24.4 13.8 14,5 6.6
t=0 51.5 8.6 31.9 11.3 13.1
t=7 35.4 18.4 33.7 18.0 32.8
EM = excellent margin, SM = submargination, OH = overhang, MO = marginal opening, and RF= restoration fracture. p < 0.01; underlined = not significant.
a nonparametric analysis of variance. The entire restoration was analyzed, but separate calculations were also done for the occlusal and proximal areas. The original margin qualities were converted to more conclusive categories according to the following scheme: Excellent mar- = Q 1 gin (EM) Submargination = MQ 2 + (SM) MQ 7 + MQll + MQ 12 Overhang (OH) = MQ 3 + MQ8 + MQ 13 + MQ 14 Marginal open- = MQ 4 + ing (M0) MQ 7 + MQ8 + MQ9 + MQ 10 Restoration = MQ 5 + fracture (RF) MQ 9 + MQll + MQ 13 Enamel fracture = MQ 6 + (EF) MQ 10 + MQ 12 + MQ 14
RESULTS
The means of the converted margin qualities are given for the entire restorations in Table 3, for the occlusal portions in Table 4, and for the proximal portions of the restorations in Table 5. Note that in all cases the percent distribution of "excellent margins" decreased significantly (p<0.01) during use. After seven mo of use, the resin composite restorations showed significantly more "submargination" (p < 0.01). Occlusin also showed a significant increase of "marginal openings" up to 16% (p < 0.01). It is also obvious that the restorations were worn more occlusally, since all materials showed a significant increase of "submargination" on the occlusal surface (p < 0.01) (Table 4). On the proximal surfaces, only Occlusin showed statistically significant increased "submargination" values. After seven mo, the "restoration margin fractures" of amalgam had increased significantly to the high value of 33% (p < 0.01) (Table 5). The material comparisons are shown in Fig. 2 and Table 6. It is
46 ROULET et at/in vivo MARGINS OF POSTERIOR COMPOSITES
obvious that both composites behaved similarly in this clinical test. There were significant differences compared with amalgam, which showed fewer "excellent margins", fewer "submarginations", more "overhangs", and more "restoration margin fractures" after seven mo (p < 0.01). DISCUSSION This study can be classified as a controlled clinical study (Letzel, 1982). If well-defined assessment criteria are used, small sample sizes are sufficient (Roulet et al., 1989; Henisch, 1989). However, due to difficulties in finding participants, the sample group in the present study is not randomized. The materials were applied with the utmost care, regardless of the time needed. Composite restorations were placed only if the cavities w e r e completely s u r r o u n d e d by enamel, so that the enamel-etching technique could be used most advantageously. The cavity form was optimized so t h a t the n e g a t i v e properties of the composite (e.g., polymerization shrinkage) would be minimized. With the incremental technique (horizontal increments), good margin qualities were expected in the proximal boxes as well. The composites were finished and polished with the Composhape diamonds (Intensiv) and/or Sofiex discs (3M) exclusively. Both procedures are the least destructive to resin composite margins (Krejci and Lutz, 1984; Noack, 1988a). These treatment criteria may be more ideal than what is true for the average clinical case. Therefore, the above results are likely to favor the composite restorations compared with what can be expected to be typical clinical behavior. Experience has shown that it is much more difficult to place composite than amalgam fillings. Tight proximal contact and proper occlusal morphology were the most difficult steps in the placement of the composites. Therefore, we agree with other studies (Leidal, 1985; Leidal et at, 1985) that the placement of posterior composites is much more timeconsuming than the placement of amalgam restorations. Quantitative margin analysis in the
SEM offers many advantages (Roulet, 1989; Roulet et al., 1989) because it is non-destructive and gives a high degree of detail reproduction. The technique is also highly reliable (Roulet et al., 1989; Henisch, 1989), and early changes can be detected, which explains the short experimental time of seven mo. The problem of this study was to produce replicas which also showed the cervical margins on the proximal surfaces. The technique that used the variable impression trays, combined with a condensation silicone impression material with a low tear strength, worked well as long as the proximal areas were clean and located supragingivally. If the margins were located subgingivally, production of proper replicas was difficult. Fourteen margin qualities were determined so that assessment would be facilitated. With these criteria, the evaluator was not forced to weight the quality decision. Since clinical combinations of different criteria often occur, it seems reasonable for combinations to be defined as single categories. A screening of the raw data has shown that it is advantageous, for the characterization of the changes, for the 14 margin qualities to be reduced to six categories. Thus, it was possible for the behavior of the materials to be shown more clearly. Since we expected different clinical wear rates in the occlusal and the proximal areas, both areas were assessed separately. Due to the high specificity of the quantitative margin analysis in the SEM, the results are comparable only with those of other studies that used the same technique. It was obvious that large restorations had a much lower percentage of "excellent margins" than the best results obtained with small class III cavities in vitro (Noack, 1988a) and in vivo (Noack, 1988b), where 88% "excellent margins" were reported at baseline and 82% "excellent margins" were still present after six mo. The baseline results of the present study (71% "excellent margins" with Occlusin and 67% "excellent margins" with Colt~ne EX-D2) are slightly lower than those reported in an in vitro study by Resch and Roulet (1986), where values from 78 - 85% "excellent mar-
TABLE 5 PERCENTMARGIN QUALITIES OF THE PROXIMAL PORTIONS OF CLASS II RESTORATIONSAT BASELINE AND AFTER SEVENMONTHS OCC EM SM OH MO RF
t=O 64 2 6.7 26.3 14.0 2.3
CEX
t=7 56.4 12.5 21.6 19.1 3.4
t=O 58.3 88.0 29.4 14.8 2.4
AM
t=7 48.9 17.9 27.0 17.0 3.2
t=O 45.6 10.2 39.5 13.5 4.9
t=7 44.2 8.1 41.5 12.2 7.0
EM = excellent margin, SM = submargination, OH = overhang, MO = marginal opening, and RF = restoration fracture. p > 0.01; underlined = not significant.
gins" were found for hybrid posterior composites. The difference can be easily explained by the greater difficulties encountered in the in vivo situation. This difference (approximately 10%) was also observed for a n t e r i o r composite r e s t o r a t i o n s (Noack, 1986). After seven mo in vivo, the number of "excellent margins" from all three materials decreased by 12%. Resch and Roulet (1986) also reported a drop of 12% with "excellent margins" after 2000 thermal cycles from 5°C to 55°C. Therefore, the results of this study support the use of thermocycling as a screening test, because it can be concluded that thermocycling is a good predictor of the clinical margin behavior of resin composite restorations. In the present study, relatively high amounts of "marginal openings" were found (Tables 3-5). The value of approximately 15% after seven mo in vivo is a clear indication that leakage occurred. With the resin composites, the percentage of marginal openings was higher in the proximal area (Table 5). These data are in contrast to data found with adhesive inlay systems. Roulet (1987a) reported fewer than 10% marginal openings after seven mo in vivo for experimental microfilled composite inlays, and Herder (1988) reported fewer than 1% marginal openings for cast glass-ceramic inlays. In the present study, significant increases of "submargination" were seen. After seven mo, the portions showing submarginated restorations increased by 9% for Occlusin and 13% for Colt~ne EX-D2. This was considerably better than the values re-
p o r t e d by Roulet (1987a) with experimental composites (43%-65% "negative ledges"). Despite the considerable improvement, these values c l e a r l y indicate t h a t the composites used are subjected to severe wear in the contact-free zone. Since it is known that wear is much more severe in the occlusal contact zone (Lutz et al., 1984), considerable wear of these materials occurs. These observations must be confirmed with measured wear data. In contrast to the composites, the decrease in the number of excellent margins of the amalgam restorations was due to the significant increase in margin fractures. Such fractures were found especially in the occlusal area, where an increase of 20% was seen. This confirms the known fact that amalgam restorations are prone to margin f r a c t u r e s ( J S r g e n s e n and W a k u m o t o , 1968; L e t z e l and Vrijhoef, 1983). Comparing the margin quality of composite restorations and amalgams is problematic, because the margin quality of both restoration types has a different clinical impact. As mentioned in the opening paragraphs, marginal openings of composite restorations are the reason for multiple problems (Br~innstrSm and Nyborg, 1971, 1973; Br~innstr5m and Vojinovic, 1976; Br~innstrSm and Nordenvall, 1978). Amalgam restorations tend to fill their marginal gaps with corrosion products (Grossman et al., 1986; Passi et aI., 1983), which seems to prevent further damage. Both composites tested showed poor margins after seven mo in vivo. Occlusin was rated slightly better after seven mo, because it had significantly more "excellent margins" "Dental Materials~January 1991 47
100
90
I-10CC,
80
i
70
!
m c
t=o
t=o
~,t=o
[ ] OCC, tffi7
CEX,t=7
L I AM, tffi7
60
%
50
n.s.= not significant
40 30
OCClGEI(
*.-l~$-a
CIEX/AM OCClAIM
i-$~ ~---@-----J
L $.2 [email protected] L...$...--=
L-I~I--~ L-ns .-~ L---ns----J
EM
Lns--J ~@--J L---.ns---J
t--lII--J L..O..j L-..@-...~
SM
Lf14--4 L@.., ~..@ ,
L--f1--J L na--J L---ns----J
OH
L--ltS--J ~ne--J L--nS----J
MO
LIIS--J LnS..J L--nS....J
=--.'41.2 [email protected] [email protected]
RF
Percent margin quality of class II restorations
Fig. 2. Percent margin qualities of class II restorationsat baseline and after seven mo. Material comparisons,Occlusaland proximal data are pooled. EM = excellent margin, SM = submarglnaUon,OH = overhang, MO = marginal opening, and RF = restorationfracture.
than Colt~ne EX-D2. However, both materials showed the same amount of "marginal openings" (16%). CONCLUSIONS
Despite very careful application of the posterior composites, by use of the adhesive technique and a horizontal incremental technique, it was not possible for good margin qualities to be obtained with the composites t e s t e d . P r e s e n c e of 16% "marginal openings" indicates restoration leakage and poor function. The rather large number of submarginated composite restorations (23% occlusally) after seven mo in vivo suggests that wear occurred. Therefore, we conclude that class II amalgam r e s t o r a t i o n s should not be replaced on a routine basis with the composites used in this study. REFERENCES
BR~NNSTROM, M. and N O R D E N V A L L , K.J. (1978): Bacterial Penetration, Pulpal Reaction and the Inner Surface
of Concise Enamel Bond. Composite Fillings in Etched and Unetched Cavities, J Dent Res 57:3-10. BRANNSTR(~M,M. and NYBORG,H. (1971): The Presence of Bacteria in Cavities Filled with Silicate Cement and Composite Resin Materials, Swed Dent J 64:149-155. BRkm~STROM,M. and NYBORG,H. (1973): Cavity Treatment with a Microbicidal Fluoride Solution: Growth of Bacteria and Effect on the Pulp, J Prosthet Dent 30:303--310. BR~STROM, M. and VoJmovIc, O. (1976): Response of the Dental Pulp to Invasion of Bacteria around Three Filling Materials, J Dent Child 43:1521. GROSSM.~_~, E.S.; WITCOMB, M.J.; and JODMrdN, A. (1986): Elements in Marginal Seals at Amalgam-Tooth Interfaces, J Dent Res 65:998--1000. HENISCH, G. (1989): E i n computergestiAtztes System zur Erfassung und Verwaltung yon Messdaten im Rahmen der quantitativen Randanalyse am Rasterelektronenmikroskop. The-
sis, Free University, Berlin. HERDER, S. (1988): In-vivo-Untersu-
48 ROULET et a£/in vivo MARGINS OF POSTERIOR COMPOSITES
chung der marginalen Adaptation adh ~ i v befestigter Glaskeramikinlays.
Thesis, Free University, Berlin. JORGE~SEN, K.D. and W~U~OTO, S. (1968): Occlusal Amalgam Fillings: Marginal Defects and Secondary Caries, Odontol Tidskr 76:43--54. KREJCI, I. and LUTZ, F. (1984): Kompositf'tillungen-das 1 real 1 des Ausarbeitens, Schweiz Mschr Zahnheilk 94:1015--1028. LEIDAL, T.I. (1985): Discussion Paper: Accomplishments and Expectations with Posterior Composite Resins. In: Composite Resin Restorative Materials, G. Vanherle and D.C. Smith, Eds., Minneapolis, MN: Minnesota Mining & Mfg. Co. LEIDAL, T.I.; SOLEM,H.; and RYKKE,M. (1985): A Clinical and Scanning Electron Microscopic Study of a New Restorative Material for Use in Posterior Teeth, Acta Odontol Scand 43:1--8. LETZEL, H. (1982): Anforderungen an die klinische Prtifung yon Kunststoff-Fiillungsmaterialien. In: Ffillungswerkstoffe auf Kunststoffbasis, M.F. Mfiller-Kalben, Ed. Leverkusen, Germany: Bayer Dental.
LETZEL, H.; VAN'T HOF, M.A.; VRIJHOEF, M.M~A.; MARSHALL,G.W.; and MARSHALL,S.J. (1989): Failure, Survival, and Reasons for Replacement of Amalgam Restorations. In: Quality Evaluation of Dental Restorations, K.J. Anusavice, Ed., Chicago: Quintessence, pp. 83-94. LETZEL, H. and VRIJHOEF, M.M.A. (1983): Survival Rate of Dental Amalgam Restorations, J Dent Res 61:269, Abstr. No. 820. L~TZ, F.; PHILLIPS, R.W.; ROULET,J.F.; and SETCOS, J.C. (1984): Potential Posterior Composites-An in vivo and in vitro Comparison of Wear, J Dent Res 63:914-920. LUTZ, F. and ROULET,J.F. (1982): Composite Resin Systems: Curing Techniques and Clinical Performance. In: Ffillungswerkstoffe auf Kunststoffbasis, M.F. Mfiller-Kalben, Ed., Leverkusen: Bayer Dental, pp. 21-29. NOACK,M. (1986): In vitro- und in vivoVergleich eines experimentellen Hybridkomposits mit einem inhomogehen Mikrofi~llerkomposit bei Klasse III-Kavitgten. Thesis, Free University, Berlin. NOACK, M. (1988a): Quantitativ Filllungsrandanalyse von Frontzahnkom-
TABLE6 STATISTICALDIFFERENCESOF MATERIALS Before TC OCC/CEX CEX/AM OCC/AM
After TC
EM
MO
SM
OH
RF
EM
MO
SM
OH
RF
NS * *
NS NS NS
NS NS NS
NS * *
NS NS NS
* * *
NS NS NS
NS * NS
NS * *
NS * *
positen im Rasterelektronenmikroskop nach thermischer Wechselbadbelastung, Dtsch Zahn(trztl Z 43:295-299. NOACK, M. (1988b): Randschluss yon Frontzahnfiillungen mit Hybrid- und Mikroffillerkompositen nach einer Tragezeit von 6 Monaten, Dtsch Zahndrztl Z 43:919-922. PASSI, P.; MUSAJO, F.; and FURLAN,C. (1983): La chiusura marginale delle otturazioni in amalgama: microanalisi dei detriti inorganici che concorrono a formare il sigillo (nota II), Dent Cadmos 51:37-48. RESCH, H. and ROULET,J.F. (1986): In vitro Marginal Integrity of MOD Fillings with Hybrid and Microfilled Composites, J Dent Res 65: 826, Abstr. No. 902. ROULET, J.F. (1987a): Degradation of Dental Polymers. Basel: Karger.
ROULET, J.F. (1987b): A Materials Scientist's View: Assessment of Wear and Marginal Integrity, Quint Int 18:543552. ROULET, J.F. (1989): Margin Quality: Criteria and Techniques for Assessment. In: Quality Evaluation of Dental Restorations, K.J. Anusavice, Ed., Chicago: Quintessence, pp. 223-241. ROULET, J.F.; REICH, T.; BLUNCK, U.; and NOACK, M. (1989): Quantitative Margin Analysis in the Scanning Electron Microscope, Scanning Microsc 3:147-159. WILSON, N.H.F.; SMITH, G.A.; and WILSON, M.A. (1986): A Clinical Trial of a Visible Light Cured Posterior Composite Restorative Material: Three-year Results, Quint Int 17:643652.
Dental Materials~January 1991 4g