- Email: [email protected]
Effect of tooth surface roughness on marginal seating and retention of complete metal crowns
Effect of tooth surface roughness on marginal seating and retention of complete metal crowns Morakot Tuntiprawon, DDS, MDSca Faculty of Dentistry, Chulalongkorn University, Bangkok, Thailand Statement of problem. Retention and marginal adaptation factors have major influence on the failure of cemented complete veneer crowns.
Purpose. This study investigated the effect of axial surface roughness on the marginal seating and retention of silver-palladium crowns luted with zinc phosphate, glass ionomer, and resin cements. Material and methods. Coarse and fine diamond stones were used to create various surface roughnesses of premolars. A milling machine was used to control the height and angle of the axial walls of tooth preparations. Ten cast metal crowns in 6 subgroups were luted with 3 cements (Phosphacap, Fuji Cap I, and Panavia 21). Marginal seating was recorded with a Digimatic indicator. Retention was determined by measuring the tensile force required to remove a metal crown with a Lloyd testing machine. Results. Two-way analysis of variance revealed statistically significant differences (P<.001) in retention for both luting cements and surface roughness. No significant difference was recorded for marginal seating relative to roughness (P=.860) and interaction effects (P=.204). Tukey-HSD tests revealed substantial differences in retention among Phosphacap, Fuji Cap I, and Panavia 21 cements. Significant differences were not confirmed in marginal seating between Fuji Cap I and Phosphacap cements with coarse diamonds, and Phosphacap and Panavia 21 cements with fine diamonds. Conclusions. The best retention for complete metal crown was demonstrated for tooth preparations ground with coarse diamonds and cemented with Panavia 21 cement. Differences in axial surface roughness had no effect on the marginal seating of the complete metal crowns. (J Prosthet Dent 1999;81:142-7.)
CLINICAL IMPLICATIONS Teeth prepared with coarse diamond stones had greater retention than those prepared with fine diamond stones. Resin cement provided the best retention, but poorest seating for complete metal crowns.
T
he longevity of fixed prostheses depends on the retention and marginal integrity of restorations.1 Marginal leakage of cements has been documented as one of the major causes of failure because of secondary dental caries.2,3 Many factors have been demonstrated to influence the marginal seating and retention of crowns, such as the size and shape of prepared teeth,4 manipulation of cement,5-9 retentive properties of cement,10-18 cement film thickness,19,20 relieving space or venting for cement,21-28 cement application,29 and roughness of dentinal surface.30 Dental diamond grinding instruments are widely used because of their high efficiency of cutting and their versatility of shapes, dimensions, and grit sizes. Large grit diamonds have provided more rapid cutting speed and have created greater tooth surface roughness.31 The duration of tooth preparation is crucial in fixed prosthodontic procedures. A shorter chair time results in greater comfort for the patient and efficiency This research was supported by a Rachadapiseksompoj Research Fund. aAssistant Professor, Department of Prosthodontics. 142 THE JOURNAL OF PROSTHETIC DENTISTRY
for the dentist. A coarse diamond stone is more efficient in preparing tooth structure than fine diamonds, with larger dentinal projections on the axial walls. Mechanical retention relies on an interlocking area between the tooth and cement, and greater crosssectional area of a dentinal projection can result in greater retention of a casting.32 However, there have been other studies that have indicated minimal differences in retentive forces between rough and smooth tooth preparations.4,30,33 The purpose of this study was to determine the marginal seating and retention of complete cast metal crowns cemented to teeth prepared with different grit sizes of diamond stones, with 3 contemporary luting agents.
MATERIAL AND METHODS Sixty extracted, intact human premolars were selected according to crowns of similar sizes, cleaned, and stored in deionized water. Roots of the specimens were notched and embedded in acrylic resin blocks at the cementoenamel junction (CEJ). All standardized tooth preparations for complete metal crowns were perVOLUME 81 NUMBER 2
TUNTIPRAWON
THE JOURNAL OF PROSTHETIC DENTISTRY
Table I. Manipulation of luting cements Cements
Batch no.
Powder/ liquid ratio
Phosphacap
532419
32 g/16 g
Fuji Cap I
290861
Panavia 21
51181
0.22 g/0.11 g (0.09 mL) 1 turn of base and catalyst paste
Mixing method
Mixing time
Amalgamator (Silamat, Vivadent, Liechtenstein, Germany) Amalgamator
10 s
Hands
5s 20-30 s
formed with a milling machine (Milling machine, KaVo EWL, Leutkirch, Germany). The horizontal plane of the occlusal surface was initially prepared with a wheel stone, and an elliptical rubber template was made to stamp on the occlusal surface to outline the cross-sectional area. High-speed cylinder diamonds (Intensiv No. 210, Viganello-Lugano, Switzerland) were selected for preliminary preparation of the teeth until their axial walls were 0.5 mm away from the external outline. A random number table was used to divide prepared teeth into 2 groups of 30 specimens. In group 1, coarse diamond stones (grit size 120 µm, 117C, Intensiv) and in group 2, fine diamond stones (grit size 50 µm, 117GS Intensiv) were used to finish all preparations with a height of 3 mm, a 6-degree convergence angle, and a shoulder finish line (Fig. 1). Each stone was used only once per specimen. The shoulder finish line was refined for all tooth preparations with a hand instrument by 1 dentist. Specimens were ultrasonically cleaned in deionized water for 10 minutes. Surface roughness of axial walls was randomly examined with a SEM. All specimens were stored in deionized water before impressions were made. Each impression was made with a polyvinyl siloxane material (Provil M, Bayer Dental, Leverkusen, Germany) in a custom impression tray, then the dies were poured with improved stone (Silky Rock Yellow, Whip Mix, Louisville, Ky.). The external outline of the 0.5-mm shoulder finish line on dies were marked with a noncarbon red pencil. Axial surfaces of the dies were painted with 4 layers of diespacer (Nice-Fit, Shofu Dental, Kyoto, Japan) to the corner of the finish line. Gold and silver die spacer colors were alternately painted to prevent repeated applications. Wax patterns for crowns were formed with blue inlay wax (Kerr Mfg. Co, Romulus, Mich.), with a thickness of approximately 0.5 to 1 mm. A 5-mm diameter ring was waxed in the center of the occlusal surface of each pattern to facilitate connection of the crown to the tensile testing machine. Wax patterns were immediately FEBRUARY 1999
Fig. 1. Standardized preparation using milling machine (KaVo EWL).
sprued and invested in inlay investment (Cristobalite, Kerr Mfg. Co.). The powder/liquid ratio of 100 gm/40 mL was mixed for 1 minute with a vibrator. The wax patterns were vaporized at 350°C for 30 minutes, and the investments were heated at 850°C for 30 minutes. Crowns were cast in a silver-palladium alloy (Degussa AG, Palliag M, Frankfurt, Germany) with an electronic induction casting machine (Degutron, Degussa AG). Sprues were removed and castings were finished externally with stones and rubber points, respectively. Internal imperfections were removed with a small round diamond stone. The inner surfaces of the specimens were then air abraded with 50 µm aluminum oxide. The castings were seated on their corresponding teeth. The 30 specimens in each group were randomly divided into 3 subgroups of 10 specimens by using a table of random number. A red vertical straight line was drawn from the casting across to the tooth with red indelible marker as a guide for placing the casting. The height of each crown was measured before cementation with a Digimatic indicator (ID-130 ME, Mitutoyo, Kawasaki, Japan) with an accuracy of ±2 µm. Three independent measurements were recorded, and the average value of each subgroup was reported in micrometers. Castings in each subgroup were luted with three cements: zinc phosphate (1Z, 2Z) (Phosphacap, Vivadent, Schaan, Liechtenstein), glass ionomer (1G, 2G) (Fuji Cap I, Shofu Dental, Tokyo, Japan), and a resin (1R, 2R) (Panavia 21, Kuraray, Osaka, Japan). The powder:liquid ratio, mixing time, and method of mixing were strictly controlled and followed the rec143
THE JOURNAL OF PROSTHETIC DENTISTRY
TUNTIPRAWON
Table II. Two-way analysis of variance for marginal seating Source of variation
Sum of squares
Degrees of freedom
Main effects 6361.100 Roughness 6.667 Cement 6354.433 Roughness-cement 695.233 Explained 7056.333
3 1 2 2 5
Mean squares
F
P
2120.367 9.995 .000 6.667 0.031 .860 3177.217 14.977 .000 347.617 1.639 .204 1411.267 6.653 .000
Table III. Two-way analysis of variance for retention Sum of squares
Degrees of freedom
Mean squares
F
P
539151.977 134426.667 404725.310 10466.514
3 1 2 2
179717.326 134426.667 202362.655 5233.257
149.051 111.489 167.833 4.340
.000 .000 .000 .018
549618.491
5
109923.698
91.167
.000
Source of variation
Fig. 2. Tensile testing using Lloyd universal testing machine.
Main effects Roughness Cement Roughnesscement Explained
test of honestly significant difference (Tukey-HSD test) were used to determine the significant differences at the 95% level of confidence.
RESULTS
Fig. 3. Mean differences (±SD) in marginal seating for each luting agent between group 1 (coarse diamond stones) and group 2 (fine diamond stones).
ommendations of the manufacturers (Table I). The same amount of cement was painted on the inner surface of each casting. The maximal finger force of the same dentist was used to seat and secure the crown until the initial setting of cement. Excess cement was then removed, and the crown height was measured at the same location. All specimens were stored in water at 37°C for 24 hours before retention testing. Each specimen was aligned in the center of the attachment on the Lloyd universal testing machine (Model LR10K, Fareham, England) (Fig. 2). A tensile load was applied with a constant crosshead speed of 2 mm/min. The maximal force used to remove the crown was recorded in Newtons. A 2-way analysis of variance (ANOVA) and Tukey’s 144
Marginal seating was determined by the differences in crown height before and after cementation. Minimal differences in crown height suggested that the crown was seated closely to the margin of a prepared tooth. Measurements between the finish line of tooth preparation and the artificial crown margin were not investigated. Mean differences and standard deviations between groups 1 and 2 are illustrated in Figure 3. Two-way ANOVA demonstrated a statistically significant difference in marginal seating among luting cements (P<.001) (Table II). No significant difference was recorded among cement-roughness combinations (P=.240) or between diamond stones (P=.860). A Tukey-HSD test revealed that the differences between Z and G in group 1 and Z and R in group 2 were not statistically significant (P>.05). Figure 4 illustrates mean differences in retention for each luting cement and diamond stone. Two-way ANOVA disclosed differences in retention (Table III) among luting cements, between diamond stones (P<.001) and between cement-stone combinations (P=.018). A Tukey-HSD test demonstrated significant differences in retention between luting cements both in groups 1 and 2 (P<.05). The significant interaction was suggested as a result of a greater difference in retention between coarse and fine diamond stones for the resinous cement than for the other 2 cements. VOLUME 81 NUMBER 2
TUNTIPRAWON
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 4. Mean differences (±SD) in retention for each luting agent between group 1 (coarse diamond stones) and group 2 (fine diamond stones).
DISCUSSION In this study, a coarse diamond stone (grit size 120 µm) created a statistically significantly more retentive tooth surface than a fine diamond stone (grit size 50 µm). Increased retention was evident when artificial crowns were luted with a cement such as zinc phosphate,34 but also with crowns luted with an adhesive cement such as glass ionomer35 and resin cement.10 The SEM confirmed that axial surfaces prepared with a coarse diamond stone (Fig. 5, A) provided longer projections than those produced with a fine diamond stone (Fig. 5, B). Differences in lengths of projects between rough and smooth surfaces ranged from 2 to 5 µm as measuring under SEM. Spaces between a number of these projections were filled with cement. Therefore it appeared that the maximal retentive force was achieved when cement had completely occupied the spaces. The greater retention of crowns from the rough surfaces may have resulted from the larger tooth-cement interlocking areas.32 Statistically significant differences in marginal seating were not revealed between tooth surfaces prepared with a coarse and a fine diamond with any of the luting agents. It is theorized that a uniform thickness of cement painted over the inner surface of the crown, efficiently flowed into the spaces between projections, and the different lengths of projections for coarse and smooth surfaces were not sufficient to affect the flow of cement and vertical crown seating. Furthermore, other factors that influenced marginal seating were also controlled; namely, size and shape of the specimen, layers of die-spacer, casting technique, manipulation of cement, same amount of painted cement, and precise finger loading. The appropriate amount of cement uniformly painted over the entire inner surface also reduced generation of hydrostatic pressure. Excessive cement may prevent an artificial crown from seating passively on the tooth. However, the resin cement (Panavia 21) resulted in the statistically poorest marginal seating, whereas glass ionomer FEBRUARY 1999
Fig. 5. Tooth surface under SEM (×10,000). A, Prepared with coarse diamond bur (grit size 120 µm). B, Prepared with fine diamond bur (grit size 50 µm).
cement (Fuji Cap I) provided the best seating both in rough and smooth tooth preparations. One explanation was that the different flow properties and film thicknesses of luting agents influenced marginal seating. Glass ionomer cement has recorded the smoothest flow rate,36,37 with a minimal film thickness of 7.24 to 20.5 µm.37 International Organization of Standardization (ISO)38 and American Dental Association (ADA)39 specifications have recommended that the greatest film thickness of zinc phosphate cement as 25 µm. However, White and Yu19 reported a thicker film thickness of 28 µm. Numerous studies have demonstrated the film thickness of resin cement was 31 µm40 to 40 to 45 µm.41 Despite their poor marginal seating, the low solubility of resinous cements in oral fluid18,32 may control the amount of microleakage that results from incomplete seating of the crown. Less microleakage has been observed for artificial crowns luted with a resin cement compared with those luted with zinc phosphate42-44 and glass ionomer cements.46 However, fluid absorption of resin cement may result in microcracks within the cement, which can reduce flexural strength.45,46 Longitudinal clinical studies are 145
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 6. Removed crown with portion of dentin attached inside.
required to evaluate this altered property of resinous cements. Crowns cemented with a resin recorded the greatest retention, whereas zinc phosphate cement had the least retention for both smooth and rough tooth preparations. In this study, both adhesive and cohesive failure of luting cements were investigated. It has been shown that generally greater tensile or shear forces are required to unseat appliances cemented with luting agents that have a high compressive strength than with cements of low compressive strength.17 Therefore properties of adhesion and compressive strength of the luting agents were the major influencing factors. The manufacturer claims that Panavia 21 cement has a shear bond strength to dentin in the range of 18 to 20 MPa after 24 hours, and also claimed a higher compressive strength with their cement (250 MPa) compared with zinc phosphate cement (130 MPa).40 Kerby47 reported a compressive strength of Panavia cement after 24 hours of only 200 MPa. The manufacturer claimed that the compressive strength of Fuji I was 225 MPa after 24 hours. It has been demonstrated that the compressive strength of glass ionomer cement was similar to dentin34; however, the compressive strength of a dental cement was critical when retention was purely mechanical interlocking48 such as zinc phosphate cement. The benefit of improved retention is less potential for dislodgment of a casting. Nevertheless, if a crown requires removal, high retentive strength may be a disadvantage. A bur is normally used to section the crown to the cemental layer to remove crowns with zinc phosphate. A rigid hand instrument is then used to spread the crown and break the cement seal. This method is not indicated for crowns cemented with resinous cements because there is a tendency to fracture the coronal tooth structure, or the core. Of the removed artificial crowns luted with resin cement in group 1, 146
TUNTIPRAWON
50% contained some detectable tooth structure within the crown (Fig. 6). Therefore it may be necessary to grind the crown down to tooth structure to remove an artificial crown luted with a resinous cement. Removal of artificial crowns constructed from base metal alloys was more arduous. Base metal alloy crowns required more chair time, cutting instruments, and perseverance from both dentist and patient. Our study suggested that coarse diamond stones were more beneficial than fine diamond stones for tooth preparation of artificial crowns, because their use resulted in greater retention and unimpaired crown seating. Furthermore, less chair time was required for tooth preparation, which is more desirable for patients. Comprehension of physical properties and disciplined manipulation of the luting cement were also essential for optimal crown seating and retention. The dentists must also record and inform patients about the luting agent used to prevent the chance of fracturing the tooth during possible subsequent crown removal.
CONCLUSIONS Tooth surfaces prepared with a coarse diamond stone (grit size 120 µm) statistically increased the retention of artificial crowns compared with tooth surfaces prepared with a fine diamond stone (grit size 50 µm). Significant differences were recorded for luting agents: zinc phosphate, glass ionomer, and resin cements. No statistically significant differences existed in the marginal seating of artificial crowns luted to rough and smooth preparations with cements in this investigation. 1. Resinous cement (Panavia 21) provided the greatest retention of crowns but poorer marginal seating was evident. 2. Glass ionomer (Fuji Cap I) cement recorded the best marginal seating for complete metal crowns. Mr Patipan Pravichpram is gratefully acknowledged for his technical assistance. I also thank Dr Root Olan and Dr Jean Wu for their comments on the manuscript.
REFERENCES 1. Tjan AH, Dunn JR, Grant BE. Marginal leakage of cast gold crowns luted with an adhesive resin cement. J Prosthet Dent 1992;67:11-5. 2. Goldman M, Laosonthorn P, White RR. Microleakage—full crowns and the dental pulp. J Endodont 1992;18:473-5. 3. Staninec M, Giles WS, Saiku JM, Hattori M. Caries penetration and cement thickness of three luting agents. Int J Prosthodont 1988;1:259-63. 4. Smith BG. The effect of the surface roughness of prepared dentin on the retention of castings. J Prosthet Dent 1970;23:187-97. 5. Shepherd WB, Leinfelder KF, Hershey HG. The effect of mixing method, slap temperature, and humidity on the properties of zinc phosphate and zinc silicophosphate cement. Angle Orthod 1978;48:219-26. 6. Bruce WL, Stevens L. Strength properties of three zinc phosphate cements mixed to two different consistencies. Aust Dent J 1989;34:132-5. 7. Zaki AF, Karl F, Leinfelder KF. Rapid mixing of zinc phosphate cement for fixed prosthodontic procedures. J Prosthet Dent 1992;67:120-5. 8. Kendzior GM, Leinfelder KF, Hershey HG. The effect of cold temperature mixing on the properties of zinc phosphate cement and zinc silicophosphate cement. Angle Orthod 1976;46:345-50.
VOLUME 81 NUMBER 2
TUNTIPRAWON
9. Pearson GJ, Atkinson AS. Effects of temperature change on the working and setting characteristics of water-based dental cements. Dent Mater 1987;3:275-9. 10. Omura I, Yamauchi J, Harada A, Wada T. Adhesive and mechanical of a new dental adhesive. J Dental Res 1984;63:223 (abstract). 11. Phillips RW, Swartz ML, Rhodes B. An evaluation of a carboxylate adhesive cement. J Am Dent Assoc 1970;81:1353-9. 12. Tjan AJ, Li T. Retention and fit of crowns cemented with an adhesive resin. J Dent Res 1990;69:123 (abstract). 13. Lee H, Swartz ML. Evaluation of a composite resin crown and bridge luting agent. J Dent Res 1972;51:756-66. 14. White SN, Sorensen JA, Kang SK, Caputo AA. Microleakage of new crown and fixed partial denture luting agents. J Prosthet Dent 1992;67:156-61. 15. Wilson AD, Crisp S, Lewis BG, McLean JW. Experimental luting agents based on the glass ionomer cements. Br Dent J 1977;142:117-22. 16. Powers JM, Johnson ZG, Craig RG. Physical and mechanical properties of zinc polycarboxylate dental cements. J Am Dent Assoc 1974;88:380-3. 17. Phillips RW. Skinner’s science of dental material. 9th ed. Philadelphia: WB Saunders; 1989. p. 479-501. 18. White SN, Kipnis V. Effect of adhesive luting agents on the marginal seating of cast restoration. J Prosthet Dent 1993;69:28-31. 19. White SN, Yu Z. Film thickness of new adhesive luting agents. J Prosthet Dent 1992;67:782-5. 20. White SN, Yu Z. The effect of the adhesive luting agent-dentinal surface interactions on film thickness. J Prosthet Dent 1992;68:49-52. 21. White SN, Yu Z, Kipnis V. The effect of seating force on film thickness of new adhesive luting agents. J Prosthet Dent 1992;68:484-9 22. Webb EL, Murray HV, Holland GA, Taylor DF. Effects of preparation relief and flow channels on seating full coverage castings during cementation. J Prosthet Dent 1983;49:777-80. 23. Kaufman EG, Coelho DH, Colin L. Factors influencing the retention of cemented gold castings. J Prosthet Dent 1961;11:487-502. 24. Vermilyea SG, Kuffler MJ, Huget EF. The effects of die relief agent on the retention of full coverage castings. J Prosthet Dent 1983;50:207-10. 25. Campagni WV, Preston JD, Reisbick MH. Measurement of paint-on die spacers used for casting relief. J Prosthet Dent 1982;47:606-11. 26. Van Northwick WT, Gettleman L. Effect of internal relief, vibration, and venting on the vertical seating of cemented crowns. J Prosthet Dent 1981;45:395-9 27. Passon C, Lambert RH, Lambert RL, Newman S. The effect of multiple layers of die-spacer on crown retention. Oper Dent 1992;17:42-9. 28. Eames WB, O’Neal SJ, Monteiro J, Miller C, Roan JD, Cohen KS. Techniques to improve the seating of castings. J Am Dent Assoc 1978;96: 432-7. 29. Assif D, Rimer Y, Aviv I. The flow of zinc phosphate cement under a fullcoverage restoration and its effect on marginal adaptation according to the location of cement application. Quintessence Int 1987;18:765-74. 30. Ayad MF, Rosenstiel SF, Salama M. Influence of tooth surface roughness and type of cement on retention of complete cast crown. J Prosthet Dent 1997;77:116-21. 31. Sturdevant CM, Roberson TM, Heymann HO, Sturdevant JR. The art and science of operative dentistry. 3rd ed. St Louis: Mosby Year Book; 1995. p. 353-7. 32. Smyd ES. Dental engineering applied to inlay and fixed bridge fabrication. J Prosthet Dent 1952;2:536-42.
FEBRUARY 1999
THE JOURNAL OF PROSTHETIC DENTISTRY
33. Darveniza M, Basford KE, Meek J, Stevens L. The effects of surface roughness and surface area on the retention of crowns luted with zinc phosphate cement. Aust Dent J 1987;32:446-57. 34. Berkson R. Dental cement: a study of its physical property of adhesion. Am J Orthod 1950;36:701-10. 35. Wilson AD, Kent BE. The glass ionomer: a new translucent dental filling material. J Appl Chem Biotech 1971;21:313-21. 36. Cook WD. Rheological studies on the setting of water-based cements. Aust Dent J 1984;29:44-9. 37. Kyrios DM, Duke ES, Windeler AS. Glass-ionomer cement film thickness and working time. J Prosthet Dent 1989;62:533-6. 38. International Organization of Standardization, Geneva. Dental waterbased cements. ISO 9917:1991(E). 39. American National Standards Institute/American Dental Association Specification No. 8 for Zinc Phosphate Cement. J Am Dent Assoc 1978; 96:121-3. 40. Staninec M, Giles WS, Saiku JM, Hattori M. Caries penetration and cement thickness of three luting agents Int J Prosthodont 1988;1:259-63. 41. Ferrari M, Cagidiaco MC, Breschi R. Evaluation of resin-bonded retainers with the scanning electron microscope. J Prosthet Dent 1988;59:160-5. 42. Johnston JF, Phillips RW, Swartz BS. A laboratory and clinical investigation of certain resin restorative and cementing materials. Part II: a twentyeight month clinical evaluation of a resin cement. J Prosthet Dent 1955;5:705-10. 43. Tjan AH, Dunn JR, Grant BE. Marginal leakage of cast gold crowns luted with an adhesive resin cement. J Prosthet Dent 1992;67:11-5. 44. Tuntiprawon M, Sermsuti-anuwat N, Promieng N. Microleakage of luting cements and light-cured composite resin. (In Thai) Chula Dent J 1996; 19:179-88. 45. Calais JG, Söderholm KJ. Influence of filler type and water exposure on flexural strength of experimental composite resins. J Dent Res 1988;67: 836-40. 46. Diaz-Arnold AM, Williams VD, Aquilino SA. The effect of film thickness on the tensile bond strength of a prosthodontic adhesive. J Prosthet Dent 1991;66:614-8. 47. Kerby RE, McGlumphy EA, Holloway JA. Some physical properties of implant abutment luting cements. Int J Prosthodont 1992;5:321-5. 48. Jørgensen KD, Holst K. The relationship between the retention of cemented veneer crowns and the crushing strength of the cement. Acta Odontol Scand 1967;25:355-9.
Reprint requests to: DR MORAKOT TUNTIPRAWON DEPARTMENT OF PROSTHODONTICS FACULTY OF DENTISTRY CHULALONGKORN UNIVERSITY HENRI DUNANT RD PRATUMWON BANGKOK 10330 THAILAND Copyright © 1999 by The Editorial Council of The Journal of Prosthetic Dentistry. 0022-3913/99/$8.00 + 0. 10/1/93805
147
Purpose. This study investigated the effect of axial surface roughness on the marginal seating and retention of silver-palladium crowns luted with zinc phosphate, glass ionomer, and resin cements. Material and methods. Coarse and fine diamond stones were used to create various surface roughnesses of premolars. A milling machine was used to control the height and angle of the axial walls of tooth preparations. Ten cast metal crowns in 6 subgroups were luted with 3 cements (Phosphacap, Fuji Cap I, and Panavia 21). Marginal seating was recorded with a Digimatic indicator. Retention was determined by measuring the tensile force required to remove a metal crown with a Lloyd testing machine. Results. Two-way analysis of variance revealed statistically significant differences (P<.001) in retention for both luting cements and surface roughness. No significant difference was recorded for marginal seating relative to roughness (P=.860) and interaction effects (P=.204). Tukey-HSD tests revealed substantial differences in retention among Phosphacap, Fuji Cap I, and Panavia 21 cements. Significant differences were not confirmed in marginal seating between Fuji Cap I and Phosphacap cements with coarse diamonds, and Phosphacap and Panavia 21 cements with fine diamonds. Conclusions. The best retention for complete metal crown was demonstrated for tooth preparations ground with coarse diamonds and cemented with Panavia 21 cement. Differences in axial surface roughness had no effect on the marginal seating of the complete metal crowns. (J Prosthet Dent 1999;81:142-7.)
CLINICAL IMPLICATIONS Teeth prepared with coarse diamond stones had greater retention than those prepared with fine diamond stones. Resin cement provided the best retention, but poorest seating for complete metal crowns.
T
he longevity of fixed prostheses depends on the retention and marginal integrity of restorations.1 Marginal leakage of cements has been documented as one of the major causes of failure because of secondary dental caries.2,3 Many factors have been demonstrated to influence the marginal seating and retention of crowns, such as the size and shape of prepared teeth,4 manipulation of cement,5-9 retentive properties of cement,10-18 cement film thickness,19,20 relieving space or venting for cement,21-28 cement application,29 and roughness of dentinal surface.30 Dental diamond grinding instruments are widely used because of their high efficiency of cutting and their versatility of shapes, dimensions, and grit sizes. Large grit diamonds have provided more rapid cutting speed and have created greater tooth surface roughness.31 The duration of tooth preparation is crucial in fixed prosthodontic procedures. A shorter chair time results in greater comfort for the patient and efficiency This research was supported by a Rachadapiseksompoj Research Fund. aAssistant Professor, Department of Prosthodontics. 142 THE JOURNAL OF PROSTHETIC DENTISTRY
for the dentist. A coarse diamond stone is more efficient in preparing tooth structure than fine diamonds, with larger dentinal projections on the axial walls. Mechanical retention relies on an interlocking area between the tooth and cement, and greater crosssectional area of a dentinal projection can result in greater retention of a casting.32 However, there have been other studies that have indicated minimal differences in retentive forces between rough and smooth tooth preparations.4,30,33 The purpose of this study was to determine the marginal seating and retention of complete cast metal crowns cemented to teeth prepared with different grit sizes of diamond stones, with 3 contemporary luting agents.
MATERIAL AND METHODS Sixty extracted, intact human premolars were selected according to crowns of similar sizes, cleaned, and stored in deionized water. Roots of the specimens were notched and embedded in acrylic resin blocks at the cementoenamel junction (CEJ). All standardized tooth preparations for complete metal crowns were perVOLUME 81 NUMBER 2
TUNTIPRAWON
THE JOURNAL OF PROSTHETIC DENTISTRY
Table I. Manipulation of luting cements Cements
Batch no.
Powder/ liquid ratio
Phosphacap
532419
32 g/16 g
Fuji Cap I
290861
Panavia 21
51181
0.22 g/0.11 g (0.09 mL) 1 turn of base and catalyst paste
Mixing method
Mixing time
Amalgamator (Silamat, Vivadent, Liechtenstein, Germany) Amalgamator
10 s
Hands
5s 20-30 s
formed with a milling machine (Milling machine, KaVo EWL, Leutkirch, Germany). The horizontal plane of the occlusal surface was initially prepared with a wheel stone, and an elliptical rubber template was made to stamp on the occlusal surface to outline the cross-sectional area. High-speed cylinder diamonds (Intensiv No. 210, Viganello-Lugano, Switzerland) were selected for preliminary preparation of the teeth until their axial walls were 0.5 mm away from the external outline. A random number table was used to divide prepared teeth into 2 groups of 30 specimens. In group 1, coarse diamond stones (grit size 120 µm, 117C, Intensiv) and in group 2, fine diamond stones (grit size 50 µm, 117GS Intensiv) were used to finish all preparations with a height of 3 mm, a 6-degree convergence angle, and a shoulder finish line (Fig. 1). Each stone was used only once per specimen. The shoulder finish line was refined for all tooth preparations with a hand instrument by 1 dentist. Specimens were ultrasonically cleaned in deionized water for 10 minutes. Surface roughness of axial walls was randomly examined with a SEM. All specimens were stored in deionized water before impressions were made. Each impression was made with a polyvinyl siloxane material (Provil M, Bayer Dental, Leverkusen, Germany) in a custom impression tray, then the dies were poured with improved stone (Silky Rock Yellow, Whip Mix, Louisville, Ky.). The external outline of the 0.5-mm shoulder finish line on dies were marked with a noncarbon red pencil. Axial surfaces of the dies were painted with 4 layers of diespacer (Nice-Fit, Shofu Dental, Kyoto, Japan) to the corner of the finish line. Gold and silver die spacer colors were alternately painted to prevent repeated applications. Wax patterns for crowns were formed with blue inlay wax (Kerr Mfg. Co, Romulus, Mich.), with a thickness of approximately 0.5 to 1 mm. A 5-mm diameter ring was waxed in the center of the occlusal surface of each pattern to facilitate connection of the crown to the tensile testing machine. Wax patterns were immediately FEBRUARY 1999
Fig. 1. Standardized preparation using milling machine (KaVo EWL).
sprued and invested in inlay investment (Cristobalite, Kerr Mfg. Co.). The powder/liquid ratio of 100 gm/40 mL was mixed for 1 minute with a vibrator. The wax patterns were vaporized at 350°C for 30 minutes, and the investments were heated at 850°C for 30 minutes. Crowns were cast in a silver-palladium alloy (Degussa AG, Palliag M, Frankfurt, Germany) with an electronic induction casting machine (Degutron, Degussa AG). Sprues were removed and castings were finished externally with stones and rubber points, respectively. Internal imperfections were removed with a small round diamond stone. The inner surfaces of the specimens were then air abraded with 50 µm aluminum oxide. The castings were seated on their corresponding teeth. The 30 specimens in each group were randomly divided into 3 subgroups of 10 specimens by using a table of random number. A red vertical straight line was drawn from the casting across to the tooth with red indelible marker as a guide for placing the casting. The height of each crown was measured before cementation with a Digimatic indicator (ID-130 ME, Mitutoyo, Kawasaki, Japan) with an accuracy of ±2 µm. Three independent measurements were recorded, and the average value of each subgroup was reported in micrometers. Castings in each subgroup were luted with three cements: zinc phosphate (1Z, 2Z) (Phosphacap, Vivadent, Schaan, Liechtenstein), glass ionomer (1G, 2G) (Fuji Cap I, Shofu Dental, Tokyo, Japan), and a resin (1R, 2R) (Panavia 21, Kuraray, Osaka, Japan). The powder:liquid ratio, mixing time, and method of mixing were strictly controlled and followed the rec143
THE JOURNAL OF PROSTHETIC DENTISTRY
TUNTIPRAWON
Table II. Two-way analysis of variance for marginal seating Source of variation
Sum of squares
Degrees of freedom
Main effects 6361.100 Roughness 6.667 Cement 6354.433 Roughness-cement 695.233 Explained 7056.333
3 1 2 2 5
Mean squares
F
P
2120.367 9.995 .000 6.667 0.031 .860 3177.217 14.977 .000 347.617 1.639 .204 1411.267 6.653 .000
Table III. Two-way analysis of variance for retention Sum of squares
Degrees of freedom
Mean squares
F
P
539151.977 134426.667 404725.310 10466.514
3 1 2 2
179717.326 134426.667 202362.655 5233.257
149.051 111.489 167.833 4.340
.000 .000 .000 .018
549618.491
5
109923.698
91.167
.000
Source of variation
Fig. 2. Tensile testing using Lloyd universal testing machine.
Main effects Roughness Cement Roughnesscement Explained
test of honestly significant difference (Tukey-HSD test) were used to determine the significant differences at the 95% level of confidence.
RESULTS
Fig. 3. Mean differences (±SD) in marginal seating for each luting agent between group 1 (coarse diamond stones) and group 2 (fine diamond stones).
ommendations of the manufacturers (Table I). The same amount of cement was painted on the inner surface of each casting. The maximal finger force of the same dentist was used to seat and secure the crown until the initial setting of cement. Excess cement was then removed, and the crown height was measured at the same location. All specimens were stored in water at 37°C for 24 hours before retention testing. Each specimen was aligned in the center of the attachment on the Lloyd universal testing machine (Model LR10K, Fareham, England) (Fig. 2). A tensile load was applied with a constant crosshead speed of 2 mm/min. The maximal force used to remove the crown was recorded in Newtons. A 2-way analysis of variance (ANOVA) and Tukey’s 144
Marginal seating was determined by the differences in crown height before and after cementation. Minimal differences in crown height suggested that the crown was seated closely to the margin of a prepared tooth. Measurements between the finish line of tooth preparation and the artificial crown margin were not investigated. Mean differences and standard deviations between groups 1 and 2 are illustrated in Figure 3. Two-way ANOVA demonstrated a statistically significant difference in marginal seating among luting cements (P<.001) (Table II). No significant difference was recorded among cement-roughness combinations (P=.240) or between diamond stones (P=.860). A Tukey-HSD test revealed that the differences between Z and G in group 1 and Z and R in group 2 were not statistically significant (P>.05). Figure 4 illustrates mean differences in retention for each luting cement and diamond stone. Two-way ANOVA disclosed differences in retention (Table III) among luting cements, between diamond stones (P<.001) and between cement-stone combinations (P=.018). A Tukey-HSD test demonstrated significant differences in retention between luting cements both in groups 1 and 2 (P<.05). The significant interaction was suggested as a result of a greater difference in retention between coarse and fine diamond stones for the resinous cement than for the other 2 cements. VOLUME 81 NUMBER 2
TUNTIPRAWON
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 4. Mean differences (±SD) in retention for each luting agent between group 1 (coarse diamond stones) and group 2 (fine diamond stones).
DISCUSSION In this study, a coarse diamond stone (grit size 120 µm) created a statistically significantly more retentive tooth surface than a fine diamond stone (grit size 50 µm). Increased retention was evident when artificial crowns were luted with a cement such as zinc phosphate,34 but also with crowns luted with an adhesive cement such as glass ionomer35 and resin cement.10 The SEM confirmed that axial surfaces prepared with a coarse diamond stone (Fig. 5, A) provided longer projections than those produced with a fine diamond stone (Fig. 5, B). Differences in lengths of projects between rough and smooth surfaces ranged from 2 to 5 µm as measuring under SEM. Spaces between a number of these projections were filled with cement. Therefore it appeared that the maximal retentive force was achieved when cement had completely occupied the spaces. The greater retention of crowns from the rough surfaces may have resulted from the larger tooth-cement interlocking areas.32 Statistically significant differences in marginal seating were not revealed between tooth surfaces prepared with a coarse and a fine diamond with any of the luting agents. It is theorized that a uniform thickness of cement painted over the inner surface of the crown, efficiently flowed into the spaces between projections, and the different lengths of projections for coarse and smooth surfaces were not sufficient to affect the flow of cement and vertical crown seating. Furthermore, other factors that influenced marginal seating were also controlled; namely, size and shape of the specimen, layers of die-spacer, casting technique, manipulation of cement, same amount of painted cement, and precise finger loading. The appropriate amount of cement uniformly painted over the entire inner surface also reduced generation of hydrostatic pressure. Excessive cement may prevent an artificial crown from seating passively on the tooth. However, the resin cement (Panavia 21) resulted in the statistically poorest marginal seating, whereas glass ionomer FEBRUARY 1999
Fig. 5. Tooth surface under SEM (×10,000). A, Prepared with coarse diamond bur (grit size 120 µm). B, Prepared with fine diamond bur (grit size 50 µm).
cement (Fuji Cap I) provided the best seating both in rough and smooth tooth preparations. One explanation was that the different flow properties and film thicknesses of luting agents influenced marginal seating. Glass ionomer cement has recorded the smoothest flow rate,36,37 with a minimal film thickness of 7.24 to 20.5 µm.37 International Organization of Standardization (ISO)38 and American Dental Association (ADA)39 specifications have recommended that the greatest film thickness of zinc phosphate cement as 25 µm. However, White and Yu19 reported a thicker film thickness of 28 µm. Numerous studies have demonstrated the film thickness of resin cement was 31 µm40 to 40 to 45 µm.41 Despite their poor marginal seating, the low solubility of resinous cements in oral fluid18,32 may control the amount of microleakage that results from incomplete seating of the crown. Less microleakage has been observed for artificial crowns luted with a resin cement compared with those luted with zinc phosphate42-44 and glass ionomer cements.46 However, fluid absorption of resin cement may result in microcracks within the cement, which can reduce flexural strength.45,46 Longitudinal clinical studies are 145
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 6. Removed crown with portion of dentin attached inside.
required to evaluate this altered property of resinous cements. Crowns cemented with a resin recorded the greatest retention, whereas zinc phosphate cement had the least retention for both smooth and rough tooth preparations. In this study, both adhesive and cohesive failure of luting cements were investigated. It has been shown that generally greater tensile or shear forces are required to unseat appliances cemented with luting agents that have a high compressive strength than with cements of low compressive strength.17 Therefore properties of adhesion and compressive strength of the luting agents were the major influencing factors. The manufacturer claims that Panavia 21 cement has a shear bond strength to dentin in the range of 18 to 20 MPa after 24 hours, and also claimed a higher compressive strength with their cement (250 MPa) compared with zinc phosphate cement (130 MPa).40 Kerby47 reported a compressive strength of Panavia cement after 24 hours of only 200 MPa. The manufacturer claimed that the compressive strength of Fuji I was 225 MPa after 24 hours. It has been demonstrated that the compressive strength of glass ionomer cement was similar to dentin34; however, the compressive strength of a dental cement was critical when retention was purely mechanical interlocking48 such as zinc phosphate cement. The benefit of improved retention is less potential for dislodgment of a casting. Nevertheless, if a crown requires removal, high retentive strength may be a disadvantage. A bur is normally used to section the crown to the cemental layer to remove crowns with zinc phosphate. A rigid hand instrument is then used to spread the crown and break the cement seal. This method is not indicated for crowns cemented with resinous cements because there is a tendency to fracture the coronal tooth structure, or the core. Of the removed artificial crowns luted with resin cement in group 1, 146
TUNTIPRAWON
50% contained some detectable tooth structure within the crown (Fig. 6). Therefore it may be necessary to grind the crown down to tooth structure to remove an artificial crown luted with a resinous cement. Removal of artificial crowns constructed from base metal alloys was more arduous. Base metal alloy crowns required more chair time, cutting instruments, and perseverance from both dentist and patient. Our study suggested that coarse diamond stones were more beneficial than fine diamond stones for tooth preparation of artificial crowns, because their use resulted in greater retention and unimpaired crown seating. Furthermore, less chair time was required for tooth preparation, which is more desirable for patients. Comprehension of physical properties and disciplined manipulation of the luting cement were also essential for optimal crown seating and retention. The dentists must also record and inform patients about the luting agent used to prevent the chance of fracturing the tooth during possible subsequent crown removal.
CONCLUSIONS Tooth surfaces prepared with a coarse diamond stone (grit size 120 µm) statistically increased the retention of artificial crowns compared with tooth surfaces prepared with a fine diamond stone (grit size 50 µm). Significant differences were recorded for luting agents: zinc phosphate, glass ionomer, and resin cements. No statistically significant differences existed in the marginal seating of artificial crowns luted to rough and smooth preparations with cements in this investigation. 1. Resinous cement (Panavia 21) provided the greatest retention of crowns but poorer marginal seating was evident. 2. Glass ionomer (Fuji Cap I) cement recorded the best marginal seating for complete metal crowns. Mr Patipan Pravichpram is gratefully acknowledged for his technical assistance. I also thank Dr Root Olan and Dr Jean Wu for their comments on the manuscript.
REFERENCES 1. Tjan AH, Dunn JR, Grant BE. Marginal leakage of cast gold crowns luted with an adhesive resin cement. J Prosthet Dent 1992;67:11-5. 2. Goldman M, Laosonthorn P, White RR. Microleakage—full crowns and the dental pulp. J Endodont 1992;18:473-5. 3. Staninec M, Giles WS, Saiku JM, Hattori M. Caries penetration and cement thickness of three luting agents. Int J Prosthodont 1988;1:259-63. 4. Smith BG. The effect of the surface roughness of prepared dentin on the retention of castings. J Prosthet Dent 1970;23:187-97. 5. Shepherd WB, Leinfelder KF, Hershey HG. The effect of mixing method, slap temperature, and humidity on the properties of zinc phosphate and zinc silicophosphate cement. Angle Orthod 1978;48:219-26. 6. Bruce WL, Stevens L. Strength properties of three zinc phosphate cements mixed to two different consistencies. Aust Dent J 1989;34:132-5. 7. Zaki AF, Karl F, Leinfelder KF. Rapid mixing of zinc phosphate cement for fixed prosthodontic procedures. J Prosthet Dent 1992;67:120-5. 8. Kendzior GM, Leinfelder KF, Hershey HG. The effect of cold temperature mixing on the properties of zinc phosphate cement and zinc silicophosphate cement. Angle Orthod 1976;46:345-50.
VOLUME 81 NUMBER 2
TUNTIPRAWON
9. Pearson GJ, Atkinson AS. Effects of temperature change on the working and setting characteristics of water-based dental cements. Dent Mater 1987;3:275-9. 10. Omura I, Yamauchi J, Harada A, Wada T. Adhesive and mechanical of a new dental adhesive. J Dental Res 1984;63:223 (abstract). 11. Phillips RW, Swartz ML, Rhodes B. An evaluation of a carboxylate adhesive cement. J Am Dent Assoc 1970;81:1353-9. 12. Tjan AJ, Li T. Retention and fit of crowns cemented with an adhesive resin. J Dent Res 1990;69:123 (abstract). 13. Lee H, Swartz ML. Evaluation of a composite resin crown and bridge luting agent. J Dent Res 1972;51:756-66. 14. White SN, Sorensen JA, Kang SK, Caputo AA. Microleakage of new crown and fixed partial denture luting agents. J Prosthet Dent 1992;67:156-61. 15. Wilson AD, Crisp S, Lewis BG, McLean JW. Experimental luting agents based on the glass ionomer cements. Br Dent J 1977;142:117-22. 16. Powers JM, Johnson ZG, Craig RG. Physical and mechanical properties of zinc polycarboxylate dental cements. J Am Dent Assoc 1974;88:380-3. 17. Phillips RW. Skinner’s science of dental material. 9th ed. Philadelphia: WB Saunders; 1989. p. 479-501. 18. White SN, Kipnis V. Effect of adhesive luting agents on the marginal seating of cast restoration. J Prosthet Dent 1993;69:28-31. 19. White SN, Yu Z. Film thickness of new adhesive luting agents. J Prosthet Dent 1992;67:782-5. 20. White SN, Yu Z. The effect of the adhesive luting agent-dentinal surface interactions on film thickness. J Prosthet Dent 1992;68:49-52. 21. White SN, Yu Z, Kipnis V. The effect of seating force on film thickness of new adhesive luting agents. J Prosthet Dent 1992;68:484-9 22. Webb EL, Murray HV, Holland GA, Taylor DF. Effects of preparation relief and flow channels on seating full coverage castings during cementation. J Prosthet Dent 1983;49:777-80. 23. Kaufman EG, Coelho DH, Colin L. Factors influencing the retention of cemented gold castings. J Prosthet Dent 1961;11:487-502. 24. Vermilyea SG, Kuffler MJ, Huget EF. The effects of die relief agent on the retention of full coverage castings. J Prosthet Dent 1983;50:207-10. 25. Campagni WV, Preston JD, Reisbick MH. Measurement of paint-on die spacers used for casting relief. J Prosthet Dent 1982;47:606-11. 26. Van Northwick WT, Gettleman L. Effect of internal relief, vibration, and venting on the vertical seating of cemented crowns. J Prosthet Dent 1981;45:395-9 27. Passon C, Lambert RH, Lambert RL, Newman S. The effect of multiple layers of die-spacer on crown retention. Oper Dent 1992;17:42-9. 28. Eames WB, O’Neal SJ, Monteiro J, Miller C, Roan JD, Cohen KS. Techniques to improve the seating of castings. J Am Dent Assoc 1978;96: 432-7. 29. Assif D, Rimer Y, Aviv I. The flow of zinc phosphate cement under a fullcoverage restoration and its effect on marginal adaptation according to the location of cement application. Quintessence Int 1987;18:765-74. 30. Ayad MF, Rosenstiel SF, Salama M. Influence of tooth surface roughness and type of cement on retention of complete cast crown. J Prosthet Dent 1997;77:116-21. 31. Sturdevant CM, Roberson TM, Heymann HO, Sturdevant JR. The art and science of operative dentistry. 3rd ed. St Louis: Mosby Year Book; 1995. p. 353-7. 32. Smyd ES. Dental engineering applied to inlay and fixed bridge fabrication. J Prosthet Dent 1952;2:536-42.
FEBRUARY 1999
THE JOURNAL OF PROSTHETIC DENTISTRY
33. Darveniza M, Basford KE, Meek J, Stevens L. The effects of surface roughness and surface area on the retention of crowns luted with zinc phosphate cement. Aust Dent J 1987;32:446-57. 34. Berkson R. Dental cement: a study of its physical property of adhesion. Am J Orthod 1950;36:701-10. 35. Wilson AD, Kent BE. The glass ionomer: a new translucent dental filling material. J Appl Chem Biotech 1971;21:313-21. 36. Cook WD. Rheological studies on the setting of water-based cements. Aust Dent J 1984;29:44-9. 37. Kyrios DM, Duke ES, Windeler AS. Glass-ionomer cement film thickness and working time. J Prosthet Dent 1989;62:533-6. 38. International Organization of Standardization, Geneva. Dental waterbased cements. ISO 9917:1991(E). 39. American National Standards Institute/American Dental Association Specification No. 8 for Zinc Phosphate Cement. J Am Dent Assoc 1978; 96:121-3. 40. Staninec M, Giles WS, Saiku JM, Hattori M. Caries penetration and cement thickness of three luting agents Int J Prosthodont 1988;1:259-63. 41. Ferrari M, Cagidiaco MC, Breschi R. Evaluation of resin-bonded retainers with the scanning electron microscope. J Prosthet Dent 1988;59:160-5. 42. Johnston JF, Phillips RW, Swartz BS. A laboratory and clinical investigation of certain resin restorative and cementing materials. Part II: a twentyeight month clinical evaluation of a resin cement. J Prosthet Dent 1955;5:705-10. 43. Tjan AH, Dunn JR, Grant BE. Marginal leakage of cast gold crowns luted with an adhesive resin cement. J Prosthet Dent 1992;67:11-5. 44. Tuntiprawon M, Sermsuti-anuwat N, Promieng N. Microleakage of luting cements and light-cured composite resin. (In Thai) Chula Dent J 1996; 19:179-88. 45. Calais JG, Söderholm KJ. Influence of filler type and water exposure on flexural strength of experimental composite resins. J Dent Res 1988;67: 836-40. 46. Diaz-Arnold AM, Williams VD, Aquilino SA. The effect of film thickness on the tensile bond strength of a prosthodontic adhesive. J Prosthet Dent 1991;66:614-8. 47. Kerby RE, McGlumphy EA, Holloway JA. Some physical properties of implant abutment luting cements. Int J Prosthodont 1992;5:321-5. 48. Jørgensen KD, Holst K. The relationship between the retention of cemented veneer crowns and the crushing strength of the cement. Acta Odontol Scand 1967;25:355-9.
Reprint requests to: DR MORAKOT TUNTIPRAWON DEPARTMENT OF PROSTHODONTICS FACULTY OF DENTISTRY CHULALONGKORN UNIVERSITY HENRI DUNANT RD PRATUMWON BANGKOK 10330 THAILAND Copyright © 1999 by The Editorial Council of The Journal of Prosthetic Dentistry. 0022-3913/99/$8.00 + 0. 10/1/93805
147