Comparison of titanium and cobalt-chromium removable partial denture clasps

Comparison of titanium and cobalt-chromium removable partial denture clasps Jay 1. Bridgeman, a Victoria A. Marker, MS, PhD, b Susan K. Hummel, DDS, c Byron W. Benson, DDS, MS, d and Larry L. Pace, DDS e

Baylor College of Dentistry, Dallas, Texas Statement o f problem. The use of titanium alloys for removable partial dentures is an increasingly popular application. While the flexibility of titanium would allow for cast clasps to be placed in deeper undercuts than advisable with cobalt-chromium, it is possible that the retentive forces of the titanium clasp might not maintain sufficient retention after repeated flexing of the clasp arm during insertion and removal of the partial denture. Purpose. This study assessed the characteristics of cast clasps made of titanium and titanium alloys to determine whether these materials are suitable alternatives for removable partial denture applications. M a t e r i a l and methods. Removable partial denture clasps at two undercut depths were fabricated from commercially pure titanium, titanium alloy (Ti-6A1-4V), and cobalt-chromium. Loss of retention force was measured as the clasps underwent 3 years of simulated cfinical use. The data were subjected to ANOVA and Scheffd's tests to determine differences. Evidence of casting defects and porosity was evaluated by radiographic examination and nonparametric statistics. SEM microscopy was used to observe surface characteristics that were described qualitatively. Results. For the 0.75 m m undercut specimens, there was less loss of retention for clasps made from pure titanium and titanium alloy than for cobalt-chromium clasps. Porosity was more apparent in the pure titanium and titanium alloy clasps than in those made from cobalt-chromium, but the amount of porosity did not correspond to evidence of fractures or permanent deformation. Conclusions. The long-term retentive resiliency of the pure titanium and titanium alloy clasps suggests that these materials are suitable for removable partial dentures. (J Prosthet Dent 1997;78:187-93.)

I n t e r e s t in titanium for dental prostheses increased in the early 1980s. 1 A m o n g the advantages o f t i t a n i u m are its biocompatibility and strength, 2,3 m a k i n g it an ext r e m e l y suitable material for dental implants. Casting difficulties have presented obstades for fixed prosthodontics, 4 however, there have been a n u m b e r o f studies on the use o f titanium (Ti) and Ti alloys for removable prosthodontics, s,6 Several c o m m e r c i a l laboratories are currently fabricating r e m o v a b l e partial d e n t u r e ( R P D ) f r a m e w o r k s f r o m t i t a n i u m and t i t a n i u m alloys.

This study was supported by Baylor College of Dentistry Student Research Funds. aFourth Year Dental Student. bAssociate Professor, Department of Biomaterials Science. CAssistant Professor, Department of Restorative Sciences. ~Associate Professor, Department of Diagnostic Sciences. eClinica[ Assistant Professor, Department of Restorative Sciences. AUGUST 1997

T i t a n i u m has a m o d u l u s o f elasticity that is lower than that for c o b a l t - c h r o m i u m , which increases its resiliency and makes it m o r e like gold alloys. 7 This p r o p e r t y w o u l d allow for the retentive clasp arms o f R P D s to be placed in deeper undercuts on a b u t m e n t s than are possible with c o b a l t - c h r o m i u m ( C o - C r ) , because titanium is less stiff than C o - C r alloy. This characteristic is useful in clinical situations w h e r e esthetics or p e r i o d o n t a l health is a prim a r y concern. T h e flexibility o f a clasp a r m affects the retention and function o f an R P D . 8 I f a material is t o o flexible, the clasp m a y n o t provide adequate r e t e n t i o n for the R P D w h e n the f r a m e w o r k design is based o n principles used for C o - C r alloys. VandenBrink et al. s c o m p a r e d various R P D clasp materials and fabrication p r o c e d u r e s including a nickel-titanium (Ni-Ti) alloy. This alloy was f o u n d to be unacceptable for an R P D clasp even w h e n a 0.8 m m d i a m e t e r was used. O n e o u t c o m e o f this study was THE jOURNAL OF PROSTHETIC DENTISTRY

187

THE JOURNAL OF PROSTHETIC DENTISTRY

BRIDGEMAN ET AL

Table I. Summary of retentive fracture and porosity data Type of clasp

Number of fractured clasps; both undercuts*

Number of permanently deformed clasps; both undercuts*

Number of claspswith small porosity; rate 2 and 3**

Number of clasps with large porosity; rate 4**

cpTi

1

2

7

2

Ti alloy

0

2

6

2

Co-Cr

2

0

0

0

*Both undercuts: 0.25 and 0.75 mm. **Rate 1 : No porosity; Rate 2: single small (< 0.5 mm) porosity or radiolucent area on one arm of the clasp; Rate 3: a single small (< 0.5 mm) porosity or radiolucent area on both arms of the clasp; Rate 4: a large porosity (> 0.5 mm) or multiple (more than two) small porosities.

the need t o compare materials in a curved clasp configuration rather than as straight specimens. Metals and alloys are known to undergo deformation and fatigue when exposed to repeated stress, 9 which is exactly the condition o f RPD clasps. Although extensive work has been performed to determine the properties o f a variety o f materials used for removable partial denture clasps, 1°-12little is known about how titanium functions in this application. 5 To assess the suitability o f titanium and titanium alloys for RPD applications, one must demonstrate that it is possible to cast these metals with minimal defects. There are several obstacles to casting titanium and titanium alloys. Special equipment is needed because of its high melting temperature (more than 1600 ° C). This high temperature is achieved with an electric arc, and the metal must be melted under argon gas because oxygen will cause embrittlement of the casting surface.3.* Another problem is that the low density of Ti and Ti alloys makes it difficult to force the metal into the mold, and thin areas such as the clasp arms o f the RPD frameworks are prone to porosity. Investigators and laboratory technicians have been working to overcome these problems with apparent success. Blackman et al. 4 measured the dimensional changes o f titanium and other alloys cast as RPD frameworks. They found that the response o f the titanium was within the range o f dimensional changes accepted for currently used base metals of the nickel-chromium variety. They concluded that more study o f the performance of titanium under conditions o f actual use was needed. This study was conducted to determine whether: (1) Ti and Ti alloy clasps provided sufficient retentive force after 3 years o f simulated clinical use, and (2) changes in the retentive forces were related to porosity, casting defects, or other metal degradation. MATERIAL

AND METHODS

Three materials were used in this study: a commercially pure titanium (cpTi), a titanium alloy (Ti-6Al-4V) (Bio-Ti, J e n e r i c / P e n t r o n , Wallingford, Conn.), and cobalt-chromium (Co-Cr) alloy (Austenal, lnc., Chicago, 111.) (Table I). The Co-Cr clasps served as the controis because this material is a standard alloy used to manufacture RPD frameworks. All castings were made by a commercial laboratory (Lord's Dental Studio, Inc., 188

Green Bay, Wis.). To ensure consistency, all clasps were waxed and cast by one laboratory technician. The wax patterns for the titanium clasps were invested in Tyvest (Jeneric/Pentron), whereas V R investment (Austenal) was used for the clasps cast in Co-Cr. The burn-out time for the Ti metals was 12 hours. A Tycast 3000 casting machine (Jeneric/Pentron) was used to cast both the cpTi and the Ti-6A1-4V alloy at 1648 ° C. For the CoCr clasps, the burn-out time was 2.5 hours, and the cases were cast in a H o w m e d i c a ECMS casting machine (Austenal, Inc.) at 1565 ° C. The project was designed to compare clasps made to function in two different undercuts. The 0.25 mm undercut was selected because it represents a c o m m o n clinical condition for clasps made as an integral part o f CoCr RPD frameworks. The second undercut condition (0.75 ram) was chosen because it represents a difficult clinical problem. With the increased demand for esthetics, more patients are demanding that the dentist "hide" the clasps o f their RPD by placing them closer to the gingiva where the undercuts tend to be larger. The stiffness o f Co-Cr clasps makes them unsuitable to be placed in large (0.75 mm) undercuts, because the clasps would place unacceptably large stresses on the abutments. To make a "hidden" clasp for a deep (0.75 ram) undercut, flexible stainless steel wire must be soldered to the CoCr RPD framework. Fabrication o f the framework out o f flexible Ti or Ti alloys is a possible alternative to soldering stainless steel clasps on Co-Cr frameworks. Thus the 0.75 m m undercut situation was included to test the retention o f the Ti clasps. Co-Cr clasps in a 0.75 mm undercut were made for comparison, even t h o u g h they are not clinically relevant. Six RPD circumferential clasps o f each material were prepared for measured undercuts o f 0.25 and 0.75 mm. All the clasps were prepared to fit one o f three simulated abutment teeth fabricated in Co-Cr. The design of the clasps was modified slightly to standardize the path o f insertion and removal o f clasps. Each clasp was cast with a mesial rest, mesial guide plane, lingual bracing arm, buccal retentive arm, and distal guide plane. The clasps were polished and fitted to their respective dies after casting. The finishing procedure was performed by one laboratory technician who used silicon carbide stones, carbide burs, rubber wheels, and polishing cornV O L U M E 78

NUMBER2

BRIDGEMAN

ET AL

THE J O U R N A L O F PROSTHETIC DENTISTRY

700

700

Z

Z v

600

o

500

Ti6AI4V cpTi --¢---

-Cr

o O

Ti6AI4V 600 500

IJ,.

400

400

¢

300

300

n,'

200

m

100 -

<~

0-

w'

c= too O <~

6

7

8

9

10

Cycling Trials

Fig. 1. Comparison of D values for clasps made for 0.25 mm undercuts. ANOVA showed no significant differences in D values from all three materials (p > 0.05). Standard deviation bars on mean values for the Co-Cr clasps are representative of standard deviation observed for three materials.

pound. O n receipt from the laboratory, the clasps were inspected and adjusted to fit by a clinician as in a normal clinical situation. Dies representing the abutments were fabricated o f metal so that they would not wear during testing procedures. The dies consisted o f a mandibular first molar with a mesial and distal guide plane, a mesial occlusal rest, and either 0.25 or 0.75 mm undercuts on the distobuccal cusp. A distal guide plane was added to the clasp to ensure a straight path o f insertion and withdrawal. The dies were also modified by cutting a mesial occlusal rest 2 mm deep that served as a stop for insertion and as an additional way to ensure a straight line o f withdrawal. The force (mN) needed to remove the fitted clasp was measured with a universal testing machine (model 1250, Instron ,Corp., Canton, Mass.), with a crosshead speed o f 10 m m / m i n u t e . After the measurement o f the retentive force o f the fitted clasps, the clasps were cycled on and off the dies 500 times to simulate insertion and removal o f the RPD. After this simulated clinical use, the force o f removal was remeasured to determine the reduction in the a m o u n t o f retentiveness remaining. Then, each clasp was readapted to its respective die with contouring pliers. This procedure simulated "adjusting" a clasp o f a partial denture. This cycle was repeated 10 times to simulate 3 years o f clinical u s e . 12 A comparison o f the force measured when the clasp was initially fitted and the force after 500 cycles was made. This change in retention was calculated by subtracting the final retentive force from the initial force to determine the delta value. Each clasp acted as its own control, so there was no need to standardize the initial force values. The changes in retentiveness in the clasps over the entire clinical simulation were monitored. The calculated results were analyzed with analysis o f variance (ANOVA) and Schefft's tests. AUGUST 1997

200

0

6

7

8

9

10

Cycling Trials

Fig. 2. Comparison of D values of clasps made for 0.75 mm undercuts. Statistical analysis (ANOVA and Scheff6's tests, p < 0.05) indicated there was significantly greater loss of retention in Co-Cr clasps versus cpTi clasps after cycles 3, 4, and 5. For all five cycles, clasps made with Ti-6AI-4V maintained significantly lower D value than Co-Cr clasps. Standard deviation bars on mean values for Ti-6AI-4V clasps are representative of standard deviation observed for three materials.

To evaluate casting defects and to detect porosity, each clasp was radiographed with size 2 film at 10 cm, with settings o f 70 kVp, 10 mA, for five impulses. The method is similar to that used by Wang and Boyle.l~ to assess casting efficiency. The processed films were examined to detect porosities in the clasps. The amount of porosity in each clasp was rated according to the following scale: (1) no porosity; (2) a single small (<0.5 mm) porosity or radiolucent area on one arm o f the clasp; (3) a single small (<0.5 mm) porosity or radiolucent area on both arms o f the clasp; and (4) a large porosity (>0.05 mm) or multiple (more than two) small porosities. To ensure that all defects were assessed, the films were examined for evidence o f a change in the density o f the metal that would appear as an intensity difference. A stepped aluminum standard was imaged with each clasp to help detect differences in density or thicknesses o f the metal. The a m o u n t o f porosity in each set o f clasps was recorded and analyzed by nonparametric statistical tests (Kruskal-Wallis and Mann-Whitney U) to determine differences in the a m o u n t o f porosity in the respective groups. After these tests were completed, a scanning electron microscope (SEM) (JSM-35CF, JEOL, Canton, Mass.) was used to detect evidence o f metal fatigue on the clasps, especially in the area o f the minor connector. Cracks and other signs o f deformation were recorded with SEM micrographs. The internal surface o f the clasp arm was examined for wear. SEM micrographs were taken o f the observable evidence of porosity and surface defects; these micrographs were evaluated qualitatively. 189

THE JOURNAL OF PROSTHETIC DENTISTRY

A

Fig. 3. Radiographs show (a) small porosity, and (b) large porosity.

RESULTS The initial and readjusted retentive forces of the clasps made for the 0.25 mm undercut dies ranged from 191 to 555 mN. Both the lowest and highest values were observed in the clasps made with cpTi alloy. The statistical analysis (ANOVA and Scheffd's test) indicated that there was no significant difference in the delta values for the change of retention after simulated clinical use for cpTi, Ti alloy, and Co-Cr clasps made for 0.25 mm undercuts. Figure i illustrates the fluctuation in delta values for the last five cycles of testing. As was expected, the Co-Cr clasps exhibited greater retentive forces (ranging from 600 to 891 mN) when placed in the 0.75 mm undercut. The high force range was undoubtedly due to the stiffness of Co-Cr clasps and would probably be unacceptable for clinical conditions. These values were significantly greater than the retentive forces measured for the clasps made with the titanium materials. The delta values reflecting the change in retention force exhibited two patterns of change over the simulated 3-year clinical life in the specimens made for the 190

BRIDGEMAN ET AL

Fig. 4. SEM micrograph of fractured surface of Co-Cr clasp arm.

Fig. 5. SEM micrograph of porosity on arm of titanium alloy clasp.

0.75 mm undercuts (Fig. 2). The Co-Cr clasps exhibited an increase in retentiveness, followed by a sharp decrease in cycles 9 and 10. Commercially pure titanium clasps produced similar patterns of retained retentiveness. The measured change in retentive force for clasps made from the titanium alloy showed little variance as a function of cyclic fatigue. After the 3-year clinical simulation, there was evidence of deformation in the clasps made from all materials tested (Table I). Fractures occurred in one of the cpTi clasps and two of the Co-Cr clasps. Two clasps made from cpTi and two Ti-6AI-4V clasps were permanently deformed and could not be readjusted to produce a clinically relevant retentive force. These seven "failed" clasps represented i9% of the total clasps and occurred randomly among the clasps made for both 0.25 and 0.75 mm undercuts. All the titanium clasps contained about the same amount of porosity. The radiographs in Figure 3 depict the typical porosity seen in cpTi and Ti alloys. One group of clasps made from the cpTi for the 0.25 mm undercut exhibited a significantly greater amount of VOLUME 78

NUMBER2

B R I D G E M A N ET AL

THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 6. SEM micrograph shows evidence of surface cracking on cpTi clasp near base of arm.

porosity than the clasps made from the same material but fit to the 0.75 m m undercut. However, this result cannot be attributed to differences in the clasp design because the same arm length and thickness were used for both the 0.25 and 0.75 m m u n d e r c u t clasps. Radiographic examination revealed that there was significantly more porosity in the Ti and Ti alloy clasps versus the Co-Cr clasps (Table I). Despite this lack o f radiographic evidence showing porosity, two Co-Cr clasps fractured. Figure 4, which is an SEM micrograph o f the fracture surface o f a failed Co-Cr clasp, shows no evidence o f porosity in this fractured surface. Porosity was evident on the surface o f Ti-6M-4V clasps during SEM examination (Fig. 5). For the cpTi clasps, both the deformed and retentive clasps exhibited evidence o f surface cracking, which typically appeared at the base o f the clasp arm (Fig. 6) and along the length o f the tissue surface (Fig. 7). SEM examination revealed that none o f the clasps exhibited any evidence o f wear. To determine to what extent the cracks may have influenced retentive forces or permanent deformation, several clasps were sectioned to determine the depth o f the cracks. In these sectioned specimens, the cracks failed to penetrate beyond the surface layer (< 100 }am) (Fig. 8). Because these cracks were limited to the surface, it is unlikely that they affected the permanent deformation o f the clasps.

DISCUSSION One o f the questions about the use o f Ti or Ti alloys for RPD clasps was thc concern that these flexible materials would not maintain sufficient retention. 8 In the specimens with the 0.25 m m undercuts, there was little fluctuation in retentive forces as a function o f the number o f cycles for the clasps made with any o f the alloys (Fig. 1). This similarity in the retention force delta values for all of the tested materials indicates that the clasps AUGUST 1997

Fig. 7. SEM micrograph shows evidence of surface cracking on cpTi clasp near tip of arm; (A) 100x magnification; (B) at higher magnification (400x).

made from Ti and Ti alloy retained sufficient strength to function as clasps at this undercut. The fact that cycling clasps on a die with a 0.25 mm undercut had less effect on the retention forces compared with cycling specimens made for the 0.75 mm undercut can be attributed to the difference in the amount o f strain the metal underwent as it was removed from the undercut. 9 One might expect the clasps with the 0.75 m m undercut to have exhibited more permanent deformation than the clasps for the 0.25 mm undercut. The findings in this study did not support this theory because the nonretentive clasps included specimens made for both undercuts. The different profiles o f the curves in Figures 1 and 2 may reflect the difference in the magnitude o f stress a clasp experiences when removed from a 0.25 mm versus a 0.75 m m undercut. In Figure 2, the increase in the delta values may be due to work hardening o f the metal with the decrease in the delta values indicating a loss o f retentive forces. This phenomena occurred as the material became less able to maintain constant retention. The magnitude o f these retentive changes was significantly greater for Co-Cr clasps than for the cpTi clasps. This is 191

THE JOURNAL OF PROSTHETIC DENTISTRY

BRIDGEMAN ET AL

one Ti clasp fractured. Another factor that could account for the permanent deformation and failure of clasps made from all the materials tested is the clasps were designed to ensure a straight path during cycling, but it was impossible to eliminate all possibilities of torquing, because the clasps were inserted and removed by hand. Any excess torquing may have influenced the outcome of particular clasps. However, these insertion and removal motions would produce forces similar to what a clasp undergoes when an RPD is removed by a patient. Thus the testing m e t h o d was deemed clinically relevant. CONCLUSIONS

Fig. 8. SEM micrograph of backscattered image shows surface cracks of sectioned cpTi clasp arm; (A) 100x magnification; (B) at higher magnification (1000x).

expected because the 0.75 mm undercut causes the CoCr clasp to be flexed more than is appropriate for this stiff material. The Ti alloy showed little change in the retentive force as a function of cyclic fatigue. This may indicate that the alloy is less susceptible to work hardening. Despite some fluctuations in the delta values, the cpTi and the Ti alloy clasps maintained a high degree of retention and showed less fatigue than the Co-Cr clasps at the 0.75 mm undercut. Thus, for esthetic application of RPDs with "hidden" clasps, the Ti alloy seemed to be a suitable material. The concerns about the difficulty of casting Ti metal and Ti alloys suggest that casting defects might be responsible for clasp failures. 4 Thus it was theoretically possible that porosity defects could have contributed to material degradation, which led to the loss o f retention that occurred in some clasps. However, examination of all the permanently deformed clasps did not reveal consistent evidence of cracks or porosity that would explain the failures. Even in the absence of detectable casting defects, two of the clasps made out of Co-Cr for the 0.25 mm undercut exhibited catastrophic failure, whereas only 192

From the findings of this study, the following conclusions were drawn. 1. The Ti-6AI-4V clasps for a 0.75 mm undercut showed the least m o u n t of work-hardening and permanent deformation, as the small change in retention of these clasps was consistent throughout the simulated 3year clinical usage. 2. When comparing the clasps made for the 0.75 mm undercuts, the overall loss of retention for both Ti and Ti-6A1-4V clasps was less than for the Co-Cr clasps. 3. For all three materials tested, clasps made for the 0.25 mm undercut exhibited similar changes in retention. 4. SEM examination of cross sections of Ti and Ti-6AI-4V clasps revealed that cracking was confined to the surface layer (< 100 pro) and thus not likely to be the cause of permanent deformation. 5. Several failed clasps showed no evidence ofporosivy or surface cracking that could have contributed to failure. We thank Mr. Reynolds Challoner, President of Lord's Dental Studio, Inc., for contributing the clasps and Mr. Kris Van at Lord's Dental Studio, Inc., for preparing all the specimens. We also want to acknowledge Mr. Phil Ford for his assistance to the project.

REFERENCES 1. Okabe T, Hero H. The use of titaniurn in dentistry. Cells Mater 1995;5:21130. 2. Barice W]. Titanium net shape technologies. Warrendale: The Metallurgical Society of AIME; 1984. p. 179-90. 3. Moser JB, Lin JH, Taira M, Greener EH. Development of dental Pd-Ti-alIoys. Dent Mater 1985;1:37-40. 4. Blackman R, Barghi N, Tran C. Dimensional changes in casting titanium removable partial denture frameworks. J Prosthet Dent 1991; 65:30915. 5. VandenBrink ]P, WoJfaardt JF, Fau[kner MG. A comparison of various removable partial denture clasp materials and fabrication procedures for placing clasps on canine and premolar teeth. J Prosthet Dent 1993;70:1808. 6. Wang RR, Fenton A. Titanium for prosthodontic applications: a review of the literature. Quintessence Int 1996;27:401-8. 7. Phillips RW. Science of dental materials. 9th ed. Philadelphia: WB Saunders; 1991. 8. Stratton R], Wiebelt FJ. Retention and retainers. In: An atlas of removable partial denture design. Chicago: Quintessence; 1988. p. 45-72. 9. Craig RJ. Mechanical properties. In: Craig RJ, editor. Restorative dental materials. 8th ed. St Louis: Mosby; 1989. p. 65-112.

VOLUME 78

NUMBER2

BRIDGEMAN ET AL

THE JOURNAL OF PROSTHETIC DENTISTRY

10. Brudvik JS, Morris HF. Stress relaxation testing. Part Ilk Influence of wire alloys, gauges, and lengths on clasp behavior. J Prosthet Dent 1981;46:3749. 11. Morris HF, Asgar K, Brudvik JS, Winkler S, Roberts EP. Stress-relaxation testing. Part IV: Clasp pattern dimensions and their influence on clasp behavior. J Prosthet Dent 1983;50;319-26. 12. Matheson GR, Brudvik JS, Nicholls JI. Behavior of wrought wire clasps after repeated permanent deformation. J Prosthet Dent 1986;55:226-31. 13. Wang RR, Boyle AM. A simple method for inspection of porosity in titanium castings. J Prosthet Dent 1993;70:275-6.

Reprint requests to: DR. VICTORIAA. MARKER DEPARTMENTOF BIOMATERIALSSCIfiNCE

THETEXASA & M UNIVERSITYSYSTEM BAYLOR COLLEGEOF DENTISTRY

PO Box 660677 3302 GASTONAVE. DALLAS, TX 75266"0677

Copyright © 1997 by The Editorial Council of The Journal of Prosthetic Dentistry. 0022-3913/97/$5.00 + 0. 10/1/83125

Effects o f fabrication, finishing, and polishing procedures on preload in prostheses using convention "gold" and plastic cylinders Carr AB, Brunski JB, Hurley E. IntJ Oral Maxillofac Im-

plants 1996;11:589-98. Purpose. Implant prosthesis retaining screw fracture or loosening is cited as the most common complication for screw-retained prostheses. Preload is the force that is created by tension within the prosthetic retaining screw and the opposite clamping load at the interface of the prosthetic cylinder and the implant abutment. An appropriate preload, considering the inherent strength of the material making up the components, will resist external forces applied to the prosthesis. This study compared preload development with as-received gold cylinders, cast-to-gold cylinders, cylinders cast from plastic patterns (plastic cylinders), and cylinders in which the screw seat had been finished and polished following casting procedures. Material and Methods. Plastic patterns (3I, Implant Innovations Inc., West Palm Beach, Fla.) and two types of gold cylinders (3I and Noble Biocare USA, Chicago, Ill.) were used because of their compatibility with the Noble Biocare implant abutment. Castings were made to the two machined gold cylinders and of the plastic cylinder. Jelenko No. 7 (Jelenko, Armonk, N.Y.) and IS-85 (Williams Dental, Ivoclar North America, Amherst, N.Y.), alloys were used for casting, since these represented low and high fusing noble alloys. Strain gauges were connected to the abutment for preload determination. Preload was determined in the as-received, cast to, cast, and finished and polished states. A 10 Ncm torque was applied to the retaining screw for preload determination; one screw was used for each specimen and each specimen was fastened 15 times. After testing of the cast-to and as-cast cylinders, the screw seat was finished and polished before retesting ofpreload. Results. Significantly higher preload is seen with as-received cylinders than with any other cylinder. Higher preload is seen with metal cylinders than with plastic cylinders. Plastic cylinders produced higher preload with low fusing alloy than with high fusing alloy. Finishing and polishing of plastic cylinders improved preload but did not achieve the level ofpreload seen with metal cylinders. Metal cylinders did not demonstrate consistent improvement with finishing and polishing. Preload with Noble Biocare metal cylinders was greater than with 3I metal cylinders after the alloy had been cast to these cylinders. Conclusions. Suboptimal preload may place a joint at a higher risk of loosening when it is placed under functional load. Numerous factors are related to the reduction ofpreload. Efforts to improve preload are encouraged especially when plastic cylinders are used. 20 references.--SE Eckert

AUGUST 1997

193