COMMERCIALLY AVAILABLE SYSTEMS

TOP ABSTRACT COMMERCIALLY AVAILABLE SYSTEMS SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Commercially available systems for the laboratory fabrication of fixed prostheses include Targis/Vectris (Ivoclar Vivadent Inc., Schaan, Liechtenstein), everStick (Stick Tech Ltd., Turku, Finland) and Sculpture/FibreKor (Pentron Laboratory Technologies LLC, Wallingford, Conn.). These materials are composed of glass fibers that are preimpregnated with a resin matrix, resulting in strength properties comparable to those of alloys.¹ Nonimpregnated materials that are commercially available for the fabrication of the substructure include polyethylene weaves such as Ribbond (Ribbond, Seattle) and Connect (KerrLab, Orange, Calif.), and glass weaves such as GlasSpan (GlasSpan Inc., Exton, Pa.). These products can be used in the dental laboratory or dental operatory.

FRC prostheses can be made with either full-coverage (extracoronal) or partial-coverage (intracoronal) retainers, or a combination of the two designs. The choice of retainer design can be based on the condition and extent of the restorations of the abutment teeth. The FRC intracoronal bridge allows for a conservative tooth preparation design when the abutment teeth are unrestored or have modest intracoronal restorations. The etched metal Maryland bridge is the only conservative, nonsurgical fixed treatment alternative, but it has demonstrated problems with debonding, graying of abutment teeth due to metal show-through and overcontoured retainer components.^2–5

In vitro studies. During the past decade, a number of in vitro studies have described the mechanical properties of a variety of fiber-reinforced composite materials used to make fixed prostheses.^6–12 Laboratory studies provide important flexure property data through the use of standard three-point loading tests of rectangular bars or load testing of simulated prostheses. These data can provide insight into the relative strength and stiffness of FRC substructure materials or the relative load that can be borne by prostheses composed of a variety of components; however, they are not likely to predict clinical performance.

For FRC prostheses, clinical performance is likely to be a function of substructure design. Given the current limitations of the in vitro models, the optimization of FRC substructure design cannot be reliably determined through laboratory tests. Fixed prostheses are subjected to complex loading conditions in the mouth, which are difficult to simulate in laboratory testing models; consequently, clinical studies are required to illustrate the actual clinical performance of experimental prostheses.

Few published articles have described the clinical performance of metal-free FRC prostheses. Overall concerns regarding these prostheses include their survival time when subjected to moisture and to repeated loads in the intraoral environment. Specific objective clinical parameters that need to be evaluated are the loss of structural and marginal integrity, resistance to occlusal wear and color stability.

Clinical studies. In the early 1990s, Altieri and colleagues¹³ conducted a pilot study to evaluate single-tooth replacement fixed prostheses composed of acrylic denture teeth and a substructure of preimpregnated, unidirectional glass fiber/polycarbonate matrix FRC, with buccally and lingually placed wing-shaped retainers supported by unprepared abutment teeth. This small study and subsequent follow-up study demonstrated that this type of FRC, despite bonding problems associated with the thermoplastic poly-carbonate matrix, exhibited no mechanical, catastrophic failures of the substructure.

More recently, Göhring and colleagues¹⁴ and Vallittu and Sevelius¹⁵ reported the preliminary results of clinical studies of fixed FRC prostheses. In both studies, prostheses were made with a preimpregnated, unidirectional FRC. In the Göhring study, all prostheses had inlay (intracoronal) retainers and were made with a modified manufacturing technique that differed from the manufacturer’s recommendations; at two years, four of the 25 prostheses examined needed to be replaced. Vallittu and Sevelius¹⁵ also studied prostheses with a variety of predominantly partial-coverage retainer configurations and found that 93 percent of all prostheses were successful for up to two years.

Full-coverage–retainer FRC prostheses may have great value in circumstances in which a bonded, metal-free prosthesis is desired for single-tooth replacement adjacent to heavily restored abutment teeth. The clinical performance of full-coverage–retainer FRC prostheses, therefore, needs to be assessed, preferably in the context of the performance of partial-coverage FRC prostheses, which have already undergone preliminary evaluation.

The purpose of this clinical study was to compare the clinical performance of FRC bridges of extracoronal, full-coverage design with that of FRC bridges of intracoronal, partial-coverage design. The relationship between substructure design and clinical performance also was evaluated.

	SUBJECTS, MATERIALS AND METHODS

TOP ABSTRACT COMMERCIALLY AVAILABLE SYSTEMS SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The subject population for this study was composed of patients seeking care at the dental clinics of the University of Connecticut School of Dental Medicine, Farmington. Subjects who had heavily restored abutment teeth (large Class II, III or V amalgam or composite restorations or cast crowns) were assigned to the extracoronal, full-coverage design group. Subjects who had unrestored or minimally restored abutment teeth (small Class I, II and III amalgam or composite restorations) were assigned to the intracoronal, partial-coverage group. Patients with a history of severe parafunctional habits were not excluded from this study, and two such subjects were enrolled.

We made 39 light and heat–polymerized fixed bridges with a substructure of preimpregnated, unidirectional FRC (FibreKor) and veneered with a particulate composite (Sculpture). The fabrication of these prostheses followed the laboratory protocol described in detail elsewhere.^1,16,17 This fabrication protocol includes the use of a hand-placement method that allows the maintenance of the air-inhibited surface between layers of the FRC substructure and between the external surface of this substructure and the particulate composite veneer (the air-inhibited surface allows chemical bonding of additional composite materials).

These fixed-bridge prostheses were made with a single pontic and were placed in 25 patients over 37 months. The study included 22 extracoronal, full-coverage retainer prostheses and 17 intracoronal, partial-coverage retainer pros-theses, which allowed for a comparison of retainer configuration. Five of the study prostheses were anterior, 28 were posterior and six were a combination of both.

Tooth preparation changes. As the study progressed, we discovered the need to make changes to the tooth preparation and substructure designs of the prostheses. For the posterior extracoronal preparation, we added an occlusal groove to the preparation design to ensure adequate space for the FRC beneath the occlusal surface. For the anterior extracoronal preparation, a lingual step was added. Without this step, the technician had difficulty placing the strip of FRC without creating an overcontoured lingual surface. Figures 1 and 2 are drawings of the tooth preparation designs that evolved from this study.^1,17

View larger version (51K): [in this window] [in a new window]   Figure 1. Schematic drawing of anterior abutment tooth preparation for the extracoronal prosthesis (to be fabricated of Sculpture/FibreKor, Pentron Laboratory Technologies LLC, Wallingford, Conn.). mm: millimeters.

View larger version (44K): [in this window] [in a new window]   Figure 2. Schematic drawing of posterior abutment tooth preparation for the extracoronal prosthesis (to be fabricated of Sculpture/FibreKor, Pentron Laboratory Technologies LLC, Wallingford, Conn.). mm: millimeters.

View larger version (70K): [in this window] [in a new window]   Figure 3. Low-volume substructure design for fiber-reinforced composite prosthesis.

View larger version (80K): [in this window] [in a new window]   Figure 4. High-volume substructure design for fiber-reinforced composite prosthesis.

The most recent evolution of this design now includes a circumferential vertical wrap of the final FRC strip in the pontic area, providing an outer layer of fibers that is perpendicular to the majority of the pontic fibers. These changes in substructure design features allowed us to study the relationship between substructure design and clinical performance. When a prosthesis failed, a new prosthesis was made for the subject with a high-volume substructure. The new replacement prosthesis continued to be evaluated over time from a new baseline.

All clinical procedures and evaluations were performed by calibrated investigators (M.F., J.M. and J.D.) who were members of the Department of Prosthodontics and Operative Dentistry at the University of Connecticut School of Dental Medicine. The detailed protocol for chairside adjustment and delivery has been described elsewhere^1,17 and included the use of a one- or two-bottle primer/adhesive system (Bond-1 or Bond-It, respectively, Pentron Clinical Technologies LLC) and a dual-cure, low-viscosity hybrid composite material (Lute-It, Pentron Clinical Technologies LLC).

We evaluated prostheses at baseline, six months and one year and then annually thereafter using four methods. The first is a slight modification of the California Dental Association, or CDA, quality evaluation system,¹⁸ which was used to evaluate surface integrity, shade, anatomical contour, marginal integrity and structural integrity. This clinical evaluation system categorizes each prosthesis parameter qualitatively from best to worst as "R," "S," "T" or "V." "R" and "S" are considered excellent and good, respectively, and a score of "T" or "V" for most parameters requires removal or repair of the prosthesis.

We measured plaque growth with the plaque index¹⁹ and gingival health with the gingival index,²⁰ along with probing depth at four sites at each abutment tooth. We analyzed the CDA and periodontal examination data with the Wilcoxon Matched Pairs Signed Ranks test and SPSS software (SPSS for Macintosh Release 10.0, SPSS Inc., Chicago). For these data, 13 (45 percent) of the 29 surviving prostheses were due for the four-year recall examination, 12 of which we examined (because prostheses were placed at staggered times, not all were due for a four-year recall examination).

We gave prostheses an overall assessment of "surviving" or "nonsurviving" based on clinical examination results and application of the CDA criteria or on the basis of reports from patients regarding a failure (and consequent loss) of the prosthesis, which was verified by clinical examination. These dichotomous survival data were analyzed with the Kaplan-Meier Survival analysis. Differences between groups were compared with the log-rank test and SPSS software. The Kaplan-Meier procedure allows inclusion of data at all points in time.

We made photographs and vinylsiloxane impressions of selected quadrants to indirectly evaluate the prostheses. The impressions were poured with improved dental stone and then with epoxy (low viscosity embedding media, Electron Microscopy Sciences, Fort Washington, Pa.). Representative scanning electron micrographs, or SEMs (model JSM-6230F, JEOL Ltd., Tokyo) were made from the epoxy models.

	RESULTS

TOP ABSTRACT COMMERCIALLY AVAILABLE SYSTEMS SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The table shows the distribution of all bridges made for this study. The prostheses are divided according to retainer design (intracoronal vs. extracoronal), substructure design (low vs. high volume) and whether the bridge was an initial placement or a replacement.

Survival analysis. Figures 5, 6, 7 and 8 show survival analysis plots. These plots show the percentage of prostheses in satisfactory clinical service (that is, surviving) as a function of time. The vertical axis represents the percentage of prostheses that have survived clinically, while the horizontal axis illustrates the time in years of survival. As is customary with the Kaplan-Meier analysis, each vertical step represents a single failed prosthesis, and the data points on the horizontal component of the plot represent active, surviving prostheses. Because of the staggered placement of study prostheses, the surviving prostheses ranged in age from 1.95 to 4.61 years, with a mean (± standard deviation, or SD) age of 3.8 ± 0.6 years.

View larger version (34K): [in this window] [in a new window]   Figure 5. Survival analysis of all fiber-reinforced composite prostheses in the clinical trial. For all patients, 29 (74 percent) of 39 prostheses survived. When patients with severe parafunctional (bruxing) habits were excluded, 28 (82 percent) of 34 prostheses survived.

View larger version (33K): [in this window] [in a new window]   Figure 6. Survival analysis according to substructure design (excluding bruxers). Twenty (95 percent) of 21 high-volume prostheses survived, while only eight (62 percent) of 13 low-volume prostheses survived.

View larger version (32K): [in this window] [in a new window]   Figure 7. Survival analysis according to retainer design (excluding bruxers). Nineteen (86 percent) of 22 extracoronal, full-coverage prostheses survived, while nine (75 percent) of 12 intracoronal, partial-coverage prostheses survived.

View larger version (31K): [in this window] [in a new window]   Figure 8. Survival analysis according to retainer design for high-volume substructures only (excluding bruxers). Fourteen (100 percent) of 14 extracoronal, full-coverage prostheses survived, while six (86 percent) of seven intracoronal, partial-coverage prostheses survived.

We made the following comparisons with the exclusion of the two subjects with severe para-functional habits. Figure 6 describes prosthesis survival according to the substructure design (low volume or high volume). There was a statistically significant difference (P < .01) in survival between the prostheses made with the low-volume substructure design and those made with the high-volume substructure design. We observed a 95 percent survival rate for the high-volume prostheses (mean ± SD observation time, 3.75 ± 0.4 years) in contrast to a 62 percent survival rate for the low-volume prostheses.

Figure 7 shows a comparison of survival between prostheses with different retainer designs (that is, extracoronal vs. intracoronal). We found no statistically significant difference (P = .4) between the two groups. In addition, there was no statistically significant difference (P = .2) when retainer designs for only the high-volume substructures were compared (Figure 8).

Low-volume substructure failures. Failure in the low-volume substructure prostheses was a result of veneer fractures down to the FRC substructure in the pontic regions. This demonstrated that the FRC substructure needed to provide better support for the particulate composite veneer. Figure 9 shows a representative example of this type of fracture. In this case, a small segment of the particulate composite veneer on the occlusal aspect of the prosthesis was lost. A large segment of the veneer that had fractured from the facial and cervical aspects of the prosthesis was recovered for microscopic analysis.

View larger version (37K): [in this window] [in a new window]   Figure 9. Clinical veneer fracture of a low-volume substructure prosthesis. SEM: Scanning electron microscopy.

Modified CDA criteria. With regard to the clinical parameters of shade, anatomical form and marginal integrity, few changes occurred from baseline to the 48-month recall examination, and all surviving bridges exhibited good overall clinical performance. We observed no statistically significant changes during this time for any of these parameters. In addition, we detected no recurrent caries. With regard to the category of surface integrity (that is, smoothness, chipping) at 24 months, two (7 percent) of the 29 surviving prostheses developed repairable surface defects and received scores of "V." At 48 months, no prosthesis received a "V" score. One prosthesis that received a "V" score at 24 months improved to an "R" score after repair; the other prosthesis was not due for the 48-month recall examination.

Loss of surface luster. We observed a loss of surface luster for all of the prostheses with "S" scores (good) and for many of the prostheses with "R" scores (excellent) shortly after placement. However, this clinical evaluation is performed after air-drying. Under normal moist conditions, the change in surface luster is not detected clinically. At baseline, 23 (89 percent) of 26 examined prostheses received an "R" score for "surface," and at 48 months, 10 (83 percent) of 12 recalled prostheses maintained this score, with all remaining prostheses receiving "S" scores. There was, however, no statistically significant difference in surface scores between baseline and 48 months. Despite the loss of luster, external surfaces with "S" scores generally exhibited a consistently smooth texture. Figures 10 and 11 illustrate the clinical appearance of a representative study prosthesis at baseline and at 48 months.

View larger version (79K): [in this window] [in a new window]   Figure 10. Extracoronal fiber-reinforced composite pros-thesis at baseline examination.

View larger version (84K): [in this window] [in a new window]   Figure 11. Extracoronal fiber-reinforced composite prosthesis at 48-month recall examination.

SEM analysis. We made SEMs from epoxy models for a detailed evaluation of the occlusal surfaces of selected study prostheses. The SEMs (Figures 12 and 13) show the occlusal surface of a representative prosthesis at baseline and 36-month evaluations. We carefully surveyed broad areas and within deep depressions or grooves that were created by the technician or dentist before prosthesis delivery. At x 25 and x 130 magnification, we observed no FRC exposed to the occlusal surface. Furthermore, landmarks seen at baseline were readily identifiable at 36 months, demonstrating very modest occlusal wear.

View larger version (134K): [in this window] [in a new window]   Figure 12. Scanning electron micrograph at occlusal surface of extracoronal fiber-reinforced composite prosthesis replica at baseline.

View larger version (142K): [in this window] [in a new window]   Figure 13. Scanning electron micrograph at occlusal surface of extracoronal fiber-reinforced composite pros-thesis replica at 36 months.

	DISCUSSION

TOP ABSTRACT COMMERCIALLY AVAILABLE SYSTEMS SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

Metal-ceramic bridges. The outcomes of this study should be viewed in the context of clinical evaluations of metal-ceramic conventional and etched metal Maryland bridges. Scurria and colleagues²¹ conducted a meta-analysis of nine conventional bridge studies, which showed a survival rate of 95 percent at four years. At 10 and 15 years, this analysis demonstrated survival rates of 87 percent and 69 percent, respectively. Creugers and Van’t Hof²² used this approach to analyze 11 studies of clinical Maryland bridges, and determined that 74 percent of the prostheses remained in function after four years.

The data from this study also have shown that prostheses made with low-volume FRC substructure designs have statistically significantly less chance of surviving (62 percent survival from 2.64 to 4.61 years) than do prostheses made with the high-volume substructure (95 percent survival from 2.77 to 4.30 years). As stated above, the low-volume substructure consists of an approximately 2-mm by 3-mm bar across the pontic region, which does not appear to provide adequate support to the particulate composite veneer bonded to the external surface of the substructure. The low-volume substructure failures were characterized clinically by loss of at least 25 percent of the occlusal veneer, down to—and sometimes including—the substructure of the pontic.

The problem appears analogous to the loss of a segment of porcelain veneer from a metal-ceramic prosthetic substructure exhibiting inadequate rigidity and vertical support. Porcelain and particulate composite both are brittle materials that require adequate support. A high volume of FRC in the prosthesis results in increased substructural rigidity and provides a broader base of support for the composite veneer. The large number of veneer fractures seen for the low-volume substructures demonstrates the clinical problem with this design.

The continued bonding of the external layer of the FRC substructure to the particulate composite in an examined fractured segment demonstrates the integrity of the interface and the value of maintaining the air-inhibited layer as the external surface of the substructure. It is interesting to note that all surviving low-volume pros-theses except one survived for more than four years. All failures occurred within the first two years. The only prosthesis in this group that did not survive more than four years is the one for a patient who died. His prosthesis survival time was stopped at 2.64 years.

We hypothesize that the combination of the support provided by the high-volume substructure design and the maintenance of the air-inhibited layer, which provides high bond strength between the substructure and veneer, likely is responsible for the good success of the FRC bridges observed in this study. The maintenance of the air-inhibited layer, and the subsequent bond between the veneer and the substructure, probably prevented fracture of the eight low-volume prostheses that remained in service.

In an in vitro fracture-resistance study of molar crowns, Behr and colleagues²³ demonstrated the advantage of maintaining the air-inhibited layer to create a high-strength bond between the veneer and substructure. They compared the Sculpture/FibreKor system used in our study with the Targis/Vectris system, in which the air-inhibited layer is intentionally removed from the external surface of the substructure as part of the manufacturer’s recommended procedure and then treated with silane coupling agent. Behr and colleagues²³ concluded that the fracture resistance of the experimental crowns depended primarily on the bond strength between the fiber substructure and the veneer material.

In their study, Göhring and colleagues^14,24 identified the maintenance of this air-inhibited layer as an important factor in the fabrication of the intracoronal FRC prostheses. They modified the manufacturer’s recommended technique for making Targis/Vectris inlay prostheses by substituting the vacuum and pressure polymerization process with light polymerization and hand placement, which allowed for maintenance of the air-inhibited layer and a durable adhesion between fiber reinforcement and veneering composite.

Vallittu and Sevelius¹⁵ reported on the clinical performance of 31 FRC fixed prostheses with primarily partial-coverage retainers. For these pros-theses, they used continuous unidirectional glass FRC preimpregnated with a thermoplastic polymer (Stick, Stick Tech Ltd.) to fabricate the substructure. The substructure was veneered with a light-polymerized particulate composite (Sinfony, 3M ESPE or Vita Zeta LC, Vita Zahn-fabrik, Bad Säckingen, Germany). At up to 24 months, no substructure failures occurred and 93 percent of the prostheses were in service, with the only failures being the result of debonding of two prostheses from the abutment teeth.

In our study, the clinical evaluation of the surviving FRC prostheses showed that the particulate composite veneer material exhibited good color stability and resistance to wear. The anatomical form on axial and occlusal surfaces exhibited little, if any, change. The surface defects that resulted in unsatisfactory modified CDA scores in two (7 percent) of 29 prostheses at the 24-month recall examination appear to be due to a fabrication error in one case and to patient parafunctional habits in the other case.

Careful examination of these cases (including casts and photographs) supports these hypotheses. In one case, a small pit and adjacent shaded region visible at the six-month recall visit suggested a void or other fabrication error when the increment of composite veneer was applied. In the second case, the patient displayed moderate para-functional habits, resulting in several chipped restored and unrestored teeth. The successful results seen in this study and the results of a similar evaluation by Vallittu and Sevelius¹⁵ are, however, in contrast to a recently published report, which noted high amounts of "chipping and delamination" with various FRC/particulate composite products.²⁵

The SEMs provide additional evidence regarding the minimal degree of occlusal wear noted during the course of this study. Some SEMs showed evidence of changes occurring in areas where the technician made incremental additions of composite veneer. These changes may be minimized through careful blending and condensing of incremental additions with specially designed brushes, as well as through the careful maintenance of the air-inhibited layer on the surface that is to receive an incremental addition. It may, however, be best to add all of the particulate composite to the occlusal aspect of the FRC substructure at one time and not in increments. The SEM evaluation did not show that FRC was visible on the occlusal surface of experimental prostheses.

Clinical concepts derived from study. The box shows the key elements for the successful use of FRC prostheses based on the data and on our clinical experience. These elements include careful case selection according to the indications and contraindications for the use of these materials.¹ Key elements for the dentist include tooth preparation design that allows for an adequate amount of FRC, an accurate interocclusal registration and proper insertion technique. An inaccurate interocclusal registration could result in considerable occlusal adjustment by the dentist at chairside and the potential for inadvertent occlusal exposure of the FRC substructure or the creation of an extremely thin layer of particulate composite veneer in functional areas.

The laboratory technician must carefully articulate casts, provide a substructure design with a large volume of FRC placed into areas that do not create esthetic or periodontal problems, and maintain the integrity of the air-inhibited layer between all layers of FRC and particulate composite.

Potential study limitations. Although this clinical study had a modest sample size and low statistical power, we were able to detect a difference between two substructure designs. Of additional concern is a potential lack of generalizability to everyday clinical practice. The prostheses followed up in this study were made with careful control of all clinical and laboratory procedures. We believe that dentists and laboratory technicians need to follow case selection criteria and procedures such as those outlined in the box and above to achieve a satisfactory clinical result.

	CONCLUSIONS

TOP ABSTRACT COMMERCIALLY AVAILABLE SYSTEMS SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The data from this prospective clinical study of 39 FRC fixed prostheses show that both extracoronal and intracoronal prostheses made with a substructure of preimpregnated, unidirectional FRC and veneered with a hybrid particulate composite provided good clinical service for four or more years. These data show that clinical survival was primarily associated with the substructure design, and that the high-volume FRC design is superior to the low-volume design. Except for loss of surface luster in many of the prostheses, clinical characteristics such as shade, anatomical form, structural integrity, marginal integrity and surface integrity were largely unchanged from baseline to 48 months.

Like all other materials, FRC cannot compensate for all clinical and laboratory errors and incorrect uses of the material. Specifically, this study demonstrated that a higher likelihood of failure is associated with the low-volume substructure design. The data obtained from a small subgroup of the sample strongly suggested that the survival of FRC prostheses may be compromised when they are subjected to severe para-functional habits. This is not unexpected and occurred in two subjects. Our clinical and laboratory experience also has shown that laboratory errors occurring during the fabrication of the particulate composite veneer also compromise the survival of FRC prostheses. Future developments in FRC prostheses should address these issues of operator and technician sensitivity.

We must emphasize that FRC-supported prostheses are not all the same. The type of FRC (preimpregnated vs. nonimpregnated and glass vs. polyethylene) design and external surface characteristics of the substructure, as well as the type of particulate composite veneer are all important factors that affect clinical performance. We believe that the current indications for extracoronal and intracoronal polymer prostheses made with a high-volume substructure composed of preimpregnated, unidirectional FRC are for short-span tooth replacement. Specifically, these prostheses are indicated when a metal substructure is not desired, when abutment teeth exhibit poor geometric retention form or both (extracoronal prostheses) and when abutment teeth are not restored or are minimally restored (intracoronal prostheses).

View larger version (105K): [in this window] [in a new window]   Dr. Freilich is an associate professor, Department of Prosthodontics and Operative Dentistry, University of Connecticut School of Dental Medicine, 263 Farmington Ave., Farmington, Conn. 06030-1615, e-mail "freilich{at}nso2.uchc.edu". Address reprint requests to Dr. Freilich.

View larger version (117K): [in this window] [in a new window]   Dr. Meiers is an associate professor, Department of Prosthodontics and Operative Dentistry, University of Connecticut School of Dental Medicine, Farmington, Conn.

View larger version (125K): [in this window] [in a new window]   Dr. Duncan is an assistant professor, Department of Prosthodontics and Operative Dentistry, University of Connecticut School of Dental Medicine, Farmington, Conn.

View larger version (120K): [in this window] [in a new window]   Ms. Eckrote is a research assistant, Center for Biomaterials, University of Connecticut School of Dental Medicine, Farmington, Conn.

View larger version (123K): [in this window] [in a new window]   Dr. Goldberg is a professor, Department of Prosthodontics and Operative Dentistry, and director, Center for Biomaterials, University of Connecticut School of Dental Medicine, Farmington, Conn.