Midterm results of a 5-year prospective clinical investigation of extended ceramic veneers

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

available at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/dema

Midterm results of a 5-year prospective clinical investigation of extended ceramic veneers Petra C. Guess a,b , Christian F.J. Stappert a,b,c,∗ a

Department of Prosthodontics, School of Dentistry, University Hospital Freiburg, Albert-Ludwigs-University, Freiburg, Germany b Department of Biomaterials and Biomimetics, New York University College of Dentistry, New York, USA c Department of Periodontology and Implant Dentistry, New York University College of Dentistry, New York, USA

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objectives. Midterm-evaluation of a prospective 5-year clinical study on long-term perfor-

Received 5 July 2007

mance and success rate of pressed-ceramic veneers with two extended preparation designs.

Received in revised form

Methods. Anterior teeth of 25 patients were restored with 66 extended veneers. Forty-two

8 September 2007

overlap veneers (OV) (incisal-edge-reduction 0.5–1.5 mm, butt-joint) and 24 full veneers

Accepted 29 September 2007

(FV) were inserted. Both veneer designs were similar in buccal (0.5 mm) and proximal (0.5–0.7 mm) chamfer preparation, but differed in palatal extension. Ceramic veneers were fabricated with IPS Empress* and adhesively luted with dual-polymerizing composite Var-

Keywords:

iolink II* (*Ivoclar Vivadent). Clinical reevaluations were performed 6, 12, 25, 39, 45, and

Ceramic

62 months after insertion of the veneers according to the modified USPHS-criteria. Abso-

Veneer

lute failures were recorded as survival-rate, relative failures demonstrated by Kaplan–Meier

Overlap veneer

success-rate.

Full veneer

Results. After an observation time up to 5 years, survival-rate of full veneers was 100%, of

Press ceramic

overlap veneers 97.5% due to one severe fracture. Kaplan–Meier-analysis of relative failures

Partial coverage restoration

resulted in a success-rate of 85% for FV and 72% for OV. Reasons for relative failures were cracks, ceramic-cohesive-fractures, and loss-of-adhesion. No significant differences were found between the two veneer groups. Secondary caries and endodontic complications did not occur. Increased clinical service time resulted in enhanced marginal discoloration and decrease of marginal adaptation. Significance. Extended pressed-ceramic veneers (both OV and FV) proved to be reliable procedures to restore larger deficits in anterior teeth. Pronounced palatal extension of full veneers was not linked to a higher failure probability. Reliable adhesive bonding, as well as ceramic fatigue and fracture resistance are considered key factors for long-term success of extended-veneer restorations. © 2007 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

All-ceramic veneers have been introduced to the dental community more than 20 years ago [1]. Since then, clinical

∗

survival rates of all-ceramic veneers have become predictable [2–8]. Long-term clinical data covering follow-up evaluation periods up to 15 years are available [9]. According to the literature, long-term success of veneers is determined by

Corresponding author at: Department of Periodontology and Implant Dentistry, Department of Biomaterials and Biomimetics, New York University College of Dentistry, Arnold and Marie Schwartz Hall of Dental Sciences, 345 East 24th Street (Room 846S), New York, 10010 NY, USA. Tel.: +1 212 998 9986; fax: +1 212 995 4244. E-mail address: [email protected] (C.F.J. Stappert). 0109-5641/$ – see front matter © 2007 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2007.09.009

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

material properties and fatigue resistance of ceramic and adhesive systems used. Further factors for clinical success are marginal adaptation of the veneer restoration and functional and morphological condition of the abutment tooth [10]. When ceramic veneers were first clinically applied, a preparation of the abutment tooth was avoided. Not only did this led to distinct over-contouring at cervical and proximal tooth surfaces, but also to higher clinical failure rates as a result of gingival inflammation and secondary caries [11]. This led to the routine practice of tooth preparation for veneer restorations. Currently, four veneer preparation designs in reference to the incisal edge of the abutment tooth are described in the literature: the ‘window preparation’, which is limited to the labial tooth surface and does not involve the incisal edge; the ‘feather preparation’ with a thin ceramic layer up to the incisal edge; the ‘bevel preparation’ welting the incisal edge in ceramic and the ‘incisal overlap preparation’ which extends into the palatal surface by a chamfer preparation [12–14]. Especially in the early days of veneer therapy, the less invasive window and feather preparation designs were mainly used for the treatment of tooth discolorations. The development of more effective bleaching procedures reduced the necessity of these veneer indications [15]. The development of ceramic materials with improved mechanical properties, progress in adhesive bonding techniques, and improved luting composites, lead to a broader range of all-ceramic veneer applications [16,17]. Treating large morphological and structural defects demanded more than a labial and incisal covering of residual tooth structure. Since full crown preparations require removal of extensive tooth structure, modifications of established veneer preparations were made creating defect oriented veneer preparation by extending the designs [18]. Ceramic veneers involving the incisal edge, proximal areas and larger parts of the palatal surface have been recognized as an alternative to full crowns in the anterior dentition. The concept of a ‘full veneer’ was first described in the literature by Crispin [19]. When treating anterior teeth showing bleaching-resistant discolorations, slight morphological deficits or misalignments with all-ceramic veneers, pressed-ceramics [4,5], feldspathic ceramics [2,6,7] and CAD/CAM systems [8] have been widely used and have been scientifically well documented. Only few data exist on the long-term behavior of extended ceramic veneers [17]. In general, most clinical studies investigated various veneer types altogether, not taking into account the influence of preparation design. Therefore, a conclusive evidence-based statement on clinical longevity of proximally and palatally extended veneers [7,20] and full veneers [17] cannot be made yet. Based on previous in vitro fatigue studies of veneer preparation designs by Stappert et al. [14,21], the present prospective clinical study investigated the long-term behavior of allceramic IPS Empress veneers with two different extended tooth coverage preparations (modified overlap and full veneer design). As a null hypothesis no influence of preparation extension on clinical long-term success of pressed-ceramic veneers was estimated.

2.

805

Materials and methods

Initial examinations and inclusion of suitable patients for the study started in 1999. Twenty-five patients were selected for this study from the existing patient cliental of the Department of Prosthodontics, University Hospital Freiburg, according to their veneer indication type. Belser et al. [16] distinguished three indication groups for ceramic veneers (types I–III). Two of these indication groups, types II and III, were investigated in the present study. Veneer indication type II was applied when extensive morphological changes were necessary. This was the case for conoidal teeth (type II a), diastemata (type II b) and insufficient anterior tooth lengths (type II c). Indication type III was applied when the crown of the abutment tooth had more extensive damage, i.e. extensive crown fractures due to trauma (type III a1) or existing restorative fillings (type III a2). All patients had to show good oral hygiene and low caries risk at the beginning of treatment. Patients with parafunctions or pronounced malocclusion (e.g. frontal edge-to-edge articulation or cross bite) were not included in the study. Large cervical wedge-shaped defects, insufficient enamel quantity, and existing root canal treatments were further exclusion criteria. According to the study protocol, a maximum of six restorations per patient was permitted. Two preparation designs, overlapping incisal edge preparation (modified overlap design; OV) and full veneer preparation (FV), were investigated. Both preparation forms included the proximal tooth surfaces, but differed in palatal extension. The extent of tooth structure defect and the functional and esthetic objectives of the therapy determined the choice of preparation design. Therapy objectives had been ascertained by means of a diagnostic wax-up at the beginning of the treatment. Veneer preparations were performed as follows: Primary preparation was effected by rough diamond burs (80 ␮m) (#837KR.314.012, #878.204.012; Komet Dental, Gebr. Brasseler, Lemgo, Germany). Finer shape-congruent diamond burs (30–40 ␮m) (#8837KR.314.012, #8878.204.012; Komet Dental) were used for finishing procedure. Extent of labial and incisal reduction was pre-determined for both preparation forms using a silicone key made from the waxup. The labial surface was axially reduced by 0.3–0.5 mm (#878.204.012, #868B.314.018; Komet Dental). Cervically, a shallow chamfer (0.5 mm) was prepared epi-gingivally. The proximal reduction was 0.5–0.7 mm. The incisal edge was shortened by a minimum of 0.5–1.5 mm for both preparation forms (#837KR.314.012; Komet Dental), depending on the defect size. For the overlapping incisal edge preparation (overlap veneer, OV) the incisal edge was extensively shortened and slightly beveled towards the labial aspect. The angle between labial surface and incisal platform was approximately 110◦ . On the palatal aspect, a right-angled contour between the incisal and the palatal surface was achieved (butt joint) (Fig. 1a and b). The palatal centric contact point of all overlap veneers (OV) remained on the natural tooth structure. Full veneer (FV) restorations differed by preparing an extensive 0.5–0.7 mm deep rounded shoulder in palatal area. Extension of palatal preparation was generally limited to the cingulum area, however, an extension was justified with large tooth defects (Fig. 1c and d). Palatal centric contact points on the ceramic sur-

806

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

Fig. 1 – Modified overlap veneer preparation (OV) (butt joint) (a) labial-incisal-proximal view; (b) palatal extension and full veneer preparation (FV) (c) labial-incisal-proximal view; (d) palatal extension.

face were avoided, when applicable. All preparation margins were restricted by enamel. Labial epi-gingival preparation and controlled preparation depth enabled adhesive cementation mainly to enamel. All inner line angles were rounded. Preparation margins were not beveled. Manufacturer’s restrictions regarding minimum material thickness (axially at least 0.5 mm and incisally 1.5 mm), were consistently maintained during preparation. Incisal ceramic thickness of 1.5–2.5 mm was ensured for overlap and full veneer restorations. Impressions were taken using a polyether impression material (Permadyne, 3M ESPE, St. Paul, MN, USA). Temporary restorations were made chair-side (Protemp 3 Garant, 3M ESPE, St. Paul, MN, USA). To insert the temporary veneers enamel was etched punctually with 37% phosphoric acid (Total-Etch, Ivoclar Vivadent, Schaan, Liechtenstein) for 30 s, followed by applying unfilled adhesive bonding Heliobond (Ivoclar Vivadent, Schaan, Liechtenstein) and light-curing upon precise positioning. Extended-veneer restorations were made of the leucitereinforced glass ceramic IPS Empress (Ivoclar Vivadent, Schaan, Liechtenstein). This press-able ceramic was processed according to the IPS Empress layering technique using IPS Empress conventional lost-wax method following the

manufacturers’ instructions. After an ideal wax-up of the extended-veneer restoration, the surface was cut back by 0.3–0.8 mm to allow for a layer of veneering porcelain. The wax-up was then embedded in an investment cylinder. The glass-ceramic ingot was pressed into the pre-heated hollow mold (furnace EP 500, Ivoclar Vivadent, Schaan, Liechtenstein) at 910–920 ◦ C. Restorations were removed from the molds and cleaned with a steam jet cleaner (EV1 SJ, Silfradent Sync., Sofia, Italy). Manufacturing individually layered veneer restorations was enabled by fusing IPS Empress dentin, incisal, and transparent veneering porcelain, as well as the IPS Empress glaze material (Ivoclar Vivadent, Schaan, Liechtenstein) onto the ceramic core structure (ceramic furnace Programat P90/P95; Ivoclar Vivadent, Schaan, Liechtenstein). Minor corrections of the glazed restorations during tryingin were made chair-side. Changes to the incisal and labial surfaces were re-polished to a high-gloss using ceramic silicone polishers (Dialite Polishing Set Ceramic, Gebr. Brasseler, Lemgo, Germany). Veneers requiring major corrections or needing complete revisions were sent to the dental laboratory. Those restorations received entirely new coats of ceramic glaze.

807

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

Insertion was carried out using a cofferdam. For cementation, the prepared stumps were first cleaned with a polishing brush and fluoride-fee cleaning paste (Pell-ex Hawe Neos Dental, Bioggio, Switzerland). The prepared tooth structure was etched with 37% phosphoric acid (Total-Etch, Ivoclar Vivadent, Schaan, Liechtenstein), dentine for 15 s and enamel for 40 s, sprayed with water for 15 s, and dried. Syntac Primer was applied to the etched surface with a brush for 15 s and blowdried after a reaction time of 10 s. Then Syntac Adhesive (Ivoclar Vivadent, Schaan, Liechtenstein) was applied for a reaction time of 10 s. Heliobond (Ivoclar-Vivadent, Schaan, Liechtenstein) was applied to enamel and dentine and blown to a thin layer. To avoid inaccuracies of fit, Heliobond was not light-polymerized before restoration placement. At the same time, IPS Empress veneers were cleaned with 99% isopropanol, and the inner surfaces were etched with 4.9% hydrofluoric acid (IPS Ceramic Etching Gel, Ivoclar Vivadent, Schaan, Liechtenstein) for 60 s. The etched ceramic surface was thoroughly sprayed with water for 60 s and dried with oil-free compressed air, followed by application of a one-component adhesive silane (Monobond-S, Ivoclar Vivadent, Schaan, Liechtenstein). After a reaction time of 60 s the silanized ceramic surface was dried with air. The dual-polymerizing composite cement Variolink II (Ivoclar Vivadent, Schaan, Liechtenstein) was used for adhesive fixation. Composite cement was applied to the prepared tooth and to the inner surfaces of the restoration. Restorations were inserted with increasing pressure (∼5–10 N). Excess cement in all marginal areas was removed immediately with foam pellets. Twenty seconds of light-polymerizing at the incisal edge ensured stabilization of the veneer while other veneer surfaces remained covered (Elipar Free Light 2; 3M ESPE, St. Paul, MN, USA) [22]. Residual cement was removed with foam pellets and Superfloss (Oral-B, London, UK). Glycerin gel (Liquid-Strip, Ivoclar Vivadent, Schaan, Liechtenstein) ensured oxygen inhibition during light-polymerizing of oral,

vestibular and proximal surfaces for 40 s each with light intensity of at least 650 mW/cm2 (Elipar Free Light 2; 3M ESPE, St. Paul, MN, USA). Residual excess cement was removed with a 15c scalpel (#371716, Bard-Parker; Becton-Dickinson, Dr. Franklin Lakes, NJ, USA). Epi-gingivally, excess cement was removed with a finishing diamond (#852EF.314.014; Komet Dental) after 24 h and coronal excess cement with flexible polishing discs (Soft-Lex Pop-On, 3M ESPE, St. Paul, USA). Restoration margins were polished with silicone polishers (#9418.204.030, #9419.204.030, #9547.204.030; Komet Dental) and inter-proximal polishing strips (Soft-Lex Finishing Strips, 3M ESPE, St. Paul, USA). Static and the dynamic occlusion were checked. A total of 66 veneers were inserted from fall 2000 until 2003. Follow-up examinations were performed between 2000 and 2006. The veneers were classified by two independent investigators according to the modified United States Public Health Service- (USPHS) criteria [23,24] (Table 1) after 6, 12, 25, 39, 45, and 62 months post insertionem (Table 2). Recall assessments were not performed by the clinician who had placed the restorations. The restorations were visually inspected with dental mirror and probe, and clinically examined with waxfree dental floss. Deviations in color match, and anatomic form were recorded and photographed. Each restoration was examined for cracks, fractures, and debonding. Pulp vitality was verified with CO2 -test. The patients were questioned about possible postoperative complaints. This study was conducted according to the Declaration of Helsinki for clinical investigations and aligned with an ethics committee (AlbertLudwigs-University, Freiburg, Germany). For statistical evaluation, a survival rate evaluating absolute failures according to the in situ criterion, and a success rate describing relative failures were determined [20]. Absolute failure was defined as clinically unacceptable fractures and cracks, which required a replacement of the entire restoration,

Table 1 – USPHS-criteria for classification of veneers Characteristics

Rating

Secondary caries

Alpha Bravo

No evidence of caries contiguous with the margin of the restoration Caries evident contiguous with the margin of the restoration

Marginal adaptation

Alpha Bravo Charlie

No visible evidence of crevice along margin; no catch or penetration of explorer Visible evidence of crevice and/or catch of explorer; no penetration of explorer Visible evidence of crevice; penetration of explorer

Marginal discoloration

Alpha Bravo

No discoloration on the margin between the restoration and the tooth structure Superficial discoloration on the margin between the restoration and the tooth structure; does not penetrate in pulpal direction Discoloration has penetrated along the margin of the restorative material in pulpal direction

Charlie Color match

Alpha Bravo Charlie

Anatomic form

Alpha Bravo Charlie

Criteria

No mismatch in color, shade and/or translucency between restoration and adjacent tooth Mismatch between restoration and tooth structure within the normal range of color, shade and/or translucency (<1 shade off; Vita shade guide) Mismatch between restoration and tooth structure outside the normal range of color, shade and/or translucency (>1 shade off; Vita shade guide) The restoration is continuous with tooth anatomy The restoration is not continuous with tooth anatomy. The restoration is slightly under- or over-contoured The restoration is not continuous with tooth anatomy. Restoration material is missing; a surface concavity is ascertainable

808

Table 2 – Modified USPHS criteria and clinical evaluation of the extended veneer restorations at baseline and after a mean follow-up period of 6, 12, 25, 39, 45, and 62 months (number of patients and restorations)

Mean recall time (min–max) Patients Restorations

Baseline

1. Recall after 6 months

2. Recall after 12 months

3. Recall after 25 months

4. Recall after 39 months

5. Recall after 45 months

6. Recall after 62 months

0

6 (4–8)

12 (10–26)

25 (19–37)

39 (30–57)

45 (42–62)

62 (60–72)

22 52

14 35

10 26

11 26

9 23

25 66

24 63

USPHS-criteria/clinical parameters

Baseline

1. Recall after 6 months

2. Recall after 12 months

3. Recall after 25 months

4. Recall after 39 months

5. Recall after 45 months

6. Recall after 62 months

OV FV n = 27 n = 8

OV n = 15

FV n = 11

OV FV n = 17 n = 9

OV FV n = 15 n = 8

OV n = 42

FV n = 24

OV n = 39

FV n = 24

OV n = 30

FV n = 22

Alpha Bravo

42 –

24 –

39 –

24 –

30 –

22 –

26 –

8 –

15 –

11 –

17 –

9 –

15 –

8 –

Marginal adaptation

Alpha Bravo Charlie

42 – –

24 – –

38 1 –

23 1 –

28 2 –

20 2 –

21 5 –

7 1 –

13 2 –

9 2 –

13 4 –

7 2 –

12 3 –

6 2 –

Marginal discoloration

Alpha Bravo Charlie

42 – –

24 – –

30 9 –

23 1 –

19 11 –

14 8 –

21 5 –

6 2 –

3 12 –

3 8 –

10 7 –

4 5 –

8 7 –

3 5 –

Color match

Alpha Bravo Charlie

42 – –

24 – –

38 1 –

24 – –

29 1 –

22 – –

25 1 –

8 – –

15 – –

11 – –

17 – –

9 – –

15 – –

7 1 –

Anatomic form

Alpha Bravo Charlie

42 – –

24 – –

36 3 –

24 – –

29 1 –

22 – –

25 1

6 2 –

12 3 –

9 2 –

16 1 –

9 – –

15 – –

8 – –

Endodontic complications

Vitality negative Percussion positive

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

– –

Crack

None Small/acceptable Large/unacceptable

42 – – –

24 – – –

39 – – –

24 – – –

30 – – –

22 – – –

25 1 – –

8 – – –

15 – – –

10 1 – –

17 – – –

9 – – –

14 1 – –

8 – – –

Fracture

None Minimal/acceptable Extensive/unacceptable

42 – –

24 – –

39 – –

24 – –

30 – –

22 – –

23 3 1

8 – –

14 1 –

11 – –

15 2 –

9 – –

14 1 –

6 2 –

Retention of the veneer

Bonded Re-bonded Lost

42

24 – –

39 – –

24 – –

30 – –

22 – –

25 – 1

8 – –

15 – –

11 – –

17 – –

9 – –

14 1 –

8 – –

–

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

Secondary caries

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

and/or secondary caries as well as endodontic complications. A relative failure was defined as minimal cohesive ceramic fractures and cracks with defined extension, which were clinically acceptable, as well as adhesion loss of a restoration which could be successfully re-bonded. Relative failures were summarized in a success rate which was calculated according to the Kaplan–Meier analysis and graphically depicted [25]. The beginning of the observation interval started with incorporation of the restoration and the end of the interval was defined by incidence of a relative failure. The success rates of the overlap and full veneer restorations were statistically compared by calibrating the confidence intervals. The modified USPHS-criteria were tabulated using the statistics program R Development Core Team 2004 (R Foundation for Statistical Computing, Vienna, Austria).

3.

Results

The study population compromised 25 patients (12 (48%) women, mean age 43 years, range 19–64 years; and thirteen (52%) men, mean age 45 years, range 20–45 years). Conoidal teeth (n = 23, 35%), insufficient anterior tooth lengths (n = 14, 21%), and crown fractures due to traumata (n = 14, 21%) were the main indications for ceramic veneer treatments. Existing composite fillings (n = 9, 14%) and diastemata were further indications (n = 6, 9%). Of the 66 inserted extended veneers, 36 (86%) overlap and 24 (100%) full veneers were luted to maxillary anterior teeth, and six (14%) overlap restorations to mandibular anterior teeth (Fig. 2). Of initially 25 patients being treated with 42 overlap and 24 full veneer restorations, 24 patients came for a follow-up examination after a mean observation period of 6 months. For the second recall appointment after a mean follow-up period of 12 months, 22 patients (52 restorations, 30 over-

809

lap, 22 full veneer) were examined. Fourteen patients were summoned for the third recall (mean follow-up period of 25 months). At the fourth, fifth, and sixth recall appointment (mean follow-up period of 39, 45, and 62 months, respectively), 10, 11, and 9 patients, respectively, were examined (Table 2). At the present recall status, four patients (eight restorations) did not participate in the study anymore, which equals a total dropout rate of 16% (12% of the restorations). The majority of patients participated in the first, second and third recall periods, reflecting veneer observations after 4–37 months. Since the veneer restorations were inserted over a period of 3 years, some restorations did not reach the fourth, fifth and sixth recall periods yet. These data will be completed in the ongoing evaluation. In the given follow-up periods no endodontic complications or postoperative complaints were ascertained. No secondary caries developed (Table 2). Results for marginal adaptation were similar for both restoration types (Table 2). From the third follow-up period onwards, a decrease in marginal adaptation according to Bravo ratings was observed. Results for marginal discoloration were similar for the overlap and the full veneer restorations during the first four follow-ups (Table 2). During the fifth and sixth follow-up period, marginal discolorations (Bravo ratings) occurred more frequently; this was particularly apparent in full veneer restorations. Color match and the anatomic form were predominantly rated as Alpha for both restoration forms for the entire observation period (Table 2). One overlap veneer restoration demonstrated an absolute failure in form of a clinically unacceptable ceramic longitudinal fracture in the mesial-buccal area of a maxillary canine (Fig. 3a). The fracture occurred after the second follow-up with a service time of 20 months and was limited to the ceramic restoration material; the underlying tooth structure was not affected. The overlap veneer restoration had to be replaced by

Fig. 2 – Distribution of overlap and full veneer restorations according to the tooth position (maxillary and mandibular).

810

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

Fig. 4 – Kaplan–Meier success probability [%] according to preparation design (overlap veneer and full veneer), stating the time interval [years] (based on relative, not absolute failures).

Fig. 3 – (a) IPS Empress overlap veneer restoration of the right upper canine. Longitudinal fracture after a service time of 20 months. (b) IPS Empress full veneer restoration of the right upper incisor. Crack after a service time of 55 months.

a conventional metal-ceramic crown and was not listed in any of the subsequent investigations. The survival rate of the overlap restorations was thus 97.6%. All full veneer restorations remained in situ (survival rate of 100%). In six patients minimal ceramic cohesive fractures (chippings) were found. The minimal fractures were located in the palatal area of two overlap and two full veneer restorations; in two further overlap restorations, ceramic chippings were observed in the incisal and in the labial-cervical part, respectively. The restorations were burnished and remained in situ. The chipping areas were observed in the subsequent followups, and showed no further changes. An increased occurrence of minimal fractures was ascertained during the third, fourth, and fifth follow-up observation after a service time of 20–71 months (Table 2). In three further patients cracks in two overlap and one full veneer restoration were observed. The cracks were located in the distal-labial-incisal and palatal area (Fig. 3b). Due to the minimal extension of the cracks and prolonged bonding, the restorations did not have to be renewed. One patient showed adhesion loss of an overlap veneer restoration on the upper left lateral incisor after a service time of 61 months. The intact restoration was re-bonded with Variolink II. Six restorations with minimal cohesive fractures (four

overlap, two full veneer restorations), three restorations with cracks (two overlap, one full veneer restoration), and one debonded restoration (one overlap restoration) were categorized as relative failures. This resulted in an estimated success probability of 0.72 [0.56–0.87] for overlap veneer restorations and 0.85 [0.70–1.00] success probability for full veneer restorations after 5 years according to Kaplan–Meier estimation method (Fig. 4). Statistical comparison of both success probabilities showed overlapping of the confidence intervals. The difference between the success probabilities of both preparation forms was not significant (significance level = 0.05).

4.

Discussion

During the present 5-year observation period IPS Empress ceramic veneers with overlap and full veneer preparation showed promising survival rates of 97.6 and 100%, respectively. All full veneer restorations remained in situ in good function. One overlap restoration demonstrated an absolute failure after an observation period of 20 months. In other studies, IPS Empress overlap restorations achieved survival probabilities of 96.5% [26] and 98.8% [4] after 4.5 and 6 years, respectively. At this point the results obtained for full veneer restorations can only be compared to a 100% survival rate [17] for feldspathic ceramic veneers with pronounced palatal extension after 4.5 years. Many authors state that fractures are the most frequent cause for clinical failure of ceramic veneer restorations [6,9]. The frequency of clinically unacceptable fractures is given as follows: Magne et al. [17], 0%; Peumans et al. [6], 2% and Dumfahrt and Schaffer [2], 3%. In the present study, one absolute failure occurred of an overlap restoration bonded to a maxillary canine (2.3%). Strong lateral extrusion contacts on the

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

veneer due to canine guidance were held to be responsible for overloading and fracture of this veneer restoration [4]. Further, failure of the adhesive bond had to be considered, since the fractured ceramic fragment detached entirely from tooth substance showing no cohesive fracture [5], and composite build up material remained intact [10]. Similar longitudinal fracture patterns were described in 1.2 and 2.2% of IPS Empress veneers with palatal extension after five and up to 12 years observation time, respectively [4,5]. Partial adhesion to dentin or extensive composite fillings, and high load during static and/or dynamic occlusion increase susceptibility to fracture. Overlap and full veneer restorations showed clinically satisfying results with estimated success probabilities of 72 and 85%, respectively. The difference between the success rates was statistically not significant. Relative failures will be discussed in particular. Clinical investigations have found that minimal ceramic cohesive fractures (chipping) mainly occurred at the incisal edge due to functional stress concentrations [4] or palatally [6]. Four overlap veneer restorations demonstrated cohesive fractures (in total 9.5%: incisal 2.4%, palatal 4.8% and labiocervical 2.4%), which were clinically acceptable. Fradeani [4] reported chippings in 1.2% of the IPS Empress veneers after 6 years. Similar results for feldspathic ceramics were obtained by Dumfahrt and Schaffer [2] (2% after 10 years) and Peumans et al. [6] (9% palatal after 10 years). In the present study, two (8.3%) of the full veneer restorations showed palatal cohesive fractures. Magne et al. [17] observed ceramic cohesive fractures in 2% of comparable extended veneers using a feldspathic ceramic material after 4.5 years. ¨ Clinical studies by Magne et al. [17] and Probster et al. [26] showed that veneer restorations with locally contained cracks, yet intact adhesive bond between tooth, composite and ceramic may remain stable for years, comparable to enamel cracks in natural teeth. In the present study cracks occurred in two overlap restorations (4.7%) in the distallabial-incisal and in the palatal area, and in one full veneer restoration (4.1%) in the distal-labial area. Similar cracks in overlap restorations have been observed in previous investi¨ gations. A clinical study by Probster et al. [26] found cracks in 2.8% of IPS Empress veneers with incisal coverage after 4.5 years. Dumfahrt and Schaffer [2] detected multiple cracks in two feldspathic ceramic veneers (1%) after 10 years. The IPS Empress full veneer restorations in the present study showed a relatively low number of cracks compared to extensive feldspathic ceramic veneer restorations. In the study of Magne et al. [17] cracks had occurred in 12% of the restorations after 4.5 years; of which 10% had slightly extended cracks and 2% were clearly visible. Cracks mainly formed in the palatal area, and some in the labial aspect. Many factors are responsible for crack development. Despite the continuous improvement of dental-ceramic materials, their low ductility, brittleness and sub-critical crack growth [27,28] are still their greatest shortcomings. The sintering process can cause residual stresses and structural flaws in the ceramic material [29]. Residual stresses in the ceramic and/or in the bonding system promotes micro-crack formation and propagation [30]. Polymerization shrinkage of the luting composite creates stress concentrations at the adhesive interface [31] and at ceramic subsurface. Exposure to fluctuating temperatures between 0 and 67 ◦ C

811

from food intake [32] induces different rates of thermal expansion of the bonded materials, enhancing water absorption of the luting composite, and thus contribute to weakening not only the bond quality but also the ceramic restoration [33]. A laboratory study of ceramic veneers by Magne et al. [34] confirmed that an increasing number of thermal cycles caused an increase in crack formation. In a clinical study of feldspathic ceramic veneers, Peumans et al. [6] compared 5-year (2%) and 10-year (21%) results and proved a clear increase of visible cracks buccally and palatally due to extended masticatory loading time. Further, crack formation in ceramic veneers can be enhanced by palatal and proximal extension of preparation [14], location of restoration margins, occlusal contact points, and layer thickness of adhesive cement and ceramic itself [6,34]. In the current study, one overlap veneer (2.3%) detached after an observation period of 62 months. The restoration was ¨ in perfect condition and was reinserted. Probster et al. [26] reported a similar loss of retention of one veneer (0.7%) and Fradeani et al. [5] of three veneers (3.3%) after 4.5 and 12 years, respectively. Secondary caries did not occur at either veneer restoration type; this concurs with other veneer studies [2,4,17,26,35]. To avoid secondary caries, great importance is attributed to preparation margins bound by enamel and do not end in composite fillings [36]. In a study by Peumans et al. [6] veneers with restoration margins located in composite fillings showed a secondary caries incidence of 10% after 10 years. All teeth maintained pulp vitality for entire investigation period of 62 months. Similar results were found in clinical studies for veneer restorations of 4.5 to 10 years [2,17]. In contrast, Peumans et al. [6], found 4% of abutment teeth with veneer restorations and extensive composite fillings to be nonvital over an observation period of 10 years. The present study cannot prove a correlation between preparation design and quality of restoration margins. Satisfactory clinical results for both veneer restoration types regarding marginal adaptation (20 and 25%, respectively, Bravo rating), according to the USPHS-criteria, were achieved using adhesive luting agent (Syntac Classic and Variolink II) over the observation period of up to 62 months. Peumans et al. [6] found 36% of the veneer restoration margins to be clinically detectable by probing and described a clear increase in Bravo ratings over an observation period of up to 10.5 years. Only 14 and 4%, respectively, of the veneers showed an excellent marginal adaptation (Alpha rating) after 5 and 10 years [6,26]. Better marginal adaptation was achieved in clinical studies for IPS Empress veneers of Fradeani et al. ¨ [4,5] and Probster et al. [26] (87.8, 97.6, and 92%, respectively, Alpha rating after 4.5, 6, and 12 years). Aging of adhesive luting agents may be responsible for decrease of marginal adaptation, as a result of primary polymerization shrinkage, different coefficients of thermal expansion of the bonding materials, micro-leakage and composite washout [6,37]. A clinical study [17] of extensive veneer restorations found higher palatal marginal discrepancies, which were attributed to tensile stresses during masticatory loading in the palatal area. However, a statistical comparison of marginal fit dependent on the tooth location of the margin showed no significant differences [17].

812

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

Especially for overlap restorations (follow-up after 45 and 62 months), a decrease in marginal adaptation was accompanied by an increase in marginal discolorations in the form of higher Bravo ratings. Marginal discolorations have been reported to be 26, 28, and 65%, respectively, for veneer restorations after an observation period of 4.5, 5, and 10 years [6,26,38]. Dumfahrt and Schaffer [2] essentially attributed the discolorations in the marginal area and their increased occurrence in the course of the observation period to the use of a dual-polymerizing luting composite (Dual Zement, IvoclarVivadent, Schaan, Liechtenstein). Marginal dentin exposure, unfavorable enamel morphology [39], exogenous deposits and plaque [2], and the viscosity of the luting composite [38] have been discussed as additional factors for marginal discolorations. A study conducted by Fradeani et al. [4,5], implemented a pure light-polymerizing luting composite and found that marginal discolorations were not a frequent complication (92.8% Alpha rating after 6 years; 86.4% Alpha rating after 12 years). Surface discolorations were removed with a finishing diamond, which also explains the higher number of Alpha ratings for marginal accuracy. The overlap and full veneer restorations showed very good results for color match, due to the high translucency of the leucite-reinforced glass ceramic [40] and the use of the layering technique. Analogous to the results of Fradeani et al. [5], a decline in color quality during follow-up period was not found for either restoration type. If however the staining technique was used for individualization of veneer restorations, a slightly impaired color adaptation after an observation ¨ period of 4.5 years was documented by Probster et al. [26]. In the present study, the incisal edge was shortened during the butt joint preparation of the overlap restorations to reduce crack formation and fractures. This butt joint preparation form enabled a thicker ceramic layer at the incisal and palatal aspect of the veneer restoration [41]. In a FEA by Magne and Douglas [42], lower tensile stress values in the palatal area were found for this preparation geometry than for the mini-chamfer. While Magne and Douglas [42] considered an extension of the ceramic veneer into palatal direction to be critical because of maximum tensile stresses in the palatal concavity, the present study found no proof for increased risk of failure. This could be explained by the fact that the restoration margins were basically not located in the palatal concavity. For the overlap preparations palatal preparation of restoration margin remained coronal to the palatal concavity. For the full veneer preparations the restoration margin was located in the convex area close to the cingulum. A 2D FEA by Magne and Douglas [42] reported considerably lower tensile stress values in this area. A higher mechanical retention and resistibility of full veneer restorations are attributed to the physical casing of the tooth [19]. Contrary to other veneer preparation forms, a larger adhesive surface may be available because of a greater enamel layer thickness in the cingulum and the proximal areas [19,43]. For extensive preparations, a reduced and more equal loading of the cement and the ceramic layer could be observed by means of 3D FEA [44]: the more extensive the preparation, the greater the tooth structure surface involved in stress distribution.

In a clinical study, Crispin [19] stated a higher longevity of full veneer restorations. This finding is in agreement with the better success rate of full veneer restorations (85%), as compared to overlap restorations (75%) in the present study. In a comparative in vitro study of full veneers [21], neither the palatal extension of the veneer, with the palatal contact point being located in the ceramic or in natural tooth substance, nor the palatal layer thickness of the ceramic in the case of enamel or dentin bonding had an influence on the fracture load of maxillary anterior teeth treated with full veneers. The clinical study of extended preparation forms by Magne et al. [17] could not prove an influence of the position of the restoration margin, the extent of the ceramic extension in the incisal edge area, and the incisal overbite on the failure rate of the veneer restorations. At present the null hypothesis cannot be rejected, as the preparation design did not significantly influence the clinical long-term success. Further clinical comparisons of overlap and full veneer restorations have not been described in the literature up until now. All-ceramic veneers based on extended preparation forms present a good alternative for preserving vital tooth structure for the treatment of larger morphological defects compared to full crowns. Cracks, ceramic cohesive fractures, and the lack of integrity of the adhesive bond may be limiting factors for the long-term success of ceramic extended veneers.

5.

Conclusions

(1) In this prospective clinical study covering an observation period of up to 62 months it was shown that pressed ceramic veneers with extended overlap and full veneer preparation are a reliable procedure for the restoration of larger defects in the anterior dentition. (2) A pronounced extension of the ceramic into the palatal tooth area in the form of a full veneer preparation could not be linked to a higher failure probability, as compared to the overlap preparation. (3) Adhesive bonding of tooth, ceramic and luting composite, and the fracture resistance of the ceramic are key factors for the long-term success of the extended veneer technique.

references

[1] Calamia JR. Etched porcelain veneers: the current state of the art. Quintessence Int 1985;16:5–12. [2] Dumfahrt H, Schaffer H. Porcelain laminate veneers. A retrospective evaluation after 1 to 10 years of service: Part II—clinical results. Int J Prosthodont 2000;13:9–18. [3] Brambilla GP, Cavalle E. Fractured incisors: a judicious restorative approach—part 2. Int Dent J 2007;57:100–8. [4] Fradeani M. Six-year follow-up with Empress veneers. Int J Periodontics Restorative Dent 1998;18:216–25. [5] Fradeani M, Redemagni M, Corrado M. Porcelain laminate veneers: 6- to 12-year clinical evaluation—a retrospective study. Int J Periodontics Restorative Dent 2005;25:9–17. [6] Peumans M, De Munck J, Fieuws S, Lambrechts P, Vanherle G, Van Meerbeek B. A prospective ten-year clinical trial of porcelain veneers. J Adhes Dent 2004;6:65–76.

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 804–813

[7] Smales RJ, Etemadi S. Long-term survival of porcelain laminate veneers using two preparation designs: a retrospective study. Int J Prosthodont 2004;17:323–6. [8] Wiedhahn K, Kerschbaum T, Fasbinder DF. Clinical long-term results with 617 Cerec veneers: a nine-year report. Int J Comput Dent 2005;8:233–46. [9] Friedman MJ. A 15-year review of porcelain veneer failure-a clinician’s observations. Compend Contin Educ Dent 1998;19:625–8 [630, 632 passim; quiz 638]. [10] Peumans M, Van Meerbeek B, Lambrechts P, Vanherle G. Porcelain veneers: a review of the literature. J Dent 2000;28:163–77. [11] Shaini FJ, Shortall AC, Marquis PM. Clinical performance of porcelain laminate veneers. A retrospective evaluation over a period of 6.5 years. J Oral Rehabil 1997;24:553–9. [12] Brunton PA, Wilson NH. Preparations for porcelain laminate veneers in general dental practice. Br Dent J 1998;184:553–6. [13] Walls AW, Steele JG, Wassell RW. Crowns and other extra-coronal restorations: porcelain laminate veneers. Br Dent J 2002;193:73–76, 79–82. [14] Stappert CF, Ozden U, Gerds T, Strub JR. Longevity and failure load of ceramic veneers with different preparation designs after exposure to masticatory simulation. J Prosthet Dent 2005;94:132–9. [15] Christensen GJ. Are veneers a conservative treatment? J Am Dent Assoc 2006;137:1721–3. [16] Belser UC, Magne P, Magne M. Ceramic laminate veneers: continuous evolution of indications. J Esthet Dent 1997;9:197–207. [17] Magne P, Perroud R, Hodges JS, Belser UC. Clinical performance of novel-design porcelain veneers for the recovery of coronal volume and length. Int J Periodontics Restorative Dent 2000;20:440–57. [18] Edelhoff D, Sorensen JA. Tooth structure removal associated with various preparation designs for anterior teeth. J Prosthet Dent 2002;87:503–9. [19] Crispin BJ. Expanding the application of facial ceramic veneers. J Calif Dent Assoc 1993;21:43–54. [20] Meijering AC, Creugers NH, Roeters FJ, Mulder J. Survival of three types of veneer restorations in a clinical trial: a 2.5-year interim evaluation. J Dent 1998;26:563–8. [21] Stappert CF, Stathopoulou N, Gerds T, Strub JR. Survival rate and fracture strength of maxillary incisors, restored with different kinds of full veneers. J Oral Rehabil 2005;32:266–72. [22] Stappert CF, Derks J, Gerds T, Strub JR. Marginal accuracy of press-ceramic full veneers with different preparation before and after mouth motion fatigue. Schweiz Monatsschr Zahnmed 2007;117:474–82. [23] Cvar J, Ryge G. Criteria for the clinical evaluation of dental restoration materials. US Public Health Service Publication No. 790 San Francisco US: Government Printing Office; 1971. [24] Cvar JF, Ryge G. Reprint of criteria for the clinical evaluation of dental restorative materials, 1971. Clin Oral Investig 2005;9:215–32. [25] Kaplan EL, Meier P. Nonparametric estimation from incomplete observation. J Am Statist Assoc 1958;53:457–81.

813

¨ [26] Probster L, Engel E, Masula B. Clinical long-term evaluation of adhesive IPS-Empress ceramic veneer restorations. Dental Labor 1999:1263–8. [27] Lawn BR, Pajares A, Zhang Y, Deng Y, Polack MA, Lloyd IK, et al. Materials design in the performance of all-ceramic crown. Biomaterials 2004;25:2885–92. [28] Lawn BR, Deng Y, Thompson VP. Use of contact testing in the characterization and design of all-ceramic crownlike layer structures: a review. J Prosthet Dent 2001;86:495–510. [29] Marx R. Modern ceramic materials for esthetic restorations—reinforcement and fracture toughness. Dtsch ¨ Zahnarztl Z 1993;48:229–36. [30] Anusavice K, Phillips R. Phillips’ science of dental materials. 10th ed. Philadelphia: W.B. Saunders Company; 1996. [31] Peumans M, Van Meerbeek B, Yoshida Y, Lambrechts P, Vanherle G. Porcelain veneers bonded to tooth structure: an ultra-morphological FE-SEM examination of the adhesive interface. Dent Mater 1999;15:105–19. [32] Palmer D, Barco M, Billy E. Temperature extremes produced orally by hot and cold liquids. J Prosthet Dent 1992;67:325–7. [33] Ortengren U, Elgh U, Spasenoska V, Milleding P, Haasum J, Karlsson S. Water sorption and flexural properties of a composite resin cement. Int J Prosthodont 2000;13:141–7. [34] Magne P, Kwon KR, Belser UC, Hodges JS, Douglas WH. Crack propensity of porcelain laminate veneers: a simulated operatory evaluation. J Prosthet Dent 1999;81:327–34. [35] Pippin DJ, Mixson JM, Soldan-Els AP. Clinical evaluation of restored maxillary incisors: veneers vs. PFM crowns. J Am Dent Assoc 1995;126:1523–9. [36] Rouse JS. Full veneer versus traditional veneer preparation: a discussion of interproximal extension. J Prosthet Dent 1997;78:545–9. [37] McKinney JE, Wu W. Chemical softening and wear of dental composites. J Dent Res 1985;64:1326–31. [38] Walls AW. The use of adhesively retained all-porcelain veneers during the management of fractured and worn anterior teeth. Part 2. Clinical results after 5 years of follow-up. Br Dent J 1995;178:337–40. [39] Ferrari M, Patroni S, Balleri P. Measurement of enamel thickness in relation to reduction for etched laminate veneers. Int J Periodontics Restorative Dent 1992;12:407–13. [40] El-Mowafy O, Brochu JF. Longevity and clinical performance of IPS-Empress ceramic restorations—a literature review. J Can Dent Assoc 2002;68:233–7. ¨ [41] Gurel G. The science and art of porcelain laminate veneers. Berlin: Quintessenz; 2003. [42] Magne P, Douglas WH. Design optimization and evolution of bonded ceramics for the anterior dentition: a finite-element analysis. Quintessence Int 1999;30:661–72. [43] Garber DA. Porcelain laminate veneers: ten years later. Part I: tooth preparation. J Esthet Dent 1993;5:57–62. [44] Zarone F, Apicella D, Sorrentino R, Ferro V, Aversa R, Apicella A. Influence of tooth preparation design on the stress distribution in maxillary central incisors restored by means of alumina porcelain veneers: a 3D-finite element analysis. Dent Mater 2005;21:1178–88.