Q U I N T E S S E N C E I N T E R N AT I O N A L

Clinical study of indirect composite resin inlays

in posterior stress-bearing preparations placed
by dental students: Results after 6 months and
1, 2, and 3 years
Juergen Manhart, DDS, Priv-Doz Dr Med Dent1/
Hong-Yan Chen, DDS, Dr Med Dent1/
Albert Mehl, DDS, Prof Dr Med Dent2/
Reinhard Hickel, DDS, Prof Dr Med Dent3

Objective: This longitudinal randomized controlled clinical trial evaluated composite resin
inlays for clinical acceptability in single- or multisurface preparations and provides a survey
of the results up to 3 years. Method and Materials: Twenty-one dental students placed
75 Artglass (Heraeus Kulzer) and 80 Charisma (Heraeus Kulzer) composite resin inlays in
Class 1 and 2 preparations in posterior teeth (89 adults). Clinical evaluation was per-
formed at baseline and up to 3 years by two other dentists using modified USPHS criteria.
Results: A total of 89.8% of Artglass and 84.1% of Charisma inlays were assessed as
clinically excellent or acceptable with predominating Alfa scores. Up to the 3-year recall,
five Artglass and 10 Charisma inlays failed mainly because of postoperative symptoms,
bulk fracture, and loss of marginal integrity. No significant differences between composite
resin materials could be detected at 3 years for all clinical criteria (P > .05). The compari-
son of restoration performance with time within both groups yielded a significant increase
in marginal discoloration (P < .05) and deterioration of marginal and restoration integrity
(P < .05) for both inlay systems. However, both changes were mainly effects of scoring
shifts from Alfa to Bravo. No significant differences (P > .05) were recorded comparing
premolars and molars. Small inlays showed significantly better outcome for some of the
tested clinical parameters (P < .05). Conclusion: Clinical assessment of Artglass and
Charisma composite resin inlays exhibited an annual failure rate of 3.4% and 5.3% that is
within the range of published data. Indirect composite inlays are a competitive restorative
procedure in stress-bearing preparations. (Quintessence Int 2010;41:399–410)

Key words: clinical study, composite resin, inlays, longevity, USPHS criteria

The rehabilitation of carious or fractured pos- problems associated with direct restorative
terior teeth using an inlay/onlay technique techniques, including, among others, inade-
was introduced to overcome some of the quate proximal or occlusal morphology, insuf-
ficient wear resistance or mechanical
properties of directly placed restorative mate-
1
Associate Professor, Department of Restorative Dentistry, rials, and the restoration of severely destroyed
School of Dentistry, Ludwig-Maximilians-University, Munich,
Germany.
teeth.1 Patients’ interest in the esthetic
2
restoration of posterior teeth has stimulated
Professor, Department of Computer-Generated Restorations,
University of Zurich, Zurich, Switzerland. the development of new, tooth-colored
3
Professor and Chair, Department of Restorative Dentistry, nonmetallic materials. Initial attempts to use
School of Dentistry, Ludwig-Maximilians-University, Munich, esthetic inlays were described at the end of
Germany.
the 19th century. This trend achieved larger
Correspondence: Dr Juergen Manhart, Department of acceptance with the introduction of restora-
Restorative Dentistry, School of Dentistry, Goethe Street 70,
80336 Munich, Germany. Fax: 49 89 5160-9302. Email: manhart@
tive materials bonded to natural tooth sub-
manhart.com strate and the growing concern about the use

VOLUME 41 • NUMBER 5 • MAY 2010 399

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

of metallic alloys.2 Esthetic alternatives to cast qualified than highly trained faculty mem-
gold inlays include composite resin and bers of university dental schools or operators
ceramic inlays. who were exposed to time constraints dur-
Today, many techniques and systems are ing everyday routine dental service, such as
available for tooth-colored inlays using both general practitioners.9,10
composite resin and all-ceramic materials.3 The aim of this ongoing prospective clin-
Esthetically, these materials are preferable ical trial was to evaluate posterior composite
alternatives to their traditional counterparts. In resin inlays that were placed by supervised
contrast to ceramic inlays, indirect composite dental students using the modified USPHS
resin restorations are less costly and more scoring system. The first null hypothesis
user-friendly.4 Composite resin inlays are usu- tested was that the clinical durability of two
ally indicated for the restoration of large composite resin inlay materials did not exhib-
defects. Compared with direct composite it significantly different results. The second
resin restorations, indirect composite resin null hypothesis tested was that the clinical
inlays feature the advantages of a limitation of performance of adhesive inlays in premolars
polymerization shrinkage to the width of the did not differ from that of molars. The third
luting gap, easier establishment of physiologic null hypothesis tested was that the clinical
interproximal contacts and occlusal anatomy, performance of adhesive composite resin
and improvement of wear resistance and inlays placed in one- or two-surface prepara-
physicomechanical properties by postcuring tions did not show significant differences
the inlay with light and/or heat. compared to a second group of multisurface
Clinical studies are needed to test these preparations.
materials in the oral environment. In contrast
to direct composite resin restorations, only a
limited number of studies have referred to the
long-term in vivo performance of composite METHOD AND MATERIALS
resin inlays as a restorative material for poste-
rior teeth. Further standardized clinical data Case selection
are necessary. For this reason, clinical trials and cavity preparation
require objective, reliable, and relevant crite- Twenty-one student operators of the Munich
ria to assess the performance of restora- Dental School in their third clinical training
tions.5,6 The US Public Health Service period placed 155 adhesive inlays in 89
(USPHS) evaluation system,7 designed origi- patients within a 6-month period under the
nally to reflect differences in acceptability supervision of three experienced clinicians
(yes/no) rather than in degrees of success, is from the university’s faculty. All students were
still the most commonly used direct method generally trained in clinical adhesive den-
for rating the quality of restorations. Recently, tistry during their first and second clinical
new recommendations for conducting con- semester and received further special train-
trolled clinical studies of dental restorative ing for the present study. The study was ori-
materials were published8; however, most of ented according to the guidelines of the
the current studies started earlier and are still CONSORT statement.11
based on modified USPHS criteria. Indication for treatment was replacement
In most controlled longitudinal studies, a of failed restorations or primary caries in
limited number of experienced clinicians, stress-bearing Class 1 and Class 2 prepara-
specially trained for the specific procedure, tions of premolars and molars. The mean age
place the restorations under almost ideal of the patients was 39.4 years (range 21 to 72
conditions. It is questionable whether these years). The laboratory composite resin
conditions match the situation that exists in Artglass and the composite resin Charisma
private dental clinics in which different levels were used (Table 1). All materials used in this
of operational skills can be found. Few longi- study were standard restorative materials in
tudinally designed clinical studies were con- the dental school at the time the restorations
ducted with operators who are either less were placed. The clinical investigation was

400 VOLUME 41 • NUMBER 5 • MAY 2010

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

Ta b l e 1 Materials, manufacturer, and composition

Composition
Material Type Manufacturer of resin matrix Filler

Artglass Polyglass composite Heraeus-Kulzer UDMA Filler content: 69 wt%

Bis-GMA Ba-Al-B-Si glass (D50 0.7
TEGDMA µm; D99 2.0 µm)
Multifunctional Highly dispersed silicon
methacrylates dioxide
Charisma Microhybrid composite Heraeus-Kulzer Bis-GMA Filler content: 78 wt%
TEGDMA Ba-Al-B-Si glass (D50 0.7
µm; D99 2.0 µm)
Highly dispersed silicon
dioxide (D99 0.01–0.04 µm)
Solid Bond 3-step etch-and- Heraeus-Kulzer Esticid-20FG Solid Bond P (Primer)
rinse adhesive 20 wt% phosphoric acid None
Solid Bond P (Primer) Solid Bond S (Sealer)
Water Ba-Al-B-F-Si glass (D50 0.7
Acetone µm; D99 < 2.0 µm): 30 wt%
Maleic acid Highly dispersed silicon
HEMA dioxide
Modified polycarboxylic acid
Solid Bond S (Sealer)
Bis-GMA
TEGDMA
HEMA
Maleic acid
Modified maleic acid
Twinlook Dual-cure resin cement Heraeus-Kulzer Bis-GMA Filler content: base 74 wt%,
TEGDMA catalyst 78 wt%
Ba-Al-B-Si glass (D50 0.7
µm; D99 2.0 µm)
Highly dispersed silicon
dioxide
2bond2 Dual-cure resin cement Heraeus-Kulzer UDMA Filler content: base 69.5
1,12-Dodecandioldi- wt%, catalyst 63.2 wt%
methacrylate Ba-Al-B-Si glass (D50 0.7
Multifunctional µm; D99 2.0 µm)
methacrylates Highly dispersed silicon
dioxide (D99 0.01–0.04 µm)
Strontium fluoride
(D99 < 1.0 µm)
(UDMA) urethane dimethacrylate; (Bis-GMA) bisphenol glycidyl methacrylate; (TEGDMA) triethylene glycol dimethacrylate; (HEMA)
hydroxyethyl methacrylate; (Ba) barium; (Al) aluminum; (B) boron; (Si) silicate; (D) diameter; (F) fluorine.

approved by an ethics committee, and each “Artglass + 2bond2,” or “Charisma + 2bond2,”

patient gave written consent to participate respectively8 (Tables 2 and 3).
before treatment. Patients receiving more than Detailed inclusion and exclusion criteria
one restoration received at least one restoration for patients or teeth are detailed as follows:
of each material. A maximum of two restora-
tions of each type were inserted into one indi- Inclusion
vidual. The two inlay materials were allocated to • Males and females at least 18 years of age
the teeth employing a random design using • Patients who are regular dental attendees
sealed envelopes that indicated the experi- and are willing/able to return to the sched-
mental groups, either “Artglass + Twinlook,” or uled postplacement assessments

VOLUME 41 • NUMBER 5 • MAY 2010 401

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

Ta b l e 2 No. and distribution of evaluated composite resin inlays at baseline

and at the 3-year recall

Premolars Molars

Material/ 1- and 2-surface Multisurface 1- and 2-surface Multisurface

resin cement restorations restorations restorations restorations

Baseline
Artglass
Twinlook (n = 30) 11 12 3 4
2bond2 (n = 45) 24 14 4 3
Charisma
2bond2 (n = 80) 22 25 18 15
3-year recall
Artglass
Twinlook (n = 23) 8 8 3 4
2bond2 (n = 26) 14 9 1 2
Charisma
2bond2 (n = 63) 19 19 13 12

Ta b l e 3 No. and size of evaluated composite resin inlays at baseline

and at the 3-year recall

Restorations

Material 1-surface 2-surface 3-surface 4-surface 5-surface

Baseline
Artglass 7 35 28 5 0
Charisma 6 34 31 7 2
3-year recall
Artglass 3 23 21 2 0
Charisma 2 30 26 3 2

• Written informed consent of patients to Exclusion

participate in the clinical study
• Patients with a high level of oral hygiene • Patients who are irregular dental attendees
(Lange approximal Plaque Index < 30% and • Patients with severe systemic diseases or
modified Sulcus Bleeding Index < 10%) allergies
• Permanent premolars and molars with • Patients with severe salivary gland dys-
Class 1 or Class 2 restorative treatment function
need, with contact to at least one neigh- • Patients maintaining an unacceptable
boring tooth and being in occlusion to standard of oral hygiene
antagonistic teeth • Teeth with severe periodontal problems
• Teeth with positive reaction to cold thermal • Nonvital teeth
stimulus and being free of clinical signs • Teeth with identifiable pulpal inflammation
and symptoms of periapical pathology or pain before treatment
• Isthmus size of the treated cavities at least • Teeth formerly or now subjected to direct
half the intercuspal distance pulp capping
• Teeth with ony initial defects

Before treatment, patients were inter-

viewed to determine whether the selected
teeth had a history of hypersensitivity.

402 VOLUME 41 • NUMBER 5 • MAY 2010

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

A local anesthetic was used for all 30 seconds and dentin for 15 seconds, fol-
patients. Teeth were cleaned with a fluoride- lowed by thorough washing of all surfaces
free prophylaxis paste and a rubber cup. All with water and subsequent drying of the
cavities were prepared according to common preparations with oil-free compressed air.
principles for adhesive inlays. Convergence Care was taken to avoid desiccation of the
angles of 10 to 12 degrees between oppos- tooth substrate. The adhesive system Solid
ing walls were prepared with 80-µm and fin- Bond (Heraeus Kulzer) was applied in all
ished with 25-µm grit diamond burs with a preparations according to the manufacturer’s
slight taper (Intensiv). Line and point angles instructions. All Charisma inlays were adhe-
were rounded; enamel and dentin margins sively luted with the dual-curing resin cement
were not beveled but prepared butt-joint. 2bond2. For Artglass inlays, two subgroups
When preparation margins extended into were built (see Table 2): 45 inlays were insert-
dentin, the teeth were included in the study ed with 2bond2 resin cement, and 30 inlays
only when rubber dam use for subsequent were luted using the dual-curing resin cement
inlay placement was still possible. The pulpal Twinlook. Excess resin cement was removed
floor was shaped to give the inlay an occlusal in all cases with an explorer, a brush, and den-
thickness of at least 1.5 mm; any undercuts tal floss interproximally. The inlays were cov-
were removed. After caries removal and cavi- ered at cavosurface margins with glycerin gel
ty preparation, teeth were reassessed for their to avoid oxygen inhibition of the luting resin
continued suitability for inclusion in the trial. A surface. Each inlay surface was light cured for
thin coat of calcium hydroxide liner (Life, Kerr 40 seconds with a polymerization light (Elipar
Italia) was applied to deep dentinal surfaces Highlight, 3M ESPE, monitored before each
in 14 Artglass and 17 Charisma cases and use at minimum 800 mW/cm2 intensity). After
covered by a punctual glass-ionomer base placement and removal of rubber dam, static
(Ketac-Bond Aplicap, 3M ESPE). Complete- and dynamic occlusion were adjusted using
arch impressions were taken with a polyether fine-grit diamond burs. Inlays were then fin-
material (Impregum F, 3M ESPE). Provisional ished with disks and strips (Sof-Lex, 3M
restorations were placed with eugenol-free ESPE) and polished (Enhance and
temporary cement (Provicol, Voco). Prismagloss composite polishing paste,
All inlays were made by a dental techni- Dentsply).
cian who was experienced in fabricating
composite resin inlays strictly following man- Evaluation of the restorations
ufacturer instructions. The inlays were The clinical status of each test tooth was
postcured in a light oven (Uni-XS, Heraeus recorded before restoration placement by
Kulzer) for 10 minutes to improve the physi- the supervised students. At baseline (14 days
cal properties. All inlays were definitively after treatment); 6 months; and 1, 2, and 3
inserted within 2 weeks after impression. years, the restored teeth were rated inde-
pendently with a mirror and probe by two
Placement of the inlays experienced faculty member clinicians not
After removal of provisional restorations, the involved with inlay placement. They were cal-
teeth were thoroughly cleaned with a pro- ibrated before the study by a joint examina-
phylaxis brush and pumice. Rubber dam tion of 20 indirect composite resin inlays
was used in all cases. After try-in of the inlays (Cohen kappa value > 0.62). To eliminate
to check proximal contacts and marginal fit, bias, the assessment was performed in a
all adhesive surfaces of the inlays were air- half-blind design in which the two clinicians
borne-particle abraded (aluminum oxide 50 had no preliminary information about the
µm, 2 bar), subsequently cleaned with type of restoration they examined.
ethanol, and air dried. A silane coupling At the 3-year recall, 63 of 89 patients with
agent (Monobond S, Vivadent) was applied 49 Artglass inlays (65%) and 63 Charisma
to all internal inlay surfaces. inlays (79%) could be evaluated (see Table 3
Enamel margins were etched using phos- and Fig 1). Missing restorations were primari-
phoric acid (Esticid-20FG, Heraeus Kulzer) for ly caused by patient dropout, while five

VOLUME 41 • NUMBER 5 • MAY 2010 403

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

Charlie = replacement of the restoration for

Individuals assessed for prevention; Delta = unacceptable, replace-
eligibility (n = 241)
ment immediately necessary). When there
86 excluded was disagreement during an evaluation, the
Not meeting the ultimate decision was made by forced con-
Enrollment inclusion criteria (n = 71)
sensus of the two examiners.15,16 Color photo-
Refused to participate
(n = 15) graphs with marked occlusal contact points
155 randomly Other reasons (n = 0) were taken.14
allocated
Statistical evaluation
Interexaminer reliability was determined by
75 allocated to Artglass 80 allocated to Charisma
inlays inlays calculating Cohen kappa value, which meas-
75 received allocated Allocation 80 received allocated ures agreement between the evaluations of
intervention intervention two raters when both are rating the same
object. Because of the ordinal structured
Follow-up at Follow-up at data, only nonparametric statistical proce-
6 mo (n = 75) 6 mo (n = 80) dures were used (P < .05). The Mann-Whitney
1 y (n = 70) 1 y (n = 75)
Follow-up U test was used to explore significant differ-
2 y (n = 64) 2 y (n = 71)
3 y (n = 49) 3 y (n = 63) ences of the 3-year results between both
26 lost to follow-up at 3 y 17 lost to follow-up at 3 y types of inlay materials for the criteria listed in
Table 4 and to analyze performance differ-
49 analyzed at 3 y 63 analyzed at 3 y ences between small versus large prepara-
26 excluded from analysis Analysis 17 excluded from analysis tions. For each material, Artglass or Charisma,
(lost to follow-up) (lost to follow-up) two classifications of restoration size were
built, one- or two-surface preparations (“small
Fig 1 Flow chart of the clinical trial participants comparing Artglass and
cavity” group) and three or more surfaces
Charisma composite resin inlays according to CONSORT statement.11
(“large cavity” group). Furthermore, perform-
ance differences between premolars versus
Ta b l e 4 Criteria and methods for the direct molars, and between Artglass inlays placed
evaluation of the restorations with Twinlook versus 2bond2 resin cement
were explored using the Mann-Whitney U
Criterion Methods of evaluation
test, as well as the performance of both mate-
Surface texture Visual and probe rials between baseline and 3 years. Because
Color match/change of restoration color Visual of the low frequency of Delta scores, the
Anatomical form of the complete surface Visual and probe
Fisher exact test was used to compute the
Anatomical form at the marginal step Visual and probe
Marginal integrity Visual and probe distribution of clinically acceptable (Alfa and
Discoloration of the margin Visual Bravo) versus unacceptable (Charlie and
Integrity of the tooth Visual and probe Delta) restorations.
Integrity of the restoration Visual and probe
Occlusion Visual (articulating paper)
Testing of sensitivity Thermal testing (CO2 ice)
Postoperative symptoms Interviewing the patient
RESULTS
Artglass and eight Charisma inlays had to be Determination of the interexaminer reliability
removed up to the 2-year recall. These failed yielded kappa values above 0.64 for all rated
restorations are included in the 112 rated criteria except “color match,” which revealed
inlays. Criteria listed in Table 4 were assessed only a low initial agreement between the
using modified USPHS criteria for the direct raters (kappa value = 0.30).
evaluation of the adhesive technique.12–14 Results of the clinical evaluation compar-
This assessment resulted in ordinally struc- ing Artglass and Charisma indirect compos-
tured data for the outcome variables (Alfa = ite resin inlays at baseline; 6-month; and 1-,
excellent result; Bravo = acceptable result; 2-, and 3-year follow-up appointments are

404 VOLUME 41 • NUMBER 5 • MAY 2010

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

Ta b l e 5 Artglass composite resin inlays: Results of the clinical evaluation (modified USPHS
scores, %) at baseline; 6-month; and 1-, 2-, and 3-year follow-up

Baseline (n = 75) 6 mo (n = 75) 1 y (n = 70) 2 y (n = 64) 3 y (n = 49)

Criteria A B A B C D A B C D A B C D A B C D

Surface texture 100 0 100 0 0 0 100 0 0 0 97 3 0 0 92 8 0 0

Color match 99 1 99 1 0 0 99 1 0 0 97 3 0 0 96 4 0 0
Anatomical form of 100 0 100 0 0 0 100 0 0 0 100 0 0 0 100 0 0 0
the complete surface
Anatomical form at the 100 0 100 0 0 0 96 4 0 0 94 6 0 0 94 6 0 0
marginal step
Marginal integrity 100 0 96 4 0 0 90 10 0 0 67 33 0 0 63 37 0 0
Discoloration of the margin 100 0 87 13 0 0 61 39 0 0 48 52 0 0 41 59 0 0
Integrity of the tooth 100 0 100 0 0 0 100 0 0 0 98 2 0 0 94 6 0 0
Integrity of the restoration 100 0 100 0 0 0 99 0 1 0 95 2 3 0 92 4 4 0
Occlusion 100 0 99 1 0 0 97 3 0 0 95 5 0 0 98 2 0 0
Testing of sensitivity 100 0 100 0 0 0 99 0 1 0 98 0 2 0 98 0 2 0
Postoperative symptoms 96 4 86 11 0 3 86 10 0 4 84 11 0 5 86 8 0 6

(A) Alfa, (B) Bravo, (C) Charlie, (D) Delta.

Ta b l e 6 Charisma composite resin inlays: Results of the clinical evaluation (modified USPHS
scores, %) at baseline; 6-month; and 1-, 2-, and 3-year follow-up

Baseline (n = 80) 6 mo (n = 80) 1 y (n = 75) 2 y (n = 71) 3 y (n = 63)

Criteria A B A B C D A B C D A B C D A B C D

Surface texture 100 0 100 0 0 0 100 0 0 0 96 4 0 0 94 6 0 0

Color match 95 5 95 5 0 0 97 3 0 0 96 4 0 0 95 5 0 0
Anatomical form of the 100 0 100 0 0 0 100 0 0 0 99 1 0 0 97 3 0 0
complete surface
Anatomical form at the 100 0 100 0 0 0 100 0 0 0 100 0 0 0 97 3 0 0
marginal step
Marginal integrity 100 0 91 8 1 0 92 6 1 1 75 23 1 1 63 30 5 2
Discoloration of the margin 100 0 76 24 0 0 69 31 0 0 54 46 0 0 54 43 3 0
Integrity of the tooth 100 0 100 0 0 0 100 0 0 0 99 0 0 1 98 0 0 2
Integrity of the restoration 100 0 96 0 3 1 94 0 3 3 90 3 3 4 87 5 3 5
Occlusion 99 1 97 3 0 0 97 3 0 0 93 7 0 0 89 11 0 0
Testing of sensitivity 100 0 99 0 0 1 98 1 0 1 99 0 0 1 98 0 0 2
Postoperative symptoms 95 5 89 10 1 0 96 3 1 0 93 6 1 0 92 6 2 0

(A) Alfa, (B) Bravo, (C) Charlie, (D) Delta.

reported in Tables 5 and 6. The Mann- were in 12 patients, were randomly distrib-
Whitney U test exhibited no significant differ- uted with regard to the student operator.
ences in any of the clinical criteria listed in The statistical analysis of cavity-size influ-
Table 4 between Artglass and Charisma ence showed for the subgroup of small
composite resin inlays at the 3-year recall. Artglass inlays a significantly better marginal
There was a trend for better occlusal contact integrity (P = .025) and significantly less mar-
point distribution in favor of Artglass, ginal discoloration (P = .017). Small Charisma
although this was not statistically significant inlays exhibited a statistically significant bet-
(P = .066). Up to 3 years, five Artglass inlays ter performance for the “integrity of the
(Fig 2) and 10 Charisma inlays failed (Table restoration” parameter (P = .022). No signifi-
7). Main failure reasons were inlay fracture, cant differences for any of the parameters
loss of marginal integrity, secondary caries, could be detected comparing the clinical per-
and loss of tooth vitality. All restorations were formance of adhesive inlays in premolars ver-
replaced at the respective follow-up time. sus molars for either Artglass or Charisma
The 15 failed composite resin inlays, which (P > .05). The influence of the composite

VOLUME 41 • NUMBER 5 • MAY 2010 405

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

Ta b l e 7 Reasons and time of failure of Artglass and Charisma indirect

composite resin inlays*

Material/ Restoration Months after

tooth (FDI) surfaces baseline USPHS score Failure type

Artglass
16 POB 6 Delta Postoperative symptoms
24 OD 6 Delta Postoperative symptoms
15 MOD 12 Charlie + Delta Sensitivity (C) + postoperative symptoms (D)
15 OD 12 Charlie Integrity of the restoration
24 MOD 24 Charlie Integrity of the restoration
Charisma
45 OD 6 Charlie Integrity of the restoration
25 MODB 6 Charlie Integrity of the restoration
24 OD 6 Charlie Marginal integrity
47 MOD 6 Delta Integrity of the restoration
37 MODB 6 Delta + Charlie Sensitivity (D) +postoperative symptoms (C)
26 OM 12 Delta Marginal integrity
37 MOD 12 Delta Integrity of the restoration
37 MOD 24 Delta + Delta Integrity of the tooth + integrity of the restoration
15 MOD 36 Charlie + Charlie Marginal integrity + marginal discoloration
16 MOD 36 Charlie + Charlie Marginal integrity + marginal discoloration
*A complete failure resulted in total replacement of the respective restoration.
Surfaces: (O) occlusal, (M) mesial, (D) distal, (B) buccal, (P) palatal/lingual.

Fig 2 Artglass inlay (MOD) in the maxillary left first Fig 3 Artglass inlay (MOD) in the mandibular left
premolar showing bulk fracture at the transition second premolar showing minor abrasion in the
from isthmus to the mesial box. The restoration was luting gap at the buccal aspect of the isthmus
scored Charlie for “integrity of the restoration,” as (rated Alfa).
the fragment was not mobile.

resin cement used to adhesively lute the deterioration of restoration integrity (P = .013),
Artglass inlays revealed no significant influ- and a significant increase of postoperative
ence on any of the recorded clinical parame- symptoms (P = .001). Charisma inlays showed
ters (P > .05) (Fig 3). after 3 years a significant deterioration of sur-
The statistical comparison between base- face texture quality (P = .023), color match
line and 3-year results (Mann-Whitney U test) (P = .049), marginal integrity (P = .001) (Fig 4),
yielded for Artglass inlays a significant deterio- restoration integrity (P = .013), and distribution
ration of surface texture quality (P = .013) and of occlusal contact points (P = .042); a significant
anatomical form at the marginal step (P = .032), increase of marginal discoloration (P = .001)
reduction of marginal integrity (P = .001), in - (Fig 5); and postoperative symptoms (P = .001).
crease of marginal discoloration (P = .001), However, these effects are mostly results of

406 VOLUME 41 • NUMBER 5 • MAY 2010

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

Fig 4 Charisma inlay (MO) in the maxillary left Fig 5 Charisma inlay (MODP) in the maxillary left
second molar with signs of marginal imperfections first molar demonstrating first signs of marginal
still being clinically acceptable (scored Bravo for discoloration at the palatal extension but still being
“marginal integrity”). clinically acceptable (scored Bravo for “marginal
discoloration”).

Alfa-Bravo shifts, meaning that most of the DISCUSSION

composite resin inlays are still clinically accept-
able and functional, except those detailed in Composite resin inlays are indicated for the
Table 7. restoration of occlusal and proximal surface
From baseline up to 3 years, 15 restora- defects. The major advantage is that most of
tions failed and were scored Charlie or Delta the composite is formed by the precured
(see Table 7). However, seven inlays failed composite resin inlay, which is inserted in the
within the first 6 months of observation. To preparation using a minimum of resin
analyze the clinical failure rate (distribution of cement, offering good control of anatomical
Charlie- and Delta-scored versus Alfa- and form and proximal contacts.4,17,18 Postcuring
Bravo-scored restorations) for Artglass ver- the inlays can further enhance the mechani-
sus Charisma inlays, small versus large cal properties.
preparations, and premolars versus molars, Four inlays (three Artglass, one Charisma)
2 ⫻ 2 tables were created and analyzed failed due to postoperative symptoms that
using Fisher exact test. No significant differ- required endodontic therapy. The risk of
ences between composite resin materials postplacement hypersensitivity has been
(P = .265), cavity size (P = .111), and tooth attributed to the method of luting and could
type (P = .134) could be detected concerning be significantly reduced by improved bond-
the failure rate. Analyzing the influence of the ing systems and resin cements, let alone the
two sresin cements on the failure rate of meticulous use of recommended techniques
Artglass inlays with Fisher exact test showed and avoidance of tooth desiccation. While in
no significant influence from the luting mate- 1990, up to 16% of hypersensitivity could be
rial (P = .200). observed with adhesive restorations,19 these
Failure rates for Artglass and Charisma figures have decreased significantly with an
inlays at the 3-year recall were 10.2% and incidence of 0% to 3% today.20 Many cases
15.9%, respectively, giving an annual failure of postoperative sensitivity resolve several
rate of 3.4% and 5.3%, respectively. weeks after restoration placement.1,21 But
there are still a number of teeth that require

VOLUME 41 • NUMBER 5 • MAY 2010 407

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

operative treatment up to vital extirpation to Parallel to others, this study found no sig-
combat symptoms and causes of postopera- nificant differences between premolars and
tive hypersensitivity.22,23 molars for any of the evaluated clinical
Bulk fracture is considered to be one of the parameters.1,29–31 However, several other
most frequent causes for restoration failure.20,24 reports indicate that premolars offer more
It can be caused by weak material properties, favorable conditions for the survival of indi-
such as insufficient polymerization rate of the rect composite resin restorations than
inlay composite resin material or insufficient molars.18,24,32–35 A premolar restoration is usu-
material thickness.25 In the present study, two ally subjected to much less occlusal stress
Artglass inlays (4%) and five Charisma inlays than a molar restoration, the access for den-
(8%) had to be replaced because of fracture, tal treatment is easier, and oral hygiene
the results being not significantly different. measures are more easily controlled by the
Four Charisma inlays (7%) had to be patient. Donly et al18 reported failures due to
replaced because of deep marginal openings secondary caries and fractures predominant-
in two cases combined with secondary caries ly in molar restorations.
formation. It has been suggested that an Artglass inlays showed a significantly bet-
increase in marginal gap size may result in ter marginal integrity and significantly less
degradation of the adhesive bond, in turn marginal discoloration, and Charisma inlays
leading to microleakage and secondary exhibited a significantly better inlay integrity in
caries.21 Secondary caries is the most fre- small preparations (one and two surfaces)
quently cited reason for failure of dental compared to large preparations (three and
restorations in general practice26 and repre- more surfaces). Because of the elastic behav-
sents up to 50% of all operative dentistry pro- ior of the composite resin, differences in coef-
cedures delivered to adults.27 In this study, ficient of thermal expansion between tooth
both inlay systems experienced significant and restoratives, and fatigue of the composite
deterioration of marginal integrity (P = .001) resin and bonding agent, negative influences
and significant increase of marginal discol- of occlusal stress factors on posterior teeth
oration (P = .001) when baseline and 3-year are discussed to be more crucial for large
data were compared. Margin wear also influ- restorations and molars, which are usually
ences marginal quality.1 The present results subjected to higher occlusal loading and
show a significant decrease of anatomical stresses at the restoration-tooth interface.
form at the marginal step (P = .032) for Barone et al1 could not detect significant dif-
Artglass inlays after 3 years. Loss of marginal ferences for composite resin inlays placed in
integrity of composite resin inlays can be one- or two-surface preparations compared to
caused at baseline by polymerization shrink- multisurface inlays after 3 years, except for the
age, deficits of resin cement application, or its parameter marginal integrity. This is consis-
faulty adaptation to cavity walls. Bravo ratings tent with the findings for Artglass inlays, which
were caused by marginal opening due to exhibited a significantly better marginal
adhesive failures during clinical service. integrity (P = .025) in small preparations.
Artglass and Charisma inlays had a significant Leirskar et al31 reported a significantly higher
change in surface texture after 3 years, com- success rate for two-surface composite resin
parable to other studies.1,28 Between baseline inlays compared to three-surface inlays and
and 3-year follow-up, a significant deteriora- resin-based onlays after 5 years.
tion for the parameters “surface texture quali- Clinical treatment needs to be based on
ty,” “anatomical form at the margin,” “margin- “confirmed clinical evidence.”10 Practice-
al integrity,” “marginal discoloration,” “integrity based research must be the future of clinical
of the restoration,” “postoperative symptoms,” research, focusing on projects rooted in gen-
and “color match” could be observed for eral dental practice and involving clinicians as
either one or both of the tested materials. practitioner-investigators to establish a link
According to Hickel et al,8 these alterations between treatment outcomes in everyday
usually occur in a medium or long-term time dental practice with experienced clinical
frame from insertion of the restorations. investigators.10 The present study employed

408 VOLUME 41 • NUMBER 5 • MAY 2010

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

carefully supervised dental students of the amalgam, 2.2% for direct composite resin
third clinical training period for the placement restorations, 1.9% for ceramic inlays, and
of the restorations.31,35,36 This design intro- 1.4% for gold inlays were reported.20 The
duced an additional variable by the relatively results of this study reveal a slightly higher
large number of operators placing the com- annual failure rate for Artglass inlays (3.4%)
posite resin inlays. However, the students but a distinctly higher annual failure rate for
were thoroughly trained in adhesive dentistry Charisma inlays (5.3%). Assuming the fail-
(under same conditions) since the beginning ures within the first 6 months result from
of their studies in theoretical lectures, prac- severe treatment faults,8 a second set of
tice-based hands-on trainings, and two pre- annual failure rates, which give a more mate-
ceding clinical courses. On the other hand, rial-based approach, could be calculated to
this approach allowed simulating everyday be 2.1% for Artglass [1 - (44/47) ⫻ 1/3] and
clinical practice during restoration placement 2.9% for Charisma [1 - (53/58) ⫻ 1/3] when
in combination with the professional evalua- the early failure cases were removed from the
tion of the composite resin inlays by research- calculation.
experienced dental faculty clinicians.
Longevity of dental restorations depends
on many factors that are patient-, material-,
and clinician-related.37 It has to be distin- CONCLUSION
guished between early failures (after weeks or
a few months), failures in a medium time Artglass and Charisma inlays showed an
frame (6 to 24 months), and late failures (after annual failure rate of 3.4% and 5.3%, respec-
2 years).8 Early failures are a result of severe tively, which is in the range of 0% to 10%
treatment faults, selecting an incorrect indica- reported in a comprehensive meta-analysis.
tion, allergic/toxic adverse effects, or postop- Although the restorations were placed by rel-
erative symptoms. Failures in a medium time atively inexperienced student operators, the
frame are typically attributed to cracked tooth acceptable survival rate qualifies the indirect
syndrome or tooth fracture, marginal discol- composite inlays as a competitive restorative
oration, restoration staining, chipping, and procedure in stress-bearing preparations.
loss of vitality.8 Late failures are predominantly
caused by bulk and tooth fractures, second-
ary caries, wear or material deterioration, or
periodontal adverse effects.20 In this study, 7 ACKNOWLEDGMENT
(2 Artglass and 5 Charisma) of 15 composite
resin inlays failed within 6 months. These early The authors would like to express their gratitude to Dr
Petra Neuerer and Dr Andrea Scheibenbogen for their
failures were caused by severe postoperative
participation in the clinical study. This study was spon-
symptoms (n = 3), bulk fractures (n = 3), and sored in part by Heraeus-Kulzer, Wehrheim, Germany.
deep marginal openings (n = 1) (see Table 7). The authors state that they have no conflict of interest.
Probably, these early failures can be attributed
to the relative shortage of experience of the
student operators. All these three failure types
might be a symptom of problems during the REFERENCES
adhesive luting procedure.
Artglass and Charisma composite resin 1. Barone A, Derchi G, Rossi A, Marconini S, Covani U.
inlays showed a success rate of 89.8% and Longitudinal clinical evaluation of bonded com-
84.1% after 3 years. The results of a compre- posite inlays: A 3-year study. Quintessence Int 2008;
39:65–71.
hensive meta-analysis on posterior restora-
2. Kelsey WP, Cavel WT, Blankenau RJ, Barkmeier WW,
tions demonstrate annual failure rates for
Wilwerding TM, Latta MA. 4-year clinical study of
posterior composite resin inlays and onlays in castable ceramic crowns. Am J Dent 1995;8:259–262.
a range from 0% to 10% with a mean value of 3. Hickel R, Kunzelmann KH. Keramikinlays und
2.9% (median 2.3%); for alternative restora- Veneers. München: Hanser-Verlag, 1997.
tions mean annual failure rates of 3.0% for

VOLUME 41 • NUMBER 5 • MAY 2010 409

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Q U I N T E S S E N C E I N T E R N AT I O N A L
Manhar t et al

4. Burke FJT, Qualtrough AJE. Aesthetic inlays: 22. Krämer N, Kunzelmann KH, Mumesohn M, Pelka M,
Composite or ceramic. Br Dent J 1994;176:53–60. Hickel R. Langzeiterfahrungen mit einem mikroge-
5. Davidson CL. Posterior composites: Criteria for assess- füllten Komposit als Inlaysystem. Dtsch Zahnärztl Z
ment. Introduction. Quintessence Int 1987;18:515. 1996;51:342–344.

6. Freilich MA, Goldberg AJ, Gilpatrick RO, Simonsen 23. Wassell RW, Walls AWG, McCabe JF. Direct compos-
RJ. Direct and indirect evaluation of posterior com- ite inlays versus conventional composite restora-
posite restorations at three years. Dent Mater 1992; tions: Three-year clinical results. Br Dent J 1995;
8:60–64. 179:343–349.

7. Ryge G, Cvar JF. Criteria for the clinical evaluation of 24. Pallesen U, Qvist V. Composite resin fillings and
dental restorative materials. US Dental Health inlays. An 11-year evaluation. Clin Oral Investig
Center, publication 7902244, 1971. San Francisco: 2003;7:71–79.
US Government Printing Office, 1971. 25. Martin N, Jedynakiewicz NM. Clinical performance
8. Hickel R, Roulet JF, Bayne S, et al. Recommendations of Cerec ceramic inlays: A systematic review. Dent
for conducting controlled clinical studies of dental Mater 1999;15:54–61.
restorative materials. Clin Oral Investig 2007;11:5–33. 26. Mjör IA, Moorhead JE, Dahl JE. Reasons for replace-
9. Botelho MG, Chan AW, Yiu EY, Tse ET. Longevity of ment of restorations in permanent teeth in general
two-unit cantilevered resin-bonded fixed partial dental practice. Int Dent J 2000;50:361–366.
dentures. Am J Dent 2002;15:295–299. 27. Mjör IA, Toffenetti F. Secondary caries: A literature
10. Mjör IA. A recurring problem: Research in restora- review with case reports. Quintessence Int
tive dentistry . . . but there is a light at the end of the 2000;31:165–179.
tunnel. J Dent Res 2004;83:92. 28. Thordrup M, Isidor F, Hörsted-Bindslev P. A 5-year clin-
11. Moher D, Schulz KF, Altman DG. The CONSORT ical study of indirect and direct resin composite and
Statement: Revised recommendations for improv- ceramic inlays. Quintessence Int 2001;32:199–205.
ing the quality of reports of parallel-group random- 29. Haas M, Arnetzl G, Wegscheider WA, Konig K,
ized trials. Ann Intern Med 2001;134:657–662. Bratschko RO. Klinische und werkstoffkundliche
12. Ryge G. Clinical criteria. Int Dent J 1980;30:347–358. Erfahrungen mit Komposit-, Keramik- und
Goldinlays. Dtsch Zahnarztl Z 1992;47:18–22.
13. Ryge G, Snyder M. Evaluating the clinical quality of
restorations. J Am Dent Assoc 1973;87:369–377. 30. Wiedmer CS, Krejci I, Lutz F. Klinische, röntgenologis-
che und rasterelektronenoptische Untersuchung
14. Ryge G, Stanford JW. Recommended format for pro-
von Kompositinlays nach fünfjähriger Funktionszeit.
tocol of clinical research program: Clinical compari-
Acta Med Dent Helv 1997;2:301–307.
son of several anterior and posterior restorative
materials. Int Dent J 1977;27:46–57. 31. Leirskar J, Nordbo H, Thoresen NR, Henaug T, der
Fehr FR. A four to six years follow-up of indirect
15. Feller RP, Ricks CL, Matthews TG, Santucci EA. Three-
resin composite inlays/onlays. Acta Odontol Scand
year clinical evaluation of composite formulations
2003;61:247–251.
for posterior teeth. J Prosthet Dent 1987;57:544–550.
32. Fuzzi M, Rappelli G. Survival rate of ceramic inlays.
16. Ryge G, Jendresen MD, Glantz PO, Mjör IA.
J Dent 1998;26:623–626.
Standardization of clinical investigators for studies
of restorative materials. Swed Dent J 1981;5: 33. Geurtsen W, Schoeler U. A 4-year retrospective clin-
235–239. ical study of Class I and Class II composite restora-
tions. J Dent 1997;25:229–232.
17. Bessing C, Lundqvist P. A 1-year clinical examination
of indirect composite resin inlays: A preliminary 34. Rykke M. Dental materials for posterior restorations.
report. Quintessence Int 1991;22:153–157. Endod Dent Traumatol 1992;8:139–148.

18. Donly KJ, Jensen ME, Triolo P, Chan D. A clinical com- 35. Scheibenbogen-Fuchsbrunner A, Manhart J,
parison of resin composite inlay and onlay posteri- Kremers L, Kunzelmann KH, Hickel R. Two-year clini-
or restorations and cast-gold restorations at 7 years. cal evaluation of direct and indirect composite
Quintessence Int 1999;30:163–168. restorations in posterior teeth. J Prosthet Dent
1999;82:391–397.
19. Hickel R. Zur Problematik hypersensibler Zähne
nach Eingliederung von Adhäsivinlays. Dtsch 36. Manhart J, Chen HY, Neuerer P, Scheibenbogen-
Zahnärztl Z 1990;45:740–742. Fuchsbrunner A, Hickel R. Three-year clinical evalua-
tion of composite and ceramic inlays. Am J Dent
20. Manhart J, Chen H, Hamm G, Hickel R. Buonocore
2001;14:95–99.
Memorial Lecture. Review of the clinical survival of
direct and indirect restorations in posterior teeth of 37. Hickel R. Glass ionomers, cermets, hybrid ionomers
the permanent dentition. Oper Dent 2004;29: and compomers—(Long-term) clinical evaluation.
481–508. Trans Acad Dent Mater 1996;9:105–129.

21. Fasbinder DJ, Dennison JB, Heys DR, Lampe K. The

clinical performance of CAD/CAM-generated com-
posite inlays. J Am Dent Assoc 2005;136:1714–1723.

410 VOLUME 41 • NUMBER 5 • MAY 2010

© 2009 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART OF THIS ARTICLE
MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.