Microleakage of composite restorations after acid or Er-YAG lasercavity treatments

L. Ceballosa, R. Osorioa,*, M. Toledanoa, G.W. Marshallb

aDepartment of Dental Materials, University of Granada, Granada, SpainbDivision of Biomaterials and Bioengineering, Department of Preventive and Restorative Dental Sciences, University of California, San Francisco, CA, USA

Received 1 February 2000; revised 4 July 2000; accepted 28 September 2000

Abstract

Objectives: The purpose of this study was to compare microleakage of Class V restorations following acid, laser or laser and acid treatment

of cavity walls.

Methods: Standardized lingual and buccal Class V preparations were made in 18 human extracted third molars. The preparations were

randomly assigned to three equal groups (n� 12). Group 1: cavities were treated with 35% phosphoric acid. Group 2: cavities were irradiated

with an Er-YAG laser at 2 Hz and 250 mJ on dentin and 300 mJ on enamel, with water cooling. Group 3: cavities were irradiated with the

laser before acid etching. Scotchbond 1 Adhesive System and Z100 resin composite were used for restorations. The specimens were stored in

water for 24 h at 378C and thermally cycled (500 £ ) between 6±608C. After 24 h immersion in 0.5% basic fuchsin, dye penetration was

recorded according to an ordinal scale. Data were analysed using non-parametric statistical tests (Kruskal±Wallis and Mann±Whitney).

Results: On the occlusal walls, microleakage in acid etched cavities was signi®cantly lower than that achieved after laser treatment

(P , 0.001) or after both treatments (P , 0.05). On the gingival walls, no statistical differences were found.

Signi®cance: Laser irradiation of enamel is not a valid alternative to acid-etching pretreatment for resin composite materials adhesion.

Acid etching alone gave the lowest microleakage at the occlusal margin. No differences were found for microleakage on gingival wall,

although lased dentin surfaces presented several characteristics that appear to be advantageous for bonding. q 2001 Academy of Dental

Materials. Published by Elsevier Science Ltd. All rights reserved.

Keywords: Microleakage; Dental adhesion; Etch; Composite; Lasers

1. Introduction

The ruby laser was developed by Maiman in 1960 and

since then the effects on dental hard tissues of different laser

systems, including the ruby [1], Nd-YAG [2] and CO2 [3]

have been investigated. These laser irradiations showed an

insuf®cient ability to cut dental hard tissues as they required

relatively high energy densities to vaporize dentin and

enamel, and caused major thermal side-effects, such as

melting, cracking or charring of these tissues and pulpal

damage [4].

The development of a new solid state laser, the Erbium-

YAG laser, provided a wavelength (l � 2.94 mm) which

matched the maximum absortion wavelength of water.

This laser, in contrast to many other available lasers, had

the ability to remove dentin and enamel more effectively

and ef®ciently [5±7], with little thermal damage [8±11],

specially in a ®eld with water spray [12,13].

The Er-YAG laser has been used clinically to prepare

cavities reporting that patients perceived it as a more

comfortable method than a bur, with a signi®cantly reduced

need for local anaesthesia [14±16].

The use of this laser to modify the surfaces of teeth to

improve bonding of restorations has also been studied

[17,18]. Several characteristics of lased dental hard tissues

have been considered advantageous: microscopically rough

surfaces without demineralization, open dentinal tubules with-

out smear layer production and dentin surface sterilization.

The purpose of this study was to compare microleakage of

Class V composite restorations following acid etching or an

Er-YAG laser treatment or both, laser and acid etching treat-

ments, in order to evaluate if this laser might substitute for

acid-etching as a pretreatment for resin composite bonding.

Dental Materials 17 (2001) 340±346

dentalmaterials

0109-5641/01/$20.00 + 0.00 q 2001 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved.

PII: S0109-5641(00)00092-0

www.elsevier.com/locate/dental

* Corresponding author. Tel.: 134-958243788; fax: 134-958244085.

E-mail address: toledano@ugr.es (R. Osorio).

2. Materials and methods

Eighteen freshly extracted, intact, caries-free human third

molars were used in this study. They were hand-scaled,

cleaned and stored in saline solution for up to 1 month at

48C.

A Class V cavity was prepared, using a kidney-shaped

template, in the buccal and lingual surfaces of each tooth,

with the gingival margin placed at the cementoenamel junc-

tion. Preparations were made with a 329 carbide bur in a

high-speed handpiece equipped with water spray. Cavities

measured 5 mm long, 3 mm wide and 2 mm deep. The

occlusal margin was beveled at 458 using a ®ne grit diamond

stone. The width of the bevel was 0.5 mm.

All the cavities were prepared with the high-speed hand-

piece before use of the Er-YAG laser and/or acid to modify

the enamel and dentin surfaces.

The specimens were randomly assigned to three equal

groups (n� 12).

2.1. Group 1

Enamel and dentin were etched with 35% phosphoric acid

gel (Scotchbond etchant, 3M Dental Products, St. Paul, MN,

USA) for 15 s and rinsed for 10 s. The dentin surface was

then blot-dried.

2.2. Group 2

Enamel and dentin surfaces were conditioned using a

pulsed Er-YAG laser (KaVo K.E.Y. Laserw, no. 002532,

KaVo, Biberach, Germany; System Aesculap Meditec,

Jena, Germany) with a wavelength of 2.94 mm, a pulse

duration of 250 ms and water cooling. Following manu-

facturer' instructions, a pulse energy of 300 mJ was used

to treat enamel and of 250 mJ was used to irradiate the

dentin. The repetition rate chosen was 2 Hz. The laser

beam was delivered through an articulated arm with a 908handpiece in a non-contact mode and the spot size was

about 0.7 mm in diameter. The laser beam was not focused,

the working distance was higher than 15 mm to reduce abla-

tion [19]. Cavity laser treatment took about 2 min.

The surface laser treatment was carried out moving the

handpiece continously above the cavity, in order to obtain a

pattern of rows and columns that overlapped. It was veri®ed

that dentin and enamel cavity surfaces were homogeneously

and completely lased before further treatment using a

stereomicroscope (Olympus SZ-4045TR Co, Tokyo, Japan).

2.3. Group 3

The surface was treated with the laser as previously

described and after that, etched with phosphoric acid for

15 s and rinsed for 10 s.

All the samples then underwent the same bonding proce-

dure with Scotchbond 1 Adhesive System (lot: 19970313,

exp: 2000-03; 3M, Dental Products. St Paul, MN, USA). A

wet-bonding technique was followed as recommended by

the manufacturer, moist dentin was clinically evidenced by

a uniform shiny surface on which water was not pooled. A

fully saturated brush tip for each coat was used, applying

two consecutive coats of Scotchbond adhesive to prepared

enamel and dentin. Later, the surface was dried gently for

2±5 s and light cured for 10 s (Optilux 400, Demetron

Research Corp, Danbury, CT). The light was tested for

light output (.600 mW/cm2) before each use with a Deme-

tron radiometer (model 100, Demetron Research Corp).

Cavities were restored with a hybrid resin composite

(Z100-A3 shade, lot: 19991120, exp: 2002-10; 3M, Dental

Products. St Paul, MN, USA) in two increments with the

®rst against the gingival wall, and light-cured for 40 s.

Excess materials were removed with a No. 170 bur,

followed by ®nishing and polishing with the Sof-lex disk

system (3M Dental Products Division, St Paul, MN, USA).

The restored teeth were stored in distilled water for 1 day

before further treatment. Then they were thermally cycled

for 500 cycles between 6 and 608C with a dwell time of 30 s,

before immersion in dye [20].

The apices of the teeth were sealed with zinc oxide euge-

nol cement and coated with a nail polish 1 mm short of the

restoration margins in order to reduce other leakage else-

where that could lead to false positive results.

The crowns of the teeth were immersed in 0.5% solution

of basic fuchsin for 24 h at room temperature.

The super®cial dye was removed with a pumice slurry

and rubber cup after removal of the specimens from the dye

solution. Teeth were then mounted in a light-curing one

component methacrylate-based resin (Technovit 7200

VLC, Kulzer, Norderstedt, Germany) to facilitate handling

during sectioning. The resin was cured for 24 h (Histolux,

EXAKT, Norderstedt, Germany), then teeth were sectioned

longitudinally with a hard tissue microtome (Exakt-appara-

tebau, Otto Herrman, Norderstedt, Germany) into 0.6 mm

thick sections to evaluate the dye penetration. The sections

were then separated, and the cut surfaces corresponding to

the most mesial, central, and most distal portion of the tooth

restoration interface were examined at the occlusal and

gingival margins with a stereomicroscope (Olympus

SZ-4045TR Co, Tokyo, Japan) at £ 16 magni®cation.

Examination of the specimens was undertaken blindly by

two observers who were unaware of the exact nature of the

restorative treatment evaluated. Consensus was obtained

between obsevers if there were con¯icts in scores.

The staining along both, occlusal or gingival, tooth

restoration interfaces was recorded according to the follow-

ing criteria:

0, no dye penetration

1, dye penetration along the interface to 1/2 the depth of

the cavity wall

2, dye penetration to the full depth of the cavity wall, but

not including the axial wall

3, penetration to and along the axial wall.

L. Ceballos et al. / Dental Materials 17 (2001) 340±346 341

After the evaluation of the dye penetration in all the cut

surfaces, the section with the maximum leakage value

recorded for each cavity was selected for the analysis,

even though the relation of leakage patterns to disease is

currently obscure.

Occlusal and gingival scores for each group of restora-

tions were compared with the Kruskal±Wallis one-way

analysis of variance (ANOVA) non-parametric statistical

test to identify any statistically signi®cant differences

between the three procedures. The Mann±Whitney U test

was performed to compare each matched pair of condition-

ing treatment and also to compare site within the specimens.

Signi®cance was considered at the 0.05 level.

2.4. Electron microscopy

Two sections of each group were used for analysis by

scanning electron microscopy (SEM). After examination of

the specimens with the stereomicroscope, an individual

impression of each interface was taken with an elastomeric

material (Aquasil LV, Dentsply Caulk, Milford, DE), the

impressions were poured up in epoxy resin (Epo-thin,

Buehler Ltd, Lake Bluff, IL). The sections were polished

with waterproof papers of decreasing abrasiveness up to

1200 grit. After polishing, surfaces were treated with phos-

phoric acid 37% for 30 s, half of them were also treated with

sodium hypoclhorite 10% during 2 min. These treatments

were done in order to completely eliminate the smear layer

and to provide evidence of in®ltration of the resin into dentin.

Sections and the epoxy casts were desiccated for 48 h

(Sample Dry Keeper Samplatec Corp., Japan) and then

mounted on aluminum stubs with carbon cement. They

were then sputter-coated with gold by means of a sputter-

coating Unit E500 (Polaron Equipment Ltd., Watford,

England) and observed under a ZEISS DSM-950 (Carl

Zeiss, Germany) scanning electron microscope at an accel-

erating voltage of 20 kV and a working distance of 13±

16 mm. The epoxy casts were observed under the SEM

and compared to the actual dentin-resin sections to control

for artefact formation.

3. Results

None of the procedures tested on this study completely

eliminated microleakage. The data showing the extent of

leakage scored for the occlusal and gingival portions of

the restorations are shown in Table 1. For the occlusal

margins all 12 sites were rated 0 or 1 for the group 1 (acid

treatment), while eight of 12 for the group 2 (laser treat-

ment) and 10 of 12 for the group 3 (laser and acid treat-

ments) had either 0 or 1. For the gingival walls, only two

sites were rated 0 for the group 1 and most of the restora-

tions showed scores 2 or 3: 10 of 12 for group 1, all of the

group 2 and 10 of 12 for the group 3.

Kruskal±Wallis one-way ANOVA indicated signi®cant

differences among the three different procedures for occlu-

sal scores (P , 0.001).

Further matched analysis by the Mann±Whitney U test

was undertaken to compare occlusal site scores of each

material which revealed statistically signi®cant differences

between group 1 and group 2 (P , 0.001), and between

group 1 and 3 (P , 0.05). In both cases, acid etching treat-

ment (group 1) revealed the least dye penetration scores.

There was no statistical difference in microleakage between

performing laser treatment alone (group2) and giving both

laser and acid treatments (group 3).

On the gingival wall, the Kruskal±Wallis test did not

show statistically signi®cant differences in microleakage

among the three different treatments (p . 0.05).

When comparing occlusal and gingival leakage in each

procedure, the Mann±Whitney U test indicated a statisti-

cally signi®cant greater leakage at the gingival wall for all

the treatments (group 1 P , 0.001, group 2 P , 0.001, and

group 3 P , 0.001).

3.1. SEM results

The micrographs of Figs. 1±6 present the main SEM

®ndings. As Fig. 1 shows, no gap between the adhesive

and the enamel was found in acid etched enamel specimens.

Resin tags were visible. Specimens that had been irradiated,

whether acid etched or not (Figs. 2 and 3), presented gaps

between the adhesive and the enamel that alternated with

L. Ceballos et al. / Dental Materials 17 (2001) 340±346342

Table 1

Microleakage scores obtained for each experimental group (n� 12)

Occlusal margin Gingival margin

0 1 2 3 0 1 2 3

Acid etched 11 1 0 0 2 0 1 9

Laser etched 1 7 2 2 0 0 1 11

Laser 1 acid etched 5 5 1 1 0 2 2 8

Fig. 1. Enamel-resin interface after etching with 35% H3PO4. Note the tight

interface with no visible gap formation ( £ 1000); polished and acid treated

SEM specimen.

small zones without gaps. In Figs. 2 and 3, no mechanical

interlocking is observed between enamel and resin.

For specimens that had been etched, laser irradiated or

not, a hybrid layer could be observed along the entire inter-

face and opened tubules were found to have been penetrated

by adhesive resin. In Fig. 4(a) an homogeneous hybrid layer

formation can be observed, as the gap is an artefact forma-

tion. Long and thin resin tags were also detected and un®lled

lateral branches were found mainly around the base of the

resin tags (Fig. 4(b)). This lateral in®ltration of resin into the

surrounding demineralized dentin ®rmly attached and inte-

grated the resin tags to the hybrid layer. For laser irradiated

specimens, resin tags were clearly evident, but there was no

hybrid layer formation (Fig. (5)a and (b)).

When specimens were acid etched after Er-YAG irradia-

tion (Fig. 6(a) and (b)), resin tag formation was observed. In

Fig. 6(a) a great number of resin tags inside the tubules were

visible. In Fig. 6(b) resin tags were shorter and not so abun-

dant. In addition, hybrid layer formation can be observed.

In all the specimens some areas of interfacial gap were

observed. Such gaps were more frequently encountered in

the axial regions and represented approximately half of the

whole cavities' interface lengths.

4. Discussion

In the present investigation all groups showed higher

leakage on the gingival than on the occlusal walls. The

reason for this difference between gingival and enamel leak-

age scores is that bonding to dentin is much more technique

and substrate-sensitive than bonding to enamel. There is no

L. Ceballos et al. / Dental Materials 17 (2001) 340±346 343

Fig. 2. Enamel-resin interface after Er-YAG laser irradiation. No mechan-

ical interlocking was observed ( £ 1000); polished and acid treated SEM

specimen.

Fig. 3. Enamel-resin interface after Er-YAG laser irradiation and etching

with 35% H3PO4 ( £ 1000); polished and acid treated SEM specimen.

Fig. 4. (a) Dentin-resin interface after etching with 35% H3PO4. Note the presence of long resin tags and hybrid layer formation. After comparing with the

epoxy cast replica the gap is an artefact formation ( £ 1000); polished, acid and sodium hypoclhorite treated SEM specimen. (b) Dentin±resin interface after

etching with 35% H3PO4. Note long resin tags and adhesive lateral branches presence ( £ 2000); polished, acid and sodium hypoclhorite treated SEM

specimen.

guarantee that bonding to dentin is as durable as to enamel

[21].

As leakage at the cervical margins of Class V resin

composite restorations is nearly always observed, the ability

of bonding systems to hybridize cementum must be ques-

tioned. The literature includes only one report of hybrid

layer formation in cementum [22]. Cagidiaco et al. [23]

suggested that the leakage observed at cervical margins

may be related to the absence of dentine tubules in the

limiting 100 mm of the cervical margin, the relatively low

number of tubules in the ®rst 200±300 mm of the gingival

¯oor, and the mainly organic nature of the dentin substrate.

Enamel, when present at the cervical margin, is usually thin,

aprismatic, and bonds less well to resins. When polymer-

ized, the resin composite shrinks towards the stronger bond

at the occlusal margin and so pulls away from the weaker

bond at the gingival margin [24].

Occlusal margins that received only an acid treatment

(group 1) showed a nearly complete absence of leakage in

contrast to those treated with laser (group 2) or laser and

acid (group 3). These results con®rm the strength of the

enamel bond produced by acid conditioning [22].

The Er-YAG laser produces microexplosions during

hard tissue ablation that results in macroscopic and

microscopic irregularities. The Er-YAG laser initially

causes vaporization of water and other hydrated organic

components of the tissue. On vaporization, internal pres-

sure builds within the tissue until the explosive destruc-

tion of inorganic substance occurs before the melting

point is reached [11].

These microirregularities make the enamel surface

microretentive and may constitute the mechanism of adhe-

sion when we do not apply an acid-etchant. As Niu et al.

[25] have reported the cavity margins at enamel appear

L. Ceballos et al. / Dental Materials 17 (2001) 340±346344

Fig. 5. (a) Dentin±resin interface after Er-YAG laser irradiation. Note the resin tags formation ( £ 1500); polished and acid treated SEM specimen.

(b): Dentin±resin interface after Er-YAG laser irradiation ( £ 2000); polished and acid treated SEM specimen.

Fig. 6. (a) Dentin±resin interface after Er-YAG laser irradiation and etching with 35% H3PO4. Note the presence of many resin tags without any gap ( £ 2000);

polished and acid treated SEM specimen. (b) Dentin±resin interface after Er-YAG laser irradiation and etching with 35% H3PO4. Note the resin tags and

hybrid layer formation ( £ 1000); polished, acid and sodium hypoclhorite treated SEM specimen.

whitened when the Er-YAG laser is applied, with a similar

appearance to the acid etching effect.

The lased beveled enamel margin tends to be irregular

and this situation could affect the sealing ability of resin

restorations (Fig. 2). In addition, the crater-like defects

that can be observed in the beveled enamel margin are easily

seen under the composite restoration at this level, which

could offer an unesthetic effect.

One of the reasons for the superior seal of the acid etched

enamel group, even when we combined this method with

laser irradiation, could be the better marginal adaptation

between the composite and enamel cavity wall (Figs. 1±

3). It is likely that a high shear bond strength, as expected

from etched enamel could better compensate the polymer-

ization shrinkage of resin composite [24].

Laser irradiation seems to be associated with more leak-

age than acid treatment, under the conditions of this experi-

ment, and is not advised for promoting a bond between resin

composite materials and enamel.

On the gingival margin none of the procedures tested in

this study completely eliminated microleakage. Statistically

signi®cant differences among the three different bonding

procedures were not found and this is consistent with the

results obtained by Wright et al. [18] and Niu et al. [25].

The application of an acid etchant produces deminerali-

zation of intertubular and peritubular dentin, resulting in a

demineralized collagen matrix. In order to create a hybrid

layer the resin has to penetrate in between these collagen

®bers and reach the undemineralized dentin surface [26]

(Fig. 4(a)).

If the surface is dried too much after washing of the

etchant acid the collagen ®bers collapse, and could obstruct

resin penetration so hybrid layer formation may be affected

[21]. Therefore, a wet bonding technique was used here. The

formation of the hybrid layer, resin tags and adhesive lateral

branches are supposed to be essential to establish a strong

bond between resin and dentin (Fig. 4(b)). This mechanism

requires the complete dissolution of the smear layer by

applying a total etch technique.

One of the advantages attributed to the use of the Er-YAG

laser as a conditioner of dentin is the absence of smear layer

production due to its thermo-mechanical mechanism of

action (Fig. 5(a)).

When the Er-YAG laser is used to treat dentin, there is no

demineralization of its surface and no collagen matrix is

exposed which is necessary for the formation of the hybrid

layer. Visuri et al. [27], suggested that the greater presence

of peritubular dentin which has a greater mineral content

than intertubular dentin, may result in better bonding to the

dentin. In their study they obtained higher shear bond

strength of composite when it was bonded to Er-YAG

laser prepared dentin than to acid etched dentin. It is impor-

tant to note that they used a bonding system (ProBOND,

Caulk/Dentsply) in which acid etching was optional as the

primer acted as a hydrophilic wetting material whose mono-

mers bound to the dentin mineral.

Another difference between acid etchant and Er-YAG

laser actions related to dentin is their effect on the morphol-

ogy of dentinal tubules. When an acid etchant is applied the

peritubular dentin is preferentially etched, resulting in

funnel shaped openings to the tubules and this morphology

may contribute with polymerization shrinkage to pull the

tags away from the walls [28]. Er-YAG laser irradiation

produces no demineralization of the peritubular dentin and

the dentinal tubules remain open with no widening [29].

(Fig. 5(b)). Visuri et al. [27] also showed by SEM analysis

that the Er-YAG laser created open dentin tubules that

allowed for the development of resin tags, in agreement

with our SEM analysis (Fig. 5(a)).

No statistical differences were found among the different

procedures for marginal leakage in the gingival area where

dentin was the major substrate. The total etch technique is a

well known procedure that has been advocated as a safe and

effective method to achieve signi®cant adhesion to dentin.

However, lased dentin surfaces present several characteris-

tics that appear to be advantageous for adhesive resin

composite bonding including dentine sterilization, opening

of dentinal tubules, a surface with microirregularities with-

out a smeared layer that promotes micromechanical bonding

and no demineralization of peritubular and intertubular

dentin. These properties are interesting as recent articles

a®rm that the hybrid layer might not be so important for

the mechanism of adhesion between bonding material and

dentin [30]. Inai et al. [31] obtained good results bonding

dentin adhesives to collagen depleted dentin and suggested

that bonding to dentin mineral could improve the durability

of the hybrid layer.

When an acid is applied after Er-YAG laser irradiation a

microirregular surface and opened tubules are obtained, so

hybrid layer and resin tag formation occurs (Fig. 6(a) and

(b)). This situation could positively in¯uence the adhesion

between dentin and resin, although no differences were

found for microleakage with the other two procedures.

It would be very interesting to study the Er-YAG laser

effect in sclerotic dentin [29], that is present in non-carious

cervical lesions, as demineralization is more dif®cult in both

the peritubular and intertubular regions as dentinal tubules

are obliterated by a mineral substance [28].

It should be noted that only one adhesive system and one

composite resin were tested. An important factor that in¯u-

ences leakage is the composite and adhesive nature. The

same reference composite was chosen in order to compare

performance of the adhesive techniques. However, adhesion

is a multifaceted problem and is not suf®ciently understood,

so that further studies of shear bond strength of acid etched

and lased treated dentin applying different adhesive systems

bonded to dry and wet dentin should be conducted.

The results obtained from this in vitro study may not be

directly extrapolated to the clinical situation as independent

long-term clinical data are required before general applica-

tion of any new methods are used in routine patient treat-

ment. Prior to this, however, improved laser conditions that

L. Ceballos et al. / Dental Materials 17 (2001) 340±346 345

permit enhanced bonding on occlusal and gingival margins

of Class V preparations should be sought from further in

vitro studies. It is unlikely that methods that permit leakage

in such in vitro tests would show superior leakage resistance

in clinical trials or clinical practice.

Acknowledgements

This research project was supported by Grant MAT98-

0937-CO2 from the ComisioÂn Interministerial de Ciencia y

TecnologõÂa, Spain. This work was also partially supported

by NIH/NIDCR Grant DE11526 from the US Public Health

Service, Bethesda, MD. The authors thank Gertrudis

GoÂmez-Villaescusa for assistance in specimen preparation.

References

[1] Adrian JC, Bernier JL, Sprague WG. Laser and dental pulp. J Am

Dent Assoc 1971;83:113±7.

[2] Dederich DN, Zakariasen KL, Tulip J. Scanning electron microscopy

analysis of canal wall dentin following Nd: YAG laser irradiation. J

Endod 1984;10:428±31.

[3] Lobene RR, Bhussry BR, Fine S. Interaction of carbon dioxide laser

irradiation with enamel and dentin. J Dent Res 1968;47:311.

[4] Widgor J, Abt E, Ashra® S, Walsh JT. The effects of lasers on dental

hard tissues. J Am Dent Assoc 1993;124:65±70.

[5] Hibst R, Keller U. Experimental studies of the application of Er:YAG

laser on dental hard substances: I. Measurement of the ablation rate.

Lasers Surg Med 1989;9:338±44.

[6] Keller U, Hibst R. Experimental studies of the application of the

Er:YAG laser on dental hard substances: II. Light microscopic and

SEM investigations. Lasers Surg Med 1989;9:345±51.

[7] Hibst R, Keller U. The mechanism of Er-YAG laser induced ablation

of dental hard substances. SPIE 1993;1880:156±62.

[8] Keller U, Hibst R. Ultrastructural changes of enamel and dentin

following Er-YAG laser radiation on teeth. Proc SPIE

1990;1200:408±15.

[9] Keller U, Hibst R. Tooth pulp reaction following Er-YAG laser appli-

cation. Proc SPIE 1991;1424:127±33.

[10] Sonntag KD, Klitzman B, Burkes EJ, Hoke J, Moshonov J. Pulpal

response to cavity preparation with the Er: YAG and Mark III free

electron lasers. Oral Surg Oral Med Oral Pathol Oral Radiol Endod

1996;81:695±702.

[11] Li ZZ, Code JE, Van de Merwe WP. Er:YAG laser ablation of enamel

and dentin of human teeth: determination of ablation rates at various

¯uences and pulse repetition rates. Lasers Surg Med 1992;12:625±30.

[12] Visuri SR, Walsh Jr JT, Wigdor HA. Erbium laser ablation of dental

hard tissue: Effect of water cooling. Lasers Surg Med 1996;18:294±

300.

[13] Burkes EJ, Hoke J, Gomes E, Wolbarsht M. Wet versus dry enamel

ablation by Er:YAG laser. J Prosthet Dent 1992;67:847±51.

[14] Keller U, Hibst R, Geurtsen W, Schilke R, Heidemann D, Klaiber B,

Raab WHM. Erbium:YAG laser application in caries therapy. Evalua-

tion of patient perception and acceptance. J Dent 1998;26:649±56.

[15] Pelagalli J, Gimbel CB, Hansen RT, Swett A, Winn II DW. Investiga-

tional study of the use of Er:YAG laser versus dental drill for caries

removal and cavity preparationÐphase I. J Clin Laser Med Surg

1997;15:109±15.

[16] Matsumoto K, Nakamura Y, Mazeki K, Kimura Y. Clinical dental

application of Er: YAG laser for class V cavity preparation. J Clin

Laser Med Surg 1996;14:123±7.

[17] Keller U, Hibst R. Effects of Er-YAG Laser on enamel bonding of

composite materials. Proc SPIE 1993;1880:163±8.

[18] Wright GZ, McConnell RJ, Keller U. Microleakage of class V compo-

site restorations prepared conventionally with those prepared with an

Er:YAG laser: a pilot study. Ped Dent 1993;15:425±6.

[19] Keller U, Hibst R. Effects of Er: YAG laser in caries treatment: a

clinical pilot study. Lasers Surg Med 1997;20:32±38.

[20] Crim GA, GarcõÂa-Godoy F. Microleakage: the effect of storage and

cycling duration. J Prosthet Dent 1987;57:574±6.

[21] Ferrari M, Goracci G, GarcõÂa-Godoy F. Bonding mechanism of three

ªone bottleº systems to conditioned and unconditioned enamel and

dentin. Am J Dent 1997;10:224±30.

[22] Tay FR, Gwinnett AJ, Pan KM, Wei SHY. Variability in microleak-

age observed in a total-etch wet-bonding technique under different

handling conditions. J Dent Res 1995;74(5):1168±78.

[23] Cagidiaco MC, Ferrari M, Vichi A, Davidson CL. Mapping of tubule

and intertubule surface areas available for bonding in Class V and in

Class II preparations. J Dent 1997;25:379±89.

[24] Asmussen E. Composite restorative resins. Composition versus wall-

to-wall polymerization contraction. Acta Odontologica Scandinava

1975;33:337±44.

[25] Niu W, Noriko Eto J, Kimura Y, Hirono Takeda F, Matsumoto K. A

study on microleakage after resin ®lling of Class V cavities prepared

by Er:YAG laser. J Clin Laser Med Surg 1998;16:227±31.

[26] Van Meerbek B, Dhem A, Goret-Nicaise M, Braem M, Lambrechts P,

Vanherle G. Comparative SEM and TEM examination of the ultra-

structure of the resin-dentin interdiffusion zone. J Dent Res

1993;72:495±501.

[27] Visuri SR, Gilbert JL, Wright DD, Wigdor HA, Walsh Jr JT. Shear

strength of composite bonded to Er:YAG laser-prepared dentin. J

Dent Res 1996;75:599±605.

[28] Marshall GW, Marshall SJ, Kinney JH, Balooch M. The dentin

substrate: structure and properties related to bonding. J Dent

1997;25:441±58.

[29] PadroÂs Fradera E, Arroyo Bote S. El laÂser de erbio-YAG en la praÂctica

odontoloÂgica general. Quintessence (ed. Esp.) 1999;12:61±70.

[30] Vargas MA, Cobb DS, Armstrong SR. Resin-dentin shear bond

strength and interfacial ultrastructure with and without a hybrid

layer. Oper Dent 1997;22:159±66.

[31] Inai N, Kanemura N, Tagami J, Watanabe LG, Marshall SJ, Marshall

GW. Adhesion between collagen depleted dentin and dentin adhe-

sives. Am J Dent 1998;11:123±7.

L. Ceballos et al. / Dental Materials 17 (2001) 340±346346