d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

available at www.sciencedirect.com

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

Cytotoxicity of four categories of dental cements

Martina Schmid-Schwap a , Alexander Franz b , Franz Knig c , Margit Bristela a , Trevor Lucas d , Eva Piehslinger a , David C. Watts e , Andreas Schedle b,
a

Department of Prosthodontics, Bernhard Gottlieb University Clinic of Dentistry, Medical University of Vienna, Austria Central Research Unit, Bernhard Gottlieb University Clinic of Dentistry, Medical University of Vienna, Austria c Section of Medical Statistics, Core Unit for Medical Statistics and Informatics, Medical University of Vienna, Austria d Center of Anatomy and Cell Biology, Medical University of Vienna, Austria e School of Dentistry and Photon Science Institute, University of Manchester, Manchester, UK
b

a r t i c l e
Article history:

i n f o

a b s t r a c t
Objectives. Assessment of dental material biocompatibility is gaining increasing importance for both patients and dentists. Dental cements may be in contact with oral soft tissues for prolonged periods of time and play an important role in prosthetic rehabilitation. The aim of the present study was to evaluate eight dental cements using a standardized L929-broblast cell culture test. Methods. For each material, fresh specimens (added to the cultures immediately after preparation) and specimens preincubated for 7 days in cell culture medium were prepared

Received 15 January 2008 Received in revised form 15 July 2008 Accepted 13 August 2008

Keywords: Toxicology Biocompatibility Cell culture Fibroblasts Dental cements

according to the manufacturers recommendations. After exposure to test specimens, cell numbers were compared to glass controls. The main outcome was a two-sided 95% condence interval for the mean value of the standardized cell number for each substance investigated. Results. Fresh specimens of all tested cements showed signicant cytotoxicity, which diminished after 7 days preincubation. Cytotoxicity of fresh adhesive and self-adhesive resin cements was lower when specimens were dual-cured compared to self-cured. A rank order of cytotoxicity was established based on mean values: Nexus 2 (dual-cured) showed least cytotoxicity, followed by Variolink II (dual-cured), Nexus 2 (self-cured), Harvard, RelyxUnicem (dual-cured), Panavia 21, Fujicem, Durelon, Variolink II (self-cured), RelyxUnicem (self-cured), Maxcem (dual-cured) and Maxcem (self-cured). When bondings were added to Nexus 2 or Variolink II specimens, a slight increase in cytotoxicity was observed. Signicance. Adhesive resin cements showed less cytotoxicity than self-adhesive and chemically setting cements. Bonding only slightly inuenced cytotoxicity of the adhesive resin cements. Dual-cured specimens of adhesive and self-adhesive resin cements showed signicantly less toxicity than self-cured specimens. 2008 Published by Elsevier Ltd on behalf of Academy of Dental Materials. All rights reserved.

Corresponding author at: Bernhard Gottlieb University Clinic of Dentistry, Whringerstrasse 25a, Vienna A-1090, Austria. Tel.: +43 1 4277 67150; fax: +43 1 4277 67159. E-mail address: andreas.schedle@meduniwien.ac.at (A. Schedle). 0109-5641/$ see front matter 2008 Published by Elsevier Ltd on behalf of Academy of Dental Materials.All rights reserved. doi:10.1016/j.dental.2008.08.002

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

361

1.

Introduction

2.
2.1.

Materials and methods

Materials used in the study

The biocompatibility of dental materials is a eld of increasing interest for both dental professionals and patients. New European directives governing medical devices (in particular Directive EU RL 93/42/EEC) have led to a reorientation of research in the dental eld. Clinical trials of dental materials, implants and dental devices are now clearly regulated and are primarily governed by the requirements of the marketing authorization system for medical devices. Clinical trials in subjects/patients may only be initiated after successful completion of nonclinical evaluations and biocompatibility studies according to European regulations and guidelines. In this respect, evaluations of material technology and compatibility studies (in particular the European ISO Standard Series 10993 which includes toxicology, allergology and aspects of occupational medicine) predominate. For such tests, comprehensive international standards have to be complied with. Recent studies have shown the biological effects of various dental materials [14]. However, cytotoxic effects on direct contact with broblasts have been well documented in cell culture studies [511]. Even ceramic materials show biological effects to varying degrees [12]. A great variety of dental composites that have enjoyed clinical success in the market for several years show similar cytotoxic behavior i.e. a moderate toxicity when tested immediately after production which diminishes after 7 days of preincubation. Several dental cements also show signicant effects in cell culture assays [9,1315]. Recently, a great variety of new resin cements has been introduced to the market that have a wide range of clinical applications. Self-adhesive, dual cure, universal resin cements have been developed that are recommended for cementation of all indirect restorations such as ceramics, metals or composite materials. Most of the resin-based luting cements are dual cure materials and can be applied with or without light curing. Conventional phosphate or glass ionomer cements are also still widely used. However, all these materials will be in close contact with oral soft tissues and play a major role in prosthetic rehabilitation. The aims of the present study were to (i) compare the cytotoxicity of eight different dental cements with a standardized test protocol, (ii) determine the inuence of 7-day specimen preincubation on cytotoxicity, (iii) examine the inuence of bonding on the cytotoxicity of adhesive resin cements and (iv) investigate the impact of dual curing compared to selfcuring on the cytotoxicity of adhesive and self-adhesive resin cements. In line with these four aims, the following null-hypotheses were formulated: (1) dual curing compared to self-curing has no impact on the cytotoxicity of adhesive or self-adhesive resin cements. (2) 7 days preincubation does not inuence the cytotoxicity of dental cements. (3) Bonding does not inuence the cytotoxicity of adhesive resin cements. (4) All eight tested dental cements show similar cytotoxicity.

Cements tested in this study fall into four distinguishable categories: category I: adhesive resin cement 1, category II: adhesive resin cement 2, category III: self-adhesive resin cements and category IV: chemically setting cements. All materials are listed in Table 1.

2.2.

Manufacture of specimens

Specimens were prepared according to the manufacturers recommendations (Table 2). Cylindrical cement specimens were prepared in white teon moulds containing 5 mm diameter cylindrical holes (cylinder height 2 mm) covered with Hostaphan (polyester foil RN 75, Mitsubishi Polyester Film GmbH, Wiesbaden, Germany) standing on a glass plate. Specimens requiring light curing were cured from one side for 20 or 40 s, respectively, with a Demetron Optilux curing light (Demetron-Kerr, Danbury, CT, USA; light irradiance 550 mW/cm2 ). The irradiance of the light was tested through the polyester lm with a radiometer (Demetron-Kerr, Danbury, CT, USA). Specimens were removed immediately after curing. All specimens were sterilized with UV-radiation for 40 min on each side. Glass specimens resembling composite specimens in diameter and height were used as negative controls.

2.3. Manufacture of resin cement specimens in combination with bonding substances

Cement specimens were combined with bonding materials. Bonding was applied on a polyethylene foil (Hostaphan , Mitsubishi Polyester Film GmbH, Wiesbaden, Germany) and light cured according to the manufacturers instructions. Thereafter, teon moulds were placed on top of the bonding materials and cements prepared as described above. After unilateral light curing from the top end of the cylindrical holes it was checked visually whether the bonding substances adhered at the bottom [16].

2.4.

Preincubation of specimens

Specimens were tested either immediately after production (fresh specimens) or after preincubation in cell culture medium; one specimen in 10 ml of Dulbeccos modied Eagle medium (DMEM; Sigma, Germany) at 37 C, pH 7.2 for 7 days. After preincubation, the culture medium was removed and specimens were applied to the broblast cell cultures.

2.5.

Culture of L929-broblasts

The murine broblast cell line L929 was obtained from American Type Culture Collection (ATCC, Rockville, MD). L929 cells were cultivated in 162 cm2 asks (Costar, Cambridge, MA) in DMEM supplemented with 10% fetal calf serum (FCS) from PAA Laboratories (Pasching, Austria), 100 U/ml penicillin and 100 mg/ml streptomycin and 2.9 mg/ml glutamine (Invitrogen, Paisley, UK) at 37 C in a fully humidied air atmosphere

362

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

Table 1 Materials tested Product and manufacturer

Negative control Positive control PVC stripsa Nexus 2 base yellowc Nexus 2 catalyst highc Opti Bond Solo Plusc Variolink II base yellowb Variolink II catalyst low viscosity/transparentb Syntacb Syntacb Heliobondb Liquid stripb Maxcem yellow shadec RelyX Unicemd Harvard cement powder quick settinge Harvard cement liquid quick settinge Durelon cement triple size powderd Durelon cement triple size liquidd FujiCEM Yellowf Panavia 21g Panavia Oxyguard IIg
a b c d e f g

Category

Specication
Glass specimen For ISO 10993-5 cytotoxicity tests Universal resin luting cement Universal resin luting cement Single component adhesive Dual-cured light-cured luting composite Dual-cured light-cured luting composite Primer Adhesive Light-curing bonding resin Glycerine gel Self-etch/self-adhesive resin cement Self-adhesive universal resin cement Zinc phosphate cement Zinc phosphate cement Carboxylate cement Carboxylate cement Reinforced glass ionomer cement Dual cure dental adhesive system Glycerine gel

Lot number
30375 442798 439579 05-1279 H35789 H33173 H24808 H04943 H11231 H34024 06-1093 232661 2122302005 2121002001 230415 231253 0511141 0398C 00521A

I I I II II II II II II III III IV IV IV IV IV IV IV

Portex Ltd. Hythe, Kent, UK. Ivoclar, Vivadent AG, Schaan, Liechtenstein. Kerr, Orange, CA, USA. 3 M ESPE AG, Seefeld, Germany. Harvard Dental GmbH, Berlin, Germany. GC Dental Products Corp. Tokyo, Japan. Kuraray Medical Inc., Okayama, Japan.

containing 5% CO2 and were passaged by brief incubation in 2.5% trypsin (Invitrogen). Cells from the fourth passage were thawed 2 weeks before each experiment and passaged twice before use.

2.10.

Statistical evaluation

2.10.1. Two-stage test

For each of the four categories (adhesive resin cement 1, adhesive resin cement 2, self-adhesive resin cements and chemically setting cements), an adaptive two-stage design concerning the sample size was employed to assure that the two-sided condence interval for the mean value of the standardized cell numbers in percent does not exceed a maximal width of 15. Within each category the following materials or material combinations were tested: Adhesive resin cement 1: Nexus 2/dual-cured, Nexus 2 dual-cured + bonding, Neuxus 2 self-cured, Nexus 2 self-cured + bonding: Adhesive resin cement 2: Variolink II dual-cured, Variolink II dual-cured + bonding, Variolink II self-cured, Variolink II self-cured + bonding: Selfadhesive resin cements: RelyX Unicem dual-cured, RelyX Unicem self-cured, Maxcem dual-cured, Maxcem self-cured. Chemically setting cements: Harvard, Panavia 21, FujiCEM, Durelon. In the rst stage of the experiment, 18 observations per cement and preincubation time were performed. The standardized cell number was referred to as the percent cell count i.e. the ratio between test cell counts and negative-control cell counts (cultures with glass specimens). Note that a lower value of the standardized cell number indicates higher cytotoxicity and vice versa. A second stage had to be performed with fresh specimens of each individual material if the standard deviation of fresh specimens was >15. The second stage sample size was determined by the maximum observed standard deviation. As a result, the nal sample size for each category was: Adhesive resin cement 1: 28, adhesive resin cement 2: 25, selfadhesive resin cements: 51 and chemically setting cements:

2.6.

Exposure of cell cultures to test specimens

L929-broblasts were exposed to freshly prepared or preincubated specimens in polystyrene 6-well tissue culture plates (Costar) for 72 h at 37 C/5% CO2 . Thereafter the specimens and the supernatant containing the dead cells were removed. Cells were then harvested with trypsin, washed with fresh medium, centrifuged and resuspended in 500 l DMEM.

2.7.

Flow cytometry

Cells were counted in a volume of 500 l DMEM over a xed time of 30 s with a ow cytometer (FACSCalibur, Becton Dickinson, San Jos, CA, USA) equipped with an argon laser tuned at 488 nm. Cell numbers after exposure to test specimens were compared to negative controls.

2.8.

Negative control

Glass specimens were used as negative controls.

2.9.

Positive control

PVC strips (Portex Ltd. Hythe, Kent, UK) for ISO 10993-5 cytotoxicity testing were used as positive controls.

Table 2 Curing methods, curing or setting time and prevention of inhibition layer methodology according to the manufacturers instructions System
Variolink II Variolink II Variolink II and Syntac Variolink II and Syntac Nexus 2 Nexus 2 Nexus 2 and Opti Bond Solo Plus Nexus 2 and Opti Bond Solo Plus Maxcem Maxcem RelyX RelyX Harvard Durelon Fujicem Panavia 21
a

Specimens
Variolink II tested alone Variolink II tested alone Variolink II tested in combination with Syntac (primer, adhesive and bonding) Variolink II tested in combination with Syntac (primer, adhesive and bonding) Nexus 2 tested alone Nexus 2 tested alone Nexus 2 tested in combination with Opti Bond Solo Plus Nexus 2 tested in combination with Opti Bond Solo Plus Maxcem tested alone Maxcem tested alone RelyX tested alone RelyX tested alone Harvard tested alone Durelon tested alone Fujicem tested alone Panavia 21 tested alone

Method of curing and curing time

Light cured for 40 s Chemically cured for 15 min Syntac: light cured for 20 s, Variolink II: light cured for 40 s

Prevention of inhibition layer

2 Hostaphanfoils Liquid-strip, 2 Hostaphanfoils 2 Hostaphanfoils

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

Syntac: light cured for 20 s, Variolink II: chemically cured for 15 min

Liquid-strip, 2 Hostaphanfoils

Light cured for 40 s Chemically cured for 10 min OBSP: light cured for 20 s, Nexus 2: light cured for 40 s OBSP: light cured for 20 s, Nexus 2: chemically cured for 10 min Light cured for 20 s Chemically cured for 6 min Light cured for 20 s Chemically cured for 10 min Chemically set for 10 min Chemically set for 7 min Chemically set for 3 min Chemically cured for 15 min

2 Hostaphanfoils 2 Hostaphanfoils 2 Hostaphanfoils 2 Hostaphanfoils 2 Hostaphanfoils 2 Hostaphanfoils 2 Hostaphanfoils 2 Hostaphanfoils 2 Hostaphanfoilsa 2 Hostaphanfoilsa 2 Hostaphanfoilsa Oxyguard II, 2 Hostaphanfoils

Air-inhibition is not an issue with these cements, as they are not free-radical systems. Hostaphan foils were used to guarantee a smooth surface of these materials.

363

364

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

35. The main outcome of this two-stage test was the mean and the corresponding 95% condence intervals (95%-CI) for all fresh substances. Additionally, the mean and 95%-CI for 7 days preincubated specimens were determined.

Table 3b Analysis of variance for self-adhesive resin cements Factor

Preincubation time (fresh vs. 7 days) Curing method (self vs. dual-cured) Test materials (Maxcem vs. RelyX Unicem) Preincubation time vs. curing method Preincubation time vs. test materials Curing method vs. test materials

p-Value
<0.0001 <0.0001 <0.0001 0.0133 0.1164 <0.0001

2.10.2. Further analysis

Each of the four categories (adhesive resin cement 1, adhesive resin cement 2, self-adhesive resin cements and chemically setting cements) was analyzed with an analysis of variance (ANOVA) to explain the effects on cytotoxicity using the standardized cell number as a dependent variable. For category adhesive resin cement 1, the factors aging time (fresh vs. 7 days), curing mode (self vs. dual-cured) and bonding (cement and bonding vs. cement without bonding) were analyzed. All interactions of order 2 were included in the model (aging time vs. curing mode, aging time vs. bonding, curing mode vs. bonding). For category adhesive resin cement 2, the ANOVA was performed in the same way. For category selfadhesive resin cements, the factors aging time, curing mode (self vs. dual-cured) and substance (Maxcem vs. RelyX Unicem) were included. All interactions of order 2 were included in this model (aging time vs. curing mode, aging time vs. substance, curing mode vs. substance). In the ANOVA for category chemically setting cements, the factors aging time and substance (Harvard, Durelon, Fujicem and Panavia 21) and the interaction aging time vs. substance were considered. In all ANOVAs, a two-sided p-value < 0.05 was considered to indicate statistical signicance. Note that in all four ANOVAs, data from the positive control were not used in analyses. For establishing a rank order of toxicity of all four categories, all the data available were pooled for one exclusive analysis and the post hoc comparison developed by Ryan [17,18], Einot and Gabriel [19] and Welsch [20] was performed. This was done separately for fresh and 7-day aged specimens. All calculations were performed with SAS Release 9.1.3.

Table 3c Analysis of variance for chemically setting cements Factor

Preincubation time (fresh vs. 7 days) Test materials (Harvard, Durelon, Fujicem and Panavia 21) Preincubation time vs. test material

p-Value
<0.0001 <0.0001

0.0297

3.

Results

The mean cytotoxicity values represented by the standardized cell numbers and the corresponding 95% condence intervals are presented for the four categories of test materials in Figs. 14. The cytotoxicity of adhesive resin cements (categories adhesive resin cement 1 and adhesive resin cement 2) was lower when specimens were dual-cured compared to self-cured specimens (Figs. 1 and 2; Table 3a, p < 0.0001 [cur-

ing method]; null-hypothesis one was rejected). For all tested conditions, cytotoxicity decreased after 7 days preincubation (Figs. 1 and 2; Table 3a, p < 0.0001 [preincubation time]; nullhypothesis two was rejected); for the category adhesive resin cement 1, a higher decrease in cytotoxicity was observed for self-cured specimens than dual-cured specimens (Fig. 1; Table 3a, p = 0.0003 [preincubation time vs. curing method]). When bondings were added to adhesive resin cements 1 or 2 for both curing modes, a slight increase in cytotoxicity was observed (Figs. 1 and 2; Table 3a, p = 0.016/p = 0.017 [bonding]; null-hypothesis three was rejected). In the category self-adhesive cements, Maxcem showed signicantly more cytotoxicity than RelyX Unicem (Fig. 3; Table 3b, p < 0.0001 [Test materials]; null-hypothesis four was rejected). The cytotoxicity of fresh Maxcem was comparable to the cytotoxicity of the positive control for both dual-cured and self-cured specimens. Again, one fresh self-adhesive resin cement (RelyX Unicem) showed less cytotoxicity when specimens were dual-cured compared to self-cured specimens (Fig. 3; Table 3b, p < 0.0001, [curing method]). Differences were not signicant after preincubation for 7 days. Preincubated dual and self-cured Maxcem specimens showed signicantly more cytotoxicity than 7 days preincubated dual-cured RelyX Unicem specimens (Fig. 3; Table 3b, p < 0.0001 [preincubation time]).

Table 3a Analysis of variance for adhesive resin cements Factor Adhesive resin cements 1
Preincubation time (fresh vs. 7 days) Curing method (self vs. dual-cured) Bonding (cement and bonding vs. cement without bonding) Preincubation time vs. curing method Preincubation time vs. bonding Curing method vs. bonding <0.0001 <0.0001 0.0158 0.0003 0.2660 0.2479

p-Value Adhesive resin cements 2

<0.0001 <0.0001 0.0167 0.4996 0.8699 0.4552

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

365

Fig. 1 Inuence of adhesive resin cement 1 (Nexus 2) on L929-broblast cell number. Specimens were produced as described in Section 2 and were added to the cultures immediately after production (fresh) or after preincubation for 7 days in cell culture medium and were incubated with L929-broblasts in 6-well culture dishes for 72 h. Cell numbers were expressed as percentage of controls (cultures with glass specimens). Bars show mean and 95% condence intervals. Sample sizes were calculated according to an adaptive two-stage design.

Fig. 3 Inuence of self-adhesive resin cements (Maxcem, RelyX Unicem) on L929-broblast cell number. Specimens were produced as described in Section 2 and were added to the cultures immediately after production (fresh) or after preincubation for 7 days in cell culture medium and were incubated with L929-broblasts in 6-well culture dishes for 72 h. Cell numbers were expressed as percentage of controls (cultures with glass specimens). Bars show mean and 95% condence intervals. Sample sizes were calculated according to an adaptive two-stage design.

In the category chemically setting cements, the test materials showed statistically signicant differences in cytotoxicity (Fig. 4; Table 3c, p < 0.0001 [Test materials]), with Panavia 21, FujiCem and Durelon exhibiting more cytotoxicity than Harvard. Again, 7 days preincubated specimens showed

Fig. 2 Inuence of adhesive resin cement 2 (Variolink II) on L929-broblast cell number. Specimens were produced as described in Section 2 and were added to the cultures immediately after production (fresh) or after preincubation for 7 days in cell culture medium and were incubated with L929-broblasts in 6-well culture dishes for 72 h. Cell numbers were expressed as percentage of controls (cultures with glass specimens). Bars show mean and 95% condence intervals. Sample sizes were calculated according to an adaptive two-stage design.

Fig. 4 Inuence of chemically setting cements (Harvard, Panavia 21, Fujicem, Durelon) on L929-broblast cell number. Specimens were produced as described in Section 2 and were added to the cultures immediately after production (fresh) or after preincubation for 7 days in cell culture medium and were incubated with L929-broblasts in 6-well culture dishes for 72 h. Cell numbers were expressed as percentage of controls (cultures with glass specimens). Bars show mean and 95% condence intervals. Sample sizes were calculated according to an adaptive two-stage design.

less toxicity than fresh specimens (Fig. 4; Table 3c, p < 0.0001 [preincubation time]). The decrease in cytotoxicity was more pronounced for Panavia 21, FujiCem and Durelon than for Harvard (Fig. 4; Table 3c, p = 0.0297 [preincubation time vs. test materials]).

366

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

Table 4 Rank order of cytotoxicity of fresh specimens based on mean values (same letters indicate no statistically signicant difference using a multiple level alpha of 5%) Substancesa
Nexus 2 dual-cured Variolink II dual-cured Nexus 2 dual-cured + bonding Variolink II dual-cured + bonding Nexus 2 self-cured Harvard cement Nexus 2 self-cured + bonding RelyX Unicem dual-cured Panavia 21 FujiCEM Durelon Variolink II self-cured Variolink II self-cured + bonding RelyX Unicem self-cured Maxcem dual-cured Maxcem self-cured
a

Category
II I II I II IV II III IV IV IV I I III III III

Mean value
79.79 70.36 70.36 63.73 58.69 57.55 49 46.2 33.24 30.2 27.92 27.34 22.95 9.88 2.84 1.89

Number of observations
28 25 28 25 28 35 28 51 35 35 35 25 25 51 51 51

REGWQ groupinga
A A A C C C E E F F F F F G G G B B B D D D

Post hoc comparison developed by Ryan [17,18], Einot and Gabriel [19] and Welsch [20].

A rank order of cytotoxicity of fresh specimens was established based on mean values (Table 4). Accordingly, a rank order for 7 days preincubated specimens is shown in Table 5. Nexus 2 cytotoxicity diminished and almost reached negative control levels with no signicant difference for all tested conditions. For Variolink II, the cytotoxicity of dual-cured specimens diminished to 95% of the control (87% when combined with bonding), whereas self-cured specimens still exhibited signicantly higher cytotoxicity (5153% of controls). Cytotoxicity of dual-cured RelyX Unicem specimens was reduced to 80% of the control and self-cured specimens to 65%. Maxcem was reduced to 50% in the self-cured mode and 55% in the dual-cured mode. Harvard cement diminished to 88% of controls, FujiCem to 79%, Durelon to 74% and Panavia 21 to 72%. In summary, these results show that fresh specimens prepared from all tested cements showed signicant cytotoxicity

which diminished after 7 days of preincubation. Whereas Nexus 2, Variolink II and RelyX Unicem were signicantly less cytotoxic when dual-cured compared to self-cured alone, Maxcem showed severe cytotoxicity in both modes. Bondings enhanced the cytotoxicity of resin cements only slightly and the conventional resin cements Nexus 2 and Variolink II were less cytotoxic than self-etching resin cements. Harvard Cement showed lower cytotoxicity than Panavia 21, Durelon and FujiCem.

4.

Discussion

Resin-based materials are known to show toxic reactions in cell cultures and various studies have shown that cytotoxicity is primarily triggered by monomers released from the material [9,2123]. The more unreacted substance the cured

Table 5 Rank order of cytotoxicity of 7 days aged specimens based on mean values (same letters indicate no statistically signicant difference using a multiple level alpha of 5%) Substancesa
Nexus 2 self-cured Variolink II dual-cured Nexus 2 dual-cured + bonding Nexus 2 dual-cured Harvard cement Variolink II dual-cured + bonding Nexus 2 self-cured + bonding RelyX Unicem dual-cured FujiCEM Durelon Panavia 21 RelyX Unicem self-cured Maxcem dual-cured Variolink II self-cured Variolink II self-cured + bonding Maxcem self-cured
a

Category
II I II II IV I II III IV IV IV III III I I III

Mean value
96.81 94.53 94.51 90.49 88.06 87.51 85.67 79.47 79.24 74.37 72.17 65.12 54.65 53.31 50.71 50.35

Number of observations
18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18

REGWQ groupinga
A A A A A A A E E E E E B B B B B B B B

C C C C C C C

D D D D D D D

F F F F F

Post hoc comparison developed by Ryan [17,18], Einot and Gabriel [19] and Welsch [20].

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

367

material contains, the higher is the toxic effect [24]. This might explain the higher toxicity levels of chemically cured and selfadhesive cements in our study. Cytotoxicity decreases with preincubation of the material in a cell culture medium prior to cell culture. Previous studies have shown that this reduction in cytotoxicity increases with time until no toxicity is detectable after 6 weeks [8,9]. This can be explained by the fact that elutable species leach from the materials, an effect that gradually decreases until below the limits of detection after 6 weeks [25,26]. Our results, showing less cytotoxicity of specimens after 7 days preincubation are in line with these observations. Cytotoxic effects of glass ionomer, zinc phosphate and carboxylate cements, similar to the results obtained in this study, have been demonstrated [9,13,15,2729]. Possible reasons for the cytotoxic effects of these cements are the release of zinc and uoride ions [27], acidity and the release of other substances [3032]. The higher toxicity of self-etching cements tested in this study compared to conventional resin cements was not elicited by a pH-change of the cell culture medium as the pH range of the cell culture medium of all tested substances was within the pH range of controls due to the buffer capacity of the cell culture medium (data not shown). In the present paper, dual-cured specimens of resin-based cements showed lower toxicity than chemically cured specimens. This concurs with the results of Aranha et al., who found that light activation reduces cytotoxicity of resin modied glass ionomer lining cements [33]. In addition, the type of curing light (halogen vs. LED) can inuence the cytotoxicity of resin cements and bondings [3436]. This concurs with the ndings of Darr und Jacobsen that dual light curing leads to a rapid increase in hardness whereas only chemically cured specimens were too soft to test in the rst 30 min [37]. Caughman et al. showed that the use of dual mode for polymerization produced better results in conversion than light curing alone for most of the cements investigated [38]. Variolink II showed only 62% conversion after chemically setting only. This might explain the higher toxicity levels for this mode in the present study. Similar results are presented by Hofmann et al. who found that the dual curing mode resulted in better mechanical properties than light activation only [39]. A moderate inuence of bonding agents on cytotoxicity was seen for the tested resin cements (Variolink II, Nexus 2) in the present investigation. It is difcult or almost impossible to compare different cytotoxicity experiments due to different experimental conditions (cell type, contact method between cells and materials and time of exposure) [40]. In addition, different methods of specimen production and different ratios of specimen size to cell layer surface/volume of cell culture medium can show differences in cytotoxicity for the same test materials [41]. With regard of these differences in experimental setups, cytotoxicity of dentin adhesive systems has been shown in other studies [42]. Another important factor for toxic effects on the dental pulp is the remaining dentin thickness [43]. The complex biology of the dentin can be simulated in vitro with the dentin barrier test [44], although this test is also only a model of the real dental situation. In xed prosthetic restorations the dentin layer is always reduced and the material might be in close contact to the dental pulp.

It was demonstrated by Franz et al. [41] and Schedle et al. [45] that L929 mouse broblasts show comparable results to primary human gingival broblasts and therefore might represent a model for gingival toxicity in vitro. Variations of the experimental design show more inuence on the results than different origins of broblasts [41]. In this study the cell line L929 was chosen, because of advantages in handling, reproducibility of results and availability compared to primary cells. The experimental setup of this study shows that dental cements are capable of eliciting biological responses. Both gingival or pulpal cells might be affected by released substances from cements. The rational for this study was to compare the cytotoxicity of several dental cements with a standardized protocol and to identify and compare their cytotoxic potential. Today, a direct correlation between the occurrence of clinical effects and cytotoxicity data does not exist. Cements are in close contact to the gingival tissue during cementation. Toxic tissue damage occurring in vivo depends on the amount of cement being in contact with oral tissues, and on the amount of components released into the aqueous environment [13]. Individual variations in sensitivity might exist. Therefore the immediate removal of cements after cementation is crucial. This study shows that dental cements are at a potential risk to cause tissue damage and this risk differs for different brands and curing modes.

5.

Conclusions

(1) The adhesive resin cements Nexus 2 and Variolink II were less cytotoxic than self-adhesive resin cements and chemically setting cements; (2) cytotoxicity of all tested cements diminished after 7 days of preincubation; (3) bondings only slightly enhanced the cytotoxicity of adhesive resin cements; (4) adhesive and self-adhesive resin cements were signicantly less cytotoxic when dual-cured compared to self-cured alone.

Acknowledgements
We would like to thank Peter Bauer for assistance with the statistical evaluation and Margit Anglmayer for skilful technical assistance.

references

[1] Tillberg A, Jrvholm B, Berglund A. Risks with dental materials. Dent Mater 2008;24:9403. [2] de Souza Costa CA, Hebling J, Randall RC. Human pulp response to resin cements used to bond inlay restorations. Dent Mater 2006;22:95462. [3] Accorinte Mde L, Loguercio AD, Reis A, Muench A, de Arajo VC. Adverse effects of human pulps after direct capping with the different components from a total-etch, three-step adhesive system. Dent Mater 2005;21:599607. [4] Sbay RK, Demirci M. Pulp reactions to dentin bonding agent as a direct capping agent. J Endod 2005;31:2014. [5] Jorge JH, Giampaolo ET, Vergani CE, Machado AL, Pavarina AC, Carlos IZ. Cytotoxicity of venture base resins: effect of water bath and microwave postpolymerization heat treatments. Int J Prosthodont 2004;17:3403.

368

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 360368

[6] Taira M, Nakao H, Matsumoto T, Takahasi J. Cytotoxic effect of methyl methacrylate on 4 cultured broblasts. Int J Prosthodont 2000;13:3115. [7] Sheridan PJ, Koka S, Ewoldsen NO, Lefebvre CA, Lavin MT. Cytotoxicity of denture base resins. Int J Prosthodont 1997;10:737. [8] Franz A, Knig F, Anglmayer M, Rausch-Fan X, Gille G, Rausch W-D, et al. Cytotxic effects of packable and nonpackable dental composites. Dent Mater 2003;19:38292. [9] Schedle A, Franz A, Rausch-Fan XH, Spittler A, Lucas T, Samorapoompichit P, et al. Cytotoxic effects of dental composites, adhesive substances, compomers and cements. Dent Mater 1998;14:42940. [10] Wataha JC, Rueggeberg FA, Lapp CA, Lewis JB, Lockwood PE, Ergle JW, et al. In vitro cytotoxicity of resin-containing restorative materials after aging in articial saliva. Clin Oral Investig 1999;3:1449. [11] Willershausen B, Schfer D, Pistorius A, Schultze R, Mann W. Inuence of resin-based restoration materials on cytotoxity in gingival broblasts. Eur J Med Res 1999;4:14955. [12] Messer RL, Lockwood PE, Wataha JC, Lewis JB, Norris S. Bouillaguet In vitro cytotoxicity of traditional versus contemporary dental ceramics. J Prosthet Dent 2003;90:4528. [13] Souza PP, Aranha AM, Hebling J, Giro EM, de Souza Costa CA. In vitro cytotoxicity and in vivo biocompatibility of contemporary resin-modied glass-ionomer cements. Dent Mater 2006;22:83844. [14] Franz A, Konradsson K, Knig F, Van Dijken JW, Schedle A. Cytotoxicity of a calcium aluminate cement in comparison with other dental cements and resin-based materials. Acta Odontol Scand 2006;64:18. [15] Lonnroth EC, Dahl JE. Cytotoxicity of liquids and powders of chemically different dental materials evaluated using dimethylthiazol diphenyltetrazolium and neutral red tests. Acta Odontol Scand 2003;61:526. [16] Franz A, Knig F, Lucas T, Watts DC, Schedle A. Cytotoxic effects of dental bonding substances as a function of degree of conversion. Dent Mater 2009;25:2329. [17] Ryan TA. Multiple comparisons in psychological research. Psychol Bull 1959;56:2647. [18] Ryan TA. Signicance tests for multiple comparison of proportions, variances and other statistics. Psychol Bull 1960;57:31828. [19] Einot I, Gabriel KR. A study of powers of several methods of multiple comparisons. J Am Stat Assoc 1975;70:351. [20] Welsch RE. Stepwise multiple comparison procedures. J Am Stat Assoc 1977;72:359. [21] Nakamura M, Imai K, Oshima H, Yoshioka S, Kawahara H. Biocompatibility test of light-cured composites in vitro. Dent Mater J 1985;4:2317. [22] Kaga M, Ito Y, Okabe T, Oguchi H, Ota M. Quantitative evaluation by measuring affected area for cytotoxicity of dental materials. Shika Zairyo Kikai 1990;9:5919. [23] Schedle A, Franz A, Rausch-Fan XH, Samorapoompichit P, Boltz-Nitulescu G, Slavicek R. Cytotoxicity testing of dental materials: composite versus amalgam. Z Stomatol Suppl 1994;6:3942. [24] Inoue K, Arikawa H, Fujii K, Niihara H, Fujita R, Tsukada G, et al. Composite restorative resins: Part 3: Cytotoxicity test to mouse broblasts in culture of UV and visible light-activated composite resins. Dent Mater 1988;7:5561.

[25] Ferracane JL, Condon JR. Rate of elution of leachable components from composite. Dent Mater 1990;6:2827. [26] Ferracane JL. Elution of leachable components from composites. J Oral Rehab 1994;21:44152. [27] Hanks CT, Anderson M, Craig RG. Cytotoxic effects of dental cements on two cell culture systems. J Oral Pathol 1981;10:10112. [28] Meryon SD, Stephens PG, Browne RM. A comparison of the in vitro cytotoxicity of two glass ionomer cements. J Dent Res 1983;62:76973. [29] Meryon SD, Browne RM. In vitro cytotoxicity of three dentine bonding agents. Cell Biochem Funct 1984;2:438. [30] Welker D, Neupert G. Vergleichende in vitro Studie zellulrer Reaktionen auf lsliche bestandteile von EBA- und Phosphat-Zement. Dtsch Zahnrztl Z 1975;30:5226. [31] Leiskar J, Helgeland K. Toxicity of some dental cements in a cell culture system. Scand J Dent Res 1977;85:4719. [32] Tyas MJ. In vitro toxicity tests of zinc phosphate cement. J Oral Rehabil 1978;5:33947. [33] Aranha AM, Giro EM, Souza PP, Hebling J, de Souza Costa CA. Effect of curing regime on the cytotoxicity of resin-modied glass ionomer lining cements applied to an odontoblast-cell line. Dent Mater 2006;22:8649. [34] Spagnuolo G, Annunziata M, Rengo S. Cytotoxicity and oxidative stress caused by dental adhesive systems cured with halogen and LED lights. Clin Oral Investig 2004;8:815. [35] Ozturk N, Usumez A, Usumez S, Ozturk B. Degree of conversion and surface of resin cement cured with different curing units. Quintessence Int 2005;36:7717. [36] Ergn G, Egilmez F, Uc tas li MB, Yilmaz S. Effect of light curing type on cytotoxicity of dentine-bonding agents. Int Endod J 2007;4:21623. [37] Darr AH, Jacobsen PH. Conversion of dual luting cements. J Oral Rehabil 1995;22:437. [38] Caughman WF, Chan DC, Rueggeberg FA. Curing potential of dual-polymerizable resin cements in simulated clinical situations. J Prosthet Dent 2001;85:47984. [39] Hofmann N, Papsthart G, Hugo B, Klaiber B. Comparison of photo-activation versus chemical or dual curing of resin-based luting cements regarding exural strength, modulus and surface hardness. J Oral Rehabil 2001;28:10228. [40] Spangberg LS. In vitro assessment of the toxicity of endodontic materials. Int Endod J 1981;14:2733. [41] Franz A, Knig F, Skolka A, Sperr W, Bauer P, Lucas T, Watts DC, Schedle A. Cytotoxicity of resin composites as a function of interface area. Dent Mater 2007;23:143846. [42] Costa CA, Vaerten MA, Edwards CA, Hanks CT. Cytotoxic effects of current dental adhesive systems on immortalized odontoblast cell line MDPC-23. Dent Mater 1999;15: 43441. [43] Galler K, Hiller KA, Ettl T, Schmalz G. Selective inuence of dentin thickness upon cytotoxicity of dentin contacting materials. J Endod 2005;31:3969. [44] Schmalz G, Schuster U, Thonemann B, Barth M, Esterbauer S. Dentin barrier test with transfected bovine pulp-derived cells. J Endod 2001;27:96102. [45] Schedle A, Samorapoompichit P, Rausch-Fan XH, Franz A, Freder W, Sperr WR, Sperr W, Ellinger A, Slavicek R, Boltz-Nitulescu G, Valent P. Response of L-929 broblasts, human gingival broblasts and human tissue mast cells to various metal cations. J Dent Res 1995;74:151320.