an evaluation of accelerated portland cement as a restorative material

Biomaterials 23 (2002) 4001–4010

An evaluation of accelerated Portland cement asa restorative material

D. Abdullaha, T.R. Pitt Fordb, S. Papaioannouc, J. Nicholsond, F. McDonaldc,*aDepartment of Restorative Dentistry, Faculty of Dentistry, Universiti Kebangsaan Malaysia, Jalan Raja Muda Abdul Aziz,

50300 Kuala Lumpur, MalaysiabDepartment of Conservative Dentistry, King’s College, GKT Dental Institute, London SE1 9RT, UK

cDepartment of Orthodontics, King’s College, GKT Dental Institute, London SE1 9RT, UKdDepartment of Materials Science, King’s College, GKT Dental Institute, London SE1 9RT, UK

Received 12 October 2001; accepted 18 April 2002

Abstract

Biocompatibility of two variants of accelerated Portland cement (APC) were investigated in vitro by observing the

cytomorphology of SaOS-2 osteosarcoma cells in the presence of test materials and the effect of these materials on the expression

of markers of bone remodelling. Glass ionomer cement (GIC), mineral trioxide aggregate (MTA) and unmodified Portland cement

(RC) were used for comparison. A direct contact assay was undertaken in four samples of each test material, collected at 12, 24, 48

and 72 h. Cell morphology was observed using scanning electron microscopy (SEM) and scored. Culture media were collected for

cytokine quantification using enzyme-linked immunosorbent assay (ELISA).

On SEM evaluation, healthy SaOS-2 cells were found adhering onto the surfaces of APC variants, RC and MTA. In contrast,

rounded and dying cells were observed on GIC. Using ELISA, levels of interleukin (IL)-1b; IL-6, IL-18 and OC were significantly

higher in APC variants compared with controls and GIC (po0:01), but these levels of cytokines were not statistically significant

compared with MTA.

The results of this study provide evidence that both APC variants are non-toxic and may have potential to promote bone healing.

Further development of APC is indicated to produce a viable dental restorative material and possibly a material for orthopaedic

application. r 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Portland cement: Cytokines: Dental material

1. Introduction

The restoration of osseous and dental defects requiresa material capable of integrating with its biologicalenvironment. Materials in close association with bonemust act as ‘osteoconductive’ materials capable ofpromoting normal bone turnover. A new class ofrestorative material called mineral trioxide aggregate(MTA) was recently introduced into the field ofbiomaterials research. The MTA is a derivative ofPortland cement with similar chemical properties [1,2]and was initially developed as a root-end dental fillingmaterial [3].

This material has since generated interest with respectto its superiority both in its biological and physicalproperties over current dental materials. It has beenreported that MTA may be a viable alternative materialin certain clinical applications such as in capping of thedental pulp tissues, root end closure, repair of rootperforation as well as a root-end filling material [4].Underlying these applications are the formidableproperties of MTA: its biocompatibility, good sealingability and the ability to promote regeneration oforiginal tissue when placed in direct contact with dentalpulp and periradicular tissues [5–9].

An ideal material should be biocompatible withdental tissues and create a local environment that isconducive to pulpal healing. Few other restorativematerials have so far been able to achieve this. Thereis already strong evidence from in vitro studiesthat prove the biocompatibility of MTA [2,9]. More

*Corresponding author. Tel.: +44-20-7955-4040; fax: +44-20-7955-

4039.

E-mail address: [email protected] (F. McDonald).

0142-9612/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.

PII: S 0 1 4 2 - 9 6 1 2 ( 0 2 ) 0 0 1 4 7 - 3

significantly, in vivo studies have shown that MTApromotes both dental and bony regeneration by pulpand periradicular tissues [4]. This ability is the mostsignificant and exciting aspect of MTA, not only indentistry where dental tissue regeneration is a highlydesired property to ensure long-term success of dentaltreatment but also in orthopaedics, where the long-termprognosis of joint replacement prostheses may poten-tially be improved with the use of cements that inducebone regeneration.

Microleakage has been a common problem in dentalrestorative treatment. To overcome microleakage, mostteeth are now restored using adhesive materials. Popularmaterials include composites, various types of glassionomer cements (GIC) and hybrid materials used inconjunction with a dentine-bonding agent. Acid etchingof enamel and dentine provides a significant way ofreducing microleakage at the tooth-restoration interface[10]. While this approach is theoretically ideal, studieshave shown that microleakage can occur with thesematerials [11–13]. However, both dye leakage [5] andbacterial leakage studies [7] have found the sealingproperty of MTA to be far superior to that of existingrestorative materials. The MTA showed excellentmarginal adaptation to cavity walls when assessed usingscanning electron microscopy (SEM) [14]. In addition,MTA possesses adequate mechanical properties that arecomparable with those of amalgam and Super EBAafter setting [15].

The main disadvantage when using MTA as a dentalcoronal restorative material is primarily due to its longsetting time of approximately 2 h [15]. For this purpose,a material should ideally have a relatively short settingtime to avoid being washed away by saliva and to reducethe possibility of the unset material irritating oraltissues. Recently, the setting time of Portland cementwas successfully reduced by adding calcium chloride(CaCl2) [1].

However, with the addition of CaCl2 to Portlandcement, there is a possibility that the biocompatibilitymight be adversely affected. Before acceleratedPortland cement (APC) is developed further forclinical usage, it is imperative that it is evaluated [16]to ensure that it is non-toxic to biological tissues in vitrobefore future in vivo tests are commenced. It is alsodesirable to ensure that the same ‘interactive’ responsecan be gained from the material by establishing ifbiologically active proteins are released from adjacentcells.

The specific objectives of this study were:

(i) to determine the surface characteristics of APC;(ii) to examine the cytomorphology of SaOS-2 cells in

the presence of the test materials; and(iii) to investigate the effects of test materials on the

expression of markers of bone remodelling.

2. Materials and methods

2.1. Preparation of materials

Variants of APC (Rugby cement (RC), Rugby, UK)were used; 10% and 15% dried CaCl2 (GPR, Hopkinsand Williams, Essex, UK) were added. Portland cement(RC), MTA (Loma Linda University, Loma Linda, CA,USA) and GIC (Fuji IX) (GC Corporation, Tokyo,Japan) were used as comparative materials.

An equal amount of 10mg of each test material wasused and mixed in the following manner:

(a) MTA (1.0 g) mixed with 0.4ml sterile distilledwater on a glass slab.

(b) GIC capsule mixed mechanically for 10 s (accord-ing to the manufacturer’s recommendation).

(c) RC: 1.0 g RC mixed with 0.4ml distilled water on aglass slab.

(d) APC1. APC with 10% CaCl2: 1.0 g RC with 0.1 g

calcium chloride mixed with 0.3ml distilledwater.

2. APC with 15% CaCl2: 1.0 g RC with 0.15 gcalcium chloride mixed with 0.3ml distilledwater.

The mixed test materials were placed on a glasscoverslip; a total of 16 coverslips per test material wasprepared for direct contact assay and 1 coverslip of eachtest material was prepared for SEM analysis of surfacecharacteristic. The materials were left to set in anincubator for 1 week at 371C, 100% humidity.

The coverslips prepared for SEM for surface char-acteristics were not immersed in culture media and weremounted directly on the SEM stubs for immediatescanning. The material-coated coverslips prepared forthe direct contact assay were placed into the sterilised 16-well plates. Prior to the assay, the 16-well plates with thematerial-coated coverslips were subjected to a contin-uous ultraviolet light sterilization for a period of 12 h.

2.2. In vitro study

The SaOS-2 cells (American Type Culture Collection[ATCC] Manassa, VA, USA; www.atcc.org) werecultured in 200ml culture flasks (75 cm2 Costar, Cam-bridge, MA) using 10ml Dubelcco’s modified Eaglemedium (DMEM) supplemented with 5% v/v heat-inactivated fetal bovine serum (HIFBS) (Sigma, Dorset,UK), 2mm l-glutamine (Sigma), 100U/ml penicillin/streptomycin (Sigma) and 100 mg/ml amphotericin (Sig-ma, Dorset, UK) in humidified 5% carbon dioxide in95% air at 371C. Following confluence, SaOS-2 cellswere trypsinised and seeded at 105 cells per 50mm wellcontaining 3ml media; each well contained the testmaterial. The cell culture technique was done under

D. Abdullah et al. / Biomaterials 23 (2002) 4001–40104002

http://www.atcc.org

strict asepsis in a flow chamber to prevent contamina-tion of media by bacteria and fungi. Sixteen well platescontaining the same number of cells were used as thecontrol. After 12, 24, 48 and 72 h, the wells were takenout and culture media were extracted entirely from fourwells of each test material and control set. Theconditioned media was collected and stored untilrequired (�201C). The cells attached to the coverslipswere fixed in 2.5% gluteraldehyde in phosphate bufferfor 1 h and then washed in 0.2m phosphate buffer (pH7.3) with sucrose for 4 h.

2.3. Scanning electron microscopy

All material-coated coverslips were prepared for SEMexamination by post-fixing in 1% Osmium tetroxide inMillonig’s constant osmolarity phosphate buffer at 41Cfor 90min in a fume cupboard. The specimens were thendehydrated using ascending grades of ethanol at roomtemperature in a fume cupboard for 60min. They werethen placed in a critical point drying (CPD) machine(Emscope CPD 750, Emitech, Ashford, Kent). Thechamber was purged several times to expel anyremaining ethanol before flooding it with liquid CO2.When it was completely full and cooled to 81C, thevalves were closed and the chamber was heated to 401Cat which point the pressure increased to 1500 psi. Theresulting gas of CO2 was then gradually ‘‘bled’’ from thechamber. After drying, the specimens were mounted onthe SEM stubs using silver paint as an adhesive. Thematerials were coated with gold to a thickness of 20 nmusing an Emscope 1500.

Each coverslip sample of APC variants and from thedirect contact assay was examined using SEM (SEM501B, Hitachi, Wokingham, Berks, UK) to observesurface characteristics and cell morphology, respec-tively. An accelerating voltage of 15 kV was used formost specimens and 7.5 kV was used when fine details ofthe material surface were needed or if the specimenswere deemed ‘powdery’. Specimens were considered asbeing ‘powdery’ when multiple disseminated discreteclumps of crystals were observed to detach from themain material bulk. Two images of each specimen wererecorded at 320� and 670� magnification using FP4125 film (Ilford, Kalmut Ltd, London).

A scoring method was devised in this study. Whenhuman osteosarcoma cell lines were examined underSEM, healthy cells have been described as being flat,possessing a well-defined morphology with cytoplasmicextensions, propagating over and adhering to thesurface of test materials [17]. Samples showing thispicture were given the highest score of 2. Less healthy ordying cells appeared round, had fewer cytoplasmicextensions, were detached from the material, andsamples showing this picture were scored as 1. Whenno surviving cells were observed, samples were scored as

0. Samples that were damaged and deemed unsuitablefor evaluation were excluded from the study anddenoted as ‘n’. Sixteen plates of each material for eachtime were used, with a total of 80 samples.

2.4. Enzyme-linked immunosorbant assay (ELISA)

The cytokine and protein assay method used in thisstudy was the quantitative ELISA (Quantikine, R&DSystems Ltd, Oxford) ‘sandwich’ technique. Each assaykit contained 96 wells with a microtitre plate coated inmonoclonal antibody to a specific cytokine in the base.When the media was added to the well, the cytokinepresent bonded to the monoclonal antibody forming acomplex. Following a time period of up to 2 h, unboundproteins were washed away and an enzyme-linkedpolyclonal antibody was added to the well that actedas a link between the cytokine–antibody complex andthe colouring agent. A colour change proportional tothe exact amount of cytokine present was seen. This wasthen quantified by comparing the samples with those ofknown dilutions supplied with each kit and measuringthe optical density at 540 nm using a plate reader (LabSystems, Multiskan RC, Oxford, UK). The proteinsassayed were interleukin (IL)-1b; IL-6, IL-18 andosteocalcin, all cytokines identified as having a role inlocal bone remodelling control.

2.5. Statistical analysis

Statistical analysis was carried out using the MannWhitney U-test to compare the values of both APCvariants, MTA and RC with the control and GICvalues.

3. Results

3.1. Surface characteristics

Samples of APC showed mixed surface character-istics. Overall, the surfaces of APC were irregular.Under 670� magnification, a typical APC sampleshowed areas consisting of either cuboidal discretecrystals or areas of granular material with a coral-likesurface character. Cells were found growing in closeproximity to both types of surfaces but more cells werefound attached to the granular surfaces. The surfaces ofRC and MTA were also irregular consisting of smallround and cuboidal crystals and the same granularcrystals seen with APC variants.

3.2. Cell morphology

Cytomorphological evaluations were undertaken ofthe 80 samples (92.5%). The scores are summarised in

D. Abdullah et al. / Biomaterials 23 (2002) 4001–4010 4003

Table 1; examples are shown of GIC (Fig. 1), RC(Fig. 2), MTA (Fig. 3) and 15% APC (Fig. 4) showingthe range of cell coverage. In general, SaOS-2 cells

in contact with RC, MTA and variants of APCappeared flat and adhered well onto material surfacesat all time intervals. Rounded or dying cells were seen

Table 1

Scores for cell morphology for different materials at varying time intervals

Material Hours

12 24 48 72

Score M Score M Score M Score M

APC 10% 22 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

APC 15% 22 2n 2 2 2 2 n 2 2 2 2 n 2 2 2 2 1 2

GIC 1 1 1 1 1 1 1 1 1 1 0 0 1 0 0 0 0 0 0 0

MTA 22 2n 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2

RC 22 1 2 2 2 2 2 n 2 2 1 2 n 2 2 1 1 1 1

M—mode score.

Symbols:

0—no surviving cell.

1—cells appear rounded, less flattened with fewer cytoplasmic extensions.

2—cells appear flat, propagated on material and exhibited intact, well-defined morphology with cytoplasmic extension.

n—not able to evaluate surface.

Fig. 1. Scanning electron micrographs of GIC with cells in contact with material. Cell outlines appear flattened with cytoplasmic processes spreading

across the material surface at 12 h (a). There is no confluence of the cells but the ones present appear relatively healthy (score 2; see text for details).

At 24 h there were less cells adherent to the surface. This is in contrast with the other materials such as MTA were there is an increase in cell number

adherent to the material. There were no cells visible at 48 h (image not shown scoring 0); the cells at 72 h appear rounded and not adherent to the

material (score 1).


on most of the surfaces of GIC at 24, 48 and 72 h(Fig. 1).

3.3. ELISA assay of osteocalcin and respective cytokines

The ELISA assays and graphs of the bone regulatorycytokines, i.e. IL-1b; IL-6, IL-18 and osteocalcin (OC)are shown (Figs. 5–8). Cells in contact with cementsbased on Portland cement, i.e. APC (10% and 15%),MTA and RC expressed higher levels of cytokines andOC after 24 h compared with control cells and cells incontact with GIC.

When individual cytokine levels were considered, cellsin contact with APC (10%) was found to produce thehighest concentration of OC (67.0 pg/ml+/�5.5) at48 h, IL-1b (76 pg/ml+/�8.3) and IL-18 (21.0 pg/ml+/�4.1) at 72 h. The levels of all the cytokines expressed bythe cells increased with time.

Cells in contact with GIC produced negligibleamounts of OC at 12 and 24 h but failed to expressany of the ILs assayed. None of the controls showed anyproduction of bone regulatory cytokines and protein.

3.4. Statistical analysis

The data obtained from ELISA were derived from arelatively small sample and were not normally distrib-uted. Limited statistical analysis of cytokine expressionwas therefore performed using a non-parametric MannWhitney U-test (SigmaStat Version II). The cytokinelevels expressed by the cells in contact with APC, MTAand RC were found to be significantly higher (po0:01)compared with the control and GIC samples. The valuesbetween the APC (10%) and APC (15%) were notsignificantly different (p > 0:05); the APC variantsproduced significantly higher levels of IL-1b and IL-6than RC (po0:05).

4. Discussion

The present study indicates that APC supports theproliferation of SaOS-2 cells in vitro and activelystimulates a biological response in these cells throughthe production of cytokines and a bone-specific protein.This implies that not only is APC non-toxic, but it also

Fig. 2. RC shown at 12 h (a), 24 h (b), 48 h (c) and 72 h (d) showing the granular texture of the material and the cells on the surface. There is a steady

increase in number of cells covering the surface, all of which appear healthy and adherent (score 2).


forms a significant step in the development of Portlandcement as a restorative material. The improvement of itssetting time by the addition of CaCl2 did not interferewith the biocompatibility and osteoconductive propertyof the parent cement. Previous research has shown thismaterial to have good sealing ability as well as adequatephysical and mechanical properties [1] and this studynow provides good evidence of biocompatibility.

The direct contact assay model used in this study wasbased on those used by Koh et al. [9] and Mitchell et al.[2] with significant modifications. A human osteosarco-ma SaOS-2 cell culture system was employed in thisstudy; these cells closely resemble the human osteoblastsin their ability to express high levels of bone markerssuch as, alkaline phosphatase, cyclic adenosine mono-phosphate second messenger response to parathyroidhormone stimulation, and the expression of osteonectin,bone sialoprotein and decorin. These cells can also formbone when implanted in vivo and form a mineralisedmatrix in long-term cultures in vitro [18]. The use of thiscell line is in contrast to previous studies that have usedMG-63 osteosarcoma cells to investigate the biocompat-ibility of MTA [2,9]. These cells are known to have fewer

characteristics of mature osteoblasts compared with theSaOS-2 cell line [18].

The selection of cytokines and bone proteins in thisstudy differed from that used previously [2], as IL-18and OC were included for quantification along with IL-1b and IL-6. These cytokines were chosen because theyare potent markers of bone remodelling, physiologicallyproduced by normal, functional osteoblasts. The IL-18is a newly discovered cytokine that mediates boneformation [19] while OC is an established marker forbone formation [20]. The OC is a bone specific proteinforming about 20% of total non-collagenous proteins,and consists of three g-carboxyglutamic acid residues;this configuration enables it to participate in theregulation of hydroxyapatite growth. It is consideredto be a signal of osteoblast function of bone formationand mineral maturation [20]. The IL-1b is a polypeptidewith a molecular weight of 17.5 kDa and is an extremelypotent and powerful bone-resorbing cytokine both invitro and in vivo [21]. The IL-6 is a polypeptide with amolecular weight of between 23 and 30 kDa, andstimulates bone resorption acting synergistically withIL-1 [21].

Fig. 3. MTA shows the typical adherence with increasing coverage of cells with time at 12 h (a), 24 h (b), 48 h (c) and 72 h (d). There appears an

increase in cell number with time.


A new scoring method was devised because theprevious scoring method [2] was not entirely satisfac-tory. The new method was based on cell morphologybecause it is a valuable indicator of the physiologicalstate of the cell. When a cell is irreversibly injured, itundergoes nuclear and cytoplasmic changes leading tocell death. The earliest morphological feature of thisprocess is cell shrinkage and this is followed by loss ofintercellular connections, cell detachment to substrate,and finally the cell fragments [22]. The previous scoringmethod [2] was based on the extent of cell cover over thesurface of the material. However, the extent of cellcoverage may not be an ideal marker of cell–materialinteraction because the quantity of cell growth and itsspread is ultimately dependent on time. Furthermore,the method failed to take account of material dislodge-ment.

SEM provides important information in establishingbiocompatibility by observing the physical appearancesof the test materials, the cells and the material–cellinteractions. During SEM evaluation of the surface, amixture of two types of appearance was observed withboth APC variants. Some areas consisted predominantlyof small, discrete cuboidal-shaped crystals while other

areas appeared to be granular with coral-like surfaces.The surface appearances found in this study are similarto that seen with MTA [9,15].

The observations made in this study indicate thataddition of CaCl2 has not altered the basic crystal formsof the parent cement. As crystals represent the arrange-ments and interactions of individual atoms and mole-cules within a material, it may be postulated that thepresence of CaCl2 has not interfered with the chemicalcomposition of the parent cement whose principleconstituents are calcium and phosphorus ions [15].

Cells have been found in close contact with APC. Thisappearance suggests that the surfaces of the material arenon-irritant and do not affect the structural integrity ofthe cell. The cells appeared flat and exhibited intact,well-defined morphology with cytoplasmic extensions.The preservation of cytoplasmic extensions is importantbecause these extensions form a vital three-dimensionalnetwork within bone [23].

The close proximity of SaOS-2 cells with APCindicates favourable interaction between the cell andmaterial. Surface characteristics of a material maycontribute to cellular function [20], particularly inosteoblasts whose normal function is greatly influenced

Fig. 4. 15% APC also shows the increase in material coverage with time at 12 h (a), 24 h (b), 48 h (c) and 72 h (d).


by attachment to specific components of the matrix [24].In addition to surface chemistry and composition, it ispossible that the surface characteristics of APC areconducive for cell attachment, perhaps by encouragingthe production of specific bone matrix proteins such asType 1 collagen [24]. This protein could prove to be auseful marker in future studies.

In comparison to APC samples, the cells in contactwith GIC showed poor morphological features typicalof irreversibly damaged and dying cells. This findingcorresponded to that in an in vitro study [25], where cellsin close contact with GIC were found to be non-viable.Leachable materials were extracted from GIC andadded to the medium used for a different cell culture;cell growth was inhibited in the presence of the extractsuggesting that GIC is potentially cytotoxic. At presentthe exact, potentially cytotoxic, constituents of GIC areunknown although the fluoride content has beensuggested [25]. However, a contradictory observationhas been reported [26]. In an SEM study investigatingthe biocompatibility of five types of GIC, healthyprimary osteoblasts were seen in close contact with fourtypes of GIC. The GIC was placed in the bottom of theculture dishes and the cell culture was subsequentlyadded. Prior to the direct contact assay, three out of thefive types of GIC were washed several times to removethe acidic components released by the material. It is

possible that the removal of these acidic componentsfrom the GIC accounted for the favourable cellmorphology.

Although SEM is the primary investigative tool inmany biocompatibility studies, ELISA provides strongand reliable evidence of cellular physiology at thebiochemical level. The role of ELISA in this study isextremely significant because the cytokine expressionscorresponded well to the results from SEM.

The addition of CaCl2 may also explain the differ-ences in the pattern of cytokine and OC expression. TheMTA maintained a similar pattern of protein expressionwith IL-1b; IL-6 and OC, where a progressive rise inconcentration occurred until 48 h when maximumconcentration was reached and this was followed by asteady reduction, while APC variants continued tomaintain high concentrations of cytokines and OC at72 h.

Bone remodelling is often described as being dividedinto three stages: activation, resorptive and boneformation [27]. It could be extrapolated from this studythat APC variants might be involved in all three stagesof bone remodelling as there appears a significantinfluence on cytokines associated with each phase. TheAPC variants stimulated the production of high

0

10

20

30

40

50

60

70

80

90

APC 10% APC 15% GIC MTA RC CONTROLTest materials

pg/m

l

12 hours 24 hours48 hours 72 hours

Fig. 5. IL-1b—mean values and standard deviations (pg/ml).0

10

20

30

40

50

60

70

80

90

APC 10% APC 15% GIC MTA RC CONTROL

Test materials

pg/m

l

12 hours 24 hours48 hours 72 hours

Fig. 6. IL-6—mean values and standard deviations (pg/ml).


concentrations of IL-1b and IL-6 in SaOS-2 cellscultures as early as 12 h following initial cell contactwith the material. Both these cytokines mediate theactivation stage of bone remodelling, where recruitmentof multipotential bone marrow cells to the interfacebetween calcified bone matrix and bone marrow occurs[27]. As these cells arrive at the site, IL-1b and IL-6promote cell differentiation into osteoclasts [18]. Re-cently, differentiated and existing mature osteoclastshave been shown to begin bone resorptive activity by thesame cytokines [18]. During the resorptive stage,osteoclasts produce lysosomal enzymes and activelypump protons into the resorption lacunae causingmatrix degradation, liberating latent cytokines that havestimulatory effects on osteoblasts [28]. The stimulationof osteoblasts marks the beginning of bone formation[27]. The production of OC, a bone formation marker,by APC variants reached its maximum concentrations inSaOS-2 cell cultures at 48 h, which is 12 h following theinitial production of IL-1b and IL-6 at the activationstage. The bone formation stage is further encouragedby high concentrations of IL-18, particularly in theperiod between 48 and 72 h. The IL-18 indirectlypromotes bone formation by inhibiting the proliferationof osteoclasts through the action of the cytokine,granulocyte/macrophage colony stimulating factor [21].

In vivo work, however, would be essential to examinethese hypotheses and this is clearly the next stage ofdevelopment.

The GIC has been proposed as an alternative bonecement to polymethylmethacrylate in orthopaedic sur-gery based on its strong adhesion to hydroxyapatitecrystals under moist conditions [29]. Results obtainedfrom the present study suggest that APC is a bettermaterial compared with GIC and could have potentialfor use as a bone cement.

This study has shown that the addition of CaCl2 as anaccelerator of the setting reaction of RC has no adverseeffect, indeed apparently the reverse, on biocompat-ibility. It now allows further development of APC toproduce a faster setting cement restorative and evenorthopaedic applications.

Acknowledgements

We would like to thank the British EndodonticSociety for their financial contribution to the study.

0

10

20

30

40

50

60

70

80


pg/m

l

12 hours 24 hours

48 hours 72 hours

Fig. 8. OC—mean values and standard deviations (pg/ml).

0

5

10

15

20

25

30


pg/m

l

12 hours 24 hours

48 hours 72 hours

Fig. 7. L-18—mean values and standard deviations (pg/ml).


References

[1] Aquilina JW. The physical properties of accelerated

Portland cement. M.Sc. Project Report, University of London,

1999.

[2] Mitchell PJC, Pitt Ford TR, Torabinejad M, McDonald F.

Osteoblast biocompatibility of mineral trioxide. Biomaterials

1999;20:167–73.

[3] Torabinejad M, Watson TF, Pitt Ford TR. Sealing ability of a

mineral trioxide aggregate when used as a root-end filling

material. J Endodont 1993;19:591–5.

[4] Torabinejad M, Chivian N. Clinical application of mineral

trioxide aggregate. J Endodont 1999;25:197–205.

[5] Torabinejad M, Higa RK, McKendry DJ, Pitt Ford TR. Dye

leakage of four root-end filling materials: effects of blood

contamination. J Endodont 1994;20:159–63.

[6] Pitt Ford TR, McKendry DJ, Torabinejad M, Hong CU,

Kariyawasam SP. Use of mineral trioxide aggregate for repair

of furcal perforations. Oral Surg, Oral Med, Oral Pathol

1995;79:756–63.

[7] Torabinejad M, Rastegar AF, Kettering JD, Pitt Ford TR.

Bacterial leakage of mineral trioxide aggregate as a root-end

filling material. J Endodont 1995;21:109–12.

[8] Bates CF, Carnes DL, Rio CE. Longitudinal sealing ability of

mineral trioxide aggregate as a root-end filling material.

J Endodont 1996;22:575–8.

[9] Koh ET, Torabinejad M, Pitt Ford TR, Brady K, McDonald F.

Mineral trioxide aggregate stimulates a biological response in

human osteoblasts. J Biomed Mater Res 1997;37:432–9.

[10] Vargas MA, Swift EJ. Microleakage of resin composites with wet

versus dry bonding. Am J Dent 1994;7:187–9.

[11] Br.annstr .om M. Communication between the oral cavity and the

dental pulp associated with the restorative treatment. Opt Dent

1984;9:57–68.

[12] Sidhu S. Sealing effectiveness of light-cured glass ionomer cement

liners. J Prosthet Dent 1992;68:891–3.

[13] Grieve A, Saunders W, Alani A. The effect of dentine bonding

agents on marginal leakage of composite restoration—long term

studies. J Oral Rehab 1993;20:11.

[14] Torabinejad M, Smith PW, Kettering JD, Pitt Ford TR.

Comparative investigation of marginal adaptation of mineral

trioxide aggregate and other commonly used root-end filling

materials. J Endodont 1995;21:295–9.

[15] Torabinejad M, Hong CU, McDonald F, Pitt Ford TR. Physical

and chemical properties of a new root-end filling material.

J Endodont 1995;2:349–53.

[16] British Standard Institute. Dentistry-preclinical evaluation of

biocompatibility of medical devices used in dentistry—test

methods for dental materials. BSI 1997 Series EN ISO 7405,

1997. p. 1–18

[17] Kue R, Sohrabi A, Nagle D, Frondoza C, Hungerford D.

Enhanced proliferation and osteocalcin production by human

osteoblast-like MG63 cells on silicon nitride ceramic discs.

Biomaterials 1999;20:1195–201.

[18] Hughes FJ, Aubin JE. Culture of cells of the osteoblast lineage.

In: Arnett TR, Henderson B, editors. Methods in bone biology.

UK: Chapman & Hall, 1998. p. 1–39.

[19] Udagawa N, Horwood NJ, Elliot J, Mackay A, Owens J,

Okamura H, Kurimoto M, Chambers TJ, Martin TJ, Gillespie

MT. Interleukin-18 (interferon-gamma-inducing factor) is pro-

duced by osteoblasts and acts via granulocytes/macrophage

colony-stimulating factor and not via interferon-gamma to inhibit

osteoclast formation. J Exp Med 1997;185:1005–12.

[20] Termine JD. Bone matrix proteins and the mineralization process.

In: Favus MJ, editor. Primer on the metabolic bone diseases and

disorders of mineral metabolism. 2nd ed. New York: Raven Press,

1993. p. 21–5.

[21] Mundy GR. Cytokines of bone. In: Mundy GR, Martin TJ,

editors. Physiology and pharmacology of bone. Berlin: Springer,

1993. p. 185–214.

[22] Boyce BD, Hughes DE, Wright KR. Methods for studying cell

death in bone. In: Arnett TR, Henderson B, editors. Methods in

bone biology. UK: Chapman & Hall, 1998. p. 127–48.

[23] Freemont AJ. Histology of mineralized tissues. In: Huskins

DWL, editor. Topics in molecular and structural Biology, 11.

Calcified tissue. London: The Macmillan Press, 1993. p. 21–40.

[24] Green J, Schotland S, Stauber DJ, Kleeman CR, Clemens TL.

Cell matrix interaction in bone: type I collagen modulates signal

transduction in osteoblasts. Am J Physiol 1995;268:C1090–103.

[25] Doherty PJ. Biocompatibility of glass ionomer cement using cell

culture techniques. Clin Mater 1991;7:335–40.

[26] Oliva A, Ragione FD, Salerno A, Riccio V, Tartaro G, Cozzolino

A, D’Amato S, Potoni G, Zappia V. Biocompatibility studies on

glass ionomer cements by primary cultures of human osteoblasts.

Biomaterials 1996;17:1351–6.

[27] Baron R. Anatomy and ultrastructure of bone. In: Favus MJ,

editor. Primer on the metabolic bone diseases and disorders of

mineral metabolism. 2nd ed. New York, USA: Raven Press, 1993.

p. 3–9.

[28] Eriksen EF, Vesterby A, Kassem M, Melsen F, Mosekilde L.

Bone remodelling and bone structure. In: Mundy GR, Martin TJ,

editors. Physiology and pharmacology of bone. Berlin: Springer,

1993. p. 67–101.

[29] Jonck LM, Grobbelaar CJ, Strating H. The biocompatibility of

glass ionomer cement in joint replacement: bulk testing. Clin

Mater 1989;4:85–107.


an evaluation of accelerated portland cement as a restorative material

Documents