Journal of Oral Rehabilitation 2001 28; 6877

Bonding agent is a decisive factor in determining the marginal leakage of dental composites subjected to thermal cycling: an in vitro study
M. MATHEW*, E. K. PARAMESWARAN NAIR* & K. KRISHNAN V
*Yenepoya Dental College, Kodial Bail, Mangalore and Sree Chitra Tirunal Institute for Medical Sciences and Technology, Poojapura, Trivandrum, Kerala, India

SUMMARY

The marginal leakage at the dentine/ composite interface in Class II composite restorations subjected to thermal cycling has been effectively evaluated using the silver staining technique. The presence of a bonding agent at the dentine/composite interface is found to improve adhesion. Scanning electron and optical microscopic observations of sectioned specimens reveal that applying a second coat of bonding agent on the

dentinal surface helps in reducing microleakage compared to a single coat application, in all three radiopaque composite/bonding agent systems studied (Z100/Single Bond, Spectrum TPH/Prime & Bond 21 and Chitra/Chitrabond 10). Thermal cycling during in vitro studies was found to provide a more appropriate representation of the adhesive behaviour of the composite in clinical situations.

Introduction
The supreme test of a dental restorative lling is its ability to maintain an unfailing margin. A microscopic space tends to exist between the restoration and the prepared cavity which results in passage of saliva and salivary products into it. Incomplete marginal sealing at the tooth restoration interface is found to result in post-placement sensitivity, marginal staining, recurrent caries and eventually the development of pulp pathology (Phillips, 1965; Going, 1972). In fact, previous studies indicate that chemical toxic leachants from restorative materials are less signicant in causing pulpal injury than bacterial leakage around the restoration margin (Brannstrom, 1986; Cox et al., 1987). Many techniques have been devised to test the cavity sealing properties of restorations both in vivo and in vitro. In vitro studies include the use of dyes, chemical tracers, radioactive isotopes, air pressure, bacteria, neutron activation analysis, scanning electron microscopy (SEM), articial caries techniques and electrical conductivity. The silver staining technique has been effectively used previously to study the degree of
2001 Blackwell Science Ltd

microleakage and was found to provide a more superior denition and accurate evaluation of microleakage (Wu et al., 1983; Krishnan, Bindhu & Manjusha, 1996). All these studies emphasized that the margins of restorations are not xed, inert with impenetrable borders but rather as Myers stated in 1966 possess dynamic microcrevices which contain a busy trafc of ions and molecules (Alani & Toh, 1997). Marginal integrity of the composite restorations is found to depend on many variables including polymerization shrinkage (Rees & Jacobsen, 1989), thermal expansion, water sorption (Retief, 1994), mechanical loading (Qvist, 1983), marginal degradation and most importantly adhesion to tooth structure. A gapfree direct composite restoration is possible only if the adhesive forces to the tooth overcome the interfacial stresses generated during the curing and afterwards. Elimination of marginal contraction gaps by dentine bonding treatments have been reported before (Shortall & Asmussen, 1988). The higher the bond strength mediated by an adhesive, the smaller are the marginal gaps formed by a composite resin in a dentine cavity
68

MARGINAL LEAKAGE OF COMPOSITES

Table 1. (a) Radiopague composites used and (b) bonding agents used
Composites used Z100 Spectrum TPH Chitra (b) Single Bond P & B 21 Chitrabond 10 Lot No. 19970702 9606006 Under trials Source 3M Co., MN, U.S.A. Dentsply DeTrey GmbH, Germany SCTIMST, Trivandrum, India 3M Co., MN, U.S.A. Dentsply DeTrey, GmbH, Germany SCTIMST, Trivandrum, India Composition BIS-GMA, TEGMA, Zr or SiO2 BIS-GMA and HMDI adduct, BIS-EMA, TEGDMA, SiO2, Ba, Al, Si glass BIS-GMA, TEGDMA, radiopaque glass, SiO2

69

19970414 9712000721 Under trials

BIS-GMA, HEMA, alcohol, copolymer of PAA and PIA PENTA, dimethacrylate resin, acetone, cetylamine uoride Primer: maleic acid/HEMA solution; resin: BIS-GMA/HEMA

BIS-GMA, bisphenol A-glycidyl methacrylate; TEGMA: triethyleneglycol methacrylate; SiO2: silica; HMDI, hexamethylene diisocyanate; BIS-EMA, bisphenol A-ethylmethacrylate; TEGDMA: triethyleneglycol dimethacrylate; HEMA, 2-hydroxyethyl methacrylate; PAA, polyacrylic acid; PIA, polyitaconic acid; PENTA, dipentaerythritol pentaacrylate monophosphate.

No. 1 2 3 4 5 6 7

Chemical Etchant gel, Scotchbond Silver nitrate crystals Developing solution Green stick compound Nail varnish Araldite 502 EM media Alumina particles (B005 mm)

Batch No. 7423 0595-595-200431 Nil JY 9 Nil DOT 12B65 40-6353-006

Source 3M Co., MN, U.S.A. SD Fine Chemicals, Bombay, India Commercial grade Pinnacle DPI, Bombay, India Commercial grade Polysciences Inc., U.S.A. Buehler Co., IL, U.S.A.

Table 2. Other chemicals used

treated with adhesives. Asmussen & Munksgaard (1985) have reported a minimum dentine adhesion of 20 MPa to ensure gap-free margins. Most modern adhesive systems are superior to their predecessors especially in terms of retention, sensitivity to substrate and other clinical co-variables. But none of these systems appear yet to be able to ensure a hermetically sealed restoration with margins free of discoloration for a long time. Use of thermocycling during in vitro microleakage testing has been found to provide results which correlate laboratory ndings more accurately with clinical performance (Crim & Mattingly, 1981). Previous studies indicate conicting reports about the effect of thermal cycling upon marginal leakage in composite restorations (Darbyshire, Messer & Douglas, 1988; Kubia & Jacobsen, 1991; Rigsby et al., 1992; Rossomando & Wendt, 1995). In this study, the silver staining technique has been used effectively to evaluate the marginal leakage at the composite/dentine interface in Class II cavity restorations subjected to thermal cycling. The importance of

bonding agent and its mode of application in reducing marginal leakage has been studied in detail for three radiopaque composite/bonding agent systems.

Table 3. Nine different composite bonding agent systems used in the study
No. of teeth used 4 4 4 6 6 6 6 6 6

Group No. I II III

Composite Z100 Spectrum TPH Chitra

Bonding agent Nil Nil Nil Single Bond P & B 21 Chitrabond 10 Single Bond P & B 21 Chitrabond 10

Single layer coating IV Z100 V Spectrum TPH VI Chitra Double layer coating VII Z100 VIII Spectrum TPH IX Chitra

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 68 77

70

M . M A T H E W et al.
Table 4. Microleakage grading pattern for systems under study

Bonding agent Z100/Single Bond

No. 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6

NBA 5 6 6 6 6 6 6 5 5 6 5

Single coat 5 5 5 5 5 5 5 4 5 5 4 5 4 4 5 5 5 5

Double coat 3 5 5 5 5 5 5 5 3 4 4 3 2 4 5 2 5

Spectrum TPH/P & B 21

Chitra/Chitrabond 10

NBA, no bonding agent. 0, No visible leakage; 1, enamel leakage only; 2, dentine leakage only but not involving dentinal oor; 3, enamel and dentinal leakage not involving the oor; 4, dentine leakage involving the oor; 5, enamel and dentinal leakage involving the oor; 6, leakage into dentinal tubules.

Materials and methods

Sample collection and storage Freshly extracted non-carious human premolar teeth were collected and stored in normal saline. They were cleaned of debris by storing them in 1% hydrogen peroxide for 24 h and subsequent washing in distilled water. Samples having cracks and ssures were discarded.

The teeth were subsequently stored in distilled water. Restorative procedure. The prepared teeth were randomly divided into nine groups as shown in Table 3. Group I. Prepared cavities were etched with etching gel (37% phosphoric acid) for 15 s, washed in distilled water for 10 s and air dried under oil free compressed air. Z100 composite was placed into the cavity in two increments. Each increment was cured for 40 s using a visible light source having a wavelength of 440 480 nm.

Materials used Three radiopaque light cured composites and three bonding agents were used in the study, details of which are shown in Table 1a and b. Other chemicals and their sources are shown in Table 2.

Methods Cavity preparation. Class II cavities (MO/DO) were prepared with dimensions 25 mm depth occlusally, 2 mm width occlusally, 35 mm depth gingivally and a 3 mm wide gingival seat using a No. 330 diamond point in a high speed handpiece* with continuous water cooling.
* NSK, Tokyo, Japan.

Fig. 1. Formation of staining at the subsurface of composites (15) viewed under an optical microscope.

Caulk the Max, L. D. Caulk Co., Milford, DE, made in Canada.

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 6877

MARGINAL LEAKAGE OF COMPOSITES dried. Two increments of Chitra composite were placed into the cavity and cured for 60 s each. In groups I, II and III no bonding agent was used. Group IV. Etching was carried out as in group I and a single coat of Single Bond was applied, air dried for 25 s and cured for 10 s. Z100 composite was placed as before. Group V. Etching was carried out as in group II and a single coat of Prime & Bond 21 (P & B) was applied, left undisturbed for 30 s, air dried for 25 s and cured for 20 s. Spectrum TPH composite was placed as before.
Fig. 2. Microcrack formation at dentine/composite interface (350) viewed under an optical microscope.

71

Group VI. Etching was carried out as in group III. Chitrabond 10 primer was applied, left undisturbed for 30 s and air dried for 20 s. A single coat of Chitrabond 10 resin was then applied, air dried for 20 s and cured for 20 s. Chitra composite was placed subsequently as before.

Fig. 3. (a) Unpolished surface at the dentine/composite interface viewed under an optical microscope. (b) Dentine/composite interface viewed under an optical microscope after polishing.

Group II. Prepared cavities were etched with etching gel for 20 s, washed in distilled water for 10 s and air dried. Two increments of Spectrum TPH composite were placed into the cavity and cured for 40 s each. Group III. Prepared cavities were etched with etching gel for 15 s, washed in distilled water for 60 s and air

Fig. 4. (a) Dentine/composite interface showing silver stained marginal gap in cavities restored with Z100 composite without bonding agent viewed under an optical microscope (350 ). (b) Dentine/composite interface showing silver stained marginal gap in cavities restored with Z100 composite without bonding agent viewed under SEM.

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 68 77

72

M . M A T H E W et al.

Fig. 5. (a) Dentine/composite interface showing silver stained marginal gap in cavities restored with Spectrum TPH composite without bonding agent viewed under an optical microscope (175 ). (b) Dentine/composite interface showing silver stained marginal gap in cavities restored with Spectrum TPH composite without bonding agent viewed under SEM.

Fig. 6. (a) Dentine/composite interface showing silver stained marginal gap in cavities restored with Chitra composite without bonding agent viewed under an optical microscope (350 ). (b) Dentine/composite interface showing silver stained marginal gap in cavities restored with Chitra composite without bonding agent viewed under SEM.

Group VII. After etching, two coats of Single Bond were applied one after the other, air dried for 25 s and cured for 10 s. Z100 was placed as before. Group VIII. After etching, one coat of P & B 21 was applied, left undisturbed for 30 s, air dried and cured for 20 s. A second coat of P & B 21 was then applied, air dried for 2 3 s and cured for 20 s. Spectrum TPH composite was placed as before. Group IX. Etching and priming was carried out as before. Chitrabond 10 resin was applied, air dried and cured for 20 s each. A second coat of the resin was then applied, air dried and cured for another 20 s each. Chitra composite was placed as before. All the lled specimens were polished with composite nishing and polishing points in a low speed (4030 rpm) air motor handpiece*. The polished speci

Shofu, Kyoto, Japan.

Fig. 7. Marginal leakage at a typical dentine enamel/composite interface showing reduced leakage at the enamel/composite interface viewed under an optical microscope (175 ).

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 6877

MARGINAL LEAKAGE OF COMPOSITES

Table 5. Quantitative measurement of marginal leakage in systems studied
NBA Bonding agent Z100/Single Bond No 1 2 3 4 E/C 15 15 10 35 1875 1 2 3 4 5 6 25 25 D/C 60 35 50 100 6125 50 50 CG 25 60 Single coat E/C 20 10 10 13.33 5 10 0 5 D/C 10 10 20 13.33 25 10 15 10 CG 30 25 50 35 25 15 25 25 Double coat E/C 10 D/C 20 CG 25

73

Mean Spectrum TPH/P & B 21

425 50 50

10 15 10 0

20 10 5

25 25 25

Mean Chitra/Chitrabond 10 1 2 3 4 5 6

25 15 5 25

50 50 25 60

50 15 25 25

5 10 0

15 20 20

225 25 25

833 10 0 0 0

75 75 0 10 15

25 20 15 50

Mean

15

45

2167

20

25

25

56

2833

All values are in microns. NBA, no bonding agent; E/C, enamel/composite interface; D/C, dentine/composite interface; CG, contraction gap.

mens were then stored in distilled water at 37 C for 24 h. Thermal cycling. All the specimens were subjected to thermal cycling at 6, 37, 54 and back to 37 C completing one cycle with a 30 s dwell time. Each sample was subjected to a minimum of 100 cycles. A Bromma Multitemp II thermostatic circulator (model LKB 2219) was used to maintain the temperature constant at 6 C, a Brabender water circulating bath (model T150) was kept at 54 C and two constant temperature water baths kept at 37 C completed the thermal cycling setup. Clear nail varnish was applied over all the thermallycycled tooth specimens except 1 mm around the restoration. Green stick compound was also used to seal the apex completely. Silver staining and embedding. Each group of teeth were placed in 10 ml each of freshly prepared 50% silver

nitrate solution for 2 h in darkness. The teeth were then retrieved, washed thoroughly in distilled water, stored in developing solution for 24 h exposed to sunlight. They were again rinsed in distilled water. The stained teeth were embedded in an epoxy resin (Araldite 502 EM media) taken in polythene cups. Curing time for the resin was 48 h. The epoxy blocks containing the embedded teeth were taken out of the plastic moulds, sectioned buccolingually into specimens of 0305 mm thickness using a low speed diamond saw**. The sectioned specimens were then polished on an Ecomet 3 variable speed grinder polisher** for 15 min on a felt cloth. Samples chosen for SEM observations were polished further with 005 m alumina slurry for 3 h. Optical microscopy and SEM studies. Selected specimens were viewed and photographed under an optical microscope (model FX 35A) to detect marginal leakage. The dentine/composite, enamel/composite interfaces and the contraction gap were observed and measurements made. A grading system for evaluating mi** Buehler Isomet, IL, U.S.A.

Made for LKB Products, Bromma, Sweden by Gebruder Haake GmbH, Germany. Brabender, Duisberg, Germany.

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 68 77

74

M . M A T H E W et al. croleakage was followed strictly (Table 4) for all the samples. Samples with gross cracks and air-bubble entrapment were rejected. For SEM analysis, the polished samples were mounted on a metallic stub, coated with gold in an ion sputtering unit (E-101, Hitachi Ion Sputter) and viewed under a scanning electron microscope (Hitachi model S 2400) at a magnication of 200 1000 .

Results and discussion

The marginal seal is a critical factor for a successful dental restoration. Poor marginal seal can result in plaque and bacterial deposition resulting in caries and periodontal disturbances. All the sectioned tooth specimens from the nine groups used in the study could be

Fig. 9. (a) Dentine/composite interface showing reduced marginal leakage in cavities restored with Spectrum TPH composite with a single coat application of P & B 21 bonding agent viewed under an optical microscope (350 ). (b) Dentine/composite interface showing reduced marginal leakage in cavities restored with Spectrum TPH composite with a single coat application of P & B 21 bonding agent viewed under SEM.

Fig. 8. (a) Dentine/composite interface showing reduced marginal leakage in cavities restored with Z100 composite with a single coat application of Single Bond bonding agent viewed under an optical microscope (175 ]. (b) Dentine/composite interface showing reduced marginal leakage in cavities restored with Z100 composite with a single coat application of Single Bond bonding agent viewed under SEM.

Hitachi, Tokyo, Japan.

viewed under the optical microscope with ease by the silver staining technique. Silver appeared mostly in brownish black layers in most of the specimens observed. The layers develop as a result of diffusion of silver ions into the porosity or marginal defects as a result of capillary action which is complete within hours. The formation of staining at the subsurface of composites (Fig. 1) may be due to the diffusion of silver ions into the composite and its subsequent precipitation as metallic silver within microcavities in the material. Penetration of silver ions and their precipitation in dentinal tubules appeared totally black. The silver staining technique can therefore be said to provide a sharp picture of the marginal penetration. Also, the degree of colour formation due to the staining of silver ions is also an indication as to the nature and degree of marginal leakage as has been previously observed by Mair (1989).

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 6877

MARGINAL LEAKAGE OF COMPOSITES

75

Fig. 10. (a) Dentine/composite interface showing reduced marginal leakage in cavities restored with Chitra composite with a single coat application of Chitrabond 10 bonding agent viewed under an optical microscope (350 ). (b) Dentine/composite interface showing reduced marginal leakage in cavities restored with Chitra composite with a single coat application of Chitrabond 10 bonding agent viewed under SEM.

Fig. 12. (a) Dentine/composite interface showing negligible marginal leakage in cavities restored with Spectrum TPH composite with a double coat application of P & B 21 bonding agent viewed under an optical microscope (350 ). (b) Dentine/composite interface showing negligible marginal leakage in cavities restored with Spectrum TPH composite with a double coat application of P & B 21 bonding agent viewed under SEM.

Fig. 11. Dentine/composite interface showing negligible marginal leakage in cavities restored with Z100 composite with a double coat application of Single Bond bonding agent viewed under an optical microscope (350 ).

Another important observation made during the study was the formation of microcracks especially at the dentine/composite interface (Fig. 2). One of the main reasons for this could be attributed to the exhaustive thermal cycling carried out on the samples. It can be said to determine the magnitude of marginal leakage into dentine/composite interface in class II cavities substantially. It is found to affect the polymerization shrinkage and crack formation at the dentinal surface to a large extent. Many of the studies carried out earlier on microleakage did not subject the samples to thermal cycling. However, our observations indicate that an accurate behaviour of the leakage properties at the dentine/composite or enamel/composite interface can be represented only in tooth samples subjected to thermal cycling.

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 68 77

76

M . M A T H E W et al. Development of the cracks was observed mainly from the pulp area due to its brittle nature which tends to disintegrate during thermal cycling and peel off while specimens are sectioned. Most of the cracks are found to travel from these disintegrated areas right up to the dentine/composite interface. So, most of our observations were limited to specimens devoid of any cracks as formation of cracks at the dentine/composite interface is likely to induce penetration of silver ions through these cracks, limiting the accurate observation of microleakage. Figure 3 a and b show the unpolished and polished surfaces of the sectioned specimen viewed under an optical microscope. Polishing the specimen in 005-mm alumina slurry using a felt cloth as an abrading medium for a period 2 3 h was found to improve the surface characteristics considerably. The common error committed during marginal leakage observations where the dentinal depth is often misinterpreted as leakage due to surface variations at composite and dentine layers could thus be avoided. Weak surface bonding at the dentine/composite interface in the absence of bonding agents in the case of the rst three groups has been evident from both SEM and photomicrographic observations which illustrate gap formation due to high polymerization shrinkage and this has resulted in heavy marginal leakage (Figs 4a, b, 5a, b and 6a, b). The mean leakage at the enamel/composite interface was small compared to the leakage at the dentine/composite interface ( 15 25 mm compared with 4565 mm at the D/C interface, Fig. 7). Table 4 illustrates the grading pattern obtained for the systems studied. The extent of quantitative leakage at the enamel/ composite interface, dentine/composite interface and the gap formation observed for all the good specimens studied in the nine groups is shown in Table 5. However, considerable reduction in leakage both at enamel and dentine interfaces with the composite was observed from optical and SEM photomicrographs (Figs 8a, b, 9a, b and 10a, b) when a recommended bonding agent (Single Bond in the case of Z100, P & B 21 in the case of Spectrum TPH, Chitrabond 10 in the case of Chitra composite) was applied. Tables 4 and 5 and the microscopic observations reveal greater adhesion of composite to the tooth surface when a bonding agent was used. The magnitude of adhesion increased with applying two coats of the bonding agent where the leakage has been minimal in most of the cases observed (Figs 11, 12a, b and 13a, b) for all systems studied. It can be observed that the values obtained for the indigenously developed Chitra composite/Chitrabond 10 system was highly comparable, and even lesser, than imported controls such as Spectrum TPH and Z100 (15 and 45 mm without bonding agent at E/C and D/C interfaces, respectively; 5 and 20 mm with one coat of bonding agent; and 25 and 56 mm with two coats of bonding agent application). Scanning electron photographs and optical photographs substantiate these observations.

Fig. 13. (a) Dentine/composite interface showing negligible marginal leakage in cavities restored with Chitra composite with a double coat application of Chitrabond 10 bonding agent viewed under an optical microscope (350 ). (b) Dentine/composite interface showing negligible marginal leakage in cavities restored with Chitra composite with a double coat application of Chitrabond 10 bonding agent viewed under SEM.

Conclusions
Application of a suitable bonding agent is necessary in minimizing the marginal leakage in restorations ensur-

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 6877

MARGINAL LEAKAGE OF COMPOSITES ing strong adhesion of the composite to the dentinal margin. Applying a second coat of bonding agent is found to improve the adhesion characteristics of the composite resulting in reduced leakage compared to a single coat application. The indigenously developed Chitra radiopaque composite/Chitrabond 10 system is found to compare well with the Z100/Single Bond and Spectrum TPH/P & B 21 systems in marginal leakage aspects.
DARBYSHIRE, P.A., MESSER, L.B. & DOUGLAS, W.H. (1988) Microleakage in Class II composite restorations bonded to dentine using thermal and load cycling. Journal of Dental Research, 67, 585. GOING, R.E. (1972) Microleakage around dental restorations: a summarizing review. Journal of American Dental Association, 84, 1349. KRISHNAN, V.K., BINDHU, D.B. & MANJUSHA, K. (1996) Studies on microleakage associated with visible light cured dental composites. Journal of Biomaterials Applications, 10, 348. KUBIA, K.E. & JACOBSEN, P.H. (1991) Microleakage in Class II composite resin restorations. Odontostomatological Progress, 44, 187. MAIR, L.H. (1989) An investigation into the permeability of composite materials. Dental Materials, 5, 109. PHILLIPS, R.W (1965) New concepts in materials used for restorative dentistry. Journal of American Dental Association, 70, 652. QVIST, V. (1983) The effect of mastication on marginal adaptation of composite restorations in vivo. Journal of Dental Research, 62, 904. REES, J.S. & JACOBSEN, P.H. (1989) Polymerization shrinkage of composite resins. Dental Materials, 5, 41. RETIEF, D.H. (1994) Do adhesives prevent microleakage? International Dental Journal, 44, 19. RIGSBY, D.F., RETIEF, D.H., BIDEZ, M.W. & RUSSEL, C.M. (1992) Effect of axial load and temperature cycling on microleakage of resin restorations. American Journal of Dentistry, 5, 155. ROSSOMANDO, K.J. & WENDT, S.L. JR (1995) Thermocycling and dwell times in microleakage evaluation for bonded restorations. Dental Materials, 11, 47. SHORTALL, A. & ASMUSSEN, E. (1988) Inuence of dentine bonding agents and a glass ionomer base on the cervical marginal seal of class II composite restorations. Scandinavian Journal of Dental Research, 96, 590. WU, W., COBB, E, DERMANN, K. & RUPP, N.W. (1983) Detecting margin leakage of dental composite restorations. Journal of Biomedical Materials Research, 17, 37.

77

Acknowledgments
The authors wish to thank the Director and Head, SCTIMST, Trivandrum for providing facilities for carrying out this work. They are grateful to Drs Mira, Annie John and Ms Sulekha for helping with the embedding and sectioning of the samples, and Mr Sreekumar for the SEM photographs. They also wish to acknowledge with thanks the free samples provided by Ms Birla, 3M Co., Bangalore.

References
ALANI, A.H. & TOH, C.G. (1997) Detection of microleakage around dental restorations: a review. Operative Dentistry, 22, 173. ASMUSSEN, E. & MUNKSGAARD, E.C. (1985) Bonding of restorative resins to dentine promoted by aqueous mixtures of aldehydes and active monomers. International Dental Journal, 35, 160. BRANNSTROM, M. (1986) The cause of post-restorative sensitivity and its prevention. Journal of Endodontics, 12, 475. COX, C.F., KEALL, C.L., KEALL, H.J., OSTRO, E. & BERGENHOLTZ, G. (1987) Biocompatibility of surface sealed dental materials against exposed pulps. Journal of Prosthetic Dentistry, 57, 1. CRIM, G.A. & MATTINGLY, S.L. (1981) Evaluation of two methods of assessing marginal leakage. Journal of Prosthetic Dentistry, 45, 160.

Correspondence to: Dr Kalliyana Krishnan V, Biomedical Technology Wing, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Poojappura, Trivandrum 695 012, Kerala, India. E-mail: bmtwing@md2.vsnl.net.in

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 68 77