JOURNAL OF ENDODONTICS Copyright © 2001 by The American Association of Endodontists
Printed in U.S.A. VOL. 27, NO. 3, MARCH 2001
Adhesion of a Glass-Ionomer Root Canal Sealer to the Root Canal Wall Siriporn Timpawat, DDS, MS, Choltacha Harnirattisai, DDS, PhD, and Pisol Senawongs, DDS, MS
compared with Grossman’s sealer, a GIC sealer showed excellent and superior adaptation to the canal wall (3). The choice of sealer may influence the outcome of endodontic therapy (6). Friedman et al. (7) reported a high success rate of using a GIC sealer in endodontically treated teeth, after 6 to 18 months recall (78.3%, with only 6.1% failure). Stewart (8) described the use of calcium hydroxide to promote healing of root fractures, followed by reinforced GIC (Ketac-silver, Espe, Premier, Norristown, PA) to strengthen the root. Barkhordor (9) successfully used GIC (Chelon-silver, Espe, GmbH) in the management of a root fracture, as a root canal sealer and then condensed into the root canal to bind the segments together. An experimental investigation into the fracture resistance of endodontically treated roots using a recently developed GIC (Ketac-Endo, Espe, Seefeld, Germany) as root canal sealer demonstrated that obturation of the canals in conjunction with a glass-ionomer sealer significantly strengthened the root, compared with roots instrumented but not obturated and those obturated with gutta-percha and Roth’s 801 Sealer (10). After the root canal space is instrumented it may not provide an ideal environment for the maximal bonding of a glass ionomer sealer to the dentin (11). During instrumentation a tenacious layer of debris (the smear layer) is formed. Some studies (12, 13) have shown that removal of the smear layer enhances adhesion of the GIC to the root canal wall. The highest tensile strength values were found following EDTA/NaOCl treatment. This study was designed to determine the tensile bond strength of a GIC root canal sealer (Ketac-Endo, Espe, Seefeld, Germany) to the root canal walls after pretreatment with various surface conditioners. Acidic surface conditioners used in restorative dentistry were compared with EDTA, which is the standard endodontic irrigant for removal of the smear layer from the canal wall. Furthermore the effects of these conditioning solutions on the smear layer were investigated using scanning electron microscopy (SEM).
Glass-ionomer root canal sealer is commonly used because of its chemical bonding and favorable physical characteristics when bonding to dentin. This study was designed to determine the tensile bond strength of a glass-ionomer sealer (Ketac Endo, Espe, Seefeld, Germany) on root canal walls after pretreatment with different conditioners. Flat inner surfaces of root canal specimens were prepared. The specimens were divided into five groups of 10 teeth, and the groups were conditioned with one of the following smear layer removal solutions: 15% EDTA/NaOCl, 10% polyacrylic acid, 35% phosphoric acid, 6% citric acid, and 5.25% NaOCl as a control. Then the exposed root canal areas were coated with Ketac-Endo. Tensile bonding was measured using a universal testing machine until ultimate failure was obtained. The groups that were treated with phosphoric acid and citric acid showed significantly higher bond strengths than the groups that were treated with 15% EDTA and polyacrylic acid (p < 0.05). Bonding to dentin without smear layer removal (NaOCl group) was too low to be measured in the testing apparatus. Scanning electron microscopy confirmed that phosphoric and citric acids were more effective in removing smear layer than EDTA or polyacrylic acid. The result supported the view that pretreatment with phosphoric acid or citric acid should be used in association with glass-ionomer root canal sealer to achieve the most effective removal of the smear layer and to provide better adhesion.
MATERIALS AND METHODS Glass-ionomer cement (GIC) is known to have favorable biological, chemical, and physical characteristics, including tissue compatibility (1, 2) and the ability to adhere to dentin and enamel (3). A GIC-based root canal sealer might exhibit long-term adhesion to dentin, which would be an obvious advantage over zinc oxideeugenol–type or epoxy resin-type sealer cements (4, 5). Also,
Fifty-five sound extracted premolar teeth with a single root canal and mature root apices were used. The teeth were randomly divided into five groups of 10 teeth each and five teeth were used for the SEM study. The root canals were instrumented using conventional hand files (K-type, Kerr, Romulus, MI) up to size 40. The roots were prepared for adhesion tests as follows: longitudinal 168
Vol. 27, No. 3, March 2001
FIG 1. Schematic section of a specimen prepared for tensile testing. The arrow represents the force applied from the universal testing machine that is perpendicular to the root canal surface.
grooves were made on the buccolingual surface of the root to a depth that almost exposed the root canal wall. Roots were then split into two parts. One half of the root was used for testing. The flat inner wall of the root canal was further prepared by smoothing the canal wall with 600 grit silicone carbide paper. Polyethylene rings 20 mm in inner diameter and 20 mm in height were filled with polymethylmethacrylate resin, and the roots were placed into the resin, leaving the flat surface of the root at the level of the upper part of polyethylene ring (Fig. 1). After the resin had set for 1 h, the cut surface of the root was ground flat with 600 grit carbide paper. Ten specimens were prepared for each conditioner treatment. The conditioners that were applied were 5.25% NaOCl (negative control, with no smear layer removal), 10% polyacrylic acid (GC Corporation, Tokyo, Japan), 15% EDTA (Largal Ultra, Septodont, Specialites Septodont, Cedex, France), followed by 5.25% NaOCl, 6% citric acid (prepared by the Biochemistry and Physiology Department, Faculty of Dentistry, Mahidol University), and 35% phosphoric acid (3M Dental Products, St. Paul, MN). Each conditioner was applied to the dentin surface for 60 s. After the surface was rinsed and dried with an oil-free air stream a plastic mold 4 mm in diameter and 10 mm in height was centered over the flat dentin and attached to the dentin by sticky wax. Ketac-Endo was mechanically mixed for 10 s, then injected into the plastic mold. A wire loop was attached to the GIC. The specimens were stored in 100% relative humidity at 37°C for 48 h before the tensile bond strength was tested. After 48 h the sticky wax was removed from the attached dentin. The specimens were transferred to a testing holder (14) that could be attached to the cross-head of the Universal Testing Machine (Instron, system ID, 5566H1612, Instron LTD, Buckinghamshire, UK). The wire loop embedded in the GIC was used to pull the GIC button off the tooth surface at a cross-head speed of 0.5 mm/min. The tensile load required to fracture the bond was recorded, and the bond strength was calculated in MPa. The statistical significance of the difference in results between groups was determined using the Newman-Keul’s test for multiple comparisons. Preparation for the scanning electron microscope: Five specimens, used to determine the extent of smear layer removal, were prepared as previously stated. One specimen had the smear layer retained (5.25% NaOCl rinse only). The other four had the smear layer removed with the four conditioning solutions as follows: 15%
Adhesion of a Glass-Ionomer Root Canal Sealer
FIG 2. Dot plots of tensile bond strengths (MPa) for each group (n ⫽ 10). The very low adhesion of samples exposed only to NaOCl led to failure before testing.
EDTA, 10% polyacrylic acid, 35% phosphoric acid, and 6% citric acid. The conditioners remained on the specimens for 60 s and were then rinsed with water for 30 s, then air-dried for 30 s as described. All specimens were mounted on aluminum stubs, vacuum-dried, and sputter-coated with gold. The root canal surfaces were observed under the scanning electron microscope (JEOL, JSM5410LV, Tokyo, Japan) at ⫻2000 magnification.
RESULTS The tensile bond strengths and the distribution of the individual values are illustrated in Fig. 2. The graph does not include data for the specimens that were treated with NaOCl alone, because the GIC detached from the root surface at the time of starting the tensile test, indicating that there was no detectable adhesion when the smear layer was not removed. In the groups with the smear layer removed bond strength ranged from ⬃0.1 to 1.4 MPa, with the greatest variation in the phosphoric acid and citric acid groups. The mean bond strengths for roots treated with phosphoric acid (mean MPa) and citric acid (mean MPa) were similar, and showed a significant difference (p ⬍ 0.05) from those treated with EDTA (mean MPa) or polyacrylic acid (mean MPa). There were no significant differences between the root surfaces treated with EDTA and polyacrylic acid (p ⬎ 0.05). Observation of the fracture surfaces after testing revealed that failure generally occurred close to the dentin surface (adhesive failure), although some specimens were fractured within the GIC sealer. SEM analysis: The control specimens cleaned only with NaOCl disclosed a smear layer covering all dentinal tubules (Fig. 3). With all conditioning agents the smear layers were removed and the dentinal tubules were opened to varying extents. Some smear layer was left when the surface of the dentin was cleaned with polyacrylic acid (Fig. 4). The majority of the tubule orifices were not entirely exposed. The surface treated with EDTA disclosed removal of almost all of the smear layers and smear plugs (Fig. 5), but some tubule orifices were not entirely exposed. Surface treatment with 35% phosphoric acid and 6% citric acid removed the entire smear layer and smear plugs (Figs. 6 and 7).
Timpawat et al.
Journal of Endodontics
FIG 3. Canal surface treated with 5.25% NaOCl. A rough smeared appearance is visible and only occasional small openings of the dentinal tubules are seen. Some scratches are present from preparing specimens with the 600 grit silicone carbide paper. (Original magnification ⫻2000.)
FIG 4. Canal surface treated with 15% EDTA. Dentinal tubules are wide open, and the canal surface is smooth and clean. There is evidence of smear plugs in tubule orifices. (Original magnification ⫻2000.)
DISCUSSION There are limitations when measuring the adhesion of cements or root canal sealer to the root canal walls. The curved area of the root canal segment does not have an ideal geometry for the measurement of tensile strengths. Weiger et al. (13) tested the adhesion of GIC (Ketac-Cem) to human radicular dentin, and recommended the use of root canal segments with large diameters to minimize the tangential forces that might occur and induce a fracture pattern at the interface. In this study we enlarged the root canals to a size 40 file and also flattened the inner surface of the root canal with 600 grit silicone carbide paper. We developed an apparatus similar to Fusayama et al.’s (14) apparatus. This apparatus has balance weights attached to eliminate the influence of the weight of the puller and enables the determination of even very low adhesion such as is encountered with dental cements. This nonpressure tensile adhesion test is similar to the clinical use of root canal cement, with no substantial pressure. A previous study (13) determined the tensile bond strength of a GIC (Ketac-Cem) on root canal walls after pretreatment with
FIG 5. Canal surface treated with 10% polyacrylic acid shows visible dentinal tubules, and the canal wall appears clean. Some dentinal tubules are occluded by a smear plug. (Original magnification ⫻2000.)
FIG 6. Canal surface treated with 35% phosphoric acid, shows completely exposed dentinal tubules and patent dentinal tubules. Some tubules appear occluded by a dentinal plug. No evidence of smear layer is present, and intertubular dentin appears to be corroded by acid. (Original magnification ⫻2000.)
different conditioners. Surface pretreatment with EDTA/NaOCl provided a mean bond strength of 2.2 MPa, whereas surfaces pretreated with NaOCl gave the lowest bond strength (0.5 MPa). Acid pretreatment resulted in intermediate values. The present study demonstrated a lower adhesion of Ketac-Endo, compared with Ketac-Cem. One reason is that Ketac-Endo is a luting cement especially formulated for endodontic use. Another possible reason is that the previous experiment measured tensile strengths on the curved inner wall of root canals rather than on the flatter surface. In this study the mean bond strengths obtained when EDTA/ NaOCl and polyacrylic acid were used for surface treatment were found to be similar, and lower than for surfaces treated with 35% phosphoric acid or 6% citric acid. This may be due to increased contact surface in the dentin due to the smear layer being removed. Scanning electron micrographs from this experiment showed that with the 6% citric acid and 35% phosphoric acid surface treatment the canal walls were cleaner, and there were more patent dentinal tubules than with polyacrylic acid or EDTA/NaOCl. This observation clearly indicates that mechanical retention may occur from penetration of cements into decalcified and opened dentinal tu-
Vol. 27, No. 3, March 2001
Adhesion of a Glass-Ionomer Root Canal Sealer
sealer may not be as important as its sealability to prevent microleakage. The results of this study suggest that citric or phosphoric acid should be considered as an alternative to EDTA for the routine removal of the smear layer, regardless of the sealer cement used. This study was supported by a grant from Mahidol University Research Fund. The authors are grateful to Professor Harold H. Messer, Department of Restorative Dentistry, School of Dental Science, University of Melbourne, Victoria, Australia, for his advice, and for reviewing and correcting this manuscript. We thank Professor Junji Tagami, Department of Operative Dentistry, Tokyo Medical and Dental University, Tokyo, Japan, for his advice on apparatus testing and for his insight in preparing the manuscript. Also we are grateful to Mr. Apiwat Rittapai, Research Officer, Faculty of Dentistry, Mahidol University, for his assistance in SEM preparation.
FIG 7. Surface treated with 6% citric acid shows open dentinal tubules with evidence of smear plugs in some dentinal tubules. All of the smear layer has been removed. (Original magnification ⫻2000.)
bules, which may contribute to the higher bond strength in those two groups. The removal of the smear layer present on instrumented dentin is recommended before coating coronal dentin with GIC to obtain maximal bond strengths (4, 15). The results from this study demonstrated that the tensile strength correlated with the extent of removal of the smear layer. Most specimens demonstrated adhesive failure that occurred close to the contact surface of dentin. Some specimens demonstrated cohesive failure, which was noted in every surface treatment group. This finding indicates that tensile strengths equal to the cohesive strength of the GIC may be attained. Voids in GIC due to air entrapment were observed in some specimens. This factor may cause a weak cohesive bonding within the material. Preventing voids in the sealer is very important, because when the root canal is obturated sealer should be adapted closely to the root canal wall. In other studies (3, 7) voids in glass-ionomer sealer were also noticed when it was used as a sealer in the root canals. It was still found to have superior adaptation to the canal wall, compared with Grossman’s sealer. We conclude from this study that smear layer should be removed by using phosphoric acid or citric acid before GIC is used as a root canal sealer to get a higher bond strength of the cement to the root canal walls. This study was in agreement with Gettleman et al. (16) who found that sealers had a strong bond when the smear layer was removed. However more studies should be conducted on other aspects, such as the adaptation of this cement, as well as clinical studies. Even when smear layer was removed with these acids, the bond strength of this sealer was low, compared with those of GIC because root canal sealer is regarded as a luting cement. In clinical practice the tensile bond strength of root canal
Drs. Timpawat, Harnirattisai, and Senawongs are affiliated with the Department of Operative Dentistry, Faculty of Dentistry, Mahidol University, Bangkok, Thailand. Address requests for reprints to Dr. Siriporn Timpawat, Department of Operative Dentistry, Faculty of Dentistry, Mahidol University, 6 Yotee Street, Phayathai, Bangkok 10400, Thailand.
References 1. Jonck LM, Grobbelaar CJ, Strating H. Biological evaluation of glass ionomer cement (Ketac-O) as an interface material in total joint replacement: a screening test. Clin Mater 1989;4:201–24. 2. Blackman R, Gross M, Seltzer S. An evaluation of the biocompatibility of a glass ionomer-silver cement in rat connective tissue. J Endodon 1989; 15:76 – 80. 3. Ray H, Seltzer S. A new glass ionomer root canal sealer. J Endodon 1991;17:598 – 603. 4. Powis DR, Folleras T, Merson SA, Wilson AD. Improved adhesion of a glass ionomer cement to dentin and enamel. J Dent Res 1982;61:146 –22. 5. Aboush YEY, Jenkins CBG. An evaluation of the bonding of glass ionomer restorative to dentin and enamel. Br Dent J 1986;161:179 – 84. 6. Orstavik D, Kerekes K, Eriksen HM. Clinical performance of three endodontic sealers. Endod Dent Traumatol 1987;3:178 – 86. 7. Friedman S, Lost C, Zarrabian M, Trope M. Evaluation of success and failure after endodontic therapy using a glass ionomer cement sealer. J Endodon 1995;21:384 –90. 8. Stewart GG. The detection and treatment of vertical root fractures. J Endodon 1988;14:47–53. 9. Barkhordar RA. Treatment of vertical root fracture: a case report. Quint Int 1991;22:707–9. 10. Trope M, Ray H. Resistance to fracture of endodontically treated roots. Oral Surg Oral Med Oral Pathol 1992;73:99 –102. 11. Trope M, Tronstad L. Resistance to fracture of endodontically treated premolars restored with glass ionomer cement or acid etch composite resin. J Endodon 1991;17:257–9. 12. Hotz P, McLean JW, Sced I, Wilson AD. The bonding of glass ionomer cements to metal and tooth substrates. Br Dent J 1977;142:41–51. 13. Weiger R, Heuchert T, Hahn R, Lost C. Adhesion of a glass ionomer cement to human radicular dentin. Endod Dent Traumatol 1995;11:214 –9. 14. Fusayama T, Masato N, Kurosaki N, Iwaki M. Non-pressure adhesion of a new adhesive restorative resin. J Dent Res 1979;58:1364 –70. 15. Beech DR, Solomon A, Bernier R. Bond strength of polycarboxylic acid cements to treated dentin. Dent Mater 1985;1:154 –7. 16. Gettleman B, Messer HH, ElDeeb ME. Adhesion of sealer cements to dentin with and without the smear layer. J Endodon 1991;1:15–20.