JOURNALOF ENDOOONTICS Copyright © 1999 by The American Association of Endodontists
Printed in U.S.A VOL. 25, No. 3, MARCH 1999
In Vitro Sealing Ability of Dentin Bonding Agents and Cavity Varnish with Amalgam as Retrofills RiJstem Kemal S/ibay, DMD, PhD, and Arzu S~ibay, DMD
The aim of this study was to test the effect on the seal produced by an amalgam retrofill of two dentin bonding agents (Amalgambond and All-Bond) and a cavity varnish (Copalite). The three materials were tested when placed before the retrofill (group I) and before and after the retrofill (group II). Ninety teeth distributed into three groups were prepared for the retrofilling. Each group was divided into subgroups of I and I1. Linear apical leakage was measured by a stereomicroscope after 1 month in dye. All materials permitted dye leakage around the retrofill, regardless of the application type. No statistical difference was found between the Amalgambond subgroups or between the All-Bond subgroups. There was less leakage in varnish group I than group I1. Amalgambond group I leaked significantly less than All-Bond group I.
several microorganisms into the root dentin tubules (6). Furthermore, Vertucci and Beatty (7) pointed out the importance of open dentinal tubules on the resected surface that allow dye penetration into the root canal. They concluded that the presence of a varnish under the apical retrofill does not prevent the dye diffusion into the canal unless the entire resected surface is coated with a cavity varnish (7). Microleakage through the open tubules of the resected root has been reported as significantly greater than through the unresected root apex, and it was concluded that necrotic and possibly infected debris leakage through the patent dentinal tubules may be the main cause of the residual inflammation after apicec tomy (8). On the other hand, newer dentin bonding agents have been shown to penetrate into the dentin tubules and react with the collagen and carbonated hydroxyapatite of the intertubular dentin (9). Rud et al. (10) reported that composite bonded to the resected root surface with a dentin bonding agent (Gluma) showed complete healing after apical surgery in 74% of cases after 1 yr, which was significantly higher than the amalgam retrofill. An in vitro study by Ambus and Munksgaard (11) reported that dentin bonding agents in combination with a composite could create a gap-free root end filling. Vignaroli et al. (12) indicated that dentin bonding agents (Amalgambond, Scotchbond Multi Purpose Plus, and All-Bond 2) significantly reduced apical microleakage when used as root-end sealants. The aim of this study was to test the sealing ability of two commercially available dentin bonding agents (Amalgambond and All-Bond) and cavity varnish (Copalite) when used with an amaL gam retrofill (Dispersalloy). The agents were placed into the rootend cavity only before the retrofill, or both into the cavity before the retrofill and on the entire root end after the retrofill.
Despite the success of conventional endodontic therapy, certain cases associated with periapical pathosis present additional challenges, such as canals with broken instruments, sectioned silver points, and large post and core preparation. Regarding the management of these various clinical complications, the literature indicates surgical endodontic therapy (I). Apicectomy, along with the placement of a retrofilling, is done mainly to provide an impervious seal at the root end against possible bacterial leakage. Complete seal of the prepared root-end cavities with a retrofill material is generally considered as the proper clinical-biological objective: to maintain the heaJing and reorganization of periradicular tissues afiter surgery (2). A variety of dental materials have been used for a surgical retrofilling procedure (3). However, amalgam is the most used retrofilling material. Some studies recommended the use of a cavity varnish before placement of an amalgam retrofill to decrease the initial apical leakage (4, 5). Varnish is thought to seal gaps that may range from 6 to 150/xm between the cavity and amalgam; this gap may in turn be filled by corrosion products as the varnish dissolves with time (2). Although the root canal is considered the main biological location for microbiological growth and efforts have been concentrated to seal it, in vitro studies showed the penetration capability of
M A T E R I A L S AND M E T H O D S A total of 96 freshly extracted single-rooted anterior human teeth with mature apices were used for this in vitro study. All teeth were stored in a physiological lactated Ringer's solution with 0.2% sodium azide in the refrigerator at 0°C. The anatomical crown of each tooth was resected before canal preparation using a #331 carbide bur in a high-speed handpiece with water cooling. The working length was determined using a #15 K-type file in the canals until it was seen at the apical foramen and then subtracting 1 ram. The apical foramen was enlarged to #35 K-type file, and the canal flared coronally to a #45 K-type file by using the step-back
S~ibay and S~ibay
filing technique at 1-mm increments. The coronal half of each root was flared using Gates Glidden burs. Each canal was irrigated with 2% sodium hypochlorite and dried with paper points. The canals were obturated using the lateral condensation technique with guttapercha (Hygenic, Hygenic Corp., Akron, OH) and zinc oxideeugenol sealer (Roth root canal sealer, Type 811, Rotb Intl. Ltd., Chicago, IL). The coronal surfaces were sealed with IRM (L. D. Caulk Div., Dentsply Intl., Inc., Milford, DE) and two coats of nail polish (Artmatic, Arthur Matney Co., Inc., Brooklyn, NY), and then placed onto the root surfaces of each root. The apices of the roots were then resected beginning from the point on the lingual surface 2 mm from the apex to the facial surface ending at 3 mm from the apex with a #331 carbide bur in a high-speed handpiece and constant water spray. Class i cavities were prepared in all of the root ends to a depth of 3 mm from the facial side with a high-speed #331 carbide bur under water cooling. Triturated amalgam alloy (Dispersalloy, Caulk/Dentspty) was used as the retrofill in all teeth. The cavities in three teeth used as negative controls were filled with the amalgam and the root-end surface covered with two coats of nail polish. Another three teeth served as positive controls; cavities were filled with the amalgam, but no nail polish was applied onto the root-end surface. The remaining 90 teeth were randomly distributed in six test groups containing 15 teeth of each. In Amalgambond group I, the cavities were treated with Amalgambond dentin bonding system (Amalgambond, Parkell, Farmingdale, NY) according to the manufacturer's recommendations and filled with the amalgam. After the initial setting of the amalgam (3 min), the resected surface (including the amalgam retrofill) was slightly resurfaced with a highspeed #331 carbide bur and water spray to remove the excess bonding system on the resected root surface. In Amalgambond group lI, the beveled root end, and the cavity were coated with the same dentin bonding system, filled with the amalgam retrofitl, and then the root end recoated with the bonding material. In the groups, All-Bond I and II, All-Bond bonding system (Bisco Dental, Inc., Itasco, IL), and the amalgam were applied in the same manner as in Amalgambond groups, respectively. For the groups Copalite I and II, the root ends were treated with the cavity varnish Copalite (Bosworth, Skokie, IL) and amalgam as in Amalgambond groups. All specimens were stored in 2% methylene blue dye for 1 month at 37°C by changing the dye solution weekly. After 30 days, the roots were rinsed for 10 rain under tap water and dried at 37°C for 1 day. The nail polish was removed from the root surface by means of scraping with a sharp razor and by slow-speed diamond on a handpiece. The roots were sectioned in a buccal-lingual plane by notching buccal and lingual grooves with a diamond disc and then split into halves with a laboratory knife. After splitting, the remaining amalgam and gutta-percha plus sealer were cleaned from the roots. Linear measurements were determined using a stereomicroscope (American Optical Corporation, Buffalo, NY) at × 15 power and calculated with a traveling microscope (Gaertner Scientific Corp., Chicago, IL) at a 1 m/, accuracy. The extension of dye leakage was measured from the point at the beveled surface longitudinally to the furthest point of dye penetration in the cavity preparation. The cavity floor was considered as the maximal dye penetration point; dye extensions beyond this area were not recorded. Data were analyzed statistically by using t test, one-way analysis of variance, and Duncan's multiple comparison test at a significance level of p < 0.05.
Journal of Endodontics TABLE 1. Extent of linear apical leakage into the root-end cavities* (n = 15) Group Amalgambond Group I Group II All-Bond Group I Group II Copalite Group I Group II
* Statistical analysis: t-test--Copalite group I versus Copalite group II significant difference (p < 0.05); ANOVA--significant difference among the group l groups (p < 0.05); Duncan's multiple comparison test--Amalgambond group I versus All-Bond group I significant difference (p < 0 05).
RESULTS Complete leakage was observed along the root-end cavity walls in the positive control teeth, whereas those teeth in the negative control showed no dye leakage. The mean ± standard deviation for each test group is shown in Table 1. T-test analysis showed that there was no statistical difference between Amalgambond group I and II (Fig. l, a and b) or between All-Bond groups l and II (Fig. 2, a and b), whereas there was significantly less leakage in Copalite group I than group II (Fig. 3, a and b) (p < 0.05). One-way analysis of variance showed that there was no statistical difference among the three group II subsets. On the other hand, one-way analysis of variance showed that there was a statistical dye leakage difference among the first groups (p < 0.05). Duncan's multiple comparison test showed that Amalgambond group I leaked significantly less than All-Bond group 1 (p < 0.05). There was no other statistical differences among the first groups.
DISCUSSION The purpose of this study was to investigate the sealing efficacy of the two bonding agents, as well as a cavity varnish along with an amalgam retrofill. By placing the bonding systems and the cavity varnish only onto the root-end cavities before the amalgam retrofill (group I), and to the cavity and the beveled surface before the retrofill also to the root end alter the retrofill (group II), two alternative treatment types with the materials along with the amalgam retrofill were tested. According to the results of the present study, two bonding agents and the cavity varnish used--along with the amalgam retrofill--all allowed dye leakage into the root-end cavities, regardless of the treatment type. There were no statistical leakage differences between the groups when the same bonding agent was used. Neither for All-Bond nor for Amalgambond was there a statistically significant dye leakage discrepancy dependent on treatment type. Interestingly, the resurfaced cavity varnish group exbibited statistically less dye leakage than the other cavity varnish group. This result obtained with cavity varnish contradicts the findings of Vertucci and Beatty (7) who observed less leakage when the beveled root end and the root-end cavity were covered with the varnish. The longer experimental time of the present study possibly caused this difference, implying that the cavity varnish is not a suitable material to seal the open dentinal tubules on the beveled surface at long-term periods.
Vol. 25, No. 3, March 1999
Dentin Bonding with Amalgam Retrofill
FIG 1. Amalgambond type I specimen showing minimal leakage on the root-end cavity (a). Amalgambond type II specimen demonstrating minimal leakage on the root-end cavity (b).
FIG 2. All-Bond type I specimen showing moderate dye leakage along the root-end cavity (a). All-Bond type II specimen demonstrating minimal to moderate dye leakage on the root-end cavity (b).
We did not observe less leakage between the treatment groups of the tested bonding agents. Before the experiment, it was thought that the application of bonding agents to the resected root ends after the retrofill would prevent dye leakage. Dimensional instability of amalgam retrofill after setting may have caused the dye leakage by distorting the overlying bonding layer. There were dye leakage patterns along the cavity walls of the bonding treated teeth in the present study. The hydrolytic effects of
water may influence the interfacial area between the amalgam and the bonding agent and cause apical dye leakage without penetrating into the dentin. Ben-Amar et al. (13) observed dye penetration mostly between the tooth and restoration interface and occasionally in dentinal tubules. They concluded that the bond between the bonding agent (Amalgambond) and amalgam may diminish much more than the bond between the bonding system and dentin after applying loading stress and thermocycling on teeth. Saiku et al.
S~bay and S~bay
Journal of Endodontics
FIG 3. Copalite type I specimen s h o w i n g minimal dye leakage (a). Copatite type II specimen demonstrating moderate dye leakage on the r o o t - e n d cavity (b).
(14) demonstrated that the pattern of dye leakage seen around Amalgambond/amalgam fillings tends to be restricted to the resinamalgam interface, whereas the leakage in cavity varnish-amalgam fillings penetrates into the dentinal tubules. The eugenol content of certain dental materials has been shown to create bigger gap formations between the bonding agent and dentin (15). On the other hand, Rud et al. (10) did not find significant shear bond strength loss in roots endodontically obturated with a eugenol-containing sealer and retrofilled with bonding resin and composite. Moreover, Woody and Davis (16) reported that there were no significant leakage differences in cavities to which bonding resin was applied and that was pretreated with either a eugenol-containing temporary cement or eugenol-free cement. The eugenol content of the root canal sealer used in this study may have increased the dye leakage in the bonding resins groups. The authors thank Dr. Yuran Tank, of the University of Istanbul, Forrest Engineering Faculty, for his help with the photography. Dr. R. K. S0bay is associate professor and Dr. A. SL~bay is a resident, Department of Endodontics, School of Dentistry, University of Istanbul, Istanbul, Turkey. Address requests for reprints to Dr. R0stem Kemal S0bay, Department of Endodontics, School of Dentistry, University of Istanbut, Capa, Istanbul 34399, Turkey.
References 1. Weine FS, Gerstain H. Periapical surgery. In: Weine FS, ed. Endodontic therapy. 2nd ed. St. Louis: CV Mosby, 1976:332-42.
2. Friedman S. Retrograde approaches in endodontic therapy. Endod Dent Traumatol 1991 ;7:97-107. 3. Cohen S, Burns RC. Pathways of the pulp. 6th ed. Part One. Chap. 9. 1994:179-218. 4. Tronstad L, Trope M, Doering A, Hasselgren G. Sealing ability of dental amalgams as retrograde fillings in endodontic therapy. J Endodon 1983;9: 551-3. 5. Coen TJ, Wong M. Varnishes: the effect of a second coat on apical root leakage of retrofill amalgams. J Endodon 1992;18:97-9. 6. ~3rstravik D, Haapasolo M. Disinfection by endodontic irrigants and dressings of experimentally infected dentinal tubules. Endod Dent Traumatol 1990;6:142-9. 7. Vertucci FJ, Beatty RG. Apical leakage associated with retrofilling techniques: a dye study. J Endodon 1986;12:331-6. 8. Ichesho WR, Ellison RL, Corcoran JF, Krause DC. A spectrophotometric analysis of dentinal leakage in the resected root. J Endodon 1991;17: 503-7. 9. Nakabayashi N, Takarada K. Effect of HEMA on bonding to dentin. Dent Mater 1992;8:125-30. 10. Rud J, Munksgaard EC, Andreasen JO, Rud V, Asmussen E. Retrograde root filling with composite and a dentin-bonding agent. 1. Endod Dent Traumatol 1991 ;7:118-25. 11. Ambus C, Munksgaard EC. Dentin bonding agents and composite retrograde root filling. Am J Dent 1993;6:35-8. 12. Vignaroli PA, Anderson RW, Pashley DH. Longitudinal evaluation of the microleakage of dentin bonding agents used to seal resected root apices. J Endodon 1995;21:509-12. 13. BemAmar A, Liberman R, Rothkoff Z, Cardash HS. Long term sealing properties of Amalgambond under amalgam restorations. Am J Dent 1994;7: 14t-3. 14. Saiku JM, Germain HA, Meiers JC. Microleakage of a dental amalgam alloy bonding agent. Qper Dent 1993;18:172-8. 15. Hansen EK, Asmussen E. Influence of temporary filling materials on effect of dentin-bonding agents. Scand J Dent Res 1987;95:516-20. 16. Woody TL, Davis RD. The effect of eugenol-containing and eugenolfree temporary cements on microleakage in resin bonded restorations. Qper Dent 1992;17:175-80.