- Email: [email protected]
In Vitro Cytotoxicity of Medicated and Nonmedicated Gutta-percha Points in Cultures of Gingival Fibroblasts
JOURNAL OF ENDODONTICS Copyright © 2002 by The American Association of Endodontists
Printed in U.S.A. VOL. 29, NO. 1, JANUARY 2003
In Vitro Cytotoxicity of Medicated and Nonmedicated Gutta-percha Points in Cultures of Gingival Fibroblasts Susanne Szep, Dr med dent, Ljiljana Grumann, Karin Ronge, Anette Schriever, Dr med dent, Myriam Schultze, Dr med dent, and Detlef Heidemann, Prof Dr med dent
erties, calcium hydroxide, which can be administered in different forms, is the most frequently used medication. The dentinal tubules allow diffusion of materials placed into the root canal system. Furthermore, contact of these materials with periodontal tissues may occur via the apical foramen or accessory root canals. In this context the toxicity of a substance, i.e. its biocompatibility in contact with surrounding tissues, plays an important role. Primary toxicity can be investigated by means of gingival fibroblasts as so-called “target cells,” whereas the reaction of in vivo tissue depends on further parameters such as salivary flow rate and resident bacterial flora, mechanical, and chemical stimuli during food intake, hormonal status, and biorhythms. Despite these limitations, differentiated primary cell cultures and established cell lines are used to arrive at conclusions relevant to the in vivo situation. There are numerous studies available that deal with the toxicity of gutta-percha points and endodontic medications in the root canal (1–5). However, the cytotoxic potential of so-called “medicated gutta-percha points” recently introduced in the market, which contain chlorhexidine or calcium hydroxide as antiseptic agents, has so far not been investigated. There are studies available relating to the active substances contained in these points, but not when in combination with a gutta-percha carrier (1, 6 – 8). It was the aim of this study to compare the cytotoxicity of two medicated and four nonmedicated gutta-percha points using primary human gingival fibroblasts. Physiological and pathological changes of the cells and reactions and growth of the cell cultures, respectively, were evaluated microscopically and statistically.
This investigation was designed to test the cellular toxicity of two medicated (Roeko activ point and Roeko Calcium Hydroxide) and four nonmedicated brands of gutta-percha (GP) points (Antaeos, DeTrey White, Roeko color, and Roeko Top color). The test points were transferred into a culture medium including the GP-point material with a concentration of 6 mg/ml, and eluates were obtained after 72 h. Five milliliters of each eluate were pipetted onto fibroblast cultures, incubated, and subsequently stained. Mitotic rates, cell densities, and the distribution of normal cells, pathologically altered and dead cells were determined and correlated with control cell cultures. Roeko activ point (containing chlorhexidine) resulted in the highest number of dead cells. The difference was statistically significant in comparison with all other materials. Concerning all parameters mentioned, the cytotoxicity of the points containing calcium hydroxide (Roeko Calcium Hydroxide) was not significantly different from all other points tested, with the exception of those containing chlorhexidine. All tested gutta-percha materials caused cytotoxic reactions in varying extents. Taking into consideration the limitations of an in vitro experiment, points containing calcium hydroxide and nonmedicated points seem to be the most recommendable products for clinical use.
MATERIALS AND METHODS Inflammation-free human gingival tissue specimen were gained from surgical and periodontal operation sites. The explants were stored in a refrigerator at 4°C overnight. They were subsequently washed twice in Hanks’ balanced salt solution, which had been complemented with 5 ml of bicarbonate and an antibiotic additive and cooled to 4°C to remove blood and granulation tissue (all materials: Gibco-Life Technologies Ltd., Paisley, Scotland). The explants were cut into small pieces of approximately 1 mm3 with a sharp scalpel (no. 15, Aesculap, Tuttlingen, Germany) and trans-
The objective of root canal fillings is to seal the entire root canal system in an hermetic and biocompatible way to prevent apical or coronal penetration of liquids and microorganisms. The most frequently used definitive root canal filling material is gutta-percha, alone or in combination with a sealer. Temporary endodontic materials, so-called medications, are often applied to achieve antiseptic effects in the root canal system. Due to its alkaline prop36
Vol. 29, No. 1, January 2003
Cytotoxicity of Gutta-Percha Points
37
TABLE 1. Test materials and their composition according to manufacturers Materials (manufacturer)
Type
Size
Composition
Weight (%)
Antaeos (VDW, Mu¨nchen, Germany)
Nonmedicated GP
ISO 40
DeTrey White (Dentspy DeTrey, Konstanz, Germany)
Nonmedicated GP
ISO 40
20 60 17 1 2 ND
Roeko color (Roeko, Langenau, Germany) Roeko Calcium Hydroxide (Roeko)
Nonmedicated GP Medicated GP
ISO 40 ISO 40
Roeko Top color (Roeko) Roeko activ point (Roeko)
Nonmedicated GP Medicated GP
ISO 40 ISO 40
gutta-percha zinc oxide barium sulfate pigments wax synthetic GP zinc oxide pigments ND gutta-percha pigments calcium hydroxide ND gutta-percha zinc oxide barium sulfate pigments chlorhexidine acetate
ND 42 1 57 ND 30 65 ND 1 4
ND ⫽ not declared; GP ⫽ gutta-percha.
ferred to surface-treated, 50-cm3 polystyrene culture bottles (Falcon, Becton & Dickinson, Heidelberg, Germany). After drying the specimens onto the culture bottles for 1 to 2 min at room temperature, 5 ml of culture medium (BM EagleBasal Medium) and 10% calf serum (both materials: Gibco-Life Technologies Ltd.) were added to each bottle. Because tissues from the oral cavity cannot be obtained in a sterile state, antibiotics (penicillin) were added to the culture medium to achieve a low content of germs. The bottles were incubated at 37°C in a gasinjection incubator (Nr, Heraeus, Hanau, Germany) with a 4% CO2 atmosphere and high air humidity (approximately 95%). The first epithelial and fibroblast cells proliferated around the explant margins after 18 –24 h. The formation of a cell monolayer was detectable after 2 to 3 days. Because the test series was to be perpetuated over a longer period of time, a pure fibroblast culture was planned for this investigation and was obtained by means of trypsinization. The gutta-percha points were added to the culture medium in a way that secured a concentration of 6 mg of gutta-percha/ml culture medium (Table 1). Five milliliters of each eluate were pipetted onto a 24-h-old fibroblast monolayer and incubated for 72 h. Subsequently the cell cultures were fixed with 98% superclean ethanol and stained by means of Pappenheim’s panoptic stain. Six cultures within a test series were prepared in this way (n ⫽ 42 Petri dishes) and compared with the control cultures (n ⫽ 6). For identification of physiological and pathological cellular changes, as well as reactions and growth of the cell cultures, the fixed and stained cultures were examined at ⫻100 to ⫻250 magnification with a contrast phase microscope (Leica, Bensheim, Germany). In each bottle, 1000 cells were counted and screened for cellular or nuclear damage. The following parameters were evaluated statistically: 1. Normal cells, including fibroblasts with mitoses or inactive nuclei. 2. Pathologically altered cells, including fibroblasts with nuclear enlargement, binucleation, nuclear anomalies, and vacuoles. 3. Dead cells, including rounded fibroblasts, cells with nuclear disintegration, dead cells, and pyknoses.
4. The mitotic rate was determined as mitoses per 1000 cells, providing information on the proliferative activity of the fibroblasts. This parameter was evaluated under “normal cells.” 5. Cell density. The null hypothesis (H0 ⫽ the different gutta-percha points show no difference in relation to the above-mentioned parameters) was tested by means of the Duncan test (NCSS 6.0.2.1) with adjusted significance level (p ⫽ 0.05).
RESULTS There was a statistically significant difference between the two medicated points; Roeko activ point showed significantly less normal cells (316 cells) than Roeko-Ca(OH)2 (979 cells). Within the group of nonmedicated points and the cell control group (977–993 cells), no significant findings were found. In connection with altered cells, there were no significant differences between the respective material groups (4 –12 cells). There were, however, statistically significant differences between the materials regarding the dead cells. Fibroblasts cultured using media exposed to medicated chlorhexidine points (Roeko activ point) displayed significantly more rounded cells (678 cells) than did medicated calcium hydroxide points (7 cells). Within the group of nonmedicated points and the cell control group (2–11 cells), no significant findings were found. Roeko activ point exhibited a significantly lower cell density (26 cells/mm2) compared with the medicated Roeko-Ca(OH)2 (109 cells/mm2). Best values were obtained within the cell control group (119 cells/mm2) with significantly differences to the group Roeko activ point. Highest mitotic rates were obtained in the cell control group (9 mitoses/1000 cells), which were significantly different from the medicated Roeko activ point (2 mitoses/1000 cells). Antaeos guttapercha points exhibited a significantly higher mitotic rate (9 mitoses/1000 cells) in comparison with the medicated Roeko activ point.
38
Szep et al.
Journal of Endodontics
FIG 1. The control cultures show a regular and dense monolayer with long, slender cells and numerous mitoses (magnification ⫻100).
FIG 2. The Roeko color cultures appear similar to the DeTrey cultures. There are degenerative changes such as cells with rounded shape (magnification ⫻100).
Cell Control
nuclear enlargement, rounding up, and pyknoses and only a few nuclear anomalies and vacuolizations.
The control cultures show the typical stellate fibroblast morphology and orientation of the cells in clusters and bundles forming a sound monolayer. There were only a few dead cells detached from the dish and a small number of changed cells in the culture, including nuclear enlargement, binucleation, and to a small degree, nuclear anomalies, pyknoses, and rounding up (Fig. 1).
Roeko Top Color The mitotic rate was comparable with the DeTrey points but much higher than in the Roeko activ point culture. The cell changes comprised binucleation, nuclear enlargement, rounding up, vacuolization, and pyknoses.
Antaeos The fibroblast monolayers showed more changed cells, i.e. nuclear enlargement, binucleation, rounding up, pyknoses, and somewhat more vacuolizations and nuclear anomalies than the control cultures.
Roeko Activ Point The gutta-percha points caused pronounced toxic reactions of the fibroblasts, which appeared as retractions, rounding up, and pyknoses. To a lesser extent there was also evidence of binucleation, nuclear enlargement, and nuclear anomalies (Fig. 3).
DeTrey White The monolayers were comparable with the control cultures. There were isolated pyknotic, polynucleated, vacuolated, and rounded up cells. The proportion of fibroblasts with nuclear enlargement appears slightly increased in comparison with the control cultures. Roeko Color The mitotic rate was smaller than in the controls but comparable with the gutta-percha points Antaeos, Roeko color, Roeko Calcium Hydroxide, and Roeko Top color. The most frequent cytopathological findings were nuclear enlargement, binucleation, vacuolization, nuclear anomalies, and pyknoses (Fig. 2). Roeko Calcium Hydroxide The mitotic rates in this type of gutta-percha points were comparable with those of the control culture and Antaeos and Roeko color points. Cytopathological changes were mainly binucleation,
DISCUSSION Cell cultures are the test design most frequently used to determine the cytotoxicity of, and local reactions to, dental materials. The two basic types of cell cultures applied are primary cultures and established cell lines. Cell types used in the literature for testing of endodontic materials are fibroblasts, epithelial cells, and lymphoblasts for primary cultures or HeLa cells and sarcoma fibroblasts (L939 cells) from mice for established cell lines (2, 7, 9). Cells derived from a primary culture are distinguished by the fact that they have been cultured for the first time. Therefore, they are characterized, much like their original tissue, by a diploid set of chromosomes, a largely unchanged metabolic status, and a high degree of differentiation. In contrast, established cell lines are morphologically and physiologically more homogenous, but because they have been passaged many times, they have lost the karyotype of their original tissue. The heterogeneity of primary cultures, which reflects different stages of physiological aging, is much better suited to mimic the in vivo situation than the homogeneity of established cell lines.
Vol. 29, No. 1, January 2003
FIG 3. All fibroblasts in contact with the eluate of points containing chlorhexidine are in the process of dying. Many have already rounded up and are floating in the supernatant (magnification ⫻100).
Another important factor in this context is the selection of the cell type best suited for simulating the in vivo situation. Human uterine carcinoma cells or sarcoma fibroblasts from mice do not optimally simulate the cells found in the human oral cavity. Therefore, a primary culture of human gingival fibroblasts as target cells was used in this investigation. The distinction made between medicated and nonmedicated gutta-percha points should be kept in mind when interpreting and discussing the results of this study. Whereas the latter points remain in the root canal over several years, medicated points are only inserted for a short period of time as carriers of a medicament. Bactericidal properties are expected from medicated rather than from nonmedicated points. Available data regarding the toxicity of nonmedicated guttapercha points allow different interpretations. Whereas Kawahara et al. (10) and Wolfson and Seltzer (4) state that gutta-percha can be judged nontoxic, Spangberg et al.(2), Das (8), Moorer et al. (3), and Pascon and Spangberg (9) arrive at the opposite conclusion. This study supports the latter conclusion, because in our tests all gutta-percha points showed basically a toxic trend. Comparing the nonmedicated points with each other and with the control cultures showed an increase in the number of dead cells, which was not statistically significant. The same was true for the increase in the number of normal cells as well as for the parameters mitotic rate and cell density, all of which were lower in the nonmedicated gutta-percha cell cultures in comparison with the control cultures but not significantly so. The lowest mitotic rate in the group of nonmedicated gutta-percha points was found in the DeTrey cultures (5 mitoses/1000 cells), although as already mentioned, there was no statistical significance detectable. As described by other authors, the differing results for the nonmedicated materials can also be explained by the antibacterial properties of the zinc oxide contained in the points (3, 11). Comparison of the zinc oxide content between the products is likewise impaired by the fact that many manufacturers do not declare the respective quantities. Marciano and Michailesco (12) and Pascon and Spangberg (9) have also referred to this problem. The medicated gutta-percha points containing chlorhexidine or calcium hydroxide as antibacterial additives behaved differently
Cytotoxicity of Gutta-Percha Points
39
from the nonmedicated points. It has been demonstrated in cytotoxicity studies that chlorhexidine strongly reduces cellular proliferation (6). Mariotti and Rumpf (6) also found that chlorhexidine prevents collagen production of human gingival fibroblasts. They reported that chlorhexidine preparations may impair wound healing at a concentration of only 0.1% (6, 7) and that more granulation tissue than collagen fibers is formed in the wound. Economides et al. (16) indicate that 0.05% chlorhexidine is 10 times more toxic than antibacterial. According to these authors, chlorhexidine of the above mentioned concentration does not meet the requirement of being maximally antibacterial but minimally toxic. According to Helgeland et al. (5), characteristic effects of chlorhexidine are denaturation of enzyme proteins and cytoplasmic coagulation (5, 13). Hennessey (13) and Koskinen et al. (14) further point to their observation of a precipitation and coagulation of cytoplasmic constituents, which was confirmed in this study. In this study, the gutta-percha points containing chlorhexidine significantly influenced all tested parameters, except the number of changed cells, which was not significantly increased. There are at present no comparable studies available that prove the cytotoxicity of points containing chlorhexidine. Studies testing the toxicology of calcium hydroxide indicate that this agent causes destruction of cell membranes as well as cellular degeneration and disintegration (8, 10). Bacteriological investigations prove that lipopolysaccharides from the bacterial cell membrane are influenced by calcium hydroxide and that the latter has stronger growth inhibitory properties than zinc oxide or chlorhexidine (11, 15). These observations were corroborated in this study, but the mitotic rates and the amount of normal cells were statistically not significantly different in the points containing calcium hydroxide compared with the control cultures and the nonmedicated points. Comparing calcium hydroxide-containing points with the points containing chlorhexidine, the statement of Podbielski et al. (11), that calcium hydroxide has stronger growth inhibitory properties than chlorhexidine, could not be corroborated in this investigation. Here, the growth inhibitory effect of the chlorhexidine points was significantly higher than that of the calcium hydroxide points. The same was true when the points containing chlorhexidine and nonmedicated points were compared. Economides et al. (16) and Larsen and Ho¨ rsted-Bindslev (17) tested the impact of points containing calcium hydroxide, and not of the pure substance, on pH changes. They found that these points deliver less Ca2⫹ ions than other calcium hydroxide formulations and therefore do not seem suitable for in vivo use. In how far the pH measurements carried out by Larsen and Ho¨ rsted-Bindslev (17) are able to mimic the in vivo situation in a suitable way should be further clarified. Drs. Szep, Schriever, Schultze, Heidemann, Mrs. Ronge, and Mrs. Grumann are affiliated with the Department of Conservative Dentistry, Johann Wolfgang Goethe University. Prof. Heidemann is director and chairman, Department of Conservative Dentistry, Johann Wolfgang Goethe University, Frankfurt am Main, Germany. Address requests for reprints to Dr. Susanne Szep, Department of Conservative Dentistry, School of Dentistry, Johann Wolfgang Goethe University, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany.
References 1. Spangberg L. Biological effects of root canal filling materials. 4. Effect in vitro of solubilised root canal filling materials on HeLa cells. Odontol Revy 1969;20:289 –99. 2. Spangberg L, Engstro¨ m B, Langeland K. Biologic effects of dental
40
Szep et al.
materials. III. Toxicity and antimicrobial effect of endodontic antiseptics in vitro. Oral Surg Oral Med Oral Pathol 1973;36:856 –71. 3. Moorer WR, Genet JM. Evidence for antibacterial activity of endodontic gutta-percha cones. Oral Surg Oral Med Oral Pathol 1982;53:503–7. 4. Wolfson EM, Seltzer S. Reaction of rat connective tissue to some gutta-percha formulations. J Endodon 1975;1:395– 402. 5. Helgeland K, Heyden G, Ro¨ lla G. Effect of chlorhexidine on animal cells in vitro. Scand J Dent Res 1971;79:209 –15. 6. Mariotti AJ, Rumpf DAH. Chlorhexidine-induced changes to human gingival fibroblast collagen and non-collagen protein production. J Periodontol 1999;70:1443– 8. 7. Goldschmidt P, Cogen R, Taubman S. Cytopathologic effects of chlorhexidine on human cells. J Periodontol 1977;11:212–5. 8. Das S. Effect of certain dental materials on human pulp tissue culture. Oral Surg Oral Med Oral Pathol 1981;52:76 – 84. 9. Pascon EA, Spangberg LSW. In vitro cytotoxicity of root canal filling materials. 1. Gutta-percha. J Endodon 1990;16:429 –33. 10. Kawahara H, Yamagami A, Nakamura M. Biological testing of dental materials by means of tissue culture. Int Dent J 1968;18:443– 67. 11. Podbielski A, Boeckh C, Haller B. Growth inhibitory activity of gutta-
Journal of Endodontics percha points containing root canal medications on common endodontic bacterial pathogens as determined by an optimized quantitative in vitro assay. J Endodon 2000;26:398 – 403. 12. Marciano J, Michailesco PM. Dental gutta-percha: chemical composition, X-ray identification, enthalpic studies, and clinical implications. J Endodon 1989;15:149 –53. 13. Hennessey TD. Antibacterial properties of Hibitane. J Clin Periodontol 1977;4:36 – 48. 14. Koskinen KP, Rahkamo A, Tuompo H. Cytotoxicity of some solutions used for root canal treatment assessed with human fibroblasts and lymphoblasts. Scand J Dent Res 1981;89:71– 8. 15. Safavi KE, Nichols FC. Alteration of biological properties of bacterial lipopolysaccharide by calcium hydroxide treatment. J Endodon 1994;20:127–9. 16. Economides N, Koulaouzidou EA, Beltes P, Kortsaris AH. In vitro release of hydroxyl ions from calcium hydroxide gutta-percha points. J Endodon 1999;25:481–2. 17. Larsen MJ, Ho¨ rsted-Bindslev P. A laboratory study evaluating the release of hydroxyl ions from various calcium hydroxide products in narrow root canal-like tubes. Int Endod J 2000;33:238 – 42.
Printed in U.S.A. VOL. 29, NO. 1, JANUARY 2003
In Vitro Cytotoxicity of Medicated and Nonmedicated Gutta-percha Points in Cultures of Gingival Fibroblasts Susanne Szep, Dr med dent, Ljiljana Grumann, Karin Ronge, Anette Schriever, Dr med dent, Myriam Schultze, Dr med dent, and Detlef Heidemann, Prof Dr med dent
erties, calcium hydroxide, which can be administered in different forms, is the most frequently used medication. The dentinal tubules allow diffusion of materials placed into the root canal system. Furthermore, contact of these materials with periodontal tissues may occur via the apical foramen or accessory root canals. In this context the toxicity of a substance, i.e. its biocompatibility in contact with surrounding tissues, plays an important role. Primary toxicity can be investigated by means of gingival fibroblasts as so-called “target cells,” whereas the reaction of in vivo tissue depends on further parameters such as salivary flow rate and resident bacterial flora, mechanical, and chemical stimuli during food intake, hormonal status, and biorhythms. Despite these limitations, differentiated primary cell cultures and established cell lines are used to arrive at conclusions relevant to the in vivo situation. There are numerous studies available that deal with the toxicity of gutta-percha points and endodontic medications in the root canal (1–5). However, the cytotoxic potential of so-called “medicated gutta-percha points” recently introduced in the market, which contain chlorhexidine or calcium hydroxide as antiseptic agents, has so far not been investigated. There are studies available relating to the active substances contained in these points, but not when in combination with a gutta-percha carrier (1, 6 – 8). It was the aim of this study to compare the cytotoxicity of two medicated and four nonmedicated gutta-percha points using primary human gingival fibroblasts. Physiological and pathological changes of the cells and reactions and growth of the cell cultures, respectively, were evaluated microscopically and statistically.
This investigation was designed to test the cellular toxicity of two medicated (Roeko activ point and Roeko Calcium Hydroxide) and four nonmedicated brands of gutta-percha (GP) points (Antaeos, DeTrey White, Roeko color, and Roeko Top color). The test points were transferred into a culture medium including the GP-point material with a concentration of 6 mg/ml, and eluates were obtained after 72 h. Five milliliters of each eluate were pipetted onto fibroblast cultures, incubated, and subsequently stained. Mitotic rates, cell densities, and the distribution of normal cells, pathologically altered and dead cells were determined and correlated with control cell cultures. Roeko activ point (containing chlorhexidine) resulted in the highest number of dead cells. The difference was statistically significant in comparison with all other materials. Concerning all parameters mentioned, the cytotoxicity of the points containing calcium hydroxide (Roeko Calcium Hydroxide) was not significantly different from all other points tested, with the exception of those containing chlorhexidine. All tested gutta-percha materials caused cytotoxic reactions in varying extents. Taking into consideration the limitations of an in vitro experiment, points containing calcium hydroxide and nonmedicated points seem to be the most recommendable products for clinical use.
MATERIALS AND METHODS Inflammation-free human gingival tissue specimen were gained from surgical and periodontal operation sites. The explants were stored in a refrigerator at 4°C overnight. They were subsequently washed twice in Hanks’ balanced salt solution, which had been complemented with 5 ml of bicarbonate and an antibiotic additive and cooled to 4°C to remove blood and granulation tissue (all materials: Gibco-Life Technologies Ltd., Paisley, Scotland). The explants were cut into small pieces of approximately 1 mm3 with a sharp scalpel (no. 15, Aesculap, Tuttlingen, Germany) and trans-
The objective of root canal fillings is to seal the entire root canal system in an hermetic and biocompatible way to prevent apical or coronal penetration of liquids and microorganisms. The most frequently used definitive root canal filling material is gutta-percha, alone or in combination with a sealer. Temporary endodontic materials, so-called medications, are often applied to achieve antiseptic effects in the root canal system. Due to its alkaline prop36
Vol. 29, No. 1, January 2003
Cytotoxicity of Gutta-Percha Points
37
TABLE 1. Test materials and their composition according to manufacturers Materials (manufacturer)
Type
Size
Composition
Weight (%)
Antaeos (VDW, Mu¨nchen, Germany)
Nonmedicated GP
ISO 40
DeTrey White (Dentspy DeTrey, Konstanz, Germany)
Nonmedicated GP
ISO 40
20 60 17 1 2 ND
Roeko color (Roeko, Langenau, Germany) Roeko Calcium Hydroxide (Roeko)
Nonmedicated GP Medicated GP
ISO 40 ISO 40
Roeko Top color (Roeko) Roeko activ point (Roeko)
Nonmedicated GP Medicated GP
ISO 40 ISO 40
gutta-percha zinc oxide barium sulfate pigments wax synthetic GP zinc oxide pigments ND gutta-percha pigments calcium hydroxide ND gutta-percha zinc oxide barium sulfate pigments chlorhexidine acetate
ND 42 1 57 ND 30 65 ND 1 4
ND ⫽ not declared; GP ⫽ gutta-percha.
ferred to surface-treated, 50-cm3 polystyrene culture bottles (Falcon, Becton & Dickinson, Heidelberg, Germany). After drying the specimens onto the culture bottles for 1 to 2 min at room temperature, 5 ml of culture medium (BM EagleBasal Medium) and 10% calf serum (both materials: Gibco-Life Technologies Ltd.) were added to each bottle. Because tissues from the oral cavity cannot be obtained in a sterile state, antibiotics (penicillin) were added to the culture medium to achieve a low content of germs. The bottles were incubated at 37°C in a gasinjection incubator (Nr, Heraeus, Hanau, Germany) with a 4% CO2 atmosphere and high air humidity (approximately 95%). The first epithelial and fibroblast cells proliferated around the explant margins after 18 –24 h. The formation of a cell monolayer was detectable after 2 to 3 days. Because the test series was to be perpetuated over a longer period of time, a pure fibroblast culture was planned for this investigation and was obtained by means of trypsinization. The gutta-percha points were added to the culture medium in a way that secured a concentration of 6 mg of gutta-percha/ml culture medium (Table 1). Five milliliters of each eluate were pipetted onto a 24-h-old fibroblast monolayer and incubated for 72 h. Subsequently the cell cultures were fixed with 98% superclean ethanol and stained by means of Pappenheim’s panoptic stain. Six cultures within a test series were prepared in this way (n ⫽ 42 Petri dishes) and compared with the control cultures (n ⫽ 6). For identification of physiological and pathological cellular changes, as well as reactions and growth of the cell cultures, the fixed and stained cultures were examined at ⫻100 to ⫻250 magnification with a contrast phase microscope (Leica, Bensheim, Germany). In each bottle, 1000 cells were counted and screened for cellular or nuclear damage. The following parameters were evaluated statistically: 1. Normal cells, including fibroblasts with mitoses or inactive nuclei. 2. Pathologically altered cells, including fibroblasts with nuclear enlargement, binucleation, nuclear anomalies, and vacuoles. 3. Dead cells, including rounded fibroblasts, cells with nuclear disintegration, dead cells, and pyknoses.
4. The mitotic rate was determined as mitoses per 1000 cells, providing information on the proliferative activity of the fibroblasts. This parameter was evaluated under “normal cells.” 5. Cell density. The null hypothesis (H0 ⫽ the different gutta-percha points show no difference in relation to the above-mentioned parameters) was tested by means of the Duncan test (NCSS 6.0.2.1) with adjusted significance level (p ⫽ 0.05).
RESULTS There was a statistically significant difference between the two medicated points; Roeko activ point showed significantly less normal cells (316 cells) than Roeko-Ca(OH)2 (979 cells). Within the group of nonmedicated points and the cell control group (977–993 cells), no significant findings were found. In connection with altered cells, there were no significant differences between the respective material groups (4 –12 cells). There were, however, statistically significant differences between the materials regarding the dead cells. Fibroblasts cultured using media exposed to medicated chlorhexidine points (Roeko activ point) displayed significantly more rounded cells (678 cells) than did medicated calcium hydroxide points (7 cells). Within the group of nonmedicated points and the cell control group (2–11 cells), no significant findings were found. Roeko activ point exhibited a significantly lower cell density (26 cells/mm2) compared with the medicated Roeko-Ca(OH)2 (109 cells/mm2). Best values were obtained within the cell control group (119 cells/mm2) with significantly differences to the group Roeko activ point. Highest mitotic rates were obtained in the cell control group (9 mitoses/1000 cells), which were significantly different from the medicated Roeko activ point (2 mitoses/1000 cells). Antaeos guttapercha points exhibited a significantly higher mitotic rate (9 mitoses/1000 cells) in comparison with the medicated Roeko activ point.
38
Szep et al.
Journal of Endodontics
FIG 1. The control cultures show a regular and dense monolayer with long, slender cells and numerous mitoses (magnification ⫻100).
FIG 2. The Roeko color cultures appear similar to the DeTrey cultures. There are degenerative changes such as cells with rounded shape (magnification ⫻100).
Cell Control
nuclear enlargement, rounding up, and pyknoses and only a few nuclear anomalies and vacuolizations.
The control cultures show the typical stellate fibroblast morphology and orientation of the cells in clusters and bundles forming a sound monolayer. There were only a few dead cells detached from the dish and a small number of changed cells in the culture, including nuclear enlargement, binucleation, and to a small degree, nuclear anomalies, pyknoses, and rounding up (Fig. 1).
Roeko Top Color The mitotic rate was comparable with the DeTrey points but much higher than in the Roeko activ point culture. The cell changes comprised binucleation, nuclear enlargement, rounding up, vacuolization, and pyknoses.
Antaeos The fibroblast monolayers showed more changed cells, i.e. nuclear enlargement, binucleation, rounding up, pyknoses, and somewhat more vacuolizations and nuclear anomalies than the control cultures.
Roeko Activ Point The gutta-percha points caused pronounced toxic reactions of the fibroblasts, which appeared as retractions, rounding up, and pyknoses. To a lesser extent there was also evidence of binucleation, nuclear enlargement, and nuclear anomalies (Fig. 3).
DeTrey White The monolayers were comparable with the control cultures. There were isolated pyknotic, polynucleated, vacuolated, and rounded up cells. The proportion of fibroblasts with nuclear enlargement appears slightly increased in comparison with the control cultures. Roeko Color The mitotic rate was smaller than in the controls but comparable with the gutta-percha points Antaeos, Roeko color, Roeko Calcium Hydroxide, and Roeko Top color. The most frequent cytopathological findings were nuclear enlargement, binucleation, vacuolization, nuclear anomalies, and pyknoses (Fig. 2). Roeko Calcium Hydroxide The mitotic rates in this type of gutta-percha points were comparable with those of the control culture and Antaeos and Roeko color points. Cytopathological changes were mainly binucleation,
DISCUSSION Cell cultures are the test design most frequently used to determine the cytotoxicity of, and local reactions to, dental materials. The two basic types of cell cultures applied are primary cultures and established cell lines. Cell types used in the literature for testing of endodontic materials are fibroblasts, epithelial cells, and lymphoblasts for primary cultures or HeLa cells and sarcoma fibroblasts (L939 cells) from mice for established cell lines (2, 7, 9). Cells derived from a primary culture are distinguished by the fact that they have been cultured for the first time. Therefore, they are characterized, much like their original tissue, by a diploid set of chromosomes, a largely unchanged metabolic status, and a high degree of differentiation. In contrast, established cell lines are morphologically and physiologically more homogenous, but because they have been passaged many times, they have lost the karyotype of their original tissue. The heterogeneity of primary cultures, which reflects different stages of physiological aging, is much better suited to mimic the in vivo situation than the homogeneity of established cell lines.
Vol. 29, No. 1, January 2003
FIG 3. All fibroblasts in contact with the eluate of points containing chlorhexidine are in the process of dying. Many have already rounded up and are floating in the supernatant (magnification ⫻100).
Another important factor in this context is the selection of the cell type best suited for simulating the in vivo situation. Human uterine carcinoma cells or sarcoma fibroblasts from mice do not optimally simulate the cells found in the human oral cavity. Therefore, a primary culture of human gingival fibroblasts as target cells was used in this investigation. The distinction made between medicated and nonmedicated gutta-percha points should be kept in mind when interpreting and discussing the results of this study. Whereas the latter points remain in the root canal over several years, medicated points are only inserted for a short period of time as carriers of a medicament. Bactericidal properties are expected from medicated rather than from nonmedicated points. Available data regarding the toxicity of nonmedicated guttapercha points allow different interpretations. Whereas Kawahara et al. (10) and Wolfson and Seltzer (4) state that gutta-percha can be judged nontoxic, Spangberg et al.(2), Das (8), Moorer et al. (3), and Pascon and Spangberg (9) arrive at the opposite conclusion. This study supports the latter conclusion, because in our tests all gutta-percha points showed basically a toxic trend. Comparing the nonmedicated points with each other and with the control cultures showed an increase in the number of dead cells, which was not statistically significant. The same was true for the increase in the number of normal cells as well as for the parameters mitotic rate and cell density, all of which were lower in the nonmedicated gutta-percha cell cultures in comparison with the control cultures but not significantly so. The lowest mitotic rate in the group of nonmedicated gutta-percha points was found in the DeTrey cultures (5 mitoses/1000 cells), although as already mentioned, there was no statistical significance detectable. As described by other authors, the differing results for the nonmedicated materials can also be explained by the antibacterial properties of the zinc oxide contained in the points (3, 11). Comparison of the zinc oxide content between the products is likewise impaired by the fact that many manufacturers do not declare the respective quantities. Marciano and Michailesco (12) and Pascon and Spangberg (9) have also referred to this problem. The medicated gutta-percha points containing chlorhexidine or calcium hydroxide as antibacterial additives behaved differently
Cytotoxicity of Gutta-Percha Points
39
from the nonmedicated points. It has been demonstrated in cytotoxicity studies that chlorhexidine strongly reduces cellular proliferation (6). Mariotti and Rumpf (6) also found that chlorhexidine prevents collagen production of human gingival fibroblasts. They reported that chlorhexidine preparations may impair wound healing at a concentration of only 0.1% (6, 7) and that more granulation tissue than collagen fibers is formed in the wound. Economides et al. (16) indicate that 0.05% chlorhexidine is 10 times more toxic than antibacterial. According to these authors, chlorhexidine of the above mentioned concentration does not meet the requirement of being maximally antibacterial but minimally toxic. According to Helgeland et al. (5), characteristic effects of chlorhexidine are denaturation of enzyme proteins and cytoplasmic coagulation (5, 13). Hennessey (13) and Koskinen et al. (14) further point to their observation of a precipitation and coagulation of cytoplasmic constituents, which was confirmed in this study. In this study, the gutta-percha points containing chlorhexidine significantly influenced all tested parameters, except the number of changed cells, which was not significantly increased. There are at present no comparable studies available that prove the cytotoxicity of points containing chlorhexidine. Studies testing the toxicology of calcium hydroxide indicate that this agent causes destruction of cell membranes as well as cellular degeneration and disintegration (8, 10). Bacteriological investigations prove that lipopolysaccharides from the bacterial cell membrane are influenced by calcium hydroxide and that the latter has stronger growth inhibitory properties than zinc oxide or chlorhexidine (11, 15). These observations were corroborated in this study, but the mitotic rates and the amount of normal cells were statistically not significantly different in the points containing calcium hydroxide compared with the control cultures and the nonmedicated points. Comparing calcium hydroxide-containing points with the points containing chlorhexidine, the statement of Podbielski et al. (11), that calcium hydroxide has stronger growth inhibitory properties than chlorhexidine, could not be corroborated in this investigation. Here, the growth inhibitory effect of the chlorhexidine points was significantly higher than that of the calcium hydroxide points. The same was true when the points containing chlorhexidine and nonmedicated points were compared. Economides et al. (16) and Larsen and Ho¨ rsted-Bindslev (17) tested the impact of points containing calcium hydroxide, and not of the pure substance, on pH changes. They found that these points deliver less Ca2⫹ ions than other calcium hydroxide formulations and therefore do not seem suitable for in vivo use. In how far the pH measurements carried out by Larsen and Ho¨ rsted-Bindslev (17) are able to mimic the in vivo situation in a suitable way should be further clarified. Drs. Szep, Schriever, Schultze, Heidemann, Mrs. Ronge, and Mrs. Grumann are affiliated with the Department of Conservative Dentistry, Johann Wolfgang Goethe University. Prof. Heidemann is director and chairman, Department of Conservative Dentistry, Johann Wolfgang Goethe University, Frankfurt am Main, Germany. Address requests for reprints to Dr. Susanne Szep, Department of Conservative Dentistry, School of Dentistry, Johann Wolfgang Goethe University, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany.
References 1. Spangberg L. Biological effects of root canal filling materials. 4. Effect in vitro of solubilised root canal filling materials on HeLa cells. Odontol Revy 1969;20:289 –99. 2. Spangberg L, Engstro¨ m B, Langeland K. Biologic effects of dental
40
Szep et al.
materials. III. Toxicity and antimicrobial effect of endodontic antiseptics in vitro. Oral Surg Oral Med Oral Pathol 1973;36:856 –71. 3. Moorer WR, Genet JM. Evidence for antibacterial activity of endodontic gutta-percha cones. Oral Surg Oral Med Oral Pathol 1982;53:503–7. 4. Wolfson EM, Seltzer S. Reaction of rat connective tissue to some gutta-percha formulations. J Endodon 1975;1:395– 402. 5. Helgeland K, Heyden G, Ro¨ lla G. Effect of chlorhexidine on animal cells in vitro. Scand J Dent Res 1971;79:209 –15. 6. Mariotti AJ, Rumpf DAH. Chlorhexidine-induced changes to human gingival fibroblast collagen and non-collagen protein production. J Periodontol 1999;70:1443– 8. 7. Goldschmidt P, Cogen R, Taubman S. Cytopathologic effects of chlorhexidine on human cells. J Periodontol 1977;11:212–5. 8. Das S. Effect of certain dental materials on human pulp tissue culture. Oral Surg Oral Med Oral Pathol 1981;52:76 – 84. 9. Pascon EA, Spangberg LSW. In vitro cytotoxicity of root canal filling materials. 1. Gutta-percha. J Endodon 1990;16:429 –33. 10. Kawahara H, Yamagami A, Nakamura M. Biological testing of dental materials by means of tissue culture. Int Dent J 1968;18:443– 67. 11. Podbielski A, Boeckh C, Haller B. Growth inhibitory activity of gutta-
Journal of Endodontics percha points containing root canal medications on common endodontic bacterial pathogens as determined by an optimized quantitative in vitro assay. J Endodon 2000;26:398 – 403. 12. Marciano J, Michailesco PM. Dental gutta-percha: chemical composition, X-ray identification, enthalpic studies, and clinical implications. J Endodon 1989;15:149 –53. 13. Hennessey TD. Antibacterial properties of Hibitane. J Clin Periodontol 1977;4:36 – 48. 14. Koskinen KP, Rahkamo A, Tuompo H. Cytotoxicity of some solutions used for root canal treatment assessed with human fibroblasts and lymphoblasts. Scand J Dent Res 1981;89:71– 8. 15. Safavi KE, Nichols FC. Alteration of biological properties of bacterial lipopolysaccharide by calcium hydroxide treatment. J Endodon 1994;20:127–9. 16. Economides N, Koulaouzidou EA, Beltes P, Kortsaris AH. In vitro release of hydroxyl ions from calcium hydroxide gutta-percha points. J Endodon 1999;25:481–2. 17. Larsen MJ, Ho¨ rsted-Bindslev P. A laboratory study evaluating the release of hydroxyl ions from various calcium hydroxide products in narrow root canal-like tubes. Int Endod J 2000;33:238 – 42.