- Email: [email protected]
Effect of IRM root end fillings on healing after replantation
0099-2399/94/2008-0381/$03.00/0 JOURNAL OF ENDODONTICS Copyright © 1994 by The American Association of Endodontists
Printed in U.S.A.
VOL. 20, NO. 8, AUGUST1994
Effect of IRM Root End Fillings on Healing after Replantation T. R. Pitt Ford, BDS, PhD, J. O. Andreasen, S. O. Dorn, DDS, and S. P. Kariyawasam
oxide-eugenol cements as root end fillings. Although some clinicians have used these cements for a long time (12), the scientific rationale for doing so has not been clearly stated. Indeed, the perceived view by many has been that these cements are too soluble for long-term use as root end fillings. In an earlier study, Andreasen and Pitt Ford (13) have indicated radiographically that the periradicular radiolucencies around roots with root end root fillings of IRM were much smaller than those around roots filled with amalgam in replanted teeth in monkeys. The aim of this study was to assess whether the apparent superiority of IRM over amalgam at the radiological level was true at the histological level.
The effect of IRM as a root end filling placed in teeth prior to replantation was examined in 21 molar teeth in monkeys. After extraction, root ends were resected, the canals contaminated with oral bacteria, root end cavities prepared, and fillings of IRM or amalgam placed prior to replantation. After 8 wk the jaws were removed and prepared for histological examination. Bacteria were demonstrated in only 9 of 15 teeth filled with IRM; 18 of the roots (60%) were associated with inflammation, which was only moderate or severe around 5 (17%), and extended >0.1 mm around only 2 roots. In contrast, of the 6 teeth filled with amalgam, all contained bacteria in the root canals and 11 roots were associated with moderate or severe inflammation, which around 8 roots extended >0.5 ram. The difference in severity of inflammation for the two materials was statistically significant (p < 0.001). The tissue response to root end fillings of IRM in replanted teeth was less severe and less extensive than that to amalgam.
M A T E R I A L S AND M E T H O D S Monkeys (Cercopithecus aethiops) that had complete root formation of the mandibular first molars were chosen. After general anesthesia with phencyclidine (2.5 mg/kg body wt, Sernylan; Parke Davis & Co, Copenhagen, Denmark) supplemented with pentobarbitone sodium (30 mg/kg body wt, Mebumal; Dak, Copenhagen, Denmark), the mandibular first molars were extracted using gentle luxation movements. After extraction, the crowns of the teeth were held by forceps while 2 mm of the apices were resected fiat with a diamond wheel under a constant flow of saline. A root end cavity was drilled to a depth of 2 m m with a round bur (diameter, 0.8 mm) under constant saline cooling. The root end cavity was then washed with saline and dried with compressed air. A file contaminated with saliva was placed in the root canal to introduce infection before the root end was filled with one of the following filling materials mixed according to the manufacturers' instructions: amalgam (Standalloy, Degussa, Frankfurt, Germany) or (Luxalloy, Degussa) and zinc oxide-eugenol-based compound: IRM (Caulk, Dentsply, Milford, DE). These filling materials were randomly allocated to the teeth. After completion of the root filling procedure, which took 15 to 20 rain, the molars were replanted and left unsplinted for 8 wk. The monkeys were then killed and the jaw section was removed. The tissue blocks were placed in 10% neutral buffered formalin before being demineralized in a 10% solution of EDTA at pH 6.9. After double infiltration in celloidin-paraffin, the tissue blocks were sectioned perpendicular to the long axis mesiodistally at a thickness of 7 #m. The teeth were
Despite amalgam being widely used as a root end root filling material in endodontic surgery, a number of clinical studies have reported a low rate of success when strict healing criteria are used (1-4). When the success rate of root end fillings with amalgam has been compared with resection of roots filled with gutta-percha and sealer, the success rate with amalgam was not as good ( 1, 3). Furthermore, when amalgam was used as a root end filling after root resection, the success rate was not improved by the presence of the filling (5, 6). Healing was not as good as it should have been and this indicates that amalgam may not have been providing a seal to bacteria. Concern has also been expressed about the stability of amalgam with the migration of metallic particles into the tissues (7, 8). A large number of alternative materials have been investigated both in vitro (9) and in vivo (10, 11). A recent clinical study has indicated that IRM is significantly more successful than amalgam (2). This has led to a renewed interest in zinc
381
382
Pitt Ford et al.
Journal of Endodontics
TABLE 1. Number of roots with inflammation, its extent and severity, the presence of epithelium, and the mean extent of root resorption for each material Material
No. of Roots
No. with Inflammation
Amalgam
12
11
IRM
30
18
No. at Each Severity
No. at Each Extent (mm)
None Few Moderate Severe
1 0 5 6
None _<0.1 _<0.2 -<0.5 >0.5
None Few Moderate Severe
12 13 4 1
None _<0.1 <_0.2 _<0.5 >0.5
1
Epithelium at Root End
Mean Extent of Root Resorption Exposing Filling
0
N/A*
2
0.33
0 3 0 8 12 16 1 1 0
* NA, not available.
sectioned in step-serial sections at 100-urn intervals. At each sectioning level, slides were stained with hematoxylin and eosin, Masson's trichrome, and Brown and Brenn stain. Sections were assessed for inflammation at the root end adjacent to the root end filling material. The severity of the inflammation was recorded as: none, no inflammatory cells; few cells, few inflammatory cells; moderate, inflammatory cells did not obscure the normal tissues; and severe, inflammatory cells replaced normal tissues. The extent of inflammation was recorded as -<0.1 mm, _<0.2 mm, _<0.5 mm, or >0.5 mm. The presence of inflammation in the adjacent bone marrow was registered. The presence or absence of root resorption exposing the root end filling material was noted, and where present the greatest extent toward the crown was recorded as part of the length of the filling (to nearest one-fourth). The presence or absence of ankylosis of the tooth in any part, noticeable dissolution of the root end filling material, giant cells, a fibrous capsule, cementum deposition on the root end filling material, or epithelium covering the root end material was noted. The presence or absence of bacteria in any part of the root canal was also assessed.
TABLE 2. Number of teeth containing bacteria, exhibiting ankylosis, and demonstrating active root resorption for each material Material
No. of Teeth
No. with Bacteria in Root Canals
No. with Ankylosis
No. with Active Root Resorption
Amalgam IRM
6 15
6 9
0 8
5 13
RESULTS A total of 12 roots in 6 teeth filled with amalgam were available for histological examination. As there was no obvious difference between the effects of the two amalgam alloys, the results were combined. Thirty roots in 15 teeth were filled with IRM. The results of individual roots are given in Table 1. When the number of roots with either none or few inflammatory cells were compared with the number with moderate or severe inflammation for the two materials, the difference was statistically significant (p < 0.001, x 2 test, Yates correction). Because it was necessary to remove the amalgam before sectioning, it was not possible to assess the extent of root resorption exposing the filling. The results applicable to individual teeth are given in Table 2. IRM Eighteen roots filled with IRM were associated with inflammation at the root end (60%); only five roots had moderate
FIG 1. Root end of a tooth filled with IRM (F) showing material clinging to canal wall (D) (hematoxylin and eosin stain; original magnification x310).
or severe inflammation and two had inflammation extending >0.1 mm. No tooth had inflammation in the bone marrow space. Bacteria were found in the root canals of 60% of teeth filled with IRM. They were not usually present in large numbers and were most easily observed in the pulp space rather than near the root end filling. It was a general observation that the IRM clung to canal walls following sectioning (Fig. 1). The tissue response next to the exposed root filling generally ranged from no inflammatory response to a few inflammatory cells which were most frequently lymphocytes, although a few epithelioid-type giant cells were noticed against the filling. In cases where resorption
Vol. 20, No. 8, August 1994
Healing after IRM Root End Filling
383
had exposed lateral parts of the filling, an inflammatory response of greater intensity was observed confined to that area alone (Fig. 2). Noticeable dissolution of the cement by cells was not observed.
Amalgam Eleven roots filled with amalgam were associated with periapical inflammation (92%); all had moderate or severe inflammation and eight had inflammation extending >0.5 mm. Inflammation in the alveolar bone marrow space was found with every tooth. The interface between amalgam and canal walls could not be examined. In a few cases where amalgam and tooth had been sectioned, it appeared that there was a marked inflammatory response adjacent to the filling (Fig. 3). Frequently, the inflammation had caused resorption of the adjacent bone. There appeared to be particles of filling material dispersed in the tissues, and this was apparently not an artifact caused by sectioning. Bacteria were found in all of the root canals of teeth filled with amalgam. DISCUSSION The advantages of this method whereby the roots were filled out of the mouth is, first, that moisture contamination of the cavity could be properly controlled. Second, there was no spread of filling material into the periradicular tissues to cause foreign body reactions separate from any response to the root end filling itself. Third, the integrity of the clinical crown was not destroyed, thus preventing bacterial contamination of the pulp space from the oral cavity. The disadvantage of this method was that the tissue responses, particularly root resorption and ankylosis, consequent on the procedures of extraction and replantation were superimposed on the tissue response to root end filling. Because 11 of the 12 roots filled with amalgam were associated with moderate or severe inflammation, the success of root end filling with amalgam in replanted teeth must be questioned. The one other root was also associated with large collections of chronic inflammatory cells but at a distance of several millimeters. The cause of this severe inflammatory response which was much more severe and extensive than that to the IRM fillings is not completely clear. It may be related, first, to the decomposition of the amalgam by corrosion, second, to the effects of bacteria or their products, or, third, to a combination of these. It is known that amalgam alloys corrode to produce products rich in tin, or in the case of high copper amalgams, copper (14). The results of this study showed no discernible difference in reaction between the two types of alloy. Careful examination of the inflammatory reactions showed a large number of macrophages, many of which contained particles, presumably corrosion products. The inflammatory responses to amalgam root end fillings appeared to differ from those to infection in the absence of amalgam, because few polymorphonuclear leukocytes were observed. Although all of the teeth with amalgam fillings contained bacteria in their root canals, it was not possible to observe bacteria in close proximity to the tissues at the root end. This may have been because the presence of amalgam masked their detection as a result of the similarity of the bacterial staining with amalgam alloy particles, or because the
FIG 2. A, IRM root end filling (F) with no inflammation in the adjacent bone marrow (original magnification x50). B, Lack of inflammation at apical part of IRM with fibrous capsule in intimate contact with the material (F) (original magnification x310). C, Inflammation lateral to IRM recently exposed by resorption of tooth (hematoxylin and eosin stain; original magnification x310).
bacteria were taken away by removal of the amalgam fillings prior to sectioning, or because no bacteria were there. It is conjectured that the infection in the root canals may have
384
Pitt Ford et al.
Journal of Endodontics
FtG 3. A, Amalgam root end filling (A) with inflammation extending into bone (original magnification x50). B, Tissue adjacent to amalgam (A) showing moderate inflammation (original magnification x310). C, Cancellous bone apical to root end periodontal ligament (P) showing severe inflammation (/) at a distance >1 mm from the root (Masson trichrome stain; original magnification x50).
accelerated corrosion of the amalgam, which then led to the severe tissue reaction; this aspect is the subject of further investigation. The tissue response to the IRM fillings was much less severe than that to amalgam after the 8-wk period. This confirms the radiographical findings reported by Andreasen and Pitt Ford (13). The possible reasons for this could be that set IRM may be far less cytotoxic; the IRM effectively sealed bacteria inside the root canal system because it was commonly observed that there was very close adaptation of the material to the cavity walls; the eugenol content of the IRM prevented the passage of bacteria alongside by killing such bacteria; the
eugenol killed bacteria in the root canal system; and if bacteria were sealed in the root canal and deprived of a nutrient supply they would die out and in support of this bacteria were only detected in 60% of teeth in contrast to being observed in all of the teeth filled with amalgam. An interesting observation of the tissue response to IRM was the more severe response to the material which was recently exposed at the sides due to resorption (that had almost certainly been initiated by the procedure of extraction and replantation). It is possible that more eugenol was being released in this area than at the true root end where excess eugenol had already been leached out. It would be interesting
Vol. 20, No. 8, August 1994
Healing after IRM Root End Filling
385
mation free. The same distant response with amalgam was observed in this study and warrants further analysis. Maher et al. (20) stated that acute inflammatory cells were immediately adjacent to the IRM but the magnification of the illustrations prevented independent observation; however, the scale of the reaction to IRM certainly appeared to be limited. The design of their study effectively excluded infection of the root canal space by filling the main root canal with guttapercha and sealer, and therefore eliminated the normal rationale for root end filling. The results of the present study would appear to reflect more accurately the clinical situation, where infection of the root canal is common. The tissue response to root end fillings of IRM in replanted teeth after 2 months was significantly less severe and less extensive than that to amalgam. FtG 4. IRM root end filling (F) with bone forming across root end; laterally there is ankylosis (hematoxylin and eosin stain; original magnification x125).
to observe the tissue response at the root end at different time periods. Ankylosis was frequently observed well away from the root end and therefore was without doubt a complication of extraction and replantation, rather than due to any effect of the root end filling material. Solubility of zinc oxide-eugenol sealers is frequently cited as a reason for avoiding their use as permanent root end fillings (15). However, much of the criticism has been related to nonresin-reinforced formulations. Indeed, Jendresen et al. (16) showed that the physical properties from strength to solubility of a resin-reinforced version were markedly improved over those of the basic cement. The tissue response to IRM has been investigated in implantation studies and shown to be favorable after 1 month (17). In one specimen, bone was actually observed in contact with IRM and this further demonstrates the biocompatibility of IRM (Fig. 4). When a pulp capping material is dissolving, tissue can be observed invading the former space (18) and particles of material can also be distinguished in macrophages, apparently being removed via the lymphatic system. In this study, there was no gross dissolution of the material and macrophage activity adjacent to the root end was at a very low level. In a few specimens scattered giant cells were observed on the surface of the material; their appearance was of epithelioid cells coating the material rather than giant cells actively engulfing the material. There is a need to assess the effect of the use of other zinc oxide-eugenol cements at the root end by histological examination since the reaction to different brands has been shown to vary (19); it is possible that the type of macrophage activity is different. This investigation is in hand using this experimental model. The tabular results and conclusions of the study reported by Maher et al. (20) appear to be open to discussion. Their illustrations showed marked chronic inflammation within 1 m m of the root end of teeth filled with amalgam, even though the tissue adjacent to the amalgam appeared to be inflam-
Dr. Pitt Ford is a member of the Department of Conservative Dental Surgery, Dental School, UMDS, Guy's Hospital, London, England. Dr. Andreasen is affiliated with University Hospital, Copenhagen, Denmark. Dr. Dorn is a member of the Department of Endodontics, University of Florida, Gainesville, FL Mr. Kariyawasam is a member of the Division of Anatomy and Cell Biology, UMDS, Guy's Hospital, London, England. Address requests for reprints to Dr. T. R., Pitt Ford, Department of Conservative Dental Surgery, UMDS, Guy's Hospital, London SE1 9RT, England.
References 1. Rud J, Andreasen JO, MOtler Jensen JE. A follow-up study of 1,000 cases treated by endodontic surgery. Int J Oral Surg 1972;1:215-28. 2. Dorn SO, Gartner AH. Retrograde filling materials: a retrospective success-failure study of amalgam, EBA, and IRM. J Endodon 1990;16:391-3. 3. Grung B, Molven O, Halse A. Periapical surgery in a Norwegian County Hospital: follow-up findings of 477 teeth. J Endodon 1990;16:411-7. 4. Frank AL, Glick DH, Patterson SS, Weine FS. Long-term evaluation of surgically placed amalgam fillings. J Endodon 1992;18:391-8. 5. Mattila K, Altonen M. A clinical and roentgenological study of apicoectomized teeth. Odontol Tidsk 1968;76:389-408. 6. Allen RK, Newton CW, Brown CE. A statistical analysis of surgical and nonsurgical endodontic retreatment cases. J Endodon 1989;15:261-6. 7. Ellender G, Ham KN, Harcourt JK. The ultrastructural localization of the corrosion products of dental amalgam. Aust Dent J 1979;24:174-7. 8. Eley BM. Tissue reactions to implanted dental amalgam, including assessment by energy dispersive x-ray micro-analysis. J Pathol 1982;138:25172. 9. Olson AK, MacPherson MG, Hartwell GR, Weller RN, Kulild JC. An in vitro evaluation of injectable thermoplasticized gutta-percha, glass ionomer, and amalgam when used as retrofilling materials. J Endodon 1990;16:361-4. 10. Pitt Ford TR, Roberts GJ. Tissue response to glass ionomer retrograde root fillings. Int Endod J 1990;23:233-8. 11. Rud J, Munksgaard EC, Andreasen JO, Rud V. Retrograde root filling with composite and a dentin-bonding agent. 2. Endod Dent Traumatol 1991 ;7:126-31. 12. Nicholls E. Retrograde filling of the root canal. Oral Surg 1962;15:46373. 13. Andreasen JO, Pitt Ford TR A radiographic study of the effect of various retrograde fillings on periapical healing after replantation. Ended Dent Traumatol (in press). 14. Jensen SJ. Corrosion products of dental amalgam. Scand J Dent Res 1982;90:239-42. 15. Weine FS. Endodontic therapy, 3rd ed. St. Louis: CV Mosby 1982:461. 16. Jendresen MD, Phillips RW, Swartz ML, Norman RD. A comparative study of four zinc oxide and eugenol formutations as restorative materials. Part I. J Prosthet Dent 1969;21:176-83. 17. Cook DJ, Taylor PP. Tissue reactions to improved zinc oxide-eugenol cements. J Dent Child 1973;40:199-207. 18. Cox CF, Bergenholtz G. Healing sequence in capped inflamed dental pulps of rhesus monkeys (Macaca mulatta). Int Endod J 1986;19:113-20. 19. Browne RM, Friend LA. An investigation into the irritant properties of some root filling materials. Arch Oral Biol 1968;13:1355-69. 20. Maher WP, Johnson RL, Hess J, Steiman HR. Biocompatibility of retrograde filling materials in the ferret canine. Amalgam and IRM. Oral Surg 1992;73:738-45.
Printed in U.S.A.
VOL. 20, NO. 8, AUGUST1994
Effect of IRM Root End Fillings on Healing after Replantation T. R. Pitt Ford, BDS, PhD, J. O. Andreasen, S. O. Dorn, DDS, and S. P. Kariyawasam
oxide-eugenol cements as root end fillings. Although some clinicians have used these cements for a long time (12), the scientific rationale for doing so has not been clearly stated. Indeed, the perceived view by many has been that these cements are too soluble for long-term use as root end fillings. In an earlier study, Andreasen and Pitt Ford (13) have indicated radiographically that the periradicular radiolucencies around roots with root end root fillings of IRM were much smaller than those around roots filled with amalgam in replanted teeth in monkeys. The aim of this study was to assess whether the apparent superiority of IRM over amalgam at the radiological level was true at the histological level.
The effect of IRM as a root end filling placed in teeth prior to replantation was examined in 21 molar teeth in monkeys. After extraction, root ends were resected, the canals contaminated with oral bacteria, root end cavities prepared, and fillings of IRM or amalgam placed prior to replantation. After 8 wk the jaws were removed and prepared for histological examination. Bacteria were demonstrated in only 9 of 15 teeth filled with IRM; 18 of the roots (60%) were associated with inflammation, which was only moderate or severe around 5 (17%), and extended >0.1 mm around only 2 roots. In contrast, of the 6 teeth filled with amalgam, all contained bacteria in the root canals and 11 roots were associated with moderate or severe inflammation, which around 8 roots extended >0.5 ram. The difference in severity of inflammation for the two materials was statistically significant (p < 0.001). The tissue response to root end fillings of IRM in replanted teeth was less severe and less extensive than that to amalgam.
M A T E R I A L S AND M E T H O D S Monkeys (Cercopithecus aethiops) that had complete root formation of the mandibular first molars were chosen. After general anesthesia with phencyclidine (2.5 mg/kg body wt, Sernylan; Parke Davis & Co, Copenhagen, Denmark) supplemented with pentobarbitone sodium (30 mg/kg body wt, Mebumal; Dak, Copenhagen, Denmark), the mandibular first molars were extracted using gentle luxation movements. After extraction, the crowns of the teeth were held by forceps while 2 mm of the apices were resected fiat with a diamond wheel under a constant flow of saline. A root end cavity was drilled to a depth of 2 m m with a round bur (diameter, 0.8 mm) under constant saline cooling. The root end cavity was then washed with saline and dried with compressed air. A file contaminated with saliva was placed in the root canal to introduce infection before the root end was filled with one of the following filling materials mixed according to the manufacturers' instructions: amalgam (Standalloy, Degussa, Frankfurt, Germany) or (Luxalloy, Degussa) and zinc oxide-eugenol-based compound: IRM (Caulk, Dentsply, Milford, DE). These filling materials were randomly allocated to the teeth. After completion of the root filling procedure, which took 15 to 20 rain, the molars were replanted and left unsplinted for 8 wk. The monkeys were then killed and the jaw section was removed. The tissue blocks were placed in 10% neutral buffered formalin before being demineralized in a 10% solution of EDTA at pH 6.9. After double infiltration in celloidin-paraffin, the tissue blocks were sectioned perpendicular to the long axis mesiodistally at a thickness of 7 #m. The teeth were
Despite amalgam being widely used as a root end root filling material in endodontic surgery, a number of clinical studies have reported a low rate of success when strict healing criteria are used (1-4). When the success rate of root end fillings with amalgam has been compared with resection of roots filled with gutta-percha and sealer, the success rate with amalgam was not as good ( 1, 3). Furthermore, when amalgam was used as a root end filling after root resection, the success rate was not improved by the presence of the filling (5, 6). Healing was not as good as it should have been and this indicates that amalgam may not have been providing a seal to bacteria. Concern has also been expressed about the stability of amalgam with the migration of metallic particles into the tissues (7, 8). A large number of alternative materials have been investigated both in vitro (9) and in vivo (10, 11). A recent clinical study has indicated that IRM is significantly more successful than amalgam (2). This has led to a renewed interest in zinc
381
382
Pitt Ford et al.
Journal of Endodontics
TABLE 1. Number of roots with inflammation, its extent and severity, the presence of epithelium, and the mean extent of root resorption for each material Material
No. of Roots
No. with Inflammation
Amalgam
12
11
IRM
30
18
No. at Each Severity
No. at Each Extent (mm)
None Few Moderate Severe
1 0 5 6
None _<0.1 _<0.2 -<0.5 >0.5
None Few Moderate Severe
12 13 4 1
None _<0.1 <_0.2 _<0.5 >0.5
1
Epithelium at Root End
Mean Extent of Root Resorption Exposing Filling
0
N/A*
2
0.33
0 3 0 8 12 16 1 1 0
* NA, not available.
sectioned in step-serial sections at 100-urn intervals. At each sectioning level, slides were stained with hematoxylin and eosin, Masson's trichrome, and Brown and Brenn stain. Sections were assessed for inflammation at the root end adjacent to the root end filling material. The severity of the inflammation was recorded as: none, no inflammatory cells; few cells, few inflammatory cells; moderate, inflammatory cells did not obscure the normal tissues; and severe, inflammatory cells replaced normal tissues. The extent of inflammation was recorded as -<0.1 mm, _<0.2 mm, _<0.5 mm, or >0.5 mm. The presence of inflammation in the adjacent bone marrow was registered. The presence or absence of root resorption exposing the root end filling material was noted, and where present the greatest extent toward the crown was recorded as part of the length of the filling (to nearest one-fourth). The presence or absence of ankylosis of the tooth in any part, noticeable dissolution of the root end filling material, giant cells, a fibrous capsule, cementum deposition on the root end filling material, or epithelium covering the root end material was noted. The presence or absence of bacteria in any part of the root canal was also assessed.
TABLE 2. Number of teeth containing bacteria, exhibiting ankylosis, and demonstrating active root resorption for each material Material
No. of Teeth
No. with Bacteria in Root Canals
No. with Ankylosis
No. with Active Root Resorption
Amalgam IRM
6 15
6 9
0 8
5 13
RESULTS A total of 12 roots in 6 teeth filled with amalgam were available for histological examination. As there was no obvious difference between the effects of the two amalgam alloys, the results were combined. Thirty roots in 15 teeth were filled with IRM. The results of individual roots are given in Table 1. When the number of roots with either none or few inflammatory cells were compared with the number with moderate or severe inflammation for the two materials, the difference was statistically significant (p < 0.001, x 2 test, Yates correction). Because it was necessary to remove the amalgam before sectioning, it was not possible to assess the extent of root resorption exposing the filling. The results applicable to individual teeth are given in Table 2. IRM Eighteen roots filled with IRM were associated with inflammation at the root end (60%); only five roots had moderate
FIG 1. Root end of a tooth filled with IRM (F) showing material clinging to canal wall (D) (hematoxylin and eosin stain; original magnification x310).
or severe inflammation and two had inflammation extending >0.1 mm. No tooth had inflammation in the bone marrow space. Bacteria were found in the root canals of 60% of teeth filled with IRM. They were not usually present in large numbers and were most easily observed in the pulp space rather than near the root end filling. It was a general observation that the IRM clung to canal walls following sectioning (Fig. 1). The tissue response next to the exposed root filling generally ranged from no inflammatory response to a few inflammatory cells which were most frequently lymphocytes, although a few epithelioid-type giant cells were noticed against the filling. In cases where resorption
Vol. 20, No. 8, August 1994
Healing after IRM Root End Filling
383
had exposed lateral parts of the filling, an inflammatory response of greater intensity was observed confined to that area alone (Fig. 2). Noticeable dissolution of the cement by cells was not observed.
Amalgam Eleven roots filled with amalgam were associated with periapical inflammation (92%); all had moderate or severe inflammation and eight had inflammation extending >0.5 mm. Inflammation in the alveolar bone marrow space was found with every tooth. The interface between amalgam and canal walls could not be examined. In a few cases where amalgam and tooth had been sectioned, it appeared that there was a marked inflammatory response adjacent to the filling (Fig. 3). Frequently, the inflammation had caused resorption of the adjacent bone. There appeared to be particles of filling material dispersed in the tissues, and this was apparently not an artifact caused by sectioning. Bacteria were found in all of the root canals of teeth filled with amalgam. DISCUSSION The advantages of this method whereby the roots were filled out of the mouth is, first, that moisture contamination of the cavity could be properly controlled. Second, there was no spread of filling material into the periradicular tissues to cause foreign body reactions separate from any response to the root end filling itself. Third, the integrity of the clinical crown was not destroyed, thus preventing bacterial contamination of the pulp space from the oral cavity. The disadvantage of this method was that the tissue responses, particularly root resorption and ankylosis, consequent on the procedures of extraction and replantation were superimposed on the tissue response to root end filling. Because 11 of the 12 roots filled with amalgam were associated with moderate or severe inflammation, the success of root end filling with amalgam in replanted teeth must be questioned. The one other root was also associated with large collections of chronic inflammatory cells but at a distance of several millimeters. The cause of this severe inflammatory response which was much more severe and extensive than that to the IRM fillings is not completely clear. It may be related, first, to the decomposition of the amalgam by corrosion, second, to the effects of bacteria or their products, or, third, to a combination of these. It is known that amalgam alloys corrode to produce products rich in tin, or in the case of high copper amalgams, copper (14). The results of this study showed no discernible difference in reaction between the two types of alloy. Careful examination of the inflammatory reactions showed a large number of macrophages, many of which contained particles, presumably corrosion products. The inflammatory responses to amalgam root end fillings appeared to differ from those to infection in the absence of amalgam, because few polymorphonuclear leukocytes were observed. Although all of the teeth with amalgam fillings contained bacteria in their root canals, it was not possible to observe bacteria in close proximity to the tissues at the root end. This may have been because the presence of amalgam masked their detection as a result of the similarity of the bacterial staining with amalgam alloy particles, or because the
FIG 2. A, IRM root end filling (F) with no inflammation in the adjacent bone marrow (original magnification x50). B, Lack of inflammation at apical part of IRM with fibrous capsule in intimate contact with the material (F) (original magnification x310). C, Inflammation lateral to IRM recently exposed by resorption of tooth (hematoxylin and eosin stain; original magnification x310).
bacteria were taken away by removal of the amalgam fillings prior to sectioning, or because no bacteria were there. It is conjectured that the infection in the root canals may have
384
Pitt Ford et al.
Journal of Endodontics
FtG 3. A, Amalgam root end filling (A) with inflammation extending into bone (original magnification x50). B, Tissue adjacent to amalgam (A) showing moderate inflammation (original magnification x310). C, Cancellous bone apical to root end periodontal ligament (P) showing severe inflammation (/) at a distance >1 mm from the root (Masson trichrome stain; original magnification x50).
accelerated corrosion of the amalgam, which then led to the severe tissue reaction; this aspect is the subject of further investigation. The tissue response to the IRM fillings was much less severe than that to amalgam after the 8-wk period. This confirms the radiographical findings reported by Andreasen and Pitt Ford (13). The possible reasons for this could be that set IRM may be far less cytotoxic; the IRM effectively sealed bacteria inside the root canal system because it was commonly observed that there was very close adaptation of the material to the cavity walls; the eugenol content of the IRM prevented the passage of bacteria alongside by killing such bacteria; the
eugenol killed bacteria in the root canal system; and if bacteria were sealed in the root canal and deprived of a nutrient supply they would die out and in support of this bacteria were only detected in 60% of teeth in contrast to being observed in all of the teeth filled with amalgam. An interesting observation of the tissue response to IRM was the more severe response to the material which was recently exposed at the sides due to resorption (that had almost certainly been initiated by the procedure of extraction and replantation). It is possible that more eugenol was being released in this area than at the true root end where excess eugenol had already been leached out. It would be interesting
Vol. 20, No. 8, August 1994
Healing after IRM Root End Filling
385
mation free. The same distant response with amalgam was observed in this study and warrants further analysis. Maher et al. (20) stated that acute inflammatory cells were immediately adjacent to the IRM but the magnification of the illustrations prevented independent observation; however, the scale of the reaction to IRM certainly appeared to be limited. The design of their study effectively excluded infection of the root canal space by filling the main root canal with guttapercha and sealer, and therefore eliminated the normal rationale for root end filling. The results of the present study would appear to reflect more accurately the clinical situation, where infection of the root canal is common. The tissue response to root end fillings of IRM in replanted teeth after 2 months was significantly less severe and less extensive than that to amalgam. FtG 4. IRM root end filling (F) with bone forming across root end; laterally there is ankylosis (hematoxylin and eosin stain; original magnification x125).
to observe the tissue response at the root end at different time periods. Ankylosis was frequently observed well away from the root end and therefore was without doubt a complication of extraction and replantation, rather than due to any effect of the root end filling material. Solubility of zinc oxide-eugenol sealers is frequently cited as a reason for avoiding their use as permanent root end fillings (15). However, much of the criticism has been related to nonresin-reinforced formulations. Indeed, Jendresen et al. (16) showed that the physical properties from strength to solubility of a resin-reinforced version were markedly improved over those of the basic cement. The tissue response to IRM has been investigated in implantation studies and shown to be favorable after 1 month (17). In one specimen, bone was actually observed in contact with IRM and this further demonstrates the biocompatibility of IRM (Fig. 4). When a pulp capping material is dissolving, tissue can be observed invading the former space (18) and particles of material can also be distinguished in macrophages, apparently being removed via the lymphatic system. In this study, there was no gross dissolution of the material and macrophage activity adjacent to the root end was at a very low level. In a few specimens scattered giant cells were observed on the surface of the material; their appearance was of epithelioid cells coating the material rather than giant cells actively engulfing the material. There is a need to assess the effect of the use of other zinc oxide-eugenol cements at the root end by histological examination since the reaction to different brands has been shown to vary (19); it is possible that the type of macrophage activity is different. This investigation is in hand using this experimental model. The tabular results and conclusions of the study reported by Maher et al. (20) appear to be open to discussion. Their illustrations showed marked chronic inflammation within 1 m m of the root end of teeth filled with amalgam, even though the tissue adjacent to the amalgam appeared to be inflam-
Dr. Pitt Ford is a member of the Department of Conservative Dental Surgery, Dental School, UMDS, Guy's Hospital, London, England. Dr. Andreasen is affiliated with University Hospital, Copenhagen, Denmark. Dr. Dorn is a member of the Department of Endodontics, University of Florida, Gainesville, FL Mr. Kariyawasam is a member of the Division of Anatomy and Cell Biology, UMDS, Guy's Hospital, London, England. Address requests for reprints to Dr. T. R., Pitt Ford, Department of Conservative Dental Surgery, UMDS, Guy's Hospital, London SE1 9RT, England.
References 1. Rud J, Andreasen JO, MOtler Jensen JE. A follow-up study of 1,000 cases treated by endodontic surgery. Int J Oral Surg 1972;1:215-28. 2. Dorn SO, Gartner AH. Retrograde filling materials: a retrospective success-failure study of amalgam, EBA, and IRM. J Endodon 1990;16:391-3. 3. Grung B, Molven O, Halse A. Periapical surgery in a Norwegian County Hospital: follow-up findings of 477 teeth. J Endodon 1990;16:411-7. 4. Frank AL, Glick DH, Patterson SS, Weine FS. Long-term evaluation of surgically placed amalgam fillings. J Endodon 1992;18:391-8. 5. Mattila K, Altonen M. A clinical and roentgenological study of apicoectomized teeth. Odontol Tidsk 1968;76:389-408. 6. Allen RK, Newton CW, Brown CE. A statistical analysis of surgical and nonsurgical endodontic retreatment cases. J Endodon 1989;15:261-6. 7. Ellender G, Ham KN, Harcourt JK. The ultrastructural localization of the corrosion products of dental amalgam. Aust Dent J 1979;24:174-7. 8. Eley BM. Tissue reactions to implanted dental amalgam, including assessment by energy dispersive x-ray micro-analysis. J Pathol 1982;138:25172. 9. Olson AK, MacPherson MG, Hartwell GR, Weller RN, Kulild JC. An in vitro evaluation of injectable thermoplasticized gutta-percha, glass ionomer, and amalgam when used as retrofilling materials. J Endodon 1990;16:361-4. 10. Pitt Ford TR, Roberts GJ. Tissue response to glass ionomer retrograde root fillings. Int Endod J 1990;23:233-8. 11. Rud J, Munksgaard EC, Andreasen JO, Rud V. Retrograde root filling with composite and a dentin-bonding agent. 2. Endod Dent Traumatol 1991 ;7:126-31. 12. Nicholls E. Retrograde filling of the root canal. Oral Surg 1962;15:46373. 13. Andreasen JO, Pitt Ford TR A radiographic study of the effect of various retrograde fillings on periapical healing after replantation. Ended Dent Traumatol (in press). 14. Jensen SJ. Corrosion products of dental amalgam. Scand J Dent Res 1982;90:239-42. 15. Weine FS. Endodontic therapy, 3rd ed. St. Louis: CV Mosby 1982:461. 16. Jendresen MD, Phillips RW, Swartz ML, Norman RD. A comparative study of four zinc oxide and eugenol formutations as restorative materials. Part I. J Prosthet Dent 1969;21:176-83. 17. Cook DJ, Taylor PP. Tissue reactions to improved zinc oxide-eugenol cements. J Dent Child 1973;40:199-207. 18. Cox CF, Bergenholtz G. Healing sequence in capped inflamed dental pulps of rhesus monkeys (Macaca mulatta). Int Endod J 1986;19:113-20. 19. Browne RM, Friend LA. An investigation into the irritant properties of some root filling materials. Arch Oral Biol 1968;13:1355-69. 20. Maher WP, Johnson RL, Hess J, Steiman HR. Biocompatibility of retrograde filling materials in the ferret canine. Amalgam and IRM. Oral Surg 1992;73:738-45.