0099-2399/91/1709-0436/$03.00/0 JOURNAL OF ENDODONTICS Copyright 9 1991 by The American Association of Endodontists
Printed in U.S.A.
VOL, 17, NO. 9, SEPTEMBER1991
Histological Characterization of Bleaching-Induced External Root Resorption in Dogs Ilan Rotstein, CD, Shimon Friedman, DMD, Chaim Mor, DMD, Jonathan Katznelson, DMD, Maurice Sommer, BDS, and Itai Bab, DMD
MATERIALS AND METHODS
External root resorption occasionally develops after intracoronal bleaching with hydrogen peroxide. In this study, an experimental model was established to study thermocatalytic bleaching-induced root resorption in dogs. Histological examination after 6 months revealed that 18% of the teeth had root resorption lesions. The lesions could be divided into three types. In type I, root excavations were associated with a dense inflammatory cell infiltrate. Type II lesions were characterized by granulation tissue formation. In type Ill, the lesions were filled with reparative cementum. The three types probably represent different phases of one process. Calcium hydroxide had no effect on the occurrence or type of resorption. The instability of hydrogen peroxide and the presence of inflammatory resorption lesions 6 months postoperatively suggest hydrogen peroxide-induced toxic radicals or denaturants as potential irritants.
The experiments presented here were performed in six 2yr-old male beagle dogs. Before each treatment, the dogs were anesthetized intravenously with 6% sodium pentobarbitone (30 mg/kg) after intramuscular sedation with 2% Rompun (Bayer, Leverkrusen, Federal Republic of Germany) (1 ml/ kg). In each dog, root canal therapy was performed in a single sitting in all 12 upper and lower incisors. The root canals were prepared with K files to ISO size #25 by using 2.5% sodium hypochlorite for irrigation. The root canals were then obturated with laterally condensed gutta-percha and AH26 sealer (De Trey, Zurich, Switzerland). The access cavities were sealed with a cotton pellet and intermediate restorative material (Caulk Co., Milford, DE). The bleaching procedure was carried out 2 wk after completion of the root canal filling. Before bleaching, a thick layer of petroleum jelly was applied to the gingiva and the teeth were isolated with a rubber dam. The access cavities were reopened, and the gutta-percha filling was removed with #3 Gates Glidden burs from the coronal third of the root canal approximately 3 m m apical to the cementoenamel junction. The dentin walls of the access cavity were cleaned with a small round carbide bur rotating at slow speed and were then rinsed with water and air dried. A small cotton pellet soaked with 3 ~1 of aqueous 30% hydrogen peroxide solution was placed in each access cavity of the 60 experimental teeth. The teeth were then subjected to heat treatment with a 1000 W photoflood lamp. The heat treatment consisted of five cycles of 2-min "on" and 1-min "off" periods. One of the third incisors in each jaw served as either the negative or positive control. In the six negative control teeth, the intracoronal cotton pellet was soaked with saline; otherwise, all of the other operative procedures were similar to those performed in the experimental teeth. In the six positive control teeth, 30% hydrogen peroxide solution (0.4 ml) was injected into the periodontal ligament at the buccal-distal aspect of the tooth. After bleaching, the access cavities were rinsed with copious amounts of water. In 30 experimental teeth, a fresh paste of calcium hydroxide (Calxyl; Otto & Co., Dirmstein, Federal Republic of Germany) was packed into the pulp chambers. In the remaining 30 experimental teeth, a dry cotton pellet was left in the pulp chamber. The access cavities of all of the teeth were then sealed with amalgam. All procedures were carried out under sterile conditions.
Intracoronal bleaching with hydrogen peroxide is occasionally associated with external root resorption (1-8). The resorptive lesions may be extensive, requiring complex operative procedures or even tooth extraction (2, 5-7, 9). In vitro studies suggested that the pathogenic mechanism involves diffusion of the hydrogen peroxide through the radicular dentin (1012). Cementum defects, mainly at the cementoenamel junction, significantly enhance the hydrogen peroxide seepage (12). Experimentally, external root resorption occurred after hydrogen peroxide bleaching in dogs (13). Calcium hydroxide, when used as an intracoronal medication, has been advocated for the treatment of inflammatory root resorption (14). Similarly, it has been proposed for the management and prevention of bleaching-induced root resorption (4, 7, 8). The purpose of this study was to establish by using histological criteria an experimental model in dogs for the study of thermocatalytic bleaching-induced external root resorption. This model has been further used to evaluate the resorption-preventing effect of calcium hydroxide.
436
Vol. 17, No. 9, September 1991
Bleaching-Induced Root Resorption
437
The dogs were examined clinically and radiographically at monthly intervals. They were sacrificed 6 months after bleaching by a lethal dose of pentobarbitone. The anterior segments of the jaws, including the incisor teeth, were separated and fixed in phosphate-buffered Formalin, decalcified in 5% formic acid Formalin, and divided into right and left halves. Each segment containing three incisors was embedded in paraplast. Serial sections (5-um thick) were cut through the coronal third of the root at 0.5-mrn intervals in horizontal planes perpendicular to the long axis of the teeth (Fig. 1). The sections were stained with hematoxylin and eosin and were subjected to regular and polarized light microscopic examination. RESULTS Six months after thermocatalytic bleaching, 10 of the experimental teeth showed external root resorption lesions. These lesions were absent from the negative control teeth. Four of the six positive control teeth exhibited resorption. Radiographically, no evidence of root resorption was observed during the follow-up period. Three teeth were excluded from the experimental groups because of fractures or iatrogenic perforation. The resorptive lesions could be categorized into three types. In type I, a dense inflammatory infiltrate was associated with root erosions which often extended beyond the cementum and went deep into the dentin (Figs. 2 and 3). Occasionally, these resorption lesions resulted in gross morphological root defects (Fig. 3). The eroded root surfaces were shallow and scalloped (Fig. 2). The inflammatory cell infiltrate consisted mainly of plasma cells and macrophages (Figs. 4 and 5). Type II lesions were characterized by the formation of vascularized connective tissue within the resorptive defects. This tissue contained various amounts of inflammatory and fibroblastic cells and occasionally resembled granulation tissue (Figs. 6 to 9). An array of cementoblast-like cells was often seen along the resorbed surface (Figs. 7 and 8). In some instances of type I and type II lesions, the resorption resulted in undermining excavations that left behind unresorbed cementum (Figs. 4 to 6). Some of the resorptive lacunae contained multinucleated hard tissue-resorbing ceils (Figs. 5 and 7). In type III lesions, the resorptive defects were filled completely or partially with reparative cementum-like matrix, restoring the root surface. This matrix appeared continuous with the preexisting cementum but could be distinguished from the underlying dentin by a reversal line (Figs. 10 to 12). When filling of the resorptive lesions was incomplete, the reparative matrix was lined by cementoblast-like cells (Fig. 10). These cells were absent in completely filled lesions where the root surface was lined by periodontal ligament-like tissue (Figs. 11 and 12). In polarized light microscopy, the reparative cementum could be distinguished from intact cementum by its densely packed short lamellae oriented parallel to the root surface. The root surfaces adjacent to the reparative cementum demonstrated numerous prominent Sharpey's fibers that were anchored deep in the cementum. In restored surfaces, these fibers were thin and scanty and extended into the reparative cementum to a distance less than one third that seen in the intact cementum (Fig. 12).
FIG 1. Micrograph of horizontal section through lower right anterior segment showing coronal third of root of incisor teeth (1st, 2rid, 3rd), alveolar bone (AB), and periodontal ligament (arrows). Root canal (doub/e arrows) contains remnants of filling material (hematoxylin and eosin; original magnification x25).
FIG 2. Type I root resorption lesion with scalloped dentinal surface
(arrows) and dense inflammatory infiltrate (//) (hematoxylin and eosin; original magnification x100).
FtG 3. Type I root resorption lesion in incisor root. Note large resorptive defect of macroscopic dimensions (arrows) (hematoxylin and eosin; original magnification x25).
438
Rotstein et al.
Journal of Endodontics
FIG 6. Type II root resorption lesion showing undermining excavation in cementum (C) and dentin (D). Note fibroblasts (arrows), cementoblast-like cells (double arrows), and macrophages (arrowheads) (hematoxylin and eosin; original magnification x400). FIG 4. Type I root resorption lesion consisting of undermining excavation (curved arrows) in cementum (C) and dentin (/9). Defect is filled with infiltrate of plasma cells (arrows) and macrophages (double arrows) (hematoxylin and eosin; original magnification x400).
FIG 7. Type II root resorption lesion with fibrous tissue (FT) and cementoblast-like cells (double arrows). Note a multinucleated giant cell sitting in resorption lacunae (arrow) and a few inflammatory cells (arrowheads) (hematoxylin and eosin; original magnification x400).
FiG 5. Type I root resorption lesion. Cross-sectional plane through undermining excavations (curved arrows) in cementum (C) and dentin (D). Note multinucleated hard tissue-resorbing cell in contact with dentin (arrow) (hematoxylin and eosin; original magnification x200).
The distribution of the resorptive lesions among the three categories is shown in Table 1. All positive controls exhibited type II lesions. Teeth subjected to thermocatalytic bleaching without a calcium hydroxide dressing showed type I and type
Vol. 17, No. 9, September 1991
Bleaching-Induced Root Resorption
439
FIG 9. Type II root resorption lesion showing scalloped dentinal surface (arrows) adjacent to inflamed fibrous tissue (FT). Note reversal line (double arrows) and cementoblast-like cells (arrowheads) alongside reparative cementum (hematoxylin and eosin; original magnification x100).
FIG 8. Type II root resorption lesion showing cemental and dentinal resorbed surface (arrows) adjacent to vascularized connective tissue (FT) containing few foci of inflammatory cells (hematoxylin and eosin; original magnification •
III lesions. Type I and type II lesions were observed in the teeth dressed with calcium hydroxide. DISCUSSION In this controlled animal study, the incidence of external root resorption after thermocatalytic bleaching was approximately 18 %, considerably higher than that found in a previous radiographic survey in humans (7). This may be related to the difference in resolution between microscopy and radiology and to the difference between animal and human studies. In this connection, it is of interest that the present radiographical follow-up failed to reveal the resorption defects seen in the histological sections. In addition, the previous human study covered 1 to 8 yr posttreatment, a period sufficiently long for healing and root repair (7). The study presented here characterizes three distinct types of root resorption lesions associated with inflammation, fibrosis/vascularization, or repair. Although all three types were detected at the same postoperative interval, it is likely that they represent consecutive phases of one process. This process may be similar to other pathological conditions of hard tissue resorption where an initial inflammatory phase is triggered by tissue injury. Once the irritation ceases, the inflammatory infiltrate is replaced by a vascularized granulation tissue which, upon maturation, gives rise to hard tissue-forming cells. In the model presented here, these cells were responsible for the reversal phase during which the radicular defects were filled with reparative cementum.
F~G10. Type III root resorption lesion showing reparative cementumlike matrix (RC). Note normal cementum (NC), reversal line in dentin (double arrows), and small aggregation of cementoblast-like cells (arrows) (hematoxylin and eosin; original magnification x200).
Bleaching agents, especially 30% hydrogen peroxide, have been suggested as stimulants triggering radicular resorptive lesions (1-3, 10-12). It was hypothesized that hydrogen peroxide diffusing into the cervical periodontal ligament through patent dentinal tubules initiates an inflammatory resorption
440
Rotstein et al.
Journal of Endodontics TABLE 1, Proportion of teeth with external root resorption lesions 6 months after thermocataiytic bleaching
Type of Resorption Lesions Inflammatory
Fibrotic
Reparative
Total
Saline control
0/6
0/6
0/6
0/6
H202 control TCB* TCB + Ca(OH)2 Total
0/6 3/29 3/28 6/57
4/6 0/29 2/28 2/57
0/6 2/29 0/28 2/57
4/6 5/29 5/28 10/57
* TCB,Thermocatalyticbleaching.
F~G 11. Type III root resorption lesion showing resorbed root surface covered with reparative cementum (RC). Note reversal line in dentin (double arrows), periodontal ligament (PDL), and alveolar bone (AB) (hematoxylin and eosin; original magnification x40).
FIG 12. Type III root resorption lesion. A, Polarized light micrograph showing Sharpey's fibers in intact cementum (arrows) and reparative cementum (double arrows). B, Regular light micrograph of the same field presented in A showing dentin (D), normal cementum (NC), reparative cementum (RC), dentinal reversal line (double arrows), and periodontal ligament (PDL) (hematoxylin and eosin; original magnification x200).
process which progresses in the presence of bacteria (1, 3). This hypothesis is supported by recent in vitro studies showing diffusion of bleaching agents through dentin, particularly in the presence of cementum defects at the cementoenamel junction (11, 12). However, it is unlikely that hydrogen peroxide directly provoked the present resorptive lesions detected 6 months postoperatively. At this time interval, most of them were associated with a dense inflammatory infiltrate, suggesting the continuous presence of an irritant. Hydrogen peroxide is an unstable solution (15) and is thus unlikely to function in this capacity. On the other hand, it could react with either an organic or inorganic component of the dentin to form highly toxic radicals or denaturants which provoke an inflammatory response (2, 16). Because the saline controls, which were subjected to heat treatment, remained intact, we suggest that the heat factor alone is insufficient for the production of these irritants. Calcium hydroxide has been proven clinically effective in arresting external inflammatory root resorption (14). When resorption occurs, a part of the cementum is lost, the dentinal tubules are exposed, and a communication between the pulp cavity and the periodontal tissues is formed. It has been proposed that calcium hydroxide placed in the pulp cavity penetrates the dentin to increase the pH in the root periphery and to promote repair (17). However, this theory was called into question as both calcium hydroxide paste and hydroxyl ions have been shown to diffuse poorly through dentin (11, 18). Several cases of bleaching-induced root resorption have been treated successfully by intracoronal application of calcium hydroxide (4, 8), though in some other instances, this therapy was ineffective (5, 6, 9). In this study, bleachinginduced resorption was found regardless of the presence of calcium hydroxide, suggesting that this preparation is unsuitable for the prevention of postbleaching root resorption. Bleaching with a sodium perborate-water paste has been advocated as an alternative unassociated with complicating root resorption lesions (19). In addition, an experimental study in dogs showed that even walking bleach with sodium perborate mixed with hydrogen peroxide was less harmful than was thermocatalytic bleaching (13). Nevertheless, the thermocatalytic technique with 30% hydrogen peroxide is faster and more effective. Therefore, it is important to determine the pathogenic mechanism of bleaching-induced resorption and provide a rational choice that involves minimal risk of root resorption without sacrificing the bleaching efficacy. Drs. Rotstein, Friedman, Mor, Katznelson,and Sommer are affiliated with the Department of Endodontics, and Dr. Bab is professor and head, Bone
Vol. 17, No. 9, September 1991 Laboratory, The Hebrew University-Hadassah Faculty of Dental Medicine, Jerusalem, Israel. Address requests for reprints to Dr. Ilan Rotstein, Department of Endodontics, Hadassah Faculty of Medicine, P.O. Box 1172, Jerusalem, 91010, Israel.
References 1. Harrington GH, Natkin E. External resorption associated with bleaching of pulpless teeth. J Endodon 1979;5:344-8. 2. Lado EA, Stanley HR, Weisman MI. Cervical resorption in bleached teeth. Oral Surg 1983;55:78-80. 3. Cvek M, Lindvall AM. External root resorption following bleaching of pulpless teeth with oxygen peroxide. Endod Dent Traumato11985;1:56-60. 4. Montgomery S. External cervical resorption after bleaching a pulpless tooth. Oral Surg 1984;57:203-6. 5. Latcham NL. Postbteaching cervical resorption. J Endodon 1986;12:262-4. 6. Goon WWY, Cohen S, Borer RF. External cervical root resorption following bleaching. J Endodon 1986;12:414-8. 7. Friedman S, Rotstein I, Libfeld H, Stabholz A, Heling I. Incidence of external root resorption and esthetic results in 58 bleached pulpless teeth. Endod Dent Traumato11988;4:23-6. 8. Gimlin DR, Schindler G. The management of postbleaching cervical resorption. J Endodon 1990;16:292-7.
Bleaching-Induced Root Resorption
441
9. Friedman S. Surgical-restorative treatment of bleaching related external root resorption. Endod Dent Traumato11989;5:63-7. 10. Kehoe JC. pH reversal following in vitro bleaching of pulpless teeth. J Endodon 1987;13:6-9. 11. Fuss Z, Szajkis S, Tagger M. Tubular permeability to calcium hydroxide and to bleaching agents. J Endodon 1989;t 5:362-4. 12. Rotstein I, Torek Y, Misgav R. Effect of cementum defects on radicular penetration of 30% H202 during intracoronal bleaching. J Endodon 1991 ;17:230-8. 13. Madison S, Walton RE, Chiles S. Cervical resorption following bleaching of endodontically treated teeth: an in vivo study [Abstract] J Endodon 1987;13:135. 14. Heithersay GS. Calcium hydroxide in the treatment of pulpless teeth with associated pathology. J Br Endod Soc 1975;8:74-93. 15. Hardman PK, Moore DL, Petteway GH. Stability of hydrogen peroxide as a bleaching agent. Gen Dent 1985;33:121-2. 16. Halliwell B, Gutteridge JMC. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem J 1984;219:1-14. 17. Tronstad L, Andreasen JO, Hasselgren G, Kristerson L, Riis I. pH changes in dental tissues after root canal filling with calcium hydroxide. J Endodon 1981 ;7:17-21. 18. Wang JD, Hume WR. Diffusion of hydrogen ion and hydroxyl ion from various sources through dentine. Int Endod J 1988;21:17-26. 19. Holmstrup G, Palm AM, Lambjerg-Hansen H. Bleaching of discoloured root-filled teeth. Endod Dent Traumato11988;4:197-201.
The Way It Was We are currently concerned about the spread of certain infectious diseases. Dental office procedures have surely changed as a result. Yet present pestilences might be considered pale in comparison to the "black death" which swept medieval Europe. Bubonic plague killed 20,000 of London's population of 200,000 in 1625. At the time of the last epidemic in 1665 the population of London was 460,000. In that 1 year, over 80,000 Londoners died of plague. The death of over 17% of a city's population in 1 year is almost beyond comprehension today. Zachariah Yeomans
Printed in U.S.A.
VOL, 17, NO. 9, SEPTEMBER1991
Histological Characterization of Bleaching-Induced External Root Resorption in Dogs Ilan Rotstein, CD, Shimon Friedman, DMD, Chaim Mor, DMD, Jonathan Katznelson, DMD, Maurice Sommer, BDS, and Itai Bab, DMD
MATERIALS AND METHODS
External root resorption occasionally develops after intracoronal bleaching with hydrogen peroxide. In this study, an experimental model was established to study thermocatalytic bleaching-induced root resorption in dogs. Histological examination after 6 months revealed that 18% of the teeth had root resorption lesions. The lesions could be divided into three types. In type I, root excavations were associated with a dense inflammatory cell infiltrate. Type II lesions were characterized by granulation tissue formation. In type Ill, the lesions were filled with reparative cementum. The three types probably represent different phases of one process. Calcium hydroxide had no effect on the occurrence or type of resorption. The instability of hydrogen peroxide and the presence of inflammatory resorption lesions 6 months postoperatively suggest hydrogen peroxide-induced toxic radicals or denaturants as potential irritants.
The experiments presented here were performed in six 2yr-old male beagle dogs. Before each treatment, the dogs were anesthetized intravenously with 6% sodium pentobarbitone (30 mg/kg) after intramuscular sedation with 2% Rompun (Bayer, Leverkrusen, Federal Republic of Germany) (1 ml/ kg). In each dog, root canal therapy was performed in a single sitting in all 12 upper and lower incisors. The root canals were prepared with K files to ISO size #25 by using 2.5% sodium hypochlorite for irrigation. The root canals were then obturated with laterally condensed gutta-percha and AH26 sealer (De Trey, Zurich, Switzerland). The access cavities were sealed with a cotton pellet and intermediate restorative material (Caulk Co., Milford, DE). The bleaching procedure was carried out 2 wk after completion of the root canal filling. Before bleaching, a thick layer of petroleum jelly was applied to the gingiva and the teeth were isolated with a rubber dam. The access cavities were reopened, and the gutta-percha filling was removed with #3 Gates Glidden burs from the coronal third of the root canal approximately 3 m m apical to the cementoenamel junction. The dentin walls of the access cavity were cleaned with a small round carbide bur rotating at slow speed and were then rinsed with water and air dried. A small cotton pellet soaked with 3 ~1 of aqueous 30% hydrogen peroxide solution was placed in each access cavity of the 60 experimental teeth. The teeth were then subjected to heat treatment with a 1000 W photoflood lamp. The heat treatment consisted of five cycles of 2-min "on" and 1-min "off" periods. One of the third incisors in each jaw served as either the negative or positive control. In the six negative control teeth, the intracoronal cotton pellet was soaked with saline; otherwise, all of the other operative procedures were similar to those performed in the experimental teeth. In the six positive control teeth, 30% hydrogen peroxide solution (0.4 ml) was injected into the periodontal ligament at the buccal-distal aspect of the tooth. After bleaching, the access cavities were rinsed with copious amounts of water. In 30 experimental teeth, a fresh paste of calcium hydroxide (Calxyl; Otto & Co., Dirmstein, Federal Republic of Germany) was packed into the pulp chambers. In the remaining 30 experimental teeth, a dry cotton pellet was left in the pulp chamber. The access cavities of all of the teeth were then sealed with amalgam. All procedures were carried out under sterile conditions.
Intracoronal bleaching with hydrogen peroxide is occasionally associated with external root resorption (1-8). The resorptive lesions may be extensive, requiring complex operative procedures or even tooth extraction (2, 5-7, 9). In vitro studies suggested that the pathogenic mechanism involves diffusion of the hydrogen peroxide through the radicular dentin (1012). Cementum defects, mainly at the cementoenamel junction, significantly enhance the hydrogen peroxide seepage (12). Experimentally, external root resorption occurred after hydrogen peroxide bleaching in dogs (13). Calcium hydroxide, when used as an intracoronal medication, has been advocated for the treatment of inflammatory root resorption (14). Similarly, it has been proposed for the management and prevention of bleaching-induced root resorption (4, 7, 8). The purpose of this study was to establish by using histological criteria an experimental model in dogs for the study of thermocatalytic bleaching-induced external root resorption. This model has been further used to evaluate the resorption-preventing effect of calcium hydroxide.
436
Vol. 17, No. 9, September 1991
Bleaching-Induced Root Resorption
437
The dogs were examined clinically and radiographically at monthly intervals. They were sacrificed 6 months after bleaching by a lethal dose of pentobarbitone. The anterior segments of the jaws, including the incisor teeth, were separated and fixed in phosphate-buffered Formalin, decalcified in 5% formic acid Formalin, and divided into right and left halves. Each segment containing three incisors was embedded in paraplast. Serial sections (5-um thick) were cut through the coronal third of the root at 0.5-mrn intervals in horizontal planes perpendicular to the long axis of the teeth (Fig. 1). The sections were stained with hematoxylin and eosin and were subjected to regular and polarized light microscopic examination. RESULTS Six months after thermocatalytic bleaching, 10 of the experimental teeth showed external root resorption lesions. These lesions were absent from the negative control teeth. Four of the six positive control teeth exhibited resorption. Radiographically, no evidence of root resorption was observed during the follow-up period. Three teeth were excluded from the experimental groups because of fractures or iatrogenic perforation. The resorptive lesions could be categorized into three types. In type I, a dense inflammatory infiltrate was associated with root erosions which often extended beyond the cementum and went deep into the dentin (Figs. 2 and 3). Occasionally, these resorption lesions resulted in gross morphological root defects (Fig. 3). The eroded root surfaces were shallow and scalloped (Fig. 2). The inflammatory cell infiltrate consisted mainly of plasma cells and macrophages (Figs. 4 and 5). Type II lesions were characterized by the formation of vascularized connective tissue within the resorptive defects. This tissue contained various amounts of inflammatory and fibroblastic cells and occasionally resembled granulation tissue (Figs. 6 to 9). An array of cementoblast-like cells was often seen along the resorbed surface (Figs. 7 and 8). In some instances of type I and type II lesions, the resorption resulted in undermining excavations that left behind unresorbed cementum (Figs. 4 to 6). Some of the resorptive lacunae contained multinucleated hard tissue-resorbing ceils (Figs. 5 and 7). In type III lesions, the resorptive defects were filled completely or partially with reparative cementum-like matrix, restoring the root surface. This matrix appeared continuous with the preexisting cementum but could be distinguished from the underlying dentin by a reversal line (Figs. 10 to 12). When filling of the resorptive lesions was incomplete, the reparative matrix was lined by cementoblast-like cells (Fig. 10). These cells were absent in completely filled lesions where the root surface was lined by periodontal ligament-like tissue (Figs. 11 and 12). In polarized light microscopy, the reparative cementum could be distinguished from intact cementum by its densely packed short lamellae oriented parallel to the root surface. The root surfaces adjacent to the reparative cementum demonstrated numerous prominent Sharpey's fibers that were anchored deep in the cementum. In restored surfaces, these fibers were thin and scanty and extended into the reparative cementum to a distance less than one third that seen in the intact cementum (Fig. 12).
FIG 1. Micrograph of horizontal section through lower right anterior segment showing coronal third of root of incisor teeth (1st, 2rid, 3rd), alveolar bone (AB), and periodontal ligament (arrows). Root canal (doub/e arrows) contains remnants of filling material (hematoxylin and eosin; original magnification x25).
FIG 2. Type I root resorption lesion with scalloped dentinal surface
(arrows) and dense inflammatory infiltrate (//) (hematoxylin and eosin; original magnification x100).
FtG 3. Type I root resorption lesion in incisor root. Note large resorptive defect of macroscopic dimensions (arrows) (hematoxylin and eosin; original magnification x25).
438
Rotstein et al.
Journal of Endodontics
FIG 6. Type II root resorption lesion showing undermining excavation in cementum (C) and dentin (D). Note fibroblasts (arrows), cementoblast-like cells (double arrows), and macrophages (arrowheads) (hematoxylin and eosin; original magnification x400). FIG 4. Type I root resorption lesion consisting of undermining excavation (curved arrows) in cementum (C) and dentin (/9). Defect is filled with infiltrate of plasma cells (arrows) and macrophages (double arrows) (hematoxylin and eosin; original magnification x400).
FIG 7. Type II root resorption lesion with fibrous tissue (FT) and cementoblast-like cells (double arrows). Note a multinucleated giant cell sitting in resorption lacunae (arrow) and a few inflammatory cells (arrowheads) (hematoxylin and eosin; original magnification x400).
FiG 5. Type I root resorption lesion. Cross-sectional plane through undermining excavations (curved arrows) in cementum (C) and dentin (D). Note multinucleated hard tissue-resorbing cell in contact with dentin (arrow) (hematoxylin and eosin; original magnification x200).
The distribution of the resorptive lesions among the three categories is shown in Table 1. All positive controls exhibited type II lesions. Teeth subjected to thermocatalytic bleaching without a calcium hydroxide dressing showed type I and type
Vol. 17, No. 9, September 1991
Bleaching-Induced Root Resorption
439
FIG 9. Type II root resorption lesion showing scalloped dentinal surface (arrows) adjacent to inflamed fibrous tissue (FT). Note reversal line (double arrows) and cementoblast-like cells (arrowheads) alongside reparative cementum (hematoxylin and eosin; original magnification x100).
FIG 8. Type II root resorption lesion showing cemental and dentinal resorbed surface (arrows) adjacent to vascularized connective tissue (FT) containing few foci of inflammatory cells (hematoxylin and eosin; original magnification •
III lesions. Type I and type II lesions were observed in the teeth dressed with calcium hydroxide. DISCUSSION In this controlled animal study, the incidence of external root resorption after thermocatalytic bleaching was approximately 18 %, considerably higher than that found in a previous radiographic survey in humans (7). This may be related to the difference in resolution between microscopy and radiology and to the difference between animal and human studies. In this connection, it is of interest that the present radiographical follow-up failed to reveal the resorption defects seen in the histological sections. In addition, the previous human study covered 1 to 8 yr posttreatment, a period sufficiently long for healing and root repair (7). The study presented here characterizes three distinct types of root resorption lesions associated with inflammation, fibrosis/vascularization, or repair. Although all three types were detected at the same postoperative interval, it is likely that they represent consecutive phases of one process. This process may be similar to other pathological conditions of hard tissue resorption where an initial inflammatory phase is triggered by tissue injury. Once the irritation ceases, the inflammatory infiltrate is replaced by a vascularized granulation tissue which, upon maturation, gives rise to hard tissue-forming cells. In the model presented here, these cells were responsible for the reversal phase during which the radicular defects were filled with reparative cementum.
F~G10. Type III root resorption lesion showing reparative cementumlike matrix (RC). Note normal cementum (NC), reversal line in dentin (double arrows), and small aggregation of cementoblast-like cells (arrows) (hematoxylin and eosin; original magnification x200).
Bleaching agents, especially 30% hydrogen peroxide, have been suggested as stimulants triggering radicular resorptive lesions (1-3, 10-12). It was hypothesized that hydrogen peroxide diffusing into the cervical periodontal ligament through patent dentinal tubules initiates an inflammatory resorption
440
Rotstein et al.
Journal of Endodontics TABLE 1, Proportion of teeth with external root resorption lesions 6 months after thermocataiytic bleaching
Type of Resorption Lesions Inflammatory
Fibrotic
Reparative
Total
Saline control
0/6
0/6
0/6
0/6
H202 control TCB* TCB + Ca(OH)2 Total
0/6 3/29 3/28 6/57
4/6 0/29 2/28 2/57
0/6 2/29 0/28 2/57
4/6 5/29 5/28 10/57
* TCB,Thermocatalyticbleaching.
F~G 11. Type III root resorption lesion showing resorbed root surface covered with reparative cementum (RC). Note reversal line in dentin (double arrows), periodontal ligament (PDL), and alveolar bone (AB) (hematoxylin and eosin; original magnification x40).
FIG 12. Type III root resorption lesion. A, Polarized light micrograph showing Sharpey's fibers in intact cementum (arrows) and reparative cementum (double arrows). B, Regular light micrograph of the same field presented in A showing dentin (D), normal cementum (NC), reparative cementum (RC), dentinal reversal line (double arrows), and periodontal ligament (PDL) (hematoxylin and eosin; original magnification x200).
process which progresses in the presence of bacteria (1, 3). This hypothesis is supported by recent in vitro studies showing diffusion of bleaching agents through dentin, particularly in the presence of cementum defects at the cementoenamel junction (11, 12). However, it is unlikely that hydrogen peroxide directly provoked the present resorptive lesions detected 6 months postoperatively. At this time interval, most of them were associated with a dense inflammatory infiltrate, suggesting the continuous presence of an irritant. Hydrogen peroxide is an unstable solution (15) and is thus unlikely to function in this capacity. On the other hand, it could react with either an organic or inorganic component of the dentin to form highly toxic radicals or denaturants which provoke an inflammatory response (2, 16). Because the saline controls, which were subjected to heat treatment, remained intact, we suggest that the heat factor alone is insufficient for the production of these irritants. Calcium hydroxide has been proven clinically effective in arresting external inflammatory root resorption (14). When resorption occurs, a part of the cementum is lost, the dentinal tubules are exposed, and a communication between the pulp cavity and the periodontal tissues is formed. It has been proposed that calcium hydroxide placed in the pulp cavity penetrates the dentin to increase the pH in the root periphery and to promote repair (17). However, this theory was called into question as both calcium hydroxide paste and hydroxyl ions have been shown to diffuse poorly through dentin (11, 18). Several cases of bleaching-induced root resorption have been treated successfully by intracoronal application of calcium hydroxide (4, 8), though in some other instances, this therapy was ineffective (5, 6, 9). In this study, bleachinginduced resorption was found regardless of the presence of calcium hydroxide, suggesting that this preparation is unsuitable for the prevention of postbleaching root resorption. Bleaching with a sodium perborate-water paste has been advocated as an alternative unassociated with complicating root resorption lesions (19). In addition, an experimental study in dogs showed that even walking bleach with sodium perborate mixed with hydrogen peroxide was less harmful than was thermocatalytic bleaching (13). Nevertheless, the thermocatalytic technique with 30% hydrogen peroxide is faster and more effective. Therefore, it is important to determine the pathogenic mechanism of bleaching-induced resorption and provide a rational choice that involves minimal risk of root resorption without sacrificing the bleaching efficacy. Drs. Rotstein, Friedman, Mor, Katznelson,and Sommer are affiliated with the Department of Endodontics, and Dr. Bab is professor and head, Bone
Vol. 17, No. 9, September 1991 Laboratory, The Hebrew University-Hadassah Faculty of Dental Medicine, Jerusalem, Israel. Address requests for reprints to Dr. Ilan Rotstein, Department of Endodontics, Hadassah Faculty of Medicine, P.O. Box 1172, Jerusalem, 91010, Israel.
References 1. Harrington GH, Natkin E. External resorption associated with bleaching of pulpless teeth. J Endodon 1979;5:344-8. 2. Lado EA, Stanley HR, Weisman MI. Cervical resorption in bleached teeth. Oral Surg 1983;55:78-80. 3. Cvek M, Lindvall AM. External root resorption following bleaching of pulpless teeth with oxygen peroxide. Endod Dent Traumato11985;1:56-60. 4. Montgomery S. External cervical resorption after bleaching a pulpless tooth. Oral Surg 1984;57:203-6. 5. Latcham NL. Postbteaching cervical resorption. J Endodon 1986;12:262-4. 6. Goon WWY, Cohen S, Borer RF. External cervical root resorption following bleaching. J Endodon 1986;12:414-8. 7. Friedman S, Rotstein I, Libfeld H, Stabholz A, Heling I. Incidence of external root resorption and esthetic results in 58 bleached pulpless teeth. Endod Dent Traumato11988;4:23-6. 8. Gimlin DR, Schindler G. The management of postbleaching cervical resorption. J Endodon 1990;16:292-7.
Bleaching-Induced Root Resorption
441
9. Friedman S. Surgical-restorative treatment of bleaching related external root resorption. Endod Dent Traumato11989;5:63-7. 10. Kehoe JC. pH reversal following in vitro bleaching of pulpless teeth. J Endodon 1987;13:6-9. 11. Fuss Z, Szajkis S, Tagger M. Tubular permeability to calcium hydroxide and to bleaching agents. J Endodon 1989;t 5:362-4. 12. Rotstein I, Torek Y, Misgav R. Effect of cementum defects on radicular penetration of 30% H202 during intracoronal bleaching. J Endodon 1991 ;17:230-8. 13. Madison S, Walton RE, Chiles S. Cervical resorption following bleaching of endodontically treated teeth: an in vivo study [Abstract] J Endodon 1987;13:135. 14. Heithersay GS. Calcium hydroxide in the treatment of pulpless teeth with associated pathology. J Br Endod Soc 1975;8:74-93. 15. Hardman PK, Moore DL, Petteway GH. Stability of hydrogen peroxide as a bleaching agent. Gen Dent 1985;33:121-2. 16. Halliwell B, Gutteridge JMC. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem J 1984;219:1-14. 17. Tronstad L, Andreasen JO, Hasselgren G, Kristerson L, Riis I. pH changes in dental tissues after root canal filling with calcium hydroxide. J Endodon 1981 ;7:17-21. 18. Wang JD, Hume WR. Diffusion of hydrogen ion and hydroxyl ion from various sources through dentine. Int Endod J 1988;21:17-26. 19. Holmstrup G, Palm AM, Lambjerg-Hansen H. Bleaching of discoloured root-filled teeth. Endod Dent Traumato11988;4:197-201.
The Way It Was We are currently concerned about the spread of certain infectious diseases. Dental office procedures have surely changed as a result. Yet present pestilences might be considered pale in comparison to the "black death" which swept medieval Europe. Bubonic plague killed 20,000 of London's population of 200,000 in 1625. At the time of the last epidemic in 1665 the population of London was 460,000. In that 1 year, over 80,000 Londoners died of plague. The death of over 17% of a city's population in 1 year is almost beyond comprehension today. Zachariah Yeomans
