- Email: [email protected]
Wet and Dry Deep Cavity Preparations Compared by a Novel Odontoblast Culture Technique
JOURNAL OF ENDODONTICS Copyright © 2001 by The American Association of Endodontists
Printed in U.S.A. VOL. 27, NO. 2, FEBRUARY 2001
Wet and Dry Deep Cavity Preparations Compared by a Novel Odontoblast Culture Technique Kimberly G. Holt, DDS, Paul D. Eleazer, DDS, MS, and James P. Scheetz, PhD
Using another approach, odontoblast cell bodies were scraped from the dentinal walls of freshly extracted human teeth and grown in culture by Schiess et al. (5). These odontoblasts first assumed the characteristic spindle shape of fibroblasts and were viable for the 30 days of the experiment. The investigators noted that the odontoblasts began mitosis and by the 12th day the cells began to change morphologically to resemble epithelium. These findings were also reported by Zussman and Iochim (6) using rat incisors. Magloire et al. (7) suggested a model for studying the odontoblast in situ. Thick longitudinal slices of freshly extracted human teeth were cultured for 30 days. Odontoblasts were reported as viable within dentinal tubules and at the pulpal surface. Trypan blue staining of nonvital cells is a common procedure used in cell culture research. Basically vital cells will not allow the stain to penetrate their cell membranes. As early as 1917 Pappenheimer (8) used trypan blue to detect nonvital cells. Van Bezooijen et al. (9) used electron microscopy to verify the accuracy of using trypan blue to differentiate vital and nonvital cells. They found the ultrastructural integrity of cells that did not stain with trypan blue was intact, whereas the organelles of nonvital cells were destroyed. Scanning electron microscopy (SEM) was used by Lee et al. (10) to verify the penetration of trypan blue stain into nonvital, damaged epithelial cells. Recently Tja¨derhane et al. (11) used the inverted tooth crown as a vessel for culture medium and found that odontoblasts remained functional for at least 5 days. This study used a modification of this technique to observe odontoblasts after deep cavity preparation using a high-speed handpiece, with and without water coolant. Trypan blue was used to disclose nonvital odontoblasts.
The human odontoblast’s unique cellular extension within dentin does not easily allow culturing by traditional methods. This study leaves these cells in their natural position in the dentin. Deep preparations were made through the occlusal surfaces of extracted human third molars. The crowns were separated from the roots and the pulps gently teased from the chambers, leaving the odontoblast layer intact. These inverted pulp chambers were then incubated in cell culture medium for 2 to 4 days. Trypan blue staining was used to detect nonvital odontoblasts, and the differentiation between vital and nonvital cells was verified by SEM and toluidine vital staining. Control teeth and areas not adjacent to the preparation showed no blue staining, indicating intact cells. Areas of nonvital cells were greatest with wider preparation. Irrigation decreased odontoblast death with wide preparations. No difference due to irrigation was detected in narrow preps. A comparison of wet preparations to which heat was applied versus dry preparations showed statistically similar results. This study provides a simple in vitro method for the study of odontoblasts with their processes intact within dentin.
MATERIALS AND METHODS
Odontoblast studies have used a variety of models often involving separation of the central pulpal tissues from the odontoblast layer. Heywood and Appleton (1) longitudinally sectioned rat incisors and removed the central pulp tissue. They verified a residual layer of odontoblasts attached to the dentin wall through light microscopy, as well as transmission electron microscopy. A retained layer of odontoblasts in dentin of both bovine and human teeth was also reported by Flieder and Fisher (2). Pincus (3) confirmed the presence of odontoblasts retained on and into the dentin wall of human teeth after pulp tissue removal. Chadha and Bishop (4) longitudinally sectioned freshly extracted third molars, and through light microscopy reported the complete retention of odontoblasts in dentin.
Intact, freshly extracted human third molars were used for this study. As a preface to the experiment, eight teeth received the treatment protocol without cavity preparation to ensure that odontoblasts would remain alive under the conditions used in this investigation. The experimental teeth were stored in sterile phosphate-buffered saline (PBS) until roots were removed, within 4 h after extraction. Each tooth was externally disinfected with 70% ethanol on a cotton gauze. Tissue, bone, and debris were curetted away and the tooth again wiped with alcohol. The teeth were radiographed to determine chamber morphology. A circumferential groove approximately 2 mm deep was cut in the root at the level of the pulp 103
104
Holt et al.
Journal of Endodontics
chamber floor, as determined by radiograph. Separation of the roots from the crown was achieved using dental extraction forceps. The pulp easily pulled free from the chamber walls during root separation or by gentle broaching. A tougher, thin layer of tissue remained on the dentinal wall. This layer was considered to be odontoblasts, with their processes remaining within the dentinal tubules. Sterile PBS was used to gently rinse the chambers. Tooth crowns with their chambers facing upward were placed in labeled, sterile 24-well plates. The culture medium for these experiments was composed of Dulbecco’s modified Eagle’s medium/F-12 medium nutrient mixture (Life Technologies, Grand Island, NY), 10% fetal bovine serum, and an antibiotic combination of 100 IU/ml penicillin G ⫹ 100 g/ml streptomycin. Enough cell culture medium was added to completely cover each tooth. The medium was changed daily under a laminar flow hood. Odontoblast cultures were incubated at 37°C in a 5% CO2 and humidified air mixture for 2 to 4 days. When ready for examination, odontoblastic cultures were rinsed gently with 1 ml PBS. Trypan blue stain, 0.4%, diluted 1:1 (vol: vol) with PBS was placed inside the chambers for 2 min. The chambers were then rinsed with 3 ml PBS to reveal any nonvital odontoblasts. A light microscope with photographic capabilities was used to record results. For experiment 1, 32 teeth had been divided into four groups of eight teeth each. Deep cavity preparations were created using a #557 fissure cross-cut bur and high-speed handpiece. Eight teeth were prepared with irrigation and a narrow preparation size (group IN). Another group of eight was prepared without irrigation and a narrow preparation size (group DN). A group was prepared with irrigation and a wide preparation size (group IW). The last eight teeth were prepared without irrigation and a wide preparation size (group DW). Narrow preparations were approximately 3 mm2 and wide preparations were 7 mm2 in size. Teeth prepared with irrigation were never desiccated. Cavity preparations in the irrigation groups were sealed with a PBS-moistened cotton pellet and Cavit (ESPE America, Norristown, PA). Preparations without irrigation were sealed with a dry cotton pellet and Cavit. In experiment 2, 10 teeth had been arbitrarily subdivided into two subgroups of five each to compare the effects of heat and desiccation preparation to heat without desiccation. Five teeth, termed group M, were prepared without irrigation and sealed with Cavit over a PBS-moistened cotton pellet to minimize hygroscopic effect of the temporary filling material on the odontoblasts. The heat without desiccation group of five (group H) was prepared wet then briefly dried with a cotton pellet and a paper cone. Heat was applied to the floor of the cavity preparation using a System B heat source (Analytic Endodontics, Orange, CA) at 300°C for 10 s and then sealed with a PBS-moistened cotton pellet and Cavit.
Measurements Using a ⫻4 stereomicroscope and a digital readout micrometer, measurements of widest and narrowest dimensions of zone of trypan blue stain were used to determine the area of cell death. To ensure uniformity of preparation depth, teeth were longitudinally sectioned through the deepest part of the cavity preparation and the remaining dentin thickness from cavity preparation to chamber roof was recorded. A single operator performed all operative procedures and measurements.
For experiment 1, a two-factor analysis of variance was computed with and without irrigation and for wide and narrow preparation sizes. Independent variables consisted of wide or narrow preparation sizes, with and without irrigation. The dependent variables were remaining dentin and odontoblastic death. Thus the analysis was done twice, once for each of the dependent variables. The independent variables were the same for both analyses. For experiment 2, an independent group t test was used to compare the desiccated cavity preparations versus those that were not desiccated. This analysis was done twice, once with remaining dentin as the dependent variable, and again with area of odontoblastic death as the dependent variable. SEM Cellular status of the odontoblasts adjacent to the cavity preparation was examined by SEM. A specimen prepared without irrigation and a wide prep size was preserved in gluteraldehyde and air-dried for SEM study. The tooth was then attached to a standard SEM mounting stub using silver conducting paint. A sputtered coating of gold was applied, and the roof of the pulpal chamber was examined with a Phillips 505 SEM. Images were recorded on Polaroid 55 film. Toluidine Blue Toluidine blue, which stains the nuclei of vital cells, was used to contrast areas of vital odontoblasts with zones of odontoblast death. Another single specimen, prepared without irrigation and a wide preparation size, was stained with toluidine blue vital stain. In this instance only vital cells appeared blue, because nonvital cells will not stain with toluidine blue. RESULTS The control teeth exhibited no evidence of odontoblastic death when stained with trypan blue. This test confirmed that the technique allows odontoblasts to live under the conditions of this study. The results with SEM and vital staining by toluidine blue also showed that odontoblasts outside the affected areas remained vital and confirmed the validity of the trypan blue staining technique for identification of nonvital odontoblasts. The lack of statistical differences in the remaining dentin categories in both experiments 1 and 2 indicated that the test methods provided approximately equal depth cavities in all groups (Table 1). With odontoblastic death as the dependent variable, there was a statistically significant interaction between the independent variables, size, and irrigant. Therefore separate analyses were done for each level of the two independent variables by use of an independent group t test. When comparing wide versus narrow cavity preparations under wet conditions, the difference was statistically significant (p ⬎ 0.05), with the wide preparation causing a larger area of odontoblastic death. With dry wide preparations, the area of cell death was also statistically significantly greater (p ⬍ 0.01), compared with dry narrow preparations. Comparing wet versus dry cutting when the preparations were wide showed a statistically significant difference (p ⬎ 0.05), with the dry preparation resulting in a greater area of odontoblastic death. The comparison of wet versus dry with narrow cavity preparations indicated no statistically significant difference (p ⬎ 0.05).
Vol. 27, No. 2, February 2001
Entire Odontoblasts In Vitro for Wet or Dry Preps
105
TABLE 1. Descriptive statistics— experiment 1 Mean
SD
n
Remaining dentin Wet irrigation Wide Prep (IW) Wet irrigation Narrow Prep (IN) Dry irrigation Wide Prep (DW) Dry irrigation Narrow Prep (DN)
1.0787 1.3900 1.2625 1.4588
0.4615 0.6324 0.3751 0.5456
8 8 8 8
Odontoblast cell death Wet irrigation Wide Prep (IW) Wet irrigation Narrow Prep (IN) Dry irrigation Wide Prep (DW) Dry irrigation Narrow Prep (DN)
1.5625 0.2318 3.3863 0.3813
1.4131 0.2414 0.6382 0.3042
8 8 8 8
IW vs. IN, p ⬍ 0.05 DW vs. DN, p ⬍ 0.01 DW vs. IW, p ⬍ 0.05 DN vs. IN, not statistically significant
FIG 1. Photograph of roof of tooth chamber. Trypan staining of nonvital odontoblasts seen as a dark circle in the center of the roof. Area stained by trypan blue measured 3.14 mm2.
In experiment 2, with all wide cavity preparations, the results showed no difference when comparing teeth heated and desiccated during preparation versus those heated after wet preparation (Table 2). As with experiment 1, both remaining dentin and area of odontoblastic death were equal (p ⬎ 0.05). Longitudinal sections showed blue stain in tubules adjacent to stained areas of the pulp chamber. No stain was observed in dentin associated with vital odontoblasts. Zones of cell death were always found in the pulp chamber roof adjacent to the deepest part of the cavity preparation. Figure 1 is a photograph taken of a narrow preparation made with irrigation. Figure 2 depicts a wide preparation without irrigant. Confirmation of cell death was accomplished by SEM and staining of vital cells. Figure 3 is a SEM at ⫻680 that shows open dentinal tubules, apparently after the death of odontoblasts from a wide, dry cavity preparation. Little cellular debris remains. The toluidine vital staining test confirmed a nonstaining section (nonvital cells), corresponding to the location of typical trypan nonvital staining area adjacent to the preparation.
central pulp tissue is apparently achievable because of the anchorage of the odontoblasts in dentin by cytoplasmic processes. Odontoblasts scraped from dentin walls and grown in tissue culture medium do not maintain their morphology, nor do they likely behave as odontoblasts in situ. The effects of cavity preparation on these cultured cells cannot be directly equated to the in vivo condition. The pilot study of eight teeth demonstrated that pulp vitality can be maintained after at least 4 h without blood flow. Although not of this magnitude, sharply limited pulpal circulation by local anesthetics have been demonstrated by Kim et al. (12). Apparently odontoblast death is not instantaneous after circulatory shutdown. The tissue culture medium takes over the functions of nutrition, oxygenation, and removal of waste products. Tjaderhane et al. (11) changed the cell culture medium daily, which obviously allowed continued cell function, probably because of good contact with the monolayer of odontoblasts. The location of the zones of odontoblastic cell death and its relationship to the preparation was verified by sectioning each tooth longitudinally through the depth of its preparation. Blue stain in tubules adjacent to pulpal zones of staining were interpreted as
DISCUSSION This model for studying the reactions of human odontoblasts to various techniques and dental materials may provide a new approach for research. Although a great many histological studies have correlated dead odontoblasts with operative trauma, this may be the first method to quantitate damage by measuring area of odontoblast death. The easy separation of the odontoblast cell body layer from the TABLE 2. Descriptive statistics— experiment 2
Remaining dentin Desiccated Not desiccated Odontoblast cell death Desiccated Not desiccated
Mean
SD
n
1.8160 1.8300
0.5422 0.5572
5 5
1.1620 0.1900
There were no statistical differences found.
1.3631 0.1922
5 5
FIG 2. Photograph of roof of the chamber showing a dark area of trypan stain (arrow #1) corresponding to nonvital odontoblasts. Arrow #2 shows vital odontoblasts, which do not stain with trypan blue. The dentinal rim is shown by arrow #3. Trypan blue-stained area measured 3.56 mm2.
106
Holt et al.
FIG 3. SEM taken at ⫻680 of a specimen prepared without irrigation and a wide preparation size. Open dentinal tubules are seen where previous trypan blue stain existed. White bar is 0.1 mm.
death of the odontoblastic process as well as the cell body. There was no blue stain in the pulpal areas or dentinal tubules in areas where the odontoblast cell bodies were considered vital or in any of the control teeth. The effects of wide cavity preparation without irrigation are shown in the SEM (Fig. 3). A confluent cell membrane ends abruptly and a well-circumscribed area of open dentinal tubules is seen. One possible explanation is the aspiration of the affected odontoblasts into their dentinal tubules as reported by Langeland (13). Brannstrom (14) found the aspirated odontoblasts underwent autolysis within 6 h. Our incubation time of 2 to 4 days allowed adequate time for the autolysis to occur, and the daily medium change and 3 ml rinsing after trypan staining could account for the absence of cellular debris. A statistically significant correlation was achieved in the group prepared without irrigation and a narrow preparation size. Wider cavity size perhaps did not allow concentration of heat in a small area or allowed more air cooling to occur. The buildup of heat without the cooling effects of air or water spray seems to increase the amount of odontoblastic death. The lack of statistically significant correlation between remaining dentin and amount of odontoblastic cell death found in all categories confirms the standardization of technique in creating similar depth cavities for study. Cutting equal depth preparations is technically challenging because of the convexity of the pulp horns in immature teeth. Wide preparations, whether with or without irrigation, caused more odontoblastic destruction than narrow preparations. Irrigation in narrow preparations created the smallest zones of odontoblastic death, and wide, dry preparations created the largest zones of odontoblastic death. Whether the heat or the desiccation plays the major role during a dry preparation was questioned. Odontoblasts of teeth that were prepared without irrigation and immediately treated with a PBSmoistened cotton pellet fared no worse statistically than odontoblasts whose teeth were prepared with irrigation and treated with
Journal of Endodontics
10 s of 300°C heat. This finding suggests that desiccation is not more harmful than heat, at least for the temperature and time of application used herein. This brief heat would arguably be similar to the best case scenario for a conservative cavity preparation. The merits of the technique used in this study are many. Extracted third molars are readily available, allowing inexpensive experimentation on human odontoblasts and avoiding the use of animals. The in vitro effects of various dental materials and techniques can be evaluated simply and quickly with this unique technique. The in vitro odontoblast tissue culture model suggested by Tja¨derhane et al. (11) and modified in this study allowed odontoblasts to remain viable for 2 to 4 days. The findings of this study suggest odontoblasts can die in response to cavity preparation. The zone of odontoblastic cell death was confirmed through trypan blue staining, SEM examination, and toluidine vital cell staining. One goal of cavity preparation is to minimize trauma to pulp cells. Dead tracts produced by odontoblast cell death could become a means of ingress for bacteria and their inflammatory products. The authors wish to thank Dr. Sven Gorr for the use of his laboratory and equipment. Dr. Holt is a former endodontic resident of the University of Louisville and is currently practicing in Plano, TX. Dr. Eleazer is associate professor, director of Postgraduate Endodontics, and interim chair of the Periodontics, Endodontics, and Dental Hygiene Department, and Dr. Scheetz is a professor in the Department of Diagnostic Sciences, Prosthodontics, and Restorative Dentistry, University of Louisville, School of Dentistry, Louisville, KY. Address requests for reprints to Dr. Paul Eleazer, University of Louisville School of Dentistry, 501 South Preston Street, Room D-35, Louisville, KY 40202.
References 1. Heywood B, Appleton J. The ultrastructure of the rat incisor odontoblast in organ culture. Arch Oral Biol 1984;27:327–9. 2. Flieder D, Fisher A. The rate of endogenous oxygen consumption in bovine dental pulp. J Dental Res 1955;34:921–3. 3. Pincus P. Some physiological data on the human dental pulp. Br Dent J 1950;89:143– 8. 4. Chadha S, Bishop M. Effect of mechanical removal of the pulp upon the retention of odontoblasts around the pulp chamber of human third molars. Arch Oral Biol 1996;41:905–9. 5. Schiess A, Okigaki T, Rounds D. Human odontoblasts in vitro. I. The harvest of homogenous cell populations. J Dent Res 1996;45:1101– 4. 6. Zussman W, Iochim H. Growth of odontoblasts in vitro. I. Dental tissue culture studies. Lab Invest 1964;13:371–7. 7. Magloire H, Joeffre A, Bleicher F. An in vitro model of human dental pulp repair. J Dent Res 1996;75:1971– 8. 8. Pappenheimer A. Experimental studies upon lymphocytes. J Exp Med 1917;25:633–50. 9. Van Bezooijen C, Van Noord, M, Knook D. The viability of parenchymal liver cells isolated from young and old rats. Mechanisms of Aging and Development 1974;3:107–19. 10. Lee R, Chambers C, O’Brodovich H. Trypan blue method for the identification of areas of damage to airway epithelium due to mechanical trauma. Scan Elect Micros 1984;3:1267–71. 11. Tja¨derhane L, Salo T, Larmas M, Overall C. A novel organ culture method to study the function of human odontoblasts in vitro: gelatinase expression by odontoblasts is differentially regulated by TGF-b1. J Dent Res 1998;7:1486 –96. 12. Kim S, Salo T, Larmas M, Overall C. Effects of local anesthetics on pulpal blood flow in dogs. J Dent Res 1984;63:650 –2. 13. Langeland K. Tissue changes in the dental pulp. Oslo: Oslo University Press, 1957. 14. Brannstrom M. The effect of dentin desiccation and aspirated odontoblasts on the pulp. J Prosthet Dent 1968;20:165.
Printed in U.S.A. VOL. 27, NO. 2, FEBRUARY 2001
Wet and Dry Deep Cavity Preparations Compared by a Novel Odontoblast Culture Technique Kimberly G. Holt, DDS, Paul D. Eleazer, DDS, MS, and James P. Scheetz, PhD
Using another approach, odontoblast cell bodies were scraped from the dentinal walls of freshly extracted human teeth and grown in culture by Schiess et al. (5). These odontoblasts first assumed the characteristic spindle shape of fibroblasts and were viable for the 30 days of the experiment. The investigators noted that the odontoblasts began mitosis and by the 12th day the cells began to change morphologically to resemble epithelium. These findings were also reported by Zussman and Iochim (6) using rat incisors. Magloire et al. (7) suggested a model for studying the odontoblast in situ. Thick longitudinal slices of freshly extracted human teeth were cultured for 30 days. Odontoblasts were reported as viable within dentinal tubules and at the pulpal surface. Trypan blue staining of nonvital cells is a common procedure used in cell culture research. Basically vital cells will not allow the stain to penetrate their cell membranes. As early as 1917 Pappenheimer (8) used trypan blue to detect nonvital cells. Van Bezooijen et al. (9) used electron microscopy to verify the accuracy of using trypan blue to differentiate vital and nonvital cells. They found the ultrastructural integrity of cells that did not stain with trypan blue was intact, whereas the organelles of nonvital cells were destroyed. Scanning electron microscopy (SEM) was used by Lee et al. (10) to verify the penetration of trypan blue stain into nonvital, damaged epithelial cells. Recently Tja¨derhane et al. (11) used the inverted tooth crown as a vessel for culture medium and found that odontoblasts remained functional for at least 5 days. This study used a modification of this technique to observe odontoblasts after deep cavity preparation using a high-speed handpiece, with and without water coolant. Trypan blue was used to disclose nonvital odontoblasts.
The human odontoblast’s unique cellular extension within dentin does not easily allow culturing by traditional methods. This study leaves these cells in their natural position in the dentin. Deep preparations were made through the occlusal surfaces of extracted human third molars. The crowns were separated from the roots and the pulps gently teased from the chambers, leaving the odontoblast layer intact. These inverted pulp chambers were then incubated in cell culture medium for 2 to 4 days. Trypan blue staining was used to detect nonvital odontoblasts, and the differentiation between vital and nonvital cells was verified by SEM and toluidine vital staining. Control teeth and areas not adjacent to the preparation showed no blue staining, indicating intact cells. Areas of nonvital cells were greatest with wider preparation. Irrigation decreased odontoblast death with wide preparations. No difference due to irrigation was detected in narrow preps. A comparison of wet preparations to which heat was applied versus dry preparations showed statistically similar results. This study provides a simple in vitro method for the study of odontoblasts with their processes intact within dentin.
MATERIALS AND METHODS
Odontoblast studies have used a variety of models often involving separation of the central pulpal tissues from the odontoblast layer. Heywood and Appleton (1) longitudinally sectioned rat incisors and removed the central pulp tissue. They verified a residual layer of odontoblasts attached to the dentin wall through light microscopy, as well as transmission electron microscopy. A retained layer of odontoblasts in dentin of both bovine and human teeth was also reported by Flieder and Fisher (2). Pincus (3) confirmed the presence of odontoblasts retained on and into the dentin wall of human teeth after pulp tissue removal. Chadha and Bishop (4) longitudinally sectioned freshly extracted third molars, and through light microscopy reported the complete retention of odontoblasts in dentin.
Intact, freshly extracted human third molars were used for this study. As a preface to the experiment, eight teeth received the treatment protocol without cavity preparation to ensure that odontoblasts would remain alive under the conditions used in this investigation. The experimental teeth were stored in sterile phosphate-buffered saline (PBS) until roots were removed, within 4 h after extraction. Each tooth was externally disinfected with 70% ethanol on a cotton gauze. Tissue, bone, and debris were curetted away and the tooth again wiped with alcohol. The teeth were radiographed to determine chamber morphology. A circumferential groove approximately 2 mm deep was cut in the root at the level of the pulp 103
104
Holt et al.
Journal of Endodontics
chamber floor, as determined by radiograph. Separation of the roots from the crown was achieved using dental extraction forceps. The pulp easily pulled free from the chamber walls during root separation or by gentle broaching. A tougher, thin layer of tissue remained on the dentinal wall. This layer was considered to be odontoblasts, with their processes remaining within the dentinal tubules. Sterile PBS was used to gently rinse the chambers. Tooth crowns with their chambers facing upward were placed in labeled, sterile 24-well plates. The culture medium for these experiments was composed of Dulbecco’s modified Eagle’s medium/F-12 medium nutrient mixture (Life Technologies, Grand Island, NY), 10% fetal bovine serum, and an antibiotic combination of 100 IU/ml penicillin G ⫹ 100 g/ml streptomycin. Enough cell culture medium was added to completely cover each tooth. The medium was changed daily under a laminar flow hood. Odontoblast cultures were incubated at 37°C in a 5% CO2 and humidified air mixture for 2 to 4 days. When ready for examination, odontoblastic cultures were rinsed gently with 1 ml PBS. Trypan blue stain, 0.4%, diluted 1:1 (vol: vol) with PBS was placed inside the chambers for 2 min. The chambers were then rinsed with 3 ml PBS to reveal any nonvital odontoblasts. A light microscope with photographic capabilities was used to record results. For experiment 1, 32 teeth had been divided into four groups of eight teeth each. Deep cavity preparations were created using a #557 fissure cross-cut bur and high-speed handpiece. Eight teeth were prepared with irrigation and a narrow preparation size (group IN). Another group of eight was prepared without irrigation and a narrow preparation size (group DN). A group was prepared with irrigation and a wide preparation size (group IW). The last eight teeth were prepared without irrigation and a wide preparation size (group DW). Narrow preparations were approximately 3 mm2 and wide preparations were 7 mm2 in size. Teeth prepared with irrigation were never desiccated. Cavity preparations in the irrigation groups were sealed with a PBS-moistened cotton pellet and Cavit (ESPE America, Norristown, PA). Preparations without irrigation were sealed with a dry cotton pellet and Cavit. In experiment 2, 10 teeth had been arbitrarily subdivided into two subgroups of five each to compare the effects of heat and desiccation preparation to heat without desiccation. Five teeth, termed group M, were prepared without irrigation and sealed with Cavit over a PBS-moistened cotton pellet to minimize hygroscopic effect of the temporary filling material on the odontoblasts. The heat without desiccation group of five (group H) was prepared wet then briefly dried with a cotton pellet and a paper cone. Heat was applied to the floor of the cavity preparation using a System B heat source (Analytic Endodontics, Orange, CA) at 300°C for 10 s and then sealed with a PBS-moistened cotton pellet and Cavit.
Measurements Using a ⫻4 stereomicroscope and a digital readout micrometer, measurements of widest and narrowest dimensions of zone of trypan blue stain were used to determine the area of cell death. To ensure uniformity of preparation depth, teeth were longitudinally sectioned through the deepest part of the cavity preparation and the remaining dentin thickness from cavity preparation to chamber roof was recorded. A single operator performed all operative procedures and measurements.
For experiment 1, a two-factor analysis of variance was computed with and without irrigation and for wide and narrow preparation sizes. Independent variables consisted of wide or narrow preparation sizes, with and without irrigation. The dependent variables were remaining dentin and odontoblastic death. Thus the analysis was done twice, once for each of the dependent variables. The independent variables were the same for both analyses. For experiment 2, an independent group t test was used to compare the desiccated cavity preparations versus those that were not desiccated. This analysis was done twice, once with remaining dentin as the dependent variable, and again with area of odontoblastic death as the dependent variable. SEM Cellular status of the odontoblasts adjacent to the cavity preparation was examined by SEM. A specimen prepared without irrigation and a wide prep size was preserved in gluteraldehyde and air-dried for SEM study. The tooth was then attached to a standard SEM mounting stub using silver conducting paint. A sputtered coating of gold was applied, and the roof of the pulpal chamber was examined with a Phillips 505 SEM. Images were recorded on Polaroid 55 film. Toluidine Blue Toluidine blue, which stains the nuclei of vital cells, was used to contrast areas of vital odontoblasts with zones of odontoblast death. Another single specimen, prepared without irrigation and a wide preparation size, was stained with toluidine blue vital stain. In this instance only vital cells appeared blue, because nonvital cells will not stain with toluidine blue. RESULTS The control teeth exhibited no evidence of odontoblastic death when stained with trypan blue. This test confirmed that the technique allows odontoblasts to live under the conditions of this study. The results with SEM and vital staining by toluidine blue also showed that odontoblasts outside the affected areas remained vital and confirmed the validity of the trypan blue staining technique for identification of nonvital odontoblasts. The lack of statistical differences in the remaining dentin categories in both experiments 1 and 2 indicated that the test methods provided approximately equal depth cavities in all groups (Table 1). With odontoblastic death as the dependent variable, there was a statistically significant interaction between the independent variables, size, and irrigant. Therefore separate analyses were done for each level of the two independent variables by use of an independent group t test. When comparing wide versus narrow cavity preparations under wet conditions, the difference was statistically significant (p ⬎ 0.05), with the wide preparation causing a larger area of odontoblastic death. With dry wide preparations, the area of cell death was also statistically significantly greater (p ⬍ 0.01), compared with dry narrow preparations. Comparing wet versus dry cutting when the preparations were wide showed a statistically significant difference (p ⬎ 0.05), with the dry preparation resulting in a greater area of odontoblastic death. The comparison of wet versus dry with narrow cavity preparations indicated no statistically significant difference (p ⬎ 0.05).
Vol. 27, No. 2, February 2001
Entire Odontoblasts In Vitro for Wet or Dry Preps
105
TABLE 1. Descriptive statistics— experiment 1 Mean
SD
n
Remaining dentin Wet irrigation Wide Prep (IW) Wet irrigation Narrow Prep (IN) Dry irrigation Wide Prep (DW) Dry irrigation Narrow Prep (DN)
1.0787 1.3900 1.2625 1.4588
0.4615 0.6324 0.3751 0.5456
8 8 8 8
Odontoblast cell death Wet irrigation Wide Prep (IW) Wet irrigation Narrow Prep (IN) Dry irrigation Wide Prep (DW) Dry irrigation Narrow Prep (DN)
1.5625 0.2318 3.3863 0.3813
1.4131 0.2414 0.6382 0.3042
8 8 8 8
IW vs. IN, p ⬍ 0.05 DW vs. DN, p ⬍ 0.01 DW vs. IW, p ⬍ 0.05 DN vs. IN, not statistically significant
FIG 1. Photograph of roof of tooth chamber. Trypan staining of nonvital odontoblasts seen as a dark circle in the center of the roof. Area stained by trypan blue measured 3.14 mm2.
In experiment 2, with all wide cavity preparations, the results showed no difference when comparing teeth heated and desiccated during preparation versus those heated after wet preparation (Table 2). As with experiment 1, both remaining dentin and area of odontoblastic death were equal (p ⬎ 0.05). Longitudinal sections showed blue stain in tubules adjacent to stained areas of the pulp chamber. No stain was observed in dentin associated with vital odontoblasts. Zones of cell death were always found in the pulp chamber roof adjacent to the deepest part of the cavity preparation. Figure 1 is a photograph taken of a narrow preparation made with irrigation. Figure 2 depicts a wide preparation without irrigant. Confirmation of cell death was accomplished by SEM and staining of vital cells. Figure 3 is a SEM at ⫻680 that shows open dentinal tubules, apparently after the death of odontoblasts from a wide, dry cavity preparation. Little cellular debris remains. The toluidine vital staining test confirmed a nonstaining section (nonvital cells), corresponding to the location of typical trypan nonvital staining area adjacent to the preparation.
central pulp tissue is apparently achievable because of the anchorage of the odontoblasts in dentin by cytoplasmic processes. Odontoblasts scraped from dentin walls and grown in tissue culture medium do not maintain their morphology, nor do they likely behave as odontoblasts in situ. The effects of cavity preparation on these cultured cells cannot be directly equated to the in vivo condition. The pilot study of eight teeth demonstrated that pulp vitality can be maintained after at least 4 h without blood flow. Although not of this magnitude, sharply limited pulpal circulation by local anesthetics have been demonstrated by Kim et al. (12). Apparently odontoblast death is not instantaneous after circulatory shutdown. The tissue culture medium takes over the functions of nutrition, oxygenation, and removal of waste products. Tjaderhane et al. (11) changed the cell culture medium daily, which obviously allowed continued cell function, probably because of good contact with the monolayer of odontoblasts. The location of the zones of odontoblastic cell death and its relationship to the preparation was verified by sectioning each tooth longitudinally through the depth of its preparation. Blue stain in tubules adjacent to pulpal zones of staining were interpreted as
DISCUSSION This model for studying the reactions of human odontoblasts to various techniques and dental materials may provide a new approach for research. Although a great many histological studies have correlated dead odontoblasts with operative trauma, this may be the first method to quantitate damage by measuring area of odontoblast death. The easy separation of the odontoblast cell body layer from the TABLE 2. Descriptive statistics— experiment 2
Remaining dentin Desiccated Not desiccated Odontoblast cell death Desiccated Not desiccated
Mean
SD
n
1.8160 1.8300
0.5422 0.5572
5 5
1.1620 0.1900
There were no statistical differences found.
1.3631 0.1922
5 5
FIG 2. Photograph of roof of the chamber showing a dark area of trypan stain (arrow #1) corresponding to nonvital odontoblasts. Arrow #2 shows vital odontoblasts, which do not stain with trypan blue. The dentinal rim is shown by arrow #3. Trypan blue-stained area measured 3.56 mm2.
106
Holt et al.
FIG 3. SEM taken at ⫻680 of a specimen prepared without irrigation and a wide preparation size. Open dentinal tubules are seen where previous trypan blue stain existed. White bar is 0.1 mm.
death of the odontoblastic process as well as the cell body. There was no blue stain in the pulpal areas or dentinal tubules in areas where the odontoblast cell bodies were considered vital or in any of the control teeth. The effects of wide cavity preparation without irrigation are shown in the SEM (Fig. 3). A confluent cell membrane ends abruptly and a well-circumscribed area of open dentinal tubules is seen. One possible explanation is the aspiration of the affected odontoblasts into their dentinal tubules as reported by Langeland (13). Brannstrom (14) found the aspirated odontoblasts underwent autolysis within 6 h. Our incubation time of 2 to 4 days allowed adequate time for the autolysis to occur, and the daily medium change and 3 ml rinsing after trypan staining could account for the absence of cellular debris. A statistically significant correlation was achieved in the group prepared without irrigation and a narrow preparation size. Wider cavity size perhaps did not allow concentration of heat in a small area or allowed more air cooling to occur. The buildup of heat without the cooling effects of air or water spray seems to increase the amount of odontoblastic death. The lack of statistically significant correlation between remaining dentin and amount of odontoblastic cell death found in all categories confirms the standardization of technique in creating similar depth cavities for study. Cutting equal depth preparations is technically challenging because of the convexity of the pulp horns in immature teeth. Wide preparations, whether with or without irrigation, caused more odontoblastic destruction than narrow preparations. Irrigation in narrow preparations created the smallest zones of odontoblastic death, and wide, dry preparations created the largest zones of odontoblastic death. Whether the heat or the desiccation plays the major role during a dry preparation was questioned. Odontoblasts of teeth that were prepared without irrigation and immediately treated with a PBSmoistened cotton pellet fared no worse statistically than odontoblasts whose teeth were prepared with irrigation and treated with
Journal of Endodontics
10 s of 300°C heat. This finding suggests that desiccation is not more harmful than heat, at least for the temperature and time of application used herein. This brief heat would arguably be similar to the best case scenario for a conservative cavity preparation. The merits of the technique used in this study are many. Extracted third molars are readily available, allowing inexpensive experimentation on human odontoblasts and avoiding the use of animals. The in vitro effects of various dental materials and techniques can be evaluated simply and quickly with this unique technique. The in vitro odontoblast tissue culture model suggested by Tja¨derhane et al. (11) and modified in this study allowed odontoblasts to remain viable for 2 to 4 days. The findings of this study suggest odontoblasts can die in response to cavity preparation. The zone of odontoblastic cell death was confirmed through trypan blue staining, SEM examination, and toluidine vital cell staining. One goal of cavity preparation is to minimize trauma to pulp cells. Dead tracts produced by odontoblast cell death could become a means of ingress for bacteria and their inflammatory products. The authors wish to thank Dr. Sven Gorr for the use of his laboratory and equipment. Dr. Holt is a former endodontic resident of the University of Louisville and is currently practicing in Plano, TX. Dr. Eleazer is associate professor, director of Postgraduate Endodontics, and interim chair of the Periodontics, Endodontics, and Dental Hygiene Department, and Dr. Scheetz is a professor in the Department of Diagnostic Sciences, Prosthodontics, and Restorative Dentistry, University of Louisville, School of Dentistry, Louisville, KY. Address requests for reprints to Dr. Paul Eleazer, University of Louisville School of Dentistry, 501 South Preston Street, Room D-35, Louisville, KY 40202.
References 1. Heywood B, Appleton J. The ultrastructure of the rat incisor odontoblast in organ culture. Arch Oral Biol 1984;27:327–9. 2. Flieder D, Fisher A. The rate of endogenous oxygen consumption in bovine dental pulp. J Dental Res 1955;34:921–3. 3. Pincus P. Some physiological data on the human dental pulp. Br Dent J 1950;89:143– 8. 4. Chadha S, Bishop M. Effect of mechanical removal of the pulp upon the retention of odontoblasts around the pulp chamber of human third molars. Arch Oral Biol 1996;41:905–9. 5. Schiess A, Okigaki T, Rounds D. Human odontoblasts in vitro. I. The harvest of homogenous cell populations. J Dent Res 1996;45:1101– 4. 6. Zussman W, Iochim H. Growth of odontoblasts in vitro. I. Dental tissue culture studies. Lab Invest 1964;13:371–7. 7. Magloire H, Joeffre A, Bleicher F. An in vitro model of human dental pulp repair. J Dent Res 1996;75:1971– 8. 8. Pappenheimer A. Experimental studies upon lymphocytes. J Exp Med 1917;25:633–50. 9. Van Bezooijen C, Van Noord, M, Knook D. The viability of parenchymal liver cells isolated from young and old rats. Mechanisms of Aging and Development 1974;3:107–19. 10. Lee R, Chambers C, O’Brodovich H. Trypan blue method for the identification of areas of damage to airway epithelium due to mechanical trauma. Scan Elect Micros 1984;3:1267–71. 11. Tja¨derhane L, Salo T, Larmas M, Overall C. A novel organ culture method to study the function of human odontoblasts in vitro: gelatinase expression by odontoblasts is differentially regulated by TGF-b1. J Dent Res 1998;7:1486 –96. 12. Kim S, Salo T, Larmas M, Overall C. Effects of local anesthetics on pulpal blood flow in dogs. J Dent Res 1984;63:650 –2. 13. Langeland K. Tissue changes in the dental pulp. Oslo: Oslo University Press, 1957. 14. Brannstrom M. The effect of dentin desiccation and aspirated odontoblasts on the pulp. J Prosthet Dent 1968;20:165.