Development of periradicular lesions in immunosuppressed rats

Development of periradicular lesions in immunosuppressed rats Phillip A. Waterman, Jr., DDS, MS, a Mahmoud Torabinejad, DMD, MSD, PhD, b Paul J. McMillan, PhD, c and James D. Kettering, PhD, d Santa Rosa, Calif. LOMA LINDA UNIVERSITY

Problem. The role of bacteria has been well established in pulpal and periapical diseases, but the contribution of the host defenses is less clear. Obiectlves. The purpose of this study was to compare periradicular lesion development in immunosuppressed rats with that in normal rats. Study Design. Fifteen rats were given weekly injections of Cytoxan (Bristol Laboratories) to suppress their immune systems. The pulps of mandibular first molars of these animals and another 15 rats that had received no medications were exposed and left open to their oral flora. The rats were killed at 2, 4, and 6 weeks. Radiographic analysis was performed by means of a computer linked to a digitizing board and stylus. In addition, specimens were decalcified, sectioned, stained, and examined under a microscope with a grid to quantify relative percentages of surface areas of bone, root, periodontal ligament, marrow spaces, soft tissue, and inflammatory infiltrate. Results. Statistical analysis showed a significantly greater radiographic bone loss in the immunosuppressed group only at 4 weeks. No significant histologic differences were found between the two groups. Conclusion. Our results suggest that reduction of circulating leukocytes may not significantly affect the development of periradicular pathosis in rats. (Oral Surg Oral Med Oral Pathol Oral Iladiol Endod 1998;85:720-5)

Development of periradicular lesions occurs as an interaction between the oral flora and the existing host defenses. 1 Kakehashi et al. 2 demonstrated the importance of bacteria when periradicular lesion formation after mechanical pulp exposure in germ-free rats is compared with that in conventional rats. The germ-free rats did not develop periradicular lesions and showed evidence of repair at the exposure sites. However, necrosis of pulpal tissue and development of periradicular lesions occurred after 15 days in the conventional rats. This was further demonstrated in monkeys by Moller et at.,3 who removed intact pulps and replaced them in the teeth under sterile or non-sterile conditions. The teeth with contaminated pulps developed periradicular lesions, whereas the sterile pulps did not.

Development of periradicular lesions occurs as a response of the host to the invasion of antigens--specifically, bacteria or their by-products. This host response can be associated with a highly vascular system of gran- ulation tissue at the root apex. Such tissue usually contains a large number of inflammatory cells and appears to be the host's method of walling off the infec-

aprivate practice, Santa Rosa, Calif.; former graduate student, School of Dentistry. bprofessor of Endodontics and Director, Graduate Endodontics, School of Dentistry. cprofessor of Anatomy, School of Medicine dprofessor of Microbiology, School of Medicine Received for publication Aug. 12, 1996; returned for revision Nov. 9, 1996; accepted for publication Dec. 3, 1997. Copyright © 1998 by Mosby, Inc. 1079-2104/98/$5.00 + 0 7/15/88231

tion from the rest of the body. The absence of bacteria in most periradicular lesions attests to its success, 4 although this has been disputed by some investigators. 5

Stern et al. 6 quantified the percentage of inflammatory cells in human periradicular granulomas. They found that macrophages were the predominant inflammatory cells, followed in descending order of prevalence by lymphocytes, plasma cells, and neutrophils. However, in 15% of the lesions, lymphocytes were the predominant cells. Stern et al. 6 further demonstrated that approximately half of the cells in the lesions were inflammatory ceils. These investigators were unable to correlate the relative presence of inflammatory cells with symptoms, duration of the lesion, or treatment. In a later study, Stern et al.7 found T-lymphocytes to be more numerous than macrophages and concluded that cell-mediated immunity must play a key role in the pathogenesis of periradicular lesions.

Kontiainen et at.8 examined the presence of various inflammatory cells in human periradicular lesions using monoclonal antibodies and an immunoperoxidase tech- nique. They reported that more than 50% of the cells were lymphocytes and that the rest were neutrophils, macrophages, and natural killer cells, in descending order of prevalence. Of the lymphocytes, more than one half were T cells with a T-cytotoxic/suppressor-to-T- helper/inducer ratio of 2:1.

In two separate studies, Yu and Stashenko 9 and Stashenko and Yu 10 examined the role of leukocytes in the pathogenesis of periradicular lesions in rats. They found lymphocytes and polymorphonuclear neutrophils

720

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY Volume 85, Number 6

to be the predominant inflammatory cells. In addition, they observed that T-lymphocytes outnumbered B- lymphocytes and hypothesized that T-lymphocytes play a significant role in the pathogenesis of periradicular lesions. This confirmed the findings of Torabinejad and Kettering, 11 who showed that T cells outnumbered B cells by a factor of 2 to 20. Even though these investigators have demonstrated the presence of all types of leukocytes in periradicular lesions, the importance of each cellular component has not been fully investi- gated. Wallstrom et al. 12 compared the formation of periradicular lesions in athymic, nude rats with that in normal rats after pulp exposure. The athymic rats are a mutant strain known to be devoid of T-lymphocytes. These investigators found no significant difference in periradicular lesion size between the two groups, and they concluded that pathogenesis of periradicular lesions is not totally dependent on T-lymphocytes but is rather a "multifactorial phenomenon"

Sallay et al. 13 compared periodontal bone destruction in immnnosuppressed rats with that in normal rats. Cyclophosphamide (Cytoxan) was used to cause leukopenia in the experimental group. Cytoxan is an alkylating agent that disrupts nucleic acid function with resulting inhibition of rapidly proliferating cells. Lymphocyte and neutrophil counts are dramatically reduced shortly after Cytoxan injection. Sallay et al. found significantly more bone destruction in this group than in the control group. In a recent study, Yamasaki et a1.14 looked at the role of neutrophils in the development of periradicular lesions; they found that neutropenia after pulp exposure had no effect but that neutropenia before pulp exposure inhibited the development of the lesion.

The purpose of our study was to investigate the development of periradicular pathosis in immunosuppressed rats and compare it to that in conventional rats.

MAIFRIAI_ AND ME'I'HODS Thirty-five 8-week-old Sprague-Dawley rats were

randomly divided into seven groups of five. The rats in groups 1, 2, and 3 were given injections of Cytoxan as an immunosuppressive agent at the beginning of the experiment and at 7-day intervals throughout the study. The medication was given at a dose of 50 mg/kg of body weight by means of intraperitoneal injections. This dose caused ideal leukopenia without a rebound effect, as had been determined during a pilot project. The effectiveness of the immunosuppression was assessed by complete blood counts done each week on all rats in groups 2, 2, and 3. The rats in groups 4, 5, and 6 received no immunosuppressive therapy and served as treated controls. The five rats in group 7 were untreated controls and received no treatment at all.

Waterman et al. 721

At time 0, each rat in groups 1 through 6 was anesthetized by an intraperitoneal injection of sodium pento- barbital (5 mg/200 gm of body weight) and mounted on a jaw retraction board. Pulpal exposures were made through the occlusal surfaces of the mandibular first molars by means of a sterile #1/2 round bur in a high- speed handpiece. The exposures were left open to the oral flora for the duration of the experiment. All rats were housed similarly and received the same diet. Every 7 days, the rats in groups 1, 2, and 3 were anesthetized and blood was drawn from them through cardiac punc- tures. A complete blood count with a differential smear was performed on each blood sample (Professional Animal Labs, Tustin, Calif.). Fifty mg of Cytoxan per kg of body weight was then given. Whole blood was also drawn each week from rats in groups 4, 5, and 6 for comparison. Blood draws were also performed on the rats in group 7 throughout the experiment.

At 24 days, the rats in groups 1 and 4 were anesthetized as described. After blood samples were taken, these rats were perfused with a buffered glutaralde- hyde/formaldehyde fixative. The mandibles were dissected after removal of soft tissues and placed in 10% buffered formalin. Similar procedures were performed at 28 days on the animals in groups 2 and 5 and at 42 days on the animals in groups 3, 6, and 7.

After the distances among tube, object and film were standardized, a radiograph of each mandible half was made at 75 kVP and 10 impulses. The films were developed at the same time by means of a single automatic film processor. Radiographic analysis was performed with a Zeiss computer that was equipped with a digitizing board and a stylus. Images of the radiographs were projected onto the digitizing board; this permitted tracing of the observed periradicular radiolucencies with the stylus. Images were analyzed in random order; the examiner was unaware of which group was being traced at any given time. Three separate tracings were performed and averaged for the final analysis.

Block sections containing the mandibular molars and their surrounding tissues were decalcified in 20% formic acid. The block sections were then dehydrated and embedded in paraffin. Serial sections, each of which was 5 gm thick, were prepared and stained with hematoxylin and eosin. In addition, one section from each group was stained with a modified Gram stain for bacteria.

The histologic sections were examined at 32x with a microscope that was equipped with an ocular grid. Probe lines across the specimen were assessed for the intersection of various tissue components, Six parame- ters quantified were bone, root (including pulp tissue within the root), periodontal ligament, marrow spaces, inflammation (areas of inflammatory cell infiltrate), and soft tissue. This last component was defined as

722 Waterman etaL ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY June 1998

Circulating Leukocytes ¢EI.i,~UL m.OO0 £ , 1 ~ = ~ ) g ...................................................................

8

v 2WEEKS 4 W E E I ~ 6WEB(B

E:~ . . . . m m ~=,,m= m :

Fig. 1. Circulating leukocytes. Mean white blood cell counts of immunosuppressed, normal treated, and untreated controls were done at 2, 4, and 6 weeks; blood draws were performed weekly.

areas of soft tissue in which bone was observed in untreated controls and no inflammatory cell infiltrate was present. A seventh parameter, denoted "lesion," was also specified; it was defined as the additive areas of soft tissue along with areas of inflammation. This was done to allow comparison of radiographic and histologic data. The measures were made by means of the DataVoice (Support Technology, Grand Terrace, Calif.) computerized data collection and analysis system. This system uses a Zeiss camera lucida and allows visualization through the microscope of an LED on the cursor of a digitizing board. The intersections of the probe line with the six tissue types were identified by voice activation of the computer. The files created were decoded and analyzed to give relative percentages of the area of each parameter type.

The results of the radiographic analysis were compared through the use of a one-way analysis of vari- ance (ANOVA); the dependent variable was area and the independent variable was group. In our histologic analysis, the immunosuppressed and treated control groups were compared at each time interval for all para- meters through the use of a two-way ANOVA.

RESULTS The results of the complete blood counts are shown in

Fig. 1. There was significant leukopenia in experimental groups 1, 2, and 3 in comparison with control groups 4, 5, 6, and 7. The results of the radiographic analysis are summarized in Fig. 2. Rats in all six treatment groups had significantly larger areas of rarefaction than those in the untreated control group. When the treated control and immunosuppressed groups were compared at each period, bone loss at 4 weeks was significantly greater in the treated control rats than in the immunosuppressed

Radiographic Bone Loss /~F.A N~Z

1 . 4 "r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ++-~

.++J 1 . 2 5 . . . . . . . . . . . . . . . .

0.8 +g

0 ,2

0 2WEI~Y~ 4 ~ 6 ~

Fig. 2. Radiographic bone loss in immunosuppressed, normal treated, and untreated control rats at 2, 4, and 6 weeks

2 Weeks T i s s u e C o m p o n e n t Percentages

SOFT mFta~ B3NS ROOT PDL W!tmOt/

Fig. 3. Relative percentages of each tissue component for immunosuppressed, normal treated, and control groups at 2 weeks.

rats (p = 0.0025). No other significant differences were found in the radiographic analysis.

The results of the histologic analysis are displayed in Figs. 3, 4, and 5. The relative percentages of root and periodontal ligament were fairly constant in all groups. Comparison of groups at each period with respect to the various tissue components was done with a two-way ANOVA. This analysis showed that significantly more bone was present in group 7 (the untreated controls) than in any other group except group 4, the 2-week treated control group. At 2 weeks, there was a tendency for greater bone loss and increased lesion size in the immunosuppressed group in comparison with the treated control group; however, this was not statistically significant. There were no significant differences between the 4-week groups. A significant statistical difference was found between the amount of bone marrow in the two 6-week groups (p = 0.01). A Pearson's correlation coefficient was calculated to deter- mine the presence or absence of correlation between the


4 Weeks Tissue Component Percentages

Fig. 4. Relative percentages of each tissue component for immunosuppressed, normal treated, and control groups at 4 weeks.

Waterman et al. 723

6 Weeks Tissue Component Percentages

~ir=A % IIF.Atl8

!

|

1

~ Pot. ~ aoFr N m

Fig. 5, Relative percentages of each tissue component for immunosuppressed, normal treated, and control groups at 6 weeks.

radiographic and the histologic data. We found a strong correlation between radiographic bone loss and the histologic "lesion" in all animals (r = 0.77, p < 0.0001).

Because all animals had been purchased from the same supplier and housed similarly, it was assumed that their bacterial oral flora were similar. The sections stained with a modified Gram stain showed bacterial presence in all necrotic root canal spaces and near the apex. No evidence of bacterial growth in the periradicular tissues was noted--not even in the immunosuppressed groups.

DISCUSSION The immunosuppressive effect of Cytoxan has been

studied extensively by Host) 5 It is an alkylating agent related to the nitrogen mustards, which disrupt nucleic acid function with resulting inhibition of rapidly proliferating cells. Use of Cytoxan causes an initial depression in lymphocyte counts; this is followed by slowly increasing levels. Control levels are usually reached after 2 weeks. In addition, a marked reduction in B-lymphocyte activity is seen with Cytoxan) 6 This effect seems to be limited to the B cells, with no inhibitory effects observed in T- lymphocyte activities. The effect on granulocytes is more dramatic, with a severe granulocytopenia followed by a more rapid recovery in approximately 1 week. This phenomenon is a dose-dependent response. Cawley et al.t7 found peripheral blood leukocyte counts to be the best index for determining resistance to infection in indi- viduals with leukopenia.

AS seen in Fig. 1, a significant reduction in the circulating leukocyte levels was obtained in the immunosuppressed groups of this study. The dosages used in this experiment are equivalent to those used in treatment of human lymphomas or leukemia. The weight of each rat was monitored each week; all rats increased in weight during the experiment.

The results of the histologic analysis suggest that possessing a normal number of leukocytes is not a

significant factor in the development of periradicular lesions in the period of this study. The statistical analyses specifically compared differences in areas of bone, inflammatory infiltrate, and pefiradicular lesion size. Fig. 6 shows examples of the histologic specimens used for analysis in each treatment group.

The mean areas of bone and lesion were comparable in the 4-week samples (Fig. 4). One immunosuppressed animal showed severe acute inflammation and a large periradicular lesion. On the basis of our results, it appears that leukopenic animals are still able to mount an acute response. The total white blood cell count for this animal was comparable to that for the rest of the immunosuppressed animals.

Fig. 5 shows that there were no differences between the 6-week groups except for a significant reduction in marrow spaces in the suppressed animals. This observation was to be expected after 6 weeks of immunosuppression by Cytoxan and is in agreement with the findings of Host. 15

A possible explanation for the results of this study may be that the host possesses a large reserve of leukocytes. Even with significantly reduced numbers of circulating leukocytes, there may still exist a supply that is adequate to keep the invading bacteria under control. The response of an immunocompetent host may normally be excessive, and a reduction in the numbers of inflammatory cells may have little effect initially. Over an extended period, the host may be unable to contain the infection; this could be determined by longer experimental times. Another factor could be that T-lymphocytes, while decreased in number in the immunosuppressed group, may retain sufficient activity and function to be effective, as was the case in the control animals.

High levels of noncellular inflammatory mediators, such as prostaglandins, leukotrienes, and cytokines, have been demonstrated in periradicular lesions, is-2°

7.?4 Waterman et aL ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY June 1998

Fig. 6. Examples of histologic samples used for data analysis. Note lesion size and inflammation in each. A, Group 1, immunosuppressed at 2 weeks. B, Group 2, immunosuppressed at 4 weeks, C, Group 3, immunosuppressed at 6 weeks. D, Group 4, treated control at 2 weeks. E, Group 5, treated control at 4 weeks. F, Group 6, treated control at 6 weeks. G, Group 7, untreated control.


These mediators possibly play a significant role in the

development of per i radicular pathosis even with a reduction in the number of leukocytes. Leukocytes are the main sources of product ion for some of these substances. However, they are also produced by other cells. For example, interleukin-1 [3 has extreme bone

resorption activity and, although produced primarily by macrophages, is also produced by many other cells. 21

We found a high correlation between the radiographic

and histologic data in this experiment (p < 0.0001). Animals showing radiographic periradicular bone loss also had comparable histologic " les ions" At 4 weeks, the treated control rats with normal leukocyte levels had

significantly more bone resorption than the immunosuppressed animals (Fig. 2). The same phenomenon was not seen histologically. A possible explanation may be that one of the rats in the 4-week immunosuppressed group showed

an extensive acute apical reaction, with resulting bone loss that was much greater than that seen in any other animal.

Inasmuch as there were only five rats in each group, this large variation may have affected the statistical results.

The lineal analysis performed on the histologic sections with the microscope/computer combination allowed for careful analysis of each tissue component. This method of histologic analysis has been demonstrated to be a effective procedure for histologic analysis. 22 After data analysis, a

comparison between the treated animals and the untreated controls showed dramatic alveolar bone desmaction in the periradicular regions of the treatment groups.

The results of this study showed no difference in the size

of developing periradicular lesions between normal rats and immunosuppressed rats at the same time interval. After placement of ligature wires, Sallay et al.13 found

significantly more periodontal bone loss in immunosup-

pressed rats. These investigators ran their experiment for only 14 days. The pattern of bacterial invasion in an exposed pulp may not be the same as that in a periodontal

pocket. This might be due to differences in flora or other conditions between infected periodontal sulci and infected

root canals. It is well established that an immunocompro- raised host exhibits an enhanced susceptibility to infections of opportunistic microbes. In our experiment the suppressed animals had reduced numbers of leukocytes and had been exposed only to bacterial contamination at the pulp exposure sites. Infections of greater severity might have resulted in more serious consequences. The recent study by Yamasaki et al.14 compared periradicular lesion development in neutropenic rats with that in normal rats. These investigators found that if rats were neutropenic before mechanical pulp exposure, there was a resulting inhibition of lesion development, but that neutropenia commencing at the same time as pulp exposure had no effect on lesion development. This is in agreement with the results of our investigation.

WaWrman et al. 725

REFERENCES 1. Torabinejad M, Eby WC, Naidorf IJ. Inflammatory and

immunological aspects of the pathogenesis of human periapical lesions. Journal of Endodontics 1985; 11:479-88.

2. Kakehashi S, Stanley HR, Fitzgerald RJ. The effects of surgical exposures of dental pulps in germ-free and conventional rats. Oral Surg Oral Med Oral Patbol 1965;20:340-9.

3. Moller AJR, Fabricius L, Dahlen G, Ohman AE, Heyden G. Influence on periapical tissues of indigenous oral bacteria and necrotic pulp tissue: an experimental study in monkeys. Scandinavian Journal of Dental Research 1981;89:475-84.

4. Langeland K, Block RM, Grossman LI. A histopathological and histobacterial study of 35 periapical endodontic surgical specimens. J Endod 1977;3:8-22.

5. Tronstadt L, Barnett F, Riso K, Slots J. Extraradicular endodontic infections. Endodontic and Dental Traumatology 1987;3:86-90.

6. Stern M, Dreizen S, Mackler B, Levy BM. Quantitative analysis of cellular composition of human periapical granulomas. J Endod 1981;7:117-22.

7. Stern M, Dreizen S, Mackler B, Selbst A, Levy BM. Isolation and characterization of inflammatory cells from the human periapical granuloma. J Dent Res 1982;61:1408-12.

8. Kontiainen S, Ranta H, Lautenschlager I. Cells infiltrating human periapical inflammatory lesions. J Oral Pathol 1986;15:544-6.

9. Yu S, Stashenko E Identification of inflammatory cells in developing rat periapical lesions. J Endod 1987;13:535-40.

10. Stashenko P, Yu S. T-helper and T-suppressor cell reversal during the development of induced rat periapical lesions. J Dent Res 1989;68:830-4.

11. Torabinejad M, Kettering J. Identification and relative concentration of B and T lymphocytes in human chronic periapical lesions. J Endod 1985;11:122-5.

12. Wallstrom JB, Torabinejad M, Kettefing JD, McMillan E The role of T cells in the pathogenesis of periapical lesions. Oral Surg Oral Med Oral Pathol 1993;76:213-8.

13. Sallay K, Sanavi E Ring I, Pham U, Behling UH, Nowotny A. Alveolar bone destruction in the immunosuppressed rat. J Periodontal Res 1982;17:263-74.

14. Yamasaki M, Kumazawa M, Kohsaka T, Nakamura H. Effect of methotrexate-induced neutropenia on rat periapical lesion. Oral Surg Oral Med Oral Pathol 1994;77:655-61.

15. Host H. Comparative effects of cyclophosphamide, nitrogen mustard and total body irradiation on survival and on white blood ceils in rats. Radiat Res 1966;27:638-51.

16. Cupps TR, Edgar LC, Fauci AS. Suppression of human B lymphocyte function by cyclophosphamide. J Immunol 1982; 128:2453-7.

17. Cawley S, Findon G, Miller TE. Peripheral blood leukocyte count as an index of defense status in the leukopenic host. J Lab Clin Med 1988;112:52-7.

18. McNicholas S, Torabinejad M, Blankenship J, Baldand L. The concentration of prostaglandin E 2 in human periradicular lesions. J Endod 1990;17:97-100.

19. Torabinejad M, Cotti E, Jung T. Concentrations of leukotriene B 4 in symptomatic and asymptomatic periapical lesions. J Endod 1992;18:205-8.

20. Stashenko R The role of immune cytokines in the pathogenesis of periapical lesions. Endod Dent Traumatol 1990;6:89-96.

21. Oppenheim J, Kovacs E, Matsushima K, Durum S. There is more than one interleukin-1. Immunol Today 1986;7:45-56.

22. McMillan PJ, Harris JG. DataVoice: a microcomputer-based general purpose voice-controlled data-collection system. Comput Biol Med 1990;20:415-9.

Reprint requests: Phillip A. Waterman, Jr., DDS, MS 3748 Montgomery Drive Santa Rosa, CA 95405

Documents

Development of periradicular lesions in immunosuppressed rats