Basic ResearchTechnology
A Comparative Study of the Biocompatibility of Three
Root-end Filling Materials in Rat Connective Tissue
Shahriar Shahi, DDS, MSc, Saeed Rahimi, DDS, MSc, Mehrdad Lotfi, DDS, MSc,
Hamid Reza Yavari, DDS, MSc and Ali Reza Gaderian, DDS, MSc
Abstract
The purpose of this study was to compare the biocom-
patibility of amalgam, gray MTA and white MTA in the
connective tissue of rats. We used 45 Sprague-Dawley
rats in this study. The rats were divided into three
groups. Root end filling materials were placed in poly-
ethylene tubes and inserted into the rats connective
tissue through incisions. The rats were sacrificed after 3
days, 1 wk, and 3 wk, respectively. Histologic samples
were sectioned in 5-m thicknesses and stained with
hematoxylin and eosin. Kruskal-Wallis test was used for
statistical analysis. The results showed that after 3
days, white MTA was more biocompatible than gray
MTA and amalgam. After 1 week, gray MTA was more
biocompatible than white MTA and Amalgam. After 3
wk, there were no significant differences between ex-
perimental groups and the control group. (J Endod
2006;32:776 780)
Key Words
Biocompatibility, endodontic surgery, MTA, root-end
filling material
From the Department of Endodontics, Tabriz Faculty of
Dentistry, Azarbaidjan Shargi, Iran.
Address requests for reprints to Dr. Shahriar Shahi, DDS,
MSC, Department of Endodontics, Tabriz Faculty of Dentistry,
Golgasht Street, Tabriz, Azarbaidjan Shargi 5166614713, Iran.
E-mail address: shahriar_shahi@yahoo.com.
0099-2399/$0 - see front matter
Copyright 2006 by the American Association of
Endodontists.
doi:10.1016/j.joen.2006.01.014
776 Shahi et al.
S everal retrospective studies have addressed the outcome of conventional, nonsur-
gical root canal therapy, and the success rates reported have ranged between 53 %
(1) and 95 % (2). However, even if 90% of all endodontic treatment is successful over
time, the reciprocal failure rate is still 10% (3). Therefore, a certain number of failures
can be expected, probably because of the persistence of bacteria and their byproducts
in the root canal system. Although many endodontic failures can be successfully re-
treated by an orthograde approach (4), root-end surgery may be the only alternative to
extraction in cases where the orthograde approach is not feasible because of the pres-
ence of posts and other permanent restorative materials in the coronal aspect of the root
canals (5).
The ultimate success of the surgery depends on the regeneration of a functional
periodontal attachment apparatus, including cementum overlying the resected root-end
surface, periodontal ligament, and alveolar bone (6). This would occur more predict-
ably when the root canal that is exposed after root end resection is filled with a material
that not only seals the canal to prevent egress of any remaining bacteria or their
by-products, but also allows for the formation of a normal periodontium across its
exterior surface (5).
Amalgam had been the most popular and widely used retrograde filling material
since the last century (7). It is easy to manipulate, readily available, well tolerated by soft
tissues, radiopaque, and initially provides a tight apical seal (7). However, its disad-
vantages are significant, including potential for mercury and other ions release, slow
setting, dimensional unstability, eventual leak from corrosion, and staining the overly-
ing soft tissues causing a tattoo (8, 9). A gray mineral trioxide aggregate (MTA) was
developed at Loma Linda University in 1993 as a root-end filling material and for repair
of lateral and furcal perforations (9, 10).
Gray MTA powder consists of fine hydrophilic particles. The principle compounds
present in this material are tricalcium silicate, tricalcium aluminate, tricalcium oxide,
and silicate oxide (7). In addition, small amounts of other mineral oxides exist that are
responsible for the chemical and physical properties of this aggregate (7).The biocom-
patibility of gray MTA has been reported using cell culture techniques and connective
tissue reactions (1113). Sarkar et al. reported the propensity of MTA to release Ca and
its ability to form hydroxyapatite and concluded that the sealing ability, biocompatibility
and dentinogenic activity of MTA is attributed to these physicochemical reactions (14).
White MTA, a new type of MTA has recently been introduced to the profession and, as a
consequence, only limited research has been carried out on the properties of this new
material (1517). It was reported that the behavior of osteoblasts is very different in
contact with the surface of white MTA compared with gray MTA (16). The aim of this
study was to histopathologically examine the biocompatibility of gray MTA, white MTA
and high-copper amalgam by implanting them into the subcutaneous connective tissue
of rats.
Materials and Methods
We used 45 male, 2 to 3 month old Sprague-Dawley rats weighing 250 20 g. All
ethical and humane criteria contained in Helsinky declaration, and all recommended
points by Institutional Animal Care and Use Committee (IACUC), in care and use of
laboratory animals were observed in all the different stages of the project. The following
materials were investigated: Group A, tooth colored mineral trioxide aggregate (white
JOE Volume 32, Number 8, August 2006

MTA) (Dentsply; Tulsa Dental, USA); group B, pro-root mineral trioxide
aggregate (gray MTA) (Dentsply); group C, high-copper amalgam (Si-
naalloy, Faghihi, Iran); group D, control group.
Rats were anesthetized with diethyl-ether. The dorsal skins of the
animals were shaved under ether anesthesia and disinfected with 5%
iodine solution. Four incisions were made on the back of each rat.
Freshly mixed test materials were placed in clean, sterile polyeth-
ylene tubes (Eastern Medikit; Haryana, India) with 0.8-mm inner diam-
eter and 8-mm length and inserted into the connective tissues. Gray
MTA, white MTA and amalgam were prepared according to the recom-
mendations of the manufacturers and were applied with a plastic and an
amalgam carrier, respectively. One polyethylene tube was placed in
each rat as control. Four incisions were made on the backs of the
animals over a length of 2-cm using no. 15 blade (Romed, Holland) in
a head-to-tail direction. The skin was reflected, and the implantation
materials were inserted into spaces created by blunt dissection. To
prevent interactions of materials, the tubes were placed at least 2 cm
away from each other. The skin was closed with 3/0 silk sutures. The
evaluations were made 3, 7, and 21 days after surgical implantation. In
each examination period 15 animals were sacrificed by administering
high doses of anesthetics. The dorsal skin was shaved, and the tubes
were excised together with connective tissues around them. The speci-
mens were kept in 10% formalin solution (37% formaldehyde; Merck
Darmstadt, Germany) for 2 wk. Sections of 5 m thickness were taken
from specimens placed in paraffin blocks and stained with hematoxylin
and eosin (H&E). Tissue sections were made longitudinally through the
midline of the tubes and four sections from each specimen were se-
lected. Histological evaluations were made under a light microscope
(Carl Zeiss, Oberkachen, Germany) at 100, 200, and 400 mag-
nifications. The observer was blinded to the procedure. Evaluation of
inflammatory reactions was performed according to Cox and Robins
criteria (18, 19): accumulation of acute and chronic inflammatory
cells, fibrin deposits, tissue edema, and vascular congestion.
The inflammatory reactions were classified: Grade I, scattered
chronic inflammatory cells; grade II, infiltration of inflammatory cells,
and wavy collagen fiber deposits and fibrosis; grade III, dense infiltra-
tion of inflammatory cells, limited areas of tissue edema and vascular
congestion; grade IV, very dense infiltration of acute and chronic in-
flammatory cells, widespread edematous areas and vascular congestion
along with fibrin deposits.
Statistical analysis was performed using Kruskal-Wallis analysis for
comparing the grades between groups and two-way ANOVA for compar-
ing the severity of inflammation at intervals and for materials. Statistical
significance was defined as p 0.05.
Results
Histologic Findings
The 3-Day Old Specimens
In the gray MTA group, 16.7% of the specimens demonstrated
grade II inflammation, which consisted of the infiltration of chronic
inflammatory cells, wavy collagen fiber deposits and fibrosis. Fifty per-
cent of the specimens demonstrated grade III inflammation (Fig. 1),
which consisted of a dense infiltration of inflammatory cells, edematous
areas, and vascular congestion. Thirty-three percent of the specimens
demonstrated grade IV inflammation, which consisted of a very dense
infiltration of acute and chronic inflammatory cells, widespread edem-
atous areas, and vascular congestion along with fibrin deposits.
In the white MTA group, 16.7% of the specimens had grade I
inflammation, which consisted of sporadic infiltration of chronic in-
flammatory cells without tissue edema. 58.3% of the specimens had
grade II (Fig. 1) and 25% had grade III inflammation.
JOE Volume 32, Number 8, August 2006
Basic ResearchTechnology
In the amalgam group, 16.7% of the specimens revealed grade II,
33.3% revealed grade III, and 50% revealed grade IV inflammation
(Fig. 1).
In the control group, 21.4% of the specimens had grade II (Fig. 1),
50% had grade III, and 28.6% had grade IV inflammation.
The 7-Day Old Specimens
In the gray MTA group, 30.8% of the specimens had grade II,
46.1% had grade III (Fig. 1), and 23.1% had grade IV inflammation.
In the white MTA group, 12.5% of the specimens revealed grade II
inflammation (Fig. 1), 37.5% revealed grade III, and 50% revealed
grade IV inflammation.
In the amalgam group, 16.7% of the specimens demonstrated
grade III and 83.3% demonstrated grade IV inflammation (Fig. 1).
In the control group, 53% of the specimens had grade II (Fig. 1)
and 47% had grade IV inflammation.
The 21-Day Old Specimens
In the gray MTA group, 25% of the specimens had grade I and 75%
had grade II inflammation (Fig. 1). In the white MTA group 62.5% of the
specimens showed grade I and 37.5% showed grade II inflammation.
In the amalgam group, 33.3% of the specimens revealed grade I,
26.6% revealed grade II, 26.6% revealed grade III, and 13.3% revealed
grade IV inflammation (Fig. 1).
In the control group, 66.7% of the specimens had grade I and 33.3
% had grade II inflammation (Fig. 1).
On the whole, an overwhelming majority of the specimens had
predominantly chronic inflammatory cells and none of the specimens
had predominantly acute inflammatory cells.
Statistical Analysis
In 3-day old specimens the mean inflammation grades were 3.33
for the amalgam group, 2.08 for white MTA group, 3.17 for gray MTA
group, and 3.07 for control group, demonstrating a statistically signif-
icant difference in the mean inflammation grades among the groups (p
0.002).
In 7-day old specimens the mean inflammation grades were 3.83
for the amalgam group, 3.75 for white MTA group, 2.92 for gray MTA
group, and 2.47 for the control group, demonstrating a statistically
significant difference in the mean inflammation grades among the
groups (p 0.001).
In 21-day old specimens the mean inflammation grades were 2.2
for the amalgam group, 1.67 for gray MTA group, 1.37 for white MTA
group, and 1.7 for the control group, demonstrating no statistically
significant difference among the groups (p 0.60) (Fig. 2).
Discussion
It is obvious that materials contacting live tissues may be irritating
to the tissue. What is important is the irritating capacity of the material
and the duration of the irritating effect of the material in the neighboring
tissues. At present, there are four classic methods for the evaluation of
the biocompatibility of materials. These methods are: (a) Cytotoxic
evaluation, (b) subcutaneous implants, (c) intra-osseous implants, and
(d) the in vivo evaluation of the peri-radicular tissue reaction in human
subjects.
The method we used in the present study was subcutaneous im-
plants method, in which the materials to be evaluated were placed in
polyethylene tubes that were implanted in the subcutaneous connective
tissue in rats. This method is one of the most common methods for the
evaluation of the biocompatibility of dental materials, which was intro-
duced by Torneck et al. in 1966 (20). The efficacy of this method was
Biocompatibility of Three Root-End Filling Materials 777
Basic ResearchTechnology

Figure 1. Histologic images. (A) 3-day GMTA. Dense infiltration of lymphocyte and plasma cell (arrow), tissue edema; 200 magnification; grade III. (B) 3-day
WMTA. Lymphocyte and plasma cell (arrow), wavy collagen fiber deposition (arrow head); 200 magnification; grade II. (C) 3-day amalgam. Fibrin deposition
(arrow head), tissue edema (black arrow), chronic inflammatory cell (white arrow); 200 magnification; grade IV. (D) 3-day control. Wavy collagen fiber
deposition (arrow head); 200 magnification; grade II. (E) 7-day GMTA. Dense chronic inflammatory cells and tissue edema (arrow); 200 magnification; grade
III. (F) 7-day WMTA. Wavy collagen fiber deposition, chronic inflammatory cells (white arrow); 200 magnification; grade II. (G) 7-day amalgam. Tissue edema
(arrow head), lymphocyte and plasma cell, fibrin deposition (white arrow); 400 magnification; grade IV. (H) 7-day control. Lymphocyte and plasma cell, wavy
collagen fiber deposition (white arrow); 200 magnification; grade II. (I) 21-day GMTA. Lymphocyte and plasma cell (arrow head), collagen fiber deposition
(arrow); 200 magnification; grade II. (J) 21-day WMTA. Scattered chronic inflammatory cells (arrow); 200 magnification; grade I. (K) 21-day amalgam.
Macrophage (lightened arrow), tissue edema (arrow head), lymphocyte (white arrow); 400 magnification; grade IV. (L) 21-day control. Wavy collagen fiber
deposition (black arrow), lymphocyte and plasma cell (white arrow); 100 magnification; grade I.

re-evaluated, elaborated and substantiated by Olsson et al. in 1981
(21). According to Olsson et al. the placement of the material to be
evaluated in the terminal portions of the polyethylene tubes prevents the
diffusion of the material into the connective tissue, simulating the situ-
ation in the root canal and, therefore, is more advantageous than the
Figure 2. The mean inflammation grades in test periods and materials.
method in which the material is directly injected into the subcutaneous
connective tissue.
In different studies using this technique, different methods have
been used to compare the microscopic sections in relation to the se-
verity of inflammation around the tubes containing the materials (12,
20 22). One of these methods is based on Standford criteria (20, 22).
In this method, the number of infiltrated inflammatory cells in different
parts of the microscopic sections is counted. Another method has been
proposed by Cox et al., in which the density of inflammatory cells (not
their number) along with the response of the tissues to the inflammatory
process, such as necrosis, is used for grading the inflammation (18).
The method we used in the current study was Cox method with an
emphasis on the histologic basis of inflammation. According to this
method, the density of inflammatory cells, the presence of tissue re-
sponse such as fibrosis, and vascular response such as congestion and
fibrin extravasation were used to determine the intensity and the chro-
nologic sequence of the inflammatory response (18, 19).
Because connective tissue inflammatory response has different
aspects, focusing the interpretation only on the number of inflammatory
cells cannot encompass all the aspects of connective tissue response
and it seems to be less comprehensive. Therefore, attention to the vas-

778 Shahi et al. JOE Volume 32, Number 8, August 2006


cular and reparative responses is a more precise and also an innovative
method for the comparison of the severity of the inflammation around
the tubes containing the material(s). A study carried out by Hong et al.
demonstrated that amalgam leads to more severe inflammation and
bone resorption compared to MTA when they are used for perforation
repair (23). Furthermore, Christopher et al. compared the reaction of
the periapical tissues in dogs to root end filling amalgam and MTA. The
study revealed that the inflammatory response around amalgam is more
severe than that around MTA (24). Another study carried out by Tor-
abinejad et al. (1997) revealed that compared to amalgam, MTA as a
retrofilling material in the incisors of monkeys produces much less
inflammation (25). Another study was carried out by Torabinejad on the
periapical tissues in dogs. In that study the response of periapical tissues
in dogs to amalgam and MTA as retrofilling materials was evaluated. The
results indicated that in the tissues adjacent to amalgam moderate to
severe inflammation was evident in all the specimens but in MTA spec-
imens mild inflammation was present in two-thirds of the specimens
and moderate inflammation was present in one-third of the specimens
(5).
The results of our study indicated that after 3 and 7 days, amalgam
produces a more severe inflammatory response compared to gray MTA
and white MTA. After 3 days the inflammatory response around the tubes
containing gray MTA was more severe than that around the white MTA-
containing tubes but after a week the situation was vice versa. In other
words, the inflammatory response was more severe around the tubes
containing white MTA compared to that around gray MTA-containing
tubes. After 3 wk, the severity of the inflammatory response around all
the tubes was the same and equal to the control group. In other words,
there was no statistically significant difference between the groups (p
0.05). These results indicate that in the conditions of our study the
biocompatibility of the two kinds of MTA after 3 days and 1 wk is better
than amalgam. Our findings are consistent with the findings of the pre-
viously mentioned studies.
The results of our study differ from the results of a study by Yaltirik
et al. (2004) on the response of rat connective tissue to polyethylene
tubes containing MTA and amalgam (12). The difference of the results
between these two studies carried out in the same manner may be
attributable to the number of specimens in each group (five specimens
in that study compared to 15 specimens in our study). Furthermore, the
difference in the way microscopic sections have been evaluated may
have contributed to the difference in results. In the study carried out by
Yaltirik et al., the system of inflammation grading was based on the
number of inflammatory cells and the presence of a fibrotic capsule in
the sections was noted but was not taken into account in the inflamma-
tion grading system and also in the statistical analysis (12). In our
method for grading the severity of the inflammation the presence of
fibrosis, vascular reactions, and repair process were taken into ac-
count. In our study, after 3 days the tissue response around the tubes
containing amalgam was more severe than that around gray MTA and
white MTA, which was statistically significant. Different reasons have
been proposed to explain the tissue irritability of amalgam. According to
Omnel, the tissue destruction around retrofilled amalgam is because of
the toxic products of amalgam electrolysis, such as zinc carbonate (26).
Torabinejad et al. demonstrated that MTA releases calcium phos-
phate and calcium oxide after it is exposed to moisture. The released
calcium oxide produces calcium hydroxide after reacting with tissue
fluids. Calcium hydroxide is a tissue irritant, giving rise to inflammation
(27, 12).
The increase in the severity of inflammation for white MTA from
day 3 to day 7 may indicate the gradual release of calcium hydroxide and
its gradual reaction with tissue moisture and as a result, the production
of calcium hydroxide between days 3 and 7. For gray MTA the process of
JOE Volume 32, Number 8, August 2006
Basic ResearchTechnology
inflammation induction is different. In this case, the most severe inflam-
mation is found in the specimens after 3 days, which decreases gradu-
ally. This difference may indicate the difference in the rate of chemical
reactions leading to calcium hydroxide release. Another reason for the
difference can be attributed to the different chemical composition of the
two types of MTA. According to a study carried out by Asgari et al.
(2005), the most significant differences were the measured concentra-
tion of Al2O3 (122%), MgO (130%), and especially FeO
(1000%) when gray MTA was compared to white MTA (28). Differ-
ences in the FeO concentration are thought to be primarily responsible
for the variation in color of white MTA in comparison with gray MTA.
FeO may induce inflammation in the first 3 days in gray MTA case but
after a week a decrease in the amount of iron oxide and its elimination
from the environment during the inflammatory process may lead to a
decrease in inflammation severity. It is evident that more research stud-
ies are required to substantiate these theories. We suggest further stud-
ies on tooth-colored MTA, especially long-term animal studies on tissue
reactions.
References
1. Pekruhn RB. The incidence of failure following single-visit endodontic therapy. J
Endod 1986;12:68 72.
2. Jokinen MA, Kotilainen R, Poikkeus P, Poikkeus R, Sorkki L. Clinical and radio-
graphic study pulpectomy and root canal therapy. Scand J Dent Res 1978;86:36373.
3. Ruddle CJ. Nonsurgical endodontic Retreatment. In: Cohen S, Burns RC, eds. Path-
ways of the pulp, 8th ed. St. Louis: CV Mosby, 2002:874.
4. Bergenholtz G, Lekholm U, Milthon R. Retreatment of endododontic filling. Scand J
Dent Res 1979;87:21724.
5. Keiser K, Johnson C, Tipon DA. Cytotoxicity of mineral trioxide aggregate using
human periodontal ligament fibroblasts. J Endod 2000;26:288 91.
6. Andreasen JO, Rud J. Modes of healing histologically after endodontic surgery in 70
cases. Int J Oral Surg 1972;1:148 60.
7. Kim S. Endodontic microsurgery. In: Cohen S, Burns RC, eds. Pathways of the pulp,
8th ed. St. Louis: CV Mosby 2002:683725.
8. Gartner AH, Dorn SO. Advances in endodontic surgery. Dent Clin North Am
1992;36:357 65.
9. Lee SJ, Monsef M, Torabinejad M. Sealing ability of mineral trioxide aggregate for
repair of lateral root perforations. J Endod 1993;19:541 4.
10. Torabinejad M, Watson TF, Pitt Ford TR. Sealing ability of a mineral trioxide aggregate
when used as a root end filling material. J Endod 1993;19:5915.
11. Torabinejad M, Hong CU, Pitt Ford TR, Kettering JD. Cytotoxicity of four root end
filling materials. J Endod 1995;21:489 92.
12. Yaltirik M, Ozbas H, Bilgic B, Issever H. Reactions of connective tissue to mineral
trioxide aggregate and amalgam. J Endod 2004;30:959.
13. Riberio DA, Duarte MAH, Matsumoto MA, Marques MEA, Salvadori DMF. Biocom-
patibility in vitro tests of mineral trioxide aggregate and regular and white Portland
cements. J Endod 2005;31:6057.
14. Sarkar NK, Caicedo R, Ritwik P, Moiseyeva R, Kavashima I. Physicochemical basis of
the biologic properties of mineral trioxide aggragate. J Endod 2005;31:97100.
15. Holland R, Souza V, Nery MJ, Faraco IM Jr, Bernabe PF, Otoboni Filho JA, Dezan EJR.
Reaction of rat connective tissue to implanted dentine tubes filled with a white min-
eral trioxide aggregate. Braz Dent J 2002;13:23 6.
16. Perez AL, Spears R, Gutmann JL, Opperman LA. Osteoblasts and MG-63 osteosarcoma
cells behave differently when in contact with pro root MTA and white MTA. Int Endod
J 2003;36:564 70.
17. Mah T, Basrani B, Santos JM, et al. Periapical inflammation affecting coronally-
inoculated dog teeth with root fillings augmented by white MTA orifice plugs. J Endod
2003;29:442 6.
18. Cox CF, Subay RK, Suzuki S, Ostro E. Biocompatibility of various dental materials,
pulp healing with a surface seal. Int J Periodontics Restorative Dent 1996;16:24151.
19. Robbins SL, Cortan RS, Kumar V. Pocket compaction to Robbins pathologic basis of
diseases, 6th ed. Philadelphia: WB Saunders, 1991:46 56.
20. Torneck CD. The reaction of rat connective tissue to polyethylene tube implants. Oral
Surg 1966;21:379 87.
21. Olsson B, Slikowski A, Langland. Subcutaneous implantation for the biological eval-
uation of endodontic materials. J Endod 1981;7:355 69.
22. Ozbas H, Yaltirik M, Bilgic B, Issever H. Reaction of connective tissue to compomers,
composite and amalgam root end filling materials. Int Endod J 2003;36:2817.
Biocompatibility of Three Root-End Filling Materials 779
Basic ResearchTechnology
23. Hong CU, Mackenedy DJ, Pitt Ford TR, Torabinejad M. Healing of furcal lesions
repaired by amalgam or MTA. J Endod 1997;23:197200.
24. Christopher F, Carnes L, Carlos E, Rio E. Longitudinal sealing ability of MTA as a root
end filling material. J Endod 1996;22:575 8.
25. Torabinejad M, Pitt Ford TR. Histologic assessment of mineral trioxide aggregate as
a root end filling material in monkeys. J Endod 1997;23:225 8.
780 Shahi et al.
26. Omnell KA. Electrolytic precipitation of zinc carbonate in the jaw, an unusual com-
plication after root resection. Oral Surg 1959;12:848 52.
27. Torabinejad M, Hong CU, MC Donald F, Pitt Ford TR. Physical and chemical proper-
ties of a new root end filling material. J Endod 1995;21:349 53.
28. Asgary S, Parirokh M, Egbbal MJ, Brink F. Chemical differences between white and
gray mineral trioxide aggregate. J Endod 2005;31:1013.
JOE Volume 32, Number 8, August 2006