Journal

of
Dentistry
Journal of Dentistry 26 (1998) 585589

Decoloration of 1% methylene blue solution in contact with dental filling

materials
M.-K. Wu a,*, E.G. Kontakiotis b, P.R. Wesselink

Department of Cariology Endodontology Pedodontology, Academic Centre for Dentistry Amsterdam (ACTA), Amsterdam, The Netherlands
b
Department of Endodontics, Dental School, University of Athens, Athens, Greece
Received 24 February 1997; accepted 16 April 1997

Abstract
Objectives: Leakage studies have been performed frequently, since a tight seal provided by various dental fillings has been considered
clinically important. The dye penetration experiment using a methylene blue solution as a tracer is one of the most common methods applied
in these types of studies. The stability of the colour of methylene blue in contact with six dental filling materials was observed.
Methods: Silicon rubber tubes and human tooth roots of 10 mm in length and 1.5 mm inner diameter were filled with amalgam, calcium
hydroxide, Cavit, Fuji II, mineral trioxide aggregate, or zinc oxide eugenol, 10 tubes or roots for each material. Groups of five tubes or roots
filled with the same material were immersed in 0.8 ml 1% methylene blue dye solution. The optical density of the methylene blue solution
before immersion and after 24, 48 and 72 h of immersion was measured in a spectrophotometer at 596 nm.
Results: The methylene blue solution was found to be decoloured over time by all the test materials (P ! 0.01) except for Fuji II, in both
silicone tubes and roots. At 24 h, the optical density value of methylene blue decreased by 73% for the Ca(OH) 2/silicone group and 84% for
the mineral trioxide aggregate/silicone group.
Conclusion: Methylene blue is decoloured by some dental filling materials, which may result in unreliable results for these materials in dye
leakage studies. ! 1998 Elsevier Science Ltd. All rights reserved.
Keywords: Leakage; Methylene blue; Decoloration

1. Introduction
It is considered to be of great clinical importance that a
tight seal is provided by all types of dental fillings, including
coronal temporary and permanent dental restorations, rootcanal fillings and root-end fillings [16]. Therefore, leakage
studies on various dental fillings comprise an important part
of contemporary dental research.
A number of leakage models have been described in the
literature, including penetration of tracers (dyes, radioisotopes or micro-organisms), the electrochemical technique[7] in which the electrical current passing through the
flaws around the filling was measured, the air pressure
device[8] by which the volume of air passing along dental
fillings in ceramic discs was measured, and the fluid
transport models[9,10] in which fluid passing along dental
* Correspondence should be addressed to: M-K. Wu, Department of
Cariology Endodontology Pedodontology, ACTA Louwesweg 1, 1066
EA Amsterdam, The Netherlands. Tel: +31 20 5188367; Telefax: +31 20
6692881; e-mail: M.Wu@acta.nl

0300-5712/98/$19.00 ! 1998 Elsevier Science Ltd. All rights reserved.

PII: S0 30 0 -5 7 12 ( 97 ) 00 0 40 - 7

fillings under a headspace pressure was measured. Although

no technique has been universally accepted, dye penetration
using a 0.22.0% methylene blue (MB) solution as a tracer
is used most commonly in these leakage studies [11].
In dye penetration experiments, one end of the filled
specimen is immersed in a dye solution. Longitudinally
splitting or cross sectioning the filled specimen at the end
of the immersion period allows the length of coloured dye
penetration along the tooth-filling junction to be measured.
However, during the experiment the dye may be decoloured by contact with the test materials, depending
upon their chemical properties. Consequently, penetration
of the decoloured dye solution along the filling may not be
seen, making the recorded length of dye penetration
unreliable. In spite of the high number of dye penetration
publications, whether the dye is decoloured by the test
material has not been studied.
Whether the dye which has penetrated into the voids
along the tooth-filling junction can be seen and measured
depends upon the extent of decoloration of the dye solution
in the voids. This will be determined by, among other

586

M.-K. Wu et al./Journal of Dentistry 26 (1998) 585589

Table 1
Decoloration of methylene blue (MB) during immersion of plastic tubes filled with various materials
Mean " s.d. optical density of 1% MB before immersion (0 h) and at 3 time intervals during immersion

Materials

Control
Amalgam
Ca(OH) 2
Cavit
Fuji II
MTA
ZnOE
a

0h

24 h

48 h

72 h

0.639 " 0.002

0.639 " 0.002
0.639 " 0.002
0.639 " 0.002
0.639 " 0.002
0.639 " 0.002
0.639 " 0.002

0.641 " 0.001

0.500 " 0.006
0.172 " 0.001
0.574 " 0.003
0.635 " 0.003
0.1 " 0.002
0.637 " 0.002

0.638 " 0.001

0.507 " 0.005
0.101 " 0.012
0.429 " 0.014
0.646 " 0.007
0.057 " 0.007
0.507 " 0.012

0.641 " 0.002

0.459 " 0.009
0.058 " 0.005
0.364 " 0.006
0.643 " 0.015
0.052 " 0.007
0.468 " 0.014

Unfilled plastic tubes.

2. Materials and methods

factors, the contact surface area between the filling material

and the dye (S) and the volume of the dye solution (V). The
larger the S, or the less the V, the higher the ratio S/V, and
the more the decoloration. Suppose a dental restoration is
2.5 mm wide and 3 mm high and a void of annulus shape
around this restoration is 5 mm wide, wide enough for a
bacterium to pass. Then by calculation S is 23.46 mm 2, V
is 0.12 ml and the S/V in the void is 196 mm 1. With this
high S/V ratio, the dye in the void can be decoloured to a
great extent. Unfortunately, it is very difficult, if not impossible, to take samples in a void and check decoloration of the
dye.
In a dye penetration experiment, many specimens are
immersed in dye solution in a large container. The dye in
the container can be decoloured by the test materials too.
Due to the large volume of the dye solution, however, the
ratio S/V in this container is low. Therefore, the decoloration
of the dye is much less significant in the container than in
the voids. The S/V ratio in the container can be raised by
increasing the surface area of the immersed material and
reducing the volume of the container and the dye.
In this study, decoloration of MB dye in the container was
observed in a mimicked dye penetration experiment. Steps
were taken to increase the S/V ratio in the container as much
as possible and the change in optical density (OD) of the
MB dye in the container was measured during immersion of
different dental filling materials. The decoloration under
low S/V conditions in the container will indicate more
decoloration under high S/V conditions in the voids.

Seventy plastic tubes (silicon rubber tubes, Holland

Emergo, Landsmeer, The Netherlands), of 1 mm wall thickness, 1.5 mm inner diameter and approximately 10 mm
long, were prepared and stored in distilled water until filling.
With use of a low-speed saw (Isomet 11-1180, Buehler
Ltd, Evaston, IL, USA), the apical 3 mm root tip and the
crown were removed from 70 fully developed human
maxillary central incisors which had been stored in 10%
formalin. The roots, each 10 mm in length, were used in
this study. The root canals were prepared with Gates Glidden drills (Maillefer, Ballaigues, Switzerland) up to size 6
(ISO size 150) under copious running water. In this way all
prepared root canals had the same diameter of 1.5 mm. The
prepared roots were stored in distilled water until filling.
The filling materials tested in this experiment were Tytin
amalgam (Kerr, Romulus, MI, USA); calcium hydroxide
(Merck, Darmstadt, Germany); zinc oxide eugenol
(ZnOE) cement (Standard Dental Producten, The Hague,
The Netherlands); Fuji II glass ionomer (GC Corporation,
Tokyo, Japan); Cavit (Espe GmbH, Seefeld, Germany), a
temporary coronal filling material containing ZnOE and
calcium sulfate; and mineral trioxide aggregate (MTA)
(Loma Linda University, Loma Linda, USA), a recently
developed root-end filling material.
Ca(OH) 2 was mixed with physiological saline at a powdersaline ratio of 6:4. The other materials were mixed
according to the manufacturers recommendations. Each

Table 2
Decoloration of methylene blue (MB) during immersion of human tooth roots filled with various materials
Materials

Control
Amalgam
Ca(OH) 2
Cavit
Fuji II
MTA
ZnOE
a

Unfilled tooth roots.

Mean " s.d. optical density of 1% MB before immersion (0 h) and at 3 time intervals during immersion
0h

24 h

48 h

72 h

0.639 " 0.002

0.639 " 0.002
0.639 " 0.002
0.639 " 0.002
0.639 " 0.002
0.639 " 0.002
0.639 " 0.002

0.725 " 0.006

0.598 " 0.007
0.214 " 0.005
0.631 " 0.007
0.753 " 0.004
0.261 " 0.010
0.639 " 0.006

0.738 " 0.008

0.597 " 0.005
0.162 " 0.004
0.614 " 0.007
0.766 " 0.003
0.242 " 0.012
0.597 " 0.007

0.728 " 0.005

0.524 " 0.009
0.147 " 0.007
0.537 " 0.008
0.778 " 0.003
0.176 " 0.009
0.509 " 0.006

587

M.-K. Wu et al./Journal of Dentistry 26 (1998) 585589

material was used to fill 10 roots and 10 plastic tubes.

Although the silicone tubes used are not rigid, the fillings
were found to have standardized shape and size in a pilot
study. All the filled roots and tubes were kept at 100%
humidity for 48 h before immersion in dye.
Ten roots and 10 plastic tubes were not filled, serving as
controls.
To mimic a dye penetration experiment, one end of the
filled or unfilled plastic tubes, as well as the apical end of
the roots were immersed in 0.8 ml of 1% MB solution (pH
7.0) in small plastic containers which were 12 mm high and
had an inner diameter of 15 mm. Each container stored five
specimens from the same group. Therefore, two containers
were used for each material. The immersion proceeded at
100% humidity and 20!C.
A pair of 0.1 ml samples of MB were taken, one from
each container, before immersion and after 24, 48 and 72 h
of immersion. The OD value of each sample, which had
been diluted with 1.9 ml distilled water, was then measured
using a 550S UVVIS spectrophotometer (Perkin-Elmer,
Norwalk, CT, USA) at 596 nm. With the SPSS (Norusis,
version 6.1, Chicago, IL, USA), the change in OD values

2. On immersion of unfilled plastic tubes (controls) or

plastic tubes filled with Fuji II, the OD values of the MB
solutions were stable at all time intervals (Fig. 1, P 0.4622
for plastic tubes, P 0.6341 for Fuji II). With immersion of
the tubes containing the other materials, the OD values of
the MB solution decreased over time (Fig. 1, P ! 0.01 for
amalgam, Cavit, Ca(OH) 2, MTA or ZnOE). At 24 h, the OD
value decreased by 73% for the Ca(OH) 2/silicone group,
and 84% for the MTA/silicone group (Table 1). These considerable decolorations could be seen by the naked eye.
On immersion of unfilled tooth roots or roots filled with
Fuji II, the OD values of the MB solutions increased over
time (Table 2, P ! 0.01 for both). With immersion of the
roots containing the other materials, the OD values of
the MB solution decreased over time (Table 2, P ! 0.01
for amalgam, Cavit, Ca(OH) 2, MTA or ZnOE).

related to each time interval was statistically analysed for

each test material using one-way ANOVA.

This explains why, in the present study, MB was

decoloured by Ca(OH) 2. Calcium oxide contained in MTA
may, with water, form Ca(OH) 2 which decolours the MB.
In several MB penetration studies, little dye penetration
was observed along Ca(OH) 2-containing materials or MTA
retrograde fillings, leading to the conclusion that they provided a tight seal [1418]. These conclusions may be questioned considering our findings that the MB dye solution

3. Results
The mean OD values before immersion and at the three
time intervals during immersion are shown in Tables 1 and
0.80

4. Discussion
MB is unstable with caustic alkali [12] and hydrolysed
to transparent thional as follows [13]:

Amalgam

Ca(OH)2

Cavit

MTA

ZnOE

Control

Fuji II

0.70
0.60

OD value

0.50
0.40
0.30
0.20
0.10
0.00

0h

24h

48h

72h

Period of immersion
Fig. 1. Optical density (OD) of 1% methylene blue changed by various dental filling materials in plastic tubes. Control: unfilled plastic tubes.

588

M.-K. Wu et al./Journal of Dentistry 26 (1998) 585589

was greatly decoloured after a 24 h-immersion of MTA or

Ca(OH) 2-filled roots.
On the other hand, MB is blue in its oxidized form, which
can be reduced and become colourless [19]:
MB 2H 2e MBH2
(oxidized form, blue) (reduced form colourless)
Therefore, amalgam, Cavit (containing ZnO) and ZnOE can
decolour the MB because they contain reducing agents, such
as Zn, Cu, Ag, etc. The MB was decoloured to a greater
extent by Cavit than by ZnOE (Fig. 1), probably because the
MB particles are adsorbed to the surfactant of the high
molecular polymers in Cavit [20].
Human teeth stored in formalin have often been used in
dye penetration studies [21]. When teeth stored in formalin
were used in dye penetration studies, the OD value of the
MB solution may be increased by the formalin-stored teeth.
However, decoloration of MB was found in contact with
amalgam, Ca(OH) 2, Cavit, MTA or ZnOE in human tooth
roots which had been stored in 10% formalin (Table 2),
indicating that the increase in OD related to formalin was
insufficient to prevent an overall decoloration with these
materials.
In the present experiment, in order to increase the S/V
ratio in the container, five filled specimens were placed
together in a small container which contained only 0.8 ml
MB solution at the beginning of the immersion period.
Removal of 0.1 ml aliquots of tracer at different time intervals further reduced the volume of tracer in the container
and slightly increased the S/V ratio. By calculation, the S/V
in the container was 0.01 mm 1 at the beginning of immersion, 0.011 mm 1 after sampling at 24 h and 0.013 mm 1
after sampling at 48 h. Decoloration of a small amount of
tracer secluded within gaps in the specimen was assumed
not significantly to affect the bulk external tracers density
and was neglected in the calculation. The MB in the container, where the S/V ratio was at least 15 000 times lower as
that in the voids (see the calculation in the Introduction),
was decoloured significantly by MTA and Ca(OH) 2 (Fig. 1),
indicating that the MB in the voids may be decoloured by
these two materials to almost colourless. The extent of
decoloration caused by amalgam, Cavit and ZnOE was
less significant (Fig. 1). However, the extent of decoloration
in the voids should be greater than that in the container.
Whether the observation of MB penetration along the latter
three materials will also be hindered by the decoloration of
dye remains unknown.
In a recent dye penetration study [22], the volume of MB
that penetrated filled roots was measured after dissolving the
root in 5 ml of 35% HNO 3. The authors found that the MB
was decoloured in 5 ml of 35% HNO 3 after 48 h and therefore decided to dissolve the root in this acid solution for only
24 h in order to keep the dye blue. In our opinion, however,
the MB that penetrated into the voids along the filling may

even decolour within 24 h because the volume of MB is so

small in that situation.
In a dye leakage experiment, the specimens are immersed
in MB solution sometimes in a large container. On one hand,
the MB solution may be decoloured in the voids during a
long immersion period of up to 2 weeks [14,23]; on the
other hand, the fresh blue dye solution in the large container
may diffuse into the voids through the immersed end of the
specimen. Therefore, in that part of the void which is close
to the immersed end of the specimen, the concentration of
the blue MB may be higher than that in the other part of the
void which is relatively far from the immersed end. This is
one reason why, in some studies, it was stated that the dye
appears to become paler in colour as it penetrates deeper
and the end-points of leakage were difficult to determine
[24,25]. The length of MB penetration recorded very often
does not reveal the full length of the void, but is determined
by some other factors including the rate of decoloration of
MB solution inside the void and the rate of diffusion of the
fresh blue MB solution from the container into the void.
This may partly explain why bacteria could penetrate over
the full length of the root-canal filling in all the specimens
tested [26,27], whereas the penetration of MB has been
reported to be less than 4.5 mm along the filling in 80%
of the some 23 measurements published during 1980
1990 [11].
Linear measurements of MB penetration have been made
after longitudinal splitting or cross sectioning. Usually only
one side of the root filling is observed after longitudinal
splitting. It was expected that more dye penetration would
be found after a series of horizontal sections were taken.
However, contrary to that expectation, less penetration of
dye was recorded after cross sectioning than after longitudinal splitting [11]. This was probably because the cooling
water used during the cross sectioning may have further
diluted the blue MB in the void and resulted in shorter linear
recordings.
With the findings in this study, it can be concluded that
the recorded length of MB penetration in dye leakage studies may not always reveal the full length of the void along
the filling, since the MB solution can be significantly decoloured by some dental filling materials. The accuracy of
these MB penetration results partly depends on the chemical
properties of the test materials and the dye used. Other dyes,
like India ink, eosin, rhodamine B, etc., have been used also
in dye leakage studies. It is recommended that the influence
of the materials tested on the optical density of the dye to be
used should be evaluated prior to the dye penetration
experiment.

Acknowledgements
The authors wish to thank Professor Dr J. F. J. Engbersen,
Faculty of Chemical Technology, University of Twente,
The Netherlands for his advice concerning the chemistry

M.-K. Wu et al./Journal of Dentistry 26 (1998) 585589

of methylene blue. This work was carried out at the Department of Cariology Endodontology Pedodontology,
Academic Centre for Dentistry Amsterdam (ACTA),
Amsterdam.

References
[1] Brannstrom M., Nyborg H. The presence of bacteria in cavities filled
with silicate cement and composite resin materials. Swedish Dental
Journal, 1971;64:149155.
[2] Goldberg J., Tanzer J., Munster E. et al. Cross-sectional clinical evaluation of recurrent enamel caries, restoration of marginal integrity,
and oral hygiene status. Journal of the American Dental Association,
1981;102:635639.
[3] Cox C. F., Keall C. L., Keall H. J. et al. Biocompatibility of surfacesealed dental materials against exposed pulps. Journal of Prosthetic
Dentistry, 1987;57:18.
[4] Seltzer S., Bender I. B. Cognitive dissonance in endodontics. Oral
Surgery Oral Medicine Oral Pathology, 1965;20:505516.
[5] Harty F. J., Parkins B. J., Wengraf A. M. Success rate of apicectomy.
A retrospective study of 1016 cases. British Dental Journal,
1970;129:407413.
[6] Bystrom A., Sundqvist G. Bacteriologic evaluation of the efficacy of
mechanical root canal instrumentation in endodontic therapy.
Scandinavian Journal of Dental Research, 1981;89:321328.
[7] Jacobsen S. M., Von Fraunhofer J. A. The investigation of microleakage in root canal therapy. An electrochemical technique. Oral
Surgery Oral Medicine Oral Pathology, 1976;42:817823.
[8] Mahler D. B., Nelson L. W. Factors affecting the marginal microleakage of amalgam. Journal of the American Dental Association,
1984;108:5154.
[9] Derkson G. D., Pashley D. H., Derkson M. E. Microleakage measurement of selected restorative materials: a new in vitro method. Journal
of Prosthetic Dentistry, 1986;56:435440.
[10] Wu M.-K., DeGee A. J., Wesselink P. R., Moorer W. R. Fluid transport and bacterial penetration along root canal fillings. International
Endodontic Journal, 1993;26:203208.
[11] Wu M.-K., Wesselink P. R. Endodontic leakage studies reconsidered:
Part I. methodology, application, and relevance. International Endodontics Journal, 1993;26:3743.
[12] Windholz, M, Budavari, S., Stroumtsos, L. Y. and Fertig, M. N. (eds.),
The Merck Index, 9th edn. Merck, Rahway, 1976.

589

[13] Coffey, S. (ed.) Rodds Chemistry of Carbon Compounds, Vol. IV,

Part H, 2nd edn. Elsevier Scientific, Amsterdam, 1978.
[14] Porkaew P., Retief D. H., Barfield R. D. et al. Effects of calcium
hydroxide paste as an intracanal medicament on apical seal. Journal
of Endodontics, 1990;16:369374.
[15] Limkangwalmongol S., Abbott P. V., Sandler A. B. Apical dye
penetration with four root canal sealers and gutta-percha using longitudinal sectioning. Journal of Endodontics, 1992;18:535539.
[16] Holland R., Alexandre A. C., Murata S. S. et al. Apical leakage following root canal dressing with calcium hydroxide. Endodontics and
Dental Traumatology, 1995;11:261263.
[17] Holland R., Murata S. S., Dezan Jr E. et al. Apical leakage after root
canal filling with an experimental calcium hydroxide gutta-percha
point. Journal of Endodontics, 1996;22:7173.
[18] Torabinejad M., Higa R. K., McKendry D. J. Dye leakage of four root
end filling materials: effects of blood contamination. Journal of Endododontics, 1994;20:159163.
[19] Morris, G. H., A Biologists Physical Chemistry, 5th edn. William
Clowes and Sons, London, 1978.
[20] Handreck G. P., Smith T. D. Adsorption of methylene blue from
aqueous solutions by ZSM-5-type zeolites and related silica polymorphs. Journal of the Chemical Society, Faraday Transactions 1,
1988;84:41914201.
[21] Mann S. R., McWalter G. M. Evaluation of apical seal and placement
control in straight and curved canals obturated by laterally condensed
and thermoplasticized guttapercha. Journal of Endodontics,
1987;13:1017.
[22] Vassiliadis L., Liolios E., Kouvas V. et al. Effect of smear layer on
coronal microleakage. Oral Surgery Oral Medicine Oral Pathology,
1996;82:315320.
[23] Madison S., Zakariasen K. L. Linear and volumetric analysis of apical
leakage in tooth prepared for posts. Journal of Endodontics,
1984;10:422427.
[24] Scott A. C., Vire D. E., Swanson R. An evaluation of the Thermafil
endodontic obturation technique. Journal of Endodontics,
1992;18:340343.
[25] Pathomvanich S., Edmunds D. H. Variation in the microleakage
produced by four different techniques in root fillings in a simulated
root canal model. International Endodontic Journal, 1996;29:156162.
[26] Torabinejad M., Ung B., Kettering J. D. In vitro bacterial penetration
of coronally unsealed endodontically treated teeth. Journal of Endodontics, 1990;16:566569.
[27] Chailertvanitkul P., Saunders W. P., Mackenzie D. The effect of
smear layer on microbial coronal leakage of gutta-percha root fillings.
International Endodontic Journal, 1996;29:242248.