Basic ResearchTechnology

Dye Extraction Results on Bacterial Leakproof Root Fillings

Gustavo De-Deus, DDS, MS,* Fernanda Leal, DDS, Juliana Soares, DDS,
Aderval S. Luna, PhD, Criatiana Murad, DDS, MS, Sandra Fidel, DDS, MS, PhD, and
Rivail Antonio Sergio Fidel, DDS, MS, PhD
Abstract
This study was designed to assess and compare the
sensitivity and sealability results between bacterial
leakage and dye extraction. Eighty mandibular incisors
were prepared, and their roots were filled as follows
(n 20): G1, lateral condensation; G2, System B; and
G3, Thermafil system. All teeth were mounted in a
2-chamber apparatus, and the coronal access was exposed to human saliva for 100 days. The remaining
bacterial leakproof specimens were randomly assigned
to create 3 new and equal groups (n 10). These
bacterial leakproof specimens were submitted to a dye
extraction setup. Each specimen was placed into a
plastic vial with 5 mL of 2% methylene blue for 48
hours. After storage, the specimens were rinsed with
tap water and dissolved in nitric acid. A sample of 100
L of the supernatant from each specimen was analyzed in a UV-Visible spectrophotometer to determine
the amount of methylene blue leakage. The log-rank
test showed no significant difference in the bacterial
leakage pattern among the groups (P .05). In the dye
extraction analysis, Kruskal-Wallis H test was unable to
detect significant differences among all experimental
groups (P .05). The 3 filling techniques displayed
similar leakage for both leakage models. Moreover, the
present study demonstrated that all bacterial leakproof
specimens leaked when submitted to the blue methylene extraction model. (J Endod 2008;34:10931095)

Key Words
Bacterial leakage, blue methylene, dye extraction, filling technique, leakage models, sealability

From the *Department of Endodontics, Veiga de Almeida

University (UVA), Rio de Janeiro, RJ, Brazil; and Department
of Endodontics and Department of Analytical Chemistry, Rio
de Janeiro State University (UERJ), Rio de Janeiro, RJ, Brazil.
Address requests for reprints to Prof Gustavo De-Deus, R.
Desembargador Renato Tavares, 11, ap.102, Ipanema, Rio de
Janeiro, RJ, Brazil 22411-060. E-mail address: endogus@
gmail.com.
0099-2399/$0 - see front matter
Copyright 2008 American Association of Endodontists.
doi:10.1016/j.joen.2008.06.003

JOE Volume 34, Number 9, September 2008

he main challenge of the laboratory-based leakage testing models is to develop

experimental setups that can provide reproducible results and clear-cut conclusions regarding the sealing ability of either the tested materials or techniques. Moreover, it is also important to be able to evaluate the laboratory findings with the real
clinical situation (1). Thus, a standardized, reliable, and reproducible method is a
crucial requirement. Currently, dye leakage, fluid transport, bacterial penetration, and
glucose leakage are the methods used most often for microleakage studies. However,
none of these tests have been universally accepted.
Although the limitations of the traditional linear dye leakage evaluations were
previously well-addressed (2), this method has still been used in some recent studies
(3). On the other hand, the dye extraction method might provide more reliable results
because it quantitatively measures all of the dye taken up into the sample (4). Camps
and Pashley (5) showed that the dye extraction method yielded the same results as fluid
transport experimental setups, while saving laboratory time.
Studies that use a bacterial tracer derived from specific cultures or from human
saliva are considered to be more reliable than tests with dye (6 9). However, bacterial
leakage studies are limited static models that do not simulate any conditions found in the
oral cavity such as temperature changes, dietary influences, and salivary flow. This
method allows an evaluation of the samples after specific periods of time while also
requiring long periods of observation and is labor-intensive.
The present study was designed to assess and compare the sensitivity and sealability results between the bacterial leakage and dye extraction methods of testing. The
general purpose of this study was to understand the relationship between bacterial and
dye leakage test results. To superimpose the data, a nondestructive bacterial leakage
model was performed. Subsequently, only the remaining bacterial leakproof specimens
were submitted to the dye extraction assay. Therefore, the core hypothesis tested was
whether bacterial leakproof specimens will reveal similar patterns of dye leakage. As a
derived purpose, the leakage expression obtained by 3 well-documented root filling
techniques was assessed.

Materials and Methods

Specimen Preparation
Eighty well-preserved extracted human mandibular incisors, with a 20 1 mm
length and straight roots, were selected from the tooth bank of the Rio de Janeiro State
University. The teeth were disinfected in 0.5% chloramine-T solution, stored in distilled
water at 4C (10), and used within 6 months after extraction.
The teeth were randomly distributed with the aid of a computer algorithm (http://
www.random.org) into 3 similar experimental groups (n 20). Ten teeth with intact
crowns served as negative controls, and 10 teeth that were not obturated served as
positive controls.
Instrumentation and Root Filling
Standard access cavities were made, the patency of each canal was confirmed, and
the working length was established by deducting 1 mm from the canal length. The root
canal was prepared by using K3 nickel-titanium rotary instruments (SybronEndo, West
Collins, CA) at 250 rpm. The final preparation had a 0.06 mm taper with a diameter of
0.25 mm at the apex. In G1, the roots were filled by using the lateral condensation
technique with a size 25 master gutta-percha cone (Diadent Group International,
Chongchong Buk Do, Korea) plus 10 accessory gutta-percha cones. In G2, the System

Dye Extraction on Bacterial Leakproof Specimens

1093

Basic ResearchTechnology
Cambridge, UK) after a kinetic assay at 550 nm wavelength with concentrated nitric acid as the blank.

Statistical Analysis
All bacterial leakage data were analyzed by descriptive statistics
with categorical outcomes. The mean day of leakage for 3 filling techniques was estimated. Nonparametric Kaplan-Meier survival curves
were constructed on the basis of the leaking specimens during the
experimental time. Specimens that did not leak were computed as censored variables. Leakage was statistically compared among the experimental groups by using the log-rank test.
For dye extraction data, a statistical analysis was performed with
Kruskal-Wallis H test and Dunn multiple comparison post hoc tests. The
alpha-type error was set at .05 for both analyses.

B technique (model 1005; EIE/Analytic, Redmond, WA) was used as

recommended by the manufacturer. In G3, a size 25 Thermafil plastic
obturator (Dentsply Tulsa Dental Products, Tulsa, OK) was heated in a
Thermaprep Oven (Dentsply Tulsa) and used. For all groups, Grossman
sealer (Endofill; Herpo Ltd, Petrpolis, RJ, Brazil) was used. The
crowns of the teeth were removed, leaving 10-mm roots, with the filled
roots blind-encoded and stored at 37C and 100% humidity for 7 days
to allow setting of the sealer.

Bacterial Leakage
The apparatus used to evaluate bacterial leakage was described
previously (6 9). Briefly, the specimens were mounted in a doublechamber apparatus, where the inferior chamber were filled with 3 mL
sterile Brain Heart Infusion (BHI; Oxoid Ltd, Basingstoke, UK). Thus,
2 mm of the resected root was immersed in the broth. Afterwards, the
reservoirs were filled with human saliva (20 mL) mixed with BHI broth
in a 1:1 (v/v) ratio and were replenished every 3 days (5). Human saliva
was collected from one individual, and the volunteer did not brush for
at least 12 hours before collection (11). The system was incubated at
37C and checked daily for the appearance of turbidity in the BHI broth
during the following 100 days.
Dye Extraction Measuring
The remaining bacterial leakproof specimens were randomly assigned with the aid of a computer algorithm (http://www.random.org)
to create 3 equal groups (n 10). Thus, 10 bacterial leakproof specimens of each filling technique were submitted to the dye extraction
setup. Both positive and negative controls were also reused from the
bacterial experiment. The specimens were washed in running water for
20 minutes and then resterilized overnight in ethylene oxide gas
(BIOXXI Sterilization Services Ltd, Rio de Janeiro, Brazil).
Each superior chamber was filled with 5 mL of 2% methylene blue
(pH 7.7), and the assembly was transferred to an incubator that provided 100% humidity at 37C for 48 hours. The specimens were demounted from the double chamber, rinsed under tap water for 15
minutes, and then stored individually in a glass vial containing 800 L
of concentrated (65 wt %) nitric acid for 3 days. Vials were centrifuged
at 3500 rpm for 4 minutes, and 100 L of the supernatant from each
was then analyzed in a UV-Visible spectrophotometer (Camspec M 330,
1094

De-Deus et al.

Results
Bacterial Leakage
No sample in the negative control group demonstrated leakage
during the 15-week experiment. On the other hand, all specimens of the
positive control group showed broth turbidity within 3 days after incubation. The log-rank test showed no statistically significant differences
among the groups (P .05). Leakage was observed in 6 of 20 samples
for G1 during the experimental period, compared with 4 of 20 teeth
from G2 and 4 of 20 samples in G3 (Fig. 1). The survival mean time for
leakage was 82 days (standard deviation [SD], 7.5 days) in G3, 76 days
in G2 (SD, 7.1 days), and 69 days in G1 (SD, 8.5 days).
Dye Extraction Leakage
No leakage was recorded in the negative control group, whereas
the positive control group had substantial leakage. Even when using
bacterial leakproof specimens, methylene blue leakage existed in every
sample. Overall, methylene blue leakage was variable in the 3 experimental
groups, ranging from 0.118 0.24. The Kruskal-Wallis H-test analysis was
unable to detect significant differences among all experimental groups (P
.05). The dye extraction results are shown in Fig. 2.

0.4

0.3

Absorbance

Figure 1. Kaplan-Meier curves showing the cumulative survival of the specimens

(percent survival), defined as the absence of bacterial leakage, between the 3
groups during the experimental time. Eighty mandibular incisors were prepared, and their roots were filled as follows (n 20): G1, lateral condensation;
G2, System B; and G3, Thermafil system.

0.2

0.1

G1

G2

G3

Figure 2. Box plots of the absorbance measures (dye extraction), which illustrate the median, minimal, and maximal traces, as well as the variance in each
experimental group. Eighty mandibular incisors were prepared, and their roots
were filled as follows (n 20): G1, lateral condensation; G2, System B; and G3,
Thermafil system.

JOE Volume 34, Number 9, September 2008

Basic ResearchTechnology
Discussion
The tested root filling techniques displayed a similar pattern of
bacterial leakage, which is not a new result. Some earlier bacterial
leakage studies demonstrated no differences among continuous methods of condensation, Thermafil and lateral condensation (7, 8). Accordantly, the present study was also unable to rank the filling techniques by
using the dye extraction leakage method. As a consequence, an answer
to the core hypothesis tested emerged as the bacterial leakproof specimens revealed similar methylene blue extraction leakage pattern results. This is an interesting finding, because it was previously established
that the methylene blue molecule was able to penetrate into areas that
might not be reached by bacteria (12, 13). In other words, methylene
blue dye is considered a more sensitive tracer than bacterial or bacteriasized tracers (12, 13). Even so, the present experimental setup was
unable to detect differences between the overall results obtained by both
leakage models.
It is worthwhile mentioning that in the bacterial leakproof specimens, microorganisms were unable to find a passage through the root
filling. This means that the bacterial leakproof root fillings were free
from a sizeable through-and-through void. Therefore, bacterial leakproof specimens can have a hypothetically acceptable pattern of filling
in terms of fill density and dentin adaptation, as hypothesized in the
current study with regard to the use of bacterial leakproof root fillings.
Taking into account the present results, 2 interesting points can be
made: (1) both leakage models can be of low sensitivity to detect differences among the filling techniques, and (2) the differences among
the specimens might be too small in terms of sealing ability. The question remains whether the comparison among endodontic materials or
techniques makes sense before the laboratory leakage models have
proved their real value. To state the matter differently, the object of study
should be whether gross or slight leakage has any biologic significance
(14). One method to elucidate this issue might be to follow the research
setup introduced by Oliver and Abbott (15), in which endodontically
treated teeth, scheduled for extraction, are used in an in vitro assay. This
approach allows accessing the level of clinical significance of the results
provided by the laboratory models. With a similar model, Susini et al.
(16) were unable to establish a positive correlation between the dye
leakage model and the in vivo presence of periapical radiolucency.
Nevertheless, using methylene blue in a dye extraction model was successfully correlated to the quality of the root canal filling. Many reasons
can be given to justify these findings. However, the bottom line is that
some association was established between the in vivo status of a root
filling and laboratory results. It is worth noting that although the results
from laboratory studies might not be directly extrapolated to the clinical
situation, it is very useful if the clinical effectiveness of the endodontic

JOE Volume 34, Number 9, September 2008

materials or techniques can be predicted by in vitro findings. However,

low evidence levels regarding in vitro leakage studies are well-documented (1, 2, 15, 16) and need to be viewed with like stimuli to perform
comparisons among leakage models. In addition, the use of innovative
research setups will help in the definition of a gold standard model for
in vitro leakage studies.
In summary, the present study demonstrated that all bacterial leakproof specimens leaked when submitted to the blue methylene extraction model. Although there are clear differences in their intrinsic features, the 3 filling techniques tested displayed a similar leakage
expression for both experimental models.

References
1. Editorial Board of the Journal of Endodontics. Wanted: a base of evidence. J Endod
2007;33:14012.
2. Wu M-K, Wesselink PR. Endodontic leakage studies reconsidered: part Imethodology,
application and relevance. Int Endod J 1993;26:37 43.
3. Zmener O, Pameijer CH, Serrano SA, Vidueira M, Macchi RL. Significance of moist
root canal dentin with the use of methacrylate-based endodontic sealers: an in vitro
coronal dye leakage study. J Endod 2008;34:76 9.
4. Hamad HA, Tordik PA, McClanahan SB. Furcation perforation repair comparing gray
and white MTA: a dye extraction study. J Endod 2006;32:337 40.
5. Camps J, Pashley D. Reliability of the dye penetration studies. J Endod
2003;29:592 4.
6. Siqueira JF Jr, Ras IN, Lopes HP, De Uzeda M. Coronal leakage of two root canal
sealers containing calcium hydroxide after exposure to human saliva. J Endod
1999;25:14 6.
7. Siqueira JF Jr, Ras IN, Favieri A, Abad EC, Castro AJR, Gahyva SM. Bacterial leakage
in coronally unsealed root canals obturated with 3 different techniques. Oral Surg
Oral Med Oral Pathol Oral Radiol Endod 2000;90:64750.
8. De-Deus G, Murad C, Paciornik S, Reis CM, Coutinho-Filho T. The effect of the
canal-filled area on the bacterial leakage of oval-shaped canals. Int Endod J
2008;41:18390.
9. De-Deus G, Brando MC, Fidel RA, Fidel SR. The sealing ability of GuttaFlow in
oval-shaped canals: an ex vivo study using a polymicrobial leakage model. Int Endod
J 2007;40:794 9.
10. De Wald JP. The use of extracted teeth for in vitro bonding studies: a review of
infection control considerations. Dent Mater 1997;13:74 81.
11. Gomes BPFA, Sato E, Ferraz CCR, Teixeira FB, Zaia AA, Souza-Filho FJ. Evaluation of
time required for recontamination of coronally sealed canals medicated with calcium
hydroxide and chlrehexidine. Int Endod J 2003;36:604 9.
12. Barthel CR, Moshonov J, Shuping G, Orstavik D. Bacterial leakage versus dye leakage
in obturated root canals. Int Endod J 1999;32:370 5.
13. Kersten HW, Moorer WR. Particles and molecules in endodontic leakage. Int Endod
J 1989;22:118 24.
14. Wimonchit S, Timpawat S, Vongsavan N. A comparison of techniques for assessment
of coronal dye leakage. J Endod 2002;28:1 4.
15. Oliver CM, Abbott PV. Correlation between clinical success and apical dye penetration. Int Endod J 2001;34:637 44.
16. Susini G, Pommel L, About I, Camps J. Lack of correlation between ex vivo apical dye
penetration and presence of apical radiolucencies. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 2006;102:e19 23.

Dye Extraction on Bacterial Leakproof Specimens

1095