Aust Endod J 2019

ORIGINAL RESEARCH

The effect of different sealer removal protocols on the bond

strength of AH plus-contaminated dentine to a bulk-fill composite
Ove A. Peters, DDS, MS, PhD1,2 ; Mark R.X. Teo, BDSc1; Justin M. Ooi, BDSc1; Ashley S.W. Foo, BDSc1;
Yu-Yao Teoh, BDSc, MPhil1 ; and Alex J. Moule, BDSc, PhD1
1 School of Dentistry, UQ Oral Health Centre, The University of Queensland, Brisbane, Queensland, Australia
2 Department of Endodontics, Arthur A. Dugoni School of Dentistry, University of the Pacific, San Francisco, California, USA

Keywords
AH Plus, bonding, composite, sealer removal,
tensile strength.
Correspondence
Ove A. Peters, Department of Endodontics,
Arthur A. Dugoni School of Dentistry,
University of The Pacific, 155 5th St, San
Francisco, CA 94103, USA. Email:
opeters@pacific.edu
doi: 10.1111/aej.12383
(Accepted for publication 22 October 2019.)
Abstract
The integrity of composite bonding for post-endodontic restorations may be
compromised by sealer contamination. This study assessed the effect of differ-
ent sealer removal regimes on the bond strength of dentine to composite resin.
Dentine surfaces were contaminated with AH Plus sealer (Dentsply DeTrey,
Konstanz, Germany), followed by removal with either dry cotton pellets, cot-
ton pellets saturated with 95% ethanol, AH Plus cleaner or external surface
preparation. Dentine surfaces were not contaminated in a positive control
group. A bulk-fill composite (SDR; Dentsply) was bonded with Prime&Bond
active universal adhesive (Dentsply) onto the prepared surface. Composite
resin–dentine beams were produced, and tensile bond strength was deter-
mined using a universal testing machine. Using an etch-and-rinse adhesive,
bond strengths varied from 21.34 to 29.11 MPa with no statistical differences
among removal protocols. In conclusion, contamination by AH Plus sealer does
not appear to substantially interfere with bond strength between dentine and a
bulk-fill composite/etch-and-rinse system.

Epoxy resin sealers, such as AH Plus (Dentsply DeTrey,

Introduction
Bacterial leakage is a common reason for the failure of
endodontic therapy (1–3); leakage may occur along the
interfaces between sealer and dentine or gutta-percha, or
through voids in the sealer (4); therefore, the sealing
ability of a root canal filling depends on the sealing ability
of the sealer (5). Moreover, a current recommended pro-
tocol following root canal filling calls for the application
of dentin bonding resin to create an orifice barrier to pro-
tect the root filling from contamination by bacteria, saliva
and endotoxins (2).
In general, bonding to dentine involves the formation
of a hybrid layer, which may be affected when the den-
tine surface is contaminated (6). Specifically, the bond
strength of dentine adhesives is lower when the dentine
surface is impregnated by oil, haemostatic agents, tempo-
rary filling material or endodontic sealer residues (7–10),
specifically eugenol-containing products. Sealer rem-
nants in the pulp chamber may also lead to crown dis-
colouration (11,12).
Konstanz, Germany), are commonly used in endodontic
practice. The advantages of these include antimicrobial
action, adhesion to dentine, long working time, ease of
mixing and good sealability (3). Currently, a common
protocol for removing excess AH Plus sealer involves its
removal with a dry cotton pellet or 95% ethanol until
the cavity looks physically clean. However, it has been
shown that this method is not particularly effective for
complete removal of remaining sealer (8,10,13). Another
solvent used primarily in root canal retreatment, Endo-
solv R (Septodont, Saint-Maur-des-Fosses, France), has
been shown to be effective in preventing the deteriora-
tion of bond strength. However, the manufacturer has
stated that it contains formamide, a potential teratogen
(10). Root canal sealer solvents may potentially nega-
tively affect the bond strength of adhesive resins due to
some chemical interaction with adhesives or dentine sub-
strates (9). These solvents may also positively affect the
bond strength of adhesive resins because they facilitate
the removal of sealer remnants more safely (9).

© 2019 Australian Society of Endodontology Inc 1

Sealer Remnant Removal and Bond Strength
AH Plus Cleaner (Dentsply DeTrey) has recently been
marketed with the aim to facilitate removal of remaining
sealer. The main components of AH Plus Cleaner are
ethanol and tertiary butanol. The efficacy of removing
sealer remnants with AH Plus Cleaner is currently
unknown. Therefore, the aim of this study was to com-
pare the efficacy of AH Plus cleaner with several other
sealer removing agents on the microtensile bond strength
of dentine to a bulk-fill composite material. Specifically,
this study tested the hypothesis that using AH Plus Clea-
ner as a removal protocol on AH Plus-contaminated den-
tine would prevent deterioration of the bond interface
between dentine and a two-step etch-and-rinse adhesive.
Materials and methods
Specimen collection and preparation of dentine
surfaces
Ten extracted intact human third molars were collected
from oral surgeons with approval from the Dental
School’s Institutional Ethical Review Board. The teeth
were stored in 0.5% Chloramine-T at room temperature
and gamma-ray sterilised (Steritech, Narangba, Qld, Aus-
tralia) in accordance with established guidelines (14).
Between processes, the teeth were maintained in Chlo-
ramine-T solution to prevent dehydration of dentine.
The roots of the teeth were sectioned horizontally
2 mm below the cemento-enamel junction and mounted
onto SEM carriers with epoxy resin. They were then sec-
tioned perpendicular to the long axis of the tooth using a
low speed saw (IsoMet; Buehler, Lake Bluff, IL, USA)
under constant water cooling then further divided into
two halves to allow each tooth to be treated with two dif-
ferent protocols (n = 20; Fig. 1a). The sections were
enclosed with two metal matrix bands. One was adapted
circumferentially around the tooth, and the other was
cut to separate the two tooth halves (Fig. 1b).
Experimental groups and application of cleaning
protocols
Each tooth half was assigned to an experimental or con-
trol group (see Table 1). A thin layer of AH Plus was
applied evenly over the intracoronal dentine surface for a
duration of 10 min in the experimental groups (Fig. 1c).
Each tooth half was randomly allocated to a removal pro-
tocol (four tooth halves per protocol) as below (a–d).
Two tooth halves were also allocated for the uncontami-
nated control group (e).
a. Dry cotton pellet (contaminated, negative control)
b. Cotton pellet soaked in 95% ethanol (EtOH)
c. Cotton pellet soaked in AH Plus Cleaner
2 © 2019 Australian Society of Endodontology Inc
O. A. Peters et al.
d. Surface preparation with a high-speed fine diamond
bur (0.5 mm depth)
e. Fresh dentine (uncontaminated, positive control)
Each removal protocol was carried out until no AH
Plus was left visible on the dentine surface (Fig. 1d). For
the uncontaminated group, AH Plus was not applied to
the dentine surface.
Bonding procedure
The treated dentine surfaces were etched with 37% phos-
phoric acid (K-etchant gel; Henry Schein Halas, New
York, NY, USA) and bonded with a two-step etch-and-
rinse adhesive Prime&Bond active universal adhesive
(Dentsply) according to the manufacturer’s instructions
(Fig. 1e). A single 4mm increment build-up of a bulk-fill
composite resin (SDR; Dentsply) was constructed on the
bonded surface and cured with a LED light-curing unit
(BA Optima 10; Henry Schein Halas, Sydney, Australia)
according to the manufacturer’s instructions (Fig. 1f).
The matrix band was removed before sectioning
(Fig. 1g).
Microtensile bond strength testing
Each tooth was sectioned vertically in a buccolingual
and mesiodistal direction using the IsoMet saw with
water cooling to produce resin–dentine beams of an
approximate 0.9 9 0.9 mm cross section (Fig. 1h,i). A
minimum of 20 beams were collected for each
removal protocol subgroup for microtensile testing.
Each beam was fixed to a custom brass jig using a
cyanoacrylate resin gel.
Beams were stressed to failure in linear tension
using a universal testing machine (Model 3360;
Instron, Norwood, MA, USA) at a crosshead speed of
1 mm min 1 (Fig. 2a,b). The maximum force required
to separate the composite resin–dentine interface was
measured in Newtons (N). After testing, the speci-
mens were inspected to ensure that the failure
occurred at the resin–dentine interface. The values for
any test where failure occurred outside the resin–den-
tine interface were discarded. Cross-sectional dimen-
sions (length and width) of the beams were measured
at the failure site to within 0.01 mm using an elec-
tronic calliper. These dimensions were used to obtain
the cross-sectional area. Measurements of the beams
were conducted by two examiners. Intra-examiner
and inter-examiner variability were determined, and
means for each comparison were found to differ 1%
or less. The data were then converted to megapascals
(MPa) by dividing Newtons over the cross-sectional
area of the respective beam.
O. A. Peters et al. Sealer Remnant Removal and Bond Strength

(a) (b) (c)

(d) (e) (f)

(g) (h) (i)

Figure 1 Workflow in microtensile bond strength testing. (a) Sectioned molar tooth with coronal dentine exposed and divided mesiodistally into two
halves. (b) Metal matrix band applied to the tooth with a piece of matrix separating the two halves. (c) AH Plus sealer applied evenly to one half, the
other half being an uncontaminated control. (d) AH Plus cleaned with a cotton pellet soaked in AH Plus Cleaner (Dentsply, Konstanz, Germany) until
dentine surface was visibly clear. (e) Phosphoric acid etchant application to both tooth halves of the tooth. (f) Composite resin build-up with bulk-fill
composite resin (SDR, Dentsply). (g) Completed composite resin build-up with matrix band separating the two halves. (h) Tooth sectioned mesiodistally
and buccolingually to approximately 0.9 9 0.9 mm resin–dentine beams. (i) Photomicrograph of typical resin–dentine beams.

Statistical analysis Results

Microtensile bond strength was grouped according to
removal protocol and an analysis made with the clearing
parameters as independent measures and the bond
strength as a dependent measure. A D’Agostino and Pear-
son normality test was conducted, and the data were
found to be not normally distributed. As a result, a non-
parametric multiple group analysis (Kruskal–Wallis test)
was used, with statistical significance set at P = 0.05. A
post hoc Dunn’s test was used to compare results
between individual groups, allowing comparisons
between the positive control group and the treatment
protocols to be made.
Microtensile bond strengths are shown in Figure 3 and
summarised in Table 1. Using an etch-and-rinse protocol,
the median tensile bond strengths for the five subgroups
varied from 21.34 to 29.56 MPa. There were no signifi-
cant differences among experimental groups; likewise,
there was no significant difference between both controls
and the experimental groups, based on Kruskal–Wallis
analysis.
Most of the sample failures in tension occurred at the
adhesive interface between composite and dentine. A
small number of failures (<10%) occurred at the interface
between the resin–dentine beams and the brass jig, likely

© 2019 Australian Society of Endodontology Inc 3

Sealer Remnant Removal and Bond Strength O. A. Peters et al.

Table 1 Summary of data for five removal protocol groups, three exper-
imental and two control groups

Cotton
pellet AH Mechanical
Cleaning Negative and Plus refinement Positive
method control EtOH Cleaner only control

Median 21.9 26.67 29.57 21.34 29.11

(MPa)
Interquartile 14.13 14.03 13.81 11.04 23.31
Range
(MPa)
Standard 11.24 9.276 10.92 9.606 12.04
Deviation
(MPa)
Number of 26 (4) 34 (4) 22 (4) 22 (4) 23 (2)
beams
(tooth
Figure 3 The effect of different removal protocols on the microtensile
halves)
bond strength (MPa) of between AH Plus-contaminated dentine and
bulk-fill composite, in comparison to negative (contaminated) and posi-
tive (uncontaminated) controls. The shaded area shows corresponding
median bond strengths for the same composite placed with increments
(data from Van Ende et al., Dental Materials 2013; 29: 269–77) (20).

due to insufficient bonding of the cyanoacrylate resin.

The values from these tests were discounted.

(a)
(b)
Figure 2 Typical specimen appearance before and after testing for ten-
sile strength. (a) Photomicrograph of resin–dentine beam attached to
custom jig in the Instron universal testing machine with cyanoacrylate
resin. (b) Photomicrograph of resin–dentine beam adhesive failure.
Discussion
Under the conditions of the microtensile bond strength
tests reported here, contamination with AH Plus did not
appear to have a consistent effect on bonding to a bulk-
fill composite with an etch-and-rinse system. Moreover,
there was no statistical difference between different
methods of cleaning the dentine surface, as long as the
dentine surface was visibly clean before bonding with a
dentine adhesive. The standard deviation was large due
to the wide range of MPa values between samples. This
was attributed to anatomical variation between teeth and
formation of the hybrid layer.
Coronal dentine was used instead of pulp chamber
dentine due to the difficulty of acquiring a consistent
dentine surface for testing. Factors such as curved areas
or abnormal substrate may have potentially adversely
affected the results of the bond strength tests (4). Intra-
radicular dentine was not used due to large variations in
root canal anatomy of third molars (10).
From the mean data, it can be appreciated that AH Plus
Cleaner had the closest average MPa compared to the
uncontaminated control. AH Plus Cleaner contains etha-
nol and tertiary butanol. Ethanol is commonly used
when cleaning the pulp chamber, and it has been shown
that it is inadequate for removing sealer remnants (15).
There are limited studies on the effect of tertiary butanol

4 © 2019 Australian Society of Endodontology Inc

O. A. Peters et al.
on dentine; hence, its ability to remove AH Plus remains
to be determined.
In contrast to previous studies (10,15), an etch-and-
rinse protocol was used in this research instead of a self-
etch bonding agent. Self-etch bonding adhesives do not
remove the smear layer formed during contact between a
bur and dentine surface during cavity preparation or car-
ies removal because they use weak acids in the primer
composition, which results in retention of a smear plug
in the hybrid layer (16). As a result, there is less of a
change in the dentinal wall structure compared to the
stronger acids used in etch-and-rinse adhesives (17).
Self-etch adhesives are also less technique sensitive com-
pared to an etch-and-rinse system, as over-etching can
lead to collapse of dentinal collagen fibrils (18).
Etch-and-rinse adhesives generally use phosphoric acid
etch. This was used in the present study to clean the den-
tine surface before bonding. Etch has been shown to dis-
solve smear layer formed during cavity preparation (19).
The removal of the smear layer could be a confounding
factor as it has been shown to improve bonding strengths
due to deeper penetration of the primer into the dentinal
tubules, resulting in a stronger adhesive interface
between dentine and composite resin (19). However, it is
unclear whether etch can dissolve AH Plus sealer. If etch
can remove AH Plus sealer, then it would explain the
lack of statistical difference between any of the removal
protocol groups.
In a previous study using SDR composite, median
microtensile bond strengths ranged between 28–42 MPa
(20). This compares well with the microtensile bond
strength of the uncontaminated control in the present
study (29.11 MPa).
The results of the microtensile bond strength test alone
do not correlate to marginal leakage and are only an indi-
rect measure to assess coronal seal (10,21,22). The ratio-
nale for using a microtensile strength test is that a better
adhesion of restorative materials to dentine increases the
opportunity for good marginal sealing, longer life of the
restoration and withstanding of mechanical stresses (17).
During chemo-mechanical preparation a smear layer is
created that may be removed with various irrigation
solutions (23). It has been shown that, after smear layer
removal, AH Plus penetrates the tubules to a large extent
and has a good adaptation to canal walls, which provide
a better seal during root canal filling (24). Mechanical
surface refinement similar to what had been suggested
earlier for eugenol-containing sealers (6) was used as a
removal protocol in this study as there is no data regard-
ing the efficacy of such surface preparation in removal of
AH Plus sealer remnants. The average MPa value for the
refinement protocol was the lowest compared to all the
other groups. This could be due to remaining sealer
© 2019 Australian Society of Endodontology Inc
Sealer Remnant Removal and Bond Strength
remaining coupled with the potential presence of debris
retained in the resin–dentine interface.
Standard clinical protocols suggest the use of irrigation
with EDTA prior to root canal filling (25). EDTA has been
shown to remove the smear layer created after mechani-
cal refinement (26). In this study, the teeth were not
treated with this solution prior to the application of AH
Plus sealer. Without smear layer removal, AH Plus sealer
may have had reduced dentine tubule penetration. This
in turn could have lead to easier sealer removal and,
therefore, skewed bond strength results.
From this ex vivo study, the method of cleaning the
pulpal chamber does not appear to be crucial when
the cavity is prepared with an etch-and-rinse adhesive
for bonding. Further tests are still required to distin-
guish etching during the normal bonding process as a
confounding factor for sealer removal. Etch could be
included as a removal protocol in combination with a
self-etching adhesive to assess its ability to remove
remaining sealer and possibly debris. An effort could
also be made to reflect standard clinical protocols
including the use of EDTA and hypochlorite rinsing
prior to the application of AH Plus sealer. SEM testing
should be utilised in future testing to visualise the for-
mation of the hybrid layer and for the presence of any
sealer remnants.
In conclusion, contamination by AH Plus sealer does
not appear to substantially interfere with bond strength
between dentine and a bulk-fill composite/etch-and-
rinse system, as long as the dentine surface is visibly
clean before placing the bonding agents. It is recom-
mended that a phosphoric acid etch is used to condition
the dentine surface before bonding composite resin to a
root canal-treated tooth.
Acknowledgements
We gratefully acknowledge expert help and support by
Drs. William Ha and Patricia Wright. We would like to
thank Dentsply Sirona for donating material samples, as
well as Drs. Zeb McNamara and Barbara Woodhouse for
providing the teeth used in this study. We are finally
indebted to Mr. Bruce Rex for creating the custom jigs
used in this study.
Authorship declaration
All authors have contributed significantly to this research
project and are in agreement with this manuscript.
Conflict of interest
We do not have any conflict of interest to declare.
5
Sealer Remnant Removal and Bond Strength
References
1. Swanson K, Madison S. An evaluation of coronal
microleakage in endodontically treated teeth. Part i. Time
periods. J Endod 1987; 13: 56–9.
2. Wolanek GA, Loushine RJ, Weller RN et al. In vitro bacte-
rial penetration of endodontically treated teeth coronally
sealed with a dentin bonding agent. J Endod 2001; 27:
354–7.
3. Heling I, Gorfil C, Slutzky H et al. Endodontic failure
caused by inadequate restorative procedures: review and
treatment recommendations. J Prosthet Dent 2002; 87:
674–8.
4. Cobankara FK, Adanr N, Belli S. Evaluation of the influ-
ence of smear layer on the apical and coronal sealing abil-
ity of two sealers. J Endod 2004; 30: 406–9.
5. Wu MK, De Gee AJ, Wesselink PR. Leakage of four root
canal sealers at different thickness. Int Endod J 1994; 27:
304–8.
6. Peters O, G€ohring T, Lutz F. Effect of eugenol-containing
sealer on marginal adaptation of dentine-bonded resin fill-
ings. Int Endod J 2000; 33: 53–9.
7. Chaiyabutr Y, Kois JC. The effects of tooth preparation
cleansing protocols on the bond strength of self-adhesive
resin luting cement to contaminated dentin. Oper Dent
2008; 33: 556–63.
8. Kuga MC, So MV, De Faria-junior NB et al. Persistence of
resinous cement residues in dentin treated with different
chemical removal protocols. Microsc Res Tech 2012; 75:
982–5.
9. Topcuoglu HS, Demirbuga S, Pala K et al. The bond
strength of adhesive resins to AH Plus contaminated den-
tin cleaned by various gutta-percha solvents. Scanning
2015; 37: 138–44.
10. Roberts S, Kim JR, Gu LS et al. The efficacy of different
sealer removal protocols on bonding of self-etching adhe-
sives to ah plus-contaminated dentin. J Endod 2009; 35:
563–7.
11. Parsons JR, Walton RE, Ricks-Williamson L. In vitro
longitudinal assessment of coronal discoloration from
endodontic sealers. J Endod 2001; 27: 699–702.
12. Plotino G, Buono L, Grande NM et al. Nonvital tooth
bleaching: a review of the literature and clinical proce-
dures. J Endod 2008; 34: 394–407.
6 © 2019 Australian Society of Endodontology Inc
O. A. Peters et al.
13. Sarac D, Bulucu B, Sarac YS, Kulunk S. The effect of den-
tin-cleaning agents on resin cement bond strength to den-
tin. J Am Dent Assoc 2008; 139: 751–8.
€
14. Armstrong S, Breschi L, Ozcan M et al. Academy of dental mate-
rials guidance on in vitro testing of dental composite bonding
effectiveness to dentin/enamel using micro-tensile bond
strength (µTBS) approach. Dent Mater 2017; 33: 133–43.
15. Kuga MC, Faria G, Rossi MA et al. Persistence of epoxy-
based sealer residues in dentin treated with different
chemical removal protocols. Scanning 2013; 35: 17–21.
16. Tay FR, Pashley DH. Aggressiveness of contemporary self-
etching systems. I: Depth of penetration beyond dentin
smear layers. Dent Mater 2001; 17: 296–308.
17. Santos JN, Carrilho MR, De Goes MF et al. Effect of chemi-
cal irrigants on the bond strength of a self-etching adhe-
sive to pulp chamber dentin. J Endod 2006; 32: 1088–90.
18. Frankenberger R, Perdigao J, Rosa BT, Lopes M. "No-bottle" vs
"multi-bottle" dentin adhesives – a microtensile bond strength
and morphological study. Dent Mater 2001; 17: 373–80.
19. Kenshima S, Francci C, Reis A et al. Conditioning effect on
dentin, resin tags and hybrid layer of different acidity self-
etch adhesives applied to thick and thin smear layer. J
Dent 2006; 34: 775–83.
20. Van Ende A, De Munck J, Van Landuyt KL et al. Bulk-fill-
ing of high c-factor posterior cavities: Effect on adhesion
to cavity-bottom dentin. Dent Mater 2013; 29: 269–77.
21. Belli S, Zhang Y, Pereira PN, Pashley DH. Adhesive sealing
of the pulp chamber. J Endod 2001; 27: 521–6.
22. Ozt€ €
urk B, Ozer F. Effect of NaOCl on bond strengths of
bonding agents to pulp chamber lateral walls. J Endod
2004; 30: 362–5.
23. Torabinejad M, Handysides R, Khademi AA, Bakland LK.
Clinical implications of the smear layer in endodontics: a
review. Oral Surg Oral Med Oral Pathol Oral Radiol
Endod 2002; 94: 658–66.
24. Chen H, Zhao X, Qiu Y et al. The tubular penetration
depth and adaption of four sealers: A scanning electron
microscopic study. Biomed Res Int 2017; 2017: 2946524.
25. Regan JD, Fleury AA. Irrigants in non-surgical endodontic
treatment. J Ir Dent Assoc 2006; 52: 84–92.
26. Crumpton BJ, Goodell GG, McClanahan SB. Effects on
smear layer and debris removal with varying volumes of
17% REDTA after rotary instrumentation. J Endod 2005;
31: 536–8.