Journal of Dentistry (2004) 32, 173–196

www.intl.elsevierhealth.com/journals/jden

REVIEW

Resin bonding to cervical sclerotic dentin: A review

Franklin R. Taya,*, David H. Pashleyb

a
Paediatric Dentistry and Orthodontics, Faculty of Dentistry, University of Hong Kong, Prince Philip Dental
Hospital, 34 Hospital Road, Hong Kong, China
b
Department of Oral Biology and Maxillofacial Pathology, Medical College of Georgia, Augusta, GA, USA

Received 10 July 2003; accepted 15 October 2003

KEYWORDS
Sclerotic cervical
dentine; Resin; Adhesive
Summary Several reports have indicated that resin bond strengths to noncarious
sclerotic cervical dentine are lower than bonds made to normal dentine. This is
thought to be due to tubule occlusion by mineral salts, preventing resin tag formation.
The purpose of this review was to critically examine what is known about the structure
of this type of dentine. Recent transmission electron microscopy revealed that in
addition to occlusion of the tubules by mineral crystals, many parts of wedge-shaped
cervical lesions contain a hypermineralised surface that resists the etching action of
both self-etching primers and phosphoric acid. This layer prevents hybridisation of the
underlying sclerotic dentine. In addition, bacteria are often detected on top of the
hypermineralised layer. Sometimes the bacteria were embedded in a partially
mineralised matrix. Acidic conditioners and resins penetrate variable distances into
these multilayered structures. Examination of both sides of the failed bonds revealed a
wide variation in fracture patterns that involved all of these structures. Microtensile
bond strengths to the occlusal, gingival and deepest portions of these wedge-shaped
lesions were significantly lower than similar areas artificially prepared in normal teeth.
When resin bonds to sclerotic dentine are extended to include peripheral sound
dentine, their bond strengths are probably high enough to permit retention of class V
restorations by adhesion, without additional retention.
q 2004 Elsevier Ltd. All rights reserved.

Introduction that our current knowledge on the variability of

The efficacy of current adhesive systems is often
evaluated based upon their ability to bond to sound
dentine. Although many dentists bond to sound
dentine, a variety of pathological dentin substrates
are encountered in the clinical practice, which
include carious and sclerotic dentin. It is ironic
*Corresponding author. Tel.: þ 86-852-28590251; fax: þ 86-
852-23933201.
E-mail address: kfctay@hknet.com
clinical bonding substrates is so limited compared
with the progress achieved in adhesive technology.
This review examined the ultrastructure and bond-
ing characteristics of one type of abnormal bonding
substrate – noncarious sclerotic dentine. Essentially,
it compared the obstacles in bonding to sound vs.
sclerotic dentine. It will be seen that there is a
tremendous difference in substrate morphology as
well as that of the resin – dentine interfaces created
by bonding to such substrates. The recent introduc-
tion of contemporary self-etching primers and

0300-5712/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jdent.2003.10.009
174
the timely all-in-one adhesives are redefining
adhesive dentistry. Because these simplified
adhesive systems are easy to use, we have placed
extra emphasis on the interfaces created by these
agents on sclerotic cervical dentine.
Noncarious cervical sclerotic dentine
Noncarious cervical sclerotic lesions was described
by Zsigmondy in 18941 as angular defects, and by
Miller in 19072 as ‘wasting of tooth tissue’ that was
characterised by a slow and gradual loss of tooth
substances resulting in smooth, wedge-shaped
defects along the cemento-enamel junction. The
multifactorial etiology of these cervical lesions has
been extensively reviewed.3 – 9 There is increasing
evidence of the possible role of eccentric occlusal
stress in the pathogenesis of these hard tissue
defects.10 – 18 Recent studies on simulation of
wedge-shaped cervical lesions using various finite
element analytical models of cuspal flexure con-
firmed the contribution of stress induction in these
so-called ‘abfraction lesions’.19 – 22 Unrestored,
angular, wedge-shaped lesions demonstrated
severe stress concentration that varied with the
location of the teeth in the oral cavity, with the
highest stress contractions around the maxillary
incisors and premolars. These stresses were only
partially relieved after these lesions were restored.
Sclerotic dentine is a clinically relevant bonding
substrate in which the dentine has been physiologi-
cally and pathologically altered, partly as the
body’s natural defense mechanism to insult, and
partly as a consequence of colonisation by the oral
microflora. Partial or complete obliteration of the
dentinal tubules with tube- or rod-like sclerotic
casts is commonly observed.23 – 28 Depending on the
degree of clinical sensitivity of the lesion, various
levels of tubular patency were observed, with most
dentinal tubules being occluded within the insensi-
tive transparent dentine regions.25,27,29 – 34
In the absence of undercut retention, cervical
sclerotic dentine was found to be more difficult to
adhere to than normal dentin both in vitro and in
vivo, even with increased etching time.35 – 39 Recent
studies showed that the sclerotic casts that
obliterated the dentinal tubules were still present
after acid-conditioning of the sclerotic dentine,
resulting in minimal or no resin tag formation.
Furthermore, the zone of resin-impregnated sclero-
tic dentine was found to be thinner than those
observed in normal dentine.25,27,40 – 43
Due to the thinness of hybrid layers in sclerotic
dentine, and the complexity of the resin-bonded
F.R. Tay, D.H. Pashley
interface, regional tensile bond strength to cervical
sclerotic root dentine with some contemporary
adhesives was found to be 20 –45% lower than those
bonded to artificial wedge-shaped lesions created in
normal cervical root dentine.40,42 This was attrib-
uted to: (a) the presence of sclerotic casts within
dentinal tubules that precluded optimal resin
infiltration into the dentinal tubules, and/or (b)
the presence of a surface hypermineralised layer
that is more resistant to acid-etching. It was
postulated that an adhesive strategy that involved
micromechanical interlocking by the formation of a
resin – dentine interdiffusion zone combined with
resin-tag formation into the dentinal tubules would
be less effective when applied to the hyperminer-
alised sclerotic dentine.42,44 Contrary to these
findings, a recent study suggested that phosphoric
acid-etching was detrimental to the bonding of
sclerotic dentine, and that sclerotic dentine that
was treated with a hydrophilic primer exhibited
better marginal adaptation of resin composites than
similarly primed normal dentine.45 These authors
recommended that the layer of sclerotic dentine be
preserved for optimal bonding in cervical lesions.
Scanning electron microscopy of such surfaces,
even with the use of field emission-type micro-
scopes, does not provide sufficient detail to under-
stand the complex subsurface structures, or to
reveal how well these structures are demineralised
during etching.28,43 This information is best
obtained using transmission electron microscopy
(TEM). In this review, extensive TEM examinations
were used to evaluate biologic variations in
sclerotic cervical dentine. This is not an exhaustive
review of the literature on noncarious cervical
lesions or in resin-bonding to sclerotic dentine.
Rather, it is an attempt to summarise a number of
studies that provide a rationale for why resin bonds
to sclerotic, noncarious, wedge-shape cervical
lesions are lower than those made to normal
dentine at those same sites. This review will be
divided into two sections. In Section 1, microstruc-
tural changes that exist in noncarious, sclerotic
cervical dentine will be summarised. This is
followed by a review on the application of adhesive
resins to this altered bonding substrate.
Microstructural changes in sclerotic
dentine
Tubular occlusion
In dentine sclerosis, tubular obliteration with
rhombohedral, whitlockite crystallites (Fig. 1A)
Resin bonding to cervical sclerotic dentin: A review
Figure 1 (A) SEM micrograph of the body of a sclerotic
lesion showing a dentinal tubule that was heavily
occluded with whitlockite crystallites (pointer). The
adjacent tubules were almost completely obliterated
with peritublar dentine. (B) A higher magnification view
of the rhombohedral whitlockite crystallites (pointers)
that were present within another dentinal tubule in the
sclerotic lesion. P: peritubular dentine; I: intertubular
dentine.
has been well documented at an ultrastructural
level.27,31,34,46 – 48 A high degree of variation could
be observed even within a single lesion. While some
tubules may be completely devoid of, or sparsely
occluded with crystallites, others may be heavily
obliterated with crystallites and/or peritubular
dentine (Fig. 1B). Toward the surface of the lesion,
these crystallites were reduced in size and formed
columns of agglomerates that completely plug the
tubular orifices. They were often referred to as
sclerotic casts (Fig. 2A). At an ultrastructural level,
these tiny, electron-dense crystallites were sur-
rounded by a tube-like membranous structure26,27
175
Figure 2 (A) SEM micrograph of sclerotic casts that
blocked the dentinal tubular orifices (pointer) along the
surface of a shiny sclerotic lesion. (B) TEM micrograph
taken from an undemineralised section of the surface of a
bonded sclerotic lesion. Tubules were obliterated with
sclerotic casts that consisted of electron-dense crystal-
lites. They were surrounded by a tube-like sheath
(pointer). A thin, electron-dense, hypermineralised
layer was also evident along the surface of the lesion
(open arrow) (from Tay et al., 2000,57 with permission).
that probably represented a mineralised form of the
lamina limitans of the dentinal tubule (Fig. 2B).
Hypermineralised surface layer in shiny
sclerotic lesions
Unlike tubular occlusion, the presence of a surface
hypermineralised layer in natural cervical sclerotic
wedge-shaped lesions has only been elucidated
through the use of microradiography30 and FTIR
photoacoustic spectroscopic analysis.32 Although it
has been speculated that the hypermineralised
176
layer is devoid of collagen fibrils,37 the ultrastruc-
tural features of this layer were only verified
recently. Fig. 3A is a toluidine blue-stained,
undemineralised thin section taken from the dee-
pest part of a wedge-shaped defect that is least
accessible to tooth brushing. In the preparation of
the specimen, the surface of the lesion was not
cleaned and was fixed in Karnovsky’s fixative prior
to the embedding protocol for TEM preparation. A
surface layer of stained, unmineralised filamentous
bacteria could be seen, beneath which was an
approximately 15 mm thick hypermineralised layer.
Figure 3 (A) Light microscopic image of an undeminer-
alised section taken from the deepest part of an
untreated noncarious cervical sclerotic lesion. B: stained,
unmineralised bacteria; HM: hypermineralised surface
layer of the lesion; SD: intact sclerotic dentine; Pointers:
mineralised bacteria ghosts. (B) Corresponding unstained
TEM micrograph of the undemineralised section. The
hypermineralised nature of the surface layer (HM) could
be recognised by comparing its electron density with the
underlying sclerotic dentine (SD) (from Tay et al., 2000,57
with permission).
F.R. Tay, D.H. Pashley
Figure 4 (A) Undemineralised TEM micrograph of a
bacterium (B) within the hypermineralised layer (HM).
The intermicrobial matrix (i.e. the material between
adjacent bacteria) was mineralised and contained plate-
like crystallites (pointer) (from Tay et al., 2000,57 with
permission).
Mineralised bacteria could vaguely be discerned
within this layer. The corresponding undeminera-
lised TEM image (Fig. 3B) shows that both the
mineralised plaque zone and the surface layer of
the lesion are hypermineralised with respect to the
underlying sclerotic dentine. At a higher magnifi-
cation (Fig. 4), plate-like minerals could be recog-
nised within intermicrobial matrix.
The ultrastructure of the surface hyperminer-
alised layer is highly variable within the deepest
part of the wedge-shaped lesions. This can be
better discerned using demineralised TEM sec-
tions. It is interesting to observe that such a
layer is still retained after complete deminerali-
sation of the underlying sclerotic dentine. Fig. 5A
shows a thick, continuous, dense hyperminera-
lised layer that contained two different species
of bacteria. One species was trapped within this
layer, while the other was found to grow on the
surface of this layer. The hypermineralised layer
in Fig. 5B, on the other hand, consists of several
thin, discontinuous layers that were sandwiched
among different species of bacteria. This suggests
that changes in the microecology of the oral
environment may have resulted in the colonisa-
tion of different species of bacteria at sequential
periods. Each colony of bacteria was, in turn,
mineralised prior to the deposition of the next
colony. It is pertinent to point out that both
specimens, from which Fig. 5A and B were taken,
were brushed cleaned with a mixture of chlor-
hexidine and pumice prior to laboratory proces-
sing. Both appeared as clean, highly shiny lesions
when examined under magnification lens.
Resin bonding to cervical sclerotic dentin: A review
Figure 5 Variation in the ultrastructure of the hyper-
mineralised layer in noncarious cervical sclerotic lesions.
(A) This demineralised TEM micrograph showed a hyper-
mineralised layer (HM) within the deepest part of a
wedge-shaped lesion that was about 14 mm thick.
Bacteria colonies were trapped inside this layer (open
arrow). Another species of bacteria (arrowhead) accu-
mulated along the surface of the hypermineralised layer.
Dentinal tubules were not occluded with sclerotic casts
and were also filled with bacteria (pointer) (from Kwomg
et al., 2002,42 with permission). (B) Another deminer-
alised TEM micrograph showing a complex, intact, hybrid
structure that consisted of several discontinuous, hyper-
mineralised layers (pointers) that were sandwiched
among different colonies of bacteria (A, B, C and D).
The entire complex is about 18 mm thick. SD: sclerotic
dentine.
177
The presence of colonising bacteria on the
surface of sclerotic wedge-shaped lesions is
consistent with the report of Spranger.6 The
microecology beneath bacterial plaque changes
over time depending on the metabolism of the
microorganisms. This results in substantial pH
fluctuations along the tooth surface.49 Bacterial
products may trigger gingival inflammation with
an increased rate of sulcular fluid flow, which in
turn, provides nutrition for the microorganisms. If
a plentiful supply of fermentable carbohydrates is
available, the microbes release organic acids that
will lower the plaque pH and tend to deminer-
alise the underlying dental hard tissue. When the
carbohydrate source is depleted, the local pH
rises due to salivary buffering, and remineralisa-
tion of the dental hard tissue and mineralisation
of the dental plaque can proceed. In the absence
of carbohydrates, these bacteria may remain
viable for prolonged periods,50 utilising glyco-
gen-like intracellular polysaccharide as a metab-
olisable source of carbon during periods of
nutrient deprivation.51 They may also metabolise
amino acids and other nitrogenous substrates,
creating ammonia and other basic chemicals that
may elevate plaque pH and promote mineralis-
ation and, perhaps, hypermineralisation.52
Along the occlusal wall of the wedge-shaped
lesions, both the surface bacterial layer and the
hypermineralised layer are usually thinner, with
the latter between 1 and 2 mm thick. Similarly, the
gingival wall of a wedge-shaped lesion, which is
usually more accessible to tooth brushing, is usually
devoid of bacteria. However, a very thin surface
hypermineralised layer may sometimes be observed.
Such a layer, if present is around 200 –300 nm thick
and may be readily discerned in undemineralised
sections by its increased electron density, as well as
the characteristic arrangement of the crystallites,
which will be discussed in Section 3.3.
Mineral distribution
Minerals present within the hypermineralised layer
are larger in size compared with those within the
underlying sclerotic dentin (Fig. 6A). Unlike crystal-
lites within the underlying dentine that are randomly
arranged, those that are found in the hyperminer-
alised layer are longitudinally aligned along the c-
axis of the crystallites. This is clearly demonstrated
in Fig. 6B. Orientation of the crystallites along their
c-axis in the surface hypermineralised layer is
analogous to the induction of cellular orientation
by low-level electrical currents53 and the alignment
of filler particles within resin composites under
178 F.R. Tay, D.H. Pashley

STEM/EDX analysis

Hypermineralisation implies that the density of the

mineral within the surface layer of the defect is
higher than that of the underlying sclerotic dentine.
This is confirmed by a qualitative STEM/EDX line
scan of the calcium and phosphorus distribution
longitudinally across the surface layer of the
defect into the underlying sclerotic dentine
(Fig. 7A and B). Also evident in the line scan is the
presence of a region of decreasing calcium and
phosphorus counts along the top 500 nm of the
hypermineralised layer that is attributed to partial

Figure 6 (A) Crystallites (pointer) with the hyperminer-

alised layer (HM) were plate-like and were aligned along
the c-axis. They were also larger than those (arrowhead)
present within the underlying sclerotic dentine (SD). (B)
The characteristic orientation of crystallites along their
c-axis was also evident in a thick hypermineralised layer
taken from the deepest part of another wedge-shaped
lesion. The lesion surface was etched by a self-etching
primer, with blunting and rounding of the partially
dissolved crystallites.

the application of an electric field.54 Dentine is only

mildly piezoelectric compared with bone.55 How-
ever, piezoelectric potentials have been reported to
be generated when teeth are subjected to parafunc-
tional loading.6 Some have speculated that polaris-
ation of the remineralised crystallites caused by
piezoelectric potentials created during eccentric
tooth flexure results in their attraction and repulsion
by dipole interaction. Dipole interaction competes
with randomisation caused by Brownian motion
and alignment occurs as the dipole interaction
predominates.56
Figure 7 (A) Brightfield STEM image showing the
location of a line scan that was performed across the
surface hypermineralised layer of a noncarious cervical
sclerotic lesion (from Tay et al., 2000,57 with permission).
(B) Qualitative energy dispersive line scans showing the
distribution of calcium and phosphorus along
the adhesive, the surface hypermineralised layer of the
lesion and the underlying intact sclerotic dentine (from
Tay et al., 2000,57 with permission).
Resin bonding to cervical sclerotic dentin: A review 179

demineralisation by a self-etching primer (Clearfil crystallites without additional ZAF correction.

Liner Bond 2V; Kuraray Medical, Inc., Tokyo, Japan)
that was used to bond to sclerotic lesions.57
Quantitative energy dispersive X-ray spectra of the
elemental content of crystallites present within (a)
the surface hypermineralised layer, (b) the under-
lying intertubular dentine, and (c) within the dentinal
tubules of the wedge-shape defect are shown in
Fig. 8.57 As the analysis was performed using 70 nm
thick undemineralised sections, this enabled esti-
mation of the calcium/phosphate ratio of these
The Ca/P ratios of the crystallites within the
hypermineralised layer and the underlying dentine
approach the theoretical value of 1.67 calculated for
hydroxyapatite.58 The larger apatite crystallites
observed in the surface hypermineralised layer are
similar to larger apatite crystallites reported in
remineralised carious dentin59,60 and cementum.61
Conversely, the Ca/P ratio of crystallites from
the sclerotic casts within the dentinal tubules is
slightly lower than the calculated value of 1.50 for

Figure 8 Energy spectra from different locations of a noncarious cervical sclerotic lesion. (a) Spectrum showing
composition of crystallites within the surface hypermineralised layer. (b) Spectrum of crystallites within the underlying
intact sclerotic dentine. (c) Spectrum of crystallites occupying the lumen of a dentinal tubule. Spectra were obtained
using a 7 nm probe for 200 live seconds at 200 kV. The relative concentration of Ca, P and Mg, and the calculated Ca/P
ratios are shown in the table beneath the spectra (modified from Tay et al., 2000,57 with permission).
180 F.R. Tay, D.H. Pashley

tricalcium phosphate.58 The additional presence of

about 5% magnesium suggests that these crystallites
are whitlockite (Mg substituted b-tricalcium phos-
phate).

Status of the collagen fibrils within

the surface hypermineralised layer

An intriguing question that has remained unan-

swered is whether the surface hypermineralised
layer is devoid of collagen.36 Using special staining
for collagen, it can be seen that the supporting
matrix for the crystallites within the hyperminer-
alised layer consists of a bed of denatured collagen
(Fig. 9A). The transition from denatured collagen
(gelatin) to intact collagen with cross-banding is
evident at the base of the hypermineralised layer,
where some of the collagen fibrils are observed to
unravel into microfibrillar subunits. 57 This
transition from banded collagen to denatured
microfibrils is further illustrated beneath a layer
of bacteria in Fig. 9B, in which unraveling of
collagen fibrils created a network of microfibrillar
strands that no longer showed cross-banding.
It is possible that colonisation of bacteria along
the surface of the wedge-shaped defect results in
the production of acidic and enzyme by-products
that demineralise and denature the collagen fibrils
(Fig. 9B). Diffusion of the enzymes from the
demineralised surface downwards probably
accounts for the transition from completely
denatured collagen to unraveled microfibrils and
finally intact collagen within the underlying sclero- Figure 9 (A) TEM micrograph from a demineralised
tic dentine. The loss of phosphoproteins62 and specimens, showing that the supporting matrix of the
subsequent remineralisation of the surface bed of hypermineralised layer (HM) consisted of a bed of
denatured collagen under the possible influence of denatured collagen microfibrils(arrow). These subunits
piezoelectric charges generated from eccentric of collagen were formed by unraveling of collagen fibrils
flexural deformation6 may result in the character- (arrowhead). Within the underlying sclerotic dentine
istic orientation of the crystallites within the (SD), intact banded collagen fibrils could be seen
denatured microfibrillar matrix of the surface (pointers) (from Tay et al., 2000,57 with permission). (B)
hypermineralised layer. A thin hypermineralised layer (HM) beneath a layer of
bacteria (B) showing the transition from intact, banded
collagen fibrils into denatured collagen (gelatine),
Summary of the microstructural changes
appearing ultrastructurally as microfibrillar strands (poin-
in noncarious sclerotic cervical dentine ter). SD: sclerotic dentine.

Sclerotic dentine is an abnormal bonding sub-

strate that exhibits a high degree of variability
both in terms of occlusion of the dentinal tubules
as well as the thickness of the surface hypermi-
neralised layer. The latter, in particular is
invariably associated with bacteria. It is possible
that bacteria are involved in the pathogenesis of
the hypermineralised layer. That is, dentine may
first require demineralisation before it can be
hypermineralised. Also, mineral crystallites can-
not accumulate without a scaffold. The presence
of bacteria, apart from demineralising the den-
tine, also denatures the existing collagen matrix,
resulting in a bed of denatured collagen. This
may act as a scaffold upon which the deminer-
alised dentin may be subsequently remineralised.
The formation of the hypermineralised layer is
probably also enhanced by the presence of high
concentrations (10 ppm) of fluoride ions.63 – 65 As
denaturing of the collagen fibrils probably
removes some of the restrictions on the size of
Resin bonding to cervical sclerotic dentin: A review
crystallite formation, this may enable larger
crystallites to be formed during the process of
remineralisation. The unique arrangement of the
crystallites further suggests the complex role of
parafunctional stress on the formation of the
surface hypermineralised layer in these natural
wedge-shaped lesions.15,23 The presence of bac-
teria, mineralised bacterial matrices, hypermi-
neralised surfaces and mineral occluded tubules
makes sclerotic cervical dentine a unique multi-
layered bonding substrate. This implies that
bonding studies that attempt to generate hyper-
mineralised dentine in vitro by immersing par-
tially demineralised dentine in a remineralising
solution containing a high fluoride content66,67 do
not simulate the actual bonding conditions that
clinicians are likely to encounter when bonding to
noncarious cervical sclerotic dentine.
Bonding adhesive resins to sclerotic
dentine
Current dentine adhesives employ two different
means to achieve the goal of micromechanical
retention between resin and dentine.68 – 70 The first
method, the total-etch or etch and rinse technique,
attempts to remove the smear layer completely via
acid-etching and rinsing. The second approach, the
self-etch technique, aims at incorporating the smear
layer as a bonding substrate.
Total-etch technique
Most self-priming, single-bottle adhesives available
to-date attempt to bond to dentine that is etched
with inorganic or organic acids. Following rinsing of
the conditioners, retention is accomplished by
means of resin-infiltration into the exposed, demi-
neralised collagen matrix to form a hybrid layer of
resin-impregnated dentine.69 – 71 Systems containing
hydrophilic primer resins solvated in acetone or
ethanol were found to produce higher bond strength
when acid-conditioned dentine was left visibly moist
prior to bonding.72 Pioneered by Kanca,73 this
technique is often referred to as ‘wet bonding’.
The benefit of wet bonding stems from the ability of
water to keep the interfibrillar channels within the
collagen network from collapsing during resin-
infiltration.74,75 These channels, which are about
20 nm wide when fully extended, (Fig. 10) must be
maintained open to facilitate optimal diffusion of
resin monomers into the demineralised intertubular
dentine.76,77
181
Figure 10 Dentine collagen fibrils from a hybrid layer
that was produced using a total-etch technique. Inter-
fibrillar channels (pointers) must be maintained open for
optimal resin infiltration. This is achieved using a wet
bonding technique.
Self-etch technique
Recent developments in dentine bonding have
reintroduced the concept of utilising the smear
layer as a bonding substrate, but with improved
formulations that could etch through the smear
layer and beyond, into the underlying dentine
matrix. Failure to etch beyond the smear layer,
exemplified by some of the early adhesives that were
applied directly to the smear layer, resulted in weak
bonds due to the complete absence of a hybrid layer
in underlying intertubular dentine.78,79 Contempor-
ary self-etch adhesives have been developed by
replacing the separate acid-conditioning step with
increased concentrations of acidic resin monomers.
Two-step self-etching primers combine etching and
priming into a single step. The primed surfaces are
subsequently covered with a more hydrophobic
adhesive layer that is light-cured. In the presence
of water as an ionising medium, these adhesives that
etch through smear layers and bond to the under-
lying intact dentine.80,81 The recent introduction of
single-step (all-in-one) self-etch adhesives rep-
resents a further reduction in bonding steps that
eliminates some of the technique sensitivity and
practitioner variability that are associated with the
use of total-etch adhesives.82 – 84
When applied to sound dentine, the milder self-
etch adhesives produce a hybridised complex
(Fig. 11) that consists of a surface zone of hybridised
smear layer and a thin, subsurface hybrid layer in the
underlying intertubular dentine.85 – 87 Despite the
presence of a hybrid layer that was generally below
2 mm thick, high initial bond strengths have
182
Figure 11 The etching effect of a self-etching primer
(Clearfil SE Bond, Kuraray) through a thick smear layer
produced in sound dentine. The substructure of the smear
layer consisted of loose, globular subunits (arrow) and
channels that were filled with water prior to resin (R)
infiltration (asterisk). The hybridised complex consisted
of a hybridised smear layer (Hs) and a thin hybrid layer
(Ha) in intact dentine (D).
been reported for sound dentine.88 – 91 The more
aggressive self-etch adhesives completely dissolve
smear plugs and demineralise dentine to the extent
that is comparable with phosphoric acid-etching.87
There has been some concern that mild self-etch
adhesives may not be able to penetrate through
thick smear layers, such as those produced clinically
by rough diamond burs.85 – 92 The use of adjunctive
phosphoric acid pre-conditioning has been
suggested as a means to improve bonding of self-
etching primers to sound dentine with thick smear
layers.93 – 96
Problems in bonding to sclerotic dentine
Irrespective of the use of a total-etch or a self-etch
technique, bonding to pathologically altered sub-
strates such as sclerotic dentine from noncarious
cervical lesions generally led to compromised
bonding.40 – 42,57 Reduced bonding efficacy was
attributed to a combination of factors that include
the obliteration of dentinal tubules with sclerotic
casts, the presence of an acid-resistant hypermi-
neralised layer, and the presence of bacteria on the
lesion surface. The presence of the hyperminer-
alised layer, bacteria and the tubular mineral casts
in sclerotic dentin are analogous to the presence of
the smear layer and smear plugs in sound dentine,
being potential diffusion barriers for primer and
resin-infiltration. The concern that a self-etching
primer may not etch through the superficial layers
F.R. Tay, D.H. Pashley
on sound dentine may likewise be applicable to
sclerotic dentine. Acid pre-conditioning prior to
the application of a self-etching primer may thus be
a viable technique when bonding to sclerotic
dentine.42 It has been shown that the hybrid layer
morphology after self-etching or wet bonding in
uninstrumented sclerotic dentine is substantially
different from those observed in sound, abraded
dentine.
Obstacles in bonding to acid-etched,
sound dentine
Adhesive strategies that rely mostly on microme-
chanical retention are hampered by obstacles that
jeopardise effective infiltration of resin into dental
tissues.97 In abraded sound dentine, the smear layer
is effectively removed in adhesive systems that
utilise a separate acid-conditioning and rinsing step
(Fig. 12). In order to achieve optimal resin-infiltra-
tion, acid-etched demineralised dentine must be
suspended in water to prevent collapsing of the
interfibrillar spaces. This is effectively accom-
plished using the wet bonding technique. Interfacial
strength is dependent upon the ability of resins to
engage the leading edge of demineralised inter-
tubular dentine.98 – 102 The collagen matrix may thus
be viewed as a diffusion barrier or obstacle for
resin-infiltration in acid-etched dentine.
Figure 12 Application of Clearfil Liner Bond 2V (Kur-
aray) to sound dentine (from an artificial wedge-shaped
lesion) that was etched with 40% phosphoric acid for 15 s.
The smear layer was completely removed. The hybrid
layer (H) was about 5 mm thick. Arrow: base of hybrid
layer; A: filled adhesive containing nanofillers; D: sound
dentine (from Kwong et al., 2000,27 with permission).
Resin bonding to cervical sclerotic dentin: A review 183

Obstacles in bonding to acid-etched

sclerotic dentine

Unlike sound dentine, application of the same

adhesive strategy to sclerotic dentine results in
substantial variation in both hybrid layer and resin
tag morphology. Potential obstacles of resin-infil-
tration into uninstrumented natural lesions include
the hypermineralised surface layer, an additional
partially mineralised surface bacterial layer and
intratubular mineral casts that are comparatively
more acid-resistant.23,25,27 As these inclusions vary
considerably along the occlusal, gingival, and the
deepest part of a wedge-shaped lesion, variation in
the ultrastructure of the resin – sclerotic dentine
interfaces are possible in these different regions.
While it is reasonable to assume that the extent of
tubular occlusion (Fig. 13A) would vary according to
the severity of dentine sclerosis,103 both the
superficial bacterial and hypermineralised layers
are found to vary from site to site, being thicker
along the deepest part of the wedge-shaped
lesions.42 The surface hypermineralised layer was
usually thinner in gingival and occlusal surfaces
than in the apical or deepest part of the wedge-
shaped defects, and was often partially or com-
pletely dissolved when phosphoric acid is applied to
sclerotic dentine (Fig. 13A). As a result, the
thickness of the hybrid layers in the gingival and
occlusal sites were similar to those seen in acid-
etched sound dentin and remained fairly consistent
at about 5 mm. Bacteria, if present, tended to be
tenaciously attached to the dentine surfaces and in Figure 13 (A) Application of One-Step (Bisco) to acid-
the dentinal tubules, and were retained even after etched sclerotic dentine (SD). This stained undeminer-
rinsing (Fig. 13B). alised section was taken from the gingival aspect of a
Thicker diffusion barriers were found within the wedge-shaped lesion. Remnants of a very thin, electron-
deepest part of wedge-shaped lesions that ham- dense, hypermineralised layer (arrowhead) could be seen
on the surface of the hybrid layer (H). Some tubules were
pered the penetration of acids through the under-
occluded with sclerotic casts (pointers). Others were
lying intact sclerotic dentine. As a result, patent (arrows) and filled with adhesive resin (A). (B)
alterations in hybrid layer morphology and thick- Stained, undemineralised TEM section showing appli-
ness were seen in these regions. Considering that cation of a filled adhesive to the occlusal aspect of an
the mean thickness of the hybrid layer was about acid-etched, sclerotic lesion. Bacteria (pointers) were
5 mm in sound dentine as well as the gingival/oc- trapped on the hybrid layer (Hd) surface and in the
clusal aspects of natural sclerotic lesions, hybrid tubules. Other tubules were filled with sclerotic casts
layer morphology in the deepest part of the wedge- (arrows).
shaped lesions may be described as ‘erratic’ in
appearance. whitlockite crystallites from the sclerotic casts may
One example is shown in Fig. 14A. The surface also be identified within dentinal tubules in the
hypermineralised layer from the deepest part of a underlying sclerotic dentine.
wedge-shaped lesion was about 3 mm thick and was Fig. 14B is a stained, demineralised TEM micro-
trapped within the resin – sclerotic dentine inter- graph taken from the deepest part of another acid-
face. The phosphoric acid apparently etched etched, wedge-shaped natural sclerotic lesion.
through the hypermineralised layer and created a The surface obstacle layers were either absent or
hybrid layer of about 2 mm thick within the completely dissolved. Within the 50 mm wide region
underlying intact sclerotic dentine. Rhombohedral of the micrograph, the thickness of the hybrid layer
184 F.R. Tay, D.H. Pashley

These ‘eccentric’ hybrid layers in abnormal dentin

substrates have never been observed in bonded
sound dentine.
Unlike sound dentine in which the morphology of
the hybrid layer remains fairly consistent as long as
a wet bonding technique is used, extensive vari-
ations are found in different specimens or locations
within cervical sclerotic lesions. Fig. 15A and B are

Figure 14 (A) Undemineralised section taken from the

deepest part of a wedge-shaped natural lesion that was
etched with 40% phosphoric acid and bonded using
Clearfil Liner Bond 2V. Hh: hybridised hypermineralised
layer that contained bacteria remnants (pointer). Hd:
hybridised sclerotic dentine that was about 2 mm thick. A:
adhesive containing nanofillers; SD: sclerotic dentine. (B)
Demineralised TEM micrograph of an ‘erratic’ hybrid
layer from the apex of a wedge-shaped lesion that was
etched with 40% phosphoric acid and bonded using
Clearfil Liner Bond 2V. The thickness of the hybrid layer
Figure 15 (A) Undemineralised TEM micrograph from
varied from 2 mm (pointers) to 5 mm (Hd). A lateral
the deepest part of an acid-etched, natural lesion. The
extension of the hybrid layer could be seen below an area
hybrid layer (Hd) in intact sclerotic dentine (SD) was 5 mm
that was not infiltrated with adhesive resin (asterisk). A:
thick. This was reduced to 2 mm thick beneath the
adhesive; SD: sclerotic dentine.
hypermineralised layer (HM). The separation of the
in sclerotic dentine changed abruptly from 2 to hypermineralised layer (arrow) was an artefact produced
5 mm. This aspect of uneven etching may also during specimen processing. A: filled adhesive. (B)
Demineralised TEM micrograph of an ‘erratic’ hybrid
be seen in areas in which the acid etches laterally
layer from the apex of a wedge-shaped lesion that was
within the subsurface sclerotic dentine,
etched with 40% phosphoric acid and bonded using
producing lateral hybrid layer extensions that Clearfil Liner Bond 2V. The thickness of the hybrid layer
are separated from areas above that are not varied from being absent (arrow) when a hyperminer-
infiltrated with resin. This feature is distinct from alised layer (HM) was present, to 5 mm (Hd) where the
incompletely resin-infiltrated hybrid layers that are latter was thin and was eroded by bacteria (B). A:
observed in air-dried, acid-etched sound dentine. adhesive; SD: sclerotic dentine.
Resin bonding to cervical sclerotic dentin: A review
demineralised, stained TEM micrographs taken
from the deepest part of the same natural lesion.
Whereas a thin hypermineralised layer and the
presence of bacteria did not prevent the pen-
etration of acid or resin into the underlying
sclerotic dentine, the thickness of the hybrid layer
was greatly reduced in the presence of a thick
hypermineralised layer (Fig. 15A). In some areas,
the hybrid layer in sclerotic dentine is reduced to
the extent that it is almost nonexistent (Fig. 15B).
These thick hypermineralised layers serve as
obstacles to diffusion and prevent the penetration
of even phosphoric acid. Reduced hybrid layer
thickness may have no correlation with reduced
regional microtensile bond strength in natural
sclerotic dentine.40,42,57 However, the presence of
thin hybrid layers should clearly be differentiated
from the total absence of hybrid layer formation in
the bonding substrate. Under such circumstances,
the adhesive will be bonding directly to the
diffusion barriers that impede acid-etching. The
resultant bond strength will depend on the strength
of the attachment of such obstacles to the
underlying sclerotic dentine. It is remarkable that
these morphological fluctuations are continuous
and vary within a very small region of a lesion that is
covered by these micrographs. Such extreme
variations in hybrid layer morphology are probably
responsible for the large standard deviations in
microtensile bond strength measurements of bond-
ing to sclerotic dentine. It is likely that these
segregated areas of deficient hybrid layer for-
mation act as weak links, or stress raisers, that
contribute to the initiation of adhesive failures in
bonded sclerotic dentine.
Obstacles to bonding in sound dentin treated
with a self-etching primer alone
The use of two-step and single-step self-etch
adhesives represents an alternative means to
acquire micromechanical retention in dentine.
They are attractive in that they may be used on
dry dentine and, after mixing, require only one
primer application, which is subsequently air-
dried rather than rinsed. The latest single-step
self-etch adhesives further incorporates all
the resin monomers, photoinitiator and tertiary
amine accelerator into a single bottle (iBond,
Heraeus Kulzer, Hanau, Germany) and eliminates
an additional mixing step. Despite the physical
appearance of thin hybrid layers and short,
hybridised smear plugs in a two-step self-etch
adhesive (Fig. 16), high initial bond strength has
been reported. This suggests that there is no
correlation between hybrid layer thickness and
185
Figure 16 Demineralised TEM micrograph of the appli-
cation of Clearfil Liner Bond 2V to sound dentine (artificial
wedge-shaped lesion). To achieve effective bonding, a
self-etching primer must not only create a hybridised
smear layer (Hs), but also a hybrid layer in the underlying
intact sound dentine (Ha). The dentinal tubule was sealed
by a hybridised smear plug (Hp). R: adhesive resin; D:
sound dentine.
bond strength as along as a uniform deminerali-
sation front is created in sound intertubular
dentine.89,98,104,105
There was concern, in normal dentine, that
thick and rough smear layers may interfere with
diffusion of self-etching primers into the under-
lying intact dentine. This may be due to the
physical presence of thick smear layers as a
diffusion barrier, or their ability to buffer the
acidic monomers, making the pH too high to
demineralise the underlying intertubular dentine.
Recent studies showed that mildly aggressive self-
etching adhesives penetrated through smear
layers up to 3 – 4 mm thick and still retained
sufficient acidity to demineralise the underlying
intertubular dentine to a depth of 0.4 – 0.5 mm.106
This suggests that either the buffering capacity of
the smear layer is weak, or the smear layer does
not impose much of a physical barrier to the
primer compared with the underlying mineralised
dentine matrix. The looseness of the surface
portion of the smear layer and/or the presence
of diffusion channels between its constituents may
facilitate diffusion of the self-etching primer
through the smear layer.107 Disaggregation of the
smear layer into globular subunits108 further
provides microchannels for diffusion of the self-
etch adhesives. These microchannels, in theory,
should be more permeable to resin monomers
than the interfibrillar spaces (ca. 20 nm) of
demineralised intertubular dentine.
186 F.R. Tay, D.H. Pashley

Obstacles in bonding to self-etching primer

treated sclerotic dentine

The smear layers in sound, cut dentine do not

impose much of a restriction to the bonding of
contemporary self-etch adhesives because of their
loose organisation. Unless they are instrumented,
shiny sclerotic cervical lesions are free of smear
layers. In lieu of smear layers, other diffusion
barriers in shiny cervical sclerotic lesions include
the much denser surface hypermineralised layer, as
well as the more loosely arranged, partially miner-
alised bacterial clusters. Some of the hyperminer-
alised layers in sclerotic dentine are so thick that
they restrict the penetration of strong inorganic
acids such as phosphoric acid that usually etches
5 mm or beyond into sound dentine. Being less
acidic in nature, a mild self-etching primer such as
Clearfil Liner Bond 2V (Kuraray) etches only 0.5 mm
into sound dentine that are covered with smear
layers. As sclerotic dentine is highly variable in its
ultrastructure, different morphologic expressions
of the resin –dentine interfaces can be anticipated
when a mild self-etching primer is applied to this
abnormal bonding substrate. This may be classified
into three categories, depending on the in terms of
its thickness and continuity of the surface hypermi-
neralised layer at different locations of the wedge-
shaped lesion.

Sclerotic dentine with a thin hypermineralised

layer (<0.5 mm thick)
Thin hypermineralised layers are usually located
along the gingival and occlusal aspects of wedge-
shaped lesions. Fig. 17A represents a straightfor-
ward situation in which a very thin hypermineralised
layer is present, without bacteria, along the occlusal
aspect of a natural wedge-shaped lesion. This layer
may be recognised by the characteristic arrange-
ment of the partially dissolved crystallite remnants.
The difference in hybrid layer morphology that
results from uneven etching is readily apparent. On
the left side, the effect of the self-etching primer is
restricted to the surface layer alone, producing a
0.1 mm thick, hybridised hypermineralised layer. On
the right, the primer etched beyond the surface
layer to form an additional layer of hybridised
dentine that was about 0.5 mm thick.
A similar uneven etching effect may be seen in
thin hypermineralised layers that contain additional
surface bacterial attachments (Fig. 17B). This
complicates etching by the fact that the self-
etching primer must first infiltrate through the
matrix between the bacteria, and then through the
hypermineralised layer in order to demineralise
Figure 17 (A) Demineralised TEM micrograph of the
gingival aspect of a wedge-shaped lesion that was treated
with Clearfil Liner Bond 2V (A). Uneven etching was
evident. On the left, only a hybridised hypermineralised
layer (Hh) was produced, within which were plate-like
crystallite remnants (pointers). On the right, a 500 nm
thick layer of hybridised sclerotic dentine (Hd) was
formed beneath. SD: sclerotic dentine. (B) The bonded
interface from the occlusal aspect of a sclerotic lesion
that contained a thin hypermineralised layer but heavy
surface bacterial deposits. The hybridised complex
consisted of: (1) a hybridised intermicrobial matrix
(Him); (2) a hybridised hypermineralised layer; and (3) a
hybridised layer of intact sclerotic dentine. (Hd). The
latter was absent on the right side of the micrograph
(arrows). SD: sclerotic dentine.
the underlying intact sclerotic dentine. Unlike the
hypermineralised layer that consists of densely
arranged crystallites, the intermicrobial matrix is
more easily penetrable by the self-etching primer.
The hybridised complex, thus consisted of three
components: the hybridised intermicrobial matrix,
the hybridised hypermineralised layer and the layer
Resin bonding to cervical sclerotic dentin: A review
Figure 18 (A) The hybridised complex of Clearfil Liner
Bond 2V to sclerotic dentine in this micrograph consisted
of only the hybridised intermicrobial matrix (Hm) and
hybridised, partially dissolved, hypermineralised layer
(Hh). Crystallite remnants, arranged in a parallel orien-
tation, were observed within the latter. No hybridised
dentine was found within the underlying sclerotic dentine
(SD). (B) This micrograph was taken from the same
section as Fig. 18A. An additional layer of hybridised
sclerotic dentine (Hd) could be observed. B: bacteria; A:
filled adhesive; SD: sclerotic dentine.
of hybridised dentine. In this micrograph, the layer
of hybridised dentine was about 0.5 – 0.8 mm on the
left, but was completely absent on the right. This
illustrates the kind of biological variation that may
be expected along the entire lesion surface. The
features are not the rare, one of a kind phenom-
enon that is observed only in one single lesion.
Fig. 18A and B are high magnifications of similar
features observed in another specimen. There were
taken from the same 1 mm £ 1 mm section of the
gingival aspect of a bonded sclerotic lesion.
187
Sclerotic dentine with a thick, continuous
hypermineralised layer (>0.5 mm)
Thicker hypermineralised layers are found along the
occlusal aspects and sometimes within the deepest
part of wedge-shaped, sclerotic lesions. As the
etching effect of a self-etching primer is limited, it
cannot etch beyond the hypermineralised layer into
the underlying sclerotic dentine (Fig. 19A). At a
higher magnification, the crystallites within the
bonded hypermineralised layer are shorter and
more sparsely arranged, when compared with the
underlying unaffected hypermineralised layer
(Fig. 19B). This partially demineralised zone was
about 0.5 mm in depth. Porosities created for resin
infiltration within this zone are reminiscent of acid-
etched, aprismatic enamel. It has been shown that
bonding to the surface hypermineralised layer alone
resulted in relatively high bond strength.42 The
ultimate strength of the entire bonded assembly,
however, depends on the strength of the attach-
ment of the hypermineralised layer to the under-
lying sclerotic dentine.
Partial demineralisation of the surface hypermi-
neralised layer also occurs in the presence of
bacteria inclusions. In Fig. 19C, an unstained
undemineralised TEM micrograph, silhouettes of
the unstained bacteria could be seen above the
partially demineralised, hypermineralised layer.
The depth of demineralisation within this layer is
comparable to the action of this self-etching primer
on sound dentine, and was about 0.5 mm thick. It is
reasonable to assume that a layer of hybridised
dentine cannot be formed in the underlying dentine
when the surface hypermineralised layer is thicker
than 0.5 mm. It appears that the self-etching primer
can easily diffuse through the intermicrobial
matrix. When such a complex is subjected to tensile
stress, it remains to be seen whether bond failure
would occur between the resin infiltrated bacteria,
the partially infiltrated hypermineralised layer, or
between the base of the hypermineralised layer and
the underlying dentine.
Sclerotic dentine with a thick, discontinuous
hypermineralised layer (>0.5 mm)
These thick hypermineralised layers are found along
the apex or deepest part of wedge-shaped lesions. As
previously described, they consist of thin, discon-
tinuous hypermineralised layers that are dispersed
among several different colonies of bacteria. These
hypermineralised layers are probably intercon-
nected to form a three-dimensional structure
(Fig. 20A). It is highly unlikely that a self-etching
primer can etch through a discontinuous hypermi-
neralised layer that is 10 –15 mm thick. At a higher
magnification, Fig. 20B shows the demineralisation
188
Figure 19 (A) Undemineralised TEM micrograph taken
from the gingival aspect of a Clearfil Liner 2-treated
sclerotic lesion. There was only partial demineralisation
(pointer) of the hyperminerlised layer (HM), with the
latter still attached to the underlying sclerotic dentine
(SD). Dentinal tubules were heavily obliterated with
sclerotic casts (arrow). A: filled adhesive. (B) A higher
magnification of Fig. 19A, showing the crystallites
F.R. Tay, D.H. Pashley
front represented by the scalloped electron-dense
edge of the partially demineralised intermicrobial
matrix. It is anticipated that bond failure occurs
within the porous regions of this thick layer that is
occupied by bacteria but not infiltrated with
adhesive resin.
Summary of obstacles in bonding to sound vs.
sclerotic dentine
Potential deterrents to resin-infiltration following
total-etching or self-etching in sound and sclerotic
dentine are summarised schematically in Fig. 21.27
The left side of the figure illustrates the response to
bonding with a self-etching primer adhesive system
alone, while the right side illustrates the response to
pre-etching sclerotic dentine with 40% phosphoric
acid prior to bonding with the same self-etching
primer (Clearfil Liner Bond 2V). The thickness of the
hybrid layer is fairly consistent both for self-etch and
wet-bonded, acid-etched sound dentine, but is much
thicker in the latter group. Conversely, application
of the same adhesive strategy to sclerotic dentine
results in substantial variation in the hybrid layer
morphology in both treatment techniques. Absence
of a hybrid layer in some parts of a lesion suggests
that both treatment protocols are ineffective in
completely overcoming the diffusion barriers in
sclerotic dentine. This situation is comparable to
the early generation dentine adhesives that were
directly applied to the smear layers in sound
dentine.78 Similar to the junction between the
partially infiltrated smear layer and the underlying
intact sound dentine,109 areas devoid of hybrid layer
formation are potential weak links that may be
responsible for the lower bond strengths observed
when boning to sclerotic dentine.
Although reduction in hybrid layer thickness
may not affect micromechanical retention, spora-
dic absence of the hybrid layer and resin tags
indicate that both treatment techniques are
(pointer) within the partially demineralised zone of the
surface hypermineralised layer (HM). The ample poros-
ities thus created within this layer for micromechanical
retention of the adhesive is comparable to that of acid-
etched aprismatic enamel. SD: sclerotic dentine.
C. Undemineralised TEM micrograph taken from the
gingival aspect of a Clearfil Liner 2V-treated sclerotic
lesion. The self-etching primer diffused through the
unstained bacteria layer (B) and partially demineralised
(pointer) and infiltrated the superficial portion of the
hypermineralised layer (HM). The basal portion of the
hypermineralised layer appeared to be firmly integrated
with the underlying sclerotic dentine (SD). A: filled
adhesive.
Resin bonding to cervical sclerotic dentin: A review
Figure 20 (A) Bonding of Clearfil Liner Bond 2V to a
thick, discontinuous layer of hypermineralised dentine
(HM). The presence of a segregated piece of hyperminer-
alised tissue (arrow) suggested that the discontinuous
layers are probably interconnected three-dimensionally,
with the porous regions occupied by different colonies of
bacteria. Dentinal tubules (pointer) within the sclerotic
dentine (SD) were patent and were arranged parallel to
the surface. (B) A higher magnification of a similar lesion.
The action of Clearfil Liner Bond 2V was limited to the
superficial part of this porous layer. The demineralisation
front was represented by the electron-dense, scalloped
junction of the partially demineralised intermicrobial
matrix (arrows). Other discontinuous layers of hypermi-
neralised tissues could be seen (pointers) that may be
connected outside the plane of this section.
inadequate in overcoming diffusion barriers in
sclerotic dentine. Physical removal of the super-
ficial obstacle layers with a bur may improve
intertubular retention. In highly sclerotic lesions
however, this may be offset by moving the
bonding interface pulpward into an area where
bonding requires increasing contribution from
189
Figure 21 A schematic representation of the potential
deterrents to resin-infiltration during total-etching or
self-etching in sound and sclerotic dentine. The left side
illustrates the responses to bonding with a self-etching
primer adhesive system alone, while the right side
illustrates the responses to pre-etching sclerotic dentine
with 40% phosphoric acid prior to bonding with the same
self-etching primer adhesive (Clearfil Liner Bond 2V)
(modified from Kwong et al., 2000,27 with permission).
intratubular resin infiltration. Moreover, the for-
mation of a smear layer that consists of acid-
resistant hypermineralised dentine chips and
whitlockite crystals derived from the sclerotic
casts also creates additional diffusion barriers for
both total-etch and self-etch adhesives.
Sclerotic dentine located at the apex of wedge-
shaped natural lesions is derived from deep
dentine. Consequently, resin tag formation should
play an important role in achieving strong immedi-
ate bond strength in the sclerotic cervical lesion.
However, resin tag formation is sporadic regardless
of the conditioning methods. Absence of intratub-
ular infiltration may even be observed in some of
the occlusal parts of natural lesions that were
etched with phosphoric acid, where the thickness
of intertubular infiltration is comparable to that
190 F.R. Tay, D.H. Pashley

present in sound dentine. Similar to the results of

Ferrari et al.110 and Prati et al.,41 lateral branches
of resin tags were rarely observed when bonded
sclerotic dentine were examined by TEM.42,57 This
is likely to be caused by the acid-resistant nature
of the mineral-dense sclerotic casts that occlude
the dentinal tubules. It has been suggested that
20% of the strength of an interfacial bond was
contributed by resin-infiltration derived from resin
tag formation and another 20% from hybridisation
of the intertubular dentine.111,112 Regional tensile
bond strength from cervical sclerotic root dentine
was found to be 20 – 45% lower than those obtained
from artificial lesions prepared in sound root
dentine.40,42 This reduction may be due to the
absence of resin tags and incomplete hybridisation
in sclerotic dentine.

Regional microtensile bond strength

evaluation

Microtensile bond strength measurements compar-

ing resin bonds to the occlusal, gingival and the
apex or deepest part of natural lesions and
artificially wedge-shaped defects created in sound
cervical dentine were reported by Kwong et al.42
using the microtensile bond test.113 Lesions were
restored with Protect Liner (Kuraray) and Clearfil
AP-X resin composite (Kuraray) following treatment
with Clearfil Liner Bond 2V, with or without
phosphoric acid pre-conditioning of the lesions.
Using the nontrimming technique developed by
Shono et al.114 beams with a mean area of
0.46 ^ 0.03 mm 2 were prepared and stressed
to failure. The use of a nontrimming technique
(Fig. 22) facilitated preparation of a series of slabs,
thus allowing more than one beam to be harvested
from each lesion.113
The mean tensile bond strengths of bonds
produced by the self-etching primer alone to natural
lesions (48.7 MPa) were 26% lower (Fig. 23) than
those from artificial lesions (65.8 MPa) when all of
the bonds were pooled, and the result was statisti-
cally significant ðp , 0:001Þ: Similarly pooled data on
bonds made using the self-etching primer with
adjunctive phosphoric acid pre-conditioning to
natural lesions (53.1 MPa) were 24% lower ðp ,
0:005Þ than those produced from artificial lesions
(69.8 MPa). Pooled data, however, showed no
significant difference among the bonds made by
self-etching or total-etching to either sound dentine
ðp ¼ 0:415Þ or sclerotic dentine ðp ¼ 0:314Þ: Of the
three factors (substrate, conditioning method and
location) tested, only the difference in the type of
Figure 22 Schematic representation of the protocol
employed for regional microtensile bond strength
evaluation.
substrate (i.e. sound dentine vs. sclerotic dentine)
was found to have a significant influence on bond
strength ðp , 0:05Þ: Multiple comparison tests
showed that there was no difference in self-etching
or total-etching sclerotic dentine except for the
gingival aspect of the lesions, in which higher bond
strengths were obtained for total-etching (Fig. 23).
These results are comparable to those of
Yoshiyama et al.40 in that lower bond strengths
were found in natural sclerotic lesions, and to the
work of Phrukkanon et al.115 showing that bonding
of a self-etching primer to sound dentine is
independent of the tubular orientation. The orien-
tation of dentinal tubules in the occlusal or upper
wall of wedge-shaped lesions is approximately
parallel to the surface, while their orientation in
the gingival wall is perpendicular to the prepared
surface.116,117 Many believed that resin tag for-
mation would be more prominent in surfaces where
Resin bonding to cervical sclerotic dentin: A review
Figure 23 Regional microtensile bond strengths in self-
etching and total-etching sound (light shade bars) or
sclerotic (dark bars) dentine. Groups with the same letter
above the bars were not statistically significant ðp ,
0:05Þ: Number of teeth from which slabs were made in
each group ¼ 10.
the tubules are oriented perpendicular to the
surface rather than parallel. However, measure-
ment of microtensile bond strengths of self-etching
primers and total-etch adhesives to these walls in
wedge-shaped cavities prepared in normal dentine
revealed significantly higher bond strengths to
dentine in which the bonded surfaces were oriented
parallel to the tubules (i.e. occlusal walls).116,117
Double-etching of dentin by phosphoric acid
followed by a self-etching primer adhesive has
been shown to increase bond strengths to enamel
but to lower bond strengths in dentine,118 that may
be caused by incomplete infiltration of the adhesive
into the phosphoric acid-etched dentine.119 Using
Clearfil Liner Bond 2, Ogata et al.120 found that
multiple applications of the primer to wedge-
shaped lesions increased bond strength due to the
weak acidity of the primer. Although this has not
been tested in sclerotic dentine, the same result
would be expected. That is, multiple applications
of weakly acidic agents using constant agitation,
should improve bonding.
TEM examination of the failed bonds exhibited by
bonds created in sclerotic dentine revealed a wide
variation in the mode of failure that included all of
the different structural components that are pre-
sent within the resin –sclerotic dentine interfaces.42
The complexity of failure modes indicates that
191
reduced bond strength in sclerotic dentine is not
related to any single factor. Similar to other
biological variations, it is possible that each factor
contributes to a variable degree in different lesions.
The summation of all these factors, however, leads
to an overall reduction in bond strength.
The presence of a partially mineralised bacterial
zone in sclerotic lesions is analogous to the
presence of a smear layer on sound, abraded
dentine. This zone is porous, allowing easy pen-
etration of acids and primers to form a zone of
hybridised intermicrobial matrix. The presence of a
hybridised intermicrobial matrix may not affect
bonding, at least in the short term, providing that
the self-etching primer can effectively etch through
this layer into the underlying bonding substrate.
This is analogous to the hybridised smear layers in
sound dentine. It remains to be seen whether the
eventual degradation of the bacteria in this layer
would lead to decrease in bond strength with time.
The large standard deviation in bond strength
results in natural sclerotic lesions may simply
reflect the large biological variation in the thick-
ness of such a layer.
The presence of a hybridised hypermineralised
layer together with an underlying zone of hybri-
dised dentine does not necessarily result in low
bond strength. This is comparable to the infiltration
of a self-etching primer through smear layer-
covered sound dentine. Provided that the acids
can penetrate the overlying diffusion barriers to
engage the underlying substrate with even a very
thin hybrid layer, strong initial bonds may still be
achieved. However, erratic bonds may be expected
when the hypermineralised layer in sclerotic dentin
is too thick for acids to etch through. Resin
attachment to the partially demineralised surface
of this layer is still strong and may be comparable
with bonding to unground, aprismatic enamel.121
However, since a layer of hybridised dentine is not
produced, the strength of the bond will be highly
dependent upon the strength of the hyperminer-
alised layer to the intact sclerotic dentine.
The fact that higher bond strength was observed
along the gingival site of total-etched natural
lesions (Fig. 23) merits further discussion. This
suggests that the inability of the adhesive to form
resin tags in tubule lumina that are blocked by
mineral deposits is an important parameter that
leads to the reduction in bond strength. If the above
hypothesis is correct, then grinding of the surface
hypermineralised layer of these cervical wedge-
shaped defects prior to bonding122 should not result
in an increase in bond strength, since the underlying
sclerotic dentine still contains dentinal tubules that
are blocked by whitlockite crystallites. Application
192
of stronger phosphoric acid to these defects could
have resulted in partial dissolution of the sclerotic
casts and/or complete removal of the surrounding
peritubular dentine, allowing resin infiltration into
the dentinal tubules. This may result in higher bond
strength along the gingival site of phosphoric acid-
etched, sclerotic dentine.
Restoring the class v sclerotic lesion
Maintaining the marginal integrity and retention of
Class V resin composite restorations without the use
of additional retention has always been a challenge
for clinicians. One major factor, already analysed is
the difficulty in bonding to sclerotic dentine.
Removing the hypermineralised surface layers by
grinding or by using stronger acids (Fig. 23) are
possible strategies to improve micromechanical
retention in sclerotic dentine. While it is possible
to produce hybrid layers in sclerotic dentine with
thin diffusion barriers, these hybrid layers become
erratic or even nonexistent in the presence of thick
barriers. As clinicians have no way of discerning
these differences at a clinical level, removal of the
surface layer of sclerotic dentin prior to bonding
should be adopted.42,122 Although such a rec-
ommendation may not result in an increase in
bond strength to sclerotic dentin, it does remove
one potential source of inconsistency that leads to
bond failure. Based on the results of a two-year
clinical trial, it has been suggested that micro-
mechanical retention by acid etching of the enamel
margin is still indispensable for the clinical success
of cervical Class V composite restorations.123 Such a
concept, however, was recently challenged by two
Japanese studies, in which the authors maintained
that sclerotic dentine, being a part of the body’s
natural defence mechanism, should be preserved as
much as possible and that acid-etching should be
avoided to promote the marginal integrity of resin
composites that are bonded to these lesions.45,124
While one may remove bacteria overgrowths
from the surface hypermineralised layer, it is not
possible to remove bacteria entirely from dentinal
tubules. This is analogous to the application of
fissure sealants to stained enamel fissures,125,126 or
the bonding of resins to the inner layer of carious
dentine.127,128 The use of bactericidal solutions
(i.e. chlorhexidine) or adhesive resins with anti-
bacterial activity129,130 would be helpful. However,
the longevity of bonds that contain dead, degrad-
able bacteria should be further investigated. This is
particularly applicable to adhesives that contain an
increasing amount of hydrophilic resin monomers
F.R. Tay, D.H. Pashley
that absorb water. Recent studies showed that both
hydrophilic resins131,132 and collagen fibrils within
the hybrid layer133 – 136 degrade upon long term
water storage.137
Conclusions
The structural complexity of noncarious sclerotic
cervical dentine is remarkable. The common pre-
sence of adherent bacteria on such surfaces and
their incorporation into the bonded restorations is
disconcerting. This raises issues such as whether
these bacteria are dormant, and whether their
confinement by adhesives will create any long term
liability. These questions have also recently been
raised in reports that bacteria are present in resin-
bonded caries-affected dentine.127,128 The pre-
sence of bacteria on these surfaces justifies the
use of 2% chlorhexidine disinfectant treatment or
the use of antibacterial adhesives to disinfect the
substrate prior to bonding.
Microtensile bond strengths of self-etching
primers to sclerotic dentine were comparable
with those made to phosphoric acid-etched
sclerotic dentine, although they were lower
than those attained with sound dentine. These
bond strengths are probably high enough to
retain class V restorations even under heavy
loads, if is bonding is performed with etching of
the enamel to create additional micromechanical
retention. Although there are clinical studies that
showed encouraging results with the use of
dentine adhesives on noncarious cervical
lesions,138 – 140 the failure rates of some specific
adhesives have been reported to be high in other
studies. For example, the retention rate for One-
Step, a total-etch single-bottle adhesive, in
noncarious cervical lesions involving sclerotic
dentine was only half of that of nonsclerotic
dentine.141 The retention rate of the same
adhesive in noncarious cervical lesions was
reduced from the original 100% at 6 months to
75% after three years.142 There are also other
clinical studies that reported more favourable
retention rates when glass-ionomer based
restorative materials were compared with den-
tine adhesives/resin composites in restoring these
lesions.143 – 145 Unfortunately, there are no TEM
studies that examine the bonding of glass-
ionomer cements and resin-modified glass-iono-
mer cements to sclerotic dentine. Admittedly,
TEM work on these types of restorative materials
that are susceptible to dehydration is difficult to
perform. However, this should be done to
Resin bonding to cervical sclerotic dentin: A review
complete our understanding of this alternative
type of chemical/micromechanical interaction70
with sclerotic dentine.
Acknowledgements
This work was supported by grant DE014911 from
the NIDCR, USA, and by grant
20003755/90800/08004/400/01, Faculty of Dentis-
try, the University of Hong Kong. The authors are
grateful to Michelle Barnes and Zinnia Pang for
secretarial support.
References
1. Zsigmondy U. Über die keilförmigen Defekte an den Facial-
flächen der Zajnhälse. Österr Ungar Vjhrschr Zahnärzte 1894;
1:439—442.
2. Miller WD. Experiments and observations on the wasting of
tooth tissue variously designated as erosion, abrasion,
chemical abrasion, denudation, etc. Dental Cosmos 1907;
49:1—23. see also pp. 109—124; 225—247; 677—689.
3. Braem M, Lambrechts P, Vanherle G. Stress-induced cervical
lesions. The Journal of Prosthetic Dentistry 1992;67:
718—722.
4. Levitch LC, Bader JD, Shugars DA, Heymann HO. Non-
carious cervical lesions. Journal of Dentistry 1994;22:
195—207.
5. Gallien GS, Kaplan I, Owens BM. A review of noncarious dental
cervical lesions. Compendium 1994;15:1366—1374.
6. Spranger H. Investigation into the genesis of angular lesions at
the cervical region of teeth. Quintessence International 1995;
26:149—154.
7. Tyas MJ. The Class V lesion: aetiology and restoration.
Australian Dental Journal 1995;40:167—170.
8. Aw TC, Lepe X, Johnson GH, Mancl L. Characteristics of
noncarious cervical lesions: a clinical investigation. Journal
of American Dental Association 2002;133:725—733.
9. Giachetti L, Ercolani E, Bambi C, Landi D. Sclerotic dentin:
aetio-pathogenetic hypotheses. Minerva Stomatologica 2002;
51:285—292.
10. Xhonga FA. Bruxism and its effect on the teeth. Journal of
Oral Rehabilitation 1977;4:65—76.
11. Lee WC, Eakle WS. Possible role of tensile stress in the
etiology of cervical erosive lesions of teeth. The Journal of
Prosthetic Dentistry 1984;2:374—380.
12. Lee WC, Eakle WS. Stress-induced cervical lesions: review of
advances in the past 10 years. The Journal of Prosthetic
Dentistry 1996;75:487—494.
13. Grippo JO. Abfractions: a new classification of hard tissue
lesions of teeth. Journal of Esthetic Dentistry 1991;3:
14—19.
14. Burke FJT, Whitehead SA, McCaughey AD. Contemporary
concepts in the pathogenesis of the Class V non-carious
lesion. Dental Update 1995;22:28—32.
15. Rees JS. The role of cuspal flexure in the development of
abfraction lesions: a finite element study. European Journal
of Oral Science 1998;106:1028—1032.
16. Rees JS. A review of the biomechanics of abfraction.
European Journal of Prosthodontics and Restorative Den-
tistry 2000;8:139—144.
193
17. Rees JS. The effect of variation in occlusal loading on the
development of abfraction lesions: a finite element study.
Journal of Oral Rehabilitation 2002;29:188—193.
18. Eramo S, Baldi M, Marci MC, Monaco A. Histopathological and
therapeutical aspects of cervical lesions. Minerva Stomato-
logica 2003;52:69—74.
19. Kuroe T, Itoh H, Caputo AA, Konuma M. Biomechanics of
cervical tooth structure lesions and their restoration.
Quintessence International 2000;1:267—274.
20. Palamara D, Palamara JE, Tyas MJ, Messer HH. Strain
patterns in cervical enamel of teeth subjected to occlusal
loading. Dental Materials 2000;16:412—419.
21. Lee HE, Lin CL, Wang CH, Cheng CH, Chang CH. Stresses at
the cervical lesion of maxillary premolar: a finite element
investigation. Journal of Dentistry 2002;30:283—290.
22. Rees JS, Hammadeh M, Jagger DC. Abfraction lesion
formation in maxillary incisors, canines and premolars: a
finite element study. European Journal of Oral Science 2003;
111:149—154.
23. Gwinnett AJ, Jendresen M. Micromorphological features of
cervical erosion after acid conditioning and its relation with
composite resin. Journal of Dental Research 1978;7:
543—549.
24. Harnirattisai C, Inokoshi S, Shimada Y, Hosoda H. Adhesive
interface between resin and etched dentin of cervical
erosion/abrasion lesions. Operational Dentistry 1993;18:
138—143.
25. Van Meerbeek B, Braem M, Lambrechts P, Vanherle G.
Morphological characterization of the interface between
resin and sclerotic dentine. Journal of Dentistry 1994;22:
141—146.
26. Yoshiyama M, Suge T, Kawasaki A, Ebisu S. Morphological
characterization of tube-like structures in hypersensitive
human radicular dentine. Journal of Dentistry 1996;24:
57—63.
27. Kwong SM, Tay FR, Yip HK, Kei LH, Pashley DH. An
ultrastructural study of the application of dentine adhesives
to acid-conditioned sclerotic dentine. Journal of Dentistry
2000;7:515—528.
28. Sakoolnamarka R, Burrow MF, Prawer S, Tyas MJ. Micro-
morphological investigation of noncarious cervical lesions
treated with demineralizing agents. Journal of Adhesive
Dentistry 2000;2:279—287.
29. Nalbandian J, Federico G, Sognnaes RF. Sclerotic age
changes in root dentin of human teeth as observed by
optical, electron, and x-ray microscopy. Journal of Dental
Research 1959;39:598—607.
30. Weber DF. Human dentine sclerosis: a microradiographic
survey. Archives of Oral Biology 1974;19:163—169.
31. Yoshiyama M, Noiri Y, Ozaki K, Uchida A, Ishikawa Y, Ishida
H. Transmission electron microscopic characterization of
hypersensitive human radicular dentin. Journal of Dental
Research 1990;9:1293—1297.
32. Mixson JM, Spencer P, Moore DL, Chappell RP, Adams S.
Surface morphology and chemical characterization of
abrasion/erosion lesions. American Journal of Dentistry
1995;8:5—9.
33. Eda S, Hasegawa H, Kawakami T. Crystals closing tubules in
sclerosed root dentin of the aged. In: Shimono M, Maeda T,
Suda H, Takahashi K, editors. Proceedings of the Inter-
national Conference on Dentin/Pulp Complex 1995 and
the International Meeting on Clinical Topics of Dentin/Pulp
Complex. Osaka: Quintessence; 1996. p. 283—284.
34. Kawakami T, Takei N, Eda S. Crystals closing tubules in
sclerosed coronal dentin of the aged. In: Shimono M, Maeda
T, Suda H, Takahashi K, editors. Proceedings of the
International Conference on Dentin/Pulp Complex 1995
194
and the International Meeting on Clinical Topics of
Dentin/Pulp Complex. Osaka: Quintessence; 1996.
p. 285—286.
35. Tyas MJ. Clinical performance of three dentine bonding
agents in Class V abrasion lesions without enamel etching.
Australian Dental Journal 1987;33:177—180.
36. Duke ES, Lindemuth J. Polymeric adhesion to dentin:
contrasting substrates. American Journal of Dentistry
1990;3:24—270.
37. Duke ES, Lindemuth J. Variability of clinical dentin
substrates. American Journal of Dentistry 1991;4:
241—246.
38. Duke ES, Robbins JW, Snyder DS. Clinical evaluation of a
dentinal adhesive system: three-year results. Quintessence
International 1991;22:889—895.
39. Van Meerbeek B, Peumans M, Gladys S, Braem M, Lambrechts
P, Vanherle G. Three-year clinical effectiveness of four
total-etch dentinal adhesive systems in cervical lesions.
Quintessence International 1996;27:775—784.
40. Yoshiyama M, Sano H, Ebisu S, Tagami J, Ciucchi B, Carvalho
RM, Johnson MH, Pashley DH. Regional strengths of bonding
agents to cervical sclerotic root dentin. Journal of Dental
Research 1996;75:1404—1413.
41. Prati C, Chersoni S, Mongiorgi R, Montanari G, Pashley DH.
Thickness and morphology of resin-infiltrated dentin layer in
young, old, and sclerotic dentin. Operational Dentistry
1999;24:66—72.
42. Kwong SM, Cheung GS, Kei LH, Itthagarun A, Smales RJ, Tay
FR, Pashley DH. Micro-tensile bond strengths to sclerotic
dentin using a self-etching and a total-etching technique.
Dental Materials 2002;18:359—369.
43. Sakoolnamarka R, Burrow MF, Tyas MJ. Micromorphological
study of resin—dentin interface of non-carious cervical
lesions. Operational Dentistry 2002;27:493—499.
44. Gwinnett AJ, Kanca JA. Interfacial morphology of resin
composite and shiny erosion lesions. American Journal of
Dentistry 1992;5:315—317.
45. Kusunoki M, Itoh K, Hisamitsu H, Wakumoto S. The efficacy of
dentine adhesive to sclerotic dentine. Journal of Dentistry
2002;30:91—97.
46. Daculsi G, LeGeros RZ, Jean A, Kerebel B. Possible physico-
chemical processes in human dentin caries. Journal of
Dental Research 1987;66:1356—1359.
47. Yoshiyama M, Masada J, Uchida A, Ishida H. Scanning
electron microscopic characterization of sensitive versus
insensitive human radicular dentin. Journal of Dental
Research 1989;68:1498—1502.
48. Schüpbach P, Lutz F, Guggenheim B. Human root caries:
histopathology of arrested lesions. Caries Research 1992;26:
153—164.
49. Scheie AA, Fejerskov O, Lingström P, Birkhed D, Manji F. Use
of palladium touch microelectrodes under field conditions
for in vivo assessment of dental plaque pH in children. Caries
Research 1992;26:44—52.
50. Svensäter G, Björnsson O, Hamilton IR. Effect of carbon
starvation and proteolytic activity on stationary-phase acid
tolerance of Streptococcus mutans. Microbiology 2001;147:
2971—2979.
51. Spatafora GA, Sheets M, June R, Luyimbazi D, Howard K,
Hulbert R, Barnard D, el Janne M, Hudson MC. Regulated
expression of the Streptococcus mutans dlt genes correlates
with intracellular polysaccharide accumulation. Journal of
Bacteriology 1999;181:2363—2372.
52. Sanz M, Newman MG. Dental plaque and calculus. In:
Nisengard RJ, Newman MG, editors. Oral microbiology
and immunology. Philadelphia: WB Saunders; 1994.
p. 320—340.
F.R. Tay, D.H. Pashley
53. Katzberg AA. The induction of cellular orientation by low-
level electrical currents. Annals of the New York Academy of
Sciences 1974;238:445—450.
54. Park C, Robertson RE. Mechanical properties of resin
composites with filler particles aligned by an electric field.
Dental Materials 1998;14:385—393.
55. Marino AA, Gross BD. Piezoelectricity in cementum, dentine
and bone. Archives of Oral Biology 1989;34:507—509.
56. Chen CS, Pohl HA. Biological dielectrophoresis: the
behaviour of lone cells in a nonuniform electric field.
Annals of the New York Academy of Sciences 1974;238:
176—185.
57. Tay FR, Kwong SM, Itthagarun A, King NM, Yip HK,
Moulding KM, Pashley DH. Bonding of a self-etching primer
to noncarious cervical sclerotic dentin: interfacial ultra-
structure and micro-tensile bond strength evaluation.
Journal of Adhesive Dentistry 2000;2:9—28.
58. Ishikawa K, Suge T, Yoshiyama M, Kawasaki A, Asaoka K,
Ebisu S. Occlusion of dentinal tubules with calcium phos-
phate using acid calcium phosphate solution followed by
neutralization. Journal of Dental Research 1994;73:
1197—1204.
59. Takuma S, Ogiwara H, Suzuki H. Electron probe and
electron microscope studies of carious dentinal lesions
with remineralized surface layer. Caries Research 1975;9:
278—285.
60. Daculsi G, Kerebel B, Le Cabellec MT, Kerebel LM.
Qualitative and quantitative data on arrested caries in
dentine. Caries Research 1979;13:190—202.
61. Tohda H, Fejerskov O, Yanagisawa T. Transmission electron
microscopy of cementum crystals correlated with Ca and F
distribution in normal and carious human root surfaces.
Journal of Dental Research 1996;75:949—954.
62. Clarkson BH, Feagin FF, McCurdy SP, Sheetz JH, Speirs R.
Effect of phosphoprotein moieties on the remineralization of
human root caries. Caries Research 1991;25:166—173.
63. Arends J, Christoffersen J, Ruben J, Jongebloed WL.
Remineralization of bovine dentine in vitro. The influence
of the F content in solution on mineral distribution. Caries
Research 1989;23:309—314.
64. Damen JJ, Buijs MJ, ten Cate JM. Fluoride-dependent
formation of mineralized layers in bovine dentin during
demineralization in vitro. Caries Research 1998;32:
435—440.
65. ten Cate JM, Damen JJ, Buijs MJ. Inhibition of dentin
demineralization by fluoride in vitro. Caries Research 1998;
32:141—147.
66. Perdigão J, Swift Jr EJ, Denehy GE, Wefel JS, Donly KJ. In
vitro bond strengths and SEM evaluation of dentin bonding
systems to different dentin substrates. Journal of Dental
Research 1994;73:44—55.
67. Goncalves SE, de Araujo MA, Damiao AJ. Dentin bond
strength: influence of laser irradiation, acid etching, and
hypermineralization. Journal of Clinical Laser Medicine and
Surgery 1999;17:77—85.
68. Pashley DH, Carvalho RM. Dentine permeability and dentine
adhesion. Journal of Dentistry 1997;25:355—372.
69. Perdigão J. Dentin bonding as a function of dentin structure.
Dental Clinics of North America 2002;46:277—301.
70. Van Meerbeek B, De Munck J, Yoshida Y, Inoue S, Vargas M,
Vijay P, Van Landuyt K, Lambrechts P, Vanherle G. Buonocore
memorial lecture. Adhesion to enamel and dentin: current
status and future challenges. Operational Dentistry 2003;28:
215—235.
71. Nakabayashi N, Kojima K, Masuhara E. The promotion of
adhesion by the infiltration of monomers into tooth
Resin bonding to cervical sclerotic dentin: A review
substrates. Journal of Biomedical Materials Research 1982;
16:265—273.
72. Gwinnett AJ, Kanca JA. Micromorphology of the bonded
dentin interface and its relationship to bond strength.
American Journal of Dentistry 1992;5:73—77.
73. Kanca JA. Resin bonding to wet substrate. 1. Bonding to
dentin. Quintessence International 1992;23:39—41.
74. Gwinnett AJ. Chemically conditioned dentin: a comparison
of conventional and environmental scanning electron
microscopy findings. Dental Materials 1994;10:150—155.
75. Titley K, Chernecky R, Maric B, Valiquette N, Smith D. The
morphology of the demineralized layer in primed dentin.
American Journal of Dentistry 1994;7:22—26.
76. Pashley DH, Zhang Y, Agee K, Rouse CJ, Carvalho RM, Russell
CM. Permeability of demineralized dentin to HEMA. Dental
Materials 2000;1:7—14.
77. Pashley DH, Agee KA, Carvalho RM, Lee KW, Tay FR, Callison
TE. Effects of water and water-free polar solvents on the
tensile properties of demineralized dentin. Dental Materials
2003;19:347—352.
78. Tao L, Pashley DH, Boyd L. The effect of different types of
smear layers on dentin and enamel bond strengths. Dental
Materials 1988;4:208—216.
79. Eick JD, Cobb CM, Chappell RP, Spencer P, Robinson SJ. The
dentinal surface: its influence on dentinal adhesion. Part I.
Quintessence International 1991;22:967—977.
80. Watanabe I, Nikaido T, Nakabayashi N. Effect of adhesion
promoting monomers on adhesion to ground dentin. Japa-
nese Journal Society for Dental Materials and Devices 1990;
9:888—893.
81. Chigira H, Manabe A, Hasegawa T, Yukitani W, Fujimitsu T,
Itoh K, Hisamitsu H, Wakumoto S. Efficacy of various
commercial dentin bonding systems. Dental Materials
1994;10:363—368.
82. Finger WJ, Balkenhol M. Practitioner variability effects on
dentin bonding with an acetone-based one-bottle adhesive.
Journal of Adhesive Dentistry 1999;1:311—314.
83. Peschke A, Blunck U, Roulet JF. Influence of incorrect
application of a water-based adhesive system on the
marginal adaptation of Class V restorations. American
Journal of Dentistry 2000;13:239—244.
84. Ferrari M, Tay FR. Technique sensitivity in bonding to vital
acid-etched dentin. Operational Dentistry 2003;28:3—8.
85. Watanabe I, Nakabayashi N, Pashley DH. Bonding to ground
dentin by a phenyl-P self-etching primer. Journal of Dental
Research 1994;73:1212—1220.
86. Nakabayashi N, Saimi Y. Bonding to intact dentin. Journal of
Dental Research 1996;75:1706—1715.
87. Tay FR, Pashley DH. Aggressiveness of contemporary self-
etching adhesives. Part I. Depth of penetration beyond
dentin smear layers. Dental Materials 2001;17:296—308.
88. Prati C, Chersoni S, Mongiorgi R, Pashley DH. Resin-
infiltrated dentin layer formation of new bonding systems.
Operational Dentistry 1998;23:185—194.
89. Yoshiyama M, Matsuo T, Ebisu S, Pashley DH. Regional bond
strengths of self-etching/self-priming adhesive systems.
Journal of Dentistry 1998;26:609—616.
90. Pereira PNR, Okuda M, Sano H, Yoshikawa T, Burrow MF,
Tagami J. Effect of intrinsic wetness and regional difference
on dentin bond strength. Dental Materials 1999;15:46—53.
91. Tay FR, Sano H, Carvalho R, Pashley EL, Pashley DH. An
ultrastructural study of the influence of acidity of self-
etching primers and smear layers thickness on bonding to
intact dentin. Journal of Adhesive Dentistry 2000;2:83—98.
92. Ogata M, Harada N, Yamaguchi S, Nakajima M, Tagami J.
Effect of self-etching primer vs phosphoric acid etchant on
195
bonding to bur-prepared dentin. Operational Dentistry
2002;27:447—454.
93. Toida T, Watanabe A, Nakabayashi N. Effect of smear layer
on bonding to dentin prepared with bur. Japanese Journal
Society for Dental Materials and Devices 1995;14:109—116.
94. Igarashi K, Toida T, Nakabayashi N. Effect of Phenyl P/HEMA
primer on bonding to demineralized dentin by phosphoric
acid. Japanese Journal of Dentistry Materials 1997;16:
55—60.
95. Miyasaka K, Nakabayashi N. Combination of EDTA condi-
tioner and Phenyl-P/HEMA self-etching primer for bonding to
dentin. Dental Materials 1999;15:153—157.
96. Tay FR, Carvalho R, Sano H, Pashley DH. Effect of smear
layers on the bonding of a self-etching primer to dentin.
Journal of Adhesive Dentistry 2000;2:99—116.
97. Eick JD, Gwinnett AJ, Pashley DH, Robinson SJ. Current
concepts on adhesion to dentin. Critical Reviews in Oral
Biology and Medicine 1997;8:306—335.
98. Gwinnett AJ, Tay FR, Pang KM, Wei SH. Quantitative
contribution of the collagen network in dentin hybridization.
American Journal of Dentistry 1996;9:140—144.
99. Coli P, Alaeddin S, Wennerberg A, Karlsson S. In vitro dentin
pretreatment: surface roughness and adhesive shear bond
strength. European Journal of Oral Sciences 1999;107:
400—413.
100. Hashimoto M, Ohno H, Kaga M, Sano H, Tay FR, Oguchi H,
Araki Y, Kubota M. Over-etching effects on micro-tensile
bond strength and failure patterns for two dentin bonding
systems. Journal of Dentistry 2002;30:99—105.
101. Wang Y, Spencer P. Quantifying adhesive penetration in
adhesive/dentin interface using confocal Raman micro-
spectroscopy. Journal of Biomedical Materials Research
2002;59:46—55.
102. Wang Y, Spencer P. Hybridization efficiency of the
adhesive/dentin interface with wet bonding. Journal of
Dental Research 2003;82:141—145.
103. Vasiliadis L, Darling AI, Levers BG. The histology of sclerotic
human root dentine. Archives of Oral Biology 1983;28:
693—700.
104. Vargas MA, Cobb DS, Armstrong SR. Resin†dentin shear bond
strength and interfacial ultrastructure with and without a
hybrid layer. Operational Dentistry 1997;22:159—166.
105. Inai N, Kanemura N, Tagami J, Watanabe LG, Marshall SJ,
Marshall GW. Adhesion between collagen depleted dentin
and dentin adhesives. American Journal of Dentistry 1998;
11:123—127.
106. Tay FR, King NM, Chan KM, Pashley DH. How can
nanoleakage occur in self-etching adhesive systems that
demineralize and infiltrate simultaneously? Journal of
Adhesive Dentistry 2002;4:255—269.
107. Pashley DH, Horner JA, Brewer PD. Interactions of condi-
tioners on the dentin surface. Operational Dentistry 1992;
5(Suppl):137—150.
108. Pashley DH, Tao L, Boyd L, King GE, Horner JA. Scanning
electron microscopy of the substructure of smear layers in
human dentine. Archives of Oral Biology 1988;33:265—270.
109. Yu XY, Wieczkowski G, Davis EL, Joynt RB. Scanning
electron microscopic study of dentinal surfaces treated
with various dentinal bonding agents. Quintessence Inter-
national 1990;21:989—999.
110. Ferrari M, Cagidiaco MC, Kugel G, Davidson CL. Dentin
infiltration by three adhesive systems in clinical and
laboratory conditions. American Journal of Dentistry
1996;9:204—240.
111. Gwinnett AJ. Quantitative contribution of resin infiltra-
tion/hybridization to dentin bonding. American Journal of
Dentistry 1993;6:7—9.
196
112. Pashley DH, Sano H, Ciucchi B, Carvalho RM, Russell CM.
Bond strength versus dentin structures: a modeling
approach. Archives of Oral Biology 1995;40:1109—1118.
113. Pashley DH, Carvalho RM, Sano H, Nakajima M, Yoshiyama
M, Shono Y, Fernandes CA, Tay F. The micro-tensile bond
test: a review. Journal of Adhesive Dentistry 1999;1:
299—309.
114. Shono Y, Ogawa T, Terashita M, Carvalho RM, Pashley EL,
Pashley DH. Regional measurement of resin—dentin bond-
ing as an array. Journal of Dental Research 1999;78:
699—705.
115. Phrukkanon S, Burrow MF, Tyas MJ. The effect of dentine
location and tubule orientation on the bond strengths
between resin and dentine. Journal of Dentistry 1999;27:
265—274.
116. Ogata M, Okuda M, Nakajima M, Pereira PN, Sano H, Tagami
J. Influence of the direction of tubules on bond strength to
dentin. Operational Dentistry 2001;26:27—35.
117. Uno S, Inoue H, Finger WJ, Inoue S, Sano H. Microtensile
bond strength evaluation of three adhesive systems in
cervical dentin cavities. Journal of Adhesive Dentistry
2001;3:333—341.
118. Torii Y, Itou K, Nishitani Y, Ishikawa K, Suzuki K. Effect of
phosphoric acid etching prior to self-etching primer
application on adhesion of resin composite to enamel and
dentin. American Journal of Dentistry 2002;15:305—308.
119. Walker MP, Wang Y, Swafford J, Evans A, Spencer P.
Influence of additional acid etch treatment on resin cement
dentin infiltration. Journal of Prosthodontics 2000;9:
77—81.
120. Ogata M, Nakajima M, Sano H, Tagami J. Effect of dentin
primer application on regional bond strength to cervical
wedge-shaped cavity walls. Operational Dentistry 1999;24:
81—88.
121. Gwinnett AJ, Garcı́a-Godoy F. Effect of etching times and
acid concentration on resin shear bond strength to primary
tooth enamel. American Journal of Dentistry 1992;5:
230—237.
122. Gwinnett AJ, Kanca JA. Interfacial morphology of resin
composite and shiny erosion lesions. American Journal of
Dentistry 1992;5:315—317.
123. Van Meerbeek B, Braem M, Lambrechts P, Vanherle G. Two-
year clinical evaluation of two dentine-adhesive systems in
cervical lesions. Journal of Dentistry 1993;21:195—202.
124. Tani C, Itoh K, Hisamitsu H, Wakumoto S. Efficacy of dentin
bonding to cervical defects. Dental Materials Journal 2001;
20:359—368.
125. Handelman SL, Shey Z. Michael Buonocore and the Eastman
Dental Center: a historic perspective on sealants. Journal
of Dental Research 1996;75:529—534.
126. Simonsen RJ. Pit and fissure sealant: review of the
literature. Pediatric Dentistry 2002;24:393—414.
127. Yoshiyama M, Tay FR, Doi J, Nishitani Y, Yamada T, Itou K,
Carvalho RM, Nakajima M, Pashley DH. Bonding of self-etch
and total-etch adhesives to carious dentin. Journal of
Dental Research 2002;81:556—560.
128. Yoshiyama M, Tay FR, Torii Y, Nishitani Y, Doi J, Itou K,
Ciucchi B, Pashley DH. Resin adhesion to carious dentin.
American Journal of Dentistry 2003;16:47—52.
129. Imazato S, Ehara A, Torii M. Ebisu. Antibacterial activity of
dentine primer containing MPDB after curing. Journal of
Dentistry 1998;26:267—271.
F.R. Tay, D.H. Pashley
130. Imazato S, Kinomoto Y, Tarumi H, Ebisu S. R Tay F.
Antibacterial activity and bonding characteristics of an
adhesive resin containing antibacterial monomer MDPB.
Dental Materials 2003;19:313—319.
131. Sano H, Yoshikawa T, Peireira PNR, Kanemura N, Morigami
M, Tagami J, Pashley DH. Long-term durability of dentin
bonds made with a self-etching primer, in vivo. Journal of
Dental Research 1999;78:906—911.
132. Tay FR, Hashimoto M, Pashley DH, Peters MC, Lai SC, Yiu
CK, Cheong C. Aging affects two modes of nanoleakage
expression in bonded dentin. Journal of Dental Research
2003;82:541—547.
133. Hashimoto M, Ohno H, Kaga M, Endo K, Sano H, Oguchi
H. In vivo degradation of resin—dentin bonds in humans
over 1 to 3 years. Journal of Dental Research 2000;79:
1385—1391.
134. Hashimoto M, Ohno H, Kaga M, Endo K, Sano H, Oguchi H.
Resin-tooth adhesive interfaces after long-term function.
American Journal of Dentistry 2001;14:211—215.
135. Hashimoto M, Ohno H, Sano H, Kaga M, Oguchi H. In vitro
degradation of resin—dentin bonds analyzed by microtensile
bond test, scanning and transmission electron microscopy.
Biomaterials 2003;24:3795—3803.
136. Hashimoto M, Ohno H, Sano H, Kaga M, Oguchi H.
Degradation patterns of different adhesives and bonding
procedures. Journal of Biomedical Materials Research
2003;66B:324—330.
137. De Munck J, Van Meerbeek B, Yoshida Y, Inoue S, Vargas M,
Suzuki K, Lambrechts P, Vanherle G. Four-year water
degradation of total-etch adhesives bonded to dentin.
Journal of Dental Research 2003;82:136—140.
138. Horsted-Bindslev P, Knudsen J, Baelum V. 3-year clinical
evaluation of modified Gluma adhesive systems in cervical
abrasion/erosion lesions. American Journal of Dentistry
1996;9:22—26.
139. Mandras RS, Thurmond JW, Latta MA, Matranga LF, Kildee
JM, Barkmeier WW. Three-year clinical evaluation of the
Clearfil Liner Bond system. Operational Dentistry 1997;22:
266—270.
140. Brunton PA, Cowan AJ, Wilson MA, Wilson NH. A three-year
evaluation of restorations placed with a smear-layer-
mediated dentin bonding agent in non-carious cervical
lesions. Journal of Adhesive Dentistry 1999;1:333—341.
141. van Dijken JW. Clinical evaluation of three adhesive
systems in class V non-carious lesions. Dental Materials
2000;16:285—291.
142. Tyas MJ, Burrow MF. Three-year clinical evaluation of one-
step in non-carious cervical lesions. American Journal of
Dentistry 2002;15:309—311.
143. Neo J, Chew CL. Direct tooth-colored materials for
noncarious lesions: a 3-year clinical report. Quintessence
International 1996;27:183—188.
144. Matis BA, Cochran M, Carlson T. Longevity of glass-ionomer
restorative materials: results of a 10-year evaluation.
Quintessence International 1996;27:373—382.
145. Brackett MG, Dib A, Brackett WW, Estrada BE, Reyes AA.
One-year clinical performance of a resin-modified glass
ionomer and a resin composite restorative material in
unprepared Class V restorations. Operational Dentistry
2002;27:112—116.