real-time measurement of dentinal fluid flow during amalgam and composite restoration

Real-time measurement of dentinal fluid flow duringamalgam and composite restoration

Sun-Young Kim a, Jack Ferracane b, Hae-Young Kim c, In-Bog Lee a,*aDepartment of Conservative Dentistry and Dental Research Institute, School of Dentistry, Seoul National University,

28-2 Yeongeon-Dong, Jongro-Gu, Seoul 110-749, South KoreabDepartment of Restorative Dentistry, Division of Biomaterials and Biomechanics, School of Dentistry,

Oregon Health & Science University, Portland, OR, USAcDepartment of Dental Hygiene, College of Health Science, Eulji University, Seoul, South Korea

j o u r n a l o f d e n t i s t r y 3 8 ( 2 0 1 0 ) 3 4 3 – 3 5 1

a r t i c l e i n f o

Article history:

Received 8 October 2009

Received in revised form

23 December 2009

Accepted 25 December 2009

Keywords:

Dentinal fluid flow

Permeability

Amalgam

Composite

Etch-and-rinse

Self-etch

s u m m a r y

Objectives: This study examined changes in the dentinal fluid flow (DFF) during restorative

procedures and compared permeability after restoration among restorative materials and

adhesives.

Methods: A class 1 cavity was prepared and restored with either amalgam (Bestaloy), or

composite (Z-250) with one of two etch-and-rinse adhesives (Scotchbond MultiPurpose: MP

and Single Bond 2: SB) or one of two self-etch adhesives (Clearfil SE Bond: CE and Easy Bond:

EB) on an extracted human third molar which was connected to a sub-nanoliter fluid flow

measuring device (NFMD) under 20 cm water pressure. DFF was measured from the intact

tooth state through the restoration procedures to 30 min after restoration, and re-measured

at 3 and 7 days post-restoration.

Results: Inward flow during cavity preparation was followed by outward flow after prepa-

ration. In amalgam restoration, the outward flow changed into an inward flow during

amalgam filling, which was followed by a slight outward flow after finishing. In composite

restoration, MP and SB showed an inward flow and outward flow for the rinsing and drying

steps, respectively. Application of a hydrophobic bonding resin in the MP and CE systems

caused a decrease in the flow rate. Air-drying of solvent for the CE and EB systems caused a

sudden outward flow, whereas light-curing of the adhesive and composite caused an abrupt

inward flow.

Conclusions: Each restorative step clearly changed the direction and the rate of the DFF

during restoration, which could be well identified with NFMD.

# 2010 Elsevier Ltd. All rights reserved.

avai lab le at www.sc iencedi rect .com

journal homepage: www.intl.elsevierhealth.com/journals/jden

1. Introduction

The hydrodynamic theory is the most widely accepted

explanation of dentin hypersensitivity.1 This theory proposes

that stimuli applied to the affected tooth cause the dentinal

tubular fluid to move in either an outward or inward direction.

This movement stimulates the mechano-receptors of the

* Corresponding author. Tel.: +82 2 2072 3953; fax: +82 2 2072 3859.E-mail address: [email protected] (I.-B. Lee).

0300-5712/$ – see front matter # 2010 Elsevier Ltd. All rights reserveddoi:10.1016/j.jdent.2009.12.008

sensory nerves in the dentin or pulp. There have been many

reports about different aspects of the dentinal fluid flow (DFF)

due to various stimuli applied to exposed dentin. For example,

cold and air-blast stimuli caused an outward flow of the

dentinal tubular fluid while heat caused an inward flow.

Osmotic pressure through the application of a hypertonic

solution caused an outward flow while tactile stimuli caused a

.

mailto:[email protected]

http://dx.doi.org/10.1016/j.jdent.2009.12.008


slightly inward flow. Mechanical loading corresponding to

occlusal force caused an inward flow, followed by an outward

flow when the mechanical loading was removed.2–5

While an amalgam restoration involves a relatively simple

procedure for cavity filling, a composite restoration involves

more complicated steps, including acid-etching, rinsing and

drying, adhesive application and light-curing, composite

filling and additional light-curing, and finishing and polishing.

These sequential restoration procedures can act on dentin as

hydrodynamic stimuli, causing the dentinal fluid to move in

either an outward or inward direction. Thus far, investigations

that show consecutive DFF during composite restoration are

rare.6,7 Furthermore, a comparison of DFF during an amalgam

restoration to that during a composite restoration has not

been done.

Continuous fluid movement through a dentin disc during a

bonding procedure,6 and DFF and cuspal displacement in

response to a resin composite restorative procedure has been

reported.7 The Flodec system (DeMarco Engineering, Geneva,

Switzerland) has been predominantly used for the study of

DFF; however, the results could not show clearly the details of

DFF for each restorative procedure due to the limitation of the

resolution (2 nL) of the instrument.6,7

In our study, a sub-nanoliter scaled fluid flow measuring

device (NFMD) capable of discriminating a volume change of

0.2 nL was fabricated to trace the DFF in real-time during

restorative procedures on an extracted tooth. The tooth was

restored with amalgam or resin composite under physiologic

pulpal pressure8 and DFF was measured during and after the

restorative procedure using the NFMD. The aim of this study

was to observe the aspects of DFF during the restoration

procedure and to compare the permeability after restoration

Fig. 1 – (a) Schematic diagram of the sub-nanoliter scaled denti

preparation.

with the different restorative materials and adhesives used.

The tested null hypotheses were that the DFF pattern did not

show any change according to each restorative procedure of

amalgam or composite restoration and that the permeability

after restoration did not show any difference among the

restoration materials and the measurement time.

2. Materials and methods

2.1. Structure and working mechanism of the NFMD

The NFMD fabricated for this study consisted of three parts: a

glass capillary and photo-sensor to detect the fluid movement;

a servomotor, lead screw, and ball nut to track the fluid

movement; and a rotary encoder and computer software to

record the data (Fig. 1a).

A water filled glass capillary with an internal diameter of

0.5 mm was connected between a water reservoir and the

tooth. A photo-sensor consisting of an infrared-light emitting

diode (EL-1CL3H, Kodensh Korea Corp., Iksan, Korea) and a

photo transistor (ST-1CL3H, Kodensh Korea Corp., Iksan,

Korea) detected the movement of an air bubble trapped within

the capillary. The voltage from the photo-sensor was input to a

comparator composed of an OP-amp (LM17741, National

Semiconductor, Santa Clara, CA, USA), which compared the

voltage from the photo-sensor with a preset reference voltage

corresponding to that of the water–air interface.

The output voltage from the comparator was input to the

servo-amplifier, which supplied the servo motor with power.

The motor rotated the lead screw, which had 1 mm of pitch,

and moved a connected ball nut to which the photo-sensor

nal fluid flow measurement system (NFMD). (b) Specimen

j o u r n a l o f d e n t i s t r y 3 8 ( 2 0 1 0 ) 3 4 3 – 3 5 1 345

was attached so that the photo-sensor was always placed at

the interface between the water and the air bubble. This

continuous negative feedback mechanism makes the move-

ment of the air bubble traceable in real-time.

The rotation of the screw was measured using a rotary

encoder (E40S-1000-3-2, Autonics, Seoul, Korea) which outputs

1000 pulses per rotation. The number of pulses was stored on a

computer at a rate of two data points/s using a data acquisition

device (PCI 6016, National Instrument, Austin, TX, USA).

As the movement distance per rotation of the lead screw is

1 mm and the resolution of the encoder is 1000 pulses per

rotation, the position resolution of the NFMD is 1 mm.

Therefore, the minimum measurable volume of water

movement is 0.196 nL [(0.25 mm � 0.25 mm � p) � 1 mm, the

diameter of the capillary = 0.5 mm].

2.2. Specimen preparation

Fifty caries-free human third molars were used in this study.

The project was approved by the Institutional Review Board of

Seoul National University Dental Hospital (CRI09005). The

third molars were extracted for surgical reasons independent

of this study. Extracted teeth were stored in 1% chloramine T

solution at 4 8C and were used within three months following

their extraction.

The root was removed 3 mm below the CEJ using a low-

speed diamond saw (Isomet, Buehler, IL, USA). The pulp tissue

in the pulp chamber was carefully removed without altering

the pre-dentin surface by using thin tissue forceps and

endodontic files. A sandblasted plexiglass square (10 mm

per side and 2 mm thick) with a hole drilled at its center was

used to mount each tooth. A metal tube 0.9 mm in diameter

was inserted into the hole, and the plexiglass was attached to

the tooth using an adhesive (Adper Scotchbond MultiPurpose,

3M ESPE, St. Paul, MN, USA) and a flowable composite

(Denflow, Vericom, Anyang, Korea) to ensure that one end

of the metal tube was located in the pulp chamber. The outer

surface of the bonded interface between the plexiglass and the

tooth on the top surface, and between the plexiglass and the

metal tube on the bottom surface was covered by nail varnish.

The specimen was placed in a rubber mold except for one third

Table 1 – The restoration materials and procedures of each gr

Restoration type Group code M

Amalgam AM Copalite varnish (C

TX, USA), Bestaloy

(Dongmyung, Seou

Composite – etch-and-rinse adhesive MP Adper Scotchbond

St. Paul, MN, USA)

(3M ESPE, St. Paul,

SB Adper Single Bond

St. Paul, MN, USA)

Composite – self-etch adhesive CE Clearfil SE Bond (K

Okayama, Japan), F

EB Adper Easy Bond (

Seefeld, Germany)

E, acid-etching for 15 s; P, primer application; A, adhesive application; PA

primer application; SEA, application of self-etch adhesive; LC, light-curin

of the exposed metal tube and was embedded with epoxy resin

to 1 mm above the CEJ (Fig. 1b).

The prepared specimen was stored in distilled water and

was connected to a water reservoir which contained distilled

water. A hydrostatic pressure of 20 cm H2O was applied to the

specimen from 24 h before the experiment was conducted.8

2.3. Measurement of the DFF during and after therestorative procedures

The prepared specimen was connected to the glass capillary

by silicon tubing filled with distilled water (Fig. 1). A

hydrostatic pressure of 20 cm H2O was applied throughout

all procedures with a water reservoir to simulate physiological

pulp pressure. Each specimen had a stabilizing time of 10 min

after being connected to the NFMD. After confirming that the

fluctuation level of the DFF was within �5 nL for a further

10 min, the restoration procedure was started.

A class 1 cavity with a mesio-distal width of 5 mm, a bucco-

lingual width of 3 mm, and an occlusal depth of 2 mm from a

central pit was prepared using an air-driven high speed

handpiece (MACH-QD, NSK, Tokyo, Japan) with 200,000–

300,000 rpm under water-coolant. The prepared cavity was

blot-dried with a wet cotton pellet and was then allowed to

remain exposed to air for 7–10 min. The average rate of the DFF

occurring for 5 min just before starting the restorative

procedure, which consisted of the varnish application for

the amalgam and the adhesive application for the composite,

was set as a baseline flow rate, i.e. internal reference, for

comparison with subsequent flow rate measurements after

restoration.

The prepared cavity was restored with either an amalgam

preceded by varnish (Copalite, Cooley & Cooley, Houston, TX,

USA), or a resin composite preceded by one of two etch-and-

rinse adhesives or one of two self-etch adhesives. The

materials and procedures used in this study are shown in

Table 1. An LED curing unit (Elipar Freelight 2, 3M ESPE, St.

Paul, MN, USA) was used to polymerize the composite and

adhesive (irradiance measured at 700 mW/cm2). The DFF

measurement was performed continuously from the intact

tooth state, throughout the cavity preparation and restorative

oup.

aterials Procedure

ooley & Cooley, Houston,

amalgam alloy

l, Korea)

Copalite application, amalgam filling

MultiPurpose (3M ESPE,

, Filtek Z-250

MN, USA)

E, P, A, LC (10 s), CF and LC (40 s):

incrementally by two horizontal layers

2 (3M ESPE,

, Filtek Z-250

E, PA, LC (10 s), CF and LC (40 s):


uraray Medical Inc.

iltek Z-250

SEP, A, LC (10 s), CF and LC (40 s):


3M ESPE AG,

, Filtek Z-250

SEA, LC (10 s), CF and LC (40 s):


, application of mixed agent with primer and adhesive; SEP, self-etch

g; CF, composite filling.

Fig. 2 – A representative curve of consecutive DFF during

amalgam restoration. Upward (positive slope) movement

vs time on graph indicates outward DFF, whereas

downward (negative slope) movement indicates inward

DFF. Gentle air stream and amalgam filling caused

significant changes in the flow rate compared to the

baseline flow rate (P < 0.05). CP, cavity preparation; ECP,

end of cavity preparation; V, copalite varnish application;

ga, gentle air stream; AF, amalgam filling; and EAC, end of

amalgam condensing.


procedures, and to 30 min after the restoration in real-time. A

timetable was made for all steps of the cavity preparation,

adhesive procedure, and filling procedure. The timetable was

used to indentify the events in DFF corresponding to each step

during the restorative procedures. To determine the rate of

DFF after restoration, the average flow rate was calculated for

the last 5 min of the 30 min recorded after restoration.

The restored specimen was continuously connected to the

water reservoir with silicon tubing while maintaining a

hydrostatic pressure of 20 cm H2O and stored in a 100%

humidity plastic box, and the flow rate was re-measured at 3

days and 7 days after restoration. The average flow rate at 3

days and 7 days was determined for 10 min after stabilization

was achieved. Reductions in the flow rate were indicated as a

percent of the decreased flow rate at 30 min, 3 days, and 7 days

post-restoration with respect to the baseline flow rate [%

reduction in flow rate = 100 � (flow ratebase line � flow ratepost-

restoration)/flow ratebase line].9

The temperature and humidity of the environment were

24 � 0.5 8C and 30 � 5%, respectively. Under consideration of

1% of maximum flow rate reduction difference among five

materials or among three repeated measurements, an

appropriate sample size of eight per group was obtained with

power of 0.82 and type one error level 0.05. Finally a sample

size of 10 teeth per group was used to avoid possible reduction

of power because of failure cases during experiment proce-

dure. A repeated measure ANOVA was conducted to analyze

whether there were differences in the flow rate for each

restorative step during restoration and whether there were

differences in the reduction of the flow rate among the

restoration materials and the post-restoration measurement

time (30 min, 3 days, and 7 days). A paired T-test was

conducted to compare the flow rate for specific restorative

step with the baseline flow rate. The level of significance was

a = 0.05. The statistical package SPSS version 14.0 (SPSS Inc.,

Chicago, IL, USA) was used for all analytical procedures.

3. Results

The specific behavior of the DFF during the restorative

procedures according to the restoration methods are shown

in Figs. 2–6. Cavity preparation with an air-driven high speed

handpiece under water cooling showed an inward flow in

most cases. The direction of the DFF then reversed to an

outward movement as soon as cavity preparation was

complete. As for the amalgam restoration, the outward flow

after cavity preparation changed to an inward flow when

amalgam condensation was performed, which changed again

to a slight outward flow once amalgam condensing was

complete (Fig. 2).

The changes in DFF during resin composite bonding

procedure of each adhesive system are shown in Figs. 3b–6b

in detail. For the etch-and-rinse adhesive systems, MP and SB

(Figs. 3 and 4), the etching gel application itself caused a

decrease in the rate of the outward flow. Water rinsing caused

an inward flow, while blot-drying caused an outward flow. The

use of air-blast to evaporate solvent from the adhesives during

the bonding procedure caused an abrupt increase in outward

flow rate, even when it was performed gently. In the case of

MP, the adhesive resin application decreased the rate of

outward flow to nearly zero or even caused a slight inward

flow.

For the self-etch adhesive systems, CE and EB (Figs. 5 and 6),

the application of a self-etching primer or a self-etching

primer/adhesive did not show a specific change in the flow

direction or the flow rate. An air-dry procedure abruptly

increased the rate of outward flow. In the case of CE,

application of the solvent free bonding resin decreased the

rate of outward flow to nearly zero or even caused a slight

inward flow.

In all resin composite restorations, all three light-curing

procedures (one for adhesive curing; two for composite curing)

caused an abrupt inward flow, and a rapid compensatory

outward flow occurred after the final light-curing. The rate of

outward flow then gradually decreased to a low rate (Figs. 3–6).

Fig. 7 shows the mean reduction in the flow rate in percent

compared to the baseline at 30 min, 3 days, and 7 days after

restoration. There was no significant difference according to

the three repeated measurement times (P = 0.469) nor among

the different types of the materials at any time period

(P = 0.226).

4. Discussion

This study is the first to measure a continuous dentinal fluid

flow during an entire restorative procedure, including cavity

preparation. While most previous studies showed individual

dentinal fluid movements responding to a single stimulus,

such as air-blasts or hot and cold conditions, this study

followed the DFF as it occurs in events such as rinsing, drying,

and adhesive applications as well as restorative procedures.

Fig. 3 – (a) A representative curve of consecutive DFF during

composite restoration with MP. (b) Magnified view of

consecutive DFF during bonding procedure of MP. Upward

(positive slope) movement vs time on graph indicates

outward DFF, whereas downward (negative slope)

movement indicates inward DFF. The steps of acid-

etching, rinse, gentle air stream, adhesive resin

application, and light-curing caused significant changes in

the flow rate compared to the baseline flow rate (P < 0.05).

CP, cavity preparation; ECP, end of cavity preparation; E,

acid-etching; r, rinse; d, blot-dry; P, primer application; ga,

gentle air; A, adhesive resin application; LC, light-curing;

C1, composite filling of first layer; and C2, composite filling

of second layer.


Moreover, this study is the first to compare the DFF behavior

during amalgam and composite restorative procedures, as

well as comparing different types of resin adhesives.

Although an outward flow was driven by a simulated pulpal

pressure of 20 cm water, the DFF direction during the cavity

preparation process was mostly inward. The inward flow

during the cavity preparation might be caused from the

movement of dentinal tubular fluid due to heat produced by

the friction of the bur against the tooth structure,10 or caused

by water entering the dentinal tubule due to the pressure of

water-coolant application from a high speed handpiece. The

inward flow changed to an outward flow shortly after the

cavity preparation was complete, possibly due to hydrostatic

pulpal pressure and water evaporation.

Amalgam restoration showed a relatively simple DFF

behavior without abrupt outward or inward flow (Fig. 2). For

example, the outward flow after cavity preparation changed to

a slight inward flow upon filling amalgam into the cavity; that

inward flow changed to a slight outward flow once amalgam

burnishing was complete. The pressure from amalgam

condensing and burnishing did not appear to cause an abrupt

change of the DFF, possibly because the force application is

localized and does not affect a majority of the tubules at any

given time. Although varnish is used to reduce the permeabil-

ity of dentin during amalgam restoration, its use is not likely to

improve the sealing of dentinal tubule significantly. Instead,

smear layer and/or smear plug was proved to provide most of

the sealing of cavity in previous studies.11–13 This study also

showed that varnish applied after cavity preparation could not

cause a significant change of the flow rate compared to the

baseline flow rate (Fig. 2).

The DFF changes measured during composite restoration

were far more complicated compared to placement of

amalgam (Figs. 3–6).

An air-blast, even when it was performed gently, increased

the rate of outward flow abruptly. The air-blast is often

performed to evaporate the organic solvent, to prevent

adhesive pooling, or to make a thick adhesive layer thin

during the bonding procedure. There are two possible

mechanisms to explain why an air-blast causes an abrupt

outward flow. First, an air-blast would increase the water

evaporation from the cavity surface, which would establish an

air–liquid interface at the dentinal tubule.14 This would cause

an elevated capillary phenomenon resulting in abrupt

outward movement of dentinal tubular fluid.15 The elevated

water height by the capillary action is given by h = 2s/grr [h:

elevated water height (m), s: water surface tension (0.07 N/m),

g: gravity (9.81 m/s2), r: water density (1000 kg/m3), r: inner

radius of capillary].16 A small capillary with a diameter of 1–

5 mm, similar to dentinal tubule, has as great an absorption

force as a 7 m water column when the mean diameter of the

dentinal tubule is assumed to be 2 mm. Second, an air-blast

would act as cold stimuli that would decrease the surface

temperature, which would shrink the volume of dentinal

tubular fluid and cause an outward movement of this fluid.

Based on this study, the DFF during the composite

restoration showed a complicated behavior with more

frequent changes in the direction of fluid movement when

compared to that occurring during the amalgam restoration.

Although it is likely that most postoperative sensitivity

associated with composite restoration is due mainly to

polymerization shrinkage,17 this frequent change in the DFF

may contribute to postoperative sensitivity afterwards by

stimulation of the sensory nerve ending during the restoration

process in advance.18

The differences between etch-and-rinse adhesives (MP and

SB) and self-etch adhesives (CE and EB) were sharply reflected

in the DFF curves during the bonding procedure. Rinsing and

drying after acid-etching moved the dentinal fluid inward and

outward, respectively, in the form of a valley-shaped flow

curve during the etch-and-rinse adhesive bonding procedure

(Figs. 3 and 4). On the other hand, the self-etch adhesive

maintained an outward flow until light-curing, although there

was a slight difference in the flow rate between each step of

the bonding procedure (Figs. 5 and 6).

A difference in the DFF behavior was also found between an

adhesive system which used a separate hydrophobic bonding

resin (i.e. MP and CE) and the simplified adhesive system in

Fig. 4 – (a) A representative curve of consecutive DFF during composite restoration with SB. (b) Magnified view of consecutive

DFF during bonding procedure of SB. Upward (positive slope) movement vs time on graph indicates outward DFF, whereas

downward (negative slope) movement indicates inward DFF. The steps of acid-etching, rinse, gentle air stream, and light-

curing caused significant changes in the flow rate compared to the baseline flow rate (P < 0.05). CP, cavity preparation; ECP,

end of cavity preparation; E, acid-etching; r, rinse; d, blot-dry; sb, SB application; ga, gentle air; LC, light-curing; C1,

composite filling of first layer; and C2, composite filling of second layer.


which primer and bonding resin were mixed in a single bottle

(i.e. SB and EB). In the case of the MP and CE, the application of

a hydrophobic bonding agent itself caused the flow rate to

decrease close to zero and finally reached a plateau or

sometimes a slight inward flow (Figs. 3b–5b). In SB and EB,

which contain a mixture of a primer and a bonding resin,

however, the application of the material itself did not affect

the outward flow rate (Figs. 4b and 6b). While these simplified

adhesives contain a higher concentration of hydrophilic resin

monomer which allows them to penetrate better into the

dentin, their ability to block the outward movement of

dentinal fluid may be weak. Previous studies reported that

simplified adhesives of the two-step etch-and-rinse type and

one-step self-etch type contain many water channels in the

adhesive layer, even after polymerization; hence, the adhesive

layer would behave as a semi-permeable membrane.19–21 This

study showed that the hydrophobic bonding resin application

itself has the effect of decreasing the rate of outward flow

immediately, which can contribute to stabilizing the initial

adhesion and sealing between the dentin and resin and cause

less pulpal reactions to bonding procedures.22,23

The inward flow during the light-curing is thought to result

from the thermal expansion of dentinal fluid caused mainly by

the heat from the light-curing unit and secondly by the

polymerization heat of the composite. In this study, light-

curing of each layer of composite was performed for 40 s, and

the temperature of the light-curing tip end increased by more

than 20 8C from room temperature, which caused an increase

of temperature by 4–5 8C within the pulp chamber in a

supplementary experiment results not shown. When the

cavity was sealed by adhesive and resin composites, fluid

expansion in the dentinal tubules and pulp chamber caused by

this temperature increase would produce an abrupt inward

flow. The return to the steep outward flow shortly after light-

curing seems to be due to a ‘‘rebounding effect’’ in which the

expanded fluid in the dentinal tubule and pulp chamber

contracts again as the temperature decreases.

Ideally, restoration of a cavity preparation should seal the

dentinal tubules completely to reduce concerns over bacterial

penetration and hypersensitivity. When the flow rate was

measured at 30 min, 3 days, and 7 days post-restoration, the

amalgam and composite restorations showed similar results,

reductions in permeability with the flow rate decreased by

88.7–97.8% compared to the baseline flow rate. Thus,

movement of fluids across the dentin surface was dramati-

cally reduced; the dentin surface was not completely sealed.

Within the limits of this study, no restorations may seal the

dentinal tubules completely under circumstance in which

physiologic pulp pressure continuously causes an outward

flow.

Fig. 5 – (a) A representative curve of consecutive DFF during composite restoration with CE. (b) Magnified view of consecutive

DFF during bonding procedure of CE. Upward (positive slope) movement vs time on graph indicates outward DFF, whereas

downward (negative slope) movement indicates inward DFF. The steps of air-dry, bond resin application, and light-curing

caused significant changes in the flow rate compared to the baseline flow rate (P < 0.05). CP, cavity preparation; ECP, end of

cavity preparation; SE, self-etch primer application; a, air-dry; A, bond resin application; LC, light-curing; C1, composite

filling of first layer; and C2, composite filling of second layer.


A hydrostatic pressure of 15–20 cm H2O has been applied to

simulate the physiologic pulp pressure in many studies since

in vivo studies showed that normal physiologic pulp pressure

corresponded to a hydrostatic pressure of about 14 cm H2O

(range: 8–22 cm H2O).6,8,24–27 The studies applying a hydrostat-

ic pressure of 15–20 cm H2O including this study may over-

estimate DFF that might occur in vivo since under most in vivo

conditions, patients have received infiltration or block local

anesthetics containing epinephrine (1:100,000) which has

been shown to decrease pulpal blood flow up to 1–2 h.28–30

As pulpal blood flow falls, pulpal tissue pressure which is the

driving force for DFF would fall, so the DFF too would fall.27 It

would be needed to perform a further study considering the

effect of local anesthetics on permeability of restorations by

lowering or removing the hydrostatic pulp pressure during

restoration and afterward giving a physiologic pulp pressure.

The decrease in the flow rate according to the type of

restorative material or adhesive did not show any significant

differences in this study. This study determined the DFF rate

at relatively short time periods of 30 min, 3 days, and 7 days

after restoration. The permeability difference between the

amalgam and the composite restoration and among the

adhesives may not be shown in such a short-term observation.

A previous study which measured the decreased rate of

permeability after an amalgam and composite restoration

through a fluid transport system reported that the amalgam

restoration showed less permeability than the composite

restoration over a long period of 3 months while it did not

show a significant difference over a short period of 1 week.31

Recent studies in which microleakage was measured also

could not find a significant difference between amalgam and

composite restoration.32,33 Additional studies simulating the

intraoral condition by thermo-cycling and/or long-term

application of the hydrostatic pulpal pressure might be

necessary to investigate the difference in the permeability

between amalgam and composite restorations.

A number of earlier studies reported significant differences

of permeability between different types of adhesive; this study

did not show any statistically significant difference among the

adhesives used. 6,25,34 The reason why this study had a

different result from the earlier studies is thought to be that

this study had several different experimental conditions. First,

the baseline used in the previous studies was an increased

permeable state of dentin produced by acid-etching, whereas

the baseline in this study was the flow rate of a prepared cavity

with a smear layer produced by cutting but before acid-

etching. Second, this study measured not only the permeabil-

ity of cured adhesive on dentin but also the permeability of a

class 1 cavity restored with composite. The cavosurface

margins of the class 1 cavity in this study consisted of

excellent enamel–resin adhesion of resin to acid-etched

enamel compared to dentin–resin adhesion; this type of

Fig. 6 – (a) A representative curve of consecutive DFF during composite restoration with EB. (b) Magnified view of consecutive

DFF during bonding procedure of EB. Upward (positive slope) movement vs time on graph indicates outward DFF, whereas

downward (negative slope) movement indicates inward DFF. Air-dry and light-curing caused significant changes in the

flow rate compared to the baseline flow rate (P < 0.05). CP, cavity preparation; ECP, end of cavity preparation; eb, EB

application; a, air-dry; LC, light-curing; C1, composite filling of first layer; and C2, composite filling of second layer.


adhesion could have a much higher sealing ability compared

to the types used in the previous studies, in which a dentin

disc or piece of exposed dentin produced by horizontal cutting

was used as a specimen. Third, polymerization shrinkage

stress or water expansion of a composite in a cavity can both

be considered as an influencing factor that affects the

permeability of the restoration. Therefore, it is difficult to

directly compare the permeability of the adhesives in this

study with previous studies. Within the limits of this study,

Fig. 7 – Percent reduction in the DFF rate of each group at

30 min, 3 days, and 7 days after restoration with respect to

the baseline rate of DFF (n = 10). There was no significant

difference according to the three repeated measurement

times (P = 0.469) nor among the different types of the

materials at any time period (P = 0.226).

the adhesives used here would not be expected to show

significant difference in dentin permeability, at least when

used in restorations with all margins in enamel.

The main influencing factors affecting water evaporation,

which caused an outward flow from the exposed dentin

surface, are temperature and humidity.14,15 The temperature

and humidity of a prepared cavity surface are not constant in

most clinical situations. A prepared cavity can be filled under

the conditions of room temperature and ambient humidity

when isolated by a rubber dam, but also under intraoral

circumstances of 37 8C and 100% humidity as the other

extreme. The behavior of DFF during a composite restoration

procedure can also be affected by the finishing and polishing

procedure using rotary instruments, as heat may be generated

during this process.

In a future study, it would be interesting to investigate

how changes in the temperature and humidity around the

cavity and clinical procedures such as finishing and polish-

ing can affect the DFF and permeability of composite

restorations.

Acknowledgement

This study was supported by a grant (no. 3-2008-0014) from the

Seoul National University Dental Hospital Research Fund.


r e f e r e n c e s

1. Brannstrom M, Linden LA, Astrom A. The hydrodynamics ofthe dental tubule and of pulp fluid. A discussion of itssignificance in relation to dentinal sensitivity. CariesResearch 1967;1:310–7.

2. Brannstrom M, Linden LA, Johnson G. Movement of dentinaland pulpal fluid caused by clinical procedures. Journal ofDental Research 1968;47:679–82.

3. Brannstrom M. The hydrodynamic theory of dentinal pain:sensation in preparations, caries, and the dentinal cracksyndrome. Journal of Endodontics 1986;12:453–7.

4. Pashley DH, Matthews WG, Zhang Y, Johnson M. Fluid shiftsacross human dentin in vitro in response to hydrodynamicstimuli. Archives of Oral Biology 1996;41:1065–72.

5. Paphangkorakit J, Osborn JW. The effect of normal occlusalforces on fluid movement through human dentine in vitro.Archives of Oral Biology 2000;45:1033–41.

6. Hashimoto M, Ito S, Tay FR, Svizero NR, Sano H, Kaga M,et al. Fluid movement across the resin–dentin interfaceduring and after bonding. Journal of Dental Research2004;83:843–8.

7. Ratih DN, Palamara JE, Messer HH. Dentinal fluid flow andcuspal displacement in response to resin compositerestorative procedures. Dental Materials 2007;23:1405–11.

8. Ciucchi B, Bouillaguet S, Holz J, Pashley D. Dentinal fluiddynamics in human teeth, in vivo. Journal of Endodontics1995;21:191–4.

9. Kolker JL, Vargas MA, Armstrong SR, Dawson DV. Effect ofdesensitizing agents on dentin permeability and dentintubule occlusion. Journal of Adhesive Dentistry 2002;4:211–21.

10. Horiuchi H, Matthews B. In-vitro observations on fluid flowthrough human dentine caused by pain-producing stimuli.Archives of Oral Biology 1973;18:275–94.

11. Pashley DH, O’Meara JA, Williams EC, Kepler EE. Dentinpermeability: effects of cavity varnishes and bases. Journal ofProsthetic Dentistry 1985;53:511–6.

12. Pashley DH, Depew DD. Effects of the smear layer. Copaliteand oxalate on microleakage. Operative Dentistry 1986;11:95–102.

13. Carrilho MR, Tay FR, Sword J, Donnelly AM, Agee KA,Nishitani Y, et al. Dentine sealing provided by smear layer/smear plugs vs. adhesive resins/resin tags. European Journalof Oral Sciences 2007;115:321–9.

14. Goodis HE, Tao L, Pashley DH. Evaporative water loss fromhuman dentine in vitro. Archives of Oral Biology 1990;35:523–7.

15. Matthews WG, Showman CD, Pashley DH. Air blast-inducedevaporative water loss from human dentine, in vitro.Archives of Oral Biology 1993;38:517–23.

16. Benenson W, Harris JW, Stocker H, Lutz H. Handbook ofphysics. 1st ed. New York: Springer; 2006. (p. 185, 795).

17. Bausch JR, de Lange K, Davidson CL, Peters A, de Gee AJ.Clinical significance of polymerization shrinkage ofcomposite resins. The Journal of Prosthetic Dentistry1982;48:59–67.

18. Pashley DH. Dynamics of the pulpo-dentin complex. CriticalReviews in Oral Biology and Medicine 1996;7:104–33.

19. Tay FR, Pashley DH, Suh BI, Carvalho RM, Itthagarun A.Single-step adhesives are permeable membranes. Journal ofDentistry 2002;30:371–82.

20. Tay FR, Pashley DH. Water treeing – a potential mechanismfor degradation of dentin adhesives. American Journal ofDentistry 2003;16:6–12.

21. Itthagarun A, Tay FR, Pashley DH, Wefel JS, Garcıa-Godoy F,Wei SH. Single-step, self-etch adhesives behave aspermeable membranes after polymerization. Part III.Evidence from fluid conductance and artificial cariesinhibition. American Journal of Dentistry 2004;17:394–400.

22. King NM, Tay FR, Pashley DH, Hashimoto M, Ito S, BrackettWW, et al. Conversion of one-step to two-step self-etchadhesives for improved efficacy and extended application.American Journal of Dentistry 2005;18:126–34.

23. Brackett WW, Ito S, Tay FR, Haisch LD, Pashley DH.Microtensile bond strength of self-etching resins: effect of ahydrophobic layer. Operative Dentistry 2005;30:733–8.

24. Sauro S, Pashley DH, Montanari M, Chersoni S, Carvalho RM,Toledano M, et al. Effect of simulated pulpal pressure ondentin permeability and adhesion of self-etch adhesives.Dental Materials 2007;23:705–13.

25. Chersoni S, Suppa P, Grandini S, Goracci C, Monticelli F, YiuC, et al. In vivo and in vitro permeability of one-step self-etch adhesives. Journal of Dental Research 2004;83:459–64.

26. Ozok AR, Wu MK, Ten Cate JM, Wesselink PR. Effect ofdentinal fluid composition on dentin demineralization invitro. Journal of Dental Research 2004;83:849–53.

27. Vongsavan N, Matthews B. Fluid flow through cat dentine invivo. Archives of Oral Biology 1992;37:175–85.

28. Olgart L, Gazelius B. Effect of adrenalin and felypressine(octapressin) on blood flow and sensory nerve activity in thetooth. Acta Odontologica Scandinavica 1977;35:69–75.

29. Kim S, Edwall L, Trowbridge H, Chien S. Effects of localanesthetics on pulpal blood flow in dogs. Journal of DentalResearch 1984;63:650–2.

30. Odor TM, Pitt Ford TR, McDonald F. Adrenaline in localanaesthesia: the effect of concentration on dental pulpalcirculation and anaesthesia. Endodontics and DentalTraumatology 1994;10:167–73.

31. Ozok AR, De Gee AJ, Wu MK, Wesselink PR. The influence ofresin composite and bonded amalgam restorations ondentine permeability in Class II cavities in vitro. DentalMaterials 2001;17:477–84.

32. Hersek N, Canay S, Akca K, Ciftci Y. Comparison ofmicroleakage properties of three different filling materials.An autoradiographic study. Journal of Oral Rehabilitation2002;29:1212–7.

33. Murray PE, Hafez AA, Smith AJ, Cox CF. Bacterialmicroleakage and pulp inflammation associated withvarious restorative materials. Dental Materials 2002;18:470–8.

34. Gregoire G, Guignes P, Millas A. Effect of self-etchingadhesives on dentin permeability in a fluid flow model.Journal of Prosthetic Dentistry 2005;93:56–63.

real-time measurement of dentinal fluid flow during amalgam and composite restoration

Documents